- 09 Apr, 2021 40 commits
-
-
Guchun Chen authored
Otherwise, below errors will be found on DIMGREY_CAVEFISH with DCN302. Error log observed in driver load: [drm:amdgpu_dm_irq_register_interrupt [amdgpu]] *ERROR* DM_IRQ: invalid irq_source: 0! Error observed in mode1_rest sequence: [ 27.265920] #PF: supervisor read access in kernel mode [ 27.265941] #PF: error_code(0x0000) - not-present page [ 27.265956] PGD 0 P4D 0 [ 27.265967] Oops: 0000 [#1] SMP NOPTI [ 27.265979] CPU: 0 PID: 1360 Comm: cat Tainted: G W 5.11.0-99b3786c1770 #20210323 [ 27.266005] Hardware name: System manufacturer System Product Name/PRIME Z390-A, BIOS 1401 11/26/2019 [ 27.266033] RIP: 0010:dal_irq_service_ack+0x25/0x60 [amdgpu] [ 27.266203] Code: 5d 5d c3 66 90 0f 1f 44 00 00 55 83 fe 61 48 89 e5 77 27 89 f0 48 8d 04 40 48 c1 e0 04 48 03 47 08 74 17 48 8b 50 28 48 89 c6 <48> 8b 52 08 48 85 d2 74 20 e8 1d 64 45 c9 5d c3 89 f2 48 c7 c7 f0 [ 27.266248] RSP: 0018:ffffa115824a3c08 EFLAGS: 00010082 [ 27.266270] RAX: ffffffffc0942c10 RBX: ffffffffc0942c10 RCX: 0000000000000000 [ 27.266288] RDX: 0000000000000000 RSI: ffffffffc0942c10 RDI: ffff88d509cba7a0 [ 27.266312] RBP: ffffa115824a3c08 R08: 0000000000000000 R09: 0000000000000001 [ 27.266335] R10: ffffa115824a3b20 R11: ffffa115824a3b58 R12: ffff88d509cba7a0 [ 27.266353] R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000000246 [ 27.266377] FS: 00007fb3e2438580(0000) GS:ffff88d50dc00000(0000) knlGS:0000000000000000 [ 27.266402] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 27.266417] CR2: 0000000000000008 CR3: 000000083e5ac006 CR4: 00000000003706f0 [ 27.266441] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 27.266464] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [ 27.266483] Call Trace: [ 27.266491] dal_irq_service_set+0x31/0x80 [amdgpu] [ 27.266639] dc_interrupt_set+0x24/0x30 [amdgpu] [ 27.266775] amdgpu_dm_set_dmub_trace_irq_state+0x22/0x30 [amdgpu] [ 27.266920] amdgpu_irq_update+0x57/0xa0 [amdgpu] [ 27.267030] amdgpu_irq_gpu_reset_resume_helper+0x64/0xa0 [amdgpu] [ 27.267135] amdgpu_do_asic_reset+0x211/0x420 [amdgpu] [ 27.267232] amdgpu_device_gpu_recover+0x517/0xa70 [amdgpu] [ 27.267325] gpu_recover_get+0x2e/0x60 [amdgpu] [ 27.267421] simple_attr_read+0x6b/0x130 [ 27.267441] debugfs_attr_read+0x49/0x70 [ 27.267454] full_proxy_read+0x5c/0x90 [ 27.267474] vfs_read+0xa1/0x190 [ 27.267486] ksys_read+0xa7/0xe0 [ 27.267501] __x64_sys_read+0x1a/0x20 [ 27.267521] do_syscall_64+0x37/0x80 [ 27.267541] entry_SYSCALL_64_after_hwframe+0x44/0xa9 [ 27.267562] RIP: 0033:0x7fb3e2356142 Signed-off-by:
Guchun Chen <guchun.chen@amd.com> Reviewed-by:
Leo (Hanghong) Ma <hanghong.ma@amd.com> Signed-off-by:
Alex Deucher <alexander.deucher@amd.com>
-
Rohit Khaire authored
Add 3 sub flags to notify guest for indirect reg access of gc, mmhub and ih The host sets these flags depending on L1 RAP version, asic and other scenarios. These flags ensure that there is compatibility between different guest/host/vbios versions. Signed-off-by:
Rohit Khaire <rohit.khaire@amd.com> Reviewed-by:
Monk Liu <monk.liu@amd.com> Acked-by:
Luben Tuikov <luben.tuikov@amd.com> Signed-off-by:
-
Alex Deucher authored
Do the same thing we do for Renoir. We can check, but since the sbios has started DPM, it will always return true which causes the driver to skip some of the SMU init when it shouldn't. Reviewed-by:
Zhan Liu <zhan.liu@amd.com> Acked-by:
Evan Quan <evan.quan@amd.com> Signed-off-by:
-
Diego Viola authored
Signed-off-by:
Diego Viola <diego.viola@gmail.com> Signed-off-by:
-
Werner Sembach authored
