[PATCH v2] drm/amdgpu: fix gpu page fault after hibernation on PF passthrough

[PATCH v2] drm/amdgpu: fix gpu page fault after hibernation on PF passthrough

[Samuel Zhang]

On PF passthrough environment, after hibernate and then resume, coralgemm
will cause gpu page fault.

Mode1 reset happens during hibernate, but partition mode is not restored
on resume, register mmCP_HYP_XCP_CTL and mmCP_PSP_XCP_CTL is not right
after resume. When CP access the MQD BO, wrong stride size is used,
this will cause out of bound access on the MQD BO, resulting page fault.

The fix is to ensure gfx_v9_4_3_switch_compute_partition() is called
when resume from a hibernation.

Signed-off-by: Samuel Zhang <guoqing.zhang@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/aqua_vanjaram.c | 3 ++-
 drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c    | 4 +++-
 2 files changed, 5 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram.c b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram.c
index 811124ff88a8..75357e8a35b2 100644
--- a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram.c
+++ b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram.c
@@ -407,7 +407,8 @@ static int aqua_vanjaram_switch_partition_mode(struct amdgpu_xcp_mgr *xcp_mgr,
 		return -EINVAL;
 	}
 
-	if (adev->kfd.init_complete && !amdgpu_in_reset(adev))
+	if (adev->kfd.init_complete && !amdgpu_in_reset(adev) &&
+		!adev->in_s4)
 		flags |= AMDGPU_XCP_OPS_KFD;
 
 	if (flags & AMDGPU_XCP_OPS_KFD) {
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
index c4c551ef6b87..a12c72213a79 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
@@ -2291,7 +2291,9 @@ static int gfx_v9_4_3_cp_resume(struct amdgpu_device *adev)
 		r = amdgpu_xcp_init(adev->xcp_mgr, num_xcp, mode);
 
 	} else {
-		if (amdgpu_xcp_query_partition_mode(adev->xcp_mgr,
+		if (adev->in_s4) /* Restore if resuming from suspend */
+			amdgpu_xcp_restore_partition_mode(adev->xcp_mgr);
+		else if (amdgpu_xcp_query_partition_mode(adev->xcp_mgr,
 						    AMDGPU_XCP_FL_NONE) ==
 		    AMDGPU_UNKNOWN_COMPUTE_PARTITION_MODE)
 			r = amdgpu_xcp_switch_partition_mode(
-- 
2.27.0

Re: [PATCH v2] drm/amdgpu: fix gpu page fault after hibernation on PF passthrough

[Lazar, Lijo]

On 11/3/2025 9:35 AM, Samuel Zhang wrote:
> On PF passthrough environment, after hibernate and then resume, coralgemm
> will cause gpu page fault.
> 
> Mode1 reset happens during hibernate, but partition mode is not restored
> on resume, register mmCP_HYP_XCP_CTL and mmCP_PSP_XCP_CTL is not right
> after resume. When CP access the MQD BO, wrong stride size is used,
> this will cause out of bound access on the MQD BO, resulting page fault.
> 
> The fix is to ensure gfx_v9_4_3_switch_compute_partition() is called
> when resume from a hibernation.
> 
> Signed-off-by: Samuel Zhang <guoqing.zhang@amd.com>
> ---
>   drivers/gpu/drm/amd/amdgpu/aqua_vanjaram.c | 3 ++-
>   drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c    | 4 +++-
>   2 files changed, 5 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram.c b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram.c
> index 811124ff88a8..75357e8a35b2 100644
> --- a/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram.c
> +++ b/drivers/gpu/drm/amd/amdgpu/aqua_vanjaram.c
> @@ -407,7 +407,8 @@ static int aqua_vanjaram_switch_partition_mode(struct amdgpu_xcp_mgr *xcp_mgr,
>   		return -EINVAL;
>   	}
>   
> -	if (adev->kfd.init_complete && !amdgpu_in_reset(adev))
> +	if (adev->kfd.init_complete && !amdgpu_in_reset(adev) &&
> +		!adev->in_s4)

The logic should be the same for a generic suspend scenario. Please make 
it in_suspend (here and in the cp_resume call).

Thanks,
Lijo

>   		flags |= AMDGPU_XCP_OPS_KFD;
>   
>   	if (flags & AMDGPU_XCP_OPS_KFD) {
> diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
> index c4c551ef6b87..a12c72213a79 100644
> --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
> +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
> @@ -2291,7 +2291,9 @@ static int gfx_v9_4_3_cp_resume(struct amdgpu_device *adev)
>   		r = amdgpu_xcp_init(adev->xcp_mgr, num_xcp, mode);
>   
>   	} else {
> -		if (amdgpu_xcp_query_partition_mode(adev->xcp_mgr,
> +		if (adev->in_s4) /* Restore if resuming from suspend */
> +			amdgpu_xcp_restore_partition_mode(adev->xcp_mgr);
> +		else if (amdgpu_xcp_query_partition_mode(adev->xcp_mgr,
>   						    AMDGPU_XCP_FL_NONE) ==
>   		    AMDGPU_UNKNOWN_COMPUTE_PARTITION_MODE)
>   			r = amdgpu_xcp_switch_partition_mode(