aboutsummaryrefslogtreecommitdiffstatshomepage
path: root/tools/perf/scripts/python/export-to-postgresql.py (unfollow)
AgeCommit message (Collapse)AuthorFilesLines
2024-09-17drm/xe: Do not run GPU page fault handler on a closed VMMatthew Brost1-0/+6
Closing a VM removes page table memory thus we shouldn't touch page tables when a VM is closed. Do not run the GPU page fault handler once the VM is closed to avoid touching page tables. Signed-off-by: Matthew Brost <matthew.brost@intel.com> Reviewed-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20240911011820.825127-1-matthew.brost@intel.com (cherry picked from commit f96dbf7c321d70834d46f3aedb75a671e839b51e) Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
2024-09-17drm/xe/bo: add some annotations in bo_put()Matthew Auld2-5/+15
If the put() triggers bo destroy then there is at least one potential sleeping lock. Also annotate bos_lock and ggtt lock. Signed-off-by: Matthew Auld <matthew.auld@intel.com> Cc: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com> Cc: Tejas Upadhyay <tejas.upadhyay@intel.com> Cc: "Thomas Hellström" <thomas.hellstrom@linux.intel.com> Reviewed-by: Matthew Brost <matthew.brost@intel.com> Reviewed-by: Tejas Upadhyay <tejas.upadhyay@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20240911155527.178910-8-matthew.auld@intel.com (cherry picked from commit 3b04c2cfd71c54117237c72f2a08ff0ae1f602e2) Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
2024-09-17drm/xe/client: use mem_type from the current resourceMatthew Auld1-6/+1
Rather extract the mem_type from the current resource. Checking the first potential placement doesn't really tell us where the bo is currently allocated, especially if there are multiple potential placements. Signed-off-by: Matthew Auld <matthew.auld@intel.com> Cc: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com> Cc: Tejas Upadhyay <tejas.upadhyay@intel.com> Cc: "Thomas Hellström" <thomas.hellstrom@linux.intel.com> Reviewed-by: Matthew Brost <matthew.brost@intel.com> Reviewed-by: Tejas Upadhyay <tejas.upadhyay@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20240911155527.178910-7-matthew.auld@intel.com (cherry picked from commit fbd73b7d2ae29ef0f604f376bcc22b886a49329e) Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
2024-09-17drm/xe/client: add missing bo locking in show_meminfo()Matthew Auld1-3/+36
bo_meminfo() wants to inspect bo state like tt and the ttm resource, however this state can change at any point leading to stuff like NPD and UAF, if the bo lock is not held. Grab the bo lock when calling bo_meminfo(), ensuring we drop any spinlocks first. In the case of object_idr we now also need to hold a ref. v2 (MattB) - Also add xe_bo_assert_held() Fixes: 0845233388f8 ("drm/xe: Implement fdinfo memory stats printing") Signed-off-by: Matthew Auld <matthew.auld@intel.com> Cc: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com> Cc: Tejas Upadhyay <tejas.upadhyay@intel.com> Cc: "Thomas Hellström" <thomas.hellstrom@linux.intel.com> Cc: <stable@vger.kernel.org> # v6.8+ Reviewed-by: Matthew Brost <matthew.brost@intel.com> Reviewed-by: Tejas Upadhyay <tejas.upadhyay@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20240911155527.178910-6-matthew.auld@intel.com (cherry picked from commit 4f63d712fa104c3ebefcb289d1e733e86d8698c7) Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
2024-09-17drm/xe/client: fix deadlock in show_meminfo()Matthew Auld1-1/+5
There is a real deadlock as well as sleeping in atomic() bug in here, if the bo put happens to be the last ref, since bo destruction wants to grab the same spinlock and sleeping locks. Fix that by dropping the ref using xe_bo_put_deferred(), and moving the final commit outside of the lock. Dropping the lock around the put is tricky since the bo can go out of scope and delete itself from the list, making it difficult to navigate to the next list entry. Fixes: 0845233388f8 ("drm/xe: Implement fdinfo memory stats printing") Closes: https://gitlab.freedesktop.org/drm/xe/kernel/-/issues/2727 Signed-off-by: Matthew Auld <matthew.auld@intel.com> Cc: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com> Cc: Tejas Upadhyay <tejas.upadhyay@intel.com> Cc: "Thomas Hellström" <thomas.hellstrom@linux.intel.com> Cc: <stable@vger.kernel.org> # v6.8+ Reviewed-by: Matthew Brost <matthew.brost@intel.com> Reviewed-by: Tejas Upadhyay <tejas.upadhyay@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20240911155527.178910-5-matthew.auld@intel.com (cherry picked from commit 0083b8e6f11d7662283a267d4ce7c966812ffd8a) Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
