diff options
374 files changed, 9948 insertions, 4614 deletions
diff --git a/Documentation/gpu/amdgpu-dc.rst b/Documentation/gpu/amdgpu-dc.rst deleted file mode 100644 index f7ff7e1309de..000000000000 --- a/Documentation/gpu/amdgpu-dc.rst +++ /dev/null @@ -1,74 +0,0 @@ -=================================== -drm/amd/display - Display Core (DC) -=================================== - -*placeholder - general description of supported platforms, what dc is, etc.* - -Because it is partially shared with other operating systems, the Display Core -Driver is divided in two pieces. - -1. **Display Core (DC)** contains the OS-agnostic components. Things like - hardware programming and resource management are handled here. -2. **Display Manager (DM)** contains the OS-dependent components. Hooks to the - amdgpu base driver and DRM are implemented here. - -It doesn't help that the entire package is frequently referred to as DC. But -with the context in mind, it should be clear. - -When CONFIG_DRM_AMD_DC is enabled, DC will be initialized by default for -supported ASICs. To force disable, set `amdgpu.dc=0` on kernel command line. -Likewise, to force enable on unsupported ASICs, set `amdgpu.dc=1`. - -To determine if DC is loaded, search dmesg for the following entry: - -``Display Core initialized with <version number here>`` - -AMDgpu Display Manager -====================== - -.. kernel-doc:: drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c - :doc: overview - -.. kernel-doc:: drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h - :internal: - -Lifecycle ---------- - -.. kernel-doc:: drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c - :doc: DM Lifecycle - -.. kernel-doc:: drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c - :functions: dm_hw_init dm_hw_fini - -Interrupts ----------- - -.. kernel-doc:: drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq.c - :doc: overview - -.. kernel-doc:: drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq.c - :internal: - -.. kernel-doc:: drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c - :functions: register_hpd_handlers dm_crtc_high_irq dm_pflip_high_irq - -Atomic Implementation ---------------------- - -.. kernel-doc:: drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c - :doc: atomic - -.. kernel-doc:: drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c - :functions: amdgpu_dm_atomic_check amdgpu_dm_atomic_commit_tail - -Display Core -============ - -**WIP** - -FreeSync Video --------------- - -.. kernel-doc:: drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c - :doc: FreeSync Video diff --git a/Documentation/gpu/amdgpu.rst b/Documentation/gpu/amdgpu.rst deleted file mode 100644 index 8ba72e898099..000000000000 --- a/Documentation/gpu/amdgpu.rst +++ /dev/null @@ -1,324 +0,0 @@ -========================= - drm/amdgpu AMDgpu driver -========================= - -The drm/amdgpu driver supports all AMD Radeon GPUs based on the Graphics Core -Next (GCN) architecture. - -Module Parameters -================= - -The amdgpu driver supports the following module parameters: - -.. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c - -Core Driver Infrastructure -========================== - -This section covers core driver infrastructure. - -.. _amdgpu_memory_domains: - -Memory Domains --------------- - -.. kernel-doc:: include/uapi/drm/amdgpu_drm.h - :doc: memory domains - -Buffer Objects --------------- - -.. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_object.c - :doc: amdgpu_object - -.. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_object.c - :internal: - -PRIME Buffer Sharing --------------------- - -.. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c - :doc: PRIME Buffer Sharing - -.. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c - :internal: - -MMU Notifier ------------- - -.. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_mn.c - :doc: MMU Notifier - -.. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_mn.c - :internal: - -AMDGPU Virtual Memory ---------------------- - -.. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c - :doc: GPUVM - -.. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c - :internal: - -Interrupt Handling ------------------- - -.. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c - :doc: Interrupt Handling - -.. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c - :internal: - -IP Blocks ------------------- - -.. kernel-doc:: drivers/gpu/drm/amd/include/amd_shared.h - :doc: IP Blocks - -.. kernel-doc:: drivers/gpu/drm/amd/include/amd_shared.h - :identifiers: amd_ip_block_type amd_ip_funcs - -AMDGPU XGMI Support -=================== - -.. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c - -AMDGPU RAS Support -================== - -The AMDGPU RAS interfaces are exposed via sysfs (for informational queries) and -debugfs (for error injection). - -RAS debugfs/sysfs Control and Error Injection Interfaces --------------------------------------------------------- - -.. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c - :doc: AMDGPU RAS debugfs control interface - -RAS Reboot Behavior for Unrecoverable Errors --------------------------------------------------------- - -.. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c - :doc: AMDGPU RAS Reboot Behavior for Unrecoverable Errors - -RAS Error Count sysfs Interface -------------------------------- - -.. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c - :doc: AMDGPU RAS sysfs Error Count Interface - -RAS EEPROM debugfs Interface ----------------------------- - -.. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c - :doc: AMDGPU RAS debugfs EEPROM table reset interface - -RAS VRAM Bad Pages sysfs Interface ----------------------------------- - -.. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c - :doc: AMDGPU RAS sysfs gpu_vram_bad_pages Interface - -Sample Code ------------ -Sample code for testing error injection can be found here: -https://cgit.freedesktop.org/mesa/drm/tree/tests/amdgpu/ras_tests.c - -This is part of the libdrm amdgpu unit tests which cover several areas of the GPU. -There are four sets of tests: - -RAS Basic Test - -The test verifies the RAS feature enabled status and makes sure the necessary sysfs and debugfs files -are present. - -RAS Query Test - -This test checks the RAS availability and enablement status for each supported IP block as well as -the error counts. - -RAS Inject Test - -This test injects errors for each IP. - -RAS Disable Test - -This test tests disabling of RAS features for each IP block. - - -GPU Power/Thermal Controls and Monitoring -========================================= - -This section covers hwmon and power/thermal controls. - -HWMON Interfaces ----------------- - -.. kernel-doc:: drivers/gpu/drm/amd/pm/amdgpu_pm.c - :doc: hwmon - -GPU sysfs Power State Interfaces --------------------------------- - -GPU power controls are exposed via sysfs files. - -power_dpm_state -~~~~~~~~~~~~~~~ - -.. kernel-doc:: drivers/gpu/drm/amd/pm/amdgpu_pm.c - :doc: power_dpm_state - -power_dpm_force_performance_level -~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ - -.. kernel-doc:: drivers/gpu/drm/amd/pm/amdgpu_pm.c - :doc: power_dpm_force_performance_level - -pp_table -~~~~~~~~ - -.. kernel-doc:: drivers/gpu/drm/amd/pm/amdgpu_pm.c - :doc: pp_table - -pp_od_clk_voltage -~~~~~~~~~~~~~~~~~ - -.. kernel-doc:: drivers/gpu/drm/amd/pm/amdgpu_pm.c - :doc: pp_od_clk_voltage - -pp_dpm_* -~~~~~~~~ - -.. kernel-doc:: drivers/gpu/drm/amd/pm/amdgpu_pm.c - :doc: pp_dpm_sclk pp_dpm_mclk pp_dpm_socclk pp_dpm_fclk pp_dpm_dcefclk pp_dpm_pcie - -pp_power_profile_mode -~~~~~~~~~~~~~~~~~~~~~ - -.. kernel-doc:: drivers/gpu/drm/amd/pm/amdgpu_pm.c - :doc: pp_power_profile_mode - -\*_busy_percent -~~~~~~~~~~~~~~~ - -.. kernel-doc:: drivers/gpu/drm/amd/pm/amdgpu_pm.c - :doc: gpu_busy_percent - -.. kernel-doc:: drivers/gpu/drm/amd/pm/amdgpu_pm.c - :doc: mem_busy_percent - -gpu_metrics -~~~~~~~~~~~~~~~~~~~~~ - -.. kernel-doc:: drivers/gpu/drm/amd/pm/amdgpu_pm.c - :doc: gpu_metrics - -GPU Product Information -======================= - -Information about the GPU can be obtained on certain cards -via sysfs - -product_name ------------- - -.. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_device.c - :doc: product_name - -product_number --------------- - -.. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_device.c - :doc: product_name - -serial_number -------------- - -.. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_device.c - :doc: serial_number - -unique_id ---------- - -.. kernel-doc:: drivers/gpu/drm/amd/pm/amdgpu_pm.c - :doc: unique_id - -GPU Memory Usage Information -============================ - -Various memory accounting can be accessed via sysfs - -mem_info_vram_total -------------------- - -.. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c - :doc: mem_info_vram_total - -mem_info_vram_used ------------------- - -.. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c - :doc: mem_info_vram_used - -mem_info_vis_vram_total ------------------------ - -.. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c - :doc: mem_info_vis_vram_total - -mem_info_vis_vram_used ----------------------- - -.. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c - :doc: mem_info_vis_vram_used - -mem_info_gtt_total ------------------- - -.. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_gtt_mgr.c - :doc: mem_info_gtt_total - -mem_info_gtt_used ------------------ - -.. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_gtt_mgr.c - :doc: mem_info_gtt_used - -PCIe Accounting Information -=========================== - -pcie_bw -------- - -.. kernel-doc:: drivers/gpu/drm/amd/pm/amdgpu_pm.c - :doc: pcie_bw - -pcie_replay_count ------------------ - -.. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_device.c - :doc: pcie_replay_count - -GPU SmartShift Information -========================== - -GPU SmartShift information via sysfs - -smartshift_apu_power --------------------- - -.. kernel-doc:: drivers/gpu/drm/amd/pm/amdgpu_pm.c - :doc: smartshift_apu_power - -smartshift_dgpu_power ---------------------- - -.. kernel-doc:: drivers/gpu/drm/amd/pm/amdgpu_pm.c - :doc: smartshift_dgpu_power - -smartshift_bias ---------------- - -.. kernel-doc:: drivers/gpu/drm/amd/pm/amdgpu_pm.c - :doc: smartshift_bias diff --git a/Documentation/gpu/amdgpu/amdgpu-glossary.rst b/Documentation/gpu/amdgpu/amdgpu-glossary.rst new file mode 100644 index 000000000000..859dcec6c6f9 --- /dev/null +++ b/Documentation/gpu/amdgpu/amdgpu-glossary.rst @@ -0,0 +1,87 @@ +=============== +AMDGPU Glossary +=============== + +Here you can find some generic acronyms used in the amdgpu driver. Notice that +we have a dedicated glossary for Display Core at +'Documentation/gpu/amdgpu/display/dc-glossary.rst'. + +.. glossary:: + + CP + Command Processor + + CPLIB + Content Protection Library + + DFS + Digital Frequency Synthesizer + + ECP + Enhanced Content Protection + + EOP + End Of Pipe/Pipeline + + GC + Graphics and Compute + + GMC + Graphic Memory Controller + + IH + Interrupt Handler + + HQD + Hardware Queue Descriptor + + IB + Indirect Buffer + + IP + Intellectual Property blocks + + KCQ + Kernel Compute Queue + + KGQ + Kernel Graphics Queue + + KIQ + Kernel Interface Queue + + MEC + MicroEngine Compute + + MES + MicroEngine Scheduler + + MMHUB + Multi-Media HUB + + MQD + Memory Queue Descriptor + + PPLib + PowerPlay Library - PowerPlay is the power management component. + + PSP + Platform Security Processor + + RCL + RunList Controller + + SDMA + System DMA + + SMU + System Management Unit + + SS + Spread Spectrum + + VCE + Video Compression Engine + + VCN + Video Codec Next diff --git a/Documentation/gpu/amdgpu/display/config_example.svg b/Documentation/gpu/amdgpu/display/config_example.svg new file mode 100644 index 000000000000..cdac9858601c --- /dev/null +++ b/Documentation/gpu/amdgpu/display/config_example.svg @@ -0,0 +1,414 @@ +<?xml version="1.0" encoding="UTF-8" standalone="no"?> +<!-- Created with Inkscape (http://www.inkscape.org/) --> + +<svg + xmlns:dc="http://purl.org/dc/elements/1.1/" + xmlns:cc="http://creativecommons.org/ns#" + xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" + xmlns:svg="http://www.w3.org/2000/svg" + xmlns="http://www.w3.org/2000/svg" + xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd" + xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape" + width="144.63406mm" + height="66.596054mm" + viewBox="0 0 144.15195 66.596054" + version="1.1" + id="svg8" + inkscape:version="0.92.5 (2060ec1f9f, 2020-04-08)" + sodipodi:docname="config_example.svg"> + <defs + id="defs2"> + <marker + inkscape:stockid="Arrow1Mend" + orient="auto" + refY="0" + refX="0" + id="Arrow1Mend" + style="overflow:visible" + inkscape:isstock="true"> + <path + id="path4547" + d="M 0,0 5,-5 -12.5,0 5,5 Z" + style="fill:#ff0000;fill-opacity:1;fill-rule:evenodd;stroke:#ff0000;stroke-width:1.00000003pt;stroke-opacity:1" + transform="matrix(-0.4,0,0,-0.4,-4,0)" + inkscape:connector-curvature="0" /> + </marker> + <marker + inkscape:stockid="Arrow1Mend" + orient="auto" + refY="0" + refX="0" + id="Arrow1Mend-3" + style="overflow:visible" + inkscape:isstock="true"> + <path + inkscape:connector-curvature="0" + id="path4547-6" + d="M 0,0 5,-5 -12.5,0 5,5 Z" + style="fill:#ff0000;fill-opacity:1;fill-rule:evenodd;stroke:#ff0000;stroke-width:1.00000003pt;stroke-opacity:1" + transform="matrix(-0.4,0,0,-0.4,-4,0)" /> + </marker> + <marker + inkscape:stockid="Arrow1Mend" + orient="auto" + refY="0" + refX="0" + id="Arrow1Mend-3-5" + style="overflow:visible" + inkscape:isstock="true"> + <path + inkscape:connector-curvature="0" + id="path4547-6-3" + d="M 0,0 5,-5 -12.5,0 5,5 Z" + style="fill:#ff0000;fill-opacity:1;fill-rule:evenodd;stroke:#ff0000;stroke-width:1.00000003pt;stroke-opacity:1" + transform="matrix(-0.4,0,0,-0.4,-4,0)" /> + </marker> + <marker + inkscape:stockid="Arrow1Mend" + orient="auto" + refY="0" + refX="0" + id="Arrow1Mend-3-5-0" + style="overflow:visible" + inkscape:isstock="true"> + <path + inkscape:connector-curvature="0" + id="path4547-6-3-6" + d="M 0,0 5,-5 -12.5,0 5,5 Z" + style="fill:#ff0000;fill-opacity:1;fill-rule:evenodd;stroke:#ff0000;stroke-width:1.00000003pt;stroke-opacity:1" + transform="matrix(-0.4,0,0,-0.4,-4,0)" /> + </marker> + <marker + inkscape:stockid="Arrow1Mend" + orient="auto" + refY="0" + refX="0" + id="Arrow1Mend-3-5-7" + style="overflow:visible" + inkscape:isstock="true"> + <path + inkscape:connector-curvature="0" + id="path4547-6-3-3" + d="M 0,0 5,-5 -12.5,0 5,5 Z" + style="fill:#ff0000;fill-opacity:1;fill-rule:evenodd;stroke:#ff0000;stroke-width:1.00000003pt;stroke-opacity:1" + transform="matrix(-0.4,0,0,-0.4,-4,0)" /> + </marker> + </defs> + <sodipodi:namedview + id="base" + pagecolor="#ffffff" + bordercolor="#666666" + borderopacity="1.0" + inkscape:pageopacity="0.0" + inkscape:pageshadow="2" + inkscape:zoom="0.98994949" + inkscape:cx="518.91791" + inkscape:cy="172.50112" + inkscape:document-units="mm" + inkscape:current-layer="layer1" + showgrid="true" + viewbox-width="209.3" + inkscape:window-width="3840" + inkscape:window-height="1136" + inkscape:window-x="0" + inkscape:window-y="27" + inkscape:window-maximized="1" + fit-margin-top="0" + fit-margin-left="0" + fit-margin-right="0" + fit-margin-bottom="0"> + <inkscape:grid + type="xygrid" + id="grid817" + originx="4.390216" + originy="-208.88856" /> + </sodipodi:namedview> + <metadata + id="metadata5"> + <rdf:RDF> + <cc:Work + rdf:about=""> + <dc:format>image/svg+xml</dc:format> + <dc:type + rdf:resource="http://purl.org/dc/dcmitype/StillImage" /> + <dc:title></dc:title> + </cc:Work> + </rdf:RDF> + </metadata> + <g + inkscape:label="Layer 1" + inkscape:groupmode="layer" + id="layer1" + transform="translate(4.4048992,-21.515392)"> + <path + style="fill:none;stroke:#000000;stroke-width:0.26458335px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1" + d="m 20.816662,35.062492 h 23.8125 v -5.291667 h 5.291667 v 5.291667 h 10.583334 v -5.291667 h 5.291667 v 5.291667 h 2.645833 v -5.291667 h 5.291667 v 5.291667 h 66.14583" + id="path4522" + inkscape:connector-curvature="0" /> + <path + style="fill:none;stroke:#000000;stroke-width:0.26458335px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1" + d="m 20.816662,48.291659 h 7.9375 v -5.291667 h 5.291667 v 5.291667 h 58.208335 v -5.291667 h 5.291666 v 5.291667 h 42.33333" + id="path4524" + inkscape:connector-curvature="0" /> + <path + style="fill:none;stroke:#000000;stroke-width:0.26458335px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1" + d="m 20.816662,61.520826 h 26.458334 v -5.291667 h 44.979168 v 5.291667 h 47.624996" + id="path4526" + inkscape:connector-curvature="0" /> + <path + style="fill:none;stroke:#000000;stroke-width:0.26458335px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1" + d="M 20.816662,72.104159 H 139.87916" + id="path4528" + inkscape:connector-curvature="0" /> + <path + style="fill:none;stroke:#000000;stroke-width:0.26458335px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1" + d="M 20.816662,77.395826 H 139.87916" + id="path4530" + inkscape:connector-curvature="0" /> + <path + style="fill:none;stroke:#000000;stroke-width:0.26458335px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1" + d="M 20.816662,82.687493 H 139.87916" + id="path4532" + inkscape:connector-curvature="0" /> + <path + style="fill:none;stroke:#000000;stroke-width:0.26458335px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1" + d="M 20.816662,87.97916 H 139.87916" + id="path4534" + inkscape:connector-curvature="0" /> + <path + style="fill:none;stroke:#ff0000;stroke-width:0.5291667;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:0.52916668, 0.52916668;stroke-dashoffset:0;stroke-opacity:1;marker-end:url(#Arrow1Mend)" + d="m 47.274996,29.770826 c 3.836215,14.933158 3.472151,27.586643 0.264583,41.010418" + id="path4536" + inkscape:connector-curvature="0" + sodipodi:nodetypes="cc" /> + <path + style="fill:none;stroke:#ff0000;stroke-width:0.5291667;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:0.52916669, 0.52916669;stroke-dashoffset:0;stroke-opacity:1;marker-end:url(#Arrow1Mend-3)" + d="m 63.149996,29.770826 c 3.836214,14.933158 5.059652,27.586642 1.852084,41.010418" + id="path4536-7" + inkscape:connector-curvature="0" + sodipodi:nodetypes="cc" /> + <path + style="fill:none;stroke:#ff0000;stroke-width:0.5291667;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:0.5291667, 0.5291667;stroke-dashoffset:0;stroke-opacity:1;marker-end:url(#Arrow1Mend-3-5)" + d="m 71.087496,29.770825 c 3.836214,14.933158 5.059652,27.586643 1.852084,41.010419" + id="path4536-7-5" + inkscape:connector-curvature="0" + sodipodi:nodetypes="cc" /> + <text + xml:space="preserve" + style="font-style:normal;font-weight:normal;font-size:10.58333397px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.26458335" + x="59.359009" + y="24.195677" + id="text6572"><tspan + sodipodi:role="line" + x="59.359009" + y="24.195677" + style="font-size:3.52777791px;line-height:5.39999962;text-align:center;text-anchor:middle;stroke-width:0.26458335" + id="tspan6574">Configurations</tspan></text> + <text + xml:space="preserve" + style="font-style:normal;font-weight:normal;font-size:10.58333397px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.26458335" + x="46.825508" + y="28.542402" + id="text6572-6"><tspan + sodipodi:role="line" + x="46.825508" + y="28.542402" + style="font-size:3.52777815px;line-height:5.39999962;text-align:center;text-anchor:middle;fill:#008000;stroke-width:0.26458335" + id="tspan6574-2">A</tspan></text> + <text + xml:space="preserve" + style="font-style:normal;font-weight:normal;font-size:10.58333397px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.26458335" + x="62.8895" + y="28.825886" + id="text6572-6-2"><tspan + sodipodi:role="line" + x="62.8895" + y="28.825886" + style="font-size:3.52777839px;line-height:5.39999962;text-align:center;text-anchor:middle;fill:#0000ff;stroke-width:0.26458335" + id="tspan6574-2-7">B</tspan></text> + <text + xml:space="preserve" + style="font-style:normal;font-weight:normal;font-size:10.58333397px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.26458335" + x="70.827003" + y="29.109362" + id="text6572-6-2-3"><tspan + sodipodi:role="line" + x="70.827003" + y="29.109362" + style="font-size:3.52777863px;line-height:5.39999962;text-align:center;text-anchor:middle;fill:#c87137;stroke-width:0.26458335" + id="tspan6574-2-7-6">C</tspan></text> + <path + style="fill:none;stroke:#ff0000;stroke-width:0.5291667;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:0.52916671, 0.52916671;stroke-dashoffset:0;stroke-opacity:1;marker-end:url(#Arrow1Mend-3-5-0)" + d="m 92.254164,42.999993 c 9.142136,12.745655 4.411987,28.608461 0.529167,38.364584" + id="path4536-7-5-2" + inkscape:connector-curvature="0" + sodipodi:nodetypes="cc" /> + <path + style="fill:none;stroke:#000000;stroke-width:0.5291667;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1" + d="m 47.274996,72.104159 v 5.291667" + id="path8053" + inkscape:connector-curvature="0" /> + <path + style="fill:none;stroke:#000000;stroke-width:0.5291667;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1" + d="m 64.472913,72.10416 v 5.291667" + id="path8053-6" + inkscape:connector-curvature="0" /> + <path + style="fill:none;stroke:#000000;stroke-width:0.5291667;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1" + d="m 72.410413,72.10416 v 5.291667" + id="path8053-6-1" + inkscape:connector-curvature="0" /> + <path + style="fill:none;stroke:#000000;stroke-width:0.5291667;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1" + d="m 92.254164,82.687494 v 5.291667" + id="path8053-6-1-8" + inkscape:connector-curvature="0" /> + <text + xml:space="preserve" + style="font-style:normal;font-weight:normal;font-size:10.58333397px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.26458335" + x="55.802444" + y="76.167412" + id="text6572-6-7"><tspan + sodipodi:role="line" + x="55.802444" + y="76.167412" + style="font-size:3.52777839px;line-height:5.39999962;text-align:center;text-anchor:middle;fill:#008000;stroke-width:0.26458335" + id="tspan6574-2-9">A</tspan></text> + <text + xml:space="preserve" + style="font-style:normal;font-weight:normal;font-size:10.58333397px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.26458335" + x="68.559143" + y="75.883926" + id="text6572-6-2-2"><tspan + sodipodi:role="line" + x="68.559143" + y="75.883926" + style="font-size:3.52777863px;line-height:5.39999962;text-align:center;text-anchor:middle;fill:#0000ff;stroke-width:0.26458335" + id="tspan6574-2-7-0">B</tspan></text> + <text + xml:space="preserve" + style="font-style:normal;font-weight:normal;font-size:10.58333397px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.26458335" + x="84.812119" + y="75.883911" + id="text6572-6-2-3-2"><tspan + sodipodi:role="line" + x="84.812119" + y="75.883911" + style="font-size:3.52777863px;line-height:5.39999962;text-align:center;text-anchor:middle;fill:#c87137;stroke-width:0.26458335" + id="tspan6574-2-7-6-3">C</tspan></text> + <text + xml:space="preserve" + style="font-style:normal;font-weight:normal;font-size:10.58333397px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.26458335" + x="98.513756" + y="86.845222" + id="text6572-6-2-3-2-7"><tspan + sodipodi:role="line" + x="98.513756" + y="86.845222" + style="font-size:3.52777863px;line-height:5.39999962;text-align:center;text-anchor:middle;fill:#c87137;stroke-width:0.26458335" + id="tspan6574-2-7-6-3-5">C</tspan></text> + <text + xml:space="preserve" + style="font-style:normal;font-weight:normal;font-size:10.58333397px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.26458335" + x="35.452015" + y="75.694931" + id="text6572-9"><tspan + sodipodi:role="line" + x="35.452015" + y="75.694931" + style="font-size:3.52777815px;line-height:5.39999962;text-align:center;text-anchor:middle;stroke-width:0.26458335" + id="tspan6574-22">Old config</tspan></text> + <text + xml:space="preserve" + style="font-style:normal;font-weight:normal;font-size:10.58333397px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.26458335" + x="55.484753" + y="86.656235" + id="text6572-9-8"><tspan + sodipodi:role="line" + x="55.484753" + y="86.656235" + style="font-size:3.52777839px;line-height:5.39999962;text-align:center;text-anchor:middle;stroke-width:0.26458335" + id="tspan6574-22-9">Old config</tspan></text> + <path + style="fill:none;stroke:#ff0000;stroke-width:0.5291667;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:0.52916671, 0.52916671;stroke-dashoffset:0;stroke-opacity:1;marker-end:url(#Arrow1Mend-3-5-7)" + d="m 92.254164,42.999993 c 4.233333,4.7625 2.645833,13.229167 0.79375,17.197917" + inkscape:connector-curvature="0" + sodipodi:nodetypes="cc" /> + <text + xml:space="preserve" + style="font-style:normal;font-weight:normal;font-size:10.58333397px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.26458335" + x="3.7020128" + y="33.550579" + id="text6572-1"><tspan + sodipodi:role="line" + x="3.7020128" + y="42.914349" + style="font-size:3.52777815px;line-height:5.39999962;text-align:center;text-anchor:middle;stroke-width:0.26458335" + id="tspan15310" /></text> + <text + xml:space="preserve" + style="font-style:normal;font-weight:normal;font-size:3.17500019px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.26458335" + x="13.366468" + y="46.590767" + id="text15316"><tspan + sodipodi:role="line" + x="13.366468" + y="46.590767" + style="text-align:center;text-anchor:middle;stroke-width:0.26458335" + id="tspan15318">VUpdate</tspan></text> + <text + xml:space="preserve" + style="font-style:normal;font-weight:normal;font-size:3.17500043px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.26458335" + x="14.45245" + y="29.676321" + id="text15316-3"><tspan + sodipodi:role="line" + id="tspan15314-1" + x="14.45245" + y="29.676321" + style="text-align:center;text-anchor:middle;stroke-width:0.26458335">Update</tspan><tspan + sodipodi:role="line" + x="14.45245" + y="33.645073" + style="text-align:center;text-anchor:middle;stroke-width:0.26458335" + id="tspan15318-9">Lock</tspan></text> + <text + xml:space="preserve" + style="font-style:normal;font-weight:normal;font-size:3.17500043px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.26458335" + x="7.5676007" + y="56.985115" + id="text15316-4"><tspan + sodipodi:role="line" + x="7.5676007" + y="56.985115" + style="text-align:center;text-anchor:middle;stroke-width:0.26458335" + id="tspan15318-7">Register update</tspan><tspan + sodipodi:role="line" + x="7.5676007" + y="60.953865" + style="text-align:center;text-anchor:middle;stroke-width:0.26458335" + id="tspan15361">Pending Status</tspan></text> + <text + xml:space="preserve" + style="font-style:normal;font-weight:normal;font-size:3.17500043px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.26458335" + x="16.074829" + y="76.167404" + id="text15316-8"><tspan + sodipodi:role="line" + x="16.074829" + y="76.167404" + style="text-align:center;text-anchor:middle;stroke-width:0.26458335" + id="tspan15318-4">Buf 0</tspan></text> + <text + xml:space="preserve" + style="font-style:normal;font-weight:normal;font-size:3.17500067px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.26458335" + x="16.156994" + y="86.089279" + id="text15316-8-5"><tspan + sodipodi:role="line" + x="16.156994" + y="86.089279" + style="text-align:center;text-anchor:middle;stroke-width:0.26458335" + id="tspan15318-4-0">Buf 1</tspan></text> + </g> +</svg> diff --git a/Documentation/gpu/amdgpu/display/dc-debug.rst b/Documentation/gpu/amdgpu/display/dc-debug.rst new file mode 100644 index 000000000000..40c55a618918 --- /dev/null +++ b/Documentation/gpu/amdgpu/display/dc-debug.rst @@ -0,0 +1,77 @@ +======================== +Display Core Debug tools +======================== + +DC Visual Confirmation +====================== + +Display core provides a feature named visual confirmation, which is a set of +bars added at the scanout time by the driver to convey some specific +information. In general, you can enable this debug option by using:: + + echo <N> > /sys/kernel/debug/dri/0/amdgpu_dm_visual_confirm + +Where `N` is an integer number for some specific scenarios that the developer +wants to enable, you will see some of these debug cases in the following +subsection. + +Multiple Planes Debug +--------------------- + +If you want to enable or debug multiple planes in a specific user-space +application, you can leverage a debug feature named visual confirm. For +enabling it, you will need:: + + echo 1 > /sys/kernel/debug/dri/0/amdgpu_dm_visual_confirm + +You need to reload your GUI to see the visual confirmation. When the plane +configuration changes or a full update occurs there will be a colored bar at +the bottom of each hardware plane being drawn on the screen. + +* The color indicates the format - For example, red is AR24 and green is NV12 +* The height of the bar indicates the index of the plane +* Pipe split can be observed if there are two bars with a difference in height + covering the same plane + +Consider the video playback case in which a video is played in a specific +plane, and the desktop is drawn in another plane. The video plane should +feature one or two green bars at the bottom of the video depending on pipe +split configuration. + +* There should **not** be any visual corruption +* There should **not** be any underflow or screen flashes +* There should **not** be any black screens +* There should **not** be any cursor corruption +* Multiple plane **may** be briefly disabled during window transitions or + resizing but should come back after the action has finished + +Pipe Split Debug +---------------- + +Sometimes we need to debug if DCN is splitting pipes correctly, and visual +confirmation is also handy for this case. Similar to the MPO case, you can use +the below command to enable visual confirmation:: + + echo 1 > /sys/kernel/debug/dri/0/amdgpu_dm_visual_confirm + +In this case, if you have a pipe split, you will see one small red bar at the +bottom of the display covering the entire display width and another bar +covering the second pipe. In other words, you will see a bit high bar in the +second pipe. + +DTN Debug +========= + +DC (DCN) provides an extensive log that dumps multiple details from our +hardware configuration. Via debugfs, you can capture those status values by +using Display Test Next (DTN) log, which can be captured via debugfs by using:: + + cat /sys/kernel/debug/dri/0/amdgpu_dm_dtn_log + +Since this log is updated accordingly with DCN status, you can also follow the +change in real-time by using something like:: + + sudo watch -d cat /sys/kernel/debug/dri/0/amdgpu_dm_dtn_log + +When reporting a bug related to DC, consider attaching this log before and +after you reproduce the bug. diff --git a/Documentation/gpu/amdgpu/display/dc-glossary.rst b/Documentation/gpu/amdgpu/display/dc-glossary.rst new file mode 100644 index 000000000000..116f5f0942fd --- /dev/null +++ b/Documentation/gpu/amdgpu/display/dc-glossary.rst @@ -0,0 +1,237 @@ +=========== +DC Glossary +=========== + +On this page, we try to keep track of acronyms related to the display +component. If you do not find what you are looking for, look at the +'Documentation/gpu/amdgpu/amdgpu-glossary.rst'; if you cannot find it anywhere, +consider asking in the amdgfx and update this page. + +.. glossary:: + + ABM + Adaptive Backlight Modulation + + APU + Accelerated Processing Unit + + ASIC + Application-Specific Integrated Circuit + + ASSR + Alternate Scrambler Seed Reset + + AZ + Azalia (HD audio DMA engine) + + BPC + Bits Per Colour/Component + + BPP + Bits Per Pixel + + Clocks + * PCLK: Pixel Clock + * SYMCLK: Symbol Clock + * SOCCLK: GPU Engine Clock + * DISPCLK: Display Clock + * DPPCLK: DPP Clock + * DCFCLK: Display Controller Fabric Clock + * REFCLK: Real Time Reference Clock + * PPLL: Pixel PLL + * FCLK: Fabric Clock + * MCLK: Memory Clock + + CRC + Cyclic Redundancy Check + + CRTC + Cathode Ray Tube Controller - commonly called "Controller" - Generates + raw stream of pixels, clocked at pixel clock + + CVT + Coordinated Video Timings + + DAL + Display Abstraction layer + + DC (Software) + Display Core + + DC (Hardware) + Display Controller + + DCC + Delta Colour Compression + + DCE + Display Controller Engine + + DCHUB + Display Controller HUB + + ARB + Arbiter + + VTG + Vertical Timing Generator + + DCN + Display Core Next + + DCCG + Display Clock Generator block + + DDC + Display Data Channel + + DIO + Display IO + + DPP + Display Pipes and Planes + + DSC + Display Stream Compression (Reduce the amount of bits to represent pixel + count while at the same pixel clock) + + dGPU + discrete GPU + + DMIF + Display Memory Interface + + DML + Display Mode Library + + DMCU + Display Micro-Controller Unit + + DMCUB + Display Micro-Controller Unit, version B + + DPCD + DisplayPort Configuration Data + + DPM(S) + Display Power Management (Signaling) + + DRR + Dynamic Refresh Rate + + DWB + Display Writeback + + FB + Frame Buffer + + FBC + Frame Buffer Compression + + FEC + Forward Error Correction + + FRL + Fixed Rate Link + + GCO + Graphical Controller Object + + GSL + Global Swap Lock + + iGPU + integrated GPU + + ISR + Interrupt Service Request + + ISV + Independent Software Vendor + + KMD + Kernel Mode Driver + + LB + Line Buffer + + LFC + Low Framerate Compensation + + LTTPR + Link Training Tunable Phy Repeater + + LUT + Lookup Table + + MALL + Memory Access at Last Level + + MC + Memory Controller + + MPC + Multiple pipes and plane combine + + MPO + Multi Plane Overlay + + MST + Multi Stream Transport + + NBP State + Northbridge Power State + + NBIO + North Bridge Input/Output + + ODM + Output Data Mapping + + OPM + Output Protection Manager + + OPP + Output Plane Processor + + OPTC + Output Pipe Timing Combiner + + OTG + Output Timing Generator + + PCON + Power Controller + + PGFSM + Power Gate Finite State Machine + + PSR + Panel Self Refresh + + SCL + Scaler + + SDP + Scalable Data Port + + SLS + Single Large Surface + + SST + Single Stream Transport + + TMDS + Transition-Minimized Differential Signaling + + TMZ + Trusted Memory Zone + + TTU + Time to Underflow + + VRR + Variable Refresh Rate + + UVD + Unified Video Decoder diff --git a/Documentation/gpu/amdgpu/display/dc_pipeline_overview.svg b/Documentation/gpu/amdgpu/display/dc_pipeline_overview.svg new file mode 100644 index 000000000000..9adecebfe65b --- /dev/null +++ b/Documentation/gpu/amdgpu/display/dc_pipeline_overview.svg @@ -0,0 +1,1125 @@ +<?xml version="1.0" encoding="UTF-8" standalone="no"?> +<!-- Created with Inkscape (http://www.inkscape.org/) --> + +<svg + xmlns:dc="http://purl.org/dc/elements/1.1/" + xmlns:cc="http://creativecommons.org/ns#" + xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" + xmlns:svg="http://www.w3.org/2000/svg" + xmlns="http://www.w3.org/2000/svg" + xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd" + xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape" + width="1296.7491" + height="741.97845" + viewBox="0 0 343.0982 196.31514" + version="1.1" + id="svg8" + inkscape:version="0.92.5 (2060ec1f9f, 2020-04-08)" + sodipodi:docname="dc_pipeline_overview.svg"> + <defs + id="defs2"> + <marker + inkscape:stockid="Arrow2Mend" + orient="auto" + refY="0" + refX="0" + id="marker8858" + style="overflow:visible" + inkscape:isstock="true"> + <path + id="path8616" + style="fill:#aa00d4;fill-opacity:1;fill-rule:evenodd;stroke:#aa00d4;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1" + d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z" + transform="scale(-0.6)" + inkscape:connector-curvature="0" /> + </marker> + <marker + inkscape:stockid="Arrow2Send" + orient="auto" + refY="0" + refX="0" + id="Arrow2Send" + style="overflow:visible" + inkscape:isstock="true"> + <path + id="path8622" + style="fill:#ff0000;fill-opacity:1;fill-rule:evenodd;stroke:#ff0000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1" + d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z" + transform="matrix(-0.3,0,0,-0.3,0.69,0)" + inkscape:connector-curvature="0" /> + </marker> + <marker + inkscape:stockid="Arrow1Lend" + orient="auto" + refY="0" + refX="0" + id="Arrow1Lend" + style="overflow:visible" + inkscape:isstock="true"> + <path + id="path8592" + d="M 0,0 5,-5 -12.5,0 5,5 Z" + style="fill:#ff0000;fill-opacity:1;fill-rule:evenodd;stroke:#ff0000;stroke-width:1.00000003pt;stroke-opacity:1" + transform="matrix(-0.8,0,0,-0.8,-10,0)" + inkscape:connector-curvature="0" /> + </marker> + <marker + inkscape:stockid="Arrow2Lend" + orient="auto" + refY="0" + refX="0" + id="Arrow2Lend" + style="overflow:visible" + inkscape:isstock="true"> + <path + id="path8610" + style="fill:#ff0000;fill-opacity:1;fill-rule:evenodd;stroke:#ff0000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1" + d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z" + transform="matrix(-1.1,0,0,-1.1,-1.1,0)" + inkscape:connector-curvature="0" /> + </marker> + <marker + inkscape:stockid="Arrow2Mend" + orient="auto" + refY="0" + refX="0" + id="Arrow2Mend" + style="overflow:visible" + inkscape:isstock="true"> + <path + id="path1200" + style="fill:#ff0000;fill-opacity:1;fill-rule:evenodd;stroke:#ff0000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1" + d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z" + transform="scale(-0.6)" + inkscape:connector-curvature="0" /> + </marker> + <marker + inkscape:stockid="Arrow2Mend" + orient="auto" + refY="0" + refX="0" + id="Arrow2Mend-8" + style="overflow:visible" + inkscape:isstock="true"> + <path + inkscape:connector-curvature="0" + id="path1200-9" + style="fill:#ff0000;fill-opacity:1;fill-rule:evenodd;stroke:#ff0000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1" + d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z" + transform="scale(-0.6)" /> + </marker> + <marker + inkscape:stockid="Arrow2Mend" + orient="auto" + refY="0" + refX="0" + id="Arrow2Mend-8-3" + style="overflow:visible" + inkscape:isstock="true"> + <path + inkscape:connector-curvature="0" + id="path1200-9-6" + style="fill:#ff0000;fill-opacity:1;fill-rule:evenodd;stroke:#ff0000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1" + d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z" + transform="scale(-0.6)" /> + </marker> + <marker + inkscape:stockid="Arrow2Mend" + orient="auto" + refY="0" + refX="0" + id="Arrow2Mend-8-3-2" + style="overflow:visible" + inkscape:isstock="true"> + <path + inkscape:connector-curvature="0" + id="path1200-9-6-9" + style="fill:#ff0000;fill-opacity:1;fill-rule:evenodd;stroke:#ff0000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1" + d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z" + transform="scale(-0.6)" /> + </marker> + <marker + inkscape:stockid="Arrow2Mend" + orient="auto" + refY="0" + refX="0" + id="Arrow2Mend-8-3-2-1" + style="overflow:visible" + inkscape:isstock="true"> + <path + inkscape:connector-curvature="0" + id="path1200-9-6-9-9" + style="fill:#ff0000;fill-opacity:1;fill-rule:evenodd;stroke:#ff0000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1" + d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z" + transform="scale(-0.6)" /> + </marker> + <marker + inkscape:stockid="Arrow2Mend" + orient="auto" + refY="0" + refX="0" + id="Arrow2Mend-8-3-2-7" + style="overflow:visible" + inkscape:isstock="true"> + <path + inkscape:connector-curvature="0" + id="path1200-9-6-9-8" + style="fill:#008000;fill-opacity:1;fill-rule:evenodd;stroke:#008000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1" + d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z" + transform="scale(-0.6)" /> + </marker> + <marker + inkscape:stockid="Arrow2Mend" + orient="auto" + refY="0" + refX="0" + id="Arrow2Mend-8-3-4" + style="overflow:visible" + inkscape:isstock="true"> + <path + inkscape:connector-curvature="0" + id="path1200-9-6-5" + style="fill:#008000;fill-opacity:1;fill-rule:evenodd;stroke:#008000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1" + d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z" + transform="scale(-0.6)" /> + </marker> + <marker + inkscape:stockid="Arrow2Mend" + orient="auto" + refY="0" + refX="0" + id="Arrow2Mend-8-0" + style="overflow:visible" + inkscape:isstock="true"> + <path + inkscape:connector-curvature="0" + id="path1200-9-3" + style="fill:#008000;fill-opacity:1;fill-rule:evenodd;stroke:#008000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1" + d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z" + transform="scale(-0.6)" /> + </marker> + <marker + inkscape:stockid="Arrow2Mend" + orient="auto" + refY="0" + refX="0" + id="Arrow2Mend-6" + style="overflow:visible" + inkscape:isstock="true"> + <path + inkscape:connector-curvature="0" + id="path1200-1" + style="fill:#008000;fill-opacity:1;fill-rule:evenodd;stroke:#008000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1" + d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z" + transform="scale(-0.6)" /> + </marker> + <marker + inkscape:stockid="Arrow2Mend" + orient="auto" + refY="0" + refX="0" + id="Arrow2Mend-8-3-2-6" + style="overflow:visible" + inkscape:isstock="true"> + <path + inkscape:connector-curvature="0" + id="path1200-9-6-9-1" + style="fill:#ff0000;fill-opacity:1;fill-rule:evenodd;stroke:#ff0000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1" + d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z" + transform="scale(-0.6)" /> + </marker> + <marker + inkscape:stockid="Arrow2Mend" + orient="auto" + refY="0" + refX="0" + id="Arrow2Mend-8-0-7" + style="overflow:visible" + inkscape:isstock="true"> + <path + inkscape:connector-curvature="0" + id="path1200-9-3-4" + style="fill:#008000;fill-opacity:1;fill-rule:evenodd;stroke:#008000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1" + d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z" + transform="scale(-0.6)" /> + </marker> + <marker + inkscape:stockid="Arrow2Mend" + orient="auto" + refY="0" + refX="0" + id="Arrow2Mend-6-3" + style="overflow:visible" + inkscape:isstock="true"> + <path + inkscape:connector-curvature="0" + id="path1200-1-0" + style="fill:#008000;fill-opacity:1;fill-rule:evenodd;stroke:#008000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1" + d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z" + transform="scale(-0.6)" /> + </marker> + <marker + inkscape:stockid="Arrow2Mend" + orient="auto" + refY="0" + refX="0" + id="Arrow2Mend-8-3-2-8" + style="overflow:visible" + inkscape:isstock="true"> + <path + inkscape:connector-curvature="0" + id="path1200-9-6-9-6" + style="fill:#ff0000;fill-opacity:1;fill-rule:evenodd;stroke:#ff0000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1" + d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z" + transform="scale(-0.6)" /> + </marker> + <marker + inkscape:stockid="Arrow2Mend" + orient="auto" + refY="0" + refX="0" + id="Arrow2Mend-3" + style="overflow:visible" + inkscape:isstock="true"> + <path + id="path1200-6" + style="fill:#ff0000;fill-opacity:1;fill-rule:evenodd;stroke:#ff0000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1" + d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z" + transform="scale(-0.6)" + inkscape:connector-curvature="0" /> + </marker> + <marker + inkscape:stockid="Arrow2Mend" + orient="auto" + refY="0" + refX="0" + id="marker8858-3" + style="overflow:visible" + inkscape:isstock="true"> + <path + id="path8616-5" + style="fill:#00ffcc;fill-opacity:1;fill-rule:evenodd;stroke:#00ffcc;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1" + d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z" + transform="scale(-0.6)" + inkscape:connector-curvature="0" /> + </marker> + <marker + inkscape:stockid="Arrow2Mend" + orient="auto" + refY="0" + refX="0" + id="Arrow2Mend-8-3-3" + style="overflow:visible" + inkscape:isstock="true"> + <path + inkscape:connector-curvature="0" + id="path1200-9-6-56" + style="fill:#ff0000;fill-opacity:1;fill-rule:evenodd;stroke:#ff0000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1" + d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z" + transform="scale(-0.6)" /> + </marker> + <marker + inkscape:stockid="Arrow2Mend" + orient="auto" + refY="0" + refX="0" + id="Arrow2Mend-8-0-2" + style="overflow:visible" + inkscape:isstock="true"> + <path + inkscape:connector-curvature="0" + id="path1200-9-3-9" + style="fill:#008000;fill-opacity:1;fill-rule:evenodd;stroke:#008000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1" + d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z" + transform="scale(-0.6)" /> + </marker> + </defs> + <sodipodi:namedview + id="base" + pagecolor="#ffffff" + bordercolor="#666666" + borderopacity="1.0" + inkscape:pageopacity="0.0" + inkscape:pageshadow="2" + inkscape:zoom="2.8" + inkscape:cx="603.80172" + inkscape:cy="404.14319" + inkscape:document-units="mm" + inkscape:current-layer="layer1" + showgrid="false" + inkscape:window-width="3840" + inkscape:window-height="2096" + inkscape:window-x="0" + inkscape:window-y="27" + inkscape:window-maximized="1" + showguides="false" + fit-margin-top="0" + fit-margin-left="0" + fit-margin-right="0" + fit-margin-bottom="0" + units="px" + inkscape:snap-global="false" /> + <metadata + id="metadata5"> + <rdf:RDF> + <cc:Work + rdf:about=""> + <dc:format>image/svg+xml</dc:format> + <dc:type + rdf:resource="http://purl.org/dc/dcmitype/StillImage" /> + <dc:title /> + </cc:Work> + </rdf:RDF> + </metadata> + <g + inkscape:label="Layer 1" + inkscape:groupmode="layer" + id="layer1" + transform="translate(419.79645,20.103767)"> + <path + style="fill:#008000;stroke:#008000;stroke-width:0.59275198;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1;marker-end:url(#Arrow2Mend-8-3-2-7)" + d="m -340.37552,57.5332 h -14.81024" + id="path1171-7-1-3-0" + inkscape:connector-curvature="0" /> + <path + style="fill:#008000;stroke:#008000;stroke-width:0.59715915;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1;marker-end:url(#Arrow2Mend-8-3-4)" + d="m -293.23443,57.5332 h -15.03129" + id="path1171-7-1-32" + inkscape:connector-curvature="0" /> + <path + style="fill:#008000;stroke:#008000;stroke-width:0.59275198;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1;marker-end:url(#Arrow2Mend-8-0)" + d="M -246.45946,57.5332 H -261.2697" + id="path1171-7-6" + inkscape:connector-curvature="0" /> + <path + style="fill:#008000;stroke:#008000;stroke-width:0.59275198;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1;marker-end:url(#Arrow2Mend-6)" + d="m -151.28623,57.5332 h -14.81024" + id="path1171-0" + inkscape:connector-curvature="0" /> + <path + style="fill:none;stroke:#ff0000;stroke-width:0.98222464;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1;marker-end:url(#Arrow2Mend-8-3-2-6)" + d="m -310.11621,-10.988713 h -35.41856" + id="path1171-7-1-3-5" + inkscape:connector-curvature="0" + sodipodi:nodetypes="cc" /> + <path + style="fill:none;stroke:#ff0000;stroke-width:1.33745635;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1;marker-end:url(#Arrow2Mend-8-3-2-1)" + d="M -174.42569,48.441117 V -10.963061 L -277.26548,-11.45916" + id="path1171-7-1-3-4" + inkscape:connector-curvature="0" + sodipodi:nodetypes="ccc" /> + <path + style="fill:none;stroke:#ff0000;stroke-width:0.95872593;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1;marker-end:url(#Arrow2Mend-8-3)" + d="m -262.79442,87.935594 h 14.32069" + id="path1171-7-1" + inkscape:connector-curvature="0" /> + <path + style="fill:none;stroke:#ff0000;stroke-width:0.97006679;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1;marker-end:url(#Arrow2Mend-8)" + d="m -309.80088,87.935594 h 14.44587" + id="path1171-7" + inkscape:connector-curvature="0" /> + <path + style="fill:none;stroke:#ff0000;stroke-width:0.96187615;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1;marker-end:url(#Arrow2Mend)" + d="m -356.45657,87.935594 h 14.20296" + id="path1171" + inkscape:connector-curvature="0" /> + <path + style="fill:none;stroke:#ff0000;stroke-width:0.96061862;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1;marker-end:url(#Arrow2Mend-8-3-2)" + d="m -167.44556,87.935594 h 14.16584" + id="path1171-7-1-3" + inkscape:connector-curvature="0" /> + <path + style="fill:none;stroke:#008000;stroke-width:0.87091714;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1;marker-end:url(#Arrow2Mend-8-0-7)" + d="M -193.82812,48.312503 V 14.168502 l -84.03577,-0.467726" + id="path1171-7-6-4" + inkscape:connector-curvature="0" + sodipodi:nodetypes="ccc" /> + <path + style="fill:none;stroke:#000000;stroke-width:0.81852055;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:4.91112339, 0.81852057;stroke-dashoffset:0;stroke-opacity:1" + d="m -133.33998,42.989657 v 5.457081" + id="path7149-3-7" + inkscape:connector-curvature="0" /> + <path + style="fill:none;stroke:#000000;stroke-width:0.81852055;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:4.91112339, 0.81852057;stroke-dashoffset:0;stroke-opacity:1" + d="m -298.69506,162.44998 v 13.31197" + id="path7149" + inkscape:connector-curvature="0" /> + <path + style="fill:none;stroke:#008080;stroke-width:0.81852055;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:0.81852057, 0.81852057;stroke-dashoffset:0;stroke-opacity:1" + d="m -242.80131,107.00907 v 9.60171" + id="path7040-5-4-7-5-6-9" + inkscape:connector-curvature="0" /> + <path + style="fill:none;stroke:#008080;stroke-width:0.81852055;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:0.81852057, 0.81852057;stroke-dashoffset:0;stroke-opacity:1" + d="m -300.34873,107.17445 v 9.6017" + id="path7040-5-4-7-5-6" + inkscape:connector-curvature="0" /> + <path + style="fill:none;stroke:#008080;stroke-width:0.81852055;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:0.81852057, 0.81852057;stroke-dashoffset:0;stroke-opacity:1" + d="m -359.26293,106.99745 v 9.60171" + id="path7040-5-4-7-5" + inkscape:connector-curvature="0" /> + <path + style="fill:none;stroke:#008080;stroke-width:0.81852055;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:0.81852057, 0.81852057;stroke-dashoffset:0;stroke-opacity:1" + d="M -369.74543,25.114933 V 37.991587" + id="path7040-5-4-7-6" + inkscape:connector-curvature="0" /> + <path + style="fill:none;stroke:#008080;stroke-width:0.91136348;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:0.91136346, 0.91136346;stroke-dashoffset:0;stroke-opacity:1" + d="M -135.17034,93.582486 V 107.10642 H -403.93077 V 37.882965 h 109.60575 V 25.225991" + id="path7038" + inkscape:connector-curvature="0" + sodipodi:nodetypes="cccccc" /> + <path + style="fill:none;stroke:#008080;stroke-width:0.81852055;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:0.81852057, 0.81852057;stroke-dashoffset:0;stroke-opacity:1" + d="M -231.106,94.010086 V 106.96943" + id="path7040" + inkscape:connector-curvature="0" /> + <path + style="fill:none;stroke:#008080;stroke-width:0.81852055;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:0.81852057, 0.81852057;stroke-dashoffset:0;stroke-opacity:1" + d="M -278.50224,93.844719 V 106.80406" + id="path7040-5" + inkscape:connector-curvature="0" /> + <path + style="fill:none;stroke:#008080;stroke-width:0.81852055;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:0.81852057, 0.81852057;stroke-dashoffset:0;stroke-opacity:1" + d="M -325.89848,93.701083 V 106.99115" + id="path7040-5-4" + inkscape:connector-curvature="0" /> + <path + style="fill:none;stroke:#008080;stroke-width:0.81852055;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:0.81852057, 0.81852057;stroke-dashoffset:0;stroke-opacity:1" + d="M -373.29471,93.899037 V 107.27179" + id="path7040-5-4-7" + inkscape:connector-curvature="0" /> + <g + id="g934" + transform="matrix(0.61872421,0,0,0.61872421,-154.16506,-3.5724799)"> + <rect + ry="2.1052283e-06" + y="84.280701" + x="-376.383" + height="72.786827" + width="49.352299" + id="rect834" + style="fill:none;stroke:#000000;stroke-width:1.16258347;stroke-linecap:round;stroke-linejoin:bevel;stroke-miterlimit:4;stroke-dasharray:none" /> + <text + id="text838" + y="95.916664" + x="-371.17261" + style="font-style:normal;font-weight:normal;font-size:10.58333302px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.26458332" + xml:space="preserve"><tspan + style="stroke-width:0.26458332" + y="95.916664" + x="-371.17261" + id="tspan836" + sodipodi:role="line">DCHUB</tspan></text> + <text + id="text846" + y="121.99702" + x="-352.74997" + style="font-style:normal;font-weight:normal;font-size:10.58333302px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.26458332" + xml:space="preserve"><tspan + style="text-align:center;text-anchor:middle;stroke-width:0.26458332" + y="121.99702" + x="-352.74997" + id="tspan844" + sodipodi:role="line">HUBP</tspan><tspan + id="tspan863" + style="text-align:center;text-anchor:middle;stroke-width:0.26458332" + y="135.22618" + x="-352.74997" + sodipodi:role="line">(n)</tspan></text> + </g> + <g + id="g942" + transform="matrix(0.61872421,0,0,0.61872421,-158.40385,-3.2216813)"> + <text + id="text838-5" + y="116.65257" + x="-269.45752" + style="font-style:normal;font-weight:normal;font-size:10.58333302px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.26458332" + xml:space="preserve"><tspan + style="text-align:center;text-anchor:middle;stroke-width:0.26458332" + y="116.65257" + x="-269.45752" + id="tspan836-3" + sodipodi:role="line">DPP</tspan><tspan + id="tspan936" + style="text-align:center;text-anchor:middle;stroke-width:0.26458332" + y="129.88174" + x="-269.45752" + sodipodi:role="line">(n)</tspan></text> + <rect + ry="2.1052283e-06" + y="83.71373" + x="-293.7952" + height="72.786827" + width="49.352303" + id="rect834-5" + style="fill:none;stroke:#000000;stroke-width:1.16258347;stroke-linecap:round;stroke-linejoin:bevel;stroke-miterlimit:4;stroke-dasharray:none" /> + </g> + <g + id="g1158" + transform="matrix(0.61872421,0,0,0.61872421,-154.34048,-6.2618995)"> + <text + id="text838-5-2" + y="128.87331" + x="-200.18195" + style="font-style:normal;font-weight:normal;font-size:10.58333302px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.26458332" + xml:space="preserve"><tspan + id="tspan936-1" + style="text-align:center;text-anchor:middle;stroke-width:0.26458332" + y="128.87331" + x="-200.18195" + sodipodi:role="line">MPC</tspan></text> + <rect + ry="2.1052283e-06" + y="88.627419" + x="-224.62555" + height="72.786827" + width="49.352303" + id="rect834-5-2" + style="fill:none;stroke:#000000;stroke-width:1.16258347;stroke-linecap:round;stroke-linejoin:bevel;stroke-miterlimit:4;stroke-dasharray:none" /> + </g> + <g + id="g1153" + transform="matrix(0.61872421,0,0,0.61872421,-108.51628,-6.4957668)"> + <text + id="text838-5-2-7" + y="129.2513" + x="-120.96272" + style="font-style:normal;font-weight:normal;font-size:10.58333302px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.26458332" + xml:space="preserve"><tspan + id="tspan936-1-0" + style="text-align:center;text-anchor:middle;stroke-width:0.26458332" + y="129.2513" + x="-120.96272" + sodipodi:role="line">OPTC</tspan></text> + <rect + ry="2.1052283e-06" + y="89.005402" + x="-145.62854" + height="72.786827" + width="49.352306" + id="rect834-5-2-9" + style="fill:none;stroke:#000000;stroke-width:1.16258347;stroke-linecap:round;stroke-linejoin:bevel;stroke-miterlimit:4;stroke-dasharray:none" /> + </g> + <g + id="g1148" + transform="matrix(0.61872421,0,0,0.61872421,-105.25474,-7.6650796)"> + <text + id="text838-5-2-7-3" + y="131.14117" + x="-48.981136" + style="font-style:normal;font-weight:normal;font-size:10.58333302px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.26458332" + xml:space="preserve"><tspan + id="tspan936-1-0-6" + style="text-align:center;text-anchor:middle;stroke-width:0.26458332" + y="131.14117" + x="-48.981136" + sodipodi:role="line">DIO</tspan></text> + <rect + ry="2.1052283e-06" + y="90.895279" + x="-73.435081" + height="72.786827" + width="49.352306" + id="rect834-5-2-9-0" + style="fill:none;stroke:#000000;stroke-width:1.16258347;stroke-linecap:round;stroke-linejoin:bevel;stroke-miterlimit:4;stroke-dasharray:none" /> + </g> + <g + id="g1133" + transform="matrix(0.61872421,0,0,0.61872421,-181.52704,-7.6650796)"> + <text + id="text838-5-2-6" + y="241.13223" + x="-286.96921" + style="font-style:normal;font-weight:normal;font-size:10.58333302px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.26458332" + xml:space="preserve"><tspan + id="tspan936-1-2" + style="text-align:center;text-anchor:middle;stroke-width:0.26458332" + y="241.13223" + x="-286.96921" + sodipodi:role="line">DCCG</tspan></text> + <rect + ry="2.1052283e-06" + y="200.88634" + x="-311.56009" + height="72.786827" + width="49.352306" + id="rect834-5-2-6" + style="fill:none;stroke:#000000;stroke-width:1.16258347;stroke-linecap:round;stroke-linejoin:bevel;stroke-miterlimit:4;stroke-dasharray:none" /> + </g> + <g + id="g1138" + transform="matrix(0.61872421,0,0,0.61872421,-181.52704,-7.6650796)"> + <text + id="text838-5-2-6-1" + y="241.81844" + x="-190.55942" + style="font-style:normal;font-weight:normal;font-size:10.58333302px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.26458332" + xml:space="preserve"><tspan + id="tspan936-1-2-8" + style="text-align:center;text-anchor:middle;stroke-width:0.26458332" + y="241.81844" + x="-190.55942" + sodipodi:role="line">DMU</tspan></text> + <rect + ry="2.1052283e-06" + y="201.6423" + x="-215.17615" + height="72.786827" + width="49.352306" + id="rect834-5-2-6-7" + style="fill:none;stroke:#000000;stroke-width:1.16258347;stroke-linecap:round;stroke-linejoin:bevel;stroke-miterlimit:4;stroke-dasharray:none" /> + </g> + <text + xml:space="preserve" + style="font-style:normal;font-weight:normal;font-size:6.54816437px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.1637041" + x="-297.75696" + y="109.44505" + id="text1063"><tspan + sodipodi:role="line" + id="tspan1061" + x="-297.75696" + y="115.23865" + style="stroke-width:0.1637041" /></text> + <g + id="g1143" + transform="matrix(0.61872421,0,0,0.61872421,-181.52704,-8.9747125)"> + <text + id="text838-5-2-6-1-9" + y="243.02728" + x="-99.967323" + style="font-style:normal;font-weight:normal;font-size:10.58333302px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.26458332" + xml:space="preserve"><tspan + id="tspan936-1-2-8-2" + style="text-align:center;text-anchor:middle;stroke-width:0.26458332" + y="243.02728" + x="-99.967323" + sodipodi:role="line">AZ</tspan></text> + <rect + ry="2.1052283e-06" + y="202.77623" + x="-124.83984" + height="72.786827" + width="49.352306" + id="rect834-5-2-6-7-0" + style="fill:none;stroke:#000000;stroke-width:1.16258347;stroke-linecap:round;stroke-linejoin:bevel;stroke-miterlimit:4;stroke-dasharray:none" /> + </g> + <g + id="g1169" + transform="matrix(0.61872421,0,0,0.61872421,-154.16506,1.4555785)"> + <text + id="text838-5-2-6-2" + y="5.9612885" + x="-348.74365" + style="font-style:normal;font-weight:normal;font-size:10.58333302px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.26458332" + xml:space="preserve"><tspan + id="tspan936-1-2-3" + style="text-align:center;text-anchor:middle;stroke-width:0.26458332" + y="5.9612885" + x="-348.74365" + sodipodi:role="line">MMHUBBUB</tspan></text> + <rect + ry="2.1010696e-06" + y="-34.142948" + x="-384.64743" + height="72.643044" + width="72.096924" + id="rect834-5-2-6-75" + style="fill:none;stroke:#000000;stroke-width:1.40378118;stroke-linecap:round;stroke-linejoin:bevel;stroke-miterlimit:4;stroke-dasharray:none" /> + </g> + <g + id="g1164" + transform="matrix(0.61872421,0,0,0.61872421,-154.16506,-7.6650796)"> + <text + id="text838-5-2-6-9" + y="13.465075" + x="-227.30836" + style="font-style:normal;font-weight:normal;font-size:10.58333302px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.26458332" + xml:space="preserve"><tspan + id="tspan936-1-2-2" + style="text-align:center;text-anchor:middle;stroke-width:0.26458332" + y="13.465075" + x="-227.30836" + sodipodi:role="line">DWB</tspan><tspan + id="tspan1128" + style="text-align:center;text-anchor:middle;stroke-width:0.26458332" + y="26.694241" + x="-227.30836" + sodipodi:role="line">(n)</tspan></text> + <rect + ry="2.1052283e-06" + y="-19.473768" + x="-251.83983" + height="72.786827" + width="49.352306" + id="rect834-5-2-6-2" + style="fill:none;stroke:#000000;stroke-width:1.16258347;stroke-linecap:round;stroke-linejoin:bevel;stroke-miterlimit:4;stroke-dasharray:none" /> + </g> + <path + style="fill:none;stroke:#000000;stroke-width:0.91371936;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:5.48231601, 0.91371934;stroke-dashoffset:0;stroke-opacity:1" + d="m -358.95963,161.63019 v 14.12431 h 250.20395 V 43.149938 H -361.845 V 25.478973" + id="path7147" + inkscape:connector-curvature="0" + sodipodi:nodetypes="cccccc" /> + <path + style="fill:none;stroke:#000000;stroke-width:0.81852055;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:4.91112339, 0.81852057;stroke-dashoffset:0;stroke-opacity:1" + d="m -242.92533,161.58513 v 14.05612" + id="path7149-3" + inkscape:connector-curvature="0" /> + <path + style="fill:none;stroke:#000000;stroke-width:0.81852055;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:4.91112339, 0.81852057;stroke-dashoffset:0;stroke-opacity:1" + d="m -184.37695,42.955607 v 5.457082" + id="path7149-3-7-4" + inkscape:connector-curvature="0" /> + <path + style="fill:none;stroke:#000000;stroke-width:0.81852055;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:4.91112339, 0.81852057;stroke-dashoffset:0;stroke-opacity:1" + d="m -277.36283,43.141644 v 5.457082" + id="path7149-3-7-5" + inkscape:connector-curvature="0" /> + <path + style="fill:none;stroke:#000000;stroke-width:0.81852055;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:4.91112339, 0.81852057;stroke-dashoffset:0;stroke-opacity:1" + d="M -325.48437,42.976278 V 48.43336" + id="path7149-3-7-2" + inkscape:connector-curvature="0" /> + <path + style="fill:none;stroke:#000000;stroke-width:0.81852055;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:4.91112339, 0.81852057;stroke-dashoffset:0;stroke-opacity:1" + d="m -361.86492,43.141644 v 5.457083" + id="path7149-3-7-54" + inkscape:connector-curvature="0" /> + <path + style="fill:#008000;stroke:#008000;stroke-width:0.46329758;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1;marker-end:url(#Arrow2Mend-6-3)" + d="m -147.58542,-8.2978166 h -9.04766" + id="path1171-0-7" + inkscape:connector-curvature="0" /> + <path + style="fill:none;stroke:#ff0000;stroke-width:0.98222464;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1;marker-end:url(#Arrow2Mend-8-3-2-8)" + d="m -157.13421,-1.6500501 h 8.66407" + id="path1171-7-1-3-8" + inkscape:connector-curvature="0" /> + <path + style="fill:none;stroke:#000000;stroke-width:0.74503672;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:4.47022031, 0.74503672;stroke-dashoffset:0;stroke-opacity:1" + d="m -148.50314,4.9845652 h -7.91265" + id="path7149-3-7-8" + inkscape:connector-curvature="0" /> + <path + style="fill:none;stroke:#008080;stroke-width:0.81852055;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:0.81852057, 0.81852057;stroke-dashoffset:0;stroke-opacity:1" + d="m -157.59442,11.623513 h 10.26991" + id="path7040-4" + inkscape:connector-curvature="0" /> + <text + xml:space="preserve" + style="font-style:normal;font-weight:normal;font-size:6.54816437px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.1637041" + x="-142.72867" + y="-6.9685979" + id="text12079"><tspan + sodipodi:role="line" + id="tspan12077" + x="-142.72867" + y="-6.9685979" + style="font-size:4.80198765px;stroke-width:0.1637041">Global sync</tspan></text> + <text + xml:space="preserve" + style="font-style:normal;font-weight:normal;font-size:6.54816437px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.1637041" + x="-142.93031" + y="0.13578746" + id="text12079-3"><tspan + sodipodi:role="line" + id="tspan12077-1" + x="-142.93031" + y="0.13578746" + style="font-size:4.80198765px;stroke-width:0.1637041">Pixel data</tspan></text> + <text + xml:space="preserve" + style="font-style:normal;font-weight:normal;font-size:6.54816437px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.1637041" + x="-142.77556" + y="6.3093324" + id="text12079-3-4"><tspan + sodipodi:role="line" + id="tspan12077-1-9" + x="-142.77556" + y="6.3093324" + style="font-size:4.80198765px;stroke-width:0.1637041">Sideband signal</tspan></text> + <text + xml:space="preserve" + style="font-style:normal;font-weight:normal;font-size:6.54816437px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.1637041" + x="-142.72867" + y="12.948278" + id="text12079-3-4-2"><tspan + sodipodi:role="line" + id="tspan12077-1-9-0" + x="-142.72867" + y="12.948278" + style="font-size:4.80198765px;stroke-width:0.1637041">Config. Bus</tspan></text> + <path + style="fill:none;stroke:#aa00d4;stroke-width:1.32291663;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1;marker-end:url(#marker8858)" + d="m -406.68795,73.185276 h 14.20296" + id="path1171-75" + inkscape:connector-curvature="0" /> + <text + xml:space="preserve" + style="font-style:normal;font-weight:normal;font-size:3.17499995px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.26458332" + x="-420.21503" + y="75.065918" + id="text8862"><tspan + sodipodi:role="line" + id="tspan8860" + x="-420.21503" + y="75.065918" + style="font-size:6.3499999px;stroke-width:0.26458332">SDP</tspan></text> + <path + style="fill:none;stroke:#00ffcc;stroke-width:1.25980031;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1;marker-end:url(#marker8858-3)" + d="m -119.19923,72.243805 h 12.88004" + id="path1171-75-6" + inkscape:connector-curvature="0" /> + <text + xml:space="preserve" + style="font-style:normal;font-weight:normal;font-size:3.17499995px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.26458332" + x="-104.87327" + y="74.54258" + id="text8862-2"><tspan + sodipodi:role="line" + id="tspan8860-9" + x="-104.87327" + y="74.54258" + style="font-size:6.3499999px;stroke-width:0.26458332">Monitor</tspan></text> + <g + id="g6280" + transform="translate(-133.43389,-37.35791)"> + <text + id="text838-5-2-7-6" + y="110.67171" + x="-97.4758" + style="font-style:normal;font-weight:normal;font-size:6.54816437px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.16370411" + xml:space="preserve"><tspan + id="tspan936-1-0-7" + style="text-align:center;text-anchor:middle;stroke-width:0.16370411" + y="110.67171" + x="-97.4758" + sodipodi:role="line">OPP</tspan></text> + <rect + ry="1.3025557e-06" + y="85.770599" + x="-112.73714" + height="45.034973" + width="30.535467" + id="rect834-5-2-9-5" + style="fill:none;stroke:#000000;stroke-width:0.71931857;stroke-linecap:round;stroke-linejoin:bevel;stroke-miterlimit:4;stroke-dasharray:none" /> + </g> + <path + style="fill:#008000;stroke:#008000;stroke-width:0.59275198;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1;marker-end:url(#Arrow2Mend-8-0-2)" + d="m -199.6735,57.600919 h -14.81024" + id="path1171-7-6-1" + inkscape:connector-curvature="0" /> + <path + style="fill:none;stroke:#ff0000;stroke-width:0.95872593;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1;marker-end:url(#Arrow2Mend-8-3-3)" + d="m -214.95012,88.003315 h 14.32069" + id="path1171-7-1-2" + inkscape:connector-curvature="0" /> + <path + style="fill:none;stroke:#008080;stroke-width:0.81852055;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:0.81852058, 0.81852058;stroke-dashoffset:0;stroke-opacity:1" + d="M -182.99565,94.057598 V 107.01694" + id="path7040-7" + inkscape:connector-curvature="0" /> + <path + style="fill:none;stroke:#000000;stroke-width:0.81852055;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:4.91112344, 0.81852058;stroke-dashoffset:0;stroke-opacity:1" + d="m -231.7616,43.563759 v 5.457082" + id="path7149-3-7-4-0" + inkscape:connector-curvature="0" /> + <g + aria-label="[" + transform="matrix(0,-1,0.74237844,0,14.567595,39.540924)" + style="font-style:normal;font-weight:normal;font-size:3.17499995px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.26458332" + id="text6872"> + <path + d="m -65.936923,-548.78511 h 8.816294 v 2.79112 h -6.82247 v 176.34952 h 6.82247 v 2.41314 h -8.816294 z" + style="font-size:50.79999924px;stroke-width:0.26458332" + id="path6874" + inkscape:connector-curvature="0" + sodipodi:nodetypes="ccccccccc" /> + </g> + <g + aria-label="[" + transform="rotate(-90,182.49521,-144.01791)" + style="font-style:normal;font-weight:normal;font-size:3.17499995px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.26458332" + id="text6872-3"> + <path + d="m -63.825546,-623.34091 h 7.228794 v 2.26195 h -5.764137 l 0,127.08032 h 5.764137 v 1.88397 h -7.228794 z" + style="font-size:50.79999924px;stroke-width:0.26458332" + id="path6874-6" + inkscape:connector-curvature="0" + sodipodi:nodetypes="ccccccccc" /> + </g> + <text + xml:space="preserve" + style="font-style:normal;font-weight:normal;font-size:3.17499995px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.26458332" + x="-359.80389" + y="99.104233" + id="text6929"><tspan + sodipodi:role="line" + id="tspan6927" + x="-359.80389" + y="99.104233" + style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:5.64444447px;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold';stroke-width:0.26458332">dc_plane</tspan></text> + <text + xml:space="preserve" + style="font-style:normal;font-weight:normal;font-size:3.17499995px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.26458332" + x="-223.56163" + y="99.142021" + id="text6933"><tspan + sodipodi:role="line" + id="tspan6931" + x="-223.56163" + y="99.142021" + style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:5.64444447px;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold';stroke-width:0.26458332">dc_stream</tspan></text> + <g + aria-label="[" + transform="matrix(0,1,1,0,153.30551,96.566025)" + style="font-style:normal;font-weight:normal;font-size:3.17499995px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.26458332" + id="text6872-35"> + <path + d="m -65.936923,-545.95029 h 8.816294 v 2.79112 h -6.898066 v 271.78851 h 6.898066 v 2.41314 h -8.816294 z" + style="font-size:50.79999924px;stroke-width:0.26458332" + id="path6874-62" + inkscape:connector-curvature="0" + sodipodi:nodetypes="ccccccccc" /> + </g> + <text + xml:space="preserve" + style="font-style:normal;font-weight:normal;font-size:3.17499995px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.26458332" + x="-267.43958" + y="28.5028" + id="text6933-9"><tspan + sodipodi:role="line" + id="tspan6931-1" + x="-267.43958" + y="28.5028" + style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:5.64444447px;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold';stroke-width:0.26458332">dc_state</tspan></text> + <g + aria-label="[" + transform="matrix(0,0.98158883,-1.0187565,0,0,-7.4835468)" + style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:6.62759447px;line-height:1.25;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold';letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.5522995" + id="text6973"> + <path + d="m 23.679381,144.30265 h 3.028123 v 1.29445 h -1.820839 v 7.91629 h 1.820839 v 1.29445 h -3.028123 z" + style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:11.78239059px;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold';stroke-width:0.5522995" + id="path6975" + inkscape:connector-curvature="0" + sodipodi:nodetypes="ccccccccc" /> + </g> + <text + xml:space="preserve" + style="font-style:normal;font-weight:normal;font-size:3.17499995px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.26458332" + x="-142.71655" + y="18.955769" + id="text6980"><tspan + sodipodi:role="line" + id="tspan6978" + x="-142.71655" + y="18.955769" + style="font-size:4.58611107px;stroke-width:0.26458332">Code struct</tspan></text> + <g + aria-label="[" + transform="rotate(-90,94.826273,-58.762727)" + style="font-style:normal;font-weight:normal;font-size:3.17499995px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.26458332" + id="text6872-35-2"> + <path + d="m -66.881863,-308.95922 h 7.115401 l 0,1.69499 h -5.197173 v 42.03568 h 5.197173 v 1.78948 h -7.115401 z" + style="font-size:50.79999924px;stroke-width:0.26458332" + id="path6874-62-7" + inkscape:connector-curvature="0" + sodipodi:nodetypes="ccccccccc" /> + </g> + <text + xml:space="preserve" + style="font-style:normal;font-weight:normal;font-size:3.17499995px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.26458332" + x="-134.09625" + y="99.354439" + id="text6933-9-0"><tspan + sodipodi:role="line" + id="tspan6931-1-9" + x="-134.09625" + y="99.354439" + style="font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:5.64444447px;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold';stroke-width:0.26458332">dc_link</tspan></text> + <g + aria-label="}" + transform="rotate(90,-145.27371,-140.09832)" + style="font-style:normal;font-weight:normal;font-size:3.17499995px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#3771c8;fill-opacity:1;stroke:none;stroke-width:0.26458332" + id="text1003"> + <path + d="m 102.58571,58.211269 h 0.86816 c 1.15755,0 3.11267,-0.177767 3.45994,-0.5333 0.35553,-0.355534 1.28925,-1.124479 1.28925,-2.306836 V -61.475482 c 0,-1.289844 0.18603,-2.228288 0.5581,-2.815332 0.37207,-0.587044 0.26105,-0.992187 1.17882,-1.215429 -0.91777,-0.206706 -0.80675,-0.603581 -1.17882,-1.190625 -0.37207,-0.587045 -0.5581,-1.529623 -0.5581,-2.827735 v -3.075781 c 0,-1.174088 -0.93372,-1.938899 -1.28925,-2.294433 -0.34727,-0.363802 -2.30239,-0.545703 -3.45994,-0.545703 h -0.86816 v -1.773536 h 0.78134 c 2.05879,0 4.63403,0.305924 5.32029,0.917774 0.69453,0.60358 1.0418,1.81901 1.0418,3.646289 v 2.976562 c 0,1.231966 0.22324,2.087728 0.66973,2.567285 0.44648,0.471289 5.80035,0.706934 6.97444,0.706934 h 0.76894 v 1.773535 h -0.76894 c -1.17409,0 -6.52796,0.239778 -6.97444,0.719336 -0.44649,0.479557 -0.66973,1.343587 -0.66973,2.59209 V 55.420742 c 0,1.827279 -0.34727,3.046842 -1.0418,3.658691 -0.68626,0.611849 -3.2615,0.917774 -5.32029,0.917774 h -0.78134 z" + style="font-size:25.39999962px;fill:#3771c8;stroke-width:0.26458332" + id="path1005" + inkscape:connector-curvature="0" + sodipodi:nodetypes="cscsscccsscsccscsscsccscsscscc" /> + </g> + <text + xml:space="preserve" + style="font-style:normal;font-weight:normal;font-size:3.17499995px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.26458332" + x="-200.59984" + y="129.60852" + id="text1010"><tspan + sodipodi:role="line" + id="tspan1008" + x="-200.59984" + y="129.60852" + style="font-style:italic;font-variant:normal;font-weight:bold;font-stretch:normal;font-size:6.3499999px;font-family:sans-serif;-inkscape-font-specification:'sans-serif Bold Italic';text-align:center;text-anchor:middle;stroke-width:0.26458332">Floating point</tspan><tspan + sodipodi:role="line" + x="-200.59984" + y="137.54602" + style="font-size:6.3499999px;text-align:center;text-anchor:middle;stroke-width:0.26458332" + id="tspan1059">calculation</tspan></text> + <g + aria-label="}" + transform="rotate(90,-94.294068,-92.593178)" + style="font-style:normal;font-weight:normal;font-size:3.17499995px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#3771c8;fill-opacity:1;stroke:none;stroke-width:0.26458332" + id="text1003-5"> + <path + d="m 102.58571,58.211269 h 0.86816 c 1.15755,0 3.11267,-0.177767 3.45994,-0.5333 0.35553,-0.355534 1.10026,-1.124479 1.10026,-2.306836 V -44.637502 c 0,-1.289844 0.18603,-2.228288 0.5581,-2.815332 0.37207,-0.587044 0.45004,-0.992187 1.36781,-1.215429 -0.91777,-0.206706 -0.99574,-0.603581 -1.36781,-1.190625 -0.37207,-0.587045 -0.5581,-1.529623 -0.5581,-2.827735 v -19.913761 c 0,-1.174088 -0.74473,-1.938899 -1.10026,-2.294433 -0.34727,-0.363802 -2.30239,-0.545703 -3.45994,-0.545703 h -0.86816 v -1.773536 h 0.78134 c 2.05879,0 4.63403,0.305924 5.32029,0.917774 0.69453,0.60358 1.0418,1.81901 1.0418,3.646289 v 19.814542 c 0,1.231966 0.22324,2.087728 0.66973,2.567285 0.44648,0.471289 1.25677,0.706934 2.43086,0.706934 h 0.76894 v 1.773535 h -0.76894 c -1.17409,0 -1.98438,0.239778 -2.43086,0.719336 -0.44649,0.479557 -0.66973,1.343587 -0.66973,2.59209 v 99.897013 c 0,1.827279 -0.34727,3.046842 -1.0418,3.658691 -0.68626,0.611849 -3.2615,0.917774 -5.32029,0.917774 h -0.78134 z" + style="font-size:25.39999962px;fill:#3771c8;stroke-width:0.26458332" + id="path1005-3" + inkscape:connector-curvature="0" + sodipodi:nodetypes="cscsscccsscsccscsscsccscsscscc" /> + </g> + <text + xml:space="preserve" + style="font-style:normal;font-weight:normal;font-size:3.17499995px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.26458332" + x="-137.43764" + y="122.46283" + id="text1010-5"><tspan + sodipodi:role="line" + id="tspan1008-6" + x="-137.43764" + y="122.46283" + style="font-size:6.3499999px;text-align:center;text-anchor:middle;stroke-width:0.26458332">bit-depth</tspan><tspan + sodipodi:role="line" + x="-137.43764" + y="130.40033" + style="font-size:6.3499999px;text-align:center;text-anchor:middle;stroke-width:0.26458332" + id="tspan1057">reduction/dither</tspan></text> + <text + xml:space="preserve" + style="font-style:normal;font-weight:normal;font-size:3.17499995px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#3771c8;fill-opacity:1;stroke:none;stroke-width:0.26458332;" + x="21.087883" + y="155.64751" + id="text1064" + transform="rotate(90)"><tspan + sodipodi:role="line" + id="tspan1062" + x="21.087883" + y="155.64751" + style="font-size:9.87777805px;stroke-width:0.26458332;fill:#3771c8;">}</tspan></text> + <text + xml:space="preserve" + style="font-style:normal;font-weight:normal;font-size:3.17499995px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.26458332" + x="-142.71655" + y="25.939869" + id="text6980-9"><tspan + sodipodi:role="line" + id="tspan6978-1" + x="-142.71655" + y="25.939869" + style="font-size:4.58611107px;stroke-width:0.26458332">Notes</tspan></text> + </g> +</svg> diff --git a/Documentation/gpu/amdgpu/display/dcn-overview.rst b/Documentation/gpu/amdgpu/display/dcn-overview.rst new file mode 100644 index 000000000000..f98624d7828e --- /dev/null +++ b/Documentation/gpu/amdgpu/display/dcn-overview.rst @@ -0,0 +1,171 @@ +======================= +Display Core Next (DCN) +======================= + +To equip our readers with the basic knowledge of how AMD Display Core Next +(DCN) works, we need to start with an overview of the hardware pipeline. Below +you can see a picture that provides a DCN overview, keep in mind that this is a +generic diagram, and we have variations per ASIC. + +.. kernel-figure:: dc_pipeline_overview.svg + +Based on this diagram, we can pass through each block and briefly describe +them: + +* **Display Controller Hub (DCHUB)**: This is the gateway between the Scalable + Data Port (SDP) and DCN. This component has multiple features, such as memory + arbitration, rotation, and cursor manipulation. + +* **Display Pipe and Plane (DPP)**: This block provides pre-blend pixel + processing such as color space conversion, linearization of pixel data, tone + mapping, and gamut mapping. + +* **Multiple Pipe/Plane Combined (MPC)**: This component performs blending of + multiple planes, using global or per-pixel alpha. + +* **Output Pixel Processing (OPP)**: Process and format pixels to be sent to + the display. + +* **Output Pipe Timing Combiner (OPTC)**: It generates time output to combine + streams or divide capabilities. CRC values are generated in this block. + +* **Display Output (DIO)**: Codify the output to the display connected to our + GPU. + +* **Display Writeback (DWB)**: It provides the ability to write the output of + the display pipe back to memory as video frames. + +* **Multi-Media HUB (MMHUBBUB)**: Memory controller interface for DMCUB and DWB + (Note that DWB is not hooked yet). + +* **DCN Management Unit (DMU)**: It provides registers with access control and + interrupts the controller to the SOC host interrupt unit. This block includes + the Display Micro-Controller Unit - version B (DMCUB), which is handled via + firmware. + +* **DCN Clock Generator Block (DCCG)**: It provides the clocks and resets + for all of the display controller clock domains. + +* **Azalia (AZ)**: Audio engine. + +The above diagram is an architecture generalization of DCN, which means that +every ASIC has variations around this base model. Notice that the display +pipeline is connected to the Scalable Data Port (SDP) via DCHUB; you can see +the SDP as the element from our Data Fabric that feeds the display pipe. + +Always approach the DCN architecture as something flexible that can be +configured and reconfigured in multiple ways; in other words, each block can be +setup or ignored accordingly with userspace demands. For example, if we +want to drive an 8k@60Hz with a DSC enabled, our DCN may require 4 DPP and 2 +OPP. It is DC's responsibility to drive the best configuration for each +specific scenario. Orchestrate all of these components together requires a +sophisticated communication interface which is highlighted in the diagram by +the edges that connect each block; from the chart, each connection between +these blocks represents: + +1. Pixel data interface (red): Represents the pixel data flow; +2. Global sync signals (green): It is a set of synchronization signals composed + by VStartup, VUpdate, and VReady; +3. Config interface: Responsible to configure blocks; +4. Sideband signals: All other signals that do not fit the previous one. + +These signals are essential and play an important role in DCN. Nevertheless, +the Global Sync deserves an extra level of detail described in the next +section. + +All of these components are represented by a data structure named dc_state. +From DCHUB to MPC, we have a representation called dc_plane; from MPC to OPTC, +we have dc_stream, and the output (DIO) is handled by dc_link. Keep in mind +that HUBP accesses a surface using a specific format read from memory, and our +dc_plane should work to convert all pixels in the plane to something that can +be sent to the display via dc_stream and dc_link. + +Front End and Back End +---------------------- + +Display pipeline can be broken down into two components that are usually +referred as **Front End (FE)** and **Back End (BE)**, where FE consists of: + +* DCHUB (Mainly referring to a subcomponent named HUBP) +* DPP +* MPC + +On the other hand, BE consist of + +* OPP +* OPTC +* DIO (DP/HDMI stream encoder and link encoder) + +OPP and OPTC are two joining blocks between FE and BE. On a side note, this is +a one-to-one mapping of the link encoder to PHY, but we can configure the DCN +to choose which link encoder to connect to which PHY. FE's main responsibility +is to change, blend and compose pixel data, while BE's job is to frame a +generic pixel stream to a specific display's pixel stream. + +Data Flow +--------- + +Initially, data is passed in from VRAM through Data Fabric (DF) in native pixel +formats. Such data format stays through till HUBP in DCHUB, where HUBP unpacks +different pixel formats and outputs them to DPP in uniform streams through 4 +channels (1 for alpha + 3 for colors). + +The Converter and Cursor (CNVC) in DPP would then normalize the data +representation and convert them to a DCN specific floating-point format (i.e., +different from the IEEE floating-point format). In the process, CNVC also +applies a degamma function to transform the data from non-linear to linear +space to relax the floating-point calculations following. Data would stay in +this floating-point format from DPP to OPP. + +Starting OPP, because color transformation and blending have been completed +(i.e alpha can be dropped), and the end sinks do not require the precision and +dynamic range that floating points provide (i.e. all displays are in integer +depth format), bit-depth reduction/dithering would kick in. In OPP, we would +also apply a regamma function to introduce the gamma removed earlier back. +Eventually, we output data in integer format at DIO. + +Global Sync +----------- + +Many DCN registers are double buffered, most importantly the surface address. +This allows us to update DCN hardware atomically for page flips, as well as +for most other updates that don't require enabling or disabling of new pipes. + +(Note: There are many scenarios when DC will decide to reserve extra pipes +in order to support outputs that need a very high pixel clock, or for +power saving purposes.) + +These atomic register updates are driven by global sync signals in DCN. In +order to understand how atomic updates interact with DCN hardware, and how DCN +signals page flip and vblank events it is helpful to understand how global sync +is programmed. + +Global sync consists of three signals, VSTARTUP, VUPDATE, and VREADY. These are +calculated by the Display Mode Library - DML (drivers/gpu/drm/amd/display/dc/dml) +based on a large number of parameters and ensure our hardware is able to feed +the DCN pipeline without underflows or hangs in any given system configuration. +The global sync signals always happen during VBlank, are independent from the +VSync signal, and do not overlap each other. + +VUPDATE is the only signal that is of interest to the rest of the driver stack +or userspace clients as it signals the point at which hardware latches to +atomically programmed (i.e. double buffered) registers. Even though it is +independent of the VSync signal we use VUPDATE to signal the VSync event as it +provides the best indication of how atomic commits and hardware interact. + +Since DCN hardware is double-buffered the DC driver is able to program the +hardware at any point during the frame. + +The below picture illustrates the global sync signals: + +.. kernel-figure:: global_sync_vblank.svg + +These signals affect core DCN behavior. Programming them incorrectly will lead +to a number of negative consequences, most of them quite catastrophic. + +The following picture shows how global sync allows for a mailbox style of +updates, i.e. it allows for multiple re-configurations between VUpdate +events where only the last configuration programmed before the VUpdate signal +becomes effective. + +.. kernel-figure:: config_example.svg diff --git a/Documentation/gpu/amdgpu/display/display-manager.rst b/Documentation/gpu/amdgpu/display/display-manager.rst new file mode 100644 index 000000000000..7ce31f89d9a0 --- /dev/null +++ b/Documentation/gpu/amdgpu/display/display-manager.rst @@ -0,0 +1,42 @@ +====================== +AMDgpu Display Manager +====================== + +.. contents:: Table of Contents + :depth: 3 + +.. kernel-doc:: drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c + :doc: overview + +.. kernel-doc:: drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h + :internal: + +Lifecycle +========= + +.. kernel-doc:: drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c + :doc: DM Lifecycle + +.. kernel-doc:: drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c + :functions: dm_hw_init dm_hw_fini + +Interrupts +========== + +.. kernel-doc:: drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq.c + :doc: overview + +.. kernel-doc:: drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_irq.c + :internal: + +.. kernel-doc:: drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c + :functions: register_hpd_handlers dm_crtc_high_irq dm_pflip_high_irq + +Atomic Implementation +===================== + +.. kernel-doc:: drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c + :doc: atomic + +.. kernel-doc:: drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c + :functions: amdgpu_dm_atomic_check amdgpu_dm_atomic_commit_tail diff --git a/Documentation/gpu/amdgpu/display/global_sync_vblank.svg b/Documentation/gpu/amdgpu/display/global_sync_vblank.svg new file mode 100644 index 000000000000..48f5dc4fd5d3 --- /dev/null +++ b/Documentation/gpu/amdgpu/display/global_sync_vblank.svg @@ -0,0 +1,485 @@ +<?xml version="1.0" encoding="UTF-8" standalone="no"?> +<!-- Created with Inkscape (http://www.inkscape.org/) --> + +<svg + xmlns:dc="http://purl.org/dc/elements/1.1/" + xmlns:cc="http://creativecommons.org/ns#" + xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" + xmlns:svg="http://www.w3.org/2000/svg" + xmlns="http://www.w3.org/2000/svg" + xmlns:sodipodi="http://sodipodi.sourceforge.net/DTD/sodipodi-0.dtd" + xmlns:inkscape="http://www.inkscape.org/namespaces/inkscape" + width="232.24133mm" + height="96.174995mm" + viewBox="0 0 232.24133 96.174995" + version="1.1" + id="svg8" + inkscape:version="0.92.5 (2060ec1f9f, 2020-04-08)" + sodipodi:docname="global_sync_vblank.svg"> + <defs + id="defs2"> + <marker + inkscape:stockid="Arrow2Mend" + orient="auto" + refY="0" + refX="0" + id="Arrow2Mend" + style="overflow:visible" + inkscape:isstock="true"> + <path + id="path862" + style="fill:#800080;fill-opacity:1;fill-rule:evenodd;stroke:#800080;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1" + d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z" + transform="scale(-0.6)" + inkscape:connector-curvature="0" /> + </marker> + <marker + inkscape:stockid="Arrow2Send" + orient="auto" + refY="0" + refX="0" + id="Arrow2Send" + style="overflow:visible" + inkscape:isstock="true"> + <path + id="path868" + style="fill:#ff00ff;fill-opacity:1;fill-rule:evenodd;stroke:#ff00ff;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1" + d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z" + transform="matrix(-0.3,0,0,-0.3,0.69,0)" + inkscape:connector-curvature="0" /> + </marker> + <marker + inkscape:stockid="Arrow2Lend" + orient="auto" + refY="0" + refX="0" + id="Arrow2Lend" + style="overflow:visible" + inkscape:isstock="true"> + <path + id="path856" + style="fill:#ff00ff;fill-opacity:1;fill-rule:evenodd;stroke:#ff00ff;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1" + d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z" + transform="matrix(-1.1,0,0,-1.1,-1.1,0)" + inkscape:connector-curvature="0" /> + </marker> + <marker + inkscape:stockid="Arrow1Lend" + orient="auto" + refY="0" + refX="0" + id="marker1719" + style="overflow:visible" + inkscape:isstock="true"> + <path + id="path1717" + d="M 0,0 5,-5 -12.5,0 5,5 Z" + style="fill:#ff00ff;fill-opacity:1;fill-rule:evenodd;stroke:#ff00ff;stroke-width:1.00000003pt;stroke-opacity:1" + transform="matrix(-0.8,0,0,-0.8,-10,0)" + inkscape:connector-curvature="0" /> + </marker> + <marker + inkscape:stockid="Arrow1Lend" + orient="auto" + refY="0" + refX="0" + id="marker1661" + style="overflow:visible" + inkscape:isstock="true"> + <path + id="path1659" + d="M 0,0 5,-5 -12.5,0 5,5 Z" + style="fill:#ff00ff;fill-opacity:1;fill-rule:evenodd;stroke:#ff00ff;stroke-width:1.00000003pt;stroke-opacity:1" + transform="matrix(-0.8,0,0,-0.8,-10,0)" + inkscape:connector-curvature="0" /> + </marker> + <marker + inkscape:isstock="true" + style="overflow:visible" + id="marker1311" + refX="0" + refY="0" + orient="auto" + inkscape:stockid="Arrow1Lend" + inkscape:collect="always"> + <path + transform="matrix(-0.8,0,0,-0.8,-10,0)" + style="fill:#ff0000;fill-opacity:1;fill-rule:evenodd;stroke:#ff0000;stroke-width:1.00000003pt;stroke-opacity:1" + d="M 0,0 5,-5 -12.5,0 5,5 Z" + id="path1309" + inkscape:connector-curvature="0" /> + </marker> + <marker + inkscape:isstock="true" + style="overflow:visible" + id="marker1253" + refX="0" + refY="0" + orient="auto" + inkscape:stockid="Arrow1Lstart"> + <path + transform="matrix(0.8,0,0,0.8,10,0)" + style="fill:#ff0000;fill-opacity:1;fill-rule:evenodd;stroke:#ff0000;stroke-width:1.00000003pt;stroke-opacity:1" + d="M 0,0 5,-5 -12.5,0 5,5 Z" + id="path1251" + inkscape:connector-curvature="0" /> + </marker> + <marker + inkscape:stockid="Arrow1Lend" + orient="auto" + refY="0" + refX="0" + id="Arrow1Lend" + style="overflow:visible" + inkscape:isstock="true" + inkscape:collect="always"> + <path + id="path838" + d="M 0,0 5,-5 -12.5,0 5,5 Z" + style="fill:#ff0000;fill-opacity:1;fill-rule:evenodd;stroke:#ff0000;stroke-width:1.00000003pt;stroke-opacity:1" + transform="matrix(-0.8,0,0,-0.8,-10,0)" + inkscape:connector-curvature="0" /> + </marker> + <marker + inkscape:stockid="Arrow1Lstart" + orient="auto" + refY="0" + refX="0" + id="Arrow1Lstart" + style="overflow:visible" + inkscape:isstock="true" + inkscape:collect="always"> + <path + id="path835" + d="M 0,0 5,-5 -12.5,0 5,5 Z" + style="fill:#ff0000;fill-opacity:1;fill-rule:evenodd;stroke:#ff0000;stroke-width:1.00000003pt;stroke-opacity:1" + transform="matrix(0.8,0,0,0.8,10,0)" + inkscape:connector-curvature="0" /> + </marker> + <marker + inkscape:stockid="Arrow1Send" + orient="auto" + refY="0" + refX="0" + id="Arrow1Send" + style="overflow:visible" + inkscape:isstock="true"> + <path + id="path850" + d="M 0,0 5,-5 -12.5,0 5,5 Z" + style="fill:#000000;fill-opacity:1;fill-rule:evenodd;stroke:#000000;stroke-width:1.00000003pt;stroke-opacity:1" + transform="matrix(-0.2,0,0,-0.2,-1.2,0)" + inkscape:connector-curvature="0" /> + </marker> + <marker + inkscape:stockid="Arrow2Sstart" + orient="auto" + refY="0" + refX="0" + id="Arrow2Sstart" + style="overflow:visible" + inkscape:isstock="true"> + <path + id="path865" + style="fill:#000000;fill-opacity:1;fill-rule:evenodd;stroke:#000000;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1" + d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z" + transform="matrix(0.3,0,0,0.3,-0.69,0)" + inkscape:connector-curvature="0" /> + </marker> + <marker + inkscape:stockid="Arrow2Mend" + orient="auto" + refY="0" + refX="0" + id="Arrow2Mend-2" + style="overflow:visible" + inkscape:isstock="true"> + <path + inkscape:connector-curvature="0" + id="path862-3" + style="fill:#800080;fill-opacity:1;fill-rule:evenodd;stroke:#800080;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1" + d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z" + transform="scale(-0.6)" /> + </marker> + <marker + inkscape:stockid="Arrow2Mend" + orient="auto" + refY="0" + refX="0" + id="Arrow2Mend-2-5" + style="overflow:visible" + inkscape:isstock="true"> + <path + inkscape:connector-curvature="0" + id="path862-3-9" + style="fill:#800080;fill-opacity:1;fill-rule:evenodd;stroke:#800080;stroke-width:0.625;stroke-linejoin:round;stroke-opacity:1" + d="M 8.7185878,4.0337352 -2.2072895,0.01601326 8.7185884,-4.0017078 c -1.7454984,2.3720609 -1.7354408,5.6174519 -6e-7,8.035443 z" + transform="scale(-0.6)" /> + </marker> + </defs> + <sodipodi:namedview + id="base" + pagecolor="#ffffff" + bordercolor="#666666" + borderopacity="1.0" + inkscape:pageopacity="0.0" + inkscape:pageshadow="2" + inkscape:zoom="1.979899" + inkscape:cx="747.52324" + inkscape:cy="319.84503" + inkscape:document-units="mm" + inkscape:current-layer="layer1" + showgrid="true" + inkscape:window-width="3840" + inkscape:window-height="2096" + inkscape:window-x="0" + inkscape:window-y="27" + inkscape:window-maximized="1" + fit-margin-top="0" + fit-margin-left="0" + fit-margin-right="0" + fit-margin-bottom="0"> + <inkscape:grid + type="xygrid" + id="grid815" + originx="15.282997" + originy="-184.54792" /> + </sodipodi:namedview> + <metadata + id="metadata5"> + <rdf:RDF> + <cc:Work + rdf:about=""> + <dc:format>image/svg+xml</dc:format> + <dc:type + rdf:resource="http://purl.org/dc/dcmitype/StillImage" /> + <dc:title /> + </cc:Work> + </rdf:RDF> + </metadata> + <g + inkscape:label="Layer 1" + inkscape:groupmode="layer" + id="layer1" + transform="translate(15.282998,-16.277083)"> + <path + style="fill:none;stroke:#000000;stroke-width:0.5291667;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1" + d="M 15.875,27.125001 V 16.541666 H 26.458333 V 27.125001 H 177.27084 V 16.541666 h 10.58333 v 10.583335 h 29.10416" + id="path817" + inkscape:connector-curvature="0" + sodipodi:nodetypes="ccccccccc" /> + <path + style="fill:none;stroke:#000000;stroke-width:0.5291667;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1" + d="M 15.875,37.708334 H 44.979166 V 48.291667 H 100.54167 V 37.708334 H 206.375 v 10.583333 h 10.58333" + id="path819" + inkscape:connector-curvature="0" /> + <path + style="fill:none;stroke:#000000;stroke-width:0.5291667;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1" + d="m 15.875,66.8125 h 97.89583 V 56.229167 h 7.9375 V 66.8125 h 92.60417" + id="path821" + inkscape:connector-curvature="0" /> + <path + style="fill:none;stroke:#000000;stroke-width:0.5291667;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1" + d="m 15.875,85.333334 c 0,0 132.29166,0 132.29166,0 V 74.75 h 15.875 v 10.583334 h 47.625" + id="path823" + inkscape:connector-curvature="0" /> + <path + style="fill:none;stroke:#000000;stroke-width:0.5291667;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1" + d="M 15.875,101.20833 H 187.85416 V 90.625 h 10.58334 v 10.58333 h 10.58333" + id="path825" + inkscape:connector-curvature="0" /> + <path + style="fill:none;stroke:#000000;stroke-width:0.5291667;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:0.52916667, 0.52916667;stroke-dashoffset:0;stroke-opacity:1" + d="M 100.54167,48.291667 V 111.79167" + id="path827" + inkscape:connector-curvature="0" /> + <path + style="fill:none;stroke:#000000;stroke-width:0.5291667;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:0.52916667, 0.52916667;stroke-dashoffset:0;stroke-opacity:1" + d="m 113.77083,66.8125 v 44.97917" + id="path829" + inkscape:connector-curvature="0" /> + <path + style="fill:none;stroke:#000000;stroke-width:0.5291667;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:0.52916667, 0.52916667;stroke-dashoffset:0;stroke-opacity:1" + d="M 206.375,48.291667 V 109.14583" + id="path831" + inkscape:connector-curvature="0" /> + <path + style="fill:none;stroke:#ff0000;stroke-width:0.26458332px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1;marker-start:url(#Arrow1Lstart);marker-end:url(#Arrow1Lend)" + d="m 100.54167,106.5 h 13.22916" + id="path833" + inkscape:connector-curvature="0" /> + <path + style="fill:none;stroke:#ff0000;stroke-width:0.26458332px;stroke-linecap:butt;stroke-linejoin:miter;stroke-opacity:1;marker-start:url(#marker1253);marker-end:url(#marker1311)" + d="M 113.77083,106.5 H 206.375" + id="path1243" + inkscape:connector-curvature="0" /> + <text + xml:space="preserve" + style="font-style:normal;font-weight:normal;font-size:3.17499995px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.26458332" + x="105.83333" + y="111.79166" + id="text1405"><tspan + sodipodi:role="line" + id="tspan1403" + x="105.83333" + y="111.79166" + style="stroke-width:0.26458332">To</tspan></text> + <text + xml:space="preserve" + style="font-style:normal;font-weight:normal;font-size:3.17499995px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.26458332" + x="145.52083" + y="111.79166" + id="text1409"><tspan + sodipodi:role="line" + id="tspan1407" + x="145.52083" + y="111.79166" + style="stroke-width:0.26458332">VStartup Period</tspan></text> + <text + xml:space="preserve" + style="font-style:normal;font-weight:normal;font-size:3.17499995px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.26458332" + x="156.01123" + y="78.71875" + id="text1413"><tspan + sodipodi:role="line" + x="156.01123" + y="78.71875" + style="font-weight:bold;text-align:center;text-anchor:middle;stroke-width:0.26458332" + id="tspan1415">VUpdate</tspan><tspan + sodipodi:role="line" + x="156.01123" + y="82.6875" + style="font-weight:bold;text-align:center;text-anchor:middle;stroke-width:0.26458332" + id="tspan1440">Width</tspan></text> + <text + xml:space="preserve" + style="font-style:normal;font-weight:normal;font-size:3.17499995px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.26458332" + x="173.77611" + y="92.703873" + id="text1413-3"><tspan + sodipodi:role="line" + id="tspan1411-6" + x="173.77611" + y="92.703873" + style="font-weight:bold;text-align:center;text-anchor:middle;stroke-width:0.26458332">VReady</tspan><tspan + sodipodi:role="line" + x="173.77611" + y="96.672623" + style="font-weight:bold;text-align:center;text-anchor:middle;stroke-width:0.26458332" + id="tspan1415-7">Offset</tspan></text> + <text + xml:space="preserve" + style="font-style:normal;font-weight:normal;font-size:3.17499995px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.26458332" + x="135.78951" + y="70.78125" + id="text1413-5"><tspan + sodipodi:role="line" + x="135.78951" + y="70.78125" + style="font-weight:bold;text-align:center;text-anchor:middle;stroke-width:0.26458332" + id="tspan1440-5">VUpdate</tspan><tspan + sodipodi:role="line" + x="135.78951" + y="74.75" + style="font-weight:bold;text-align:center;text-anchor:middle;stroke-width:0.26458332" + id="tspan1465">Offset</tspan></text> + <text + xml:space="preserve" + style="font-style:normal;font-weight:normal;font-size:3.17499995px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.26458332" + x="137.39433" + y="48.291664" + id="text1479"><tspan + sodipodi:role="line" + id="tspan1477" + x="137.39433" + y="48.291664" + style="font-weight:bold;stroke-width:0.26458332">VSTARTUP_START</tspan></text> + <text + xml:space="preserve" + style="font-style:normal;font-weight:normal;font-size:3.17499995px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.26458332" + x="-5.4806676" + y="22.778271" + id="text1479-1"><tspan + sodipodi:role="line" + id="tspan1477-2" + x="-5.4806676" + y="22.778271" + style="font-weight:bold;font-size:4.93888903px;stroke-width:0.26458332">VSYNC</tspan></text> + <text + xml:space="preserve" + style="font-style:normal;font-weight:normal;font-size:3.17499995px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.26458332" + x="-9.3767252" + y="45.64584" + id="text1479-1-7"><tspan + sodipodi:role="line" + id="tspan1477-2-0" + x="-9.3767252" + y="45.64584" + style="font-weight:bold;font-size:5.64444447px;stroke-width:0.26458332">VBlank</tspan></text> + <text + xml:space="preserve" + style="font-style:normal;font-weight:normal;font-size:3.17499995px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.26458332" + x="-15.310558" + y="64.92263" + id="text1479-1-7-9"><tspan + sodipodi:role="line" + id="tspan1477-2-0-3" + x="-15.310558" + y="64.92263" + style="font-weight:bold;font-size:5.64444447px;stroke-width:0.26458332">VStartup</tspan></text> + <text + xml:space="preserve" + style="font-style:normal;font-weight:normal;font-size:3.17499995px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.26458332" + x="-14.17781" + y="85.144356" + id="text1479-1-7-9-6"><tspan + sodipodi:role="line" + id="tspan1477-2-0-3-0" + x="-14.17781" + y="85.144356" + style="font-weight:bold;font-size:5.64444447px;stroke-width:0.26458332">VUpdate</tspan></text> + <text + xml:space="preserve" + style="font-style:normal;font-weight:normal;font-size:3.17499995px;line-height:1.25;font-family:sans-serif;letter-spacing:0px;word-spacing:0px;fill:#000000;fill-opacity:1;stroke:none;stroke-width:0.26458332" + x="-11.052421" + y="101.39733" + id="text1479-1-7-9-6-6"><tspan + sodipodi:role="line" + id="tspan1477-2-0-3-0-2" + x="-11.052421" + y="101.39733" + style="font-weight:bold;font-size:5.64444447px;stroke-width:0.26458332">VReady</tspan></text> + <g + id="g5189" + transform="translate(269.875,-14.287499)"> + <path + sodipodi:nodetypes="cc" + inkscape:connector-curvature="0" + id="path5143" + d="m -202.40625,45.645828 3.96875,-7.9375" + style="fill:none;stroke:#000000;stroke-width:0.5291667;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1" /> + <path + sodipodi:nodetypes="cc" + inkscape:connector-curvature="0" + id="path5143-2" + d="m -199.76042,45.645828 3.96874,-7.937499" + style="fill:none;stroke:#000000;stroke-width:0.52916676;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1" /> + </g> + <g + id="g5189-3" + transform="translate(268.55209,7.9375003)"> + <path + sodipodi:nodetypes="cc" + inkscape:connector-curvature="0" + id="path5143-6" + d="m -202.40625,45.645828 3.96875,-7.9375" + style="fill:none;stroke:#000000;stroke-width:0.5291667;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1" /> + <path + sodipodi:nodetypes="cc" + inkscape:connector-curvature="0" + id="path5143-2-1" + d="m -199.76042,45.645828 3.96874,-7.937499" + style="fill:none;stroke:#000000;stroke-width:0.52916676;stroke-linecap:butt;stroke-linejoin:miter;stroke-miterlimit:4;stroke-dasharray:none;stroke-opacity:1" /> + </g> + </g> +</svg> diff --git a/Documentation/gpu/amdgpu/display/index.rst b/Documentation/gpu/amdgpu/display/index.rst new file mode 100644 index 000000000000..c1fb2fb3c710 --- /dev/null +++ b/Documentation/gpu/amdgpu/display/index.rst @@ -0,0 +1,31 @@ +.. _amdgpu-display-core: + +=================================== +drm/amd/display - Display Core (DC) +=================================== + +AMD display engine is partially shared with other operating systems; for this +reason, our Display Core Driver is divided into two pieces: + +1. **Display Core (DC)** contains the OS-agnostic components. Things like + hardware programming and resource management are handled here. +2. **Display Manager (DM)** contains the OS-dependent components. Hooks to the + amdgpu base driver and DRM are implemented here. + +The display pipe is responsible for "scanning out" a rendered frame from the +GPU memory (also called VRAM, FrameBuffer, etc.) to a display. In other words, +it would: + +1. Read frame information from memory; +2. Perform required transformation; +3. Send pixel data to sink devices. + +If you want to learn more about our driver details, take a look at the below +table of content: + +.. toctree:: + + display-manager.rst + dc-debug.rst + dcn-overview.rst + dc-glossary.rst diff --git a/Documentation/gpu/amdgpu/driver-core.rst b/Documentation/gpu/amdgpu/driver-core.rst new file mode 100644 index 000000000000..ebf5932845a9 --- /dev/null +++ b/Documentation/gpu/amdgpu/driver-core.rst @@ -0,0 +1,182 @@ +============================ + Core Driver Infrastructure +============================ + +GPU Hardware Structure +====================== + +Each ASIC is a collection of hardware blocks. We refer to them as +"IPs" (Intellectual Property blocks). Each IP encapsulates certain +functionality. IPs are versioned and can also be mixed and matched. +E.g., you might have two different ASICs that both have System DMA (SDMA) 5.x IPs. +The driver is arranged by IPs. There are driver components to handle +the initialization and operation of each IP. There are also a bunch +of smaller IPs that don't really need much if any driver interaction. +Those end up getting lumped into the common stuff in the soc files. +The soc files (e.g., vi.c, soc15.c nv.c) contain code for aspects of +the SoC itself rather than specific IPs. E.g., things like GPU resets +and register access functions are SoC dependent. + +An APU contains more than just CPU and GPU, it also contains all of +the platform stuff (audio, usb, gpio, etc.). Also, a lot of +components are shared between the CPU, platform, and the GPU (e.g., +SMU, PSP, etc.). Specific components (CPU, GPU, etc.) usually have +their interface to interact with those common components. For things +like S0i3 there is a ton of coordination required across all the +components, but that is probably a bit beyond the scope of this +section. + +With respect to the GPU, we have the following major IPs: + +GMC (Graphics Memory Controller) + This was a dedicated IP on older pre-vega chips, but has since + become somewhat decentralized on vega and newer chips. They now + have dedicated memory hubs for specific IPs or groups of IPs. We + still treat it as a single component in the driver however since + the programming model is still pretty similar. This is how the + different IPs on the GPU get the memory (VRAM or system memory). + It also provides the support for per process GPU virtual address + spaces. + +IH (Interrupt Handler) + This is the interrupt controller on the GPU. All of the IPs feed + their interrupts into this IP and it aggregates them into a set of + ring buffers that the driver can parse to handle interrupts from + different IPs. + +PSP (Platform Security Processor) + This handles security policy for the SoC and executes trusted + applications, and validates and loads firmwares for other blocks. + +SMU (System Management Unit) + This is the power management microcontroller. It manages the entire + SoC. The driver interacts with it to control power management + features like clocks, voltages, power rails, etc. + +DCN (Display Controller Next) + This is the display controller. It handles the display hardware. + It is described in more details in :ref:`Display Core <amdgpu-display-core>`. + +SDMA (System DMA) + This is a multi-purpose DMA engine. The kernel driver uses it for + various things including paging and GPU page table updates. It's also + exposed to userspace for use by user mode drivers (OpenGL, Vulkan, + etc.) + +GC (Graphics and Compute) + This is the graphics and compute engine, i.e., the block that + encompasses the 3D pipeline and and shader blocks. This is by far the + largest block on the GPU. The 3D pipeline has tons of sub-blocks. In + addition to that, it also contains the CP microcontrollers (ME, PFP, + CE, MEC) and the RLC microcontroller. It's exposed to userspace for + user mode drivers (OpenGL, Vulkan, OpenCL, etc.) + +VCN (Video Core Next) + This is the multi-media engine. It handles video and image encode and + decode. It's exposed to userspace for user mode drivers (VA-API, + OpenMAX, etc.) + +Graphics and Compute Microcontrollers +------------------------------------- + +CP (Command Processor) + The name for the hardware block that encompasses the front end of the + GFX/Compute pipeline. Consists mainly of a bunch of microcontrollers + (PFP, ME, CE, MEC). The firmware that runs on these microcontrollers + provides the driver interface to interact with the GFX/Compute engine. + + MEC (MicroEngine Compute) + This is the microcontroller that controls the compute queues on the + GFX/compute engine. + + MES (MicroEngine Scheduler) + This is a new engine for managing queues. This is currently unused. + +RLC (RunList Controller) + This is another microcontroller in the GFX/Compute engine. It handles + power management related functionality within the GFX/Compute engine. + The name is a vestige of old hardware where it was originally added + and doesn't really have much relation to what the engine does now. + +Driver Structure +================ + +In general, the driver has a list of all of the IPs on a particular +SoC and for things like init/fini/suspend/resume, more or less just +walks the list and handles each IP. + +Some useful constructs: + +KIQ (Kernel Interface Queue) + This is a control queue used by the kernel driver to manage other gfx + and compute queues on the GFX/compute engine. You can use it to + map/unmap additional queues, etc. + +IB (Indirect Buffer) + A command buffer for a particular engine. Rather than writing + commands directly to the queue, you can write the commands into a + piece of memory and then put a pointer to the memory into the queue. + The hardware will then follow the pointer and execute the commands in + the memory, then returning to the rest of the commands in the ring. + +.. _amdgpu_memory_domains: + +Memory Domains +============== + +.. kernel-doc:: include/uapi/drm/amdgpu_drm.h + :doc: memory domains + +Buffer Objects +============== + +.. