shinys000114/linux
1
0
Fork 0
mirror of https://github.com/hardkernel/linux.git synced 2026-06-07 19:30:30 +09:00
Code Issues Packages Projects Releases Wiki Activity

drm/amdkfd: Use same SQ prefetch setting as amdgpu

Browse Source 
commit d56b1980d7 upstream.

0 causes instruction fetch stall at cache line boundary under some
conditions on Navi10. A non-zero prefetch is the preferred default
in any case.

Fixes soft hang in Luxmark.

Signed-off-by: Jay Cornwall <jay.cornwall@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This commit is contained in:
Jay Cornwall
2020-10-17 08:38:43 -05:00
committed by Greg Kroah-Hartman
parent d417026c40
commit 1e460aa735
1 changed files with 3 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
drivers/gpu/drm/amd/amdkfd/kfd_device_queue_manager_v10.c
View File

@@ -58,8 +58,9 @@ static int update_qpd_v10(struct device_queue_manager *dqm,
/* check if sh_mem_config register already configured */
if (qpd->sh_mem_config == 0) {
qpd->sh_mem_config =
SH_MEM_ALIGNMENT_MODE_UNALIGNED <<
SH_MEM_CONFIG__ALIGNMENT_MODE__SHIFT;
(SH_MEM_ALIGNMENT_MODE_UNALIGNED <<
SH_MEM_CONFIG__ALIGNMENT_MODE__SHIFT) |
(3 << SH_MEM_CONFIG__INITIAL_INST_PREFETCH__SHIFT);
#if 0
/* TODO:
* This shouldn't be an issue with Navi10. Verify.