2024-09-17drm/xe/vram: fix ccs offset calculationMatthew Auld1-0/+1
Spec says SW is expected to round up to the nearest 128K, if not already aligned for the CC unit view of CCS. We are seeing the assert sometimes pop on BMG to tell us that there is a hole between GSM and CCS, as well as popping other asserts with having a vram size with strange alignment, which is likely caused by misaligned offset here. v2 (Shuicheng): - Do the round_up() on final SW address. BSpec: 68023 Fixes: b5c2ca0372dc ("drm/xe/xe2hpg: Determine flat ccs offset for vram") Signed-off-by: Matthew Auld <matthew.auld@intel.com> Cc: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com> Cc: Akshata Jahagirdar <akshata.jahagirdar@intel.com> Cc: Lucas De Marchi <lucas.demarchi@intel.com> Cc: Shuicheng Lin <shuicheng.lin@intel.com> Cc: Matt Roper <matthew.d.roper@intel.com> Cc: stable@vger.kernel.org # v6.10+ Reviewed-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com> Tested-by: Shuicheng Lin <shuicheng.lin@intel.com> Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20240916084911.13119-2-matthew.auld@intel.com Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com> (cherry picked from commit 37173392741c425191b959acb3adf70c9a4610c0) Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
2024-09-17drm/xe/guc: Fix GUC_{SUBMIT,FIRMWARE}_VER helper macrosMichal Wajdeczko1-2/+4
Those macros rely on non-existing MAKE_VER_STRUCT macro, while the correct one that should be used is named MAKE_GUC_VER_STRUCT. Fixes: 4eb0aab6e443 ("drm/xe/guc: Bump minimum required GuC version to v70.29.2") Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com> Cc: Julia Filipchuk <julia.filipchuk@intel.com> Cc: John Harrison <John.C.Harrison@Intel.com> Reviewed-by: Michał Winiarski <michal.winiarski@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20240912203817.1880-2-michal.wajdeczko@intel.com (cherry picked from commit 02fdf821ed79f59c40d766a85947aa7cc25d4364) Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
2024-09-16drm/i915/dp: Fix AUX IO power enabling for eDP PSRImre Deak3-1/+22
Panel Self Refresh on eDP requires the AUX IO power to be enabled whenever the output (main link) is enabled. This is required by the AUX_PHY_WAKE/ML_PHY_LOCK signaling initiated by the HW automatically to re-enable the main link after it got disabled in power saving states (see eDP v1.4b, sections 5.1, 6.1.3.3.1.1). The Panel Replay mode on non-eDP outputs on the other hand is only supported by keeping the main link active, thus not requiring the above AUX_PHY_WAKE/ML_PHY_LOCK signaling (eDP v1.4b, section 6.1.3.3.1.2). Thus enabling the AUX IO power for this case is not required either. Based on the above enable the AUX IO power only for eDP/PSR outputs. Bspec: 49274, 53370 v2: - Add a TODO comment to adjust the requirement for AUX IO based on whether the ALPM/main-link off mode gets enabled. (Rodrigo) Cc: Animesh Manna <animesh.manna@intel.com> Fixes: b8cf5b5d266e ("drm/i915/panelreplay: Initializaton and compute config for panel replay") Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com> Signed-off-by: Imre Deak <imre.deak@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20240910111847.2995725-1-imre.deak@intel.com (cherry picked from commit f7c2ed9d4ce80a2570c492825de239dc8b500f2e) Signed-off-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
2024-09-16drm/i915/display: BMG supports UHBR13.5Arun R Murthy1-2/+11
UHBR20 is not supported by battlemage and the maximum link rate supported is UHBR13.5 v2: Replace IS_DGFX with IS_BATTLEMAGE (Jani) HSD: 16023263677 Signed-off-by: Arun R Murthy <arun.r.murthy@intel.com> Reviewed-by: Mika Kahola <mika.kahola@intel.com> Fixes: 98b1c87a5e51 ("drm/i915/xe2hpd: Set maximum DP rate to UHBR13.5") Signed-off-by: Suraj Kandpal <suraj.kandpal@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20240827081205.136569-1-arun.r.murthy@intel.com (cherry picked from commit 9c2338ac4543e0fab3a1e0f9f025591e0f0d9f8f) Signed-off-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
2024-09-16drm/i915/psr: Do not wait for PSR being idle on on Panel ReplayJouni Högander1-11/+2
We do not have ALPM on DP Panel Replay. Due to this SRD_STATUS[SRD State] doesn't change from SRDENT_ON after Panel Replay is enabled until it gets disabled. On eDP Panel Replay DEEP_SLEEP is not reached. _psr2_ready_for_pipe_update_locked is waiting DEEP_SLEEP bit getting reset. Take these into account in Panel Replay code by not waiting PSR getting idle after enabling VBI. Fixes: 29fb595d4875 ("drm/i915/psr: Panel replay uses SRD_STATUS to track it's status") Cc: Animesh Manna <animesh.manna@intel.com> Signed-off-by: Jouni Högander <jouni.hogander@intel.com> Reviewed-by: Animesh Manna <animesh.manna@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20240906070033.289015-5-jouni.hogander@intel.com (cherry picked from commit a2d98feb4b0013ef4f9db0d8f642a8ac1f5ecbb9) Signed-off-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