When encoder validation of a display mode fails, retry with less bandwidth heavy YCbCr420 color mode, if available. This enables some HDMI 1.4 setups to support 4k60Hz output, which previously failed silently. On some setups, while the monitor and the gpu support display modes with pixel clocks of up to 600MHz, the link encoder might not. This prevents YCbCr444 and RGB encoding for 4k60Hz, but YCbCr420 encoding might still be possible. However, which color mode is used is decided before the link encoder capabilities are checked. This patch fixes the problem by retrying to find a display mode with YCbCr420 enforced and using it, if it is valid. Reviewed-by:
Harry Wentland <harry.wentland@amd.com> Signed-off-by:
Werner Sembach <wse@tuxedocomputers.com> Signed-off-by:
-
Aric Cyr authored
This version brings along following features/fixes: - LTTPR improvements - Disable MALL when SMU not present - Fix bug in HW that causes P-State to hang when DPG is enabled in certain conditions - Update code path for enabling DPG - Update display endpoint control path - Add dynamic link encoder selection - Fix debugfs MST topology to dump from the root MST node - Enable DP DSC Compliance automation for Linux - ASSR is enabled only when DPCD is supported and the display connected is internal - Added kernel trace event to print real-time refresh rate value to debug VRR issues Signed-off-by:
Aric Cyr <aric.cyr@amd.com> Acked-by:
Anson Jacob <Anson.Jacob@amd.com> Tested-by:
Dan Wheeler <daniel.wheeler@amd.com> Signed-off-by:
-
Chris Park authored
[Why] Bring-up purpose code to disable DMUB calling into SMU and timeout for MALL when SMU is not present. [How] Disable MALL when SMU is not present. Signed-off-by:
Chris Park <Chris.Park@amd.com> Reviewed-by:
Nicholas Kazlauskas <Nicholas.Kazlauskas@amd.com> Acked-by:
-
Anthony Koo authored
Signed-off-by:
Anthony Koo <Anthony.Koo@amd.com> Acked-by:
-
Qingqing Zhuo authored
[Why&How] Add log for easier debug purposes. Signed-off-by:
Qingqing Zhuo <qingqing.zhuo@amd.com> Reviewed-by:
-
Wesley Chalmers authored
[WHY] There is a bug in HW that causes P-State to hang when DPG is enabled in certain conditions. [HOW] The solution is to force MIN_TTU_VBLANK register to maximum value whenever DPG has been enabled. Make stream do a full update on test pattern change, so that the TTUs get updated. When DPG is enabled, update the ttu_regs.min_ttu_vblank field of each pipe in the stream's topology to the maximum value (0xffffff). v2: squash in build fix for when DCN is not defined (Alex) Signed-off-by:
Wesley Chalmers <Wesley.Chalmers@amd.com> Reviewed-by:
Tony Cheng <Tony.Cheng@amd.com> Acked-by:
-
Wesley Chalmers authored
[WHY] We want to make enabling test pattern a part of the stream update code path. This change is the first step towards that goal. Signed-off-by:
-
Jimmy Kizito authored
[Why] Some display endpoints may be dynamically mapped to the link encoders which drive them. [How] Update the code paths for display enabling/disabling to accommodate the dynamic association between links and link encoders. Signed-off-by:
Jimmy Kizito <Jimmy.Kizito@amd.com> Reviewed-by:
Jun Lei <Jun.Lei@amd.com> Acked-by:
-
Jimmy Kizito authored
[Why] Some display endpoints may be programmably mapped to compatible link encoders. The assignment of link encoders to links has to be dynamic to accommodate the increased flexibility in comparison to conventional display endpoints. [How] - Add link encoder assignment tracking variables. - Execute link encoder assignment algorithm before enabling link and release link encoders from links once they are disabled. Signed-off-by:
-
Eryk Brol authored