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_object.c + :doc: amdgpu_object + +.. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_object.c + :internal: + +PRIME Buffer Sharing +==================== + +.. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c + :doc: PRIME Buffer Sharing + +.. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c + :internal: + +MMU Notifier +============ + +.. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_mn.c + :doc: MMU Notifier + +.. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_mn.c + :internal: + +AMDGPU Virtual Memory +===================== + +.. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c + :doc: GPUVM + +.. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c + :internal: + +Interrupt Handling +================== + +.. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c + :doc: Interrupt Handling + +.. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c + :internal: + +IP Blocks +========= + +.. kernel-doc:: drivers/gpu/drm/amd/include/amd_shared.h + :doc: IP Blocks + +.. kernel-doc:: drivers/gpu/drm/amd/include/amd_shared.h + :identifiers: amd_ip_block_type amd_ip_funcs diff --git a/Documentation/gpu/amdgpu/driver-misc.rst b/Documentation/gpu/amdgpu/driver-misc.rst new file mode 100644 index 000000000000..e3d6b2fa2493 --- /dev/null +++ b/Documentation/gpu/amdgpu/driver-misc.rst @@ -0,0 +1,112 @@ +================================ + Misc AMDGPU driver information +================================ + +GPU Product Information +======================= + +Information about the GPU can be obtained on certain cards +via sysfs + +product_name +------------ + +.. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_device.c + :doc: product_name + +product_number +-------------- + +.. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_device.c + :doc: product_name + +serial_number +------------- + +.. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_device.c + :doc: serial_number + +unique_id +--------- + +.. kernel-doc:: drivers/gpu/drm/amd/pm/amdgpu_pm.c + :doc: unique_id + +GPU Memory Usage Information +============================ + +Various memory accounting can be accessed via sysfs + +mem_info_vram_total +------------------- + +.. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c + :doc: mem_info_vram_total + +mem_info_vram_used +------------------ + +.. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c + :doc: mem_info_vram_used + +mem_info_vis_vram_total +----------------------- + +.. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c + :doc: mem_info_vis_vram_total + +mem_info_vis_vram_used +---------------------- + +.. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_vram_mgr.c + :doc: mem_info_vis_vram_used + +mem_info_gtt_total +------------------ + +.. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_gtt_mgr.c + :doc: mem_info_gtt_total + +mem_info_gtt_used +----------------- + +.. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_gtt_mgr.c + :doc: mem_info_gtt_used + +PCIe Accounting Information +=========================== + +pcie_bw +------- + +.. kernel-doc:: drivers/gpu/drm/amd/pm/amdgpu_pm.c + :doc: pcie_bw + +pcie_replay_count +----------------- + +.. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_device.c + :doc: pcie_replay_count + +GPU SmartShift Information +========================== + +GPU SmartShift information via sysfs + +smartshift_apu_power +-------------------- + +.. kernel-doc:: drivers/gpu/drm/amd/pm/amdgpu_pm.c + :doc: smartshift_apu_power + +smartshift_dgpu_power +--------------------- + +.. kernel-doc:: drivers/gpu/drm/amd/pm/amdgpu_pm.c + :doc: smartshift_dgpu_power + +smartshift_bias +--------------- + +.. kernel-doc:: drivers/gpu/drm/amd/pm/amdgpu_pm.c + :doc: smartshift_bias diff --git a/Documentation/gpu/amdgpu/index.rst b/Documentation/gpu/amdgpu/index.rst new file mode 100644 index 000000000000..a24e1cfa7407 --- /dev/null +++ b/Documentation/gpu/amdgpu/index.rst @@ -0,0 +1,17 @@ +========================== + drm/amdgpu AMDgpu driver +========================== + +The drm/amdgpu driver supports all AMD Radeon GPUs based on the Graphics Core +Next (GCN) architecture. + +.. toctree:: + + module-parameters + driver-core + display/index + xgmi + ras + thermal + driver-misc + amdgpu-glossary diff --git a/Documentation/gpu/amdgpu/module-parameters.rst b/Documentation/gpu/amdgpu/module-parameters.rst new file mode 100644 index 000000000000..ea538c8dda35 --- /dev/null +++ b/Documentation/gpu/amdgpu/module-parameters.rst @@ -0,0 +1,7 @@ +=================== + Module Parameters +=================== + +The amdgpu driver supports the following module parameters: + +.. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c diff --git a/Documentation/gpu/amdgpu/ras.rst b/Documentation/gpu/amdgpu/ras.rst new file mode 100644 index 000000000000..047f76e395cf --- /dev/null +++ b/Documentation/gpu/amdgpu/ras.rst @@ -0,0 +1,62 @@ +==================== + AMDGPU RAS Support +==================== + +The AMDGPU RAS interfaces are exposed via sysfs (for informational queries) and +debugfs (for error injection). + +RAS debugfs/sysfs Control and Error Injection Interfaces +======================================================== + +.. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c + :doc: AMDGPU RAS debugfs control interface + +RAS Reboot Behavior for Unrecoverable Errors +============================================ + +.. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c + :doc: AMDGPU RAS Reboot Behavior for Unrecoverable Errors + +RAS Error Count sysfs Interface +=============================== + +.. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c + :doc: AMDGPU RAS sysfs Error Count Interface + +RAS EEPROM debugfs Interface +============================ + +.. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c + :doc: AMDGPU RAS debugfs EEPROM table reset interface + +RAS VRAM Bad Pages sysfs Interface +================================== + +.. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c + :doc: AMDGPU RAS sysfs gpu_vram_bad_pages Interface + +Sample Code +=========== +Sample code for testing error injection can be found here: +https://cgit.freedesktop.org/mesa/drm/tree/tests/amdgpu/ras_tests.c + +This is part of the libdrm amdgpu unit tests which cover several areas of the GPU. +There are four sets of tests: + +RAS Basic Test + +The test verifies the RAS feature enabled status and makes sure the necessary sysfs and debugfs files +are present. + +RAS Query Test + +This test checks the RAS availability and enablement status for each supported IP block as well as +the error counts. + +RAS Inject Test + +This test injects errors for each IP. + +RAS Disable Test + +This test tests disabling of RAS features for each IP block. diff --git a/Documentation/gpu/amdgpu/thermal.rst b/Documentation/gpu/amdgpu/thermal.rst new file mode 100644 index 000000000000..8aeb0186c9ef --- /dev/null +++ b/Documentation/gpu/amdgpu/thermal.rst @@ -0,0 +1,65 @@ +=========================================== + GPU Power/Thermal Controls and Monitoring +=========================================== + +HWMON Interfaces +================ + +.. kernel-doc:: drivers/gpu/drm/amd/pm/amdgpu_pm.c + :doc: hwmon + +GPU sysfs Power State Interfaces +================================ + +GPU power controls are exposed via sysfs files. + +power_dpm_state +--------------- + +.. kernel-doc:: drivers/gpu/drm/amd/pm/amdgpu_pm.c + :doc: power_dpm_state + +power_dpm_force_performance_level +--------------------------------- + +.. kernel-doc:: drivers/gpu/drm/amd/pm/amdgpu_pm.c + :doc: power_dpm_force_performance_level + +pp_table +-------- + +.. kernel-doc:: drivers/gpu/drm/amd/pm/amdgpu_pm.c + :doc: pp_table + +pp_od_clk_voltage +----------------- + +.. kernel-doc:: drivers/gpu/drm/amd/pm/amdgpu_pm.c + :doc: pp_od_clk_voltage + +pp_dpm_* +-------- + +.. kernel-doc:: drivers/gpu/drm/amd/pm/amdgpu_pm.c + :doc: pp_dpm_sclk pp_dpm_mclk pp_dpm_socclk pp_dpm_fclk pp_dpm_dcefclk pp_dpm_pcie + +pp_power_profile_mode +--------------------- + +.. kernel-doc:: drivers/gpu/drm/amd/pm/amdgpu_pm.c + :doc: pp_power_profile_mode + +\*_busy_percent +--------------- + +.. kernel-doc:: drivers/gpu/drm/amd/pm/amdgpu_pm.c + :doc: gpu_busy_percent + +.. kernel-doc:: drivers/gpu/drm/amd/pm/amdgpu_pm.c + :doc: mem_busy_percent + +gpu_metrics +----------- + +.. kernel-doc:: drivers/gpu/drm/amd/pm/amdgpu_pm.c + :doc: gpu_metrics diff --git a/Documentation/gpu/amdgpu/xgmi.rst b/Documentation/gpu/amdgpu/xgmi.rst new file mode 100644 index 000000000000..23f2856f4524 --- /dev/null +++ b/Documentation/gpu/amdgpu/xgmi.rst @@ -0,0 +1,5 @@ +===================== + AMDGPU XGMI Support +===================== + +.. kernel-doc:: drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c diff --git a/Documentation/gpu/drivers.rst b/Documentation/gpu/drivers.rst index b4a0ed3ca961..3a52f48215a3 100644 --- a/Documentation/gpu/drivers.rst +++ b/Documentation/gpu/drivers.rst @@ -4,8 +4,7 @@ GPU Driver Documentation .. toctree:: - amdgpu - amdgpu-dc + amdgpu/index i915 mcde meson diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu.h b/drivers/gpu/drm/amd/amdgpu/amdgpu.h index 9f017663ac50..9eb7afa276f2 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu.h @@ -1095,7 +1095,9 @@ struct amdgpu_device { pci_channel_state_t pci_channel_state; struct amdgpu_reset_control *reset_cntl; - uint32_t ip_versions[HW_ID_MAX][HWIP_MAX_INSTANCE]; + uint32_t ip_versions[MAX_HWIP][HWIP_MAX_INSTANCE]; + + bool ram_is_direct_mapped; }; static inline struct amdgpu_device *drm_to_adev(struct drm_device *ddev) @@ -1316,6 +1318,8 @@ void amdgpu_device_flush_hdp(struct amdgpu_device *adev, void amdgpu_device_invalidate_hdp(struct amdgpu_device *adev, struct amdgpu_ring *ring); +void amdgpu_device_halt(struct amdgpu_device *adev); + /* atpx handler */ #if defined(CONFIG_VGA_SWITCHEROO) void amdgpu_register_atpx_handler(void); @@ -1359,8 +1363,6 @@ int amdgpu_device_resume(struct drm_device *dev, bool fbcon); u32 amdgpu_get_vblank_counter_kms(struct drm_crtc *crtc); int amdgpu_enable_vblank_kms(struct drm_crtc *crtc); void amdgpu_disable_vblank_kms(struct drm_crtc *crtc); -long amdgpu_kms_compat_ioctl(struct file *filp, unsigned int cmd, - unsigned long arg); int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file *filp); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c index 5df89a295177..f9bab963a948 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gpuvm.c @@ -708,10 +708,12 @@ static int kfd_mem_attach(struct amdgpu_device *adev, struct kgd_mem *mem, pr_debug("\t add VA 0x%llx - 0x%llx to vm %p\n", va, va + bo_size, vm); - if (adev == bo_adev || (mem->domain == AMDGPU_GEM_DOMAIN_VRAM && - amdgpu_xgmi_same_hive(adev, bo_adev))) { - /* Mappings on the local GPU and VRAM mappings in the - * local hive share the original BO + if (adev == bo_adev || + (amdgpu_ttm_tt_get_usermm(mem->bo->tbo.ttm) && adev->ram_is_direct_mapped) || + (mem->domain == AMDGPU_GEM_DOMAIN_VRAM && amdgpu_xgmi_same_hive(adev, bo_adev))) { + /* Mappings on the local GPU, or VRAM mappings in the + * local hive, or userptr mapping IOMMU direct map mode + * share the original BO */ attachment[i]->type = KFD_MEM_ATT_SHARED; bo[i] = mem->bo; @@ -1559,13 +1561,8 @@ int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu( ret = init_user_pages(*mem, user_addr); if (ret) goto allocate_init_user_pages_failed; - } - - if (offset) - *offset = amdgpu_bo_mmap_offset(bo); - - if (flags & (KFD_IOC_ALLOC_MEM_FLAGS_DOORBELL | - KFD_IOC_ALLOC_MEM_FLAGS_MMIO_REMAP)) { + } else if (flags & (KFD_IOC_ALLOC_MEM_FLAGS_DOORBELL | + KFD_IOC_ALLOC_MEM_FLAGS_MMIO_REMAP)) { ret = amdgpu_amdkfd_gpuvm_pin_bo(bo, AMDGPU_GEM_DOMAIN_GTT); if (ret) { pr_err("Pinning MMIO/DOORBELL BO during ALLOC FAILED\n"); @@ -1575,11 +1572,14 @@ int amdgpu_amdkfd_gpuvm_alloc_memory_of_gpu( bo->preferred_domains = AMDGPU_GEM_DOMAIN_GTT; } + if (offset) + *offset = amdgpu_bo_mmap_offset(bo); + return 0; allocate_init_user_pages_failed: - remove_kgd_mem_from_kfd_bo_list(*mem, avm->process_info); err_pin_bo: + remove_kgd_mem_from_kfd_bo_list(*mem, avm->process_info); drm_vma_node_revoke(&gobj->vma_node, drm_priv); err_node_allow: /* Don't unreserve system mem limit twice */ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c index 97178b307ed6..4d4ddf026faf 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_atomfirmware.c @@ -470,8 +470,8 @@ bool amdgpu_atomfirmware_dynamic_boot_config_supported(struct amdgpu_device *ade /** * amdgpu_atomfirmware_ras_rom_addr -- Get the RAS EEPROM addr from VBIOS - * adev: amdgpu_device pointer - * i2c_address: pointer to u8; if not NULL, will contain + * @adev: amdgpu_device pointer + * @i2c_address: pointer to u8; if not NULL, will contain * the RAS EEPROM address if the function returns true * * Return true if VBIOS supports RAS EEPROM address reporting, diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c index 7abe9500c0c6..d6d986be906a 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_atpx_handler.c @@ -11,6 +11,7 @@ #include <linux/pci.h> #include <linux/delay.h> +#include "amdgpu.h" #include "amd_acpi.h" #define AMDGPU_PX_QUIRK_FORCE_ATPX (1 << 0) @@ -165,7 +166,7 @@ static void amdgpu_atpx_parse_functions(struct amdgpu_atpx_functions *f, u32 mas } /** - * amdgpu_atpx_validate_functions - validate ATPX functions + * amdgpu_atpx_validate - validate ATPX functions * * @atpx: amdgpu atpx struct * diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c index 0de66f59adb8..c16a2704ced6 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_connectors.c @@ -108,7 +108,7 @@ int amdgpu_connector_get_monitor_bpc(struct drm_connector *connector) case DRM_MODE_CONNECTOR_DVII: case DRM_MODE_CONNECTOR_HDMIB: if (amdgpu_connector->use_digital) { - if (drm_detect_hdmi_monitor(amdgpu_connector_edid(connector))) { + if (connector->display_info.is_hdmi) { if (connector->display_info.bpc) bpc = connector->display_info.bpc; } @@ -116,7 +116,7 @@ int amdgpu_connector_get_monitor_bpc(struct drm_connector *connector) break; case DRM_MODE_CONNECTOR_DVID: case DRM_MODE_CONNECTOR_HDMIA: - if (drm_detect_hdmi_monitor(amdgpu_connector_edid(connector))) { + if (connector->display_info.is_hdmi) { if (connector->display_info.bpc) bpc = connector->display_info.bpc; } @@ -125,7 +125,7 @@ int amdgpu_connector_get_monitor_bpc(struct drm_connector *connector) dig_connector = amdgpu_connector->con_priv; if ((dig_connector->dp_sink_type == CONNECTOR_OBJECT_ID_DISPLAYPORT) || (dig_connector->dp_sink_type == CONNECTOR_OBJECT_ID_eDP) || - drm_detect_hdmi_monitor(amdgpu_connector_edid(connector))) { + connector->display_info.is_hdmi) { if (connector->display_info.bpc) bpc = connector->display_info.bpc; } @@ -149,7 +149,7 @@ int amdgpu_connector_get_monitor_bpc(struct drm_connector *connector) break; } - if (drm_detect_hdmi_monitor(amdgpu_connector_edid(connector))) { + if (connector->display_info.is_hdmi) { /* * Pre DCE-8 hw can't handle > 12 bpc, and more than 12 bpc doesn't make * much sense without support for > 12 bpc framebuffers. RGB 4:4:4 at @@ -315,8 +315,10 @@ static void amdgpu_connector_get_edid(struct drm_connector *connector) if (!amdgpu_connector->edid) { /* some laptops provide a hardcoded edid in rom for LCDs */ if (((connector->connector_type == DRM_MODE_CONNECTOR_LVDS) || - (connector->connector_type == DRM_MODE_CONNECTOR_eDP))) + (connector->connector_type == DRM_MODE_CONNECTOR_eDP))) { amdgpu_connector->edid = amdgpu_connector_get_hardcoded_edid(adev); + drm_connector_update_edid_property(connector, amdgpu_connector->edid); + } } } @@ -326,6 +328,7 @@ static void amdgpu_connector_free_edid(struct drm_connector *connector) kfree(amdgpu_connector->edid); amdgpu_connector->edid = NULL; + drm_connector_update_edid_property(connector, NULL); } static int amdgpu_connector_ddc_get_modes(struct drm_connector *connector) @@ -387,6 +390,9 @@ amdgpu_connector_lcd_native_mode(struct drm_encoder *encoder) native_mode->vdisplay != 0 && native_mode->clock != 0) { mode = drm_mode_duplicate(dev, native_mode); + if (!mode) + return NULL; + mode->type = DRM_MODE_TYPE_PREFERRED | DRM_MODE_TYPE_DRIVER; drm_mode_set_name(mode); @@ -401,6 +407,9 @@ amdgpu_connector_lcd_native_mode(struct drm_encoder *encoder) * simpler. */ mode = drm_cvt_mode(dev, native_mode->hdisplay, native_mode->vdisplay, 60, true, false, false); + if (!mode) + return NULL; + mode->type = DRM_MODE_TYPE_PREFERRED | DRM_MODE_TYPE_DRIVER; DRM_DEBUG_KMS("Adding cvt approximation of native panel mode %s\n", mode->name); } @@ -1171,7 +1180,7 @@ static enum drm_mode_status amdgpu_connector_dvi_mode_valid(struct drm_connector (amdgpu_connector->connector_object_id == CONNECTOR_OBJECT_ID_DUAL_LINK_DVI_D) || (amdgpu_connector->connector_object_id == CONNECTOR_OBJECT_ID_HDMI_TYPE_B)) { return MODE_OK; - } else if (drm_detect_hdmi_monitor(amdgpu_connector_edid(connector))) { + } else if (connector->display_info.is_hdmi) { /* HDMI 1.3+ supports max clock of 340 Mhz */ if (mode->clock > 340000) return MODE_CLOCK_HIGH; @@ -1463,7 +1472,7 @@ static enum drm_mode_status amdgpu_connector_dp_mode_valid(struct drm_connector (amdgpu_dig_connector->dp_sink_type == CONNECTOR_OBJECT_ID_eDP)) { return amdgpu_atombios_dp_mode_valid_helper(connector, mode); } else { - if (drm_detect_hdmi_monitor(amdgpu_connector_edid(connector))) { + if (connector->display_info.is_hdmi) { /* HDMI 1.3+ supports max clock of 340 Mhz */ if (mode->clock > 340000) return MODE_CLOCK_HIGH; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c index 164d6a9e9fbb..25e2e5bf90eb 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_debugfs.c @@ -1618,6 +1618,9 @@ int amdgpu_debugfs_init(struct amdgpu_device *adev) if (!debugfs_initialized()) return 0; + debugfs_create_x32("amdgpu_smu_debug", 0600, root, + &adev->pm.smu_debug_mask); + ent = debugfs_create_file("amdgpu_preempt_ib", 0600, root, adev, &fops_ib_preempt); if (IS_ERR(ent)) { diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c index 90d22a376632..8219e40240ce 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c @@ -30,6 +30,7 @@ #include <linux/module.h> #include <linux/console.h> #include <linux/slab.h> +#include <linux/iommu.h> #include <drm/drm_atomic_helper.h> #include <drm/drm_probe_helper.h> @@ -331,7 +332,7 @@ void amdgpu_device_mm_access(struct amdgpu_device *adev, loff_t pos, } /** - * amdgpu_device_vram_access - access vram by vram aperature + * amdgpu_device_aper_access - access vram by vram aperature * * @adev: amdgpu_device pointer * @pos: offset of the buffer in vram @@ -550,11 +551,11 @@ void amdgpu_device_wreg(struct amdgpu_device *adev, trace_amdgpu_device_wreg(adev->pdev->device, reg, v); } -/* +/** * amdgpu_mm_wreg_mmio_rlc - write register either with mmio or with RLC path if in range * * this function is invoked only the debugfs register access - * */ + */ void amdgpu_mm_wreg_mmio_rlc(struct amdgpu_device *adev, uint32_t reg, uint32_t v) { @@ -1100,7 +1101,7 @@ static void amdgpu_device_wb_fini(struct amdgpu_device *adev) } /** - * amdgpu_device_wb_init- Init Writeback driver info and allocate memory + * amdgpu_device_wb_init - Init Writeback driver info and allocate memory * * @adev: amdgpu_device pointer * @@ -2657,6 +2658,36 @@ static int amdgpu_device_ip_late_init(struct amdgpu_device *adev) return 0; } +/** + * amdgpu_device_smu_fini_early - smu hw_fini wrapper + * + * @adev: amdgpu_device pointer + * + * For ASICs need to disable SMC first + */ +static void amdgpu_device_smu_fini_early(struct amdgpu_device *adev) +{ + int i, r; + + if (adev->ip_versions[GC_HWIP][0] > IP_VERSION(9, 0, 0)) + return; + + for (i = 0; i < adev->num_ip_blocks; i++) { + if (!adev->ip_blocks[i].status.hw) + continue; + if (adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_SMC) { + r = adev->ip_blocks[i].version->funcs->hw_fini((void *)adev); + /* XXX handle errors */ + if (r) { + DRM_DEBUG("hw_fini of IP block <%s> failed %d\n", + adev->ip_blocks[i].version->funcs->name, r); + } + adev->ip_blocks[i].status.hw = false; + break; + } + } +} + static int amdgpu_device_ip_fini_early(struct amdgpu_device *adev) { int i, r; @@ -2677,21 +2708,8 @@ static int amdgpu_device_ip_fini_early(struct amdgpu_device *adev) amdgpu_device_set_pg_state(adev, AMD_PG_STATE_UNGATE); amdgpu_device_set_cg_state(adev, AMD_CG_STATE_UNGATE); - /* need to disable SMC first */ - for (i = 0; i < adev->num_ip_blocks; i++) { - if (!adev->ip_blocks[i].status.hw) - continue; - if (adev->ip_blocks[i].version->type == AMD_IP_BLOCK_TYPE_SMC) { - r = adev->ip_blocks[i].version->funcs->hw_fini((void *)adev); - /* XXX handle errors */ - if (r) { - DRM_DEBUG("hw_fini of IP block <%s> failed %d\n", - adev->ip_blocks[i].version->funcs->name, r); - } - adev->ip_blocks[i].status.hw = false; - break; - } - } + /* Workaroud for ASICs need to disable SMC first */ + amdgpu_device_smu_fini_early(adev); for (i = adev->num_ip_blocks - 1; i >= 0; i--) { if (!adev->ip_blocks[i].status.hw) @@ -2733,8 +2751,6 @@ static int amdgpu_device_ip_fini(struct amdgpu_device *adev) if (amdgpu_sriov_vf(adev) && adev->virt.ras_init_done) amdgpu_virt_release_ras_err_handler_data(adev); - amdgpu_ras_pre_fini(adev); - if (adev->gmc.xgmi.num_physical_nodes > 1) amdgpu_xgmi_remove_device(adev); @@ -3367,6 +3383,22 @@ static int amdgpu_device_get_job_timeout_settings(struct amdgpu_device *adev) return ret; } +/** + * amdgpu_device_check_iommu_direct_map - check if RAM direct mapped to GPU + * + * @adev: amdgpu_device pointer + * + * RAM direct mapped to GPU if IOMMU is not enabled or is pass through mode + */ +static void amdgpu_device_check_iommu_direct_map(struct amdgpu_device *adev) +{ + struct iommu_domain *domain; + + domain = iommu_get_domain_for_dev(adev->dev); + if (!domain || domain->type == IOMMU_DOMAIN_IDENTITY) + adev->ram_is_direct_mapped = true; +} + static const struct attribute *amdgpu_dev_attributes[] = { &dev_attr_product_name.attr, &dev_attr_product_number.attr, @@ -3770,6 +3802,8 @@ fence_driver_init: queue_delayed_work(system_wq, &mgpu_info.delayed_reset_work, msecs_to_jiffies(AMDGPU_RESUME_MS)); + amdgpu_device_check_iommu_direct_map(adev); + return 0; release_ras_con: @@ -3803,7 +3837,7 @@ static void amdgpu_device_unmap_mmio(struct amdgpu_device *adev) } /** - * amdgpu_device_fini - tear down the driver + * amdgpu_device_fini_hw - tear down the driver * * @adev: amdgpu_device pointer * @@ -3844,6 +3878,9 @@ void amdgpu_device_fini_hw(struct amdgpu_device *adev) amdgpu_ucode_sysfs_fini(adev); sysfs_remove_files(&adev->dev->kobj, amdgpu_dev_attributes); + /* disable ras feature must before hw fini */ + amdgpu_ras_pre_fini(adev); + amdgpu_device_ip_fini_early(adev); amdgpu_irq_fini_hw(adev); @@ -4284,6 +4321,9 @@ static int amdgpu_device_reset_sriov(struct amdgpu_device *adev, bool from_hypervisor) { int r; + struct amdgpu_hive_info *hive = NULL; + + amdgpu_amdkfd_pre_reset(adev); amdgpu_amdkfd_pre_reset(adev); @@ -4312,8 +4352,19 @@ static int amdgpu_device_reset_sriov(struct amdgpu_device *adev, if (r) goto error; - amdgpu_irq_gpu_reset_resume_helper(adev); - r = amdgpu_ib_ring_tests(adev); + hive = amdgpu_get_xgmi_hive(adev); + /* Update PSP FW topology after reset */ + if (hive && adev->gmc.xgmi.num_physical_nodes > 1) + r = amdgpu_xgmi_update_topology(hive, adev); + + if (hive) + amdgpu_put_xgmi_hive(hive); + + if (!r) { + amdgpu_irq_gpu_reset_resume_helper(adev); + r = amdgpu_ib_ring_tests(adev); + amdgpu_amdkfd_post_reset(adev); + } error: if (!r && adev->virt.gim_feature & AMDGIM_FEATURE_GIM_FLR_VRAMLOST) { @@ -4745,7 +4796,7 @@ static int amdgpu_device_lock_hive_adev(struct amdgpu_device *adev, struct amdgp { struct amdgpu_device *tmp_adev = NULL; - if (adev->gmc.xgmi.num_physical_nodes > 1) { + if (!amdgpu_sriov_vf(adev) && (adev->gmc.xgmi.num_physical_nodes > 1)) { if (!hive) { dev_err(adev->dev, "Hive is NULL while device has multiple xgmi nodes"); return -ENODEV; @@ -4957,7 +5008,8 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev, * We always reset all schedulers for device and all devices for XGMI * hive so that should take care of them too. */ - hive = amdgpu_get_xgmi_hive(adev); + if (!amdgpu_sriov_vf(adev)) + hive = amdgpu_get_xgmi_hive(adev); if (hive) { if (atomic_cmpxchg(&hive->in_reset, 0, 1) != 0) { DRM_INFO("Bailing on TDR for s_job:%llx, hive: %llx as another already in progress", @@ -4998,7 +5050,7 @@ int amdgpu_device_gpu_recover(struct amdgpu_device *adev, * to put adev in the 1st position. */ INIT_LIST_HEAD(&device_list); - if (adev->gmc.xgmi.num_physical_nodes > 1) { + if (!amdgpu_sriov_vf(adev) && (adev->gmc.xgmi.num_physical_nodes > 1)) { list_for_each_entry(tmp_adev, &hive->device_list, gmc.xgmi.head) list_add_tail(&tmp_adev->reset_list, &device_list); if (!list_is_first(&adev->reset_list, &device_list)) @@ -5632,3 +5684,42 @@ void amdgpu_device_invalidate_hdp(struct amdgpu_device *adev, amdgpu_asic_invalidate_hdp(adev, ring); } + +/** + * amdgpu_device_halt() - bring hardware to some kind of halt state + * + * @adev: amdgpu_device pointer + * + * Bring hardware to some kind of halt state so that no one can touch it + * any more. It will help to maintain error context when error occurred. + * Compare to a simple hang, the system will keep stable at least for SSH + * access. Then it should be trivial to inspect the hardware state and + * see what's going on. Implemented as following: + * + * 1. drm_dev_unplug() makes device inaccessible to user space(IOCTLs, etc), + * clears all CPU mappings to device, disallows remappings through page faults + * 2. amdgpu_irq_disable_all() disables all interrupts + * 3. amdgpu_fence_driver_hw_fini() signals all HW fences + * 4. set adev->no_hw_access to avoid potential crashes after setp 5 + * 5. amdgpu_device_unmap_mmio() clears all MMIO mappings + * 6. pci_disable_device() and pci_wait_for_pending_transaction() + * flush any in flight DMA operations + */ +void amdgpu_device_halt(struct amdgpu_device *adev) +{ + struct pci_dev *pdev = adev->pdev; + struct drm_device *ddev = adev_to_drm(adev); + + drm_dev_unplug(ddev); + + amdgpu_irq_disable_all(adev); + + amdgpu_fence_driver_hw_fini(adev); + + adev->no_hw_access = true; + + amdgpu_device_unmap_mmio(adev); + + pci_disable_device(pdev); + pci_wait_for_pending_transaction(pdev); +} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c index ea00090b3fb3..5bc072220f03 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.c @@ -67,7 +67,8 @@ #include "smuio_v11_0_6.h" #include "smuio_v13_0.h" -MODULE_FIRMWARE("amdgpu/ip_discovery.bin"); +#define FIRMWARE_IP_DISCOVERY "amdgpu/ip_discovery.bin" +MODULE_FIRMWARE(FIRMWARE_IP_DISCOVERY); #define mmRCC_CONFIG_MEMSIZE 0xde3 #define mmMM_INDEX 0x0 @@ -179,7 +180,7 @@ static int hw_id_map[MAX_HWIP] = { [DCI_HWIP] = DCI_HWID, }; -static int amdgpu_discovery_read_binary(struct amdgpu_device *adev, uint8_t *binary) +static int amdgpu_discovery_read_binary_from_vram(struct amdgpu_device *adev, uint8_t *binary) { uint64_t vram_size = (uint64_t)RREG32(mmRCC_CONFIG_MEMSIZE) << 20; uint64_t pos = vram_size - DISCOVERY_TMR_OFFSET; @@ -189,6 +190,34 @@ static int amdgpu_discovery_read_binary(struct amdgpu_device *adev, uint8_t *bin return 0; } +static int amdgpu_discovery_read_binary_from_file(struct amdgpu_device *adev, uint8_t *binary) +{ + const struct firmware *fw; + const char *fw_name; + int r; + + switch (amdgpu_discovery) { + case 2: + fw_name = FIRMWARE_IP_DISCOVERY; + break; + default: + dev_warn(adev->dev, "amdgpu_discovery is not set properly\n"); + return -EINVAL; + } + + r = request_firmware(&fw, fw_name, adev->dev); + if (r) { + dev_err(adev->dev, "can't load firmware \"%s\"\n", + fw_name); + return r; + } + + memcpy((u8 *)binary, (u8 *)fw->data, adev->mman.discovery_tmr_size); + release_firmware(fw); + + return 0; +} + static uint16_t amdgpu_discovery_calculate_checksum(uint8_t *data, uint32_t size) { uint16_t checksum = 0; @@ -206,13 +235,20 @@ static inline bool amdgpu_discovery_verify_checksum(uint8_t *data, uint32_t size return !!(amdgpu_discovery_calculate_checksum(data, size) == expected); } +static inline bool amdgpu_discovery_verify_binary_signature(uint8_t *binary) +{ + struct binary_header *bhdr; + bhdr = (struct binary_header *)binary; + + return (le32_to_cpu(bhdr->binary_signature) == BINARY_SIGNATURE); +} + static int amdgpu_discovery_init(struct amdgpu_device *adev) { struct table_info *info; struct binary_header *bhdr; struct ip_discovery_header *ihdr; struct gpu_info_header *ghdr; - const struct firmware *fw; uint16_t offset; uint16_t size; uint16_t checksum; @@ -223,31 +259,32 @@ static int amdgpu_discovery_init(struct amdgpu_device *adev) if (!adev->mman.discovery_bin) return -ENOMEM; - if (amdgpu_discovery == 2) { - r = request_firmware(&fw, "amdgpu/ip_discovery.bin", adev->dev); - if (r) - goto get_from_vram; - dev_info(adev->dev, "Using IP discovery from file\n"); - memcpy((u8 *)adev->mman.discovery_bin, (u8 *)fw->data, - adev->mman.discovery_tmr_size); - release_firmware(fw); - } else { -get_from_vram: - r = amdgpu_discovery_read_binary(adev, adev->mman.discovery_bin); + r = amdgpu_discovery_read_binary_from_vram(adev, adev->mman.discovery_bin); + if (r) { + dev_err(adev->dev, "failed to read ip discovery binary from vram\n"); + r = -EINVAL; + goto out; + } + + if(!amdgpu_discovery_verify_binary_signature(adev->mman.discovery_bin)) { + dev_warn(adev->dev, "get invalid ip discovery binary signature from vram\n"); + /* retry read ip discovery binary from file */ + r = amdgpu_discovery_read_binary_from_file(adev, adev->mman.discovery_bin); if (r) { - DRM_ERROR("failed to read ip discovery binary\n"); + dev_err(adev->dev, "failed to read ip discovery binary from file\n"); + r = -EINVAL; + goto out; + } + /* check the ip discovery binary signature */ + if(!amdgpu_discovery_verify_binary_signature(adev->mman.discovery_bin)) { + dev_warn(adev->dev, "get invalid ip discovery binary signature from file\n"); + r = -EINVAL; goto out; } } bhdr = (struct binary_header *)adev->mman.discovery_bin; - if (le32_to_cpu(bhdr->binary_signature) != BINARY_SIGNATURE) { - DRM_ERROR("invalid ip discovery binary signature\n"); - r = -EINVAL; - goto out; - } - offset = offsetof(struct binary_header, binary_checksum) + sizeof(bhdr->binary_checksum); size = le16_to_cpu(bhdr->binary_size) - offset; @@ -255,7 +292,7 @@ get_from_vram: if (!amdgpu_discovery_verify_checksum(adev->mman.discovery_bin + offset, size, checksum)) { - DRM_ERROR("invalid ip discovery binary checksum\n"); + dev_err(adev->dev, "invalid ip discovery binary checksum\n"); r = -EINVAL; goto out; } @@ -266,14 +303,14 @@ get_from_vram: ihdr = (struct ip_discovery_header *)(adev->mman.discovery_bin + offset); if (le32_to_cpu(ihdr->signature) != DISCOVERY_TABLE_SIGNATURE) { - DRM_ERROR("invalid ip discovery data table signature\n"); + dev_err(adev->dev, "invalid ip discovery data table signature\n"); r = -EINVAL; goto out; } if (!amdgpu_discovery_verify_checksum(adev->mman.discovery_bin + offset, le16_to_cpu(ihdr->size), checksum)) { - DRM_ERROR("invalid ip discovery data table checksum\n"); + dev_err(adev->dev, "invalid ip discovery data table checksum\n"); r = -EINVAL; goto out; } @@ -285,7 +322,7 @@ get_from_vram: if (!amdgpu_discovery_verify_checksum(adev->mman.discovery_bin + offset, le32_to_cpu(ghdr->size), checksum)) { - DRM_ERROR("invalid gc data table checksum\n"); + dev_err(adev->dev, "invalid gc data table checksum\n"); r = -EINVAL; goto out; } @@ -379,8 +416,18 @@ int amdgpu_discovery_reg_base_init(struct amdgpu_device *adev) ip->major, ip->minor, ip->revision); - if (le16_to_cpu(ip->hw_id) == VCN_HWID) + if (le16_to_cpu(ip->hw_id) == VCN_HWID) { + /* Bit [5:0]: original revision value + * Bit [7:6]: en/decode capability: + * 0b00 : VCN function normally + * 0b10 : encode is disabled + * 0b01 : decode is disabled + */ + adev->vcn.vcn_config[adev->vcn.num_vcn_inst] = + ip->revision & 0xc0; + ip->revision &= ~0xc0; adev->vcn.num_vcn_inst++; + } if (le16_to_cpu(ip->hw_id) == SDMA0_HWID || le16_to_cpu(ip->hw_id) == SDMA1_HWID || le16_to_cpu(ip->hw_id) == SDMA2_HWID || @@ -472,14 +519,6 @@ int amdgpu_discovery_get_ip_version(struct amdgpu_device *adev, int hw_id, int n return -EINVAL; } - -int amdgpu_discovery_get_vcn_version(struct amdgpu_device *adev, int vcn_instance, - int *major, int *minor, int *revision) -{ - return amdgpu_discovery_get_ip_version(adev, VCN_HWID, - vcn_instance, major, minor, revision); -} - void amdgpu_discovery_harvest_ip(struct amdgpu_device *adev) { struct binary_header *bhdr; @@ -917,7 +956,6 @@ static int amdgpu_discovery_set_mm_ip_blocks(struct amdgpu_device *adev) break; case IP_VERSION(3, 0, 0): case IP_VERSION(3, 0, 16): - case IP_VERSION(3, 0, 64): case IP_VERSION(3, 1, 1): case IP_VERSION(3, 0, 2): case IP_VERSION(3, 0, 192): diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.h index 0ea029e3b850..14537cec19db 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_discovery.h @@ -33,8 +33,6 @@ void amdgpu_discovery_harvest_ip(struct amdgpu_device *adev); int amdgpu_discovery_get_ip_version(struct amdgpu_device *adev, int hw_id, int number_instance, int *major, int *minor, int *revision); -int amdgpu_discovery_get_vcn_version(struct amdgpu_device *adev, int vcn_instance, - int *major, int *minor, int *revision); int amdgpu_discovery_get_gfx_info(struct amdgpu_device *adev); int amdgpu_discovery_set_ip_blocks(struct amdgpu_device *adev); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c index 18cc7155e667..82011e75ed85 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_display.c @@ -1360,7 +1360,7 @@ bool amdgpu_display_crtc_scaling_mode_fixup(struct drm_crtc *crtc, if ((!(mode->flags & DRM_MODE_FLAG_INTERLACE)) && ((amdgpu_encoder->underscan_type == UNDERSCAN_ON) || ((amdgpu_encoder->underscan_type == UNDERSCAN_AUTO) && - drm_detect_hdmi_monitor(amdgpu_connector_edid(connector)) && + connector->display_info.is_hdmi && amdgpu_display_is_hdtv_mode(mode)))) { if (amdgpu_encoder->underscan_hborder != 0) amdgpu_crtc->h_border = amdgpu_encoder->underscan_hborder; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c index 3a6f125c6dc9..02099058d0d6 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.c @@ -313,9 +313,12 @@ module_param_named(dpm, amdgpu_dpm, int, 0444); /** * DOC: fw_load_type (int) - * Set different firmware loading type for debugging (0 = direct, 1 = SMU, 2 = PSP). The default is -1 (auto). + * Set different firmware loading type for debugging, if supported. + * Set to 0 to force direct loading if supported by the ASIC. Set + * to -1 to select the default loading mode for the ASIC, as defined + * by the driver. The default is -1 (auto). */ -MODULE_PARM_DESC(fw_load_type, "firmware loading type (0 = direct, 1 = SMU, 2 = PSP, -1 = auto)"); +MODULE_PARM_DESC(fw_load_type, "firmware loading type (0 = force direct if supported, -1 = auto)"); module_param_named(fw_load_type, amdgpu_fw_load_type, int, 0444); /** diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.h index e3a4f7048042..8178323e4bef 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_drv.h @@ -45,4 +45,7 @@ long amdgpu_drm_ioctl(struct file *filp, unsigned int cmd, unsigned long arg); +long amdgpu_kms_compat_ioctl(struct file *filp, + unsigned int cmd, unsigned long arg); + #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_encoders.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_encoders.c index af4ef84e27a7..c96e458ed088 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_encoders.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_encoders.c @@ -222,7 +222,7 @@ bool amdgpu_dig_monitor_is_duallink(struct drm_encoder *encoder, case DRM_MODE_CONNECTOR_HDMIB: if (amdgpu_connector->use_digital) { /* HDMI 1.3 supports up to 340 Mhz over single link */ - if (drm_detect_hdmi_monitor(amdgpu_connector_edid(connector))) { + if (connector->display_info.is_hdmi) { if (pixel_clock > 340000) return true; else @@ -244,7 +244,7 @@ bool amdgpu_dig_monitor_is_duallink(struct drm_encoder *encoder, return false; else { /* HDMI 1.3 supports up to 340 Mhz over single link */ - if (drm_detect_hdmi_monitor(amdgpu_connector_edid(connector))) { + if (connector->display_info.is_hdmi) { if (pixel_clock > 340000) return true; else diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c index 08478fce00f2..2430d6223c2d 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.c @@ -350,6 +350,7 @@ static inline uint64_t amdgpu_gmc_fault_key(uint64_t addr, uint16_t pasid) * amdgpu_gmc_filter_faults - filter VM faults * * @adev: amdgpu device structure + * @ih: interrupt ring that the fault received from * @addr: address of the VM fault * @pasid: PASID of the process causing the fault * @timestamp: timestamp of the fault @@ -358,7 +359,8 @@ static inline uint64_t amdgpu_gmc_fault_key(uint64_t addr, uint16_t pasid) * True if the fault was filtered and should not be processed further. * False if the fault is a new one and needs to be handled. */ -bool amdgpu_gmc_filter_faults(struct amdgpu_device *adev, uint64_t addr, +bool amdgpu_gmc_filter_faults(struct amdgpu_device *adev, + struct amdgpu_ih_ring *ih, uint64_t addr, uint16_t pasid, uint64_t timestamp) { struct amdgpu_gmc *gmc = &adev->gmc; @@ -366,6 +368,10 @@ bool amdgpu_gmc_filter_faults(struct amdgpu_device *adev, uint64_t addr, struct amdgpu_gmc_fault *fault; uint32_t hash; + /* Stale retry fault if timestamp goes backward */ + if (amdgpu_ih_ts_after(timestamp, ih->processed_timestamp)) + return true; + /* If we don't have space left in the ring buffer return immediately */ stamp = max(timestamp, AMDGPU_GMC_FAULT_TIMEOUT + 1) - AMDGPU_GMC_FAULT_TIMEOUT; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h index e55201134a01..8458cebc6d5b 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gmc.h @@ -316,7 +316,8 @@ void amdgpu_gmc_gart_location(struct amdgpu_device *adev, struct amdgpu_gmc *mc); void amdgpu_gmc_agp_location(struct amdgpu_device *adev, struct amdgpu_gmc *mc); -bool amdgpu_gmc_filter_faults(struct amdgpu_device *adev, uint64_t addr, +bool amdgpu_gmc_filter_faults(struct amdgpu_device *adev, + struct amdgpu_ih_ring *ih, uint64_t addr, uint16_t pasid, uint64_t timestamp); void amdgpu_gmc_filter_faults_remove(struct amdgpu_device *adev, uint64_t addr, uint16_t pasid); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c index 0c7963dfacad..3df146579ad9 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.c @@ -164,52 +164,32 @@ void amdgpu_ih_ring_write(struct amdgpu_ih_ring *ih, const uint32_t *iv, } } -/* Waiter helper that checks current rptr matches or passes checkpoint wptr */ -static bool amdgpu_ih_has_checkpoint_processed(struct amdgpu_device *adev, - struct amdgpu_ih_ring *ih, - uint32_t checkpoint_wptr, - uint32_t *prev_rptr) -{ - uint32_t cur_rptr = ih->rptr | (*prev_rptr & ~ih->ptr_mask); - - /* rptr has wrapped. */ - if (cur_rptr < *prev_rptr) - cur_rptr += ih->ptr_mask + 1; - *prev_rptr = cur_rptr; - - /* check ring is empty to workaround missing wptr overflow flag */ - return cur_rptr >= checkpoint_wptr || - (cur_rptr & ih->ptr_mask) == amdgpu_ih_get_wptr(adev, ih); -} - /** - * amdgpu_ih_wait_on_checkpoint_process - wait to process IVs up to checkpoint + * amdgpu_ih_wait_on_checkpoint_process_ts - wait to process IVs up to checkpoint * * @adev: amdgpu_device pointer * @ih: ih ring to process * * Used to ensure ring has processed IVs up to the checkpoint write pointer. */ -int amdgpu_ih_wait_on_checkpoint_process(struct amdgpu_device *adev, +int amdgpu_ih_wait_on_checkpoint_process_ts(struct amdgpu_device *adev, struct amdgpu_ih_ring *ih) { - uint32_t checkpoint_wptr, rptr; + uint32_t checkpoint_wptr; + uint64_t checkpoint_ts; + long timeout = HZ; if (!ih->enabled || adev->shutdown) return -ENODEV; checkpoint_wptr = amdgpu_ih_get_wptr(adev, ih); - /* Order wptr with rptr. */ + /* Order wptr with ring data. */ rmb(); - rptr = READ_ONCE(ih->rptr); - - /* wptr has wrapped. */ - if (rptr > checkpoint_wptr) - checkpoint_wptr += ih->ptr_mask + 1; + checkpoint_ts = amdgpu_ih_decode_iv_ts(adev, ih, checkpoint_wptr, -1); - return wait_event_interruptible(ih->wait_process, - amdgpu_ih_has_checkpoint_processed(adev, ih, - checkpoint_wptr, &rptr)); + return wait_event_interruptible_timeout(ih->wait_process, + amdgpu_ih_ts_after(checkpoint_ts, ih->processed_timestamp) || + ih->rptr == amdgpu_ih_get_wptr(adev, ih), timeout); } /** @@ -299,3 +279,18 @@ void amdgpu_ih_decode_iv_helper(struct amdgpu_device *adev, /* wptr/rptr are in bytes! */ ih->rptr += 32; } + +uint64_t amdgpu_ih_decode_iv_ts_helper(struct amdgpu_ih_ring *ih, u32 rptr, + signed int offset) +{ + uint32_t iv_size = 32; + uint32_t ring_index; + uint32_t dw1, dw2; + + rptr += iv_size * offset; + ring_index = (rptr & ih->ptr_mask) >> 2; + + dw1 = le32_to_cpu(ih->ring[ring_index + 1]); + dw2 = le32_to_cpu(ih->ring[ring_index + 2]); + return dw1 | ((u64)(dw2 & 0xffff) << 32); +} diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.h index 0649b59830a5..dd1c2eded6b9 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ih.h @@ -68,20 +68,30 @@ struct amdgpu_ih_ring { /* For waiting on IH processing at checkpoint. */ wait_queue_head_t wait_process; + uint64_t processed_timestamp; }; +/* return true if time stamp t2 is after t1 with 48bit wrap around */ +#define amdgpu_ih_ts_after(t1, t2) \ + (((int64_t)((t2) << 16) - (int64_t)((t1) << 16)) > 0LL) + /* provided by the ih block */ struct amdgpu_ih_funcs { /* ring read/write ptr handling, called from interrupt context */ u32 (*get_wptr)(struct amdgpu_device *adev, struct amdgpu_ih_ring *ih); void (*decode_iv)(struct amdgpu_device *adev, struct amdgpu_ih_ring *ih, struct amdgpu_iv_entry *entry); + uint64_t (*decode_iv_ts)(struct amdgpu_ih_ring *ih, u32 rptr, + signed int offset); void (*set_rptr)(struct amdgpu_device *adev, struct amdgpu_ih_ring *ih); }; #define amdgpu_ih_get_wptr(adev, ih) (adev)->irq.ih_funcs->get_wptr((adev), (ih)) #define amdgpu_ih_decode_iv(adev, iv) \ (adev)->irq.ih_funcs->decode_iv((adev), (ih), (iv)) +#define amdgpu_ih_decode_iv_ts(adev, ih, rptr, offset) \ + (WARN_ON_ONCE(!(adev)->irq.ih_funcs->decode_iv_ts) ? 0 : \ + (adev)->irq.ih_funcs->decode_iv_ts((ih), (rptr), (offset))) #define amdgpu_ih_set_rptr(adev, ih) (adev)->irq.ih_funcs->set_rptr((adev), (ih)) int amdgpu_ih_ring_init(struct amdgpu_device *adev, struct amdgpu_ih_ring *ih, @@ -89,10 +99,12 @@ int amdgpu_ih_ring_init(struct amdgpu_device *adev, struct amdgpu_ih_ring *ih, void amdgpu_ih_ring_fini(struct amdgpu_device *adev, struct amdgpu_ih_ring *ih); void amdgpu_ih_ring_write(struct amdgpu_ih_ring *ih, const uint32_t *iv, unsigned int num_dw); -int amdgpu_ih_wait_on_checkpoint_process(struct amdgpu_device *adev, - struct amdgpu_ih_ring *ih); +int amdgpu_ih_wait_on_checkpoint_process_ts(struct amdgpu_device *adev, + struct amdgpu_ih_ring *ih); int amdgpu_ih_process(struct amdgpu_device *adev, struct amdgpu_ih_ring *ih); void amdgpu_ih_decode_iv_helper(struct amdgpu_device *adev, struct amdgpu_ih_ring *ih, struct amdgpu_iv_entry *entry); +uint64_t amdgpu_ih_decode_iv_ts_helper(struct amdgpu_ih_ring *ih, u32 rptr, + signed int offset); #endif diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ioc32.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ioc32.c index 5cf142e849bb..a1cbd7c3deb2 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ioc32.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ioc32.c @@ -1,4 +1,4 @@ -/** +/* * \file amdgpu_ioc32.c * * 32-bit ioctl compatibility routines for the AMDGPU DRM. @@ -37,12 +37,9 @@ long amdgpu_kms_compat_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) { unsigned int nr = DRM_IOCTL_NR(cmd); - int ret; if (nr < DRM_COMMAND_BASE) return drm_compat_ioctl(filp, cmd, arg); - ret = amdgpu_drm_ioctl(filp, cmd, arg); - - return ret; + return amdgpu_drm_ioctl(filp, cmd, arg); } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c index 4f3c62adccbd..f5cbc2747ac6 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_irq.c @@ -390,7 +390,7 @@ void amdgpu_irq_fini_hw(struct amdgpu_device *adev) } /** - * amdgpu_irq_fini - shut down interrupt handling + * amdgpu_irq_fini_sw - shut down interrupt handling * * @adev: amdgpu device pointer * @@ -528,6 +528,9 @@ void amdgpu_irq_dispatch(struct amdgpu_device *adev, /* Send it to amdkfd as well if it isn't already handled */ if (!handled) amdgpu_amdkfd_interrupt(adev, entry.iv_entry); + + if (amdgpu_ih_ts_after(ih->processed_timestamp, entry.timestamp)) + ih->processed_timestamp = entry.timestamp; } /** diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_pll.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_pll.c index 4eaec446b49d..0bb2466d539a 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_pll.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_pll.c @@ -69,6 +69,7 @@ static void amdgpu_pll_reduce_ratio(unsigned *nom, unsigned *den, /** * amdgpu_pll_get_fb_ref_div - feedback and ref divider calculation * + * @adev: amdgpu_device pointer * @nom: nominator * @den: denominator * @post_div: post divider @@ -106,6 +107,7 @@ static void amdgpu_pll_get_fb_ref_div(struct amdgpu_device *adev, unsigned int n /** * amdgpu_pll_compute - compute PLL paramaters * + * @adev: amdgpu_device pointer * @pll: information about the PLL * @freq: requested frequency * @dot_clock_p: resulting pixel clock diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_pmu.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_pmu.c index 82e9ecf84352..71ee361d0972 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_pmu.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_pmu.c @@ -233,6 +233,10 @@ static void amdgpu_perf_start(struct perf_event *event, int flags) if (WARN_ON_ONCE(!(hwc->state & PERF_HES_STOPPED))) return; + if ((!pe->adev->df.funcs) || + (!pe->adev->df.funcs->pmc_start)) + return; + WARN_ON_ONCE(!(hwc->state & PERF_HES_UPTODATE)); hwc->state = 0; @@ -268,6 +272,10 @@ static void amdgpu_perf_read(struct perf_event *event) pmu); u64 count, prev; + if ((!pe->adev->df.funcs) || + (!pe->adev->df.funcs->pmc_get_count)) + return; + do { prev = local64_read(&hwc->prev_count); @@ -297,6 +305,10 @@ static void amdgpu_perf_stop(struct perf_event *event, int flags) if (hwc->state & PERF_HES_UPTODATE) return; + if ((!pe->adev->df.funcs) || + (!pe->adev->df.funcs->pmc_stop)) + return; + switch (hwc->config_base) { case AMDGPU_PMU_EVENT_CONFIG_TYPE_DF: case AMDGPU_PMU_EVENT_CONFIG_TYPE_XGMI: @@ -326,6 +338,10 @@ static int amdgpu_perf_add(struct perf_event *event, int flags) struct amdgpu_pmu_entry, pmu); + if ((!pe->adev->df.funcs) || + (!pe->adev->df.funcs->pmc_start)) + return -EINVAL; + switch (pe->pmu_perf_type) { case AMDGPU_PMU_PERF_TYPE_DF: hwc->config_base = AMDGPU_PMU_EVENT_CONFIG_TYPE_DF; @@ -371,6 +387,9 @@ static void amdgpu_perf_del(struct perf_event *event, int flags) struct amdgpu_pmu_entry *pe = container_of(event->pmu, struct amdgpu_pmu_entry, pmu); + if ((!pe->adev->df.funcs) || + (!pe->adev->df.funcs->pmc_stop)) + return; amdgpu_perf_stop(event, PERF_EF_UPDATE); diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_preempt_mgr.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_preempt_mgr.c index d02c8637f909..786afe4f58f9 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_preempt_mgr.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_preempt_mgr.c @@ -59,7 +59,7 @@ static DEVICE_ATTR_RO(mem_info_preempt_used); * @man: TTM memory type manager * @tbo: TTM BO we need this range for * @place: placement flags and restrictions - * @mem: the resulting mem object + * @res: TTM memory object * * Dummy, just count the space used without allocating resources or any limit. */ @@ -85,7 +85,7 @@ static int amdgpu_preempt_mgr_new(struct ttm_resource_manager *man, * amdgpu_preempt_mgr_del - free ranges * * @man: TTM memory type manager - * @mem: TTM memory object + * @res: TTM memory object * * Free the allocated GTT again. */ diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c index c641f84649d6..dee17a0e1187 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_psp.c @@ -518,7 +518,7 @@ static struct psp_gfx_cmd_resp *acquire_psp_cmd_buf(struct psp_context *psp) return cmd; } -void release_psp_cmd_buf(struct psp_context *psp) +static void release_psp_cmd_buf(struct psp_context *psp) { mutex_unlock(&psp->mutex); } @@ -2017,12 +2017,16 @@ static int psp_hw_start(struct psp_context *psp) return ret; } + if (amdgpu_sriov_vf(adev) && amdgpu_in_reset(adev)) + goto skip_pin_bo; + ret = psp_tmr_init(psp); if (ret) { DRM_ERROR("PSP tmr init failed!\n"); return ret; } +skip_pin_bo: /* * For ASICs with DF Cstate management centralized * to PMFW, TMR setup should be performed after PMFW @@ -2452,6 +2456,18 @@ skip_memalloc: return ret; } + if (amdgpu_sriov_vf(adev) && amdgpu_in_reset(adev)) { + if (adev->gmc.xgmi.num_physical_nodes > 1) { + ret = psp_xgmi_initialize(psp, false, true); + /* Warning the XGMI seesion initialize failure + * Instead of stop driver initialization + */ + if (ret) + dev_err(psp->adev->dev, + "XGMI: Failed to initialize XGMI session\n"); + } + } + if (psp->ta_fw) { ret = psp_ras_initialize(psp); if (ret) diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c index 46910e7b2927..cd9e5914944b 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c @@ -867,9 +867,9 @@ static int amdgpu_ras_enable_all_features(struct amdgpu_device *adev, /* feature ctl end */ -void amdgpu_ras_mca_query_error_status(struct amdgpu_device *adev, - struct ras_common_if *ras_block, - struct ras_err_data *err_data) +static void amdgpu_ras_mca_query_error_status(struct amdgpu_device *adev, + struct ras_common_if *ras_block, + struct ras_err_data *err_data) { switch (ras_block->sub_block_index) { case AMDGPU_RAS_MCA_BLOCK__MP0: @@ -1161,9 +1161,9 @@ int amdgpu_ras_error_inject(struct amdgpu_device *adev, /** * amdgpu_ras_query_error_count -- Get error counts of all IPs - * adev: pointer to AMD GPU device - * ce_count: pointer to an integer to be set to the count of correctible errors. - * ue_count: pointer to an integer to be set to the count of uncorrectible + * @adev: pointer to AMD GPU device + * @ce_count: pointer to an integer to be set to the count of correctible errors. + * @ue_count: pointer to an integer to be set to the count of uncorrectible * errors. * * If set, @ce_count or @ue_count, count and return the corresponding @@ -1747,6 +1747,16 @@ static void amdgpu_ras_log_on_err_counter(struct amdgpu_device *adev) if (info.head.block == AMDGPU_RAS_BLOCK__PCIE_BIF) continue; + /* + * this is a workaround for aldebaran, skip send msg to + * smu to get ecc_info table due to smu handle get ecc + * info table failed temporarily. + * should be removed until smu fix handle ecc_info table. + */ + if ((info.head.block == AMDGPU_RAS_BLOCK__UMC) && + (adev->ip_versions[MP1_HWIP][0] == IP_VERSION(13, 0, 2))) + continue; + amdgpu_ras_query_error_status(adev, &info); } } @@ -2362,7 +2372,11 @@ int amdgpu_ras_init(struct amdgpu_device *adev) } /* Init poison supported flag, the default value is false */ - if (adev->df.funcs && + if (adev->gmc.xgmi.connected_to_cpu) { + /* enabled by default when GPU is connected to CPU */ + con->poison_supported = true; + } + else if (adev->df.funcs && adev->df.funcs->query_ras_poison_mode && adev->umc.ras_funcs && adev->umc.ras_funcs->query_ras_poison_mode) { @@ -2503,7 +2517,6 @@ void amdgpu_ras_late_fini(struct amdgpu_device *adev, amdgpu_ras_sysfs_remove(adev, ras_block); if (ih_info->cb) amdgpu_ras_interrupt_remove_handler(adev, ih_info); - amdgpu_ras_feature_enable(adev, ras_block, 0); } /* do some init work after IP late init as dependence. diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c index 688bef1649b5..344f711ad144 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vce.c @@ -434,7 +434,6 @@ void amdgpu_vce_free_handles(struct amdgpu_device *adev, struct drm_file *filp) * * @ring: ring we should submit the msg to * @handle: VCE session handle to use - * @bo: amdgpu object for which we query the offset * @fence: optional fence to return * * Open up a stream for HW test diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c index 585961c2f5f2..9a19a6a57b23 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c @@ -286,20 +286,13 @@ int amdgpu_vcn_sw_fini(struct amdgpu_device *adev) bool amdgpu_vcn_is_disabled_vcn(struct amdgpu_device *adev, enum vcn_ring_type type, uint32_t vcn_instance) { bool ret = false; + int vcn_config = adev->vcn.vcn_config[vcn_instance]; - int major; - int minor; - int revision; - - /* if cannot find IP data, then this VCN does not exist */ - if (amdgpu_discovery_get_vcn_version(adev, vcn_instance, &major, &minor, &revision) != 0) - return true; - - if ((type == VCN_ENCODE_RING) && (revision & VCN_BLOCK_ENCODE_DISABLE_MASK)) { + if ((type == VCN_ENCODE_RING) && (vcn_config & VCN_BLOCK_ENCODE_DISABLE_MASK)) { ret = true; - } else if ((type == VCN_DECODE_RING) && (revision & VCN_BLOCK_DECODE_DISABLE_MASK)) { + } else if ((type == VCN_DECODE_RING) && (vcn_config & VCN_BLOCK_DECODE_DISABLE_MASK)) { ret = true; - } else if ((type == VCN_UNIFIED_RING) && (revision & VCN_BLOCK_QUEUE_DISABLE_MASK)) { + } else if ((type == VCN_UNIFIED_RING) && (vcn_config & VCN_BLOCK_QUEUE_DISABLE_MASK)) { ret = true; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h index bfa27ea94804..5d3728b027d3 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h @@ -235,6 +235,7 @@ struct amdgpu_vcn { uint8_t num_vcn_inst; struct amdgpu_vcn_inst inst[AMDGPU_MAX_VCN_INSTANCES]; + uint8_t vcn_config[AMDGPU_MAX_VCN_INSTANCES]; struct amdgpu_vcn_reg internal; struct mutex vcn_pg_lock; struct mutex vcn1_jpeg1_workaround; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vkms.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vkms.c index ac9a8cd21c4b..2dcc68e04e84 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vkms.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vkms.c @@ -16,6 +16,8 @@ #include "ivsrcid/ivsrcid_vislands30.h" #include "amdgpu_vkms.h" #include "amdgpu_display.h" +#include "atom.h" +#include "amdgpu_irq.h" /** * DOC: amdgpu_vkms @@ -41,16 +43,16 @@ static const u32 amdgpu_vkms_formats[] = { static enum hrtimer_restart amdgpu_vkms_vblank_simulate(struct hrtimer *timer) { - struct amdgpu_vkms_output *output = container_of(timer, - struct amdgpu_vkms_output, - vblank_hrtimer); - struct drm_crtc *crtc = &output->crtc; + struct amdgpu_crtc *amdgpu_crtc = container_of(timer, struct amdgpu_crtc, vblank_timer); + struct drm_crtc *crtc = &amdgpu_crtc->base; + struct amdgpu_vkms_output *output = drm_crtc_to_amdgpu_vkms_output(crtc); u64 ret_overrun; bool ret; - ret_overrun = hrtimer_forward_now(&output->vblank_hrtimer, + ret_overrun = hrtimer_forward_now(&amdgpu_crtc->vblank_timer, output->period_ns); - WARN_ON(ret_overrun != 1); + if (ret_overrun != 1) + DRM_WARN("%s: vblank timer overrun\n", __func__); ret = drm_crtc_handle_vblank(crtc); if (!ret) @@ -65,22 +67,21 @@ static int amdgpu_vkms_enable_vblank(struct drm_crtc *crtc) unsigned int pipe = drm_crtc_index(crtc); struct drm_vblank_crtc *vblank = &dev->vblank[pipe]; struct amdgpu_vkms_output *out = drm_crtc_to_amdgpu_vkms_output(crtc); + struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc); drm_calc_timestamping_constants(crtc, &crtc->mode); - hrtimer_init(&out->vblank_hrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); - out->vblank_hrtimer.function = &amdgpu_vkms_vblank_simulate; out->period_ns = ktime_set(0, vblank->framedur_ns); - hrtimer_start(&out->vblank_hrtimer, out->period_ns, HRTIMER_MODE_REL); + hrtimer_start(&amdgpu_crtc->vblank_timer, out->period_ns, HRTIMER_MODE_REL); return 0; } static void amdgpu_vkms_disable_vblank(struct drm_crtc *crtc) { - struct amdgpu_vkms_output *out = drm_crtc_to_amdgpu_vkms_output(crtc); + struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc); - hrtimer_cancel(&out->vblank_hrtimer); + hrtimer_cancel(&amdgpu_crtc->vblank_timer); } static bool amdgpu_vkms_get_vblank_timestamp(struct drm_crtc *crtc, @@ -92,13 +93,14 @@ static bool amdgpu_vkms_get_vblank_timestamp(struct drm_crtc *crtc, unsigned int pipe = crtc->index; struct amdgpu_vkms_output *output = drm_crtc_to_amdgpu_vkms_output(crtc); struct drm_vblank_crtc *vblank = &dev->vblank[pipe]; + struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc); if (!READ_ONCE(vblank->enabled)) { *vblank_time = ktime_get(); return true; } - *vblank_time = READ_ONCE(output->vblank_hrtimer.node.expires); + *vblank_time = READ_ONCE(amdgpu_crtc->vblank_timer.node.expires); if (WARN_ON(*vblank_time == vblank->time)) return true; @@ -165,6 +167,8 @@ static const struct drm_crtc_helper_funcs amdgpu_vkms_crtc_helper_funcs = { static int amdgpu_vkms_crtc_init(struct drm_device *dev, struct drm_crtc *crtc, struct drm_plane *primary, struct drm_plane *cursor) { + struct amdgpu_device *adev = drm_to_adev(dev); + struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc); int ret; ret = drm_crtc_init_with_planes(dev, crtc, primary, cursor, @@ -176,6 +180,17 @@ static int amdgpu_vkms_crtc_init(struct drm_device *dev, struct drm_crtc *crtc, drm_crtc_helper_add(crtc, &amdgpu_vkms_crtc_helper_funcs); + amdgpu_crtc->crtc_id = drm_crtc_index(crtc); + adev->mode_info.crtcs[drm_crtc_index(crtc)] = amdgpu_crtc; + + amdgpu_crtc->pll_id = ATOM_PPLL_INVALID; + amdgpu_crtc->encoder = NULL; + amdgpu_crtc->connector = NULL; + amdgpu_crtc->vsync_timer_enabled = AMDGPU_IRQ_STATE_DISABLE; + + hrtimer_init(&amdgpu_crtc->vblank_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); + amdgpu_crtc->vblank_timer.function = &amdgpu_vkms_vblank_simulate; + return ret; } @@ -375,6 +390,7 @@ static struct drm_plane *amdgpu_vkms_plane_init(struct drm_device *dev, int index) { struct drm_plane *plane; + uint64_t modifiers[] = {DRM_FORMAT_MOD_LINEAR, DRM_FORMAT_MOD_INVALID}; int ret; plane = kzalloc(sizeof(*plane), GFP_KERNEL); @@ -385,7 +401,7 @@ static struct drm_plane *amdgpu_vkms_plane_init(struct drm_device *dev, &amdgpu_vkms_plane_funcs, amdgpu_vkms_formats, ARRAY_SIZE(amdgpu_vkms_formats), - NULL, type, NULL); + modifiers, type, NULL); if (ret) { kfree(plane); return ERR_PTR(ret); @@ -396,12 +412,12 @@ static struct drm_plane *amdgpu_vkms_plane_init(struct drm_device *dev, return plane; } -int amdgpu_vkms_output_init(struct drm_device *dev, - struct amdgpu_vkms_output *output, int index) +static int amdgpu_vkms_output_init(struct drm_device *dev, struct + amdgpu_vkms_output *output, int index) { struct drm_connector *connector = &output->connector; struct drm_encoder *encoder = &output->encoder; - struct drm_crtc *crtc = &output->crtc; + struct drm_crtc *crtc = &output->crtc.base; struct drm_plane *primary, *cursor = NULL; int ret; @@ -465,6 +481,11 @@ static int amdgpu_vkms_sw_init(void *handle) int r, i; struct amdgpu_device *adev = (struct amdgpu_device *)handle; + adev->amdgpu_vkms_output = kcalloc(adev->mode_info.num_crtc, + sizeof(struct amdgpu_vkms_output), GFP_KERNEL); + if (!adev->amdgpu_vkms_output) + return -ENOMEM; + adev_to_drm(adev)->max_vblank_count = 0; adev_to_drm(adev)->mode_config.funcs = &amdgpu_vkms_mode_funcs; @@ -481,10 +502,6 @@ static int amdgpu_vkms_sw_init(void *handle) if (r) return r; - adev->amdgpu_vkms_output = kcalloc(adev->mode_info.num_crtc, sizeof(struct amdgpu_vkms_output), GFP_KERNEL); - if (!adev->amdgpu_vkms_output) - return -ENOMEM; - /* allocate crtcs, encoders, connectors */ for (i = 0; i < adev->mode_info.num_crtc; i++) { r = amdgpu_vkms_output_init(adev_to_drm(adev), &adev->amdgpu_vkms_output[i], i); @@ -504,15 +521,16 @@ static int amdgpu_vkms_sw_fini(void *handle) int i = 0; for (i = 0; i < adev->mode_info.num_crtc; i++) - if (adev->amdgpu_vkms_output[i].vblank_hrtimer.function) - hrtimer_cancel(&adev->amdgpu_vkms_output[i].vblank_hrtimer); - - kfree(adev->mode_info.bios_hardcoded_edid); - kfree(adev->amdgpu_vkms_output); + if (adev->mode_info.crtcs[i]) + hrtimer_cancel(&adev->mode_info.crtcs[i]->vblank_timer); drm_kms_helper_poll_fini(adev_to_drm(adev)); + drm_mode_config_cleanup(adev_to_drm(adev)); adev->mode_info.mode_config_initialized = false; + + kfree(adev->mode_info.bios_hardcoded_edid); + kfree(adev->amdgpu_vkms_output); return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vkms.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vkms.h index 97f1b79c0724..4f8722ff37c2 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vkms.h +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vkms.h @@ -10,15 +10,14 @@ #define YRES_MAX 16384 #define drm_crtc_to_amdgpu_vkms_output(target) \ - container_of(target, struct amdgpu_vkms_output, crtc) + container_of(target, struct amdgpu_vkms_output, crtc.base) extern const struct amdgpu_ip_block_version amdgpu_vkms_ip_block; struct amdgpu_vkms_output { - struct drm_crtc crtc; + struct amdgpu_crtc crtc; struct drm_encoder encoder; struct drm_connector connector; - struct hrtimer vblank_hrtimer; ktime_t period_ns; struct drm_pending_vblank_event *event; }; diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c index a96ae4c0e040..b37fc7d7d2c7 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c @@ -53,7 +53,7 @@ * can be mapped as snooped (cached system pages) or unsnooped * (uncached system pages). * Each VM has an ID associated with it and there is a page table - * associated with each VMID. When execting a command buffer, + * associated with each VMID. When executing a command buffer, * the kernel tells the the ring what VMID to use for that command * buffer. VMIDs are allocated dynamically as commands are submitted. * The userspace drivers maintain their own address space and the kernel diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c index 567df2db23ac..a38c6a747fa4 100644 --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_xgmi.c @@ -265,6 +265,11 @@ static ssize_t amdgpu_xgmi_show_error(struct device *dev, ficaa_pie_ctl_in = AMDGPU_XGMI_SET_FICAA(0x200); ficaa_pie_status_in = AMDGPU_XGMI_SET_FICAA(0x208); + if ((!adev->df.funcs) || + (!adev->df.funcs->get_fica) || + (!adev->df.funcs->set_fica)) + return -EINVAL; + fica_out = adev->df.funcs->get_fica(adev, ficaa_pie_ctl_in); if (fica_out != 0x1f) pr_err("xGMI error counters not enabled!\n"); diff --git a/drivers/gpu/drm/amd/amdgpu/atombios_encoders.c b/drivers/gpu/drm/amd/amdgpu/atombios_encoders.c index 6134ed964027..a92d86e12718 100644 --- a/drivers/gpu/drm/amd/amdgpu/atombios_encoders.c +++ b/drivers/gpu/drm/amd/amdgpu/atombios_encoders.c @@ -469,7 +469,7 @@ int amdgpu_atombios_encoder_get_encoder_mode(struct drm_encoder *encoder) if (amdgpu_connector->use_digital && (amdgpu_connector->audio == AMDGPU_AUDIO_ENABLE)) return ATOM_ENCODER_MODE_HDMI; - else if (drm_detect_hdmi_monitor(amdgpu_connector_edid(connector)) && + else if (connector->display_info.is_hdmi && (amdgpu_connector->audio == AMDGPU_AUDIO_AUTO)) return ATOM_ENCODER_MODE_HDMI; else if (amdgpu_connector->use_digital) @@ -488,7 +488,7 @@ int amdgpu_atombios_encoder_get_encoder_mode(struct drm_encoder *encoder) if (amdgpu_audio != 0) { if (amdgpu_connector->audio == AMDGPU_AUDIO_ENABLE) return ATOM_ENCODER_MODE_HDMI; - else if (drm_detect_hdmi_monitor(amdgpu_connector_edid(connector)) && + else if (connector->display_info.is_hdmi && (amdgpu_connector->audio == AMDGPU_AUDIO_AUTO)) return ATOM_ENCODER_MODE_HDMI; else @@ -506,7 +506,7 @@ int amdgpu_atombios_encoder_get_encoder_mode(struct drm_encoder *encoder) } else if (amdgpu_audio != 0) { if (amdgpu_connector->audio == AMDGPU_AUDIO_ENABLE) return ATOM_ENCODER_MODE_HDMI; - else if (drm_detect_hdmi_monitor(amdgpu_connector_edid(connector)) && + else if (connector->display_info.is_hdmi && (amdgpu_connector->audio == AMDGPU_AUDIO_AUTO)) return ATOM_ENCODER_MODE_HDMI; else diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c index b305fd39874f..edb3e3b08eed 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c @@ -3070,8 +3070,8 @@ static void gfx_v9_0_init_pg(struct amdgpu_device *adev) AMD_PG_SUPPORT_CP | AMD_PG_SUPPORT_GDS | AMD_PG_SUPPORT_RLC_SMU_HS)) { - WREG32(mmRLC_JUMP_TABLE_RESTORE, - adev->gfx.rlc.cp_table_gpu_addr >> 8); + WREG32_SOC15(GC, 0, mmRLC_JUMP_TABLE_RESTORE, + adev->gfx.rlc.cp_table_gpu_addr >> 8); gfx_v9_0_init_gfx_power_gating(adev); } } diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c index 480e41847d7c..ec4d5e15b766 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v1_0.c @@ -162,7 +162,6 @@ static void gfxhub_v1_0_init_tlb_regs(struct amdgpu_device *adev) ENABLE_ADVANCED_DRIVER_MODEL, 1); tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, SYSTEM_APERTURE_UNMAPPED_ACCESS, 0); - tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, ECO_BITS, 0); tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, MTYPE, MTYPE_UC);/* XXX for emulation. */ tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, ATC_EN, 1); diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c index 14c1c1a297dd..6e0ace2fbfab 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_0.c @@ -196,7 +196,6 @@ static void gfxhub_v2_0_init_tlb_regs(struct amdgpu_device *adev) ENABLE_ADVANCED_DRIVER_MODEL, 1); tmp = REG_SET_FIELD(tmp, GCMC_VM_MX_L1_TLB_CNTL, SYSTEM_APERTURE_UNMAPPED_ACCESS, 0); - tmp = REG_SET_FIELD(tmp, GCMC_VM_MX_L1_TLB_CNTL, ECO_BITS, 0); tmp = REG_SET_FIELD(tmp, GCMC_VM_MX_L1_TLB_CNTL, MTYPE, MTYPE_UC); /* UC, uncached */ diff --git a/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_1.c b/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_1.c index e80d1dc43079..b4eddf6e98a6 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_1.c +++ b/drivers/gpu/drm/amd/amdgpu/gfxhub_v2_1.c @@ -197,7 +197,6 @@ static void gfxhub_v2_1_init_tlb_regs(struct amdgpu_device *adev) ENABLE_ADVANCED_DRIVER_MODEL, 1); tmp = REG_SET_FIELD(tmp, GCMC_VM_MX_L1_TLB_CNTL, SYSTEM_APERTURE_UNMAPPED_ACCESS, 0); - tmp = REG_SET_FIELD(tmp, GCMC_VM_MX_L1_TLB_CNTL, ECO_BITS, 0); tmp = REG_SET_FIELD(tmp, GCMC_VM_MX_L1_TLB_CNTL, MTYPE, MTYPE_UC); /* UC, uncached */ diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c index 3ec5ff5a6dbe..ae46eb35b3d7 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c @@ -107,7 +107,7 @@ static int gmc_v10_0_process_interrupt(struct amdgpu_device *adev, /* Process it onyl if it's the first fault for this address */ if (entry->ih != &adev->irq.ih_soft && - amdgpu_gmc_filter_faults(adev, addr, entry->pasid, + amdgpu_gmc_filter_faults(adev, entry->ih, addr, entry->pasid, entry->timestamp)) return 1; @@ -992,10 +992,14 @@ static int gmc_v10_0_gart_enable(struct amdgpu_device *adev) return -EINVAL; } + if (amdgpu_sriov_vf(adev) && amdgpu_in_reset(adev)) + goto skip_pin_bo; + r = amdgpu_gart_table_vram_pin(adev); if (r) return r; +skip_pin_bo: r = adev->gfxhub.funcs->gart_enable(adev); if (r) return r; diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c index cb82404df534..a5471923b3f6 100644 --- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c +++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c @@ -523,7 +523,7 @@ static int gmc_v9_0_process_interrupt(struct amdgpu_device *adev, /* Process it onyl if it's the first fault for this address */ if (entry->ih != &adev->irq.ih_soft && - amdgpu_gmc_filter_faults(adev, addr, entry->pasid, + amdgpu_gmc_filter_faults(adev, entry->ih, addr, entry->pasid, entry->timestamp)) return 1; @@ -1294,7 +1294,8 @@ static int gmc_v9_0_late_init(void *handle) if (!amdgpu_sriov_vf(adev) && (adev->ip_versions[UMC_HWIP][0] == IP_VERSION(6, 0, 0))) { if (!(adev->ras_enabled & (1 << AMDGPU_RAS_BLOCK__UMC))) { - if (adev->df.funcs->enable_ecc_force_par_wr_rmw) + if (adev->df.funcs && + adev->df.funcs->enable_ecc_force_par_wr_rmw) adev->df.funcs->enable_ecc_force_par_wr_rmw(adev, false); } } @@ -1505,9 +1506,11 @@ static int gmc_v9_0_sw_init(void *handle) chansize = 64; else chansize = 128; - - numchan = adev->df.funcs->get_hbm_channel_number(adev); - adev->gmc.vram_width = numchan * chansize; + if (adev->df.funcs && + adev->df.funcs->get_hbm_channel_number) { + numchan = adev->df.funcs->get_hbm_channel_number(adev); + adev->gmc.vram_width = numchan * chansize; + } } adev->gmc.vram_type = vram_type; @@ -1714,10 +1717,14 @@ static int gmc_v9_0_gart_enable(struct amdgpu_device *adev) return -EINVAL; } + if (amdgpu_sriov_vf(adev) && amdgpu_in_reset(adev)) + goto skip_pin_bo; + r = amdgpu_gart_table_vram_pin(adev); if (r) return r; +skip_pin_bo: r = adev->gfxhub.funcs->gart_enable(adev); if (r) return r; @@ -1742,7 +1749,7 @@ static int gmc_v9_0_hw_init(void *handle) { struct amdgpu_device *adev = (struct amdgpu_device *)handle; bool value; - int r, i; + int i; /* The sequence of these two function calls matters.*/ gmc_v9_0_init_golden_registers(adev); @@ -1777,9 +1784,7 @@ static int gmc_v9_0_hw_init(void *handle) if (adev->umc.funcs && adev->umc.funcs->init_registers) adev->umc.funcs->init_registers(adev); - r = gmc_v9_0_gart_enable(adev); - - return r; + return gmc_v9_0_gart_enable(adev); } /** @@ -1808,6 +1813,14 @@ static int gmc_v9_0_hw_fini(void *handle) return 0; } + /* + * Pair the operations did in gmc_v9_0_hw_init and thus maintain + * a correct cached state for GMC. Otherwise, the "gate" again + * operation on S3 resuming will fail due to wrong cached state. + */ + if (adev->mmhub.funcs->update_power_gating) + adev->mmhub.funcs->update_power_gating(adev, false); + amdgpu_irq_put(adev, &adev->gmc.ecc_irq, 0); amdgpu_irq_put(adev, &adev->gmc.vm_fault, 0); diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c index a99953833820..1da2ec692057 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c @@ -145,7 +145,6 @@ static void mmhub_v1_0_init_tlb_regs(struct amdgpu_device *adev) ENABLE_ADVANCED_DRIVER_MODEL, 1); tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, SYSTEM_APERTURE_UNMAPPED_ACCESS, 0); - tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, ECO_BITS, 0); tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, MTYPE, MTYPE_UC);/* XXX for emulation. */ tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, ATC_EN, 1); @@ -302,10 +301,10 @@ static void mmhub_v1_0_update_power_gating(struct amdgpu_device *adev, if (amdgpu_sriov_vf(adev)) return; - if (enable && adev->pg_flags & AMD_PG_SUPPORT_MMHUB) { - amdgpu_dpm_set_powergating_by_smu(adev, AMD_IP_BLOCK_TYPE_GMC, true); - - } + if (adev->pg_flags & AMD_PG_SUPPORT_MMHUB) + amdgpu_dpm_set_powergating_by_smu(adev, + AMD_IP_BLOCK_TYPE_GMC, + enable); } static int mmhub_v1_0_gart_enable(struct amdgpu_device *adev) diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_7.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_7.c index f80a14a1b82d..f5f7181f9af5 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_7.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_7.c @@ -165,7 +165,6 @@ static void mmhub_v1_7_init_tlb_regs(struct amdgpu_device *adev) ENABLE_ADVANCED_DRIVER_MODEL, 1); tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, SYSTEM_APERTURE_UNMAPPED_ACCESS, 0); - tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, ECO_BITS, 0); tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, MTYPE, MTYPE_UC);/* XXX for emulation. */ tmp = REG_SET_FIELD(tmp, MC_VM_MX_L1_TLB_CNTL, ATC_EN, 1); diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c index 25f8e93e5ec3..3718ff610ab2 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v2_0.c @@ -267,7 +267,6 @@ static void mmhub_v2_0_init_tlb_regs(struct amdgpu_device *adev) ENABLE_ADVANCED_DRIVER_MODEL, 1); tmp = REG_SET_FIELD(tmp, MMMC_VM_MX_L1_TLB_CNTL, SYSTEM_APERTURE_UNMAPPED_ACCESS, 0); - tmp = REG_SET_FIELD(tmp, MMMC_VM_MX_L1_TLB_CNTL, ECO_BITS, 0); tmp = REG_SET_FIELD(tmp, MMMC_VM_MX_L1_TLB_CNTL, MTYPE, MTYPE_UC); /* UC, uncached */ diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v2_3.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v2_3.c index a11d60ec6321..9e16da28505a 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v2_3.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v2_3.c @@ -194,7 +194,6 @@ static void mmhub_v2_3_init_tlb_regs(struct amdgpu_device *adev) ENABLE_ADVANCED_DRIVER_MODEL, 1); tmp = REG_SET_FIELD(tmp, MMMC_VM_MX_L1_TLB_CNTL, SYSTEM_APERTURE_UNMAPPED_ACCESS, 0); - tmp = REG_SET_FIELD(tmp, MMMC_VM_MX_L1_TLB_CNTL, ECO_BITS, 0); tmp = REG_SET_FIELD(tmp, MMMC_VM_MX_L1_TLB_CNTL, MTYPE, MTYPE_UC); /* UC, uncached */ diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.c index c4ef822bbe8c..ff49eeaf7882 100644 --- a/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.c +++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v9_4.c @@ -190,8 +190,6 @@ static void mmhub_v9_4_init_tlb_regs(struct amdgpu_device *adev, int hubid) tmp = REG_SET_FIELD(tmp, VMSHAREDVC0_MC_VM_MX_L1_TLB_CNTL, SYSTEM_APERTURE_UNMAPPED_ACCESS, 0); tmp = REG_SET_FIELD(tmp, VMSHAREDVC0_MC_VM_MX_L1_TLB_CNTL, - ECO_BITS, 0); - tmp = REG_SET_FIELD(tmp, VMSHAREDVC0_MC_VM_MX_L1_TLB_CNTL, MTYPE, MTYPE_UC);/* XXX for emulation. */ tmp = REG_SET_FIELD(tmp, VMSHAREDVC0_MC_VM_MX_L1_TLB_CNTL, ATC_EN, 1); diff --git a/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c b/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c index 23b066bcffb2..0077e738db31 100644 --- a/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c +++ b/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.c @@ -252,11 +252,12 @@ static void xgpu_ai_mailbox_flr_work(struct work_struct *work) * otherwise the mailbox msg will be ruined/reseted by * the VF FLR. */ - if (!down_write_trylock(&adev->reset_sem)) + if (atomic_cmpxchg(&adev->in_gpu_reset, 0, 1) != 0) return; + down_write(&adev->reset_sem); + amdgpu_virt_fini_data_exchange(adev); - atomic_set(&adev->in_gpu_reset, 1); xgpu_ai_mailbox_trans_msg(adev, IDH_READY_TO_RESET, 0, 0, 0); diff --git a/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.h b/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.h index bd3b23171579..f9aa4d0bb638 100644 --- a/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.h +++ b/drivers/gpu/drm/amd/amdgpu/mxgpu_ai.h @@ -26,7 +26,7 @@ #define AI_MAILBOX_POLL_ACK_TIMEDOUT 500 #define AI_MAILBOX_POLL_MSG_TIMEDOUT 6000 -#define AI_MAILBOX_POLL_FLR_TIMEDOUT 5000 +#define AI_MAILBOX_POLL_FLR_TIMEDOUT 10000 #define AI_MAILBOX_POLL_MSG_REP_MAX 11 enum idh_request { diff --git a/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.c b/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.c index a35e6d87e537..477d0dde19c5 100644 --- a/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.c +++ b/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.c @@ -281,11 +281,12 @@ static void xgpu_nv_mailbox_flr_work(struct work_struct *work) * otherwise the mailbox msg will be ruined/reseted by * the VF FLR. */ - if (!down_write_trylock(&adev->reset_sem)) + if (atomic_cmpxchg(&adev->in_gpu_reset, 0, 1) != 0) return; + down_write(&adev->reset_sem); + amdgpu_virt_fini_data_exchange(adev); - atomic_set(&adev->in_gpu_reset, 1); xgpu_nv_mailbox_trans_msg(adev, IDH_READY_TO_RESET, 0, 0, 0); diff --git a/drivers/gpu/drm/amd/amdgpu/navi10_ih.c b/drivers/gpu/drm/amd/amdgpu/navi10_ih.c index 38241cf0e1f1..8ce5b8ca1fd7 100644 --- a/drivers/gpu/drm/amd/amdgpu/navi10_ih.c +++ b/drivers/gpu/drm/amd/amdgpu/navi10_ih.c @@ -716,6 +716,7 @@ static const struct amd_ip_funcs navi10_ih_ip_funcs = { static const struct amdgpu_ih_funcs navi10_ih_funcs = { .get_wptr = navi10_ih_get_wptr, .decode_iv = amdgpu_ih_decode_iv_helper, + .decode_iv_ts = amdgpu_ih_decode_iv_ts_helper, .set_rptr = navi10_ih_set_rptr }; diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c b/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c index 853d1511b889..81e033549dda 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v5_0.c @@ -481,8 +481,6 @@ static void sdma_v5_0_ring_emit_ib(struct amdgpu_ring *ring, * sdma_v5_0_ring_emit_mem_sync - flush the IB by graphics cache rinse * * @ring: amdgpu ring pointer - * @job: job to retrieve vmid from - * @ib: IB object to schedule * * flush the IB by graphics cache rinse. */ diff --git a/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c b/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c index 4d4d1aa51b8a..4f546f632223 100644 --- a/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c +++ b/drivers/gpu/drm/amd/amdgpu/sdma_v5_2.c @@ -368,8 +368,6 @@ static void sdma_v5_2_ring_emit_ib(struct amdgpu_ring *ring, * sdma_v5_2_ring_emit_mem_sync - flush the IB by graphics cache rinse * * @ring: amdgpu ring pointer - * @job: job to retrieve vmid from - * @ib: IB object to schedule * * flush the IB by graphics cache rinse. */ diff --git a/drivers/gpu/drm/amd/amdgpu/soc15.c b/drivers/gpu/drm/amd/amdgpu/soc15.c index de9b55383e9f..0fc1747e4a70 100644 --- a/drivers/gpu/drm/amd/amdgpu/soc15.c +++ b/drivers/gpu/drm/amd/amdgpu/soc15.c @@ -744,7 +744,7 @@ static void soc15_reg_base_init(struct amdgpu_device *adev) vega10_reg_base_init(adev); break; case CHIP_RENOIR: - /* It's safe to do ip discovery here for Renior, + /* It's safe to do ip discovery here for Renoir, * it doesn't support SRIOV. */ if (amdgpu_discovery) { r = amdgpu_discovery_reg_base_init(adev); @@ -1238,7 +1238,9 @@ static int soc15_common_sw_init(void *handle) if (amdgpu_sriov_vf(adev)) xgpu_ai_mailbox_add_irq_id(adev); - adev->df.funcs->sw_init(adev); + if (adev->df.funcs && + adev->df.funcs->sw_init) + adev->df.funcs->sw_init(adev); return 0; } @@ -1250,7 +1252,10 @@ static int soc15_common_sw_fini(void *handle) if (adev->nbio.ras_funcs && adev->nbio.ras_funcs->ras_fini) adev->nbio.ras_funcs->ras_fini(adev); - adev->df.funcs->sw_fini(adev); + + if (adev->df.funcs && + adev->df.funcs->sw_fini) + adev->df.funcs->sw_fini(adev); return 0; } diff --git a/drivers/gpu/drm/amd/amdgpu/vega10_ih.c b/drivers/gpu/drm/amd/amdgpu/vega10_ih.c index a9ca6988009e..3070466f54e1 100644 --- a/drivers/gpu/drm/amd/amdgpu/vega10_ih.c +++ b/drivers/gpu/drm/amd/amdgpu/vega10_ih.c @@ -640,6 +640,7 @@ const struct amd_ip_funcs vega10_ih_ip_funcs = { static const struct amdgpu_ih_funcs vega10_ih_funcs = { .get_wptr = vega10_ih_get_wptr, .decode_iv = amdgpu_ih_decode_iv_helper, + .decode_iv_ts = amdgpu_ih_decode_iv_ts_helper, .set_rptr = vega10_ih_set_rptr }; diff --git a/drivers/gpu/drm/amd/amdgpu/vega20_ih.c b/drivers/gpu/drm/amd/amdgpu/vega20_ih.c index f51dfc38ac65..3b4eb8285943 100644 --- a/drivers/gpu/drm/amd/amdgpu/vega20_ih.c +++ b/drivers/gpu/drm/amd/amdgpu/vega20_ih.c @@ -688,6 +688,7 @@ const struct amd_ip_funcs vega20_ih_ip_funcs = { static const struct amdgpu_ih_funcs vega20_ih_funcs = { .get_wptr = vega20_ih_get_wptr, .decode_iv = amdgpu_ih_decode_iv_helper, + .decode_iv_ts = amdgpu_ih_decode_iv_ts_helper, .set_rptr = vega20_ih_set_rptr }; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device.c b/drivers/gpu/drm/amd/amdkfd/kfd_device.c index e1294fba0c26..facc28f58c1f 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_device.