aboutsummaryrefslogtreecommitdiff
path: root/drivers/gpu
AgeCommit message (Collapse)AuthorFilesLines
2021-10-05drm/amdgpu: handle the case of pci_channel_io_frozen only in amdgpu_pci_resumeGuchun Chen2-0/+7
In current code, when a PCI error state pci_channel_io_normal is detectd, it will report PCI_ERS_RESULT_CAN_RECOVER status to PCI driver, and PCI driver will continue the execution of PCI resume callback report_resume by pci_walk_bridge, and the callback will go into amdgpu_pci_resume finally, where write lock is releasd unconditionally without acquiring such lock first. In this case, a deadlock will happen when other threads start to acquire the read lock. To fix this, add a member in amdgpu_device strucutre to cache pci_channel_state, and only continue the execution in amdgpu_pci_resume when it's pci_channel_io_frozen. Fixes: c9a6b82f45e2 ("drm/amdgpu: Implement DPC recovery") Suggested-by: Andrey Grodzovsky <[email protected]> Signed-off-by: Guchun Chen <[email protected]> Reviewed-by: Andrey Grodzovsky <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2021-10-05drm/amdgpu: init iommu after amdkfd device initYifan Zhang1-4/+4
This patch is to fix clinfo failure in Raven/Picasso: Number of platforms: 1 Platform Profile: FULL_PROFILE Platform Version: OpenCL 2.2 AMD-APP (3364.0) Platform Name: AMD Accelerated Parallel Processing Platform Vendor: Advanced Micro Devices, Inc. Platform Extensions: cl_khr_icd cl_amd_event_callback Platform Name: AMD Accelerated Parallel Processing Number of devices: 0 Signed-off-by: Yifan Zhang <[email protected]> Reviewed-by: James Zhu <[email protected]> Tested-by: James Zhu <[email protected]> Acked-by: Felix Kuehling <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2021-10-05drm/amdkfd: remove redundant iommu cleanup codeYifan Zhang1-7/+1
kfd_resume doesn't involve iommu operation, remove redundant iommu cleanup code. Signed-off-by: Yifan Zhang <[email protected]> Reviewed-by: James Zhu <[email protected]> Tested-by: James Zhu <[email protected]> Acked-by: Felix Kuehling <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2021-10-05drm/amdgpu/display: fix dependencies for DRM_AMD_DC_SIAlex Deucher1-0/+2
Depends on DRM_AMDGPU_SI and DRM_AMD_DC Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2021-10-05drm/amdgpu: During s0ix don't wait to signal GFXOFFLijo Lazar1-2/+12
In the rare event when GFX IP suspend coincides with a s0ix entry, don't schedule a delayed work, instead signal PMFW immediately to allow GFXOFF entry. GFXOFF is a prerequisite for s0ix entry. PMFW needs to be signaled about GFXOFF status before amd-pmc module passes OS HINT to PMFW telling that everything is ready for a safe s0ix entry. Bug: https://gitlab.freedesktop.org/drm/amd/-/issues/1712 Signed-off-by: Lijo Lazar <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Reviewed-by: Mario Limonciello <[email protected]> Signed-off-by: Alex Deucher <[email protected]> Cc: [email protected]
2021-10-05drm/amdkfd: fix a potential ttm->sg memory leakLang Yu1-0/+1
Memory is allocated for ttm->sg by kmalloc in kfd_mem_dmamap_userptr, but isn't freed by kfree in kfd_mem_dmaunmap_userptr. Free it! Fixes: 264fb4d332f5 ("drm/amdgpu: Add multi-GPU DMA mapping helpers") Signed-off-by: Lang Yu <[email protected]> Reviewed-by: Felix Kuehling <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2021-10-05drm/amd/display: USB4 bring up set correct addressJude Shih1-1/+1
[Why] YELLOW_CARP_B0 address was not correct [How] Set YELLOW_CARP_B0 to 0x1A. Reviewed-by: Nicholas Kazlauskas <[email protected]> Acked-by: Wayne Lin <[email protected]> Signed-off-by: Jude Shih <[email protected]> Signed-off-by: Alex Deucher <[email protected]> Cc: [email protected]
2021-10-05drm/amd/display: Fix DCN3 B0 DP Alt MappingLiu, Zhan1-0/+6
[Why] DCN3 B0 has a mux, which redirects PHYC and PHYD to PHYF and PHYG. [How] Fix DIG mapping. Reviewed-by: Charlene Liu <[email protected]> Acked-by: Rodrigo Siqueira <[email protected]> Signed-off-by: Zhan Liu <[email protected]> Signed-off-by: Alex Deucher <[email protected]> Cc: [email protected] (cherry picked from commit 4b7786d87fb3adf3e534c4f1e4f824d8700b786b)
2021-10-05drm/amd/display: Fix B0 USB-C DP Alt modeLiu, Zhan4-2/+70
[Why] Starting from B0, along with RDPCSTX, RDPCSPIPE registers are also used. [How] Make sure RDPCSPIPE registers are programmed correctly. Reviewed-by: Charlene Liu <[email protected]> Acked-by: Rodrigo Siqueira <[email protected]> Signed-off-by: Zhan Liu <[email protected]> Signed-off-by: Alex Deucher <[email protected]> Cc: [email protected] (cherry picked from commit bdd1a21b52557ea8f61d0a5dc2f77151b576eb70)
