summaryrefslogtreecommitdiff
path: root/sys/dev/pci/drm
AgeCommit message (Collapse)Author
2021-10-12Remove misleading uvm reference counting that isn't actually used.Mark Kettenis
Make sure uvm_obj_init() is only called once. Call uvm_obj_destroy() when we release the GEM object that wraps an uvm object for which we called uvm_obj_init(). ok mpi@, jsg@
2021-10-07drm/i915/request: fix early tracepointsJonathan Gray
From Matthew Auld d35d95e8b9da638d27bce9552262e0c486138343 in linux 5.10.y/5.10.71 c83ff0186401169eb27ce5057d820b7a863455c3 in mainline linux
2021-10-07drm/amdgpu: correct initial cp_hqd_quantum for gfx9Jonathan Gray
From Hawking Zhang 9f382e1edf90ae03be43dbd4976c2a332cd7ce2d in linux 5.10.y/5.10.71 9f52c25f59b504a29dda42d83ac1e24d2af535d4 in mainline linux
2021-10-07drm/amd/display: Pass PCI deviceid into DCJonathan Gray
From Charlene Liu c331fad63b6d527193ae8b7c056b2f10fef53c81 in linux 5.10.y/5.10.71 d942856865c733ff60450de9691af796ad71d7bc in mainline linux
2021-10-02Extend workaround for reset on context closure from gen 7-8 to gen 4-8Jonathan Gray
as asavvycomputist@disroot.org reported this occurs on gm45 (gen 4).
2021-10-01amd/display: downgrade validation failure log levelJonathan Gray
From Simon Ser 526261c1b706fec0ea80ce9f14c8fe8468bee34d in linux 5.10.y/5.10.70 7bbee36d71502ab9a341505da89a017c7ae2e6b2 in mainline linux
2021-10-01treewide: Change list_sort to use const pointersJonathan Gray
From Sami Tolvanen 55e6f8b3c0f5cc600df12ddd0371d2703b910fd7 in linux 5.10.y/5.10.70 4f0f586bf0c898233d8f316f471a21db2abd522d in mainline linux
2021-10-01drm/amd/pm: Update intermediate power state for SIJonathan Gray
From Lijo Lazar 68d4fbe6220cd1f3d07cab0a4901e62f8c12cc68 in linux 5.10.y/5.10.70 ab39d3cef526ba09c4c6923b4cd7e6ec1c5d4faa in mainline linux
2021-09-29syncJonathan Gray
2021-09-29drm/amdgpu: add some additional RDNA2 PCI IDsJonathan Gray
From Alex Deucher 8f0c93f454bd7ab04eaec1d3c436c4c7c2378f07 in mainline linux
2021-09-27drm/amdgpu: Disable PCIE_DPM on Intel RKL PlatformJonathan Gray
From Koba Ko 45bd9dd1bee8aedc4cbd409b1ba7f9b4f941eea6 in linux 5.10.y/5.10.69 b3dc549986eb7b38eba4a144e979dc93f386751f in mainline linux
2021-09-27drm/amd/amdgpu: Increase HWIP_MAX_INSTANCE to 10Jonathan Gray
From Ernst Sjoestrand 8f95553f0016c3994d9c022b5af4a1a433d6714e in linux 5.10.y/5.10.68 67a44e659888569a133a8f858c8230e9d7aad1d5 in mainline linux
2021-09-27drm/amd/display: Update bounding box states (v2)Jonathan Gray
From Jerry (Fangzhi) Zuo b80a99e048275d566d63f2463a2f640065ccbf75 in linux 5.10.y/5.10.67 a7a9d11e12fcc32160d55e8612e72e5ab51b15dc in mainline linux
2021-09-27drm/amd/display: Update number of DCN3 clock statesJonathan Gray
From Aurabindo Pillai 583c4f3d09c3e980a683b59febbb0c775bdff1db in linux 5.10.y/5.10.67 0bbf06d888734041e813b916d7821acd4f72005a in mainline linux
2021-09-27drm/amdgpu: Fix BUG_ON assertJonathan Gray