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_device.c @@ -53,680 +53,242 @@ extern const struct kfd2kgd_calls aldebaran_kfd2kgd; extern const struct kfd2kgd_calls gfx_v10_kfd2kgd; extern const struct kfd2kgd_calls gfx_v10_3_kfd2kgd; -#ifdef KFD_SUPPORT_IOMMU_V2 -static const struct kfd_device_info kaveri_device_info = { - .asic_name = "kaveri", - .gfx_target_version = 70000, - .max_pasid_bits = 16, - /* max num of queues for KV.TODO should be a dynamic value */ - .max_no_of_hqd = 24, - .doorbell_size = 4, - .ih_ring_entry_size = 4 * sizeof(uint32_t), - .event_interrupt_class = &event_interrupt_class_cik, - .num_of_watch_points = 4, - .mqd_size_aligned = MQD_SIZE_ALIGNED, - .supports_cwsr = false, - .needs_iommu_device = true, - .needs_pci_atomics = false, - .num_sdma_queues_per_engine = 2, -}; - -static const struct kfd_device_info carrizo_device_info = { - .asic_name = "carrizo", - .gfx_target_version = 80001, - .max_pasid_bits = 16, - /* max num of queues for CZ.TODO should be a dynamic value */ - .max_no_of_hqd = 24, - .doorbell_size = 4, - .ih_ring_entry_size = 4 * sizeof(uint32_t), - .event_interrupt_class = &event_interrupt_class_cik, - .num_of_watch_points = 4, - .mqd_size_aligned = MQD_SIZE_ALIGNED, - .supports_cwsr = true, - .needs_iommu_device = true, - .needs_pci_atomics = false, - .num_sdma_queues_per_engine = 2, -}; - -static const struct kfd_device_info raven_device_info = { - .asic_name = "raven", - .gfx_target_version = 90002, - .max_pasid_bits = 16, - .max_no_of_hqd = 24, - .doorbell_size = 8, - .ih_ring_entry_size = 8 * sizeof(uint32_t), - .event_interrupt_class = &event_interrupt_class_v9, - .num_of_watch_points = 4, - .mqd_size_aligned = MQD_SIZE_ALIGNED, - .supports_cwsr = true, - .needs_iommu_device = true, - .needs_pci_atomics = true, - .num_sdma_queues_per_engine = 2, -}; -#endif - -#ifdef CONFIG_DRM_AMDGPU_CIK -static const struct kfd_device_info hawaii_device_info = { - .asic_name = "hawaii", - .gfx_target_version = 70001, - .max_pasid_bits = 16, - /* max num of queues for KV.TODO should be a dynamic value */ - .max_no_of_hqd = 24, - .doorbell_size = 4, - .ih_ring_entry_size = 4 * sizeof(uint32_t), - .event_interrupt_class = &event_interrupt_class_cik, - .num_of_watch_points = 4, - .mqd_size_aligned = MQD_SIZE_ALIGNED, - .supports_cwsr = false, - .needs_iommu_device = false, - .needs_pci_atomics = false, - .num_sdma_queues_per_engine = 2, -}; -#endif - -static const struct kfd_device_info tonga_device_info = { - .asic_name = "tonga", - .gfx_target_version = 80002, - .max_pasid_bits = 16, - .max_no_of_hqd = 24, - .doorbell_size = 4, - .ih_ring_entry_size = 4 * sizeof(uint32_t), - .event_interrupt_class = &event_interrupt_class_cik, - .num_of_watch_points = 4, - .mqd_size_aligned = MQD_SIZE_ALIGNED, - .supports_cwsr = false, - .needs_iommu_device = false, - .needs_pci_atomics = true, - .num_sdma_queues_per_engine = 2, -}; - -static const struct kfd_device_info fiji_device_info = { - .asic_name = "fiji", - .gfx_target_version = 80003, - .max_pasid_bits = 16, - .max_no_of_hqd = 24, - .doorbell_size = 4, - .ih_ring_entry_size = 4 * sizeof(uint32_t), - .event_interrupt_class = &event_interrupt_class_cik, - .num_of_watch_points = 4, - .mqd_size_aligned = MQD_SIZE_ALIGNED, - .supports_cwsr = true, - .needs_iommu_device = false, - .needs_pci_atomics = true, - .num_sdma_queues_per_engine = 2, -}; - -static const struct kfd_device_info fiji_vf_device_info = { - .asic_name = "fiji", - .gfx_target_version = 80003, - .max_pasid_bits = 16, - .max_no_of_hqd = 24, - .doorbell_size = 4, - .ih_ring_entry_size = 4 * sizeof(uint32_t), - .event_interrupt_class = &event_interrupt_class_cik, - .num_of_watch_points = 4, - .mqd_size_aligned = MQD_SIZE_ALIGNED, - .supports_cwsr = true, - .needs_iommu_device = false, - .needs_pci_atomics = false, - .num_sdma_queues_per_engine = 2, -}; - - -static const struct kfd_device_info polaris10_device_info = { - .asic_name = "polaris10", - .gfx_target_version = 80003, - .max_pasid_bits = 16, - .max_no_of_hqd = 24, - .doorbell_size = 4, - .ih_ring_entry_size = 4 * sizeof(uint32_t), - .event_interrupt_class = &event_interrupt_class_cik, - .num_of_watch_points = 4, - .mqd_size_aligned = MQD_SIZE_ALIGNED, - .supports_cwsr = true, - .needs_iommu_device = false, - .needs_pci_atomics = true, - .num_sdma_queues_per_engine = 2, -}; - -static const struct kfd_device_info polaris10_vf_device_info = { - .asic_name = "polaris10", - .gfx_target_version = 80003, - .max_pasid_bits = 16, - .max_no_of_hqd = 24, - .doorbell_size = 4, - .ih_ring_entry_size = 4 * sizeof(uint32_t), - .event_interrupt_class = &event_interrupt_class_cik, - .num_of_watch_points = 4, - .mqd_size_aligned = MQD_SIZE_ALIGNED, - .supports_cwsr = true, - .needs_iommu_device = false, - .needs_pci_atomics = false, - .num_sdma_queues_per_engine = 2, -}; - -static const struct kfd_device_info polaris11_device_info = { - .asic_name = "polaris11", - .gfx_target_version = 80003, - .max_pasid_bits = 16, - .max_no_of_hqd = 24, - .doorbell_size = 4, - .ih_ring_entry_size = 4 * sizeof(uint32_t), - .event_interrupt_class = &event_interrupt_class_cik, - .num_of_watch_points = 4, - .mqd_size_aligned = MQD_SIZE_ALIGNED, - .supports_cwsr = true, - .needs_iommu_device = false, - .needs_pci_atomics = true, - .num_sdma_queues_per_engine = 2, -}; - -static const struct kfd_device_info polaris12_device_info = { - .asic_name = "polaris12", - .gfx_target_version = 80003, - .max_pasid_bits = 16, - .max_no_of_hqd = 24, - .doorbell_size = 4, - .ih_ring_entry_size = 4 * sizeof(uint32_t), - .event_interrupt_class = &event_interrupt_class_cik, - .num_of_watch_points = 4, - .mqd_size_aligned = MQD_SIZE_ALIGNED, - .supports_cwsr = true, - .needs_iommu_device = false, - .needs_pci_atomics = true, - .num_sdma_queues_per_engine = 2, -}; - -static const struct kfd_device_info vegam_device_info = { - .asic_name = "vegam", - .gfx_target_version = 80003, - .max_pasid_bits = 16, - .max_no_of_hqd = 24, - .doorbell_size = 4, - .ih_ring_entry_size = 4 * sizeof(uint32_t), - .event_interrupt_class = &event_interrupt_class_cik, - .num_of_watch_points = 4, - .mqd_size_aligned = MQD_SIZE_ALIGNED, - .supports_cwsr = true, - .needs_iommu_device = false, - .needs_pci_atomics = true, - .num_sdma_queues_per_engine = 2, -}; - -static const struct kfd_device_info vega10_device_info = { - .asic_name = "vega10", - .gfx_target_version = 90000, - .max_pasid_bits = 16, - .max_no_of_hqd = 24, - .doorbell_size = 8, - .ih_ring_entry_size = 8 * sizeof(uint32_t), - .event_interrupt_class = &event_interrupt_class_v9, - .num_of_watch_points = 4, - .mqd_size_aligned = MQD_SIZE_ALIGNED, - .supports_cwsr = true, - .needs_iommu_device = false, - .needs_pci_atomics = false, - .num_sdma_queues_per_engine = 2, -}; - -static const struct kfd_device_info vega10_vf_device_info = { - .asic_name = "vega10", - .gfx_target_version = 90000, - .max_pasid_bits = 16, - .max_no_of_hqd = 24, - .doorbell_size = 8, - .ih_ring_entry_size = 8 * sizeof(uint32_t), - .event_interrupt_class = &event_interrupt_class_v9, - .num_of_watch_points = 4, - .mqd_size_aligned = MQD_SIZE_ALIGNED, - .supports_cwsr = true, - .needs_iommu_device = false, - .needs_pci_atomics = false, - .num_sdma_queues_per_engine = 2, -}; - -static const struct kfd_device_info vega12_device_info = { - .asic_name = "vega12", - .gfx_target_version = 90004, - .max_pasid_bits = 16, - .max_no_of_hqd = 24, - .doorbell_size = 8, - .ih_ring_entry_size = 8 * sizeof(uint32_t), - .event_interrupt_class = &event_interrupt_class_v9, - .num_of_watch_points = 4, - .mqd_size_aligned = MQD_SIZE_ALIGNED, - .supports_cwsr = true, - .needs_iommu_device = false, - .needs_pci_atomics = false, - .num_sdma_queues_per_engine = 2, -}; - -static const struct kfd_device_info vega20_device_info = { - .asic_name = "vega20", - .gfx_target_version = 90006, - .max_pasid_bits = 16, - .max_no_of_hqd = 24, - .doorbell_size = 8, - .ih_ring_entry_size = 8 * sizeof(uint32_t), - .event_interrupt_class = &event_interrupt_class_v9, - .num_of_watch_points = 4, - .mqd_size_aligned = MQD_SIZE_ALIGNED, - .supports_cwsr = true, - .needs_iommu_device = false, - .needs_pci_atomics = false, - .num_sdma_queues_per_engine = 8, -}; - -static const struct kfd_device_info arcturus_device_info = { - .asic_name = "arcturus", - .gfx_target_version = 90008, - .max_pasid_bits = 16, - .max_no_of_hqd = 24, - .doorbell_size = 8, - .ih_ring_entry_size = 8 * sizeof(uint32_t), - .event_interrupt_class = &event_interrupt_class_v9, - .num_of_watch_points = 4, - .mqd_size_aligned = MQD_SIZE_ALIGNED, - .supports_cwsr = true, - .needs_iommu_device = false, - .needs_pci_atomics = false, - .num_sdma_queues_per_engine = 8, -}; - -static const struct kfd_device_info aldebaran_device_info = { - .asic_name = "aldebaran", - .gfx_target_version = 90010, - .max_pasid_bits = 16, - .max_no_of_hqd = 24, - .doorbell_size = 8, - .ih_ring_entry_size = 8 * sizeof(uint32_t), - .event_interrupt_class = &event_interrupt_class_v9, - .num_of_watch_points = 4, - .mqd_size_aligned = MQD_SIZE_ALIGNED, - .supports_cwsr = true, - .needs_iommu_device = false, - .needs_pci_atomics = false, - .num_sdma_queues_per_engine = 8, -}; - -static const struct kfd_device_info renoir_device_info = { - .asic_name = "renoir", - .gfx_target_version = 90012, - .max_pasid_bits = 16, - .max_no_of_hqd = 24, - .doorbell_size = 8, - .ih_ring_entry_size = 8 * sizeof(uint32_t), - .event_interrupt_class = &event_interrupt_class_v9, - .num_of_watch_points = 4, - .mqd_size_aligned = MQD_SIZE_ALIGNED, - .supports_cwsr = true, - .needs_iommu_device = false, - .needs_pci_atomics = false, - .num_sdma_queues_per_engine = 2, -}; - -static const struct kfd_device_info navi10_device_info = { - .asic_name = "navi10", - .gfx_target_version = 100100, - .max_pasid_bits = 16, - .max_no_of_hqd = 24, - .doorbell_size = 8, - .ih_ring_entry_size = 8 * sizeof(uint32_t), - .event_interrupt_class = &event_interrupt_class_v9, - .num_of_watch_points = 4, - .mqd_size_aligned = MQD_SIZE_ALIGNED, - .needs_iommu_device = false, - .supports_cwsr = true, - .needs_pci_atomics = true, - .no_atomic_fw_version = 145, - .num_sdma_queues_per_engine = 8, -}; - -static const struct kfd_device_info navi12_device_info = { - .asic_name = "navi12", - .gfx_target_version = 100101, - .max_pasid_bits = 16, - .max_no_of_hqd = 24, - .doorbell_size = 8, - .ih_ring_entry_size = 8 * sizeof(uint32_t), - .event_interrupt_class = &event_interrupt_class_v9, - .num_of_watch_points = 4, - .mqd_size_aligned = MQD_SIZE_ALIGNED, - .needs_iommu_device = false, - .supports_cwsr = true, - .needs_pci_atomics = true, - .no_atomic_fw_version = 145, - .num_sdma_queues_per_engine = 8, -}; - -static const struct kfd_device_info navi14_device_info = { - .asic_name = "navi14", - .gfx_target_version = 100102, - .max_pasid_bits = 16, - .max_no_of_hqd = 24, - .doorbell_size = 8, - .ih_ring_entry_size = 8 * sizeof(uint32_t), - .event_interrupt_class = &event_interrupt_class_v9, - .num_of_watch_points = 4, - .mqd_size_aligned = MQD_SIZE_ALIGNED, - .needs_iommu_device = false, - .supports_cwsr = true, - .needs_pci_atomics = true, - .no_atomic_fw_version = 145, - .num_sdma_queues_per_engine = 8, -}; - -static const struct kfd_device_info sienna_cichlid_device_info = { - .asic_name = "sienna_cichlid", - .gfx_target_version = 100300, - .max_pasid_bits = 16, - .max_no_of_hqd = 24, - .doorbell_size = 8, - .ih_ring_entry_size = 8 * sizeof(uint32_t), - .event_interrupt_class = &event_interrupt_class_v9, - .num_of_watch_points = 4, - .mqd_size_aligned = MQD_SIZE_ALIGNED, - .needs_iommu_device = false, - .supports_cwsr = true, - .needs_pci_atomics = true, - .no_atomic_fw_version = 92, - .num_sdma_queues_per_engine = 8, -}; - -static const struct kfd_device_info navy_flounder_device_info = { - .asic_name = "navy_flounder", - .gfx_target_version = 100301, - .max_pasid_bits = 16, - .max_no_of_hqd = 24, - .doorbell_size = 8, - .ih_ring_entry_size = 8 * sizeof(uint32_t), - .event_interrupt_class = &event_interrupt_class_v9, - .num_of_watch_points = 4, - .mqd_size_aligned = MQD_SIZE_ALIGNED, - .needs_iommu_device = false, - .supports_cwsr = true, - .needs_pci_atomics = true, - .no_atomic_fw_version = 92, - .num_sdma_queues_per_engine = 8, -}; - -static const struct kfd_device_info vangogh_device_info = { - .asic_name = "vangogh", - .gfx_target_version = 100303, - .max_pasid_bits = 16, - .max_no_of_hqd = 24, - .doorbell_size = 8, - .ih_ring_entry_size = 8 * sizeof(uint32_t), - .event_interrupt_class = &event_interrupt_class_v9, - .num_of_watch_points = 4, - .mqd_size_aligned = MQD_SIZE_ALIGNED, - .needs_iommu_device = false, - .supports_cwsr = true, - .needs_pci_atomics = true, - .no_atomic_fw_version = 92, - .num_sdma_queues_per_engine = 2, -}; - -static const struct kfd_device_info dimgrey_cavefish_device_info = { - .asic_name = "dimgrey_cavefish", - .gfx_target_version = 100302, - .max_pasid_bits = 16, - .max_no_of_hqd = 24, - .doorbell_size = 8, - .ih_ring_entry_size = 8 * sizeof(uint32_t), - .event_interrupt_class = &event_interrupt_class_v9, - .num_of_watch_points = 4, - .mqd_size_aligned = MQD_SIZE_ALIGNED, - .needs_iommu_device = false, - .supports_cwsr = true, - .needs_pci_atomics = true, - .no_atomic_fw_version = 92, - .num_sdma_queues_per_engine = 8, -}; - -static const struct kfd_device_info beige_goby_device_info = { - .asic_name = "beige_goby", - .gfx_target_version = 100304, - .max_pasid_bits = 16, - .max_no_of_hqd = 24, - .doorbell_size = 8, - .ih_ring_entry_size = 8 * sizeof(uint32_t), - .event_interrupt_class = &event_interrupt_class_v9, - .num_of_watch_points = 4, - .mqd_size_aligned = MQD_SIZE_ALIGNED, - .needs_iommu_device = false, - .supports_cwsr = true, - .needs_pci_atomics = true, - .no_atomic_fw_version = 92, - .num_sdma_queues_per_engine = 8, -}; - -static const struct kfd_device_info yellow_carp_device_info = { - .asic_name = "yellow_carp", - .gfx_target_version = 100305, - .max_pasid_bits = 16, - .max_no_of_hqd = 24, - .doorbell_size = 8, - .ih_ring_entry_size = 8 * sizeof(uint32_t), - .event_interrupt_class = &event_interrupt_class_v9, - .num_of_watch_points = 4, - .mqd_size_aligned = MQD_SIZE_ALIGNED, - .needs_iommu_device = false, - .supports_cwsr = true, - .needs_pci_atomics = true, - .no_atomic_fw_version = 92, - .num_sdma_queues_per_engine = 2, -}; - -static const struct kfd_device_info cyan_skillfish_device_info = { - .asic_name = "cyan_skillfish", - .gfx_target_version = 100103, - .max_pasid_bits = 16, - .max_no_of_hqd = 24, - .doorbell_size = 8, - .ih_ring_entry_size = 8 * sizeof(uint32_t), - .event_interrupt_class = &event_interrupt_class_v9, - .num_of_watch_points = 4, - .mqd_size_aligned = MQD_SIZE_ALIGNED, - .needs_iommu_device = false, - .supports_cwsr = true, - .needs_pci_atomics = true, - .num_sdma_queues_per_engine = 8, -}; - static int kfd_gtt_sa_init(struct kfd_dev *kfd, unsigned int buf_size, unsigned int chunk_size); static void kfd_gtt_sa_fini(struct kfd_dev *kfd); static int kfd_resume(struct kfd_dev *kfd); +static void kfd_device_info_init(struct kfd_dev *kfd, + bool vf, uint32_t gfx_target_version) +{ + uint32_t gc_version = KFD_GC_VERSION(kfd); + uint32_t sdma_version = kfd->adev->ip_versions[SDMA0_HWIP][0]; + uint32_t asic_type = kfd->adev->asic_type; + + kfd->device_info.max_pasid_bits = 16; + kfd->device_info.max_no_of_hqd = 24; + kfd->device_info.num_of_watch_points = 4; + kfd->device_info.mqd_size_aligned = MQD_SIZE_ALIGNED; + kfd->device_info.gfx_target_version = gfx_target_version; + + if (KFD_IS_SOC15(kfd)) { + kfd->device_info.doorbell_size = 8; + kfd->device_info.ih_ring_entry_size = 8 * sizeof(uint32_t); + kfd->device_info.event_interrupt_class = &event_interrupt_class_v9; + kfd->device_info.supports_cwsr = true; + + if ((sdma_version >= IP_VERSION(4, 0, 0) && + sdma_version <= IP_VERSION(4, 2, 0)) || + sdma_version == IP_VERSION(5, 2, 1) || + sdma_version == IP_VERSION(5, 2, 3)) + kfd->device_info.num_sdma_queues_per_engine = 2; + else + kfd->device_info.num_sdma_queues_per_engine = 8; + + /* Raven */ + if (gc_version == IP_VERSION(9, 1, 0) || + gc_version == IP_VERSION(9, 2, 2)) + kfd->device_info.needs_iommu_device = true; + + if (gc_version < IP_VERSION(11, 0, 0)) { + /* Navi2x+, Navi1x+ */ + if (gc_version >= IP_VERSION(10, 3, 0)) + kfd->device_info.no_atomic_fw_version = 92; + else if (gc_version >= IP_VERSION(10, 1, 1)) + kfd->device_info.no_atomic_fw_version = 145; + + /* Navi1x+ */ + if (gc_version >= IP_VERSION(10, 1, 1)) + kfd->device_info.needs_pci_atomics = true; + } + } else { + kfd->device_info.doorbell_size = 4; + kfd->device_info.ih_ring_entry_size = 4 * sizeof(uint32_t); + kfd->device_info.event_interrupt_class = &event_interrupt_class_cik; + kfd->device_info.num_sdma_queues_per_engine = 2; + + if (asic_type != CHIP_KAVERI && + asic_type != CHIP_HAWAII && + asic_type != CHIP_TONGA) + kfd->device_info.supports_cwsr = true; + + if (asic_type == CHIP_KAVERI || + asic_type == CHIP_CARRIZO) + kfd->device_info.needs_iommu_device = true; + + if (asic_type != CHIP_HAWAII && !vf) + kfd->device_info.needs_pci_atomics = true; + } +} + struct kfd_dev *kgd2kfd_probe(struct amdgpu_device *adev, bool vf) { - struct kfd_dev *kfd; - const struct kfd_device_info *device_info; - const struct kfd2kgd_calls *f2g; + struct kfd_dev *kfd = NULL; + const struct kfd2kgd_calls *f2g = NULL; struct pci_dev *pdev = adev->pdev; + uint32_t gfx_target_version = 0; switch (adev->asic_type) { #ifdef KFD_SUPPORT_IOMMU_V2 #ifdef CONFIG_DRM_AMDGPU_CIK case CHIP_KAVERI: - if (vf) - device_info = NULL; - else - device_info = &kaveri_device_info; - f2g = &gfx_v7_kfd2kgd; + gfx_target_version = 70000; + if (!vf) + f2g = &gfx_v7_kfd2kgd; break; #endif case CHIP_CARRIZO: - if (vf) - device_info = NULL; - else - device_info = &carrizo_device_info; - f2g = &gfx_v8_kfd2kgd; + gfx_target_version = 80001; + if (!vf) + f2g = &gfx_v8_kfd2kgd; break; #endif #ifdef CONFIG_DRM_AMDGPU_CIK case CHIP_HAWAII: - if (vf) - device_info = NULL; - else - device_info = &hawaii_device_info; - f2g = &gfx_v7_kfd2kgd; + gfx_target_version = 70001; + if (!amdgpu_exp_hw_support) + pr_info( + "KFD support on Hawaii is experimental. See modparam exp_hw_support\n" + ); + else if (!vf) + f2g = &gfx_v7_kfd2kgd; break; #endif case CHIP_TONGA: - if (vf) - device_info = NULL; - else - device_info = &tonga_device_info; - f2g = &gfx_v8_kfd2kgd; + gfx_target_version = 80002; + if (!vf) + f2g = &gfx_v8_kfd2kgd; break; case CHIP_FIJI: - if (vf) - device_info = &fiji_vf_device_info; - else - device_info = &fiji_device_info; + gfx_target_version = 80003; f2g = &gfx_v8_kfd2kgd; break; case CHIP_POLARIS10: - if (vf) - device_info = &polaris10_vf_device_info; - else - device_info = &polaris10_device_info; + gfx_target_version = 80003; f2g = &gfx_v8_kfd2kgd; break; case CHIP_POLARIS11: - if (vf) - device_info = NULL; - else - device_info = &polaris11_device_info; - f2g = &gfx_v8_kfd2kgd; + gfx_target_version = 80003; + if (!vf) + f2g = &gfx_v8_kfd2kgd; break; case CHIP_POLARIS12: - if (vf) - device_info = NULL; - else - device_info = &polaris12_device_info; - f2g = &gfx_v8_kfd2kgd; + gfx_target_version = 80003; + if (!vf) + f2g = &gfx_v8_kfd2kgd; break; case CHIP_VEGAM: - if (vf) - device_info = NULL; - else - device_info = &vegam_device_info; - f2g = &gfx_v8_kfd2kgd; + gfx_target_version = 80003; + if (!vf) + f2g = &gfx_v8_kfd2kgd; break; default: switch (adev->ip_versions[GC_HWIP][0]) { + /* Vega 10 */ case IP_VERSION(9, 0, 1): - if (vf) - device_info = &vega10_vf_device_info; - else - device_info = &vega10_device_info; + gfx_target_version = 90000; f2g = &gfx_v9_kfd2kgd; break; #ifdef KFD_SUPPORT_IOMMU_V2 + /* Raven */ case IP_VERSION(9, 1, 0): case IP_VERSION(9, 2, 2): - if (vf) - device_info = NULL; - else - device_info = &raven_device_info; - f2g = &gfx_v9_kfd2kgd; + gfx_target_version = 90002; + if (!vf) + f2g = &gfx_v9_kfd2kgd; break; #endif + /* Vega12 */ case IP_VERSION(9, 2, 1): - if (vf) - device_info = NULL; - else - device_info = &vega12_device_info; - f2g = &gfx_v9_kfd2kgd; + gfx_target_version = 90004; + if (!vf) + f2g = &gfx_v9_kfd2kgd; break; + /* Renoir */ case IP_VERSION(9, 3, 0): - if (vf) - device_info = NULL; - else - device_info = &renoir_device_info; - f2g = &gfx_v9_kfd2kgd; + gfx_target_version = 90012; + if (!vf) + f2g = &gfx_v9_kfd2kgd; break; + /* Vega20 */ case IP_VERSION(9, 4, 0): - if (vf) - device_info = NULL; - else - device_info = &vega20_device_info; - f2g = &gfx_v9_kfd2kgd; + gfx_target_version = 90006; + if (!vf) + f2g = &gfx_v9_kfd2kgd; break; + /* Arcturus */ case IP_VERSION(9, 4, 1): - device_info = &arcturus_device_info; + gfx_target_version = 90008; f2g = &arcturus_kfd2kgd; break; + /* Aldebaran */ case IP_VERSION(9, 4, 2): - device_info = &aldebaran_device_info; + gfx_target_version = 90010; f2g = &aldebaran_kfd2kgd; break; + /* Navi10 */ case IP_VERSION(10, 1, 10): - if (vf) - device_info = NULL; - else - device_info = &navi10_device_info; - f2g = &gfx_v10_kfd2kgd; + gfx_target_version = 100100; + if (!vf) + f2g = &gfx_v10_kfd2kgd; break; + /* Navi12 */ case IP_VERSION(10, 1, 2): - device_info = &navi12_device_info; + gfx_target_version = 100101; f2g = &gfx_v10_kfd2kgd; break; + /* Navi14 */ case IP_VERSION(10, 1, 1): - if (vf) - device_info = NULL; - else - device_info = &navi14_device_info; - f2g = &gfx_v10_kfd2kgd; + gfx_target_version = 100102; + if (!vf) + f2g = &gfx_v10_kfd2kgd; break; + /* Cyan Skillfish */ case IP_VERSION(10, 1, 3): - if (vf) - device_info = NULL; - else - device_info = &cyan_skillfish_device_info; - f2g = &gfx_v10_kfd2kgd; + gfx_target_version = 100103; + if (!vf) + f2g = &gfx_v10_kfd2kgd; break; + /* Sienna Cichlid */ case IP_VERSION(10, 3, 0): - device_info = &sienna_cichlid_device_info; + gfx_target_version = 100300; f2g = &gfx_v10_3_kfd2kgd; break; + /* Navy Flounder */ case IP_VERSION(10, 3, 2): - device_info = &navy_flounder_device_info; + gfx_target_version = 100301; f2g = &gfx_v10_3_kfd2kgd; break; + /* Van Gogh */ case IP_VERSION(10, 3, 1): - if (vf) - device_info = NULL; - else - device_info = &vangogh_device_info; - f2g = &gfx_v10_3_kfd2kgd; + gfx_target_version = 100303; + if (!vf) + f2g = &gfx_v10_3_kfd2kgd; break; + /* Dimgrey Cavefish */ case IP_VERSION(10, 3, 4): - device_info = &dimgrey_cavefish_device_info; + gfx_target_version = 100302; f2g = &gfx_v10_3_kfd2kgd; break; + /* Beige Goby */ case IP_VERSION(10, 3, 5): - device_info = &beige_goby_device_info; + gfx_target_version = 100304; f2g = &gfx_v10_3_kfd2kgd; break; + /* Yellow Carp */ case IP_VERSION(10, 3, 3): - if (vf) - device_info = NULL; - else - device_info = &yellow_carp_device_info; - f2g = &gfx_v10_3_kfd2kgd; + gfx_target_version = 100305; + if (!vf) + f2g = &gfx_v10_3_kfd2kgd; break; default: - return NULL; + break; } break; } - if (!device_info || !f2g) { + if (!f2g) { if (adev->ip_versions[GC_HWIP][0]) dev_err(kfd_device, "GC IP %06x %s not supported in kfd\n", adev->ip_versions[GC_HWIP][0], vf ? "VF" : ""); @@ -741,7 +303,7 @@ struct kfd_dev *kgd2kfd_probe(struct amdgpu_device *adev, bool vf) return NULL; kfd->adev = adev; - kfd->device_info = device_info; + kfd_device_info_init(kfd, vf, gfx_target_version); kfd->pdev = pdev; kfd->init_complete = false; kfd->kfd2kgd = f2g; @@ -760,7 +322,7 @@ struct kfd_dev *kgd2kfd_probe(struct amdgpu_device *adev, bool vf) static void kfd_cwsr_init(struct kfd_dev *kfd) { - if (cwsr_enable && kfd->device_info->supports_cwsr) { + if (cwsr_enable && kfd->device_info.supports_cwsr) { if (KFD_GC_VERSION(kfd) < IP_VERSION(9, 0, 1)) { BUILD_BUG_ON(sizeof(cwsr_trap_gfx8_hex) > PAGE_SIZE); kfd->cwsr_isa = cwsr_trap_gfx8_hex; @@ -844,14 +406,14 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd, */ kfd->pci_atomic_requested = amdgpu_amdkfd_have_atomics_support(kfd->adev); if (!kfd->pci_atomic_requested && - kfd->device_info->needs_pci_atomics && - (!kfd->device_info->no_atomic_fw_version || - kfd->mec_fw_version < kfd->device_info->no_atomic_fw_version)) { + kfd->device_info.needs_pci_atomics && + (!kfd->device_info.no_atomic_fw_version || + kfd->mec_fw_version < kfd->device_info.no_atomic_fw_version)) { dev_info(kfd_device, "skipped device %x:%x, PCI rejects atomics %d<%d\n", kfd->pdev->vendor, kfd->pdev->device, kfd->mec_fw_version, - kfd->device_info->no_atomic_fw_version); + kfd->device_info.no_atomic_fw_version); return false; } @@ -868,7 +430,7 @@ bool kgd2kfd_device_init(struct kfd_dev *kfd, /* calculate max size of mqds needed for queues */ size = max_num_of_queues_per_device * - kfd->device_info->mqd_size_aligned; + kfd->device_info.mqd_size_aligned; /* * calculate max size of runlist packet. @@ -1143,7 +705,7 @@ void kgd2kfd_interrupt(struct kfd_dev *kfd, const void *ih_ring_entry) if (!kfd->init_complete) return; - if (kfd->device_info->ih_ring_entry_size > sizeof(patched_ihre)) { + if (kfd->device_info.ih_ring_entry_size > sizeof(patched_ihre)) { dev_err_once(kfd_device, "Ring entry too small\n"); return; } diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c index 2af2b3268171..dd0b952f0173 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager.c @@ -108,13 +108,13 @@ static unsigned int get_num_all_sdma_engines(struct device_queue_manager *dqm) unsigned int get_num_sdma_queues(struct device_queue_manager *dqm) { return kfd_get_num_sdma_engines(dqm->dev) * - dqm->dev->device_info->num_sdma_queues_per_engine; + dqm->dev->device_info.num_sdma_queues_per_engine; } unsigned int get_num_xgmi_sdma_queues(struct device_queue_manager *dqm) { return kfd_get_num_xgmi_sdma_engines(dqm->dev) * - dqm->dev->device_info->num_sdma_queues_per_engine; + dqm->dev->device_info.num_sdma_queues_per_engine; } void program_sh_mem_settings(struct device_queue_manager *dqm, @@ -1838,7 +1838,7 @@ static int allocate_hiq_sdma_mqd(struct device_queue_manager *dqm) struct kfd_mem_obj *mem_obj = &dqm->hiq_sdma_mqd; uint32_t size = dqm->mqd_mgrs[KFD_MQD_TYPE_SDMA]->mqd_size * get_num_all_sdma_engines(dqm) * - dev->device_info->num_sdma_queues_per_engine + + dev->device_info.num_sdma_queues_per_engine + dqm->mqd_mgrs[KFD_MQD_TYPE_HIQ]->mqd_size; retval = amdgpu_amdkfd_alloc_gtt_mem(dev->adev, size, @@ -2082,7 +2082,7 @@ int dqm_debugfs_hqds(struct seq_file *m, void *data) for (pipe = 0; pipe < get_num_all_sdma_engines(dqm); pipe++) { for (queue = 0; - queue < dqm->dev->device_info->num_sdma_queues_per_engine; + queue < dqm->dev->device_info.num_sdma_queues_per_engine; queue++) { r = dqm->dev->kfd2kgd->hqd_sdma_dump( dqm->dev->adev, pipe, queue, &dump, &n_regs); diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_doorbell.c b/drivers/gpu/drm/amd/amdkfd/kfd_doorbell.c index 768d153acff4..0dbcf54657ed 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_doorbell.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_doorbell.c @@ -48,7 +48,7 @@ /* # of doorbell bytes allocated for each process. */ size_t kfd_doorbell_process_slice(struct kfd_dev *kfd) { - return roundup(kfd->device_info->doorbell_size * + return roundup(kfd->device_info.doorbell_size * KFD_MAX_NUM_OF_QUEUES_PER_PROCESS, PAGE_SIZE); } @@ -180,7 +180,7 @@ void __iomem *kfd_get_kernel_doorbell(struct kfd_dev *kfd, if (inx >= KFD_MAX_NUM_OF_QUEUES_PER_PROCESS) return NULL; - inx *= kfd->device_info->doorbell_size / sizeof(u32); + inx *= kfd->device_info.doorbell_size / sizeof(u32); /* * Calculating the kernel doorbell offset using the first @@ -201,7 +201,7 @@ void kfd_release_kernel_doorbell(struct kfd_dev *kfd, u32 __iomem *db_addr) unsigned int inx; inx = (unsigned int)(db_addr - kfd->doorbell_kernel_ptr) - * sizeof(u32) / kfd->device_info->doorbell_size; + * sizeof(u32) / kfd->device_info.doorbell_size; mutex_lock(&kfd->doorbell_mutex); __clear_bit(inx, kfd->doorbell_available_index); @@ -239,7 +239,7 @@ unsigned int kfd_get_doorbell_dw_offset_in_bar(struct kfd_dev *kfd, return kfd->doorbell_base_dw_offset + pdd->doorbell_index * kfd_doorbell_process_slice(kfd) / sizeof(u32) + - doorbell_id * kfd->device_info->doorbell_size / sizeof(u32); + doorbell_id * kfd->device_info.doorbell_size / sizeof(u32); } uint64_t kfd_get_number_elems(struct kfd_dev *kfd) diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c index 20512a4e9a91..deb64168c9e8 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_int_process_v9.c @@ -135,7 +135,7 @@ static bool event_interrupt_isr_v9(struct kfd_dev *dev, *patched_flag = true; memcpy(patched_ihre, ih_ring_entry, - dev->device_info->ih_ring_entry_size); + dev->device_info.ih_ring_entry_size); pasid = dev->dqm->vmid_pasid[vmid]; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_interrupt.c b/drivers/gpu/drm/amd/amdkfd/kfd_interrupt.c index bc47f6a44456..81887c2013c9 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_interrupt.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_interrupt.c @@ -54,7 +54,7 @@ int kfd_interrupt_init(struct kfd_dev *kfd) int r; r = kfifo_alloc(&kfd->ih_fifo, - KFD_IH_NUM_ENTRIES * kfd->device_info->ih_ring_entry_size, + KFD_IH_NUM_ENTRIES * kfd->device_info.ih_ring_entry_size, GFP_KERNEL); if (r) { dev_err(kfd_chardev(), "Failed to allocate IH fifo\n"); @@ -114,8 +114,8 @@ bool enqueue_ih_ring_entry(struct kfd_dev *kfd, const void *ih_ring_entry) int count; count = kfifo_in(&kfd->ih_fifo, ih_ring_entry, - kfd->device_info->ih_ring_entry_size); - if (count != kfd->device_info->ih_ring_entry_size) { + kfd->device_info.ih_ring_entry_size); + if (count != kfd->device_info.ih_ring_entry_size) { dev_err_ratelimited(kfd_chardev(), "Interrupt ring overflow, dropping interrupt %d\n", count); @@ -133,11 +133,11 @@ static bool dequeue_ih_ring_entry(struct kfd_dev *kfd, void *ih_ring_entry) int count; count = kfifo_out(&kfd->ih_fifo, ih_ring_entry, - kfd->device_info->ih_ring_entry_size); + kfd->device_info.ih_ring_entry_size); - WARN_ON(count && count != kfd->device_info->ih_ring_entry_size); + WARN_ON(count && count != kfd->device_info.ih_ring_entry_size); - return count == kfd->device_info->ih_ring_entry_size; + return count == kfd->device_info.ih_ring_entry_size; } static void interrupt_wq(struct work_struct *work) @@ -146,13 +146,13 @@ static void interrupt_wq(struct work_struct *work) interrupt_work); uint32_t ih_ring_entry[KFD_MAX_RING_ENTRY_SIZE]; - if (dev->device_info->ih_ring_entry_size > sizeof(ih_ring_entry)) { + if (dev->device_info.ih_ring_entry_size > sizeof(ih_ring_entry)) { dev_err_once(kfd_chardev(), "Ring entry too small\n"); return; } while (dequeue_ih_ring_entry(dev, ih_ring_entry)) - dev->device_info->event_interrupt_class->interrupt_wq(dev, + dev->device_info.event_interrupt_class->interrupt_wq(dev, ih_ring_entry); } @@ -163,7 +163,7 @@ bool interrupt_is_wanted(struct kfd_dev *dev, /* integer and bitwise OR so there is no boolean short-circuiting */ unsigned int wanted = 0; - wanted |= dev->device_info->event_interrupt_class->interrupt_isr(dev, + wanted |= dev->device_info.event_interrupt_class->interrupt_isr(dev, ih_ring_entry, patched_ihre, flag); return wanted != 0; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_iommu.c b/drivers/gpu/drm/amd/amdkfd/kfd_iommu.c index 73f2257acc23..66ad8d0b8f7f 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_iommu.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_iommu.c @@ -89,7 +89,7 @@ int kfd_iommu_device_init(struct kfd_dev *kfd) } pasid_limit = min_t(unsigned int, - (unsigned int)(1 << kfd->device_info->max_pasid_bits), + (unsigned int)(1 << kfd->device_info.max_pasid_bits), iommu_info.max_pasids); if (!kfd_set_pasid_limit(pasid_limit)) { diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c b/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c index 406479a369a9..16f8bc4ca7f6 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_kernel_queue.c @@ -111,7 +111,7 @@ static bool kq_initialize(struct kernel_queue *kq, struct kfd_dev *dev, kq->rptr_kernel = kq->rptr_mem->cpu_ptr; kq->rptr_gpu_addr = kq->rptr_mem->gpu_addr; - retval = kfd_gtt_sa_allocate(dev, dev->device_info->doorbell_size, + retval = kfd_gtt_sa_allocate(dev, dev->device_info.doorbell_size, &kq->wptr_mem); if (retval != 0) @@ -297,7 +297,7 @@ void kq_submit_packet(struct kernel_queue *kq) } pr_debug("\n"); #endif - if (kq->dev->device_info->doorbell_size == 8) { + if (kq->dev->device_info.doorbell_size == 8) { *kq->wptr64_kernel = kq->pending_wptr64; write_kernel_doorbell64(kq->queue->properties.doorbell_ptr, kq->pending_wptr64); @@ -310,7 +310,7 @@ void kq_submit_packet(struct kernel_queue *kq) void kq_rollback_packet(struct kernel_queue *kq) { - if (kq->dev->device_info->doorbell_size == 8) { + if (kq->dev->device_info.doorbell_size == 8) { kq->pending_wptr64 = *kq->wptr64_kernel; kq->pending_wptr = *kq->wptr_kernel % (kq->queue->properties.queue_size / 4); diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c index d84cec0022b1..48c2f2b6e217 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_migrate.c @@ -108,8 +108,8 @@ error_free: * svm_migrate_copy_memory_gart - sdma copy data between ram and vram * * @adev: amdgpu device the sdma ring running - * @src: source page address array - * @dst: destination page address array + * @sys: system DMA pointer to be copied + * @vram: vram destination DMA pointer * @npages: number of pages to copy * @direction: enum MIGRATION_COPY_DIR * @mfence: output, sdma fence to signal after sdma is done diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c index 7b4118915bf6..e2825ad4d699 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_mqd_manager.c @@ -71,7 +71,7 @@ struct kfd_mem_obj *allocate_sdma_mqd(struct kfd_dev *dev, return NULL; offset = (q->sdma_engine_id * - dev->device_info->num_sdma_queues_per_engine + + dev->device_info.num_sdma_queues_per_engine + q->sdma_queue_id) * dev->dqm->mqd_mgrs[KFD_MQD_TYPE_SDMA]->mqd_size; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_vi.c b/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_vi.c index 08442e7d9944..3c0658e32e93 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_vi.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_packet_manager_vi.c @@ -110,8 +110,8 @@ static int pm_runlist_vi(struct packet_manager *pm, uint32_t *buffer, return 0; } -int pm_set_resources_vi(struct packet_manager *pm, uint32_t *buffer, - struct scheduling_resources *res) +static int pm_set_resources_vi(struct packet_manager *pm, uint32_t *buffer, + struct scheduling_resources *res) { struct pm4_mes_set_resources *packet; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h index 7ea528941951..0c3f911e3bf4 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_priv.h +++ b/drivers/gpu/drm/amd/amdkfd/kfd_priv.h @@ -195,7 +195,6 @@ struct kfd_event_interrupt_class { }; struct kfd_device_info { - const char *asic_name; uint32_t gfx_target_version; const struct kfd_event_interrupt_class *event_interrupt_class; unsigned int max_pasid_bits; @@ -231,7 +230,7 @@ struct kfd_vmid_info { struct kfd_dev { struct amdgpu_device *adev; - const struct kfd_device_info *device_info; + struct kfd_device_info device_info; struct pci_dev *pdev; struct drm_device *ddev; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process.c b/drivers/gpu/drm/amd/amdkfd/kfd_process.c index d4c8a6948a9f..f1930ff2c74a 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_process.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_process.c @@ -251,14 +251,13 @@ cleanup: } /** - * @kfd_get_cu_occupancy - Collect number of waves in-flight on this device + * kfd_get_cu_occupancy - Collect number of waves in-flight on this device * by current process. Translates acquired wave count into number of compute units * that are occupied. * - * @atr: Handle of attribute that allows reporting of wave count. The attribute + * @attr: Handle of attribute that allows reporting of wave count. The attribute * handle encapsulates GPU device it is associated with, thereby allowing collection * of waves in flight, etc - * * @buffer: Handle of user provided buffer updated with wave count * * Return: Number of bytes written to user buffer or an error value @@ -1011,7 +1010,7 @@ static void kfd_process_destroy_pdds(struct kfd_process *p) free_pages((unsigned long)pdd->qpd.cwsr_kaddr, get_order(KFD_CWSR_TBA_TMA_SIZE)); - kfree(pdd->qpd.doorbell_bitmap); + bitmap_free(pdd->qpd.doorbell_bitmap); idr_destroy(&pdd->alloc_idr); kfd_free_process_doorbells(pdd->dev, pdd->doorbell_index); @@ -1433,9 +1432,8 @@ static int init_doorbell_bitmap(struct qcm_process_device *qpd, if (!KFD_IS_SOC15(dev)) return 0; - qpd->doorbell_bitmap = - kzalloc(DIV_ROUND_UP(KFD_MAX_NUM_OF_QUEUES_PER_PROCESS, - BITS_PER_BYTE), GFP_KERNEL); + qpd->doorbell_bitmap = bitmap_zalloc(KFD_MAX_NUM_OF_QUEUES_PER_PROCESS, + GFP_KERNEL); if (!qpd->doorbell_bitmap) return -ENOMEM; @@ -1447,9 +1445,9 @@ static int init_doorbell_bitmap(struct qcm_process_device *qpd, for (i = 0; i < KFD_MAX_NUM_OF_QUEUES_PER_PROCESS / 2; i++) { if (i >= range_start && i <= range_end) { - set_bit(i, qpd->doorbell_bitmap); - set_bit(i + KFD_QUEUE_DOORBELL_MIRROR_OFFSET, - qpd->doorbell_bitmap); + __set_bit(i, qpd->doorbell_bitmap); + __set_bit(i + KFD_QUEUE_DOORBELL_MIRROR_OFFSET, + qpd->doorbell_bitmap); } } diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c index 4f8464658daf..5e5c84a8e1ef 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_process_queue_manager.c @@ -135,9 +135,8 @@ void kfd_process_dequeue_from_all_devices(struct kfd_process *p) int pqm_init(struct process_queue_manager *pqm, struct kfd_process *p) { INIT_LIST_HEAD(&pqm->queues); - pqm->queue_slot_bitmap = - kzalloc(DIV_ROUND_UP(KFD_MAX_NUM_OF_QUEUES_PER_PROCESS, - BITS_PER_BYTE), GFP_KERNEL); + pqm->queue_slot_bitmap = bitmap_zalloc(KFD_MAX_NUM_OF_QUEUES_PER_PROCESS, + GFP_KERNEL); if (!pqm->queue_slot_bitmap) return -ENOMEM; pqm->process = p; @@ -159,7 +158,7 @@ void pqm_uninit(struct process_queue_manager *pqm) kfree(pqn); } - kfree(pqm->queue_slot_bitmap); + bitmap_free(pqm->queue_slot_bitmap); pqm->queue_slot_bitmap = NULL; } @@ -220,7 +219,7 @@ int pqm_create_queue(struct process_queue_manager *pqm, * Hence we also check the type as well */ if ((pdd->qpd.is_debug) || (type == KFD_QUEUE_TYPE_DIQ)) - max_queues = dev->device_info->max_no_of_hqd/2; + max_queues = dev->device_info.max_no_of_hqd/2; if (pdd->qpd.queue_count >= max_queues) return -ENOSPC; diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c index 755265f6c53b..7e92dcea4ce8 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c @@ -704,6 +704,61 @@ svm_range_apply_attrs(struct kfd_process *p, struct svm_range *prange, } } +static bool +svm_range_is_same_attrs(struct kfd_process *p, struct svm_range *prange, + uint32_t nattr, struct kfd_ioctl_svm_attribute *attrs) +{ + uint32_t i; + int gpuidx; + + for (i = 0; i < nattr; i++) { + switch (attrs[i].type) { + case KFD_IOCTL_SVM_ATTR_PREFERRED_LOC: + if (prange->preferred_loc != attrs[i].value) + return false; + break; + case KFD_IOCTL_SVM_ATTR_PREFETCH_LOC: + /* Prefetch should always trigger a migration even + * if the value of the attribute didn't change. + */ + return false; + case KFD_IOCTL_SVM_ATTR_ACCESS: + case KFD_IOCTL_SVM_ATTR_ACCESS_IN_PLACE: + case KFD_IOCTL_SVM_ATTR_NO_ACCESS: + gpuidx = kfd_process_gpuidx_from_gpuid(p, + attrs[i].value); + if (attrs[i].type == KFD_IOCTL_SVM_ATTR_NO_ACCESS) { + if (test_bit(gpuidx, prange->bitmap_access) || + test_bit(gpuidx, prange->bitmap_aip)) + return false; + } else if (attrs[i].type == KFD_IOCTL_SVM_ATTR_ACCESS) { + if (!test_bit(gpuidx, prange->bitmap_access)) + return false; + } else { + if (!test_bit(gpuidx, prange->bitmap_aip)) + return false; + } + break; + case KFD_IOCTL_SVM_ATTR_SET_FLAGS: + if ((prange->flags & attrs[i].value) != attrs[i].value) + return false; + break; + case KFD_IOCTL_SVM_ATTR_CLR_FLAGS: + if ((prange->flags & attrs[i].value) != 0) + return false; + break; + case KFD_IOCTL_SVM_ATTR_GRANULARITY: + if (prange->granularity != attrs[i].value) + return false; + break; + default: + WARN_ONCE(1, "svm_range_check_attrs wasn't called?"); + } + } + + return true; +} + /** * svm_range_debug_dump - print all range information from svms * @svms: svm range list header @@ -741,14 +796,6 @@ static void svm_range_debug_dump(struct svm_range_list *svms) } } -static bool -svm_range_is_same_attrs(struct svm_range *old, struct svm_range *new) -{ - return (old->prefetch_loc == new->prefetch_loc && - old->flags == new->flags && - old->granularity == new->granularity); -} - static int svm_range_split_array(void *ppnew, void *ppold, size_t size, uint64_t old_start, uint64_t old_n, @@ -941,7 +988,7 @@ svm_range_split(struct svm_range *prange, uint64_t start, uint64_t last, } static int -svm_range_split_tail(struct svm_range *prange, struct svm_range *new, +svm_range_split_tail(struct svm_range *prange, uint64_t new_last, struct list_head *insert_list) { struct svm_range *tail; @@ -953,7 +1000,7 @@ svm_range_split_tail(struct svm_range *prange, struct svm_range *new, } static int -svm_range_split_head(struct svm_range *prange, struct svm_range *new, +svm_range_split_head(struct svm_range *prange, uint64_t new_start, struct list_head *insert_list) { struct svm_range *head; @@ -1169,7 +1216,6 @@ svm_range_map_to_gpu(struct amdgpu_device *adev, struct amdgpu_vm *vm, unsigned long npages, bool readonly, dma_addr_t *dma_addr, struct amdgpu_device *bo_adev, struct dma_fence **fence) { - struct amdgpu_bo_va bo_va; bool table_freed = false; uint64_t pte_flags; unsigned long last_start; @@ -1182,9 +1228,6 @@ svm_range_map_to_gpu(struct amdgpu_device *adev, struct amdgpu_vm *vm, pr_debug("svms 0x%p [0x%lx 0x%lx] readonly %d\n", prange->svms, last_start, last_start + npages - 1, readonly); - if (prange->svm_bo && prange->ttm_res) - bo_va.is_xgmi = amdgpu_xgmi_same_hive(adev, bo_adev); - for (i = offset; i < offset + npages; i++) { last_domain = dma_addr[i] & SVM_RANGE_VRAM_DOMAIN; dma_addr[i] &= ~SVM_RANGE_VRAM_DOMAIN; @@ -1650,6 +1693,10 @@ out_reschedule: /** * svm_range_evict - evict svm range + * @prange: svm range structure + * @mm: current process mm_struct + * @start: starting process queue number + * @last: last process queue number * * Stop all queues of the process to ensure GPU doesn't access the memory, then * return to let CPU evict the buffer and proceed CPU pagetable update. @@ -1754,46 +1801,49 @@ static struct svm_range *svm_range_clone(struct svm_range *old) } /** - * svm_range_handle_overlap - split overlap ranges - * @svms: svm range list header - * @new: range added with this attributes - * @start: range added start address, in pages - * @last: range last address, in pages - * @update_list: output, the ranges attributes are updated. For set_attr, this - * will do validation and map to GPUs. For unmap, this will be - * removed and unmap from GPUs - * @insert_list: output, the ranges will be inserted into svms, attributes are - * not changes. For set_attr, this will add into svms. - * @remove_list:output, the ranges will be removed from svms - * @left: the remaining range after overlap, For set_attr, this will be added - * as new range. + * svm_range_add - add svm range and handle overlap + * @p: the range add to this process svms + * @start: page size aligned + * @size: page size aligned + * @nattr: number of attributes + * @attrs: array of attributes + * @update_list: output, the ranges need validate and update GPU mapping + * @insert_list: output, the ranges need insert to svms + * @remove_list: output, the ranges are replaced and need remove from svms * - * Total have 5 overlap cases. + * Check if the virtual address range has overlap with any existing ranges, + * split partly overlapping ranges and add new ranges in the gaps. All changes + * should be applied to the range_list and interval tree transactionally. If + * any range split or allocation fails, the entire update fails. Therefore any + * existing overlapping svm_ranges are cloned and the original svm_ranges left + * unchanged. * - * This function handles overlap of an address interval with existing - * struct svm_ranges for applying new attributes. This may require - * splitting existing struct svm_ranges. All changes should be applied to - * the range_list and interval tree transactionally. If any split operation - * fails, the entire update fails. Therefore the existing overlapping - * svm_ranges are cloned and the original svm_ranges left unchanged. If the - * transaction succeeds, the modified clones are added and the originals - * freed. Otherwise the clones are removed and the old svm_ranges remain. + * If the transaction succeeds, the caller can update and insert clones and + * new ranges, then free the originals. * - * Context: The caller must hold svms->lock + * Otherwise the caller can free the clones and new ranges, while the old + * svm_ranges remain unchanged. + * + * Context: Process context, caller must hold svms->lock + * + * Return: + * 0 - OK, otherwise error code */ static int -svm_range_handle_overlap(struct svm_range_list *svms, struct svm_range *new, - unsigned long start, unsigned long last, - struct list_head *update_list, - struct list_head *insert_list, - struct list_head *remove_list, - unsigned long *left) +svm_range_add(struct kfd_process *p, uint64_t start, uint64_t size, + uint32_t nattr, struct kfd_ioctl_svm_attribute *attrs, + struct list_head *update_list, struct list_head *insert_list, + struct list_head *remove_list) { + unsigned long last = start + size - 1UL; + struct svm_range_list *svms = &p->svms; struct interval_tree_node *node; struct svm_range *prange; struct svm_range *tmp; int r = 0; + pr_debug("svms 0x%p [0x%llx 0x%lx]\n", &p->svms, start, last); + INIT_LIST_HEAD(update_list); INIT_LIST_HEAD(insert_list); INIT_LIST_HEAD(remove_list); @@ -1801,18 +1851,24 @@ svm_range_handle_overlap(struct svm_range_list *svms, struct svm_range *new, node = interval_tree_iter_first(&svms->objects, start, last); while (node) { struct interval_tree_node *next; - struct svm_range *old; unsigned long next_start; pr_debug("found overlap node [0x%lx 0x%lx]\n", node->start, node->last); - old = container_of(node, struct svm_range, it_node); + prange = container_of(node, struct svm_range, it_node); next = interval_tree_iter_next(node, start, last); next_start = min(node->last, last) + 1; - if (node->start < start || node->last > last) { - /* node intersects the updated range, clone+split it */ + if (svm_range_is_same_attrs(p, prange, nattr, attrs)) { + /* nothing to do */ + } else if (node->start < start || node->last > last) { + /* node intersects the update range and its attributes + * will change. Clone and split it, apply updates only + * to the overlapping part + */ + struct svm_range *old = prange; + prange = svm_range_clone(old); if (!prange) { r = -ENOMEM; @@ -1821,17 +1877,18 @@ svm_range_handle_overlap(struct svm_range_list *svms, struct svm_range *new, list_add(&old->remove_list, remove_list); list_add(&prange->insert_list, insert_list); + list_add(&prange->update_list, update_list); if (node->start < start) { pr_debug("change old range start\n"); - r = svm_range_split_head(prange, new, start, + r = svm_range_split_head(prange, start, insert_list); if (r) goto out; } if (node->last > last) { pr_debug("change old range last\n"); - r = svm_range_split_tail(prange, new, last, + r = svm_range_split_tail(prange, last, insert_list); if (r) goto out; @@ -1840,16 +1897,12 @@ svm_range_handle_overlap(struct svm_range_list *svms, struct svm_range *new, /* The node is contained within start..last, * just update it */ - prange = old; - } - - if (!svm_range_is_same_attrs(prange, new)) list_add(&prange->update_list, update_list); + } /* insert a new node if needed */ if (node->start > start) { - prange = svm_range_new(prange->svms, start, - node->start - 1); + prange = svm_range_new(svms, start, node->start - 1); if (!prange) { r = -ENOMEM; goto out; @@ -1863,8 +1916,16 @@ svm_range_handle_overlap(struct svm_range_list *svms, struct svm_range *new, start = next_start; } - if (left && start <= last) - *left = last - start + 1; + /* add a final range at the end if needed */ + if (start <= last) { + prange = svm_range_new(svms, start, last); + if (!prange) { + r = -ENOMEM; + goto out; + } + list_add(&prange->insert_list, insert_list); + list_add(&prange->update_list, update_list); + } out: if (r) @@ -1970,7 +2031,7 @@ restart: pr_debug("drain retry fault gpu %d svms %p\n", i, svms); - amdgpu_ih_wait_on_checkpoint_process(pdd->dev->adev, + amdgpu_ih_wait_on_checkpoint_process_ts(pdd->dev->adev, &pdd->dev->adev->irq.ih1); pr_debug("drain retry fault gpu %d svms 0x%p done\n", i, svms); } @@ -2161,6 +2222,9 @@ svm_range_unmap_from_cpu(struct mm_struct *mm, struct svm_range *prange, /** * svm_range_cpu_invalidate_pagetables - interval notifier callback + * @mni: mmu_interval_notifier struct + * @range: mmu_notifier_range struct + * @cur_seq: value to pass to mmu_interval_set_seq() * * If event is MMU_NOTIFY_UNMAP, this is from CPU unmap range, otherwise, it * is from migration, or CPU page invalidation callback. @@ -2884,59 +2948,6 @@ svm_range_is_valid(struct kfd_process *p, uint64_t start, uint64_t size) } /** - * svm_range_add - add svm range and handle overlap - * @p: the range add to this process svms - * @start: page size aligned - * @size: page size aligned - * @nattr: number of attributes - * @attrs: array of attributes - * @update_list: output, the ranges need validate and update GPU mapping - * @insert_list: output, the ranges need insert to svms - * @remove_list: output, the ranges are replaced and need remove from svms - * - * Check if the virtual address range has overlap with the registered ranges, - * split the overlapped range, copy and adjust pages address and vram nodes in - * old and new ranges. - * - * Context: Process context, caller must hold svms->lock - * - * Return: - * 0 - OK, otherwise error code - */ -static int -svm_range_add(struct kfd_process *p, uint64_t start, uint64_t size, - uint32_t nattr, struct kfd_ioctl_svm_attribute *attrs, - struct list_head *update_list, struct list_head *insert_list, - struct list_head *remove_list) -{ - uint64_t last = start + size - 1UL; - struct svm_range_list *svms; - struct svm_range new = {0}; - struct svm_range *prange; - unsigned long left = 0; - int r = 0; - - pr_debug("svms 0x%p [0x%llx 0x%llx]\n", &p->svms, start, last); - - svm_range_apply_attrs(p, &new, nattr, attrs); - - svms = &p->svms; - - r = svm_range_handle_overlap(svms, &new, start, last, update_list, - insert_list, remove_list, &left); - if (r) - return r; - - if (left) { - prange = svm_range_new(svms, last - left + 1, last); - list_add(&prange->insert_list, insert_list); - list_add(&prange->update_list, update_list); - } - - return 0; -} - -/** * svm_range_best_prefetch_location - decide the best prefetch location * @prange: svm range structure * diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c index 2d44b26b6657..948fbb39336e 100644 --- a/drivers/gpu/drm/amd/amdkfd/kfd_topology.c +++ b/drivers/gpu/drm/amd/amdkfd/kfd_topology.c @@ -503,7 +503,7 @@ static ssize_t node_show(struct kobject *kobj, struct attribute *attr, if (dev->gpu) { log_max_watch_addr = - __ilog2_u32(dev->gpu->device_info->num_of_watch_points); + __ilog2_u32(dev->gpu->device_info.num_of_watch_points); if (log_max_watch_addr) { dev->node_props.capability |= @@ -1285,6 +1285,8 @@ int kfd_topology_add_device(struct kfd_dev *gpu) void *crat_image = NULL; size_t image_size = 0; int proximity_domain; + int i; + const char *asic_name = amdgpu_asic_name[gpu->adev->asic_type]; INIT_LIST_HEAD(&temp_topology_device_list); @@ -1370,13 +1372,17 @@ int kfd_topology_add_device(struct kfd_dev *gpu) amdgpu_amdkfd_get_cu_info(dev->gpu->adev, &cu_info); - strncpy(dev->node_props.name, gpu->device_info->asic_name, - KFD_TOPOLOGY_PUBLIC_NAME_SIZE); + for (i = 0; i < KFD_TOPOLOGY_PUBLIC_NAME_SIZE-1; i++) { + dev->node_props.name[i] = __tolower(asic_name[i]); + if (asic_name[i] == '\0') + break; + } + dev->node_props.name[i] = '\0'; dev->node_props.simd_arrays_per_engine = cu_info.num_shader_arrays_per_engine; - dev->node_props.gfx_target_version = gpu->device_info->gfx_target_version; + dev->node_props.gfx_target_version = gpu->device_info.gfx_target_version; dev->node_props.vendor_id = gpu->pdev->vendor; dev->node_props.device_id = gpu->pdev->device; dev->node_props.capability |= @@ -1396,7 +1402,7 @@ int kfd_topology_add_device(struct kfd_dev *gpu) dev->node_props.num_sdma_xgmi_engines = kfd_get_num_xgmi_sdma_engines(gpu); dev->node_props.num_sdma_queues_per_engine = - gpu->device_info->num_sdma_queues_per_engine; + gpu->device_info.num_sdma_queues_per_engine; dev->node_props.num_gws = (dev->gpu->gws && dev->gpu->dqm->sched_policy != KFD_SCHED_POLICY_NO_HWS) ? dev->gpu->adev->gds.gws_size : 0; @@ -1572,7 +1578,7 @@ void kfd_double_confirm_iommu_support(struct kfd_dev *gpu) gpu->use_iommu_v2 = false; - if (!gpu->device_info->needs_iommu_device) + if (!gpu->device_info.needs_iommu_device) return; down_read(&topology_lock); diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c index 116a280d8a20..2f0b14f8f833 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c @@ -624,7 +624,7 @@ static void dm_dcn_vertical_interrupt0_high_irq(void *interrupt_params) #endif /* CONFIG_DRM_AMD_SECURE_DISPLAY */ /** - * dmub_aux_setconfig_reply_callback - Callback for AUX or SET_CONFIG command. + * dmub_aux_setconfig_callback - Callback for AUX or SET_CONFIG command. * @adev: amdgpu_device pointer * @notify: dmub notification structure * @@ -632,7 +632,8 @@ static void dm_dcn_vertical_interrupt0_high_irq(void *interrupt_params) * Copies dmub notification to DM which is to be read by AUX command. * issuing thread and also signals the event to wake up the thread. */ -void dmub_aux_setconfig_callback(struct amdgpu_device *adev, struct dmub_notification *notify) +static void dmub_aux_setconfig_callback(struct amdgpu_device *adev, + struct dmub_notification *notify) { if (adev->dm.dmub_notify) memcpy(adev->dm.dmub_notify, notify, sizeof(struct dmub_notification)); @@ -648,7 +649,8 @@ void dmub_aux_setconfig_callback(struct amdgpu_device *adev, struct dmub_notific * Dmub Hpd interrupt processing callback. Gets displayindex through the * ink index and calls helper to do the processing. */ -void dmub_hpd_callback(struct amdgpu_device *adev, struct dmub_notification *notify) +static void dmub_hpd_callback(struct amdgpu_device *adev, + struct dmub_notification *notify) { struct amdgpu_dm_connector *aconnector; struct amdgpu_dm_connector *hpd_aconnector = NULL; @@ -705,8 +707,10 @@ void dmub_hpd_callback(struct amdgpu_device *adev, struct dmub_notification *not * to dmub interrupt handling thread * Return: true if successfully registered, false if there is existing registration */ -bool register_dmub_notify_callback(struct amdgpu_device *adev, enum dmub_notification_type type, -dmub_notify_interrupt_callback_t callback, bool dmub_int_thread_offload) +static bool register_dmub_notify_callback(struct amdgpu_device *adev, + enum dmub_notification_type type, + dmub_notify_interrupt_callback_t callback, + bool dmub_int_thread_offload) { if (callback != NULL && type < ARRAY_SIZE(adev->dm.dmub_thread_offload)) { adev->dm.dmub_callback[type] = callback; @@ -1050,6 +1054,11 @@ static int dm_dmub_hw_init(struct amdgpu_device *adev) return 0; } + /* Reset DMCUB if it was previously running - before we overwrite its memory. */ + status = dmub_srv_hw_reset(dmub_srv); + if (status != DMUB_STATUS_OK) + DRM_WARN("Error resetting DMUB HW: %d\n", status); + hdr = (const struct dmcub_firmware_header_v1_0 *)dmub_fw->data; fw_inst_const = dmub_fw->data + @@ -1453,6 +1462,13 @@ static int amdgpu_dm_init(struct amdgpu_device *adev) if (amdgpu_dc_feature_mask & DC_EDP_NO_POWER_SEQUENCING) init_data.flags.edp_no_power_sequencing = true; +#ifdef CONFIG_DRM_AMD_DC_DCN + if (amdgpu_dc_feature_mask & DC_DISABLE_LTTPR_DP1_4A) + init_data.flags.allow_lttpr_non_transparent_mode.bits.DP1_4A = true; + if (amdgpu_dc_feature_mask & DC_DISABLE_LTTPR_DP2_0) + init_data.flags.allow_lttpr_non_transparent_mode.bits.DP2_0 = true; +#endif + init_data.flags.power_down_display_on_boot = true; if (check_seamless_boot_capability(adev)) { @@ -2950,13 +2966,12 @@ void amdgpu_dm_update_connector_after_detect( aconnector->edid = (struct edid *)sink->dc_edid.raw_edid; - drm_connector_update_edid_property(connector, - aconnector->edid); if (aconnector->dc_link->aux_mode) drm_dp_cec_set_edid(&aconnector->dm_dp_aux.aux, aconnector->edid); } + drm_connector_update_edid_property(connector, aconnector->edid); amdgpu_dm_update_freesync_caps(connector, aconnector->edid); update_connector_ext_caps(aconnector); } else { @@ -6061,10 +6076,12 @@ static void update_dsc_caps(struct amdgpu_dm_connector *aconnector, if (aconnector->dc_link && (sink->sink_signal == SIGNAL_TYPE_DISPLAY_PORT || sink->sink_signal == SIGNAL_TYPE_EDP)) { - dc_dsc_parse_dsc_dpcd(aconnector->dc_link->ctx->dc, - aconnector->dc_link->dpcd_caps.dsc_caps.dsc_basic_caps.raw, - aconnector->dc_link->dpcd_caps.dsc_caps.dsc_branch_decoder_caps.raw, - dsc_caps); + if (sink->link->dpcd_caps.dongle_type == DISPLAY_DONGLE_NONE || + sink->link->dpcd_caps.dongle_type == DISPLAY_DONGLE_DP_HDMI_CONVERTER) + dc_dsc_parse_dsc_dpcd(aconnector->dc_link->ctx->dc, + aconnector->dc_link->dpcd_caps.dsc_caps.dsc_basic_caps.raw, + aconnector->dc_link->dpcd_caps.dsc_caps.dsc_branch_decoder_caps.raw, + dsc_caps); } } @@ -6134,6 +6151,8 @@ static void apply_dsc_policy_for_stream(struct amdgpu_dm_connector *aconnector, uint32_t link_bandwidth_kbps; uint32_t max_dsc_target_bpp_limit_override = 0; struct dc *dc = sink->ctx->dc; + uint32_t max_supported_bw_in_kbps, timing_bw_in_kbps; + uint32_t dsc_max_supported_bw_in_kbps; link_bandwidth_kbps = dc_link_bandwidth_kbps(aconnector->dc_link, dc_link_get_link_cap(aconnector->dc_link)); @@ -6152,16 +6171,37 @@ static void apply_dsc_policy_for_stream(struct amdgpu_dm_connector *aconnector, apply_dsc_policy_for_edp(aconnector, sink, stream, dsc_caps, max_dsc_target_bpp_limit_override); } else if (aconnector->dc_link && sink->sink_signal == SIGNAL_TYPE_DISPLAY_PORT) { - - if (dc_dsc_compute_config(aconnector->dc_link->ctx->dc->res_pool->dscs[0], + if (sink->link->dpcd_caps.dongle_type == DISPLAY_DONGLE_NONE) { + if (dc_dsc_compute_config(aconnector->dc_link->ctx->dc->res_pool->dscs[0], dsc_caps, aconnector->dc_link->ctx->dc->debug.dsc_min_slice_height_override, max_dsc_target_bpp_limit_override, link_bandwidth_kbps, &stream->timing, &stream->timing.dsc_cfg)) { - stream->timing.flags.DSC = 1; - DRM_DEBUG_DRIVER("%s: [%s] DSC is selected from SST RX\n", __func__, drm_connector->name); + stream->timing.flags.DSC = 1; + DRM_DEBUG_DRIVER("%s: [%s] DSC is selected from SST RX\n", + __func__, drm_connector->name); + } + } else if (sink->link->dpcd_caps.dongle_type == DISPLAY_DONGLE_DP_HDMI_CONVERTER) { + timing_bw_in_kbps = dc_bandwidth_in_kbps_from_timing(&stream->timing); + max_supported_bw_in_kbps = link_bandwidth_kbps; + dsc_max_supported_bw_in_kbps = link_bandwidth_kbps; + + if (timing_bw_in_kbps > max_supported_bw_in_kbps && + max_supported_bw_in_kbps > 0 && + dsc_max_supported_bw_in_kbps > 0) + if (dc_dsc_compute_config(aconnector->dc_link->ctx->dc->res_pool->dscs[0], + dsc_caps, + aconnector->dc_link->ctx->dc->debug.dsc_min_slice_height_override, + max_dsc_target_bpp_limit_override, + dsc_max_supported_bw_in_kbps, + &stream->timing, + &stream->timing.dsc_cfg)) { + stream->timing.flags.DSC = 1; + DRM_DEBUG_DRIVER("%s: [%s] DSC is selected from DP-HDMI PCON\n", + __func__, drm_connector->name); + } } } @@ -8305,15 +8345,8 @@ void amdgpu_dm_connector_init_helper(struct amdgpu_display_manager *dm, break; case DRM_MODE_CONNECTOR_DisplayPort: aconnector->base.polled = DRM_CONNECTOR_POLL_HPD; - if (link->is_dig_mapping_flexible && - link->dc->res_pool->funcs->link_encs_assign) { - link->link_enc = - link_enc_cfg_get_link_enc_used_by_link(link->ctx->dc, link); - if (!link->link_enc) - link->link_enc = - link_enc_cfg_get_next_avail_link_enc(link->ctx->dc); - } - + link->link_enc = dp_get_link_enc(link); + ASSERT(link->link_enc); if (link->link_enc) aconnector->base.ycbcr_420_allowed = link->link_enc->features.dp_ycbcr420_supported ? true : false; @@ -10716,6 +10749,24 @@ static int dm_update_plane_state(struct dc *dc, return ret; } +static void dm_get_oriented_plane_size(struct drm_plane_state *plane_state, + int *src_w, int *src_h) +{ + switch (plane_state->rotation & DRM_MODE_ROTATE_MASK) { + case DRM_MODE_ROTATE_90: + case DRM_MODE_ROTATE_270: + *src_w = plane_state->src_h >> 16; + *src_h = plane_state->src_w >> 16; + break; + case DRM_MODE_ROTATE_0: + case DRM_MODE_ROTATE_180: + default: + *src_w = plane_state->src_w >> 16; + *src_h = plane_state->src_h >> 16; + break; + } +} + static int dm_check_crtc_cursor(struct drm_atomic_state *state, struct drm_crtc *crtc, struct drm_crtc_state *new_crtc_state) @@ -10724,6 +10775,8 @@ static int dm_check_crtc_cursor(struct drm_atomic_state *state, struct drm_plane_state *new_cursor_state, *new_underlying_state; int i; int cursor_scale_w, cursor_scale_h, underlying_scale_w, underlying_scale_h; + int cursor_src_w, cursor_src_h; + int underlying_src_w, underlying_src_h; /* On DCE and DCN there is no dedicated hardware cursor plane. We get a * cursor per pipe but it's going to inherit the scaling and @@ -10735,10 +10788,9 @@ static int dm_check_crtc_cursor(struct drm_atomic_state *state, return 0; } - cursor_scale_w = new_cursor_state->crtc_w * 1000 / - (new_cursor_state->src_w >> 16); - cursor_scale_h = new_cursor_state->crtc_h * 1000 / - (new_cursor_state->src_h >> 16); + dm_get_oriented_plane_size(new_cursor_state, &cursor_src_w, &cursor_src_h); + cursor_scale_w = new_cursor_state->crtc_w * 1000 / cursor_src_w; + cursor_scale_h = new_cursor_state->crtc_h * 1000 / cursor_src_h; for_each_new_plane_in_state_reverse(state, underlying, new_underlying_state, i) { /* Narrow down to non-cursor planes on the same CRTC as the cursor */ @@ -10749,10 +10801,10 @@ static int dm_check_crtc_cursor(struct drm_atomic_state *state, if (!new_underlying_state->fb) continue; - underlying_scale_w = new_underlying_state->crtc_w * 1000 / - (new_underlying_state->src_w >> 16); - underlying_scale_h = new_underlying_state->crtc_h * 1000 / - (new_underlying_state->src_h >> 16); + dm_get_oriented_plane_size(new_underlying_state, + &underlying_src_w, &underlying_src_h); + underlying_scale_w = new_underlying_state->crtc_w * 1000 / underlying_src_w; + underlying_scale_h = new_underlying_state->crtc_h * 1000 / underlying_src_h; if (cursor_scale_w != underlying_scale_w || cursor_scale_h != underlying_scale_h) { @@ -11271,7 +11323,7 @@ static bool dm_edid_parser_send_cea(struct amdgpu_display_manager *dm, sizeof(cmd.edid_cea) - sizeof(cmd.edid_cea.header); input->offset = offset; input->length = length; - input->total_length = total_length; + input->cea_total_length = total_length; memcpy(input->payload, data, length); res = dc_dmub_srv_cmd_with_reply_data(dm->dc->ctx->dmub_srv, &cmd); @@ -11578,8 +11630,10 @@ uint32_t dm_read_reg_func(const struct dc_context *ctx, uint32_t address, return value; } -int amdgpu_dm_set_dmub_async_sync_status(bool is_cmd_aux, struct dc_context *ctx, - uint8_t status_type, uint32_t *operation_result) +static int amdgpu_dm_set_dmub_async_sync_status(bool is_cmd_aux, + struct dc_context *ctx, + uint8_t status_type, + uint32_t *operation_result) { struct amdgpu_device *adev = ctx->driver_context; int return_status = -1; diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h index bb65f41d1a59..c98e402eab0c 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.h @@ -50,9 +50,9 @@ #define AMDGPU_DMUB_NOTIFICATION_MAX 5 -/** +/* * DMUB Async to Sync Mechanism Status - **/ + */ #define DMUB_ASYNC_TO_SYNC_ACCESS_FAIL 1 #define DMUB_ASYNC_TO_SYNC_ACCESS_TIMEOUT 2 #define DMUB_ASYNC_TO_SYNC_ACCESS_SUCCESS 3 diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c index a022e5bb30a5..a71177305bcd 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_color.c @@ -285,8 +285,12 @@ static int __set_input_tf(struct dc_transfer_func *func, } /** + * amdgpu_dm_verify_lut_sizes + * @crtc_state: the DRM CRTC state + * * Verifies that the Degamma and Gamma LUTs attached to the |crtc_state| are of * the expected size. + * * Returns 0 on success. */ int amdgpu_dm_verify_lut_sizes(const struct drm_crtc_state *crtc_state) diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c index 0277685864c5..26719efa5396 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c @@ -824,6 +824,48 @@ static int dmub_fw_state_show(struct seq_file *m, void *data) return seq_write(m, state_base, state_size); } +/* psr_capability_show() - show eDP panel PSR capability + * + * The read function: sink_psr_capability_show + * Shows if sink has PSR capability or not. + * If yes - the PSR version is appended + * + * cat /sys/kernel/debug/dri/0/eDP-X/psr_capability + * + * Expected output: + * "Sink support: no\n" - if panel doesn't support PSR + * "Sink support: yes [0x01]\n" - if panel supports PSR1 + * "Driver support: no\n" - if driver doesn't support PSR + * "Driver support: yes [0x01]\n" - if driver supports PSR1 + */ +static int psr_capability_show(struct seq_file *m, void *data) +{ + struct drm_connector *connector = m->private; + struct amdgpu_dm_connector *aconnector = to_amdgpu_dm_connector(connector); + struct dc_link *link = aconnector->dc_link; + + if (!link) + return -ENODEV; + + if (link->type == dc_connection_none) + return -ENODEV; + + if (!(link->connector_signal & SIGNAL_TYPE_EDP)) + return -ENODEV; + + seq_printf(m, "Sink support: %s", yesno(link->dpcd_caps.psr_caps.psr_version != 0)); + if (link->dpcd_caps.psr_caps.psr_version) + seq_printf(m, " [0x%02x]", link->dpcd_caps.psr_caps.psr_version); + seq_puts(m, "\n"); + + seq_printf(m, "Driver support: %s", yesno(link->psr_settings.psr_feature_enabled)); + if (link->psr_settings.psr_version) + seq_printf(m, " [0x%02x]", link->psr_settings.psr_version); + seq_puts(m, "\n"); + + return 0; +} + /* * Returns the current and maximum output bpc for the connector. * Example usage: cat /sys/kernel/debug/dri/0/DP-1/output_bpc @@ -2467,6 +2509,7 @@ DEFINE_SHOW_ATTRIBUTE(dp_lttpr_status); DEFINE_SHOW_ATTRIBUTE(hdcp_sink_capability); #endif DEFINE_SHOW_ATTRIBUTE(internal_display); +DEFINE_SHOW_ATTRIBUTE(psr_capability); static const struct file_operations dp_dsc_clock_en_debugfs_fops = { .owner = THIS_MODULE, @@ -2712,6 +2755,138 @@ static const struct { {"internal_display", &internal_display_fops} }; +/* + * Returns supported customized link rates by this eDP panel. + * Example usage: cat /sys/kernel/debug/dri/0/eDP-x/ilr_setting + */ +static int edp_ilr_show(struct seq_file *m, void *unused) +{ + struct amdgpu_dm_connector *aconnector = to_amdgpu_dm_connector(m->private); + struct dc_link *link = aconnector->dc_link; + uint8_t supported_link_rates[16]; + uint32_t link_rate_in_khz; + uint32_t entry = 0; + uint8_t dpcd_rev; + + memset(supported_link_rates, 0, sizeof(supported_link_rates)); + dm_helpers_dp_read_dpcd(link->ctx, link, DP_SUPPORTED_LINK_RATES, + supported_link_rates, sizeof(supported_link_rates)); + + dpcd_rev = link->dpcd_caps.dpcd_rev.raw; + + if (dpcd_rev >= DP_DPCD_REV_13 && + (supported_link_rates[entry+1] != 0 || supported_link_rates[entry] != 0)) { + + for (entry = 0; entry < 16; entry += 2) { + link_rate_in_khz = (supported_link_rates[entry+1] * 0x100 + + supported_link_rates[entry]) * 200; + seq_printf(m, "[%d] %d kHz\n", entry/2, link_rate_in_khz); + } + } else { + seq_printf(m, "ILR is not supported by this eDP panel.\n"); + } + + return 0; +} + +/* + * Set supported customized link rate to eDP panel. + * + * echo <lane_count> <link_rate option> > ilr_setting + * + * for example, supported ILR : [0] 1620000 kHz [1] 2160000 kHz [2] 2430000 kHz ... + * echo 4 1 > /sys/kernel/debug/dri/0/eDP-x/ilr_setting + * to set 4 lanes and 2.16 GHz + */ +static ssize_t edp_ilr_write(struct file *f, const char __user *buf, + size_t size, loff_t *pos) +{ + struct amdgpu_dm_connector *connector = file_inode(f)->i_private; + struct dc_link *link = connector->dc_link; + struct amdgpu_device *adev = drm_to_adev(connector->base.dev); + struct dc *dc = (struct dc *)link->dc; + struct dc_link_settings prefer_link_settings; + char *wr_buf = NULL; + const uint32_t wr_buf_size = 40; + /* 0: lane_count; 1: link_rate */ + int max_param_num = 2; + uint8_t param_nums = 0; + long param[2]; + bool valid_input = true; + + if (size == 0) + return -EINVAL; + + wr_buf = kcalloc(wr_buf_size, sizeof(char), GFP_KERNEL); + if (!wr_buf) + return -ENOMEM; + + if (parse_write_buffer_into_params(wr_buf, wr_buf_size, + (long *)param, buf, + max_param_num, + ¶m_nums)) { + kfree(wr_buf); + return -EINVAL; + } + + if (param_nums <= 0) { + kfree(wr_buf); + return -EINVAL; + } + + switch (param[0]) { + case LANE_COUNT_ONE: + case LANE_COUNT_TWO: + case LANE_COUNT_FOUR: + break; + default: + valid_input = false; + break; + } + + if (param[1] >= link->dpcd_caps.edp_supported_link_rates_count) + valid_input = false; + + if (!valid_input) { + kfree(wr_buf); + DRM_DEBUG_DRIVER("Invalid Input value. No HW will be programmed\n"); + prefer_link_settings.use_link_rate_set = false; + dc_link_set_preferred_training_settings(dc, NULL, NULL, link, true); + return size; + } + + /* save user force lane_count, link_rate to preferred settings + * spread spectrum will not be changed + */ + prefer_link_settings.link_spread = link->cur_link_settings.link_spread; + prefer_link_settings.lane_count = param[0]; + prefer_link_settings.use_link_rate_set = true; + prefer_link_settings.link_rate_set = param[1]; + prefer_link_settings.link_rate = link->dpcd_caps.edp_supported_link_rates[param[1]]; + + mutex_lock(&adev->dm.dc_lock); + dc_link_set_preferred_training_settings(dc, &prefer_link_settings, + NULL, link, false); + mutex_unlock(&adev->dm.dc_lock); + + kfree(wr_buf); + return size; +} + +static int edp_ilr_open(struct inode *inode, struct file *file) +{ + return single_open(file, edp_ilr_show, inode->i_private); +} + +static const struct file_operations edp_ilr_debugfs_fops = { + .owner = THIS_MODULE, + .open = edp_ilr_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, + .write = edp_ilr_write +}; + void connector_debugfs_init(struct amdgpu_dm_connector *connector) { int i; @@ -2726,11 +2901,14 @@ void connector_debugfs_init(struct amdgpu_dm_connector *connector) } } if (connector->base.connector_type == DRM_MODE_CONNECTOR_eDP) { + debugfs_create_file_unsafe("psr_capability", 0444, dir, connector, &psr_capability_fops); debugfs_create_file_unsafe("psr_state", 0444, dir, connector, &psr_fops); debugfs_create_file("amdgpu_current_backlight_pwm", 0444, dir, connector, ¤t_backlight_fops); debugfs_create_file("amdgpu_target_backlight_pwm", 0444, dir, connector, &target_backlight_fops); + debugfs_create_file("ilr_setting", 0644, dir, connector, + &edp_ilr_debugfs_fops); } for (i = 0; i < ARRAY_SIZE(connector_debugfs_entries); i++) { @@ -2909,10 +3087,13 @@ static int crc_win_update_set(void *data, u64 val) struct amdgpu_device *adev = drm_to_adev(new_crtc->dev); struct crc_rd_work *crc_rd_wrk = adev->dm.crc_rd_wrk; + if (!crc_rd_wrk) + return 0; + if (val) { spin_lock_irq(&adev_to_drm(adev)->event_lock); spin_lock_irq(&crc_rd_wrk->crc_rd_work_lock); - if (crc_rd_wrk && crc_rd_wrk->crtc) { + if (crc_rd_wrk->crtc) { old_crtc = crc_rd_wrk->crtc; old_acrtc = to_amdgpu_crtc(old_crtc); } @@ -3190,6 +3371,32 @@ static int disable_hpd_get(void *data, u64 *val) DEFINE_DEBUGFS_ATTRIBUTE(disable_hpd_ops, disable_hpd_get, disable_hpd_set, "%llu\n"); +#if defined(CONFIG_DRM_AMD_DC_DCN) +/* + * Temporary w/a to force sst sequence in M42D DP2 mst receiver + * Example usage: echo 1 > /sys/kernel/debug/dri/0/amdgpu_dm_dp_set_mst_en_for_sst + */ +static int dp_force_sst_set(void *data, u64 val) +{ + struct amdgpu_device *adev = data; + + adev->dm.dc->debug.set_mst_en_for_sst = val; + + return 0; +} + +static int dp_force_sst_get(void *data, u64 *val) +{ + struct amdgpu_device *adev = data; + + *val = adev->dm.dc->debug.set_mst_en_for_sst; + + return 0; +} +DEFINE_DEBUGFS_ATTRIBUTE(dp_set_mst_en_for_sst_ops, dp_force_sst_get, + dp_force_sst_set, "%llu\n"); +#endif + /* * Sets the DC visual confirm debug option from the given string. * Example usage: echo 1 > /sys/kernel/debug/dri/0/amdgpu_visual_confirm @@ -3299,6 +3506,10 @@ void dtn_debugfs_init(struct amdgpu_device *adev) adev, &mst_topo_fops); debugfs_create_file("amdgpu_dm_dtn_log", 0644, root, adev, &dtn_log_fops); +#if defined(CONFIG_DRM_AMD_DC_DCN) + debugfs_create_file("amdgpu_dm_dp_set_mst_en_for_sst", 0644, root, adev, + &dp_set_mst_en_for_sst_ops); +#endif debugfs_create_file_unsafe("amdgpu_dm_visual_confirm", 0644, root, adev, &visual_confirm_fops); diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c index 72a2e84645df..29f07c26d080 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_helpers.c @@ -83,16 +83,17 @@ static int amdgpu_dm_patch_edid_caps(struct dc_edid_caps *edid_caps) * void * */ enum dc_edid_status dm_helpers_parse_edid_caps( - struct dc_context *ctx, + struct dc_link *link, const struct dc_edid *edid, struct dc_edid_caps *edid_caps) { + struct amdgpu_dm_connector *aconnector = link->priv; + struct drm_connector *connector = &aconnector->base; struct edid *edid_buf = (struct edid *) edid->raw_edid; struct cea_sad *sads; int sad_count = -1; int sadb_count = -1; int i = 0; - int j = 0; uint8_t *sadb = NULL; enum dc_edid_status result = EDID_OK; @@ -111,23 +112,11 @@ enum dc_edid_status dm_helpers_parse_edid_caps( edid_caps->manufacture_week = edid_buf->mfg_week; edid_caps->manufacture_year = edid_buf->mfg_year; - /* One of the four detailed_timings stores the monitor name. It's - * stored in an array of length 13. */ - for (i = 0; i < 4; i++) { - if (edid_buf->detailed_timings[i].data.other_data.type == 0xfc) { - while (j < 13 && edid_buf->detailed_timings[i].data.other_data.data.str.str[j]) { - if (edid_buf->detailed_timings[i].data.other_data.data.str.str[j] == '\n') - break; - - edid_caps->display_name[j] = - edid_buf->detailed_timings[i].data.other_data.data.str.str[j]; - j++; - } - } - } + drm_edid_get_monitor_name(edid_buf, + edid_caps->display_name, + AUDIO_INFO_DISPLAY_NAME_SIZE_IN_CHARS); - edid_caps->edid_hdmi = drm_detect_hdmi_monitor( - (struct edid *) edid->raw_edid); + edid_caps->edid_hdmi = connector->display_info.is_hdmi; sad_count = drm_edid_to_sad((struct edid *) edid->raw_edid, &sads); if (sad_count <= 0) @@ -585,8 +574,17 @@ bool dm_helpers_dp_write_dsc_enable( } if (stream->signal == SIGNAL_TYPE_DISPLAY_PORT || stream->signal == SIGNAL_TYPE_EDP) { - ret = dm_helpers_dp_write_dpcd(ctx, stream->link, DP_DSC_ENABLE, &enable_dsc, 1); - DC_LOG_DC("Send DSC %s to sst display\n", enable_dsc ? "enable" : "disable"); +#if defined(CONFIG_DRM_AMD_DC_DCN) + if (stream->sink->link->dpcd_caps.dongle_type == DISPLAY_DONGLE_NONE) { +#endif + ret = dm_helpers_dp_write_dpcd(ctx, stream->link, DP_DSC_ENABLE, &enable_dsc, 1); + DC_LOG_DC("Send DSC %s to SST RX\n", enable_dsc ? "enable" : "disable"); +#if defined(CONFIG_DRM_AMD_DC_DCN) + } else if (stream->sink->link->dpcd_caps.dongle_type == DISPLAY_DONGLE_DP_HDMI_CONVERTER) { + ret = dm_helpers_dp_write_dpcd(ctx, stream->link, DP_DSC_ENABLE, &enable_dsc, 1); + DC_LOG_DC("Send DSC %s to DP-HDMI PCON\n", enable_dsc ? "enable" : "disable"); + } +#endif } return (ret > 0); @@ -650,14 +648,8 @@ enum dc_edid_status dm_helpers_read_local_edid( /* We don't need the original edid anymore */ kfree(edid); - /* connector->display_info will be parsed from EDID and saved - * into drm_connector->display_info from edid by call stack - * below: - * drm_parse_ycbcr420_deep_color_info - * drm_parse_hdmi_forum_vsdb - * drm_parse_cea_ext - * drm_add_display_info - * drm_connector_update_edid_property + /* connector->display_info is parsed from EDID and saved + * into drm_connector->display_info * * drm_connector->display_info will be used by amdgpu_dm funcs, * like fill_stream_properties_from_drm_display_mode @@ -665,7 +657,7 @@ enum dc_edid_status dm_helpers_read_local_edid( amdgpu_dm_update_connector_after_detect(aconnector); edid_status = dm_helpers_parse_edid_caps( - ctx, + link, &sink->dc_edid, &sink->edid_caps); diff --git a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.c b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.c index c022e56f9459..c510638b4f99 100644 --- a/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.c +++ b/drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_psr.c @@ -26,6 +26,73 @@ #include "amdgpu_dm_psr.h" #include "dc.h" #include "dm_helpers.h" +#include "amdgpu_dm.h" + +static bool link_get_psr_caps(struct dc_link *link) +{ + uint8_t psr_dpcd_data[EDP_PSR_RECEIVER_CAP_SIZE]; + uint8_t edp_rev_dpcd_data; + + + + if (!dm_helpers_dp_read_dpcd(NULL, link, DP_PSR_SUPPORT, + psr_dpcd_data, sizeof(psr_dpcd_data))) + return false; + + if (!dm_helpers_dp_read_dpcd(NULL, link, DP_EDP_DPCD_REV, + &edp_rev_dpcd_data, sizeof(edp_rev_dpcd_data))) + return false; + + link->dpcd_caps.psr_caps.psr_version = psr_dpcd_data[0]; + link->dpcd_caps.psr_caps.edp_revision = edp_rev_dpcd_data; + +#ifdef CONFIG_DRM_AMD_DC_DCN + if (link->dpcd_caps.psr_caps.psr_version > 0x1) { + uint8_t alpm_dpcd_data; + uint8_t su_granularity_dpcd_data; + + if (!dm_helpers_dp_read_dpcd(NULL, link, DP_RECEIVER_ALPM_CAP, + &alpm_dpcd_data, sizeof(alpm_dpcd_data))) + return false; + + if (!dm_helpers_dp_read_dpcd(NULL, link, DP_PSR2_SU_Y_GRANULARITY, + &su_granularity_dpcd_data, sizeof(su_granularity_dpcd_data))) + return false; + + link->dpcd_caps.psr_caps.y_coordinate_required = psr_dpcd_data[1] & DP_PSR2_SU_Y_COORDINATE_REQUIRED; + link->dpcd_caps.psr_caps.su_granularity_required = psr_dpcd_data[1] & DP_PSR2_SU_GRANULARITY_REQUIRED; + + link->dpcd_caps.psr_caps.alpm_cap = alpm_dpcd_data & DP_ALPM_CAP; + link->dpcd_caps.psr_caps.standby_support = alpm_dpcd_data & (1 << 1); + + link->dpcd_caps.psr_caps.su_y_granularity = su_granularity_dpcd_data; + } +#endif + return true; +} + +#ifdef CONFIG_DRM_AMD_DC_DCN +static bool link_supports_psrsu(struct dc_link *link) +{ + struct dc *dc = link->ctx->dc; + + if (!dc->caps.dmcub_support) + return false; + + if (dc->ctx->dce_version < DCN_VERSION_3_1) + return false; + + if (!link->dpcd_caps.psr_caps.alpm_cap || + !link->dpcd_caps.psr_caps.y_coordinate_required) + return false; + + if (link->dpcd_caps.psr_caps.su_granularity_required && + !link->dpcd_caps.psr_caps.su_y_granularity) + return false; + + return true; +} +#endif /* * amdgpu_dm_set_psr_caps() - set link psr capabilities @@ -34,26 +101,34 @@ */ void amdgpu_dm_set_psr_caps(struct dc_link *link) { - uint8_t dpcd_data[EDP_PSR_RECEIVER_CAP_SIZE]; - if (!(link->connector_signal & SIGNAL_TYPE_EDP)) return; + if (link->type == dc_connection_none) return; - if (dm_helpers_dp_read_dpcd(NULL, link, DP_PSR_SUPPORT, - dpcd_data, sizeof(dpcd_data))) { - link->dpcd_caps.psr_caps.psr_version = dpcd_data[0]; - - if (dpcd_data[0] == 0) { - link->psr_settings.psr_version = DC_PSR_VERSION_UNSUPPORTED; - link->psr_settings.psr_feature_enabled = false; - } else { + + if (!link_get_psr_caps(link)) { + DRM_ERROR("amdgpu: Failed to read PSR Caps!\n"); + return; + } + + if (link->dpcd_caps.psr_caps.psr_version == 0) { + link->psr_settings.psr_version = DC_PSR_VERSION_UNSUPPORTED; + link->psr_settings.psr_feature_enabled = false; + + } else { +#ifdef CONFIG_DRM_AMD_DC_DCN + if (link_supports_psrsu(link)) + link->psr_settings.psr_version = DC_PSR_VERSION_SU_1; + else +#endif link->psr_settings.psr_version = DC_PSR_VERSION_1; - link->psr_settings.psr_feature_enabled = true; - } - DRM_INFO("PSR support:%d\n", link->psr_settings.psr_feature_enabled); + link->psr_settings.psr_feature_enabled = true; } + + DRM_INFO("PSR support:%d\n", link->psr_settings.psr_feature_enabled); + } /* diff --git a/drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c b/drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c index c8b0a2f05b4d..ec19678a0702 100644 --- a/drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c +++ b/drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c @@ -739,7 +739,9 @@ static void hack_bounding_box(struct dcn_bw_internal_vars *v, hack_force_pipe_split(v, context->streams[0]->timing.pix_clk_100hz); } -unsigned int get_highest_allowed_voltage_level(uint32_t chip_family, uint32_t hw_internal_rev, uint32_t pci_revision_id) +static unsigned int get_highest_allowed_voltage_level(uint32_t chip_family, + uint32_t hw_internal_rev, + uint32_t pci_revision_id) { /* for low power RV2 variants, the highest voltage level we want is 0 */ if ((chip_family == FAMILY_RV) && diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn10/rv1_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn10/rv1_clk_mgr.c index 76ec8ec92efd..60761ff3cbf1 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn10/rv1_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn10/rv1_clk_mgr.c @@ -34,7 +34,7 @@ #include "rv1_clk_mgr_vbios_smu.h" #include "rv1_clk_mgr_clk.h" -void rv1_init_clocks(struct clk_mgr *clk_mgr) +static void rv1_init_clocks(struct clk_mgr *clk_mgr) { memset(&(clk_mgr->clks), 0, sizeof(struct dc_clocks)); } diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn10/rv1_clk_mgr_vbios_smu.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn10/rv1_clk_mgr_vbios_smu.c index fe18bb9e19aa..06bab24d8e27 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn10/rv1_clk_mgr_vbios_smu.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn10/rv1_clk_mgr_vbios_smu.c @@ -28,6 +28,8 @@ #include "reg_helper.h" #include <linux/delay.h> +#include "rv1_clk_mgr_vbios_smu.h" + #define MAX_INSTANCE 5 #define MAX_SEGMENT 5 diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn20/dcn20_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn20/dcn20_clk_mgr.c index 2108bff49d4e..9f35f2e8f971 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn20/dcn20_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn20/dcn20_clk_mgr.c @@ -409,7 +409,7 @@ void dcn2_init_clocks(struct clk_mgr *clk_mgr) clk_mgr->clks.prev_p_state_change_support = true; } -void dcn2_enable_pme_wa(struct clk_mgr *clk_mgr_base) +static void dcn2_enable_pme_wa(struct clk_mgr *clk_mgr_base) { struct clk_mgr_internal *clk_mgr = TO_CLK_MGR_INTERNAL(clk_mgr_base); struct pp_smu_funcs_nv *pp_smu = NULL; diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn201/dcn201_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn201/dcn201_clk_mgr.c index db9950244c7b..fbdd0a92d146 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn201/dcn201_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn201/dcn201_clk_mgr.c @@ -74,42 +74,6 @@ static const struct clk_mgr_mask clk_mgr_mask = { CLK_COMMON_MASK_SH_LIST_DCN201_BASE(_MASK) }; -void dcn201_update_clocks_vbios(struct clk_mgr *clk_mgr, - struct dc_state *context, - bool safe_to_lower) -{ - struct dc_clocks *new_clocks = &context->bw_ctx.bw.dcn.clk; - - bool update_dppclk = false; - bool update_dispclk = false; - - if (should_set_clock(safe_to_lower, new_clocks->dppclk_khz, clk_mgr->clks.dppclk_khz)) { - clk_mgr->clks.dppclk_khz = new_clocks->dppclk_khz; - update_dppclk = true; - } - - if (should_set_clock(safe_to_lower, new_clocks->dispclk_khz, clk_mgr->clks.dispclk_khz)) { - clk_mgr->clks.dispclk_khz = new_clocks->dispclk_khz; - update_dispclk = true; - } - - if (update_dppclk || update_dispclk) { - struct bp_set_dce_clock_parameters dce_clk_params; - struct dc_bios *bp = clk_mgr->ctx->dc_bios; - - if (update_dispclk) { - memset(&dce_clk_params, 0, sizeof(dce_clk_params)); - dce_clk_params.target_clock_frequency = new_clocks->dispclk_khz; - dce_clk_params.pll_id = CLOCK_SOURCE_ID_DFS; - dce_clk_params.clock_type = DCECLOCK_TYPE_DISPLAY_CLOCK; - bp->funcs->set_dce_clock(bp, &dce_clk_params); - } - /* currently there is no DCECLOCK_TYPE_DPPCLK type defined in VBIOS interface. - * vbios program DPPCLK to the same DispCLK limitation - */ - } -} - static void dcn201_init_clocks(struct clk_mgr *clk_mgr) { memset(&(clk_mgr->clks), 0, sizeof(struct dc_clocks)); @@ -126,10 +90,8 @@ static void dcn201_update_clocks(struct clk_mgr *clk_mgr_base, struct clk_mgr_internal *clk_mgr = TO_CLK_MGR_INTERNAL(clk_mgr_base); struct dc_clocks *new_clocks = &context->bw_ctx.bw.dcn.clk; struct dc *dc = clk_mgr_base->ctx->dc; - int display_count; bool update_dppclk = false; bool update_dispclk = false; - bool enter_display_off = false; bool dpp_clock_lowered = false; bool force_reset = false; bool p_state_change_support; @@ -145,10 +107,7 @@ static void dcn201_update_clocks(struct clk_mgr *clk_mgr_base, dcn2_read_clocks_from_hw_dentist(clk_mgr_base); } - display_count = clk_mgr_helper_get_active_display_cnt(dc, context); - - if (display_count == 0) - enter_display_off = true; + clk_mgr_helper_get_active_display_cnt(dc, context); if (should_set_clock(safe_to_lower, new_clocks->phyclk_khz, clk_mgr_base->clks.phyclk_khz)) clk_mgr_base->clks.phyclk_khz = new_clocks->phyclk_khz; diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.c index ac2d4c4f04e4..fbda42313bfe 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr.c @@ -56,9 +56,7 @@ /* TODO: evaluate how to lower or disable all dcn clocks in screen off case */ -int rn_get_active_display_cnt_wa( - struct dc *dc, - struct dc_state *context) +static int rn_get_active_display_cnt_wa(struct dc *dc, struct dc_state *context) { int i, display_count; bool tmds_present = false; @@ -77,7 +75,8 @@ int rn_get_active_display_cnt_wa( const struct dc_link *link = dc->links[i]; /* abusing the fact that the dig and phy are coupled to see if the phy is enabled */ - if (link->link_enc->funcs->is_dig_enabled(link->link_enc)) + if (link->link_enc->funcs->is_dig_enabled && + link->link_enc->funcs->is_dig_enabled(link->link_enc)) display_count++; } @@ -88,7 +87,7 @@ int rn_get_active_display_cnt_wa( return display_count; } -void rn_set_low_power_state(struct clk_mgr *clk_mgr_base) +static void rn_set_low_power_state(struct clk_mgr *clk_mgr_base) { struct clk_mgr_internal *clk_mgr = TO_CLK_MGR_INTERNAL(clk_mgr_base); @@ -122,7 +121,7 @@ static void rn_update_clocks_update_dpp_dto(struct clk_mgr_internal *clk_mgr, } -void rn_update_clocks(struct clk_mgr *clk_mgr_base, +static void rn_update_clocks(struct clk_mgr *clk_mgr_base, struct dc_state *context, bool safe_to_lower) { @@ -437,25 +436,14 @@ static void rn_dump_clk_registers(struct clk_state_registers_and_bypass *regs_an } } -/* This function produce translated logical clk state values*/ -void rn_get_clk_states(struct clk_mgr *clk_mgr_base, struct clk_states *s) -{ - struct clk_state_registers_and_bypass sb = { 0 }; - struct clk_log_info log_info = { 0 }; - - rn_dump_clk_registers(&sb, clk_mgr_base, &log_info); - - s->dprefclk_khz = sb.dprefclk * 1000; -} - -void rn_enable_pme_wa(struct clk_mgr *clk_mgr_base) +static void rn_enable_pme_wa(struct clk_mgr *clk_mgr_base) { struct clk_mgr_internal *clk_mgr = TO_CLK_MGR_INTERNAL(clk_mgr_base); rn_vbios_smu_enable_pme_wa(clk_mgr); } -void rn_init_clocks(struct clk_mgr *clk_mgr) +static void rn_init_clocks(struct clk_mgr *clk_mgr) { memset(&(clk_mgr->clks), 0, sizeof(struct dc_clocks)); // Assumption is that boot state always supports pstate diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr_vbios_smu.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr_vbios_smu.c index 9f7eed6688c4..8161a6ae410d 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr_vbios_smu.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn21/rn_clk_mgr_vbios_smu.c @@ -33,6 +33,8 @@ #include "mp/mp_12_0_0_offset.h" #include "mp/mp_12_0_0_sh_mask.h" +#include "rn_clk_mgr_vbios_smu.h" + #define REG(reg_name) \ (MP0_BASE.instance[0].segment[mm ## reg_name ## _BASE_IDX] + mm ## reg_name) @@ -86,7 +88,9 @@ static uint32_t rn_smu_wait_for_response(struct clk_mgr_internal *clk_mgr, unsig } -int rn_vbios_smu_send_msg_with_param(struct clk_mgr_internal *clk_mgr, unsigned int msg_id, unsigned int param) +static int rn_vbios_smu_send_msg_with_param(struct clk_mgr_internal *clk_mgr, + unsigned int msg_id, + unsigned int param) { uint32_t result; diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn30/dcn30_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn30/dcn30_clk_mgr.c index 1861a147a7fa..f977f29907df 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn30/dcn30_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn30/dcn30_clk_mgr.c @@ -252,6 +252,7 @@ static void dcn3_update_clocks(struct clk_mgr *clk_mgr_base, bool update_dispclk = false; bool enter_display_off = false; bool dpp_clock_lowered = false; + bool update_pstate_unsupported_clk = false; struct dmcu *dmcu = clk_mgr_base->ctx->dc->res_pool->dmcu; bool force_reset = false; bool update_uclk = false; @@ -299,13 +300,28 @@ static void dcn3_update_clocks(struct clk_mgr *clk_mgr_base, clk_mgr_base->clks.prev_p_state_change_support = clk_mgr_base->clks.p_state_change_support; total_plane_count = clk_mgr_helper_get_active_plane_cnt(dc, context); p_state_change_support = new_clocks->p_state_change_support || (total_plane_count == 0); - if (should_update_pstate_support(safe_to_lower, p_state_change_support, clk_mgr_base->clks.p_state_change_support)) { + + // invalidate the current P-State forced min in certain dc_mode_softmax situations + if (dc->clk_mgr->dc_mode_softmax_enabled && safe_to_lower && !p_state_change_support) { + if ((new_clocks->dramclk_khz <= dc->clk_mgr->bw_params->dc_mode_softmax_memclk * 1000) != + (clk_mgr_base->clks.dramclk_khz <= dc->clk_mgr->bw_params->dc_mode_softmax_memclk * 1000)) + update_pstate_unsupported_clk = true; + } + + if (should_update_pstate_support(safe_to_lower, p_state_change_support, clk_mgr_base->clks.p_state_change_support) || + update_pstate_unsupported_clk) { clk_mgr_base->clks.p_state_change_support = p_state_change_support; /* to disable P-State switching, set UCLK min = max */ - if (!clk_mgr_base->clks.p_state_change_support) - dcn30_smu_set_hard_min_by_freq(clk_mgr, PPCLK_UCLK, + if (!clk_mgr_base->clks.p_state_change_support) { + if (dc->clk_mgr->dc_mode_softmax_enabled && + new_clocks->dramclk_khz <= dc->clk_mgr->bw_params->dc_mode_softmax_memclk * 1000) + dcn30_smu_set_hard_min_by_freq(clk_mgr, PPCLK_UCLK, + dc->clk_mgr->bw_params->dc_mode_softmax_memclk); + else + dcn30_smu_set_hard_min_by_freq(clk_mgr, PPCLK_UCLK, clk_mgr_base->bw_params->clk_table.entries[clk_mgr_base->bw_params->clk_table.num_entries - 1].memclk_mhz); + } } /* Always update saved value, even if new value not set due to P-State switching unsupported */ @@ -421,6 +437,24 @@ static void dcn3_set_hard_max_memclk(struct clk_mgr *clk_mgr_base) clk_mgr_base->bw_params->clk_table.entries[clk_mgr_base->bw_params->clk_table.num_entries - 1].memclk_mhz); } +static void dcn3_set_max_memclk(struct clk_mgr *clk_mgr_base, unsigned int memclk_mhz) +{ + struct clk_mgr_internal *clk_mgr = TO_CLK_MGR_INTERNAL(clk_mgr_base); + + if (!clk_mgr->smu_present) + return; + + dcn30_smu_set_hard_max_by_freq(clk_mgr, PPCLK_UCLK, memclk_mhz); +} +static void dcn3_set_min_memclk(struct clk_mgr *clk_mgr_base, unsigned int memclk_mhz) +{ + struct clk_mgr_internal *clk_mgr = TO_CLK_MGR_INTERNAL(clk_mgr_base); + + if (!clk_mgr->smu_present) + return; + dcn30_smu_set_hard_min_by_freq(clk_mgr, PPCLK_UCLK, memclk_mhz); +} + /* Get current memclk states, update bounding box */ static void dcn3_get_memclk_states_from_smu(struct clk_mgr *clk_mgr_base) { @@ -436,6 +470,8 @@ static void dcn3_get_memclk_states_from_smu(struct clk_mgr *clk_mgr_base) &num_levels); clk_mgr_base->bw_params->clk_table.num_entries = num_levels ? num_levels : 1; + clk_mgr_base->bw_params->dc_mode_softmax_memclk = dcn30_smu_get_dc_mode_max_dpm_freq(clk_mgr, PPCLK_UCLK); + /* Refresh bounding box */ clk_mgr_base->ctx->dc->res_pool->funcs->update_bw_bounding_box( clk_mgr->base.ctx->dc, clk_mgr_base->bw_params); @@ -505,6 +541,8 @@ static struct clk_mgr_funcs dcn3_funcs = { .notify_wm_ranges = dcn3_notify_wm_ranges, .set_hard_min_memclk = dcn3_set_hard_min_memclk, .set_hard_max_memclk = dcn3_set_hard_max_memclk, + .set_max_memclk = dcn3_set_max_memclk, + .set_min_memclk = dcn3_set_min_memclk, .get_memclk_states_from_smu = dcn3_get_memclk_states_from_smu, .are_clock_states_equal = dcn3_are_clock_states_equal, .enable_pme_wa = dcn3_enable_pme_wa, diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn301/dcn301_smu.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn301/dcn301_smu.c index 6ea642615854..d9920d91838d 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn301/dcn301_smu.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn301/dcn301_smu.c @@ -88,9 +88,9 @@ static uint32_t dcn301_smu_wait_for_response(struct clk_mgr_internal *clk_mgr, u return res_val; } -int dcn301_smu_send_msg_with_param( - struct clk_mgr_internal *clk_mgr, - unsigned int msg_id, unsigned int param) +static int dcn301_smu_send_msg_with_param(struct clk_mgr_internal *clk_mgr, + unsigned int msg_id, + unsigned int param) { uint32_t result; diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn301/vg_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn301/vg_clk_mgr.c index 3eee32faa208..8f78e62b28b7 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn301/vg_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn301/vg_clk_mgr.c @@ -89,9 +89,9 @@ static int vg_get_active_display_cnt_wa( return display_count; } -void vg_update_clocks(struct clk_mgr *clk_mgr_base, - struct dc_state *context, - bool safe_to_lower) +static void vg_update_clocks(struct clk_mgr *clk_mgr_base, + struct dc_state *context, + bool safe_to_lower) { struct clk_mgr_internal *clk_mgr = TO_CLK_MGR_INTERNAL(clk_mgr_base); struct dc_clocks *new_clocks = &context->bw_ctx.bw.dcn.clk; @@ -367,18 +367,6 @@ static void vg_dump_clk_registers(struct clk_state_registers_and_bypass *regs_an } } -/* This function produce translated logical clk state values*/ -void vg_get_clk_states(struct clk_mgr *clk_mgr_base, struct clk_states *s) -{ - - struct clk_state_registers_and_bypass sb = { 0 }; - struct clk_log_info log_info = { 0 }; - - vg_dump_clk_registers(&sb, clk_mgr_base, &log_info); - - s->dprefclk_khz = sb.dprefclk * 1000; -} - static void vg_enable_pme_wa(struct clk_mgr *clk_mgr_base) { struct clk_mgr_internal *clk_mgr = TO_CLK_MGR_INTERNAL(clk_mgr_base); @@ -386,7 +374,7 @@ static void vg_enable_pme_wa(struct clk_mgr *clk_mgr_base) dcn301_smu_enable_pme_wa(clk_mgr); } -void vg_init_clocks(struct clk_mgr *clk_mgr) +static void vg_init_clocks(struct clk_mgr *clk_mgr) { memset(&(clk_mgr->clks), 0, sizeof(struct dc_clocks)); // Assumption is that boot state always supports pstate diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_clk_mgr.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_clk_mgr.c index a13ff1783b9b..412cc6a716f7 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_clk_mgr.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_clk_mgr.c @@ -540,10 +540,9 @@ static unsigned int find_clk_for_voltage( return clock; } -void dcn31_clk_mgr_helper_populate_bw_params( - struct clk_mgr_internal *clk_mgr, - struct integrated_info *bios_info, - const DpmClocks_t *clock_table) +static void dcn31_clk_mgr_helper_populate_bw_params(struct clk_mgr_internal *clk_mgr, + struct integrated_info *bios_info, + const DpmClocks_t *clock_table) { int i, j; struct clk_bw_params *bw_params = clk_mgr->base.bw_params; diff --git a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_smu.c b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_smu.c index 8c2b77eb9459..b7ace235a2d5 100644 --- a/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_smu.c +++ b/drivers/gpu/drm/amd/display/dc/clk_mgr/dcn31/dcn31_smu.c @@ -95,9 +95,9 @@ static uint32_t dcn31_smu_wait_for_response(struct clk_mgr_internal *clk_mgr, un return res_val; } -int dcn31_smu_send_msg_with_param( - struct clk_mgr_internal *clk_mgr, - unsigned int msg_id, unsigned int param) +static int dcn31_smu_send_msg_with_param(struct clk_mgr_internal *clk_mgr, + unsigned int msg_id, + unsigned int param) { uint32_t result; diff --git a/drivers/gpu/drm/amd/display/dc/core/dc.c b/drivers/gpu/drm/amd/display/dc/core/dc.c index 17b7408d84b7..c250f6de5136 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc.c @@ -809,7 +809,7 @@ void dc_stream_set_static_screen_params(struct dc *dc, static void dc_destruct(struct dc *dc) { // reset link encoder assignment table on destruct - if (dc->res_pool->funcs->link_encs_assign) + if (dc->res_pool && dc->res_pool->funcs->link_encs_assign) link_enc_cfg_init(dc, dc->current_state); if (dc->current_state) { @@ -1422,22 +1422,29 @@ static void program_timing_sync( status->timing_sync_info.master = false; } - /* remove any other unblanked pipes as they have already been synced */ - for (j = j + 1; j < group_size; j++) { - bool is_blanked; - if (pipe_set[j]->stream_res.opp->funcs->dpg_is_blanked) - is_blanked = - pipe_set[j]->stream_res.opp->funcs->dpg_is_blanked(pipe_set[j]->stream_res.opp); - else - is_blanked = - pipe_set[j]->stream_res.tg->funcs->is_blanked(pipe_set[j]->stream_res.tg); - if (!is_blanked) { - group_size--; - pipe_set[j] = pipe_set[group_size]; - j--; + /* remove any other pipes that are already been synced */ + if (dc->config.use_pipe_ctx_sync_logic) { + /* check pipe's syncd to decide which pipe to be removed */ + for (j = 1; j < group_size; j++) { + if (pipe_set[j]->pipe_idx_syncd == pipe_set[0]->pipe_idx_syncd) { + group_size--; + pipe_set[j] = pipe_set[group_size]; + j--; + } else + /* link slave pipe's syncd with master pipe */ + pipe_set[j]->pipe_idx_syncd = pipe_set[0]->pipe_idx_syncd; } - } + } else { + /* remove any other pipes by checking valid plane */ + for (j = j + 1; j < group_size; j++) { + if (pipe_set[j]->plane_state) { + group_size--; + pipe_set[j] = pipe_set[group_size]; + j--; + } + } + } if (group_size > 1) { if (sync_type == TIMING_SYNCHRONIZABLE) { @@ -2994,12 +3001,12 @@ static void commit_planes_for_stream(struct dc *dc, #ifdef CONFIG_DRM_AMD_DC_DCN if (dc->debug.validate_dml_output) { for (i = 0; i < dc->res_pool->pipe_count; i++) { - struct pipe_ctx cur_pipe = context->res_ctx.pipe_ctx[i]; - if (cur_pipe.stream == NULL) + struct pipe_ctx *cur_pipe = &context->res_ctx.pipe_ctx[i]; + if (cur_pipe->stream == NULL) continue; - cur_pipe.plane_res.hubp->funcs->validate_dml_output( - cur_pipe.plane_res.hubp, dc->ctx, + cur_pipe->plane_res.hubp->funcs->validate_dml_output( + cur_pipe->plane_res.hubp, dc->ctx, &context->res_ctx.pipe_ctx[i].rq_regs, &context->res_ctx.pipe_ctx[i].dlg_regs, &context->res_ctx.pipe_ctx[i].ttu_regs); @@ -3441,7 +3448,7 @@ struct dc_sink *dc_link_add_remote_sink( goto fail_add_sink; edid_status = dm_helpers_parse_edid_caps( - link->ctx, + link, &dc_sink->dc_edid, &dc_sink->edid_caps); @@ -3598,6 +3605,98 @@ void dc_lock_memory_clock_frequency(struct dc *dc) core_link_enable_stream(dc->current_state, &dc->current_state->res_ctx.pipe_ctx[i]); } +static void blank_and_force_memclk(struct dc *dc, bool apply, unsigned int memclk_mhz) +{ + struct dc_state *context = dc->current_state; + struct hubp *hubp; + struct pipe_ctx *pipe; + int i; + + for (i = 0; i < dc->res_pool->pipe_count; i++) { + pipe = &context->res_ctx.pipe_ctx[i]; + + if (pipe->stream != NULL) { + dc->hwss.disable_pixel_data(dc, pipe, true); + + // wait for double buffer + pipe->stream_res.tg->funcs->wait_for_state(pipe->stream_res.tg, CRTC_STATE_VACTIVE); + pipe->stream_res.tg->funcs->wait_for_state(pipe->stream_res.tg, CRTC_STATE_VBLANK); + pipe->stream_res.tg->funcs->wait_for_state(pipe->stream_res.tg, CRTC_STATE_VACTIVE); + + hubp = pipe->plane_res.hubp; + hubp->funcs->set_blank_regs(hubp, true); + } + } + + dc->clk_mgr->funcs->set_max_memclk(dc->clk_mgr, memclk_mhz); + dc->clk_mgr->funcs->set_min_memclk(dc->clk_mgr, memclk_mhz); + + for (i = 0; i < dc->res_pool->pipe_count; i++) { + pipe = &context->res_ctx.pipe_ctx[i]; + + if (pipe->stream != NULL) { + dc->hwss.disable_pixel_data(dc, pipe, false); + + hubp = pipe->plane_res.hubp; + hubp->funcs->set_blank_regs(hubp, false); + } + } +} + + +/** + * dc_enable_dcmode_clk_limit() - lower clocks in dc (battery) mode + * @dc: pointer to dc of the dm calling this + * @enable: True = transition to DC mode, false = transition back to AC mode + * + * Some SoCs define additional clock limits when in DC mode, DM should + * invoke this function when the platform undergoes a power source transition + * so DC can apply/unapply the limit. This interface may be disruptive to + * the onscreen content. + * + * Context: Triggered by OS through DM interface, or manually by escape calls. + * Need to hold a dclock when doing so. + * + * Return: none (void function) + * + */ +void dc_enable_dcmode_clk_limit(struct dc *dc, bool enable) +{ + uint32_t hw_internal_rev = dc->ctx->asic_id.hw_internal_rev; + unsigned int softMax, maxDPM, funcMin; + bool p_state_change_support; + + if (!ASICREV_IS_BEIGE_GOBY_P(hw_internal_rev)) + return; + + softMax = dc->clk_mgr->bw_params->dc_mode_softmax_memclk; + maxDPM = dc->clk_mgr->bw_params->clk_table.entries[dc->clk_mgr->bw_params->clk_table.num_entries - 1].memclk_mhz; + funcMin = (dc->clk_mgr->clks.dramclk_khz + 999) / 1000; + p_state_change_support = dc->clk_mgr->clks.p_state_change_support; + + if (enable && !dc->clk_mgr->dc_mode_softmax_enabled) { + if (p_state_change_support) { + if (funcMin <= softMax) + dc->clk_mgr->funcs->set_max_memclk(dc->clk_mgr, softMax); + // else: No-Op + } else { + if (funcMin <= softMax) + blank_and_force_memclk(dc, true, softMax); + // else: No-Op + } + } else if (!enable && dc->clk_mgr->dc_mode_softmax_enabled) { + if (p_state_change_support) { + if (funcMin <= softMax) + dc->clk_mgr->funcs->set_max_memclk(dc->clk_mgr, maxDPM); + // else: No-Op + } else { + if (funcMin <= softMax) + blank_and_force_memclk(dc, true, maxDPM); + // else: No-Op + } + } + dc->clk_mgr->dc_mode_softmax_enabled = enable; +} bool dc_is_plane_eligible_for_idle_optimizations(struct dc *dc, struct dc_plane_state *plane, struct dc_cursor_attributes *cursor_attr) { diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link.c b/drivers/gpu/drm/amd/display/dc/core/dc_link.c index faa0bc308fc8..3d75f56a939c 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_link.