diff options
180 files changed, 17658 insertions, 5701 deletions
diff --git a/Documentation/DocBook/drm.tmpl b/Documentation/DocBook/drm.tmpl index 196b8b9dba11..b0300529ab13 100644 --- a/Documentation/DocBook/drm.tmpl +++ b/Documentation/DocBook/drm.tmpl @@ -6,11 +6,36 @@ <bookinfo> <title>Linux DRM Developer's Guide</title> + <authorgroup> + <author> + <firstname>Jesse</firstname> + <surname>Barnes</surname> + <contrib>Initial version</contrib> + <affiliation> + <orgname>Intel Corporation</orgname> + <address> + <email>jesse.barnes@intel.com</email> + </address> + </affiliation> + </author> + <author> + <firstname>Laurent</firstname> + <surname>Pinchart</surname> + <contrib>Driver internals</contrib> + <affiliation> + <orgname>Ideas on board SPRL</orgname> + <address> + <email>laurent.pinchart@ideasonboard.com</email> + </address> + </affiliation> + </author> + </authorgroup> + <copyright> <year>2008-2009</year> - <holder> - Intel Corporation (Jesse Barnes <jesse.barnes@intel.com>) - </holder> + <year>2012</year> + <holder>Intel Corporation</holder> + <holder>Laurent Pinchart</holder> </copyright> <legalnotice> @@ -20,6 +45,17 @@ the kernel source COPYING file. </para> </legalnotice> + + <revhistory> + <!-- Put document revisions here, newest first. --> + <revision> + <revnumber>1.0</revnumber> + <date>2012-07-13</date> + <authorinitials>LP</authorinitials> + <revremark>Added extensive documentation about driver internals. + </revremark> + </revision> + </revhistory> </bookinfo> <toc></toc> @@ -72,342 +108,361 @@ submission & fencing, suspend/resume support, and DMA services. </para> - <para> - The core of every DRM driver is struct drm_driver. Drivers - typically statically initialize a drm_driver structure, - then pass it to drm_init() at load time. - </para> <!-- Internals: driver init --> <sect1> - <title>Driver initialization</title> - <para> - Before calling the DRM initialization routines, the driver must - first create and fill out a struct drm_driver structure. - </para> - <programlisting> - static struct drm_driver driver = { - /* Don't use MTRRs here; the Xserver or userspace app should - * deal with them for Intel hardware. - */ - .driver_features = - DRIVER_USE_AGP | DRIVER_REQUIRE_AGP | - DRIVER_HAVE_IRQ | DRIVER_IRQ_SHARED | DRIVER_MODESET, - .load = i915_driver_load, - .unload = i915_driver_unload, - .firstopen = i915_driver_firstopen, - .lastclose = i915_driver_lastclose, - .preclose = i915_driver_preclose, - .save = i915_save, - .restore = i915_restore, - .device_is_agp = i915_driver_device_is_agp, - .get_vblank_counter = i915_get_vblank_counter, - .enable_vblank = i915_enable_vblank, - .disable_vblank = i915_disable_vblank, - .irq_preinstall = i915_driver_irq_preinstall, - .irq_postinstall = i915_driver_irq_postinstall, - .irq_uninstall = i915_driver_irq_uninstall, - .irq_handler = i915_driver_irq_handler, - .reclaim_buffers = drm_core_reclaim_buffers, - .get_map_ofs = drm_core_get_map_ofs, - .get_reg_ofs = drm_core_get_reg_ofs, - .fb_probe = intelfb_probe, - .fb_remove = intelfb_remove, - .fb_resize = intelfb_resize, - .master_create = i915_master_create, - .master_destroy = i915_master_destroy, -#if defined(CONFIG_DEBUG_FS) - .debugfs_init = i915_debugfs_init, - .debugfs_cleanup = i915_debugfs_cleanup, -#endif - .gem_init_object = i915_gem_init_object, - .gem_free_object = i915_gem_free_object, - .gem_vm_ops = &i915_gem_vm_ops, - .ioctls = i915_ioctls, - .fops = { - .owner = THIS_MODULE, - .open = drm_open, - .release = drm_release, - .ioctl = drm_ioctl, - .mmap = drm_mmap, - .poll = drm_poll, - .fasync = drm_fasync, -#ifdef CONFIG_COMPAT - .compat_ioctl = i915_compat_ioctl, -#endif - .llseek = noop_llseek, - }, - .pci_driver = { - .name = DRIVER_NAME, - .id_table = pciidlist, - .probe = probe, - .remove = __devexit_p(drm_cleanup_pci), - }, - .name = DRIVER_NAME, - .desc = DRIVER_DESC, - .date = DRIVER_DATE, - .major = DRIVER_MAJOR, - .minor = DRIVER_MINOR, - .patchlevel = DRIVER_PATCHLEVEL, - }; - </programlisting> - <para> - In the example above, taken from the i915 DRM driver, the driver - sets several flags indicating what core features it supports; - we go over the individual callbacks in later sections. Since - flags indicate which features your driver supports to the DRM - core, you need to set most of them prior to calling drm_init(). Some, - like DRIVER_MODESET can be set later based on user supplied parameters, - but that's the exception rather than the rule. - </para> - <variablelist> - <title>Driver flags</title> - <varlistentry> - <term>DRIVER_USE_AGP</term> - <listitem><para> - Driver uses AGP interface - </para></listitem> - </varlistentry> - <varlistentry> - <term>DRIVER_REQUIRE_AGP</term> - <listitem><para> - Driver needs AGP interface to function. - </para></listitem> - </varlistentry> - <varlistentry> - <term>DRIVER_USE_MTRR</term> - <listitem> - <para> - Driver uses MTRR interface for mapping memory. Deprecated. - </para> - </listitem> - </varlistentry> - <varlistentry> - <term>DRIVER_PCI_DMA</term> - <listitem><para> - Driver is capable of PCI DMA. Deprecated. - </para></listitem> - </varlistentry> - <varlistentry> - <term>DRIVER_SG</term> - <listitem><para> - Driver can perform scatter/gather DMA. Deprecated. - </para></listitem> - </varlistentry> - <varlistentry> - <term>DRIVER_HAVE_DMA</term> - <listitem><para>Driver supports DMA. Deprecated.</para></listitem> - </varlistentry> - <varlistentry> - <term>DRIVER_HAVE_IRQ</term><term>DRIVER_IRQ_SHARED</term> - <listitem> - <para> - DRIVER_HAVE_IRQ indicates whether the driver has an IRQ - handler. DRIVER_IRQ_SHARED indicates whether the device & - handler support shared IRQs (note that this is required of - PCI drivers). - </para> - </listitem> - </varlistentry> - <varlistentry> - <term>DRIVER_DMA_QUEUE</term> - <listitem> - <para> - Should be set if the driver queues DMA requests and completes them - asynchronously. Deprecated. - </para> - </listitem> - </varlistentry> - <varlistentry> - <term>DRIVER_FB_DMA</term> - <listitem> - <para> - Driver supports DMA to/from the framebuffer. Deprecated. - </para> - </listitem> - </varlistentry> - <varlistentry> - <term>DRIVER_MODESET</term> - <listitem> - <para> - Driver supports mode setting interfaces. - </para> - </listitem> - </varlistentry> - </variablelist> - <para> - In this specific case, the driver requires AGP and supports - IRQs. DMA, as discussed later, is handled by device-specific ioctls - in this case. It also supports the kernel mode setting APIs, though - unlike in the actual i915 driver source, this example unconditionally - exports KMS capability. + <title>Driver Initialization</title> + <para> + At the core of every DRM driver is a <structname>drm_driver</structname> + structure. Drivers typically statically initialize a drm_driver structure, + and then pass it to one of the <function>drm_*_init()</function> functions + to register it with the DRM subsystem. </para> - </sect1> - - <!-- Internals: driver load --> - - <sect1> - <title>Driver load</title> - <para> - In the previous section, we saw what a typical drm_driver - structure might look like. One of the more important fields in - the structure is the hook for the load function. - </para> - <programlisting> - static struct drm_driver driver = { - ... - .load = i915_driver_load, - ... - }; - </programlisting> - <para> - The load function has many responsibilities: allocating a driver - private structure, specifying supported performance counters, - configuring the device (e.g. mapping registers & command - buffers), initializing the memory manager, and setting up the - initial output configuration. - </para> - <para> - If compatibility is a concern (e.g. with drivers converted over - to the new interfaces from the old ones), care must be taken to - prevent device initialization and control that is incompatible with - currently active userspace drivers. For instance, if user - level mode setting drivers are in use, it would be problematic - to perform output discovery & configuration at load time. - Likewise, if user-level drivers unaware of memory management are - in use, memory management and command buffer setup may need to - be omitted. These requirements are driver-specific, and care - needs to be taken to keep both old and new applications and - libraries working. The i915 driver supports the "modeset" - module parameter to control whether advanced features are - enabled at load time or in legacy fashion. + <para> + The <structname>drm_driver</structname> structure contains static + information that describes the driver and features it supports, and + pointers to methods that the DRM core will call to implement the DRM API. + We will first go through the <structname>drm_driver</structname> static + information fields, and will then describe individual operations in + details as they get used in later sections. </para> - <sect2> - <title>Driver private & performance counters</title> - <para> - The driver private hangs off the main drm_device structure and - can be used for tracking various device-specific bits of - information, like register offsets, command buffer status, - register state for suspend/resume, etc. At load time, a - driver may simply allocate one and set drm_device.dev_priv - appropriately; it should be freed and drm_device.dev_priv set - to NULL when the driver is unloaded. - </para> + <title>Driver Information</title> + <sect3> + <title>Driver Features</title> + <para> + Drivers inform the DRM core about their requirements and supported + features by setting appropriate flags in the + <structfield>driver_features</structfield> field. Since those flags + influence the DRM core behaviour since registration time, most of them + must be set to registering the <structname>drm_driver</structname> + instance. + </para> + <synopsis>u32 driver_features;</synopsis> + <variablelist> + <title>Driver Feature Flags</title> + <varlistentry> + <term>DRIVER_USE_AGP</term> + <listitem><para> + Driver uses AGP interface, the DRM core will manage AGP resources. + </para></listitem> + </varlistentry> + <varlistentry> + <term>DRIVER_REQUIRE_AGP</term> + <listitem><para> + Driver needs AGP interface to function. AGP initialization failure + will become a fatal error. + </para></listitem> + </varlistentry> + <varlistentry> + <term>DRIVER_USE_MTRR</term> + <listitem><para> + Driver uses MTRR interface for mapping memory, the DRM core will + manage MTRR resources. Deprecated. + </para></listitem> + </varlistentry> + <varlistentry> + <term>DRIVER_PCI_DMA</term> + <listitem><para> + Driver is capable of PCI DMA, mapping of PCI DMA buffers to + userspace will be enabled. Deprecated. + </para></listitem> + </varlistentry> + <varlistentry> + <term>DRIVER_SG</term> + <listitem><para> + Driver can perform scatter/gather DMA, allocation and mapping of + scatter/gather buffers will be enabled. Deprecated. + </para></listitem> + </varlistentry> + <varlistentry> + <term>DRIVER_HAVE_DMA</term> + <listitem><para> + Driver supports DMA, the userspace DMA API will be supported. + Deprecated. + </para></listitem> + </varlistentry> + <varlistentry> + <term>DRIVER_HAVE_IRQ</term><term>DRIVER_IRQ_SHARED</term> + <listitem><para> + DRIVER_HAVE_IRQ indicates whether the driver has an IRQ handler. The + DRM core will automatically register an interrupt handler when the + flag is set. DRIVER_IRQ_SHARED indicates whether the device & + handler support shared IRQs (note that this is required of PCI + drivers). + </para></listitem> + </varlistentry> + <varlistentry> + <term>DRIVER_IRQ_VBL</term> + <listitem><para>Unused. Deprecated.</para></listitem> + </varlistentry> + <varlistentry> + <term>DRIVER_DMA_QUEUE</term> + <listitem><para> + Should be set if the driver queues DMA requests and completes them + asynchronously. Deprecated. + </para></listitem> + </varlistentry> + <varlistentry> + <term>DRIVER_FB_DMA</term> + <listitem><para> + Driver supports DMA to/from the framebuffer, mapping of frambuffer + DMA buffers to userspace will be supported. Deprecated. + </para></listitem> + </varlistentry> + <varlistentry> + <term>DRIVER_IRQ_VBL2</term> + <listitem><para>Unused. Deprecated.</para></listitem> + </varlistentry> + <varlistentry> + <term>DRIVER_GEM</term> + <listitem><para> + Driver use the GEM memory manager. + </para></listitem> + </varlistentry> + <varlistentry> + <term>DRIVER_MODESET</term> + <listitem><para> + Driver supports mode setting interfaces (KMS). + </para></listitem> + </varlistentry> + <varlistentry> + <term>DRIVER_PRIME</term> + <listitem><para> + Driver implements DRM PRIME buffer sharing. + </para></listitem> + </varlistentry> + </variablelist> + </sect3> + <sect3> + <title>Major, Minor and Patchlevel</title> + <synopsis>int major; +int minor; +int patchlevel;</synopsis> + <para> + The DRM core identifies driver versions by a major, minor and patch + level triplet. The information is printed to the kernel log at + initialization time and passed to userspace through the + DRM_IOCTL_VERSION ioctl. + </para> + <para> + The major and minor numbers are also used to verify the requested driver + API version passed to DRM_IOCTL_SET_VERSION. When the driver API changes + between minor versions, applications can call DRM_IOCTL_SET_VERSION to + select a specific version of the API. If the requested major isn't equal + to the driver major, or the requested minor is larger than the driver + minor, the DRM_IOCTL_SET_VERSION call will return an error. Otherwise + the driver's set_version() method will be called with the requested + version. + </para> + </sect3> + <sect3> + <title>Name, Description and Date</title> + <synopsis>char *name; +char *desc; +char *date;</synopsis> + <para> + The driver name is printed to the kernel log at initialization time, + used for IRQ registration and passed to userspace through + DRM_IOCTL_VERSION. + </para> + <para> + The driver description is a purely informative string passed to + userspace through the DRM_IOCTL_VERSION ioctl and otherwise unused by + the kernel. + </para> + <para> + The driver date, formatted as YYYYMMDD, is meant to identify the date of + the latest modification to the driver. However, as most drivers fail to + update it, its value is mostly useless. The DRM core prints it to the + kernel log at initialization time and passes it to userspace through the + DRM_IOCTL_VERSION ioctl. + </para> + </sect3> + </sect2> + <sect2> + <title>Driver Load</title> <para> - The DRM supports several counters which may be used for rough - performance characterization. Note that the DRM stat counter - system is not often used by applications, and supporting - additional counters is completely optional. + The <methodname>load</methodname> method is the driver and device + initialization entry point. The method is responsible for allocating and + initializing driver private data, specifying supported performance + counters, performing resource allocation and mapping (e.g. acquiring + clocks, mapping registers or allocating command buffers), initializing + the memory manager (<xref linkend="drm-memory-management"/>), installing + the IRQ handler (<xref linkend="drm-irq-registration"/>), setting up + vertical blanking handling (<xref linkend="drm-vertical-blank"/>), mode + setting (<xref linkend="drm-mode-setting"/>) and initial output + configuration (<xref linkend="drm-kms-init"/>). </para> + <note><para> + If compatibility is a concern (e.g. with drivers converted over from + User Mode Setting to Kernel Mode Setting), care must be taken to prevent + device initialization and control that is incompatible with currently + active userspace drivers. For instance, if user level mode setting + drivers are in use, it would be problematic to perform output discovery + & configuration at load time. Likewise, if user-level drivers + unaware of memory management are in use, memory management and command + buffer setup may need to be omitted. These requirements are + driver-specific, and care needs to be taken to keep both old and new + applications and libraries working. + </para></note> + <synopsis>int (*load) (struct drm_device *, unsigned long flags);</synopsis> <para> - These interfaces are deprecated and should not be used. If performance - monitoring is desired, the developer should investigate and - potentially enhance the kernel perf and tracing infrastructure to export - GPU related performance information for consumption by performance - monitoring tools and applications. + The method takes two arguments, a pointer to the newly created + <structname>drm_device</structname> and flags. The flags are used to + pass the <structfield>driver_data</structfield> field of the device id + corresponding to the device passed to <function>drm_*_init()</function>. + Only PCI devices currently use this, USB and platform DRM drivers have + their <methodname>load</methodname> method called with flags to 0. </para> + <sect3> + <title>Driver Private & Performance Counters</title> + <para> + The driver private hangs off the main + <structname>drm_device</structname> structure and can be used for + tracking various device-specific bits of information, like register + offsets, command buffer status, register state for suspend/resume, etc. + At load time, a driver may simply allocate one and set + <structname>drm_device</structname>.<structfield>dev_priv</structfield> + appropriately; it should be freed and + <structname>drm_device</structname>.<structfield>dev_priv</structfield> + set to NULL when the driver is unloaded. + </para> + <para> + DRM supports several counters which were used for rough performance + characterization. This stat counter system is deprecated and should not + be used. If performance monitoring is desired, the developer should + investigate and potentially enhance the kernel perf and tracing + infrastructure to export GPU related performance information for + consumption by performance monitoring tools and applications. + </para> + </sect3> + <sect3 id="drm-irq-registration"> + <title>IRQ Registration</title> + <para> + The DRM core tries to facilitate IRQ handler registration and + unregistration by providing <function>drm_irq_install</function> and + <function>drm_irq_uninstall</function> functions. Those functions only + support a single interrupt per device. + </para> + <!--!Fdrivers/char/drm/drm_irq.c drm_irq_install--> + <para> + Both functions get the device IRQ by calling + <function>drm_dev_to_irq</function>. This inline function will call a + bus-specific operation to retrieve the IRQ number. For platform devices, + <function>platform_get_irq</function>(..., 0) is used to retrieve the + IRQ number. + </para> + <para> + <function>drm_irq_install</function> starts by calling the + <methodname>irq_preinstall</methodname> driver operation. The operation + is optional and must make sure that the interrupt will not get fired by + clearing all pending interrupt flags or disabling the interrupt. + </para> + <para> + The IRQ will then be requested by a call to + <function>request_irq</function>. If the DRIVER_IRQ_SHARED driver + feature flag is set, a shared (IRQF_SHARED) IRQ handler will be + requested. + </para> + <para> + The IRQ handler function must be provided as the mandatory irq_handler + driver operation. It will get passed directly to + <function>request_irq</function> and thus has the same prototype as all + IRQ handlers. It will get called with a pointer to the DRM device as the + second argument. + </para> + <para> + Finally the function calls the optional + <methodname>irq_postinstall</methodname> driver operation. The operation + usually enables interrupts (excluding the vblank interrupt, which is + enabled separately), but drivers may choose to enable/disable interrupts + at a different time. + </para> + <para> + <function>drm_irq_uninstall</function> is similarly used to uninstall an + IRQ handler. It starts by waking up all processes waiting on a vblank + interrupt to make sure they don't hang, and then calls the optional + <methodname>irq_uninstall</methodname> driver operation. The operation + must disable all hardware interrupts. Finally the function frees the IRQ + by calling <function>free_irq</function>. + </para> + </sect3> + <sect3> + <title>Memory Manager Initialization</title> + <para> + Every DRM driver requires a memory manager which must be initialized at + load time. DRM currently contains two memory managers, the Translation + Table Manager (TTM) and the Graphics Execution Manager (GEM). + This document describes the use of the GEM memory manager only. See + <xref linkend="drm-memory-management"/> for details. + </para> + </sect3> + <sect3> + <title>Miscellaneous Device Configuration</title> + <para> + Another task that may be necessary for PCI devices during configuration + is mapping the video BIOS. On many devices, the VBIOS describes device + configuration, LCD panel timings (if any), and contains flags indicating + device state. Mapping the BIOS can be done using the pci_map_rom() call, + a convenience function that takes care of mapping the actual ROM, + whether it has been shadowed into memory (typically at address 0xc0000) + or exists on the PCI device in the ROM BAR. Note that after the ROM has + been mapped and any necessary information has been extracted, it should + be unmapped; on many devices, the ROM address decoder is shared with + other BARs, so leaving it mapped could cause undesired behaviour like + hangs or memory corruption. + <!--!Fdrivers/pci/rom.c pci_map_rom--> + </para> + </sect3> </sect2> + </sect1> - <sect2> - <title>Configuring the device</title> - <para> - Obviously, device configuration is device-specific. - However, there are several common operations: finding a - device's PCI resources, mapping them, and potentially setting - up an IRQ handler. - </para> - <para> - Finding & mapping resources is fairly straightforward. The - DRM wrapper functions, drm_get_resource_start() and - drm_get_resource_len(), may be used to find BARs on the given - drm_device struct. Once those values have been retrieved, the - driver load function can call drm_addmap() to create a new - mapping for the BAR in question. Note that you probably want a - drm_local_map_t in your driver private structure to track any - mappings you create. -<!-- !Fdrivers/gpu/drm/drm_bufs.c drm_get_resource_* --> -<!-- !Finclude/drm/drmP.h drm_local_map_t --> - </para> - <para> - if compatibility with other operating systems isn't a concern - (DRM drivers can run under various BSD variants and OpenSolaris), - native Linux calls may be used for the above, e.g. pci_resource_* - and iomap*/iounmap. See the Linux device driver book for more - info. - </para> - <para> - Once you have a register map, you may use the DRM_READn() and - DRM_WRITEn() macros to access the registers on your device, or - use driver-specific versions to offset into your MMIO space - relative to a driver-specific base pointer (see I915_READ for - an example). - </para> - <para> - If your device supports interrupt generation, you may want to - set up an interrupt handler when the driver is loaded. This - is done using the drm_irq_install() function. If your device - supports vertical blank interrupts, it should call - drm_vblank_init() to initialize the core vblank handling code before - enabling interrupts on your device. This ensures the vblank related - structures are allocated and allows the core to handle vblank events. - </para> -<!--!Fdrivers/char/drm/drm_irq.c drm_irq_install--> - <para> - Once your interrupt handler is registered (it uses your - drm_driver.irq_handler as the actual interrupt handling - function), you can safely enable interrupts on your device, - assuming any other state your interrupt handler uses is also - initialized. - </para> - <para> - Another task that may be necessary during configuration is - mapping the video BIOS. On many devices, the VBIOS describes - device configuration, LCD panel timings (if any), and contains - flags indicating device state. Mapping the BIOS can be done - using the pci_map_rom() call, a convenience function that - takes care of mapping the actual ROM, whether it has been - shadowed into memory (typically at address 0xc0000) or exists - on the PCI device in the ROM BAR. Note that after the ROM - has been mapped and any necessary information has been extracted, - it should be unmapped; on many devices, the ROM address decoder is - shared with other BARs, so leaving it mapped could cause - undesired behavior like hangs or memory corruption. -<!--!Fdrivers/pci/rom.c pci_map_rom--> - </para> - </sect2> + <!-- Internals: memory management --> + <sect1 id="drm-memory-management"> + <title>Memory management</title> + <para> + Modern Linux systems require large amount of graphics memory to store + frame buffers, textures, vertices and other graphics-related data. Given + the very dynamic nature of many of that data, managing graphics memory + efficiently is thus crucial for the graphics stack and plays a central + role in the DRM infrastructure. + </para> + <para> + The DRM core includes two memory managers, namely Translation Table Maps + (TTM) and Graphics Execution Manager (GEM). TTM was the first DRM memory + manager to be developed and tried to be a one-size-fits-them all + solution. It provides a single userspace API to accomodate the need of + all hardware, supporting both Unified Memory Architecture (UMA) devices + and devices with dedicated video RAM (i.e. most discrete video cards). + This resulted in a large, complex piece of code that turned out to be + hard to use for driver development. + </para> + <para> + GEM started as an Intel-sponsored project in reaction to TTM's + complexity. Its design philosophy is completely different: instead of + providing a solution to every graphics memory-related problems, GEM + identified common code between drivers and created a support library to + share it. GEM has simpler initialization and execution requirements than + TTM, but has no video RAM management capabitilies and is thus limited to + UMA devices. + </para> <sect2> - <title>Memory manager initialization</title> - <para> - In order to allocate command buffers, cursor memory, scanout - buffers, etc., as well as support the latest features provided - by packages like Mesa and the X.Org X server, your driver - should support a memory manager. - </para> + <title>The Translation Table Manager (TTM)</title> <para> - If your driver supports memory management (it should!), you - need to set that up at load time as well. How you initialize - it depends on which memory manager you're using: TTM or GEM. + TTM design background and information belongs here. </para> <sect3> <title>TTM initialization</title> - <para> - TTM (for Translation Table Manager) manages video memory and - aperture space for graphics devices. TTM supports both UMA devices - and devices with dedicated video RAM (VRAM), i.e. most discrete - graphics devices. If your device has dedicated RAM, supporting - TTM is desirable. TTM also integrates tightly with your - driver-specific buffer execution function. See the radeon - driver for examples. - </para> - <para> - The core TTM structure is the ttm_bo_driver struct. It contains - several fields with function pointers for initializing the TTM, - allocating and freeing memory, waiting for command completion - and fence synchronization, and memory migration. See the - radeon_ttm.c file for an example of usage. + <warning><para>This section is outdated.</para></warning> + <para> + Drivers wishing to support TTM must fill out a drm_bo_driver + structure. The structure contains several fields with function + pointers for initializing the TTM, allocating and freeing memory, + waiting for command completion and fence synchronization, and memory + migration. See the radeon_ttm.c file for an example of usage. </para> <para> The ttm_global_reference structure is made up of several fields: @@ -445,82 +500,1081 @@ count for the TTM, which will call your initialization function. </para> </sect3> + </sect2> + <sect2 id="drm-gem"> + <title>The Graphics Execution Manager (GEM)</title> + <para> + The GEM design approach has resulted in a memory manager that doesn't + provide full coverage of all (or even all common) use cases in its + userspace or kernel API. GEM exposes a set of standard memory-related + operations to userspace and a set of helper functions to drivers, and let + drivers implement hardware-specific operations with their own private API. + </para> + <para> + The GEM userspace API is described in the + <ulink url="http://lwn.net/Articles/283798/"><citetitle>GEM - the Graphics + Execution Manager</citetitle></ulink> article on LWN. While slightly + outdated, the document provides a good overview of the GEM API principles. + Buffer allocation and read and write operations, described as part of the + common GEM API, are currently implemented using driver-specific ioctls. + </para> + <para> + GEM is data-agnostic. It manages abstract buffer objects without knowing + what individual buffers contain. APIs that require knowledge of buffer + contents or purpose, such as buffer allocation or synchronization + primitives, are thus outside of the scope of GEM and must be implemented + using driver-specific ioctls. + </para> + <para> + On a fundamental level, GEM involves several operations: + <itemizedlist> + <listitem>Memory allocation and freeing</listitem> + <listitem>Command execution</listitem> + <listitem>Aperture management at command execution time</listitem> + </itemizedlist> + Buffer object allocation is relatively straightforward and largely + provided by Linux's shmem layer, which provides memory to back each + object. + </para> + <para> + Device-specific operations, such as command execution, pinning, buffer + read & write, mapping, and domain ownership transfers are left to + driver-specific ioctls. + </para> + <sect3> + <title>GEM Initialization</title> + <para> + Drivers that use GEM must set the DRIVER_GEM bit in the struct + <structname>drm_driver</structname> + <structfield>driver_features</structfield> field. The DRM core will + then automatically initialize the GEM core before calling the + <methodname>load</methodname> operation. Behind the scene, this will + create a DRM Memory Manager object which provides an address space + pool for object allocation. + </para> + <para> + In a KMS configuration, drivers need to allocate and initialize a + command ring buffer following core GEM initialization if required by + the hardware. UMA devices usually have what is called a "stolen" + memory region, which provides space for the initial framebuffer and + large, contiguous memory regions required by the device. This space is + typically not managed by GEM, and must be initialized separately into + its own DRM MM object. + </para> + </sect3> <sect3> - <title>GEM initialization</title> - <para> - GEM is an alternative to TTM, designed specifically for UMA - devices. It has simpler initialization and execution requirements - than TTM, but has no VRAM management capability. Core GEM - is initialized by calling drm_mm_init() to create - a GTT DRM MM object, which provides an address space pool for - object allocation. In a KMS configuration, the driver - needs to allocate and initialize a command ring buffer following - core GEM initialization. A UMA device usually has what is called a - "stolen" memory region, which provides space for the initial - framebuffer and large, contiguous memory regions required by the - device. This space is not typically managed by GEM, and it must - be initialized separately into its own DRM MM object. - </para> - <para> - Initialization is driver-specific. In the case of Intel - integrated graphics chips like 965GM, GEM initialization can - be done by calling the internal GEM init function, - i915_gem_do_init(). Since the 965GM is a UMA device - (i.e. it doesn't have dedicated VRAM), GEM manages - making regular RAM available for GPU operations. Memory set - aside by the BIOS (called "stolen" memory by the i915 - driver) is managed by the DRM memrange allocator; the - rest of the aperture is managed by GEM. - <programlisting> - /* Basic memrange allocator for stolen space (aka vram) */ - drm_memrange_init(&dev_priv->vram, 0, prealloc_size); - /* Let GEM Manage from end of prealloc space to end of aperture */ - i915_gem_do_init(dev, prealloc_size, agp_size); - </programlisting> -<!--!Edrivers/char/drm/drm_memrange.c--> - </para> - <para> - Once the memory manager has been set up, we may allocate the - command buffer. In the i915 case, this is also done with a - GEM function, i915_gem_init_ringbuffer(). - </para> + <title>GEM Objects Creation</title> + <para> + GEM splits creation of GEM objects and allocation of the memory that + backs them in two distinct operations. + </para> + <para> + GEM objects are represented by an instance of struct + <structname>drm_gem_object</structname>. Drivers usually need to extend + GEM objects with private information and thus create a driver-specific + GEM object structure type that embeds an instance of struct + <structname>drm_gem_object</structname>. + </para> + <para> + To create a GEM object, a driver allocates memory for an instance of its + specific GEM object type and initializes the embedded struct + <structname>drm_gem_object</structname> with a call to + <function>drm_gem_object_init</function>. The function takes a pointer to + the DRM device, a pointer to the GEM object and the buffer object size + in bytes. + </para> + <para> + GEM uses shmem to allocate anonymous pageable memory. + <function>drm_gem_object_init</function> will create an shmfs file of + the requested size and store it into the struct + <structname>drm_gem_object</structname> <structfield>filp</structfield> + field. The memory is used as either main storage for the object when the + graphics hardware uses system memory directly or as a backing store + otherwise. + </para> + <para> + Drivers are responsible for the actual physical pages allocation by + calling <function>shmem_read_mapping_page_gfp</function> for each page. + Note that they can decide to allocate pages when initializing the GEM + object, or to delay allocation until the memory is needed (for instance + when a page fault occurs as a result of a userspace memory access or + when the driver needs to start a DMA transfer involving the memory). + </para> + <para> + Anonymous pageable memory allocation is not always desired, for instance + when the hardware requires physically contiguous system memory as is + often the case in embedded devices. Drivers can create GEM objects with + no shmfs backing (called private GEM objects) by initializing them with + a call to <function>drm_gem_private_object_init</function> instead of + <function>drm_gem_object_init</function>. Storage for private GEM + objects must be managed by drivers. + </para> + <para> + Drivers that do not need to extend GEM objects with private information + can call the <function>drm_gem_object_alloc</function> function to + allocate and initialize a struct <structname>drm_gem_object</structname> + instance. The GEM core will call the optional driver + <methodname>gem_init_object</methodname> operation after initializing + the GEM object with <function>drm_gem_object_init</function>. + <synopsis>int (*gem_init_object) (struct drm_gem_object *obj);</synopsis> + </para> + <para> + No alloc-and-init function exists for private GEM objects. + </para> + </sect3> + <sect3> + <title>GEM Objects Lifetime</title> + <para> + All GEM objects are reference-counted by the GEM core. References can be + acquired and release by <function>calling drm_gem_object_reference</function> + and <function>drm_gem_object_unreference</function> respectively. The + caller must hold the <structname>drm_device</structname> + <structfield>struct_mutex</structfield> lock. As a convenience, GEM + provides the <function>drm_gem_object_reference_unlocked</function> and + <function>drm_gem_object_unreference_unlocked</function> functions that + can be called without holding the lock. + </para> + <para> + When the last reference to a GEM object is released the GEM core calls + the <structname>drm_driver</structname> + <methodname>gem_free_object</methodname> operation. That operation is + mandatory for GEM-enabled drivers and must free the GEM object and all + associated resources. + </para> + <para> + <synopsis>void (*gem_free_object) (struct drm_gem_object *obj);</synopsis> + Drivers are responsible for freeing all GEM object resources, including + the resources created by the GEM core. If an mmap offset has been + created for the object (in which case + <structname>drm_gem_object</structname>::<structfield>map_list</structfield>::<structfield>map</structfield> + is not NULL) it must be freed by a call to + <function>drm_gem_free_mmap_offset</function>. The shmfs backing store + must be released by calling <function>drm_gem_object_release</function> + (that function can safely be called if no shmfs backing store has been + created). + </para> + </sect3> + <sect3> + <title>GEM Objects Naming</title> + <para> + Communication between userspace and the kernel refers to GEM objects + using local handles, global names or, more recently, file descriptors. + All of those are 32-bit integer values; the usual Linux kernel limits + apply to the file descriptors. + </para> + <para> + GEM handles are local to a DRM file. Applications get a handle to a GEM + object through a driver-specific ioctl, and can use that handle to refer + to the GEM object in other standard or driver-specific ioctls. Closing a + DRM file handle frees all its GEM handles and dereferences the + associated GEM objects. + </para> + <para> + To create a handle for a GEM object drivers call + <function>drm_gem_handle_create</function>. The function takes a pointer + to the DRM file and the GEM object and returns a locally unique handle. + When the handle is no longer needed drivers delete it with a call to + <function>drm_gem_handle_delete</function>. Finally the GEM object + associated with a handle can be retrieved by a call to + <function>drm_gem_object_lookup</function>. + </para> + <para> + Handles don't take ownership of GEM objects, they only take a reference + to the object that will be dropped when the handle is destroyed. To + avoid leaking GEM objects, drivers must make sure they drop the + reference(s) they own (such as the initial reference taken at object + creation time) as appropriate, without any special consideration for the + handle. For example, in the particular case of combined GEM object and + handle creation in the implementation of the + <methodname>dumb_create</methodname> operation, drivers must drop the + initial reference to the GEM object before returning the handle. + </para> + <para> + GEM names are similar in purpose to handles but are not local to DRM + files. They can be passed between processes to reference a GEM object + globally. Names can't be used directly to refer to objects in the DRM + API, applications must convert handles to names and names to handles + using the DRM_IOCTL_GEM_FLINK and DRM_IOCTL_GEM_OPEN ioctls + respectively. The conversion is handled by the DRM core without any + driver-specific support. + </para> + <para> + Similar to global names, GEM file descriptors are also used to share GEM + objects across processes. They offer additional security: as file + descriptors must be explictly sent over UNIX domain sockets to be shared + between applications, they can't be guessed like the globally unique GEM + names. + </para> + <para> + Drivers that support GEM file descriptors, also known as the DRM PRIME + API, must set the DRIVER_PRIME bit in the struct + <structname>drm_driver</structname> + <structfield>driver_features</structfield> field, and implement the + <methodname>prime_handle_to_fd</methodname> and + <methodname>prime_fd_to_handle</methodname> operations. + </para> + <para> + <synopsis>int (*prime_handle_to_fd)(struct drm_device *dev, + struct drm_file *file_priv, uint32_t handle, + uint32_t flags, int *prime_fd); + int (*prime_fd_to_handle)(struct drm_device *dev, + struct drm_file *file_priv, int prime_fd, + uint32_t *handle);</synopsis> + Those two operations convert a handle to a PRIME file descriptor and + vice versa. Drivers must use the kernel dma-buf buffer sharing framework + to manage the PRIME file descriptors. + </para> + <para> + While non-GEM drivers must implement the operations themselves, GEM + drivers must use the <function>drm_gem_prime_handle_to_fd</function> + and <function>drm_gem_prime_fd_to_handle</function> helper functions. + Those helpers rely on the driver + <methodname>gem_prime_export</methodname> and + <methodname>gem_prime_import</methodname> operations to create a dma-buf + instance from a GEM object (dma-buf exporter role) and to create a GEM + object from a dma-buf instance (dma-buf importer role). + </para> + <para> + <synopsis>struct dma_buf * (*gem_prime_export)(struct drm_device *dev, + struct drm_gem_object *obj, + int flags); + struct drm_gem_object * (*gem_prime_import)(struct drm_device *dev, + struct dma_buf *dma_buf);</synopsis> + These two operations are mandatory for GEM drivers that support DRM + PRIME. + </para> + </sect3> + <sect3 id="drm-gem-objects-mapping"> + <title>GEM Objects Mapping</title> + <para> + Because mapping operations are fairly heavyweight GEM favours + read/write-like access to buffers, implemented through driver-specific + ioctls, over mapping buffers to userspace. However, when random access + to the buffer is needed (to perform software rendering for instance), + direct access to the object can be more efficient. + </para> + <para> + The mmap system call can't be used directly to map GEM objects, as they + don't have their own file handle. Two alternative methods currently + co-exist to map GEM objects to userspace. The first method uses a + driver-specific ioctl to perform the mapping operation, calling + <function>do_mmap</function> under the hood. This is often considered + dubious, seems to be discouraged for new GEM-enabled drivers, and will + thus not be described here. + </para> + <para> + The second method uses the mmap system call on the DRM file handle. + <synopsis>void *mmap(void *addr, size_t length, int prot, int flags, int fd, + off_t offset);</synopsis> + DRM identifies the GEM object to be mapped by a fake offset passed + through the mmap offset argument. Prior to being mapped, a GEM object + must thus be associated with a fake offset. To do so, drivers must call + <function>drm_gem_create_mmap_offset</function> on the object. The + function allocates a fake offset range from a pool and stores the + offset divided by PAGE_SIZE in + <literal>obj->map_list.hash.key</literal>. Care must be taken not to + call <function>drm_gem_create_mmap_offset</function> if a fake offset + has already been allocated for the object. This can be tested by + <literal>obj->map_list.map</literal> being non-NULL. + </para> + <para> + Once allocated, the fake offset value + (<literal>obj->map_list.hash.key << PAGE_SHIFT</literal>) + must be passed to the application in a driver-specific way and can then + be used as the mmap offset argument. + </para> + <para> + The GEM core provides a helper method <function>drm_gem_mmap</function> + to handle object mapping. The method can be set directly as the mmap + file operation handler. It will look up the GEM object based on the + offset value and set the VMA operations to the + <structname>drm_driver</structname> <structfield>gem_vm_ops</structfield> + field. Note that <function>drm_gem_mmap</function> doesn't map memory to + userspace, but relies on the driver-provided fault handler to map pages + individually. + </para> + <para> + To use <function>drm_gem_mmap</function>, drivers must fill the struct + <structname>drm_driver</structname> <structfield>gem_vm_ops</structfield> + field with a pointer to VM operations. + </para> + <para> + <synopsis>struct vm_operations_struct *gem_vm_ops + + struct vm_operations_struct { + void (*open)(struct vm_area_struct * area); + void (*close)(struct vm_area_struct * area); + int (*fault)(struct vm_area_struct *vma, struct vm_fault *vmf); + };</synopsis> + </para> + <para> + The <methodname>open</methodname> and <methodname>close</methodname> + operations must update the GEM object reference count. Drivers can use + the <function>drm_gem_vm_open</function> and + <function>drm_gem_vm_close</function> helper functions directly as open + and close handlers. + </para> + <para> + The fault operation handler is responsible for mapping individual pages + to userspace when a page fault occurs. Depending on the memory + allocation scheme, drivers can allocate pages at fault time, or can + decide to allocate memory for the GEM object at the time the object is + created. + </para> + <para> + Drivers that want to map the GEM object upfront instead of handling page + faults can implement their own mmap file operation handler. + </para> + </sect3> + <sect3> + <title>Dumb GEM Objects</title> + <para> + The GEM API doesn't standardize GEM objects creation and leaves it to + driver-specific ioctls. While not an issue for full-fledged graphics + stacks that include device-specific userspace components (in libdrm for + instance), this limit makes DRM-based early boot graphics unnecessarily + complex. + </para> + <para> + Dumb GEM objects partly alleviate the problem by providing a standard + API to create dumb buffers suitable for scanout, which can then be used + to create KMS frame buffers. + </para> + <para> + To support dumb GEM objects drivers must implement the + <methodname>dumb_create</methodname>, + <methodname>dumb_destroy</methodname> and + <methodname>dumb_map_offset</methodname> operations. + </para> + <itemizedlist> + <listitem> + <synopsis>int (*dumb_create)(struct drm_file *file_priv, struct drm_device *dev, + struct drm_mode_create_dumb *args);</synopsis> + <para> + The <methodname>dumb_create</methodname> operation creates a GEM + object suitable for scanout based on the width, height and depth + from the struct <structname>drm_mode_create_dumb</structname> + argument. It fills the argument's <structfield>handle</structfield>, + <structfield>pitch</structfield> and <structfield>size</structfield> + fields with a handle for the newly created GEM object and its line + pitch and size in bytes. + </para> + </listitem> + <listitem> + <synopsis>int (*dumb_destroy)(struct drm_file *file_priv, struct drm_device *dev, + uint32_t handle);</synopsis> + <para> + The <methodname>dumb_destroy</methodname> operation destroys a dumb + GEM object created by <methodname>dumb_create</methodname>. + </para> + </listitem> + <listitem> + <synopsis>int (*dumb_map_offset)(struct drm_file *file_priv, struct drm_device *dev, + uint32_t handle, uint64_t *offset);</synopsis> + <para> + The <methodname>dumb_map_offset</methodname> operation associates an + mmap fake offset with the GEM object given by the handle and returns + it. Drivers must use the + <function>drm_gem_create_mmap_offset</function> function to + associate the fake offset as described in + <xref linkend="drm-gem-objects-mapping"/>. + </para> + </listitem> + </itemizedlist> + </sect3> + <sect3> + <title>Memory Coherency</title> + <para> + When mapped to the device or used in a command buffer, backing pages + for an object are flushed to memory and marked write combined so as to + be coherent with the GPU. Likewise, if the CPU accesses an object + after the GPU has finished rendering to the object, then the object + must be made coherent with the CPU's view of memory, usually involving + GPU cache flushing of various kinds. This core CPU<->GPU + coherency management is provided by a device-specific ioctl, which + evaluates an object's current domain and performs any necessary + flushing or synchronization to put the object into the desired + coherency domain (note that the object may be busy, i.e. an active + render target; in that case, setting the domain blocks the client and + waits for rendering to complete before performing any necessary + flushing operations). + </para> + </sect3> + <sect3> + <title>Command Execution</title> + <para> + Perhaps the most important GEM function for GPU devices is providing a + command execution interface to clients. Client programs construct + command buffers containing references to previously allocated memory + objects, and then submit them to GEM. At that point, GEM takes care to + bind all the objects into the GTT, execute the buffer, and provide + necessary synchronization between clients accessing the same buffers. + This often involves evicting some objects from the GTT and re-binding + others (a fairly expensive operation), and providing relocation + support which hides fixed GTT offsets from clients. Clients must take + care not to submit command buffers that reference more objects than + can fit in the GTT; otherwise, GEM will reject them and no rendering + will occur. Similarly, if several objects in the buffer require fence + registers to be allocated for correct rendering (e.g. 2D blits on + pre-965 chips), care must be taken not to require more fence registers + than are available to the client. Such resource management should be + abstracted from the client in libdrm. + </para> </sect3> </sect2> + </sect1> + + <!-- Internals: mode setting --> + <sect1 id="drm-mode-setting"> + <title>Mode Setting</title> + <para> + Drivers must initialize the mode setting core by calling + <function>drm_mode_config_init</function> on the DRM device. The function + initializes the <structname>drm_device</structname> + <structfield>mode_config</structfield> field and never fails. Once done, + mode configuration must be setup by initializing the following fields. + </para> + <itemizedlist> + <listitem> + <synopsis>int min_width, min_height; +int max_width, max_height;</synopsis> + <para> + Minimum and maximum width and height of the frame buffers in pixel + units. + </para> + </listitem> + <listitem> + <synopsis>struct drm_mode_config_funcs *funcs;</synopsis> + <para>Mode setting functions.</para> + </listitem> + </itemizedlist> <sect2> - <title>Output configuration</title> + <title>Frame Buffer Creation</title> + <synopsis>struct drm_framebuffer *(*fb_create)(struct drm_device *dev, + struct drm_file *file_priv, + struct drm_mode_fb_cmd2 *mode_cmd);</synopsis> <para> - The final initialization task is output configuration. This involves: - <itemizedlist> - <listitem> - Finding and initializing the CRTCs, encoders, and connectors - for the device. - </listitem> - <listitem> - Creating an initial configuration. - </listitem> - <listitem> - Registering a framebuffer console driver. - </listitem> - </itemizedlist> + Frame buffers are abstract memory objects that provide a source of + pixels to scanout to a CRTC. Applications explicitly request the + creation of frame buffers through the DRM_IOCTL_MODE_ADDFB(2) ioctls and + receive an opaque handle that can be passed to the KMS CRTC control, + plane configuration and page flip functions. + </para> + <para> + Frame buffers rely on the underneath memory manager for low-level memory + operations. When creating a frame buffer applications pass a memory + handle (or a list of memory handles for multi-planar formats) through + the <parameter>drm_mode_fb_cmd2</parameter> argument. This document + assumes that the driver uses GEM, those handles thus reference GEM + objects. + </para> + <para> + Drivers must first validate the requested frame buffer parameters passed + through the mode_cmd argument. In particular this is where invalid + sizes, pixel formats or pitches can be caught. + </para> + <para> + If the parameters are deemed valid, drivers then create, initialize and + return an instance of struct <structname>drm_framebuffer</structname>. + If desired the instance can be embedded in a larger driver-specific + structure. The new instance is initialized with a call to + <function>drm_framebuffer_init</function> which takes a pointer to DRM + frame buffer operations (struct + <structname>drm_framebuffer_funcs</structname>). Frame buffer operations are + <itemizedlist> + <listitem> + <synopsis>int (*create_handle)(struct drm_framebuffer *fb, + struct drm_file *file_priv, unsigned int *handle);</synopsis> + <para> + Create a handle to the frame buffer underlying memory object. If + the frame buffer uses a multi-plane format, the handle will + reference the memory object associated with the first plane. + </para> + <para> + Drivers call <function>drm_gem_handle_create</function> to create + the handle. + </para> + </listitem> + <listitem> + <synopsis>void (*destroy)(struct drm_framebuffer *framebuffer);</synopsis> + <para> + Destroy the frame buffer object and frees all associated + resources. Drivers must call + <function>drm_framebuffer_cleanup</function> to free resources + allocated by the DRM core for the frame buffer object, and must + make sure to unreference all memory objects associated with the + frame buffer. Handles created by the + <methodname>create_handle</methodname> operation are released by + the DRM core. + </para> + </listitem> + <listitem> + <synopsis>int (*dirty)(struct drm_framebuffer *framebuffer, + struct drm_file *file_priv, unsigned flags, unsigned color, + struct drm_clip_rect *clips, unsigned num_clips);</synopsis> + <para> + This optional operation notifies the driver that a region of the + frame buffer has changed in response to a DRM_IOCTL_MODE_DIRTYFB + ioctl call. + </para> + </listitem> + </itemizedlist> + </para> + <para> + After initializing the <structname>drm_framebuffer</structname> + instance drivers must fill its <structfield>width</structfield>, + <structfield>height</structfield>, <structfield>pitches</structfield>, + <structfield>offsets</structfield>, <structfield>depth</structfield>, + <structfield>bits_per_pixel</structfield> and + <structfield>pixel_format</structfield> fields from the values passed + through the <parameter>drm_mode_fb_cmd2</parameter> argument. They + should call the <function>drm_helper_mode_fill_fb_struct</function> + helper function to do so. + </para> + </sect2> + <sect2> + <title>Output Polling</title> + <synopsis>void (*output_poll_changed)(struct drm_device *dev);</synopsis> + <para> + This operation notifies the driver that the status of one or more + connectors has changed. Drivers that use the fb helper can just call the + <function>drm_fb_helper_hotplug_event</function> function to handle this + operation. + </para> + </sect2> + </sect1> + + <!-- Internals: kms initialization and cleanup --> + + <sect1 id="drm-kms-init"> + <title>KMS Initialization and Cleanup</title> + <para> + A KMS device is abstracted and exposed as a set of planes, CRTCs, encoders + and connectors. KMS drivers must thus create and initialize all those + objects at load time after initializing mode setting. + </para> + <sect2> + <title>CRTCs (struct <structname>drm_crtc</structname>)</title> + <para> + A CRTC is an abstraction representing a part of the chip that contains a + pointer to a scanout buffer. Therefore, the number of CRTCs available + determines how many independent scanout buffers can be active at any + given time. The CRTC structure contains several fields to support this: + a pointer to some video memory (abstracted as a frame buffer object), a + display mode, and an (x, y) offset into the video memory to support + panning or configurations where one piece of video memory spans multiple + CRTCs. </para> <sect3> - <title>Output discovery and initialization</title> - <para> - Several core functions exist to create CRTCs, encoders, and - connectors, namely: drm_crtc_init(), drm_connector_init(), and - drm_encoder_init(), along with several "helper" functions to - perform common tasks. - </para> - <para> - Connectors should be registered with sysfs once they've been - detected and initialized, using the - drm_sysfs_connector_add() function. Likewise, when they're - removed from the system, they should be destroyed with - drm_sysfs_connector_remove(). - </para> - <programlisting> -<![CDATA[ + <title>CRTC Initialization</title> + <para> + A KMS device must create and register at least one struct + <structname>drm_crtc</structname> instance. The instance is allocated + and zeroed by the driver, possibly as part of a larger structure, and + registered with a call to <function>drm_crtc_init</function> with a + pointer to CRTC functions. + </para> + </sect3> + <sect3> + <title>CRTC Operations</title> + <sect4> + <title>Set Configuration</title> + <synopsis>int (*set_config)(struct drm_mode_set *set);</synopsis> + <para> + Apply a new CRTC configuration to the device. The configuration + specifies a CRTC, a frame buffer to scan out from, a (x,y) position in + the frame buffer, a display mode and an array of connectors to drive + with the CRTC if possible. + </para> + <para> + If the frame buffer specified in the configuration is NULL, the driver + must detach all encoders connected to the CRTC and all connectors + attached to those encoders and disable them. + </para> + <para> + This operation is called with the mode config lock held. + </para> + <note><para> + FIXME: How should set_config interact with DPMS? If the CRTC is + suspended, should it be resumed? + </para></note> + </sect4> + <sect4> + <title>Page Flipping</title> + <synopsis>int (*page_flip)(struct drm_crtc *crtc, struct drm_framebuffer *fb, + struct drm_pending_vblank_event *event);</synopsis> + <para> + Schedule a page flip to the given frame buffer for the CRTC. This + operation is called with the mode config mutex held. + </para> + <para> + Page flipping is a synchronization mechanism that replaces the frame + buffer being scanned out by the CRTC with a new frame buffer during + vertical blanking, avoiding tearing. When an application requests a page + flip the DRM core verifies that the new frame buffer is large enough to + be scanned out by the CRTC in the currently configured mode and then + calls the CRTC <methodname>page_flip</methodname> operation with a + pointer to the new frame buffer. + </para> + <para> + The <methodname>page_flip</methodname> operation schedules a page flip. + Once any pending rendering targetting the new frame buffer has + completed, the CRTC will be reprogrammed to display that frame buffer + after the next vertical refresh. The operation must return immediately + without waiting for rendering or page flip to complete and must block + any new rendering to the frame buffer until the page flip completes. + </para> + <para> + If a page flip is already pending, the + <methodname>page_flip</methodname> operation must return + -<errorname>EBUSY</errorname>. + </para> + <para> + To synchronize page flip to vertical blanking the driver will likely + need to enable vertical blanking interrupts. It should call + <function>drm_vblank_get</function> for that purpose, and call + <function>drm_vblank_put</function> after the page flip completes. + </para> + <para> + If the application has requested to be notified when page flip completes + the <methodname>page_flip</methodname> operation will be called with a + non-NULL <parameter>event</parameter> argument pointing to a + <structname>drm_pending_vblank_event</structname> instance. Upon page + flip completion the driver must fill the + <parameter>event</parameter>::<structfield>event</structfield> + <structfield>sequence</structfield>, <structfield>tv_sec</structfield> + and <structfield>tv_usec</structfield> fields with the associated + vertical blanking count and timestamp, add the event to the + <parameter>drm_file</parameter> list of events to be signaled, and wake + up any waiting process. This can be performed with + <programlisting><![CDATA[ + struct timeval now; + + event->event.sequence = drm_vblank_count_and_time(..., &now); + event->event.tv_sec = now.tv_sec; + event->event.tv_usec = now.tv_usec; + + spin_lock_irqsave(&dev->event_lock, flags); + list_add_tail(&event->base.link, &event->base.file_priv->event_list); + wake_up_interruptible(&event->base.file_priv->event_wait); + spin_unlock_irqrestore(&dev->event_lock, flags); + ]]></programlisting> + </para> + <note><para> + FIXME: Could drivers that don't need to wait for rendering to complete + just add the event to <literal>dev->vblank_event_list</literal> and + let the DRM core handle everything, as for "normal" vertical blanking + events? + </para></note> + <para> + While waiting for the page flip to complete, the + <literal>event->base.link</literal> list head can be used freely by + the driver to store the pending event in a driver-specific list. + </para> + <para> + If the file handle is closed before the event is signaled, drivers must + take care to destroy the event in their + <methodname>preclose</methodname> operation (and, if needed, call + <function>drm_vblank_put</function>). + </para> + </sect4> + <sect4> + <title>Miscellaneous</title> + <itemizedlist> + <listitem> + <synopsis>void (*gamma_set)(struct drm_crtc *crtc, u16 *r, u16 *g, u16 *b, + uint32_t start, uint32_t size);</synopsis> + <para> + Apply a gamma table to the device. The operation is optional. + </para> + </listitem> + <listitem> + <synopsis>void (*destroy)(struct drm_crtc *crtc);</synopsis> + <para> + Destroy the CRTC when not needed anymore. See + <xref linkend="drm-kms-init"/>. + </para> + </listitem> + </itemizedlist> + </sect4> + </sect3> + </sect2> + <sect2> + <title>Planes (struct <structname>drm_plane</structname>)</title> + <para> + A plane represents an image source that can be blended with or overlayed + on top of a CRTC during the scanout process. Planes are associated with + a frame buffer to crop a portion of the image memory (source) and + optionally scale it to a destination size. The result is then blended + with or overlayed on top of a CRTC. + </para> + <sect3> + <title>Plane Initialization</title> + <para> + Planes are optional. To create a plane, a KMS drivers allocates and + zeroes an instances of struct <structname>drm_plane</structname> + (possibly as part of a larger structure) and registers it with a call + to <function>drm_plane_init</function>. The function takes a bitmask + of the CRTCs that can be associated with the plane, a pointer to the + plane functions and a list of format supported formats. + </para> + </sect3> + <sect3> + <title>Plane Operations</title> + <itemizedlist> + <listitem> + <synopsis>int (*update_plane)(struct drm_plane *plane, struct drm_crtc *crtc, + struct drm_framebuffer *fb, int crtc_x, int crtc_y, + unsigned int crtc_w, unsigned int crtc_h, + uint32_t src_x, uint32_t src_y, + uint32_t src_w, uint32_t src_h);</synopsis> + <para> + Enable and configure the plane to use the given CRTC and frame buffer. + </para> + <para> + The source rectangle in frame buffer memory coordinates is given by + the <parameter>src_x</parameter>, <parameter>src_y</parameter>, + <parameter>src_w</parameter> and <parameter>src_h</parameter> + parameters (as 16.16 fixed point values). Devices that don't support + subpixel plane coordinates can ignore the fractional part. + </para> + <para> + The destination rectangle in CRTC coordinates is given by the + <parameter>crtc_x</parameter>, <parameter>crtc_y</parameter>, + <parameter>crtc_w</parameter> and <parameter>crtc_h</parameter> + parameters (as integer values). Devices scale the source rectangle to + the destination rectangle. If scaling is not supported, and the source + rectangle size doesn't match the destination rectangle size, the + driver must return a -<errorname>EINVAL</errorname> error. + </para> + </listitem> + <listitem> + <synopsis>int (*disable_plane)(struct drm_plane *plane);</synopsis> + <para> + Disable the plane. The DRM core calls this method in response to a + DRM_IOCTL_MODE_SETPLANE ioctl call with the frame buffer ID set to 0. + Disabled planes must not be processed by the CRTC. + </para> + </listitem> + <listitem> + <synopsis>void (*destroy)(struct drm_plane *plane);</synopsis> + <para> + Destroy the plane when not needed anymore. See + <xref linkend="drm-kms-init"/>. + </para> + </listitem> + </itemizedlist> + </sect3> + </sect2> + <sect2> + <title>Encoders (struct <structname>drm_encoder</structname>)</title> + <para> + An encoder takes pixel data from a CRTC and converts it to a format + suitable for any attached connectors. On some devices, it may be + possible to have a CRTC send data to more than one encoder. In that + case, both encoders would receive data from the same scanout buffer, + resulting in a "cloned" display configuration across the connectors + attached to each encoder. + </para> + <sect3> + <title>Encoder Initialization</title> + <para> + As for CRTCs, a KMS driver must create, initialize and register at + least one struct <structname>drm_encoder</structname> instance. The + instance is allocated and zeroed by the driver, possibly as part of a + larger structure. + </para> + <para> + Drivers must initialize the struct <structname>drm_encoder</structname> + <structfield>possible_crtcs</structfield> and + <structfield>possible_clones</structfield> fields before registering the + encoder. Both fields are bitmasks of respectively the CRTCs that the + encoder can be connected to, and sibling encoders candidate for cloning. + </para> + <para> + After being initialized, the encoder must be registered with a call to + <function>drm_encoder_init</function>. The function takes a pointer to + the encoder functions and an encoder type. Supported types are + <itemizedlist> + <listitem> + DRM_MODE_ENCODER_DAC for VGA and analog on DVI-I/DVI-A + </listitem> + <listitem> + DRM_MODE_ENCODER_TMDS for DVI, HDMI and (embedded) DisplayPort + </listitem> + <listitem> + DRM_MODE_ENCODER_LVDS for display panels + </listitem> + <listitem> + DRM_MODE_ENCODER_TVDAC for TV output (Composite, S-Video, Component, + SCART) + </listitem> + <listitem> + DRM_MODE_ENCODER_VIRTUAL for virtual machine displays + </listitem> + </itemizedlist> + </para> + <para> + Encoders must be attached to a CRTC to be used. DRM drivers leave + encoders unattached at initialization time. Applications (or the fbdev + compatibility layer when implemented) are responsible for attaching the + encoders they want to use to a CRTC. + </para> + </sect3> + <sect3> + <title>Encoder Operations</title> + <itemizedlist> + <listitem> + <synopsis>void (*destroy)(struct drm_encoder *encoder);</synopsis> + <para> + Called to destroy the encoder when not needed anymore. See + <xref linkend="drm-kms-init"/>. + </para> + </listitem> + </itemizedlist> + </sect3> + </sect2> + <sect2> + <title>Connectors (struct <structname>drm_connector</structname>)</title> + <para> + A connector is the final destination for pixel data on a device, and + usually connects directly to an external display device like a monitor + or laptop panel. A connector can only be attached to one encoder at a + time. The connector is also the structure where information about the + attached display is kept, so it contains fields for display data, EDID + data, DPMS & connection status, and information about modes + supported on the attached displays. + </para> + <sect3> + <title>Connector Initialization</title> + <para> + Finally a KMS driver must create, initialize, register and attach at + least one struct <structname>drm_connector</structname> instance. The + instance is created as other KMS objects and initialized by setting the + following fields. + </para> + <variablelist> + <varlistentry> + <term><structfield>interlace_allowed</structfield></term> + <listitem><para> + Whether the connector can handle interlaced modes. + </para></listitem> + </varlistentry> + <varlistentry> + <term><structfield>doublescan_allowed</structfield></term> + <listitem><para> + Whether the connector can handle doublescan. + </para></listitem> + </varlistentry> + <varlistentry> + <term><structfield>display_info + </structfield></term> + <listitem><para> + Display information is filled from EDID information when a display + is detected. For non hot-pluggable displays such as flat panels in + embedded systems, the driver should initialize the + <structfield>display_info</structfield>.<structfield>width_mm</structfield> + and + <structfield>display_info</structfield>.<structfield>height_mm</structfield> + fields with the physical size of the display. + </para></listitem> + </varlistentry> + <varlistentry> + <term id="drm-kms-connector-polled"><structfield>polled</structfield></term> + <listitem><para> + Connector polling mode, a combination of + <variablelist> + <varlistentry> + <term>DRM_CONNECTOR_POLL_HPD</term> + <listitem><para> + The connector generates hotplug events and doesn't need to be + periodically polled. The CONNECT and DISCONNECT flags must not + be set together with the HPD flag. + </para></listitem> + </varlistentry> + <varlistentry> + <term>DRM_CONNECTOR_POLL_CONNECT</term> + <listitem><para> + Periodically poll the connector for connection. + </para></listitem> + </varlistentry> + <varlistentry> + <term>DRM_CONNECTOR_POLL_DISCONNECT</term> + <listitem><para> + Periodically poll the connector for disconnection. + </para></listitem> + </varlistentry> + </variablelist> + Set to 0 for connectors that don't support connection status + discovery. + </para></listitem> + </varlistentry> + </variablelist> + <para> + The connector is then registered with a call to + <function>drm_connector_init</function> with a pointer to the connector + functions and a connector type, and exposed through sysfs with a call to + <function>drm_sysfs_connector_add</function>. + </para> + <para> + Supported connector types are + <itemizedlist> + <listitem>DRM_MODE_CONNECTOR_VGA</listitem> + <listitem>DRM_MODE_CONNECTOR_DVII</listitem> + <listitem>DRM_MODE_CONNECTOR_DVID</listitem> + <listitem>DRM_MODE_CONNECTOR_DVIA</listitem> + <listitem>DRM_MODE_CONNECTOR_Composite</listitem> + <listitem>DRM_MODE_CONNECTOR_SVIDEO</listitem> + <listitem>DRM_MODE_CONNECTOR_LVDS</listitem> + <listitem>DRM_MODE_CONNECTOR_Component</listitem> + <listitem>DRM_MODE_CONNECTOR_9PinDIN</listitem> + <listitem>DRM_MODE_CONNECTOR_DisplayPort</listitem> + <listitem>DRM_MODE_CONNECTOR_HDMIA</listitem> + <listitem>DRM_MODE_CONNECTOR_HDMIB</listitem> + <listitem>DRM_MODE_CONNECTOR_TV</listitem> + <listitem>DRM_MODE_CONNECTOR_eDP</listitem> + <listitem>DRM_MODE_CONNECTOR_VIRTUAL</listitem> + </itemizedlist> + </para> + <para> + Connectors must be attached to an encoder to be used. For devices that + map connectors to encoders 1:1, the connector should be attached at + initialization time with a call to + <function>drm_mode_connector_attach_encoder</function>. The driver must + also set the <structname>drm_connector</structname> + <structfield>encoder</structfield> field to point to the attached + encoder. + </para> + <para> + Finally, drivers must initialize the connectors state change detection + with a call to <function>drm_kms_helper_poll_init</function>. If at + least one connector is pollable but can't generate hotplug interrupts + (indicated by the DRM_CONNECTOR_POLL_CONNECT and + DRM_CONNECTOR_POLL_DISCONNECT connector flags), a delayed work will + automatically be queued to periodically poll for changes. Connectors + that can generate hotplug interrupts must be marked with the + DRM_CONNECTOR_POLL_HPD flag instead, and their interrupt handler must + call <function>drm_helper_hpd_irq_event</function>. The function will + queue a delayed work to check the state of all connectors, but no + periodic polling will be done. + </para> + </sect3> + <sect3> + <title>Connector Operations</title> + <note><para> + Unless otherwise state, all operations are mandatory. + </para></note> + <sect4> + <title>DPMS</title> + <synopsis>void (*dpms)(struct drm_connector *connector, int mode);</synopsis> + <para> + The DPMS operation sets the power state of a connector. The mode + argument is one of + <itemizedlist> + <listitem><para>DRM_MODE_DPMS_ON</para></listitem> + <listitem><para>DRM_MODE_DPMS_STANDBY</para></listitem> + <listitem><para>DRM_MODE_DPMS_SUSPEND</para></listitem> + <listitem><para>DRM_MODE_DPMS_OFF</para></listitem> + </itemizedlist> + </para> + <para> + In all but DPMS_ON mode the encoder to which the connector is attached + should put the display in low-power mode by driving its signals + appropriately. If more than one connector is attached to the encoder + care should be taken not to change the power state of other displays as + a side effect. Low-power mode should be propagated to the encoders and + CRTCs when all related connectors are put in low-power mode. + </para> + </sect4> + <sect4> + <title>Modes</title> + <synopsis>int (*fill_modes)(struct drm_connector *connector, uint32_t max_width, + uint32_t max_height);</synopsis> + <para> + Fill the mode list with all supported modes for the connector. If the + <parameter>max_width</parameter> and <parameter>max_height</parameter> + arguments are non-zero, the implementation must ignore all modes wider + than <parameter>max_width</parameter> or higher than + <parameter>max_height</parameter>. + </para> + <para> + The connector must also fill in this operation its + <structfield>display_info</structfield> + <structfield>width_mm</structfield> and + <structfield>height_mm</structfield> fields with the connected display + physical size in millimeters. The fields should be set to 0 if the value + isn't known or is not applicable (for instance for projector devices). + </para> + </sect4> + <sect4> + <title>Connection Status</title> + <para> + The connection status is updated through polling or hotplug events when + supported (see <xref linkend="drm-kms-connector-polled"/>). The status + value is reported to userspace through ioctls and must not be used + inside the driver, as it only gets initialized by a call to + <function>drm_mode_getconnector</function> from userspace. + </para> + <synopsis>enum drm_connector_status (*detect)(struct drm_connector *connector, + bool force);</synopsis> + <para> + Check to see if anything is attached to the connector. The + <parameter>force</parameter> parameter is set to false whilst polling or + to true when checking the connector due to user request. + <parameter>force</parameter> can be used by the driver to avoid + expensive, destructive operations during automated probing. + </para> + <para> + Return connector_status_connected if something is connected to the + connector, connector_status_disconnected if nothing is connected and + connector_status_unknown if the connection state isn't known. + </para> + <para> + Drivers should only return connector_status_connected if the connection + status has really been probed as connected. Connectors that can't detect + the connection status, or failed connection status probes, should return + connector_status_unknown. + </para> + </sect4> + <sect4> + <title>Miscellaneous</title> + <itemizedlist> + <listitem> + <synopsis>void (*destroy)(struct drm_connector *connector);</synopsis> + <para> + Destroy the connector when not needed anymore. See + <xref linkend="drm-kms-init"/>. + </para> + </listitem> + </itemizedlist> + </sect4> + </sect3> + </sect2> + <sect2> + <title>Cleanup</title> + <para> + The DRM core manages its objects' lifetime. When an object is not needed + anymore the core calls its destroy function, which must clean up and + free every resource allocated for the object. Every + <function>drm_*_init</function> call must be matched with a + corresponding <function>drm_*_cleanup</function> call to cleanup CRTCs + (<function>drm_crtc_cleanup</function>), planes + (<function>drm_plane_cleanup</function>), encoders + (<function>drm_encoder_cleanup</function>) and connectors + (<function>drm_connector_cleanup</function>). Furthermore, connectors + that have been added to sysfs must be removed by a call to + <function>drm_sysfs_connector_remove</function> before calling + <function>drm_connector_cleanup</function>. + </para> + <para> + Connectors state change detection must be cleanup up with a call to + <function>drm_kms_helper_poll_fini</function>. + </para> + </sect2> + <sect2> + <title>Output discovery and initialization example</title> + <programlisting><![CDATA[ void intel_crt_init(struct drm_device *dev) { struct drm_connector *connector; @@ -556,252 +1610,741 @@ void intel_crt_init(struct drm_device *dev) drm_connector_helper_add(connector, &intel_crt_connector_helper_funcs); drm_sysfs_connector_add(connector); -} -]]> - </programlisting> - <para> - In the example above (again, taken from the i915 driver), a - CRT connector and encoder combination is created. A device-specific - i2c bus is also created for fetching EDID data and - performing monitor detection. Once the process is complete, - the new connector is registered with sysfs to make its - properties available to applications. - </para> - <sect4> - <title>Helper functions and core functions</title> - <para> - Since many PC-class graphics devices have similar display output - designs, the DRM provides a set of helper functions to make - output management easier. The core helper routines handle - encoder re-routing and the disabling of unused functions following - mode setting. Using the helpers is optional, but recommended for - devices with PC-style architectures (i.e. a set of display planes - for feeding pixels to encoders which are in turn routed to - connectors). Devices with more complex requirements needing - finer grained management may opt to use the core callbacks - directly. - </para> - <para> - [Insert typical diagram here.] [Insert OMAP style config here.] - </para> - </sect4> - <para> - Each encoder object needs to provide: - <itemizedlist> - <listitem> - A DPMS (basically on/off) function. - </listitem> - <listitem> - A mode-fixup function (for converting requested modes into - native hardware timings). - </listitem> - <listitem> - Functions (prepare, set, and commit) for use by the core DRM - helper functions. - </listitem> - </itemizedlist> - Connector helpers need to provide functions (mode-fetch, validity, - and encoder-matching) for returning an ideal encoder for a given - connector. The core connector functions include a DPMS callback, - save/restore routines (deprecated), detection, mode probing, - property handling, and cleanup functions. - </para> -<!--!Edrivers/char/drm/drm_crtc.h--> -<!--!Edrivers/char/drm/drm_crtc.c--> -<!--!Edrivers/char/drm/drm_crtc_helper.c--> - </sect3> +}]]></programlisting> + <para> + In the example above (taken from the i915 driver), a CRTC, connector and + encoder combination is created. A device-specific i2c bus is also + created for fetching EDID data and performing monitor detection. Once + the process is complete, the new connector is registered with sysfs to + make its properties available to applications. + </para> </sect2> </sect1> - <!-- Internals: vblank handling --> + <!-- Internals: mid-layer helper functions --> <sect1> - <title>VBlank event handling</title> + <title>Mid-layer Helper Functions</title> <para> - The DRM core exposes two vertical blank related ioctls: - <variablelist> - <varlistentry> - <term>DRM_IOCTL_WAIT_VBLANK</term> - <listitem> - <para> - This takes a struct drm_wait_vblank structure as its argument, - and it is used to block or request a signal when a specified - vblank event occurs. - </para> - </listitem> - </varlistentry> - <varlistentry> - <term>DRM_IOCTL_MODESET_CTL</term> - <listitem> - <para> - This should be called by application level drivers before and - after mode setting, since on many devices the vertical blank - counter is reset at that time. Internally, the DRM snapshots - the last vblank count when the ioctl is called with the - _DRM_PRE_MODESET command, so that the counter won't go backwards - (which is dealt with when _DRM_POST_MODESET is used). - </para> - </listitem> - </varlistentry> - </variablelist> -<!--!Edrivers/char/drm/drm_irq.c--> + The CRTC, encoder and connector functions provided by the drivers + implement the DRM API. They're called by the DRM core and ioctl handlers + to handle device state changes and configuration request. As implementing + those functions often requires logic not specific to drivers, mid-layer + helper functions are available to avoid duplicating boilerplate code. + </para> + <para> + The DRM core contains one mid-layer implementation. The mid-layer provides + implementations of several CRTC, encoder and connector functions (called + from the top of the mid-layer) that pre-process requests and call + lower-level functions provided by the driver (at the bottom of the + mid-layer). For instance, the + <function>drm_crtc_helper_set_config</function> function can be used to + fill the struct <structname>drm_crtc_funcs</structname> + <structfield>set_config</structfield> field. When called, it will split + the <methodname>set_config</methodname> operation in smaller, simpler + operations and call the driver to handle them. </para> <para> - To support the functions above, the DRM core provides several - helper functions for tracking vertical blank counters, and - requires drivers to provide several callbacks: - get_vblank_counter(), enable_vblank() and disable_vblank(). The - core uses get_vblank_counter() to keep the counter accurate - across interrupt disable periods. It should return the current - vertical blank event count, which is often tracked in a device - register. The enable and disable vblank callbacks should enable - and disable vertical blank interrupts, respectively. In the - absence of DRM clients waiting on vblank events, the core DRM - code uses the disable_vblank() function to disable - interrupts, which saves power. They are re-enabled again when - a client calls the vblank wait ioctl above. + To use the mid-layer, drivers call <function>drm_crtc_helper_add</function>, + <function>drm_encoder_helper_add</function> and + <function>drm_connector_helper_add</function> functions to install their + mid-layer bottom operations handlers, and fill the + <structname>drm_crtc_funcs</structname>, + <structname>drm_encoder_funcs</structname> and + <structname>drm_connector_funcs</structname> structures with pointers to + the mid-layer top API functions. Installing the mid-layer bottom operation + handlers is best done right after registering the corresponding KMS object. </para> <para> - A device that doesn't provide a count register may simply use an - internal atomic counter incremented on every vertical blank - interrupt (and then treat the enable_vblank() and disable_vblank() - callbacks as no-ops). + The mid-layer is not split between CRTC, encoder and connector operations. + To use it, a driver must provide bottom functions for all of the three KMS + entities. </para> + <sect2> + <title>Helper Functions</title> + <itemizedlist> + <listitem> + <synopsis>int drm_crtc_helper_set_config(struct drm_mode_set *set);</synopsis> + <para> + The <function>drm_crtc_helper_set_config</function> helper function + is a CRTC <methodname>set_config</methodname> implementation. It + first tries to locate the best encoder for each connector by calling + the connector <methodname>best_encoder</methodname> helper + operation. + </para> + <para> + After locating the appropriate encoders, the helper function will + call the <methodname>mode_fixup</methodname> encoder and CRTC helper + operations to adjust the requested mode, or reject it completely in + which case an error will be returned to the application. If the new + configuration after mode adjustment is identical to the current + configuration the helper function will return without performing any + other operation. + </para> + <para> + If the adjusted mode is identical to the current mode but changes to + the frame buffer need to be applied, the + <function>drm_crtc_helper_set_config</function> function will call + the CRTC <methodname>mode_set_base</methodname> helper operation. If + the adjusted mode differs from the current mode, or if the + <methodname>mode_set_base</methodname> helper operation is not + provided, the helper function performs a full mode set sequence by + calling the <methodname>prepare</methodname>, + <methodname>mode_set</methodname> and + <methodname>commit</methodname> CRTC and encoder helper operations, + in that order. + </para> + </listitem> + <listitem> + <synopsis>void drm_helper_connector_dpms(struct drm_connector *connector, int mode);</synopsis> + <para> + The <function>drm_helper_connector_dpms</function> helper function + is a connector <methodname>dpms</methodname> implementation that + tracks power state of connectors. To use the function, drivers must + provide <methodname>dpms</methodname> helper operations for CRTCs + and encoders to apply the DPMS state to the device. + </para> + <para> + The mid-layer doesn't track the power state of CRTCs and encoders. + The <methodname>dpms</methodname> helper operations can thus be + called with a mode identical to the currently active mode. + </para> + </listitem> + <listitem> + <synopsis>int drm_helper_probe_single_connector_modes(struct drm_connector *connector, + uint32_t maxX, uint32_t maxY);</synopsis> + <para> + The <function>drm_helper_probe_single_connector_modes</function> helper + function is a connector <methodname>fill_modes</methodname> + implementation that updates the connection status for the connector + and then retrieves a list of modes by calling the connector + <methodname>get_modes</methodname> helper operation. + </para> + <para> + The function filters out modes larger than + <parameter>max_width</parameter> and <parameter>max_height</parameter> + if specified. It then calls the connector + <methodname>mode_valid</methodname> helper operation for each mode in + the probed list to check whether the mode is valid for the connector. + </para> + </listitem> + </itemizedlist> + </sect2> + <sect2> + <title>CRTC Helper Operations</title> + <itemizedlist> + <listitem id="drm-helper-crtc-mode-fixup"> + <synopsis>bool (*mode_fixup)(struct drm_crtc *crtc, + const struct drm_display_mode *mode, + struct drm_display_mode *adjusted_mode);</synopsis> + <para> + Let CRTCs adjust the requested mode or reject it completely. This + operation returns true if the mode is accepted (possibly after being + adjusted) or false if it is rejected. + </para> + <para> + The <methodname>mode_fixup</methodname> operation should reject the + mode if it can't reasonably use it. The definition of "reasonable" + is currently fuzzy in this context. One possible behaviour would be + to set the adjusted mode to the panel timings when a fixed-mode + panel is used with hardware capable of scaling. Another behaviour + would be to accept any input mode and adjust it to the closest mode + supported by the hardware (FIXME: This needs to be clarified). + </para> + </listitem> + <listitem> + <synopsis>int (*mode_set_base)(struct drm_crtc *crtc, int x, int y, + struct drm_framebuffer *old_fb)</synopsis> + <para> + Move the CRTC on the current frame buffer (stored in + <literal>crtc->fb</literal>) to position (x,y). Any of the frame + buffer, x position or y position may have been modified. + </para> + <para> + This helper operation is optional. If not provided, the + <function>drm_crtc_helper_set_config</function> function will fall + back to the <methodname>mode_set</methodname> helper operation. + </para> + <note><para> + FIXME: Why are x and y passed as arguments, as they can be accessed + through <literal>crtc->x</literal> and + <literal>crtc->y</literal>? + </para></note> + </listitem> + <listitem> + <synopsis>void (*prepare)(struct drm_crtc *crtc);</synopsis> + <para> + Prepare the CRTC for mode setting. This operation is called after + validating the requested mode. Drivers use it to perform + device-specific operations required before setting the new mode. + </para> + </listitem> + <listitem> + <synopsis>int (*mode_set)(struct drm_crtc *crtc, struct drm_display_mode *mode, + struct drm_display_mode *adjusted_mode, int x, int y, + struct drm_framebuffer *old_fb);</synopsis> + <para> + Set a new mode, position and frame buffer. Depending on the device + requirements, the mode can be stored internally by the driver and + applied in the <methodname>commit</methodname> operation, or + programmed to the hardware immediately. + </para> + <para> + The <methodname>mode_set</methodname> operation returns 0 on success + or a negative error code if an error occurs. + </para> + </listitem> + <listitem> + <synopsis>void (*commit)(struct drm_crtc *crtc);</synopsis> + <para> + Commit a mode. This operation is called after setting the new mode. + Upon return the device must use the new mode and be fully + operational. + </para> + </listitem> + </itemizedlist> + </sect2> + <sect2> + <title>Encoder Helper Operations</title> + <itemizedlist> + <listitem> + <synopsis>bool (*mode_fixup)(struct drm_encoder *encoder, + const struct drm_display_mode *mode, + struct drm_display_mode *adjusted_mode);</synopsis> + <note><para> + FIXME: The mode argument be const, but the i915 driver modifies + mode->clock in <function>intel_dp_mode_fixup</function>. + </para></note> + <para> + Let encoders adjust the requested mode or reject it completely. This + operation returns true if the mode is accepted (possibly after being + adjusted) or false if it is rejected. See the + <link linkend="drm-helper-crtc-mode-fixup">mode_fixup CRTC helper + operation</link> for an explanation of the allowed adjustments. + </para> + </listitem> + <listitem> + <synopsis>void (*prepare)(struct drm_encoder *encoder);</synopsis> + <para> + Prepare the encoder for mode setting. This operation is called after + validating the requested mode. Drivers use it to perform + device-specific operations required before setting the new mode. + </para> + </listitem> + <listitem> + <synopsis>void (*mode_set)(struct drm_encoder *encoder, + struct drm_display_mode *mode, + struct drm_display_mode *adjusted_mode);</synopsis> + <para> + Set a new mode. Depending on the device requirements, the mode can + be stored internally by the driver and applied in the + <methodname>commit</methodname> operation, or programmed to the + hardware immediately. + </para> + </listitem> + <listitem> + <synopsis>void (*commit)(struct drm_encoder *encoder);</synopsis> + <para> + Commit a mode. This operation is called after setting the new mode. + Upon return the device must use the new mode and be fully + operational. + </para> + </listitem> + </itemizedlist> + </sect2> + <sect2> + <title>Connector Helper Operations</title> + <itemizedlist> + <listitem> + <synopsis>struct drm_encoder *(*best_encoder)(struct drm_connector *connector);</synopsis> + <para> + Return a pointer to the best encoder for the connecter. Device that + map connectors to encoders 1:1 simply return the pointer to the + associated encoder. This operation is mandatory. + </para> + </listitem> + <listitem> + <synopsis>int (*get_modes)(struct drm_connector *connector);</synopsis> + <para> + Fill the connector's <structfield>probed_modes</structfield> list + by parsing EDID data with <function>drm_add_edid_modes</function> or + calling <function>drm_mode_probed_add</function> directly for every + supported mode and return the number of modes it has detected. This + operation is mandatory. + </para> + <para> + When adding modes manually the driver creates each mode with a call to + <function>drm_mode_create</function> and must fill the following fields. + <itemizedlist> + <listitem> + <synopsis>__u32 type;</synopsis> + <para> + Mode type bitmask, a combination of + <variablelist> + <varlistentry> + <term>DRM_MODE_TYPE_BUILTIN</term> + <listitem><para>not used?</para></listitem> + </varlistentry> + <varlistentry> + <term>DRM_MODE_TYPE_CLOCK_C</term> + <listitem><para>not used?</para></listitem> + </varlistentry> + <varlistentry> + <term>DRM_MODE_TYPE_CRTC_C</term> + <listitem><para>not used?</para></listitem> + </varlistentry> + <varlistentry> + <term> + DRM_MODE_TYPE_PREFERRED - The preferred mode for the connector + </term> + <listitem> + <para>not used?</para> + </listitem> + </varlistentry> + <varlistentry> + <term>DRM_MODE_TYPE_DEFAULT</term> + <listitem><para>not used?</para></listitem> + </varlistentry> + <varlistentry> + <term>DRM_MODE_TYPE_USERDEF</term> + <listitem><para>not used?</para></listitem> + </varlistentry> + <varlistentry> + <term>DRM_MODE_TYPE_DRIVER</term> + <listitem> + <para> + The mode has been created by the driver (as opposed to + to user-created modes). + </para> + </listitem> + </varlistentry> + </variablelist> + Drivers must set the DRM_MODE_TYPE_DRIVER bit for all modes they + create, and set the DRM_MODE_TYPE_PREFERRED bit for the preferred + mode. + </para> + </listitem> + <listitem> + <synopsis>__u32 clock;</synopsis> + <para>Pixel clock frequency in kHz unit</para> + </listitem> + <listitem> + <synopsis>__u16 hdisplay, hsync_start, hsync_end, htotal; + __u16 vdisplay, vsync_start, vsync_end, vtotal;</synopsis> + <para>Horizontal and vertical timing information</para> + <screen><![CDATA[ + Active Front Sync Back + Region Porch Porch + <-----------------------><----------------><-------------><--------------> + + //////////////////////| + ////////////////////// | + ////////////////////// |.................. ................ + _______________ + + <----- [hv]display -----> + <------------- [hv]sync_start ------------> + <--------------------- [hv]sync_end ---------------------> + <-------------------------------- [hv]total -----------------------------> +]]></screen> + </listitem> + <listitem> + <synopsis>__u16 hskew; + __u16 vscan;</synopsis> + <para>Unknown</para> + </listitem> + <listitem> + <synopsis>__u32 flags;</synopsis> + <para> + Mode flags, a combination of + <variablelist> + <varlistentry> + <term>DRM_MODE_FLAG_PHSYNC</term> + <listitem><para> + Horizontal sync is active high + </para></listitem> + </varlistentry> + <varlistentry> + <term>DRM_MODE_FLAG_NHSYNC</term> + <listitem><para> + Horizontal sync is active low + </para></listitem> + </varlistentry> + <varlistentry> + <term>DRM_MODE_FLAG_PVSYNC</term> + <listitem><para> + Vertical sync is active high + </para></listitem> + </varlistentry> + <varlistentry> + <term>DRM_MODE_FLAG_NVSYNC</term> + <listitem><para> + Vertical sync is active low + </para></listitem> + </varlistentry> + <varlistentry> + <term>DRM_MODE_FLAG_INTERLACE</term> + <listitem><para> + Mode is interlaced + </para></listitem> + </varlistentry> + <varlistentry> + <term>DRM_MODE_FLAG_DBLSCAN</term> + <listitem><para> + Mode uses doublescan + </para></listitem> + </varlistentry> + <varlistentry> + <term>DRM_MODE_FLAG_CSYNC</term> + <listitem><para> + Mode uses composite sync + </para></listitem> + </varlistentry> + <varlistentry> + <term>DRM_MODE_FLAG_PCSYNC</term> + <listitem><para> + Composite sync is active high + </para></listitem> + </varlistentry> + <varlistentry> + <term>DRM_MODE_FLAG_NCSYNC</term> + <listitem><para> + Composite sync is active low + </para></listitem> + </varlistentry> + <varlistentry> + <term>DRM_MODE_FLAG_HSKEW</term> + <listitem><para> + hskew provided (not used?) + </para></listitem> + </varlistentry> + <varlistentry> + <term>DRM_MODE_FLAG_BCAST</term> + <listitem><para> + not used? + </para></listitem> + </varlistentry> + <varlistentry> + <term>DRM_MODE_FLAG_PIXMUX</term> + <listitem><para> + not used? + </para></listitem> + </varlistentry> + <varlistentry> + <term>DRM_MODE_FLAG_DBLCLK</term> + <listitem><para> + not used? + </para></listitem> + </varlistentry> + <varlistentry> + <term>DRM_MODE_FLAG_CLKDIV2</term> + <listitem><para> + ? + </para></listitem> + </varlistentry> + </variablelist> + </para> + <para> + Note that modes marked with the INTERLACE or DBLSCAN flags will be + filtered out by + <function>drm_helper_probe_single_connector_modes</function> if + the connector's <structfield>interlace_allowed</structfield> or + <structfield>doublescan_allowed</structfield> field is set to 0. + </para> + </listitem> + <listitem> + <synopsis>char name[DRM_DISPLAY_MODE_LEN];</synopsis> + <para> + Mode name. The driver must call + <function>drm_mode_set_name</function> to fill the mode name from + <structfield>hdisplay</structfield>, + <structfield>vdisplay</structfield> and interlace flag after + filling the corresponding fields. + </para> + </listitem> + </itemizedlist> + </para> + <para> + The <structfield>vrefresh</structfield> value is computed by + <function>drm_helper_probe_single_connector_modes</function>. + </para> + <para> + When parsing EDID data, <function>drm_add_edid_modes</function> fill the + connector <structfield>display_info</structfield> + <structfield>width_mm</structfield> and + <structfield>height_mm</structfield> fields. When creating modes + manually the <methodname>get_modes</methodname> helper operation must + set the <structfield>display_info</structfield> + <structfield>width_mm</structfield> and + <structfield>height_mm</structfield> fields if they haven't been set + already (for instance at initilization time when a fixed-size panel is + attached to the connector). The mode <structfield>width_mm</structfield> + and <structfield>height_mm</structfield> fields are only used internally + during EDID parsing and should not be set when creating modes manually. + </para> + </listitem> + <listitem> + <synopsis>int (*mode_valid)(struct drm_connector *connector, + struct drm_display_mode *mode);</synopsis> + <para> + Verify whether a mode is valid for the connector. Return MODE_OK for + supported modes and one of the enum drm_mode_status values (MODE_*) + for unsupported modes. This operation is mandatory. + </para> + <para> + As the mode rejection reason is currently not used beside for + immediately removing the unsupported mode, an implementation can + return MODE_BAD regardless of the exact reason why the mode is not + valid. + </para> + <note><para> + Note that the <methodname>mode_valid</methodname> helper operation is + only called for modes detected by the device, and + <emphasis>not</emphasis> for modes set by the user through the CRTC + <methodname>set_config</methodname> operation. + </para></note> + </listitem> + </itemizedlist> + </sect2> </sect1> - <sect1> - <title>Memory management</title> + <!-- Internals: vertical blanking --> + + <sect1 id="drm-vertical-blank"> + <title>Vertical Blanking</title> + <para> + Vertical blanking plays a major role in graphics rendering. To achieve + tear-free display, users must synchronize page flips and/or rendering to + vertical blanking. The DRM API offers ioctls to perform page flips + synchronized to vertical blanking and wait for vertical blanking. + </para> + <para> + The DRM core handles most of the vertical blanking management logic, which + involves filtering out spurious interrupts, keeping race-free blanking + counters, coping with counter wrap-around and resets and keeping use + counts. It relies on the driver to generate vertical blanking interrupts + and optionally provide a hardware vertical blanking counter. Drivers must + implement the following operations. + </para> + <itemizedlist> + <listitem> + <synopsis>int (*enable_vblank) (struct drm_device *dev, int crtc); +void (*disable_vblank) (struct drm_device *dev, int crtc);</synopsis> + <para> + Enable or disable vertical blanking interrupts for the given CRTC. + </para> + </listitem> + <listitem> + <synopsis>u32 (*get_vblank_counter) (struct drm_device *dev, int crtc);</synopsis> + <para> + Retrieve the value of the vertical blanking counter for the given + CRTC. If the hardware maintains a vertical blanking counter its value + should be returned. Otherwise drivers can use the + <function>drm_vblank_count</function> helper function to handle this + operation. + </para> + </listitem> + </itemizedlist> <para> - The memory manager lies at the heart of many DRM operations; it - is required to support advanced client features like OpenGL - pbuffers. The DRM currently contains two memory managers: TTM - and GEM. + Drivers must initialize the vertical blanking handling core with a call to + <function>drm_vblank_init</function> in their + <methodname>load</methodname> operation. The function will set the struct + <structname>drm_device</structname> + <structfield>vblank_disable_allowed</structfield> field to 0. This will + keep vertical blanking interrupts enabled permanently until the first mode + set operation, where <structfield>vblank_disable_allowed</structfield> is + set to 1. The reason behind this is not clear. Drivers can set the field + to 1 after <function>calling drm_vblank_init</function> to make vertical + blanking interrupts dynamically managed from the beginning. </para> + <para> + Vertical blanking interrupts can be enabled by the DRM core or by drivers + themselves (for instance to handle page flipping operations). The DRM core + maintains a vertical blanking use count to ensure that the interrupts are + not disabled while a user still needs them. To increment the use count, + drivers call <function>drm_vblank_get</function>. Upon return vertical + blanking interrupts are guaranteed to be enabled. + </para> + <para> + To decrement the use count drivers call + <function>drm_vblank_put</function>. Only when the use count drops to zero + will the DRM core disable the vertical blanking interrupts after a delay + by scheduling a timer. The delay is accessible through the vblankoffdelay + module parameter or the <varname>drm_vblank_offdelay</varname> global + variable and expressed in milliseconds. Its default value is 5000 ms. + </para> + <para> + When a vertical blanking interrupt occurs drivers only need to call the + <function>drm_handle_vblank</function> function to account for the + interrupt. + </para> + <para> + Resources allocated by <function>drm_vblank_init</function> must be freed + with a call to <function>drm_vblank_cleanup</function> in the driver + <methodname>unload</methodname> operation handler. + </para> + </sect1> + + <!-- Internals: open/close, file operations and ioctls --> + <sect1> + <title>Open/Close, File Operations and IOCTLs</title> <sect2> - <title>The Translation Table Manager (TTM)</title> + <title>Open and Close</title> + <synopsis>int (*firstopen) (struct drm_device *); +void (*lastclose) (struct drm_device *); +int (*open) (struct drm_device *, struct drm_file *); +void (*preclose) (struct drm_device *, struct drm_file *); +void (*postclose) (struct drm_device *, struct drm_file *);</synopsis> + <abstract>Open and close handlers. None of those methods are mandatory. + </abstract> <para> - TTM was developed by Tungsten Graphics, primarily by Thomas - Hellström, and is intended to be a flexible, high performance - graphics memory manager. + The <methodname>firstopen</methodname> method is called by the DRM core + when an application opens a device that has no other opened file handle. + Similarly the <methodname>lastclose</methodname> method is called when + the last application holding a file handle opened on the device closes + it. Both methods are mostly used for UMS (User Mode Setting) drivers to + acquire and release device resources which should be done in the + <methodname>load</methodname> and <methodname>unload</methodname> + methods for KMS drivers. </para> <para> - Drivers wishing to support TTM must fill out a drm_bo_driver - structure. + Note that the <methodname>lastclose</methodname> method is also called + at module unload time or, for hot-pluggable devices, when the device is + unplugged. The <methodname>firstopen</methodname> and + <methodname>lastclose</methodname> calls can thus be unbalanced. </para> <para> - TTM design background and information belongs here. + The <methodname>open</methodname> method is called every time the device + is opened by an application. Drivers can allocate per-file private data + in this method and store them in the struct + <structname>drm_file</structname> <structfield>driver_priv</structfield> + field. Note that the <methodname>open</methodname> method is called + before <methodname>firstopen</methodname>. + </para> + <para> + The close operation is split into <methodname>preclose</methodname> and + <methodname>postclose</methodname> methods. Drivers must stop and + cleanup all per-file operations in the <methodname>preclose</methodname> + method. For instance pending vertical blanking and page flip events must + be cancelled. No per-file operation is allowed on the file handle after + returning from the <methodname>preclose</methodname> method. + </para> + <para> + Finally the <methodname>postclose</methodname> method is called as the + last step of the close operation, right before calling the + <methodname>lastclose</methodname> method if no other open file handle + exists for the device. Drivers that have allocated per-file private data + in the <methodname>open</methodname> method should free it here. + </para> + <para> + The <methodname>lastclose</methodname> method should restore CRTC and + plane properties to default value, so that a subsequent open of the + device will not inherit state from the previous user. </para> </sect2> - <sect2> - <title>The Graphics Execution Manager (GEM)</title> + <title>File Operations</title> + <synopsis>const struct file_operations *fops</synopsis> + <abstract>File operations for the DRM device node.</abstract> <para> - GEM is an Intel project, authored by Eric Anholt and Keith - Packard. It provides simpler interfaces than TTM, and is well - suited for UMA devices. + Drivers must define the file operations structure that forms the DRM + userspace API entry point, even though most of those operations are + implemented in the DRM core. The <methodname>open</methodname>, + <methodname>release</methodname> and <methodname>ioctl</methodname> + operations are handled by + <programlisting> + .owner = THIS_MODULE, + .open = drm_open, + .release = drm_release, + .unlocked_ioctl = drm_ioctl, + #ifdef CONFIG_COMPAT + .compat_ioctl = drm_compat_ioctl, + #endif + </programlisting> </para> <para> - GEM-enabled drivers must provide gem_init_object() and - gem_free_object() callbacks to support the core memory - allocation routines. They should also provide several driver-specific - ioctls to support command execution, pinning, buffer - read & write, mapping, and domain ownership transfers. + Drivers that implement private ioctls that requires 32/64bit + compatibility support must provide their own + <methodname>compat_ioctl</methodname> handler that processes private + ioctls and calls <function>drm_compat_ioctl</function> for core ioctls. </para> <para> - On a fundamental level, GEM involves several operations: - <itemizedlist> - <listitem>Memory allocation and freeing</listitem> - <listitem>Command execution</listitem> - <listitem>Aperture management at command execution time</listitem> - </itemizedlist> - Buffer object allocation is relatively - straightforward and largely provided by Linux's shmem layer, which - provides memory to back each object. When mapped into the GTT - or used in a command buffer, the backing pages for an object are - flushed to memory and marked write combined so as to be coherent - with the GPU. Likewise, if the CPU accesses an object after the GPU - has finished rendering to the object, then the object must be made - coherent with the CPU's view - of memory, usually involving GPU cache flushing of various kinds. - This core CPU<->GPU coherency management is provided by a - device-specific ioctl, which evaluates an object's current domain and - performs any necessary flushing or synchronization to put the object - into the desired coherency domain (note that the object may be busy, - i.e. an active render target; in that case, setting the domain - blocks the client and waits for rendering to complete before - performing any necessary flushing operations). - </para> - <para> - Perhaps the most important GEM function is providing a command - execution interface to clients. Client programs construct command - buffers containing references to previously allocated memory objects, - and then submit them to GEM. At that point, GEM takes care to bind - all the objects into the GTT, execute the buffer, and provide - necessary synchronization between clients accessing the same buffers. - This often involves evicting some objects from the GTT and re-binding - others (a fairly expensive operation), and providing relocation - support which hides fixed GTT offsets from clients. Clients must - take care not to submit command buffers that reference more objects - than can fit in the GTT; otherwise, GEM will reject them and no rendering - will occur. Similarly, if several objects in the buffer require - fence registers to be allocated for correct rendering (e.g. 2D blits - on pre-965 chips), care must be taken not to require more fence - registers than are available to the client. Such resource management - should be abstracted from the client in libdrm. + The <methodname>read</methodname> and <methodname>poll</methodname> + operations provide support for reading DRM events and polling them. They + are implemented by + <programlisting> + .poll = drm_poll, + .read = drm_read, + .fasync = drm_fasync, + .llseek = no_llseek, + </programlisting> + </para> + <para> + The memory mapping implementation varies depending on how the driver + manages memory. Pre-GEM drivers will use <function>drm_mmap</function>, + while GEM-aware drivers will use <function>drm_gem_mmap</function>. See + <xref linkend="drm-gem"/>. + <programlisting> + .mmap = drm_gem_mmap, + </programlisting> + </para> + <para> + No other file operation is supported by the DRM API. + </para> + </sect2> + <sect2> + <title>IOCTLs</title> + <synopsis>struct drm_ioctl_desc *ioctls; +int num_ioctls;</synopsis> + <abstract>Driver-specific ioctls descriptors table.</abstract> + <para> + Driver-specific ioctls numbers start at DRM_COMMAND_BASE. The ioctls + descriptors table is indexed by the ioctl number offset from the base + value. Drivers can use the DRM_IOCTL_DEF_DRV() macro to initialize the + table entries. + </para> + <para> + <programlisting>DRM_IOCTL_DEF_DRV(ioctl, func, flags)</programlisting> + <para> + <parameter>ioctl</parameter> is the ioctl name. Drivers must define + the DRM_##ioctl and DRM_IOCTL_##ioctl macros to the ioctl number + offset from DRM_COMMAND_BASE and the ioctl number respectively. The + first macro is private to the device while the second must be exposed + to userspace in a public header. + </para> + <para> + <parameter>func</parameter> is a pointer to the ioctl handler function + compatible with the <type>drm_ioctl_t</type> type. + <programlisting>typedef int drm_ioctl_t(struct drm_device *dev, void *data, + struct drm_file *file_priv);</programlisting> + </para> + <para> + <parameter>flags</parameter> is a bitmask combination of the following + values. It restricts how the ioctl is allowed to be called. + <itemizedlist> + <listitem><para> + DRM_AUTH - Only authenticated callers allowed + </para></listitem> + <listitem><para> + DRM_MASTER - The ioctl can only be called on the master file + handle + </para></listitem> + <listitem><para> + DRM_ROOT_ONLY - Only callers with the SYSADMIN capability allowed + </para></listitem> + <listitem><para> + DRM_CONTROL_ALLOW - The ioctl can only be called on a control + device + </para></listitem> + <listitem><para> + DRM_UNLOCKED - The ioctl handler will be called without locking + the DRM global mutex + </para></listitem> + </itemizedlist> + </para> </para> </sect2> - - </sect1> - - <!-- Output management --> - <sect1> - <title>Output management</title> - <para> - At the core of the DRM output management code is a set of - structures representing CRTCs, encoders, and connectors. - </para> - <para> - A CRTC is an abstraction representing a part of the chip that - contains a pointer to a scanout buffer. Therefore, the number - of CRTCs available determines how many independent scanout - buffers can be active at any given time. The CRTC structure - contains several fields to support this: a pointer to some video - memory, a display mode, and an (x, y) offset into the video - memory to support panning or configurations where one piece of - video memory spans multiple CRTCs. - </para> - <para> - An encoder takes pixel data from a CRTC and converts it to a - format suitable for any attached connectors. On some devices, - it may be possible to have a CRTC send data to more than one - encoder. In that case, both encoders would receive data from - the same scanout buffer, resulting in a "cloned" display - configuration across the connectors attached to each encoder. - </para> - <para> - A connector is the final destination for pixel data on a device, - and usually connects directly to an external display device like - a monitor or laptop panel. A connector can only be attached to - one encoder at a time. The connector is also the structure - where information about the attached display is kept, so it - contains fields for display data, EDID data, DPMS & - connection status, and information about modes supported on the - attached displays. - </para> -<!--!Edrivers/char/drm/drm_crtc.c--> - </sect1> - - <sect1> - <title>Framebuffer management</title> - <para> - Clients need to provide a framebuffer object which provides a source - of pixels for a CRTC to deliver to the encoder(s) and ultimately the - connector(s). A framebuffer is fundamentally a driver-specific memory - object, made into an opaque handle by the DRM's addfb() function. - Once a framebuffer has been created this way, it may be passed to the - KMS mode setting routines for use in a completed configuration. - </para> </sect1> <sect1> @@ -812,15 +2355,24 @@ void intel_crt_init(struct drm_device *dev) </para> </sect1> + <!-- Internals: suspend/resume --> + <sect1> - <title>Suspend/resume</title> + <title>Suspend/Resume</title> + <para> + The DRM core provides some suspend/resume code, but drivers wanting full + suspend/resume support should provide save() and restore() functions. + These are called at suspend, hibernate, or resume time, and should perform + any state save or restore required by your device across suspend or + hibernate states. + </para> + <synopsis>int (*suspend) (struct drm_device *, pm_message_t state); +int (*resume) (struct drm_device *);</synopsis> <para> - The DRM core provides some suspend/resume code, but drivers - wanting full suspend/resume support should provide save() and - restore() functions. These are called at suspend, - hibernate, or resume time, and should perform any state save or - restore required by your device across suspend or hibernate - states. + Those are legacy suspend and resume methods. New driver should use the + power management interface provided by their bus type (usually through + the struct <structname>device_driver</structname> dev_pm_ops) and set + these methods to NULL. </para> </sect1> @@ -833,6 +2385,35 @@ void intel_crt_init(struct drm_device *dev) </sect1> </chapter> +<!-- TODO + +- Add a glossary +- Document the struct_mutex catch-all lock +- Document connector properties + +- Why is the load method optional? +- What are drivers supposed to set the initial display state to, and how? + Connector's DPMS states are not initialized and are thus equal to + DRM_MODE_DPMS_ON. The fbcon compatibility layer calls + drm_helper_disable_unused_functions(), which disables unused encoders and + CRTCs, but doesn't touch the connectors' DPMS state, and + drm_helper_connector_dpms() in reaction to fbdev blanking events. Do drivers + that don't implement (or just don't use) fbcon compatibility need to call + those functions themselves? +- KMS drivers must call drm_vblank_pre_modeset() and drm_vblank_post_modeset() + around mode setting. Should this be done in the DRM core? +- vblank_disable_allowed is set to 1 in the first drm_vblank_post_modeset() + call and never set back to 0. It seems to be safe to permanently set it to 1 + in drm_vblank_init() for KMS driver, and it might be safe for UMS drivers as + well. This should be investigated. +- crtc and connector .save and .restore operations are only used internally in + drivers, should they be removed from the core? +- encoder mid-layer .save and .restore operations are only used internally in + drivers, should they be removed from the core? +- encoder mid-layer .detect operation is only used internally in drivers, + should it be removed from the core? +--> + <!-- External interfaces --> <chapter id="drmExternals"> @@ -853,6 +2434,42 @@ void intel_crt_init(struct drm_device *dev) Cover generic ioctls and sysfs layout here. We only need high-level info, since man pages should cover the rest. </para> + + <!-- External: vblank handling --> + + <sect1> + <title>VBlank event handling</title> + <para> + The DRM core exposes two vertical blank related ioctls: + <variablelist> + <varlistentry> + <term>DRM_IOCTL_WAIT_VBLANK</term> + <listitem> + <para> + This takes a struct drm_wait_vblank structure as its argument, + and it is used to block or request a signal when a specified + vblank event occurs. + </para> + </listitem> + </varlistentry> + <varlistentry> + <term>DRM_IOCTL_MODESET_CTL</term> + <listitem> + <para> + This should be called by application level drivers before and + after mode setting, since on many devices the vertical blank + counter is reset at that time. Internally, the DRM snapshots + the last vblank count when the ioctl is called with the + _DRM_PRE_MODESET command, so that the counter won't go backwards + (which is dealt with when _DRM_POST_MODESET is used). + </para> + </listitem> + </varlistentry> + </variablelist> +<!--!Edrivers/char/drm/drm_irq.c--> + </para> + </sect1> + </chapter> <!-- API reference --> diff --git a/drivers/acpi/video.c b/drivers/acpi/video.c index 1e0a9e17c31d..f94d4c818fc7 100644 --- a/drivers/acpi/video.c +++ b/drivers/acpi/video.c @@ -1448,8 +1448,7 @@ static void acpi_video_bus_notify(struct acpi_device *device, u32 event) case ACPI_VIDEO_NOTIFY_SWITCH: /* User requested a switch, * most likely via hotkey. */ acpi_bus_generate_proc_event(device, event, 0); - if (!acpi_notifier_call_chain(device, event, 0)) - keycode = KEY_SWITCHVIDEOMODE; + keycode = KEY_SWITCHVIDEOMODE; break; case ACPI_VIDEO_NOTIFY_PROBE: /* User plugged in or removed a video @@ -1479,8 +1478,9 @@ static void acpi_video_bus_notify(struct acpi_device *device, u32 event) break; } - if (event != ACPI_VIDEO_NOTIFY_SWITCH) - acpi_notifier_call_chain(device, event, 0); + if (acpi_notifier_call_chain(device, event, 0)) + /* Something vetoed the keypress. */ + keycode = 0; if (keycode) { input_report_key(input, keycode, 1); diff --git a/drivers/char/agp/intel-gtt.c b/drivers/char/agp/intel-gtt.c index 58e32f7c3229..e01f5eaaec82 100644 --- a/drivers/char/agp/intel-gtt.c +++ b/drivers/char/agp/intel-gtt.c @@ -84,40 +84,33 @@ static struct _intel_private { #define IS_IRONLAKE intel_private.driver->is_ironlake #define HAS_PGTBL_EN intel_private.driver->has_pgtbl_enable -int intel_gtt_map_memory(struct page **pages, unsigned int num_entries, - struct scatterlist **sg_list, int *num_sg) +static int intel_gtt_map_memory(struct page **pages, + unsigned int num_entries, + struct sg_table *st) { - struct sg_table st; struct scatterlist *sg; int i; - if (*sg_list) - return 0; /* already mapped (for e.g. resume */ - DBG("try mapping %lu pages\n", (unsigned long)num_entries); - if (sg_alloc_table(&st, num_entries, GFP_KERNEL)) + if (sg_alloc_table(st, num_entries, GFP_KERNEL)) goto err; - *sg_list = sg = st.sgl; - - for (i = 0 ; i < num_entries; i++, sg = sg_next(sg)) + for_each_sg(st->sgl, sg, num_entries, i) sg_set_page(sg, pages[i], PAGE_SIZE, 0); - *num_sg = pci_map_sg(intel_private.pcidev, *sg_list, - num_entries, PCI_DMA_BIDIRECTIONAL); - if (unlikely(!*num_sg)) + if (!pci_map_sg(intel_private.pcidev, + st->sgl, st->nents, PCI_DMA_BIDIRECTIONAL)) goto err; return 0; err: - sg_free_table(&st); + sg_free_table(st); return -ENOMEM; } -EXPORT_SYMBOL(intel_gtt_map_memory); -void intel_gtt_unmap_memory(struct scatterlist *sg_list, int num_sg) +static void intel_gtt_unmap_memory(struct scatterlist *sg_list, int num_sg) { struct sg_table st; DBG("try unmapping %lu pages\n", (unsigned long)mem->page_count); @@ -130,7 +123,6 @@ void intel_gtt_unmap_memory(struct scatterlist *sg_list, int num_sg) sg_free_table(&st); } -EXPORT_SYMBOL(intel_gtt_unmap_memory); static void intel_fake_agp_enable(struct agp_bridge_data *bridge, u32 mode) { @@ -674,9 +666,14 @@ static int intel_gtt_init(void) gtt_map_size = intel_private.base.gtt_total_entries * 4; - intel_private.gtt = ioremap(intel_private.gtt_bus_addr, - gtt_map_size); - if (!intel_private.gtt) { + intel_private.gtt = NULL; + if (INTEL_GTT_GEN < 6) + intel_private.gtt = ioremap_wc(intel_private.gtt_bus_addr, + gtt_map_size); + if (intel_private.gtt == NULL) + intel_private.gtt = ioremap(intel_private.gtt_bus_addr, + gtt_map_size); + if (intel_private.gtt == NULL) { intel_private.driver->cleanup(); iounmap(intel_private.registers); return -ENOMEM; @@ -879,8 +876,7 @@ static bool i830_check_flags(unsigned int flags) return false; } -void intel_gtt_insert_sg_entries(struct scatterlist *sg_list, - unsigned int sg_len, +void intel_gtt_insert_sg_entries(struct sg_table *st, unsigned int pg_start, unsigned int flags) { @@ -892,12 +888,11 @@ void intel_gtt_insert_sg_entries(struct scatterlist *sg_list, /* sg may merge pages, but we have to separate * per-page addr for GTT */ - for_each_sg(sg_list, sg, sg_len, i) { + for_each_sg(st->sgl, sg, st->nents, i) { len = sg_dma_len(sg) >> PAGE_SHIFT; for (m = 0; m < len; m++) { dma_addr_t addr = sg_dma_address(sg) + (m << PAGE_SHIFT); - intel_private.driver->write_entry(addr, - j, flags); + intel_private.driver->write_entry(addr, j, flags); j++; } } @@ -905,8 +900,10 @@ void intel_gtt_insert_sg_entries(struct scatterlist *sg_list, } EXPORT_SYMBOL(intel_gtt_insert_sg_entries); -void intel_gtt_insert_pages(unsigned int first_entry, unsigned int num_entries, - struct page **pages, unsigned int flags) +static void intel_gtt_insert_pages(unsigned int first_entry, + unsigned int num_entries, + struct page **pages, + unsigned int flags) { int i, j; @@ -917,7 +914,6 @@ void intel_gtt_insert_pages(unsigned int first_entry, unsigned int num_entries, } readl(intel_private.gtt+j-1); } -EXPORT_SYMBOL(intel_gtt_insert_pages); static int intel_fake_agp_insert_entries(struct agp_memory *mem, off_t pg_start, int type) @@ -953,13 +949,15 @@ static int intel_fake_agp_insert_entries(struct agp_memory *mem, global_cache_flush(); if (intel_private.base.needs_dmar) { - ret = intel_gtt_map_memory(mem->pages, mem->page_count, - &mem->sg_list, &mem->num_sg); + struct sg_table st; + + ret = intel_gtt_map_memory(mem->pages, mem->page_count, &st); if (ret != 0) return ret; - intel_gtt_insert_sg_entries(mem->sg_list, mem->num_sg, - pg_start, type); + intel_gtt_insert_sg_entries(&st, pg_start, type); + mem->sg_list = st.sgl; + mem->num_sg = st.nents; } else intel_gtt_insert_pages(pg_start, mem->page_count, mem->pages, type); diff --git a/drivers/gpu/drm/Kconfig b/drivers/gpu/drm/Kconfig index 90e28081712d..18321b68b880 100644 --- a/drivers/gpu/drm/Kconfig +++ b/drivers/gpu/drm/Kconfig @@ -22,7 +22,7 @@ menuconfig DRM config DRM_USB tristate depends on DRM - depends on USB_ARCH_HAS_HCD + depends on USB_SUPPORT && USB_ARCH_HAS_HCD select USB config DRM_KMS_HELPER @@ -54,6 +54,21 @@ config DRM_TTM GPU memory types. Will be enabled automatically if a device driver uses it. +config DRM_GEM_CMA_HELPER + bool + depends on DRM + help + Choose this if you need the GEM CMA helper functions + +config DRM_KMS_CMA_HELPER + bool + select DRM_GEM_CMA_HELPER + select FB_SYS_FILLRECT + select FB_SYS_COPYAREA + select FB_SYS_IMAGEBLIT + help + Choose this if you need the KMS CMA helper functions + config DRM_TDFX tristate "3dfx Banshee/Voodoo3+" depends on DRM && PCI @@ -193,3 +208,5 @@ source "drivers/gpu/drm/ast/Kconfig" source "drivers/gpu/drm/mgag200/Kconfig" source "drivers/gpu/drm/cirrus/Kconfig" + +source "drivers/gpu/drm/shmobile/Kconfig" diff --git a/drivers/gpu/drm/Makefile b/drivers/gpu/drm/Makefile index f65f65ed0ddf..2ff5cefe9ead 100644 --- a/drivers/gpu/drm/Makefile +++ b/drivers/gpu/drm/Makefile @@ -15,11 +15,13 @@ drm-y := drm_auth.o drm_buffer.o drm_bufs.o drm_cache.o \ drm_trace_points.o drm_global.o drm_prime.o drm-$(CONFIG_COMPAT) += drm_ioc32.o +drm-$(CONFIG_DRM_GEM_CMA_HELPER) += drm_gem_cma_helper.o drm-usb-y := drm_usb.o drm_kms_helper-y := drm_fb_helper.o drm_crtc_helper.o drm_dp_i2c_helper.o drm_kms_helper-$(CONFIG_DRM_LOAD_EDID_FIRMWARE) += drm_edid_load.o +drm_kms_helper-$(CONFIG_DRM_KMS_CMA_HELPER) += drm_fb_cma_helper.o obj-$(CONFIG_DRM_KMS_HELPER) += drm_kms_helper.o @@ -45,4 +47,5 @@ obj-$(CONFIG_DRM_EXYNOS) +=exynos/ obj-$(CONFIG_DRM_GMA500) += gma500/ obj-$(CONFIG_DRM_UDL) += udl/ obj-$(CONFIG_DRM_AST) += ast/ +obj-$(CONFIG_DRM_SHMOBILE) +=shmobile/ obj-y += i2c/ diff --git a/drivers/gpu/drm/ast/ast_drv.h b/drivers/gpu/drm/ast/ast_drv.h index d4af9edcbb97..20a437f88780 100644 --- a/drivers/gpu/drm/ast/ast_drv.h +++ b/drivers/gpu/drm/ast/ast_drv.h @@ -94,7 +94,6 @@ struct ast_private { struct drm_global_reference mem_global_ref; struct ttm_bo_global_ref bo_global_ref; struct ttm_bo_device bdev; - atomic_t validate_sequence; } ttm; struct drm_gem_object *cursor_cache; diff --git a/drivers/gpu/drm/ast/ast_mode.c b/drivers/gpu/drm/ast/ast_mode.c index a712cafcfa1d..e87f1ff223f4 100644 --- a/drivers/gpu/drm/ast/ast_mode.c +++ b/drivers/gpu/drm/ast/ast_mode.c @@ -582,7 +582,6 @@ static const struct drm_crtc_helper_funcs ast_crtc_helper_funcs = { .mode_set_base = ast_crtc_mode_set_base, .disable = ast_crtc_disable, .load_lut = ast_crtc_load_lut, - .disable = ast_crtc_disable, .prepare = ast_crtc_prepare, .commit = ast_crtc_commit, @@ -737,6 +736,7 @@ static int ast_get_modes(struct drm_connector *connector) if (edid) { drm_mode_connector_update_edid_property(&ast_connector->base, edid); ret = drm_add_edid_modes(connector, edid); + kfree(edid); return ret; } else drm_mode_connector_update_edid_property(&ast_connector->base, NULL); diff --git a/drivers/gpu/drm/cirrus/cirrus_drv.h b/drivers/gpu/drm/cirrus/cirrus_drv.h index 64ea597cb6d3..5d045647a3fc 100644 --- a/drivers/gpu/drm/cirrus/cirrus_drv.h +++ b/drivers/gpu/drm/cirrus/cirrus_drv.h @@ -143,7 +143,6 @@ struct cirrus_device { struct drm_global_reference mem_global_ref; struct ttm_bo_global_ref bo_global_ref; struct ttm_bo_device bdev; - atomic_t validate_sequence; } ttm; bool mm_inited; }; diff --git a/drivers/gpu/drm/drm_cache.c b/drivers/gpu/drm/drm_cache.c index 08758e061478..4a4274b348b6 100644 --- a/drivers/gpu/drm/drm_cache.c +++ b/drivers/gpu/drm/drm_cache.c @@ -37,12 +37,13 @@ drm_clflush_page(struct page *page) { uint8_t *page_virtual; unsigned int i; + const int size = boot_cpu_data.x86_clflush_size; if (unlikely(page == NULL)) return; page_virtual = kmap_atomic(page); - for (i = 0; i < PAGE_SIZE; i += boot_cpu_data.x86_clflush_size) + for (i = 0; i < PAGE_SIZE; i += size) clflush(page_virtual + i); kunmap_atomic(page_virtual); } @@ -100,6 +101,31 @@ drm_clflush_pages(struct page *pages[], unsigned long num_pages) EXPORT_SYMBOL(drm_clflush_pages); void +drm_clflush_sg(struct sg_table *st) +{ +#if defined(CONFIG_X86) + if (cpu_has_clflush) { + struct scatterlist *sg; + int i; + + mb(); + for_each_sg(st->sgl, sg, st->nents, i) + drm_clflush_page(sg_page(sg)); + mb(); + + return; + } + + if (on_each_cpu(drm_clflush_ipi_handler, NULL, 1) != 0) + printk(KERN_ERR "Timed out waiting for cache flush.\n"); +#else + printk(KERN_ERR "Architecture has no drm_cache.c support\n"); + WARN_ON_ONCE(1); +#endif +} +EXPORT_SYMBOL(drm_clflush_sg); + +void drm_clflush_virt_range(char *addr, unsigned long length) { #if defined(CONFIG_X86) diff --git a/drivers/gpu/drm/drm_crtc.c b/drivers/gpu/drm/drm_crtc.c index 6fbfc244748f..9c346a50379f 100644 --- a/drivers/gpu/drm/drm_crtc.c +++ b/drivers/gpu/drm/drm_crtc.c @@ -294,6 +294,8 @@ int drm_framebuffer_init(struct drm_device *dev, struct drm_framebuffer *fb, { int ret; + kref_init(&fb->refcount); + ret = drm_mode_object_get(dev, &fb->base, DRM_MODE_OBJECT_FB); if (ret) return ret; @@ -307,6 +309,38 @@ int drm_framebuffer_init(struct drm_device *dev, struct drm_framebuffer *fb, } EXPORT_SYMBOL(drm_framebuffer_init); +static void drm_framebuffer_free(struct kref *kref) +{ + struct drm_framebuffer *fb = + container_of(kref, struct drm_framebuffer, refcount); + fb->funcs->destroy(fb); +} + +/** + * drm_framebuffer_unreference - unref a framebuffer + * + * LOCKING: + * Caller must hold mode config lock. + */ +void drm_framebuffer_unreference(struct drm_framebuffer *fb) +{ + struct drm_device *dev = fb->dev; + DRM_DEBUG("FB ID: %d\n", fb->base.id); + WARN_ON(!mutex_is_locked(&dev->mode_config.mutex)); + kref_put(&fb->refcount, drm_framebuffer_free); +} +EXPORT_SYMBOL(drm_framebuffer_unreference); + +/** + * drm_framebuffer_reference - incr the fb refcnt + */ +void drm_framebuffer_reference(struct drm_framebuffer *fb) +{ + DRM_DEBUG("FB ID: %d\n", fb->base.id); + kref_get(&fb->refcount); +} +EXPORT_SYMBOL(drm_framebuffer_reference); + /** * drm_framebuffer_cleanup - remove a framebuffer object * @fb: framebuffer to remove @@ -320,6 +354,32 @@ EXPORT_SYMBOL(drm_framebuffer_init); void drm_framebuffer_cleanup(struct drm_framebuffer *fb) { struct drm_device *dev = fb->dev; + /* + * This could be moved to drm_framebuffer_remove(), but for + * debugging is nice to keep around the list of fb's that are + * no longer associated w/ a drm_file but are not unreferenced + * yet. (i915 and omapdrm have debugfs files which will show + * this.) + */ + drm_mode_object_put(dev, &fb->base); + list_del(&fb->head); + dev->mode_config.num_fb--; +} +EXPORT_SYMBOL(drm_framebuffer_cleanup); + +/** + * drm_framebuffer_remove - remove and unreference a framebuffer object + * @fb: framebuffer to remove + * + * LOCKING: + * Caller must hold mode config lock. + * + * Scans all the CRTCs and planes in @dev's mode_config. If they're + * using @fb, removes it, setting it to NULL. + */ +void drm_framebuffer_remove(struct drm_framebuffer *fb) +{ + struct drm_device *dev = fb->dev; struct drm_crtc *crtc; struct drm_plane *plane; struct drm_mode_set set; @@ -350,11 +410,11 @@ void drm_framebuffer_cleanup(struct drm_framebuffer *fb) } } - drm_mode_object_put(dev, &fb->base); - list_del(&fb->head); - dev->mode_config.num_fb--; + list_del(&fb->filp_head); + + drm_framebuffer_unreference(fb); } -EXPORT_SYMBOL(drm_framebuffer_cleanup); +EXPORT_SYMBOL(drm_framebuffer_remove); /** * drm_crtc_init - Initialise a new CRTC object @@ -377,6 +437,7 @@ int drm_crtc_init(struct drm_device *dev, struct drm_crtc *crtc, crtc->dev = dev; crtc->funcs = funcs; + crtc->invert_dimensions = false; mutex_lock(&dev->mode_config.mutex); @@ -1031,11 +1092,7 @@ void drm_mode_config_cleanup(struct drm_device *dev) } list_for_each_entry_safe(fb, fbt, &dev->mode_config.fb_list, head) { - fb->funcs->destroy(fb); - } - - list_for_each_entry_safe(crtc, ct, &dev->mode_config.crtc_list, head) { - crtc->funcs->destroy(crtc); + drm_framebuffer_remove(fb); } list_for_each_entry_safe(plane, plt, &dev->mode_config.plane_list, @@ -1043,6 +1100,10 @@ void drm_mode_config_cleanup(struct drm_device *dev) plane->funcs->destroy(plane); } + list_for_each_entry_safe(crtc, ct, &dev->mode_config.crtc_list, head) { + crtc->funcs->destroy(crtc); + } + idr_remove_all(&dev->mode_config.crtc_idr); idr_destroy(&dev->mode_config.crtc_idr); } @@ -1852,6 +1913,7 @@ int drm_mode_setcrtc(struct drm_device *dev, void *data, DRM_DEBUG_KMS("[CRTC:%d]\n", crtc->base.id); if (crtc_req->mode_valid) { + int hdisplay, vdisplay; /* If we have a mode we need a framebuffer. */ /* If we pass -1, set the mode with the currently bound fb */ if (crtc_req->fb_id == -1) { @@ -1887,14 +1949,20 @@ int drm_mode_setcrtc(struct drm_device *dev, void *data, drm_mode_set_crtcinfo(mode, CRTC_INTERLACE_HALVE_V); - if (mode->hdisplay > fb->width || - mode->vdisplay > fb->height || - crtc_req->x > fb->width - mode->hdisplay || - crtc_req->y > fb->height - mode->vdisplay) { - DRM_DEBUG_KMS("Invalid CRTC viewport %ux%u+%u+%u for fb size %ux%u.