From Andrey Grodzovsky 7b1abace16a9dff6804d4eb94750beb60d9502b4 in linux 5.10.y/5.10.67 ea7acd7c5967542353430947f3faf699e70602e5 in mainline linux
2021-09-27drm/dp_mst: Fix return code on sideband message failureJonathan Gray
From Rajkumar Subbiah bb693c114e8b53e3e0b8228be218d907d35959a5 in linux 5.10.y/5.10.67 92bd92c44d0d9be5dcbcda315b4be4b909ed9740 in mainline linux
2021-09-27drm/amdkfd: Account for SH/SE count when setting up cu masks.Jonathan Gray
From Sean Keely 0e9f4492219f8f991163691aad43897da8478c4e in linux 5.10.y/5.10.67 1ec06c2dee679e9f089e78ed20cb74ee90155f61 in mainline linux
2021-09-27drm/display: fix possible null-pointer dereference in dcn10_set_clock()Jonathan Gray
From Tuo Li 83449db3aac0895147eac723bf23d0739720b968 in linux 5.10.y/5.10.67 554594567b1fa3da74f88ec7b2dc83d000c58e98 in mainline linux
2021-09-27gpu: drm: amd: amdgpu: amdgpu_i2c: fix possible uninitialized-variable ↵Jonathan Gray
access in amdgpu_i2c_router_select_ddc_port() From Tuo Li 2254383788ff93a423e20068333b9f8376d56cb4 in linux 5.10.y/5.10.67 a211260c34cfadc6068fece8c9e99e0fe1e2a2b6 in mainline linux
2021-09-27drm/amd/display: fix incorrect CM/TF programming sequence in dwbJonathan Gray
From Roy Chan 63ebc1f1df813ebb40d19449c356480555008166 in linux 5.10.y/5.10.67 781e1e23131cce56fb557e6ec2260480a6bd08cc in mainline linux
2021-09-27drm/amd/display: fix missing writeback disablement if plane is removedJonathan Gray
From Roy Chan d763afc4ea2b251217ec87cf4c1e006c9f0aef99 in linux 5.10.y/5.10.67 82367e7f22d085092728f45fd5fbb15e3fb997c0 in mainline linux
2021-09-27drm/amd/amdgpu: Update debugfs link_settings output link_rate field in hexJonathan Gray
From Anson Jacob 6f51f4241253974a6a147daecd5c20beb7450330 in linux 5.10.y/5.10.67 1a394b3c3de2577f200cb623c52a5c2b82805cec in mainline linux
2021-09-27drm/amdgpu: Fix a printing messageJonathan Gray
From Oak Zeng a5999d18a8d8c4c767c60d67fe6a6fe51b9a203d in linux 5.10.y/5.10.67 95f71f12aa45d65b7f2ccab95569795edffd379a in mainline linux
2021-09-27drm/amd/display: Fix timer_per_pixel unit errorJonathan Gray
From Oliver Logush f462a39eb8334b52e332cc0cbffb705660b7d87b in linux 5.10.y/5.10.67 23e55639b87fb16a9f0f66032ecb57060df6c46c in mainline linux
2021-09-27drm: protect drm_master pointers in drm_lease.cJonathan Gray
From Desmond Cheong Zhi Xi 34609faad0c9f9f08d4b59d25c94b78bf5710d93 in linux 5.10.y/5.10.67 56f0729a510f92151682ff6c89f69724d5595d6e in mainline linux
2021-09-27drm: serialize drm_file.master with a new spinlockJonathan Gray
From Desmond Cheong Zhi Xi 06a553a99bacb00d3bc25f79e75c8e0fbf7a5025 in linux 5.10.y/5.10.67 0b0860a3cf5eccf183760b1177a1dcdb821b0b66 in mainline linux
2021-09-27drm: avoid blocking in drm_clients_info's rcu sectionJonathan Gray
From Desmond Cheong Zhi Xi 54e51d288b38377e8cd645a83e1ad08cc9d20ccc in linux 5.10.y/5.10.67 5eff9585de220cdd131237f5665db5e6c6bdf590 in mainline linux
2021-09-27drm/amdgpu: Fix amdgpu_ras_eeprom_init()Jonathan Gray