2024-09-12drm/xe: Fix missing conversion to xe_display_pm_runtime_resumeMaarten Lankhorst1-1/+1
This error path was missed when converting away from xe_display_pm_resume with second argument. Fixes: 66a0f6b9f5fc ("drm/xe/display: handle HPD polling in display runtime suspend/resume") Cc: Arun R Murthy <arun.r.murthy@intel.com> Cc: Vinod Govindapillai <vinod.govindapillai@intel.com> Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com> Reviewed-by: Vinod Govindapillai <vinod.govindapillai@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20240905150052.174895-2-maarten.lankhorst@linux.intel.com (cherry picked from commit 474f64cb988a410db8a0b779d6afdaa2a7fc5759) Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
2024-09-12drm/xe/xe2hpg: Add Wa_15016589081Tejas Upadhyay2-0/+5
Wa_15016589081 applies to xe2_hpg renderCS V2(Gustavo) - rename bit macro Signed-off-by: Tejas Upadhyay <tejas.upadhyay@intel.com> Reviewed-by: Gustavo Sousa <gustavo.sousa@intel.com> Reviewed-by: Himal Prasad Ghimiray <himal.prasad.ghimiray@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20240904101333.2049655-1-tejas.upadhyay@intel.com Signed-off-by: Nirmoy Das <nirmoy.das@intel.com> (cherry picked from commit 9db969b36b2fbca13ad4088aff725ebd5e8142f5) Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
2024-09-12drm/xe: Don't keep stale pointer to bo->ggtt_nodeMichal Wajdeczko1-2/+5
When we fail to map a BO in the GGTT, we release our GGTT node placeholder, but leave stale bo->ggtt_node pointer to it, which triggers an assert immediately followed by a crash, due to UAF: [ ] xe 0000:00:02.0: [drm] Assertion `bo->ggtt_node->base.size == bo->size` failed! [ ] WARNING: CPU: 4 PID: 126 at drivers/gpu/drm/xe/xe_ggtt.c:689 xe_ggtt_remove_bo+0x1d9/0x250 [xe] [ ] RIP: 0010:xe_ggtt_remove_bo+0x1d9/0x250 [xe] [ ] Call Trace: [ ] <TASK> [ ] ? __warn+0x88/0x190 [ ] ? xe_ggtt_remove_bo+0x1d9/0x250 [xe] [ ] ? report_bug+0x1c3/0x1d0 [ ] ? handle_bug+0x42/0x70 [ ] ? exc_invalid_op+0x14/0x70 [ ] ? asm_exc_invalid_op+0x16/0x20 [ ] ? xe_ggtt_remove_bo+0x1d9/0x250 [xe] [ ] ? xe_ggtt_remove_bo+0x1d9/0x250 [xe] [ ] xe_ttm_bo_destroy+0x11f/0x260 [xe] [ ] ? ttm_bo_release+0x31c/0x350 [ttm] [ ] ? __mutex_unlock_slowpath+0x35/0x270 [ ] __xe_bo_create_locked+0x4a0/0x550 [xe] [ ] ? mark_held_locks+0x49/0x80 [ ] xe_bo_create_pin_map_at+0x37/0x200 [xe] [ ] xe_bo_create_pin_map+0x11/0x20 [xe] While around, for similar reason, also don't keep an error pointer if we fail to allocate ggtt_node placeholder. Fixes: 34e804220f69 ("drm/xe: Make xe_ggtt_node struct independent") Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20240906220348.1836-1-michal.wajdeczko@intel.com (cherry picked from commit f2710d95724ebbfa35d6d4b82017eeab70994509) Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
2024-09-12drm/xe: fix missing 'xe_vm_put'Dafna Hirschfeld1-1/+3
Fix memleak caused by missing xe_vm_put Fixes: 852856e3b6f6 ("drm/xe: Use reserved copy engine for user binds on faulting devices") Signed-off-by: Dafna Hirschfeld <dhirschfeld@habana.ai> Reviewed-by: Nirmoy Das <nirmoy.das@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20240901044227.1177211-1-dhirschfeld@habana.ai Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com> (cherry picked from commit 249df8cbecf0ab4877eab66cae857748631831a9) Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
2024-09-12drm/xe: fix build warning with CONFIG_PM=nArnd Bergmann1-0/+4
The 'runtime_status' field is an implementation detail of the power management code, so a device driver should not normally touch this: drivers/gpu/drm/xe/xe_pm.c: In function 'xe_pm_suspending_or_resuming': drivers/gpu/drm/xe/xe_pm.c:606:26: error: 'struct dev_pm_info' has no member named 'runtime_status' 606 | return dev->power.runtime_status == RPM_SUSPENDING || | ^ drivers/gpu/drm/xe/xe_pm.c:607:27: error: 'struct dev_pm_info' has no member named 'runtime_status' 607 | dev->power.runtime_status == RPM_RESUMING; | ^ drivers/gpu/drm/xe/xe_pm.c:608:1: error: control reaches end of non-void function [-Werror=return-type] Add an #ifdef check to avoid the build regression. Fixes: ad92f5231261 ("drm/xe: Suppress missing outer rpm protection warning") Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com> Signed-off-by: Arnd Bergmann <arnd@arndb.de> Link: https://patchwork.freedesktop.org/patch/msgid/20240909202521.1018439-1-arnd@kernel.org Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com> (cherry picked from commit 1c129ed07de47684ff2471e32b52fa823533aa06) Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
