Merge tag 'amd-drm-next-6.10-2024-04-13' of https://gitlab.freedesktop.org/agd5f/linux into drm-next

amd-drm-next-6.10-2024-04-13:

amdgpu:
- HDCP fixes
- ODM fixes
- RAS fixes
- Devcoredump improvements
- Misc code cleanups
- Expose VCN activity via sysfs
- SMY 13.0.x updates
- Enable fast updates on DCN 3.1.4
- Add dclk and vclk reporting on additional devices
- Add ACA RAS infrastructure
- Implement TLB flush fence
- EEPROM handling fixes
- SMUIO 14.0.2 support
- SMU 14.0.1 Updates
- Sync page table freeing with TLB flushes
- DML2 refactor
- DC debug improvements
- SR-IOV fixes
- Suspend and Resume fixes
- DCN 3.5.x Updates
- Z8 fixes
- UMSCH fixes
- GPU reset fixes
- HDP fix for second GFX pipe on GC 10.x
- Enable secondary GFX pipe on GC 10.3
- Refactor and clean up BACO/BOCO/BAMACO handling
- VCN partitioning fix
- DC DWB fixes
- VSC SDP fixes
- DCN 3.1.6 fix
- GC 11.5 fixes
- Remove invalid TTM resource start check
- DCN 1.0 fixes

amdkfd:
- MQD handling cleanup
- Preemption handling fixes for XCDs
- TLB flush fix for GC 9.4.2
- Properly clean up workqueue during module unload
- Fix memory leak process create failure
- Range check CP bad op exception targets to avoid reporting invalid exceptions to userspace

radeon:
- Misc code cleanups

From: Alex Deucher <alexander.deucher@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20240413213708.3427038-1-alexander.deucher@amd.com
Signed-off-by: Dave Airlie <airlied@redhat.com>

1 files changed, 30 insertions, 11 deletions

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c
index 7a4eae36778a..54ab51a4ada7 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_virt.c
@@ -32,6 +32,7 @@
 
 #include "amdgpu.h"
 #include "amdgpu_ras.h"
+#include "amdgpu_reset.h"
 #include "vi.h"
 #include "soc15.h"
 #include "nv.h"
@@ -424,7 +425,7 @@ static int amdgpu_virt_read_pf2vf_data(struct amdgpu_device *adev)
 		return -EINVAL;
 
 	if (pf2vf_info->size > 1024) {
-		DRM_ERROR("invalid pf2vf message size\n");
+		dev_err(adev->dev, "invalid pf2vf message size: 0x%x\n", pf2vf_info->size);
 		return -EINVAL;
 	}
 
@@ -435,7 +436,9 @@ static int amdgpu_virt_read_pf2vf_data(struct amdgpu_device *adev)
 			adev->virt.fw_reserve.p_pf2vf, pf2vf_info->size,
 			adev->virt.fw_reserve.checksum_key, checksum);
 		if (checksum != checkval) {
-			DRM_ERROR("invalid pf2vf message\n");
+			dev_err(adev->dev,
+				"invalid pf2vf message: header checksum=0x%x calculated checksum=0x%x\n",
+				checksum, checkval);
 			return -EINVAL;
 		}
 
@@ -449,7 +452,9 @@ static int amdgpu_virt_read_pf2vf_data(struct amdgpu_device *adev)
 			adev->virt.fw_reserve.p_pf2vf, pf2vf_info->size,
 			0, checksum);
 		if (checksum != checkval) {
-			DRM_ERROR("invalid pf2vf message\n");
+			dev_err(adev->dev,
+				"invalid pf2vf message: header checksum=0x%x calculated checksum=0x%x\n",
+				checksum, checkval);
 			return -EINVAL;
 		}
 
@@ -485,7 +490,7 @@ static int amdgpu_virt_read_pf2vf_data(struct amdgpu_device *adev)
 			((struct amd_sriov_msg_pf2vf_info *)pf2vf_info)->uuid;
 		break;
 	default:
-		DRM_ERROR("invalid pf2vf version\n");
+		dev_err(adev->dev, "invalid pf2vf version: 0x%x\n", pf2vf_info->version);
 		return -EINVAL;
 	}
 
@@ -571,6 +576,11 @@ static int amdgpu_virt_write_vf2pf_data(struct amdgpu_device *adev)
 	vf2pf_info->decode_usage = 0;
 
 	vf2pf_info->dummy_page_addr = (uint64_t)adev->dummy_page_addr;
+	vf2pf_info->mes_info_addr = (uint64_t)adev->mes.resource_1_gpu_addr;
+
+	if (adev->mes.resource_1) {
+		vf2pf_info->mes_info_size = adev->mes.resource_1->tbo.base.size;
+	}
 	vf2pf_info->checksum =
 		amd_sriov_msg_checksum(
 		vf2pf_info, vf2pf_info->header.size, 0, 0);
@@ -584,8 +594,22 @@ static void amdgpu_virt_update_vf2pf_work_item(struct work_struct *work)
 	int ret;
 
 	ret = amdgpu_virt_read_pf2vf_data(adev);
-	if (ret)
+	if (ret) {
+		adev->virt.vf2pf_update_retry_cnt++;
+		if ((adev->virt.vf2pf_update_retry_cnt >= AMDGPU_VF2PF_UPDATE_MAX_RETRY_LIMIT) &&
+		    amdgpu_sriov_runtime(adev) && !amdgpu_in_reset(adev)) {
+			amdgpu_ras_set_fed(adev, true);
+			if (amdgpu_reset_domain_schedule(adev->reset_domain,
+							  &adev->virt.flr_work))
+				return;
+			else
+				dev_err(adev->dev, "Failed to queue work! at %s", __func__);
+		}
+
 		goto out;
+	}
+
+	adev->virt.vf2pf_update_retry_cnt = 0;
 	amdgpu_virt_write_vf2pf_data(adev);
 
 out:
@@ -606,6 +630,7 @@ void amdgpu_virt_init_data_exchange(struct amdgpu_device *adev)
 	adev->virt.fw_reserve.p_pf2vf = NULL;
 	adev->virt.fw_reserve.p_vf2pf = NULL;
 	adev->virt.vf2pf_update_interval_ms = 0;
+	adev->virt.vf2pf_update_retry_cnt = 0;
 
 	if (adev->mman.fw_vram_usage_va && adev->mman.drv_vram_usage_va) {
 		DRM_WARN("Currently fw_vram and drv_vram should not have values at the same time!");
@@ -705,12 +730,6 @@ void amdgpu_detect_virtualization(struct amdgpu_device *adev)
 			adev->virt.caps |= AMDGPU_PASSTHROUGH_MODE;
 	}
 
-	if (amdgpu_sriov_vf(adev) && adev->asic_type == CHIP_SIENNA_CICHLID)
-		/* VF MMIO access (except mailbox range) from CPU
-		 * will be blocked during sriov runtime
-		 */
-		adev->virt.caps |= AMDGPU_VF_MMIO_ACCESS_PROTECT;
-
 	/* we have the ability to check now */
 	if (amdgpu_sriov_vf(adev)) {
 		switch (adev->asic_type) {