[why] The drm dump_topology function was previously called on all DP connectors. This resulted in empty topology dumps for those connectors which weren't root MST nodes. [how] Make sure we only dump topology from the root MST node. Signed-off-by:
Eryk Brol <eryk.brol@amd.com> Reviewed-by:
Aurabindo Jayamohanan Pillai <Aurabindo.Pillai@amd.com> Acked-by:
-
Wesley Chalmers authored
[WHY] Some systems can enable LTTPR through bits in BIOS, while other systems can be configured at boot to enable LTTPR. Some configs enable Non-Transparent mode, while others enable Transparent mode. Signed-off-by:
-
Wesley Chalmers authored
[WHY] There are three possible modes for LTTPR: - Non-LTTPR mode, where AUX timeout is 400 us and no per-hop link training is done - LTTPR Transparent mode, where AUX timeout is 3200 us and no per-hop link training is done - LTTPR Non-Transparent mode, where AUX timeout is 3200 us and per-hop link training is done [HOW] Use an enum instead of a bool to track LTTPR state; modify comparisons accordingly. Signed-off-by:
-
Wesley Chalmers authored
[WHY] The logic to toggle LTTPR transparent/non-transparent requires 2 flags provided by BIOS [HOW] Repurpose the interface to get dce caps so both LTTPR querying functions can use them. Signed-off-by:
-
Krunoslav Kovac authored
[Why&How] Renaming structure to better indicate its meaning. Signed-off-by:
Krunoslav Kovac <Krunoslav.Kovac@amd.com> Reviewed-by:
-
Vladimir Stempen authored
[why] Word "remainder" was misspelled as "reminder" in reduceSizeAndFraction method variable. [how] Fix the spelling. Signed-off-by:
Vladimir Stempen <vladimir.stempen@amd.com> Reviewed-by:
Bindu R <Bindu.R@amd.com> Acked-by:
-
Wenjing Liu authored
Signed-off-by:
Wenjing Liu <wenjing.liu@amd.com> Reviewed-by:
George Shen <George.Shen@amd.com> Acked-by:
-
Wenjing Liu authored
Signed-off-by:
-
Qingqing Zhuo authored
[Why] Color depth data is not parsed during test requests. [How] Update display color depth according to color depth request from the test equipment. Signed-off-by:
-
Stylon Wang authored
[Why] ASSR enabling only considers capability declared in DPCD. We also need to check whether the connector is internal. [How] ASSR enabling need to check both DPCD capability and internal display flag. Signed-off-by:
Stylon Wang <stylon.wang@amd.com> Reviewed-by:
-
Leo (Hanghong) Ma authored
[Why] Static analysis on linux-next has found a potential null pointer dereference; [How] Refactor the function, add ASSERT and remove the unnecessary check. Signed-off-by:
-
Rodrigo Siqueira authored
When we have to debug VRR issues, we usually want to know the current refresh rate; for this reason, it is handy to have a way to check in real-time the refresh rate value. This commit introduces a kernel trace that can provide such information. Signed-off-by:
Rodrigo Siqueira <Rodrigo.Siqueira@amd.com> Reviewed-by:
-
Wesley Chalmers authored
[WHY] Some platforms will have LTTPR capabilities forced on by VBIOS flags; the functions added here will access those flags. Signed-off-by:
-
Feifei Xu authored
This message is not needed on Aldebaran. Signed-off-by:
Feifei Xu <Feifei.Xu@amd.com> Reviewed-by:
Lijo Lazar <lijo.lazar@amd.com> Signed-off-by:
-
Chengming Gui authored
