* [PATCH v2] drm/amdgpu/mes: zero the sdma_hqd_mask of 2nd SDMA engine for SDMA 6.0.1
@ 2022-08-31 0:56 Yifan Zhang
2022-08-31 2:20 ` Huang, Tim
0 siblings, 1 reply; 2+ messages in thread
From: Yifan Zhang @ 2022-08-31 0:56 UTC (permalink / raw)
To: amd-gfx; +Cc: Alexander.Deucher, Tim.Huang, Jack.Xiao, Xiaojian.Du, Yifan Zhang
there is only one SDMA engine in SDMA 6.0.1, the sdma_hqd_mask has to be
zeroed for the 2nd engine, otherwise MES scheduler will consider 2nd
engine exists and map/unmap SDMA queues to the non-existent engine.
Signed-off-by: Yifan Zhang <yifan1.zhang@amd.com>
---
drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c
index fe82b8b19a4e..0c546245793b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c
@@ -181,6 +181,9 @@ int amdgpu_mes_init(struct amdgpu_device *adev)
for (i = 0; i < AMDGPU_MES_MAX_SDMA_PIPES; i++) {
if (adev->ip_versions[SDMA0_HWIP][0] < IP_VERSION(6, 0, 0))
adev->mes.sdma_hqd_mask[i] = i ? 0 : 0x3fc;
+ /* zero sdma_hqd_mask for non-existent engine */
+ else if (adev->sdma.num_instances == 1)
+ adev->mes.sdma_hqd_mask[i] = i ? 0 : 0xfc;
else
adev->mes.sdma_hqd_mask[i] = 0xfc;
}
--
2.37.1
^ permalink raw reply related [flat|nested] 2+ messages in thread
* RE: [PATCH v2] drm/amdgpu/mes: zero the sdma_hqd_mask of 2nd SDMA engine for SDMA 6.0.1
2022-08-31 0:56 [PATCH v2] drm/amdgpu/mes: zero the sdma_hqd_mask of 2nd SDMA engine for SDMA 6.0.1 Yifan Zhang
@ 2022-08-31 2:20 ` Huang, Tim
0 siblings, 0 replies; 2+ messages in thread
From: Huang, Tim @ 2022-08-31 2:20 UTC (permalink / raw)
To: Zhang, Yifan, amd-gfx; +Cc: Deucher, Alexander, Xiao, Jack, Du, Xiaojian
[AMD Official Use Only - General]
Reviewed-by: Tim Huang <Tim.Huang@amd.com>
Best Regards,
Tim Huang
-----Original Message-----
From: Zhang, Yifan <Yifan1.Zhang@amd.com>
Sent: Wednesday, August 31, 2022 8:56 AM
To: amd-gfx@lists.freedesktop.org
Cc: Huang, Tim <Tim.Huang@amd.com>; Deucher, Alexander <Alexander.Deucher@amd.com>; Du, Xiaojian <Xiaojian.Du@amd.com>; Xiao, Jack <Jack.Xiao@amd.com>; Zhang, Yifan <Yifan1.Zhang@amd.com>
Subject: [PATCH v2] drm/amdgpu/mes: zero the sdma_hqd_mask of 2nd SDMA engine for SDMA 6.0.1
there is only one SDMA engine in SDMA 6.0.1, the sdma_hqd_mask has to be zeroed for the 2nd engine, otherwise MES scheduler will consider 2nd engine exists and map/unmap SDMA queues to the non-existent engine.
Signed-off-by: Yifan Zhang <yifan1.zhang@amd.com>
---
drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c
index fe82b8b19a4e..0c546245793b 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c
@@ -181,6 +181,9 @@ int amdgpu_mes_init(struct amdgpu_device *adev)
for (i = 0; i < AMDGPU_MES_MAX_SDMA_PIPES; i++) {
if (adev->ip_versions[SDMA0_HWIP][0] < IP_VERSION(6, 0, 0))
adev->mes.sdma_hqd_mask[i] = i ? 0 : 0x3fc;
+ /* zero sdma_hqd_mask for non-existent engine */
+ else if (adev->sdma.num_instances == 1)
+ adev->mes.sdma_hqd_mask[i] = i ? 0 : 0xfc;
else
adev->mes.sdma_hqd_mask[i] = 0xfc;
}
--
2.37.1
^ permalink raw reply related [flat|nested] 2+ messages in thread
end of thread, other threads:[~2022-08-31 2:20 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2022-08-31 0:56 [PATCH v2] drm/amdgpu/mes: zero the sdma_hqd_mask of 2nd SDMA engine for SDMA 6.0.1 Yifan Zhang
2022-08-31 2:20 ` Huang, Tim
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.