| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
This structure is never used outside the file, so make it locally
scoped.
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
Use DEFINE_SHOW_ATTRIBUTE helper macro to simplify the debugfs code for
the status and status_all entries. No functional change.
Signed-off-by: Liu Shixin <liushixin2@huawei.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
Add entries for NVDEC to the Tegra234 SID table.
Signed-off-by: Mikko Perttunen <mperttunen@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
On Tegra234, engines that are programmed through Host1x channels can
be attached to either the NISO0 or NISO1 SMMU. Because of that, when
selecting a context device to use with an engine, we need to select
one that is also attached to the same SMMU.
Add a parameter to host1x_memory_context_alloc to specify which device
we are allocating a context for, and use it to pick an appropriate
context device.
Signed-off-by: Mikko Perttunen <mperttunen@nvidia.com>
[treding@nvidia.com: update !IOMMU_API stub signature]
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
Since commit c7e3ca515e78 ("iommu/tegra: gart: Do not register with
bus") quite some time ago, the GART driver has effectively disabled
itself to avoid issues with the GPU driver expecting it to work in ways
that it doesn't. As of commit 57365a04c921 ("iommu: Move bus setup to
IOMMU device registration") that bodge no longer works, but really the
GPU driver should be responsible for its own behaviour anyway. Make the
workaround explicit.
Reported-by: Jon Hunter <jonathanh@nvidia.com>
Suggested-by: Dmitry Osipenko <digetx@gmail.com>
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Tested-by: Jon Hunter <jonathanh@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
Use bitmap_zalloc()/bitmap_free() instead of hand-writing them.
It is less verbose and it improves the semantic.
While at it, remove a useless bitmap_zero() call. The bitmap is already
zero'ed when allocated.
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
host1x_cdma_push_wide() had the assumptions that the last parameter word
was a NOP opcode, and that NOP opcodes could be used in all situations.
Neither are true with the new job opcode sequence, so adjust the
function to not have these assumptions, and instead place an early
RESTART opcode when necessary to jump back to the beginning of the
pushbuffer.
Signed-off-by: Mikko Perttunen <mperttunen@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
During the refactoring of channel_submit(), assignment of syncval was
moved but it is also used in channel_submit(). Add this assignment back
to channel_submit() as well.
Signed-off-by: Mikko Perttunen <mperttunen@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
Conditional registration is a problem for other subsystems which may
unwittingly try to interact with host1x_context_device_bus_type in an
uninitialised state on non-Tegra platforms. A look under /sys/bus on a
typical system already reveals plenty of entries from enabled but
otherwise irrelevant configs, so lets keep things simple and register
our context bus unconditionally too.
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Reviewed-by: Mikko Perttunen <mperttunen@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
When MLOCK enforcement is enabled, the 0-word write currently done
is rejected by the hardware outside of an MLOCK region. As such,
on these chips, which also have the newer, more convenient RESTART_W
opcode, use that instead to skip over the timed out job.
Signed-off-by: Mikko Perttunen <mperttunen@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
With the full-featured opcode sequence using MLOCKs, we need to also
unlock those MLOCKs in the event of a timeout. However, it turns out
that on Tegra186/Tegra194, by default, we don't need to do this;
furthermore, on Tegra234 it is much simpler to do; so only implement
this on Tegra234 for the time being.
Signed-off-by: Mikko Perttunen <mperttunen@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
For new (Tegra186+) SoCs, use a new ('full-featured') job opcode
sequence that is compatible with virtualization. In particular,
the Host1x hardware in Tegra234 is more strict regarding the sequence,
requiring ACQUIRE_MLOCK-SETCLASS-SETSTREAMID opcodes to occur in
that sequence without gaps (except for SETPAYLOAD), so let's do it
properly in one go now.
Signed-off-by: Mikko Perttunen <mperttunen@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
Add device data and chip headers for Tegra234.
Signed-off-by: Mikko Perttunen <mperttunen@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
On Tegra234, each Host1x VM has 8 interrupt lines. Each syncpoint
can be configured with which interrupt line should be used for
threshold interrupt, allowing for load balancing.
For now, to keep backwards compatibility, just set all syncpoints
to the first interrupt.
Signed-off-by: Mikko Perttunen <mperttunen@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
Host1x on Tegra234 does not have a software-controllable reset line.
As such, don't bail out if we don't find one in the device tree.