\n", - mode->hdisplay, mode->vdisplay, - crtc_req->x, crtc_req->y, - fb->width, fb->height); + hdisplay = mode->hdisplay; + vdisplay = mode->vdisplay; + + if (crtc->invert_dimensions) + swap(hdisplay, vdisplay); + + if (hdisplay > fb->width || + vdisplay > fb->height || + crtc_req->x > fb->width - hdisplay || + crtc_req->y > fb->height - vdisplay) { + DRM_DEBUG_KMS("Invalid fb size %ux%u for CRTC viewport %ux%u+%d+%d%s.\n", + fb->width, fb->height, + hdisplay, vdisplay, crtc_req->x, crtc_req->y, + crtc->invert_dimensions ? " (inverted)" : ""); ret = -ENOSPC; goto out; } @@ -2169,6 +2237,8 @@ static int format_check(const struct drm_mode_fb_cmd2 *r) case DRM_FORMAT_NV21: case DRM_FORMAT_NV16: case DRM_FORMAT_NV61: + case DRM_FORMAT_NV24: + case DRM_FORMAT_NV42: case DRM_FORMAT_YUV410: case DRM_FORMAT_YVU410: case DRM_FORMAT_YUV411: @@ -2335,11 +2405,7 @@ int drm_mode_rmfb(struct drm_device *dev, goto out; } - /* TODO release all crtc connected to the framebuffer */ - /* TODO unhock the destructor from the buffer object */ - - list_del(&fb->filp_head); - fb->funcs->destroy(fb); + drm_framebuffer_remove(fb); out: mutex_unlock(&dev->mode_config.mutex); @@ -2489,8 +2555,7 @@ void drm_fb_release(struct drm_file *priv) mutex_lock(&dev->mode_config.mutex); list_for_each_entry_safe(fb, tfb, &priv->fbs, filp_head) { - list_del(&fb->filp_head); - fb->funcs->destroy(fb); + drm_framebuffer_remove(fb); } mutex_unlock(&dev->mode_config.mutex); } @@ -3489,6 +3554,7 @@ int drm_mode_page_flip_ioctl(struct drm_device *dev, struct drm_framebuffer *fb; struct drm_pending_vblank_event *e = NULL; unsigned long flags; + int hdisplay, vdisplay; int ret = -EINVAL; if (page_flip->flags & ~DRM_MODE_PAGE_FLIP_FLAGS || @@ -3518,14 +3584,19 @@ int drm_mode_page_flip_ioctl(struct drm_device *dev, goto out; fb = obj_to_fb(obj); - if (crtc->mode.hdisplay > fb->width || - crtc->mode.vdisplay > fb->height || - crtc->x > fb->width - crtc->mode.hdisplay || - crtc->y > fb->height - crtc->mode.vdisplay) { - DRM_DEBUG_KMS("Invalid fb size %ux%u for CRTC viewport %ux%u+%d+%d.\n", - fb->width, fb->height, - crtc->mode.hdisplay, crtc->mode.vdisplay, - crtc->x, crtc->y); + hdisplay = crtc->mode.hdisplay; + vdisplay = crtc->mode.vdisplay; + + if (crtc->invert_dimensions) + swap(hdisplay, vdisplay); + + if (hdisplay > fb->width || + vdisplay > fb->height || + crtc->x > fb->width - hdisplay || + crtc->y > fb->height - vdisplay) { + DRM_DEBUG_KMS("Invalid fb size %ux%u for CRTC viewport %ux%u+%d+%d%s.\n", + fb->width, fb->height, hdisplay, vdisplay, crtc->x, crtc->y, + crtc->invert_dimensions ? " (inverted)" : ""); ret = -ENOSPC; goto out; } @@ -3718,6 +3789,8 @@ int drm_format_num_planes(uint32_t format) case DRM_FORMAT_NV21: case DRM_FORMAT_NV16: case DRM_FORMAT_NV61: + case DRM_FORMAT_NV24: + case DRM_FORMAT_NV42: return 2; default: return 1; @@ -3751,6 +3824,8 @@ int drm_format_plane_cpp(uint32_t format, int plane) case DRM_FORMAT_NV21: case DRM_FORMAT_NV16: case DRM_FORMAT_NV61: + case DRM_FORMAT_NV24: + case DRM_FORMAT_NV42: return plane ? 2 : 1; case DRM_FORMAT_YUV410: case DRM_FORMAT_YVU410: diff --git a/drivers/gpu/drm/drm_drv.c b/drivers/gpu/drm/drm_drv.c index 9238de4009fa..a2c6eb6e0dd8 100644 --- a/drivers/gpu/drm/drm_drv.c +++ b/drivers/gpu/drm/drm_drv.c @@ -140,10 +140,10 @@ static struct drm_ioctl_desc drm_ioctls[] = { DRM_IOCTL_DEF(DRM_IOCTL_PRIME_HANDLE_TO_FD, drm_prime_handle_to_fd_ioctl, DRM_AUTH|DRM_UNLOCKED), DRM_IOCTL_DEF(DRM_IOCTL_PRIME_FD_TO_HANDLE, drm_prime_fd_to_handle_ioctl, DRM_AUTH|DRM_UNLOCKED), - DRM_IOCTL_DEF(DRM_IOCTL_MODE_GETPLANERESOURCES, drm_mode_getplane_res, DRM_MASTER|DRM_CONTROL_ALLOW|DRM_UNLOCKED), + DRM_IOCTL_DEF(DRM_IOCTL_MODE_GETPLANERESOURCES, drm_mode_getplane_res, DRM_CONTROL_ALLOW|DRM_UNLOCKED), DRM_IOCTL_DEF(DRM_IOCTL_MODE_GETCRTC, drm_mode_getcrtc, DRM_CONTROL_ALLOW|DRM_UNLOCKED), DRM_IOCTL_DEF(DRM_IOCTL_MODE_SETCRTC, drm_mode_setcrtc, DRM_MASTER|DRM_CONTROL_ALLOW|DRM_UNLOCKED), - DRM_IOCTL_DEF(DRM_IOCTL_MODE_GETPLANE, drm_mode_getplane, DRM_MASTER|DRM_CONTROL_ALLOW|DRM_UNLOCKED), + DRM_IOCTL_DEF(DRM_IOCTL_MODE_GETPLANE, drm_mode_getplane, DRM_CONTROL_ALLOW|DRM_UNLOCKED), DRM_IOCTL_DEF(DRM_IOCTL_MODE_SETPLANE, drm_mode_setplane, DRM_MASTER|DRM_CONTROL_ALLOW|DRM_UNLOCKED), DRM_IOCTL_DEF(DRM_IOCTL_MODE_CURSOR, drm_mode_cursor_ioctl, DRM_MASTER|DRM_CONTROL_ALLOW|DRM_UNLOCKED), DRM_IOCTL_DEF(DRM_IOCTL_MODE_GETGAMMA, drm_mode_gamma_get_ioctl, DRM_UNLOCKED), @@ -152,19 +152,19 @@ static struct drm_ioctl_desc drm_ioctls[] = { DRM_IOCTL_DEF(DRM_IOCTL_MODE_GETCONNECTOR, drm_mode_getconnector, DRM_CONTROL_ALLOW|DRM_UNLOCKED), DRM_IOCTL_DEF(DRM_IOCTL_MODE_ATTACHMODE, drm_mode_attachmode_ioctl, DRM_MASTER|DRM_CONTROL_ALLOW|DRM_UNLOCKED), DRM_IOCTL_DEF(DRM_IOCTL_MODE_DETACHMODE, drm_mode_detachmode_ioctl, DRM_MASTER|DRM_CONTROL_ALLOW|DRM_UNLOCKED), - DRM_IOCTL_DEF(DRM_IOCTL_MODE_GETPROPERTY, drm_mode_getproperty_ioctl, DRM_MASTER | DRM_CONTROL_ALLOW|DRM_UNLOCKED), + DRM_IOCTL_DEF(DRM_IOCTL_MODE_GETPROPERTY, drm_mode_getproperty_ioctl, DRM_CONTROL_ALLOW|DRM_UNLOCKED), DRM_IOCTL_DEF(DRM_IOCTL_MODE_SETPROPERTY, drm_mode_connector_property_set_ioctl, DRM_MASTER|DRM_CONTROL_ALLOW|DRM_UNLOCKED), - DRM_IOCTL_DEF(DRM_IOCTL_MODE_GETPROPBLOB, drm_mode_getblob_ioctl, DRM_MASTER|DRM_CONTROL_ALLOW|DRM_UNLOCKED), + DRM_IOCTL_DEF(DRM_IOCTL_MODE_GETPROPBLOB, drm_mode_getblob_ioctl, DRM_CONTROL_ALLOW|DRM_UNLOCKED), DRM_IOCTL_DEF(DRM_IOCTL_MODE_GETFB, drm_mode_getfb, DRM_CONTROL_ALLOW|DRM_UNLOCKED), - DRM_IOCTL_DEF(DRM_IOCTL_MODE_ADDFB, drm_mode_addfb, DRM_MASTER|DRM_CONTROL_ALLOW|DRM_UNLOCKED), - DRM_IOCTL_DEF(DRM_IOCTL_MODE_ADDFB2, drm_mode_addfb2, DRM_MASTER|DRM_CONTROL_ALLOW|DRM_UNLOCKED), - DRM_IOCTL_DEF(DRM_IOCTL_MODE_RMFB, drm_mode_rmfb, DRM_MASTER|DRM_CONTROL_ALLOW|DRM_UNLOCKED), + DRM_IOCTL_DEF(DRM_IOCTL_MODE_ADDFB, drm_mode_addfb, DRM_CONTROL_ALLOW|DRM_UNLOCKED), + DRM_IOCTL_DEF(DRM_IOCTL_MODE_ADDFB2, drm_mode_addfb2, DRM_CONTROL_ALLOW|DRM_UNLOCKED), + DRM_IOCTL_DEF(DRM_IOCTL_MODE_RMFB, drm_mode_rmfb, DRM_CONTROL_ALLOW|DRM_UNLOCKED), DRM_IOCTL_DEF(DRM_IOCTL_MODE_PAGE_FLIP, drm_mode_page_flip_ioctl, DRM_MASTER|DRM_CONTROL_ALLOW|DRM_UNLOCKED), DRM_IOCTL_DEF(DRM_IOCTL_MODE_DIRTYFB, drm_mode_dirtyfb_ioctl, DRM_MASTER|DRM_CONTROL_ALLOW|DRM_UNLOCKED), - DRM_IOCTL_DEF(DRM_IOCTL_MODE_CREATE_DUMB, drm_mode_create_dumb_ioctl, DRM_MASTER|DRM_CONTROL_ALLOW|DRM_UNLOCKED), - DRM_IOCTL_DEF(DRM_IOCTL_MODE_MAP_DUMB, drm_mode_mmap_dumb_ioctl, DRM_MASTER|DRM_CONTROL_ALLOW|DRM_UNLOCKED), - DRM_IOCTL_DEF(DRM_IOCTL_MODE_DESTROY_DUMB, drm_mode_destroy_dumb_ioctl, DRM_MASTER|DRM_CONTROL_ALLOW|DRM_UNLOCKED), - DRM_IOCTL_DEF(DRM_IOCTL_MODE_OBJ_GETPROPERTIES, drm_mode_obj_get_properties_ioctl, DRM_MASTER|DRM_CONTROL_ALLOW|DRM_UNLOCKED), + DRM_IOCTL_DEF(DRM_IOCTL_MODE_CREATE_DUMB, drm_mode_create_dumb_ioctl, DRM_CONTROL_ALLOW|DRM_UNLOCKED), + DRM_IOCTL_DEF(DRM_IOCTL_MODE_MAP_DUMB, drm_mode_mmap_dumb_ioctl, DRM_CONTROL_ALLOW|DRM_UNLOCKED), + DRM_IOCTL_DEF(DRM_IOCTL_MODE_DESTROY_DUMB, drm_mode_destroy_dumb_ioctl, DRM_CONTROL_ALLOW|DRM_UNLOCKED), + DRM_IOCTL_DEF(DRM_IOCTL_MODE_OBJ_GETPROPERTIES, drm_mode_obj_get_properties_ioctl, DRM_CONTROL_ALLOW|DRM_UNLOCKED), DRM_IOCTL_DEF(DRM_IOCTL_MODE_OBJ_SETPROPERTY, drm_mode_obj_set_property_ioctl, DRM_MASTER|DRM_CONTROL_ALLOW|DRM_UNLOCKED), }; diff --git a/drivers/gpu/drm/drm_edid.c b/drivers/gpu/drm/drm_edid.c index b7ee230572b7..289c9b18c0d3 100644 --- a/drivers/gpu/drm/drm_edid.c +++ b/drivers/gpu/drm/drm_edid.c @@ -161,7 +161,7 @@ MODULE_PARM_DESC(edid_fixup, * Sanity check the EDID block (base or extension). Return 0 if the block * doesn't check out, or 1 if it's valid. */ -bool drm_edid_block_valid(u8 *raw_edid, int block) +bool drm_edid_block_valid(u8 *raw_edid, int block, bool print_bad_edid) { int i; u8 csum = 0; @@ -184,7 +184,9 @@ bool drm_edid_block_valid(u8 *raw_edid, int block) for (i = 0; i < EDID_LENGTH; i++) csum += raw_edid[i]; if (csum) { - DRM_ERROR("EDID checksum is invalid, remainder is %d\n", csum); + if (print_bad_edid) { + DRM_ERROR("EDID checksum is invalid, remainder is %d\n", csum); + } /* allow CEA to slide through, switches mangle this */ if (raw_edid[0] != 0x02) @@ -210,7 +212,7 @@ bool drm_edid_block_valid(u8 *raw_edid, int block) return 1; bad: - if (raw_edid) { + if (raw_edid && print_bad_edid) { printk(KERN_ERR "Raw EDID:\n"); print_hex_dump(KERN_ERR, " \t", DUMP_PREFIX_NONE, 16, 1, raw_edid, EDID_LENGTH, false); @@ -234,7 +236,7 @@ bool drm_edid_is_valid(struct edid *edid) return false; for (i = 0; i <= edid->extensions; i++) - if (!drm_edid_block_valid(raw + i * EDID_LENGTH, i)) + if (!drm_edid_block_valid(raw + i * EDID_LENGTH, i, true)) return false; return true; @@ -257,6 +259,8 @@ drm_do_probe_ddc_edid(struct i2c_adapter *adapter, unsigned char *buf, int block, int len) { unsigned char start = block * EDID_LENGTH; + unsigned char segment = block >> 1; + unsigned char xfers = segment ? 3 : 2; int ret, retries = 5; /* The core i2c driver will automatically retry the transfer if the @@ -268,6 +272,11 @@ drm_do_probe_ddc_edid(struct i2c_adapter *adapter, unsigned char *buf, do { struct i2c_msg msgs[] = { { + .addr = DDC_SEGMENT_ADDR, + .flags = 0, + .len = 1, + .buf = &segment, + }, { .addr = DDC_ADDR, .flags = 0, .len = 1, @@ -279,15 +288,21 @@ drm_do_probe_ddc_edid(struct i2c_adapter *adapter, unsigned char *buf, .buf = buf, } }; - ret = i2c_transfer(adapter, msgs, 2); + + /* + * Avoid sending the segment addr to not upset non-compliant ddc + * monitors. + */ + ret = i2c_transfer(adapter, &msgs[3 - xfers], xfers); + if (ret == -ENXIO) { DRM_DEBUG_KMS("drm: skipping non-existent adapter %s\n", adapter->name); break; } - } while (ret != 2 && --retries); + } while (ret != xfers && --retries); - return ret == 2 ? 0 : -1; + return ret == xfers ? 0 : -1; } static bool drm_edid_is_zero(u8 *in_edid, int length) @@ -306,6 +321,7 @@ drm_do_get_edid(struct drm_connector *connector, struct i2c_adapter *adapter) { int i, j = 0, valid_extensions = 0; u8 *block, *new; + bool print_bad_edid = !connector->bad_edid_counter || (drm_debug & DRM_UT_KMS); if ((block = kmalloc(EDID_LENGTH, GFP_KERNEL)) == NULL) return NULL; @@ -314,7 +330,7 @@ drm_do_get_edid(struct drm_connector *connector, struct i2c_adapter *adapter) for (i = 0; i < 4; i++) { if (drm_do_probe_ddc_edid(adapter, block, 0, EDID_LENGTH)) goto out; - if (drm_edid_block_valid(block, 0)) + if (drm_edid_block_valid(block, 0, print_bad_edid)) break; if (i == 0 && drm_edid_is_zero(block, EDID_LENGTH)) { connector->null_edid_counter++; @@ -339,7 +355,7 @@ drm_do_get_edid(struct drm_connector *connector, struct i2c_adapter *adapter) block + (valid_extensions + 1) * EDID_LENGTH, j, EDID_LENGTH)) goto out; - if (drm_edid_block_valid(block + (valid_extensions + 1) * EDID_LENGTH, j)) { + if (drm_edid_block_valid(block + (valid_extensions + 1) * EDID_LENGTH, j, print_bad_edid)) { valid_extensions++; break; } @@ -362,8 +378,11 @@ drm_do_get_edid(struct drm_connector *connector, struct i2c_adapter *adapter) return block; carp: - dev_warn(connector->dev->dev, "%s: EDID block %d invalid.\n", - drm_get_connector_name(connector), j); + if (print_bad_edid) { + dev_warn(connector->dev->dev, "%s: EDID block %d invalid.\n", + drm_get_connector_name(connector), j); + } + connector->bad_edid_counter++; out: kfree(block); @@ -402,10 +421,7 @@ struct edid *drm_get_edid(struct drm_connector *connector, if (drm_probe_ddc(adapter)) edid = (struct edid *)drm_do_get_edid(connector, adapter); - connector->display_info.raw_edid = (char *)edid; - return edid; - } EXPORT_SYMBOL(drm_get_edid); @@ -1523,16 +1539,57 @@ do_cea_modes (struct drm_connector *connector, u8 *db, u8 len) } static int +cea_db_payload_len(const u8 *db) +{ + return db[0] & 0x1f; +} + +static int +cea_db_tag(const u8 *db) +{ + return db[0] >> 5; +} + +static int +cea_revision(const u8 *cea) +{ + return cea[1]; +} + +static int +cea_db_offsets(const u8 *cea, int *start, int *end) +{ + /* Data block offset in CEA extension block */ + *start = 4; + *end = cea[2]; + if (*end == 0) + *end = 127; + if (*end < 4 || *end > 127) + return -ERANGE; + return 0; +} + +#define for_each_cea_db(cea, i, start, end) \ + for ((i) = (start); (i) < (end) && (i) + cea_db_payload_len(&(cea)[(i)]) < (end); (i) += cea_db_payload_len(&(cea)[(i)]) + 1) + +static int add_cea_modes(struct drm_connector *connector, struct edid *edid) { u8 * cea = drm_find_cea_extension(edid); u8 * db, dbl; int modes = 0; - if (cea && cea[1] >= 3) { - for (db = cea + 4; db < cea + cea[2]; db += dbl + 1) { - dbl = db[0] & 0x1f; - if (((db[0] & 0xe0) >> 5) == VIDEO_BLOCK) + if (cea && cea_revision(cea) >= 3) { + int i, start, end; + + if (cea_db_offsets(cea, &start, &end)) + return 0; + + for_each_cea_db(cea, i, start, end) { + db = &cea[i]; + dbl = cea_db_payload_len(db); + + if (cea_db_tag(db) == VIDEO_BLOCK) modes += do_cea_modes (connector, db+1, dbl); } } @@ -1541,19 +1598,28 @@ add_cea_modes(struct drm_connector *connector, struct edid *edid) } static void -parse_hdmi_vsdb(struct drm_connector *connector, uint8_t *db) +parse_hdmi_vsdb(struct drm_connector *connector, const u8 *db) { - connector->eld[5] |= (db[6] >> 7) << 1; /* Supports_AI */ - - connector->dvi_dual = db[6] & 1; - connector->max_tmds_clock = db[7] * 5; + u8 len = cea_db_payload_len(db); - connector->latency_present[0] = db[8] >> 7; - connector->latency_present[1] = (db[8] >> 6) & 1; - connector->video_latency[0] = db[9]; - connector->audio_latency[0] = db[10]; - connector->video_latency[1] = db[11]; - connector->audio_latency[1] = db[12]; + if (len >= 6) { + connector->eld[5] |= (db[6] >> 7) << 1; /* Supports_AI */ + connector->dvi_dual = db[6] & 1; + } + if (len >= 7) + connector->max_tmds_clock = db[7] * 5; + if (len >= 8) { + connector->latency_present[0] = db[8] >> 7; + connector->latency_present[1] = (db[8] >> 6) & 1; + } + if (len >= 9) + connector->video_latency[0] = db[9]; + if (len >= 10) + connector->audio_latency[0] = db[10]; + if (len >= 11) + connector->video_latency[1] = db[11]; + if (len >= 12) + connector->audio_latency[1] = db[12]; DRM_LOG_KMS("HDMI: DVI dual %d, " "max TMDS clock %d, " @@ -1577,6 +1643,21 @@ monitor_name(struct detailed_timing *t, void *data) *(u8 **)data = t->data.other_data.data.str.str; } +static bool cea_db_is_hdmi_vsdb(const u8 *db) +{ + int hdmi_id; + + if (cea_db_tag(db) != VENDOR_BLOCK) + return false; + + if (cea_db_payload_len(db) < 5) + return false; + + hdmi_id = db[1] | (db[2] << 8) | (db[3] << 16); + + return hdmi_id == HDMI_IDENTIFIER; +} + /** * drm_edid_to_eld - build ELD from EDID * @connector: connector corresponding to the HDMI/DP sink @@ -1623,29 +1704,40 @@ void drm_edid_to_eld(struct drm_connector *connector, struct edid *edid) eld[18] = edid->prod_code[0]; eld[19] = edid->prod_code[1]; - if (cea[1] >= 3) - for (db = cea + 4; db < cea + cea[2]; db += dbl + 1) { - dbl = db[0] & 0x1f; - - switch ((db[0] & 0xe0) >> 5) { + if (cea_revision(cea) >= 3) { + int i, start, end; + + if (cea_db_offsets(cea, &start, &end)) { + start = 0; + end = 0; + } + + for_each_cea_db(cea, i, start, end) { + db = &cea[i]; + dbl = cea_db_payload_len(db); + + switch (cea_db_tag(db)) { case AUDIO_BLOCK: /* Audio Data Block, contains SADs */ sad_count = dbl / 3; - memcpy(eld + 20 + mnl, &db[1], dbl); + if (dbl >= 1) + memcpy(eld + 20 + mnl, &db[1], dbl); break; case SPEAKER_BLOCK: - /* Speaker Allocation Data Block */ - eld[7] = db[1]; + /* Speaker Allocation Data Block */ + if (dbl >= 1) + eld[7] = db[1]; break; case VENDOR_BLOCK: /* HDMI Vendor-Specific Data Block */ - if (db[1] == 0x03 && db[2] == 0x0c && db[3] == 0) + if (cea_db_is_hdmi_vsdb(db)) parse_hdmi_vsdb(connector, db); break; default: break; } } + } eld[5] |= sad_count << 4; eld[2] = (20 + mnl + sad_count * 3 + 3) / 4; @@ -1723,38 +1815,26 @@ EXPORT_SYMBOL(drm_select_eld); bool drm_detect_hdmi_monitor(struct edid *edid) { u8 *edid_ext; - int i, hdmi_id; + int i; int start_offset, end_offset; - bool is_hdmi = false; edid_ext = drm_find_cea_extension(edid); if (!edid_ext) - goto end; + return false; - /* Data block offset in CEA extension block */ - start_offset = 4; - end_offset = edid_ext[2]; + if (cea_db_offsets(edid_ext, &start_offset, &end_offset)) + return false; /* * Because HDMI identifier is in Vendor Specific Block, * search it from all data blocks of CEA extension. */ - for (i = start_offset; i < end_offset; - /* Increased by data block len */ - i += ((edid_ext[i] & 0x1f) + 1)) { - /* Find vendor specific block */ - if ((edid_ext[i] >> 5) == VENDOR_BLOCK) { - hdmi_id = edid_ext[i + 1] | (edid_ext[i + 2] << 8) | - edid_ext[i + 3] << 16; - /* Find HDMI identifier */ - if (hdmi_id == HDMI_IDENTIFIER) - is_hdmi = true; - break; - } + for_each_cea_db(edid_ext, i, start_offset, end_offset) { + if (cea_db_is_hdmi_vsdb(&edid_ext[i])) + return true; } -end: - return is_hdmi; + return false; } EXPORT_SYMBOL(drm_detect_hdmi_monitor); @@ -1786,15 +1866,13 @@ bool drm_detect_monitor_audio(struct edid *edid) goto end; } - /* Data block offset in CEA extension block */ - start_offset = 4; - end_offset = edid_ext[2]; + if (cea_db_offsets(edid_ext, &start_offset, &end_offset)) + goto end; - for (i = start_offset; i < end_offset; - i += ((edid_ext[i] & 0x1f) + 1)) { - if ((edid_ext[i] >> 5) == AUDIO_BLOCK) { + for_each_cea_db(edid_ext, i, start_offset, end_offset) { + if (cea_db_tag(&edid_ext[i]) == AUDIO_BLOCK) { has_audio = true; - for (j = 1; j < (edid_ext[i] & 0x1f); j += 3) + for (j = 1; j < cea_db_payload_len(&edid_ext[i]) + 1; j += 3) DRM_DEBUG_KMS("CEA audio format %d\n", (edid_ext[i + j] >> 3) & 0xf); goto end; diff --git a/drivers/gpu/drm/drm_edid_load.c b/drivers/gpu/drm/drm_edid_load.c index 0303935d10e2..ea9cdab3d431 100644 --- a/drivers/gpu/drm/drm_edid_load.c +++ b/drivers/gpu/drm/drm_edid_load.c @@ -114,8 +114,8 @@ static u8 generic_edid[GENERIC_EDIDS][128] = { }, }; -static int edid_load(struct drm_connector *connector, char *name, - char *connector_name) +static u8 *edid_load(struct drm_connector *connector, char *name, + char *connector_name) { const struct firmware *fw; struct platform_device *pdev; @@ -123,6 +123,7 @@ static int edid_load(struct drm_connector *connector, char *name, int fwsize, expected; int builtin = 0, err = 0; int i, valid_extensions = 0; + bool print_bad_edid = !connector->bad_edid_counter || (drm_debug & DRM_UT_KMS); pdev = platform_device_register_simple(connector_name, -1, NULL, 0); if (IS_ERR(pdev)) { @@ -173,7 +174,8 @@ static int edid_load(struct drm_connector *connector, char *name, } memcpy(edid, fwdata, fwsize); - if (!drm_edid_block_valid(edid, 0)) { + if (!drm_edid_block_valid(edid, 0, print_bad_edid)) { + connector->bad_edid_counter++; DRM_ERROR("Base block of EDID firmware \"%s\" is invalid ", name); kfree(edid); @@ -185,7 +187,7 @@ static int edid_load(struct drm_connector *connector, char *name, if (i != valid_extensions + 1) memcpy(edid + (valid_extensions + 1) * EDID_LENGTH, edid + i * EDID_LENGTH, EDID_LENGTH); - if (drm_edid_block_valid(edid + i * EDID_LENGTH, i)) + if (drm_edid_block_valid(edid + i * EDID_LENGTH, i, print_bad_edid)) valid_extensions++; } @@ -205,7 +207,6 @@ static int edid_load(struct drm_connector *connector, char *name, edid = new_edid; } - connector->display_info.raw_edid = edid; DRM_INFO("Got %s EDID base block and %d extension%s from " "\"%s\" for connector \"%s\"\n", builtin ? "built-in" : "external", valid_extensions, valid_extensions == 1 ? "" : "s", @@ -215,7 +216,10 @@ relfw_out: release_firmware(fw); out: - return err; + if (err) + return ERR_PTR(err); + + return edid; } int drm_load_edid_firmware(struct drm_connector *connector) @@ -223,6 +227,7 @@ int drm_load_edid_firmware(struct drm_connector *connector) char *connector_name = drm_get_connector_name(connector); char *edidname = edid_firmware, *last, *colon; int ret; + struct edid *edid; if (*edidname == '\0') return 0; @@ -240,13 +245,13 @@ int drm_load_edid_firmware(struct drm_connector *connector) if (*last == '\n') *last = '\0'; - ret = edid_load(connector, edidname, connector_name); - if (ret) + edid = (struct edid *) edid_load(connector, edidname, connector_name); + if (IS_ERR_OR_NULL(edid)) return 0; - drm_mode_connector_update_edid_property(connector, - (struct edid *) connector->display_info.raw_edid); + drm_mode_connector_update_edid_property(connector, edid); + ret = drm_add_edid_modes(connector, edid); + kfree(edid); - return drm_add_edid_modes(connector, (struct edid *) - connector->display_info.raw_edid); + return ret; } diff --git a/drivers/gpu/drm/drm_edid_modes.h b/drivers/gpu/drm/drm_edid_modes.h index ff98a7eb38dd..57459b316adc 100644 --- a/drivers/gpu/drm/drm_edid_modes.h +++ b/drivers/gpu/drm/drm_edid_modes.h @@ -89,7 +89,7 @@ static const struct drm_display_mode drm_dmt_modes[] = { 976, 1088, 0, 480, 486, 494, 517, 0, DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) }, /* 1024x768@43Hz, interlace */ - { DRM_MODE("1024x768", DRM_MODE_TYPE_DRIVER, 44900, 1024, 1032, + { DRM_MODE("1024x768i", DRM_MODE_TYPE_DRIVER, 44900, 1024, 1032, 1208, 1264, 0, 768, 768, 772, 817, 0, DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC | DRM_MODE_FLAG_INTERLACE) }, @@ -395,7 +395,7 @@ static const struct drm_display_mode edid_est_modes[] = { { DRM_MODE("1024x768", DRM_MODE_TYPE_DRIVER, 65000, 1024, 1048, 1184, 1344, 0, 768, 771, 777, 806, 0, DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) }, /* 1024x768@60Hz */ - { DRM_MODE("1024x768", DRM_MODE_TYPE_DRIVER,44900, 1024, 1032, + { DRM_MODE("1024x768i", DRM_MODE_TYPE_DRIVER,44900, 1024, 1032, 1208, 1264, 0, 768, 768, 776, 817, 0, DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC | DRM_MODE_FLAG_INTERLACE) }, /* 1024x768@43Hz */ { DRM_MODE("832x624", DRM_MODE_TYPE_DRIVER, 57284, 832, 864, @@ -506,17 +506,17 @@ static const struct drm_display_mode edid_cea_modes[] = { 1430, 1650, 0, 720, 725, 730, 750, 0, DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) }, /* 5 - 1920x1080i@60Hz */ - { DRM_MODE("1920x1080", DRM_MODE_TYPE_DRIVER, 74250, 1920, 2008, + { DRM_MODE("1920x1080i", DRM_MODE_TYPE_DRIVER, 74250, 1920, 2008, 2052, 2200, 0, 1080, 1084, 1094, 1125, 0, DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC | DRM_MODE_FLAG_INTERLACE) }, /* 6 - 1440x480i@60Hz */ - { DRM_MODE("1440x480", DRM_MODE_TYPE_DRIVER, 27000, 1440, 1478, + { DRM_MODE("1440x480i", DRM_MODE_TYPE_DRIVER, 27000, 1440, 1478, 1602, 1716, 0, 480, 488, 494, 525, 0, DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC | DRM_MODE_FLAG_INTERLACE | DRM_MODE_FLAG_DBLCLK) }, /* 7 - 1440x480i@60Hz */ - { DRM_MODE("1440x480", DRM_MODE_TYPE_DRIVER, 27000, 1440, 1478, + { DRM_MODE("1440x480i", DRM_MODE_TYPE_DRIVER, 27000, 1440, 1478, 1602, 1716, 0, 480, 488, 494, 525, 0, DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC | DRM_MODE_FLAG_INTERLACE | DRM_MODE_FLAG_DBLCLK) }, @@ -531,12 +531,12 @@ static const struct drm_display_mode edid_cea_modes[] = { DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC | DRM_MODE_FLAG_DBLCLK) }, /* 10 - 2880x480i@60Hz */ - { DRM_MODE("2880x480", DRM_MODE_TYPE_DRIVER, 54000, 2880, 2956, + { DRM_MODE("2880x480i", DRM_MODE_TYPE_DRIVER, 54000, 2880, 2956, 3204, 3432, 0, 480, 488, 494, 525, 0, DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC | DRM_MODE_FLAG_INTERLACE) }, /* 11 - 2880x480i@60Hz */ - { DRM_MODE("2880x480", DRM_MODE_TYPE_DRIVER, 54000, 2880, 2956, + { DRM_MODE("2880x480i", DRM_MODE_TYPE_DRIVER, 54000, 2880, 2956, 3204, 3432, 0, 480, 488, 494, 525, 0, DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC | DRM_MODE_FLAG_INTERLACE) }, @@ -573,17 +573,17 @@ static const struct drm_display_mode edid_cea_modes[] = { 1760, 1980, 0, 720, 725, 730, 750, 0, DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC) }, /* 20 - 1920x1080i@50Hz */ - { DRM_MODE("1920x1080", DRM_MODE_TYPE_DRIVER, 74250, 1920, 2448, + { DRM_MODE("1920x1080i", DRM_MODE_TYPE_DRIVER, 74250, 1920, 2448, 2492, 2640, 0, 1080, 1084, 1094, 1125, 0, DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC | DRM_MODE_FLAG_INTERLACE) }, /* 21 - 1440x576i@50Hz */ - { DRM_MODE("1440x576", DRM_MODE_TYPE_DRIVER, 27000, 1440, 1464, + { DRM_MODE("1440x576i", DRM_MODE_TYPE_DRIVER, 27000, 1440, 1464, 1590, 1728, 0, 576, 580, 586, 625, 0, DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC | DRM_MODE_FLAG_INTERLACE | DRM_MODE_FLAG_DBLCLK) }, /* 22 - 1440x576i@50Hz */ - { DRM_MODE("1440x576", DRM_MODE_TYPE_DRIVER, 27000, 1440, 1464, + { DRM_MODE("1440x576i", DRM_MODE_TYPE_DRIVER, 27000, 1440, 1464, 1590, 1728, 0, 576, 580, 586, 625, 0, DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC | DRM_MODE_FLAG_INTERLACE | DRM_MODE_FLAG_DBLCLK) }, @@ -598,12 +598,12 @@ static const struct drm_display_mode edid_cea_modes[] = { DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC | DRM_MODE_FLAG_DBLCLK) }, /* 25 - 2880x576i@50Hz */ - { DRM_MODE("2880x576", DRM_MODE_TYPE_DRIVER, 54000, 2880, 2928, + { DRM_MODE("2880x576i", DRM_MODE_TYPE_DRIVER, 54000, 2880, 2928, 3180, 3456, 0, 576, 580, 586, 625, 0, DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC | DRM_MODE_FLAG_INTERLACE) }, /* 26 - 2880x576i@50Hz */ - { DRM_MODE("2880x576", DRM_MODE_TYPE_DRIVER, 54000, 2880, 2928, + { DRM_MODE("2880x576i", DRM_MODE_TYPE_DRIVER, 54000, 2880, 2928, 3180, 3456, 0, 576, 580, 586, 625, 0, DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC | DRM_MODE_FLAG_INTERLACE) }, @@ -656,12 +656,12 @@ static const struct drm_display_mode edid_cea_modes[] = { 3184, 3456, 0, 576, 581, 586, 625, 0, DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) }, /* 39 - 1920x1080i@50Hz */ - { DRM_MODE("1920x1080", DRM_MODE_TYPE_DRIVER, 72000, 1920, 1952, + { DRM_MODE("1920x1080i", DRM_MODE_TYPE_DRIVER, 72000, 1920, 1952, 2120, 2304, 0, 1080, 1126, 1136, 1250, 0, DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_NVSYNC | DRM_MODE_FLAG_INTERLACE) }, /* 40 - 1920x1080i@100Hz */ - { DRM_MODE("1920x1080", DRM_MODE_TYPE_DRIVER, 148500, 1920, 2448, + { DRM_MODE("1920x1080i", DRM_MODE_TYPE_DRIVER, 148500, 1920, 2448, 2492, 2640, 0, 1080, 1084, 1094, 1125, 0, DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC | DRM_MODE_FLAG_INTERLACE) }, @@ -688,7 +688,7 @@ static const struct drm_display_mode edid_cea_modes[] = { DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC | DRM_MODE_FLAG_DBLCLK) }, /* 46 - 1920x1080i@120Hz */ - { DRM_MODE("1920x1080", DRM_MODE_TYPE_DRIVER, 148500, 1920, 2008, + { DRM_MODE("1920x1080i", DRM_MODE_TYPE_DRIVER, 148500, 1920, 2008, 2052, 2200, 0, 1080, 1084, 1094, 1125, 0, DRM_MODE_FLAG_PHSYNC | DRM_MODE_FLAG_PVSYNC | DRM_MODE_FLAG_INTERLACE) }, @@ -705,12 +705,12 @@ static const struct drm_display_mode edid_cea_modes[] = { 798, 858, 0, 480, 489, 495, 525, 0, DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) }, /* 50 - 1440x480i@120Hz */ - { DRM_MODE("1440x480", DRM_MODE_TYPE_DRIVER, 54000, 1440, 1478, + { DRM_MODE("1440x480i", DRM_MODE_TYPE_DRIVER, 54000, 1440, 1478, 1602, 1716, 0, 480, 488, 494, 525, 0, DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC | DRM_MODE_FLAG_INTERLACE | DRM_MODE_FLAG_DBLCLK) }, /* 51 - 1440x480i@120Hz */ - { DRM_MODE("1440x480", DRM_MODE_TYPE_DRIVER, 54000, 1440, 1478, + { DRM_MODE("1440x480i", DRM_MODE_TYPE_DRIVER, 54000, 1440, 1478, 1602, 1716, 0, 480, 488, 494, 525, 0, DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC | DRM_MODE_FLAG_INTERLACE | DRM_MODE_FLAG_DBLCLK) }, @@ -723,12 +723,12 @@ static const struct drm_display_mode edid_cea_modes[] = { 796, 864, 0, 576, 581, 586, 625, 0, DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) }, /* 54 - 1440x576i@200Hz */ - { DRM_MODE("1440x576", DRM_MODE_TYPE_DRIVER, 108000, 1440, 1464, + { DRM_MODE("1440x576i", DRM_MODE_TYPE_DRIVER, 108000, 1440, 1464, 1590, 1728, 0, 576, 580, 586, 625, 0, DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC | DRM_MODE_FLAG_INTERLACE | DRM_MODE_FLAG_DBLCLK) }, /* 55 - 1440x576i@200Hz */ - { DRM_MODE("1440x576", DRM_MODE_TYPE_DRIVER, 108000, 1440, 1464, + { DRM_MODE("1440x576i", DRM_MODE_TYPE_DRIVER, 108000, 1440, 1464, 1590, 1728, 0, 576, 580, 586, 625, 0, DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC | DRM_MODE_FLAG_INTERLACE | DRM_MODE_FLAG_DBLCLK) }, @@ -741,12 +741,12 @@ static const struct drm_display_mode edid_cea_modes[] = { 798, 858, 0, 480, 489, 495, 525, 0, DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) }, /* 58 - 1440x480i@240 */ - { DRM_MODE("1440x480", DRM_MODE_TYPE_DRIVER, 108000, 1440, 1478, + { DRM_MODE("1440x480i", DRM_MODE_TYPE_DRIVER, 108000, 1440, 1478, 1602, 1716, 0, 480, 488, 494, 525, 0, DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC | DRM_MODE_FLAG_INTERLACE | DRM_MODE_FLAG_DBLCLK) }, /* 59 - 1440x480i@240 */ - { DRM_MODE("1440x480", DRM_MODE_TYPE_DRIVER, 108000, 1440, 1478, + { DRM_MODE("1440x480i", DRM_MODE_TYPE_DRIVER, 108000, 1440, 1478, 1602, 1716, 0, 480, 488, 494, 525, 0, DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC | DRM_MODE_FLAG_INTERLACE | DRM_MODE_FLAG_DBLCLK) }, diff --git a/drivers/gpu/drm/drm_fb_cma_helper.c b/drivers/gpu/drm/drm_fb_cma_helper.c new file mode 100644 index 000000000000..09e11a5d921a --- /dev/null +++ b/drivers/gpu/drm/drm_fb_cma_helper.c @@ -0,0 +1,406 @@ +/* + * drm kms/fb cma (contiguous memory allocator) helper functions + * + * Copyright (C) 2012 Analog Device Inc. + * Author: Lars-Peter Clausen <lars@metafoo.de> + * + * Based on udl_fbdev.c + * Copyright (C) 2012 Red Hat + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version 2 + * of the License, or (at your option) any later version. + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + */ + +#include <drm/drmP.h> +#include <drm/drm_crtc.h> +#include <drm/drm_fb_helper.h> +#include <drm/drm_crtc_helper.h> +#include <drm/drm_gem_cma_helper.h> +#include <drm/drm_fb_cma_helper.h> +#include <linux/module.h> + +struct drm_fb_cma { + struct drm_framebuffer fb; + struct drm_gem_cma_object *obj[4]; +}; + +struct drm_fbdev_cma { + struct drm_fb_helper fb_helper; + struct drm_fb_cma *fb; +}; + +static inline struct drm_fbdev_cma *to_fbdev_cma(struct drm_fb_helper *helper) +{ + return container_of(helper, struct drm_fbdev_cma, fb_helper); +} + +static inline struct drm_fb_cma *to_fb_cma(struct drm_framebuffer *fb) +{ + return container_of(fb, struct drm_fb_cma, fb); +} + +static void drm_fb_cma_destroy(struct drm_framebuffer *fb) +{ + struct drm_fb_cma *fb_cma = to_fb_cma(fb); + int i; + + for (i = 0; i < 4; i++) { + if (fb_cma->obj[i]) + drm_gem_object_unreference_unlocked(&fb_cma->obj[i]->base); + } + + drm_framebuffer_cleanup(fb); + kfree(fb_cma); +} + +static int drm_fb_cma_create_handle(struct drm_framebuffer *fb, + struct drm_file *file_priv, unsigned int *handle) +{ + struct drm_fb_cma *fb_cma = to_fb_cma(fb); + + return drm_gem_handle_create(file_priv, + &fb_cma->obj[0]->base, handle); +} + +static struct drm_framebuffer_funcs drm_fb_cma_funcs = { + .destroy = drm_fb_cma_destroy, + .create_handle = drm_fb_cma_create_handle, +}; + +static struct drm_fb_cma *drm_fb_cma_alloc(struct drm_device *dev, + struct drm_mode_fb_cmd2 *mode_cmd, struct drm_gem_cma_object **obj, + unsigned int num_planes) +{ + struct drm_fb_cma *fb_cma; + int ret; + int i; + + fb_cma = kzalloc(sizeof(*fb_cma), GFP_KERNEL); + if (!fb_cma) + return ERR_PTR(-ENOMEM); + + ret = drm_framebuffer_init(dev, &fb_cma->fb, &drm_fb_cma_funcs); + if (ret) { + dev_err(dev->dev, "Failed to initalize framebuffer: %d\n", ret); + kfree(fb_cma); + return ERR_PTR(ret); + } + + drm_helper_mode_fill_fb_struct(&fb_cma->fb, mode_cmd); + + for (i = 0; i < num_planes; i++) + fb_cma->obj[i] = obj[i]; + + return fb_cma; +} + +/** + * drm_fb_cma_create() - (struct drm_mode_config_funcs *)->fb_create callback function + * + * If your hardware has special alignment or pitch requirements these should be + * checked before calling this function. + */ +struct drm_framebuffer *drm_fb_cma_create(struct drm_device *dev, + struct drm_file *file_priv, struct drm_mode_fb_cmd2 *mode_cmd) +{ + struct drm_fb_cma *fb_cma; + struct drm_gem_cma_object *objs[4]; + struct drm_gem_object *obj; + unsigned int hsub; + unsigned int vsub; + int ret; + int i; + + hsub = drm_format_horz_chroma_subsampling(mode_cmd->pixel_format); + vsub = drm_format_vert_chroma_subsampling(mode_cmd->pixel_format); + + for (i = 0; i < drm_format_num_planes(mode_cmd->pixel_format); i++) { + unsigned int width = mode_cmd->width / (i ? hsub : 1); + unsigned int height = mode_cmd->height / (i ? vsub : 1); + unsigned int min_size; + + obj = drm_gem_object_lookup(dev, file_priv, mode_cmd->handles[i]); + if (!obj) { + dev_err(dev->dev, "Failed to lookup GEM object\n"); + ret = -ENXIO; + goto err_gem_object_unreference; + } + + min_size = (height - 1) * mode_cmd->pitches[i] + + width * drm_format_plane_cpp(mode_cmd->pixel_format, i) + + mode_cmd->offsets[i]; + + if (obj->size < min_size) { + drm_gem_object_unreference_unlocked(obj); + ret = -EINVAL; + goto err_gem_object_unreference; + } + objs[i] = to_drm_gem_cma_obj(obj); + } + + fb_cma = drm_fb_cma_alloc(dev, mode_cmd, objs, i); + if (IS_ERR(fb_cma)) { + ret = PTR_ERR(fb_cma); + goto err_gem_object_unreference; + } + + return &fb_cma->fb; + +err_gem_object_unreference: + for (i--; i >= 0; i--) + drm_gem_object_unreference_unlocked(&objs[i]->base); + return ERR_PTR(ret); +} +EXPORT_SYMBOL_GPL(drm_fb_cma_create); + +/** + * drm_fb_cma_get_gem_obj() - Get CMA GEM object for framebuffer + * @fb: The framebuffer + * @plane: Which plane + * + * Return the CMA GEM object for given framebuffer. + * + * This function will usually be called from the CRTC callback functions. + */ +struct drm_gem_cma_object *drm_fb_cma_get_gem_obj(struct drm_framebuffer *fb, + unsigned int plane) +{ + struct drm_fb_cma *fb_cma = to_fb_cma(fb); + + if (plane >= 4) + return NULL; + + return fb_cma->obj[plane]; +} +EXPORT_SYMBOL_GPL(drm_fb_cma_get_gem_obj); + +static struct fb_ops drm_fbdev_cma_ops = { + .owner = THIS_MODULE, + .fb_fillrect = sys_fillrect, + .fb_copyarea = sys_copyarea, + .fb_imageblit = sys_imageblit, + .fb_check_var = drm_fb_helper_check_var, + .fb_set_par = drm_fb_helper_set_par, + .fb_blank = drm_fb_helper_blank, + .fb_pan_display = drm_fb_helper_pan_display, + .fb_setcmap = drm_fb_helper_setcmap, +}; + +static int drm_fbdev_cma_create(struct drm_fb_helper *helper, + struct drm_fb_helper_surface_size *sizes) +{ + struct drm_fbdev_cma *fbdev_cma = to_fbdev_cma(helper); + struct drm_mode_fb_cmd2 mode_cmd = { 0 }; + struct drm_device *dev = helper->dev; + struct drm_gem_cma_object *obj; + struct drm_framebuffer *fb; + unsigned int bytes_per_pixel; + unsigned long offset; + struct fb_info *fbi; + size_t size; + int ret; + + DRM_DEBUG_KMS("surface width(%d), height(%d) and bpp(%d\n", + sizes->surface_width, sizes->surface_height, + sizes->surface_bpp); + + bytes_per_pixel = DIV_ROUND_UP(sizes->surface_bpp, 8); + + mode_cmd.width = sizes->surface_width; + mode_cmd.height = sizes->surface_height; + mode_cmd.pitches[0] = sizes->surface_width * bytes_per_pixel; + mode_cmd.pixel_format = drm_mode_legacy_fb_format(sizes->surface_bpp, + sizes->surface_depth); + + size = mode_cmd.pitches[0] * mode_cmd.height; + obj = drm_gem_cma_create(dev, size); + if (!obj) + return -ENOMEM; + + fbi = framebuffer_alloc(0, dev->dev); + if (!fbi) { + dev_err(dev->dev, "Failed to allocate framebuffer info.\n"); + ret = -ENOMEM; + goto err_drm_gem_cma_free_object; + } + + fbdev_cma->fb = drm_fb_cma_alloc(dev, &mode_cmd, &obj, 1); + if (IS_ERR(fbdev_cma->fb)) { + dev_err(dev->dev, "Failed to allocate DRM framebuffer.\n"); + ret = PTR_ERR(fbdev_cma->fb); + goto err_framebuffer_release; + } + + fb = &fbdev_cma->fb->fb; + helper->fb = fb; + helper->fbdev = fbi; + + fbi->par = helper; + fbi->flags = FBINFO_FLAG_DEFAULT; + fbi->fbops = &drm_fbdev_cma_ops; + + ret = fb_alloc_cmap(&fbi->cmap, 256, 0); + if (ret) { + dev_err(dev->dev, "Failed to allocate color map.\n"); + goto err_drm_fb_cma_destroy; + } + + drm_fb_helper_fill_fix(fbi, fb->pitches[0], fb->depth); + drm_fb_helper_fill_var(fbi, helper, fb->width, fb->height); + + offset = fbi->var.xoffset * bytes_per_pixel; + offset += fbi->var.yoffset * fb->pitches[0]; + + dev->mode_config.fb_base = (resource_size_t)obj->paddr; + fbi->screen_base = obj->vaddr + offset; + fbi->fix.smem_start = (unsigned long)(obj->paddr + offset); + fbi->screen_size = size; + fbi->fix.smem_len = size; + + return 0; + +err_drm_fb_cma_destroy: + drm_fb_cma_destroy(fb); +err_framebuffer_release: + framebuffer_release(fbi); +err_drm_gem_cma_free_object: + drm_gem_cma_free_object(&obj->base); + return ret; +} + +static int drm_fbdev_cma_probe(struct drm_fb_helper *helper, + struct drm_fb_helper_surface_size *sizes) +{ + int ret = 0; + + if (!helper->fb) { + ret = drm_fbdev_cma_create(helper, sizes); + if (ret < 0) + return ret; + ret = 1; + } + + return ret; +} + +static struct drm_fb_helper_funcs drm_fb_cma_helper_funcs = { + .fb_probe = drm_fbdev_cma_probe, +}; + +/** + * drm_fbdev_cma_init() - Allocate and initializes a drm_fbdev_cma struct + * @dev: DRM device + * @preferred_bpp: Preferred bits per pixel for the device + * @num_crtc: Number of CRTCs + * @max_conn_count: Maximum number of connectors + * + * Returns a newly allocated drm_fbdev_cma struct or a ERR_PTR. + */ +struct drm_fbdev_cma *drm_fbdev_cma_init(struct drm_device *dev, + unsigned int preferred_bpp, unsigned int num_crtc, + unsigned int max_conn_count) +{ + struct drm_fbdev_cma *fbdev_cma; + struct drm_fb_helper *helper; + int ret; + + fbdev_cma = kzalloc(sizeof(*fbdev_cma), GFP_KERNEL); + if (!fbdev_cma) { + dev_err(dev->dev, "Failed to allocate drm fbdev.\n"); + return ERR_PTR(-ENOMEM); + } + + fbdev_cma->fb_helper.funcs = &drm_fb_cma_helper_funcs; + helper = &fbdev_cma->fb_helper; + + ret = drm_fb_helper_init(dev, helper, num_crtc, max_conn_count); + if (ret < 0) { + dev_err(dev->dev, "Failed to initialize drm fb helper.\n"); + goto err_free; + } + + ret = drm_fb_helper_single_add_all_connectors(helper); + if (ret < 0) { + dev_err(dev->dev, "Failed to add connectors.\n"); + goto err_drm_fb_helper_fini; + + } + + ret = drm_fb_helper_initial_config(helper, preferred_bpp); + if (ret < 0) { + dev_err(dev->dev, "Failed to set inital hw configuration.\n"); + goto err_drm_fb_helper_fini; + } + + return fbdev_cma; + +err_drm_fb_helper_fini: + drm_fb_helper_fini(helper); +err_free: + kfree(fbdev_cma); + + return ERR_PTR(ret); +} +EXPORT_SYMBOL_GPL(drm_fbdev_cma_init); + +/** + * drm_fbdev_cma_fini() - Free drm_fbdev_cma struct + * @fbdev_cma: The drm_fbdev_cma struct + */ +void drm_fbdev_cma_fini(struct drm_fbdev_cma *fbdev_cma) +{ + if (fbdev_cma->fb_helper.fbdev) { + struct fb_info *info; + int ret; + + info = fbdev_cma->fb_helper.fbdev; + ret = unregister_framebuffer(info); + if (ret < 0) + DRM_DEBUG_KMS("failed unregister_framebuffer()\n"); + + if (info->cmap.len) + fb_dealloc_cmap(&info->cmap); + + framebuffer_release(info); + } + + if (fbdev_cma->fb) + drm_fb_cma_destroy(&fbdev_cma->fb->fb); + + drm_fb_helper_fini(&fbdev_cma->fb_helper); + kfree(fbdev_cma); +} +EXPORT_SYMBOL_GPL(drm_fbdev_cma_fini); + +/** + * drm_fbdev_cma_restore_mode() - Restores initial framebuffer mode + * @fbdev_cma: The drm_fbdev_cma struct, may be NULL + * + * This function is usually called from the DRM drivers lastclose callback. + */ +void drm_fbdev_cma_restore_mode(struct drm_fbdev_cma *fbdev_cma) +{ + if (fbdev_cma) + drm_fb_helper_restore_fbdev_mode(&fbdev_cma->fb_helper); +} +EXPORT_SYMBOL_GPL(drm_fbdev_cma_restore_mode); + +/** + * drm_fbdev_cma_hotplug_event() - Poll for hotpulug events + * @fbdev_cma: The drm_fbdev_cma struct, may be NULL + * + * This function is usually called from the DRM drivers output_poll_changed + * callback. + */ +void drm_fbdev_cma_hotplug_event(struct drm_fbdev_cma *fbdev_cma) +{ + if (fbdev_cma) + drm_fb_helper_hotplug_event(&fbdev_cma->fb_helper); +} +EXPORT_SYMBOL_GPL(drm_fbdev_cma_hotplug_event); diff --git a/drivers/gpu/drm/drm_fb_helper.c b/drivers/gpu/drm/drm_fb_helper.c index f546d1e8af82..b5d05f58c15a 100644 --- a/drivers/gpu/drm/drm_fb_helper.c +++ b/drivers/gpu/drm/drm_fb_helper.c @@ -236,7 +236,7 @@ bool drm_fb_helper_restore_fbdev_mode(struct drm_fb_helper *fb_helper) } EXPORT_SYMBOL(drm_fb_helper_restore_fbdev_mode); -bool drm_fb_helper_force_kernel_mode(void) +static bool drm_fb_helper_force_kernel_mode(void) { bool ret, error = false; struct drm_fb_helper *helper; @@ -330,7 +330,7 @@ static void drm_fb_helper_dpms(struct fb_info *info, int dpms_mode) /* Walk the connectors & encoders on this fb turning them on/off */ for (j = 0; j < fb_helper->connector_count; j++) { connector = fb_helper->connector_info[j]->connector; - drm_helper_connector_dpms(connector, dpms_mode); + connector->funcs->dpms(connector, dpms_mode); drm_connector_property_set_value(connector, dev->mode_config.dpms_property, dpms_mode); } @@ -1230,7 +1230,6 @@ static void drm_setup_crtcs(struct drm_fb_helper *fb_helper) struct drm_device *dev = fb_helper->dev; struct drm_fb_helper_crtc **crtcs; struct drm_display_mode **modes; - struct drm_encoder *encoder; struct drm_mode_set *modeset; bool *enabled; int width, height; @@ -1241,11 +1240,6 @@ static void drm_setup_crtcs(struct drm_fb_helper *fb_helper) width = dev->mode_config.max_width; height = dev->mode_config.max_height; - /* clean out all the encoder/crtc combos */ - list_for_each_entry(encoder, &dev->mode_config.encoder_list, head) { - encoder->crtc = NULL; - } - crtcs = kcalloc(dev->mode_config.num_connector, sizeof(struct drm_fb_helper_crtc *), GFP_KERNEL); modes = kcalloc(dev->mode_config.num_connector, diff --git a/drivers/gpu/drm/drm_gem_cma_helper.c b/drivers/gpu/drm/drm_gem_cma_helper.c new file mode 100644 index 000000000000..1aa8fee1e865 --- /dev/null +++ b/drivers/gpu/drm/drm_gem_cma_helper.c @@ -0,0 +1,251 @@ +/* + * drm gem CMA (contiguous memory allocator) helper functions + * + * Copyright (C) 2012 Sascha Hauer, Pengutronix + * + * Based on Samsung Exynos code + * + * Copyright (c) 2011 Samsung Electronics Co., Ltd. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version 2 + * of the License, or (at your option) any later version. + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + */ + +#include <linux/mm.h> +#include <linux/slab.h> +#include <linux/mutex.h> +#include <linux/export.h> +#include <linux/dma-mapping.h> + +#include <drm/drmP.h> +#include <drm/drm.h> +#include <drm/drm_gem_cma_helper.h> + +static unsigned int get_gem_mmap_offset(struct drm_gem_object *obj) +{ + return (unsigned int)obj->map_list.hash.key << PAGE_SHIFT; +} + +static void drm_gem_cma_buf_destroy(struct drm_device *drm, + struct drm_gem_cma_object *cma_obj) +{ + dma_free_writecombine(drm->dev, cma_obj->base.size, cma_obj->vaddr, + cma_obj->paddr); +} + +/* + * drm_gem_cma_create - allocate an object with the given size + * + * returns a struct drm_gem_cma_object* on success or ERR_PTR values + * on failure. + */ +struct drm_gem_cma_object *drm_gem_cma_create(struct drm_device *drm, + unsigned int size) +{ + struct drm_gem_cma_object *cma_obj; + struct drm_gem_object *gem_obj; + int ret; + + size = round_up(size, PAGE_SIZE); + + cma_obj = kzalloc(sizeof(*cma_obj), GFP_KERNEL); + if (!cma_obj) + return ERR_PTR(-ENOMEM); + + cma_obj->vaddr = dma_alloc_writecombine(drm->dev, size, + &cma_obj->paddr, GFP_KERNEL | __GFP_NOWARN); + if (!cma_obj->vaddr) { + dev_err(drm->dev, "failed to allocate buffer with size %d\n", size); + ret = -ENOMEM; + goto err_dma_alloc; + } + + gem_obj = &cma_obj->base; + + ret = drm_gem_object_init(drm, gem_obj, size); + if (ret) + goto err_obj_init; + + ret = drm_gem_create_mmap_offset(gem_obj); + if (ret) + goto err_create_mmap_offset; + + return cma_obj; + +err_create_mmap_offset: + drm_gem_object_release(gem_obj); + +err_obj_init: + drm_gem_cma_buf_destroy(drm, cma_obj); + +err_dma_alloc: + kfree(cma_obj); + + return ERR_PTR(ret); +} +EXPORT_SYMBOL_GPL(drm_gem_cma_create); + +/* + * drm_gem_cma_create_with_handle - allocate an object with the given + * size and create a gem handle on it + * + * returns a struct drm_gem_cma_object* on success or ERR_PTR values + * on failure. + */ +static struct drm_gem_cma_object *drm_gem_cma_create_with_handle( + struct drm_file *file_priv, + struct drm_device *drm, unsigned int size, + unsigned int *handle) +{ + struct drm_gem_cma_object *cma_obj; + struct drm_gem_object *gem_obj; + int ret; + + cma_obj = drm_gem_cma_create(drm, size); + if (IS_ERR(cma_obj)) + return cma_obj; + + gem_obj = &cma_obj->base; + + /* + * allocate a id of idr table where the obj is registered + * and handle has the id what user can see. + */ + ret = drm_gem_handle_create(file_priv, gem_obj, handle); + if (ret) + goto err_handle_create; + + /* drop reference from allocate - handle holds it now. */ + drm_gem_object_unreference_unlocked(gem_obj); + + return cma_obj; + +err_handle_create: + drm_gem_cma_free_object(gem_obj); + + return ERR_PTR(ret); +} + +/* + * drm_gem_cma_free_object - (struct drm_driver)->gem_free_object callback + * function + */ +void drm_gem_cma_free_object(struct drm_gem_object *gem_obj) +{ + struct drm_gem_cma_object *cma_obj; + + if (gem_obj->map_list.map) + drm_gem_free_mmap_offset(gem_obj); + + drm_gem_object_release(gem_obj); + + cma_obj = to_drm_gem_cma_obj(gem_obj); + + drm_gem_cma_buf_destroy(gem_obj->dev, cma_obj); + + kfree(cma_obj); +} +EXPORT_SYMBOL_GPL(drm_gem_cma_free_object); + +/* + * drm_gem_cma_dumb_create - (struct drm_driver)->dumb_create callback + * function + * + * This aligns the pitch and size arguments to the minimum required. wrap + * this into your own function if you need bigger alignment. + */ +int drm_gem_cma_dumb_create(struct drm_file *file_priv, + struct drm_device *dev, struct drm_mode_create_dumb *args) +{ + struct drm_gem_cma_object *cma_obj; + int min_pitch = DIV_ROUND_UP(args->width * args->bpp, 8); + + if (args->pitch < min_pitch) + args->pitch = min_pitch; + + if (args->size < args->pitch * args->height) + args->size = args->pitch * args->height; + + cma_obj = drm_gem_cma_create_with_handle(file_priv, dev, + args->size, &args->handle); + if (IS_ERR(cma_obj)) + return PTR_ERR(cma_obj); + + return 0; +} +EXPORT_SYMBOL_GPL(drm_gem_cma_dumb_create); + +/* + * drm_gem_cma_dumb_map_offset - (struct drm_driver)->dumb_map_offset callback + * function + */ +int drm_gem_cma_dumb_map_offset(struct drm_file *file_priv, + struct drm_device *drm, uint32_t handle, uint64_t *offset) +{ + struct drm_gem_object *gem_obj; + + mutex_lock(&drm->struct_mutex); + + gem_obj = drm_gem_object_lookup(drm, file_priv, handle); + if (!gem_obj) { + dev_err(drm->dev, "failed to lookup gem object\n"); + mutex_unlock(&drm->struct_mutex); + return -EINVAL; + } + + *offset = get_gem_mmap_offset(gem_obj); + + drm_gem_object_unreference(gem_obj); + + mutex_unlock(&drm->struct_mutex); + + return 0; +} +EXPORT_SYMBOL_GPL(drm_gem_cma_dumb_map_offset); + +const struct vm_operations_struct drm_gem_cma_vm_ops = { + .open = drm_gem_vm_open, + .close = drm_gem_vm_close, +}; +EXPORT_SYMBOL_GPL(drm_gem_cma_vm_ops); + +/* + * drm_gem_cma_mmap - (struct file_operation)->mmap callback function + */ +int drm_gem_cma_mmap(struct file *filp, struct vm_area_struct *vma) +{ + struct drm_gem_object *gem_obj; + struct drm_gem_cma_object *cma_obj; + int ret; + + ret = drm_gem_mmap(filp, vma); + if (ret) + return ret; + + gem_obj = vma->vm_private_data; + cma_obj = to_drm_gem_cma_obj(gem_obj); + + ret = remap_pfn_range(vma, vma->vm_start, cma_obj->paddr >> PAGE_SHIFT, + vma->vm_end - vma->vm_start, vma->vm_page_prot); + if (ret) + drm_gem_vm_close(vma); + + return ret; +} +EXPORT_SYMBOL_GPL(drm_gem_cma_mmap); + +/* + * drm_gem_cma_dumb_destroy - (struct drm_driver)->dumb_destroy callback function + */ +int drm_gem_cma_dumb_destroy(struct drm_file *file_priv, + struct drm_device *drm, unsigned int handle) +{ + return drm_gem_handle_delete(file_priv, handle); +} +EXPORT_SYMBOL_GPL(drm_gem_cma_dumb_destroy); diff --git a/drivers/gpu/drm/drm_irq.c b/drivers/gpu/drm/drm_irq.c index 03f16f352fe2..076c4a86ff84 100644 --- a/drivers/gpu/drm/drm_irq.c +++ b/drivers/gpu/drm/drm_irq.c @@ -1236,7 +1236,7 @@ done: return ret; } -void drm_handle_vblank_events(struct drm_device *dev, int crtc) +static void drm_handle_vblank_events(struct drm_device *dev, int crtc) { struct drm_pending_vblank_event *e, *t; struct timeval now; diff --git a/drivers/gpu/drm/drm_vm.c b/drivers/gpu/drm/drm_vm.c index 961ee08927fe..6fed21502313 100644 --- a/drivers/gpu/drm/drm_vm.c +++ b/drivers/gpu/drm/drm_vm.c @@ -62,7 +62,7 @@ static pgprot_t drm_io_prot(uint32_t map_type, struct vm_area_struct *vma) tmp = pgprot_writecombine(tmp); else tmp = pgprot_noncached(tmp); -#elif defined(__sparc__) || defined(__arm__) +#elif defined(__sparc__) || defined(__arm__) || defined(__mips__) tmp = pgprot_noncached(tmp); #endif return tmp; @@ -619,20 +619,11 @@ int drm_mmap_locked(struct file *filp, struct vm_area_struct *vma) offset = drm_core_get_reg_ofs(dev); vma->vm_flags |= VM_IO; /* not in core dump */ vma->vm_page_prot = drm_io_prot(map->type, vma); -#if !defined(__arm__) if (io_remap_pfn_range(vma, vma->vm_start, (map->offset + offset) >> PAGE_SHIFT, vma->vm_end - vma->vm_start, vma->vm_page_prot)) return -EAGAIN; -#else - if (remap_pfn_range(vma, vma->vm_start, - (map->offset + offset) >> PAGE_SHIFT, - vma->vm_end - vma->vm_start, - vma->vm_page_prot)) - return -EAGAIN; -#endif - DRM_DEBUG(" Type = %d; start = 0x%lx, end = 0x%lx," " offset = 0x%llx\n", map->type, diff --git a/drivers/gpu/drm/exynos/exynos_drm_connector.c b/drivers/gpu/drm/exynos/exynos_drm_connector.c index d9568198c300..9dce3b9c3896 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_connector.c +++ b/drivers/gpu/drm/exynos/exynos_drm_connector.c @@ -147,9 +147,7 @@ static int exynos_drm_connector_get_modes(struct drm_connector *connector) drm_mode_connector_update_edid_property(connector, edid); count = drm_add_edid_modes(connector, edid); - - kfree(connector->display_info.raw_edid); - connector->display_info.raw_edid = edid; + kfree(edid); } else { struct drm_display_mode *mode = drm_mode_create(connector->dev); struct exynos_drm_panel_info *panel; diff --git a/drivers/gpu/drm/exynos/exynos_drm_fbdev.c b/drivers/gpu/drm/exynos/exynos_drm_fbdev.c index d5586cc75163..f4ac43356583 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_fbdev.c +++ b/drivers/gpu/drm/exynos/exynos_drm_fbdev.c @@ -266,8 +266,8 @@ static void exynos_drm_fbdev_destroy(struct drm_device *dev, /* release drm framebuffer and real buffer */ if (fb_helper->fb && fb_helper->fb->funcs) { fb = fb_helper->fb; - if (fb && fb->funcs->destroy) - fb->funcs->destroy(fb); + if (fb) + drm_framebuffer_remove(fb); } /* release linux framebuffer */ diff --git a/drivers/gpu/drm/exynos/exynos_drm_vidi.c b/drivers/gpu/drm/exynos/exynos_drm_vidi.c index 537027a74fd5..e364165f1a2a 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_vidi.c +++ b/drivers/gpu/drm/exynos/exynos_drm_vidi.c @@ -102,7 +102,6 @@ static int vidi_get_edid(struct device *dev, struct drm_connector *connector, u8 *edid, int len) { struct vidi_context *ctx = get_vidi_context(dev); - struct edid *raw_edid; DRM_DEBUG_KMS("%s\n", __FILE__); @@ -115,18 +114,6 @@ static int vidi_get_edid(struct device *dev, struct drm_connector *connector, return -EFAULT; } - raw_edid = kzalloc(len, GFP_KERNEL); - if (!raw_edid) { - DRM_DEBUG_KMS("failed to allocate raw_edid.\n"); - return -ENOMEM; - } - - memcpy(raw_edid, ctx->raw_edid, min((1 + ctx->raw_edid->extensions) - * EDID_LENGTH, len)); - - /* attach the edid data to connector. */ - connector->display_info.raw_edid = (char *)raw_edid; - memcpy(edid, ctx->raw_edid, min((1 + ctx->raw_edid->extensions) * EDID_LENGTH, len)); diff --git a/drivers/gpu/drm/gma500/Makefile b/drivers/gpu/drm/gma500/Makefile index abfa2a93f0d0..7a2d40a5c1e1 100644 --- a/drivers/gpu/drm/gma500/Makefile +++ b/drivers/gpu/drm/gma500/Makefile @@ -3,7 +3,7 @@ # ccflags-y += -I$(srctree)/include/drm -gma500_gfx-y += gem_glue.o \ +gma500_gfx-y += \ accel_2d.o \ backlight.o \ framebuffer.o \ @@ -30,7 +30,8 @@ gma500_gfx-$(CONFIG_DRM_GMA3600) += cdv_device.o \ cdv_intel_crt.o \ cdv_intel_display.o \ cdv_intel_hdmi.o \ - cdv_intel_lvds.o + cdv_intel_lvds.o \ + cdv_intel_dp.o gma500_gfx-$(CONFIG_DRM_GMA600) += oaktrail_device.o \ oaktrail_crtc.o \ diff --git a/drivers/gpu/drm/gma500/backlight.c b/drivers/gpu/drm/gma500/backlight.c index 20793951fcac..143eba3309c5 100644 --- a/drivers/gpu/drm/gma500/backlight.c +++ b/drivers/gpu/drm/gma500/backlight.c @@ -26,10 +26,55 @@ #include "intel_bios.h" #include "power.h" +static void do_gma_backlight_set(struct drm_device *dev) +{ +#ifdef CONFIG_BACKLIGHT_CLASS_DEVICE + struct drm_psb_private *dev_priv = dev->dev_private; + backlight_update_status(dev_priv->backlight_device); +#endif +} + +void gma_backlight_enable(struct drm_device *dev) +{ +#ifdef CONFIG_BACKLIGHT_CLASS_DEVICE + struct drm_psb_private *dev_priv = dev->dev_private; + dev_priv->backlight_enabled = true; + if (dev_priv->backlight_device) { + dev_priv->backlight_device->props.brightness = dev_priv->backlight_level; + do_gma_backlight_set(dev); + } +#endif +} + +void gma_backlight_disable(struct drm_device *dev) +{ +#ifdef CONFIG_BACKLIGHT_CLASS_DEVICE + struct drm_psb_private *dev_priv = dev->dev_private; + dev_priv->backlight_enabled = false; + if (dev_priv->backlight_device) { + dev_priv->backlight_device->props.brightness = 0; + do_gma_backlight_set(dev); + } +#endif +} + +void gma_backlight_set(struct drm_device *dev, int v) +{ +#ifdef CONFIG_BACKLIGHT_CLASS_DEVICE + struct drm_psb_private *dev_priv = dev->dev_private; + dev_priv->backlight_level = v; + if (dev_priv->backlight_device && dev_priv->backlight_enabled) { + dev_priv->backlight_device->props.brightness = v; + do_gma_backlight_set(dev); + } +#endif +} + int gma_backlight_init(struct drm_device *dev) { #ifdef CONFIG_BACKLIGHT_CLASS_DEVICE struct drm_psb_private *dev_priv = dev->dev_private; + dev_priv->backlight_enabled = true; return dev_priv->ops->backlight_init(dev); #else return 0; diff --git a/drivers/gpu/drm/gma500/cdv_device.c b/drivers/gpu/drm/gma500/cdv_device.c index b7e7b49d8f62..bfc2f397019a 100644 --- a/drivers/gpu/drm/gma500/cdv_device.c +++ b/drivers/gpu/drm/gma500/cdv_device.c @@ -58,10 +58,17 @@ static int cdv_output_init(struct drm_device *dev) cdv_intel_lvds_init(dev, &dev_priv->mode_dev); /* These bits indicate HDMI not SDVO on CDV */ - if (REG_READ(SDVOB) & SDVO_DETECTED) + if (REG_READ(SDVOB) & SDVO_DETECTED) { cdv_hdmi_init(dev, &dev_priv->mode_dev, SDVOB); - if (REG_READ(SDVOC) & SDVO_DETECTED) + if (REG_READ(DP_B) & DP_DETECTED) + cdv_intel_dp_init(dev, &dev_priv->mode_dev, DP_B); + } + + if (REG_READ(SDVOC) & SDVO_DETECTED) { cdv_hdmi_init(dev, &dev_priv->mode_dev, SDVOC); + if (REG_READ(DP_C) & DP_DETECTED) + cdv_intel_dp_init(dev, &dev_priv->mode_dev, DP_C); + } return 0; } @@ -163,6 +170,7 @@ static int cdv_backlight_init(struct drm_device *dev) cdv_get_brightness(cdv_backlight_device); backlight_update_status(cdv_backlight_device); dev_priv->backlight_device = cdv_backlight_device; + dev_priv->backlight_enabled = true; return 0; } @@ -449,6 +457,7 @@ static void cdv_get_core_freq(struct drm_device *dev) case 6: case 7: dev_priv->core_freq = 266; + break; default: dev_priv->core_freq = 0; } @@ -488,6 +497,65 @@ static void cdv_hotplug_enable(struct drm_device *dev, bool on) } } +static const char *force_audio_names[] = { + "off", + "auto", + "on", +}; + +void cdv_intel_attach_force_audio_property(struct drm_connector *connector) +{ + struct drm_device *dev = connector->dev; + struct drm_psb_private *dev_priv = dev->dev_private; + struct drm_property *prop; + int i; + + prop = dev_priv->force_audio_property; + if (prop == NULL) { + prop = drm_property_create(dev, DRM_MODE_PROP_ENUM, + "audio", + ARRAY_SIZE(force_audio_names)); + if (prop == NULL) + return; + + for (i = 0; i < ARRAY_SIZE(force_audio_names); i++) + drm_property_add_enum(prop, i, i-1, force_audio_names[i]); + + dev_priv->force_audio_property = prop; + } + drm_connector_attach_property(connector, prop, 0); +} + + +static const char *broadcast_rgb_names[] = { + "Full", + "Limited 16:235", +}; + +void cdv_intel_attach_broadcast_rgb_property(struct drm_connector *connector) +{ + struct drm_device *dev = connector->dev; + struct drm_psb_private *dev_priv = dev->dev_private; + struct drm_property *prop; + int i; + + prop = dev_priv->broadcast_rgb_property; + if (prop == NULL) { + prop = drm_property_create(dev, DRM_MODE_PROP_ENUM, + "Broadcast RGB", + ARRAY_SIZE(broadcast_rgb_names)); + if (prop == NULL) + return; + + for (i = 0; i < ARRAY_SIZE(broadcast_rgb_names); i++) + drm_property_add_enum(prop, i, i, broadcast_rgb_names[i]); + + dev_priv->broadcast_rgb_property = prop; + } + + drm_connector_attach_property(connector, prop, 0); +} + /* Cedarview */ static const struct psb_offset cdv_regmap[2] = { { diff --git a/drivers/gpu/drm/gma500/cdv_intel_display.c b/drivers/gpu/drm/gma500/cdv_intel_display.c index a68509ba22a8..3cfd0931fbfb 100644 --- a/drivers/gpu/drm/gma500/cdv_intel_display.c +++ b/drivers/gpu/drm/gma500/cdv_intel_display.c @@ -57,15 +57,26 @@ struct cdv_intel_clock_t { struct cdv_intel_limit_t { struct cdv_intel_range_t dot, vco, n, m, m1, m2, p, p1; struct cdv_intel_p2_t p2; + bool (*find_pll)(const struct cdv_intel_limit_t *, struct drm_crtc *, + int, int, struct cdv_intel_clock_t *); }; +static bool cdv_intel_find_best_PLL(const struct cdv_intel_limit_t *limit, + struct drm_crtc *crtc, int target, int refclk, + struct cdv_intel_clock_t *best_clock); +static bool cdv_intel_find_dp_pll(const struct cdv_intel_limit_t *limit, struct drm_crtc *crtc, int target, + int refclk, + struct cdv_intel_clock_t *best_clock); + #define CDV_LIMIT_SINGLE_LVDS_96 0 #define CDV_LIMIT_SINGLE_LVDS_100 1 #define CDV_LIMIT_DAC_HDMI_27 2 #define CDV_LIMIT_DAC_HDMI_96 3 +#define CDV_LIMIT_DP_27 4 +#define CDV_LIMIT_DP_100 5 static const struct cdv_intel_limit_t cdv_intel_limits[] = { - { /* CDV_SIGNLE_LVDS_96MHz */ + { /* CDV_SINGLE_LVDS_96MHz */ .dot = {.min = 20000, .max = 115500}, .vco = {.min = 1800000, .max = 3600000}, .n = {.min = 2, .max = 6}, @@ -76,6 +87,7 @@ static const struct cdv_intel_limit_t cdv_intel_limits[] = { .p1 = {.min = 2, .max = 10}, .p2 = {.dot_limit = 200000, .p2_slow = 14, .p2_fast = 14}, + .find_pll = cdv_intel_find_best_PLL, }, { /* CDV_SINGLE_LVDS_100MHz */ .dot = {.min = 20000, .max = 115500}, @@ -90,6 +102,7 @@ static const struct cdv_intel_limit_t cdv_intel_limits[] = { * is 80-224Mhz. Prefer single channel as much as possible. */ .p2 = {.dot_limit = 200000, .p2_slow = 14, .p2_fast = 14}, + .find_pll = cdv_intel_find_best_PLL, }, { /* CDV_DAC_HDMI_27MHz */ .dot = {.min = 20000, .max = 400000}, @@ -101,6 +114,7 @@ static const struct cdv_intel_limit_t cdv_intel_limits[] = { .p = {.min = 5, .max = 90}, .p1 = {.min = 1, .max = 9}, .p2 = {.dot_limit = 225000, .p2_slow = 10, .p2_fast = 5}, + .find_pll = cdv_intel_find_best_PLL, }, { /* CDV_DAC_HDMI_96MHz */ .dot = {.min = 20000, .max = 400000}, @@ -112,7 +126,32 @@ static const struct cdv_intel_limit_t cdv_intel_limits[] = { .p = {.min = 5, .max = 100}, .p1 = {.min = 1, .max = 10}, .p2 = {.dot_limit = 225000, .p2_slow = 10, .p2_fast = 5}, + .find_pll = cdv_intel_find_best_PLL, + }, + { /* CDV_DP_27MHz */ + .dot = {.min = 160000, .max = 272000}, + .vco = {.min = 1809000, .max = 3564000}, + .n = {.min = 1, .max = 1}, + .m = {.min = 67, .max = 132}, + .m1 = {.min = 0, .max = 0}, + .m2 = {.min = 65, .max = 130}, + .p = {.min = 5, .max = 90}, + .p1 = {.min = 1, .max = 9}, + .p2 = {.dot_limit = 225000, .p2_slow = 10, .p2_fast = 10}, + .find_pll = cdv_intel_find_dp_pll, }, + { /* CDV_DP_100MHz */ + .dot = {.min = 160000, .max = 272000}, + .vco = {.min = 1800000, .max = 3600000}, + .n = {.min = 2, .max = 6}, + .m = {.min = 60, .max = 164}, + .m1 = {.min = 0, .max = 0}, + .m2 = {.min = 58, .max = 162}, + .p = {.min = 5, .max = 100}, + .p1 = {.min = 1, .max = 10}, + .p2 = {.dot_limit = 225000, .p2_slow = 10, .p2_fast = 10}, + .find_pll = cdv_intel_find_dp_pll, + } }; #define _wait_for(COND, MS, W) ({ \ @@ -132,7 +171,7 @@ static const struct cdv_intel_limit_t cdv_intel_limits[] = { #define wait_for(COND, MS) _wait_for(COND, MS, 1) -static int cdv_sb_read(struct drm_device *dev, u32 reg, u32 *val) +int cdv_sb_read(struct drm_device *dev, u32 reg, u32 *val) { int ret; @@ -159,7 +198,7 @@ static int cdv_sb_read(struct drm_device *dev, u32 reg, u32 *val) return 0; } -static int cdv_sb_write(struct drm_device *dev, u32 reg, u32 val) +int cdv_sb_write(struct drm_device *dev, u32 reg, u32 val) { int ret; static bool dpio_debug = true; @@ -201,7 +240,7 @@ static int cdv_sb_write(struct drm_device *dev, u32 reg, u32 val) /* Reset the DPIO configuration register. The BIOS does this at every * mode set. */ -static void cdv_sb_reset(struct drm_device *dev) +void cdv_sb_reset(struct drm_device *dev) { REG_WRITE(DPIO_CFG, 0); @@ -216,7 +255,7 @@ static void cdv_sb_reset(struct drm_device *dev) */ static int cdv_dpll_set_clock_cdv(struct drm_device *dev, struct drm_crtc *crtc, - struct cdv_intel_clock_t *clock, bool is_lvds) + struct cdv_intel_clock_t *clock, bool is_lvds, u32 ddi_select) { struct psb_intel_crtc *psb_crtc = to_psb_intel_crtc(crtc); int pipe = psb_crtc->pipe; @@ -259,7 +298,7 @@ cdv_dpll_set_clock_cdv(struct drm_device *dev, struct drm_crtc *crtc, ref_value &= ~(REF_CLK_MASK); /* use DPLL_A for pipeB on CRT/HDMI */ - if (pipe == 1 && !is_lvds) { + if (pipe == 1 && !is_lvds && !(ddi_select & DP_MASK)) { DRM_DEBUG_KMS("use DPLLA for pipe B\n"); ref_value |= REF_CLK_DPLLA; } else { @@ -336,30 +375,33 @@ cdv_dpll_set_clock_cdv(struct drm_device *dev, struct drm_crtc *crtc, if (ret) return ret; - lane_reg = PSB_LANE0; - cdv_sb_read(dev, lane_reg, &lane_value); - lane_value &= ~(LANE_PLL_MASK); - lane_value |= LANE_PLL_ENABLE | LANE_PLL_PIPE(pipe); - cdv_sb_write(dev, lane_reg, lane_value); - - lane_reg = PSB_LANE1; - cdv_sb_read(dev, lane_reg, &lane_value); - lane_value &= ~(LANE_PLL_MASK); - lane_value |= LANE_PLL_ENABLE | LANE_PLL_PIPE(pipe); - cdv_sb_write(dev, lane_reg, lane_value); - - lane_reg = PSB_LANE2; - cdv_sb_read(dev, lane_reg, &lane_value); - lane_value &= ~(LANE_PLL_MASK); - lane_value |= LANE_PLL_ENABLE | LANE_PLL_PIPE(pipe); - cdv_sb_write(dev, lane_reg, lane_value); - - lane_reg = PSB_LANE3; - cdv_sb_read(dev, lane_reg, &lane_value); - lane_value &= ~(LANE_PLL_MASK); - lane_value |= LANE_PLL_ENABLE | LANE_PLL_PIPE(pipe); - cdv_sb_write(dev, lane_reg, lane_value); - + if (ddi_select) { + if ((ddi_select & DDI_MASK) == DDI0_SELECT) { + lane_reg = PSB_LANE0; + cdv_sb_read(dev, lane_reg, &lane_value); + lane_value &= ~(LANE_PLL_MASK); + lane_value |= LANE_PLL_ENABLE | LANE_PLL_PIPE(pipe); + cdv_sb_write(dev, lane_reg, lane_value); + + lane_reg = PSB_LANE1; + cdv_sb_read(dev, lane_reg, &lane_value); + lane_value &= ~(LANE_PLL_MASK); + lane_value |= LANE_PLL_ENABLE | LANE_PLL_PIPE(pipe); + cdv_sb_write(dev, lane_reg, lane_value); + } else { + lane_reg = PSB_LANE2; + cdv_sb_read(dev, lane_reg, &lane_value); + lane_value &= ~(LANE_PLL_MASK); + lane_value |= LANE_PLL_ENABLE | LANE_PLL_PIPE(pipe); + cdv_sb_write(dev, lane_reg, lane_value); + + lane_reg = PSB_LANE3; + cdv_sb_read(dev, lane_reg, &lane_value); + lane_value &= ~(LANE_PLL_MASK); + lane_value |= LANE_PLL_ENABLE | LANE_PLL_PIPE(pipe); + cdv_sb_write(dev, lane_reg, lane_value); + } + } return 0; } @@ -396,6 +438,12 @@ static const struct cdv_intel_limit_t *cdv_intel_limit(struct drm_crtc *crtc, limit = &cdv_intel_limits[CDV_LIMIT_SINGLE_LVDS_96]; else limit = &cdv_intel_limits[CDV_LIMIT_SINGLE_LVDS_100]; + } else if (psb_intel_pipe_has_type(crtc, INTEL_OUTPUT_DISPLAYPORT) || + psb_intel_pipe_has_type(crtc, INTEL_OUTPUT_EDP)) { + if (refclk == 27000) + limit = &cdv_intel_limits[CDV_LIMIT_DP_27]; + else + limit = &cdv_intel_limits[CDV_LIMIT_DP_100]; } else { if (refclk == 27000) limit = &cdv_intel_limits[CDV_LIMIT_DAC_HDMI_27]; @@ -438,13 +486,12 @@ static bool cdv_intel_PLL_is_valid(struct drm_crtc *crtc, return true; } -static bool cdv_intel_find_best_PLL(struct drm_crtc *crtc, int target, - int refclk, - struct cdv_intel_clock_t *best_clock) +static bool cdv_intel_find_best_PLL(const struct cdv_intel_limit_t *limit, + struct drm_crtc *crtc, int target, int refclk, + struct cdv_intel_clock_t *best_clock) { struct drm_device *dev = crtc->dev; struct cdv_intel_clock_t clock; - const struct cdv_intel_limit_t *limit = cdv_intel_limit(crtc, refclk); int err = target; @@ -498,6 +545,49 @@ static bool cdv_intel_find_best_PLL(struct drm_crtc *crtc, int target, return err != target; } +static bool cdv_intel_find_dp_pll(const struct cdv_intel_limit_t *limit, struct drm_crtc *crtc, int target, + int refclk, + struct cdv_intel_clock_t *best_clock) +{ + struct cdv_intel_clock_t clock; + if (refclk == 27000) { + if (target < 200000) { + clock.p1 = 2; + clock.p2 = 10; + clock.n = 1; + clock.m1 = 0; + clock.m2 = 118; + } else { + clock.p1 = 1; + clock.p2 = 10; + clock.n = 1; + clock.m1 = 0; + clock.m2 = 98; + } + } else if (refclk == 100000) { + if (target < 200000) { + clock.p1 = 2; + clock.p2 = 10; + clock.n = 5; + clock.m1 = 0; + clock.m2 = 160; + } else { + clock.p1 = 1; + clock.p2 = 10; + clock.n = 5; + clock.m1 = 0; + clock.m2 = 133; + } + } else + return false; + clock.m = clock.m2 + 2; + clock.p = clock.p1 * clock.p2; + clock.vco = (refclk * clock.m) / clock.n; + clock.dot = clock.vco / clock.p; + memcpy(best_clock, &clock, sizeof(struct cdv_intel_clock_t)); + return true; +} + static int cdv_intel_pipe_set_base(struct drm_crtc *crtc, int x, int y, struct drm_framebuffer *old_fb) { @@ -791,7 +881,7 @@ static void cdv_intel_crtc_dpms(struct drm_crtc *crtc, int mode) case DRM_MODE_DPMS_STANDBY: case DRM_MODE_DPMS_SUSPEND: if (psb_intel_crtc->active) - return; + break; psb_intel_crtc->active = true; @@ -835,17 +925,15 @@ static void cdv_intel_crtc_dpms(struct drm_crtc *crtc, int mode) REG_WRITE(map->status, temp); REG_READ(map->status); - cdv_intel_update_watermark(dev, crtc); cdv_intel_crtc_load_lut(crtc); /* Give the overlay scaler a chance to enable * if it's on this pipe */ /* psb_intel_crtc_dpms_video(crtc, true); TODO */ - psb_intel_crtc->crtc_enable = true; break; case DRM_MODE_DPMS_OFF: if (!psb_intel_crtc->active) - return; + break; psb_intel_crtc->active = false; @@ -892,10 +980,9 @@ static void cdv_intel_crtc_dpms(struct drm_crtc *crtc, int mode) /* Wait for the clocks to turn off. */ udelay(150); - cdv_intel_update_watermark(dev, crtc); - psb_intel_crtc->crtc_enable = false; break; } + cdv_intel_update_watermark(dev, crtc); /*Set FIFO Watermarks*/ REG_WRITE(DSPARB, 0x3F3E); } @@ -952,9 +1039,12 @@ static int cdv_intel_crtc_mode_set(struct drm_crtc *crtc, u32 dpll = 0, dspcntr, pipeconf; bool ok; bool is_crt = false, is_lvds = false, is_tv = false; - bool is_hdmi = false; + bool is_hdmi = false, is_dp = false; struct drm_mode_config *mode_config = &dev->mode_config; struct drm_connector *connector; + const struct cdv_intel_limit_t *limit; + u32 ddi_select = 0; + bool is_edp = false; list_for_each_entry(connector, &mode_config->connector_list, head) { struct psb_intel_encoder *psb_intel_encoder = @@ -964,6 +1054,7 @@ static int cdv_intel_crtc_mode_set(struct drm_crtc *crtc, || connector->encoder->crtc != crtc) continue; + ddi_select = psb_intel_encoder->ddi_select; switch (psb_intel_encoder->type) { case INTEL_OUTPUT_LVDS: is_lvds = true; @@ -977,6 +1068,15 @@ static int cdv_intel_crtc_mode_set(struct drm_crtc *crtc, case INTEL_OUTPUT_HDMI: is_hdmi = true; break; + case INTEL_OUTPUT_DISPLAYPORT: + is_dp = true; + break; + case INTEL_OUTPUT_EDP: + is_edp = true; + break; + default: + DRM_ERROR("invalid output type.\n"); + return 0; } } @@ -986,6 +1086,20 @@ static int cdv_intel_crtc_mode_set(struct drm_crtc *crtc, else /* high-end sku, 27/100 mhz */ refclk = 27000; + if (is_dp || is_edp) { + /* + * Based on the spec the low-end SKU has only CRT/LVDS. So it is + * unnecessary to consider it for DP/eDP. + * On the high-end SKU, it will use the 27/100M reference clk + * for DP/eDP. When using SSC clock, the ref clk is 100MHz.Otherwise + * it will be 27MHz. From the VBIOS code it seems that the pipe A choose + * 27MHz for DP/eDP while the Pipe B chooses the 100MHz. + */ + if (pipe == 0) + refclk = 27000; + else + refclk = 100000; + } if (is_lvds && dev_priv->lvds_use_ssc) { refclk = dev_priv->lvds_ssc_freq * 1000; @@ -993,8 +1107,10 @@ static int cdv_intel_crtc_mode_set(struct drm_crtc *crtc, } drm_mode_debug_printmodeline(adjusted_mode); + + limit = cdv_intel_limit(crtc, refclk); - ok = cdv_intel_find_best_PLL(crtc, adjusted_mode->clock, refclk, + ok = limit->find_pll(limit, crtc, adjusted_mode->clock, refclk, &clock); if (!ok) { dev_err(dev->dev, "Couldn't find PLL settings for mode!\n"); @@ -1009,6 +1125,15 @@ static int cdv_intel_crtc_mode_set(struct drm_crtc *crtc, } /* dpll |= PLL_REF_INPUT_DREFCLK; */ + if (is_dp || is_edp) { + cdv_intel_dp_set_m_n(crtc, mode, adjusted_mode); + } else { + REG_WRITE(PIPE_GMCH_DATA_M(pipe), 0); + REG_WRITE(PIPE_GMCH_DATA_N(pipe), 0); + REG_WRITE(PIPE_DP_LINK_M(pipe), 0); + REG_WRITE(PIPE_DP_LINK_N(pipe), 0); + } + dpll |= DPLL_SYNCLOCK_ENABLE; /* if (is_lvds) dpll |= DPLLB_MODE_LVDS; @@ -1019,6 +1144,31 @@ static int cdv_intel_crtc_mode_set(struct drm_crtc *crtc, /* setup pipeconf */ pipeconf = REG_READ(map->conf); + pipeconf &= ~(PIPE_BPC_MASK); + if (is_edp) { + switch (dev_priv->edp.bpp) { + case 24: + pipeconf |= PIPE_8BPC; + break; + case 18: + pipeconf |= PIPE_6BPC; + break; + case 30: + pipeconf |= PIPE_10BPC; + break; + default: + pipeconf |= PIPE_8BPC; + break; + } + } else if (is_lvds) { + /* the BPC will be 6 if it is 18-bit LVDS panel */ + if ((REG_READ(LVDS) & LVDS_A3_POWER_MASK) == LVDS_A3_POWER_UP) + pipeconf |= PIPE_8BPC; + else + pipeconf |= PIPE_6BPC; + } else + pipeconf |= PIPE_8BPC; + /* Set up the display plane register */ dspcntr = DISPPLANE_GAMMA_ENABLE; @@ -1033,7 +1183,7 @@ static int cdv_intel_crtc_mode_set(struct drm_crtc *crtc, REG_WRITE(map->dpll, dpll | DPLL_VGA_MODE_DIS | DPLL_SYNCLOCK_ENABLE); REG_READ(map->dpll); - cdv_dpll_set_clock_cdv(dev, crtc, &clock, is_lvds); + cdv_dpll_set_clock_cdv(dev, crtc, &clock, is_lvds, ddi_select); udelay(150); diff --git a/drivers/gpu/drm/gma500/cdv_intel_dp.c b/drivers/gpu/drm/gma500/cdv_intel_dp.c new file mode 100644 index 000000000000..c9abc06ef680 --- /dev/null +++ b/drivers/gpu/drm/gma500/cdv_intel_dp.c @@ -0,0 +1,1951 @@ +/* + * Copyright © 2012 Intel Corporation + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice (including the next + * paragraph) shall be included in all copies or substantial portions of the + * Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING + * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS + * IN THE SOFTWARE. + * + * Authors: + * Keith Packard <keithp@keithp.com> + * + */ + +#include <linux/i2c.h> +#include <linux/slab.h> +#include "drmP.h" +#include "drm.h" +#include "drm_crtc.h" +#include "drm_crtc_helper.h" +#include "psb_drv.h" +#include "psb_intel_drv.h" +#include "psb_intel_reg.h" +#include "drm_dp_helper.h" + +#define _wait_for(COND, MS, W) ({ \ + unsigned long timeout__ = jiffies + msecs_to_jiffies(MS); \ + int ret__ = 0; \ + while (! (COND)) { \ + if (time_after(jiffies, timeout__)) { \ + ret__ = -ETIMEDOUT; \ + break; \ + } \ + if (W && !in_dbg_master()) msleep(W); \ + } \ + ret__; \ +}) + +#define wait_for(COND, MS) _wait_for(COND, MS, 1) + +#define DP_LINK_STATUS_SIZE 6 +#define DP_LINK_CHECK_TIMEOUT (10 * 1000) + +#define DP_LINK_CONFIGURATION_SIZE 9 + +#define CDV_FAST_LINK_TRAIN 1 + +struct cdv_intel_dp { + uint32_t output_reg; + uint32_t DP; + uint8_t link_configuration[DP_LINK_CONFIGURATION_SIZE]; + bool has_audio; + int force_audio; + uint32_t color_range; + uint8_t link_bw; + uint8_t lane_count; + uint8_t dpcd[4]; + struct psb_intel_encoder *encoder; + struct i2c_adapter adapter; + struct i2c_algo_dp_aux_data algo; + uint8_t train_set[4]; + uint8_t link_status[DP_LINK_STATUS_SIZE]; + int panel_power_up_delay; + int panel_power_down_delay; + int panel_power_cycle_delay; + int backlight_on_delay; + int backlight_off_delay; + struct drm_display_mode *panel_fixed_mode; /* for eDP */ + bool panel_on; +}; + +struct ddi_regoff { + uint32_t PreEmph1; + uint32_t PreEmph2; + uint32_t VSwing1; + uint32_t VSwing2; + uint32_t VSwing3; + uint32_t VSwing4; + uint32_t VSwing5; +}; + +static struct ddi_regoff ddi_DP_train_table[] = { + {.PreEmph1 = 0x812c, .PreEmph2 = 0x8124, .VSwing1 = 0x8154, + .VSwing2 = 0x8148, .VSwing3 = 0x814C, .VSwing4 = 0x8150, + .VSwing5 = 0x8158,}, + {.PreEmph1 = 0x822c, .PreEmph2 = 0x8224, .VSwing1 = 0x8254, + .VSwing2 = 0x8248, .VSwing3 = 0x824C, .VSwing4 = 0x8250, + .VSwing5 = 0x8258,}, +}; + +static uint32_t dp_vswing_premph_table[] = { + 0x55338954, 0x4000, + 0x554d8954, 0x2000, + 0x55668954, 0, + 0x559ac0d4, 0x6000, +}; +/** + * is_edp - is the given port attached to an eDP panel (either CPU or PCH) + * @intel_dp: DP struct + * + * If a CPU or PCH DP output is attached to an eDP panel, this function + * will return true, and false otherwise. + */ +static bool is_edp(struct psb_intel_encoder *encoder) +{ + return encoder->type == INTEL_OUTPUT_EDP; +} + + +static void cdv_intel_dp_start_link_train(struct psb_intel_encoder *encoder); +static void cdv_intel_dp_complete_link_train(struct psb_intel_encoder *encoder); +static void cdv_intel_dp_link_down(struct psb_intel_encoder *encoder); + +static int +cdv_intel_dp_max_lane_count(struct psb_intel_encoder *encoder) +{ + struct cdv_intel_dp *intel_dp = encoder->dev_priv; + int max_lane_count = 4; + + if (intel_dp->dpcd[DP_DPCD_REV] >= 0x11) { + max_lane_count = intel_dp->dpcd[DP_MAX_LANE_COUNT] & 0x1f; + switch (max_lane_count) { + case 1: case 2: case 4: + break; + default: + max_lane_count = 4; + } + } + return max_lane_count; +} + +static int +cdv_intel_dp_max_link_bw(struct psb_intel_encoder *encoder) +{ + struct cdv_intel_dp *intel_dp = encoder->dev_priv; + int max_link_bw = intel_dp->dpcd[DP_MAX_LINK_RATE]; + + switch (max_link_bw) { + case DP_LINK_BW_1_62: + case DP_LINK_BW_2_7: + break; + default: + max_link_bw = DP_LINK_BW_1_62; + break; + } + return max_link_bw; +} + +static int +cdv_intel_dp_link_clock(uint8_t link_bw) +{ + if (link_bw == DP_LINK_BW_2_7) + return 270000; + else + return 162000; +} + +static int +cdv_intel_dp_link_required(int pixel_clock, int bpp) +{ + return (pixel_clock * bpp + 7) / 8; +} + +static int +cdv_intel_dp_max_data_rate(int max_link_clock, int max_lanes) +{ + return (max_link_clock * max_lanes * 19) / 20; +} + +static void cdv_intel_edp_panel_vdd_on(struct psb_intel_encoder *intel_encoder) +{ + struct drm_device *dev = intel_encoder->base.dev; + struct cdv_intel_dp *intel_dp = intel_encoder->dev_priv; + u32 pp; + + if (intel_dp->panel_on) { + DRM_DEBUG_KMS("Skip VDD on because of panel on\n"); + return; + } + DRM_DEBUG_KMS("\n"); + + pp = REG_READ(PP_CONTROL); + + pp |= EDP_FORCE_VDD; + REG_WRITE(PP_CONTROL, pp); + REG_READ(PP_CONTROL); + msleep(intel_dp->panel_power_up_delay); +} + +static void cdv_intel_edp_panel_vdd_off(struct psb_intel_encoder *intel_encoder) +{ + struct drm_device *dev = intel_encoder->base.dev; + u32 pp; + + DRM_DEBUG_KMS("\n"); + pp = REG_READ(PP_CONTROL); + + pp &= ~EDP_FORCE_VDD; + REG_WRITE(PP_CONTROL, pp); + REG_READ(PP_CONTROL); + +} + +/* Returns true if the panel was already on when called */ +static bool cdv_intel_edp_panel_on(struct psb_intel_encoder *intel_encoder) +{ + struct drm_device *dev = intel_encoder->base.dev; + struct cdv_intel_dp *intel_dp = intel_encoder->dev_priv; + u32 pp, idle_on_mask = PP_ON | PP_SEQUENCE_NONE; + + if (intel_dp->panel_on) + return true; + + DRM_DEBUG_KMS("\n"); + pp = REG_READ(PP_CONTROL); + pp &= ~PANEL_UNLOCK_MASK; + + pp |= (PANEL_UNLOCK_REGS | POWER_TARGET_ON); + REG_WRITE(PP_CONTROL, pp); + REG_READ(PP_CONTROL); + + if (wait_for(((REG_READ(PP_STATUS) & idle_on_mask) == idle_on_mask), 1000)) { + DRM_DEBUG_KMS("Error in Powering up eDP panel, status %x\n", REG_READ(PP_STATUS)); + intel_dp->panel_on = false; + } else + intel_dp->panel_on = true; + msleep(intel_dp->panel_power_up_delay); + + return false; +} + +static void cdv_intel_edp_panel_off (struct psb_intel_encoder *intel_encoder) +{ + struct drm_device *dev = intel_encoder->base.dev; + u32 pp, idle_off_mask = PP_ON ; + struct cdv_intel_dp *intel_dp = intel_encoder->dev_priv; + + DRM_DEBUG_KMS("\n"); + + pp = REG_READ(PP_CONTROL); + + if ((pp & POWER_TARGET_ON) == 0) + return; + + intel_dp->panel_on = false; + pp &= ~PANEL_UNLOCK_MASK; + /* ILK workaround: disable reset around power sequence */ + + pp &= ~POWER_TARGET_ON; + pp &= ~EDP_FORCE_VDD; + pp &= ~EDP_BLC_ENABLE; + REG_WRITE(PP_CONTROL, pp); + REG_READ(PP_CONTROL); + DRM_DEBUG_KMS("PP_STATUS %x\n", REG_READ(PP_STATUS)); + + if (wait_for((REG_READ(PP_STATUS) & idle_off_mask) == 0, 1000)) { + DRM_DEBUG_KMS("Error in turning off Panel\n"); + } + + msleep(intel_dp->panel_power_cycle_delay); + DRM_DEBUG_KMS("Over\n"); +} + +static void cdv_intel_edp_backlight_on (struct psb_intel_encoder *intel_encoder) +{ + struct drm_device *dev = intel_encoder->base.dev; + u32 pp; + + DRM_DEBUG_KMS("\n"); + /* + * If we enable the backlight right away following a panel power + * on, we may see slight flicker as the panel syncs with the eDP + * link. So delay a bit to make sure the image is solid before + * allowing it to appear. + */ + msleep(300); + pp = REG_READ(PP_CONTROL); + + pp |= EDP_BLC_ENABLE; + REG_WRITE(PP_CONTROL, pp); + gma_backlight_enable(dev); +} + +static void cdv_intel_edp_backlight_off (struct psb_intel_encoder *intel_encoder) +{ + struct drm_device *dev = intel_encoder->base.dev; + struct cdv_intel_dp *intel_dp = intel_encoder->dev_priv; + u32 pp; + + DRM_DEBUG_KMS("\n"); + gma_backlight_disable(dev); + msleep(10); + pp = REG_READ(PP_CONTROL); + + pp &= ~EDP_BLC_ENABLE; + REG_WRITE(PP_CONTROL, pp); + msleep(intel_dp->backlight_off_delay); +} + +static int +cdv_intel_dp_mode_valid(struct drm_connector *connector, + struct drm_display_mode *mode) +{ + struct psb_intel_encoder *encoder = psb_intel_attached_encoder(connector); + struct cdv_intel_dp *intel_dp = encoder->dev_priv; + int max_link_clock = cdv_intel_dp_link_clock(cdv_intel_dp_max_link_bw(encoder)); + int max_lanes = cdv_intel_dp_max_lane_count(encoder); + struct drm_psb_private *dev_priv = connector->dev->dev_private; + + if (is_edp(encoder) && intel_dp->panel_fixed_mode) { + if (mode->hdisplay > intel_dp->panel_fixed_mode->hdisplay) + return MODE_PANEL; + if (mode->vdisplay > intel_dp->panel_fixed_mode->vdisplay) + return MODE_PANEL; + } + + /* only refuse the mode on non eDP since we have seen some weird eDP panels + which are outside spec tolerances but somehow work by magic */ + if (!is_edp(encoder) && + (cdv_intel_dp_link_required(mode->clock, dev_priv->edp.bpp) + > cdv_intel_dp_max_data_rate(max_link_clock, max_lanes))) + return MODE_CLOCK_HIGH; + + if (is_edp(encoder)) { + if (cdv_intel_dp_link_required(mode->clock, 24) + > cdv_intel_dp_max_data_rate(max_link_clock, max_lanes)) + return MODE_CLOCK_HIGH; + + } + if (mode->clock < 10000) + return MODE_CLOCK_LOW; + + return MODE_OK; +} + +static uint32_t +pack_aux(uint8_t *src, int src_bytes) +{ + int i; + uint32_t v = 0; + + if (src_bytes > 4) + src_bytes = 4; + for (i = 0; i < src_bytes; i++) + v |= ((uint32_t) src[i]) << ((3-i) * 8); + return v; +} + +static void +unpack_aux(uint32_t src, uint8_t *dst, int dst_bytes) +{ + int i; + if (dst_bytes > 4) + dst_bytes = 4; + for (i = 0; i < dst_bytes; i++) + dst[i] = src >> ((3-i) * 8); +} + +static int +cdv_intel_dp_aux_ch(struct psb_intel_encoder *encoder, + uint8_t *send, int send_bytes, + uint8_t *recv, int recv_size) +{ + struct cdv_intel_dp *intel_dp = encoder->dev_priv; + uint32_t output_reg = intel_dp->output_reg; + struct drm_device *dev = encoder->base.dev; + uint32_t ch_ctl = output_reg + 0x10; + uint32_t ch_data = ch_ctl + 4; + int i; + int recv_bytes; + uint32_t status; + uint32_t aux_clock_divider; + int try, precharge; + + /* The clock divider is based off the hrawclk, + * and would like to run at 2MHz. So, take the + * hrawclk value and divide by 2 and use that + * On CDV platform it uses 200MHz as hrawclk. + * + */ + aux_clock_divider = 200 / 2; + + precharge = 4; + if (is_edp(encoder)) + precharge = 10; + + if (REG_READ(ch_ctl) & DP_AUX_CH_CTL_SEND_BUSY) { + DRM_ERROR("dp_aux_ch not started status 0x%08x\n", + REG_READ(ch_ctl)); + return -EBUSY; + } + + /* Must try at least 3 times according to DP spec */ + for (try = 0; try < 5; try++) { + /* Load the send data into the aux channel data registers */ + for (i = 0; i < send_bytes; i += 4) + REG_WRITE(ch_data + i, + pack_aux(send + i, send_bytes - i)); + + /* Send the command and wait for it to complete */ + REG_WRITE(ch_ctl, + DP_AUX_CH_CTL_SEND_BUSY | + DP_AUX_CH_CTL_TIME_OUT_400us | + (send_bytes << DP_AUX_CH_CTL_MESSAGE_SIZE_SHIFT) | + (precharge << DP_AUX_CH_CTL_PRECHARGE_2US_SHIFT) | + (aux_clock_divider << DP_AUX_CH_CTL_BIT_CLOCK_2X_SHIFT) | + DP_AUX_CH_CTL_DONE | + DP_AUX_CH_CTL_TIME_OUT_ERROR | + DP_AUX_CH_CTL_RECEIVE_ERROR); + for (;;) { + status = REG_READ(ch_ctl); + if ((status & DP_AUX_CH_CTL_SEND_BUSY) == 0) + break; + udelay(100); + } + + /* Clear done status and any errors */ + REG_WRITE(ch_ctl, + status | + DP_AUX_CH_CTL_DONE | + DP_AUX_CH_CTL_TIME_OUT_ERROR | + DP_AUX_CH_CTL_RECEIVE_ERROR); + if (status & DP_AUX_CH_CTL_DONE) + break; + } + + if ((status & DP_AUX_CH_CTL_DONE) == 0) { + DRM_ERROR("dp_aux_ch not done status 0x%08x\n", status); + return -EBUSY; + } + + /* Check for timeout or receive error. + * Timeouts occur when the sink is not connected + */ + if (status & DP_AUX_CH_CTL_RECEIVE_ERROR) { + DRM_ERROR("dp_aux_ch receive error status 0x%08x\n", status); + return -EIO; + } + + /* Timeouts occur when the device isn't connected, so they're + * "normal" -- don't fill the kernel log with these */ + if (status & DP_AUX_CH_CTL_TIME_OUT_ERROR) { + DRM_DEBUG_KMS("dp_aux_ch timeout status 0x%08x\n", status); + return -ETIMEDOUT; + } + + /* Unload any bytes sent back from the other side */ + recv_bytes = ((status & DP_AUX_CH_CTL_MESSAGE_SIZE_MASK) >> + DP_AUX_CH_CTL_MESSAGE_SIZE_SHIFT); + if (recv_bytes > recv_size) + recv_bytes = recv_size; + + for (i = 0; i < recv_bytes; i += 4) + unpack_aux(REG_READ(ch_data + i), + recv + i, recv_bytes - i); + + return recv_bytes; +} + +/* Write data to the aux channel in native mode */ +static int +cdv_intel_dp_aux_native_write(struct psb_intel_encoder *encoder, + uint16_t address, uint8_t *send, int send_bytes) +{ + int ret; + uint8_t msg[20]; + int msg_bytes; + uint8_t ack; + + if (send_bytes > 16) + return -1; + msg[0] = AUX_NATIVE_WRITE << 4; + msg[1] = address >> 8; + msg[2] = address & 0xff; + msg[3] = send_bytes - 1; + memcpy(&msg[4], send, send_bytes); + msg_bytes = send_bytes + 4; + for (;;) { + ret = cdv_intel_dp_aux_ch(encoder, msg, msg_bytes, &ack, 1); + if (ret < 0) + return ret; + if ((ack & AUX_NATIVE_REPLY_MASK) == AUX_NATIVE_REPLY_ACK) + break; + else if ((ack & AUX_NATIVE_REPLY_MASK) == AUX_NATIVE_REPLY_DEFER) + udelay(100); + else + return -EIO; + } + return send_bytes; +} + +/* Write a single byte to the aux channel in native mode */ +static int +cdv_intel_dp_aux_native_write_1(struct psb_intel_encoder *encoder, + uint16_t address, uint8_t byte) +{ + return cdv_intel_dp_aux_native_write(encoder, address, &byte, 1); +} + +/* read bytes from a native aux channel */ +static int +cdv_intel_dp_aux_native_read(struct psb_intel_encoder *encoder, + uint16_t address, uint8_t *recv, int recv_bytes) +{ + uint8_t msg[4]; + int msg_bytes; + uint8_t reply[20]; + int reply_bytes; + uint8_t ack; + int ret; + + msg[0] = AUX_NATIVE_READ << 4; + msg[1] = address >> 8; + msg[2] = address & 0xff; + msg[3] = recv_bytes - 1; + + msg_bytes = 4; + reply_bytes = recv_bytes + 1; + + for (;;) { + ret = cdv_intel_dp_aux_ch(encoder, msg, msg_bytes, + reply, reply_bytes); + if (ret == 0) + return -EPROTO; + if (ret < 0) + return ret; + ack = reply[0]; + if ((ack & AUX_NATIVE_REPLY_MASK) == AUX_NATIVE_REPLY_ACK) { + memcpy(recv, reply + 1, ret - 1); + return ret - 1; + } + else if ((ack & AUX_NATIVE_REPLY_MASK) == AUX_NATIVE_REPLY_DEFER) + udelay(100); + else + return -EIO; + } +} + +static int +cdv_intel_dp_i2c_aux_ch(struct i2c_adapter *adapter, int mode, + uint8_t write_byte, uint8_t *read_byte) +{ + struct i2c_algo_dp_aux_data *algo_data = adapter->algo_data; + struct cdv_intel_dp *intel_dp = container_of(adapter, + struct cdv_intel_dp, + adapter); + struct psb_intel_encoder *encoder = intel_dp->encoder; + uint16_t address = algo_data->address; + uint8_t msg[5]; + uint8_t reply[2]; + unsigned retry; + int msg_bytes; + int reply_bytes; + int ret; + + /* Set up the command byte */ + if (mode & MODE_I2C_READ) + msg[0] = AUX_I2C_READ << 4; + else + msg[0] = AUX_I2C_WRITE << 4; + + if (!(mode & MODE_I2C_STOP)) + msg[0] |= AUX_I2C_MOT << 4; + + msg[1] = address >> 8; + msg[2] = address; + + switch (mode) { + case MODE_I2C_WRITE: + msg[3] = 0; + msg[4] = write_byte; + msg_bytes = 5; + reply_bytes = 1; + break; + case MODE_I2C_READ: + msg[3] = 0; + msg_bytes = 4; + reply_bytes = 2; + break; + default: + msg_bytes = 3; + reply_bytes = 1; + break; + } + + for (retry = 0; retry < 5; retry++) { + ret = cdv_intel_dp_aux_ch(encoder, + msg, msg_bytes, + reply, reply_bytes); + if (ret < 0) { + DRM_DEBUG_KMS("aux_ch failed %d\n", ret); + return ret; + } + + switch (reply[0] & AUX_NATIVE_REPLY_MASK) { + case AUX_NATIVE_REPLY_ACK: + /* I2C-over-AUX Reply field is only valid + * when paired with AUX ACK. + */ + break; + case AUX_NATIVE_REPLY_NACK: + DRM_DEBUG_KMS("aux_ch native nack\n"); + return -EREMOTEIO; + case AUX_NATIVE_REPLY_DEFER: + udelay(100); + continue; + default: + DRM_ERROR("aux_ch invalid native reply 0x%02x\n", + reply[0]); + return -EREMOTEIO; + } + + switch (reply[0] & AUX_I2C_REPLY_MASK) { + case AUX_I2C_REPLY_ACK: + if (mode == MODE_I2C_READ) { + *read_byte = reply[1]; + } + return reply_bytes - 1; + case AUX_I2C_REPLY_NACK: + DRM_DEBUG_KMS("aux_i2c nack\n"); + return -EREMOTEIO; + case AUX_I2C_REPLY_DEFER: + DRM_DEBUG_KMS("aux_i2c defer\n"); + udelay(100); + break; + default: + DRM_ERROR("aux_i2c invalid reply 0x%02x\n", reply[0]); + return -EREMOTEIO; + } + } + + DRM_ERROR("too many retries, giving up\n"); + return -EREMOTEIO; +} + +static int +cdv_intel_dp_i2c_init(struct psb_intel_connector *connector, struct psb_intel_encoder *encoder, const char *name) +{ + struct cdv_intel_dp *intel_dp = encoder->dev_priv; + int ret; + + DRM_DEBUG_KMS("i2c_init %s\n", name); + + intel_dp->algo.running = false; + intel_dp->algo.address = 0; + intel_dp->algo.aux_ch = cdv_intel_dp_i2c_aux_ch; + + memset(&intel_dp->adapter, '\0', sizeof (intel_dp->adapter)); + intel_dp->adapter.owner = THIS_MODULE; + intel_dp->adapter.class = I2C_CLASS_DDC; + strncpy (intel_dp->adapter.name, name, sizeof(intel_dp->adapter.name) - 1); + intel_dp->adapter.name[sizeof(intel_dp->adapter.name) - 1] = '\0'; + intel_dp->adapter.algo_data = &intel_dp->algo; + intel_dp->adapter.dev.parent = &connector->base.kdev; + + if (is_edp(encoder)) + cdv_intel_edp_panel_vdd_on(encoder); + ret = i2c_dp_aux_add_bus(&intel_dp->adapter); + if (is_edp(encoder)) + cdv_intel_edp_panel_vdd_off(encoder); + + return ret; +} + +void cdv_intel_fixed_panel_mode(struct drm_display_mode *fixed_mode, + struct drm_display_mode *adjusted_mode) +{ + adjusted_mode->hdisplay = fixed_mode->hdisplay; + adjusted_mode->hsync_start = fixed_mode->hsync_start; + adjusted_mode->hsync_end = fixed_mode->hsync_end; + adjusted_mode->htotal = fixed_mode->htotal; + + adjusted_mode->vdisplay = fixed_mode->vdisplay; + adjusted_mode->vsync_start = fixed_mode->vsync_start; + adjusted_mode->vsync_end = fixed_mode->vsync_end; + adjusted_mode->vtotal = fixed_mode->vtotal; + + adjusted_mode->clock = fixed_mode->clock; + + drm_mode_set_crtcinfo(adjusted_mode, CRTC_INTERLACE_HALVE_V); +} + +static bool +cdv_intel_dp_mode_fixup(struct drm_encoder *encoder, const struct drm_display_mode *mode, + struct drm_display_mode *adjusted_mode) +{ + struct drm_psb_private *dev_priv = encoder->dev->dev_private; + struct psb_intel_encoder *intel_encoder = to_psb_intel_encoder(encoder); + struct cdv_intel_dp *intel_dp = intel_encoder->dev_priv; + int lane_count, clock; + int max_lane_count = cdv_intel_dp_max_lane_count(intel_encoder); + int max_clock = cdv_intel_dp_max_link_bw(intel_encoder) == DP_LINK_BW_2_7 ? 1 : 0; + static int bws[2] = { DP_LINK_BW_1_62, DP_LINK_BW_2_7 }; + int refclock = mode->clock; + int bpp = 24; + + if (is_edp(intel_encoder) && intel_dp->panel_fixed_mode) { + cdv_intel_fixed_panel_mode(intel_dp->panel_fixed_mode, adjusted_mode); + refclock = intel_dp->panel_fixed_mode->clock; + bpp = dev_priv->edp.bpp; + } + + for (lane_count = 1; lane_count <= max_lane_count; lane_count <<= 1) { + for (clock = max_clock; clock >= 0; clock--) { + int link_avail = cdv_intel_dp_max_data_rate(cdv_intel_dp_link_clock(bws[clock]), lane_count); + + if (cdv_intel_dp_link_required(refclock, bpp) <= link_avail) { + intel_dp->link_bw = bws[clock]; + intel_dp->lane_count = lane_count; + adjusted_mode->clock = cdv_intel_dp_link_clock(intel_dp->link_bw); + DRM_DEBUG_KMS("Display port link bw %02x lane " + "count %d clock %d\n", + intel_dp->link_bw, intel_dp->lane_count, + adjusted_mode->clock); + return true; + } + } + } + if (is_edp(intel_encoder)) { + /* okay we failed just pick the highest */ + intel_dp->lane_count = max_lane_count; + intel_dp->link_bw = bws[max_clock]; + adjusted_mode->clock = cdv_intel_dp_link_clock(intel_dp->link_bw); + DRM_DEBUG_KMS("Force picking display port link bw %02x lane " + "count %d clock %d\n", + intel_dp->link_bw, intel_dp->lane_count, + adjusted_mode->clock); + + return true; + } + return false; +} + +struct cdv_intel_dp_m_n { + uint32_t tu; + uint32_t gmch_m; + uint32_t gmch_n; + uint32_t link_m; + uint32_t link_n; +}; + +static void +cdv_intel_reduce_ratio(uint32_t *num, uint32_t *den) +{ + /* + while (*num > 0xffffff || *den > 0xffffff) { + *num >>= 1; + *den >>= 1; + }*/ + uint64_t value, m; + m = *num; + value = m * (0x800000); + m = do_div(value, *den); + *num = value; + *den = 0x800000; +} + +static void +cdv_intel_dp_compute_m_n(int bpp, + int nlanes, + int pixel_clock, + int link_clock, + struct cdv_intel_dp_m_n *m_n) +{ + m_n->tu = 64; + m_n->gmch_m = (pixel_clock * bpp + 7) >> 3; + m_n->gmch_n = link_clock * nlanes; + cdv_intel_reduce_ratio(&m_n->gmch_m, &m_n->gmch_n); + m_n->link_m = pixel_clock; + m_n->link_n = link_clock; + cdv_intel_reduce_ratio(&m_n->link_m, &m_n->link_n); +} + +void +cdv_intel_dp_set_m_n(struct drm_crtc *crtc, struct drm_display_mode *mode, + struct drm_display_mode *adjusted_mode) +{ + struct drm_device *dev = crtc->dev; + struct drm_psb_private *dev_priv = dev->dev_private; + struct drm_mode_config *mode_config = &dev->mode_config; + struct drm_encoder *encoder; + struct psb_intel_crtc *intel_crtc = to_psb_intel_crtc(crtc); + int lane_count = 4, bpp = 24; + struct cdv_intel_dp_m_n m_n; + int pipe = intel_crtc->pipe; + + /* + * Find the lane count in the intel_encoder private + */ + list_for_each_entry(encoder, &mode_config->encoder_list, head) { + struct psb_intel_encoder *intel_encoder; + struct cdv_intel_dp *intel_dp; + + if (encoder->crtc != crtc) + continue; + + intel_encoder = to_psb_intel_encoder(encoder); + intel_dp = intel_encoder->dev_priv; + if (intel_encoder->type == INTEL_OUTPUT_DISPLAYPORT) { + lane_count = intel_dp->lane_count; + break; + } else if (is_edp(intel_encoder)) { + lane_count = intel_dp->lane_count; + bpp = dev_priv->edp.bpp; + break; + } + } + + /* + * Compute the GMCH and Link ratios. The '3' here is + * the number of bytes_per_pixel post-LUT, which we always + * set up for 8-bits of R/G/B, or 3 bytes total. + */ + cdv_intel_dp_compute_m_n(bpp, lane_count, + mode->clock, adjusted_mode->clock, &m_n); + + { + REG_WRITE(PIPE_GMCH_DATA_M(pipe), + ((m_n.tu - 1) << PIPE_GMCH_DATA_M_TU_SIZE_SHIFT) | + m_n.gmch_m); + REG_WRITE(PIPE_GMCH_DATA_N(pipe), m_n.gmch_n); + REG_WRITE(PIPE_DP_LINK_M(pipe), m_n.link_m); + REG_WRITE(PIPE_DP_LINK_N(pipe), m_n.link_n); + } +} + +static void +cdv_intel_dp_mode_set(struct drm_encoder *encoder, struct drm_display_mode *mode, + struct drm_display_mode *adjusted_mode) +{ + struct psb_intel_encoder *intel_encoder = to_psb_intel_encoder(encoder); + struct drm_crtc *crtc = encoder->crtc; + struct psb_intel_crtc *intel_crtc = to_psb_intel_crtc(crtc); + struct cdv_intel_dp *intel_dp = intel_encoder->dev_priv; + struct drm_device *dev = encoder->dev; + + intel_dp->DP = DP_VOLTAGE_0_4 | DP_PRE_EMPHASIS_0; + intel_dp->DP |= intel_dp->color_range; + + if (adjusted_mode->flags & DRM_MODE_FLAG_PHSYNC) + intel_dp->DP |= DP_SYNC_HS_HIGH; + if (adjusted_mode->flags & DRM_MODE_FLAG_PVSYNC) + intel_dp->DP |= DP_SYNC_VS_HIGH; + + intel_dp->DP |= DP_LINK_TRAIN_OFF; + + switch (intel_dp->lane_count) { + case 1: + intel_dp->DP |= DP_PORT_WIDTH_1; + break; + case 2: + intel_dp->DP |= DP_PORT_WIDTH_2; + break; + case 4: + intel_dp->DP |= DP_PORT_WIDTH_4; + break; + } + if (intel_dp->has_audio) + intel_dp->DP |= DP_AUDIO_OUTPUT_ENABLE; + + memset(intel_dp->link_configuration, 0, DP_LINK_CONFIGURATION_SIZE); + intel_dp->link_configuration[0] = intel_dp->link_bw; + intel_dp->link_configuration[1] = intel_dp->lane_count; + + /* + * Check for DPCD version > 1.1 and enhanced framing support + */ + if (intel_dp->dpcd[DP_DPCD_REV] >= 0x11 && + (intel_dp->dpcd[DP_MAX_LANE_COUNT] & DP_ENHANCED_FRAME_CAP)) { + intel_dp->link_configuration[1] |= DP_LANE_COUNT_ENHANCED_FRAME_EN; + intel_dp->DP |= DP_ENHANCED_FRAMING; + } + + /* CPT DP's pipe select is decided in TRANS_DP_CTL */ + if (intel_crtc->pipe == 1) + intel_dp->DP |= DP_PIPEB_SELECT; + + REG_WRITE(intel_dp->output_reg, (intel_dp->DP | DP_PORT_EN)); + DRM_DEBUG_KMS("DP expected reg is %x\n", intel_dp->DP); + if (is_edp(intel_encoder)) { + uint32_t pfit_control; + cdv_intel_edp_panel_on(intel_encoder); + + if (mode->hdisplay != adjusted_mode->hdisplay || + mode->vdisplay != adjusted_mode->vdisplay) + pfit_control = PFIT_ENABLE; + else + pfit_control = 0; + + pfit_control |= intel_crtc->pipe << PFIT_PIPE_SHIFT; + + REG_WRITE(PFIT_CONTROL, pfit_control); + } +} + + +/* If the sink supports it, try to set the power state appropriately */ +static void cdv_intel_dp_sink_dpms(struct psb_intel_encoder *encoder, int mode) +{ + struct cdv_intel_dp *intel_dp = encoder->dev_priv; + int ret, i; + + /* Should have a valid DPCD by this point */ + if (intel_dp->dpcd[DP_DPCD_REV] < 0x11) + return; + + if (mode != DRM_MODE_DPMS_ON) { + ret = cdv_intel_dp_aux_native_write_1(encoder, DP_SET_POWER, + DP_SET_POWER_D3); + if (ret != 1) + DRM_DEBUG_DRIVER("failed to write sink power state\n"); + } else { + /* + * When turning on, we need to retry for 1ms to give the sink + * time to wake up. + */ + for (i = 0; i < 3; i++) { + ret = cdv_intel_dp_aux_native_write_1(encoder, + DP_SET_POWER, + DP_SET_POWER_D0); + if (ret == 1) + break; + udelay(1000); + } + } +} + +static void cdv_intel_dp_prepare(struct drm_encoder *encoder) +{ + struct psb_intel_encoder *intel_encoder = to_psb_intel_encoder(encoder); + int edp = is_edp(intel_encoder); + + if (edp) { + cdv_intel_edp_backlight_off(intel_encoder); + cdv_intel_edp_panel_off(intel_encoder); + cdv_intel_edp_panel_vdd_on(intel_encoder); + } + /* Wake up the sink first */ + cdv_intel_dp_sink_dpms(intel_encoder, DRM_MODE_DPMS_ON); + cdv_intel_dp_link_down(intel_encoder); + if (edp) + cdv_intel_edp_panel_vdd_off(intel_encoder); +} + +static void cdv_intel_dp_commit(struct drm_encoder *encoder) +{ + struct psb_intel_encoder *intel_encoder = to_psb_intel_encoder(encoder); + int edp = is_edp(intel_encoder); + + if (edp) + cdv_intel_edp_panel_on(intel_encoder); + cdv_intel_dp_start_link_train(intel_encoder); + cdv_intel_dp_complete_link_train(intel_encoder); + if (edp) + cdv_intel_edp_backlight_on(intel_encoder); +} + +static void +cdv_intel_dp_dpms(struct drm_encoder *encoder, int mode) +{ + struct psb_intel_encoder *intel_encoder = to_psb_intel_encoder(encoder); + struct cdv_intel_dp *intel_dp = intel_encoder->dev_priv; + struct drm_device *dev = encoder->dev; + uint32_t dp_reg = REG_READ(intel_dp->output_reg); + int edp = is_edp(intel_encoder); + + if (mode != DRM_MODE_DPMS_ON) { + if (edp) { + cdv_intel_edp_backlight_off(intel_encoder); + cdv_intel_edp_panel_vdd_on(intel_encoder); + } + cdv_intel_dp_sink_dpms(intel_encoder, mode); + cdv_intel_dp_link_down(intel_encoder); + if (edp) { + cdv_intel_edp_panel_vdd_off(intel_encoder); + cdv_intel_edp_panel_off(intel_encoder); + } + } else { + if (edp) + cdv_intel_edp_panel_on(intel_encoder); + cdv_intel_dp_sink_dpms(intel_encoder, mode); + if (!(dp_reg & DP_PORT_EN)) { + cdv_intel_dp_start_link_train(intel_encoder); + cdv_intel_dp_complete_link_train(intel_encoder); + } + if (edp) + cdv_intel_edp_backlight_on(intel_encoder); + } +} + +/* + * Native read with retry for link status and receiver capability reads for + * cases where the sink may still be asleep. + */ +static bool +cdv_intel_dp_aux_native_read_retry(struct psb_intel_encoder *encoder, uint16_t address, + uint8_t *recv, int recv_bytes) +{ + int ret, i; + + /* + * Sinks are *supposed* to come up within 1ms from an off state, + * but we're also supposed to retry 3 times per the spec. + */ + for (i = 0; i < 3; i++) { + ret = cdv_intel_dp_aux_native_read(encoder, address, recv, + recv_bytes); + if (ret == recv_bytes) + return true; + udelay(1000); + } + + return false; +} + +/* + * Fetch AUX CH registers 0x202 - 0x207 which contain + * link status information + */ +static bool +cdv_intel_dp_get_link_status(struct psb_intel_encoder *encoder) +{ + struct cdv_intel_dp *intel_dp = encoder->dev_priv; + return cdv_intel_dp_aux_native_read_retry(encoder, + DP_LANE0_1_STATUS, + intel_dp->link_status, + DP_LINK_STATUS_SIZE); +} + +static uint8_t +cdv_intel_dp_link_status(uint8_t link_status[DP_LINK_STATUS_SIZE], + int r) +{ + return link_status[r - DP_LANE0_1_STATUS]; +} + +static uint8_t +cdv_intel_get_adjust_request_voltage(uint8_t link_status[DP_LINK_STATUS_SIZE], + int lane) +{ + int i = DP_ADJUST_REQUEST_LANE0_1 + (lane >> 1); + int s = ((lane & 1) ? + DP_ADJUST_VOLTAGE_SWING_LANE1_SHIFT : + DP_ADJUST_VOLTAGE_SWING_LANE0_SHIFT); + uint8_t l = cdv_intel_dp_link_status(link_status, i); + + return ((l >> s) & 3) << DP_TRAIN_VOLTAGE_SWING_SHIFT; +} + +static uint8_t +cdv_intel_get_adjust_request_pre_emphasis(uint8_t link_status[DP_LINK_STATUS_SIZE], + int lane) +{ + int i = DP_ADJUST_REQUEST_LANE0_1 + (lane >> 1); + int s = ((lane & 1) ? + DP_ADJUST_PRE_EMPHASIS_LANE1_SHIFT : + DP_ADJUST_PRE_EMPHASIS_LANE0_SHIFT); + uint8_t l = cdv_intel_dp_link_status(link_status, i); + + return ((l >> s) & 3) << DP_TRAIN_PRE_EMPHASIS_SHIFT; +} + + +#if 0 +static char *voltage_names[] = { + "0.4V", "0.6V", "0.8V", "1.2V" +}; +static char *pre_emph_names[] = { + "0dB", "3.5dB", "6dB", "9.5dB" +}; +static char *link_train_names[] = { + "pattern 1", "pattern 2", "idle", "off" +}; +#endif + +#define CDV_DP_VOLTAGE_MAX DP_TRAIN_VOLTAGE_SWING_1200 +/* +static uint8_t +cdv_intel_dp_pre_emphasis_max(uint8_t voltage_swing) +{ + switch (voltage_swing & DP_TRAIN_VOLTAGE_SWING_MASK) { + case DP_TRAIN_VOLTAGE_SWING_400: + return DP_TRAIN_PRE_EMPHASIS_6; + case DP_TRAIN_VOLTAGE_SWING_600: + return DP_TRAIN_PRE_EMPHASIS_6; + case DP_TRAIN_VOLTAGE_SWING_800: + return DP_TRAIN_PRE_EMPHASIS_3_5; + case DP_TRAIN_VOLTAGE_SWING_1200: + default: + return DP_TRAIN_PRE_EMPHASIS_0; + } +} +*/ +static void +cdv_intel_get_adjust_train(struct psb_intel_encoder *encoder) +{ + struct cdv_intel_dp *intel_dp = encoder->dev_priv; + uint8_t v = 0; + uint8_t p = 0; + int lane; + + for (lane = 0; lane < intel_dp->lane_count; lane++) { + uint8_t this_v = cdv_intel_get_adjust_request_voltage(intel_dp->link_status, lane); + uint8_t this_p = cdv_intel_get_adjust_request_pre_emphasis(intel_dp->link_status, lane); + + if (this_v > v) + v = this_v; + if (this_p > p) + p = this_p; + } + + if (v >= CDV_DP_VOLTAGE_MAX) + v = CDV_DP_VOLTAGE_MAX | DP_TRAIN_MAX_SWING_REACHED; + + if (p == DP_TRAIN_PRE_EMPHASIS_MASK) + p |= DP_TRAIN_MAX_PRE_EMPHASIS_REACHED; + + for (lane = 0; lane < 4; lane++) + intel_dp->train_set[lane] = v | p; +} + + +static uint8_t +cdv_intel_get_lane_status(uint8_t link_status[DP_LINK_STATUS_SIZE], + int lane) +{ + int i = DP_LANE0_1_STATUS + (lane >> 1); + int s = (lane & 1) * 4; + uint8_t l = cdv_intel_dp_link_status(link_status, i); + + return (l >> s) & 0xf; +} + +/* Check for clock recovery is done on all channels */ +static bool +cdv_intel_clock_recovery_ok(uint8_t link_status[DP_LINK_STATUS_SIZE], int lane_count) +{ + int lane; + uint8_t lane_status; + + for (lane = 0; lane < lane_count; lane++) { + lane_status = cdv_intel_get_lane_status(link_status, lane); + if ((lane_status & DP_LANE_CR_DONE) == 0) + return false; + } + return true; +} + +/* Check to see if channel eq is done on all channels */ +#define CHANNEL_EQ_BITS (DP_LANE_CR_DONE|\ + DP_LANE_CHANNEL_EQ_DONE|\ + DP_LANE_SYMBOL_LOCKED) +static bool +cdv_intel_channel_eq_ok(struct psb_intel_encoder *encoder) +{ + struct cdv_intel_dp *intel_dp = encoder->dev_priv; + uint8_t lane_align; + uint8_t lane_status; + int lane; + + lane_align = cdv_intel_dp_link_status(intel_dp->link_status, + DP_LANE_ALIGN_STATUS_UPDATED); + if ((lane_align & DP_INTERLANE_ALIGN_DONE) == 0) + return false; + for (lane = 0; lane < intel_dp->lane_count; lane++) { + lane_status = cdv_intel_get_lane_status(intel_dp->link_status, lane); + if ((lane_status & CHANNEL_EQ_BITS) != CHANNEL_EQ_BITS) + return false; + } + return true; +} + +static bool +cdv_intel_dp_set_link_train(struct psb_intel_encoder *encoder, + uint32_t dp_reg_value, + uint8_t dp_train_pat) +{ + + struct drm_device *dev = encoder->base.dev; + int ret; + struct cdv_intel_dp *intel_dp = encoder->dev_priv; + + REG_WRITE(intel_dp->output_reg, dp_reg_value); + REG_READ(intel_dp->output_reg); + + ret = cdv_intel_dp_aux_native_write_1(encoder, + DP_TRAINING_PATTERN_SET, + dp_train_pat); + + if (ret != 1) { + DRM_DEBUG_KMS("Failure in setting link pattern %x\n", + dp_train_pat); + return false; + } + + return true; +} + + +static bool +cdv_intel_dplink_set_level(struct psb_intel_encoder *encoder, + uint8_t dp_train_pat) +{ + + int ret; + struct cdv_intel_dp *intel_dp = encoder->dev_priv; + + ret = cdv_intel_dp_aux_native_write(encoder, + DP_TRAINING_LANE0_SET, + intel_dp->train_set, + intel_dp->lane_count); + + if (ret != intel_dp->lane_count) { + DRM_DEBUG_KMS("Failure in setting level %d, lane_cnt= %d\n", + intel_dp->train_set[0], intel_dp->lane_count); + return false; + } + return true; +} + +static void +cdv_intel_dp_set_vswing_premph(struct psb_intel_encoder *encoder, uint8_t signal_level) +{ + struct drm_device *dev = encoder->base.dev; + struct cdv_intel_dp *intel_dp = encoder->dev_priv; + struct ddi_regoff *ddi_reg; + int vswing, premph, index; + + if (intel_dp->output_reg == DP_B) + ddi_reg = &ddi_DP_train_table[0]; + else + ddi_reg = &ddi_DP_train_table[1]; + + vswing = (signal_level & DP_TRAIN_VOLTAGE_SWING_MASK); + premph = ((signal_level & DP_TRAIN_PRE_EMPHASIS_MASK)) >> + DP_TRAIN_PRE_EMPHASIS_SHIFT; + + if (vswing + premph > 3) + return; +#ifdef CDV_FAST_LINK_TRAIN + return; +#endif + DRM_DEBUG_KMS("Test2\n"); + //return ; + cdv_sb_reset(dev); + /* ;Swing voltage programming + ;gfx_dpio_set_reg(0xc058, 0x0505313A) */ + cdv_sb_write(dev, ddi_reg->VSwing5, 0x0505313A); + + /* ;gfx_dpio_set_reg(0x8154, 0x43406055) */ + cdv_sb_write(dev, ddi_reg->VSwing1, 0x43406055); + + /* ;gfx_dpio_set_reg(0x8148, 0x55338954) + * The VSwing_PreEmph table is also considered based on the vswing/premp + */ + index = (vswing + premph) * 2; + if (premph == 1 && vswing == 1) { + cdv_sb_write(dev, ddi_reg->VSwing2, 0x055738954); + } else + cdv_sb_write(dev, ddi_reg->VSwing2, dp_vswing_premph_table[index]); + + /* ;gfx_dpio_set_reg(0x814c, 0x40802040) */ + if ((vswing + premph) == DP_TRAIN_VOLTAGE_SWING_1200) + cdv_sb_write(dev, ddi_reg->VSwing3, 0x70802040); + else + cdv_sb_write(dev, ddi_reg->VSwing3, 0x40802040); + + /* ;gfx_dpio_set_reg(0x8150, 0x2b405555) */ + /* cdv_sb_write(dev, ddi_reg->VSwing4, 0x2b405555); */ + + /* ;gfx_dpio_set_reg(0x8154, 0xc3406055) */ + cdv_sb_write(dev, ddi_reg->VSwing1, 0xc3406055); + + /* ;Pre emphasis programming + * ;gfx_dpio_set_reg(0xc02c, 0x1f030040) + */ + cdv_sb_write(dev, ddi_reg->PreEmph1, 0x1f030040); + + /* ;gfx_dpio_set_reg(0x8124, 0x00004000) */ + index = 2 * premph + 1; + cdv_sb_write(dev, ddi_reg->PreEmph2, dp_vswing_premph_table[index]); + return; +} + + +/* Enable corresponding port and start training pattern 1 */ +static void +cdv_intel_dp_start_link_train(struct psb_intel_encoder *encoder) +{ + struct drm_device *dev = encoder->base.dev; + struct cdv_intel_dp *intel_dp = encoder->dev_priv; + int i; + uint8_t voltage; + bool clock_recovery = false; + int tries; + u32 reg; + uint32_t DP = intel_dp->DP; + + DP |= DP_PORT_EN; + DP &= ~DP_LINK_TRAIN_MASK; + + reg = DP; + reg |= DP_LINK_TRAIN_PAT_1; + /* Enable output, wait for it to become active */ + REG_WRITE(intel_dp->output_reg, reg); + REG_READ(intel_dp->output_reg); + psb_intel_wait_for_vblank(dev); + + DRM_DEBUG_KMS("Link config\n"); + /* Write the link configuration data */ + cdv_intel_dp_aux_native_write(encoder, DP_LINK_BW_SET, + intel_dp->link_configuration, + 2); + + memset(intel_dp->train_set, 0, 4); + voltage = 0; + tries = 0; + clock_recovery = false; + + DRM_DEBUG_KMS("Start train\n"); + reg = DP | DP_LINK_TRAIN_PAT_1; + + + for (;;) { + /* Use intel_dp->train_set[0] to set the voltage and pre emphasis values */ + DRM_DEBUG_KMS("DP Link Train Set %x, Link_config %x, %x\n", + intel_dp->train_set[0], + intel_dp->link_configuration[0], + intel_dp->link_configuration[1]); + + if (!cdv_intel_dp_set_link_train(encoder, reg, DP_TRAINING_PATTERN_1)) { + DRM_DEBUG_KMS("Failure in aux-transfer setting pattern 1\n"); + } + cdv_intel_dp_set_vswing_premph(encoder, intel_dp->train_set[0]); + /* Set training pattern 1 */ + + cdv_intel_dplink_set_level(encoder, DP_TRAINING_PATTERN_1); + + udelay(200); + if (!cdv_intel_dp_get_link_status(encoder)) + break; + + DRM_DEBUG_KMS("DP Link status %x, %x, %x, %x, %x, %x\n", + intel_dp->link_status[0], intel_dp->link_status[1], intel_dp->link_status[2], + intel_dp->link_status[3], intel_dp->link_status[4], intel_dp->link_status[5]); + + if (cdv_intel_clock_recovery_ok(intel_dp->link_status, intel_dp->lane_count)) { + DRM_DEBUG_KMS("PT1 train is done\n"); + clock_recovery = true; + break; + } + + /* Check to see if we've tried the max voltage */ + for (i = 0; i < intel_dp->lane_count; i++) + if ((intel_dp->train_set[i] & DP_TRAIN_MAX_SWING_REACHED) == 0) + break; + if (i == intel_dp->lane_count) + break; + + /* Check to see if we've tried the same voltage 5 times */ + if ((intel_dp->train_set[0] & DP_TRAIN_VOLTAGE_SWING_MASK) == voltage) { + ++tries; + if (tries == 5) + break; + } else + tries = 0; + voltage = intel_dp->train_set[0] & DP_TRAIN_VOLTAGE_SWING_MASK; + + /* Compute new intel_dp->train_set as requested by target */ + cdv_intel_get_adjust_train(encoder); + + } + + if (!clock_recovery) { + DRM_DEBUG_KMS("failure in DP patter 1 training, train set %x\n", intel_dp->train_set[0]); + } + + intel_dp->DP = DP; +} + +static void +cdv_intel_dp_complete_link_train(struct psb_intel_encoder *encoder) +{ + struct drm_device *dev = encoder->base.dev; + struct cdv_intel_dp *intel_dp = encoder->dev_priv; + bool channel_eq = false; + int tries, cr_tries; + u32 reg; + uint32_t DP = intel_dp->DP; + + /* channel equalization */ + tries = 0; + cr_tries = 0; + channel_eq = false; + + DRM_DEBUG_KMS("\n"); + reg = DP | DP_LINK_TRAIN_PAT_2; + + for (;;) { + + DRM_DEBUG_KMS("DP Link Train Set %x, Link_config %x, %x\n", + intel_dp->train_set[0], + intel_dp->link_configuration[0], + intel_dp->link_configuration[1]); + /* channel eq pattern */ + + if (!cdv_intel_dp_set_link_train(encoder, reg, + DP_TRAINING_PATTERN_2)) { + DRM_DEBUG_KMS("Failure in aux-transfer setting pattern 2\n"); + } + /* Use intel_dp->train_set[0] to set the voltage and pre emphasis values */ + + if (cr_tries > 5) { + DRM_ERROR("failed to train DP, aborting\n"); + cdv_intel_dp_link_down(encoder); + break; + } + + cdv_intel_dp_set_vswing_premph(encoder, intel_dp->train_set[0]); + + cdv_intel_dplink_set_level(encoder, DP_TRAINING_PATTERN_2); + + udelay(1000); + if (!cdv_intel_dp_get_link_status(encoder)) + break; + + DRM_DEBUG_KMS("DP Link status %x, %x, %x, %x, %x, %x\n", + intel_dp->link_status[0], intel_dp->link_status[1], intel_dp->link_status[2], + intel_dp->link_status[3], intel_dp->link_status[4], intel_dp->link_status[5]); + + /* Make sure clock is still ok */ + if (!cdv_intel_clock_recovery_ok(intel_dp->link_status, intel_dp->lane_count)) { + cdv_intel_dp_start_link_train(encoder); + cr_tries++; + continue; + } + + if (cdv_intel_channel_eq_ok(encoder)) { + DRM_DEBUG_KMS("PT2 train is done\n"); + channel_eq = true; + break; + } + + /* Try 5 times, then try clock recovery if that fails */ + if (tries > 5) { + cdv_intel_dp_link_down(encoder); + cdv_intel_dp_start_link_train(encoder); + tries = 0; + cr_tries++; + continue; + } + + /* Compute new intel_dp->train_set as requested by target */ + cdv_intel_get_adjust_train(encoder); + ++tries; + + } + + reg = DP | DP_LINK_TRAIN_OFF; + + REG_WRITE(intel_dp->output_reg, reg); + REG_READ(intel_dp->output_reg); + cdv_intel_dp_aux_native_write_1(encoder, + DP_TRAINING_PATTERN_SET, DP_TRAINING_PATTERN_DISABLE); +} + +static void +cdv_intel_dp_link_down(struct psb_intel_encoder *encoder) +{ + struct drm_device *dev = encoder->base.dev; + struct cdv_intel_dp *intel_dp = encoder->dev_priv; + uint32_t DP = intel_dp->DP; + + if ((REG_READ(intel_dp->output_reg) & DP_PORT_EN) == 0) + return; + + DRM_DEBUG_KMS("\n"); + + + { + DP &= ~DP_LINK_TRAIN_MASK; + REG_WRITE(intel_dp->output_reg, DP | DP_LINK_TRAIN_PAT_IDLE); + } + REG_READ(intel_dp->output_reg); + + msleep(17); + + REG_WRITE(intel_dp->output_reg, DP & ~DP_PORT_EN); + REG_READ(intel_dp->output_reg); +} + +static enum drm_connector_status +cdv_dp_detect(struct psb_intel_encoder *encoder) +{ + struct cdv_intel_dp *intel_dp = encoder->dev_priv; + enum drm_connector_status status; + + status = connector_status_disconnected; + if (cdv_intel_dp_aux_native_read(encoder, 0x000, intel_dp->dpcd, + sizeof (intel_dp->dpcd)) == sizeof (intel_dp->dpcd)) + { + if (intel_dp->dpcd[DP_DPCD_REV] != 0) + status = connector_status_connected; + } + if (status == connector_status_connected) + DRM_DEBUG_KMS("DPCD: Rev=%x LN_Rate=%x LN_CNT=%x LN_DOWNSP=%x\n", + intel_dp->dpcd[0], intel_dp->dpcd[1], + intel_dp->dpcd[2], intel_dp->dpcd[3]); + return status; +} + +/** + * Uses CRT_HOTPLUG_EN and CRT_HOTPLUG_STAT to detect DP connection. + * + * \return true if DP port is connected. + * \return false if DP port is disconnected. + */ +static enum drm_connector_status +cdv_intel_dp_detect(struct drm_connector *connector, bool force) +{ + struct psb_intel_encoder *encoder = psb_intel_attached_encoder(connector); + struct cdv_intel_dp *intel_dp = encoder->dev_priv; + enum drm_connector_status status; + struct edid *edid = NULL; + int edp = is_edp(encoder); + + intel_dp->has_audio = false; + + if (edp) + cdv_intel_edp_panel_vdd_on(encoder); + status = cdv_dp_detect(encoder); + if (status != connector_status_connected) { + if (edp) + cdv_intel_edp_panel_vdd_off(encoder); + return status; + } + + if (intel_dp->force_audio) { + intel_dp->has_audio = intel_dp->force_audio > 0; + } else { + edid = drm_get_edid(connector, &intel_dp->adapter); + if (edid) { + intel_dp->has_audio = drm_detect_monitor_audio(edid); + kfree(edid); + } + } + if (edp) + cdv_intel_edp_panel_vdd_off(encoder); + + return connector_status_connected; +} + +static int cdv_intel_dp_get_modes(struct drm_connector *connector) +{ + struct psb_intel_encoder *intel_encoder = psb_intel_attached_encoder(connector); + struct cdv_intel_dp *intel_dp = intel_encoder->dev_priv; + struct edid *edid = NULL; + int ret = 0; + int edp = is_edp(intel_encoder); + + + edid = drm_get_edid(connector, &intel_dp->adapter); + if (edid) { + drm_mode_connector_update_edid_property(connector, edid); + ret = drm_add_edid_modes(connector, edid); + kfree(edid); + } + + if (is_edp(intel_encoder)) { + struct drm_device *dev = connector->dev; + struct drm_psb_private *dev_priv = dev->dev_private; + + cdv_intel_edp_panel_vdd_off(intel_encoder); + if (ret) { + if (edp && !intel_dp->panel_fixed_mode) { + struct drm_display_mode *newmode; + list_for_each_entry(newmode, &connector->probed_modes, + head) { + if (newmode->type & DRM_MODE_TYPE_PREFERRED) { + intel_dp->panel_fixed_mode = + drm_mode_duplicate(dev, newmode); + break; + } + } + } + + return ret; + } + if (!intel_dp->panel_fixed_mode && dev_priv->lfp_lvds_vbt_mode) { + intel_dp->panel_fixed_mode = + drm_mode_duplicate(dev, dev_priv->lfp_lvds_vbt_mode); + if (intel_dp->panel_fixed_mode) { + intel_dp->panel_fixed_mode->type |= + DRM_MODE_TYPE_PREFERRED; + } + } + if (intel_dp->panel_fixed_mode != NULL) { + struct drm_display_mode *mode; + mode = drm_mode_duplicate(dev, intel_dp->panel_fixed_mode); + drm_mode_probed_add(connector, mode); + return 1; + } + } + + return ret; +} + +static bool +cdv_intel_dp_detect_audio(struct drm_connector *connector) +{ + struct psb_intel_encoder *encoder = psb_intel_attached_encoder(connector); + struct cdv_intel_dp *intel_dp = encoder->dev_priv; + struct edid *edid; + bool has_audio = false; + int edp = is_edp(encoder); + + if (edp) + cdv_intel_edp_panel_vdd_on(encoder); + + edid = drm_get_edid(connector, &intel_dp->adapter); + if (edid) { + has_audio = drm_detect_monitor_audio(edid); + kfree(edid); + } + if (edp) + cdv_intel_edp_panel_vdd_off(encoder); + + return has_audio; +} + +static int +cdv_intel_dp_set_property(struct drm_connector *connector, + struct drm_property *property, + uint64_t val) +{ + struct drm_psb_private *dev_priv = connector->dev->dev_private; + struct psb_intel_encoder *encoder = psb_intel_attached_encoder(connector); + struct cdv_intel_dp *intel_dp = encoder->dev_priv; + int ret; + + ret = drm_connector_property_set_value(connector, property, val); + if (ret) + return ret; + + if (property == dev_priv->force_audio_property) { + int i = val; + bool has_audio; + + if (i == intel_dp->force_audio) + return 0; + + intel_dp->force_audio = i; + + if (i == 0) + has_audio = cdv_intel_dp_detect_audio(connector); + else + has_audio = i > 0; + + if (has_audio == intel_dp->has_audio) + return 0; + + intel_dp->has_audio = has_audio; + goto done; + } + + if (property == dev_priv->broadcast_rgb_property) { + if (val == !!intel_dp->color_range) + return 0; + + intel_dp->color_range = val ? DP_COLOR_RANGE_16_235 : 0; + goto done; + } + + return -EINVAL; + +done: + if (encoder->base.crtc) { + struct drm_crtc *crtc = encoder->base.crtc; + drm_crtc_helper_set_mode(crtc, &crtc->mode, + crtc->x, crtc->y, + crtc->fb); + } + + return 0; +} + +static void +cdv_intel_dp_destroy(struct drm_connector *connector) +{ + struct psb_intel_encoder *psb_intel_encoder = + psb_intel_attached_encoder(connector); + struct cdv_intel_dp *intel_dp = psb_intel_encoder->dev_priv; + + if (is_edp(psb_intel_encoder)) { + /* cdv_intel_panel_destroy_backlight(connector->dev); */ + if (intel_dp->panel_fixed_mode) { + kfree(intel_dp->panel_fixed_mode); + intel_dp->panel_fixed_mode = NULL; + } + } + i2c_del_adapter(&intel_dp->adapter); + drm_sysfs_connector_remove(connector); + drm_connector_cleanup(connector); + kfree(connector); +} + +static void cdv_intel_dp_encoder_destroy(struct drm_encoder *encoder) +{ + drm_encoder_cleanup(encoder); +} + +static const struct drm_encoder_helper_funcs cdv_intel_dp_helper_funcs = { + .dpms = cdv_intel_dp_dpms, + .mode_fixup = cdv_intel_dp_mode_fixup, + .prepare = cdv_intel_dp_prepare, + .mode_set = cdv_intel_dp_mode_set, + .commit = cdv_intel_dp_commit, +}; + +static const struct drm_connector_funcs cdv_intel_dp_connector_funcs = { + .dpms = drm_helper_connector_dpms, + .detect = cdv_intel_dp_detect, + .fill_modes = drm_helper_probe_single_connector_modes, + .set_property = cdv_intel_dp_set_property, + .destroy = cdv_intel_dp_destroy, +}; + +static const struct drm_connector_helper_funcs cdv_intel_dp_connector_helper_funcs = { + .get_modes = cdv_intel_dp_get_modes, + .mode_valid = cdv_intel_dp_mode_valid, + .best_encoder = psb_intel_best_encoder, +}; + +static const struct drm_encoder_funcs cdv_intel_dp_enc_funcs = { + .destroy = cdv_intel_dp_encoder_destroy, +}; + + +static void cdv_intel_dp_add_properties(struct drm_connector *connector) +{ + cdv_intel_attach_force_audio_property(connector); + cdv_intel_attach_broadcast_rgb_property(connector); +} + +/* check the VBT to see whether the eDP is on DP-D port */ +static bool cdv_intel_dpc_is_edp(struct drm_device *dev) +{ + struct drm_psb_private *dev_priv = dev->dev_private; + struct child_device_config *p_child; + int i; + + if (!dev_priv->child_dev_num) + return false; + + for (i = 0; i < dev_priv->child_dev_num; i++) { + p_child = dev_priv->child_dev + i; + + if (p_child->dvo_port == PORT_IDPC && + p_child->device_type == DEVICE_TYPE_eDP) + return true; + } + return false; +} + +/* Cedarview display clock gating + + We need this disable dot get correct behaviour while enabling + DP/eDP. TODO - investigate if we can turn it back to normality + after enabling */ +static void cdv_disable_intel_clock_gating(struct drm_device *dev) +{ + u32 reg_value; + reg_value = REG_READ(DSPCLK_GATE_D); + + reg_value |= (DPUNIT_PIPEB_GATE_DISABLE | + DPUNIT_PIPEA_GATE_DISABLE | + DPCUNIT_CLOCK_GATE_DISABLE | + DPLSUNIT_CLOCK_GATE_DISABLE | + DPOUNIT_CLOCK_GATE_DISABLE | + DPIOUNIT_CLOCK_GATE_DISABLE); + + REG_WRITE(DSPCLK_GATE_D, reg_value); + + udelay(500); +} + +void +cdv_intel_dp_init(struct drm_device *dev, struct psb_intel_mode_device *mode_dev, int output_reg) +{ + struct psb_intel_encoder *psb_intel_encoder; + struct psb_intel_connector *psb_intel_connector; + struct drm_connector *connector; + struct drm_encoder *encoder; + struct cdv_intel_dp *intel_dp; + const char *name = NULL; + int type = DRM_MODE_CONNECTOR_DisplayPort; + + psb_intel_encoder = kzalloc(sizeof(struct psb_intel_encoder), GFP_KERNEL); + if (!psb_intel_encoder) + return; + psb_intel_connector = kzalloc(sizeof(struct psb_intel_connector), GFP_KERNEL); + if (!psb_intel_connector) + goto err_connector; + intel_dp = kzalloc(sizeof(struct cdv_intel_dp), GFP_KERNEL); + if (!intel_dp) + goto err_priv; + + if ((output_reg == DP_C) && cdv_intel_dpc_is_edp(dev)) + type = DRM_MODE_CONNECTOR_eDP; + + connector = &psb_intel_connector->base; + encoder = &psb_intel_encoder->base; + + drm_connector_init(dev, connector, &cdv_intel_dp_connector_funcs, type); + drm_encoder_init(dev, encoder, &cdv_intel_dp_enc_funcs, DRM_MODE_ENCODER_TMDS); + + psb_intel_connector_attach_encoder(psb_intel_connector, psb_intel_encoder); + + if (type == DRM_MODE_CONNECTOR_DisplayPort) + psb_intel_encoder->type = INTEL_OUTPUT_DISPLAYPORT; + else + psb_intel_encoder->type = INTEL_OUTPUT_EDP; + + + psb_intel_encoder->dev_priv=intel_dp; + intel_dp->encoder = psb_intel_encoder; + intel_dp->output_reg = output_reg; + + drm_encoder_helper_add(encoder, &cdv_intel_dp_helper_funcs); + drm_connector_helper_add(connector, &cdv_intel_dp_connector_helper_funcs); + + connector->polled = DRM_CONNECTOR_POLL_HPD; + connector->interlace_allowed = false; + connector->doublescan_allowed = false; + + drm_sysfs_connector_add(connector); + + /* Set up the DDC bus. */ + switch (output_reg) { + case DP_B: + name = "DPDDC-B"; + psb_intel_encoder->ddi_select = (DP_MASK | DDI0_SELECT); + break; + case DP_C: + name = "DPDDC-C"; + psb_intel_encoder->ddi_select = (DP_MASK | DDI1_SELECT); + break; + } + + cdv_disable_intel_clock_gating(dev); + + cdv_intel_dp_i2c_init(psb_intel_connector, psb_intel_encoder, name); + /* FIXME:fail check */ + cdv_intel_dp_add_properties(connector); + + if (is_edp(psb_intel_encoder)) { + int ret; + struct edp_power_seq cur; + u32 pp_on, pp_off, pp_div; + u32 pwm_ctrl; + + pp_on = REG_READ(PP_CONTROL); + pp_on &= ~PANEL_UNLOCK_MASK; + pp_on |= PANEL_UNLOCK_REGS; + + REG_WRITE(PP_CONTROL, pp_on); + + pwm_ctrl = REG_READ(BLC_PWM_CTL2); + pwm_ctrl |= PWM_PIPE_B; + REG_WRITE(BLC_PWM_CTL2, pwm_ctrl); + + pp_on = REG_READ(PP_ON_DELAYS); + pp_off = REG_READ(PP_OFF_DELAYS); + pp_div = REG_READ(PP_DIVISOR); + + /* Pull timing values out of registers */ + cur.t1_t3 = (pp_on & PANEL_POWER_UP_DELAY_MASK) >> + PANEL_POWER_UP_DELAY_SHIFT; + + cur.t8 = (pp_on & PANEL_LIGHT_ON_DELAY_MASK) >> + PANEL_LIGHT_ON_DELAY_SHIFT; + + cur.t9 = (pp_off & PANEL_LIGHT_OFF_DELAY_MASK) >> + PANEL_LIGHT_OFF_DELAY_SHIFT; + + cur.t10 = (pp_off & PANEL_POWER_DOWN_DELAY_MASK) >> + PANEL_POWER_DOWN_DELAY_SHIFT; + + cur.t11_t12 = ((pp_div & PANEL_POWER_CYCLE_DELAY_MASK) >> + PANEL_POWER_CYCLE_DELAY_SHIFT); + + DRM_DEBUG_KMS("cur t1_t3 %d t8 %d t9 %d t10 %d t11_t12 %d\n", + cur.t1_t3, cur.t8, cur.t9, cur.t10, cur.t11_t12); + + + intel_dp->panel_power_up_delay = cur.t1_t3 / 10; + intel_dp->backlight_on_delay = cur.t8 / 10; + intel_dp->backlight_off_delay = cur.t9 / 10; + intel_dp->panel_power_down_delay = cur.t10 / 10; + intel_dp->panel_power_cycle_delay = (cur.t11_t12 - 1) * 100; + + DRM_DEBUG_KMS("panel power up delay %d, power down delay %d, power cycle delay %d\n", + intel_dp->panel_power_up_delay, intel_dp->panel_power_down_delay, + intel_dp->panel_power_cycle_delay); + + DRM_DEBUG_KMS("backlight on delay %d, off delay %d\n", + intel_dp->backlight_on_delay, intel_dp->backlight_off_delay); + + + cdv_intel_edp_panel_vdd_on(psb_intel_encoder); + ret = cdv_intel_dp_aux_native_read(psb_intel_encoder, DP_DPCD_REV, + intel_dp->dpcd, + sizeof(intel_dp->dpcd)); + cdv_intel_edp_panel_vdd_off(psb_intel_encoder); + if (ret == 0) { + /* if this fails, presume the device is a ghost */ + DRM_INFO("failed to retrieve link info, disabling eDP\n"); + cdv_intel_dp_encoder_destroy(encoder); + cdv_intel_dp_destroy(connector); + goto err_priv; + } else { + DRM_DEBUG_KMS("DPCD: Rev=%x LN_Rate=%x LN_CNT=%x LN_DOWNSP=%x\n", + intel_dp->dpcd[0], intel_dp->dpcd[1], + intel_dp->dpcd[2], intel_dp->dpcd[3]); + + } + /* The CDV reference driver moves pnale backlight setup into the displays that + have a backlight: this is a good idea and one we should probably adopt, however + we need to migrate all the drivers before we can do that */ + /*cdv_intel_panel_setup_backlight(dev); */ + } + return; + +err_priv: + kfree(psb_intel_connector); +err_connector: + kfree(psb_intel_encoder); +} diff --git a/drivers/gpu/drm/gma500/cdv_intel_hdmi.c b/drivers/gpu/drm/gma500/cdv_intel_hdmi.c index a86f87b9ddde..7272a461edfe 100644 --- a/drivers/gpu/drm/gma500/cdv_intel_hdmi.c +++ b/drivers/gpu/drm/gma500/cdv_intel_hdmi.c @@ -139,8 +139,6 @@ static enum drm_connector_status cdv_hdmi_detect( { struct psb_intel_encoder *psb_intel_encoder = psb_intel_attached_encoder(connector); - struct psb_intel_connector *psb_intel_connector = - to_psb_intel_connector(connector); struct mid_intel_hdmi_priv *hdmi_priv = psb_intel_encoder->dev_priv; struct edid *edid = NULL; enum drm_connector_status status = connector_status_disconnected; @@ -157,8 +155,6 @@ static enum drm_connector_status cdv_hdmi_detect( hdmi_priv->has_hdmi_audio = drm_detect_monitor_audio(edid); } - - psb_intel_connector->base.display_info.raw_edid = NULL; kfree(edid); } return status; @@ -352,9 +348,11 @@ void cdv_hdmi_init(struct drm_device *dev, switch (reg) { case SDVOB: ddc_bus = GPIOE; + psb_intel_encoder->ddi_select = DDI0_SELECT; break; case SDVOC: ddc_bus = GPIOD; + psb_intel_encoder->ddi_select = DDI1_SELECT; break; default: DRM_ERROR("unknown reg 0x%x for HDMI\n", reg); diff --git a/drivers/gpu/drm/gma500/cdv_intel_lvds.c b/drivers/gpu/drm/gma500/cdv_intel_lvds.c index c7f9468b74ba..b362dd39bf5a 100644 --- a/drivers/gpu/drm/gma500/cdv_intel_lvds.c +++ b/drivers/gpu/drm/gma500/cdv_intel_lvds.c @@ -506,16 +506,8 @@ static int cdv_intel_lvds_set_property(struct drm_connector *connector, property, value)) return -1; - else { -#ifdef CONFIG_BACKLIGHT_CLASS_DEVICE - struct drm_psb_private *dev_priv = - encoder->dev->dev_private; - struct backlight_device *bd = - dev_priv->backlight_device; - bd->props.brightness = value; - backlight_update_status(bd); -#endif - } + else + gma_backlight_set(encoder->dev, value); } else if (!strcmp(property->name, "DPMS") && encoder) { struct drm_encoder_helper_funcs *helpers = encoder->helper_private; diff --git a/drivers/gpu/drm/gma500/framebuffer.c b/drivers/gpu/drm/gma500/framebuffer.c index 5732b5702e1c..884ba73ac6ce 100644 --- a/drivers/gpu/drm/gma500/framebuffer.c +++ b/drivers/gpu/drm/gma500/framebuffer.c @@ -764,6 +764,13 @@ static void psb_setup_outputs(struct drm_device *dev) crtc_mask = dev_priv->ops->hdmi_mask; clone_mask = (1 << INTEL_OUTPUT_HDMI); break; + case INTEL_OUTPUT_DISPLAYPORT: + crtc_mask = (1 << 0) | (1 << 1); + clone_mask = (1 << INTEL_OUTPUT_DISPLAYPORT); + break; + case INTEL_OUTPUT_EDP: + crtc_mask = (1 << 1); + clone_mask = (1 << INTEL_OUTPUT_EDP); } encoder->possible_crtcs = crtc_mask; encoder->possible_clones = diff --git a/drivers/gpu/drm/gma500/gem.c b/drivers/gpu/drm/gma500/gem.c index fc7d144bc2d3..df20546a2a34 100644 --- a/drivers/gpu/drm/gma500/gem.c +++ b/drivers/gpu/drm/gma500/gem.c @@ -36,7 +36,12 @@ int psb_gem_init_object(struct drm_gem_object *obj) void psb_gem_free_object(struct drm_gem_object *obj) { struct gtt_range *gtt = container_of(obj, struct gtt_range, gem); - drm_gem_object_release_wrap(obj); + + /* Remove the list map if one is present */ + if (obj->map_list.map) + drm_gem_free_mmap_offset(obj); + drm_gem_object_release(obj); + /* This must occur last as it frees up the memory of the GEM object */ psb_gtt_free_range(obj->dev, gtt); } @@ -77,7 +82,7 @@ int psb_gem_dumb_map_gtt(struct drm_file *file, struct drm_device *dev, /* Make it mmapable */ if (!obj->map_list.map) { - ret = gem_create_mmap_offset(obj); + ret = drm_gem_create_mmap_offset(obj); if (ret) goto out; } diff --git a/drivers/gpu/drm/gma500/gem_glue.c b/drivers/gpu/drm/gma500/gem_glue.c deleted file mode 100644 index 3c17634f6061..000000000000 --- a/drivers/gpu/drm/gma500/gem_glue.c +++ /dev/null @@ -1,90 +0,0 @@ -/************************************************************************** - * Copyright (c) 2011, Intel Corporation. - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or modify it - * under the terms and conditions of the GNU General Public License, - * version 2, as published by the Free Software Foundation. - * - * This program is distributed in the hope it will be useful, but WITHOUT - * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or - * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for - * more details. - * - * You should have received a copy of the GNU General Public License along with - * this program; if not, write to the Free Software Foundation, Inc., - * 51 Franklin St - Fifth Floor, Boston, MA 02110-1301 USA. - * - **************************************************************************/ - -#include <drm/drmP.h> -#include <drm/drm.h> -#include "gem_glue.h" - -void drm_gem_object_release_wrap(struct drm_gem_object *obj) -{ - /* Remove the list map if one is present */ - if (obj->map_list.map) { - struct drm_gem_mm *mm = obj->dev->mm_private; - struct drm_map_list *list = &obj->map_list; - drm_ht_remove_item(&mm->offset_hash, &list->hash); - drm_mm_put_block(list->file_offset_node); - kfree(list->map); - list->map = NULL; - } - drm_gem_object_release(obj); -} - -/** - * gem_create_mmap_offset - invent an mmap offset - * @obj: our object - * - * Standard implementation of offset generation for mmap as is - * duplicated in several drivers. This belongs in GEM. - */ -int gem_create_mmap_offset(struct drm_gem_object *obj) -{ - struct drm_device *dev = obj->dev; - struct drm_gem_mm *mm = dev->mm_private; - struct drm_map_list *list; - struct drm_local_map *map; - int ret; - - list = &obj->map_list; - list->map = kzalloc(sizeof(struct drm_map_list), GFP_KERNEL); - if (list->map == NULL) - return -ENOMEM; - map = list->map; - map->type = _DRM_GEM; - map->size = obj->size; - map->handle = obj; - - list->file_offset_node = drm_mm_search_free(&mm->offset_manager, - obj->size / PAGE_SIZE, 0, 0); - if (!list->file_offset_node) { - dev_err(dev->dev, "failed to allocate offset for bo %d\n", - obj->name); - ret = -ENOSPC; - goto free_it; - } - list->file_offset_node = drm_mm_get_block(list->file_offset_node, - obj->size / PAGE_SIZE, 0); - if (!list->file_offset_node) { - ret = -ENOMEM; - goto free_it; - } - list->hash.key = list->file_offset_node->start; - ret = drm_ht_insert_item(&mm->offset_hash, &list->hash); - if (ret) { - dev_err(dev->dev, "failed to add to map hash\n"); - goto free_mm; - } - return 0; - -free_mm: - drm_mm_put_block(list->file_offset_node); -free_it: - kfree(list->map); - list->map = NULL; - return ret; -} diff --git a/drivers/gpu/drm/gma500/gem_glue.h b/drivers/gpu/drm/gma500/gem_glue.h deleted file mode 100644 index ce5ce30f74db..000000000000 --- a/drivers/gpu/drm/gma500/gem_glue.h +++ /dev/null @@ -1,2 +0,0 @@ -extern void drm_gem_object_release_wrap(struct drm_gem_object *obj); -extern int gem_create_mmap_offset(struct drm_gem_object *obj); diff --git a/drivers/gpu/drm/gma500/intel_bios.c b/drivers/gpu/drm/gma500/intel_bios.c index 8d7caf0f363e..4fb79cf00ed8 100644 --- a/drivers/gpu/drm/gma500/intel_bios.c +++ b/drivers/gpu/drm/gma500/intel_bios.c @@ -54,6 +54,98 @@ static void *find_section(struct bdb_header *bdb, int section_id) return NULL; } +static void +parse_edp(struct drm_psb_private *dev_priv, struct bdb_header *bdb) +{ + struct bdb_edp *edp; + struct edp_power_seq *edp_pps; + struct edp_link_params *edp_link_params; + uint8_t panel_type; + + edp = find_section(bdb, BDB_EDP); + + dev_priv->edp.bpp = 18; + if (!edp) { + if (dev_priv->edp.support) { + DRM_DEBUG_KMS("No eDP BDB found but eDP panel supported, assume %dbpp panel color depth.