Re: [PATCH] drm/amdgpu: correct single device PCIe reset flow for DPC

["Lazar, Lijo" <lijo.lazar@amd.com>]

On 10-Apr-26 1:00 PM, Ce Sun wrote:
> For triggering the dpc event with a single device, we still need
> to set the in_link_reset flag and the dpc status.
> 
> Signed-off-by: Ce Sun <cesun102@amd.com>

Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>

Thanks,
Lijo

> ---
>   drivers/gpu/drm/amd/amdgpu/amdgpu_device.c | 11 +++++++----
>   drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c    |  3 ++-
>   2 files changed, 9 insertions(+), 5 deletions(-)
> 
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
> index deb41c095b59..df1f6b070ea1 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
> @@ -5518,8 +5518,6 @@ static void amdgpu_device_recovery_prepare(struct amdgpu_device *adev,
>   			list_add_tail(&tmp_adev->reset_list, device_list);
>   			if (adev->shutdown)
>   				tmp_adev->shutdown = true;
> -			if (amdgpu_reset_in_dpc(adev))
> -				tmp_adev->pcie_reset_ctx.in_link_reset = true;
>   		}
>   		if (!list_is_first(&adev->reset_list, device_list))
>   			list_rotate_to_front(&adev->reset_list, device_list);
> @@ -6291,6 +6289,9 @@ pci_ers_result_t amdgpu_pci_error_detected(struct pci_dev *pdev, pci_channel_sta
>   			amdgpu_reset_set_dpc_status(adev, true);
>   
>   			mutex_lock(&hive->hive_lock);
> +		} else {
> +			if (amdgpu_device_bus_status_check(adev))
> +				amdgpu_reset_set_dpc_status(adev, true);
>   		}
>   		memset(&reset_context, 0, sizeof(reset_context));
>   		INIT_LIST_HEAD(&device_list);
> @@ -6411,6 +6412,7 @@ pci_ers_result_t amdgpu_pci_slot_reset(struct pci_dev *pdev)
>   		list_for_each_entry(tmp_adev, &hive->device_list, gmc.xgmi.head)
>   			tmp_adev->pcie_reset_ctx.in_link_reset = true;
>   	} else {
> +		adev->pcie_reset_ctx.in_link_reset = true;
>   		set_bit(AMDGPU_SKIP_HW_RESET, &reset_context.flags);
>   	}
>   
> @@ -6467,9 +6469,10 @@ void amdgpu_pci_resume(struct pci_dev *pdev)
>   			tmp_adev->pcie_reset_ctx.in_link_reset = false;
>   			list_add_tail(&tmp_adev->reset_list, &device_list);
>   		}
> -	} else
> +	} else {
> +		adev->pcie_reset_ctx.in_link_reset = false;
>   		list_add_tail(&adev->reset_list, &device_list);
> -
> +	}
>   	amdgpu_device_sched_resume(&device_list, NULL, NULL);
>   	amdgpu_device_gpu_resume(adev, &device_list, false);
>   	amdgpu_device_recovery_put_reset_lock(adev, &device_list);
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
> index 03d95dca93d7..debb82a2e031 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.c
> @@ -34,6 +34,7 @@
>   #include "amdgpu.h"
>   #include "amdgpu_pm.h"
>   #include "amdgpu_vcn.h"
> +#include "amdgpu_reset.h"
>   #include "soc15d.h"
>   
>   /* Firmware Names */
> @@ -361,7 +362,7 @@ int amdgpu_vcn_suspend(struct amdgpu_device *adev, int i)
>   
>   	/* err_event_athub and dpc recovery will corrupt VCPU buffer, so we need to
>   	 * restore fw data and clear buffer in amdgpu_vcn_resume() */
> -	if (in_ras_intr || adev->pcie_reset_ctx.in_link_reset)
> +	if (in_ras_intr || amdgpu_reset_in_dpc(adev))
>   		return 0;
>   
>   	return amdgpu_vcn_save_vcpu_bo_inst(adev, i);