2021-10-05drm/i915: Extend the async flip VT-d w/a to skl/bxtVille Syrjälä2-0/+17
Looks like skl/bxt/derivatives also need the plane stride stretch w/a when using async flips and VT-d is enabled, or else we get corruption on screen. To my surprise this was even documented in bspec, but only as a note on the CHICHKEN_PIPESL register description rather than on the w/a list. So very much the same thing as on HSW/BDW, except the bits moved yet again. Cc: [email protected] Cc: Karthik B S <[email protected]> Fixes: 55ea1cb178ef ("drm/i915: Enable async flips in i915") Signed-off-by: Ville Syrjälä <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected] Reviewed-by: Matt Roper <[email protected]> (cherry picked from commit d08df3b0bdb25546e86dc9a6c4e3ec0c43832299) Signed-off-by: Jani Nikula <[email protected]>
2021-10-05drm/i915/bdb: Fix version checkLukasz Majczak2-6/+21
With patch "drm/i915/vbt: Fix backlight parsing for VBT 234+" the size of bdb_lfp_backlight_data structure has been increased, causing if-statement in the parse_lfp_backlight function that comapres this structure size to the one retrieved from BDB, always to fail for older revisions. This patch calculates expected size of the structure for a given BDB version and compares it with the value gathered from BDB. Tested on Chromebook Pixelbook (Nocturne) (reports bdb->version = 221) Fixes: d381baad29b4 ("drm/i915/vbt: Fix backlight parsing for VBT 234+") Tested-by: Lukasz Majczak <[email protected]> Signed-off-by: Lukasz Majczak <[email protected]> Reviewed-by: José Roberto de Souza <[email protected]> Signed-off-by: José Roberto de Souza <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected] (cherry picked from commit 4378daf5d04eed59724e6d0e74755e17dce2e105) Signed-off-by: Jani Nikula <[email protected]>
2021-10-05drm/i915/tc: Fix TypeC port init/resume time sanitizationImre Deak3-18/+20
Atm during driver loading and system resume TypeC ports are accessed before their HW/SW state is synced. Move the TypeC port sanitization to the encoder's sync_state hook to fix this. v2: Handle the encoder disabled case in gen11_dsi_sync_state() as well (Jose, Jani) Fixes: f9e76a6e68d3 ("drm/i915: Add an encoder hook to sanitize its state during init/resume") Cc: José Roberto de Souza <[email protected]> Cc: Ville Syrjälä <[email protected]> Signed-off-by: Imre Deak <[email protected]> Reviewed-by: José Roberto de Souza <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected] (cherry picked from commit 7194dc998dfffca096c30b3cd39625158608992d) Signed-off-by: Jani Nikula <[email protected]>
2021-10-05drm/i915: Fix runtime pm handling in i915_gem_shrinkMaarten Lankhorst1-2/+5
We forgot to call intel_runtime_pm_put on error, fix it! Signed-off-by: Maarten Lankhorst <[email protected]> Fixes: cf41a8f1dc1e ("drm/i915: Finally remove obj->mm.lock.") Cc: Thomas Hellström <[email protected]> Cc: Daniel Vetter <[email protected]> Cc: <[email protected]> # v5.13+ Reviewed-by: Thomas Hellström <[email protected]> Reviewed-by: Niranjana Vishwanathapura <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected] (cherry picked from commit 239f3c2ee18376587026efecaea5250fa5926d20) Signed-off-by: Jani Nikula <[email protected]>
2021-10-05drm/i915/audio: Use BIOS provided value for RKL HDA linkKai-Heng Feng1-2/+3
Commit 989634fb49ad ("drm/i915/audio: set HDA link parameters in driver") makes HDMI audio on Lenovo P350 disappear. So in addition to TGL, extend the logic to RKL to use BIOS provided value to fix the regression. Fixes: 989634fb49ad ("drm/i915/audio: set HDA link parameters in driver") Reviewed-by: Kai Vehmanen <[email protected]> Signed-off-by: Kai-Heng Feng <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected] (cherry picked from commit c6b40ee330fe09b332715bb7ec1467e4fcbe2e65) Signed-off-by: Jani Nikula <[email protected]>
2021-10-04treewide: Replace the use of mem_encrypt_active() with cc_platform_has()Tom Lendacky4-8/+10
Replace uses of mem_encrypt_active() with calls to cc_platform_has() with the CC_ATTR_MEM_ENCRYPT attribute. Remove the implementation of mem_encrypt_active() across all arches. For s390, since the default implementation of the cc_platform_has() matches the s390 implementation of mem_encrypt_active(), cc_platform_has() does not need to be implemented in s390 (the config option ARCH_HAS_CC_PLATFORM is not set). Signed-off-by: Tom Lendacky <[email protected]> Signed-off-by: Borislav Petkov <[email protected]> Link: https://lkml.kernel.org/r/[email protected]
2021-10-02drm/i915: fix blank screen booting crashesHugh Dickins1-2/+3