\n", + dev_priv->edp.bpp); + } + return; + } + + panel_type = dev_priv->panel_type; + switch ((edp->color_depth >> (panel_type * 2)) & 3) { + case EDP_18BPP: + dev_priv->edp.bpp = 18; + break; + case EDP_24BPP: + dev_priv->edp.bpp = 24; + break; + case EDP_30BPP: + dev_priv->edp.bpp = 30; + break; + } + + /* Get the eDP sequencing and link info */ + edp_pps = &edp->power_seqs[panel_type]; + edp_link_params = &edp->link_params[panel_type]; + + dev_priv->edp.pps = *edp_pps; + + DRM_DEBUG_KMS("EDP timing in vbt t1_t3 %d t8 %d t9 %d t10 %d t11_t12 %d\n", + dev_priv->edp.pps.t1_t3, dev_priv->edp.pps.t8, + dev_priv->edp.pps.t9, dev_priv->edp.pps.t10, + dev_priv->edp.pps.t11_t12); + + dev_priv->edp.rate = edp_link_params->rate ? DP_LINK_BW_2_7 : + DP_LINK_BW_1_62; + switch (edp_link_params->lanes) { + case 0: + dev_priv->edp.lanes = 1; + break; + case 1: + dev_priv->edp.lanes = 2; + break; + case 3: + default: + dev_priv->edp.lanes = 4; + break; + } + DRM_DEBUG_KMS("VBT reports EDP: Lane_count %d, Lane_rate %d, Bpp %d\n", + dev_priv->edp.lanes, dev_priv->edp.rate, dev_priv->edp.bpp); + + switch (edp_link_params->preemphasis) { + case 0: + dev_priv->edp.preemphasis = DP_TRAIN_PRE_EMPHASIS_0; + break; + case 1: + dev_priv->edp.preemphasis = DP_TRAIN_PRE_EMPHASIS_3_5; + break; + case 2: + dev_priv->edp.preemphasis = DP_TRAIN_PRE_EMPHASIS_6; + break; + case 3: + dev_priv->edp.preemphasis = DP_TRAIN_PRE_EMPHASIS_9_5; + break; + } + switch (edp_link_params->vswing) { + case 0: + dev_priv->edp.vswing = DP_TRAIN_VOLTAGE_SWING_400; + break; + case 1: + dev_priv->edp.vswing = DP_TRAIN_VOLTAGE_SWING_600; + break; + case 2: + dev_priv->edp.vswing = DP_TRAIN_VOLTAGE_SWING_800; + break; + case 3: + dev_priv->edp.vswing = DP_TRAIN_VOLTAGE_SWING_1200; + break; + } + DRM_DEBUG_KMS("VBT reports EDP: VSwing %d, Preemph %d\n", + dev_priv->edp.vswing, dev_priv->edp.preemphasis); +} + static u16 get_blocksize(void *p) { @@ -154,6 +246,8 @@ static void parse_lfp_panel_data(struct drm_psb_private *dev_priv, return; dev_priv->lvds_dither = lvds_options->pixel_dither; + dev_priv->panel_type = lvds_options->panel_type; + if (lvds_options->panel_type == 0xff) return; @@ -340,6 +434,9 @@ parse_driver_features(struct drm_psb_private *dev_priv, if (!driver) return; + if (driver->lvds_config == BDB_DRIVER_FEATURE_EDP) + dev_priv->edp.support = 1; + /* This bit means to use 96Mhz for DPLL_A or not */ if (driver->primary_lfp_id) dev_priv->dplla_96mhz = true; @@ -437,6 +534,9 @@ int psb_intel_init_bios(struct drm_device *dev) size_t size; int i; + + dev_priv->panel_type = 0xff; + /* XXX Should this validation be moved to intel_opregion.c? */ if (dev_priv->opregion.vbt) { struct vbt_header *vbt = dev_priv->opregion.vbt; @@ -477,6 +577,7 @@ int psb_intel_init_bios(struct drm_device *dev) parse_sdvo_device_mapping(dev_priv, bdb); parse_device_mapping(dev_priv, bdb); parse_backlight_data(dev_priv, bdb); + parse_edp(dev_priv, bdb); if (bios) pci_unmap_rom(pdev, bios); diff --git a/drivers/gpu/drm/gma500/intel_bios.h b/drivers/gpu/drm/gma500/intel_bios.h index 2e95523b84b1..c6267c98c9e7 100644 --- a/drivers/gpu/drm/gma500/intel_bios.h +++ b/drivers/gpu/drm/gma500/intel_bios.h @@ -23,6 +23,7 @@ #define _I830_BIOS_H_ #include <drm/drmP.h> +#include <drm/drm_dp_helper.h> struct vbt_header { u8 signature[20]; /**< Always starts with 'VBT$' */ @@ -93,6 +94,7 @@ struct vbios_data { #define BDB_SDVO_LVDS_PNP_IDS 24 #define BDB_SDVO_LVDS_POWER_SEQ 25 #define BDB_TV_OPTIONS 26 +#define BDB_EDP 27 #define BDB_LVDS_OPTIONS 40 #define BDB_LVDS_LFP_DATA_PTRS 41 #define BDB_LVDS_LFP_DATA 42 @@ -391,6 +393,11 @@ struct bdb_sdvo_lvds_options { u8 panel_misc_bits_4; } __attribute__((packed)); +#define BDB_DRIVER_FEATURE_NO_LVDS 0 +#define BDB_DRIVER_FEATURE_INT_LVDS 1 +#define BDB_DRIVER_FEATURE_SDVO_LVDS 2 +#define BDB_DRIVER_FEATURE_EDP 3 + struct bdb_driver_features { u8 boot_dev_algorithm:1; u8 block_display_switch:1; @@ -431,6 +438,45 @@ struct bdb_driver_features { u8 custom_vbt_version; } __attribute__((packed)); +#define EDP_18BPP 0 +#define EDP_24BPP 1 +#define EDP_30BPP 2 +#define EDP_RATE_1_62 0 +#define EDP_RATE_2_7 1 +#define EDP_LANE_1 0 +#define EDP_LANE_2 1 +#define EDP_LANE_4 3 +#define EDP_PREEMPHASIS_NONE 0 +#define EDP_PREEMPHASIS_3_5dB 1 +#define EDP_PREEMPHASIS_6dB 2 +#define EDP_PREEMPHASIS_9_5dB 3 +#define EDP_VSWING_0_4V 0 +#define EDP_VSWING_0_6V 1 +#define EDP_VSWING_0_8V 2 +#define EDP_VSWING_1_2V 3 + +struct edp_power_seq { + u16 t1_t3; + u16 t8; + u16 t9; + u16 t10; + u16 t11_t12; +} __attribute__ ((packed)); + +struct edp_link_params { + u8 rate:4; + u8 lanes:4; + u8 preemphasis:4; + u8 vswing:4; +} __attribute__ ((packed)); + +struct bdb_edp { + struct edp_power_seq power_seqs[16]; + u32 color_depth; + u32 sdrrs_msa_timing_delay; + struct edp_link_params link_params[16]; +} __attribute__ ((packed)); + extern int psb_intel_init_bios(struct drm_device *dev); extern void psb_intel_destroy_bios(struct drm_device *dev); diff --git a/drivers/gpu/drm/gma500/mdfld_dsi_output.c b/drivers/gpu/drm/gma500/mdfld_dsi_output.c index 5675d93b4205..32dba2ab53e1 100644 --- a/drivers/gpu/drm/gma500/mdfld_dsi_output.c +++ b/drivers/gpu/drm/gma500/mdfld_dsi_output.c @@ -299,17 +299,8 @@ static int mdfld_dsi_connector_set_property(struct drm_connector *connector, if (drm_connector_property_set_value(connector, property, value)) goto set_prop_error; - else { -#ifdef CONFIG_BACKLIGHT_CLASS_DEVICE - struct backlight_device *psb_bd; - - psb_bd = mdfld_get_backlight_device(); - if (psb_bd) { - psb_bd->props.brightness = value; - mdfld_set_brightness(psb_bd); - } -#endif - } + else + gma_backlight_set(encoder->dev, value); } set_prop_done: return 0; diff --git a/drivers/gpu/drm/gma500/mid_bios.c b/drivers/gpu/drm/gma500/mid_bios.c index b2a790bd9899..850cd3fbb969 100644 --- a/drivers/gpu/drm/gma500/mid_bios.c +++ b/drivers/gpu/drm/gma500/mid_bios.c @@ -118,20 +118,20 @@ static void mid_get_pci_revID(struct drm_psb_private *dev_priv) dev_priv->platform_rev_id); } -struct vbt_header { +struct mid_vbt_header { u32 signature; u8 revision; } __packed; /* The same for r0 and r1 */ struct vbt_r0 { - struct vbt_header vbt_header; + struct mid_vbt_header vbt_header; u8 size; u8 checksum; } __packed; struct vbt_r10 { - struct vbt_header vbt_header; + struct mid_vbt_header vbt_header; u8 checksum; u16 size; u8 panel_count; @@ -281,7 +281,7 @@ static void mid_get_vbt_data(struct drm_psb_private *dev_priv) struct drm_device *dev = dev_priv->dev; u32 addr; u8 __iomem *vbt_virtual; - struct vbt_header vbt_header; + struct mid_vbt_header vbt_header; struct pci_dev *pci_gfx_root = pci_get_bus_and_slot(0, PCI_DEVFN(2, 0)); int ret = -1; diff --git a/drivers/gpu/drm/gma500/oaktrail_hdmi.c b/drivers/gpu/drm/gma500/oaktrail_hdmi.c index 2eb3dc4e9c9b..69e51e903f35 100644 --- a/drivers/gpu/drm/gma500/oaktrail_hdmi.c +++ b/drivers/gpu/drm/gma500/oaktrail_hdmi.c @@ -252,7 +252,6 @@ static int oaktrail_hdmi_get_modes(struct drm_connector *connector) if (edid) { drm_mode_connector_update_edid_property(connector, edid); ret = drm_add_edid_modes(connector, edid); - connector->display_info.raw_edid = NULL; } /* diff --git a/drivers/gpu/drm/gma500/opregion.c b/drivers/gpu/drm/gma500/opregion.c index c430bd424681..ad0d6de938f3 100644 --- a/drivers/gpu/drm/gma500/opregion.c +++ b/drivers/gpu/drm/gma500/opregion.c @@ -166,8 +166,7 @@ static u32 asle_set_backlight(struct drm_device *dev, u32 bclp) if (config_enabled(CONFIG_BACKLIGHT_CLASS_DEVICE)) { int max = bd->props.max_brightness; - bd->props.brightness = bclp * max / 255; - backlight_update_status(bd); + gma_backlight_set(dev, bclp * max / 255); } asle->cblv = (bclp * 0x64) / 0xff | ASLE_CBLV_VALID; diff --git a/drivers/gpu/drm/gma500/psb_device.c b/drivers/gpu/drm/gma500/psb_device.c index 5971bc82b765..f1432f096e58 100644 --- a/drivers/gpu/drm/gma500/psb_device.c +++ b/drivers/gpu/drm/gma500/psb_device.c @@ -290,6 +290,7 @@ static void psb_get_core_freq(struct drm_device *dev) case 6: case 7: dev_priv->core_freq = 266; + break; default: dev_priv->core_freq = 0; } diff --git a/drivers/gpu/drm/gma500/psb_drv.h b/drivers/gpu/drm/gma500/psb_drv.h index 1bd115ecefe1..223ff5b1b5c4 100644 --- a/drivers/gpu/drm/gma500/psb_drv.h +++ b/drivers/gpu/drm/gma500/psb_drv.h @@ -24,10 +24,10 @@ #include <drm/drmP.h> #include "drm_global.h" -#include "gem_glue.h" #include "gma_drm.h" #include "psb_reg.h" #include "psb_intel_drv.h" +#include "intel_bios.h" #include "gtt.h" #include "power.h" #include "opregion.h" @@ -613,6 +613,8 @@ struct drm_psb_private { */ struct backlight_device *backlight_device; struct drm_property *backlight_property; + bool backlight_enabled; + int backlight_level; uint32_t blc_adj1; uint32_t blc_adj2; @@ -640,6 +642,19 @@ struct drm_psb_private { int mdfld_panel_id; bool dplla_96mhz; /* DPLL data from the VBT */ + + struct { + int rate; + int lanes; + int preemphasis; + int vswing; + + bool initialized; + bool support; + int bpp; + struct edp_power_seq pps; + } edp; + uint8_t panel_type; }; @@ -796,6 +811,9 @@ extern int psb_fbdev_init(struct drm_device *dev); /* backlight.c */ int gma_backlight_init(struct drm_device *dev); void gma_backlight_exit(struct drm_device *dev); +void gma_backlight_disable(struct drm_device *dev); +void gma_backlight_enable(struct drm_device *dev); +void gma_backlight_set(struct drm_device *dev, int v); /* oaktrail_crtc.c */ extern const struct drm_crtc_helper_funcs oaktrail_helper_funcs; diff --git a/drivers/gpu/drm/gma500/psb_intel_drv.h b/drivers/gpu/drm/gma500/psb_intel_drv.h index ebe1a28f60e1..90f2d11e686b 100644 --- a/drivers/gpu/drm/gma500/psb_intel_drv.h +++ b/drivers/gpu/drm/gma500/psb_intel_drv.h @@ -29,10 +29,6 @@ * Display related stuff */ -/* store information about an Ixxx DVO */ -/* The i830->i865 use multiple DVOs with multiple i2cs */ -/* the i915, i945 have a single sDVO i2c bus - which is different */ -#define MAX_OUTPUTS 6 /* maximum connectors per crtcs in the mode set */ #define INTELFB_CONN_LIMIT 4 @@ -69,6 +65,8 @@ #define INTEL_OUTPUT_HDMI 6 #define INTEL_OUTPUT_MIPI 7 #define INTEL_OUTPUT_MIPI2 8 +#define INTEL_OUTPUT_DISPLAYPORT 9 +#define INTEL_OUTPUT_EDP 10 #define INTEL_DVO_CHIP_NONE 0 #define INTEL_DVO_CHIP_LVDS 1 @@ -133,6 +131,11 @@ struct psb_intel_encoder { void (*hot_plug)(struct psb_intel_encoder *); int crtc_mask; int clone_mask; + u32 ddi_select; /* Channel info */ +#define DDI0_SELECT 0x01 +#define DDI1_SELECT 0x02 +#define DP_MASK 0x8000 +#define DDI_MASK 0x03 void *dev_priv; /* For sdvo_priv, lvds_priv, etc... */ /* FIXME: Either make SDVO and LVDS store it's i2c here or give CDV it's @@ -190,7 +193,6 @@ struct psb_intel_crtc { u32 mode_flags; bool active; - bool crtc_enable; /* Saved Crtc HW states */ struct psb_intel_crtc_state *crtc_state; @@ -285,4 +287,20 @@ extern void gma_intel_gmbus_set_speed(struct i2c_adapter *adapter, int speed); extern void gma_intel_gmbus_force_bit(struct i2c_adapter *adapter, bool force_bit); extern void gma_intel_teardown_gmbus(struct drm_device *dev); +/* DP support */ +extern void cdv_intel_dp_init(struct drm_device *dev, struct psb_intel_mode_device *mode_dev, int output_reg); +extern void cdv_intel_dp_set_m_n(struct drm_crtc *crtc, + struct drm_display_mode *mode, + struct drm_display_mode *adjusted_mode); + +extern void psb_intel_attach_force_audio_property(struct drm_connector *connector); +extern void psb_intel_attach_broadcast_rgb_property(struct drm_connector *connector); + +extern int cdv_sb_read(struct drm_device *dev, u32 reg, u32 *val); +extern int cdv_sb_write(struct drm_device *dev, u32 reg, u32 val); +extern void cdv_sb_reset(struct drm_device *dev); + +extern void cdv_intel_attach_force_audio_property(struct drm_connector *connector); +extern void cdv_intel_attach_broadcast_rgb_property(struct drm_connector *connector); + #endif /* __INTEL_DRV_H__ */ diff --git a/drivers/gpu/drm/gma500/psb_intel_lvds.c b/drivers/gpu/drm/gma500/psb_intel_lvds.c index 37adc9edf974..2a4c3a9e33e3 100644 --- a/drivers/gpu/drm/gma500/psb_intel_lvds.c +++ b/drivers/gpu/drm/gma500/psb_intel_lvds.c @@ -630,17 +630,8 @@ int psb_intel_lvds_set_property(struct drm_connector *connector, property, value)) goto set_prop_error; - else { -#ifdef CONFIG_BACKLIGHT_CLASS_DEVICE - struct drm_psb_private *devp = - encoder->dev->dev_private; - struct backlight_device *bd = devp->backlight_device; - if (bd) { - bd->props.brightness = value; - backlight_update_status(bd); - } -#endif - } + else + gma_backlight_set(encoder->dev, value); } else if (!strcmp(property->name, "DPMS")) { struct drm_encoder_helper_funcs *hfuncs = encoder->helper_private; diff --git a/drivers/gpu/drm/gma500/psb_intel_reg.h b/drivers/gpu/drm/gma500/psb_intel_reg.h index 8e8c8efb0a89..d914719c4b60 100644 --- a/drivers/gpu/drm/gma500/psb_intel_reg.h +++ b/drivers/gpu/drm/gma500/psb_intel_reg.h @@ -173,15 +173,46 @@ #define PP_SEQUENCE_ON (1 << 28) #define PP_SEQUENCE_OFF (2 << 28) #define PP_SEQUENCE_MASK 0x30000000 +#define PP_CYCLE_DELAY_ACTIVE (1 << 27) +#define PP_SEQUENCE_STATE_ON_IDLE (1 << 3) +#define PP_SEQUENCE_STATE_MASK 0x0000000f + #define PP_CONTROL 0x61204 #define POWER_TARGET_ON (1 << 0) - +#define PANEL_UNLOCK_REGS (0xabcd << 16) +#define PANEL_UNLOCK_MASK (0xffff << 16) +#define EDP_FORCE_VDD (1 << 3) +#define EDP_BLC_ENABLE (1 << 2) +#define PANEL_POWER_RESET (1 << 1) +#define PANEL_POWER_OFF (0 << 0) +#define PANEL_POWER_ON (1 << 0) + +/* Poulsbo/Oaktrail */ #define LVDSPP_ON 0x61208 #define LVDSPP_OFF 0x6120c #define PP_CYCLE 0x61210 +/* Cedartrail */ #define PP_ON_DELAYS 0x61208 /* Cedartrail */ +#define PANEL_PORT_SELECT_MASK (3 << 30) +#define PANEL_PORT_SELECT_LVDS (0 << 30) +#define PANEL_PORT_SELECT_EDP (1 << 30) +#define PANEL_POWER_UP_DELAY_MASK (0x1fff0000) +#define PANEL_POWER_UP_DELAY_SHIFT 16 +#define PANEL_LIGHT_ON_DELAY_MASK (0x1fff) +#define PANEL_LIGHT_ON_DELAY_SHIFT 0 + #define PP_OFF_DELAYS 0x6120c /* Cedartrail */ +#define PANEL_POWER_DOWN_DELAY_MASK (0x1fff0000) +#define PANEL_POWER_DOWN_DELAY_SHIFT 16 +#define PANEL_LIGHT_OFF_DELAY_MASK (0x1fff) +#define PANEL_LIGHT_OFF_DELAY_SHIFT 0 + +#define PP_DIVISOR 0x61210 /* Cedartrail */ +#define PP_REFERENCE_DIVIDER_MASK (0xffffff00) +#define PP_REFERENCE_DIVIDER_SHIFT 8 +#define PANEL_POWER_CYCLE_DELAY_MASK (0x1f) +#define PANEL_POWER_CYCLE_DELAY_SHIFT 0 #define PFIT_CONTROL 0x61230 #define PFIT_ENABLE (1 << 31) @@ -1282,6 +1313,10 @@ No status bits are changed. # define VRHUNIT_CLOCK_GATE_DISABLE (1 << 28) /* Fixed value on CDV */ # define DPOUNIT_CLOCK_GATE_DISABLE (1 << 11) # define DPIOUNIT_CLOCK_GATE_DISABLE (1 << 6) +# define DPUNIT_PIPEB_GATE_DISABLE (1 << 30) +# define DPUNIT_PIPEA_GATE_DISABLE (1 << 25) +# define DPCUNIT_CLOCK_GATE_DISABLE (1 << 24) +# define DPLSUNIT_CLOCK_GATE_DISABLE (1 << 13) #define RAMCLK_GATE_D 0x6210 @@ -1347,5 +1382,165 @@ No status bits are changed. #define LANE_PLL_ENABLE (0x3 << 20) #define LANE_PLL_PIPE(p) (((p) == 0) ? (1 << 21) : (0 << 21)) +#define DP_B 0x64100 +#define DP_C 0x64200 + +#define DP_PORT_EN (1 << 31) +#define DP_PIPEB_SELECT (1 << 30) +#define DP_PIPE_MASK (1 << 30) + +/* Link training mode - select a suitable mode for each stage */ +#define DP_LINK_TRAIN_PAT_1 (0 << 28) +#define DP_LINK_TRAIN_PAT_2 (1 << 28) +#define DP_LINK_TRAIN_PAT_IDLE (2 << 28) +#define DP_LINK_TRAIN_OFF (3 << 28) +#define DP_LINK_TRAIN_MASK (3 << 28) +#define DP_LINK_TRAIN_SHIFT 28 + +/* Signal voltages. These are mostly controlled by the other end */ +#define DP_VOLTAGE_0_4 (0 << 25) +#define DP_VOLTAGE_0_6 (1 << 25) +#define DP_VOLTAGE_0_8 (2 << 25) +#define DP_VOLTAGE_1_2 (3 << 25) +#define DP_VOLTAGE_MASK (7 << 25) +#define DP_VOLTAGE_SHIFT 25 + +/* Signal pre-emphasis levels, like voltages, the other end tells us what + * they want + */ +#define DP_PRE_EMPHASIS_0 (0 << 22) +#define DP_PRE_EMPHASIS_3_5 (1 << 22) +#define DP_PRE_EMPHASIS_6 (2 << 22) +#define DP_PRE_EMPHASIS_9_5 (3 << 22) +#define DP_PRE_EMPHASIS_MASK (7 << 22) +#define DP_PRE_EMPHASIS_SHIFT 22 + +/* How many wires to use. I guess 3 was too hard */ +#define DP_PORT_WIDTH_1 (0 << 19) +#define DP_PORT_WIDTH_2 (1 << 19) +#define DP_PORT_WIDTH_4 (3 << 19) +#define DP_PORT_WIDTH_MASK (7 << 19) + +/* Mystic DPCD version 1.1 special mode */ +#define DP_ENHANCED_FRAMING (1 << 18) + +/** locked once port is enabled */ +#define DP_PORT_REVERSAL (1 << 15) + +/** sends the clock on lane 15 of the PEG for debug */ +#define DP_CLOCK_OUTPUT_ENABLE (1 << 13) + +#define DP_SCRAMBLING_DISABLE (1 << 12) +#define DP_SCRAMBLING_DISABLE_IRONLAKE (1 << 7) + +/** limit RGB values to avoid confusing TVs */ +#define DP_COLOR_RANGE_16_235 (1 << 8) + +/** Turn on the audio link */ +#define DP_AUDIO_OUTPUT_ENABLE (1 << 6) + +/** vs and hs sync polarity */ +#define DP_SYNC_VS_HIGH (1 << 4) +#define DP_SYNC_HS_HIGH (1 << 3) + +/** A fantasy */ +#define DP_DETECTED (1 << 2) + +/** The aux channel provides a way to talk to the + * signal sink for DDC etc. Max packet size supported + * is 20 bytes in each direction, hence the 5 fixed + * data registers + */ +#define DPB_AUX_CH_CTL 0x64110 +#define DPB_AUX_CH_DATA1 0x64114 +#define DPB_AUX_CH_DATA2 0x64118 +#define DPB_AUX_CH_DATA3 0x6411c +#define DPB_AUX_CH_DATA4 0x64120 +#define DPB_AUX_CH_DATA5 0x64124 + +#define DPC_AUX_CH_CTL 0x64210 +#define DPC_AUX_CH_DATA1 0x64214 +#define DPC_AUX_CH_DATA2 0x64218 +#define DPC_AUX_CH_DATA3 0x6421c +#define DPC_AUX_CH_DATA4 0x64220 +#define DPC_AUX_CH_DATA5 0x64224 + +#define DP_AUX_CH_CTL_SEND_BUSY (1 << 31) +#define DP_AUX_CH_CTL_DONE (1 << 30) +#define DP_AUX_CH_CTL_INTERRUPT (1 << 29) +#define DP_AUX_CH_CTL_TIME_OUT_ERROR (1 << 28) +#define DP_AUX_CH_CTL_TIME_OUT_400us (0 << 26) +#define DP_AUX_CH_CTL_TIME_OUT_600us (1 << 26) +#define DP_AUX_CH_CTL_TIME_OUT_800us (2 << 26) +#define DP_AUX_CH_CTL_TIME_OUT_1600us (3 << 26) +#define DP_AUX_CH_CTL_TIME_OUT_MASK (3 << 26) +#define DP_AUX_CH_CTL_RECEIVE_ERROR (1 << 25) +#define DP_AUX_CH_CTL_MESSAGE_SIZE_MASK (0x1f << 20) +#define DP_AUX_CH_CTL_MESSAGE_SIZE_SHIFT 20 +#define DP_AUX_CH_CTL_PRECHARGE_2US_MASK (0xf << 16) +#define DP_AUX_CH_CTL_PRECHARGE_2US_SHIFT 16 +#define DP_AUX_CH_CTL_AUX_AKSV_SELECT (1 << 15) +#define DP_AUX_CH_CTL_MANCHESTER_TEST (1 << 14) +#define DP_AUX_CH_CTL_SYNC_TEST (1 << 13) +#define DP_AUX_CH_CTL_DEGLITCH_TEST (1 << 12) +#define DP_AUX_CH_CTL_PRECHARGE_TEST (1 << 11) +#define DP_AUX_CH_CTL_BIT_CLOCK_2X_MASK (0x7ff) +#define DP_AUX_CH_CTL_BIT_CLOCK_2X_SHIFT 0 + +/* + * Computing GMCH M and N values for the Display Port link + * + * GMCH M/N = dot clock * bytes per pixel / ls_clk * # of lanes + * + * ls_clk (we assume) is the DP link clock (1.62 or 2.7 GHz) + * + * The GMCH value is used internally + * + * bytes_per_pixel is the number of bytes coming out of the plane, + * which is after the LUTs, so we want the bytes for our color format. + * For our current usage, this is always 3, one byte for R, G and B. + */ + +#define _PIPEA_GMCH_DATA_M 0x70050 +#define _PIPEB_GMCH_DATA_M 0x71050 + +/* Transfer unit size for display port - 1, default is 0x3f (for TU size 64) */ +#define PIPE_GMCH_DATA_M_TU_SIZE_MASK (0x3f << 25) +#define PIPE_GMCH_DATA_M_TU_SIZE_SHIFT 25 + +#define PIPE_GMCH_DATA_M_MASK (0xffffff) + +#define _PIPEA_GMCH_DATA_N 0x70054 +#define _PIPEB_GMCH_DATA_N 0x71054 +#define PIPE_GMCH_DATA_N_MASK (0xffffff) + +/* + * Computing Link M and N values for the Display Port link + * + * Link M / N = pixel_clock / ls_clk + * + * (the DP spec calls pixel_clock the 'strm_clk') + * + * The Link value is transmitted in the Main Stream + * Attributes and VB-ID. + */ + +#define _PIPEA_DP_LINK_M 0x70060 +#define _PIPEB_DP_LINK_M 0x71060 +#define PIPEA_DP_LINK_M_MASK (0xffffff) + +#define _PIPEA_DP_LINK_N 0x70064 +#define _PIPEB_DP_LINK_N 0x71064 +#define PIPEA_DP_LINK_N_MASK (0xffffff) + +#define PIPE_GMCH_DATA_M(pipe) _PIPE(pipe, _PIPEA_GMCH_DATA_M, _PIPEB_GMCH_DATA_M) +#define PIPE_GMCH_DATA_N(pipe) _PIPE(pipe, _PIPEA_GMCH_DATA_N, _PIPEB_GMCH_DATA_N) +#define PIPE_DP_LINK_M(pipe) _PIPE(pipe, _PIPEA_DP_LINK_M, _PIPEB_DP_LINK_M) +#define PIPE_DP_LINK_N(pipe) _PIPE(pipe, _PIPEA_DP_LINK_N, _PIPEB_DP_LINK_N) + +#define PIPE_BPC_MASK (7 << 5) +#define PIPE_8BPC (0 << 5) +#define PIPE_10BPC (1 << 5) +#define PIPE_6BPC (2 << 5) #endif diff --git a/drivers/gpu/drm/gma500/psb_intel_sdvo.c b/drivers/gpu/drm/gma500/psb_intel_sdvo.c index 0466c7b985f8..d35f93ba3a85 100644 --- a/drivers/gpu/drm/gma500/psb_intel_sdvo.c +++ b/drivers/gpu/drm/gma500/psb_intel_sdvo.c @@ -1292,7 +1292,6 @@ psb_intel_sdvo_get_analog_edid(struct drm_connector *connector) return drm_get_edid(connector, &dev_priv->gmbus[dev_priv->crt_ddc_pin].adapter); - return NULL; } static enum drm_connector_status @@ -1343,7 +1342,6 @@ psb_intel_sdvo_hdmi_sink_detect(struct drm_connector *connector) } } else status = connector_status_disconnected; - connector->display_info.raw_edid = NULL; kfree(edid); } @@ -1404,7 +1402,6 @@ psb_intel_sdvo_detect(struct drm_connector *connector, bool force) ret = connector_status_disconnected; else ret = connector_status_connected; - connector->display_info.raw_edid = NULL; kfree(edid); } else ret = connector_status_connected; @@ -1453,7 +1450,6 @@ static void psb_intel_sdvo_get_ddc_modes(struct drm_connector *connector) drm_add_edid_modes(connector, edid); } - connector->display_info.raw_edid = NULL; kfree(edid); } } diff --git a/drivers/gpu/drm/i2c/ch7006_drv.c b/drivers/gpu/drm/i2c/ch7006_drv.c index 36d952280c50..599099fe76e3 100644 --- a/drivers/gpu/drm/i2c/ch7006_drv.c +++ b/drivers/gpu/drm/i2c/ch7006_drv.c @@ -427,15 +427,10 @@ static int ch7006_remove(struct i2c_client *client) return 0; } -static int ch7006_suspend(struct i2c_client *client, pm_message_t mesg) +static int ch7006_resume(struct device *dev) { - ch7006_dbg(client, "\n"); - - return 0; -} + struct i2c_client *client = to_i2c_client(dev); -static int ch7006_resume(struct i2c_client *client) -{ ch7006_dbg(client, "\n"); ch7006_write(client, 0x3d, 0x0); @@ -499,15 +494,18 @@ static struct i2c_device_id ch7006_ids[] = { }; MODULE_DEVICE_TABLE(i2c, ch7006_ids); +static const struct dev_pm_ops ch7006_pm_ops = { + .resume = ch7006_resume, +}; + static struct drm_i2c_encoder_driver ch7006_driver = { .i2c_driver = { .probe = ch7006_probe, .remove = ch7006_remove, - .suspend = ch7006_suspend, - .resume = ch7006_resume, .driver = { .name = "ch7006", + .pm = &ch7006_pm_ops, }, .id_table = ch7006_ids, diff --git a/drivers/gpu/drm/i915/Makefile b/drivers/gpu/drm/i915/Makefile index b0bacdba6d7e..0f2c5493242b 100644 --- a/drivers/gpu/drm/i915/Makefile +++ b/drivers/gpu/drm/i915/Makefile @@ -40,6 +40,7 @@ i915-y := i915_drv.o i915_dma.o i915_irq.o \ dvo_ivch.o \ dvo_tfp410.o \ dvo_sil164.o \ + dvo_ns2501.o \ i915_gem_dmabuf.o i915-$(CONFIG_COMPAT) += i915_ioc32.o diff --git a/drivers/gpu/drm/i915/dvo.h b/drivers/gpu/drm/i915/dvo.h index 58914691a77b..74b5efccfdb1 100644 --- a/drivers/gpu/drm/i915/dvo.h +++ b/drivers/gpu/drm/i915/dvo.h @@ -58,13 +58,12 @@ struct intel_dvo_dev_ops { void (*create_resources)(struct intel_dvo_device *dvo); /* - * Turn on/off output or set intermediate power levels if available. + * Turn on/off output. * - * Unsupported intermediate modes drop to the lower power setting. - * If the mode is DPMSModeOff, the output must be disabled, - * as the DPLL may be disabled afterwards. + * Because none of our dvo drivers support an intermediate power levels, + * we don't expose this in the interfac. */ - void (*dpms)(struct intel_dvo_device *dvo, int mode); + void (*dpms)(struct intel_dvo_device *dvo, bool enable); /* * Callback for testing a video mode for a given output. @@ -115,6 +114,12 @@ struct intel_dvo_dev_ops { */ enum drm_connector_status (*detect)(struct intel_dvo_device *dvo); + /* + * Probe the current hw status, returning true if the connected output + * is active. + */ + bool (*get_hw_state)(struct intel_dvo_device *dev); + /** * Query the device for the modes it provides. * @@ -140,5 +145,6 @@ extern struct intel_dvo_dev_ops ch7xxx_ops; extern struct intel_dvo_dev_ops ivch_ops; extern struct intel_dvo_dev_ops tfp410_ops; extern struct intel_dvo_dev_ops ch7017_ops; +extern struct intel_dvo_dev_ops ns2501_ops; #endif /* _INTEL_DVO_H */ diff --git a/drivers/gpu/drm/i915/dvo_ch7017.c b/drivers/gpu/drm/i915/dvo_ch7017.c index 1ca799a1e1fc..86b27d1d90c2 100644 --- a/drivers/gpu/drm/i915/dvo_ch7017.c +++ b/drivers/gpu/drm/i915/dvo_ch7017.c @@ -163,7 +163,7 @@ struct ch7017_priv { }; static void ch7017_dump_regs(struct intel_dvo_device *dvo); -static void ch7017_dpms(struct intel_dvo_device *dvo, int mode); +static void ch7017_dpms(struct intel_dvo_device *dvo, bool enable); static bool ch7017_read(struct intel_dvo_device *dvo, u8 addr, u8 *val) { @@ -309,7 +309,7 @@ static void ch7017_mode_set(struct intel_dvo_device *dvo, lvds_power_down = CH7017_LVDS_POWER_DOWN_DEFAULT_RESERVED | (mode->hdisplay & 0x0700) >> 8; - ch7017_dpms(dvo, DRM_MODE_DPMS_OFF); + ch7017_dpms(dvo, false); ch7017_write(dvo, CH7017_HORIZONTAL_ACTIVE_PIXEL_INPUT, horizontal_active_pixel_input); ch7017_write(dvo, CH7017_HORIZONTAL_ACTIVE_PIXEL_OUTPUT, @@ -331,7 +331,7 @@ static void ch7017_mode_set(struct intel_dvo_device *dvo, } /* set the CH7017 power state */ -static void ch7017_dpms(struct intel_dvo_device *dvo, int mode) +static void ch7017_dpms(struct intel_dvo_device *dvo, bool enable) { uint8_t val; @@ -345,7 +345,7 @@ static void ch7017_dpms(struct intel_dvo_device *dvo, int mode) CH7017_DAC3_POWER_DOWN | CH7017_TV_POWER_DOWN_EN); - if (mode == DRM_MODE_DPMS_ON) { + if (enable) { /* Turn on the LVDS */ ch7017_write(dvo, CH7017_LVDS_POWER_DOWN, val & ~CH7017_LVDS_POWER_DOWN_EN); @@ -359,6 +359,18 @@ static void ch7017_dpms(struct intel_dvo_device *dvo, int mode) msleep(20); } +static bool ch7017_get_hw_state(struct intel_dvo_device *dvo) +{ + uint8_t val; + + ch7017_read(dvo, CH7017_LVDS_POWER_DOWN, &val); + + if (val & CH7017_LVDS_POWER_DOWN_EN) + return false; + else + return true; +} + static void ch7017_dump_regs(struct intel_dvo_device *dvo) { uint8_t val; @@ -396,6 +408,7 @@ struct intel_dvo_dev_ops ch7017_ops = { .mode_valid = ch7017_mode_valid, .mode_set = ch7017_mode_set, .dpms = ch7017_dpms, + .get_hw_state = ch7017_get_hw_state, .dump_regs = ch7017_dump_regs, .destroy = ch7017_destroy, }; diff --git a/drivers/gpu/drm/i915/dvo_ch7xxx.c b/drivers/gpu/drm/i915/dvo_ch7xxx.c index 4a036600e806..38f3a6cb8c7d 100644 --- a/drivers/gpu/drm/i915/dvo_ch7xxx.c +++ b/drivers/gpu/drm/i915/dvo_ch7xxx.c @@ -289,14 +289,26 @@ static void ch7xxx_mode_set(struct intel_dvo_device *dvo, } /* set the CH7xxx power state */ -static void ch7xxx_dpms(struct intel_dvo_device *dvo, int mode) +static void ch7xxx_dpms(struct intel_dvo_device *dvo, bool enable) { - if (mode == DRM_MODE_DPMS_ON) + if (enable) ch7xxx_writeb(dvo, CH7xxx_PM, CH7xxx_PM_DVIL | CH7xxx_PM_DVIP); else ch7xxx_writeb(dvo, CH7xxx_PM, CH7xxx_PM_FPD); } +static bool ch7xxx_get_hw_state(struct intel_dvo_device *dvo) +{ + u8 val; + + ch7xxx_readb(dvo, CH7xxx_PM, &val); + + if (val & CH7xxx_PM_FPD) + return false; + else + return true; +} + static void ch7xxx_dump_regs(struct intel_dvo_device *dvo) { int i; @@ -326,6 +338,7 @@ struct intel_dvo_dev_ops ch7xxx_ops = { .mode_valid = ch7xxx_mode_valid, .mode_set = ch7xxx_mode_set, .dpms = ch7xxx_dpms, + .get_hw_state = ch7xxx_get_hw_state, .dump_regs = ch7xxx_dump_regs, .destroy = ch7xxx_destroy, }; diff --git a/drivers/gpu/drm/i915/dvo_ivch.c b/drivers/gpu/drm/i915/dvo_ivch.c index 04f2893d5e3c..baaf65bf0bdd 100644 --- a/drivers/gpu/drm/i915/dvo_ivch.c +++ b/drivers/gpu/drm/i915/dvo_ivch.c @@ -288,7 +288,7 @@ static enum drm_mode_status ivch_mode_valid(struct intel_dvo_device *dvo, } /** Sets the power state of the panel connected to the ivch */ -static void ivch_dpms(struct intel_dvo_device *dvo, int mode) +static void ivch_dpms(struct intel_dvo_device *dvo, bool enable) { int i; uint16_t vr01, vr30, backlight; @@ -297,13 +297,13 @@ static void ivch_dpms(struct intel_dvo_device *dvo, int mode) if (!ivch_read(dvo, VR01, &vr01)) return; - if (mode == DRM_MODE_DPMS_ON) + if (enable) backlight = 1; else backlight = 0; ivch_write(dvo, VR80, backlight); - if (mode == DRM_MODE_DPMS_ON) + if (enable) vr01 |= VR01_LCD_ENABLE | VR01_DVO_ENABLE; else vr01 &= ~(VR01_LCD_ENABLE | VR01_DVO_ENABLE); @@ -315,7 +315,7 @@ static void ivch_dpms(struct intel_dvo_device *dvo, int mode) if (!ivch_read(dvo, VR30, &vr30)) break; - if (((vr30 & VR30_PANEL_ON) != 0) == (mode == DRM_MODE_DPMS_ON)) + if (((vr30 & VR30_PANEL_ON) != 0) == enable) break; udelay(1000); } @@ -323,6 +323,20 @@ static void ivch_dpms(struct intel_dvo_device *dvo, int mode) udelay(16 * 1000); } +static bool ivch_get_hw_state(struct intel_dvo_device *dvo) +{ + uint16_t vr01; + + /* Set the new power state of the panel. */ + if (!ivch_read(dvo, VR01, &vr01)) + return false; + + if (vr01 & VR01_LCD_ENABLE) + return true; + else + return false; +} + static void ivch_mode_set(struct intel_dvo_device *dvo, struct drm_display_mode *mode, struct drm_display_mode *adjusted_mode) @@ -413,6 +427,7 @@ static void ivch_destroy(struct intel_dvo_device *dvo) struct intel_dvo_dev_ops ivch_ops = { .init = ivch_init, .dpms = ivch_dpms, + .get_hw_state = ivch_get_hw_state, .mode_valid = ivch_mode_valid, .mode_set = ivch_mode_set, .detect = ivch_detect, diff --git a/drivers/gpu/drm/i915/dvo_ns2501.c b/drivers/gpu/drm/i915/dvo_ns2501.c new file mode 100644 index 000000000000..c4a255be6979 --- /dev/null +++ b/drivers/gpu/drm/i915/dvo_ns2501.c @@ -0,0 +1,588 @@ +/* + * + * Copyright (c) 2012 Gilles Dartiguelongue, Thomas Richter + * + * All Rights Reserved. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the + * "Software"), to deal in the Software without restriction, including + * without limitation the rights to use, copy, modify, merge, publish, + * distribute, sub license, and/or sell copies of the Software, and to + * permit persons to whom the Software is furnished to do so, subject to + * the following conditions: + * + * The above copyright notice and this permission notice (including the + * next paragraph) shall be included in all copies or substantial portions + * of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS + * OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF + * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NON-INFRINGEMENT. + * IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR + * ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, + * TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE + * SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE. + * + */ + +#include "dvo.h" +#include "i915_reg.h" +#include "i915_drv.h" + +#define NS2501_VID 0x1305 +#define NS2501_DID 0x6726 + +#define NS2501_VID_LO 0x00 +#define NS2501_VID_HI 0x01 +#define NS2501_DID_LO 0x02 +#define NS2501_DID_HI 0x03 +#define NS2501_REV 0x04 +#define NS2501_RSVD 0x05 +#define NS2501_FREQ_LO 0x06 +#define NS2501_FREQ_HI 0x07 + +#define NS2501_REG8 0x08 +#define NS2501_8_VEN (1<<5) +#define NS2501_8_HEN (1<<4) +#define NS2501_8_DSEL (1<<3) +#define NS2501_8_BPAS (1<<2) +#define NS2501_8_RSVD (1<<1) +#define NS2501_8_PD (1<<0) + +#define NS2501_REG9 0x09 +#define NS2501_9_VLOW (1<<7) +#define NS2501_9_MSEL_MASK (0x7<<4) +#define NS2501_9_TSEL (1<<3) +#define NS2501_9_RSEN (1<<2) +#define NS2501_9_RSVD (1<<1) +#define NS2501_9_MDI (1<<0) + +#define NS2501_REGC 0x0c + +struct ns2501_priv { + //I2CDevRec d; + bool quiet; + int reg_8_shadow; + int reg_8_set; + // Shadow registers for i915 + int dvoc; + int pll_a; + int srcdim; + int fw_blc; +}; + +#define NSPTR(d) ((NS2501Ptr)(d->DriverPrivate.ptr)) + +/* + * For reasons unclear to me, the ns2501 at least on the Fujitsu/Siemens + * laptops does not react on the i2c bus unless + * both the PLL is running and the display is configured in its native + * resolution. + * This function forces the DVO on, and stores the registers it touches. + * Afterwards, registers are restored to regular values. + * + * This is pretty much a hack, though it works. + * Without that, ns2501_readb and ns2501_writeb fail + * when switching the resolution. + */ + +static void enable_dvo(struct intel_dvo_device *dvo) +{ + struct ns2501_priv *ns = (struct ns2501_priv *)(dvo->dev_priv); + struct i2c_adapter *adapter = dvo->i2c_bus; + struct intel_gmbus *bus = container_of(adapter, + struct intel_gmbus, + adapter); + struct drm_i915_private *dev_priv = bus->dev_priv; + + DRM_DEBUG_KMS("%s: Trying to re-enable the DVO\n", __FUNCTION__); + + ns->dvoc = I915_READ(DVO_C); + ns->pll_a = I915_READ(_DPLL_A); + ns->srcdim = I915_READ(DVOC_SRCDIM); + ns->fw_blc = I915_READ(FW_BLC); + + I915_WRITE(DVOC, 0x10004084); + I915_WRITE(_DPLL_A, 0xd0820000); + I915_WRITE(DVOC_SRCDIM, 0x400300); // 1024x768 + I915_WRITE(FW_BLC, 0x1080304); + + I915_WRITE(DVOC, 0x90004084); +} + +/* + * Restore the I915 registers modified by the above + * trigger function. + */ +static void restore_dvo(struct intel_dvo_device *dvo) +{ + struct i2c_adapter *adapter = dvo->i2c_bus; + struct intel_gmbus *bus = container_of(adapter, + struct intel_gmbus, + adapter); + struct drm_i915_private *dev_priv = bus->dev_priv; + struct ns2501_priv *ns = (struct ns2501_priv *)(dvo->dev_priv); + + I915_WRITE(DVOC, ns->dvoc); + I915_WRITE(_DPLL_A, ns->pll_a); + I915_WRITE(DVOC_SRCDIM, ns->srcdim); + I915_WRITE(FW_BLC, ns->fw_blc); +} + +/* +** Read a register from the ns2501. +** Returns true if successful, false otherwise. +** If it returns false, it might be wise to enable the +** DVO with the above function. +*/ +static bool ns2501_readb(struct intel_dvo_device *dvo, int addr, uint8_t * ch) +{ + struct ns2501_priv *ns = dvo->dev_priv; + struct i2c_adapter *adapter = dvo->i2c_bus; + u8 out_buf[2]; + u8 in_buf[2]; + + struct i2c_msg msgs[] = { + { + .addr = dvo->slave_addr, + .flags = 0, + .len = 1, + .buf = out_buf, + }, + { + .addr = dvo->slave_addr, + .flags = I2C_M_RD, + .len = 1, + .buf = in_buf, + } + }; + + out_buf[0] = addr; + out_buf[1] = 0; + + if (i2c_transfer(adapter, msgs, 2) == 2) { + *ch = in_buf[0]; + return true; + }; + + if (!ns->quiet) { + DRM_DEBUG_KMS + ("Unable to read register 0x%02x from %s:0x%02x.\n", addr, + adapter->name, dvo->slave_addr); + } + + return false; +} + +/* +** Write a register to the ns2501. +** Returns true if successful, false otherwise. +** If it returns false, it might be wise to enable the +** DVO with the above function. +*/ +static bool ns2501_writeb(struct intel_dvo_device *dvo, int addr, uint8_t ch) +{ + struct ns2501_priv *ns = dvo->dev_priv; + struct i2c_adapter *adapter = dvo->i2c_bus; + uint8_t out_buf[2]; + + struct i2c_msg msg = { + .addr = dvo->slave_addr, + .flags = 0, + .len = 2, + .buf = out_buf, + }; + + out_buf[0] = addr; + out_buf[1] = ch; + + if (i2c_transfer(adapter, &msg, 1) == 1) { + return true; + } + + if (!ns->quiet) { + DRM_DEBUG_KMS("Unable to write register 0x%02x to %s:%d\n", + addr, adapter->name, dvo->slave_addr); + } + + return false; +} + +/* National Semiconductor 2501 driver for chip on i2c bus + * scan for the chip on the bus. + * Hope the VBIOS initialized the PLL correctly so we can + * talk to it. If not, it will not be seen and not detected. + * Bummer! + */ +static bool ns2501_init(struct intel_dvo_device *dvo, + struct i2c_adapter *adapter) +{ + /* this will detect the NS2501 chip on the specified i2c bus */ + struct ns2501_priv *ns; + unsigned char ch; + + ns = kzalloc(sizeof(struct ns2501_priv), GFP_KERNEL); + if (ns == NULL) + return false; + + dvo->i2c_bus = adapter; + dvo->dev_priv = ns; + ns->quiet = true; + + if (!ns2501_readb(dvo, NS2501_VID_LO, &ch)) + goto out; + + if (ch != (NS2501_VID & 0xff)) { + DRM_DEBUG_KMS("ns2501 not detected got %d: from %s Slave %d.\n", + ch, adapter->name, dvo->slave_addr); + goto out; + } + + if (!ns2501_readb(dvo, NS2501_DID_LO, &ch)) + goto out; + + if (ch != (NS2501_DID & 0xff)) { + DRM_DEBUG_KMS("ns2501 not detected got %d: from %s Slave %d.\n", + ch, adapter->name, dvo->slave_addr); + goto out; + } + ns->quiet = false; + ns->reg_8_set = 0; + ns->reg_8_shadow = + NS2501_8_PD | NS2501_8_BPAS | NS2501_8_VEN | NS2501_8_HEN; + + DRM_DEBUG_KMS("init ns2501 dvo controller successfully!\n"); + return true; + +out: + kfree(ns); + return false; +} + +static enum drm_connector_status ns2501_detect(struct intel_dvo_device *dvo) +{ + /* + * This is a Laptop display, it doesn't have hotplugging. + * Even if not, the detection bit of the 2501 is unreliable as + * it only works for some display types. + * It is even more unreliable as the PLL must be active for + * allowing reading from the chiop. + */ + return connector_status_connected; +} + +static enum drm_mode_status ns2501_mode_valid(struct intel_dvo_device *dvo, + struct drm_display_mode *mode) +{ + DRM_DEBUG_KMS + ("%s: is mode valid (hdisplay=%d,htotal=%d,vdisplay=%d,vtotal=%d)\n", + __FUNCTION__, mode->hdisplay, mode->htotal, mode->vdisplay, + mode->vtotal); + + /* + * Currently, these are all the modes I have data from. + * More might exist. Unclear how to find the native resolution + * of the panel in here so we could always accept it + * by disabling the scaler. + */ + if ((mode->hdisplay == 800 && mode->vdisplay == 600) || + (mode->hdisplay == 640 && mode->vdisplay == 480) || + (mode->hdisplay == 1024 && mode->vdisplay == 768)) { + return MODE_OK; + } else { + return MODE_ONE_SIZE; /* Is this a reasonable error? */ + } +} + +static void ns2501_mode_set(struct intel_dvo_device *dvo, + struct drm_display_mode *mode, + struct drm_display_mode *adjusted_mode) +{ + bool ok; + bool restore = false; + struct ns2501_priv *ns = (struct ns2501_priv *)(dvo->dev_priv); + + DRM_DEBUG_KMS + ("%s: set mode (hdisplay=%d,htotal=%d,vdisplay=%d,vtotal=%d).\n", + __FUNCTION__, mode->hdisplay, mode->htotal, mode->vdisplay, + mode->vtotal); + + /* + * Where do I find the native resolution for which scaling is not required??? + * + * First trigger the DVO on as otherwise the chip does not appear on the i2c + * bus. + */ + do { + ok = true; + + if (mode->hdisplay == 800 && mode->vdisplay == 600) { + /* mode 277 */ + ns->reg_8_shadow &= ~NS2501_8_BPAS; + DRM_DEBUG_KMS("%s: switching to 800x600\n", + __FUNCTION__); + + /* + * No, I do not know where this data comes from. + * It is just what the video bios left in the DVO, so + * I'm just copying it here over. + * This also means that I cannot support any other modes + * except the ones supported by the bios. + */ + ok &= ns2501_writeb(dvo, 0x11, 0xc8); // 0xc7 also works. + ok &= ns2501_writeb(dvo, 0x1b, 0x19); + ok &= ns2501_writeb(dvo, 0x1c, 0x62); // VBIOS left 0x64 here, but 0x62 works nicer + ok &= ns2501_writeb(dvo, 0x1d, 0x02); + + ok &= ns2501_writeb(dvo, 0x34, 0x03); + ok &= ns2501_writeb(dvo, 0x35, 0xff); + + ok &= ns2501_writeb(dvo, 0x80, 0x27); + ok &= ns2501_writeb(dvo, 0x81, 0x03); + ok &= ns2501_writeb(dvo, 0x82, 0x41); + ok &= ns2501_writeb(dvo, 0x83, 0x05); + + ok &= ns2501_writeb(dvo, 0x8d, 0x02); + ok &= ns2501_writeb(dvo, 0x8e, 0x04); + ok &= ns2501_writeb(dvo, 0x8f, 0x00); + + ok &= ns2501_writeb(dvo, 0x90, 0xfe); /* vertical. VBIOS left 0xff here, but 0xfe works better */ + ok &= ns2501_writeb(dvo, 0x91, 0x07); + ok &= ns2501_writeb(dvo, 0x94, 0x00); + ok &= ns2501_writeb(dvo, 0x95, 0x00); + + ok &= ns2501_writeb(dvo, 0x96, 0x00); + + ok &= ns2501_writeb(dvo, 0x99, 0x00); + ok &= ns2501_writeb(dvo, 0x9a, 0x88); + + ok &= ns2501_writeb(dvo, 0x9c, 0x23); /* Looks like first and last line of the image. */ + ok &= ns2501_writeb(dvo, 0x9d, 0x00); + ok &= ns2501_writeb(dvo, 0x9e, 0x25); + ok &= ns2501_writeb(dvo, 0x9f, 0x03); + + ok &= ns2501_writeb(dvo, 0xa4, 0x80); + + ok &= ns2501_writeb(dvo, 0xb6, 0x00); + + ok &= ns2501_writeb(dvo, 0xb9, 0xc8); /* horizontal? */ + ok &= ns2501_writeb(dvo, 0xba, 0x00); /* horizontal? */ + + ok &= ns2501_writeb(dvo, 0xc0, 0x05); /* horizontal? */ + ok &= ns2501_writeb(dvo, 0xc1, 0xd7); + + ok &= ns2501_writeb(dvo, 0xc2, 0x00); + ok &= ns2501_writeb(dvo, 0xc3, 0xf8); + + ok &= ns2501_writeb(dvo, 0xc4, 0x03); + ok &= ns2501_writeb(dvo, 0xc5, 0x1a); + + ok &= ns2501_writeb(dvo, 0xc6, 0x00); + ok &= ns2501_writeb(dvo, 0xc7, 0x73); + ok &= ns2501_writeb(dvo, 0xc8, 0x02); + + } else if (mode->hdisplay == 640 && mode->vdisplay == 480) { + /* mode 274 */ + DRM_DEBUG_KMS("%s: switching to 640x480\n", + __FUNCTION__); + /* + * No, I do not know where this data comes from. + * It is just what the video bios left in the DVO, so + * I'm just copying it here over. + * This also means that I cannot support any other modes + * except the ones supported by the bios. + */ + ns->reg_8_shadow &= ~NS2501_8_BPAS; + + ok &= ns2501_writeb(dvo, 0x11, 0xa0); + ok &= ns2501_writeb(dvo, 0x1b, 0x11); + ok &= ns2501_writeb(dvo, 0x1c, 0x54); + ok &= ns2501_writeb(dvo, 0x1d, 0x03); + + ok &= ns2501_writeb(dvo, 0x34, 0x03); + ok &= ns2501_writeb(dvo, 0x35, 0xff); + + ok &= ns2501_writeb(dvo, 0x80, 0xff); + ok &= ns2501_writeb(dvo, 0x81, 0x07); + ok &= ns2501_writeb(dvo, 0x82, 0x3d); + ok &= ns2501_writeb(dvo, 0x83, 0x05); + + ok &= ns2501_writeb(dvo, 0x8d, 0x02); + ok &= ns2501_writeb(dvo, 0x8e, 0x10); + ok &= ns2501_writeb(dvo, 0x8f, 0x00); + + ok &= ns2501_writeb(dvo, 0x90, 0xff); /* vertical */ + ok &= ns2501_writeb(dvo, 0x91, 0x07); + ok &= ns2501_writeb(dvo, 0x94, 0x00); + ok &= ns2501_writeb(dvo, 0x95, 0x00); + + ok &= ns2501_writeb(dvo, 0x96, 0x05); + + ok &= ns2501_writeb(dvo, 0x99, 0x00); + ok &= ns2501_writeb(dvo, 0x9a, 0x88); + + ok &= ns2501_writeb(dvo, 0x9c, 0x24); + ok &= ns2501_writeb(dvo, 0x9d, 0x00); + ok &= ns2501_writeb(dvo, 0x9e, 0x25); + ok &= ns2501_writeb(dvo, 0x9f, 0x03); + + ok &= ns2501_writeb(dvo, 0xa4, 0x84); + + ok &= ns2501_writeb(dvo, 0xb6, 0x09); + + ok &= ns2501_writeb(dvo, 0xb9, 0xa0); /* horizontal? */ + ok &= ns2501_writeb(dvo, 0xba, 0x00); /* horizontal? */ + + ok &= ns2501_writeb(dvo, 0xc0, 0x05); /* horizontal? */ + ok &= ns2501_writeb(dvo, 0xc1, 0x90); + + ok &= ns2501_writeb(dvo, 0xc2, 0x00); + ok &= ns2501_writeb(dvo, 0xc3, 0x0f); + + ok &= ns2501_writeb(dvo, 0xc4, 0x03); + ok &= ns2501_writeb(dvo, 0xc5, 0x16); + + ok &= ns2501_writeb(dvo, 0xc6, 0x00); + ok &= ns2501_writeb(dvo, 0xc7, 0x02); + ok &= ns2501_writeb(dvo, 0xc8, 0x02); + + } else if (mode->hdisplay == 1024 && mode->vdisplay == 768) { + /* mode 280 */ + DRM_DEBUG_KMS("%s: switching to 1024x768\n", + __FUNCTION__); + /* + * This might or might not work, actually. I'm silently + * assuming here that the native panel resolution is + * 1024x768. If not, then this leaves the scaler disabled + * generating a picture that is likely not the expected. + * + * Problem is that I do not know where to take the panel + * dimensions from. + * + * Enable the bypass, scaling not required. + * + * The scaler registers are irrelevant here.... + * + */ + ns->reg_8_shadow |= NS2501_8_BPAS; + ok &= ns2501_writeb(dvo, 0x37, 0x44); + } else { + /* + * Data not known. Bummer! + * Hopefully, the code should not go here + * as mode_OK delivered no other modes. + */ + ns->reg_8_shadow |= NS2501_8_BPAS; + } + ok &= ns2501_writeb(dvo, NS2501_REG8, ns->reg_8_shadow); + + if (!ok) { + if (restore) + restore_dvo(dvo); + enable_dvo(dvo); + restore = true; + } + } while (!ok); + /* + * Restore the old i915 registers before + * forcing the ns2501 on. + */ + if (restore) + restore_dvo(dvo); +} + +/* set the NS2501 power state */ +static bool ns2501_get_hw_state(struct intel_dvo_device *dvo) +{ + unsigned char ch; + + if (!ns2501_readb(dvo, NS2501_REG8, &ch)) + return false; + + if (ch & NS2501_8_PD) + return true; + else + return false; +} + +/* set the NS2501 power state */ +static void ns2501_dpms(struct intel_dvo_device *dvo, bool enable) +{ + bool ok; + bool restore = false; + struct ns2501_priv *ns = (struct ns2501_priv *)(dvo->dev_priv); + unsigned char ch; + + DRM_DEBUG_KMS("%s: Trying set the dpms of the DVO to %i\n", + __FUNCTION__, enable); + + ch = ns->reg_8_shadow; + + if (enable) + ch |= NS2501_8_PD; + else + ch &= ~NS2501_8_PD; + + if (ns->reg_8_set == 0 || ns->reg_8_shadow != ch) { + ns->reg_8_set = 1; + ns->reg_8_shadow = ch; + + do { + ok = true; + ok &= ns2501_writeb(dvo, NS2501_REG8, ch); + ok &= + ns2501_writeb(dvo, 0x34, + enable ? 0x03 : 0x00); + ok &= + ns2501_writeb(dvo, 0x35, + enable ? 0xff : 0x00); + if (!ok) { + if (restore) + restore_dvo(dvo); + enable_dvo(dvo); + restore = true; + } + } while (!ok); + + if (restore) + restore_dvo(dvo); + } +} + +static void ns2501_dump_regs(struct intel_dvo_device *dvo) +{ + uint8_t val; + + ns2501_readb(dvo, NS2501_FREQ_LO, &val); + DRM_LOG_KMS("NS2501_FREQ_LO: 0x%02x\n", val); + ns2501_readb(dvo, NS2501_FREQ_HI, &val); + DRM_LOG_KMS("NS2501_FREQ_HI: 0x%02x\n", val); + ns2501_readb(dvo, NS2501_REG8, &val); + DRM_LOG_KMS("NS2501_REG8: 0x%02x\n", val); + ns2501_readb(dvo, NS2501_REG9, &val); + DRM_LOG_KMS("NS2501_REG9: 0x%02x\n", val); + ns2501_readb(dvo, NS2501_REGC, &val); + DRM_LOG_KMS("NS2501_REGC: 0x%02x\n", val); +} + +static void ns2501_destroy(struct intel_dvo_device *dvo) +{ + struct ns2501_priv *ns = dvo->dev_priv; + + if (ns) { + kfree(ns); + dvo->dev_priv = NULL; + } +} + +struct intel_dvo_dev_ops ns2501_ops = { + .init = ns2501_init, + .detect = ns2501_detect, + .mode_valid = ns2501_mode_valid, + .mode_set = ns2501_mode_set, + .dpms = ns2501_dpms, + .get_hw_state = ns2501_get_hw_state, + .dump_regs = ns2501_dump_regs, + .destroy = ns2501_destroy, +}; diff --git a/drivers/gpu/drm/i915/dvo_sil164.c b/drivers/gpu/drm/i915/dvo_sil164.c index a0b13a6f619d..4debd32e3e4c 100644 --- a/drivers/gpu/drm/i915/dvo_sil164.c +++ b/drivers/gpu/drm/i915/dvo_sil164.c @@ -208,7 +208,7 @@ static void sil164_mode_set(struct intel_dvo_device *dvo, } /* set the SIL164 power state */ -static void sil164_dpms(struct intel_dvo_device *dvo, int mode) +static void sil164_dpms(struct intel_dvo_device *dvo, bool enable) { int ret; unsigned char ch; @@ -217,7 +217,7 @@ static void sil164_dpms(struct intel_dvo_device *dvo, int mode) if (ret == false) return; - if (mode == DRM_MODE_DPMS_ON) + if (enable) ch |= SIL164_8_PD; else ch &= ~SIL164_8_PD; @@ -226,6 +226,21 @@ static void sil164_dpms(struct intel_dvo_device *dvo, int mode) return; } +static bool sil164_get_hw_state(struct intel_dvo_device *dvo) +{ + int ret; + unsigned char ch; + + ret = sil164_readb(dvo, SIL164_REG8, &ch); + if (ret == false) + return false; + + if (ch & SIL164_8_PD) + return true; + else + return false; +} + static void sil164_dump_regs(struct intel_dvo_device *dvo) { uint8_t val; @@ -258,6 +273,7 @@ struct intel_dvo_dev_ops sil164_ops = { .mode_valid = sil164_mode_valid, .mode_set = sil164_mode_set, .dpms = sil164_dpms, + .get_hw_state = sil164_get_hw_state, .dump_regs = sil164_dump_regs, .destroy = sil164_destroy, }; diff --git a/drivers/gpu/drm/i915/dvo_tfp410.c b/drivers/gpu/drm/i915/dvo_tfp410.c index aa2cd3ec54aa..e17f1b07e915 100644 --- a/drivers/gpu/drm/i915/dvo_tfp410.c +++ b/drivers/gpu/drm/i915/dvo_tfp410.c @@ -234,14 +234,14 @@ static void tfp410_mode_set(struct intel_dvo_device *dvo, } /* set the tfp410 power state */ -static void tfp410_dpms(struct intel_dvo_device *dvo, int mode) +static void tfp410_dpms(struct intel_dvo_device *dvo, bool enable) { uint8_t ctl1; if (!tfp410_readb(dvo, TFP410_CTL_1, &ctl1)) return; - if (mode == DRM_MODE_DPMS_ON) + if (enable) ctl1 |= TFP410_CTL_1_PD; else ctl1 &= ~TFP410_CTL_1_PD; @@ -249,6 +249,19 @@ static void tfp410_dpms(struct intel_dvo_device *dvo, int mode) tfp410_writeb(dvo, TFP410_CTL_1, ctl1); } +static bool tfp410_get_hw_state(struct intel_dvo_device *dvo) +{ + uint8_t ctl1; + + if (!tfp410_readb(dvo, TFP410_CTL_1, &ctl1)) + return false; + + if (ctl1 & TFP410_CTL_1_PD) + return true; + else + return false; +} + static void tfp410_dump_regs(struct intel_dvo_device *dvo) { uint8_t val, val2; @@ -299,6 +312,7 @@ struct intel_dvo_dev_ops tfp410_ops = { .mode_valid = tfp410_mode_valid, .mode_set = tfp410_mode_set, .dpms = tfp410_dpms, + .get_hw_state = tfp410_get_hw_state, .dump_regs = tfp410_dump_regs, .destroy = tfp410_destroy, }; diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c index 359f6e8b9b00..7a4b3f3aad3f 100644 --- a/drivers/gpu/drm/i915/i915_debugfs.c +++ b/drivers/gpu/drm/i915/i915_debugfs.c @@ -44,7 +44,6 @@ enum { ACTIVE_LIST, - FLUSHING_LIST, INACTIVE_LIST, PINNED_LIST, }; @@ -62,28 +61,11 @@ static int i915_capabilities(struct seq_file *m, void *data) seq_printf(m, "gen: %d\n", info->gen); seq_printf(m, "pch: %d\n", INTEL_PCH_TYPE(dev)); -#define B(x) seq_printf(m, #x ": %s\n", yesno(info->x)) - B(is_mobile); - B(is_i85x); - B(is_i915g); - B(is_i945gm); - B(is_g33); - B(need_gfx_hws); - B(is_g4x); - B(is_pineview); - B(is_broadwater); - B(is_crestline); - B(has_fbc); - B(has_pipe_cxsr); - B(has_hotplug); - B(cursor_needs_physical); - B(has_overlay); - B(overlay_needs_physical); - B(supports_tv); - B(has_bsd_ring); - B(has_blt_ring); - B(has_llc); -#undef B +#define DEV_INFO_FLAG(x) seq_printf(m, #x ": %s\n", yesno(info->x)) +#define DEV_INFO_SEP ; + DEV_INFO_FLAGS; +#undef DEV_INFO_FLAG +#undef DEV_INFO_SEP return 0; } @@ -121,20 +103,23 @@ static const char *cache_level_str(int type) static void describe_obj(struct seq_file *m, struct drm_i915_gem_object *obj) { - seq_printf(m, "%p: %s%s %8zdKiB %04x %04x %d %d%s%s%s", + seq_printf(m, "%p: %s%s %8zdKiB %04x %04x %d %d %d%s%s%s", &obj->base, get_pin_flag(obj), get_tiling_flag(obj), obj->base.size / 1024, obj->base.read_domains, obj->base.write_domain, - obj->last_rendering_seqno, + obj->last_read_seqno, + obj->last_write_seqno, obj->last_fenced_seqno, cache_level_str(obj->cache_level), obj->dirty ? " dirty" : "", obj->madv == I915_MADV_DONTNEED ? " purgeable" : ""); if (obj->base.name) seq_printf(m, " (name: %d)", obj->base.name); + if (obj->pin_count) + seq_printf(m, " (pinned x %d)", obj->pin_count); if (obj->fence_reg != I915_FENCE_REG_NONE) seq_printf(m, " (fence: %d)", obj->fence_reg); if (obj->gtt_space != NULL) @@ -177,10 +162,6 @@ static int i915_gem_object_list_info(struct seq_file *m, void *data) seq_printf(m, "Inactive:\n"); head = &dev_priv->mm.inactive_list; break; - case FLUSHING_LIST: - seq_printf(m, "Flushing:\n"); - head = &dev_priv->mm.flushing_list; - break; default: mutex_unlock(&dev->struct_mutex); return -EINVAL; @@ -218,8 +199,8 @@ static int i915_gem_object_info(struct seq_file *m, void* data) struct drm_info_node *node = (struct drm_info_node *) m->private; struct drm_device *dev = node->minor->dev; struct drm_i915_private *dev_priv = dev->dev_private; - u32 count, mappable_count; - size_t size, mappable_size; + u32 count, mappable_count, purgeable_count; + size_t size, mappable_size, purgeable_size; struct drm_i915_gem_object *obj; int ret; @@ -232,13 +213,12 @@ static int i915_gem_object_info(struct seq_file *m, void* data) dev_priv->mm.object_memory); size = count = mappable_size = mappable_count = 0; - count_objects(&dev_priv->mm.gtt_list, gtt_list); + count_objects(&dev_priv->mm.bound_list, gtt_list); seq_printf(m, "%u [%u] objects, %zu [%zu] bytes in gtt\n", count, mappable_count, size, mappable_size); size = count = mappable_size = mappable_count = 0; count_objects(&dev_priv->mm.active_list, mm_list); - count_objects(&dev_priv->mm.flushing_list, mm_list); seq_printf(m, " %u [%u] active objects, %zu [%zu] bytes\n", count, mappable_count, size, mappable_size); @@ -247,8 +227,16 @@ static int i915_gem_object_info(struct seq_file *m, void* data) seq_printf(m, " %u [%u] inactive objects, %zu [%zu] bytes\n", count, mappable_count, size, mappable_size); + size = count = purgeable_size = purgeable_count = 0; + list_for_each_entry(obj, &dev_priv->mm.unbound_list, gtt_list) { + size += obj->base.size, ++count; + if (obj->madv == I915_MADV_DONTNEED) + purgeable_size += obj->base.size, ++purgeable_count; + } + seq_printf(m, "%u unbound objects, %zu bytes\n", count, size); + size = count = mappable_size = mappable_count = 0; - list_for_each_entry(obj, &dev_priv->mm.gtt_list, gtt_list) { + list_for_each_entry(obj, &dev_priv->mm.bound_list, gtt_list) { if (obj->fault_mappable) { size += obj->gtt_space->size; ++count; @@ -257,7 +245,13 @@ static int i915_gem_object_info(struct seq_file *m, void* data) mappable_size += obj->gtt_space->size; ++mappable_count; } + if (obj->madv == I915_MADV_DONTNEED) { + purgeable_size += obj->base.size; + ++purgeable_count; + } } + seq_printf(m, "%u purgeable objects, %zu bytes\n", + purgeable_count, purgeable_size); seq_printf(m, "%u pinned mappable objects, %zu bytes\n", mappable_count, mappable_size); seq_printf(m, "%u fault mappable objects, %zu bytes\n", @@ -286,7 +280,7 @@ static int i915_gem_gtt_info(struct seq_file *m, void* data) return ret; total_obj_size = total_gtt_size = count = 0; - list_for_each_entry(obj, &dev_priv->mm.gtt_list, gtt_list) { + list_for_each_entry(obj, &dev_priv->mm.bound_list, gtt_list) { if (list == PINNED_LIST && obj->pin_count == 0) continue; @@ -359,40 +353,22 @@ static int i915_gem_request_info(struct seq_file *m, void *data) struct drm_info_node *node = (struct drm_info_node *) m->private; struct drm_device *dev = node->minor->dev; drm_i915_private_t *dev_priv = dev->dev_private; + struct intel_ring_buffer *ring; struct drm_i915_gem_request *gem_request; - int ret, count; + int ret, count, i; ret = mutex_lock_interruptible(&dev->struct_mutex); if (ret) return ret; count = 0; - if (!list_empty(&dev_priv->ring[RCS].request_list)) { - seq_printf(m, "Render requests:\n"); - list_for_each_entry(gem_request, - &dev_priv->ring[RCS].request_list, - list) { - seq_printf(m, " %d @ %d\n", - gem_request->seqno, - (int) (jiffies - gem_request->emitted_jiffies)); - } - count++; - } - if (!list_empty(&dev_priv->ring[VCS].request_list)) { - seq_printf(m, "BSD requests:\n"); - list_for_each_entry(gem_request, - &dev_priv->ring[VCS].request_list, - list) { - seq_printf(m, " %d @ %d\n", - gem_request->seqno, - (int) (jiffies - gem_request->emitted_jiffies)); - } - count++; - } - if (!list_empty(&dev_priv->ring[BCS].request_list)) { - seq_printf(m, "BLT requests:\n"); + for_each_ring(ring, dev_priv, i) { + if (list_empty(&ring->request_list)) + continue; + + seq_printf(m, "%s requests:\n", ring->name); list_for_each_entry(gem_request, - &dev_priv->ring[BCS].request_list, + &ring->request_list, list) { seq_printf(m, " %d @ %d\n", gem_request->seqno, @@ -413,7 +389,7 @@ static void i915_ring_seqno_info(struct seq_file *m, { if (ring->get_seqno) { seq_printf(m, "Current sequence (%s): %d\n", - ring->name, ring->get_seqno(ring)); + ring->name, ring->get_seqno(ring, false)); } } @@ -422,14 +398,15 @@ static int i915_gem_seqno_info(struct seq_file *m, void *data) struct drm_info_node *node = (struct drm_info_node *) m->private; struct drm_device *dev = node->minor->dev; drm_i915_private_t *dev_priv = dev->dev_private; + struct intel_ring_buffer *ring; int ret, i; ret = mutex_lock_interruptible(&dev->struct_mutex); if (ret) return ret; - for (i = 0; i < I915_NUM_RINGS; i++) - i915_ring_seqno_info(m, &dev_priv->ring[i]); + for_each_ring(ring, dev_priv, i) + i915_ring_seqno_info(m, ring); mutex_unlock(&dev->struct_mutex); @@ -442,6 +419,7 @@ static int i915_interrupt_info(struct seq_file *m, void *data) struct drm_info_node *node = (struct drm_info_node *) m->private; struct drm_device *dev = node->minor->dev; drm_i915_private_t *dev_priv = dev->dev_private; + struct intel_ring_buffer *ring; int ret, i, pipe; ret = mutex_lock_interruptible(&dev->struct_mutex); @@ -519,13 +497,13 @@ static int i915_interrupt_info(struct seq_file *m, void *data) } seq_printf(m, "Interrupts received: %d\n", atomic_read(&dev_priv->irq_received)); - for (i = 0; i < I915_NUM_RINGS; i++) { + for_each_ring(ring, dev_priv, i) { if (IS_GEN6(dev) || IS_GEN7(dev)) { - seq_printf(m, "Graphics Interrupt mask (%s): %08x\n", - dev_priv->ring[i].name, - I915_READ_IMR(&dev_priv->ring[i])); + seq_printf(m, + "Graphics Interrupt mask (%s): %08x\n", + ring->name, I915_READ_IMR(ring)); } - i915_ring_seqno_info(m, &dev_priv->ring[i]); + i915_ring_seqno_info(m, ring); } mutex_unlock(&dev->struct_mutex); @@ -548,7 +526,8 @@ static int i915_gem_fence_regs_info(struct seq_file *m, void *data) for (i = 0; i < dev_priv->num_fence_regs; i++) { struct drm_i915_gem_object *obj = dev_priv->fence_regs[i].obj; - seq_printf(m, "Fenced object[%2d] = ", i); + seq_printf(m, "Fence %d, pin count = %d, object = ", + i, dev_priv->fence_regs[i].pin_count); if (obj == NULL) seq_printf(m, "unused"); else @@ -630,12 +609,12 @@ static void print_error_buffers(struct seq_file *m, seq_printf(m, "%s [%d]:\n", name, count); while (count--) { - seq_printf(m, " %08x %8u %04x %04x %08x%s%s%s%s%s%s%s", + seq_printf(m, " %08x %8u %04x %04x %x %x%s%s%s%s%s%s%s", err->gtt_offset, err->size, err->read_domains, err->write_domain, - err->seqno, + err->rseqno, err->wseqno, pin_flag(err->pinned), tiling_flag(err->tiling), dirty_flag(err->dirty), @@ -667,10 +646,9 @@ static void i915_ring_error_state(struct seq_file *m, seq_printf(m, " IPEIR: 0x%08x\n", error->ipeir[ring]); seq_printf(m, " IPEHR: 0x%08x\n", error->ipehr[ring]); seq_printf(m, " INSTDONE: 0x%08x\n", error->instdone[ring]); - if (ring == RCS && INTEL_INFO(dev)->gen >= 4) { - seq_printf(m, " INSTDONE1: 0x%08x\n", error->instdone1); + if (ring == RCS && INTEL_INFO(dev)->gen >= 4) seq_printf(m, " BBADDR: 0x%08llx\n", error->bbaddr); - } + if (INTEL_INFO(dev)->gen >= 4) seq_printf(m, " INSTPS: 0x%08x\n", error->instps[ring]); seq_printf(m, " INSTPM: 0x%08x\n", error->instpm[ring]); @@ -719,11 +697,17 @@ static int i915_error_state(struct seq_file *m, void *unused) for (i = 0; i < dev_priv->num_fence_regs; i++) seq_printf(m, " fence[%d] = %08llx\n", i, error->fence[i]); + for (i = 0; i < ARRAY_SIZE(error->extra_instdone); i++) + seq_printf(m, " INSTDONE_%d: 0x%08x\n", i, error->extra_instdone[i]); + if (INTEL_INFO(dev)->gen >= 6) { seq_printf(m, "ERROR: 0x%08x\n", error->error); seq_printf(m, "DONE_REG: 0x%08x\n", error->done_reg); } + if (INTEL_INFO(dev)->gen == 7) + seq_printf(m, "ERR_INT: 0x%08x\n", error->err_int); + for_each_ring(ring, dev_priv, i) i915_ring_error_state(m, dev, error, i); @@ -799,10 +783,14 @@ i915_error_state_write(struct file *filp, struct seq_file *m = filp->private_data; struct i915_error_state_file_priv *error_priv = m->private; struct drm_device *dev = error_priv->dev; + int ret; DRM_DEBUG_DRIVER("Resetting error state\n"); - mutex_lock(&dev->struct_mutex); + ret = mutex_lock_interruptible(&dev->struct_mutex); + if (ret) + return ret; + i915_destroy_error_state(dev); mutex_unlock(&dev->struct_mutex); @@ -926,7 +914,7 @@ static int i915_cur_delayinfo(struct seq_file *m, void *unused) seq_printf(m, "Render p-state limit: %d\n", rp_state_limits & 0xff); seq_printf(m, "CAGF: %dMHz\n", ((rpstat & GEN6_CAGF_MASK) >> - GEN6_CAGF_SHIFT) * 50); + GEN6_CAGF_SHIFT) * GT_FREQUENCY_MULTIPLIER); seq_printf(m, "RP CUR UP EI: %dus\n", rpupei & GEN6_CURICONT_MASK); seq_printf(m, "RP CUR UP: %dus\n", rpcurup & @@ -942,15 +930,15 @@ static int i915_cur_delayinfo(struct seq_file *m, void *unused) max_freq = (rp_state_cap & 0xff0000) >> 16; seq_printf(m, "Lowest (RPN) frequency: %dMHz\n", - max_freq * 50); + max_freq * GT_FREQUENCY_MULTIPLIER); max_freq = (rp_state_cap & 0xff00) >> 8; seq_printf(m, "Nominal (RP1) frequency: %dMHz\n", - max_freq * 50); + max_freq * GT_FREQUENCY_MULTIPLIER); max_freq = rp_state_cap & 0xff; seq_printf(m, "Max non-overclocked (RP0) frequency: %dMHz\n", - max_freq * 50); + max_freq * GT_FREQUENCY_MULTIPLIER); } else { seq_printf(m, "no P-state info available\n"); } @@ -1292,7 +1280,8 @@ static int i915_ring_freq_table(struct seq_file *m, void *unused) seq_printf(m, "GPU freq (MHz)\tEffective CPU freq (MHz)\n"); - for (gpu_freq = dev_priv->min_delay; gpu_freq <= dev_priv->max_delay; + for (gpu_freq = dev_priv->rps.min_delay; + gpu_freq <= dev_priv->rps.max_delay; gpu_freq++) { I915_WRITE(GEN6_PCODE_DATA, gpu_freq); I915_WRITE(GEN6_PCODE_MAILBOX, GEN6_PCODE_READY | @@ -1303,7 +1292,7 @@ static int i915_ring_freq_table(struct seq_file *m, void *unused) continue; } ia_freq = I915_READ(GEN6_PCODE_DATA); - seq_printf(m, "%d\t\t%d\n", gpu_freq * 50, ia_freq * 100); + seq_printf(m, "%d\t\t%d\n", gpu_freq * GT_FREQUENCY_MULTIPLIER, ia_freq * 100); } mutex_unlock(&dev->struct_mutex); @@ -1472,8 +1461,12 @@ static int i915_swizzle_info(struct seq_file *m, void *data) struct drm_info_node *node = (struct drm_info_node *) m->private; struct drm_device *dev = node->minor->dev; struct drm_i915_private *dev_priv = dev->dev_private; + int ret; + + ret = mutex_lock_interruptible(&dev->struct_mutex); + if (ret) + return ret; - mutex_lock(&dev->struct_mutex); seq_printf(m, "bit6 swizzle for X-tiling = %s\n", swizzle_string(dev_priv->mm.bit_6_swizzle_x)); seq_printf(m, "bit6 swizzle for Y-tiling = %s\n", @@ -1520,9 +1513,7 @@ static int i915_ppgtt_info(struct seq_file *m, void *data) if (INTEL_INFO(dev)->gen == 6) seq_printf(m, "GFX_MODE: 0x%08x\n", I915_READ(GFX_MODE)); - for (i = 0; i < I915_NUM_RINGS; i++) { - ring = &dev_priv->ring[i]; - + for_each_ring(ring, dev_priv, i) { seq_printf(m, "%s\n", ring->name); if (INTEL_INFO(dev)->gen == 7) seq_printf(m, "GFX_MODE: 0x%08x\n", I915_READ(RING_MODE_GEN7(ring))); @@ -1674,7 +1665,7 @@ i915_ring_stop_write(struct file *filp, struct drm_device *dev = filp->private_data; struct drm_i915_private *dev_priv = dev->dev_private; char buf[20]; - int val = 0; + int val = 0, ret; if (cnt > 0) { if (cnt > sizeof(buf) - 1) @@ -1689,7 +1680,10 @@ i915_ring_stop_write(struct file *filp, DRM_DEBUG_DRIVER("Stopping rings 0x%08x\n", val); - mutex_lock(&dev->struct_mutex); + ret = mutex_lock_interruptible(&dev->struct_mutex); + if (ret) + return ret; + dev_priv->stop_rings = val; mutex_unlock(&dev->struct_mutex); @@ -1713,10 +1707,18 @@ i915_max_freq_read(struct file *filp, struct drm_device *dev = filp->private_data; drm_i915_private_t *dev_priv = dev->dev_private; char buf[80]; - int len; + int len, ret; + + if (!(IS_GEN6(dev) || IS_GEN7(dev))) + return -ENODEV; + + ret = mutex_lock_interruptible(&dev->struct_mutex); + if (ret) + return ret; len = snprintf(buf, sizeof(buf), - "max freq: %d\n", dev_priv->max_delay * 50); + "max freq: %d\n", dev_priv->rps.max_delay * GT_FREQUENCY_MULTIPLIER); + mutex_unlock(&dev->struct_mutex); if (len > sizeof(buf)) len = sizeof(buf); @@ -1733,7 +1735,10 @@ i915_max_freq_write(struct file *filp, struct drm_device *dev = filp->private_data; struct drm_i915_private *dev_priv = dev->dev_private; char buf[20]; - int val = 1; + int val = 1, ret; + + if (!(IS_GEN6(dev) || IS_GEN7(dev))) + return -ENODEV; if (cnt > 0) { if (cnt > sizeof(buf) - 1) @@ -1748,12 +1753,17 @@ i915_max_freq_write(struct file *filp, DRM_DEBUG_DRIVER("Manually setting max freq to %d\n", val); + ret = mutex_lock_interruptible(&dev->struct_mutex); + if (ret) + return ret; + /* * Turbo will still be enabled, but won't go above the set value. */ - dev_priv->max_delay = val / 50; + dev_priv->rps.max_delay = val / GT_FREQUENCY_MULTIPLIER; - gen6_set_rps(dev, val / 50); + gen6_set_rps(dev, val / GT_FREQUENCY_MULTIPLIER); + mutex_unlock(&dev->struct_mutex); return cnt; } @@ -1773,10 +1783,18 @@ i915_min_freq_read(struct file *filp, char __user *ubuf, size_t max, struct drm_device *dev = filp->private_data; drm_i915_private_t *dev_priv = dev->dev_private; char buf[80]; - int len; + int len, ret; + + if (!(IS_GEN6(dev) || IS_GEN7(dev))) + return -ENODEV; + + ret = mutex_lock_interruptible(&dev->struct_mutex); + if (ret) + return ret; len = snprintf(buf, sizeof(buf), - "min freq: %d\n", dev_priv->min_delay * 50); + "min freq: %d\n", dev_priv->rps.min_delay * GT_FREQUENCY_MULTIPLIER); + mutex_unlock(&dev->struct_mutex); if (len > sizeof(buf)) len = sizeof(buf); @@ -1791,7 +1809,10 @@ i915_min_freq_write(struct file *filp, const char __user *ubuf, size_t cnt, struct drm_device *dev = filp->private_data; struct drm_i915_private *dev_priv = dev->dev_private; char buf[20]; - int val = 1; + int val = 1, ret; + + if (!(IS_GEN6(dev) || IS_GEN7(dev))) + return -ENODEV; if (cnt > 0) { if (cnt > sizeof(buf) - 1) @@ -1806,12 +1827,17 @@ i915_min_freq_write(struct file *filp, const char __user *ubuf, size_t cnt, DRM_DEBUG_DRIVER("Manually setting min freq to %d\n", val); + ret = mutex_lock_interruptible(&dev->struct_mutex); + if (ret) + return ret; + /* * Turbo will still be enabled, but won't go below the set value. */ - dev_priv->min_delay = val / 50; + dev_priv->rps.min_delay = val / GT_FREQUENCY_MULTIPLIER; - gen6_set_rps(dev, val / 50); + gen6_set_rps(dev, val / GT_FREQUENCY_MULTIPLIER); + mutex_unlock(&dev->struct_mutex); return cnt; } @@ -1834,9 +1860,15 @@ i915_cache_sharing_read(struct file *filp, drm_i915_private_t *dev_priv = dev->dev_private; char buf[80]; u32 snpcr; - int len; + int len, ret; + + if (!(IS_GEN6(dev) || IS_GEN7(dev))) + return -ENODEV; + + ret = mutex_lock_interruptible(&dev->struct_mutex); + if (ret) + return ret; - mutex_lock(&dev_priv->dev->struct_mutex); snpcr = I915_READ(GEN6_MBCUNIT_SNPCR); mutex_unlock(&dev_priv->dev->struct_mutex); @@ -1862,6 +1894,9 @@ i915_cache_sharing_write(struct file *filp, u32 snpcr; int val = 1; + if (!(IS_GEN6(dev) || IS_GEN7(dev))) + return -ENODEV; + if (cnt > 0) { if (cnt > sizeof(buf) - 1) return -EINVAL; @@ -1925,16 +1960,11 @@ static int i915_forcewake_open(struct inode *inode, struct file *file) { struct drm_device *dev = inode->i_private; struct drm_i915_private *dev_priv = dev->dev_private; - int ret; if (INTEL_INFO(dev)->gen < 6) return 0; - ret = mutex_lock_interruptible(&dev->struct_mutex); - if (ret) - return ret; gen6_gt_force_wake_get(dev_priv); - mutex_unlock(&dev->struct_mutex); return 0; } @@ -1947,16 +1977,7 @@ static int i915_forcewake_release(struct inode *inode, struct file *file) if (INTEL_INFO(dev)->gen < 6) return 0; - /* - * It's bad that we can potentially hang userspace if struct_mutex gets - * forever stuck. However, if we cannot acquire this lock it means that - * almost certainly the driver has hung, is not unload-able. Therefore - * hanging here is probably a minor inconvenience not to be seen my - * almost every user. - */ - mutex_lock(&dev->struct_mutex); gen6_gt_force_wake_put(dev_priv); - mutex_unlock(&dev->struct_mutex); return 0; } @@ -2006,7 +2027,6 @@ static struct drm_info_list i915_debugfs_list[] = { {"i915_gem_gtt", i915_gem_gtt_info, 0}, {"i915_gem_pinned", i915_gem_gtt_info, 0, (void *) PINNED_LIST}, {"i915_gem_active", i915_gem_object_list_info, 0, (void *) ACTIVE_LIST}, - {"i915_gem_flushing", i915_gem_object_list_info, 0, (void *) FLUSHING_LIST}, {"i915_gem_inactive", i915_gem_object_list_info, 0, (void *) INACTIVE_LIST}, {"i915_gem_pageflip", i915_gem_pageflip_info, 0}, {"i915_gem_request", i915_gem_request_info, 0}, @@ -2067,6 +2087,7 @@ int i915_debugfs_init(struct drm_minor *minor) &i915_cache_sharing_fops); if (ret) return ret; + ret = i915_debugfs_create(minor->debugfs_root, minor, "i915_ring_stop", &i915_ring_stop_fops); diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c index 914c0dfabe60..ffbc9156c792 100644 --- a/drivers/gpu/drm/i915/i915_dma.c +++ b/drivers/gpu/drm/i915/i915_dma.c @@ -235,10 +235,10 @@ static int i915_initialize(struct drm_device * dev, drm_i915_init_t * init) } } - dev_priv->cpp = init->cpp; - dev_priv->back_offset = init->back_offset; - dev_priv->front_offset = init->front_offset; - dev_priv->current_page = 0; + dev_priv->dri1.cpp = init->cpp; + dev_priv->dri1.back_offset = init->back_offset; + dev_priv->dri1.front_offset = init->front_offset; + dev_priv->dri1.current_page = 0; if (master_priv->sarea_priv) master_priv->sarea_priv->pf_current_page = 0; @@ -575,7 +575,7 @@ static int i915_dispatch_flip(struct drm_device * dev) DRM_DEBUG_DRIVER("%s: page=%d pfCurrentPage=%d\n", __func__, - dev_priv->current_page, + dev_priv->dri1.current_page, master_priv->sarea_priv->pf_current_page); i915_kernel_lost_context(dev); @@ -589,12 +589,12 @@ static int i915_dispatch_flip(struct drm_device * dev) OUT_RING(CMD_OP_DISPLAYBUFFER_INFO | ASYNC_FLIP); OUT_RING(0); - if (dev_priv->current_page == 0) { - OUT_RING(dev_priv->back_offset); - dev_priv->current_page = 1; + if (dev_priv->dri1.current_page == 0) { + OUT_RING(dev_priv->dri1.back_offset); + dev_priv->dri1.current_page = 1; } else { - OUT_RING(dev_priv->front_offset); - dev_priv->current_page = 0; + OUT_RING(dev_priv->dri1.front_offset); + dev_priv->dri1.current_page = 0; } OUT_RING(0); @@ -613,7 +613,7 @@ static int i915_dispatch_flip(struct drm_device * dev) ADVANCE_LP_RING(); } - master_priv->sarea_priv->pf_current_page = dev_priv->current_page; + master_priv->sarea_priv->pf_current_page = dev_priv->dri1.current_page; return 0; } @@ -1009,6 +1009,12 @@ static int i915_getparam(struct drm_device *dev, void *data, case I915_PARAM_HAS_WAIT_TIMEOUT: value = 1; break; + case I915_PARAM_HAS_SEMAPHORES: + value = i915_semaphore_is_enabled(dev); + break; + case I915_PARAM_HAS_PRIME_VMAP_FLUSH: + value = 1; + break; default: DRM_DEBUG_DRIVER("Unknown parameter %d\n", param->param); @@ -1425,6 +1431,21 @@ static void i915_kick_out_firmware_fb(struct drm_i915_private *dev_priv) kfree(ap); } +static void i915_dump_device_info(struct drm_i915_private *dev_priv) +{ + const struct intel_device_info *info = dev_priv->info; + +#define DEV_INFO_FLAG(name) info->name ? #name "," : "" +#define DEV_INFO_SEP , + DRM_DEBUG_DRIVER("i915 device info: gen=%i, pciid=0x%04x flags=" + "%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s%s", + info->gen, + dev_priv->dev->pdev->device, + DEV_INFO_FLAGS); +#undef DEV_INFO_FLAG +#undef DEV_INFO_SEP +} + /** * i915_driver_load - setup chip and create an initial config * @dev: DRM device @@ -1440,7 +1461,7 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags) { struct drm_i915_private *dev_priv; struct intel_device_info *info; - int ret = 0, mmio_bar; + int ret = 0, mmio_bar, mmio_size; uint32_t aperture_size; info = (struct intel_device_info *) flags; @@ -1449,7 +1470,6 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags) if (info->gen >= 6 && !drm_core_check_feature(dev, DRIVER_MODESET)) return -ENODEV; - /* i915 has 4 more counters */ dev->counters += 4; dev->types[6] = _DRM_STAT_IRQ; @@ -1465,6 +1485,8 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags) dev_priv->dev = dev; dev_priv->info = info; + i915_dump_device_info(dev_priv); + if (i915_get_bridge_dev(dev)) { ret = -EIO; goto free_priv; @@ -1504,7 +1526,19 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags) dma_set_coherent_mask(&dev->pdev->dev, DMA_BIT_MASK(32)); mmio_bar = IS_GEN2(dev) ? 1 : 0; - dev_priv->regs = pci_iomap(dev->pdev, mmio_bar, 0); + /* Before gen4, the registers and the GTT are behind different BARs. + * However, from gen4 onwards, the registers and the GTT are shared + * in the same BAR, so we want to restrict this ioremap from + * clobbering the GTT which we want ioremap_wc instead. Fortunately, + * the register BAR remains the same size for all the earlier + * generations up to Ironlake. + */ + if (info->gen < 5) + mmio_size = 512*1024; + else + mmio_size = 2*1024*1024; + + dev_priv->regs = pci_iomap(dev->pdev, mmio_bar, mmio_size); if (!dev_priv->regs) { DRM_ERROR("failed to map registers\n"); ret = -EIO; @@ -1536,11 +1570,9 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags) * * All tasks on the workqueue are expected to acquire the dev mutex * so there is no point in running more than one instance of the - * workqueue at any time: max_active = 1 and NON_REENTRANT. + * workqueue at any time. Use an ordered one. */ - dev_priv->wq = alloc_workqueue("i915", - WQ_UNBOUND | WQ_NON_REENTRANT, - 1); + dev_priv->wq = alloc_ordered_workqueue("i915", 0); if (dev_priv->wq == NULL) { DRM_ERROR("Failed to create our workqueue.\n"); ret = -ENOMEM; @@ -1586,7 +1618,7 @@ int i915_driver_load(struct drm_device *dev, unsigned long flags) spin_lock_init(&dev_priv->irq_lock); spin_lock_init(&dev_priv->error_lock); - spin_lock_init(&dev_priv->rps_lock); + spin_lock_init(&dev_priv->rps.lock); spin_lock_init(&dev_priv->dpio_lock); if (IS_IVYBRIDGE(dev) || IS_HASWELL(dev)) @@ -1836,6 +1868,8 @@ struct drm_ioctl_desc i915_ioctls[] = { DRM_IOCTL_DEF_DRV(I915_GEM_PIN, i915_gem_pin_ioctl, DRM_AUTH|DRM_ROOT_ONLY|DRM_UNLOCKED), DRM_IOCTL_DEF_DRV(I915_GEM_UNPIN, i915_gem_unpin_ioctl, DRM_AUTH|DRM_ROOT_ONLY|DRM_UNLOCKED), DRM_IOCTL_DEF_DRV(I915_GEM_BUSY, i915_gem_busy_ioctl, DRM_AUTH|DRM_UNLOCKED), + DRM_IOCTL_DEF_DRV(I915_GEM_SET_CACHING, i915_gem_set_caching_ioctl, DRM_UNLOCKED), + DRM_IOCTL_DEF_DRV(I915_GEM_GET_CACHING, i915_gem_get_caching_ioctl, DRM_UNLOCKED), DRM_IOCTL_DEF_DRV(I915_GEM_THROTTLE, i915_gem_throttle_ioctl, DRM_AUTH|DRM_UNLOCKED), DRM_IOCTL_DEF_DRV(I915_GEM_ENTERVT, i915_gem_entervt_ioctl, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY|DRM_UNLOCKED), DRM_IOCTL_DEF_DRV(I915_GEM_LEAVEVT, i915_gem_leavevt_ioctl, DRM_AUTH|DRM_MASTER|DRM_ROOT_ONLY|DRM_UNLOCKED), @@ -1858,6 +1892,7 @@ struct drm_ioctl_desc i915_ioctls[] = { DRM_IOCTL_DEF_DRV(I915_GEM_WAIT, i915_gem_wait_ioctl, DRM_AUTH|DRM_UNLOCKED), DRM_IOCTL_DEF_DRV(I915_GEM_CONTEXT_CREATE, i915_gem_context_create_ioctl, DRM_UNLOCKED), DRM_IOCTL_DEF_DRV(I915_GEM_CONTEXT_DESTROY, i915_gem_context_destroy_ioctl, DRM_UNLOCKED), + DRM_IOCTL_DEF_DRV(I915_REG_READ, i915_reg_read_ioctl, DRM_UNLOCKED), }; int i915_max_ioctl = DRM_ARRAY_SIZE(i915_ioctls); diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c index a24ffbe97c01..a7837e556945 100644 --- a/drivers/gpu/drm/i915/i915_drv.c +++ b/drivers/gpu/drm/i915/i915_drv.c @@ -470,6 +470,9 @@ static int i915_drm_freeze(struct drm_device *dev) "GEM idle failed, resume might fail\n"); return error; } + + intel_modeset_disable(dev); + drm_irq_uninstall(dev); } @@ -543,13 +546,9 @@ static int i915_drm_thaw(struct drm_device *dev) mutex_unlock(&dev->struct_mutex); intel_modeset_init_hw(dev); + intel_modeset_setup_hw_state(dev); drm_mode_config_reset(dev); drm_irq_install(dev); - - /* Resume the modeset for every activated CRTC */ - mutex_lock(&dev->mode_config.mutex); - drm_helper_resume_force_mode(dev); - mutex_unlock(&dev->mode_config.mutex); } intel_opregion_init(dev); @@ -1060,7 +1059,7 @@ static bool IS_DISPLAYREG(u32 reg) * This should make it easier to transition modules over to the * new register block scheme, since we can do it incrementally. */ - if (reg >= 0x180000) + if (reg >= VLV_DISPLAY_BASE) return false; if (reg >= RENDER_RING_BASE && @@ -1174,9 +1173,59 @@ void i915_write##x(struct drm_i915_private *dev_priv, u32 reg, u##x val) { \ if (unlikely(__fifo_ret)) { \ gen6_gt_check_fifodbg(dev_priv); \ } \ + if (IS_HASWELL(dev_priv->dev) && (I915_READ_NOTRACE(GEN7_ERR_INT) & ERR_INT_MMIO_UNCLAIMED)) { \ + DRM_ERROR("Unclaimed write to %x\n", reg); \ + writel(ERR_INT_MMIO_UNCLAIMED, dev_priv->regs + GEN7_ERR_INT); \ + } \ } __i915_write(8, b) __i915_write(16, w) __i915_write(32, l) __i915_write(64, q) #undef __i915_write + +static const struct register_whitelist { + uint64_t offset; + uint32_t size; + uint32_t gen_bitmask; /* support gens, 0x10 for 4, 0x30 for 4 and 5, etc. */ +} whitelist[] = { + { RING_TIMESTAMP(RENDER_RING_BASE), 8, 0xF0 }, +}; + +int i915_reg_read_ioctl(struct drm_device *dev, + void *data, struct drm_file *file) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + struct drm_i915_reg_read *reg = data; + struct register_whitelist const *entry = whitelist; + int i; + + for (i = 0; i < ARRAY_SIZE(whitelist); i++, entry++) { + if (entry->offset == reg->offset && + (1 << INTEL_INFO(dev)->gen & entry->gen_bitmask)) + break; + } + + if (i == ARRAY_SIZE(whitelist)) + return -EINVAL; + + switch (entry->size) { + case 8: + reg->val = I915_READ64(reg->offset); + break; + case 4: + reg->val = I915_READ(reg->offset); + break; + case 2: + reg->val = I915_READ16(reg->offset); + break; + case 1: + reg->val = I915_READ8(reg->offset); + break; + default: + WARN_ON(1); + return -EINVAL; + } + + return 0; +} diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index 627fe35781b4..4f2831aa5fed 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -109,6 +109,7 @@ struct intel_pch_pll { #define WATCH_COHERENCY 0 #define WATCH_LISTS 0 +#define WATCH_GTT 0 #define I915_GEM_PHYS_CURSOR_0 1 #define I915_GEM_PHYS_CURSOR_1 2 @@ -195,9 +196,10 @@ struct drm_i915_error_state { u32 cpu_ring_head[I915_NUM_RINGS]; u32 cpu_ring_tail[I915_NUM_RINGS]; u32 error; /* gen6+ */ + u32 err_int; /* gen7 */ u32 instpm[I915_NUM_RINGS]; u32 instps[I915_NUM_RINGS]; - u32 instdone1; + u32 extra_instdone[I915_NUM_INSTDONE_REG]; u32 seqno[I915_NUM_RINGS]; u64 bbaddr; u32 fault_reg[I915_NUM_RINGS]; @@ -221,7 +223,7 @@ struct drm_i915_error_state { struct drm_i915_error_buffer { u32 size; u32 name; - u32 seqno; + u32 rseqno, wseqno; u32 gtt_offset; u32 read_domains; u32 write_domain; @@ -239,7 +241,6 @@ struct drm_i915_error_state { }; struct drm_i915_display_funcs { - void (*dpms)(struct drm_crtc *crtc, int mode); bool (*fbc_enabled)(struct drm_device *dev); void (*enable_fbc)(struct drm_crtc *crtc, unsigned long interval); void (*disable_fbc)(struct drm_device *dev); @@ -248,7 +249,6 @@ struct drm_i915_display_funcs { void (*update_wm)(struct drm_device *dev); void (*update_sprite_wm)(struct drm_device *dev, int pipe, uint32_t sprite_width, int pixel_size); - void (*sanitize_pm)(struct drm_device *dev); void (*update_linetime_wm)(struct drm_device *dev, int pipe, struct drm_display_mode *mode); int (*crtc_mode_set)(struct drm_crtc *crtc, @@ -256,6 +256,8 @@ struct drm_i915_display_funcs { struct drm_display_mode *adjusted_mode, int x, int y, struct drm_framebuffer *old_fb); + void (*crtc_enable)(struct drm_crtc *crtc); + void (*crtc_disable)(struct drm_crtc *crtc); void (*off)(struct drm_crtc *crtc); void (*write_eld)(struct drm_connector *connector, struct drm_crtc *crtc); @@ -279,6 +281,32 @@ struct drm_i915_gt_funcs { void (*force_wake_put)(struct drm_i915_private *dev_priv); }; +#define DEV_INFO_FLAGS \ + DEV_INFO_FLAG(is_mobile) DEV_INFO_SEP \ + DEV_INFO_FLAG(is_i85x) DEV_INFO_SEP \ + DEV_INFO_FLAG(is_i915g) DEV_INFO_SEP \ + DEV_INFO_FLAG(is_i945gm) DEV_INFO_SEP \ + DEV_INFO_FLAG(is_g33) DEV_INFO_SEP \ + DEV_INFO_FLAG(need_gfx_hws) DEV_INFO_SEP \ + DEV_INFO_FLAG(is_g4x) DEV_INFO_SEP \ + DEV_INFO_FLAG(is_pineview) DEV_INFO_SEP \ + DEV_INFO_FLAG(is_broadwater) DEV_INFO_SEP \ + DEV_INFO_FLAG(is_crestline) DEV_INFO_SEP \ + DEV_INFO_FLAG(is_ivybridge) DEV_INFO_SEP \ + DEV_INFO_FLAG(is_valleyview) DEV_INFO_SEP \ + DEV_INFO_FLAG(is_haswell) DEV_INFO_SEP \ + DEV_INFO_FLAG(has_force_wake) DEV_INFO_SEP \ + DEV_INFO_FLAG(has_fbc) DEV_INFO_SEP \ + DEV_INFO_FLAG(has_pipe_cxsr) DEV_INFO_SEP \ + DEV_INFO_FLAG(has_hotplug) DEV_INFO_SEP \ + DEV_INFO_FLAG(cursor_needs_physical) DEV_INFO_SEP \ + DEV_INFO_FLAG(has_overlay) DEV_INFO_SEP \ + DEV_INFO_FLAG(overlay_needs_physical) DEV_INFO_SEP \ + DEV_INFO_FLAG(supports_tv) DEV_INFO_SEP \ + DEV_INFO_FLAG(has_bsd_ring) DEV_INFO_SEP \ + DEV_INFO_FLAG(has_blt_ring) DEV_INFO_SEP \ + DEV_INFO_FLAG(has_llc) + struct intel_device_info { u8 gen; u8 is_mobile:1; @@ -402,12 +430,6 @@ typedef struct drm_i915_private { struct resource mch_res; - unsigned int cpp; - int back_offset; - int front_offset; - int current_page; - int page_flipping; - atomic_t irq_received; /* protects the irq masks */ @@ -425,7 +447,6 @@ typedef struct drm_i915_private { u32 hotplug_supported_mask; struct work_struct hotplug_work; - unsigned int sr01, adpa, ppcr, dvob, dvoc, lvds; int num_pipe; int num_pch_pll; @@ -434,8 +455,7 @@ typedef struct drm_i915_private { struct timer_list hangcheck_timer; int hangcheck_count; uint32_t last_acthd[I915_NUM_RINGS]; - uint32_t last_instdone; - uint32_t last_instdone1; + uint32_t prev_instdone[I915_NUM_INSTDONE_REG]; unsigned int stop_rings; @@ -666,7 +686,13 @@ typedef struct drm_i915_private { struct drm_mm gtt_space; /** List of all objects in gtt_space. Used to restore gtt * mappings on resume */ - struct list_head gtt_list; + struct list_head bound_list; + /** + * List of objects which are not bound to the GTT (thus + * are idle and not used by the GPU) but still have + * (presumably uncached) pages still attached. + */ + struct list_head unbound_list; /** Usable portion of the GTT for GEM */ unsigned long gtt_start; @@ -696,17 +722,6 @@ typedef struct drm_i915_private { struct list_head active_list; /** - * List of objects which are not in the ringbuffer but which - * still have a write_domain which needs to be flushed before - * unbinding. - * - * last_rendering_seqno is 0 while an object is in this list. - * - * A reference is held on the buffer while on this list. - */ - struct list_head flushing_list; - - /** * LRU list of objects which are not in the ringbuffer and * are ready to unbind, but are still in the GTT. * @@ -775,6 +790,12 @@ typedef struct drm_i915_private { struct { unsigned allow_batchbuffer : 1; u32 __iomem *gfx_hws_cpu_addr; + + unsigned int cpp; + int back_offset; + int front_offset; + int current_page; + int page_flipping; } dri1; /* Kernel Modesetting */ @@ -796,9 +817,6 @@ typedef struct drm_i915_private { bool lvds_downclock_avail; /* indicates the reduced downclock for LVDS*/ int lvds_downclock; - struct work_struct idle_work; - struct timer_list idle_timer; - bool busy; u16 orig_clock; int child_dev_num; struct child_device_config *child_dev; @@ -807,26 +825,41 @@ typedef struct drm_i915_private { bool mchbar_need_disable; - struct work_struct rps_work; - spinlock_t rps_lock; - u32 pm_iir; - - u8 cur_delay; - u8 min_delay; - u8 max_delay; - u8 fmax; - u8 fstart; - - u64 last_count1; - unsigned long last_time1; - unsigned long chipset_power; - u64 last_count2; - struct timespec last_time2; - unsigned long gfx_power; - int c_m; - int r_t; - u8 corr; - spinlock_t *mchdev_lock; + /* gen6+ rps state */ + struct { + struct work_struct work; + u32 pm_iir; + /* lock - irqsave spinlock that protectects the work_struct and + * pm_iir. */ + spinlock_t lock; + + /* The below variables an all the rps hw state are protected by + * dev->struct mutext. */ + u8 cur_delay; + u8 min_delay; + u8 max_delay; + } rps; + + /* ilk-only ips/rps state. Everything in here is protected by the global + * mchdev_lock in intel_pm.c */ + struct { + u8 cur_delay; + u8 min_delay; + u8 max_delay; + u8 fmax; + u8 fstart; + + u64 last_count1; + unsigned long last_time1; + unsigned long chipset_power; + u64 last_count2; + struct timespec last_time2; + unsigned long gfx_power; + u8 corr; + + int c_m; + int r_t; + } ips; enum no_fbc_reason no_fbc_reason; @@ -861,30 +894,48 @@ enum hdmi_force_audio { }; enum i915_cache_level { - I915_CACHE_NONE, + I915_CACHE_NONE = 0, I915_CACHE_LLC, - I915_CACHE_LLC_MLC, /* gen6+ */ + I915_CACHE_LLC_MLC, /* gen6+, in docs at least! */ +}; + +struct drm_i915_gem_object_ops { + /* Interface between the GEM object and its backing storage. + * get_pages() is called once prior to the use of the associated set + * of pages before to binding them into the GTT, and put_pages() is + * called after we no longer need them. As we expect there to be + * associated cost with migrating pages between the backing storage + * and making them available for the GPU (e.g. clflush), we may hold + * onto the pages after they are no longer referenced by the GPU + * in case they may be used again shortly (for example migrating the + * pages to a different memory domain within the GTT). put_pages() + * will therefore most likely be called when the object itself is + * being released or under memory pressure (where we attempt to + * reap pages for the shrinker). + */ + int (*get_pages)(struct drm_i915_gem_object *); + void (*put_pages)(struct drm_i915_gem_object *); }; struct drm_i915_gem_object { struct drm_gem_object base; + const struct drm_i915_gem_object_ops *ops; + /** Current space allocated to this object in the GTT, if any. */ struct drm_mm_node *gtt_space; struct list_head gtt_list; - /** This object's place on the active/flushing/inactive lists */ + /** This object's place on the active/inactive lists */ struct list_head ring_list; struct list_head mm_list; - /** This object's place on GPU write list */ - struct list_head gpu_write_list; /** This object's place in the batchbuffer or on the eviction list */ struct list_head exec_list; /** - * This is set if the object is on the active or flushing lists - * (has pending rendering), and is not set if it's on inactive (ready - * to be unbound). + * This is set if the object is on the active lists (has pending + * rendering and so a non-zero seqno), and is not set if it i s on + * inactive (ready to be unbound) list. */ unsigned int active:1; @@ -895,12 +946,6 @@ struct drm_i915_gem_object { unsigned int dirty:1; /** - * This is set if the object has been written to since the last - * GPU flush. - */ - unsigned int pending_gpu_write:1; - - /** * Fence register bits (if any) for this object. Will be set * as needed when mapped into the GTT. * Protected by dev->struct_mutex. @@ -961,17 +1006,12 @@ struct drm_i915_gem_object { unsigned int has_aliasing_ppgtt_mapping:1; unsigned int has_global_gtt_mapping:1; + unsigned int has_dma_mapping:1; - struct page **pages; - - /** - * DMAR support - */ - struct scatterlist *sg_list; - int num_sg; + struct sg_table *pages; + int pages_pin_count; /* prime dma-buf support */ - struct sg_table *sg_table; void *dma_buf_vmapping; int vmapping_count; @@ -992,7 +1032,8 @@ struct drm_i915_gem_object { struct intel_ring_buffer *ring; /** Breadcrumb of last rendering to the buffer. */ - uint32_t last_rendering_seqno; + uint32_t last_read_seqno; + uint32_t last_write_seqno; /** Breadcrumb of last fenced GPU access to the buffer. */ uint32_t last_fenced_seqno; @@ -1135,6 +1176,10 @@ struct drm_i915_file_private { #define HAS_FORCE_WAKE(dev) (INTEL_INFO(dev)->has_force_wake) +#define HAS_L3_GPU_CACHE(dev) (IS_IVYBRIDGE(dev) || IS_HASWELL(dev)) + +#define GT_FREQUENCY_MULTIPLIER 50 + #include "i915_trace.h" /** @@ -1256,6 +1301,10 @@ int i915_gem_unpin_ioctl(struct drm_device *dev, void *data, struct drm_file *file_priv); int i915_gem_busy_ioctl(struct drm_device *dev, void *data, struct drm_file *file_priv); +int i915_gem_get_caching_ioctl(struct drm_device *dev, void *data, + struct drm_file *file); +int i915_gem_set_caching_ioctl(struct drm_device *dev, void *data, + struct drm_file *file); int i915_gem_throttle_ioctl(struct drm_device *dev, void *data, struct drm_file *file_priv); int i915_gem_madvise_ioctl(struct drm_device *dev, void *data, @@ -1274,24 +1323,42 @@ int i915_gem_wait_ioctl(struct drm_device *dev, void *data, struct drm_file *file_priv); void i915_gem_load(struct drm_device *dev); int i915_gem_init_object(struct drm_gem_object *obj); -int __must_check i915_gem_flush_ring(struct intel_ring_buffer *ring, - uint32_t invalidate_domains, - uint32_t flush_domains); +void i915_gem_object_init(struct drm_i915_gem_object *obj, + const struct drm_i915_gem_object_ops *ops); struct drm_i915_gem_object *i915_gem_alloc_object(struct drm_device *dev, size_t size); void i915_gem_free_object(struct drm_gem_object *obj); int __must_check i915_gem_object_pin(struct drm_i915_gem_object *obj, uint32_t alignment, - bool map_and_fenceable); + bool map_and_fenceable, + bool nonblocking); void i915_gem_object_unpin(struct drm_i915_gem_object *obj); int __must_check i915_gem_object_unbind(struct drm_i915_gem_object *obj); void i915_gem_release_mmap(struct drm_i915_gem_object *obj); void i915_gem_lastclose(struct drm_device *dev); -int i915_gem_object_get_pages_gtt(struct drm_i915_gem_object *obj, - gfp_t gfpmask); +int __must_check i915_gem_object_get_pages(struct drm_i915_gem_object *obj); +static inline struct page *i915_gem_object_get_page(struct drm_i915_gem_object *obj, int n) +{ + struct scatterlist *sg = obj->pages->sgl; + while (n >= SG_MAX_SINGLE_ALLOC) { + sg = sg_chain_ptr(sg + SG_MAX_SINGLE_ALLOC - 1); + n -= SG_MAX_SINGLE_ALLOC - 1; + } + return sg_page(sg+n); +} +static inline void i915_gem_object_pin_pages(struct drm_i915_gem_object *obj) +{ + BUG_ON(obj->pages == NULL); + obj->pages_pin_count++; +} +static inline void i915_gem_object_unpin_pages(struct drm_i915_gem_object *obj) +{ + BUG_ON(obj->pages_pin_count == 0); + obj->pages_pin_count--; +} + int __must_check i915_mutex_lock_interruptible(struct drm_device *dev); -int __must_check i915_gem_object_wait_rendering(struct drm_i915_gem_object *obj); int i915_gem_object_sync(struct drm_i915_gem_object *obj, struct intel_ring_buffer *to); void i915_gem_object_move_to_active(struct drm_i915_gem_object *obj, @@ -1358,9 +1425,9 @@ void i915_gem_init_ppgtt(struct drm_device *dev); void i915_gem_cleanup_ringbuffer(struct drm_device *dev); int __must_check i915_gpu_idle(struct drm_device *dev); int __must_check i915_gem_idle(struct drm_device *dev); -int __must_check i915_add_request(struct intel_ring_buffer *ring, - struct drm_file *file, - struct drm_i915_gem_request *request); +int i915_add_request(struct intel_ring_buffer *ring, + struct drm_file *file, + struct drm_i915_gem_request *request); int __must_check i915_wait_seqno(struct intel_ring_buffer *ring, uint32_t seqno); int i915_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf); @@ -1429,8 +1496,11 @@ void i915_gem_init_global_gtt(struct drm_device *dev, /* i915_gem_evict.c */ int __must_check i915_gem_evict_something(struct drm_device *dev, int min_size, - unsigned alignment, bool mappable); -int i915_gem_evict_everything(struct drm_device *dev, bool purgeable_only); + unsigned alignment, + unsigned cache_level, + bool mappable, + bool nonblock); +int i915_gem_evict_everything(struct drm_device *dev); /* i915_gem_stolen.c */ int i915_gem_init_stolen(struct drm_device *dev); @@ -1519,6 +1589,7 @@ extern void intel_modeset_init(struct drm_device *dev); extern void intel_modeset_gem_init(struct drm_device *dev); extern void intel_modeset_cleanup(struct drm_device *dev); extern int intel_modeset_vga_set_state(struct drm_device *dev, bool state); +extern void intel_modeset_setup_hw_state(struct drm_device *dev); extern bool intel_fbc_enabled(struct drm_device *dev); extern void intel_disable_fbc(struct drm_device *dev); extern bool ironlake_set_drps(struct drm_device *dev, u8 val); @@ -1529,6 +1600,8 @@ extern int intel_trans_dp_port_sel(struct drm_crtc *crtc); extern int intel_enable_rc6(const struct drm_device *dev); extern bool i915_semaphore_is_enabled(struct drm_device *dev); +int i915_reg_read_ioctl(struct drm_device *dev, void *data, + struct drm_file *file); /* overlay */ #ifdef CONFIG_DEBUG_FS diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index 274d25de521e..365a7dc8a4a8 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -37,12 +37,12 @@ #include <linux/pci.h> #include <linux/dma-buf.h> -static __must_check int i915_gem_object_flush_gpu_write_domain(struct drm_i915_gem_object *obj); static void i915_gem_object_flush_gtt_write_domain(struct drm_i915_gem_object *obj); static void i915_gem_object_flush_cpu_write_domain(struct drm_i915_gem_object *obj); static __must_check int i915_gem_object_bind_to_gtt(struct drm_i915_gem_object *obj, unsigned alignment, - bool map_and_fenceable); + bool map_and_fenceable, + bool nonblocking); static int i915_gem_phys_pwrite(struct drm_device *dev, struct drm_i915_gem_object *obj, struct drm_i915_gem_pwrite *args, @@ -56,6 +56,8 @@ static void i915_gem_object_update_fence(struct drm_i915_gem_object *obj, static int i915_gem_inactive_shrink(struct shrinker *shrinker, struct shrink_control *sc); +static long i915_gem_purge(struct drm_i915_private *dev_priv, long target); +static void i915_gem_shrink_all(struct drm_i915_private *dev_priv); static void i915_gem_object_truncate(struct drm_i915_gem_object *obj); static inline void i915_gem_object_fence_lost(struct drm_i915_gem_object *obj) @@ -141,7 +143,7 @@ int i915_mutex_lock_interruptible(struct drm_device *dev) static inline bool i915_gem_object_is_inactive(struct drm_i915_gem_object *obj) { - return !obj->active; + return obj->gtt_space && !obj->active; } int @@ -180,7 +182,7 @@ i915_gem_get_aperture_ioctl(struct drm_device *dev, void *data, pinned = 0; mutex_lock(&dev->struct_mutex); - list_for_each_entry(obj, &dev_priv->mm.gtt_list, gtt_list) + list_for_each_entry(obj, &dev_priv->mm.bound_list, gtt_list) if (obj->pin_count) pinned += obj->gtt_space->size; mutex_unlock(&dev->struct_mutex); @@ -341,7 +343,7 @@ shmem_pread_fast(struct page *page, int shmem_page_offset, int page_length, page_length); kunmap_atomic(vaddr); - return ret; + return ret ? -EFAULT : 0; } static void @@ -392,7 +394,7 @@ shmem_pread_slow(struct page *page, int shmem_page_offset, int page_length, page_length); kunmap(page); - return ret; + return ret ? - EFAULT : 0; } static int @@ -401,7 +403,6 @@ i915_gem_shmem_pread(struct drm_device *dev, struct drm_i915_gem_pread *args, struct drm_file *file) { - struct address_space *mapping = obj->base.filp->f_path.dentry->d_inode->i_mapping; char __user *user_data; ssize_t remain; loff_t offset; @@ -410,7 +411,8 @@ i915_gem_shmem_pread(struct drm_device *dev, int hit_slowpath = 0; int prefaulted = 0; int needs_clflush = 0; - int release_page; + struct scatterlist *sg; + int i; user_data = (char __user *) (uintptr_t) args->data_ptr; remain = args->size; @@ -424,16 +426,30 @@ i915_gem_shmem_pread(struct drm_device *dev, * anyway again before the next pread happens. */ if (obj->cache_level == I915_CACHE_NONE) needs_clflush = 1; - ret = i915_gem_object_set_to_gtt_domain(obj, false); - if (ret) - return ret; + if (obj->gtt_space) { + ret = i915_gem_object_set_to_gtt_domain(obj, false); + if (ret) + return ret; + } } + ret = i915_gem_object_get_pages(obj); + if (ret) + return ret; + + i915_gem_object_pin_pages(obj); + offset = args->offset; - while (remain > 0) { + for_each_sg(obj->pages->sgl, sg, obj->pages->nents, i) { struct page *page; + if (i < offset >> PAGE_SHIFT) + continue; + + if (remain <= 0) + break; + /* Operation in this page * * shmem_page_offset = offset within page in shmem file @@ -444,18 +460,7 @@ i915_gem_shmem_pread(struct drm_device *dev, if ((shmem_page_offset + page_length) > PAGE_SIZE) page_length = PAGE_SIZE - shmem_page_offset; - if (obj->pages) { - page = obj->pages[offset >> PAGE_SHIFT]; - release_page = 0; - } else { - page = shmem_read_mapping_page(mapping, offset >> PAGE_SHIFT); - if (IS_ERR(page)) { - ret = PTR_ERR(page); - goto out; - } - release_page = 1; - } - + page = sg_page(sg); page_do_bit17_swizzling = obj_do_bit17_swizzling && (page_to_phys(page) & (1 << 17)) != 0; @@ -466,7 +471,6 @@ i915_gem_shmem_pread(struct drm_device *dev, goto next_page; hit_slowpath = 1; - page_cache_get(page); mutex_unlock(&dev->struct_mutex); if (!prefaulted) { @@ -484,16 +488,12 @@ i915_gem_shmem_pread(struct drm_device *dev, needs_clflush); mutex_lock(&dev->struct_mutex); - page_cache_release(page); + next_page: mark_page_accessed(page); - if (release_page) - page_cache_release(page); - if (ret) { - ret = -EFAULT; + if (ret) goto out; - } remain -= page_length; user_data += page_length; @@ -501,6 +501,8 @@ next_page: } out: + i915_gem_object_unpin_pages(obj); + if (hit_slowpath) { /* Fixup: Kill any reinstated backing storage pages */ if (obj->madv == __I915_MADV_PURGED) @@ -606,7 +608,7 @@ i915_gem_gtt_pwrite_fast(struct drm_device *dev, char __user *user_data; int page_offset, page_length, ret; - ret = i915_gem_object_pin(obj, 0, true); + ret = i915_gem_object_pin(obj, 0, true, true); if (ret) goto out; @@ -686,7 +688,7 @@ shmem_pwrite_fast(struct page *page, int shmem_page_offset, int page_length, page_length); kunmap_atomic(vaddr); - return ret; + return ret ? -EFAULT : 0; } /* Only difference to the fast-path function is that this can handle bit17 @@ -720,7 +722,7 @@ shmem_pwrite_slow(struct page *page, int shmem_page_offset, int page_length, page_do_bit17_swizzling); kunmap(page); - return ret; + return ret ? -EFAULT : 0; } static int @@ -729,7 +731,6 @@ i915_gem_shmem_pwrite(struct drm_device *dev, struct drm_i915_gem_pwrite *args, struct drm_file *file) { - struct address_space *mapping = obj->base.filp->f_path.dentry->d_inode->i_mapping; ssize_t remain; loff_t offset; char __user *user_data; @@ -738,7 +739,8 @@ i915_gem_shmem_pwrite(struct drm_device *dev, int hit_slowpath = 0; int needs_clflush_after = 0; int needs_clflush_before = 0; - int release_page; + int i; + struct scatterlist *sg; user_data = (char __user *) (uintptr_t) args->data_ptr; remain = args->size; @@ -752,9 +754,11 @@ i915_gem_shmem_pwrite(struct drm_device *dev, * right away and we therefore have to clflush anyway. */ if (obj->cache_level == I915_CACHE_NONE) needs_clflush_after = 1; - ret = i915_gem_object_set_to_gtt_domain(obj, true); - if (ret) - return ret; + if (obj->gtt_space) { + ret = i915_gem_object_set_to_gtt_domain(obj, true); + if (ret) + return ret; + } } /* Same trick applies for invalidate partially written cachelines before * writing. */ @@ -762,13 +766,25 @@ i915_gem_shmem_pwrite(struct drm_device *dev, && obj->cache_level == I915_CACHE_NONE) needs_clflush_before = 1; + ret = i915_gem_object_get_pages(obj); + if (ret) + return ret; + + i915_gem_object_pin_pages(obj); + offset = args->offset; obj->dirty = 1; - while (remain > 0) { + for_each_sg(obj->pages->sgl, sg, obj->pages->nents, i) { struct page *page; int partial_cacheline_write; + if (i < offset >> PAGE_SHIFT) + continue; + + if (remain <= 0) + break; + /* Operation in this page * * shmem_page_offset = offset within page in shmem file @@ -787,18 +803,7 @@ i915_gem_shmem_pwrite(struct drm_device *dev, ((shmem_page_offset | page_length) & (boot_cpu_data.x86_clflush_size - 1)); - if (obj->pages) { - page = obj->pages[offset >> PAGE_SHIFT]; - release_page = 0; - } else { - page = shmem_read_mapping_page(mapping, offset >> PAGE_SHIFT); - if (IS_ERR(page)) { - ret = PTR_ERR(page); - goto out; - } - release_page = 1; - } - + page = sg_page(sg); page_do_bit17_swizzling = obj_do_bit17_swizzling && (page_to_phys(page) & (1 << 17)) != 0; @@ -810,26 +815,20 @@ i915_gem_shmem_pwrite(struct drm_device *dev, goto next_page; hit_slowpath = 1; - page_cache_get(page); mutex_unlock(&dev->struct_mutex); - ret = shmem_pwrite_slow(page, shmem_page_offset, page_length, user_data, page_do_bit17_swizzling, partial_cacheline_write, needs_clflush_after); mutex_lock(&dev->struct_mutex); - page_cache_release(page); + next_page: set_page_dirty(page); mark_page_accessed(page); - if (release_page) - page_cache_release(page); - if (ret) { - ret = -EFAULT; + if (ret) goto out; - } remain -= page_length; user_data += page_length; @@ -837,6 +836,8 @@ next_page: } out: + i915_gem_object_unpin_pages(obj); + if (hit_slowpath) { /* Fixup: Kill any reinstated backing storage pages */ if (obj->madv == __I915_MADV_PURGED) @@ -920,10 +921,8 @@ i915_gem_pwrite_ioctl(struct drm_device *dev, void *data, goto out; } - if (obj->gtt_space && - obj->cache_level == I915_CACHE_NONE && + if (obj->cache_level == I915_CACHE_NONE && obj->tiling_mode == I915_TILING_NONE && - obj->map_and_fenceable && obj->base.write_domain != I915_GEM_DOMAIN_CPU) { ret = i915_gem_gtt_pwrite_fast(dev, obj, args, file); /* Note that the gtt paths might fail with non-page-backed user @@ -931,7 +930,7 @@ i915_gem_pwrite_ioctl(struct drm_device *dev, void *data, * textures). Fallback to the shmem path in that case. */ } - if (ret == -EFAULT) + if (ret == -EFAULT || ret == -ENOSPC) ret = i915_gem_shmem_pwrite(dev, obj, args, file); out: @@ -941,6 +940,240 @@ unlock: return ret; } +int +i915_gem_check_wedge(struct drm_i915_private *dev_priv, + bool interruptible) +{ + if (atomic_read(&dev_priv->mm.wedged)) { + struct completion *x = &dev_priv->error_completion; + bool recovery_complete; + unsigned long flags; + + /* Give the error handler a chance to run. */ + spin_lock_irqsave(&x->wait.lock, flags); + recovery_complete = x->done > 0; + spin_unlock_irqrestore(&x->wait.lock, flags); + + /* Non-interruptible callers can't handle -EAGAIN, hence return + * -EIO unconditionally for these. */ + if (!interruptible) + return -EIO; + + /* Recovery complete, but still wedged means reset failure. */ + if (recovery_complete) + return -EIO; + + return -EAGAIN; + } + + return 0; +} + +/* + * Compare seqno against outstanding lazy request. Emit a request if they are + * equal. + */ +static int +i915_gem_check_olr(struct intel_ring_buffer *ring, u32 seqno) +{ + int ret; + + BUG_ON(!mutex_is_locked(&ring->dev->struct_mutex)); + + ret = 0; + if (seqno == ring->outstanding_lazy_request) + ret = i915_add_request(ring, NULL, NULL); + + return ret; +} + +/** + * __wait_seqno - wait until execution of seqno has finished + * @ring: the ring expected to report seqno + * @seqno: duh! + * @interruptible: do an interruptible wait (normally yes) + * @timeout: in - how long to wait (NULL forever); out - how much time remaining + * + * Returns 0 if the seqno was found within the alloted time. Else returns the + * errno with remaining time filled in timeout argument. + */ +static int __wait_seqno(struct intel_ring_buffer *ring, u32 seqno, + bool interruptible, struct timespec *timeout) +{ + drm_i915_private_t *dev_priv = ring->dev->dev_private; + struct timespec before, now, wait_time={1,0}; + unsigned long timeout_jiffies; + long end; + bool wait_forever = true; + int ret; + + if (i915_seqno_passed(ring->get_seqno(ring, true), seqno)) + return 0; + + trace_i915_gem_request_wait_begin(ring, seqno); + + if (timeout != NULL) { + wait_time = *timeout; + wait_forever = false; + } + + timeout_jiffies = timespec_to_jiffies(&wait_time); + + if (WARN_ON(!ring->irq_get(ring))) + return -ENODEV; + + /* Record current time in case interrupted by signal, or wedged * */ + getrawmonotonic(&before); + +#define EXIT_COND \ + (i915_seqno_passed(ring->get_seqno(ring, false), seqno) || \ + atomic_read(&dev_priv->mm.wedged)) + do { + if (interruptible) + end = wait_event_interruptible_timeout(ring->irq_queue, + EXIT_COND, + timeout_jiffies); + else + end = wait_event_timeout(ring->irq_queue, EXIT_COND, + timeout_jiffies); + + ret = i915_gem_check_wedge(dev_priv, interruptible); + if (ret) + end = ret; + } while (end == 0 && wait_forever); + + getrawmonotonic(&now); + + ring->irq_put(ring); + trace_i915_gem_request_wait_end(ring, seqno); +#undef EXIT_COND + + if (timeout) { + struct timespec sleep_time = timespec_sub(now, before); + *timeout = timespec_sub(*timeout, sleep_time); + } + + switch (end) { + case -EIO: + case -EAGAIN: /* Wedged */ + case -ERESTARTSYS: /* Signal */ + return (int)end; + case 0: /* Timeout */ + if (timeout) + set_normalized_timespec(timeout, 0, 0); + return -ETIME; + default: /* Completed */ + WARN_ON(end < 0); /* We're not aware of other errors */ + return 0; + } +} + +/** + * Waits for a sequence number to be signaled, and cleans up the + * request and object lists appropriately for that event. + */ +int +i915_wait_seqno(struct intel_ring_buffer *ring, uint32_t seqno) +{ + struct drm_device *dev = ring->dev; + struct drm_i915_private *dev_priv = dev->dev_private; + bool interruptible = dev_priv->mm.interruptible; + int ret; + + BUG_ON(!mutex_is_locked(&dev->struct_mutex)); + BUG_ON(seqno == 0); + + ret = i915_gem_check_wedge(dev_priv, interruptible); + if (ret) + return ret; + + ret = i915_gem_check_olr(ring, seqno); + if (ret) + return ret; + + return __wait_seqno(ring, seqno, interruptible, NULL); +} + +/** + * Ensures that all rendering to the object has completed and the object is + * safe to unbind from the GTT or access from the CPU. + */ +static __must_check int +i915_gem_object_wait_rendering(struct drm_i915_gem_object *obj, + bool readonly) +{ + struct intel_ring_buffer *ring = obj->ring; + u32 seqno; + int ret; + + seqno = readonly ? obj->last_write_seqno : obj->last_read_seqno; + if (seqno == 0) + return 0; + + ret = i915_wait_seqno(ring, seqno); + if (ret) + return ret; + + i915_gem_retire_requests_ring(ring); + + /* Manually manage the write flush as we may have not yet + * retired the buffer. + */ + if (obj->last_write_seqno && + i915_seqno_passed(seqno, obj->last_write_seqno)) { + obj->last_write_seqno = 0; + obj->base.write_domain &= ~I915_GEM_GPU_DOMAINS; + } + + return 0; +} + +/* A nonblocking variant of the above wait. This is a highly dangerous routine + * as the object state may change during this call. + */ +static __must_check int +i915_gem_object_wait_rendering__nonblocking(struct drm_i915_gem_object *obj, + bool readonly) +{ + struct drm_device *dev = obj->base.dev; + struct drm_i915_private *dev_priv = dev->dev_private; + struct intel_ring_buffer *ring = obj->ring; + u32 seqno; + int ret; + + BUG_ON(!mutex_is_locked(&dev->struct_mutex)); + BUG_ON(!dev_priv->mm.interruptible); + + seqno = readonly ? obj->last_write_seqno : obj->last_read_seqno; + if (seqno == 0) + return 0; + + ret = i915_gem_check_wedge(dev_priv, true); + if (ret) + return ret; + + ret = i915_gem_check_olr(ring, seqno); + if (ret) + return ret; + + mutex_unlock(&dev->struct_mutex); + ret = __wait_seqno(ring, seqno, true, NULL); + mutex_lock(&dev->struct_mutex); + + i915_gem_retire_requests_ring(ring); + + /* Manually manage the write flush as we may have not yet + * retired the buffer. + */ + if (obj->last_write_seqno && + i915_seqno_passed(seqno, obj->last_write_seqno)) { + obj->last_write_seqno = 0; + obj->base.write_domain &= ~I915_GEM_GPU_DOMAINS; + } + + return ret; +} + /** * Called when user space prepares to use an object with the CPU, either * through the mmap ioctl's mapping or a GTT mapping. @@ -978,6 +1211,14 @@ i915_gem_set_domain_ioctl(struct drm_device *dev, void *data, goto unlock; } + /* Try to flush the object off the GPU without holding the lock. + * We will repeat the flush holding the lock in the normal manner + * to catch cases where we are gazumped. + */ + ret = i915_gem_object_wait_rendering__nonblocking(obj, !write_domain); + if (ret) + goto unref; + if (read_domains & I915_GEM_DOMAIN_GTT) { ret = i915_gem_object_set_to_gtt_domain(obj, write_domain != 0); @@ -991,6 +1232,7 @@ i915_gem_set_domain_ioctl(struct drm_device *dev, void *data, ret = i915_gem_object_set_to_cpu_domain(obj, write_domain != 0); } +unref: drm_gem_object_unreference(&obj->base); unlock: mutex_unlock(&dev->struct_mutex); @@ -1110,7 +1352,7 @@ int i915_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf) goto unlock; } if (!obj->gtt_space) { - ret = i915_gem_object_bind_to_gtt(obj, 0, true); + ret = i915_gem_object_bind_to_gtt(obj, 0, true, false); if (ret) goto unlock; @@ -1271,6 +1513,42 @@ i915_gem_get_unfenced_gtt_alignment(struct drm_device *dev, return i915_gem_get_gtt_size(dev, size, tiling_mode); } +static int i915_gem_object_create_mmap_offset(struct drm_i915_gem_object *obj) +{ + struct drm_i915_private *dev_priv = obj->base.dev->dev_private; + int ret; + + if (obj->base.map_list.map) + return 0; + + ret = drm_gem_create_mmap_offset(&obj->base); + if (ret != -ENOSPC) + return ret; + + /* Badly fragmented mmap space? The only way we can recover + * space is by destroying unwanted objects. We can't randomly release + * mmap_offsets as userspace expects them to be persistent for the + * lifetime of the objects. The closest we can is to release the + * offsets on purgeable objects by truncating it and marking it purged, + * which prevents userspace from ever using that object again. + */ + i915_gem_purge(dev_priv, obj->base.size >> PAGE_SHIFT); + ret = drm_gem_create_mmap_offset(&obj->base); + if (ret != -ENOSPC) + return ret; + + i915_gem_shrink_all(dev_priv); + return drm_gem_create_mmap_offset(&obj->base); +} + +static void i915_gem_object_free_mmap_offset(struct drm_i915_gem_object *obj) +{ + if (!obj->base.map_list.map) + return; + + drm_gem_free_mmap_offset(&obj->base); +} + int i915_gem_mmap_gtt(struct drm_file *file, struct drm_device *dev, @@ -1302,11 +1580,9 @@ i915_gem_mmap_gtt(struct drm_file *file, goto out; } - if (!obj->base.map_list.map) { - ret = drm_gem_create_mmap_offset(&obj->base); - if (ret) - goto out; - } + ret = i915_gem_object_create_mmap_offset(obj); + if (ret) + goto out; *offset = (u64)obj->base.map_list.hash.key << PAGE_SHIFT; @@ -1341,83 +1617,245 @@ i915_gem_mmap_gtt_ioctl(struct drm_device *dev, void *data, return i915_gem_mmap_gtt(file, dev, args->handle, &args->offset); } -int -i915_gem_object_get_pages_gtt(struct drm_i915_gem_object *obj, - gfp_t gfpmask) +/* Immediately discard the backing storage */ +static void +i915_gem_object_truncate(struct drm_i915_gem_object *obj) { - int page_count, i; - struct address_space *mapping; struct inode *inode; - struct page *page; - if (obj->pages || obj->sg_table) - return 0; + i915_gem_object_free_mmap_offset(obj); - /* Get the list of pages out of our struct file. They'll be pinned - * at this point until we release them. - */ - page_count = obj->base.size / PAGE_SIZE; - BUG_ON(obj->pages != NULL); - obj->pages = drm_malloc_ab(page_count, sizeof(struct page *)); - if (obj->pages == NULL) - return -ENOMEM; + if (obj->base.filp == NULL) + return; + /* Our goal here is to return as much of the memory as + * is possible back to the system as we are called from OOM. + * To do this we must instruct the shmfs to drop all of its + * backing pages, *now*. + */ inode = obj->base.filp->f_path.dentry->d_inode; - mapping = inode->i_mapping; - gfpmask |= mapping_gfp_mask(mapping); - - for (i = 0; i < page_count; i++) { - page = shmem_read_mapping_page_gfp(mapping, i, gfpmask); - if (IS_ERR(page)) - goto err_pages; - - obj->pages[i] = page; - } - - if (i915_gem_object_needs_bit17_swizzle(obj)) - i915_gem_object_do_bit_17_swizzle(obj); - - return 0; + shmem_truncate_range(inode, 0, (loff_t)-1); -err_pages: - while (i--) - page_cache_release(obj->pages[i]); + obj->madv = __I915_MADV_PURGED; +} - drm_free_large(obj->pages); - obj->pages = NULL; - return PTR_ERR(page); +static inline int +i915_gem_object_is_purgeable(struct drm_i915_gem_object *obj) +{ + return obj->madv == I915_MADV_DONTNEED; } static void i915_gem_object_put_pages_gtt(struct drm_i915_gem_object *obj) { int page_count = obj->base.size / PAGE_SIZE; - int i; - - if (!obj->pages) - return; + struct scatterlist *sg; + int ret, i; BUG_ON(obj->madv == __I915_MADV_PURGED); + ret = i915_gem_object_set_to_cpu_domain(obj, true); + if (ret) { + /* In the event of a disaster, abandon all caches and + * hope for the best. + */ + WARN_ON(ret != -EIO); + i915_gem_clflush_object(obj); + obj->base.read_domains = obj->base.write_domain = I915_GEM_DOMAIN_CPU; + } + if (i915_gem_object_needs_bit17_swizzle(obj)) i915_gem_object_save_bit_17_swizzle(obj); if (obj->madv == I915_MADV_DONTNEED) obj->dirty = 0; - for (i = 0; i < page_count; i++) { + for_each_sg(obj->pages->sgl, sg, page_count, i) { + struct page *page = sg_page(sg); + if (obj->dirty) - set_page_dirty(obj->pages[i]); + set_page_dirty(page); if (obj->madv == I915_MADV_WILLNEED) - mark_page_accessed(obj->pages[i]); + mark_page_accessed(page); - page_cache_release(obj->pages[i]); + page_cache_release(page); } obj->dirty = 0; - drm_free_large(obj->pages); + sg_free_table(obj->pages); + kfree(obj->pages); +} + +static int +i915_gem_object_put_pages(struct drm_i915_gem_object *obj) +{ + const struct drm_i915_gem_object_ops *ops = obj->ops; + + if (obj->pages == NULL) + return 0; + + BUG_ON(obj->gtt_space); + + if (obj->pages_pin_count) + return -EBUSY; + + ops->put_pages(obj); obj->pages = NULL; + + list_del(&obj->gtt_list); + if (i915_gem_object_is_purgeable(obj)) + i915_gem_object_truncate(obj); + + return 0; +} + +static long +i915_gem_purge(struct drm_i915_private *dev_priv, long target) +{ + struct drm_i915_gem_object *obj, *next; + long count = 0; + + list_for_each_entry_safe(obj, next, + &dev_priv->mm.unbound_list, + gtt_list) { + if (i915_gem_object_is_purgeable(obj) && + i915_gem_object_put_pages(obj) == 0) { + count += obj->base.size >> PAGE_SHIFT; + if (count >= target) + return count; + } + } + + list_for_each_entry_safe(obj, next, + &dev_priv->mm.inactive_list, + mm_list) { + if (i915_gem_object_is_purgeable(obj) && + i915_gem_object_unbind(obj) == 0 && + i915_gem_object_put_pages(obj) == 0) { + count += obj->base.size >> PAGE_SHIFT; + if (count >= target) + return count; + } + } + + return count; +} + +static void +i915_gem_shrink_all(struct drm_i915_private *dev_priv) +{ + struct drm_i915_gem_object *obj, *next; + + i915_gem_evict_everything(dev_priv->dev); + + list_for_each_entry_safe(obj, next, &dev_priv->mm.unbound_list, gtt_list) + i915_gem_object_put_pages(obj); +} + +static int +i915_gem_object_get_pages_gtt(struct drm_i915_gem_object *obj) +{ + struct drm_i915_private *dev_priv = obj->base.dev->dev_private; + int page_count, i; + struct address_space *mapping; + struct sg_table *st; + struct scatterlist *sg; + struct page *page; + gfp_t gfp; + + /* Assert that the object is not currently in any GPU domain. As it + * wasn't in the GTT, there shouldn't be any way it could have been in + * a GPU cache + */ + BUG_ON(obj->base.read_domains & I915_GEM_GPU_DOMAINS); + BUG_ON(obj->base.write_domain & I915_GEM_GPU_DOMAINS); + + st = kmalloc(sizeof(*st), GFP_KERNEL); + if (st == NULL) + return -ENOMEM; + + page_count = obj->base.size / PAGE_SIZE; + if (sg_alloc_table(st, page_count, GFP_KERNEL)) { + sg_free_table(st); + kfree(st); + return -ENOMEM; + } + + /* Get the list of pages out of our struct file. They'll be pinned + * at this point until we release them. + * + * Fail silently without starting the shrinker + */ + mapping = obj->base.filp->f_path.dentry->d_inode->i_mapping; + gfp = mapping_gfp_mask(mapping); + gfp |= __GFP_NORETRY | __GFP_NOWARN; + gfp &= ~(__GFP_IO | __GFP_WAIT); + for_each_sg(st->sgl, sg, page_count, i) { + page = shmem_read_mapping_page_gfp(mapping, i, gfp); + if (IS_ERR(page)) { + i915_gem_purge(dev_priv, page_count); + page = shmem_read_mapping_page_gfp(mapping, i, gfp); + } + if (IS_ERR(page)) { + /* We've tried hard to allocate the memory by reaping + * our own buffer, now let the real VM do its job and + * go down in flames if truly OOM. + */ + gfp &= ~(__GFP_NORETRY | __GFP_NOWARN); + gfp |= __GFP_IO | __GFP_WAIT; + + i915_gem_shrink_all(dev_priv); + page = shmem_read_mapping_page_gfp(mapping, i, gfp); + if (IS_ERR(page)) + goto err_pages; + + gfp |= __GFP_NORETRY | __GFP_NOWARN; + gfp &= ~(__GFP_IO | __GFP_WAIT); + } + + sg_set_page(sg, page, PAGE_SIZE, 0); + } + + if (i915_gem_object_needs_bit17_swizzle(obj)) + i915_gem_object_do_bit_17_swizzle(obj); + + obj->pages = st; + return 0; + +err_pages: + for_each_sg(st->sgl, sg, i, page_count) + page_cache_release(sg_page(sg)); + sg_free_table(st); + kfree(st); + return PTR_ERR(page); +} + +/* Ensure that the associated pages are gathered from the backing storage + * and pinned into our object. i915_gem_object_get_pages() may be called + * multiple times before they are released by a single call to + * i915_gem_object_put_pages() - once the pages are no longer referenced + * either as a result of memory pressure (reaping pages under the shrinker) + * or as the object is itself released. + */ +int +i915_gem_object_get_pages(struct drm_i915_gem_object *obj) +{ + struct drm_i915_private *dev_priv = obj->base.dev->dev_private; + const struct drm_i915_gem_object_ops *ops = obj->ops; + int ret; + + if (obj->pages) + return 0; + + BUG_ON(obj->pages_pin_count); + + ret = ops->get_pages(obj); + if (ret) + return ret; + + list_add_tail(&obj->gtt_list, &dev_priv->mm.unbound_list); + return 0; } void @@ -1441,7 +1879,7 @@ i915_gem_object_move_to_active(struct drm_i915_gem_object *obj, list_move_tail(&obj->mm_list, &dev_priv->mm.active_list); list_move_tail(&obj->ring_list, &ring->active_list); - obj->last_rendering_seqno = seqno; + obj->last_read_seqno = seqno; if (obj->fenced_gpu_access) { obj->last_fenced_seqno = seqno; @@ -1458,97 +1896,35 @@ i915_gem_object_move_to_active(struct drm_i915_gem_object *obj, } static void -i915_gem_object_move_off_active(struct drm_i915_gem_object *obj) -{ - list_del_init(&obj->ring_list); - obj->last_rendering_seqno = 0; - obj->last_fenced_seqno = 0; -} - -static void -i915_gem_object_move_to_flushing(struct drm_i915_gem_object *obj) +i915_gem_object_move_to_inactive(struct drm_i915_gem_object *obj) { struct drm_device *dev = obj->base.dev; - drm_i915_private_t *dev_priv = dev->dev_private; + struct drm_i915_private *dev_priv = dev->dev_private; + BUG_ON(obj->base.write_domain & ~I915_GEM_GPU_DOMAINS); BUG_ON(!obj->active); - list_move_tail(&obj->mm_list, &dev_priv->mm.flushing_list); - - i915_gem_object_move_off_active(obj); -} -static void -i915_gem_object_move_to_inactive(struct drm_i915_gem_object *obj) -{ - struct drm_device *dev = obj->base.dev; - struct drm_i915_private *dev_priv = dev->dev_private; + if (obj->pin_count) /* are we a framebuffer? */ + intel_mark_fb_idle(obj); list_move_tail(&obj->mm_list, &dev_priv->mm.inactive_list); - BUG_ON(!list_empty(&obj->gpu_write_list)); - BUG_ON(!obj->active); + list_del_init(&obj->ring_list); obj->ring = NULL; - i915_gem_object_move_off_active(obj); + obj->last_read_seqno = 0; + obj->last_write_seqno = 0; + obj->base.write_domain = 0; + + obj->last_fenced_seqno = 0; obj->fenced_gpu_access = false; obj->active = 0; - obj->pending_gpu_write = false; drm_gem_object_unreference(&obj->base); WARN_ON(i915_verify_lists(dev)); } -/* Immediately discard the backing storage */ -static void -i915_gem_object_truncate(struct drm_i915_gem_object *obj) -{ - struct inode *inode; - - /* Our goal here is to return as much of the memory as - * is possible back to the system as we are called from OOM. - * To do this we must instruct the shmfs to drop all of its - * backing pages, *now*. - */ - inode = obj->base.filp->f_path.dentry->d_inode; - shmem_truncate_range(inode, 0, (loff_t)-1); - - if (obj->base.map_list.map) - drm_gem_free_mmap_offset(&obj->base); - - obj->madv = __I915_MADV_PURGED; -} - -static inline int -i915_gem_object_is_purgeable(struct drm_i915_gem_object *obj) -{ - return obj->madv == I915_MADV_DONTNEED; -} - -static void -i915_gem_process_flushing_list(struct intel_ring_buffer *ring, - uint32_t flush_domains) -{ - struct drm_i915_gem_object *obj, *next; - - list_for_each_entry_safe(obj, next, - &ring->gpu_write_list, - gpu_write_list) { - if (obj->base.write_domain & flush_domains) { - uint32_t old_write_domain = obj->base.write_domain; - - obj->base.write_domain = 0; - list_del_init(&obj->gpu_write_list); - i915_gem_object_move_to_active(obj, ring, - i915_gem_next_request_seqno(ring)); - - trace_i915_gem_object_change_domain(obj, - obj->base.read_domains, - old_write_domain); - } - } -} - static u32 i915_gem_get_seqno(struct drm_device *dev) { @@ -1589,15 +1965,16 @@ i915_add_request(struct intel_ring_buffer *ring, * is that the flush _must_ happen before the next request, no matter * what. */ - if (ring->gpu_caches_dirty) { - ret = i915_gem_flush_ring(ring, 0, I915_GEM_GPU_DOMAINS); - if (ret) - return ret; + ret = intel_ring_flush_all_caches(ring); + if (ret) + return ret; - ring->gpu_caches_dirty = false; + if (request == NULL) { + request = kmalloc(sizeof(*request), GFP_KERNEL); + if (request == NULL) + return -ENOMEM; } - BUG_ON(request == NULL); seqno = i915_gem_next_request_seqno(ring); /* Record the position of the start of the request so that @@ -1608,8 +1985,10 @@ i915_add_request(struct intel_ring_buffer *ring, request_ring_position = intel_ring_get_tail(ring); ret = ring->add_request(ring, &seqno); - if (ret) - return ret; + if (ret) { + kfree(request); + return ret; + } trace_i915_gem_request_add(ring, seqno); @@ -1619,6 +1998,7 @@ i915_add_request(struct intel_ring_buffer *ring, request->emitted_jiffies = jiffies; was_empty = list_empty(&ring->request_list); list_add_tail(&request->list, &ring->request_list); + request->file_priv = NULL; if (file) { struct drm_i915_file_private *file_priv = file->driver_priv; @@ -1638,13 +2018,13 @@ i915_add_request(struct intel_ring_buffer *ring, jiffies + msecs_to_jiffies(DRM_I915_HANGCHECK_PERIOD)); } - if (was_empty) + if (was_empty) { queue_delayed_work(dev_priv->wq, &dev_priv->mm.retire_work, HZ); + intel_mark_busy(dev_priv->dev); + } } - WARN_ON(!list_empty(&ring->gpu_write_list)); - return 0; } @@ -1686,8 +2066,6 @@ static void i915_gem_reset_ring_lists(struct drm_i915_private *dev_priv, struct drm_i915_gem_object, ring_list); - obj->base.write_domain = 0; - list_del_init(&obj->gpu_write_list); i915_gem_object_move_to_inactive(obj); } } @@ -1723,20 +2101,6 @@ void i915_gem_reset(struct drm_device *dev) for_each_ring(ring, dev_priv, i) i915_gem_reset_ring_lists(dev_priv, ring); - /* Remove anything from the flushing lists. The GPU cache is likely - * to be lost on reset along with the data, so simply move the - * lost bo to the inactive list. - */ - while (!list_empty(&dev_priv->mm.flushing_list)) { - obj = list_first_entry(&dev_priv->mm.flushing_list, - struct drm_i915_gem_object, - mm_list); - - obj->base.write_domain = 0; - list_del_init(&obj->gpu_write_list); - i915_gem_object_move_to_inactive(obj); - } - /* Move everything out of the GPU domains to ensure we do any * necessary invalidation upon reuse. */ @@ -1765,7 +2129,7 @@ i915_gem_retire_requests_ring(struct intel_ring_buffer *ring) WARN_ON(i915_verify_lists(ring->dev)); - seqno = ring->get_seqno(ring); + seqno = ring->get_seqno(ring, true); for (i = 0; i < ARRAY_SIZE(ring->sync_seqno); i++) if (seqno >= ring->sync_seqno[i]) @@ -1804,13 +2168,10 @@ i915_gem_retire_requests_ring(struct intel_ring_buffer *ring) struct drm_i915_gem_object, ring_list); - if (!i915_seqno_passed(seqno, obj->last_rendering_seqno)) + if (!i915_seqno_passed(seqno, obj->last_read_seqno)) break; - if (obj->base.write_domain != 0) - i915_gem_object_move_to_flushing(obj); - else - i915_gem_object_move_to_inactive(obj); + i915_gem_object_move_to_inactive(obj); } if (unlikely(ring->trace_irq_seqno && @@ -1859,216 +2220,20 @@ i915_gem_retire_work_handler(struct work_struct *work) */ idle = true; for_each_ring(ring, dev_priv, i) { - if (ring->gpu_caches_dirty) { - struct drm_i915_gem_request *request; - - request = kzalloc(sizeof(*request), GFP_KERNEL); - if (request == NULL || - i915_add_request(ring, NULL, request)) - kfree(request); - } + if (ring->gpu_caches_dirty) + i915_add_request(ring, NULL, NULL); idle &= list_empty(&ring->request_list); } if (!dev_priv->mm.suspended && !idle) queue_delayed_work(dev_priv->wq, &dev_priv->mm.retire_work, HZ); + if (idle) + intel_mark_idle(dev); mutex_unlock(&dev->struct_mutex); } -int -i915_gem_check_wedge(struct drm_i915_private *dev_priv, - bool interruptible) -{ - if (atomic_read(&dev_priv->mm.wedged)) { - struct completion *x = &dev_priv->error_completion; - bool recovery_complete; - unsigned long flags; - - /* Give the error handler a chance to run. */ - spin_lock_irqsave(&x->wait.lock, flags); - recovery_complete = x->done > 0; - spin_unlock_irqrestore(&x->wait.lock, flags); - - /* Non-interruptible callers can't handle -EAGAIN, hence return - * -EIO unconditionally for these. */ - if (!interruptible) - return -EIO; - - /* Recovery complete, but still wedged means reset failure. */ - if (recovery_complete) - return -EIO; - - return -EAGAIN; - } - - return 0; -} - -/* - * Compare seqno against outstanding lazy request. Emit a request if they are - * equal. - */ -static int -i915_gem_check_olr(struct intel_ring_buffer *ring, u32 seqno) -{ - int ret = 0; - - BUG_ON(!mutex_is_locked(&ring->dev->struct_mutex)); - - if (seqno == ring->outstanding_lazy_request) { - struct drm_i915_gem_request *request; - - request = kzalloc(sizeof(*request), GFP_KERNEL); - if (request == NULL) - return -ENOMEM; - - ret = i915_add_request(ring, NULL, request); - if (ret) { - kfree(request); - return ret; - } - - BUG_ON(seqno != request->seqno); - } - - return ret; -} - -/** - * __wait_seqno - wait until execution of seqno has finished - * @ring: the ring expected to report seqno - * @seqno: duh! - * @interruptible: do an interruptible wait (normally yes) - * @timeout: in - how long to wait (NULL forever); out - how much time remaining - * - * Returns 0 if the seqno was found within the alloted time. Else returns the - * errno with remaining time filled in timeout argument. - */ -static int __wait_seqno(struct intel_ring_buffer *ring, u32 seqno, - bool interruptible, struct timespec *timeout) -{ - drm_i915_private_t *dev_priv = ring->dev->dev_private; - struct timespec before, now, wait_time={1,0}; - unsigned long timeout_jiffies; - long end; - bool wait_forever = true; - int ret; - - if (i915_seqno_passed(ring->get_seqno(ring), seqno)) - return 0; - - trace_i915_gem_request_wait_begin(ring, seqno); - - if (timeout != NULL) { - wait_time = *timeout; - wait_forever = false; - } - - timeout_jiffies = timespec_to_jiffies(&wait_time); - - if (WARN_ON(!ring->irq_get(ring))) - return -ENODEV; - - /* Record current time in case interrupted by signal, or wedged * */ - getrawmonotonic(&before); - -#define EXIT_COND \ - (i915_seqno_passed(ring->get_seqno(ring), seqno) || \ - atomic_read(&dev_priv->mm.wedged)) - do { - if (interruptible) - end = wait_event_interruptible_timeout(ring->irq_queue, - EXIT_COND, - timeout_jiffies); - else - end = wait_event_timeout(ring->irq_queue, EXIT_COND, - timeout_jiffies); - - ret = i915_gem_check_wedge(dev_priv, interruptible); - if (ret) - end = ret; - } while (end == 0 && wait_forever); - - getrawmonotonic(&now); - - ring->irq_put(ring); - trace_i915_gem_request_wait_end(ring, seqno); -#undef EXIT_COND - - if (timeout) { - struct timespec sleep_time = timespec_sub(now, before); - *timeout = timespec_sub(*timeout, sleep_time); - } - - switch (end) { - case -EIO: - case -EAGAIN: /* Wedged */ - case -ERESTARTSYS: /* Signal */ - return (int)end; - case 0: /* Timeout */ - if (timeout) - set_normalized_timespec(timeout, 0, 0); - return -ETIME; - default: /* Completed */ - WARN_ON(end < 0); /* We're not aware of other errors */ - return 0; - } -} - -/** - * Waits for a sequence number to be signaled, and cleans up the - * request and object lists appropriately for that event. - */ -int -i915_wait_seqno(struct intel_ring_buffer *ring, uint32_t seqno) -{ - drm_i915_private_t *dev_priv = ring->dev->dev_private; - int ret = 0; - - BUG_ON(seqno == 0); - - ret = i915_gem_check_wedge(dev_priv, dev_priv->mm.interruptible); - if (ret) - return ret; - - ret = i915_gem_check_olr(ring, seqno); - if (ret) - return ret; - - ret = __wait_seqno(ring, seqno, dev_priv->mm.interruptible, NULL); - - return ret; -} - -/** - * Ensures that all rendering to the object has completed and the object is - * safe to unbind from the GTT or access from the CPU. - */ -int -i915_gem_object_wait_rendering(struct drm_i915_gem_object *obj) -{ - int ret; - - /* This function only exists to support waiting for existing rendering, - * not for emitting required flushes. - */ - BUG_ON((obj->base.write_domain & I915_GEM_GPU_DOMAINS) != 0); - - /* If there is rendering queued on the buffer being evicted, wait for - * it. - */ - if (obj->active) { - ret = i915_wait_seqno(obj->ring, obj->last_rendering_seqno); - if (ret) - return ret; - i915_gem_retire_requests_ring(obj->ring); - } - - return 0; -} - /** * Ensures that an object will eventually get non-busy by flushing any required * write domains, emitting any outstanding lazy request and retiring and @@ -2080,14 +2245,10 @@ i915_gem_object_flush_active(struct drm_i915_gem_object *obj) int ret; if (obj->active) { - ret = i915_gem_object_flush_gpu_write_domain(obj); + ret = i915_gem_check_olr(obj->ring, obj->last_read_seqno); if (ret) return ret; - ret = i915_gem_check_olr(obj->ring, - obj->last_rendering_seqno); - if (ret) - return ret; i915_gem_retire_requests_ring(obj->ring); } @@ -2147,7 +2308,7 @@ i915_gem_wait_ioctl(struct drm_device *dev, void *data, struct drm_file *file) goto out; if (obj->active) { - seqno = obj->last_rendering_seqno; + seqno = obj->last_read_seqno; ring = obj->ring; } @@ -2202,11 +2363,11 @@ i915_gem_object_sync(struct drm_i915_gem_object *obj, return 0; if (to == NULL || !i915_semaphore_is_enabled(obj->base.dev)) - return i915_gem_object_wait_rendering(obj); + return i915_gem_object_wait_rendering(obj, false); idx = intel_ring_sync_index(from, to); - seqno = obj->last_rendering_seqno; + seqno = obj->last_read_seqno; if (seqno <= from->sync_seqno[idx]) return 0; @@ -2260,6 +2421,8 @@ i915_gem_object_unbind(struct drm_i915_gem_object *obj) if (obj->pin_count) return -EBUSY; + BUG_ON(obj->pages == NULL); + ret = i915_gem_object_finish_gpu(obj); if (ret) return ret; @@ -2270,22 +2433,6 @@ i915_gem_object_unbind(struct drm_i915_gem_object *obj) i915_gem_object_finish_gtt(obj); - /* Move the object to the CPU domain to ensure that - * any possible CPU writes while it's not in the GTT - * are flushed when we go to remap it. - */ - if (ret == 0) - ret = i915_gem_object_set_to_cpu_domain(obj, 1); - if (ret == -ERESTARTSYS) - return ret; - if (ret) { - /* In the event of a disaster, abandon all caches and - * hope for the best. - */ - i915_gem_clflush_object(obj); - obj->base.read_domains = obj->base.write_domain = I915_GEM_DOMAIN_CPU; - } - /* release the fence reg _after_ flushing */ ret = i915_gem_object_put_fence(obj); if (ret) @@ -2301,10 +2448,8 @@ i915_gem_object_unbind(struct drm_i915_gem_object *obj) } i915_gem_gtt_finish_object(obj); - i915_gem_object_put_pages_gtt(obj); - - list_del_init(&obj->gtt_list); - list_del_init(&obj->mm_list); + list_del(&obj->mm_list); + list_move_tail(&obj->gtt_list, &dev_priv->mm.unbound_list); /* Avoid an unnecessary call to unbind on rebind. */ obj->map_and_fenceable = true; @@ -2312,48 +2457,14 @@ i915_gem_object_unbind(struct drm_i915_gem_object *obj) obj->gtt_space = NULL; obj->gtt_offset = 0; - if (i915_gem_object_is_purgeable(obj)) - i915_gem_object_truncate(obj); - - return ret; -} - -int -i915_gem_flush_ring(struct intel_ring_buffer *ring, - uint32_t invalidate_domains, - uint32_t flush_domains) -{ - int ret; - - if (((invalidate_domains | flush_domains) & I915_GEM_GPU_DOMAINS) == 0) - return 0; - - trace_i915_gem_ring_flush(ring, invalidate_domains, flush_domains); - - ret = ring->flush(ring, invalidate_domains, flush_domains); - if (ret) - return ret; - - if (flush_domains & I915_GEM_GPU_DOMAINS) - i915_gem_process_flushing_list(ring, flush_domains); - return 0; } static int i915_ring_idle(struct intel_ring_buffer *ring) { - int ret; - - if (list_empty(&ring->gpu_write_list) && list_empty(&ring->active_list)) + if (list_empty(&ring->active_list)) return 0; - if (!list_empty(&ring->gpu_write_list)) { - ret = i915_gem_flush_ring(ring, - I915_GEM_GPU_DOMAINS, I915_GEM_GPU_DOMAINS); - if (ret) - return ret; - } - return i915_wait_seqno(ring, i915_gem_next_request_seqno(ring)); } @@ -2372,10 +2483,6 @@ int i915_gpu_idle(struct drm_device *dev) ret = i915_ring_idle(ring); if (ret) return ret; - - /* Is the device fubar? */ - if (WARN_ON(!list_empty(&ring->gpu_write_list))) - return -EBUSY; } return 0; @@ -2548,21 +2655,8 @@ static void i915_gem_object_update_fence(struct drm_i915_gem_object *obj, static int i915_gem_object_flush_fence(struct drm_i915_gem_object *obj) { - int ret; - - if (obj->fenced_gpu_access) { - if (obj->base.write_domain & I915_GEM_GPU_DOMAINS) { - ret = i915_gem_flush_ring(obj->ring, - 0, obj->base.write_domain); - if (ret) - return ret; - } - - obj->fenced_gpu_access = false; - } - if (obj->last_fenced_seqno) { - ret = i915_wait_seqno(obj->ring, obj->last_fenced_seqno); + int ret = i915_wait_seqno(obj->ring, obj->last_fenced_seqno); if (ret) return ret; @@ -2575,6 +2669,7 @@ i915_gem_object_flush_fence(struct drm_i915_gem_object *obj) if (obj->base.read_domains & I915_GEM_DOMAIN_GTT) mb(); + obj->fenced_gpu_access = false; return 0; } @@ -2694,18 +2789,88 @@ i915_gem_object_get_fence(struct drm_i915_gem_object *obj) return 0; } +static bool i915_gem_valid_gtt_space(struct drm_device *dev, + struct drm_mm_node *gtt_space, + unsigned long cache_level) +{ + struct drm_mm_node *other; + + /* On non-LLC machines we have to be careful when putting differing + * types of snoopable memory together to avoid the prefetcher + * crossing memory domains and dieing. + */ + if (HAS_LLC(dev)) + return true; + + if (gtt_space == NULL) + return true; + + if (list_empty(>t_space->node_list)) + return true; + + other = list_entry(gtt_space->node_list.prev, struct drm_mm_node, node_list); + if (other->allocated && !other->hole_follows && other->color != cache_level) + return false; + + other = list_entry(gtt_space->node_list.next, struct drm_mm_node, node_list); + if (other->allocated && !gtt_space->hole_follows && other->color != cache_level) + return false; + + return true; +} + +static void i915_gem_verify_gtt(struct drm_device *dev) +{ +#if WATCH_GTT + struct drm_i915_private *dev_priv = dev->dev_private; + struct drm_i915_gem_object *obj; + int err = 