Signed-off-by: Mikko Perttunen <mperttunen@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
Program virtualization tables specifying which VMs have access to which
Host1x hardware resources. Programming these has become mandatory in
Tegra234.
For now, since the driver does not operate as a Host1x hypervisor, we
basically allow access to everything to everyone.
Signed-off-by: Mikko Perttunen <mperttunen@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
Refactor 'regs' property loading using devm_platform_ioremap_*
and add loading of the 'common' region found on Tegra234.
Signed-off-by: Mikko Perttunen <mperttunen@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
Host1x class information and opcodes are unchanged or backwards
compatible across SoCs so let's not duplicate them for each one
but have them in a shared header file.
At the same time, add opcode functions for acquire/release_mlock.
Signed-off-by: Mikko Perttunen <mperttunen@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
Add code to do stream ID switching at the beginning of a job. The
stream ID is switched to the stream ID specified by the context
passed in the job structure.
Before switching the stream ID, an OP_DONE wait is done on the
channel's engine to ensure that there is no residual ongoing
work that might do DMA using the new stream ID.
Signed-off-by: Mikko Perttunen <mperttunen@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
Add code to register context devices from device tree, allocate them
out and manage their refcounts.
Signed-off-by: Mikko Perttunen <mperttunen@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
The context bus is a "dummy" bus that contains struct devices that
correspond to IOMMU contexts assigned through Host1x to processes.
Even when host1x itself is built as a module, the bus is registered
in built-in code so that the built-in ARM SMMU driver is able to
reference it.
Signed-off-by: Mikko Perttunen <mperttunen@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
When the host1x syncpts status is dumped via the debugfs, syncpts that
have been allocated but not yet used are not shown and so currently it
is not possible to see all the allocated syncpts. Update the path for
dumping the syncpt status via the debugfs to show all allocated syncpts
even if they have not been used yet. Note that when the syncpt status
is dumped by the kernel itself for debugging only the active syncpt are
shown.
Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
Buffer mappings used in job submissions are usually small and not
rapidly reused as opposed to framebuffers (which are usually large and
rapidly reused, for example when page-flipping between double-buffered
framebuffers). Avoid going through the mapping cache for these buffers
since the cache would also lead to leaks if nobody is ever releasing
the cache's last reference. For DRM/KMS these last references are
dropped when the framebuffers are removed and therefore no longer
needed.
While at it, also add a note about the need to explicitly remove the
final reference to the mapping in the cache.
Reviewed-by: Jon Hunter <jonathanh@nvidia.com>
Tested-by: Jon Hunter <jonathanh@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
Add a missing 'host1x_channel_list_free()' call in the remove function,
as already done in the error handling path of the probe function.
Fixes: 8474b02531c4 ("gpu: host1x: Refactor channel allocation code")
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
Add the missing 'host1x_bo_cache_destroy()' call in the error handling
path of the probe, as already done in the remove function.
In order to simplify the error handling, move the 'host1x_bo_cache_init()'
call after all the devm_ function.
Fixes: 1f39b1dfa53c ("drm/tegra: Implement buffer object cache")
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
The new TegraDRM UAPI uses syncpoint waiting with timeout set to
zero to indicate reading the syncpoint value. To support that we
need to return the syncpoint value always when waiting.
Fixes: 44e961381354 ("drm/tegra: Implement syncpoint wait UAPI")
Signed-off-by: Mikko Perttunen <mperttunen@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
Tegra186+ hangs if host1x hardware is disabled at a kernel boot time
because we touch hardware before runtime PM is resumed. Move sync point
assignment initialization to the RPM-resume callback. Older SoCs were
unaffected because they skip that sync point initialization.
Tested-by: Jon Hunter <jonathanh@nvidia.com> # T186
Reported-by: Jon Hunter <jonathanh@nvidia.com> # T186
Fixes: 6b6776e2ab8a ("gpu: host1x: Add initial runtime PM and OPP support")
Signed-off-by: Dmitry Osipenko <digetx@gmail.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
Host1x DMA buffer isn't mapped properly when CONFIG_ARM_DMA_USE_IOMMU=y.
The memory management code of Host1x driver has a longstanding overhaul
overdue and it's not obvious where the problem is in this case. Hence
let's add back the old workaround which we already had sometime before.
It explicitly detaches Host1x device from the offending implicit IOMMU
domain. This fixes a completely broken Host1x DMA in case of ARM32
multiplatform kernel config.