5.15-rc1 crashes with blank screen when booting up on two ThinkPads using i915. Bisections converge convincingly, but arrive at different and suprising "culprits", none of them the actual culprit. netconsole (with init_netconsole() hacked to call i915_init() when logging has started, instead of by module_init()) tells the story: kernel BUG at drivers/gpu/drm/i915/i915_sw_fence.c:245! with RSI: ffffffff814d408b pointing to sw_fence_dummy_notify(). I've been building with CONFIG_CC_OPTIMIZE_FOR_SIZE=y, and that function needs to be 4-byte aligned. Fixes: 62eaf0ae217d ("drm/i915/guc: Support request cancellation") Signed-off-by: Hugh Dickins <[email protected]> Tested-by: Steven Rostedt (VMware) <[email protected]> Signed-off-by: Linus Torvalds <[email protected]>
2021-10-01drm/msm/submit: fix overflow check on 64-bit architecturesArnd Bergmann1-1/+2
The overflow check does causes a warning from clang-14 when 'sz' is a type that is smaller than size_t: drivers/gpu/drm/msm/msm_gem_submit.c:217:10: error: result of comparison of constant 18446744073709551615 with expression of type 'unsigned int' is always false [-Werror,-Wtautological-constant-out-of-range-compare] if (sz == SIZE_MAX) { Change the type accordingly. Fixes: 20224d715a88 ("drm/msm/submit: Move copy_from_user ahead of locking bos") Signed-off-by: Arnd Bergmann <[email protected]> Reviewed-by: Stephen Boyd <[email protected]> Link: https://lore.kernel.org/r/[email protected] Signed-off-by: Rob Clark <[email protected]>
2021-10-01drm/msm: One sched entity per process per priorityRob Clark3-21/+75
Some userspace apps make assumptions that rendering against multiple contexts within the same process (from the same thread, with appropriate MakeCurrent() calls) provides sufficient synchronization without any external synchronization (ie. glFenceSync()/glWaitSync()). Since a submitqueue maps to a gl/vk context, having multiple sched entities of the same priority only works with implicit sync enabled. To fix this, limit things to a single sched entity per priority level per process. An alternative would be sharing submitqueues between contexts in userspace, but tracking of per-context faults (ie. GL_EXT_robustness) is already done at the submitqueue level, so this is not an option. Signed-off-by: Rob Clark <[email protected]>
2021-10-01drm/msm: A bit more docs + cleanupRob Clark2-45/+57
msm_file_private is more gpu related, and in the next commit it will need access to other GPU specific #defines. While we're at it, add some comments. Signed-off-by: Rob Clark <[email protected]>
2021-10-01drm/msm/a6xx: Track current ctx by seqnoRob Clark4-4/+17
In theory a context can be destroyed and a new one allocated at the same address, making the pointer comparision to detect when we don't need to update the current pagetables invalid. Instead assign a sequence number to each context on creation, and use this for the check. Fixes: 84c31ee16f90 ("drm/msm/a6xx: Add support for per-instance pagetables") Signed-off-by: Rob Clark <[email protected]>
2021-10-01drm/msm/a6xx: Serialize GMU communicationRob Clark3-6/+43
I've seen some crashes in our crash reporting that *look* like multiple threads stomping on each other while communicating with GMU. So wrap all those paths in a lock. Signed-off-by: Rob Clark <[email protected]>
2021-10-01drm/msm: Fix crash on dev file closeRob Clark2-4/+3
If the device file was opened prior to fw being available (such as from initrd before rootfs is mounted, when the initrd does not contain GPU fw), that would cause a later crash when the dev file is closed due to unitialized submitqueues list: CPU: 4 PID: 263 Comm: plymouthd Tainted: G W 5.15.0-rc2-next-20210924 #2 Hardware name: LENOVO 81JL/LNVNB161216, BIOS 9UCN33WW(V2.06) 06/ 4/2019 pstate: 60400005 (nZCv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--) pc : msm_submitqueue_close+0x30/0x190 [msm] lr : msm_postclose+0x54/0xf0 [msm] sp : ffff80001074bb80 x29: ffff80001074bb80 x28: ffff03ad80c4db80 x27: ffff03ad80dc5ab0 x26: 0000000000000000 x25: ffff03ad80dc5af8 x24: ffff03ad81e90800 x23: 0000000000000000 x22: ffff03ad81e90800 x21: ffff03ad8b35e788 x20: ffff03ad81e90878 x19: 0000000000000000 x18: 0000000000000000 x17: 0000000000000000 x16: ffffda15f14f7940 x15: 0000000000000000 x14: 0000000000000000 x13: 0000000000000001 x12: 0000000000000040 x11: 0000000000000000 x10: 0000000000000000 x9 : ffffda15cd18ff88 x8 : ffff03ad80c4db80 x7 : 0000000000000228 x6 : 0000000000000000 x5 : 1793a4e807e636bd x4 : ffff03ad80c4db80 x3 : ffff03ad81e90878 x2 : 0000000000000000 x1 : ffff03ad80c4db80 x0 : 0000000000000000 Call trace: msm_submitqueue_close+0x30/0x190 [msm] msm_postclose+0x54/0xf0 [msm] drm_file_free.part.0+0x1cc/0x2e0 [drm] drm_close_helper.isra.0+0x74/0x84 [drm] drm_release+0x78/0x120 [drm] __fput+0x78/0x23c ____fput+0x1c/0x30 task_work_run+0xcc/0x22c do_exit+0x304/0x9f4 do_group_exit+0x44/0xb0 __wake_up_parent+0x0/0x3c invoke_syscall+0x50/0x120 el0_svc_common.constprop.0+0x4c/0xf4 do_el0_svc+0x30/0x9c el0_svc+0x20/0x60 el0t_64_sync_handler+0xe8/0xf0 el0t_64_sync+0x1a0/0x1a4 Code: aa0003f5 a90153f3 f8408eb3 aa1303e0 (f85e8674) ---[ end trace 39b2fa37509a2be2 ]--- Fixing recursive fault but reboot is needed! Fixes: 86c2a0f000c1 drm/msm: ("Small submitqueue creation cleanup") Reported-by: Steev Klimaszewski <[email protected]> Signed-off-by: Rob Clark <[email protected]>
