diff options
| author | Lancelot SIX <lancelot.six@amd.com> | 2025-01-28 19:16:49 +0000 |
|---|---|---|
| committer | Greg Kroah-Hartman <gregkh@linuxfoundation.org> | 2025-02-27 04:30:16 -0800 |
| commit | e68f4717a2928aa499d07a4836c2b47b7a626b06 (patch) | |
| tree | fe44b33f420ed6673b33d60d9a433a8a0f46ced3 /drivers/gpu | |
| parent | 1d83efc7e7dd43974f187d7c6b677f226af2e718 (diff) | |
| download | linux-e68f4717a2928aa499d07a4836c2b47b7a626b06.tar.gz linux-e68f4717a2928aa499d07a4836c2b47b7a626b06.tar.bz2 linux-e68f4717a2928aa499d07a4836c2b47b7a626b06.zip | |
drm/amdkfd: Ensure consistent barrier state saved in gfx12 trap handler
[ Upstream commit d584198a6fe4c51f4aa88ad72f258f8961a0f11c ]
It is possible for some waves in a workgroup to finish their save
sequence before the group leader has had time to capture the workgroup
barrier state. When this happens, having those waves exit do impact the
barrier state. As a consequence, the state captured by the group leader
is invalid, and is eventually incorrectly restored.
This patch proposes to have all waves in a workgroup wait for each other
at the end of their save sequence (just before calling s_endpgm_saved).
Signed-off-by: Lancelot SIX <lancelot.six@amd.com>
Reviewed-by: Jay Cornwall <jay.cornwall@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org # 6.12.x
Signed-off-by: Sasha Levin <sashal@kernel.org>
Diffstat (limited to 'drivers/gpu')
| -rw-r--r-- | drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler.h | 3 | ||||
| -rw-r--r-- | drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler_gfx12.asm | 4 |
2 files changed, 6 insertions, 1 deletions
diff --git a/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler.h b/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler.h index 02f7ba8c93cd..7062f12b5b75 100644 --- a/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler.h +++ b/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler.h @@ -4117,7 +4117,8 @@ static const uint32_t cwsr_trap_gfx12_hex[] = { 0x0000ffff, 0x8bfe7e7e, 0x8bea6a6a, 0xb97af804, 0xbe804ec2, 0xbf94fffe, - 0xbe804a6c, 0xbfb10000, + 0xbe804a6c, 0xbe804ec2, + 0xbf94fffe, 0xbfb10000, 0xbf9f0000, 0xbf9f0000, 0xbf9f0000, 0xbf9f0000, 0xbf9f0000, 0x00000000, diff --git a/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler_gfx12.asm b/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler_gfx12.asm index 1740e98c6719..7b9d36e5fa43 100644 --- a/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler_gfx12.asm +++ b/drivers/gpu/drm/amd/amdkfd/cwsr_trap_handler_gfx12.asm @@ -1049,6 +1049,10 @@ L_SKIP_BARRIER_RESTORE: s_rfe_b64 s_restore_pc_lo //Return to the main shader program and resume execution L_END_PGM: + // Make sure that no wave of the workgroup can exit the trap handler + // before the workgroup barrier state is saved. + s_barrier_signal -2 + s_barrier_wait -2 s_endpgm_saved end |