0; + + list_for_each_entry(obj, &dev_priv->mm.gtt_list, gtt_list) { + if (obj->gtt_space == NULL) { + printk(KERN_ERR "object found on GTT list with no space reserved\n"); + err++; + continue; + } + + if (obj->cache_level != obj->gtt_space->color) { + printk(KERN_ERR "object reserved space [%08lx, %08lx] with wrong color, cache_level=%x, color=%lx\n", + obj->gtt_space->start, + obj->gtt_space->start + obj->gtt_space->size, + obj->cache_level, + obj->gtt_space->color); + err++; + continue; + } + + if (!i915_gem_valid_gtt_space(dev, + obj->gtt_space, + obj->cache_level)) { + printk(KERN_ERR "invalid GTT space found at [%08lx, %08lx] - color=%x\n", + obj->gtt_space->start, + obj->gtt_space->start + obj->gtt_space->size, + obj->cache_level); + err++; + continue; + } + } + + WARN_ON(err); +#endif +} + /** * Finds free space in the GTT aperture and binds the object there. */ static int i915_gem_object_bind_to_gtt(struct drm_i915_gem_object *obj, unsigned alignment, - bool map_and_fenceable) + bool map_and_fenceable, + bool nonblocking) { struct drm_device *dev = obj->base.dev; drm_i915_private_t *dev_priv = dev->dev_private; struct drm_mm_node *free_space; - gfp_t gfpmask = __GFP_NORETRY | __GFP_NOWARN; u32 size, fence_size, fence_alignment, unfenced_alignment; bool mappable, fenceable; int ret; @@ -2745,89 +2910,67 @@ i915_gem_object_bind_to_gtt(struct drm_i915_gem_object *obj, return -E2BIG; } + ret = i915_gem_object_get_pages(obj); + if (ret) + return ret; + search_free: if (map_and_fenceable) free_space = - drm_mm_search_free_in_range(&dev_priv->mm.gtt_space, - size, alignment, - 0, dev_priv->mm.gtt_mappable_end, - 0); + drm_mm_search_free_in_range_color(&dev_priv->mm.gtt_space, + size, alignment, obj->cache_level, + 0, dev_priv->mm.gtt_mappable_end, + false); else - free_space = drm_mm_search_free(&dev_priv->mm.gtt_space, - size, alignment, 0); + free_space = drm_mm_search_free_color(&dev_priv->mm.gtt_space, + size, alignment, obj->cache_level, + false); if (free_space != NULL) { if (map_and_fenceable) obj->gtt_space = drm_mm_get_block_range_generic(free_space, - size, alignment, 0, + size, alignment, obj->cache_level, 0, dev_priv->mm.gtt_mappable_end, - 0); + false); else obj->gtt_space = - drm_mm_get_block(free_space, size, alignment); + drm_mm_get_block_generic(free_space, + size, alignment, obj->cache_level, + false); } if (obj->gtt_space == NULL) { - /* If the gtt is empty and we're still having trouble - * fitting our object in, we're out of memory. - */ ret = i915_gem_evict_something(dev, size, alignment, - map_and_fenceable); + obj->cache_level, + map_and_fenceable, + nonblocking); if (ret) return ret; goto search_free; } - - ret = i915_gem_object_get_pages_gtt(obj, gfpmask); - if (ret) { + if (WARN_ON(!i915_gem_valid_gtt_space(dev, + obj->gtt_space, + obj->cache_level))) { drm_mm_put_block(obj->gtt_space); obj->gtt_space = NULL; - - if (ret == -ENOMEM) { - /* first try to reclaim some memory by clearing the GTT */ - ret = i915_gem_evict_everything(dev, false); - if (ret) { - /* now try to shrink everyone else */ - if (gfpmask) { - gfpmask = 0; - goto search_free; - } - - return -ENOMEM; - } - - goto search_free; - } - - return ret; + return -EINVAL; } + ret = i915_gem_gtt_prepare_object(obj); if (ret) { - i915_gem_object_put_pages_gtt(obj); drm_mm_put_block(obj->gtt_space); obj->gtt_space = NULL; - - if (i915_gem_evict_everything(dev, false)) - return ret; - - goto search_free; + return ret; } if (!dev_priv->mm.aliasing_ppgtt) i915_gem_gtt_bind_object(obj, obj->cache_level); - list_add_tail(&obj->gtt_list, &dev_priv->mm.gtt_list); + list_move_tail(&obj->gtt_list, &dev_priv->mm.bound_list); list_add_tail(&obj->mm_list, &dev_priv->mm.inactive_list); - /* Assert that the object is not currently in any GPU domain. As it - * wasn't in the GTT, there shouldn't be any way it could have been in - * a GPU cache - */ - BUG_ON(obj->base.read_domains & I915_GEM_GPU_DOMAINS); - BUG_ON(obj->base.write_domain & I915_GEM_GPU_DOMAINS); - obj->gtt_offset = obj->gtt_space->start; fenceable = @@ -2840,6 +2983,7 @@ i915_gem_object_bind_to_gtt(struct drm_i915_gem_object *obj, obj->map_and_fenceable = mappable && fenceable; trace_i915_gem_object_bind(obj, map_and_fenceable); + i915_gem_verify_gtt(dev); return 0; } @@ -2866,18 +3010,7 @@ i915_gem_clflush_object(struct drm_i915_gem_object *obj) trace_i915_gem_object_clflush(obj); - drm_clflush_pages(obj->pages, obj->base.size / PAGE_SIZE); -} - -/** Flushes any GPU write domain for the object if it's dirty. */ -static int -i915_gem_object_flush_gpu_write_domain(struct drm_i915_gem_object *obj) -{ - if ((obj->base.write_domain & I915_GEM_GPU_DOMAINS) == 0) - return 0; - - /* Queue the GPU write cache flushing we need. */ - return i915_gem_flush_ring(obj->ring, 0, obj->base.write_domain); + drm_clflush_sg(obj->pages); } /** Flushes the GTT write domain for the object if it's dirty. */ @@ -2946,16 +3079,10 @@ i915_gem_object_set_to_gtt_domain(struct drm_i915_gem_object *obj, bool write) if (obj->base.write_domain == I915_GEM_DOMAIN_GTT) return 0; - ret = i915_gem_object_flush_gpu_write_domain(obj); + ret = i915_gem_object_wait_rendering(obj, !write); if (ret) return ret; - if (obj->pending_gpu_write || write) { - ret = i915_gem_object_wait_rendering(obj); - if (ret) - return ret; - } - i915_gem_object_flush_cpu_write_domain(obj); old_write_domain = obj->base.write_domain; @@ -2998,6 +3125,12 @@ int i915_gem_object_set_cache_level(struct drm_i915_gem_object *obj, return -EBUSY; } + if (!i915_gem_valid_gtt_space(dev, obj->gtt_space, cache_level)) { + ret = i915_gem_object_unbind(obj); + if (ret) + return ret; + } + if (obj->gtt_space) { ret = i915_gem_object_finish_gpu(obj); if (ret) @@ -3009,7 +3142,7 @@ int i915_gem_object_set_cache_level(struct drm_i915_gem_object *obj, * registers with snooped memory, so relinquish any fences * currently pointing to our region in the aperture. */ - if (INTEL_INFO(obj->base.dev)->gen < 6) { + if (INTEL_INFO(dev)->gen < 6) { ret = i915_gem_object_put_fence(obj); if (ret) return ret; @@ -3020,6 +3153,8 @@ int i915_gem_object_set_cache_level(struct drm_i915_gem_object *obj, if (obj->has_aliasing_ppgtt_mapping) i915_ppgtt_bind_object(dev_priv->mm.aliasing_ppgtt, obj, cache_level); + + obj->gtt_space->color = cache_level; } if (cache_level == I915_CACHE_NONE) { @@ -3046,9 +3181,72 @@ int i915_gem_object_set_cache_level(struct drm_i915_gem_object *obj, } obj->cache_level = cache_level; + i915_gem_verify_gtt(dev); return 0; } +int i915_gem_get_caching_ioctl(struct drm_device *dev, void *data, + struct drm_file *file) +{ + struct drm_i915_gem_caching *args = data; + struct drm_i915_gem_object *obj; + int ret; + + ret = i915_mutex_lock_interruptible(dev); + if (ret) + return ret; + + obj = to_intel_bo(drm_gem_object_lookup(dev, file, args->handle)); + if (&obj->base == NULL) { + ret = -ENOENT; + goto unlock; + } + + args->caching = obj->cache_level != I915_CACHE_NONE; + + drm_gem_object_unreference(&obj->base); +unlock: + mutex_unlock(&dev->struct_mutex); + return ret; +} + +int i915_gem_set_caching_ioctl(struct drm_device *dev, void *data, + struct drm_file *file) +{ + struct drm_i915_gem_caching *args = data; + struct drm_i915_gem_object *obj; + enum i915_cache_level level; + int ret; + + ret = i915_mutex_lock_interruptible(dev); + if (ret) + return ret; + + switch (args->caching) { + case I915_CACHING_NONE: + level = I915_CACHE_NONE; + break; + case I915_CACHING_CACHED: + level = I915_CACHE_LLC; + break; + default: + return -EINVAL; + } + + obj = to_intel_bo(drm_gem_object_lookup(dev, file, args->handle)); + if (&obj->base == NULL) { + ret = -ENOENT; + goto unlock; + } + + ret = i915_gem_object_set_cache_level(obj, level); + + drm_gem_object_unreference(&obj->base); +unlock: + mutex_unlock(&dev->struct_mutex); + return ret; +} + /* * Prepare buffer for display plane (scanout, cursors, etc). * Can be called from an uninterruptible phase (modesetting) and allows @@ -3062,10 +3260,6 @@ i915_gem_object_pin_to_display_plane(struct drm_i915_gem_object *obj, u32 old_read_domains, old_write_domain; int ret; - ret = i915_gem_object_flush_gpu_write_domain(obj); - if (ret) - return ret; - if (pipelined != obj->ring) { ret = i915_gem_object_sync(obj, pipelined); if (ret) @@ -3089,7 +3283,7 @@ i915_gem_object_pin_to_display_plane(struct drm_i915_gem_object *obj, * (e.g. libkms for the bootup splash), we have to ensure that we * always use map_and_fenceable for all scanout buffers. */ - ret = i915_gem_object_pin(obj, alignment, true); + ret = i915_gem_object_pin(obj, alignment, true, false); if (ret) return ret; @@ -3101,7 +3295,7 @@ i915_gem_object_pin_to_display_plane(struct drm_i915_gem_object *obj, /* It should now be out of any other write domains, and we can update * the domain values for our changes. */ - BUG_ON((obj->base.write_domain & ~I915_GEM_DOMAIN_GTT) != 0); + obj->base.write_domain = 0; obj->base.read_domains |= I915_GEM_DOMAIN_GTT; trace_i915_gem_object_change_domain(obj, @@ -3119,13 +3313,7 @@ i915_gem_object_finish_gpu(struct drm_i915_gem_object *obj) if ((obj->base.read_domains & I915_GEM_GPU_DOMAINS) == 0) return 0; - if (obj->base.write_domain & I915_GEM_GPU_DOMAINS) { - ret = i915_gem_flush_ring(obj->ring, 0, obj->base.write_domain); - if (ret) - return ret; - } - - ret = i915_gem_object_wait_rendering(obj); + ret = i915_gem_object_wait_rendering(obj, false); if (ret) return ret; @@ -3149,16 +3337,10 @@ i915_gem_object_set_to_cpu_domain(struct drm_i915_gem_object *obj, bool write) if (obj->base.write_domain == I915_GEM_DOMAIN_CPU) return 0; - ret = i915_gem_object_flush_gpu_write_domain(obj); + ret = i915_gem_object_wait_rendering(obj, !write); if (ret) return ret; - if (write || obj->pending_gpu_write) { - ret = i915_gem_object_wait_rendering(obj); - if (ret) - return ret; - } - i915_gem_object_flush_gtt_write_domain(obj); old_write_domain = obj->base.write_domain; @@ -3238,7 +3420,8 @@ i915_gem_ring_throttle(struct drm_device *dev, struct drm_file *file) int i915_gem_object_pin(struct drm_i915_gem_object *obj, uint32_t alignment, - bool map_and_fenceable) + bool map_and_fenceable, + bool nonblocking) { int ret; @@ -3263,7 +3446,8 @@ i915_gem_object_pin(struct drm_i915_gem_object *obj, if (obj->gtt_space == NULL) { ret = i915_gem_object_bind_to_gtt(obj, alignment, - map_and_fenceable); + map_and_fenceable, + nonblocking); if (ret) return ret; } @@ -3321,7 +3505,7 @@ i915_gem_pin_ioctl(struct drm_device *dev, void *data, obj->user_pin_count++; obj->pin_filp = file; if (obj->user_pin_count == 1) { - ret = i915_gem_object_pin(obj, args->alignment, true); + ret = i915_gem_object_pin(obj, args->alignment, true, false); if (ret) goto out; } @@ -3401,6 +3585,10 @@ i915_gem_busy_ioctl(struct drm_device *dev, void *data, ret = i915_gem_object_flush_active(obj); args->busy = obj->active; + if (obj->ring) { + BUILD_BUG_ON(I915_NUM_RINGS > 16); + args->busy |= intel_ring_flag(obj->ring) << 16; + } drm_gem_object_unreference(&obj->base); unlock: @@ -3449,9 +3637,8 @@ i915_gem_madvise_ioctl(struct drm_device *dev, void *data, if (obj->madv != __I915_MADV_PURGED) obj->madv = args->madv; - /* if the object is no longer bound, discard its backing storage */ - if (i915_gem_object_is_purgeable(obj) && - obj->gtt_space == NULL) + /* if the object is no longer attached, discard its backing storage */ + if (i915_gem_object_is_purgeable(obj) && obj->pages == NULL) i915_gem_object_truncate(obj); args->retained = obj->madv != __I915_MADV_PURGED; @@ -3463,10 +3650,32 @@ unlock: return ret; } +void i915_gem_object_init(struct drm_i915_gem_object *obj, + const struct drm_i915_gem_object_ops *ops) +{ + INIT_LIST_HEAD(&obj->mm_list); + INIT_LIST_HEAD(&obj->gtt_list); + INIT_LIST_HEAD(&obj->ring_list); + INIT_LIST_HEAD(&obj->exec_list); + + obj->ops = ops; + + obj->fence_reg = I915_FENCE_REG_NONE; + obj->madv = I915_MADV_WILLNEED; + /* Avoid an unnecessary call to unbind on the first bind. */ + obj->map_and_fenceable = true; + + i915_gem_info_add_obj(obj->base.dev->dev_private, obj->base.size); +} + +static const struct drm_i915_gem_object_ops i915_gem_object_ops = { + .get_pages = i915_gem_object_get_pages_gtt, + .put_pages = i915_gem_object_put_pages_gtt, +}; + struct drm_i915_gem_object *i915_gem_alloc_object(struct drm_device *dev, size_t size) { - struct drm_i915_private *dev_priv = dev->dev_private; struct drm_i915_gem_object *obj; struct address_space *mapping; u32 mask; @@ -3490,7 +3699,7 @@ struct drm_i915_gem_object *i915_gem_alloc_object(struct drm_device *dev, mapping = obj->base.filp->f_path.dentry->d_inode->i_mapping; mapping_set_gfp_mask(mapping, mask); - i915_gem_info_add_obj(dev_priv, size); + i915_gem_object_init(obj, &i915_gem_object_ops); obj->base.write_domain = I915_GEM_DOMAIN_CPU; obj->base.read_domains = I915_GEM_DOMAIN_CPU; @@ -3512,17 +3721,6 @@ struct drm_i915_gem_object *i915_gem_alloc_object(struct drm_device *dev, } else obj->cache_level = I915_CACHE_NONE; - obj->base.driver_private = NULL; - obj->fence_reg = I915_FENCE_REG_NONE; - INIT_LIST_HEAD(&obj->mm_list); - INIT_LIST_HEAD(&obj->gtt_list); - INIT_LIST_HEAD(&obj->ring_list); - INIT_LIST_HEAD(&obj->exec_list); - INIT_LIST_HEAD(&obj->gpu_write_list); - obj->madv = I915_MADV_WILLNEED; - /* Avoid an unnecessary call to unbind on the first bind. */ - obj->map_and_fenceable = true; - return obj; } @@ -3541,9 +3739,6 @@ void i915_gem_free_object(struct drm_gem_object *gem_obj) trace_i915_gem_object_destroy(obj); - if (gem_obj->import_attach) - drm_prime_gem_destroy(gem_obj, obj->sg_table); - if (obj->phys_obj) i915_gem_detach_phys_object(dev, obj); @@ -3559,8 +3754,14 @@ void i915_gem_free_object(struct drm_gem_object *gem_obj) dev_priv->mm.interruptible = was_interruptible; } - if (obj->base.map_list.map) - drm_gem_free_mmap_offset(&obj->base); + obj->pages_pin_count = 0; + i915_gem_object_put_pages(obj); + i915_gem_object_free_mmap_offset(obj); + + BUG_ON(obj->pages); + + if (obj->base.import_attach) + drm_prime_gem_destroy(&obj->base, NULL); drm_gem_object_release(&obj->base); i915_gem_info_remove_obj(dev_priv, obj->base.size); @@ -3591,7 +3792,7 @@ i915_gem_idle(struct drm_device *dev) /* Under UMS, be paranoid and evict. */ if (!drm_core_check_feature(dev, DRIVER_MODESET)) - i915_gem_evict_everything(dev, false); + i915_gem_evict_everything(dev); i915_gem_reset_fences(dev); @@ -3892,7 +4093,6 @@ i915_gem_entervt_ioctl(struct drm_device *dev, void *data, } BUG_ON(!list_empty(&dev_priv->mm.active_list)); - BUG_ON(!list_empty(&dev_priv->mm.flushing_list)); BUG_ON(!list_empty(&dev_priv->mm.inactive_list)); mutex_unlock(&dev->struct_mutex); @@ -3940,7 +4140,6 @@ init_ring_lists(struct intel_ring_buffer *ring) { INIT_LIST_HEAD(&ring->active_list); INIT_LIST_HEAD(&ring->request_list); - INIT_LIST_HEAD(&ring->gpu_write_list); } void @@ -3950,10 +4149,10 @@ i915_gem_load(struct drm_device *dev) drm_i915_private_t *dev_priv = dev->dev_private; INIT_LIST_HEAD(&dev_priv->mm.active_list); - INIT_LIST_HEAD(&dev_priv->mm.flushing_list); INIT_LIST_HEAD(&dev_priv->mm.inactive_list); + INIT_LIST_HEAD(&dev_priv->mm.unbound_list); + INIT_LIST_HEAD(&dev_priv->mm.bound_list); INIT_LIST_HEAD(&dev_priv->mm.fence_list); - INIT_LIST_HEAD(&dev_priv->mm.gtt_list); for (i = 0; i < I915_NUM_RINGS; i++) init_ring_lists(&dev_priv->ring[i]); for (i = 0; i < I915_MAX_NUM_FENCES; i++) @@ -4198,18 +4397,6 @@ void i915_gem_release(struct drm_device *dev, struct drm_file *file) } static int -i915_gpu_is_active(struct drm_device *dev) -{ - drm_i915_private_t *dev_priv = dev->dev_private; - int lists_empty; - - lists_empty = list_empty(&dev_priv->mm.flushing_list) && - list_empty(&dev_priv->mm.active_list); - - return !lists_empty; -} - -static int i915_gem_inactive_shrink(struct shrinker *shrinker, struct shrink_control *sc) { struct drm_i915_private *dev_priv = @@ -4217,60 +4404,27 @@ i915_gem_inactive_shrink(struct shrinker *shrinker, struct shrink_control *sc) struct drm_i915_private, mm.inactive_shrinker); struct drm_device *dev = dev_priv->dev; - struct drm_i915_gem_object *obj, *next; + struct drm_i915_gem_object *obj; int nr_to_scan = sc->nr_to_scan; int cnt; if (!mutex_trylock(&dev->struct_mutex)) return 0; - /* "fast-path" to count number of available objects */ - if (nr_to_scan == 0) { - cnt = 0; - list_for_each_entry(obj, - &dev_priv->mm.inactive_list, - mm_list) - cnt++; - mutex_unlock(&dev->struct_mutex); - return cnt / 100 * sysctl_vfs_cache_pressure; - } - -rescan: - /* first scan for clean buffers */ - i915_gem_retire_requests(dev); - - list_for_each_entry_safe(obj, next, - &dev_priv->mm.inactive_list, - mm_list) { - if (i915_gem_object_is_purgeable(obj)) { - if (i915_gem_object_unbind(obj) == 0 && - --nr_to_scan == 0) - break; - } + if (nr_to_scan) { + nr_to_scan -= i915_gem_purge(dev_priv, nr_to_scan); + if (nr_to_scan > 0) + i915_gem_shrink_all(dev_priv); } - /* second pass, evict/count anything still on the inactive list */ cnt = 0; - list_for_each_entry_safe(obj, next, - &dev_priv->mm.inactive_list, - mm_list) { - if (nr_to_scan && - i915_gem_object_unbind(obj) == 0) - nr_to_scan--; - else - cnt++; - } + list_for_each_entry(obj, &dev_priv->mm.unbound_list, gtt_list) + if (obj->pages_pin_count == 0) + cnt += obj->base.size >> PAGE_SHIFT; + list_for_each_entry(obj, &dev_priv->mm.bound_list, gtt_list) + if (obj->pin_count == 0 && obj->pages_pin_count == 0) + cnt += obj->base.size >> PAGE_SHIFT; - if (nr_to_scan && i915_gpu_is_active(dev)) { - /* - * We are desperate for pages, so as a last resort, wait - * for the GPU to finish and discard whatever we can. - * This has a dramatic impact to reduce the number of - * OOM-killer events whilst running the GPU aggressively. - */ - if (i915_gpu_idle(dev) == 0) - goto rescan; - } mutex_unlock(&dev->struct_mutex); - return cnt / 100 * sysctl_vfs_cache_pressure; + return cnt; } diff --git a/drivers/gpu/drm/i915/i915_gem_context.c b/drivers/gpu/drm/i915/i915_gem_context.c index a9d58d72bb4d..4aa7ecf77ede 100644 --- a/drivers/gpu/drm/i915/i915_gem_context.c +++ b/drivers/gpu/drm/i915/i915_gem_context.c @@ -97,8 +97,7 @@ static struct i915_hw_context * i915_gem_context_get(struct drm_i915_file_private *file_priv, u32 id); -static int do_switch(struct drm_i915_gem_object *from_obj, - struct i915_hw_context *to, u32 seqno); +static int do_switch(struct i915_hw_context *to); static int get_context_size(struct drm_device *dev) { @@ -113,7 +112,10 @@ static int get_context_size(struct drm_device *dev) break; case 7: reg = I915_READ(GEN7_CXT_SIZE); - ret = GEN7_CXT_TOTAL_SIZE(reg) * 64; + if (IS_HASWELL(dev)) + ret = HSW_CXT_TOTAL_SIZE(reg) * 64; + else + ret = GEN7_CXT_TOTAL_SIZE(reg) * 64; break; default: BUG(); @@ -219,20 +221,21 @@ static int create_default_context(struct drm_i915_private *dev_priv) * default context. */ dev_priv->ring[RCS].default_context = ctx; - ret = i915_gem_object_pin(ctx->obj, CONTEXT_ALIGN, false); - if (ret) { - do_destroy(ctx); - return ret; - } + ret = i915_gem_object_pin(ctx->obj, CONTEXT_ALIGN, false, false); + if (ret) + goto err_destroy; - ret = do_switch(NULL, ctx, 0); - if (ret) { - i915_gem_object_unpin(ctx->obj); - do_destroy(ctx); - } else { - DRM_DEBUG_DRIVER("Default HW context loaded\n"); - } + ret = do_switch(ctx); + if (ret) + goto err_unpin; + DRM_DEBUG_DRIVER("Default HW context loaded\n"); + return 0; + +err_unpin: + i915_gem_object_unpin(ctx->obj); +err_destroy: + do_destroy(ctx); return ret; } @@ -359,18 +362,19 @@ mi_set_context(struct intel_ring_buffer *ring, return ret; } -static int do_switch(struct drm_i915_gem_object *from_obj, - struct i915_hw_context *to, - u32 seqno) +static int do_switch(struct i915_hw_context *to) { - struct intel_ring_buffer *ring = NULL; + struct intel_ring_buffer *ring = to->ring; + struct drm_i915_gem_object *from_obj = ring->last_context_obj; u32 hw_flags = 0; int ret; - BUG_ON(to == NULL); BUG_ON(from_obj != NULL && from_obj->pin_count == 0); - ret = i915_gem_object_pin(to->obj, CONTEXT_ALIGN, false); + if (from_obj == to->obj) + return 0; + + ret = i915_gem_object_pin(to->obj, CONTEXT_ALIGN, false, false); if (ret) return ret; @@ -393,7 +397,6 @@ static int do_switch(struct drm_i915_gem_object *from_obj, else if (WARN_ON_ONCE(from_obj == to->obj)) /* not yet expected */ hw_flags |= MI_FORCE_RESTORE; - ring = to->ring; ret = mi_set_context(ring, to, hw_flags); if (ret) { i915_gem_object_unpin(to->obj); @@ -407,6 +410,7 @@ static int do_switch(struct drm_i915_gem_object *from_obj, * MI_SET_CONTEXT instead of when the next seqno has completed. */ if (from_obj != NULL) { + u32 seqno = i915_gem_next_request_seqno(ring); from_obj->base.read_domains = I915_GEM_DOMAIN_INSTRUCTION; i915_gem_object_move_to_active(from_obj, ring, seqno); /* As long as MI_SET_CONTEXT is serializing, ie. it flushes the @@ -417,7 +421,7 @@ static int do_switch(struct drm_i915_gem_object *from_obj, * swapped, but there is no way to do that yet. */ from_obj->dirty = 1; - BUG_ON(from_obj->ring != to->ring); + BUG_ON(from_obj->ring != ring); i915_gem_object_unpin(from_obj); drm_gem_object_unreference(&from_obj->base); @@ -448,9 +452,7 @@ int i915_switch_context(struct intel_ring_buffer *ring, int to_id) { struct drm_i915_private *dev_priv = ring->dev->dev_private; - struct drm_i915_file_private *file_priv = NULL; struct i915_hw_context *to; - struct drm_i915_gem_object *from_obj = ring->last_context_obj; if (dev_priv->hw_contexts_disabled) return 0; @@ -458,21 +460,18 @@ int i915_switch_context(struct intel_ring_buffer *ring, if (ring != &dev_priv->ring[RCS]) return 0; - if (file) - file_priv = file->driver_priv; - if (to_id == DEFAULT_CONTEXT_ID) { to = ring->default_context; } else { - to = i915_gem_context_get(file_priv, to_id); + if (file == NULL) + return -EINVAL; + + to = i915_gem_context_get(file->driver_priv, to_id); if (to == NULL) return -ENOENT; } - if (from_obj == to->obj) - return 0; - - return do_switch(from_obj, to, i915_gem_next_request_seqno(to->ring)); + return do_switch(to); } int i915_gem_context_create_ioctl(struct drm_device *dev, void *data, diff --git a/drivers/gpu/drm/i915/i915_gem_dmabuf.c b/drivers/gpu/drm/i915/i915_gem_dmabuf.c index aa308e1337db..ca3497e1108c 100644 --- a/drivers/gpu/drm/i915/i915_gem_dmabuf.c +++ b/drivers/gpu/drm/i915/i915_gem_dmabuf.c @@ -28,35 +28,62 @@ #include <linux/dma-buf.h> static struct sg_table *i915_gem_map_dma_buf(struct dma_buf_attachment *attachment, - enum dma_data_direction dir) + enum dma_data_direction dir) { struct drm_i915_gem_object *obj = attachment->dmabuf->priv; - struct drm_device *dev = obj->base.dev; - int npages = obj->base.size / PAGE_SIZE; - struct sg_table *sg = NULL; - int ret; - int nents; + struct sg_table *st; + struct scatterlist *src, *dst; + int ret, i; - ret = i915_mutex_lock_interruptible(dev); + ret = i915_mutex_lock_interruptible(obj->base.dev); if (ret) return ERR_PTR(ret); - if (!obj->pages) { - ret = i915_gem_object_get_pages_gtt(obj, __GFP_NORETRY | __GFP_NOWARN); - if (ret) - goto out; + ret = i915_gem_object_get_pages(obj); + if (ret) { + st = ERR_PTR(ret); + goto out; + } + + /* Copy sg so that we make an independent mapping */ + st = kmalloc(sizeof(struct sg_table), GFP_KERNEL); + if (st == NULL) { + st = ERR_PTR(-ENOMEM); + goto out; } - /* link the pages into an SG then map the sg */ - sg = drm_prime_pages_to_sg(obj->pages, npages); - nents = dma_map_sg(attachment->dev, sg->sgl, sg->nents, dir); + ret = sg_alloc_table(st, obj->pages->nents, GFP_KERNEL); + if (ret) { + kfree(st); + st = ERR_PTR(ret); + goto out; + } + + src = obj->pages->sgl; + dst = st->sgl; + for (i = 0; i < obj->pages->nents; i++) { + sg_set_page(dst, sg_page(src), PAGE_SIZE, 0); + dst = sg_next(dst); + src = sg_next(src); + } + + if (!dma_map_sg(attachment->dev, st->sgl, st->nents, dir)) { + sg_free_table(st); + kfree(st); + st = ERR_PTR(-ENOMEM); + goto out; + } + + i915_gem_object_pin_pages(obj); + out: - mutex_unlock(&dev->struct_mutex); - return sg; + mutex_unlock(&obj->base.dev->struct_mutex); + return st; } static void i915_gem_unmap_dma_buf(struct dma_buf_attachment *attachment, - struct sg_table *sg, enum dma_data_direction dir) + struct sg_table *sg, + enum dma_data_direction dir) { dma_unmap_sg(attachment->dev, sg->sgl, sg->nents, dir); sg_free_table(sg); @@ -78,7 +105,9 @@ static void *i915_gem_dmabuf_vmap(struct dma_buf *dma_buf) { struct drm_i915_gem_object *obj = dma_buf->priv; struct drm_device *dev = obj->base.dev; - int ret; + struct scatterlist *sg; + struct page **pages; + int ret, i; ret = i915_mutex_lock_interruptible(dev); if (ret) @@ -89,24 +118,34 @@ static void *i915_gem_dmabuf_vmap(struct dma_buf *dma_buf) goto out_unlock; } - if (!obj->pages) { - ret = i915_gem_object_get_pages_gtt(obj, __GFP_NORETRY | __GFP_NOWARN); - if (ret) { - mutex_unlock(&dev->struct_mutex); - return ERR_PTR(ret); - } - } + ret = i915_gem_object_get_pages(obj); + if (ret) + goto error; - obj->dma_buf_vmapping = vmap(obj->pages, obj->base.size / PAGE_SIZE, 0, PAGE_KERNEL); - if (!obj->dma_buf_vmapping) { - DRM_ERROR("failed to vmap object\n"); - goto out_unlock; - } + ret = -ENOMEM; + + pages = drm_malloc_ab(obj->pages->nents, sizeof(struct page *)); + if (pages == NULL) + goto error; + + for_each_sg(obj->pages->sgl, sg, obj->pages->nents, i) + pages[i] = sg_page(sg); + + obj->dma_buf_vmapping = vmap(pages, obj->pages->nents, 0, PAGE_KERNEL); + drm_free_large(pages); + + if (!obj->dma_buf_vmapping) + goto error; obj->vmapping_count = 1; + i915_gem_object_pin_pages(obj); out_unlock: mutex_unlock(&dev->struct_mutex); return obj->dma_buf_vmapping; + +error: + mutex_unlock(&dev->struct_mutex); + return ERR_PTR(ret); } static void i915_gem_dmabuf_vunmap(struct dma_buf *dma_buf, void *vaddr) @@ -119,10 +158,11 @@ static void i915_gem_dmabuf_vunmap(struct dma_buf *dma_buf, void *vaddr) if (ret) return; - --obj->vmapping_count; - if (obj->vmapping_count == 0) { + if (--obj->vmapping_count == 0) { vunmap(obj->dma_buf_vmapping); obj->dma_buf_vmapping = NULL; + + i915_gem_object_unpin_pages(obj); } mutex_unlock(&dev->struct_mutex); } @@ -151,6 +191,22 @@ static int i915_gem_dmabuf_mmap(struct dma_buf *dma_buf, struct vm_area_struct * return -EINVAL; } +static int i915_gem_begin_cpu_access(struct dma_buf *dma_buf, size_t start, size_t length, enum dma_data_direction direction) +{ + struct drm_i915_gem_object *obj = dma_buf->priv; + struct drm_device *dev = obj->base.dev; + int ret; + bool write = (direction == DMA_BIDIRECTIONAL || direction == DMA_TO_DEVICE); + + ret = i915_mutex_lock_interruptible(dev); + if (ret) + return ret; + + ret = i915_gem_object_set_to_cpu_domain(obj, write); + mutex_unlock(&dev->struct_mutex); + return ret; +} + static const struct dma_buf_ops i915_dmabuf_ops = { .map_dma_buf = i915_gem_map_dma_buf, .unmap_dma_buf = i915_gem_unmap_dma_buf, @@ -162,25 +218,47 @@ static const struct dma_buf_ops i915_dmabuf_ops = { .mmap = i915_gem_dmabuf_mmap, .vmap = i915_gem_dmabuf_vmap, .vunmap = i915_gem_dmabuf_vunmap, + .begin_cpu_access = i915_gem_begin_cpu_access, }; struct dma_buf *i915_gem_prime_export(struct drm_device *dev, - struct drm_gem_object *gem_obj, int flags) + struct drm_gem_object *gem_obj, int flags) { struct drm_i915_gem_object *obj = to_intel_bo(gem_obj); - return dma_buf_export(obj, &i915_dmabuf_ops, - obj->base.size, 0600); + return dma_buf_export(obj, &i915_dmabuf_ops, obj->base.size, 0600); +} + +static int i915_gem_object_get_pages_dmabuf(struct drm_i915_gem_object *obj) +{ + struct sg_table *sg; + + sg = dma_buf_map_attachment(obj->base.import_attach, DMA_BIDIRECTIONAL); + if (IS_ERR(sg)) + return PTR_ERR(sg); + + obj->pages = sg; + obj->has_dma_mapping = true; + return 0; } +static void i915_gem_object_put_pages_dmabuf(struct drm_i915_gem_object *obj) +{ + dma_buf_unmap_attachment(obj->base.import_attach, + obj->pages, DMA_BIDIRECTIONAL); + obj->has_dma_mapping = false; +} + +static const struct drm_i915_gem_object_ops i915_gem_object_dmabuf_ops = { + .get_pages = i915_gem_object_get_pages_dmabuf, + .put_pages = i915_gem_object_put_pages_dmabuf, +}; + struct drm_gem_object *i915_gem_prime_import(struct drm_device *dev, - struct dma_buf *dma_buf) + struct dma_buf *dma_buf) { struct dma_buf_attachment *attach; - struct sg_table *sg; struct drm_i915_gem_object *obj; - int npages; - int size; int ret; /* is this one of own objects? */ @@ -198,34 +276,24 @@ struct drm_gem_object *i915_gem_prime_import(struct drm_device *dev, if (IS_ERR(attach)) return ERR_CAST(attach); - sg = dma_buf_map_attachment(attach, DMA_BIDIRECTIONAL); - if (IS_ERR(sg)) { - ret = PTR_ERR(sg); - goto fail_detach; - } - - size = dma_buf->size; - npages = size / PAGE_SIZE; obj = kzalloc(sizeof(*obj), GFP_KERNEL); if (obj == NULL) { ret = -ENOMEM; - goto fail_unmap; + goto fail_detach; } - ret = drm_gem_private_object_init(dev, &obj->base, size); + ret = drm_gem_private_object_init(dev, &obj->base, dma_buf->size); if (ret) { kfree(obj); - goto fail_unmap; + goto fail_detach; } - obj->sg_table = sg; + i915_gem_object_init(obj, &i915_gem_object_dmabuf_ops); obj->base.import_attach = attach; return &obj->base; -fail_unmap: - dma_buf_unmap_attachment(attach, sg, DMA_BIDIRECTIONAL); fail_detach: dma_buf_detach(dma_buf, attach); return ERR_PTR(ret); diff --git a/drivers/gpu/drm/i915/i915_gem_evict.c b/drivers/gpu/drm/i915/i915_gem_evict.c index eba0308f10e3..a2d8acde8550 100644 --- a/drivers/gpu/drm/i915/i915_gem_evict.c +++ b/drivers/gpu/drm/i915/i915_gem_evict.c @@ -44,7 +44,8 @@ mark_free(struct drm_i915_gem_object *obj, struct list_head *unwind) int i915_gem_evict_something(struct drm_device *dev, int min_size, - unsigned alignment, bool mappable) + unsigned alignment, unsigned cache_level, + bool mappable, bool nonblocking) { drm_i915_private_t *dev_priv = dev->dev_private; struct list_head eviction_list, unwind_list; @@ -79,11 +80,11 @@ i915_gem_evict_something(struct drm_device *dev, int min_size, INIT_LIST_HEAD(&unwind_list); if (mappable) drm_mm_init_scan_with_range(&dev_priv->mm.gtt_space, - min_size, alignment, 0, + min_size, alignment, cache_level, 0, dev_priv->mm.gtt_mappable_end); else drm_mm_init_scan(&dev_priv->mm.gtt_space, - min_size, alignment, 0); + min_size, alignment, cache_level); /* First see if there is a large enough contiguous idle region... */ list_for_each_entry(obj, &dev_priv->mm.inactive_list, mm_list) { @@ -91,29 +92,16 @@ i915_gem_evict_something(struct drm_device *dev, int min_size, goto found; } - /* Now merge in the soon-to-be-expired objects... */ - list_for_each_entry(obj, &dev_priv->mm.active_list, mm_list) { - /* Does the object require an outstanding flush? */ - if (obj->base.write_domain) - continue; - - if (mark_free(obj, &unwind_list)) - goto found; - } + if (nonblocking) + goto none; - /* Finally add anything with a pending flush (in order of retirement) */ - list_for_each_entry(obj, &dev_priv->mm.flushing_list, mm_list) { - if (mark_free(obj, &unwind_list)) - goto found; - } + /* Now merge in the soon-to-be-expired objects... */ list_for_each_entry(obj, &dev_priv->mm.active_list, mm_list) { - if (!obj->base.write_domain) - continue; - if (mark_free(obj, &unwind_list)) goto found; } +none: /* Nothing found, clean up and bail out! */ while (!list_empty(&unwind_list)) { obj = list_first_entry(&unwind_list, @@ -164,7 +152,7 @@ found: } int -i915_gem_evict_everything(struct drm_device *dev, bool purgeable_only) +i915_gem_evict_everything(struct drm_device *dev) { drm_i915_private_t *dev_priv = dev->dev_private; struct drm_i915_gem_object *obj, *next; @@ -172,12 +160,11 @@ i915_gem_evict_everything(struct drm_device *dev, bool purgeable_only) int ret; lists_empty = (list_empty(&dev_priv->mm.inactive_list) && - list_empty(&dev_priv->mm.flushing_list) && list_empty(&dev_priv->mm.active_list)); if (lists_empty) return -ENOSPC; - trace_i915_gem_evict_everything(dev, purgeable_only); + trace_i915_gem_evict_everything(dev); /* The gpu_idle will flush everything in the write domain to the * active list. Then we must move everything off the active list @@ -189,16 +176,11 @@ i915_gem_evict_everything(struct drm_device *dev, bool purgeable_only) i915_gem_retire_requests(dev); - BUG_ON(!list_empty(&dev_priv->mm.flushing_list)); - /* Having flushed everything, unbind() should never raise an error */ list_for_each_entry_safe(obj, next, - &dev_priv->mm.inactive_list, mm_list) { - if (!purgeable_only || obj->madv != I915_MADV_WILLNEED) { - if (obj->pin_count == 0) - WARN_ON(i915_gem_object_unbind(obj)); - } - } + &dev_priv->mm.inactive_list, mm_list) + if (obj->pin_count == 0) + WARN_ON(i915_gem_object_unbind(obj)); return 0; } diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c index ff2819ea0813..6a2f3e50c714 100644 --- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c +++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c @@ -34,180 +34,6 @@ #include "intel_drv.h" #include <linux/dma_remapping.h> -struct change_domains { - uint32_t invalidate_domains; - uint32_t flush_domains; - uint32_t flush_rings; - uint32_t flips; -}; - -/* - * Set the next domain for the specified object. This - * may not actually perform the necessary flushing/invaliding though, - * as that may want to be batched with other set_domain operations - * - * This is (we hope) the only really tricky part of gem. The goal - * is fairly simple -- track which caches hold bits of the object - * and make sure they remain coherent. A few concrete examples may - * help to explain how it works. For shorthand, we use the notation - * (read_domains, write_domain), e.g. (CPU, CPU) to indicate the - * a pair of read and write domain masks. - * - * Case 1: the batch buffer - * - * 1. Allocated - * 2. Written by CPU - * 3. Mapped to GTT - * 4. Read by GPU - * 5. Unmapped from GTT - * 6. Freed - * - * Let's take these a step at a time - * - * 1. Allocated - * Pages allocated from the kernel may still have - * cache contents, so we set them to (CPU, CPU) always. - * 2. Written by CPU (using pwrite) - * The pwrite function calls set_domain (CPU, CPU) and - * this function does nothing (as nothing changes) - * 3. Mapped by GTT - * This function asserts that the object is not - * currently in any GPU-based read or write domains - * 4. Read by GPU - * i915_gem_execbuffer calls set_domain (COMMAND, 0). - * As write_domain is zero, this function adds in the - * current read domains (CPU+COMMAND, 0). - * flush_domains is set to CPU. - * invalidate_domains is set to COMMAND - * clflush is run to get data out of the CPU caches - * then i915_dev_set_domain calls i915_gem_flush to - * emit an MI_FLUSH and drm_agp_chipset_flush - * 5. Unmapped from GTT - * i915_gem_object_unbind calls set_domain (CPU, CPU) - * flush_domains and invalidate_domains end up both zero - * so no flushing/invalidating happens - * 6. Freed - * yay, done - * - * Case 2: The shared render buffer - * - * 1. Allocated - * 2. Mapped to GTT - * 3. Read/written by GPU - * 4. set_domain to (CPU,CPU) - * 5. Read/written by CPU - * 6. Read/written by GPU - * - * 1. Allocated - * Same as last example, (CPU, CPU) - * 2. Mapped to GTT - * Nothing changes (assertions find that it is not in the GPU) - * 3. Read/written by GPU - * execbuffer calls set_domain (RENDER, RENDER) - * flush_domains gets CPU - * invalidate_domains gets GPU - * clflush (obj) - * MI_FLUSH and drm_agp_chipset_flush - * 4. set_domain (CPU, CPU) - * flush_domains gets GPU - * invalidate_domains gets CPU - * wait_rendering (obj) to make sure all drawing is complete. - * This will include an MI_FLUSH to get the data from GPU - * to memory - * clflush (obj) to invalidate the CPU cache - * Another MI_FLUSH in i915_gem_flush (eliminate this somehow?) - * 5. Read/written by CPU - * cache lines are loaded and dirtied - * 6. Read written by GPU - * Same as last GPU access - * - * Case 3: The constant buffer - * - * 1. Allocated - * 2. Written by CPU - * 3. Read by GPU - * 4. Updated (written) by CPU again - * 5. Read by GPU - * - * 1. Allocated - * (CPU, CPU) - * 2. Written by CPU - * (CPU, CPU) - * 3. Read by GPU - * (CPU+RENDER, 0) - * flush_domains = CPU - * invalidate_domains = RENDER - * clflush (obj) - * MI_FLUSH - * drm_agp_chipset_flush - * 4. Updated (written) by CPU again - * (CPU, CPU) - * flush_domains = 0 (no previous write domain) - * invalidate_domains = 0 (no new read domains) - * 5. Read by GPU - * (CPU+RENDER, 0) - * flush_domains = CPU - * invalidate_domains = RENDER - * clflush (obj) - * MI_FLUSH - * drm_agp_chipset_flush - */ -static void -i915_gem_object_set_to_gpu_domain(struct drm_i915_gem_object *obj, - struct intel_ring_buffer *ring, - struct change_domains *cd) -{ - uint32_t invalidate_domains = 0, flush_domains = 0; - - /* - * If the object isn't moving to a new write domain, - * let the object stay in multiple read domains - */ - if (obj->base.pending_write_domain == 0) - obj->base.pending_read_domains |= obj->base.read_domains; - - /* - * Flush the current write domain if - * the new read domains don't match. Invalidate - * any read domains which differ from the old - * write domain - */ - if (obj->base.write_domain && - (((obj->base.write_domain != obj->base.pending_read_domains || - obj->ring != ring)) || - (obj->fenced_gpu_access && !obj->pending_fenced_gpu_access))) { - flush_domains |= obj->base.write_domain; - invalidate_domains |= - obj->base.pending_read_domains & ~obj->base.write_domain; - } - /* - * Invalidate any read caches which may have - * stale data. That is, any new read domains. - */ - invalidate_domains |= obj->base.pending_read_domains & ~obj->base.read_domains; - if ((flush_domains | invalidate_domains) & I915_GEM_DOMAIN_CPU) - i915_gem_clflush_object(obj); - - if (obj->base.pending_write_domain) - cd->flips |= atomic_read(&obj->pending_flip); - - /* The actual obj->write_domain will be updated with - * pending_write_domain after we emit the accumulated flush for all - * of our domain changes in execbuffers (which clears objects' - * write_domains). So if we have a current write domain that we - * aren't changing, set pending_write_domain to that. - */ - if (flush_domains == 0 && obj->base.pending_write_domain == 0) - obj->base.pending_write_domain = obj->base.write_domain; - - cd->invalidate_domains |= invalidate_domains; - cd->flush_domains |= flush_domains; - if (flush_domains & I915_GEM_GPU_DOMAINS) - cd->flush_rings |= intel_ring_flag(obj->ring); - if (invalidate_domains & I915_GEM_GPU_DOMAINS) - cd->flush_rings |= intel_ring_flag(ring); -} - struct eb_objects { int and; struct hlist_head buckets[0]; @@ -218,6 +44,7 @@ eb_create(int size) { struct eb_objects *eb; int count = PAGE_SIZE / sizeof(struct hlist_head) / 2; + BUILD_BUG_ON(!is_power_of_2(PAGE_SIZE / sizeof(struct hlist_head))); while (count > size) count >>= 1; eb = kzalloc(count*sizeof(struct hlist_head) + @@ -269,6 +96,7 @@ eb_destroy(struct eb_objects *eb) static inline int use_cpu_reloc(struct drm_i915_gem_object *obj) { return (obj->base.write_domain == I915_GEM_DOMAIN_CPU || + !obj->map_and_fenceable || obj->cache_level != I915_CACHE_NONE); } @@ -383,7 +211,8 @@ i915_gem_execbuffer_relocate_entry(struct drm_i915_gem_object *obj, if (ret) return ret; - vaddr = kmap_atomic(obj->pages[reloc->offset >> PAGE_SHIFT]); + vaddr = kmap_atomic(i915_gem_object_get_page(obj, + reloc->offset >> PAGE_SHIFT)); *(uint32_t *)(vaddr + page_offset) = reloc->delta; kunmap_atomic(vaddr); } else { @@ -504,7 +333,8 @@ i915_gem_execbuffer_relocate(struct drm_device *dev, return ret; } -#define __EXEC_OBJECT_HAS_FENCE (1<<31) +#define __EXEC_OBJECT_HAS_PIN (1<<31) +#define __EXEC_OBJECT_HAS_FENCE (1<<30) static int need_reloc_mappable(struct drm_i915_gem_object *obj) @@ -514,9 +344,10 @@ need_reloc_mappable(struct drm_i915_gem_object *obj) } static int -pin_and_fence_object(struct drm_i915_gem_object *obj, - struct intel_ring_buffer *ring) +i915_gem_execbuffer_reserve_object(struct drm_i915_gem_object *obj, + struct intel_ring_buffer *ring) { + struct drm_i915_private *dev_priv = obj->base.dev->dev_private; struct drm_i915_gem_exec_object2 *entry = obj->exec_entry; bool has_fenced_gpu_access = INTEL_INFO(ring->dev)->gen < 4; bool need_fence, need_mappable; @@ -528,15 +359,17 @@ pin_and_fence_object(struct drm_i915_gem_object *obj, obj->tiling_mode != I915_TILING_NONE; need_mappable = need_fence || need_reloc_mappable(obj); - ret = i915_gem_object_pin(obj, entry->alignment, need_mappable); + ret = i915_gem_object_pin(obj, entry->alignment, need_mappable, false); if (ret) return ret; + entry->flags |= __EXEC_OBJECT_HAS_PIN; + if (has_fenced_gpu_access) { if (entry->flags & EXEC_OBJECT_NEEDS_FENCE) { ret = i915_gem_object_get_fence(obj); if (ret) - goto err_unpin; + return ret; if (i915_gem_object_pin_fence(obj)) entry->flags |= __EXEC_OBJECT_HAS_FENCE; @@ -545,12 +378,35 @@ pin_and_fence_object(struct drm_i915_gem_object *obj, } } + /* Ensure ppgtt mapping exists if needed */ + if (dev_priv->mm.aliasing_ppgtt && !obj->has_aliasing_ppgtt_mapping) { + i915_ppgtt_bind_object(dev_priv->mm.aliasing_ppgtt, + obj, obj->cache_level); + + obj->has_aliasing_ppgtt_mapping = 1; + } + entry->offset = obj->gtt_offset; return 0; +} -err_unpin: - i915_gem_object_unpin(obj); - return ret; +static void +i915_gem_execbuffer_unreserve_object(struct drm_i915_gem_object *obj) +{ + struct drm_i915_gem_exec_object2 *entry; + + if (!obj->gtt_space) + return; + + entry = obj->exec_entry; + + if (entry->flags & __EXEC_OBJECT_HAS_FENCE) + i915_gem_object_unpin_fence(obj); + + if (entry->flags & __EXEC_OBJECT_HAS_PIN) + i915_gem_object_unpin(obj); + + entry->flags &= ~(__EXEC_OBJECT_HAS_FENCE | __EXEC_OBJECT_HAS_PIN); } static int @@ -558,11 +414,10 @@ i915_gem_execbuffer_reserve(struct intel_ring_buffer *ring, struct drm_file *file, struct list_head *objects) { - drm_i915_private_t *dev_priv = ring->dev->dev_private; struct drm_i915_gem_object *obj; - int ret, retry; - bool has_fenced_gpu_access = INTEL_INFO(ring->dev)->gen < 4; struct list_head ordered_objects; + bool has_fenced_gpu_access = INTEL_INFO(ring->dev)->gen < 4; + int retry; INIT_LIST_HEAD(&ordered_objects); while (!list_empty(objects)) { @@ -587,6 +442,7 @@ i915_gem_execbuffer_reserve(struct intel_ring_buffer *ring, obj->base.pending_read_domains = 0; obj->base.pending_write_domain = 0; + obj->pending_fenced_gpu_access = false; } list_splice(&ordered_objects, objects); @@ -599,12 +455,12 @@ i915_gem_execbuffer_reserve(struct intel_ring_buffer *ring, * 2. Bind new objects. * 3. Decrement pin count. * - * This avoid unnecessary unbinding of later objects in order to makr + * This avoid unnecessary unbinding of later objects in order to make * room for the earlier objects *unless* we need to defragment. */ retry = 0; do { - ret = 0; + int ret = 0; /* Unbind any ill-fitting objects or pin. */ list_for_each_entry(obj, objects, exec_list) { @@ -624,7 +480,7 @@ i915_gem_execbuffer_reserve(struct intel_ring_buffer *ring, (need_mappable && !obj->map_and_fenceable)) ret = i915_gem_object_unbind(obj); else - ret = pin_and_fence_object(obj, ring); + ret = i915_gem_execbuffer_reserve_object(obj, ring); if (ret) goto err; } @@ -634,77 +490,22 @@ i915_gem_execbuffer_reserve(struct intel_ring_buffer *ring, if (obj->gtt_space) continue; - ret = pin_and_fence_object(obj, ring); - if (ret) { - int ret_ignore; - - /* This can potentially raise a harmless - * -EINVAL if we failed to bind in the above - * call. It cannot raise -EINTR since we know - * that the bo is freshly bound and so will - * not need to be flushed or waited upon. - */ - ret_ignore = i915_gem_object_unbind(obj); - (void)ret_ignore; - WARN_ON(obj->gtt_space); - break; - } + ret = i915_gem_execbuffer_reserve_object(obj, ring); + if (ret) + goto err; } - /* Decrement pin count for bound objects */ - list_for_each_entry(obj, objects, exec_list) { - struct drm_i915_gem_exec_object2 *entry; - - if (!obj->gtt_space) - continue; - - entry = obj->exec_entry; - if (entry->flags & __EXEC_OBJECT_HAS_FENCE) { - i915_gem_object_unpin_fence(obj); - entry->flags &= ~__EXEC_OBJECT_HAS_FENCE; - } - - i915_gem_object_unpin(obj); - - /* ... and ensure ppgtt mapping exist if needed. */ - if (dev_priv->mm.aliasing_ppgtt && !obj->has_aliasing_ppgtt_mapping) { - i915_ppgtt_bind_object(dev_priv->mm.aliasing_ppgtt, - obj, obj->cache_level); +err: /* Decrement pin count for bound objects */ + list_for_each_entry(obj, objects, exec_list) + i915_gem_execbuffer_unreserve_object(obj); - obj->has_aliasing_ppgtt_mapping = 1; - } - } - - if (ret != -ENOSPC || retry > 1) + if (ret != -ENOSPC || retry++) return ret; - /* First attempt, just clear anything that is purgeable. - * Second attempt, clear the entire GTT. - */ - ret = i915_gem_evict_everything(ring->dev, retry == 0); + ret = i915_gem_evict_everything(ring->dev); if (ret) return ret; - - retry++; } while (1); - -err: - list_for_each_entry_continue_reverse(obj, objects, exec_list) { - struct drm_i915_gem_exec_object2 *entry; - - if (!obj->gtt_space) - continue; - - entry = obj->exec_entry; - if (entry->flags & __EXEC_OBJECT_HAS_FENCE) { - i915_gem_object_unpin_fence(obj); - entry->flags &= ~__EXEC_OBJECT_HAS_FENCE; - } - - i915_gem_object_unpin(obj); - } - - return ret; } static int @@ -810,18 +611,6 @@ err: return ret; } -static void -i915_gem_execbuffer_flush(struct drm_device *dev, - uint32_t invalidate_domains, - uint32_t flush_domains) -{ - if (flush_domains & I915_GEM_DOMAIN_CPU) - intel_gtt_chipset_flush(); - - if (flush_domains & I915_GEM_DOMAIN_GTT) - wmb(); -} - static int i915_gem_execbuffer_wait_for_flips(struct intel_ring_buffer *ring, u32 flips) { @@ -854,48 +643,45 @@ i915_gem_execbuffer_wait_for_flips(struct intel_ring_buffer *ring, u32 flips) return 0; } - static int i915_gem_execbuffer_move_to_gpu(struct intel_ring_buffer *ring, struct list_head *objects) { struct drm_i915_gem_object *obj; - struct change_domains cd; + uint32_t flush_domains = 0; + uint32_t flips = 0; int ret; - memset(&cd, 0, sizeof(cd)); - list_for_each_entry(obj, objects, exec_list) - i915_gem_object_set_to_gpu_domain(obj, ring, &cd); - - if (cd.invalidate_domains | cd.flush_domains) { - i915_gem_execbuffer_flush(ring->dev, - cd.invalidate_domains, - cd.flush_domains); - } - - if (cd.flips) { - ret = i915_gem_execbuffer_wait_for_flips(ring, cd.flips); + list_for_each_entry(obj, objects, exec_list) { + ret = i915_gem_object_sync(obj, ring); if (ret) return ret; + + if (obj->base.write_domain & I915_GEM_DOMAIN_CPU) + i915_gem_clflush_object(obj); + + if (obj->base.pending_write_domain) + flips |= atomic_read(&obj->pending_flip); + + flush_domains |= obj->base.write_domain; } - list_for_each_entry(obj, objects, exec_list) { - ret = i915_gem_object_sync(obj, ring); + if (flips) { + ret = i915_gem_execbuffer_wait_for_flips(ring, flips); if (ret) return ret; } + if (flush_domains & I915_GEM_DOMAIN_CPU) + intel_gtt_chipset_flush(); + + if (flush_domains & I915_GEM_DOMAIN_GTT) + wmb(); + /* Unconditionally invalidate gpu caches and ensure that we do flush * any residual writes from the previous batch. */ - ret = i915_gem_flush_ring(ring, - I915_GEM_GPU_DOMAINS, - ring->gpu_caches_dirty ? I915_GEM_GPU_DOMAINS : 0); - if (ret) - return ret; - - ring->gpu_caches_dirty = false; - return 0; + return intel_ring_invalidate_all_caches(ring); } static bool @@ -943,9 +729,8 @@ i915_gem_execbuffer_move_to_active(struct list_head *objects, struct drm_i915_gem_object *obj; list_for_each_entry(obj, objects, exec_list) { - u32 old_read = obj->base.read_domains; - u32 old_write = obj->base.write_domain; - + u32 old_read = obj->base.read_domains; + u32 old_write = obj->base.write_domain; obj->base.read_domains = obj->base.pending_read_domains; obj->base.write_domain = obj->base.pending_write_domain; @@ -954,17 +739,13 @@ i915_gem_execbuffer_move_to_active(struct list_head *objects, i915_gem_object_move_to_active(obj, ring, seqno); if (obj->base.write_domain) { obj->dirty = 1; - obj->pending_gpu_write = true; - list_move_tail(&obj->gpu_write_list, - &ring->gpu_write_list); + obj->last_write_seqno = seqno; if (obj->pin_count) /* check for potential scanout */ - intel_mark_busy(ring->dev, obj); + intel_mark_fb_busy(obj); } trace_i915_gem_object_change_domain(obj, old_read, old_write); } - - intel_mark_busy(ring->dev, NULL); } static void @@ -972,16 +753,11 @@ i915_gem_execbuffer_retire_commands(struct drm_device *dev, struct drm_file *file, struct intel_ring_buffer *ring) { - struct drm_i915_gem_request *request; - /* Unconditionally force add_request to emit a full flush. */ ring->gpu_caches_dirty = true; /* Add a breadcrumb for the completion of the batch buffer */ - request = kzalloc(sizeof(*request), GFP_KERNEL); - if (request == NULL || i915_add_request(ring, file, request)) { - kfree(request); - } + (void)i915_add_request(ring, file, NULL); } static int @@ -1327,8 +1103,7 @@ i915_gem_execbuffer(struct drm_device *dev, void *data, return -ENOMEM; } ret = copy_from_user(exec_list, - (struct drm_i915_relocation_entry __user *) - (uintptr_t) args->buffers_ptr, + (void __user *)(uintptr_t)args->buffers_ptr, sizeof(*exec_list) * args->buffer_count); if (ret != 0) { DRM_DEBUG("copy %d exec entries failed %d\n", @@ -1367,8 +1142,7 @@ i915_gem_execbuffer(struct drm_device *dev, void *data, for (i = 0; i < args->buffer_count; i++) exec_list[i].offset = exec2_list[i].offset; /* ... and back out to userspace */ - ret = copy_to_user((struct drm_i915_relocation_entry __user *) - (uintptr_t) args->buffers_ptr, + ret = copy_to_user((void __user *)(uintptr_t)args->buffers_ptr, exec_list, sizeof(*exec_list) * args->buffer_count); if (ret) { @@ -1422,8 +1196,7 @@ i915_gem_execbuffer2(struct drm_device *dev, void *data, ret = i915_gem_do_execbuffer(dev, data, file, args, exec2_list); if (!ret) { /* Copy the new buffer offsets back to the user's exec list. */ - ret = copy_to_user((struct drm_i915_relocation_entry __user *) - (uintptr_t) args->buffers_ptr, + ret = copy_to_user((void __user *)(uintptr_t)args->buffers_ptr, exec2_list, sizeof(*exec2_list) * args->buffer_count); if (ret) { diff --git a/drivers/gpu/drm/i915/i915_gem_gtt.c b/drivers/gpu/drm/i915/i915_gem_gtt.c index 60815b861ec2..47e427e4bc67 100644 --- a/drivers/gpu/drm/i915/i915_gem_gtt.c +++ b/drivers/gpu/drm/i915/i915_gem_gtt.c @@ -167,8 +167,7 @@ void i915_gem_cleanup_aliasing_ppgtt(struct drm_device *dev) } static void i915_ppgtt_insert_sg_entries(struct i915_hw_ppgtt *ppgtt, - struct scatterlist *sg_list, - unsigned sg_len, + const struct sg_table *pages, unsigned first_entry, uint32_t pte_flags) { @@ -180,12 +179,12 @@ static void i915_ppgtt_insert_sg_entries(struct i915_hw_ppgtt *ppgtt, struct scatterlist *sg; /* init sg walking */ - sg = sg_list; + sg = pages->sgl; i = 0; segment_len = sg_dma_len(sg) >> PAGE_SHIFT; m = 0; - while (i < sg_len) { + while (i < pages->nents) { pt_vaddr = kmap_atomic(ppgtt->pt_pages[act_pd]); for (j = first_pte; j < I915_PPGTT_PT_ENTRIES; j++) { @@ -194,13 +193,11 @@ static void i915_ppgtt_insert_sg_entries(struct i915_hw_ppgtt *ppgtt, pt_vaddr[j] = pte | pte_flags; /* grab the next page */ - m++; - if (m == segment_len) { - sg = sg_next(sg); - i++; - if (i == sg_len) + if (++m == segment_len) { + if (++i == pages->nents) break; + sg = sg_next(sg); segment_len = sg_dma_len(sg) >> PAGE_SHIFT; m = 0; } @@ -213,44 +210,10 @@ static void i915_ppgtt_insert_sg_entries(struct i915_hw_ppgtt *ppgtt, } } -static void i915_ppgtt_insert_pages(struct i915_hw_ppgtt *ppgtt, - unsigned first_entry, unsigned num_entries, - struct page **pages, uint32_t pte_flags) -{ - uint32_t *pt_vaddr, pte; - unsigned act_pd = first_entry / I915_PPGTT_PT_ENTRIES; - unsigned first_pte = first_entry % I915_PPGTT_PT_ENTRIES; - unsigned last_pte, i; - dma_addr_t page_addr; - - while (num_entries) { - last_pte = first_pte + num_entries; - last_pte = min_t(unsigned, last_pte, I915_PPGTT_PT_ENTRIES); - - pt_vaddr = kmap_atomic(ppgtt->pt_pages[act_pd]); - - for (i = first_pte; i < last_pte; i++) { - page_addr = page_to_phys(*pages); - pte = GEN6_PTE_ADDR_ENCODE(page_addr); - pt_vaddr[i] = pte | pte_flags; - - pages++; - } - - kunmap_atomic(pt_vaddr); - - num_entries -= last_pte - first_pte; - first_pte = 0; - act_pd++; - } -} - void i915_ppgtt_bind_object(struct i915_hw_ppgtt *ppgtt, struct drm_i915_gem_object *obj, enum i915_cache_level cache_level) { - struct drm_device *dev = obj->base.dev; - struct drm_i915_private *dev_priv = dev->dev_private; uint32_t pte_flags = GEN6_PTE_VALID; switch (cache_level) { @@ -261,7 +224,7 @@ void i915_ppgtt_bind_object(struct i915_hw_ppgtt *ppgtt, pte_flags |= GEN6_PTE_CACHE_LLC; break; case I915_CACHE_NONE: - if (IS_HASWELL(dev)) + if (IS_HASWELL(obj->base.dev)) pte_flags |= HSW_PTE_UNCACHED; else pte_flags |= GEN6_PTE_UNCACHED; @@ -270,26 +233,10 @@ void i915_ppgtt_bind_object(struct i915_hw_ppgtt *ppgtt, BUG(); } - if (obj->sg_table) { - i915_ppgtt_insert_sg_entries(ppgtt, - obj->sg_table->sgl, - obj->sg_table->nents, - obj->gtt_space->start >> PAGE_SHIFT, - pte_flags); - } else if (dev_priv->mm.gtt->needs_dmar) { - BUG_ON(!obj->sg_list); - - i915_ppgtt_insert_sg_entries(ppgtt, - obj->sg_list, - obj->num_sg, - obj->gtt_space->start >> PAGE_SHIFT, - pte_flags); - } else - i915_ppgtt_insert_pages(ppgtt, - obj->gtt_space->start >> PAGE_SHIFT, - obj->base.size >> PAGE_SHIFT, - obj->pages, - pte_flags); + i915_ppgtt_insert_sg_entries(ppgtt, + obj->pages, + obj->gtt_space->start >> PAGE_SHIFT, + pte_flags); } void i915_ppgtt_unbind_object(struct i915_hw_ppgtt *ppgtt, @@ -351,7 +298,7 @@ void i915_gem_restore_gtt_mappings(struct drm_device *dev) intel_gtt_clear_range(dev_priv->mm.gtt_start / PAGE_SIZE, (dev_priv->mm.gtt_end - dev_priv->mm.gtt_start) / PAGE_SIZE); - list_for_each_entry(obj, &dev_priv->mm.gtt_list, gtt_list) { + list_for_each_entry(obj, &dev_priv->mm.bound_list, gtt_list) { i915_gem_clflush_object(obj); i915_gem_gtt_bind_object(obj, obj->cache_level); } @@ -361,44 +308,26 @@ void i915_gem_restore_gtt_mappings(struct drm_device *dev) int i915_gem_gtt_prepare_object(struct drm_i915_gem_object *obj) { - struct drm_device *dev = obj->base.dev; - struct drm_i915_private *dev_priv = dev->dev_private; - - /* don't map imported dma buf objects */ - if (dev_priv->mm.gtt->needs_dmar && !obj->sg_table) - return intel_gtt_map_memory(obj->pages, - obj->base.size >> PAGE_SHIFT, - &obj->sg_list, - &obj->num_sg); - else + if (obj->has_dma_mapping) return 0; + + if (!dma_map_sg(&obj->base.dev->pdev->dev, + obj->pages->sgl, obj->pages->nents, + PCI_DMA_BIDIRECTIONAL)) + return -ENOSPC; + + return 0; } void i915_gem_gtt_bind_object(struct drm_i915_gem_object *obj, enum i915_cache_level cache_level) { struct drm_device *dev = obj->base.dev; - struct drm_i915_private *dev_priv = dev->dev_private; unsigned int agp_type = cache_level_to_agp_type(dev, cache_level); - if (obj->sg_table) { - intel_gtt_insert_sg_entries(obj->sg_table->sgl, - obj->sg_table->nents, - obj->gtt_space->start >> PAGE_SHIFT, - agp_type); - } else if (dev_priv->mm.gtt->needs_dmar) { - BUG_ON(!obj->sg_list); - - intel_gtt_insert_sg_entries(obj->sg_list, - obj->num_sg, - obj->gtt_space->start >> PAGE_SHIFT, - agp_type); - } else - intel_gtt_insert_pages(obj->gtt_space->start >> PAGE_SHIFT, - obj->base.size >> PAGE_SHIFT, - obj->pages, - agp_type); - + intel_gtt_insert_sg_entries(obj->pages, + obj->gtt_space->start >> PAGE_SHIFT, + agp_type); obj->has_global_gtt_mapping = 1; } @@ -418,14 +347,31 @@ void i915_gem_gtt_finish_object(struct drm_i915_gem_object *obj) interruptible = do_idling(dev_priv); - if (obj->sg_list) { - intel_gtt_unmap_memory(obj->sg_list, obj->num_sg); - obj->sg_list = NULL; - } + if (!obj->has_dma_mapping) + dma_unmap_sg(&dev->pdev->dev, + obj->pages->sgl, obj->pages->nents, + PCI_DMA_BIDIRECTIONAL); undo_idling(dev_priv, interruptible); } +static void i915_gtt_color_adjust(struct drm_mm_node *node, + unsigned long color, + unsigned long *start, + unsigned long *end) +{ + if (node->color != color) + *start += 4096; + + if (!list_empty(&node->node_list)) { + node = list_entry(node->node_list.next, + struct drm_mm_node, + node_list); + if (node->allocated && node->color != color) + *end -= 4096; + } +} + void i915_gem_init_global_gtt(struct drm_device *dev, unsigned long start, unsigned long mappable_end, @@ -435,6 +381,8 @@ void i915_gem_init_global_gtt(struct drm_device *dev, /* Substract the guard page ... */ drm_mm_init(&dev_priv->mm.gtt_space, start, end - start - PAGE_SIZE); + if (!HAS_LLC(dev)) + dev_priv->mm.gtt_space.color_adjust = i915_gtt_color_adjust; dev_priv->mm.gtt_start = start; dev_priv->mm.gtt_mappable_end = mappable_end; diff --git a/drivers/gpu/drm/i915/i915_gem_tiling.c b/drivers/gpu/drm/i915/i915_gem_tiling.c index b964df51cec7..8093ecd2ea31 100644 --- a/drivers/gpu/drm/i915/i915_gem_tiling.c +++ b/drivers/gpu/drm/i915/i915_gem_tiling.c @@ -470,18 +470,20 @@ i915_gem_swizzle_page(struct page *page) void i915_gem_object_do_bit_17_swizzle(struct drm_i915_gem_object *obj) { + struct scatterlist *sg; int page_count = obj->base.size >> PAGE_SHIFT; int i; if (obj->bit_17 == NULL) return; - for (i = 0; i < page_count; i++) { - char new_bit_17 = page_to_phys(obj->pages[i]) >> 17; + for_each_sg(obj->pages->sgl, sg, page_count, i) { + struct page *page = sg_page(sg); + char new_bit_17 = page_to_phys(page) >> 17; if ((new_bit_17 & 0x1) != (test_bit(i, obj->bit_17) != 0)) { - i915_gem_swizzle_page(obj->pages[i]); - set_page_dirty(obj->pages[i]); + i915_gem_swizzle_page(page); + set_page_dirty(page); } } } @@ -489,6 +491,7 @@ i915_gem_object_do_bit_17_swizzle(struct drm_i915_gem_object *obj) void i915_gem_object_save_bit_17_swizzle(struct drm_i915_gem_object *obj) { + struct scatterlist *sg; int page_count = obj->base.size >> PAGE_SHIFT; int i; @@ -502,8 +505,9 @@ i915_gem_object_save_bit_17_swizzle(struct drm_i915_gem_object *obj) } } - for (i = 0; i < page_count; i++) { - if (page_to_phys(obj->pages[i]) & (1 << 17)) + for_each_sg(obj->pages->sgl, sg, page_count, i) { + struct page *page = sg_page(sg); + if (page_to_phys(page) & (1 << 17)) __set_bit(i, obj->bit_17); else __clear_bit(i, obj->bit_17); diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c index 5249640cce13..d7f0066538a9 100644 --- a/drivers/gpu/drm/i915/i915_irq.c +++ b/drivers/gpu/drm/i915/i915_irq.c @@ -296,11 +296,21 @@ static void i915_hotplug_work_func(struct work_struct *work) drm_helper_hpd_irq_event(dev); } -static void i915_handle_rps_change(struct drm_device *dev) +/* defined intel_pm.c */ +extern spinlock_t mchdev_lock; + +static void ironlake_handle_rps_change(struct drm_device *dev) { drm_i915_private_t *dev_priv = dev->dev_private; u32 busy_up, busy_down, max_avg, min_avg; - u8 new_delay = dev_priv->cur_delay; + u8 new_delay; + unsigned long flags; + + spin_lock_irqsave(&mchdev_lock, flags); + + I915_WRITE16(MEMINTRSTS, I915_READ(MEMINTRSTS)); + + new_delay = dev_priv->ips.cur_delay; I915_WRITE16(MEMINTRSTS, MEMINT_EVAL_CHG); busy_up = I915_READ(RCPREVBSYTUPAVG); @@ -310,19 +320,21 @@ static void i915_handle_rps_change(struct drm_device *dev) /* Handle RCS change request from hw */ if (busy_up > max_avg) { - if (dev_priv->cur_delay != dev_priv->max_delay) - new_delay = dev_priv->cur_delay - 1; - if (new_delay < dev_priv->max_delay) - new_delay = dev_priv->max_delay; + if (dev_priv->ips.cur_delay != dev_priv->ips.max_delay) + new_delay = dev_priv->ips.cur_delay - 1; + if (new_delay < dev_priv->ips.max_delay) + new_delay = dev_priv->ips.max_delay; } else if (busy_down < min_avg) { - if (dev_priv->cur_delay != dev_priv->min_delay) - new_delay = dev_priv->cur_delay + 1; - if (new_delay > dev_priv->min_delay) - new_delay = dev_priv->min_delay; + if (dev_priv->ips.cur_delay != dev_priv->ips.min_delay) + new_delay = dev_priv->ips.cur_delay + 1; + if (new_delay > dev_priv->ips.min_delay) + new_delay = dev_priv->ips.min_delay; } if (ironlake_set_drps(dev, new_delay)) - dev_priv->cur_delay = new_delay; + dev_priv->ips.cur_delay = new_delay; + + spin_unlock_irqrestore(&mchdev_lock, flags); return; } @@ -335,7 +347,7 @@ static void notify_ring(struct drm_device *dev, if (ring->obj == NULL) return; - trace_i915_gem_request_complete(ring, ring->get_seqno(ring)); + trace_i915_gem_request_complete(ring, ring->get_seqno(ring, false)); wake_up_all(&ring->irq_queue); if (i915_enable_hangcheck) { @@ -349,16 +361,16 @@ static void notify_ring(struct drm_device *dev, static void gen6_pm_rps_work(struct work_struct *work) { drm_i915_private_t *dev_priv = container_of(work, drm_i915_private_t, - rps_work); + rps.work); u32 pm_iir, pm_imr; u8 new_delay; - spin_lock_irq(&dev_priv->rps_lock); - pm_iir = dev_priv->pm_iir; - dev_priv->pm_iir = 0; + spin_lock_irq(&dev_priv->rps.lock); + pm_iir = dev_priv->rps.pm_iir; + dev_priv->rps.pm_iir = 0; pm_imr = I915_READ(GEN6_PMIMR); I915_WRITE(GEN6_PMIMR, 0); - spin_unlock_irq(&dev_priv->rps_lock); + spin_unlock_irq(&dev_priv->rps.lock); if ((pm_iir & GEN6_PM_DEFERRED_EVENTS) == 0) return; @@ -366,11 +378,17 @@ static void gen6_pm_rps_work(struct work_struct *work) mutex_lock(&dev_priv->dev->struct_mutex); if (pm_iir & GEN6_PM_RP_UP_THRESHOLD) - new_delay = dev_priv->cur_delay + 1; + new_delay = dev_priv->rps.cur_delay + 1; else - new_delay = dev_priv->cur_delay - 1; + new_delay = dev_priv->rps.cur_delay - 1; - gen6_set_rps(dev_priv->dev, new_delay); + /* sysfs frequency interfaces may have snuck in while servicing the + * interrupt + */ + if (!(new_delay > dev_priv->rps.max_delay || + new_delay < dev_priv->rps.min_delay)) { + gen6_set_rps(dev_priv->dev, new_delay); + } mutex_unlock(&dev_priv->dev->struct_mutex); } @@ -444,7 +462,7 @@ static void ivybridge_handle_parity_error(struct drm_device *dev) drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private; unsigned long flags; - if (!IS_IVYBRIDGE(dev)) + if (!HAS_L3_GPU_CACHE(dev)) return; spin_lock_irqsave(&dev_priv->irq_lock, flags); @@ -488,19 +506,19 @@ static void gen6_queue_rps_work(struct drm_i915_private *dev_priv, * IIR bits should never already be set because IMR should * prevent an interrupt from being shown in IIR. The warning * displays a case where we've unsafely cleared - * dev_priv->pm_iir. Although missing an interrupt of the same + * dev_priv->rps.pm_iir. Although missing an interrupt of the same * type is not a problem, it displays a problem in the logic. * - * The mask bit in IMR is cleared by rps_work. + * The mask bit in IMR is cleared by dev_priv->rps.work. */ - spin_lock_irqsave(&dev_priv->rps_lock, flags); - dev_priv->pm_iir |= pm_iir; - I915_WRITE(GEN6_PMIMR, dev_priv->pm_iir); + spin_lock_irqsave(&dev_priv->rps.lock, flags); + dev_priv->rps.pm_iir |= pm_iir; + I915_WRITE(GEN6_PMIMR, dev_priv->rps.pm_iir); POSTING_READ(GEN6_PMIMR); - spin_unlock_irqrestore(&dev_priv->rps_lock, flags); + spin_unlock_irqrestore(&dev_priv->rps.lock, flags); - queue_work(dev_priv->wq, &dev_priv->rps_work); + queue_work(dev_priv->wq, &dev_priv->rps.work); } static irqreturn_t valleyview_irq_handler(DRM_IRQ_ARGS) @@ -793,10 +811,8 @@ static irqreturn_t ironlake_irq_handler(DRM_IRQ_ARGS) ibx_irq_handler(dev, pch_iir); } - if (de_iir & DE_PCU_EVENT) { - I915_WRITE16(MEMINTRSTS, I915_READ(MEMINTRSTS)); - i915_handle_rps_change(dev); - } + if (IS_GEN5(dev) && de_iir & DE_PCU_EVENT) + ironlake_handle_rps_change(dev); if (IS_GEN6(dev) && pm_iir & GEN6_PM_DEFERRED_EVENTS) gen6_queue_rps_work(dev_priv, pm_iir); @@ -843,26 +859,55 @@ static void i915_error_work_func(struct work_struct *work) } } +/* NB: please notice the memset */ +static void i915_get_extra_instdone(struct drm_device *dev, + uint32_t *instdone) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + memset(instdone, 0, sizeof(*instdone) * I915_NUM_INSTDONE_REG); + + switch(INTEL_INFO(dev)->gen) { + case 2: + case 3: + instdone[0] = I915_READ(INSTDONE); + break; + case 4: + case 5: + case 6: + instdone[0] = I915_READ(INSTDONE_I965); + instdone[1] = I915_READ(INSTDONE1); + break; + default: + WARN_ONCE(1, "Unsupported platform\n"); + case 7: + instdone[0] = I915_READ(GEN7_INSTDONE_1); + instdone[1] = I915_READ(GEN7_SC_INSTDONE); + instdone[2] = I915_READ(GEN7_SAMPLER_INSTDONE); + instdone[3] = I915_READ(GEN7_ROW_INSTDONE); + break; + } +} + #ifdef CONFIG_DEBUG_FS static struct drm_i915_error_object * i915_error_object_create(struct drm_i915_private *dev_priv, struct drm_i915_gem_object *src) { struct drm_i915_error_object *dst; - int page, page_count; + int i, count; u32 reloc_offset; if (src == NULL || src->pages == NULL) return NULL; - page_count = src->base.size / PAGE_SIZE; + count = src->base.size / PAGE_SIZE; - dst = kmalloc(sizeof(*dst) + page_count * sizeof(u32 *), GFP_ATOMIC); + dst = kmalloc(sizeof(*dst) + count * sizeof(u32 *), GFP_ATOMIC); if (dst == NULL) return NULL; reloc_offset = src->gtt_offset; - for (page = 0; page < page_count; page++) { + for (i = 0; i < count; i++) { unsigned long flags; void *d; @@ -885,30 +930,33 @@ i915_error_object_create(struct drm_i915_private *dev_priv, memcpy_fromio(d, s, PAGE_SIZE); io_mapping_unmap_atomic(s); } else { + struct page *page; void *s; - drm_clflush_pages(&src->pages[page], 1); + page = i915_gem_object_get_page(src, i); + + drm_clflush_pages(&page, 1); - s = kmap_atomic(src->pages[page]); + s = kmap_atomic(page); memcpy(d, s, PAGE_SIZE); kunmap_atomic(s); - drm_clflush_pages(&src->pages[page], 1); + drm_clflush_pages(&page, 1); } local_irq_restore(flags); - dst->pages[page] = d; + dst->pages[i] = d; reloc_offset += PAGE_SIZE; } - dst->page_count = page_count; + dst->page_count = count; dst->gtt_offset = src->gtt_offset; return dst; unwind: - while (page--) - kfree(dst->pages[page]); + while (i--) + kfree(dst->pages[i]); kfree(dst); return NULL; } @@ -949,7 +997,8 @@ static void capture_bo(struct drm_i915_error_buffer *err, { err->size = obj->base.size; err->name = obj->base.name; - err->seqno = obj->last_rendering_seqno; + err->rseqno = obj->last_read_seqno; + err->wseqno = obj->last_write_seqno; err->gtt_offset = obj->gtt_offset; err->read_domains = obj->base.read_domains; err->write_domain = obj->base.write_domain; @@ -1039,12 +1088,12 @@ i915_error_first_batchbuffer(struct drm_i915_private *dev_priv, if (!ring->get_seqno) return NULL; - seqno = ring->get_seqno(ring); + seqno = ring->get_seqno(ring, false); list_for_each_entry(obj, &dev_priv->mm.active_list, mm_list) { if (obj->ring != ring) continue; - if (i915_seqno_passed(seqno, obj->last_rendering_seqno)) + if (i915_seqno_passed(seqno, obj->last_read_seqno)) continue; if ((obj->base.read_domains & I915_GEM_DOMAIN_COMMAND) == 0) @@ -1080,10 +1129,8 @@ static void i915_record_ring_state(struct drm_device *dev, error->ipehr[ring->id] = I915_READ(RING_IPEHR(ring->mmio_base)); error->instdone[ring->id] = I915_READ(RING_INSTDONE(ring->mmio_base)); error->instps[ring->id] = I915_READ(RING_INSTPS(ring->mmio_base)); - if (ring->id == RCS) { - error->instdone1 = I915_READ(INSTDONE1); + if (ring->id == RCS) error->bbaddr = I915_READ64(BB_ADDR); - } } else { error->faddr[ring->id] = I915_READ(DMA_FADD_I8XX); error->ipeir[ring->id] = I915_READ(IPEIR); @@ -1093,7 +1140,7 @@ static void i915_record_ring_state(struct drm_device *dev, error->waiting[ring->id] = waitqueue_active(&ring->irq_queue); error->instpm[ring->id] = I915_READ(RING_INSTPM(ring->mmio_base)); - error->seqno[ring->id] = ring->get_seqno(ring); + error->seqno[ring->id] = ring->get_seqno(ring, false); error->acthd[ring->id] = intel_ring_get_active_head(ring); error->head[ring->id] = I915_READ_HEAD(ring); error->tail[ring->id] = I915_READ_TAIL(ring); @@ -1199,6 +1246,11 @@ static void i915_capture_error_state(struct drm_device *dev) error->done_reg = I915_READ(DONE_REG); } + if (INTEL_INFO(dev)->gen == 7) + error->err_int = I915_READ(GEN7_ERR_INT); + + i915_get_extra_instdone(dev, error->extra_instdone); + i915_gem_record_fences(dev, error); i915_gem_record_rings(dev, error); @@ -1210,7 +1262,7 @@ static void i915_capture_error_state(struct drm_device *dev) list_for_each_entry(obj, &dev_priv->mm.active_list, mm_list) i++; error->active_bo_count = i; - list_for_each_entry(obj, &dev_priv->mm.gtt_list, gtt_list) + list_for_each_entry(obj, &dev_priv->mm.bound_list, gtt_list) if (obj->pin_count) i++; error->pinned_bo_count = i - error->active_bo_count; @@ -1235,7 +1287,7 @@ static void i915_capture_error_state(struct drm_device *dev) error->pinned_bo_count = capture_pinned_bo(error->pinned_bo, error->pinned_bo_count, - &dev_priv->mm.gtt_list); + &dev_priv->mm.bound_list); do_gettimeofday(&error->time); @@ -1274,24 +1326,26 @@ void i915_destroy_error_state(struct drm_device *dev) static void i915_report_and_clear_eir(struct drm_device *dev) { struct drm_i915_private *dev_priv = dev->dev_private; + uint32_t instdone[I915_NUM_INSTDONE_REG]; u32 eir = I915_READ(EIR); - int pipe; + int pipe, i; if (!eir) return; pr_err("render error detected, EIR: 0x%08x\n", eir); + i915_get_extra_instdone(dev, instdone); + if (IS_G4X(dev)) { if (eir & (GM45_ERROR_MEM_PRIV | GM45_ERROR_CP_PRIV)) { u32 ipeir = I915_READ(IPEIR_I965); pr_err(" IPEIR: 0x%08x\n", I915_READ(IPEIR_I965)); pr_err(" IPEHR: 0x%08x\n", I915_READ(IPEHR_I965)); - pr_err(" INSTDONE: 0x%08x\n", - I915_READ(INSTDONE_I965)); + for (i = 0; i < ARRAY_SIZE(instdone); i++) + pr_err(" INSTDONE_%d: 0x%08x\n", i, instdone[i]); pr_err(" INSTPS: 0x%08x\n", I915_READ(INSTPS)); - pr_err(" INSTDONE1: 0x%08x\n", I915_READ(INSTDONE1)); pr_err(" ACTHD: 0x%08x\n", I915_READ(ACTHD_I965)); I915_WRITE(IPEIR_I965, ipeir); POSTING_READ(IPEIR_I965); @@ -1325,12 +1379,13 @@ static void i915_report_and_clear_eir(struct drm_device *dev) if (eir & I915_ERROR_INSTRUCTION) { pr_err("instruction error\n"); pr_err(" INSTPM: 0x%08x\n", I915_READ(INSTPM)); + for (i = 0; i < ARRAY_SIZE(instdone); i++) + pr_err(" INSTDONE_%d: 0x%08x\n", i, instdone[i]); if (INTEL_INFO(dev)->gen < 4) { u32 ipeir = I915_READ(IPEIR); pr_err(" IPEIR: 0x%08x\n", I915_READ(IPEIR)); pr_err(" IPEHR: 0x%08x\n", I915_READ(IPEHR)); - pr_err(" INSTDONE: 0x%08x\n", I915_READ(INSTDONE)); pr_err(" ACTHD: 0x%08x\n", I915_READ(ACTHD)); I915_WRITE(IPEIR, ipeir); POSTING_READ(IPEIR); @@ -1339,10 +1394,7 @@ static void i915_report_and_clear_eir(struct drm_device *dev) pr_err(" IPEIR: 0x%08x\n", I915_READ(IPEIR_I965)); pr_err(" IPEHR: 0x%08x\n", I915_READ(IPEHR_I965)); - pr_err(" INSTDONE: 0x%08x\n", - I915_READ(INSTDONE_I965)); pr_err(" INSTPS: 0x%08x\n", I915_READ(INSTPS)); - pr_err(" INSTDONE1: 0x%08x\n", I915_READ(INSTDONE1)); pr_err(" ACTHD: 0x%08x\n", I915_READ(ACTHD_I965)); I915_WRITE(IPEIR_I965, ipeir); POSTING_READ(IPEIR_I965); @@ -1590,7 +1642,8 @@ ring_last_seqno(struct intel_ring_buffer *ring) static bool i915_hangcheck_ring_idle(struct intel_ring_buffer *ring, bool *err) { if (list_empty(&ring->request_list) || - i915_seqno_passed(ring->get_seqno(ring), ring_last_seqno(ring))) { + i915_seqno_passed(ring->get_seqno(ring, false), + ring_last_seqno(ring))) { /* Issue a wake-up to catch stuck h/w. */ if (waitqueue_active(&ring->irq_queue)) { DRM_ERROR("Hangcheck timer elapsed... %s idle\n", @@ -1656,7 +1709,7 @@ void i915_hangcheck_elapsed(unsigned long data) { struct drm_device *dev = (struct drm_device *)data; drm_i915_private_t *dev_priv = dev->dev_private; - uint32_t acthd[I915_NUM_RINGS], instdone, instdone1; + uint32_t acthd[I915_NUM_RINGS], instdone[I915_NUM_INSTDONE_REG]; struct intel_ring_buffer *ring; bool err = false, idle; int i; @@ -1684,25 +1737,16 @@ void i915_hangcheck_elapsed(unsigned long data) return; } - if (INTEL_INFO(dev)->gen < 4) { - instdone = I915_READ(INSTDONE); - instdone1 = 0; - } else { - instdone = I915_READ(INSTDONE_I965); - instdone1 = I915_READ(INSTDONE1); - } - + i915_get_extra_instdone(dev, instdone); if (memcmp(dev_priv->last_acthd, acthd, sizeof(acthd)) == 0 && - dev_priv->last_instdone == instdone && - dev_priv->last_instdone1 == instdone1) { + memcmp(dev_priv->prev_instdone, instdone, sizeof(instdone)) == 0) { if (i915_hangcheck_hung(dev)) return; } else { dev_priv->hangcheck_count = 0; memcpy(dev_priv->last_acthd, acthd, sizeof(acthd)); - dev_priv->last_instdone = instdone; - dev_priv->last_instdone1 = instdone1; + memcpy(dev_priv->prev_instdone, instdone, sizeof(instdone)); } repeat: @@ -2647,7 +2691,7 @@ void intel_irq_init(struct drm_device *dev) INIT_WORK(&dev_priv->hotplug_work, i915_hotplug_work_func); INIT_WORK(&dev_priv->error_work, i915_error_work_func); - INIT_WORK(&dev_priv->rps_work, gen6_pm_rps_work); + INIT_WORK(&dev_priv->rps.work, gen6_pm_rps_work); INIT_WORK(&dev_priv->parity_error_work, ivybridge_parity_work); dev->driver->get_vblank_counter = i915_get_vblank_counter; diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h index 28725ce5b82c..7637824c6a7d 100644 --- a/drivers/gpu/drm/i915/i915_reg.h +++ b/drivers/gpu/drm/i915/i915_reg.h @@ -450,6 +450,7 @@ #define RING_ACTHD(base) ((base)+0x74) #define RING_NOPID(base) ((base)+0x94) #define RING_IMR(base) ((base)+0xa8) +#define RING_TIMESTAMP(base) ((base)+0x358) #define TAIL_ADDR 0x001FFFF8 #define HEAD_WRAP_COUNT 0xFFE00000 #define HEAD_WRAP_ONE 0x00200000 @@ -478,6 +479,11 @@ #define IPEIR_I965 0x02064 #define IPEHR_I965 0x02068 #define INSTDONE_I965 0x0206c +#define GEN7_INSTDONE_1 0x0206c +#define GEN7_SC_INSTDONE 0x07100 +#define GEN7_SAMPLER_INSTDONE 0x0e160 +#define GEN7_ROW_INSTDONE 0x0e164 +#define I915_NUM_INSTDONE_REG 4 #define RING_IPEIR(base) ((base)+0x64) #define RING_IPEHR(base) ((base)+0x68) #define RING_INSTDONE(base) ((base)+0x6c) @@ -500,6 +506,8 @@ #define DMA_FADD_I8XX 0x020d0 #define ERROR_GEN6 0x040a0 +#define GEN7_ERR_INT 0x44040 +#define ERR_INT_MMIO_UNCLAIMED (1<<13) /* GM45+ chicken bits -- debug workaround bits that may be required * for various sorts of correct behavior. The top 16 bits of each are @@ -529,6 +537,8 @@ #define GFX_PSMI_GRANULARITY (1<<10) #define GFX_PPGTT_ENABLE (1<<9) +#define VLV_DISPLAY_BASE 0x180000 + #define SCPD0 0x0209c /* 915+ only */ #define IER 0x020a0 #define IIR 0x020a4 @@ -1496,6 +1506,14 @@ GEN7_CXT_EXTENDED_SIZE(ctx_reg) + \ GEN7_CXT_GT1_SIZE(ctx_reg) + \ GEN7_CXT_VFSTATE_SIZE(ctx_reg)) +#define HSW_CXT_POWER_SIZE(ctx_reg) ((ctx_reg >> 26) & 0x3f) +#define HSW_CXT_RING_SIZE(ctx_reg) ((ctx_reg >> 23) & 0x7) +#define HSW_CXT_RENDER_SIZE(ctx_reg) ((ctx_reg >> 15) & 0xff) +#define HSW_CXT_TOTAL_SIZE(ctx_reg) (HSW_CXT_POWER_SIZE(ctx_reg) + \ + HSW_CXT_RING_SIZE(ctx_reg) + \ + HSW_CXT_RENDER_SIZE(ctx_reg) + \ + GEN7_CXT_VFSTATE_SIZE(ctx_reg)) + /* * Overlay regs @@ -1549,12 +1567,35 @@ /* VGA port control */ #define ADPA 0x61100 +#define PCH_ADPA 0xe1100 +#define VLV_ADPA (VLV_DISPLAY_BASE + ADPA) + #define ADPA_DAC_ENABLE (1<<31) #define ADPA_DAC_DISABLE 0 #define ADPA_PIPE_SELECT_MASK (1<<30) #define ADPA_PIPE_A_SELECT 0 #define ADPA_PIPE_B_SELECT (1<<30) #define ADPA_PIPE_SELECT(pipe) ((pipe) << 30) +/* CPT uses bits 29:30 for pch transcoder select */ +#define ADPA_CRT_HOTPLUG_MASK 0x03ff0000 /* bit 25-16 */ +#define ADPA_CRT_HOTPLUG_MONITOR_NONE (0<<24) +#define ADPA_CRT_HOTPLUG_MONITOR_MASK (3<<24) +#define ADPA_CRT_HOTPLUG_MONITOR_COLOR (3<<24) +#define ADPA_CRT_HOTPLUG_MONITOR_MONO (2<<24) +#define ADPA_CRT_HOTPLUG_ENABLE (1<<23) +#define ADPA_CRT_HOTPLUG_PERIOD_64 (0<<22) +#define ADPA_CRT_HOTPLUG_PERIOD_128 (1<<22) +#define ADPA_CRT_HOTPLUG_WARMUP_5MS (0<<21) +#define ADPA_CRT_HOTPLUG_WARMUP_10MS (1<<21) +#define ADPA_CRT_HOTPLUG_SAMPLE_2S (0<<20) +#define ADPA_CRT_HOTPLUG_SAMPLE_4S (1<<20) +#define ADPA_CRT_HOTPLUG_VOLTAGE_40 (0<<18) +#define ADPA_CRT_HOTPLUG_VOLTAGE_50 (1<<18) +#define ADPA_CRT_HOTPLUG_VOLTAGE_60 (2<<18) +#define ADPA_CRT_HOTPLUG_VOLTAGE_70 (3<<18) +#define ADPA_CRT_HOTPLUG_VOLREF_325MV (0<<17) +#define ADPA_CRT_HOTPLUG_VOLREF_475MV (1<<17) +#define ADPA_CRT_HOTPLUG_FORCE_TRIGGER (1<<16) #define ADPA_USE_VGA_HVPOLARITY (1<<15) #define ADPA_SETS_HVPOLARITY 0 #define ADPA_VSYNC_CNTL_DISABLE (1<<11) @@ -1753,6 +1794,10 @@ /* Video Data Island Packet control */ #define VIDEO_DIP_DATA 0x61178 +/* Read the description of VIDEO_DIP_DATA (before Haswel) or VIDEO_DIP_ECC + * (Haswell and newer) to see which VIDEO_DIP_DATA byte corresponds to each byte + * of the infoframe structure specified by CEA-861. */ +#define VIDEO_DIP_DATA_SIZE 32 #define VIDEO_DIP_CTL 0x61170 /* Pre HSW: */ #define VIDEO_DIP_ENABLE (1 << 31) @@ -3889,31 +3934,6 @@ #define FDI_PLL_CTL_1 0xfe000 #define FDI_PLL_CTL_2 0xfe004 -/* CRT */ -#define PCH_ADPA 0xe1100 -#define ADPA_TRANS_SELECT_MASK (1<<30) -#define ADPA_TRANS_A_SELECT 0 -#define ADPA_TRANS_B_SELECT (1<<30) -#define ADPA_CRT_HOTPLUG_MASK 0x03ff0000 /* bit 25-16 */ -#define ADPA_CRT_HOTPLUG_MONITOR_NONE (0<<24) -#define ADPA_CRT_HOTPLUG_MONITOR_MASK (3<<24) -#define ADPA_CRT_HOTPLUG_MONITOR_COLOR (3<<24) -#define ADPA_CRT_HOTPLUG_MONITOR_MONO (2<<24) -#define ADPA_CRT_HOTPLUG_ENABLE (1<<23) -#define ADPA_CRT_HOTPLUG_PERIOD_64 (0<<22) -#define ADPA_CRT_HOTPLUG_PERIOD_128 (1<<22) -#define ADPA_CRT_HOTPLUG_WARMUP_5MS (0<<21) -#define ADPA_CRT_HOTPLUG_WARMUP_10MS (1<<21) -#define ADPA_CRT_HOTPLUG_SAMPLE_2S (0<<20) -#define ADPA_CRT_HOTPLUG_SAMPLE_4S (1<<20) -#define ADPA_CRT_HOTPLUG_VOLTAGE_40 (0<<18) -#define ADPA_CRT_HOTPLUG_VOLTAGE_50 (1<<18) -#define ADPA_CRT_HOTPLUG_VOLTAGE_60 (2<<18) -#define ADPA_CRT_HOTPLUG_VOLTAGE_70 (3<<18) -#define ADPA_CRT_HOTPLUG_VOLREF_325MV (0<<17) -#define ADPA_CRT_HOTPLUG_VOLREF_475MV (1<<17) -#define ADPA_CRT_HOTPLUG_FORCE_TRIGGER (1<<16) - /* or SDVOB */ #define HDMIB 0xe1140 #define PORT_ENABLE (1 << 31) @@ -4021,6 +4041,8 @@ #define PORT_TRANS_C_SEL_CPT (2<<29) #define PORT_TRANS_SEL_MASK (3<<29) #define PORT_TRANS_SEL_CPT(pipe) ((pipe) << 29) +#define PORT_TO_PIPE(val) (((val) & (1<<30)) >> 30) +#define PORT_TO_PIPE_CPT(val) (((val) & PORT_TRANS_SEL_MASK) >> 29) #define TRANS_DP_CTL_A 0xe0300 #define TRANS_DP_CTL_B 0xe1300 @@ -4239,7 +4261,15 @@ #define G4X_HDMIW_HDMIEDID 0x6210C #define IBX_HDMIW_HDMIEDID_A 0xE2050 +#define IBX_HDMIW_HDMIEDID_B 0xE2150 +#define IBX_HDMIW_HDMIEDID(pipe) _PIPE(pipe, \ + IBX_HDMIW_HDMIEDID_A, \ + IBX_HDMIW_HDMIEDID_B) #define IBX_AUD_CNTL_ST_A 0xE20B4 +#define IBX_AUD_CNTL_ST_B 0xE21B4 +#define IBX_AUD_CNTL_ST(pipe) _PIPE(pipe, \ + IBX_AUD_CNTL_ST_A, \ + IBX_AUD_CNTL_ST_B) #define IBX_ELD_BUFFER_SIZE (0x1f << 10) #define IBX_ELD_ADDRESS (0x1f << 5) #define IBX_ELD_ACK (1 << 4) @@ -4248,7 +4278,15 @@ #define IBX_CP_READYB (1 << 1) #define CPT_HDMIW_HDMIEDID_A 0xE5050 +#define CPT_HDMIW_HDMIEDID_B 0xE5150 +#define CPT_HDMIW_HDMIEDID(pipe) _PIPE(pipe, \ + CPT_HDMIW_HDMIEDID_A, \ + CPT_HDMIW_HDMIEDID_B) #define CPT_AUD_CNTL_ST_A 0xE50B4 +#define CPT_AUD_CNTL_ST_B 0xE51B4 +#define CPT_AUD_CNTL_ST(pipe) _PIPE(pipe, \ + CPT_AUD_CNTL_ST_A, \ + CPT_AUD_CNTL_ST_B) #define CPT_AUD_CNTRL_ST2 0xE50C0 /* These are the 4 32-bit write offset registers for each stream @@ -4258,7 +4296,15 @@ #define GEN7_SO_WRITE_OFFSET(n) (0x5280 + (n) * 4) #define IBX_AUD_CONFIG_A 0xe2000 +#define IBX_AUD_CONFIG_B 0xe2100 +#define IBX_AUD_CFG(pipe) _PIPE(pipe, \ + IBX_AUD_CONFIG_A, \ + IBX_AUD_CONFIG_B) #define CPT_AUD_CONFIG_A 0xe5000 +#define CPT_AUD_CONFIG_B 0xe5100 +#define CPT_AUD_CFG(pipe) _PIPE(pipe, \ + CPT_AUD_CONFIG_A, \ + CPT_AUD_CONFIG_B) #define AUD_CONFIG_N_VALUE_INDEX (1 << 29) #define AUD_CONFIG_N_PROG_ENABLE (1 << 28) #define AUD_CONFIG_UPPER_N_SHIFT 20 @@ -4269,195 +4315,233 @@ #define AUD_CONFIG_PIXEL_CLOCK_HDMI (0xf << 16) #define AUD_CONFIG_DISABLE_NCTS (1 << 3) +/* HSW Audio */ +#define HSW_AUD_CONFIG_A 0x65000 /* Audio Configuration Transcoder A */ +#define HSW_AUD_CONFIG_B 0x65100 /* Audio Configuration Transcoder B */ +#define HSW_AUD_CFG(pipe) _PIPE(pipe, \ + HSW_AUD_CONFIG_A, \ + HSW_AUD_CONFIG_B) + +#define HSW_AUD_MISC_CTRL_A 0x65010 /* Audio Misc Control Convert 1 */ +#define HSW_AUD_MISC_CTRL_B 0x65110 /* Audio Misc Control Convert 2 */ +#define HSW_AUD_MISC_CTRL(pipe) _PIPE(pipe, \ + HSW_AUD_MISC_CTRL_A, \ + HSW_AUD_MISC_CTRL_B) + +#define HSW_AUD_DIP_ELD_CTRL_ST_A 0x650b4 /* Audio DIP and ELD Control State Transcoder A */ +#define HSW_AUD_DIP_ELD_CTRL_ST_B 0x651b4 /* Audio DIP and ELD Control State Transcoder B */ +#define HSW_AUD_DIP_ELD_CTRL(pipe) _PIPE(pipe, \ + HSW_AUD_DIP_ELD_CTRL_ST_A, \ + HSW_AUD_DIP_ELD_CTRL_ST_B) + +/* Audio Digital Converter */ +#define HSW_AUD_DIG_CNVT_1 0x65080 /* Audio Converter 1 */ +#define HSW_AUD_DIG_CNVT_2 0x65180 /* Audio Converter 1 */ +#define AUD_DIG_CNVT(pipe) _PIPE(pipe, \ + HSW_AUD_DIG_CNVT_1, \ + HSW_AUD_DIG_CNVT_2) +#define DIP_PORT_SEL_MASK 0x3 + +#define HSW_AUD_EDID_DATA_A 0x65050 +#define HSW_AUD_EDID_DATA_B 0x65150 +#define HSW_AUD_EDID_DATA(pipe) _PIPE(pipe, \ + HSW_AUD_EDID_DATA_A, \ + HSW_AUD_EDID_DATA_B) + +#define HSW_AUD_PIPE_CONV_CFG 0x6507c /* Audio pipe and converter configs */ +#define HSW_AUD_PIN_ELD_CP_VLD 0x650c0 /* Audio ELD and CP Ready Status */ +#define AUDIO_INACTIVE_C (1<<11) +#define AUDIO_INACTIVE_B (1<<7) +#define AUDIO_INACTIVE_A (1<<3) +#define AUDIO_OUTPUT_ENABLE_A (1<<2) +#define AUDIO_OUTPUT_ENABLE_B (1<<6) +#define AUDIO_OUTPUT_ENABLE_C (1<<10) +#define AUDIO_ELD_VALID_A (1<<0) +#define AUDIO_ELD_VALID_B (1<<4) +#define AUDIO_ELD_VALID_C (1<<8) +#define AUDIO_CP_READY_A (1<<1) +#define AUDIO_CP_READY_B (1<<5) +#define AUDIO_CP_READY_C (1<<9) + /* HSW Power Wells */ -#define HSW_PWR_WELL_CTL1 0x45400 /* BIOS */ -#define HSW_PWR_WELL_CTL2 0x45404 /* Driver */ -#define HSW_PWR_WELL_CTL3 0x45408 /* KVMR */ -#define HSW_PWR_WELL_CTL4 0x4540C /* Debug */ -#define HSW_PWR_WELL_ENABLE (1<<31) -#define HSW_PWR_WELL_STATE (1<<30) -#define HSW_PWR_WELL_CTL5 0x45410 +#define HSW_PWR_WELL_CTL1 0x45400 /* BIOS */ +#define HSW_PWR_WELL_CTL2 0x45404 /* Driver */ +#define HSW_PWR_WELL_CTL3 0x45408 /* KVMR */ +#define HSW_PWR_WELL_CTL4 0x4540C /* Debug */ +#define HSW_PWR_WELL_ENABLE (1<<31) +#define HSW_PWR_WELL_STATE (1<<30) +#define HSW_PWR_WELL_CTL5 0x45410 #define HSW_PWR_WELL_ENABLE_SINGLE_STEP (1<<31) #define HSW_PWR_WELL_PWR_GATE_OVERRIDE (1<<20) -#define HSW_PWR_WELL_FORCE_ON (1<<19) -#define HSW_PWR_WELL_CTL6 0x45414 +#define HSW_PWR_WELL_FORCE_ON (1<<19) +#define HSW_PWR_WELL_CTL6 0x45414 /* Per-pipe DDI Function Control */ -#define PIPE_DDI_FUNC_CTL_A 0x60400 -#define PIPE_DDI_FUNC_CTL_B 0x61400 -#define PIPE_DDI_FUNC_CTL_C 0x62400 +#define PIPE_DDI_FUNC_CTL_A 0x60400 +#define PIPE_DDI_FUNC_CTL_B 0x61400 +#define PIPE_DDI_FUNC_CTL_C 0x62400 #define PIPE_DDI_FUNC_CTL_EDP 0x6F400 -#define DDI_FUNC_CTL(pipe) _PIPE(pipe, \ - PIPE_DDI_FUNC_CTL_A, \ - PIPE_DDI_FUNC_CTL_B) +#define DDI_FUNC_CTL(pipe) _PIPE(pipe, PIPE_DDI_FUNC_CTL_A, \ + PIPE_DDI_FUNC_CTL_B) #define PIPE_DDI_FUNC_ENABLE (1<<31) /* Those bits are ignored by pipe EDP since it can only connect to DDI A */ -#define PIPE_DDI_PORT_MASK (7<<28) -#define PIPE_DDI_SELECT_PORT(x) ((x)<<28) -#define PIPE_DDI_MODE_SELECT_HDMI (0<<24) -#define PIPE_DDI_MODE_SELECT_DVI (1<<24) +#define PIPE_DDI_PORT_MASK (7<<28) +#define PIPE_DDI_SELECT_PORT(x) ((x)<<28) +#define PIPE_DDI_MODE_SELECT_MASK (7<<24) +#define PIPE_DDI_MODE_SELECT_HDMI (0<<24) +#define PIPE_DDI_MODE_SELECT_DVI (1<<24) #define PIPE_DDI_MODE_SELECT_DP_SST (2<<24) #define PIPE_DDI_MODE_SELECT_DP_MST (3<<24) -#define PIPE_DDI_MODE_SELECT_FDI (4<<24) -#define PIPE_DDI_BPC_8 (0<<20) -#define PIPE_DDI_BPC_10 (1<<20) -#define PIPE_DDI_BPC_6 (2<<20) -#define PIPE_DDI_BPC_12 (3<<20) -#define PIPE_DDI_BFI_ENABLE (1<<4) -#define PIPE_DDI_PORT_WIDTH_X1 (0<<1) -#define PIPE_DDI_PORT_WIDTH_X2 (1<<1) -#define PIPE_DDI_PORT_WIDTH_X4 (3<<1) +#define PIPE_DDI_MODE_SELECT_FDI (4<<24) +#define PIPE_DDI_BPC_MASK (7<<20) +#define PIPE_DDI_BPC_8 (0<<20) +#define PIPE_DDI_BPC_10 (1<<20) +#define PIPE_DDI_BPC_6 (2<<20) +#define PIPE_DDI_BPC_12 (3<<20) +#define PIPE_DDI_PVSYNC (1<<17) +#define PIPE_DDI_PHSYNC (1<<16) +#define PIPE_DDI_BFI_ENABLE (1<<4) +#define PIPE_DDI_PORT_WIDTH_X1 (0<<1) +#define PIPE_DDI_PORT_WIDTH_X2 (1<<1) +#define PIPE_DDI_PORT_WIDTH_X4 (3<<1) /* DisplayPort Transport Control */ #define DP_TP_CTL_A 0x64040 #define DP_TP_CTL_B 0x64140 -#define DP_TP_CTL(port) _PORT(port, \ - DP_TP_CTL_A, \ - DP_TP_CTL_B) -#define DP_TP_CTL_ENABLE (1<<31) -#define DP_TP_CTL_MODE_SST (0<<27) -#define DP_TP_CTL_MODE_MST (1<<27) +#define DP_TP_CTL(port) _PORT(port, DP_TP_CTL_A, DP_TP_CTL_B) +#define DP_TP_CTL_ENABLE (1<<31) +#define DP_TP_CTL_MODE_SST (0<<27) +#define DP_TP_CTL_MODE_MST (1<<27) #define DP_TP_CTL_ENHANCED_FRAME_ENABLE (1<<18) -#define DP_TP_CTL_FDI_AUTOTRAIN (1<<15) +#define DP_TP_CTL_FDI_AUTOTRAIN (1<<15) #define DP_TP_CTL_LINK_TRAIN_MASK (7<<8) #define DP_TP_CTL_LINK_TRAIN_PAT1 (0<<8) #define DP_TP_CTL_LINK_TRAIN_PAT2 (1<<8) -#define DP_TP_CTL_LINK_TRAIN_NORMAL (3<<8) +#define DP_TP_CTL_LINK_TRAIN_NORMAL (3<<8) /* DisplayPort Transport Status */ #define DP_TP_STATUS_A 0x64044 #define DP_TP_STATUS_B 0x64144 -#define DP_TP_STATUS(port) _PORT(port, \ - DP_TP_STATUS_A, \ - DP_TP_STATUS_B) +#define DP_TP_STATUS(port) _PORT(port, DP_TP_STATUS_A, DP_TP_STATUS_B) #define DP_TP_STATUS_AUTOTRAIN_DONE (1<<12) /* DDI Buffer Control */ #define DDI_BUF_CTL_A 0x64000 #define DDI_BUF_CTL_B 0x64100 -#define DDI_BUF_CTL(port) _PORT(port, \ - DDI_BUF_CTL_A, \ - DDI_BUF_CTL_B) -#define DDI_BUF_CTL_ENABLE (1<<31) +#define DDI_BUF_CTL(port) _PORT(port, DDI_BUF_CTL_A, DDI_BUF_CTL_B) +#define DDI_BUF_CTL_ENABLE (1<<31) #define DDI_BUF_EMP_400MV_0DB_HSW (0<<24) /* Sel0 */ -#define DDI_BUF_EMP_400MV_3_5DB_HSW (1<<24) /* Sel1 */ +#define DDI_BUF_EMP_400MV_3_5DB_HSW (1<<24) /* Sel1 */ #define DDI_BUF_EMP_400MV_6DB_HSW (2<<24) /* Sel2 */ -#define DDI_BUF_EMP_400MV_9_5DB_HSW (3<<24) /* Sel3 */ +#define DDI_BUF_EMP_400MV_9_5DB_HSW (3<<24) /* Sel3 */ #define DDI_BUF_EMP_600MV_0DB_HSW (4<<24) /* Sel4 */ -#define DDI_BUF_EMP_600MV_3_5DB_HSW (5<<24) /* Sel5 */ +#define DDI_BUF_EMP_600MV_3_5DB_HSW (5<<24) /* Sel5 */ #define DDI_BUF_EMP_600MV_6DB_HSW (6<<24) /* Sel6 */ #define DDI_BUF_EMP_800MV_0DB_HSW (7<<24) /* Sel7 */ -#define DDI_BUF_EMP_800MV_3_5DB_HSW (8<<24) /* Sel8 */ -#define DDI_BUF_EMP_MASK (0xf<<24) -#define DDI_BUF_IS_IDLE (1<<7) -#define DDI_PORT_WIDTH_X1 (0<<1) -#define DDI_PORT_WIDTH_X2 (1<<1) -#define DDI_PORT_WIDTH_X4 (3<<1) +#define DDI_BUF_EMP_800MV_3_5DB_HSW (8<<24) /* Sel8 */ +#define DDI_BUF_EMP_MASK (0xf<<24) +#define DDI_BUF_IS_IDLE (1<<7) +#define DDI_PORT_WIDTH_X1 (0<<1) +#define DDI_PORT_WIDTH_X2 (1<<1) +#define DDI_PORT_WIDTH_X4 (3<<1) #define DDI_INIT_DISPLAY_DETECTED (1<<0) /* DDI Buffer Translations */ #define DDI_BUF_TRANS_A 0x64E00 #define DDI_BUF_TRANS_B 0x64E60 -#define DDI_BUF_TRANS(port) _PORT(port, \ - DDI_BUF_TRANS_A, \ - DDI_BUF_TRANS_B) +#define DDI_BUF_TRANS(port) _PORT(port, DDI_BUF_TRANS_A, DDI_BUF_TRANS_B) /* Sideband Interface (SBI) is programmed indirectly, via * SBI_ADDR, which contains the register offset; and SBI_DATA, * which contains the payload */ -#define SBI_ADDR 0xC6000 -#define SBI_DATA 0xC6004 +#define SBI_ADDR 0xC6000 +#define SBI_DATA 0xC6004 #define SBI_CTL_STAT 0xC6008 #define SBI_CTL_OP_CRRD (0x6<<8) #define SBI_CTL_OP_CRWR (0x7<<8) #define SBI_RESPONSE_FAIL (0x1<<1) -#define SBI_RESPONSE_SUCCESS (0x0<<1) -#define SBI_BUSY (0x1<<0) -#define SBI_READY (0x0<<0) +#define SBI_RESPONSE_SUCCESS (0x0<<1) +#define SBI_BUSY (0x1<<0) +#define SBI_READY (0x0<<0) /* SBI offsets */ -#define SBI_SSCDIVINTPHASE6 0x0600 +#define SBI_SSCDIVINTPHASE6 0x0600 #define SBI_SSCDIVINTPHASE_DIVSEL_MASK ((0x7f)<<1) #define SBI_SSCDIVINTPHASE_DIVSEL(x) ((x)<<1) #define SBI_SSCDIVINTPHASE_INCVAL_MASK ((0x7f)<<8) #define SBI_SSCDIVINTPHASE_INCVAL(x) ((x)<<8) -#define SBI_SSCDIVINTPHASE_DIR(x) ((x)<<15) +#define SBI_SSCDIVINTPHASE_DIR(x) ((x)<<15) #define SBI_SSCDIVINTPHASE_PROPAGATE (1<<0) -#define SBI_SSCCTL 0x020c +#define SBI_SSCCTL 0x020c #define SBI_SSCCTL6 0x060C -#define SBI_SSCCTL_DISABLE (1<<0) +#define SBI_SSCCTL_DISABLE (1<<0) #define SBI_SSCAUXDIV6 0x0610 #define SBI_SSCAUXDIV_FINALDIV2SEL(x) ((x)<<4) -#define SBI_DBUFF0 0x2a00 +#define SBI_DBUFF0 0x2a00 /* LPT PIXCLK_GATE */ -#define PIXCLK_GATE 0xC6020 -#define PIXCLK_GATE_UNGATE 1<<0 -#define PIXCLK_GATE_GATE 0<<0 +#define PIXCLK_GATE 0xC6020 +#define PIXCLK_GATE_UNGATE (1<<0) +#define PIXCLK_GATE_GATE (0<<0) /* SPLL */ -#define SPLL_CTL 0x46020 +#define SPLL_CTL 0x46020 #define SPLL_PLL_ENABLE (1<<31) #define SPLL_PLL_SCC (1<<28) #define SPLL_PLL_NON_SCC (2<<28) -#define SPLL_PLL_FREQ_810MHz (0<<26) -#define SPLL_PLL_FREQ_1350MHz (1<<26) +#define SPLL_PLL_FREQ_810MHz (0<<26) +#define SPLL_PLL_FREQ_1350MHz (1<<26) /* WRPLL */ -#define WRPLL_CTL1 0x46040 -#define WRPLL_CTL2 0x46060 -#define WRPLL_PLL_ENABLE (1<<31) -#define WRPLL_PLL_SELECT_SSC (0x01<<28) -#define WRPLL_PLL_SELECT_NON_SCC (0x02<<28) +#define WRPLL_CTL1 0x46040 +#define WRPLL_CTL2 0x46060 +#define WRPLL_PLL_ENABLE (1<<31) +#define WRPLL_PLL_SELECT_SSC (0x01<<28) +#define WRPLL_PLL_SELECT_NON_SCC (0x02<<28) #define WRPLL_PLL_SELECT_LCPLL_2700 (0x03<<28) /* WRPLL divider programming */ -#define WRPLL_DIVIDER_REFERENCE(x) ((x)<<0) -#define WRPLL_DIVIDER_POST(x) ((x)<<8) -#define WRPLL_DIVIDER_FEEDBACK(x) ((x)<<16) +#define WRPLL_DIVIDER_REFERENCE(x) ((x)<<0) +#define WRPLL_DIVIDER_POST(x) ((x)<<8) +#define WRPLL_DIVIDER_FEEDBACK(x) ((x)<<16) /* Port clock selection */ #define PORT_CLK_SEL_A 0x46100 #define PORT_CLK_SEL_B 0x46104 -#define PORT_CLK_SEL(port) _PORT(port, \ - PORT_CLK_SEL_A, \ - PORT_CLK_SEL_B) +#define PORT_CLK_SEL(port) _PORT(port, PORT_CLK_SEL_A, PORT_CLK_SEL_B) #define PORT_CLK_SEL_LCPLL_2700 (0<<29) #define PORT_CLK_SEL_LCPLL_1350 (1<<29) #define PORT_CLK_SEL_LCPLL_810 (2<<29) -#define PORT_CLK_SEL_SPLL (3<<29) +#define PORT_CLK_SEL_SPLL (3<<29) #define PORT_CLK_SEL_WRPLL1 (4<<29) #define PORT_CLK_SEL_WRPLL2 (5<<29) /* Pipe clock selection */ #define PIPE_CLK_SEL_A 0x46140 #define PIPE_CLK_SEL_B 0x46144 -#define PIPE_CLK_SEL(pipe) _PIPE(pipe, \ - PIPE_CLK_SEL_A, \ - PIPE_CLK_SEL_B) +#define PIPE_CLK_SEL(pipe) _PIPE(pipe, PIPE_CLK_SEL_A, PIPE_CLK_SEL_B) /* For each pipe, we need to select the corresponding port clock */ -#define PIPE_CLK_SEL_DISABLED (0x0<<29) -#define PIPE_CLK_SEL_PORT(x) ((x+1)<<29) +#define PIPE_CLK_SEL_DISABLED (0x0<<29) +#define PIPE_CLK_SEL_PORT(x) ((x+1)<<29) /* LCPLL Control */ -#define LCPLL_CTL 0x130040 +#define LCPLL_CTL 0x130040 #define LCPLL_PLL_DISABLE (1<<31) #define LCPLL_PLL_LOCK (1<<30) -#define LCPLL_CD_CLOCK_DISABLE (1<<25) +#define LCPLL_CD_CLOCK_DISABLE (1<<25) #define LCPLL_CD2X_CLOCK_DISABLE (1<<23) /* Pipe WM_LINETIME - watermark line time */ #define PIPE_WM_LINETIME_A 0x45270 #define PIPE_WM_LINETIME_B 0x45274 -#define PIPE_WM_LINETIME(pipe) _PIPE(pipe, \ - PIPE_WM_LINETIME_A, \ - PIPE_WM_LINETIME_B) -#define PIPE_WM_LINETIME_MASK (0x1ff) -#define PIPE_WM_LINETIME_TIME(x) ((x)) +#define PIPE_WM_LINETIME(pipe) _PIPE(pipe, PIPE_WM_LINETIME_A, \ + PIPE_WM_LINETIME_B) +#define PIPE_WM_LINETIME_MASK (0x1ff) +#define PIPE_WM_LINETIME_TIME(x) ((x)) #define PIPE_WM_LINETIME_IPS_LINETIME_MASK (0x1ff<<16) -#define PIPE_WM_LINETIME_IPS_LINETIME(x) ((x)<<16) +#define PIPE_WM_LINETIME_IPS_LINETIME(x) ((x)<<16) /* SFUSE_STRAP */ -#define SFUSE_STRAP 0xc2014 +#define SFUSE_STRAP 0xc2014 #define SFUSE_STRAP_DDIB_DETECTED (1<<2) #define SFUSE_STRAP_DDIC_DETECTED (1<<1) #define SFUSE_STRAP_DDID_DETECTED (1<<0) diff --git a/drivers/gpu/drm/i915/i915_sysfs.c b/drivers/gpu/drm/i915/i915_sysfs.c index 7631807a2788..903eebd2117a 100644 --- a/drivers/gpu/drm/i915/i915_sysfs.c +++ b/drivers/gpu/drm/i915/i915_sysfs.c @@ -46,32 +46,32 @@ static u32 calc_residency(struct drm_device *dev, const u32 reg) } static ssize_t -show_rc6_mask(struct device *dev, struct device_attribute *attr, char *buf) +show_rc6_mask(struct device *kdev, struct device_attribute *attr, char *buf) { - struct drm_minor *dminor = container_of(dev, struct drm_minor, kdev); + struct drm_minor *dminor = container_of(kdev, struct drm_minor, kdev); return snprintf(buf, PAGE_SIZE, "%x", intel_enable_rc6(dminor->dev)); } static ssize_t -show_rc6_ms(struct device *dev, struct device_attribute *attr, char *buf) +show_rc6_ms(struct device *kdev, struct device_attribute *attr, char *buf) { - struct drm_minor *dminor = container_of(dev, struct drm_minor, kdev); + struct drm_minor *dminor = container_of(kdev, struct drm_minor, kdev); u32 rc6_residency = calc_residency(dminor->dev, GEN6_GT_GFX_RC6); return snprintf(buf, PAGE_SIZE, "%u", rc6_residency); } static ssize_t -show_rc6p_ms(struct device *dev, struct device_attribute *attr, char *buf) +show_rc6p_ms(struct device *kdev, struct device_attribute *attr, char *buf) { - struct drm_minor *dminor = container_of(dev, struct drm_minor, kdev); + struct drm_minor *dminor = container_of(kdev, struct drm_minor, kdev); u32 rc6p_residency = calc_residency(dminor->dev, GEN6_GT_GFX_RC6p); return snprintf(buf, PAGE_SIZE, "%u", rc6p_residency); } static ssize_t -show_rc6pp_ms(struct device *dev, struct device_attribute *attr, char *buf) +show_rc6pp_ms(struct device *kdev, struct device_attribute *attr, char *buf) { - struct drm_minor *dminor = container_of(dev, struct drm_minor, kdev); + struct drm_minor *dminor = container_of(kdev, struct drm_minor, kdev); u32 rc6pp_residency = calc_residency(dminor->dev, GEN6_GT_GFX_RC6pp); return snprintf(buf, PAGE_SIZE, "%u", rc6pp_residency); } @@ -93,6 +93,7 @@ static struct attribute_group rc6_attr_group = { .name = power_group_name, .attrs = rc6_attrs }; +#endif static int l3_access_valid(struct drm_device *dev, loff_t offset) { @@ -202,37 +203,214 @@ static struct bin_attribute dpf_attrs = { .mmap = NULL }; +static ssize_t gt_cur_freq_mhz_show(struct device *kdev, + struct device_attribute *attr, char *buf) +{ + struct drm_minor *minor = container_of(kdev, struct drm_minor, kdev); + struct drm_device *dev = minor->dev; + struct drm_i915_private *dev_priv = dev->dev_private; + int ret; + + ret = i915_mutex_lock_interruptible(dev); + if (ret) + return ret; + + ret = dev_priv->rps.cur_delay * GT_FREQUENCY_MULTIPLIER; + mutex_unlock(&dev->struct_mutex); + + return snprintf(buf, PAGE_SIZE, "%d", ret); +} + +static ssize_t gt_max_freq_mhz_show(struct device *kdev, struct device_attribute *attr, char *buf) +{ + struct drm_minor *minor = container_of(kdev, struct drm_minor, kdev); + struct drm_device *dev = minor->dev; + struct drm_i915_private *dev_priv = dev->dev_private; + int ret; + + ret = i915_mutex_lock_interruptible(dev); + if (ret) + return ret; + + ret = dev_priv->rps.max_delay * GT_FREQUENCY_MULTIPLIER; + mutex_unlock(&dev->struct_mutex); + + return snprintf(buf, PAGE_SIZE, "%d", ret); +} + +static ssize_t gt_max_freq_mhz_store(struct device *kdev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + struct drm_minor *minor = container_of(kdev, struct drm_minor, kdev); + struct drm_device *dev = minor->dev; + struct drm_i915_private *dev_priv = dev->dev_private; + u32 val, rp_state_cap, hw_max, hw_min; + ssize_t ret; + + ret = kstrtou32(buf, 0, &val); + if (ret) + return ret; + + val /= GT_FREQUENCY_MULTIPLIER; + + ret = mutex_lock_interruptible(&dev->struct_mutex); + if (ret) + return ret; + + rp_state_cap = I915_READ(GEN6_RP_STATE_CAP); + hw_max = (rp_state_cap & 0xff); + hw_min = ((rp_state_cap & 0xff0000) >> 16); + + if (val < hw_min || val > hw_max || val < dev_priv->rps.min_delay) { + mutex_unlock(&dev->struct_mutex); + return -EINVAL; + } + + if (dev_priv->rps.cur_delay > val) + gen6_set_rps(dev_priv->dev, val); + + dev_priv->rps.max_delay = val; + + mutex_unlock(&dev->struct_mutex); + + return count; +} + +static ssize_t gt_min_freq_mhz_show(struct device *kdev, struct device_attribute *attr, char *buf) +{ + struct drm_minor *minor = container_of(kdev, struct drm_minor, kdev); + struct drm_device *dev = minor->dev; + struct drm_i915_private *dev_priv = dev->dev_private; + int ret; + + ret = i915_mutex_lock_interruptible(dev); + if (ret) + return ret; + + ret = dev_priv->rps.min_delay * GT_FREQUENCY_MULTIPLIER; + mutex_unlock(&dev->struct_mutex); + + return snprintf(buf, PAGE_SIZE, "%d", ret); +} + +static ssize_t gt_min_freq_mhz_store(struct device *kdev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + struct drm_minor *minor = container_of(kdev, struct drm_minor, kdev); + struct drm_device *dev = minor->dev; + struct drm_i915_private *dev_priv = dev->dev_private; + u32 val, rp_state_cap, hw_max, hw_min; + ssize_t ret; + + ret = kstrtou32(buf, 0, &val); + if (ret) + return ret; + + val /= GT_FREQUENCY_MULTIPLIER; + + ret = mutex_lock_interruptible(&dev->struct_mutex); + if (ret) + return ret; + + rp_state_cap = I915_READ(GEN6_RP_STATE_CAP); + hw_max = (rp_state_cap & 0xff); + hw_min = ((rp_state_cap & 0xff0000) >> 16); + + if (val < hw_min || val > hw_max || val > dev_priv->rps.max_delay) { + mutex_unlock(&dev->struct_mutex); + return -EINVAL; + } + + if (dev_priv->rps.cur_delay < val) + gen6_set_rps(dev_priv->dev, val); + + dev_priv->rps.min_delay = val; + + mutex_unlock(&dev->struct_mutex); + + return count; + +} + +static DEVICE_ATTR(gt_cur_freq_mhz, S_IRUGO, gt_cur_freq_mhz_show, NULL); +static DEVICE_ATTR(gt_max_freq_mhz, S_IRUGO | S_IWUSR, gt_max_freq_mhz_show, gt_max_freq_mhz_store); +static DEVICE_ATTR(gt_min_freq_mhz, S_IRUGO | S_IWUSR, gt_min_freq_mhz_show, gt_min_freq_mhz_store); + + +static ssize_t gt_rp_mhz_show(struct device *kdev, struct device_attribute *attr, char *buf); +static DEVICE_ATTR(gt_RP0_freq_mhz, S_IRUGO, gt_rp_mhz_show, NULL); +static DEVICE_ATTR(gt_RP1_freq_mhz, S_IRUGO, gt_rp_mhz_show, NULL); +static DEVICE_ATTR(gt_RPn_freq_mhz, S_IRUGO, gt_rp_mhz_show, NULL); + +/* For now we have a static number of RP states */ +static ssize_t gt_rp_mhz_show(struct device *kdev, struct device_attribute *attr, char *buf) +{ + struct drm_minor *minor = container_of(kdev, struct drm_minor, kdev); + struct drm_device *dev = minor->dev; + struct drm_i915_private *dev_priv = dev->dev_private; + u32 val, rp_state_cap; + ssize_t ret; + + ret = mutex_lock_interruptible(&dev->struct_mutex); + if (ret) + return ret; + rp_state_cap = I915_READ(GEN6_RP_STATE_CAP); + mutex_unlock(&dev->struct_mutex); + + if (attr == &dev_attr_gt_RP0_freq_mhz) { + val = ((rp_state_cap & 0x0000ff) >> 0) * GT_FREQUENCY_MULTIPLIER; + } else if (attr == &dev_attr_gt_RP1_freq_mhz) { + val = ((rp_state_cap & 0x00ff00) >> 8) * GT_FREQUENCY_MULTIPLIER; + } else if (attr == &dev_attr_gt_RPn_freq_mhz) { + val = ((rp_state_cap & 0xff0000) >> 16) * GT_FREQUENCY_MULTIPLIER; + } else { + BUG(); + } + return snprintf(buf, PAGE_SIZE, "%d", val); +} + +static const struct attribute *gen6_attrs[] = { + &dev_attr_gt_cur_freq_mhz.attr, + &dev_attr_gt_max_freq_mhz.attr, + &dev_attr_gt_min_freq_mhz.attr, + &dev_attr_gt_RP0_freq_mhz.attr, + &dev_attr_gt_RP1_freq_mhz.attr, + &dev_attr_gt_RPn_freq_mhz.attr, + NULL, +}; + void i915_setup_sysfs(struct drm_device *dev) { int ret; +#ifdef CONFIG_PM if (INTEL_INFO(dev)->gen >= 6) { ret = sysfs_merge_group(&dev->primary->kdev.kobj, &rc6_attr_group); if (ret) DRM_ERROR("RC6 residency sysfs setup failed\n"); } - - if (IS_IVYBRIDGE(dev)) { +#endif + if (HAS_L3_GPU_CACHE(dev)) { ret = device_create_bin_file(&dev->primary->kdev, &dpf_attrs); if (ret) DRM_ERROR("l3 parity sysfs setup failed\n"); } + + if (INTEL_INFO(dev)->gen >= 6) { + ret = sysfs_create_files(&dev->primary->kdev.kobj, gen6_attrs); + if (ret) + DRM_ERROR("gen6 sysfs setup failed\n"); + } } void i915_teardown_sysfs(struct drm_device *dev) { + sysfs_remove_files(&dev->primary->kdev.kobj, gen6_attrs); device_remove_bin_file(&dev->primary->kdev, &dpf_attrs); +#ifdef CONFIG_PM sysfs_unmerge_group(&dev->primary->kdev.kobj, &rc6_attr_group); +#endif } -#else -void i915_setup_sysfs(struct drm_device *dev) -{ - return; -} - -void i915_teardown_sysfs(struct drm_device *dev) -{ - return; -} -#endif /* CONFIG_PM */ diff --git a/drivers/gpu/drm/i915/i915_trace.h b/drivers/gpu/drm/i915/i915_trace.h index fe90b3a84a6d..8134421b89a6 100644 --- a/drivers/gpu/drm/i915/i915_trace.h +++ b/drivers/gpu/drm/i915/i915_trace.h @@ -214,22 +214,18 @@ TRACE_EVENT(i915_gem_evict, ); TRACE_EVENT(i915_gem_evict_everything, - TP_PROTO(struct drm_device *dev, bool purgeable), - TP_ARGS(dev, purgeable), + TP_PROTO(struct drm_device *dev), + TP_ARGS(dev), TP_STRUCT__entry( __field(u32, dev) - __field(bool, purgeable) ), TP_fast_assign( __entry->dev = dev->primary->index; - __entry->purgeable = purgeable; ), - TP_printk("dev=%d%s", - __entry->dev, - __entry->purgeable ? ", purgeable only" : "") + TP_printk("dev=%d", __entry->dev) ); TRACE_EVENT(i915_gem_ring_dispatch, @@ -434,6 +430,21 @@ TRACE_EVENT(i915_reg_rw, (u32)(__entry->val >> 32)) ); +TRACE_EVENT(intel_gpu_freq_change, + TP_PROTO(u32 freq), + TP_ARGS(freq), + + TP_STRUCT__entry( + __field(u32, freq) + ), + + TP_fast_assign( + __entry->freq = freq; + ), + + TP_printk("new_freq=%u", __entry->freq) +); + #endif /* _I915_TRACE_H_ */ /* This part must be outside protection */ diff --git a/drivers/gpu/drm/i915/intel_crt.c b/drivers/gpu/drm/i915/intel_crt.c index 23bdc8cd1458..c42b9809f86d 100644 --- a/drivers/gpu/drm/i915/intel_crt.c +++ b/drivers/gpu/drm/i915/intel_crt.c @@ -47,6 +47,7 @@ struct intel_crt { struct intel_encoder base; bool force_hotplug_required; + u32 adpa_reg; }; static struct intel_crt *intel_attached_crt(struct drm_connector *connector) @@ -55,42 +56,68 @@ static struct intel_crt *intel_attached_crt(struct drm_connector *connector) struct intel_crt, base); } -static void pch_crt_dpms(struct drm_encoder *encoder, int mode) +static struct intel_crt *intel_encoder_to_crt(struct intel_encoder *encoder) { - struct drm_device *dev = encoder->dev; + return container_of(encoder, struct intel_crt, base); +} + +static bool intel_crt_get_hw_state(struct intel_encoder *encoder, + enum pipe *pipe) +{ + struct drm_device *dev = encoder->base.dev; struct drm_i915_private *dev_priv = dev->dev_private; + struct intel_crt *crt = intel_encoder_to_crt(encoder); + u32 tmp; + + tmp = I915_READ(crt->adpa_reg); + + if (!(tmp & ADPA_DAC_ENABLE)) + return false; + + if (HAS_PCH_CPT(dev)) + *pipe = PORT_TO_PIPE_CPT(tmp); + else + *pipe = PORT_TO_PIPE(tmp); + + return true; +} + +static void intel_disable_crt(struct intel_encoder *encoder) +{ + struct drm_i915_private *dev_priv = encoder->base.dev->dev_private; + struct intel_crt *crt = intel_encoder_to_crt(encoder); u32 temp; - temp = I915_READ(PCH_ADPA); + temp = I915_READ(crt->adpa_reg); + temp &= ~(ADPA_HSYNC_CNTL_DISABLE | ADPA_VSYNC_CNTL_DISABLE); temp &= ~ADPA_DAC_ENABLE; + I915_WRITE(crt->adpa_reg, temp); +} - switch (mode) { - case DRM_MODE_DPMS_ON: - temp |= ADPA_DAC_ENABLE; - break; - case DRM_MODE_DPMS_STANDBY: - case DRM_MODE_DPMS_SUSPEND: - case DRM_MODE_DPMS_OFF: - /* Just leave port enable cleared */ - break; - } +static void intel_enable_crt(struct intel_encoder *encoder) +{ + struct drm_i915_private *dev_priv = encoder->base.dev->dev_private; + struct intel_crt *crt = intel_encoder_to_crt(encoder); + u32 temp; - I915_WRITE(PCH_ADPA, temp); + temp = I915_READ(crt->adpa_reg); + temp |= ADPA_DAC_ENABLE; + I915_WRITE(crt->adpa_reg, temp); } -static void gmch_crt_dpms(struct drm_encoder *encoder, int mode) +/* Note: The caller is required to filter out dpms modes not supported by the + * platform. */ +static void intel_crt_set_dpms(struct intel_encoder *encoder, int mode) { - struct drm_device *dev = encoder->dev; + struct drm_device *dev = encoder->base.dev; struct drm_i915_private *dev_priv = dev->dev_private; + struct intel_crt *crt = intel_encoder_to_crt(encoder); u32 temp; - temp = I915_READ(ADPA); + temp = I915_READ(crt->adpa_reg); temp &= ~(ADPA_HSYNC_CNTL_DISABLE | ADPA_VSYNC_CNTL_DISABLE); temp &= ~ADPA_DAC_ENABLE; - if (IS_VALLEYVIEW(dev) && mode != DRM_MODE_DPMS_ON) - mode = DRM_MODE_DPMS_OFF; - switch (mode) { case DRM_MODE_DPMS_ON: temp |= ADPA_DAC_ENABLE; @@ -106,7 +133,51 @@ static void gmch_crt_dpms(struct drm_encoder *encoder, int mode) break; } - I915_WRITE(ADPA, temp); + I915_WRITE(crt->adpa_reg, temp); +} + +static void intel_crt_dpms(struct drm_connector *connector, int mode) +{ + struct drm_device *dev = connector->dev; + struct intel_encoder *encoder = intel_attached_encoder(connector); + struct drm_crtc *crtc; + int old_dpms; + + /* PCH platforms and VLV only support on/off. */ + if (INTEL_INFO(dev)->gen < 5 && mode != DRM_MODE_DPMS_ON) + mode = DRM_MODE_DPMS_OFF; + + if (mode == connector->dpms) + return; + + old_dpms = connector->dpms; + connector->dpms = mode; + + /* Only need to change hw state when actually enabled */ + crtc = encoder->base.crtc; + if (!crtc) { + encoder->connectors_active = false; + return; + } + + /* We need the pipe to run for anything but OFF. */ + if (mode == DRM_MODE_DPMS_OFF) + encoder->connectors_active = false; + else + encoder->connectors_active = true; + + if (mode < old_dpms) { + /* From off to on, enable the pipe first. */ + intel_crtc_update_dpms(crtc); + + intel_crt_set_dpms(encoder, mode); + } else { + intel_crt_set_dpms(encoder, mode); + + intel_crtc_update_dpms(crtc); + } + + intel_modeset_check_state(connector->dev); } static int intel_crt_mode_valid(struct drm_connector *connector, @@ -145,19 +216,15 @@ static void intel_crt_mode_set(struct drm_encoder *encoder, struct drm_device *dev = encoder->dev; struct drm_crtc *crtc = encoder->crtc; + struct intel_crt *crt = + intel_encoder_to_crt(to_intel_encoder(encoder)); struct intel_crtc *intel_crtc = to_intel_crtc(crtc); struct drm_i915_private *dev_priv = dev->dev_private; int dpll_md_reg; u32 adpa, dpll_md; - u32 adpa_reg; dpll_md_reg = DPLL_MD(intel_crtc->pipe); - if (HAS_PCH_SPLIT(dev)) - adpa_reg = PCH_ADPA; - else - adpa_reg = ADPA; - /* * Disable separate mode multiplier used when cloning SDVO to CRT * XXX this needs to be adjusted when we really are cloning @@ -185,7 +252,7 @@ static void intel_crt_mode_set(struct drm_encoder *encoder, if (!HAS_PCH_SPLIT(dev)) I915_WRITE(BCLRPAT(intel_crtc->pipe), 0); - I915_WRITE(adpa_reg, adpa); + I915_WRITE(crt->adpa_reg, adpa); } static bool intel_ironlake_crt_detect_hotplug(struct drm_connector *connector) @@ -545,14 +612,12 @@ intel_crt_detect(struct drm_connector *connector, bool force) return connector->status; /* for pre-945g platforms use load detect */ - if (intel_get_load_detect_pipe(&crt->base, connector, NULL, - &tmp)) { + if (intel_get_load_detect_pipe(connector, NULL, &tmp)) { if (intel_crt_detect_ddc(connector)) status = connector_status_connected; else status = intel_crt_load_detect(crt); - intel_release_load_detect_pipe(&crt->base, connector, - &tmp); + intel_release_load_detect_pipe(connector, &tmp); } else status = connector_status_unknown; @@ -603,25 +668,15 @@ static void intel_crt_reset(struct drm_connector *connector) * Routines for controlling stuff on the analog port */ -static const struct drm_encoder_helper_funcs pch_encoder_funcs = { +static const struct drm_encoder_helper_funcs crt_encoder_funcs = { .mode_fixup = intel_crt_mode_fixup, - .prepare = intel_encoder_prepare, - .commit = intel_encoder_commit, .mode_set = intel_crt_mode_set, - .dpms = pch_crt_dpms, -}; - -static const struct drm_encoder_helper_funcs gmch_encoder_funcs = { - .mode_fixup = intel_crt_mode_fixup, - .prepare = intel_encoder_prepare, - .commit = intel_encoder_commit, - .mode_set = intel_crt_mode_set, - .dpms = gmch_crt_dpms, + .disable = intel_encoder_noop, }; static const struct drm_connector_funcs intel_crt_connector_funcs = { .reset = intel_crt_reset, - .dpms = drm_helper_connector_dpms, + .dpms = intel_crt_dpms, .detect = intel_crt_detect, .fill_modes = drm_helper_probe_single_connector_modes, .destroy = intel_crt_destroy, @@ -662,7 +717,6 @@ void intel_crt_init(struct drm_device *dev) struct intel_crt *crt; struct intel_connector *intel_connector; struct drm_i915_private *dev_priv = dev->dev_private; - const struct drm_encoder_helper_funcs *encoder_helper_funcs; /* Skip machines without VGA that falsely report hotplug events */ if (dmi_check_system(intel_no_crt)) @@ -688,13 +742,11 @@ void intel_crt_init(struct drm_device *dev) intel_connector_attach_encoder(intel_connector, &crt->base); crt->base.type = INTEL_OUTPUT_ANALOG; - crt->base.clone_mask = (1 << INTEL_SDVO_NON_TV_CLONE_BIT | - 1 << INTEL_ANALOG_CLONE_BIT | - 1 << INTEL_SDVO_LVDS_CLONE_BIT); + crt->base.cloneable = true; if (IS_HASWELL(dev)) crt->base.crtc_mask = (1 << 0); else - crt->base.crtc_mask = (1 << 0) | (1 << 1); + crt->base.crtc_mask = (1 << 0) | (1 << 1) | (1 << 2); if (IS_GEN2(dev)) connector->interlace_allowed = 0; @@ -703,11 +755,18 @@ void intel_crt_init(struct drm_device *dev) connector->doublescan_allowed = 0; if (HAS_PCH_SPLIT(dev)) - encoder_helper_funcs = &pch_encoder_funcs; + crt->adpa_reg = PCH_ADPA; + else if (IS_VALLEYVIEW(dev)) + crt->adpa_reg = VLV_ADPA; else - encoder_helper_funcs = &gmch_encoder_funcs; + crt->adpa_reg = ADPA; + + crt->base.disable = intel_disable_crt; + crt->base.enable = intel_enable_crt; + crt->base.get_hw_state = intel_crt_get_hw_state; + intel_connector->get_hw_state = intel_connector_get_hw_state; - drm_encoder_helper_add(&crt->base.base, encoder_helper_funcs); + drm_encoder_helper_add(&crt->base.base, &crt_encoder_funcs); drm_connector_helper_add(connector, &intel_crt_connector_helper_funcs); drm_sysfs_connector_add(connector); diff --git a/drivers/gpu/drm/i915/intel_ddi.c b/drivers/gpu/drm/i915/intel_ddi.c index 933c74859172..bfe375466a0e 100644 --- a/drivers/gpu/drm/i915/intel_ddi.c +++ b/drivers/gpu/drm/i915/intel_ddi.c @@ -250,7 +250,7 @@ void intel_ddi_init(struct drm_device *dev, enum port port) case PORT_B: case PORT_C: case PORT_D: - intel_hdmi_init(dev, DDI_BUF_CTL(port)); + intel_hdmi_init(dev, DDI_BUF_CTL(port), port); break; default: DRM_DEBUG_DRIVER("No handlers defined for port %d, skipping DDI initialization\n", @@ -267,7 +267,8 @@ struct wrpll_tmds_clock { u16 r2; /* Reference divider */ }; -/* Table of matching values for WRPLL clocks programming for each frequency */ +/* Table of matching values for WRPLL clocks programming for each frequency. + * The code assumes this table is sorted. */ static const struct wrpll_tmds_clock wrpll_tmds_clock_table[] = { {19750, 38, 25, 18}, {20000, 48, 32, 18}, @@ -277,7 +278,6 @@ static const struct wrpll_tmds_clock wrpll_tmds_clock_table[] = { {23000, 36, 23, 15}, {23500, 40, 40, 23}, {23750, 26, 16, 14}, - {23750, 26, 16, 14}, {24000, 36, 24, 15}, {25000, 36, 25, 15}, {25175, 26, 40, 33}, @@ -437,7 +437,6 @@ static const struct wrpll_tmds_clock wrpll_tmds_clock_table[] = { {108000, 8, 24, 15}, {108108, 8, 173, 108}, {109000, 6, 23, 19}, - {109000, 6, 23, 19}, {110000, 6, 22, 18}, {110013, 6, 22, 18}, {110250, 8, 49, 30}, @@ -614,7 +613,6 @@ static const struct wrpll_tmds_clock wrpll_tmds_clock_table[] = { {218250, 4, 42, 26}, {218750, 4, 34, 21}, {219000, 4, 47, 29}, - {219000, 4, 47, 29}, {220000, 4, 44, 27}, {220640, 4, 49, 30}, {220750, 4, 36, 22}, @@ -658,7 +656,7 @@ void intel_ddi_mode_set(struct drm_encoder *encoder, struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(encoder); int port = intel_hdmi->ddi_port; int pipe = intel_crtc->pipe; - int p, n2, r2, valid=0; + int p, n2, r2; u32 temp, i; /* On Haswell, we need to enable the clocks and prepare DDI function to @@ -666,26 +664,23 @@ void intel_ddi_mode_set(struct drm_encoder *encoder, */ DRM_DEBUG_KMS("Preparing HDMI DDI mode for Haswell on port %c, pipe %c\n", port_name(port), pipe_name(pipe)); - for (i=0; i < ARRAY_SIZE(wrpll_tmds_clock_table); i++) { - if (crtc->mode.clock == wrpll_tmds_clock_table[i].clock) { - p = wrpll_tmds_clock_table[i].p; - n2 = wrpll_tmds_clock_table[i].n2; - r2 = wrpll_tmds_clock_table[i].r2; + for (i = 0; i < ARRAY_SIZE(wrpll_tmds_clock_table); i++) + if (crtc->mode.clock <= wrpll_tmds_clock_table[i].clock) + break; - DRM_DEBUG_KMS("WR PLL clock: found settings for %dKHz refresh rate: p=%d, n2=%d, r2=%d\n", - crtc->mode.clock, - p, n2, r2); + if (i == ARRAY_SIZE(wrpll_tmds_clock_table)) + i--; - valid = 1; - break; - } - } + p = wrpll_tmds_clock_table[i].p; + n2 = wrpll_tmds_clock_table[i].n2; + r2 = wrpll_tmds_clock_table[i].r2; - if (!valid) { - DRM_ERROR("Unable to find WR PLL clock settings for %dKHz refresh rate\n", - crtc->mode.clock); - return; - } + if (wrpll_tmds_clock_table[i].clock != crtc->mode.clock) + DRM_INFO("WR PLL: using settings for %dKHz on %dKHz mode\n", + wrpll_tmds_clock_table[i].clock, crtc->mode.clock); + + DRM_DEBUG_KMS("WR PLL: %dKHz refresh rate with p=%d, n2=%d r2=%d\n", + crtc->mode.clock, p, n2, r2); /* Enable LCPLL if disabled */ temp = I915_READ(LCPLL_CTL); @@ -718,46 +713,107 @@ void intel_ddi_mode_set(struct drm_encoder *encoder, /* Proper support for digital audio needs a new logic and a new set * of registers, so we leave it for future patch bombing. */ - DRM_DEBUG_DRIVER("HDMI audio on pipe %c not yet supported on DDI\n", + DRM_DEBUG_DRIVER("HDMI audio on pipe %c on DDI\n", pipe_name(intel_crtc->pipe)); + + /* write eld */ + DRM_DEBUG_DRIVER("HDMI audio: write eld information\n"); + intel_write_eld(encoder, adjusted_mode); } /* Enable PIPE_DDI_FUNC_CTL for the pipe to work in HDMI mode */ - temp = I915_READ(DDI_FUNC_CTL(pipe)); - temp &= ~PIPE_DDI_PORT_MASK; - temp &= ~PIPE_DDI_BPC_12; - temp |= PIPE_DDI_SELECT_PORT(port) | - PIPE_DDI_MODE_SELECT_HDMI | - ((intel_crtc->bpp > 24) ? - PIPE_DDI_BPC_12 : - PIPE_DDI_BPC_8) | - PIPE_DDI_FUNC_ENABLE; + temp = PIPE_DDI_FUNC_ENABLE | PIPE_DDI_SELECT_PORT(port); + + switch (intel_crtc->bpp) { + case 18: + temp |= PIPE_DDI_BPC_6; + break; + case 24: + temp |= PIPE_DDI_BPC_8; + break; + case 30: + temp |= PIPE_DDI_BPC_10; + break; + case 36: + temp |= PIPE_DDI_BPC_12; + break; + default: + WARN(1, "%d bpp unsupported by pipe DDI function\n", + intel_crtc->bpp); + } + + if (intel_hdmi->has_hdmi_sink) + temp |= PIPE_DDI_MODE_SELECT_HDMI; + else + temp |= PIPE_DDI_MODE_SELECT_DVI; + + if (adjusted_mode->flags & DRM_MODE_FLAG_PVSYNC) + temp |= PIPE_DDI_PVSYNC; + if (adjusted_mode->flags & DRM_MODE_FLAG_PHSYNC) + temp |= PIPE_DDI_PHSYNC; I915_WRITE(DDI_FUNC_CTL(pipe), temp); intel_hdmi->set_infoframes(encoder, adjusted_mode); } -void intel_ddi_dpms(struct drm_encoder *encoder, int mode) +bool intel_ddi_get_hw_state(struct intel_encoder *encoder, + enum pipe *pipe) { - struct drm_device *dev = encoder->dev; + struct drm_device *dev = encoder->base.dev; struct drm_i915_private *dev_priv = dev->dev_private; - struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(encoder); + struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(&encoder->base); + u32 tmp; + int i; + + tmp = I915_READ(DDI_BUF_CTL(intel_hdmi->ddi_port)); + + if (!(tmp & DDI_BUF_CTL_ENABLE)) + return false; + + for_each_pipe(i) { + tmp = I915_READ(DDI_FUNC_CTL(i)); + + if ((tmp & PIPE_DDI_PORT_MASK) + == PIPE_DDI_SELECT_PORT(intel_hdmi->ddi_port)) { + *pipe = i; + return true; + } + } + + DRM_DEBUG_KMS("No pipe for ddi port %i found\n", intel_hdmi->ddi_port); + + return true; +} + +void intel_enable_ddi(struct intel_encoder *encoder) +{ + struct drm_device *dev = encoder->base.dev; + struct drm_i915_private *dev_priv = dev->dev_private; + struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(&encoder->base); int port = intel_hdmi->ddi_port; u32 temp; temp = I915_READ(DDI_BUF_CTL(port)); - - if (mode != DRM_MODE_DPMS_ON) { - temp &= ~DDI_BUF_CTL_ENABLE; - } else { - temp |= DDI_BUF_CTL_ENABLE; - } + temp |= DDI_BUF_CTL_ENABLE; /* Enable DDI_BUF_CTL. In HDMI/DVI mode, the port width, * and swing/emphasis values are ignored so nothing special needs * to be done besides enabling the port. */ - I915_WRITE(DDI_BUF_CTL(port), - temp); + I915_WRITE(DDI_BUF_CTL(port), temp); +} + +void intel_disable_ddi(struct intel_encoder *encoder) +{ + struct drm_device *dev = encoder->base.dev; + struct drm_i915_private *dev_priv = dev->dev_private; + struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(&encoder->base); + int port = intel_hdmi->ddi_port; + u32 temp; + + temp = I915_READ(DDI_BUF_CTL(port)); + temp &= ~DDI_BUF_CTL_ENABLE; + + I915_WRITE(DDI_BUF_CTL(port), temp); } diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c index c040aee1341c..6f5aafa1b633 100644 --- a/drivers/gpu/drm/i915/intel_display.c +++ b/drivers/gpu/drm/i915/intel_display.c @@ -1006,7 +1006,7 @@ void intel_wait_for_pipe_off(struct drm_device *dev, int pipe) /* Wait for the Pipe State to go off */ if (wait_for((I915_READ(reg) & I965_PIPECONF_ACTIVE) == 0, 100)) - DRM_DEBUG_KMS("pipe_off wait timed out\n"); + WARN(1, "pipe_off wait timed out\n"); } else { u32 last_line, line_mask; int reg = PIPEDSL(pipe); @@ -1024,7 +1024,7 @@ void intel_wait_for_pipe_off(struct drm_device *dev, int pipe) } while (((I915_READ(reg) & line_mask) != last_line) && time_after(timeout, jiffies)); if (time_after(jiffies, timeout)) - DRM_DEBUG_KMS("pipe_off wait timed out\n"); + WARN(1, "pipe_off wait timed out\n"); } } @@ -1431,6 +1431,8 @@ static void assert_pch_ports_disabled(struct drm_i915_private *dev_priv, * protect mechanism may be enabled. * * Note! This is for pre-ILK only. + * + * Unfortunately needed by dvo_ns2501 since the dvo depends on it running. */ static void intel_enable_pll(struct drm_i915_private *dev_priv, enum pipe pipe) { @@ -1860,59 +1862,6 @@ static void intel_disable_plane(struct drm_i915_private *dev_priv, intel_wait_for_vblank(dev_priv->dev, pipe); } -static void disable_pch_dp(struct drm_i915_private *dev_priv, - enum pipe pipe, int reg, u32 port_sel) -{ - u32 val = I915_READ(reg); - if (dp_pipe_enabled(dev_priv, pipe, port_sel, val)) { - DRM_DEBUG_KMS("Disabling pch dp %x on pipe %d\n", reg, pipe); - I915_WRITE(reg, val & ~DP_PORT_EN); - } -} - -static void disable_pch_hdmi(struct drm_i915_private *dev_priv, - enum pipe pipe, int reg) -{ - u32 val = I915_READ(reg); - if (hdmi_pipe_enabled(dev_priv, pipe, val)) { - DRM_DEBUG_KMS("Disabling pch HDMI %x on pipe %d\n", - reg, pipe); - I915_WRITE(reg, val & ~PORT_ENABLE); - } -} - -/* Disable any ports connected to this transcoder */ -static void intel_disable_pch_ports(struct drm_i915_private *dev_priv, - enum pipe pipe) -{ - u32 reg, val; - - val = I915_READ(PCH_PP_CONTROL); - I915_WRITE(PCH_PP_CONTROL, val | PANEL_UNLOCK_REGS); - - disable_pch_dp(dev_priv, pipe, PCH_DP_B, TRANS_DP_PORT_SEL_B); - disable_pch_dp(dev_priv, pipe, PCH_DP_C, TRANS_DP_PORT_SEL_C); - disable_pch_dp(dev_priv, pipe, PCH_DP_D, TRANS_DP_PORT_SEL_D); - - reg = PCH_ADPA; - val = I915_READ(reg); - if (adpa_pipe_enabled(dev_priv, pipe, val)) - I915_WRITE(reg, val & ~ADPA_DAC_ENABLE); - - reg = PCH_LVDS; - val = I915_READ(reg); - if (lvds_pipe_enabled(dev_priv, pipe, val)) { - DRM_DEBUG_KMS("disable lvds on pipe %d val 0x%08x\n", pipe, val); - I915_WRITE(reg, val & ~LVDS_PORT_EN); - POSTING_READ(reg); - udelay(100); - } - - disable_pch_hdmi(dev_priv, pipe, HDMIB); - disable_pch_hdmi(dev_priv, pipe, HDMIC); - disable_pch_hdmi(dev_priv, pipe, HDMID); -} - int intel_pin_and_fence_fb_obj(struct drm_device *dev, struct drm_i915_gem_object *obj, @@ -2201,16 +2150,17 @@ intel_finish_fb(struct drm_framebuffer *old_fb) static int intel_pipe_set_base(struct drm_crtc *crtc, int x, int y, - struct drm_framebuffer *old_fb) + struct drm_framebuffer *fb) { struct drm_device *dev = crtc->dev; struct drm_i915_private *dev_priv = dev->dev_private; struct drm_i915_master_private *master_priv; struct intel_crtc *intel_crtc = to_intel_crtc(crtc); + struct drm_framebuffer *old_fb; int ret; /* no fb bound */ - if (!crtc->fb) { + if (!fb) { DRM_ERROR("No FB bound\n"); return 0; } @@ -2224,7 +2174,7 @@ intel_pipe_set_base(struct drm_crtc *crtc, int x, int y, mutex_lock(&dev->struct_mutex); ret = intel_pin_and_fence_fb_obj(dev, - to_intel_framebuffer(crtc->fb)->obj, + to_intel_framebuffer(fb)->obj, NULL); if (ret != 0) { mutex_unlock(&dev->struct_mutex); @@ -2232,17 +2182,22 @@ intel_pipe_set_base(struct drm_crtc *crtc, int x, int y, return ret; } - if (old_fb) - intel_finish_fb(old_fb); + if (crtc->fb) + intel_finish_fb(crtc->fb); - ret = dev_priv->display.update_plane(crtc, crtc->fb, x, y); + ret = dev_priv->display.update_plane(crtc, fb, x, y); if (ret) { - intel_unpin_fb_obj(to_intel_framebuffer(crtc->fb)->obj); + intel_unpin_fb_obj(to_intel_framebuffer(fb)->obj); mutex_unlock(&dev->struct_mutex); DRM_ERROR("failed to update base address\n"); return ret; } + old_fb = crtc->fb; + crtc->fb = fb; + crtc->x = x; + crtc->y = y; + if (old_fb) { intel_wait_for_vblank(dev, intel_crtc->pipe); intel_unpin_fb_obj(to_intel_framebuffer(old_fb)->obj); @@ -2709,11 +2664,10 @@ static void ivb_manual_fdi_link_train(struct drm_crtc *crtc) DRM_DEBUG_KMS("FDI train done.\n"); } -static void ironlake_fdi_pll_enable(struct drm_crtc *crtc) +static void ironlake_fdi_pll_enable(struct intel_crtc *intel_crtc) { - struct drm_device *dev = crtc->dev; + struct drm_device *dev = intel_crtc->base.dev; struct drm_i915_private *dev_priv = dev->dev_private; - struct intel_crtc *intel_crtc = to_intel_crtc(crtc); int pipe = intel_crtc->pipe; u32 reg, temp; @@ -2754,6 +2708,35 @@ static void ironlake_fdi_pll_enable(struct drm_crtc *crtc) } } +static void ironlake_fdi_pll_disable(struct intel_crtc *intel_crtc) +{ + struct drm_device *dev = intel_crtc->base.dev; + struct drm_i915_private *dev_priv = dev->dev_private; + int pipe = intel_crtc->pipe; + u32 reg, temp; + + /* Switch from PCDclk to Rawclk */ + reg = FDI_RX_CTL(pipe); + temp = I915_READ(reg); + I915_WRITE(reg, temp & ~FDI_PCDCLK); + + /* Disable CPU FDI TX PLL */ + reg = FDI_TX_CTL(pipe); + temp = I915_READ(reg); + I915_WRITE(reg, temp & ~FDI_TX_PLL_ENABLE); + + POSTING_READ(reg); + udelay(100); + + reg = FDI_RX_CTL(pipe); + temp = I915_READ(reg); + I915_WRITE(reg, temp & ~FDI_RX_PLL_ENABLE); + + /* Wait for the clocks to turn off. */ + POSTING_READ(reg); + udelay(100); +} + static void cpt_phase_pointer_disable(struct drm_device *dev, int pipe) { struct drm_i915_private *dev_priv = dev->dev_private; @@ -2838,13 +2821,13 @@ static void intel_crtc_wait_for_pending_flips(struct drm_crtc *crtc) static bool intel_crtc_driving_pch(struct drm_crtc *crtc) { struct drm_device *dev = crtc->dev; - struct intel_encoder *encoder; + struct intel_encoder *intel_encoder; /* * If there's a non-PCH eDP on this crtc, it must be DP_A, and that * must be driven by its own crtc; no sharing is possible. */ - for_each_encoder_on_crtc(dev, crtc, encoder) { + for_each_encoder_on_crtc(dev, crtc, intel_encoder) { /* On Haswell, LPT PCH handles the VGA connection via FDI, and Haswell * CPU handles all others */ @@ -2852,19 +2835,19 @@ static bool intel_crtc_driving_pch(struct drm_crtc *crtc) /* It is still unclear how this will work on PPT, so throw up a warning */ WARN_ON(!HAS_PCH_LPT(dev)); - if (encoder->type == DRM_MODE_ENCODER_DAC) { + if (intel_encoder->type == INTEL_OUTPUT_ANALOG) { DRM_DEBUG_KMS("Haswell detected DAC encoder, assuming is PCH\n"); return true; } else { DRM_DEBUG_KMS("Haswell detected encoder %d, assuming is CPU\n", - encoder->type); + intel_encoder->type); return false; } } - switch (encoder->type) { + switch (intel_encoder->type) { case INTEL_OUTPUT_EDP: - if (!intel_encoder_is_pch_edp(&encoder->base)) + if (!intel_encoder_is_pch_edp(&intel_encoder->base)) return false; continue; } @@ -3181,11 +3164,14 @@ static void ironlake_crtc_enable(struct drm_crtc *crtc) struct drm_device *dev = crtc->dev; struct drm_i915_private *dev_priv = dev->dev_private; struct intel_crtc *intel_crtc = to_intel_crtc(crtc); + struct intel_encoder *encoder; int pipe = intel_crtc->pipe; int plane = intel_crtc->plane; u32 temp; bool is_pch_port; + WARN_ON(!crtc->enabled); + if (intel_crtc->active) return; @@ -3200,10 +3186,16 @@ static void ironlake_crtc_enable(struct drm_crtc *crtc) is_pch_port = intel_crtc_driving_pch(crtc); - if (is_pch_port) - ironlake_fdi_pll_enable(crtc); - else - ironlake_fdi_disable(crtc); + if (is_pch_port) { + ironlake_fdi_pll_enable(intel_crtc); + } else { + assert_fdi_tx_disabled(dev_priv, pipe); + assert_fdi_rx_disabled(dev_priv, pipe); + } + + for_each_encoder_on_crtc(dev, crtc, encoder) + if (encoder->pre_enable) + encoder->pre_enable(encoder); /* Enable panel fitting for LVDS */ if (dev_priv->pch_pf_size && @@ -3234,6 +3226,12 @@ static void ironlake_crtc_enable(struct drm_crtc *crtc) mutex_unlock(&dev->struct_mutex); intel_crtc_update_cursor(crtc, true); + + for_each_encoder_on_crtc(dev, crtc, encoder) + encoder->enable(encoder); + + if (HAS_PCH_CPT(dev)) + intel_cpt_verify_modeset(dev, intel_crtc->pipe); } static void ironlake_crtc_disable(struct drm_crtc *crtc) @@ -3241,13 +3239,18 @@ static void ironlake_crtc_disable(struct drm_crtc *crtc) struct drm_device *dev = crtc->dev; struct drm_i915_private *dev_priv = dev->dev_private; struct intel_crtc *intel_crtc = to_intel_crtc(crtc); + struct intel_encoder *encoder; int pipe = intel_crtc->pipe; int plane = intel_crtc->plane; u32 reg, temp; + if (!intel_crtc->active) return; + for_each_encoder_on_crtc(dev, crtc, encoder) + encoder->disable(encoder); + intel_crtc_wait_for_pending_flips(crtc); drm_vblank_off(dev, pipe); intel_crtc_update_cursor(crtc, false); @@ -3263,14 +3266,11 @@ static void ironlake_crtc_disable(struct drm_crtc *crtc) I915_WRITE(PF_CTL(pipe), 0); I915_WRITE(PF_WIN_SZ(pipe), 0); - ironlake_fdi_disable(crtc); + for_each_encoder_on_crtc(dev, crtc, encoder) + if (encoder->post_disable) + encoder->post_disable(encoder); - /* This is a horrible layering violation; we should be doing this in - * the connector/encoder ->prepare instead, but we don't always have - * enough information there about the config to know whether it will - * actually be necessary or just cause undesired flicker. - */ - intel_disable_pch_ports(dev_priv, pipe); + ironlake_fdi_disable(crtc); intel_disable_transcoder(dev_priv, pipe); @@ -3304,26 +3304,7 @@ static void ironlake_crtc_disable(struct drm_crtc *crtc) /* disable PCH DPLL */ intel_disable_pch_pll(intel_crtc); - /* Switch from PCDclk to Rawclk */ - reg = FDI_RX_CTL(pipe); - temp = I915_READ(reg); - I915_WRITE(reg, temp & ~FDI_PCDCLK); - - /* Disable CPU FDI TX PLL */ - reg = FDI_TX_CTL(pipe); - temp = I915_READ(reg); - I915_WRITE(reg, temp & ~FDI_TX_PLL_ENABLE); - - POSTING_READ(reg); - udelay(100); - - reg = FDI_RX_CTL(pipe); - temp = I915_READ(reg); - I915_WRITE(reg, temp & ~FDI_RX_PLL_ENABLE); - - /* Wait for the clocks to turn off. */ - POSTING_READ(reg); - udelay(100); + ironlake_fdi_pll_disable(intel_crtc); intel_crtc->active = false; intel_update_watermarks(dev); @@ -3333,30 +3314,6 @@ static void ironlake_crtc_disable(struct drm_crtc *crtc) mutex_unlock(&dev->struct_mutex); } -static void ironlake_crtc_dpms(struct drm_crtc *crtc, int mode) -{ - struct intel_crtc *intel_crtc = to_intel_crtc(crtc); - int pipe = intel_crtc->pipe; - int plane = intel_crtc->plane; - - /* XXX: When our outputs are all unaware of DPMS modes other than off - * and on, we should map those modes to DRM_MODE_DPMS_OFF in the CRTC. - */ - switch (mode) { - case DRM_MODE_DPMS_ON: - case DRM_MODE_DPMS_STANDBY: - case DRM_MODE_DPMS_SUSPEND: - DRM_DEBUG_KMS("crtc %d/%d dpms on\n", pipe, plane); - ironlake_crtc_enable(crtc); - break; - - case DRM_MODE_DPMS_OFF: - DRM_DEBUG_KMS("crtc %d/%d dpms off\n", pipe, plane); - ironlake_crtc_disable(crtc); - break; - } -} - static void ironlake_crtc_off(struct drm_crtc *crtc) { struct intel_crtc *intel_crtc = to_intel_crtc(crtc); @@ -3386,9 +3343,12 @@ static void i9xx_crtc_enable(struct drm_crtc *crtc) struct drm_device *dev = crtc->dev; struct drm_i915_private *dev_priv = dev->dev_private; struct intel_crtc *intel_crtc = to_intel_crtc(crtc); + struct intel_encoder *encoder; int pipe = intel_crtc->pipe; int plane = intel_crtc->plane; + WARN_ON(!crtc->enabled); + if (intel_crtc->active) return; @@ -3405,6 +3365,9 @@ static void i9xx_crtc_enable(struct drm_crtc *crtc) /* Give the overlay scaler a chance to enable if it's on this pipe */ intel_crtc_dpms_overlay(intel_crtc, true); intel_crtc_update_cursor(crtc, true); + + for_each_encoder_on_crtc(dev, crtc, encoder) + encoder->enable(encoder); } static void i9xx_crtc_disable(struct drm_crtc *crtc) @@ -3412,12 +3375,17 @@ static void i9xx_crtc_disable(struct drm_crtc *crtc) struct drm_device *dev = crtc->dev; struct drm_i915_private *dev_priv = dev->dev_private; struct intel_crtc *intel_crtc = to_intel_crtc(crtc); + struct intel_encoder *encoder; int pipe = intel_crtc->pipe; int plane = intel_crtc->plane; + if (!intel_crtc->active) return; + for_each_encoder_on_crtc(dev, crtc, encoder) + encoder->disable(encoder); + /* Give the overlay scaler a chance to disable if it's on this pipe */ intel_crtc_wait_for_pending_flips(crtc); drm_vblank_off(dev, pipe); @@ -3436,45 +3404,17 @@ static void i9xx_crtc_disable(struct drm_crtc *crtc) intel_update_watermarks(dev); } -static void i9xx_crtc_dpms(struct drm_crtc *crtc, int mode) -{ - /* XXX: When our outputs are all unaware of DPMS modes other than off - * and on, we should map those modes to DRM_MODE_DPMS_OFF in the CRTC. - */ - switch (mode) { - case DRM_MODE_DPMS_ON: - case DRM_MODE_DPMS_STANDBY: - case DRM_MODE_DPMS_SUSPEND: - i9xx_crtc_enable(crtc); - break; - case DRM_MODE_DPMS_OFF: - i9xx_crtc_disable(crtc); - break; - } -} - static void i9xx_crtc_off(struct drm_crtc *crtc) { } -/** - * Sets the power management mode of the pipe and plane. - */ -static void intel_crtc_dpms(struct drm_crtc *crtc, int mode) +static void intel_crtc_update_sarea(struct drm_crtc *crtc, + bool enabled) { struct drm_device *dev = crtc->dev; - struct drm_i915_private *dev_priv = dev->dev_private; struct drm_i915_master_private *master_priv; struct intel_crtc *intel_crtc = to_intel_crtc(crtc); int pipe = intel_crtc->pipe; - bool enabled; - - if (intel_crtc->dpms_mode == mode) - return; - - intel_crtc->dpms_mode = mode; - - dev_priv->display.dpms(crtc, mode); if (!dev->primary->master) return; @@ -3483,8 +3423,6 @@ static void intel_crtc_dpms(struct drm_crtc *crtc, int mode) if (!master_priv->sarea_priv) return; - enabled = crtc->enabled && mode != DRM_MODE_DPMS_OFF; - switch (pipe) { case 0: master_priv->sarea_priv->pipeA_w = enabled ? crtc->mode.hdisplay : 0; @@ -3500,13 +3438,42 @@ static void intel_crtc_dpms(struct drm_crtc *crtc, int mode) } } +/** + * Sets the power management mode of the pipe and plane. + */ +void intel_crtc_update_dpms(struct drm_crtc *crtc) +{ + struct drm_device *dev = crtc->dev; + struct drm_i915_private *dev_priv = dev->dev_private; + struct intel_encoder *intel_encoder; + bool enable = false; + + for_each_encoder_on_crtc(dev, crtc, intel_encoder) + enable |= intel_encoder->connectors_active; + + if (enable) + dev_priv->display.crtc_enable(crtc); + else + dev_priv->display.crtc_disable(crtc); + + intel_crtc_update_sarea(crtc, enable); +} + +static void intel_crtc_noop(struct drm_crtc *crtc) +{ +} + static void intel_crtc_disable(struct drm_crtc *crtc) { - struct drm_crtc_helper_funcs *crtc_funcs = crtc->helper_private; struct drm_device *dev = crtc->dev; + struct drm_connector *connector; struct drm_i915_private *dev_priv = dev->dev_private; - crtc_funcs->dpms(crtc, DRM_MODE_DPMS_OFF); + /* crtc should still be enabled when we disable it. */ + WARN_ON(!crtc->enabled); + + dev_priv->display.crtc_disable(crtc); + intel_crtc_update_sarea(crtc, false); dev_priv->display.off(crtc); assert_plane_disabled(dev->dev_private, to_intel_crtc(crtc)->plane); @@ -3516,63 +3483,128 @@ static void intel_crtc_disable(struct drm_crtc *crtc) mutex_lock(&dev->struct_mutex); intel_unpin_fb_obj(to_intel_framebuffer(crtc->fb)->obj); mutex_unlock(&dev->struct_mutex); + crtc->fb = NULL; + } + + /* Update computed state. */ + list_for_each_entry(connector, &dev->mode_config.connector_list, head) { + if (!connector->encoder || !connector->encoder->crtc) + continue; + + if (connector->encoder->crtc != crtc) + continue; + + connector->dpms = DRM_MODE_DPMS_OFF; + to_intel_encoder(connector->encoder)->connectors_active = false; } } -/* Prepare for a mode set. - * - * Note we could be a lot smarter here. We need to figure out which outputs - * will be enabled, which disabled (in short, how the config will changes) - * and perform the minimum necessary steps to accomplish that, e.g. updating - * watermarks, FBC configuration, making sure PLLs are programmed correctly, - * panel fitting is in the proper state, etc. - */ -static void i9xx_crtc_prepare(struct drm_crtc *crtc) +void intel_modeset_disable(struct drm_device *dev) { - i9xx_crtc_disable(crtc); + struct drm_crtc *crtc; + + list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) { + if (crtc->enabled) + intel_crtc_disable(crtc); + } } -static void i9xx_crtc_commit(struct drm_crtc *crtc) +void intel_encoder_noop(struct drm_encoder *encoder) { - i9xx_crtc_enable(crtc); } -static void ironlake_crtc_prepare(struct drm_crtc *crtc) +void intel_encoder_destroy(struct drm_encoder *encoder) { - ironlake_crtc_disable(crtc); + struct intel_encoder *intel_encoder = to_intel_encoder(encoder); + + drm_encoder_cleanup(encoder); + kfree(intel_encoder); } -static void ironlake_crtc_commit(struct drm_crtc *crtc) +/* Simple dpms helper for encodres with just one connector, no cloning and only + * one kind of off state. It clamps all !ON modes to fully OFF and changes the + * state of the entire output pipe. */ +void intel_encoder_dpms(struct intel_encoder *encoder, int mode) { - ironlake_crtc_enable(crtc); + if (mode == DRM_MODE_DPMS_ON) { + encoder->connectors_active = true; + + intel_crtc_update_dpms(encoder->base.crtc); + } else { + encoder->connectors_active = false; + + intel_crtc_update_dpms(encoder->base.crtc); + } } -void intel_encoder_prepare(struct drm_encoder *encoder) +/* Cross check the actual hw state with our own modeset state tracking (and it's + * internal consistency). */ +static void intel_connector_check_state(struct intel_connector *connector) { - struct drm_encoder_helper_funcs *encoder_funcs = encoder->helper_private; - /* lvds has its own version of prepare see intel_lvds_prepare */ - encoder_funcs->dpms(encoder, DRM_MODE_DPMS_OFF); + if (connector->get_hw_state(connector)) { + struct intel_encoder *encoder = connector->encoder; + struct drm_crtc *crtc; + bool encoder_enabled; + enum pipe pipe; + + DRM_DEBUG_KMS("[CONNECTOR:%d:%s]\n", + connector->base.base.id, + drm_get_connector_name(&connector->base)); + + WARN(connector->base.dpms == DRM_MODE_DPMS_OFF, + "wrong connector dpms state\n"); + WARN(connector->base.encoder != &encoder->base, + "active connector not linked to encoder\n"); + WARN(!encoder->connectors_active, + "encoder->connectors_active not set\n"); + + encoder_enabled = encoder->get_hw_state(encoder, &pipe); + WARN(!encoder_enabled, "encoder not enabled\n"); + if (WARN_ON(!encoder->base.crtc)) + return; + + crtc = encoder->base.crtc; + + WARN(!crtc->enabled, "crtc not enabled\n"); + WARN(!to_intel_crtc(crtc)->active, "crtc not active\n"); + WARN(pipe != to_intel_crtc(crtc)->pipe, + "encoder active on the wrong pipe\n"); + } } -void intel_encoder_commit(struct drm_encoder *encoder) +/* Even simpler default implementation, if there's really no special case to + * consider. */ +void intel_connector_dpms(struct drm_connector *connector, int mode) { - struct drm_encoder_helper_funcs *encoder_funcs = encoder->helper_private; - struct drm_device *dev = encoder->dev; - struct intel_crtc *intel_crtc = to_intel_crtc(encoder->crtc); + struct intel_encoder *encoder = intel_attached_encoder(connector); - /* lvds has its own version of commit see intel_lvds_commit */ - encoder_funcs->dpms(encoder, DRM_MODE_DPMS_ON); + /* All the simple cases only support two dpms states. */ + if (mode != DRM_MODE_DPMS_ON) + mode = DRM_MODE_DPMS_OFF; - if (HAS_PCH_CPT(dev)) - intel_cpt_verify_modeset(dev, intel_crtc->pipe); + if (mode == connector->dpms) + return; + + connector->dpms = mode; + + /* Only need to change hw state when actually enabled */ + if (encoder->base.crtc) + intel_encoder_dpms(encoder, mode); + else + WARN_ON(encoder->connectors_active != false); + + intel_modeset_check_state(connector->dev); } -void intel_encoder_destroy(struct drm_encoder *encoder) +/* Simple connector->get_hw_state implementation for encoders that support only + * one connector and no cloning and hence the encoder state determines the state + * of the connector. */ +bool intel_connector_get_hw_state(struct intel_connector *connector) { - struct intel_encoder *intel_encoder = to_intel_encoder(encoder); + enum pipe pipe = 0; + struct intel_encoder *encoder = connector->encoder; - drm_encoder_cleanup(encoder); - kfree(intel_encoder); + return encoder->get_hw_state(encoder, &pipe); } static bool intel_crtc_mode_fixup(struct drm_crtc *crtc, @@ -3593,6 +3625,13 @@ static bool intel_crtc_mode_fixup(struct drm_crtc *crtc, if (!