2021-10-01Merge tag 'exynos-drm-fixes-for-v5.15-rc4' of ↵Daniel Vetter9-31/+9
git://git.kernel.org/pub/scm/linux/kernel/git/daeinki/drm-exynos into drm-fixes One cleanup - Use devm_platform_ioremap_resource() helper function instead of old one. Signed-off-by: Daniel Vetter <[email protected]> From: Inki Dae <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
2021-10-01Merge tag 'amd-drm-fixes-5.15-2021-09-29' of ↵Daniel Vetter7-11/+53
https://gitlab.freedesktop.org/agd5f/linux into drm-fixes amd-drm-fixes-5.15-2021-09-29: amdgpu: - gart pin count fix - eDP flicker fix - GFX9 MQD fix - Display fixes - Tiling flags fix for pre-GFX9 - SDMA resume fix for S0ix Signed-off-by: Daniel Vetter <[email protected]> From: Alex Deucher <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
2021-10-01Merge tag 'drm-intel-fixes-2021-09-30' of ↵Daniel Vetter5-23/+14
git://anongit.freedesktop.org/drm/drm-intel into drm-fixes drm/i915 fixes for v5.15-rc4: - Fix GVT scheduler ww lock usage - Fix pdfdocs documentation build - Fix request early tracepoints - Fix an invalid warning from rps worker Signed-off-by: Daniel Vetter <[email protected]> From: Jani Nikula <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
2021-10-01KVM: x86: add config for non-kvm users of page trackingDavid Stevens1-0/+1
Add a config option that allows kvm to determine whether or not there are any external users of page tracking. Signed-off-by: David Stevens <[email protected]> Message-Id: <[email protected]> Signed-off-by: Paolo Bonzini <[email protected]>
2021-09-28drm/amdgpu: force exit gfxoff on sdma resume for rmb s0ixPrike Liang1-0/+8
In the s2idle stress test sdma resume fail occasionally,in the failed case GPU is in the gfxoff state.This issue may introduce by firmware miss handle doorbell S/R and now temporary fix the issue by forcing exit gfxoff for sdma resume. Signed-off-by: Prike Liang <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]> Cc: [email protected]
2021-09-28drm/amdgpu: check tiling flags when creating FB on GFX8-Simon Ser1-0/+31
On GFX9+, format modifiers are always enabled and ensure the frame-buffers can be scanned out at ADDFB2 time. On GFX8-, format modifiers are not supported and no other check is performed. This means ADDFB2 IOCTLs will succeed even if the tiling isn't supported for scan-out, and will result in garbage displayed on screen [1]. Fix this by adding a check for tiling flags for GFX8 and older. The check is taken from radeonsi in Mesa (see how is_displayable is populated in gfx6_compute_surface). Changes in v2: use drm_WARN_ONCE instead of drm_WARN (Michel) [1]: https://github.com/swaywm/wlroots/issues/3185 Signed-off-by: Simon Ser <[email protected]> Acked-by: Michel Dänzer <[email protected]> Cc: Alex Deucher <[email protected]> Cc: Harry Wentland <[email protected]> Cc: Nicholas Kazlauskas <[email protected]> Cc: Bas Nieuwenhuizen <[email protected]> Signed-off-by: Alex Deucher <[email protected]> Cc: [email protected]
2021-09-28drm/amd/display: Pass PCI deviceid into DCCharlene Liu1-0/+1
[why] pci deviceid not passed to dal dc, without proper break, dcn2.x falls into dcn3.x code path [how] pass in pci deviceid, and break once dal_version initialized. Reviewed-by: Zhan Liu <[email protected]> Acked-by: Anson Jacob <[email protected]> Signed-off-by: Charlene Liu <[email protected]> Tested-by: Daniel Wheeler <[email protected]> Signed-off-by: Alex Deucher <[email protected]> Cc: [email protected]
2021-09-28drm/amd/display: initialize backlight_ramping_override to falseJosip Pavic1-0/+1
[Why] Stack variable params.backlight_ramping_override is uninitialized, so it contains junk data [How] Initialize the variable to false Reviewed-by: Roman Li <[email protected]> Acked-by: Anson Jacob <[email protected]> Signed-off-by: Josip Pavic <[email protected]> Tested-by: Daniel Wheeler <[email protected]> Signed-off-by: Alex Deucher <[email protected]> Cc: [email protected]
2021-09-28drm/amdgpu: correct initial cp_hqd_quantum for gfx9Hawking Zhang1-1/+1