From Luben Tuikov 10a135969fd7419695c003ddb67ef8a7820a808b in linux 5.10.y/5.10.67 dce4400e6516d18313d23de45b5be8a18980b00e in mainline linux
2021-09-22Add a workaround for machines where the framebuffer size reported by theMark Kettenis
hardware is incorrect. In this case, make sure the amount of "stolen" memory is at least as large as the EFI framebuffer such that the driver doesn't use this memory until we've switched to the framebuffer allocated by the amdgpu(4) driver. Needs further investigation why the size reported by the hardware is incorrect. Tested by djm@ ok jsg@, deraadt@
2021-09-16drm/amdgpu/acp: Make PM domain really workJonathan Gray
From Kai-Heng Feng 1f60072320b5f8071946e4b765cbf78a34d22a67 in linux 5.10.y/5.10.65 aff890288de2d818e4f83ec40c9315e2d735df07 in mainline linux
2021-09-09drm: Copy drm_wait_vblank to user before returningJonathan Gray
From Mark Yacoub 6fd6e20520ccd05a1ac3321404dd498cc28576cb in linux 5.10.y/5.10.62 fa0b1ef5f7a694f48e00804a391245f3471aa155 in mainline linux
2021-09-09drm/amd/pm: change the workload type for some cardsJonathan Gray
From Kenneth Feng b00ca567579a4c2f9a4cd6f9a63946f793e8b506 in linux 5.10.y/5.10.62 93c5701b00d50d192ce2247cb10d6c0b3fe25cd8 in mainline linux
2021-09-09Revert "drm/amd/pm: fix workload mismatch on vega10"Jonathan Gray
From Kenneth Feng 3c37ec4350220a548ffc6753646913899e86b1c7 in linux 5.10.y/5.10.62 2fd31689f9e44af949f60ff4f8aca013e628ab81 in mainline linux
2021-09-09drm/i915: Fix syncmap memory leakJonathan Gray
From Matthew Brost 257ea8a5edc04d5199db83137fbaa24e1de98e9e in linux 5.10.y/5.10.62 a63bcf08f0efb5348105bb8e0e1e8c6671077753 in mainline linux
2021-09-09drm/amdgpu: Cancel delayed work when GFXOFF is disabledJonathan Gray
From Michel Daenzer da3067eadcc156b742657c0694beae0a7c49d157 in linux 5.10.y/5.10.62 32bc8f8373d2d6a681c96e4b25dca60d4d1c6016 in mainline linux
2021-09-09Don't print references to Intel's website for downloading firmwareJoshua Stein
when loading fails, it will just confuse people ok jsg
2021-09-05Pass `uobj' to uvmfault_unlockall() at the end of the fault handler.Martin Pieuchot
This is currently a NOOP but will become necessary to unlock the UVM object with the upcoing "vmobjlock" diff. Tested by patrick@ and robert@ ok jsg@
2021-09-02Initialize the uvm object when we actually need it such that we canMark Kettenis
use uvm_obj_init() to initialize the pager ops and initial reference count. This will help future uvm unlocking diffs. ok mpi@, jsg@
2021-08-28Remove bogus code that tries to unmap GTT pages in a codepath that clearlyMark Kettenis
doesn't deal with non-GTT mappings. What the Linux code does here isn't possible on OpenBSD and probably unecessary. Seems to fix a crash reported by sthen@ ok jsg@
2021-08-27drm/amd/display: workaround for hard hang on HPD on native DPJonathan Gray