(adjusted_mode->private_flags & INTEL_MODE_CRTC_TIMINGS_SET)) drm_mode_set_crtcinfo(adjusted_mode, 0); + /* WaPruneModeWithIncorrectHsyncOffset: Cantiga+ cannot handle modes + * with a hsync front porch of 0. + */ + if ((INTEL_INFO(dev)->gen > 4 || IS_G4X(dev)) && + adjusted_mode->hsync_start == adjusted_mode->hdisplay) + return false; + return true; } @@ -3728,6 +3767,7 @@ static inline bool intel_panel_use_ssc(struct drm_i915_private *dev_priv) * true if they don't match). */ static bool intel_choose_pipe_bpp_dither(struct drm_crtc *crtc, + struct drm_framebuffer *fb, unsigned int *pipe_bpp, struct drm_display_mode *mode) { @@ -3797,7 +3837,7 @@ static bool intel_choose_pipe_bpp_dither(struct drm_crtc *crtc, * also stays within the max display bpc discovered above. */ - switch (crtc->fb->depth) { + switch (fb->depth) { case 8: bpc = 8; /* since we go through a colormap */ break; @@ -4216,7 +4256,7 @@ static int i9xx_crtc_mode_set(struct drm_crtc *crtc, struct drm_display_mode *mode, struct drm_display_mode *adjusted_mode, int x, int y, - struct drm_framebuffer *old_fb) + struct drm_framebuffer *fb) { struct drm_device *dev = crtc->dev; struct drm_i915_private *dev_priv = dev->dev_private; @@ -4406,7 +4446,7 @@ static int i9xx_crtc_mode_set(struct drm_crtc *crtc, I915_WRITE(DSPCNTR(plane), dspcntr); POSTING_READ(DSPCNTR(plane)); - ret = intel_pipe_set_base(crtc, x, y, old_fb); + ret = intel_pipe_set_base(crtc, x, y, fb); intel_update_watermarks(dev); @@ -4560,24 +4600,130 @@ static int ironlake_get_refclk(struct drm_crtc *crtc) return 120000; } +static void ironlake_set_pipeconf(struct drm_crtc *crtc, + struct drm_display_mode *adjusted_mode, + bool dither) +{ + struct drm_i915_private *dev_priv = crtc->dev->dev_private; + struct intel_crtc *intel_crtc = to_intel_crtc(crtc); + int pipe = intel_crtc->pipe; + uint32_t val; + + val = I915_READ(PIPECONF(pipe)); + + val &= ~PIPE_BPC_MASK; + switch (intel_crtc->bpp) { + case 18: + val |= PIPE_6BPC; + break; + case 24: + val |= PIPE_8BPC; + break; + case 30: + val |= PIPE_10BPC; + break; + case 36: + val |= PIPE_12BPC; + break; + default: + val |= PIPE_8BPC; + break; + } + + val &= ~(PIPECONF_DITHER_EN | PIPECONF_DITHER_TYPE_MASK); + if (dither) + val |= (PIPECONF_DITHER_EN | PIPECONF_DITHER_TYPE_SP); + + val &= ~PIPECONF_INTERLACE_MASK; + if (adjusted_mode->flags & DRM_MODE_FLAG_INTERLACE) + val |= PIPECONF_INTERLACED_ILK; + else + val |= PIPECONF_PROGRESSIVE; + + I915_WRITE(PIPECONF(pipe), val); + POSTING_READ(PIPECONF(pipe)); +} + +static bool ironlake_compute_clocks(struct drm_crtc *crtc, + struct drm_display_mode *adjusted_mode, + intel_clock_t *clock, + bool *has_reduced_clock, + intel_clock_t *reduced_clock) +{ + struct drm_device *dev = crtc->dev; + struct drm_i915_private *dev_priv = dev->dev_private; + struct intel_encoder *intel_encoder; + int refclk; + const intel_limit_t *limit; + bool ret, is_sdvo = false, is_tv = false, is_lvds = false; + + for_each_encoder_on_crtc(dev, crtc, intel_encoder) { + switch (intel_encoder->type) { + case INTEL_OUTPUT_LVDS: + is_lvds = true; + break; + case INTEL_OUTPUT_SDVO: + case INTEL_OUTPUT_HDMI: + is_sdvo = true; + if (intel_encoder->needs_tv_clock) + is_tv = true; + break; + case INTEL_OUTPUT_TVOUT: + is_tv = true; + break; + } + } + + refclk = ironlake_get_refclk(crtc); + + /* + * Returns a set of divisors for the desired target clock with the given + * refclk, or FALSE. The returned values represent the clock equation: + * reflck * (5 * (m1 + 2) + (m2 + 2)) / (n + 2) / p1 / p2. + */ + limit = intel_limit(crtc, refclk); + ret = limit->find_pll(limit, crtc, adjusted_mode->clock, refclk, NULL, + clock); + if (!ret) + return false; + + if (is_lvds && dev_priv->lvds_downclock_avail) { + /* + * Ensure we match the reduced clock's P to the target clock. + * If the clocks don't match, we can't switch the display clock + * by using the FP0/FP1. In such case we will disable the LVDS + * downclock feature. + */ + *has_reduced_clock = limit->find_pll(limit, crtc, + dev_priv->lvds_downclock, + refclk, + clock, + reduced_clock); + } + + if (is_sdvo && is_tv) + i9xx_adjust_sdvo_tv_clock(adjusted_mode, clock); + + return true; +} + static int ironlake_crtc_mode_set(struct drm_crtc *crtc, struct drm_display_mode *mode, struct drm_display_mode *adjusted_mode, int x, int y, - struct drm_framebuffer *old_fb) + struct drm_framebuffer *fb) { struct drm_device *dev = crtc->dev; struct drm_i915_private *dev_priv = dev->dev_private; struct intel_crtc *intel_crtc = to_intel_crtc(crtc); int pipe = intel_crtc->pipe; int plane = intel_crtc->plane; - int refclk, num_connectors = 0; + int num_connectors = 0; intel_clock_t clock, reduced_clock; - u32 dpll, fp = 0, fp2 = 0, dspcntr, pipeconf; + u32 dpll, fp = 0, fp2 = 0; bool ok, has_reduced_clock = false, is_sdvo = false; bool is_crt = false, is_lvds = false, is_tv = false, is_dp = false; struct intel_encoder *encoder, *edp_encoder = NULL; - const intel_limit_t *limit; int ret; struct fdi_m_n m_n = {0}; u32 temp; @@ -4619,16 +4765,8 @@ static int ironlake_crtc_mode_set(struct drm_crtc *crtc, num_connectors++; } - refclk = ironlake_get_refclk(crtc); - - /* - * Returns a set of divisors for the desired target clock with the given - * refclk, or FALSE. The returned values represent the clock equation: - * reflck * (5 * (m1 + 2) + (m2 + 2)) / (n + 2) / p1 / p2. - */ - limit = intel_limit(crtc, refclk); - ok = limit->find_pll(limit, crtc, adjusted_mode->clock, refclk, NULL, - &clock); + ok = ironlake_compute_clocks(crtc, adjusted_mode, &clock, + &has_reduced_clock, &reduced_clock); if (!ok) { DRM_ERROR("Couldn't find PLL settings for mode!\n"); return -EINVAL; @@ -4637,24 +4775,6 @@ static int ironlake_crtc_mode_set(struct drm_crtc *crtc, /* Ensure that the cursor is valid for the new mode before changing... */ intel_crtc_update_cursor(crtc, true); - if (is_lvds && dev_priv->lvds_downclock_avail) { - /* - * Ensure we match the reduced clock's P to the target clock. - * If the clocks don't match, we can't switch the display clock - * by using the FP0/FP1. In such case we will disable the LVDS - * downclock feature. - */ - has_reduced_clock = limit->find_pll(limit, crtc, - dev_priv->lvds_downclock, - refclk, - &clock, - &reduced_clock); - } - - if (is_sdvo && is_tv) - i9xx_adjust_sdvo_tv_clock(adjusted_mode, &clock); - - /* FDI link */ pixel_multiplier = intel_mode_get_pixel_multiplier(adjusted_mode); lane = 0; @@ -4682,32 +4802,17 @@ static int ironlake_crtc_mode_set(struct drm_crtc *crtc, target_clock = adjusted_mode->clock; /* determine panel color depth */ - temp = I915_READ(PIPECONF(pipe)); - temp &= ~PIPE_BPC_MASK; - dither = intel_choose_pipe_bpp_dither(crtc, &pipe_bpp, mode); - switch (pipe_bpp) { - case 18: - temp |= PIPE_6BPC; - break; - case 24: - temp |= PIPE_8BPC; - break; - case 30: - temp |= PIPE_10BPC; - break; - case 36: - temp |= PIPE_12BPC; - break; - default: + dither = intel_choose_pipe_bpp_dither(crtc, fb, &pipe_bpp, mode); + if (is_lvds && dev_priv->lvds_dither) + dither = true; + + if (pipe_bpp != 18 && pipe_bpp != 24 && pipe_bpp != 30 && + pipe_bpp != 36) { WARN(1, "intel_choose_pipe_bpp returned invalid value %d\n", - pipe_bpp); - temp |= PIPE_8BPC; + pipe_bpp); pipe_bpp = 24; - break; } - intel_crtc->bpp = pipe_bpp; - I915_WRITE(PIPECONF(pipe), temp); if (!lane) { /* @@ -4791,12 +4896,6 @@ static int ironlake_crtc_mode_set(struct drm_crtc *crtc, else dpll |= PLL_REF_INPUT_DREFCLK; - /* setup pipeconf */ - pipeconf = I915_READ(PIPECONF(pipe)); - - /* Set up the display plane register */ - dspcntr = DISPPLANE_GAMMA_ENABLE; - DRM_DEBUG_KMS("Mode for pipe %d:\n", pipe); drm_mode_debug_printmodeline(mode); @@ -4856,12 +4955,6 @@ static int ironlake_crtc_mode_set(struct drm_crtc *crtc, I915_WRITE(PCH_LVDS, temp); } - pipeconf &= ~PIPECONF_DITHER_EN; - pipeconf &= ~PIPECONF_DITHER_TYPE_MASK; - if ((is_lvds && dev_priv->lvds_dither) || dither) { - pipeconf |= PIPECONF_DITHER_EN; - pipeconf |= PIPECONF_DITHER_TYPE_SP; - } if (is_dp && !is_cpu_edp) { intel_dp_set_m_n(crtc, mode, adjusted_mode); } else { @@ -4897,9 +4990,7 @@ static int ironlake_crtc_mode_set(struct drm_crtc *crtc, } } - pipeconf &= ~PIPECONF_INTERLACE_MASK; if (adjusted_mode->flags & DRM_MODE_FLAG_INTERLACE) { - pipeconf |= PIPECONF_INTERLACED_ILK; /* the chip adds 2 halflines automatically */ adjusted_mode->crtc_vtotal -= 1; adjusted_mode->crtc_vblank_end -= 1; @@ -4907,7 +4998,6 @@ static int ironlake_crtc_mode_set(struct drm_crtc *crtc, adjusted_mode->crtc_hsync_start - adjusted_mode->crtc_htotal/2); } else { - pipeconf |= PIPECONF_PROGRESSIVE; I915_WRITE(VSYNCSHIFT(pipe), 0); } @@ -4945,15 +5035,15 @@ static int ironlake_crtc_mode_set(struct drm_crtc *crtc, if (is_cpu_edp) ironlake_set_pll_edp(crtc, adjusted_mode->clock); - I915_WRITE(PIPECONF(pipe), pipeconf); - POSTING_READ(PIPECONF(pipe)); + ironlake_set_pipeconf(crtc, adjusted_mode, dither); intel_wait_for_vblank(dev, pipe); - I915_WRITE(DSPCNTR(plane), dspcntr); + /* Set up the display plane register */ + I915_WRITE(DSPCNTR(plane), DISPPLANE_GAMMA_ENABLE); POSTING_READ(DSPCNTR(plane)); - ret = intel_pipe_set_base(crtc, x, y, old_fb); + ret = intel_pipe_set_base(crtc, x, y, fb); intel_update_watermarks(dev); @@ -4966,7 +5056,7 @@ static int intel_crtc_mode_set(struct drm_crtc *crtc, struct drm_display_mode *mode, struct drm_display_mode *adjusted_mode, int x, int y, - struct drm_framebuffer *old_fb) + struct drm_framebuffer *fb) { struct drm_device *dev = crtc->dev; struct drm_i915_private *dev_priv = dev->dev_private; @@ -4977,14 +5067,9 @@ static int intel_crtc_mode_set(struct drm_crtc *crtc, drm_vblank_pre_modeset(dev, pipe); ret = dev_priv->display.crtc_mode_set(crtc, mode, adjusted_mode, - x, y, old_fb); + x, y, fb); drm_vblank_post_modeset(dev, pipe); - if (ret) - intel_crtc->dpms_mode = DRM_MODE_DPMS_OFF; - else - intel_crtc->dpms_mode = DRM_MODE_DPMS_ON; - return ret; } @@ -5057,6 +5142,91 @@ static void g4x_write_eld(struct drm_connector *connector, I915_WRITE(G4X_AUD_CNTL_ST, i); } +static void haswell_write_eld(struct drm_connector *connector, + struct drm_crtc *crtc) +{ + struct drm_i915_private *dev_priv = connector->dev->dev_private; + uint8_t *eld = connector->eld; + struct drm_device *dev = crtc->dev; + uint32_t eldv; + uint32_t i; + int len; + int pipe = to_intel_crtc(crtc)->pipe; + int tmp; + + int hdmiw_hdmiedid = HSW_AUD_EDID_DATA(pipe); + int aud_cntl_st = HSW_AUD_DIP_ELD_CTRL(pipe); + int aud_config = HSW_AUD_CFG(pipe); + int aud_cntrl_st2 = HSW_AUD_PIN_ELD_CP_VLD; + + + DRM_DEBUG_DRIVER("HDMI: Haswell Audio initialize....\n"); + + /* Audio output enable */ + DRM_DEBUG_DRIVER("HDMI audio: enable codec\n"); + tmp = I915_READ(aud_cntrl_st2); + tmp |= (AUDIO_OUTPUT_ENABLE_A << (pipe * 4)); + I915_WRITE(aud_cntrl_st2, tmp); + + /* Wait for 1 vertical blank */ + intel_wait_for_vblank(dev, pipe); + + /* Set ELD valid state */ + tmp = I915_READ(aud_cntrl_st2); + DRM_DEBUG_DRIVER("HDMI audio: pin eld vld status=0x%8x\n", tmp); + tmp |= (AUDIO_ELD_VALID_A << (pipe * 4)); + I915_WRITE(aud_cntrl_st2, tmp); + tmp = I915_READ(aud_cntrl_st2); + DRM_DEBUG_DRIVER("HDMI audio: eld vld status=0x%8x\n", tmp); + + /* Enable HDMI mode */ + tmp = I915_READ(aud_config); + DRM_DEBUG_DRIVER("HDMI audio: audio conf: 0x%8x\n", tmp); + /* clear N_programing_enable and N_value_index */ + tmp &= ~(AUD_CONFIG_N_VALUE_INDEX | AUD_CONFIG_N_PROG_ENABLE); + I915_WRITE(aud_config, tmp); + + DRM_DEBUG_DRIVER("ELD on pipe %c\n", pipe_name(pipe)); + + eldv = AUDIO_ELD_VALID_A << (pipe * 4); + + if (intel_pipe_has_type(crtc, INTEL_OUTPUT_DISPLAYPORT)) { + DRM_DEBUG_DRIVER("ELD: DisplayPort detected\n"); + eld[5] |= (1 << 2); /* Conn_Type, 0x1 = DisplayPort */ + I915_WRITE(aud_config, AUD_CONFIG_N_VALUE_INDEX); /* 0x1 = DP */ + } else + I915_WRITE(aud_config, 0); + + if (intel_eld_uptodate(connector, + aud_cntrl_st2, eldv, + aud_cntl_st, IBX_ELD_ADDRESS, + hdmiw_hdmiedid)) + return; + + i = I915_READ(aud_cntrl_st2); + i &= ~eldv; + I915_WRITE(aud_cntrl_st2, i); + + if (!eld[0]) + return; + + i = I915_READ(aud_cntl_st); + i &= ~IBX_ELD_ADDRESS; + I915_WRITE(aud_cntl_st, i); + i = (i >> 29) & DIP_PORT_SEL_MASK; /* DIP_Port_Select, 0x1 = PortB */ + DRM_DEBUG_DRIVER("port num:%d\n", i); + + len = min_t(uint8_t, eld[2], 21); /* 84 bytes of hw ELD buffer */ + DRM_DEBUG_DRIVER("ELD size %d\n", len); + for (i = 0; i < len; i++) + I915_WRITE(hdmiw_hdmiedid, *((uint32_t *)eld + i)); + + i = I915_READ(aud_cntrl_st2); + i |= eldv; + I915_WRITE(aud_cntrl_st2, i); + +} + static void ironlake_write_eld(struct drm_connector *connector, struct drm_crtc *crtc) { @@ -5069,28 +5239,24 @@ static void ironlake_write_eld(struct drm_connector *connector, int aud_config; int aud_cntl_st; int aud_cntrl_st2; + int pipe = to_intel_crtc(crtc)->pipe; if (HAS_PCH_IBX(connector->dev)) { - hdmiw_hdmiedid = IBX_HDMIW_HDMIEDID_A; - aud_config = IBX_AUD_CONFIG_A; - aud_cntl_st = IBX_AUD_CNTL_ST_A; + hdmiw_hdmiedid = IBX_HDMIW_HDMIEDID(pipe); + aud_config = IBX_AUD_CFG(pipe); + aud_cntl_st = IBX_AUD_CNTL_ST(pipe); aud_cntrl_st2 = IBX_AUD_CNTL_ST2; } else { - hdmiw_hdmiedid = CPT_HDMIW_HDMIEDID_A; - aud_config = CPT_AUD_CONFIG_A; - aud_cntl_st = CPT_AUD_CNTL_ST_A; + hdmiw_hdmiedid = CPT_HDMIW_HDMIEDID(pipe); + aud_config = CPT_AUD_CFG(pipe); + aud_cntl_st = CPT_AUD_CNTL_ST(pipe); aud_cntrl_st2 = CPT_AUD_CNTRL_ST2; } - i = to_intel_crtc(crtc)->pipe; - hdmiw_hdmiedid += i * 0x100; - aud_cntl_st += i * 0x100; - aud_config += i * 0x100; - - DRM_DEBUG_DRIVER("ELD on pipe %c\n", pipe_name(i)); + DRM_DEBUG_DRIVER("ELD on pipe %c\n", pipe_name(pipe)); i = I915_READ(aud_cntl_st); - i = (i >> 29) & 0x3; /* DIP_Port_Select, 0x1 = PortB */ + i = (i >> 29) & DIP_PORT_SEL_MASK; /* DIP_Port_Select, 0x1 = PortB */ if (!i) { DRM_DEBUG_DRIVER("Audio directed to unknown port\n"); /* operate blindly on all ports */ @@ -5337,8 +5503,6 @@ static int intel_crtc_cursor_set(struct drm_crtc *crtc, uint32_t addr; int ret; - DRM_DEBUG_KMS("\n"); - /* if we want to turn off the cursor ignore width and height */ if (!handle) { DRM_DEBUG_KMS("cursor off\n"); @@ -5584,17 +5748,18 @@ mode_fits_in_fbdev(struct drm_device *dev, return fb; } -bool intel_get_load_detect_pipe(struct intel_encoder *intel_encoder, - struct drm_connector *connector, +bool intel_get_load_detect_pipe(struct drm_connector *connector, struct drm_display_mode *mode, struct intel_load_detect_pipe *old) { struct intel_crtc *intel_crtc; + struct intel_encoder *intel_encoder = + intel_attached_encoder(connector); struct drm_crtc *possible_crtc; struct drm_encoder *encoder = &intel_encoder->base; struct drm_crtc *crtc = NULL; struct drm_device *dev = encoder->dev; - struct drm_framebuffer *old_fb; + struct drm_framebuffer *fb; int i = -1; DRM_DEBUG_KMS("[CONNECTOR:%d:%s], [ENCODER:%d:%s]\n", @@ -5615,21 +5780,12 @@ bool intel_get_load_detect_pipe(struct intel_encoder *intel_encoder, if (encoder->crtc) { crtc = encoder->crtc; - intel_crtc = to_intel_crtc(crtc); - old->dpms_mode = intel_crtc->dpms_mode; + old->dpms_mode = connector->dpms; old->load_detect_temp = false; /* Make sure the crtc and connector are running */ - if (intel_crtc->dpms_mode != DRM_MODE_DPMS_ON) { - struct drm_encoder_helper_funcs *encoder_funcs; - struct drm_crtc_helper_funcs *crtc_funcs; - - crtc_funcs = crtc->helper_private; - crtc_funcs->dpms(crtc, DRM_MODE_DPMS_ON); - - encoder_funcs = encoder->helper_private; - encoder_funcs->dpms(encoder, DRM_MODE_DPMS_ON); - } + if (connector->dpms != DRM_MODE_DPMS_ON) + connector->funcs->dpms(connector, DRM_MODE_DPMS_ON); return true; } @@ -5653,19 +5809,17 @@ bool intel_get_load_detect_pipe(struct intel_encoder *intel_encoder, return false; } - encoder->crtc = crtc; - connector->encoder = encoder; + intel_encoder->new_crtc = to_intel_crtc(crtc); + to_intel_connector(connector)->new_encoder = intel_encoder; intel_crtc = to_intel_crtc(crtc); - old->dpms_mode = intel_crtc->dpms_mode; + old->dpms_mode = connector->dpms; old->load_detect_temp = true; old->release_fb = NULL; if (!mode) mode = &load_detect_mode; - old_fb = crtc->fb; - /* We need a framebuffer large enough to accommodate all accesses * that the plane may generate whilst we perform load detection. * We can not rely on the fbcon either being present (we get called @@ -5673,50 +5827,52 @@ bool intel_get_load_detect_pipe(struct intel_encoder *intel_encoder, * not even exist) or that it is large enough to satisfy the * requested mode. */ - crtc->fb = mode_fits_in_fbdev(dev, mode); - if (crtc->fb == NULL) { + fb = mode_fits_in_fbdev(dev, mode); + if (fb == NULL) { DRM_DEBUG_KMS("creating tmp fb for load-detection\n"); - crtc->fb = intel_framebuffer_create_for_mode(dev, mode, 24, 32); - old->release_fb = crtc->fb; + fb = intel_framebuffer_create_for_mode(dev, mode, 24, 32); + old->release_fb = fb; } else DRM_DEBUG_KMS("reusing fbdev for load-detection framebuffer\n"); - if (IS_ERR(crtc->fb)) { + if (IS_ERR(fb)) { DRM_DEBUG_KMS("failed to allocate framebuffer for load-detection\n"); - crtc->fb = old_fb; - return false; + goto fail; } - if (!drm_crtc_helper_set_mode(crtc, mode, 0, 0, old_fb)) { + if (!intel_set_mode(crtc, mode, 0, 0, fb)) { DRM_DEBUG_KMS("failed to set mode on load-detect pipe\n"); if (old->release_fb) old->release_fb->funcs->destroy(old->release_fb); - crtc->fb = old_fb; - return false; + goto fail; } /* let the connector get through one full cycle before testing */ intel_wait_for_vblank(dev, intel_crtc->pipe); return true; +fail: + connector->encoder = NULL; + encoder->crtc = NULL; + return false; } -void intel_release_load_detect_pipe(struct intel_encoder *intel_encoder, - struct drm_connector *connector, +void intel_release_load_detect_pipe(struct drm_connector *connector, struct intel_load_detect_pipe *old) { + struct intel_encoder *intel_encoder = + intel_attached_encoder(connector); struct drm_encoder *encoder = &intel_encoder->base; - struct drm_device *dev = encoder->dev; - struct drm_crtc *crtc = encoder->crtc; - struct drm_encoder_helper_funcs *encoder_funcs = encoder->helper_private; - struct drm_crtc_helper_funcs *crtc_funcs = crtc->helper_private; DRM_DEBUG_KMS("[CONNECTOR:%d:%s], [ENCODER:%d:%s]\n", connector->base.id, drm_get_connector_name(connector), encoder->base.id, drm_get_encoder_name(encoder)); if (old->load_detect_temp) { - connector->encoder = NULL; - drm_helper_disable_unused_functions(dev); + struct drm_crtc *crtc = encoder->crtc; + + to_intel_connector(connector)->new_encoder = NULL; + intel_encoder->new_crtc = NULL; + intel_set_mode(crtc, NULL, 0, 0, NULL); if (old->release_fb) old->release_fb->funcs->destroy(old->release_fb); @@ -5725,10 +5881,8 @@ void intel_release_load_detect_pipe(struct intel_encoder *intel_encoder, } /* Switch crtc and encoder back off if necessary */ - if (old->dpms_mode != DRM_MODE_DPMS_ON) { - encoder_funcs->dpms(encoder, old->dpms_mode); - crtc_funcs->dpms(crtc, old->dpms_mode); - } + if (old->dpms_mode != DRM_MODE_DPMS_ON) + connector->funcs->dpms(connector, old->dpms_mode); } /* Returns the clock of the currently programmed mode of the given pipe. */ @@ -5850,46 +6004,6 @@ struct drm_display_mode *intel_crtc_mode_get(struct drm_device *dev, return mode; } -#define GPU_IDLE_TIMEOUT 500 /* ms */ - -/* When this timer fires, we've been idle for awhile */ -static void intel_gpu_idle_timer(unsigned long arg) -{ - struct drm_device *dev = (struct drm_device *)arg; - drm_i915_private_t *dev_priv = dev->dev_private; - - if (!list_empty(&dev_priv->mm.active_list)) { - /* Still processing requests, so just re-arm the timer. */ - mod_timer(&dev_priv->idle_timer, jiffies + - msecs_to_jiffies(GPU_IDLE_TIMEOUT)); - return; - } - - dev_priv->busy = false; - queue_work(dev_priv->wq, &dev_priv->idle_work); -} - -#define CRTC_IDLE_TIMEOUT 1000 /* ms */ - -static void intel_crtc_idle_timer(unsigned long arg) -{ - struct intel_crtc *intel_crtc = (struct intel_crtc *)arg; - struct drm_crtc *crtc = &intel_crtc->base; - drm_i915_private_t *dev_priv = crtc->dev->dev_private; - struct intel_framebuffer *intel_fb; - - intel_fb = to_intel_framebuffer(crtc->fb); - if (intel_fb && intel_fb->obj->active) { - /* The framebuffer is still being accessed by the GPU. */ - mod_timer(&intel_crtc->idle_timer, jiffies + - msecs_to_jiffies(CRTC_IDLE_TIMEOUT)); - return; - } - - intel_crtc->busy = false; - queue_work(dev_priv->wq, &dev_priv->idle_work); -} - static void intel_increase_pllclock(struct drm_crtc *crtc) { struct drm_device *dev = crtc->dev; @@ -5919,10 +6033,6 @@ static void intel_increase_pllclock(struct drm_crtc *crtc) if (dpll & DISPLAY_RATE_SELECT_FPA1) DRM_DEBUG_DRIVER("failed to upclock LVDS!\n"); } - - /* Schedule downclock */ - mod_timer(&intel_crtc->idle_timer, jiffies + - msecs_to_jiffies(CRTC_IDLE_TIMEOUT)); } static void intel_decrease_pllclock(struct drm_crtc *crtc) @@ -5961,89 +6071,46 @@ static void intel_decrease_pllclock(struct drm_crtc *crtc) } -/** - * intel_idle_update - adjust clocks for idleness - * @work: work struct - * - * Either the GPU or display (or both) went idle. Check the busy status - * here and adjust the CRTC and GPU clocks as necessary. - */ -static void intel_idle_update(struct work_struct *work) +void intel_mark_busy(struct drm_device *dev) { - drm_i915_private_t *dev_priv = container_of(work, drm_i915_private_t, - idle_work); - struct drm_device *dev = dev_priv->dev; + i915_update_gfx_val(dev->dev_private); +} + +void intel_mark_idle(struct drm_device *dev) +{ +} + +void intel_mark_fb_busy(struct drm_i915_gem_object *obj) +{ + struct drm_device *dev = obj->base.dev; struct drm_crtc *crtc; - struct intel_crtc *intel_crtc; if (!i915_powersave) return; - mutex_lock(&dev->struct_mutex); - - i915_update_gfx_val(dev_priv); - list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) { - /* Skip inactive CRTCs */ if (!crtc->fb) continue; - intel_crtc = to_intel_crtc(crtc); - if (!intel_crtc->busy) - intel_decrease_pllclock(crtc); + if (to_intel_framebuffer(crtc->fb)->obj == obj) + intel_increase_pllclock(crtc); } - - - mutex_unlock(&dev->struct_mutex); } -/** - * intel_mark_busy - mark the GPU and possibly the display busy - * @dev: drm device - * @obj: object we're operating on - * - * Callers can use this function to indicate that the GPU is busy processing - * commands. If @obj matches one of the CRTC objects (i.e. it's a scanout - * buffer), we'll also mark the display as busy, so we know to increase its - * clock frequency. - */ -void intel_mark_busy(struct drm_device *dev, struct drm_i915_gem_object *obj) +void intel_mark_fb_idle(struct drm_i915_gem_object *obj) { - drm_i915_private_t *dev_priv = dev->dev_private; - struct drm_crtc *crtc = NULL; - struct intel_framebuffer *intel_fb; - struct intel_crtc *intel_crtc; - - if (!drm_core_check_feature(dev, DRIVER_MODESET)) - return; - - if (!dev_priv->busy) { - intel_sanitize_pm(dev); - dev_priv->busy = true; - } else - mod_timer(&dev_priv->idle_timer, jiffies + - msecs_to_jiffies(GPU_IDLE_TIMEOUT)); + struct drm_device *dev = obj->base.dev; + struct drm_crtc *crtc; - if (obj == NULL) + if (!i915_powersave) return; list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) { if (!crtc->fb) continue; - intel_crtc = to_intel_crtc(crtc); - intel_fb = to_intel_framebuffer(crtc->fb); - if (intel_fb->obj == obj) { - if (!intel_crtc->busy) { - /* Non-busy -> busy, upclock */ - intel_increase_pllclock(crtc); - intel_crtc->busy = true; - } else { - /* Busy -> busy, put off timer */ - mod_timer(&intel_crtc->idle_timer, jiffies + - msecs_to_jiffies(CRTC_IDLE_TIMEOUT)); - } - } + if (to_intel_framebuffer(crtc->fb)->obj == obj) + intel_decrease_pllclock(crtc); } } @@ -6394,7 +6461,7 @@ static int intel_gen7_queue_flip(struct drm_device *dev, default: WARN_ONCE(1, "unknown plane in flip command\n"); ret = -ENODEV; - goto err; + goto err_unpin; } ret = intel_ring_begin(ring, 4); @@ -6502,7 +6569,7 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc, goto cleanup_pending; intel_disable_fbc(dev); - intel_mark_busy(dev, obj); + intel_mark_fb_busy(obj); mutex_unlock(&dev->struct_mutex); trace_i915_flip_request(intel_crtc->plane, obj); @@ -6527,81 +6594,807 @@ free_work: return ret; } -static void intel_sanitize_modesetting(struct drm_device *dev, - int pipe, int plane) +static struct drm_crtc_helper_funcs intel_helper_funcs = { + .mode_set_base_atomic = intel_pipe_set_base_atomic, + .load_lut = intel_crtc_load_lut, + .disable = intel_crtc_noop, +}; + +bool intel_encoder_check_is_cloned(struct intel_encoder *encoder) { - struct drm_i915_private *dev_priv = dev->dev_private; - u32 reg, val; - int i; + struct intel_encoder *other_encoder; + struct drm_crtc *crtc = &encoder->new_crtc->base; - /* Clear any frame start delays used for debugging left by the BIOS */ - for_each_pipe(i) { - reg = PIPECONF(i); - I915_WRITE(reg, I915_READ(reg) & ~PIPECONF_FRAME_START_DELAY_MASK); + if (WARN_ON(!crtc)) + return false; + + list_for_each_entry(other_encoder, + &crtc->dev->mode_config.encoder_list, + base.head) { + + if (&other_encoder->new_crtc->base != crtc || + encoder == other_encoder) + continue; + else + return true; } - if (HAS_PCH_SPLIT(dev)) - return; + return false; +} - /* Who knows what state these registers were left in by the BIOS or - * grub? - * - * If we leave the registers in a conflicting state (e.g. with the - * display plane reading from the other pipe than the one we intend - * to use) then when we attempt to teardown the active mode, we will - * not disable the pipes and planes in the correct order -- leaving - * a plane reading from a disabled pipe and possibly leading to - * undefined behaviour. +static bool intel_encoder_crtc_ok(struct drm_encoder *encoder, + struct drm_crtc *crtc) +{ + struct drm_device *dev; + struct drm_crtc *tmp; + int crtc_mask = 1; + + WARN(!crtc, "checking null crtc?\n"); + + dev = crtc->dev; + + list_for_each_entry(tmp, &dev->mode_config.crtc_list, head) { + if (tmp == crtc) + break; + crtc_mask <<= 1; + } + + if (encoder->possible_crtcs & crtc_mask) + return true; + return false; +} + +/** + * intel_modeset_update_staged_output_state + * + * Updates the staged output configuration state, e.g. after we've read out the + * current hw state. + */ +static void intel_modeset_update_staged_output_state(struct drm_device *dev) +{ + struct intel_encoder *encoder; + struct intel_connector *connector; + + list_for_each_entry(connector, &dev->mode_config.connector_list, + base.head) { + connector->new_encoder = + to_intel_encoder(connector->base.encoder); + } + + list_for_each_entry(encoder, &dev->mode_config.encoder_list, + base.head) { + encoder->new_crtc = + to_intel_crtc(encoder->base.crtc); + } +} + +/** + * intel_modeset_commit_output_state + * + * This function copies the stage display pipe configuration to the real one. + */ +static void intel_modeset_commit_output_state(struct drm_device *dev) +{ + struct intel_encoder *encoder; + struct intel_connector *connector; + + list_for_each_entry(connector, &dev->mode_config.connector_list, + base.head) { + connector->base.encoder = &connector->new_encoder->base; + } + + list_for_each_entry(encoder, &dev->mode_config.encoder_list, + base.head) { + encoder->base.crtc = &encoder->new_crtc->base; + } +} + +static struct drm_display_mode * +intel_modeset_adjusted_mode(struct drm_crtc *crtc, + struct drm_display_mode *mode) +{ + struct drm_device *dev = crtc->dev; + struct drm_display_mode *adjusted_mode; + struct drm_encoder_helper_funcs *encoder_funcs; + struct intel_encoder *encoder; + + adjusted_mode = drm_mode_duplicate(dev, mode); + if (!adjusted_mode) + return ERR_PTR(-ENOMEM); + + /* Pass our mode to the connectors and the CRTC to give them a chance to + * adjust it according to limitations or connector properties, and also + * a chance to reject the mode entirely. */ + list_for_each_entry(encoder, &dev->mode_config.encoder_list, + base.head) { - reg = DSPCNTR(plane); - val = I915_READ(reg); + if (&encoder->new_crtc->base != crtc) + continue; + encoder_funcs = encoder->base.helper_private; + if (!(encoder_funcs->mode_fixup(&encoder->base, mode, + adjusted_mode))) { + DRM_DEBUG_KMS("Encoder fixup failed\n"); + goto fail; + } + } - if ((val & DISPLAY_PLANE_ENABLE) == 0) - return; - if (!!(val & DISPPLANE_SEL_PIPE_MASK) == pipe) - return; + if (!(intel_crtc_mode_fixup(crtc, mode, adjusted_mode))) { + DRM_DEBUG_KMS("CRTC fixup failed\n"); + goto fail; + } + DRM_DEBUG_KMS("[CRTC:%d]\n", crtc->base.id); - /* This display plane is active and attached to the other CPU pipe. */ - pipe = !pipe; + return adjusted_mode; +fail: + drm_mode_destroy(dev, adjusted_mode); + return ERR_PTR(-EINVAL); +} - /* Disable the plane and wait for it to stop reading from the pipe. */ - intel_disable_plane(dev_priv, plane, pipe); - intel_disable_pipe(dev_priv, pipe); +/* Computes which crtcs are affected and sets the relevant bits in the mask. For + * simplicity we use the crtc's pipe number (because it's easier to obtain). */ +static void +intel_modeset_affected_pipes(struct drm_crtc *crtc, unsigned *modeset_pipes, + unsigned *prepare_pipes, unsigned *disable_pipes) +{ + struct intel_crtc *intel_crtc; + struct drm_device *dev = crtc->dev; + struct intel_encoder *encoder; + struct intel_connector *connector; + struct drm_crtc *tmp_crtc; + + *disable_pipes = *modeset_pipes = *prepare_pipes = 0; + + /* Check which crtcs have changed outputs connected to them, these need + * to be part of the prepare_pipes mask. We don't (yet) support global + * modeset across multiple crtcs, so modeset_pipes will only have one + * bit set at most. */ + list_for_each_entry(connector, &dev->mode_config.connector_list, + base.head) { + if (connector->base.encoder == &connector->new_encoder->base) + continue; + + if (connector->base.encoder) { + tmp_crtc = connector->base.encoder->crtc; + + *prepare_pipes |= 1 << to_intel_crtc(tmp_crtc)->pipe; + } + + if (connector->new_encoder) + *prepare_pipes |= + 1 << connector->new_encoder->new_crtc->pipe; + } + + list_for_each_entry(encoder, &dev->mode_config.encoder_list, + base.head) { + if (encoder->base.crtc == &encoder->new_crtc->base) + continue; + + if (encoder->base.crtc) { + tmp_crtc = encoder->base.crtc; + + *prepare_pipes |= 1 << to_intel_crtc(tmp_crtc)->pipe; + } + + if (encoder->new_crtc) + *prepare_pipes |= 1 << encoder->new_crtc->pipe; + } + + /* Check for any pipes that will be fully disabled ... */ + list_for_each_entry(intel_crtc, &dev->mode_config.crtc_list, + base.head) { + bool used = false; + + /* Don't try to disable disabled crtcs. */ + if (!intel_crtc->base.enabled) + continue; + + list_for_each_entry(encoder, &dev->mode_config.encoder_list, + base.head) { + if (encoder->new_crtc == intel_crtc) + used = true; + } + + if (!used) + *disable_pipes |= 1 << intel_crtc->pipe; + } + + + /* set_mode is also used to update properties on life display pipes. */ + intel_crtc = to_intel_crtc(crtc); + if (crtc->enabled) + *prepare_pipes |= 1 << intel_crtc->pipe; + + /* We only support modeset on one single crtc, hence we need to do that + * only for the passed in crtc iff we change anything else than just + * disable crtcs. + * + * This is actually not true, to be fully compatible with the old crtc + * helper we automatically disable _any_ output (i.e. doesn't need to be + * connected to the crtc we're modesetting on) if it's disconnected. + * Which is a rather nutty api (since changed the output configuration + * without userspace's explicit request can lead to confusion), but + * alas. Hence we currently need to modeset on all pipes we prepare. */ + if (*prepare_pipes) + *modeset_pipes = *prepare_pipes; + + /* ... and mask these out. */ + *modeset_pipes &= ~(*disable_pipes); + *prepare_pipes &= ~(*disable_pipes); } -static void intel_crtc_reset(struct drm_crtc *crtc) +static bool intel_crtc_in_use(struct drm_crtc *crtc) { + struct drm_encoder *encoder; struct drm_device *dev = crtc->dev; - struct intel_crtc *intel_crtc = to_intel_crtc(crtc); - /* Reset flags back to the 'unknown' status so that they - * will be correctly set on the initial modeset. + list_for_each_entry(encoder, &dev->mode_config.encoder_list, head) + if (encoder->crtc == crtc) + return true; + + return false; +} + +static void +intel_modeset_update_state(struct drm_device *dev, unsigned prepare_pipes) +{ + struct intel_encoder *intel_encoder; + struct intel_crtc *intel_crtc; + struct drm_connector *connector; + + list_for_each_entry(intel_encoder, &dev->mode_config.encoder_list, + base.head) { + if (!intel_encoder->base.crtc) + continue; + + intel_crtc = to_intel_crtc(intel_encoder->base.crtc); + + if (prepare_pipes & (1 << intel_crtc->pipe)) + intel_encoder->connectors_active = false; + } + + intel_modeset_commit_output_state(dev); + + /* Update computed state. */ + list_for_each_entry(intel_crtc, &dev->mode_config.crtc_list, + base.head) { + intel_crtc->base.enabled = intel_crtc_in_use(&intel_crtc->base); + } + + list_for_each_entry(connector, &dev->mode_config.connector_list, head) { + if (!connector->encoder || !connector->encoder->crtc) + continue; + + intel_crtc = to_intel_crtc(connector->encoder->crtc); + + if (prepare_pipes & (1 << intel_crtc->pipe)) { + struct drm_property *dpms_property = + dev->mode_config.dpms_property; + + connector->dpms = DRM_MODE_DPMS_ON; + drm_connector_property_set_value(connector, + dpms_property, + DRM_MODE_DPMS_ON); + + intel_encoder = to_intel_encoder(connector->encoder); + intel_encoder->connectors_active = true; + } + } + +} + +#define for_each_intel_crtc_masked(dev, mask, intel_crtc) \ + list_for_each_entry((intel_crtc), \ + &(dev)->mode_config.crtc_list, \ + base.head) \ + if (mask & (1 <<(intel_crtc)->pipe)) \ + +void +intel_modeset_check_state(struct drm_device *dev) +{ + struct intel_crtc *crtc; + struct intel_encoder *encoder; + struct intel_connector *connector; + + list_for_each_entry(connector, &dev->mode_config.connector_list, + base.head) { + /* This also checks the encoder/connector hw state with the + * ->get_hw_state callbacks. */ + intel_connector_check_state(connector); + + WARN(&connector->new_encoder->base != connector->base.encoder, + "connector's staged encoder doesn't match current encoder\n"); + } + + list_for_each_entry(encoder, &dev->mode_config.encoder_list, + base.head) { + bool enabled = false; + bool active = false; + enum pipe pipe, tracked_pipe; + + DRM_DEBUG_KMS("[ENCODER:%d:%s]\n", + encoder->base.base.id, + drm_get_encoder_name(&encoder->base)); + + WARN(&encoder->new_crtc->base != encoder->base.crtc, + "encoder's stage crtc doesn't match current crtc\n"); + WARN(encoder->connectors_active && !encoder->base.crtc, + "encoder's active_connectors set, but no crtc\n"); + + list_for_each_entry(connector, &dev->mode_config.connector_list, + base.head) { + if (connector->base.encoder != &encoder->base) + continue; + enabled = true; + if (connector->base.dpms != DRM_MODE_DPMS_OFF) + active = true; + } + WARN(!!encoder->base.crtc != enabled, + "encoder's enabled state mismatch " + "(expected %i, found %i)\n", + !!encoder->base.crtc, enabled); + WARN(active && !encoder->base.crtc, + "active encoder with no crtc\n"); + + WARN(encoder->connectors_active != active, + "encoder's computed active state doesn't match tracked active state " + "(expected %i, found %i)\n", active, encoder->connectors_active); + + active = encoder->get_hw_state(encoder, &pipe); + WARN(active != encoder->connectors_active, + "encoder's hw state doesn't match sw tracking " + "(expected %i, found %i)\n", + encoder->connectors_active, active); + + if (!encoder->base.crtc) + continue; + + tracked_pipe = to_intel_crtc(encoder->base.crtc)->pipe; + WARN(active && pipe != tracked_pipe, + "active encoder's pipe doesn't match" + "(expected %i, found %i)\n", + tracked_pipe, pipe); + + } + + list_for_each_entry(crtc, &dev->mode_config.crtc_list, + base.head) { + bool enabled = false; + bool active = false; + + DRM_DEBUG_KMS("[CRTC:%d]\n", + crtc->base.base.id); + + WARN(crtc->active && !crtc->base.enabled, + "active crtc, but not enabled in sw tracking\n"); + + list_for_each_entry(encoder, &dev->mode_config.encoder_list, + base.head) { + if (encoder->base.crtc != &crtc->base) + continue; + enabled = true; + if (encoder->connectors_active) + active = true; + } + WARN(active != crtc->active, + "crtc's computed active state doesn't match tracked active state " + "(expected %i, found %i)\n", active, crtc->active); + WARN(enabled != crtc->base.enabled, + "crtc's computed enabled state doesn't match tracked enabled state " + "(expected %i, found %i)\n", enabled, crtc->base.enabled); + + assert_pipe(dev->dev_private, crtc->pipe, crtc->active); + } +} + +bool intel_set_mode(struct drm_crtc *crtc, + struct drm_display_mode *mode, + int x, int y, struct drm_framebuffer *fb) +{ + struct drm_device *dev = crtc->dev; + drm_i915_private_t *dev_priv = dev->dev_private; + struct drm_display_mode *adjusted_mode, saved_mode, saved_hwmode; + struct drm_encoder_helper_funcs *encoder_funcs; + struct drm_encoder *encoder; + struct intel_crtc *intel_crtc; + unsigned disable_pipes, prepare_pipes, modeset_pipes; + bool ret = true; + + intel_modeset_affected_pipes(crtc, &modeset_pipes, + &prepare_pipes, &disable_pipes); + + DRM_DEBUG_KMS("set mode pipe masks: modeset: %x, prepare: %x, disable: %x\n", + modeset_pipes, prepare_pipes, disable_pipes); + + for_each_intel_crtc_masked(dev, disable_pipes, intel_crtc) + intel_crtc_disable(&intel_crtc->base); + + saved_hwmode = crtc->hwmode; + saved_mode = crtc->mode; + + /* Hack: Because we don't (yet) support global modeset on multiple + * crtcs, we don't keep track of the new mode for more than one crtc. + * Hence simply check whether any bit is set in modeset_pipes in all the + * pieces of code that are not yet converted to deal with mutliple crtcs + * changing their mode at the same time. */ + adjusted_mode = NULL; + if (modeset_pipes) { + adjusted_mode = intel_modeset_adjusted_mode(crtc, mode); + if (IS_ERR(adjusted_mode)) { + return false; + } + } + + for_each_intel_crtc_masked(dev, prepare_pipes, intel_crtc) { + if (intel_crtc->base.enabled) + dev_priv->display.crtc_disable(&intel_crtc->base); + } + + /* crtc->mode is already used by the ->mode_set callbacks, hence we need + * to set it here already despite that we pass it down the callchain. */ - intel_crtc->dpms_mode = -1; + if (modeset_pipes) + crtc->mode = *mode; - /* We need to fix up any BIOS configuration that conflicts with - * our expectations. + /* Only after disabling all output pipelines that will be changed can we + * update the the output configuration. */ + intel_modeset_update_state(dev, prepare_pipes); + + /* Set up the DPLL and any encoders state that needs to adjust or depend + * on the DPLL. */ - intel_sanitize_modesetting(dev, intel_crtc->pipe, intel_crtc->plane); + for_each_intel_crtc_masked(dev, modeset_pipes, intel_crtc) { + ret = !intel_crtc_mode_set(&intel_crtc->base, + mode, adjusted_mode, + x, y, fb); + if (!ret) + goto done; + + list_for_each_entry(encoder, &dev->mode_config.encoder_list, head) { + + if (encoder->crtc != &intel_crtc->base) + continue; + + DRM_DEBUG_KMS("[ENCODER:%d:%s] set [MODE:%d:%s]\n", + encoder->base.id, drm_get_encoder_name(encoder), + mode->base.id, mode->name); + encoder_funcs = encoder->helper_private; + encoder_funcs->mode_set(encoder, mode, adjusted_mode); + } + } + + /* Now enable the clocks, plane, pipe, and connectors that we set up. */ + for_each_intel_crtc_masked(dev, prepare_pipes, intel_crtc) + dev_priv->display.crtc_enable(&intel_crtc->base); + + if (modeset_pipes) { + /* Store real post-adjustment hardware mode. */ + crtc->hwmode = *adjusted_mode; + + /* Calculate and store various constants which + * are later needed by vblank and swap-completion + * timestamping. They are derived from true hwmode. + */ + drm_calc_timestamping_constants(crtc); + } + + /* FIXME: add subpixel order */ +done: + drm_mode_destroy(dev, adjusted_mode); + if (!ret && crtc->enabled) { + crtc->hwmode = saved_hwmode; + crtc->mode = saved_mode; + } else { + intel_modeset_check_state(dev); + } + + return ret; } -static struct drm_crtc_helper_funcs intel_helper_funcs = { - .dpms = intel_crtc_dpms, - .mode_fixup = intel_crtc_mode_fixup, - .mode_set = intel_crtc_mode_set, - .mode_set_base = intel_pipe_set_base, - .mode_set_base_atomic = intel_pipe_set_base_atomic, - .load_lut = intel_crtc_load_lut, - .disable = intel_crtc_disable, -}; +#undef for_each_intel_crtc_masked + +static void intel_set_config_free(struct intel_set_config *config) +{ + if (!config) + return; + + kfree(config->save_connector_encoders); + kfree(config->save_encoder_crtcs); + kfree(config); +} + +static int intel_set_config_save_state(struct drm_device *dev, + struct intel_set_config *config) +{ + struct drm_encoder *encoder; + struct drm_connector *connector; + int count; + + config->save_encoder_crtcs = + kcalloc(dev->mode_config.num_encoder, + sizeof(struct drm_crtc *), GFP_KERNEL); + if (!config->save_encoder_crtcs) + return -ENOMEM; + + config->save_connector_encoders = + kcalloc(dev->mode_config.num_connector, + sizeof(struct drm_encoder *), GFP_KERNEL); + if (!config->save_connector_encoders) + return -ENOMEM; + + /* Copy data. Note that driver private data is not affected. + * Should anything bad happen only the expected state is + * restored, not the drivers personal bookkeeping. + */ + count = 0; + list_for_each_entry(encoder, &dev->mode_config.encoder_list, head) { + config->save_encoder_crtcs[count++] = encoder->crtc; + } + + count = 0; + list_for_each_entry(connector, &dev->mode_config.connector_list, head) { + config->save_connector_encoders[count++] = connector->encoder; + } + + return 0; +} + +static void intel_set_config_restore_state(struct drm_device *dev, + struct intel_set_config *config) +{ + struct intel_encoder *encoder; + struct intel_connector *connector; + int count; + + count = 0; + list_for_each_entry(encoder, &dev->mode_config.encoder_list, base.head) { + encoder->new_crtc = + to_intel_crtc(config->save_encoder_crtcs[count++]); + } + + count = 0; + list_for_each_entry(connector, &dev->mode_config.connector_list, base.head) { + connector->new_encoder = + to_intel_encoder(config->save_connector_encoders[count++]); + } +} + +static void +intel_set_config_compute_mode_changes(struct drm_mode_set *set, + struct intel_set_config *config) +{ + + /* We should be able to check here if the fb has the same properties + * and then just flip_or_move it */ + if (set->crtc->fb != set->fb) { + /* If we have no fb then treat it as a full mode set */ + if (set->crtc->fb == NULL) { + DRM_DEBUG_KMS("crtc has no fb, full mode set\n"); + config->mode_changed = true; + } else if (set->fb == NULL) { + config->mode_changed = true; + } else if (set->fb->depth != set->crtc->fb->depth) { + config->mode_changed = true; + } else if (set->fb->bits_per_pixel != + set->crtc->fb->bits_per_pixel) { + config->mode_changed = true; + } else + config->fb_changed = true; + } + + if (set->fb && (set->x != set->crtc->x || set->y != set->crtc->y)) + config->fb_changed = true; + + if (set->mode && !drm_mode_equal(set->mode, &set->crtc->mode)) { + DRM_DEBUG_KMS("modes are different, full mode set\n"); + drm_mode_debug_printmodeline(&set->crtc->mode); + drm_mode_debug_printmodeline(set->mode); + config->mode_changed = true; + } +} + +static int +intel_modeset_stage_output_state(struct drm_device *dev, + struct drm_mode_set *set, + struct intel_set_config *config) +{ + struct drm_crtc *new_crtc; + struct intel_connector *connector; + struct intel_encoder *encoder; + int count, ro; + + /* The upper layers ensure that we either disabl a crtc or have a list + * of connectors. For paranoia, double-check this. */ + WARN_ON(!set->fb && (set->num_connectors != 0)); + WARN_ON(set->fb && (set->num_connectors == 0)); + + count = 0; + list_for_each_entry(connector, &dev->mode_config.connector_list, + base.head) { + /* Otherwise traverse passed in connector list and get encoders + * for them. */ + for (ro = 0; ro < set->num_connectors; ro++) { + if (set->connectors[ro] == &connector->base) { + connector->new_encoder = connector->encoder; + break; + } + } + + /* If we disable the crtc, disable all its connectors. Also, if + * the connector is on the changing crtc but not on the new + * connector list, disable it. */ + if ((!set->fb || ro == set->num_connectors) && + connector->base.encoder && + connector->base.encoder->crtc == set->crtc) { + connector->new_encoder = NULL; + + DRM_DEBUG_KMS("[CONNECTOR:%d:%s] to [NOCRTC]\n", + connector->base.base.id, + drm_get_connector_name(&connector->base)); + } + + + if (&connector->new_encoder->base != connector->base.encoder) { + DRM_DEBUG_KMS("encoder changed, full mode switch\n"); + config->mode_changed = true; + } + + /* Disable all disconnected encoders. */ + if (connector->base.status == connector_status_disconnected) + connector->new_encoder = NULL; + } + /* connector->new_encoder is now updated for all connectors. */ + + /* Update crtc of enabled connectors. */ + count = 0; + list_for_each_entry(connector, &dev->mode_config.connector_list, + base.head) { + if (!connector->new_encoder) + continue; + + new_crtc = connector->new_encoder->base.crtc; + + for (ro = 0; ro < set->num_connectors; ro++) { + if (set->connectors[ro] == &connector->base) + new_crtc = set->crtc; + } + + /* Make sure the new CRTC will work with the encoder */ + if (!intel_encoder_crtc_ok(&connector->new_encoder->base, + new_crtc)) { + return -EINVAL; + } + connector->encoder->new_crtc = to_intel_crtc(new_crtc); + + DRM_DEBUG_KMS("[CONNECTOR:%d:%s] to [CRTC:%d]\n", + connector->base.base.id, + drm_get_connector_name(&connector->base), + new_crtc->base.id); + } + + /* Check for any encoders that needs to be disabled. */ + list_for_each_entry(encoder, &dev->mode_config.encoder_list, + base.head) { + list_for_each_entry(connector, + &dev->mode_config.connector_list, + base.head) { + if (connector->new_encoder == encoder) { + WARN_ON(!connector->new_encoder->new_crtc); + + goto next_encoder; + } + } + encoder->new_crtc = NULL; +next_encoder: + /* Only now check for crtc changes so we don't miss encoders + * that will be disabled. */ + if (&encoder->new_crtc->base != encoder->base.crtc) { + DRM_DEBUG_KMS("crtc changed, full mode switch\n"); + config->mode_changed = true; + } + } + /* Now we've also updated encoder->new_crtc for all encoders. */ + + return 0; +} + +static int intel_crtc_set_config(struct drm_mode_set *set) +{ + struct drm_device *dev; + struct drm_mode_set save_set; + struct intel_set_config *config; + int ret; + + BUG_ON(!set); + BUG_ON(!set->crtc); + BUG_ON(!set->crtc->helper_private); + + if (!set->mode) + set->fb = NULL; + + /* The fb helper likes to play gross jokes with ->mode_set_config. + * Unfortunately the crtc helper doesn't do much at all for this case, + * so we have to cope with this madness until the fb helper is fixed up. */ + if (set->fb && set->num_connectors == 0) + return 0; + + if (set->fb) { + DRM_DEBUG_KMS("[CRTC:%d] [FB:%d] #connectors=%d (x y) (%i %i)\n", + set->crtc->base.id, set->fb->base.id, + (int)set->num_connectors, set->x, set->y); + } else { + DRM_DEBUG_KMS("[CRTC:%d] [NOFB]\n", set->crtc->base.id); + } + + dev = set->crtc->dev; + + ret = -ENOMEM; + config = kzalloc(sizeof(*config), GFP_KERNEL); + if (!config) + goto out_config; + + ret = intel_set_config_save_state(dev, config); + if (ret) + goto out_config; + + save_set.crtc = set->crtc; + save_set.mode = &set->crtc->mode; + save_set.x = set->crtc->x; + save_set.y = set->crtc->y; + save_set.fb = set->crtc->fb; + + /* Compute whether we need a full modeset, only an fb base update or no + * change at all. In the future we might also check whether only the + * mode changed, e.g. for LVDS where we only change the panel fitter in + * such cases. */ + intel_set_config_compute_mode_changes(set, config); + + ret = intel_modeset_stage_output_state(dev, set, config); + if (ret) + goto fail; + + if (config->mode_changed) { + if (set->mode) { + DRM_DEBUG_KMS("attempting to set mode from" + " userspace\n"); + drm_mode_debug_printmodeline(set->mode); + } + + if (!intel_set_mode(set->crtc, set->mode, + set->x, set->y, set->fb)) { + DRM_ERROR("failed to set mode on [CRTC:%d]\n", + set->crtc->base.id); + ret = -EINVAL; + goto fail; + } + } else if (config->fb_changed) { + ret = intel_pipe_set_base(set->crtc, + set->x, set->y, set->fb); + } + + intel_set_config_free(config); + + return 0; + +fail: + intel_set_config_restore_state(dev, config); + + /* Try to restore the config */ + if (config->mode_changed && + !intel_set_mode(save_set.crtc, save_set.mode, + save_set.x, save_set.y, save_set.fb)) + DRM_ERROR("failed to restore config after modeset failure\n"); + +out_config: + intel_set_config_free(config); + return ret; +} static const struct drm_crtc_funcs intel_crtc_funcs = { - .reset = intel_crtc_reset, .cursor_set = intel_crtc_cursor_set, .cursor_move = intel_crtc_cursor_move, .gamma_set = intel_crtc_gamma_set, - .set_config = drm_crtc_helper_set_config, + .set_config = intel_crtc_set_config, .destroy = intel_crtc_destroy, .page_flip = intel_crtc_page_flip, }; @@ -6655,24 +7448,9 @@ static void intel_crtc_init(struct drm_device *dev, int pipe) dev_priv->plane_to_crtc_mapping[intel_crtc->plane] = &intel_crtc->base; dev_priv->pipe_to_crtc_mapping[intel_crtc->pipe] = &intel_crtc->base; - intel_crtc_reset(&intel_crtc->base); - intel_crtc->active = true; /* force the pipe off on setup_init_config */ intel_crtc->bpp = 24; /* default for pre-Ironlake */ - if (HAS_PCH_SPLIT(dev)) { - intel_helper_funcs.prepare = ironlake_crtc_prepare; - intel_helper_funcs.commit = ironlake_crtc_commit; - } else { - intel_helper_funcs.prepare = i9xx_crtc_prepare; - intel_helper_funcs.commit = i9xx_crtc_commit; - } - drm_crtc_helper_add(&intel_crtc->base, &intel_helper_funcs); - - intel_crtc->busy = false; - - setup_timer(&intel_crtc->idle_timer, intel_crtc_idle_timer, - (unsigned long)intel_crtc); } int intel_get_pipe_from_crtc_id(struct drm_device *dev, void *data, @@ -6699,15 +7477,23 @@ int intel_get_pipe_from_crtc_id(struct drm_device *dev, void *data, return 0; } -static int intel_encoder_clones(struct drm_device *dev, int type_mask) +static int intel_encoder_clones(struct intel_encoder *encoder) { - struct intel_encoder *encoder; + struct drm_device *dev = encoder->base.dev; + struct intel_encoder *source_encoder; int index_mask = 0; int entry = 0; - list_for_each_entry(encoder, &dev->mode_config.encoder_list, base.head) { - if (type_mask & encoder->clone_mask) + list_for_each_entry(source_encoder, + &dev->mode_config.encoder_list, base.head) { + + if (encoder == source_encoder) index_mask |= (1 << entry); + + /* Intel hw has only one MUX where enocoders could be cloned. */ + if (encoder->cloneable && source_encoder->cloneable) + index_mask |= (1 << entry); + entry++; } @@ -6748,10 +7534,10 @@ static void intel_setup_outputs(struct drm_device *dev) dpd_is_edp = intel_dpd_is_edp(dev); if (has_edp_a(dev)) - intel_dp_init(dev, DP_A); + intel_dp_init(dev, DP_A, PORT_A); if (dpd_is_edp && (I915_READ(PCH_DP_D) & DP_DETECTED)) - intel_dp_init(dev, PCH_DP_D); + intel_dp_init(dev, PCH_DP_D, PORT_D); } intel_crt_init(dev); @@ -6782,22 +7568,22 @@ static void intel_setup_outputs(struct drm_device *dev) /* PCH SDVOB multiplex with HDMIB */ found = intel_sdvo_init(dev, PCH_SDVOB, true); if (!found) - intel_hdmi_init(dev, HDMIB); + intel_hdmi_init(dev, HDMIB, PORT_B); if (!found && (I915_READ(PCH_DP_B) & DP_DETECTED)) - intel_dp_init(dev, PCH_DP_B); + intel_dp_init(dev, PCH_DP_B, PORT_B); } if (I915_READ(HDMIC) & PORT_DETECTED) - intel_hdmi_init(dev, HDMIC); + intel_hdmi_init(dev, HDMIC, PORT_C); if (!dpd_is_edp && I915_READ(HDMID) & PORT_DETECTED) - intel_hdmi_init(dev, HDMID); + intel_hdmi_init(dev, HDMID, PORT_D); if (I915_READ(PCH_DP_C) & DP_DETECTED) - intel_dp_init(dev, PCH_DP_C); + intel_dp_init(dev, PCH_DP_C, PORT_C); if (!dpd_is_edp && (I915_READ(PCH_DP_D) & DP_DETECTED)) - intel_dp_init(dev, PCH_DP_D); + intel_dp_init(dev, PCH_DP_D, PORT_D); } else if (IS_VALLEYVIEW(dev)) { int found; @@ -6805,17 +7591,17 @@ static void intel_setup_outputs(struct drm_device *dev) /* SDVOB multiplex with HDMIB */ found = intel_sdvo_init(dev, SDVOB, true); if (!found) - intel_hdmi_init(dev, SDVOB); + intel_hdmi_init(dev, SDVOB, PORT_B); if (!found && (I915_READ(DP_B) & DP_DETECTED)) - intel_dp_init(dev, DP_B); + intel_dp_init(dev, DP_B, PORT_B); } if (I915_READ(SDVOC) & PORT_DETECTED) - intel_hdmi_init(dev, SDVOC); + intel_hdmi_init(dev, SDVOC, PORT_C); /* Shares lanes with HDMI on SDVOC */ if (I915_READ(DP_C) & DP_DETECTED) - intel_dp_init(dev, DP_C); + intel_dp_init(dev, DP_C, PORT_C); } else if (SUPPORTS_DIGITAL_OUTPUTS(dev)) { bool found = false; @@ -6824,12 +7610,12 @@ static void intel_setup_outputs(struct drm_device *dev) found = intel_sdvo_init(dev, SDVOB, true); if (!found && SUPPORTS_INTEGRATED_HDMI(dev)) { DRM_DEBUG_KMS("probing HDMI on SDVOB\n"); - intel_hdmi_init(dev, SDVOB); + intel_hdmi_init(dev, SDVOB, PORT_B); } if (!found && SUPPORTS_INTEGRATED_DP(dev)) { DRM_DEBUG_KMS("probing DP_B\n"); - intel_dp_init(dev, DP_B); + intel_dp_init(dev, DP_B, PORT_B); } } @@ -6844,18 +7630,18 @@ static void intel_setup_outputs(struct drm_device *dev) if (SUPPORTS_INTEGRATED_HDMI(dev)) { DRM_DEBUG_KMS("probing HDMI on SDVOC\n"); - intel_hdmi_init(dev, SDVOC); + intel_hdmi_init(dev, SDVOC, PORT_C); } if (SUPPORTS_INTEGRATED_DP(dev)) { DRM_DEBUG_KMS("probing DP_C\n"); - intel_dp_init(dev, DP_C); + intel_dp_init(dev, DP_C, PORT_C); } } if (SUPPORTS_INTEGRATED_DP(dev) && (I915_READ(DP_D) & DP_DETECTED)) { DRM_DEBUG_KMS("probing DP_D\n"); - intel_dp_init(dev, DP_D); + intel_dp_init(dev, DP_D, PORT_D); } } else if (IS_GEN2(dev)) intel_dvo_init(dev); @@ -6866,12 +7652,9 @@ static void intel_setup_outputs(struct drm_device *dev) list_for_each_entry(encoder, &dev->mode_config.encoder_list, base.head) { encoder->base.possible_crtcs = encoder->crtc_mask; encoder->base.possible_clones = - intel_encoder_clones(dev, encoder->clone_mask); + intel_encoder_clones(encoder); } - /* disable all the possible outputs/crtcs before entering KMS mode */ - drm_helper_disable_unused_functions(dev); - if (HAS_PCH_IBX(dev) || HAS_PCH_CPT(dev)) ironlake_init_pch_refclk(dev); } @@ -6973,13 +7756,15 @@ static void intel_init_display(struct drm_device *dev) /* We always want a DPMS function */ if (HAS_PCH_SPLIT(dev)) { - dev_priv->display.dpms = ironlake_crtc_dpms; dev_priv->display.crtc_mode_set = ironlake_crtc_mode_set; + dev_priv->display.crtc_enable = ironlake_crtc_enable; + dev_priv->display.crtc_disable = ironlake_crtc_disable; dev_priv->display.off = ironlake_crtc_off; dev_priv->display.update_plane = ironlake_update_plane; } else { - dev_priv->display.dpms = i9xx_crtc_dpms; dev_priv->display.crtc_mode_set = i9xx_crtc_mode_set; + dev_priv->display.crtc_enable = i9xx_crtc_enable; + dev_priv->display.crtc_disable = i9xx_crtc_disable; dev_priv->display.off = i9xx_crtc_off; dev_priv->display.update_plane = i9xx_update_plane; } @@ -7023,7 +7808,7 @@ static void intel_init_display(struct drm_device *dev) dev_priv->display.write_eld = ironlake_write_eld; } else if (IS_HASWELL(dev)) { dev_priv->display.fdi_link_train = hsw_fdi_link_train; - dev_priv->display.write_eld = ironlake_write_eld; + dev_priv->display.write_eld = haswell_write_eld; } else dev_priv->display.update_wm = NULL; } else if (IS_G4X(dev)) { @@ -7101,21 +7886,16 @@ static struct intel_quirk intel_quirks[] = { /* HP Mini needs pipe A force quirk (LP: #322104) */ { 0x27ae, 0x103c, 0x361a, quirk_pipea_force }, - /* Thinkpad R31 needs pipe A force quirk */ - { 0x3577, 0x1014, 0x0505, quirk_pipea_force }, /* Toshiba Protege R-205, S-209 needs pipe A force quirk */ { 0x2592, 0x1179, 0x0001, quirk_pipea_force }, - /* ThinkPad X30 needs pipe A force quirk (LP: #304614) */ - { 0x3577, 0x1014, 0x0513, quirk_pipea_force }, - /* ThinkPad X40 needs pipe A force quirk */ - /* ThinkPad T60 needs pipe A force quirk (bug #16494) */ { 0x2782, 0x17aa, 0x201a, quirk_pipea_force }, /* 855 & before need to leave pipe A & dpll A up */ { 0x3582, PCI_ANY_ID, PCI_ANY_ID, quirk_pipea_force }, { 0x2562, PCI_ANY_ID, PCI_ANY_ID, quirk_pipea_force }, + { 0x3577, PCI_ANY_ID, PCI_ANY_ID, quirk_pipea_force }, /* Lenovo U160 cannot use SSC on LVDS */ { 0x0046, 0x17aa, 0x3920, quirk_ssc_force_disable }, @@ -7231,10 +8011,251 @@ void intel_modeset_init(struct drm_device *dev) /* Just disable it once at startup */ i915_disable_vga(dev); intel_setup_outputs(dev); +} - INIT_WORK(&dev_priv->idle_work, intel_idle_update); - setup_timer(&dev_priv->idle_timer, intel_gpu_idle_timer, - (unsigned long)dev); +static void +intel_connector_break_all_links(struct intel_connector *connector) +{ + connector->base.dpms = DRM_MODE_DPMS_OFF; + connector->base.encoder = NULL; + connector->encoder->connectors_active = false; + connector->encoder->base.crtc = NULL; +} + +static void intel_enable_pipe_a(struct drm_device *dev) +{ + struct intel_connector *connector; + struct drm_connector *crt = NULL; + struct intel_load_detect_pipe load_detect_temp; + + /* We can't just switch on the pipe A, we need to set things up with a + * proper mode and output configuration. As a gross hack, enable pipe A + * by enabling the load detect pipe once. */ + list_for_each_entry(connector, + &dev->mode_config.connector_list, + base.head) { + if (connector->encoder->type == INTEL_OUTPUT_ANALOG) { + crt = &connector->base; + break; + } + } + + if (!crt) + return; + + if (intel_get_load_detect_pipe(crt, NULL, &load_detect_temp)) + intel_release_load_detect_pipe(crt, &load_detect_temp); + + +} + +static void intel_sanitize_crtc(struct intel_crtc *crtc) +{ + struct drm_device *dev = crtc->base.dev; + struct drm_i915_private *dev_priv = dev->dev_private; + u32 reg, val; + + /* Clear any frame start delays used for debugging left by the BIOS */ + reg = PIPECONF(crtc->pipe); + I915_WRITE(reg, I915_READ(reg) & ~PIPECONF_FRAME_START_DELAY_MASK); + + /* We need to sanitize the plane -> pipe mapping first because this will + * disable the crtc (and hence change the state) if it is wrong. */ + if (!HAS_PCH_SPLIT(dev)) { + struct intel_connector *connector; + bool plane; + + reg = DSPCNTR(crtc->plane); + val = I915_READ(reg); + + if ((val & DISPLAY_PLANE_ENABLE) == 0 && + (!!(val & DISPPLANE_SEL_PIPE_MASK) == crtc->pipe)) + goto ok; + + DRM_DEBUG_KMS("[CRTC:%d] wrong plane connection detected!\n", + crtc->base.base.id); + + /* Pipe has the wrong plane attached and the plane is active. + * Temporarily change the plane mapping and disable everything + * ... */ + plane = crtc->plane; + crtc->plane = !plane; + dev_priv->display.crtc_disable(&crtc->base); + crtc->plane = plane; + + /* ... and break all links. */ + list_for_each_entry(connector, &dev->mode_config.connector_list, + base.head) { + if (connector->encoder->base.crtc != &crtc->base) + continue; + + intel_connector_break_all_links(connector); + } + + WARN_ON(crtc->active); + crtc->base.enabled = false; + } +ok: + + if (dev_priv->quirks & QUIRK_PIPEA_FORCE && + crtc->pipe == PIPE_A && !crtc->active) { + /* BIOS forgot to enable pipe A, this mostly happens after + * resume. Force-enable the pipe to fix this, the update_dpms + * call below we restore the pipe to the right state, but leave + * the required bits on. */ + intel_enable_pipe_a(dev); + } + + /* Adjust the state of the output pipe according to whether we + * have active connectors/encoders. */ + intel_crtc_update_dpms(&crtc->base); + + if (crtc->active != crtc->base.enabled) { + struct intel_encoder *encoder; + + /* This can happen either due to bugs in the get_hw_state + * functions or because the pipe is force-enabled due to the + * pipe A quirk. */ + DRM_DEBUG_KMS("[CRTC:%d] hw state adjusted, was %s, now %s\n", + crtc->base.base.id, + crtc->base.enabled ? "enabled" : "disabled", + crtc->active ? "enabled" : "disabled"); + + crtc->base.enabled = crtc->active; + + /* Because we only establish the connector -> encoder -> + * crtc links if something is active, this means the + * crtc is now deactivated. Break the links. connector + * -> encoder links are only establish when things are + * actually up, hence no need to break them. */ + WARN_ON(crtc->active); + + for_each_encoder_on_crtc(dev, &crtc->base, encoder) { + WARN_ON(encoder->connectors_active); + encoder->base.crtc = NULL; + } + } +} + +static void intel_sanitize_encoder(struct intel_encoder *encoder) +{ + struct intel_connector *connector; + struct drm_device *dev = encoder->base.dev; + + /* We need to check both for a crtc link (meaning that the + * encoder is active and trying to read from a pipe) and the + * pipe itself being active. */ + bool has_active_crtc = encoder->base.crtc && + to_intel_crtc(encoder->base.crtc)->active; + + if (encoder->connectors_active && !has_active_crtc) { + DRM_DEBUG_KMS("[ENCODER:%d:%s] has active connectors but no active pipe!\n", + encoder->base.base.id, + drm_get_encoder_name(&encoder->base)); + + /* Connector is active, but has no active pipe. This is + * fallout from our resume register restoring. Disable + * the encoder manually again. */ + if (encoder->base.crtc) { + DRM_DEBUG_KMS("[ENCODER:%d:%s] manually disabled\n", + encoder->base.base.id, + drm_get_encoder_name(&encoder->base)); + encoder->disable(encoder); + } + + /* Inconsistent output/port/pipe state happens presumably due to + * a bug in one of the get_hw_state functions. Or someplace else + * in our code, like the register restore mess on resume. Clamp + * things to off as a safer default. */ + list_for_each_entry(connector, + &dev->mode_config.connector_list, + base.head) { + if (connector->encoder != encoder) + continue; + + intel_connector_break_all_links(connector); + } + } + /* Enabled encoders without active connectors will be fixed in + * the crtc fixup. */ +} + +/* Scan out the current hw modeset state, sanitizes it and maps it into the drm + * and i915 state tracking structures. */ +void intel_modeset_setup_hw_state(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + enum pipe pipe; + u32 tmp; + struct intel_crtc *crtc; + struct intel_encoder *encoder; + struct intel_connector *connector; + + for_each_pipe(pipe) { + crtc = to_intel_crtc(dev_priv->pipe_to_crtc_mapping[pipe]); + + tmp = I915_READ(PIPECONF(pipe)); + if (tmp & PIPECONF_ENABLE) + crtc->active = true; + else + crtc->active = false; + + crtc->base.enabled = crtc->active; + + DRM_DEBUG_KMS("[CRTC:%d] hw state readout: %s\n", + crtc->base.base.id, + crtc->active ? "enabled" : "disabled"); + } + + list_for_each_entry(encoder, &dev->mode_config.encoder_list, + base.head) { + pipe = 0; + + if (encoder->get_hw_state(encoder, &pipe)) { + encoder->base.crtc = + dev_priv->pipe_to_crtc_mapping[pipe]; + } else { + encoder->base.crtc = NULL; + } + + encoder->connectors_active = false; + DRM_DEBUG_KMS("[ENCODER:%d:%s] hw state readout: %s, pipe=%i\n", + encoder->base.base.id, + drm_get_encoder_name(&encoder->base), + encoder->base.crtc ? "enabled" : "disabled", + pipe); + } + + list_for_each_entry(connector, &dev->mode_config.connector_list, + base.head) { + if (connector->get_hw_state(connector)) { + connector->base.dpms = DRM_MODE_DPMS_ON; + connector->encoder->connectors_active = true; + connector->base.encoder = &connector->encoder->base; + } else { + connector->base.dpms = DRM_MODE_DPMS_OFF; + connector->base.encoder = NULL; + } + DRM_DEBUG_KMS("[CONNECTOR:%d:%s] hw state readout: %s\n", + connector->base.base.id, + drm_get_connector_name(&connector->base), + connector->base.encoder ? "enabled" : "disabled"); + } + + /* HW state is read out, now we need to sanitize this mess. */ + list_for_each_entry(encoder, &dev->mode_config.encoder_list, + base.head) { + intel_sanitize_encoder(encoder); + } + + for_each_pipe(pipe) { + crtc = to_intel_crtc(dev_priv->pipe_to_crtc_mapping[pipe]); + intel_sanitize_crtc(crtc); + } + + intel_modeset_update_staged_output_state(dev); + + intel_modeset_check_state(dev); } void intel_modeset_gem_init(struct drm_device *dev) @@ -7242,6 +8263,8 @@ void intel_modeset_gem_init(struct drm_device *dev) intel_modeset_init_hw(dev); intel_setup_overlay(dev); + + intel_modeset_setup_hw_state(dev); } void intel_modeset_cleanup(struct drm_device *dev) @@ -7280,19 +8303,11 @@ void intel_modeset_cleanup(struct drm_device *dev) * enqueue unpin/hotplug work. */ drm_irq_uninstall(dev); cancel_work_sync(&dev_priv->hotplug_work); - cancel_work_sync(&dev_priv->rps_work); + cancel_work_sync(&dev_priv->rps.work); /* flush any delayed tasks or pending work */ flush_scheduled_work(); - /* Shut off idle work before the crtcs get freed. */ - list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) { - intel_crtc = to_intel_crtc(crtc); - del_timer_sync(&intel_crtc->idle_timer); - } - del_timer_sync(&dev_priv->idle_timer); - cancel_work_sync(&dev_priv->idle_work); - drm_mode_config_cleanup(dev); } @@ -7338,7 +8353,7 @@ struct intel_display_error_state { u32 position; u32 base; u32 size; - } cursor[2]; + } cursor[I915_MAX_PIPES]; struct intel_pipe_error_state { u32 conf; @@ -7350,7 +8365,7 @@ struct intel_display_error_state { u32 vtotal; u32 vblank; u32 vsync; - } pipe[2]; + } pipe[I915_MAX_PIPES]; struct intel_plane_error_state { u32 control; @@ -7360,7 +8375,7 @@ struct intel_display_error_state { u32 addr; u32 surface; u32 tile_offset; - } plane[2]; + } plane[I915_MAX_PIPES]; }; struct intel_display_error_state * @@ -7374,7 +8389,7 @@ intel_display_capture_error_state(struct drm_device *dev) if (error == NULL) return NULL; - for (i = 0; i < 2; i++) { + for_each_pipe(i) { error->cursor[i].control = I915_READ(CURCNTR(i)); error->cursor[i].position = I915_READ(CURPOS(i)); error->cursor[i].base = I915_READ(CURBASE(i)); @@ -7407,9 +8422,11 @@ intel_display_print_error_state(struct seq_file *m, struct drm_device *dev, struct intel_display_error_state *error) { + drm_i915_private_t *dev_priv = dev->dev_private; int i; - for (i = 0; i < 2; i++) { + seq_printf(m, "Num Pipes: %d\n", dev_priv->num_pipe); + for_each_pipe(i) { seq_printf(m, "Pipe [%d]:\n", i); seq_printf(m, " CONF: %08x\n", error->pipe[i].conf); seq_printf(m, " SRC: %08x\n", error->pipe[i].source); diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c index ace757af9133..1474f84fdbd0 100644 --- a/drivers/gpu/drm/i915/intel_dp.c +++ b/drivers/gpu/drm/i915/intel_dp.c @@ -36,42 +36,10 @@ #include "intel_drv.h" #include "i915_drm.h" #include "i915_drv.h" -#include "drm_dp_helper.h" -#define DP_RECEIVER_CAP_SIZE 0xf #define DP_LINK_STATUS_SIZE 6 #define DP_LINK_CHECK_TIMEOUT (10 * 1000) -#define DP_LINK_CONFIGURATION_SIZE 9 - -struct intel_dp { - struct intel_encoder base; - uint32_t output_reg; - uint32_t DP; - uint8_t link_configuration[DP_LINK_CONFIGURATION_SIZE]; - bool has_audio; - enum hdmi_force_audio force_audio; - uint32_t color_range; - int dpms_mode; - uint8_t link_bw; - uint8_t lane_count; - uint8_t dpcd[DP_RECEIVER_CAP_SIZE]; - struct i2c_adapter adapter; - struct i2c_algo_dp_aux_data algo; - bool is_pch_edp; - uint8_t train_set[4]; - int panel_power_up_delay; - int panel_power_down_delay; - int panel_power_cycle_delay; - int backlight_on_delay; - int backlight_off_delay; - struct drm_display_mode *panel_fixed_mode; /* for eDP */ - struct delayed_work panel_vdd_work; - bool want_panel_vdd; - struct edid *edid; /* cached EDID for eDP */ - int edid_mode_count; -}; - /** * is_edp - is the given port attached to an eDP panel (either CPU or PCH) * @intel_dp: DP struct @@ -840,9 +808,6 @@ intel_dp_set_m_n(struct drm_crtc *crtc, struct drm_display_mode *mode, } } -static void ironlake_edp_pll_on(struct drm_encoder *encoder); -static void ironlake_edp_pll_off(struct drm_encoder *encoder); - static void intel_dp_mode_set(struct drm_encoder *encoder, struct drm_display_mode *mode, struct drm_display_mode *adjusted_mode) @@ -853,14 +818,6 @@ intel_dp_mode_set(struct drm_encoder *encoder, struct drm_display_mode *mode, struct drm_crtc *crtc = intel_dp->base.base.crtc; struct intel_crtc *intel_crtc = to_intel_crtc(crtc); - /* Turn on the eDP PLL if needed */ - if (is_edp(intel_dp)) { - if (!is_pch_edp(intel_dp)) - ironlake_edp_pll_on(encoder); - else - ironlake_edp_pll_off(encoder); - } - /* * There are four kinds of DP registers: * @@ -882,10 +839,8 @@ intel_dp_mode_set(struct drm_encoder *encoder, struct drm_display_mode *mode, * supposed to be read-only. */ intel_dp->DP = I915_READ(intel_dp->output_reg) & DP_DETECTED; - intel_dp->DP |= DP_VOLTAGE_0_4 | DP_PRE_EMPHASIS_0; /* Handle DP bits in common between all three register formats */ - intel_dp->DP |= DP_VOLTAGE_0_4 | DP_PRE_EMPHASIS_0; switch (intel_dp->lane_count) { @@ -932,7 +887,6 @@ intel_dp_mode_set(struct drm_encoder *encoder, struct drm_display_mode *mode, intel_dp->DP |= intel_crtc->pipe << 29; /* don't miss out required setting for eDP */ - intel_dp->DP |= DP_PLL_ENABLE; if (adjusted_mode->clock < 200000) intel_dp->DP |= DP_PLL_FREQ_160MHZ; else @@ -954,7 +908,6 @@ intel_dp_mode_set(struct drm_encoder *encoder, struct drm_display_mode *mode, if (is_cpu_edp(intel_dp)) { /* don't miss out required setting for eDP */ - intel_dp->DP |= DP_PLL_ENABLE; if (adjusted_mode->clock < 200000) intel_dp->DP |= DP_PLL_FREQ_160MHZ; else @@ -1225,27 +1178,49 @@ static void ironlake_edp_backlight_off(struct intel_dp *intel_dp) msleep(intel_dp->backlight_off_delay); } -static void ironlake_edp_pll_on(struct drm_encoder *encoder) +static void ironlake_edp_pll_on(struct intel_dp *intel_dp) { - struct drm_device *dev = encoder->dev; + struct drm_device *dev = intel_dp->base.base.dev; + struct drm_crtc *crtc = intel_dp->base.base.crtc; struct drm_i915_private *dev_priv = dev->dev_private; u32 dpa_ctl; + assert_pipe_disabled(dev_priv, + to_intel_crtc(crtc)->pipe); + DRM_DEBUG_KMS("\n"); dpa_ctl = I915_READ(DP_A); - dpa_ctl |= DP_PLL_ENABLE; - I915_WRITE(DP_A, dpa_ctl); + WARN(dpa_ctl & DP_PLL_ENABLE, "dp pll on, should be off\n"); + WARN(dpa_ctl & DP_PORT_EN, "dp port still on, should be off\n"); + + /* We don't adjust intel_dp->DP while tearing down the link, to + * facilitate link retraining (e.g. after hotplug). Hence clear all + * enable bits here to ensure that we don't enable too much. */ + intel_dp->DP &= ~(DP_PORT_EN | DP_AUDIO_OUTPUT_ENABLE); + intel_dp->DP |= DP_PLL_ENABLE; + I915_WRITE(DP_A, intel_dp->DP); POSTING_READ(DP_A); udelay(200); } -static void ironlake_edp_pll_off(struct drm_encoder *encoder) +static void ironlake_edp_pll_off(struct intel_dp *intel_dp) { - struct drm_device *dev = encoder->dev; + struct drm_device *dev = intel_dp->base.base.dev; + struct drm_crtc *crtc = intel_dp->base.base.crtc; struct drm_i915_private *dev_priv = dev->dev_private; u32 dpa_ctl; + assert_pipe_disabled(dev_priv, + to_intel_crtc(crtc)->pipe); + dpa_ctl = I915_READ(DP_A); + WARN((dpa_ctl & DP_PLL_ENABLE) == 0, + "dp pll off, should be on\n"); + WARN(dpa_ctl & DP_PORT_EN, "dp port still on, should be off\n"); + + /* We can't rely on the value tracked for the DP register in + * intel_dp->DP because link_down must not change that (otherwise link + * re-training will fail. */ dpa_ctl &= ~DP_PLL_ENABLE; I915_WRITE(DP_A, dpa_ctl); POSTING_READ(DP_A); @@ -1282,10 +1257,57 @@ static void intel_dp_sink_dpms(struct intel_dp *intel_dp, int mode) } } -static void intel_dp_prepare(struct drm_encoder *encoder) +static bool intel_dp_get_hw_state(struct intel_encoder *encoder, + enum pipe *pipe) { - struct intel_dp *intel_dp = enc_to_intel_dp(encoder); + struct intel_dp *intel_dp = enc_to_intel_dp(&encoder->base); + struct drm_device *dev = encoder->base.dev; + struct drm_i915_private *dev_priv = dev->dev_private; + u32 tmp = I915_READ(intel_dp->output_reg); + if (!(tmp & DP_PORT_EN)) + return false; + + if (is_cpu_edp(intel_dp) && IS_GEN7(dev)) { + *pipe = PORT_TO_PIPE_CPT(tmp); + } else if (!HAS_PCH_CPT(dev) || is_cpu_edp(intel_dp)) { + *pipe = PORT_TO_PIPE(tmp); + } else { + u32 trans_sel; + u32 trans_dp; + int i; + + switch (intel_dp->output_reg) { + case PCH_DP_B: + trans_sel = TRANS_DP_PORT_SEL_B; + break; + case PCH_DP_C: + trans_sel = TRANS_DP_PORT_SEL_C; + break; + case PCH_DP_D: + trans_sel = TRANS_DP_PORT_SEL_D; + break; + default: + return true; + } + + for_each_pipe(i) { + trans_dp = I915_READ(TRANS_DP_CTL(i)); + if ((trans_dp & TRANS_DP_PORT_SEL_MASK) == trans_sel) { + *pipe = i; + return true; + } + } + } + + DRM_DEBUG_KMS("No pipe for dp port 0x%x found\n", intel_dp->output_reg); + + return true; +} + +static void intel_disable_dp(struct intel_encoder *encoder) +{ + struct intel_dp *intel_dp = enc_to_intel_dp(&encoder->base); /* Make sure the panel is off before trying to change the mode. But also * ensure that we have vdd while we switch off the panel. */ @@ -1293,14 +1315,31 @@ static void intel_dp_prepare(struct drm_encoder *encoder) ironlake_edp_backlight_off(intel_dp); intel_dp_sink_dpms(intel_dp, DRM_MODE_DPMS_ON); ironlake_edp_panel_off(intel_dp); - intel_dp_link_down(intel_dp); + + /* cpu edp my only be disable _after_ the cpu pipe/plane is disabled. */ + if (!is_cpu_edp(intel_dp)) + intel_dp_link_down(intel_dp); } -static void intel_dp_commit(struct drm_encoder *encoder) +static void intel_post_disable_dp(struct intel_encoder *encoder) { - struct intel_dp *intel_dp = enc_to_intel_dp(encoder); - struct drm_device *dev = encoder->dev; - struct intel_crtc *intel_crtc = to_intel_crtc(intel_dp->base.base.crtc); + struct intel_dp *intel_dp = enc_to_intel_dp(&encoder->base); + + if (is_cpu_edp(intel_dp)) { + intel_dp_link_down(intel_dp); + ironlake_edp_pll_off(intel_dp); + } +} + +static void intel_enable_dp(struct intel_encoder *encoder) +{ + struct intel_dp *intel_dp = enc_to_intel_dp(&encoder->base); + struct drm_device *dev = encoder->base.dev; + struct drm_i915_private *dev_priv = dev->dev_private; + uint32_t dp_reg = I915_READ(intel_dp->output_reg); + + if (WARN_ON(dp_reg & DP_PORT_EN)) + return; ironlake_edp_panel_vdd_on(intel_dp); intel_dp_sink_dpms(intel_dp, DRM_MODE_DPMS_ON); @@ -1309,47 +1348,14 @@ static void intel_dp_commit(struct drm_encoder *encoder) ironlake_edp_panel_vdd_off(intel_dp, true); intel_dp_complete_link_train(intel_dp); ironlake_edp_backlight_on(intel_dp); - - intel_dp->dpms_mode = DRM_MODE_DPMS_ON; - - if (HAS_PCH_CPT(dev)) - intel_cpt_verify_modeset(dev, intel_crtc->pipe); } -static void -intel_dp_dpms(struct drm_encoder *encoder, int mode) +static void intel_pre_enable_dp(struct intel_encoder *encoder) { - struct intel_dp *intel_dp = enc_to_intel_dp(encoder); - struct drm_device *dev = encoder->dev; - struct drm_i915_private *dev_priv = dev->dev_private; - uint32_t dp_reg = I915_READ(intel_dp->output_reg); + struct intel_dp *intel_dp = enc_to_intel_dp(&encoder->base); - if (mode != DRM_MODE_DPMS_ON) { - /* Switching the panel off requires vdd. */ - ironlake_edp_panel_vdd_on(intel_dp); - ironlake_edp_backlight_off(intel_dp); - intel_dp_sink_dpms(intel_dp, mode); - ironlake_edp_panel_off(intel_dp); - intel_dp_link_down(intel_dp); - - if (is_cpu_edp(intel_dp)) - ironlake_edp_pll_off(encoder); - } else { - if (is_cpu_edp(intel_dp)) - ironlake_edp_pll_on(encoder); - - ironlake_edp_panel_vdd_on(intel_dp); - intel_dp_sink_dpms(intel_dp, mode); - if (!(dp_reg & DP_PORT_EN)) { - intel_dp_start_link_train(intel_dp); - ironlake_edp_panel_on(intel_dp); - ironlake_edp_panel_vdd_off(intel_dp, true); - intel_dp_complete_link_train(intel_dp); - } else - ironlake_edp_panel_vdd_off(intel_dp, false); - ironlake_edp_backlight_on(intel_dp); - } - intel_dp->dpms_mode = mode; + if (is_cpu_edp(intel_dp)) + ironlake_edp_pll_on(intel_dp); } /* @@ -1668,6 +1674,45 @@ intel_dp_set_link_train(struct intel_dp *intel_dp, struct drm_i915_private *dev_priv = dev->dev_private; int ret; + if (HAS_PCH_CPT(dev) && (IS_GEN7(dev) || !is_cpu_edp(intel_dp))) { + dp_reg_value &= ~DP_LINK_TRAIN_MASK_CPT; + + switch (dp_train_pat & DP_TRAINING_PATTERN_MASK) { + case DP_TRAINING_PATTERN_DISABLE: + dp_reg_value |= DP_LINK_TRAIN_OFF_CPT; + break; + case DP_TRAINING_PATTERN_1: + dp_reg_value |= DP_LINK_TRAIN_PAT_1_CPT; + break; + case DP_TRAINING_PATTERN_2: + dp_reg_value |= DP_LINK_TRAIN_PAT_2_CPT; + break; + case DP_TRAINING_PATTERN_3: + DRM_ERROR("DP training pattern 3 not supported\n"); + dp_reg_value |= DP_LINK_TRAIN_PAT_2_CPT; + break; + } + + } else { + dp_reg_value &= ~DP_LINK_TRAIN_MASK; + + switch (dp_train_pat & DP_TRAINING_PATTERN_MASK) { + case DP_TRAINING_PATTERN_DISABLE: + dp_reg_value |= DP_LINK_TRAIN_OFF; + break; + case DP_TRAINING_PATTERN_1: + dp_reg_value |= DP_LINK_TRAIN_PAT_1; + break; + case DP_TRAINING_PATTERN_2: + dp_reg_value |= DP_LINK_TRAIN_PAT_2; + break; + case DP_TRAINING_PATTERN_3: + DRM_ERROR("DP training pattern 3 not supported\n"); + dp_reg_value |= DP_LINK_TRAIN_PAT_2; + break; + } + } + I915_WRITE(intel_dp->output_reg, dp_reg_value); POSTING_READ(intel_dp->output_reg); @@ -1675,12 +1720,15 @@ intel_dp_set_link_train(struct intel_dp *intel_dp, DP_TRAINING_PATTERN_SET, dp_train_pat); - ret = intel_dp_aux_native_write(intel_dp, - DP_TRAINING_LANE0_SET, - intel_dp->train_set, - intel_dp->lane_count); - if (ret != intel_dp->lane_count) - return false; + if ((dp_train_pat & DP_TRAINING_PATTERN_MASK) != + DP_TRAINING_PATTERN_DISABLE) { + ret = intel_dp_aux_native_write(intel_dp, + DP_TRAINING_LANE0_SET, + intel_dp->train_set, + intel_dp->lane_count); + if (ret != intel_dp->lane_count) + return false; + } return true; } @@ -1690,26 +1738,12 @@ static void intel_dp_start_link_train(struct intel_dp *intel_dp) { struct drm_device *dev = intel_dp->base.base.dev; - struct drm_i915_private *dev_priv = dev->dev_private; - struct intel_crtc *intel_crtc = to_intel_crtc(intel_dp->base.base.crtc); int i; uint8_t voltage; bool clock_recovery = false; int voltage_tries, loop_tries; - u32 reg; uint32_t DP = intel_dp->DP; - /* - * On CPT we have to enable the port in training pattern 1, which - * will happen below in intel_dp_set_link_train. Otherwise, enable - * the port and wait for it to become active. - */ - if (!HAS_PCH_CPT(dev)) { - I915_WRITE(intel_dp->output_reg, intel_dp->DP); - POSTING_READ(intel_dp->output_reg); - intel_wait_for_vblank(dev, intel_crtc->pipe); - } - /* Write the link configuration data */ intel_dp_aux_native_write(intel_dp, DP_LINK_BW_SET, intel_dp->link_configuration, @@ -1717,10 +1751,6 @@ intel_dp_start_link_train(struct intel_dp *intel_dp) DP |= DP_PORT_EN; - if (HAS_PCH_CPT(dev) && (IS_GEN7(dev) || !is_cpu_edp(intel_dp))) - DP &= ~DP_LINK_TRAIN_MASK_CPT; - else - DP &= ~DP_LINK_TRAIN_MASK; memset(intel_dp->train_set, 0, 4); voltage = 0xff; voltage_tries = 0; @@ -1744,12 +1774,7 @@ intel_dp_start_link_train(struct intel_dp *intel_dp) DP = (DP & ~(DP_VOLTAGE_MASK|DP_PRE_EMPHASIS_MASK)) | signal_levels; } - if (HAS_PCH_CPT(dev) && (IS_GEN7(dev) || !is_cpu_edp(intel_dp))) - reg = DP | DP_LINK_TRAIN_PAT_1_CPT; - else - reg = DP | DP_LINK_TRAIN_PAT_1; - - if (!intel_dp_set_link_train(intel_dp, reg, + if (!intel_dp_set_link_train(intel_dp, DP, DP_TRAINING_PATTERN_1 | DP_LINK_SCRAMBLING_DISABLE)) break; @@ -1804,10 +1829,8 @@ static void intel_dp_complete_link_train(struct intel_dp *intel_dp) { struct drm_device *dev = intel_dp->base.base.dev; - struct drm_i915_private *dev_priv = dev->dev_private; bool channel_eq = false; int tries, cr_tries; - u32 reg; uint32_t DP = intel_dp->DP; /* channel equalization */ @@ -1836,13 +1859,8 @@ intel_dp_complete_link_train(struct intel_dp *intel_dp) DP = (DP & ~(DP_VOLTAGE_MASK|DP_PRE_EMPHASIS_MASK)) | signal_levels; } - if (HAS_PCH_CPT(dev) && (IS_GEN7(dev) || !is_cpu_edp(intel_dp))) - reg = DP | DP_LINK_TRAIN_PAT_2_CPT; - else - reg = DP | DP_LINK_TRAIN_PAT_2; - /* channel eq pattern */ - if (!intel_dp_set_link_train(intel_dp, reg, + if (!intel_dp_set_link_train(intel_dp, DP, DP_TRAINING_PATTERN_2 | DP_LINK_SCRAMBLING_DISABLE)) break; @@ -1877,15 +1895,7 @@ intel_dp_complete_link_train(struct intel_dp *intel_dp) ++tries; } - if (HAS_PCH_CPT(dev) && (IS_GEN7(dev) || !is_cpu_edp(intel_dp))) - reg = DP | DP_LINK_TRAIN_OFF_CPT; - else - reg = DP | DP_LINK_TRAIN_OFF; - - I915_WRITE(intel_dp->output_reg, reg); - POSTING_READ(intel_dp->output_reg); - intel_dp_aux_native_write_1(intel_dp, - DP_TRAINING_PATTERN_SET, DP_TRAINING_PATTERN_DISABLE); + intel_dp_set_link_train(intel_dp, DP, DP_TRAINING_PATTERN_DISABLE); } static void @@ -1895,18 +1905,11 @@ intel_dp_link_down(struct intel_dp *intel_dp) struct drm_i915_private *dev_priv = dev->dev_private; uint32_t DP = intel_dp->DP; - if ((I915_READ(intel_dp->output_reg) & DP_PORT_EN) == 0) + if (WARN_ON((I915_READ(intel_dp->output_reg) & DP_PORT_EN) == 0)) return; DRM_DEBUG_KMS("\n"); - if (is_edp(intel_dp)) { - DP &= ~DP_PLL_ENABLE; - I915_WRITE(intel_dp->output_reg, DP); - POSTING_READ(intel_dp->output_reg); - udelay(100); - } - if (HAS_PCH_CPT(dev) && (IS_GEN7(dev) || !is_cpu_edp(intel_dp))) { DP &= ~DP_LINK_TRAIN_MASK_CPT; I915_WRITE(intel_dp->output_reg, DP | DP_LINK_TRAIN_PAT_IDLE_CPT); @@ -1918,13 +1921,6 @@ intel_dp_link_down(struct intel_dp *intel_dp) msleep(17); - if (is_edp(intel_dp)) { - if (HAS_PCH_CPT(dev) && (IS_GEN7(dev) || !is_cpu_edp(intel_dp))) - DP |= DP_LINK_TRAIN_OFF_CPT; - else - DP |= DP_LINK_TRAIN_OFF; - } - if (HAS_PCH_IBX(dev) && I915_READ(intel_dp->output_reg) & DP_PIPEB_SELECT) { struct drm_crtc *crtc = intel_dp->base.base.crtc; @@ -2033,10 +2029,10 @@ intel_dp_check_link_status(struct intel_dp *intel_dp) u8 sink_irq_vector; u8 link_status[DP_LINK_STATUS_SIZE]; - if (intel_dp->dpms_mode != DRM_MODE_DPMS_ON) + if (!intel_dp->base.connectors_active) return; - if (!intel_dp->base.base.crtc) + if (WARN_ON(!intel_dp->base.base.crtc)) return; /* Try to read receiver status if the link appears to be up */ @@ -2160,7 +2156,6 @@ intel_dp_get_edid_modes(struct drm_connector *connector, struct i2c_adapter *ada ret = drm_add_edid_modes(connector, intel_dp->edid); drm_edid_to_eld(connector, intel_dp->edid); - connector->display_info.raw_edid = NULL; return intel_dp->edid_mode_count; } @@ -2206,7 +2201,6 @@ intel_dp_detect(struct drm_connector *connector, bool force) edid = intel_dp_get_edid(connector, &intel_dp->adapter); if (edid) { intel_dp->has_audio = drm_detect_monitor_audio(edid); - connector->display_info.raw_edid = NULL; kfree(edid); } } @@ -2271,8 +2265,6 @@ intel_dp_detect_audio(struct drm_connector *connector) edid = intel_dp_get_edid(connector, &intel_dp->adapter); if (edid) { has_audio = drm_detect_monitor_audio(edid); - - connector->display_info.raw_edid = NULL; kfree(edid); } @@ -2326,9 +2318,8 @@ intel_dp_set_property(struct drm_connector *connector, done: if (intel_dp->base.base.crtc) { struct drm_crtc *crtc = intel_dp->base.base.crtc; - drm_crtc_helper_set_mode(crtc, &crtc->mode, - crtc->x, crtc->y, - crtc->fb); + intel_set_mode(crtc, &crtc->mode, + crtc->x, crtc->y, crtc->fb); } return 0; @@ -2362,15 +2353,13 @@ static void intel_dp_encoder_destroy(struct drm_encoder *encoder) } static const struct drm_encoder_helper_funcs intel_dp_helper_funcs = { - .dpms = intel_dp_dpms, .mode_fixup = intel_dp_mode_fixup, - .prepare = intel_dp_prepare, .mode_set = intel_dp_mode_set, - .commit = intel_dp_commit, + .disable = intel_encoder_noop, }; static const struct drm_connector_funcs intel_dp_connector_funcs = { - .dpms = drm_helper_connector_dpms, + .dpms = intel_connector_dpms, .detect = intel_dp_detect, .fill_modes = drm_helper_probe_single_connector_modes, .set_property = intel_dp_set_property, @@ -2441,7 +2430,7 @@ intel_dp_add_properties(struct intel_dp *intel_dp, struct drm_connector *connect } void -intel_dp_init(struct drm_device *dev, int output_reg) +intel_dp_init(struct drm_device *dev, int output_reg, enum port port) { struct drm_i915_private *dev_priv = dev->dev_private; struct drm_connector *connector; @@ -2456,7 +2445,9 @@ intel_dp_init(struct drm_device *dev, int output_reg) return; intel_dp->output_reg = output_reg; - intel_dp->dpms_mode = -1; + intel_dp->port = port; + /* Preserve the current hw state. */ + intel_dp->DP = I915_READ(intel_dp->output_reg); intel_connector = kzalloc(sizeof(struct intel_connector), GFP_KERNEL); if (!intel_connector) { @@ -2483,18 +2474,10 @@ intel_dp_init(struct drm_device *dev, int output_reg) connector->polled = DRM_CONNECTOR_POLL_HPD; - if (output_reg == DP_B || output_reg == PCH_DP_B) - intel_encoder->clone_mask = (1 << INTEL_DP_B_CLONE_BIT); - else if (output_reg == DP_C || output_reg == PCH_DP_C) - intel_encoder->clone_mask = (1 << INTEL_DP_C_CLONE_BIT); - else if (output_reg == DP_D || output_reg == PCH_DP_D) - intel_encoder->clone_mask = (1 << INTEL_DP_D_CLONE_BIT); + intel_encoder->cloneable = false; - if (is_edp(intel_dp)) { - intel_encoder->clone_mask = (1 << INTEL_EDP_CLONE_BIT); - INIT_DELAYED_WORK(&intel_dp->panel_vdd_work, - ironlake_panel_vdd_work); - } + INIT_DELAYED_WORK(&intel_dp->panel_vdd_work, + ironlake_panel_vdd_work); intel_encoder->crtc_mask = (1 << 0) | (1 << 1) | (1 << 2); @@ -2508,29 +2491,33 @@ intel_dp_init(struct drm_device *dev, int output_reg) intel_connector_attach_encoder(intel_connector, intel_encoder); drm_sysfs_connector_add(connector); + intel_encoder->enable = intel_enable_dp; + intel_encoder->pre_enable = intel_pre_enable_dp; + intel_encoder->disable = intel_disable_dp; + intel_encoder->post_disable = intel_post_disable_dp; + intel_encoder->get_hw_state = intel_dp_get_hw_state; + intel_connector->get_hw_state = intel_connector_get_hw_state; + /* Set up the DDC bus. */ - switch (output_reg) { - case DP_A: - name = "DPDDC-A"; - break; - case DP_B: - case PCH_DP_B: - dev_priv->hotplug_supported_mask |= - DPB_HOTPLUG_INT_STATUS; - name = "DPDDC-B"; - break; - case DP_C: - case PCH_DP_C: - dev_priv->hotplug_supported_mask |= - DPC_HOTPLUG_INT_STATUS; - name = "DPDDC-C"; - break; - case DP_D: - case PCH_DP_D: - dev_priv->hotplug_supported_mask |= - DPD_HOTPLUG_INT_STATUS; - name = "DPDDC-D"; - break; + switch (port) { + case PORT_A: + name = "DPDDC-A"; + break; + case PORT_B: + dev_priv->hotplug_supported_mask |= DPB_HOTPLUG_INT_STATUS; + name = "DPDDC-B"; + break; + case PORT_C: + dev_priv->hotplug_supported_mask |= DPC_HOTPLUG_INT_STATUS; + name = "DPDDC-C"; + break; + case PORT_D: + dev_priv->hotplug_supported_mask |= DPD_HOTPLUG_INT_STATUS; + name = "DPDDC-D"; + break; + default: + WARN(1, "Invalid port %c\n", port_name(port)); + break; } /* Cache some DPCD data in the eDP case */ diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h index cd54cf88a28f..351fd7179cd7 100644 --- a/drivers/gpu/drm/i915/intel_drv.h +++ b/drivers/gpu/drm/i915/intel_drv.h @@ -31,6 +31,7 @@ #include "drm_crtc.h" #include "drm_crtc_helper.h" #include "drm_fb_helper.h" +#include "drm_dp_helper.h" #define _wait_for(COND, MS, W) ({ \ unsigned long timeout__ = jiffies + msecs_to_jiffies(MS); \ @@ -40,7 +41,11 @@ ret__ = -ETIMEDOUT; \ break; \ } \ - if (W && drm_can_sleep()) msleep(W); \ + if (W && drm_can_sleep()) { \ + msleep(W); \ + } else { \ + cpu_relax(); \ + } \ } \ ret__; \ }) @@ -90,25 +95,6 @@ #define INTEL_OUTPUT_DISPLAYPORT 7 #define INTEL_OUTPUT_EDP 8 -/* Intel Pipe Clone Bit */ -#define INTEL_HDMIB_CLONE_BIT 1 -#define INTEL_HDMIC_CLONE_BIT 2 -#define INTEL_HDMID_CLONE_BIT 3 -#define INTEL_HDMIE_CLONE_BIT 4 -#define INTEL_HDMIF_CLONE_BIT 5 -#define INTEL_SDVO_NON_TV_CLONE_BIT 6 -#define INTEL_SDVO_TV_CLONE_BIT 7 -#define INTEL_SDVO_LVDS_CLONE_BIT 8 -#define INTEL_ANALOG_CLONE_BIT 9 -#define INTEL_TV_CLONE_BIT 10 -#define INTEL_DP_B_CLONE_BIT 11 -#define INTEL_DP_C_CLONE_BIT 12 -#define INTEL_DP_D_CLONE_BIT 13 -#define INTEL_LVDS_CLONE_BIT 14 -#define INTEL_DVO_TMDS_CLONE_BIT 15 -#define INTEL_DVO_LVDS_CLONE_BIT 16 -#define INTEL_EDP_CLONE_BIT 17 - #define INTEL_DVO_CHIP_NONE 0 #define INTEL_DVO_CHIP_LVDS 1 #define INTEL_DVO_CHIP_TMDS 2 @@ -151,16 +137,48 @@ struct intel_fbdev { struct intel_encoder { struct drm_encoder base; + /* + * The new crtc this encoder will be driven from. Only differs from + * base->crtc while a modeset is in progress. + */ + struct intel_crtc *new_crtc; + int type; bool needs_tv_clock; + /* + * Intel hw has only one MUX where encoders could be clone, hence a + * simple flag is enough to compute the possible_clones mask. + */ + bool cloneable; + bool connectors_active; void (*hot_plug)(struct intel_encoder *); + void (*pre_enable)(struct intel_encoder *); + void (*enable)(struct intel_encoder *); + void (*disable)(struct intel_encoder *); + void (*post_disable)(struct intel_encoder *); + /* Read out the current hw state of this connector, returning true if + * the encoder is active. If the encoder is enabled it also set the pipe + * it is connected to in the pipe parameter. */ + bool (*get_hw_state)(struct intel_encoder *, enum pipe *pipe); int crtc_mask; - int clone_mask; }; struct intel_connector { struct drm_connector base; + /* + * The fixed encoder this connector is connected to. + */ struct intel_encoder *encoder; + + /* + * The new encoder this connector will be driven. Only differs from + * encoder while a modeset is in progress. + */ + struct intel_encoder *new_encoder; + + /* Reads out the current hw, returning true if the connector is enabled + * and active (i.e. dpms ON state). */ + bool (*get_hw_state)(struct intel_connector *); }; struct intel_crtc { @@ -168,11 +186,13 @@ struct intel_crtc { enum pipe pipe; enum plane plane; u8 lut_r[256], lut_g[256], lut_b[256]; - int dpms_mode; - bool active; /* is the crtc on? independent of the dpms mode */ + /* + * Whether the crtc and the connected output pipeline is active. Implies + * that crtc->enabled is set, i.e. the current mode configuration has + * some outputs connected to this crtc. + */ + bool active; bool primary_disabled; /* is the crtc obscured by a plane? */ - bool busy; /* is scanout buffer being updated frequently? */ - struct timer_list idle_timer; bool lowfreq_avail; struct intel_overlay *overlay; struct intel_unpin_work *unpin_work; @@ -311,6 +331,37 @@ struct intel_hdmi { struct drm_display_mode *adjusted_mode); }; +#define DP_RECEIVER_CAP_SIZE 0xf +#define DP_LINK_CONFIGURATION_SIZE 9 + +struct intel_dp { + struct intel_encoder base; + uint32_t output_reg; + uint32_t DP; + uint8_t link_configuration[DP_LINK_CONFIGURATION_SIZE]; + bool has_audio; + enum hdmi_force_audio force_audio; + enum port port; + uint32_t color_range; + uint8_t link_bw; + uint8_t lane_count; + uint8_t dpcd[DP_RECEIVER_CAP_SIZE]; + struct i2c_adapter adapter; + struct i2c_algo_dp_aux_data algo; + bool is_pch_edp; + uint8_t train_set[4]; + int panel_power_up_delay; + int panel_power_down_delay; + int panel_power_cycle_delay; + int backlight_on_delay; + int backlight_off_delay; + struct drm_display_mode *panel_fixed_mode; /* for eDP */ + struct delayed_work panel_vdd_work; + bool want_panel_vdd; + struct edid *edid; /* cached EDID for eDP */ + int edid_mode_count; +}; + static inline struct drm_crtc * intel_get_crtc_for_pipe(struct drm_device *dev, int pipe) { @@ -350,17 +401,21 @@ extern void intel_attach_force_audio_property(struct drm_connector *connector); extern void intel_attach_broadcast_rgb_property(struct drm_connector *connector); extern void intel_crt_init(struct drm_device *dev); -extern void intel_hdmi_init(struct drm_device *dev, int sdvox_reg); +extern void intel_hdmi_init(struct drm_device *dev, + int sdvox_reg, enum port port); extern struct intel_hdmi *enc_to_intel_hdmi(struct drm_encoder *encoder); extern void intel_dip_infoframe_csum(struct dip_infoframe *avi_if); extern bool intel_sdvo_init(struct drm_device *dev, uint32_t sdvo_reg, bool is_sdvob); extern void intel_dvo_init(struct drm_device *dev); extern void intel_tv_init(struct drm_device *dev); -extern void intel_mark_busy(struct drm_device *dev, - struct drm_i915_gem_object *obj); +extern void intel_mark_busy(struct drm_device *dev); +extern void intel_mark_idle(struct drm_device *dev); +extern void intel_mark_fb_busy(struct drm_i915_gem_object *obj); +extern void intel_mark_fb_idle(struct drm_i915_gem_object *obj); extern bool intel_lvds_init(struct drm_device *dev); -extern void intel_dp_init(struct drm_device *dev, int dp_reg); +extern void intel_dp_init(struct drm_device *dev, int output_reg, + enum port port); void intel_dp_set_m_n(struct drm_crtc *crtc, struct drm_display_mode *mode, struct drm_display_mode *adjusted_mode); @@ -373,8 +428,6 @@ extern int intel_plane_init(struct drm_device *dev, enum pipe pipe); extern void intel_flush_display_plane(struct drm_i915_private *dev_priv, enum plane plane); -void intel_sanitize_pm(struct drm_device *dev); - /* intel_panel.c */ extern void intel_fixed_panel_mode(struct drm_display_mode *fixed_mode, struct drm_display_mode *adjusted_mode); @@ -391,10 +444,27 @@ extern void intel_panel_disable_backlight(struct drm_device *dev); extern void intel_panel_destroy_backlight(struct drm_device *dev); extern enum drm_connector_status intel_panel_detect(struct drm_device *dev); +struct intel_set_config { + struct drm_encoder **save_connector_encoders; + struct drm_crtc **save_encoder_crtcs; + + bool fb_changed; + bool mode_changed; +}; + +extern bool intel_set_mode(struct drm_crtc *crtc, struct drm_display_mode *mode, + int x, int y, struct drm_framebuffer *old_fb); +extern void intel_modeset_disable(struct drm_device *dev); extern void intel_crtc_load_lut(struct drm_crtc *crtc); -extern void intel_encoder_prepare(struct drm_encoder *encoder); -extern void intel_encoder_commit(struct drm_encoder *encoder); +extern void intel_crtc_update_dpms(struct drm_crtc *crtc); +extern void intel_encoder_noop(struct drm_encoder *encoder); extern void intel_encoder_destroy(struct drm_encoder *encoder); +extern void intel_encoder_dpms(struct intel_encoder *encoder, int mode); +extern bool intel_encoder_check_is_cloned(struct intel_encoder *encoder); +extern void intel_connector_dpms(struct drm_connector *, int mode); +extern bool intel_connector_get_hw_state(struct intel_connector *connector); +extern void intel_modeset_check_state(struct drm_device *dev); + static inline struct intel_encoder *intel_attached_encoder(struct drm_connector *connector) { @@ -417,12 +487,10 @@ struct intel_load_detect_pipe { bool load_detect_temp; int dpms_mode; }; -extern bool intel_get_load_detect_pipe(struct intel_encoder *intel_encoder, - struct drm_connector *connector, +extern bool intel_get_load_detect_pipe(struct drm_connector *connector, struct drm_display_mode *mode, struct intel_load_detect_pipe *old); -extern void intel_release_load_detect_pipe(struct intel_encoder *intel_encoder, - struct drm_connector *connector, +extern void intel_release_load_detect_pipe(struct drm_connector *connector, struct intel_load_detect_pipe *old); extern void intelfb_restore(void); @@ -503,7 +571,10 @@ extern void intel_disable_gt_powersave(struct drm_device *dev); extern void gen6_gt_check_fifodbg(struct drm_i915_private *dev_priv); extern void ironlake_teardown_rc6(struct drm_device *dev); -extern void intel_ddi_dpms(struct drm_encoder *encoder, int mode); +extern void intel_enable_ddi(struct intel_encoder *encoder); +extern void intel_disable_ddi(struct intel_encoder *encoder); +extern bool intel_ddi_get_hw_state(struct intel_encoder *encoder, + enum pipe *pipe); extern void intel_ddi_mode_set(struct drm_encoder *encoder, struct drm_display_mode *mode, struct drm_display_mode *adjusted_mode); diff --git a/drivers/gpu/drm/i915/intel_dvo.c b/drivers/gpu/drm/i915/intel_dvo.c index 36c542e5036b..4f1fdcc44005 100644 --- a/drivers/gpu/drm/i915/intel_dvo.c +++ b/drivers/gpu/drm/i915/intel_dvo.c @@ -37,6 +37,7 @@ #define SIL164_ADDR 0x38 #define CH7xxx_ADDR 0x76 #define TFP410_ADDR 0x38 +#define NS2501_ADDR 0x38 static const struct intel_dvo_device intel_dvo_devices[] = { { @@ -74,7 +75,14 @@ static const struct intel_dvo_device intel_dvo_devices[] = { .slave_addr = 0x75, .gpio = GMBUS_PORT_DPB, .dev_ops = &ch7017_ops, - } + }, + { + .type = INTEL_DVO_CHIP_TMDS, + .name = "ns2501", + .dvo_reg = DVOC, + .slave_addr = NS2501_ADDR, + .dev_ops = &ns2501_ops, + } }; struct intel_dvo { @@ -97,22 +105,91 @@ static struct intel_dvo *intel_attached_dvo(struct drm_connector *connector) struct intel_dvo, base); } -static void intel_dvo_dpms(struct drm_encoder *encoder, int mode) +static bool intel_dvo_connector_get_hw_state(struct intel_connector *connector) { - struct drm_i915_private *dev_priv = encoder->dev->dev_private; - struct intel_dvo *intel_dvo = enc_to_intel_dvo(encoder); + struct intel_dvo *intel_dvo = intel_attached_dvo(&connector->base); + + return intel_dvo->dev.dev_ops->get_hw_state(&intel_dvo->dev); +} + +static bool intel_dvo_get_hw_state(struct intel_encoder *encoder, + enum pipe *pipe) +{ + struct drm_device *dev = encoder->base.dev; + struct drm_i915_private *dev_priv = dev->dev_private; + struct intel_dvo *intel_dvo = enc_to_intel_dvo(&encoder->base); + u32 tmp; + + tmp = I915_READ(intel_dvo->dev.dvo_reg); + + if (!(tmp & DVO_ENABLE)) + return false; + + *pipe = PORT_TO_PIPE(tmp); + + return true; +} + +static void intel_disable_dvo(struct intel_encoder *encoder) +{ + struct drm_i915_private *dev_priv = encoder->base.dev->dev_private; + struct intel_dvo *intel_dvo = enc_to_intel_dvo(&encoder->base); + u32 dvo_reg = intel_dvo->dev.dvo_reg; + u32 temp = I915_READ(dvo_reg); + + intel_dvo->dev.dev_ops->dpms(&intel_dvo->dev, false); + I915_WRITE(dvo_reg, temp & ~DVO_ENABLE); + I915_READ(dvo_reg); +} + +static void intel_enable_dvo(struct intel_encoder *encoder) +{ + struct drm_i915_private *dev_priv = encoder->base.dev->dev_private; + struct intel_dvo *intel_dvo = enc_to_intel_dvo(&encoder->base); u32 dvo_reg = intel_dvo->dev.dvo_reg; u32 temp = I915_READ(dvo_reg); + I915_WRITE(dvo_reg, temp | DVO_ENABLE); + I915_READ(dvo_reg); + intel_dvo->dev.dev_ops->dpms(&intel_dvo->dev, true); +} + +static void intel_dvo_dpms(struct drm_connector *connector, int mode) +{ + struct intel_dvo *intel_dvo = intel_attached_dvo(connector); + struct drm_crtc *crtc; + + /* dvo supports only 2 dpms states. */ + if (mode != DRM_MODE_DPMS_ON) + mode = DRM_MODE_DPMS_OFF; + + if (mode == connector->dpms) + return; + + connector->dpms = mode; + + /* Only need to change hw state when actually enabled */ + crtc = intel_dvo->base.base.crtc; + if (!crtc) { + intel_dvo->base.connectors_active = false; + return; + } + if (mode == DRM_MODE_DPMS_ON) { - I915_WRITE(dvo_reg, temp | DVO_ENABLE); - I915_READ(dvo_reg); - intel_dvo->dev.dev_ops->dpms(&intel_dvo->dev, mode); + intel_dvo->base.connectors_active = true; + + intel_crtc_update_dpms(crtc); + + intel_dvo->dev.dev_ops->dpms(&intel_dvo->dev, true); } else { - intel_dvo->dev.dev_ops->dpms(&intel_dvo->dev, mode); - I915_WRITE(dvo_reg, temp & ~DVO_ENABLE); - I915_READ(dvo_reg); + intel_dvo->dev.dev_ops->dpms(&intel_dvo->dev, false); + + intel_dvo->base.connectors_active = false; + + intel_crtc_update_dpms(crtc); } + + intel_modeset_check_state(connector->dev); } static int intel_dvo_mode_valid(struct drm_connector *connector, @@ -267,15 +344,13 @@ static void intel_dvo_destroy(struct drm_connector *connector) } static const struct drm_encoder_helper_funcs intel_dvo_helper_funcs = { - .dpms = intel_dvo_dpms, .mode_fixup = intel_dvo_mode_fixup, - .prepare = intel_encoder_prepare, .mode_set = intel_dvo_mode_set, - .commit = intel_encoder_commit, + .disable = intel_encoder_noop, }; static const struct drm_connector_funcs intel_dvo_connector_funcs = { - .dpms = drm_helper_connector_dpms, + .dpms = intel_dvo_dpms, .detect = intel_dvo_detect, .destroy = intel_dvo_destroy, .fill_modes = drm_helper_probe_single_connector_modes, @@ -364,6 +439,11 @@ void intel_dvo_init(struct drm_device *dev) drm_encoder_init(dev, &intel_encoder->base, &intel_dvo_enc_funcs, encoder_type); + intel_encoder->disable = intel_disable_dvo; + intel_encoder->enable = intel_enable_dvo; + intel_encoder->get_hw_state = intel_dvo_get_hw_state; + intel_connector->get_hw_state = intel_dvo_connector_get_hw_state; + /* Now, try to find a controller */ for (i = 0; i < ARRAY_SIZE(intel_dvo_devices); i++) { struct drm_connector *connector = &intel_connector->base; @@ -396,17 +476,14 @@ void intel_dvo_init(struct drm_device *dev) intel_encoder->crtc_mask = (1 << 0) | (1 << 1); switch (dvo->type) { case INTEL_DVO_CHIP_TMDS: - intel_encoder->clone_mask = - (1 << INTEL_DVO_TMDS_CLONE_BIT) | - (1 << INTEL_ANALOG_CLONE_BIT); + intel_encoder->cloneable = true; drm_connector_init(dev, connector, &intel_dvo_connector_funcs, DRM_MODE_CONNECTOR_DVII); encoder_type = DRM_MODE_ENCODER_TMDS; break; case INTEL_DVO_CHIP_LVDS: - intel_encoder->clone_mask = - (1 << INTEL_DVO_LVDS_CLONE_BIT); + intel_encoder->cloneable = false; drm_connector_init(dev, connector, &intel_dvo_connector_funcs, DRM_MODE_CONNECTOR_LVDS); diff --git a/drivers/gpu/drm/i915/intel_hdmi.c b/drivers/gpu/drm/i915/intel_hdmi.c index 12dc3308ab8c..08f2b63d740a 100644 --- a/drivers/gpu/drm/i915/intel_hdmi.c +++ b/drivers/gpu/drm/i915/intel_hdmi.c @@ -151,6 +151,9 @@ static void g4x_write_infoframe(struct drm_encoder *encoder, I915_WRITE(VIDEO_DIP_DATA, *data); data++; } + /* Write every possible data byte to force correct ECC calculation. */ + for (; i < VIDEO_DIP_DATA_SIZE; i += 4) + I915_WRITE(VIDEO_DIP_DATA, 0); mmiowb(); val |= g4x_infoframe_enable(frame); @@ -186,6 +189,9 @@ static void ibx_write_infoframe(struct drm_encoder *encoder, I915_WRITE(TVIDEO_DIP_DATA(intel_crtc->pipe), *data); data++; } + /* Write every possible data byte to force correct ECC calculation. */ + for (; i < VIDEO_DIP_DATA_SIZE; i += 4) + I915_WRITE(TVIDEO_DIP_DATA(intel_crtc->pipe), 0); mmiowb(); val |= g4x_infoframe_enable(frame); @@ -224,6 +230,9 @@ static void cpt_write_infoframe(struct drm_encoder *encoder, I915_WRITE(TVIDEO_DIP_DATA(intel_crtc->pipe), *data); data++; } + /* Write every possible data byte to force correct ECC calculation. */ + for (; i < VIDEO_DIP_DATA_SIZE; i += 4) + I915_WRITE(TVIDEO_DIP_DATA(intel_crtc->pipe), 0); mmiowb(); val |= g4x_infoframe_enable(frame); @@ -259,6 +268,9 @@ static void vlv_write_infoframe(struct drm_encoder *encoder, I915_WRITE(VLV_TVIDEO_DIP_DATA(intel_crtc->pipe), *data); data++; } + /* Write every possible data byte to force correct ECC calculation. */ + for (; i < VIDEO_DIP_DATA_SIZE; i += 4) + I915_WRITE(VLV_TVIDEO_DIP_DATA(intel_crtc->pipe), 0); mmiowb(); val |= g4x_infoframe_enable(frame); @@ -292,6 +304,9 @@ static void hsw_write_infoframe(struct drm_encoder *encoder, I915_WRITE(data_reg + i, *data); data++; } + /* Write every possible data byte to force correct ECC calculation. */ + for (; i < VIDEO_DIP_DATA_SIZE; i += 4) + I915_WRITE(data_reg + i, 0); mmiowb(); val |= hsw_infoframe_enable(frame); @@ -377,6 +392,7 @@ static void g4x_set_infoframes(struct drm_encoder *encoder, port = VIDEO_DIP_PORT_C; break; default: + BUG(); return; } @@ -435,6 +451,7 @@ static void ibx_set_infoframes(struct drm_encoder *encoder, port = VIDEO_DIP_PORT_D; break; default: + BUG(); return; } @@ -601,15 +618,36 @@ static void intel_hdmi_mode_set(struct drm_encoder *encoder, intel_hdmi->set_infoframes(encoder, adjusted_mode); } -static void intel_hdmi_dpms(struct drm_encoder *encoder, int mode) +static bool intel_hdmi_get_hw_state(struct intel_encoder *encoder, + enum pipe *pipe) { - struct drm_device *dev = encoder->dev; + struct drm_device *dev = encoder->base.dev; struct drm_i915_private *dev_priv = dev->dev_private; - struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(encoder); + struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(&encoder->base); + u32 tmp; + + tmp = I915_READ(intel_hdmi->sdvox_reg); + + if (!(tmp & SDVO_ENABLE)) + return false; + + if (HAS_PCH_CPT(dev)) + *pipe = PORT_TO_PIPE_CPT(tmp); + else + *pipe = PORT_TO_PIPE(tmp); + + return true; +} + +static void intel_enable_hdmi(struct intel_encoder *encoder) +{ + struct drm_device *dev = encoder->base.dev; + struct drm_i915_private *dev_priv = dev->dev_private; + struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(&encoder->base); u32 temp; u32 enable_bits = SDVO_ENABLE; - if (intel_hdmi->has_audio || mode != DRM_MODE_DPMS_ON) + if (intel_hdmi->has_audio) enable_bits |= SDVO_AUDIO_ENABLE; temp = I915_READ(intel_hdmi->sdvox_reg); @@ -617,31 +655,12 @@ static void intel_hdmi_dpms(struct drm_encoder *encoder, int mode) /* HW workaround for IBX, we need to move the port to transcoder A * before disabling it. */ if (HAS_PCH_IBX(dev)) { - struct drm_crtc *crtc = encoder->crtc; + struct drm_crtc *crtc = encoder->base.crtc; int pipe = crtc ? to_intel_crtc(crtc)->pipe : -1; - if (mode != DRM_MODE_DPMS_ON) { - if (temp & SDVO_PIPE_B_SELECT) { - temp &= ~SDVO_PIPE_B_SELECT; - I915_WRITE(intel_hdmi->sdvox_reg, temp); - POSTING_READ(intel_hdmi->sdvox_reg); - - /* Again we need to write this twice. */ - I915_WRITE(intel_hdmi->sdvox_reg, temp); - POSTING_READ(intel_hdmi->sdvox_reg); - - /* Transcoder selection bits only update - * effectively on vblank. */ - if (crtc) - intel_wait_for_vblank(dev, pipe); - else - msleep(50); - } - } else { - /* Restore the transcoder select bit. */ - if (pipe == PIPE_B) - enable_bits |= SDVO_PIPE_B_SELECT; - } + /* Restore the transcoder select bit. */ + if (pipe == PIPE_B) + enable_bits |= SDVO_PIPE_B_SELECT; } /* HW workaround, need to toggle enable bit off and on for 12bpc, but @@ -652,12 +671,64 @@ static void intel_hdmi_dpms(struct drm_encoder *encoder, int mode) POSTING_READ(intel_hdmi->sdvox_reg); } - if (mode != DRM_MODE_DPMS_ON) { - temp &= ~enable_bits; - } else { - temp |= enable_bits; + temp |= enable_bits; + + I915_WRITE(intel_hdmi->sdvox_reg, temp); + POSTING_READ(intel_hdmi->sdvox_reg); + + /* HW workaround, need to write this twice for issue that may result + * in first write getting masked. + */ + if (HAS_PCH_SPLIT(dev)) { + I915_WRITE(intel_hdmi->sdvox_reg, temp); + POSTING_READ(intel_hdmi->sdvox_reg); + } +} + +static void intel_disable_hdmi(struct intel_encoder *encoder) +{ + struct drm_device *dev = encoder->base.dev; + struct drm_i915_private *dev_priv = dev->dev_private; + struct intel_hdmi *intel_hdmi = enc_to_intel_hdmi(&encoder->base); + u32 temp; + u32 enable_bits = SDVO_ENABLE | SDVO_AUDIO_ENABLE; + + temp = I915_READ(intel_hdmi->sdvox_reg); + + /* HW workaround for IBX, we need to move the port to transcoder A + * before disabling it. */ + if (HAS_PCH_IBX(dev)) { + struct drm_crtc *crtc = encoder->base.crtc; + int pipe = crtc ? to_intel_crtc(crtc)->pipe : -1; + + if (temp & SDVO_PIPE_B_SELECT) { + temp &= ~SDVO_PIPE_B_SELECT; + I915_WRITE(intel_hdmi->sdvox_reg, temp); + POSTING_READ(intel_hdmi->sdvox_reg); + + /* Again we need to write this twice. */ + I915_WRITE(intel_hdmi->sdvox_reg, temp); + POSTING_READ(intel_hdmi->sdvox_reg); + + /* Transcoder selection bits only update + * effectively on vblank. */ + if (crtc) + intel_wait_for_vblank(dev, pipe); + else + msleep(50); + } + } + + /* HW workaround, need to toggle enable bit off and on for 12bpc, but + * we do this anyway which shows more stable in testing. + */ + if (HAS_PCH_SPLIT(dev)) { + I915_WRITE(intel_hdmi->sdvox_reg, temp & ~SDVO_ENABLE); + POSTING_READ(intel_hdmi->sdvox_reg); } + temp &= ~enable_bits; + I915_WRITE(intel_hdmi->sdvox_reg, temp); POSTING_READ(intel_hdmi->sdvox_reg); @@ -737,7 +808,6 @@ intel_hdmi_detect(struct drm_connector *connector, bool force) drm_detect_hdmi_monitor(edid); intel_hdmi->has_audio = drm_detect_monitor_audio(edid); } - connector->display_info.raw_edid = NULL; kfree(edid); } @@ -778,8 +848,6 @@ intel_hdmi_detect_audio(struct drm_connector *connector) if (edid) { if (edid->input & DRM_EDID_INPUT_DIGITAL) has_audio = drm_detect_monitor_audio(edid); - - connector->display_info.raw_edid = NULL; kfree(edid); } @@ -833,9 +901,8 @@ intel_hdmi_set_property(struct drm_connector *connector, done: if (intel_hdmi->base.base.crtc) { struct drm_crtc *crtc = intel_hdmi->base.base.crtc; - drm_crtc_helper_set_mode(crtc, &crtc->mode, - crtc->x, crtc->y, - crtc->fb); + intel_set_mode(crtc, &crtc->mode, + crtc->x, crtc->y, crtc->fb); } return 0; @@ -849,23 +916,19 @@ static void intel_hdmi_destroy(struct drm_connector *connector) } static const struct drm_encoder_helper_funcs intel_hdmi_helper_funcs_hsw = { - .dpms = intel_ddi_dpms, .mode_fixup = intel_hdmi_mode_fixup, - .prepare = intel_encoder_prepare, .mode_set = intel_ddi_mode_set, - .commit = intel_encoder_commit, + .disable = intel_encoder_noop, }; static const struct drm_encoder_helper_funcs intel_hdmi_helper_funcs = { - .dpms = intel_hdmi_dpms, .mode_fixup = intel_hdmi_mode_fixup, - .prepare = intel_encoder_prepare, .mode_set = intel_hdmi_mode_set, - .commit = intel_encoder_commit, + .disable = intel_encoder_noop, }; static const struct drm_connector_funcs intel_hdmi_connector_funcs = { - .dpms = drm_helper_connector_dpms, + .dpms = intel_connector_dpms, .detect = intel_hdmi_detect, .fill_modes = drm_helper_probe_single_connector_modes, .set_property = intel_hdmi_set_property, @@ -889,7 +952,7 @@ intel_hdmi_add_properties(struct intel_hdmi *intel_hdmi, struct drm_connector *c intel_attach_broadcast_rgb_property(connector); } -void intel_hdmi_init(struct drm_device *dev, int sdvox_reg) +void intel_hdmi_init(struct drm_device *dev, int sdvox_reg, enum port port) { struct drm_i915_private *dev_priv = dev->dev_private; struct drm_connector *connector; @@ -923,48 +986,25 @@ void intel_hdmi_init(struct drm_device *dev, int sdvox_reg) connector->doublescan_allowed = 0; intel_encoder->crtc_mask = (1 << 0) | (1 << 1) | (1 << 2); - /* Set up the DDC bus. */ - if (sdvox_reg == SDVOB) { - intel_encoder->clone_mask = (1 << INTEL_HDMIB_CLONE_BIT); - intel_hdmi->ddc_bus = GMBUS_PORT_DPB; - dev_priv->hotplug_supported_mask |= HDMIB_HOTPLUG_INT_STATUS; - } else if (sdvox_reg == SDVOC) { - intel_encoder->clone_mask = (1 << INTEL_HDMIC_CLONE_BIT); - intel_hdmi->ddc_bus = GMBUS_PORT_DPC; - dev_priv->hotplug_supported_mask |= HDMIC_HOTPLUG_INT_STATUS; - } else if (sdvox_reg == HDMIB) { - intel_encoder->clone_mask = (1 << INTEL_HDMID_CLONE_BIT); - intel_hdmi->ddc_bus = GMBUS_PORT_DPB; - dev_priv->hotplug_supported_mask |= HDMIB_HOTPLUG_INT_STATUS; - } else if (sdvox_reg == HDMIC) { - intel_encoder->clone_mask = (1 << INTEL_HDMIE_CLONE_BIT); - intel_hdmi->ddc_bus = GMBUS_PORT_DPC; - dev_priv->hotplug_supported_mask |= HDMIC_HOTPLUG_INT_STATUS; - } else if (sdvox_reg == HDMID) { - intel_encoder->clone_mask = (1 << INTEL_HDMIF_CLONE_BIT); - intel_hdmi->ddc_bus = GMBUS_PORT_DPD; - dev_priv->hotplug_supported_mask |= HDMID_HOTPLUG_INT_STATUS; - } else if (sdvox_reg == DDI_BUF_CTL(PORT_B)) { - DRM_DEBUG_DRIVER("LPT: detected output on DDI B\n"); - intel_encoder->clone_mask = (1 << INTEL_HDMIB_CLONE_BIT); + intel_encoder->cloneable = false; + + intel_hdmi->ddi_port = port; + switch (port) { + case PORT_B: intel_hdmi->ddc_bus = GMBUS_PORT_DPB; - intel_hdmi->ddi_port = PORT_B; dev_priv->hotplug_supported_mask |= HDMIB_HOTPLUG_INT_STATUS; - } else if (sdvox_reg == DDI_BUF_CTL(PORT_C)) { - DRM_DEBUG_DRIVER("LPT: detected output on DDI C\n"); - intel_encoder->clone_mask = (1 << INTEL_HDMIC_CLONE_BIT); + break; + case PORT_C: intel_hdmi->ddc_bus = GMBUS_PORT_DPC; - intel_hdmi->ddi_port = PORT_C; dev_priv->hotplug_supported_mask |= HDMIC_HOTPLUG_INT_STATUS; - } else if (sdvox_reg == DDI_BUF_CTL(PORT_D)) { - DRM_DEBUG_DRIVER("LPT: detected output on DDI D\n"); - intel_encoder->clone_mask = (1 << INTEL_HDMID_CLONE_BIT); + break; + case PORT_D: intel_hdmi->ddc_bus = GMBUS_PORT_DPD; - intel_hdmi->ddi_port = PORT_D; dev_priv->hotplug_supported_mask |= HDMID_HOTPLUG_INT_STATUS; - } else { - /* If we got an unknown sdvox_reg, things are pretty much broken - * in a way that we should let the kernel know about it */ + break; + case PORT_A: + /* Internal port only for eDP. */ + default: BUG(); } @@ -987,10 +1027,21 @@ void intel_hdmi_init(struct drm_device *dev, int sdvox_reg) intel_hdmi->set_infoframes = cpt_set_infoframes; } - if (IS_HASWELL(dev)) - drm_encoder_helper_add(&intel_encoder->base, &intel_hdmi_helper_funcs_hsw); - else - drm_encoder_helper_add(&intel_encoder->base, &intel_hdmi_helper_funcs); + if (IS_HASWELL(dev)) { + intel_encoder->enable = intel_enable_ddi; + intel_encoder->disable = intel_disable_ddi; + intel_encoder->get_hw_state = intel_ddi_get_hw_state; + drm_encoder_helper_add(&intel_encoder->base, + &intel_hdmi_helper_funcs_hsw); + } else { + intel_encoder->enable = intel_enable_hdmi; + intel_encoder->disable = intel_disable_hdmi; + intel_encoder->get_hw_state = intel_hdmi_get_hw_state; + drm_encoder_helper_add(&intel_encoder->base, + &intel_hdmi_helper_funcs); + } + intel_connector->get_hw_state = intel_connector_get_hw_state; + intel_hdmi_add_properties(intel_hdmi, connector); diff --git a/drivers/gpu/drm/i915/intel_lvds.c b/drivers/gpu/drm/i915/intel_lvds.c index e9a6f6aaed85..40d72bd64e11 100644 --- a/drivers/gpu/drm/i915/intel_lvds.c +++ b/drivers/gpu/drm/i915/intel_lvds.c @@ -65,13 +65,40 @@ static struct intel_lvds *intel_attached_lvds(struct drm_connector *connector) struct intel_lvds, base); } +static bool intel_lvds_get_hw_state(struct intel_encoder *encoder, + enum pipe *pipe) +{ + struct drm_device *dev = encoder->base.dev; + struct drm_i915_private *dev_priv = dev->dev_private; + u32 lvds_reg, tmp; + + if (HAS_PCH_SPLIT(dev)) { + lvds_reg = PCH_LVDS; + } else { + lvds_reg = LVDS; + } + + tmp = I915_READ(lvds_reg); + + if (!(tmp & LVDS_PORT_EN)) + return false; + + if (HAS_PCH_CPT(dev)) + *pipe = PORT_TO_PIPE_CPT(tmp); + else + *pipe = PORT_TO_PIPE(tmp); + + return true; +} + /** * Sets the power state for the panel. */ -static void intel_lvds_enable(struct intel_lvds *intel_lvds) +static void intel_enable_lvds(struct intel_encoder *encoder) { - struct drm_device *dev = intel_lvds->base.base.dev; - struct intel_crtc *intel_crtc = to_intel_crtc(intel_lvds->base.base.crtc); + struct drm_device *dev = encoder->base.dev; + struct intel_lvds *intel_lvds = to_intel_lvds(&encoder->base); + struct intel_crtc *intel_crtc = to_intel_crtc(encoder->base.crtc); struct drm_i915_private *dev_priv = dev->dev_private; u32 ctl_reg, lvds_reg, stat_reg; @@ -111,9 +138,10 @@ static void intel_lvds_enable(struct intel_lvds *intel_lvds) intel_panel_enable_backlight(dev, intel_crtc->pipe); } -static void intel_lvds_disable(struct intel_lvds *intel_lvds) +static void intel_disable_lvds(struct intel_encoder *encoder) { - struct drm_device *dev = intel_lvds->base.base.dev; + struct drm_device *dev = encoder->base.dev; + struct intel_lvds *intel_lvds = to_intel_lvds(&encoder->base); struct drm_i915_private *dev_priv = dev->dev_private; u32 ctl_reg, lvds_reg, stat_reg; @@ -142,18 +170,6 @@ static void intel_lvds_disable(struct intel_lvds *intel_lvds) POSTING_READ(lvds_reg); } -static void intel_lvds_dpms(struct drm_encoder *encoder, int mode) -{ - struct intel_lvds *intel_lvds = to_intel_lvds(encoder); - - if (mode == DRM_MODE_DPMS_ON) - intel_lvds_enable(intel_lvds); - else - intel_lvds_disable(intel_lvds); - - /* XXX: We never power down the LVDS pairs. */ -} - static int intel_lvds_mode_valid(struct drm_connector *connector, struct drm_display_mode *mode) { @@ -234,9 +250,8 @@ static bool intel_lvds_mode_fixup(struct drm_encoder *encoder, { struct drm_device *dev = encoder->dev; struct drm_i915_private *dev_priv = dev->dev_private; - struct intel_crtc *intel_crtc = to_intel_crtc(encoder->crtc); struct intel_lvds *intel_lvds = to_intel_lvds(encoder); - struct intel_encoder *tmp_encoder; + struct intel_crtc *intel_crtc = intel_lvds->base.new_crtc; u32 pfit_control = 0, pfit_pgm_ratios = 0, border = 0; int pipe; @@ -246,14 +261,8 @@ static bool intel_lvds_mode_fixup(struct drm_encoder *encoder, return false; } - /* Should never happen!! */ - for_each_encoder_on_crtc(dev, encoder->crtc, tmp_encoder) { - if (&tmp_encoder->base != encoder) { - DRM_ERROR("Can't enable LVDS and another " - "encoder on the same pipe\n"); - return false; - } - } + if (intel_encoder_check_is_cloned(&intel_lvds->base)) + return false; /* * We have timings from the BIOS for the panel, put them in @@ -405,23 +414,6 @@ out: return true; } -static void intel_lvds_prepare(struct drm_encoder *encoder) -{ - struct intel_lvds *intel_lvds = to_intel_lvds(encoder); - - intel_lvds_disable(intel_lvds); -} - -static void intel_lvds_commit(struct drm_encoder *encoder) -{ - struct intel_lvds *intel_lvds = to_intel_lvds(encoder); - - /* Always do a full power on as we do not know what state - * we were left in. - */ - intel_lvds_enable(intel_lvds); -} - static void intel_lvds_mode_set(struct drm_encoder *encoder, struct drm_display_mode *mode, struct drm_display_mode *adjusted_mode) @@ -535,7 +527,7 @@ static int intel_lid_notify(struct notifier_block *nb, unsigned long val, dev_priv->modeset_on_lid = 0; mutex_lock(&dev->mode_config.mutex); - drm_helper_resume_force_mode(dev); + intel_modeset_check_state(dev); mutex_unlock(&dev->mode_config.mutex); return NOTIFY_OK; @@ -587,8 +579,8 @@ static int intel_lvds_set_property(struct drm_connector *connector, * If the CRTC is enabled, the display will be changed * according to the new panel fitting mode. */ - drm_crtc_helper_set_mode(crtc, &crtc->mode, - crtc->x, crtc->y, crtc->fb); + intel_set_mode(crtc, &crtc->mode, + crtc->x, crtc->y, crtc->fb); } } @@ -596,11 +588,9 @@ static int intel_lvds_set_property(struct drm_connector *connector, } static const struct drm_encoder_helper_funcs intel_lvds_helper_funcs = { - .dpms = intel_lvds_dpms, .mode_fixup = intel_lvds_mode_fixup, - .prepare = intel_lvds_prepare, .mode_set = intel_lvds_mode_set, - .commit = intel_lvds_commit, + .disable = intel_encoder_noop, }; static const struct drm_connector_helper_funcs intel_lvds_connector_helper_funcs = { @@ -610,7 +600,7 @@ static const struct drm_connector_helper_funcs intel_lvds_connector_helper_funcs }; static const struct drm_connector_funcs intel_lvds_connector_funcs = { - .dpms = drm_helper_connector_dpms, + .dpms = intel_connector_dpms, .detect = intel_lvds_detect, .fill_modes = drm_helper_probe_single_connector_modes, .set_property = intel_lvds_set_property, @@ -972,10 +962,15 @@ bool intel_lvds_init(struct drm_device *dev) drm_encoder_init(dev, &intel_encoder->base, &intel_lvds_enc_funcs, DRM_MODE_ENCODER_LVDS); + intel_encoder->enable = intel_enable_lvds; + intel_encoder->disable = intel_disable_lvds; + intel_encoder->get_hw_state = intel_lvds_get_hw_state; + intel_connector->get_hw_state = intel_connector_get_hw_state; + intel_connector_attach_encoder(intel_connector, intel_encoder); intel_encoder->type = INTEL_OUTPUT_LVDS; - intel_encoder->clone_mask = (1 << INTEL_LVDS_CLONE_BIT); + intel_encoder->cloneable = false; if (HAS_PCH_SPLIT(dev)) intel_encoder->crtc_mask = (1 << 0) | (1 << 1) | (1 << 2); else if (IS_GEN4(dev)) diff --git a/drivers/gpu/drm/i915/intel_modes.c b/drivers/gpu/drm/i915/intel_modes.c index 29b72593fbb2..4bc1c0fc342a 100644 --- a/drivers/gpu/drm/i915/intel_modes.c +++ b/drivers/gpu/drm/i915/intel_modes.c @@ -45,7 +45,6 @@ int intel_connector_update_modes(struct drm_connector *connector, drm_mode_connector_update_edid_property(connector, edid); ret = drm_add_edid_modes(connector, edid); drm_edid_to_eld(connector, edid); - connector->display_info.raw_edid = NULL; kfree(edid); return ret; diff --git a/drivers/gpu/drm/i915/intel_opregion.c b/drivers/gpu/drm/i915/intel_opregion.c index 18bd0af855dc..e27c17012628 100644 --- a/drivers/gpu/drm/i915/intel_opregion.c +++ b/drivers/gpu/drm/i915/intel_opregion.c @@ -427,6 +427,25 @@ blind_set: goto end; } +static void intel_setup_cadls(struct drm_device *dev) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + struct intel_opregion *opregion = &dev_priv->opregion; + int i = 0; + u32 disp_id; + + /* Initialize the CADL field by duplicating the DIDL values. + * Technically, this is not always correct as display outputs may exist, + * but not active. This initialization is necessary for some Clevo + * laptops that check this field before processing the brightness and + * display switching hotkeys. Just like DIDL, CADL is NULL-terminated if + * there are less than eight devices. */ + do { + disp_id = ioread32(&opregion->acpi->didl[i]); + iowrite32(disp_id, &opregion->acpi->cadl[i]); + } while (++i < 8 && disp_id != 0); +} + void intel_opregion_init(struct drm_device *dev) { struct drm_i915_private *dev_priv = dev->dev_private; @@ -436,8 +455,10 @@ void intel_opregion_init(struct drm_device *dev) return; if (opregion->acpi) { - if (drm_core_check_feature(dev, DRIVER_MODESET)) + if (drm_core_check_feature(dev, DRIVER_MODESET)) { intel_didl_outputs(dev); + intel_setup_cadls(dev); + } /* Notify BIOS we are ready to handle ACPI video ext notifs. * Right now, all the events are handled by the ACPI video module. diff --git a/drivers/gpu/drm/i915/intel_overlay.c b/drivers/gpu/drm/i915/intel_overlay.c index 830d0dd610e1..afd0f30ab882 100644 --- a/drivers/gpu/drm/i915/intel_overlay.c +++ b/drivers/gpu/drm/i915/intel_overlay.c @@ -235,54 +235,6 @@ static int intel_overlay_do_wait_request(struct intel_overlay *overlay, return 0; } -/* Workaround for i830 bug where pipe a must be enable to change control regs */ -static int -i830_activate_pipe_a(struct drm_device *dev) -{ - drm_i915_private_t *dev_priv = dev->dev_private; - struct intel_crtc *crtc; - struct drm_crtc_helper_funcs *crtc_funcs; - struct drm_display_mode vesa_640x480 = { - DRM_MODE("640x480", DRM_MODE_TYPE_DRIVER, 25175, 640, 656, - 752, 800, 0, 480, 489, 492, 525, 0, - DRM_MODE_FLAG_NHSYNC | DRM_MODE_FLAG_NVSYNC) - }, *mode; - - crtc = to_intel_crtc(dev_priv->pipe_to_crtc_mapping[0]); - if (crtc->dpms_mode == DRM_MODE_DPMS_ON) - return 0; - - /* most i8xx have pipe a forced on, so don't trust dpms mode */ - if (I915_READ(_PIPEACONF) & PIPECONF_ENABLE) - return 0; - - crtc_funcs = crtc->base.helper_private; - if (crtc_funcs->dpms == NULL) - return 0; - - DRM_DEBUG_DRIVER("Enabling pipe A in order to enable overlay\n"); - - mode = drm_mode_duplicate(dev, &vesa_640x480); - - if (!drm_crtc_helper_set_mode(&crtc->base, mode, - crtc->base.x, crtc->base.y, - crtc->base.fb)) - return 0; - - crtc_funcs->dpms(&crtc->base, DRM_MODE_DPMS_ON); - return 1; -} - -static void -i830_deactivate_pipe_a(struct drm_device *dev) -{ - drm_i915_private_t *dev_priv = dev->dev_private; - struct drm_crtc *crtc = dev_priv->pipe_to_crtc_mapping[0]; - struct drm_crtc_helper_funcs *crtc_funcs = crtc->helper_private; - - crtc_funcs->dpms(crtc, DRM_MODE_DPMS_OFF); -} - /* overlay needs to be disable in OCMD reg */ static int intel_overlay_on(struct intel_overlay *overlay) { @@ -290,17 +242,12 @@ static int intel_overlay_on(struct intel_overlay *overlay) struct drm_i915_private *dev_priv = dev->dev_private; struct intel_ring_buffer *ring = &dev_priv->ring[RCS]; struct drm_i915_gem_request *request; - int pipe_a_quirk = 0; int ret; BUG_ON(overlay->active); overlay->active = 1; - if (IS_I830(dev)) { - pipe_a_quirk = i830_activate_pipe_a(dev); - if (pipe_a_quirk < 0) - return pipe_a_quirk; - } + WARN_ON(IS_I830(dev) && !