2024-09-12drm/xe: Suppress missing outer rpm protection warningRodrigo Vivi1-1/+16
Do not raise a WARN if we are likely within suspending or resuming path. This is likely this false positive: rpm_status: 0000:03:00.0 status=RPM_SUSPENDING console: xe_bo_evict_all (called from suspend) xe_sched_job_create: dev=0000:03:00.0, ... xe_sched_job_exec: dev=0000:03:00.0, ... xe_pm_runtime_put: dev=0000:03:00.0, ... xe_sched_job_run: dev=0000:03:00.0, ... rpm_usage: 0000:03:00.0 flags-0 cnt-2 ... rpm_usage: 0000:03:00.0 flags-0 cnt-2 ... rpm_usage: 0000:03:00.0 flags-0 cnt-2 ... console: xe 0000:03:00.0: [drm] Missing outer runtime PM protection console: xe_guc_ct_send+0x15/0x50 [xe] console: guc_exec_queue_run_job+0x1509/0x3950 [xe] [snip] console: drm_sched_run_job_work+0x649/0xc20 At this point, BOs are getting evicted from VRAM with rpm usage-counter = 2, but rpm status = SUSPENDING. The xe->pm_callback_task won't be equal 'current' because this call is coming from a work queue. So, pm_runtime_get_if_active() will be called and return 0 because rpm status != ACTIVE (but equal SUSPENDING or RESUMING). v2: Still get the reference even on non suspending/resuming path (Jonathan, Brost). Cc: Matthew Brost <matthew.brost@intel.com> Cc: Matthew Auld <matthew.auld@intel.com> Reviewed-by: Jonathan Cavitt <jonathan.cavitt@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20240905140215.56404-1-rodrigo.vivi@intel.com Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com> (cherry picked from commit cb85e39dc5d1717fab82810984cce0e54712a3c2) Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
2024-09-12drm/xe: prevent potential UAF in pf_provision_vf_ggtt()Matthew Auld1-4/+4
The node ptr can point to an already freed ptr, if we hit the path with an already allocated node. We later dereference that pointer with: xe_gt_assert(gt, !xe_ggtt_node_allocated(node)); which is a potential UAF. Fix this by not stashing the ptr for node. Also since it is likely a bad idea to leave config->ggtt_region pointing to a stale ptr, also set that to NULL by calling pf_release_vf_config_ggtt() instead of pf_release_ggtt(). Fixes: 34e804220f69 ("drm/xe: Make xe_ggtt_node struct independent") Signed-off-by: Matthew Auld <matthew.auld@intel.com> Cc: Matthew Brost <matthew.brost@intel.com> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com> Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20240828104341.180111-2-matthew.auld@intel.com (cherry picked from commit 89076b5a8b4e0a01040585e156a0b014cd472fd3) Signed-off-by: Lucas De Marchi <lucas.demarchi@intel.com>
2024-09-11drm/amd/display: Add all planes on CRTC to state for overlay cursorLeo Li1-0/+11
[Why] DC has a special commit path for native cursor, which use the built-in cursor pipe within DCN planes. This update path does not require all enabled planes to be added to the list of surface updates sent to DC. This is not the case for overlay cursor; it uses the same path as MPO commits. This update path requires all enabled planes to be added to the list of surface updates sent to DC. Otherwise, DC will disable planes not inside the list. [How] If overlay cursor is needed, add all planes on the same CRTC as this cursor to the atomic state. This is already done for non-cursor planes (MPO), just before the added lines. Fixes: 1b04dcca4fb1 ("drm/amd/display: Introduce overlay cursor mode") Closes: https://lore.kernel.org/lkml/f68020a3-c413-482d-beb2-5432d98a1d3e@amd.com Reviewed-by: Harry Wentland <harry.wentland@amd.com> Signed-off-by: Leo Li <sunpeng.li@amd.com> Tested-by: Mikhail Gavrilov <mikhail.v.gavrilov@gmail.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-09-11drm/i915/bios: fix printk format widthJani Nikula1-1/+1
s/0x04%x/0x%04x/ to use 0 prefixed width 4 instead of printing 04 verbatim. Fixes: 51f5748179d4 ("drm/i915/bios: create fake child devices on missing VBT") Cc: stable@vger.kernel.org # v5.13+ Reviewed-by: Vandita Kulkarni <vandita.kulkarni@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20240905112519.4186408-1-jani.nikula@intel.com Signed-off-by: Jani Nikula <jani.nikula@intel.com> (cherry picked from commit 54df34c5a2439b481f066476e67bfa21a0a640e5) Signed-off-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
2024-09-11drm/i915/display: Fix BMG CCS modifiersJuha-Pekka Heikkila1-0/+3