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_link.c @@ -2766,8 +2766,23 @@ static bool dp_active_dongle_validate_timing( return false; } +#if defined(CONFIG_DRM_AMD_DC_DCN) + if (dongle_caps->dp_hdmi_frl_max_link_bw_in_kbps > 0) { // DP to HDMI FRL converter + struct dc_crtc_timing outputTiming = *timing; + + if (timing->flags.DSC && !timing->dsc_cfg.is_frl) + /* DP input has DSC, HDMI FRL output doesn't have DSC, remove DSC from output timing */ + outputTiming.flags.DSC = 0; + if (dc_bandwidth_in_kbps_from_timing(&outputTiming) > dongle_caps->dp_hdmi_frl_max_link_bw_in_kbps) + return false; + } else { // DP to HDMI TMDS converter + if (get_timing_pixel_clock_100hz(timing) > (dongle_caps->dp_hdmi_max_pixel_clk_in_khz * 10)) + return false; + } +#else if (get_timing_pixel_clock_100hz(timing) > (dongle_caps->dp_hdmi_max_pixel_clk_in_khz * 10)) return false; +#endif #if defined(CONFIG_DRM_AMD_DC_DCN) } @@ -3401,7 +3416,8 @@ static void dc_log_vcp_x_y(const struct dc_link *link, struct fixed31_32 avg_tim /* * Payload allocation/deallocation for SST introduced in DP2.0 */ -enum dc_status dc_link_update_sst_payload(struct pipe_ctx *pipe_ctx, bool allocate) +static enum dc_status dc_link_update_sst_payload(struct pipe_ctx *pipe_ctx, + bool allocate) { struct dc_stream_state *stream = pipe_ctx->stream; struct dc_link *link = stream->link; diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c b/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c index 297553074bfd..8a35370da867 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_link_dp.c @@ -100,6 +100,7 @@ static const struct dp_lt_fallback_entry dp_lt_fallbacks[] = { #endif static bool decide_fallback_link_setting( + struct dc_link *link, struct dc_link_settings initial_link_settings, struct dc_link_settings *current_link_setting, enum link_training_result training_result); @@ -398,6 +399,223 @@ static uint8_t get_dpcd_link_rate(const struct dc_link_settings *link_settings) } #endif +static void vendor_specific_lttpr_wa_one_start(struct dc_link *link) +{ + const uint8_t vendor_lttpr_write_data[4] = {0x1, 0x50, 0x63, 0xff}; + const uint8_t offset = dp_convert_to_count( + link->dpcd_caps.lttpr_caps.phy_repeater_cnt); + uint32_t vendor_lttpr_write_address = 0xF004F; + + if (offset != 0xFF) + vendor_lttpr_write_address += + ((DP_REPEATER_CONFIGURATION_AND_STATUS_SIZE) * (offset - 1)); + + /* W/A for certain LTTPR to reset their lane settings, part one of two */ + core_link_write_dpcd( + link, + vendor_lttpr_write_address, + &vendor_lttpr_write_data[0], + sizeof(vendor_lttpr_write_data)); +} + +static void vendor_specific_lttpr_wa_one_end( + struct dc_link *link, + uint8_t retry_count) +{ + const uint8_t vendor_lttpr_write_data[4] = {0x1, 0x50, 0x63, 0x0}; + const uint8_t offset = dp_convert_to_count( + link->dpcd_caps.lttpr_caps.phy_repeater_cnt); + uint32_t vendor_lttpr_write_address = 0xF004F; + + if (!retry_count) { + if (offset != 0xFF) + vendor_lttpr_write_address += + ((DP_REPEATER_CONFIGURATION_AND_STATUS_SIZE) * (offset - 1)); + + /* W/A for certain LTTPR to reset their lane settings, part two of two */ + core_link_write_dpcd( + link, + vendor_lttpr_write_address, + &vendor_lttpr_write_data[0], + sizeof(vendor_lttpr_write_data)); + } +} + +static void vendor_specific_lttpr_wa_one_two( + struct dc_link *link, + const uint8_t rate) +{ + if (link->apply_vendor_specific_lttpr_link_rate_wa) { + uint8_t toggle_rate = 0x0; + + if (rate == 0x6) + toggle_rate = 0xA; + else + toggle_rate = 0x6; + + if (link->vendor_specific_lttpr_link_rate_wa == rate) { + /* W/A for certain LTTPR to reset internal state for link training */ + core_link_write_dpcd( + link, + DP_LINK_BW_SET, + &toggle_rate, + 1); + } + + /* Store the last attempted link rate for this link */ + link->vendor_specific_lttpr_link_rate_wa = rate; + } +} + +static void vendor_specific_lttpr_wa_three( + struct dc_link *link, + union lane_adjust dpcd_lane_adjust[LANE_COUNT_DP_MAX]) +{ + const uint8_t vendor_lttpr_write_data_vs[3] = {0x0, 0x53, 0x63}; + const uint8_t vendor_lttpr_write_data_pe[3] = {0x0, 0x54, 0x63}; + const uint8_t offset = dp_convert_to_count( + link->dpcd_caps.lttpr_caps.phy_repeater_cnt); + uint32_t vendor_lttpr_write_address = 0xF004F; + uint32_t vendor_lttpr_read_address = 0xF0053; + uint8_t dprx_vs = 0; + uint8_t dprx_pe = 0; + uint8_t lane; + + if (offset != 0xFF) { + vendor_lttpr_write_address += + ((DP_REPEATER_CONFIGURATION_AND_STATUS_SIZE) * (offset - 1)); + vendor_lttpr_read_address += + ((DP_REPEATER_CONFIGURATION_AND_STATUS_SIZE) * (offset - 1)); + } + + /* W/A to read lane settings requested by DPRX */ + core_link_write_dpcd( + link, + vendor_lttpr_write_address, + &vendor_lttpr_write_data_vs[0], + sizeof(vendor_lttpr_write_data_vs)); + core_link_read_dpcd( + link, + vendor_lttpr_read_address, + &dprx_vs, + 1); + core_link_write_dpcd( + link, + vendor_lttpr_write_address, + &vendor_lttpr_write_data_pe[0], + sizeof(vendor_lttpr_write_data_pe)); + core_link_read_dpcd( + link, + vendor_lttpr_read_address, + &dprx_pe, + 1); + + for (lane = 0; lane < LANE_COUNT_DP_MAX; lane++) { + dpcd_lane_adjust[lane].bits.VOLTAGE_SWING_LANE = (dprx_vs >> (2 * lane)) & 0x3; + dpcd_lane_adjust[lane].bits.PRE_EMPHASIS_LANE = (dprx_pe >> (2 * lane)) & 0x3; + } +} + +static void vendor_specific_lttpr_wa_three_dpcd( + struct dc_link *link, + union dpcd_training_lane dpcd_lane_adjust[LANE_COUNT_DP_MAX]) +{ + union lane_adjust lane_adjust[LANE_COUNT_DP_MAX]; + uint8_t lane = 0; + + vendor_specific_lttpr_wa_three(link, lane_adjust); + + for (lane = 0; lane < LANE_COUNT_DP_MAX; lane++) { + dpcd_lane_adjust[lane].bits.VOLTAGE_SWING_SET = lane_adjust[lane].bits.VOLTAGE_SWING_LANE; + dpcd_lane_adjust[lane].bits.PRE_EMPHASIS_SET = lane_adjust[lane].bits.PRE_EMPHASIS_LANE; + } +} + +static void vendor_specific_lttpr_wa_four( + struct dc_link *link, + bool apply_wa) +{ + const uint8_t vendor_lttpr_write_data_one[4] = {0x1, 0x55, 0x63, 0x8}; + const uint8_t vendor_lttpr_write_data_two[4] = {0x1, 0x55, 0x63, 0x0}; + const uint8_t offset = dp_convert_to_count( + link->dpcd_caps.lttpr_caps.phy_repeater_cnt); + uint32_t vendor_lttpr_write_address = 0xF004F; +#if defined(CONFIG_DRM_AMD_DC_DP2_0) + uint8_t sink_status = 0; + uint8_t i; +#endif + + if (offset != 0xFF) + vendor_lttpr_write_address += + ((DP_REPEATER_CONFIGURATION_AND_STATUS_SIZE) * (offset - 1)); + + /* W/A to pass through DPCD write of TPS=0 to DPRX */ + if (apply_wa) { + core_link_write_dpcd( + link, + vendor_lttpr_write_address, + &vendor_lttpr_write_data_one[0], + sizeof(vendor_lttpr_write_data_one)); + } + + /* clear training pattern set */ + dpcd_set_training_pattern(link, DP_TRAINING_PATTERN_VIDEOIDLE); + + if (apply_wa) { + core_link_write_dpcd( + link, + vendor_lttpr_write_address, + &vendor_lttpr_write_data_two[0], + sizeof(vendor_lttpr_write_data_two)); + } + +#if defined(CONFIG_DRM_AMD_DC_DP2_0) + /* poll for intra-hop disable */ + for (i = 0; i < 10; i++) { + if ((core_link_read_dpcd(link, DP_SINK_STATUS, &sink_status, 1) == DC_OK) && + (sink_status & DP_INTRA_HOP_AUX_REPLY_INDICATION) == 0) + break; + udelay(1000); + } +#endif +} + +static void vendor_specific_lttpr_wa_five( + struct dc_link *link, + const union dpcd_training_lane dpcd_lane_adjust[LANE_COUNT_DP_MAX], + uint8_t lane_count) +{ + const uint32_t vendor_lttpr_write_address = 0xF004F; + const uint8_t vendor_lttpr_write_data_reset[4] = {0x1, 0x50, 0x63, 0xFF}; + uint8_t vendor_lttpr_write_data_vs[4] = {0x1, 0x51, 0x63, 0x0}; + uint8_t vendor_lttpr_write_data_pe[4] = {0x1, 0x52, 0x63, 0x0}; + uint8_t lane = 0; + + for (lane = 0; lane < lane_count; lane++) { + vendor_lttpr_write_data_vs[3] |= + dpcd_lane_adjust[lane].bits.VOLTAGE_SWING_SET << (2 * lane); + vendor_lttpr_write_data_pe[3] |= + dpcd_lane_adjust[lane].bits.PRE_EMPHASIS_SET << (2 * lane); + } + + /* Force LTTPR to output desired VS and PE */ + core_link_write_dpcd( + link, + vendor_lttpr_write_address, + &vendor_lttpr_write_data_reset[0], + sizeof(vendor_lttpr_write_data_reset)); + core_link_write_dpcd( + link, + vendor_lttpr_write_address, + &vendor_lttpr_write_data_vs[0], + sizeof(vendor_lttpr_write_data_vs)); + core_link_write_dpcd( + link, + vendor_lttpr_write_address, + &vendor_lttpr_write_data_pe[0], + sizeof(vendor_lttpr_write_data_pe)); +} + enum dc_status dpcd_set_link_settings( struct dc_link *link, const struct link_training_settings *lt_settings) @@ -452,6 +670,15 @@ enum dc_status dpcd_set_link_settings( #else rate = (uint8_t) (lt_settings->link_settings.link_rate); #endif + if (link->dc->debug.apply_vendor_specific_lttpr_wa && + (link->chip_caps & EXT_DISPLAY_PATH_CAPS__DP_FIXED_VS_EN) && + link->lttpr_mode == LTTPR_MODE_TRANSPARENT) + vendor_specific_lttpr_wa_one_start(link); + + if (link->dc->debug.apply_vendor_specific_lttpr_wa && + (link->chip_caps & EXT_DISPLAY_PATH_CAPS__DP_FIXED_VS_EN)) + vendor_specific_lttpr_wa_one_two(link, rate); + status = core_link_write_dpcd(link, DP_LINK_BW_SET, &rate, 1); } @@ -1211,6 +1438,12 @@ static enum link_training_result perform_channel_equalization_sequence( dp_translate_training_aux_read_interval( link->dpcd_caps.lttpr_caps.aux_rd_interval[offset - 1]); + if (link->dc->debug.apply_vendor_specific_lttpr_wa && + (link->chip_caps & EXT_DISPLAY_PATH_CAPS__DP_FIXED_VS_EN) && + link->lttpr_mode == LTTPR_MODE_TRANSPARENT) { + wait_time_microsec = 16000; + } + dp_wait_for_training_aux_rd_interval( link, wait_time_microsec); @@ -1314,6 +1547,11 @@ static enum link_training_result perform_clock_recovery_sequence( if (link->lttpr_mode == LTTPR_MODE_NON_TRANSPARENT) wait_time_microsec = TRAINING_AUX_RD_INTERVAL; + if (link->dc->debug.apply_vendor_specific_lttpr_wa && + (link->chip_caps & EXT_DISPLAY_PATH_CAPS__DP_FIXED_VS_EN)) { + wait_time_microsec = 16000; + } + dp_wait_for_training_aux_rd_interval( link, wait_time_microsec); @@ -1329,6 +1567,13 @@ static enum link_training_result perform_clock_recovery_sequence( dpcd_lane_adjust, offset); + if (link->dc->debug.apply_vendor_specific_lttpr_wa && + (link->chip_caps & EXT_DISPLAY_PATH_CAPS__DP_FIXED_VS_EN) && + link->lttpr_mode == LTTPR_MODE_TRANSPARENT) { + vendor_specific_lttpr_wa_one_end(link, retry_count); + vendor_specific_lttpr_wa_three(link, dpcd_lane_adjust); + } + /* 5. check CR done*/ if (dp_is_cr_done(lane_count, dpcd_lane_status)) return LINK_TRAINING_SUCCESS; @@ -2203,7 +2448,14 @@ enum link_training_result dc_link_dp_perform_link_training( <_settings); /* reset previous training states */ - dpcd_exit_training_mode(link); + if (link->dc->debug.apply_vendor_specific_lttpr_wa && + (link->chip_caps & EXT_DISPLAY_PATH_CAPS__DP_FIXED_VS_EN) && + link->lttpr_mode == LTTPR_MODE_TRANSPARENT) { + link->apply_vendor_specific_lttpr_link_rate_wa = true; + vendor_specific_lttpr_wa_four(link, true); + } else { + dpcd_exit_training_mode(link); + } /* configure link prior to entering training mode */ dpcd_configure_lttpr_mode(link, <_settings); @@ -2223,8 +2475,17 @@ enum link_training_result dc_link_dp_perform_link_training( else ASSERT(0); - /* exit training mode and switch to video idle */ - dpcd_exit_training_mode(link); + /* exit training mode */ + if (link->dc->debug.apply_vendor_specific_lttpr_wa && + (link->chip_caps & EXT_DISPLAY_PATH_CAPS__DP_FIXED_VS_EN) && + link->lttpr_mode == LTTPR_MODE_TRANSPARENT) { + link->apply_vendor_specific_lttpr_link_rate_wa = false; + vendor_specific_lttpr_wa_four(link, (status != LINK_TRAINING_SUCCESS)); + } else { + dpcd_exit_training_mode(link); + } + + /* switch to video idle */ if ((status == LINK_TRAINING_SUCCESS) || !skip_video_pattern) status = dp_transition_to_video_idle(link, <_settings, @@ -2349,7 +2610,7 @@ bool perform_link_training_with_retries( uint32_t req_bw; uint32_t link_bw; - decide_fallback_link_setting(*link_setting, ¤t_setting, status); + decide_fallback_link_setting(link, *link_setting, ¤t_setting, status); /* Fail link training if reduced link bandwidth no longer meets * stream requirements. */ @@ -2864,7 +3125,7 @@ bool dp_verify_link_cap( * based on the actual mode we're driving */ dp_disable_link_phy(link, link->connector_signal); - } while (!success && decide_fallback_link_setting( + } while (!success && decide_fallback_link_setting(link, initial_link_settings, cur, status)); /* Link Training failed for all Link Settings @@ -3116,6 +3377,7 @@ static bool decide_fallback_link_setting_max_bw_policy( * and no further fallback could be done */ static bool decide_fallback_link_setting( + struct dc_link *link, struct dc_link_settings initial_link_settings, struct dc_link_settings *current_link_setting, enum link_training_result training_result) @@ -3123,7 +3385,8 @@ static bool decide_fallback_link_setting( if (!current_link_setting) return false; #if defined(CONFIG_DRM_AMD_DC_DCN) - if (dp_get_link_encoding_format(&initial_link_settings) == DP_128b_132b_ENCODING) + if (dp_get_link_encoding_format(&initial_link_settings) == DP_128b_132b_ENCODING || + link->dc->debug.force_dp2_lt_fallback_method) return decide_fallback_link_setting_max_bw_policy(&initial_link_settings, current_link_setting); #endif @@ -3581,7 +3844,6 @@ static bool handle_hpd_irq_psr_sink(struct dc_link *link) &psr_configuration.raw, sizeof(psr_configuration.raw)); - if (psr_configuration.bits.ENABLE) { unsigned char dpcdbuf[3] = {0}; union psr_error_status psr_error_status; @@ -3613,10 +3875,12 @@ static bool handle_hpd_irq_psr_sink(struct dc_link *link) sizeof(psr_error_status.raw)); /* PSR error, disable and re-enable PSR */ - allow_active = false; - dc_link_set_psr_allow_active(link, &allow_active, true, false, NULL); - allow_active = true; - dc_link_set_psr_allow_active(link, &allow_active, true, false, NULL); + if (link->psr_settings.psr_allow_active) { + allow_active = false; + dc_link_set_psr_allow_active(link, &allow_active, true, false, NULL); + allow_active = true; + dc_link_set_psr_allow_active(link, &allow_active, true, false, NULL); + } return true; } else if (psr_sink_psr_status.bits.SINK_SELF_REFRESH_STATUS == @@ -3694,6 +3958,13 @@ static void dp_test_send_phy_test_pattern(struct dc_link *link) &dpcd_lane_adjustment[0].raw, sizeof(dpcd_lane_adjustment)); + if (link->dc->debug.apply_vendor_specific_lttpr_wa && + (link->chip_caps & EXT_DISPLAY_PATH_CAPS__DP_FIXED_VS_EN) && + link->lttpr_mode == LTTPR_MODE_TRANSPARENT) + vendor_specific_lttpr_wa_three_dpcd( + link, + link_training_settings.dpcd_lane_settings); + /*get post cursor 2 parameters * For DP 1.1a or eariler, this DPCD register's value is 0 * For DP 1.2 or later: @@ -4313,6 +4584,56 @@ static int translate_dpcd_max_bpc(enum dpcd_downstream_port_max_bpc bpc) return -1; } +#if defined(CONFIG_DRM_AMD_DC_DCN) +uint32_t dc_link_bw_kbps_from_raw_frl_link_rate_data(uint8_t bw) +{ + switch (bw) { + case 0b001: + return 9000000; + case 0b010: + return 18000000; + case 0b011: + return 24000000; + case 0b100: + return 32000000; + case 0b101: + return 40000000; + case 0b110: + return 48000000; + } + + return 0; +} + +/** + * Return PCON's post FRL link training supported BW if its non-zero, otherwise return max_supported_frl_bw. + */ +static uint32_t intersect_frl_link_bw_support( + const uint32_t max_supported_frl_bw_in_kbps, + const union hdmi_encoded_link_bw hdmi_encoded_link_bw) +{ + uint32_t supported_bw_in_kbps = max_supported_frl_bw_in_kbps; + + // HDMI_ENCODED_LINK_BW bits are only valid if HDMI Link Configuration bit is 1 (FRL mode) + if (hdmi_encoded_link_bw.bits.FRL_MODE) { + if (hdmi_encoded_link_bw.bits.BW_48Gbps) + supported_bw_in_kbps = 48000000; + else if (hdmi_encoded_link_bw.bits.BW_40Gbps) + supported_bw_in_kbps = 40000000; + else if (hdmi_encoded_link_bw.bits.BW_32Gbps) + supported_bw_in_kbps = 32000000; + else if (hdmi_encoded_link_bw.bits.BW_24Gbps) + supported_bw_in_kbps = 24000000; + else if (hdmi_encoded_link_bw.bits.BW_18Gbps) + supported_bw_in_kbps = 18000000; + else if (hdmi_encoded_link_bw.bits.BW_9Gbps) + supported_bw_in_kbps = 9000000; + } + + return supported_bw_in_kbps; +} +#endif + static void read_dp_device_vendor_id(struct dc_link *link) { struct dp_device_vendor_id dp_id; @@ -4424,6 +4745,27 @@ static void get_active_converter_info( translate_dpcd_max_bpc( hdmi_color_caps.bits.MAX_BITS_PER_COLOR_COMPONENT); +#if defined(CONFIG_DRM_AMD_DC_DCN) + if (link->dc->caps.hdmi_frl_pcon_support) { + union hdmi_encoded_link_bw hdmi_encoded_link_bw; + + link->dpcd_caps.dongle_caps.dp_hdmi_frl_max_link_bw_in_kbps = + dc_link_bw_kbps_from_raw_frl_link_rate_data( + hdmi_color_caps.bits.MAX_ENCODED_LINK_BW_SUPPORT); + + // Intersect reported max link bw support with the supported link rate post FRL link training + if (core_link_read_dpcd(link, DP_PCON_HDMI_POST_FRL_STATUS, + &hdmi_encoded_link_bw.raw, sizeof(hdmi_encoded_link_bw)) == DC_OK) { + link->dpcd_caps.dongle_caps.dp_hdmi_frl_max_link_bw_in_kbps = intersect_frl_link_bw_support( + link->dpcd_caps.dongle_caps.dp_hdmi_frl_max_link_bw_in_kbps, + hdmi_encoded_link_bw); + } + + if (link->dpcd_caps.dongle_caps.dp_hdmi_frl_max_link_bw_in_kbps > 0) + link->dpcd_caps.dongle_caps.extendedCapValid = true; + } +#endif + if (link->dpcd_caps.dongle_caps.dp_hdmi_max_pixel_clk_in_khz != 0) link->dpcd_caps.dongle_caps.extendedCapValid = true; } @@ -5400,8 +5742,18 @@ bool dc_link_dp_set_test_pattern( if (is_dp_phy_pattern(test_pattern)) { /* Set DPCD Lane Settings before running test pattern */ if (p_link_settings != NULL) { - dp_set_hw_lane_settings(link, p_link_settings, DPRX); - dpcd_set_lane_settings(link, p_link_settings, DPRX); + if (link->dc->debug.apply_vendor_specific_lttpr_wa && + (link->chip_caps & EXT_DISPLAY_PATH_CAPS__DP_FIXED_VS_EN) && + link->lttpr_mode == LTTPR_MODE_TRANSPARENT) { + dpcd_set_lane_settings(link, p_link_settings, DPRX); + vendor_specific_lttpr_wa_five( + link, + p_link_settings->dpcd_lane_settings, + p_link_settings->link_settings.lane_count); + } else { + dp_set_hw_lane_settings(link, p_link_settings, DPRX); + dpcd_set_lane_settings(link, p_link_settings, DPRX); + } } /* Blank stream if running test pattern */ @@ -5825,6 +6177,23 @@ void dp_set_fec_enable(struct dc_link *link, bool enable) } } +struct link_encoder *dp_get_link_enc(struct dc_link *link) +{ + struct link_encoder *link_enc; + + link_enc = link->link_enc; + if (link->is_dig_mapping_flexible && + link->dc->res_pool->funcs->link_encs_assign) { + link_enc = link_enc_cfg_get_link_enc_used_by_link(link->ctx->dc, + link); + if (!link->link_enc) + link_enc = link_enc_cfg_get_next_avail_link_enc( + link->ctx->dc); + } + + return link_enc; +} + void dpcd_set_source_specific_data(struct dc_link *link) { if (!link->dc->vendor_signature.is_valid) { @@ -6289,3 +6658,14 @@ bool is_dp_128b_132b_signal(struct pipe_ctx *pipe_ctx) dc_is_dp_signal(pipe_ctx->stream->signal)); } #endif + +void edp_panel_backlight_power_on(struct dc_link *link) +{ + if (link->connector_signal != SIGNAL_TYPE_EDP) + return; + + link->dc->hwss.edp_power_control(link, true); + link->dc->hwss.edp_wait_for_hpd_ready(link, true); + if (link->dc->hwss.edp_backlight_control) + link->dc->hwss.edp_backlight_control(link, true); +} diff --git a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c index ce8f7f4fa2b7..8b6b035bfa9c 100644 --- a/drivers/gpu/drm/amd/display/dc/core/dc_resource.c +++ b/drivers/gpu/drm/amd/display/dc/core/dc_resource.c @@ -734,6 +734,10 @@ static void calculate_split_count_and_index(struct pipe_ctx *pipe_ctx, int *spli (*split_idx)++; split_pipe = split_pipe->top_pipe; } + + /* MPO window on right side of ODM split */ + if (split_pipe && split_pipe->prev_odm_pipe && !pipe_ctx->prev_odm_pipe) + (*split_idx)++; } else { /*Get odm split index*/ struct pipe_ctx *split_pipe = pipe_ctx->prev_odm_pipe; @@ -780,7 +784,11 @@ static void calculate_recout(struct pipe_ctx *pipe_ctx) /* * Only the leftmost ODM pipe should be offset by a nonzero distance */ - if (!pipe_ctx->prev_odm_pipe || split_idx == split_count) { + if (pipe_ctx->top_pipe && pipe_ctx->top_pipe->prev_odm_pipe && !pipe_ctx->prev_odm_pipe) { + /* MPO window on right side of ODM split */ + data->recout.x = stream->dst.x + (surf_clip.x - stream->dst.width/2) * + stream->dst.width / stream->src.width; + } else if (!pipe_ctx->prev_odm_pipe || split_idx == split_count) { data->recout.x = stream->dst.x; if (stream->src.x < surf_clip.x) data->recout.x += (surf_clip.x - stream->src.x) * stream->dst.width @@ -978,6 +986,8 @@ static void calculate_inits_and_viewports(struct pipe_ctx *pipe_ctx) * stream->dst.height / stream->src.height; if (pipe_ctx->prev_odm_pipe && split_idx) ro_lb = data->h_active * split_idx - recout_full_x; + else if (pipe_ctx->top_pipe && pipe_ctx->top_pipe->prev_odm_pipe) + ro_lb = data->h_active * split_idx - recout_full_x + data->recout.x; else ro_lb = data->recout.x - recout_full_x; ro_tb = data->recout.y - recout_full_y; @@ -1076,6 +1086,9 @@ bool resource_build_scaling_params(struct pipe_ctx *pipe_ctx) timing->v_border_top + timing->v_border_bottom; if (pipe_ctx->next_odm_pipe || pipe_ctx->prev_odm_pipe) pipe_ctx->plane_res.scl_data.h_active /= get_num_odm_splits(pipe_ctx) + 1; + /* ODM + windows MPO, where window is on either right or left ODM half */ + else if (pipe_ctx->top_pipe && (pipe_ctx->top_pipe->next_odm_pipe || pipe_ctx->top_pipe->prev_odm_pipe)) + pipe_ctx->plane_res.scl_data.h_active /= get_num_odm_splits(pipe_ctx->top_pipe) + 1; /* depends on h_active */ calculate_recout(pipe_ctx); @@ -1084,11 +1097,6 @@ bool resource_build_scaling_params(struct pipe_ctx *pipe_ctx) /* depends on scaling ratios and recout, does not calculate offset yet */ calculate_viewport_size(pipe_ctx); - /* Stopgap for validation of ODM + MPO on one side of screen case */ - if (pipe_ctx->plane_res.scl_data.viewport.height < 1 || - pipe_ctx->plane_res.scl_data.viewport.width < 1) - return false; - /* * LB calculations depend on vp size, h/v_active and scaling ratios * Setting line buffer pixel depth to 24bpp yields banding @@ -1437,23 +1445,54 @@ bool dc_add_plane_to_context( if (head_pipe != free_pipe) { tail_pipe = resource_get_tail_pipe(&context->res_ctx, head_pipe); ASSERT(tail_pipe); - free_pipe->stream_res.tg = tail_pipe->stream_res.tg; - free_pipe->stream_res.abm = tail_pipe->stream_res.abm; - free_pipe->stream_res.opp = tail_pipe->stream_res.opp; - free_pipe->stream_res.stream_enc = tail_pipe->stream_res.stream_enc; - free_pipe->stream_res.audio = tail_pipe->stream_res.audio; - free_pipe->clock_source = tail_pipe->clock_source; - free_pipe->top_pipe = tail_pipe; - tail_pipe->bottom_pipe = free_pipe; - if (!free_pipe->next_odm_pipe && tail_pipe->next_odm_pipe && tail_pipe->next_odm_pipe->bottom_pipe) { - free_pipe->next_odm_pipe = tail_pipe->next_odm_pipe->bottom_pipe; - tail_pipe->next_odm_pipe->bottom_pipe->prev_odm_pipe = free_pipe; - } - if (!free_pipe->prev_odm_pipe && tail_pipe->prev_odm_pipe && tail_pipe->prev_odm_pipe->bottom_pipe) { - free_pipe->prev_odm_pipe = tail_pipe->prev_odm_pipe->bottom_pipe; - tail_pipe->prev_odm_pipe->bottom_pipe->next_odm_pipe = free_pipe; + + /* ODM + window MPO, where MPO window is on right half only */ + if (free_pipe->plane_state && + (free_pipe->plane_state->clip_rect.x >= free_pipe->stream->src.width/2) && + tail_pipe->next_odm_pipe) { + free_pipe->stream_res.tg = tail_pipe->next_odm_pipe->stream_res.tg; + free_pipe->stream_res.abm = tail_pipe->next_odm_pipe->stream_res.abm; + free_pipe->stream_res.opp = tail_pipe->next_odm_pipe->stream_res.opp; + free_pipe->stream_res.stream_enc = tail_pipe->next_odm_pipe->stream_res.stream_enc; + free_pipe->stream_res.audio = tail_pipe->next_odm_pipe->stream_res.audio; + free_pipe->clock_source = tail_pipe->next_odm_pipe->clock_source; + + free_pipe->top_pipe = tail_pipe->next_odm_pipe; + tail_pipe->next_odm_pipe->bottom_pipe = free_pipe; + } else { + free_pipe->stream_res.tg = tail_pipe->stream_res.tg; + free_pipe->stream_res.abm = tail_pipe->stream_res.abm; + free_pipe->stream_res.opp = tail_pipe->stream_res.opp; + free_pipe->stream_res.stream_enc = tail_pipe->stream_res.stream_enc; + free_pipe->stream_res.audio = tail_pipe->stream_res.audio; + free_pipe->clock_source = tail_pipe->clock_source; + + free_pipe->top_pipe = tail_pipe; + tail_pipe->bottom_pipe = free_pipe; + + if (!free_pipe->next_odm_pipe && tail_pipe->next_odm_pipe && tail_pipe->next_odm_pipe->bottom_pipe) { + free_pipe->next_odm_pipe = tail_pipe->next_odm_pipe->bottom_pipe; + tail_pipe->next_odm_pipe->bottom_pipe->prev_odm_pipe = free_pipe; + } + if (!free_pipe->prev_odm_pipe && tail_pipe->prev_odm_pipe && tail_pipe->prev_odm_pipe->bottom_pipe) { + free_pipe->prev_odm_pipe = tail_pipe->prev_odm_pipe->bottom_pipe; + tail_pipe->prev_odm_pipe->bottom_pipe->next_odm_pipe = free_pipe; + } } } + + /* ODM + window MPO, where MPO window is on left half only */ + if (free_pipe->plane_state && + (free_pipe->plane_state->clip_rect.x + free_pipe->plane_state->clip_rect.width <= + free_pipe->stream->src.x + free_pipe->stream->src.width/2)) { + break; + } + /* ODM + window MPO, where MPO window is on right half only */ + if (free_pipe->plane_state && + (free_pipe->plane_state->clip_rect.x >= free_pipe->stream->src.width/2)) { + break; + } + head_pipe = head_pipe->next_odm_pipe; } /* assign new surfaces*/ @@ -3122,3 +3161,57 @@ struct hpo_dp_link_encoder *resource_get_unused_hpo_dp_link_encoder( return enc; } #endif + +void reset_syncd_pipes_from_disabled_pipes(struct dc *dc, + struct dc_state *context) +{ + int i, j; + struct pipe_ctx *pipe_ctx_old, *pipe_ctx, *pipe_ctx_syncd; + + /* If pipe backend is reset, need to reset pipe syncd status */ + for (i = 0; i < dc->res_pool->pipe_count; i++) { + pipe_ctx_old = &dc->current_state->res_ctx.pipe_ctx[i]; + pipe_ctx = &context->res_ctx.pipe_ctx[i]; + + if (!pipe_ctx_old->stream) + continue; + + if (pipe_ctx_old->top_pipe || pipe_ctx_old->prev_odm_pipe) + continue; + + if (!pipe_ctx->stream || + pipe_need_reprogram(pipe_ctx_old, pipe_ctx)) { + + /* Reset all the syncd pipes from the disabled pipe */ + for (j = 0; j < dc->res_pool->pipe_count; j++) { + pipe_ctx_syncd = &context->res_ctx.pipe_ctx[j]; + if ((GET_PIPE_SYNCD_FROM_PIPE(pipe_ctx_syncd) == pipe_ctx_old->pipe_idx) || + !IS_PIPE_SYNCD_VALID(pipe_ctx_syncd)) + SET_PIPE_SYNCD_TO_PIPE(pipe_ctx_syncd, j); + } + } + } +} + +void check_syncd_pipes_for_disabled_master_pipe(struct dc *dc, + struct dc_state *context, + uint8_t disabled_master_pipe_idx) +{ + int i; + struct pipe_ctx *pipe_ctx, *pipe_ctx_check; + + pipe_ctx = &context->res_ctx.pipe_ctx[disabled_master_pipe_idx]; + if ((GET_PIPE_SYNCD_FROM_PIPE(pipe_ctx) != disabled_master_pipe_idx) || + !IS_PIPE_SYNCD_VALID(pipe_ctx)) + SET_PIPE_SYNCD_TO_PIPE(pipe_ctx, disabled_master_pipe_idx); + + /* for the pipe disabled, check if any slave pipe exists and assert */ + for (i = 0; i < dc->res_pool->pipe_count; i++) { + pipe_ctx_check = &context->res_ctx.pipe_ctx[i]; + + if ((GET_PIPE_SYNCD_FROM_PIPE(pipe_ctx_check) == disabled_master_pipe_idx) && + IS_PIPE_SYNCD_VALID(pipe_ctx_check) && (i != disabled_master_pipe_idx)) + DC_ERR("DC: Failure: pipe_idx[%d] syncd with disabled master pipe_idx[%d]\n", + i, disabled_master_pipe_idx); + } +} diff --git a/drivers/gpu/drm/amd/display/dc/dc.h b/drivers/gpu/drm/amd/display/dc/dc.h index bf2878235dba..18e59d635ca2 100644 --- a/drivers/gpu/drm/amd/display/dc/dc.h +++ b/drivers/gpu/drm/amd/display/dc/dc.h @@ -47,7 +47,7 @@ struct aux_payload; struct set_config_cmd_payload; struct dmub_notification; -#define DC_VER "3.2.163" +#define DC_VER "3.2.166" #define MAX_SURFACES 3 #define MAX_PLANES 6 @@ -75,6 +75,16 @@ enum dc_plane_type { DC_PLANE_TYPE_DCN_UNIVERSAL, }; +// Sizes defined as multiples of 64KB +enum det_size { + DET_SIZE_DEFAULT = 0, + DET_SIZE_192KB = 3, + DET_SIZE_256KB = 4, + DET_SIZE_320KB = 5, + DET_SIZE_384KB = 6 +}; + + struct dc_plane_cap { enum dc_plane_type type; uint32_t blends_with_above : 1; @@ -187,6 +197,7 @@ struct dc_caps { struct dc_color_caps color; #if defined(CONFIG_DRM_AMD_DC_DCN) bool dp_hpo; + bool hdmi_frl_pcon_support; #endif bool edp_dsc_support; bool vbios_lttpr_aware; @@ -333,6 +344,7 @@ struct dc_config { uint8_t vblank_alignment_max_frame_time_diff; bool is_asymmetric_memory; bool is_single_rank_dimm; + bool use_pipe_ctx_sync_logic; }; enum visual_confirm { @@ -510,7 +522,8 @@ union dpia_debug_options { uint32_t force_non_lttpr:1; uint32_t extend_aux_rd_interval:1; uint32_t disable_mst_dsc_work_around:1; - uint32_t reserved:28; + uint32_t hpd_delay_in_ms:12; + uint32_t reserved:16; } bits; uint32_t raw; }; @@ -678,6 +691,7 @@ struct dc_debug_options { /* TODO - remove once tested */ bool legacy_dp2_lt; bool set_mst_en_for_sst; + bool force_dp2_lt_fallback_method; #endif union mem_low_power_enable_options enable_mem_low_power; union root_clock_optimization_options root_clock_optimization; @@ -690,11 +704,14 @@ struct dc_debug_options { /* FEC/PSR1 sequence enable delay in 100us */ uint8_t fec_enable_delay_in100us; bool enable_driver_sequence_debug; + enum det_size crb_alloc_policy; + int crb_alloc_policy_min_disp_count; #if defined(CONFIG_DRM_AMD_DC_DCN) bool disable_z10; bool enable_sw_cntl_psr; union dpia_debug_options dpia_debug; #endif + bool apply_vendor_specific_lttpr_wa; }; struct gpu_info_soc_bounding_box_v1_0; @@ -1295,6 +1312,11 @@ struct dc_sink_dsc_caps { // 'true' if these are virtual DPCD's DSC caps (immediately upstream of sink in MST topology), // 'false' if they are sink's DSC caps bool is_virtual_dpcd_dsc; +#if defined(CONFIG_DRM_AMD_DC_DCN) + // 'true' if MST topology supports DSC passthrough for sink + // 'false' if MST topology does not support DSC passthrough + bool is_dsc_passthrough_supported; +#endif struct dsc_dec_dpcd_caps dsc_dec_caps; }; @@ -1410,6 +1432,9 @@ void dc_unlock_memory_clock_frequency(struct dc *dc); */ void dc_lock_memory_clock_frequency(struct dc *dc); +/* set soft max for memclk, to be used for AC/DC switching clock limitations */ +void dc_enable_dcmode_clk_limit(struct dc *dc, bool enable); + /* cleanup on driver unload */ void dc_hardware_release(struct dc *dc); diff --git a/drivers/gpu/drm/amd/display/dc/dc_dp_types.h b/drivers/gpu/drm/amd/display/dc/dc_dp_types.h index e68e9a86a4d9..353dac420f34 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_dp_types.h +++ b/drivers/gpu/drm/amd/display/dc/dc_dp_types.h @@ -378,7 +378,14 @@ enum dpcd_downstream_port_detailed_type { union dwnstream_port_caps_byte2 { struct { uint8_t MAX_BITS_PER_COLOR_COMPONENT:2; +#if defined(CONFIG_DRM_AMD_DC_DCN) + uint8_t MAX_ENCODED_LINK_BW_SUPPORT:3; + uint8_t SOURCE_CONTROL_MODE_SUPPORT:1; + uint8_t CONCURRENT_LINK_BRING_UP_SEQ_SUPPORT:1; + uint8_t RESERVED:1; +#else uint8_t RESERVED:6; +#endif } bits; uint8_t raw; }; @@ -416,6 +423,30 @@ union dwnstream_port_caps_byte3_hdmi { uint8_t raw; }; +#if defined(CONFIG_DRM_AMD_DC_DCN) +union hdmi_sink_encoded_link_bw_support { + struct { + uint8_t HDMI_SINK_ENCODED_LINK_BW_SUPPORT:3; + uint8_t RESERVED:5; + } bits; + uint8_t raw; +}; + +union hdmi_encoded_link_bw { + struct { + uint8_t FRL_MODE:1; // Bit 0 + uint8_t BW_9Gbps:1; + uint8_t BW_18Gbps:1; + uint8_t BW_24Gbps:1; + uint8_t BW_32Gbps:1; + uint8_t BW_40Gbps:1; + uint8_t BW_48Gbps:1; + uint8_t RESERVED:1; // Bit 7 + } bits; + uint8_t raw; +}; +#endif + /*4-byte structure for detailed capabilities of a down-stream port (DP-to-TMDS converter).*/ union dwnstream_portxcaps { @@ -852,6 +883,15 @@ struct psr_caps { unsigned char psr_version; unsigned int psr_rfb_setup_time; bool psr_exit_link_training_required; + unsigned char edp_revision; + unsigned char support_ver; + bool su_granularity_required; + bool y_coordinate_required; + uint8_t su_y_granularity; + bool alpm_cap; + bool standby_support; + uint8_t rate_control_caps; + unsigned int psr_power_opt_flag; }; /* Length of router topology ID read from DPCD in bytes. */ diff --git a/drivers/gpu/drm/amd/display/dc/dc_hw_types.h b/drivers/gpu/drm/amd/display/dc/dc_hw_types.h index 52355fe6994c..eac34f591a3f 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_hw_types.h +++ b/drivers/gpu/drm/amd/display/dc/dc_hw_types.h @@ -741,6 +741,9 @@ struct dc_dsc_config { uint32_t version_minor; /* DSC minor version. Full version is formed as 1.version_minor. */ bool ycbcr422_simple; /* Tell DSC engine to convert YCbCr 4:2:2 to 'YCbCr 4:2:2 simple'. */ int32_t rc_buffer_size; /* DSC RC buffer block size in bytes */ +#if defined(CONFIG_DRM_AMD_DC_DCN) + bool is_frl; /* indicate if DSC is applied based on HDMI FRL sink's capability */ +#endif bool is_dp; /* indicate if DSC is applied based on DP's capability */ }; struct dc_crtc_timing { diff --git a/drivers/gpu/drm/amd/display/dc/dc_link.h b/drivers/gpu/drm/amd/display/dc/dc_link.h index 37af564c4b33..4ebba641538b 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_link.h +++ b/drivers/gpu/drm/amd/display/dc/dc_link.h @@ -186,6 +186,10 @@ struct dc_link { /* Drive settings read from integrated info table */ struct dc_lane_settings bios_forced_drive_settings; + /* Vendor specific LTTPR workaround variables */ + uint8_t vendor_specific_lttpr_link_rate_wa; + bool apply_vendor_specific_lttpr_link_rate_wa; + /* MST record stream using this link */ struct link_flags { bool dp_keep_receiver_powered; @@ -447,6 +451,7 @@ bool dc_link_is_fec_supported(const struct dc_link *link); bool dc_link_should_enable_fec(const struct dc_link *link); #if defined(CONFIG_DRM_AMD_DC_DCN) +uint32_t dc_link_bw_kbps_from_raw_frl_link_rate_data(uint8_t bw); enum dp_link_encoding dc_link_dp_mst_decide_link_encoding_format(const struct dc_link *link); #endif #endif /* DC_LINK_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/dc_types.h b/drivers/gpu/drm/amd/display/dc/dc_types.h index 388457ffc0a8..0285a4b38d05 100644 --- a/drivers/gpu/drm/amd/display/dc/dc_types.h +++ b/drivers/gpu/drm/amd/display/dc/dc_types.h @@ -430,6 +430,7 @@ struct dc_dongle_caps { uint32_t dp_hdmi_max_bpc; uint32_t dp_hdmi_max_pixel_clk_in_khz; #if defined(CONFIG_DRM_AMD_DC_DCN) + uint32_t dp_hdmi_frl_max_link_bw_in_kbps; struct dc_dongle_dfp_cap_ext dfp_cap_ext; #endif }; @@ -950,6 +951,7 @@ enum dc_gpu_mem_alloc_type { enum dc_psr_version { DC_PSR_VERSION_1 = 0, + DC_PSR_VERSION_SU_1 = 1, DC_PSR_VERSION_UNSUPPORTED = 0xFFFFFFFF, }; diff --git a/drivers/gpu/drm/amd/display/dc/dce/dce_link_encoder.c b/drivers/gpu/drm/amd/display/dc/dce/dce_link_encoder.c index 1e77ffee71b3..f1c61d5aee6c 100644 --- a/drivers/gpu/drm/amd/display/dc/dce/dce_link_encoder.c +++ b/drivers/gpu/drm/amd/display/dc/dce/dce_link_encoder.c @@ -788,8 +788,9 @@ static bool dce110_link_encoder_validate_hdmi_output( crtc_timing->pixel_encoding == PIXEL_ENCODING_YCBCR420) return false; - if (!enc110->base.features.flags.bits.HDMI_6GB_EN && - adjusted_pix_clk_khz >= 300000) + if ((!enc110->base.features.flags.bits.HDMI_6GB_EN || + enc110->base.ctx->dc->debug.hdmi20_disable) && + adjusted_pix_clk_khz >= 300000) return false; if (enc110->base.ctx->dc->debug.hdmi20_disable && crtc_timing->pixel_encoding == PIXEL_ENCODING_YCBCR420) diff --git a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c index 3d421583e9ca..f1593186e964 100644 --- a/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c +++ b/drivers/gpu/drm/amd/display/dc/dce110/dce110_hw_sequencer.c @@ -69,6 +69,8 @@ #include "dcn10/dcn10_hw_sequencer.h" +#include "dce110_hw_sequencer.h" + #define GAMMA_HW_POINTS_NUM 256 /* @@ -1564,6 +1566,10 @@ static enum dc_status apply_single_controller_ctx_to_hw( &pipe_ctx->stream->audio_info); } + /* make sure no pipes syncd to the pipe being enabled */ + if (!pipe_ctx->stream->apply_seamless_boot_optimization && dc->config.use_pipe_ctx_sync_logic) + check_syncd_pipes_for_disabled_master_pipe(dc, context, pipe_ctx->pipe_idx); + #if defined(CONFIG_DRM_AMD_DC_DCN) /* DCN3.1 FPGA Workaround * Need to enable HPO DP Stream Encoder before setting OTG master enable. @@ -1792,7 +1798,6 @@ void dce110_enable_accelerated_mode(struct dc *dc, struct dc_state *context) struct dc_stream_state *edp_streams[MAX_NUM_EDP]; struct dc_link *edp_link_with_sink = NULL; struct dc_link *edp_link = NULL; - struct dc_stream_state *edp_stream = NULL; struct dce_hwseq *hws = dc->hwseq; int edp_with_sink_num; int edp_num; @@ -1813,27 +1818,29 @@ void dce110_enable_accelerated_mode(struct dc *dc, struct dc_state *context) get_edp_streams(context, edp_streams, &edp_stream_num); // Check fastboot support, disable on DCE8 because of blank screens - if (edp_num && dc->ctx->dce_version != DCE_VERSION_8_0 && + if (edp_num && edp_stream_num && dc->ctx->dce_version != DCE_VERSION_8_0 && dc->ctx->dce_version != DCE_VERSION_8_1 && dc->ctx->dce_version != DCE_VERSION_8_3) { for (i = 0; i < edp_num; i++) { edp_link = edp_links[i]; + if (edp_link != edp_streams[0]->link) + continue; // enable fastboot if backend is enabled on eDP - if (edp_link->link_enc->funcs->is_dig_enabled(edp_link->link_enc)) { - /* Set optimization flag on eDP stream*/ - if (edp_stream_num && edp_link->link_status.link_active) { - edp_stream = edp_streams[0]; - can_apply_edp_fast_boot = !is_edp_ilr_optimization_required(edp_stream->link, &edp_stream->timing); - edp_stream->apply_edp_fast_boot_optimization = can_apply_edp_fast_boot; - if (can_apply_edp_fast_boot) - DC_LOG_EVENT_LINK_TRAINING("eDP fast boot disabled to optimize link rate\n"); - - break; - } + if (edp_link->link_enc->funcs->is_dig_enabled && + edp_link->link_enc->funcs->is_dig_enabled(edp_link->link_enc) && + edp_link->link_status.link_active) { + struct dc_stream_state *edp_stream = edp_streams[0]; + + can_apply_edp_fast_boot = !is_edp_ilr_optimization_required(edp_stream->link, &edp_stream->timing); + edp_stream->apply_edp_fast_boot_optimization = can_apply_edp_fast_boot; + if (can_apply_edp_fast_boot) + DC_LOG_EVENT_LINK_TRAINING("eDP fast boot disabled to optimize link rate\n"); + + break; } } // We are trying to enable eDP, don't power down VDD - if (edp_stream_num && can_apply_edp_fast_boot) + if (can_apply_edp_fast_boot) keep_edp_vdd_on = true; } @@ -2294,6 +2301,10 @@ enum dc_status dce110_apply_ctx_to_hw( enum dc_status status; int i; + /* reset syncd pipes from disabled pipes */ + if (dc->config.use_pipe_ctx_sync_logic) + reset_syncd_pipes_from_disabled_pipes(dc, context); + /* Reset old context */ /* look up the targets that have been removed since last commit */ hws->funcs.reset_hw_ctx_wrap(dc, context); diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.c index 91fdfcd8a14e..db7ca4b0cdb9 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp.c @@ -119,14 +119,6 @@ void dpp_read_state(struct dpp *dpp_base, } } -/* Program gamut remap in bypass mode */ -void dpp_set_gamut_remap_bypass(struct dcn10_dpp *dpp) -{ - REG_SET(CM_GAMUT_REMAP_CONTROL, 0, - CM_GAMUT_REMAP_MODE, 0); - /* Gamut remap in bypass */ -} - #define IDENTITY_RATIO(ratio) (dc_fixpt_u2d19(ratio) == (1 << 19)) bool dpp1_get_optimal_number_of_taps( diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp_dscl.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp_dscl.c index e31a6f1516bb..f607a0e28f14 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp_dscl.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_dpp_dscl.c @@ -89,51 +89,6 @@ enum dscl_mode_sel { DSCL_MODE_DSCL_BYPASS = 6 }; -static void dpp1_dscl_set_overscan( - struct dcn10_dpp *dpp, - const struct scaler_data *data) -{ - uint32_t left = data->recout.x; - uint32_t top = data->recout.y; - - int right = data->h_active - data->recout.x - data->recout.width; - int bottom = data->v_active - data->recout.y - data->recout.height; - - if (right < 0) { - BREAK_TO_DEBUGGER(); - right = 0; - } - if (bottom < 0) { - BREAK_TO_DEBUGGER(); - bottom = 0; - } - - REG_SET_2(DSCL_EXT_OVERSCAN_LEFT_RIGHT, 0, - EXT_OVERSCAN_LEFT, left, - EXT_OVERSCAN_RIGHT, right); - - REG_SET_2(DSCL_EXT_OVERSCAN_TOP_BOTTOM, 0, - EXT_OVERSCAN_BOTTOM, bottom, - EXT_OVERSCAN_TOP, top); -} - -static void dpp1_dscl_set_otg_blank( - struct dcn10_dpp *dpp, const struct scaler_data *data) -{ - uint32_t h_blank_start = data->h_active; - uint32_t h_blank_end = 0; - uint32_t v_blank_start = data->v_active; - uint32_t v_blank_end = 0; - - REG_SET_2(OTG_H_BLANK, 0, - OTG_H_BLANK_START, h_blank_start, - OTG_H_BLANK_END, h_blank_end); - - REG_SET_2(OTG_V_BLANK, 0, - OTG_V_BLANK_START, v_blank_start, - OTG_V_BLANK_END, v_blank_end); -} - static int dpp1_dscl_get_pixel_depth_val(enum lb_pixel_depth depth) { if (depth == LB_PIXEL_DEPTH_30BPP) @@ -555,58 +510,6 @@ static enum lb_memory_config dpp1_dscl_find_lb_memory_config(struct dcn10_dpp *d return LB_MEMORY_CONFIG_0; } -void dpp1_dscl_set_scaler_auto_scale( - struct dpp *dpp_base, - const struct scaler_data *scl_data) -{ - enum lb_memory_config lb_config; - struct dcn10_dpp *dpp = TO_DCN10_DPP(dpp_base); - enum dscl_mode_sel dscl_mode = dpp1_dscl_get_dscl_mode( - dpp_base, scl_data, dpp_base->ctx->dc->debug.always_scale); - bool ycbcr = scl_data->format >= PIXEL_FORMAT_VIDEO_BEGIN - && scl_data->format <= PIXEL_FORMAT_VIDEO_END; - - dpp1_dscl_set_overscan(dpp, scl_data); - - dpp1_dscl_set_otg_blank(dpp, scl_data); - - REG_UPDATE(SCL_MODE, DSCL_MODE, dscl_mode); - - if (dscl_mode == DSCL_MODE_DSCL_BYPASS) - return; - - lb_config = dpp1_dscl_find_lb_memory_config(dpp, scl_data); - dpp1_dscl_set_lb(dpp, &scl_data->lb_params, lb_config); - - if (dscl_mode == DSCL_MODE_SCALING_444_BYPASS) - return; - - /* TODO: v_min */ - REG_SET_3(DSCL_AUTOCAL, 0, - AUTOCAL_MODE, AUTOCAL_MODE_AUTOSCALE, - AUTOCAL_NUM_PIPE, 0, - AUTOCAL_PIPE_ID, 0); - - /* Black offsets */ - if (ycbcr) - REG_SET_2(SCL_BLACK_OFFSET, 0, - SCL_BLACK_OFFSET_RGB_Y, BLACK_OFFSET_RGB_Y, - SCL_BLACK_OFFSET_CBCR, BLACK_OFFSET_CBCR); - else - - REG_SET_2(SCL_BLACK_OFFSET, 0, - SCL_BLACK_OFFSET_RGB_Y, BLACK_OFFSET_RGB_Y, - SCL_BLACK_OFFSET_CBCR, BLACK_OFFSET_RGB_Y); - - REG_SET_4(SCL_TAP_CONTROL, 0, - SCL_V_NUM_TAPS, scl_data->taps.v_taps - 1, - SCL_H_NUM_TAPS, scl_data->taps.h_taps - 1, - SCL_V_NUM_TAPS_C, scl_data->taps.v_taps_c - 1, - SCL_H_NUM_TAPS_C, scl_data->taps.h_taps_c - 1); - - dpp1_dscl_set_scl_filter(dpp, scl_data, ycbcr); -} - static void dpp1_dscl_set_manual_ratio_init( struct dcn10_dpp *dpp, const struct scaler_data *data) diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c index a2b925cc4132..2d470f524367 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_hw_sequencer.c @@ -77,9 +77,9 @@ #define PGFSM_POWER_ON 0 #define PGFSM_POWER_OFF 2 -void print_microsec(struct dc_context *dc_ctx, - struct dc_log_buffer_ctx *log_ctx, - uint32_t ref_cycle) +static void print_microsec(struct dc_context *dc_ctx, + struct dc_log_buffer_ctx *log_ctx, + uint32_t ref_cycle) { const uint32_t ref_clk_mhz = dc_ctx->dc->res_pool->ref_clocks.dchub_ref_clock_inKhz / 1000; static const unsigned int frac = 1000; @@ -132,7 +132,8 @@ static void log_mpc_crc(struct dc *dc, REG_READ(DPP_TOP0_DPP_CRC_VAL_B_A), REG_READ(DPP_TOP0_DPP_CRC_VAL_R_G)); } -void dcn10_log_hubbub_state(struct dc *dc, struct dc_log_buffer_ctx *log_ctx) +static void dcn10_log_hubbub_state(struct dc *dc, + struct dc_log_buffer_ctx *log_ctx) { struct dc_context *dc_ctx = dc->ctx; struct dcn_hubbub_wm wm; @@ -1972,10 +1973,9 @@ static bool wait_for_reset_trigger_to_occur( return rc; } -uint64_t reduceSizeAndFraction( - uint64_t *numerator, - uint64_t *denominator, - bool checkUint32Bounary) +static uint64_t reduceSizeAndFraction(uint64_t *numerator, + uint64_t *denominator, + bool checkUint32Bounary) { int i; bool ret = checkUint32Bounary == false; @@ -2023,7 +2023,7 @@ uint64_t reduceSizeAndFraction( return ret; } -bool is_low_refresh_rate(struct pipe_ctx *pipe) +static bool is_low_refresh_rate(struct pipe_ctx *pipe) { uint32_t master_pipe_refresh_rate = pipe->stream->timing.pix_clk_100hz * 100 / @@ -2032,7 +2032,8 @@ bool is_low_refresh_rate(struct pipe_ctx *pipe) return master_pipe_refresh_rate <= 30; } -uint8_t get_clock_divider(struct pipe_ctx *pipe, bool account_low_refresh_rate) +static uint8_t get_clock_divider(struct pipe_ctx *pipe, + bool account_low_refresh_rate) { uint32_t clock_divider = 1; uint32_t numpipes = 1; @@ -2052,10 +2053,8 @@ uint8_t get_clock_divider(struct pipe_ctx *pipe, bool account_low_refresh_rate) return clock_divider; } -int dcn10_align_pixel_clocks( - struct dc *dc, - int group_size, - struct pipe_ctx *grouped_pipes[]) +static int dcn10_align_pixel_clocks(struct dc *dc, int group_size, + struct pipe_ctx *grouped_pipes[]) { struct dc_context *dc_ctx = dc->ctx; int i, master = -1, embedded = -1; @@ -2344,7 +2343,7 @@ static void mmhub_read_vm_context0_settings(struct dcn10_hubp *hubp1, } -void dcn10_program_pte_vm(struct dce_hwseq *hws, struct hubp *hubp) +static void dcn10_program_pte_vm(struct dce_hwseq *hws, struct hubp *hubp) { struct dcn10_hubp *hubp1 = TO_DCN10_HUBP(hubp); struct vm_system_aperture_param apt = {0}; diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.c index 2dc4b4e4ba02..f4b34c110eae 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_link_encoder.c @@ -646,8 +646,9 @@ static bool dcn10_link_encoder_validate_hdmi_output( crtc_timing->pixel_encoding == PIXEL_ENCODING_YCBCR420) return false; - if (!enc10->base.features.flags.bits.HDMI_6GB_EN && - adjusted_pix_clk_100hz >= 3000000) + if ((!enc10->base.features.flags.bits.HDMI_6GB_EN || + enc10->base.ctx->dc->debug.hdmi20_disable) && + adjusted_pix_clk_100hz >= 3000000) return false; if (enc10->base.ctx->dc->debug.hdmi20_disable && crtc_timing->pixel_encoding == PIXEL_ENCODING_YCBCR420) diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_opp.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_opp.c index d54d731415d7..2c409356f512 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_opp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_opp.c @@ -348,36 +348,6 @@ void opp1_program_stereo( */ } -void opp1_program_oppbuf( - struct output_pixel_processor *opp, - struct oppbuf_params *oppbuf) -{ - struct dcn10_opp *oppn10 = TO_DCN10_OPP(opp); - - /* Program the oppbuf active width to be the frame width from mpc */ - REG_UPDATE(OPPBUF_CONTROL, OPPBUF_ACTIVE_WIDTH, oppbuf->active_width); - - /* Specifies the number of segments in multi-segment mode (DP-MSO operation) - * description "In 1/2/4 segment mode, specifies the horizontal active width in pixels of the display panel. - * In 4 segment split left/right mode, specifies the horizontal 1/2 active width in pixels of the display panel. - * Used to determine segment boundaries in multi-segment mode. Used to determine the width of the vertical active space in 3D frame packed modes. - * OPPBUF_ACTIVE_WIDTH must be integer divisible by the total number of segments." - */ - REG_UPDATE(OPPBUF_CONTROL, OPPBUF_DISPLAY_SEGMENTATION, oppbuf->mso_segmentation); - - /* description "Specifies the number of overlap pixels (1-8 overlapping pixels supported), used in multi-segment mode (DP-MSO operation)" */ - REG_UPDATE(OPPBUF_CONTROL, OPPBUF_OVERLAP_PIXEL_NUM, oppbuf->mso_overlap_pixel_num); - - /* description "Specifies the number of times a pixel is replicated (0-15 pixel replications supported). - * A value of 0 disables replication. The total number of times a pixel is output is OPPBUF_PIXEL_REPETITION + 1." - */ - REG_UPDATE(OPPBUF_CONTROL, OPPBUF_PIXEL_REPETITION, oppbuf->pixel_repetition); - - /* Controls the number of padded pixels at the end of a segment */ - if (REG(OPPBUF_CONTROL1)) - REG_UPDATE(OPPBUF_CONTROL1, OPPBUF_NUM_SEGMENT_PADDED_PIXELS, oppbuf->num_segment_padded_pixels); -} - void opp1_pipe_clock_control(struct output_pixel_processor *opp, bool enable) { struct dcn10_opp *oppn10 = TO_DCN10_OPP(opp); diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c index 3d2a2848857a..b1671b00ce40 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_optc.c @@ -132,22 +132,6 @@ void optc1_setup_vertical_interrupt2( } /** - * Vupdate keepout can be set to a window to block the update lock for that pipe from changing. - * Start offset begins with vstartup and goes for x number of clocks, - * end offset starts from end of vupdate to x number of clocks. - */ -void optc1_set_vupdate_keepout(struct timing_generator *optc, - struct vupdate_keepout_params *params) -{ - struct optc *optc1 = DCN10TG_FROM_TG(optc); - - REG_SET_3(OTG_VUPDATE_KEEPOUT, 0, - MASTER_UPDATE_LOCK_VUPDATE_KEEPOUT_START_OFFSET, params->start_offset, - MASTER_UPDATE_LOCK_VUPDATE_KEEPOUT_END_OFFSET, params->end_offset, - OTG_MASTER_UPDATE_LOCK_VUPDATE_KEEPOUT_EN, params->enable); -} - -/** * program_timing_generator used by mode timing set * Program CRTC Timing Registers - OTG_H_*, OTG_V_*, Pixel repetition. * Including SYNC. Call BIOS command table to program Timings. @@ -876,7 +860,7 @@ void optc1_set_static_screen_control( OTG_STATIC_SCREEN_FRAME_COUNT, num_frames); } -void optc1_setup_manual_trigger(struct timing_generator *optc) +static void optc1_setup_manual_trigger(struct timing_generator *optc) { struct optc *optc1 = DCN10TG_FROM_TG(optc); @@ -894,7 +878,7 @@ void optc1_setup_manual_trigger(struct timing_generator *optc) OTG_TRIGA_CLEAR, 1); } -void optc1_program_manual_trigger(struct timing_generator *optc) +static void optc1_program_manual_trigger(struct timing_generator *optc) { struct optc *optc1 = DCN10TG_FROM_TG(optc); diff --git a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c index 19a2dd619ec7..858b72149897 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn10/dcn10_resource.c @@ -686,9 +686,8 @@ static struct output_pixel_processor *dcn10_opp_create( return &opp->base; } -struct dce_aux *dcn10_aux_engine_create( - struct dc_context *ctx, - uint32_t inst) +static struct dce_aux *dcn10_aux_engine_create(struct dc_context *ctx, + uint32_t inst) { struct aux_engine_dce110 *aux_engine = kzalloc(sizeof(struct aux_engine_dce110), GFP_KERNEL); @@ -724,9 +723,8 @@ static const struct dce_i2c_mask i2c_masks = { I2C_COMMON_MASK_SH_LIST_DCE110(_MASK) }; -struct dce_i2c_hw *dcn10_i2c_hw_create( - struct dc_context *ctx, - uint32_t inst) +static struct dce_i2c_hw *dcn10_i2c_hw_create(struct dc_context *ctx, + uint32_t inst) { struct dce_i2c_hw *dce_i2c_hw = kzalloc(sizeof(struct dce_i2c_hw), GFP_KERNEL); @@ -805,7 +803,7 @@ static const struct encoder_feature_support link_enc_feature = { .flags.bits.IS_TPS4_CAPABLE = true }; -struct link_encoder *dcn10_link_encoder_create( +static struct link_encoder *dcn10_link_encoder_create( const struct encoder_init_data *enc_init_data) { struct dcn10_link_encoder *enc10 = @@ -847,7 +845,7 @@ static struct panel_cntl *dcn10_panel_cntl_create(const struct panel_cntl_init_d return &panel_cntl->base; } -struct clock_source *dcn10_clock_source_create( +static struct clock_source *dcn10_clock_source_create( struct dc_context *ctx, struct dc_bios *bios, enum clock_source_id id, @@ -945,7 +943,7 @@ static const struct resource_create_funcs res_create_maximus_funcs = { .create_hwseq = dcn10_hwseq_create, }; -void dcn10_clock_source_destroy(struct clock_source **clk_src) +static void dcn10_clock_source_destroy(struct clock_source **clk_src) { kfree(TO_DCE110_CLK_SRC(*clk_src)); *clk_src = NULL; @@ -1122,7 +1120,7 @@ static enum dc_status build_mapped_resource( return DC_OK; } -enum dc_status dcn10_add_stream_to_ctx( +static enum dc_status dcn10_add_stream_to_ctx( struct dc *dc, struct dc_state *new_ctx, struct dc_stream_state *dc_stream) diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp.c index a9e420c7d75a..970b65efeac1 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dpp.c @@ -251,20 +251,6 @@ static void dpp2_cnv_setup ( } -void dpp2_cnv_set_bias_scale( - struct dpp *dpp_base, - struct dc_bias_and_scale *bias_and_scale) -{ - struct dcn20_dpp *dpp = TO_DCN20_DPP(dpp_base); - - REG_UPDATE(FCNV_FP_BIAS_R, FCNV_FP_BIAS_R, bias_and_scale->bias_red); - REG_UPDATE(FCNV_FP_BIAS_G, FCNV_FP_BIAS_G, bias_and_scale->bias_green); - REG_UPDATE(FCNV_FP_BIAS_B, FCNV_FP_BIAS_B, bias_and_scale->bias_blue); - REG_UPDATE(FCNV_FP_SCALE_R, FCNV_FP_SCALE_R, bias_and_scale->scale_red); - REG_UPDATE(FCNV_FP_SCALE_G, FCNV_FP_SCALE_G, bias_and_scale->scale_green); - REG_UPDATE(FCNV_FP_SCALE_B, FCNV_FP_SCALE_B, bias_and_scale->scale_blue); -} - /*compute the maximum number of lines that we can fit in the line buffer*/ void dscl2_calc_lb_num_partitions( const struct scaler_data *scl_data, diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dwb_scl.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dwb_scl.c index 880954ac0b02..994fb732a7cb 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dwb_scl.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_dwb_scl.c @@ -527,7 +527,7 @@ static const uint16_t filter_12tap_16p_183[108] = { 0, 84, 16328, 16032, 416, 1944, 1944, 416, 16032, 16328, 84, 0, }; -const uint16_t *wbscl_get_filter_3tap_16p(struct fixed31_32 ratio) +static const uint16_t *wbscl_get_filter_3tap_16p(struct fixed31_32 ratio) { if (ratio.value < dc_fixpt_one.value) return filter_3tap_16p_upscale; @@ -539,7 +539,7 @@ const uint16_t *wbscl_get_filter_3tap_16p(struct fixed31_32 ratio) return filter_3tap_16p_183; } -const uint16_t *wbscl_get_filter_4tap_16p(struct fixed31_32 ratio) +static const uint16_t *wbscl_get_filter_4tap_16p(struct fixed31_32 ratio) { if (ratio.value < dc_fixpt_one.value) return filter_4tap_16p_upscale; diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubp.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubp.c index 5adf42a7cc27..dc1752e9f461 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubp.c @@ -192,9 +192,8 @@ void hubp2_vready_at_or_After_vsync(struct hubp *hubp, REG_UPDATE(DCHUBP_CNTL, HUBP_VREADY_AT_OR_AFTER_VSYNC, value); } -void hubp2_program_requestor( - struct hubp *hubp, - struct _vcs_dpi_display_rq_regs_st *rq_regs) +static void hubp2_program_requestor(struct hubp *hubp, + struct _vcs_dpi_display_rq_regs_st *rq_regs) { struct dcn20_hubp *hubp2 = TO_DCN20_HUBP(hubp); @@ -930,6 +929,16 @@ bool hubp2_is_flip_pending(struct hubp *hubp) void hubp2_set_blank(struct hubp *hubp, bool blank) { + hubp2_set_blank_regs(hubp, blank); + + if (blank) { + hubp->mpcc_id = 0xf; + hubp->opp_id = OPP_ID_INVALID; + } +} + +void hubp2_set_blank_regs(struct hubp *hubp, bool blank) +{ struct dcn20_hubp *hubp2 = TO_DCN20_HUBP(hubp); uint32_t blank_en = blank ? 1 : 0; @@ -951,9 +960,6 @@ void hubp2_set_blank(struct hubp *hubp, bool blank) HUBP_NO_OUTSTANDING_REQ, 1, 1, 200); } - - hubp->mpcc_id = 0xf; - hubp->opp_id = OPP_ID_INVALID; } } @@ -1285,7 +1291,7 @@ void hubp2_read_state(struct hubp *hubp) } -void hubp2_validate_dml_output(struct hubp *hubp, +static void hubp2_validate_dml_output(struct hubp *hubp, struct dc_context *ctx, struct _vcs_dpi_display_rq_regs_st *dml_rq_regs, struct _vcs_dpi_display_dlg_regs_st *dml_dlg_attr, @@ -1603,6 +1609,7 @@ static struct hubp_funcs dcn20_hubp_funcs = { .hubp_setup_interdependent = hubp2_setup_interdependent, .hubp_set_vm_system_aperture_settings = hubp2_set_vm_system_aperture_settings, .set_blank = hubp2_set_blank, + .set_blank_regs = hubp2_set_blank_regs, .dcc_control = hubp2_dcc_control, .mem_program_viewport = min_set_viewport, .set_cursor_attributes = hubp2_cursor_set_attributes, diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubp.h b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubp.h index eea2254b15e4..9204c3ef323b 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubp.h +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hubp.h @@ -330,6 +330,7 @@ void hubp2_program_surface_config( bool hubp2_is_flip_pending(struct hubp *hubp); void hubp2_set_blank(struct hubp *hubp, bool blank); +void hubp2_set_blank_regs(struct hubp *hubp, bool blank); void hubp2_cursor_set_position( struct hubp *hubp, diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c index e6af99ae3d9f..a17fe8ab2904 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.c @@ -615,6 +615,11 @@ void dcn20_disable_plane(struct dc *dc, struct pipe_ctx *pipe_ctx) pipe_ctx->pipe_idx); } +void dcn20_disable_pixel_data(struct dc *dc, struct pipe_ctx *pipe_ctx, bool blank) +{ + dcn20_blank_pixel_data(dc, pipe_ctx, blank); +} + static int calc_mpc_flow_ctrl_cnt(const struct dc_stream_state *stream, int opp_cnt) { @@ -1080,10 +1085,8 @@ static void dcn20_power_on_plane( } } -void dcn20_enable_plane( - struct dc *dc, - struct pipe_ctx *pipe_ctx, - struct dc_state *context) +static void dcn20_enable_plane(struct dc *dc, struct pipe_ctx *pipe_ctx, + struct dc_state *context) { //if (dc->debug.sanity_checks) { // dcn10_verify_allow_pstate_change_high(dc); @@ -1842,6 +1845,11 @@ void dcn20_optimize_bandwidth( dc->res_pool->ref_clocks.dchub_ref_clock_inKhz / 1000, true); + if (dc->clk_mgr->dc_mode_softmax_enabled) + if (dc->clk_mgr->clks.dramclk_khz > dc->clk_mgr->bw_params->dc_mode_softmax_memclk * 1000 && + context->bw_ctx.bw.dcn.clk.dramclk_khz <= dc->clk_mgr->bw_params->dc_mode_softmax_memclk * 1000) + dc->clk_mgr->funcs->set_max_memclk(dc->clk_mgr, dc->clk_mgr->bw_params->dc_mode_softmax_memclk); + dc->clk_mgr->funcs->update_clocks( dc->clk_mgr, context, diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.h b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.h index 6bba191cd33e..33a36c02b2f8 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.h +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_hwseq.h @@ -53,6 +53,10 @@ void dcn20_enable_stream(struct pipe_ctx *pipe_ctx); void dcn20_unblank_stream(struct pipe_ctx *pipe_ctx, struct dc_link_settings *link_settings); void dcn20_disable_plane(struct dc *dc, struct pipe_ctx *pipe_ctx); +void dcn20_disable_pixel_data( + struct dc *dc, + struct pipe_ctx *pipe_ctx, + bool blank); void dcn20_blank_pixel_data( struct dc *dc, struct pipe_ctx *pipe_ctx, diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_init.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_init.c index 5cfd4b0afea5..91e4885b743e 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_init.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_init.c @@ -27,6 +27,8 @@ #include "dcn10/dcn10_hw_sequencer.h" #include "dcn20_hwseq.h" +#include "dcn20_init.h" + static const struct hw_sequencer_funcs dcn20_funcs = { .program_gamut_remap = dcn10_program_gamut_remap, .init_hw = dcn10_init_hw, diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_mpc.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_mpc.c index 947eb0df3f12..15734db0cdea 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_mpc.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_mpc.c @@ -400,10 +400,9 @@ static void mpc20_program_ogam_pwl( } -void apply_DEDCN20_305_wa( - struct mpc *mpc, - int mpcc_id, enum dc_lut_mode current_mode, - enum dc_lut_mode next_mode) +static void apply_DEDCN20_305_wa(struct mpc *mpc, int mpcc_id, + enum dc_lut_mode current_mode, + enum dc_lut_mode next_mode) { struct dcn20_mpc *mpc20 = TO_DCN20_MPC(mpc); @@ -525,7 +524,7 @@ static void mpc2_init_mpcc(struct mpcc *mpcc, int mpcc_inst) mpcc->sm_cfg.enable = false; } -struct mpcc *mpc2_get_mpcc_for_dpp(struct mpc_tree *tree, int dpp_id) +static struct mpcc *mpc2_get_mpcc_for_dpp(struct mpc_tree *tree, int dpp_id) { struct mpcc *tmp_mpcc = tree->opp_list; diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_optc.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_optc.c index 8c34751b0e58..0340fdd3f5fb 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_optc.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_optc.c @@ -73,21 +73,6 @@ bool optc2_enable_crtc(struct timing_generator *optc) } /** - * DRR double buffering control to select buffer point - * for V_TOTAL, H_TOTAL, VTOTAL_MIN, VTOTAL_MAX, VTOTAL_MIN_SEL and VTOTAL_MAX_SEL registers - * Options: anytime, start of frame, dp start of frame (range timing) - */ -void optc2_set_timing_db_mode(struct timing_generator *optc, bool enable) -{ - struct optc *optc1 = DCN10TG_FROM_TG(optc); - - uint32_t blank_data_double_buffer_enable = enable ? 1 : 0; - - REG_UPDATE(OTG_DOUBLE_BUFFER_CONTROL, - OTG_RANGE_TIMING_DBUF_UPDATE_MODE, blank_data_double_buffer_enable); -} - -/** *For the below, I'm not sure how your GSL parameters are stored in your env, * so I will assume a gsl_params struct for now */ @@ -110,30 +95,6 @@ void optc2_set_gsl(struct timing_generator *optc, } -/* Use the gsl allow flip as the master update lock */ -void optc2_use_gsl_as_master_update_lock(struct timing_generator *optc, - const struct gsl_params *params) -{ - struct optc *optc1 = DCN10TG_FROM_TG(optc); - - REG_UPDATE(OTG_GSL_CONTROL, - OTG_MASTER_UPDATE_LOCK_GSL_EN, params->master_update_lock_gsl_en); -} - -/* You can control the GSL timing by limiting GSL to a window (X,Y) */ -void optc2_set_gsl_window(struct timing_generator *optc, - const struct gsl_params *params) -{ - struct optc *optc1 = DCN10TG_FROM_TG(optc); - - REG_SET_2(OTG_GSL_WINDOW_X, 0, - OTG_GSL_WINDOW_START_X, params->gsl_window_start_x, - OTG_GSL_WINDOW_END_X, params->gsl_window_end_x); - REG_SET_2(OTG_GSL_WINDOW_Y, 0, - OTG_GSL_WINDOW_START_Y, params->gsl_window_start_y, - OTG_GSL_WINDOW_END_Y, params->gsl_window_end_y); -} - void optc2_set_gsl_source_select( struct timing_generator *optc, int group_idx, @@ -156,18 +117,6 @@ void optc2_set_gsl_source_select( } } -/* DSC encoder frame start controls: x = h position, line_num = # of lines from vstartup */ -void optc2_set_dsc_encoder_frame_start(struct timing_generator *optc, - int x_position, - int line_num) -{ - struct optc *optc1 = DCN10TG_FROM_TG(optc); - - REG_SET_2(OTG_DSC_START_POSITION, 0, - OTG_DSC_START_POSITION_X, x_position, - OTG_DSC_START_POSITION_LINE_NUM, line_num); -} - /* Set DSC-related configuration. * dsc_mode: 0 disables DSC, other values enable DSC in specified format * sc_bytes_per_pixel: Bytes per pixel in u3.28 format @@ -293,8 +242,8 @@ void optc2_get_optc_source(struct timing_generator *optc, *num_of_src_opp = 1; } -void optc2_set_dwb_source(struct timing_generator *optc, - uint32_t dwb_pipe_inst) +static void optc2_set_dwb_source(struct timing_generator *optc, + uint32_t dwb_pipe_inst) { struct optc *optc1 = DCN10TG_FROM_TG(optc); @@ -306,7 +255,7 @@ void optc2_set_dwb_source(struct timing_generator *optc, OPTC_DWB1_SOURCE_SELECT, optc->inst); } -void optc2_align_vblanks( +static void optc2_align_vblanks( struct timing_generator *optc_master, struct timing_generator *optc_slave, uint32_t master_pixel_clock_100Hz, diff --git a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c index 3883f918b3bb..40b122a708ef 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn20/dcn20_resource.c @@ -3796,6 +3796,8 @@ static bool dcn20_resource_construct( dc->caps.color.mpc.ogam_rom_caps.hlg = 0; dc->caps.color.mpc.ocsc = 1; + dc->caps.hdmi_frl_pcon_support = true; + if (dc->ctx->dce_environment == DCE_ENV_PRODUCTION_DRV) { dc->debug = debug_defaults_drv; } else if (dc->ctx->dce_environment == DCE_ENV_FPGA_MAXIMUS) { diff --git a/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_dccg.c b/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_dccg.c index f5bf04f7da25..9a3402148fde 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_dccg.c +++ b/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_dccg.c @@ -44,7 +44,8 @@ #define DC_LOGGER \ dccg->ctx->logger -void dccg201_update_dpp_dto(struct dccg *dccg, int dpp_inst, int req_dppclk) +static void dccg201_update_dpp_dto(struct dccg *dccg, int dpp_inst, + int req_dppclk) { /* vbios handles it */ } diff --git a/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_hubp.c b/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_hubp.c index 6b6f74d4afd1..35dd4bac242a 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_hubp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_hubp.c @@ -55,7 +55,7 @@ static void hubp201_program_surface_config( hubp1_program_pixel_format(hubp, format); } -void hubp201_program_deadline( +static void hubp201_program_deadline( struct hubp *hubp, struct _vcs_dpi_display_dlg_regs_st *dlg_attr, struct _vcs_dpi_display_ttu_regs_st *ttu_attr) @@ -63,9 +63,8 @@ void hubp201_program_deadline( hubp1_program_deadline(hubp, dlg_attr, ttu_attr); } -void hubp201_program_requestor( - struct hubp *hubp, - struct _vcs_dpi_display_rq_regs_st *rq_regs) +static void hubp201_program_requestor(struct hubp *hubp, + struct _vcs_dpi_display_rq_regs_st *rq_regs) { struct dcn201_hubp *hubp201 = TO_DCN201_HUBP(hubp); diff --git a/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_link_encoder.c b/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_link_encoder.c index a65e8f7801db..7f9ec59ef443 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_link_encoder.c +++ b/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_link_encoder.c @@ -50,8 +50,8 @@ #define IND_REG(index) \ (enc10->link_regs->index) -void dcn201_link_encoder_get_max_link_cap(struct link_encoder *enc, - struct dc_link_settings *link_settings) +static void dcn201_link_encoder_get_max_link_cap(struct link_encoder *enc, + struct dc_link_settings *link_settings) { uint32_t value1, value2; struct dcn10_link_encoder *enc10 = TO_DCN10_LINK_ENC(enc); @@ -66,7 +66,7 @@ void dcn201_link_encoder_get_max_link_cap(struct link_encoder *enc, } } -bool dcn201_link_encoder_is_in_alt_mode(struct link_encoder *enc) +static bool dcn201_link_encoder_is_in_alt_mode(struct link_encoder *enc) { uint32_t value; struct dcn10_link_encoder *enc10 = TO_DCN10_LINK_ENC(enc); diff --git a/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_resource.c b/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_resource.c index 0fa381088d1d..d6acf9a8590a 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn201/dcn201_resource.c @@ -672,9 +672,8 @@ static struct output_pixel_processor *dcn201_opp_create( return &opp->base; } -struct dce_aux *dcn201_aux_engine_create( - struct dc_context *ctx, - uint32_t inst) +static struct dce_aux *dcn201_aux_engine_create(struct dc_context *ctx, + uint32_t inst) { struct aux_engine_dce110 *aux_engine = kzalloc(sizeof(struct aux_engine_dce110), GFP_ATOMIC); @@ -706,9 +705,8 @@ static const struct dce_i2c_mask i2c_masks = { I2C_COMMON_MASK_SH_LIST_DCN2(_MASK) }; -struct dce_i2c_hw *dcn201_i2c_hw_create( - struct dc_context *ctx, - uint32_t inst) +static struct dce_i2c_hw *dcn201_i2c_hw_create(struct dc_context *ctx, + uint32_t inst) { struct dce_i2c_hw *dce_i2c_hw = kzalloc(sizeof(struct dce_i2c_hw), GFP_ATOMIC); @@ -789,7 +787,7 @@ static const struct encoder_feature_support link_enc_feature = { .flags.bits.IS_TPS4_CAPABLE = true }; -struct link_encoder *dcn201_link_encoder_create( +static struct link_encoder *dcn201_link_encoder_create( const struct encoder_init_data *enc_init_data) { struct dcn20_link_encoder *enc20 = @@ -811,7 +809,7 @@ struct link_encoder *dcn201_link_encoder_create( return &enc10->base; } -struct clock_source *dcn201_clock_source_create( +static struct clock_source *dcn201_clock_source_create( struct dc_context *ctx, struct dc_bios *bios, enum clock_source_id id, @@ -906,7 +904,7 @@ static const struct resource_create_funcs res_create_maximus_funcs = { .create_hwseq = dcn201_hwseq_create, }; -void dcn201_clock_source_destroy(struct clock_source **clk_src) +static void dcn201_clock_source_destroy(struct clock_source **clk_src) { kfree(TO_DCE110_CLK_SRC(*clk_src)); *clk_src = NULL; diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubbub.c b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubbub.c index 36044cb8ec83..c5e200d09038 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubbub.c +++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubbub.c @@ -680,7 +680,7 @@ void hubbub21_wm_read_state(struct hubbub *hubbub, DCHUBBUB_ARB_ALLOW_DRAM_CLK_CHANGE_WATERMARK_D, &s->dram_clk_chanage); } -void hubbub21_apply_DEDCN21_147_wa(struct hubbub *hubbub) +static void hubbub21_apply_DEDCN21_147_wa(struct hubbub *hubbub) { struct dcn20_hubbub *hubbub1 = TO_DCN20_HUBBUB(hubbub); uint32_t prog_wm_value; diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubp.c b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubp.c index 3de1bcf9b3d8..58e459c7e7d3 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_hubp.c @@ -183,7 +183,7 @@ static void hubp21_setup( } -void hubp21_set_viewport( +static void hubp21_set_viewport( struct hubp *hubp, const struct rect *viewport, const struct rect *viewport_c) @@ -225,8 +225,8 @@ void hubp21_set_viewport( SEC_VIEWPORT_Y_START_C, viewport_c->y); } -void hubp21_set_vm_system_aperture_settings(struct hubp *hubp, - struct vm_system_aperture_param *apt) +static void hubp21_set_vm_system_aperture_settings(struct hubp *hubp, + struct vm_system_aperture_param *apt) { struct dcn21_hubp *hubp21 = TO_DCN21_HUBP(hubp); @@ -248,7 +248,7 @@ void hubp21_set_vm_system_aperture_settings(struct hubp *hubp, SYSTEM_ACCESS_MODE, 0x3); } -void hubp21_validate_dml_output(struct hubp *hubp, +static void hubp21_validate_dml_output(struct hubp *hubp, struct dc_context *ctx, struct _vcs_dpi_display_rq_regs_st *dml_rq_regs, struct _vcs_dpi_display_dlg_regs_st *dml_dlg_attr, @@ -664,7 +664,8 @@ static void program_surface_flip_and_addr(struct hubp *hubp, struct surface_flip flip_regs->DCSURF_PRIMARY_SURFACE_ADDRESS); } -void dmcub_PLAT_54186_wa(struct hubp *hubp, struct surface_flip_registers *flip_regs) +static void dmcub_PLAT_54186_wa(struct hubp *hubp, + struct surface_flip_registers *flip_regs) { struct dc_dmub_srv *dmcub = hubp->ctx->dmub_srv; struct dcn21_hubp *hubp21 = TO_DCN21_HUBP(hubp); @@ -697,7 +698,7 @@ void dmcub_PLAT_54186_wa(struct hubp *hubp, struct surface_flip_registers *flip_ PERF_TRACE(); // TODO: remove after performance is stable. } -bool hubp21_program_surface_flip_and_addr( +static bool hubp21_program_surface_flip_and_addr( struct hubp *hubp, const struct dc_plane_address *address, bool flip_immediate) @@ -805,7 +806,7 @@ bool hubp21_program_surface_flip_and_addr( return true; } -void hubp21_init(struct hubp *hubp) +static void hubp21_init(struct hubp *hubp) { // DEDCN21-133: Inconsistent row starting line for flip between DPTE and Meta // This is a chicken bit to enable the ECO fix. diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_init.c b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_init.c index 54c11ba550ae..b270f0b194dc 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_init.c +++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_init.c @@ -28,6 +28,8 @@ #include "dcn20/dcn20_hwseq.h" #include "dcn21_hwseq.h" +#include "dcn21_init.h" + static const struct hw_sequencer_funcs dcn21_funcs = { .program_gamut_remap = dcn10_program_gamut_remap, .init_hw = dcn10_init_hw, diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_link_encoder.c b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_link_encoder.c index aa46c35b05a2..0a1ba6e7081c 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_link_encoder.c +++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_link_encoder.c @@ -203,7 +203,7 @@ static bool update_cfg_data( return true; } -bool dcn21_link_encoder_acquire_phy(struct link_encoder *enc) +static bool dcn21_link_encoder_acquire_phy(struct link_encoder *enc) { struct dcn10_link_encoder *enc10 = TO_DCN10_LINK_ENC(enc); int value; @@ -277,7 +277,7 @@ void dcn21_link_encoder_enable_dp_output( } -void dcn21_link_encoder_enable_dp_mst_output( +static void dcn21_link_encoder_enable_dp_mst_output( struct link_encoder *enc, const struct dc_link_settings *link_settings, enum clock_source_id clock_source) @@ -288,9 +288,8 @@ void dcn21_link_encoder_enable_dp_mst_output( dcn10_link_encoder_enable_dp_mst_output(enc, link_settings, clock_source); } -void dcn21_link_encoder_disable_output( - struct link_encoder *enc, - enum signal_type signal) +static void dcn21_link_encoder_disable_output(struct link_encoder *enc, + enum signal_type signal) { dcn10_link_encoder_disable_output(enc, signal); diff --git a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c index d452a0d1777e..ca1bbc942fd4 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn21/dcn21_resource.c @@ -784,9 +784,8 @@ static const struct dce_i2c_mask i2c_masks = { I2C_COMMON_MASK_SH_LIST_DCN2(_MASK) }; -struct dce_i2c_hw *dcn21_i2c_hw_create( - struct dc_context *ctx, - uint32_t inst) +static struct dce_i2c_hw *dcn21_i2c_hw_create(struct dc_context *ctx, + uint32_t inst) { struct dce_i2c_hw *dce_i2c_hw = kzalloc(sizeof(struct dce_i2c_hw), GFP_KERNEL); @@ -1093,7 +1092,7 @@ static void patch_bounding_box(struct dc *dc, struct _vcs_dpi_soc_bounding_box_s } } -void dcn21_calculate_wm( +static void dcn21_calculate_wm( struct dc *dc, struct dc_state *context, display_e2e_pipe_params_st *pipes, int *out_pipe_cnt, @@ -1390,7 +1389,7 @@ validate_out: * with DC_FP_START()/DC_FP_END(). Use the same approach as for * dcn20_validate_bandwidth in dcn20_resource.c. */ -bool dcn21_validate_bandwidth(struct dc *dc, struct dc_state *context, +static bool dcn21_validate_bandwidth(struct dc *dc, struct dc_state *context, bool fast_validate) { bool voltage_supported; @@ -1480,8 +1479,8 @@ static struct hubbub *dcn21_hubbub_create(struct dc_context *ctx) return &hubbub->base; } -struct output_pixel_processor *dcn21_opp_create( - struct dc_context *ctx, uint32_t inst) +static struct output_pixel_processor *dcn21_opp_create(struct dc_context *ctx, + uint32_t inst) { struct dcn20_opp *opp = kzalloc(sizeof(struct dcn20_opp), GFP_KERNEL); @@ -1496,9 +1495,8 @@ struct output_pixel_processor *dcn21_opp_create( return &opp->base; } -struct timing_generator *dcn21_timing_generator_create( - struct dc_context *ctx, - uint32_t instance) +static struct timing_generator *dcn21_timing_generator_create(struct dc_context *ctx, + uint32_t instance) { struct optc *tgn10 = kzalloc(sizeof(struct optc), GFP_KERNEL); @@ -1518,7 +1516,7 @@ struct timing_generator *dcn21_timing_generator_create( return &tgn10->base; } -struct mpc *dcn21_mpc_create(struct dc_context *ctx) +static struct mpc *dcn21_mpc_create(struct dc_context *ctx) { struct dcn20_mpc *mpc20 = kzalloc(sizeof(struct dcn20_mpc), GFP_KERNEL); @@ -1545,8 +1543,8 @@ static void read_dce_straps( } -struct display_stream_compressor *dcn21_dsc_create( - struct dc_context *ctx, uint32_t inst) +static struct display_stream_compressor *dcn21_dsc_create(struct dc_context *ctx, + uint32_t inst) { struct dcn20_dsc *dsc = kzalloc(sizeof(struct dcn20_dsc), GFP_KERNEL); @@ -1683,9 +1681,8 @@ static struct dc_cap_funcs cap_funcs = { .get_dcc_compression_cap = dcn20_get_dcc_compression_cap }; -struct stream_encoder *dcn21_stream_encoder_create( - enum engine_id eng_id, - struct dc_context *ctx) +static struct stream_encoder *dcn21_stream_encoder_create(enum engine_id eng_id, + struct dc_context *ctx) { struct dcn10_stream_encoder *enc1 = kzalloc(sizeof(struct dcn10_stream_encoder), GFP_KERNEL); @@ -1917,7 +1914,7 @@ static int dcn21_populate_dml_pipes_from_context( return pipe_cnt; } -enum dc_status dcn21_patch_unknown_plane_state(struct dc_plane_state *plane_state) +static enum dc_status dcn21_patch_unknown_plane_state(struct dc_plane_state *plane_state) { enum dc_status result = DC_OK; @@ -2028,6 +2025,8 @@ static bool dcn21_resource_construct( dc->caps.color.mpc.ogam_rom_caps.hlg = 0; dc->caps.color.mpc.ocsc = 1; + dc->caps.hdmi_frl_pcon_support = true; + if (dc->ctx->dce_environment == DCE_ENV_PRODUCTION_DRV) dc->debug = debug_defaults_drv; else if (dc->ctx->dce_environment == DCE_ENV_FPGA_MAXIMUS) { diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_dio_stream_encoder.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_dio_stream_encoder.c index 7aa9aaf5db4c..8daa12730bc1 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_dio_stream_encoder.c +++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_dio_stream_encoder.c @@ -50,22 +50,6 @@ enc1->base.ctx -void convert_dc_info_packet_to_128( - const struct dc_info_packet *info_packet, - struct dc_info_packet_128 *info_packet_128) -{ - unsigned int i; - - info_packet_128->hb0 = info_packet->hb0; - info_packet_128->hb1 = info_packet->hb1; - info_packet_128->hb2 = info_packet->hb2; - info_packet_128->hb3 = info_packet->hb3; - - for (i = 0; i < 32; i++) { - info_packet_128->sb[i] = info_packet->sb[i]; - } - -} static void enc3_update_hdmi_info_packet( struct dcn10_stream_encoder *enc1, uint32_t packet_index, @@ -489,7 +473,7 @@ static void enc3_dp_set_odm_combine( } /* setup stream encoder in dvi mode */ -void enc3_stream_encoder_dvi_set_stream_attribute( +static void enc3_stream_encoder_dvi_set_stream_attribute( struct stream_encoder *enc, struct dc_crtc_timing *crtc_timing, bool is_dual_link) diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_dpp.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_dpp.c index c1d967ed6551..ab3918c0a15b 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_dpp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_dpp.c @@ -41,8 +41,7 @@ dpp->tf_shift->field_name, dpp->tf_mask->field_name -void dpp30_read_state(struct dpp *dpp_base, - struct dcn_dpp_state *s) +static void dpp30_read_state(struct dpp *dpp_base, struct dcn_dpp_state *s) { struct dcn20_dpp *dpp = TO_DCN20_DPP(dpp_base); @@ -373,7 +372,7 @@ void dpp3_set_cursor_attributes( } -bool dpp3_get_optimal_number_of_taps( +static bool dpp3_get_optimal_number_of_taps( struct dpp *dpp, struct scaler_data *scl_data, const struct scaling_taps *in_taps) @@ -474,22 +473,7 @@ bool dpp3_get_optimal_number_of_taps( return true; } -void dpp3_cnv_set_bias_scale( - struct dpp *dpp_base, - struct dc_bias_and_scale *bias_and_scale) -{ - struct dcn3_dpp *dpp = TO_DCN30_DPP(dpp_base); - - REG_UPDATE(FCNV_FP_BIAS_R, FCNV_FP_BIAS_R, bias_and_scale->bias_red); - REG_UPDATE(FCNV_FP_BIAS_G, FCNV_FP_BIAS_G, bias_and_scale->bias_green); - REG_UPDATE(FCNV_FP_BIAS_B, FCNV_FP_BIAS_B, bias_and_scale->bias_blue); - REG_UPDATE(FCNV_FP_SCALE_R, FCNV_FP_SCALE_R, bias_and_scale->scale_red); - REG_UPDATE(FCNV_FP_SCALE_G, FCNV_FP_SCALE_G, bias_and_scale->scale_green); - REG_UPDATE(FCNV_FP_SCALE_B, FCNV_FP_SCALE_B, bias_and_scale->scale_blue); -} - -void dpp3_deferred_update( - struct dpp *dpp_base) +static void dpp3_deferred_update(struct dpp *dpp_base) { int bypass_state; struct dcn3_dpp *dpp = TO_DCN30_DPP(dpp_base); @@ -751,8 +735,8 @@ static enum dc_lut_mode dpp3_get_blndgam_current(struct dpp *dpp_base) return mode; } -bool dpp3_program_blnd_lut( - struct dpp *dpp_base, const struct pwl_params *params) +static bool dpp3_program_blnd_lut(struct dpp *dpp_base, + const struct pwl_params *params) { enum dc_lut_mode current_mode; enum dc_lut_mode next_mode; @@ -1164,9 +1148,8 @@ static void dpp3_program_shaper_lutb_settings( } -bool dpp3_program_shaper( - struct dpp *dpp_base, - const struct pwl_params *params) +static bool dpp3_program_shaper(struct dpp *dpp_base, + const struct pwl_params *params) { enum dc_lut_mode current_mode; enum dc_lut_mode next_mode; @@ -1355,9 +1338,8 @@ static void dpp3_select_3dlut_ram_mask( REG_SET(CM_3DLUT_INDEX, 0, CM_3DLUT_INDEX, 0); } -bool dpp3_program_3dlut( - struct dpp *dpp_base, - struct tetrahedral_params *params) +static bool dpp3_program_3dlut(struct dpp *dpp_base, + struct tetrahedral_params *params) { enum dc_lut_mode mode; bool is_17x17x17; diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hubp.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hubp.c index eac08926b574..6a4dcafb9bba 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hubp.c +++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hubp.c @@ -490,6 +490,7 @@ static struct hubp_funcs dcn30_hubp_funcs = { .hubp_setup_interdependent = hubp2_setup_interdependent, .hubp_set_vm_system_aperture_settings = hubp3_set_vm_system_aperture_settings, .set_blank = hubp2_set_blank, + .set_blank_regs = hubp2_set_blank_regs, .dcc_control = hubp3_dcc_control, .mem_program_viewport = min_set_viewport, .set_cursor_attributes = hubp2_cursor_set_attributes, diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c index 3e99bb9c70ab..1db1ca19411d 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c +++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.c @@ -344,6 +344,17 @@ void dcn30_enable_writeback( dwb->funcs->enable(dwb, &wb_info->dwb_params); } +void dcn30_prepare_bandwidth(struct dc *dc, + struct dc_state *context) +{ + if (dc->clk_mgr->dc_mode_softmax_enabled) + if (dc->clk_mgr->clks.dramclk_khz <= dc->clk_mgr->bw_params->dc_mode_softmax_memclk * 1000 && + context->bw_ctx.bw.dcn.clk.dramclk_khz > dc->clk_mgr->bw_params->dc_mode_softmax_memclk * 1000) + dc->clk_mgr->funcs->set_max_memclk(dc->clk_mgr, dc->clk_mgr->bw_params->clk_table.entries[dc->clk_mgr->bw_params->clk_table.num_entries - 1].memclk_mhz); + + dcn20_prepare_bandwidth(dc, context); +} + void dcn30_disable_writeback( struct dc *dc, unsigned int dwb_pipe_inst) diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.h b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.h index e9a0005288d3..73e7b690e82c 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.h +++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_hwseq.h @@ -27,7 +27,7 @@ #define __DC_HWSS_DCN30_H__ #include "hw_sequencer_private.h" - +#include "dcn20/dcn20_hwseq.h" struct dc; void dcn30_init_hw(struct dc *dc); @@ -47,6 +47,9 @@ void dcn30_disable_writeback( struct dc *dc, unsigned int dwb_pipe_inst); +void dcn30_prepare_bandwidth(struct dc *dc, + struct dc_state *context); + bool dcn30_mmhubbub_warmup( struct dc *dc, unsigned int num_dwb, diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_init.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_init.c index 93f32a312fee..bb347319de83 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_init.c +++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_init.c @@ -29,6 +29,8 @@ #include "dcn21/dcn21_hwseq.h" #include "dcn30_hwseq.h" +#include "dcn30_init.h" + static const struct hw_sequencer_funcs dcn30_funcs = { .program_gamut_remap = dcn10_program_gamut_remap, .init_hw = dcn30_init_hw, @@ -53,6 +55,7 @@ static const struct hw_sequencer_funcs dcn30_funcs = { .enable_audio_stream = dce110_enable_audio_stream, .disable_audio_stream = dce110_disable_audio_stream, .disable_plane = dcn20_disable_plane, + .disable_pixel_data = dcn20_disable_pixel_data, .pipe_control_lock = dcn20_pipe_control_lock, .interdependent_update_lock = dcn10_lock_all_pipes, .cursor_lock = dcn10_cursor_lock, diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_mmhubbub.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_mmhubbub.c index 1c4b171c68ad..7a93eff183d9 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_mmhubbub.c +++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_mmhubbub.c @@ -100,7 +100,7 @@ static void mmhubbub3_warmup_mcif(struct mcif_wb *mcif_wb, REG_UPDATE(MMHUBBUB_WARMUP_CONTROL_STATUS, MMHUBBUB_WARMUP_EN, false); } -void mmhubbub3_config_mcif_buf(struct mcif_wb *mcif_wb, +static void mmhubbub3_config_mcif_buf(struct mcif_wb *mcif_wb, struct mcif_buf_params *params, unsigned int dest_height) { diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_mpc.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_mpc.c index 95149734378b..0ce0d6165f43 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_mpc.c +++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_mpc.c @@ -1362,7 +1362,7 @@ uint32_t mpcc3_acquire_rmu(struct mpc *mpc, int mpcc_id, int rmu_idx) return -1; } -int mpcc3_release_rmu(struct mpc *mpc, int mpcc_id) +static int mpcc3_release_rmu(struct mpc *mpc, int mpcc_id) { struct dcn30_mpc *mpc30 = TO_DCN30_MPC(mpc); int rmu_idx; diff --git a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c index 79a66e0c4303..369ceeeddc7e 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn30/dcn30_resource.c @@ -816,7 +816,7 @@ static const struct dc_plane_cap plane_cap = { .argb8888 = true, .nv12 = true, .fp16 = true, - .p010 = false, + .p010 = true, .ayuv = false, }, @@ -875,7 +875,7 @@ static const struct dc_debug_options debug_defaults_diags = { .use_max_lb = true }; -void dcn30_dpp_destroy(struct dpp **dpp) +static void dcn30_dpp_destroy(struct dpp **dpp) { kfree(TO_DCN20_DPP(*dpp)); *dpp = NULL; @@ -992,7 +992,7 @@ static struct mpc *dcn30_mpc_create( return &mpc30->base; } -struct hubbub *dcn30_hubbub_create(struct dc_context *ctx) +static struct hubbub *dcn30_hubbub_create(struct dc_context *ctx) { int i; @@ -1143,9 +1143,8 @@ static struct afmt *dcn30_afmt_create( return &afmt3->base; } -struct stream_encoder *dcn30_stream_encoder_create( - enum engine_id eng_id, - struct dc_context *ctx) +static struct stream_encoder *dcn30_stream_encoder_create(enum engine_id eng_id, + struct dc_context *ctx) { struct dcn10_stream_encoder *enc1; struct vpg *vpg; @@ -1179,8 +1178,7 @@ struct stream_encoder *dcn30_stream_encoder_create( return &enc1->base; } -struct dce_hwseq *dcn30_hwseq_create( - struct dc_context *ctx) +static struct dce_hwseq *dcn30_hwseq_create(struct dc_context *ctx) { struct dce_hwseq *hws = kzalloc(sizeof(struct dce_hwseq), GFP_KERNEL); @@ -1763,6 +1761,17 @@ static bool dcn30_split_stream_for_mpc_or_odm( int pipe_idx = sec_pipe->pipe_idx; const struct resource_pool *pool = dc->res_pool; + if (pri_pipe->plane_state) { + /* ODM + window MPO, where MPO window is on left half only */ + if (pri_pipe->plane_state->clip_rect.x + pri_pipe->plane_state->clip_rect.width <= + pri_pipe->stream->src.x + pri_pipe->stream->src.width/2) + return true; + + /* ODM + window MPO, where MPO window is on right half only */ + if (pri_pipe->plane_state->clip_rect.x >= pri_pipe->stream->src.width/2) + return true; + } + *sec_pipe = *pri_pipe; sec_pipe->pipe_idx = pipe_idx; @@ -2639,6 +2648,8 @@ static bool dcn30_resource_construct( dc->caps.color.mpc.ogam_rom_caps.hlg = 0; dc->caps.color.mpc.ocsc = 1; + dc->caps.hdmi_frl_pcon_support = true; + /* read VBIOS LTTPR caps */ { if (ctx->dc_bios->funcs->get_lttpr_caps) { diff --git a/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_init.c b/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_init.c index e85b695f2351..3d42a1a337ec 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_init.c +++ b/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_init.c @@ -30,6 +30,8 @@ #include "dcn30/dcn30_hwseq.h" #include "dcn301_hwseq.h" +#include "dcn301_init.h" + static const struct hw_sequencer_funcs dcn301_funcs = { .program_gamut_remap = dcn10_program_gamut_remap, .init_hw = dcn10_init_hw, diff --git a/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_panel_cntl.c b/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_panel_cntl.c index 736bda30abc3..ad0df1a72a90 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_panel_cntl.c +++ b/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_panel_cntl.c @@ -93,7 +93,7 @@ static unsigned int dcn301_get_16_bit_backlight_from_pwm(struct panel_cntl *pane return (uint32_t)(current_backlight); } -uint32_t dcn301_panel_cntl_hw_init(struct panel_cntl *panel_cntl) +static uint32_t dcn301_panel_cntl_hw_init(struct panel_cntl *panel_cntl) { struct dcn301_panel_cntl *dcn301_panel_cntl = TO_DCN301_PANEL_CNTL(panel_cntl); uint32_t value; @@ -147,7 +147,7 @@ uint32_t dcn301_panel_cntl_hw_init(struct panel_cntl *panel_cntl) return current_backlight; } -void dcn301_panel_cntl_destroy(struct panel_cntl **panel_cntl) +static void dcn301_panel_cntl_destroy(struct panel_cntl **panel_cntl) { struct dcn301_panel_cntl *dcn301_panel_cntl = TO_DCN301_PANEL_CNTL(*panel_cntl); @@ -155,7 +155,7 @@ void dcn301_panel_cntl_destroy(struct panel_cntl **panel_cntl) *panel_cntl = NULL; } -bool dcn301_is_panel_backlight_on(struct panel_cntl *panel_cntl) +static bool dcn301_is_panel_backlight_on(struct panel_cntl *panel_cntl) { struct dcn301_panel_cntl *dcn301_panel_cntl = TO_DCN301_PANEL_CNTL(panel_cntl); uint32_t value; @@ -165,7 +165,7 @@ bool dcn301_is_panel_backlight_on(struct panel_cntl *panel_cntl) return value; } -bool dcn301_is_panel_powered_on(struct panel_cntl *panel_cntl) +static bool dcn301_is_panel_powered_on(struct panel_cntl *panel_cntl) { struct dcn301_panel_cntl *dcn301_panel_cntl = TO_DCN301_PANEL_CNTL(panel_cntl); uint32_t pwr_seq_state, dig_on, dig_on_ovrd; @@ -177,7 +177,7 @@ bool dcn301_is_panel_powered_on(struct panel_cntl *panel_cntl) return (pwr_seq_state == 1) || (dig_on == 1 && dig_on_ovrd == 1); } -void dcn301_store_backlight_level(struct panel_cntl *panel_cntl) +static void dcn301_store_backlight_level(struct panel_cntl *panel_cntl) { struct dcn301_panel_cntl *dcn301_panel_cntl = TO_DCN301_PANEL_CNTL(panel_cntl); diff --git a/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_resource.c b/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_resource.c index 7abc36a4ff76..b4001233867c 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn301/dcn301_resource.c @@ -656,7 +656,7 @@ static const struct dc_plane_cap plane_cap = { .argb8888 = true, .nv12 = true, .fp16 = true, - .p010 = false, + .p010 = true, .ayuv = false, }, @@ -717,15 +717,13 @@ static const struct dc_debug_options debug_defaults_diags = { .use_max_lb = false, }; -void dcn301_dpp_destroy(struct dpp **dpp) +static void dcn301_dpp_destroy(struct dpp **dpp) { kfree(TO_DCN20_DPP(*dpp)); *dpp = NULL; } -struct dpp *dcn301_dpp_create( - struct dc_context *ctx, - uint32_t inst) +static struct dpp *dcn301_dpp_create(struct dc_context *ctx, uint32_t inst) { struct dcn3_dpp *dpp = kzalloc(sizeof(struct dcn3_dpp), GFP_KERNEL); @@ -741,8 +739,8 @@ struct dpp *dcn301_dpp_create( kfree(dpp); return NULL; } -struct output_pixel_processor *dcn301_opp_create( - struct dc_context *ctx, uint32_t inst) +static struct output_pixel_processor *dcn301_opp_create(struct dc_context *ctx, + uint32_t inst) { struct dcn20_opp *opp = kzalloc(sizeof(struct dcn20_opp), GFP_KERNEL); @@ -757,9 +755,7 @@ struct output_pixel_processor *dcn301_opp_create( return &opp->base; } -struct dce_aux *dcn301_aux_engine_create( - struct dc_context *ctx, - uint32_t inst) +static struct dce_aux *dcn301_aux_engine_create(struct dc_context *ctx, uint32_t inst) { struct aux_engine_dce110 *aux_engine = kzalloc(sizeof(struct aux_engine_dce110), GFP_KERNEL); @@ -793,9 +789,7 @@ static const struct dce_i2c_mask i2c_masks = { I2C_COMMON_MASK_SH_LIST_DCN2(_MASK) }; -struct dce_i2c_hw *dcn301_i2c_hw_create( - struct dc_context *ctx, - uint32_t inst) +static struct dce_i2c_hw *dcn301_i2c_hw_create(struct dc_context *ctx, uint32_t inst) { struct dce_i2c_hw *dce_i2c_hw = kzalloc(sizeof(struct dce_i2c_hw), GFP_KERNEL); @@ -829,7 +823,7 @@ static struct mpc *dcn301_mpc_create( return &mpc30->base; } -struct hubbub *dcn301_hubbub_create(struct dc_context *ctx) +static struct hubbub *dcn301_hubbub_create(struct dc_context *ctx) { int i; @@ -860,9 +854,8 @@ struct hubbub *dcn301_hubbub_create(struct dc_context *ctx) return &hubbub3->base; } -struct timing_generator *dcn301_timing_generator_create( - struct dc_context *ctx, - uint32_t instance) +static struct timing_generator *dcn301_timing_generator_create( + struct dc_context *ctx, uint32_t instance) { struct optc *tgn10 = kzalloc(sizeof(struct optc), GFP_KERNEL); @@ -894,7 +887,7 @@ static const struct encoder_feature_support link_enc_feature = { .flags.bits.IS_TPS4_CAPABLE = true }; -struct link_encoder *dcn301_link_encoder_create( +static struct link_encoder *dcn301_link_encoder_create( const struct encoder_init_data *enc_init_data) { struct dcn20_link_encoder *enc20 = @@ -915,7 +908,7 @@ struct link_encoder *dcn301_link_encoder_create( return &enc20->enc10.base; } -struct panel_cntl *dcn301_panel_cntl_create(const struct panel_cntl_init_data *init_data) +static struct panel_cntl *dcn301_panel_cntl_create(const struct panel_cntl_init_data *init_data) { struct dcn301_panel_cntl *panel_cntl = kzalloc(sizeof(struct dcn301_panel_cntl), GFP_KERNEL); @@ -997,9 +990,8 @@ static struct afmt *dcn301_afmt_create( return &afmt3->base; } -struct stream_encoder *dcn301_stream_encoder_create( - enum engine_id eng_id, - struct dc_context *ctx) +static struct stream_encoder *dcn301_stream_encoder_create(enum engine_id eng_id, + struct dc_context *ctx) { struct dcn10_stream_encoder *enc1; struct vpg *vpg; @@ -1033,8 +1025,7 @@ struct stream_encoder *dcn301_stream_encoder_create( return &enc1->base; } -struct dce_hwseq *dcn301_hwseq_create( - struct dc_context *ctx) +static struct dce_hwseq *dcn301_hwseq_create(struct dc_context *ctx) { struct dce_hwseq *hws = kzalloc(sizeof(struct dce_hwseq), GFP_KERNEL); @@ -1182,9 +1173,7 @@ static void dcn301_destruct(struct dcn301_resource_pool *pool) dcn_dccg_destroy(&pool->base.dccg); } -struct hubp *dcn301_hubp_create( - struct dc_context *ctx, - uint32_t inst) +static struct hubp *dcn301_hubp_create(struct dc_context *ctx, uint32_t inst) { struct dcn20_hubp *hubp2 = kzalloc(sizeof(struct dcn20_hubp), GFP_KERNEL); @@ -1201,7 +1190,7 @@ struct hubp *dcn301_hubp_create( return NULL; } -bool dcn301_dwbc_create(struct dc_context *ctx, struct resource_pool *pool) +static bool dcn301_dwbc_create(struct dc_context *ctx, struct resource_pool *pool) { int i; uint32_t pipe_count = pool->res_cap->num_dwb; @@ -1226,7 +1215,7 @@ bool dcn301_dwbc_create(struct dc_context *ctx, struct resource_pool *pool) return true; } -bool dcn301_mmhubbub_create(struct dc_context *ctx, struct resource_pool *pool) +static bool dcn301_mmhubbub_create(struct dc_context *ctx, struct resource_pool *pool) { int i; uint32_t pipe_count = pool->res_cap->num_dwb; diff --git a/drivers/gpu/drm/amd/display/dc/dcn302/dcn302_init.c b/drivers/gpu/drm/amd/display/dc/dcn302/dcn302_init.c index d88b9011c502..eb375f30f5bc 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn302/dcn302_init.c +++ b/drivers/gpu/drm/amd/display/dc/dcn302/dcn302_init.c @@ -29,6 +29,8 @@ #include "dc.h" +#include "dcn302_init.h" + void dcn302_hw_sequencer_construct(struct dc *dc) { dcn30_hw_sequencer_construct(dc); diff --git a/drivers/gpu/drm/amd/display/dc/dcn302/dcn302_resource.c b/drivers/gpu/drm/amd/display/dc/dcn302/dcn302_resource.c index 058f5d71e037..003e95368672 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn302/dcn302_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn302/dcn302_resource.c @@ -276,7 +276,7 @@ static const struct dc_plane_cap plane_cap = { .argb8888 = true, .nv12 = true, .fp16 = true, - .p010 = false, + .p010 = true, .ayuv = false, }, .max_upscale_factor = { diff --git a/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_init.c b/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_init.c index aa5dbbade2bd..f499f8ab5e47 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_init.c +++ b/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_init.c @@ -9,6 +9,8 @@ #include "dcn30/dcn30_init.h" #include "dc.h" +#include "dcn303_init.h" + void dcn303_hw_sequencer_construct(struct dc *dc) { dcn30_hw_sequencer_construct(dc); diff --git a/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_resource.c b/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_resource.c index 7024aeb0884c..01ba9d656c72 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn303/dcn303_resource.c @@ -254,7 +254,7 @@ static const struct dc_plane_cap plane_cap = { .argb8888 = true, .nv12 = true, .fp16 = true, - .p010 = false, + .p010 = true, .ayuv = false, }, .max_upscale_factor = { diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dccg.c b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dccg.c index 815481a3ef54..ea4f8e06b07c 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dccg.c +++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dccg.c @@ -462,7 +462,7 @@ void dccg31_set_physymclk( } /* Controls the generation of pixel valid for OTG in (OTG -> HPO case) */ -void dccg31_set_dtbclk_dto( +static void dccg31_set_dtbclk_dto( struct dccg *dccg, int dtbclk_inst, int req_dtbclk_khz, diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dio_link_encoder.c b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dio_link_encoder.c index ee6f13bef377..71c359f9cdd2 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dio_link_encoder.c +++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_dio_link_encoder.c @@ -67,6 +67,39 @@ #define MIN(X, Y) ((X) < (Y) ? (X) : (Y)) #endif +static uint8_t phy_id_from_transmitter(enum transmitter t) +{ + uint8_t phy_id; + + switch (t) { + case TRANSMITTER_UNIPHY_A: + phy_id = 0; + break; + case TRANSMITTER_UNIPHY_B: + phy_id = 1; + break; + case TRANSMITTER_UNIPHY_C: + phy_id = 2; + break; + case TRANSMITTER_UNIPHY_D: + phy_id = 3; + break; + case TRANSMITTER_UNIPHY_E: + phy_id = 4; + break; + case TRANSMITTER_UNIPHY_F: + phy_id = 5; + break; + case TRANSMITTER_UNIPHY_G: + phy_id = 6; + break; + default: + phy_id = 0; + break; + } + return phy_id; +} + void dcn31_link_encoder_set_dio_phy_mux( struct link_encoder *enc, enum encoder_type_select sel, @@ -141,7 +174,7 @@ void dcn31_link_encoder_set_dio_phy_mux( } } -void enc31_hw_init(struct link_encoder *enc) +static void enc31_hw_init(struct link_encoder *enc) { struct dcn10_link_encoder *enc10 = TO_DCN10_LINK_ENC(enc); @@ -536,57 +569,45 @@ void dcn31_link_encoder_disable_output( bool dcn31_link_encoder_is_in_alt_mode(struct link_encoder *enc) { struct dcn10_link_encoder *enc10 = TO_DCN10_LINK_ENC(enc); - uint32_t dp_alt_mode_disable; + struct dc_dmub_srv *dc_dmub_srv = enc->ctx->dmub_srv; + union dmub_rb_cmd cmd; bool is_usb_c_alt_mode = false; - if (enc->features.flags.bits.DP_IS_USB_C) { - if (enc->ctx->asic_id.hw_internal_rev != YELLOW_CARP_B0) { - // [Note] no need to check hw_internal_rev once phy mux selection is ready - REG_GET(RDPCSTX_PHY_CNTL6, RDPCS_PHY_DPALT_DISABLE, &dp_alt_mode_disable); - } else { - /* - * B0 phys use a new set of registers to check whether alt mode is disabled. - * if value == 1 alt mode is disabled, otherwise it is enabled. - */ - if ((enc10->base.transmitter == TRANSMITTER_UNIPHY_A) - || (enc10->base.transmitter == TRANSMITTER_UNIPHY_B) - || (enc10->base.transmitter == TRANSMITTER_UNIPHY_E)) { - REG_GET(RDPCSTX_PHY_CNTL6, RDPCS_PHY_DPALT_DISABLE, &dp_alt_mode_disable); - } else { - // [Note] need to change TRANSMITTER_UNIPHY_C/D to F/G once phy mux selection is ready - REG_GET(RDPCSPIPE_PHY_CNTL6, RDPCS_PHY_DPALT_DISABLE, &dp_alt_mode_disable); - } - } + if (enc->features.flags.bits.DP_IS_USB_C && dc_dmub_srv) { + memset(&cmd, 0, sizeof(cmd)); + cmd.query_dp_alt.header.type = DMUB_CMD__VBIOS; + cmd.query_dp_alt.header.sub_type = DMUB_CMD__VBIOS_TRANSMITTER_QUERY_DP_ALT; + cmd.query_dp_alt.header.payload_bytes = sizeof(cmd.panel_cntl.data); + cmd.query_dp_alt.data.phy_id = phy_id_from_transmitter(enc10->base.transmitter); - is_usb_c_alt_mode = (dp_alt_mode_disable == 0); + if (!dc_dmub_srv_cmd_with_reply_data(dc_dmub_srv, &cmd)) + return false; + + is_usb_c_alt_mode = (cmd.query_dp_alt.data.is_dp_alt_disable == 0); } return is_usb_c_alt_mode; } -void dcn31_link_encoder_get_max_link_cap(struct link_encoder *enc, - struct dc_link_settings *link_settings) +void dcn31_link_encoder_get_max_link_cap(struct link_encoder *enc, struct dc_link_settings *link_settings) { struct dcn10_link_encoder *enc10 = TO_DCN10_LINK_ENC(enc); - uint32_t is_in_usb_c_dp4_mode = 0; + struct dc_dmub_srv *dc_dmub_srv = enc->ctx->dmub_srv; + union dmub_rb_cmd cmd; dcn10_link_encoder_get_max_link_cap(enc, link_settings); - /* in usb c dp2 mode, max lane count is 2 */ - if (enc->funcs->is_in_alt_mode && enc->funcs->is_in_alt_mode(enc)) { - if (enc->ctx->asic_id.hw_internal_rev != YELLOW_CARP_B0) { - // [Note] no need to check hw_internal_rev once phy mux selection is ready - REG_GET(RDPCSTX_PHY_CNTL6, RDPCS_PHY_DPALT_DP4, &is_in_usb_c_dp4_mode); - } else { - if ((enc10->base.transmitter == TRANSMITTER_UNIPHY_A) - || (enc10->base.transmitter == TRANSMITTER_UNIPHY_B) - || (enc10->base.transmitter == TRANSMITTER_UNIPHY_E)) { - REG_GET(RDPCSTX_PHY_CNTL6, RDPCS_PHY_DPALT_DP4, &is_in_usb_c_dp4_mode); - } else { - REG_GET(RDPCSPIPE_PHY_CNTL6, RDPCS_PHY_DPALT_DP4, &is_in_usb_c_dp4_mode); - } - } - if (!is_in_usb_c_dp4_mode) + if (enc->features.flags.bits.DP_IS_USB_C && dc_dmub_srv) { + memset(&cmd, 0, sizeof(cmd)); + cmd.query_dp_alt.header.type = DMUB_CMD__VBIOS; + cmd.query_dp_alt.header.sub_type = DMUB_CMD__VBIOS_TRANSMITTER_QUERY_DP_ALT; + cmd.query_dp_alt.header.payload_bytes = sizeof(cmd.panel_cntl.data); + cmd.query_dp_alt.data.phy_id = phy_id_from_transmitter(enc10->base.transmitter); + + if (!dc_dmub_srv_cmd_with_reply_data(dc_dmub_srv, &cmd)) + return; + + if (cmd.query_dp_alt.data.is_usb && cmd.query_dp_alt.data.is_dp4 == 0) link_settings->lane_count = MIN(LANE_COUNT_TWO, link_settings->lane_count); } } diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_init.c b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_init.c index e175b6cc0125..7a7a8c5edabd 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_init.c +++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_init.c @@ -31,6 +31,8 @@ #include "dcn301/dcn301_hwseq.h" #include "dcn31/dcn31_hwseq.h" +#include "dcn31_init.h" + static const struct hw_sequencer_funcs dcn31_funcs = { .program_gamut_remap = dcn10_program_gamut_remap, .init_hw = dcn31_init_hw, @@ -101,6 +103,7 @@ static const struct hw_sequencer_funcs dcn31_funcs = { .z10_restore = dcn31_z10_restore, .z10_save_init = dcn31_z10_save_init, .set_disp_pattern_generator = dcn30_set_disp_pattern_generator, + .exit_optimized_pwr_state = dcn21_exit_optimized_pwr_state, .update_visual_confirm_color = dcn20_update_visual_confirm_color, }; diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_panel_cntl.c b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_panel_cntl.c index 3b3721386571..83ece02380a8 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_panel_cntl.c +++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_panel_cntl.c @@ -65,7 +65,7 @@ static uint32_t dcn31_get_16_bit_backlight_from_pwm(struct panel_cntl *panel_cnt return cmd.panel_cntl.data.current_backlight; } -uint32_t dcn31_panel_cntl_hw_init(struct panel_cntl *panel_cntl) +static uint32_t dcn31_panel_cntl_hw_init(struct panel_cntl *panel_cntl) { struct dcn31_panel_cntl *dcn31_panel_cntl = TO_DCN31_PANEL_CNTL(panel_cntl); struct dc_dmub_srv *dc_dmub_srv = panel_cntl->ctx->dmub_srv; @@ -96,7 +96,7 @@ uint32_t dcn31_panel_cntl_hw_init(struct panel_cntl *panel_cntl) return cmd.panel_cntl.data.current_backlight; } -void dcn31_panel_cntl_destroy(struct panel_cntl **panel_cntl) +static void dcn31_panel_cntl_destroy(struct panel_cntl **panel_cntl) { struct dcn31_panel_cntl *dcn31_panel_cntl = TO_DCN31_PANEL_CNTL(*panel_cntl); @@ -104,7 +104,7 @@ void dcn31_panel_cntl_destroy(struct panel_cntl **panel_cntl) *panel_cntl = NULL; } -bool dcn31_is_panel_backlight_on(struct panel_cntl *panel_cntl) +static bool dcn31_is_panel_backlight_on(struct panel_cntl *panel_cntl) { union dmub_rb_cmd cmd; @@ -114,7 +114,7 @@ bool dcn31_is_panel_backlight_on(struct panel_cntl *panel_cntl) return cmd.panel_cntl.data.is_backlight_on; } -bool dcn31_is_panel_powered_on(struct panel_cntl *panel_cntl) +static bool dcn31_is_panel_powered_on(struct panel_cntl *panel_cntl) { union dmub_rb_cmd cmd; @@ -124,7 +124,7 @@ bool dcn31_is_panel_powered_on(struct panel_cntl *panel_cntl) return cmd.panel_cntl.data.is_powered_on; } -void dcn31_store_backlight_level(struct panel_cntl *panel_cntl) +static void dcn31_store_backlight_level(struct panel_cntl *panel_cntl) { union dmub_rb_cmd cmd; diff --git a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_resource.c b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_resource.c index 88e040687940..29ee3c22b0ab 100644 --- a/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_resource.c +++ b/drivers/gpu/drm/amd/display/dc/dcn31/dcn31_resource.c @@ -485,7 +485,8 @@ static const struct dcn31_apg_mask apg_mask = { SE_DCN3_REG_LIST(id)\ } -static const struct dcn10_stream_enc_registers stream_enc_regs[] = { +/* Some encoders won't be initialized here - but they're logical, not physical. */ +static const struct dcn10_stream_enc_registers stream_enc_regs[ENGINE_ID_COUNT] = { stream_enc_regs(0), stream_enc_regs(1), stream_enc_regs(2), @@ -968,7 +969,7 @@ static const struct dc_plane_cap plane_cap = { .argb8888 = true, .nv12 = true, .fp16 = true, - .p010 = false, + .p010 = true, .ayuv = false, }, @@ -1023,6 +1024,7 @@ static const struct dc_debug_options debug_defaults_drv = { }, .optimize_edp_link_rate = true, .enable_sw_cntl_psr = true, + .apply_vendor_specific_lttpr_wa = true, }; static const struct dc_debug_options debug_defaults_diags = { @@ -1270,7 +1272,7 @@ static struct link_encoder *dcn31_link_enc_create_minimal( return &enc20->enc10.base; } -struct panel_cntl *dcn31_panel_cntl_create(const struct panel_cntl_init_data *init_data) +static struct panel_cntl *dcn31_panel_cntl_create(const struct panel_cntl_init_data *init_data) { struct dcn31_panel_cntl *panel_cntl = kzalloc(sizeof(struct dcn31_panel_cntl), GFP_KERNEL); @@ -1774,6 +1776,7 @@ static int dcn31_populate_dml_pipes_from_context( int i, pipe_cnt; struct resource_context *res_ctx = &context->res_ctx; struct pipe_ctx *pipe; + bool upscaled = false; dcn20_populate_dml_pipes_from_context(dc, context, pipes, fast_validate); @@ -1785,6 +1788,11 @@ static int dcn31_populate_dml_pipes_from_context( pipe = &res_ctx->pipe_ctx[i]; timing = &pipe->stream->timing; + if (pipe->plane_state && + (pipe->plane_state->src_rect.height < pipe->plane_state->dst_rect.height || + pipe->plane_state->src_rect.width < pipe->plane_state->dst_rect.width)) + upscaled = true; + /* * Immediate flip can be set dynamically after enabling the plane. * We need to require support for immediate flip or underflow can be @@ -1829,6 +1837,11 @@ static int dcn31_populate_dml_pipes_from_context( context->bw_ctx.dml.ip.det_buffer_size_kbytes = 192; pipes[0].pipe.src.unbounded_req_mode = true; } + } else if (context->stream_count >= dc->debug.crb_alloc_policy_min_disp_count + && dc->debug.crb_alloc_policy > DET_SIZE_DEFAULT) { + context->bw_ctx.dml.ip.det_buffer_size_kbytes = dc->debug.crb_alloc_policy * 64; + } else if (context->stream_count >= 3 && upscaled) { + context->bw_ctx.dml.ip.det_buffer_size_kbytes = 192; } return pipe_cnt; @@ -2199,6 +2212,7 @@ static bool dcn31_resource_construct( dc->caps.post_blend_color_processing = true; dc->caps.force_dp_tps4_for_cp2520 = true; dc->caps.dp_hpo = true; + dc->caps.hdmi_frl_pcon_support = true; dc->caps.edp_dsc_support = true; dc->caps.extended_aux_timeout_support = true; dc->caps.dmcub_support = true; @@ -2238,6 +2252,9 @@ static bool dcn31_resource_construct( dc->caps.color.mpc.ogam_rom_caps.hlg = 0; dc->caps.color.mpc.ocsc = 1; + /* Use pipe context based otg sync logic */ + dc->config.use_pipe_ctx_sync_logic = true; + /* read VBIOS LTTPR caps */ { if (ctx->dc_bios->funcs->get_lttpr_caps) { diff --git a/drivers/gpu/drm/amd/display/dc/dm_helpers.h b/drivers/gpu/drm/amd/display/dc/dm_helpers.h index 0fe66b080a03..7f94e3f70d7f 100644 --- a/drivers/gpu/drm/amd/display/dc/dm_helpers.h +++ b/drivers/gpu/drm/amd/display/dc/dm_helpers.h @@ -59,7 +59,7 @@ void dm_helpers_free_gpu_mem( void *pvMem); enum dc_edid_status dm_helpers_parse_edid_caps( - struct dc_context *ctx, + struct dc_link *link, const struct dc_edid *edid, struct dc_edid_caps *edid_caps); diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_rq_dlg_calc_21.c b/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_rq_dlg_calc_21.c index 46c433c0bcb0..8bc27de4c104 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_rq_dlg_calc_21.