From Qingqing Zhuo 2e6cc93e1b8cf3ec2966961c1e98722ee7281023 in linux 5.10.y/5.10.61 c4152b297d56d3696ad0a9003169bc5b98ad7b72 in mainline linux
2021-08-27drm/amd/display: Fix Dynamic bpp issue with 8K30 with Navi 1XJonathan Gray
From Bing Guo dcc8c5fb8d8595f5061c7b000ca1d16449a5e865 in linux 5.10.y/5.10.61 06050a0f01dbac2ca33145ef19a72041206ea983 in mainline linux
2021-08-27drm/amdgpu: fix the doorbell missing when in CGPG issue for renoir.Jonathan Gray
From Yifan Zhang 7525f2e4de0069983497a9d3eab1ca9813ae1b4b in linux 5.10.y/5.10.61 1c0539a6fc8a4a4b77278e35d763073890de96b9 in mainline linux
2021-08-18drm/i915: Only access SFC_DONE when media domain is not fused offJonathan Gray
From Matt Roper 65395b053d03cb662e63cbf2c7e0faef8c15cb8c in linux 5.10.y/5.10.60 24d032e2359e3abc926b3d423f49a7c33e0b7836 in mainline linux
2021-08-18drm/amdgpu: don't enable baco on boco platforms in runpmJonathan Gray
From Alex Deucher 95de3592f87e46df63119dd52b4a0e544e519c6b in linux 5.10.y/5.10.60 202ead5a3c589b0594a75cb99f080174f6851fed in mainline linux
2021-08-18drm/amd/display: use GFP_ATOMIC in amdgpu_dm_irq_schedule_workJonathan Gray
From Anson Jacob bd80d11a516c78fb74d11e69c67082f36f8ef8e3 in linux 5.10.y/5.10.60 0cde63a8fc4d9f9f580c297211fd05f91c0fd66d in mainline linux
2021-08-18drm/amd/display: Remove invalid assert for ODM + MPC caseJonathan Gray
From Eric Bernstein ae311a7418f13be375c29ec4178baa51cd9101ba in linux 5.10.y/5.10.60 c90f6263f58a28c3d97b83679d6fd693b33dfd4e in mainline linux
2021-08-14call task_del() from destroy_work_on_stack()Jonathan Gray
prompted by jcs@ reporting a protection fault trap in drm_mode_rmfb_work_fn() while playing a youtube video in firefox on a kaby lake machine. He later saw the same trace on tiger lake. The previous attempt to avoid this situation by changing work flush functions from taskq_barrier() to taskq_del_barrier() resulted in suspend sometimes not working on various intel based thinkpads. The only code we build which calls destroy_work_on_stack() is in drm_framebuffer.c so the scope of this change is more limited. Linux only uses destroy_work_on_stack() for debugging so the workqueue behaviour still doesn't match. This version is confirmed to not break suspend on x260 by sthen@ and x280 by tb@ and still avoids the original problem according to jcs@
2021-08-12drm/amdgpu/display: only enable aux backlight control for OLED panelsJonathan Gray
From Alex Deucher bb65051dcd1fd380a73ca52c87f89522e15bf62d in linux 5.10.y/5.10.58 f2ad3accefc63e72e9932e141c21875cc04beec8 in mainline linux
2021-08-12drm/i915: Correct SFC_DONE register offsetJonathan Gray
From Matt Roper 7397034905acaecbc64f6838779bdc81667e682f in linux 5.10.y/5.10.58 9c9c6d0ab08acfe41c9f7efa72c4ad3f133a266b in mainline linux
2021-08-12drm/amdgpu/display: fix DMUB firmware version infoJonathan Gray
From Shirish S dd3f7c5c890450ab2ad6f269a3fdf7bcd6fc2908 in linux 5.10.y/5.10.58 0e99e960ce6d5ff586fc0733bc393c087f52c27b in mainline linux