Let I915_FORMAT_MOD_4_TILED_BMG_CCS show up as supported modifier Fixes: 97c6efb36497 ("drm/i915/display: Plane capability for 64k phys alignment") Signed-off-by: Juha-Pekka Heikkila <juhapekka.heikkila@gmail.com> Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20240902074021.459480-1-juhapekka.heikkila@gmail.com Signed-off-by: Maarten Lankhorst,,, <maarten.lankhorst@linux.intel.com> (cherry picked from commit c4d37c54c3739530f8585ccf064fb712913f8375) Signed-off-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
2024-09-10drm/amdgpu: get rid of bogus includes of fdtable.hAl Viro2-2/+0
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-09-10drm/amdkfd: CRIU fixesAl Viro1-18/+46
Instead of trying to use close_fd() on failure exits, just have criu_get_prime_handle() store the file reference without inserting it into descriptor table. Then, once the callers are past the last failure exit, they can go and either insert all those file references into the corresponding slots of descriptor table, or drop all those file references and free the unused descriptors. Reviewed-by: Felix Kuehling <felix.kuehling@amd.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-09-10drm/amdgpu: fix a race in kfd_mem_export_dmabuf()Al Viro1-9/+3
Using drm_gem_prime_handle_to_fd() to set dmabuf up and insert it into descriptor table, only to have it looked up by file descriptor and remove it from descriptor table is not just too convoluted - it's racy; another thread might have modified the descriptor table while we'd been going through that song and dance. Switch kfd_mem_export_dmabuf() to using drm_gem_prime_handle_to_dmabuf() and leave the descriptor table alone... Reviewed-by: Felix Kuehling <felix.kuehling@amd.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-09-10drm: new helper: drm_gem_prime_handle_to_dmabuf()Al Viro2-30/+57
Once something had been put into descriptor table, the only thing you can do with it is returning descriptor to userland - you can't withdraw it on subsequent failure exit, etc. You certainly can't count upon it staying in the same slot of descriptor table - another thread could've played with close(2)/dup2(2)/whatnot. drm_gem_prime_handle_to_fd() creates a dmabuf, allocates a descriptor and attaches dmabuf's file to it (the last two steps are done in dma_buf_fd()). That's nice when all you are going to do is passing a descriptor to userland. If you just need to work with the resulting object or have something else to be done that might fail, drm_gem_prime_handle_to_fd() is racy. The problem is analogous to one with anon_inode_getfd(), and solution is similar to what anon_inode_getfile() provides. Add drm_gem_prime_handle_to_dmabuf() - the "set dmabuf up" parts of drm_gem_prime_handle_to_fd() without the descriptor-related ones. Instead of inserting into descriptor table and returning the file descriptor it just returns the struct file. drm_gem_prime_handle_to_fd() becomes a wrapper for it. Other users will be introduced in the next commit. Acked-by: Thomas Zimmermann <tzimmermann@suse.de> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-09-10drm/amdgpu/atomfirmware: Silence UBSAN warningAlex Deucher1-2/+2
Per the comments, these are variable sized arrays. Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/3613 Reviewed-by: Harry Wentland <harry.wentland@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-09-10drm/amdgpu: Fix kdoc entry in 'amdgpu_vm_cpu_prepare'Srinivasan Shanmugam1-2/+1
This commit updates described non-existent parameters 'resv' and 'sync_mode', and failed to describe the existing 'sync' parameter. Fixes the below with gcc W=1: drivers/gpu/drm/amd/amdgpu/amdgpu_vm_cpu.c:50: warning: Function parameter or struct member 'sync' not described in 'amdgpu_vm_cpu_prepare' drivers/gpu/drm/amd/amdgpu/amdgpu_vm_cpu.c:50: warning: Excess function parameter 'resv' description in 'amdgpu_vm_cpu_prepare' drivers/gpu/drm/amd/amdgpu/amdgpu_vm_cpu.c:50: warning: Excess function parameter 'sync_mode' description in 'amdgpu_vm_cpu_prepare' Cc: Christian König <christian.koenig@amd.com> Cc: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-09-10drm/amd/amdgpu: apply command submission parser for JPEG v1David (Ming Qiang) Wu2-1/+86
Similar to jpeg_v2_dec_ring_parse_cs() but it has different register ranges and a few other registers access. Acked-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: David (Ming Qiang) Wu <David.Wu3@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-09-10drm/amd/amdgpu: apply command submission parser for JPEG v2+David (Ming Qiang) Wu10-64/+78
This patch extends the same cs parser from JPEG v4.0.3 to other JPEG versions (v2 and above). Rename to more common name as jpeg_v2_dec_ring_parse_cs() from jpeg_v4_0_3_dec_ring_parse_cs(). Acked-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: David (Ming Qiang) Wu <David.Wu3@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-09-10drm/amd/pm: fix the pp_dpm_pcie issue on smu v14.0.2/3Kenneth Feng1-0/+3
fix the pp_dpm_pcie issue on smu v14.0.2/3 as below: 0: 2.5GT/s, x4 250Mhz 1: 8.0GT/s, x4 616Mhz * 2: 8.0GT/s, x4 1143Mhz * the middle level can be removed since it is always skipped on smu v14.0.2/3 Signed-off-by: Kenneth Feng <kenneth.feng@amd.com> Acked-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-09-10drm/amd/pm: update the features set on smu v14.0.2/3Kenneth Feng2-1/+19
update the features set on smu v14.0.2/3 Signed-off-by: Kenneth Feng <kenneth.feng@amd.com> Reviewed-by: Yang Wang <kevinyang.wang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-09-10drm/amdkfd: Fix resource leak in criu restore queueJesse Zhang1-0/+1
To avoid memory leaks, release q_extra_data when exiting the restore queue. v2: Correct the proto (Alex) Signed-off-by: Jesse Zhang <jesse.zhang@amd.com> Reviewed-by: Tim Huang <tim.huang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-09-10drm/amd/display: Do not reset planes based on crtc zpos_changedLeo Li1-1/+1
[Why] drm_normalize_zpos will set the crtc_state->zpos_changed to 1 if any of it's assigned planes changes zpos, or is removed/added from it. To have amdgpu_dm request a plane reset on this is too broad. For example, if only the cursor plane was moved from one crtc to another, the crtc's zpos_changed will be set to true. But that does not mean that the underlying primary plane requires a reset. [How] Narrow it down so that only the plane that has a change in zpos will require a reset. As a future TODO, we can further optimize this by only requiring a reset on z-order change. Z-order is different from z-pos, since a zpos change doesn't necessarily mean the z-ordering changed, and DC should only require a reset if the z-ordering changed. For example, the following zpos update does not change z-ordering: Plane A: zpos 2 -> 3 Plane B: zpos 1 -> 2 => Plane A is still on top of plane B: no reset needed Whereas this one does change z-ordering: Plane A: zpos 2 -> 1 Plane B: zpos 1 -> 2 => Plane A changed from on top, to below plane B: reset needed Fixes: 38e0c3df6dbd ("drm/amd/display: Move PRIMARY plane zpos higher") Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/3569 Signed-off-by: Leo Li <sunpeng.li@amd.com> Reviewed-by: Harry Wentland <harry.wentland@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-09-08Linux 6.11-rc7Linus Torvalds1-1/+1
2024-09-06KVM: x86: don't fall through case statements without annotationsLinus Torvalds1-0/+1
clang warns on this because it has an unannotated fall-through between cases: arch/x86/kvm/x86.c:4819:2: error: unannotated fall-through between switch labels [-Werror,-Wimplicit-fallthrough] and while we could annotate it as a fallthrough, the proper fix is to just add the break for this case, instead of falling through to the default case and the break there. gcc also has that warning, but it looks like gcc only warns for the cases where they fall through to "real code", rather than to just a break. Odd. Fixes: d30d9ee94cc0 ("KVM: x86: Only advertise KVM_CAP_READONLY_MEM when supported by VM") Cc: Paolo Bonzini <pbonzini@redhat.com> Cc: Tom Dohrmann <erbse.13@gmx.de> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2024-09-06drm/amdgpu: drop redundant W=1 warnings from MakefileJani Nikula1-17/+1
Since commit a61ddb4393ad ("drm: enable (most) W=1 warnings by default across the subsystem"), most of the extra warnings in the driver Makefile are redundant. Remove them. Note that -Wmissing-declarations and -Wmissing-prototypes are always enabled by default in scripts/Makefile.extrawarn. Reviewed-by: Hamza Mahfooz <hamza.mahfooz@amd.com> Signed-off-by: Jani Nikula <jani.nikula@intel.com> Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-09-06drm/amdgpu: revert "use CPU for page table update if SDMA is unavailable"Christian König1-6/+0
That is clearly not something we should do upstream. The SDMA is mandatory for the driver to work correctly. We could do this for emulation and bringup, but in those cases the engineer should probably enabled CPU based updates manually. This reverts commit 62eefd10ac1c7e976bda47ff311bd87cee40ab8d. Signed-off-by: Christian König <christian.koenig@amd.com> Reviewed-by: Yifan Zhang <yifan1.zhang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-09-06drm/amdgpu/mes11: Indent an if statmentDan Carpenter1-1/+1