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn21/display_rq_dlg_calc_21.c @@ -1711,14 +1711,6 @@ void dml21_rq_dlg_get_dlg_reg( dml_print("DML_DLG: Calculation for pipe[%d] end\n", pipe_idx); } -void dml_rq_dlg_get_arb_params(struct display_mode_lib *mode_lib, display_arb_params_st *arb_param) -{ - memset(arb_param, 0, sizeof(*arb_param)); - arb_param->max_req_outstanding = 256; - arb_param->min_req_outstanding = 68; - arb_param->sat_level_us = 60; -} - static void calculate_ttu_cursor( struct display_mode_lib *mode_lib, double *refcyc_per_req_delivery_pre_cur, diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.c b/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.c index 7e937bdcea00..6feb23432f8d 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dcn31/display_mode_vba_31.c @@ -422,62 +422,8 @@ static void CalculateUrgentBurstFactor( static void UseMinimumDCFCLK( struct display_mode_lib *mode_lib, - int MaxInterDCNTileRepeaters, int MaxPrefetchMode, - double FinalDRAMClockChangeLatency, - double SREnterPlusExitTime, - int ReturnBusWidth, - int RoundTripPingLatencyCycles, - int ReorderingBytes, - int PixelChunkSizeInKByte, - int MetaChunkSize, - bool GPUVMEnable, - int GPUVMMaxPageTableLevels, - bool HostVMEnable, - int NumberOfActivePlanes, - double HostVMMinPageSize, - int HostVMMaxNonCachedPageTableLevels, - bool DynamicMetadataVMEnabled, - enum immediate_flip_requirement ImmediateFlipRequirement, - bool ProgressiveToInterlaceUnitInOPP, - double MaxAveragePercentOfIdealFabricAndSDPPortBWDisplayCanUseInNormalSystemOperation, - double PercentOfIdealFabricAndSDPPortBWReceivedAfterUrgLatency, - int VTotal[], - int VActive[], - int DynamicMetadataTransmittedBytes[], - int DynamicMetadataLinesBeforeActiveRequired[], - bool Interlace[], - double RequiredDPPCLK[][2][DC__NUM_DPP__MAX], - double RequiredDISPCLK[][2], - double UrgLatency[], - unsigned int NoOfDPP[][2][DC__NUM_DPP__MAX], - double ProjectedDCFCLKDeepSleep[][2], - double MaximumVStartup[][2][DC__NUM_DPP__MAX], - double TotalVActivePixelBandwidth[][2], - double TotalVActiveCursorBandwidth[][2], - double TotalMetaRowBandwidth[][2], - double TotalDPTERowBandwidth[][2], - unsigned int TotalNumberOfActiveDPP[][2], - unsigned int TotalNumberOfDCCActiveDPP[][2], - int dpte_group_bytes[], - double PrefetchLinesY[][2][DC__NUM_DPP__MAX], - double PrefetchLinesC[][2][DC__NUM_DPP__MAX], - int swath_width_luma_ub_all_states[][2][DC__NUM_DPP__MAX], - int swath_width_chroma_ub_all_states[][2][DC__NUM_DPP__MAX], - int BytePerPixelY[], - int BytePerPixelC[], - int HTotal[], - double PixelClock[], - double PDEAndMetaPTEBytesPerFrame[][2][DC__NUM_DPP__MAX], - double DPTEBytesPerRow[][2][DC__NUM_DPP__MAX], - double MetaRowBytes[][2][DC__NUM_DPP__MAX], - bool DynamicMetadataEnable[], - double VActivePixelBandwidth[][2][DC__NUM_DPP__MAX], - double VActiveCursorBandwidth[][2][DC__NUM_DPP__MAX], - double ReadBandwidthLuma[], - double ReadBandwidthChroma[], - double DCFCLKPerState[], - double DCFCLKState[][2]); + int ReorderingBytes); static void CalculatePixelDeliveryTimes( unsigned int NumberOfActivePlanes, @@ -3949,6 +3895,102 @@ static double TruncToValidBPP( return BPP_INVALID; } +static noinline void CalculatePrefetchSchedulePerPlane( + struct display_mode_lib *mode_lib, + double HostVMInefficiencyFactor, + int i, + unsigned j, + unsigned k) +{ + struct vba_vars_st *v = &mode_lib->vba; + Pipe myPipe; + + myPipe.DPPCLK = v->RequiredDPPCLK[i][j][k]; + myPipe.DISPCLK = v->RequiredDISPCLK[i][j]; + myPipe.PixelClock = v->PixelClock[k]; + myPipe.DCFCLKDeepSleep = v->ProjectedDCFCLKDeepSleep[i][j]; + myPipe.DPPPerPlane = v->NoOfDPP[i][j][k]; + myPipe.ScalerEnabled = v->ScalerEnabled[k]; + myPipe.SourceScan = v->SourceScan[k]; + myPipe.BlockWidth256BytesY = v->Read256BlockWidthY[k]; + myPipe.BlockHeight256BytesY = v->Read256BlockHeightY[k]; + myPipe.BlockWidth256BytesC = v->Read256BlockWidthC[k]; + myPipe.BlockHeight256BytesC = v->Read256BlockHeightC[k]; + myPipe.InterlaceEnable = v->Interlace[k]; + myPipe.NumberOfCursors = v->NumberOfCursors[k]; + myPipe.VBlank = v->VTotal[k] - v->VActive[k]; + myPipe.HTotal = v->HTotal[k]; + myPipe.DCCEnable = v->DCCEnable[k]; + myPipe.ODMCombineIsEnabled = v->ODMCombineEnablePerState[i][k] == dm_odm_combine_mode_4to1 + || v->ODMCombineEnablePerState[i][k] == dm_odm_combine_mode_2to1; + myPipe.SourcePixelFormat = v->SourcePixelFormat[k]; + myPipe.BytePerPixelY = v->BytePerPixelY[k]; + myPipe.BytePerPixelC = v->BytePerPixelC[k]; + myPipe.ProgressiveToInterlaceUnitInOPP = v->ProgressiveToInterlaceUnitInOPP; + v->NoTimeForPrefetch[i][j][k] = CalculatePrefetchSchedule( + mode_lib, + HostVMInefficiencyFactor, + &myPipe, + v->DSCDelayPerState[i][k], + v->DPPCLKDelaySubtotal + v->DPPCLKDelayCNVCFormater, + v->DPPCLKDelaySCL, + v->DPPCLKDelaySCLLBOnly, + v->DPPCLKDelayCNVCCursor, + v->DISPCLKDelaySubtotal, + v->SwathWidthYThisState[k] / v->HRatio[k], + v->OutputFormat[k], + v->MaxInterDCNTileRepeaters, + dml_min(v->MaxVStartup, v->MaximumVStartup[i][j][k]), + v->MaximumVStartup[i][j][k], + v->GPUVMMaxPageTableLevels, + v->GPUVMEnable, + v->HostVMEnable, + v->HostVMMaxNonCachedPageTableLevels, + v->HostVMMinPageSize, + v->DynamicMetadataEnable[k], + v->DynamicMetadataVMEnabled, + v->DynamicMetadataLinesBeforeActiveRequired[k], + v->DynamicMetadataTransmittedBytes[k], + v->UrgLatency[i], + v->ExtraLatency, + v->TimeCalc, + v->PDEAndMetaPTEBytesPerFrame[i][j][k], + v->MetaRowBytes[i][j][k], + v->DPTEBytesPerRow[i][j][k], + v->PrefetchLinesY[i][j][k], + v->SwathWidthYThisState[k], + v->PrefillY[k], + v->MaxNumSwY[k], + v->PrefetchLinesC[i][j][k], + v->SwathWidthCThisState[k], + v->PrefillC[k], + v->MaxNumSwC[k], + v->swath_width_luma_ub_this_state[k], + v->swath_width_chroma_ub_this_state[k], + v->SwathHeightYThisState[k], + v->SwathHeightCThisState[k], + v->TWait, + &v->DSTXAfterScaler[k], + &v->DSTYAfterScaler[k], + &v->LineTimesForPrefetch[k], + &v->PrefetchBW[k], + &v->LinesForMetaPTE[k], + &v->LinesForMetaAndDPTERow[k], + &v->VRatioPreY[i][j][k], + &v->VRatioPreC[i][j][k], + &v->RequiredPrefetchPixelDataBWLuma[i][j][k], + &v->RequiredPrefetchPixelDataBWChroma[i][j][k], + &v->NoTimeForDynamicMetadata[i][j][k], + &v->Tno_bw[k], + &v->prefetch_vmrow_bw[k], + &v->dummy7[k], + &v->dummy8[k], + &v->dummy13[k], + &v->VUpdateOffsetPix[k], + &v->VUpdateWidthPix[k], + &v->VReadyOffsetPix[k]); +} + void dml31_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_lib) { struct vba_vars_st *v = &mode_lib->vba; @@ -5079,66 +5121,8 @@ void dml31_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l } } - if (v->UseMinimumRequiredDCFCLK == true) { - UseMinimumDCFCLK( - mode_lib, - v->MaxInterDCNTileRepeaters, - MaxPrefetchMode, - v->DRAMClockChangeLatency, - v->SREnterPlusExitTime, - v->ReturnBusWidth, - v->RoundTripPingLatencyCycles, - ReorderingBytes, - v->PixelChunkSizeInKByte, - v->MetaChunkSize, - v->GPUVMEnable, - v->GPUVMMaxPageTableLevels, - v->HostVMEnable, - v->NumberOfActivePlanes, - v->HostVMMinPageSize, - v->HostVMMaxNonCachedPageTableLevels, - v->DynamicMetadataVMEnabled, - v->ImmediateFlipRequirement[0], - v->ProgressiveToInterlaceUnitInOPP, - v->MaxAveragePercentOfIdealFabricAndSDPPortBWDisplayCanUseInNormalSystemOperation, - v->PercentOfIdealFabricAndSDPPortBWReceivedAfterUrgLatency, - v->VTotal, - v->VActive, - v->DynamicMetadataTransmittedBytes, - v->DynamicMetadataLinesBeforeActiveRequired, - v->Interlace, - v->RequiredDPPCLK, - v->RequiredDISPCLK, - v->UrgLatency, - v->NoOfDPP, - v->ProjectedDCFCLKDeepSleep, - v->MaximumVStartup, - v->TotalVActivePixelBandwidth, - v->TotalVActiveCursorBandwidth, - v->TotalMetaRowBandwidth, - v->TotalDPTERowBandwidth, - v->TotalNumberOfActiveDPP, - v->TotalNumberOfDCCActiveDPP, - v->dpte_group_bytes, - v->PrefetchLinesY, - v->PrefetchLinesC, - v->swath_width_luma_ub_all_states, - v->swath_width_chroma_ub_all_states, - v->BytePerPixelY, - v->BytePerPixelC, - v->HTotal, - v->PixelClock, - v->PDEAndMetaPTEBytesPerFrame, - v->DPTEBytesPerRow, - v->MetaRowBytes, - v->DynamicMetadataEnable, - v->VActivePixelBandwidth, - v->VActiveCursorBandwidth, - v->ReadBandwidthLuma, - v->ReadBandwidthChroma, - v->DCFCLKPerState, - v->DCFCLKState); - } + if (v->UseMinimumRequiredDCFCLK == true) + UseMinimumDCFCLK(mode_lib, MaxPrefetchMode, ReorderingBytes); for (i = 0; i < v->soc.num_states; ++i) { for (j = 0; j <= 1; ++j) { @@ -5276,92 +5260,9 @@ void dml31_ModeSupportAndSystemConfigurationFull(struct display_mode_lib *mode_l v->SREnterPlusExitTime); for (k = 0; k < v->NumberOfActivePlanes; k++) { - Pipe myPipe; - - myPipe.DPPCLK = v->RequiredDPPCLK[i][j][k]; - myPipe.DISPCLK = v->RequiredDISPCLK[i][j]; - myPipe.PixelClock = v->PixelClock[k]; - myPipe.DCFCLKDeepSleep = v->ProjectedDCFCLKDeepSleep[i][j]; - myPipe.DPPPerPlane = v->NoOfDPP[i][j][k]; - myPipe.ScalerEnabled = v->ScalerEnabled[k]; - myPipe.SourceScan = v->SourceScan[k]; - myPipe.BlockWidth256BytesY = v->Read256BlockWidthY[k]; - myPipe.BlockHeight256BytesY = v->Read256BlockHeightY[k]; - myPipe.BlockWidth256BytesC = v->Read256BlockWidthC[k]; - myPipe.BlockHeight256BytesC = v->Read256BlockHeightC[k]; - myPipe.InterlaceEnable = v->Interlace[k]; - myPipe.NumberOfCursors = v->NumberOfCursors[k]; - myPipe.VBlank = v->VTotal[k] - v->VActive[k]; - myPipe.HTotal = v->HTotal[k]; - myPipe.DCCEnable = v->DCCEnable[k]; - myPipe.ODMCombineIsEnabled = v->ODMCombineEnablePerState[i][k] == dm_odm_combine_mode_4to1 - || v->ODMCombineEnablePerState[i][k] == dm_odm_combine_mode_2to1; - myPipe.SourcePixelFormat = v->SourcePixelFormat[k]; - myPipe.BytePerPixelY = v->BytePerPixelY[k]; - myPipe.BytePerPixelC = v->BytePerPixelC[k]; - myPipe.ProgressiveToInterlaceUnitInOPP = v->ProgressiveToInterlaceUnitInOPP; - v->NoTimeForPrefetch[i][j][k] = CalculatePrefetchSchedule( - mode_lib, - HostVMInefficiencyFactor, - &myPipe, - v->DSCDelayPerState[i][k], - v->DPPCLKDelaySubtotal + v->DPPCLKDelayCNVCFormater, - v->DPPCLKDelaySCL, - v->DPPCLKDelaySCLLBOnly, - v->DPPCLKDelayCNVCCursor, - v->DISPCLKDelaySubtotal, - v->SwathWidthYThisState[k] / v->HRatio[k], - v->OutputFormat[k], - v->MaxInterDCNTileRepeaters, - dml_min(v->MaxVStartup, v->MaximumVStartup[i][j][k]), - v->MaximumVStartup[i][j][k], - v->GPUVMMaxPageTableLevels, - v->GPUVMEnable, - v->HostVMEnable, - v->HostVMMaxNonCachedPageTableLevels, - v->HostVMMinPageSize, - v->DynamicMetadataEnable[k], - v->DynamicMetadataVMEnabled, - v->DynamicMetadataLinesBeforeActiveRequired[k], - v->DynamicMetadataTransmittedBytes[k], - v->UrgLatency[i], - v->ExtraLatency, - v->TimeCalc, - v->PDEAndMetaPTEBytesPerFrame[i][j][k], - v->MetaRowBytes[i][j][k], - v->DPTEBytesPerRow[i][j][k], - v->PrefetchLinesY[i][j][k], - v->SwathWidthYThisState[k], - v->PrefillY[k], - v->MaxNumSwY[k], - v->PrefetchLinesC[i][j][k], - v->SwathWidthCThisState[k], - v->PrefillC[k], - v->MaxNumSwC[k], - v->swath_width_luma_ub_this_state[k], - v->swath_width_chroma_ub_this_state[k], - v->SwathHeightYThisState[k], - v->SwathHeightCThisState[k], - v->TWait, - &v->DSTXAfterScaler[k], - &v->DSTYAfterScaler[k], - &v->LineTimesForPrefetch[k], - &v->PrefetchBW[k], - &v->LinesForMetaPTE[k], - &v->LinesForMetaAndDPTERow[k], - &v->VRatioPreY[i][j][k], - &v->VRatioPreC[i][j][k], - &v->RequiredPrefetchPixelDataBWLuma[i][j][k], - &v->RequiredPrefetchPixelDataBWChroma[i][j][k], - &v->NoTimeForDynamicMetadata[i][j][k], - &v->Tno_bw[k], - &v->prefetch_vmrow_bw[k], - &v->dummy7[k], - &v->dummy8[k], - &v->dummy13[k], - &v->VUpdateOffsetPix[k], - &v->VUpdateWidthPix[k], - &v->VReadyOffsetPix[k]); + CalculatePrefetchSchedulePerPlane(mode_lib, + HostVMInefficiencyFactor, + i, j, k); } for (k = 0; k < v->NumberOfActivePlanes; k++) { @@ -7249,69 +7150,15 @@ static double CalculateUrgentLatency( static void UseMinimumDCFCLK( struct display_mode_lib *mode_lib, - int MaxInterDCNTileRepeaters, int MaxPrefetchMode, - double FinalDRAMClockChangeLatency, - double SREnterPlusExitTime, - int ReturnBusWidth, - int RoundTripPingLatencyCycles, - int ReorderingBytes, - int PixelChunkSizeInKByte, - int MetaChunkSize, - bool GPUVMEnable, - int GPUVMMaxPageTableLevels, - bool HostVMEnable, - int NumberOfActivePlanes, - double HostVMMinPageSize, - int HostVMMaxNonCachedPageTableLevels, - bool DynamicMetadataVMEnabled, - enum immediate_flip_requirement ImmediateFlipRequirement, - bool ProgressiveToInterlaceUnitInOPP, - double MaxAveragePercentOfIdealFabricAndSDPPortBWDisplayCanUseInNormalSystemOperation, - double PercentOfIdealFabricAndSDPPortBWReceivedAfterUrgLatency, - int VTotal[], - int VActive[], - int DynamicMetadataTransmittedBytes[], - int DynamicMetadataLinesBeforeActiveRequired[], - bool Interlace[], - double RequiredDPPCLK[][2][DC__NUM_DPP__MAX], - double RequiredDISPCLK[][2], - double UrgLatency[], - unsigned int NoOfDPP[][2][DC__NUM_DPP__MAX], - double ProjectedDCFCLKDeepSleep[][2], - double MaximumVStartup[][2][DC__NUM_DPP__MAX], - double TotalVActivePixelBandwidth[][2], - double TotalVActiveCursorBandwidth[][2], - double TotalMetaRowBandwidth[][2], - double TotalDPTERowBandwidth[][2], - unsigned int TotalNumberOfActiveDPP[][2], - unsigned int TotalNumberOfDCCActiveDPP[][2], - int dpte_group_bytes[], - double PrefetchLinesY[][2][DC__NUM_DPP__MAX], - double PrefetchLinesC[][2][DC__NUM_DPP__MAX], - int swath_width_luma_ub_all_states[][2][DC__NUM_DPP__MAX], - int swath_width_chroma_ub_all_states[][2][DC__NUM_DPP__MAX], - int BytePerPixelY[], - int BytePerPixelC[], - int HTotal[], - double PixelClock[], - double PDEAndMetaPTEBytesPerFrame[][2][DC__NUM_DPP__MAX], - double DPTEBytesPerRow[][2][DC__NUM_DPP__MAX], - double MetaRowBytes[][2][DC__NUM_DPP__MAX], - bool DynamicMetadataEnable[], - double VActivePixelBandwidth[][2][DC__NUM_DPP__MAX], - double VActiveCursorBandwidth[][2][DC__NUM_DPP__MAX], - double ReadBandwidthLuma[], - double ReadBandwidthChroma[], - double DCFCLKPerState[], - double DCFCLKState[][2]) + int ReorderingBytes) { struct vba_vars_st *v = &mode_lib->vba; int dummy1, i, j, k; double NormalEfficiency, dummy2, dummy3; double TotalMaxPrefetchFlipDPTERowBandwidth[DC__VOLTAGE_STATES][2]; - NormalEfficiency = PercentOfIdealFabricAndSDPPortBWReceivedAfterUrgLatency / 100.0; + NormalEfficiency = v->PercentOfIdealFabricAndSDPPortBWReceivedAfterUrgLatency / 100.0; for (i = 0; i < v->soc.num_states; ++i) { for (j = 0; j <= 1; ++j) { double PixelDCFCLKCyclesRequiredInPrefetch[DC__NUM_DPP__MAX]; @@ -7329,61 +7176,61 @@ static void UseMinimumDCFCLK( double MinimumTvmPlus2Tr0; TotalMaxPrefetchFlipDPTERowBandwidth[i][j] = 0; - for (k = 0; k < NumberOfActivePlanes; ++k) { + for (k = 0; k < v->NumberOfActivePlanes; ++k) { TotalMaxPrefetchFlipDPTERowBandwidth[i][j] = TotalMaxPrefetchFlipDPTERowBandwidth[i][j] - + NoOfDPP[i][j][k] * DPTEBytesPerRow[i][j][k] / (15.75 * HTotal[k] / PixelClock[k]); + + v->NoOfDPP[i][j][k] * v->DPTEBytesPerRow[i][j][k] / (15.75 * v->HTotal[k] / v->PixelClock[k]); } - for (k = 0; k <= NumberOfActivePlanes - 1; ++k) { - NoOfDPPState[k] = NoOfDPP[i][j][k]; + for (k = 0; k <= v->NumberOfActivePlanes - 1; ++k) { + NoOfDPPState[k] = v->NoOfDPP[i][j][k]; } - MinimumTWait = CalculateTWait(MaxPrefetchMode, FinalDRAMClockChangeLatency, UrgLatency[i], SREnterPlusExitTime); - NonDPTEBandwidth = TotalVActivePixelBandwidth[i][j] + TotalVActiveCursorBandwidth[i][j] + TotalMetaRowBandwidth[i][j]; - DPTEBandwidth = (HostVMEnable == true || ImmediateFlipRequirement == dm_immediate_flip_required) ? - TotalMaxPrefetchFlipDPTERowBandwidth[i][j] : TotalDPTERowBandwidth[i][j]; + MinimumTWait = CalculateTWait(MaxPrefetchMode, v->FinalDRAMClockChangeLatency, v->UrgLatency[i], v->SREnterPlusExitTime); + NonDPTEBandwidth = v->TotalVActivePixelBandwidth[i][j] + v->TotalVActiveCursorBandwidth[i][j] + v->TotalMetaRowBandwidth[i][j]; + DPTEBandwidth = (v->HostVMEnable == true || v->ImmediateFlipRequirement[0] == dm_immediate_flip_required) ? + TotalMaxPrefetchFlipDPTERowBandwidth[i][j] : v->TotalDPTERowBandwidth[i][j]; DCFCLKRequiredForAverageBandwidth = dml_max3( - ProjectedDCFCLKDeepSleep[i][j], - (NonDPTEBandwidth + TotalDPTERowBandwidth[i][j]) / ReturnBusWidth - / (MaxAveragePercentOfIdealFabricAndSDPPortBWDisplayCanUseInNormalSystemOperation / 100), - (NonDPTEBandwidth + DPTEBandwidth / NormalEfficiency) / NormalEfficiency / ReturnBusWidth); + v->ProjectedDCFCLKDeepSleep[i][j], + (NonDPTEBandwidth + v->TotalDPTERowBandwidth[i][j]) / v->ReturnBusWidth + / (v->MaxAveragePercentOfIdealFabricAndSDPPortBWDisplayCanUseInNormalSystemOperation / 100), + (NonDPTEBandwidth + DPTEBandwidth / NormalEfficiency) / NormalEfficiency / v->ReturnBusWidth); ExtraLatencyBytes = CalculateExtraLatencyBytes( ReorderingBytes, - TotalNumberOfActiveDPP[i][j], - PixelChunkSizeInKByte, - TotalNumberOfDCCActiveDPP[i][j], - MetaChunkSize, - GPUVMEnable, - HostVMEnable, - NumberOfActivePlanes, + v->TotalNumberOfActiveDPP[i][j], + v->PixelChunkSizeInKByte, + v->TotalNumberOfDCCActiveDPP[i][j], + v->MetaChunkSize, + v->GPUVMEnable, + v->HostVMEnable, + v->NumberOfActivePlanes, NoOfDPPState, - dpte_group_bytes, + v->dpte_group_bytes, 1, - HostVMMinPageSize, - HostVMMaxNonCachedPageTableLevels); - ExtraLatencyCycles = RoundTripPingLatencyCycles + __DML_ARB_TO_RET_DELAY__ + ExtraLatencyBytes / NormalEfficiency / ReturnBusWidth; - for (k = 0; k < NumberOfActivePlanes; ++k) { + v->HostVMMinPageSize, + v->HostVMMaxNonCachedPageTableLevels); + ExtraLatencyCycles = v->RoundTripPingLatencyCycles + __DML_ARB_TO_RET_DELAY__ + ExtraLatencyBytes / NormalEfficiency / v->ReturnBusWidth; + for (k = 0; k < v->NumberOfActivePlanes; ++k) { double DCFCLKCyclesRequiredInPrefetch; double ExpectedPrefetchBWAcceleration; double PrefetchTime; - PixelDCFCLKCyclesRequiredInPrefetch[k] = (PrefetchLinesY[i][j][k] * swath_width_luma_ub_all_states[i][j][k] * BytePerPixelY[k] - + PrefetchLinesC[i][j][k] * swath_width_chroma_ub_all_states[i][j][k] * BytePerPixelC[k]) / NormalEfficiency / ReturnBusWidth; + PixelDCFCLKCyclesRequiredInPrefetch[k] = (v->PrefetchLinesY[i][j][k] * v->swath_width_luma_ub_all_states[i][j][k] * v->BytePerPixelY[k] + + v->PrefetchLinesC[i][j][k] * v->swath_width_chroma_ub_all_states[i][j][k] * v->BytePerPixelC[k]) / NormalEfficiency / v->ReturnBusWidth; DCFCLKCyclesRequiredInPrefetch = 2 * ExtraLatencyCycles / NoOfDPPState[k] - + PDEAndMetaPTEBytesPerFrame[i][j][k] / NormalEfficiency / NormalEfficiency / ReturnBusWidth * (GPUVMMaxPageTableLevels > 2 ? 1 : 0) - + 2 * DPTEBytesPerRow[i][j][k] / NormalEfficiency / NormalEfficiency / ReturnBusWidth - + 2 * MetaRowBytes[i][j][k] / NormalEfficiency / ReturnBusWidth + PixelDCFCLKCyclesRequiredInPrefetch[k]; - PrefetchPixelLinesTime[k] = dml_max(PrefetchLinesY[i][j][k], PrefetchLinesC[i][j][k]) * HTotal[k] / PixelClock[k]; - ExpectedPrefetchBWAcceleration = (VActivePixelBandwidth[i][j][k] + VActiveCursorBandwidth[i][j][k]) - / (ReadBandwidthLuma[k] + ReadBandwidthChroma[k]); + + v->PDEAndMetaPTEBytesPerFrame[i][j][k] / NormalEfficiency / NormalEfficiency / v->ReturnBusWidth * (v->GPUVMMaxPageTableLevels > 2 ? 1 : 0) + + 2 * v->DPTEBytesPerRow[i][j][k] / NormalEfficiency / NormalEfficiency / v->ReturnBusWidth + + 2 * v->MetaRowBytes[i][j][k] / NormalEfficiency / v->ReturnBusWidth + PixelDCFCLKCyclesRequiredInPrefetch[k]; + PrefetchPixelLinesTime[k] = dml_max(v->PrefetchLinesY[i][j][k], v->PrefetchLinesC[i][j][k]) * v->HTotal[k] / v->PixelClock[k]; + ExpectedPrefetchBWAcceleration = (v->VActivePixelBandwidth[i][j][k] + v->VActiveCursorBandwidth[i][j][k]) + / (v->ReadBandwidthLuma[k] + v->ReadBandwidthChroma[k]); DynamicMetadataVMExtraLatency[k] = - (GPUVMEnable == true && DynamicMetadataEnable[k] == true && DynamicMetadataVMEnabled == true) ? - UrgLatency[i] * GPUVMMaxPageTableLevels * (HostVMEnable == true ? HostVMMaxNonCachedPageTableLevels + 1 : 1) : 0; - PrefetchTime = (MaximumVStartup[i][j][k] - 1) * HTotal[k] / PixelClock[k] - MinimumTWait - - UrgLatency[i] - * ((GPUVMMaxPageTableLevels <= 2 ? GPUVMMaxPageTableLevels : GPUVMMaxPageTableLevels - 2) - * (HostVMEnable == true ? HostVMMaxNonCachedPageTableLevels + 1 : 1) - 1) + (v->GPUVMEnable == true && v->DynamicMetadataEnable[k] == true && v->DynamicMetadataVMEnabled == true) ? + v->UrgLatency[i] * v->GPUVMMaxPageTableLevels * (v->HostVMEnable == true ? v->HostVMMaxNonCachedPageTableLevels + 1 : 1) : 0; + PrefetchTime = (v->MaximumVStartup[i][j][k] - 1) * v->HTotal[k] / v->PixelClock[k] - MinimumTWait + - v->UrgLatency[i] + * ((v->GPUVMMaxPageTableLevels <= 2 ? v->GPUVMMaxPageTableLevels : v->GPUVMMaxPageTableLevels - 2) + * (v->HostVMEnable == true ? v->HostVMMaxNonCachedPageTableLevels + 1 : 1) - 1) - DynamicMetadataVMExtraLatency[k]; if (PrefetchTime > 0) { @@ -7392,14 +7239,14 @@ static void UseMinimumDCFCLK( / (PrefetchTime * PixelDCFCLKCyclesRequiredInPrefetch[k] / DCFCLKCyclesRequiredInPrefetch); DCFCLKRequiredForPeakBandwidthPerPlane[k] = NoOfDPPState[k] * PixelDCFCLKCyclesRequiredInPrefetch[k] / PrefetchPixelLinesTime[k] * dml_max(1.0, ExpectedVRatioPrefetch) * dml_max(1.0, ExpectedVRatioPrefetch / 4) * ExpectedPrefetchBWAcceleration; - if (HostVMEnable == true || ImmediateFlipRequirement == dm_immediate_flip_required) { + if (v->HostVMEnable == true || v->ImmediateFlipRequirement[0] == dm_immediate_flip_required) { DCFCLKRequiredForPeakBandwidthPerPlane[k] = DCFCLKRequiredForPeakBandwidthPerPlane[k] - + NoOfDPPState[k] * DPTEBandwidth / NormalEfficiency / NormalEfficiency / ReturnBusWidth; + + NoOfDPPState[k] * DPTEBandwidth / NormalEfficiency / NormalEfficiency / v->ReturnBusWidth; } } else { - DCFCLKRequiredForPeakBandwidthPerPlane[k] = DCFCLKPerState[i]; + DCFCLKRequiredForPeakBandwidthPerPlane[k] = v->DCFCLKPerState[i]; } - if (DynamicMetadataEnable[k] == true) { + if (v->DynamicMetadataEnable[k] == true) { double TSetupPipe; double TdmbfPipe; double TdmsksPipe; @@ -7407,17 +7254,17 @@ static void UseMinimumDCFCLK( double AllowedTimeForUrgentExtraLatency; CalculateVupdateAndDynamicMetadataParameters( - MaxInterDCNTileRepeaters, - RequiredDPPCLK[i][j][k], - RequiredDISPCLK[i][j], - ProjectedDCFCLKDeepSleep[i][j], - PixelClock[k], - HTotal[k], - VTotal[k] - VActive[k], - DynamicMetadataTransmittedBytes[k], - DynamicMetadataLinesBeforeActiveRequired[k], - Interlace[k], - ProgressiveToInterlaceUnitInOPP, + v->MaxInterDCNTileRepeaters, + v->RequiredDPPCLK[i][j][k], + v->RequiredDISPCLK[i][j], + v->ProjectedDCFCLKDeepSleep[i][j], + v->PixelClock[k], + v->HTotal[k], + v->VTotal[k] - v->VActive[k], + v->DynamicMetadataTransmittedBytes[k], + v->DynamicMetadataLinesBeforeActiveRequired[k], + v->Interlace[k], + v->ProgressiveToInterlaceUnitInOPP, &TSetupPipe, &TdmbfPipe, &TdmecPipe, @@ -7425,31 +7272,31 @@ static void UseMinimumDCFCLK( &dummy1, &dummy2, &dummy3); - AllowedTimeForUrgentExtraLatency = MaximumVStartup[i][j][k] * HTotal[k] / PixelClock[k] - MinimumTWait - TSetupPipe - TdmbfPipe - TdmecPipe + AllowedTimeForUrgentExtraLatency = v->MaximumVStartup[i][j][k] * v->HTotal[k] / v->PixelClock[k] - MinimumTWait - TSetupPipe - TdmbfPipe - TdmecPipe - TdmsksPipe - DynamicMetadataVMExtraLatency[k]; if (AllowedTimeForUrgentExtraLatency > 0) { DCFCLKRequiredForPeakBandwidthPerPlane[k] = dml_max( DCFCLKRequiredForPeakBandwidthPerPlane[k], ExtraLatencyCycles / AllowedTimeForUrgentExtraLatency); } else { - DCFCLKRequiredForPeakBandwidthPerPlane[k] = DCFCLKPerState[i]; + DCFCLKRequiredForPeakBandwidthPerPlane[k] = v->DCFCLKPerState[i]; } } } DCFCLKRequiredForPeakBandwidth = 0; - for (k = 0; k <= NumberOfActivePlanes - 1; ++k) { + for (k = 0; k <= v->NumberOfActivePlanes - 1; ++k) { DCFCLKRequiredForPeakBandwidth = DCFCLKRequiredForPeakBandwidth + DCFCLKRequiredForPeakBandwidthPerPlane[k]; } - MinimumTvmPlus2Tr0 = UrgLatency[i] - * (GPUVMEnable == true ? - (HostVMEnable == true ? - (GPUVMMaxPageTableLevels + 2) * (HostVMMaxNonCachedPageTableLevels + 1) - 1 : GPUVMMaxPageTableLevels + 1) : + MinimumTvmPlus2Tr0 = v->UrgLatency[i] + * (v->GPUVMEnable == true ? + (v->HostVMEnable == true ? + (v->GPUVMMaxPageTableLevels + 2) * (v->HostVMMaxNonCachedPageTableLevels + 1) - 1 : v->GPUVMMaxPageTableLevels + 1) : 0); - for (k = 0; k < NumberOfActivePlanes; ++k) { + for (k = 0; k < v->NumberOfActivePlanes; ++k) { double MaximumTvmPlus2Tr0PlusTsw; - MaximumTvmPlus2Tr0PlusTsw = (MaximumVStartup[i][j][k] - 2) * HTotal[k] / PixelClock[k] - MinimumTWait - DynamicMetadataVMExtraLatency[k]; + MaximumTvmPlus2Tr0PlusTsw = (v->MaximumVStartup[i][j][k] - 2) * v->HTotal[k] / v->PixelClock[k] - MinimumTWait - DynamicMetadataVMExtraLatency[k]; if (MaximumTvmPlus2Tr0PlusTsw <= MinimumTvmPlus2Tr0 + PrefetchPixelLinesTime[k] / 4) { - DCFCLKRequiredForPeakBandwidth = DCFCLKPerState[i]; + DCFCLKRequiredForPeakBandwidth = v->DCFCLKPerState[i]; } else { DCFCLKRequiredForPeakBandwidth = dml_max3( DCFCLKRequiredForPeakBandwidth, @@ -7457,7 +7304,7 @@ static void UseMinimumDCFCLK( (2 * ExtraLatencyCycles + PixelDCFCLKCyclesRequiredInPrefetch[k]) / (MaximumTvmPlus2Tr0PlusTsw - MinimumTvmPlus2Tr0)); } } - DCFCLKState[i][j] = dml_min(DCFCLKPerState[i], 1.05 * dml_max(DCFCLKRequiredForAverageBandwidth, DCFCLKRequiredForPeakBandwidth)); + v->DCFCLKState[i][j] = dml_min(v->DCFCLKPerState[i], 1.05 * dml_max(DCFCLKRequiredForAverageBandwidth, DCFCLKRequiredForPeakBandwidth)); } } } diff --git a/drivers/gpu/drm/amd/display/dc/dml/dml_wrapper.c b/drivers/gpu/drm/amd/display/dc/dml/dml_wrapper.c index ece34b0b8a46..91810aaee5a3 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dml_wrapper.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dml_wrapper.c @@ -1223,8 +1223,8 @@ static void dml_full_validate_bw_helper(struct dc *dc, *pipe_cnt = dml_populate_dml_pipes_from_context(dc, context, pipes, false); *vlevel = dml_get_voltage_level(&context->bw_ctx.dml, pipes, *pipe_cnt); if (*vlevel < context->bw_ctx.dml.soc.num_states) { - memset(split, 0, sizeof(split)); - memset(merge, 0, sizeof(merge)); + memset(split, 0, MAX_PIPES * sizeof(*split)); + memset(merge, 0, MAX_PIPES * sizeof(*merge)); *vlevel = dml_validate_apply_pipe_split_flags(dc, context, *vlevel, split, merge); } diff --git a/drivers/gpu/drm/amd/display/dc/dml/dsc/rc_calc_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dsc/rc_calc_fpu.c index 122ba291a7ef..ec636d06e18c 100644 --- a/drivers/gpu/drm/amd/display/dc/dml/dsc/rc_calc_fpu.c +++ b/drivers/gpu/drm/amd/display/dc/dml/dsc/rc_calc_fpu.c @@ -93,7 +93,7 @@ static void get_qp_set(qp_set qps, enum colour_mode cm, enum bits_per_comp bpc, TABLE_CASE(420, 12, min); } - if (table == 0) + if (!table) return; index = (bpp - table[0].bpp) * 2; diff --git a/drivers/gpu/drm/amd/display/dc/inc/core_types.h b/drivers/gpu/drm/amd/display/dc/inc/core_types.h index 6fc6488c54c0..f3c0e70073da 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/core_types.h +++ b/drivers/gpu/drm/amd/display/dc/inc/core_types.h @@ -367,6 +367,7 @@ struct pipe_ctx { struct pll_settings pll_settings; uint8_t pipe_idx; + uint8_t pipe_idx_syncd; struct pipe_ctx *top_pipe; struct pipe_ctx *bottom_pipe; diff --git a/drivers/gpu/drm/amd/display/dc/inc/dc_link_dp.h b/drivers/gpu/drm/amd/display/dc/inc/dc_link_dp.h index a6d3d859754a..52bdfea7897b 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/dc_link_dp.h +++ b/drivers/gpu/drm/amd/display/dc/inc/dc_link_dp.h @@ -170,6 +170,7 @@ uint8_t dc_dp_initialize_scrambling_data_symbols( enum dc_status dp_set_fec_ready(struct dc_link *link, bool ready); void dp_set_fec_enable(struct dc_link *link, bool enable); +struct link_encoder *dp_get_link_enc(struct dc_link *link); bool dp_set_dsc_enable(struct pipe_ctx *pipe_ctx, bool enable); bool dp_set_dsc_pps_sdp(struct pipe_ctx *pipe_ctx, bool enable, bool immediate_update); void dp_set_dsc_on_stream(struct pipe_ctx *pipe_ctx, bool enable); @@ -217,4 +218,5 @@ bool is_dp_128b_132b_signal(struct pipe_ctx *pipe_ctx); void reset_dp_hpo_stream_encoders_for_link(struct dc_link *link); bool dp_retrieve_lttpr_cap(struct dc_link *link); +void edp_panel_backlight_power_on(struct dc_link *link); #endif /* __DC_LINK_DP_H__ */ diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h b/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h index a17e5de3b100..c920c4b6077d 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/clk_mgr.h @@ -211,6 +211,8 @@ struct dummy_pstate_entry { struct clk_bw_params { unsigned int vram_type; unsigned int num_channels; + unsigned int dispclk_vco_khz; + unsigned int dc_mode_softmax_memclk; struct clk_limit_table clk_table; struct wm_table wm_table; struct dummy_pstate_entry dummy_pstate_table[4]; @@ -261,6 +263,10 @@ struct clk_mgr_funcs { /* Send message to PMFW to set hard max memclk frequency to highest DPM */ void (*set_hard_max_memclk)(struct clk_mgr *clk_mgr); + /* Custom set a memclk freq range*/ + void (*set_max_memclk)(struct clk_mgr *clk_mgr, unsigned int memclk_mhz); + void (*set_min_memclk)(struct clk_mgr *clk_mgr, unsigned int memclk_mhz); + /* Get current memclk states from PMFW, update relevant structures */ void (*get_memclk_states_from_smu)(struct clk_mgr *clk_mgr); @@ -274,6 +280,7 @@ struct clk_mgr { struct dc_clocks clks; bool psr_allow_active_cache; bool force_smu_not_present; + bool dc_mode_softmax_enabled; int dprefclk_khz; // Used by program pixel clock in clock source funcs, need to figureout where this goes int dentist_vco_freq_khz; struct clk_state_registers_and_bypass boot_snapshot; diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h b/drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h index 80e1a32bc63d..2c031586f4e6 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw/hubp.h @@ -139,6 +139,7 @@ struct hubp_funcs { bool (*hubp_is_flip_pending)(struct hubp *hubp); void (*set_blank)(struct hubp *hubp, bool blank); + void (*set_blank_regs)(struct hubp *hubp, bool blank); void (*set_hubp_blank_en)(struct hubp *hubp, bool blank); void (*set_cursor_attributes)( diff --git a/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h b/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h index d50f4bd06b5d..05053f3b4ab7 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h +++ b/drivers/gpu/drm/amd/display/dc/inc/hw_sequencer.h @@ -64,6 +64,7 @@ struct hw_sequencer_funcs { enum dc_status (*apply_ctx_to_hw)(struct dc *dc, struct dc_state *context); void (*disable_plane)(struct dc *dc, struct pipe_ctx *pipe_ctx); + void (*disable_pixel_data)(struct dc *dc, struct pipe_ctx *pipe_ctx, bool blank); void (*apply_ctx_for_surface)(struct dc *dc, const struct dc_stream_state *stream, int num_planes, struct dc_state *context); diff --git a/drivers/gpu/drm/amd/display/dc/inc/resource.h b/drivers/gpu/drm/amd/display/dc/inc/resource.h index 372c0898facd..c208925f8247 100644 --- a/drivers/gpu/drm/amd/display/dc/inc/resource.h +++ b/drivers/gpu/drm/amd/display/dc/inc/resource.h @@ -34,6 +34,10 @@ #define MEMORY_TYPE_HBM 2 +#define IS_PIPE_SYNCD_VALID(pipe) ((((pipe)->pipe_idx_syncd) & 0x80)?1:0) +#define GET_PIPE_SYNCD_FROM_PIPE(pipe) ((pipe)->pipe_idx_syncd & 0x7F) +#define SET_PIPE_SYNCD_TO_PIPE(pipe, pipe_syncd) ((pipe)->pipe_idx_syncd = (0x80 | pipe_syncd)) + enum dce_version resource_parse_asic_id( struct hw_asic_id asic_id); @@ -206,4 +210,11 @@ struct hpo_dp_link_encoder *resource_get_unused_hpo_dp_link_encoder( const struct resource_pool *pool); #endif +void reset_syncd_pipes_from_disabled_pipes(struct dc *dc, + struct dc_state *context); + +void check_syncd_pipes_for_disabled_master_pipe(struct dc *dc, + struct dc_state *context, + uint8_t disabled_master_pipe_idx); + #endif /* DRIVERS_GPU_DRM_AMD_DC_DEV_DC_INC_RESOURCE_H_ */ diff --git a/drivers/gpu/drm/amd/display/dc/irq/dce110/irq_service_dce110.c b/drivers/gpu/drm/amd/display/dc/irq/dce110/irq_service_dce110.c index 378cc11aa047..6b5fedd9ace0 100644 --- a/drivers/gpu/drm/amd/display/dc/irq/dce110/irq_service_dce110.c +++ b/drivers/gpu/drm/amd/display/dc/irq/dce110/irq_service_dce110.c @@ -185,16 +185,18 @@ bool dal_irq_service_dummy_set(struct irq_service *irq_service, const struct irq_source_info *info, bool enable) { - DC_LOG_ERROR("%s: called for non-implemented irq source\n", - __func__); + DC_LOG_ERROR("%s: called for non-implemented irq source, src_id=%u, ext_id=%u\n", + __func__, info->src_id, info->ext_id); + return false; } bool dal_irq_service_dummy_ack(struct irq_service *irq_service, const struct irq_source_info *info) { - DC_LOG_ERROR("%s: called for non-implemented irq source\n", - __func__); + DC_LOG_ERROR("%s: called for non-implemented irq source, src_id=%u, ext_id=%u\n", + __func__, info->src_id, info->ext_id); + return false; } diff --git a/drivers/gpu/drm/amd/display/dc/irq/dcn10/irq_service_dcn10.c b/drivers/gpu/drm/amd/display/dc/irq/dcn10/irq_service_dcn10.c index 34f43cb650f8..cf072e2347d3 100644 --- a/drivers/gpu/drm/amd/display/dc/irq/dcn10/irq_service_dcn10.c +++ b/drivers/gpu/drm/amd/display/dc/irq/dcn10/irq_service_dcn10.c @@ -40,10 +40,9 @@ #include "ivsrcid/dcn/irqsrcs_dcn_1_0.h" -enum dc_irq_source to_dal_irq_source_dcn10( - struct irq_service *irq_service, - uint32_t src_id, - uint32_t ext_id) +static enum dc_irq_source to_dal_irq_source_dcn10(struct irq_service *irq_service, + uint32_t src_id, + uint32_t ext_id) { switch (src_id) { case DCN_1_0__SRCID__DC_D1_OTG_VSTARTUP: diff --git a/drivers/gpu/drm/amd/display/dc/irq/dcn201/irq_service_dcn201.c b/drivers/gpu/drm/amd/display/dc/irq/dcn201/irq_service_dcn201.c index a47f68634fc3..aa708b61142f 100644 --- a/drivers/gpu/drm/amd/display/dc/irq/dcn201/irq_service_dcn201.c +++ b/drivers/gpu/drm/amd/display/dc/irq/dcn201/irq_service_dcn201.c @@ -39,10 +39,9 @@ #include "ivsrcid/dcn/irqsrcs_dcn_1_0.h" -enum dc_irq_source to_dal_irq_source_dcn201( - struct irq_service *irq_service, - uint32_t src_id, - uint32_t ext_id) +static enum dc_irq_source to_dal_irq_source_dcn201(struct irq_service *irq_service, + uint32_t src_id, + uint32_t ext_id) { switch (src_id) { case DCN_1_0__SRCID__DC_D1_OTG_VSTARTUP: diff --git a/drivers/gpu/drm/amd/display/dc/irq/dcn21/irq_service_dcn21.c b/drivers/gpu/drm/amd/display/dc/irq/dcn21/irq_service_dcn21.c index 78940cb20e10..235294534c43 100644 --- a/drivers/gpu/drm/amd/display/dc/irq/dcn21/irq_service_dcn21.c +++ b/drivers/gpu/drm/amd/display/dc/irq/dcn21/irq_service_dcn21.c @@ -40,10 +40,9 @@ #include "ivsrcid/dcn/irqsrcs_dcn_1_0.h" -enum dc_irq_source to_dal_irq_source_dcn21( - struct irq_service *irq_service, - uint32_t src_id, - uint32_t ext_id) +static enum dc_irq_source to_dal_irq_source_dcn21(struct irq_service *irq_service, + uint32_t src_id, + uint32_t ext_id) { switch (src_id) { case DCN_1_0__SRCID__DC_D1_OTG_VSTARTUP: diff --git a/drivers/gpu/drm/amd/display/dc/irq/dcn31/irq_service_dcn31.c b/drivers/gpu/drm/amd/display/dc/irq/dcn31/irq_service_dcn31.c index 38e0ade60c7b..1b88e4e627fd 100644 --- a/drivers/gpu/drm/amd/display/dc/irq/dcn31/irq_service_dcn31.c +++ b/drivers/gpu/drm/amd/display/dc/irq/dcn31/irq_service_dcn31.c @@ -36,10 +36,9 @@ #include "ivsrcid/dcn/irqsrcs_dcn_1_0.h" -enum dc_irq_source to_dal_irq_source_dcn31( - struct irq_service *irq_service, - uint32_t src_id, - uint32_t ext_id) +static enum dc_irq_source to_dal_irq_source_dcn31(struct irq_service *irq_service, + uint32_t src_id, + uint32_t ext_id) { switch (src_id) { case DCN_1_0__SRCID__DC_D1_OTG_VSTARTUP: diff --git a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h index 7eec65090862..afe9bc839b45 100644 --- a/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h +++ b/drivers/gpu/drm/amd/display/dmub/inc/dmub_cmd.h @@ -46,10 +46,10 @@ /* Firmware versioning. */ #ifdef DMUB_EXPOSE_VERSION -#define DMUB_FW_VERSION_GIT_HASH 0x465e619a +#define DMUB_FW_VERSION_GIT_HASH 0xc99a4517 #define DMUB_FW_VERSION_MAJOR 0 #define DMUB_FW_VERSION_MINOR 0 -#define DMUB_FW_VERSION_REVISION 94 +#define DMUB_FW_VERSION_REVISION 97 #define DMUB_FW_VERSION_TEST 0 #define DMUB_FW_VERSION_VBIOS 0 #define DMUB_FW_VERSION_HOTFIX 0 @@ -408,7 +408,14 @@ enum dmub_cmd_vbios_type { * Enables or disables power gating. */ DMUB_CMD__VBIOS_ENABLE_DISP_POWER_GATING = 3, + /** + * Controls embedded panels. + */ DMUB_CMD__VBIOS_LVTMA_CONTROL = 15, + /** + * Query DP alt status on a transmitter. + */ + DMUB_CMD__VBIOS_TRANSMITTER_QUERY_DP_ALT = 26, }; //============================================================================== @@ -2394,6 +2401,24 @@ struct dmub_rb_cmd_lvtma_control { }; /** + * Data passed in/out in a DMUB_CMD__VBIOS_TRANSMITTER_QUERY_DP_ALT command. + */ +struct dmub_rb_cmd_transmitter_query_dp_alt_data { + uint8_t phy_id; /**< 0=UNIPHYA, 1=UNIPHYB, 2=UNIPHYC, 3=UNIPHYD, 4=UNIPHYE, 5=UNIPHYF */ + uint8_t is_usb; /**< is phy is usb */ + uint8_t is_dp_alt_disable; /**< is dp alt disable */ + uint8_t is_dp4; /**< is dp in 4 lane */ +}; + +/** + * Definition of a DMUB_CMD__VBIOS_TRANSMITTER_QUERY_DP_ALT command. + */ +struct dmub_rb_cmd_transmitter_query_dp_alt { + struct dmub_cmd_header header; /**< header */ + struct dmub_rb_cmd_transmitter_query_dp_alt_data data; /**< payload */ +}; + +/** * Maximum number of bytes a chunk sent to DMUB for parsing */ #define DMUB_EDID_CEA_DATA_CHUNK_BYTES 8 @@ -2404,7 +2429,7 @@ struct dmub_rb_cmd_lvtma_control { struct dmub_cmd_send_edid_cea { uint16_t offset; /**< offset into the CEA block */ uint8_t length; /**< number of bytes in payload to copy as part of CEA block */ - uint16_t total_length; /**< total length of the CEA block */ + uint16_t cea_total_length; /**< total length of the CEA block */ uint8_t payload[DMUB_EDID_CEA_DATA_CHUNK_BYTES]; /**< data chunk of the CEA block */ uint8_t pad[3]; /**< padding and for future expansion */ }; @@ -2601,6 +2626,10 @@ union dmub_rb_cmd { */ struct dmub_rb_cmd_lvtma_control lvtma_control; /** + * Definition of a DMUB_CMD__VBIOS_TRANSMITTER_QUERY_DP_ALT command. + */ + struct dmub_rb_cmd_transmitter_query_dp_alt query_dp_alt; + /** * Definition of a DMUB_CMD__DPIA_DIG1_CONTROL command. */ struct dmub_rb_cmd_dig1_dpia_control dig1_dpia_control; diff --git a/drivers/gpu/drm/amd/include/amd_shared.h b/drivers/gpu/drm/amd/include/amd_shared.h index 4b9e68a79f06..f57a1478f0fe 100644 --- a/drivers/gpu/drm/amd/include/amd_shared.h +++ b/drivers/gpu/drm/amd/include/amd_shared.h @@ -231,6 +231,8 @@ enum DC_FEATURE_MASK { DC_DISABLE_FRACTIONAL_PWM_MASK = (1 << 2), //0x4, disabled by default DC_PSR_MASK = (1 << 3), //0x8, disabled by default for dcn < 3.1 DC_EDP_NO_POWER_SEQUENCING = (1 << 4), //0x10, disabled by default + DC_DISABLE_LTTPR_DP1_4A = (1 << 5), //0x20, disabled by default + DC_DISABLE_LTTPR_DP2_0 = (1 << 6), //0x40, disabled by default }; enum DC_DEBUG_MASK { diff --git a/drivers/gpu/drm/amd/include/cyan_skillfish_ip_offset.h b/drivers/gpu/drm/amd/include/cyan_skillfish_ip_offset.h index 9cb5f3631c60..ce79e5de8ce3 100644 --- a/drivers/gpu/drm/amd/include/cyan_skillfish_ip_offset.h +++ b/drivers/gpu/drm/amd/include/cyan_skillfish_ip_offset.h @@ -25,15 +25,15 @@ #define MAX_SEGMENT 5 -struct IP_BASE_INSTANCE +struct IP_BASE_INSTANCE { unsigned int segment[MAX_SEGMENT]; -}; - -struct IP_BASE +} __maybe_unused; + +struct IP_BASE { struct IP_BASE_INSTANCE instance[MAX_INSTANCE]; -}; +} __maybe_unused; static const struct IP_BASE ATHUB_BASE ={ { { { 0x00000C00, 0, 0, 0, 0 } }, diff --git a/drivers/gpu/drm/amd/include/kgd_pp_interface.h b/drivers/gpu/drm/amd/include/kgd_pp_interface.h index bac15c466733..5c0867ebcfce 100644 --- a/drivers/gpu/drm/amd/include/kgd_pp_interface.h +++ b/drivers/gpu/drm/amd/include/kgd_pp_interface.h @@ -153,6 +153,10 @@ enum PP_SMC_POWER_PROFILE { PP_SMC_POWER_PROFILE_COUNT, }; +extern const char * const amdgpu_pp_profile_name[PP_SMC_POWER_PROFILE_COUNT]; + + + enum { PP_GROUP_UNKNOWN = 0, PP_GROUP_GFX = 1, diff --git a/drivers/gpu/drm/amd/include/yellow_carp_offset.h b/drivers/gpu/drm/amd/include/yellow_carp_offset.h index 76b9eb3f441d..28a56b56bcaf 100644 --- a/drivers/gpu/drm/amd/include/yellow_carp_offset.h +++ b/drivers/gpu/drm/amd/include/yellow_carp_offset.h @@ -9,12 +9,12 @@ struct IP_BASE_INSTANCE { unsigned int segment[MAX_SEGMENT]; -}; +} __maybe_unused; struct IP_BASE { struct IP_BASE_INSTANCE instance[MAX_INSTANCE]; -}; +} __maybe_unused; static const struct IP_BASE ACP_BASE = { { { { 0x02403800, 0x00480000, 0, 0, 0, 0 } }, diff --git a/drivers/gpu/drm/amd/pm/amdgpu_pm.c b/drivers/gpu/drm/amd/pm/amdgpu_pm.c index 49df4c20f09e..082539c70fd4 100644 --- a/drivers/gpu/drm/amd/pm/amdgpu_pm.c +++ b/drivers/gpu/drm/amd/pm/amdgpu_pm.c @@ -82,6 +82,16 @@ static const struct hwmon_temp_label { {PP_TEMP_MEM, "mem"}, }; +const char * const amdgpu_pp_profile_name[] = { + "BOOTUP_DEFAULT", + "3D_FULL_SCREEN", + "POWER_SAVING", + "VIDEO", + "VR", + "COMPUTE", + "CUSTOM" +}; + /** * DOC: power_dpm_state * diff --git a/drivers/gpu/drm/amd/pm/inc/amdgpu_dpm.h b/drivers/gpu/drm/amd/pm/inc/amdgpu_dpm.h index 16e3f72d31b9..c464a045000d 100644 --- a/drivers/gpu/drm/amd/pm/inc/amdgpu_dpm.h +++ b/drivers/gpu/drm/amd/pm/inc/amdgpu_dpm.h @@ -423,6 +423,9 @@ enum ip_power_state { POWER_STATE_OFF, }; +/* Used to mask smu debug modes */ +#define SMU_DEBUG_HALT_ON_ERROR 0x1 + struct amdgpu_pm { struct mutex mutex; u32 current_sclk; @@ -460,6 +463,11 @@ struct amdgpu_pm { struct list_head pm_attr_list; atomic_t pwr_state[AMD_IP_BLOCK_TYPE_NUM]; + + /* + * 0 = disabled (default), otherwise enable corresponding debug mode + */ + uint32_t smu_debug_mask; }; #define R600_SSTU_DFLT 0 diff --git a/drivers/gpu/drm/amd/pm/inc/amdgpu_smu.h b/drivers/gpu/drm/amd/pm/inc/amdgpu_smu.h index f738f7dc20c9..2b9b9a7ba97a 100644 --- a/drivers/gpu/drm/amd/pm/inc/amdgpu_smu.h +++ b/drivers/gpu/drm/amd/pm/inc/amdgpu_smu.h @@ -481,6 +481,7 @@ struct stb_context { }; #define WORKLOAD_POLICY_MAX 7 + struct smu_context { struct amdgpu_device *adev; diff --git a/drivers/gpu/drm/amd/pm/powerplay/amd_powerplay.c b/drivers/gpu/drm/amd/pm/powerplay/amd_powerplay.c index 20cb234d5061..3ab67b232cd4 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/amd_powerplay.c +++ b/drivers/gpu/drm/amd/pm/powerplay/amd_powerplay.c @@ -1328,7 +1328,12 @@ static int pp_set_powergating_by_smu(void *handle, pp_dpm_powergate_vce(handle, gate); break; case AMD_IP_BLOCK_TYPE_GMC: - pp_dpm_powergate_mmhub(handle); + /* + * For now, this is only used on PICASSO. + * And only "gate" operation is supported. + */ + if (gate) + pp_dpm_powergate_mmhub(handle); break; case AMD_IP_BLOCK_TYPE_GFX: ret = pp_dpm_powergate_gfx(handle, gate); diff --git a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu10_hwmgr.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu10_hwmgr.c index 1f406f21b452..9ddd8491ff00 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu10_hwmgr.c +++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu10_hwmgr.c @@ -1439,13 +1439,6 @@ static int smu10_get_power_profile_mode(struct pp_hwmgr *hwmgr, char *buf) {70, 90, 0, 0,}, {30, 60, 0, 6,}, }; - static const char *profile_name[6] = { - "BOOTUP_DEFAULT", - "3D_FULL_SCREEN", - "POWER_SAVING", - "VIDEO", - "VR", - "COMPUTE"}; static const char *title[6] = {"NUM", "MODE_NAME", "BUSY_SET_POINT", @@ -1463,7 +1456,7 @@ static int smu10_get_power_profile_mode(struct pp_hwmgr *hwmgr, char *buf) for (i = 0; i <= PP_SMC_POWER_PROFILE_COMPUTE; i++) size += sysfs_emit_at(buf, size, "%3d %14s%s: %14d %3d %10d %14d\n", - i, profile_name[i], (i == hwmgr->power_profile_mode) ? "*" : " ", + i, amdgpu_pp_profile_name[i], (i == hwmgr->power_profile_mode) ? "*" : " ", profile_mode_setting[i][0], profile_mode_setting[i][1], profile_mode_setting[i][2], profile_mode_setting[i][3]); diff --git a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu7_hwmgr.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu7_hwmgr.c index 611969bf4520..cd99db0dc2be 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu7_hwmgr.c +++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/smu7_hwmgr.c @@ -5498,14 +5498,6 @@ static int smu7_get_power_profile_mode(struct pp_hwmgr *hwmgr, char *buf) uint32_t i, size = 0; uint32_t len; - static const char *profile_name[7] = {"BOOTUP_DEFAULT", - "3D_FULL_SCREEN", - "POWER_SAVING", - "VIDEO", - "VR", - "COMPUTE", - "CUSTOM"}; - static const char *title[8] = {"NUM", "MODE_NAME", "SCLK_UP_HYST", @@ -5529,7 +5521,7 @@ static int smu7_get_power_profile_mode(struct pp_hwmgr *hwmgr, char *buf) for (i = 0; i < len; i++) { if (i == hwmgr->power_profile_mode) { size += sysfs_emit_at(buf, size, "%3d %14s %s: %8d %16d %16d %16d %16d %16d\n", - i, profile_name[i], "*", + i, amdgpu_pp_profile_name[i], "*", data->current_profile_setting.sclk_up_hyst, data->current_profile_setting.sclk_down_hyst, data->current_profile_setting.sclk_activity, @@ -5540,12 +5532,12 @@ static int smu7_get_power_profile_mode(struct pp_hwmgr *hwmgr, char *buf) } if (smu7_profiling[i].bupdate_sclk) size += sysfs_emit_at(buf, size, "%3d %16s: %8d %16d %16d ", - i, profile_name[i], smu7_profiling[i].sclk_up_hyst, + i, amdgpu_pp_profile_name[i], smu7_profiling[i].sclk_up_hyst, smu7_profiling[i].sclk_down_hyst, smu7_profiling[i].sclk_activity); else size += sysfs_emit_at(buf, size, "%3d %16s: %8s %16s %16s ", - i, profile_name[i], "-", "-", "-"); + i, amdgpu_pp_profile_name[i], "-", "-", "-"); if (smu7_profiling[i].bupdate_mclk) size += sysfs_emit_at(buf, size, "%16d %16d %16d\n", diff --git a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_hwmgr.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_hwmgr.c index e6336654c565..3f040be0d158 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_hwmgr.c +++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega10_hwmgr.c @@ -5097,13 +5097,6 @@ static int vega10_get_power_profile_mode(struct pp_hwmgr *hwmgr, char *buf) {70, 90, 0, 0,}, {30, 60, 0, 6,}, }; - static const char *profile_name[7] = {"BOOTUP_DEFAULT", - "3D_FULL_SCREEN", - "POWER_SAVING", - "VIDEO", - "VR", - "COMPUTE", - "CUSTOM"}; static const char *title[6] = {"NUM", "MODE_NAME", "BUSY_SET_POINT", @@ -5121,11 +5114,12 @@ static int vega10_get_power_profile_mode(struct pp_hwmgr *hwmgr, char *buf) for (i = 0; i < PP_SMC_POWER_PROFILE_CUSTOM; i++) size += sysfs_emit_at(buf, size, "%3d %14s%s: %14d %3d %10d %14d\n", - i, profile_name[i], (i == hwmgr->power_profile_mode) ? "*" : " ", + i, amdgpu_pp_profile_name[i], (i == hwmgr->power_profile_mode) ? "*" : " ", profile_mode_setting[i][0], profile_mode_setting[i][1], profile_mode_setting[i][2], profile_mode_setting[i][3]); + size += sysfs_emit_at(buf, size, "%3d %14s%s: %14d %3d %10d %14d\n", i, - profile_name[i], (i == hwmgr->power_profile_mode) ? "*" : " ", + amdgpu_pp_profile_name[i], (i == hwmgr->power_profile_mode) ? "*" : " ", data->custom_profile_mode[0], data->custom_profile_mode[1], data->custom_profile_mode[2], data->custom_profile_mode[3]); return size; diff --git a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega20_hwmgr.c b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega20_hwmgr.c index 85d55ab4e369..97b3ad369046 100644 --- a/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega20_hwmgr.c +++ b/drivers/gpu/drm/amd/pm/powerplay/hwmgr/vega20_hwmgr.c @@ -3980,14 +3980,6 @@ static int vega20_get_power_profile_mode(struct pp_hwmgr *hwmgr, char *buf) DpmActivityMonitorCoeffInt_t activity_monitor; uint32_t i, size = 0; uint16_t workload_type = 0; - static const char *profile_name[] = { - "BOOTUP_DEFAULT", - "3D_FULL_SCREEN", - "POWER_SAVING", - "VIDEO", - "VR", - "COMPUTE", - "CUSTOM"}; static const char *title[] = { "PROFILE_INDEX(NAME)", "CLOCK_TYPE(NAME)", @@ -4021,7 +4013,7 @@ static int vega20_get_power_profile_mode(struct pp_hwmgr *hwmgr, char *buf) return result); size += sysfs_emit_at(buf, size, "%2d %14s%s:\n", - i, profile_name[i], (i == hwmgr->power_profile_mode) ? "*" : " "); + i, amdgpu_pp_profile_name[i], (i == hwmgr->power_profile_mode) ? "*" : " "); size += sysfs_emit_at(buf, size, "%19s %d(%13s) %7d %7d %7d %7d %7d %7d %7d %7d %7d\n", " ", diff --git a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c index edcf2738748a..2d718c30c8eb 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c +++ b/drivers/gpu/drm/amd/pm/swsmu/amdgpu_smu.c @@ -277,8 +277,12 @@ static int smu_dpm_set_power_gate(void *handle, struct smu_context *smu = handle; int ret = 0; - if (!smu->pm_enabled || !smu->adev->pm.dpm_enabled) + if (!smu->pm_enabled || !smu->adev->pm.dpm_enabled) { + dev_WARN(smu->adev->dev, + "SMU uninitialized but power %s requested for %u!\n", + gate ? "gate" : "ungate", block_type); return -EOPNOTSUPP; + } switch (block_type) { /* @@ -1346,7 +1350,6 @@ static int smu_hw_init(void *handle) } if (smu->is_apu) { - smu_powergate_sdma(&adev->smu, false); smu_dpm_set_vcn_enable(smu, true); smu_dpm_set_jpeg_enable(smu, true); smu_set_gfx_cgpg(&adev->smu, true); @@ -1508,10 +1511,6 @@ static int smu_hw_fini(void *handle) if (amdgpu_sriov_vf(adev)&& !amdgpu_sriov_is_pp_one_vf(adev)) return 0; - if (smu->is_apu) { - smu_powergate_sdma(&adev->smu, true); - } - smu_dpm_set_vcn_enable(smu, false); smu_dpm_set_jpeg_enable(smu, false); diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c index fd1d30a93db5..58bc387fb279 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/arcturus_ppt.c @@ -295,16 +295,6 @@ static int arcturus_allocate_dpm_context(struct smu_context *smu) return -ENOMEM; smu_dpm->dpm_context_size = sizeof(struct smu_11_0_dpm_context); - smu_dpm->dpm_current_power_state = kzalloc(sizeof(struct smu_power_state), - GFP_KERNEL); - if (!smu_dpm->dpm_current_power_state) - return -ENOMEM; - - smu_dpm->dpm_request_power_state = kzalloc(sizeof(struct smu_power_state), - GFP_KERNEL); - if (!smu_dpm->dpm_request_power_state) - return -ENOMEM; - return 0; } @@ -1389,14 +1379,6 @@ static int arcturus_get_power_profile_mode(struct smu_context *smu, char *buf) { DpmActivityMonitorCoeffInt_t activity_monitor; - static const char *profile_name[] = { - "BOOTUP_DEFAULT", - "3D_FULL_SCREEN", - "POWER_SAVING", - "VIDEO", - "VR", - "COMPUTE", - "CUSTOM"}; static const char *title[] = { "PROFILE_INDEX(NAME)", "CLOCK_TYPE(NAME)", @@ -1453,7 +1435,7 @@ static int arcturus_get_power_profile_mode(struct smu_context *smu, } size += sysfs_emit_at(buf, size, "%2d %14s%s\n", - i, profile_name[i], (i == smu->power_profile_mode) ? "*" : " "); + i, amdgpu_pp_profile_name[i], (i == smu->power_profile_mode) ? "*" : " "); if (smu_version >= 0x360d00) { size += sysfs_emit_at(buf, size, "%19s %d(%13s) %7d %7d %7d %7d %7d %7d %7d %7d %7d\n", diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c index 60a557068ea4..2bb7816b245a 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/navi10_ppt.c @@ -1713,14 +1713,6 @@ static int navi10_get_power_profile_mode(struct smu_context *smu, char *buf) DpmActivityMonitorCoeffInt_t activity_monitor; uint32_t i, size = 0; int16_t workload_type = 0; - static const char *profile_name[] = { - "BOOTUP_DEFAULT", - "3D_FULL_SCREEN", - "POWER_SAVING", - "VIDEO", - "VR", - "COMPUTE", - "CUSTOM"}; static const char *title[] = { "PROFILE_INDEX(NAME)", "CLOCK_TYPE(NAME)", @@ -1759,7 +1751,7 @@ static int navi10_get_power_profile_mode(struct smu_context *smu, char *buf) } size += sysfs_emit_at(buf, size, "%2d %14s%s:\n", - i, profile_name[i], (i == smu->power_profile_mode) ? "*" : " "); + i, amdgpu_pp_profile_name[i], (i == smu->power_profile_mode) ? "*" : " "); size += sysfs_emit_at(buf, size, "%19s %d(%13s) %7d %7d %7d %7d %7d %7d %7d %7d %7d\n", " ", diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c index a673e05853fe..777f717c37ae 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/sienna_cichlid_ppt.c @@ -1349,14 +1349,6 @@ static int sienna_cichlid_get_power_profile_mode(struct smu_context *smu, char * &(activity_monitor_external.DpmActivityMonitorCoeffInt); uint32_t i, size = 0; int16_t workload_type = 0; - static const char *profile_name[] = { - "BOOTUP_DEFAULT", - "3D_FULL_SCREEN", - "POWER_SAVING", - "VIDEO", - "VR", - "COMPUTE", - "CUSTOM"}; static const char *title[] = { "PROFILE_INDEX(NAME)", "CLOCK_TYPE(NAME)", @@ -1395,7 +1387,7 @@ static int sienna_cichlid_get_power_profile_mode(struct smu_context *smu, char * } size += sysfs_emit_at(buf, size, "%2d %14s%s:\n", - i, profile_name[i], (i == smu->power_profile_mode) ? "*" : " "); + i, amdgpu_pp_profile_name[i], (i == smu->power_profile_mode) ? "*" : " "); size += sysfs_emit_at(buf, size, "%19s %d(%13s) %7d %7d %7d %7d %7d %7d %7d %7d %7d\n", " ", diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu11/vangogh_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu11/vangogh_ppt.c index c02ed65ffa38..5cb07ed227fb 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu11/vangogh_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu11/vangogh_ppt.c @@ -1039,14 +1039,6 @@ failed: static int vangogh_get_power_profile_mode(struct smu_context *smu, char *buf) { - static const char *profile_name[] = { - "BOOTUP_DEFAULT", - "3D_FULL_SCREEN", - "POWER_SAVING", - "VIDEO", - "VR", - "COMPUTE", - "CUSTOM"}; uint32_t i, size = 0; int16_t workload_type = 0; @@ -1066,7 +1058,7 @@ static int vangogh_get_power_profile_mode(struct smu_context *smu, continue; size += sysfs_emit_at(buf, size, "%2d %14s%s\n", - i, profile_name[i], (i == smu->power_profile_mode) ? "*" : " "); + i, amdgpu_pp_profile_name[i], (i == smu->power_profile_mode) ? "*" : " "); } return size; diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu12/renoir_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu12/renoir_ppt.c index 145f13b8c977..25c4b135f830 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu12/renoir_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu12/renoir_ppt.c @@ -1095,14 +1095,6 @@ static int renoir_set_watermarks_table( static int renoir_get_power_profile_mode(struct smu_context *smu, char *buf) { - static const char *profile_name[] = { - "BOOTUP_DEFAULT", - "3D_FULL_SCREEN", - "POWER_SAVING", - "VIDEO", - "VR", - "COMPUTE", - "CUSTOM"}; uint32_t i, size = 0; int16_t workload_type = 0; @@ -1121,7 +1113,7 @@ static int renoir_get_power_profile_mode(struct smu_context *smu, continue; size += sysfs_emit_at(buf, size, "%2d %14s%s\n", - i, profile_name[i], (i == smu->power_profile_mode) ? "*" : " "); + i, amdgpu_pp_profile_name[i], (i == smu->power_profile_mode) ? "*" : " "); } return size; diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu12/smu_v12_0.c b/drivers/gpu/drm/amd/pm/swsmu/smu12/smu_v12_0.c index d60b8c5e8715..43028f2cd28b 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu12/smu_v12_0.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu12/smu_v12_0.c @@ -191,6 +191,9 @@ int smu_v12_0_fini_smc_tables(struct smu_context *smu) kfree(smu_table->watermarks_table); smu_table->watermarks_table = NULL; + kfree(smu_table->gpu_metrics_table); + smu_table->gpu_metrics_table = NULL; + return 0; } diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c index 6e781cee8bb6..0907da022197 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c @@ -262,16 +262,6 @@ static int aldebaran_allocate_dpm_context(struct smu_context *smu) return -ENOMEM; smu_dpm->dpm_context_size = sizeof(struct smu_13_0_dpm_context); - smu_dpm->dpm_current_power_state = kzalloc(sizeof(struct smu_power_state), - GFP_KERNEL); - if (!smu_dpm->dpm_current_power_state) - return -ENOMEM; - - smu_dpm->dpm_request_power_state = kzalloc(sizeof(struct smu_power_state), - GFP_KERNEL); - if (!smu_dpm->dpm_request_power_state) - return -ENOMEM; - return 0; } diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c index 55421ea622fb..677a246212f9 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c @@ -196,6 +196,7 @@ int smu_v13_0_check_fw_status(struct smu_context *smu) int smu_v13_0_check_fw_version(struct smu_context *smu) { + struct amdgpu_device *adev = smu->adev; uint32_t if_version = 0xff, smu_version = 0xff; uint16_t smu_major; uint8_t smu_minor, smu_debug; @@ -208,6 +209,8 @@ int smu_v13_0_check_fw_version(struct smu_context *smu) smu_major = (smu_version >> 16) & 0xffff; smu_minor = (smu_version >> 8) & 0xff; smu_debug = (smu_version >> 0) & 0xff; + if (smu->is_apu) + adev->pm.fw_version = smu_version; switch (smu->adev->ip_versions[MP1_HWIP][0]) { case IP_VERSION(13, 0, 2): @@ -224,9 +227,6 @@ int smu_v13_0_check_fw_version(struct smu_context *smu) break; } - dev_info(smu->adev->dev, "smu fw reported version = 0x%08x (%d.%d.%d)\n", - smu_version, smu_major, smu_minor, smu_debug); - /* * 1. if_version mismatch is not critical as our fw is designed * to be backward compatible. diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c b/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c index 500af6f8adcb..ee1a312fd497 100644 --- a/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c +++ b/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.c @@ -94,7 +94,7 @@ static void smu_cmn_read_arg(struct smu_context *smu, /** * __smu_cmn_poll_stat -- poll for a status from the SMU - * smu: a pointer to SMU context + * @smu: a pointer to SMU context * * Returns the status of the SMU, which could be, * 0, the SMU is busy with your command; @@ -257,10 +257,11 @@ int smu_cmn_send_msg_without_waiting(struct smu_context *smu, uint16_t msg_index, uint32_t param) { + struct amdgpu_device *adev = smu->adev; u32 reg; int res; - if (smu->adev->no_hw_access) + if (adev->no_hw_access) return 0; reg = __smu_cmn_poll_stat(smu); @@ -272,6 +273,12 @@ int smu_cmn_send_msg_without_waiting(struct smu_context *smu, __smu_cmn_send_msg(smu, msg_index, param); res = 0; Out: + if (unlikely(adev->pm.smu_debug_mask & SMU_DEBUG_HALT_ON_ERROR) && + res && (res != -ETIME)) { + amdgpu_device_halt(adev); + WARN_ON(1); + } + return res; } @@ -288,9 +295,18 @@ Out: int smu_cmn_wait_for_response(struct smu_context *smu) { u32 reg; + int res; reg = __smu_cmn_poll_stat(smu); - return __smu_cmn_reg2errno(smu, reg); + res = __smu_cmn_reg2errno(smu, reg); + + if (unlikely(smu->adev->pm.smu_debug_mask & SMU_DEBUG_HALT_ON_ERROR) && + res && (res != -ETIME)) { + amdgpu_device_halt(smu->adev); + WARN_ON(1); + } + + return res; } /** @@ -328,10 +344,11 @@ int smu_cmn_send_smc_msg_with_param(struct smu_context *smu, uint32_t param, uint32_t *read_arg) { + struct amdgpu_device *adev = smu->adev; int res, index; u32 reg; - if (smu->adev->no_hw_access) + if (adev->no_hw_access) return 0; index = smu_cmn_to_asic_specific_index(smu, @@ -357,6 +374,11 @@ int smu_cmn_send_smc_msg_with_param(struct smu_context *smu, if (read_arg) smu_cmn_read_arg(smu, read_arg); Out: + if (unlikely(adev->pm.smu_debug_mask & SMU_DEBUG_HALT_ON_ERROR) && res) { + amdgpu_device_halt(adev); + WARN_ON(1); + } + mutex_unlock(&smu->message_lock); return res; } diff --git a/drivers/gpu/drm/etnaviv/etnaviv_drv.c b/drivers/gpu/drm/etnaviv/etnaviv_drv.c index 7dcc6392792d..0b756ecb1bc2 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_drv.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_drv.c @@ -589,6 +589,7 @@ static int compare_str(struct device *dev, void *data) static int etnaviv_pdev_probe(struct platform_device *pdev) { struct device *dev = &pdev->dev; + struct device_node *first_node = NULL; struct component_match *match = NULL; if (!dev->platform_data) { @@ -598,6 +599,9 @@ static int etnaviv_pdev_probe(struct platform_device *pdev) if (!of_device_is_available(core_node)) continue; + if (!first_node) + first_node = core_node; + drm_of_component_match_add(&pdev->dev, &match, compare_of, core_node); } @@ -609,6 +613,32 @@ static int etnaviv_pdev_probe(struct platform_device *pdev) component_match_add(dev, &match, compare_str, names[i]); } + /* + * PTA and MTLB can have 40 bit base addresses, but + * unfortunately, an entry in the MTLB can only point to a + * 32 bit base address of a STLB. Moreover, to initialize the + * MMU we need a command buffer with a 32 bit address because + * without an MMU there is only an indentity mapping between + * the internal 32 bit addresses and the bus addresses. + * + * To make things easy, we set the dma_coherent_mask to 32 + * bit to make sure we are allocating the command buffers and + * TLBs in the lower 4 GiB address space. + */ + if (dma_set_mask(&pdev->dev, DMA_BIT_MASK(40)) || + dma_set_coherent_mask(&pdev->dev, DMA_BIT_MASK(32))) { + dev_dbg(&pdev->dev, "No suitable DMA available\n"); + return -ENODEV; + } + + /* + * Apply the same DMA configuration to the virtual etnaviv + * device as the GPU we found. This assumes that all Vivante + * GPUs in the system share the same DMA constraints. + */ + if (first_node) + of_dma_configure(&pdev->dev, first_node, true); + return component_master_add_with_match(dev, &etnaviv_master_ops, match); } @@ -653,21 +683,12 @@ static int __init etnaviv_init(void) if (!of_device_is_available(np)) continue; - pdev = platform_device_alloc("etnaviv", -1); + pdev = platform_device_alloc("etnaviv", PLATFORM_DEVID_NONE); if (!pdev) { ret = -ENOMEM; of_node_put(np); goto unregister_platform_driver; } - pdev->dev.coherent_dma_mask = DMA_BIT_MASK(40); - pdev->dev.dma_mask = &pdev->dev.coherent_dma_mask; - - /* - * Apply the same DMA configuration to the virtual etnaviv - * device as the GPU we found. This assumes that all Vivante - * GPUs in the system share the same DMA constraints. - */ - of_dma_configure(&pdev->dev, np, true); ret = platform_device_add(pdev); if (ret) { diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gem_submit.c b/drivers/gpu/drm/etnaviv/etnaviv_gem_submit.c index b5e8ce86dbe7..b03c20c14ca1 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_gem_submit.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_gem_submit.c @@ -469,6 +469,12 @@ int etnaviv_ioctl_gem_submit(struct drm_device *dev, void *data, return -EINVAL; } + if (args->stream_size > SZ_64K || args->nr_relocs > SZ_64K || + args->nr_bos > SZ_64K || args->nr_pmrs > 128) { + DRM_ERROR("submit arguments out of size limits\n"); + return -EINVAL; + } + /* * Copy the command submission and bo array to kernel space in * one go, and do this outside of any locks. diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gpu.c b/drivers/gpu/drm/etnaviv/etnaviv_gpu.c index 242a5fd8b932..ba5fd012a40a 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_gpu.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_gpu.c @@ -1658,7 +1658,7 @@ etnaviv_gpu_cooling_set_cur_state(struct thermal_cooling_device *cdev, return 0; } -static struct thermal_cooling_device_ops cooling_ops = { +static const struct thermal_cooling_device_ops cooling_ops = { .get_max_state = etnaviv_gpu_cooling_get_max_state, .get_cur_state = etnaviv_gpu_cooling_get_cur_state, .set_cur_state = etnaviv_gpu_cooling_set_cur_state, diff --git a/drivers/gpu/drm/etnaviv/etnaviv_gpu.h b/drivers/gpu/drm/etnaviv/etnaviv_gpu.h index 1c75c8ed5bce..85eddd492774 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_gpu.h +++ b/drivers/gpu/drm/etnaviv/etnaviv_gpu.h @@ -130,6 +130,7 @@ struct etnaviv_gpu { /* hang detection */ u32 hangcheck_dma_addr; + u32 hangcheck_fence; void __iomem *mmio; int irq; diff --git a/drivers/gpu/drm/etnaviv/etnaviv_sched.c b/drivers/gpu/drm/etnaviv/etnaviv_sched.c index 180bb633d5c5..58f593b278c1 100644 --- a/drivers/gpu/drm/etnaviv/etnaviv_sched.c +++ b/drivers/gpu/drm/etnaviv/etnaviv_sched.c @@ -107,8 +107,10 @@ static enum drm_gpu_sched_stat etnaviv_sched_timedout_job(struct drm_sched_job */ dma_addr = gpu_read(gpu, VIVS_FE_DMA_ADDRESS); change = dma_addr - gpu->hangcheck_dma_addr; - if (change < 0 || change > 16) { + if (gpu->completed_fence != gpu->hangcheck_fence || + change < 0 || change > 16) { gpu->hangcheck_dma_addr = dma_addr; + gpu->hangcheck_fence = gpu->completed_fence; goto out_no_timeout; } diff --git a/drivers/gpu/drm/exynos/exynos_drm_drv.c b/drivers/gpu/drm/exynos/exynos_drm_drv.c index d8f1cf4d6b69..9743b6b17447 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_drv.c +++ b/drivers/gpu/drm/exynos/exynos_drm_drv.c @@ -102,16 +102,7 @@ static const struct drm_ioctl_desc exynos_ioctls[] = { DRM_RENDER_ALLOW), }; -static const struct file_operations exynos_drm_driver_fops = { - .owner = THIS_MODULE, - .open = drm_open, - .mmap = exynos_drm_gem_mmap, - .poll = drm_poll, - .read = drm_read, - .unlocked_ioctl = drm_ioctl, - .compat_ioctl = drm_compat_ioctl, - .release = drm_release, -}; +DEFINE_DRM_GEM_FOPS(exynos_drm_driver_fops); static const struct drm_driver exynos_drm_driver = { .driver_features = DRIVER_MODESET | DRIVER_GEM @@ -124,7 +115,7 @@ static const struct drm_driver exynos_drm_driver = { .prime_fd_to_handle = drm_gem_prime_fd_to_handle, .gem_prime_import = exynos_drm_gem_prime_import, .gem_prime_import_sg_table = exynos_drm_gem_prime_import_sg_table, - .gem_prime_mmap = exynos_drm_gem_prime_mmap, + .gem_prime_mmap = drm_gem_prime_mmap, .ioctls = exynos_ioctls, .num_ioctls = ARRAY_SIZE(exynos_ioctls), .fops = &exynos_drm_driver_fops, diff --git a/drivers/gpu/drm/exynos/exynos_drm_dsi.c b/drivers/gpu/drm/exynos/exynos_drm_dsi.c index 8d137857818c..32a36572b894 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_dsi.c +++ b/drivers/gpu/drm/exynos/exynos_drm_dsi.c @@ -13,7 +13,6 @@ #include <linux/gpio/consumer.h> #include <linux/irq.h> #include <linux/of_device.h> -#include <linux/of_gpio.h> #include <linux/of_graph.h> #include <linux/phy/phy.h> #include <linux/regulator/consumer.h> @@ -265,7 +264,7 @@ struct exynos_dsi { struct clk **clks; struct regulator_bulk_data supplies[2]; int irq; - int te_gpio; + struct gpio_desc *te_gpio; u32 pll_clk_rate; u32 burst_clk_rate; @@ -1298,14 +1297,14 @@ static void exynos_dsi_enable_irq(struct exynos_dsi *dsi) { enable_irq(dsi->irq); - if (gpio_is_valid(dsi->te_gpio)) - enable_irq(gpio_to_irq(dsi->te_gpio)); + if (dsi->te_gpio) + enable_irq(gpiod_to_irq(dsi->te_gpio)); } static void exynos_dsi_disable_irq(struct exynos_dsi *dsi) { - if (gpio_is_valid(dsi->te_gpio)) - disable_irq(gpio_to_irq(dsi->te_gpio)); + if (dsi->te_gpio) + disable_irq(gpiod_to_irq(dsi->te_gpio)); disable_irq(dsi->irq); } @@ -1335,42 +1334,31 @@ static int exynos_dsi_register_te_irq(struct exynos_dsi *dsi, int ret; int te_gpio_irq; - dsi->te_gpio = of_get_named_gpio(panel->of_node, "te-gpios", 0); - if (dsi->te_gpio == -ENOENT) - return 0; - - if (!gpio_is_valid(dsi->te_gpio)) { - ret = dsi->te_gpio; - dev_err(dsi->dev, "cannot get te-gpios, %d\n", ret); - goto out; - } - - ret = gpio_request(dsi->te_gpio, "te_gpio"); - if (ret) { - dev_err(dsi->dev, "gpio request failed with %d\n", ret); - goto out; + dsi->te_gpio = devm_gpiod_get_optional(dsi->dev, "te", GPIOD_IN); + if (IS_ERR(dsi->te_gpio)) { + dev_err(dsi->dev, "gpio request failed with %ld\n", + PTR_ERR(dsi->te_gpio)); + return PTR_ERR(dsi->te_gpio); } - te_gpio_irq = gpio_to_irq(dsi->te_gpio); + te_gpio_irq = gpiod_to_irq(dsi->te_gpio); ret = request_threaded_irq(te_gpio_irq, exynos_dsi_te_irq_handler, NULL, IRQF_TRIGGER_RISING | IRQF_NO_AUTOEN, "TE", dsi); if (ret) { dev_err(dsi->dev, "request interrupt failed with %d\n", ret); - gpio_free(dsi->te_gpio); - goto out; + gpiod_put(dsi->te_gpio); + return ret; } -out: - return ret; + return 0; } static void exynos_dsi_unregister_te_irq(struct exynos_dsi *dsi) { - if (gpio_is_valid(dsi->te_gpio)) { - free_irq(gpio_to_irq(dsi->te_gpio), dsi); - gpio_free(dsi->te_gpio); - dsi->te_gpio = -ENOENT; + if (dsi->te_gpio) { + free_irq(gpiod_to_irq(dsi->te_gpio), dsi); + gpiod_put(dsi->te_gpio); } } @@ -1745,9 +1733,6 @@ static int exynos_dsi_probe(struct platform_device *pdev) if (!dsi) return -ENOMEM; - /* To be checked as invalid one */ - dsi->te_gpio = -ENOENT; - init_completion(&dsi->completed); spin_lock_init(&dsi->transfer_lock); INIT_LIST_HEAD(&dsi->transfer_list); diff --git a/drivers/gpu/drm/exynos/exynos_drm_fbdev.c b/drivers/gpu/drm/exynos/exynos_drm_fbdev.c index 5147f5929be7..02c97b9ca926 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_fbdev.c +++ b/drivers/gpu/drm/exynos/exynos_drm_fbdev.c @@ -15,6 +15,7 @@ #include <drm/drm_crtc.h> #include <drm/drm_fb_helper.h> #include <drm/drm_fourcc.h> +#include <drm/drm_prime.h> #include <drm/drm_probe_helper.h> #include <drm/exynos_drm.h> @@ -39,25 +40,8 @@ static int exynos_drm_fb_mmap(struct fb_info *info, struct drm_fb_helper *helper = info->par; struct exynos_drm_fbdev *exynos_fbd = to_exynos_fbdev(helper); struct exynos_drm_gem *exynos_gem = exynos_fbd->exynos_gem; - unsigned long vm_size; - int ret; - - vma->vm_flags |= VM_IO | VM_DONTEXPAND | VM_DONTDUMP; - - vm_size = vma->vm_end - vma->vm_start; - - if (vm_size > exynos_gem->size) - return -EINVAL; - ret = dma_mmap_attrs(to_dma_dev(helper->dev), vma, exynos_gem->cookie, - exynos_gem->dma_addr, exynos_gem->size, - exynos_gem->dma_attrs); - if (ret < 0) { - DRM_DEV_ERROR(to_dma_dev(helper->dev), "failed to mmap.\n"); - return ret; - } - - return 0; + return drm_gem_prime_mmap(&exynos_gem->base, vma); } static const struct fb_ops exynos_drm_fb_ops = { diff --git a/drivers/gpu/drm/exynos/exynos_drm_fimc.c b/drivers/gpu/drm/exynos/exynos_drm_fimc.c index ecfd82d0afb7..023f54ee61a8 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_fimc.c +++ b/drivers/gpu/drm/exynos/exynos_drm_fimc.c @@ -782,8 +782,8 @@ static int fimc_set_prescaler(struct fimc_context *ctx, struct fimc_scaler *sc, sc->hratio = (src_w << 14) / (dst_w << hfactor); sc->vratio = (src_h << 14) / (dst_h << vfactor); - sc->up_h = (dst_w >= src_w) ? true : false; - sc->up_v = (dst_h >= src_h) ? true : false; + sc->up_h = (dst_w >= src_w); + sc->up_v = (dst_h >= src_h); DRM_DEV_DEBUG_KMS(ctx->dev, "hratio[%d]vratio[%d]up_h[%d]up_v[%d]\n", sc->hratio, sc->vratio, sc->up_h, sc->up_v); diff --git a/drivers/gpu/drm/exynos/exynos_drm_gem.c b/drivers/gpu/drm/exynos/exynos_drm_gem.c index 0a0c042a3155..3e493f48e0d4 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_gem.c +++ b/drivers/gpu/drm/exynos/exynos_drm_gem.c @@ -20,6 +20,8 @@ MODULE_IMPORT_NS(DMA_BUF); +static int exynos_drm_gem_mmap(struct drm_gem_object *obj, struct vm_area_struct *vma); + static int exynos_drm_alloc_buf(struct exynos_drm_gem *exynos_gem, bool kvmap) { struct drm_device *dev = exynos_gem->base.dev; @@ -138,6 +140,7 @@ static const struct vm_operations_struct exynos_drm_gem_vm_ops = { static const struct drm_gem_object_funcs exynos_drm_gem_object_funcs = { .free = exynos_drm_gem_free_object, .get_sg_table = exynos_drm_gem_prime_get_sg_table, + .mmap = exynos_drm_gem_mmap, .vm_ops = &exynos_drm_gem_vm_ops, }; @@ -357,12 +360,16 @@ int exynos_drm_gem_dumb_create(struct drm_file *file_priv, return 0; } -static int exynos_drm_gem_mmap_obj(struct drm_gem_object *obj, - struct vm_area_struct *vma) +static int exynos_drm_gem_mmap(struct drm_gem_object *obj, struct vm_area_struct *vma) { struct exynos_drm_gem *exynos_gem = to_exynos_gem(obj); int ret; + if (obj->import_attach) + return dma_buf_mmap(obj->dma_buf, vma, 0); + + vma->vm_flags |= VM_IO | VM_DONTEXPAND | VM_DONTDUMP; + DRM_DEV_DEBUG_KMS(to_dma_dev(obj->dev), "flags = 0x%x\n", exynos_gem->flags); @@ -388,26 +395,6 @@ err_close_vm: return ret; } -int exynos_drm_gem_mmap(struct file *filp, struct vm_area_struct *vma) -{ - struct drm_gem_object *obj; - int ret; - - /* set vm_area_struct. */ - ret = drm_gem_mmap(filp, vma); - if (ret < 0) { - DRM_ERROR("failed to mmap.\n"); - return ret; - } - - obj = vma->vm_private_data; - - if (obj->import_attach) - return dma_buf_mmap(obj->dma_buf, vma, 0); - - return exynos_drm_gem_mmap_obj(obj, vma); -} - /* low-level interface prime helpers */ struct drm_gem_object *exynos_drm_gem_prime_import(struct drm_device *dev, struct dma_buf *dma_buf) @@ -469,15 +456,3 @@ exynos_drm_gem_prime_import_sg_table(struct drm_device *dev, exynos_gem->sgt = sgt; return &exynos_gem->base; } - -int exynos_drm_gem_prime_mmap(struct drm_gem_object *obj, - struct vm_area_struct *vma) -{ - int ret; - - ret = drm_gem_mmap_obj(obj, obj->size, vma); - if (ret < 0) - return ret; - - return exynos_drm_gem_mmap_obj(obj, vma); -} diff --git a/drivers/gpu/drm/exynos/exynos_drm_gem.h b/drivers/gpu/drm/exynos/exynos_drm_gem.h index a23272fb96fb..79d7e1a87419 100644 --- a/drivers/gpu/drm/exynos/exynos_drm_gem.h +++ b/drivers/gpu/drm/exynos/exynos_drm_gem.h @@ -96,9 +96,6 @@ int exynos_drm_gem_dumb_create(struct drm_file *file_priv, struct drm_device *dev, struct drm_mode_create_dumb *args); -/* set vm_flags and we can change the vm attribute to other one at here. */ -int exynos_drm_gem_mmap(struct file *filp, struct vm_area_struct *vma); - /* low-level interface prime helpers */ struct drm_gem_object *exynos_drm_gem_prime_import(struct drm_device *dev, struct dma_buf *dma_buf); @@ -107,7 +104,5 @@ struct drm_gem_object * exynos_drm_gem_prime_import_sg_table(struct drm_device *dev, struct dma_buf_attachment *attach, struct sg_table *sgt); -int exynos_drm_gem_prime_mmap(struct drm_gem_object *obj, - struct vm_area_struct *vma); #endif diff --git a/drivers/gpu/drm/i915/Makefile b/drivers/gpu/drm/i915/Makefile index 6ddd2d2bbaaf..1b62b9f65196 100644 --- a/drivers/gpu/drm/i915/Makefile +++ b/drivers/gpu/drm/i915/Makefile @@ -163,6 +163,7 @@ i915-y += \ i915_active.o \ i915_buddy.o \ i915_cmd_parser.o \ + i915_deps.o \ i915_gem_evict.o \ i915_gem_gtt.o \ i915_gem_ww.o \ diff --git a/drivers/gpu/drm/i915/display/intel_atomic_plane.c b/drivers/gpu/drm/i915/display/intel_atomic_plane.c index 89005628cc3a..c2c512cd8ec0 100644 --- a/drivers/gpu/drm/i915/display/intel_atomic_plane.c +++ b/drivers/gpu/drm/i915/display/intel_atomic_plane.c @@ -819,7 +819,7 @@ intel_prepare_plane_fb(struct drm_plane *_plane, * maximum clocks following a vblank miss (see do_rps_boost()). */ if (!state->rps_interactive) { - intel_rps_mark_interactive(&dev_priv->gt.rps, true); + intel_rps_mark_interactive(&to_gt(dev_priv)->rps, true); state->rps_interactive = true; } @@ -853,7 +853,7 @@ intel_cleanup_plane_fb(struct drm_plane *plane, return; if (state->rps_interactive) { - intel_rps_mark_interactive(&dev_priv->gt.rps, false); + intel_rps_mark_interactive(&to_gt(dev_priv)->rps, false); state->rps_interactive = false; } diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c index 6fbad5c6cc71..bf7ce684dd8e 100644 --- a/drivers/gpu/drm/i915/display/intel_display.c +++ b/drivers/gpu/drm/i915/display/intel_display.c @@ -843,7 +843,7 @@ __intel_display_resume(struct drm_device *dev, static bool gpu_reset_clobbers_display(struct drm_i915_private *dev_priv) { return (INTEL_INFO(dev_priv)->gpu_reset_clobbers_display && - intel_has_gpu_reset(&dev_priv->gt)); + intel_has_gpu_reset(to_gt(dev_priv))); } void intel_display_prepare_reset(struct drm_i915_private *dev_priv) @@ -862,14 +862,14 @@ void intel_display_prepare_reset(struct drm_i915_private *dev_priv) return; /* We have a modeset vs reset deadlock, defensively unbreak it. */ - set_bit(I915_RESET_MODESET, &dev_priv->gt.reset.flags); + set_bit(I915_RESET_MODESET, &to_gt(dev_priv)->reset.flags); smp_mb__after_atomic(); - wake_up_bit(&dev_priv->gt.reset.flags, I915_RESET_MODESET); + wake_up_bit(&to_gt(dev_priv)->reset.flags, I915_RESET_MODESET); if (atomic_read(&dev_priv->gpu_error.pending_fb_pin)) { drm_dbg_kms(&dev_priv->drm, "Modeset potentially stuck, unbreaking through wedging\n"); - intel_gt_set_wedged(&dev_priv->gt); + intel_gt_set_wedged(to_gt(dev_priv)); } /* @@ -920,7 +920,7 @@ void intel_display_finish_reset(struct drm_i915_private *dev_priv) return; /* reset doesn't touch the display */ - if (!test_bit(I915_RESET_MODESET, &dev_priv->gt.reset.flags)) + if (!test_bit(I915_RESET_MODESET, &to_gt(dev_priv)->reset.flags)) return; state = fetch_and_zero(&dev_priv->modeset_restore_state); @@ -958,7 +958,7 @@ unlock: drm_modeset_acquire_fini(ctx); mutex_unlock(&dev->mode_config.mutex); - clear_bit_unlock(I915_RESET_MODESET, &dev_priv->gt.reset.flags); + clear_bit_unlock(I915_RESET_MODESET, &to_gt(dev_priv)->reset.flags); } static void icl_set_pipe_chicken(const struct intel_crtc_state *crtc_state) @@ -8513,19 +8513,19 @@ static void intel_atomic_commit_fence_wait(struct intel_atomic_state *intel_stat for (;;) { prepare_to_wait(&intel_state->commit_ready.wait, &wait_fence, TASK_UNINTERRUPTIBLE); - prepare_to_wait(bit_waitqueue(&dev_priv->gt.reset.flags, + prepare_to_wait(bit_waitqueue(&to_gt(dev_priv)->reset.flags, I915_RESET_MODESET), &wait_reset, TASK_UNINTERRUPTIBLE); if (i915_sw_fence_done(&intel_state->commit_ready) || - test_bit(I915_RESET_MODESET, &dev_priv->gt.reset.flags)) + test_bit(I915_RESET_MODESET, &to_gt(dev_priv)->reset.flags)) break; schedule(); } finish_wait(&intel_state->commit_ready.wait, &wait_fence); - finish_wait(bit_waitqueue(&dev_priv->gt.reset.flags, + finish_wait(bit_waitqueue(&to_gt(dev_priv)->reset.flags, I915_RESET_MODESET), &wait_reset); } diff --git a/drivers/gpu/drm/i915/display/intel_dpt.c b/drivers/gpu/drm/i915/display/intel_dpt.c index 963ca7155b06..8f674745e7e0 100644 --- a/drivers/gpu/drm/i915/display/intel_dpt.c +++ b/drivers/gpu/drm/i915/display/intel_dpt.c @@ -264,7 +264,7 @@ intel_dpt_create(struct intel_framebuffer *fb) vm = &dpt->vm; - vm->gt = &i915->gt; + vm->gt = to_gt(i915); vm->i915 = i915; vm->dma = i915->drm.dev; vm->total = (size / sizeof(gen8_pte_t)) * I915_GTT_PAGE_SIZE; @@ -279,8 +279,6 @@ intel_dpt_create(struct intel_framebuffer *fb) vm->vma_ops.bind_vma = dpt_bind_vma; vm->vma_ops.unbind_vma = dpt_unbind_vma; - vm->vma_ops.set_pages = ggtt_set_pages; - vm->vma_ops.clear_pages = clear_pages; vm->pte_encode = gen8_ggtt_pte_encode; diff --git a/drivers/gpu/drm/i915/display/intel_overlay.c b/drivers/gpu/drm/i915/display/intel_overlay.c index 7e3f5c6ca484..1a376e9a1ff3 100644 --- a/drivers/gpu/drm/i915/display/intel_overlay.c +++ b/drivers/gpu/drm/i915/display/intel_overlay.c @@ -1382,7 +1382,7 @@ void intel_overlay_setup(struct drm_i915_private *dev_priv) if (!HAS_OVERLAY(dev_priv)) return; - engine = dev_priv->gt.engine[RCS0]; + engine = to_gt(dev_priv)->engine[RCS0]; if (!engine || !engine->kernel_context) return; diff --git a/drivers/gpu/drm/i915/display/skl_universal_plane.c b/drivers/gpu/drm/i915/display/skl_universal_plane.c index d5359cf3d270..93a385396512 100644 --- a/drivers/gpu/drm/i915/display/skl_universal_plane.c +++ b/drivers/gpu/drm/i915/display/skl_universal_plane.c @@ -1737,7 +1737,7 @@ static bool bo_has_valid_encryption(struct drm_i915_gem_object *obj) { struct drm_i915_private *i915 = to_i915(obj->base.dev); - return intel_pxp_key_check(&i915->gt.pxp, obj, false) == 0; + return intel_pxp_key_check(&to_gt(i915)->pxp, obj, false) == 0; } static bool pxp_is_borked(struct drm_i915_gem_object *obj) diff --git a/drivers/gpu/drm/i915/gem/i915_gem_context.c b/drivers/gpu/drm/i915/gem/i915_gem_context.c index 347dab952e90..cad3f0b2be9e 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_context.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_context.c @@ -237,7 +237,7 @@ static int proto_context_set_persistence(struct drm_i915_private *i915, * colateral damage, and we should not pretend we can by * exposing the interface. */ - if (!intel_has_reset_engine(&i915->gt)) + if (!intel_has_reset_engine(to_gt(i915))) return -ENODEV; pc->user_flags &= ~BIT(UCONTEXT_PERSISTENCE); @@ -254,7 +254,7 @@ static int proto_context_set_protected(struct drm_i915_private *i915, if (!protected) { pc->uses_protected_content = false; - } else if (!intel_pxp_is_enabled(&i915->gt.pxp)) { + } else if (!intel_pxp_is_enabled(&to_gt(i915)->pxp)) { ret = -ENODEV; } else if ((pc->user_flags & BIT(UCONTEXT_RECOVERABLE)) || !(pc->user_flags & BIT(UCONTEXT_BANNABLE))) { @@ -268,8 +268,8 @@ static int proto_context_set_protected(struct drm_i915_private *i915, */ pc->pxp_wakeref = intel_runtime_pm_get(&i915->runtime_pm); - if (!intel_pxp_is_active(&i915->gt.pxp)) - ret = intel_pxp_start(&i915->gt.pxp); + if (!intel_pxp_is_active(&to_gt(i915)->pxp)) + ret = intel_pxp_start(&to_gt(i915)->pxp); } return ret; @@ -571,7 +571,7 @@ set_proto_ctx_engines_parallel_submit(struct i915_user_extension __user *base, intel_engine_mask_t prev_mask; /* FIXME: This is NIY for execlists */ - if (!(intel_uc_uses_guc_submission(&i915->gt.uc))) + if (!(intel_uc_uses_guc_submission(&to_gt(i915)->uc))) return -ENODEV; if (get_user(slot, &ext->engine_index)) @@ -833,7 +833,7 @@ static int set_proto_ctx_sseu(struct drm_i915_file_private *fpriv, sseu = &pc->legacy_rcs_sseu; } - ret = i915_gem_user_to_context_sseu(&i915->gt, &user_sseu, sseu); + ret = i915_gem_user_to_context_sseu(to_gt(i915), &user_sseu, sseu); if (ret) return ret; @@ -1044,7 +1044,7 @@ static struct i915_gem_engines *alloc_engines(unsigned int count) static struct i915_gem_engines *default_engines(struct i915_gem_context *ctx, struct intel_sseu rcs_sseu) { - const struct intel_gt *gt = &ctx->i915->gt; + const struct intel_gt *gt = to_gt(ctx->i915); struct intel_engine_cs *engine; struct i915_gem_engines *e, *err; enum intel_engine_id id; @@ -1521,7 +1521,7 @@ static int __context_set_persistence(struct i915_gem_context *ctx, bool state) * colateral damage, and we should not pretend we can by * exposing the interface. */ - if (!intel_has_reset_engine(&ctx->i915->gt)) + if (!intel_has_reset_engine(to_gt(ctx->i915))) return -ENODEV; i915_gem_context_clear_persistence(ctx); @@ -1559,7 +1559,7 @@ i915_gem_create_context(struct drm_i915_private *i915, } else if (HAS_FULL_PPGTT(i915)) { struct i915_ppgtt *ppgtt; - ppgtt = i915_ppgtt_create(&i915->gt, 0); + ppgtt = i915_ppgtt_create(to_gt(i915), 0); if (IS_ERR(ppgtt)) { drm_dbg(&i915->drm, "PPGTT setup failed (%ld)\n", PTR_ERR(ppgtt)); @@ -1742,7 +1742,7 @@ int i915_gem_vm_create_ioctl(struct drm_device *dev, void *data, if (args->flags) return -EINVAL; - ppgtt = i915_ppgtt_create(&i915->gt, 0); + ppgtt = i915_ppgtt_create(to_gt(i915), 0); if (IS_ERR(ppgtt)) return PTR_ERR(ppgtt); @@ -2194,7 +2194,7 @@ int i915_gem_context_create_ioctl(struct drm_device *dev, void *data, if (args->flags & I915_CONTEXT_CREATE_FLAGS_UNKNOWN) return -EINVAL; - ret = intel_gt_terminally_wedged(&i915->gt); + ret = intel_gt_terminally_wedged(to_gt(i915)); if (ret) return ret; diff --git a/drivers/gpu/drm/i915/gem/i915_gem_create.c b/drivers/gpu/drm/i915/gem/i915_gem_create.c index 8955d6abcef1..9402d4bf4ffc 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_create.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_create.c @@ -379,7 +379,7 @@ static int ext_set_protected(struct i915_user_extension __user *base, void *data if (ext.flags) return -EINVAL; - if (!intel_pxp_is_enabled(&ext_data->i915->gt.pxp)) + if (!intel_pxp_is_enabled(&to_gt(ext_data->i915)->pxp)) return -ENODEV; ext_data->flags |= I915_BO_PROTECTED; diff --git a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c index 60ee60f7bb09..e7f548a22970 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c @@ -1098,6 +1098,47 @@ static inline struct i915_ggtt *cache_to_ggtt(struct reloc_cache *cache) return &i915->ggtt; } +static void reloc_cache_unmap(struct reloc_cache *cache) +{ + void *vaddr; + + if (!cache->vaddr) + return; + + vaddr = unmask_page(cache->vaddr); + if (cache->vaddr & KMAP) + kunmap_atomic(vaddr); + else + io_mapping_unmap_atomic((void __iomem *)vaddr); +} + +static void reloc_cache_remap(struct reloc_cache *cache, + struct drm_i915_gem_object *obj) +{ + void *vaddr; + + if (!cache->vaddr) + return; + + if (cache->vaddr & KMAP) { + struct page *page = i915_gem_object_get_page(obj, cache->page); + + vaddr = kmap_atomic(page); + cache->vaddr = unmask_flags(cache->vaddr) | + (unsigned long)vaddr; + } else { + struct i915_ggtt *ggtt = cache_to_ggtt(cache); + unsigned long offset; + + offset = cache->node.start; + if (!drm_mm_node_allocated(&cache->node)) + offset += cache->page << PAGE_SHIFT; + + cache->vaddr = (unsigned long) + io_mapping_map_atomic_wc(&ggtt->iomap, offset); + } +} + static void reloc_cache_reset(struct reloc_cache *cache, struct i915_execbuffer *eb) { void *vaddr; @@ -1362,10 +1403,17 @@ eb_relocate_entry(struct i915_execbuffer *eb, * batchbuffers. */ if (reloc->write_domain == I915_GEM_DOMAIN_INSTRUCTION && - GRAPHICS_VER(eb->i915) == 6) { + GRAPHICS_VER(eb->i915) == 6 && + !i915_vma_is_bound(target->vma, I915_VMA_GLOBAL_BIND)) { + struct i915_vma *vma = target->vma; + + reloc_cache_unmap(&eb->reloc_cache); + mutex_lock(&vma->vm->mutex); err = i915_vma_bind(target->vma, target->vma->obj->cache_level, PIN_GLOBAL, NULL); + mutex_unlock(&vma->vm->mutex); + reloc_cache_remap(&eb->reloc_cache, ev->vma->obj); if (err) return err; } @@ -2361,9 +2409,9 @@ static int eb_submit(struct i915_execbuffer *eb) return err; } -static int num_vcs_engines(const struct drm_i915_private *i915) +static int num_vcs_engines(struct drm_i915_private *i915) { - return hweight_long(VDBOX_MASK(&i915->gt)); + return hweight_long(VDBOX_MASK(to_gt(i915))); } /* diff --git a/drivers/gpu/drm/i915/gem/i915_gem_mman.c b/drivers/gpu/drm/i915/gem/i915_gem_mman.c index 39bb15eafc07..aaf970c37aa2 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_mman.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_mman.c @@ -73,7 +73,7 @@ i915_gem_mmap_ioctl(struct drm_device *dev, void *data, if (args->flags & ~(I915_MMAP_WC)) return -EINVAL; - if (args->flags & I915_MMAP_WC && !boot_cpu_has(X86_FEATURE_PAT)) + if (args->flags & I915_MMAP_WC && !pat_enabled()) return -ENODEV; obj = i915_gem_object_lookup(file, args->handle); @@ -646,7 +646,7 @@ mmap_offset_attach(struct drm_i915_gem_object *obj, goto insert; /* Attempt to reap some mmap space from dead objects */ - err = intel_gt_retire_requests_timeout(&i915->gt, MAX_SCHEDULE_TIMEOUT, + err = intel_gt_retire_requests_timeout(to_gt(i915), MAX_SCHEDULE_TIMEOUT, NULL); if (err) goto err; @@ -737,7 +737,7 @@ i915_gem_dumb_mmap_offset(struct drm_file *file, if (HAS_LMEM(to_i915(dev))) mmap_type = I915_MMAP_TYPE_FIXED; - else if (boot_cpu_has(X86_FEATURE_PAT)) + else if (pat_enabled()) mmap_type = I915_MMAP_TYPE_WC; else if (!i915_ggtt_has_aperture(&to_i915(dev)->ggtt)) return -ENODEV; @@ -793,7 +793,7 @@ i915_gem_mmap_offset_ioctl(struct drm_device *dev, void *data, break; case I915_MMAP_OFFSET_WC: - if (!boot_cpu_has(X86_FEATURE_PAT)) + if (!pat_enabled()) return -ENODEV; type = I915_MMAP_TYPE_WC; break; @@ -803,7 +803,7 @@ i915_gem_mmap_offset_ioctl(struct drm_device *dev, void *data, break; case I915_MMAP_OFFSET_UC: - if (!boot_cpu_has(X86_FEATURE_PAT)) + if (!pat_enabled()) return -ENODEV; type = I915_MMAP_TYPE_UC; break; diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object.c b/drivers/gpu/drm/i915/gem/i915_gem_object.c index 5fac9b560b73..d87b508b59b1 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_object.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_object.c @@ -262,6 +262,8 @@ static void __i915_gem_object_free_mmaps(struct drm_i915_gem_object *obj) */ void __i915_gem_object_pages_fini(struct drm_i915_gem_object *obj) { + assert_object_held(obj); + if (!list_empty(&obj->vma.list)) { struct i915_vma *vma; @@ -328,7 +330,16 @@ static void __i915_gem_free_objects(struct drm_i915_private *i915, obj->ops->delayed_free(obj); continue; } + + if (!i915_gem_object_trylock(obj, NULL)) { + /* busy, toss it back to the pile */ + if (llist_add(&obj->freed, &i915->mm.free_list)) + queue_delayed_work(i915->wq, &i915->mm.free_work, msecs_to_jiffies(10)); + continue; + } + __i915_gem_object_pages_fini(obj); + i915_gem_object_unlock(obj); __i915_gem_free_object(obj); /* But keep the pointer alive for RCU-protected lookups */ @@ -348,7 +359,7 @@ void i915_gem_flush_free_objects(struct drm_i915_private *i915) static void __i915_gem_free_work(struct work_struct *work) { struct drm_i915_private *i915 = - container_of(work, struct drm_i915_private, mm.free_work); + container_of(work, struct drm_i915_private, mm.free_work.work); i915_gem_flush_free_objects(i915); } @@ -380,7 +391,7 @@ static void i915_gem_free_object(struct drm_gem_object *gem_obj) */ if (llist_add(&obj->freed, &i915->mm.free_list)) - queue_work(i915->wq, &i915->mm.free_work); + queue_delayed_work(i915->wq, &i915->mm.free_work, 0); } void __i915_gem_object_flush_frontbuffer(struct drm_i915_gem_object *obj, @@ -705,7 +716,7 @@ bool i915_gem_object_placement_possible(struct drm_i915_gem_object *obj, void i915_gem_init__objects(struct drm_i915_private *i915) { - INIT_WORK(&i915->mm.free_work, __i915_gem_free_work); + INIT_DELAYED_WORK(&i915->mm.free_work, __i915_gem_free_work); } void i915_objects_module_exit(void) diff --git a/drivers/gpu/drm/i915/gem/i915_gem_object.h b/drivers/gpu/drm/i915/gem/i915_gem_object.h index 66f20b803b01..f66d46882ea7 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_object.h +++ b/drivers/gpu/drm/i915/gem/i915_gem_object.h @@ -210,9 +210,13 @@ static inline int i915_gem_object_lock_interruptible(struct drm_i915_gem_object return __i915_gem_object_lock(obj, ww, true); } -static inline bool i915_gem_object_trylock(struct drm_i915_gem_object *obj) +static inline bool i915_gem_object_trylock(struct drm_i915_gem_object *obj, + struct i915_gem_ww_ctx *ww) { - return dma_resv_trylock(obj->base.resv); + if (!ww) + return dma_resv_trylock(obj->base.resv); + else + return ww_mutex_trylock(&obj->base.resv->lock, &ww->ctx); } static inline void i915_gem_object_unlock(struct drm_i915_gem_object *obj) diff --git a/drivers/gpu/drm/i915/gem/i915_gem_pages.c b/drivers/gpu/drm/i915/gem/i915_gem_pages.c index 49c6e55c68ce..89b70f5cde7a 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_pages.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_pages.c @@ -424,8 +424,7 @@ void *i915_gem_object_pin_map(struct drm_i915_gem_object *obj, goto err_unpin; } - if (GEM_WARN_ON(type == I915_MAP_WC && - !static_cpu_has(X86_FEATURE_PAT))) + if (GEM_WARN_ON(type == I915_MAP_WC && !pat_enabled())) ptr = ERR_PTR(-ENODEV); else if (i915_gem_object_has_struct_page(obj)) ptr = i915_gem_object_map_page(obj, type); diff --git a/drivers/gpu/drm/i915/gem/i915_gem_phys.c b/drivers/gpu/drm/i915/gem/i915_gem_phys.c index 7986612f48fa..ca6faffcc496 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_phys.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_phys.c @@ -19,6 +19,7 @@ static int i915_gem_object_get_pages_phys(struct drm_i915_gem_object *obj) { struct address_space *mapping = obj->base.filp->f_mapping; + struct drm_i915_private *i915 = to_i915(obj->base.dev); struct scatterlist *sg; struct sg_table *st; dma_addr_t dma; @@ -73,7 +74,7 @@ static int i915_gem_object_get_pages_phys(struct drm_i915_gem_object *obj) dst += PAGE_SIZE; } - intel_gt_chipset_flush(&to_i915(obj->base.dev)->gt); + intel_gt_chipset_flush(to_gt(i915)); /* We're no longer struct page backed */ obj->mem_flags &= ~I915_BO_FLAG_STRUCT_PAGE; @@ -140,6 +141,7 @@ int i915_gem_object_pwrite_phys(struct drm_i915_gem_object *obj, { void *vaddr = sg_page(obj->mm.pages->sgl) + args->offset; char __user *user_data = u64_to_user_ptr(args->data_ptr); + struct drm_i915_private *i915 = to_i915(obj->base.dev); int err; err = i915_gem_object_wait(obj, @@ -159,7 +161,7 @@ int i915_gem_object_pwrite_phys(struct drm_i915_gem_object *obj, return -EFAULT; drm_clflush_virt_range(vaddr, args->size); - intel_gt_chipset_flush(&to_i915(obj->base.dev)->gt); + intel_gt_chipset_flush(to_gt(i915)); i915_gem_object_flush_frontbuffer(obj, ORIGIN_CPU); return 0; diff --git a/drivers/gpu/drm/i915/gem/i915_gem_pm.c b/drivers/gpu/drm/i915/gem/i915_gem_pm.c index 726b40e1fbb0..ac56124760e1 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_pm.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_pm.c @@ -35,7 +35,7 @@ void i915_gem_suspend(struct drm_i915_private *i915) * state. Fortunately, the kernel_context is disposable and we do * not rely on its state. */ - intel_gt_suspend_prepare(&i915->gt); + intel_gt_suspend_prepare(to_gt(i915)); i915_gem_drain_freed_objects(i915); } @@ -153,7 +153,7 @@ void i915_gem_suspend_late(struct drm_i915_private *i915) * machine in an unusable condition. */ - intel_gt_suspend_late(&i915->gt); + intel_gt_suspend_late(to_gt(i915)); spin_lock_irqsave(&i915->mm.obj_lock, flags); for (phase = phases; *phase; phase++) { @@ -223,7 +223,7 @@ void i915_gem_resume(struct drm_i915_private *i915) * guarantee that the context image is complete. So let's just reset * it and start again. */ - intel_gt_resume(&i915->gt); + intel_gt_resume(to_gt(i915)); ret = lmem_restore(i915, I915_TTM_BACKUP_ALLOW_GPU); GEM_WARN_ON(ret); diff --git a/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c b/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c index 157a9765f483..cc927e49d21f 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_shrinker.c @@ -36,8 +36,8 @@ static bool can_release_pages(struct drm_i915_gem_object *obj) return swap_available() || obj->mm.madv == I915_MADV_DONTNEED; } -static bool unsafe_drop_pages(struct drm_i915_gem_object *obj, - unsigned long shrink, bool trylock_vm) +static int drop_pages(struct drm_i915_gem_object *obj, + unsigned long shrink, bool trylock_vm) { unsigned long flags; @@ -153,7 +153,7 @@ i915_gem_shrink(struct i915_gem_ww_ctx *ww, */ if (shrink & I915_SHRINK_ACTIVE) /* Retire requests to unpin all idle contexts */ - intel_gt_retire_requests(&i915->gt); + intel_gt_retire_requests(to_gt(i915)); /* * As we may completely rewrite the (un)bound list whilst unbinding @@ -214,26 +214,24 @@ i915_gem_shrink(struct i915_gem_ww_ctx *ww, spin_unlock_irqrestore(&i915->mm.obj_lock, flags); - err = 0; - if (unsafe_drop_pages(obj, shrink, trylock_vm)) { - /* May arrive from get_pages on another bo */ - if (!ww) { - if (!i915_gem_object_trylock(obj)) - goto skip; - } else { - err = i915_gem_object_lock(obj, ww); - if (err) - goto skip; - } - - if (!__i915_gem_object_put_pages(obj)) { - if (!try_to_writeback(obj, shrink)) - count += obj->base.size >> PAGE_SHIFT; - } - if (!ww) - i915_gem_object_unlock(obj); + /* May arrive from get_pages on another bo */ + if (!ww) { + if (!i915_gem_object_trylock(obj, NULL)) + goto skip; + } else { + err = i915_gem_object_lock(obj, ww); + if (err) + goto skip; } + if (drop_pages(obj, shrink, trylock_vm) && + !__i915_gem_object_put_pages(obj) && + !try_to_writeback(obj, shrink)) + count += obj->base.size >> PAGE_SHIFT; + + if (!ww) + i915_gem_object_unlock(obj); + scanned += obj->base.size >> PAGE_SHIFT; skip: i915_gem_object_put(obj); @@ -407,12 +405,18 @@ i915_gem_shrinker_vmap(struct notifier_block *nb, unsigned long event, void *ptr list_for_each_entry_safe(vma, next, &i915->ggtt.vm.bound_list, vm_link) { unsigned long count = vma->node.size >> PAGE_SHIFT; + struct drm_i915_gem_object *obj = vma->obj; if (!vma->iomap || i915_vma_is_active(vma)) continue; + if (!i915_gem_object_trylock(obj, NULL)) + continue; + if (__i915_vma_unbind(vma) == 0) freed_pages += count; + + i915_gem_object_unlock(obj); } mutex_unlock(&i915->ggtt.vm.mutex); diff --git a/drivers/gpu/drm/i915/gem/i915_gem_stolen.c b/drivers/gpu/drm/i915/gem/i915_gem_stolen.c index bce03d74a0b4..7df50fd6cc7b 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_stolen.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_stolen.c @@ -488,6 +488,9 @@ static int i915_gem_init_stolen(struct intel_memory_region *mem) return 0; } + /* Exclude the reserved region from driver use */ + mem->region.end = reserved_base - 1; + /* It is possible for the reserved area to end before the end of stolen * memory, so just consider the start. */ reserved_total = stolen_top - reserved_base; @@ -653,7 +656,7 @@ static int __i915_gem_object_create_stolen(struct intel_memory_region *mem, cache_level = HAS_LLC(mem->i915) ? I915_CACHE_LLC : I915_CACHE_NONE; i915_gem_object_set_cache_coherency(obj, cache_level); - if (WARN_ON(!i915_gem_object_trylock(obj))) + if (WARN_ON(!i915_gem_object_trylock(obj, NULL))) return -EBUSY; i915_gem_object_init_memory_region(obj, mem); @@ -780,6 +783,7 @@ i915_gem_stolen_lmem_setup(struct drm_i915_private *i915, u16 type, struct intel_uncore *uncore = &i915->uncore; struct pci_dev *pdev = to_pci_dev(i915->drm.dev); struct intel_memory_region *mem; + resource_size_t min_page_size; resource_size_t io_start; resource_size_t lmem_size; u64 lmem_base; @@ -791,8 +795,11 @@ i915_gem_stolen_lmem_setup(struct drm_i915_private *i915, u16 type, lmem_size = pci_resource_len(pdev, 2) - lmem_base; io_start = pci_resource_start(pdev, 2) + lmem_base; + min_page_size = HAS_64K_PAGES(i915) ? I915_GTT_PAGE_SIZE_64K : + I915_GTT_PAGE_SIZE_4K; + mem = intel_memory_region_create(i915, lmem_base, lmem_size, - I915_GTT_PAGE_SIZE_4K, io_start, + min_page_size, io_start, type, instance, &i915_region_stolen_lmem_ops); if (IS_ERR(mem)) diff --git a/drivers/gpu/drm/i915/gem/i915_gem_throttle.c b/drivers/gpu/drm/i915/gem/i915_gem_throttle.c index 1929d6cf4150..75501db71041 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_throttle.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_throttle.c @@ -38,12 +38,13 @@ i915_gem_throttle_ioctl(struct drm_device *dev, void *data, { const unsigned long recent_enough = jiffies - DRM_I915_THROTTLE_JIFFIES; struct drm_i915_file_private *file_priv = file->driver_priv; + struct drm_i915_private *i915 = to_i915(dev); struct i915_gem_context *ctx; unsigned long idx; long ret; /* ABI: return -EIO if already wedged */ - ret = intel_gt_terminally_wedged(&to_i915(dev)->gt); + ret = intel_gt_terminally_wedged(to_gt(i915)); if (ret) return ret; diff --git a/drivers/gpu/drm/i915/gem/i915_gem_ttm.c b/drivers/gpu/drm/i915/gem/i915_gem_ttm.c index 218a9b3037c7..923cc7ad8d70 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_ttm.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_ttm.c @@ -166,7 +166,7 @@ static int i915_ttm_tt_shmem_populate(struct ttm_device *bdev, struct intel_memory_region *mr = i915->mm.regions[INTEL_MEMORY_SYSTEM]; struct i915_ttm_tt *i915_tt = container_of(ttm, typeof(*i915_tt), ttm); const unsigned int max_segment = i915_sg_segment_size(); - const size_t size = ttm->num_pages << PAGE_SHIFT; + const size_t size = (size_t)ttm->num_pages << PAGE_SHIFT; struct file *filp = i915_tt->filp; struct sgt_iter sgt_iter; struct sg_table *st; diff --git a/drivers/gpu/drm/i915/gem/i915_gem_ttm_move.c b/drivers/gpu/drm/i915/gem/i915_gem_ttm_move.c index 80df9f592407..ee9612a3ee5e 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_ttm_move.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_ttm_move.c @@ -3,10 +3,9 @@ * Copyright © 2021 Intel Corporation */ -#include <linux/dma-fence-array.h> - #include <drm/ttm/ttm_bo_driver.h> +#include "i915_deps.h" #include "i915_drv.h" #include "intel_memory_region.h" #include "intel_region_ttm.h" @@ -43,234 +42,6 @@ void i915_ttm_migrate_set_failure_modes(bool gpu_migration, } #endif -/** - * DOC: Set of utilities to dynamically collect dependencies and - * eventually coalesce them into a single fence which is fed into - * the GT migration code, since it only accepts a single dependency - * fence. - * The single fence returned from these utilities, in the case of - * dependencies from multiple fence contexts, a struct dma_fence_array, - * since the i915 request code can break that up and await the individual - * fences. - * - * Once we can do async unbinding, this is also needed to coalesce - * the migration fence with the unbind fences. - * - * While collecting the individual dependencies, we store the refcounted - * struct dma_fence pointers in a realloc-managed pointer array, since - * that can be easily fed into a dma_fence_array. Other options are - * available, like for example an xarray for similarity with drm/sched. - * Can be changed easily if needed. - * - * A struct i915_deps need to be initialized using i915_deps_init(). - * If i915_deps_add_dependency() or i915_deps_add_resv() return an - * error code they will internally call i915_deps_fini(), which frees - * all internal references and allocations. After a call to - * i915_deps_to_fence(), or i915_deps_sync(), the struct should similarly - * be viewed as uninitialized. - * - * We might want to break this out into a separate file as a utility. - */ - -#define I915_DEPS_MIN_ALLOC_CHUNK 8U - -/** - * struct i915_deps - Collect dependencies into a single dma-fence - * @single: Storage for pointer if the collection is a single fence. - * @fence: Allocated array of fence pointers if more than a single fence; - * otherwise points to the address of @single. - * @num_deps: Current number of dependency fences. - * @fences_size: Size of the @fences array in number of pointers. - * @gfp: Allocation mode. - */ -struct i915_deps { - struct dma_fence *single; - struct dma_fence **fences; - unsigned int num_deps; - unsigned int fences_size; - gfp_t gfp; -}; - -static void i915_deps_reset_fences(struct i915_deps *deps) -{ - if (deps->fences != &deps->single) - kfree(deps->fences); - deps->num_deps = 0; - deps->fences_size = 1; - deps->fences = &deps->single; -} - -static void i915_deps_init(struct i915_deps *deps, gfp_t gfp) -{ - deps->fences = NULL; - deps->gfp = gfp; - i915_deps_reset_fences(deps); -} - -static void i915_deps_fini(struct i915_deps *deps) -{ - unsigned int i; - - for (i = 0; i < deps->num_deps; ++i) - dma_fence_put(deps->fences[i]); - - if (deps->fences != &deps->single) - kfree(deps->fences); -} - -static int i915_deps_grow(struct i915_deps *deps, struct dma_fence *fence, - const struct ttm_operation_ctx *ctx) -{ - int ret; - - if (deps->num_deps >= deps->fences_size) { - unsigned int new_size = 2 * deps->fences_size; - struct dma_fence **new_fences; - - new_size = max(new_size, I915_DEPS_MIN_ALLOC_CHUNK); - new_fences = kmalloc_array(new_size, sizeof(*new_fences), deps->gfp); - if (!new_fences) - goto sync; - - memcpy(new_fences, deps->fences, - deps->fences_size * sizeof(*new_fences)); - swap(new_fences, deps->fences); - if (new_fences != &deps->single) - kfree(new_fences); - deps->fences_size = new_size; - } - deps->fences[deps->num_deps++] = dma_fence_get(fence); - return 0; - -sync: - if (ctx->no_wait_gpu && !dma_fence_is_signaled(fence)) { - ret = -EBUSY; - goto unref; - } - - ret = dma_fence_wait(fence, ctx->interruptible); - if (ret) - goto unref; - - ret = fence->error; - if (ret) - goto unref; - - return 0; - -unref: - i915_deps_fini(deps); - return ret; -} - -static int i915_deps_sync(struct i915_deps *deps, - const struct ttm_operation_ctx *ctx) -{ - struct dma_fence **fences = deps->fences; - unsigned int i; - int ret = 0; - - for (i = 0; i < deps->num_deps; ++i, ++fences) { - if (ctx->no_wait_gpu && !dma_fence_is_signaled(*fences)) { - ret = -EBUSY; - break; - } - - ret = dma_fence_wait(*fences, ctx->interruptible); - if (!ret) - ret = (*fences)->error; - if (ret) - break; - } - - i915_deps_fini(deps); - return ret; -} - -static int i915_deps_add_dependency(struct i915_deps *deps, - struct dma_fence *fence, - const struct ttm_operation_ctx *ctx) -{ - unsigned int i; - int ret; - - if (!fence) - return 0; - - if (dma_fence_is_signaled(fence)) { - ret = fence->error; - if (ret) - i915_deps_fini(deps); - return ret; - } - - for (i = 0; i < deps->num_deps; ++i) { - struct dma_fence *entry = deps->fences[i]; - - if (!entry->context || entry->context != fence->context) - continue; - - if (dma_fence_is_later(fence, entry)) { - dma_fence_put(entry); - deps->fences[i] = dma_fence_get(fence); - } - - return 0; - } - - return i915_deps_grow(deps, fence, ctx); -} - -static struct dma_fence *i915_deps_to_fence(struct i915_deps *deps, - const struct ttm_operation_ctx *ctx) -{ - struct dma_fence_array *array; - - if (deps->num_deps == 0) - return NULL; - - if (deps->num_deps == 1) { - deps->num_deps = 0; - return deps->fences[0]; - } - - /* - * TODO: Alter the allocation mode here to not try too hard to - * make things async. - */ - array = dma_fence_array_create(deps->num_deps, deps->fences, 0, 0, - false); - if (!array) - return ERR_PTR(i915_deps_sync(deps, ctx)); - - deps->fences = NULL; - i915_deps_reset_fences(deps); - - return &array->base; -} - -static int i915_deps_add_resv(struct i915_deps *deps, struct dma_resv *resv, - bool all, const bool no_excl, - const struct ttm_operation_ctx *ctx) -{ - struct dma_resv_iter iter; - struct dma_fence *fence; - - dma_resv_assert_held(resv); - dma_resv_for_each_fence(&iter, resv, all, fence) { - int ret; - - if (no_excl && dma_resv_iter_is_exclusive(&iter)) - continue; - - ret = i915_deps_add_dependency(deps, fence, ctx); - if (ret) - return ret; - } - - return 0; -} - static enum i915_cache_level i915_ttm_cache_level(struct drm_i915_private *i915, struct ttm_resource *res, struct ttm_tt *ttm) @@ -387,7 +158,7 @@ static struct dma_fence *i915_ttm_accel_move(struct ttm_buffer_object *bo, struct ttm_resource *dst_mem, struct ttm_tt *dst_ttm, struct sg_table *dst_st, - struct dma_fence *dep) + const struct i915_deps *deps) { struct drm_i915_private *i915 = container_of(bo->bdev, typeof(*i915), bdev); @@ -397,7 +168,7 @@ static struct dma_fence *i915_ttm_accel_move(struct ttm_buffer_object *bo, enum i915_cache_level src_level, dst_level; int ret; - if (!i915->gt.migrate.context || intel_gt_is_wedged(&i915->gt)) + if (!to_gt(i915)->migrate.context || intel_gt_is_wedged(to_gt(i915))) return ERR_PTR(-EINVAL); /* With fail_gpu_migration, we always perform a GPU clear. */ @@ -410,8 +181,8 @@ static struct dma_fence *i915_ttm_accel_move(struct ttm_buffer_object *bo, !I915_SELFTEST_ONLY(fail_gpu_migration)) return ERR_PTR(-EINVAL); - intel_engine_pm_get(i915->gt.migrate.context->engine); - ret = intel_context_migrate_clear(i915->gt.migrate.context, dep, + intel_engine_pm_get(to_gt(i915)->migrate.context->engine); + ret = intel_context_migrate_clear(to_gt(i915)->migrate.context, deps, dst_st->sgl, dst_level, i915_ttm_gtt_binds_lmem(dst_mem), 0, &rq); @@ -423,9 +194,9 @@ static struct dma_fence *i915_ttm_accel_move(struct ttm_buffer_object *bo, return ERR_CAST(src_rsgt); src_level = i915_ttm_cache_level(i915, bo->resource, src_ttm); - intel_engine_pm_get(i915->gt.migrate.context->engine); - ret = intel_context_migrate_copy(i915->gt.migrate.context, - dep, src_rsgt->table.sgl, + intel_engine_pm_get(to_gt(i915)->migrate.context->engine); + ret = intel_context_migrate_copy(to_gt(i915)->migrate.context, + deps, src_rsgt->table.sgl, src_level, i915_ttm_gtt_binds_lmem(bo->resource), dst_st->sgl, dst_level, @@ -435,7 +206,7 @@ static struct dma_fence *i915_ttm_accel_move(struct ttm_buffer_object *bo, i915_refct_sgt_put(src_rsgt); } - intel_engine_pm_put(i915->gt.migrate.context->engine); + intel_engine_pm_put(to_gt(i915)->migrate.context->engine); if (ret && rq) { i915_request_wait(rq, 0, MAX_SCHEDULE_TIMEOUT); @@ -610,10 +381,11 @@ i915_ttm_memcpy_work_arm(struct i915_ttm_memcpy_work *work, } static struct dma_fence * -__i915_ttm_move(struct ttm_buffer_object *bo, bool clear, +__i915_ttm_move(struct ttm_buffer_object *bo, + const struct ttm_operation_ctx *ctx, bool clear, struct ttm_resource *dst_mem, struct ttm_tt *dst_ttm, struct i915_refct_sgt *dst_rsgt, bool allow_accel, - struct dma_fence *move_dep) + const struct i915_deps *move_deps) { struct i915_ttm_memcpy_work *copy_work = NULL; struct i915_ttm_memcpy_arg _arg, *arg = &_arg; @@ -621,7 +393,7 @@ __i915_ttm_move(struct ttm_buffer_object *bo, bool clear, if (allow_accel) { fence = i915_ttm_accel_move(bo, clear, dst_mem, dst_ttm, - &dst_rsgt->table, move_dep); + &dst_rsgt->table, move_deps); /* * We only need to intercept the error when moving to lmem. @@ -655,8 +427,8 @@ __i915_ttm_move(struct ttm_buffer_object *bo, bool clear, if (!IS_ERR(fence)) goto out; - } else if (move_dep) { - int err = dma_fence_wait(move_dep, true); + } else if (move_deps) { + int err = i915_deps_sync(move_deps, ctx); if (err) return ERR_PTR(err); @@ -680,29 +452,17 @@ out: return fence; } -static struct dma_fence *prev_fence(struct ttm_buffer_object *bo, - struct ttm_operation_ctx *ctx) +static int +prev_deps(struct ttm_buffer_object *bo, struct ttm_operation_ctx *ctx, + struct i915_deps *deps) { - struct i915_deps deps; int ret; - /* - * Instead of trying hard with GFP_KERNEL to allocate memory, - * the dependency collection will just sync if it doesn't - * succeed. - */ - i915_deps_init(&deps, GFP_KERNEL | __GFP_NORETRY | __GFP_NOWARN); - ret = i915_deps_add_dependency(&deps, bo->moving, ctx); + ret = i915_deps_add_dependency(deps, bo->moving, ctx); if (!ret) - /* - * TODO: Only await excl fence here, and shared fences before - * signaling the migration fence. - */ - ret = i915_deps_add_resv(&deps, bo->base.resv, true, false, ctx); - if (ret) - return ERR_PTR(ret); + ret = i915_deps_add_resv(deps, bo->base.resv, ctx); - return i915_deps_to_fence(&deps, ctx); + return ret; } /** @@ -756,16 +516,18 @@ int i915_ttm_move(struct ttm_buffer_object *bo, bool evict, clear = !i915_ttm_cpu_maps_iomem(bo->resource) && (!ttm || !ttm_tt_is_populated(ttm)); if (!