didn't read the value of mmCP_HQD_QUANTUM from correct register offset Signed-off-by: Hawking Zhang <[email protected]> Reviewed-by: Le Ma <[email protected]> Signed-off-by: Alex Deucher <[email protected]> Cc: [email protected]
2021-09-28drm/amd/display: Fix Display Flicker on embedded panelsPraful Swarnakar1-8/+7
[Why] ASSR is dependent on Signed PSP Verstage to enable Content Protection for eDP panels. Unsigned PSP verstage is used during development phase causing ASSR to FAIL. As a result, link training is performed with DP_PANEL_MODE_DEFAULT instead of DP_PANEL_MODE_EDP for eDP panels that causes display flicker on some panels. [How] - Do not change panel mode, if ASSR is disabled - Just report and continue to perform eDP link training with right settings further. Signed-off-by: Praful Swarnakar <[email protected]> Reviewed-by: Harry Wentland <[email protected]> Signed-off-by: Alex Deucher <[email protected]> Cc: [email protected]
2021-09-28drm/amdgpu: fix gart.bo pin_count leakLeslie Shi2-2/+4
gmc_v{9,10}_0_gart_disable() isn't called matched with correspoding gart_enbale function in SRIOV case. This will lead to gart.bo pin_count leak on driver unload. Cc: Hawking Zhang <[email protected]> Signed-off-by: Leslie Shi <[email protected]> Signed-off-by: Guchun Chen <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2021-09-28Merge tag 'drm/tegra/for-5.15-rc3' of ↵Dave Airlie4-12/+5
ssh://git.freedesktop.org/git/tegra/linux into drm-fixes drm/tegra: Fixes for v5.15-rc3 This contains a fix for an UAPI error that happens when no IOMMU is enabled, as well as a few build warning fixes and plugs a potential memory leak. Signed-off-by: Dave Airlie <[email protected]> From: Thierry Reding <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
2021-09-27drm/i915: Remove warning from the rps workerTejas Upadhyay1-2/+0
In commit 4e5c8a99e1cb ("drm/i915: Drop i915_request.lock requirement for intel_rps_boost()"), we decoupled the rps worker from the pm so that we could avoid the synchronization penalty which makes the assertion liable to run too early. Which makes warning invalid hence removed. Fixes: 4e5c8a99e1cb ("drm/i915: Drop i915_request.lock requirement for intel_rps_boost()") Reviewed-by: Chris Wilson <[email protected]> Signed-off-by: Tejas Upadhyay <[email protected]> Signed-off-by: Matt Roper <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/20210914090412.1393498-1-tejaskumarx.surendrakumar.upadhyay@intel.com (cherry picked from commit a837a0686308d95ad9c48d32b4dfe86a17dc98c2) Signed-off-by: Jani Nikula <[email protected]>
2021-09-27drm/i915/request: fix early tracepointsMatthew Auld1-9/+2
Currently we blow up in trace_dma_fence_init, when calling into get_driver_name or get_timeline_name, since both the engine and context might be NULL(or contain some garbage address) in the case of newly allocated slab objects via the request ctor. Note that we also use SLAB_TYPESAFE_BY_RCU here, which allows requests to be immediately freed, but delay freeing the underlying page by an RCU grace period. With this scheme requests can be re-allocated, at the same time as they are also being read by some lockless RCU lookup mechanism. In the ctor case, which is only called for new slab objects(i.e allocate new page and call the ctor for each object) it's safe to reset the context/engine prior to calling into dma_fence_init, since we can be certain that no one is doing an RCU lookup which might depend on peeking at the engine/context, like in active_engine(), since the object can't yet be externally visible. In the recycled case(which might also be externally visible) the request refcount always transitions from 0->1 after we set the context/engine etc, which should ensure it's valid to dereference the engine for example, when doing an RCU list-walk, so long as we can also increment the refcount first. If the refcount is already zero, then the request is considered complete/released. If it's non-zero, then the request might be in the process of being re-allocated, or potentially still in flight, however after successfully incrementing the refcount, it's possible to carefully inspect the request state, to determine if the request is still what we were looking for. Note that all externally visible requests returned to the cache must have zero refcount. One possible fix then is to move dma_fence_init out from the request ctor. Originally this was how it was done, but it was moved in: commit 855e39e65cfc33a73724f1cc644ffc5754864a20 Author: Chris Wilson <[email protected]> Date: Mon Feb 3 09:41:48 2020 +0000 drm/i915: Initialise basic fence before acquiring seqno where it looks like intel_timeline_get_seqno() relied