Indent the "break" statement one more tab. Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-09-06drm/amdkfd: Document and define SVM events message macroPhilip Yang2-36/+109
Document how to use SMI system management interface to enable and receive SVM events. Document SVM event triggers. Define SVM events message string format macro that could be used by user mode for sscanf to parse the event. Add it to uAPI header file to make it obvious that is changing uAPI in future. No functional changes. Signed-off-by: Philip Yang <Philip.Yang@amd.com> Reviewed-by: James Zhu <James.Zhu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-09-06drm/amdkfd: Select reset method for poison handlingHawking Zhang1-8/+32
Driver mode-2 is only supported by relative new smc firmware. Signed-off-by: Hawking Zhang <Hawking.Zhang@amd.com> Reviewed-by: Tao Zhou <tao.zhou1@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-09-06drm/amdkfd: fix missed queue reset on queue destroyJonathan Kim1-2/+3
If a queue is being destroyed but causes a HWS hang on removal, the KFD may issue an unnecessary gpu reset if the destroyed queue can be fixed by a queue reset. This is because the queue has been removed from the KFD's queue list prior to the preemption action on destroy so the reset call will fail to match the HQD PQ reset information against the KFD's queue record to do the actual reset. To fix this, deactivate the queue prior to preemption since it's being destroyed anyways and remove the queue from the KFD's queue list after preemption. Signed-off-by: Jonathan Kim <jonathan.kim@amd.com> Reviewed-by: Felix Kuehling <felix.kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-09-06drm/amdgpu: Surface svm_default_granularity, a RW module parameterRamesh Errabolu4-7/+39
Enables users to update SVM's default granularity, used in buffer migration and handling of recoverable page faults. Param value is set in terms of log(numPages(buffer)), e.g. 9 for a 2 MIB buffer Signed-off-by: Ramesh Errabolu <Ramesh.Errabolu@amd.com> Reviewed-by: Philip Yang <Philip.Yang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-09-06drm/amdgpu: fix queue reset issue by mmioJesse Zhang1-0/+1
Initialize the queue type before resetting the queue using mmio. Signed-off-by: Jesse Zhang <jesse.zhang@amd.com> Acked-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-09-06drm/amd/display: Add kdoc entry for 'program_isharp_1dlut' in 'dpp401_dscl_program_isharp'Srinivasan Shanmugam1-0/+1
Added a descriptor for the 'program_isharp_1dlut' parameter, which is a flag used to determine whether to program the isharp 1D LUT. Fixes the below with gcc W=1: drivers/gpu/drm/amd/amdgpu/../display/dc/dpp/dcn401/dcn401_dpp_dscl.c:963: warning: Function parameter or struct member 'program_isharp_1dlut' not described in 'dpp401_dscl_program_isharp' Cc: Tom Chung <chiahsuan.chung@amd.com> Cc: Rodrigo Siqueira <Rodrigo.Siqueira@amd.com> Cc: Roman Li <roman.li@amd.com> Cc: Alex Hung <alex.hung@amd.com> Cc: Aurabindo Pillai <aurabindo.pillai@amd.com> Cc: Harry Wentland <harry.wentland@amd.com> Cc: Hamza Mahfooz <hamza.mahfooz@amd.com> Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com> Reviewed-by: Tom Chung <chiahsuan.chung@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-09-06drm/amdgpu: Replace 'amdgpu_job_submit_direct' with 'drm_sched_entity' in cleaner shaderSrinivasan Shanmugam1-16/+19
This commit replaces the use of amdgpu_job_submit_direct which submits the job to the ring directly, with drm_sched_entity in the cleaner shader job submission process. The change allows the GPU scheduler to manage the cleaner shader job. - The job is then submitted to the GPU using the drm_sched_entity_push_job function, which allows the GPU scheduler to manage the job. This change improves the reliability of the cleaner shader job submission process by leveraging the capabilities of the GPU scheduler. Fixes: d361ad5d2fc0 ("drm/amdgpu: Add sysfs interface for running cleaner shader") Cc: Christian König <christian.koenig@amd.com> Cc: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com> Suggested-by: Christian König <christian.koenig@amd.com> Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-09-06drm/amdgpu/: Add missing kdoc entry in amdgpu_vm_handle_fault functionSrinivasan Shanmugam1-0/+1