(clear && ttm && !(ttm->page_flags & TTM_TT_FLAG_ZERO_ALLOC))) { - struct dma_fence *dep = prev_fence(bo, ctx); + struct i915_deps deps; - if (IS_ERR(dep)) { + i915_deps_init(&deps, GFP_KERNEL | __GFP_NORETRY | __GFP_NOWARN); + ret = prev_deps(bo, ctx, &deps); + if (ret) { i915_refct_sgt_put(dst_rsgt); - return PTR_ERR(dep); + return ret; } - migration_fence = __i915_ttm_move(bo, clear, dst_mem, bo->ttm, - dst_rsgt, true, dep); - dma_fence_put(dep); + migration_fence = __i915_ttm_move(bo, ctx, clear, dst_mem, bo->ttm, + dst_rsgt, true, &deps); + i915_deps_fini(&deps); } /* We can possibly get an -ERESTARTSYS here */ @@ -826,47 +588,38 @@ int i915_gem_obj_copy_ttm(struct drm_i915_gem_object *dst, .interruptible = intr, }; struct i915_refct_sgt *dst_rsgt; - struct dma_fence *copy_fence, *dep_fence; + struct dma_fence *copy_fence; struct i915_deps deps; - int ret, shared_err; + int ret; assert_object_held(dst); assert_object_held(src); i915_deps_init(&deps, GFP_KERNEL | __GFP_NORETRY | __GFP_NOWARN); - /* - * We plan to add a shared fence only for the source. If that - * fails, we await all source fences before commencing - * the copy instead of only the exclusive. - */ - shared_err = dma_resv_reserve_shared(src_bo->base.resv, 1); - ret = i915_deps_add_resv(&deps, dst_bo->base.resv, true, false, &ctx); - if (!ret) - ret = i915_deps_add_resv(&deps, src_bo->base.resv, - !!shared_err, false, &ctx); + ret = dma_resv_reserve_shared(src_bo->base.resv, 1); if (ret) return ret; - dep_fence = i915_deps_to_fence(&deps, &ctx); - if (IS_ERR(dep_fence)) - return PTR_ERR(dep_fence); + ret = i915_deps_add_resv(&deps, dst_bo->base.resv, &ctx); + if (ret) + return ret; + + ret = i915_deps_add_resv(&deps, src_bo->base.resv, &ctx); + if (ret) + return ret; dst_rsgt = i915_ttm_resource_get_st(dst, dst_bo->resource); - copy_fence = __i915_ttm_move(src_bo, false, dst_bo->resource, + copy_fence = __i915_ttm_move(src_bo, &ctx, false, dst_bo->resource, dst_bo->ttm, dst_rsgt, allow_accel, - dep_fence); + &deps); + i915_deps_fini(&deps); i915_refct_sgt_put(dst_rsgt); if (IS_ERR_OR_NULL(copy_fence)) return PTR_ERR_OR_ZERO(copy_fence); dma_resv_add_excl_fence(dst_bo->base.resv, copy_fence); - - /* If we failed to reserve a shared slot, add an exclusive fence */ - if (shared_err) - dma_resv_add_excl_fence(src_bo->base.resv, copy_fence); - else - dma_resv_add_shared_fence(src_bo->base.resv, copy_fence); + dma_resv_add_shared_fence(src_bo->base.resv, copy_fence); dma_fence_put(copy_fence); diff --git a/drivers/gpu/drm/i915/gem/i915_gem_userptr.c b/drivers/gpu/drm/i915/gem/i915_gem_userptr.c index 3173c9f9a040..3cc01c30dd62 100644 --- a/drivers/gpu/drm/i915/gem/i915_gem_userptr.c +++ b/drivers/gpu/drm/i915/gem/i915_gem_userptr.c @@ -529,7 +529,7 @@ i915_gem_userptr_ioctl(struct drm_device *dev, * On almost all of the older hw, we cannot tell the GPU that * a page is readonly. */ - if (!dev_priv->gt.vm->has_read_only) + if (!to_gt(dev_priv)->vm->has_read_only) return -ENODEV; } diff --git a/drivers/gpu/drm/i915/gem/selftests/huge_pages.c b/drivers/gpu/drm/i915/gem/selftests/huge_pages.c index c69c7d45aabc..11f0aa65f8a3 100644 --- a/drivers/gpu/drm/i915/gem/selftests/huge_pages.c +++ b/drivers/gpu/drm/i915/gem/selftests/huge_pages.c @@ -1705,7 +1705,7 @@ int i915_gem_huge_page_mock_selftests(void) mkwrite_device_info(dev_priv)->ppgtt_type = INTEL_PPGTT_FULL; mkwrite_device_info(dev_priv)->ppgtt_size = 48; - ppgtt = i915_ppgtt_create(&dev_priv->gt, 0); + ppgtt = i915_ppgtt_create(to_gt(dev_priv), 0); if (IS_ERR(ppgtt)) { err = PTR_ERR(ppgtt); goto out_unlock; @@ -1747,7 +1747,7 @@ int i915_gem_huge_page_live_selftests(struct drm_i915_private *i915) return 0; } - if (intel_gt_is_wedged(&i915->gt)) + if (intel_gt_is_wedged(to_gt(i915))) return 0; return i915_live_subtests(tests, i915); diff --git a/drivers/gpu/drm/i915/gem/selftests/i915_gem_client_blt.c b/drivers/gpu/drm/i915/gem/selftests/i915_gem_client_blt.c index 8402ed925a69..75947e9dada2 100644 --- a/drivers/gpu/drm/i915/gem/selftests/i915_gem_client_blt.c +++ b/drivers/gpu/drm/i915/gem/selftests/i915_gem_client_blt.c @@ -592,7 +592,7 @@ int i915_gem_client_blt_live_selftests(struct drm_i915_private *i915) SUBTEST(igt_client_tiled_blits), }; - if (intel_gt_is_wedged(&i915->gt)) + if (intel_gt_is_wedged(to_gt(i915))) return 0; return i915_live_subtests(tests, i915); diff --git a/drivers/gpu/drm/i915/gem/selftests/i915_gem_context.c b/drivers/gpu/drm/i915/gem/selftests/i915_gem_context.c index 21b71568cd5f..3f41fe5ec9d4 100644 --- a/drivers/gpu/drm/i915/gem/selftests/i915_gem_context.c +++ b/drivers/gpu/drm/i915/gem/selftests/i915_gem_context.c @@ -90,7 +90,7 @@ static int live_nop_switch(void *arg) } if (i915_request_wait(rq, 0, 10 * HZ) < 0) { pr_err("Failed to populated %d contexts\n", nctx); - intel_gt_set_wedged(&i915->gt); + intel_gt_set_wedged(to_gt(i915)); i915_request_put(rq); err = -EIO; goto out_file; @@ -146,7 +146,7 @@ static int live_nop_switch(void *arg) if (i915_request_wait(rq, 0, HZ / 5) < 0) { pr_err("Switching between %ld contexts timed out\n", prime); - intel_gt_set_wedged(&i915->gt); + intel_gt_set_wedged(to_gt(i915)); i915_request_put(rq); break; } @@ -1223,7 +1223,7 @@ __igt_ctx_sseu(struct drm_i915_private *i915, return 0; if (flags & TEST_RESET) - igt_global_reset_lock(&i915->gt); + igt_global_reset_lock(to_gt(i915)); obj = i915_gem_object_create_internal(i915, PAGE_SIZE); if (IS_ERR(obj)) { @@ -1306,7 +1306,7 @@ out_put: out_unlock: if (flags & TEST_RESET) - igt_global_reset_unlock(&i915->gt); + igt_global_reset_unlock(to_gt(i915)); if (ret) pr_err("%s: Failed with %d!\n", name, ret); @@ -1481,10 +1481,10 @@ static int check_scratch(struct i915_address_space *vm, u64 offset) static int write_to_scratch(struct i915_gem_context *ctx, struct intel_engine_cs *engine, + struct drm_i915_gem_object *obj, u64 offset, u32 value) { struct drm_i915_private *i915 = ctx->i915; - struct drm_i915_gem_object *obj; struct i915_address_space *vm; struct i915_request *rq; struct i915_vma *vma; @@ -1497,15 +1497,9 @@ static int write_to_scratch(struct i915_gem_context *ctx, if (err) return err; - obj = i915_gem_object_create_internal(i915, PAGE_SIZE); - if (IS_ERR(obj)) - return PTR_ERR(obj); - cmd = i915_gem_object_pin_map_unlocked(obj, I915_MAP_WB); - if (IS_ERR(cmd)) { - err = PTR_ERR(cmd); - goto out; - } + if (IS_ERR(cmd)) + return PTR_ERR(cmd); *cmd++ = MI_STORE_DWORD_IMM_GEN4; if (GRAPHICS_VER(i915) >= 8) { @@ -1569,17 +1563,19 @@ err_unpin: i915_vma_unpin(vma); out_vm: i915_vm_put(vm); -out: - i915_gem_object_put(obj); + + if (!err) + err = i915_gem_object_wait(obj, 0, MAX_SCHEDULE_TIMEOUT); + return err; } static int read_from_scratch(struct i915_gem_context *ctx, struct intel_engine_cs *engine, + struct drm_i915_gem_object *obj, u64 offset, u32 *value) { struct drm_i915_private *i915 = ctx->i915; - struct drm_i915_gem_object *obj; struct i915_address_space *vm; const u32 result = 0x100; struct i915_request *rq; @@ -1594,10 +1590,6 @@ static int read_from_scratch(struct i915_gem_context *ctx, if (err) return err; - obj = i915_gem_object_create_internal(i915, PAGE_SIZE); - if (IS_ERR(obj)) - return PTR_ERR(obj); - if (GRAPHICS_VER(i915) >= 8) { const u32 GPR0 = engine->mmio_base + 0x600; @@ -1615,7 +1607,7 @@ static int read_from_scratch(struct i915_gem_context *ctx, cmd = i915_gem_object_pin_map_unlocked(obj, I915_MAP_WB); if (IS_ERR(cmd)) { err = PTR_ERR(cmd); - goto out; + goto err_unpin; } memset(cmd, POISON_INUSE, PAGE_SIZE); @@ -1651,7 +1643,7 @@ static int read_from_scratch(struct i915_gem_context *ctx, cmd = i915_gem_object_pin_map_unlocked(obj, I915_MAP_WB); if (IS_ERR(cmd)) { err = PTR_ERR(cmd); - goto out; + goto err_unpin; } memset(cmd, POISON_INUSE, PAGE_SIZE); @@ -1722,8 +1714,10 @@ err_unpin: i915_vma_unpin(vma); out_vm: i915_vm_put(vm); -out: - i915_gem_object_put(obj); + + if (!err) + err = i915_gem_object_wait(obj, 0, MAX_SCHEDULE_TIMEOUT); + return err; } @@ -1757,6 +1751,7 @@ static int igt_vm_isolation(void *arg) { struct drm_i915_private *i915 = arg; struct i915_gem_context *ctx_a, *ctx_b; + struct drm_i915_gem_object *obj_a, *obj_b; unsigned long num_engines, count; struct intel_engine_cs *engine; struct igt_live_test t; @@ -1810,6 +1805,18 @@ static int igt_vm_isolation(void *arg) vm_total = ctx_a->vm->total; GEM_BUG_ON(ctx_b->vm->total != vm_total); + obj_a = i915_gem_object_create_internal(i915, PAGE_SIZE); + if (IS_ERR(obj_a)) { + err = PTR_ERR(obj_a); + goto out_file; + } + + obj_b = i915_gem_object_create_internal(i915, PAGE_SIZE); + if (IS_ERR(obj_b)) { + err = PTR_ERR(obj_b); + goto put_a; + } + count = 0; num_engines = 0; for_each_uabi_engine(engine, i915) { @@ -1832,13 +1839,13 @@ static int igt_vm_isolation(void *arg) I915_GTT_PAGE_SIZE, vm_total, sizeof(u32), alignof_dword); - err = write_to_scratch(ctx_a, engine, + err = write_to_scratch(ctx_a, engine, obj_a, offset, 0xdeadbeef); if (err == 0) - err = read_from_scratch(ctx_b, engine, + err = read_from_scratch(ctx_b, engine, obj_b, offset, &value); if (err) - goto out_file; + goto put_b; if (value != expected) { pr_err("%s: Read %08x from scratch (offset 0x%08x_%08x), after %lu reads!\n", @@ -1847,7 +1854,7 @@ static int igt_vm_isolation(void *arg) lower_32_bits(offset), this); err = -EINVAL; - goto out_file; + goto put_b; } this++; @@ -1858,6 +1865,10 @@ static int igt_vm_isolation(void *arg) pr_info("Checked %lu scratch offsets across %lu engines\n", count, num_engines); +put_b: + i915_gem_object_put(obj_b); +put_a: + i915_gem_object_put(obj_a); out_file: if (igt_live_test_end(&t)) err = -EIO; @@ -1877,7 +1888,7 @@ int i915_gem_context_live_selftests(struct drm_i915_private *i915) SUBTEST(igt_vm_isolation), }; - if (intel_gt_is_wedged(&i915->gt)) + if (intel_gt_is_wedged(to_gt(i915))) return 0; return i915_live_subtests(tests, i915); diff --git a/drivers/gpu/drm/i915/gem/selftests/i915_gem_migrate.c b/drivers/gpu/drm/i915/gem/selftests/i915_gem_migrate.c index 4b8e6b098659..ecb691c81d1e 100644 --- a/drivers/gpu/drm/i915/gem/selftests/i915_gem_migrate.c +++ b/drivers/gpu/drm/i915/gem/selftests/i915_gem_migrate.c @@ -261,5 +261,5 @@ int i915_gem_migrate_live_selftests(struct drm_i915_private *i915) if (!HAS_LMEM(i915)) return 0; - return intel_gt_live_subtests(tests, &i915->gt); + return intel_gt_live_subtests(tests, to_gt(i915)); } diff --git a/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c b/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c index 6d30cdfa80f3..743e6ab2c40b 100644 --- a/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c +++ b/drivers/gpu/drm/i915/gem/selftests/i915_gem_mman.c @@ -84,6 +84,7 @@ static int check_partial_mapping(struct drm_i915_gem_object *obj, struct rnd_state *prng) { const unsigned long npages = obj->base.size / PAGE_SIZE; + struct drm_i915_private *i915 = to_i915(obj->base.dev); struct i915_ggtt_view view; struct i915_vma *vma; unsigned long page; @@ -141,7 +142,7 @@ static int check_partial_mapping(struct drm_i915_gem_object *obj, if (offset >= obj->base.size) goto out; - intel_gt_flush_ggtt_writes(&to_i915(obj->base.dev)->gt); + intel_gt_flush_ggtt_writes(to_gt(i915)); p = i915_gem_object_get_page(obj, offset >> PAGE_SHIFT); cpu = kmap(p) + offset_in_page(offset); @@ -175,6 +176,7 @@ static int check_partial_mappings(struct drm_i915_gem_object *obj, { const unsigned int nreal = obj->scratch / PAGE_SIZE; const unsigned long npages = obj->base.size / PAGE_SIZE; + struct drm_i915_private *i915 = to_i915(obj->base.dev); struct i915_vma *vma; unsigned long page; int err; @@ -234,7 +236,7 @@ static int check_partial_mappings(struct drm_i915_gem_object *obj, if (offset >= obj->base.size) continue; - intel_gt_flush_ggtt_writes(&to_i915(obj->base.dev)->gt); + intel_gt_flush_ggtt_writes(to_gt(i915)); p = i915_gem_object_get_page(obj, offset >> PAGE_SHIFT); cpu = kmap(p) + offset_in_page(offset); @@ -616,14 +618,14 @@ static bool assert_mmap_offset(struct drm_i915_private *i915, static void disable_retire_worker(struct drm_i915_private *i915) { i915_gem_driver_unregister__shrinker(i915); - intel_gt_pm_get(&i915->gt); - cancel_delayed_work_sync(&i915->gt.requests.retire_work); + intel_gt_pm_get(to_gt(i915)); + cancel_delayed_work_sync(&to_gt(i915)->requests.retire_work); } static void restore_retire_worker(struct drm_i915_private *i915) { igt_flush_test(i915); - intel_gt_pm_put(&i915->gt); + intel_gt_pm_put(to_gt(i915)); i915_gem_driver_register__shrinker(i915); } @@ -651,8 +653,8 @@ static int igt_mmap_offset_exhaustion(void *arg) /* Disable background reaper */ disable_retire_worker(i915); - GEM_BUG_ON(!i915->gt.awake); - intel_gt_retire_requests(&i915->gt); + GEM_BUG_ON(!to_gt(i915)->awake); + intel_gt_retire_requests(to_gt(i915)); i915_gem_drain_freed_objects(i915); /* Trim the device mmap space to only a page */ @@ -728,7 +730,7 @@ static int igt_mmap_offset_exhaustion(void *arg) /* Now fill with busy dead objects that we expect to reap */ for (loop = 0; loop < 3; loop++) { - if (intel_gt_is_wedged(&i915->gt)) + if (intel_gt_is_wedged(to_gt(i915))) break; obj = i915_gem_object_create_internal(i915, PAGE_SIZE); @@ -942,7 +944,7 @@ static int __igt_mmap(struct drm_i915_private *i915, } if (type == I915_MMAP_TYPE_GTT) - intel_gt_flush_ggtt_writes(&i915->gt); + intel_gt_flush_ggtt_writes(to_gt(i915)); err = wc_check(obj); if (err == -ENXIO) @@ -1049,7 +1051,7 @@ static int __igt_mmap_access(struct drm_i915_private *i915, goto out_unmap; } - intel_gt_flush_ggtt_writes(&i915->gt); + intel_gt_flush_ggtt_writes(to_gt(i915)); err = access_process_vm(current, addr, &x, sizeof(x), 0); if (err != sizeof(x)) { @@ -1065,7 +1067,7 @@ static int __igt_mmap_access(struct drm_i915_private *i915, goto out_unmap; } - intel_gt_flush_ggtt_writes(&i915->gt); + intel_gt_flush_ggtt_writes(to_gt(i915)); err = __get_user(y, ptr); if (err) { @@ -1165,7 +1167,7 @@ static int __igt_mmap_gpu(struct drm_i915_private *i915, } if (type == I915_MMAP_TYPE_GTT) - intel_gt_flush_ggtt_writes(&i915->gt); + intel_gt_flush_ggtt_writes(to_gt(i915)); for_each_uabi_engine(engine, i915) { struct i915_request *rq; diff --git a/drivers/gpu/drm/i915/gt/gen6_ppgtt.c b/drivers/gpu/drm/i915/gt/gen6_ppgtt.c index 4a166d25fe60..6e9292918bfc 100644 --- a/drivers/gpu/drm/i915/gt/gen6_ppgtt.c +++ b/drivers/gpu/drm/i915/gt/gen6_ppgtt.c @@ -269,19 +269,6 @@ static void gen6_ppgtt_cleanup(struct i915_address_space *vm) free_pd(&ppgtt->base.vm, ppgtt->base.pd); } -static int pd_vma_set_pages(struct i915_vma *vma) -{ - vma->pages = ERR_PTR(-ENODEV); - return 0; -} - -static void pd_vma_clear_pages(struct i915_vma *vma) -{ - GEM_BUG_ON(!vma->pages); - - vma->pages = NULL; -} - static void pd_vma_bind(struct i915_address_space *vm, struct i915_vm_pt_stash *stash, struct i915_vma *vma, @@ -321,8 +308,6 @@ static void pd_vma_unbind(struct i915_address_space *vm, struct i915_vma *vma) } static const struct i915_vma_ops pd_vma_ops = { - .set_pages = pd_vma_set_pages, - .clear_pages = pd_vma_clear_pages, .bind_vma = pd_vma_bind, .unbind_vma = pd_vma_unbind, }; @@ -454,6 +439,7 @@ struct i915_ppgtt *gen6_ppgtt_create(struct intel_gt *gt) ppgtt->base.vm.cleanup = gen6_ppgtt_cleanup; ppgtt->base.vm.alloc_pt_dma = alloc_pt_dma; + ppgtt->base.vm.alloc_scratch_dma = alloc_pt_dma; ppgtt->base.vm.pte_encode = ggtt->vm.pte_encode; err = gen6_ppgtt_init_scratch(ppgtt); diff --git a/drivers/gpu/drm/i915/gt/gen8_ppgtt.c b/drivers/gpu/drm/i915/gt/gen8_ppgtt.c index 95c02096a61b..b012c50f7ce7 100644 --- a/drivers/gpu/drm/i915/gt/gen8_ppgtt.c +++ b/drivers/gpu/drm/i915/gt/gen8_ppgtt.c @@ -776,10 +776,29 @@ struct i915_ppgtt *gen8_ppgtt_create(struct intel_gt *gt, */ ppgtt->vm.has_read_only = !IS_GRAPHICS_VER(gt->i915, 11, 12); - if (HAS_LMEM(gt->i915)) + if (HAS_LMEM(gt->i915)) { ppgtt->vm.alloc_pt_dma = alloc_pt_lmem; - else + + /* + * On some platforms the hw has dropped support for 4K GTT pages + * when dealing with LMEM, and due to the design of 64K GTT + * pages in the hw, we can only mark the *entire* page-table as + * operating in 64K GTT mode, since the enable bit is still on + * the pde, and not the pte. And since we still need to allow + * 4K GTT pages for SMEM objects, we can't have a "normal" 4K + * page-table with scratch pointing to LMEM, since that's + * undefined from the hw pov. The simplest solution is to just + * move the 64K scratch page to SMEM on such platforms and call + * it a day, since that should work for all configurations. + */ + if (HAS_64K_PAGES(gt->i915)) + ppgtt->vm.alloc_scratch_dma = alloc_pt_dma; + else + ppgtt->vm.alloc_scratch_dma = alloc_pt_lmem; + } else { ppgtt->vm.alloc_pt_dma = alloc_pt_dma; + ppgtt->vm.alloc_scratch_dma = alloc_pt_dma; + } err = gen8_init_scratch(&ppgtt->vm); if (err) diff --git a/drivers/gpu/drm/i915/gt/intel_context.h b/drivers/gpu/drm/i915/gt/intel_context.h index 246c37d72cd7..d8c74bbf9aae 100644 --- a/drivers/gpu/drm/i915/gt/intel_context.h +++ b/drivers/gpu/drm/i915/gt/intel_context.h @@ -211,7 +211,8 @@ static inline void intel_context_enter(struct intel_context *ce) static inline void intel_context_mark_active(struct intel_context *ce) { - lockdep_assert_held(&ce->timeline->mutex); + lockdep_assert(lockdep_is_held(&ce->timeline->mutex) || + test_bit(CONTEXT_IS_PARKING, &ce->flags)); ++ce->active_count; } diff --git a/drivers/gpu/drm/i915/gt/intel_context_types.h b/drivers/gpu/drm/i915/gt/intel_context_types.h index 9e0177dc5484..30cd81ad8911 100644 --- a/drivers/gpu/drm/i915/gt/intel_context_types.h +++ b/drivers/gpu/drm/i915/gt/intel_context_types.h @@ -118,6 +118,7 @@ struct intel_context { #define CONTEXT_LRCA_DIRTY 9 #define CONTEXT_GUC_INIT 10 #define CONTEXT_PERMA_PIN 11 +#define CONTEXT_IS_PARKING 12 struct { u64 timeout_us; diff --git a/drivers/gpu/drm/i915/gt/intel_engine_pm.c b/drivers/gpu/drm/i915/gt/intel_engine_pm.c index a1334b48dde7..b0a4a2dbe3ee 100644 --- a/drivers/gpu/drm/i915/gt/intel_engine_pm.c +++ b/drivers/gpu/drm/i915/gt/intel_engine_pm.c @@ -26,7 +26,7 @@ static void dbg_poison_ce(struct intel_context *ce) int type = i915_coherent_map_type(ce->engine->i915, obj, true); void *map; - if (!i915_gem_object_trylock(obj)) + if (!i915_gem_object_trylock(obj, NULL)) return; map = i915_gem_object_pin_map(obj, type); @@ -80,39 +80,6 @@ static int __engine_unpark(struct intel_wakeref *wf) return 0; } -#if IS_ENABLED(CONFIG_LOCKDEP) - -static unsigned long __timeline_mark_lock(struct intel_context *ce) -{ - unsigned long flags; - - local_irq_save(flags); - mutex_acquire(&ce->timeline->mutex.dep_map, 2, 0, _THIS_IP_); - - return flags; -} - -static void __timeline_mark_unlock(struct intel_context *ce, - unsigned long flags) -{ - mutex_release(&ce->timeline->mutex.dep_map, _THIS_IP_); - local_irq_restore(flags); -} - -#else - -static unsigned long __timeline_mark_lock(struct intel_context *ce) -{ - return 0; -} - -static void __timeline_mark_unlock(struct intel_context *ce, - unsigned long flags) -{ -} - -#endif /* !IS_ENABLED(CONFIG_LOCKDEP) */ - static void duration(struct dma_fence *fence, struct dma_fence_cb *cb) { struct i915_request *rq = to_request(fence); @@ -159,7 +126,6 @@ static bool switch_to_kernel_context(struct intel_engine_cs *engine) { struct intel_context *ce = engine->kernel_context; struct i915_request *rq; - unsigned long flags; bool result = true; /* @@ -214,7 +180,7 @@ static bool switch_to_kernel_context(struct intel_engine_cs *engine) * engine->wakeref.count, we may see the request completion and retire * it causing an underflow of the engine->wakeref. */ - flags = __timeline_mark_lock(ce); + set_bit(CONTEXT_IS_PARKING, &ce->flags); GEM_BUG_ON(atomic_read(&ce->timeline->active_count) < 0); rq = __i915_request_create(ce, GFP_NOWAIT); @@ -246,7 +212,7 @@ static bool switch_to_kernel_context(struct intel_engine_cs *engine) result = false; out_unlock: - __timeline_mark_unlock(ce, flags); + clear_bit(CONTEXT_IS_PARKING, &ce->flags); return result; } diff --git a/drivers/gpu/drm/i915/gt/intel_engine_user.c b/drivers/gpu/drm/i915/gt/intel_engine_user.c index 8f8bea08e734..9ce85a845105 100644 --- a/drivers/gpu/drm/i915/gt/intel_engine_user.c +++ b/drivers/gpu/drm/i915/gt/intel_engine_user.c @@ -116,7 +116,7 @@ static void set_scheduler_caps(struct drm_i915_private *i915) disabled |= (I915_SCHEDULER_CAP_ENABLED | I915_SCHEDULER_CAP_PRIORITY); - if (intel_uc_uses_guc_submission(&i915->gt.uc)) + if (intel_uc_uses_guc_submission(&to_gt(i915)->uc)) enabled |= I915_SCHEDULER_CAP_STATIC_PRIORITY_MAP; for (i = 0; i < ARRAY_SIZE(map); i++) { diff --git a/drivers/gpu/drm/i915/gt/intel_ggtt.c b/drivers/gpu/drm/i915/gt/intel_ggtt.c index cbc6d2b1fd9e..5263dda7f8d5 100644 --- a/drivers/gpu/drm/i915/gt/intel_ggtt.c +++ b/drivers/gpu/drm/i915/gt/intel_ggtt.c @@ -22,9 +22,6 @@ #include "intel_gtt.h" #include "gen8_ppgtt.h" -static int -i915_get_ggtt_vma_pages(struct i915_vma *vma); - static void i915_ggtt_color_adjust(const struct drm_mm_node *node, unsigned long color, u64 *start, @@ -892,21 +889,6 @@ static int ggtt_probe_common(struct i915_ggtt *ggtt, u64 size) return 0; } -int ggtt_set_pages(struct i915_vma *vma) -{ - int ret; - - GEM_BUG_ON(vma->pages); - - ret = i915_get_ggtt_vma_pages(vma); - if (ret) - return ret; - - vma->page_sizes = vma->obj->mm.page_sizes; - - return 0; -} - static void gen6_gmch_remove(struct i915_address_space *vm) { struct i915_ggtt *ggtt = i915_vm_to_ggtt(vm); @@ -941,6 +923,7 @@ static int gen8_gmch_probe(struct i915_ggtt *ggtt) size = gen8_get_total_gtt_size(snb_gmch_ctl); ggtt->vm.alloc_pt_dma = alloc_pt_dma; + ggtt->vm.alloc_scratch_dma = alloc_pt_dma; ggtt->vm.lmem_pt_obj_flags = I915_BO_ALLOC_PM_EARLY; ggtt->vm.total = (size / sizeof(gen8_pte_t)) * I915_GTT_PAGE_SIZE; @@ -967,8 +950,6 @@ static int gen8_gmch_probe(struct i915_ggtt *ggtt) ggtt->vm.vma_ops.bind_vma = ggtt_bind_vma; ggtt->vm.vma_ops.unbind_vma = ggtt_unbind_vma; - ggtt->vm.vma_ops.set_pages = ggtt_set_pages; - ggtt->vm.vma_ops.clear_pages = clear_pages; ggtt->vm.pte_encode = gen8_ggtt_pte_encode; @@ -1094,6 +1075,7 @@ static int gen6_gmch_probe(struct i915_ggtt *ggtt) ggtt->vm.total = (size / sizeof(gen6_pte_t)) * I915_GTT_PAGE_SIZE; ggtt->vm.alloc_pt_dma = alloc_pt_dma; + ggtt->vm.alloc_scratch_dma = alloc_pt_dma; ggtt->vm.clear_range = nop_clear_range; if (!HAS_FULL_PPGTT(i915) || intel_scanout_needs_vtd_wa(i915)) @@ -1117,8 +1099,6 @@ static int gen6_gmch_probe(struct i915_ggtt *ggtt) ggtt->vm.vma_ops.bind_vma = ggtt_bind_vma; ggtt->vm.vma_ops.unbind_vma = ggtt_unbind_vma; - ggtt->vm.vma_ops.set_pages = ggtt_set_pages; - ggtt->vm.vma_ops.clear_pages = clear_pages; return ggtt_probe_common(ggtt, size); } @@ -1146,6 +1126,7 @@ static int i915_gmch_probe(struct i915_ggtt *ggtt) (struct resource)DEFINE_RES_MEM(gmadr_base, ggtt->mappable_end); ggtt->vm.alloc_pt_dma = alloc_pt_dma; + ggtt->vm.alloc_scratch_dma = alloc_pt_dma; if (needs_idle_maps(i915)) { drm_notice(&i915->drm, @@ -1162,8 +1143,6 @@ static int i915_gmch_probe(struct i915_ggtt *ggtt) ggtt->vm.vma_ops.bind_vma = ggtt_bind_vma; ggtt->vm.vma_ops.unbind_vma = ggtt_unbind_vma; - ggtt->vm.vma_ops.set_pages = ggtt_set_pages; - ggtt->vm.vma_ops.clear_pages = clear_pages; if (unlikely(ggtt->do_idle_maps)) drm_notice(&i915->drm, @@ -1229,7 +1208,7 @@ int i915_ggtt_probe_hw(struct drm_i915_private *i915) { int ret; - ret = ggtt_probe_hw(&i915->ggtt, &i915->gt); + ret = ggtt_probe_hw(&i915->ggtt, to_gt(i915)); if (ret) return ret; @@ -1333,382 +1312,3 @@ void i915_ggtt_resume(struct i915_ggtt *ggtt) intel_ggtt_restore_fences(ggtt); } - -static struct scatterlist * -rotate_pages(struct drm_i915_gem_object *obj, unsigned int offset, - unsigned int width, unsigned int height, - unsigned int src_stride, unsigned int dst_stride, - struct sg_table *st, struct scatterlist *sg) -{ - unsigned int column, row; - unsigned int src_idx; - - for (column = 0; column < width; column++) { - unsigned int left; - - src_idx = src_stride * (height - 1) + column + offset; - for (row = 0; row < height; row++) { - st->nents++; - /* - * We don't need the pages, but need to initialize - * the entries so the sg list can be happily traversed. - * The only thing we need are DMA addresses. - */ - sg_set_page(sg, NULL, I915_GTT_PAGE_SIZE, 0); - sg_dma_address(sg) = - i915_gem_object_get_dma_address(obj, src_idx); - sg_dma_len(sg) = I915_GTT_PAGE_SIZE; - sg = sg_next(sg); - src_idx -= src_stride; - } - - left = (dst_stride - height) * I915_GTT_PAGE_SIZE; - - if (!left) - continue; - - st->nents++; - - /* - * The DE ignores the PTEs for the padding tiles, the sg entry - * here is just a conenience to indicate how many padding PTEs - * to insert at this spot. - */ - sg_set_page(sg, NULL, left, 0); - sg_dma_address(sg) = 0; - sg_dma_len(sg) = left; - sg = sg_next(sg); - } - - return sg; -} - -static noinline struct sg_table * -intel_rotate_pages(struct intel_rotation_info *rot_info, - struct drm_i915_gem_object *obj) -{ - unsigned int size = intel_rotation_info_size(rot_info); - struct drm_i915_private *i915 = to_i915(obj->base.dev); - struct sg_table *st; - struct scatterlist *sg; - int ret = -ENOMEM; - int i; - - /* Allocate target SG list. */ - st = kmalloc(sizeof(*st), GFP_KERNEL); - if (!st) - goto err_st_alloc; - - ret = sg_alloc_table(st, size, GFP_KERNEL); - if (ret) - goto err_sg_alloc; - - st->nents = 0; - sg = st->sgl; - - for (i = 0 ; i < ARRAY_SIZE(rot_info->plane); i++) - sg = rotate_pages(obj, rot_info->plane[i].offset, - rot_info->plane[i].width, rot_info->plane[i].height, - rot_info->plane[i].src_stride, - rot_info->plane[i].dst_stride, - st, sg); - - return st; - -err_sg_alloc: - kfree(st); -err_st_alloc: - - drm_dbg(&i915->drm, "Failed to create rotated mapping for object size %zu! (%ux%u tiles, %u pages)\n", - obj->base.size, rot_info->plane[0].width, - rot_info->plane[0].height, size); - - return ERR_PTR(ret); -} - -static struct scatterlist * -add_padding_pages(unsigned int count, - struct sg_table *st, struct scatterlist *sg) -{ - st->nents++; - - /* - * The DE ignores the PTEs for the padding tiles, the sg entry - * here is just a convenience to indicate how many padding PTEs - * to insert at this spot. - */ - sg_set_page(sg, NULL, count * I915_GTT_PAGE_SIZE, 0); - sg_dma_address(sg) = 0; - sg_dma_len(sg) = count * I915_GTT_PAGE_SIZE; - sg = sg_next(sg); - - return sg; -} - -static struct scatterlist * -remap_tiled_color_plane_pages(struct drm_i915_gem_object *obj, - unsigned int offset, unsigned int alignment_pad, - unsigned int width, unsigned int height, - unsigned int src_stride, unsigned int dst_stride, - struct sg_table *st, struct scatterlist *sg, - unsigned int *gtt_offset) -{ - unsigned int row; - - if (!width || !height) - return sg; - - if (alignment_pad) - sg = add_padding_pages(alignment_pad, st, sg); - - for (row = 0; row < height; row++) { - unsigned int left = width * I915_GTT_PAGE_SIZE; - - while (left) { - dma_addr_t addr; - unsigned int length; - - /* - * We don't need the pages, but need to initialize - * the entries so the sg list can be happily traversed. - * The only thing we need are DMA addresses. - */ - - addr = i915_gem_object_get_dma_address_len(obj, offset, &length); - - length = min(left, length); - - st->nents++; - - sg_set_page(sg, NULL, length, 0); - sg_dma_address(sg) = addr; - sg_dma_len(sg) = length; - sg = sg_next(sg); - - offset += length / I915_GTT_PAGE_SIZE; - left -= length; - } - - offset += src_stride - width; - - left = (dst_stride - width) * I915_GTT_PAGE_SIZE; - - if (!left) - continue; - - sg = add_padding_pages(left >> PAGE_SHIFT, st, sg); - } - - *gtt_offset += alignment_pad + dst_stride * height; - - return sg; -} - -static struct scatterlist * -remap_contiguous_pages(struct drm_i915_gem_object *obj, - unsigned int obj_offset, - unsigned int count, - struct sg_table *st, struct scatterlist *sg) -{ - struct scatterlist *iter; - unsigned int offset; - - iter = i915_gem_object_get_sg_dma(obj, obj_offset, &offset); - GEM_BUG_ON(!iter); - - do { - unsigned int len; - - len = min(sg_dma_len(iter) - (offset << PAGE_SHIFT), - count << PAGE_SHIFT); - sg_set_page(sg, NULL, len, 0); - sg_dma_address(sg) = - sg_dma_address(iter) + (offset << PAGE_SHIFT); - sg_dma_len(sg) = len; - - st->nents++; - count -= len >> PAGE_SHIFT; - if (count == 0) - return sg; - - sg = __sg_next(sg); - iter = __sg_next(iter); - offset = 0; - } while (1); -} - -static struct scatterlist * -remap_linear_color_plane_pages(struct drm_i915_gem_object *obj, - unsigned int obj_offset, unsigned int alignment_pad, - unsigned int size, - struct sg_table *st, struct scatterlist *sg, - unsigned int *gtt_offset) -{ - if (!size) - return sg; - - if (alignment_pad) - sg = add_padding_pages(alignment_pad, st, sg); - - sg = remap_contiguous_pages(obj, obj_offset, size, st, sg); - sg = sg_next(sg); - - *gtt_offset += alignment_pad + size; - - return sg; -} - -static struct scatterlist * -remap_color_plane_pages(const struct intel_remapped_info *rem_info, - struct drm_i915_gem_object *obj, - int color_plane, - struct sg_table *st, struct scatterlist *sg, - unsigned int *gtt_offset) -{ - unsigned int alignment_pad = 0; - - if (rem_info->plane_alignment) - alignment_pad = ALIGN(*gtt_offset, rem_info->plane_alignment) - *gtt_offset; - - if (rem_info->plane[color_plane].linear) - sg = remap_linear_color_plane_pages(obj, - rem_info->plane[color_plane].offset, - alignment_pad, - rem_info->plane[color_plane].size, - st, sg, - gtt_offset); - - else - sg = remap_tiled_color_plane_pages(obj, - rem_info->plane[color_plane].offset, - alignment_pad, - rem_info->plane[color_plane].width, - rem_info->plane[color_plane].height, - rem_info->plane[color_plane].src_stride, - rem_info->plane[color_plane].dst_stride, - st, sg, - gtt_offset); - - return sg; -} - -static noinline struct sg_table * -intel_remap_pages(struct intel_remapped_info *rem_info, - struct drm_i915_gem_object *obj) -{ - unsigned int size = intel_remapped_info_size(rem_info); - struct drm_i915_private *i915 = to_i915(obj->base.dev); - struct sg_table *st; - struct scatterlist *sg; - unsigned int gtt_offset = 0; - int ret = -ENOMEM; - int i; - - /* Allocate target SG list. */ - st = kmalloc(sizeof(*st), GFP_KERNEL); - if (!st) - goto err_st_alloc; - - ret = sg_alloc_table(st, size, GFP_KERNEL); - if (ret) - goto err_sg_alloc; - - st->nents = 0; - sg = st->sgl; - - for (i = 0 ; i < ARRAY_SIZE(rem_info->plane); i++) - sg = remap_color_plane_pages(rem_info, obj, i, st, sg, >t_offset); - - i915_sg_trim(st); - - return st; - -err_sg_alloc: - kfree(st); -err_st_alloc: - - drm_dbg(&i915->drm, "Failed to create remapped mapping for object size %zu! (%ux%u tiles, %u pages)\n", - obj->base.size, rem_info->plane[0].width, - rem_info->plane[0].height, size); - - return ERR_PTR(ret); -} - -static noinline struct sg_table * -intel_partial_pages(const struct i915_ggtt_view *view, - struct drm_i915_gem_object *obj) -{ - struct sg_table *st; - struct scatterlist *sg; - unsigned int count = view->partial.size; - int ret = -ENOMEM; - - st = kmalloc(sizeof(*st), GFP_KERNEL); - if (!st) - goto err_st_alloc; - - ret = sg_alloc_table(st, count, GFP_KERNEL); - if (ret) - goto err_sg_alloc; - - st->nents = 0; - - sg = remap_contiguous_pages(obj, view->partial.offset, count, st, st->sgl); - - sg_mark_end(sg); - i915_sg_trim(st); /* Drop any unused tail entries. */ - - return st; - -err_sg_alloc: - kfree(st); -err_st_alloc: - return ERR_PTR(ret); -} - -static int -i915_get_ggtt_vma_pages(struct i915_vma *vma) -{ - int ret; - - /* - * The vma->pages are only valid within the lifespan of the borrowed - * obj->mm.pages. When the obj->mm.pages sg_table is regenerated, so - * must be the vma->pages. A simple rule is that vma->pages must only - * be accessed when the obj->mm.pages are pinned. - */ - GEM_BUG_ON(!i915_gem_object_has_pinned_pages(vma->obj)); - - switch (vma->ggtt_view.type) { - default: - GEM_BUG_ON(vma->ggtt_view.type); - fallthrough; - case I915_GGTT_VIEW_NORMAL: - vma->pages = vma->obj->mm.pages; - return 0; - - case I915_GGTT_VIEW_ROTATED: - vma->pages = - intel_rotate_pages(&vma->ggtt_view.rotated, vma->obj); - break; - - case I915_GGTT_VIEW_REMAPPED: - vma->pages = - intel_remap_pages(&vma->ggtt_view.remapped, vma->obj); - break; - - case I915_GGTT_VIEW_PARTIAL: - vma->pages = intel_partial_pages(&vma->ggtt_view, vma->obj); - break; - } - - ret = 0; - if (IS_ERR(vma->pages)) { - ret = PTR_ERR(vma->pages); - vma->pages = NULL; - drm_err(&vma->vm->i915->drm, - "Failed to get pages for VMA view type %u (%d)!\n", - vma->ggtt_view.type, ret); - } - return ret; -} diff --git a/drivers/gpu/drm/i915/gt/intel_gt.c b/drivers/gpu/drm/i915/gt/intel_gt.c index f2422d48be32..f98f0fb21efb 100644 --- a/drivers/gpu/drm/i915/gt/intel_gt.c +++ b/drivers/gpu/drm/i915/gt/intel_gt.c @@ -25,11 +25,8 @@ #include "shmem_utils.h" #include "pxp/intel_pxp.h" -void intel_gt_init_early(struct intel_gt *gt, struct drm_i915_private *i915) +void __intel_gt_init_early(struct intel_gt *gt, struct drm_i915_private *i915) { - gt->i915 = i915; - gt->uncore = &i915->uncore; - spin_lock_init(>->irq_lock); INIT_LIST_HEAD(>->closed_vma); @@ -48,6 +45,12 @@ void intel_gt_init_early(struct intel_gt *gt, struct drm_i915_private *i915) intel_rps_init_early(>->rps); } +void intel_gt_init_early(struct intel_gt *gt, struct drm_i915_private *i915) +{ + gt->i915 = i915; + gt->uncore = &i915->uncore; +} + int intel_gt_probe_lmem(struct intel_gt *gt) { struct drm_i915_private *i915 = gt->i915; diff --git a/drivers/gpu/drm/i915/gt/intel_gt.h b/drivers/gpu/drm/i915/gt/intel_gt.h index 74e771871a9b..3ace129eb2af 100644 --- a/drivers/gpu/drm/i915/gt/intel_gt.h +++ b/drivers/gpu/drm/i915/gt/intel_gt.h @@ -35,6 +35,7 @@ static inline struct intel_gt *huc_to_gt(struct intel_huc *huc) } void intel_gt_init_early(struct intel_gt *gt, struct drm_i915_private *i915); +void __intel_gt_init_early(struct intel_gt *gt, struct drm_i915_private *i915); void intel_gt_init_hw_early(struct intel_gt *gt, struct i915_ggtt *ggtt); int intel_gt_probe_lmem(struct intel_gt *gt); int intel_gt_init_mmio(struct intel_gt *gt); diff --git a/drivers/gpu/drm/i915/gt/intel_gt_buffer_pool.c b/drivers/gpu/drm/i915/gt/intel_gt_buffer_pool.c index acc49c56a9f3..9db3dcbd917f 100644 --- a/drivers/gpu/drm/i915/gt/intel_gt_buffer_pool.c +++ b/drivers/gpu/drm/i915/gt/intel_gt_buffer_pool.c @@ -9,11 +9,6 @@ #include "intel_engine_pm.h" #include "intel_gt_buffer_pool.h" -static struct intel_gt *to_gt(struct intel_gt_buffer_pool *pool) -{ - return container_of(pool, struct intel_gt, buffer_pool); -} - static struct list_head * bucket_for_size(struct intel_gt_buffer_pool *pool, size_t sz) { @@ -141,7 +136,7 @@ static struct intel_gt_buffer_pool_node * node_create(struct intel_gt_buffer_pool *pool, size_t sz, enum i915_map_type type) { - struct intel_gt *gt = to_gt(pool); + struct intel_gt *gt = container_of(pool, struct intel_gt, buffer_pool); struct intel_gt_buffer_pool_node *node; struct drm_i915_gem_object *obj; diff --git a/drivers/gpu/drm/i915/gt/intel_gt_debugfs.h b/drivers/gpu/drm/i915/gt/intel_gt_debugfs.h index e307ceb99031..17e79b735cfe 100644 --- a/drivers/gpu/drm/i915/gt/intel_gt_debugfs.h +++ b/drivers/gpu/drm/i915/gt/intel_gt_debugfs.h @@ -10,11 +10,7 @@ struct intel_gt; -#define DEFINE_INTEL_GT_DEBUGFS_ATTRIBUTE(__name) \ - static int __name ## _open(struct inode *inode, struct file *file) \ -{ \ - return single_open(file, __name ## _show, inode->i_private); \ -} \ +#define __GT_DEBUGFS_ATTRIBUTE_FOPS(__name) \ static const struct file_operations __name ## _fops = { \ .owner = THIS_MODULE, \ .open = __name ## _open, \ @@ -23,6 +19,21 @@ static const struct file_operations __name ## _fops = { \ .release = single_release, \ } +#define DEFINE_INTEL_GT_DEBUGFS_ATTRIBUTE(__name) \ +static int __name ## _open(struct inode *inode, struct file *file) \ +{ \ + return single_open(file, __name ## _show, inode->i_private); \ +} \ +__GT_DEBUGFS_ATTRIBUTE_FOPS(__name) + +#define DEFINE_INTEL_GT_DEBUGFS_ATTRIBUTE_WITH_SIZE(__name, __size_vf) \ +static int __name ## _open(struct inode *inode, struct file *file) \ +{ \ + return single_open_size(file, __name ## _show, inode->i_private, \ + __size_vf(inode->i_private)); \ +} \ +__GT_DEBUGFS_ATTRIBUTE_FOPS(__name) + void intel_gt_debugfs_register(struct intel_gt *gt); struct intel_gt_debugfs_file { diff --git a/drivers/gpu/drm/i915/gt/intel_gtt.c b/drivers/gpu/drm/i915/gt/intel_gtt.c index 9fee968d57db..a94be0306464 100644 --- a/drivers/gpu/drm/i915/gt/intel_gtt.c +++ b/drivers/gpu/drm/i915/gt/intel_gtt.c @@ -224,19 +224,6 @@ void i915_address_space_init(struct i915_address_space *vm, int subclass) INIT_LIST_HEAD(&vm->bound_list); } -void clear_pages(struct i915_vma *vma) -{ - GEM_BUG_ON(!vma->pages); - - if (vma->pages != vma->obj->mm.pages) { - sg_free_table(vma->pages); - kfree(vma->pages); - } - vma->pages = NULL; - - memset(&vma->page_sizes, 0, sizeof(vma->page_sizes)); -} - void *__px_vaddr(struct drm_i915_gem_object *p) { enum i915_map_type type; @@ -302,7 +289,7 @@ int setup_scratch_page(struct i915_address_space *vm) do { struct drm_i915_gem_object *obj; - obj = vm->alloc_pt_dma(vm, size); + obj = vm->alloc_scratch_dma(vm, size); if (IS_ERR(obj)) goto skip; @@ -338,6 +325,18 @@ skip: if (size == I915_GTT_PAGE_SIZE_4K) return -ENOMEM; + /* + * If we need 64K minimum GTT pages for device local-memory, + * like on XEHPSDV, then we need to fail the allocation here, + * otherwise we can't safely support the insertion of + * local-memory pages for this vm, since the HW expects the + * correct physical alignment and size when the page-table is + * operating in 64K GTT mode, which includes any scratch PTEs, + * since userspace can still touch them. + */ + if (HAS_64K_PAGES(vm->i915)) + return -ENOMEM; + size = I915_GTT_PAGE_SIZE_4K; } while (1); } diff --git a/drivers/gpu/drm/i915/gt/intel_gtt.h b/drivers/gpu/drm/i915/gt/intel_gtt.h index 51afe66d00f2..177b42b935a1 100644 --- a/drivers/gpu/drm/i915/gt/intel_gtt.h +++ b/drivers/gpu/drm/i915/gt/intel_gtt.h @@ -209,9 +209,6 @@ struct i915_vma_ops { */ void (*unbind_vma)(struct i915_address_space *vm, struct i915_vma *vma); - - int (*set_pages)(struct i915_vma *vma); - void (*clear_pages)(struct i915_vma *vma); }; struct i915_address_space { @@ -268,6 +265,8 @@ struct i915_address_space { struct drm_i915_gem_object * (*alloc_pt_dma)(struct i915_address_space *vm, int sz); + struct drm_i915_gem_object * + (*alloc_scratch_dma)(struct i915_address_space *vm, int sz); u64 (*pte_encode)(dma_addr_t addr, enum i915_cache_level level, @@ -599,10 +598,6 @@ release_pd_entry(struct i915_page_directory * const pd, const struct drm_i915_gem_object * const scratch); void gen6_ggtt_invalidate(struct i915_ggtt *ggtt); -int ggtt_set_pages(struct i915_vma *vma); -int ppgtt_set_pages(struct i915_vma *vma); -void clear_pages(struct i915_vma *vma); - void ppgtt_bind_vma(struct i915_address_space *vm, struct i915_vm_pt_stash *stash, struct i915_vma *vma, diff --git a/drivers/gpu/drm/i915/gt/intel_migrate.c b/drivers/gpu/drm/i915/gt/intel_migrate.c index 19a01878fee3..18b44af56969 100644 --- a/drivers/gpu/drm/i915/gt/intel_migrate.c +++ b/drivers/gpu/drm/i915/gt/intel_migrate.c @@ -404,7 +404,7 @@ static int emit_copy(struct i915_request *rq, int size) int intel_context_migrate_copy(struct intel_context *ce, - struct dma_fence *await, + const struct i915_deps *deps, struct scatterlist *src, enum i915_cache_level src_cache_level, bool src_is_lmem, @@ -431,8 +431,8 @@ intel_context_migrate_copy(struct intel_context *ce, goto out_ce; } - if (await) { - err = i915_request_await_dma_fence(rq, await); + if (deps) { + err = i915_request_await_deps(rq, deps); if (err) goto out_rq; @@ -442,7 +442,7 @@ intel_context_migrate_copy(struct intel_context *ce, goto out_rq; } - await = NULL; + deps = NULL; } /* The PTE updates + copy must not be interrupted. */ @@ -525,7 +525,7 @@ static int emit_clear(struct i915_request *rq, int size, u32 value) int intel_context_migrate_clear(struct intel_context *ce, - struct dma_fence *await, + const struct i915_deps *deps, struct scatterlist *sg, enum i915_cache_level cache_level, bool is_lmem, @@ -550,8 +550,8 @@ intel_context_migrate_clear(struct intel_context *ce, goto out_ce; } - if (await) { - err = i915_request_await_dma_fence(rq, await); + if (deps) { + err = i915_request_await_deps(rq, deps); if (err) goto out_rq; @@ -561,7 +561,7 @@ intel_context_migrate_clear(struct intel_context *ce, goto out_rq; } - await = NULL; + deps = NULL; } /* The PTE updates + clear must not be interrupted. */ @@ -599,7 +599,7 @@ out_ce: int intel_migrate_copy(struct intel_migrate *m, struct i915_gem_ww_ctx *ww, - struct dma_fence *await, + const struct i915_deps *deps, struct scatterlist *src, enum i915_cache_level src_cache_level, bool src_is_lmem, @@ -624,7 +624,7 @@ int intel_migrate_copy(struct intel_migrate *m, if (err) goto out; - err = intel_context_migrate_copy(ce, await, + err = intel_context_migrate_copy(ce, deps, src, src_cache_level, src_is_lmem, dst, dst_cache_level, dst_is_lmem, out); @@ -638,7 +638,7 @@ out: int intel_migrate_clear(struct intel_migrate *m, struct i915_gem_ww_ctx *ww, - struct dma_fence *await, + const struct i915_deps *deps, struct scatterlist *sg, enum i915_cache_level cache_level, bool is_lmem, @@ -661,7 +661,7 @@ intel_migrate_clear(struct intel_migrate *m, if (err) goto out; - err = intel_context_migrate_clear(ce, await, sg, cache_level, + err = intel_context_migrate_clear(ce, deps, sg, cache_level, is_lmem, value, out); intel_context_unpin(ce); diff --git a/drivers/gpu/drm/i915/gt/intel_migrate.h b/drivers/gpu/drm/i915/gt/intel_migrate.h index 4e18e755a00b..ccc677ec4aa3 100644 --- a/drivers/gpu/drm/i915/gt/intel_migrate.h +++ b/drivers/gpu/drm/i915/gt/intel_migrate.h @@ -11,6 +11,7 @@ #include "intel_migrate_types.h" struct dma_fence; +struct i915_deps; struct i915_request; struct i915_gem_ww_ctx; struct intel_gt; @@ -23,7 +24,7 @@ struct intel_context *intel_migrate_create_context(struct intel_migrate *m); int intel_migrate_copy(struct intel_migrate *m, struct i915_gem_ww_ctx *ww, - struct dma_fence *await, + const struct i915_deps *deps, struct scatterlist *src, enum i915_cache_level src_cache_level, bool src_is_lmem, @@ -33,7 +34,7 @@ int intel_migrate_copy(struct intel_migrate *m, struct i915_request **out); int intel_context_migrate_copy(struct intel_context *ce, - struct dma_fence *await, + const struct i915_deps *deps, struct scatterlist *src, enum i915_cache_level src_cache_level, bool src_is_lmem, @@ -45,7 +46,7 @@ int intel_context_migrate_copy(struct intel_context *ce, int intel_migrate_clear(struct intel_migrate *m, struct i915_gem_ww_ctx *ww, - struct dma_fence *await, + const struct i915_deps *deps, struct scatterlist *sg, enum i915_cache_level cache_level, bool is_lmem, @@ -53,7 +54,7 @@ intel_migrate_clear(struct intel_migrate *m, struct i915_request **out); int intel_context_migrate_clear(struct intel_context *ce, - struct dma_fence *await, + const struct i915_deps *deps, struct scatterlist *sg, enum i915_cache_level cache_level, bool is_lmem, diff --git a/drivers/gpu/drm/i915/gt/intel_ppgtt.c b/drivers/gpu/drm/i915/gt/intel_ppgtt.c index 4396bfd630d8..083b3090c69c 100644 --- a/drivers/gpu/drm/i915/gt/intel_ppgtt.c +++ b/drivers/gpu/drm/i915/gt/intel_ppgtt.c @@ -289,16 +289,6 @@ void i915_vm_free_pt_stash(struct i915_address_space *vm, } } -int ppgtt_set_pages(struct i915_vma *vma) -{ - GEM_BUG_ON(vma->pages); - - vma->pages = vma->obj->mm.pages; - vma->page_sizes = vma->obj->mm.page_sizes; - - return 0; -} - void ppgtt_init(struct i915_ppgtt *ppgtt, struct intel_gt *gt, unsigned long lmem_pt_obj_flags) { @@ -315,6 +305,4 @@ void ppgtt_init(struct i915_ppgtt *ppgtt, struct intel_gt *gt, ppgtt->vm.vma_ops.bind_vma = ppgtt_bind_vma; ppgtt->vm.vma_ops.unbind_vma = ppgtt_unbind_vma; - ppgtt->vm.vma_ops.set_pages = ppgtt_set_pages; - ppgtt->vm.vma_ops.clear_pages = clear_pages; } diff --git a/drivers/gpu/drm/i915/gt/intel_region_lmem.c b/drivers/gpu/drm/i915/gt/intel_region_lmem.c index 9ea49e0a27c0..fde2dcb59809 100644 --- a/drivers/gpu/drm/i915/gt/intel_region_lmem.c +++ b/drivers/gpu/drm/i915/gt/intel_region_lmem.c @@ -197,6 +197,7 @@ static struct intel_memory_region *setup_lmem(struct intel_gt *gt) struct intel_uncore *uncore = gt->uncore; struct pci_dev *pdev = to_pci_dev(i915->drm.dev); struct intel_memory_region *mem; + resource_size_t min_page_size; resource_size_t io_start; resource_size_t lmem_size; int err; @@ -211,10 +212,12 @@ static struct intel_memory_region *setup_lmem(struct intel_gt *gt) if (GEM_WARN_ON(lmem_size > pci_resource_len(pdev, 2))) return ERR_PTR(-ENODEV); + min_page_size = HAS_64K_PAGES(i915) ? I915_GTT_PAGE_SIZE_64K : + I915_GTT_PAGE_SIZE_4K; mem = intel_memory_region_create(i915, 0, lmem_size, - I915_GTT_PAGE_SIZE_4K, + min_page_size, io_start, INTEL_MEMORY_LOCAL, 0, diff --git a/drivers/gpu/drm/i915/gt/intel_rps.c b/drivers/gpu/drm/i915/gt/intel_rps.c index 07ff7ba7b2b7..54e7df788dbf 100644 --- a/drivers/gpu/drm/i915/gt/intel_rps.c +++ b/drivers/gpu/drm/i915/gt/intel_rps.c @@ -2226,6 +2226,65 @@ u32 intel_rps_read_state_cap(struct intel_rps *rps) return intel_uncore_read(uncore, GEN6_RP_STATE_CAP); } +static void intel_rps_set_manual(struct intel_rps *rps, bool enable) +{ + struct intel_uncore *uncore = rps_to_uncore(rps); + u32 state = enable ? GEN9_RPSWCTL_ENABLE : GEN9_RPSWCTL_DISABLE; + + /* Allow punit to process software requests */ + intel_uncore_write(uncore, GEN6_RP_CONTROL, state); +} + +void intel_rps_raise_unslice(struct intel_rps *rps) +{ + struct intel_uncore *uncore = rps_to_uncore(rps); + u32 rp0_unslice_req; + + mutex_lock(&rps->lock); + + if (rps_uses_slpc(rps)) { + /* RP limits have not been initialized yet for SLPC path */ + rp0_unslice_req = ((intel_rps_read_state_cap(rps) >> 0) + & 0xff) * GEN9_FREQ_SCALER; + + intel_rps_set_manual(rps, true); + intel_uncore_write(uncore, GEN6_RPNSWREQ, + ((rp0_unslice_req << + GEN9_SW_REQ_UNSLICE_RATIO_SHIFT) | + GEN9_IGNORE_SLICE_RATIO)); + intel_rps_set_manual(rps, false); + } else { + intel_rps_set(rps, rps->rp0_freq); + } + + mutex_unlock(&rps->lock); +} + +void intel_rps_lower_unslice(struct intel_rps *rps) +{ + struct intel_uncore *uncore = rps_to_uncore(rps); + u32 rpn_unslice_req; + + mutex_lock(&rps->lock); + + if (rps_uses_slpc(rps)) { + /* RP limits have not been initialized yet for SLPC path */ + rpn_unslice_req = ((intel_rps_read_state_cap(rps) >> 16) + & 0xff) * GEN9_FREQ_SCALER; + + intel_rps_set_manual(rps, true); + intel_uncore_write(uncore, GEN6_RPNSWREQ, + ((rpn_unslice_req << + GEN9_SW_REQ_UNSLICE_RATIO_SHIFT) | + GEN9_IGNORE_SLICE_RATIO)); + intel_rps_set_manual(rps, false); + } else { + intel_rps_set(rps, rps->min_freq); + } + + mutex_unlock(&rps->lock); +} + /* External interface for intel_ips.ko */ static struct drm_i915_private __rcu *ips_mchdev; @@ -2302,7 +2361,7 @@ unsigned long i915_read_mch_val(void) return 0; with_intel_runtime_pm(&i915->runtime_pm, wakeref) { - struct intel_ips *ips = &i915->gt.rps.ips; + struct intel_ips *ips = &to_gt(i915)->rps.ips; spin_lock_irq(&mchdev_lock); chipset_val = __ips_chipset_val(ips); @@ -2329,7 +2388,7 @@ bool i915_gpu_raise(void) if (!i915) return false; - rps = &i915->gt.rps; + rps = &to_gt(i915)->rps; spin_lock_irq(&mchdev_lock); if (rps->max_freq_softlimit < rps->max_freq) @@ -2356,7 +2415,7 @@ bool i915_gpu_lower(void) if (!i915) return false; - rps = &i915->gt.rps; + rps = &to_gt(i915)->rps; spin_lock_irq(&mchdev_lock); if (rps->max_freq_softlimit > rps->min_freq) @@ -2382,7 +2441,7 @@ bool i915_gpu_busy(void) if (!i915) return false; - ret = i915->gt.awake; + ret = to_gt(i915)->awake; drm_dev_put(&i915->drm); return ret; @@ -2405,11 +2464,11 @@ bool i915_gpu_turbo_disable(void) if (!i915) return false; - rps = &i915->gt.rps; + rps = &to_gt(i915)->rps; spin_lock_irq(&mchdev_lock); rps->max_freq_softlimit = rps->min_freq; - ret = !__gen5_rps_set(&i915->gt.rps, rps->min_freq); + ret = !__gen5_rps_set(&to_gt(i915)->rps, rps->min_freq); spin_unlock_irq(&mchdev_lock); drm_dev_put(&i915->drm); diff --git a/drivers/gpu/drm/i915/gt/intel_rps.h b/drivers/gpu/drm/i915/gt/intel_rps.h index aee12f37d38a..c6d76a3d1331 100644 --- a/drivers/gpu/drm/i915/gt/intel_rps.h +++ b/drivers/gpu/drm/i915/gt/intel_rps.h @@ -45,6 +45,8 @@ u32 intel_rps_get_rpn_frequency(struct intel_rps *rps); u32 intel_rps_read_punit_req(struct intel_rps *rps); u32 intel_rps_read_punit_req_frequency(struct intel_rps *rps); u32 intel_rps_read_state_cap(struct intel_rps *rps); +void intel_rps_raise_unslice(struct intel_rps *rps); +void intel_rps_lower_unslice(struct intel_rps *rps); void gen5_rps_irq_handler(struct intel_rps *rps); void gen6_rps_irq_handler(struct intel_rps *rps, u32 pm_iir); diff --git a/drivers/gpu/drm/i915/gt/intel_workarounds.c b/drivers/gpu/drm/i915/gt/intel_workarounds.c index 3113266c286e..ab3277a3d593 100644 --- a/drivers/gpu/drm/i915/gt/intel_workarounds.c +++ b/drivers/gpu/drm/i915/gt/intel_workarounds.c @@ -929,7 +929,7 @@ hsw_gt_workarounds_init(struct intel_gt *gt, struct i915_wa_list *wal) static void gen9_wa_init_mcr(struct drm_i915_private *i915, struct i915_wa_list *wal) { - const struct sseu_dev_info *sseu = &i915->gt.info.sseu; + const struct sseu_dev_info *sseu = &to_gt(i915)->info.sseu; unsigned int slice, subslice; u32 mcr, mcr_mask; diff --git a/drivers/gpu/drm/i915/gt/mock_engine.c b/drivers/gpu/drm/i915/gt/mock_engine.c index bb99fc03f503..c0637bf799a3 100644 --- a/drivers/gpu/drm/i915/gt/mock_engine.c +++ b/drivers/gpu/drm/i915/gt/mock_engine.c @@ -17,7 +17,7 @@ static int mock_timeline_pin(struct intel_timeline *tl) { int err; - if (WARN_ON(!i915_gem_object_trylock(tl->hwsp_ggtt->obj))) + if (WARN_ON(!i915_gem_object_trylock(tl->hwsp_ggtt->obj, NULL))) return -EBUSY; err = intel_timeline_pin_map(tl); @@ -345,7 +345,7 @@ struct intel_engine_cs *mock_engine(struct drm_i915_private *i915, struct mock_engine *engine; GEM_BUG_ON(id >= I915_NUM_ENGINES); - GEM_BUG_ON(!i915->gt.uncore); + GEM_BUG_ON(!to_gt(i915)->uncore); engine = kzalloc(sizeof(*engine) + PAGE_SIZE, GFP_KERNEL); if (!engine) @@ -353,8 +353,8 @@ struct intel_engine_cs *mock_engine(struct drm_i915_private *i915, /* minimal engine setup for requests */ engine->base.i915 = i915; - engine->base.gt = &i915->gt; - engine->base.uncore = i915->gt.uncore; + engine->base.gt = to_gt(i915); + engine->base.uncore = to_gt(i915)->uncore; snprintf(engine->base.name, sizeof(engine->base.name), "%s", name); engine->base.id = id; engine->base.mask = BIT(id); @@ -377,8 +377,8 @@ struct intel_engine_cs *mock_engine(struct drm_i915_private *i915, engine->base.release = mock_engine_release; - i915->gt.engine[id] = &engine->base; - i915->gt.engine_class[0][id] = &engine->base; + to_gt(i915)->engine[id] = &engine->base; + to_gt(i915)->engine_class[0][id] = &engine->base; /* fake hw queue */ spin_lock_init(&engine->hw_lock); diff --git a/drivers/gpu/drm/i915/gt/selftest_context.c b/drivers/gpu/drm/i915/gt/selftest_context.c index fa7b99a671dd..76fbae358072 100644 --- a/drivers/gpu/drm/i915/gt/selftest_context.c +++ b/drivers/gpu/drm/i915/gt/selftest_context.c @@ -442,7 +442,7 @@ int intel_context_live_selftests(struct drm_i915_private *i915) SUBTEST(live_active_context), SUBTEST(live_remote_context), }; - struct intel_gt *gt = &i915->gt; + struct intel_gt *gt = to_gt(i915); if (intel_gt_is_wedged(gt)) return 0; diff --git a/drivers/gpu/drm/i915/gt/selftest_engine.c b/drivers/gpu/drm/i915/gt/selftest_engine.c index 262764f6d90a..57fea9ea1705 100644 --- a/drivers/gpu/drm/i915/gt/selftest_engine.c +++ b/drivers/gpu/drm/i915/gt/selftest_engine.c @@ -12,7 +12,7 @@ int intel_engine_live_selftests(struct drm_i915_private *i915) live_engine_pm_selftests, NULL, }; - struct intel_gt *gt = &i915->gt; + struct intel_gt *gt = to_gt(i915); typeof(*tests) *fn; for (fn = tests; *fn; fn++) { diff --git a/drivers/gpu/drm/i915/gt/selftest_engine_cs.c b/drivers/gpu/drm/i915/gt/selftest_engine_cs.c index 64abf5feabfa..1b75f478d1b8 100644 --- a/drivers/gpu/drm/i915/gt/selftest_engine_cs.c +++ b/drivers/gpu/drm/i915/gt/selftest_engine_cs.c @@ -361,10 +361,10 @@ int intel_engine_cs_perf_selftests(struct drm_i915_private *i915) SUBTEST(perf_mi_noop), }; - if (intel_gt_is_wedged(&i915->gt)) + if (intel_gt_is_wedged(to_gt(i915))) return 0; - return intel_gt_live_subtests(tests, &i915->gt); + return intel_gt_live_subtests(tests, to_gt(i915)); } static int intel_mmio_bases_check(void *arg) diff --git a/drivers/gpu/drm/i915/gt/selftest_engine_heartbeat.c b/drivers/gpu/drm/i915/gt/selftest_engine_heartbeat.c index 6e6e4d747cca..273d440a53e3 100644 --- a/drivers/gpu/drm/i915/gt/selftest_engine_heartbeat.c +++ b/drivers/gpu/drm/i915/gt/selftest_engine_heartbeat.c @@ -378,13 +378,13 @@ int intel_heartbeat_live_selftests(struct drm_i915_private *i915) int saved_hangcheck; int err; - if (intel_gt_is_wedged(&i915->gt)) + if (intel_gt_is_wedged(to_gt(i915))) return 0; saved_hangcheck = i915->params.enable_hangcheck; i915->params.enable_hangcheck = INT_MAX; - err = intel_gt_live_subtests(tests, &i915->gt); + err = intel_gt_live_subtests(tests, to_gt(i915)); i915->params.enable_hangcheck = saved_hangcheck; return err; diff --git a/drivers/gpu/drm/i915/gt/selftest_engine_pm.c b/drivers/gpu/drm/i915/gt/selftest_engine_pm.c index 75f6efc9882f..8af261831470 100644 --- a/drivers/gpu/drm/i915/gt/selftest_engine_pm.c +++ b/drivers/gpu/drm/i915/gt/selftest_engine_pm.c @@ -229,7 +229,7 @@ static int __spin_until_busier(struct intel_engine_cs *engine, ktime_t busyness) start = ktime_get(); while (intel_engine_get_busy_time(engine, &unused) == busyness) { dt = ktime_get() - start; - if (dt > 500000) { + if (dt > 10000000) { pr_err("active wait timed out %lld\n", dt); ENGINE_TRACE(engine, "active wait time out %lld\n", dt); return -ETIME; diff --git a/drivers/gpu/drm/i915/gt/selftest_execlists.c b/drivers/gpu/drm/i915/gt/selftest_execlists.c index b367ecfa42de..e10da897e07a 100644 --- a/drivers/gpu/drm/i915/gt/selftest_execlists.c +++ b/drivers/gpu/drm/i915/gt/selftest_execlists.c @@ -4502,11 +4502,11 @@ int intel_execlists_live_selftests(struct drm_i915_private *i915) SUBTEST(live_virtual_reset), }; - if (i915->gt.submission_method != INTEL_SUBMISSION_ELSP) + if (to_gt(i915)->submission_method != INTEL_SUBMISSION_ELSP) return 0; - if (intel_gt_is_wedged(&i915->gt)) + if (intel_gt_is_wedged(to_gt(i915))) return 0; - return intel_gt_live_subtests(tests, &i915->gt); + return intel_gt_live_subtests(tests, to_gt(i915)); } diff --git a/drivers/gpu/drm/i915/gt/selftest_gt_pm.c b/drivers/gpu/drm/i915/gt/selftest_gt_pm.c index 55c5cdb99f45..8bf62a5826cc 100644 --- a/drivers/gpu/drm/i915/gt/selftest_gt_pm.c +++ b/drivers/gpu/drm/i915/gt/selftest_gt_pm.c @@ -193,10 +193,10 @@ int intel_gt_pm_live_selftests(struct drm_i915_private *i915) SUBTEST(live_gt_resume), }; - if (intel_gt_is_wedged(&i915->gt)) + if (intel_gt_is_wedged(to_gt(i915))) return 0; - return intel_gt_live_subtests(tests, &i915->gt); + return intel_gt_live_subtests(tests, to_gt(i915)); } int intel_gt_pm_late_selftests(struct drm_i915_private *i915) @@ -210,8 +210,8 @@ int intel_gt_pm_late_selftests(struct drm_i915_private *i915) SUBTEST(live_rc6_ctx_wa), }; - if (intel_gt_is_wedged(&i915->gt)) + if (intel_gt_is_wedged(to_gt(i915))) return 0; - return intel_gt_live_subtests(tests, &i915->gt); + return intel_gt_live_subtests(tests, to_gt(i915)); } diff --git a/drivers/gpu/drm/i915/gt/selftest_hangcheck.c b/drivers/gpu/drm/i915/gt/selftest_hangcheck.c index e5ad4d5a91c0..15d63435ec4d 100644 --- a/drivers/gpu/drm/i915/gt/selftest_hangcheck.c +++ b/drivers/gpu/drm/i915/gt/selftest_hangcheck.c @@ -2018,7 +2018,7 @@ int intel_hangcheck_live_selftests(struct drm_i915_private *i915) SUBTEST(igt_reset_evict_fence), SUBTEST(igt_handle_error), }; - struct intel_gt *gt = &i915->gt; + struct intel_gt *gt = to_gt(i915); intel_wakeref_t wakeref; int err; diff --git a/drivers/gpu/drm/i915/gt/selftest_lrc.c b/drivers/gpu/drm/i915/gt/selftest_lrc.c index b0977a3b699b..618c905daa19 100644 --- a/drivers/gpu/drm/i915/gt/selftest_lrc.c +++ b/drivers/gpu/drm/i915/gt/selftest_lrc.c @@ -1847,5 +1847,5 @@ int intel_lrc_live_selftests(struct drm_i915_private *i915) if (!HAS_LOGICAL_RING_CONTEXTS(i915)) return 0; - return intel_gt_live_subtests(tests, &i915->gt); + return intel_gt_live_subtests(tests, to_gt(i915)); } diff --git a/drivers/gpu/drm/i915/gt/selftest_migrate.c b/drivers/gpu/drm/i915/gt/selftest_migrate.c index e21787301bbd..fa4293d2944f 100644 --- a/drivers/gpu/drm/i915/gt/selftest_migrate.c +++ b/drivers/gpu/drm/i915/gt/selftest_migrate.c @@ -442,7 +442,7 @@ int intel_migrate_live_selftests(struct drm_i915_private *i915) SUBTEST(thread_global_copy), SUBTEST(thread_global_clear), }; - struct intel_gt *gt = &i915->gt; + struct intel_gt *gt = to_gt(i915); if (!gt->migrate.context) return 0; @@ -465,7 +465,7 @@ create_init_lmem_internal(struct intel_gt *gt, size_t sz, bool try_lmem) return obj; } - i915_gem_object_trylock(obj); + i915_gem_object_trylock(obj, NULL); err = i915_gem_object_pin_pages(obj); if (err) { i915_gem_object_unlock(obj); @@ -658,7 +658,7 @@ int intel_migrate_perf_selftests(struct drm_i915_private *i915) SUBTEST(perf_clear_blt), SUBTEST(perf_copy_blt), }; - struct intel_gt *gt = &i915->gt; + struct intel_gt *gt = to_gt(i915); if (intel_gt_is_wedged(gt)) return 0; diff --git a/drivers/gpu/drm/i915/gt/selftest_mocs.c b/drivers/gpu/drm/i915/gt/selftest_mocs.c index 13d25bf2a94a..c1d861333c44 100644 --- a/drivers/gpu/drm/i915/gt/selftest_mocs.c +++ b/drivers/gpu/drm/i915/gt/selftest_mocs.c @@ -451,5 +451,5 @@ int intel_mocs_live_selftests(struct drm_i915_private *i915) if (!get_mocs_settings(i915, &table)) return 0; - return intel_gt_live_subtests(tests, &i915->gt); + return intel_gt_live_subtests(tests, to_gt(i915)); } diff --git a/drivers/gpu/drm/i915/gt/selftest_reset.c b/drivers/gpu/drm/i915/gt/selftest_reset.c index 7a50c9f4071b..8a873f6bda7f 100644 --- a/drivers/gpu/drm/i915/gt/selftest_reset.c +++ b/drivers/gpu/drm/i915/gt/selftest_reset.c @@ -376,7 +376,7 @@ int intel_reset_live_selftests(struct drm_i915_private *i915) SUBTEST(igt_atomic_reset), SUBTEST(igt_atomic_engine_reset), }; - struct intel_gt *gt = &i915->gt; + struct intel_gt *gt = to_gt(i915); if (!intel_has_gpu_reset(gt)) return 0; diff --git a/drivers/gpu/drm/i915/gt/selftest_ring_submission.c b/drivers/gpu/drm/i915/gt/selftest_ring_submission.c index 041954408d0f..70f9ac1ec2c7 100644 --- a/drivers/gpu/drm/i915/gt/selftest_ring_submission.c +++ b/drivers/gpu/drm/i915/gt/selftest_ring_submission.c @@ -291,8 +291,8 @@ int intel_ring_submission_live_selftests(struct drm_i915_private *i915) SUBTEST(live_ctx_switch_wa), }; - if (i915->gt.submission_method > INTEL_SUBMISSION_RING) + if (to_gt(i915)->submission_method > INTEL_SUBMISSION_RING) return 0; - return intel_gt_live_subtests(tests, &i915->gt); + return intel_gt_live_subtests(tests, to_gt(i915)); } diff --git a/drivers/gpu/drm/i915/gt/selftest_slpc.c b/drivers/gpu/drm/i915/gt/selftest_slpc.c index 9334bad131a2..b768cea5943d 100644 --- a/drivers/gpu/drm/i915/gt/selftest_slpc.c +++ b/drivers/gpu/drm/i915/gt/selftest_slpc.c @@ -39,7 +39,7 @@ static int slpc_set_max_freq(struct intel_guc_slpc *slpc, u32 freq) static int live_slpc_clamp_min(void *arg) { struct drm_i915_private *i915 = arg; - struct intel_gt *gt = &i915->gt; + struct intel_gt *gt = to_gt(i915); struct intel_guc_slpc *slpc = >->uc.guc.slpc; struct intel_rps *rps = >->rps; struct intel_engine_cs *engine; @@ -166,7 +166,7 @@ static int live_slpc_clamp_min(void *arg) static int live_slpc_clamp_max(void *arg) { struct drm_i915_private *i915 = arg; - struct intel_gt *gt = &i915->gt; + struct intel_gt *gt = to_gt(i915); struct intel_guc_slpc *slpc; struct intel_rps *rps; struct intel_engine_cs *engine; @@ -304,7 +304,7 @@ int intel_slpc_live_selftests(struct drm_i915_private *i915) SUBTEST(live_slpc_clamp_min), }; - if (intel_gt_is_wedged(&i915->gt)) + if (intel_gt_is_wedged(to_gt(i915))) return 0; return i915_live_subtests(tests, i915); diff --git a/drivers/gpu/drm/i915/gt/selftest_timeline.c b/drivers/gpu/drm/i915/gt/selftest_timeline.c index d0b6a3afcf44..e2eb686a9763 100644 --- a/drivers/gpu/drm/i915/gt/selftest_timeline.c +++ b/drivers/gpu/drm/i915/gt/selftest_timeline.c @@ -159,7 +159,7 @@ static int mock_hwsp_freelist(void *arg) INIT_RADIX_TREE(&state.cachelines, GFP_KERNEL); state.prng = I915_RND_STATE_INITIALIZER(i915_selftest.random_seed); - state.gt = &i915->gt; + state.gt = to_gt(i915); /* * Create a bunch of timelines and check that their HWSP do not overlap. @@ -1416,8 +1416,8 @@ int intel_timeline_live_selftests(struct drm_i915_private *i915) SUBTEST(live_hwsp_rollover_user), }; - if (intel_gt_is_wedged(&i915->gt)) + if (intel_gt_is_wedged(to_gt(i915))) return 0; - return intel_gt_live_subtests(tests, &i915->gt); + return intel_gt_live_subtests(tests, to_gt(i915)); } diff --git a/drivers/gpu/drm/i915/gt/selftest_workarounds.c b/drivers/gpu/drm/i915/gt/selftest_workarounds.c index 962e91ba3be4..0287c2573c51 100644 --- a/drivers/gpu/drm/i915/gt/selftest_workarounds.c +++ b/drivers/gpu/drm/i915/gt/selftest_workarounds.c @@ -1387,8 +1387,8 @@ int intel_workarounds_live_selftests(struct drm_i915_private *i915) SUBTEST(live_engine_reset_workarounds), }; - if (intel_gt_is_wedged(&i915->gt)) + if (intel_gt_is_wedged(to_gt(i915))) return 0; - return intel_gt_live_subtests(tests, &i915->gt); + return intel_gt_live_subtests(tests, to_gt(i915)); } diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc.h b/drivers/gpu/drm/i915/gt/uc/intel_guc.h index 1cb46098030d..f9240d4baa69 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_guc.h +++ b/drivers/gpu/drm/i915/gt/uc/intel_guc.h @@ -95,6 +95,11 @@ struct intel_guc { */ struct ida guc_ids; /** + * @num_guc_ids: Number of guc_ids, selftest feature to be able + * to reduce this number while testing. + */ + int num_guc_ids; + /** * @guc_ids_bitmap: used to allocate new guc_ids, multi-lrc */ unsigned long *guc_ids_bitmap; @@ -202,6 +207,13 @@ struct intel_guc { */ struct delayed_work work; } timestamp; + +#ifdef CONFIG_DRM_I915_SELFTEST + /** + * @number_guc_id_stolen: The number of guc_ids that have been stolen + */ + int number_guc_id_stolen; +#endif }; static inline struct intel_guc *log_to_guc(struct intel_guc_log *log) diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c index a0cc34be7b56..aa6dd6415202 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_ct.c @@ -523,6 +523,15 @@ static inline bool ct_deadlocked(struct intel_guc_ct *ct) CT_ERROR(ct, "Communication stalled for %lld ms, desc status=%#x,%#x\n", ktime_ms_delta(ktime_get(), ct->stall_time), send->status, recv->status); + CT_ERROR(ct, "H2G Space: %u (Bytes)\n", + atomic_read(&ct->ctbs.send.space) * 4); + CT_ERROR(ct, "Head: %u (Dwords)\n", ct->ctbs.send.desc->head); + CT_ERROR(ct, "Tail: %u (Dwords)\n", ct->ctbs.send.desc->tail); + CT_ERROR(ct, "G2H Space: %u (Bytes)\n", + atomic_read(&ct->ctbs.recv.space) * 4); + CT_ERROR(ct, "Head: %u\n (Dwords)", ct->ctbs.recv.desc->head); + CT_ERROR(ct, "Tail: %u\n (Dwords)", ct->ctbs.recv.desc->tail); + ct->ctbs.send.broken = true; } @@ -582,12 +591,19 @@ static inline bool h2g_has_room(struct intel_guc_ct *ct, u32 len_dw) static int has_room_nb(struct intel_guc_ct *ct, u32 h2g_dw, u32 g2h_dw) { + bool h2g = h2g_has_room(ct, h2g_dw); + bool g2h = g2h_has_room(ct, g2h_dw); + lockdep_assert_held(&ct->ctbs.send.lock); - if (unlikely(!h2g_has_room(ct, h2g_dw) || !g2h_has_room(ct, g2h_dw))) { + if (unlikely(!h2g || !g2h)) { if (ct->stall_time == KTIME_MAX) ct->stall_time = ktime_get(); + /* Be paranoid and kick G2H tasklet to free credits */ + if (!g2h) + tasklet_hi_schedule(&ct->receive_tasklet); + if (unlikely(ct_deadlocked(ct))) return -EPIPE; else diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_fw.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_fw.c index 196424be0998..31420ce1ce6b 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_guc_fw.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_fw.c @@ -40,9 +40,8 @@ static void guc_prepare_xfer(struct intel_uncore *uncore) } } -/* Copy RSA signature from the fw image to HW for verification */ -static int guc_xfer_rsa(struct intel_uc_fw *guc_fw, - struct intel_uncore *uncore) +static int guc_xfer_rsa_mmio(struct intel_uc_fw *guc_fw, + struct intel_uncore *uncore) { u32 rsa[UOS_RSA_SCRATCH_COUNT]; size_t copied; @@ -58,6 +57,27 @@ static int guc_xfer_rsa(struct intel_uc_fw *guc_fw, return 0; } +static int guc_xfer_rsa_vma(struct intel_uc_fw *guc_fw, + struct intel_uncore *uncore) +{ + struct intel_guc *guc = container_of(guc_fw, struct intel_guc, fw); + + intel_uncore_write(uncore, UOS_RSA_SCRATCH(0), + intel_guc_ggtt_offset(guc, guc_fw->rsa_data)); + + return 0; +} + +/* Copy RSA signature from the fw image to HW for verification */ +static int guc_xfer_rsa(struct intel_uc_fw *guc_fw, + struct intel_uncore *uncore) +{ + if (guc_fw->rsa_data) + return guc_xfer_rsa_vma(guc_fw, uncore); + else + return guc_xfer_rsa_mmio(guc_fw, uncore); +} + /* * Read the GuC status register (GUC_STATUS) and store it in the * specified location; then return a boolean indicating whether @@ -142,7 +162,10 @@ int intel_guc_fw_upload(struct intel_guc *guc) /* * Note that GuC needs the CSS header plus uKernel code to be copied * by the DMA engine in one operation, whereas the RSA signature is - * loaded via MMIO. + * loaded separately, either by copying it to the UOS_RSA_SCRATCH + * register (if key size <= 256) or through a ggtt-pinned vma (if key + * size > 256). The RSA size and therefore the way we provide it to the + * HW is fixed for each platform and hard-coded in the bootrom. */ ret = guc_xfer_rsa(&guc->fw, uncore); if (ret) @@ -164,6 +187,6 @@ int intel_guc_fw_upload(struct intel_guc *guc) return 0; out: - intel_uc_fw_change_status(&guc->fw, INTEL_UC_FIRMWARE_FAIL); + intel_uc_fw_change_status(&guc->fw, INTEL_UC_FIRMWARE_LOAD_FAIL); return ret; } diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_log.h b/drivers/gpu/drm/i915/gt/uc/intel_guc_log.h index ac1ee1d5ce10..fe6ab7550a14 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_guc_log.h +++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_log.h @@ -15,9 +15,12 @@ struct intel_guc; -#ifdef CONFIG_DRM_I915_DEBUG_GUC +#if defined(CONFIG_DRM_I915_DEBUG_GUC) #define CRASH_BUFFER_SIZE SZ_2M #define DEBUG_BUFFER_SIZE SZ_16M +#elif defined(CONFIG_DRM_I915_DEBUG_GEM) +#define CRASH_BUFFER_SIZE SZ_1M +#define DEBUG_BUFFER_SIZE SZ_2M #else #define CRASH_BUFFER_SIZE SZ_8K #define DEBUG_BUFFER_SIZE SZ_64K diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_log_debugfs.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_log_debugfs.c index 46026c2c1722..ddfbe334689f 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_guc_log_debugfs.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_log_debugfs.c @@ -10,28 +10,80 @@ #include "intel_guc.h" #include "intel_guc_log.h" #include "intel_guc_log_debugfs.h" +#include "intel_uc.h" + +static u32 obj_to_guc_log_dump_size(struct drm_i915_gem_object *obj) +{ + u32 size; + + if (!obj) + return PAGE_SIZE; + + /* "0x%08x 0x%08x 0x%08x 0x%08x\n" => 16 bytes -> 44 chars => x2.75 */ + size = ((obj->base.size * 11) + 3) / 4; + + /* Add padding for final blank line, any extra header info, etc. */ + size = PAGE_ALIGN(size + PAGE_SIZE); + + return size; +} + +static u32 guc_log_dump_size(struct intel_guc_log *log) +{ + struct intel_guc *guc = log_to_guc(log); + + if (!intel_guc_is_supported(guc)) + return PAGE_SIZE; + + if (!log->vma) + return PAGE_SIZE; + + return obj_to_guc_log_dump_size(log->vma->obj); +} static int guc_log_dump_show(struct seq_file *m, void *data) { struct drm_printer p = drm_seq_file_printer(m); + int ret; - return intel_guc_log_dump(m->private, &p, false); + ret = intel_guc_log_dump(m->private, &p, false); + + if (IS_ENABLED(CONFIG_DRM_I915_DEBUG_GEM) && seq_has_overflowed(m)) + pr_warn_once("preallocated size:%zx for %s exceeded\n", + m->size, __func__); + + return ret; +} +DEFINE_INTEL_GT_DEBUGFS_ATTRIBUTE_WITH_SIZE(guc_log_dump, guc_log_dump_size); + +static u32 guc_load_err_dump_size(struct intel_guc_log *log) +{ + struct intel_guc *guc = log_to_guc(log); + struct intel_uc *uc = container_of(guc, struct intel_uc, guc); + + if (!intel_guc_is_supported(guc)) + return PAGE_SIZE; + + return obj_to_guc_log_dump_size(uc->load_err_log); } -DEFINE_INTEL_GT_DEBUGFS_ATTRIBUTE(guc_log_dump); static int guc_load_err_log_dump_show(struct seq_file *m, void *data) { struct drm_printer p = drm_seq_file_printer(m); + if (IS_ENABLED(CONFIG_DRM_I915_DEBUG_GEM) && seq_has_overflowed(m)) + pr_warn_once("preallocated size:%zx for %s exceeded\n", + m->size, __func__); + return intel_guc_log_dump(m->private, &p, true); } -DEFINE_INTEL_GT_DEBUGFS_ATTRIBUTE(guc_load_err_log_dump); +DEFINE_INTEL_GT_DEBUGFS_ATTRIBUTE_WITH_SIZE(guc_load_err_log_dump, guc_load_err_dump_size); static int guc_log_level_get(void *data, u64 *val) { struct intel_guc_log *log = data; - if (!intel_guc_is_used(log_to_guc(log))) + if (!log->vma) return -ENODEV; *val = intel_guc_log_get_level(log); @@ -43,7 +95,7 @@ static int guc_log_level_set(void *data, u64 val) { struct intel_guc_log *log = data; - if (!intel_guc_is_used(log_to_guc(log))) + if (!log->vma) return -ENODEV; return intel_guc_log_set_level(log, val); diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_slpc.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_slpc.c index 22c1c12369f2..13b27b8ff74e 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_guc_slpc.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_slpc.c @@ -623,7 +623,7 @@ int intel_guc_slpc_enable(struct intel_guc_slpc *slpc) if (unlikely(ret < 0)) return ret; - intel_guc_pm_intrmsk_enable(&i915->gt); + intel_guc_pm_intrmsk_enable(to_gt(i915)); slpc_get_rp_values(slpc); diff --git a/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c b/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c index 1f9d4fde421f..e7517206af82 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_guc_submission.c @@ -145,7 +145,8 @@ guc_create_parallel(struct intel_engine_cs **engines, * use should be low and 1/16 should be sufficient. Minimum of 32 guc_ids for * multi-lrc. */ -#define NUMBER_MULTI_LRC_GUC_ID (GUC_MAX_LRC_DESCRIPTORS / 16) +#define NUMBER_MULTI_LRC_GUC_ID(guc) \ + ((guc)->submission_state.num_guc_ids / 16) /* * Below is a set of functions which control the GuC scheduling state which @@ -1040,8 +1041,6 @@ static void scrub_guc_desc_for_outstanding_g2h(struct intel_guc *guc) spin_unlock(&ce->guc_state.lock); - GEM_BUG_ON(!do_put && !destroyed); - if (pending_enable || destroyed || deregister) { decr_outstanding_submission_g2h(guc); if (deregister) @@ -1206,7 +1205,7 @@ static ktime_t guc_engine_busyness(struct intel_engine_cs *engine, ktime_t *now) * start_gt_clk is derived from GuC state. To get a consistent * view of activity, we query the GuC state only if gt is awake. */ - if (intel_gt_pm_get_if_awake(gt) && !in_reset) { + if (!in_reset && intel_gt_pm_get_if_awake(gt)) { stats_saved = *stats; gt_stamp_saved = guc->timestamp.gt_stamp; guc_update_engine_gt_clks(engine); @@ -1777,7 +1776,7 @@ int intel_guc_submission_init(struct intel_guc *guc) destroyed_worker_func); guc->submission_state.guc_ids_bitmap = - bitmap_zalloc(NUMBER_MULTI_LRC_GUC_ID, GFP_KERNEL); + bitmap_zalloc(NUMBER_MULTI_LRC_GUC_ID(guc), GFP_KERNEL); if (!guc->submission_state.guc_ids_bitmap) return -ENOMEM; @@ -1871,13 +1870,13 @@ static int new_guc_id(struct intel_guc *guc, struct intel_context *ce) if (intel_context_is_parent(ce)) ret = bitmap_find_free_region(guc->submission_state.guc_ids_bitmap, - NUMBER_MULTI_LRC_GUC_ID, + NUMBER_MULTI_LRC_GUC_ID(guc), order_base_2(ce->parallel.number_children + 1)); else ret = ida_simple_get(&guc->submission_state.guc_ids, - NUMBER_MULTI_LRC_GUC_ID, - GUC_MAX_LRC_DESCRIPTORS, + NUMBER_MULTI_LRC_GUC_ID(guc), + guc->submission_state.num_guc_ids, GFP_KERNEL | __GFP_RETRY_MAYFAIL | __GFP_NOWARN); if (unlikely(ret < 0)) @@ -1935,14 +1934,18 @@ static int steal_guc_id(struct intel_guc *guc, struct intel_context *ce) GEM_BUG_ON(intel_context_is_parent(cn)); list_del_init(&cn->guc_id.link); - ce->guc_id = cn->guc_id; + ce->guc_id.id = cn->guc_id.id; - spin_lock(&ce->guc_state.lock); + spin_lock(&cn->guc_state.lock); clr_context_registered(cn); - spin_unlock(&ce->guc_state.lock); + spin_unlock(&cn->guc_state.lock); set_context_guc_id_invalid(cn); +#ifdef CONFIG_DRM_I915_SELFTEST + guc->number_guc_id_stolen++; +#endif + return 0; } else { return -EAGAIN; @@ -2646,7 +2649,6 @@ static inline void guc_lrc_desc_unpin(struct intel_context *ce) unsigned long flags; bool disabled; - lockdep_assert_held(&guc->submission_state.lock); GEM_BUG_ON(!intel_gt_pm_is_awake(gt)); GEM_BUG_ON(!lrc_desc_registered(guc, ce->guc_id.id)); GEM_BUG_ON(ce != __get_context(guc, ce->guc_id.id)); @@ -2662,7 +2664,7 @@ static inline void guc_lrc_desc_unpin(struct intel_context *ce) } spin_unlock_irqrestore(&ce->guc_state.lock, flags); if (unlikely(disabled)) { - __release_guc_id(guc, ce); + release_guc_id(guc, ce); __guc_context_destroy(ce); return; } @@ -2696,36 +2698,48 @@ static void __guc_context_destroy(struct intel_context *ce) static void guc_flush_destroyed_contexts(struct intel_guc *guc) { - struct intel_context *ce, *cn; + struct intel_context *ce; unsigned long flags; GEM_BUG_ON(!submission_disabled(guc) && guc_submission_initialized(guc)); - spin_lock_irqsave(&guc->submission_state.lock, flags); - list_for_each_entry_safe(ce, cn, - &guc->submission_state.destroyed_contexts, - destroyed_link) { - list_del_init(&ce->destroyed_link); - __release_guc_id(guc, ce); + while (!list_empty(&guc->submission_state.destroyed_contexts)) { + spin_lock_irqsave(&guc->submission_state.lock, flags); + ce = list_first_entry_or_null(&guc->submission_state.destroyed_contexts, + struct intel_context, + destroyed_link); + if (ce) + list_del_init(&ce->destroyed_link); + spin_unlock_irqrestore(&guc->submission_state.lock, flags); + + if (!ce) + break; + + release_guc_id(guc, ce); __guc_context_destroy(ce); } - spin_unlock_irqrestore(&guc->submission_state.lock, flags); } static void deregister_destroyed_contexts(struct intel_guc *guc) { - struct intel_context *ce, *cn; + struct intel_context *ce; unsigned long flags; - spin_lock_irqsave(&guc->submission_state.lock, flags); - list_for_each_entry_safe(ce, cn, - &guc->submission_state.destroyed_contexts, - destroyed_link) { - list_del_init(&ce->destroyed_link); + while (!list_empty(&guc->submission_state.destroyed_contexts)) { + spin_lock_irqsave(&guc->submission_state.lock, flags); + ce = list_first_entry_or_null(&guc->submission_state.destroyed_contexts, + struct intel_context, + destroyed_link); + if (ce) + list_del_init(&ce->destroyed_link); + spin_unlock_irqrestore(&guc->submission_state.lock, flags); + + if (!ce) + break; + guc_lrc_desc_unpin(ce); } - spin_unlock_irqrestore(&guc->submission_state.lock, flags); } static void destroyed_worker_func(struct work_struct *w) @@ -3770,6 +3784,7 @@ static bool __guc_submission_selected(struct intel_guc *guc) void intel_guc_submission_init_early(struct intel_guc *guc) { + guc->submission_state.num_guc_ids = GUC_MAX_LRC_DESCRIPTORS; guc->submission_supported = __guc_submission_supported(guc); guc->submission_selected = __guc_submission_selected(guc); } @@ -4018,11 +4033,12 @@ int intel_guc_engine_failure_process_msg(struct intel_guc *guc, const u32 *msg, u32 len) { struct intel_engine_cs *engine; + struct intel_gt *gt = guc_to_gt(guc); u8 guc_class, instance; u32 reason; if (unlikely(len != 3)) { - drm_err(&guc_to_gt(guc)->i915->drm, "Invalid length %u", len); + drm_err(>->i915->drm, "Invalid length %u", len); return -EPROTO; } @@ -4032,12 +4048,19 @@ int intel_guc_engine_failure_process_msg(struct intel_guc *guc, engine = guc_lookup_engine(guc, guc_class, instance); if (unlikely(!engine)) { - drm_err(&guc_to_gt(guc)->i915->drm, + drm_err(>->i915->drm, "Invalid engine %d:%d", guc_class, instance); return -EPROTO; } - intel_gt_handle_error(guc_to_gt(guc), engine->mask, + /* + * This is an unexpected failure of a hardware feature. So, log a real + * error message not just the informational that comes with the reset. + */ + drm_err(>->i915->drm, "GuC engine reset request failed on %d:%d (%s) because 0x%08X", + guc_class, instance, engine->name, reason); + + intel_gt_handle_error(gt, engine->mask, I915_ERROR_CAPTURE, "GuC failed to reset %s (reason=0x%08x)\n", engine->name, reason); diff --git a/drivers/gpu/drm/i915/gt/uc/intel_huc.c b/drivers/gpu/drm/i915/gt/uc/intel_huc.c index ff4b6869b80b..d10b227ac4aa 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_huc.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_huc.c @@ -54,65 +54,6 @@ void intel_huc_init_early(struct intel_huc *huc) } } -static int intel_huc_rsa_data_create(struct intel_huc *huc) -{ - struct intel_gt *gt = huc_to_gt(huc); - struct intel_guc *guc = >->uc.guc; - struct i915_vma *vma; - size_t copied; - void *vaddr; - int err; - - err = i915_inject_probe_error(gt->i915, -ENXIO); - if (err) - return err; - - /* - * HuC firmware will sit above GUC_GGTT_TOP and will not map - * through GTT. Unfortunately, this means GuC cannot perform - * the HuC auth. as the rsa offset now falls within the GuC - * inaccessible range. We resort to perma-pinning an additional - * vma within the accessible range that only contains the rsa - * signature. The GuC can use this extra pinning to perform - * the authentication since its GGTT offset will be GuC - * accessible. - */ - GEM_BUG_ON(huc->fw.rsa_size > PAGE_SIZE); - vma = intel_guc_allocate_vma(guc, PAGE_SIZE); - if (IS_ERR(vma)) - return PTR_ERR(vma); - - vaddr = i915_gem_object_pin_map_unlocked(vma->obj, - i915_coherent_map_type(gt->i915, - vma->obj, true)); - if (IS_ERR(vaddr)) { - i915_vma_unpin_and_release(&vma, 0); - err = PTR_ERR(vaddr); - goto unpin_out; - } - - copied = intel_uc_fw_copy_rsa(&huc->fw, vaddr, vma->size); - i915_gem_object_unpin_map(vma->obj); - - if (copied < huc->fw.rsa_size) { - err = -ENOMEM; - goto unpin_out; - } - - huc->rsa_data = vma; - - return 0; - -unpin_out: - i915_vma_unpin_and_release(&vma, 0); - return err; -} - -static void intel_huc_rsa_data_destroy(struct intel_huc *huc) -{ - i915_vma_unpin_and_release(&huc->rsa_data, 0); -} - int intel_huc_init(struct intel_huc *huc) { struct drm_i915_private *i915 = huc_to_gt(huc)->i915; @@ -122,21 +63,10 @@ int intel_huc_init(struct intel_huc *huc) if (err) goto out; - /* - * HuC firmware image is outside GuC accessible range. - * Copy the RSA signature out of the image into - * a perma-pinned region set aside for it - */ - err = intel_huc_rsa_data_create(huc); - if (err) - goto out_fini; - intel_uc_fw_change_status(&huc->fw, INTEL_UC_FIRMWARE_LOADABLE); return 0; -out_fini: - intel_uc_fw_fini(&huc->fw); out: i915_probe_error(i915, "failed with %d\n", err); return err; @@ -147,7 +77,6 @@ void intel_huc_fini(struct intel_huc *huc) if (!intel_uc_fw_is_loadable(&huc->fw)) return; - intel_huc_rsa_data_destroy(huc); intel_uc_fw_fini(&huc->fw); } @@ -177,7 +106,7 @@ int intel_huc_auth(struct intel_huc *huc) goto fail; ret = intel_guc_auth_huc(guc, - intel_guc_ggtt_offset(guc, huc->rsa_data)); + intel_guc_ggtt_offset(guc, huc->fw.rsa_data)); if (ret) { DRM_ERROR("HuC: GuC did not ack Auth request %d\n", ret); goto fail; @@ -199,7 +128,7 @@ int intel_huc_auth(struct intel_huc *huc) fail: i915_probe_error(gt->i915, "HuC: Authentication failed %d\n", ret); - intel_uc_fw_change_status(&huc->fw, INTEL_UC_FIRMWARE_FAIL); + intel_uc_fw_change_status(&huc->fw, INTEL_UC_FIRMWARE_LOAD_FAIL); return ret; } diff --git a/drivers/gpu/drm/i915/gt/uc/intel_huc.h b/drivers/gpu/drm/i915/gt/uc/intel_huc.h index daee43b661d4..ae8c8a6c8cc8 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_huc.h +++ b/drivers/gpu/drm/i915/gt/uc/intel_huc.h @@ -15,8 +15,6 @@ struct intel_huc { struct intel_uc_fw fw; /* HuC-specific additions */ - struct i915_vma *rsa_data; - struct { i915_reg_t reg; u32 mask; diff --git a/drivers/gpu/drm/i915/gt/uc/intel_uc.c b/drivers/gpu/drm/i915/gt/uc/intel_uc.c index 8f17005ce85f..09ed29df67bc 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_uc.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_uc.c @@ -8,6 +8,7 @@ #include "intel_guc.h" #include "intel_guc_ads.h" #include "intel_guc_submission.h" +#include "gt/intel_rps.h" #include "intel_uc.h" #include "i915_drv.h" @@ -462,6 +463,8 @@ static int __uc_init_hw(struct intel_uc *uc) else attempts = 1; + intel_rps_raise_unslice(&uc_to_gt(uc)->rps); + while (attempts--) { /* * Always reset the GuC just before (re)loading, so @@ -499,6 +502,9 @@ static int __uc_init_hw(struct intel_uc *uc) ret = intel_guc_slpc_enable(&guc->slpc); if (ret) goto err_submission; + } else { + /* Restore GT back to RPn for non-SLPC path */ + intel_rps_lower_unslice(&uc_to_gt(uc)->rps); } drm_info(&i915->drm, "%s firmware %s version %u.%u %s:%s\n", @@ -529,6 +535,9 @@ err_submission: err_log_capture: __uc_capture_load_err_log(uc); err_out: + /* Return GT back to RPn */ + intel_rps_lower_unslice(&uc_to_gt(uc)->rps); + __uc_sanitize(uc); if (!ret) { diff --git a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c index 3aa87be4f2e4..a5af05bde6f2 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c +++ b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.c @@ -48,22 +48,39 @@ void intel_uc_fw_change_status(struct intel_uc_fw *uc_fw, * Note that RKL and ADL-S have the same GuC/HuC device ID's and use the same * firmware as TGL. */ -#define INTEL_UC_FIRMWARE_DEFS(fw_def, guc_def, huc_def) \ - fw_def(ALDERLAKE_P, 0, guc_def(adlp, 62, 0, 3), huc_def(tgl, 7, 9, 3)) \ - fw_def(ALDERLAKE_S, 0, guc_def(tgl, 62, 0, 0), huc_def(tgl, 7, 9, 3)) \ - fw_def(DG1, 0, guc_def(dg1, 62, 0, 0), huc_def(dg1, 7, 9, 3)) \ - fw_def(ROCKETLAKE, 0, guc_def(tgl, 62, 0, 0), huc_def(tgl, 7, 9, 3)) \ - fw_def(TIGERLAKE, 0, guc_def(tgl, 62, 0, 0), huc_def(tgl, 7, 9, 3)) \ - fw_def(JASPERLAKE, 0, guc_def(ehl, 62, 0, 0), huc_def(ehl, 9, 0, 0)) \ - fw_def(ELKHARTLAKE, 0, guc_def(ehl, 62, 0, 0), huc_def(ehl, 9, 0, 0)) \ - fw_def(ICELAKE, 0, guc_def(icl, 62, 0, 0), huc_def(icl, 9, 0, 0)) \ - fw_def(COMETLAKE, 5, guc_def(cml, 62, 0, 0), huc_def(cml, 4, 0, 0)) \ - fw_def(COMETLAKE, 0, guc_def(kbl, 62, 0, 0), huc_def(kbl, 4, 0, 0)) \ - fw_def(COFFEELAKE, 0, guc_def(kbl, 62, 0, 0), huc_def(kbl, 4, 0, 0)) \ - fw_def(GEMINILAKE, 0, guc_def(glk, 62, 0, 0), huc_def(glk, 4, 0, 0)) \ - fw_def(KABYLAKE, 0, guc_def(kbl, 62, 0, 0), huc_def(kbl, 4, 0, 0)) \ - fw_def(BROXTON, 0, guc_def(bxt, 62, 0, 0), huc_def(bxt, 2, 0, 0)) \ - fw_def(SKYLAKE, 0, guc_def(skl, 62, 0, 0), huc_def(skl, 2, 0, 0)) +#define INTEL_GUC_FIRMWARE_DEFS(fw_def, guc_def) \ + fw_def(ALDERLAKE_P, 0, guc_def(adlp, 62, 0, 3)) \ + fw_def(ALDERLAKE_S, 0, guc_def(tgl, 62, 0, 0)) \ + fw_def(DG1, 0, guc_def(dg1, 62, 0, 0)) \ + fw_def(ROCKETLAKE, 0, guc_def(tgl, 62, 0, 0)) \ + fw_def(TIGERLAKE, 0, guc_def(tgl, 62, 0, 0)) \ + fw_def(JASPERLAKE, 0, guc_def(ehl, 62, 0, 0)) \ + fw_def(ELKHARTLAKE, 0, guc_def(ehl, 62, 0, 0)) \ + fw_def(ICELAKE, 0, guc_def(icl, 62, 0, 0)) \ + fw_def(COMETLAKE, 5, guc_def(cml, 62, 0, 0)) \ + fw_def(COMETLAKE, 0, guc_def(kbl, 62, 0, 0)) \ + fw_def(COFFEELAKE, 0, guc_def(kbl, 62, 0, 0)) \ + fw_def(GEMINILAKE, 0, guc_def(glk, 62, 0, 0)) \ + fw_def(KABYLAKE, 0, guc_def(kbl, 62, 0, 0)) \ + fw_def(BROXTON, 0, guc_def(bxt, 62, 0, 0)) \ + fw_def(SKYLAKE, 0, guc_def(skl, 62, 0, 0)) + +#define INTEL_HUC_FIRMWARE_DEFS(fw_def, huc_def) \ + fw_def(ALDERLAKE_P, 0, huc_def(tgl, 7, 9, 3)) \ + fw_def(ALDERLAKE_S, 0, huc_def(tgl, 7, 9, 3)) \ + fw_def(DG1, 0, huc_def(dg1, 7, 9, 3)) \ + fw_def(ROCKETLAKE, 0, huc_def(tgl, 7, 9, 3)) \ + fw_def(TIGERLAKE, 0, huc_def(tgl, 7, 9, 3)) \ + fw_def(JASPERLAKE, 0, huc_def(ehl, 9, 0, 0)) \ + fw_def(ELKHARTLAKE, 0, huc_def(ehl, 9, 0, 0)) \ + fw_def(ICELAKE, 0, huc_def(icl, 9, 0, 0)) \ + fw_def(COMETLAKE, 5, huc_def(cml, 4, 0, 0)) \ + fw_def(COMETLAKE, 0, huc_def(kbl, 4, 0, 0)) \ + fw_def(COFFEELAKE, 0, huc_def(kbl, 4, 0, 0)) \ + fw_def(GEMINILAKE, 0, huc_def(glk, 4, 0, 0)) \ + fw_def(KABYLAKE, 0, huc_def(kbl, 4, 0, 0)) \ + fw_def(BROXTON, 0, huc_def(bxt, 2, 0, 0)) \ + fw_def(SKYLAKE, 0, huc_def(skl, 2, 0, 0)) #define __MAKE_UC_FW_PATH(prefix_, name_, major_, minor_, patch_) \ "i915/" \ @@ -79,11 +96,11 @@ void intel_uc_fw_change_status(struct intel_uc_fw *uc_fw, __MAKE_UC_FW_PATH(prefix_, "_huc_", major_, minor_, bld_num_) /* All blobs need to be declared via MODULE_FIRMWARE() */ -#define INTEL_UC_MODULE_FW(platform_, revid_, guc_, huc_) \ - MODULE_FIRMWARE(guc_); \ - MODULE_FIRMWARE(huc_); +#define INTEL_UC_MODULE_FW(platform_, revid_, uc_) \ + MODULE_FIRMWARE(uc_); -INTEL_UC_FIRMWARE_DEFS(INTEL_UC_MODULE_FW, MAKE_GUC_FW_PATH, MAKE_HUC_FW_PATH) +INTEL_GUC_FIRMWARE_DEFS(INTEL_UC_MODULE_FW, MAKE_GUC_FW_PATH) +INTEL_HUC_FIRMWARE_DEFS(INTEL_UC_MODULE_FW, MAKE_HUC_FW_PATH) /* The below structs and macros are used to iterate across the list of blobs */ struct __packed uc_fw_blob { @@ -106,31 +123,47 @@ struct __packed uc_fw_blob { struct __packed uc_fw_platform_requirement { enum intel_platform p; u8 rev; /* first platform rev using this FW */ - const struct uc_fw_blob blobs[INTEL_UC_FW_NUM_TYPES]; + const struct uc_fw_blob blob; }; -#define MAKE_FW_LIST(platform_, revid_, guc_, huc_) \ +#define MAKE_FW_LIST(platform_, revid_, uc_) \ { \ .p = INTEL_##platform_, \ .rev = revid_, \ - .blobs[INTEL_UC_FW_TYPE_GUC] = guc_, \ - .blobs[INTEL_UC_FW_TYPE_HUC] = huc_, \ + .blob = uc_, \ }, +struct fw_blobs_by_type { + const struct uc_fw_platform_requirement *blobs; + u32 count; +}; + static void __uc_fw_auto_select(struct drm_i915_private *i915, struct intel_uc_fw *uc_fw) { - static const struct uc_fw_platform_requirement fw_blobs[] = { - INTEL_UC_FIRMWARE_DEFS(MAKE_FW_LIST, GUC_FW_BLOB, HUC_FW_BLOB) + static const struct uc_fw_platform_requirement blobs_guc[] = { + INTEL_GUC_FIRMWARE_DEFS(MAKE_FW_LIST, GUC_FW_BLOB) + }; + static const struct uc_fw_platform_requirement blobs_huc[] = { + INTEL_HUC_FIRMWARE_DEFS(MAKE_FW_LIST, HUC_FW_BLOB) }; + static const struct fw_blobs_by_type blobs_all[INTEL_UC_FW_NUM_TYPES] = { + [INTEL_UC_FW_TYPE_GUC] = { blobs_guc, ARRAY_SIZE(blobs_guc) }, + [INTEL_UC_FW_TYPE_HUC] = { blobs_huc, ARRAY_SIZE(blobs_huc) }, + }; + static const struct uc_fw_platform_requirement *fw_blobs; enum intel_platform p = INTEL_INFO(i915)->platform; + u32 fw_count; u8 rev = INTEL_REVID(i915); int i; - for (i = 0; i < ARRAY_SIZE(fw_blobs) && p <= fw_blobs[i].p; i++) { + GEM_BUG_ON(uc_fw->type >= ARRAY_SIZE(blobs_all)); + fw_blobs = blobs_all[uc_fw->type].blobs; + fw_count = blobs_all[uc_fw->type].count; + + for (i = 0; i < fw_count && p <= fw_blobs[i].p; i++) { if (p == fw_blobs[i].p && rev >= fw_blobs[i].rev) { - const struct uc_fw_blob *blob = - &fw_blobs[i].blobs[uc_fw->type]; + const struct uc_fw_blob *blob = &fw_blobs[i].blob; uc_fw->path = blob->path; uc_fw->major_ver_wanted = blob->major; uc_fw->minor_ver_wanted = blob->minor; @@ -140,7 +173,7 @@ __uc_fw_auto_select(struct drm_i915_private *i915, struct intel_uc_fw *uc_fw) /* make sure the list is ordered as expected */ if (IS_ENABLED(CONFIG_DRM_I915_SELFTEST)) { - for (i = 1; i < ARRAY_SIZE(fw_blobs); i++) { + for (i = 1; i < fw_count; i++) { if (fw_blobs[i].p < fw_blobs[i - 1].p) continue; @@ -322,13 +355,6 @@ int intel_uc_fw_fetch(struct intel_uc_fw *uc_fw) uc_fw->ucode_size = (css->size_dw - css->header_size_dw) * sizeof(u32); /* now RSA */ - if (unlikely(css->key_size_dw != UOS_RSA_SCRATCH_COUNT)) { - drm_warn(&i915->drm, "%s firmware %s: unexpected key size: %u != %u\n", - intel_uc_fw_type_repr(uc_fw->type), uc_fw->path, - css->key_size_dw, UOS_RSA_SCRATCH_COUNT); - err = -EPROTO; - goto fail; - } uc_fw->rsa_size = css->key_size_dw * sizeof(u32); /* At least, it should have header, uCode and RSA. Size of all three. */ @@ -540,10 +566,79 @@ fail: i915_probe_error(gt->i915, "Failed to load %s firmware %s (%d)\n", intel_uc_fw_type_repr(uc_fw->type), uc_fw->path, err); - intel_uc_fw_change_status(uc_fw, INTEL_UC_FIRMWARE_FAIL); + intel_uc_fw_change_status(uc_fw, INTEL_UC_FIRMWARE_LOAD_FAIL); return err; } +static inline bool uc_fw_need_rsa_in_memory(struct intel_uc_fw *uc_fw) +{ + /* + * The HW reads the GuC RSA from memory if the key size is > 256 bytes, + * while it reads it from the 64 RSA registers if it is smaller. + * The HuC RSA is always read from memory. + */ + return uc_fw->type == INTEL_UC_FW_TYPE_HUC || uc_fw->rsa_size > 256; +} + +static int uc_fw_rsa_data_create(struct intel_uc_fw *uc_fw) +{ + struct intel_gt *gt = __uc_fw_to_gt(uc_fw); + struct i915_vma *vma; + size_t copied; + void *vaddr; + int err; + + err = i915_inject_probe_error(gt->i915, -ENXIO); + if (err) + return err; + + if (!uc_fw_need_rsa_in_memory(uc_fw)) + return 0; + + /* + * uC firmwares will sit above GUC_GGTT_TOP and will not map through + * GGTT. Unfortunately, this means that the GuC HW cannot perform the uC + * authentication from memory, as the RSA offset now falls within the + * GuC inaccessible range. We resort to perma-pinning an additional vma + * within the accessible range that only contains the RSA signature. + * The GuC HW can use this extra pinning to perform the authentication + * since its GGTT offset will be GuC accessible. + */ + GEM_BUG_ON(uc_fw->rsa_size > PAGE_SIZE); + vma = intel_guc_allocate_vma(>->uc.guc, PAGE_SIZE); + if (IS_ERR(vma)) + return PTR_ERR(vma); + + vaddr = i915_gem_object_pin_map_unlocked(vma->obj, + i915_coherent_map_type(gt->i915, vma->obj, true)); + if (IS_ERR(vaddr)) { + i915_vma_unpin_and_release(&vma, 0); + err = PTR_ERR(vaddr); + goto unpin_out; + } + + copied = intel_uc_fw_copy_rsa(uc_fw, vaddr, vma->size); + i915_gem_object_unpin_map(vma->obj); + + if (copied < uc_fw->rsa_size) { + err = -ENOMEM; + goto unpin_out; + } + + uc_fw->rsa_data = vma; + + return 0; + +unpin_out: + i915_vma_unpin_and_release(&vma, 0); + return err; +} + +static void uc_fw_rsa_data_destroy(struct intel_uc_fw *uc_fw) +{ + i915_vma_unpin_and_release(&uc_fw->rsa_data, 0); +} + int intel_uc_fw_init(struct intel_uc_fw *uc_fw) { int err; @@ -558,14 +653,29 @@ int intel_uc_fw_init(struct intel_uc_fw *uc_fw) if (err) { DRM_DEBUG_DRIVER("%s fw pin-pages err=%d\n", intel_uc_fw_type_repr(uc_fw->type), err); - intel_uc_fw_change_status(uc_fw, INTEL_UC_FIRMWARE_FAIL); + goto out; } + err = uc_fw_rsa_data_create(uc_fw); + if (err) { + DRM_DEBUG_DRIVER("%s fw rsa data creation failed, err=%d\n", + intel_uc_fw_type_repr(uc_fw->type), err); + goto out_unpin; + } + + return 0; + +out_unpin: + i915_gem_object_unpin_pages(uc_fw->obj); +out: + intel_uc_fw_change_status(uc_fw, INTEL_UC_FIRMWARE_INIT_FAIL); return err; } void intel_uc_fw_fini(struct intel_uc_fw *uc_fw) { + uc_fw_rsa_data_destroy(uc_fw); + if (i915_gem_object_has_pinned_pages(uc_fw->obj)) i915_gem_object_unpin_pages(uc_fw->obj); diff --git a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.h b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.h index 1e00bf65639e..d9d1dc0b4cbb 100644 --- a/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.h +++ b/drivers/gpu/drm/i915/gt/uc/intel_uc_fw.h @@ -32,11 +32,12 @@ struct intel_gt; * | | MISSING <--/ | \--> ERROR | * | fetch | V | * | | AVAILABLE | - * +------------+- | -+ + * +------------+- | \ -+ + * | | | \--> INIT FAIL | * | init | V | * | | /------> LOADABLE <----<-----------\ | * +------------+- \ / \ \ \ -+ - * | | FAIL <--< \--> TRANSFERRED \ | + * | | LOAD FAIL <--< \--> TRANSFERRED \ | * | upload | \ / \ / | * | | \---------/ \--> RUNNING | * +------------+---------------------------------------------------+ @@ -50,8 +51,9 @@ enum intel_uc_fw_status { INTEL_UC_FIRMWARE_MISSING, /* blob not found on the system */ INTEL_UC_FIRMWARE_ERROR, /* invalid format or version */ INTEL_UC_FIRMWARE_AVAILABLE, /* blob found and copied in mem */ + INTEL_UC_FIRMWARE_INIT_FAIL, /* failed to prepare fw objects for load */ INTEL_UC_FIRMWARE_LOADABLE, /* all fw-required objects are ready */ - INTEL_UC_FIRMWARE_FAIL, /* failed to xfer or init/auth the fw */ + INTEL_UC_FIRMWARE_LOAD_FAIL, /* failed to xfer or init/auth the fw */ INTEL_UC_FIRMWARE_TRANSFERRED, /* dma xfer done */ INTEL_UC_FIRMWARE_RUNNING /* init/auth done */ }; @@ -84,6 +86,7 @@ struct intel_uc_fw { * or during a GT reset (mutex guarantees single threaded). */ struct i915_vma dummy; + struct i915_vma *rsa_data; /* * The firmware build process will generate a version header file with major and @@ -130,10 +133,12 @@ const char *intel_uc_fw_status_repr(enum intel_uc_fw_status status) return "ERROR"; case INTEL_UC_FIRMWARE_AVAILABLE: return "AVAILABLE"; + case INTEL_UC_FIRMWARE_INIT_FAIL: + return "INIT FAIL"; case INTEL_UC_FIRMWARE_LOADABLE: return "LOADABLE"; - case INTEL_UC_FIRMWARE_FAIL: - return "FAIL"; + case INTEL_UC_FIRMWARE_LOAD_FAIL: + return "LOAD FAIL"; case INTEL_UC_FIRMWARE_TRANSFERRED: return "TRANSFERRED"; case INTEL_UC_FIRMWARE_RUNNING: @@ -155,7 +160,8 @@ static inline int intel_uc_fw_status_to_error(enum intel_uc_fw_status status) return -ENOENT; case INTEL_UC_FIRMWARE_ERROR: return -ENOEXEC; - case INTEL_UC_FIRMWARE_FAIL: + case INTEL_UC_FIRMWARE_INIT_FAIL: + case INTEL_UC_FIRMWARE_LOAD_FAIL: return -EIO; case INTEL_UC_FIRMWARE_SELECTED: return -ESTALE; diff --git a/drivers/gpu/drm/i915/gt/uc/selftest_guc.c b/drivers/gpu/drm/i915/gt/uc/selftest_guc.c index fb0e4a7bd8ca..d3327b802b76 100644 --- a/drivers/gpu/drm/i915/gt/uc/selftest_guc.c +++ b/drivers/gpu/drm/i915/gt/uc/selftest_guc.c @@ -3,8 +3,21 @@ * Copyright �� 2021 Intel Corporation */ +#include "selftests/igt_spinner.h" #include "selftests/intel_scheduler_helpers.h" +static int request_add_spin(struct i915_request *rq, struct igt_spinner *spin) +{ + int err = 0; + + i915_request_get(rq); + i915_request_add(rq); + if (spin && !igt_wait_for_spinner(spin, rq)) + err = -ETIMEDOUT; + + return err; +} + static struct i915_request *nop_user_request(struct intel_context *ce, struct i915_request *from) { @@ -110,12 +123,172 @@ err: return ret; } +/* + * intel_guc_steal_guc_ids - Test to exhaust all guc_ids and then steal one + * + * This test creates a spinner which is used to block all subsequent submissions + * until it completes. Next, a loop creates a context and a NOP request each + * iteration until the guc_ids are exhausted (request creation returns -EAGAIN). + * The spinner is ended, unblocking all requests created in the loop. At this + * point all guc_ids are exhausted but are available to steal. Try to create + * another request which should successfully steal a guc_id. Wait on last + * request to complete, idle GPU, verify a guc_id was stolen via a counter, and + * exit the test. Test also artificially reduces the number of guc_ids so the + * test runs in a timely manner. + */ +static int intel_guc_steal_guc_ids(void *arg) +{ + struct intel_gt *gt = arg; + struct intel_guc *guc = >->uc.guc; + int ret, sv, context_index = 0; + intel_wakeref_t wakeref; + struct intel_engine_cs *engine; + struct intel_context **ce; + struct igt_spinner spin; + struct i915_request *spin_rq = NULL, *rq, *last = NULL; + int number_guc_id_stolen = guc->number_guc_id_stolen; + + ce = kzalloc(sizeof(*ce) * GUC_MAX_LRC_DESCRIPTORS, GFP_KERNEL); + if (!ce) { + pr_err("Context array allocation failed\n"); + return -ENOMEM; + } + + wakeref = intel_runtime_pm_get(gt->uncore->rpm); + engine = intel_selftest_find_any_engine(gt); + sv = guc->submission_state.num_guc_ids; + guc->submission_state.num_guc_ids = 4096; + + /* Create spinner to block requests in below loop */ + ce[context_index] = intel_context_create(engine); + if (IS_ERR(ce[context_index])) { + ret = PTR_ERR(ce[context_index]); + ce[context_index] = NULL; + pr_err("Failed to create context: %d\n", ret); + goto err_wakeref; + } + ret = igt_spinner_init(&spin, engine->gt); + if (ret) { + pr_err("Failed to create spinner: %d\n", ret); + goto err_contexts; + } + spin_rq = igt_spinner_create_request(&spin, ce[context_index], + MI_ARB_CHECK); + if (IS_ERR(spin_rq)) { + ret = PTR_ERR(spin_rq); + pr_err("Failed to create spinner request: %d\n", ret); + goto err_contexts; + } + ret = request_add_spin(spin_rq, &spin); + if (ret) { + pr_err("Failed to add Spinner request: %d\n", ret); + goto err_spin_rq; + } + + /* Use all guc_ids */ + while (ret != -EAGAIN) { + ce[++context_index] = intel_context_create(engine); + if (IS_ERR(ce[context_index])) { + ret = PTR_ERR(ce[context_index--]); + ce[context_index] = NULL; + pr_err("Failed to create context: %d\n", ret); + goto err_spin_rq; + } + + rq = nop_user_request(ce[context_index], spin_rq); + if (IS_ERR(rq)) { + ret = PTR_ERR(rq); + rq = NULL; + if (ret != -EAGAIN) { + pr_err("Failed to create request, %d: %d\n", + context_index, ret); + goto err_spin_rq; + } + } else { + if (last) + i915_request_put(last); + last = rq; + } + } + + /* Release blocked requests */ + igt_spinner_end(&spin); + ret = intel_selftest_wait_for_rq(spin_rq); + if (ret) { + pr_err("Spin request failed to complete: %d\n", ret); + i915_request_put(last); + goto err_spin_rq; + } + i915_request_put(spin_rq); + igt_spinner_fini(&spin); + spin_rq = NULL; + + /* Wait for last request */ + ret = i915_request_wait(last, 0, HZ * 30); + i915_request_put(last); + if (ret < 0) { + pr_err("Last request failed to complete: %d\n", ret); + goto err_spin_rq; + } + + /* Try to steal guc_id */ + rq = nop_user_request(ce[context_index], NULL); + if (IS_ERR(rq)) { + ret = PTR_ERR(rq); + pr_err("Failed to steal guc_id, %d: %d\n", context_index, ret); + goto err_spin_rq; + } + + /* Wait for request with stolen guc_id */ + ret = i915_request_wait(rq, 0, HZ); + i915_request_put(rq); + if (ret < 0) { + pr_err("Request with stolen guc_id failed to complete: %d\n", + ret); + goto err_spin_rq; + } + + /* Wait for idle */ + ret = intel_gt_wait_for_idle(gt, HZ * 30); + if (ret < 0) { + pr_err("GT failed to idle: %d\n", ret); + goto err_spin_rq; + } + + /* Verify a guc_id was stolen */ + if (guc->number_guc_id_stolen == number_guc_id_stolen) { + pr_err("No guc_id was stolen"); + ret = -EINVAL; + } else { + ret = 0; + } + +err_spin_rq: + if (spin_rq) { + igt_spinner_end(&spin); + intel_selftest_wait_for_rq(spin_rq); + i915_request_put(spin_rq); + igt_spinner_fini(&spin); + intel_gt_wait_for_idle(gt, HZ * 30); + } +err_contexts: + for (; context_index >= 0 && ce[context_index]; --context_index) + intel_context_put(ce[context_index]); +err_wakeref: + intel_runtime_pm_put(gt->uncore->rpm, wakeref); + kfree(ce); + guc->submission_state.num_guc_ids = sv; + + return ret; +} + int intel_guc_live_selftests(struct drm_i915_private *i915) { static const struct i915_subtest tests[] = { SUBTEST(intel_guc_scrub_ctbs), + SUBTEST(intel_guc_steal_guc_ids), }; - struct intel_gt *gt = &i915->gt; + struct intel_gt *gt = to_gt(i915); if (intel_gt_is_wedged(gt)) return 0; diff --git a/drivers/gpu/drm/i915/gt/uc/selftest_guc_multi_lrc.c b/drivers/gpu/drm/i915/gt/uc/selftest_guc_multi_lrc.c index 50953c8e8b53..1297ddbf7f88 100644 --- a/drivers/gpu/drm/i915/gt/uc/selftest_guc_multi_lrc.c +++ b/drivers/gpu/drm/i915/gt/uc/selftest_guc_multi_lrc.c @@ -167,7 +167,7 @@ int intel_guc_multi_lrc_live_selftests(struct drm_i915_private *i915) static const struct i915_subtest tests[] = { SUBTEST(intel_guc_multi_lrc_basic), }; - struct intel_gt *gt = &i915->gt; + struct intel_gt *gt = to_gt(i915); if (intel_gt_is_wedged(gt)) return 0; diff --git a/drivers/gpu/drm/i915/gvt/gvt.c b/drivers/gpu/drm/i915/gvt/gvt.c index cbac409f6c8a..f0b69e4dcb52 100644 --- a/drivers/gpu/drm/i915/gvt/gvt.c +++ b/drivers/gpu/drm/i915/gvt/gvt.c @@ -205,7 +205,7 @@ int intel_gvt_init_device(struct drm_i915_private *i915) spin_lock_init(&gvt->scheduler.mmio_context_lock); mutex_init(&gvt->lock); mutex_init(&gvt->sched_lock); - gvt->gt = &i915->gt; + gvt->gt = to_gt(i915); i915->gvt = gvt; init_device_info(gvt); diff --git a/drivers/gpu/drm/i915/gvt/scheduler.c b/drivers/gpu/drm/i915/gvt/scheduler.c index 6c804102528b..42a0c9ae0a73 100644 --- a/drivers/gpu/drm/i915/gvt/scheduler.c +++ b/drivers/gpu/drm/i915/gvt/scheduler.c @@ -1386,7 +1386,7 @@ int intel_vgpu_setup_submission(struct intel_vgpu *vgpu) enum intel_engine_id i; int ret; - ppgtt = i915_ppgtt_create(&i915->gt, I915_BO_ALLOC_PM_EARLY); + ppgtt = i915_ppgtt_create(to_gt(i915), I915_BO_ALLOC_PM_EARLY); if (IS_ERR(ppgtt)) return PTR_ERR(ppgtt); diff --git a/drivers/gpu/drm/i915/i915_active.c b/drivers/gpu/drm/i915/i915_active.c index 3103c1e1fd14..ee2b3a375362 100644 --- a/drivers/gpu/drm/i915/i915_active.c +++ b/drivers/gpu/drm/i915/i915_active.c @@ -426,8 +426,9 @@ replace_barrier(struct i915_active *ref, struct i915_active_fence *active) return true; } -int i915_active_ref(struct i915_active *ref, u64 idx, struct dma_fence *fence) +int i915_active_add_request(struct i915_active *ref, struct i915_request *rq) { + struct dma_fence *fence = &rq->fence; struct i915_active_fence *active; int err; @@ -436,7 +437,7 @@ int i915_active_ref(struct i915_active *ref, u64 idx, struct dma_fence *fence) if (err) return err; - active = active_instance(ref, idx); + active = active_instance(ref, i915_request_timeline(rq)->fence_context); if (!active) { err = -ENOMEM; goto out; @@ -477,29 +478,6 @@ __i915_active_set_fence(struct i915_active *ref, return prev; } -static struct i915_active_fence * -__active_fence(struct i915_active *ref, u64 idx) -{ - struct active_node *it; - - it = __active_lookup(ref, idx); - if (unlikely(!it)) { /* Contention with parallel tree builders! */ - spin_lock_irq(&ref->tree_lock); - it = __active_lookup(ref, idx); - spin_unlock_irq(&ref->tree_lock); - } - GEM_BUG_ON(!it); /* slot must be preallocated */ - - return &it->base; -} - -struct dma_fence * -__i915_active_ref(struct i915_active *ref, u64 idx, struct dma_fence *fence) -{ - /* Only valid while active, see i915_active_acquire_for_context() */ - return __i915_active_set_fence(ref, __active_fence(ref, idx), fence); -} - struct dma_fence * i915_active_set_exclusive(struct i915_active *ref, struct dma_fence *f) { diff --git a/drivers/gpu/drm/i915/i915_active.h b/drivers/gpu/drm/i915/i915_active.h index 5fcdb0e2bc9e..7eb44132183a 100644 --- a/drivers/gpu/drm/i915/i915_active.h +++ b/drivers/gpu/drm/i915/i915_active.h @@ -164,26 +164,11 @@ void __i915_active_init(struct i915_active *ref, __i915_active_init(ref, active, retire, flags, &__mkey, &__wkey); \ } while (0) -struct dma_fence * -__i915_active_ref(struct i915_active *ref, u64 idx, struct dma_fence *fence); -int i915_active_ref(struct i915_active *ref, u64 idx, struct dma_fence *fence); - -static inline int -i915_active_add_request(struct i915_active *ref, struct i915_request *rq) -{ - return i915_active_ref(ref, - i915_request_timeline(rq)->fence_context, - &rq->fence); -} +int i915_active_add_request(struct i915_active *ref, struct i915_request *rq); struct dma_fence * i915_active_set_exclusive(struct i915_active *ref, struct dma_fence *f); -static inline bool i915_active_has_exclusive(struct i915_active *ref) -{ - return rcu_access_pointer(ref->excl.fence); -} - int __i915_active_wait(struct i915_active *ref, int state); static inline int i915_active_wait(struct i915_active *ref) { diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c index bafb902269de..e0e052cdf8b8 100644 --- a/drivers/gpu/drm/i915/i915_debugfs.c +++ b/drivers/gpu/drm/i915/i915_debugfs.c @@ -65,7 +65,7 @@ static int i915_capabilities(struct seq_file *m, void *data) intel_device_info_print_static(INTEL_INFO(i915), &p); intel_device_info_print_runtime(RUNTIME_INFO(i915), &p); i915_print_iommu_status(i915, &p); - intel_gt_info_print(&i915->gt.info, &p); + intel_gt_info_print(&to_gt(i915)->info, &p); intel_driver_caps_print(&i915->caps, &p); kernel_param_lock(THIS_MODULE); @@ -293,7 +293,7 @@ static int i915_gpu_info_open(struct inode *inode, struct file *file) gpu = NULL; with_intel_runtime_pm(&i915->runtime_pm, wakeref) - gpu = i915_gpu_coredump(&i915->gt, ALL_ENGINES); + gpu = i915_gpu_coredump(to_gt(i915), ALL_ENGINES); if (IS_ERR(gpu)) return PTR_ERR(gpu); @@ -351,7 +351,7 @@ static const struct file_operations i915_error_state_fops = { static int i915_frequency_info(struct seq_file *m, void *unused) { struct drm_i915_private *i915 = node_to_i915(m->private); - struct intel_gt *gt = &i915->gt; + struct intel_gt *gt = to_gt(i915); struct drm_printer p = drm_seq_file_printer(m); intel_gt_pm_frequency_dump(gt, &p); @@ -439,11 +439,11 @@ static int i915_swizzle_info(struct seq_file *m, void *data) static int i915_rps_boost_info(struct seq_file *m, void *data) { struct drm_i915_private *dev_priv = node_to_i915(m->private); - struct intel_rps *rps = &dev_priv->gt.rps; + struct intel_rps *rps = &to_gt(dev_priv)->rps; seq_printf(m, "RPS enabled? %s\n", yesno(intel_rps_is_enabled(rps))); seq_printf(m, "RPS active? %s\n", yesno(intel_rps_is_active(rps))); - seq_printf(m, "GPU busy? %s\n", yesno(dev_priv->gt.awake)); + seq_printf(m, "GPU busy? %s\n", yesno(to_gt(dev_priv)->awake)); seq_printf(m, "Boosts outstanding? %d\n", atomic_read(&rps->num_waiters)); seq_printf(m, "Interactive? %d\n", READ_ONCE(rps->power.interactive)); @@ -476,7 +476,7 @@ static int i915_runtime_pm_status(struct seq_file *m, void *unused) seq_printf(m, "Runtime power status: %s\n", enableddisabled(!dev_priv->power_domains.init_wakeref)); - seq_printf(m, "GPU idle: %s\n", yesno(!dev_priv->gt.awake)); + seq_printf(m, "GPU idle: %s\n", yesno(!to_gt(dev_priv)->awake)); seq_printf(m, "IRQs disabled: %s\n", yesno(!intel_irqs_enabled(dev_priv))); #ifdef CONFIG_PM @@ -508,18 +508,18 @@ static int i915_engine_info(struct seq_file *m, void *unused) wakeref = intel_runtime_pm_get(&i915->runtime_pm); seq_printf(m, "GT awake? %s [%d], %llums\n", - yesno(i915->gt.awake), - atomic_read(&i915->gt.wakeref.count), - ktime_to_ms(intel_gt_get_awake_time(&i915->gt))); + yesno(to_gt(i915)->awake), + atomic_read(&to_gt(i915)->wakeref.count), + ktime_to_ms(intel_gt_get_awake_time(to_gt(i915)))); seq_printf(m, "CS timestamp frequency: %u Hz, %d ns\n", - i915->gt.clock_frequency, - i915->gt.clock_period_ns); + to_gt(i915)->clock_frequency, + to_gt(i915)->clock_period_ns); p = drm_seq_file_printer(m); for_each_uabi_engine(engine, i915) intel_engine_dump(engine, &p, "%s\n", engine->name); - intel_gt_show_timelines(&i915->gt, &p, i915_request_show_with_schedule); + intel_gt_show_timelines(to_gt(i915), &p, i915_request_show_with_schedule); intel_runtime_pm_put(&i915->runtime_pm, wakeref); @@ -558,14 +558,14 @@ static int i915_wedged_get(void *data, u64 *val) { struct drm_i915_private *i915 = data; - return intel_gt_debugfs_reset_show(&i915->gt, val); + return intel_gt_debugfs_reset_show(to_gt(i915), val); } static int i915_wedged_set(void *data, u64 val) { struct drm_i915_private *i915 = data; - return intel_gt_debugfs_reset_store(&i915->gt, val); + return intel_gt_debugfs_reset_store(to_gt(i915), val); } DEFINE_SIMPLE_ATTRIBUTE(i915_wedged_fops, @@ -581,7 +581,7 @@ i915_perf_noa_delay_set(void *data, u64 val) * This would lead to infinite waits as we're doing timestamp * difference on the CS with only 32bits. */ - if (intel_gt_ns_to_clock_interval(&i915->gt, val) > U32_MAX) + if (intel_gt_ns_to_clock_interval(to_gt(i915), val) > U32_MAX) return -EINVAL; atomic64_set(&i915->perf.noa_programming_delay, val); @@ -666,16 +666,18 @@ static int i915_drop_caches_set(void *data, u64 val) { struct drm_i915_private *i915 = data; + unsigned int flags; int ret; DRM_DEBUG("Dropping caches: 0x%08llx [0x%08llx]\n", val, val & DROP_ALL); - ret = gt_drop_caches(&i915->gt, val); + ret = gt_drop_caches(to_gt(i915), val); if (ret) return ret; fs_reclaim_acquire(GFP_KERNEL); + flags = memalloc_noreclaim_save(); if (val & DROP_BOUND) i915_gem_shrink(NULL, i915, LONG_MAX, NULL, I915_SHRINK_BOUND); @@ -684,6 +686,7 @@ i915_drop_caches_set(void *data, u64 val) if (val & DROP_SHRINK_ALL) i915_gem_shrink_all(i915); + memalloc_noreclaim_restore(flags); fs_reclaim_release(GFP_KERNEL); if (val & DROP_RCU) @@ -702,7 +705,7 @@ DEFINE_SIMPLE_ATTRIBUTE(i915_drop_caches_fops, static int i915_sseu_status(struct seq_file *m, void *unused) { struct drm_i915_private *i915 = node_to_i915(m->private); - struct intel_gt *gt = &i915->gt; + struct intel_gt *gt = to_gt(i915); return intel_sseu_status(m, gt); } @@ -711,14 +714,14 @@ static int i915_forcewake_open(struct inode *inode, struct file *file) { struct drm_i915_private *i915 = inode->i_private; - return intel_gt_pm_debugfs_forcewake_user_open(&i915->gt); + return intel_gt_pm_debugfs_forcewake_user_open(to_gt(i915)); } static int i915_forcewake_release(struct inode *inode, struct file *file) { struct drm_i915_private *i915 = inode->i_private; - return intel_gt_pm_debugfs_forcewake_user_release(&i915->gt); + return intel_gt_pm_debugfs_forcewake_user_release(to_gt(i915)); } static const struct file_operations i915_forcewake_fops = { diff --git a/drivers/gpu/drm/i915/i915_debugfs_params.c b/drivers/gpu/drm/i915/i915_debugfs_params.c index 20424275d41e..783c8676eee2 100644 --- a/drivers/gpu/drm/i915/i915_debugfs_params.c +++ b/drivers/gpu/drm/i915/i915_debugfs_params.c @@ -40,8 +40,8 @@ static int notify_guc(struct drm_i915_private *i915) { int ret = 0; - if (intel_uc_uses_guc_submission(&i915->gt.uc)) - ret = intel_guc_global_policies_update(&i915->gt.uc.guc); + if (intel_uc_uses_guc_submission(&to_gt(i915)->uc)) + ret = intel_guc_global_policies_update(&to_gt(i915)->uc.guc); return ret; } diff --git a/drivers/gpu/drm/i915/i915_deps.c b/drivers/gpu/drm/i915/i915_deps.c new file mode 100644 index 000000000000..999210b37325 --- /dev/null +++ b/drivers/gpu/drm/i915/i915_deps.c @@ -0,0 +1,237 @@ +// SPDX-License-Identifier: MIT +/* + * Copyright © 2021 Intel Corporation + */ + +#include <linux/dma-fence.h> +#include <linux/slab.h> + +#include <drm/ttm/ttm_bo_api.h> + +#include "i915_deps.h" + +/** + * DOC: Set of utilities to dynamically collect dependencies into a + * structure which is fed into the GT migration code. + * + * Once we can do async unbinding, this is also needed to coalesce + * the migration fence with the unbind fences if these are coalesced + * post-migration. + * + * While collecting the individual dependencies, we store the refcounted + * struct dma_fence pointers in a realloc-managed pointer array, since + * that can be easily fed into a dma_fence_array. Other options are + * available, like for example an xarray for similarity with drm/sched. + * Can be changed easily if needed. + * + * A struct i915_deps need to be initialized using i915_deps_init(). + * If i915_deps_add_dependency() or i915_deps_add_resv() return an + * error code they will internally call i915_deps_fini(), which frees + * all internal references and allocations. + */ + +/* Min number of fence pointers in the array when an allocation occurs. */ +#define I915_DEPS_MIN_ALLOC_CHUNK 8U + +static void i915_deps_reset_fences(struct i915_deps *deps) +{ + if (deps->fences != &deps->single) + kfree(deps->fences); + deps->num_deps = 0; + deps->fences_size = 1; + deps->fences = &deps->single; +} + +/** + * i915_deps_init - Initialize an i915_deps structure + * @deps: Pointer to the i915_deps structure to initialize. + * @gfp: The allocation mode for subsequenst allocations. + */ +void i915_deps_init(struct i915_deps *deps, gfp_t gfp) +{ + deps->fences = NULL; + deps->gfp = gfp; + i915_deps_reset_fences(deps); +} + +/** + * i915_deps_fini - Finalize an i915_deps structure + * @deps: Pointer to the i915_deps structure to finalize. + * + * This function drops all fence references taken, conditionally frees and + * then resets the fences array. + */ +void i915_deps_fini(struct i915_deps *deps) +{ + unsigned int i; + + for (i = 0; i < deps->num_deps; ++i) + dma_fence_put(deps->fences[i]); + + if (deps->fences != &deps->single) + kfree(deps->fences); +} + +static int i915_deps_grow(struct i915_deps *deps, struct dma_fence *fence, + const struct ttm_operation_ctx *ctx) +{ + int ret; + + if (deps->num_deps >= deps->fences_size) { + unsigned int new_size = 2 * deps->fences_size; + struct dma_fence **new_fences; + + new_size = max(new_size, I915_DEPS_MIN_ALLOC_CHUNK); + new_fences = kmalloc_array(new_size, sizeof(*new_fences), deps->gfp); + if (!new_fences) + goto sync; + + memcpy(new_fences, deps->fences, + deps->fences_size * sizeof(*new_fences)); + swap(new_fences, deps->fences); + if (new_fences != &deps->single) + kfree(new_fences); + deps->fences_size = new_size; + } + deps->fences[deps->num_deps++] = dma_fence_get(fence); + return 0; + +sync: + if (ctx->no_wait_gpu && !dma_fence_is_signaled(fence)) { + ret = -EBUSY; + goto unref; + } + + ret = dma_fence_wait(fence, ctx->interruptible); + if (ret) + goto unref; + + ret = fence->error; + if (ret) + goto unref; + + return 0; + +unref: + i915_deps_fini(deps); + return ret; +} + +/** + * i915_deps_sync - Wait for all the fences in the dependency collection + * @deps: Pointer to the i915_deps structure the fences of which to wait for. + * @ctx: Pointer to a struct ttm_operation_ctx indicating how the waits + * should be performed. + * + * This function waits for fences in the dependency collection. If it + * encounters an error during the wait or a fence error, the wait for + * further fences is aborted and the error returned. + * + * Return: Zero if successful, Negative error code on error. + */ +int i915_deps_sync(const struct i915_deps *deps, const struct ttm_operation_ctx *ctx) +{ + struct dma_fence **fences = deps->fences; + unsigned int i; + int ret = 0; + + for (i = 0; i < deps->num_deps; ++i, ++fences) { + if (ctx->no_wait_gpu && !dma_fence_is_signaled(*fences)) { + ret = -EBUSY; + break; + } + + ret = dma_fence_wait(*fences, ctx->interruptible); + if (!ret) + ret = (*fences)->error; + if (ret) + break; + } + + return ret; +} + +/** + * i915_deps_add_dependency - Add a fence to the dependency collection + * @deps: Pointer to the i915_deps structure a fence is to be added to. + * @fence: The fence to add. + * @ctx: Pointer to a struct ttm_operation_ctx indicating how waits are to + * be performed if waiting. + * + * Adds a fence to the dependency collection, and takes a reference on it. + * If the fence context is not zero and there was a later fence from the + * same fence context already added, then the fence is not added to the + * dependency collection. If the fence context is not zero and there was + * an earlier fence already added, then the fence will replace the older + * fence from the same context and the reference on the earlier fence will + * be dropped. + * If there is a failure to allocate memory to accommodate the new fence to + * be added, the new fence will instead be waited for and an error may + * be returned; depending on the value of @ctx, or if there was a fence + * error. If an error was returned, the dependency collection will be + * finalized and all fence reference dropped. + * + * Return: 0 if success. Negative error code on error. + */ +int i915_deps_add_dependency(struct i915_deps *deps, + struct dma_fence *fence, + const struct ttm_operation_ctx *ctx) +{ + unsigned int i; + int ret; + + if (!fence) + return 0; + + if (dma_fence_is_signaled(fence)) { + ret = fence->error; + if (ret) + i915_deps_fini(deps); + return ret; + } + + for (i = 0; i < deps->num_deps; ++i) { + struct dma_fence *entry = deps->fences[i]; + + if (!entry->context || entry->context != fence->context) + continue; + + if (dma_fence_is_later(fence, entry)) { + dma_fence_put(entry); + deps->fences[i] = dma_fence_get(fence); + } + + return 0; + } + + return i915_deps_grow(deps, fence, ctx); +} + +/** + * i915_deps_add_resv - Add the fences of a reservation object to a dependency + * collection. + * @deps: Pointer to the i915_deps structure a fence is to be added to. + * @resv: The reservation object, then fences of which to add. + * @ctx: Pointer to a struct ttm_operation_ctx indicating how waits are to + * be performed if waiting. + * + * Calls i915_deps_add_depencency() on the indicated fences of @resv. + * + * Return: Zero on success. Negative error code on error. + */ +int i915_deps_add_resv(struct i915_deps *deps, struct dma_resv *resv, + const struct ttm_operation_ctx *ctx) +{ + struct dma_resv_iter iter; + struct dma_fence *fence; + + dma_resv_assert_held(resv); + dma_resv_for_each_fence(&iter, resv, true, fence) { + int ret = i915_deps_add_dependency(deps, fence, ctx); + + if (ret) + return ret; + } + + return 0; +} diff --git a/drivers/gpu/drm/i915/i915_deps.h b/drivers/gpu/drm/i915/i915_deps.h new file mode 100644 index 000000000000..d76c0106c910 --- /dev/null +++ b/drivers/gpu/drm/i915/i915_deps.h @@ -0,0 +1,45 @@ +/* SPDX-License-Identifier: MIT */ +/* + * Copyright © 2021 Intel Corporation + */ + +#ifndef _I915_DEPS_H_ +#define _I915_DEPS_H_ + +#include <linux/types.h> + +struct ttm_operation_ctx; +struct dma_fence; +struct dma_resv; + +/** + * struct i915_deps - Collect dependencies into a single dma-fence + * @single: Storage for pointer if the collection is a single fence. + * @fences: Allocated array of fence pointers if more than a single fence; + * otherwise points to the address of @single. + * @num_deps: Current number of dependency fences. + * @fences_size: Size of the @fences array in number of pointers. + * @gfp: Allocation mode. + */ +struct i915_deps { + struct dma_fence *single; + struct dma_fence **fences; + unsigned int num_deps; + unsigned int fences_size; + gfp_t gfp; +}; + +void i915_deps_init(struct i915_deps *deps, gfp_t gfp); + +void i915_deps_fini(struct i915_deps *deps); + +int i915_deps_add_dependency(struct i915_deps *deps, + struct dma_fence *fence, + const struct ttm_operation_ctx *ctx); + +int i915_deps_add_resv(struct i915_deps *deps, struct dma_resv *resv, + const struct ttm_operation_ctx *ctx); + +int i915_deps_sync(const struct i915_deps *deps, + const struct ttm_operation_ctx *ctx); +#endif diff --git a/drivers/gpu/drm/i915/i915_driver.c b/drivers/gpu/drm/i915/i915_driver.c index e9125f14b3d1..95174938b160 100644 --- a/drivers/gpu/drm/i915/i915_driver.c +++ b/drivers/gpu/drm/i915/i915_driver.c @@ -291,7 +291,7 @@ static void intel_detect_preproduction_hw(struct drm_i915_private *dev_priv) static void sanitize_gpu(struct drm_i915_private *i915) { if (!INTEL_INFO(i915)->gpu_reset_clobbers_display) - __intel_gt_reset(&i915->gt, ALL_ENGINES); + __intel_gt_reset(to_gt(i915), ALL_ENGINES); } /** @@ -314,8 +314,9 @@ static int i915_driver_early_probe(struct drm_i915_private *dev_priv) intel_device_info_subplatform_init(dev_priv); intel_step_init(dev_priv); + intel_gt_init_early(to_gt(dev_priv), dev_priv); intel_uncore_mmio_debug_init_early(&dev_priv->mmio_debug); - intel_uncore_init_early(&dev_priv->uncore, dev_priv); + intel_uncore_init_early(&dev_priv->uncore, to_gt(dev_priv)); spin_lock_init(&dev_priv->irq_lock); spin_lock_init(&dev_priv->gpu_error.lock); @@ -346,7 +347,7 @@ static int i915_driver_early_probe(struct drm_i915_private *dev_priv) intel_wopcm_init_early(&dev_priv->wopcm); - intel_gt_init_early(&dev_priv->gt, dev_priv); + __intel_gt_init_early(to_gt(dev_priv), dev_priv); i915_gem_init_early(dev_priv); @@ -367,7 +368,7 @@ static int i915_driver_early_probe(struct drm_i915_private *dev_priv) err_gem: i915_gem_cleanup_early(dev_priv); - intel_gt_driver_late_release(&dev_priv->gt); + intel_gt_driver_late_release(to_gt(dev_priv)); intel_region_ttm_device_fini(dev_priv); err_ttm: vlv_suspend_cleanup(dev_priv); @@ -386,7 +387,7 @@ static void i915_driver_late_release(struct drm_i915_private *dev_priv) intel_irq_fini(dev_priv); intel_power_domains_cleanup(dev_priv); i915_gem_cleanup_early(dev_priv); - intel_gt_driver_late_release(&dev_priv->gt); + intel_gt_driver_late_release(to_gt(dev_priv)); intel_region_ttm_device_fini(dev_priv); vlv_suspend_cleanup(dev_priv); i915_workqueues_cleanup(dev_priv); @@ -429,7 +430,7 @@ static int i915_driver_mmio_probe(struct drm_i915_private *dev_priv) intel_setup_mchbar(dev_priv); intel_device_info_runtime_init(dev_priv); - ret = intel_gt_init_mmio(&dev_priv->gt); + ret = intel_gt_init_mmio(to_gt(dev_priv)); if (ret) goto err_uncore; @@ -586,9 +587,9 @@ static int i915_driver_hw_probe(struct drm_i915_private *dev_priv) if (ret) goto err_ggtt; - intel_gt_init_hw_early(&dev_priv->gt, &dev_priv->ggtt); + intel_gt_init_hw_early(to_gt(dev_priv), &dev_priv->ggtt); - ret = intel_gt_probe_lmem(&dev_priv->gt); + ret = intel_gt_probe_lmem(to_gt(dev_priv)); if (ret) goto err_mem_regions; @@ -701,7 +702,7 @@ static void i915_driver_register(struct drm_i915_private *dev_priv) /* Depends on sysfs having been initialized */ i915_perf_register(dev_priv); - intel_gt_driver_register(&dev_priv->gt); + intel_gt_driver_register(to_gt(dev_priv)); intel_display_driver_register(dev_priv); @@ -729,7 +730,7 @@ static void i915_driver_unregister(struct drm_i915_private *dev_priv) intel_display_driver_unregister(dev_priv); - intel_gt_driver_unregister(&dev_priv->gt); + intel_gt_driver_unregister(to_gt(dev_priv)); i915_perf_unregister(dev_priv); i915_pmu_unregister(dev_priv); @@ -762,7 +763,7 @@ static void i915_welcome_messages(struct drm_i915_private *dev_priv) intel_device_info_print_static(INTEL_INFO(dev_priv), &p); intel_device_info_print_runtime(RUNTIME_INFO(dev_priv), &p); i915_print_iommu_status(dev_priv, &p); - intel_gt_info_print(&dev_priv->gt.info, &p); + intel_gt_info_print(&to_gt(dev_priv)->info, &p); } if (IS_ENABLED(CONFIG_DRM_I915_DEBUG)) @@ -1384,7 +1385,7 @@ static int i915_drm_resume_early(struct drm_device *dev) intel_uncore_resume_early(&dev_priv->uncore); - intel_gt_check_and_clear_faults(&dev_priv->gt); + intel_gt_check_and_clear_faults(to_gt(dev_priv)); intel_display_power_resume_early(dev_priv); @@ -1567,7 +1568,7 @@ static int intel_runtime_suspend(struct device *kdev) */ i915_gem_runtime_suspend(dev_priv); - intel_gt_runtime_suspend(&dev_priv->gt); + intel_gt_runtime_suspend(to_gt(dev_priv)); intel_runtime_pm_disable_interrupts(dev_priv); @@ -1583,7 +1584,7 @@ static int intel_runtime_suspend(struct device *kdev) intel_runtime_pm_enable_interrupts(dev_priv); - intel_gt_runtime_resume(&dev_priv->gt); + intel_gt_runtime_resume(to_gt(dev_priv)); enable_rpm_wakeref_asserts(rpm); @@ -1671,7 +1672,7 @@ static int intel_runtime_resume(struct device *kdev) * No point of rolling back things in case of an error, as the best * we can do is to hope that things will still work (and disable RPM). */ - intel_gt_runtime_resume(&dev_priv->gt); + intel_gt_runtime_resume(to_gt(dev_priv)); /* * On VLV/CHV display interrupts are part of the display diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index 58aa55bc7461..0c70ab08fc0c 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -473,7 +473,7 @@ struct i915_gem_mm { * List of objects which are pending destruction. */ struct llist_head free_list; - struct work_struct free_work; + struct delayed_work free_work; /** * Count of objects pending destructions. Used to skip needlessly * waiting on an RCU barrier if no objects are waiting to be freed. @@ -1005,7 +1005,7 @@ struct drm_i915_private { struct i915_perf perf; /* Abstract the submission mechanism (legacy ringbuffer or execlists) away */ - struct intel_gt gt; + struct intel_gt gt0; struct { struct i915_gem_contexts { @@ -1077,6 +1077,11 @@ static inline struct drm_i915_private *pdev_to_i915(struct pci_dev *pdev) return pci_get_drvdata(pdev); } +static inline struct intel_gt *to_gt(struct drm_i915_private *i915) +{ + return &i915->gt0; +} + /* Simple iterator over all initialised engines */ #define for_each_engine(engine__, dev_priv__, id__) \ for ((id__) = 0; \ @@ -1527,6 +1532,14 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915, #define HAS_MSLICES(dev_priv) \ (INTEL_INFO(dev_priv)->has_mslices) +/* + * Set this flag, when platform requires 64K GTT page sizes or larger for + * device local memory access. Also this flag implies that we require or + * at least support the compact PT layout for the ppGTT when using the 64K + * GTT pages. + */ +#define HAS_64K_PAGES(dev_priv) (INTEL_INFO(dev_priv)->has_64k_pages) + #define HAS_IPC(dev_priv) (INTEL_INFO(dev_priv)->display.has_ipc) #define HAS_REGION(i915, i) (INTEL_INFO(i915)->memory_regions & (i)) @@ -1540,7 +1553,7 @@ IS_SUBPLATFORM(const struct drm_i915_private *i915, #define HAS_PXP(dev_priv) ((IS_ENABLED(CONFIG_DRM_I915_PXP) && \ INTEL_INFO(dev_priv)->has_pxp) && \ - VDBOX_MASK(&dev_priv->gt)) + VDBOX_MASK(to_gt(dev_priv))) #define HAS_GMCH(dev_priv) (INTEL_INFO(dev_priv)->display.has_gmch) @@ -1623,7 +1636,7 @@ static inline void i915_gem_drain_freed_objects(struct drm_i915_private *i915) * armed the work again. */ while (atomic_read(&i915->mm.free_count)) { - flush_work(&i915->mm.free_work); + flush_delayed_work(&i915->mm.free_work); flush_delayed_work(&i915->bdev.wq); rcu_barrier(); } @@ -1657,13 +1670,10 @@ i915_gem_object_ggtt_pin_ww(struct drm_i915_gem_object *obj, const struct i915_ggtt_view *view, u64 size, u64 alignment, u64 flags); -static inline struct i915_vma * __must_check +struct i915_vma * __must_check i915_gem_object_ggtt_pin(struct drm_i915_gem_object *obj, const struct i915_ggtt_view *view, - u64 size, u64 alignment, u64 flags) -{ - return i915_gem_object_ggtt_pin_ww(obj, NULL, view, size, alignment, flags); -} + u64 size, u64 alignment, u64 flags); int i915_gem_object_unbind(struct drm_i915_gem_object *obj, unsigned long flags); diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index 527228d4da7e..915bf431f320 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -877,6 +877,8 @@ i915_gem_object_ggtt_pin_ww(struct drm_i915_gem_object *obj, struct i915_vma *vma; int ret; + GEM_WARN_ON(!ww); + if (flags & PIN_MAPPABLE && (!view || view->type == I915_GGTT_VIEW_NORMAL)) { /* @@ -936,10 +938,7 @@ new_vma: return ERR_PTR(ret); } - if (ww) - ret = i915_vma_pin_ww(vma, ww, size, alignment, flags | PIN_GLOBAL); - else - ret = i915_vma_pin(vma, size, alignment, flags | PIN_GLOBAL); + ret = i915_vma_pin_ww(vma, ww, size, alignment, flags | PIN_GLOBAL); if (ret) return ERR_PTR(ret); @@ -959,6 +958,29 @@ new_vma: return vma; } +struct i915_vma * __must_check +i915_gem_object_ggtt_pin(struct drm_i915_gem_object *obj, + const struct i915_ggtt_view *view, + u64 size, u64 alignment, u64 flags) +{ + struct i915_gem_ww_ctx ww; + struct i915_vma *ret; + int err; + + for_i915_gem_ww(&ww, err, true) { + err = i915_gem_object_lock(obj, &ww); + if (err) + continue; + + ret = i915_gem_object_ggtt_pin_ww(obj, &ww, view, size, + alignment, flags); + if (IS_ERR(ret)) + err = PTR_ERR(ret); + } + + return err ? ERR_PTR(err) : ret; +} + int i915_gem_madvise_ioctl(struct drm_device *dev, void *data, struct drm_file *file_priv) @@ -1049,7 +1071,7 @@ int i915_gem_init(struct drm_i915_private *dev_priv) if (ret) return ret; - intel_uc_fetch_firmwares(&dev_priv->gt.uc); + intel_uc_fetch_firmwares(&to_gt(dev_priv)->uc); intel_wopcm_init(&dev_priv->wopcm); ret = i915_init_ggtt(dev_priv); @@ -1069,7 +1091,7 @@ int i915_gem_init(struct drm_i915_private *dev_priv) */ intel_init_clock_gating(dev_priv); - ret = intel_gt_init(&dev_priv->gt); + ret = intel_gt_init(to_gt(dev_priv)); if (ret) goto err_unlock; @@ -1085,7 +1107,7 @@ err_unlock: i915_gem_drain_workqueue(dev_priv); if (ret != -EIO) - intel_uc_cleanup_firmwares(&dev_priv->gt.uc); + intel_uc_cleanup_firmwares(&to_gt(dev_priv)->uc); if (ret == -EIO) { /* @@ -1093,10 +1115,10 @@ err_unlock: * as wedged. But we only want to do this when the GPU is angry, * for all other failure, such as an allocation failure, bail. */ - if (!intel_gt_is_wedged(&dev_priv->gt)) { + if (!intel_gt_is_wedged(to_gt(dev_priv))) { i915_probe_error(dev_priv, "Failed to initialize GPU, declaring it wedged!\n"); - intel_gt_set_wedged(&dev_priv->gt); + intel_gt_set_wedged(to_gt(dev_priv)); } /* Minimal basic recovery for KMS */ @@ -1127,7 +1149,7 @@ void i915_gem_driver_remove(struct drm_i915_private *dev_priv) intel_wakeref_auto_fini(&dev_priv->ggtt.userfault_wakeref); i915_gem_suspend_late(dev_priv); - intel_gt_driver_remove(&dev_priv->gt); + intel_gt_driver_remove(to_gt(dev_priv)); dev_priv->uabi_engines = RB_ROOT; /* Flush any outstanding unpin_work. */ @@ -1138,9 +1160,9 @@ void i915_gem_driver_remove(struct drm_i915_private *dev_priv) void i915_gem_driver_release(struct drm_i915_private *dev_priv) { - intel_gt_driver_release(&dev_priv->gt); + intel_gt_driver_release(to_gt(dev_priv)); - intel_uc_cleanup_firmwares(&dev_priv->gt.uc); + intel_uc_cleanup_firmwares(&to_gt(dev_priv)->uc); i915_gem_drain_freed_objects(dev_priv); diff --git a/drivers/gpu/drm/i915/i915_getparam.c b/drivers/gpu/drm/i915/i915_getparam.c index 77490cb5ff9c..7f80ad247bc8 100644 --- a/drivers/gpu/drm/i915/i915_getparam.c +++ b/drivers/gpu/drm/i915/i915_getparam.c @@ -13,7 +13,7 @@ int i915_getparam_ioctl(struct drm_device *dev, void *data, { struct drm_i915_private *i915 = to_i915(dev); struct pci_dev *pdev = to_pci_dev(dev->dev); - const struct sseu_dev_info *sseu = &i915->gt.info.sseu; + const struct sseu_dev_info *sseu = &to_gt(i915)->info.sseu; drm_i915_getparam_t *param = data; int value = 0; @@ -82,8 +82,8 @@ int i915_getparam_ioctl(struct drm_device *dev, void *data, break; case I915_PARAM_HAS_GPU_RESET: value = i915->params.enable_hangcheck && - intel_has_gpu_reset(&i915->gt); - if (value && intel_has_reset_engine(&i915->gt)) + intel_has_gpu_reset(to_gt(i915)); + if (value && intel_has_reset_engine(to_gt(i915))) value = 2; break; case I915_PARAM_HAS_RESOURCE_STREAMER: @@ -96,7 +96,7 @@ int i915_getparam_ioctl(struct drm_device *dev, void *data, value = sseu->min_eu_in_pool; break; case I915_PARAM_HUC_STATUS: - value = intel_huc_check_status(&i915->gt.uc.huc); + value = intel_huc_check_status(&to_gt(i915)->uc.huc); if (value < 0) return value; break; @@ -158,7 +158,7 @@ int i915_getparam_ioctl(struct drm_device *dev, void *data, return -ENODEV; break; case I915_PARAM_CS_TIMESTAMP_FREQUENCY: - value = i915->gt.clock_frequency; + value = to_gt(i915)->clock_frequency; break; case I915_PARAM_MMAP_GTT_COHERENT: value = INTEL_INFO(i915)->has_coherent_ggtt; diff --git a/drivers/gpu/drm/i915/i915_gpu_error.c b/drivers/gpu/drm/i915/i915_gpu_error.c index 96d2d99f5b98..5ae812d60abe 100644 --- a/drivers/gpu/drm/i915/i915_gpu_error.c +++ b/drivers/gpu/drm/i915/i915_gpu_error.c @@ -505,7 +505,7 @@ static void error_print_context(struct drm_i915_error_state_buf *m, const char *header, const struct i915_gem_context_coredump *ctx) { - const u32 period = m->i915->gt.clock_period_ns; + const u32 period = to_gt(m->i915)->clock_period_ns; err_printf(m, "%s%s[%d] prio %d, guilty %d active %d, runtime total %lluns, avg %lluns\n", header, ctx->comm, ctx->pid, ctx->sched_attr.priority, @@ -1415,18 +1415,15 @@ static struct i915_vma_coredump * create_vma_coredump(const struct intel_gt *gt, struct i915_vma *vma, const char *name, struct i915_vma_compress *compress) { - struct i915_vma_coredump *ret = NULL; + struct i915_vma_coredump *ret; struct i915_vma_snapshot tmp; - bool lockdep_cookie; if (!vma) return NULL; + GEM_WARN_ON(!i915_vma_is_pinned(vma)); i915_vma_snapshot_init_onstack(&tmp, vma, name); - if (i915_vma_snapshot_resource_pin(&tmp, &lockdep_cookie)) { - ret = i915_vma_coredump_create(gt, &tmp, compress); - i915_vma_snapshot_resource_unpin(&tmp, lockdep_cookie); - } + ret = i915_vma_coredump_create(gt, &tmp, compress); i915_vma_snapshot_put_onstack(&tmp); return ret; @@ -1849,7 +1846,7 @@ i915_gpu_coredump_alloc(struct drm_i915_private *i915, gfp_t gfp) error->time = ktime_get_real(); error->boottime = ktime_get_boottime(); - error->uptime = ktime_sub(ktime_get(), i915->gt.last_init_time); + error->uptime = ktime_sub(ktime_get(), to_gt(i915)->last_init_time); error->capture = jiffies; capture_gen(error); diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c index 5b98fb0532b5..21f75b069fa8 100644 --- a/drivers/gpu/drm/i915/i915_irq.c +++ b/drivers/gpu/drm/i915/i915_irq.c @@ -1040,7 +1040,7 @@ static void ivb_parity_work(struct work_struct *work) { struct drm_i915_private *dev_priv = container_of(work, typeof(*dev_priv), l3_parity.error_work); - struct intel_gt *gt = &dev_priv->gt; + struct intel_gt *gt = to_gt(dev_priv); u32 error_status, row, bank, subbank; char *parity_event[6]; u32 misccpctl; @@ -1718,9 +1718,9 @@ static irqreturn_t valleyview_irq_handler(int irq, void *arg) intel_uncore_write(&dev_priv->uncore, VLV_MASTER_IER, MASTER_INTERRUPT_ENABLE); if (gt_iir) - gen6_gt_irq_handler(&dev_priv->gt, gt_iir); + gen6_gt_irq_handler(to_gt(dev_priv), gt_iir); if (pm_iir) - gen6_rps_irq_handler(&dev_priv->gt.rps, pm_iir); + gen6_rps_irq_handler(&to_gt(dev_priv)->rps, pm_iir); if (hotplug_status) i9xx_hpd_irq_handler(dev_priv, hotplug_status); @@ -1777,7 +1777,7 @@ static irqreturn_t cherryview_irq_handler(int irq, void *arg) ier = intel_uncore_read(&dev_priv->uncore, VLV_IER); intel_uncore_write(&dev_priv->uncore, VLV_IER, 0); - gen8_gt_irq_handler(&dev_priv->gt, master_ctl); + gen8_gt_irq_handler(to_gt(dev_priv), master_ctl); if (iir & I915_DISPLAY_PORT_INTERRUPT) hotplug_status = i9xx_hpd_irq_ack(dev_priv); @@ -2108,7 +2108,7 @@ static void ilk_display_irq_handler(struct drm_i915_private *dev_priv, } if (DISPLAY_VER(dev_priv) == 5 && de_iir & DE_PCU_EVENT) - gen5_rps_irq_handler(&dev_priv->gt.rps); + gen5_rps_irq_handler(&to_gt(dev_priv)->rps); } static void ivb_display_irq_handler(struct drm_i915_private *dev_priv, @@ -2189,9 +2189,9 @@ static irqreturn_t ilk_irq_handler(int irq, void *arg) if (gt_iir) { raw_reg_write(regs, GTIIR, gt_iir); if (GRAPHICS_VER(i915) >= 6) - gen6_gt_irq_handler(&i915->gt, gt_iir); + gen6_gt_irq_handler(to_gt(i915), gt_iir); else - gen5_gt_irq_handler(&i915->gt, gt_iir); + gen5_gt_irq_handler(to_gt(i915), gt_iir); ret = IRQ_HANDLED; } @@ -2209,7 +2209,7 @@ static irqreturn_t ilk_irq_handler(int irq, void *arg) u32 pm_iir = raw_reg_read(regs, GEN6_PMIIR); if (pm_iir) { raw_reg_write(regs, GEN6_PMIIR, pm_iir); - gen6_rps_irq_handler(&i915->gt.rps, pm_iir); + gen6_rps_irq_handler(&to_gt(i915)->rps, pm_iir); ret = IRQ_HANDLED; } } @@ -2635,7 +2635,7 @@ static irqreturn_t gen8_irq_handler(int irq, void *arg) } /* Find, queue (onto bottom-halves), then clear each source */ - gen8_gt_irq_handler(&dev_priv->gt, master_ctl); + gen8_gt_irq_handler(to_gt(dev_priv), master_ctl); /* IRQs are synced during runtime_suspend, we don't require a wakeref */ if (master_ctl & ~GEN8_GT_IRQS) { @@ -2715,7 +2715,7 @@ static irqreturn_t gen11_irq_handler(int irq, void *arg) { struct drm_i915_private *i915 = arg; void __iomem * const regs = i915->uncore.regs; - struct intel_gt *gt = &i915->gt; + struct intel_gt *gt = to_gt(i915); u32 master_ctl; u32 gu_misc_iir; @@ -2771,7 +2771,7 @@ static inline void dg1_master_intr_enable(void __iomem * const regs) static irqreturn_t dg1_irq_handler(int irq, void *arg) { struct drm_i915_private * const i915 = arg; - struct intel_gt *gt = &i915->gt; + struct intel_gt *gt = to_gt(i915); void __iomem * const regs = gt->uncore->regs; u32 master_tile_ctl, master_ctl; u32 gu_misc_iir; @@ -3075,7 +3075,7 @@ static void ilk_irq_reset(struct drm_i915_private *dev_priv) intel_uncore_write(uncore, EDP_PSR_IIR, 0xffffffff); } - gen5_gt_irq_reset(&dev_priv->gt); + gen5_gt_irq_reset(to_gt(dev_priv)); ibx_irq_reset(dev_priv); } @@ -3085,7 +3085,7 @@ static void valleyview_irq_reset(struct drm_i915_private *dev_priv) intel_uncore_write(&dev_priv->uncore, VLV_MASTER_IER, 0); intel_uncore_posting_read(&dev_priv->uncore, VLV_MASTER_IER); - gen5_gt_irq_reset(&dev_priv->gt); + gen5_gt_irq_reset(to_gt(dev_priv)); spin_lock_irq(&dev_priv->irq_lock); if (dev_priv->display_irqs_enabled) @@ -3119,7 +3119,7 @@ static void gen8_irq_reset(struct drm_i915_private *dev_priv) gen8_master_intr_disable(dev_priv->uncore.regs); - gen8_gt_irq_reset(&dev_priv->gt); + gen8_gt_irq_reset(to_gt(dev_priv)); gen8_display_irq_reset(dev_priv); GEN3_IRQ_RESET(uncore, GEN8_PCU_); @@ -3173,7 +3173,7 @@ static void gen11_display_irq_reset(struct drm_i915_private *dev_priv) static void gen11_irq_reset(struct drm_i915_private *dev_priv) { - struct intel_gt *gt = &dev_priv->gt; + struct intel_gt *gt = to_gt(dev_priv); struct intel_uncore *uncore = gt->uncore; gen11_master_intr_disable(dev_priv->uncore.regs); @@ -3187,7 +3187,7 @@ static void gen11_irq_reset(struct drm_i915_private *dev_priv) static void dg1_irq_reset(struct drm_i915_private *dev_priv) { - struct intel_gt *gt = &dev_priv->gt; + struct intel_gt *gt = to_gt(dev_priv); struct intel_uncore *uncore = gt->uncore; dg1_master_intr_disable(dev_priv->uncore.regs); @@ -3252,7 +3252,7 @@ static void cherryview_irq_reset(struct drm_i915_private *dev_priv) intel_uncore_write(&dev_priv->uncore, GEN8_MASTER_IRQ, 0); intel_uncore_posting_read(&dev_priv->uncore, GEN8_MASTER_IRQ); - gen8_gt_irq_reset(&dev_priv->gt); + gen8_gt_irq_reset(to_gt(dev_priv)); GEN3_IRQ_RESET(uncore, GEN8_PCU_); @@ -3709,7 +3709,7 @@ static void ilk_irq_postinstall(struct drm_i915_private *dev_priv) ibx_irq_postinstall(dev_priv); - gen5_gt_irq_postinstall(&dev_priv->gt); + gen5_gt_irq_postinstall(to_gt(dev_priv)); GEN3_IRQ_INIT(uncore, DE, dev_priv->irq_mask, display_mask | extra_mask); @@ -3746,7 +3746,7 @@ void valleyview_disable_display_irqs(struct drm_i915_private *dev_priv) static void valleyview_irq_postinstall(struct drm_i915_private *dev_priv) { - gen5_gt_irq_postinstall(&dev_priv->gt); + gen5_gt_irq_postinstall(to_gt(dev_priv)); spin_lock_irq(&dev_priv->irq_lock); if (dev_priv->display_irqs_enabled) @@ -3852,7 +3852,7 @@ static void gen8_irq_postinstall(struct drm_i915_private *dev_priv) else if (HAS_PCH_SPLIT(dev_priv)) ibx_irq_postinstall(dev_priv); - gen8_gt_irq_postinstall(&dev_priv->gt); + gen8_gt_irq_postinstall(to_gt(dev_priv)); gen8_de_irq_postinstall(dev_priv); gen8_master_intr_enable(dev_priv->uncore.regs); @@ -3871,7 +3871,7 @@ static void gen11_de_irq_postinstall(struct drm_i915_private *dev_priv) static void gen11_irq_postinstall(struct drm_i915_private *dev_priv) { - struct intel_gt *gt = &dev_priv->gt; + struct intel_gt *gt = to_gt(dev_priv); struct intel_uncore *uncore = gt->uncore; u32 gu_misc_masked = GEN11_GU_MISC_GSE; @@ -3889,7 +3889,7 @@ static void gen11_irq_postinstall(struct drm_i915_private *dev_priv) static void dg1_irq_postinstall(struct drm_i915_private *dev_priv) { - struct intel_gt *gt = &dev_priv->gt; + struct intel_gt *gt = to_gt(dev_priv); struct intel_uncore *uncore = gt->uncore; u32 gu_misc_masked = GEN11_GU_MISC_GSE; @@ -3910,7 +3910,7 @@ static void dg1_irq_postinstall(struct drm_i915_private *dev_priv) static void cherryview_irq_postinstall(struct drm_i915_private *dev_priv) { - gen8_gt_irq_postinstall(&dev_priv->gt); + gen8_gt_irq_postinstall(to_gt(dev_priv)); spin_lock_irq(&dev_priv->irq_lock); if (dev_priv->display_irqs_enabled) @@ -4073,7 +4073,7 @@ static irqreturn_t i8xx_irq_handler(int irq, void *arg) intel_uncore_write16(&dev_priv->uncore, GEN2_IIR, iir); if (iir & I915_USER_INTERRUPT) - intel_engine_cs_irq(dev_priv->gt.engine[RCS0], iir); + intel_engine_cs_irq(to_gt(dev_priv)->engine[RCS0], iir); if (iir & I915_MASTER_ERROR_INTERRUPT) i8xx_error_irq_handler(dev_priv, eir, eir_stuck); @@ -4181,7 +4181,7 @@ static irqreturn_t i915_irq_handler(int irq, void *arg) intel_uncore_write(&dev_priv->uncore, GEN2_IIR, iir); if (iir & I915_USER_INTERRUPT) - intel_engine_cs_irq(dev_priv->gt.engine[RCS0], iir); + intel_engine_cs_irq(to_gt(dev_priv)->engine[RCS0], iir); if (iir & I915_MASTER_ERROR_INTERRUPT) i9xx_error_irq_handler(dev_priv, eir, eir_stuck); @@ -4326,11 +4326,11 @@ static irqreturn_t i965_irq_handler(int irq, void *arg) intel_uncore_write(&dev_priv->uncore, GEN2_IIR, iir); if (iir & I915_USER_INTERRUPT) - intel_engine_cs_irq(dev_priv->gt.engine[RCS0], + intel_engine_cs_irq(to_gt(dev_priv)->engine[RCS0], iir); if (iir & I915_BSD_USER_INTERRUPT) - intel_engine_cs_irq(dev_priv->gt.engine[VCS0], + intel_engine_cs_irq(to_gt(dev_priv)->engine[VCS0], iir >> 25); if (iir & I915_MASTER_ERROR_INTERRUPT) @@ -4381,7 +4381,7 @@ void intel_irq_init(struct drm_i915_private *dev_priv) /* pre-gen11 the guc irqs bits are in the upper 16 bits of the pm reg */ if (HAS_GT_UC(dev_priv) && GRAPHICS_VER(dev_priv) < 11) - dev_priv->gt.pm_guc_events = GUC_INTR_GUC2HOST << 16; + to_gt(dev_priv)->pm_guc_events = GUC_INTR_GUC2HOST << 16; if (!HAS_DISPLAY(dev_priv)) return; diff --git a/drivers/gpu/drm/i915/i915_params.c b/drivers/gpu/drm/i915/i915_params.c index e07f4cfea63a..525ae832aa9a 100644 --- a/drivers/gpu/drm/i915/i915_params.c +++ b/drivers/gpu/drm/i915/i915_params.c @@ -140,6 +140,9 @@ i915_param_named_unsafe(invert_brightness, int, 0400, i915_param_named(disable_display, bool, 0400, "Disable display (default: false)"); +i915_param_named(memtest, bool, 0400, + "Perform a read/write test of all device memory on module load (default: off)"); + i915_param_named(mmio_debug, int, 0400, "Enable the MMIO debug code for the first N failures (default: off). " "This may negatively affect performance."); diff --git a/drivers/gpu/drm/i915/i915_params.h b/drivers/gpu/drm/i915/i915_params.h index 8d725b64592d..c9d53ff910a0 100644 --- a/drivers/gpu/drm/i915/i915_params.h +++ b/drivers/gpu/drm/i915/i915_params.h @@ -64,6 +64,7 @@ struct drm_printer; param(char *, guc_firmware_path, NULL, 0400) \ param(char *, huc_firmware_path, NULL, 0400) \ param(char *, dmc_firmware_path, NULL, 0400) \ + param(bool, memtest, false, 0400) \ param(int, mmio_debug, -IS_ENABLED(CONFIG_DRM_I915_DEBUG_MMIO), 0600) \ param(int, edp_vswing, 0, 0400) \ param(unsigned int, reset, 3, 0600) \ diff --git a/drivers/gpu/drm/i915/i915_pci.c b/drivers/gpu/drm/i915/i915_pci.c index ae36dfd77dcf..261294df535c 100644 --- a/drivers/gpu/drm/i915/i915_pci.c +++ b/drivers/gpu/drm/i915/i915_pci.c @@ -1027,6 +1027,7 @@ static const struct intel_device_info xehpsdv_info = { DGFX_FEATURES, PLATFORM(INTEL_XEHPSDV), .display = { }, + .has_64k_pages = 1, .platform_engine_mask = BIT(RCS0) | BIT(BCS0) | BIT(VECS0) | BIT(VECS1) | BIT(VECS2) | BIT(VECS3) | @@ -1044,6 +1045,7 @@ static const struct intel_device_info dg2_info = { .graphics.rel = 55, .media.rel = 55, PLATFORM(INTEL_DG2), + .has_64k_pages = 1, .platform_engine_mask = BIT(RCS0) | BIT(BCS0) | BIT(VECS0) | BIT(VECS1) | diff --git a/drivers/gpu/drm/i915/i915_perf.c b/drivers/gpu/drm/i915/i915_perf.c index 2f01b8c0284c..170bba913c30 100644 --- a/drivers/gpu/drm/i915/i915_perf.c +++ b/drivers/gpu/drm/i915/i915_perf.c @@ -4443,7 +4443,7 @@ void i915_perf_init(struct drm_i915_private *i915) mutex_init(&perf->lock); /* Choose a representative limit */ - oa_sample_rate_hard_limit = i915->gt.clock_frequency / 2; + oa_sample_rate_hard_limit = to_gt(i915)->clock_frequency / 2; mutex_init(&perf->metrics_lock); idr_init_base(&perf->metrics_idr, 1); diff --git a/drivers/gpu/drm/i915/i915_pmu.c b/drivers/gpu/drm/i915/i915_pmu.c index 0b488d49694c..ea655161793e 100644 --- a/drivers/gpu/drm/i915/i915_pmu.c +++ b/drivers/gpu/drm/i915/i915_pmu.c @@ -210,8 +210,8 @@ static void init_rc6(struct i915_pmu *pmu) struct drm_i915_private *i915 = container_of(pmu, typeof(*i915), pmu); intel_wakeref_t wakeref; - with_intel_runtime_pm(i915->gt.uncore->rpm, wakeref) { - pmu->sample[__I915_SAMPLE_RC6].cur = __get_rc6(&i915->gt); + with_intel_runtime_pm(to_gt(i915)->uncore->rpm, wakeref) { + pmu->sample[__I915_SAMPLE_RC6].cur = __get_rc6(to_gt(i915)); pmu->sample[__I915_SAMPLE_RC6_LAST_REPORTED].cur = pmu->sample[__I915_SAMPLE_RC6].cur; pmu->sleep_last = ktime_get_raw(); @@ -222,7 +222,7 @@ static void park_rc6(struct drm_i915_private *i915) { struct i915_pmu *pmu = &i915->pmu; - pmu->sample[__I915_SAMPLE_RC6].cur = __get_rc6(&i915->gt); + pmu->sample[__I915_SAMPLE_RC6].cur = __get_rc6(to_gt(i915)); pmu->sleep_last = ktime_get_raw(); } @@ -419,7 +419,7 @@ static enum hrtimer_restart i915_sample(struct hrtimer *hrtimer) struct drm_i915_private *i915 = container_of(hrtimer, struct drm_i915_private, pmu.timer); struct i915_pmu *pmu = &i915->pmu; - struct intel_gt *gt = &i915->gt; + struct intel_gt *gt = to_gt(i915); unsigned int period_ns; ktime_t now; @@ -476,7 +476,7 @@ engine_event_status(struct intel_engine_cs *engine, static int config_status(struct drm_i915_private *i915, u64 config) { - struct intel_gt *gt = &i915->gt; + struct intel_gt *gt = to_gt(i915); switch (config) { case I915_PMU_ACTUAL_FREQUENCY: @@ -601,10 +601,10 @@ static u64 __i915_pmu_event_read(struct perf_event *event) val = READ_ONCE(pmu->irq_count); break; case I915_PMU_RC6_RESIDENCY: - val = get_rc6(&i915->gt); + val = get_rc6(to_gt(i915)); break; case I915_PMU_SOFTWARE_GT_AWAKE_TIME: - val = ktime_to_ns(intel_gt_get_awake_time(&i915->gt)); + val = ktime_to_ns(intel_gt_get_awake_time(to_gt(i915))); break; } } diff --git a/drivers/gpu/drm/i915/i915_query.c b/drivers/gpu/drm/i915/i915_query.c index 51b368be0fc4..2dfbc22857a3 100644 --- a/drivers/gpu/drm/i915/i915_query.c +++ b/drivers/gpu/drm/i915/i915_query.c @@ -31,7 +31,7 @@ static int copy_query_item(void *query_hdr, size_t query_sz, static int query_topology_info(struct drm_i915_private *dev_priv, struct drm_i915_query_item *query_item) { - const struct sseu_dev_info *sseu = &dev_priv->gt.info.sseu; + const struct sseu_dev_info *sseu = &to_gt(dev_priv)->info.sseu; struct drm_i915_query_topology_info topo; u32 slice_length, subslice_length, eu_length, total_length; int ret; diff --git a/drivers/gpu/drm/i915/i915_reg.h b/drivers/gpu/drm/i915/i915_reg.h index d27ba273cc68..4c28dadf8d69 100644 --- a/drivers/gpu/drm/i915/i915_reg.h +++ b/drivers/gpu/drm/i915/i915_reg.h @@ -9417,6 +9417,7 @@ enum { #define GEN6_OFFSET(x) ((x) << 19) #define GEN6_AGGRESSIVE_TURBO (0 << 15) #define GEN9_SW_REQ_UNSLICE_RATIO_SHIFT 23 +#define GEN9_IGNORE_SLICE_RATIO (0 << 0) #define GEN6_RC_VIDEO_FREQ _MMIO(0xA00C) #define GEN6_RC_CONTROL _MMIO(0xA090) @@ -9452,6 +9453,9 @@ enum { #define GEN6_RP_UP_BUSY_CONT (0x4 << 3) #define GEN6_RP_DOWN_IDLE_AVG (0x2 << 0) #define GEN6_RP_DOWN_IDLE_CONT (0x1 << 0) +#define GEN6_RPSWCTL_SHIFT 9 +#define GEN9_RPSWCTL_ENABLE (0x2 << GEN6_RPSWCTL_SHIFT) +#define GEN9_RPSWCTL_DISABLE (0x0 << GEN6_RPSWCTL_SHIFT) #define GEN6_RP_UP_THRESHOLD _MMIO(0xA02C) #define GEN6_RP_DOWN_THRESHOLD _MMIO(0xA030) #define GEN6_RP_CUR_UP_EI _MMIO(0xA050) diff --git a/drivers/gpu/drm/i915/i915_request.c b/drivers/gpu/drm/i915/i915_request.c index ad175d662b4e..76cf5ac91e94 100644 --- a/drivers/gpu/drm/i915/i915_request.c +++ b/drivers/gpu/drm/i915/i915_request.c @@ -42,6 +42,7 @@ #include "gt/intel_rps.h" #include "i915_active.h" +#include "i915_deps.h" #include "i915_drv.h" #include "i915_trace.h" #include "intel_pm.h" @@ -308,6 +309,7 @@ void i915_request_free_capture_list(struct i915_capture_list *capture) struct i915_capture_list *next = capture->next; i915_vma_snapshot_put(capture->vma_snapshot); + kfree(capture); capture = next; } } @@ -1543,6 +1545,27 @@ i915_request_await_dma_fence(struct i915_request *rq, struct dma_fence *fence) } /** + * i915_request_await_deps - set this request to (async) wait upon a struct + * i915_deps dma_fence collection + * @rq: request we are wishing to use + * @deps: The struct i915_deps containing the dependencies. + * + * Returns 0 if successful, negative error code on error. + */ +int i915_request_await_deps(struct i915_request *rq, const struct i915_deps *deps) +{ + int i, err; + + for (i = 0; i < deps->num_deps; ++i) { + err = i915_request_await_dma_fence(rq, deps->fences[i]); + if (err) + return err; + } + + return 0; +} + +/** * i915_request_await_object - set this request to (async) wait upon a bo * @to: request we are wishing to use * @obj: object which may be in use on another ring. diff --git a/drivers/gpu/drm/i915/i915_request.h b/drivers/gpu/drm/i915/i915_request.h index 0ed01979491b..170ee78c2858 100644 --- a/drivers/gpu/drm/i915/i915_request.h +++ b/drivers/gpu/drm/i915/i915_request.h @@ -47,6 +47,7 @@ struct drm_file; struct drm_i915_gem_object; struct drm_printer; +struct i915_deps; struct i915_request; #if IS_ENABLED(CONFIG_DRM_I915_CAPTURE_ERROR) @@ -411,6 +412,7 @@ int i915_request_await_object(struct i915_request *to, bool write); int i915_request_await_dma_fence(struct i915_request *rq, struct dma_fence *fence); +int i915_request_await_deps(struct i915_request *rq, const struct i915_deps *deps); int i915_request_await_execution(struct i915_request *rq, struct dma_fence *fence); @@ -657,7 +659,8 @@ i915_request_timeline(const struct i915_request *rq) { /* Valid only while the request is being constructed (or retired). */ return rcu_dereference_protected(rq->timeline, - lockdep_is_held(&rcu_access_pointer(rq->timeline)->mutex)); + lockdep_is_held(&rcu_access_pointer(rq->timeline)->mutex) || + test_bit(CONTEXT_IS_PARKING, &rq->context->flags)); } static inline struct i915_gem_context * diff --git a/drivers/gpu/drm/i915/i915_sysfs.c b/drivers/gpu/drm/i915/i915_sysfs.c index 59d441cedc75..fae4d1f4f275 100644 --- a/drivers/gpu/drm/i915/i915_sysfs.c +++ b/drivers/gpu/drm/i915/i915_sysfs.c @@ -52,7 +52,7 @@ static u32 calc_residency(struct drm_i915_private *dev_priv, u64 res = 0; with_intel_runtime_pm(&dev_priv->runtime_pm, wakeref) - res = intel_rc6_residency_us(&dev_priv->gt.rc6, reg); + res = intel_rc6_residency_us(&to_gt(dev_priv)->rc6, reg); return DIV_ROUND_CLOSEST_ULL(res, 1000); } @@ -260,7 +260,7 @@ static ssize_t gt_act_freq_mhz_show(struct device *kdev, struct device_attribute *attr, char *buf) { struct drm_i915_private *i915 = kdev_minor_to_i915(kdev); - struct intel_rps *rps = &i915->gt.rps; + struct intel_rps *rps = &to_gt(i915)->rps; return sysfs_emit(buf, "%d\n", intel_rps_read_actual_frequency(rps)); } @@ -269,7 +269,7 @@ static ssize_t gt_cur_freq_mhz_show(struct device *kdev, struct device_attribute *attr, char *buf) { struct drm_i915_private *i915 = kdev_minor_to_i915(kdev); - struct intel_rps *rps = &i915->gt.rps; + struct intel_rps *rps = &to_gt(i915)->rps; return sysfs_emit(buf, "%d\n", intel_rps_get_requested_frequency(rps)); } @@ -277,7 +277,7 @@ static ssize_t gt_cur_freq_mhz_show(struct device *kdev, static ssize_t gt_boost_freq_mhz_show(struct device *kdev, struct device_attribute *attr, char *buf) { struct drm_i915_private *i915 = kdev_minor_to_i915(kdev); - struct intel_rps *rps = &i915->gt.rps; + struct intel_rps *rps = &to_gt(i915)->rps; return sysfs_emit(buf, "%d\n", intel_rps_get_boost_frequency(rps)); } @@ -287,7 +287,7 @@ static ssize_t gt_boost_freq_mhz_store(struct device *kdev, const char *buf, size_t count) { struct drm_i915_private *dev_priv = kdev_minor_to_i915(kdev); - struct intel_rps *rps = &dev_priv->gt.rps; + struct intel_rps *rps = &to_gt(dev_priv)->rps; ssize_t ret; u32 val; @@ -304,7 +304,7 @@ static ssize_t vlv_rpe_freq_mhz_show(struct device *kdev, struct device_attribute *attr, char *buf) { struct drm_i915_private *dev_priv = kdev_minor_to_i915(kdev); - struct intel_rps *rps = &dev_priv->gt.rps; + struct intel_rps *rps = &to_gt(dev_priv)->rps; return sysfs_emit(buf, "%d\n", intel_gpu_freq(rps, rps->efficient_freq)); } @@ -312,7 +312,7 @@ static ssize_t vlv_rpe_freq_mhz_show(struct device *kdev, static ssize_t gt_max_freq_mhz_show(struct device *kdev, struct device_attribute *attr, char *buf) { struct drm_i915_private *dev_priv = kdev_minor_to_i915(kdev); - struct intel_gt *gt = &dev_priv->gt; + struct intel_gt *gt = to_gt(dev_priv); struct intel_rps *rps = >->rps; return sysfs_emit(buf, "%d\n", intel_rps_get_max_frequency(rps)); @@ -323,7 +323,7 @@ static ssize_t gt_max_freq_mhz_store(struct device *kdev, const char *buf, size_t count) { struct drm_i915_private *dev_priv = kdev_minor_to_i915(kdev); - struct intel_gt *gt = &dev_priv->gt; + struct intel_gt *gt = to_gt(dev_priv); struct intel_rps *rps = >->rps; ssize_t ret; u32 val; @@ -340,7 +340,7 @@ static ssize_t gt_max_freq_mhz_store(struct device *kdev, static ssize_t gt_min_freq_mhz_show(struct device *kdev, struct device_attribute *attr, char *buf) { struct drm_i915_private *i915 = kdev_minor_to_i915(kdev); - struct intel_gt *gt = &i915->gt; + struct intel_gt *gt = to_gt(i915); struct intel_rps *rps = >->rps; return sysfs_emit(buf, "%d\n", intel_rps_get_min_frequency(rps)); @@ -351,7 +351,7 @@ static ssize_t gt_min_freq_mhz_store(struct device *kdev, const char *buf, size_t count) { struct drm_i915_private *i915 = kdev_minor_to_i915(kdev); - struct intel_rps *rps = &i915->gt.rps; + struct intel_rps *rps = &to_gt(i915)->rps; ssize_t ret; u32 val; @@ -381,7 +381,7 @@ static DEVICE_ATTR(gt_RPn_freq_mhz, S_IRUGO, gt_rp_mhz_show, NULL); static ssize_t gt_rp_mhz_show(struct device *kdev, struct device_attribute *attr, char *buf) { struct drm_i915_private *dev_priv = kdev_minor_to_i915(kdev); - struct intel_rps *rps = &dev_priv->gt.rps; + struct intel_rps *rps = &to_gt(dev_priv)->rps; u32 val; if (attr == &dev_attr_gt_RP0_freq_mhz) diff --git a/drivers/gpu/drm/i915/i915_vma.c b/drivers/gpu/drm/i915/i915_vma.c index 927f0d4f8e11..29a858c53bdd 100644 --- a/drivers/gpu/drm/i915/i915_vma.c +++ b/drivers/gpu/drm/i915/i915_vma.c @@ -109,7 +109,6 @@ vma_create(struct drm_i915_gem_object *obj, return ERR_PTR(-ENOMEM); kref_init(&vma->ref); - mutex_init(&vma->pages_mutex); vma->vm = i915_vm_get(vm); vma->ops = &vm->vma_ops; vma->obj = obj; @@ -356,22 +355,18 @@ int i915_vma_wait_for_bind(struct i915_vma *vma) #if IS_ENABLED(CONFIG_DRM_I915_DEBUG_GEM) static int i915_vma_verify_bind_complete(struct i915_vma *vma) { - int err = 0; - - if (i915_active_has_exclusive(&vma->active)) { - struct dma_fence *fence = - i915_active_fence_get(&vma->active.excl); + struct dma_fence *fence = i915_active_fence_get(&vma->active.excl); + int err; - if (!fence) - return 0; + if (!fence) + return 0; - if (dma_fence_is_signaled(fence)) - err = fence->error; - else - err = -EBUSY; + if (dma_fence_is_signaled(fence)) + err = fence->error; + else + err = -EBUSY; - dma_fence_put(fence); - } + dma_fence_put(fence); return err; } @@ -398,6 +393,7 @@ int i915_vma_bind(struct i915_vma *vma, u32 bind_flags; u32 vma_flags; + lockdep_assert_held(&vma->vm->mutex); GEM_BUG_ON(!drm_mm_node_allocated(&vma->node)); GEM_BUG_ON(vma->size > vma->node.size); @@ -419,7 +415,7 @@ int i915_vma_bind(struct i915_vma *vma, if (bind_flags == 0) return 0; - GEM_BUG_ON(!vma->pages); + GEM_BUG_ON(!atomic_read(&vma->pages_count)); trace_i915_vma_bind(vma, bind_flags); if (work && bind_flags & vma->vm->bind_async_flags) { @@ -820,10 +816,337 @@ unpinned: return pinned; } -static int vma_get_pages(struct i915_vma *vma) +static struct scatterlist * +rotate_pages(struct drm_i915_gem_object *obj, unsigned int offset, + unsigned int width, unsigned int height, + unsigned int src_stride, unsigned int dst_stride, + struct sg_table *st, struct scatterlist *sg) { - int err = 0; - bool pinned_pages = true; + unsigned int column, row; + unsigned int src_idx; + + for (column = 0; column < width; column++) { + unsigned int left; + + src_idx = src_stride * (height - 1) + column + offset; + for (row = 0; row < height; row++) { + st->nents++; + /* + * We don't need the pages, but need to initialize + * the entries so the sg list can be happily traversed. + * The only thing we need are DMA addresses. + */ + sg_set_page(sg, NULL, I915_GTT_PAGE_SIZE, 0); + sg_dma_address(sg) = + i915_gem_object_get_dma_address(obj, src_idx); + sg_dma_len(sg) = I915_GTT_PAGE_SIZE; + sg = sg_next(sg); + src_idx -= src_stride; + } + + left = (dst_stride - height) * I915_GTT_PAGE_SIZE; + + if (!left) + continue; + + st->nents++; + + /* + * The DE ignores the PTEs for the padding tiles, the sg entry + * here is just a conenience to indicate how many padding PTEs + * to insert at this spot. + */ + sg_set_page(sg, NULL, left, 0); + sg_dma_address(sg) = 0; + sg_dma_len(sg) = left; + sg = sg_next(sg); + } + + return sg; +} + +static noinline struct sg_table * +intel_rotate_pages(struct intel_rotation_info *rot_info, + struct drm_i915_gem_object *obj) +{ + unsigned int size = intel_rotation_info_size(rot_info); + struct drm_i915_private *i915 = to_i915(obj->base.dev); + struct sg_table *st; + struct scatterlist *sg; + int ret = -ENOMEM; + int i; + + /* Allocate target SG list. */ + st = kmalloc(sizeof(*st), GFP_KERNEL); + if (!st) + goto err_st_alloc; + + ret = sg_alloc_table(st, size, GFP_KERNEL); + if (ret) + goto err_sg_alloc; + + st->nents = 0; + sg = st->sgl; + + for (i = 0 ; i < ARRAY_SIZE(rot_info->plane); i++) + sg = rotate_pages(obj, rot_info->plane[i].offset, + rot_info->plane[i].width, rot_info->plane[i].height, + rot_info->plane[i].src_stride, + rot_info->plane[i].dst_stride, + st, sg); + + return st; + +err_sg_alloc: + kfree(st); +err_st_alloc: + + drm_dbg(&i915->drm, "Failed to create rotated mapping for object size %zu! (%ux%u tiles, %u pages)\n", + obj->base.size, rot_info->plane[0].width, + rot_info->plane[0].height, size); + + return ERR_PTR(ret); +} + +static struct scatterlist * +remap_pages(struct drm_i915_gem_object *obj, + unsigned int offset, unsigned int alignment_pad, + unsigned int width, unsigned int height, + unsigned int src_stride, unsigned int dst_stride, + struct sg_table *st, struct scatterlist *sg) +{ + unsigned int row; + + if (!width || !height) + return sg; + + if (alignment_pad) { + st->nents++; + + /* + * The DE ignores the PTEs for the padding tiles, the sg entry + * here is just a convenience to indicate how many padding PTEs + * to insert at this spot. + */ + sg_set_page(sg, NULL, alignment_pad * 4096, 0); + sg_dma_address(sg) = 0; + sg_dma_len(sg) = alignment_pad * 4096; + sg = sg_next(sg); + } + + for (row = 0; row < height; row++) { + unsigned int left = width * I915_GTT_PAGE_SIZE; + + while (left) { + dma_addr_t addr; + unsigned int length; + + /* + * We don't need the pages, but need to initialize + * the entries so the sg list can be happily traversed. + * The only thing we need are DMA addresses. + */ + + addr = i915_gem_object_get_dma_address_len(obj, offset, &length); + + length = min(left, length); + + st->nents++; + + sg_set_page(sg, NULL, length, 0); + sg_dma_address(sg) = addr; + sg_dma_len(sg) = length; + sg = sg_next(sg); + + offset += length / I915_GTT_PAGE_SIZE; + left -= length; + } + + offset += src_stride - width; + + left = (dst_stride - width) * I915_GTT_PAGE_SIZE; + + if (!left) + continue; + + st->nents++; + + /* + * The DE ignores the PTEs for the padding tiles, the sg entry + * here is just a conenience to indicate how many padding PTEs + * to insert at this spot. + */ + sg_set_page(sg, NULL, left, 0); + sg_dma_address(sg) = 0; + sg_dma_len(sg) = left; + sg = sg_next(sg); + } + + return sg; +} + +static noinline struct sg_table * +intel_remap_pages(struct intel_remapped_info *rem_info, + struct drm_i915_gem_object *obj) +{ + unsigned int size = intel_remapped_info_size(rem_info); + struct drm_i915_private *i915 = to_i915(obj->base.dev); + struct sg_table *st; + struct scatterlist *sg; + unsigned int gtt_offset = 0; + int ret = -ENOMEM; + int i; + + /* Allocate target SG list. */ + st = kmalloc(sizeof(*st), GFP_KERNEL); + if (!st) + goto err_st_alloc; + + ret = sg_alloc_table(st, size, GFP_KERNEL); + if (ret) + goto err_sg_alloc; + + st->nents = 0; + sg = st->sgl; + + for (i = 0 ; i < ARRAY_SIZE(rem_info->plane); i++) { + unsigned int alignment_pad = 0; + + if (rem_info->plane_alignment) + alignment_pad = ALIGN(gtt_offset, rem_info->plane_alignment) - gtt_offset; + + sg = remap_pages(obj, + rem_info->plane[i].offset, alignment_pad, + rem_info->plane[i].width, rem_info->plane[i].height, + rem_info->plane[i].src_stride, rem_info->plane[i].dst_stride, + st, sg); + + gtt_offset += alignment_pad + + rem_info->plane[i].dst_stride * rem_info->plane[i].height; + } + + i915_sg_trim(st); + + return st; + +err_sg_alloc: + kfree(st); +err_st_alloc: + + drm_dbg(&i915->drm, "Failed to create remapped mapping for object size %zu! (%ux%u tiles, %u pages)\n", + obj->base.size, rem_info->plane[0].width, + rem_info->plane[0].height, size); + + return ERR_PTR(ret); +} + +static noinline struct sg_table * +intel_partial_pages(const struct i915_ggtt_view *view, + struct drm_i915_gem_object *obj) +{ + struct sg_table *st; + struct scatterlist *sg, *iter; + unsigned int count = view->partial.size; + unsigned int offset; + int ret = -ENOMEM; + + st = kmalloc(sizeof(*st), GFP_KERNEL); + if (!st) + goto err_st_alloc; + + ret = sg_alloc_table(st, count, GFP_KERNEL); + if (ret) + goto err_sg_alloc; + + iter = i915_gem_object_get_sg_dma(obj, view->partial.offset, &offset); + GEM_BUG_ON(!iter); + + sg = st->sgl; + st->nents = 0; + do { + unsigned int len; + + len = min(sg_dma_len(iter) - (offset << PAGE_SHIFT), + count << PAGE_SHIFT); + sg_set_page(sg, NULL, len, 0); + sg_dma_address(sg) = + sg_dma_address(iter) + (offset << PAGE_SHIFT); + sg_dma_len(sg) = len; + + st->nents++; + count -= len >> PAGE_SHIFT; + if (count == 0) { + sg_mark_end(sg); + i915_sg_trim(st); /* Drop any unused tail entries. */ + + return st; + } + + sg = __sg_next(sg); + iter = __sg_next(iter); + offset = 0; + } while (1); + +err_sg_alloc: + kfree(st); +err_st_alloc: + return ERR_PTR(ret); +} + +static int +__i915_vma_get_pages(struct i915_vma *vma) +{ + struct sg_table *pages; + int ret; + + /* + * The vma->pages are only valid within the lifespan of the borrowed + * obj->mm.pages. When the obj->mm.pages sg_table is regenerated, so + * must be the vma->pages. A simple rule is that vma->pages must only + * be accessed when the obj->mm.pages are pinned. + */ + GEM_BUG_ON(!i915_gem_object_has_pinned_pages(vma->obj)); + + switch (vma->ggtt_view.type) { + default: + GEM_BUG_ON(vma->ggtt_view.type); + fallthrough; + case I915_GGTT_VIEW_NORMAL: + pages = vma->obj->mm.pages; + break; + + case I915_GGTT_VIEW_ROTATED: + pages = + intel_rotate_pages(&vma->ggtt_view.rotated, vma->obj); + break; + + case I915_GGTT_VIEW_REMAPPED: + pages = + intel_remap_pages(&vma->ggtt_view.remapped, vma->obj); + break; + + case I915_GGTT_VIEW_PARTIAL: + pages = intel_partial_pages(&vma->ggtt_view, vma->obj); + break; + } + + ret = 0; + if (IS_ERR(pages)) { + ret = PTR_ERR(pages); + pages = NULL; + drm_err(&vma->vm->i915->drm, + "Failed to get pages for VMA view type %u (%d)!\n", + vma->ggtt_view.type, ret); + } + + vma->pages = pages; + + return ret; +} + +I915_SELFTEST_EXPORT int i915_vma_get_pages(struct i915_vma *vma) +{ + int err; if (atomic_add_unless(&vma->pages_count, 1, 0)) return 0; @@ -832,25 +1155,17 @@ static int vma_get_pages(struct i915_vma *vma) if (err) return err; - /* Allocations ahoy! */ - if (mutex_lock_interruptible(&vma->pages_mutex)) { - err = -EINTR; - goto unpin; - } + err = __i915_vma_get_pages(vma); + if (err) + goto err_unpin; - if (!atomic_read(&vma->pages_count)) { - err = vma->ops->set_pages(vma); - if (err) - goto unlock; - pinned_pages = false; - } + vma->page_sizes = vma->obj->mm.page_sizes; atomic_inc(&vma->pages_count); -unlock: - mutex_unlock(&vma->pages_mutex); -unpin: - if (pinned_pages) - __i915_gem_object_unpin_pages(vma->obj); + return 0; + +err_unpin: + __i915_gem_object_unpin_pages(vma->obj); return err; } @@ -858,18 +1173,31 @@ unpin: static void __vma_put_pages(struct i915_vma *vma, unsigned int count) { /* We allocate under vma_get_pages, so beware the shrinker */ - mutex_lock_nested(&vma->pages_mutex, SINGLE_DEPTH_NESTING); + struct sg_table *pages = READ_ONCE(vma->pages); + GEM_BUG_ON(atomic_read(&vma->pages_count) < count); + if (atomic_sub_return(count, &vma->pages_count) == 0) { - vma->ops->clear_pages(vma); - GEM_BUG_ON(vma->pages); + /* + * The atomic_sub_return is a read barrier for the READ_ONCE of + * vma->pages above. + * + * READ_ONCE is safe because this is either called from the same + * function (i915_vma_pin_ww), or guarded by vma->vm->mutex. + * + * TODO: We're leaving vma->pages dangling, until vma->obj->resv + * lock is required. + */ + if (pages != vma->obj->mm.pages) { + sg_free_table(pages); + kfree(pages); + } i915_gem_object_unpin_pages(vma->obj); } - mutex_unlock(&vma->pages_mutex); } -static void vma_put_pages(struct i915_vma *vma) +I915_SELFTEST_EXPORT void i915_vma_put_pages(struct i915_vma *vma) { if (atomic_add_unless(&vma->pages_count, -1, 1)) return; @@ -900,10 +1228,8 @@ int i915_vma_pin_ww(struct i915_vma *vma, struct i915_gem_ww_ctx *ww, unsigned int bound; int err; -#ifdef CONFIG_PROVE_LOCKING - if (debug_locks && !WARN_ON(!ww)) - assert_vma_held(vma); -#endif + assert_vma_held(vma); + GEM_BUG_ON(!ww); BUILD_BUG_ON(PIN_GLOBAL != I915_VMA_GLOBAL_BIND); BUILD_BUG_ON(PIN_USER != I915_VMA_LOCAL_BIND); @@ -914,7 +1240,7 @@ int i915_vma_pin_ww(struct i915_vma *vma, struct i915_gem_ww_ctx *ww, if (try_qad_pin(vma, flags & I915_VMA_BIND_MASK)) return 0; - err = vma_get_pages(vma); + err = i915_vma_get_pages(vma); if (err) return err; @@ -1042,10 +1368,11 @@ err_fence: err_rpm: if (wakeref) intel_runtime_pm_put(&vma->vm->i915->runtime_pm, wakeref); + if (moving) dma_fence_put(moving); - vma_put_pages(vma); + i915_vma_put_pages(vma); return err; } @@ -1060,23 +1387,15 @@ static void flush_idle_contexts(struct intel_gt *gt) intel_gt_wait_for_idle(gt, MAX_SCHEDULE_TIMEOUT); } -int i915_ggtt_pin(struct i915_vma *vma, struct i915_gem_ww_ctx *ww, - u32 align, unsigned int flags) +static int __i915_ggtt_pin(struct i915_vma *vma, struct i915_gem_ww_ctx *ww, + u32 align, unsigned int flags) { struct i915_address_space *vm = vma->vm; int err; - GEM_BUG_ON(!i915_vma_is_ggtt(vma)); - -#ifdef CONFIG_LOCKDEP - WARN_ON(!ww && dma_resv_held(vma->obj->base.resv)); -#endif - do { - if (ww) - err = i915_vma_pin_ww(vma, ww, 0, align, flags | PIN_GLOBAL); - else - err = i915_vma_pin(vma, 0, align, flags | PIN_GLOBAL); + err = i915_vma_pin_ww(vma, ww, 0, align, flags | PIN_GLOBAL); + if (err != -ENOSPC) { if (!err) { err = i915_vma_wait_for_bind(vma); @@ -1095,6 +1414,30 @@ int i915_ggtt_pin(struct i915_vma *vma, struct i915_gem_ww_ctx *ww, } while (1); } +int i915_ggtt_pin(struct i915_vma *vma, struct i915_gem_ww_ctx *ww, + u32 align, unsigned int flags) +{ + struct i915_gem_ww_ctx _ww; + int err; + + GEM_BUG_ON(!i915_vma_is_ggtt(vma)); + + if (ww) + return __i915_ggtt_pin(vma, ww, align, flags); + +#ifdef CONFIG_LOCKDEP + WARN_ON(dma_resv_held(vma->obj->base.resv)); +#endif + + for_i915_gem_ww(&_ww, err, true) { + err = i915_gem_object_lock(vma->obj, &_ww); + if (!err) + err = __i915_ggtt_pin(vma, &_ww, align, flags); + } + + return err; +} + static void __vma_close(struct i915_vma *vma, struct intel_gt *gt) { /* @@ -1249,7 +1592,7 @@ __i915_request_await_bind(struct i915_request *rq, struct i915_vma *vma) return __i915_request_await_exclusive(rq, &vma->active); } -int __i915_vma_move_to_active(struct i915_vma *vma, struct i915_request *rq) +static int __i915_vma_move_to_active(struct i915_vma *vma, struct i915_request *rq) { int err; diff --git a/drivers/gpu/drm/i915/i915_vma.h b/drivers/gpu/drm/i915/i915_vma.h index 4033aa08d5e4..32719431b3df 100644 --- a/drivers/gpu/drm/i915/i915_vma.h +++ b/drivers/gpu/drm/i915/i915_vma.h @@ -55,8 +55,6 @@ static inline bool i915_vma_is_active(const struct i915_vma *vma) /* do not reserve memory to prevent deadlocks */ #define __EXEC_OBJECT_NO_RESERVE BIT(31) -int __must_check __i915_vma_move_to_active(struct i915_vma *vma, - struct i915_request *rq); int __must_check _i915_vma_move_to_active(struct i915_vma *vma, struct i915_request *rq, struct dma_fence *fence, @@ -433,4 +431,7 @@ static inline int i915_vma_sync(struct i915_vma *vma) void i915_vma_module_exit(void); int i915_vma_module_init(void); +I915_SELFTEST_DECLARE(int i915_vma_get_pages(struct i915_vma *vma)); +I915_SELFTEST_DECLARE(void i915_vma_put_pages(struct i915_vma *vma)); + #endif diff --git a/drivers/gpu/drm/i915/i915_vma_types.h b/drivers/gpu/drm/i915/i915_vma_types.h index f03fa96a1701..ca575e129ced 100644 --- a/drivers/gpu/drm/i915/i915_vma_types.h +++ b/drivers/gpu/drm/i915/i915_vma_types.h @@ -270,7 +270,6 @@ struct i915_vma { #define I915_VMA_PAGES_BIAS 24 #define I915_VMA_PAGES_ACTIVE (BIT(24) | 1) atomic_t pages_count; /* number of active binds to the pages */ - struct mutex pages_mutex; /* protect acquire/release of backing pages */ /** * Support different GGTT views into the same object. diff --git a/drivers/gpu/drm/i915/intel_device_info.h b/drivers/gpu/drm/i915/intel_device_info.h index d9ac6540d058..78597d382445 100644 --- a/drivers/gpu/drm/i915/intel_device_info.h +++ b/drivers/gpu/drm/i915/intel_device_info.h @@ -126,6 +126,7 @@ enum intel_ppgtt_type { func(is_dgfx); \ /* Keep has_* in alphabetical order */ \ func(has_64bit_reloc); \ + func(has_64k_pages); \ func(gpu_reset_clobbers_display); \ func(has_reset_engine); \ func(has_global_mocs); \ diff --git a/drivers/gpu/drm/i915/intel_gvt.c b/drivers/gpu/drm/i915/intel_gvt.c index 4e70c1a9ef2e..cf6e98962d82 100644 --- a/drivers/gpu/drm/i915/intel_gvt.c +++ b/drivers/gpu/drm/i915/intel_gvt.c @@ -109,7 +109,7 @@ int intel_gvt_init(struct drm_i915_private *dev_priv) return 0; } - if (intel_uc_wants_guc_submission(&dev_priv->gt.uc)) { + if (intel_uc_wants_guc_submission(&to_gt(dev_priv)->uc)) { drm_err(&dev_priv->drm, "i915 GVT-g loading failed due to Graphics virtualization is not yet supported with GuC submission\n"); return -EIO; diff --git a/drivers/gpu/drm/i915/intel_memory_region.c b/drivers/gpu/drm/i915/intel_memory_region.c index b43121609e25..c70d7e286a51 100644 --- a/drivers/gpu/drm/i915/intel_memory_region.c +++ b/drivers/gpu/drm/i915/intel_memory_region.c @@ -3,6 +3,8 @@ * Copyright © 2019 Intel Corporation */ +#include <linux/prandom.h> + #include "intel_memory_region.h" #include "i915_drv.h" #include "i915_ttm_buddy_manager.h" @@ -29,6 +31,110 @@ static const struct { }, }; +static int __iopagetest(struct intel_memory_region *mem, + u8 __iomem *va, int pagesize, + u8 value, resource_size_t offset, + const void *caller) +{ + int byte = prandom_u32_max(pagesize); + u8 result[3]; + + memset_io(va, value, pagesize); /* or GPF! */ + wmb(); + + result[0] = ioread8(va); + result[1] = ioread8(va + byte); + result[2] = ioread8(va + pagesize - 1); + if (memchr_inv(result, value, sizeof(result))) { + dev_err(mem->i915->drm.dev, + "Failed to read back from memory region:%pR at [%pa + %pa] for %ps; wrote %x, read (%x, %x, %x)\n", + &mem->region, &mem->io_start, &offset, caller, + value, result[0], result[1], result[2]); + return -EINVAL; + } + + return 0; +} + +static int iopagetest(struct intel_memory_region *mem, + resource_size_t offset, + const void *caller) +{ + const u8 val[] = { 0x0, 0xa5, 0xc3, 0xf0 }; + void __iomem *va; + int err; + int i; + + va = ioremap_wc(mem->io_start + offset, PAGE_SIZE); + if (!va) { + dev_err(mem->i915->drm.dev, + "Failed to ioremap memory region [%pa + %pa] for %ps\n", + &mem->io_start, &offset, caller); + return -EFAULT; + } + + for (i = 0; i < ARRAY_SIZE(val); i++) { + err = __iopagetest(mem, va, PAGE_SIZE, val[i], offset, caller); + if (err) + break; + + err = __iopagetest(mem, va, PAGE_SIZE, ~val[i], offset, caller); + if (err) + break; + } + + iounmap(va); + return err; +} + +static resource_size_t random_page(resource_size_t last) +{ + /* Limited to low 44b (16TiB), but should suffice for a spot check */ + return prandom_u32_max(last >> PAGE_SHIFT) << PAGE_SHIFT; +} + +static int iomemtest(struct intel_memory_region *mem, + bool test_all, + const void *caller) +{ + resource_size_t last = resource_size(&mem->region) - PAGE_SIZE; + resource_size_t page; + int err; + + /* + * Quick test to check read/write access to the iomap (backing store). + * + * Write a byte, read it back. If the iomapping fails, we expect + * a GPF preventing further execution. If the backing store does not + * exist, the read back will return garbage. We check a couple of pages, + * the first and last of the specified region to confirm the backing + * store + iomap does cover the entire memory region; and we check + * a random offset within as a quick spot check for bad memory. + */ + + if (test_all) { + for (page = 0; page <= last; page += PAGE_SIZE) { + err = iopagetest(mem, page, caller); + if (err) + return err; + } + } else { + err = iopagetest(mem, 0, caller); + if (err) + return err; + + err = iopagetest(mem, last, caller); + if (err) + return err; + + err = iopagetest(mem, random_page(last), caller); + if (err) + return err; + } + + return 0; +} + struct intel_memory_region * intel_memory_region_lookup(struct drm_i915_private *i915, u16 class, u16 instance) @@ -90,6 +196,21 @@ void intel_memory_region_debug(struct intel_memory_region *mr, &mr->total, &mr->avail); } +static int intel_memory_region_memtest(struct intel_memory_region *mem, + void *caller) +{ + struct drm_i915_private *i915 = mem->i915; + int err = 0; + + if (!mem->io_start) + return 0; + + if (IS_ENABLED(CONFIG_DRM_I915_DEBUG_GEM) || i915->params.memtest) + err = iomemtest(mem, i915->params.memtest, caller); + + return err; +} + struct intel_memory_region * intel_memory_region_create(struct drm_i915_private *i915, resource_size_t start, @@ -126,8 +247,15 @@ intel_memory_region_create(struct drm_i915_private *i915, goto err_free; } + err = intel_memory_region_memtest(mem, (void *)_RET_IP_); + if (err) + goto err_release; + return mem; +err_release: + if (mem->ops->release) + mem->ops->release(mem); err_free: kfree(mem); return ERR_PTR(err); diff --git a/drivers/gpu/drm/i915/intel_uncore.c b/drivers/gpu/drm/i915/intel_uncore.c index abdac78d3976..fc25ebf1a593 100644 --- a/drivers/gpu/drm/i915/intel_uncore.c +++ b/drivers/gpu/drm/i915/intel_uncore.c @@ -2061,12 +2061,13 @@ void intel_uncore_cleanup_mmio(struct intel_uncore *uncore) } void intel_uncore_init_early(struct intel_uncore *uncore, - struct drm_i915_private *i915) + struct intel_gt *gt) { spin_lock_init(&uncore->lock); - uncore->i915 = i915; - uncore->rpm = &i915->runtime_pm; - uncore->debug = &i915->mmio_debug; + uncore->i915 = gt->i915; + uncore->gt = gt; + uncore->rpm = >->i915->runtime_pm; + uncore->debug = >->i915->mmio_debug; } static void uncore_raw_init(struct intel_uncore *uncore) diff --git a/drivers/gpu/drm/i915/intel_uncore.h b/drivers/gpu/drm/i915/intel_uncore.h index d1d17b04e29f..210fe2a71612 100644 --- a/drivers/gpu/drm/i915/intel_uncore.h +++ b/drivers/gpu/drm/i915/intel_uncore.h @@ -129,6 +129,7 @@ struct intel_uncore { void __iomem *regs; struct drm_i915_private *i915; + struct intel_gt *gt; struct intel_runtime_pm *rpm; spinlock_t lock; /** lock is also taken in irq contexts. */ @@ -217,7 +218,7 @@ u32 intel_uncore_read_with_mcr_steering(struct intel_uncore *uncore, void intel_uncore_mmio_debug_init_early(struct intel_uncore_mmio_debug *mmio_debug); void intel_uncore_init_early(struct intel_uncore *uncore, - struct drm_i915_private *i915); + struct intel_gt *gt); int intel_uncore_setup_mmio(struct intel_uncore *uncore); int intel_uncore_init_mmio(struct intel_uncore *uncore); void intel_uncore_prune_engine_fw_domains(struct intel_uncore *uncore, diff --git a/drivers/gpu/drm/i915/intel_wopcm.c b/drivers/gpu/drm/i915/intel_wopcm.c index 5e511bb891f9..f06d21005106 100644 --- a/drivers/gpu/drm/i915/intel_wopcm.c +++ b/drivers/gpu/drm/i915/intel_wopcm.c @@ -220,7 +220,7 @@ static bool __wopcm_regs_locked(struct intel_uncore *uncore, void intel_wopcm_init(struct intel_wopcm *wopcm) { struct drm_i915_private *i915 = wopcm_to_i915(wopcm); - struct intel_gt *gt = &i915->gt; + struct intel_gt *gt = to_gt(i915); u32 guc_fw_size = intel_uc_fw_get_upload_size(>->uc.guc.fw); u32 huc_fw_size = intel_uc_fw_get_upload_size(>->uc.huc.fw); u32 ctx_rsvd = context_reserved_size(i915); diff --git a/drivers/gpu/drm/i915/pxp/intel_pxp_tee.c b/drivers/gpu/drm/i915/pxp/intel_pxp_tee.c index 5d169624ad60..195b2323ec00 100644 --- a/drivers/gpu/drm/i915/pxp/intel_pxp_tee.c +++ b/drivers/gpu/drm/i915/pxp/intel_pxp_tee.c @@ -16,7 +16,9 @@ static inline struct intel_pxp *i915_dev_to_pxp(struct device *i915_kdev) { - return &kdev_to_i915(i915_kdev)->gt.pxp; + struct drm_i915_private *i915 = kdev_to_i915(i915_kdev); + + return &to_gt(i915)->pxp; } static int intel_pxp_tee_io_message(struct intel_pxp *pxp, diff --git a/drivers/gpu/drm/i915/selftests/i915_active.c b/drivers/gpu/drm/i915/selftests/i915_active.c index 61bf4560d8af..2dac9be1de58 100644 --- a/drivers/gpu/drm/i915/selftests/i915_active.c +++ b/drivers/gpu/drm/i915/selftests/i915_active.c @@ -254,7 +254,7 @@ int i915_active_live_selftests(struct drm_i915_private *i915) SUBTEST(live_active_barrier), }; - if (intel_gt_is_wedged(&i915->gt)) + if (intel_gt_is_wedged(to_gt(i915))) return 0; return i915_subtests(tests, i915); diff --git a/drivers/gpu/drm/i915/selftests/i915_gem.c b/drivers/gpu/drm/i915/selftests/i915_gem.c index 152d9ab135b1..b5576888cd78 100644 --- a/drivers/gpu/drm/i915/selftests/i915_gem.c +++ b/drivers/gpu/drm/i915/selftests/i915_gem.c @@ -248,7 +248,7 @@ int i915_gem_live_selftests(struct drm_i915_private *i915) SUBTEST(igt_gem_ww_ctx), }; - if (intel_gt_is_wedged(&i915->gt)) + if (intel_gt_is_wedged(to_gt(i915))) return 0; return i915_live_subtests(tests, i915); diff --git a/drivers/gpu/drm/i915/selftests/i915_gem_evict.c b/drivers/gpu/drm/i915/selftests/i915_gem_evict.c index 7e0658a77659..75b709c26dd3 100644 --- a/drivers/gpu/drm/i915/selftests/i915_gem_evict.c +++ b/drivers/gpu/drm/i915/selftests/i915_gem_evict.c @@ -545,7 +545,7 @@ int i915_gem_evict_mock_selftests(void) return -ENOMEM; with_intel_runtime_pm(&i915->runtime_pm, wakeref) - err = i915_subtests(tests, &i915->gt); + err = i915_subtests(tests, to_gt(i915)); mock_destroy_device(i915); return err; @@ -557,8 +557,8 @@ int i915_gem_evict_live_selftests(struct drm_i915_private *i915) SUBTEST(igt_evict_contexts), }; - if (intel_gt_is_wedged(&i915->gt)) + if (intel_gt_is_wedged(to_gt(i915))) return 0; - return intel_gt_live_subtests(tests, &i915->gt); + return intel_gt_live_subtests(tests, to_gt(i915)); } diff --git a/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c b/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c index 46f4236039a9..575705c3bce9 100644 --- a/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c +++ b/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c @@ -155,7 +155,7 @@ static int igt_ppgtt_alloc(void *arg) if (!HAS_PPGTT(dev_priv)) return 0; - ppgtt = i915_ppgtt_create(&dev_priv->gt, 0); + ppgtt = i915_ppgtt_create(to_gt(dev_priv), 0); if (IS_ERR(ppgtt)) return PTR_ERR(ppgtt); @@ -1053,7 +1053,7 @@ static int exercise_ppgtt(struct drm_i915_private *dev_priv, if (IS_ERR(file)) return PTR_ERR(file); - ppgtt = i915_ppgtt_create(&dev_priv->gt, 0); + ppgtt = i915_ppgtt_create(to_gt(dev_priv), 0); if (IS_ERR(ppgtt)) { err = PTR_ERR(ppgtt); goto out_free; @@ -1275,7 +1275,7 @@ static void track_vma_bind(struct i915_vma *vma) __i915_gem_object_pin_pages(obj); - GEM_BUG_ON(vma->pages); + GEM_BUG_ON(atomic_read(&vma->pages_count)); atomic_set(&vma->pages_count, I915_VMA_PAGES_ACTIVE); __i915_gem_object_pin_pages(obj); vma->pages = obj->mm.pages; @@ -1953,7 +1953,9 @@ static int igt_cs_tlb(void *arg) goto end; } - err = vma->ops->set_pages(vma); + i915_gem_object_lock(bbe, NULL); + err = i915_vma_get_pages(vma); + i915_gem_object_unlock(bbe); if (err) goto end; @@ -1994,7 +1996,7 @@ end_ww: i915_request_put(rq); } - vma->ops->clear_pages(vma); + i915_vma_put_pages(vma); err = context_sync(ce); if (err) { @@ -2009,7 +2011,9 @@ end_ww: goto end; } - err = vma->ops->set_pages(vma); + i915_gem_object_lock(act, NULL); + err = i915_vma_get_pages(vma); + i915_gem_object_unlock(act); if (err) goto end; @@ -2047,7 +2051,7 @@ end_ww: } end_spin(batch, count - 1); - vma->ops->clear_pages(vma); + i915_vma_put_pages(vma); err = context_sync(ce); if (err) { diff --git a/drivers/gpu/drm/i915/selftests/i915_perf.c b/drivers/gpu/drm/i915/selftests/i915_perf.c index 9e9a6cb1d9e5..88db2e3d81d0 100644 --- a/drivers/gpu/drm/i915/selftests/i915_perf.c +++ b/drivers/gpu/drm/i915/selftests/i915_perf.c @@ -424,7 +424,7 @@ int i915_perf_live_selftests(struct drm_i915_private *i915) if (!perf->metrics_kobj || !perf->ops.enable_metric_set) return 0; - if (intel_gt_is_wedged(&i915->gt)) + if (intel_gt_is_wedged(to_gt(i915))) return 0; err = alloc_empty_config(&i915->perf); diff --git a/drivers/gpu/drm/i915/selftests/i915_request.c b/drivers/gpu/drm/i915/selftests/i915_request.c index 9979ef9197cd..92a859b34190 100644 --- a/drivers/gpu/drm/i915/selftests/i915_request.c +++ b/drivers/gpu/drm/i915/selftests/i915_request.c @@ -841,7 +841,7 @@ static struct i915_vma *empty_batch(struct drm_i915_private *i915) __i915_gem_object_flush_map(obj, 0, 64); i915_gem_object_unpin_map(obj); - intel_gt_chipset_flush(&i915->gt); + intel_gt_chipset_flush(to_gt(i915)); vma = i915_vma_instance(obj, &i915->ggtt.vm, NULL); if (IS_ERR(vma)) { @@ -982,7 +982,7 @@ static struct i915_vma *recursive_batch(struct drm_i915_private *i915) if (IS_ERR(obj)) return ERR_CAST(obj); - vma = i915_vma_instance(obj, i915->gt.vm, NULL); + vma = i915_vma_instance(obj, to_gt(i915)->vm, NULL); if (IS_ERR(vma)) { err = PTR_ERR(vma); goto err; @@ -1014,7 +1014,7 @@ static struct i915_vma *recursive_batch(struct drm_i915_private *i915) __i915_gem_object_flush_map(obj, 0, 64); i915_gem_object_unpin_map(obj); - intel_gt_chipset_flush(&i915->gt); + intel_gt_chipset_flush(to_gt(i915)); return vma; @@ -1700,7 +1700,7 @@ int i915_request_live_selftests(struct drm_i915_private *i915) SUBTEST(live_breadcrumbs_smoketest), }; - if (intel_gt_is_wedged(&i915->gt)) + if (intel_gt_is_wedged(to_gt(i915))) return 0; return i915_subtests(tests, i915); @@ -3091,7 +3091,7 @@ int i915_request_perf_selftests(struct drm_i915_private *i915) SUBTEST(perf_parallel_engines), }; - if (intel_gt_is_wedged(&i915->gt)) + if (intel_gt_is_wedged(to_gt(i915))) return 0; return i915_subtests(tests, i915); diff --git a/drivers/gpu/drm/i915/selftests/i915_selftest.c b/drivers/gpu/drm/i915/selftests/i915_selftest.c index 484759c9409c..2d6d7bd13c3c 100644 --- a/drivers/gpu/drm/i915/selftests/i915_selftest.c +++ b/drivers/gpu/drm/i915/selftests/i915_selftest.c @@ -298,10 +298,10 @@ int __i915_live_setup(void *data) struct drm_i915_private *i915 = data; /* The selftests expect an idle system */ - if (intel_gt_pm_wait_for_idle(&i915->gt)) + if (intel_gt_pm_wait_for_idle(to_gt(i915))) return -EIO; - return intel_gt_terminally_wedged(&i915->gt); + return intel_gt_terminally_wedged(to_gt(i915)); } int __i915_live_teardown(int err, void *data) diff --git a/drivers/gpu/drm/i915/selftests/i915_vma.c b/drivers/gpu/drm/i915/selftests/i915_vma.c index 1f10fe36619b..5c5809dfe9b2 100644 --- a/drivers/gpu/drm/i915/selftests/i915_vma.c +++ b/drivers/gpu/drm/i915/selftests/i915_vma.c @@ -691,7 +691,11 @@ static int igt_vma_rotate_remap(void *arg) } i915_vma_unpin(vma); - + err = i915_vma_unbind(vma); + if (err) { + pr_err("Unbinding returned %i\n", err); + goto out_object; + } cond_resched(); } } @@ -848,6 +852,11 @@ static int igt_vma_partial(void *arg) i915_vma_unpin(vma); nvma++; + err = i915_vma_unbind(vma); + if (err) { + pr_err("Unbinding returned %i\n", err); + goto out_object; + } cond_resched(); } @@ -882,6 +891,12 @@ static int igt_vma_partial(void *arg) i915_vma_unpin(vma); + err = i915_vma_unbind(vma); + if (err) { + pr_err("Unbinding returned %i\n", err); + goto out_object; + } + count = 0; list_for_each_entry(vma, &obj->vma.list, obj_link) count++; diff --git a/drivers/gpu/drm/i915/selftests/igt_flush_test.c b/drivers/gpu/drm/i915/selftests/igt_flush_test.c index a6c71fca61aa..b84594601d30 100644 --- a/drivers/gpu/drm/i915/selftests/igt_flush_test.c +++ b/drivers/gpu/drm/i915/selftests/igt_flush_test.c @@ -14,7 +14,7 @@ int igt_flush_test(struct drm_i915_private *i915) { - struct intel_gt *gt = &i915->gt; + struct intel_gt *gt = to_gt(i915); int ret = intel_gt_is_wedged(gt) ? -EIO : 0; cond_resched(); diff --git a/drivers/gpu/drm/i915/selftests/igt_live_test.c b/drivers/gpu/drm/i915/selftests/igt_live_test.c index 1c721542e277..72b58b66692a 100644 --- a/drivers/gpu/drm/i915/selftests/igt_live_test.c +++ b/drivers/gpu/drm/i915/selftests/igt_live_test.c @@ -16,7 +16,7 @@ int igt_live_test_begin(struct igt_live_test *t, const char *func, const char *name) { - struct intel_gt *gt = &i915->gt; + struct intel_gt *gt = to_gt(i915); struct intel_engine_cs *engine; enum intel_engine_id id; int err; @@ -57,7 +57,7 @@ int igt_live_test_end(struct igt_live_test *t) return -EIO; } - for_each_engine(engine, &i915->gt, id) { + for_each_engine(engine, to_gt(i915), id) { if (t->reset_engine[id] == i915_reset_engine_count(&i915->gpu_error, engine)) continue; diff --git a/drivers/gpu/drm/i915/selftests/intel_memory_region.c b/drivers/gpu/drm/i915/selftests/intel_memory_region.c index 0d5df0dc7212..8255561ff853 100644 --- a/drivers/gpu/drm/i915/selftests/intel_memory_region.c +++ b/drivers/gpu/drm/i915/selftests/intel_memory_region.c @@ -1217,7 +1217,7 @@ int intel_memory_region_live_selftests(struct drm_i915_private *i915) return 0; } - if (intel_gt_is_wedged(&i915->gt)) + if (intel_gt_is_wedged(to_gt(i915))) return 0; return i915_live_subtests(tests, i915); @@ -1229,7 +1229,7 @@ int intel_memory_region_perf_selftests(struct drm_i915_private *i915) SUBTEST(perf_memcpy), }; - if (intel_gt_is_wedged(&i915->gt)) + if (intel_gt_is_wedged(to_gt(i915))) return 0; return i915_live_subtests(tests, i915); diff --git a/drivers/gpu/drm/i915/selftests/intel_uncore.c b/drivers/gpu/drm/i915/selftests/intel_uncore.c index bc8128170a99..cdd196783535 100644 --- a/drivers/gpu/drm/i915/selftests/intel_uncore.c +++ b/drivers/gpu/drm/i915/selftests/intel_uncore.c @@ -344,5 +344,5 @@ int intel_uncore_live_selftests(struct drm_i915_private *i915) SUBTEST(live_forcewake_domains), }; - return intel_gt_live_subtests(tests, &i915->gt); + return intel_gt_live_subtests(tests, to_gt(i915)); } diff --git a/drivers/gpu/drm/i915/selftests/mock_gem_device.c b/drivers/gpu/drm/i915/selftests/mock_gem_device.c index d0e2e61de8d4..8aa7b1d33865 100644 --- a/drivers/gpu/drm/i915/selftests/mock_gem_device.c +++ b/drivers/gpu/drm/i915/selftests/mock_gem_device.c @@ -45,7 +45,7 @@ void mock_device_flush(struct drm_i915_private *i915) { - struct intel_gt *gt = &i915->gt; + struct intel_gt *gt = to_gt(i915); struct intel_engine_cs *engine; enum intel_engine_id id; @@ -64,7 +64,7 @@ static void mock_device_release(struct drm_device *dev) goto out; mock_device_flush(i915); - intel_gt_driver_remove(&i915->gt); + intel_gt_driver_remove(to_gt(i915)); i915_gem_drain_workqueue(i915); i915_gem_drain_freed_objects(i915); @@ -73,7 +73,7 @@ static void mock_device_release(struct drm_device *dev) destroy_workqueue(i915->wq); intel_region_ttm_device_fini(i915); - intel_gt_driver_late_release(&i915->gt); + intel_gt_driver_late_release(to_gt(i915)); intel_memory_regions_driver_release(i915); drm_mode_config_cleanup(&i915->drm); @@ -175,14 +175,14 @@ struct drm_i915_private *mock_gem_device(void) mkwrite_device_info(i915)->memory_regions = REGION_SMEM; intel_memory_regions_hw_probe(i915); - mock_uncore_init(&i915->uncore, i915); - spin_lock_init(&i915->gpu_error.lock); i915_gem_init__mm(i915); - intel_gt_init_early(&i915->gt, i915); - atomic_inc(&i915->gt.wakeref.count); /* disable; no hw support */ - i915->gt.awake = -ENODEV; + intel_gt_init_early(to_gt(i915), i915); + __intel_gt_init_early(to_gt(i915), i915); + mock_uncore_init(&i915->uncore, i915); + atomic_inc(&to_gt(i915)->wakeref.count); /* disable; no hw support */ + to_gt(i915)->awake = -ENODEV; ret = intel_region_ttm_device_init(i915); if (ret) @@ -195,19 +195,19 @@ struct drm_i915_private *mock_gem_device(void) mock_init_contexts(i915); mock_init_ggtt(i915, &i915->ggtt); - i915->gt.vm = i915_vm_get(&i915->ggtt.vm); + to_gt(i915)->vm = i915_vm_get(&i915->ggtt.vm); mkwrite_device_info(i915)->platform_engine_mask = BIT(0); - i915->gt.info.engine_mask = BIT(0); + to_gt(i915)->info.engine_mask = BIT(0); - i915->gt.engine[RCS0] = mock_engine(i915, "mock", RCS0); - if (!i915->gt.engine[RCS0]) + to_gt(i915)->engine[RCS0] = mock_engine(i915, "mock", RCS0); + if (!to_gt(i915)->engine[RCS0]) goto err_unlock; - if (mock_engine_init(i915->gt.engine[RCS0])) + if (mock_engine_init(to_gt(i915)->engine[RCS0])) goto err_context; - __clear_bit(I915_WEDGED, &i915->gt.reset.flags); + __clear_bit(I915_WEDGED, &to_gt(i915)->reset.flags); intel_engines_driver_register(i915); i915->do_release = true; @@ -216,13 +216,13 @@ struct drm_i915_private *mock_gem_device(void) return i915; err_context: - intel_gt_driver_remove(&i915->gt); + intel_gt_driver_remove(to_gt(i915)); err_unlock: destroy_workqueue(i915->wq); err_drv: intel_region_ttm_device_fini(i915); err_ttm: - intel_gt_driver_late_release(&i915->gt); + intel_gt_driver_late_release(to_gt(i915)); intel_memory_regions_driver_release(i915); drm_mode_config_cleanup(&i915->drm); mock_destroy_device(i915); diff --git a/drivers/gpu/drm/i915/selftests/mock_gtt.c b/drivers/gpu/drm/i915/selftests/mock_gtt.c index cc047ec594f9..1802baf80a17 100644 --- a/drivers/gpu/drm/i915/selftests/mock_gtt.c +++ b/drivers/gpu/drm/i915/selftests/mock_gtt.c @@ -70,7 +70,7 @@ struct i915_ppgtt *mock_ppgtt(struct drm_i915_private *i915, const char *name) if (!ppgtt) return NULL; - ppgtt->vm.gt = &i915->gt; + ppgtt->vm.gt = to_gt(i915); ppgtt->vm.i915 = i915; ppgtt->vm.total = round_down(U64_MAX, PAGE_SIZE); ppgtt->vm.dma = i915->drm.dev; @@ -78,6 +78,7 @@ struct i915_ppgtt *mock_ppgtt(struct drm_i915_private *i915, const char *name) i915_address_space_init(&ppgtt->vm, VM_CLASS_PPGTT); ppgtt->vm.alloc_pt_dma = alloc_pt_dma; + ppgtt->vm.alloc_scratch_dma = alloc_pt_dma; ppgtt->vm.clear_range = mock_clear_range; ppgtt->vm.insert_page = mock_insert_page; @@ -86,8 +87,6 @@ struct i915_ppgtt *mock_ppgtt(struct drm_i915_private *i915, const char *name) ppgtt->vm.vma_ops.bind_vma = mock_bind_ppgtt; ppgtt->vm.vma_ops.unbind_vma = mock_unbind_ppgtt; - ppgtt->vm.vma_ops.set_pages = ppgtt_set_pages; - ppgtt->vm.vma_ops.clear_pages = clear_pages; return ppgtt; } @@ -109,7 +108,7 @@ void mock_init_ggtt(struct drm_i915_private *i915, struct i915_ggtt *ggtt) { memset(ggtt, 0, sizeof(*ggtt)); - ggtt->vm.gt = &i915->gt; + ggtt->vm.gt = to_gt(i915); ggtt->vm.i915 = i915; ggtt->vm.is_ggtt = true; @@ -118,6 +117,7 @@ void mock_init_ggtt(struct drm_i915_private *i915, struct i915_ggtt *ggtt) ggtt->vm.total = 4096 * PAGE_SIZE; ggtt->vm.alloc_pt_dma = alloc_pt_dma; + ggtt->vm.alloc_scratch_dma = alloc_pt_dma; ggtt->vm.clear_range = mock_clear_range; ggtt->vm.insert_page = mock_insert_page; @@ -126,11 +126,9 @@ void mock_init_ggtt(struct drm_i915_private *i915, struct i915_ggtt *ggtt) ggtt->vm.vma_ops.bind_vma = mock_bind_ggtt; ggtt->vm.vma_ops.unbind_vma = mock_unbind_ggtt; - ggtt->vm.vma_ops.set_pages = ggtt_set_pages; - ggtt->vm.vma_ops.clear_pages = clear_pages; i915_address_space_init(&ggtt->vm, VM_CLASS_GGTT); - i915->gt.ggtt = ggtt; + to_gt(i915)->ggtt = ggtt; } void mock_fini_ggtt(struct i915_ggtt *ggtt) diff --git a/drivers/gpu/drm/i915/selftests/mock_uncore.c b/drivers/gpu/drm/i915/selftests/mock_uncore.c index ca57e4008701..f2d6be5e1230 100644 --- a/drivers/gpu/drm/i915/selftests/mock_uncore.c +++ b/drivers/gpu/drm/i915/selftests/mock_uncore.c @@ -42,7 +42,7 @@ __nop_read(64) void mock_uncore_init(struct intel_uncore *uncore, struct drm_i915_private *i915) { - intel_uncore_init_early(uncore, i915); + intel_uncore_init_early(uncore, to_gt(i915)); ASSIGN_RAW_WRITE_MMIO_VFUNCS(uncore, nop); ASSIGN_RAW_READ_MMIO_VFUNCS(uncore, nop); diff --git a/drivers/gpu/drm/radeon/radeon_kms.c b/drivers/gpu/drm/radeon/radeon_kms.c index 7afe28408085..e2488559cc9f 100644 --- a/drivers/gpu/drm/radeon/radeon_kms.c +++ b/drivers/gpu/drm/radeon/radeon_kms.c @@ -648,6 +648,8 @@ void radeon_driver_lastclose_kms(struct drm_device *dev) int radeon_driver_open_kms(struct drm_device *dev, struct drm_file *file_priv) { struct radeon_device *rdev = dev->dev_private; + struct radeon_fpriv *fpriv; + struct radeon_vm *vm; int r; file_priv->driver_priv = NULL; @@ -660,8 +662,6 @@ int radeon_driver_open_kms(struct drm_device *dev, struct drm_file *file_priv) /* new gpu have virtual address space support */ if (rdev->family >= CHIP_CAYMAN) { - struct radeon_fpriv *fpriv; - struct radeon_vm *vm; fpriv = kzalloc(sizeof(*fpriv), GFP_KERNEL); if (unlikely(!fpriv)) { @@ -672,35 +672,39 @@ int radeon_driver_open_kms(struct drm_device *dev, struct drm_file *file_priv) if (rdev->accel_working) { vm = &fpriv->vm; r = radeon_vm_init(rdev, vm); - if (r) { - kfree(fpriv); - goto out_suspend; - } + if (r) + goto out_fpriv; r = radeon_bo_reserve(rdev->ring_tmp_bo.bo, false); - if (r) { - radeon_vm_fini(rdev, vm); - kfree(fpriv); - goto out_suspend; - } + if (r) + goto out_vm_fini; /* map the ib pool buffer read only into * virtual address space */ vm->ib_bo_va = radeon_vm_bo_add(rdev, vm, rdev->ring_tmp_bo.bo); + if (!vm->ib_bo_va) { + r = -ENOMEM; + goto out_vm_fini; + } + r = radeon_vm_bo_set_addr(rdev, vm->ib_bo_va, RADEON_VA_IB_OFFSET, RADEON_VM_PAGE_READABLE | RADEON_VM_PAGE_SNOOPED); - if (r) { - radeon_vm_fini(rdev, vm); - kfree(fpriv); - goto out_suspend; - } + if (r) + goto out_vm_fini; } file_priv->driver_priv = fpriv; } + if (!r) + goto out_suspend; + +out_vm_fini: + radeon_vm_fini(rdev, vm); +out_fpriv: + kfree(fpriv); out_suspend: pm_runtime_mark_last_busy(dev->dev); pm_runtime_put_autosuspend(dev->dev); diff --git a/drivers/gpu/drm/radeon/radeon_vce.c b/drivers/gpu/drm/radeon/radeon_vce.c index 511a942e851d..ca4a36464340 100644 --- a/drivers/gpu/drm/radeon/radeon_vce.c +++ b/drivers/gpu/drm/radeon/radeon_vce.c @@ -513,7 +513,7 @@ int radeon_vce_cs_reloc(struct radeon_cs_parser *p, int lo, int hi, * @allocated: allocated a new handle? * * Validates the handle and return the found session index or -EINVAL - * we we don't have another free session index. + * we don't have another free session index. */ static int radeon_vce_validate_handle(struct radeon_cs_parser *p, uint32_t handle, bool *allocated) diff --git a/drivers/gpu/drm/tegra/Kconfig b/drivers/gpu/drm/tegra/Kconfig index 1650a448eabd..8cf5aeb9db6c 100644 --- a/drivers/gpu/drm/tegra/Kconfig +++ b/drivers/gpu/drm/tegra/Kconfig @@ -12,6 +12,9 @@ config DRM_TEGRA select INTERCONNECT select IOMMU_IOVA select CEC_CORE if CEC_NOTIFIER + select SND_SIMPLE_CARD if SND_SOC_TEGRA20_SPDIF + select SND_SOC_HDMI_CODEC if SND_SOC_TEGRA20_SPDIF + select SND_AUDIO_GRAPH_CARD if SND_SOC_TEGRA20_SPDIF help Choose this option if you have an NVIDIA Tegra SoC. diff --git a/drivers/gpu/drm/tegra/Makefile b/drivers/gpu/drm/tegra/Makefile index d801909182cf..df6cc986aeba 100644 --- a/drivers/gpu/drm/tegra/Makefile +++ b/drivers/gpu/drm/tegra/Makefile @@ -23,7 +23,8 @@ tegra-drm-y := \ gr2d.o \ gr3d.o \ falcon.o \ - vic.o + vic.o \ + nvdec.o tegra-drm-y += trace.o diff --git a/drivers/gpu/drm/tegra/dc.c b/drivers/gpu/drm/tegra/dc.c index a29d64f87563..eb70eee8992a 100644 --- a/drivers/gpu/drm/tegra/dc.c +++ b/drivers/gpu/drm/tegra/dc.c @@ -11,9 +11,12 @@ #include <linux/interconnect.h> #include <linux/module.h> #include <linux/of_device.h> +#include <linux/pm_domain.h> +#include <linux/pm_opp.h> #include <linux/pm_runtime.h> #include <linux/reset.h> +#include <soc/tegra/common.h> #include <soc/tegra/pmc.h> #include <drm/drm_atomic.h> @@ -890,11 +893,9 @@ static int tegra_cursor_atomic_check(struct drm_plane *plane, return 0; } -static void tegra_cursor_atomic_update(struct drm_plane *plane, - struct drm_atomic_state *state) +static void __tegra_cursor_atomic_update(struct drm_plane *plane, + struct drm_plane_state *new_state) { - struct drm_plane_state *new_state = drm_atomic_get_new_plane_state(state, - plane); struct tegra_plane_state *tegra_plane_state = to_tegra_plane_state(new_state); struct tegra_dc *dc = to_tegra_dc(new_state->crtc); struct tegra_drm *tegra = plane->dev->dev_private; @@ -990,6 +991,14 @@ static void tegra_cursor_atomic_update(struct drm_plane *plane, tegra_dc_writel(dc, value, DC_DISP_CURSOR_POSITION); } +static void tegra_cursor_atomic_update(struct drm_plane *plane, + struct drm_atomic_state *state) +{ + struct drm_plane_state *new_state = drm_atomic_get_new_plane_state(state, plane); + + __tegra_cursor_atomic_update(plane, new_state); +} + static void tegra_cursor_atomic_disable(struct drm_plane *plane, struct drm_atomic_state *state) { @@ -1009,12 +1018,78 @@ static void tegra_cursor_atomic_disable(struct drm_plane *plane, tegra_dc_writel(dc, value, DC_DISP_DISP_WIN_OPTIONS); } +static int tegra_cursor_atomic_async_check(struct drm_plane *plane, struct drm_atomic_state *state) +{ + struct drm_plane_state *new_state = drm_atomic_get_new_plane_state(state, plane); + struct drm_crtc_state *crtc_state; + int min_scale, max_scale; + int err; + + crtc_state = drm_atomic_get_existing_crtc_state(state, new_state->crtc); + if (WARN_ON(!crtc_state)) + return -EINVAL; + + if (!crtc_state->active) + return -EINVAL; + + if (plane->state->crtc != new_state->crtc || + plane->state->src_w != new_state->src_w || + plane->state->src_h != new_state->src_h || + plane->state->crtc_w != new_state->crtc_w || + plane->state->crtc_h != new_state->crtc_h || + plane->state->fb != new_state->fb || + plane->state->fb == NULL) + return -EINVAL; + + min_scale = (1 << 16) / 8; + max_scale = (8 << 16) / 1; + + err = drm_atomic_helper_check_plane_state(new_state, crtc_state, min_scale, max_scale, + true, true); + if (err < 0) + return err; + + if (new_state->visible != plane->state->visible) + return -EINVAL; + + return 0; +} + +static void tegra_cursor_atomic_async_update(struct drm_plane *plane, + struct drm_atomic_state *state) +{ + struct drm_plane_state *new_state = drm_atomic_get_new_plane_state(state, plane); + struct tegra_dc *dc = to_tegra_dc(new_state->crtc); + + plane->state->src_x = new_state->src_x; + plane->state->src_y = new_state->src_y; + plane->state->crtc_x = new_state->crtc_x; + plane->state->crtc_y = new_state->crtc_y; + + if (new_state->visible) { + struct tegra_plane *p = to_tegra_plane(plane); + u32 value; + + __tegra_cursor_atomic_update(plane, new_state); + + value = (WIN_A_ACT_REQ << p->index) << 8 | GENERAL_UPDATE; + tegra_dc_writel(dc, value, DC_CMD_STATE_CONTROL); + (void)tegra_dc_readl(dc, DC_CMD_STATE_CONTROL); + + value = (WIN_A_ACT_REQ << p->index) | GENERAL_ACT_REQ; + tegra_dc_writel(dc, value, DC_CMD_STATE_CONTROL); + (void)tegra_dc_readl(dc, DC_CMD_STATE_CONTROL); + } +} + static const struct drm_plane_helper_funcs tegra_cursor_plane_helper_funcs = { .prepare_fb = tegra_plane_prepare_fb, .cleanup_fb = tegra_plane_cleanup_fb, .atomic_check = tegra_cursor_atomic_check, .atomic_update = tegra_cursor_atomic_update, .atomic_disable = tegra_cursor_atomic_disable, + .atomic_async_check = tegra_cursor_atomic_async_check, + .atomic_async_update = tegra_cursor_atomic_async_update, }; static const uint64_t linear_modifiers[] = { @@ -1267,9 +1342,9 @@ static struct drm_plane *tegra_dc_add_planes(struct drm_device *drm, err = PTR_ERR(planes[i]); while (i--) - tegra_plane_funcs.destroy(planes[i]); + planes[i]->funcs->destroy(planes[i]); - tegra_plane_funcs.destroy(primary); + primary->funcs->destroy(primary); return ERR_PTR(err); } } @@ -1762,10 +1837,55 @@ int tegra_dc_state_setup_clock(struct tegra_dc *dc, return 0; } -static void tegra_dc_commit_state(struct tegra_dc *dc, - struct tegra_dc_state *state) +static void tegra_dc_update_voltage_state(struct tegra_dc *dc, + struct tegra_dc_state *state) +{ + unsigned long rate, pstate; + struct dev_pm_opp *opp; + int err; + + if (!dc->has_opp_table) + return; + + /* calculate actual