on some of the rq->fence state, but that is no longer the case since: commit 12ca695d2c1ed26b2dcbb528b42813bd0f216cfc Author: Maarten Lankhorst <[email protected]> Date: Tue Mar 23 16:49:50 2021 +0100 drm/i915: Do not share hwsp across contexts any more, v8. intel_timeline_get_seqno() could also be cleaned up slightly by dropping the request argument. Moving dma_fence_init back out of the ctor, should ensure we have enough of the request initialised in case of trace_dma_fence_init. Functionally this should be the same, and is effectively what we were already open coding before, except now we also assign the fence->lock and fence->ops, but since these are invariant for recycled requests(which might be externally visible), and will therefore already hold the same value, it shouldn't matter. An alternative fix, since we don't yet have a fully initialised request when in the ctor, is just setting the context/engine as NULL, but this does require adding some extra handling in get_driver_name etc. v2(Daniel): - Try to make the commit message less confusing Fixes: 855e39e65cfc ("drm/i915: Initialise basic fence before acquiring seqno") Signed-off-by: Matthew Auld <[email protected]> Cc: Michael Mason <[email protected]> Cc: Daniel Vetter <[email protected]> Reviewed-by: Daniel Vetter <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected] (cherry picked from commit be988eaee1cb208c4445db46bc3ceaf75f586f0b) Signed-off-by: Jani Nikula <[email protected]>
2021-09-27drm/i915/guc, docs: Fix pdfdocs build error by removing nested gridAkira Yokosawa2-10/+10
Nested grids in grid-table cells are not specified as proper ReST constructs. Commit 572f2a5cd974 ("drm/i915/guc: Update firmware to v62.0.0") added a couple of kerneldoc tables of the form: +---+-------+------------------------------------------------------+ | 1 | 31:0 | +------------------------------------------------+ | +---+-------+ | | | |...| | | Embedded `HXG Message`_ | | +---+-------+ | | | | n | 31:0 | +------------------------------------------------+ | +---+-------+------------------------------------------------------+ For "make htmldocs", they happen to work as one might expect, but they are incompatible with "make latexdocs" and "make pdfdocs", and cause the generated gpu.tex file to become incomplete and unbuildable by xelatex. Restore the compatibility by removing those nested grids in the tables. Size comparison of generated gpu.tex: Sphinx 2.4.4 Sphinx 4.2.0 v5.14: 3238686 3841631 v5.15-rc1: 376270 432729 with this fix: 3377846 3998095 Fixes: 572f2a5cd974 ("drm/i915/guc: Update firmware to v62.0.0") Cc: John Harrison <[email protected]> Cc: Michal Wajdeczko <[email protected]> Cc: Matthew Brost <[email protected]> Cc: Daniele Ceraolo Spurio <[email protected]> Cc: Matt Roper <[email protected]> Cc: Jonathan Corbet <[email protected]> Signed-off-by: Akira Yokosawa <[email protected]> Signed-off-by: Jani Nikula <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected] (cherry picked from commit 017792a041183c4f5ec595f386e76a40ed728cfc) Signed-off-by: Jani Nikula <[email protected]>
2021-09-27Merge tag 'gvt-fixes-2021-09-18' of https://github.com/intel/gvt-linux into ↵Jani Nikula1-2/+2
drm-intel-fixes gvt-fixes-2021-09-18 - ww locking fix from Zhi Signed-off-by: Jani Nikula <[email protected]> From: Zhenyu Wang <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
2021-09-25drm/mga/mga_ioc32: Use struct_group() for memcpy() regionKees Cook1-13/+14
In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memcpy(), memmove(), and memset(), avoid intentionally writing across neighboring fields. Use struct_group() in struct drm32_mga_init around members chipset, sgram, maccess, fb_cpp, front_offset, front_pitch, back_offset, back_pitch, depth_cpp, depth_offset, depth_pitch, texture_offset, and texture_size, so they can be referenced together. This will allow memcpy() and sizeof() to more easily reason about sizes, improve readability, and avoid future warnings about writing beyond the end of chipset. "pahole" shows no size nor member offset changes to struct drm32_mga_init. "objdump -d" shows no meaningful object code changes (i.e. only source line number induced differences and optimizations). Note that since this is a UAPI header, __struct_group() is used directly. Cc: David Airlie <[email protected]> Cc: Lee Jones <[email protected]> Cc: [email protected] Signed-off-by: Kees Cook <[email protected]> Acked-by: Daniel Vetter <[email protected]> Link: https://lore.kernel.org/lkml/[email protected]
2021-09-24Merge tag 'drm-fixes-2021-09-24' of git://anongit.freedesktop.org/drm/drmLinus Torvalds9-30/+55