This commit adds a description for the 'ts' parameter in the amdgpu_vm_handle_fault function's comment block. Fixes the below with gcc W=1: drivers/gpu/drm/amd/amdgpu/amdgpu_vm.c:2781: warning: Function parameter or struct member 'ts' not described in 'amdgpu_vm_handle_fault' Cc: Xiaogang.Chen <Xiaogang.Chen@amd.com> Cc: Christian König <christian.koenig@amd.com> Cc: Alex Deucher <alexander.deucher@amd.com> Reported-by: kernel test robot <lkp@intel.com> Closes: https://lore.kernel.org/oe-kbuild-all/202408251419.vgZHg3GV-lkp@intel.com/ Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com> Reviewed-by: Xiaogang Chen <Xiaogang.Chen@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-09-06drm/amd/display: fix dccg root clock optimization related hangQili Lu4-5/+15
[Why] enable dpp rcg before we disable dppclk in hw_init cause system hang/reboot [How] we remove dccg rcg related code from init into a separate function and call it after we init pipe Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> Signed-off-by: Qili Lu <qili.lu@amd.com> Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-09-06drm/amd/display: Refactor dccg35_get_other_enabled_symclk_feNicholas Susanto1-40/+25
[Why] Function used to check the number of FEs connected to the current BE. This was then used to determine if the symclk could be disabled, if all FEs were disconnected. However, the function would skip over the primary FE and return 0 when the primary FE was still connected. This caused black screens on driver disable with an MST daisy chain hooked up. [How] Refactor the function to correctly return the number of FEs connected to the input BE. Also, rename it for clarity. Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> Signed-off-by: Nicholas Susanto <Nicholas.Susanto@amd.com> Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-09-06drm/amdgpu: Normalize reg offsets on JPEG v4.0.3Lijo Lazar1-21/+15
On VFs and SOCs with GC 9.4.4, VCN RRMT is disabled. Only local register offsets should be used on JPEG v4.0.3 as they cannot handle remote access to other AIDs. Since only local offsets are used, the special write to MCM_ADDR register is no longer needed. Signed-off-by: Lijo Lazar <lijo.lazar@amd.com> Reviewed-by: Sathishkumar S <sathishkumar.sundararaju@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-09-06drm/amd/display: Avoid race between dcn35_set_drr() and dc_state_destruct()Tobias Jakobi1-8/+12
dc_state_destruct() nulls the resource context of the DC state. The pipe context passed to dcn35_set_drr() is a member of this resource context. If dc_state_destruct() is called parallel to the IRQ processing (which calls dcn35_set_drr() at some point), we can end up using already nulled function callback fields of struct stream_resource. The logic in dcn35_set_drr() already tries to avoid this, by checking tg against NULL. But if the nulling happens exactly after the NULL check and before the next access, then we get a race. Avoid this by copying tg first to a local variable, and then use this variable for all the operations. This should work, as long as nobody frees the resource pool where the timing generators live. Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/3142 Fixes: 06ad7e164256 ("drm/amd/display: Destroy DC context while keeping DML and DML2") Signed-off-by: Tobias Jakobi <tjakobi@math.uni-bielefeld.de> Reviewed-by: Harry Wentland <harry.wentland@amd.com> Signed-off-by: Mario Limonciello <mario.limonciello@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2024-09-06drm/amd/display: Avoid race between dcn10_set_drr() and dc_state_destruct()Tobias Jakobi1-8/+12
dc_state_destruct() nulls the resource context of the DC state. The pipe context passed to dcn10_set_drr() is a member of this resource context. If dc_state_destruct() is called parallel to the IRQ processing (which calls dcn10_set_drr() at some point), we can end up using already nulled function callback fields of struct stream_resource. The logic in dcn10_set_drr() already tries to avoid this, by checking tg against NULL. But if the nulling happens exactly after the NULL check and before the next access, then we get a race. Avoid this by copying tg first to a local variable, and then use this variable for all the operations. This should work, as long as nobody frees the resource pool where the timing generators live. Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/3142 Fixes: 06ad7e164256 ("drm/amd/display: Destroy DC context while keeping DML and DML2") Signed-off-by: Tobias Jakobi <tjakobi@math.uni-bielefeld.de> Tested-by: Raoul van Rüschen <raoul.van.rueschen@gmail.com> Tested-by: Christopher Snowhill <chris@kode54.net> Reviewed-by: Harry Wentland <harry.wentland@amd.com> Tested-by: Sefa Eyeoglu <contact@scrumplex.net> Signed-off-by: Mario Limonciello <mario.limonciello@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Copyright © 1996 – 2023 Jason A. Donenfeld. All Rights Reverse Engineered.