pixel clock rate which depends on internal divider */ + rate = DIV_ROUND_UP(clk_get_rate(dc->clk) * 2, state->div + 2); + + /* find suitable OPP for the rate */ + opp = dev_pm_opp_find_freq_ceil(dc->dev, &rate); + + /* + * Very high resolution modes may results in a clock rate that is + * above the characterized maximum. In this case it's okay to fall + * back to the characterized maximum. + */ + if (opp == ERR_PTR(-ERANGE)) + opp = dev_pm_opp_find_freq_floor(dc->dev, &rate); + + if (IS_ERR(opp)) { + dev_err(dc->dev, "failed to find OPP for %luHz: %pe\n", + rate, opp); + return; + } + + pstate = dev_pm_opp_get_required_pstate(opp, 0); + dev_pm_opp_put(opp); + + /* + * The minimum core voltage depends on the pixel clock rate (which + * depends on internal clock divider of the CRTC) and not on the + * rate of the display controller clock. This is why we're not using + * dev_pm_opp_set_rate() API and instead controlling the power domain + * directly. + */ + err = dev_pm_genpd_set_performance_state(dc->dev, pstate); + if (err) + dev_err(dc->dev, "failed to set power domain state to %lu: %d\n", + pstate, err); +} + +static void tegra_dc_set_clock_rate(struct tegra_dc *dc, + struct tegra_dc_state *state) { - u32 value; int err; err = clk_set_parent(dc->clk, state->clk); @@ -1797,10 +1917,7 @@ static void tegra_dc_commit_state(struct tegra_dc *dc, state->div); DRM_DEBUG_KMS("pclk: %lu\n", state->pclk); - if (!dc->soc->has_nvdisplay) { - value = SHIFT_CLK_DIVIDER(state->div) | PIXEL_CLK_DIVIDER_PCD1; - tegra_dc_writel(dc, value, DC_DISP_DISP_CLOCK_CONTROL); - } + tegra_dc_update_voltage_state(dc, state); } static void tegra_dc_stop(struct tegra_dc *dc) @@ -1991,6 +2108,13 @@ static void tegra_crtc_atomic_disable(struct drm_crtc *crtc, err = host1x_client_suspend(&dc->client); if (err < 0) dev_err(dc->dev, "failed to suspend: %d\n", err); + + if (dc->has_opp_table) { + err = dev_pm_genpd_set_performance_state(dc->dev, 0); + if (err) + dev_err(dc->dev, + "failed to clear power domain state: %d\n", err); + } } static void tegra_crtc_atomic_enable(struct drm_crtc *crtc, @@ -2002,6 +2126,9 @@ static void tegra_crtc_atomic_enable(struct drm_crtc *crtc, u32 value; int err; + /* apply PLL changes */ + tegra_dc_set_clock_rate(dc, crtc_state); + err = host1x_client_resume(&dc->client); if (err < 0) { dev_err(dc->dev, "failed to resume: %d\n", err); @@ -2076,8 +2203,11 @@ static void tegra_crtc_atomic_enable(struct drm_crtc *crtc, else tegra_dc_writel(dc, 0, DC_DISP_BORDER_COLOR); - /* apply PLL and pixel clock changes */ - tegra_dc_commit_state(dc, crtc_state); + /* apply pixel clock changes */ + if (!dc->soc->has_nvdisplay) { + value = SHIFT_CLK_DIVIDER(crtc_state->div) | PIXEL_CLK_DIVIDER_PCD1; + tegra_dc_writel(dc, value, DC_DISP_DISP_CLOCK_CONTROL); + } /* program display mode */ tegra_dc_set_timings(dc, mode); @@ -2107,6 +2237,12 @@ static void tegra_crtc_atomic_enable(struct drm_crtc *crtc, tegra_dc_writel(dc, value, DC_COM_RG_UNDERFLOW); } + if (dc->rgb) { + /* XXX: parameterize? */ + value = SC0_H_QUALIFIER_NONE | SC1_H_QUALIFIER_NONE; + tegra_dc_writel(dc, value, DC_DISP_SHIFT_CLOCK_OPTIONS); + } + tegra_dc_commit(dc); drm_crtc_vblank_on(crtc); @@ -2685,6 +2821,7 @@ static const struct tegra_dc_soc_info tegra20_dc_soc_info = { .has_win_b_vfilter_mem_client = true, .has_win_c_without_vert_filter = true, .plane_tiled_memory_bandwidth_x2 = false, + .has_pll_d2_out0 = false, }; static const struct tegra_dc_soc_info tegra30_dc_soc_info = { @@ -2707,6 +2844,7 @@ static const struct tegra_dc_soc_info tegra30_dc_soc_info = { .has_win_b_vfilter_mem_client = true, .has_win_c_without_vert_filter = false, .plane_tiled_memory_bandwidth_x2 = true, + .has_pll_d2_out0 = true, }; static const struct tegra_dc_soc_info tegra114_dc_soc_info = { @@ -2729,6 +2867,7 @@ static const struct tegra_dc_soc_info tegra114_dc_soc_info = { .has_win_b_vfilter_mem_client = false, .has_win_c_without_vert_filter = false, .plane_tiled_memory_bandwidth_x2 = true, + .has_pll_d2_out0 = true, }; static const struct tegra_dc_soc_info tegra124_dc_soc_info = { @@ -2751,6 +2890,7 @@ static const struct tegra_dc_soc_info tegra124_dc_soc_info = { .has_win_b_vfilter_mem_client = false, .has_win_c_without_vert_filter = false, .plane_tiled_memory_bandwidth_x2 = false, + .has_pll_d2_out0 = true, }; static const struct tegra_dc_soc_info tegra210_dc_soc_info = { @@ -2773,6 +2913,7 @@ static const struct tegra_dc_soc_info tegra210_dc_soc_info = { .has_win_b_vfilter_mem_client = false, .has_win_c_without_vert_filter = false, .plane_tiled_memory_bandwidth_x2 = false, + .has_pll_d2_out0 = true, }; static const struct tegra_windowgroup_soc tegra186_dc_wgrps[] = { @@ -2823,6 +2964,7 @@ static const struct tegra_dc_soc_info tegra186_dc_soc_info = { .wgrps = tegra186_dc_wgrps, .num_wgrps = ARRAY_SIZE(tegra186_dc_wgrps), .plane_tiled_memory_bandwidth_x2 = false, + .has_pll_d2_out0 = false, }; static const struct tegra_windowgroup_soc tegra194_dc_wgrps[] = { @@ -2873,6 +3015,7 @@ static const struct tegra_dc_soc_info tegra194_dc_soc_info = { .wgrps = tegra194_dc_wgrps, .num_wgrps = ARRAY_SIZE(tegra194_dc_wgrps), .plane_tiled_memory_bandwidth_x2 = false, + .has_pll_d2_out0 = false, }; static const struct of_device_id tegra_dc_of_match[] = { @@ -2973,6 +3116,23 @@ static int tegra_dc_couple(struct tegra_dc *dc) return 0; } +static int tegra_dc_init_opp_table(struct tegra_dc *dc) +{ + struct tegra_core_opp_params opp_params = {}; + int err; + + err = devm_tegra_core_dev_init_opp_table(dc->dev, &opp_params); + if (err && err != -ENODEV) + return err; + + if (err) + dc->has_opp_table = false; + else + dc->has_opp_table = true; + + return 0; +} + static int tegra_dc_probe(struct platform_device *pdev) { u64 dma_mask = dma_get_mask(pdev->dev.parent); @@ -3038,6 +3198,10 @@ static int tegra_dc_probe(struct platform_device *pdev) tegra_powergate_power_off(dc->powergate); } + err = tegra_dc_init_opp_table(dc); + if (err < 0) + return err; + dc->regs = devm_platform_ioremap_resource(pdev, 0); if (IS_ERR(dc->regs)) return PTR_ERR(dc->regs); diff --git a/drivers/gpu/drm/tegra/dc.h b/drivers/gpu/drm/tegra/dc.h index 40378308d527..3f91a10ea6c7 100644 --- a/drivers/gpu/drm/tegra/dc.h +++ b/drivers/gpu/drm/tegra/dc.h @@ -76,6 +76,7 @@ struct tegra_dc_soc_info { bool has_win_b_vfilter_mem_client; bool has_win_c_without_vert_filter; bool plane_tiled_memory_bandwidth_x2; + bool has_pll_d2_out0; }; struct tegra_dc { @@ -100,6 +101,8 @@ struct tegra_dc { struct drm_info_list *debugfs_files; const struct tegra_dc_soc_info *soc; + + bool has_opp_table; }; static inline struct tegra_dc * diff --git a/drivers/gpu/drm/tegra/drm.c b/drivers/gpu/drm/tegra/drm.c index 8d37d6b00562..e9de91a4e7e8 100644 --- a/drivers/gpu/drm/tegra/drm.c +++ b/drivers/gpu/drm/tegra/drm.c @@ -10,6 +10,7 @@ #include <linux/iommu.h> #include <linux/module.h> #include <linux/platform_device.h> +#include <linux/pm_runtime.h> #include <drm/drm_aperture.h> #include <drm/drm_atomic.h> @@ -21,6 +22,10 @@ #include <drm/drm_prime.h> #include <drm/drm_vblank.h> +#if IS_ENABLED(CONFIG_ARM_DMA_USE_IOMMU) +#include <asm/dma-iommu.h> +#endif + #include "dc.h" #include "drm.h" #include "gem.h" @@ -116,6 +121,7 @@ static int tegra_drm_open(struct drm_device *drm, struct drm_file *filp) static void tegra_drm_context_free(struct tegra_drm_context *context) { context->client->ops->close_channel(context); + pm_runtime_put(context->client->base.dev); kfree(context); } @@ -427,13 +433,20 @@ static int tegra_client_open(struct tegra_drm_file *fpriv, { int err; + err = pm_runtime_resume_and_get(client->base.dev); + if (err) + return err; + err = client->ops->open_channel(client, context); - if (err < 0) + if (err < 0) { + pm_runtime_put(client->base.dev); return err; + } err = idr_alloc(&fpriv->legacy_contexts, context, 1, 0, GFP_KERNEL); if (err < 0) { client->ops->close_channel(context); + pm_runtime_put(client->base.dev); return err; } @@ -936,6 +949,17 @@ int host1x_client_iommu_attach(struct host1x_client *client) struct iommu_group *group = NULL; int err; +#if IS_ENABLED(CONFIG_ARM_DMA_USE_IOMMU) + if (client->dev->archdata.mapping) { + struct dma_iommu_mapping *mapping = + to_dma_iommu_mapping(client->dev); + arm_iommu_detach_device(client->dev); + arm_iommu_release_mapping(mapping); + + domain = iommu_get_domain_for_dev(client->dev); + } +#endif + /* * If the host1x client is already attached to an IOMMU domain that is * not the shared IOMMU domain, don't try to attach it to a different @@ -1344,15 +1368,18 @@ static const struct of_device_id host1x_drm_subdevs[] = { { .compatible = "nvidia,tegra210-sor", }, { .compatible = "nvidia,tegra210-sor1", }, { .compatible = "nvidia,tegra210-vic", }, + { .compatible = "nvidia,tegra210-nvdec", }, { .compatible = "nvidia,tegra186-display", }, { .compatible = "nvidia,tegra186-dc", }, { .compatible = "nvidia,tegra186-sor", }, { .compatible = "nvidia,tegra186-sor1", }, { .compatible = "nvidia,tegra186-vic", }, + { .compatible = "nvidia,tegra186-nvdec", }, { .compatible = "nvidia,tegra194-display", }, { .compatible = "nvidia,tegra194-dc", }, { .compatible = "nvidia,tegra194-sor", }, { .compatible = "nvidia,tegra194-vic", }, + { .compatible = "nvidia,tegra194-nvdec", }, { /* sentinel */ } }; @@ -1376,6 +1403,7 @@ static struct platform_driver * const drivers[] = { &tegra_gr2d_driver, &tegra_gr3d_driver, &tegra_vic_driver, + &tegra_nvdec_driver, }; static int __init host1x_drm_init(void) diff --git a/drivers/gpu/drm/tegra/drm.h b/drivers/gpu/drm/tegra/drm.h index 8b28327c931c..fc0a19554eac 100644 --- a/drivers/gpu/drm/tegra/drm.h +++ b/drivers/gpu/drm/tegra/drm.h @@ -202,5 +202,6 @@ extern struct platform_driver tegra_sor_driver; extern struct platform_driver tegra_gr2d_driver; extern struct platform_driver tegra_gr3d_driver; extern struct platform_driver tegra_vic_driver; +extern struct platform_driver tegra_nvdec_driver; #endif /* HOST1X_DRM_H */ diff --git a/drivers/gpu/drm/tegra/gem.c b/drivers/gpu/drm/tegra/gem.c index d38fd7e12b57..fce0e52973c2 100644 --- a/drivers/gpu/drm/tegra/gem.c +++ b/drivers/gpu/drm/tegra/gem.c @@ -23,86 +23,97 @@ MODULE_IMPORT_NS(DMA_BUF); -static void tegra_bo_put(struct host1x_bo *bo) +static unsigned int sg_dma_count_chunks(struct scatterlist *sgl, unsigned int nents) { - struct tegra_bo *obj = host1x_to_tegra_bo(bo); + dma_addr_t next = ~(dma_addr_t)0; + unsigned int count = 0, i; + struct scatterlist *s; - drm_gem_object_put(&obj->gem); -} + for_each_sg(sgl, s, nents, i) { + /* sg_dma_address(s) is only valid for entries that have sg_dma_len(s) != 0. */ + if (!sg_dma_len(s)) + continue; -/* XXX move this into lib/scatterlist.c? */ -static int sg_alloc_table_from_sg(struct sg_table *sgt, struct scatterlist *sg, - unsigned int nents, gfp_t gfp_mask) -{ - struct scatterlist *dst; - unsigned int i; - int err; + if (sg_dma_address(s) != next) { + next = sg_dma_address(s) + sg_dma_len(s); + count++; + } + } - err = sg_alloc_table(sgt, nents, gfp_mask); - if (err < 0) - return err; + return count; +} - dst = sgt->sgl; +static inline unsigned int sgt_dma_count_chunks(struct sg_table *sgt) +{ + return sg_dma_count_chunks(sgt->sgl, sgt->nents); +} - for (i = 0; i < nents; i++) { - sg_set_page(dst, sg_page(sg), sg->length, 0); - dst = sg_next(dst); - sg = sg_next(sg); - } +static void tegra_bo_put(struct host1x_bo *bo) +{ + struct tegra_bo *obj = host1x_to_tegra_bo(bo); - return 0; + drm_gem_object_put(&obj->gem); } -static struct sg_table *tegra_bo_pin(struct device *dev, struct host1x_bo *bo, - dma_addr_t *phys) +static struct host1x_bo_mapping *tegra_bo_pin(struct device *dev, struct host1x_bo *bo, + enum dma_data_direction direction) { struct tegra_bo *obj = host1x_to_tegra_bo(bo); - struct sg_table *sgt; + struct drm_gem_object *gem = &obj->gem; + struct host1x_bo_mapping *map; int err; + map = kzalloc(sizeof(*map), GFP_KERNEL); + if (!map) + return ERR_PTR(-ENOMEM); + + kref_init(&map->ref); + map->bo = host1x_bo_get(bo); + map->direction = direction; + map->dev = dev; + /* - * If we've manually mapped the buffer object through the IOMMU, make - * sure to return the IOVA address of our mapping. - * - * Similarly, for buffers that have been allocated by the DMA API the - * physical address can be used for devices that are not attached to - * an IOMMU. For these devices, callers must pass a valid pointer via - * the @phys argument. - * - * Imported buffers were also already mapped at import time, so the - * existing mapping can be reused. + * Imported buffers need special treatment to satisfy the semantics of DMA-BUF. */ - if (phys) { - *phys = obj->iova; - return NULL; + if (gem->import_attach) { + struct dma_buf *buf = gem->import_attach->dmabuf; + + map->attach = dma_buf_attach(buf, dev); + if (IS_ERR(map->attach)) { + err = PTR_ERR(map->attach); + goto free; + } + + map->sgt = dma_buf_map_attachment(map->attach, direction); + if (IS_ERR(map->sgt)) { + dma_buf_detach(buf, map->attach); + err = PTR_ERR(map->sgt); + goto free; + } + + err = sgt_dma_count_chunks(map->sgt); + map->size = gem->size; + + goto out; } /* * If we don't have a mapping for this buffer yet, return an SG table * so that host1x can do the mapping for us via the DMA API. */ - sgt = kzalloc(sizeof(*sgt), GFP_KERNEL); - if (!sgt) - return ERR_PTR(-ENOMEM); + map->sgt = kzalloc(sizeof(*map->sgt), GFP_KERNEL); + if (!map->sgt) { + err = -ENOMEM; + goto free; + } if (obj->pages) { /* * If the buffer object was allocated from the explicit IOMMU * API code paths, construct an SG table from the pages. */ - err = sg_alloc_table_from_pages(sgt, obj->pages, obj->num_pages, - 0, obj->gem.size, GFP_KERNEL); - if (err < 0) - goto free; - } else if (obj->sgt) { - /* - * If the buffer object already has an SG table but no pages - * were allocated for it, it means the buffer was imported and - * the SG table needs to be copied to avoid overwriting any - * other potential users of the original SG table. - */ - err = sg_alloc_table_from_sg(sgt, obj->sgt->sgl, - obj->sgt->orig_nents, GFP_KERNEL); + err = sg_alloc_table_from_pages(map->sgt, obj->pages, obj->num_pages, 0, gem->size, + GFP_KERNEL); if (err < 0) goto free; } else { @@ -111,25 +122,53 @@ static struct sg_table *tegra_bo_pin(struct device *dev, struct host1x_bo *bo, * not imported, it had to be allocated with the DMA API, so * the DMA API helper can be used. */ - err = dma_get_sgtable(dev, sgt, obj->vaddr, obj->iova, - obj->gem.size); + err = dma_get_sgtable(dev, map->sgt, obj->vaddr, obj->iova, gem->size); if (err < 0) goto free; } - return sgt; + err = dma_map_sgtable(dev, map->sgt, direction, 0); + if (err) + goto free_sgt; + +out: + /* + * If we've manually mapped the buffer object through the IOMMU, make sure to return the + * existing IOVA address of our mapping. + */ + if (!obj->mm) { + map->phys = sg_dma_address(map->sgt->sgl); + map->chunks = err; + } else { + map->phys = obj->iova; + map->chunks = 1; + } + + map->size = gem->size; + return map; + +free_sgt: + sg_free_table(map->sgt); free: - kfree(sgt); + kfree(map->sgt); + kfree(map); return ERR_PTR(err); } -static void tegra_bo_unpin(struct device *dev, struct sg_table *sgt) +static void tegra_bo_unpin(struct host1x_bo_mapping *map) { - if (sgt) { - sg_free_table(sgt); - kfree(sgt); + if (map->attach) { + dma_buf_unmap_attachment(map->attach, map->sgt, map->direction); + dma_buf_detach(map->attach->dmabuf, map->attach); + } else { + dma_unmap_sgtable(map->dev, map->sgt, map->direction, 0); + sg_free_table(map->sgt); + kfree(map->sgt); } + + host1x_bo_put(map->bo); + kfree(map); } static void *tegra_bo_mmap(struct host1x_bo *bo) @@ -452,8 +491,18 @@ free: void tegra_bo_free_object(struct drm_gem_object *gem) { struct tegra_drm *tegra = gem->dev->dev_private; + struct host1x_bo_mapping *mapping, *tmp; struct tegra_bo *bo = to_tegra_bo(gem); + /* remove all mappings of this buffer object from any caches */ + list_for_each_entry_safe(mapping, tmp, &bo->base.mappings, list) { + if (mapping->cache) + host1x_bo_unpin(mapping); + else + dev_err(gem->dev->dev, "mapping %p stale for device %s\n", mapping, + dev_name(mapping->dev)); + } + if (tegra->domain) tegra_bo_iommu_unmap(tegra, bo); diff --git a/drivers/gpu/drm/tegra/gr2d.c b/drivers/gpu/drm/tegra/gr2d.c index de288cba3905..e3bb4c99ed39 100644 --- a/drivers/gpu/drm/tegra/gr2d.c +++ b/drivers/gpu/drm/tegra/gr2d.c @@ -4,14 +4,25 @@ */ #include <linux/clk.h> +#include <linux/delay.h> #include <linux/iommu.h> #include <linux/module.h> #include <linux/of_device.h> +#include <linux/pm_runtime.h> +#include <linux/reset.h> + +#include <soc/tegra/common.h> #include "drm.h" #include "gem.h" #include "gr2d.h" +enum { + RST_MC, + RST_GR2D, + RST_GR2D_MAX, +}; + struct gr2d_soc { unsigned int version; }; @@ -21,6 +32,9 @@ struct gr2d { struct host1x_channel *channel; struct clk *clk; + struct reset_control_bulk_data resets[RST_GR2D_MAX]; + unsigned int nresets; + const struct gr2d_soc *soc; DECLARE_BITMAP(addr_regs, GR2D_NUM_REGS); @@ -56,15 +70,22 @@ static int gr2d_init(struct host1x_client *client) goto free; } + pm_runtime_enable(client->dev); + pm_runtime_use_autosuspend(client->dev); + pm_runtime_set_autosuspend_delay(client->dev, 200); + err = tegra_drm_register_client(dev->dev_private, drm); if (err < 0) { dev_err(client->dev, "failed to register client: %d\n", err); - goto detach; + goto disable_rpm; } return 0; -detach: +disable_rpm: + pm_runtime_dont_use_autosuspend(client->dev); + pm_runtime_force_suspend(client->dev); + host1x_client_iommu_detach(client); free: host1x_syncpt_put(client->syncpts[0]); @@ -85,10 +106,15 @@ static int gr2d_exit(struct host1x_client *client) if (err < 0) return err; + pm_runtime_dont_use_autosuspend(client->dev); + pm_runtime_force_suspend(client->dev); + host1x_client_iommu_detach(client); host1x_syncpt_put(client->syncpts[0]); host1x_channel_put(gr2d->channel); + gr2d->channel = NULL; + return 0; } @@ -190,6 +216,27 @@ static const u32 gr2d_addr_regs[] = { GR2D_VA_BASE_ADDR_SB, }; +static int gr2d_get_resets(struct device *dev, struct gr2d *gr2d) +{ + int err; + + gr2d->resets[RST_MC].id = "mc"; + gr2d->resets[RST_GR2D].id = "2d"; + gr2d->nresets = RST_GR2D_MAX; + + err = devm_reset_control_bulk_get_optional_exclusive_released( + dev, gr2d->nresets, gr2d->resets); + if (err) { + dev_err(dev, "failed to get reset: %d\n", err); + return err; + } + + if (WARN_ON(!gr2d->resets[RST_GR2D].rstc)) + return -ENOENT; + + return 0; +} + static int gr2d_probe(struct platform_device *pdev) { struct device *dev = &pdev->dev; @@ -202,6 +249,8 @@ static int gr2d_probe(struct platform_device *pdev) if (!gr2d) return -ENOMEM; + platform_set_drvdata(pdev, gr2d); + gr2d->soc = of_device_get_match_data(dev); syncpts = devm_kzalloc(dev, sizeof(*syncpts), GFP_KERNEL); @@ -214,11 +263,9 @@ static int gr2d_probe(struct platform_device *pdev) return PTR_ERR(gr2d->clk); } - err = clk_prepare_enable(gr2d->clk); - if (err) { - dev_err(dev, "cannot turn on clock\n"); + err = gr2d_get_resets(dev, gr2d); + if (err) return err; - } INIT_LIST_HEAD(&gr2d->client.base.list); gr2d->client.base.ops = &gr2d_client_ops; @@ -231,10 +278,13 @@ static int gr2d_probe(struct platform_device *pdev) gr2d->client.version = gr2d->soc->version; gr2d->client.ops = &gr2d_ops; + err = devm_tegra_core_dev_init_opp_table_common(dev); + if (err) + return err; + err = host1x_client_register(&gr2d->client.base); if (err < 0) { dev_err(dev, "failed to register host1x client: %d\n", err); - clk_disable_unprepare(gr2d->clk); return err; } @@ -242,8 +292,6 @@ static int gr2d_probe(struct platform_device *pdev) for (i = 0; i < ARRAY_SIZE(gr2d_addr_regs); i++) set_bit(gr2d_addr_regs[i], gr2d->addr_regs); - platform_set_drvdata(pdev, gr2d); - return 0; } @@ -259,15 +307,100 @@ static int gr2d_remove(struct platform_device *pdev) return err; } + return 0; +} + +static int __maybe_unused gr2d_runtime_suspend(struct device *dev) +{ + struct gr2d *gr2d = dev_get_drvdata(dev); + int err; + + host1x_channel_stop(gr2d->channel); + reset_control_bulk_release(gr2d->nresets, gr2d->resets); + + /* + * GR2D module shouldn't be reset while hardware is idling, otherwise + * host1x's cmdproc will stuck on trying to access any G2 register + * after reset. GR2D module could be either hot-reset or reset after + * power-gating of the HEG partition. Hence we will put in reset only + * the memory client part of the module, the HEG GENPD will take care + * of resetting GR2D module across power-gating. + * + * On Tegra20 there is no HEG partition, but it's okay to have + * undetermined h/w state since userspace is expected to reprogram + * the state on each job submission anyways. + */ + err = reset_control_acquire(gr2d->resets[RST_MC].rstc); + if (err) { + dev_err(dev, "failed to acquire MC reset: %d\n", err); + goto acquire_reset; + } + + err = reset_control_assert(gr2d->resets[RST_MC].rstc); + reset_control_release(gr2d->resets[RST_MC].rstc); + if (err) { + dev_err(dev, "failed to assert MC reset: %d\n", err); + goto acquire_reset; + } + clk_disable_unprepare(gr2d->clk); return 0; + +acquire_reset: + reset_control_bulk_acquire(gr2d->nresets, gr2d->resets); + reset_control_bulk_deassert(gr2d->nresets, gr2d->resets); + + return err; } +static int __maybe_unused gr2d_runtime_resume(struct device *dev) +{ + struct gr2d *gr2d = dev_get_drvdata(dev); + int err; + + err = reset_control_bulk_acquire(gr2d->nresets, gr2d->resets); + if (err) { + dev_err(dev, "failed to acquire reset: %d\n", err); + return err; + } + + err = clk_prepare_enable(gr2d->clk); + if (err) { + dev_err(dev, "failed to enable clock: %d\n", err); + goto release_reset; + } + + usleep_range(2000, 4000); + + /* this is a reset array which deasserts both 2D MC and 2D itself */ + err = reset_control_bulk_deassert(gr2d->nresets, gr2d->resets); + if (err) { + dev_err(dev, "failed to deassert reset: %d\n", err); + goto disable_clk; + } + + return 0; + +disable_clk: + clk_disable_unprepare(gr2d->clk); +release_reset: + reset_control_bulk_release(gr2d->nresets, gr2d->resets); + + return err; +} + +static const struct dev_pm_ops tegra_gr2d_pm = { + SET_RUNTIME_PM_OPS(gr2d_runtime_suspend, gr2d_runtime_resume, NULL) + SET_SYSTEM_SLEEP_PM_OPS(pm_runtime_force_suspend, + pm_runtime_force_resume) +}; + struct platform_driver tegra_gr2d_driver = { .driver = { .name = "tegra-gr2d", .of_match_table = gr2d_match, + .pm = &tegra_gr2d_pm, }, .probe = gr2d_probe, .remove = gr2d_remove, diff --git a/drivers/gpu/drm/tegra/gr3d.c b/drivers/gpu/drm/tegra/gr3d.c index 24442ade0da3..a1fd3113ea96 100644 --- a/drivers/gpu/drm/tegra/gr3d.c +++ b/drivers/gpu/drm/tegra/gr3d.c @@ -5,32 +5,47 @@ */ #include <linux/clk.h> +#include <linux/delay.h> #include <linux/host1x.h> #include <linux/iommu.h> #include <linux/module.h> #include <linux/of_device.h> #include <linux/platform_device.h> +#include <linux/pm_domain.h> +#include <linux/pm_opp.h> +#include <linux/pm_runtime.h> #include <linux/reset.h> +#include <soc/tegra/common.h> #include <soc/tegra/pmc.h> #include "drm.h" #include "gem.h" #include "gr3d.h" +enum { + RST_MC, + RST_GR3D, + RST_MC2, + RST_GR3D2, + RST_GR3D_MAX, +}; + struct gr3d_soc { unsigned int version; + unsigned int num_clocks; + unsigned int num_resets; }; struct gr3d { struct tegra_drm_client client; struct host1x_channel *channel; - struct clk *clk_secondary; - struct clk *clk; - struct reset_control *rst_secondary; - struct reset_control *rst; const struct gr3d_soc *soc; + struct clk_bulk_data *clocks; + unsigned int nclocks; + struct reset_control_bulk_data resets[RST_GR3D_MAX]; + unsigned int nresets; DECLARE_BITMAP(addr_regs, GR3D_NUM_REGS); }; @@ -65,15 +80,22 @@ static int gr3d_init(struct host1x_client *client) goto free; } + pm_runtime_enable(client->dev); + pm_runtime_use_autosuspend(client->dev); + pm_runtime_set_autosuspend_delay(client->dev, 200); + err = tegra_drm_register_client(dev->dev_private, drm); if (err < 0) { dev_err(client->dev, "failed to register client: %d\n", err); - goto detach; + goto disable_rpm; } return 0; -detach: +disable_rpm: + pm_runtime_dont_use_autosuspend(client->dev); + pm_runtime_force_suspend(client->dev); + host1x_client_iommu_detach(client); free: host1x_syncpt_put(client->syncpts[0]); @@ -93,10 +115,15 @@ static int gr3d_exit(struct host1x_client *client) if (err < 0) return err; + pm_runtime_dont_use_autosuspend(client->dev); + pm_runtime_force_suspend(client->dev); + host1x_client_iommu_detach(client); host1x_syncpt_put(client->syncpts[0]); host1x_channel_put(gr3d->channel); + gr3d->channel = NULL; + return 0; } @@ -155,14 +182,20 @@ static const struct tegra_drm_client_ops gr3d_ops = { static const struct gr3d_soc tegra20_gr3d_soc = { .version = 0x20, + .num_clocks = 1, + .num_resets = 2, }; static const struct gr3d_soc tegra30_gr3d_soc = { .version = 0x30, + .num_clocks = 2, + .num_resets = 4, }; static const struct gr3d_soc tegra114_gr3d_soc = { .version = 0x35, + .num_clocks = 1, + .num_resets = 2, }; static const struct of_device_id tegra_gr3d_match[] = { @@ -278,69 +311,216 @@ static const u32 gr3d_addr_regs[] = { GR3D_GLOBAL_SAMP23SURFADDR(15), }; -static int gr3d_probe(struct platform_device *pdev) +static int gr3d_power_up_legacy_domain(struct device *dev, const char *name, + unsigned int id) { - struct device_node *np = pdev->dev.of_node; - struct host1x_syncpt **syncpts; - struct gr3d *gr3d; + struct gr3d *gr3d = dev_get_drvdata(dev); + struct reset_control *reset; + struct clk *clk; unsigned int i; int err; - gr3d = devm_kzalloc(&pdev->dev, sizeof(*gr3d), GFP_KERNEL); - if (!gr3d) - return -ENOMEM; - - gr3d->soc = of_device_get_match_data(&pdev->dev); + /* + * Tegra20 device-tree doesn't specify 3d clock name and there is only + * one clock for Tegra20. Tegra30+ device-trees always specified names + * for the clocks. + */ + if (gr3d->nclocks == 1) { + if (id == TEGRA_POWERGATE_3D1) + return 0; + + clk = gr3d->clocks[0].clk; + } else { + for (i = 0; i < gr3d->nclocks; i++) { + if (WARN_ON(!gr3d->clocks[i].id)) + continue; + + if (!strcmp(gr3d->clocks[i].id, name)) { + clk = gr3d->clocks[i].clk; + break; + } + } - syncpts = devm_kzalloc(&pdev->dev, sizeof(*syncpts), GFP_KERNEL); - if (!syncpts) - return -ENOMEM; + if (WARN_ON(i == gr3d->nclocks)) + return -EINVAL; + } - gr3d->clk = devm_clk_get(&pdev->dev, NULL); - if (IS_ERR(gr3d->clk)) { - dev_err(&pdev->dev, "cannot get clock\n"); - return PTR_ERR(gr3d->clk); + /* + * We use array of resets, which includes MC resets, and MC + * reset shouldn't be asserted while hardware is gated because + * MC flushing will fail for gated hardware. Hence for legacy + * PD we request the individual reset separately. + */ + reset = reset_control_get_exclusive_released(dev, name); + if (IS_ERR(reset)) + return PTR_ERR(reset); + + err = reset_control_acquire(reset); + if (err) { + dev_err(dev, "failed to acquire %s reset: %d\n", name, err); + } else { + err = tegra_powergate_sequence_power_up(id, clk, reset); + reset_control_release(reset); } - gr3d->rst = devm_reset_control_get(&pdev->dev, "3d"); - if (IS_ERR(gr3d->rst)) { - dev_err(&pdev->dev, "cannot get reset\n"); - return PTR_ERR(gr3d->rst); + reset_control_put(reset); + if (err) + return err; + + /* + * tegra_powergate_sequence_power_up() leaves clocks enabled, + * while GENPD not. Hence keep clock-enable balanced. + */ + clk_disable_unprepare(clk); + + return 0; +} + +static void gr3d_del_link(void *link) +{ + device_link_del(link); +} + +static int gr3d_init_power(struct device *dev, struct gr3d *gr3d) +{ + static const char * const opp_genpd_names[] = { "3d0", "3d1", NULL }; + const u32 link_flags = DL_FLAG_STATELESS | DL_FLAG_PM_RUNTIME; + struct device **opp_virt_devs, *pd_dev; + struct device_link *link; + unsigned int i; + int err; + + err = of_count_phandle_with_args(dev->of_node, "power-domains", + "#power-domain-cells"); + if (err < 0) { + if (err != -ENOENT) + return err; + + /* + * Older device-trees don't use GENPD. In this case we should + * toggle power domain manually. + */ + err = gr3d_power_up_legacy_domain(dev, "3d", + TEGRA_POWERGATE_3D); + if (err) + return err; + + err = gr3d_power_up_legacy_domain(dev, "3d2", + TEGRA_POWERGATE_3D1); + if (err) + return err; + + return 0; } - if (of_device_is_compatible(np, "nvidia,tegra30-gr3d")) { - gr3d->clk_secondary = devm_clk_get(&pdev->dev, "3d2"); - if (IS_ERR(gr3d->clk_secondary)) { - dev_err(&pdev->dev, "cannot get secondary clock\n"); - return PTR_ERR(gr3d->clk_secondary); + /* + * The PM domain core automatically attaches a single power domain, + * otherwise it skips attaching completely. We have a single domain + * on Tegra20 and two domains on Tegra30+. + */ + if (dev->pm_domain) + return 0; + + err = devm_pm_opp_attach_genpd(dev, opp_genpd_names, &opp_virt_devs); + if (err) + return err; + + for (i = 0; opp_genpd_names[i]; i++) { + pd_dev = opp_virt_devs[i]; + if (!pd_dev) { + dev_err(dev, "failed to get %s power domain\n", + opp_genpd_names[i]); + return -EINVAL; } - gr3d->rst_secondary = devm_reset_control_get(&pdev->dev, - "3d2"); - if (IS_ERR(gr3d->rst_secondary)) { - dev_err(&pdev->dev, "cannot get secondary reset\n"); - return PTR_ERR(gr3d->rst_secondary); + link = device_link_add(dev, pd_dev, link_flags); + if (!link) { + dev_err(dev, "failed to link to %s\n", dev_name(pd_dev)); + return -EINVAL; } + + err = devm_add_action_or_reset(dev, gr3d_del_link, link); + if (err) + return err; } - err = tegra_powergate_sequence_power_up(TEGRA_POWERGATE_3D, gr3d->clk, - gr3d->rst); + return 0; +} + +static int gr3d_get_clocks(struct device *dev, struct gr3d *gr3d) +{ + int err; + + err = devm_clk_bulk_get_all(dev, &gr3d->clocks); if (err < 0) { - dev_err(&pdev->dev, "failed to power up 3D unit\n"); + dev_err(dev, "failed to get clock: %d\n", err); return err; } + gr3d->nclocks = err; - if (gr3d->clk_secondary) { - err = tegra_powergate_sequence_power_up(TEGRA_POWERGATE_3D1, - gr3d->clk_secondary, - gr3d->rst_secondary); - if (err < 0) { - dev_err(&pdev->dev, - "failed to power up secondary 3D unit\n"); - return err; - } + if (gr3d->nclocks != gr3d->soc->num_clocks) { + dev_err(dev, "invalid number of clocks: %u\n", gr3d->nclocks); + return -ENOENT; + } + + return 0; +} + +static int gr3d_get_resets(struct device *dev, struct gr3d *gr3d) +{ + int err; + + gr3d->resets[RST_MC].id = "mc"; + gr3d->resets[RST_MC2].id = "mc2"; + gr3d->resets[RST_GR3D].id = "3d"; + gr3d->resets[RST_GR3D2].id = "3d2"; + gr3d->nresets = gr3d->soc->num_resets; + + err = devm_reset_control_bulk_get_optional_exclusive_released( + dev, gr3d->nresets, gr3d->resets); + if (err) { + dev_err(dev, "failed to get reset: %d\n", err); + return err; } + if (WARN_ON(!gr3d->resets[RST_GR3D].rstc) || + WARN_ON(!gr3d->resets[RST_GR3D2].rstc && gr3d->nresets == 4)) + return -ENOENT; + + return 0; +} + +static int gr3d_probe(struct platform_device *pdev) +{ + struct host1x_syncpt **syncpts; + struct gr3d *gr3d; + unsigned int i; + int err; + + gr3d = devm_kzalloc(&pdev->dev, sizeof(*gr3d), GFP_KERNEL); + if (!gr3d) + return -ENOMEM; + + platform_set_drvdata(pdev, gr3d); + + gr3d->soc = of_device_get_match_data(&pdev->dev); + + syncpts = devm_kzalloc(&pdev->dev, sizeof(*syncpts), GFP_KERNEL); + if (!syncpts) + return -ENOMEM; + + err = gr3d_get_clocks(&pdev->dev, gr3d); + if (err) + return err; + + err = gr3d_get_resets(&pdev->dev, gr3d); + if (err) + return err; + + err = gr3d_init_power(&pdev->dev, gr3d); + if (err) + return err; + INIT_LIST_HEAD(&gr3d->client.base.list); gr3d->client.base.ops = &gr3d_client_ops; gr3d->client.base.dev = &pdev->dev; @@ -352,6 +532,10 @@ static int gr3d_probe(struct platform_device *pdev) gr3d->client.version = gr3d->soc->version; gr3d->client.ops = &gr3d_ops; + err = devm_tegra_core_dev_init_opp_table_common(&pdev->dev); + if (err) + return err; + err = host1x_client_register(&gr3d->client.base); if (err < 0) { dev_err(&pdev->dev, "failed to register host1x client: %d\n", @@ -363,8 +547,6 @@ static int gr3d_probe(struct platform_device *pdev) for (i = 0; i < ARRAY_SIZE(gr3d_addr_regs); i++) set_bit(gr3d_addr_regs[i], gr3d->addr_regs); - platform_set_drvdata(pdev, gr3d); - return 0; } @@ -380,23 +562,80 @@ static int gr3d_remove(struct platform_device *pdev) return err; } - if (gr3d->clk_secondary) { - reset_control_assert(gr3d->rst_secondary); - tegra_powergate_power_off(TEGRA_POWERGATE_3D1); - clk_disable_unprepare(gr3d->clk_secondary); + return 0; +} + +static int __maybe_unused gr3d_runtime_suspend(struct device *dev) +{ + struct gr3d *gr3d = dev_get_drvdata(dev); + int err; + + host1x_channel_stop(gr3d->channel); + + err = reset_control_bulk_assert(gr3d->nresets, gr3d->resets); + if (err) { + dev_err(dev, "failed to assert reset: %d\n", err); + return err; + } + + usleep_range(10, 20); + + /* + * Older device-trees don't specify MC resets and power-gating can't + * be done safely in that case. Hence we will keep the power ungated + * for older DTBs. For newer DTBs, GENPD will perform the power-gating. + */ + + clk_bulk_disable_unprepare(gr3d->nclocks, gr3d->clocks); + reset_control_bulk_release(gr3d->nresets, gr3d->resets); + + return 0; +} + +static int __maybe_unused gr3d_runtime_resume(struct device *dev) +{ + struct gr3d *gr3d = dev_get_drvdata(dev); + int err; + + err = reset_control_bulk_acquire(gr3d->nresets, gr3d->resets); + if (err) { + dev_err(dev, "failed to acquire reset: %d\n", err); + return err; + } + + err = clk_bulk_prepare_enable(gr3d->nclocks, gr3d->clocks); + if (err) { + dev_err(dev, "failed to enable clock: %d\n", err); + goto release_reset; } - reset_control_assert(gr3d->rst); - tegra_powergate_power_off(TEGRA_POWERGATE_3D); - clk_disable_unprepare(gr3d->clk); + err = reset_control_bulk_deassert(gr3d->nresets, gr3d->resets); + if (err) { + dev_err(dev, "failed to deassert reset: %d\n", err); + goto disable_clk; + } return 0; + +disable_clk: + clk_bulk_disable_unprepare(gr3d->nclocks, gr3d->clocks); +release_reset: + reset_control_bulk_release(gr3d->nresets, gr3d->resets); + + return err; } +static const struct dev_pm_ops tegra_gr3d_pm = { + SET_RUNTIME_PM_OPS(gr3d_runtime_suspend, gr3d_runtime_resume, NULL) + SET_SYSTEM_SLEEP_PM_OPS(pm_runtime_force_suspend, + pm_runtime_force_resume) +}; + struct platform_driver tegra_gr3d_driver = { .driver = { .name = "tegra-gr3d", .of_match_table = tegra_gr3d_match, + .pm = &tegra_gr3d_pm, }, .probe = gr3d_probe, .remove = gr3d_remove, diff --git a/drivers/gpu/drm/tegra/hdmi.c b/drivers/gpu/drm/tegra/hdmi.c index e5d2a4026028..8845af5d325f 100644 --- a/drivers/gpu/drm/tegra/hdmi.c +++ b/drivers/gpu/drm/tegra/hdmi.c @@ -11,10 +11,14 @@ #include <linux/math64.h> #include <linux/module.h> #include <linux/of_device.h> +#include <linux/pm_opp.h> #include <linux/pm_runtime.h> #include <linux/regulator/consumer.h> #include <linux/reset.h> +#include <soc/tegra/common.h> +#include <sound/hdmi-codec.h> + #include <drm/drm_atomic_helper.h> #include <drm/drm_crtc.h> #include <drm/drm_debugfs.h> @@ -78,6 +82,9 @@ struct tegra_hdmi { bool dvi; struct drm_info_list *debugfs_files; + + struct platform_device *audio_pdev; + struct mutex audio_lock; }; static inline struct tegra_hdmi * @@ -360,6 +367,18 @@ static const struct tmds_config tegra124_tmds_config[] = { }, }; +static void tegra_hdmi_audio_lock(struct tegra_hdmi *hdmi) +{ + mutex_lock(&hdmi->audio_lock); + disable_irq(hdmi->irq); +} + +static void tegra_hdmi_audio_unlock(struct tegra_hdmi *hdmi) +{ + enable_irq(hdmi->irq); + mutex_unlock(&hdmi->audio_lock); +} + static int tegra_hdmi_get_audio_config(unsigned int audio_freq, unsigned int pix_clock, struct tegra_hdmi_audio_config *config) @@ -829,6 +848,23 @@ static void tegra_hdmi_setup_tmds(struct tegra_hdmi *hdmi, HDMI_NV_PDISP_SOR_IO_PEAK_CURRENT); } +static int tegra_hdmi_reconfigure_audio(struct tegra_hdmi *hdmi) +{ + int err; + + err = tegra_hdmi_setup_audio(hdmi); + if (err < 0) { + tegra_hdmi_disable_audio_infoframe(hdmi); + tegra_hdmi_disable_audio(hdmi); + } else { + tegra_hdmi_setup_audio_infoframe(hdmi); + tegra_hdmi_enable_audio_infoframe(hdmi); + tegra_hdmi_enable_audio(hdmi); + } + + return err; +} + static bool tegra_output_is_hdmi(struct tegra_output *output) { struct edid *edid; @@ -1135,6 +1171,8 @@ static void tegra_hdmi_encoder_disable(struct drm_encoder *encoder) u32 value; int err; + tegra_hdmi_audio_lock(hdmi); + /* * The following accesses registers of the display controller, so make * sure it's only executed when the output is attached to one. @@ -1159,6 +1197,10 @@ static void tegra_hdmi_encoder_disable(struct drm_encoder *encoder) tegra_hdmi_writel(hdmi, 0, HDMI_NV_PDISP_INT_ENABLE); tegra_hdmi_writel(hdmi, 0, HDMI_NV_PDISP_INT_MASK); + hdmi->pixel_clock = 0; + + tegra_hdmi_audio_unlock(hdmi); + err = host1x_client_suspend(&hdmi->client); if (err < 0) dev_err(hdmi->dev, "failed to suspend: %d\n", err); @@ -1182,6 +1224,8 @@ static void tegra_hdmi_encoder_enable(struct drm_encoder *encoder) return; } + tegra_hdmi_audio_lock(hdmi); + /* * Enable and unmask the HDA codec SCRATCH0 register interrupt. This * is used for interoperability between the HDA codec driver and the @@ -1195,7 +1239,7 @@ static void tegra_hdmi_encoder_enable(struct drm_encoder *encoder) h_back_porch = mode->htotal - mode->hsync_end; h_front_porch = mode->hsync_start - mode->hdisplay; - err = clk_set_rate(hdmi->clk, hdmi->pixel_clock); + err = dev_pm_opp_set_rate(hdmi->dev, hdmi->pixel_clock); if (err < 0) { dev_err(hdmi->dev, "failed to set HDMI clock frequency: %d\n", err); @@ -1387,6 +1431,8 @@ static void tegra_hdmi_encoder_enable(struct drm_encoder *encoder) } /* TODO: add HDCP support */ + + tegra_hdmi_audio_unlock(hdmi); } static int @@ -1416,6 +1462,91 @@ static const struct drm_encoder_helper_funcs tegra_hdmi_encoder_helper_funcs = { .atomic_check = tegra_hdmi_encoder_atomic_check, }; +static int tegra_hdmi_hw_params(struct device *dev, void *data, + struct hdmi_codec_daifmt *fmt, + struct hdmi_codec_params *hparms) +{ + struct tegra_hdmi *hdmi = data; + int ret = 0; + + tegra_hdmi_audio_lock(hdmi); + + hdmi->format.sample_rate = hparms->sample_rate; + hdmi->format.channels = hparms->channels; + + if (hdmi->pixel_clock && !hdmi->dvi) + ret = tegra_hdmi_reconfigure_audio(hdmi); + + tegra_hdmi_audio_unlock(hdmi); + + return ret; +} + +static int tegra_hdmi_audio_startup(struct device *dev, void *data) +{ + struct tegra_hdmi *hdmi = data; + int ret; + + ret = host1x_client_resume(&hdmi->client); + if (ret < 0) + dev_err(hdmi->dev, "failed to resume: %d\n", ret); + + return ret; +} + +static void tegra_hdmi_audio_shutdown(struct device *dev, void *data) +{ + struct tegra_hdmi *hdmi = data; + int ret; + + tegra_hdmi_audio_lock(hdmi); + + hdmi->format.sample_rate = 0; + hdmi->format.channels = 0; + + tegra_hdmi_audio_unlock(hdmi); + + ret = host1x_client_suspend(&hdmi->client); + if (ret < 0) + dev_err(hdmi->dev, "failed to suspend: %d\n", ret); +} + +static const struct hdmi_codec_ops tegra_hdmi_codec_ops = { + .hw_params = tegra_hdmi_hw_params, + .audio_startup = tegra_hdmi_audio_startup, + .audio_shutdown = tegra_hdmi_audio_shutdown, +}; + +static int tegra_hdmi_codec_register(struct tegra_hdmi *hdmi) +{ + struct hdmi_codec_pdata codec_data = {}; + + if (hdmi->config->has_hda) + return 0; + + codec_data.ops = &tegra_hdmi_codec_ops; + codec_data.data = hdmi; + codec_data.spdif = 1; + + hdmi->audio_pdev = platform_device_register_data(hdmi->dev, + HDMI_CODEC_DRV_NAME, + PLATFORM_DEVID_AUTO, + &codec_data, + sizeof(codec_data)); + if (IS_ERR(hdmi->audio_pdev)) + return PTR_ERR(hdmi->audio_pdev); + + hdmi->format.channels = 2; + + return 0; +} + +static void tegra_hdmi_codec_unregister(struct tegra_hdmi *hdmi) +{ + if (hdmi->audio_pdev) + platform_device_unregister(hdmi->audio_pdev); +} + static int tegra_hdmi_init(struct host1x_client *client) { struct tegra_hdmi *hdmi = host1x_client_to_hdmi(client); @@ -1453,28 +1584,47 @@ static int tegra_hdmi_init(struct host1x_client *client) if (err < 0) { dev_err(client->dev, "failed to enable HDMI regulator: %d\n", err); - return err; + goto output_exit; } err = regulator_enable(hdmi->pll); if (err < 0) { dev_err(hdmi->dev, "failed to enable PLL regulator: %d\n", err); - return err; + goto disable_hdmi; } err = regulator_enable(hdmi->vdd); if (err < 0) { dev_err(hdmi->dev, "failed to enable VDD regulator: %d\n", err); - return err; + goto disable_pll; + } + + err = tegra_hdmi_codec_register(hdmi); + if (err < 0) { + dev_err(hdmi->dev, "failed to register audio codec: %d\n", err); + goto disable_vdd; } return 0; + +disable_vdd: + regulator_disable(hdmi->vdd); +disable_pll: + regulator_disable(hdmi->pll); +disable_hdmi: + regulator_disable(hdmi->hdmi); +output_exit: + tegra_output_exit(&hdmi->output); + + return err; } static int tegra_hdmi_exit(struct host1x_client *client) { struct tegra_hdmi *hdmi = host1x_client_to_hdmi(client); + tegra_hdmi_codec_unregister(hdmi); + tegra_output_exit(&hdmi->output); regulator_disable(hdmi->vdd); @@ -1599,7 +1749,6 @@ static irqreturn_t tegra_hdmi_irq(int irq, void *data) { struct tegra_hdmi *hdmi = data; u32 value; - int err; value = tegra_hdmi_readl(hdmi, HDMI_NV_PDISP_INT_STATUS); tegra_hdmi_writel(hdmi, value, HDMI_NV_PDISP_INT_STATUS); @@ -1614,16 +1763,7 @@ static irqreturn_t tegra_hdmi_irq(int irq, void *data) format = value & SOR_AUDIO_HDA_CODEC_SCRATCH0_FMT_MASK; tegra_hda_parse_format(format, &hdmi->format); - - err = tegra_hdmi_setup_audio(hdmi); - if (err < 0) { - tegra_hdmi_disable_audio_infoframe(hdmi); - tegra_hdmi_disable_audio(hdmi); - } else { - tegra_hdmi_setup_audio_infoframe(hdmi); - tegra_hdmi_enable_audio_infoframe(hdmi); - tegra_hdmi_enable_audio(hdmi); - } + tegra_hdmi_reconfigure_audio(hdmi); } else { tegra_hdmi_disable_audio_infoframe(hdmi); tegra_hdmi_disable_audio(hdmi); @@ -1651,6 +1791,8 @@ static int tegra_hdmi_probe(struct platform_device *pdev) hdmi->stereo = false; hdmi->dvi = false; + mutex_init(&hdmi->audio_lock); + hdmi->clk = devm_clk_get(&pdev->dev, NULL); if (IS_ERR(hdmi->clk)) { dev_err(&pdev->dev, "failed to get clock\n"); @@ -1732,7 +1874,14 @@ static int tegra_hdmi_probe(struct platform_device *pdev) } platform_set_drvdata(pdev, hdmi); - pm_runtime_enable(&pdev->dev); + + err = devm_pm_runtime_enable(&pdev->dev); + if (err) + return err; + + err = devm_tegra_core_dev_init_opp_table_common(&pdev->dev); + if (err) + return err; INIT_LIST_HEAD(&hdmi->client.list); hdmi->client.ops = &hdmi_client_ops; @@ -1753,8 +1902,6 @@ static int tegra_hdmi_remove(struct platform_device *pdev) struct tegra_hdmi *hdmi = platform_get_drvdata(pdev); int err; - pm_runtime_disable(&pdev->dev); - err = host1x_client_unregister(&hdmi->client); if (err < 0) { dev_err(&pdev->dev, "failed to unregister host1x client: %d\n", diff --git a/drivers/gpu/drm/tegra/hub.h b/drivers/gpu/drm/tegra/hub.h index 3efa1be07ff8..23c4b2115ed1 100644 --- a/drivers/gpu/drm/tegra/hub.h +++ b/drivers/gpu/drm/tegra/hub.h @@ -72,7 +72,6 @@ to_tegra_display_hub_state(struct drm_private_state *priv) return container_of(priv, struct tegra_display_hub_state, base); } -struct tegra_dc; struct tegra_plane; int tegra_display_hub_prepare(struct tegra_display_hub *hub); diff --git a/drivers/gpu/drm/tegra/nvdec.c b/drivers/gpu/drm/tegra/nvdec.c new file mode 100644 index 000000000000..79e1e88203cf --- /dev/null +++ b/drivers/gpu/drm/tegra/nvdec.c @@ -0,0 +1,466 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (c) 2015-2021, NVIDIA Corporation. + */ + +#include <linux/clk.h> +#include <linux/delay.h> +#include <linux/host1x.h> +#include <linux/iommu.h> +#include <linux/module.h> +#include <linux/of.h> +#include <linux/of_device.h> +#include <linux/of_platform.h> +#include <linux/platform_device.h> +#include <linux/pm_runtime.h> +#include <linux/reset.h> + +#include <soc/tegra/pmc.h> + +#include "drm.h" +#include "falcon.h" +#include "vic.h" + +struct nvdec_config { + const char *firmware; + unsigned int version; + bool supports_sid; +}; + +struct nvdec { + struct falcon falcon; + + void __iomem *regs; + struct tegra_drm_client client; + struct host1x_channel *channel; + struct device *dev; + struct clk *clk; + + /* Platform configuration */ + const struct nvdec_config *config; +}; + +static inline struct nvdec *to_nvdec(struct tegra_drm_client *client) +{ + return container_of(client, struct nvdec, client); +} + +static inline void nvdec_writel(struct nvdec *nvdec, u32 value, + unsigned int offset) +{ + writel(value, nvdec->regs + offset); +} + +static int nvdec_boot(struct nvdec *nvdec) +{ +#ifdef CONFIG_IOMMU_API + struct iommu_fwspec *spec = dev_iommu_fwspec_get(nvdec->dev); +#endif + int err; + +#ifdef CONFIG_IOMMU_API + if (nvdec->config->supports_sid && spec) { + u32 value; + + value = TRANSCFG_ATT(1, TRANSCFG_SID_FALCON) | TRANSCFG_ATT(0, TRANSCFG_SID_HW); + nvdec_writel(nvdec, value, VIC_TFBIF_TRANSCFG); + + if (spec->num_ids > 0) { + value = spec->ids[0] & 0xffff; + + nvdec_writel(nvdec, value, VIC_THI_STREAMID0); + nvdec_writel(nvdec, value, VIC_THI_STREAMID1); + } + } +#endif + + err = falcon_boot(&nvdec->falcon); + if (err < 0) + return err; + + err = falcon_wait_idle(&nvdec->falcon); + if (err < 0) { + dev_err(nvdec->dev, "falcon boot timed out\n"); + return err; + } + + return 0; +} + +static int nvdec_init(struct host1x_client *client) +{ + struct tegra_drm_client *drm = host1x_to_drm_client(client); + struct drm_device *dev = dev_get_drvdata(client->host); + struct tegra_drm *tegra = dev->dev_private; + struct nvdec *nvdec = to_nvdec(drm); + int err; + + err = host1x_client_iommu_attach(client); + if (err < 0 && err != -ENODEV) { + dev_err(nvdec->dev, "failed to attach to domain: %d\n", err); + return err; + } + + nvdec->channel = host1x_channel_request(client); + if (!nvdec->channel) { + err = -ENOMEM; + goto detach; + } + + client->syncpts[0] = host1x_syncpt_request(client, 0); + if (!client->syncpts[0]) { + err = -ENOMEM; + goto free_channel; + } + + pm_runtime_enable(client->dev); + pm_runtime_use_autosuspend(client->dev); + pm_runtime_set_autosuspend_delay(client->dev, 500); + + err = tegra_drm_register_client(tegra, drm); + if (err < 0) + goto disable_rpm; + + /* + * Inherit the DMA parameters (such as maximum segment size) from the + * parent host1x device. + */ + client->dev->dma_parms = client->host->dma_parms; + + return 0; + +disable_rpm: + pm_runtime_dont_use_autosuspend(client->dev); + pm_runtime_force_suspend(client->dev); + + host1x_syncpt_put(client->syncpts[0]); +free_channel: + host1x_channel_put(nvdec->channel); +detach: + host1x_client_iommu_detach(client); + + return err; +} + +static int nvdec_exit(struct host1x_client *client) +{ + struct tegra_drm_client *drm = host1x_to_drm_client(client); + struct drm_device *dev = dev_get_drvdata(client->host); + struct tegra_drm *tegra = dev->dev_private; + struct nvdec *nvdec = to_nvdec(drm); + int err; + + /* avoid a dangling pointer just in case this disappears */ + client->dev->dma_parms = NULL; + + err = tegra_drm_unregister_client(tegra, drm); + if (err < 0) + return err; + + pm_runtime_dont_use_autosuspend(client->dev); + pm_runtime_force_suspend(client->dev); + + host1x_syncpt_put(client->syncpts[0]); + host1x_channel_put(nvdec->channel); + host1x_client_iommu_detach(client); + + nvdec->channel = NULL; + + if (client->group) { + dma_unmap_single(nvdec->dev, nvdec->falcon.firmware.phys, + nvdec->falcon.firmware.size, DMA_TO_DEVICE); + tegra_drm_free(tegra, nvdec->falcon.firmware.size, + nvdec->falcon.firmware.virt, + nvdec->falcon.firmware.iova); + } else { + dma_free_coherent(nvdec->dev, nvdec->falcon.firmware.size, + nvdec->falcon.firmware.virt, + nvdec->falcon.firmware.iova); + } + + return 0; +} + +static const struct host1x_client_ops nvdec_client_ops = { + .init = nvdec_init, + .exit = nvdec_exit, +}; + +static int nvdec_load_firmware(struct nvdec *nvdec) +{ + struct host1x_client *client = &nvdec->client.base; + struct tegra_drm *tegra = nvdec->client.drm; + dma_addr_t iova; + size_t size; + void *virt; + int err; + + if (nvdec->falcon.firmware.virt) + return 0; + + err = falcon_read_firmware(&nvdec->falcon, nvdec->config->firmware); + if (err < 0) + return err; + + size = nvdec->falcon.firmware.size; + + if (!client->group) { + virt = dma_alloc_coherent(nvdec->dev, size, &iova, GFP_KERNEL); + + err = dma_mapping_error(nvdec->dev, iova); + if (err < 0) + return err; + } else { + virt = tegra_drm_alloc(tegra, size, &iova); + } + + nvdec->falcon.firmware.virt = virt; + nvdec->falcon.firmware.iova = iova; + + err = falcon_load_firmware(&nvdec->falcon); + if (err < 0) + goto cleanup; + + /* + * In this case we have received an IOVA from the shared domain, so we + * need to make sure to get the physical address so that the DMA API + * knows what memory pages to flush the cache for. + */ + if (client->group) { + dma_addr_t phys; + + phys = dma_map_single(nvdec->dev, virt, size, DMA_TO_DEVICE); + + err = dma_mapping_error(nvdec->dev, phys); + if (err < 0) + goto cleanup; + + nvdec->falcon.firmware.phys = phys; + } + + return 0; + +cleanup: + if (!client->group) + dma_free_coherent(nvdec->dev, size, virt, iova); + else + tegra_drm_free(tegra, size, virt, iova); + + return err; +} + + +static __maybe_unused int nvdec_runtime_resume(struct device *dev) +{ + struct nvdec *nvdec = dev_get_drvdata(dev); + int err; + + err = clk_prepare_enable(nvdec->clk); + if (err < 0) + return err; + + usleep_range(10, 20); + + err = nvdec_load_firmware(nvdec); + if (err < 0) + goto disable; + + err = nvdec_boot(nvdec); + if (err < 0) + goto disable; + + return 0; + +disable: + clk_disable_unprepare(nvdec->clk); + return err; +} + +static __maybe_unused int nvdec_runtime_suspend(struct device *dev) +{ + struct nvdec *nvdec = dev_get_drvdata(dev); + + host1x_channel_stop(nvdec->channel); + + clk_disable_unprepare(nvdec->clk); + + return 0; +} + +static int nvdec_open_channel(struct tegra_drm_client *client, + struct tegra_drm_context *context) +{ + struct nvdec *nvdec = to_nvdec(client); + + context->channel = host1x_channel_get(nvdec->channel); + if (!context->channel) + return -ENOMEM; + + return 0; +} + +static void nvdec_close_channel(struct tegra_drm_context *context) +{ + host1x_channel_put(context->channel); +} + +static const struct tegra_drm_client_ops nvdec_ops = { + .open_channel = nvdec_open_channel, + .close_channel = nvdec_close_channel, + .submit = tegra_drm_submit, +}; + +#define NVIDIA_TEGRA_210_NVDEC_FIRMWARE "nvidia/tegra210/nvdec.bin" + +static const struct nvdec_config nvdec_t210_config = { + .firmware = NVIDIA_TEGRA_210_NVDEC_FIRMWARE, + .version = 0x21, + .supports_sid = false, +}; + +#define NVIDIA_TEGRA_186_NVDEC_FIRMWARE "nvidia/tegra186/nvdec.bin" + +static const struct nvdec_config nvdec_t186_config = { + .firmware = NVIDIA_TEGRA_186_NVDEC_FIRMWARE, + .version = 0x18, + .supports_sid = true, +}; + +#define NVIDIA_TEGRA_194_NVDEC_FIRMWARE "nvidia/tegra194/nvdec.bin" + +static const struct nvdec_config nvdec_t194_config = { + .firmware = NVIDIA_TEGRA_194_NVDEC_FIRMWARE, + .version = 0x19, + .supports_sid = true, +}; + +static const struct of_device_id tegra_nvdec_of_match[] = { + { .compatible = "nvidia,tegra210-nvdec", .data = &nvdec_t210_config }, + { .compatible = "nvidia,tegra186-nvdec", .data = &nvdec_t186_config }, + { .compatible = "nvidia,tegra194-nvdec", .data = &nvdec_t194_config }, + { }, +}; +MODULE_DEVICE_TABLE(of, tegra_nvdec_of_match); + +static int nvdec_probe(struct platform_device *pdev) +{ + struct device *dev = &pdev->dev; + struct host1x_syncpt **syncpts; + struct nvdec *nvdec; + u32 host_class; + int err; + + /* inherit DMA mask from host1x parent */ + err = dma_coerce_mask_and_coherent(dev, *dev->parent->dma_mask); + if (err < 0) { + dev_err(&pdev->dev, "failed to set DMA mask: %d\n", err); + return err; + } + + nvdec = devm_kzalloc(dev, sizeof(*nvdec), GFP_KERNEL); + if (!nvdec) + return -ENOMEM; + + nvdec->config = of_device_get_match_data(dev); + + syncpts = devm_kzalloc(dev, sizeof(*syncpts), GFP_KERNEL); + if (!syncpts) + return -ENOMEM; + + nvdec->regs = devm_platform_get_and_ioremap_resource(pdev, 0, NULL); + if (IS_ERR(nvdec->regs)) + return PTR_ERR(nvdec->regs); + + nvdec->clk = devm_clk_get(dev, NULL); + if (IS_ERR(nvdec->clk)) { + dev_err(&pdev->dev, "failed to get clock\n"); + return PTR_ERR(nvdec->clk); + } + + err = clk_set_rate(nvdec->clk, ULONG_MAX); + if (err < 0) { + dev_err(&pdev->dev, "failed to set clock rate\n"); + return err; + } + + err = of_property_read_u32(dev->of_node, "nvidia,host1x-class", &host_class); + if (err < 0) + host_class = HOST1X_CLASS_NVDEC; + + nvdec->falcon.dev = dev; + nvdec->falcon.regs = nvdec->regs; + + err = falcon_init(&nvdec->falcon); + if (err < 0) + return err; + + platform_set_drvdata(pdev, nvdec); + + INIT_LIST_HEAD(&nvdec->client.base.list); + nvdec->client.base.ops = &nvdec_client_ops; + nvdec->client.base.dev = dev; + nvdec->client.base.class = host_class; + nvdec->client.base.syncpts = syncpts; + nvdec->client.base.num_syncpts = 1; + nvdec->dev = dev; + + INIT_LIST_HEAD(&nvdec->client.list); + nvdec->client.version = nvdec->config->version; + nvdec->client.ops = &nvdec_ops; + + err = host1x_client_register(&nvdec->client.base); + if (err < 0) { + dev_err(dev, "failed to register host1x client: %d\n", err); + goto exit_falcon; + } + + return 0; + +exit_falcon: + falcon_exit(&nvdec->falcon); + + return err; +} + +static int nvdec_remove(struct platform_device *pdev) +{ + struct nvdec *nvdec = platform_get_drvdata(pdev); + int err; + + err = host1x_client_unregister(&nvdec->client.base); + if (err < 0) { + dev_err(&pdev->dev, "failed to unregister host1x client: %d\n", + err); + return err; + } + + falcon_exit(&nvdec->falcon); + + return 0; +} + +static const struct dev_pm_ops nvdec_pm_ops = { + SET_RUNTIME_PM_OPS(nvdec_runtime_suspend, nvdec_runtime_resume, NULL) + SET_SYSTEM_SLEEP_PM_OPS(pm_runtime_force_suspend, + pm_runtime_force_resume) +}; + +struct platform_driver tegra_nvdec_driver = { + .driver = { + .name = "tegra-nvdec", + .of_match_table = tegra_nvdec_of_match, + .pm = &nvdec_pm_ops + }, + .probe = nvdec_probe, + .remove = nvdec_remove, +}; + +#if IS_ENABLED(CONFIG_ARCH_TEGRA_210_SOC) +MODULE_FIRMWARE(NVIDIA_TEGRA_210_NVDEC_FIRMWARE); +#endif +#if IS_ENABLED(CONFIG_ARCH_TEGRA_186_SOC) +MODULE_FIRMWARE(NVIDIA_TEGRA_186_NVDEC_FIRMWARE); +#endif +#if IS_ENABLED(CONFIG_ARCH_TEGRA_194_SOC) +MODULE_FIRMWARE(NVIDIA_TEGRA_194_NVDEC_FIRMWARE); +#endif diff --git a/drivers/gpu/drm/tegra/plane.c b/drivers/gpu/drm/tegra/plane.c index 16a1cdc28657..321cb1f13da6 100644 --- a/drivers/gpu/drm/tegra/plane.c +++ b/drivers/gpu/drm/tegra/plane.c @@ -74,7 +74,7 @@ tegra_plane_atomic_duplicate_state(struct drm_plane *plane) for (i = 0; i < 3; i++) { copy->iova[i] = DMA_MAPPING_ERROR; - copy->sgt[i] = NULL; + copy->map[i] = NULL; } return ©->base; @@ -138,55 +138,37 @@ const struct drm_plane_funcs tegra_plane_funcs = { static int tegra_dc_pin(struct tegra_dc *dc, struct tegra_plane_state *state) { - struct iommu_domain *domain = iommu_get_domain_for_dev(dc->dev); unsigned int i; int err; for (i = 0; i < state->base.fb->format->num_planes; i++) { struct tegra_bo *bo = tegra_fb_get_plane(state->base.fb, i); - dma_addr_t phys_addr, *phys; - struct sg_table *sgt; + struct host1x_bo_mapping *map; - /* - * If we're not attached to a domain, we already stored the - * physical address when the buffer was allocated. If we're - * part of a group that's shared between all display - * controllers, we've also already mapped the framebuffer - * through the SMMU. In both cases we can short-circuit the - * code below and retrieve the stored IOV address. - */ - if (!domain || dc->client.group) - phys = &phys_addr; - else - phys = NULL; - - sgt = host1x_bo_pin(dc->dev, &bo->base, phys); - if (IS_ERR(sgt)) { - err = PTR_ERR(sgt); + map = host1x_bo_pin(dc->dev, &bo->base, DMA_TO_DEVICE, &dc->client.cache); + if (IS_ERR(map)) { + err = PTR_ERR(map); goto unpin; } - if (sgt) { - err = dma_map_sgtable(dc->dev, sgt, DMA_TO_DEVICE, 0); - if (err) - goto unpin; - + if (!dc->client.group) { /* * The display controller needs contiguous memory, so * fail if the buffer is discontiguous and we fail to * map its SG table to a single contiguous chunk of * I/O virtual memory. */ - if (sgt->nents > 1) { + if (map->chunks > 1) { err = -EINVAL; goto unpin; } - state->iova[i] = sg_dma_address(sgt->sgl); - state->sgt[i] = sgt; + state->iova[i] = map->phys; } else { - state->iova[i] = phys_addr; + state->iova[i] = bo->iova; } + + state->map[i] = map; } return 0; @@ -195,15 +177,9 @@ unpin: dev_err(dc->dev, "failed to map plane %u: %d\n", i, err); while (i--) { - struct tegra_bo *bo = tegra_fb_get_plane(state->base.fb, i); - struct sg_table *sgt = state->sgt[i]; - - if (sgt) - dma_unmap_sgtable(dc->dev, sgt, DMA_TO_DEVICE, 0); - - host1x_bo_unpin(dc->dev, &bo->base, sgt); + host1x_bo_unpin(state->map[i]); state->iova[i] = DMA_MAPPING_ERROR; - state->sgt[i] = NULL; + state->map[i] = NULL; } return err; @@ -214,15 +190,9 @@ static void tegra_dc_unpin(struct tegra_dc *dc, struct tegra_plane_state *state) unsigned int i; for (i = 0; i < state->base.fb->format->num_planes; i++) { - struct tegra_bo *bo = tegra_fb_get_plane(state->base.fb, i); - struct sg_table *sgt = state->sgt[i]; - - if (sgt) - dma_unmap_sgtable(dc->dev, sgt, DMA_TO_DEVICE, 0); - - host1x_bo_unpin(dc->dev, &bo->base, sgt); + host1x_bo_unpin(state->map[i]); state->iova[i] = DMA_MAPPING_ERROR; - state->sgt[i] = NULL; + state->map[i] = NULL; } } @@ -230,11 +200,14 @@ int tegra_plane_prepare_fb(struct drm_plane *plane, struct drm_plane_state *state) { struct tegra_dc *dc = to_tegra_dc(state->crtc); + int err; if (!state->fb) return 0; - drm_gem_plane_helper_prepare_fb(plane, state); + err = drm_gem_plane_helper_prepare_fb(plane, state); + if (err < 0) + return err; return tegra_dc_pin(dc, to_tegra_plane_state(state)); } diff --git a/drivers/gpu/drm/tegra/plane.h b/drivers/gpu/drm/tegra/plane.h index d9470780c803..dfb20712fbd7 100644 --- a/drivers/gpu/drm/tegra/plane.h +++ b/drivers/gpu/drm/tegra/plane.h @@ -43,7 +43,7 @@ struct tegra_plane_legacy_blending_state { struct tegra_plane_state { struct drm_plane_state base; - struct sg_table *sgt[3]; + struct host1x_bo_mapping *map[3]; dma_addr_t iova[3]; struct tegra_bo_tiling tiling; diff --git a/drivers/gpu/drm/tegra/rgb.c b/drivers/gpu/drm/tegra/rgb.c index 606c78a2b988..ff8fce36d2aa 100644 --- a/drivers/gpu/drm/tegra/rgb.c +++ b/drivers/gpu/drm/tegra/rgb.c @@ -17,6 +17,8 @@ struct tegra_rgb { struct tegra_output output; struct tegra_dc *dc; + struct clk *pll_d_out0; + struct clk *pll_d2_out0; struct clk *clk_parent; struct clk *clk; }; @@ -116,13 +118,21 @@ static void tegra_rgb_encoder_enable(struct drm_encoder *encoder) DISP_ORDER_RED_BLUE; tegra_dc_writel(rgb->dc, value, DC_DISP_DISP_INTERFACE_CONTROL); - /* XXX: parameterize? */ - value = SC0_H_QUALIFIER_NONE | SC1_H_QUALIFIER_NONE; - tegra_dc_writel(rgb->dc, value, DC_DISP_SHIFT_CLOCK_OPTIONS); - tegra_dc_commit(rgb->dc); } +static bool tegra_rgb_pll_rate_change_allowed(struct tegra_rgb *rgb) +{ + if (!rgb->pll_d2_out0) + return false; + + if (!clk_is_match(rgb->clk_parent, rgb->pll_d_out0) && + !clk_is_match(rgb->clk_parent, rgb->pll_d2_out0)) + return false; + + return true; +} + static int tegra_rgb_encoder_atomic_check(struct drm_encoder *encoder, struct drm_crtc_state *crtc_state, @@ -151,8 +161,17 @@ tegra_rgb_encoder_atomic_check(struct drm_encoder *encoder, * and hope that the desired frequency can be matched (or at least * matched sufficiently close that the panel will still work). */ - div = ((clk_get_rate(rgb->clk) * 2) / pclk) - 2; - pclk = 0; + if (tegra_rgb_pll_rate_change_allowed(rgb)) { + /* + * Set display controller clock to x2 of PCLK in order to + * produce higher resolution pulse positions. + */ + div = 2; + pclk *= 2; + } else { + div = ((clk_get_rate(rgb->clk) * 2) / pclk) - 2; + pclk = 0; + } err = tegra_dc_state_setup_clock(dc, crtc_state, rgb->clk_parent, pclk, div); @@ -210,6 +229,22 @@ int tegra_dc_rgb_probe(struct tegra_dc *dc) return err; } + rgb->pll_d_out0 = clk_get_sys(NULL, "pll_d_out0"); + if (IS_ERR(rgb->pll_d_out0)) { + err = PTR_ERR(rgb->pll_d_out0); + dev_err(dc->dev, "failed to get pll_d_out0: %d\n", err); + return err; + } + + if (dc->soc->has_pll_d2_out0) { + rgb->pll_d2_out0 = clk_get_sys(NULL, "pll_d2_out0"); + if (IS_ERR(rgb->pll_d2_out0)) { + err = PTR_ERR(rgb->pll_d2_out0); + dev_err(dc->dev, "failed to get pll_d2_out0: %d\n", err); + return err; + } + } + dc->rgb = &rgb->output; return 0; @@ -217,9 +252,15 @@ int tegra_dc_rgb_probe(struct tegra_dc *dc) int tegra_dc_rgb_remove(struct tegra_dc *dc) { + struct tegra_rgb *rgb; + if (!dc->rgb) return 0; + rgb = to_rgb(dc->rgb); + clk_put(rgb->pll_d2_out0); + clk_put(rgb->pll_d_out0); + tegra_output_remove(dc->rgb); dc->rgb = NULL; diff --git a/drivers/gpu/drm/tegra/submit.c b/drivers/gpu/drm/tegra/submit.c index 776f825df52f..6d6dd8c35475 100644 --- a/drivers/gpu/drm/tegra/submit.c +++ b/drivers/gpu/drm/tegra/submit.c @@ -64,33 +64,62 @@ static void gather_bo_put(struct host1x_bo *host_bo) kref_put(&bo->ref, gather_bo_release); } -static struct sg_table * -gather_bo_pin(struct device *dev, struct host1x_bo *host_bo, dma_addr_t *phys) +static struct host1x_bo_mapping * +gather_bo_pin(struct device *dev, struct host1x_bo *bo, enum dma_data_direction direction) { - struct gather_bo *bo = container_of(host_bo, struct gather_bo, base); - struct sg_table *sgt; + struct gather_bo *gather = container_of(bo, struct gather_bo, base); + struct host1x_bo_mapping *map; int err; - sgt = kzalloc(sizeof(*sgt), GFP_KERNEL); - if (!sgt) + map = kzalloc(sizeof(*map), GFP_KERNEL); + if (!map) return ERR_PTR(-ENOMEM); - err = dma_get_sgtable(bo->dev, sgt, bo->gather_data, bo->gather_data_dma, - bo->gather_data_words * 4); - if (err) { - kfree(sgt); - return ERR_PTR(err); + kref_init(&map->ref); + map->bo = host1x_bo_get(bo); + map->direction = direction; + map->dev = dev; + + map->sgt = kzalloc(sizeof(*map->sgt), GFP_KERNEL); + if (!map->sgt) { + err = -ENOMEM; + goto free; } - return sgt; + err = dma_get_sgtable(gather->dev, map->sgt, gather->gather_data, gather->gather_data_dma, + gather->gather_data_words * 4); + if (err) + goto free_sgt; + + err = dma_map_sgtable(dev, map->sgt, direction, 0); + if (err) + goto free_sgt; + + map->phys = sg_dma_address(map->sgt->sgl); + map->size = gather->gather_data_words * 4; + map->chunks = err; + + return map; + +free_sgt: + sg_free_table(map->sgt); + kfree(map->sgt); +free: + kfree(map); + return ERR_PTR(err); } -static void gather_bo_unpin(struct device *dev, struct sg_table *sgt) +static void gather_bo_unpin(struct host1x_bo_mapping *map) { - if (sgt) { - sg_free_table(sgt); - kfree(sgt); - } + if (!map) + return; + + dma_unmap_sgtable(map->dev, map->sgt, map->direction, 0); + sg_free_table(map->sgt); + kfree(map->sgt); + host1x_bo_put(map->bo); + + kfree(map); } static void *gather_bo_mmap(struct host1x_bo *host_bo) @@ -475,8 +504,8 @@ static void release_job(struct host1x_job *job) kfree(job_data->used_mappings); kfree(job_data); - if (pm_runtime_enabled(client->base.dev)) - pm_runtime_put_autosuspend(client->base.dev); + pm_runtime_mark_last_busy(client->base.dev); + pm_runtime_put_autosuspend(client->base.dev); } int tegra_drm_ioctl_channel_submit(struct drm_device *drm, void *data, @@ -560,12 +589,10 @@ int tegra_drm_ioctl_channel_submit(struct drm_device *drm, void *data, } /* Boot engine. */ - if (pm_runtime_enabled(context->client->base.dev)) { - err = pm_runtime_resume_and_get(context->client->base.dev); - if (err < 0) { - SUBMIT_ERR(context, "could not power up engine: %d", err); - goto unpin_job; - } + err = pm_runtime_resume_and_get(context->client->base.dev); + if (err < 0) { + SUBMIT_ERR(context, "could not power up engine: %d", err); + goto unpin_job; } job->user_data = job_data; diff --git a/drivers/gpu/drm/tegra/uapi.c b/drivers/gpu/drm/tegra/uapi.c index 690a339c52ec..9ab9179d2026 100644 --- a/drivers/gpu/drm/tegra/uapi.c +++ b/drivers/gpu/drm/tegra/uapi.c @@ -17,11 +17,7 @@ static void tegra_drm_mapping_release(struct kref *ref) struct tegra_drm_mapping *mapping = container_of(ref, struct tegra_drm_mapping, ref); - if (mapping->sgt) - dma_unmap_sgtable(mapping->dev, mapping->sgt, mapping->direction, - DMA_ATTR_SKIP_CPU_SYNC); - - host1x_bo_unpin(mapping->dev, mapping->bo, mapping->sgt); + host1x_bo_unpin(mapping->map); host1x_bo_put(mapping->bo); kfree(mapping); @@ -159,6 +155,7 @@ int tegra_drm_ioctl_channel_map(struct drm_device *drm, void *data, struct drm_f struct drm_tegra_channel_map *args = data; struct tegra_drm_mapping *mapping; struct tegra_drm_context *context; + enum dma_data_direction direction; int err = 0; if (args->flags & ~DRM_TEGRA_CHANNEL_MAP_READ_WRITE) @@ -180,68 +177,53 @@ int tegra_drm_ioctl_channel_map(struct drm_device *drm, void *data, struct drm_f kref_init(&mapping->ref); - mapping->dev = context->client->base.dev; mapping->bo = tegra_gem_lookup(file, args->handle); if (!mapping->bo) { err = -EINVAL; - goto unlock; + goto free; } - if (context->client->base.group) { - /* IOMMU domain managed directly using IOMMU API */ - host1x_bo_pin(mapping->dev, mapping->bo, &mapping->iova); - } else { - switch (args->flags & DRM_TEGRA_CHANNEL_MAP_READ_WRITE) { - case DRM_TEGRA_CHANNEL_MAP_READ_WRITE: - mapping->direction = DMA_BIDIRECTIONAL; - break; - - case DRM_TEGRA_CHANNEL_MAP_WRITE: - mapping->direction = DMA_FROM_DEVICE; - break; - - case DRM_TEGRA_CHANNEL_MAP_READ: - mapping->direction = DMA_TO_DEVICE; - break; + switch (args->flags & DRM_TEGRA_CHANNEL_MAP_READ_WRITE) { + case DRM_TEGRA_CHANNEL_MAP_READ_WRITE: + direction = DMA_BIDIRECTIONAL; + break; - default: - return -EINVAL; - } + case DRM_TEGRA_CHANNEL_MAP_WRITE: + direction = DMA_FROM_DEVICE; + break; - mapping->sgt = host1x_bo_pin(mapping->dev, mapping->bo, NULL); - if (IS_ERR(mapping->sgt)) { - err = PTR_ERR(mapping->sgt); - goto put_gem; - } + case DRM_TEGRA_CHANNEL_MAP_READ: + direction = DMA_TO_DEVICE; + break; - err = dma_map_sgtable(mapping->dev, mapping->sgt, mapping->direction, - DMA_ATTR_SKIP_CPU_SYNC); - if (err) - goto unpin; + default: + err = -EINVAL; + goto put_gem; + } - mapping->iova = sg_dma_address(mapping->sgt->sgl); + mapping->map = host1x_bo_pin(context->client->base.dev, mapping->bo, direction, NULL); + if (IS_ERR(mapping->map)) { + err = PTR_ERR(mapping->map); + goto put_gem; } + mapping->iova = mapping->map->phys; mapping->iova_end = mapping->iova + host1x_to_tegra_bo(mapping->bo)->gem.size; err = xa_alloc(&context->mappings, &args->mapping, mapping, XA_LIMIT(1, U32_MAX), GFP_KERNEL); if (err < 0) - goto unmap; + goto unpin; mutex_unlock(&fpriv->lock); return 0; -unmap: - if (mapping->sgt) { - dma_unmap_sgtable(mapping->dev, mapping->sgt, mapping->direction, - DMA_ATTR_SKIP_CPU_SYNC); - } unpin: - host1x_bo_unpin(mapping->dev, mapping->bo, mapping->sgt); + host1x_bo_unpin(mapping->map); put_gem: host1x_bo_put(mapping->bo); +free: kfree(mapping); unlock: mutex_unlock(&fpriv->lock); diff --git a/drivers/gpu/drm/tegra/uapi.h b/drivers/gpu/drm/tegra/uapi.h index 12adad770ad3..92ff1e44ff15 100644 --- a/drivers/gpu/drm/tegra/uapi.h +++ b/drivers/gpu/drm/tegra/uapi.h @@ -27,10 +27,9 @@ struct tegra_drm_file { struct tegra_drm_mapping { struct kref ref; - struct device *dev; + struct host1x_bo_mapping *map; struct host1x_bo *bo; - struct sg_table *sgt; - enum dma_data_direction direction; + dma_addr_t iova; dma_addr_t iova_end; }; diff --git a/drivers/gpu/drm/tegra/vic.c b/drivers/gpu/drm/tegra/vic.c index c02010ff2b7f..1e342fa3d27b 100644 --- a/drivers/gpu/drm/tegra/vic.c +++ b/drivers/gpu/drm/tegra/vic.c @@ -5,6 +5,7 @@ #include <linux/clk.h> #include <linux/delay.h> +#include <linux/dma-mapping.h> #include <linux/host1x.h> #include <linux/iommu.h> #include <linux/module.h> @@ -151,9 +152,13 @@ static int vic_init(struct host1x_client *client) goto free_channel; } + pm_runtime_enable(client->dev); + pm_runtime_use_autosuspend(client->dev); + pm_runtime_set_autosuspend_delay(client->dev, 500); + err = tegra_drm_register_client(tegra, drm); if (err < 0) - goto free_syncpt; + goto disable_rpm; /* * Inherit the DMA parameters (such as maximum segment size) from the @@ -163,7 +168,10 @@ static int vic_init(struct host1x_client *client) return 0; -free_syncpt: +disable_rpm: + pm_runtime_dont_use_autosuspend(client->dev); + pm_runtime_force_suspend(client->dev); + host1x_syncpt_put(client->syncpts[0]); free_channel: host1x_channel_put(vic->channel); @@ -188,10 +196,15 @@ static int vic_exit(struct host1x_client *client) if (err < 0) return err; + pm_runtime_dont_use_autosuspend(client->dev); + pm_runtime_force_suspend(client->dev); + host1x_syncpt_put(client->syncpts[0]); host1x_channel_put(vic->channel); host1x_client_iommu_detach(client); + vic->channel = NULL; + if (client->group) { dma_unmap_single(vic->dev, vic->falcon.firmware.phys, vic->falcon.firmware.size, DMA_TO_DEVICE); @@ -232,12 +245,12 @@ static int vic_load_firmware(struct vic *vic) if (!client->group) { virt = dma_alloc_coherent(vic->dev, size, &iova, GFP_KERNEL); - - err = dma_mapping_error(vic->dev, iova); - if (err < 0) - return err; + if (!virt) + return -ENOMEM; } else { virt = tegra_drm_alloc(tegra, size, &iova); + if (IS_ERR(virt)) + return PTR_ERR(virt); } vic->falcon.firmware.virt = virt; @@ -315,6 +328,8 @@ static int vic_runtime_suspend(struct device *dev) struct vic *vic = dev_get_drvdata(dev); int err; + host1x_channel_stop(vic->channel); + err = reset_control_assert(vic->rst); if (err < 0) return err; @@ -330,27 +345,17 @@ static int vic_open_channel(struct tegra_drm_client *client, struct tegra_drm_context *context) { struct vic *vic = to_vic(client); - int err; - - err = pm_runtime_resume_and_get(vic->dev); - if (err < 0) - return err; context->channel = host1x_channel_get(vic->channel); - if (!context->channel) { - pm_runtime_put(vic->dev); + if (!context->channel) return -ENOMEM; - } return 0; } static void vic_close_channel(struct tegra_drm_context *context) { - struct vic *vic = to_vic(context->client); - host1x_channel_put(context->channel); - pm_runtime_put(vic->dev); } static const struct tegra_drm_client_ops vic_ops = { @@ -441,6 +446,12 @@ static int vic_probe(struct platform_device *pdev) return PTR_ERR(vic->clk); } + err = clk_set_rate(vic->clk, ULONG_MAX); + if (err < 0) { + dev_err(&pdev->dev, "failed to set clock rate\n"); + return err; + } + if (!dev->pm_domain) { vic->rst = devm_reset_control_get(dev, "vic"); if (IS_ERR(vic->rst)) { @@ -476,17 +487,8 @@ static int vic_probe(struct platform_device *pdev) goto exit_falcon; } - pm_runtime_enable(&pdev->dev); - if (!pm_runtime_enabled(&pdev->dev)) { - err = vic_runtime_resume(&pdev->dev); - if (err < 0) - goto unregister_client; - } - return 0; -unregister_client: - host1x_client_unregister(&vic->client.base); exit_falcon: falcon_exit(&vic->falcon); @@ -505,11 +507,6 @@ static int vic_remove(struct platform_device *pdev) return err; } - if (pm_runtime_enabled(&pdev->dev)) - pm_runtime_disable(&pdev->dev); - else - vic_runtime_suspend(&pdev->dev); - falcon_exit(&vic->falcon); return 0; @@ -517,6 +514,8 @@ static int vic_remove(struct platform_device *pdev) static const struct dev_pm_ops vic_pm_ops = { SET_RUNTIME_PM_OPS(vic_runtime_suspend, vic_runtime_resume, NULL) + SET_SYSTEM_SLEEP_PM_OPS(pm_runtime_force_suspend, + pm_runtime_force_resume) }; struct platform_driver tegra_vic_driver = { diff --git a/drivers/gpu/host1x/Kconfig b/drivers/gpu/host1x/Kconfig index 6dab94adf25e..6815b4db17c1 100644 --- a/drivers/gpu/host1x/Kconfig +++ b/drivers/gpu/host1x/Kconfig @@ -2,6 +2,7 @@ config TEGRA_HOST1X tristate "NVIDIA Tegra host1x driver" depends on ARCH_TEGRA || (ARM && COMPILE_TEST) + select DMA_SHARED_BUFFER select IOMMU_IOVA help Driver for the NVIDIA Tegra host1x hardware. diff --git a/drivers/gpu/host1x/bus.c b/drivers/gpu/host1x/bus.c index 218e3718fd68..bdee16a0bb8e 100644 --- a/drivers/gpu/host1x/bus.c +++ b/drivers/gpu/host1x/bus.c @@ -5,6 +5,7 @@ */ #include <linux/debugfs.h> +#include <linux/dma-mapping.h> #include <linux/host1x.h> #include <linux/of.h> #include <linux/seq_file.h> @@ -742,6 +743,7 @@ EXPORT_SYMBOL(host1x_driver_unregister); */ void __host1x_client_init(struct host1x_client *client, struct lock_class_key *key) { + host1x_bo_cache_init(&client->cache); INIT_LIST_HEAD(&client->list); __mutex_init(&client->lock, "host1x client lock", key); client->usecount = 0; @@ -761,7 +763,6 @@ EXPORT_SYMBOL(host1x_client_exit); /** * __host1x_client_register() - register a host1x client * @client: host1x client - * @key: lock class key for the client-specific mutex * * Registers a host1x client with each host1x controller instance. Note that * each client will only match their parent host1x controller and will only be @@ -830,6 +831,8 @@ int host1x_client_unregister(struct host1x_client *client) mutex_unlock(&clients_lock); + host1x_bo_cache_destroy(&client->cache); + return 0; } EXPORT_SYMBOL(host1x_client_unregister); @@ -904,3 +907,78 @@ unlock: return err; } EXPORT_SYMBOL(host1x_client_resume); + +struct host1x_bo_mapping *host1x_bo_pin(struct device *dev, struct host1x_bo *bo, + enum dma_data_direction dir, + struct host1x_bo_cache *cache) +{ + struct host1x_bo_mapping *mapping; + + if (cache) { + mutex_lock(&cache->lock); + + list_for_each_entry(mapping, &cache->mappings, entry) { + if (mapping->bo == bo && mapping->direction == dir) { + kref_get(&mapping->ref); + goto unlock; + } + } + } + + mapping = bo->ops->pin(dev, bo, dir); + if (IS_ERR(mapping)) + goto unlock; + + spin_lock(&mapping->bo->lock); + list_add_tail(&mapping->list, &bo->mappings); + spin_unlock(&mapping->bo->lock); + + if (cache) { + INIT_LIST_HEAD(&mapping->entry); + mapping->cache = cache; + + list_add_tail(&mapping->entry, &cache->mappings); + + /* bump reference count to track the copy in the cache */ + kref_get(&mapping->ref); + } + +unlock: + if (cache) + mutex_unlock(&cache->lock); + + return mapping; +} +EXPORT_SYMBOL(host1x_bo_pin); + +static void __host1x_bo_unpin(struct kref *ref) +{ + struct host1x_bo_mapping *mapping = to_host1x_bo_mapping(ref); + + /* + * When the last reference of the mapping goes away, make sure to remove the mapping from + * the cache. + */ + if (mapping->cache) + list_del(&mapping->entry); + + spin_lock(&mapping->bo->lock); + list_del(&mapping->list); + spin_unlock(&mapping->bo->lock); + + mapping->bo->ops->unpin(mapping); +} + +void host1x_bo_unpin(struct host1x_bo_mapping *mapping) +{ + struct host1x_bo_cache *cache = mapping->cache; + + if (cache) + mutex_lock(&cache->lock); + + kref_put(&mapping->ref, __host1x_bo_unpin); + + if (cache) + mutex_unlock(&cache->lock); +} +EXPORT_SYMBOL(host1x_bo_unpin); diff --git a/drivers/gpu/host1x/channel.c b/drivers/gpu/host1x/channel.c index 4cd212bb570d..2a9a3a8d5931 100644 --- a/drivers/gpu/host1x/channel.c +++ b/drivers/gpu/host1x/channel.c @@ -75,6 +75,14 @@ struct host1x_channel *host1x_channel_get_index(struct host1x *host, return ch; } +void host1x_channel_stop(struct host1x_channel *channel) +{ + struct host1x *host = dev_get_drvdata(channel->dev->parent); + + host1x_hw_cdma_stop(host, &channel->cdma); +} +EXPORT_SYMBOL(host1x_channel_stop); + static void release_channel(struct kref *kref) { struct host1x_channel *channel = diff --git a/drivers/gpu/host1x/debug.c b/drivers/gpu/host1x/debug.c index 8a14880c61bb..18d9c8d206e3 100644 --- a/drivers/gpu/host1x/debug.c +++ b/drivers/gpu/host1x/debug.c @@ -7,6 +7,7 @@ */ #include <linux/debugfs.h> +#include <linux/pm_runtime.h> #include <linux/seq_file.h> #include <linux/uaccess.h> @@ -52,6 +53,11 @@ static int show_channel(struct host1x_channel *ch, void *data, bool show_fifo) { struct host1x *m = dev_get_drvdata(ch->dev->parent); struct output *o = data; + int err; + + err = pm_runtime_resume_and_get(m->dev); + if (err < 0) + return err; mutex_lock(&ch->cdma.lock); mutex_lock(&debug_lock); @@ -64,6 +70,8 @@ static int show_channel(struct host1x_channel *ch, void *data, bool show_fifo) mutex_unlock(&debug_lock); mutex_unlock(&ch->cdma.lock); + pm_runtime_put(m->dev); + return 0; } @@ -71,9 +79,14 @@ static void show_syncpts(struct host1x *m, struct output *o) { struct list_head *pos; unsigned int i; + int err; host1x_debug_output(o, "---- syncpts ----\n"); + err = pm_runtime_resume_and_get(m->dev); + if (err < 0) + return; + for (i = 0; i < host1x_syncpt_nb_pts(m); i++) { u32 max = host1x_syncpt_read_max(m->syncpt + i); u32 min = host1x_syncpt_load(m->syncpt + i); @@ -101,6 +114,8 @@ static void show_syncpts(struct host1x *m, struct output *o) base_val); } + pm_runtime_put(m->dev); + host1x_debug_output(o, "\n"); } diff --git a/drivers/gpu/host1x/dev.c b/drivers/gpu/host1x/dev.c index fbb6447b8659..6994f8c0e02e 100644 --- a/drivers/gpu/host1x/dev.c +++ b/drivers/gpu/host1x/dev.c @@ -6,18 +6,26 @@ */ #include <linux/clk.h> +#include <linux/delay.h> #include <linux/dma-mapping.h> #include <linux/io.h> #include <linux/list.h> #include <linux/module.h> #include <linux/of_device.h> #include <linux/of.h> +#include <linux/pm_runtime.h> #include <linux/slab.h> +#include <soc/tegra/common.h> + #define CREATE_TRACE_POINTS #include <trace/events/host1x.h> #undef CREATE_TRACE_POINTS +#if IS_ENABLED(CONFIG_ARM_DMA_USE_IOMMU) +#include <asm/dma-iommu.h> +#endif + #include "bus.h" #include "channel.h" #include "debug.h" @@ -132,6 +140,12 @@ static const struct host1x_sid_entry tegra186_sid_table[] = { .offset = 0x30, .limit = 0x34 }, + { + /* NVDEC */ + .base = 0x1b00, + .offset = 0x30, + .limit = 0x34 + }, }; static const struct host1x_info host1x06_info = { @@ -156,6 +170,18 @@ static const struct host1x_sid_entry tegra194_sid_table[] = { .offset = 0x30, .limit = 0x34 }, + { + /* NVDEC */ + .base = 0x1b00, + .offset = 0x30, + .limit = 0x34 + }, + { + /* NVDEC1 */ + .base = 0x1bc0, + .offset = 0x30, + .limit = 0x34 + }, }; static const struct host1x_info host1x07_info = { @@ -190,6 +216,9 @@ static void host1x_setup_sid_table(struct host1x *host) const struct host1x_info *info = host->info; unsigned int i; + if (!info->has_hypervisor) + return; + for (i = 0; i < info->num_sid_entries; i++) { const struct host1x_sid_entry *entry = &info->sid_table[i]; @@ -238,6 +267,17 @@ static struct iommu_domain *host1x_iommu_attach(struct host1x *host) struct iommu_domain *domain = iommu_get_domain_for_dev(host->dev); int err; +#if IS_ENABLED(CONFIG_ARM_DMA_USE_IOMMU) + if (host->dev->archdata.mapping) { + struct dma_iommu_mapping *mapping = + to_dma_iommu_mapping(host->dev); + arm_iommu_detach_device(host->dev); + arm_iommu_release_mapping(mapping); + + domain = iommu_get_domain_for_dev(host->dev); + } +#endif + /* * We may not always want to enable IOMMU support (for example if the * host1x firewall is already enabled and we don't support addressing @@ -347,6 +387,27 @@ static void host1x_iommu_exit(struct host1x *host) } } +static int host1x_get_resets(struct host1x *host) +{ + int err; + + host->resets[0].id = "mc"; + host->resets[1].id = "host1x"; + host->nresets = ARRAY_SIZE(host->resets); + + err = devm_reset_control_bulk_get_optional_exclusive_released( + host->dev, host->nresets, host->resets); + if (err) { + dev_err(host->dev, "failed to get reset: %d\n", err); + return err; + } + + if (WARN_ON(!host->resets[1].rstc)) + return -ENOENT; + + return 0; +} + static int host1x_probe(struct platform_device *pdev) { struct host1x *host; @@ -386,6 +447,7 @@ static int host1x_probe(struct platform_device *pdev) if (syncpt_irq < 0) return syncpt_irq; + host1x_bo_cache_init(&host->cache); mutex_init(&host->devices_lock); INIT_LIST_HEAD(&host->devices); INIT_LIST_HEAD(&host->list); @@ -423,12 +485,9 @@ static int host1x_probe(struct platform_device *pdev) return err; } - host->rst = devm_reset_control_get(&pdev->dev, "host1x"); - if (IS_ERR(host->rst)) { - err = PTR_ERR(host->rst); - dev_err(&pdev->dev, "failed to get reset: %d\n", err); + err = host1x_get_resets(host); + if (err) return err; - } err = host1x_iommu_init(host); if (err < 0) { @@ -443,22 +502,10 @@ static int host1x_probe(struct platform_device *pdev) goto iommu_exit; } - err = clk_prepare_enable(host->clk); - if (err < 0) { - dev_err(&pdev->dev, "failed to enable clock\n"); - goto free_channels; - } - - err = reset_control_deassert(host->rst); - if (err < 0) { - dev_err(&pdev->dev, "failed to deassert reset: %d\n", err); - goto unprepare_disable; - } - err = host1x_syncpt_init(host); if (err) { dev_err(&pdev->dev, "failed to initialize syncpts\n"); - goto reset_assert; + goto free_channels; } err = host1x_intr_init(host, syncpt_irq); @@ -467,10 +514,18 @@ static int host1x_probe(struct platform_device *pdev) goto deinit_syncpt; } - host1x_debug_init(host); + pm_runtime_enable(&pdev->dev); + + err = devm_tegra_core_dev_init_opp_table_common(&pdev->dev); + if (err) + goto pm_disable; - if (host->info->has_hypervisor) - host1x_setup_sid_table(host); + /* the driver's code isn't ready yet for the dynamic RPM */ + err = pm_runtime_resume_and_get(&pdev->dev); + if (err) + goto pm_disable; + + host1x_debug_init(host); err = host1x_register(host); if (err < 0) @@ -486,13 +541,14 @@ unregister: host1x_unregister(host); deinit_debugfs: host1x_debug_deinit(host); + + pm_runtime_put_sync_suspend(&pdev->dev); +pm_disable: + pm_runtime_disable(&pdev->dev); + host1x_intr_deinit(host); deinit_syncpt: host1x_syncpt_deinit(host); -reset_assert: - reset_control_assert(host->rst); -unprepare_disable: - clk_disable_unprepare(host->clk); free_channels: host1x_channel_list_free(&host->channel_list); iommu_exit: @@ -507,19 +563,94 @@ static int host1x_remove(struct platform_device *pdev) host1x_unregister(host); host1x_debug_deinit(host); + + pm_runtime_force_suspend(&pdev->dev); + host1x_intr_deinit(host); host1x_syncpt_deinit(host); - reset_control_assert(host->rst); - clk_disable_unprepare(host->clk); host1x_iommu_exit(host); + host1x_bo_cache_destroy(&host->cache); return 0; } +static int __maybe_unused host1x_runtime_suspend(struct device *dev) +{ + struct host1x *host = dev_get_drvdata(dev); + int err; + + host1x_intr_stop(host); + host1x_syncpt_save(host); + + err = reset_control_bulk_assert(host->nresets, host->resets); + if (err) { + dev_err(dev, "failed to assert reset: %d\n", err); + goto resume_host1x; + } + + usleep_range(1000, 2000); + + clk_disable_unprepare(host->clk); + reset_control_bulk_release(host->nresets, host->resets); + + return 0; + +resume_host1x: + host1x_setup_sid_table(host); + host1x_syncpt_restore(host); + host1x_intr_start(host); + + return err; +} + +static int __maybe_unused host1x_runtime_resume(struct device *dev) +{ + struct host1x *host = dev_get_drvdata(dev); + int err; + + err = reset_control_bulk_acquire(host->nresets, host->resets); + if (err) { + dev_err(dev, "failed to acquire reset: %d\n", err); + return err; + } + + err = clk_prepare_enable(host->clk); + if (err) { + dev_err(dev, "failed to enable clock: %d\n", err); + goto release_reset; + } + + err = reset_control_bulk_deassert(host->nresets, host->resets); + if (err < 0) { + dev_err(dev, "failed to deassert reset: %d\n", err); + goto disable_clk; + } + + host1x_setup_sid_table(host); + host1x_syncpt_restore(host); + host1x_intr_start(host); + + return 0; + +disable_clk: + clk_disable_unprepare(host->clk); +release_reset: + reset_control_bulk_release(host->nresets, host->resets); + + return err; +} + +static const struct dev_pm_ops host1x_pm_ops = { + SET_RUNTIME_PM_OPS(host1x_runtime_suspend, host1x_runtime_resume, + NULL) + /* TODO: add system suspend-resume once driver will be ready for that */ +}; + static struct platform_driver tegra_host1x_driver = { .driver = { .name = "tegra-host1x", .of_match_table = host1x_of_match, + .pm = &host1x_pm_ops, }, .probe = host1x_probe, .remove = host1x_remove, diff --git a/drivers/gpu/host1x/dev.h b/drivers/gpu/host1x/dev.h index fa6d4bc46e98..ca4b082f0cd4 100644 --- a/drivers/gpu/host1x/dev.h +++ b/drivers/gpu/host1x/dev.h @@ -118,7 +118,8 @@ struct host1x { struct host1x_syncpt_base *bases; struct device *dev; struct clk *clk; - struct reset_control *rst; + struct reset_control_bulk_data resets[2]; + unsigned int nresets; struct iommu_group *group; struct iommu_domain *domain; @@ -149,6 +150,8 @@ struct host1x { struct list_head list; struct device_dma_parameters dma_parms; + + struct host1x_bo_cache cache; }; void host1x_hypervisor_writel(struct host1x *host1x, u32 r, u32 v); diff --git a/drivers/gpu/host1x/hw/channel_hw.c b/drivers/gpu/host1x/hw/channel_hw.c index 1999780a7203..6b40e9af1e88 100644 --- a/drivers/gpu/host1x/hw/channel_hw.c +++ b/drivers/gpu/host1x/hw/channel_hw.c @@ -159,6 +159,27 @@ static void host1x_channel_set_streamid(struct host1x_channel *channel) #endif } +static void host1x_enable_gather_filter(struct host1x_channel *ch) +{ +#if HOST1X_HW >= 6 + struct host1x *host = dev_get_drvdata(ch->dev->parent); + u32 val; + + if (!host->hv_regs) + return; + + val = host1x_hypervisor_readl( + host, HOST1X_HV_CH_KERNEL_FILTER_GBUFFER(ch->id / 32)); + val |= BIT(ch->id % 32); + host1x_hypervisor_writel( + host, val, HOST1X_HV_CH_KERNEL_FILTER_GBUFFER(ch->id / 32)); +#elif HOST1X_HW >= 4 + host1x_ch_writel(ch, + HOST1X_CHANNEL_CHANNELCTRL_KERNEL_FILTER_GBUFFER(1), + HOST1X_CHANNEL_CHANNELCTRL); +#endif +} + static int channel_submit(struct host1x_job *job) { struct host1x_channel *ch = job->channel; @@ -190,6 +211,7 @@ static int channel_submit(struct host1x_job *job) } host1x_channel_set_streamid(ch); + host1x_enable_gather_filter(ch); /* begin a CDMA submit */ err = host1x_cdma_begin(&ch->cdma, job); @@ -249,27 +271,6 @@ error: return err; } -static void enable_gather_filter(struct host1x *host, - struct host1x_channel *ch) -{ -#if HOST1X_HW >= 6 - u32 val; - - if (!host->hv_regs) - return; - - val = host1x_hypervisor_readl( - host, HOST1X_HV_CH_KERNEL_FILTER_GBUFFER(ch->id / 32)); - val |= BIT(ch->id % 32); - host1x_hypervisor_writel( - host, val, HOST1X_HV_CH_KERNEL_FILTER_GBUFFER(ch->id / 32)); -#elif HOST1X_HW >= 4 - host1x_ch_writel(ch, - HOST1X_CHANNEL_CHANNELCTRL_KERNEL_FILTER_GBUFFER(1), - HOST1X_CHANNEL_CHANNELCTRL); -#endif -} - static int host1x_channel_init(struct host1x_channel *ch, struct host1x *dev, unsigned int index) { @@ -278,7 +279,6 @@ static int host1x_channel_init(struct host1x_channel *ch, struct host1x *dev, #else ch->regs = dev->regs + index * 0x100; #endif - enable_gather_filter(dev, ch); return 0; } diff --git a/drivers/gpu/host1x/intr.c b/drivers/gpu/host1x/intr.c index 45b6be927ec4..965ba21818b1 100644 --- a/drivers/gpu/host1x/intr.c +++ b/drivers/gpu/host1x/intr.c @@ -297,14 +297,11 @@ int host1x_intr_init(struct host1x *host, unsigned int irq_sync) "host1x_sp_%02u", id); } - host1x_intr_start(host); - return 0; } void host1x_intr_deinit(struct host1x *host) { - host1x_intr_stop(host); } void host1x_intr_start(struct host1x *host) diff --git a/drivers/gpu/host1x/job.c b/drivers/gpu/host1x/job.c index 0eef6df7c89e..5e8c183167b7 100644 --- a/drivers/gpu/host1x/job.c +++ b/drivers/gpu/host1x/job.c @@ -134,20 +134,20 @@ EXPORT_SYMBOL(host1x_job_add_wait); static unsigned int pin_job(struct host1x *host, struct host1x_job *job) { + unsigned long mask = HOST1X_RELOC_READ | HOST1X_RELOC_WRITE; struct host1x_client *client = job->client; struct device *dev = client->dev; struct host1x_job_gather *g; - struct iommu_domain *domain; - struct sg_table *sgt; unsigned int i; int err; - domain = iommu_get_domain_for_dev(dev); job->num_unpins = 0; for (i = 0; i < job->num_relocs; i++) { struct host1x_reloc *reloc = &job->relocs[i]; - dma_addr_t phys_addr, *phys; + enum dma_data_direction direction; + struct host1x_bo_mapping *map; + struct host1x_bo *bo; reloc->target.bo = host1x_bo_get(reloc->target.bo); if (!reloc->target.bo) { @@ -155,64 +155,44 @@ static unsigned int pin_job(struct host1x *host, struct host1x_job *job) goto unpin; } - /* - * If the client device is not attached to an IOMMU, the - * physical address of the buffer object can be used. - * - * Similarly, when an IOMMU domain is shared between all - * host1x clients, the IOVA is already available, so no - * need to map the buffer object again. - * - * XXX Note that this isn't always safe to do because it - * relies on an assumption that no cache maintenance is - * needed on the buffer objects. - */ - if (!domain || client->group) - phys = &phys_addr; - else - phys = NULL; - - sgt = host1x_bo_pin(dev, reloc->target.bo, phys); - if (IS_ERR(sgt)) { - err = PTR_ERR(sgt); - goto unpin; - } + bo = reloc->target.bo; - if (sgt) { - unsigned long mask = HOST1X_RELOC_READ | - HOST1X_RELOC_WRITE; - enum dma_data_direction dir; - - switch (reloc->flags & mask) { - case HOST1X_RELOC_READ: - dir = DMA_TO_DEVICE; - break; + switch (reloc->flags & mask) { + case HOST1X_RELOC_READ: + direction = DMA_TO_DEVICE; + break; - case HOST1X_RELOC_WRITE: - dir = DMA_FROM_DEVICE; - break; + case HOST1X_RELOC_WRITE: + direction = DMA_FROM_DEVICE; + break; - case HOST1X_RELOC_READ | HOST1X_RELOC_WRITE: - dir = DMA_BIDIRECTIONAL; - break; + case HOST1X_RELOC_READ | HOST1X_RELOC_WRITE: + direction = DMA_BIDIRECTIONAL; + break; - default: - err = -EINVAL; - goto unpin; - } + default: + err = -EINVAL; + goto unpin; + } - err = dma_map_sgtable(dev, sgt, dir, 0); - if (err) - goto unpin; + map = host1x_bo_pin(dev, bo, direction, &client->cache); + if (IS_ERR(map)) { + err = PTR_ERR(map); + goto unpin; + } - job->unpins[job->num_unpins].dev = dev; - job->unpins[job->num_unpins].dir = dir; - phys_addr = sg_dma_address(sgt->sgl); + /* + * host1x clients are generally not able to do scatter-gather themselves, so fail + * if the buffer is discontiguous and we fail to map its SG table to a single + * contiguous chunk of I/O virtual memory. + */ + if (map->chunks > 1) { + err = -EINVAL; + goto unpin; } - job->addr_phys[job->num_unpins] = phys_addr; - job->unpins[job->num_unpins].bo = reloc->target.bo; - job->unpins[job->num_unpins].sgt = sgt; + job->addr_phys[job->num_unpins] = map->phys; + job->unpins[job->num_unpins].map = map; job->num_unpins++; } @@ -224,12 +204,11 @@ static unsigned int pin_job(struct host1x *host, struct host1x_job *job) return 0; for (i = 0; i < job->num_cmds; i++) { + struct host1x_bo_mapping *map; size_t gather_size = 0; struct scatterlist *sg; - dma_addr_t phys_addr; unsigned long shift; struct iova *alloc; - dma_addr_t *phys; unsigned int j; if (job->cmds[i].is_wait) @@ -243,25 +222,16 @@ static unsigned int pin_job(struct host1x *host, struct host1x_job *job) goto unpin; } - /** - * If the host1x is not attached to an IOMMU, there is no need - * to map the buffer object for the host1x, since the physical - * address can simply be used. - */ - if (!iommu_get_domain_for_dev(host->dev)) - phys = &phys_addr; - else - phys = NULL; - - sgt = host1x_bo_pin(host->dev, g->bo, phys); - if (IS_ERR(sgt)) { - err = PTR_ERR(sgt); - goto put; + map = host1x_bo_pin(host->dev, g->bo, DMA_TO_DEVICE, &host->cache); + if (IS_ERR(map)) { + err = PTR_ERR(map); + goto unpin; } if (host->domain) { - for_each_sgtable_sg(sgt, sg, j) + for_each_sgtable_sg(map->sgt, sg, j) gather_size += sg->length; + gather_size = iova_align(&host->iova, gather_size); shift = iova_shift(&host->iova); @@ -272,33 +242,23 @@ static unsigned int pin_job(struct host1x *host, struct host1x_job *job) goto put; } - err = iommu_map_sgtable(host->domain, - iova_dma_addr(&host->iova, alloc), - sgt, IOMMU_READ); + err = iommu_map_sgtable(host->domain, iova_dma_addr(&host->iova, alloc), + map->sgt, IOMMU_READ); if (err == 0) { __free_iova(&host->iova, alloc); err = -EINVAL; goto put; } - job->unpins[job->num_unpins].size = gather_size; - phys_addr = iova_dma_addr(&host->iova, alloc); - } else if (sgt) { - err = dma_map_sgtable(host->dev, sgt, DMA_TO_DEVICE, 0); - if (err) - goto put; - - job->unpins[job->num_unpins].dir = DMA_TO_DEVICE; - job->unpins[job->num_unpins].dev = host->dev; - phys_addr = sg_dma_address(sgt->sgl); + map->phys = iova_dma_addr(&host->iova, alloc); + map->size = gather_size; } - job->addr_phys[job->num_unpins] = phys_addr; - job->gather_addr_phys[i] = phys_addr; - - job->unpins[job->num_unpins].bo = g->bo; - job->unpins[job->num_unpins].sgt = sgt; + job->addr_phys[job->num_unpins] = map->phys; + job->unpins[job->num_unpins].map = map; job->num_unpins++; + + job->gather_addr_phys[i] = map->phys; } return 0; @@ -690,22 +650,16 @@ void host1x_job_unpin(struct host1x_job *job) unsigned int i; for (i = 0; i < job->num_unpins; i++) { - struct host1x_job_unpin_data *unpin = &job->unpins[i]; - struct device *dev = unpin->dev ?: host->dev; - struct sg_table *sgt = unpin->sgt; - - if (!job->enable_firewall && unpin->size && host->domain) { - iommu_unmap(host->domain, job->addr_phys[i], - unpin->size); - free_iova(&host->iova, - iova_pfn(&host->iova, job->addr_phys[i])); - } + struct host1x_bo_mapping *map = job->unpins[i].map; + struct host1x_bo *bo = map->bo; - if (unpin->dev && sgt) - dma_unmap_sgtable(unpin->dev, sgt, unpin->dir, 0); + if (!job->enable_firewall && map->size && host->domain) { + iommu_unmap(host->domain, job->addr_phys[i], map->size); + free_iova(&host->iova, iova_pfn(&host->iova, job->addr_phys[i])); + } - host1x_bo_unpin(dev, unpin->bo, sgt); - host1x_bo_put(unpin->bo); + host1x_bo_unpin(map); + host1x_bo_put(bo); } job->num_unpins = 0; diff --git a/drivers/gpu/host1x/job.h b/drivers/gpu/host1x/job.h index b4428c5495c9..dad5a1946693 100644 --- a/drivers/gpu/host1x/job.h +++ b/drivers/gpu/host1x/job.h @@ -35,11 +35,7 @@ struct host1x_job_cmd { }; struct host1x_job_unpin_data { - struct host1x_bo *bo; - struct sg_table *sgt; - struct device *dev; - size_t size; - enum dma_data_direction dir; + struct host1x_bo_mapping *map; }; /* diff --git a/drivers/gpu/host1x/syncpt.c b/drivers/gpu/host1x/syncpt.c index d198a10848c6..e08e331e46ae 100644 --- a/drivers/gpu/host1x/syncpt.c +++ b/drivers/gpu/host1x/syncpt.c @@ -143,6 +143,8 @@ void host1x_syncpt_restore(struct host1x *host) for (i = 0; i < host1x_syncpt_nb_bases(host); i++) host1x_hw_syncpt_restore_wait_base(host, sp_base + i); + host1x_hw_syncpt_enable_protection(host); + wmb(); } @@ -366,9 +368,6 @@ int host1x_syncpt_init(struct host1x *host) host->syncpt = syncpt; host->bases = bases; - host1x_syncpt_restore(host); - host1x_hw_syncpt_enable_protection(host); - /* Allocate sync point to use for clearing waits for expired fences */ host->nop_sp = host1x_syncpt_alloc(host, 0, "reserved-nop"); if (!host->nop_sp) diff --git a/drivers/soc/tegra/common.c b/drivers/soc/tegra/common.c index cd33e99249c3..35c882da55fc 100644 --- a/drivers/soc/tegra/common.c +++ b/drivers/soc/tegra/common.c @@ -10,6 +10,7 @@ #include <linux/export.h> #include <linux/of.h> #include <linux/pm_opp.h> +#include <linux/pm_runtime.h> #include <soc/tegra/common.h> #include <soc/tegra/fuse.h> @@ -43,6 +44,7 @@ static int tegra_core_dev_init_opp_state(struct device *dev) { unsigned long rate; struct clk *clk; + bool rpm_enabled; int err; clk = devm_clk_get(dev, NULL); @@ -57,8 +59,31 @@ static int tegra_core_dev_init_opp_state(struct device *dev) return -EINVAL; } + /* + * Runtime PM of the device must be enabled in order to set up + * GENPD's performance properly because GENPD core checks whether + * device is suspended and this check doesn't work while RPM is + * disabled. This makes sure the OPP vote below gets cached in + * GENPD for the device. Instead, the vote is done the next time + * the device gets runtime resumed. + */ + rpm_enabled = pm_runtime_enabled(dev); + if (!rpm_enabled) + pm_runtime_enable(dev); + + /* should never happen in practice */ + if (!pm_runtime_enabled(dev)) { + dev_WARN(dev, "failed to enable runtime PM\n"); + pm_runtime_disable(dev); + return -EINVAL; + } + /* first dummy rate-setting initializes voltage vote */ err = dev_pm_opp_set_rate(dev, rate); + + if (!rpm_enabled) + pm_runtime_disable(dev); + if (err) { dev_err(dev, "failed to initialize OPP clock: %d\n", err); return err; diff --git a/include/linux/host1x.h b/include/linux/host1x.h index 7bccf589aba7..e8dc5bc41f79 100644 --- a/include/linux/host1x.h +++ b/include/linux/host1x.h @@ -7,6 +7,8 @@ #define __LINUX_HOST1X_H #include <linux/device.h> +#include <linux/dma-direction.h> +#include <linux/spinlock.h> #include <linux/types.h> enum host1x_class { @@ -15,6 +17,8 @@ enum host1x_class { HOST1X_CLASS_GR2D_SB = 0x52, HOST1X_CLASS_VIC = 0x5D, HOST1X_CLASS_GR3D = 0x60, + HOST1X_CLASS_NVDEC = 0xF0, + HOST1X_CLASS_NVDEC1 = 0xF5, }; struct host1x; @@ -24,6 +28,28 @@ struct iommu_group; u64 host1x_get_dma_mask(struct host1x *host1x); /** + * struct host1x_bo_cache - host1x buffer object cache + * @mappings: list of mappings + * @lock: synchronizes accesses to the list of mappings + */ +struct host1x_bo_cache { + struct list_head mappings; + struct mutex lock; +}; + +static inline void host1x_bo_cache_init(struct host1x_bo_cache *cache) +{ + INIT_LIST_HEAD(&cache->mappings); + mutex_init(&cache->lock); +} + +static inline void host1x_bo_cache_destroy(struct host1x_bo_cache *cache) +{ + /* XXX warn if not empty? */ + mutex_destroy(&cache->lock); +} + +/** * struct host1x_client_ops - host1x client operations * @early_init: host1x client early initialization code * @init: host1x client initialization code @@ -73,6 +99,8 @@ struct host1x_client { struct host1x_client *parent; unsigned int usecount; struct mutex lock; + + struct host1x_bo_cache cache; }; /* @@ -82,23 +110,48 @@ struct host1x_client { struct host1x_bo; struct sg_table; +struct host1x_bo_mapping { + struct kref ref; + struct dma_buf_attachment *attach; + enum dma_data_direction direction; + struct list_head list; + struct host1x_bo *bo; + struct sg_table *sgt; + unsigned int chunks; + struct device *dev; + dma_addr_t phys; + size_t size; + + struct host1x_bo_cache *cache; + struct list_head entry; +}; + +static inline struct host1x_bo_mapping *to_host1x_bo_mapping(struct kref *ref) +{ + return container_of(ref, struct host1x_bo_mapping, ref); +} + struct host1x_bo_ops { struct host1x_bo *(*get)(struct host1x_bo *bo); void (*put)(struct host1x_bo *bo); - struct sg_table *(*pin)(struct device *dev, struct host1x_bo *bo, - dma_addr_t *phys); - void (*unpin)(struct device *dev, struct sg_table *sgt); + struct host1x_bo_mapping *(*pin)(struct device *dev, struct host1x_bo *bo, + enum dma_data_direction dir); + void (*unpin)(struct host1x_bo_mapping *map); void *(*mmap)(struct host1x_bo *bo); void (*munmap)(struct host1x_bo *bo, void *addr); }; struct host1x_bo { const struct host1x_bo_ops *ops; + struct list_head mappings; + spinlock_t lock; }; static inline void host1x_bo_init(struct host1x_bo *bo, const struct host1x_bo_ops *ops) { + INIT_LIST_HEAD(&bo->mappings); + spin_lock_init(&bo->lock); bo->ops = ops; } @@ -112,18 +165,10 @@ static inline void host1x_bo_put(struct host1x_bo *bo) bo->ops->put(bo); } -static inline struct sg_table *host1x_bo_pin(struct device *dev, - struct host1x_bo *bo, - dma_addr_t *phys) -{ - return bo->ops->pin(dev, bo, phys); -} - -static inline void host1x_bo_unpin(struct device *dev, struct host1x_bo *bo, - struct sg_table *sgt) -{ - bo->ops->unpin(dev, sgt); -} +struct host1x_bo_mapping *host1x_bo_pin(struct device *dev, struct host1x_bo *bo, + enum dma_data_direction dir, + struct host1x_bo_cache *cache); +void host1x_bo_unpin(struct host1x_bo_mapping *map); static inline void *host1x_bo_mmap(struct host1x_bo *bo) { @@ -181,6 +226,7 @@ struct host1x_job; struct host1x_channel *host1x_channel_request(struct host1x_client *client); struct host1x_channel *host1x_channel_get(struct host1x_channel *channel); +void host1x_channel_stop(struct host1x_channel *channel); void host1x_channel_put(struct host1x_channel *channel); int host1x_job_submit(struct host1x_job *job); diff --git a/include/soc/tegra/common.h b/include/soc/tegra/common.h index af41ad80ec21..8ec1ac07fc85 100644 --- a/include/soc/tegra/common.h +++ b/include/soc/tegra/common.h @@ -39,4 +39,19 @@ devm_tegra_core_dev_init_opp_table(struct device *dev, } #endif +static inline int +devm_tegra_core_dev_init_opp_table_common(struct device *dev) +{ + struct tegra_core_opp_params opp_params = {}; + int err; + + opp_params.init_state = true; + + err = devm_tegra_core_dev_init_opp_table(dev, &opp_params); + if (err != -ENODEV) + return err; + + return 0; +} + #endif /* __SOC_TEGRA_COMMON_H__ */ diff --git a/include/uapi/drm/amdgpu_drm.h b/include/uapi/drm/amdgpu_drm.h index 26e45fc5eb1a..0b94ec7b73e7 100644 --- a/include/uapi/drm/amdgpu_drm.h +++ b/include/uapi/drm/amdgpu_drm.h @@ -80,7 +80,7 @@ extern "C" { * * %AMDGPU_GEM_DOMAIN_GTT GPU accessible system memory, mapped into the * GPU's virtual address space via gart. Gart memory linearizes non-contiguous - * pages of system memory, allows GPU access system memory in a linezrized + * pages of system memory, allows GPU access system memory in a linearized * fashion. * * %AMDGPU_GEM_DOMAIN_VRAM Local video memory. For APUs, it is memory |
