Linux kernel mirror (for testing) git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git
kernel os linux
1
fork

Configure Feed

Select the types of activity you want to include in your feed.

drm/amd: Disable MES LR compute W/A

A workaround was introduced in commit 1fb710793ce2 ("drm/amdgpu: Enable
MES lr_compute_wa by default") to help with some hangs observed in gfx1151.

This WA didn't fully fix the issue. It was actually fixed by adjusting
the VGPR size to the correct value that matched the hardware in commit
b42f3bf9536c ("drm/amdkfd: bump minimum vgpr size for gfx1151").

There are reports of instability on other products with newer GC microcode
versions, and I believe they're caused by this workaround. As we don't
need the workaround any more, remove it.

Fixes: b42f3bf9536c ("drm/amdkfd: bump minimum vgpr size for gfx1151")
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 9973e64bd6ee7642860a6f3b6958cbf14e89cabd)
Cc: stable@vger.kernel.org

authored by

Mario Limonciello and committed by
Alex Deucher
6b0d8129 b57c4ec9

-10
-5
drivers/gpu/drm/amd/amdgpu/mes_v11_0.c
··· 720 720 mes_set_hw_res_pkt.enable_reg_active_poll = 1; 721 721 mes_set_hw_res_pkt.enable_level_process_quantum_check = 1; 722 722 mes_set_hw_res_pkt.oversubscription_timer = 50; 723 - if ((mes->adev->mes.sched_version & AMDGPU_MES_VERSION_MASK) >= 0x7f) 724 - mes_set_hw_res_pkt.enable_lr_compute_wa = 1; 725 - else 726 - dev_info_once(mes->adev->dev, 727 - "MES FW version must be >= 0x7f to enable LR compute workaround.\n"); 728 723 729 724 if (amdgpu_mes_log_enable) { 730 725 mes_set_hw_res_pkt.enable_mes_event_int_logging = 1;
-5
drivers/gpu/drm/amd/amdgpu/mes_v12_0.c
··· 779 779 mes_set_hw_res_pkt.use_different_vmid_compute = 1; 780 780 mes_set_hw_res_pkt.enable_reg_active_poll = 1; 781 781 mes_set_hw_res_pkt.enable_level_process_quantum_check = 1; 782 - if ((mes->adev->mes.sched_version & AMDGPU_MES_VERSION_MASK) >= 0x82) 783 - mes_set_hw_res_pkt.enable_lr_compute_wa = 1; 784 - else 785 - dev_info_once(adev->dev, 786 - "MES FW version must be >= 0x82 to enable LR compute workaround.\n"); 787 782 788 783 /* 789 784 * Keep oversubscribe timer for sdma . When we have unmapped doorbell