Cc: stable@vger.kernel.org
Fixes: af1cbfb9bf0f ("gpu: host1x: Support DMA mapping of buffers")
Signed-off-by: Dmitry Osipenko <digetx@gmail.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
Add host1x_channel_stop() which waits till channel becomes idle and then
stops the channel hardware. This is needed for supporting suspend/resume
by host1x drivers since the hardware state is lost after power-gating,
thus the channel needs to be stopped before client enters into suspend.
Tested-by: Peter Geis <pgwipeout@gmail.com> # Ouya T30
Tested-by: Paul Fertser <fercerpav@gmail.com> # PAZ00 T20
Tested-by: Nicolas Chauvet <kwizart@gmail.com> # PAZ00 T20 and TK1 T124
Tested-by: Matt Merhar <mattmerhar@protonmail.com> # Ouya T30
Signed-off-by: Dmitry Osipenko <digetx@gmail.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
Add runtime PM and OPP support to the Host1x driver. For the starter we
will keep host1x always-on because dynamic power management require a major
refactoring of the driver code since lot's of code paths are missing the
RPM handling and we're going to remove some of these paths in the future.
Reviewed-by: Ulf Hansson <ulf.hansson@linaro.org>
Tested-by: Peter Geis <pgwipeout@gmail.com> # Ouya T30
Tested-by: Paul Fertser <fercerpav@gmail.com> # PAZ00 T20
Tested-by: Nicolas Chauvet <kwizart@gmail.com> # PAZ00 T20 and TK1 T124
Tested-by: Matt Merhar <mattmerhar@protonmail.com> # Ouya T30
Signed-off-by: Dmitry Osipenko <digetx@gmail.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
Host1x seems to be relying on picking up dma-mapping.h transitively from
iova.h, which has no reason to include it in the first place. Fix the
former issue before we totally break things by fixing the latter one.
CC: Thierry Reding <thierry.reding@gmail.com>
CC: Mikko Perttunen <mperttunen@nvidia.com>
CC: dri-devel@lists.freedesktop.org
Signed-off-by: Robin Murphy <robin.murphy@arm.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
Linking fails when dma-buf is disabled:
ld.lld: error: undefined symbol: dma_fence_release
>>> referenced by fence.c
>>> gpu/host1x/fence.o:(host1x_syncpt_fence_enable_signaling) in archive drivers/built-in.a
>>> referenced by fence.c
>>> gpu/host1x/fence.o:(host1x_fence_signal) in archive drivers/built-in.a
>>> referenced by fence.c
>>> gpu/host1x/fence.o:(do_fence_timeout) in archive drivers/built-in.a
Fixes: 687db2207b1b ("gpu: host1x: Add DMA fence implementation")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Mikko Perttunen <mperttunen@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
Fix kernel-doc warning in host1x:
../drivers/gpu/host1x/bus.c:774: warning: Excess function parameter 'key' description in '__host1x_client_register'
Fixes: 0cfe5a6e758f ("gpu: host1x: Split up client initalization and registration")
Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
Cc: Thierry Reding <thierry.reding@gmail.com>
Cc: dri-devel@lists.freedesktop.org
Cc: linux-tegra@vger.kernel.org
Cc: David Airlie <airlied@linux.ie>
Cc: Daniel Vetter <daniel@ffwll.ch>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
Add support for booting and using NVDEC on Tegra210, Tegra186
and Tegra194 to the Host1x and TegraDRM drivers. Booting in
secure mode is not currently supported.
Signed-off-by: Mikko Perttunen <mperttunen@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
This cache is used to avoid mapping and unmapping buffer objects
unnecessarily. Mappings are cached per client and stay hot until
the buffer object is destroyed.
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
DMA-BUF requires that each device that accesses a DMA-BUF attaches to it
separately. To do so the host1x_bo_pin() and host1x_bo_unpin() functions
need to be reimplemented so that they can return a mapping, which either
represents an attachment or a map of the driver's own GEM object.
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
The memory allocated for a DMA fence could be leaked if the code failed
to allocate the waiter object. Make sure to release the fence allocation
on failure.
Reported-by: kernel test robot <lkp@intel.com>
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
The DEFINE_SPINLOCK macro creates a global spinlock symbol that is visible
to the whole kernel. This is unintended in the code, fix it.
