drm/amdgpu: Implement Runtime Bad Page query for VFs

Host will send a notification when new bad pages are available.

Uopn guest request, the first 256 bad page addresses
will be placed into the PF2VF region.
Guest should pause the PF2VF worker thread while
the copy is in progress.

Reviewed-by: Shravan Kumar Gande <Shravankumar.Gande@amd.com>
Signed-off-by: Victor Skvortsov <victor.skvortsov@amd.com>
Signed-off-by: Ellen Pan <yunru.pan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

1 files changed, 28 insertions, 0 deletions

diff --git a/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.c b/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.c
index 5aadf24cb202..74a50c0036ef 100644
--- a/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.c
+++ b/drivers/gpu/drm/amd/amdgpu/mxgpu_nv.c
@@ -187,6 +187,9 @@ send_request:
 	case IDH_REQ_RAS_CPER_DUMP:
 		event = IDH_RAS_CPER_DUMP_READY;
 		break;
+	case IDH_REQ_RAS_BAD_PAGES:
+		event = IDH_RAS_BAD_PAGES_READY;
+		break;
 	default:
 		break;
 	}
@@ -342,6 +345,19 @@ static void xgpu_nv_mailbox_flr_work(struct work_struct *work)
 	}
 }
 
+static void xgpu_nv_mailbox_bad_pages_work(struct work_struct *work)
+{
+	struct amdgpu_virt *virt = container_of(work, struct amdgpu_virt, bad_pages_work);
+	struct amdgpu_device *adev = container_of(virt, struct amdgpu_device, virt);
+
+	if (down_read_trylock(&adev->reset_domain->sem)) {
+		amdgpu_virt_fini_data_exchange(adev);
+		amdgpu_virt_request_bad_pages(adev);
+		amdgpu_virt_init_data_exchange(adev);
+		up_read(&adev->reset_domain->sem);
+	}
+}
+
 static int xgpu_nv_set_mailbox_rcv_irq(struct amdgpu_device *adev,
 				       struct amdgpu_irq_src *src,
 				       unsigned type,
@@ -366,6 +382,11 @@ static int xgpu_nv_mailbox_rcv_irq(struct amdgpu_device *adev,
 	enum idh_event event = xgpu_nv_mailbox_peek_msg(adev);
 
 	switch (event) {
+	case IDH_RAS_BAD_PAGES_NOTIFICATION:
+		xgpu_nv_mailbox_send_ack(adev);
+		if (amdgpu_sriov_runtime(adev))
+			schedule_work(&adev->virt.bad_pages_work);
+		break;
 	case IDH_FLR_NOTIFICATION:
 		if (amdgpu_sriov_runtime(adev))
 			WARN_ONCE(!amdgpu_reset_domain_schedule(adev->reset_domain,
@@ -436,6 +457,7 @@ int xgpu_nv_mailbox_get_irq(struct amdgpu_device *adev)
 	}
 
 	INIT_WORK(&adev->virt.flr_work, xgpu_nv_mailbox_flr_work);
+	INIT_WORK(&adev->virt.bad_pages_work, xgpu_nv_mailbox_bad_pages_work);
 
 	return 0;
 }
@@ -480,6 +502,11 @@ static int xgpu_nv_req_ras_cper_dump(struct amdgpu_device *adev, u64 vf_rptr)
 		adev, IDH_REQ_RAS_CPER_DUMP, vf_rptr_hi, vf_rptr_lo, 0);
 }
 
+static int xgpu_nv_req_ras_bad_pages(struct amdgpu_device *adev)
+{
+	return xgpu_nv_send_access_requests(adev, IDH_REQ_RAS_BAD_PAGES);
+}
+
 const struct amdgpu_virt_ops xgpu_nv_virt_ops = {
 	.req_full_gpu	= xgpu_nv_request_full_gpu_access,
 	.rel_full_gpu	= xgpu_nv_release_full_gpu_access,
@@ -492,4 +519,5 @@ const struct amdgpu_virt_ops xgpu_nv_virt_ops = {
 	.rcvd_ras_intr = xgpu_nv_rcvd_ras_intr,
 	.req_ras_err_count = xgpu_nv_req_ras_err_count,
 	.req_ras_cper_dump = xgpu_nv_req_ras_cper_dump,
+	.req_bad_pages = xgpu_nv_req_ras_bad_pages,
 };