(dev_priv->quirks & QUIRK_PIPEA_FORCE)); request = kzalloc(sizeof(*request), GFP_KERNEL); if (request == NULL) { @@ -322,9 +269,6 @@ static int intel_overlay_on(struct intel_overlay *overlay) ret = intel_overlay_do_wait_request(overlay, request, NULL); out: - if (pipe_a_quirk) - i830_deactivate_pipe_a(dev); - return ret; } @@ -1439,7 +1383,7 @@ void intel_setup_overlay(struct drm_device *dev) } overlay->flip_addr = reg_bo->phys_obj->handle->busaddr; } else { - ret = i915_gem_object_pin(reg_bo, PAGE_SIZE, true); + ret = i915_gem_object_pin(reg_bo, PAGE_SIZE, true, false); if (ret) { DRM_ERROR("failed to pin overlay register bo\n"); goto out_free_bo; diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c index ba8a27b1757a..d69f8f49beb5 100644 --- a/drivers/gpu/drm/i915/intel_pm.c +++ b/drivers/gpu/drm/i915/intel_pm.c @@ -31,6 +31,8 @@ #include "../../../platform/x86/intel_ips.h" #include <linux/module.h> +#define FORCEWAKE_ACK_TIMEOUT_MS 2 + /* FBC, or Frame Buffer Compression, is a technique employed to compress the * framebuffer contents in-memory, aiming at reducing the required bandwidth * during in-memory transfers and, therefore, reduce the power packet. @@ -593,7 +595,7 @@ static void i915_ironlake_get_mem_freq(struct drm_device *dev) break; } - dev_priv->r_t = dev_priv->mem_freq; + dev_priv->ips.r_t = dev_priv->mem_freq; switch (csipll & 0x3ff) { case 0x00c: @@ -625,11 +627,11 @@ static void i915_ironlake_get_mem_freq(struct drm_device *dev) } if (dev_priv->fsb_freq == 3200) { - dev_priv->c_m = 0; + dev_priv->ips.c_m = 0; } else if (dev_priv->fsb_freq > 3200 && dev_priv->fsb_freq <= 4800) { - dev_priv->c_m = 1; + dev_priv->ips.c_m = 1; } else { - dev_priv->c_m = 2; + dev_priv->ips.c_m = 2; } } @@ -2138,7 +2140,7 @@ intel_alloc_context_page(struct drm_device *dev) return NULL; } - ret = i915_gem_object_pin(ctx, 4096, true); + ret = i915_gem_object_pin(ctx, 4096, true, false); if (ret) { DRM_ERROR("failed to pin power context: %d\n", ret); goto err_unref; @@ -2160,11 +2162,22 @@ err_unref: return NULL; } +/** + * Lock protecting IPS related data structures + */ +DEFINE_SPINLOCK(mchdev_lock); + +/* Global for IPS driver to get at the current i915 device. Protected by + * mchdev_lock. */ +static struct drm_i915_private *i915_mch_dev; + bool ironlake_set_drps(struct drm_device *dev, u8 val) { struct drm_i915_private *dev_priv = dev->dev_private; u16 rgvswctl; + assert_spin_locked(&mchdev_lock); + rgvswctl = I915_READ16(MEMSWCTL); if (rgvswctl & MEMCTL_CMD_STS) { DRM_DEBUG("gpu busy, RCS change rejected\n"); @@ -2188,6 +2201,8 @@ static void ironlake_enable_drps(struct drm_device *dev) u32 rgvmodectl = I915_READ(MEMMODECTL); u8 fmax, fmin, fstart, vstart; + spin_lock_irq(&mchdev_lock); + /* Enable temp reporting */ I915_WRITE16(PMMISC, I915_READ(PMMISC) | MCPPCE_EN); I915_WRITE16(TSC1, I915_READ(TSC1) | TSE); @@ -2211,12 +2226,12 @@ static void ironlake_enable_drps(struct drm_device *dev) vstart = (I915_READ(PXVFREQ_BASE + (fstart * 4)) & PXVFREQ_PX_MASK) >> PXVFREQ_PX_SHIFT; - dev_priv->fmax = fmax; /* IPS callback will increase this */ - dev_priv->fstart = fstart; + dev_priv->ips.fmax = fmax; /* IPS callback will increase this */ + dev_priv->ips.fstart = fstart; - dev_priv->max_delay = fstart; - dev_priv->min_delay = fmin; - dev_priv->cur_delay = fstart; + dev_priv->ips.max_delay = fstart; + dev_priv->ips.min_delay = fmin; + dev_priv->ips.cur_delay = fstart; DRM_DEBUG_DRIVER("fmax: %d, fmin: %d, fstart: %d\n", fmax, fmin, fstart); @@ -2233,23 +2248,29 @@ static void ironlake_enable_drps(struct drm_device *dev) rgvmodectl |= MEMMODE_SWMODE_EN; I915_WRITE(MEMMODECTL, rgvmodectl); - if (wait_for((I915_READ(MEMSWCTL) & MEMCTL_CMD_STS) == 0, 10)) + if (wait_for_atomic((I915_READ(MEMSWCTL) & MEMCTL_CMD_STS) == 0, 10)) DRM_ERROR("stuck trying to change perf mode\n"); - msleep(1); + mdelay(1); ironlake_set_drps(dev, fstart); - dev_priv->last_count1 = I915_READ(0x112e4) + I915_READ(0x112e8) + + dev_priv->ips.last_count1 = I915_READ(0x112e4) + I915_READ(0x112e8) + I915_READ(0x112e0); - dev_priv->last_time1 = jiffies_to_msecs(jiffies); - dev_priv->last_count2 = I915_READ(0x112f4); - getrawmonotonic(&dev_priv->last_time2); + dev_priv->ips.last_time1 = jiffies_to_msecs(jiffies); + dev_priv->ips.last_count2 = I915_READ(0x112f4); + getrawmonotonic(&dev_priv->ips.last_time2); + + spin_unlock_irq(&mchdev_lock); } static void ironlake_disable_drps(struct drm_device *dev) { struct drm_i915_private *dev_priv = dev->dev_private; - u16 rgvswctl = I915_READ16(MEMSWCTL); + u16 rgvswctl; + + spin_lock_irq(&mchdev_lock); + + rgvswctl = I915_READ16(MEMSWCTL); /* Ack interrupts, disable EFC interrupt */ I915_WRITE(MEMINTREN, I915_READ(MEMINTREN) & ~MEMINT_EVAL_CHG_EN); @@ -2259,31 +2280,54 @@ static void ironlake_disable_drps(struct drm_device *dev) I915_WRITE(DEIMR, I915_READ(DEIMR) | DE_PCU_EVENT); /* Go back to the starting frequency */ - ironlake_set_drps(dev, dev_priv->fstart); - msleep(1); + ironlake_set_drps(dev, dev_priv->ips.fstart); + mdelay(1); rgvswctl |= MEMCTL_CMD_STS; I915_WRITE(MEMSWCTL, rgvswctl); - msleep(1); + mdelay(1); + spin_unlock_irq(&mchdev_lock); } -void gen6_set_rps(struct drm_device *dev, u8 val) +/* There's a funny hw issue where the hw returns all 0 when reading from + * GEN6_RP_INTERRUPT_LIMITS. Hence we always need to compute the desired value + * ourselves, instead of doing a rmw cycle (which might result in us clearing + * all limits and the gpu stuck at whatever frequency it is at atm). + */ +static u32 gen6_rps_limits(struct drm_i915_private *dev_priv, u8 *val) { - struct drm_i915_private *dev_priv = dev->dev_private; u32 limits; limits = 0; - if (val >= dev_priv->max_delay) - val = dev_priv->max_delay; - else - limits |= dev_priv->max_delay << 24; - if (val <= dev_priv->min_delay) - val = dev_priv->min_delay; - else - limits |= dev_priv->min_delay << 16; + if (*val >= dev_priv->rps.max_delay) + *val = dev_priv->rps.max_delay; + limits |= dev_priv->rps.max_delay << 24; + + /* Only set the down limit when we've reached the lowest level to avoid + * getting more interrupts, otherwise leave this clear. This prevents a + * race in the hw when coming out of rc6: There's a tiny window where + * the hw runs at the minimal clock before selecting the desired + * frequency, if the down threshold expires in that window we will not + * receive a down interrupt. */ + if (*val <= dev_priv->rps.min_delay) { + *val = dev_priv->rps.min_delay; + limits |= dev_priv->rps.min_delay << 16; + } - if (val == dev_priv->cur_delay) + return limits; +} + +void gen6_set_rps(struct drm_device *dev, u8 val) +{ + struct drm_i915_private *dev_priv = dev->dev_private; + u32 limits = gen6_rps_limits(dev_priv, &val); + + WARN_ON(!mutex_is_locked(&dev->struct_mutex)); + WARN_ON(val > dev_priv->rps.max_delay); + WARN_ON(val < dev_priv->rps.min_delay); + + if (val == dev_priv->rps.cur_delay) return; I915_WRITE(GEN6_RPNSWREQ, @@ -2296,7 +2340,11 @@ void gen6_set_rps(struct drm_device *dev, u8 val) */ I915_WRITE(GEN6_RP_INTERRUPT_LIMITS, limits); - dev_priv->cur_delay = val; + POSTING_READ(GEN6_RPNSWREQ); + + dev_priv->rps.cur_delay = val; + + trace_intel_gpu_freq_change(val * 50); } static void gen6_disable_rps(struct drm_device *dev) @@ -2312,40 +2360,40 @@ static void gen6_disable_rps(struct drm_device *dev) * register (PMIMR) to mask PM interrupts. The only risk is in leaving * stale bits in PMIIR and PMIMR which gen6_enable_rps will clean up. */ - spin_lock_irq(&dev_priv->rps_lock); - dev_priv->pm_iir = 0; - spin_unlock_irq(&dev_priv->rps_lock); + spin_lock_irq(&dev_priv->rps.lock); + dev_priv->rps.pm_iir = 0; + spin_unlock_irq(&dev_priv->rps.lock); I915_WRITE(GEN6_PMIIR, I915_READ(GEN6_PMIIR)); } int intel_enable_rc6(const struct drm_device *dev) { - /* - * Respect the kernel parameter if it is set - */ + /* Respect the kernel parameter if it is set */ if (i915_enable_rc6 >= 0) return i915_enable_rc6; - /* - * Disable RC6 on Ironlake - */ - if (INTEL_INFO(dev)->gen == 5) - return 0; + if (INTEL_INFO(dev)->gen == 5) { +#ifdef CONFIG_INTEL_IOMMU + /* Disable rc6 on ilk if VT-d is on. */ + if (intel_iommu_gfx_mapped) + return false; +#endif + DRM_DEBUG_DRIVER("Ironlake: only RC6 available\n"); + return INTEL_RC6_ENABLE; + } - /* On Haswell, only RC6 is available. So let's enable it by default to - * provide better testing and coverage since the beginning. - */ - if (IS_HASWELL(dev)) + if (IS_HASWELL(dev)) { + DRM_DEBUG_DRIVER("Haswell: only RC6 available\n"); return INTEL_RC6_ENABLE; + } - /* - * Disable rc6 on Sandybridge - */ + /* snb/ivb have more than one rc6 state. */ if (INTEL_INFO(dev)->gen == 6) { DRM_DEBUG_DRIVER("Sandybridge: deep RC6 disabled\n"); return INTEL_RC6_ENABLE; } + DRM_DEBUG_DRIVER("RC6 and deep RC6 enabled\n"); return (INTEL_RC6_ENABLE | INTEL_RC6p_ENABLE); } @@ -2383,9 +2431,9 @@ static void gen6_enable_rps(struct drm_device *dev) gt_perf_status = I915_READ(GEN6_GT_PERF_STATUS); /* In units of 100MHz */ - dev_priv->max_delay = rp_state_cap & 0xff; - dev_priv->min_delay = (rp_state_cap & 0xff0000) >> 16; - dev_priv->cur_delay = 0; + dev_priv->rps.max_delay = rp_state_cap & 0xff; + dev_priv->rps.min_delay = (rp_state_cap & 0xff0000) >> 16; + dev_priv->rps.cur_delay = 0; /* disable the counters and set deterministic thresholds */ I915_WRITE(GEN6_RC_CONTROL, 0); @@ -2438,8 +2486,8 @@ static void gen6_enable_rps(struct drm_device *dev) I915_WRITE(GEN6_RP_DOWN_TIMEOUT, 1000000); I915_WRITE(GEN6_RP_INTERRUPT_LIMITS, - dev_priv->max_delay << 24 | - dev_priv->min_delay << 16); + dev_priv->rps.max_delay << 24 | + dev_priv->rps.min_delay << 16); I915_WRITE(GEN6_RP_UP_THRESHOLD, 59400); I915_WRITE(GEN6_RP_DOWN_THRESHOLD, 245000); @@ -2477,7 +2525,7 @@ static void gen6_enable_rps(struct drm_device *dev) 500)) DRM_ERROR("timeout waiting for pcode mailbox to finish\n"); if (pcu_mbox & (1<<31)) { /* OC supported */ - dev_priv->max_delay = pcu_mbox & 0xff; + dev_priv->rps.max_delay = pcu_mbox & 0xff; DRM_DEBUG_DRIVER("overclocking supported, adjusting frequency max to %dMHz\n", pcu_mbox * 50); } @@ -2485,10 +2533,10 @@ static void gen6_enable_rps(struct drm_device *dev) /* requires MSI enabled */ I915_WRITE(GEN6_PMIER, GEN6_PM_DEFERRED_EVENTS); - spin_lock_irq(&dev_priv->rps_lock); - WARN_ON(dev_priv->pm_iir != 0); + spin_lock_irq(&dev_priv->rps.lock); + WARN_ON(dev_priv->rps.pm_iir != 0); I915_WRITE(GEN6_PMIMR, 0); - spin_unlock_irq(&dev_priv->rps_lock); + spin_unlock_irq(&dev_priv->rps.lock); /* enable all PM interrupts */ I915_WRITE(GEN6_PMINTRMSK, 0); @@ -2520,9 +2568,9 @@ static void gen6_update_ring_freq(struct drm_device *dev) * to use for memory access. We do this by specifying the IA frequency * the PCU should use as a reference to determine the ring frequency. */ - for (gpu_freq = dev_priv->max_delay; gpu_freq >= dev_priv->min_delay; + for (gpu_freq = dev_priv->rps.max_delay; gpu_freq >= dev_priv->rps.min_delay; gpu_freq--) { - int diff = dev_priv->max_delay - gpu_freq; + int diff = dev_priv->rps.max_delay - gpu_freq; /* * For GPU frequencies less than 750MHz, just use the lowest @@ -2686,14 +2734,16 @@ static const struct cparams { { 0, 800, 231, 23784 }, }; -unsigned long i915_chipset_val(struct drm_i915_private *dev_priv) +static unsigned long __i915_chipset_val(struct drm_i915_private *dev_priv) { u64 total_count, diff, ret; u32 count1, count2, count3, m = 0, c = 0; unsigned long now = jiffies_to_msecs(jiffies), diff1; int i; - diff1 = now - dev_priv->last_time1; + assert_spin_locked(&mchdev_lock); + + diff1 = now - dev_priv->ips.last_time1; /* Prevent division-by-zero if we are asking too fast. * Also, we don't get interesting results if we are polling @@ -2701,7 +2751,7 @@ unsigned long i915_chipset_val(struct drm_i915_private *dev_priv) * in such cases. */ if (diff1 <= 10) - return dev_priv->chipset_power; + return dev_priv->ips.chipset_power; count1 = I915_READ(DMIEC); count2 = I915_READ(DDREC); @@ -2710,16 +2760,16 @@ unsigned long i915_chipset_val(struct drm_i915_private *dev_priv) total_count = count1 + count2 + count3; /* FIXME: handle per-counter overflow */ - if (total_count < dev_priv->last_count1) { - diff = ~0UL - dev_priv->last_count1; + if (total_count < dev_priv->ips.last_count1) { + diff = ~0UL - dev_priv->ips.last_count1; diff += total_count; } else { - diff = total_count - dev_priv->last_count1; + diff = total_count - dev_priv->ips.last_count1; } for (i = 0; i < ARRAY_SIZE(cparams); i++) { - if (cparams[i].i == dev_priv->c_m && - cparams[i].t == dev_priv->r_t) { + if (cparams[i].i == dev_priv->ips.c_m && + cparams[i].t == dev_priv->ips.r_t) { m = cparams[i].m; c = cparams[i].c; break; @@ -2730,14 +2780,30 @@ unsigned long i915_chipset_val(struct drm_i915_private *dev_priv) ret = ((m * diff) + c); ret = div_u64(ret, 10); - dev_priv->last_count1 = total_count; - dev_priv->last_time1 = now; + dev_priv->ips.last_count1 = total_count; + dev_priv->ips.last_time1 = now; - dev_priv->chipset_power = ret; + dev_priv->ips.chipset_power = ret; return ret; } +unsigned long i915_chipset_val(struct drm_i915_private *dev_priv) +{ + unsigned long val; + + if (dev_priv->info->gen != 5) + return 0; + + spin_lock_irq(&mchdev_lock); + + val = __i915_chipset_val(dev_priv); + + spin_unlock_irq(&mchdev_lock); + + return val; +} + unsigned long i915_mch_val(struct drm_i915_private *dev_priv) { unsigned long m, x, b; @@ -2894,18 +2960,17 @@ static u16 pvid_to_extvid(struct drm_i915_private *dev_priv, u8 pxvid) return v_table[pxvid].vd; } -void i915_update_gfx_val(struct drm_i915_private *dev_priv) +static void __i915_update_gfx_val(struct drm_i915_private *dev_priv) { struct timespec now, diff1; u64 diff; unsigned long diffms; u32 count; - if (dev_priv->info->gen != 5) - return; + assert_spin_locked(&mchdev_lock); getrawmonotonic(&now); - diff1 = timespec_sub(now, dev_priv->last_time2); + diff1 = timespec_sub(now, dev_priv->ips.last_time2); /* Don't divide by 0 */ diffms = diff1.tv_sec * 1000 + diff1.tv_nsec / 1000000; @@ -2914,28 +2979,42 @@ void i915_update_gfx_val(struct drm_i915_private *dev_priv) count = I915_READ(GFXEC); - if (count < dev_priv->last_count2) { - diff = ~0UL - dev_priv->last_count2; + if (count < dev_priv->ips.last_count2) { + diff = ~0UL - dev_priv->ips.last_count2; diff += count; } else { - diff = count - dev_priv->last_count2; + diff = count - dev_priv->ips.last_count2; } - dev_priv->last_count2 = count; - dev_priv->last_time2 = now; + dev_priv->ips.last_count2 = count; + dev_priv->ips.last_time2 = now; /* More magic constants... */ diff = diff * 1181; diff = div_u64(diff, diffms * 10); - dev_priv->gfx_power = diff; + dev_priv->ips.gfx_power = diff; } -unsigned long i915_gfx_val(struct drm_i915_private *dev_priv) +void i915_update_gfx_val(struct drm_i915_private *dev_priv) +{ + if (dev_priv->info->gen != 5) + return; + + spin_lock_irq(&mchdev_lock); + + __i915_update_gfx_val(dev_priv); + + spin_unlock_irq(&mchdev_lock); +} + +static unsigned long __i915_gfx_val(struct drm_i915_private *dev_priv) { unsigned long t, corr, state1, corr2, state2; u32 pxvid, ext_v; - pxvid = I915_READ(PXVFREQ_BASE + (dev_priv->cur_delay * 4)); + assert_spin_locked(&mchdev_lock); + + pxvid = I915_READ(PXVFREQ_BASE + (dev_priv->rps.cur_delay * 4)); pxvid = (pxvid >> 24) & 0x7f; ext_v = pvid_to_extvid(dev_priv, pxvid); @@ -2955,27 +3034,31 @@ unsigned long i915_gfx_val(struct drm_i915_private *dev_priv) corr = corr * ((150142 * state1) / 10000 - 78642); corr /= 100000; - corr2 = (corr * dev_priv->corr); + corr2 = (corr * dev_priv->ips.corr); state2 = (corr2 * state1) / 10000; state2 /= 100; /* convert to mW */ - i915_update_gfx_val(dev_priv); + __i915_update_gfx_val(dev_priv); - return dev_priv->gfx_power + state2; + return dev_priv->ips.gfx_power + state2; } -/* Global for IPS driver to get at the current i915 device */ -static struct drm_i915_private *i915_mch_dev; -/* - * Lock protecting IPS related data structures - * - i915_mch_dev - * - dev_priv->max_delay - * - dev_priv->min_delay - * - dev_priv->fmax - * - dev_priv->gpu_busy - */ -static DEFINE_SPINLOCK(mchdev_lock); +unsigned long i915_gfx_val(struct drm_i915_private *dev_priv) +{ + unsigned long val; + + if (dev_priv->info->gen != 5) + return 0; + + spin_lock_irq(&mchdev_lock); + + val = __i915_gfx_val(dev_priv); + + spin_unlock_irq(&mchdev_lock); + + return val; +} /** * i915_read_mch_val - return value for IPS use @@ -2988,18 +3071,18 @@ unsigned long i915_read_mch_val(void) struct drm_i915_private *dev_priv; unsigned long chipset_val, graphics_val, ret = 0; - spin_lock(&mchdev_lock); + spin_lock_irq(&mchdev_lock); if (!i915_mch_dev) goto out_unlock; dev_priv = i915_mch_dev; - chipset_val = i915_chipset_val(dev_priv); - graphics_val = i915_gfx_val(dev_priv); + chipset_val = __i915_chipset_val(dev_priv); + graphics_val = __i915_gfx_val(dev_priv); ret = chipset_val + graphics_val; out_unlock: - spin_unlock(&mchdev_lock); + spin_unlock_irq(&mchdev_lock); return ret; } @@ -3015,18 +3098,18 @@ bool i915_gpu_raise(void) struct drm_i915_private *dev_priv; bool ret = true; - spin_lock(&mchdev_lock); + spin_lock_irq(&mchdev_lock); if (!i915_mch_dev) { ret = false; goto out_unlock; } dev_priv = i915_mch_dev; - if (dev_priv->max_delay > dev_priv->fmax) - dev_priv->max_delay--; + if (dev_priv->ips.max_delay > dev_priv->ips.fmax) + dev_priv->ips.max_delay--; out_unlock: - spin_unlock(&mchdev_lock); + spin_unlock_irq(&mchdev_lock); return ret; } @@ -3043,18 +3126,18 @@ bool i915_gpu_lower(void) struct drm_i915_private *dev_priv; bool ret = true; - spin_lock(&mchdev_lock); + spin_lock_irq(&mchdev_lock); if (!i915_mch_dev) { ret = false; goto out_unlock; } dev_priv = i915_mch_dev; - if (dev_priv->max_delay < dev_priv->min_delay) - dev_priv->max_delay++; + if (dev_priv->ips.max_delay < dev_priv->ips.min_delay) + dev_priv->ips.max_delay++; out_unlock: - spin_unlock(&mchdev_lock); + spin_unlock_irq(&mchdev_lock); return ret; } @@ -3068,17 +3151,20 @@ EXPORT_SYMBOL_GPL(i915_gpu_lower); bool i915_gpu_busy(void) { struct drm_i915_private *dev_priv; + struct intel_ring_buffer *ring; bool ret = false; + int i; - spin_lock(&mchdev_lock); + spin_lock_irq(&mchdev_lock); if (!i915_mch_dev) goto out_unlock; dev_priv = i915_mch_dev; - ret = dev_priv->busy; + for_each_ring(ring, dev_priv, i) + ret |= !list_empty(&ring->request_list); out_unlock: - spin_unlock(&mchdev_lock); + spin_unlock_irq(&mchdev_lock); return ret; } @@ -3095,20 +3181,20 @@ bool i915_gpu_turbo_disable(void) struct drm_i915_private *dev_priv; bool ret = true; - spin_lock(&mchdev_lock); + spin_lock_irq(&mchdev_lock); if (!i915_mch_dev) { ret = false; goto out_unlock; } dev_priv = i915_mch_dev; - dev_priv->max_delay = dev_priv->fstart; + dev_priv->ips.max_delay = dev_priv->ips.fstart; - if (!ironlake_set_drps(dev_priv->dev, dev_priv->fstart)) + if (!ironlake_set_drps(dev_priv->dev, dev_priv->ips.fstart)) ret = false; out_unlock: - spin_unlock(&mchdev_lock); + spin_unlock_irq(&mchdev_lock); return ret; } @@ -3136,19 +3222,20 @@ ips_ping_for_i915_load(void) void intel_gpu_ips_init(struct drm_i915_private *dev_priv) { - spin_lock(&mchdev_lock); + /* We only register the i915 ips part with intel-ips once everything is + * set up, to avoid intel-ips sneaking in and reading bogus values. */ + spin_lock_irq(&mchdev_lock); i915_mch_dev = dev_priv; - dev_priv->mchdev_lock = &mchdev_lock; - spin_unlock(&mchdev_lock); + spin_unlock_irq(&mchdev_lock); ips_ping_for_i915_load(); } void intel_gpu_ips_teardown(void) { - spin_lock(&mchdev_lock); + spin_lock_irq(&mchdev_lock); i915_mch_dev = NULL; - spin_unlock(&mchdev_lock); + spin_unlock_irq(&mchdev_lock); } static void intel_init_emon(struct drm_device *dev) { @@ -3218,7 +3305,7 @@ static void intel_init_emon(struct drm_device *dev) lcfuse = I915_READ(LCFUSE02); - dev_priv->corr = (lcfuse & LCFUSE_HIV_MASK); + dev_priv->ips.corr = (lcfuse & LCFUSE_HIV_MASK); } void intel_disable_gt_powersave(struct drm_device *dev) @@ -3731,42 +3818,6 @@ void intel_init_clock_gating(struct drm_device *dev) dev_priv->display.init_pch_clock_gating(dev); } -static void gen6_sanitize_pm(struct drm_device *dev) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - u32 limits, delay, old; - - gen6_gt_force_wake_get(dev_priv); - - old = limits = I915_READ(GEN6_RP_INTERRUPT_LIMITS); - /* Make sure we continue to get interrupts - * until we hit the minimum or maximum frequencies. - */ - limits &= ~(0x3f << 16 | 0x3f << 24); - delay = dev_priv->cur_delay; - if (delay < dev_priv->max_delay) - limits |= (dev_priv->max_delay & 0x3f) << 24; - if (delay > dev_priv->min_delay) - limits |= (dev_priv->min_delay & 0x3f) << 16; - - if (old != limits) { - /* Note that the known failure case is to read back 0. */ - DRM_DEBUG_DRIVER("Power management discrepancy: GEN6_RP_INTERRUPT_LIMITS " - "expected %08x, was %08x\n", limits, old); - I915_WRITE(GEN6_RP_INTERRUPT_LIMITS, limits); - } - - gen6_gt_force_wake_put(dev_priv); -} - -void intel_sanitize_pm(struct drm_device *dev) -{ - struct drm_i915_private *dev_priv = dev->dev_private; - - if (dev_priv->display.sanitize_pm) - dev_priv->display.sanitize_pm(dev); -} - /* Starting with Haswell, we have different power wells for * different parts of the GPU. This attempts to enable them all. */ @@ -3852,7 +3903,6 @@ void intel_init_pm(struct drm_device *dev) dev_priv->display.update_wm = NULL; } dev_priv->display.init_clock_gating = gen6_init_clock_gating; - dev_priv->display.sanitize_pm = gen6_sanitize_pm; } else if (IS_IVYBRIDGE(dev)) { /* FIXME: detect B0+ stepping and use auto training */ if (SNB_READ_WM0_LATENCY()) { @@ -3864,7 +3914,6 @@ void intel_init_pm(struct drm_device *dev) dev_priv->display.update_wm = NULL; } dev_priv->display.init_clock_gating = ivybridge_init_clock_gating; - dev_priv->display.sanitize_pm = gen6_sanitize_pm; } else if (IS_HASWELL(dev)) { if (SNB_READ_WM0_LATENCY()) { dev_priv->display.update_wm = sandybridge_update_wm; @@ -3876,7 +3925,6 @@ void intel_init_pm(struct drm_device *dev) dev_priv->display.update_wm = NULL; } dev_priv->display.init_clock_gating = haswell_init_clock_gating; - dev_priv->display.sanitize_pm = gen6_sanitize_pm; } else dev_priv->display.update_wm = NULL; } else if (IS_VALLEYVIEW(dev)) { @@ -3955,14 +4003,16 @@ static void __gen6_gt_force_wake_get(struct drm_i915_private *dev_priv) else forcewake_ack = FORCEWAKE_ACK; - if (wait_for_atomic_us((I915_READ_NOTRACE(forcewake_ack) & 1) == 0, 500)) - DRM_ERROR("Force wake wait timed out\n"); + if (wait_for_atomic((I915_READ_NOTRACE(forcewake_ack) & 1) == 0, + FORCEWAKE_ACK_TIMEOUT_MS)) + DRM_ERROR("Timed out waiting for forcewake old ack to clear.\n"); I915_WRITE_NOTRACE(FORCEWAKE, 1); - POSTING_READ(FORCEWAKE); + POSTING_READ(ECOBUS); /* something from same cacheline, but !FORCEWAKE */ - if (wait_for_atomic_us((I915_READ_NOTRACE(forcewake_ack) & 1), 500)) - DRM_ERROR("Force wake wait timed out\n"); + if (wait_for_atomic((I915_READ_NOTRACE(forcewake_ack) & 1), + FORCEWAKE_ACK_TIMEOUT_MS)) + DRM_ERROR("Timed out waiting for forcewake to ack request.\n"); __gen6_gt_wait_for_thread_c0(dev_priv); } @@ -3976,14 +4026,16 @@ static void __gen6_gt_force_wake_mt_get(struct drm_i915_private *dev_priv) else forcewake_ack = FORCEWAKE_MT_ACK; - if (wait_for_atomic_us((I915_READ_NOTRACE(forcewake_ack) & 1) == 0, 500)) - DRM_ERROR("Force wake wait timed out\n"); + if (wait_for_atomic((I915_READ_NOTRACE(forcewake_ack) & 1) == 0, + FORCEWAKE_ACK_TIMEOUT_MS)) + DRM_ERROR("Timed out waiting for forcewake old ack to clear.\n"); I915_WRITE_NOTRACE(FORCEWAKE_MT, _MASKED_BIT_ENABLE(1)); - POSTING_READ(FORCEWAKE_MT); + POSTING_READ(ECOBUS); /* something from same cacheline, but !FORCEWAKE */ - if (wait_for_atomic_us((I915_READ_NOTRACE(forcewake_ack) & 1), 500)) - DRM_ERROR("Force wake wait timed out\n"); + if (wait_for_atomic((I915_READ_NOTRACE(forcewake_ack) & 1), + FORCEWAKE_ACK_TIMEOUT_MS)) + DRM_ERROR("Timed out waiting for forcewake to ack request.\n"); __gen6_gt_wait_for_thread_c0(dev_priv); } @@ -4016,14 +4068,14 @@ void gen6_gt_check_fifodbg(struct drm_i915_private *dev_priv) static void __gen6_gt_force_wake_put(struct drm_i915_private *dev_priv) { I915_WRITE_NOTRACE(FORCEWAKE, 0); - POSTING_READ(FORCEWAKE); + /* gen6_gt_check_fifodbg doubles as the POSTING_READ */ gen6_gt_check_fifodbg(dev_priv); } static void __gen6_gt_force_wake_mt_put(struct drm_i915_private *dev_priv) { I915_WRITE_NOTRACE(FORCEWAKE_MT, _MASKED_BIT_DISABLE(1)); - POSTING_READ(FORCEWAKE_MT); + /* gen6_gt_check_fifodbg doubles as the POSTING_READ */ gen6_gt_check_fifodbg(dev_priv); } @@ -4062,24 +4114,24 @@ int __gen6_gt_wait_for_fifo(struct drm_i915_private *dev_priv) static void vlv_force_wake_get(struct drm_i915_private *dev_priv) { - /* Already awake? */ - if ((I915_READ(0x130094) & 0xa1) == 0xa1) - return; + if (wait_for_atomic((I915_READ_NOTRACE(FORCEWAKE_ACK_VLV) & 1) == 0, + FORCEWAKE_ACK_TIMEOUT_MS)) + DRM_ERROR("Timed out waiting for forcewake old ack to clear.\n"); - I915_WRITE_NOTRACE(FORCEWAKE_VLV, 0xffffffff); - POSTING_READ(FORCEWAKE_VLV); + I915_WRITE_NOTRACE(FORCEWAKE_VLV, _MASKED_BIT_ENABLE(1)); - if (wait_for_atomic_us((I915_READ_NOTRACE(FORCEWAKE_ACK_VLV) & 1), 500)) - DRM_ERROR("Force wake wait timed out\n"); + if (wait_for_atomic((I915_READ_NOTRACE(FORCEWAKE_ACK_VLV) & 1), + FORCEWAKE_ACK_TIMEOUT_MS)) + DRM_ERROR("Timed out waiting for forcewake to ack request.\n"); __gen6_gt_wait_for_thread_c0(dev_priv); } static void vlv_force_wake_put(struct drm_i915_private *dev_priv) { - I915_WRITE_NOTRACE(FORCEWAKE_VLV, 0xffff0000); - /* FIXME: confirm VLV behavior with Punit folks */ - POSTING_READ(FORCEWAKE_VLV); + I915_WRITE_NOTRACE(FORCEWAKE_VLV, _MASKED_BIT_DISABLE(1)); + /* The below doubles as a POSTING_READ */ + gen6_gt_check_fifodbg(dev_priv); } void intel_gt_init(struct drm_device *dev) diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.c b/drivers/gpu/drm/i915/intel_ringbuffer.c index e2a73b38abe9..984a0c5fbf5d 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.c +++ b/drivers/gpu/drm/i915/intel_ringbuffer.c @@ -262,6 +262,83 @@ gen6_render_ring_flush(struct intel_ring_buffer *ring, return 0; } +static int +gen7_render_ring_cs_stall_wa(struct intel_ring_buffer *ring) +{ + int ret; + + ret = intel_ring_begin(ring, 4); + if (ret) + return ret; + + intel_ring_emit(ring, GFX_OP_PIPE_CONTROL(4)); + intel_ring_emit(ring, PIPE_CONTROL_CS_STALL | + PIPE_CONTROL_STALL_AT_SCOREBOARD); + intel_ring_emit(ring, 0); + intel_ring_emit(ring, 0); + intel_ring_advance(ring); + + return 0; +} + +static int +gen7_render_ring_flush(struct intel_ring_buffer *ring, + u32 invalidate_domains, u32 flush_domains) +{ + u32 flags = 0; + struct pipe_control *pc = ring->private; + u32 scratch_addr = pc->gtt_offset + 128; + int ret; + + /* + * Ensure that any following seqno writes only happen when the render + * cache is indeed flushed. + * + * Workaround: 4th PIPE_CONTROL command (except the ones with only + * read-cache invalidate bits set) must have the CS_STALL bit set. We + * don't try to be clever and just set it unconditionally. + */ + flags |= PIPE_CONTROL_CS_STALL; + + /* Just flush everything. Experiments have shown that reducing the + * number of bits based on the write domains has little performance + * impact. + */ + if (flush_domains) { + flags |= PIPE_CONTROL_RENDER_TARGET_CACHE_FLUSH; + flags |= PIPE_CONTROL_DEPTH_CACHE_FLUSH; + } + if (invalidate_domains) { + flags |= PIPE_CONTROL_TLB_INVALIDATE; + flags |= PIPE_CONTROL_INSTRUCTION_CACHE_INVALIDATE; + flags |= PIPE_CONTROL_TEXTURE_CACHE_INVALIDATE; + flags |= PIPE_CONTROL_VF_CACHE_INVALIDATE; + flags |= PIPE_CONTROL_CONST_CACHE_INVALIDATE; + flags |= PIPE_CONTROL_STATE_CACHE_INVALIDATE; + /* + * TLB invalidate requires a post-sync write. + */ + flags |= PIPE_CONTROL_QW_WRITE; + + /* Workaround: we must issue a pipe_control with CS-stall bit + * set before a pipe_control command that has the state cache + * invalidate bit set. */ + gen7_render_ring_cs_stall_wa(ring); + } + + ret = intel_ring_begin(ring, 4); + if (ret) + return ret; + + intel_ring_emit(ring, GFX_OP_PIPE_CONTROL(4)); + intel_ring_emit(ring, flags); + intel_ring_emit(ring, scratch_addr | PIPE_CONTROL_GLOBAL_GTT); + intel_ring_emit(ring, 0); + intel_ring_advance(ring); + + return 0; +} + static void ring_write_tail(struct intel_ring_buffer *ring, u32 value) { @@ -382,12 +459,12 @@ init_pipe_control(struct intel_ring_buffer *ring) i915_gem_object_set_cache_level(obj, I915_CACHE_LLC); - ret = i915_gem_object_pin(obj, 4096, true); + ret = i915_gem_object_pin(obj, 4096, true, false); if (ret) goto err_unref; pc->gtt_offset = obj->gtt_offset; - pc->cpu_page = kmap(obj->pages[0]); + pc->cpu_page = kmap(sg_page(obj->pages->sgl)); if (pc->cpu_page == NULL) goto err_unpin; @@ -414,7 +491,8 @@ cleanup_pipe_control(struct intel_ring_buffer *ring) return; obj = pc->obj; - kunmap(obj->pages[0]); + + kunmap(sg_page(obj->pages->sgl)); i915_gem_object_unpin(obj); drm_gem_object_unreference(&obj->base); @@ -462,7 +540,7 @@ static int init_render_ring(struct intel_ring_buffer *ring) if (INTEL_INFO(dev)->gen >= 6) I915_WRITE(INSTPM, _MASKED_BIT_ENABLE(INSTPM_FORCE_ORDERING)); - if (IS_IVYBRIDGE(dev)) + if (HAS_L3_GPU_CACHE(dev)) I915_WRITE_IMR(ring, ~GEN6_RENDER_L3_PARITY_ERROR); return ret; @@ -628,26 +706,24 @@ pc_render_add_request(struct intel_ring_buffer *ring, } static u32 -gen6_ring_get_seqno(struct intel_ring_buffer *ring) +gen6_ring_get_seqno(struct intel_ring_buffer *ring, bool lazy_coherency) { - struct drm_device *dev = ring->dev; - /* Workaround to force correct ordering between irq and seqno writes on * ivb (and maybe also on snb) by reading from a CS register (like * ACTHD) before reading the status page. */ - if (IS_GEN6(dev) || IS_GEN7(dev)) + if (!lazy_coherency) intel_ring_get_active_head(ring); return intel_read_status_page(ring, I915_GEM_HWS_INDEX); } static u32 -ring_get_seqno(struct intel_ring_buffer *ring) +ring_get_seqno(struct intel_ring_buffer *ring, bool lazy_coherency) { return intel_read_status_page(ring, I915_GEM_HWS_INDEX); } static u32 -pc_render_get_seqno(struct intel_ring_buffer *ring) +pc_render_get_seqno(struct intel_ring_buffer *ring, bool lazy_coherency) { struct pipe_control *pc = ring->private; return pc->cpu_page[0]; @@ -852,7 +928,7 @@ gen6_ring_get_irq(struct intel_ring_buffer *ring) spin_lock_irqsave(&dev_priv->irq_lock, flags); if (ring->irq_refcount++ == 0) { - if (IS_IVYBRIDGE(dev) && ring->id == RCS) + if (HAS_L3_GPU_CACHE(dev) && ring->id == RCS) I915_WRITE_IMR(ring, ~(ring->irq_enable_mask | GEN6_RENDER_L3_PARITY_ERROR)); else @@ -875,7 +951,7 @@ gen6_ring_put_irq(struct intel_ring_buffer *ring) spin_lock_irqsave(&dev_priv->irq_lock, flags); if (--ring->irq_refcount == 0) { - if (IS_IVYBRIDGE(dev) && ring->id == RCS) + if (HAS_L3_GPU_CACHE(dev) && ring->id == RCS) I915_WRITE_IMR(ring, ~GEN6_RENDER_L3_PARITY_ERROR); else I915_WRITE_IMR(ring, ~0); @@ -951,7 +1027,7 @@ static void cleanup_status_page(struct intel_ring_buffer *ring) if (obj == NULL) return; - kunmap(obj->pages[0]); + kunmap(sg_page(obj->pages->sgl)); i915_gem_object_unpin(obj); drm_gem_object_unreference(&obj->base); ring->status_page.obj = NULL; @@ -972,13 +1048,13 @@ static int init_status_page(struct intel_ring_buffer *ring) i915_gem_object_set_cache_level(obj, I915_CACHE_LLC); - ret = i915_gem_object_pin(obj, 4096, true); + ret = i915_gem_object_pin(obj, 4096, true, false); if (ret != 0) { goto err_unref; } ring->status_page.gfx_addr = obj->gtt_offset; - ring->status_page.page_addr = kmap(obj->pages[0]); + ring->status_page.page_addr = kmap(sg_page(obj->pages->sgl)); if (ring->status_page.page_addr == NULL) { ret = -ENOMEM; goto err_unpin; @@ -1010,7 +1086,6 @@ static int intel_init_ring_buffer(struct drm_device *dev, ring->dev = dev; INIT_LIST_HEAD(&ring->active_list); INIT_LIST_HEAD(&ring->request_list); - INIT_LIST_HEAD(&ring->gpu_write_list); ring->size = 32 * PAGE_SIZE; init_waitqueue_head(&ring->irq_queue); @@ -1030,7 +1105,7 @@ static int intel_init_ring_buffer(struct drm_device *dev, ring->obj = obj; - ret = i915_gem_object_pin(obj, PAGE_SIZE, true); + ret = i915_gem_object_pin(obj, PAGE_SIZE, true, false); if (ret) goto err_unref; @@ -1379,7 +1454,9 @@ int intel_init_render_ring_buffer(struct drm_device *dev) if (INTEL_INFO(dev)->gen >= 6) { ring->add_request = gen6_add_request; - ring->flush = gen6_render_ring_flush; + ring->flush = gen7_render_ring_flush; + if (INTEL_INFO(dev)->gen == 6) + ring->flush = gen6_render_ring_flush; ring->irq_get = gen6_ring_get_irq; ring->irq_put = gen6_ring_put_irq; ring->irq_enable_mask = GT_USER_INTERRUPT; @@ -1481,7 +1558,6 @@ int intel_render_ring_init_dri(struct drm_device *dev, u64 start, u32 size) ring->dev = dev; INIT_LIST_HEAD(&ring->active_list); INIT_LIST_HEAD(&ring->request_list); - INIT_LIST_HEAD(&ring->gpu_write_list); ring->size = size; ring->effective_size = ring->size; @@ -1574,3 +1650,41 @@ int intel_init_blt_ring_buffer(struct drm_device *dev) return intel_init_ring_buffer(dev, ring); } + +int +intel_ring_flush_all_caches(struct intel_ring_buffer *ring) +{ + int ret; + + if (!ring->gpu_caches_dirty) + return 0; + + ret = ring->flush(ring, 0, I915_GEM_GPU_DOMAINS); + if (ret) + return ret; + + trace_i915_gem_ring_flush(ring, 0, I915_GEM_GPU_DOMAINS); + + ring->gpu_caches_dirty = false; + return 0; +} + +int +intel_ring_invalidate_all_caches(struct intel_ring_buffer *ring) +{ + uint32_t flush_domains; + int ret; + + flush_domains = 0; + if (ring->gpu_caches_dirty) + flush_domains = I915_GEM_GPU_DOMAINS; + + ret = ring->flush(ring, I915_GEM_GPU_DOMAINS, flush_domains); + if (ret) + return ret; + + trace_i915_gem_ring_flush(ring, I915_GEM_GPU_DOMAINS, flush_domains); + + ring->gpu_caches_dirty = false; + return 0; +} diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.h b/drivers/gpu/drm/i915/intel_ringbuffer.h index 1d3c81fdad92..2ea7a311a1f0 100644 --- a/drivers/gpu/drm/i915/intel_ringbuffer.h +++ b/drivers/gpu/drm/i915/intel_ringbuffer.h @@ -72,7 +72,14 @@ struct intel_ring_buffer { u32 flush_domains); int (*add_request)(struct intel_ring_buffer *ring, u32 *seqno); - u32 (*get_seqno)(struct intel_ring_buffer *ring); + /* Some chipsets are not quite as coherent as advertised and need + * an expensive kick to force a true read of the up-to-date seqno. + * However, the up-to-date seqno is not always required and the last + * seen value is good enough. Note that the seqno will always be + * monotonic, even if not coherent. + */ + u32 (*get_seqno)(struct intel_ring_buffer *ring, + bool lazy_coherency); int (*dispatch_execbuffer)(struct intel_ring_buffer *ring, u32 offset, u32 length); void (*cleanup)(struct intel_ring_buffer *ring); @@ -101,15 +108,6 @@ struct intel_ring_buffer { struct list_head request_list; /** - * List of objects currently pending a GPU write flush. - * - * All elements on this list will belong to either the - * active_list or flushing_list, last_rendering_seqno can - * be used to differentiate between the two elements. - */ - struct list_head gpu_write_list; - - /** * Do we have some not yet emitted requests outstanding? */ u32 outstanding_lazy_request; @@ -204,6 +202,8 @@ static inline void intel_ring_emit(struct intel_ring_buffer *ring, void intel_ring_advance(struct intel_ring_buffer *ring); u32 intel_ring_get_seqno(struct intel_ring_buffer *ring); +int intel_ring_flush_all_caches(struct intel_ring_buffer *ring); +int intel_ring_invalidate_all_caches(struct intel_ring_buffer *ring); int intel_init_render_ring_buffer(struct drm_device *dev); int intel_init_bsd_ring_buffer(struct drm_device *dev); diff --git a/drivers/gpu/drm/i915/intel_sdvo.c b/drivers/gpu/drm/i915/intel_sdvo.c index 123afd357611..39c319827f91 100644 --- a/drivers/gpu/drm/i915/intel_sdvo.c +++ b/drivers/gpu/drm/i915/intel_sdvo.c @@ -97,7 +97,7 @@ struct intel_sdvo { /* * Hotplug activation bits for this device */ - uint8_t hotplug_active[2]; + uint16_t hotplug_active; /** * This is used to select the color range of RBG outputs in HDMI mode. @@ -628,6 +628,14 @@ static bool intel_sdvo_set_active_outputs(struct intel_sdvo *intel_sdvo, &outputs, sizeof(outputs)); } +static bool intel_sdvo_get_active_outputs(struct intel_sdvo *intel_sdvo, + u16 *outputs) +{ + return intel_sdvo_get_value(intel_sdvo, + SDVO_CMD_GET_ACTIVE_OUTPUTS, + outputs, sizeof(*outputs)); +} + static bool intel_sdvo_set_encoder_power_state(struct intel_sdvo *intel_sdvo, int mode) { @@ -1142,51 +1150,132 @@ static void intel_sdvo_mode_set(struct drm_encoder *encoder, intel_sdvo_write_sdvox(intel_sdvo, sdvox); } -static void intel_sdvo_dpms(struct drm_encoder *encoder, int mode) +static bool intel_sdvo_connector_get_hw_state(struct intel_connector *connector) { - struct drm_device *dev = encoder->dev; + struct intel_sdvo_connector *intel_sdvo_connector = + to_intel_sdvo_connector(&connector->base); + struct intel_sdvo *intel_sdvo = intel_attached_sdvo(&connector->base); + u16 active_outputs; + + intel_sdvo_get_active_outputs(intel_sdvo, &active_outputs); + + if (active_outputs & intel_sdvo_connector->output_flag) + return true; + else + return false; +} + +static bool intel_sdvo_get_hw_state(struct intel_encoder *encoder, + enum pipe *pipe) +{ + struct drm_device *dev = encoder->base.dev; struct drm_i915_private *dev_priv = dev->dev_private; - struct intel_sdvo *intel_sdvo = to_intel_sdvo(encoder); - struct intel_crtc *intel_crtc = to_intel_crtc(encoder->crtc); + struct intel_sdvo *intel_sdvo = to_intel_sdvo(&encoder->base); + u32 tmp; + + tmp = I915_READ(intel_sdvo->sdvo_reg); + + if (!(tmp & SDVO_ENABLE)) + return false; + + if (HAS_PCH_CPT(dev)) + *pipe = PORT_TO_PIPE_CPT(tmp); + else + *pipe = PORT_TO_PIPE(tmp); + + return true; +} + +static void intel_disable_sdvo(struct intel_encoder *encoder) +{ + struct drm_i915_private *dev_priv = encoder->base.dev->dev_private; + struct intel_sdvo *intel_sdvo = to_intel_sdvo(&encoder->base); + u32 temp; + + intel_sdvo_set_active_outputs(intel_sdvo, 0); + if (0) + intel_sdvo_set_encoder_power_state(intel_sdvo, + DRM_MODE_DPMS_OFF); + + temp = I915_READ(intel_sdvo->sdvo_reg); + if ((temp & SDVO_ENABLE) != 0) { + intel_sdvo_write_sdvox(intel_sdvo, temp & ~SDVO_ENABLE); + } +} + +static void intel_enable_sdvo(struct intel_encoder *encoder) +{ + struct drm_device *dev = encoder->base.dev; + struct drm_i915_private *dev_priv = dev->dev_private; + struct intel_sdvo *intel_sdvo = to_intel_sdvo(&encoder->base); + struct intel_crtc *intel_crtc = to_intel_crtc(encoder->base.crtc); u32 temp; + bool input1, input2; + int i; + u8 status; + + temp = I915_READ(intel_sdvo->sdvo_reg); + if ((temp & SDVO_ENABLE) == 0) + intel_sdvo_write_sdvox(intel_sdvo, temp | SDVO_ENABLE); + for (i = 0; i < 2; i++) + intel_wait_for_vblank(dev, intel_crtc->pipe); + + status = intel_sdvo_get_trained_inputs(intel_sdvo, &input1, &input2); + /* Warn if the device reported failure to sync. + * A lot of SDVO devices fail to notify of sync, but it's + * a given it the status is a success, we succeeded. + */ + if (status == SDVO_CMD_STATUS_SUCCESS && !input1) { + DRM_DEBUG_KMS("First %s output reported failure to " + "sync\n", SDVO_NAME(intel_sdvo)); + } + + if (0) + intel_sdvo_set_encoder_power_state(intel_sdvo, + DRM_MODE_DPMS_ON); + intel_sdvo_set_active_outputs(intel_sdvo, intel_sdvo->attached_output); +} + +static void intel_sdvo_dpms(struct drm_connector *connector, int mode) +{ + struct drm_crtc *crtc; + struct intel_sdvo *intel_sdvo = intel_attached_sdvo(connector); + + /* dvo supports only 2 dpms states. */ + if (mode != DRM_MODE_DPMS_ON) + mode = DRM_MODE_DPMS_OFF; + + if (mode == connector->dpms) + return; + + connector->dpms = mode; + + /* Only need to change hw state when actually enabled */ + crtc = intel_sdvo->base.base.crtc; + if (!crtc) { + intel_sdvo->base.connectors_active = false; + return; + } if (mode != DRM_MODE_DPMS_ON) { intel_sdvo_set_active_outputs(intel_sdvo, 0); if (0) intel_sdvo_set_encoder_power_state(intel_sdvo, mode); - if (mode == DRM_MODE_DPMS_OFF) { - temp = I915_READ(intel_sdvo->sdvo_reg); - if ((temp & SDVO_ENABLE) != 0) { - intel_sdvo_write_sdvox(intel_sdvo, temp & ~SDVO_ENABLE); - } - } + intel_sdvo->base.connectors_active = false; + + intel_crtc_update_dpms(crtc); } else { - bool input1, input2; - int i; - u8 status; - - temp = I915_READ(intel_sdvo->sdvo_reg); - if ((temp & SDVO_ENABLE) == 0) - intel_sdvo_write_sdvox(intel_sdvo, temp | SDVO_ENABLE); - for (i = 0; i < 2; i++) - intel_wait_for_vblank(dev, intel_crtc->pipe); - - status = intel_sdvo_get_trained_inputs(intel_sdvo, &input1, &input2); - /* Warn if the device reported failure to sync. - * A lot of SDVO devices fail to notify of sync, but it's - * a given it the status is a success, we succeeded. - */ - if (status == SDVO_CMD_STATUS_SUCCESS && !input1) { - DRM_DEBUG_KMS("First %s output reported failure to " - "sync\n", SDVO_NAME(intel_sdvo)); - } + intel_sdvo->base.connectors_active = true; + + intel_crtc_update_dpms(crtc); if (0) intel_sdvo_set_encoder_power_state(intel_sdvo, mode); intel_sdvo_set_active_outputs(intel_sdvo, intel_sdvo->attached_output); } - return; + + intel_modeset_check_state(connector->dev); } static int intel_sdvo_mode_valid(struct drm_connector *connector, @@ -1251,25 +1340,29 @@ static bool intel_sdvo_get_capabilities(struct intel_sdvo *intel_sdvo, struct in return true; } -static int intel_sdvo_supports_hotplug(struct intel_sdvo *intel_sdvo) +static uint16_t intel_sdvo_get_hotplug_support(struct intel_sdvo *intel_sdvo) { struct drm_device *dev = intel_sdvo->base.base.dev; - u8 response[2]; + uint16_t hotplug; /* HW Erratum: SDVO Hotplug is broken on all i945G chips, there's noise * on the line. */ if (IS_I945G(dev) || IS_I945GM(dev)) - return false; + return 0; - return intel_sdvo_get_value(intel_sdvo, SDVO_CMD_GET_HOT_PLUG_SUPPORT, - &response, 2) && response[0]; + if (!intel_sdvo_get_value(intel_sdvo, SDVO_CMD_GET_HOT_PLUG_SUPPORT, + &hotplug, sizeof(hotplug))) + return 0; + + return hotplug; } static void intel_sdvo_enable_hotplug(struct intel_encoder *encoder) { struct intel_sdvo *intel_sdvo = to_intel_sdvo(&encoder->base); - intel_sdvo_write_cmd(intel_sdvo, SDVO_CMD_SET_ACTIVE_HOT_PLUG, &intel_sdvo->hotplug_active, 2); + intel_sdvo_write_cmd(intel_sdvo, SDVO_CMD_SET_ACTIVE_HOT_PLUG, + &intel_sdvo->hotplug_active, 2); } static bool @@ -1345,7 +1438,6 @@ intel_sdvo_tmds_sink_detect(struct drm_connector *connector) } } else status = connector_status_disconnected; - connector->display_info.raw_edid = NULL; kfree(edid); } @@ -1419,7 +1511,6 @@ intel_sdvo_detect(struct drm_connector *connector, bool force) else ret = connector_status_disconnected; - connector->display_info.raw_edid = NULL; kfree(edid); } else ret = connector_status_connected; @@ -1465,7 +1556,6 @@ static void intel_sdvo_get_ddc_modes(struct drm_connector *connector) drm_add_edid_modes(connector, edid); } - connector->display_info.raw_edid = NULL; kfree(edid); } } @@ -1837,8 +1927,8 @@ set_value: done: if (intel_sdvo->base.base.crtc) { struct drm_crtc *crtc = intel_sdvo->base.base.crtc; - drm_crtc_helper_set_mode(crtc, &crtc->mode, crtc->x, - crtc->y, crtc->fb); + intel_set_mode(crtc, &crtc->mode, + crtc->x, crtc->y, crtc->fb); } return 0; @@ -1846,15 +1936,13 @@ done: } static const struct drm_encoder_helper_funcs intel_sdvo_helper_funcs = { - .dpms = intel_sdvo_dpms, .mode_fixup = intel_sdvo_mode_fixup, - .prepare = intel_encoder_prepare, .mode_set = intel_sdvo_mode_set, - .commit = intel_encoder_commit, + .disable = intel_encoder_noop, }; static const struct drm_connector_funcs intel_sdvo_connector_funcs = { - .dpms = drm_helper_connector_dpms, + .dpms = intel_sdvo_dpms, .detect = intel_sdvo_detect, .fill_modes = drm_helper_probe_single_connector_modes, .set_property = intel_sdvo_set_property, @@ -2026,6 +2114,7 @@ intel_sdvo_connector_init(struct intel_sdvo_connector *connector, connector->base.base.interlace_allowed = 1; connector->base.base.doublescan_allowed = 0; connector->base.base.display_info.subpixel_order = SubPixelHorizontalRGB; + connector->base.get_hw_state = intel_sdvo_connector_get_hw_state; intel_connector_attach_encoder(&connector->base, &encoder->base); drm_sysfs_connector_add(&connector->base.base); @@ -2064,17 +2153,18 @@ intel_sdvo_dvi_init(struct intel_sdvo *intel_sdvo, int device) intel_connector = &intel_sdvo_connector->base; connector = &intel_connector->base; - if (intel_sdvo_supports_hotplug(intel_sdvo) & (1 << device)) { + if (intel_sdvo_get_hotplug_support(intel_sdvo) & + intel_sdvo_connector->output_flag) { connector->polled = DRM_CONNECTOR_POLL_HPD; - intel_sdvo->hotplug_active[0] |= 1 << device; + intel_sdvo->hotplug_active |= intel_sdvo_connector->output_flag; /* Some SDVO devices have one-shot hotplug interrupts. * Ensure that they get re-enabled when an interrupt happens. */ intel_encoder->hot_plug = intel_sdvo_enable_hotplug; intel_sdvo_enable_hotplug(intel_encoder); - } - else + } else { connector->polled = DRM_CONNECTOR_POLL_CONNECT | DRM_CONNECTOR_POLL_DISCONNECT; + } encoder->encoder_type = DRM_MODE_ENCODER_TMDS; connector->connector_type = DRM_MODE_CONNECTOR_DVID; @@ -2082,8 +2172,7 @@ intel_sdvo_dvi_init(struct intel_sdvo *intel_sdvo, int device) connector->connector_type = DRM_MODE_CONNECTOR_HDMIA; intel_sdvo->is_hdmi = true; } - intel_sdvo->base.clone_mask = ((1 << INTEL_SDVO_NON_TV_CLONE_BIT) | - (1 << INTEL_ANALOG_CLONE_BIT)); + intel_sdvo->base.cloneable = true; intel_sdvo_connector_init(intel_sdvo_connector, intel_sdvo); if (intel_sdvo->is_hdmi) @@ -2114,7 +2203,7 @@ intel_sdvo_tv_init(struct intel_sdvo *intel_sdvo, int type) intel_sdvo->is_tv = true; intel_sdvo->base.needs_tv_clock = true; - intel_sdvo->base.clone_mask = 1 << INTEL_SDVO_TV_CLONE_BIT; + intel_sdvo->base.cloneable = false; intel_sdvo_connector_init(intel_sdvo_connector, intel_sdvo); @@ -2157,8 +2246,7 @@ intel_sdvo_analog_init(struct intel_sdvo *intel_sdvo, int device) intel_sdvo_connector->output_flag = SDVO_OUTPUT_RGB1; } - intel_sdvo->base.clone_mask = ((1 << INTEL_SDVO_NON_TV_CLONE_BIT) | - (1 << INTEL_ANALOG_CLONE_BIT)); + intel_sdvo->base.cloneable = true; intel_sdvo_connector_init(intel_sdvo_connector, intel_sdvo); @@ -2190,8 +2278,10 @@ intel_sdvo_lvds_init(struct intel_sdvo *intel_sdvo, int device) intel_sdvo_connector->output_flag = SDVO_OUTPUT_LVDS1; } - intel_sdvo->base.clone_mask = ((1 << INTEL_ANALOG_CLONE_BIT) | - (1 << INTEL_SDVO_LVDS_CLONE_BIT)); + /* SDVO LVDS is cloneable because the SDVO encoder does the upscaling, + * as opposed to native LVDS, where we upscale with the panel-fitter + * (and hence only the native LVDS resolution could be cloned). */ + intel_sdvo->base.cloneable = true; intel_sdvo_connector_init(intel_sdvo_connector, intel_sdvo); if (!intel_sdvo_create_enhance_property(intel_sdvo, intel_sdvo_connector)) @@ -2576,6 +2666,10 @@ bool intel_sdvo_init(struct drm_device *dev, uint32_t sdvo_reg, bool is_sdvob) drm_encoder_helper_add(&intel_encoder->base, &intel_sdvo_helper_funcs); + intel_encoder->disable = intel_disable_sdvo; + intel_encoder->enable = intel_enable_sdvo; + intel_encoder->get_hw_state = intel_sdvo_get_hw_state; + /* In default case sdvo lvds is false */ if (!intel_sdvo_get_capabilities(intel_sdvo, &intel_sdvo->caps)) goto err; @@ -2590,7 +2684,7 @@ bool intel_sdvo_init(struct drm_device *dev, uint32_t sdvo_reg, bool is_sdvob) /* Only enable the hotplug irq if we need it, to work around noisy * hotplug lines. */ - if (intel_sdvo->hotplug_active[0]) + if (intel_sdvo->hotplug_active) dev_priv->hotplug_supported_mask |= hotplug_mask; intel_sdvo_select_ddc_bus(dev_priv, intel_sdvo, sdvo_reg); diff --git a/drivers/gpu/drm/i915/intel_tv.c b/drivers/gpu/drm/i915/intel_tv.c index befce6c49704..d2c5c8f3baf3 100644 --- a/drivers/gpu/drm/i915/intel_tv.c +++ b/drivers/gpu/drm/i915/intel_tv.c @@ -836,22 +836,37 @@ static struct intel_tv *intel_attached_tv(struct drm_connector *connector) base); } +static bool +intel_tv_get_hw_state(struct intel_encoder *encoder, enum pipe *pipe) +{ + struct drm_device *dev = encoder->base.dev; + struct drm_i915_private *dev_priv = dev->dev_private; + u32 tmp = I915_READ(TV_CTL); + + if (!(tmp & TV_ENC_ENABLE)) + return false; + + *pipe = PORT_TO_PIPE(tmp); + + return true; +} + static void -intel_tv_dpms(struct drm_encoder *encoder, int mode) +intel_enable_tv(struct intel_encoder *encoder) { - struct drm_device *dev = encoder->dev; + struct drm_device *dev = encoder->base.dev; struct drm_i915_private *dev_priv = dev->dev_private; - switch (mode) { - case DRM_MODE_DPMS_ON: - I915_WRITE(TV_CTL, I915_READ(TV_CTL) | TV_ENC_ENABLE); - break; - case DRM_MODE_DPMS_STANDBY: - case DRM_MODE_DPMS_SUSPEND: - case DRM_MODE_DPMS_OFF: - I915_WRITE(TV_CTL, I915_READ(TV_CTL) & ~TV_ENC_ENABLE); - break; - } + I915_WRITE(TV_CTL, I915_READ(TV_CTL) | TV_ENC_ENABLE); +} + +static void +intel_disable_tv(struct intel_encoder *encoder) +{ + struct drm_device *dev = encoder->base.dev; + struct drm_i915_private *dev_priv = dev->dev_private; + + I915_WRITE(TV_CTL, I915_READ(TV_CTL) & ~TV_ENC_ENABLE); } static const struct tv_mode * @@ -895,17 +910,14 @@ intel_tv_mode_fixup(struct drm_encoder *encoder, const struct drm_display_mode *mode, struct drm_display_mode *adjusted_mode) { - struct drm_device *dev = encoder->dev; struct intel_tv *intel_tv = enc_to_intel_tv(encoder); const struct tv_mode *tv_mode = intel_tv_mode_find(intel_tv); - struct intel_encoder *other_encoder; if (!tv_mode) return false; - for_each_encoder_on_crtc(dev, encoder->crtc, other_encoder) - if (&other_encoder->base != encoder) - return false; + if (intel_encoder_check_is_cloned(&intel_tv->base)) + return false; adjusted_mode->clock = tv_mode->clock; return true; @@ -1303,12 +1315,9 @@ intel_tv_detect(struct drm_connector *connector, bool force) if (force) { struct intel_load_detect_pipe tmp; - if (intel_get_load_detect_pipe(&intel_tv->base, connector, - &mode, &tmp)) { + if (intel_get_load_detect_pipe(connector, &mode, &tmp)) { type = intel_tv_detect_type(intel_tv, connector); - intel_release_load_detect_pipe(&intel_tv->base, - connector, - &tmp); + intel_release_load_detect_pipe(connector, &tmp); } else return connector_status_unknown; } else @@ -1474,22 +1483,20 @@ intel_tv_set_property(struct drm_connector *connector, struct drm_property *prop } if (changed && crtc) - drm_crtc_helper_set_mode(crtc, &crtc->mode, crtc->x, - crtc->y, crtc->fb); + intel_set_mode(crtc, &crtc->mode, + crtc->x, crtc->y, crtc->fb); out: return ret; } static const struct drm_encoder_helper_funcs intel_tv_helper_funcs = { - .dpms = intel_tv_dpms, .mode_fixup = intel_tv_mode_fixup, - .prepare = intel_encoder_prepare, .mode_set = intel_tv_mode_set, - .commit = intel_encoder_commit, + .disable = intel_encoder_noop, }; static const struct drm_connector_funcs intel_tv_connector_funcs = { - .dpms = drm_helper_connector_dpms, + .dpms = intel_connector_dpms, .detect = intel_tv_detect, .destroy = intel_tv_destroy, .set_property = intel_tv_set_property, @@ -1619,10 +1626,15 @@ intel_tv_init(struct drm_device *dev) drm_encoder_init(dev, &intel_encoder->base, &intel_tv_enc_funcs, DRM_MODE_ENCODER_TVDAC); + intel_encoder->enable = intel_enable_tv; + intel_encoder->disable = intel_disable_tv; + intel_encoder->get_hw_state = intel_tv_get_hw_state; + intel_connector->get_hw_state = intel_connector_get_hw_state; + intel_connector_attach_encoder(intel_connector, intel_encoder); intel_encoder->type = INTEL_OUTPUT_TVOUT; intel_encoder->crtc_mask = (1 << 0) | (1 << 1); - intel_encoder->clone_mask = (1 << INTEL_TV_CLONE_BIT); + intel_encoder->cloneable = false; intel_encoder->base.possible_crtcs = ((1 << 0) | (1 << 1)); intel_encoder->base.possible_clones = (1 << INTEL_OUTPUT_TVOUT); intel_tv->type = DRM_MODE_CONNECTOR_Unknown; diff --git a/drivers/gpu/drm/mgag200/mgag200_drv.h b/drivers/gpu/drm/mgag200/mgag200_drv.h index 6f13b3563234..d22cbbf3a202 100644 --- a/drivers/gpu/drm/mgag200/mgag200_drv.h +++ b/drivers/gpu/drm/mgag200/mgag200_drv.h @@ -195,7 +195,6 @@ struct mga_device { struct drm_global_reference mem_global_ref; struct ttm_bo_global_ref bo_global_ref; struct ttm_bo_device bdev; - atomic_t validate_sequence; } ttm; u32 reg_1e24; /* SE model number */ diff --git a/drivers/gpu/drm/mgag200/mgag200_mode.c b/drivers/gpu/drm/mgag200/mgag200_mode.c index b69642d5d850..c7420e83c0b9 100644 --- a/drivers/gpu/drm/mgag200/mgag200_mode.c +++ b/drivers/gpu/drm/mgag200/mgag200_mode.c @@ -1399,7 +1399,6 @@ static int mga_vga_get_modes(struct drm_connector *connector) if (edid) { drm_mode_connector_update_edid_property(connector, edid); ret = drm_add_edid_modes(connector, edid); - connector->display_info.raw_edid = NULL; kfree(edid); } return ret; diff --git a/drivers/gpu/drm/nouveau/nouveau_dp.c b/drivers/gpu/drm/nouveau/nouveau_dp.c index e1387bd73d76..72266ae91fa1 100644 --- a/drivers/gpu/drm/nouveau/nouveau_dp.c +++ b/drivers/gpu/drm/nouveau/nouveau_dp.c @@ -166,9 +166,7 @@ dp_link_train_update(struct drm_device *dev, struct dp_state *dp, u32 delay) if (ret) return ret; - NV_DEBUG(drm, "status %02x %02x %02x %02x %02x %02x\n", - dp->stat[0], dp->stat[1], dp->stat[2], dp->stat[3], - dp->stat[4], dp->stat[5]); + NV_DEBUG(drm, "status %*ph\n", 6, dp->stat); return 0; } diff --git a/drivers/gpu/drm/radeon/atombios_crtc.c b/drivers/gpu/drm/radeon/atombios_crtc.c index 2817101fb167..96184d02c8d9 100644 --- a/drivers/gpu/drm/radeon/atombios_crtc.c +++ b/drivers/gpu/drm/radeon/atombios_crtc.c @@ -83,25 +83,19 @@ static void atombios_scaler_setup(struct drm_crtc *crtc) struct radeon_crtc *radeon_crtc = to_radeon_crtc(crtc); ENABLE_SCALER_PS_ALLOCATION args; int index = GetIndexIntoMasterTable(COMMAND, EnableScaler); - + struct radeon_encoder *radeon_encoder = + to_radeon_encoder(radeon_crtc->encoder); /* fixme - fill in enc_priv for atom dac */ enum radeon_tv_std tv_std = TV_STD_NTSC; bool is_tv = false, is_cv = false; - struct drm_encoder *encoder; if (!ASIC_IS_AVIVO(rdev) && radeon_crtc->crtc_id) return; - list_for_each_entry(encoder, &dev->mode_config.encoder_list, head) { - /* find tv std */ - if (encoder->crtc == crtc) { - struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder); - if (radeon_encoder->active_device & ATOM_DEVICE_TV_SUPPORT) { - struct radeon_encoder_atom_dac *tv_dac = radeon_encoder->enc_priv; - tv_std = tv_dac->tv_std; - is_tv = true; - } - } + if (radeon_encoder->active_device & ATOM_DEVICE_TV_SUPPORT) { + struct radeon_encoder_atom_dac *tv_dac = radeon_encoder->enc_priv; + tv_std = tv_dac->tv_std; + is_tv = true; } memset(&args, 0, sizeof(args)); @@ -533,99 +527,87 @@ union adjust_pixel_clock { }; static u32 atombios_adjust_pll(struct drm_crtc *crtc, - struct drm_display_mode *mode, - struct radeon_pll *pll, - bool ss_enabled, - struct radeon_atom_ss *ss) + struct drm_display_mode *mode) { + struct radeon_crtc *radeon_crtc = to_radeon_crtc(crtc); struct drm_device *dev = crtc->dev; struct radeon_device *rdev = dev->dev_private; - struct drm_encoder *encoder = NULL; - struct radeon_encoder *radeon_encoder = NULL; - struct drm_connector *connector = NULL; + struct drm_encoder *encoder = radeon_crtc->encoder; + struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder); + struct drm_connector *connector = radeon_get_connector_for_encoder(encoder); u32 adjusted_clock = mode->clock; - int encoder_mode = 0; + int encoder_mode = atombios_get_encoder_mode(encoder); u32 dp_clock = mode->clock; - int bpc = 8; - bool is_duallink = false; + int bpc = radeon_get_monitor_bpc(connector); + bool is_duallink = radeon_dig_monitor_is_duallink(encoder, mode->clock); /* reset the pll flags */ - pll->flags = 0; + radeon_crtc->pll_flags = 0; if (ASIC_IS_AVIVO(rdev)) { if ((rdev->family == CHIP_RS600) || (rdev->family == CHIP_RS690) || (rdev->family == CHIP_RS740)) - pll->flags |= (/*RADEON_PLL_USE_FRAC_FB_DIV |*/ - RADEON_PLL_PREFER_CLOSEST_LOWER); + radeon_crtc->pll_flags |= (/*RADEON_PLL_USE_FRAC_FB_DIV |*/ + RADEON_PLL_PREFER_CLOSEST_LOWER); if (ASIC_IS_DCE32(rdev) && mode->clock > 200000) /* range limits??? */ - pll->flags |= RADEON_PLL_PREFER_HIGH_FB_DIV; + radeon_crtc->pll_flags |= RADEON_PLL_PREFER_HIGH_FB_DIV; else - pll->flags |= RADEON_PLL_PREFER_LOW_REF_DIV; + radeon_crtc->pll_flags |= RADEON_PLL_PREFER_LOW_REF_DIV; if (rdev->family < CHIP_RV770) - pll->flags |= RADEON_PLL_PREFER_MINM_OVER_MAXP; + radeon_crtc->pll_flags |= RADEON_PLL_PREFER_MINM_OVER_MAXP; /* use frac fb div on APUs */ if (ASIC_IS_DCE41(rdev) || ASIC_IS_DCE61(rdev)) - pll->flags |= RADEON_PLL_USE_FRAC_FB_DIV; + radeon_crtc->pll_flags |= RADEON_PLL_USE_FRAC_FB_DIV; } else { - pll->flags |= RADEON_PLL_LEGACY; + radeon_crtc->pll_flags |= RADEON_PLL_LEGACY; if (mode->clock > 200000) /* range limits??? */ - pll->flags |= RADEON_PLL_PREFER_HIGH_FB_DIV; + radeon_crtc->pll_flags |= RADEON_PLL_PREFER_HIGH_FB_DIV; else - pll->flags |= RADEON_PLL_PREFER_LOW_REF_DIV; + radeon_crtc->pll_flags |= RADEON_PLL_PREFER_LOW_REF_DIV; } - list_for_each_entry(encoder, &dev->mode_config.encoder_list, head) { - if (encoder->crtc == crtc) { - radeon_encoder = to_radeon_encoder(encoder); - connector = radeon_get_connector_for_encoder(encoder); - bpc = radeon_get_monitor_bpc(connector); - encoder_mode = atombios_get_encoder_mode(encoder); - is_duallink = radeon_dig_monitor_is_duallink(encoder, mode->clock); - if ((radeon_encoder->devices & (ATOM_DEVICE_LCD_SUPPORT | ATOM_DEVICE_DFP_SUPPORT)) || - (radeon_encoder_get_dp_bridge_encoder_id(encoder) != ENCODER_OBJECT_ID_NONE)) { - if (connector) { - struct radeon_connector *radeon_connector = to_radeon_connector(connector); - struct radeon_connector_atom_dig *dig_connector = - radeon_connector->con_priv; - - dp_clock = dig_connector->dp_clock; - } - } + if ((radeon_encoder->devices & (ATOM_DEVICE_LCD_SUPPORT | ATOM_DEVICE_DFP_SUPPORT)) || + (radeon_encoder_get_dp_bridge_encoder_id(encoder) != ENCODER_OBJECT_ID_NONE)) { + if (connector) { + struct radeon_connector *radeon_connector = to_radeon_connector(connector); + struct radeon_connector_atom_dig *dig_connector = + radeon_connector->con_priv; - /* use recommended ref_div for ss */ - if (radeon_encoder->devices & (ATOM_DEVICE_LCD_SUPPORT)) { - if (ss_enabled) { - if (ss->refdiv) { - pll->flags |= RADEON_PLL_USE_REF_DIV; - pll->reference_div = ss->refdiv; - if (ASIC_IS_AVIVO(rdev)) - pll->flags |= RADEON_PLL_USE_FRAC_FB_DIV; - } - } - } + dp_clock = dig_connector->dp_clock; + } + } - if (ASIC_IS_AVIVO(rdev)) { - /* DVO wants 2x pixel clock if the DVO chip is in 12 bit mode */ - if (radeon_encoder->encoder_id == ENCODER_OBJECT_ID_INTERNAL_KLDSCP_DVO1) - adjusted_clock = mode->clock * 2; - if (radeon_encoder->active_device & (ATOM_DEVICE_TV_SUPPORT)) - pll->flags |= RADEON_PLL_PREFER_CLOSEST_LOWER; - if (radeon_encoder->devices & (ATOM_DEVICE_LCD_SUPPORT)) - pll->flags |= RADEON_PLL_IS_LCD; - } else { - if (encoder->encoder_type != DRM_MODE_ENCODER_DAC) - pll->flags |= RADEON_PLL_NO_ODD_POST_DIV; - if (encoder->encoder_type == DRM_MODE_ENCODER_LVDS) - pll->flags |= RADEON_PLL_USE_REF_DIV; + /* use recommended ref_div for ss */ + if (radeon_encoder->devices & (ATOM_DEVICE_LCD_SUPPORT)) { + if (radeon_crtc->ss_enabled) { + if (radeon_crtc->ss.refdiv) { + radeon_crtc->pll_flags |= RADEON_PLL_USE_REF_DIV; + radeon_crtc->pll_reference_div = radeon_crtc->ss.refdiv; + if (ASIC_IS_AVIVO(rdev)) + radeon_crtc->pll_flags |= RADEON_PLL_USE_FRAC_FB_DIV; } - break; } } + if (ASIC_IS_AVIVO(rdev)) { + /* DVO wants 2x pixel clock if the DVO chip is in 12 bit mode */ + if (radeon_encoder->encoder_id == ENCODER_OBJECT_ID_INTERNAL_KLDSCP_DVO1) + adjusted_clock = mode->clock * 2; + if (radeon_encoder->active_device & (ATOM_DEVICE_TV_SUPPORT)) + radeon_crtc->pll_flags |= RADEON_PLL_PREFER_CLOSEST_LOWER; + if (radeon_encoder->devices & (ATOM_DEVICE_LCD_SUPPORT)) + radeon_crtc->pll_flags |= RADEON_PLL_IS_LCD; + } else { + if (encoder->encoder_type != DRM_MODE_ENCODER_DAC) + radeon_crtc->pll_flags |= RADEON_PLL_NO_ODD_POST_DIV; + if (encoder->encoder_type == DRM_MODE_ENCODER_LVDS) + radeon_crtc->pll_flags |= RADEON_PLL_USE_REF_DIV; + } + /* DCE3+ has an AdjustDisplayPll that will adjust the pixel clock * accordingly based on the encoder/transmitter to work around * special hw requirements. @@ -650,7 +632,7 @@ static u32 atombios_adjust_pll(struct drm_crtc *crtc, args.v1.usPixelClock = cpu_to_le16(mode->clock / 10); args.v1.ucTransmitterID = radeon_encoder->encoder_id; args.v1.ucEncodeMode = encoder_mode; - if (ss_enabled && ss->percentage) + if (radeon_crtc->ss_enabled && radeon_crtc->ss.percentage) args.v1.ucConfig |= ADJUST_DISPLAY_CONFIG_SS_ENABLE; @@ -663,7 +645,7 @@ static u32 atombios_adjust_pll(struct drm_crtc *crtc, args.v3.sInput.ucTransmitterID = radeon_encoder->encoder_id; args.v3.sInput.ucEncodeMode = encoder_mode; args.v3.sInput.ucDispPllConfig = 0; - if (ss_enabled && ss->percentage) + if (radeon_crtc->ss_enabled && radeon_crtc->ss.percentage) args.v3.sInput.ucDispPllConfig |= DISPPLL_CONFIG_SS_ENABLE; if (ENCODER_MODE_IS_DP(encoder_mode)) { @@ -695,14 +677,14 @@ static u32 atombios_adjust_pll(struct drm_crtc *crtc, index, (uint32_t *)&args); adjusted_clock = le32_to_cpu(args.v3.sOutput.ulDispPllFreq) * 10; if (args.v3.sOutput.ucRefDiv) { - pll->flags |= RADEON_PLL_USE_FRAC_FB_DIV; - pll->flags |= RADEON_PLL_USE_REF_DIV; - pll->reference_div = args.v3.sOutput.ucRefDiv; + radeon_crtc->pll_flags |= RADEON_PLL_USE_FRAC_FB_DIV; + radeon_crtc->pll_flags |= RADEON_PLL_USE_REF_DIV; + radeon_crtc->pll_reference_div = args.v3.sOutput.ucRefDiv; } if (args.v3.sOutput.ucPostDiv) { - pll->flags |= RADEON_PLL_USE_FRAC_FB_DIV; - pll->flags |= RADEON_PLL_USE_POST_DIV; - pll->post_div = args.v3.sOutput.ucPostDiv; + radeon_crtc->pll_flags |= RADEON_PLL_USE_FRAC_FB_DIV; + radeon_crtc->pll_flags |= RADEON_PLL_USE_POST_DIV; + radeon_crtc->pll_post_div = args.v3.sOutput.ucPostDiv; } break; default: @@ -837,7 +819,10 @@ static void atombios_crtc_program_pll(struct drm_crtc *crtc, args.v3.ucFracFbDiv = frac_fb_div; args.v3.ucPostDiv = post_div; args.v3.ucPpll = pll_id; - args.v3.ucMiscInfo = (pll_id << 2); + if (crtc_id == ATOM_CRTC2) + args.v3.ucMiscInfo = PIXEL_CLOCK_MISC_CRTC_SEL_CRTC2; + else + args.v3.ucMiscInfo = PIXEL_CLOCK_MISC_CRTC_SEL_CRTC1; if (ss_enabled && (ss->type & ATOM_EXTERNAL_SS_MASK)) args.v3.ucMiscInfo |= PIXEL_CLOCK_MISC_REF_DIV_SRC; args.v3.ucTransmitterId = encoder_id; @@ -907,58 +892,29 @@ static void atombios_crtc_program_pll(struct drm_crtc *crtc, atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); } -static void atombios_crtc_set_pll(struct drm_crtc *crtc, struct drm_display_mode *mode) +static bool atombios_crtc_prepare_pll(struct drm_crtc *crtc, struct drm_display_mode *mode) { struct radeon_crtc *radeon_crtc = to_radeon_crtc(crtc); struct drm_device *dev = crtc->dev; struct radeon_device *rdev = dev->dev_private; - struct drm_encoder *encoder = NULL; - struct radeon_encoder *radeon_encoder = NULL; - u32 pll_clock = mode->clock; - u32 ref_div = 0, fb_div = 0, frac_fb_div = 0, post_div = 0; - struct radeon_pll *pll; - u32 adjusted_clock; - int encoder_mode = 0; - struct radeon_atom_ss ss; - bool ss_enabled = false; - int bpc = 8; + struct radeon_encoder *radeon_encoder = + to_radeon_encoder(radeon_crtc->encoder); + int encoder_mode = atombios_get_encoder_mode(radeon_crtc->encoder); - list_for_each_entry(encoder, &dev->mode_config.encoder_list, head) { - if (encoder->crtc == crtc) { - radeon_encoder = to_radeon_encoder(encoder); - encoder_mode = atombios_get_encoder_mode(encoder); - break; - } - } - - if (!radeon_encoder) - return; - - switch (radeon_crtc->pll_id) { - case ATOM_PPLL1: - pll = &rdev->clock.p1pll; - break; - case ATOM_PPLL2: - pll = &rdev->clock.p2pll; - break; - case ATOM_DCPLL: - case ATOM_PPLL_INVALID: - default: - pll = &rdev->clock.dcpll; - break; - } + radeon_crtc->bpc = 8; + radeon_crtc->ss_enabled = false; if ((radeon_encoder->active_device & (ATOM_DEVICE_LCD_SUPPORT | ATOM_DEVICE_DFP_SUPPORT)) || - (radeon_encoder_get_dp_bridge_encoder_id(encoder) != ENCODER_OBJECT_ID_NONE)) { + (radeon_encoder_get_dp_bridge_encoder_id(radeon_crtc->encoder) != ENCODER_OBJECT_ID_NONE)) { struct radeon_encoder_atom_dig *dig = radeon_encoder->enc_priv; struct drm_connector *connector = - radeon_get_connector_for_encoder(encoder); + radeon_get_connector_for_encoder(radeon_crtc->encoder); struct radeon_connector *radeon_connector = to_radeon_connector(connector); struct radeon_connector_atom_dig *dig_connector = radeon_connector->con_priv; int dp_clock; - bpc = radeon_get_monitor_bpc(connector); + radeon_crtc->bpc = radeon_get_monitor_bpc(connector); switch (encoder_mode) { case ATOM_ENCODER_MODE_DP_MST: @@ -966,45 +922,54 @@ static void atombios_crtc_set_pll(struct drm_crtc *crtc, struct drm_display_mode /* DP/eDP */ dp_clock = dig_connector->dp_clock / 10; if (ASIC_IS_DCE4(rdev)) - ss_enabled = - radeon_atombios_get_asic_ss_info(rdev, &ss, + radeon_crtc->ss_enabled = + radeon_atombios_get_asic_ss_info(rdev, &radeon_crtc->ss, ASIC_INTERNAL_SS_ON_DP, dp_clock); else { if (dp_clock == 16200) { - ss_enabled = - radeon_atombios_get_ppll_ss_info(rdev, &ss, + radeon_crtc->ss_enabled = + radeon_atombios_get_ppll_ss_info(rdev, + &radeon_crtc->ss, ATOM_DP_SS_ID2); - if (!ss_enabled) - ss_enabled = - radeon_atombios_get_ppll_ss_info(rdev, &ss, + if (!radeon_crtc->ss_enabled) + radeon_crtc->ss_enabled = + radeon_atombios_get_ppll_ss_info(rdev, + &radeon_crtc->ss, ATOM_DP_SS_ID1); } else - ss_enabled = - radeon_atombios_get_ppll_ss_info(rdev, &ss, + radeon_crtc->ss_enabled = + radeon_atombios_get_ppll_ss_info(rdev, + &radeon_crtc->ss, ATOM_DP_SS_ID1); } break; case ATOM_ENCODER_MODE_LVDS: if (ASIC_IS_DCE4(rdev)) - ss_enabled = radeon_atombios_get_asic_ss_info(rdev, &ss, - dig->lcd_ss_id, - mode->clock / 10); + radeon_crtc->ss_enabled = + radeon_atombios_get_asic_ss_info(rdev, + &radeon_crtc->ss, + dig->lcd_ss_id, + mode->clock / 10); else - ss_enabled = radeon_atombios_get_ppll_ss_info(rdev, &ss, - dig->lcd_ss_id); + radeon_crtc->ss_enabled = + radeon_atombios_get_ppll_ss_info(rdev, + &radeon_crtc->ss, + dig->lcd_ss_id); break; case ATOM_ENCODER_MODE_DVI: if (ASIC_IS_DCE4(rdev)) - ss_enabled = - radeon_atombios_get_asic_ss_info(rdev, &ss, + radeon_crtc->ss_enabled = + radeon_atombios_get_asic_ss_info(rdev, + &radeon_crtc->ss, ASIC_INTERNAL_SS_ON_TMDS, mode->clock / 10); break; case ATOM_ENCODER_MODE_HDMI: if (ASIC_IS_DCE4(rdev)) - ss_enabled = - radeon_atombios_get_asic_ss_info(rdev, &ss, + radeon_crtc->ss_enabled = + radeon_atombios_get_asic_ss_info(rdev, + &radeon_crtc->ss, ASIC_INTERNAL_SS_ON_HDMI, mode->clock / 10); break; @@ -1014,43 +979,80 @@ static void atombios_crtc_set_pll(struct drm_crtc *crtc, struct drm_display_mode } /* adjust pixel clock as needed */ - adjusted_clock = atombios_adjust_pll(crtc, mode, pll, ss_enabled, &ss); + radeon_crtc->adjusted_clock = atombios_adjust_pll(crtc, mode); + + return true; +} + +static void atombios_crtc_set_pll(struct drm_crtc *crtc, struct drm_display_mode *mode) +{ + struct radeon_crtc *radeon_crtc = to_radeon_crtc(crtc); + struct drm_device *dev = crtc->dev; + struct radeon_device *rdev = dev->dev_private; + struct radeon_encoder *radeon_encoder = + to_radeon_encoder(radeon_crtc->encoder); + u32 pll_clock = mode->clock; + u32 ref_div = 0, fb_div = 0, frac_fb_div = 0, post_div = 0; + struct radeon_pll *pll; + int encoder_mode = atombios_get_encoder_mode(radeon_crtc->encoder); + + switch (radeon_crtc->pll_id) { + case ATOM_PPLL1: + pll = &rdev->clock.p1pll; + break; + case ATOM_PPLL2: + pll = &rdev->clock.p2pll; + break; + case ATOM_DCPLL: + case ATOM_PPLL_INVALID: + default: + pll = &rdev->clock.dcpll; + break; + } + + /* update pll params */ + pll->flags = radeon_crtc->pll_flags; + pll->reference_div = radeon_crtc->pll_reference_div; + pll->post_div = radeon_crtc->pll_post_div; if (radeon_encoder->active_device & (ATOM_DEVICE_TV_SUPPORT)) /* TV seems to prefer the legacy algo on some boards */ - radeon_compute_pll_legacy(pll, adjusted_clock, &pll_clock, &fb_div, &frac_fb_div, - &ref_div, &post_div); + radeon_compute_pll_legacy(pll, radeon_crtc->adjusted_clock, &pll_clock, + &fb_div, &frac_fb_div, &ref_div, &post_div); else if (ASIC_IS_AVIVO(rdev)) - radeon_compute_pll_avivo(pll, adjusted_clock, &pll_clock, &fb_div, &frac_fb_div, - &ref_div, &post_div); + radeon_compute_pll_avivo(pll, radeon_crtc->adjusted_clock, &pll_clock, + &fb_div, &frac_fb_div, &ref_div, &post_div); else - radeon_compute_pll_legacy(pll, adjusted_clock, &pll_clock, &fb_div, &frac_fb_div, - &ref_div, &post_div); + radeon_compute_pll_legacy(pll, radeon_crtc->adjusted_clock, &pll_clock, + &fb_div, &frac_fb_div, &ref_div, &post_div); - atombios_crtc_program_ss(rdev, ATOM_DISABLE, radeon_crtc->pll_id, radeon_crtc->crtc_id, &ss); + atombios_crtc_program_ss(rdev, ATOM_DISABLE, radeon_crtc->pll_id, + radeon_crtc->crtc_id, &radeon_crtc->ss); atombios_crtc_program_pll(crtc, radeon_crtc->crtc_id, radeon_crtc->pll_id, encoder_mode, radeon_encoder->encoder_id, mode->clock, - ref_div, fb_div, frac_fb_div, post_div, bpc, ss_enabled, &ss); + ref_div, fb_div, frac_fb_div, post_div, + radeon_crtc->bpc, radeon_crtc->ss_enabled, &radeon_crtc->ss); - if (ss_enabled) { + if (radeon_crtc->ss_enabled) { /* calculate ss amount and step size */ if (ASIC_IS_DCE4(rdev)) { u32 step_size; - u32 amount = (((fb_div * 10) + frac_fb_div) * ss.percentage) / 10000; - ss.amount = (amount / 10) & ATOM_PPLL_SS_AMOUNT_V2_FBDIV_MASK; - ss.amount |= ((amount - (amount / 10)) << ATOM_PPLL_SS_AMOUNT_V2_NFRAC_SHIFT) & + u32 amount = (((fb_div * 10) + frac_fb_div) * radeon_crtc->ss.percentage) / 10000; + radeon_crtc->ss.amount = (amount / 10) & ATOM_PPLL_SS_AMOUNT_V2_FBDIV_MASK; + radeon_crtc->ss.amount |= ((amount - (amount / 10)) << ATOM_PPLL_SS_AMOUNT_V2_NFRAC_SHIFT) & ATOM_PPLL_SS_AMOUNT_V2_NFRAC_MASK; - if (ss.type & ATOM_PPLL_SS_TYPE_V2_CENTRE_SPREAD) - step_size = (4 * amount * ref_div * (ss.rate * 2048)) / + if (radeon_crtc->ss.type & ATOM_PPLL_SS_TYPE_V2_CENTRE_SPREAD) + step_size = (4 * amount * ref_div * (radeon_crtc->ss.rate * 2048)) / (125 * 25 * pll->reference_freq / 100); else - step_size = (2 * amount * ref_div * (ss.rate * 2048)) / + step_size = (2 * amount * ref_div * (radeon_crtc->ss.rate * 2048)) / (125 * 25 * pll->reference_freq / 100); - ss.step = step_size; + radeon_crtc->ss.step = step_size; } - atombios_crtc_program_ss(rdev, ATOM_ENABLE, radeon_crtc->pll_id, radeon_crtc->crtc_id, &ss); + atombios_crtc_program_ss(rdev, ATOM_ENABLE, radeon_crtc->pll_id, + radeon_crtc->crtc_id, &radeon_crtc->ss); } } @@ -1479,85 +1481,251 @@ static void radeon_legacy_atom_fixup(struct drm_crtc *crtc) } } +/** + * radeon_get_pll_use_mask - look up a mask of which pplls are in use + * + * @crtc: drm crtc + * + * Returns the mask of which PPLLs (Pixel PLLs) are in use. + */ +static u32 radeon_get_pll_use_mask(struct drm_crtc *crtc) +{ + struct drm_device *dev = crtc->dev; + struct drm_crtc *test_crtc; + struct radeon_crtc *test_radeon_crtc; + u32 pll_in_use = 0; + + list_for_each_entry(test_crtc, &dev->mode_config.crtc_list, head) { + if (crtc == test_crtc) + continue; + + test_radeon_crtc = to_radeon_crtc(test_crtc); + if (test_radeon_crtc->pll_id != ATOM_PPLL_INVALID) + pll_in_use |= (1 << test_radeon_crtc->pll_id); + } + return pll_in_use; +} + +/** + * radeon_get_shared_dp_ppll - return the PPLL used by another crtc for DP + * + * @crtc: drm crtc + * + * Returns the PPLL (Pixel PLL) used by another crtc/encoder which is + * also in DP mode. For DP, a single PPLL can be used for all DP + * crtcs/encoders. + */ +static int radeon_get_shared_dp_ppll(struct drm_crtc *crtc) +{ + struct drm_device *dev = crtc->dev; + struct drm_crtc *test_crtc; + struct radeon_crtc *test_radeon_crtc; + + list_for_each_entry(test_crtc, &dev->mode_config.crtc_list, head) { + if (crtc == test_crtc) + continue; + test_radeon_crtc = to_radeon_crtc(test_crtc); + if (test_radeon_crtc->encoder && + ENCODER_MODE_IS_DP(atombios_get_encoder_mode(test_radeon_crtc->encoder))) { + /* for DP use the same PLL for all */ + if (test_radeon_crtc->pll_id != ATOM_PPLL_INVALID) + return test_radeon_crtc->pll_id; + } + } + return ATOM_PPLL_INVALID; +} + +/** + * radeon_get_shared_nondp_ppll - return the PPLL used by another non-DP crtc + * + * @crtc: drm crtc + * @encoder: drm encoder + * + * Returns the PPLL (Pixel PLL) used by another non-DP crtc/encoder which can + * be shared (i.e., same clock). + */ +static int radeon_get_shared_nondp_ppll(struct drm_crtc *crtc) +{ + struct radeon_crtc *radeon_crtc = to_radeon_crtc(crtc); + struct drm_device *dev = crtc->dev; + struct drm_crtc *test_crtc; + struct radeon_crtc *test_radeon_crtc; + u32 adjusted_clock, test_adjusted_clock; + + adjusted_clock = radeon_crtc->adjusted_clock; + + if (adjusted_clock == 0) + return ATOM_PPLL_INVALID; + + list_for_each_entry(test_crtc, &dev->mode_config.crtc_list, head) { + if (crtc == test_crtc) + continue; + test_radeon_crtc = to_radeon_crtc(test_crtc); + if (test_radeon_crtc->encoder && + !ENCODER_MODE_IS_DP(atombios_get_encoder_mode(test_radeon_crtc->encoder))) { + /* check if we are already driving this connector with another crtc */ + if (test_radeon_crtc->connector == radeon_crtc->connector) { + /* if we are, return that pll */ + if (test_radeon_crtc->pll_id != ATOM_PPLL_INVALID) + return test_radeon_crtc->pll_id; + } + /* for non-DP check the clock */ + test_adjusted_clock = test_radeon_crtc->adjusted_clock; + if ((crtc->mode.clock == test_crtc->mode.clock) && + (adjusted_clock == test_adjusted_clock) && + (radeon_crtc->ss_enabled == test_radeon_crtc->ss_enabled) && + (test_radeon_crtc->pll_id != ATOM_PPLL_INVALID)) + return test_radeon_crtc->pll_id; + } + } + return ATOM_PPLL_INVALID; +} + +/** + * radeon_atom_pick_pll - Allocate a PPLL for use by the crtc. + * + * @crtc: drm crtc + * + * Returns the PPLL (Pixel PLL) to be used by the crtc. For DP monitors + * a single PPLL can be used for all DP crtcs/encoders. For non-DP + * monitors a dedicated PPLL must be used. If a particular board has + * an external DP PLL, return ATOM_PPLL_INVALID to skip PLL programming + * as there is no need to program the PLL itself. If we are not able to + * allocate a PLL, return ATOM_PPLL_INVALID to skip PLL programming to + * avoid messing up an existing monitor. + * + * Asic specific PLL information + * + * DCE 6.1 + * - PPLL2 is only available to UNIPHYA (both DP and non-DP) + * - PPLL0, PPLL1 are available for UNIPHYB/C/D/E/F (both DP and non-DP) + * + * DCE 6.0 + * - PPLL0 is available to all UNIPHY (DP only) + * - PPLL1, PPLL2 are available for all UNIPHY (both DP and non-DP) and DAC + * + * DCE 5.0 + * - DCPLL is available to all UNIPHY (DP only) + * - PPLL1, PPLL2 are available for all UNIPHY (both DP and non-DP) and DAC + * + * DCE 3.0/4.0/4.1 + * - PPLL1, PPLL2 are available for all UNIPHY (both DP and non-DP) and DAC + * + */ static int radeon_atom_pick_pll(struct drm_crtc *crtc) { struct radeon_crtc *radeon_crtc = to_radeon_crtc(crtc); struct drm_device *dev = crtc->dev; struct radeon_device *rdev = dev->dev_private; - struct drm_encoder *test_encoder; - struct drm_crtc *test_crtc; - uint32_t pll_in_use = 0; + struct radeon_encoder *radeon_encoder = + to_radeon_encoder(radeon_crtc->encoder); + u32 pll_in_use; + int pll; if (ASIC_IS_DCE61(rdev)) { - list_for_each_entry(test_encoder, &dev->mode_config.encoder_list, head) { - if (test_encoder->crtc && (test_encoder->crtc == crtc)) { - struct radeon_encoder *test_radeon_encoder = - to_radeon_encoder(test_encoder); - struct radeon_encoder_atom_dig *dig = - test_radeon_encoder->enc_priv; - - if ((test_radeon_encoder->encoder_id == - ENCODER_OBJECT_ID_INTERNAL_UNIPHY) && - (dig->linkb == false)) /* UNIPHY A uses PPLL2 */ - return ATOM_PPLL2; + struct radeon_encoder_atom_dig *dig = + radeon_encoder->enc_priv; + + if ((radeon_encoder->encoder_id == ENCODER_OBJECT_ID_INTERNAL_UNIPHY) && + (dig->linkb == false)) + /* UNIPHY A uses PPLL2 */ + return ATOM_PPLL2; + else if (ENCODER_MODE_IS_DP(atombios_get_encoder_mode(radeon_crtc->encoder))) { + /* UNIPHY B/C/D/E/F */ + if (rdev->clock.dp_extclk) + /* skip PPLL programming if using ext clock */ + return ATOM_PPLL_INVALID; + else { + /* use the same PPLL for all DP monitors */ + pll = radeon_get_shared_dp_ppll(crtc); + if (pll != ATOM_PPLL_INVALID) + return pll; } + } else { + /* use the same PPLL for all monitors with the same clock */ + pll = radeon_get_shared_nondp_ppll(crtc); + if (pll != ATOM_PPLL_INVALID) + return pll; } /* UNIPHY B/C/D/E/F */ - list_for_each_entry(test_crtc, &dev->mode_config.crtc_list, head) { - struct radeon_crtc *radeon_test_crtc; - - if (crtc == test_crtc) - continue; - - radeon_test_crtc = to_radeon_crtc(test_crtc); - if ((radeon_test_crtc->pll_id == ATOM_PPLL0) || - (radeon_test_crtc->pll_id == ATOM_PPLL1)) - pll_in_use |= (1 << radeon_test_crtc->pll_id); - } - if (!(pll_in_use & 4)) + pll_in_use = radeon_get_pll_use_mask(crtc); + if (!(pll_in_use & (1 << ATOM_PPLL0))) return ATOM_PPLL0; - return ATOM_PPLL1; + if (!(pll_in_use & (1 << ATOM_PPLL1))) + return ATOM_PPLL1; + DRM_ERROR("unable to allocate a PPLL\n"); + return ATOM_PPLL_INVALID; } else if (ASIC_IS_DCE4(rdev)) { - list_for_each_entry(test_encoder, &dev->mode_config.encoder_list, head) { - if (test_encoder->crtc && (test_encoder->crtc == crtc)) { - /* in DP mode, the DP ref clock can come from PPLL, DCPLL, or ext clock, - * depending on the asic: - * DCE4: PPLL or ext clock - * DCE5: DCPLL or ext clock - * - * Setting ATOM_PPLL_INVALID will cause SetPixelClock to skip - * PPLL/DCPLL programming and only program the DP DTO for the - * crtc virtual pixel clock. - */ - if (ENCODER_MODE_IS_DP(atombios_get_encoder_mode(test_encoder))) { - if (rdev->clock.dp_extclk) - return ATOM_PPLL_INVALID; - else if (ASIC_IS_DCE6(rdev)) - return ATOM_PPLL0; - else if (ASIC_IS_DCE5(rdev)) - return ATOM_DCPLL; - } + /* in DP mode, the DP ref clock can come from PPLL, DCPLL, or ext clock, + * depending on the asic: + * DCE4: PPLL or ext clock + * DCE5: PPLL, DCPLL, or ext clock + * DCE6: PPLL, PPLL0, or ext clock + * + * Setting ATOM_PPLL_INVALID will cause SetPixelClock to skip + * PPLL/DCPLL programming and only program the DP DTO for the + * crtc virtual pixel clock. + */ + if (ENCODER_MODE_IS_DP(atombios_get_encoder_mode(radeon_crtc->encoder))) { + if (rdev->clock.dp_extclk) + /* skip PPLL programming if using ext clock */ + return ATOM_PPLL_INVALID; + else if (ASIC_IS_DCE6(rdev)) + /* use PPLL0 for all DP */ + return ATOM_PPLL0; + else if (ASIC_IS_DCE5(rdev)) + /* use DCPLL for all DP */ + return ATOM_DCPLL; + else { + /* use the same PPLL for all DP monitors */ + pll = radeon_get_shared_dp_ppll(crtc); + if (pll != ATOM_PPLL_INVALID) + return pll; } + } else { + /* use the same PPLL for all monitors with the same clock */ + pll = radeon_get_shared_nondp_ppll(crtc); + if (pll != ATOM_PPLL_INVALID) + return pll; } - - /* otherwise, pick one of the plls */ - list_for_each_entry(test_crtc, &dev->mode_config.crtc_list, head) { - struct radeon_crtc *radeon_test_crtc; - - if (crtc == test_crtc) - continue; - - radeon_test_crtc = to_radeon_crtc(test_crtc); - if ((radeon_test_crtc->pll_id >= ATOM_PPLL1) && - (radeon_test_crtc->pll_id <= ATOM_PPLL2)) - pll_in_use |= (1 << radeon_test_crtc->pll_id); - } - if (!(pll_in_use & 1)) + /* all other cases */ + pll_in_use = radeon_get_pll_use_mask(crtc); + if (!(pll_in_use & (1 << ATOM_PPLL2))) + return ATOM_PPLL2; + if (!(pll_in_use & (1 << ATOM_PPLL1))) return ATOM_PPLL1; - return ATOM_PPLL2; - } else - return radeon_crtc->crtc_id; - + DRM_ERROR("unable to allocate a PPLL\n"); + return ATOM_PPLL_INVALID; + } else { + if (ASIC_IS_AVIVO(rdev)) { + /* in DP mode, the DP ref clock can come from either PPLL + * depending on the asic: + * DCE3: PPLL1 or PPLL2 + */ + if (ENCODER_MODE_IS_DP(atombios_get_encoder_mode(radeon_crtc->encoder))) { + /* use the same PPLL for all DP monitors */ + pll = radeon_get_shared_dp_ppll(crtc); + if (pll != ATOM_PPLL_INVALID) + return pll; + } else { + /* use the same PPLL for all monitors with the same clock */ + pll = radeon_get_shared_nondp_ppll(crtc); + if (pll != ATOM_PPLL_INVALID) + return pll; + } + /* all other cases */ + pll_in_use = radeon_get_pll_use_mask(crtc); + if (!(pll_in_use & (1 << ATOM_PPLL2))) + return ATOM_PPLL2; + if (!(pll_in_use & (1 << ATOM_PPLL1))) + return ATOM_PPLL1; + DRM_ERROR("unable to allocate a PPLL\n"); + return ATOM_PPLL_INVALID; + } else { + /* on pre-R5xx asics, the crtc to pll mapping is hardcoded */ + return radeon_crtc->crtc_id; + } + } } void radeon_atom_disp_eng_pll_init(struct radeon_device *rdev) @@ -1588,18 +1756,13 @@ int atombios_crtc_mode_set(struct drm_crtc *crtc, struct radeon_crtc *radeon_crtc = to_radeon_crtc(crtc); struct drm_device *dev = crtc->dev; struct radeon_device *rdev = dev->dev_private; - struct drm_encoder *encoder; + struct radeon_encoder *radeon_encoder = + to_radeon_encoder(radeon_crtc->encoder); bool is_tvcv = false; - list_for_each_entry(encoder, &dev->mode_config.encoder_list, head) { - /* find tv std */ - if (encoder->crtc == crtc) { - struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder); - if (radeon_encoder->active_device & - (ATOM_DEVICE_TV_SUPPORT | ATOM_DEVICE_CV_SUPPORT)) - is_tvcv = true; - } - } + if (radeon_encoder->active_device & + (ATOM_DEVICE_TV_SUPPORT | ATOM_DEVICE_CV_SUPPORT)) + is_tvcv = true; atombios_crtc_set_pll(crtc, adjusted_mode); @@ -1626,8 +1789,34 @@ static bool atombios_crtc_mode_fixup(struct drm_crtc *crtc, const struct drm_display_mode *mode, struct drm_display_mode *adjusted_mode) { + struct radeon_crtc *radeon_crtc = to_radeon_crtc(crtc); + struct drm_device *dev = crtc->dev; + struct drm_encoder *encoder; + + /* assign the encoder to the radeon crtc to avoid repeated lookups later */ + list_for_each_entry(encoder, &dev->mode_config.encoder_list, head) { + if (encoder->crtc == crtc) { + radeon_crtc->encoder = encoder; + radeon_crtc->connector = radeon_get_connector_for_encoder(encoder); + break; + } + } + if ((radeon_crtc->encoder == NULL) || (radeon_crtc->connector == NULL)) { + radeon_crtc->encoder = NULL; + radeon_crtc->connector = NULL; + return false; + } if (!radeon_crtc_scaling_mode_fixup(crtc, mode, adjusted_mode)) return false; + if (!atombios_crtc_prepare_pll(crtc, adjusted_mode)) + return false; + /* pick pll */ + radeon_crtc->pll_id = radeon_atom_pick_pll(crtc); + /* if we can't get a PPLL for a non-DP encoder, fail */ + if ((radeon_crtc->pll_id == ATOM_PPLL_INVALID) && + !ENCODER_MODE_IS_DP(atombios_get_encoder_mode(radeon_crtc->encoder))) + return false; + return true; } @@ -1638,8 +1827,6 @@ static void atombios_crtc_prepare(struct drm_crtc *crtc) struct radeon_device *rdev = dev->dev_private; radeon_crtc->in_mode_set = true; - /* pick pll */ - radeon_crtc->pll_id = radeon_atom_pick_pll(crtc); /* disable crtc pair power gating before programming */ if (ASIC_IS_DCE6(rdev)) @@ -1697,7 +1884,10 @@ static void atombios_crtc_disable(struct drm_crtc *crtc) break; } done: - radeon_crtc->pll_id = -1; + radeon_crtc->pll_id = ATOM_PPLL_INVALID; + radeon_crtc->adjusted_clock = 0; + radeon_crtc->encoder = NULL; + radeon_crtc->connector = NULL; } static const struct drm_crtc_helper_funcs atombios_helper_funcs = { @@ -1746,6 +1936,9 @@ void radeon_atombios_init_crtc(struct drm_device *dev, else radeon_crtc->crtc_offset = 0; } - radeon_crtc->pll_id = -1; + radeon_crtc->pll_id = ATOM_PPLL_INVALID; + radeon_crtc->adjusted_clock = 0; + radeon_crtc->encoder = NULL; + radeon_crtc->connector = NULL; drm_crtc_helper_add(&radeon_crtc->base, &atombios_helper_funcs); } diff --git a/drivers/gpu/drm/radeon/atombios_dp.c b/drivers/gpu/drm/radeon/atombios_dp.c index 3623b98ed3fe..d48224b0d6b4 100644 --- a/drivers/gpu/drm/radeon/atombios_dp.c +++ b/drivers/gpu/drm/radeon/atombios_dp.c @@ -653,9 +653,7 @@ static bool radeon_dp_get_link_status(struct radeon_connector *radeon_connector, return false; } - DRM_DEBUG_KMS("link status %02x %02x %02x %02x %02x %02x\n", - link_status[0], link_status[1], link_status[2], - link_status[3], link_status[4], link_status[5]); + DRM_DEBUG_KMS("link status %*ph\n", 6, link_status); return true; } diff --git a/drivers/gpu/drm/radeon/atombios_encoders.c b/drivers/gpu/drm/radeon/atombios_encoders.c index 6e8803a1170c..806cbcc94fdd 100644 --- a/drivers/gpu/drm/radeon/atombios_encoders.c +++ b/drivers/gpu/drm/radeon/atombios_encoders.c @@ -28,9 +28,251 @@ #include "radeon_drm.h" #include "radeon.h" #include "atom.h" +#include <linux/backlight.h> extern int atom_debug; +static u8 +radeon_atom_get_backlight_level_from_reg(struct radeon_device *rdev) +{ + u8 backlight_level; + u32 bios_2_scratch; + + if (rdev->family >= CHIP_R600) + bios_2_scratch = RREG32(R600_BIOS_2_SCRATCH); + else + bios_2_scratch = RREG32(RADEON_BIOS_2_SCRATCH); + + backlight_level = ((bios_2_scratch & ATOM_S2_CURRENT_BL_LEVEL_MASK) >> + ATOM_S2_CURRENT_BL_LEVEL_SHIFT); + + return backlight_level; +} + +static void +radeon_atom_set_backlight_level_to_reg(struct radeon_device *rdev, + u8 backlight_level) +{ + u32 bios_2_scratch; + + if (rdev->family >= CHIP_R600) + bios_2_scratch = RREG32(R600_BIOS_2_SCRATCH); + else + bios_2_scratch = RREG32(RADEON_BIOS_2_SCRATCH); + + bios_2_scratch &= ~ATOM_S2_CURRENT_BL_LEVEL_MASK; + bios_2_scratch |= ((backlight_level << ATOM_S2_CURRENT_BL_LEVEL_SHIFT) & + ATOM_S2_CURRENT_BL_LEVEL_MASK); + + if (rdev->family >= CHIP_R600) + WREG32(R600_BIOS_2_SCRATCH, bios_2_scratch); + else + WREG32(RADEON_BIOS_2_SCRATCH, bios_2_scratch); +} + +u8 +atombios_get_backlight_level(struct radeon_encoder *radeon_encoder) +{ + struct drm_device *dev = radeon_encoder->base.dev; + struct radeon_device *rdev = dev->dev_private; + + if (!(rdev->mode_info.firmware_flags & ATOM_BIOS_INFO_BL_CONTROLLED_BY_GPU)) + return 0; + + return radeon_atom_get_backlight_level_from_reg(rdev); +} + +void +atombios_set_backlight_level(struct radeon_encoder *radeon_encoder, u8 level) +{ + struct drm_encoder *encoder = &radeon_encoder->base; + struct drm_device *dev = radeon_encoder->base.dev; + struct radeon_device *rdev = dev->dev_private; + struct radeon_encoder_atom_dig *dig; + DISPLAY_DEVICE_OUTPUT_CONTROL_PS_ALLOCATION args; + int index; + + if (!(rdev->mode_info.firmware_flags & ATOM_BIOS_INFO_BL_CONTROLLED_BY_GPU)) + return; + + if ((radeon_encoder->devices & (ATOM_DEVICE_LCD_SUPPORT)) && + radeon_encoder->enc_priv) { + dig = radeon_encoder->enc_priv; + dig->backlight_level = level; + radeon_atom_set_backlight_level_to_reg(rdev, dig->backlight_level); + + switch (radeon_encoder->encoder_id) { + case ENCODER_OBJECT_ID_INTERNAL_LVDS: + case ENCODER_OBJECT_ID_INTERNAL_LVTM1: + index = GetIndexIntoMasterTable(COMMAND, LCD1OutputControl); + if (dig->backlight_level == 0) { + args.ucAction = ATOM_LCD_BLOFF; + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + } else { + args.ucAction = ATOM_LCD_BL_BRIGHTNESS_CONTROL; + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + args.ucAction = ATOM_LCD_BLON; + atom_execute_table(rdev->mode_info.atom_context, index, (uint32_t *)&args); + } + break; + case ENCODER_OBJECT_ID_INTERNAL_UNIPHY: + case ENCODER_OBJECT_ID_INTERNAL_KLDSCP_LVTMA: + case ENCODER_OBJECT_ID_INTERNAL_UNIPHY1: + case ENCODER_OBJECT_ID_INTERNAL_UNIPHY2: + if (dig->backlight_level == 0) + atombios_dig_transmitter_setup(encoder, ATOM_TRANSMITTER_ACTION_LCD_BLOFF, 0, 0); + else { + atombios_dig_transmitter_setup(encoder, ATOM_TRANSMITTER_ACTION_BL_BRIGHTNESS_CONTROL, 0, 0); + atombios_dig_transmitter_setup(encoder, ATOM_TRANSMITTER_ACTION_LCD_BLON, 0, 0); + } + break; + default: + break; + } + } +} + +#if defined(CONFIG_BACKLIGHT_CLASS_DEVICE) || defined(CONFIG_BACKLIGHT_CLASS_DEVICE_MODULE) + +static u8 radeon_atom_bl_level(struct backlight_device *bd) +{ + u8 level; + + /* Convert brightness to hardware level */ + if (bd->props.brightness < 0) + level = 0; + else if (bd->props.brightness > RADEON_MAX_BL_LEVEL) + level = RADEON_MAX_BL_LEVEL; + else + level = bd->props.brightness; + + return level; +} + +static int radeon_atom_backlight_update_status(struct backlight_device *bd) +{ + struct radeon_backlight_privdata *pdata = bl_get_data(bd); + struct radeon_encoder *radeon_encoder = pdata->encoder; + + atombios_set_backlight_level(radeon_encoder, radeon_atom_bl_level(bd)); + + return 0; +} + +static int radeon_atom_backlight_get_brightness(struct backlight_device *bd) +{ + struct radeon_backlight_privdata *pdata = bl_get_data(bd); + struct radeon_encoder *radeon_encoder = pdata->encoder; + struct drm_device *dev = radeon_encoder->base.dev; + struct radeon_device *rdev = dev->dev_private; + + return radeon_atom_get_backlight_level_from_reg(rdev); +} + +static const struct backlight_ops radeon_atom_backlight_ops = { + .get_brightness = radeon_atom_backlight_get_brightness, + .update_status = radeon_atom_backlight_update_status, +}; + +void radeon_atom_backlight_init(struct radeon_encoder *radeon_encoder, + struct drm_connector *drm_connector) +{ + struct drm_device *dev = radeon_encoder->base.dev; + struct radeon_device *rdev = dev->dev_private; + struct backlight_device *bd; + struct backlight_properties props; + struct radeon_backlight_privdata *pdata; + struct radeon_encoder_atom_dig *dig; + u8 backlight_level; + + if (!radeon_encoder->enc_priv) + return; + + if (!rdev->is_atom_bios) + return; + + if (!(rdev->mode_info.firmware_flags & ATOM_BIOS_INFO_BL_CONTROLLED_BY_GPU)) + return; + + pdata = kmalloc(sizeof(struct radeon_backlight_privdata), GFP_KERNEL); + if (!pdata) { + DRM_ERROR("Memory allocation failed\n"); + goto error; + } + + memset(&props, 0, sizeof(props)); + props.max_brightness = RADEON_MAX_BL_LEVEL; + props.type = BACKLIGHT_RAW; + bd = backlight_device_register("radeon_bl", &drm_connector->kdev, + pdata, &radeon_atom_backlight_ops, &props); + if (IS_ERR(bd)) { + DRM_ERROR("Backlight registration failed\n"); + goto error; + } + + pdata->encoder = radeon_encoder; + + backlight_level = radeon_atom_get_backlight_level_from_reg(rdev); + + dig = radeon_encoder->enc_priv; + dig->bl_dev = bd; + + bd->props.brightness = radeon_atom_backlight_get_brightness(bd); + bd->props.power = FB_BLANK_UNBLANK; + backlight_update_status(bd); + + DRM_INFO("radeon atom DIG backlight initialized\n"); + + return; + +error: + kfree(pdata); + return; +} + +static void radeon_atom_backlight_exit(struct radeon_encoder *radeon_encoder) +{ + struct drm_device *dev = radeon_encoder->base.dev; + struct radeon_device *rdev = dev->dev_private; + struct backlight_device *bd = NULL; + struct radeon_encoder_atom_dig *dig; + + if (!radeon_encoder->enc_priv) + return; + + if (!rdev->is_atom_bios) + return; + + if (!(rdev->mode_info.firmware_flags & ATOM_BIOS_INFO_BL_CONTROLLED_BY_GPU)) + return; + + dig = radeon_encoder->enc_priv; + bd = dig->bl_dev; + dig->bl_dev = NULL; + + if (bd) { + struct radeon_legacy_backlight_privdata *pdata; + + pdata = bl_get_data(bd); + backlight_device_unregister(bd); + kfree(pdata); + + DRM_INFO("radeon atom LVDS backlight unloaded\n"); + } +} + +#else /* !CONFIG_BACKLIGHT_CLASS_DEVICE */ + +void radeon_atom_backlight_init(struct radeon_encoder *encoder) +{ +} + +static void radeon_atom_backlight_exit(struct radeon_encoder *encoder) +{ +} + +#endif + /* evil but including atombios.h is much worse */ bool radeon_atom_get_tv_timings(struct radeon_device *rdev, int index, struct drm_display_mode *mode); @@ -209,6 +451,32 @@ atombios_tv_setup(struct drm_encoder *encoder, int action) } +static u8 radeon_atom_get_bpc(struct drm_encoder *encoder) +{ + struct drm_connector *connector = radeon_get_connector_for_encoder(encoder); + int bpc = 8; + + if (connector) + bpc = radeon_get_monitor_bpc(connector); + + switch (bpc) { + case 0: + return PANEL_BPC_UNDEFINE; + case 6: + return PANEL_6BIT_PER_COLOR; + case 8: + default: + return PANEL_8BIT_PER_COLOR; + case 10: + return PANEL_10BIT_PER_COLOR; + case 12: + return PANEL_12BIT_PER_COLOR; + case 16: + return PANEL_16BIT_PER_COLOR; + } +} + + union dvo_encoder_control { ENABLE_EXTERNAL_TMDS_ENCODER_PS_ALLOCATION ext_tmds; DVO_ENCODER_CONTROL_PS_ALLOCATION dvo; @@ -406,7 +674,8 @@ atombios_get_encoder_mode(struct drm_encoder *encoder) return ATOM_ENCODER_MODE_DP; /* DVO is always DVO */ - if (radeon_encoder->encoder_id == ATOM_ENCODER_MODE_DVO) + if ((radeon_encoder->encoder_id == ENCODER_OBJECT_ID_INTERNAL_DVO1) || + (radeon_encoder->encoder_id == ENCODER_OBJECT_ID_INTERNAL_KLDSCP_DVO1)) return ATOM_ENCODER_MODE_DVO; connector = radeon_get_connector_for_encoder(encoder); @@ -535,7 +804,6 @@ atombios_dig_encoder_setup(struct drm_encoder *encoder, int action, int panel_mo int dp_clock = 0; int dp_lane_count = 0; int hpd_id = RADEON_HPD_NONE; - int bpc = 8; if (connector) { struct radeon_connector *radeon_connector = to_radeon_connector(connector); @@ -545,7 +813,6 @@ atombios_dig_encoder_setup(struct drm_encoder *encoder, int action, int panel_mo dp_clock = dig_connector->dp_clock; dp_lane_count = dig_connector->dp_lane_count; hpd_id = radeon_connector->hpd.hpd; - bpc = radeon_get_monitor_bpc(connector); } /* no dig encoder assigned */ @@ -612,37 +879,17 @@ atombios_dig_encoder_setup(struct drm_encoder *encoder, int action, int panel_mo else args.v3.ucEncoderMode = atombios_get_encoder_mode(encoder); - if (ENCODER_MODE_IS_DP(args.v1.ucEncoderMode)) + if (ENCODER_MODE_IS_DP(args.v3.ucEncoderMode)) args.v3.ucLaneNum = dp_lane_count; else if (radeon_dig_monitor_is_duallink(encoder, radeon_encoder->pixel_clock)) args.v3.ucLaneNum = 8; else args.v3.ucLaneNum = 4; - if (ENCODER_MODE_IS_DP(args.v1.ucEncoderMode) && (dp_clock == 270000)) + if (ENCODER_MODE_IS_DP(args.v3.ucEncoderMode) && (dp_clock == 270000)) args.v1.ucConfig |= ATOM_ENCODER_CONFIG_V3_DPLINKRATE_2_70GHZ; args.v3.acConfig.ucDigSel = dig->dig_encoder; - switch (bpc) { - case 0: - args.v3.ucBitPerColor = PANEL_BPC_UNDEFINE; - break; - case 6: - args.v3.ucBitPerColor = PANEL_6BIT_PER_COLOR; - break; - case 8: - default: - args.v3.ucBitPerColor = PANEL_8BIT_PER_COLOR; - break; - case 10: - args.v3.ucBitPerColor = PANEL_10BIT_PER_COLOR; - break; - case 12: - args.v3.ucBitPerColor = PANEL_12BIT_PER_COLOR; - break; - case 16: - args.v3.ucBitPerColor = PANEL_16BIT_PER_COLOR; - break; - } + args.v3.ucBitPerColor = radeon_atom_get_bpc(encoder); break; case 4: args.v4.ucAction = action; @@ -652,41 +899,21 @@ atombios_dig_encoder_setup(struct drm_encoder *encoder, int action, int panel_mo else args.v4.ucEncoderMode = atombios_get_encoder_mode(encoder); - if (ENCODER_MODE_IS_DP(args.v1.ucEncoderMode)) + if (ENCODER_MODE_IS_DP(args.v4.ucEncoderMode)) args.v4.ucLaneNum = dp_lane_count; else if (radeon_dig_monitor_is_duallink(encoder, radeon_encoder->pixel_clock)) args.v4.ucLaneNum = 8; else args.v4.ucLaneNum = 4; - if (ENCODER_MODE_IS_DP(args.v1.ucEncoderMode)) { + if (ENCODER_MODE_IS_DP(args.v4.ucEncoderMode)) { if (dp_clock == 270000) args.v1.ucConfig |= ATOM_ENCODER_CONFIG_V4_DPLINKRATE_2_70GHZ; else if (dp_clock == 540000) args.v1.ucConfig |= ATOM_ENCODER_CONFIG_V4_DPLINKRATE_5_40GHZ; } args.v4.acConfig.ucDigSel = dig->dig_encoder; - switch (bpc) { - case 0: - args.v4.ucBitPerColor = PANEL_BPC_UNDEFINE; - break; - case 6: - args.v4.ucBitPerColor = PANEL_6BIT_PER_COLOR; - break; - case 8: - default: - args.v4.ucBitPerColor = PANEL_8BIT_PER_COLOR; - break; - case 10: - args.v4.ucBitPerColor = PANEL_10BIT_PER_COLOR; - break; - case 12: - args.v4.ucBitPerColor = PANEL_12BIT_PER_COLOR; - break; - case 16: - args.v4.ucBitPerColor = PANEL_16BIT_PER_COLOR; - break; - } + args.v4.ucBitPerColor = radeon_atom_get_bpc(encoder); if (hpd_id == RADEON_HPD_NONE) args.v4.ucHPD_ID = 0; else @@ -799,8 +1026,7 @@ atombios_dig_transmitter_setup(struct drm_encoder *encoder, int action, uint8_t args.v1.asMode.ucLaneSet = lane_set; } else { if (is_dp) - args.v1.usPixelClock = - cpu_to_le16(dp_clock / 10); + args.v1.usPixelClock = cpu_to_le16(dp_clock / 10); else if (radeon_dig_monitor_is_duallink(encoder, radeon_encoder->pixel_clock)) args.v1.usPixelClock = cpu_to_le16((radeon_encoder->pixel_clock / 2) / 10); else @@ -857,8 +1083,7 @@ atombios_dig_transmitter_setup(struct drm_encoder *encoder, int action, uint8_t args.v2.asMode.ucLaneSet = lane_set; } else { if (is_dp) - args.v2.usPixelClock = - cpu_to_le16(dp_clock / 10); + args.v2.usPixelClock = cpu_to_le16(dp_clock / 10); else if (radeon_dig_monitor_is_duallink(encoder, radeon_encoder->pixel_clock)) args.v2.usPixelClock = cpu_to_le16((radeon_encoder->pixel_clock / 2) / 10); else @@ -900,8 +1125,7 @@ atombios_dig_transmitter_setup(struct drm_encoder *encoder, int action, uint8_t args.v3.asMode.ucLaneSet = lane_set; } else { if (is_dp) - args.v3.usPixelClock = - cpu_to_le16(dp_clock / 10); + args.v3.usPixelClock = cpu_to_le16(dp_clock / 10); else if (radeon_dig_monitor_is_duallink(encoder, radeon_encoder->pixel_clock)) args.v3.usPixelClock = cpu_to_le16((radeon_encoder->pixel_clock / 2) / 10); else @@ -960,8 +1184,7 @@ atombios_dig_transmitter_setup(struct drm_encoder *encoder, int action, uint8_t args.v4.asMode.ucLaneSet = lane_set; } else { if (is_dp) - args.v4.usPixelClock = - cpu_to_le16(dp_clock / 10); + args.v4.usPixelClock = cpu_to_le16(dp_clock / 10); else if (radeon_dig_monitor_is_duallink(encoder, radeon_encoder->pixel_clock)) args.v4.usPixelClock = cpu_to_le16((radeon_encoder->pixel_clock / 2) / 10); else @@ -1147,7 +1370,6 @@ atombios_external_encoder_setup(struct drm_encoder *encoder, int dp_lane_count = 0; int connector_object_id = 0; u32 ext_enum = (ext_radeon_encoder->encoder_enum & ENUM_ID_MASK) >> ENUM_ID_SHIFT; - int bpc = 8; if (action == EXTERNAL_ENCODER_ACTION_V3_ENCODER_INIT) connector = radeon_get_connector_for_encoder_init(encoder); @@ -1163,7 +1385,6 @@ atombios_external_encoder_setup(struct drm_encoder *encoder, dp_lane_count = dig_connector->dp_lane_count; connector_object_id = (radeon_connector->connector_object_id & OBJECT_ID_MASK) >> OBJECT_ID_SHIFT; - bpc = radeon_get_monitor_bpc(connector); } memset(&args, 0, sizeof(args)); @@ -1221,27 +1442,7 @@ atombios_external_encoder_setup(struct drm_encoder *encoder, args.v3.sExtEncoder.ucConfig |= EXTERNAL_ENCODER_CONFIG_V3_ENCODER3; break; } - switch (bpc) { - case 0: - args.v3.sExtEncoder.ucBitPerColor = PANEL_BPC_UNDEFINE; - break; - case 6: - args.v3.sExtEncoder.ucBitPerColor = PANEL_6BIT_PER_COLOR; - break; - case 8: - default: - args.v3.sExtEncoder.ucBitPerColor = PANEL_8BIT_PER_COLOR; - break; - case 10: - args.v3.sExtEncoder.ucBitPerColor = PANEL_10BIT_PER_COLOR; - break; - case 12: - args.v3.sExtEncoder.ucBitPerColor = PANEL_12BIT_PER_COLOR; - break; - case 16: - args.v3.sExtEncoder.ucBitPerColor = PANEL_16BIT_PER_COLOR; - break; - } + args.v3.sExtEncoder.ucBitPerColor = radeon_atom_get_bpc(encoder); break; default: DRM_ERROR("Unknown table version: %d, %d\n", frev, crev); @@ -2286,6 +2487,8 @@ static const struct drm_encoder_helper_funcs radeon_atom_dac_helper_funcs = { void radeon_enc_destroy(struct drm_encoder *encoder) { struct radeon_encoder *radeon_encoder = to_radeon_encoder(encoder); + if (radeon_encoder->devices & (ATOM_DEVICE_LCD_SUPPORT)) + radeon_atom_backlight_exit(radeon_encoder); kfree(radeon_encoder->enc_priv); drm_encoder_cleanup(encoder); kfree(radeon_encoder); @@ -2295,7 +2498,7 @@ static const struct drm_encoder_funcs radeon_atom_enc_funcs = { .destroy = radeon_enc_destroy, }; -struct radeon_encoder_atom_dac * +static struct radeon_encoder_atom_dac * radeon_atombios_set_dac_info(struct radeon_encoder *radeon_encoder) { struct drm_device *dev = radeon_encoder->base.dev; @@ -2309,7 +2512,7 @@ radeon_atombios_set_dac_info(struct radeon_encoder *radeon_encoder) return dac; } -struct radeon_encoder_atom_dig * +static struct radeon_encoder_atom_dig * radeon_atombios_set_dig_info(struct radeon_encoder *radeon_encoder) { int encoder_enum = (radeon_encoder->encoder_enum & ENUM_ID_MASK) >> ENUM_ID_SHIFT; diff --git a/drivers/gpu/drm/radeon/evergreen.c b/drivers/gpu/drm/radeon/evergreen.c index e93b80a6d4e9..c4ded396b78d 100644 --- a/drivers/gpu/drm/radeon/evergreen.c +++ b/drivers/gpu/drm/radeon/evergreen.c @@ -37,6 +37,16 @@ #define EVERGREEN_PFP_UCODE_SIZE 1120 #define EVERGREEN_PM4_UCODE_SIZE 1376 +static const u32 crtc_offsets[6] = +{ + EVERGREEN_CRTC0_REGISTER_OFFSET, + EVERGREEN_CRTC1_REGISTER_OFFSET, + EVERGREEN_CRTC2_REGISTER_OFFSET, + EVERGREEN_CRTC3_REGISTER_OFFSET, + EVERGREEN_CRTC4_REGISTER_OFFSET, + EVERGREEN_CRTC5_REGISTER_OFFSET +}; + static void evergreen_gpu_init(struct radeon_device *rdev); void evergreen_fini(struct radeon_device *rdev); void evergreen_pcie_gen2_enable(struct radeon_device *rdev); @@ -109,17 +119,19 @@ void evergreen_fix_pci_max_read_req_size(struct radeon_device *rdev) */ void dce4_wait_for_vblank(struct radeon_device *rdev, int crtc) { - struct radeon_crtc *radeon_crtc = rdev->mode_info.crtcs[crtc]; int i; - if (RREG32(EVERGREEN_CRTC_CONTROL + radeon_crtc->crtc_offset) & EVERGREEN_CRTC_MASTER_EN) { + if (crtc >= rdev->num_crtc) + return; + + if (RREG32(EVERGREEN_CRTC_CONTROL + crtc_offsets[crtc]) & EVERGREEN_CRTC_MASTER_EN) { for (i = 0; i < rdev->usec_timeout; i++) { - if (!(RREG32(EVERGREEN_CRTC_STATUS + radeon_crtc->crtc_offset) & EVERGREEN_CRTC_V_BLANK)) + if (!(RREG32(EVERGREEN_CRTC_STATUS + crtc_offsets[crtc]) & EVERGREEN_CRTC_V_BLANK)) break; udelay(1); } for (i = 0; i < rdev->usec_timeout; i++) { - if (RREG32(EVERGREEN_CRTC_STATUS + radeon_crtc->crtc_offset) & EVERGREEN_CRTC_V_BLANK) + if (RREG32(EVERGREEN_CRTC_STATUS + crtc_offsets[crtc]) & EVERGREEN_CRTC_V_BLANK) break; udelay(1); } @@ -314,6 +326,64 @@ void sumo_pm_init_profile(struct radeon_device *rdev) } /** + * btc_pm_init_profile - Initialize power profiles callback. + * + * @rdev: radeon_device pointer + * + * Initialize the power states used in profile mode + * (BTC, cayman). + * Used for profile mode only. + */ +void btc_pm_init_profile(struct radeon_device *rdev) +{ + int idx; + + /* default */ + rdev->pm.profiles[PM_PROFILE_DEFAULT_IDX].dpms_off_ps_idx = rdev->pm.default_power_state_index; + rdev->pm.profiles[PM_PROFILE_DEFAULT_IDX].dpms_on_ps_idx = rdev->pm.default_power_state_index; + rdev->pm.profiles[PM_PROFILE_DEFAULT_IDX].dpms_off_cm_idx = 0; + rdev->pm.profiles[PM_PROFILE_DEFAULT_IDX].dpms_on_cm_idx = 2; + /* starting with BTC, there is one state that is used for both + * MH and SH. Difference is that we always use the high clock index for + * mclk. + */ + if (rdev->flags & RADEON_IS_MOBILITY) + idx = radeon_pm_get_type_index(rdev, POWER_STATE_TYPE_BATTERY, 0); + else + idx = radeon_pm_get_type_index(rdev, POWER_STATE_TYPE_PERFORMANCE, 0); + /* low sh */ + rdev->pm.profiles[PM_PROFILE_LOW_SH_IDX].dpms_off_ps_idx = idx; + rdev->pm.profiles[PM_PROFILE_LOW_SH_IDX].dpms_on_ps_idx = idx; + rdev->pm.profiles[PM_PROFILE_LOW_SH_IDX].dpms_off_cm_idx = 0; + rdev->pm.profiles[PM_PROFILE_LOW_SH_IDX].dpms_on_cm_idx = 0; + /* mid sh */ + rdev->pm.profiles[PM_PROFILE_MID_SH_IDX].dpms_off_ps_idx = idx; + rdev->pm.profiles[PM_PROFILE_MID_SH_IDX].dpms_on_ps_idx = idx; + rdev->pm.profiles[PM_PROFILE_MID_SH_IDX].dpms_off_cm_idx = 0; + rdev->pm.profiles[PM_PROFILE_MID_SH_IDX].dpms_on_cm_idx = 1; + /* high sh */ + rdev->pm.profiles[PM_PROFILE_HIGH_SH_IDX].dpms_off_ps_idx = idx; + rdev->pm.profiles[PM_PROFILE_HIGH_SH_IDX].dpms_on_ps_idx = idx; + rdev->pm.profiles[PM_PROFILE_HIGH_SH_IDX].dpms_off_cm_idx = 0; + rdev->pm.profiles[PM_PROFILE_HIGH_SH_IDX].dpms_on_cm_idx = 2; + /* low mh */ + rdev->pm.profiles[PM_PROFILE_LOW_MH_IDX].dpms_off_ps_idx = idx; + rdev->pm.profiles[PM_PROFILE_LOW_MH_IDX].dpms_on_ps_idx = idx; + rdev->pm.profiles[PM_PROFILE_LOW_MH_IDX].dpms_off_cm_idx = 0; + rdev->pm.profiles[PM_PROFILE_LOW_MH_IDX].dpms_on_cm_idx = 0; + /* mid mh */ + rdev->pm.profiles[PM_PROFILE_MID_MH_IDX].dpms_off_ps_idx = idx; + rdev->pm.profiles[PM_PROFILE_MID_MH_IDX].dpms_on_ps_idx = idx; + rdev->pm.profiles[PM_PROFILE_MID_MH_IDX].dpms_off_cm_idx = 0; + rdev->pm.profiles[PM_PROFILE_MID_MH_IDX].dpms_on_cm_idx = 1; + /* high mh */ + rdev->pm.profiles[PM_PROFILE_HIGH_MH_IDX].dpms_off_ps_idx = idx; + rdev->pm.profiles[PM_PROFILE_HIGH_MH_IDX].dpms_on_ps_idx = idx; + rdev->pm.profiles[PM_PROFILE_HIGH_MH_IDX].dpms_off_cm_idx = 0; + rdev->pm.profiles[PM_PROFILE_HIGH_MH_IDX].dpms_on_cm_idx = 2; +} + +/** * evergreen_pm_misc - set additional pm hw parameters callback. * * @rdev: radeon_device pointer @@ -1109,7 +1179,7 @@ void evergreen_pcie_gart_tlb_flush(struct radeon_device *rdev) } } -int evergreen_pcie_gart_enable(struct radeon_device *rdev) +static int evergreen_pcie_gart_enable(struct radeon_device *rdev) { u32 tmp; int r; @@ -1168,7 +1238,7 @@ int evergreen_pcie_gart_enable(struct radeon_device *rdev) return 0; } -void evergreen_pcie_gart_disable(struct radeon_device *rdev) +static void evergreen_pcie_gart_disable(struct radeon_device *rdev) { u32 tmp; @@ -1193,7 +1263,7 @@ void evergreen_pcie_gart_disable(struct radeon_device *rdev) radeon_gart_table_vram_unpin(rdev); } -void evergreen_pcie_gart_fini(struct radeon_device *rdev) +static void evergreen_pcie_gart_fini(struct radeon_device *rdev) { evergreen_pcie_gart_disable(rdev); radeon_gart_table_vram_free(rdev); @@ -1201,7 +1271,7 @@ void evergreen_pcie_gart_fini(struct radeon_device *rdev) } -void evergreen_agp_enable(struct radeon_device *rdev) +static void evergreen_agp_enable(struct radeon_device *rdev) { u32 tmp; @@ -1229,116 +1299,103 @@ void evergreen_agp_enable(struct radeon_device *rdev) void evergreen_mc_stop(struct radeon_device *rdev, struct evergreen_mc_save *save) { + u32 crtc_enabled, tmp, frame_count, blackout; + int i, j; + save->vga_render_control = RREG32(VGA_RENDER_CONTROL); save->vga_hdp_control = RREG32(VGA_HDP_CONTROL); - /* Stop all video */ + /* disable VGA render */ WREG32(VGA_RENDER_CONTROL, 0); - WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC0_REGISTER_OFFSET, 1); - WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC1_REGISTER_OFFSET, 1); - if (rdev->num_crtc >= 4) { - WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC2_REGISTER_OFFSET, 1); - WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC3_REGISTER_OFFSET, 1); - } - if (rdev->num_crtc >= 6) { - WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC4_REGISTER_OFFSET, 1); - WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC5_REGISTER_OFFSET, 1); - } - WREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC0_REGISTER_OFFSET, 0); - WREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC1_REGISTER_OFFSET, 0); - if (rdev->num_crtc >= 4) { - WREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC2_REGISTER_OFFSET, 0); - WREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC3_REGISTER_OFFSET, 0); - } - if (rdev->num_crtc >= 6) { - WREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC4_REGISTER_OFFSET, 0); - WREG32(EVERGREEN_CRTC_CONTROL + EVERGREEN_CRTC5_REGISTER_OFFSET, 0); - } - WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC0_REGISTER_OFFSET, 0); - WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC1_REGISTER_OFFSET, 0); - if (rdev->num_crtc >= 4) { - WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC2_REGISTER_OFFSET, 0); - WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC3_REGISTER_OFFSET, 0); - } - if (rdev->num_crtc >= 6) { - WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC4_REGISTER_OFFSET, 0); - WREG32(EVERGREEN_CRTC_UPDATE_LOCK + EVERGREEN_CRTC5_REGISTER_OFFSET, 0); + /* blank the display controllers */ + for (i = 0; i < rdev->num_crtc; i++) { + crtc_enabled = RREG32(EVERGREEN_CRTC_CONTROL + crtc_offsets[i]) & EVERGREEN_CRTC_MASTER_EN; + if (crtc_enabled) { + save->crtc_enabled[i] = true; + if (ASIC_IS_DCE6(rdev)) { + tmp = RREG32(EVERGREEN_CRTC_BLANK_CONTROL + crtc_offsets[i]); + if (!(tmp & EVERGREEN_CRTC_BLANK_DATA_EN)) { + radeon_wait_for_vblank(rdev, i); + tmp |= EVERGREEN_CRTC_BLANK_DATA_EN; + WREG32(EVERGREEN_CRTC_BLANK_CONTROL + crtc_offsets[i], tmp); + } + } else { + tmp = RREG32(EVERGREEN_CRTC_CONTROL + crtc_offsets[i]); + if (!(tmp & EVERGREEN_CRTC_DISP_READ_REQUEST_DISABLE)) { + radeon_wait_for_vblank(rdev, i); + tmp |= EVERGREEN_CRTC_DISP_READ_REQUEST_DISABLE; + WREG32(EVERGREEN_CRTC_CONTROL + crtc_offsets[i], tmp); + } + } + /* wait for the next frame */ + frame_count = radeon_get_vblank_counter(rdev, i); + for (j = 0; j < rdev->usec_timeout; j++) { + if (radeon_get_vblank_counter(rdev, i) != frame_count) + break; + udelay(1); + } + } } - WREG32(D1VGA_CONTROL, 0); - WREG32(D2VGA_CONTROL, 0); - if (rdev->num_crtc >= 4) { - WREG32(EVERGREEN_D3VGA_CONTROL, 0); - WREG32(EVERGREEN_D4VGA_CONTROL, 0); - } - if (rdev->num_crtc >= 6) { - WREG32(EVERGREEN_D5VGA_CONTROL, 0); - WREG32(EVERGREEN_D6VGA_CONTROL, 0); + radeon_mc_wait_for_idle(rdev); + + blackout = RREG32(MC_SHARED_BLACKOUT_CNTL); + if ((blackout & BLACKOUT_MODE_MASK) != 1) { + /* Block CPU access */ + WREG32(BIF_FB_EN, 0); + /* blackout the MC */ + blackout &= ~BLACKOUT_MODE_MASK; + WREG32(MC_SHARED_BLACKOUT_CNTL, blackout | 1); } } void evergreen_mc_resume(struct radeon_device *rdev, struct evergreen_mc_save *save) { - WREG32(EVERGREEN_GRPH_PRIMARY_SURFACE_ADDRESS_HIGH + EVERGREEN_CRTC0_REGISTER_OFFSET, - upper_32_bits(rdev->mc.vram_start)); - WREG32(EVERGREEN_GRPH_SECONDARY_SURFACE_ADDRESS_HIGH + EVERGREEN_CRTC0_REGISTER_OFFSET, - upper_32_bits(rdev->mc.vram_start)); - WREG32(EVERGREEN_GRPH_PRIMARY_SURFACE_ADDRESS + EVERGREEN_CRTC0_REGISTER_OFFSET, - (u32)rdev->mc.vram_start); - WREG32(EVERGREEN_GRPH_SECONDARY_SURFACE_ADDRESS + EVERGREEN_CRTC0_REGISTER_OFFSET, - (u32)rdev->mc.vram_start); - - WREG32(EVERGREEN_GRPH_PRIMARY_SURFACE_ADDRESS_HIGH + EVERGREEN_CRTC1_REGISTER_OFFSET, - upper_32_bits(rdev->mc.vram_start)); - WREG32(EVERGREEN_GRPH_SECONDARY_SURFACE_ADDRESS_HIGH + EVERGREEN_CRTC1_REGISTER_OFFSET, - upper_32_bits(rdev->mc.vram_start)); - WREG32(EVERGREEN_GRPH_PRIMARY_SURFACE_ADDRESS + EVERGREEN_CRTC1_REGISTER_OFFSET, - (u32)rdev->mc.vram_start); - WREG32(EVERGREEN_GRPH_SECONDARY_SURFACE_ADDRESS + EVERGREEN_CRTC1_REGISTER_OFFSET, - (u32)rdev->mc.vram_start); - - if (rdev->num_crtc >= 4) { - WREG32(EVERGREEN_GRPH_PRIMARY_SURFACE_ADDRESS_HIGH + EVERGREEN_CRTC2_REGISTER_OFFSET, - upper_32_bits(rdev->mc.vram_start)); - WREG32(EVERGREEN_GRPH_SECONDARY_SURFACE_ADDRESS_HIGH + EVERGREEN_CRTC2_REGISTER_OFFSET, - upper_32_bits(rdev->mc.vram_start)); - WREG32(EVERGREEN_GRPH_PRIMARY_SURFACE_ADDRESS + EVERGREEN_CRTC2_REGISTER_OFFSET, - (u32)rdev->mc.vram_start); - WREG32(EVERGREEN_GRPH_SECONDARY_SURFACE_ADDRESS + EVERGREEN_CRTC2_REGISTER_OFFSET, - (u32)rdev->mc.vram_start); - - WREG32(EVERGREEN_GRPH_PRIMARY_SURFACE_ADDRESS_HIGH + EVERGREEN_CRTC3_REGISTER_OFFSET, - upper_32_bits(rdev->mc.vram_start)); - WREG32(EVERGREEN_GRPH_SECONDARY_SURFACE_ADDRESS_HIGH + EVERGREEN_CRTC3_REGISTER_OFFSET, - upper_32_bits(rdev->mc.vram_start)); - WREG32(EVERGREEN_GRPH_PRIMARY_SURFACE_ADDRESS + EVERGREEN_CRTC3_REGISTER_OFFSET, - (u32)rdev->mc.vram_start); - WREG32(EVERGREEN_GRPH_SECONDARY_SURFACE_ADDRESS + EVERGREEN_CRTC3_REGISTER_OFFSET, - (u32)rdev->mc.vram_start); - } - if (rdev->num_crtc >= 6) { - WREG32(EVERGREEN_GRPH_PRIMARY_SURFACE_ADDRESS_HIGH + EVERGREEN_CRTC4_REGISTER_OFFSET, - upper_32_bits(rdev->mc.vram_start)); - WREG32(EVERGREEN_GRPH_SECONDARY_SURFACE_ADDRESS_HIGH + EVERGREEN_CRTC4_REGISTER_OFFSET, - upper_32_bits(rdev->mc.vram_start)); - WREG32(EVERGREEN_GRPH_PRIMARY_SURFACE_ADDRESS + EVERGREEN_CRTC4_REGISTER_OFFSET, - (u32)rdev->mc.vram_start); - WREG32(EVERGREEN_GRPH_SECONDARY_SURFACE_ADDRESS + EVERGREEN_CRTC4_REGISTER_OFFSET, - (u32)rdev->mc.vram_start); + u32 tmp, frame_count; + int i, j; - WREG32(EVERGREEN_GRPH_PRIMARY_SURFACE_ADDRESS_HIGH + EVERGREEN_CRTC5_REGISTER_OFFSET, + /* update crtc base addresses */ + for (i = 0; i < rdev->num_crtc; i++) { + WREG32(EVERGREEN_GRPH_PRIMARY_SURFACE_ADDRESS_HIGH + crtc_offsets[i], upper_32_bits(rdev->mc.vram_start)); - WREG32(EVERGREEN_GRPH_SECONDARY_SURFACE_ADDRESS_HIGH + EVERGREEN_CRTC5_REGISTER_OFFSET, + WREG32(EVERGREEN_GRPH_SECONDARY_SURFACE_ADDRESS_HIGH + crtc_offsets[i], upper_32_bits(rdev->mc.vram_start)); - WREG32(EVERGREEN_GRPH_PRIMARY_SURFACE_ADDRESS + EVERGREEN_CRTC5_REGISTER_OFFSET, + WREG32(EVERGREEN_GRPH_PRIMARY_SURFACE_ADDRESS + crtc_offsets[i], (u32)rdev->mc.vram_start); - WREG32(EVERGREEN_GRPH_SECONDARY_SURFACE_ADDRESS + EVERGREEN_CRTC5_REGISTER_OFFSET, + WREG32(EVERGREEN_GRPH_SECONDARY_SURFACE_ADDRESS + crtc_offsets[i], (u32)rdev->mc.vram_start); } - WREG32(EVERGREEN_VGA_MEMORY_BASE_ADDRESS_HIGH, upper_32_bits(rdev->mc.vram_start)); WREG32(EVERGREEN_VGA_MEMORY_BASE_ADDRESS, (u32)rdev->mc.vram_start); - /* Unlock host access */ + + /* unblackout the MC */ + tmp = RREG32(MC_SHARED_BLACKOUT_CNTL); + tmp &= ~BLACKOUT_MODE_MASK; + WREG32(MC_SHARED_BLACKOUT_CNTL, tmp); + /* allow CPU access */ + WREG32(BIF_FB_EN, FB_READ_EN | FB_WRITE_EN); + + for (i = 0; i < rdev->num_crtc; i++) { + if (save->crtc_enabled) { + if (ASIC_IS_DCE6(rdev)) { + tmp = RREG32(EVERGREEN_CRTC_BLANK_CONTROL + crtc_offsets[i]); + tmp |= EVERGREEN_CRTC_BLANK_DATA_EN; + WREG32(EVERGREEN_CRTC_BLANK_CONTROL + crtc_offsets[i], tmp); + } else { + tmp = RREG32(EVERGREEN_CRTC_CONTROL + crtc_offsets[i]); + tmp &= ~EVERGREEN_CRTC_DISP_READ_REQUEST_DISABLE; + WREG32(EVERGREEN_CRTC_CONTROL + crtc_offsets[i], tmp); + } + /* wait for the next frame */ + frame_count = radeon_get_vblank_counter(rdev, i); + for (j = 0; j < rdev->usec_timeout; j++) { + if (radeon_get_vblank_counter(rdev, i) != frame_count) + break; + udelay(1); + } + } + } + /* Unlock vga access */ WREG32(VGA_HDP_CONTROL, save->vga_hdp_control); mdelay(1); WREG32(VGA_RENDER_CONTROL, save->vga_render_control); @@ -1557,7 +1614,7 @@ static int evergreen_cp_start(struct radeon_device *rdev) return 0; } -int evergreen_cp_resume(struct radeon_device *rdev) +static int evergreen_cp_resume(struct radeon_device *rdev) { struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX]; u32 tmp; @@ -2333,22 +2390,10 @@ int evergreen_asic_reset(struct radeon_device *rdev) u32 evergreen_get_vblank_counter(struct radeon_device *rdev, int crtc) { - switch (crtc) { - case 0: - return RREG32(CRTC_STATUS_FRAME_COUNT + EVERGREEN_CRTC0_REGISTER_OFFSET); - case 1: - return RREG32(CRTC_STATUS_FRAME_COUNT + EVERGREEN_CRTC1_REGISTER_OFFSET); - case 2: - return RREG32(CRTC_STATUS_FRAME_COUNT + EVERGREEN_CRTC2_REGISTER_OFFSET); - case 3: - return RREG32(CRTC_STATUS_FRAME_COUNT + EVERGREEN_CRTC3_REGISTER_OFFSET); - case 4: - return RREG32(CRTC_STATUS_FRAME_COUNT + EVERGREEN_CRTC4_REGISTER_OFFSET); - case 5: - return RREG32(CRTC_STATUS_FRAME_COUNT + EVERGREEN_CRTC5_REGISTER_OFFSET); - default: + if (crtc >= rdev->num_crtc) return 0; - } + else + return RREG32(CRTC_STATUS_FRAME_COUNT + crtc_offsets[crtc]); } void evergreen_disable_interrupt_state(struct radeon_device *rdev) @@ -2541,10 +2586,6 @@ int evergreen_irq_set(struct radeon_device *rdev) DRM_DEBUG("evergreen_irq_set: hdmi 5\n"); afmt6 |= AFMT_AZ_FORMAT_WTRIG_MASK; } - if (rdev->irq.gui_idle) { - DRM_DEBUG("gui idle\n"); - grbm_int_cntl |= GUI_IDLE_INT_ENABLE; - } if (rdev->family >= CHIP_CAYMAN) { cayman_cp_int_cntl_setup(rdev, 0, cp_int_cntl); @@ -2726,7 +2767,7 @@ static void evergreen_irq_ack(struct radeon_device *rdev) } } -void evergreen_irq_disable(struct radeon_device *rdev) +static void evergreen_irq_disable(struct radeon_device *rdev) { r600_disable_interrupts(rdev); /* Wait and acknowledge irq */ @@ -3079,7 +3120,6 @@ restart_ih: break; case 233: /* GUI IDLE */ DRM_DEBUG("IH: GUI idle\n"); - wake_up(&rdev->irq.idle_queue); break; default: DRM_DEBUG("Unhandled interrupt: %d %d\n", src_id, src_data); diff --git a/drivers/gpu/drm/radeon/evergreen_cs.c b/drivers/gpu/drm/radeon/evergreen_cs.c index e44a62a07fe3..2ceab2b52d69 100644 --- a/drivers/gpu/drm/radeon/evergreen_cs.c +++ b/drivers/gpu/drm/radeon/evergreen_cs.c @@ -846,6 +846,16 @@ static int evergreen_cs_track_validate_texture(struct radeon_cs_parser *p, return -EINVAL; } + if (!mipmap) { + if (llevel) { + dev_warn(p->dev, "%s:%i got NULL MIP_ADDRESS relocation\n", + __func__, __LINE__); + return -EINVAL; + } else { + return 0; /* everything's ok */ + } + } + /* check mipmap size */ for (i = 1; i <= llevel; i++) { unsigned w, h, d; @@ -995,7 +1005,7 @@ static int evergreen_cs_track_check(struct radeon_cs_parser *p) * Assume that chunk_ib_index is properly set. Will return -EINVAL * if packet is bigger than remaining ib size. or if packets is unknown. **/ -int evergreen_cs_packet_parse(struct radeon_cs_parser *p, +static int evergreen_cs_packet_parse(struct radeon_cs_parser *p, struct radeon_cs_packet *pkt, unsigned idx) { @@ -1081,6 +1091,27 @@ static int evergreen_cs_packet_next_reloc(struct radeon_cs_parser *p, } /** + * evergreen_cs_packet_next_is_pkt3_nop() - test if the next packet is NOP + * @p: structure holding the parser context. + * + * Check if the next packet is a relocation packet3. + **/ +static bool evergreen_cs_packet_next_is_pkt3_nop(struct radeon_cs_parser *p) +{ + struct radeon_cs_packet p3reloc; + int r; + + r = evergreen_cs_packet_parse(p, &p3reloc, p->idx); + if (r) { + return false; + } + if (p3reloc.type != PACKET_TYPE3 || p3reloc.opcode != PACKET3_NOP) { + return false; + } + return true; +} + +/** * evergreen_cs_packet_next_vline() - parse userspace VLINE packet * @parser: parser structure holding parsing context. * @@ -2330,7 +2361,7 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p, for (i = 0; i < (pkt->count / 8); i++) { struct radeon_bo *texture, *mipmap; u32 toffset, moffset; - u32 size, offset; + u32 size, offset, mip_address, tex_dim; switch (G__SQ_CONSTANT_TYPE(radeon_get_ib_value(p, idx+1+(i*8)+7))) { case SQ_TEX_VTX_VALID_TEXTURE: @@ -2359,14 +2390,28 @@ static int evergreen_packet3_check(struct radeon_cs_parser *p, } texture = reloc->robj; toffset = (u32)((reloc->lobj.gpu_offset >> 8) & 0xffffffff); + /* tex mip base */ - r = evergreen_cs_packet_next_reloc(p, &reloc); - if (r) { - DRM_ERROR("bad SET_RESOURCE (tex)\n"); - return -EINVAL; + tex_dim = ib[idx+1+(i*8)+0] & 0x7; + mip_address = ib[idx+1+(i*8)+3]; + + if ((tex_dim == SQ_TEX_DIM_2D_MSAA || tex_dim == SQ_TEX_DIM_2D_ARRAY_MSAA) && + !mip_address && + !evergreen_cs_packet_next_is_pkt3_nop(p)) { + /* MIP_ADDRESS should point to FMASK for an MSAA texture. + * It should be 0 if FMASK is disabled. */ + moffset = 0; + mipmap = NULL; + } else { + r = evergreen_cs_packet_next_reloc(p, &reloc); + if (r) { + DRM_ERROR("bad SET_RESOURCE (tex)\n"); + return -EINVAL; + } + moffset = (u32)((reloc->lobj.gpu_offset >> 8) & 0xffffffff); + mipmap = reloc->robj; } - moffset = (u32)((reloc->lobj.gpu_offset >> 8) & 0xffffffff); - mipmap = reloc->robj; + r = evergreen_cs_track_validate_texture(p, texture, mipmap, idx+1+(i*8)); if (r) return r; diff --git a/drivers/gpu/drm/radeon/evergreen_reg.h b/drivers/gpu/drm/radeon/evergreen_reg.h index 8beac1065025..034f4c22e5db 100644 --- a/drivers/gpu/drm/radeon/evergreen_reg.h +++ b/drivers/gpu/drm/radeon/evergreen_reg.h @@ -218,6 +218,8 @@ #define EVERGREEN_CRTC_CONTROL 0x6e70 # define EVERGREEN_CRTC_MASTER_EN (1 << 0) # define EVERGREEN_CRTC_DISP_READ_REQUEST_DISABLE (1 << 24) +#define EVERGREEN_CRTC_BLANK_CONTROL 0x6e74 +# define EVERGREEN_CRTC_BLANK_DATA_EN (1 << 8) #define EVERGREEN_CRTC_STATUS 0x6e8c # define EVERGREEN_CRTC_V_BLANK (1 << 0) #define EVERGREEN_CRTC_STATUS_POSITION 0x6e90 diff --git a/drivers/gpu/drm/radeon/evergreend.h b/drivers/gpu/drm/radeon/evergreend.h index 79347855d9bf..df542f1a5dfb 100644 --- a/drivers/gpu/drm/radeon/evergreend.h +++ b/drivers/gpu/drm/radeon/evergreend.h @@ -87,6 +87,10 @@ #define CONFIG_MEMSIZE 0x5428 +#define BIF_FB_EN 0x5490 +#define FB_READ_EN (1 << 0) +#define FB_WRITE_EN (1 << 1) + #define CP_COHER_BASE 0x85F8 #define CP_STALLED_STAT1 0x8674 #define CP_STALLED_STAT2 0x8678 @@ -430,6 +434,9 @@ #define NOOFCHAN_MASK 0x00003000 #define MC_SHARED_CHREMAP 0x2008 +#define MC_SHARED_BLACKOUT_CNTL 0x20ac +#define BLACKOUT_MODE_MASK 0x00000007 + #define MC_ARB_RAMCFG 0x2760 #define NOOFBANK_SHIFT 0 #define NOOFBANK_MASK 0x00000003 diff --git a/drivers/gpu/drm/radeon/ni.c b/drivers/gpu/drm/radeon/ni.c index 853800e8582f..9a46f7d4e61f 100644 --- a/drivers/gpu/drm/radeon/ni.c +++ b/drivers/gpu/drm/radeon/ni.c @@ -726,7 +726,7 @@ void cayman_pcie_gart_tlb_flush(struct radeon_device *rdev) WREG32(VM_INVALIDATE_REQUEST, 1); } -int cayman_pcie_gart_enable(struct radeon_device *rdev) +static int cayman_pcie_gart_enable(struct radeon_device *rdev) { int i, r; @@ -782,7 +782,7 @@ int cayman_pcie_gart_enable(struct radeon_device *rdev) (u32)(rdev->dummy_page.addr >> 12)); WREG32(VM_CONTEXT1_CNTL2, 0); WREG32(VM_CONTEXT1_CNTL, 0); - WREG32(VM_CONTEXT1_CNTL, ENABLE_CONTEXT | PAGE_TABLE_DEPTH(0) | + WREG32(VM_CONTEXT1_CNTL, ENABLE_CONTEXT | PAGE_TABLE_DEPTH(1) | RANGE_PROTECTION_FAULT_ENABLE_DEFAULT); cayman_pcie_gart_tlb_flush(rdev); @@ -793,7 +793,7 @@ int cayman_pcie_gart_enable(struct radeon_device *rdev) return 0; } -void cayman_pcie_gart_disable(struct radeon_device *rdev) +static void cayman_pcie_gart_disable(struct radeon_device *rdev) { /* Disable all tables */ WREG32(VM_CONTEXT0_CNTL, 0); @@ -813,7 +813,7 @@ void cayman_pcie_gart_disable(struct radeon_device *rdev) radeon_gart_table_vram_unpin(rdev); } -void cayman_pcie_gart_fini(struct radeon_device *rdev) +static void cayman_pcie_gart_fini(struct radeon_device *rdev) { cayman_pcie_gart_disable(rdev); radeon_gart_table_vram_free(rdev); @@ -879,12 +879,13 @@ void cayman_ring_ib_execute(struct radeon_device *rdev, struct radeon_ib *ib) #endif (ib->gpu_addr & 0xFFFFFFFC)); radeon_ring_write(ring, upper_32_bits(ib->gpu_addr) & 0xFF); - radeon_ring_write(ring, ib->length_dw | (ib->vm_id << 24)); + radeon_ring_write(ring, ib->length_dw | + (ib->vm ? (ib->vm->id << 24) : 0)); /* flush read cache over gart for this vmid */ radeon_ring_write(ring, PACKET3(PACKET3_SET_CONFIG_REG, 1)); radeon_ring_write(ring, (CP_COHER_CNTL2 - PACKET3_SET_CONFIG_REG_START) >> 2); - radeon_ring_write(ring, ib->vm_id); + radeon_ring_write(ring, ib->vm ? ib->vm->id : 0); radeon_ring_write(ring, PACKET3(PACKET3_SURFACE_SYNC, 3)); radeon_ring_write(ring, PACKET3_TC_ACTION_ENA | PACKET3_SH_ACTION_ENA); radeon_ring_write(ring, 0xFFFFFFFF); @@ -1004,7 +1005,7 @@ static void cayman_cp_fini(struct radeon_device *rdev) radeon_scratch_free(rdev, ring->rptr_save_reg); } -int cayman_cp_resume(struct radeon_device *rdev) +static int cayman_cp_resume(struct radeon_device *rdev) { static const int ridx[] = { RADEON_RING_TYPE_GFX_INDEX, @@ -1496,53 +1497,16 @@ void cayman_vm_fini(struct radeon_device *rdev) { } -int cayman_vm_bind(struct radeon_device *rdev, struct radeon_vm *vm, int id) -{ - WREG32(VM_CONTEXT0_PAGE_TABLE_START_ADDR + (id << 2), 0); - WREG32(VM_CONTEXT0_PAGE_TABLE_END_ADDR + (id << 2), vm->last_pfn); - WREG32(VM_CONTEXT0_PAGE_TABLE_BASE_ADDR + (id << 2), vm->pt_gpu_addr >> 12); - /* flush hdp cache */ - WREG32(HDP_MEM_COHERENCY_FLUSH_CNTL, 0x1); - /* bits 0-7 are the VM contexts0-7 */ - WREG32(VM_INVALIDATE_REQUEST, 1 << id); - return 0; -} - -void cayman_vm_unbind(struct radeon_device *rdev, struct radeon_vm *vm) -{ - WREG32(VM_CONTEXT0_PAGE_TABLE_START_ADDR + (vm->id << 2), 0); - WREG32(VM_CONTEXT0_PAGE_TABLE_END_ADDR + (vm->id << 2), 0); - WREG32(VM_CONTEXT0_PAGE_TABLE_BASE_ADDR + (vm->id << 2), 0); - /* flush hdp cache */ - WREG32(HDP_MEM_COHERENCY_FLUSH_CNTL, 0x1); - /* bits 0-7 are the VM contexts0-7 */ - WREG32(VM_INVALIDATE_REQUEST, 1 << vm->id); -} - -void cayman_vm_tlb_flush(struct radeon_device *rdev, struct radeon_vm *vm) -{ - if (vm->id == -1) - return; - - /* flush hdp cache */ - WREG32(HDP_MEM_COHERENCY_FLUSH_CNTL, 0x1); - /* bits 0-7 are the VM contexts0-7 */ - WREG32(VM_INVALIDATE_REQUEST, 1 << vm->id); -} - -#define R600_PTE_VALID (1 << 0) +#define R600_ENTRY_VALID (1 << 0) #define R600_PTE_SYSTEM (1 << 1) #define R600_PTE_SNOOPED (1 << 2) #define R600_PTE_READABLE (1 << 5) #define R600_PTE_WRITEABLE (1 << 6) -uint32_t cayman_vm_page_flags(struct radeon_device *rdev, - struct radeon_vm *vm, - uint32_t flags) +uint32_t cayman_vm_page_flags(struct radeon_device *rdev, uint32_t flags) { uint32_t r600_flags = 0; - - r600_flags |= (flags & RADEON_VM_PAGE_VALID) ? R600_PTE_VALID : 0; + r600_flags |= (flags & RADEON_VM_PAGE_VALID) ? R600_ENTRY_VALID : 0; r600_flags |= (flags & RADEON_VM_PAGE_READABLE) ? R600_PTE_READABLE : 0; r600_flags |= (flags & RADEON_VM_PAGE_WRITEABLE) ? R600_PTE_WRITEABLE : 0; if (flags & RADEON_VM_PAGE_SYSTEM) { @@ -1552,12 +1516,76 @@ uint32_t cayman_vm_page_flags(struct radeon_device *rdev, return r600_flags; } -void cayman_vm_set_page(struct radeon_device *rdev, struct radeon_vm *vm, - unsigned pfn, uint64_t addr, uint32_t flags) +/** + * cayman_vm_set_page - update the page tables using the CP + * + * @rdev: radeon_device pointer + * @pe: addr of the page entry + * @addr: dst addr to write into pe + * @count: number of page entries to update + * @incr: increase next addr by incr bytes + * @flags: access flags + * + * Update the page tables using the CP (cayman-si). + */ +void cayman_vm_set_page(struct radeon_device *rdev, uint64_t pe, + uint64_t addr, unsigned count, + uint32_t incr, uint32_t flags) +{ + struct radeon_ring *ring = &rdev->ring[rdev->asic->vm.pt_ring_index]; + uint32_t r600_flags = cayman_vm_page_flags(rdev, flags); + int i; + + radeon_ring_write(ring, PACKET3(PACKET3_ME_WRITE, 1 + count * 2)); + radeon_ring_write(ring, pe); + radeon_ring_write(ring, upper_32_bits(pe) & 0xff); + for (i = 0; i < count; ++i) { + uint64_t value = 0; + if (flags & RADEON_VM_PAGE_SYSTEM) { + value = radeon_vm_map_gart(rdev, addr); + value &= 0xFFFFFFFFFFFFF000ULL; + addr += incr; + + } else if (flags & RADEON_VM_PAGE_VALID) { + value = addr; + addr += incr; + } + + value |= r600_flags; + radeon_ring_write(ring, value); + radeon_ring_write(ring, upper_32_bits(value)); + } +} + +/** + * cayman_vm_flush - vm flush using the CP + * + * @rdev: radeon_device pointer + * + * Update the page table base and flush the VM TLB + * using the CP (cayman-si). + */ +void cayman_vm_flush(struct radeon_device *rdev, int ridx, struct radeon_vm *vm) { - void __iomem *ptr = (void *)vm->pt; + struct radeon_ring *ring = &rdev->ring[ridx]; + + if (vm == NULL) + return; + + radeon_ring_write(ring, PACKET0(VM_CONTEXT0_PAGE_TABLE_START_ADDR + (vm->id << 2), 0)); + radeon_ring_write(ring, 0); + + radeon_ring_write(ring, PACKET0(VM_CONTEXT0_PAGE_TABLE_END_ADDR + (vm->id << 2), 0)); + radeon_ring_write(ring, vm->last_pfn); - addr = addr & 0xFFFFFFFFFFFFF000ULL; - addr |= flags; - writeq(addr, ptr + (pfn * 8)); + radeon_ring_write(ring, PACKET0(VM_CONTEXT0_PAGE_TABLE_BASE_ADDR + (vm->id << 2), 0)); + radeon_ring_write(ring, vm->pd_gpu_addr >> 12); + + /* flush hdp cache */ + radeon_ring_write(ring, PACKET0(HDP_MEM_COHERENCY_FLUSH_CNTL, 0)); + radeon_ring_write(ring, 0x1); + + /* bits 0-7 are the VM contexts0-7 */ + radeon_ring_write(ring, PACKET0(VM_INVALIDATE_REQUEST, 0)); + radeon_ring_write(ring, 1 << vm->id); } diff --git a/drivers/gpu/drm/radeon/nid.h b/drivers/gpu/drm/radeon/nid.h index 870db340d377..2423d1b5d385 100644 --- a/drivers/gpu/drm/radeon/nid.h +++ b/drivers/gpu/drm/radeon/nid.h @@ -585,6 +585,7 @@ #define PACKET3_SET_CONTEXT_REG_INDIRECT 0x73 #define PACKET3_SET_RESOURCE_INDIRECT 0x74 #define PACKET3_SET_APPEND_CNT 0x75 +#define PACKET3_ME_WRITE 0x7A #endif diff --git a/drivers/gpu/drm/radeon/r100.c b/drivers/gpu/drm/radeon/r100.c index 8d7e33a0b243..b41237bf884b 100644 --- a/drivers/gpu/drm/radeon/r100.c +++ b/drivers/gpu/drm/radeon/r100.c @@ -80,10 +80,12 @@ MODULE_FIRMWARE(FIRMWARE_R520); */ void r100_wait_for_vblank(struct radeon_device *rdev, int crtc) { - struct radeon_crtc *radeon_crtc = rdev->mode_info.crtcs[crtc]; int i; - if (radeon_crtc->crtc_id == 0) { + if (crtc >= rdev->num_crtc) + return; + + if (crtc == 0) { if (RREG32(RADEON_CRTC_GEN_CNTL) & RADEON_CRTC_EN) { for (i = 0; i < rdev->usec_timeout; i++) { if (!(RREG32(RADEON_CRTC_STATUS) & RADEON_CRTC_VBLANK_CUR)) @@ -698,9 +700,6 @@ int r100_irq_set(struct radeon_device *rdev) if (atomic_read(&rdev->irq.ring_int[RADEON_RING_TYPE_GFX_INDEX])) { tmp |= RADEON_SW_INT_ENABLE; } - if (rdev->irq.gui_idle) { - tmp |= RADEON_GUI_IDLE_MASK; - } if (rdev->irq.crtc_vblank_int[0] || atomic_read(&rdev->irq.pflip[0])) { tmp |= RADEON_CRTC_VBLANK_MASK; @@ -737,12 +736,6 @@ static uint32_t r100_irq_ack(struct radeon_device *rdev) RADEON_CRTC_VBLANK_STAT | RADEON_CRTC2_VBLANK_STAT | RADEON_FP_DETECT_STAT | RADEON_FP2_DETECT_STAT; - /* the interrupt works, but the status bit is permanently asserted */ - if (rdev->irq.gui_idle && radeon_gui_idle(rdev)) { - if (!rdev->irq.gui_idle_acked) - irq_mask |= RADEON_GUI_IDLE_STAT; - } - if (irqs) { WREG32(RADEON_GEN_INT_STATUS, irqs); } @@ -754,9 +747,6 @@ int r100_irq_process(struct radeon_device *rdev) uint32_t status, msi_rearm; bool queue_hotplug = false; - /* reset gui idle ack. the status bit is broken */ - rdev->irq.gui_idle_acked = false; - status = r100_irq_ack(rdev); if (!status) { return IRQ_NONE; @@ -769,11 +759,6 @@ int r100_irq_process(struct radeon_device *rdev) if (status & RADEON_SW_INT_TEST) { radeon_fence_process(rdev, RADEON_RING_TYPE_GFX_INDEX); } - /* gui idle interrupt */ - if (status & RADEON_GUI_IDLE_STAT) { - rdev->irq.gui_idle_acked = true; - wake_up(&rdev->irq.idle_queue); - } /* Vertical blank interrupts */ if (status & RADEON_CRTC_VBLANK_STAT) { if (rdev->irq.crtc_vblank_int[0]) { @@ -803,8 +788,6 @@ int r100_irq_process(struct radeon_device *rdev) } status = r100_irq_ack(rdev); } - /* reset gui idle ack. the status bit is broken */ - rdev->irq.gui_idle_acked = false; if (queue_hotplug) schedule_work(&rdev->hotplug_work); if (rdev->msi_enabled) { @@ -2530,7 +2513,7 @@ void r100_cs_track_clear(struct radeon_device *rdev, struct r100_cs_track *track /* * Global GPU functions */ -void r100_errata(struct radeon_device *rdev) +static void r100_errata(struct radeon_device *rdev) { rdev->pll_errata = 0; @@ -2545,51 +2528,7 @@ void r100_errata(struct radeon_device *rdev) } } -/* Wait for vertical sync on primary CRTC */ -void r100_gpu_wait_for_vsync(struct radeon_device *rdev) -{ - uint32_t crtc_gen_cntl, tmp; - int i; - - crtc_gen_cntl = RREG32(RADEON_CRTC_GEN_CNTL); - if ((crtc_gen_cntl & RADEON_CRTC_DISP_REQ_EN_B) || - !(crtc_gen_cntl & RADEON_CRTC_EN)) { - return; - } - /* Clear the CRTC_VBLANK_SAVE bit */ - WREG32(RADEON_CRTC_STATUS, RADEON_CRTC_VBLANK_SAVE_CLEAR); - for (i = 0; i < rdev->usec_timeout; i++) { - tmp = RREG32(RADEON_CRTC_STATUS); - if (tmp & RADEON_CRTC_VBLANK_SAVE) { - return; - } - DRM_UDELAY(1); - } -} - -/* Wait for vertical sync on secondary CRTC */ -void r100_gpu_wait_for_vsync2(struct radeon_device *rdev) -{ - uint32_t crtc2_gen_cntl, tmp; - int i; - - crtc2_gen_cntl = RREG32(RADEON_CRTC2_GEN_CNTL); - if ((crtc2_gen_cntl & RADEON_CRTC2_DISP_REQ_EN_B) || - !(crtc2_gen_cntl & RADEON_CRTC2_EN)) - return; - - /* Clear the CRTC_VBLANK_SAVE bit */ - WREG32(RADEON_CRTC2_STATUS, RADEON_CRTC2_VBLANK_SAVE_CLEAR); - for (i = 0; i < rdev->usec_timeout; i++) { - tmp = RREG32(RADEON_CRTC2_STATUS); - if (tmp & RADEON_CRTC2_VBLANK_SAVE) { - return; - } - DRM_UDELAY(1); - } -} - -int r100_rbbm_fifo_wait_for_entry(struct radeon_device *rdev, unsigned n) +static int r100_rbbm_fifo_wait_for_entry(struct radeon_device *rdev, unsigned n) { unsigned i; uint32_t tmp; @@ -2950,7 +2889,7 @@ void r100_vga_set_state(struct radeon_device *rdev, bool state) WREG32(RADEON_CONFIG_CNTL, temp); } -void r100_mc_init(struct radeon_device *rdev) +static void r100_mc_init(struct radeon_device *rdev) { u64 base; @@ -3022,7 +2961,7 @@ void r100_pll_wreg(struct radeon_device *rdev, uint32_t reg, uint32_t v) r100_pll_errata_after_data(rdev); } -void r100_set_safe_registers(struct radeon_device *rdev) +static void r100_set_safe_registers(struct radeon_device *rdev) { if (ASIC_IS_RN50(rdev)) { rdev->config.r100.reg_safe_bm = rn50_reg_safe_bm; @@ -3817,9 +3756,10 @@ int r100_ib_test(struct radeon_device *rdev, struct radeon_ring *ring) return r; } WREG32(scratch, 0xCAFEDEAD); - r = radeon_ib_get(rdev, RADEON_RING_TYPE_GFX_INDEX, &ib, 256); + r = radeon_ib_get(rdev, RADEON_RING_TYPE_GFX_INDEX, &ib, NULL, 256); if (r) { - return r; + DRM_ERROR("radeon: failed to get ib (%d).\n", r); + goto free_scratch; } ib.ptr[0] = PACKET0(scratch, 0); ib.ptr[1] = 0xDEADBEEF; @@ -3832,13 +3772,13 @@ int r100_ib_test(struct radeon_device *rdev, struct radeon_ring *ring) ib.length_dw = 8; r = radeon_ib_schedule(rdev, &ib, NULL); if (r) { - radeon_scratch_free(rdev, scratch); - radeon_ib_free(rdev, &ib); - return r; + DRM_ERROR("radeon: failed to schedule ib (%d).\n", r); + goto free_ib; } r = radeon_fence_wait(ib.fence, false); if (r) { - return r; + DRM_ERROR("radeon: fence wait failed (%d).\n", r); + goto free_ib; } for (i = 0; i < rdev->usec_timeout; i++) { tmp = RREG32(scratch); @@ -3854,8 +3794,10 @@ int r100_ib_test(struct radeon_device *rdev, struct radeon_ring *ring) scratch, tmp); r = -EINVAL; } - radeon_scratch_free(rdev, scratch); +free_ib: radeon_ib_free(rdev, &ib); +free_scratch: + radeon_scratch_free(rdev, scratch); return r; } @@ -3964,7 +3906,7 @@ static void r100_mc_program(struct radeon_device *rdev) r100_mc_resume(rdev, &save); } -void r100_clock_startup(struct radeon_device *rdev) +static void r100_clock_startup(struct radeon_device *rdev) { u32 tmp; diff --git a/drivers/gpu/drm/radeon/r300.c b/drivers/gpu/drm/radeon/r300.c index 646a1927dda7..4949bfc14b58 100644 --- a/drivers/gpu/drm/radeon/r300.c +++ b/drivers/gpu/drm/radeon/r300.c @@ -296,7 +296,7 @@ void r300_ring_start(struct radeon_device *rdev, struct radeon_ring *ring) radeon_ring_unlock_commit(rdev, ring); } -void r300_errata(struct radeon_device *rdev) +static void r300_errata(struct radeon_device *rdev) { rdev->pll_errata = 0; @@ -322,7 +322,7 @@ int r300_mc_wait_for_idle(struct radeon_device *rdev) return -1; } -void r300_gpu_init(struct radeon_device *rdev) +static void r300_gpu_init(struct radeon_device *rdev) { uint32_t gb_tile_config, tmp; diff --git a/drivers/gpu/drm/radeon/r520.c b/drivers/gpu/drm/radeon/r520.c index 079d3c52c08a..28b4f871aaf4 100644 --- a/drivers/gpu/drm/radeon/r520.c +++ b/drivers/gpu/drm/radeon/r520.c @@ -119,7 +119,7 @@ static void r520_vram_get_type(struct radeon_device *rdev) rdev->mc.vram_width *= 2; } -void r520_mc_init(struct radeon_device *rdev) +static void r520_mc_init(struct radeon_device *rdev) { r520_vram_get_type(rdev); @@ -131,7 +131,7 @@ void r520_mc_init(struct radeon_device *rdev) radeon_update_bandwidth_info(rdev); } -void r520_mc_program(struct radeon_device *rdev) +static void r520_mc_program(struct radeon_device *rdev) { struct rv515_mc_save save; diff --git a/drivers/gpu/drm/radeon/r600.c b/drivers/gpu/drm/radeon/r600.c index d79c639ae739..39b743fff791 100644 --- a/drivers/gpu/drm/radeon/r600.c +++ b/drivers/gpu/drm/radeon/r600.c @@ -98,7 +98,7 @@ int r600_debugfs_mc_info_init(struct radeon_device *rdev); /* r600,rv610,rv630,rv620,rv635,rv670 */ int r600_mc_wait_for_idle(struct radeon_device *rdev); -void r600_gpu_init(struct radeon_device *rdev); +static void r600_gpu_init(struct radeon_device *rdev); void r600_fini(struct radeon_device *rdev); void r600_irq_disable(struct radeon_device *rdev); static void r600_pcie_gen2_enable(struct radeon_device *rdev); @@ -881,7 +881,7 @@ int r600_pcie_gart_init(struct radeon_device *rdev) return radeon_gart_table_vram_alloc(rdev); } -int r600_pcie_gart_enable(struct radeon_device *rdev) +static int r600_pcie_gart_enable(struct radeon_device *rdev) { u32 tmp; int r, i; @@ -938,7 +938,7 @@ int r600_pcie_gart_enable(struct radeon_device *rdev) return 0; } -void r600_pcie_gart_disable(struct radeon_device *rdev) +static void r600_pcie_gart_disable(struct radeon_device *rdev) { u32 tmp; int i; @@ -971,14 +971,14 @@ void r600_pcie_gart_disable(struct radeon_device *rdev) radeon_gart_table_vram_unpin(rdev); } -void r600_pcie_gart_fini(struct radeon_device *rdev) +static void r600_pcie_gart_fini(struct radeon_device *rdev) { radeon_gart_fini(rdev); r600_pcie_gart_disable(rdev); radeon_gart_table_vram_free(rdev); } -void r600_agp_enable(struct radeon_device *rdev) +static void r600_agp_enable(struct radeon_device *rdev) { u32 tmp; int i; @@ -1158,7 +1158,7 @@ static void r600_vram_gtt_location(struct radeon_device *rdev, struct radeon_mc } } -int r600_mc_init(struct radeon_device *rdev) +static int r600_mc_init(struct radeon_device *rdev) { u32 tmp; int chansize, numchan; @@ -1258,7 +1258,7 @@ void r600_vram_scratch_fini(struct radeon_device *rdev) * reset, it's up to the caller to determine if the GPU needs one. We * might add an helper function to check that. */ -int r600_gpu_soft_reset(struct radeon_device *rdev) +static int r600_gpu_soft_reset(struct radeon_device *rdev) { struct rv515_mc_save save; u32 grbm_busy_mask = S_008010_VC_BUSY(1) | S_008010_VGT_BUSY_NO_DMA(1) | @@ -1433,7 +1433,7 @@ int r600_count_pipe_bits(uint32_t val) return ret; } -void r600_gpu_init(struct radeon_device *rdev) +static void r600_gpu_init(struct radeon_device *rdev) { u32 tiling_config; u32 ramcfg; @@ -2347,7 +2347,7 @@ void r600_clear_surface_reg(struct radeon_device *rdev, int reg) /* FIXME: implement */ } -int r600_startup(struct radeon_device *rdev) +static int r600_startup(struct radeon_device *rdev) { struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX]; int r; @@ -2635,10 +2635,10 @@ int r600_ib_test(struct radeon_device *rdev, struct radeon_ring *ring) return r; } WREG32(scratch, 0xCAFEDEAD); - r = radeon_ib_get(rdev, ring->idx, &ib, 256); + r = radeon_ib_get(rdev, ring->idx, &ib, NULL, 256); if (r) { DRM_ERROR("radeon: failed to get ib (%d).