Pull drm fixes from Dave Airlie: "Quiet week this week, just some i915 and amd fixes, just getting ready for my all nighter maintainer summit! Summary: i915: - Fix ADL-P memory bandwidth parameters - Fix memory corruption due to a double free - Fix memory leak in DMC firmware handling amdgpu: - Update MAINTAINERS entry for powerplay - Fix empty macros - SI DPM fix amdkfd: - SVM fixes - DMA mapping fix" * tag 'drm-fixes-2021-09-24' of git://anongit.freedesktop.org/drm/drm: drm/amdkfd: fix svm_migrate_fini warning drm/amdkfd: handle svm migrate init error drm/amd/pm: Update intermediate power state for SI drm/amdkfd: fix dma mapping leaking warning drm/amdkfd: SVM map to gpus check vma boundary MAINTAINERS: fix up entry for AMD Powerplay drm/amd/display: fix empty debug macros drm/i915: Free all DMC payloads drm/i915: Move __i915_gem_free_object to ttm_bo_destroy drm/i915: Update memory bandwidth parameters
2021-09-24Merge tag 'drm-intel-fixes-2021-09-23' of ↵Dave Airlie3-8/+25
git://anongit.freedesktop.org/drm/drm-intel into drm-fixes drm/i915 fixes for v5.15-rc3: - Fix ADL-P memory bandwidth parameters - Fix memory corruption due to a double free - Fix memory leak in DMC firmware handling Signed-off-by: Dave Airlie <[email protected]> From: Jani Nikula <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected]
2021-09-23drm/amdkfd: fix svm_migrate_fini warningPhilip Yang3-15/+4
Device manager releases device-specific resources when a driver disconnects from a device, devm_memunmap_pages and devm_release_mem_region calls in svm_migrate_fini are redundant. It causes below warning trace after patch "drm/amdgpu: Split amdgpu_device_fini into early and late", so remove function svm_migrate_fini. BUG: https://gitlab.freedesktop.org/drm/amd/-/issues/1718 WARNING: CPU: 1 PID: 3646 at drivers/base/devres.c:795 devm_release_action+0x51/0x60 Call Trace: ? memunmap_pages+0x360/0x360 svm_migrate_fini+0x2d/0x60 [amdgpu] kgd2kfd_device_exit+0x23/0xa0 [amdgpu] amdgpu_amdkfd_device_fini_sw+0x1d/0x30 [amdgpu] amdgpu_device_fini_sw+0x45/0x290 [amdgpu] amdgpu_driver_release_kms+0x12/0x30 [amdgpu] drm_dev_release+0x20/0x40 [drm] release_nodes+0x196/0x1e0 device_release_driver_internal+0x104/0x1d0 driver_detach+0x47/0x90 bus_remove_driver+0x7a/0xd0 pci_unregister_driver+0x3d/0x90 amdgpu_exit+0x11/0x20 [amdgpu] Signed-off-by: Philip Yang <[email protected]> Reviewed-by: Felix Kuehling <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2021-09-23drm/amdkfd: handle svm migrate init errorPhilip Yang1-0/+3
If svm migration init failed to create pgmap for device memory, set pgmap type to 0 to disable device SVM support capability. Signed-off-by: Philip Yang <[email protected]> Reviewed-by: Felix Kuehling <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2021-09-23drm/amd/pm: Update intermediate power state for SILijo Lazar1-0/+2
Update the current state as boot state during dpm initialization. During the subsequent initialization, set_power_state gets called to transition to the final power state. set_power_state refers to values from the current state and without current state populated, it could result in NULL pointer dereference. For ex: on platforms where PCI speed change is supported through ACPI ATCS method, the link speed of current state needs to be queried before deciding on changing to final power state's link speed. The logic to query ATCS-support was broken on certain platforms. The issue became visible when broken ATCS-support logic got fixed with commit f9b7f3703ff9 ("drm/amdgpu/acpi: make ATPX/ATCS structures global (v2)"). Bug: https://gitlab.freedesktop.org/drm/amd/-/issues/1698 Signed-off-by: Lijo Lazar <[email protected]> Reviewed-by: Hawking Zhang <[email protected]> Acked-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]> Cc: [email protected]
2021-09-23drm/amdkfd: fix dma mapping leaking warningPhilip Yang1-4/+14
For xnack off, restore work dma unmap previous system memory page, and dma map the updated system memory page to update GPU mapping, this is not dma mapping leaking, remove the WARN_ONCE for dma mapping leaking. prange->dma_addr store the VRAM page pfn after the range migrated to VRAM, should not dma unmap VRAM page when updating GPU mapping or remove prange. Add helper svm_is_valid_dma_mapping_addr to check VRAM page and error cases. Mask out SVM_RANGE_VRAM_DOMAIN flag in dma_addr before calling amdgpu vm update to avoid BUG_ON(*addr & 0xFFFF00000000003FULL), and set it again immediately after. This flag is used to know the type of page later to dma unmapping system memory page. Fixes: 1d5dbfe6c06a ("drm/amdkfd: classify and map mixed svm range pages in GPU") Signed-off-by: Philip Yang <[email protected]> Reviewed-by: Felix Kuehling <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2021-09-23drm/amdkfd: SVM map to gpus check vma boundaryPhilip Yang1-1/+5