When resume from gpu reset, need set MP1 state to UNLOAD before reload SMU FW otherwise will cause following errors: [ 121.642772] [drm] reserve 0x400000 from 0x87fec00000 for PSP TMR [ 123.801051] [drm] failed to load ucode id (24) [ 123.801055] [drm] psp command (0x6) failed and response status is (0x0) [ 123.801214] [drm:psp_load_smu_fw [amdgpu]] *ERROR* PSP load smu failed! [ 123.801398] [drm:psp_resume [amdgpu]] *ERROR* PSP resume failed [ 123.801536] [drm:amdgpu_device_fw_loading [amdgpu]] *ERROR* resume of IP block <psp> failed -22 [ 123.801632] amdgpu 0000:04:00.0: amdgpu: GPU reset(9) failed [ 123.801691] amdgpu 0000:07:00.0: amdgpu: GPU reset(9) failed [ 123.802899] amdgpu 0000:04:00.0: amdgpu: GPU reset end with ret = -22 v2: add error info and including ALDEBARAN also Signed-off-by:
Chengming Gui <Jack.Gui@amd.com> Reviewed-and-tested-by:
-
Lijo Lazar authored
If reset handler is not implemented, reset error before proceeding. Fixes issue with the following trace - [ 106.508592] amdgpu 0000:b1:00.0: amdgpu: ASIC reset failed with error, -38 for drm dev, 0000:b1:00.0 [ 106.508972] amdgpu 0000:b1:00.0: amdgpu: GPU reset succeeded, trying to resume [ 106.509116] [drm] PCIE GART of 512M enabled. [ 106.509120] [drm] PTB located at 0x0000008000000000 [ 106.509136] [drm] VRAM is lost due to GPU reset! [ 106.509332] [drm] PSP is resuming... Signed-off-by:
-
Nikola Cornij authored
[why] This was a regression introduced by commit: drm/amd/display: Skip modeset for front porch change Due to the change how timing parameters were set, scaled modes would cause a black screen on some eDP panels. Would probably apply to other displays (i.e. even non-eDP) that only have scaled modes, but such case is not that usual for external displays. [how] Pick up crtc frame dimensions when programming the timing unless it's FreeSync video mode. Fixes: 6f59f229 ("drm/amd/display: Skip modeset for front porch change") Signed-off-by:Nikola Cornij <nikola.cornij@amd.com> Reviewed-by:
-
Alex Sierra authored
Starting Arcturus, it supports ih reroute through mmio directly in bare metal environment. This is also valid for newer asics such as Aldebaran. Signed-off-by:
Alex Sierra <alex.sierra@amd.com> Acked-by:
Christian König <christian.koenig@amd.com> Reviewed-by:
Hawking Zhang <Hawking.Zhang@amd.com> Reviewed-by:
Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by:
-
Qu Huang authored
Amdgpu driver uses 4-byte data type as DQM fence memory, and transmits GPU address of fence memory to microcode through query status PM4 message. However, query status PM4 message definition and microcode processing are all processed according to 8 bytes. Fence memory only allocates 4 bytes of memory, but microcode does write 8 bytes of memory, so there is a memory corruption. Changes since v1: * Change dqm->fence_addr as a u64 pointer to fix this issue, also fix up query_status and amdkfd_fence_wait_timeout function uses 64 bit fence value to make them consistent. Signed-off-by:
Qu Huang <jinsdb@126.com> Reviewed-by:
-
Nirmoy Das authored
Offset calculation wasn't correct as start addresses are in pfn not in bytes. Signed-off-by:
Nirmoy Das <nirmoy.das@amd.com> Reviewed-by:
-
Tom St Denis authored
Signed-off-by:
Tom St Denis <tom.stdenis@amd.com> Reviewed-by:
-
Lijo Lazar authored
Firmware returns zero-based max level, increment by one to get total levels. This fixes the issue of not showing all levels and current frequency when frequency is at max DPM level. Signed-off-by:
-
Evan Quan authored
No need to have special handling for swSMU supported ASICs. Signed-off-by:
-
Evan Quan authored
No need to have special handling for swSMU supported ASICs. Signed-off-by:
-
Evan Quan authored
Add "static" declarations for those APIs used internally. Signed-off-by:
-
Evan Quan authored
No need to have special handling for swSMU supported ASICs. Signed-off-by:
-
Evan Quan authored
Correct the check for vblank short. Signed-off-by:
-