\n", r); - return r; + goto free_scratch; } ib.ptr[0] = PACKET3(PACKET3_SET_CONFIG_REG, 1); ib.ptr[1] = ((scratch - PACKET3_SET_CONFIG_REG_OFFSET) >> 2); @@ -2646,15 +2646,13 @@ int r600_ib_test(struct radeon_device *rdev, struct radeon_ring *ring) ib.length_dw = 3; r = radeon_ib_schedule(rdev, &ib, NULL); if (r) { - radeon_scratch_free(rdev, scratch); - radeon_ib_free(rdev, &ib); DRM_ERROR("radeon: failed to schedule ib (%d).\n", r); - return r; + goto free_ib; } r = radeon_fence_wait(ib.fence, false); if (r) { DRM_ERROR("radeon: fence wait failed (%d).\n", r); - return r; + goto free_ib; } for (i = 0; i < rdev->usec_timeout; i++) { tmp = RREG32(scratch); @@ -2669,8 +2667,10 @@ int r600_ib_test(struct radeon_device *rdev, struct radeon_ring *ring) scratch, tmp); r = -EINVAL; } - radeon_scratch_free(rdev, scratch); +free_ib: radeon_ib_free(rdev, &ib); +free_scratch: + radeon_scratch_free(rdev, scratch); return r; } @@ -3088,10 +3088,6 @@ int r600_irq_set(struct radeon_device *rdev) DRM_DEBUG("r600_irq_set: hdmi 0\n"); hdmi1 |= HDMI0_AZ_FORMAT_WTRIG_MASK; } - if (rdev->irq.gui_idle) { - DRM_DEBUG("gui idle\n"); - grbm_int_cntl |= GUI_IDLE_INT_ENABLE; - } WREG32(CP_INT_CNTL, cp_int_cntl); WREG32(DxMODE_INT_MASK, mode_int); @@ -3475,7 +3471,6 @@ restart_ih: break; case 233: /* GUI IDLE */ DRM_DEBUG("IH: GUI idle\n"); - wake_up(&rdev->irq.idle_queue); break; default: DRM_DEBUG("Unhandled interrupt: %d %d\n", src_id, src_data); diff --git a/drivers/gpu/drm/radeon/r600_blit.c b/drivers/gpu/drm/radeon/r600_blit.c index 3c031a48205d..661fec2a2cc1 100644 --- a/drivers/gpu/drm/radeon/r600_blit.c +++ b/drivers/gpu/drm/radeon/r600_blit.c @@ -489,31 +489,36 @@ set_default_state(drm_radeon_private_t *dev_priv) ADVANCE_RING(); } -static uint32_t i2f(uint32_t input) +/* 23 bits of float fractional data */ +#define I2F_FRAC_BITS 23 +#define I2F_MASK ((1 << I2F_FRAC_BITS) - 1) + +/* + * Converts unsigned integer into 32-bit IEEE floating point representation. + * Will be exact from 0 to 2^24. Above that, we round towards zero + * as the fractional bits will not fit in a float. (It would be better to + * round towards even as the fpu does, but that is slower.) + */ +__pure uint32_t int2float(uint32_t x) { - u32 result, i, exponent, fraction; - - if ((input & 0x3fff) == 0) - result = 0; /* 0 is a special case */ - else { - exponent = 140; /* exponent biased by 127; */ - fraction = (input & 0x3fff) << 10; /* cheat and only - handle numbers below 2^^15 */ - for (i = 0; i < 14; i++) { - if (fraction & 0x800000) - break; - else { - fraction = fraction << 1; /* keep - shifting left until top bit = 1 */ - exponent = exponent - 1; - } - } - result = exponent << 23 | (fraction & 0x7fffff); /* mask - off top bit; assumed 1 */ - } - return result; -} + uint32_t msb, exponent, fraction; + + /* Zero is special */ + if (!x) return 0; + + /* Get location of the most significant bit */ + msb = __fls(x); + /* + * Use a rotate instead of a shift because that works both leftwards + * and rightwards due to the mod(32) behaviour. This means we don't + * need to check to see if we are above 2^24 or not. + */ + fraction = ror32(x, (msb - I2F_FRAC_BITS) & 0x1f) & I2F_MASK; + exponent = (127 + msb) << I2F_FRAC_BITS; + + return fraction + exponent; +} static int r600_nomm_get_vb(struct drm_device *dev) { @@ -632,20 +637,20 @@ r600_blit_copy(struct drm_device *dev, vb = r600_nomm_get_vb_ptr(dev); } - vb[0] = i2f(dst_x); + vb[0] = int2float(dst_x); vb[1] = 0; - vb[2] = i2f(src_x); + vb[2] = int2float(src_x); vb[3] = 0; - vb[4] = i2f(dst_x); - vb[5] = i2f(h); - vb[6] = i2f(src_x); - vb[7] = i2f(h); + vb[4] = int2float(dst_x); + vb[5] = int2float(h); + vb[6] = int2float(src_x); + vb[7] = int2float(h); - vb[8] = i2f(dst_x + cur_size); - vb[9] = i2f(h); - vb[10] = i2f(src_x + cur_size); - vb[11] = i2f(h); + vb[8] = int2float(dst_x + cur_size); + vb[9] = int2float(h); + vb[10] = int2float(src_x + cur_size); + vb[11] = int2float(h); /* src */ set_tex_resource(dev_priv, FMT_8, @@ -721,20 +726,20 @@ r600_blit_copy(struct drm_device *dev, vb = r600_nomm_get_vb_ptr(dev); } - vb[0] = i2f(dst_x / 4); + vb[0] = int2float(dst_x / 4); vb[1] = 0; - vb[2] = i2f(src_x / 4); + vb[2] = int2float(src_x / 4); vb[3] = 0; - vb[4] = i2f(dst_x / 4); - vb[5] = i2f(h); - vb[6] = i2f(src_x / 4); - vb[7] = i2f(h); + vb[4] = int2float(dst_x / 4); + vb[5] = int2float(h); + vb[6] = int2float(src_x / 4); + vb[7] = int2float(h); - vb[8] = i2f((dst_x + cur_size) / 4); - vb[9] = i2f(h); - vb[10] = i2f((src_x + cur_size) / 4); - vb[11] = i2f(h); + vb[8] = int2float((dst_x + cur_size) / 4); + vb[9] = int2float(h); + vb[10] = int2float((src_x + cur_size) / 4); + vb[11] = int2float(h); /* src */ set_tex_resource(dev_priv, FMT_8_8_8_8, @@ -804,20 +809,20 @@ r600_blit_swap(struct drm_device *dev, dx2 = dx + w; dy2 = dy + h; - vb[0] = i2f(dx); - vb[1] = i2f(dy); - vb[2] = i2f(sx); - vb[3] = i2f(sy); + vb[0] = int2float(dx); + vb[1] = int2float(dy); + vb[2] = int2float(sx); + vb[3] = int2float(sy); - vb[4] = i2f(dx); - vb[5] = i2f(dy2); - vb[6] = i2f(sx); - vb[7] = i2f(sy2); + vb[4] = int2float(dx); + vb[5] = int2float(dy2); + vb[6] = int2float(sx); + vb[7] = int2float(sy2); - vb[8] = i2f(dx2); - vb[9] = i2f(dy2); - vb[10] = i2f(sx2); - vb[11] = i2f(sy2); + vb[8] = int2float(dx2); + vb[9] = int2float(dy2); + vb[10] = int2float(sx2); + vb[11] = int2float(sy2); switch(cpp) { case 4: diff --git a/drivers/gpu/drm/radeon/r600_blit_kms.c b/drivers/gpu/drm/radeon/r600_blit_kms.c index 2bef8549ddfe..1c7ed3a5d045 100644 --- a/drivers/gpu/drm/radeon/r600_blit_kms.c +++ b/drivers/gpu/drm/radeon/r600_blit_kms.c @@ -455,46 +455,6 @@ set_default_state(struct radeon_device *rdev) radeon_ring_write(ring, sq_stack_resource_mgmt_2); } -#define I2F_MAX_BITS 15 -#define I2F_MAX_INPUT ((1 << I2F_MAX_BITS) - 1) -#define I2F_SHIFT (24 - I2F_MAX_BITS) - -/* - * Converts unsigned integer into 32-bit IEEE floating point representation. - * Conversion is not universal and only works for the range from 0 - * to 2^I2F_MAX_BITS-1. Currently we only use it with inputs between - * 0 and 16384 (inclusive), so I2F_MAX_BITS=15 is enough. If necessary, - * I2F_MAX_BITS can be increased, but that will add to the loop iterations - * and slow us down. Conversion is done by shifting the input and counting - * down until the first 1 reaches bit position 23. The resulting counter - * and the shifted input are, respectively, the exponent and the fraction. - * The sign is always zero. - */ -static uint32_t i2f(uint32_t input) -{ - u32 result, i, exponent, fraction; - - WARN_ON_ONCE(input > I2F_MAX_INPUT); - - if ((input & I2F_MAX_INPUT) == 0) - result = 0; - else { - exponent = 126 + I2F_MAX_BITS; - fraction = (input & I2F_MAX_INPUT) << I2F_SHIFT; - - for (i = 0; i < I2F_MAX_BITS; i++) { - if (fraction & 0x800000) - break; - else { - fraction = fraction << 1; - exponent = exponent - 1; - } - } - result = exponent << 23 | (fraction & 0x7fffff); - } - return result; -} - int r600_blit_init(struct radeon_device *rdev) { u32 obj_size; @@ -766,14 +726,14 @@ void r600_kms_blit_copy(struct radeon_device *rdev, vb_cpu_addr[3] = 0; vb_cpu_addr[4] = 0; - vb_cpu_addr[5] = i2f(h); + vb_cpu_addr[5] = int2float(h); vb_cpu_addr[6] = 0; - vb_cpu_addr[7] = i2f(h); + vb_cpu_addr[7] = int2float(h); - vb_cpu_addr[8] = i2f(w); - vb_cpu_addr[9] = i2f(h); - vb_cpu_addr[10] = i2f(w); - vb_cpu_addr[11] = i2f(h); + vb_cpu_addr[8] = int2float(w); + vb_cpu_addr[9] = int2float(h); + vb_cpu_addr[10] = int2float(w); + vb_cpu_addr[11] = int2float(h); rdev->r600_blit.primitives.set_tex_resource(rdev, FMT_8_8_8_8, w, h, w, src_gpu_addr, size_in_bytes); diff --git a/drivers/gpu/drm/radeon/r600_blit_shaders.h b/drivers/gpu/drm/radeon/r600_blit_shaders.h index f437d36dd98c..2f3ce7a75976 100644 --- a/drivers/gpu/drm/radeon/r600_blit_shaders.h +++ b/drivers/gpu/drm/radeon/r600_blit_shaders.h @@ -35,4 +35,5 @@ extern const u32 r6xx_default_state[]; extern const u32 r6xx_ps_size, r6xx_vs_size; extern const u32 r6xx_default_size, r7xx_default_size; +__pure uint32_t int2float(uint32_t x); #endif diff --git a/drivers/gpu/drm/radeon/r600_cs.c b/drivers/gpu/drm/radeon/r600_cs.c index f37676d7f217..853f05ced1b1 100644 --- a/drivers/gpu/drm/radeon/r600_cs.c +++ b/drivers/gpu/drm/radeon/r600_cs.c @@ -847,7 +847,7 @@ static int r600_cs_track_check(struct radeon_cs_parser *p) * Assume that chunk_ib_index is properly set. Will return -EINVAL * if packet is bigger than remaining ib size. or if packets is unknown. **/ -int r600_cs_packet_parse(struct radeon_cs_parser *p, +static int r600_cs_packet_parse(struct radeon_cs_parser *p, struct radeon_cs_packet *pkt, unsigned idx) { @@ -2180,7 +2180,8 @@ static int r600_packet3_check(struct radeon_cs_parser *p, } break; case PACKET3_STRMOUT_BASE_UPDATE: - if (p->family < CHIP_RV770) { + /* RS780 and RS880 also need this */ + if (p->family < CHIP_RS780) { DRM_ERROR("STRMOUT_BASE_UPDATE only supported on 7xx\n"); return -EINVAL; } diff --git a/drivers/gpu/drm/radeon/r600_hdmi.c b/drivers/gpu/drm/radeon/r600_hdmi.c index e3558c3ef24a..857a7d7862fb 100644 --- a/drivers/gpu/drm/radeon/r600_hdmi.c +++ b/drivers/gpu/drm/radeon/r600_hdmi.c @@ -53,7 +53,7 @@ enum r600_hdmi_iec_status_bits { AUDIO_STATUS_LEVEL = 0x80 }; -struct radeon_hdmi_acr r600_hdmi_predefined_acr[] = { +static const struct radeon_hdmi_acr r600_hdmi_predefined_acr[] = { /* 32kHz 44.1kHz 48kHz */ /* Clock N CTS N CTS N CTS */ { 25174, 4576, 28125, 7007, 31250, 6864, 28125 }, /* 25,20/1.001 MHz */ diff --git a/drivers/gpu/drm/radeon/radeon.h b/drivers/gpu/drm/radeon/radeon.h index 59a15315ae9f..b04c06444d8b 100644 --- a/drivers/gpu/drm/radeon/radeon.h +++ b/drivers/gpu/drm/radeon/radeon.h @@ -123,6 +123,7 @@ extern int radeon_lockup_timeout; #define CAYMAN_RING_TYPE_CP2_INDEX 2 /* hardcode those limit for now */ +#define RADEON_VA_IB_OFFSET (1 << 20) #define RADEON_VA_RESERVED_SIZE (8 << 20) #define RADEON_IB_VM_MAX_SIZE (64 << 10) @@ -253,6 +254,22 @@ static inline struct radeon_fence *radeon_fence_later(struct radeon_fence *a, } } +static inline bool radeon_fence_is_earlier(struct radeon_fence *a, + struct radeon_fence *b) +{ + if (!a) { + return false; + } + + if (!b) { + return true; + } + + BUG_ON(a->ring != b->ring); + + return a->seq < b->seq; +} + /* * Tiling registers */ @@ -275,18 +292,20 @@ struct radeon_mman { /* bo virtual address in a specific vm */ struct radeon_bo_va { - /* bo list is protected by bo being reserved */ + /* protected by bo being reserved */ struct list_head bo_list; - /* vm list is protected by vm mutex */ - struct list_head vm_list; - /* constant after initialization */ - struct radeon_vm *vm; - struct radeon_bo *bo; uint64_t soffset; uint64_t eoffset; uint32_t flags; - struct radeon_fence *fence; bool valid; + unsigned ref_count; + + /* protected by vm mutex */ + struct list_head vm_list; + + /* constant after initialization */ + struct radeon_vm *vm; + struct radeon_bo *bo; }; struct radeon_bo { @@ -566,9 +585,6 @@ struct radeon_irq { atomic_t pflip[RADEON_MAX_CRTCS]; wait_queue_head_t vblank_queue; bool hpd[RADEON_MAX_HPD_PINS]; - bool gui_idle; - bool gui_idle_acked; - wait_queue_head_t idle_queue; bool afmt[RADEON_MAX_AFMT_BLOCKS]; union radeon_irq_stat_regs stat_regs; }; @@ -583,7 +599,6 @@ void radeon_irq_kms_enable_afmt(struct radeon_device *rdev, int block); void radeon_irq_kms_disable_afmt(struct radeon_device *rdev, int block); void radeon_irq_kms_enable_hpd(struct radeon_device *rdev, unsigned hpd_mask); void radeon_irq_kms_disable_hpd(struct radeon_device *rdev, unsigned hpd_mask); -int radeon_irq_kms_wait_gui_idle(struct radeon_device *rdev); /* * CP & rings. @@ -596,7 +611,7 @@ struct radeon_ib { uint32_t *ptr; int ring; struct radeon_fence *fence; - unsigned vm_id; + struct radeon_vm *vm; bool is_const_ib; struct radeon_fence *sync_to[RADEON_NUM_RINGS]; struct radeon_semaphore *semaphore; @@ -632,41 +647,38 @@ struct radeon_ring { /* * VM */ + +/* maximum number of VMIDs */ +#define RADEON_NUM_VM 16 + +/* defines number of bits in page table versus page directory, + * a page is 4KB so we have 12 bits offset, 9 bits in the page + * table and the remaining 19 bits are in the page directory */ +#define RADEON_VM_BLOCK_SIZE 9 + +/* number of entries in page table */ +#define RADEON_VM_PTE_COUNT (1 << RADEON_VM_BLOCK_SIZE) + struct radeon_vm { struct list_head list; struct list_head va; - int id; + unsigned id; unsigned last_pfn; - u64 pt_gpu_addr; - u64 *pt; + u64 pd_gpu_addr; struct radeon_sa_bo *sa_bo; struct mutex mutex; /* last fence for cs using this vm */ struct radeon_fence *fence; -}; - -struct radeon_vm_funcs { - int (*init)(struct radeon_device *rdev); - void (*fini)(struct radeon_device *rdev); - /* cs mutex must be lock for schedule_ib */ - int (*bind)(struct radeon_device *rdev, struct radeon_vm *vm, int id); - void (*unbind)(struct radeon_device *rdev, struct radeon_vm *vm); - void (*tlb_flush)(struct radeon_device *rdev, struct radeon_vm *vm); - uint32_t (*page_flags)(struct radeon_device *rdev, - struct radeon_vm *vm, - uint32_t flags); - void (*set_page)(struct radeon_device *rdev, struct radeon_vm *vm, - unsigned pfn, uint64_t addr, uint32_t flags); + /* last flush or NULL if we still need to flush */ + struct radeon_fence *last_flush; }; struct radeon_vm_manager { struct mutex lock; struct list_head lru_vm; - uint32_t use_bitmap; + struct radeon_fence *active[RADEON_NUM_VM]; struct radeon_sa_manager sa_manager; uint32_t max_pfn; - /* fields constant after init */ - const struct radeon_vm_funcs *funcs; /* number of VMIDs */ unsigned nvm; /* vram base address for page table entry */ @@ -738,7 +750,8 @@ struct si_rlc { }; int radeon_ib_get(struct radeon_device *rdev, int ring, - struct radeon_ib *ib, unsigned size); + struct radeon_ib *ib, struct radeon_vm *vm, + unsigned size); void radeon_ib_free(struct radeon_device *rdev, struct radeon_ib *ib); int radeon_ib_schedule(struct radeon_device *rdev, struct radeon_ib *ib, struct radeon_ib *const_ib); @@ -1131,6 +1144,15 @@ struct radeon_asic { void (*tlb_flush)(struct radeon_device *rdev); int (*set_page)(struct radeon_device *rdev, int i, uint64_t addr); } gart; + struct { + int (*init)(struct radeon_device *rdev); + void (*fini)(struct radeon_device *rdev); + + u32 pt_ring_index; + void (*set_page)(struct radeon_device *rdev, uint64_t pe, + uint64_t addr, unsigned count, + uint32_t incr, uint32_t flags); + } vm; /* ring specific callbacks */ struct { void (*ib_execute)(struct radeon_device *rdev, struct radeon_ib *ib); @@ -1143,6 +1165,7 @@ struct radeon_asic { int (*ring_test)(struct radeon_device *rdev, struct radeon_ring *cp); int (*ib_test)(struct radeon_device *rdev, struct radeon_ring *cp); bool (*is_lockup)(struct radeon_device *rdev, struct radeon_ring *cp); + void (*vm_flush)(struct radeon_device *rdev, int ridx, struct radeon_vm *vm); } ring[RADEON_NUM_RINGS]; /* irqs */ struct { @@ -1157,6 +1180,10 @@ struct radeon_asic { u32 (*get_vblank_counter)(struct radeon_device *rdev, int crtc); /* wait for vblank */ void (*wait_for_vblank)(struct radeon_device *rdev, int crtc); + /* set backlight level */ + void (*set_backlight_level)(struct radeon_encoder *radeon_encoder, u8 level); + /* get backlight level */ + u8 (*get_backlight_level)(struct radeon_encoder *radeon_encoder); } display; /* copy functions for bo handling */ struct { @@ -1428,6 +1455,56 @@ struct r600_vram_scratch { u64 gpu_addr; }; +/* + * ACPI + */ +struct radeon_atif_notification_cfg { + bool enabled; + int command_code; +}; + +struct radeon_atif_notifications { + bool display_switch; + bool expansion_mode_change; + bool thermal_state; + bool forced_power_state; + bool system_power_state; + bool display_conf_change; + bool px_gfx_switch; + bool brightness_change; + bool dgpu_display_event; +}; + +struct radeon_atif_functions { + bool system_params; + bool sbios_requests; + bool select_active_disp; + bool lid_state; + bool get_tv_standard; + bool set_tv_standard; + bool get_panel_expansion_mode; + bool set_panel_expansion_mode; + bool temperature_change; + bool graphics_device_types; +}; + +struct radeon_atif { + struct radeon_atif_notifications notifications; + struct radeon_atif_functions functions; + struct radeon_atif_notification_cfg notification_cfg; + struct radeon_encoder *encoder_for_bl; +}; + +struct radeon_atcs_functions { + bool get_ext_state; + bool pcie_perf_req; + bool pcie_dev_rdy; + bool pcie_bus_width; +}; + +struct radeon_atcs { + struct radeon_atcs_functions functions; +}; /* * Core structure, functions and helpers. @@ -1520,6 +1597,9 @@ struct radeon_device { /* virtual memory */ struct radeon_vm_manager vm_manager; struct mutex gpu_clock_mutex; + /* ACPI interface */ + struct radeon_atif atif; + struct radeon_atcs atcs; }; int radeon_device_init(struct radeon_device *rdev, @@ -1683,15 +1763,21 @@ void radeon_ring_write(struct radeon_ring *ring, uint32_t v); #define radeon_asic_reset(rdev) (rdev)->asic->asic_reset((rdev)) #define radeon_gart_tlb_flush(rdev) (rdev)->asic->gart.tlb_flush((rdev)) #define radeon_gart_set_page(rdev, i, p) (rdev)->asic->gart.set_page((rdev), (i), (p)) +#define radeon_asic_vm_init(rdev) (rdev)->asic->vm.init((rdev)) +#define radeon_asic_vm_fini(rdev) (rdev)->asic->vm.fini((rdev)) +#define radeon_asic_vm_set_page(rdev, pe, addr, count, incr, flags) ((rdev)->asic->vm.set_page((rdev), (pe), (addr), (count), (incr), (flags))) #define radeon_ring_start(rdev, r, cp) (rdev)->asic->ring[(r)].ring_start((rdev), (cp)) #define radeon_ring_test(rdev, r, cp) (rdev)->asic->ring[(r)].ring_test((rdev), (cp)) #define radeon_ib_test(rdev, r, cp) (rdev)->asic->ring[(r)].ib_test((rdev), (cp)) #define radeon_ring_ib_execute(rdev, r, ib) (rdev)->asic->ring[(r)].ib_execute((rdev), (ib)) #define radeon_ring_ib_parse(rdev, r, ib) (rdev)->asic->ring[(r)].ib_parse((rdev), (ib)) #define radeon_ring_is_lockup(rdev, r, cp) (rdev)->asic->ring[(r)].is_lockup((rdev), (cp)) +#define radeon_ring_vm_flush(rdev, r, vm) (rdev)->asic->ring[(r)].vm_flush((rdev), (r), (vm)) #define radeon_irq_set(rdev) (rdev)->asic->irq.set((rdev)) #define radeon_irq_process(rdev) (rdev)->asic->irq.process((rdev)) #define radeon_get_vblank_counter(rdev, crtc) (rdev)->asic->display.get_vblank_counter((rdev), (crtc)) +#define radeon_set_backlight_level(rdev, e, l) (rdev)->asic->display.set_backlight_level((e), (l)) +#define radeon_get_backlight_level(rdev, e) (rdev)->asic->display.get_backlight_level((e)) #define radeon_fence_ring_emit(rdev, r, fence) (rdev)->asic->ring[(r)].emit_fence((rdev), (fence)) #define radeon_semaphore_ring_emit(rdev, r, cp, semaphore, emit_wait) (rdev)->asic->ring[(r)].emit_semaphore((rdev), (cp), (semaphore), (emit_wait)) #define radeon_copy_blit(rdev, s, d, np, f) (rdev)->asic->copy.blit((rdev), (s), (d), (np), (f)) @@ -1759,22 +1845,30 @@ int radeon_vm_manager_init(struct radeon_device *rdev); void radeon_vm_manager_fini(struct radeon_device *rdev); int radeon_vm_init(struct radeon_device *rdev, struct radeon_vm *vm); void radeon_vm_fini(struct radeon_device *rdev, struct radeon_vm *vm); -int radeon_vm_bind(struct radeon_device *rdev, struct radeon_vm *vm); -void radeon_vm_unbind(struct radeon_device *rdev, struct radeon_vm *vm); +int radeon_vm_alloc_pt(struct radeon_device *rdev, struct radeon_vm *vm); +struct radeon_fence *radeon_vm_grab_id(struct radeon_device *rdev, + struct radeon_vm *vm, int ring); +void radeon_vm_fence(struct radeon_device *rdev, + struct radeon_vm *vm, + struct radeon_fence *fence); +uint64_t radeon_vm_map_gart(struct radeon_device *rdev, uint64_t addr); int radeon_vm_bo_update_pte(struct radeon_device *rdev, struct radeon_vm *vm, struct radeon_bo *bo, struct ttm_mem_reg *mem); void radeon_vm_bo_invalidate(struct radeon_device *rdev, struct radeon_bo *bo); -int radeon_vm_bo_add(struct radeon_device *rdev, - struct radeon_vm *vm, - struct radeon_bo *bo, - uint64_t offset, - uint32_t flags); +struct radeon_bo_va *radeon_vm_bo_find(struct radeon_vm *vm, + struct radeon_bo *bo); +struct radeon_bo_va *radeon_vm_bo_add(struct radeon_device *rdev, + struct radeon_vm *vm, + struct radeon_bo *bo); +int radeon_vm_bo_set_addr(struct radeon_device *rdev, + struct radeon_bo_va *bo_va, + uint64_t offset, + uint32_t flags); int radeon_vm_bo_rmv(struct radeon_device *rdev, - struct radeon_vm *vm, - struct radeon_bo *bo); + struct radeon_bo_va *bo_va); /* audio */ void r600_audio_update_hdmi(struct work_struct *work); @@ -1832,12 +1926,14 @@ extern void evergreen_hdmi_setmode(struct drm_encoder *encoder, struct drm_displ extern int ni_init_microcode(struct radeon_device *rdev); extern int ni_mc_load_microcode(struct radeon_device *rdev); -/* radeon_acpi.c */ -#if defined(CONFIG_ACPI) -extern int radeon_acpi_init(struct radeon_device *rdev); -#else -static inline int radeon_acpi_init(struct radeon_device *rdev) { return 0; } -#endif +/* radeon_acpi.c */ +#if defined(CONFIG_ACPI) +extern int radeon_acpi_init(struct radeon_device *rdev); +extern void radeon_acpi_fini(struct radeon_device *rdev); +#else +static inline int radeon_acpi_init(struct radeon_device *rdev) { return 0; } +static inline void radeon_acpi_fini(struct radeon_device *rdev) { } +#endif #include "radeon_object.h" diff --git a/drivers/gpu/drm/radeon/radeon_acpi.c b/drivers/gpu/drm/radeon/radeon_acpi.c index 3516a6081dcf..c3976eb341bf 100644 --- a/drivers/gpu/drm/radeon/radeon_acpi.c +++ b/drivers/gpu/drm/radeon/radeon_acpi.c @@ -1,35 +1,120 @@ +/* + * Copyright 2012 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + */ + #include <linux/pci.h> #include <linux/acpi.h> #include <linux/slab.h> +#include <linux/power_supply.h> #include <acpi/acpi_drivers.h> #include <acpi/acpi_bus.h> +#include <acpi/video.h> #include "drmP.h" #include "drm.h" #include "drm_sarea.h" #include "drm_crtc_helper.h" #include "radeon.h" +#include "radeon_acpi.h" +#include "atom.h" #include <linux/vga_switcheroo.h> +#define ACPI_AC_CLASS "ac_adapter" + +extern void radeon_pm_acpi_event_handler(struct radeon_device *rdev); + +struct atif_verify_interface { + u16 size; /* structure size in bytes (includes size field) */ + u16 version; /* version */ + u32 notification_mask; /* supported notifications mask */ + u32 function_bits; /* supported functions bit vector */ +} __packed; + +struct atif_system_params { + u16 size; /* structure size in bytes (includes size field) */ + u32 valid_mask; /* valid flags mask */ + u32 flags; /* flags */ + u8 command_code; /* notify command code */ +} __packed; + +struct atif_sbios_requests { + u16 size; /* structure size in bytes (includes size field) */ + u32 pending; /* pending sbios requests */ + u8 panel_exp_mode; /* panel expansion mode */ + u8 thermal_gfx; /* thermal state: target gfx controller */ + u8 thermal_state; /* thermal state: state id (0: exit state, non-0: state) */ + u8 forced_power_gfx; /* forced power state: target gfx controller */ + u8 forced_power_state; /* forced power state: state id */ + u8 system_power_src; /* system power source */ + u8 backlight_level; /* panel backlight level (0-255) */ +} __packed; + +#define ATIF_NOTIFY_MASK 0x3 +#define ATIF_NOTIFY_NONE 0 +#define ATIF_NOTIFY_81 1 +#define ATIF_NOTIFY_N 2 + +struct atcs_verify_interface { + u16 size; /* structure size in bytes (includes size field) */ + u16 version; /* version */ + u32 function_bits; /* supported functions bit vector */ +} __packed; + /* Call the ATIF method + */ +/** + * radeon_atif_call - call an ATIF method * - * Note: currently we discard the output + * @handle: acpi handle + * @function: the ATIF function to execute + * @params: ATIF function params + * + * Executes the requested ATIF function (all asics). + * Returns a pointer to the acpi output buffer. */ -static int radeon_atif_call(acpi_handle handle) +static union acpi_object *radeon_atif_call(acpi_handle handle, int function, + struct acpi_buffer *params) { acpi_status status; union acpi_object atif_arg_elements[2]; struct acpi_object_list atif_arg; - struct acpi_buffer buffer = { ACPI_ALLOCATE_BUFFER, NULL}; + struct acpi_buffer buffer = { ACPI_ALLOCATE_BUFFER, NULL }; atif_arg.count = 2; atif_arg.pointer = &atif_arg_elements[0]; atif_arg_elements[0].type = ACPI_TYPE_INTEGER; - atif_arg_elements[0].integer.value = 0; - atif_arg_elements[1].type = ACPI_TYPE_INTEGER; - atif_arg_elements[1].integer.value = 0; + atif_arg_elements[0].integer.value = function; + + if (params) { + atif_arg_elements[1].type = ACPI_TYPE_BUFFER; + atif_arg_elements[1].buffer.length = params->length; + atif_arg_elements[1].buffer.pointer = params->pointer; + } else { + /* We need a second fake parameter */ + atif_arg_elements[1].type = ACPI_TYPE_INTEGER; + atif_arg_elements[1].integer.value = 0; + } status = acpi_evaluate_object(handle, "ATIF", &atif_arg, &buffer); @@ -38,17 +123,434 @@ static int radeon_atif_call(acpi_handle handle) DRM_DEBUG_DRIVER("failed to evaluate ATIF got %s\n", acpi_format_exception(status)); kfree(buffer.pointer); - return 1; + return NULL; } - kfree(buffer.pointer); - return 0; + return buffer.pointer; +} + +/** + * radeon_atif_parse_notification - parse supported notifications + * + * @n: supported notifications struct + * @mask: supported notifications mask from ATIF + * + * Use the supported notifications mask from ATIF function + * ATIF_FUNCTION_VERIFY_INTERFACE to determine what notifications + * are supported (all asics). + */ +static void radeon_atif_parse_notification(struct radeon_atif_notifications *n, u32 mask) +{ + n->display_switch = mask & ATIF_DISPLAY_SWITCH_REQUEST_SUPPORTED; + n->expansion_mode_change = mask & ATIF_EXPANSION_MODE_CHANGE_REQUEST_SUPPORTED; + n->thermal_state = mask & ATIF_THERMAL_STATE_CHANGE_REQUEST_SUPPORTED; + n->forced_power_state = mask & ATIF_FORCED_POWER_STATE_CHANGE_REQUEST_SUPPORTED; + n->system_power_state = mask & ATIF_SYSTEM_POWER_SOURCE_CHANGE_REQUEST_SUPPORTED; + n->display_conf_change = mask & ATIF_DISPLAY_CONF_CHANGE_REQUEST_SUPPORTED; + n->px_gfx_switch = mask & ATIF_PX_GFX_SWITCH_REQUEST_SUPPORTED; + n->brightness_change = mask & ATIF_PANEL_BRIGHTNESS_CHANGE_REQUEST_SUPPORTED; + n->dgpu_display_event = mask & ATIF_DGPU_DISPLAY_EVENT_SUPPORTED; +} + +/** + * radeon_atif_parse_functions - parse supported functions + * + * @f: supported functions struct + * @mask: supported functions mask from ATIF + * + * Use the supported functions mask from ATIF function + * ATIF_FUNCTION_VERIFY_INTERFACE to determine what functions + * are supported (all asics). + */ +static void radeon_atif_parse_functions(struct radeon_atif_functions *f, u32 mask) +{ + f->system_params = mask & ATIF_GET_SYSTEM_PARAMETERS_SUPPORTED; + f->sbios_requests = mask & ATIF_GET_SYSTEM_BIOS_REQUESTS_SUPPORTED; + f->select_active_disp = mask & ATIF_SELECT_ACTIVE_DISPLAYS_SUPPORTED; + f->lid_state = mask & ATIF_GET_LID_STATE_SUPPORTED; + f->get_tv_standard = mask & ATIF_GET_TV_STANDARD_FROM_CMOS_SUPPORTED; + f->set_tv_standard = mask & ATIF_SET_TV_STANDARD_IN_CMOS_SUPPORTED; + f->get_panel_expansion_mode = mask & ATIF_GET_PANEL_EXPANSION_MODE_FROM_CMOS_SUPPORTED; + f->set_panel_expansion_mode = mask & ATIF_SET_PANEL_EXPANSION_MODE_IN_CMOS_SUPPORTED; + f->temperature_change = mask & ATIF_TEMPERATURE_CHANGE_NOTIFICATION_SUPPORTED; + f->graphics_device_types = mask & ATIF_GET_GRAPHICS_DEVICE_TYPES_SUPPORTED; +} + +/** + * radeon_atif_verify_interface - verify ATIF + * + * @handle: acpi handle + * @atif: radeon atif struct + * + * Execute the ATIF_FUNCTION_VERIFY_INTERFACE ATIF function + * to initialize ATIF and determine what features are supported + * (all asics). + * returns 0 on success, error on failure. + */ +static int radeon_atif_verify_interface(acpi_handle handle, + struct radeon_atif *atif) +{ + union acpi_object *info; + struct atif_verify_interface output; + size_t size; + int err = 0; + + info = radeon_atif_call(handle, ATIF_FUNCTION_VERIFY_INTERFACE, NULL); + if (!info) + return -EIO; + + memset(&output, 0, sizeof(output)); + + size = *(u16 *) info->buffer.pointer; + if (size < 12) { + DRM_INFO("ATIF buffer is too small: %lu\n", size); + err = -EINVAL; + goto out; + } + size = min(sizeof(output), size); + + memcpy(&output, info->buffer.pointer, size); + + /* TODO: check version? */ + DRM_DEBUG_DRIVER("ATIF version %u\n", output.version); + + radeon_atif_parse_notification(&atif->notifications, output.notification_mask); + radeon_atif_parse_functions(&atif->functions, output.function_bits); + +out: + kfree(info); + return err; +} + +/** + * radeon_atif_get_notification_params - determine notify configuration + * + * @handle: acpi handle + * @n: atif notification configuration struct + * + * Execute the ATIF_FUNCTION_GET_SYSTEM_PARAMETERS ATIF function + * to determine if a notifier is used and if so which one + * (all asics). This is either Notify(VGA, 0x81) or Notify(VGA, n) + * where n is specified in the result if a notifier is used. + * Returns 0 on success, error on failure. + */ +static int radeon_atif_get_notification_params(acpi_handle handle, + struct radeon_atif_notification_cfg *n) +{ + union acpi_object *info; + struct atif_system_params params; + size_t size; + int err = 0; + + info = radeon_atif_call(handle, ATIF_FUNCTION_GET_SYSTEM_PARAMETERS, NULL); + if (!info) { + err = -EIO; + goto out; + } + + size = *(u16 *) info->buffer.pointer; + if (size < 10) { + err = -EINVAL; + goto out; + } + + memset(¶ms, 0, sizeof(params)); + size = min(sizeof(params), size); + memcpy(¶ms, info->buffer.pointer, size); + + DRM_DEBUG_DRIVER("SYSTEM_PARAMS: mask = %#x, flags = %#x\n", + params.flags, params.valid_mask); + params.flags = params.flags & params.valid_mask; + + if ((params.flags & ATIF_NOTIFY_MASK) == ATIF_NOTIFY_NONE) { + n->enabled = false; + n->command_code = 0; + } else if ((params.flags & ATIF_NOTIFY_MASK) == ATIF_NOTIFY_81) { + n->enabled = true; + n->command_code = 0x81; + } else { + if (size < 11) { + err = -EINVAL; + goto out; + } + n->enabled = true; + n->command_code = params.command_code; + } + +out: + DRM_DEBUG_DRIVER("Notification %s, command code = %#x\n", + (n->enabled ? "enabled" : "disabled"), + n->command_code); + kfree(info); + return err; +} + +/** + * radeon_atif_get_sbios_requests - get requested sbios event + * + * @handle: acpi handle + * @req: atif sbios request struct + * + * Execute the ATIF_FUNCTION_GET_SYSTEM_BIOS_REQUESTS ATIF function + * to determine what requests the sbios is making to the driver + * (all asics). + * Returns 0 on success, error on failure. + */ +static int radeon_atif_get_sbios_requests(acpi_handle handle, + struct atif_sbios_requests *req) +{ + union acpi_object *info; + size_t size; + int count = 0; + + info = radeon_atif_call(handle, ATIF_FUNCTION_GET_SYSTEM_BIOS_REQUESTS, NULL); + if (!info) + return -EIO; + + size = *(u16 *)info->buffer.pointer; + if (size < 0xd) { + count = -EINVAL; + goto out; + } + memset(req, 0, sizeof(*req)); + + size = min(sizeof(*req), size); + memcpy(req, info->buffer.pointer, size); + DRM_DEBUG_DRIVER("SBIOS pending requests: %#x\n", req->pending); + + count = hweight32(req->pending); + +out: + kfree(info); + return count; +} + +/** + * radeon_atif_handler - handle ATIF notify requests + * + * @rdev: radeon_device pointer + * @event: atif sbios request struct + * + * Checks the acpi event and if it matches an atif event, + * handles it. + * Returns NOTIFY code + */ +int radeon_atif_handler(struct radeon_device *rdev, + struct acpi_bus_event *event) +{ + struct radeon_atif *atif = &rdev->atif; + struct atif_sbios_requests req; + acpi_handle handle; + int count; + + DRM_DEBUG_DRIVER("event, device_class = %s, type = %#x\n", + event->device_class, event->type); + + if (strcmp(event->device_class, ACPI_VIDEO_CLASS) != 0) + return NOTIFY_DONE; + + if (!atif->notification_cfg.enabled || + event->type != atif->notification_cfg.command_code) + /* Not our event */ + return NOTIFY_DONE; + + /* Check pending SBIOS requests */ + handle = DEVICE_ACPI_HANDLE(&rdev->pdev->dev); + count = radeon_atif_get_sbios_requests(handle, &req); + + if (count <= 0) + return NOTIFY_DONE; + + DRM_DEBUG_DRIVER("ATIF: %d pending SBIOS requests\n", count); + + if (req.pending & ATIF_PANEL_BRIGHTNESS_CHANGE_REQUEST) { + struct radeon_encoder *enc = atif->encoder_for_bl; + + if (enc) { + DRM_DEBUG_DRIVER("Changing brightness to %d\n", + req.backlight_level); + + radeon_set_backlight_level(rdev, enc, req.backlight_level); + + if (rdev->is_atom_bios) { + struct radeon_encoder_atom_dig *dig = enc->enc_priv; + backlight_force_update(dig->bl_dev, + BACKLIGHT_UPDATE_HOTKEY); + } else { + struct radeon_encoder_lvds *dig = enc->enc_priv; + backlight_force_update(dig->bl_dev, + BACKLIGHT_UPDATE_HOTKEY); + } + } + } + /* TODO: check other events */ + + /* We've handled the event, stop the notifier chain. The ACPI interface + * overloads ACPI_VIDEO_NOTIFY_PROBE, we don't want to send that to + * userspace if the event was generated only to signal a SBIOS + * request. + */ + return NOTIFY_BAD; +} + +/* Call the ATCS method + */ +/** + * radeon_atcs_call - call an ATCS method + * + * @handle: acpi handle + * @function: the ATCS function to execute + * @params: ATCS function params + * + * Executes the requested ATCS function (all asics). + * Returns a pointer to the acpi output buffer. + */ +static union acpi_object *radeon_atcs_call(acpi_handle handle, int function, + struct acpi_buffer *params) +{ + acpi_status status; + union acpi_object atcs_arg_elements[2]; + struct acpi_object_list atcs_arg; + struct acpi_buffer buffer = { ACPI_ALLOCATE_BUFFER, NULL }; + + atcs_arg.count = 2; + atcs_arg.pointer = &atcs_arg_elements[0]; + + atcs_arg_elements[0].type = ACPI_TYPE_INTEGER; + atcs_arg_elements[0].integer.value = function; + + if (params) { + atcs_arg_elements[1].type = ACPI_TYPE_BUFFER; + atcs_arg_elements[1].buffer.length = params->length; + atcs_arg_elements[1].buffer.pointer = params->pointer; + } else { + /* We need a second fake parameter */ + atcs_arg_elements[1].type = ACPI_TYPE_INTEGER; + atcs_arg_elements[1].integer.value = 0; + } + + status = acpi_evaluate_object(handle, "ATCS", &atcs_arg, &buffer); + + /* Fail only if calling the method fails and ATIF is supported */ + if (ACPI_FAILURE(status) && status != AE_NOT_FOUND) { + DRM_DEBUG_DRIVER("failed to evaluate ATCS got %s\n", + acpi_format_exception(status)); + kfree(buffer.pointer); + return NULL; + } + + return buffer.pointer; +} + +/** + * radeon_atcs_parse_functions - parse supported functions + * + * @f: supported functions struct + * @mask: supported functions mask from ATCS + * + * Use the supported functions mask from ATCS function + * ATCS_FUNCTION_VERIFY_INTERFACE to determine what functions + * are supported (all asics). + */ +static void radeon_atcs_parse_functions(struct radeon_atcs_functions *f, u32 mask) +{ + f->get_ext_state = mask & ATCS_GET_EXTERNAL_STATE_SUPPORTED; + f->pcie_perf_req = mask & ATCS_PCIE_PERFORMANCE_REQUEST_SUPPORTED; + f->pcie_dev_rdy = mask & ATCS_PCIE_DEVICE_READY_NOTIFICATION_SUPPORTED; + f->pcie_bus_width = mask & ATCS_SET_PCIE_BUS_WIDTH_SUPPORTED; +} + +/** + * radeon_atcs_verify_interface - verify ATCS + * + * @handle: acpi handle + * @atcs: radeon atcs struct + * + * Execute the ATCS_FUNCTION_VERIFY_INTERFACE ATCS function + * to initialize ATCS and determine what features are supported + * (all asics). + * returns 0 on success, error on failure. + */ +static int radeon_atcs_verify_interface(acpi_handle handle, + struct radeon_atcs *atcs) +{ + union acpi_object *info; + struct atcs_verify_interface output; + size_t size; + int err = 0; + + info = radeon_atcs_call(handle, ATCS_FUNCTION_VERIFY_INTERFACE, NULL); + if (!info) + return -EIO; + + memset(&output, 0, sizeof(output)); + + size = *(u16 *) info->buffer.pointer; + if (size < 8) { + DRM_INFO("ATCS buffer is too small: %lu\n", size); + err = -EINVAL; + goto out; + } + size = min(sizeof(output), size); + + memcpy(&output, info->buffer.pointer, size); + + /* TODO: check version? */ + DRM_DEBUG_DRIVER("ATCS version %u\n", output.version); + + radeon_atcs_parse_functions(&atcs->functions, output.function_bits); + +out: + kfree(info); + return err; +} + +/** + * radeon_acpi_event - handle notify events + * + * @nb: notifier block + * @val: val + * @data: acpi event + * + * Calls relevant radeon functions in response to various + * acpi events. + * Returns NOTIFY code + */ +static int radeon_acpi_event(struct notifier_block *nb, + unsigned long val, + void *data) +{ + struct radeon_device *rdev = container_of(nb, struct radeon_device, acpi_nb); + struct acpi_bus_event *entry = (struct acpi_bus_event *)data; + + if (strcmp(entry->device_class, ACPI_AC_CLASS) == 0) { + if (power_supply_is_system_supplied() > 0) + DRM_DEBUG_DRIVER("pm: AC\n"); + else + DRM_DEBUG_DRIVER("pm: DC\n"); + + radeon_pm_acpi_event_handler(rdev); + } + + /* Check for pending SBIOS requests */ + return radeon_atif_handler(rdev, entry); } /* Call all ACPI methods here */ +/** + * radeon_acpi_init - init driver acpi support + * + * @rdev: radeon_device pointer + * + * Verifies the AMD ACPI interfaces and registers with the acpi + * notifier chain (all asics). + * Returns 0 on success, error on failure. + */ int radeon_acpi_init(struct radeon_device *rdev) { acpi_handle handle; + struct radeon_atif *atif = &rdev->atif; + struct radeon_atcs *atcs = &rdev->atcs; int ret; /* Get the device handle */ @@ -58,11 +560,90 @@ int radeon_acpi_init(struct radeon_device *rdev) if (!ASIC_IS_AVIVO(rdev) || !rdev->bios || !handle) return 0; + /* Call the ATCS method */ + ret = radeon_atcs_verify_interface(handle, atcs); + if (ret) { + DRM_DEBUG_DRIVER("Call to ATCS verify_interface failed: %d\n", ret); + } + /* Call the ATIF method */ - ret = radeon_atif_call(handle); - if (ret) - return ret; + ret = radeon_atif_verify_interface(handle, atif); + if (ret) { + DRM_DEBUG_DRIVER("Call to ATIF verify_interface failed: %d\n", ret); + goto out; + } + + if (atif->notifications.brightness_change) { + struct drm_encoder *tmp; + struct radeon_encoder *target = NULL; + + /* Find the encoder controlling the brightness */ + list_for_each_entry(tmp, &rdev->ddev->mode_config.encoder_list, + head) { + struct radeon_encoder *enc = to_radeon_encoder(tmp); + + if ((enc->devices & (ATOM_DEVICE_LCD_SUPPORT)) && + enc->enc_priv) { + if (rdev->is_atom_bios) { + struct radeon_encoder_atom_dig *dig = enc->enc_priv; + if (dig->bl_dev) { + target = enc; + break; + } + } else { + struct radeon_encoder_lvds *dig = enc->enc_priv; + if (dig->bl_dev) { + target = enc; + break; + } + } + } + } + + atif->encoder_for_bl = target; + if (!target) { + /* Brightness change notification is enabled, but we + * didn't find a backlight controller, this should + * never happen. + */ + DRM_ERROR("Cannot find a backlight controller\n"); + } + } - return 0; + if (atif->functions.sbios_requests && !atif->functions.system_params) { + /* XXX check this workraround, if sbios request function is + * present we have to see how it's configured in the system + * params + */ + atif->functions.system_params = true; + } + + if (atif->functions.system_params) { + ret = radeon_atif_get_notification_params(handle, + &atif->notification_cfg); + if (ret) { + DRM_DEBUG_DRIVER("Call to GET_SYSTEM_PARAMS failed: %d\n", + ret); + /* Disable notification */ + atif->notification_cfg.enabled = false; + } + } + +out: + rdev->acpi_nb.notifier_call = radeon_acpi_event; + register_acpi_notifier(&rdev->acpi_nb); + + return ret; } +/** + * radeon_acpi_fini - tear down driver acpi support + * + * @rdev: radeon_device pointer + * + * Unregisters with the acpi notifier chain (all asics). + */ +void radeon_acpi_fini(struct radeon_device *rdev) +{ + unregister_acpi_notifier(&rdev->acpi_nb); +} diff --git a/drivers/gpu/drm/radeon/radeon_acpi.h b/drivers/gpu/drm/radeon/radeon_acpi.h new file mode 100644 index 000000000000..be4af76f213d --- /dev/null +++ b/drivers/gpu/drm/radeon/radeon_acpi.h @@ -0,0 +1,445 @@ +/* + * Copyright 2012 Advanced Micro Devices, Inc. + * + * Permission is hereby granted, free of charge, to any person obtaining a + * copy of this software and associated documentation files (the "Software"), + * to deal in the Software without restriction, including without limitation + * the rights to use, copy, modify, merge, publish, distribute, sublicense, + * and/or sell copies of the Software, and to permit persons to whom the + * Software is furnished to do so, subject to the following conditions: + * + * The above copyright notice and this permission notice shall be included in + * all copies or substantial portions of the Software. + * + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL + * THE COPYRIGHT HOLDER(S) OR AUTHOR(S) BE LIABLE FOR ANY CLAIM, DAMAGES OR + * OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, + * ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR + * OTHER DEALINGS IN THE SOFTWARE. + * + */ + +#ifndef RADEON_ACPI_H +#define RADEON_ACPI_H + +struct radeon_device; +struct acpi_bus_event; + +int radeon_atif_handler(struct radeon_device *rdev, + struct acpi_bus_event *event); + +/* AMD hw uses four ACPI control methods: + * 1. ATIF + * ARG0: (ACPI_INTEGER) function code + * ARG1: (ACPI_BUFFER) parameter buffer, 256 bytes + * OUTPUT: (ACPI_BUFFER) output buffer, 256 bytes + * ATIF provides an entry point for the gfx driver to interact with the sbios. + * The AMD ACPI notification mechanism uses Notify (VGA, 0x81) or a custom + * notification. Which notification is used as indicated by the ATIF Control + * Method GET_SYSTEM_PARAMETERS. When the driver receives Notify (VGA, 0x81) or + * a custom notification it invokes ATIF Control Method GET_SYSTEM_BIOS_REQUESTS + * to identify pending System BIOS requests and associated parameters. For + * example, if one of the pending requests is DISPLAY_SWITCH_REQUEST, the driver + * will perform display device detection and invoke ATIF Control Method + * SELECT_ACTIVE_DISPLAYS. + * + * 2. ATPX + * ARG0: (ACPI_INTEGER) function code + * ARG1: (ACPI_BUFFER) parameter buffer, 256 bytes + * OUTPUT: (ACPI_BUFFER) output buffer, 256 bytes + * ATPX methods are used on PowerXpress systems to handle mux switching and + * discrete GPU power control. + * + * 3. ATRM + * ARG0: (ACPI_INTEGER) offset of vbios rom data + * ARG1: (ACPI_BUFFER) size of the buffer to fill (up to 4K). + * OUTPUT: (ACPI_BUFFER) output buffer + * ATRM provides an interfacess to access the discrete GPU vbios image on + * PowerXpress systems with multiple GPUs. + * + * 4. ATCS + * ARG0: (ACPI_INTEGER) function code + * ARG1: (ACPI_BUFFER) parameter buffer, 256 bytes + * OUTPUT: (ACPI_BUFFER) output buffer, 256 bytes + * ATCS provides an interface to AMD chipset specific functionality. + * + */ +/* ATIF */ +#define ATIF_FUNCTION_VERIFY_INTERFACE 0x0 +/* ARG0: ATIF_FUNCTION_VERIFY_INTERFACE + * ARG1: none + * OUTPUT: + * WORD - structure size in bytes (includes size field) + * WORD - version + * DWORD - supported notifications mask + * DWORD - supported functions bit vector + */ +/* Notifications mask */ +# define ATIF_DISPLAY_SWITCH_REQUEST_SUPPORTED (1 << 0) +# define ATIF_EXPANSION_MODE_CHANGE_REQUEST_SUPPORTED (1 << 1) +# define ATIF_THERMAL_STATE_CHANGE_REQUEST_SUPPORTED (1 << 2) +# define ATIF_FORCED_POWER_STATE_CHANGE_REQUEST_SUPPORTED (1 << 3) +# define ATIF_SYSTEM_POWER_SOURCE_CHANGE_REQUEST_SUPPORTED (1 << 4) +# define ATIF_DISPLAY_CONF_CHANGE_REQUEST_SUPPORTED (1 << 5) +# define ATIF_PX_GFX_SWITCH_REQUEST_SUPPORTED (1 << 6) +# define ATIF_PANEL_BRIGHTNESS_CHANGE_REQUEST_SUPPORTED (1 << 7) +# define ATIF_DGPU_DISPLAY_EVENT_SUPPORTED (1 << 8) +/* supported functions vector */ +# define ATIF_GET_SYSTEM_PARAMETERS_SUPPORTED (1 << 0) +# define ATIF_GET_SYSTEM_BIOS_REQUESTS_SUPPORTED (1 << 1) +# define ATIF_SELECT_ACTIVE_DISPLAYS_SUPPORTED (1 << 2) +# define ATIF_GET_LID_STATE_SUPPORTED (1 << 3) +# define ATIF_GET_TV_STANDARD_FROM_CMOS_SUPPORTED (1 << 4) +# define ATIF_SET_TV_STANDARD_IN_CMOS_SUPPORTED (1 << 5) +# define ATIF_GET_PANEL_EXPANSION_MODE_FROM_CMOS_SUPPORTED (1 << 6) +# define ATIF_SET_PANEL_EXPANSION_MODE_IN_CMOS_SUPPORTED (1 << 7) +# define ATIF_TEMPERATURE_CHANGE_NOTIFICATION_SUPPORTED (1 << 12) +# define ATIF_GET_GRAPHICS_DEVICE_TYPES_SUPPORTED (1 << 14) +#define ATIF_FUNCTION_GET_SYSTEM_PARAMETERS 0x1 +/* ARG0: ATIF_FUNCTION_GET_SYSTEM_PARAMETERS + * ARG1: none + * OUTPUT: + * WORD - structure size in bytes (includes size field) + * DWORD - valid flags mask + * DWORD - flags + * + * OR + * + * WORD - structure size in bytes (includes size field) + * DWORD - valid flags mask + * DWORD - flags + * BYTE - notify command code + * + * flags + * bits 1:0: + * 0 - Notify(VGA, 0x81) is not used for notification + * 1 - Notify(VGA, 0x81) is used for notification + * 2 - Notify(VGA, n) is used for notification where + * n (0xd0-0xd9) is specified in notify command code. + * bit 2: + * 1 - lid changes not reported though int10 + */ +#define ATIF_FUNCTION_GET_SYSTEM_BIOS_REQUESTS 0x2 +/* ARG0: ATIF_FUNCTION_GET_SYSTEM_BIOS_REQUESTS + * ARG1: none + * OUTPUT: + * WORD - structure size in bytes (includes size field) + * DWORD - pending sbios requests + * BYTE - panel expansion mode + * BYTE - thermal state: target gfx controller + * BYTE - thermal state: state id (0: exit state, non-0: state) + * BYTE - forced power state: target gfx controller + * BYTE - forced power state: state id + * BYTE - system power source + * BYTE - panel backlight level (0-255) + */ +/* pending sbios requests */ +# define ATIF_DISPLAY_SWITCH_REQUEST (1 << 0) +# define ATIF_EXPANSION_MODE_CHANGE_REQUEST (1 << 1) +# define ATIF_THERMAL_STATE_CHANGE_REQUEST (1 << 2) +# define ATIF_FORCED_POWER_STATE_CHANGE_REQUEST (1 << 3) +# define ATIF_SYSTEM_POWER_SOURCE_CHANGE_REQUEST (1 << 4) +# define ATIF_DISPLAY_CONF_CHANGE_REQUEST (1 << 5) +# define ATIF_PX_GFX_SWITCH_REQUEST (1 << 6) +# define ATIF_PANEL_BRIGHTNESS_CHANGE_REQUEST (1 << 7) +# define ATIF_DGPU_DISPLAY_EVENT (1 << 8) +/* panel expansion mode */ +# define ATIF_PANEL_EXPANSION_DISABLE 0 +# define ATIF_PANEL_EXPANSION_FULL 1 +# define ATIF_PANEL_EXPANSION_ASPECT 2 +/* target gfx controller */ +# define ATIF_TARGET_GFX_SINGLE 0 +# define ATIF_TARGET_GFX_PX_IGPU 1 +# define ATIF_TARGET_GFX_PX_DGPU 2 +/* system power source */ +# define ATIF_POWER_SOURCE_AC 1 +# define ATIF_POWER_SOURCE_DC 2 +# define ATIF_POWER_SOURCE_RESTRICTED_AC_1 3 +# define ATIF_POWER_SOURCE_RESTRICTED_AC_2 4 +#define ATIF_FUNCTION_SELECT_ACTIVE_DISPLAYS 0x3 +/* ARG0: ATIF_FUNCTION_SELECT_ACTIVE_DISPLAYS + * ARG1: + * WORD - structure size in bytes (includes size field) + * WORD - selected displays + * WORD - connected displays + * OUTPUT: + * WORD - structure size in bytes (includes size field) + * WORD - selected displays + */ +# define ATIF_LCD1 (1 << 0) +# define ATIF_CRT1 (1 << 1) +# define ATIF_TV (1 << 2) +# define ATIF_DFP1 (1 << 3) +# define ATIF_CRT2 (1 << 4) +# define ATIF_LCD2 (1 << 5) +# define ATIF_DFP2 (1 << 7) +# define ATIF_CV (1 << 8) +# define ATIF_DFP3 (1 << 9) +# define ATIF_DFP4 (1 << 10) +# define ATIF_DFP5 (1 << 11) +# define ATIF_DFP6 (1 << 12) +#define ATIF_FUNCTION_GET_LID_STATE 0x4 +/* ARG0: ATIF_FUNCTION_GET_LID_STATE + * ARG1: none + * OUTPUT: + * WORD - structure size in bytes (includes size field) + * BYTE - lid state (0: open, 1: closed) + * + * GET_LID_STATE only works at boot and resume, for general lid + * status, use the kernel provided status + */ +#define ATIF_FUNCTION_GET_TV_STANDARD_FROM_CMOS 0x5 +/* ARG0: ATIF_FUNCTION_GET_TV_STANDARD_FROM_CMOS + * ARG1: none + * OUTPUT: + * WORD - structure size in bytes (includes size field) + * BYTE - 0 + * BYTE - TV standard + */ +# define ATIF_TV_STD_NTSC 0 +# define ATIF_TV_STD_PAL 1 +# define ATIF_TV_STD_PALM 2 +# define ATIF_TV_STD_PAL60 3 +# define ATIF_TV_STD_NTSCJ 4 +# define ATIF_TV_STD_PALCN 5 +# define ATIF_TV_STD_PALN 6 +# define ATIF_TV_STD_SCART_RGB 9 +#define ATIF_FUNCTION_SET_TV_STANDARD_IN_CMOS 0x6 +/* ARG0: ATIF_FUNCTION_SET_TV_STANDARD_IN_CMOS + * ARG1: + * WORD - structure size in bytes (includes size field) + * BYTE - 0 + * BYTE - TV standard + * OUTPUT: none + */ +#define ATIF_FUNCTION_GET_PANEL_EXPANSION_MODE_FROM_CMOS 0x7 +/* ARG0: ATIF_FUNCTION_GET_PANEL_EXPANSION_MODE_FROM_CMOS + * ARG1: none + * OUTPUT: + * WORD - structure size in bytes (includes size field) + * BYTE - panel expansion mode + */ +#define ATIF_FUNCTION_SET_PANEL_EXPANSION_MODE_IN_CMOS 0x8 +/* ARG0: ATIF_FUNCTION_SET_PANEL_EXPANSION_MODE_IN_CMOS + * ARG1: + * WORD - structure size in bytes (includes size field) + * BYTE - panel expansion mode + * OUTPUT: none + */ +#define ATIF_FUNCTION_TEMPERATURE_CHANGE_NOTIFICATION 0xD +/* ARG0: ATIF_FUNCTION_TEMPERATURE_CHANGE_NOTIFICATION + * ARG1: + * WORD - structure size in bytes (includes size field) + * WORD - gfx controller id + * BYTE - current temperature (degress Celsius) + * OUTPUT: none + */ +#define ATIF_FUNCTION_GET_GRAPHICS_DEVICE_TYPES 0xF +/* ARG0: ATIF_FUNCTION_GET_GRAPHICS_DEVICE_TYPES + * ARG1: none + * OUTPUT: + * WORD - number of gfx devices + * WORD - device structure size in bytes (excludes device size field) + * DWORD - flags \ + * WORD - bus number } repeated structure + * WORD - device number / + */ +/* flags */ +# define ATIF_PX_REMOVABLE_GRAPHICS_DEVICE (1 << 0) +# define ATIF_XGP_PORT (1 << 1) +# define ATIF_VGA_ENABLED_GRAPHICS_DEVICE (1 << 2) +# define ATIF_XGP_PORT_IN_DOCK (1 << 3) + +/* ATPX */ +#define ATPX_FUNCTION_VERIFY_INTERFACE 0x0 +/* ARG0: ATPX_FUNCTION_VERIFY_INTERFACE + * ARG1: none + * OUTPUT: + * WORD - structure size in bytes (includes size field) + * WORD - version + * DWORD - supported functions bit vector + */ +/* supported functions vector */ +# define ATPX_GET_PX_PARAMETERS_SUPPORTED (1 << 0) +# define ATPX_POWER_CONTROL_SUPPORTED (1 << 1) +# define ATPX_DISPLAY_MUX_CONTROL_SUPPORTED (1 << 2) +# define ATPX_I2C_MUX_CONTROL_SUPPORTED (1 << 3) +# define ATPX_GRAPHICS_DEVICE_SWITCH_START_NOTIFICATION_SUPPORTED (1 << 4) +# define ATPX_GRAPHICS_DEVICE_SWITCH_END_NOTIFICATION_SUPPORTED (1 << 5) +# define ATPX_GET_DISPLAY_CONNECTORS_MAPPING_SUPPORTED (1 << 7) +# define ATPX_GET_DISPLAY_DETECTION_PORTS_SUPPORTED (1 << 8) +#define ATPX_FUNCTION_GET_PX_PARAMETERS 0x1 +/* ARG0: ATPX_FUNCTION_GET_PX_PARAMETERS + * ARG1: none + * OUTPUT: + * WORD - structure size in bytes (includes size field) + * DWORD - valid flags mask + * DWORD - flags + */ +/* flags */ +# define ATPX_LVDS_I2C_AVAILABLE_TO_BOTH_GPUS (1 << 0) +# define ATPX_CRT1_I2C_AVAILABLE_TO_BOTH_GPUS (1 << 1) +# define ATPX_DVI1_I2C_AVAILABLE_TO_BOTH_GPUS (1 << 2) +# define ATPX_CRT1_RGB_SIGNAL_MUXED (1 << 3) +# define ATPX_TV_SIGNAL_MUXED (1 << 4) +# define ATPX_DFP_SIGNAL_MUXED (1 << 5) +# define ATPX_SEPARATE_MUX_FOR_I2C (1 << 6) +# define ATPX_DYNAMIC_PX_SUPPORTED (1 << 7) +# define ATPX_ACF_NOT_SUPPORTED (1 << 8) +# define ATPX_FIXED_NOT_SUPPORTED (1 << 9) +# define ATPX_DYNAMIC_DGPU_POWER_OFF_SUPPORTED (1 << 10) +# define ATPX_DGPU_REQ_POWER_FOR_DISPLAYS (1 << 11) +#define ATPX_FUNCTION_POWER_CONTROL 0x2 +/* ARG0: ATPX_FUNCTION_POWER_CONTROL + * ARG1: + * WORD - structure size in bytes (includes size field) + * BYTE - dGPU power state (0: power off, 1: power on) + * OUTPUT: none + */ +#define ATPX_FUNCTION_DISPLAY_MUX_CONTROL 0x3 +/* ARG0: ATPX_FUNCTION_DISPLAY_MUX_CONTROL + * ARG1: + * WORD - structure size in bytes (includes size field) + * WORD - display mux control (0: iGPU, 1: dGPU) + * OUTPUT: none + */ +# define ATPX_INTEGRATED_GPU 0 +# define ATPX_DISCRETE_GPU 1 +#define ATPX_FUNCTION_I2C_MUX_CONTROL 0x4 +/* ARG0: ATPX_FUNCTION_I2C_MUX_CONTROL + * ARG1: + * WORD - structure size in bytes (includes size field) + * WORD - i2c/aux/hpd mux control (0: iGPU, 1: dGPU) + * OUTPUT: none + */ +#define ATPX_FUNCTION_GRAPHICS_DEVICE_SWITCH_START_NOTIFICATION 0x5 +/* ARG0: ATPX_FUNCTION_GRAPHICS_DEVICE_SWITCH_START_NOTIFICATION + * ARG1: + * WORD - structure size in bytes (includes size field) + * WORD - target gpu (0: iGPU, 1: dGPU) + * OUTPUT: none + */ +#define ATPX_FUNCTION_GRAPHICS_DEVICE_SWITCH_END_NOTIFICATION 0x6 +/* ARG0: ATPX_FUNCTION_GRAPHICS_DEVICE_SWITCH_END_NOTIFICATION + * ARG1: + * WORD - structure size in bytes (includes size field) + * WORD - target gpu (0: iGPU, 1: dGPU) + * OUTPUT: none + */ +#define ATPX_FUNCTION_GET_DISPLAY_CONNECTORS_MAPPING 0x8 +/* ARG0: ATPX_FUNCTION_GET_DISPLAY_CONNECTORS_MAPPING + * ARG1: none + * OUTPUT: + * WORD - number of display connectors + * WORD - connector structure size in bytes (excludes connector size field) + * BYTE - flags \ + * BYTE - ATIF display vector bit position } repeated + * BYTE - adapter id (0: iGPU, 1-n: dGPU ordered by pcie bus number) } structure + * WORD - connector ACPI id / + */ +/* flags */ +# define ATPX_DISPLAY_OUTPUT_SUPPORTED_BY_ADAPTER_ID_DEVICE (1 << 0) +# define ATPX_DISPLAY_HPD_SUPPORTED_BY_ADAPTER_ID_DEVICE (1 << 1) +# define ATPX_DISPLAY_I2C_SUPPORTED_BY_ADAPTER_ID_DEVICE (1 << 2) +#define ATPX_FUNCTION_GET_DISPLAY_DETECTION_PORTS 0x9 +/* ARG0: ATPX_FUNCTION_GET_DISPLAY_DETECTION_PORTS + * ARG1: none + * OUTPUT: + * WORD - number of HPD/DDC ports + * WORD - port structure size in bytes (excludes port size field) + * BYTE - ATIF display vector bit position \ + * BYTE - hpd id } reapeated structure + * BYTE - ddc id / + * + * available on A+A systems only + */ +/* hpd id */ +# define ATPX_HPD_NONE 0 +# define ATPX_HPD1 1 +# define ATPX_HPD2 2 +# define ATPX_HPD3 3 +# define ATPX_HPD4 4 +# define ATPX_HPD5 5 +# define ATPX_HPD6 6 +/* ddc id */ +# define ATPX_DDC_NONE 0 +# define ATPX_DDC1 1 +# define ATPX_DDC2 2 +# define ATPX_DDC3 3 +# define ATPX_DDC4 4 +# define ATPX_DDC5 5 +# define ATPX_DDC6 6 +# define ATPX_DDC7 7 +# define ATPX_DDC8 8 + +/* ATCS */ +#define ATCS_FUNCTION_VERIFY_INTERFACE 0x0 +/* ARG0: ATCS_FUNCTION_VERIFY_INTERFACE + * ARG1: none + * OUTPUT: + * WORD - structure size in bytes (includes size field) + * WORD - version + * DWORD - supported functions bit vector + */ +/* supported functions vector */ +# define ATCS_GET_EXTERNAL_STATE_SUPPORTED (1 << 0) +# define ATCS_PCIE_PERFORMANCE_REQUEST_SUPPORTED (1 << 1) +# define ATCS_PCIE_DEVICE_READY_NOTIFICATION_SUPPORTED (1 << 2) +# define ATCS_SET_PCIE_BUS_WIDTH_SUPPORTED (1 << 3) +#define ATCS_FUNCTION_GET_EXTERNAL_STATE 0x1 +/* ARG0: ATCS_FUNCTION_GET_EXTERNAL_STATE + * ARG1: none + * OUTPUT: + * WORD - structure size in bytes (includes size field) + * DWORD - valid flags mask + * DWORD - flags (0: undocked, 1: docked) + */ +/* flags */ +# define ATCS_DOCKED (1 << 0) +#define ATCS_FUNCTION_PCIE_PERFORMANCE_REQUEST 0x2 +/* ARG0: ATCS_FUNCTION_PCIE_PERFORMANCE_REQUEST + * ARG1: + * WORD - structure size in bytes (includes size field) + * WORD - client id (bit 2-0: func num, 7-3: dev num, 15-8: bus num) + * WORD - valid flags mask + * WORD - flags + * BYTE - request type + * BYTE - performance request + * OUTPUT: + * WORD - structure size in bytes (includes size field) + * BYTE - return value + */ +/* flags */ +# define ATCS_ADVERTISE_CAPS (1 << 0) +# define ATCS_WAIT_FOR_COMPLETION (1 << 1) +/* request type */ +# define ATCS_PCIE_LINK_SPEED 1 +/* performance request */ +# define ATCS_REMOVE 0 +# define ATCS_FORCE_LOW_POWER 1 +# define ATCS_PERF_LEVEL_1 2 /* PCIE Gen 1 */ +# define ATCS_PERF_LEVEL_2 3 /* PCIE Gen 2 */ +# define ATCS_PERF_LEVEL_3 4 /* PCIE Gen 3 */ +/* return value */ +# define ATCS_REQUEST_REFUSED 1 +# define ATCS_REQUEST_COMPLETE 2 +# define ATCS_REQUEST_IN_PROGRESS 3 +#define ATCS_FUNCTION_PCIE_DEVICE_READY_NOTIFICATION 0x3 +/* ARG0: ATCS_FUNCTION_PCIE_DEVICE_READY_NOTIFICATION + * ARG1: none + * OUTPUT: none + */ +#define ATCS_FUNCTION_SET_PCIE_BUS_WIDTH 0x4 +/* ARG0: ATCS_FUNCTION_SET_PCIE_BUS_WIDTH + * ARG1: + * WORD - structure size in bytes (includes size field) + * WORD - client id (bit 2-0: func num, 7-3: dev num, 15-8: bus num) + * BYTE - number of active lanes + * OUTPUT: + * WORD - structure size in bytes (includes size field) + * BYTE - number of active lanes + */ + +#endif diff --git a/drivers/gpu/drm/radeon/radeon_asic.c b/drivers/gpu/drm/radeon/radeon_asic.c index 973417c4b014..654520b95ab7 100644 --- a/drivers/gpu/drm/radeon/radeon_asic.c +++ b/drivers/gpu/drm/radeon/radeon_asic.c @@ -198,6 +198,8 @@ static struct radeon_asic r100_asic = { .bandwidth_update = &r100_bandwidth_update, .get_vblank_counter = &r100_get_vblank_counter, .wait_for_vblank = &r100_wait_for_vblank, + .set_backlight_level = &radeon_legacy_set_backlight_level, + .get_backlight_level = &radeon_legacy_get_backlight_level, }, .copy = { .blit = &r100_copy_blit, @@ -272,6 +274,8 @@ static struct radeon_asic r200_asic = { .bandwidth_update = &r100_bandwidth_update, .get_vblank_counter = &r100_get_vblank_counter, .wait_for_vblank = &r100_wait_for_vblank, + .set_backlight_level = &radeon_legacy_set_backlight_level, + .get_backlight_level = &radeon_legacy_get_backlight_level, }, .copy = { .blit = &r100_copy_blit, @@ -346,6 +350,8 @@ static struct radeon_asic r300_asic = { .bandwidth_update = &r100_bandwidth_update, .get_vblank_counter = &r100_get_vblank_counter, .wait_for_vblank = &r100_wait_for_vblank, + .set_backlight_level = &radeon_legacy_set_backlight_level, + .get_backlight_level = &radeon_legacy_get_backlight_level, }, .copy = { .blit = &r100_copy_blit, @@ -420,6 +426,8 @@ static struct radeon_asic r300_asic_pcie = { .bandwidth_update = &r100_bandwidth_update, .get_vblank_counter = &r100_get_vblank_counter, .wait_for_vblank = &r100_wait_for_vblank, + .set_backlight_level = &radeon_legacy_set_backlight_level, + .get_backlight_level = &radeon_legacy_get_backlight_level, }, .copy = { .blit = &r100_copy_blit, @@ -494,6 +502,8 @@ static struct radeon_asic r420_asic = { .bandwidth_update = &r100_bandwidth_update, .get_vblank_counter = &r100_get_vblank_counter, .wait_for_vblank = &r100_wait_for_vblank, + .set_backlight_level = &atombios_set_backlight_level, + .get_backlight_level = &atombios_get_backlight_level, }, .copy = { .blit = &r100_copy_blit, @@ -568,6 +578,8 @@ static struct radeon_asic rs400_asic = { .bandwidth_update = &r100_bandwidth_update, .get_vblank_counter = &r100_get_vblank_counter, .wait_for_vblank = &r100_wait_for_vblank, + .set_backlight_level = &radeon_legacy_set_backlight_level, + .get_backlight_level = &radeon_legacy_get_backlight_level, }, .copy = { .blit = &r100_copy_blit, @@ -642,6 +654,8 @@ static struct radeon_asic rs600_asic = { .bandwidth_update = &rs600_bandwidth_update, .get_vblank_counter = &rs600_get_vblank_counter, .wait_for_vblank = &avivo_wait_for_vblank, + .set_backlight_level = &atombios_set_backlight_level, + .get_backlight_level = &atombios_get_backlight_level, }, .copy = { .blit = &r100_copy_blit, @@ -716,6 +730,8 @@ static struct radeon_asic rs690_asic = { .get_vblank_counter = &rs600_get_vblank_counter, .bandwidth_update = &rs690_bandwidth_update, .wait_for_vblank = &avivo_wait_for_vblank, + .set_backlight_level = &atombios_set_backlight_level, + .get_backlight_level = &atombios_get_backlight_level, }, .copy = { .blit = &r100_copy_blit, @@ -790,6 +806,8 @@ static struct radeon_asic rv515_asic = { .get_vblank_counter = &rs600_get_vblank_counter, .bandwidth_update = &rv515_bandwidth_update, .wait_for_vblank = &avivo_wait_for_vblank, + .set_backlight_level = &atombios_set_backlight_level, + .get_backlight_level = &atombios_get_backlight_level, }, .copy = { .blit = &r100_copy_blit, @@ -864,6 +882,8 @@ static struct radeon_asic r520_asic = { .bandwidth_update = &rv515_bandwidth_update, .get_vblank_counter = &rs600_get_vblank_counter, .wait_for_vblank = &avivo_wait_for_vblank, + .set_backlight_level = &atombios_set_backlight_level, + .get_backlight_level = &atombios_get_backlight_level, }, .copy = { .blit = &r100_copy_blit, @@ -937,6 +957,8 @@ static struct radeon_asic r600_asic = { .bandwidth_update = &rv515_bandwidth_update, .get_vblank_counter = &rs600_get_vblank_counter, .wait_for_vblank = &avivo_wait_for_vblank, + .set_backlight_level = &atombios_set_backlight_level, + .get_backlight_level = &atombios_get_backlight_level, }, .copy = { .blit = &r600_copy_blit, @@ -1010,6 +1032,8 @@ static struct radeon_asic rs780_asic = { .bandwidth_update = &rs690_bandwidth_update, .get_vblank_counter = &rs600_get_vblank_counter, .wait_for_vblank = &avivo_wait_for_vblank, + .set_backlight_level = &atombios_set_backlight_level, + .get_backlight_level = &atombios_get_backlight_level, }, .copy = { .blit = &r600_copy_blit, @@ -1083,6 +1107,8 @@ static struct radeon_asic rv770_asic = { .bandwidth_update = &rv515_bandwidth_update, .get_vblank_counter = &rs600_get_vblank_counter, .wait_for_vblank = &avivo_wait_for_vblank, + .set_backlight_level = &atombios_set_backlight_level, + .get_backlight_level = &atombios_get_backlight_level, }, .copy = { .blit = &r600_copy_blit, @@ -1156,6 +1182,8 @@ static struct radeon_asic evergreen_asic = { .bandwidth_update = &evergreen_bandwidth_update, .get_vblank_counter = &evergreen_get_vblank_counter, .wait_for_vblank = &dce4_wait_for_vblank, + .set_backlight_level = &atombios_set_backlight_level, + .get_backlight_level = &atombios_get_backlight_level, }, .copy = { .blit = &r600_copy_blit, @@ -1229,6 +1257,8 @@ static struct radeon_asic sumo_asic = { .bandwidth_update = &evergreen_bandwidth_update, .get_vblank_counter = &evergreen_get_vblank_counter, .wait_for_vblank = &dce4_wait_for_vblank, + .set_backlight_level = &atombios_set_backlight_level, + .get_backlight_level = &atombios_get_backlight_level, }, .copy = { .blit = &r600_copy_blit, @@ -1302,6 +1332,8 @@ static struct radeon_asic btc_asic = { .bandwidth_update = &evergreen_bandwidth_update, .get_vblank_counter = &evergreen_get_vblank_counter, .wait_for_vblank = &dce4_wait_for_vblank, + .set_backlight_level = &atombios_set_backlight_level, + .get_backlight_level = &atombios_get_backlight_level, }, .copy = { .blit = &r600_copy_blit, @@ -1325,7 +1357,7 @@ static struct radeon_asic btc_asic = { .misc = &evergreen_pm_misc, .prepare = &evergreen_pm_prepare, .finish = &evergreen_pm_finish, - .init_profile = &r600_pm_init_profile, + .init_profile = &btc_pm_init_profile, .get_dynpm_state = &r600_pm_get_dynpm_state, .get_engine_clock = &radeon_atom_get_engine_clock, .set_engine_clock = &radeon_atom_set_engine_clock, @@ -1342,16 +1374,6 @@ static struct radeon_asic btc_asic = { }, }; -static const struct radeon_vm_funcs cayman_vm_funcs = { - .init = &cayman_vm_init, - .fini = &cayman_vm_fini, - .bind = &cayman_vm_bind, - .unbind = &cayman_vm_unbind, - .tlb_flush = &cayman_vm_tlb_flush, - .page_flags = &cayman_vm_page_flags, - .set_page = &cayman_vm_set_page, -}; - static struct radeon_asic cayman_asic = { .init = &cayman_init, .fini = &cayman_fini, @@ -1366,6 +1388,12 @@ static struct radeon_asic cayman_asic = { .tlb_flush = &cayman_pcie_gart_tlb_flush, .set_page = &rs600_gart_set_page, }, + .vm = { + .init = &cayman_vm_init, + .fini = &cayman_vm_fini, + .pt_ring_index = RADEON_RING_TYPE_GFX_INDEX, + .set_page = &cayman_vm_set_page, + }, .ring = { [RADEON_RING_TYPE_GFX_INDEX] = { .ib_execute = &cayman_ring_ib_execute, @@ -1376,6 +1404,7 @@ static struct radeon_asic cayman_asic = { .ring_test = &r600_ring_test, .ib_test = &r600_ib_test, .is_lockup = &evergreen_gpu_is_lockup, + .vm_flush = &cayman_vm_flush, }, [CAYMAN_RING_TYPE_CP1_INDEX] = { .ib_execute = &cayman_ring_ib_execute, @@ -1386,6 +1415,7 @@ static struct radeon_asic cayman_asic = { .ring_test = &r600_ring_test, .ib_test = &r600_ib_test, .is_lockup = &evergreen_gpu_is_lockup, + .vm_flush = &cayman_vm_flush, }, [CAYMAN_RING_TYPE_CP2_INDEX] = { .ib_execute = &cayman_ring_ib_execute, @@ -1396,6 +1426,7 @@ static struct radeon_asic cayman_asic = { .ring_test = &r600_ring_test, .ib_test = &r600_ib_test, .is_lockup = &evergreen_gpu_is_lockup, + .vm_flush = &cayman_vm_flush, } }, .irq = { @@ -1406,6 +1437,8 @@ static struct radeon_asic cayman_asic = { .bandwidth_update = &evergreen_bandwidth_update, .get_vblank_counter = &evergreen_get_vblank_counter, .wait_for_vblank = &dce4_wait_for_vblank, + .set_backlight_level = &atombios_set_backlight_level, + .get_backlight_level = &atombios_get_backlight_level, }, .copy = { .blit = &r600_copy_blit, @@ -1429,7 +1462,7 @@ static struct radeon_asic cayman_asic = { .misc = &evergreen_pm_misc, .prepare = &evergreen_pm_prepare, .finish = &evergreen_pm_finish, - .init_profile = &r600_pm_init_profile, + .init_profile = &btc_pm_init_profile, .get_dynpm_state = &r600_pm_get_dynpm_state, .get_engine_clock = &radeon_atom_get_engine_clock, .set_engine_clock = &radeon_atom_set_engine_clock, @@ -1460,6 +1493,12 @@ static struct radeon_asic trinity_asic = { .tlb_flush = &cayman_pcie_gart_tlb_flush, .set_page = &rs600_gart_set_page, }, + .vm = { + .init = &cayman_vm_init, + .fini = &cayman_vm_fini, + .pt_ring_index = RADEON_RING_TYPE_GFX_INDEX, + .set_page = &cayman_vm_set_page, + }, .ring = { [RADEON_RING_TYPE_GFX_INDEX] = { .ib_execute = &cayman_ring_ib_execute, @@ -1470,6 +1509,7 @@ static struct radeon_asic trinity_asic = { .ring_test = &r600_ring_test, .ib_test = &r600_ib_test, .is_lockup = &evergreen_gpu_is_lockup, + .vm_flush = &cayman_vm_flush, }, [CAYMAN_RING_TYPE_CP1_INDEX] = { .ib_execute = &cayman_ring_ib_execute, @@ -1480,6 +1520,7 @@ static struct radeon_asic trinity_asic = { .ring_test = &r600_ring_test, .ib_test = &r600_ib_test, .is_lockup = &evergreen_gpu_is_lockup, + .vm_flush = &cayman_vm_flush, }, [CAYMAN_RING_TYPE_CP2_INDEX] = { .ib_execute = &cayman_ring_ib_execute, @@ -1490,6 +1531,7 @@ static struct radeon_asic trinity_asic = { .ring_test = &r600_ring_test, .ib_test = &r600_ib_test, .is_lockup = &evergreen_gpu_is_lockup, + .vm_flush = &cayman_vm_flush, } }, .irq = { @@ -1500,6 +1542,8 @@ static struct radeon_asic trinity_asic = { .bandwidth_update = &dce6_bandwidth_update, .get_vblank_counter = &evergreen_get_vblank_counter, .wait_for_vblank = &dce4_wait_for_vblank, + .set_backlight_level = &atombios_set_backlight_level, + .get_backlight_level = &atombios_get_backlight_level, }, .copy = { .blit = &r600_copy_blit, @@ -1540,16 +1584,6 @@ static struct radeon_asic trinity_asic = { }, }; -static const struct radeon_vm_funcs si_vm_funcs = { - .init = &si_vm_init, - .fini = &si_vm_fini, - .bind = &si_vm_bind, - .unbind = &si_vm_unbind, - .tlb_flush = &si_vm_tlb_flush, - .page_flags = &cayman_vm_page_flags, - .set_page = &cayman_vm_set_page, -}; - static struct radeon_asic si_asic = { .init = &si_init, .fini = &si_fini, @@ -1564,6 +1598,12 @@ static struct radeon_asic si_asic = { .tlb_flush = &si_pcie_gart_tlb_flush, .set_page = &rs600_gart_set_page, }, + .vm = { + .init = &si_vm_init, + .fini = &si_vm_fini, + .pt_ring_index = RADEON_RING_TYPE_GFX_INDEX, + .set_page = &si_vm_set_page, + }, .ring = { [RADEON_RING_TYPE_GFX_INDEX] = { .ib_execute = &si_ring_ib_execute, @@ -1574,6 +1614,7 @@ static struct radeon_asic si_asic = { .ring_test = &r600_ring_test, .ib_test = &r600_ib_test, .is_lockup = &si_gpu_is_lockup, + .vm_flush = &si_vm_flush, }, [CAYMAN_RING_TYPE_CP1_INDEX] = { .ib_execute = &si_ring_ib_execute, @@ -1584,6 +1625,7 @@ static struct radeon_asic si_asic = { .ring_test = &r600_ring_test, .ib_test = &r600_ib_test, .is_lockup = &si_gpu_is_lockup, + .vm_flush = &si_vm_flush, }, [CAYMAN_RING_TYPE_CP2_INDEX] = { .ib_execute = &si_ring_ib_execute, @@ -1594,6 +1636,7 @@ static struct radeon_asic si_asic = { .ring_test = &r600_ring_test, .ib_test = &r600_ib_test, .is_lockup = &si_gpu_is_lockup, + .vm_flush = &si_vm_flush, } }, .irq = { @@ -1604,6 +1647,8 @@ static struct radeon_asic si_asic = { .bandwidth_update = &dce6_bandwidth_update, .get_vblank_counter = &evergreen_get_vblank_counter, .wait_for_vblank = &dce4_wait_for_vblank, + .set_backlight_level = &atombios_set_backlight_level, + .get_backlight_level = &atombios_get_backlight_level, }, .copy = { .blit = NULL, @@ -1697,6 +1742,7 @@ int radeon_asic_init(struct radeon_device *rdev) rdev->asic->pm.set_engine_clock = &radeon_legacy_set_engine_clock; rdev->asic->pm.get_memory_clock = &radeon_legacy_get_memory_clock; rdev->asic->pm.set_memory_clock = NULL; + rdev->asic->display.set_backlight_level = &radeon_legacy_set_backlight_level; } break; case CHIP_RS400: @@ -1769,13 +1815,11 @@ int radeon_asic_init(struct radeon_device *rdev) rdev->asic = &cayman_asic; /* set num crtcs */ rdev->num_crtc = 6; - rdev->vm_manager.funcs = &cayman_vm_funcs; break; case CHIP_ARUBA: rdev->asic = &trinity_asic; /* set num crtcs */ rdev->num_crtc = 4; - rdev->vm_manager.funcs = &cayman_vm_funcs; break; case CHIP_TAHITI: case CHIP_PITCAIRN: @@ -1783,7 +1827,6 @@ int radeon_asic_init(struct radeon_device *rdev) rdev->asic = &si_asic; /* set num crtcs */ rdev->num_crtc = 6; - rdev->vm_manager.funcs = &si_vm_funcs; break; default: /* FIXME: not supported yet */ diff --git a/drivers/gpu/drm/radeon/radeon_asic.h b/drivers/gpu/drm/radeon/radeon_asic.h index 18c38d14c8cd..5e3a0e5c6be1 100644 --- a/drivers/gpu/drm/radeon/radeon_asic.h +++ b/drivers/gpu/drm/radeon/radeon_asic.h @@ -42,6 +42,12 @@ uint32_t radeon_atom_get_memory_clock(struct radeon_device *rdev); void radeon_atom_set_memory_clock(struct radeon_device *rdev, uint32_t mem_clock); void radeon_atom_set_clock_gating(struct radeon_device *rdev, int enable); +void atombios_set_backlight_level(struct radeon_encoder *radeon_encoder, u8 level); +u8 atombios_get_backlight_level(struct radeon_encoder *radeon_encoder); +void radeon_legacy_set_backlight_level(struct radeon_encoder *radeon_encoder, u8 level); +u8 radeon_legacy_get_backlight_level(struct radeon_encoder *radeon_encoder); + + /* * r100,rv100,rs100,rv200,rs200 */ @@ -389,6 +395,7 @@ void r700_cp_fini(struct radeon_device *rdev); struct evergreen_mc_save { u32 vga_render_control; u32 vga_hdp_control; + bool crtc_enabled[RADEON_MAX_CRTCS]; }; void evergreen_pcie_gart_tlb_flush(struct radeon_device *rdev); @@ -413,6 +420,7 @@ extern void evergreen_pm_misc(struct radeon_device *rdev); extern void evergreen_pm_prepare(struct radeon_device *rdev); extern void evergreen_pm_finish(struct radeon_device *rdev); extern void sumo_pm_init_profile(struct radeon_device *rdev); +extern void btc_pm_init_profile(struct radeon_device *rdev); extern void evergreen_pre_page_flip(struct radeon_device *rdev, int crtc); extern u32 evergreen_page_flip(struct radeon_device *rdev, int crtc, u64 crtc_base); extern void evergreen_post_page_flip(struct radeon_device *rdev, int crtc); @@ -435,14 +443,11 @@ int cayman_asic_reset(struct radeon_device *rdev); void cayman_ring_ib_execute(struct radeon_device *rdev, struct radeon_ib *ib); int cayman_vm_init(struct radeon_device *rdev); void cayman_vm_fini(struct radeon_device *rdev); -int cayman_vm_bind(struct radeon_device *rdev, struct radeon_vm *vm, int id); -void cayman_vm_unbind(struct radeon_device *rdev, struct radeon_vm *vm); -void cayman_vm_tlb_flush(struct radeon_device *rdev, struct radeon_vm *vm); -uint32_t cayman_vm_page_flags(struct radeon_device *rdev, - struct radeon_vm *vm, - uint32_t flags); -void cayman_vm_set_page(struct radeon_device *rdev, struct radeon_vm *vm, - unsigned pfn, uint64_t addr, uint32_t flags); +void cayman_vm_flush(struct radeon_device *rdev, int ridx, struct radeon_vm *vm); +uint32_t cayman_vm_page_flags(struct radeon_device *rdev, uint32_t flags); +void cayman_vm_set_page(struct radeon_device *rdev, uint64_t pe, + uint64_t addr, unsigned count, + uint32_t incr, uint32_t flags); int evergreen_ib_parse(struct radeon_device *rdev, struct radeon_ib *ib); /* DCE6 - SI */ @@ -465,9 +470,10 @@ int si_irq_set(struct radeon_device *rdev); int si_irq_process(struct radeon_device *rdev); int si_vm_init(struct radeon_device *rdev); void si_vm_fini(struct radeon_device *rdev); -int si_vm_bind(struct radeon_device *rdev, struct radeon_vm *vm, int id); -void si_vm_unbind(struct radeon_device *rdev, struct radeon_vm *vm); -void si_vm_tlb_flush(struct radeon_device *rdev, struct radeon_vm *vm); +void si_vm_set_page(struct radeon_device *rdev, uint64_t pe, + uint64_t addr, unsigned count, + uint32_t incr, uint32_t flags); +void si_vm_flush(struct radeon_device *rdev, int ridx, struct radeon_vm *vm); int si_ib_parse(struct radeon_device *rdev, struct radeon_ib *ib); uint64_t si_get_gpu_clock(struct radeon_device *rdev); diff --git a/drivers/gpu/drm/radeon/radeon_atombios.c b/drivers/gpu/drm/radeon/radeon_atombios.c index d67d4f3eb6f4..01b90b4f5e22 100644 --- a/drivers/gpu/drm/radeon/radeon_atombios.c +++ b/drivers/gpu/drm/radeon/radeon_atombios.c @@ -1254,6 +1254,10 @@ bool radeon_atom_get_clock_info(struct drm_device *dev) if (rdev->clock.max_pixel_clock == 0) rdev->clock.max_pixel_clock = 40000; + /* not technically a clock, but... */ + rdev->mode_info.firmware_flags = + le16_to_cpu(firmware_info->info.usFirmwareCapability.susAccess); + return true; } @@ -2005,7 +2009,8 @@ static int radeon_atombios_parse_power_table_1_3(struct radeon_device *rdev) power_info = (union power_info *)(mode_info->atom_context->bios + data_offset); /* add the i2c bus for thermal/fan chip */ - if (power_info->info.ucOverdriveThermalController > 0) { + if ((power_info->info.ucOverdriveThermalController > 0) && + (power_info->info.ucOverdriveThermalController < ARRAY_SIZE(thermal_controller_names))) { DRM_INFO("Possible %s thermal controller at 0x%02x\n", thermal_controller_names[power_info->info.ucOverdriveThermalController], power_info->info.ucOverdriveControllerAddress >> 1); @@ -2209,7 +2214,7 @@ static void radeon_atombios_add_pplib_thermal_controller(struct radeon_device *r (controller->ucType == ATOM_PP_THERMALCONTROLLER_EMC2103_WITH_INTERNAL)) { DRM_INFO("Special thermal controller config\n"); - } else { + } else if (controller->ucType < ARRAY_SIZE(pp_lib_thermal_controller_names)) { DRM_INFO("Possible %s thermal controller at 0x%02x %s fan control\n", pp_lib_thermal_controller_names[controller->ucType], controller->ucI2cAddress >> 1, @@ -2224,6 +2229,12 @@ static void radeon_atombios_add_pplib_thermal_controller(struct radeon_device *r strlcpy(info.type, name, sizeof(info.type)); i2c_new_device(&rdev->pm.i2c_bus->adapter, &info); } + } else { + DRM_INFO("Unknown thermal controller type %d at 0x%02x %s fan control\n", + controller->ucType, + controller->ucI2cAddress >> 1, + (controller->ucFanParameters & + ATOM_PP_FANPARAMETERS_NOFAN) ? "without" : "with"); } } } diff --git a/drivers/gpu/drm/radeon/radeon_atpx_handler.c b/drivers/gpu/drm/radeon/radeon_atpx_handler.c index 2a2cf0b88a28..582e99449c12 100644 --- a/drivers/gpu/drm/radeon/radeon_atpx_handler.c +++ b/drivers/gpu/drm/radeon/radeon_atpx_handler.c @@ -12,30 +12,62 @@ #include <acpi/acpi_bus.h> #include <linux/pci.h> -#define ATPX_VERSION 0 -#define ATPX_GPU_PWR 2 -#define ATPX_MUX_SELECT 3 -#define ATPX_I2C_MUX_SELECT 4 -#define ATPX_SWITCH_START 5 -#define ATPX_SWITCH_END 6 - -#define ATPX_INTEGRATED 0 -#define ATPX_DISCRETE 1 +#include "radeon_acpi.h" + +struct radeon_atpx_functions { + bool px_params; + bool power_cntl; + bool disp_mux_cntl; + bool i2c_mux_cntl; + bool switch_start; + bool switch_end; + bool disp_connectors_mapping; + bool disp_detetion_ports; +}; -#define ATPX_MUX_IGD 0 -#define ATPX_MUX_DISCRETE 1 +struct radeon_atpx { + acpi_handle handle; + struct radeon_atpx_functions functions; +}; static struct radeon_atpx_priv { bool atpx_detected; /* handle for device - and atpx */ acpi_handle dhandle; - acpi_handle atpx_handle; + struct radeon_atpx atpx; } radeon_atpx_priv; -static int radeon_atpx_get_version(acpi_handle handle) +struct atpx_verify_interface { + u16 size; /* structure size in bytes (includes size field) */ + u16 version; /* version */ + u32 function_bits; /* supported functions bit vector */ +} __packed; + +struct atpx_power_control { + u16 size; + u8 dgpu_state; +} __packed; + +struct atpx_mux { + u16 size; + u16 mux; +} __packed; + +/** + * radeon_atpx_call - call an ATPX method + * + * @handle: acpi handle + * @function: the ATPX function to execute + * @params: ATPX function params + * + * Executes the requested ATPX function (all asics). + * Returns a pointer to the acpi output buffer. + */ +static union acpi_object *radeon_atpx_call(acpi_handle handle, int function, + struct acpi_buffer *params) { acpi_status status; - union acpi_object atpx_arg_elements[2], *obj; + union acpi_object atpx_arg_elements[2]; struct acpi_object_list atpx_arg; struct acpi_buffer buffer = { ACPI_ALLOCATE_BUFFER, NULL }; @@ -43,99 +75,292 @@ static int radeon_atpx_get_version(acpi_handle handle) atpx_arg.pointer = &atpx_arg_elements[0]; atpx_arg_elements[0].type = ACPI_TYPE_INTEGER; - atpx_arg_elements[0].integer.value = ATPX_VERSION; + atpx_arg_elements[0].integer.value = function; + + if (params) { + atpx_arg_elements[1].type = ACPI_TYPE_BUFFER; + atpx_arg_elements[1].buffer.length = params->length; + atpx_arg_elements[1].buffer.pointer = params->pointer; + } else { + /* We need a second fake parameter */ + atpx_arg_elements[1].type = ACPI_TYPE_INTEGER; + atpx_arg_elements[1].integer.value = 0; + } - atpx_arg_elements[1].type = ACPI_TYPE_INTEGER; - atpx_arg_elements[1].integer.value = ATPX_VERSION; + status = acpi_evaluate_object(handle, "ATPX", &atpx_arg, &buffer); - status = acpi_evaluate_object(handle, NULL, &atpx_arg, &buffer); - if (ACPI_FAILURE(status)) { - printk("%s: failed to call ATPX: %s\n", __func__, acpi_format_exception(status)); - return -ENOSYS; + /* Fail only if calling the method fails and ATPX is supported */ + if (ACPI_FAILURE(status) && status != AE_NOT_FOUND) { + printk("failed to evaluate ATPX got %s\n", + acpi_format_exception(status)); + kfree(buffer.pointer); + return NULL; } - obj = (union acpi_object *)buffer.pointer; - if (obj && (obj->type == ACPI_TYPE_BUFFER)) - printk(KERN_INFO "radeon atpx: version is %d\n", *((u8 *)(obj->buffer.pointer) + 2)); - kfree(buffer.pointer); - return 0; + + return buffer.pointer; } -static int radeon_atpx_execute(acpi_handle handle, int cmd_id, u16 value) +/** + * radeon_atpx_parse_functions - parse supported functions + * + * @f: supported functions struct + * @mask: supported functions mask from ATPX + * + * Use the supported functions mask from ATPX function + * ATPX_FUNCTION_VERIFY_INTERFACE to determine what functions + * are supported (all asics). + */ +static void radeon_atpx_parse_functions(struct radeon_atpx_functions *f, u32 mask) { - acpi_status status; - union acpi_object atpx_arg_elements[2]; - struct acpi_object_list atpx_arg; - struct acpi_buffer buffer = { ACPI_ALLOCATE_BUFFER, NULL }; - uint8_t buf[4] = {0}; - - if (!handle) - return -EINVAL; - - atpx_arg.count = 2; - atpx_arg.pointer = &atpx_arg_elements[0]; + f->px_params = mask & ATPX_GET_PX_PARAMETERS_SUPPORTED; + f->power_cntl = mask & ATPX_POWER_CONTROL_SUPPORTED; + f->disp_mux_cntl = mask & ATPX_DISPLAY_MUX_CONTROL_SUPPORTED; + f->i2c_mux_cntl = mask & ATPX_I2C_MUX_CONTROL_SUPPORTED; + f->switch_start = mask & ATPX_GRAPHICS_DEVICE_SWITCH_START_NOTIFICATION_SUPPORTED; + f->switch_end = mask & ATPX_GRAPHICS_DEVICE_SWITCH_END_NOTIFICATION_SUPPORTED; + f->disp_connectors_mapping = mask & ATPX_GET_DISPLAY_CONNECTORS_MAPPING_SUPPORTED; + f->disp_detetion_ports = mask & ATPX_GET_DISPLAY_DETECTION_PORTS_SUPPORTED; +} - atpx_arg_elements[0].type = ACPI_TYPE_INTEGER; - atpx_arg_elements[0].integer.value = cmd_id; +/** + * radeon_atpx_verify_interface - verify ATPX + * + * @handle: acpi handle + * @atpx: radeon atpx struct + * + * Execute the ATPX_FUNCTION_VERIFY_INTERFACE ATPX function + * to initialize ATPX and determine what features are supported + * (all asics). + * returns 0 on success, error on failure. + */ +static int radeon_atpx_verify_interface(struct radeon_atpx *atpx) +{ + union acpi_object *info; + struct atpx_verify_interface output; + size_t size; + int err = 0; + + info = radeon_atpx_call(atpx->handle, ATPX_FUNCTION_VERIFY_INTERFACE, NULL); + if (!info) + return -EIO; + + memset(&output, 0, sizeof(output)); + + size = *(u16 *) info->buffer.pointer; + if (size < 8) { + printk("ATPX buffer is too small: %lu\n", size); + err = -EINVAL; + goto out; + } + size = min(sizeof(output), size); - buf[2] = value & 0xff; - buf[3] = (value >> 8) & 0xff; + memcpy(&output, info->buffer.pointer, size); - atpx_arg_elements[1].type = ACPI_TYPE_BUFFER; - atpx_arg_elements[1].buffer.length = 4; - atpx_arg_elements[1].buffer.pointer = buf; + /* TODO: check version? */ + printk("ATPX version %u\n", output.version); - status = acpi_evaluate_object(handle, NULL, &atpx_arg, &buffer); - if (ACPI_FAILURE(status)) { - printk("%s: failed to call ATPX: %s\n", __func__, acpi_format_exception(status)); - return -ENOSYS; - } - kfree(buffer.pointer); + radeon_atpx_parse_functions(&atpx->functions, output.function_bits); - return 0; +out: + kfree(info); + return err; } -static int radeon_atpx_set_discrete_state(acpi_handle handle, int state) +/** + * radeon_atpx_set_discrete_state - power up/down discrete GPU + * + * @atpx: atpx info struct + * @state: discrete GPU state (0 = power down, 1 = power up) + * + * Execute the ATPX_FUNCTION_POWER_CONTROL ATPX function to + * power down/up the discrete GPU (all asics). + * Returns 0 on success, error on failure. + */ +static int radeon_atpx_set_discrete_state(struct radeon_atpx *atpx, u8 state) { - return radeon_atpx_execute(handle, ATPX_GPU_PWR, state); + struct acpi_buffer params; + union acpi_object *info; + struct atpx_power_control input; + + if (atpx->functions.power_cntl) { + input.size = 3; + input.dgpu_state = state; + params.length = input.size; + params.pointer = &input; + info = radeon_atpx_call(atpx->handle, + ATPX_FUNCTION_POWER_CONTROL, + ¶ms); + if (!info) + return -EIO; + kfree(info); + } + return 0; } -static int radeon_atpx_switch_mux(acpi_handle handle, int mux_id) +/** + * radeon_atpx_switch_disp_mux - switch display mux + * + * @atpx: atpx info struct + * @mux_id: mux state (0 = integrated GPU, 1 = discrete GPU) + * + * Execute the ATPX_FUNCTION_DISPLAY_MUX_CONTROL ATPX function to + * switch the display mux between the discrete GPU and integrated GPU + * (all asics). + * Returns 0 on success, error on failure. + */ +static int radeon_atpx_switch_disp_mux(struct radeon_atpx *atpx, u16 mux_id) { - return radeon_atpx_execute(handle, ATPX_MUX_SELECT, mux_id); + struct acpi_buffer params; + union acpi_object *info; + struct atpx_mux input; + + if (atpx->functions.disp_mux_cntl) { + input.size = 4; + input.mux = mux_id; + params.length = input.size; + params.pointer = &input; + info = radeon_atpx_call(atpx->handle, + ATPX_FUNCTION_DISPLAY_MUX_CONTROL, + ¶ms); + if (!info) + return -EIO; + kfree(info); + } + return 0; } -static int radeon_atpx_switch_i2c_mux(acpi_handle handle, int mux_id) +/** + * radeon_atpx_switch_i2c_mux - switch i2c/hpd mux + * + * @atpx: atpx info struct + * @mux_id: mux state (0 = integrated GPU, 1 = discrete GPU) + * + * Execute the ATPX_FUNCTION_I2C_MUX_CONTROL ATPX function to + * switch the i2c/hpd mux between the discrete GPU and integrated GPU + * (all asics). + * Returns 0 on success, error on failure. + */ +static int radeon_atpx_switch_i2c_mux(struct radeon_atpx *atpx, u16 mux_id) { - return radeon_atpx_execute(handle, ATPX_I2C_MUX_SELECT, mux_id); + struct acpi_buffer params; + union acpi_object *info; + struct atpx_mux input; + + if (atpx->functions.i2c_mux_cntl) { + input.size = 4; + input.mux = mux_id; + params.length = input.size; + params.pointer = &input; + info = radeon_atpx_call(atpx->handle, + ATPX_FUNCTION_I2C_MUX_CONTROL, + ¶ms); + if (!info) + return -EIO; + kfree(info); + } + return 0; } -static int radeon_atpx_switch_start(acpi_handle handle, int gpu_id) +/** + * radeon_atpx_switch_start - notify the sbios of a GPU switch + * + * @atpx: atpx info struct + * @mux_id: mux state (0 = integrated GPU, 1 = discrete GPU) + * + * Execute the ATPX_FUNCTION_GRAPHICS_DEVICE_SWITCH_START_NOTIFICATION ATPX + * function to notify the sbios that a switch between the discrete GPU and + * integrated GPU has begun (all asics). + * Returns 0 on success, error on failure. + */ +static int radeon_atpx_switch_start(struct radeon_atpx *atpx, u16 mux_id) { - return radeon_atpx_execute(handle, ATPX_SWITCH_START, gpu_id); + struct acpi_buffer params; + union acpi_object *info; + struct atpx_mux input; + + if (atpx->functions.switch_start) { + input.size = 4; + input.mux = mux_id; + params.length = input.size; + params.pointer = &input; + info = radeon_atpx_call(atpx->handle, + ATPX_FUNCTION_GRAPHICS_DEVICE_SWITCH_START_NOTIFICATION, + ¶ms); + if (!info) + return -EIO; + kfree(info); + } + return 0; } -static int radeon_atpx_switch_end(acpi_handle handle, int gpu_id) +/** + * radeon_atpx_switch_end - notify the sbios of a GPU switch + * + * @atpx: atpx info struct + * @mux_id: mux state (0 = integrated GPU, 1 = discrete GPU) + * + * Execute the ATPX_FUNCTION_GRAPHICS_DEVICE_SWITCH_END_NOTIFICATION ATPX + * function to notify the sbios that a switch between the discrete GPU and + * integrated GPU has ended (all asics). + * Returns 0 on success, error on failure. + */ +static int radeon_atpx_switch_end(struct radeon_atpx *atpx, u16 mux_id) { - return radeon_atpx_execute(handle, ATPX_SWITCH_END, gpu_id); + struct acpi_buffer params; + union acpi_object *info; + struct atpx_mux input; + + if (atpx->functions.switch_end) { + input.size = 4; + input.mux = mux_id; + params.length = input.size; + params.pointer = &input; + info = radeon_atpx_call(atpx->handle, + ATPX_FUNCTION_GRAPHICS_DEVICE_SWITCH_END_NOTIFICATION, + ¶ms); + if (!info) + return -EIO; + kfree(info); + } + return 0; } +/** + * radeon_atpx_switchto - switch to the requested GPU + * + * @id: GPU to switch to + * + * Execute the necessary ATPX functions to switch between the discrete GPU and + * integrated GPU (all asics). + * Returns 0 on success, error on failure. + */ static int radeon_atpx_switchto(enum vga_switcheroo_client_id id) { - int gpu_id; + u16 gpu_id; if (id == VGA_SWITCHEROO_IGD) - gpu_id = ATPX_INTEGRATED; + gpu_id = ATPX_INTEGRATED_GPU; else - gpu_id = ATPX_DISCRETE; + gpu_id = ATPX_DISCRETE_GPU; - radeon_atpx_switch_start(radeon_atpx_priv.atpx_handle, gpu_id); - radeon_atpx_switch_mux(radeon_atpx_priv.atpx_handle, gpu_id); - radeon_atpx_switch_i2c_mux(radeon_atpx_priv.atpx_handle, gpu_id); - radeon_atpx_switch_end(radeon_atpx_priv.atpx_handle, gpu_id); + radeon_atpx_switch_start(&radeon_atpx_priv.atpx, gpu_id); + radeon_atpx_switch_disp_mux(&radeon_atpx_priv.atpx, gpu_id); + radeon_atpx_switch_i2c_mux(&radeon_atpx_priv.atpx, gpu_id); + radeon_atpx_switch_end(&radeon_atpx_priv.atpx, gpu_id); return 0; } +/** + * radeon_atpx_switchto - switch to the requested GPU + * + * @id: GPU to switch to + * @state: requested power state (0 = off, 1 = on) + * + * Execute the necessary ATPX function to power down/up the discrete GPU + * (all asics). + * Returns 0 on success, error on failure. + */ static int radeon_atpx_power_state(enum vga_switcheroo_client_id id, enum vga_switcheroo_state state) { @@ -143,10 +368,18 @@ static int radeon_atpx_power_state(enum vga_switcheroo_client_id id, if (id == VGA_SWITCHEROO_IGD) return 0; - radeon_atpx_set_discrete_state(radeon_atpx_priv.atpx_handle, state); + radeon_atpx_set_discrete_state(&radeon_atpx_priv.atpx, state); return 0; } +/** + * radeon_atpx_pci_probe_handle - look up the ATRM and ATPX handles + * + * @pdev: pci device + * + * Look up the ATPX and ATRM handles (all asics). + * Returns true if the handles are found, false if not. + */ static bool radeon_atpx_pci_probe_handle(struct pci_dev *pdev) { acpi_handle dhandle, atpx_handle; @@ -161,18 +394,30 @@ static bool radeon_atpx_pci_probe_handle(struct pci_dev *pdev) return false; radeon_atpx_priv.dhandle = dhandle; - radeon_atpx_priv.atpx_handle = atpx_handle; + radeon_atpx_priv.atpx.handle = atpx_handle; return true; } +/** + * radeon_atpx_init - verify the ATPX interface + * + * Verify the ATPX interface (all asics). + * Returns 0 on success, error on failure. + */ static int radeon_atpx_init(void) { /* set up the ATPX handle */ - - radeon_atpx_get_version(radeon_atpx_priv.atpx_handle); - return 0; + return radeon_atpx_verify_interface(&radeon_atpx_priv.atpx); } +/** + * radeon_atpx_get_client_id - get the client id + * + * @pdev: pci device + * + * look up whether we are the integrated or discrete GPU (all asics). + * Returns the client id. + */ static int radeon_atpx_get_client_id(struct pci_dev *pdev) { if (radeon_atpx_priv.dhandle == DEVICE_ACPI_HANDLE(&pdev->dev)) @@ -188,6 +433,12 @@ static struct vga_switcheroo_handler radeon_atpx_handler = { .get_client_id = radeon_atpx_get_client_id, }; +/** + * radeon_atpx_detect - detect whether we have PX + * + * Check if we have a PX system (all asics). + * Returns true if we have a PX system, false if not. + */ static bool radeon_atpx_detect(void) { char acpi_method_name[255] = { 0 }; @@ -203,7 +454,7 @@ static bool radeon_atpx_detect(void) } if (has_atpx && vga_count == 2) { - acpi_get_name(radeon_atpx_priv.atpx_handle, ACPI_FULL_PATHNAME, &buffer); + acpi_get_name(radeon_atpx_priv.atpx.handle, ACPI_FULL_PATHNAME, &buffer); printk(KERN_INFO "VGA switcheroo: detected switching method %s handle\n", acpi_method_name); radeon_atpx_priv.atpx_detected = true; @@ -212,6 +463,11 @@ static bool radeon_atpx_detect(void) return false; } +/** + * radeon_register_atpx_handler - register with vga_switcheroo + * + * Register the PX callbacks with vga_switcheroo (all asics). + */ void radeon_register_atpx_handler(void) { bool r; @@ -224,6 +480,11 @@ void radeon_register_atpx_handler(void) vga_switcheroo_register_handler(&radeon_atpx_handler); } +/** + * radeon_unregister_atpx_handler - unregister with vga_switcheroo + * + * Unregister the PX callbacks with vga_switcheroo (all asics). + */ void radeon_unregister_atpx_handler(void) { vga_switcheroo_unregister_handler(); diff --git a/drivers/gpu/drm/radeon/radeon_combios.c b/drivers/gpu/drm/radeon/radeon_combios.c index f75247d42ffd..8a73f0758903 100644 --- a/drivers/gpu/drm/radeon/radeon_combios.c +++ b/drivers/gpu/drm/radeon/radeon_combios.c @@ -3319,15 +3319,6 @@ static void combios_write_ram_size(struct drm_device *dev) WREG32(RADEON_CONFIG_MEMSIZE, mem_size); } -void radeon_combios_dyn_clk_setup(struct drm_device *dev, int enable) -{ - uint16_t dyn_clk_info = - combios_get_table_offset(dev, COMBIOS_DYN_CLK_1_TABLE); - - if (dyn_clk_info) - combios_parse_pll_table(dev, dyn_clk_info); -} - void radeon_combios_asic_init(struct drm_device *dev) { struct radeon_device *rdev = dev->dev_private; diff --git a/drivers/gpu/drm/radeon/radeon_connectors.c b/drivers/gpu/drm/radeon/radeon_connectors.c index 895e628b60f8..69a142fc3d1d 100644 --- a/drivers/gpu/drm/radeon/radeon_connectors.c +++ b/drivers/gpu/drm/radeon/radeon_connectors.c @@ -40,10 +40,6 @@ radeon_atombios_connected_scratch_regs(struct drm_connector *connector, struct drm_encoder *encoder, bool connected); -extern void -radeon_legacy_backlight_init(struct radeon_encoder *radeon_encoder, - struct drm_connector *drm_connector); - void radeon_connector_hotplug(struct drm_connector *connector) { struct drm_device *dev = connector->dev; @@ -198,7 +194,7 @@ radeon_connector_update_scratch_regs(struct drm_connector *connector, enum drm_c } } -struct drm_encoder *radeon_find_encoder(struct drm_connector *connector, int encoder_type) +static struct drm_encoder *radeon_find_encoder(struct drm_connector *connector, int encoder_type) { struct drm_mode_object *obj; struct drm_encoder *encoder; @@ -219,7 +215,7 @@ struct drm_encoder *radeon_find_encoder(struct drm_connector *connector, int enc return NULL; } -struct drm_encoder *radeon_best_single_encoder(struct drm_connector *connector) +static struct drm_encoder *radeon_best_single_encoder(struct drm_connector *connector) { int enc_id = connector->encoder_ids[0]; struct drm_mode_object *obj; @@ -370,7 +366,7 @@ static void radeon_add_common_modes(struct drm_encoder *encoder, struct drm_conn } } -int radeon_connector_set_property(struct drm_connector *connector, struct drm_property *property, +static int radeon_connector_set_property(struct drm_connector *connector, struct drm_property *property, uint64_t val) { struct drm_device *dev = connector->dev; @@ -691,13 +687,13 @@ static int radeon_lvds_set_property(struct drm_connector *connector, } -struct drm_connector_helper_funcs radeon_lvds_connector_helper_funcs = { +static const struct drm_connector_helper_funcs radeon_lvds_connector_helper_funcs = { .get_modes = radeon_lvds_get_modes, .mode_valid = radeon_lvds_mode_valid, .best_encoder = radeon_best_single_encoder, }; -struct drm_connector_funcs radeon_lvds_connector_funcs = { +static const struct drm_connector_funcs radeon_lvds_connector_funcs = { .dpms = drm_helper_connector_dpms, .detect = radeon_lvds_detect, .fill_modes = drm_helper_probe_single_connector_modes, @@ -809,13 +805,13 @@ radeon_vga_detect(struct drm_connector *connector, bool force) return ret; } -struct drm_connector_helper_funcs radeon_vga_connector_helper_funcs = { +static const struct drm_connector_helper_funcs radeon_vga_connector_helper_funcs = { .get_modes = radeon_vga_get_modes, .mode_valid = radeon_vga_mode_valid, .best_encoder = radeon_best_single_encoder, }; -struct drm_connector_funcs radeon_vga_connector_funcs = { +static const struct drm_connector_funcs radeon_vga_connector_funcs = { .dpms = drm_helper_connector_dpms, .detect = radeon_vga_detect, .fill_modes = drm_helper_probe_single_connector_modes, @@ -879,13 +875,13 @@ radeon_tv_detect(struct drm_connector *connector, bool force) return ret; } -struct drm_connector_helper_funcs radeon_tv_connector_helper_funcs = { +static const struct drm_connector_helper_funcs radeon_tv_connector_helper_funcs = { .get_modes = radeon_tv_get_modes, .mode_valid = radeon_tv_mode_valid, .best_encoder = radeon_best_single_encoder, }; -struct drm_connector_funcs radeon_tv_connector_funcs = { +static const struct drm_connector_funcs radeon_tv_connector_funcs = { .dpms = drm_helper_connector_dpms, .detect = radeon_tv_detect, .fill_modes = drm_helper_probe_single_connector_modes, @@ -1089,7 +1085,7 @@ out: } /* okay need to be smart in here about which encoder to pick */ -struct drm_encoder *radeon_dvi_encoder(struct drm_connector *connector) +static struct drm_encoder *radeon_dvi_encoder(struct drm_connector *connector) { int enc_id = connector->encoder_ids[0]; struct radeon_connector *radeon_connector = to_radeon_connector(connector); @@ -1179,13 +1175,13 @@ static int radeon_dvi_mode_valid(struct drm_connector *connector, return MODE_OK; } -struct drm_connector_helper_funcs radeon_dvi_connector_helper_funcs = { +static const struct drm_connector_helper_funcs radeon_dvi_connector_helper_funcs = { .get_modes = radeon_dvi_get_modes, .mode_valid = radeon_dvi_mode_valid, .best_encoder = radeon_dvi_encoder, }; -struct drm_connector_funcs radeon_dvi_connector_funcs = { +static const struct drm_connector_funcs radeon_dvi_connector_funcs = { .dpms = drm_helper_connector_dpms, .detect = radeon_dvi_detect, .fill_modes = drm_helper_probe_single_connector_modes, @@ -1462,13 +1458,13 @@ static int radeon_dp_mode_valid(struct drm_connector *connector, } } -struct drm_connector_helper_funcs radeon_dp_connector_helper_funcs = { +static const struct drm_connector_helper_funcs radeon_dp_connector_helper_funcs = { .get_modes = radeon_dp_get_modes, .mode_valid = radeon_dp_mode_valid, .best_encoder = radeon_dvi_encoder, }; -struct drm_connector_funcs radeon_dp_connector_funcs = { +static const struct drm_connector_funcs radeon_dp_connector_funcs = { .dpms = drm_helper_connector_dpms, .detect = radeon_dp_detect, .fill_modes = drm_helper_probe_single_connector_modes, @@ -2008,15 +2004,4 @@ radeon_add_legacy_connector(struct drm_device *dev, connector->polled = DRM_CONNECTOR_POLL_HPD; connector->display_info.subpixel_order = subpixel_order; drm_sysfs_connector_add(connector); - if (connector_type == DRM_MODE_CONNECTOR_LVDS) { - struct drm_encoder *drm_encoder; - - list_for_each_entry(drm_encoder, &dev->mode_config.encoder_list, head) { - struct radeon_encoder *radeon_encoder; - - radeon_encoder = to_radeon_encoder(drm_encoder); - if (radeon_encoder->encoder_id == ENCODER_OBJECT_ID_INTERNAL_LVDS) - radeon_legacy_backlight_init(radeon_encoder, connector); - } - } } diff --git a/drivers/gpu/drm/radeon/radeon_cs.c b/drivers/gpu/drm/radeon/radeon_cs.c index b4a0db24f4dd..d59eb59cdb81 100644 --- a/drivers/gpu/drm/radeon/radeon_cs.c +++ b/drivers/gpu/drm/radeon/radeon_cs.c @@ -32,7 +32,7 @@ void r100_cs_dump_packet(struct radeon_cs_parser *p, struct radeon_cs_packet *pkt); -int radeon_cs_parser_relocs(struct radeon_cs_parser *p) +static int radeon_cs_parser_relocs(struct radeon_cs_parser *p) { struct drm_device *ddev = p->rdev->ddev; struct radeon_cs_chunk *chunk; @@ -115,19 +115,27 @@ static int radeon_cs_get_ring(struct radeon_cs_parser *p, u32 ring, s32 priority return 0; } +static void radeon_cs_sync_to(struct radeon_cs_parser *p, + struct radeon_fence *fence) +{ + struct radeon_fence *other; + + if (!fence) + return; + + other = p->ib.sync_to[fence->ring]; + p->ib.sync_to[fence->ring] = radeon_fence_later(fence, other); +} + static void radeon_cs_sync_rings(struct radeon_cs_parser *p) { int i; for (i = 0; i < p->nrelocs; i++) { - struct radeon_fence *a, *b; - - if (!p->relocs[i].robj || !p->relocs[i].robj->tbo.sync_obj) + if (!p->relocs[i].robj) continue; - a = p->relocs[i].robj->tbo.sync_obj; - b = p->ib.sync_to[a->ring]; - p->ib.sync_to[a->ring] = radeon_fence_later(a, b); + radeon_cs_sync_to(p, p->relocs[i].robj->tbo.sync_obj); } } @@ -278,30 +286,6 @@ int radeon_cs_parser_init(struct radeon_cs_parser *p, void *data) return 0; } -static void radeon_bo_vm_fence_va(struct radeon_cs_parser *parser, - struct radeon_fence *fence) -{ - struct radeon_fpriv *fpriv = parser->filp->driver_priv; - struct radeon_vm *vm = &fpriv->vm; - struct radeon_bo_list *lobj; - - if (parser->chunk_ib_idx == -1) { - return; - } - if ((parser->cs_flags & RADEON_CS_USE_VM) == 0) { - return; - } - - list_for_each_entry(lobj, &parser->validated, tv.head) { - struct radeon_bo_va *bo_va; - struct radeon_bo *rbo = lobj->bo; - - bo_va = radeon_bo_va(rbo, vm); - radeon_fence_unref(&bo_va->fence); - bo_va->fence = radeon_fence_ref(fence); - } -} - /** * cs_parser_fini() - clean parser states * @parser: parser structure holding parsing context. @@ -315,8 +299,6 @@ static void radeon_cs_parser_fini(struct radeon_cs_parser *parser, int error) unsigned i; if (!error) { - /* fence all bo va before ttm_eu_fence_buffer_objects so bo are still reserved */ - radeon_bo_vm_fence_va(parser, parser->ib.fence); ttm_eu_fence_buffer_objects(&parser->validated, parser->ib.fence); } else { @@ -363,7 +345,7 @@ static int radeon_cs_ib_chunk(struct radeon_device *rdev, * uncached). */ r = radeon_ib_get(rdev, parser->ring, &parser->ib, - ib_chunk->length_dw * 4); + NULL, ib_chunk->length_dw * 4); if (r) { DRM_ERROR("Failed to get ib !\n"); return r; @@ -380,7 +362,6 @@ static int radeon_cs_ib_chunk(struct radeon_device *rdev, return r; } radeon_cs_sync_rings(parser); - parser->ib.vm_id = 0; r = radeon_ib_schedule(rdev, &parser->ib, NULL); if (r) { DRM_ERROR("Failed to schedule IB !\n"); @@ -391,10 +372,15 @@ static int radeon_cs_ib_chunk(struct radeon_device *rdev, static int radeon_bo_vm_update_pte(struct radeon_cs_parser *parser, struct radeon_vm *vm) { + struct radeon_device *rdev = parser->rdev; struct radeon_bo_list *lobj; struct radeon_bo *bo; int r; + r = radeon_vm_bo_update_pte(rdev, vm, rdev->ring_tmp_bo.bo, &rdev->ring_tmp_bo.bo->tbo.mem); + if (r) { + return r; + } list_for_each_entry(lobj, &parser->validated, tv.head) { bo = lobj->bo; r = radeon_vm_bo_update_pte(parser->rdev, vm, bo, &bo->tbo.mem); @@ -426,7 +412,7 @@ static int radeon_cs_ib_vm_chunk(struct radeon_device *rdev, return -EINVAL; } r = radeon_ib_get(rdev, parser->ring, &parser->const_ib, - ib_chunk->length_dw * 4); + vm, ib_chunk->length_dw * 4); if (r) { DRM_ERROR("Failed to get const ib !\n"); return r; @@ -450,7 +436,7 @@ static int radeon_cs_ib_vm_chunk(struct radeon_device *rdev, return -EINVAL; } r = radeon_ib_get(rdev, parser->ring, &parser->ib, - ib_chunk->length_dw * 4); + vm, ib_chunk->length_dw * 4); if (r) { DRM_ERROR("Failed to get ib !\n"); return r; @@ -468,7 +454,7 @@ static int radeon_cs_ib_vm_chunk(struct radeon_device *rdev, mutex_lock(&rdev->vm_manager.lock); mutex_lock(&vm->mutex); - r = radeon_vm_bind(rdev, vm); + r = radeon_vm_alloc_pt(rdev, vm); if (r) { goto out; } @@ -477,32 +463,21 @@ static int radeon_cs_ib_vm_chunk(struct radeon_device *rdev, goto out; } radeon_cs_sync_rings(parser); - - parser->ib.vm_id = vm->id; - /* ib pool is bind at 0 in virtual address space, - * so gpu_addr is the offset inside the pool bo - */ - parser->ib.gpu_addr = parser->ib.sa_bo->soffset; + radeon_cs_sync_to(parser, vm->fence); + radeon_cs_sync_to(parser, radeon_vm_grab_id(rdev, vm, parser->ring)); if ((rdev->family >= CHIP_TAHITI) && (parser->chunk_const_ib_idx != -1)) { - parser->const_ib.vm_id = vm->id; - /* ib pool is bind at 0 in virtual address space, - * so gpu_addr is the offset inside the pool bo - */ - parser->const_ib.gpu_addr = parser->const_ib.sa_bo->soffset; r = radeon_ib_schedule(rdev, &parser->ib, &parser->const_ib); } else { r = radeon_ib_schedule(rdev, &parser->ib, NULL); } -out: if (!r) { - if (vm->fence) { - radeon_fence_unref(&vm->fence); - } - vm->fence = radeon_fence_ref(parser->ib.fence); + radeon_vm_fence(rdev, vm, parser->ib.fence); } + +out: mutex_unlock(&vm->mutex); mutex_unlock(&rdev->vm_manager.lock); return r; diff --git a/drivers/gpu/drm/radeon/radeon_device.c b/drivers/gpu/drm/radeon/radeon_device.c index 7a3daebd732d..64a42647f08a 100644 --- a/drivers/gpu/drm/radeon/radeon_device.c +++ b/drivers/gpu/drm/radeon/radeon_device.c @@ -842,7 +842,7 @@ static unsigned int radeon_vga_set_decode(void *cookie, bool state) * Validates certain module parameters and updates * the associated values used by the driver (all asics). */ -void radeon_check_arguments(struct radeon_device *rdev) +static void radeon_check_arguments(struct radeon_device *rdev) { /* vramlimit must be a power of two */ switch (radeon_vram_limit) { @@ -1013,13 +1013,11 @@ int radeon_device_init(struct radeon_device *rdev, init_rwsem(&rdev->pm.mclk_lock); init_rwsem(&rdev->exclusive_lock); init_waitqueue_head(&rdev->irq.vblank_queue); - init_waitqueue_head(&rdev->irq.idle_queue); r = radeon_gem_init(rdev); if (r) return r; /* initialize vm here */ mutex_init(&rdev->vm_manager.lock); - rdev->vm_manager.use_bitmap = 1; rdev->vm_manager.max_pfn = 1 << 20; INIT_LIST_HEAD(&rdev->vm_manager.lru_vm); @@ -1284,6 +1282,13 @@ int radeon_resume_kms(struct drm_device *dev) if (rdev->is_atom_bios) { radeon_atom_encoder_init(rdev); radeon_atom_disp_eng_pll_init(rdev); + /* turn on the BL */ + if (rdev->mode_info.bl_encoder) { + u8 bl_level = radeon_get_backlight_level(rdev, + rdev->mode_info.bl_encoder); + radeon_set_backlight_level(rdev, rdev->mode_info.bl_encoder, + bl_level); + } } /* reset hpd state */ radeon_hpd_init(rdev); diff --git a/drivers/gpu/drm/radeon/radeon_drv.c b/drivers/gpu/drm/radeon/radeon_drv.c index 8c593ea82c41..27ece75b4789 100644 --- a/drivers/gpu/drm/radeon/radeon_drv.c +++ b/drivers/gpu/drm/radeon/radeon_drv.c @@ -64,9 +64,11 @@ * 2.20.0 - r600-si: RADEON_INFO_TIMESTAMP query * 2.21.0 - r600-r700: FMASK and CMASK * 2.22.0 - r600 only: RESOLVE_BOX allowed + * 2.23.0 - allow STRMOUT_BASE_UPDATE on RS780 and RS880 + * 2.24.0 - eg only: allow MIP_ADDRESS=0 for MSAA textures */ #define KMS_DRIVER_MAJOR 2 -#define KMS_DRIVER_MINOR 22 +#define KMS_DRIVER_MINOR 24 #define KMS_DRIVER_PATCHLEVEL 0 int radeon_driver_load_kms(struct drm_device *dev, unsigned long flags); int radeon_driver_unload_kms(struct drm_device *dev); diff --git a/drivers/gpu/drm/radeon/radeon_encoders.c b/drivers/gpu/drm/radeon/radeon_encoders.c index 74670696277d..e66c807df9e6 100644 --- a/drivers/gpu/drm/radeon/radeon_encoders.c +++ b/drivers/gpu/drm/radeon/radeon_encoders.c @@ -29,6 +29,14 @@ #include "radeon.h" #include "atom.h" +extern void +radeon_legacy_backlight_init(struct radeon_encoder *radeon_encoder, + struct drm_connector *drm_connector); +extern void +radeon_atom_backlight_init(struct radeon_encoder *radeon_encoder, + struct drm_connector *drm_connector); + + static uint32_t radeon_encoder_clones(struct drm_encoder *encoder) { struct drm_device *dev = encoder->dev; @@ -153,6 +161,7 @@ radeon_get_encoder_enum(struct drm_device *dev, uint32_t supported_device, uint8 void radeon_link_encoder_connector(struct drm_device *dev) { + struct radeon_device *rdev = dev->dev_private; struct drm_connector *connector; struct radeon_connector *radeon_connector; struct drm_encoder *encoder; @@ -163,8 +172,16 @@ radeon_link_encoder_connector(struct drm_device *dev) radeon_connector = to_radeon_connector(connector); list_for_each_entry(encoder, &dev->mode_config.encoder_list, head) { radeon_encoder = to_radeon_encoder(encoder); - if (radeon_encoder->devices & radeon_connector->devices) + if (radeon_encoder->devices & radeon_connector->devices) { drm_mode_connector_attach_encoder(connector, encoder); + if (radeon_encoder->devices & (ATOM_DEVICE_LCD_SUPPORT)) { + if (rdev->is_atom_bios) + radeon_atom_backlight_init(radeon_encoder, connector); + else + radeon_legacy_backlight_init(radeon_encoder, connector); + rdev->mode_info.bl_encoder = radeon_encoder; + } + } } } } diff --git a/drivers/gpu/drm/radeon/radeon_fb.c b/drivers/gpu/drm/radeon/radeon_fb.c index 5906914a78bc..6ada72c6d7a1 100644 --- a/drivers/gpu/drm/radeon/radeon_fb.c +++ b/drivers/gpu/drm/radeon/radeon_fb.c @@ -316,22 +316,6 @@ static int radeon_fb_find_or_create_single(struct drm_fb_helper *helper, return new_fb; } -static char *mode_option; -int radeon_parse_options(char *options) -{ - char *this_opt; - - if (!options || !*options) - return 0; - - while ((this_opt = strsep(&options, ",")) != NULL) { - if (!*this_opt) - continue; - mode_option = this_opt; - } - return 0; -} - void radeon_fb_output_poll_changed(struct radeon_device *rdev) { drm_fb_helper_hotplug_event(&rdev->mode_info.rfbdev->helper); diff --git a/drivers/gpu/drm/radeon/radeon_fence.c b/drivers/gpu/drm/radeon/radeon_fence.c index 2a59375dbe52..1eb3db52a36e 100644 --- a/drivers/gpu/drm/radeon/radeon_fence.c +++ b/drivers/gpu/drm/radeon/radeon_fence.c @@ -399,7 +399,7 @@ int radeon_fence_wait(struct radeon_fence *fence, bool intr) return 0; } -bool radeon_fence_any_seq_signaled(struct radeon_device *rdev, u64 *seq) +static bool radeon_fence_any_seq_signaled(struct radeon_device *rdev, u64 *seq) { unsigned i; diff --git a/drivers/gpu/drm/radeon/radeon_gart.c b/drivers/gpu/drm/radeon/radeon_gart.c index bb3b7fe05ccd..753b7ca3c807 100644 --- a/drivers/gpu/drm/radeon/radeon_gart.c +++ b/drivers/gpu/drm/radeon/radeon_gart.c @@ -423,6 +423,18 @@ void radeon_gart_fini(struct radeon_device *rdev) */ /** + * radeon_vm_directory_size - returns the size of the page directory in bytes + * + * @rdev: radeon_device pointer + * + * Calculate the size of the page directory in bytes (cayman+). + */ +static unsigned radeon_vm_directory_size(struct radeon_device *rdev) +{ + return (rdev->vm_manager.max_pfn >> RADEON_VM_BLOCK_SIZE) * 8; +} + +/** * radeon_vm_manager_init - init the vm manager * * @rdev: radeon_device pointer @@ -435,12 +447,15 @@ int radeon_vm_manager_init(struct radeon_device *rdev) struct radeon_vm *vm; struct radeon_bo_va *bo_va; int r; + unsigned size; if (!rdev->vm_manager.enabled) { - /* mark first vm as always in use, it's the system one */ /* allocate enough for 2 full VM pts */ + size = RADEON_GPU_PAGE_ALIGN(radeon_vm_directory_size(rdev)); + size += RADEON_GPU_PAGE_ALIGN(rdev->vm_manager.max_pfn * 8); + size *= 2; r = radeon_sa_bo_manager_init(rdev, &rdev->vm_manager.sa_manager, - rdev->vm_manager.max_pfn * 8 * 2, + size, RADEON_GEM_DOMAIN_VRAM); if (r) { dev_err(rdev->dev, "failed to allocate vm bo (%dKB)\n", @@ -448,10 +463,10 @@ int radeon_vm_manager_init(struct radeon_device *rdev) return r; } - r = rdev->vm_manager.funcs->init(rdev); + r = radeon_asic_vm_init(rdev); if (r) return r; - + rdev->vm_manager.enabled = true; r = radeon_sa_bo_manager_start(rdev, &rdev->vm_manager.sa_manager); @@ -461,73 +476,36 @@ int radeon_vm_manager_init(struct radeon_device *rdev) /* restore page table */ list_for_each_entry(vm, &rdev->vm_manager.lru_vm, list) { - if (vm->id == -1) + if (vm->sa_bo == NULL) continue; list_for_each_entry(bo_va, &vm->va, vm_list) { - struct ttm_mem_reg *mem = NULL; - if (bo_va->valid) - mem = &bo_va->bo->tbo.mem; - bo_va->valid = false; - r = radeon_vm_bo_update_pte(rdev, vm, bo_va->bo, mem); - if (r) { - DRM_ERROR("Failed to update pte for vm %d!\n", vm->id); - } - } - - r = rdev->vm_manager.funcs->bind(rdev, vm, vm->id); - if (r) { - DRM_ERROR("Failed to bind vm %d!\n", vm->id); } } return 0; } -/* global mutex must be lock */ /** - * radeon_vm_unbind_locked - unbind a specific vm + * radeon_vm_free_pt - free the page table for a specific vm * * @rdev: radeon_device pointer * @vm: vm to unbind * - * Unbind the requested vm (cayman+). - * Wait for use of the VM to finish, then unbind the page table, - * and free the page table memory. + * Free the page table of a specific vm (cayman+). + * + * Global and local mutex must be lock! */ -static void radeon_vm_unbind_locked(struct radeon_device *rdev, +static void radeon_vm_free_pt(struct radeon_device *rdev, struct radeon_vm *vm) { struct radeon_bo_va *bo_va; - if (vm->id == -1) { + if (!vm->sa_bo) return; - } - /* wait for vm use to end */ - while (vm->fence) { - int r; - r = radeon_fence_wait(vm->fence, false); - if (r) - DRM_ERROR("error while waiting for fence: %d\n", r); - if (r == -EDEADLK) { - mutex_unlock(&rdev->vm_manager.lock); - r = radeon_gpu_reset(rdev); - mutex_lock(&rdev->vm_manager.lock); - if (!r) - continue; - } - break; - } - radeon_fence_unref(&vm->fence); - - /* hw unbind */ - rdev->vm_manager.funcs->unbind(rdev, vm); - rdev->vm_manager.use_bitmap &= ~(1 << vm->id); list_del_init(&vm->list); - vm->id = -1; - radeon_sa_bo_free(rdev, &vm->sa_bo, NULL); - vm->pt = NULL; + radeon_sa_bo_free(rdev, &vm->sa_bo, vm->fence); list_for_each_entry(bo_va, &vm->va, vm_list) { bo_va->valid = false; @@ -544,16 +522,22 @@ static void radeon_vm_unbind_locked(struct radeon_device *rdev, void radeon_vm_manager_fini(struct radeon_device *rdev) { struct radeon_vm *vm, *tmp; + int i; if (!rdev->vm_manager.enabled) return; mutex_lock(&rdev->vm_manager.lock); - /* unbind all active vm */ + /* free all allocated page tables */ list_for_each_entry_safe(vm, tmp, &rdev->vm_manager.lru_vm, list) { - radeon_vm_unbind_locked(rdev, vm); + mutex_lock(&vm->mutex); + radeon_vm_free_pt(rdev, vm); + mutex_unlock(&vm->mutex); } - rdev->vm_manager.funcs->fini(rdev); + for (i = 0; i < RADEON_NUM_VM; ++i) { + radeon_fence_unref(&rdev->vm_manager.active[i]); + } + radeon_asic_vm_fini(rdev); mutex_unlock(&rdev->vm_manager.lock); radeon_sa_bo_manager_suspend(rdev, &rdev->vm_manager.sa_manager); @@ -561,46 +545,34 @@ void radeon_vm_manager_fini(struct radeon_device *rdev) rdev->vm_manager.enabled = false; } -/* global mutex must be locked */ /** - * radeon_vm_unbind - locked version of unbind - * - * @rdev: radeon_device pointer - * @vm: vm to unbind - * - * Locked version that wraps radeon_vm_unbind_locked (cayman+). - */ -void radeon_vm_unbind(struct radeon_device *rdev, struct radeon_vm *vm) -{ - mutex_lock(&vm->mutex); - radeon_vm_unbind_locked(rdev, vm); - mutex_unlock(&vm->mutex); -} - -/* global and local mutex must be locked */ -/** - * radeon_vm_bind - bind a page table to a VMID + * radeon_vm_alloc_pt - allocates a page table for a VM * * @rdev: radeon_device pointer * @vm: vm to bind * - * Bind the requested vm (cayman+). - * Suballocate memory for the page table, allocate a VMID - * and bind the page table to it, and finally start to populate - * the page table. + * Allocate a page table for the requested vm (cayman+). + * Also starts to populate the page table. * Returns 0 for success, error for failure. + * + * Global and local mutex must be locked! */ -int radeon_vm_bind(struct radeon_device *rdev, struct radeon_vm *vm) +int radeon_vm_alloc_pt(struct radeon_device *rdev, struct radeon_vm *vm) { struct radeon_vm *vm_evict; - unsigned i; - int id = -1, r; + int r; + u64 *pd_addr; + int tables_size; if (vm == NULL) { return -EINVAL; } - if (vm->id != -1) { + /* allocate enough to cover the current VM size */ + tables_size = RADEON_GPU_PAGE_ALIGN(radeon_vm_directory_size(rdev)); + tables_size += RADEON_GPU_PAGE_ALIGN(vm->last_pfn * 8); + + if (vm->sa_bo != NULL) { /* update lru */ list_del_init(&vm->list); list_add_tail(&vm->list, &rdev->vm_manager.lru_vm); @@ -609,98 +581,215 @@ int radeon_vm_bind(struct radeon_device *rdev, struct radeon_vm *vm) retry: r = radeon_sa_bo_new(rdev, &rdev->vm_manager.sa_manager, &vm->sa_bo, - RADEON_GPU_PAGE_ALIGN(vm->last_pfn * 8), - RADEON_GPU_PAGE_SIZE, false); - if (r) { + tables_size, RADEON_GPU_PAGE_SIZE, false); + if (r == -ENOMEM) { if (list_empty(&rdev->vm_manager.lru_vm)) { return r; } vm_evict = list_first_entry(&rdev->vm_manager.lru_vm, struct radeon_vm, list); - radeon_vm_unbind(rdev, vm_evict); + mutex_lock(&vm_evict->mutex); + radeon_vm_free_pt(rdev, vm_evict); + mutex_unlock(&vm_evict->mutex); goto retry; + + } else if (r) { + return r; } - vm->pt = radeon_sa_bo_cpu_addr(vm->sa_bo); - vm->pt_gpu_addr = radeon_sa_bo_gpu_addr(vm->sa_bo); - memset(vm->pt, 0, RADEON_GPU_PAGE_ALIGN(vm->last_pfn * 8)); -retry_id: - /* search for free vm */ - for (i = 0; i < rdev->vm_manager.nvm; i++) { - if (!(rdev->vm_manager.use_bitmap & (1 << i))) { - id = i; - break; + pd_addr = radeon_sa_bo_cpu_addr(vm->sa_bo); + vm->pd_gpu_addr = radeon_sa_bo_gpu_addr(vm->sa_bo); + memset(pd_addr, 0, tables_size); + + list_add_tail(&vm->list, &rdev->vm_manager.lru_vm); + return radeon_vm_bo_update_pte(rdev, vm, rdev->ring_tmp_bo.bo, + &rdev->ring_tmp_bo.bo->tbo.mem); +} + +/** + * radeon_vm_grab_id - allocate the next free VMID + * + * @rdev: radeon_device pointer + * @vm: vm to allocate id for + * @ring: ring we want to submit job to + * + * Allocate an id for the vm (cayman+). + * Returns the fence we need to sync to (if any). + * + * Global and local mutex must be locked! + */ +struct radeon_fence *radeon_vm_grab_id(struct radeon_device *rdev, + struct radeon_vm *vm, int ring) +{ + struct radeon_fence *best[RADEON_NUM_RINGS] = {}; + unsigned choices[2] = {}; + unsigned i; + + /* check if the id is still valid */ + if (vm->fence && vm->fence == rdev->vm_manager.active[vm->id]) + return NULL; + + /* we definately need to flush */ + radeon_fence_unref(&vm->last_flush); + + /* skip over VMID 0, since it is the system VM */ + for (i = 1; i < rdev->vm_manager.nvm; ++i) { + struct radeon_fence *fence = rdev->vm_manager.active[i]; + + if (fence == NULL) { + /* found a free one */ + vm->id = i; + return NULL; + } + + if (radeon_fence_is_earlier(fence, best[fence->ring])) { + best[fence->ring] = fence; + choices[fence->ring == ring ? 