Fixes: 687db2207b1b ("gpu: host1x: Add DMA fence implementation")
Signed-off-by: Dmitry Osipenko <digetx@gmail.com>
Reviewed-by: Mikko Perttunen <mperttunen@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
Show the values of the DMASTART and DMAEND registers when dumping status
to help with failure analysis.
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
Dumping the full CDMA push buffer takes a long time and isn't very
useful since most of the contents are not relevant. Instead only show
the CDMA push buffer entries associated with current jobs.
While at it, tweak the indentation a bit to make the output more
readable.
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
The host1x debug code uses a mix of phys_addr_t, dma_addr_t and u32 to
represent addresses. However, these addresses are always DMA addresses
so use the appropriate type.
This fixes some issues with how these addresses are displayed, because
they could be truncated in some cases and not show the full address.
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
The new UAPI will have its own firewall, and we don't want to run
the firewall in the Host1x driver for those jobs. As such, add a
parameter to host1x_job_alloc to specify if we want to skip the
firewall in the Host1x driver.
Signed-off-by: Mikko Perttunen <mperttunen@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
Add support for inserting syncpoint waits in the CDMA pushbuffer.
These waits need to be done in HOST1X class, while gather submitted
by the application execute in engine class.
Support is added by converting the gather list of job into a command
list that can include both gathers and waits. When the job is
submitted, these commands are pushed as the appropriate opcodes
on the CDMA pushbuffer.
Also supported are waits relative to the start of the job,
which are useful for jobs doing multiple things with an engine
that doesn't natively support pipelining.
While at it, use 32-bit waits on chips that support them.
Signed-off-by: Mikko Perttunen <mperttunen@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
Add a callback field to the job structure, to be called just before
the job is to be freed. This allows the job's submitter to clean
up any of its own state, like decrement runtime PM refcounts.
Signed-off-by: Mikko Perttunen <mperttunen@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
Add a new property for jobs to enable or disable recovery i.e.
CPU increments of syncpoints to max value on job timeout. This
allows for a more solid model for hanged jobs, where userspace
doesn't need to guess if a syncpoint increment happened because
the job completed, or because job timeout was triggered.
On job timeout, we stop the channel, NOP all future jobs on the
channel using the same syncpoint, mark the syncpoint as locked
and resume the channel from the next job, if any.
The future jobs are NOPed, since because we don't do the CPU
increments, the value of the syncpoint is no longer synchronized,
and any waiters would become confused if a future job incremented
the syncpoint. The syncpoint is marked locked to ensure that any
future jobs cannot increment the syncpoint either, until the
application has recognized the situation and reallocated the
syncpoint.
Signed-off-by: Mikko Perttunen <mperttunen@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
Add an implementation of dma_fences based on syncpoints. Syncpoint
interrupts are used to signal fences. Additionally, after
software signaling has been enabled, a 30 second timeout is started.
If the syncpoint threshold is not reached within this period,
the fence is signalled with an -ETIMEDOUT error code. This is to
allow fences that would never reach their syncpoint threshold to
be cleaned up. The timeout can potentially be removed in the future
after job tracking code has been refactored.
Signed-off-by: Mikko Perttunen <mperttunen@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
In some cases we may need to initialize the host1x client first before
registering it. This commit adds a new helper that will do nothing but
the initialization of the data structure.
At the same time, the initialization is removed from the registration
function. Note, however, that for simplicity we explicitly initialize
the client when the host1x_client_register() function is called, as
opposed to the low-level __host1x_client_register() function. This
allows existing callers to remain unchanged.
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
These callbacks can be used by client drivers to run code during early
init and during late exit. Early init callbacks are run prior to the
regular init callbacks while late exit callbacks run after the regular
exit callbacks.
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
Syncpoint interrupts are not working as expected on Tegra194. The
problem is that the syncpoint interrupt threshold being used is the
global interrupt threshold and not the virtual interrupt threshold.
Fix this by using the virtual interrupt threshold which aligns with
downstream.
Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
Signed-off-by: Mikko Perttunen <mperttunen@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
Move the assignment of the ref out-pointer in host1x_intr_add_action
to happen within the spinlock. With the current arrangement,
it is possible for the waiter to complete before the assignment
has happened, which breaks horribly if the waiter completion
callback tries to use the reference.
In practice, there is currently no situation where this issue can
manifest -- it was first noticed with the upcoming DMA fence
implementation patches. As such this doesn't need to be backported.
Signed-off-by: Mikko Perttunen <mperttunen@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