SVM range may includes multiple VMAs with different vm_flags, if prange page index is the last page of the VMA offset + npages, update GPU mapping to create GPU page table with same VMA access permission. Signed-off-by: Philip Yang <[email protected]> Reviewed-by: Felix Kuehling <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2021-09-23drm/amd/display: fix empty debug macrosArnd Bergmann1-2/+2
Using an empty macro expansion as a conditional expression produces a W=1 warning: drivers/gpu/drm/amd/amdgpu/../display/dc/dce/dce_aux.c: In function 'dce_aux_transfer_with_retries': drivers/gpu/drm/amd/amdgpu/../display/dc/dce/dce_aux.c:775:156: error: suggest braces around empty body in an 'if' statement [-Werror=empty-body] 775 | "dce_aux_transfer_with_retries: AUX_RET_SUCCESS: AUX_TRANSACTION_REPLY_I2C_OVER_AUX_DEFER"); | ^ drivers/gpu/drm/amd/amdgpu/../display/dc/dce/dce_aux.c:783:155: error: suggest braces around empty body in an 'if' statement [-Werror=empty-body] 783 | "dce_aux_transfer_with_retries: AUX_RET_SUCCESS: AUX_TRANSACTION_REPLY_I2C_OVER_AUX_NACK"); | ^ Expand it to "do { } while (0)" instead to make the expression more robust and avoid the warning. Fixes: 56aca2309301 ("drm/amd/display: Add AUX I2C tracing.") Signed-off-by: Arnd Bergmann <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2021-09-21drm/i915: Free all DMC payloadsChris Wilson1-1/+4
Free all the DMC payloads, not just DMC_MAIN. unreferenced object 0xffff88ff32d4d800 (size 1024): comm "kworker/1:5", pid 701, jiffies 4294904239 (age 109.736s) hex dump (first 32 bytes): 40 40 00 0c 03 00 00 00 00 00 00 00 00 00 00 00 @@.............. 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ backtrace: [<00000000ba9d0d95>] dmc_load_work_fn+0x34d/0x510 [i915] [<000000001049fcab>] process_one_work+0x261/0x550 [<00000000eeb995ac>] worker_thread+0x49/0x3c0 [<0000000021031dc3>] kthread+0x10b/0x140 [<000000004a0f69ee>] ret_from_fork+0x1f/0x30 unreferenced object 0xffff88ff0bde4000 (size 1024): comm "kworker/0:3", pid 708, jiffies 4294904469 (age 108.816s) hex dump (first 32 bytes): 40 40 00 0c 01 00 00 00 00 00 00 00 00 00 00 00 @@.............. 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................ backtrace: [<00000000ba9d0d95>] dmc_load_work_fn+0x34d/0x510 [i915] [<000000001049fcab>] process_one_work+0x261/0x550 [<00000000eeb995ac>] worker_thread+0x49/0x3c0 [<0000000021031dc3>] kthread+0x10b/0x140 [<000000004a0f69ee>] ret_from_fork+0x1f/0x30 Fixes: 3d5928a168a9 ("drm/i915/xelpd: Pipe A DMC plugging") Cc: Anusha Srivatsa <[email protected]> Cc: José Roberto de Souza <[email protected]> Signed-off-by: Chris Wilson <[email protected]> Signed-off-by: Lucas De Marchi <[email protected]> Reviewed-by: José Roberto de Souza <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected] (cherry picked from commit 064b877dff4252ced91a1c8b1f129073f2991f6e) Signed-off-by: Jani Nikula <[email protected]>
2021-09-21drm/i915: Move __i915_gem_free_object to ttm_bo_destroyMaarten Lankhorst1-4/+5
When we implement delayed destroy, we may have a second call to the delete_mem_notify() handler, while free_object() only should be called once. Move it to bo->destroy(), to ensure it's only called once. This fixes some weird memory corruption issues with delayed destroy when async eviction is used. Signed-off-by: Maarten Lankhorst <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected] Fixes: 213d50927763 ("drm/i915/ttm: Introduce a TTM i915 gem object backend") Cc: Thomas Hellström <[email protected]> Reviewed-by: Thomas Hellström <[email protected]> (cherry picked from commit 48b0961269546716c3232748bf37e64e49fb866c) Signed-off-by: Jani Nikula <[email protected]>
2021-09-21drm/i915: Update memory bandwidth parametersRadhakrishna Sripada1-3/+16
Earlier while calculating derated bw we would use 90% of the calculated bw. Starting ADL-P we use a non standard derating. Updating the formulae to reflect the same. Bspec: 64631 v2: Use the new derating value only for ADL-P(MattR) Fixes: 4d32fe2f14a7 ("drm/i915/adl_p: Update memory bandwidth parameters") Cc: Matt Roper <[email protected]> Signed-off-by: Radhakrishna Sripada <[email protected]> Reviewed-by: Matt Roper <[email protected]> Signed-off-by: Matt Roper <[email protected]> Link: https://patchwork.freedesktop.org/patch/msgid/[email protected] (cherry picked from commit f6d66fc8cf5f673ea76407be84dc17dbb3eda108) Signed-off-by: Jani Nikula <[email protected]>
Powered by cgit, Gruvboxed by Blaster4385.