0 : 1] = i; } } - /* evict vm if necessary */ - if (id == -1) { - vm_evict = list_first_entry(&rdev->vm_manager.lru_vm, struct radeon_vm, list); - radeon_vm_unbind(rdev, vm_evict); - goto retry_id; + + for (i = 0; i < 2; ++i) { + if (choices[i]) { + vm->id = choices[i]; + return rdev->vm_manager.active[choices[i]]; + } } - /* do hw bind */ - r = rdev->vm_manager.funcs->bind(rdev, vm, id); - if (r) { - radeon_sa_bo_free(rdev, &vm->sa_bo, NULL); - return r; + /* should never happen */ + BUG(); + return NULL; +} + +/** + * radeon_vm_fence - remember fence for vm + * + * @rdev: radeon_device pointer + * @vm: vm we want to fence + * @fence: fence to remember + * + * Fence the vm (cayman+). + * Set the fence used to protect page table and id. + * + * Global and local mutex must be locked! + */ +void radeon_vm_fence(struct radeon_device *rdev, + struct radeon_vm *vm, + struct radeon_fence *fence) +{ + radeon_fence_unref(&rdev->vm_manager.active[vm->id]); + rdev->vm_manager.active[vm->id] = radeon_fence_ref(fence); + + radeon_fence_unref(&vm->fence); + vm->fence = radeon_fence_ref(fence); +} + +/** + * radeon_vm_bo_find - find the bo_va for a specific vm & bo + * + * @vm: requested vm + * @bo: requested buffer object + * + * Find @bo inside the requested vm (cayman+). + * Search inside the @bos vm list for the requested vm + * Returns the found bo_va or NULL if none is found + * + * Object has to be reserved! + */ +struct radeon_bo_va *radeon_vm_bo_find(struct radeon_vm *vm, + struct radeon_bo *bo) +{ + struct radeon_bo_va *bo_va; + + list_for_each_entry(bo_va, &bo->va, bo_list) { + if (bo_va->vm == vm) { + return bo_va; + } } - rdev->vm_manager.use_bitmap |= 1 << id; - vm->id = id; - list_add_tail(&vm->list, &rdev->vm_manager.lru_vm); - return radeon_vm_bo_update_pte(rdev, vm, rdev->ring_tmp_bo.bo, - &rdev->ring_tmp_bo.bo->tbo.mem); + return NULL; } -/* object have to be reserved */ /** * radeon_vm_bo_add - add a bo to a specific vm * * @rdev: radeon_device pointer * @vm: requested vm * @bo: radeon buffer object - * @offset: requested offset of the buffer in the VM address space - * @flags: attributes of pages (read/write/valid/etc.) * * Add @bo into the requested vm (cayman+). - * Add @bo to the list of bos associated with the vm and validate - * the offset requested within the vm address space. - * Returns 0 for success, error for failure. + * Add @bo to the list of bos associated with the vm + * Returns newly added bo_va or NULL for failure + * + * Object has to be reserved! */ -int radeon_vm_bo_add(struct radeon_device *rdev, - struct radeon_vm *vm, - struct radeon_bo *bo, - uint64_t offset, - uint32_t flags) +struct radeon_bo_va *radeon_vm_bo_add(struct radeon_device *rdev, + struct radeon_vm *vm, + struct radeon_bo *bo) { - struct radeon_bo_va *bo_va, *tmp; - struct list_head *head; - uint64_t size = radeon_bo_size(bo), last_offset = 0; - unsigned last_pfn; + struct radeon_bo_va *bo_va; bo_va = kzalloc(sizeof(struct radeon_bo_va), GFP_KERNEL); if (bo_va == NULL) { - return -ENOMEM; + return NULL; } bo_va->vm = vm; bo_va->bo = bo; - bo_va->soffset = offset; - bo_va->eoffset = offset + size; - bo_va->flags = flags; + bo_va->soffset = 0; + bo_va->eoffset = 0; + bo_va->flags = 0; bo_va->valid = false; + bo_va->ref_count = 1; INIT_LIST_HEAD(&bo_va->bo_list); INIT_LIST_HEAD(&bo_va->vm_list); - /* make sure object fit at this offset */ - if (bo_va->soffset >= bo_va->eoffset) { - kfree(bo_va); - return -EINVAL; - } - last_pfn = bo_va->eoffset / RADEON_GPU_PAGE_SIZE; - if (last_pfn > rdev->vm_manager.max_pfn) { - kfree(bo_va); - dev_err(rdev->dev, "va above limit (0x%08X > 0x%08X)\n", - last_pfn, rdev->vm_manager.max_pfn); - return -EINVAL; + mutex_lock(&vm->mutex); + list_add(&bo_va->vm_list, &vm->va); + list_add_tail(&bo_va->bo_list, &bo->va); + mutex_unlock(&vm->mutex); + + return bo_va; +} + +/** + * radeon_vm_bo_set_addr - set bos virtual address inside a vm + * + * @rdev: radeon_device pointer + * @bo_va: bo_va to store the address + * @soffset: requested offset of the buffer in the VM address space + * @flags: attributes of pages (read/write/valid/etc.) + * + * Set offset of @bo_va (cayman+). + * Validate and set the offset requested within the vm address space. + * Returns 0 for success, error for failure. + * + * Object has to be reserved! + */ +int radeon_vm_bo_set_addr(struct radeon_device *rdev, + struct radeon_bo_va *bo_va, + uint64_t soffset, + uint32_t flags) +{ + uint64_t size = radeon_bo_size(bo_va->bo); + uint64_t eoffset, last_offset = 0; + struct radeon_vm *vm = bo_va->vm; + struct radeon_bo_va *tmp; + struct list_head *head; + unsigned last_pfn; + + if (soffset) { + /* make sure object fit at this offset */ + eoffset = soffset + size; + if (soffset >= eoffset) { + return -EINVAL; + } + + last_pfn = eoffset / RADEON_GPU_PAGE_SIZE; + if (last_pfn > rdev->vm_manager.max_pfn) { + dev_err(rdev->dev, "va above limit (0x%08X > 0x%08X)\n", + last_pfn, rdev->vm_manager.max_pfn); + return -EINVAL; + } + + } else { + eoffset = last_pfn = 0; } mutex_lock(&vm->mutex); @@ -713,7 +802,7 @@ int radeon_vm_bo_add(struct radeon_device *rdev, if (last_pfn > vm->last_pfn) { /* grow va space 32M by 32M */ unsigned align = ((32 << 20) >> 12) - 1; - radeon_vm_unbind_locked(rdev, vm); + radeon_vm_free_pt(rdev, vm); vm->last_pfn = (last_pfn + align) & ~align; } mutex_unlock(&rdev->vm_manager.lock); @@ -721,68 +810,60 @@ int radeon_vm_bo_add(struct radeon_device *rdev, head = &vm->va; last_offset = 0; list_for_each_entry(tmp, &vm->va, vm_list) { - if (bo_va->soffset >= last_offset && bo_va->eoffset < tmp->soffset) { + if (bo_va == tmp) { + /* skip over currently modified bo */ + continue; + } + + if (soffset >= last_offset && eoffset <= tmp->soffset) { /* bo can be added before this one */ break; } - if (bo_va->soffset >= tmp->soffset && bo_va->soffset < tmp->eoffset) { + if (eoffset > tmp->soffset && soffset < tmp->eoffset) { /* bo and tmp overlap, invalid offset */ dev_err(rdev->dev, "bo %p va 0x%08X conflict with (bo %p 0x%08X 0x%08X)\n", - bo, (unsigned)bo_va->soffset, tmp->bo, + bo_va->bo, (unsigned)bo_va->soffset, tmp->bo, (unsigned)tmp->soffset, (unsigned)tmp->eoffset); - kfree(bo_va); mutex_unlock(&vm->mutex); return -EINVAL; } last_offset = tmp->eoffset; head = &tmp->vm_list; } - list_add(&bo_va->vm_list, head); - list_add_tail(&bo_va->bo_list, &bo->va); + + bo_va->soffset = soffset; + bo_va->eoffset = eoffset; + bo_va->flags = flags; + bo_va->valid = false; + list_move(&bo_va->vm_list, head); + mutex_unlock(&vm->mutex); return 0; } /** - * radeon_vm_get_addr - get the physical address of the page + * radeon_vm_map_gart - get the physical address of a gart page * * @rdev: radeon_device pointer - * @mem: ttm mem - * @pfn: pfn + * @addr: the unmapped addr * * Look up the physical address of the page that the pte resolves * to (cayman+). * Returns the physical address of the page. */ -static u64 radeon_vm_get_addr(struct radeon_device *rdev, - struct ttm_mem_reg *mem, - unsigned pfn) +uint64_t radeon_vm_map_gart(struct radeon_device *rdev, uint64_t addr) { - u64 addr = 0; - - switch (mem->mem_type) { - case TTM_PL_VRAM: - addr = (mem->start << PAGE_SHIFT); - addr += pfn * RADEON_GPU_PAGE_SIZE; - addr += rdev->vm_manager.vram_base_offset; - break; - case TTM_PL_TT: - /* offset inside page table */ - addr = mem->start << PAGE_SHIFT; - addr += pfn * RADEON_GPU_PAGE_SIZE; - addr = addr >> PAGE_SHIFT; - /* page table offset */ - addr = rdev->gart.pages_addr[addr]; - /* in case cpu page size != gpu page size*/ - addr += (pfn * RADEON_GPU_PAGE_SIZE) & (~PAGE_MASK); - break; - default: - break; - } - return addr; + uint64_t result; + + /* page table offset */ + result = rdev->gart.pages_addr[addr >> PAGE_SHIFT]; + + /* in case cpu page size != gpu page size*/ + result |= addr & (~PAGE_MASK); + + return result; } -/* object have to be reserved & global and local mutex must be locked */ /** * radeon_vm_bo_update_pte - map a bo into the vm page table * @@ -793,103 +874,160 @@ static u64 radeon_vm_get_addr(struct radeon_device *rdev, * * Fill in the page table entries for @bo (cayman+). * Returns 0 for success, -EINVAL for failure. + * + * Object have to be reserved & global and local mutex must be locked! */ int radeon_vm_bo_update_pte(struct radeon_device *rdev, struct radeon_vm *vm, struct radeon_bo *bo, struct ttm_mem_reg *mem) { + unsigned ridx = rdev->asic->vm.pt_ring_index; + struct radeon_ring *ring = &rdev->ring[ridx]; + struct radeon_semaphore *sem = NULL; struct radeon_bo_va *bo_va; - unsigned ngpu_pages, i; - uint64_t addr = 0, pfn; - uint32_t flags; + unsigned nptes, npdes, ndw; + uint64_t pe, addr; + uint64_t pfn; + int r; /* nothing to do if vm isn't bound */ - if (vm->id == -1) + if (vm->sa_bo == NULL) return 0; - bo_va = radeon_bo_va(bo, vm); + bo_va = radeon_vm_bo_find(vm, bo); if (bo_va == NULL) { dev_err(rdev->dev, "bo %p not in vm %p\n", bo, vm); return -EINVAL; } - if (bo_va->valid && mem) + if (!bo_va->soffset) { + dev_err(rdev->dev, "bo %p don't has a mapping in vm %p\n", + bo, vm); + return -EINVAL; + } + + if ((bo_va->valid && mem) || (!bo_va->valid && mem == NULL)) return 0; - ngpu_pages = radeon_bo_ngpu_pages(bo); bo_va->flags &= ~RADEON_VM_PAGE_VALID; bo_va->flags &= ~RADEON_VM_PAGE_SYSTEM; if (mem) { + addr = mem->start << PAGE_SHIFT; if (mem->mem_type != TTM_PL_SYSTEM) { bo_va->flags |= RADEON_VM_PAGE_VALID; bo_va->valid = true; } if (mem->mem_type == TTM_PL_TT) { bo_va->flags |= RADEON_VM_PAGE_SYSTEM; + } else { + addr += rdev->vm_manager.vram_base_offset; } + } else { + addr = 0; + bo_va->valid = false; } - pfn = bo_va->soffset / RADEON_GPU_PAGE_SIZE; - flags = rdev->vm_manager.funcs->page_flags(rdev, bo_va->vm, bo_va->flags); - for (i = 0, addr = 0; i < ngpu_pages; i++) { - if (mem && bo_va->valid) { - addr = radeon_vm_get_addr(rdev, mem, i); + + if (vm->fence && radeon_fence_signaled(vm->fence)) { + radeon_fence_unref(&vm->fence); + } + + if (vm->fence && vm->fence->ring != ridx) { + r = radeon_semaphore_create(rdev, &sem); + if (r) { + return r; } - rdev->vm_manager.funcs->set_page(rdev, bo_va->vm, i + pfn, addr, flags); } - rdev->vm_manager.funcs->tlb_flush(rdev, bo_va->vm); + + /* estimate number of dw needed */ + /* reserve space for 32-bit padding */ + ndw = 32; + + nptes = radeon_bo_ngpu_pages(bo); + + pfn = (bo_va->soffset / RADEON_GPU_PAGE_SIZE); + + /* handle cases where a bo spans several pdes */ + npdes = (ALIGN(pfn + nptes, RADEON_VM_PTE_COUNT) - + (pfn & ~(RADEON_VM_PTE_COUNT - 1))) >> RADEON_VM_BLOCK_SIZE; + + /* reserve space for one header for every 2k dwords */ + ndw += (nptes >> 11) * 3; + /* reserve space for pte addresses */ + ndw += nptes * 2; + + /* reserve space for one header for every 2k dwords */ + ndw += (npdes >> 11) * 3; + /* reserve space for pde addresses */ + ndw += npdes * 2; + + r = radeon_ring_lock(rdev, ring, ndw); + if (r) { + return r; + } + + if (sem && radeon_fence_need_sync(vm->fence, ridx)) { + radeon_semaphore_sync_rings(rdev, sem, vm->fence->ring, ridx); + radeon_fence_note_sync(vm->fence, ridx); + } + + /* update page table entries */ + pe = vm->pd_gpu_addr; + pe += radeon_vm_directory_size(rdev); + pe += (bo_va->soffset / RADEON_GPU_PAGE_SIZE) * 8; + + radeon_asic_vm_set_page(rdev, pe, addr, nptes, + RADEON_GPU_PAGE_SIZE, bo_va->flags); + + /* update page directory entries */ + addr = pe; + + pe = vm->pd_gpu_addr; + pe += ((bo_va->soffset / RADEON_GPU_PAGE_SIZE) >> RADEON_VM_BLOCK_SIZE) * 8; + + radeon_asic_vm_set_page(rdev, pe, addr, npdes, + RADEON_VM_PTE_COUNT * 8, RADEON_VM_PAGE_VALID); + + radeon_fence_unref(&vm->fence); + r = radeon_fence_emit(rdev, &vm->fence, ridx); + if (r) { + radeon_ring_unlock_undo(rdev, ring); + return r; + } + radeon_ring_unlock_commit(rdev, ring); + radeon_semaphore_free(rdev, &sem, vm->fence); + radeon_fence_unref(&vm->last_flush); return 0; } -/* object have to be reserved */ /** * radeon_vm_bo_rmv - remove a bo to a specific vm * * @rdev: radeon_device pointer - * @vm: requested vm - * @bo: radeon buffer object + * @bo_va: requested bo_va * - * Remove @bo from the requested vm (cayman+). - * Remove @bo from the list of bos associated with the vm and - * remove the ptes for @bo in the page table. + * Remove @bo_va->bo from the requested vm (cayman+). + * Remove @bo_va->bo from the list of bos associated with the bo_va->vm and + * remove the ptes for @bo_va in the page table. * Returns 0 for success. + * + * Object have to be reserved! */ int radeon_vm_bo_rmv(struct radeon_device *rdev, - struct radeon_vm *vm, - struct radeon_bo *bo) + struct radeon_bo_va *bo_va) { - struct radeon_bo_va *bo_va; int r; - bo_va = radeon_bo_va(bo, vm); - if (bo_va == NULL) - return 0; - - /* wait for va use to end */ - while (bo_va->fence) { - r = radeon_fence_wait(bo_va->fence, false); - if (r) { - DRM_ERROR("error while waiting for fence: %d\n", r); - } - if (r == -EDEADLK) { - r = radeon_gpu_reset(rdev); - if (!r) - continue; - } - break; - } - radeon_fence_unref(&bo_va->fence); - mutex_lock(&rdev->vm_manager.lock); - mutex_lock(&vm->mutex); - radeon_vm_bo_update_pte(rdev, vm, bo, NULL); + mutex_lock(&bo_va->vm->mutex); + r = radeon_vm_bo_update_pte(rdev, bo_va->vm, bo_va->bo, NULL); mutex_unlock(&rdev->vm_manager.lock); list_del(&bo_va->vm_list); - mutex_unlock(&vm->mutex); + mutex_unlock(&bo_va->vm->mutex); list_del(&bo_va->bo_list); kfree(bo_va); - return 0; + return r; } /** @@ -925,27 +1063,23 @@ void radeon_vm_bo_invalidate(struct radeon_device *rdev, */ int radeon_vm_init(struct radeon_device *rdev, struct radeon_vm *vm) { + struct radeon_bo_va *bo_va; int r; - vm->id = -1; + vm->id = 0; vm->fence = NULL; + vm->last_pfn = 0; mutex_init(&vm->mutex); INIT_LIST_HEAD(&vm->list); INIT_LIST_HEAD(&vm->va); - /* SI requires equal sized PTs for all VMs, so always set - * last_pfn to max_pfn. cayman allows variable sized - * pts so we can grow then as needed. Once we switch - * to two level pts we can unify this again. - */ - if (rdev->family >= CHIP_TAHITI) - vm->last_pfn = rdev->vm_manager.max_pfn; - else - vm->last_pfn = 0; + /* map the ib pool buffer at 0 in virtual address space, set * read only */ - r = radeon_vm_bo_add(rdev, vm, rdev->ring_tmp_bo.bo, 0, - RADEON_VM_PAGE_READABLE | RADEON_VM_PAGE_SNOOPED); + bo_va = radeon_vm_bo_add(rdev, vm, rdev->ring_tmp_bo.bo); + r = radeon_vm_bo_set_addr(rdev, bo_va, RADEON_VA_IB_OFFSET, + RADEON_VM_PAGE_READABLE | + RADEON_VM_PAGE_SNOOPED); return r; } @@ -965,7 +1099,7 @@ void radeon_vm_fini(struct radeon_device *rdev, struct radeon_vm *vm) mutex_lock(&rdev->vm_manager.lock); mutex_lock(&vm->mutex); - radeon_vm_unbind_locked(rdev, vm); + radeon_vm_free_pt(rdev, vm); mutex_unlock(&rdev->vm_manager.lock); /* remove all bo at this point non are busy any more because unbind @@ -973,10 +1107,9 @@ void radeon_vm_fini(struct radeon_device *rdev, struct radeon_vm *vm) */ r = radeon_bo_reserve(rdev->ring_tmp_bo.bo, false); if (!r) { - bo_va = radeon_bo_va(rdev->ring_tmp_bo.bo, vm); + bo_va = radeon_vm_bo_find(vm, rdev->ring_tmp_bo.bo); list_del_init(&bo_va->bo_list); list_del_init(&bo_va->vm_list); - radeon_fence_unref(&bo_va->fence); radeon_bo_unreserve(rdev->ring_tmp_bo.bo); kfree(bo_va); } @@ -988,10 +1121,11 @@ void radeon_vm_fini(struct radeon_device *rdev, struct radeon_vm *vm) r = radeon_bo_reserve(bo_va->bo, false); if (!r) { list_del_init(&bo_va->bo_list); - radeon_fence_unref(&bo_va->fence); radeon_bo_unreserve(bo_va->bo); kfree(bo_va); } } + radeon_fence_unref(&vm->fence); + radeon_fence_unref(&vm->last_flush); mutex_unlock(&vm->mutex); } diff --git a/drivers/gpu/drm/radeon/radeon_gem.c b/drivers/gpu/drm/radeon/radeon_gem.c index 1b57b0058ad6..6579befa5101 100644 --- a/drivers/gpu/drm/radeon/radeon_gem.c +++ b/drivers/gpu/drm/radeon/radeon_gem.c @@ -124,6 +124,30 @@ void radeon_gem_fini(struct radeon_device *rdev) */ int radeon_gem_object_open(struct drm_gem_object *obj, struct drm_file *file_priv) { + struct radeon_bo *rbo = gem_to_radeon_bo(obj); + struct radeon_device *rdev = rbo->rdev; + struct radeon_fpriv *fpriv = file_priv->driver_priv; + struct radeon_vm *vm = &fpriv->vm; + struct radeon_bo_va *bo_va; + int r; + + if (rdev->family < CHIP_CAYMAN) { + return 0; + } + + r = radeon_bo_reserve(rbo, false); + if (r) { + return r; + } + + bo_va = radeon_vm_bo_find(vm, rbo); + if (!bo_va) { + bo_va = radeon_vm_bo_add(rdev, vm, rbo); + } else { + ++bo_va->ref_count; + } + radeon_bo_unreserve(rbo); + return 0; } @@ -134,16 +158,25 @@ void radeon_gem_object_close(struct drm_gem_object *obj, struct radeon_device *rdev = rbo->rdev; struct radeon_fpriv *fpriv = file_priv->driver_priv; struct radeon_vm *vm = &fpriv->vm; + struct radeon_bo_va *bo_va; + int r; if (rdev->family < CHIP_CAYMAN) { return; } - if (radeon_bo_reserve(rbo, false)) { - dev_err(rdev->dev, "leaking bo va because we fail to reserve bo\n"); + r = radeon_bo_reserve(rbo, true); + if (r) { + dev_err(rdev->dev, "leaking bo va because " + "we fail to reserve bo (%d)\n", r); return; } - radeon_vm_bo_rmv(rdev, vm, rbo); + bo_va = radeon_vm_bo_find(vm, rbo); + if (bo_va) { + if (--bo_va->ref_count == 0) { + radeon_vm_bo_rmv(rdev, bo_va); + } + } radeon_bo_unreserve(rbo); } @@ -459,19 +492,24 @@ int radeon_gem_va_ioctl(struct drm_device *dev, void *data, drm_gem_object_unreference_unlocked(gobj); return r; } + bo_va = radeon_vm_bo_find(&fpriv->vm, rbo); + if (!bo_va) { + args->operation = RADEON_VA_RESULT_ERROR; + drm_gem_object_unreference_unlocked(gobj); + return -ENOENT; + } + switch (args->operation) { case RADEON_VA_MAP: - bo_va = radeon_bo_va(rbo, &fpriv->vm); - if (bo_va) { + if (bo_va->soffset) { args->operation = RADEON_VA_RESULT_VA_EXIST; args->offset = bo_va->soffset; goto out; } - r = radeon_vm_bo_add(rdev, &fpriv->vm, rbo, - args->offset, args->flags); + r = radeon_vm_bo_set_addr(rdev, bo_va, args->offset, args->flags); break; case RADEON_VA_UNMAP: - r = radeon_vm_bo_rmv(rdev, &fpriv->vm, rbo); + r = radeon_vm_bo_set_addr(rdev, bo_va, 0, 0); break; default: break; diff --git a/drivers/gpu/drm/radeon/radeon_ioc32.c b/drivers/gpu/drm/radeon/radeon_ioc32.c index 48b7cea31e08..c4bb2269be10 100644 --- a/drivers/gpu/drm/radeon/radeon_ioc32.c +++ b/drivers/gpu/drm/radeon/radeon_ioc32.c @@ -369,7 +369,7 @@ static int compat_radeon_cp_setparam(struct file *file, unsigned int cmd, #define compat_radeon_cp_setparam NULL #endif /* X86_64 || IA64 */ -drm_ioctl_compat_t *radeon_compat_ioctls[] = { +static drm_ioctl_compat_t *radeon_compat_ioctls[] = { [DRM_RADEON_CP_INIT] = compat_radeon_cp_init, [DRM_RADEON_CLEAR] = compat_radeon_cp_clear, [DRM_RADEON_STIPPLE] = compat_radeon_cp_stipple, diff --git a/drivers/gpu/drm/radeon/radeon_irq_kms.c b/drivers/gpu/drm/radeon/radeon_irq_kms.c index afaa1727abd2..b06fa5936100 100644 --- a/drivers/gpu/drm/radeon/radeon_irq_kms.c +++ b/drivers/gpu/drm/radeon/radeon_irq_kms.c @@ -99,7 +99,6 @@ void radeon_driver_irq_preinstall_kms(struct drm_device *dev) /* Disable *all* interrupts */ for (i = 0; i < RADEON_NUM_RINGS; i++) atomic_set(&rdev->irq.ring_int[i], 0); - rdev->irq.gui_idle = false; for (i = 0; i < RADEON_MAX_HPD_PINS; i++) rdev->irq.hpd[i] = false; for (i = 0; i < RADEON_MAX_CRTCS; i++) { @@ -147,7 +146,6 @@ void radeon_driver_irq_uninstall_kms(struct drm_device *dev) /* Disable *all* interrupts */ for (i = 0; i < RADEON_NUM_RINGS; i++) atomic_set(&rdev->irq.ring_int[i], 0); - rdev->irq.gui_idle = false; for (i = 0; i < RADEON_MAX_HPD_PINS; i++) rdev->irq.hpd[i] = false; for (i = 0; i < RADEON_MAX_CRTCS; i++) { @@ -204,6 +202,16 @@ static bool radeon_msi_ok(struct radeon_device *rdev) (rdev->pdev->subsystem_device == 0x01fd)) return true; + /* Gateway RS690 only seems to work with MSIs. */ + if ((rdev->pdev->device == 0x791f) && + (rdev->pdev->subsystem_vendor == 0x107b) && + (rdev->pdev->subsystem_device == 0x0185)) + return true; + + /* try and enable MSIs by default on all RS690s */ + if (rdev->family == CHIP_RS690) + return true; + /* RV515 seems to have MSI issues where it loses * MSI rearms occasionally. This leads to lockups and freezes. * disable it by default. @@ -457,34 +465,3 @@ void radeon_irq_kms_disable_hpd(struct radeon_device *rdev, unsigned hpd_mask) spin_unlock_irqrestore(&rdev->irq.lock, irqflags); } -/** - * radeon_irq_kms_wait_gui_idle - waits for drawing engine to be idle - * - * @rdev: radeon device pointer - * - * Enabled the GUI idle interrupt and waits for it to fire (r6xx+). - * This is currently used to make sure the 3D engine is idle for power - * management, but should be replaces with proper fence waits. - * GUI idle interrupts don't work very well on pre-r6xx hw and it also - * does not take into account other aspects of the chip that may be busy. - * DO NOT USE GOING FORWARD. - */ -int radeon_irq_kms_wait_gui_idle(struct radeon_device *rdev) -{ - unsigned long irqflags; - int r; - - spin_lock_irqsave(&rdev->irq.lock, irqflags); - rdev->irq.gui_idle = true; - radeon_irq_set(rdev); - spin_unlock_irqrestore(&rdev->irq.lock, irqflags); - - r = wait_event_timeout(rdev->irq.idle_queue, radeon_gui_idle(rdev), - msecs_to_jiffies(RADEON_WAIT_IDLE_TIMEOUT)); - - spin_lock_irqsave(&rdev->irq.lock, irqflags); - rdev->irq.gui_idle = false; - radeon_irq_set(rdev); - spin_unlock_irqrestore(&rdev->irq.lock, irqflags); - return r; -} diff --git a/drivers/gpu/drm/radeon/radeon_kms.c b/drivers/gpu/drm/radeon/radeon_kms.c index 414b4acf6947..cb8e94d1a2b2 100644 --- a/drivers/gpu/drm/radeon/radeon_kms.c +++ b/drivers/gpu/drm/radeon/radeon_kms.c @@ -51,6 +51,7 @@ int radeon_driver_unload_kms(struct drm_device *dev) if (rdev == NULL) return 0; + radeon_acpi_fini(rdev); radeon_modeset_fini(rdev); radeon_device_fini(rdev); kfree(rdev); @@ -103,11 +104,6 @@ int radeon_driver_load_kms(struct drm_device *dev, unsigned long flags) goto out; } - /* Call ACPI methods */ - acpi_status = radeon_acpi_init(rdev); - if (acpi_status) - dev_dbg(&dev->pdev->dev, "Error during ACPI methods call\n"); - /* Again modeset_init should fail only on fatal error * otherwise it should provide enough functionalities * for shadowfb to run @@ -115,6 +111,17 @@ int radeon_driver_load_kms(struct drm_device *dev, unsigned long flags) r = radeon_modeset_init(rdev); if (r) dev_err(&dev->pdev->dev, "Fatal error during modeset init\n"); + + /* Call ACPI methods: require modeset init + * but failure is not fatal + */ + if (!r) { + acpi_status = radeon_acpi_init(rdev); + if (acpi_status) + dev_dbg(&dev->pdev->dev, + "Error during ACPI methods call\n"); + } + out: if (r) radeon_driver_unload_kms(dev); diff --git a/drivers/gpu/drm/radeon/radeon_legacy_crtc.c b/drivers/gpu/drm/radeon/radeon_legacy_crtc.c index 94b4a1c12893..5677a424b585 100644 --- a/drivers/gpu/drm/radeon/radeon_legacy_crtc.c +++ b/drivers/gpu/drm/radeon/radeon_legacy_crtc.c @@ -206,11 +206,6 @@ static void radeon_legacy_rmx_mode_set(struct drm_crtc *crtc, WREG32(RADEON_FP_CRTC_V_TOTAL_DISP, fp_crtc_v_total_disp); } -void radeon_restore_common_regs(struct drm_device *dev) -{ - /* don't need this yet */ -} - static void radeon_pll_wait_for_read_update_complete(struct drm_device *dev) { struct radeon_device *rdev = dev->dev_private; @@ -295,7 +290,7 @@ static uint8_t radeon_compute_pll_gain(uint16_t ref_freq, uint16_t ref_div, return 1; } -void radeon_crtc_dpms(struct drm_crtc *crtc, int mode) +static void radeon_crtc_dpms(struct drm_crtc *crtc, int mode) { struct radeon_crtc *radeon_crtc = to_radeon_crtc(crtc); struct drm_device *dev = crtc->dev; diff --git a/drivers/gpu/drm/radeon/radeon_legacy_encoders.c b/drivers/gpu/drm/radeon/radeon_legacy_encoders.c index 670e9910f869..8ad9c5f16014 100644 --- a/drivers/gpu/drm/radeon/radeon_legacy_encoders.c +++ b/drivers/gpu/drm/radeon/radeon_legacy_encoders.c @@ -271,13 +271,6 @@ static const struct drm_encoder_helper_funcs radeon_legacy_lvds_helper_funcs = { #if defined(CONFIG_BACKLIGHT_CLASS_DEVICE) || defined(CONFIG_BACKLIGHT_CLASS_DEVICE_MODULE) -#define MAX_RADEON_LEVEL 0xFF - -struct radeon_backlight_privdata { - struct radeon_encoder *encoder; - uint8_t negative; -}; - static uint8_t radeon_legacy_lvds_level(struct backlight_device *bd) { struct radeon_backlight_privdata *pdata = bl_get_data(bd); @@ -286,21 +279,33 @@ static uint8_t radeon_legacy_lvds_level(struct backlight_device *bd) /* Convert brightness to hardware level */ if (bd->props.brightness < 0) level = 0; - else if (bd->props.brightness > MAX_RADEON_LEVEL) - level = MAX_RADEON_LEVEL; + else if (bd->props.brightness > RADEON_MAX_BL_LEVEL) + level = RADEON_MAX_BL_LEVEL; else level = bd->props.brightness; if (pdata->negative) - level = MAX_RADEON_LEVEL - level; + level = RADEON_MAX_BL_LEVEL - level; return level; } -static int radeon_legacy_backlight_update_status(struct backlight_device *bd) +u8 +radeon_legacy_get_backlight_level(struct radeon_encoder *radeon_encoder) +{ + struct drm_device *dev = radeon_encoder->base.dev; + struct radeon_device *rdev = dev->dev_private; + u8 backlight_level; + + backlight_level = (RREG32(RADEON_LVDS_GEN_CNTL) >> + RADEON_LVDS_BL_MOD_LEVEL_SHIFT) & 0xff; + + return backlight_level; +} + +void +radeon_legacy_set_backlight_level(struct radeon_encoder *radeon_encoder, u8 level) { - struct radeon_backlight_privdata *pdata = bl_get_data(bd); - struct radeon_encoder *radeon_encoder = pdata->encoder; struct drm_device *dev = radeon_encoder->base.dev; struct radeon_device *rdev = dev->dev_private; int dpms_mode = DRM_MODE_DPMS_ON; @@ -308,19 +313,31 @@ static int radeon_legacy_backlight_update_status(struct backlight_device *bd) if (radeon_encoder->enc_priv) { if (rdev->is_atom_bios) { struct radeon_encoder_atom_dig *lvds = radeon_encoder->enc_priv; - dpms_mode = lvds->dpms_mode; - lvds->backlight_level = radeon_legacy_lvds_level(bd); + if (lvds->backlight_level > 0) + dpms_mode = lvds->dpms_mode; + else + dpms_mode = DRM_MODE_DPMS_OFF; + lvds->backlight_level = level; } else { struct radeon_encoder_lvds *lvds = radeon_encoder->enc_priv; - dpms_mode = lvds->dpms_mode; - lvds->backlight_level = radeon_legacy_lvds_level(bd); + if (lvds->backlight_level > 0) + dpms_mode = lvds->dpms_mode; + else + dpms_mode = DRM_MODE_DPMS_OFF; + lvds->backlight_level = level; } } - if (bd->props.brightness > 0) - radeon_legacy_lvds_update(&radeon_encoder->base, dpms_mode); - else - radeon_legacy_lvds_update(&radeon_encoder->base, DRM_MODE_DPMS_OFF); + radeon_legacy_lvds_update(&radeon_encoder->base, dpms_mode); +} + +static int radeon_legacy_backlight_update_status(struct backlight_device *bd) +{ + struct radeon_backlight_privdata *pdata = bl_get_data(bd); + struct radeon_encoder *radeon_encoder = pdata->encoder; + + radeon_legacy_set_backlight_level(radeon_encoder, + radeon_legacy_lvds_level(bd)); return 0; } @@ -336,7 +353,7 @@ static int radeon_legacy_backlight_get_brightness(struct backlight_device *bd) backlight_level = (RREG32(RADEON_LVDS_GEN_CNTL) >> RADEON_LVDS_BL_MOD_LEVEL_SHIFT) & 0xff; - return pdata->negative ? MAX_RADEON_LEVEL - backlight_level : backlight_level; + return pdata->negative ? RADEON_MAX_BL_LEVEL - backlight_level : backlight_level; } static const struct backlight_ops radeon_backlight_ops = { @@ -370,7 +387,7 @@ void radeon_legacy_backlight_init(struct radeon_encoder *radeon_encoder, } memset(&props, 0, sizeof(props)); - props.max_brightness = MAX_RADEON_LEVEL; + props.max_brightness = RADEON_MAX_BL_LEVEL; props.type = BACKLIGHT_RAW; bd = backlight_device_register("radeon_bl", &drm_connector->kdev, pdata, &radeon_backlight_ops, &props); @@ -449,7 +466,7 @@ static void radeon_legacy_backlight_exit(struct radeon_encoder *radeon_encoder) } if (bd) { - struct radeon_legacy_backlight_privdata *pdata; + struct radeon_backlight_privdata *pdata; pdata = bl_get_data(bd); backlight_device_unregister(bd); diff --git a/drivers/gpu/drm/radeon/radeon_mode.h b/drivers/gpu/drm/radeon/radeon_mode.h index d56978949f34..527761801590 100644 --- a/drivers/gpu/drm/radeon/radeon_mode.h +++ b/drivers/gpu/drm/radeon/radeon_mode.h @@ -252,8 +252,23 @@ struct radeon_mode_info { /* pointer to fbdev info structure */ struct radeon_fbdev *rfbdev; + /* firmware flags */ + u16 firmware_flags; + /* pointer to backlight encoder */ + struct radeon_encoder *bl_encoder; }; +#define RADEON_MAX_BL_LEVEL 0xFF + +#if defined(CONFIG_BACKLIGHT_CLASS_DEVICE) || defined(CONFIG_BACKLIGHT_CLASS_DEVICE_MODULE) + +struct radeon_backlight_privdata { + struct radeon_encoder *encoder; + uint8_t negative; +}; + +#endif + #define MAX_H_CODE_TIMING_LEN 32 #define MAX_V_CODE_TIMING_LEN 32 @@ -269,6 +284,18 @@ struct radeon_tv_regs { uint16_t v_code_timing[MAX_V_CODE_TIMING_LEN]; }; +struct radeon_atom_ss { + uint16_t percentage; + uint8_t type; + uint16_t step; + uint8_t delay; + uint8_t range; + uint8_t refdiv; + /* asic_ss */ + uint16_t rate; + uint16_t amount; +}; + struct radeon_crtc { struct drm_crtc base; int crtc_id; @@ -293,6 +320,16 @@ struct radeon_crtc { /* page flipping */ struct radeon_unpin_work *unpin_work; int deferred_flip_completion; + /* pll sharing */ + struct radeon_atom_ss ss; + bool ss_enabled; + u32 adjusted_clock; + int bpc; + u32 pll_reference_div; + u32 pll_post_div; + u32 pll_flags; + struct drm_encoder *encoder; + struct drm_connector *connector; }; struct radeon_encoder_primary_dac { @@ -346,18 +383,6 @@ struct radeon_encoder_ext_tmds { }; /* spread spectrum */ -struct radeon_atom_ss { - uint16_t percentage; - uint8_t type; - uint16_t step; - uint8_t delay; - uint8_t range; - uint8_t refdiv; - /* asic_ss */ - uint16_t rate; - uint16_t amount; -}; - struct radeon_encoder_atom_dig { bool linkb; /* atom dig */ diff --git a/drivers/gpu/drm/radeon/radeon_object.c b/drivers/gpu/drm/radeon/radeon_object.c index 9024e7222839..a236795dc69a 100644 --- a/drivers/gpu/drm/radeon/radeon_object.c +++ b/drivers/gpu/drm/radeon/radeon_object.c @@ -52,7 +52,7 @@ void radeon_bo_clear_va(struct radeon_bo *bo) list_for_each_entry_safe(bo_va, tmp, &bo->va, bo_list) { /* remove from all vm address space */ - radeon_vm_bo_rmv(bo->rdev, bo_va->vm, bo); + radeon_vm_bo_rmv(bo->rdev, bo_va); } } @@ -627,18 +627,17 @@ int radeon_bo_wait(struct radeon_bo *bo, u32 *mem_type, bool no_wait) /** * radeon_bo_reserve - reserve bo * @bo: bo structure - * @no_wait: don't sleep while trying to reserve (return -EBUSY) + * @no_intr: don't return -ERESTARTSYS on pending signal * * Returns: - * -EBUSY: buffer is busy and @no_wait is true * -ERESTARTSYS: A wait for the buffer to become unreserved was interrupted by * a signal. Release all buffer reservations and return to user-space. */ -int radeon_bo_reserve(struct radeon_bo *bo, bool no_wait) +int radeon_bo_reserve(struct radeon_bo *bo, bool no_intr) { int r; - r = ttm_bo_reserve(&bo->tbo, true, no_wait, false, 0); + r = ttm_bo_reserve(&bo->tbo, !no_intr, false, false, 0); if (unlikely(r != 0)) { if (r != -ERESTARTSYS) dev_err(bo->rdev->dev, "%p reserve failed\n", bo); @@ -646,16 +645,3 @@ int radeon_bo_reserve(struct radeon_bo *bo, bool no_wait) } return 0; } - -/* object have to be reserved */ -struct radeon_bo_va *radeon_bo_va(struct radeon_bo *rbo, struct radeon_vm *vm) -{ - struct radeon_bo_va *bo_va; - - list_for_each_entry(bo_va, &rbo->va, bo_list) { - if (bo_va->vm == vm) { - return bo_va; - } - } - return NULL; -} diff --git a/drivers/gpu/drm/radeon/radeon_object.h b/drivers/gpu/drm/radeon/radeon_object.h index 17fb99f177cf..93cd491fff2e 100644 --- a/drivers/gpu/drm/radeon/radeon_object.h +++ b/drivers/gpu/drm/radeon/radeon_object.h @@ -52,7 +52,7 @@ static inline unsigned radeon_mem_type_to_domain(u32 mem_type) return 0; } -int radeon_bo_reserve(struct radeon_bo *bo, bool no_wait); +int radeon_bo_reserve(struct radeon_bo *bo, bool no_intr); static inline void radeon_bo_unreserve(struct radeon_bo *bo) { @@ -141,8 +141,6 @@ extern void radeon_bo_move_notify(struct ttm_buffer_object *bo, struct ttm_mem_reg *mem); extern int radeon_bo_fault_reserve_notify(struct ttm_buffer_object *bo); extern int radeon_bo_get_surface_reg(struct radeon_bo *bo); -extern struct radeon_bo_va *radeon_bo_va(struct radeon_bo *rbo, - struct radeon_vm *vm); /* * sub allocation diff --git a/drivers/gpu/drm/radeon/radeon_pm.c b/drivers/gpu/drm/radeon/radeon_pm.c index 7ae606600107..bc2e7050a9d8 100644 --- a/drivers/gpu/drm/radeon/radeon_pm.c +++ b/drivers/gpu/drm/radeon/radeon_pm.c @@ -24,9 +24,6 @@ #include "radeon.h" #include "avivod.h" #include "atom.h" -#ifdef CONFIG_ACPI -#include <linux/acpi.h> -#endif #include <linux/power_supply.h> #include <linux/hwmon.h> #include <linux/hwmon-sysfs.h> @@ -36,7 +33,7 @@ #define RADEON_WAIT_VBLANK_TIMEOUT 200 static const char *radeon_pm_state_type_name[5] = { - "Default", + "", "Powersave", "Battery", "Balanced", @@ -50,8 +47,6 @@ static bool radeon_pm_debug_check_in_vbl(struct radeon_device *rdev, bool finish static void radeon_pm_update_profile(struct radeon_device *rdev); static void radeon_pm_set_clocks(struct radeon_device *rdev); -#define ACPI_AC_CLASS "ac_adapter" - int radeon_pm_get_type_index(struct radeon_device *rdev, enum radeon_pm_state_type ps_type, int instance) @@ -70,33 +65,17 @@ int radeon_pm_get_type_index(struct radeon_device *rdev, return rdev->pm.default_power_state_index; } -#ifdef CONFIG_ACPI -static int radeon_acpi_event(struct notifier_block *nb, - unsigned long val, - void *data) +void radeon_pm_acpi_event_handler(struct radeon_device *rdev) { - struct radeon_device *rdev = container_of(nb, struct radeon_device, acpi_nb); - struct acpi_bus_event *entry = (struct acpi_bus_event *)data; - - if (strcmp(entry->device_class, ACPI_AC_CLASS) == 0) { - if (power_supply_is_system_supplied() > 0) - DRM_DEBUG_DRIVER("pm: AC\n"); - else - DRM_DEBUG_DRIVER("pm: DC\n"); - - if (rdev->pm.pm_method == PM_METHOD_PROFILE) { - if (rdev->pm.profile == PM_PROFILE_AUTO) { - mutex_lock(&rdev->pm.mutex); - radeon_pm_update_profile(rdev); - radeon_pm_set_clocks(rdev); - mutex_unlock(&rdev->pm.mutex); - } + if (rdev->pm.pm_method == PM_METHOD_PROFILE) { + if (rdev->pm.profile == PM_PROFILE_AUTO) { + mutex_lock(&rdev->pm.mutex); + radeon_pm_update_profile(rdev); + radeon_pm_set_clocks(rdev); + mutex_unlock(&rdev->pm.mutex); } } - - return NOTIFY_OK; } -#endif static void radeon_pm_update_profile(struct radeon_device *rdev) { @@ -188,8 +167,21 @@ static void radeon_set_power_state(struct radeon_device *rdev) if (sclk > rdev->pm.default_sclk) sclk = rdev->pm.default_sclk; - mclk = rdev->pm.power_state[rdev->pm.requested_power_state_index]. - clock_info[rdev->pm.requested_clock_mode_index].mclk; + /* starting with BTC, there is one state that is used for both + * MH and SH. Difference is that we always use the high clock index for + * mclk. + */ + if ((rdev->pm.pm_method == PM_METHOD_PROFILE) && + (rdev->family >= CHIP_BARTS) && + rdev->pm.active_crtc_count && + ((rdev->pm.profile_index == PM_PROFILE_MID_MH_IDX) || + (rdev->pm.profile_index == PM_PROFILE_LOW_MH_IDX))) + mclk = rdev->pm.power_state[rdev->pm.requested_power_state_index]. + clock_info[rdev->pm.profiles[PM_PROFILE_HIGH_MH_IDX].dpms_on_cm_idx].mclk; + else + mclk = rdev->pm.power_state[rdev->pm.requested_power_state_index]. + clock_info[rdev->pm.requested_clock_mode_index].mclk; + if (mclk > rdev->pm.default_mclk) mclk = rdev->pm.default_mclk; @@ -253,18 +245,13 @@ static void radeon_pm_set_clocks(struct radeon_device *rdev) down_write(&rdev->pm.mclk_lock); mutex_lock(&rdev->ring_lock); - /* gui idle int has issues on older chips it seems */ - if (rdev->family >= CHIP_R600) { - if (rdev->irq.installed) { - /* wait for GPU to become idle */ - radeon_irq_kms_wait_gui_idle(rdev); - } - } else { - struct radeon_ring *ring = &rdev->ring[RADEON_RING_TYPE_GFX_INDEX]; - if (ring->ready) { - radeon_fence_wait_empty_locked(rdev, RADEON_RING_TYPE_GFX_INDEX); - } + /* wait for the rings to drain */ + for (i = 0; i < RADEON_NUM_RINGS; i++) { + struct radeon_ring *ring = &rdev->ring[i]; + if (ring->ready) + radeon_fence_wait_empty_locked(rdev, i); } + radeon_unmap_vram_bos(rdev); if (rdev->irq.installed) { @@ -320,17 +307,15 @@ static void radeon_pm_print_states(struct radeon_device *rdev) for (j = 0; j < power_state->num_clock_modes; j++) { clock_info = &(power_state->clock_info[j]); if (rdev->flags & RADEON_IS_IGP) - DRM_DEBUG_DRIVER("\t\t%d e: %d%s\n", - j, - clock_info->sclk * 10, - clock_info->flags & RADEON_PM_MODE_NO_DISPLAY ? "\tNo display only" : ""); + DRM_DEBUG_DRIVER("\t\t%d e: %d\n", + j, + clock_info->sclk * 10); else - DRM_DEBUG_DRIVER("\t\t%d e: %d\tm: %d\tv: %d%s\n", - j, - clock_info->sclk * 10, - clock_info->mclk * 10, - clock_info->voltage.voltage, - clock_info->flags & RADEON_PM_MODE_NO_DISPLAY ? "\tNo display only" : ""); + DRM_DEBUG_DRIVER("\t\t%d e: %d\tm: %d\tv: %d\n", + j, + clock_info->sclk * 10, + clock_info->mclk * 10, + clock_info->voltage.voltage); } } } @@ -547,7 +532,9 @@ void radeon_pm_suspend(struct radeon_device *rdev) void radeon_pm_resume(struct radeon_device *rdev) { /* set up the default clocks if the MC ucode is loaded */ - if (ASIC_IS_DCE5(rdev) && rdev->mc_fw) { + if ((rdev->family >= CHIP_BARTS) && + (rdev->family <= CHIP_CAYMAN) && + rdev->mc_fw) { if (rdev->pm.default_vddc) radeon_atom_set_voltage(rdev, rdev->pm.default_vddc, SET_VOLTAGE_TYPE_ASIC_VDDC); @@ -602,7 +589,9 @@ int radeon_pm_init(struct radeon_device *rdev) radeon_pm_print_states(rdev); radeon_pm_init_profile(rdev); /* set up the default clocks if the MC ucode is loaded */ - if (ASIC_IS_DCE5(rdev) && rdev->mc_fw) { + if ((rdev->family >= CHIP_BARTS) && + (rdev->family <= CHIP_CAYMAN) && + rdev->mc_fw) { if (rdev->pm.default_vddc) radeon_atom_set_voltage(rdev, rdev->pm.default_vddc, SET_VOLTAGE_TYPE_ASIC_VDDC); @@ -632,10 +621,6 @@ int radeon_pm_init(struct radeon_device *rdev) if (ret) DRM_ERROR("failed to create device file for power method\n"); -#ifdef CONFIG_ACPI - rdev->acpi_nb.notifier_call = radeon_acpi_event; - register_acpi_notifier(&rdev->acpi_nb); -#endif if (radeon_debugfs_pm_init(rdev)) { DRM_ERROR("Failed to register debugfs file for PM!\n"); } @@ -666,9 +651,6 @@ void radeon_pm_fini(struct radeon_device *rdev) device_remove_file(rdev->dev, &dev_attr_power_profile); device_remove_file(rdev->dev, &dev_attr_power_method); -#ifdef CONFIG_ACPI - unregister_acpi_notifier(&rdev->acpi_nb); -#endif } if (rdev->pm.power_state) diff --git a/drivers/gpu/drm/radeon/radeon_ring.c b/drivers/gpu/drm/radeon/radeon_ring.c index 43c431a2686d..028508859a3b 100644 --- a/drivers/gpu/drm/radeon/radeon_ring.c +++ b/drivers/gpu/drm/radeon/radeon_ring.c @@ -43,7 +43,7 @@ * produce command buffers which are send to the kernel and * put in IBs for execution by the requested ring. */ -int radeon_debugfs_sa_init(struct radeon_device *rdev); +static int radeon_debugfs_sa_init(struct radeon_device *rdev); /** * radeon_ib_get - request an IB (Indirect Buffer) @@ -58,7 +58,8 @@ int radeon_debugfs_sa_init(struct radeon_device *rdev); * Returns 0 on success, error on failure. */ int radeon_ib_get(struct radeon_device *rdev, int ring, - struct radeon_ib *ib, unsigned size) + struct radeon_ib *ib, struct radeon_vm *vm, + unsigned size) { int i, r; @@ -76,8 +77,15 @@ int radeon_ib_get(struct radeon_device *rdev, int ring, ib->ring = ring; ib->fence = NULL; ib->ptr = radeon_sa_bo_cpu_addr(ib->sa_bo); - ib->gpu_addr = radeon_sa_bo_gpu_addr(ib->sa_bo); - ib->vm_id = 0; + ib->vm = vm; + if (vm) { + /* ib pool is bound at RADEON_VA_IB_OFFSET in virtual address + * space and soffset is the offset inside the pool bo + */ + ib->gpu_addr = ib->sa_bo->soffset + RADEON_VA_IB_OFFSET; + } else { + ib->gpu_addr = radeon_sa_bo_gpu_addr(ib->sa_bo); + } ib->is_const_ib = false; for (i = 0; i < RADEON_NUM_RINGS; ++i) ib->sync_to[i] = NULL; @@ -152,6 +160,10 @@ int radeon_ib_schedule(struct radeon_device *rdev, struct radeon_ib *ib, if (!need_sync) { radeon_semaphore_free(rdev, &ib->semaphore, NULL); } + /* if we can't remember our last VM flush then flush now! */ + if (ib->vm && !ib->vm->last_flush) { + radeon_ring_vm_flush(rdev, ib->ring, ib->vm); + } if (const_ib) { radeon_ring_ib_execute(rdev, const_ib->ring, const_ib); radeon_semaphore_free(rdev, &const_ib->semaphore, NULL); @@ -166,6 +178,10 @@ int radeon_ib_schedule(struct radeon_device *rdev, struct radeon_ib *ib, if (const_ib) { const_ib->fence = radeon_fence_ref(ib->fence); } + /* we just flushed the VM, remember that */ + if (ib->vm && !ib->vm->last_flush) { + ib->vm->last_flush = radeon_fence_ref(ib->fence); + } radeon_ring_unlock_commit(rdev, ring); return 0; } @@ -275,7 +291,7 @@ int radeon_ib_ring_tests(struct radeon_device *rdev) * wptr. The GPU then starts fetching commands and executes * them until the pointers are equal again. */ -int radeon_debugfs_ring_init(struct radeon_device *rdev, struct radeon_ring *ring); +static int radeon_debugfs_ring_init(struct radeon_device *rdev, struct radeon_ring *ring); /** * radeon_ring_write - write a value to the ring @@ -803,7 +819,7 @@ static struct drm_info_list radeon_debugfs_sa_list[] = { #endif -int radeon_debugfs_ring_init(struct radeon_device *rdev, struct radeon_ring *ring) +static int radeon_debugfs_ring_init(struct radeon_device *rdev, struct radeon_ring *ring) { #if defined(CONFIG_DEBUG_FS) unsigned i; @@ -823,7 +839,7 @@ int radeon_debugfs_ring_init(struct radeon_device *rdev, struct radeon_ring *rin return 0; } -int radeon_debugfs_sa_init(struct radeon_device *rdev) +static int radeon_debugfs_sa_init(struct radeon_device *rdev) { #if defined(CONFIG_DEBUG_FS) return radeon_debugfs_add_files(rdev, radeon_debugfs_sa_list, 1); diff --git a/drivers/gpu/drm/radeon/radeon_sa.c b/drivers/gpu/drm/radeon/radeon_sa.c index 4e771240fdd0..105fde69d045 100644 --- a/drivers/gpu/drm/radeon/radeon_sa.c +++ b/drivers/gpu/drm/radeon/radeon_sa.c @@ -316,7 +316,7 @@ int radeon_sa_bo_new(struct radeon_device *rdev, { struct radeon_fence *fences[RADEON_NUM_RINGS]; unsigned tries[RADEON_NUM_RINGS]; - int i, r = -ENOMEM; + int i, r; BUG_ON(align > RADEON_GPU_PAGE_SIZE); BUG_ON(size > sa_manager->size); @@ -331,7 +331,7 @@ int radeon_sa_bo_new(struct radeon_device *rdev, INIT_LIST_HEAD(&(*sa_bo)->flist); spin_lock(&sa_manager->wq.lock); - while(1) { + do { for (i = 0; i < RADEON_NUM_RINGS; ++i) { fences[i] = NULL; tries[i] = 0; @@ -349,26 +349,22 @@ int radeon_sa_bo_new(struct radeon_device *rdev, /* see if we can skip over some allocations */ } while (radeon_sa_bo_next_hole(sa_manager, fences, tries)); - if (!block) { - break; - } - spin_unlock(&sa_manager->wq.lock); r = radeon_fence_wait_any(rdev, fences, false); spin_lock(&sa_manager->wq.lock); /* if we have nothing to wait for block */ - if (r == -ENOENT) { + if (r == -ENOENT && block) { r = wait_event_interruptible_locked( sa_manager->wq, radeon_sa_event(sa_manager, size, align) ); + + } else if (r == -ENOENT) { + r = -ENOMEM; } - if (r) { - goto out_err; - } - }; -out_err: + } while (!r); + spin_unlock(&sa_manager->wq.lock); kfree(*sa_bo); *sa_bo = NULL; diff --git a/drivers/gpu/drm/radeon/radeon_test.c b/drivers/gpu/drm/radeon/radeon_test.c index 7c16540c10ff..587c09a00ba2 100644 --- a/drivers/gpu/drm/radeon/radeon_test.c +++ b/drivers/gpu/drm/radeon/radeon_test.c @@ -313,7 +313,7 @@ out_cleanup: printk(KERN_WARNING "Error while testing ring sync (%d).\n", r); } -void radeon_test_ring_sync2(struct radeon_device *rdev, +static void radeon_test_ring_sync2(struct radeon_device *rdev, struct radeon_ring *ringA, struct radeon_ring *ringB, struct radeon_ring *ringC) diff --git a/drivers/gpu/drm/radeon/radeon_ttm.c b/drivers/gpu/drm/radeon/radeon_ttm.c index 5b71c716d83f..5ebe1b3e5db2 100644 --- a/drivers/gpu/drm/radeon/radeon_ttm.c +++ b/drivers/gpu/drm/radeon/radeon_ttm.c @@ -549,7 +549,7 @@ static struct ttm_backend_func radeon_backend_func = { .destroy = &radeon_ttm_backend_destroy, }; -struct ttm_tt *radeon_ttm_tt_create(struct ttm_bo_device *bdev, +static struct ttm_tt *radeon_ttm_tt_create(struct ttm_bo_device *bdev, unsigned long size, uint32_t page_flags, struct page *dummy_read_page) { diff --git a/drivers/gpu/drm/radeon/rs400.c b/drivers/gpu/drm/radeon/rs400.c index 2752f7f78237..73051ce3121e 100644 --- a/drivers/gpu/drm/radeon/rs400.c +++ b/drivers/gpu/drm/radeon/rs400.c @@ -242,7 +242,7 @@ int rs400_mc_wait_for_idle(struct radeon_device *rdev) return -1; } -void rs400_gpu_init(struct radeon_device *rdev) +static void rs400_gpu_init(struct radeon_device *rdev) { /* FIXME: is this correct ? */ r420_pipes_init(rdev); @@ -252,7 +252,7 @@ void rs400_gpu_init(struct radeon_device *rdev) } } -void rs400_mc_init(struct radeon_device *rdev) +static void rs400_mc_init(struct radeon_device *rdev) { u64 base; @@ -370,7 +370,7 @@ static int rs400_debugfs_pcie_gart_info_init(struct radeon_device *rdev) #endif } -void rs400_mc_program(struct radeon_device *rdev) +static void rs400_mc_program(struct radeon_device *rdev) { struct r100_mc_save save; diff --git a/drivers/gpu/drm/radeon/rs600.c b/drivers/gpu/drm/radeon/rs600.c index 5301b3df8466..dc8d021a999b 100644 --- a/drivers/gpu/drm/radeon/rs600.c +++ b/drivers/gpu/drm/radeon/rs600.c @@ -43,22 +43,30 @@ #include "rs600_reg_safe.h" -void rs600_gpu_init(struct radeon_device *rdev); +static void rs600_gpu_init(struct radeon_device *rdev); int rs600_mc_wait_for_idle(struct radeon_device *rdev); +static const u32 crtc_offsets[2] = +{ + 0, + AVIVO_D2CRTC_H_TOTAL - AVIVO_D1CRTC_H_TOTAL +}; + void avivo_wait_for_vblank(struct radeon_device *rdev, int crtc) { - struct radeon_crtc *radeon_crtc = rdev->mode_info.crtcs[crtc]; int i; - if (RREG32(AVIVO_D1CRTC_CONTROL + radeon_crtc->crtc_offset) & AVIVO_CRTC_EN) { + if (crtc >= rdev->num_crtc) + return; + + if (RREG32(AVIVO_D1CRTC_CONTROL + crtc_offsets[crtc]) & AVIVO_CRTC_EN) { for (i = 0; i < rdev->usec_timeout; i++) { - if (!(RREG32(AVIVO_D1CRTC_STATUS + radeon_crtc->crtc_offset) & AVIVO_D1CRTC_V_BLANK)) + if (!(RREG32(AVIVO_D1CRTC_STATUS + crtc_offsets[crtc]) & AVIVO_D1CRTC_V_BLANK)) break; udelay(1); } for (i = 0; i < rdev->usec_timeout; i++) { - if (RREG32(AVIVO_D1CRTC_STATUS + radeon_crtc->crtc_offset) & AVIVO_D1CRTC_V_BLANK) + if (RREG32(AVIVO_D1CRTC_STATUS + crtc_offsets[crtc]) & AVIVO_D1CRTC_V_BLANK) break; udelay(1); } @@ -424,7 +432,7 @@ void rs600_gart_tlb_flush(struct radeon_device *rdev) tmp = RREG32_MC(R_000100_MC_PT0_CNTL); } -int rs600_gart_init(struct radeon_device *rdev) +static int rs600_gart_init(struct radeon_device *rdev) { int r; @@ -506,7 +514,7 @@ static int rs600_gart_enable(struct radeon_device *rdev) return 0; } -void rs600_gart_disable(struct radeon_device *rdev) +static void rs600_gart_disable(struct radeon_device *rdev) { u32 tmp; @@ -517,7 +525,7 @@ void rs600_gart_disable(struct radeon_device *rdev) radeon_gart_table_vram_unpin(rdev); } -void rs600_gart_fini(struct radeon_device *rdev) +static void rs600_gart_fini(struct radeon_device *rdev) { radeon_gart_fini(rdev); rs600_gart_disable(rdev); @@ -567,9 +575,6 @@ int rs600_irq_set(struct radeon_device *rdev) if (atomic_read(&rdev->irq.ring_int[RADEON_RING_TYPE_GFX_INDEX])) { tmp |= S_000040_SW_INT_EN(1); } - if (rdev->irq.gui_idle) { - tmp |= S_000040_GUI_IDLE(1); - } if (rdev->irq.crtc_vblank_int[0] || atomic_read(&rdev->irq.pflip[0])) { mode_int |= S_006540_D1MODE_VBLANK_INT_MASK(1); @@ -602,12 +607,6 @@ static inline u32 rs600_irq_ack(struct radeon_device *rdev) uint32_t irq_mask = S_000044_SW_INT(1); u32 tmp; - /* the interrupt works, but the status bit is permanently asserted */ - if (rdev->irq.gui_idle && radeon_gui_idle(rdev)) { - if (!rdev->irq.gui_idle_acked) - irq_mask |= S_000044_GUI_IDLE_STAT(1); - } - if (G_000044_DISPLAY_INT_STAT(irqs)) { rdev->irq.stat_regs.r500.disp_int = RREG32(R_007EDC_DISP_INTERRUPT_STATUS); if (G_007EDC_LB_D1_VBLANK_INTERRUPT(rdev->irq.stat_regs.r500.disp_int)) { @@ -667,9 +666,6 @@ int rs600_irq_process(struct radeon_device *rdev) bool queue_hotplug = false; bool queue_hdmi = false; - /* reset gui idle ack. the status bit is broken */ - rdev->irq.gui_idle_acked = false; - status = rs600_irq_ack(rdev); if (!status && !rdev->irq.stat_regs.r500.disp_int && @@ -683,11 +679,6 @@ int rs600_irq_process(struct radeon_device *rdev) if (G_000044_SW_INT(status)) { radeon_fence_process(rdev, RADEON_RING_TYPE_GFX_INDEX); } - /* GUI idle */ - if (G_000040_GUI_IDLE(status)) { - rdev->irq.gui_idle_acked = true; - wake_up(&rdev->irq.idle_queue); - } /* Vertical blank interrupts */ if (G_007EDC_LB_D1_VBLANK_INTERRUPT(rdev->irq.stat_regs.r500.disp_int)) { if (rdev->irq.crtc_vblank_int[0]) { @@ -721,8 +712,6 @@ int rs600_irq_process(struct radeon_device *rdev) } status = rs600_irq_ack(rdev); } - /* reset gui idle ack. the status bit is broken */ - rdev->irq.gui_idle_acked = false; if (queue_hotplug) schedule_work(&rdev->hotplug_work); if (queue_hdmi) @@ -764,7 +753,7 @@ int rs600_mc_wait_for_idle(struct radeon_device *rdev) return -1; } -void rs600_gpu_init(struct radeon_device *rdev) +static void rs600_gpu_init(struct radeon_device *rdev) { r420_pipes_init(rdev); /* Wait for mc idle */ @@ -772,7 +761,7 @@ void rs600_gpu_init(struct radeon_device *rdev) dev_warn(rdev->dev, "Wait MC idle timeout before updating MC.\n"); } -void rs600_mc_init(struct radeon_device *rdev) +static void rs600_mc_init(struct radeon_device *rdev) { u64 base; @@ -834,7 +823,7 @@ void rs600_mc_wreg(struct radeon_device *rdev, uint32_t reg, uint32_t v) WREG32(R_000074_MC_IND_DATA, v); } -void rs600_debugfs(struct radeon_device *rdev) +static void rs600_debugfs(struct radeon_device *rdev) { if (r100_debugfs_rbbm_init(rdev)) DRM_ERROR("Failed to register debugfs file for RBBM !\n"); diff --git a/drivers/gpu/drm/radeon/rs690.c b/drivers/gpu/drm/radeon/rs690.c index 3b663fcfe061..5cd5aceb69fa 100644 --- a/drivers/gpu/drm/radeon/rs690.c +++ b/drivers/gpu/drm/radeon/rs690.c @@ -145,7 +145,7 @@ void rs690_pm_info(struct radeon_device *rdev) rdev->pm.sideport_bandwidth.full = dfixed_div(rdev->pm.sideport_bandwidth, tmp); } -void rs690_mc_init(struct radeon_device *rdev) +static void rs690_mc_init(struct radeon_device *rdev) { u64 base; @@ -224,7 +224,7 @@ struct rs690_watermark { fixed20_12 sclk; }; -void rs690_crtc_bandwidth_compute(struct radeon_device *rdev, +static void rs690_crtc_bandwidth_compute(struct radeon_device *rdev, struct radeon_crtc *crtc, struct rs690_watermark *wm) { @@ -581,7 +581,7 @@ void rs690_mc_wreg(struct radeon_device *rdev, uint32_t reg, uint32_t v) WREG32(R_000078_MC_INDEX, 0x7F); } -void rs690_mc_program(struct radeon_device *rdev) +static void rs690_mc_program(struct radeon_device *rdev) { struct rv515_mc_save save; diff --git a/drivers/gpu/drm/radeon/rv515.c b/drivers/gpu/drm/radeon/rv515.c index aa8ef491ef3c..2d75d30be5b4 100644 --- a/drivers/gpu/drm/radeon/rv515.c +++ b/drivers/gpu/drm/radeon/rv515.c @@ -35,9 +35,9 @@ #include "rv515_reg_safe.h" /* This files gather functions specifics to: rv515 */ -int rv515_debugfs_pipes_info_init(struct radeon_device *rdev); -int rv515_debugfs_ga_info_init(struct radeon_device *rdev); -void rv515_gpu_init(struct radeon_device *rdev); +static int rv515_debugfs_pipes_info_init(struct radeon_device *rdev); +static int rv515_debugfs_ga_info_init(struct radeon_device *rdev); +static void rv515_gpu_init(struct radeon_device *rdev); int rv515_mc_wait_for_idle(struct radeon_device *rdev); void rv515_debugfs(struct radeon_device *rdev) @@ -143,7 +143,7 @@ void rv515_vga_render_disable(struct radeon_device *rdev) RREG32(R_000300_VGA_RENDER_CONTROL) & C_000300_VGA_VSTATUS_CNTL); } -void rv515_gpu_init(struct radeon_device *rdev) +static void rv515_gpu_init(struct radeon_device *rdev) { unsigned pipe_select_current, gb_pipe_select, tmp; @@ -189,7 +189,7 @@ static void rv515_vram_get_type(struct radeon_device *rdev) } } -void rv515_mc_init(struct radeon_device *rdev) +static void rv515_mc_init(struct radeon_device *rdev) { rv515_vram_get_type(rdev); @@ -261,7 +261,7 @@ static struct drm_info_list rv515_ga_info_list[] = { }; #endif -int rv515_debugfs_pipes_info_init(struct radeon_device *rdev) +static int rv515_debugfs_pipes_info_init(struct radeon_device *rdev) { #if defined(CONFIG_DEBUG_FS) return radeon_debugfs_add_files(rdev, rv515_pipes_info_list, 1); @@ -270,7 +270,7 @@ int rv515_debugfs_pipes_info_init(struct radeon_device *rdev) #endif } -int rv515_debugfs_ga_info_init(struct radeon_device *rdev) +static int rv515_debugfs_ga_info_init(struct radeon_device *rdev) { #if defined(CONFIG_DEBUG_FS) return radeon_debugfs_add_files(rdev, rv515_ga_info_list, 1); @@ -310,7 +310,7 @@ void rv515_mc_resume(struct radeon_device *rdev, struct rv515_mc_save *save) WREG32(R_000300_VGA_RENDER_CONTROL, save->vga_render_control); } -void rv515_mc_program(struct radeon_device *rdev) +static void rv515_mc_program(struct radeon_device *rdev) { struct rv515_mc_save save; @@ -787,7 +787,7 @@ struct rv515_watermark { fixed20_12 sclk; }; -void rv515_crtc_bandwidth_compute(struct radeon_device *rdev, +static void rv515_crtc_bandwidth_compute(struct radeon_device *rdev, struct radeon_crtc *crtc, struct rv515_watermark *wm) { diff --git a/drivers/gpu/drm/radeon/rv770.c b/drivers/gpu/drm/radeon/rv770.c index ca8ffec10ff6..2469afe11b85 100644 --- a/drivers/gpu/drm/radeon/rv770.c +++ b/drivers/gpu/drm/radeon/rv770.c @@ -124,7 +124,7 @@ void rv770_pm_misc(struct radeon_device *rdev) /* * GART */ -int rv770_pcie_gart_enable(struct radeon_device *rdev) +static int rv770_pcie_gart_enable(struct radeon_device *rdev) { u32 tmp; int r, i; @@ -175,7 +175,7 @@ int rv770_pcie_gart_enable(struct radeon_device *rdev) return 0; } -void rv770_pcie_gart_disable(struct radeon_device *rdev) +static void rv770_pcie_gart_disable(struct radeon_device *rdev) { u32 tmp; int i; @@ -201,7 +201,7 @@ void rv770_pcie_gart_disable(struct radeon_device *rdev) radeon_gart_table_vram_unpin(rdev); } -void rv770_pcie_gart_fini(struct radeon_device *rdev) +static void rv770_pcie_gart_fini(struct radeon_device *rdev) { radeon_gart_fini(rdev); rv770_pcie_gart_disable(rdev); @@ -209,7 +209,7 @@ void rv770_pcie_gart_fini(struct radeon_device *rdev) } -void rv770_agp_enable(struct radeon_device *rdev) +static void rv770_agp_enable(struct radeon_device *rdev) { u32 tmp; int i; @@ -839,7 +839,7 @@ void r700_vram_gtt_location(struct radeon_device *rdev, struct radeon_mc *mc) } } -int rv770_mc_init(struct radeon_device *rdev) +static int rv770_mc_init(struct radeon_device *rdev) { u32 tmp; int chansize, numchan; diff --git a/drivers/gpu/drm/radeon/si.c b/drivers/gpu/drm/radeon/si.c index 0139e227e3c7..c76825ffa37f 100644 --- a/drivers/gpu/drm/radeon/si.c +++ b/drivers/gpu/drm/radeon/si.c @@ -1806,13 +1806,14 @@ void si_ring_ib_execute(struct radeon_device *rdev, struct radeon_ib *ib) #endif (ib->gpu_addr & 0xFFFFFFFC)); radeon_ring_write(ring, upper_32_bits(ib->gpu_addr) & 0xFFFF); - radeon_ring_write(ring, ib->length_dw | (ib->vm_id << 24)); + radeon_ring_write(ring, ib->length_dw | + (ib->vm ? (ib->vm->id << 24) : 0)); if (!ib->is_const_ib) { /* flush read cache over gart for this vmid */ radeon_ring_write(ring, PACKET3(PACKET3_SET_CONFIG_REG, 1)); radeon_ring_write(ring, (CP_COHER_CNTL2 - PACKET3_SET_CONFIG_REG_START) >> 2); - radeon_ring_write(ring, ib->vm_id); + radeon_ring_write(ring, ib->vm ? ib->vm->id : 0); radeon_ring_write(ring, PACKET3(PACKET3_SURFACE_SYNC, 3)); radeon_ring_write(ring, PACKET3_TCL1_ACTION_ENA | PACKET3_TC_ACTION_ENA | @@ -2363,7 +2364,7 @@ void si_pcie_gart_tlb_flush(struct radeon_device *rdev) WREG32(VM_INVALIDATE_REQUEST, 1); } -int si_pcie_gart_enable(struct radeon_device *rdev) +static int si_pcie_gart_enable(struct radeon_device *rdev) { int r, i; @@ -2425,7 +2426,7 @@ int si_pcie_gart_enable(struct radeon_device *rdev) WREG32(VM_CONTEXT1_PROTECTION_FAULT_DEFAULT_ADDR, (u32)(rdev->dummy_page.addr >> 12)); WREG32(VM_CONTEXT1_CNTL2, 0); - WREG32(VM_CONTEXT1_CNTL, ENABLE_CONTEXT | PAGE_TABLE_DEPTH(0) | + WREG32(VM_CONTEXT1_CNTL, ENABLE_CONTEXT | PAGE_TABLE_DEPTH(1) | RANGE_PROTECTION_FAULT_ENABLE_DEFAULT); si_pcie_gart_tlb_flush(rdev); @@ -2436,7 +2437,7 @@ int si_pcie_gart_enable(struct radeon_device *rdev) return 0; } -void si_pcie_gart_disable(struct radeon_device *rdev) +static void si_pcie_gart_disable(struct radeon_device *rdev) { /* Disable all tables */ WREG32(VM_CONTEXT0_CNTL, 0); @@ -2455,7 +2456,7 @@ void si_pcie_gart_disable(struct radeon_device *rdev) radeon_gart_table_vram_unpin(rdev); } -void si_pcie_gart_fini(struct radeon_device *rdev) +static void si_pcie_gart_fini(struct radeon_device *rdev) { si_pcie_gart_disable(rdev); radeon_gart_table_vram_free(rdev); @@ -2788,41 +2789,84 @@ void si_vm_fini(struct radeon_device *rdev) { } -int si_vm_bind(struct radeon_device *rdev, struct radeon_vm *vm, int id) +/** + * si_vm_set_page - update the page tables using the CP + * + * @rdev: radeon_device pointer + * @pe: addr of the page entry + * @addr: dst addr to write into pe + * @count: number of page entries to update + * @incr: increase next addr by incr bytes + * @flags: access flags + * + * Update the page tables using the CP (cayman-si). + */ +void si_vm_set_page(struct radeon_device *rdev, uint64_t pe, + uint64_t addr, unsigned count, + uint32_t incr, uint32_t flags) { - if (id < 8) - WREG32(VM_CONTEXT0_PAGE_TABLE_BASE_ADDR + (id << 2), vm->pt_gpu_addr >> 12); - else - WREG32(VM_CONTEXT8_PAGE_TABLE_BASE_ADDR + ((id - 8) << 2), - vm->pt_gpu_addr >> 12); - /* flush hdp cache */ - WREG32(HDP_MEM_COHERENCY_FLUSH_CNTL, 0x1); - /* bits 0-15 are the VM contexts0-15 */ - WREG32(VM_INVALIDATE_REQUEST, 1 << id); - return 0; + struct radeon_ring *ring = &rdev->ring[rdev->asic->vm.pt_ring_index]; + uint32_t r600_flags = cayman_vm_page_flags(rdev, flags); + int i; + uint64_t value; + + radeon_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 2 + count * 2)); + radeon_ring_write(ring, (WRITE_DATA_ENGINE_SEL(0) | + WRITE_DATA_DST_SEL(1))); + radeon_ring_write(ring, pe); + radeon_ring_write(ring, upper_32_bits(pe)); + for (i = 0; i < count; ++i) { + if (flags & RADEON_VM_PAGE_SYSTEM) { + value = radeon_vm_map_gart(rdev, addr); + value &= 0xFFFFFFFFFFFFF000ULL; + } else if (flags & RADEON_VM_PAGE_VALID) + value = addr; + else + value = 0; + addr += incr; + value |= r600_flags; + radeon_ring_write(ring, value); + radeon_ring_write(ring, upper_32_bits(value)); + } } -void si_vm_unbind(struct radeon_device *rdev, struct radeon_vm *vm) +void si_vm_flush(struct radeon_device *rdev, int ridx, struct radeon_vm *vm) { - if (vm->id < 8) - WREG32(VM_CONTEXT0_PAGE_TABLE_BASE_ADDR + (vm->id << 2), 0); - else - WREG32(VM_CONTEXT8_PAGE_TABLE_BASE_ADDR + ((vm->id - 8) << 2), 0); - /* flush hdp cache */ - WREG32(HDP_MEM_COHERENCY_FLUSH_CNTL, 0x1); - /* bits 0-15 are the VM contexts0-15 */ - WREG32(VM_INVALIDATE_REQUEST, 1 << vm->id); -} + struct radeon_ring *ring = &rdev->ring[ridx]; -void si_vm_tlb_flush(struct radeon_device *rdev, struct radeon_vm *vm) -{ - if (vm->id == -1) + if (vm == NULL) return; + /* write new base address */ + radeon_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3)); + radeon_ring_write(ring, (WRITE_DATA_ENGINE_SEL(0) | + WRITE_DATA_DST_SEL(0))); + + if (vm->id < 8) { + radeon_ring_write(ring, + (VM_CONTEXT0_PAGE_TABLE_BASE_ADDR + (vm->id << 2)) >> 2); + } else { + radeon_ring_write(ring, + (VM_CONTEXT8_PAGE_TABLE_BASE_ADDR + ((vm->id - 8) << 2)) >> 2); + } + radeon_ring_write(ring, 0); + radeon_ring_write(ring, vm->pd_gpu_addr >> 12); + /* flush hdp cache */ - WREG32(HDP_MEM_COHERENCY_FLUSH_CNTL, 0x1); + radeon_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3)); + radeon_ring_write(ring, (WRITE_DATA_ENGINE_SEL(0) | + WRITE_DATA_DST_SEL(0))); + radeon_ring_write(ring, HDP_MEM_COHERENCY_FLUSH_CNTL >> 2); + radeon_ring_write(ring, 0); + radeon_ring_write(ring, 0x1); + /* bits 0-15 are the VM contexts0-15 */ - WREG32(VM_INVALIDATE_REQUEST, 1 << vm->id); + radeon_ring_write(ring, PACKET3(PACKET3_WRITE_DATA, 3)); + radeon_ring_write(ring, (WRITE_DATA_ENGINE_SEL(0) | + WRITE_DATA_DST_SEL(0))); + radeon_ring_write(ring, VM_INVALIDATE_REQUEST >> 2); + radeon_ring_write(ring, 0); + radeon_ring_write(ring, 1 << vm->id); } /* @@ -3199,10 +3243,6 @@ int si_irq_set(struct radeon_device *rdev) DRM_DEBUG("si_irq_set: hpd 6\n"); hpd6 |= DC_HPDx_INT_EN; } - if (rdev->irq.gui_idle) { - DRM_DEBUG("gui idle\n"); - grbm_int_cntl |= GUI_IDLE_INT_ENABLE; - } WREG32(CP_INT_CNTL_RING0, cp_int_cntl); WREG32(CP_INT_CNTL_RING1, cp_int_cntl1); @@ -3658,7 +3698,6 @@ restart_ih: break; case 233: /* GUI IDLE */ DRM_DEBUG("IH: GUI idle\n"); - wake_up(&rdev->irq.idle_queue); break; default: DRM_DEBUG("Unhandled interrupt: %d %d\n", src_id, src_data); diff --git a/drivers/gpu/drm/radeon/sid.h b/drivers/gpu/drm/radeon/sid.h index ef4815c27b1c..7d2a20e56577 100644 --- a/drivers/gpu/drm/radeon/sid.h +++ b/drivers/gpu/drm/radeon/sid.h @@ -812,6 +812,21 @@ #define PACKET3_DRAW_INDEX_OFFSET_2 0x35 #define PACKET3_DRAW_INDEX_MULTI_ELEMENT 0x36 #define PACKET3_WRITE_DATA 0x37 +#define WRITE_DATA_DST_SEL(x) ((x) << 8) + /* 0 - register + * 1 - memory (sync - via GRBM) + * 2 - tc/l2 + * 3 - gds + * 4 - reserved + * 5 - memory (async - direct) + */ +#define WR_ONE_ADDR (1 << 16) +#define WR_CONFIRM (1 << 20) +#define WRITE_DATA_ENGINE_SEL(x) ((x) << 30) + /* 0 - me + * 1 - pfp + * 2 - ce + */ #define PACKET3_DRAW_INDEX_INDIRECT_MULTI 0x38 #define PACKET3_MEM_SEMAPHORE 0x39 #define PACKET3_MPEG_INDEX 0x3A diff --git a/drivers/gpu/drm/savage/savage_bci.c b/drivers/gpu/drm/savage/savage_bci.c index 1efbb9075837..c89aef420971 100644 --- a/drivers/gpu/drm/savage/savage_bci.c +++ b/drivers/gpu/drm/savage/savage_bci.c @@ -547,6 +547,8 @@ int savage_driver_load(struct drm_device *dev, unsigned long chipset) dev_priv->chipset = (enum savage_family)chipset; + pci_set_master(dev->pdev); + return 0; } diff --git a/drivers/gpu/drm/shmobile/Kconfig b/drivers/gpu/drm/shmobile/Kconfig new file mode 100644 index 000000000000..7e7d52b2a2fc --- /dev/null +++ b/drivers/gpu/drm/shmobile/Kconfig @@ -0,0 +1,10 @@ +config DRM_SHMOBILE + tristate "DRM Support for SH Mobile" + depends on DRM && (SUPERH || ARCH_SHMOBILE) + select DRM_KMS_HELPER + select DRM_KMS_CMA_HELPER + select DRM_GEM_CMA_HELPER + help + Choose this option if you have an SH Mobile chipset. + If M is selected the module will be called shmob-drm. + diff --git a/drivers/gpu/drm/shmobile/Makefile b/drivers/gpu/drm/shmobile/Makefile new file mode 100644 index 000000000000..4c3eeb355630 --- /dev/null +++ b/drivers/gpu/drm/shmobile/Makefile @@ -0,0 +1,7 @@ +shmob-drm-y := shmob_drm_backlight.o \ + shmob_drm_crtc.o \ + shmob_drm_drv.o \ + shmob_drm_kms.o \ + shmob_drm_plane.o + +obj-$(CONFIG_DRM_SHMOBILE) += shmob-drm.o diff --git a/drivers/gpu/drm/shmobile/shmob_drm_backlight.c b/drivers/gpu/drm/shmobile/shmob_drm_backlight.c new file mode 100644 index 000000000000..463aee18f774 --- /dev/null +++ b/drivers/gpu/drm/shmobile/shmob_drm_backlight.c @@ -0,0 +1,90 @@ +/* + * shmob_drm_backlight.c -- SH Mobile DRM Backlight + * + * Copyright (C) 2012 Renesas Corporation + * + * Laurent Pinchart (laurent.pinchart@ideasonboard.com) + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + */ + +#include <linux/backlight.h> + +#include "shmob_drm_backlight.h" +#include "shmob_drm_crtc.h" +#include "shmob_drm_drv.h" + +static int shmob_drm_backlight_update(struct backlight_device *bdev) +{ + struct shmob_drm_connector *scon = bl_get_data(bdev); + struct shmob_drm_device *sdev = scon->connector.dev->dev_private; + const struct shmob_drm_backlight_data *bdata = &sdev->pdata->backlight; + int brightness = bdev->props.brightness; + + if (bdev->props.power != FB_BLANK_UNBLANK || + bdev->props.state & BL_CORE_SUSPENDED) + brightness = 0; + + return bdata->set_brightness(brightness); +} + +static int shmob_drm_backlight_get_brightness(struct backlight_device *bdev) +{ + struct shmob_drm_connector *scon = bl_get_data(bdev); + struct shmob_drm_device *sdev = scon->connector.dev->dev_private; + const struct shmob_drm_backlight_data *bdata = &sdev->pdata->backlight; + + return bdata->get_brightness(); +} + +static const struct backlight_ops shmob_drm_backlight_ops = { + .options = BL_CORE_SUSPENDRESUME, + .update_status = shmob_drm_backlight_update, + .get_brightness = shmob_drm_backlight_get_brightness, +}; + +void shmob_drm_backlight_dpms(struct shmob_drm_connector *scon, int mode) +{ + if (scon->backlight == NULL) + return; + + scon->backlight->props.power = mode == DRM_MODE_DPMS_ON + ? FB_BLANK_UNBLANK : FB_BLANK_POWERDOWN; + backlight_update_status(scon->backlight); +} + +int shmob_drm_backlight_init(struct shmob_drm_connector *scon) +{ + struct shmob_drm_device *sdev = scon->connector.dev->dev_private; + const struct shmob_drm_backlight_data *bdata = &sdev->pdata->backlight; + struct drm_connector *connector = &scon->connector; + struct drm_device *dev = connector->dev; + struct backlight_device *backlight; + + if (!bdata->max_brightness) + return 0; + + backlight = backlight_device_register(bdata->name, dev->dev, scon, + &shmob_drm_backlight_ops, NULL); + if (IS_ERR(backlight)) { + dev_err(dev->dev, "unable to register backlight device: %ld\n", + PTR_ERR(backlight)); + return PTR_ERR(backlight); + } + + backlight->props.max_brightness = bdata->max_brightness; + backlight->props.brightness = bdata->max_brightness; + backlight->props.power = FB_BLANK_POWERDOWN; + backlight_update_status(backlight); + + scon->backlight = backlight; + return 0; +} + +void shmob_drm_backlight_exit(struct shmob_drm_connector *scon) +{ + backlight_device_unregister(scon->backlight); +} diff --git a/drivers/gpu/drm/shmobile/shmob_drm_backlight.h b/drivers/gpu/drm/shmobile/shmob_drm_backlight.h new file mode 100644 index 000000000000..9477595d2ff3 --- /dev/null +++ b/drivers/gpu/drm/shmobile/shmob_drm_backlight.h @@ -0,0 +1,23 @@ +/* + * shmob_drm_backlight.h -- SH Mobile DRM Backlight + * + * Copyright (C) 2012 Renesas Corporation + * + * Laurent Pinchart (laurent.pinchart@ideasonboard.com) + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + */ + +#ifndef __SHMOB_DRM_BACKLIGHT_H__ +#define __SHMOB_DRM_BACKLIGHT_H__ + +struct shmob_drm_connector; + +void shmob_drm_backlight_dpms(struct shmob_drm_connector *scon, int mode); +int shmob_drm_backlight_init(struct shmob_drm_connector *scon); +void shmob_drm_backlight_exit(struct shmob_drm_connector *scon); + +#endif /* __SHMOB_DRM_BACKLIGHT_H__ */ diff --git a/drivers/gpu/drm/shmobile/shmob_drm_crtc.c b/drivers/gpu/drm/shmobile/shmob_drm_crtc.c new file mode 100644 index 000000000000..0e7a9306bd0c --- /dev/null +++ b/drivers/gpu/drm/shmobile/shmob_drm_crtc.c @@ -0,0 +1,763 @@ +/* + * shmob_drm_crtc.c -- SH Mobile DRM CRTCs + * + * Copyright (C) 2012 Renesas Corporation + * + * Laurent Pinchart (laurent.pinchart@ideasonboard.com) + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + */ + +#include <linux/backlight.h> +#include <linux/clk.h> + +#include <drm/drmP.h> +#include <drm/drm_crtc.h> +#include <drm/drm_crtc_helper.h> +#include <drm/drm_fb_cma_helper.h> +#include <drm/drm_gem_cma_helper.h> + +#include <video/sh_mobile_meram.h> + +#include "shmob_drm_backlight.h" +#include "shmob_drm_crtc.h" +#include "shmob_drm_drv.h" +#include "shmob_drm_kms.h" +#include "shmob_drm_plane.h" +#include "shmob_drm_regs.h" + +/* + * TODO: panel support + */ + +/* ----------------------------------------------------------------------------- + * Clock management + */ + +static void shmob_drm_clk_on(struct shmob_drm_device *sdev) +{ + if (sdev->clock) + clk_enable(sdev->clock); +#if 0 + if (sdev->meram_dev && sdev->meram_dev->pdev) + pm_runtime_get_sync(&sdev->meram_dev->pdev->dev); +#endif +} + +static void shmob_drm_clk_off(struct shmob_drm_device *sdev) +{ +#if 0 + if (sdev->meram_dev && sdev->meram_dev->pdev) + pm_runtime_put_sync(&sdev->meram_dev->pdev->dev); +#endif + if (sdev->clock) + clk_disable(sdev->clock); +} + +/* ----------------------------------------------------------------------------- + * CRTC + */ + +static void shmob_drm_crtc_setup_geometry(struct shmob_drm_crtc *scrtc) +{ + struct drm_crtc *crtc = &scrtc->crtc; + struct shmob_drm_device *sdev = crtc->dev->dev_private; + const struct shmob_drm_interface_data *idata = &sdev->pdata->iface; + const struct drm_display_mode *mode = &crtc->mode; + u32 value; + + value = sdev->ldmt1r + | ((mode->flags & DRM_MODE_FLAG_PVSYNC) ? 0 : LDMT1R_VPOL) + | ((mode->flags & DRM_MODE_FLAG_PHSYNC) ? 0 : LDMT1R_HPOL) + | ((idata->flags & SHMOB_DRM_IFACE_FL_DWPOL) ? LDMT1R_DWPOL : 0) + | ((idata->flags & SHMOB_DRM_IFACE_FL_DIPOL) ? LDMT1R_DIPOL : 0) + | ((idata->flags & SHMOB_DRM_IFACE_FL_DAPOL) ? LDMT1R_DAPOL : 0) + | ((idata->flags & SHMOB_DRM_IFACE_FL_HSCNT) ? LDMT1R_HSCNT : 0) + | ((idata->flags & SHMOB_DRM_IFACE_FL_DWCNT) ? LDMT1R_DWCNT : 0); + lcdc_write(sdev, LDMT1R, value); + + if (idata->interface >= SHMOB_DRM_IFACE_SYS8A && + idata->interface <= SHMOB_DRM_IFACE_SYS24) { + /* Setup SYS bus. */ + value = (idata->sys.cs_setup << LDMT2R_CSUP_SHIFT) + | (idata->sys.vsync_active_high ? LDMT2R_RSV : 0) + | (idata->sys.vsync_dir_input ? LDMT2R_VSEL : 0) + | (idata->sys.write_setup << LDMT2R_WCSC_SHIFT) + | (idata->sys.write_cycle << LDMT2R_WCEC_SHIFT) + | (idata->sys.write_strobe << LDMT2R_WCLW_SHIFT); + lcdc_write(sdev, LDMT2R, value); + + value = (idata->sys.read_latch << LDMT3R_RDLC_SHIFT) + | (idata->sys.read_setup << LDMT3R_RCSC_SHIFT) + | (idata->sys.read_cycle << LDMT3R_RCEC_SHIFT) + | (idata->sys.read_strobe << LDMT3R_RCLW_SHIFT); + lcdc_write(sdev, LDMT3R, value); + } + + value = ((mode->hdisplay / 8) << 16) /* HDCN */ + | (mode->htotal / 8); /* HTCN */ + lcdc_write(sdev, LDHCNR, value); + + value = (((mode->hsync_end - mode->hsync_start) / 8) << 16) /* HSYNW */ + | (mode->hsync_start / 8); /* HSYNP */ + lcdc_write(sdev, LDHSYNR, value); + + value = ((mode->hdisplay & 7) << 24) | ((mode->htotal & 7) << 16) + | (((mode->hsync_end - mode->hsync_start) & 7) << 8) + | (mode->hsync_start & 7); + lcdc_write(sdev, LDHAJR, value); + + value = ((mode->vdisplay) << 16) /* VDLN */ + | mode->vtotal; /* VTLN */ + lcdc_write(sdev, LDVLNR, value); + + value = ((mode->vsync_end - mode->vsync_start) << 16) /* VSYNW */ + | mode->vsync_start; /* VSYNP */ + lcdc_write(sdev, LDVSYNR, value); +} + +static void shmob_drm_crtc_start_stop(struct shmob_drm_crtc *scrtc, bool start) +{ + struct shmob_drm_device *sdev = scrtc->crtc.dev->dev_private; + u32 value; + + value = lcdc_read(sdev, LDCNT2R); + if (start) + lcdc_write(sdev, LDCNT2R, value | LDCNT2R_DO); + else + lcdc_write(sdev, LDCNT2R, value & ~LDCNT2R_DO); + + /* Wait until power is applied/stopped. */ + while (1) { + value = lcdc_read(sdev, LDPMR) & LDPMR_LPS; + if ((start && value) || (!start && !value)) + break; + + cpu_relax(); + } + + if (!start) { + /* Stop the dot clock. */ + lcdc_write(sdev, LDDCKSTPR, LDDCKSTPR_DCKSTP); + } +} + +/* + * shmob_drm_crtc_start - Configure and start the LCDC + * @scrtc: the SH Mobile CRTC + * + * Configure and start the LCDC device. External devices (clocks, MERAM, panels, + * ...) are not touched by this function. + */ +static void shmob_drm_crtc_start(struct shmob_drm_crtc *scrtc) +{ + struct drm_crtc *crtc = &scrtc->crtc; + struct shmob_drm_device *sdev = crtc->dev->dev_private; + const struct shmob_drm_interface_data *idata = &sdev->pdata->iface; + const struct shmob_drm_format_info *format; + struct drm_device *dev = sdev->ddev; + struct drm_plane *plane; + u32 value; + + if (scrtc->started) + return; + + format = shmob_drm_format_info(crtc->fb->pixel_format); + if (WARN_ON(format == NULL)) + return; + + /* Enable clocks before accessing the hardware. */ + shmob_drm_clk_on(sdev); + + /* Reset and enable the LCDC. */ + lcdc_write(sdev, LDCNT2R, lcdc_read(sdev, LDCNT2R) | LDCNT2R_BR); + lcdc_wait_bit(sdev, LDCNT2R, LDCNT2R_BR, 0); + lcdc_write(sdev, LDCNT2R, LDCNT2R_ME); + + /* Stop the LCDC first and disable all interrupts. */ + shmob_drm_crtc_start_stop(scrtc, false); + lcdc_write(sdev, LDINTR, 0); + + /* Configure power supply, dot clocks and start them. */ + lcdc_write(sdev, LDPMR, 0); + + value = sdev->lddckr; + if (idata->clk_div) { + /* FIXME: sh7724 can only use 42, 48, 54 and 60 for the divider + * denominator. + */ + lcdc_write(sdev, LDDCKPAT1R, 0); + lcdc_write(sdev, LDDCKPAT2R, (1 << (idata->clk_div / 2)) - 1); + + if (idata->clk_div == 1) + value |= LDDCKR_MOSEL; + else + value |= idata->clk_div; + } + + lcdc_write(sdev, LDDCKR, value); + lcdc_write(sdev, LDDCKSTPR, 0); + lcdc_wait_bit(sdev, LDDCKSTPR, ~0, 0); + + /* TODO: Setup SYS panel */ + + /* Setup geometry, format, frame buffer memory and operation mode. */ + shmob_drm_crtc_setup_geometry(scrtc); + + /* TODO: Handle YUV colorspaces. Hardcode REC709 for now. */ + lcdc_write(sdev, LDDFR, format->lddfr | LDDFR_CF1); + lcdc_write(sdev, LDMLSR, scrtc->line_size); + lcdc_write(sdev, LDSA1R, scrtc->dma[0]); + if (format->yuv) + lcdc_write(sdev, LDSA2R, scrtc->dma[1]); + lcdc_write(sdev, LDSM1R, 0); + + /* Word and long word swap. */ + switch (format->fourcc) { + case DRM_FORMAT_RGB565: + case DRM_FORMAT_NV21: + case DRM_FORMAT_NV61: + case DRM_FORMAT_NV42: + value = LDDDSR_LS | LDDDSR_WS; + break; + case DRM_FORMAT_RGB888: + case DRM_FORMAT_NV12: + case DRM_FORMAT_NV16: + case DRM_FORMAT_NV24: + value = LDDDSR_LS | LDDDSR_WS | LDDDSR_BS; + break; + case DRM_FORMAT_ARGB8888: + default: + value = LDDDSR_LS; + break; + } + lcdc_write(sdev, LDDDSR, value); + + /* Setup planes. */ + list_for_each_entry(plane, &dev->mode_config.plane_list, head) { + if (plane->crtc == crtc) + shmob_drm_plane_setup(plane); + } + + /* Enable the display output. */ + lcdc_write(sdev, LDCNT1R, LDCNT1R_DE); + + shmob_drm_crtc_start_stop(scrtc, true); + + scrtc->started = true; +} + +static void shmob_drm_crtc_stop(struct shmob_drm_crtc *scrtc) +{ + struct drm_crtc *crtc = &scrtc->crtc; + struct shmob_drm_device *sdev = crtc->dev->dev_private; + + if (!scrtc->started) + return; + + /* Disable the MERAM cache. */ + if (scrtc->cache) { + sh_mobile_meram_cache_free(sdev->meram, scrtc->cache); + scrtc->cache = NULL; + } + + /* Stop the LCDC. */ + shmob_drm_crtc_start_stop(scrtc, false); + + /* Disable the display output. */ + lcdc_write(sdev, LDCNT1R, 0); + + /* Stop clocks. */ + shmob_drm_clk_off(sdev); + + scrtc->started = false; +} + +void shmob_drm_crtc_suspend(struct shmob_drm_crtc *scrtc) +{ + shmob_drm_crtc_stop(scrtc); +} + +void shmob_drm_crtc_resume(struct shmob_drm_crtc *scrtc) +{ + if (scrtc->dpms != DRM_MODE_DPMS_ON) + return; + + shmob_drm_crtc_start(scrtc); +} + +static void shmob_drm_crtc_compute_base(struct shmob_drm_crtc *scrtc, + int x, int y) +{ + struct drm_crtc *crtc = &scrtc->crtc; + struct drm_framebuffer *fb = crtc->fb; + struct shmob_drm_device *sdev = crtc->dev->dev_private; + struct drm_gem_cma_object *gem; + unsigned int bpp; + + bpp = scrtc->format->yuv ? 8 : scrtc->format->bpp; + gem = drm_fb_cma_get_gem_obj(fb, 0); + scrtc->dma[0] = gem->paddr + fb->offsets[0] + + y * fb->pitches[0] + x * bpp / 8; + + if (scrtc->format->yuv) { + bpp = scrtc->format->bpp - 8; + gem = drm_fb_cma_get_gem_obj(fb, 1); + scrtc->dma[1] = gem->paddr + fb->offsets[1] + + y / (bpp == 4 ? 2 : 1) * fb->pitches[1] + + x * (bpp == 16 ? 2 : 1); + } + + if (scrtc->cache) + sh_mobile_meram_cache_update(sdev->meram, scrtc->cache, + scrtc->dma[0], scrtc->dma[1], + &scrtc->dma[0], &scrtc->dma[1]); +} + +static void shmob_drm_crtc_update_base(struct shmob_drm_crtc *scrtc) +{ + struct drm_crtc *crtc = &scrtc->crtc; + struct shmob_drm_device *sdev = crtc->dev->dev_private; + + shmob_drm_crtc_compute_base(scrtc, crtc->x, crtc->y); + + lcdc_write_mirror(sdev, LDSA1R, scrtc->dma[0]); + if (scrtc->format->yuv) + lcdc_write_mirror(sdev, LDSA2R, scrtc->dma[1]); + + lcdc_write(sdev, LDRCNTR, lcdc_read(sdev, LDRCNTR) ^ LDRCNTR_MRS); +} + +#define to_shmob_crtc(c) container_of(c, struct shmob_drm_crtc, crtc) + +static void shmob_drm_crtc_dpms(struct drm_crtc *crtc, int mode) +{ + struct shmob_drm_crtc *scrtc = to_shmob_crtc(crtc); + + if (scrtc->dpms == mode) + return; + + if (mode == DRM_MODE_DPMS_ON) + shmob_drm_crtc_start(scrtc); + else + shmob_drm_crtc_stop(scrtc); + + scrtc->dpms = mode; +} + +static bool shmob_drm_crtc_mode_fixup(struct drm_crtc *crtc, + const struct drm_display_mode *mode, + struct drm_display_mode *adjusted_mode) +{ + return true; +} + +static void shmob_drm_crtc_mode_prepare(struct drm_crtc *crtc) +{ + shmob_drm_crtc_dpms(crtc, DRM_MODE_DPMS_OFF); +} + +static int shmob_drm_crtc_mode_set(struct drm_crtc *crtc, + struct drm_display_mode *mode, + struct drm_display_mode *adjusted_mode, + int x, int y, + struct drm_framebuffer *old_fb) +{ + struct shmob_drm_crtc *scrtc = to_shmob_crtc(crtc); + struct shmob_drm_device *sdev = crtc->dev->dev_private; + const struct sh_mobile_meram_cfg *mdata = sdev->pdata->meram; + const struct shmob_drm_format_info *format; + void *cache; + + format = shmob_drm_format_info(crtc->fb->pixel_format); + if (format == NULL) { + dev_dbg(sdev->dev, "mode_set: unsupported format %08x\n", + crtc->fb->pixel_format); + return -EINVAL; + } + + scrtc->format = format; + scrtc->line_size = crtc->fb->pitches[0]; + + if (sdev->meram) { + /* Enable MERAM cache if configured. We need to de-init + * configured ICBs before we can re-initialize them. + */ + if (scrtc->cache) { + sh_mobile_meram_cache_free(sdev->meram, scrtc->cache); + scrtc->cache = NULL; + } + + cache = sh_mobile_meram_cache_alloc(sdev->meram, mdata, + crtc->fb->pitches[0], + adjusted_mode->vdisplay, + format->meram, + &scrtc->line_size); + if (!IS_ERR(cache)) + scrtc->cache = cache; + } + + shmob_drm_crtc_compute_base(scrtc, x, y); + + return 0; +} + +static void shmob_drm_crtc_mode_commit(struct drm_crtc *crtc) +{ + shmob_drm_crtc_dpms(crtc, DRM_MODE_DPMS_ON); +} + +static int shmob_drm_crtc_mode_set_base(struct drm_crtc *crtc, int x, int y, + struct drm_framebuffer *old_fb) +{ + shmob_drm_crtc_update_base(to_shmob_crtc(crtc)); + + return 0; +} + +static const struct drm_crtc_helper_funcs crtc_helper_funcs = { + .dpms = shmob_drm_crtc_dpms, + .mode_fixup = shmob_drm_crtc_mode_fixup, + .prepare = shmob_drm_crtc_mode_prepare, + .commit = shmob_drm_crtc_mode_commit, + .mode_set = shmob_drm_crtc_mode_set, + .mode_set_base = shmob_drm_crtc_mode_set_base, +}; + +void shmob_drm_crtc_cancel_page_flip(struct shmob_drm_crtc *scrtc, + struct drm_file *file) +{ + struct drm_pending_vblank_event *event; + struct drm_device *dev = scrtc->crtc.dev; + unsigned long flags; + + /* Destroy the pending vertical blanking event associated with the + * pending page flip, if any, and disable vertical blanking interrupts. + */ + spin_lock_irqsave(&dev->event_lock, flags); + event = scrtc->event; + if (event && event->base.file_priv == file) { + scrtc->event = NULL; + event->base.destroy(&event->base); + drm_vblank_put(dev, 0); + } + spin_unlock_irqrestore(&dev->event_lock, flags); +} + +void shmob_drm_crtc_finish_page_flip(struct shmob_drm_crtc *scrtc) +{ + struct drm_pending_vblank_event *event; + struct drm_device *dev = scrtc->crtc.dev; + struct timeval vblanktime; + unsigned long flags; + + spin_lock_irqsave(&dev->event_lock, flags); + event = scrtc->event; + scrtc->event = NULL; + spin_unlock_irqrestore(&dev->event_lock, flags); + + if (event == NULL) + return; + + event->event.sequence = drm_vblank_count_and_time(dev, 0, &vblanktime); + event->event.tv_sec = vblanktime.tv_sec; + event->event.tv_usec = vblanktime.tv_usec; + + spin_lock_irqsave(&dev->event_lock, flags); + list_add_tail(&event->base.link, &event->base.file_priv->event_list); + wake_up_interruptible(&event->base.file_priv->event_wait); + spin_unlock_irqrestore(&dev->event_lock, flags); + + drm_vblank_put(dev, 0); +} + +static int shmob_drm_crtc_page_flip(struct drm_crtc *crtc, + struct drm_framebuffer *fb, + struct drm_pending_vblank_event *event) +{ + struct shmob_drm_crtc *scrtc = to_shmob_crtc(crtc); + struct drm_device *dev = scrtc->crtc.dev; + unsigned long flags; + + spin_lock_irqsave(&dev->event_lock, flags); + if (scrtc->event != NULL) { + spin_unlock_irqrestore(&dev->event_lock, flags); + return -EBUSY; + } + spin_unlock_irqrestore(&dev->event_lock, flags); + + crtc->fb = fb; + shmob_drm_crtc_update_base(scrtc); + + if (event) { + event->pipe = 0; + spin_lock_irqsave(&dev->event_lock, flags); + scrtc->event = event; + spin_unlock_irqrestore(&dev->event_lock, flags); + drm_vblank_get(dev, 0); + } + + return 0; +} + +static const struct drm_crtc_funcs crtc_funcs = { + .destroy = drm_crtc_cleanup, + .set_config = drm_crtc_helper_set_config, + .page_flip = shmob_drm_crtc_page_flip, +}; + +int shmob_drm_crtc_create(struct shmob_drm_device *sdev) +{ + struct drm_crtc *crtc = &sdev->crtc.crtc; + int ret; + + sdev->crtc.dpms = DRM_MODE_DPMS_OFF; + + ret = drm_crtc_init(sdev->ddev, crtc, &crtc_funcs); + if (ret < 0) + return ret; + + drm_crtc_helper_add(crtc, &crtc_helper_funcs); + + return 0; +} + +/* ----------------------------------------------------------------------------- + * Encoder + */ + +#define to_shmob_encoder(e) \ + container_of(e, struct shmob_drm_encoder, encoder) + +static void shmob_drm_encoder_dpms(struct drm_encoder *encoder, int mode) +{ + struct shmob_drm_encoder *senc = to_shmob_encoder(encoder); + struct shmob_drm_device *sdev = encoder->dev->dev_private; + struct shmob_drm_connector *scon = &sdev->connector; + + if (senc->dpms == mode) + return; + + shmob_drm_backlight_dpms(scon, mode); + + senc->dpms = mode; +} + +static bool shmob_drm_encoder_mode_fixup(struct drm_encoder *encoder, + const struct drm_display_mode *mode, + struct drm_display_mode *adjusted_mode) +{ + struct drm_device *dev = encoder->dev; + struct shmob_drm_device *sdev = dev->dev_private; + struct drm_connector *connector = &sdev->connector.connector; + const struct drm_display_mode *panel_mode; + + if (list_empty(&connector->modes)) { + dev_dbg(dev->dev, "mode_fixup: empty modes list\n"); + return false; + } + + /* The flat panel mode is fixed, just copy it to the adjusted mode. */ + panel_mode = list_first_entry(&connector->modes, + struct drm_display_mode, head); + drm_mode_copy(adjusted_mode, panel_mode); + + return true; +} + +static void shmob_drm_encoder_mode_prepare(struct drm_encoder *encoder) +{ + /* No-op, everything is handled in the CRTC code. */ +} + +static void shmob_drm_encoder_mode_set(struct drm_encoder *encoder, + struct drm_display_mode *mode, + struct drm_display_mode *adjusted_mode) +{ + /* No-op, everything is handled in the CRTC code. */ +} + +static void shmob_drm_encoder_mode_commit(struct drm_encoder *encoder) +{ + /* No-op, everything is handled in the CRTC code. */ +} + +static const struct drm_encoder_helper_funcs encoder_helper_funcs = { + .dpms = shmob_drm_encoder_dpms, + .mode_fixup = shmob_drm_encoder_mode_fixup, + .prepare = shmob_drm_encoder_mode_prepare, + .commit = shmob_drm_encoder_mode_commit, + .mode_set = shmob_drm_encoder_mode_set, +}; + +static void shmob_drm_encoder_destroy(struct drm_encoder *encoder) +{ + drm_encoder_cleanup(encoder); +} + +static const struct drm_encoder_funcs encoder_funcs = { + .destroy = shmob_drm_encoder_destroy, +}; + +int shmob_drm_encoder_create(struct shmob_drm_device *sdev) +{ + struct drm_encoder *encoder = &sdev->encoder.encoder; + int ret; + + sdev->encoder.dpms = DRM_MODE_DPMS_OFF; + + encoder->possible_crtcs = 1; + + ret = drm_encoder_init(sdev->ddev, encoder, &encoder_funcs, + DRM_MODE_ENCODER_LVDS); + if (ret < 0) + return ret; + + drm_encoder_helper_add(encoder, &encoder_helper_funcs); + + return 0; +} + +void shmob_drm_crtc_enable_vblank(struct shmob_drm_device *sdev, bool enable) +{ + unsigned long flags; + u32 ldintr; + + /* Be careful not to acknowledge any pending interrupt. */ + spin_lock_irqsave(&sdev->irq_lock, flags); + ldintr = lcdc_read(sdev, LDINTR) | LDINTR_STATUS_MASK; + if (enable) + ldintr |= LDINTR_VEE; + else + ldintr &= ~LDINTR_VEE; + lcdc_write(sdev, LDINTR, ldintr); + spin_unlock_irqrestore(&sdev->irq_lock, flags); +} + +/* ----------------------------------------------------------------------------- + * Connector + */ + +#define to_shmob_connector(c) \ + container_of(c, struct shmob_drm_connector, connector) + +static int shmob_drm_connector_get_modes(struct drm_connector *connector) +{ + struct shmob_drm_device *sdev = connector->dev->dev_private; + struct drm_display_mode *mode; + + mode = drm_mode_create(connector->dev); + if (mode == NULL) + return 0; + + mode->type = DRM_MODE_TYPE_PREFERRED | DRM_MODE_TYPE_DRIVER; + mode->clock = sdev->pdata->panel.mode.clock; + mode->hdisplay = sdev->pdata->panel.mode.hdisplay; + mode->hsync_start = sdev->pdata->panel.mode.hsync_start; + mode->hsync_end = sdev->pdata->panel.mode.hsync_end; + mode->htotal = sdev->pdata->panel.mode.htotal; + mode->vdisplay = sdev->pdata->panel.mode.vdisplay; + mode->vsync_start = sdev->pdata->panel.mode.vsync_start; + mode->vsync_end = sdev->pdata->panel.mode.vsync_end; + mode->vtotal = sdev->pdata->panel.mode.vtotal; + mode->flags = sdev->pdata->panel.mode.flags; + + drm_mode_set_name(mode); + drm_mode_probed_add(connector, mode); + + connector->display_info.width_mm = sdev->pdata->panel.width_mm; + connector->display_info.height_mm = sdev->pdata->panel.height_mm; + + return 1; +} + +static int shmob_drm_connector_mode_valid(struct drm_connector *connector, + struct drm_display_mode *mode) +{ + return MODE_OK; +} + +static struct drm_encoder * +shmob_drm_connector_best_encoder(struct drm_connector *connector) +{ + struct shmob_drm_connector *scon = to_shmob_connector(connector); + + return scon->encoder; +} + +static const struct drm_connector_helper_funcs connector_helper_funcs = { + .get_modes = shmob_drm_connector_get_modes, + .mode_valid = shmob_drm_connector_mode_valid, + .best_encoder = shmob_drm_connector_best_encoder, +}; + +static void shmob_drm_connector_destroy(struct drm_connector *connector) +{ + struct shmob_drm_connector *scon = to_shmob_connector(connector); + + shmob_drm_backlight_exit(scon); + drm_sysfs_connector_remove(connector); + drm_connector_cleanup(connector); +} + +static enum drm_connector_status +shmob_drm_connector_detect(struct drm_connector *connector, bool force) +{ + return connector_status_connected; +} + +static const struct drm_connector_funcs connector_funcs = { + .dpms = drm_helper_connector_dpms, + .detect = shmob_drm_connector_detect, + .fill_modes = drm_helper_probe_single_connector_modes, + .destroy = shmob_drm_connector_destroy, +}; + +int shmob_drm_connector_create(struct shmob_drm_device *sdev, + struct drm_encoder *encoder) +{ + struct drm_connector *connector = &sdev->connector.connector; + int ret; + + sdev->connector.encoder = encoder; + + connector->display_info.width_mm = sdev->pdata->panel.width_mm; + connector->display_info.height_mm = sdev->pdata->panel.height_mm; + + ret = drm_connector_init(sdev->ddev, connector, &connector_funcs, + DRM_MODE_CONNECTOR_LVDS); + if (ret < 0) + return ret; + + drm_connector_helper_add(connector, &connector_helper_funcs); + ret = drm_sysfs_connector_add(connector); + if (ret < 0) + goto err_cleanup; + + ret = shmob_drm_backlight_init(&sdev->connector); + if (ret < 0) + goto err_sysfs; + + ret = drm_mode_connector_attach_encoder(connector, encoder); + if (ret < 0) + goto err_backlight; + + connector->encoder = encoder; + + drm_helper_connector_dpms(connector, DRM_MODE_DPMS_OFF); + drm_connector_property_set_value(connector, + sdev->ddev->mode_config.dpms_property, DRM_MODE_DPMS_OFF); + + return 0; + +err_backlight: + shmob_drm_backlight_exit(&sdev->connector); +err_sysfs: + drm_sysfs_connector_remove(connector); +err_cleanup: + drm_connector_cleanup(connector); + return ret; +} diff --git a/drivers/gpu/drm/shmobile/shmob_drm_crtc.h b/drivers/gpu/drm/shmobile/shmob_drm_crtc.h new file mode 100644 index 000000000000..e5bd109c4c38 --- /dev/null +++ b/drivers/gpu/drm/shmobile/shmob_drm_crtc.h @@ -0,0 +1,60 @@ +/* + * shmob_drm_crtc.h -- SH Mobile DRM CRTCs + * + * Copyright (C) 2012 Renesas Corporation + * + * Laurent Pinchart (laurent.pinchart@ideasonboard.com) + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + */ + +#ifndef __SHMOB_DRM_CRTC_H__ +#define __SHMOB_DRM_CRTC_H__ + +#include <drm/drmP.h> +#include <drm/drm_crtc.h> + +struct backlight_device; +struct shmob_drm_device; + +struct shmob_drm_crtc { + struct drm_crtc crtc; + + struct drm_pending_vblank_event *event; + int dpms; + + const struct shmob_drm_format_info *format; + void *cache; + unsigned long dma[2]; + unsigned int line_size; + bool started; +}; + +struct shmob_drm_encoder { + struct drm_encoder encoder; + int dpms; +}; + +struct shmob_drm_connector { + struct drm_connector connector; + struct drm_encoder *encoder; + + struct backlight_device *backlight; +}; + +int shmob_drm_crtc_create(struct shmob_drm_device *sdev); +void shmob_drm_crtc_enable_vblank(struct shmob_drm_device *sdev, bool enable); +void shmob_drm_crtc_cancel_page_flip(struct shmob_drm_crtc *scrtc, + struct drm_file *file); +void shmob_drm_crtc_finish_page_flip(struct shmob_drm_crtc *scrtc); +void shmob_drm_crtc_suspend(struct shmob_drm_crtc *scrtc); +void shmob_drm_crtc_resume(struct shmob_drm_crtc *scrtc); + +int shmob_drm_encoder_create(struct shmob_drm_device *sdev); +int shmob_drm_connector_create(struct shmob_drm_device *sdev, + struct drm_encoder *encoder); + +#endif /* __SHMOB_DRM_CRTC_H__ */ diff --git a/drivers/gpu/drm/shmobile/shmob_drm_drv.c b/drivers/gpu/drm/shmobile/shmob_drm_drv.c new file mode 100644 index 000000000000..c71d493fd0c5 --- /dev/null +++ b/drivers/gpu/drm/shmobile/shmob_drm_drv.c @@ -0,0 +1,361 @@ +/* + * shmob_drm_drv.c -- SH Mobile DRM driver + * + * Copyright (C) 2012 Renesas Corporation + * + * Laurent Pinchart (laurent.pinchart@ideasonboard.com) + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + */ + +#include <linux/clk.h> +#include <linux/io.h> +#include <linux/mm.h> +#include <linux/module.h> +#include <linux/platform_device.h> +#include <linux/pm.h> +#include <linux/slab.h> + +#include <drm/drmP.h> +#include <drm/drm_crtc_helper.h> +#include <drm/drm_gem_cma_helper.h> + +#include "shmob_drm_crtc.h" +#include "shmob_drm_drv.h" +#include "shmob_drm_kms.h" +#include "shmob_drm_plane.h" +#include "shmob_drm_regs.h" + +/* ----------------------------------------------------------------------------- + * Hardware initialization + */ + +static int __devinit shmob_drm_init_interface(struct shmob_drm_device *sdev) +{ + static const u32 ldmt1r[] = { + [SHMOB_DRM_IFACE_RGB8] = LDMT1R_MIFTYP_RGB8, + [SHMOB_DRM_IFACE_RGB9] = LDMT1R_MIFTYP_RGB9, + [SHMOB_DRM_IFACE_RGB12A] = LDMT1R_MIFTYP_RGB12A, + [SHMOB_DRM_IFACE_RGB12B] = LDMT1R_MIFTYP_RGB12B, + [SHMOB_DRM_IFACE_RGB16] = LDMT1R_MIFTYP_RGB16, + [SHMOB_DRM_IFACE_RGB18] = LDMT1R_MIFTYP_RGB18, + [SHMOB_DRM_IFACE_RGB24] = LDMT1R_MIFTYP_RGB24, + [SHMOB_DRM_IFACE_YUV422] = LDMT1R_MIFTYP_YCBCR, + [SHMOB_DRM_IFACE_SYS8A] = LDMT1R_IFM | LDMT1R_MIFTYP_SYS8A, + [SHMOB_DRM_IFACE_SYS8B] = LDMT1R_IFM | LDMT1R_MIFTYP_SYS8B, + [SHMOB_DRM_IFACE_SYS8C] = LDMT1R_IFM | LDMT1R_MIFTYP_SYS8C, + [SHMOB_DRM_IFACE_SYS8D] = LDMT1R_IFM | LDMT1R_MIFTYP_SYS8D, + [SHMOB_DRM_IFACE_SYS9] = LDMT1R_IFM | LDMT1R_MIFTYP_SYS9, + [SHMOB_DRM_IFACE_SYS12] = LDMT1R_IFM | LDMT1R_MIFTYP_SYS12, + [SHMOB_DRM_IFACE_SYS16A] = LDMT1R_IFM | LDMT1R_MIFTYP_SYS16A, + [SHMOB_DRM_IFACE_SYS16B] = LDMT1R_IFM | LDMT1R_MIFTYP_SYS16B, + [SHMOB_DRM_IFACE_SYS16C] = LDMT1R_IFM | LDMT1R_MIFTYP_SYS16C, + [SHMOB_DRM_IFACE_SYS18] = LDMT1R_IFM | LDMT1R_MIFTYP_SYS18, + [SHMOB_DRM_IFACE_SYS24] = LDMT1R_IFM | LDMT1R_MIFTYP_SYS24, + }; + + if (sdev->pdata->iface.interface >= ARRAY_SIZE(ldmt1r)) { + dev_err(sdev->dev, "invalid interface type %u\n", + sdev->pdata->iface.interface); + return -EINVAL; + } + + sdev->ldmt1r = ldmt1r[sdev->pdata->iface.interface]; + return 0; +} + +static int __devinit shmob_drm_setup_clocks(struct shmob_drm_device *sdev, + enum shmob_drm_clk_source clksrc) +{ + struct clk *clk; + char *clkname; + + switch (clksrc) { + case SHMOB_DRM_CLK_BUS: + clkname = "bus_clk"; + sdev->lddckr = LDDCKR_ICKSEL_BUS; + break; + case SHMOB_DRM_CLK_PERIPHERAL: + clkname = "peripheral_clk"; + sdev->lddckr = LDDCKR_ICKSEL_MIPI; + break; + case SHMOB_DRM_CLK_EXTERNAL: + clkname = NULL; + sdev->lddckr = LDDCKR_ICKSEL_HDMI; + break; + default: + return -EINVAL; + } + + clk = clk_get(sdev->dev, clkname); + if (IS_ERR(clk)) { + dev_err(sdev->dev, "cannot get dot clock %s\n", clkname); + return PTR_ERR(clk); + } + + sdev->clock = clk; + return 0; +} + +/* ----------------------------------------------------------------------------- + * DRM operations + */ + +static int shmob_drm_unload(struct drm_device *dev) +{ + struct shmob_drm_device *sdev = dev->dev_private; + + drm_kms_helper_poll_fini(dev); + drm_mode_config_cleanup(dev); + drm_vblank_cleanup(dev); + drm_irq_uninstall(dev); + + if (sdev->clock) + clk_put(sdev->clock); + + if (sdev->mmio) + iounmap(sdev->mmio); + + dev->dev_private = NULL; + kfree(sdev); + + return 0; +} + +static int shmob_drm_load(struct drm_device *dev, unsigned long flags) +{ + struct shmob_drm_platform_data *pdata = dev->dev->platform_data; + struct platform_device *pdev = dev->platformdev; + struct shmob_drm_device *sdev; + struct resource *res; + unsigned int i; + int ret; + + if (pdata == NULL) { + dev_err(dev->dev, "no platform data\n"); + return -EINVAL; + } + + sdev = kzalloc(sizeof(*sdev), GFP_KERNEL); + if (sdev == NULL) { + dev_err(dev->dev, "failed to allocate private data\n"); + return -ENOMEM; + } + + sdev->dev = &pdev->dev; + sdev->pdata = pdata; + spin_lock_init(&sdev->irq_lock); + + sdev->ddev = dev; + dev->dev_private = sdev; + + /* I/O resources and clocks */ + res = platform_get_resource(pdev, IORESOURCE_MEM, 0); + if (res == NULL) { + dev_err(&pdev->dev, "failed to get memory resource\n"); + ret = -EINVAL; + goto done; + } + + sdev->mmio = ioremap_nocache(res->start, resource_size(res)); + if (sdev->mmio == NULL) { + dev_err(&pdev->dev, "failed to remap memory resource\n"); + ret = -ENOMEM; + goto done; + } + + ret = shmob_drm_setup_clocks(sdev, pdata->clk_source); + if (ret < 0) + goto done; + + ret = shmob_drm_init_interface(sdev); + if (ret < 0) + goto done; + + ret = shmob_drm_modeset_init(sdev); + if (ret < 0) { + dev_err(&pdev->dev, "failed to initialize mode setting\n"); + goto done; + } + + for (i = 0; i < 4; ++i) { + ret = shmob_drm_plane_create(sdev, i); + if (ret < 0) { + dev_err(&pdev->dev, "failed to create plane %u\n", i); + goto done; + } + } + + ret = drm_vblank_init(dev, 1); + if (ret < 0) { + dev_err(&pdev->dev, "failed to initialize vblank\n"); + goto done; + } + + ret = drm_irq_install(dev); + if (ret < 0) { + dev_err(&pdev->dev, "failed to install IRQ handler\n"); + goto done; + } + +done: + if (ret) + shmob_drm_unload(dev); + + return ret; +} + +static void shmob_drm_preclose(struct drm_device *dev, struct drm_file *file) +{ + struct shmob_drm_device *sdev = dev->dev_private; + + shmob_drm_crtc_cancel_page_flip(&sdev->crtc, file); +} + +static irqreturn_t shmob_drm_irq(int irq, void *arg) +{ + struct drm_device *dev = arg; + struct shmob_drm_device *sdev = dev->dev_private; + unsigned long flags; + u32 status; + + /* Acknowledge interrupts. Putting interrupt enable and interrupt flag + * bits in the same register is really brain-dead design and requires + * taking a spinlock. + */ + spin_lock_irqsave(&sdev->irq_lock, flags); + status = lcdc_read(sdev, LDINTR); + lcdc_write(sdev, LDINTR, status ^ LDINTR_STATUS_MASK); + spin_unlock_irqrestore(&sdev->irq_lock, flags); + + if (status & LDINTR_VES) { + drm_handle_vblank(dev, 0); + shmob_drm_crtc_finish_page_flip(&sdev->crtc); + } + + return IRQ_HANDLED; +} + +static int shmob_drm_enable_vblank(struct drm_device *dev, int crtc) +{ + struct shmob_drm_device *sdev = dev->dev_private; + + shmob_drm_crtc_enable_vblank(sdev, true); + + return 0; +} + +static void shmob_drm_disable_vblank(struct drm_device *dev, int crtc) +{ + struct shmob_drm_device *sdev = dev->dev_private; + + shmob_drm_crtc_enable_vblank(sdev, false); +} + +static const struct file_operations shmob_drm_fops = { + .owner = THIS_MODULE, + .open = drm_open, + .release = drm_release, + .unlocked_ioctl = drm_ioctl, +#ifdef CONFIG_COMPAT + .compat_ioctl = drm_compat_ioctl, +#endif + .poll = drm_poll, + .read = drm_read, + .fasync = drm_fasync, + .llseek = no_llseek, + .mmap = drm_gem_cma_mmap, +}; + +static struct drm_driver shmob_drm_driver = { + .driver_features = DRIVER_HAVE_IRQ | DRIVER_GEM | DRIVER_MODESET, + .load = shmob_drm_load, + .unload = shmob_drm_unload, + .preclose = shmob_drm_preclose, + .irq_handler = shmob_drm_irq, + .get_vblank_counter = drm_vblank_count, + .enable_vblank = shmob_drm_enable_vblank, + .disable_vblank = shmob_drm_disable_vblank, + .gem_free_object = drm_gem_cma_free_object, + .gem_vm_ops = &drm_gem_cma_vm_ops, + .dumb_create = drm_gem_cma_dumb_create, + .dumb_map_offset = drm_gem_cma_dumb_map_offset, + .dumb_destroy = drm_gem_cma_dumb_destroy, + .fops = &shmob_drm_fops, + .name = "shmob-drm", + .desc = "Renesas SH Mobile DRM", + .date = "20120424", + .major = 1, + .minor = 0, +}; + +/* ----------------------------------------------------------------------------- + * Power management + */ + +#if CONFIG_PM_SLEEP +static int shmob_drm_pm_suspend(struct device *dev) +{ + struct platform_device *pdev = to_platform_device(dev); + struct drm_device *ddev = platform_get_drvdata(pdev); + struct shmob_drm_device *sdev = ddev->dev_private; + + drm_kms_helper_poll_disable(ddev); + shmob_drm_crtc_suspend(&sdev->crtc); + + return 0; +} + +static int shmob_drm_pm_resume(struct device *dev) +{ + struct platform_device *pdev = to_platform_device(dev); + struct drm_device *ddev = platform_get_drvdata(pdev); + struct shmob_drm_device *sdev = ddev->dev_private; + + mutex_lock(&sdev->ddev->mode_config.mutex); + shmob_drm_crtc_resume(&sdev->crtc); + mutex_unlock(&sdev->ddev->mode_config.mutex); + + drm_kms_helper_poll_enable(sdev->ddev); + return 0; +} +#endif + +static const struct dev_pm_ops shmob_drm_pm_ops = { + SET_SYSTEM_SLEEP_PM_OPS(shmob_drm_pm_suspend, shmob_drm_pm_resume) +}; + +/* ----------------------------------------------------------------------------- + * Platform driver + */ + +static int __devinit shmob_drm_probe(struct platform_device *pdev) +{ + return drm_platform_init(&shmob_drm_driver, pdev); +} + +static int __devexit shmob_drm_remove(struct platform_device *pdev) +{ + drm_platform_exit(&shmob_drm_driver, pdev); + + return 0; +} + +static struct platform_driver shmob_drm_platform_driver = { + .probe = shmob_drm_probe, + .remove = __devexit_p(shmob_drm_remove), + .driver = { + .owner = THIS_MODULE, + .name = "shmob-drm", + .pm = &shmob_drm_pm_ops, + }, +}; + +module_platform_driver(shmob_drm_platform_driver); + +MODULE_AUTHOR("Laurent Pinchart <laurent.pinchart@ideasonboard.com>"); +MODULE_DESCRIPTION("Renesas SH Mobile DRM Driver"); +MODULE_LICENSE("GPL"); diff --git a/drivers/gpu/drm/shmobile/shmob_drm_drv.h b/drivers/gpu/drm/shmobile/shmob_drm_drv.h new file mode 100644 index 000000000000..4d46b811b5a7 --- /dev/null +++ b/drivers/gpu/drm/shmobile/shmob_drm_drv.h @@ -0,0 +1,47 @@ +/* + * shmob_drm.h -- SH Mobile DRM driver + * + * Copyright (C) 2012 Renesas Corporation + * + * Laurent Pinchart (laurent.pinchart@ideasonboard.com) + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + */ + +#ifndef __SHMOB_DRM_DRV_H__ +#define __SHMOB_DRM_DRV_H__ + +#include <linux/kernel.h> +#include <linux/platform_data/shmob_drm.h> +#include <linux/spinlock.h> + +#include "shmob_drm_crtc.h" + +struct clk; +struct device; +struct drm_device; +struct sh_mobile_meram_info; + +struct shmob_drm_device { + struct device *dev; + const struct shmob_drm_platform_data *pdata; + + void __iomem *mmio; + struct clk *clock; + struct sh_mobile_meram_info *meram; + u32 lddckr; + u32 ldmt1r; + + spinlock_t irq_lock; /* Protects hardware LDINTR register */ + + struct drm_device *ddev; + + struct shmob_drm_crtc crtc; + struct shmob_drm_encoder encoder; + struct shmob_drm_connector connector; +}; + +#endif /* __SHMOB_DRM_DRV_H__ */ diff --git a/drivers/gpu/drm/shmobile/shmob_drm_kms.c b/drivers/gpu/drm/shmobile/shmob_drm_kms.c new file mode 100644 index 000000000000..c291ee385b4f --- /dev/null +++ b/drivers/gpu/drm/shmobile/shmob_drm_kms.c @@ -0,0 +1,160 @@ +/* + * shmob_drm_kms.c -- SH Mobile DRM Mode Setting + * + * Copyright (C) 2012 Renesas Corporation + * + * Laurent Pinchart (laurent.pinchart@ideasonboard.com) + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + */ + +#include <drm/drmP.h> +#include <drm/drm_crtc.h> +#include <drm/drm_crtc_helper.h> +#include <drm/drm_fb_cma_helper.h> +#include <drm/drm_gem_cma_helper.h> + +#include <video/sh_mobile_meram.h> + +#include "shmob_drm_crtc.h" +#include "shmob_drm_drv.h" +#include "shmob_drm_kms.h" +#include "shmob_drm_regs.h" + +/* ----------------------------------------------------------------------------- + * Format helpers + */ + +static const struct shmob_drm_format_info shmob_drm_format_infos[] = { + { + .fourcc = DRM_FORMAT_RGB565, + .bpp = 16, + .yuv = false, + .lddfr = LDDFR_PKF_RGB16, + .meram = SH_MOBILE_MERAM_PF_RGB, + }, { + .fourcc = DRM_FORMAT_RGB888, + .bpp = 24, + .yuv = false, + .lddfr = LDDFR_PKF_RGB24, + .meram = SH_MOBILE_MERAM_PF_RGB, + }, { + .fourcc = DRM_FORMAT_ARGB8888, + .bpp = 32, + .yuv = false, + .lddfr = LDDFR_PKF_ARGB32, + .meram = SH_MOBILE_MERAM_PF_RGB, + }, { + .fourcc = DRM_FORMAT_NV12, + .bpp = 12, + .yuv = true, + .lddfr = LDDFR_CC | LDDFR_YF_420, + .meram = SH_MOBILE_MERAM_PF_NV, + }, { + .fourcc = DRM_FORMAT_NV21, + .bpp = 12, + .yuv = true, + .lddfr = LDDFR_CC | LDDFR_YF_420, + .meram = SH_MOBILE_MERAM_PF_NV, + }, { + .fourcc = DRM_FORMAT_NV16, + .bpp = 16, + .yuv = true, + .lddfr = LDDFR_CC | LDDFR_YF_422, + .meram = SH_MOBILE_MERAM_PF_NV, + }, { + .fourcc = DRM_FORMAT_NV61, + .bpp = 16, + .yuv = true, + .lddfr = LDDFR_CC | LDDFR_YF_422, + .meram = SH_MOBILE_MERAM_PF_NV, + }, { + .fourcc = DRM_FORMAT_NV24, + .bpp = 24, + .yuv = true, + .lddfr = LDDFR_CC | LDDFR_YF_444, + .meram = SH_MOBILE_MERAM_PF_NV24, + }, { + .fourcc = DRM_FORMAT_NV42, + .bpp = 24, + .yuv = true, + .lddfr = LDDFR_CC | LDDFR_YF_444, + .meram = SH_MOBILE_MERAM_PF_NV24, + }, +}; + +const struct shmob_drm_format_info *shmob_drm_format_info(u32 fourcc) +{ + unsigned int i; + + for (i = 0; i < ARRAY_SIZE(shmob_drm_format_infos); ++i) { + if (shmob_drm_format_infos[i].fourcc == fourcc) + return &shmob_drm_format_infos[i]; + } + + return NULL; +} + +/* ----------------------------------------------------------------------------- + * Frame buffer + */ + +static struct drm_framebuffer * +shmob_drm_fb_create(struct drm_device *dev, struct drm_file *file_priv, + struct drm_mode_fb_cmd2 *mode_cmd) +{ + const struct shmob_drm_format_info *format; + + format = shmob_drm_format_info(mode_cmd->pixel_format); + if (format == NULL) { + dev_dbg(dev->dev, "unsupported pixel format %08x\n", + mode_cmd->pixel_format); + return ERR_PTR(-EINVAL); + } + + if (mode_cmd->pitches[0] & 7 || mode_cmd->pitches[0] >= 65536) { + dev_dbg(dev->dev, "valid pitch value %u\n", + mode_cmd->pitches[0]); + return ERR_PTR(-EINVAL); + } + + if (format->yuv) { + unsigned int chroma_cpp = format->bpp == 24 ? 2 : 1; + + if (mode_cmd->pitches[1] != mode_cmd->pitches[0] * chroma_cpp) { + dev_dbg(dev->dev, + "luma and chroma pitches do not match\n"); + return ERR_PTR(-EINVAL); + } + } + + return drm_fb_cma_create(dev, file_priv, mode_cmd); +} + +static const struct drm_mode_config_funcs shmob_drm_mode_config_funcs = { + .fb_create = shmob_drm_fb_create, +}; + +int shmob_drm_modeset_init(struct shmob_drm_device *sdev) +{ + drm_mode_config_init(sdev->ddev); + + shmob_drm_crtc_create(sdev); + shmob_drm_encoder_create(sdev); + shmob_drm_connector_create(sdev, &sdev->encoder.encoder); + + drm_kms_helper_poll_init(sdev->ddev); + + sdev->ddev->mode_config.min_width = 0; + sdev->ddev->mode_config.min_height = 0; + sdev->ddev->mode_config.max_width = 4095; + sdev->ddev->mode_config.max_height = 4095; + sdev->ddev->mode_config.funcs = &shmob_drm_mode_config_funcs; + + drm_helper_disable_unused_functions(sdev->ddev); + + return 0; +} diff --git a/drivers/gpu/drm/shmobile/shmob_drm_kms.h b/drivers/gpu/drm/shmobile/shmob_drm_kms.h new file mode 100644 index 000000000000..9495c9111308 --- /dev/null +++ b/drivers/gpu/drm/shmobile/shmob_drm_kms.h @@ -0,0 +1,34 @@ +/* + * shmob_drm_kms.h -- SH Mobile DRM Mode Setting + * + * Copyright (C) 2012 Renesas Corporation + * + * Laurent Pinchart (laurent.pinchart@ideasonboard.com) + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + */ + +#ifndef __SHMOB_DRM_KMS_H__ +#define __SHMOB_DRM_KMS_H__ + +#include <linux/types.h> + +struct drm_gem_cma_object; +struct shmob_drm_device; + +struct shmob_drm_format_info { + u32 fourcc; + unsigned int bpp; + bool yuv; + u32 lddfr; + unsigned int meram; +}; + +const struct shmob_drm_format_info *shmob_drm_format_info(u32 fourcc); + +int shmob_drm_modeset_init(struct shmob_drm_device *sdev); + +#endif /* __SHMOB_DRM_KMS_H__ */ diff --git a/drivers/gpu/drm/shmobile/shmob_drm_plane.c b/drivers/gpu/drm/shmobile/shmob_drm_plane.c new file mode 100644 index 000000000000..e1eb899b0288 --- /dev/null +++ b/drivers/gpu/drm/shmobile/shmob_drm_plane.c @@ -0,0 +1,268 @@ +/* + * shmob_drm_plane.c -- SH Mobile DRM Planes + * + * Copyright (C) 2012 Renesas Corporation + * + * Laurent Pinchart (laurent.pinchart@ideasonboard.com) + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + */ + +#include <drm/drmP.h> +#include <drm/drm_crtc.h> +#include <drm/drm_crtc_helper.h> +#include <drm/drm_fb_cma_helper.h> +#include <drm/drm_gem_cma_helper.h> + +#include <video/sh_mobile_meram.h> + +#include "shmob_drm_drv.h" +#include "shmob_drm_kms.h" +#include "shmob_drm_plane.h" +#include "shmob_drm_regs.h" + +struct shmob_drm_plane { + struct drm_plane plane; + unsigned int index; + unsigned int alpha; + + const struct shmob_drm_format_info *format; + unsigned long dma[2]; + + unsigned int src_x; + unsigned int src_y; + unsigned int crtc_x; + unsigned int crtc_y; + unsigned int crtc_w; + unsigned int crtc_h; +}; + +#define to_shmob_plane(p) container_of(p, struct shmob_drm_plane, plane) + +static void shmob_drm_plane_compute_base(struct shmob_drm_plane *splane, + struct drm_framebuffer *fb, + int x, int y) +{ + struct drm_gem_cma_object *gem; + unsigned int bpp; + + bpp = splane->format->yuv ? 8 : splane->format->bpp; + gem = drm_fb_cma_get_gem_obj(fb, 0); + splane->dma[0] = gem->paddr + fb->offsets[0] + + y * fb->pitches[0] + x * bpp / 8; + + if (splane->format->yuv) { + bpp = splane->format->bpp - 8; + gem = drm_fb_cma_get_gem_obj(fb, 1); + splane->dma[1] = gem->paddr + fb->offsets[1] + + y / (bpp == 4 ? 2 : 1) * fb->pitches[1] + + x * (bpp == 16 ? 2 : 1); + } +} + +static void __shmob_drm_plane_setup(struct shmob_drm_plane *splane, + struct drm_framebuffer *fb) +{ + struct shmob_drm_device *sdev = splane->plane.dev->dev_private; + u32 format; + + /* TODO: Support ROP3 mode */ + format = LDBBSIFR_EN | (splane->alpha << LDBBSIFR_LAY_SHIFT); + + switch (splane->format->fourcc) { + case DRM_FORMAT_RGB565: + case DRM_FORMAT_NV21: + case DRM_FORMAT_NV61: + case DRM_FORMAT_NV42: + format |= LDBBSIFR_SWPL | LDBBSIFR_SWPW; + break; + case DRM_FORMAT_RGB888: + case DRM_FORMAT_NV12: + case DRM_FORMAT_NV16: + case DRM_FORMAT_NV24: + format |= LDBBSIFR_SWPL | LDBBSIFR_SWPW | LDBBSIFR_SWPB; + break; + case DRM_FORMAT_ARGB8888: + default: + format |= LDBBSIFR_SWPL; + break; + } + + switch (splane->format->fourcc) { + case DRM_FORMAT_RGB565: + format |= LDBBSIFR_AL_1 | LDBBSIFR_RY | LDBBSIFR_RPKF_RGB16; + break; + case DRM_FORMAT_RGB888: + format |= LDBBSIFR_AL_1 | LDBBSIFR_RY | LDBBSIFR_RPKF_RGB24; + break; + case DRM_FORMAT_ARGB8888: + format |= LDBBSIFR_AL_PK | LDBBSIFR_RY | LDDFR_PKF_ARGB32; + break; + case DRM_FORMAT_NV12: + case DRM_FORMAT_NV21: + format |= LDBBSIFR_AL_1 | LDBBSIFR_CHRR_420; + break; + case DRM_FORMAT_NV16: + case DRM_FORMAT_NV61: + format |= LDBBSIFR_AL_1 | LDBBSIFR_CHRR_422; + break; + case DRM_FORMAT_NV24: + case DRM_FORMAT_NV42: + format |= LDBBSIFR_AL_1 | LDBBSIFR_CHRR_444; + break; + } + +#define plane_reg_dump(sdev, splane, reg) \ + dev_dbg(sdev->ddev->dev, "%s(%u): %s 0x%08x 0x%08x\n", __func__, \ + splane->index, #reg, \ + lcdc_read(sdev, reg(splane->index)), \ + lcdc_read(sdev, reg(splane->index) + LCDC_SIDE_B_OFFSET)) + + plane_reg_dump(sdev, splane, LDBnBSIFR); + plane_reg_dump(sdev, splane, LDBnBSSZR); + plane_reg_dump(sdev, splane, LDBnBLOCR); + plane_reg_dump(sdev, splane, LDBnBSMWR); + plane_reg_dump(sdev, splane, LDBnBSAYR); + plane_reg_dump(sdev, splane, LDBnBSACR); + + lcdc_write(sdev, LDBCR, LDBCR_UPC(splane->index)); + dev_dbg(sdev->ddev->dev, "%s(%u): %s 0x%08x\n", __func__, splane->index, + "LDBCR", lcdc_read(sdev, LDBCR)); + + lcdc_write(sdev, LDBnBSIFR(splane->index), format); + + lcdc_write(sdev, LDBnBSSZR(splane->index), + (splane->crtc_h << LDBBSSZR_BVSS_SHIFT) | + (splane->crtc_w << LDBBSSZR_BHSS_SHIFT)); + lcdc_write(sdev, LDBnBLOCR(splane->index), + (splane->crtc_y << LDBBLOCR_CVLC_SHIFT) | + (splane->crtc_x << LDBBLOCR_CHLC_SHIFT)); + lcdc_write(sdev, LDBnBSMWR(splane->index), + fb->pitches[0] << LDBBSMWR_BSMW_SHIFT); + + shmob_drm_plane_compute_base(splane, fb, splane->src_x, splane->src_y); + + lcdc_write(sdev, LDBnBSAYR(splane->index), splane->dma[0]); + if (splane->format->yuv) + lcdc_write(sdev, LDBnBSACR(splane->index), splane->dma[1]); + + lcdc_write(sdev, LDBCR, + LDBCR_UPF(splane->index) | LDBCR_UPD(splane->index)); + dev_dbg(sdev->ddev->dev, "%s(%u): %s 0x%08x\n", __func__, splane->index, + "LDBCR", lcdc_read(sdev, LDBCR)); + + plane_reg_dump(sdev, splane, LDBnBSIFR); + plane_reg_dump(sdev, splane, LDBnBSSZR); + plane_reg_dump(sdev, splane, LDBnBLOCR); + plane_reg_dump(sdev, splane, LDBnBSMWR); + plane_reg_dump(sdev, splane, LDBnBSAYR); + plane_reg_dump(sdev, splane, LDBnBSACR); +} + +void shmob_drm_plane_setup(struct drm_plane *plane) +{ + struct shmob_drm_plane *splane = to_shmob_plane(plane); + + if (plane->fb == NULL || !plane->enabled) + return; + + __shmob_drm_plane_setup(splane, plane->fb); +} + +static int +shmob_drm_plane_update(struct drm_plane *plane, struct drm_crtc *crtc, + struct drm_framebuffer *fb, int crtc_x, int crtc_y, + unsigned int crtc_w, unsigned int crtc_h, + uint32_t src_x, uint32_t src_y, + uint32_t src_w, uint32_t src_h) +{ + struct shmob_drm_plane *splane = to_shmob_plane(plane); + struct shmob_drm_device *sdev = plane->dev->dev_private; + const struct shmob_drm_format_info *format; + + format = shmob_drm_format_info(fb->pixel_format); + if (format == NULL) { + dev_dbg(sdev->dev, "update_plane: unsupported format %08x\n", + fb->pixel_format); + return -EINVAL; + } + + if (src_w >> 16 != crtc_w || src_h >> 16 != crtc_h) { + dev_dbg(sdev->dev, "%s: scaling not supported\n", __func__); + return -EINVAL; + } + + splane->format = format; + + splane->src_x = src_x >> 16; + splane->src_y = src_y >> 16; + splane->crtc_x = crtc_x; + splane->crtc_y = crtc_y; + splane->crtc_w = crtc_w; + splane->crtc_h = crtc_h; + + __shmob_drm_plane_setup(splane, fb); + return 0; +} + +static int shmob_drm_plane_disable(struct drm_plane *plane) +{ + struct shmob_drm_plane *splane = to_shmob_plane(plane); + struct shmob_drm_device *sdev = plane->dev->dev_private; + + splane->format = NULL; + + lcdc_write(sdev, LDBnBSIFR(splane->index), 0); + return 0; +} + +static void shmob_drm_plane_destroy(struct drm_plane *plane) +{ + struct shmob_drm_plane *splane = to_shmob_plane(plane); + + shmob_drm_plane_disable(plane); + drm_plane_cleanup(plane); + kfree(splane); +} + +static const struct drm_plane_funcs shmob_drm_plane_funcs = { + .update_plane = shmob_drm_plane_update, + .disable_plane = shmob_drm_plane_disable, + .destroy = shmob_drm_plane_destroy, +}; + +static const uint32_t formats[] = { + DRM_FORMAT_RGB565, + DRM_FORMAT_RGB888, + DRM_FORMAT_ARGB8888, + DRM_FORMAT_NV12, + DRM_FORMAT_NV21, + DRM_FORMAT_NV16, + DRM_FORMAT_NV61, + DRM_FORMAT_NV24, + DRM_FORMAT_NV42, +}; + +int shmob_drm_plane_create(struct shmob_drm_device *sdev, unsigned int index) +{ + struct shmob_drm_plane *splane; + int ret; + + splane = kzalloc(sizeof(*splane), GFP_KERNEL); + if (splane == NULL) + return -ENOMEM; + + splane->index = index; + splane->alpha = 255; + + ret = drm_plane_init(sdev->ddev, &splane->plane, 1, + &shmob_drm_plane_funcs, formats, + ARRAY_SIZE(formats), false); + if (ret < 0) + kfree(splane); + + return ret; +} diff --git a/drivers/gpu/drm/shmobile/shmob_drm_plane.h b/drivers/gpu/drm/shmobile/shmob_drm_plane.h new file mode 100644 index 000000000000..99623d05e3b0 --- /dev/null +++ b/drivers/gpu/drm/shmobile/shmob_drm_plane.h @@ -0,0 +1,22 @@ +/* + * shmob_drm_plane.h -- SH Mobile DRM Planes + * + * Copyright (C) 2012 Renesas Corporation + * + * Laurent Pinchart (laurent.pinchart@ideasonboard.com) + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + */ + +#ifndef __SHMOB_DRM_PLANE_H__ +#define __SHMOB_DRM_PLANE_H__ + +struct shmob_drm_device; + +int shmob_drm_plane_create(struct shmob_drm_device *sdev, unsigned int index); +void shmob_drm_plane_setup(struct drm_plane *plane); + +#endif /* __SHMOB_DRM_PLANE_H__ */ diff --git a/drivers/gpu/drm/shmobile/shmob_drm_regs.h b/drivers/gpu/drm/shmobile/shmob_drm_regs.h new file mode 100644 index 000000000000..7923cdd6368e --- /dev/null +++ b/drivers/gpu/drm/shmobile/shmob_drm_regs.h @@ -0,0 +1,311 @@ +/* + * shmob_drm_regs.h -- SH Mobile DRM registers + * + * Copyright (C) 2012 Renesas Corporation + * + * Laurent Pinchart (laurent.pinchart@ideasonboard.com) + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + */ + +#ifndef __SHMOB_DRM_REGS_H__ +#define __SHMOB_DRM_REGS_H__ + +#include <linux/io.h> + +/* Register definitions */ +#define LDDCKPAT1R 0x400 +#define LDDCKPAT2R 0x404 +#define LDDCKR 0x410 +#define LDDCKR_ICKSEL_BUS (0 << 16) +#define LDDCKR_ICKSEL_MIPI (1 << 16) +#define LDDCKR_ICKSEL_HDMI (2 << 16) +#define LDDCKR_ICKSEL_EXT (3 << 16) +#define LDDCKR_ICKSEL_MASK (7 << 16) +#define LDDCKR_MOSEL (1 << 6) +#define LDDCKSTPR 0x414 +#define LDDCKSTPR_DCKSTS (1 << 16) +#define LDDCKSTPR_DCKSTP (1 << 0) +#define LDMT1R 0x418 +#define LDMT1R_VPOL (1 << 28) +#define LDMT1R_HPOL (1 << 27) +#define LDMT1R_DWPOL (1 << 26) +#define LDMT1R_DIPOL (1 << 25) +#define LDMT1R_DAPOL (1 << 24) +#define LDMT1R_HSCNT (1 << 17) +#define LDMT1R_DWCNT (1 << 16) +#define LDMT1R_IFM (1 << 12) +#define LDMT1R_MIFTYP_RGB8 (0x0 << 0) +#define LDMT1R_MIFTYP_RGB9 (0x4 << 0) +#define LDMT1R_MIFTYP_RGB12A (0x5 << 0) +#define LDMT1R_MIFTYP_RGB12B (0x6 << 0) +#define LDMT1R_MIFTYP_RGB16 (0x7 << 0) +#define LDMT1R_MIFTYP_RGB18 (0xa << 0) +#define LDMT1R_MIFTYP_RGB24 (0xb << 0) +#define LDMT1R_MIFTYP_YCBCR (0xf << 0) +#define LDMT1R_MIFTYP_SYS8A (0x0 << 0) +#define LDMT1R_MIFTYP_SYS8B (0x1 << 0) +#define LDMT1R_MIFTYP_SYS8C (0x2 << 0) +#define LDMT1R_MIFTYP_SYS8D (0x3 << 0) +#define LDMT1R_MIFTYP_SYS9 (0x4 << 0) +#define LDMT1R_MIFTYP_SYS12 (0x5 << 0) +#define LDMT1R_MIFTYP_SYS16A (0x7 << 0) +#define LDMT1R_MIFTYP_SYS16B (0x8 << 0) +#define LDMT1R_MIFTYP_SYS16C (0x9 << 0) +#define LDMT1R_MIFTYP_SYS18 (0xa << 0) +#define LDMT1R_MIFTYP_SYS24 (0xb << 0) +#define LDMT1R_MIFTYP_MASK (0xf << 0) +#define LDMT2R 0x41c +#define LDMT2R_CSUP_MASK (7 << 26) +#define LDMT2R_CSUP_SHIFT 26 +#define LDMT2R_RSV (1 << 25) +#define LDMT2R_VSEL (1 << 24) +#define LDMT2R_WCSC_MASK (0xff << 16) +#define LDMT2R_WCSC_SHIFT 16 +#define LDMT2R_WCEC_MASK (0xff << 8) +#define LDMT2R_WCEC_SHIFT 8 +#define LDMT2R_WCLW_MASK (0xff << 0) +#define LDMT2R_WCLW_SHIFT 0 +#define LDMT3R 0x420 +#define LDMT3R_RDLC_MASK (0x3f << 24) +#define LDMT3R_RDLC_SHIFT 24 +#define LDMT3R_RCSC_MASK (0xff << 16) +#define LDMT3R_RCSC_SHIFT 16 +#define LDMT3R_RCEC_MASK (0xff << 8) +#define LDMT3R_RCEC_SHIFT 8 +#define LDMT3R_RCLW_MASK (0xff << 0) +#define LDMT3R_RCLW_SHIFT 0 +#define LDDFR 0x424 +#define LDDFR_CF1 (1 << 18) +#define LDDFR_CF0 (1 << 17) +#define LDDFR_CC (1 << 16) +#define LDDFR_YF_420 (0 << 8) +#define LDDFR_YF_422 (1 << 8) +#define LDDFR_YF_444 (2 << 8) +#define LDDFR_YF_MASK (3 << 8) +#define LDDFR_PKF_ARGB32 (0x00 << 0) +#define LDDFR_PKF_RGB16 (0x03 << 0) +#define LDDFR_PKF_RGB24 (0x0b << 0) +#define LDDFR_PKF_MASK (0x1f << 0) +#define LDSM1R 0x428 +#define LDSM1R_OS (1 << 0) +#define LDSM2R 0x42c +#define LDSM2R_OSTRG (1 << 0) +#define LDSA1R 0x430 +#define LDSA2R 0x434 +#define LDMLSR 0x438 +#define LDWBFR 0x43c +#define LDWBCNTR 0x440 +#define LDWBAR 0x444 +#define LDHCNR 0x448 +#define LDHSYNR 0x44c +#define LDVLNR 0x450 +#define LDVSYNR 0x454 +#define LDHPDR 0x458 +#define LDVPDR 0x45c +#define LDPMR 0x460 +#define LDPMR_LPS (3 << 0) +#define LDINTR 0x468 +#define LDINTR_FE (1 << 10) +#define LDINTR_VSE (1 << 9) +#define LDINTR_VEE (1 << 8) +#define LDINTR_FS (1 << 2) +#define LDINTR_VSS (1 << 1) +#define LDINTR_VES (1 << 0) +#define LDINTR_STATUS_MASK (0xff << 0) +#define LDSR 0x46c +#define LDSR_MSS (1 << 10) +#define LDSR_MRS (1 << 8) +#define LDSR_AS (1 << 1) +#define LDCNT1R 0x470 +#define LDCNT1R_DE (1 << 0) +#define LDCNT2R 0x474 +#define LDCNT2R_BR (1 << 8) +#define LDCNT2R_MD (1 << 3) +#define LDCNT2R_SE (1 << 2) +#define LDCNT2R_ME (1 << 1) +#define LDCNT2R_DO (1 << 0) +#define LDRCNTR 0x478 +#define LDRCNTR_SRS (1 << 17) +#define LDRCNTR_SRC (1 << 16) +#define LDRCNTR_MRS (1 << 1) +#define LDRCNTR_MRC (1 << 0) +#define LDDDSR 0x47c +#define LDDDSR_LS (1 << 2) +#define LDDDSR_WS (1 << 1) +#define LDDDSR_BS (1 << 0) +#define LDHAJR 0x4a0 + +#define LDDWD0R 0x800 +#define LDDWDxR_WDACT (1 << 28) +#define LDDWDxR_RSW (1 << 24) +#define LDDRDR 0x840 +#define LDDRDR_RSR (1 << 24) +#define LDDRDR_DRD_MASK (0x3ffff << 0) +#define LDDWAR 0x900 +#define LDDWAR_WA (1 << 0) +#define LDDRAR 0x904 +#define LDDRAR_RA (1 << 0) + +#define LDBCR 0xb00 +#define LDBCR_UPC(n) (1 << ((n) + 16)) +#define LDBCR_UPF(n) (1 << ((n) + 8)) +#define LDBCR_UPD(n) (1 << ((n) + 0)) +#define LDBnBSIFR(n) (0xb20 + (n) * 0x20 + 0x00) +#define LDBBSIFR_EN (1 << 31) +#define LDBBSIFR_VS (1 << 29) +#define LDBBSIFR_BRSEL (1 << 28) +#define LDBBSIFR_MX (1 << 27) +#define LDBBSIFR_MY (1 << 26) +#define LDBBSIFR_CV3 (3 << 24) +#define LDBBSIFR_CV2 (2 << 24) +#define LDBBSIFR_CV1 (1 << 24) +#define LDBBSIFR_CV0 (0 << 24) +#define LDBBSIFR_CV_MASK (3 << 24) +#define LDBBSIFR_LAY_MASK (0xff << 16) +#define LDBBSIFR_LAY_SHIFT 16 +#define LDBBSIFR_ROP3_MASK (0xff << 16) +#define LDBBSIFR_ROP3_SHIFT 16 +#define LDBBSIFR_AL_PL8 (3 << 14) +#define LDBBSIFR_AL_PL1 (2 << 14) +#define LDBBSIFR_AL_PK (1 << 14) +#define LDBBSIFR_AL_1 (0 << 14) +#define LDBBSIFR_AL_MASK (3 << 14) +#define LDBBSIFR_SWPL (1 << 10) +#define LDBBSIFR_SWPW (1 << 9) +#define LDBBSIFR_SWPB (1 << 8) +#define LDBBSIFR_RY (1 << 7) +#define LDBBSIFR_CHRR_420 (2 << 0) +#define LDBBSIFR_CHRR_422 (1 << 0) +#define LDBBSIFR_CHRR_444 (0 << 0) +#define LDBBSIFR_RPKF_ARGB32 (0x00 << 0) +#define LDBBSIFR_RPKF_RGB16 (0x03 << 0) +#define LDBBSIFR_RPKF_RGB24 (0x0b << 0) +#define LDBBSIFR_RPKF_MASK (0x1f << 0) +#define LDBnBSSZR(n) (0xb20 + (n) * 0x20 + 0x04) +#define LDBBSSZR_BVSS_MASK (0xfff << 16) +#define LDBBSSZR_BVSS_SHIFT 16 +#define LDBBSSZR_BHSS_MASK (0xfff << 0) +#define LDBBSSZR_BHSS_SHIFT 0 +#define LDBnBLOCR(n) (0xb20 + (n) * 0x20 + 0x08) +#define LDBBLOCR_CVLC_MASK (0xfff << 16) +#define LDBBLOCR_CVLC_SHIFT 16 +#define LDBBLOCR_CHLC_MASK (0xfff << 0) +#define LDBBLOCR_CHLC_SHIFT 0 +#define LDBnBSMWR(n) (0xb20 + (n) * 0x20 + 0x0c) +#define LDBBSMWR_BSMWA_MASK (0xffff << 16) +#define LDBBSMWR_BSMWA_SHIFT 16 +#define LDBBSMWR_BSMW_MASK (0xffff << 0) +#define LDBBSMWR_BSMW_SHIFT 0 +#define LDBnBSAYR(n) (0xb20 + (n) * 0x20 + 0x10) +#define LDBBSAYR_FG1A_MASK (0xff << 24) +#define LDBBSAYR_FG1A_SHIFT 24 +#define LDBBSAYR_FG1R_MASK (0xff << 16) +#define LDBBSAYR_FG1R_SHIFT 16 +#define LDBBSAYR_FG1G_MASK (0xff << 8) +#define LDBBSAYR_FG1G_SHIFT 8 +#define LDBBSAYR_FG1B_MASK (0xff << 0) +#define LDBBSAYR_FG1B_SHIFT 0 +#define LDBnBSACR(n) (0xb20 + (n) * 0x20 + 0x14) +#define LDBBSACR_FG2A_MASK (0xff << 24) +#define LDBBSACR_FG2A_SHIFT 24 +#define LDBBSACR_FG2R_MASK (0xff << 16) +#define LDBBSACR_FG2R_SHIFT 16 +#define LDBBSACR_FG2G_MASK (0xff << 8) +#define LDBBSACR_FG2G_SHIFT 8 +#define LDBBSACR_FG2B_MASK (0xff << 0) +#define LDBBSACR_FG2B_SHIFT 0 +#define LDBnBSAAR(n) (0xb20 + (n) * 0x20 + 0x18) +#define LDBBSAAR_AP_MASK (0xff << 24) +#define LDBBSAAR_AP_SHIFT 24 +#define LDBBSAAR_R_MASK (0xff << 16) +#define LDBBSAAR_R_SHIFT 16 +#define LDBBSAAR_GY_MASK (0xff << 8) +#define LDBBSAAR_GY_SHIFT 8 +#define LDBBSAAR_B_MASK (0xff << 0) +#define LDBBSAAR_B_SHIFT 0 +#define LDBnBPPCR(n) (0xb20 + (n) * 0x20 + 0x1c) +#define LDBBPPCR_AP_MASK (0xff << 24) +#define LDBBPPCR_AP_SHIFT 24 +#define LDBBPPCR_R_MASK (0xff << 16) +#define LDBBPPCR_R_SHIFT 16 +#define LDBBPPCR_GY_MASK (0xff << 8) +#define LDBBPPCR_GY_SHIFT 8 +#define LDBBPPCR_B_MASK (0xff << 0) +#define LDBBPPCR_B_SHIFT 0 +#define LDBnBBGCL(n) (0xb10 + (n) * 0x04) +#define LDBBBGCL_BGA_MASK (0xff << 24) +#define LDBBBGCL_BGA_SHIFT 24 +#define LDBBBGCL_BGR_MASK (0xff << 16) +#define LDBBBGCL_BGR_SHIFT 16 +#define LDBBBGCL_BGG_MASK (0xff << 8) +#define LDBBBGCL_BGG_SHIFT 8 +#define LDBBBGCL_BGB_MASK (0xff << 0) +#define LDBBBGCL_BGB_SHIFT 0 + +#define LCDC_SIDE_B_OFFSET 0x1000 +#define LCDC_MIRROR_OFFSET 0x2000 + +static inline bool lcdc_is_banked(u32 reg) +{ + switch (reg) { + case LDMT1R: + case LDMT2R: + case LDMT3R: + case LDDFR: + case LDSM1R: + case LDSA1R: + case LDSA2R: + case LDMLSR: + case LDWBFR: + case LDWBCNTR: + case LDWBAR: + case LDHCNR: + case LDHSYNR: + case LDVLNR: + case LDVSYNR: + case LDHPDR: + case LDVPDR: + case LDHAJR: + return true; + default: + return reg >= LDBnBBGCL(0) && reg <= LDBnBPPCR(3); + } +} + +static inline void lcdc_write_mirror(struct shmob_drm_device *sdev, u32 reg, + u32 data) +{ + iowrite32(data, sdev->mmio + reg + LCDC_MIRROR_OFFSET); +} + +static inline void lcdc_write(struct shmob_drm_device *sdev, u32 reg, u32 data) +{ + iowrite32(data, sdev->mmio + reg); + if (lcdc_is_banked(reg)) + iowrite32(data, sdev->mmio + reg + LCDC_SIDE_B_OFFSET); +} + +static inline u32 lcdc_read(struct shmob_drm_device *sdev, u32 reg) +{ + return ioread32(sdev->mmio + reg); +} + +static inline int lcdc_wait_bit(struct shmob_drm_device *sdev, u32 reg, + u32 mask, u32 until) +{ + unsigned long timeout = jiffies + msecs_to_jiffies(5); + + while ((lcdc_read(sdev, reg) & mask) != until) { + if (time_after(jiffies, timeout)) + return -ETIMEDOUT; + cpu_relax(); + } + + return 0; +} + +#endif /* __SHMOB_DRM_REGS_H__ */ diff --git a/drivers/gpu/drm/ttm/ttm_bo_util.c b/drivers/gpu/drm/ttm/ttm_bo_util.c index f8187ead7b37..0df71eacd587 100644 --- a/drivers/gpu/drm/ttm/ttm_bo_util.c +++ b/drivers/gpu/drm/ttm/ttm_bo_util.c @@ -472,7 +472,7 @@ pgprot_t ttm_io_prot(uint32_t caching_flags, pgprot_t tmp) else tmp = pgprot_noncached(tmp); #endif -#if defined(__sparc__) +#if defined(__sparc__) || defined(__mips__) if (!(caching_flags & TTM_PL_FLAG_CACHED)) tmp = pgprot_noncached(tmp); #endif diff --git a/drivers/gpu/drm/ttm/ttm_page_alloc_dma.c b/drivers/gpu/drm/ttm/ttm_page_alloc_dma.c index 4f9e548b2eec..969d76532088 100644 --- a/drivers/gpu/drm/ttm/ttm_page_alloc_dma.c +++ b/drivers/gpu/drm/ttm/ttm_page_alloc_dma.c @@ -1060,7 +1060,7 @@ int ttm_dma_page_alloc_init(struct ttm_mem_global *glob, unsigned max_pages) _manager = kzalloc(sizeof(*_manager), GFP_KERNEL); if (!_manager) - goto err_manager; + goto err; mutex_init(&_manager->lock); INIT_LIST_HEAD(&_manager->pools); @@ -1078,9 +1078,6 @@ int ttm_dma_page_alloc_init(struct ttm_mem_global *glob, unsigned max_pages) } ttm_dma_pool_mm_shrink_init(_manager); return 0; -err_manager: - kfree(_manager); - _manager = NULL; err: return ret; } diff --git a/drivers/gpu/drm/ttm/ttm_tt.c b/drivers/gpu/drm/ttm/ttm_tt.c index fa09daf9a50c..11c9369d4208 100644 --- a/drivers/gpu/drm/ttm/ttm_tt.c +++ b/drivers/gpu/drm/ttm/ttm_tt.c @@ -290,8 +290,6 @@ int ttm_tt_swapin(struct ttm_tt *ttm) struct file *swap_storage; struct page *from_page; struct page *to_page; - void *from_virtual; - void *to_virtual; int i; int ret = -ENOMEM; @@ -311,11 +309,7 @@ int ttm_tt_swapin(struct ttm_tt *ttm) goto out_err; preempt_disable(); - from_virtual = kmap_atomic(from_page); - to_virtual = kmap_atomic(to_page); - memcpy(to_virtual, from_virtual, PAGE_SIZE); - kunmap_atomic(to_virtual); - kunmap_atomic(from_virtual); + copy_highpage(to_page, from_page); preempt_enable(); page_cache_release(from_page); } @@ -336,8 +330,6 @@ int ttm_tt_swapout(struct ttm_tt *ttm, struct file *persistent_swap_storage) struct file *swap_storage; struct page *from_page; struct page *to_page; - void *from_virtual; - void *to_virtual; int i; int ret = -ENOMEM; @@ -367,11 +359,7 @@ int ttm_tt_swapout(struct ttm_tt *ttm, struct file *persistent_swap_storage) goto out_err; } preempt_disable(); - from_virtual = kmap_atomic(from_page); - to_virtual = kmap_atomic(to_page); - memcpy(to_virtual, from_virtual, PAGE_SIZE); - kunmap_atomic(to_virtual); - kunmap_atomic(from_virtual); + copy_highpage(to_page, from_page); preempt_enable(); set_page_dirty(to_page); mark_page_accessed(to_page); diff --git a/drivers/gpu/drm/udl/udl_connector.c b/drivers/gpu/drm/udl/udl_connector.c index 8d9dc44f1f94..d568aa4b10d1 100644 --- a/drivers/gpu/drm/udl/udl_connector.c +++ b/drivers/gpu/drm/udl/udl_connector.c @@ -57,11 +57,8 @@ static int udl_get_modes(struct drm_connector *connector) edid = (struct edid *)udl_get_edid(udl); - connector->display_info.raw_edid = (char *)edid; - drm_mode_connector_update_edid_property(connector, edid); ret = drm_add_edid_modes(connector, edid); - connector->display_info.raw_edid = NULL; kfree(edid); return ret; } diff --git a/drivers/gpu/drm/udl/udl_encoder.c b/drivers/gpu/drm/udl/udl_encoder.c index 0731ab2e6c06..0476bfe06d14 100644 --- a/drivers/gpu/drm/udl/udl_encoder.c +++ b/drivers/gpu/drm/udl/udl_encoder.c @@ -16,7 +16,7 @@ #include "udl_drv.h" /* dummy encoder */ -void udl_enc_destroy(struct drm_encoder *encoder) +static void udl_enc_destroy(struct drm_encoder *encoder) { drm_encoder_cleanup(encoder); kfree(encoder); diff --git a/drivers/gpu/drm/udl/udl_fb.c b/drivers/gpu/drm/udl/udl_fb.c index ce9a61179925..0e47e9614b78 100644 --- a/drivers/gpu/drm/udl/udl_fb.c +++ b/drivers/gpu/drm/udl/udl_fb.c @@ -13,6 +13,7 @@ #include <linux/module.h> #include <linux/slab.h> #include <linux/fb.h> +#include <linux/dma-buf.h> #include "drmP.h" #include "drm.h" @@ -356,12 +357,12 @@ static struct fb_ops udlfb_ops = { .fb_release = udl_fb_release, }; -void udl_crtc_fb_gamma_set(struct drm_crtc *crtc, u16 red, u16 green, +static void udl_crtc_fb_gamma_set(struct drm_crtc *crtc, u16 red, u16 green, u16 blue, int regno) { } -void udl_crtc_fb_gamma_get(struct drm_crtc *crtc, u16 *red, u16 *green, +static void udl_crtc_fb_gamma_get(struct drm_crtc *crtc, u16 *red, u16 *green, u16 *blue, int regno) { *red = 0; @@ -377,16 +378,33 @@ static int udl_user_framebuffer_dirty(struct drm_framebuffer *fb, { struct udl_framebuffer *ufb = to_udl_fb(fb); int i; + int ret = 0; if (!ufb->active_16) return 0; + if (ufb->obj->base.import_attach) { + ret = dma_buf_begin_cpu_access(ufb->obj->base.import_attach->dmabuf, + 0, ufb->obj->base.size, + DMA_FROM_DEVICE); + if (ret) + return ret; + } + for (i = 0; i < num_clips; i++) { - udl_handle_damage(ufb, clips[i].x1, clips[i].y1, + ret = udl_handle_damage(ufb, clips[i].x1, clips[i].y1, clips[i].x2 - clips[i].x1, clips[i].y2 - clips[i].y1); + if (ret) + break; } - return 0; + + if (ufb->obj->base.import_attach) { + dma_buf_end_cpu_access(ufb->obj->base.import_attach->dmabuf, + 0, ufb->obj->base.size, + DMA_FROM_DEVICE); + } + return ret; } static void udl_user_framebuffer_destroy(struct drm_framebuffer *fb) diff --git a/drivers/gpu/drm/udl/udl_gem.c b/drivers/gpu/drm/udl/udl_gem.c index 291ecc145585..47b256375831 100644 --- a/drivers/gpu/drm/udl/udl_gem.c +++ b/drivers/gpu/drm/udl/udl_gem.c @@ -181,11 +181,6 @@ int udl_gem_vmap(struct udl_gem_object *obj) int ret; if (obj->base.import_attach) { - ret = dma_buf_begin_cpu_access(obj->base.import_attach->dmabuf, - 0, obj->base.size, DMA_BIDIRECTIONAL); - if (ret) - return -EINVAL; - obj->vmapping = dma_buf_vmap(obj->base.import_attach->dmabuf); if (!obj->vmapping) return -ENOMEM; @@ -206,8 +201,6 @@ void udl_gem_vunmap(struct udl_gem_object *obj) { if (obj->base.import_attach) { dma_buf_vunmap(obj->base.import_attach->dmabuf, obj->vmapping); - dma_buf_end_cpu_access(obj->base.import_attach->dmabuf, 0, - obj->base.size, DMA_BIDIRECTIONAL); return; } diff --git a/drivers/gpu/drm/udl/udl_main.c b/drivers/gpu/drm/udl/udl_main.c index 4c2d836a0893..40bf468c46c2 100644 --- a/drivers/gpu/drm/udl/udl_main.c +++ b/drivers/gpu/drm/udl/udl_main.c @@ -41,11 +41,8 @@ static int udl_parse_vendor_descriptor(struct drm_device *dev, total_len = usb_get_descriptor(usbdev, 0x5f, /* vendor specific */ 0, desc, MAX_VENDOR_DESCRIPTOR_SIZE); if (total_len > 5) { - DRM_INFO("vendor descriptor length:%x data:%02x %02x %02x %02x" \ - "%02x %02x %02x %02x %02x %02x %02x\n", - total_len, desc[0], - desc[1], desc[2], desc[3], desc[4], desc[5], desc[6], - desc[7], desc[8], desc[9], desc[10]); + DRM_INFO("vendor descriptor length:%x data:%*ph\n", + total_len, 11, desc); if ((desc[0] != total_len) || /* descriptor length */ (desc[1] != 0x5f) || /* vendor descriptor type */ diff --git a/drivers/gpu/drm/udl/udl_modeset.c b/drivers/gpu/drm/udl/udl_modeset.c index 9159d48d1dfd..8f258158c1fe 100644 --- a/drivers/gpu/drm/udl/udl_modeset.c +++ b/drivers/gpu/drm/udl/udl_modeset.c @@ -391,7 +391,7 @@ static const struct drm_crtc_funcs udl_crtc_funcs = { .destroy = udl_crtc_destroy, }; -int udl_crtc_init(struct drm_device *dev) +static int udl_crtc_init(struct drm_device *dev) { struct drm_crtc *crtc; diff --git a/drivers/gpu/drm/udl/udl_transfer.c b/drivers/gpu/drm/udl/udl_transfer.c index b9320e2608dd..f9fff657a90b 100644 --- a/drivers/gpu/drm/udl/udl_transfer.c +++ b/drivers/gpu/drm/udl/udl_transfer.c @@ -126,10 +126,10 @@ static void udl_compress_hline16( while ((pixel_end > pixel) && (cmd_buffer_end - MIN_RLX_CMD_BYTES > cmd)) { - uint8_t *raw_pixels_count_byte = 0; - uint8_t *cmd_pixels_count_byte = 0; - const u8 *raw_pixel_start = 0; - const u8 *cmd_pixel_start, *cmd_pixel_end = 0; + uint8_t *raw_pixels_count_byte = NULL; + uint8_t *cmd_pixels_count_byte = NULL; + const u8 *raw_pixel_start = NULL; + const u8 *cmd_pixel_start, *cmd_pixel_end = NULL; prefetchw((void *) cmd); /* pull in one cache line at least */ diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c index ba2c35dbf10e..09f44c5dd49d 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_drv.c @@ -438,7 +438,6 @@ static int vmw_driver_load(struct drm_device *dev, unsigned long chipset) DRM_ERROR("Failed allocating a device private struct.\n"); return -ENOMEM; } - memset(dev_priv, 0, sizeof(*dev_priv)); pci_set_master(dev->pdev); diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c index c50724bd30f6..54743943d8b3 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_kms.c @@ -483,7 +483,6 @@ static int do_surface_dirty_sou(struct vmw_private *dev_priv, } /* only need to do this once */ - memset(cmd, 0, fifo_size); cmd->header.id = cpu_to_le32(SVGA_3D_CMD_BLIT_SURFACE_TO_SCREEN); cmd->header.size = cpu_to_le32(fifo_size - sizeof(cmd->header)); diff --git a/drivers/staging/omapdrm/omap_connector.c b/drivers/staging/omapdrm/omap_connector.c index 55e9c8655850..38be186c249a 100644 --- a/drivers/staging/omapdrm/omap_connector.c +++ b/drivers/staging/omapdrm/omap_connector.c @@ -200,14 +200,11 @@ static int omap_connector_get_modes(struct drm_connector *connector) drm_mode_connector_update_edid_property( connector, edid); n = drm_add_edid_modes(connector, edid); - kfree(connector->display_info.raw_edid); - connector->display_info.raw_edid = edid; } else { drm_mode_connector_update_edid_property( connector, NULL); - connector->display_info.raw_edid = NULL; - kfree(edid); } + kfree(edid); } else { struct drm_display_mode *mode = drm_mode_create(dev); struct omap_video_timings timings = {0}; diff --git a/drivers/staging/omapdrm/omap_fbdev.c b/drivers/staging/omapdrm/omap_fbdev.c index 8c6ed3b0c6f6..8a027bb77d97 100644 --- a/drivers/staging/omapdrm/omap_fbdev.c +++ b/drivers/staging/omapdrm/omap_fbdev.c @@ -276,7 +276,7 @@ fail: if (fbi) framebuffer_release(fbi); if (fb) - fb->funcs->destroy(fb); + drm_framebuffer_remove(fb); } return ret; @@ -401,7 +401,7 @@ void omap_fbdev_free(struct drm_device *dev) /* this will free the backing object */ if (fbdev->fb) - fbdev->fb->funcs->destroy(fbdev->fb); + drm_framebuffer_remove(fbdev->fb); kfree(fbdev); diff --git a/include/drm/drmP.h b/include/drm/drmP.h index d6b67bb9075f..d5f0c163eef1 100644 --- a/include/drm/drmP.h +++ b/include/drm/drmP.h @@ -1367,6 +1367,7 @@ extern int drm_remove_magic(struct drm_master *master, drm_magic_t magic); /* Cache management (drm_cache.c) */ void drm_clflush_pages(struct page *pages[], unsigned long num_pages); +void drm_clflush_sg(struct sg_table *st); void drm_clflush_virt_range(char *addr, unsigned long length); /* Locking IOCTL support (drm_lock.h) */ diff --git a/include/drm/drm_crtc.h b/include/drm/drm_crtc.h index bfacf0d5a225..8e405b82707d 100644 --- a/include/drm/drm_crtc.h +++ b/include/drm/drm_crtc.h @@ -215,11 +215,10 @@ struct drm_display_info { u32 color_formats; u8 cea_rev; - - char *raw_edid; /* if any */ }; struct drm_framebuffer_funcs { + /* note: use drm_framebuffer_remove() */ void (*destroy)(struct drm_framebuffer *framebuffer); int (*create_handle)(struct drm_framebuffer *fb, struct drm_file *file_priv, @@ -244,6 +243,16 @@ struct drm_framebuffer_funcs { struct drm_framebuffer { struct drm_device *dev; + /* + * Note that the fb is refcounted for the benefit of driver internals, + * for example some hw, disabling a CRTC/plane is asynchronous, and + * scanout does not actually complete until the next vblank. So some + * cleanup (like releasing the reference(s) on the backing GEM bo(s)) + * should be deferred. In cases like this, the driver would like to + * hold a ref to the fb even though it has already been removed from + * userspace perspective. + */ + struct kref refcount; struct list_head head; struct drm_mode_object base; const struct drm_framebuffer_funcs *funcs; @@ -359,6 +368,9 @@ struct drm_crtc_funcs { * @enabled: is this CRTC enabled? * @mode: current mode timings * @hwmode: mode timings as programmed to hw regs + * @invert_dimensions: for purposes of error checking crtc vs fb sizes, + * invert the width/height of the crtc. This is used if the driver + * is performing 90 or 270 degree rotated scanout * @x: x position on screen * @y: y position on screen * @funcs: CRTC control functions @@ -392,6 +404,8 @@ struct drm_crtc { */ struct drm_display_mode hwmode; + bool invert_dimensions; + int x, y; const struct drm_crtc_funcs *funcs; @@ -593,6 +607,7 @@ struct drm_connector { int video_latency[2]; /* [0]: progressive, [1]: interlaced */ int audio_latency[2]; int null_edid_counter; /* needed to workaround some HW bugs where we get all 0s */ + unsigned bad_edid_counter; }; /** @@ -920,6 +935,9 @@ extern void drm_framebuffer_set_object(struct drm_device *dev, extern int drm_framebuffer_init(struct drm_device *dev, struct drm_framebuffer *fb, const struct drm_framebuffer_funcs *funcs); +extern void drm_framebuffer_unreference(struct drm_framebuffer *fb); +extern void drm_framebuffer_reference(struct drm_framebuffer *fb); +extern void drm_framebuffer_remove(struct drm_framebuffer *fb); extern void drm_framebuffer_cleanup(struct drm_framebuffer *fb); extern int drmfb_probe(struct drm_device *dev, struct drm_crtc *crtc); extern int drmfb_remove(struct drm_device *dev, struct drm_framebuffer *fb); @@ -1035,7 +1053,7 @@ extern int drm_add_modes_noedid(struct drm_connector *connector, int hdisplay, int vdisplay); extern int drm_edid_header_is_valid(const u8 *raw_edid); -extern bool drm_edid_block_valid(u8 *raw_edid, int block); +extern bool drm_edid_block_valid(u8 *raw_edid, int block, bool print_bad_edid); extern bool drm_edid_is_valid(struct edid *edid); struct drm_display_mode *drm_mode_find_dmt(struct drm_device *dev, int hsize, int vsize, int fresh, diff --git a/include/drm/drm_fb_cma_helper.h b/include/drm/drm_fb_cma_helper.h new file mode 100644 index 000000000000..76c709837543 --- /dev/null +++ b/include/drm/drm_fb_cma_helper.h @@ -0,0 +1,27 @@ +#ifndef __DRM_FB_CMA_HELPER_H__ +#define __DRM_FB_CMA_HELPER_H__ + +struct drm_fbdev_cma; +struct drm_gem_cma_object; + +struct drm_framebuffer; +struct drm_device; +struct drm_file; +struct drm_mode_fb_cmd2; + +struct drm_fbdev_cma *drm_fbdev_cma_init(struct drm_device *dev, + unsigned int preferred_bpp, unsigned int num_crtc, + unsigned int max_conn_count); +void drm_fbdev_cma_fini(struct drm_fbdev_cma *fbdev_cma); + +void drm_fbdev_cma_restore_mode(struct drm_fbdev_cma *fbdev_cma); +void drm_fbdev_cma_hotplug_event(struct drm_fbdev_cma *fbdev_cma); + +struct drm_framebuffer *drm_fb_cma_create(struct drm_device *dev, + struct drm_file *file_priv, struct drm_mode_fb_cmd2 *mode_cmd); + +struct drm_gem_cma_object *drm_fb_cma_get_gem_obj(struct drm_framebuffer *fb, + unsigned int plane); + +#endif + diff --git a/include/drm/drm_fourcc.h b/include/drm/drm_fourcc.h index f4621184a9b4..646ae5f39f42 100644 --- a/include/drm/drm_fourcc.h +++ b/include/drm/drm_fourcc.h @@ -106,6 +106,8 @@ #define DRM_FORMAT_NV21 fourcc_code('N', 'V', '2', '1') /* 2x2 subsampled Cb:Cr plane */ #define DRM_FORMAT_NV16 fourcc_code('N', 'V', '1', '6') /* 2x1 subsampled Cr:Cb plane */ #define DRM_FORMAT_NV61 fourcc_code('N', 'V', '6', '1') /* 2x1 subsampled Cb:Cr plane */ +#define DRM_FORMAT_NV24 fourcc_code('N', 'V', '2', '4') /* non-subsampled Cr:Cb plane */ +#define DRM_FORMAT_NV42 fourcc_code('N', 'V', '4', '2') /* non-subsampled Cb:Cr plane */ /* special NV12 tiled format */ #define DRM_FORMAT_NV12MT fourcc_code('T', 'M', '1', '2') /* 2x2 subsampled Cr:Cb plane 64x32 macroblocks */ diff --git a/include/drm/drm_gem_cma_helper.h b/include/drm/drm_gem_cma_helper.h new file mode 100644 index 000000000000..f0f6b1af25ad --- /dev/null +++ b/include/drm/drm_gem_cma_helper.h @@ -0,0 +1,44 @@ +#ifndef __DRM_GEM_CMA_HELPER_H__ +#define __DRM_GEM_CMA_HELPER_H__ + +struct drm_gem_cma_object { + struct drm_gem_object base; + dma_addr_t paddr; + void *vaddr; +}; + +static inline struct drm_gem_cma_object * +to_drm_gem_cma_obj(struct drm_gem_object *gem_obj) +{ + return container_of(gem_obj, struct drm_gem_cma_object, base); +} + +/* free gem object. */ +void drm_gem_cma_free_object(struct drm_gem_object *gem_obj); + +/* create memory region for drm framebuffer. */ +int drm_gem_cma_dumb_create(struct drm_file *file_priv, + struct drm_device *drm, struct drm_mode_create_dumb *args); + +/* map memory region for drm framebuffer to user space. */ +int drm_gem_cma_dumb_map_offset(struct drm_file *file_priv, + struct drm_device *drm, uint32_t handle, uint64_t *offset); + +/* set vm_flags and we can change the vm attribute to other one at here. */ +int drm_gem_cma_mmap(struct file *filp, struct vm_area_struct *vma); + +/* + * destroy memory region allocated. + * - a gem handle and physical memory region pointed by a gem object + * would be released by drm_gem_handle_delete(). + */ +int drm_gem_cma_dumb_destroy(struct drm_file *file_priv, + struct drm_device *drm, unsigned int handle); + +/* allocate physical memory. */ +struct drm_gem_cma_object *drm_gem_cma_create(struct drm_device *drm, + unsigned int size); + +extern const struct vm_operations_struct drm_gem_cma_vm_ops; + +#endif /* __DRM_GEM_CMA_HELPER_H__ */ diff --git a/include/drm/drm_sarea.h b/include/drm/drm_sarea.h index ee5389d22c64..1d1a858a203d 100644 --- a/include/drm/drm_sarea.h +++ b/include/drm/drm_sarea.h @@ -37,6 +37,8 @@ /* SAREA area needs to be at least a page */ #if defined(__alpha__) #define SAREA_MAX 0x2000U +#elif defined(__mips__) +#define SAREA_MAX 0x4000U #elif defined(__ia64__) #define SAREA_MAX 0x10000U /* 64kB */ #else diff --git a/include/drm/i915_drm.h b/include/drm/i915_drm.h index 8cc70837f929..e737607e055e 100644 --- a/include/drm/i915_drm.h +++ b/include/drm/i915_drm.h @@ -203,6 +203,9 @@ typedef struct _drm_i915_sarea { #define DRM_I915_GEM_WAIT 0x2c #define DRM_I915_GEM_CONTEXT_CREATE 0x2d #define DRM_I915_GEM_CONTEXT_DESTROY 0x2e +#define DRM_I915_GEM_SET_CACHING 0x2f +#define DRM_I915_GEM_GET_CACHING 0x30 +#define DRM_I915_REG_READ 0x31 #define DRM_IOCTL_I915_INIT DRM_IOW( DRM_COMMAND_BASE + DRM_I915_INIT, drm_i915_init_t) #define DRM_IOCTL_I915_FLUSH DRM_IO ( DRM_COMMAND_BASE + DRM_I915_FLUSH) @@ -227,6 +230,8 @@ typedef struct _drm_i915_sarea { #define DRM_IOCTL_I915_GEM_PIN DRM_IOWR(DRM_COMMAND_BASE + DRM_I915_GEM_PIN, struct drm_i915_gem_pin) #define DRM_IOCTL_I915_GEM_UNPIN DRM_IOW(DRM_COMMAND_BASE + DRM_I915_GEM_UNPIN, struct drm_i915_gem_unpin) #define DRM_IOCTL_I915_GEM_BUSY DRM_IOWR(DRM_COMMAND_BASE + DRM_I915_GEM_BUSY, struct drm_i915_gem_busy) +#define DRM_IOCTL_I915_GEM_SET_CACHING DRM_IOW(DRM_COMMAND_BASE + DRM_I915_GEM_SET_CACHING, struct drm_i915_gem_caching) +#define DRM_IOCTL_I915_GEM_GET_CACHING DRM_IOWR(DRM_COMMAND_BASE + DRM_I915_GEM_GET_CACHING, struct drm_i915_gem_caching) #define DRM_IOCTL_I915_GEM_THROTTLE DRM_IO ( DRM_COMMAND_BASE + DRM_I915_GEM_THROTTLE) #define DRM_IOCTL_I915_GEM_ENTERVT DRM_IO(DRM_COMMAND_BASE + DRM_I915_GEM_ENTERVT) #define DRM_IOCTL_I915_GEM_LEAVEVT DRM_IO(DRM_COMMAND_BASE + DRM_I915_GEM_LEAVEVT) @@ -249,6 +254,7 @@ typedef struct _drm_i915_sarea { #define DRM_IOCTL_I915_GEM_WAIT DRM_IOWR(DRM_COMMAND_BASE + DRM_I915_GEM_WAIT, struct drm_i915_gem_wait) #define DRM_IOCTL_I915_GEM_CONTEXT_CREATE DRM_IOWR (DRM_COMMAND_BASE + DRM_I915_GEM_CONTEXT_CREATE, struct drm_i915_gem_context_create) #define DRM_IOCTL_I915_GEM_CONTEXT_DESTROY DRM_IOW (DRM_COMMAND_BASE + DRM_I915_GEM_CONTEXT_DESTROY, struct drm_i915_gem_context_destroy) +#define DRM_IOCTL_I915_REG_READ DRM_IOWR (DRM_COMMAND_BASE + DRM_I915_REG_READ, struct drm_i915_reg_read) /* Allow drivers to submit batchbuffers directly to hardware, relying * on the security mechanisms provided by hardware. @@ -305,6 +311,9 @@ typedef struct drm_i915_irq_wait { #define I915_PARAM_HAS_LLC 17 #define I915_PARAM_HAS_ALIASING_PPGTT 18 #define I915_PARAM_HAS_WAIT_TIMEOUT 19 +#define I915_PARAM_HAS_SEMAPHORES 20 +#define I915_PARAM_HAS_PRIME_VMAP_FLUSH 21 +#define I915_PARAM_RSVD_FOR_FUTURE_USE 22 typedef struct drm_i915_getparam { int param; @@ -698,10 +707,31 @@ struct drm_i915_gem_busy { /** Handle of the buffer to check for busy */ __u32 handle; - /** Return busy status (1 if busy, 0 if idle) */ + /** Return busy status (1 if busy, 0 if idle). + * The high word is used to indicate on which rings the object + * currently resides: + * 16:31 - busy (r or r/w) rings (16 render, 17 bsd, 18 blt, etc) + */ __u32 busy; }; +#define I915_CACHING_NONE 0 +#define I915_CACHING_CACHED 1 + +struct drm_i915_gem_caching { + /** + * Handle of the buffer to set/get the caching level of. */ + __u32 handle; + + /** + * Cacheing level to apply or return value + * + * bits0-15 are for generic caching control (i.e. the above defined + * values). bits16-31 are reserved for platform-specific variations + * (e.g. l3$ caching on gen7). */ + __u32 caching; +}; + #define I915_TILING_NONE 0 #define I915_TILING_X 1 #define I915_TILING_Y 2 @@ -918,4 +948,8 @@ struct drm_i915_gem_context_destroy { __u32 pad; }; +struct drm_i915_reg_read { + __u64 offset; + __u64 val; /* Return value */ +}; #endif /* _I915_DRM_H_ */ diff --git a/include/drm/intel-gtt.h b/include/drm/intel-gtt.h index 8e29d551bb3c..2e37e9f02e71 100644 --- a/include/drm/intel-gtt.h +++ b/include/drm/intel-gtt.h @@ -30,16 +30,10 @@ void intel_gmch_remove(void); bool intel_enable_gtt(void); void intel_gtt_chipset_flush(void); -void intel_gtt_unmap_memory(struct scatterlist *sg_list, int num_sg); -void intel_gtt_clear_range(unsigned int first_entry, unsigned int num_entries); -int intel_gtt_map_memory(struct page **pages, unsigned int num_entries, - struct scatterlist **sg_list, int *num_sg); -void intel_gtt_insert_sg_entries(struct scatterlist *sg_list, - unsigned int sg_len, +void intel_gtt_insert_sg_entries(struct sg_table *st, unsigned int pg_start, unsigned int flags); -void intel_gtt_insert_pages(unsigned int first_entry, unsigned int num_entries, - struct page **pages, unsigned int flags); +void intel_gtt_clear_range(unsigned int first_entry, unsigned int num_entries); /* Special gtt memory types */ #define AGP_DCACHE_MEMORY 1 diff --git a/include/linux/platform_data/shmob_drm.h b/include/linux/platform_data/shmob_drm.h new file mode 100644 index 000000000000..7c686d335c12 --- /dev/null +++ b/include/linux/platform_data/shmob_drm.h @@ -0,0 +1,99 @@ +/* + * shmob_drm.h -- SH Mobile DRM driver + * + * Copyright (C) 2012 Renesas Corporation + * + * Laurent Pinchart (laurent.pinchart@ideasonboard.com) + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + */ + +#ifndef __SHMOB_DRM_H__ +#define __SHMOB_DRM_H__ + +#include <linux/kernel.h> + +#include <drm/drm_mode.h> + +struct sh_mobile_meram_cfg; +struct sh_mobile_meram_info; + +enum shmob_drm_clk_source { + SHMOB_DRM_CLK_BUS, + SHMOB_DRM_CLK_PERIPHERAL, + SHMOB_DRM_CLK_EXTERNAL, +}; + +enum shmob_drm_interface { + SHMOB_DRM_IFACE_RGB8, /* 24bpp, 8:8:8 */ + SHMOB_DRM_IFACE_RGB9, /* 18bpp, 9:9 */ + SHMOB_DRM_IFACE_RGB12A, /* 24bpp, 12:12 */ + SHMOB_DRM_IFACE_RGB12B, /* 12bpp */ + SHMOB_DRM_IFACE_RGB16, /* 16bpp */ + SHMOB_DRM_IFACE_RGB18, /* 18bpp */ + SHMOB_DRM_IFACE_RGB24, /* 24bpp */ + SHMOB_DRM_IFACE_YUV422, /* 16bpp */ + SHMOB_DRM_IFACE_SYS8A, /* 24bpp, 8:8:8 */ + SHMOB_DRM_IFACE_SYS8B, /* 18bpp, 8:8:2 */ + SHMOB_DRM_IFACE_SYS8C, /* 18bpp, 2:8:8 */ + SHMOB_DRM_IFACE_SYS8D, /* 16bpp, 8:8 */ + SHMOB_DRM_IFACE_SYS9, /* 18bpp, 9:9 */ + SHMOB_DRM_IFACE_SYS12, /* 24bpp, 12:12 */ + SHMOB_DRM_IFACE_SYS16A, /* 16bpp */ + SHMOB_DRM_IFACE_SYS16B, /* 18bpp, 16:2 */ + SHMOB_DRM_IFACE_SYS16C, /* 18bpp, 2:16 */ + SHMOB_DRM_IFACE_SYS18, /* 18bpp */ + SHMOB_DRM_IFACE_SYS24, /* 24bpp */ +}; + +struct shmob_drm_backlight_data { + const char *name; + int max_brightness; + int (*get_brightness)(void); + int (*set_brightness)(int brightness); +}; + +struct shmob_drm_panel_data { + unsigned int width_mm; /* Panel width in mm */ + unsigned int height_mm; /* Panel height in mm */ + struct drm_mode_modeinfo mode; +}; + +struct shmob_drm_sys_interface_data { + unsigned int read_latch:6; + unsigned int read_setup:8; + unsigned int read_cycle:8; + unsigned int read_strobe:8; + unsigned int write_setup:8; + unsigned int write_cycle:8; + unsigned int write_strobe:8; + unsigned int cs_setup:3; + unsigned int vsync_active_high:1; + unsigned int vsync_dir_input:1; +}; + +#define SHMOB_DRM_IFACE_FL_DWPOL (1 << 0) /* Rising edge dot clock data latch */ +#define SHMOB_DRM_IFACE_FL_DIPOL (1 << 1) /* Active low display enable */ +#define SHMOB_DRM_IFACE_FL_DAPOL (1 << 2) /* Active low display data */ +#define SHMOB_DRM_IFACE_FL_HSCNT (1 << 3) /* Disable HSYNC during VBLANK */ +#define SHMOB_DRM_IFACE_FL_DWCNT (1 << 4) /* Disable dotclock during blanking */ + +struct shmob_drm_interface_data { + enum shmob_drm_interface interface; + struct shmob_drm_sys_interface_data sys; + unsigned int clk_div; + unsigned int flags; +}; + +struct shmob_drm_platform_data { + enum shmob_drm_clk_source clk_source; + struct shmob_drm_interface_data iface; + struct shmob_drm_panel_data panel; + struct shmob_drm_backlight_data backlight; + const struct sh_mobile_meram_cfg *meram; +}; + +#endif /* __SHMOB_DRM_H__ */ |