From: YiPeng Chai <YiPeng.Chai@amd.com>
Date: Thu, 30 Oct 2025 08:49:14 +0000 (+0800)
Subject: drm/amd/ras: Add sriov ras preprocessing before gpu reset
X-Git-Url: http://git.ipfire.org/cgi-bin/gitweb.cgi?a=commitdiff_plain;h=73c6c22694430089a6f39f1677fbf598785376d3;p=thirdparty%2Fkernel%2Flinux.git

drm/amd/ras: Add sriov ras preprocessing before gpu reset

Sriov host may clear all VF commands registered to auto
update list during VF reset, set ecc.auto_uUpdate block
to false before VF reset, and after VF reset is complete,
RAS_CMD__GET_ALL_BLOCK_ECC_STATUS command will be re-registered
to auto update list of sriov host.

Signed-off-by: YiPeng Chai <YiPeng.Chai@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---

diff --git a/drivers/gpu/drm/amd/ras/ras_mgr/amdgpu_ras_mgr.c b/drivers/gpu/drm/amd/ras/ras_mgr/amdgpu_ras_mgr.c
index 4ce337b6e0e8..cb7fbc791c3c 100644
--- a/drivers/gpu/drm/amd/ras/ras_mgr/amdgpu_ras_mgr.c
+++ b/drivers/gpu/drm/amd/ras/ras_mgr/amdgpu_ras_mgr.c
@@ -642,6 +642,9 @@ int amdgpu_ras_mgr_handle_ras_cmd(struct amdgpu_device *adev,
 
 int amdgpu_ras_mgr_pre_reset(struct amdgpu_device *adev)
 {
+	if (amdgpu_sriov_vf(adev))
+		return amdgpu_virt_ras_pre_reset(adev);
+
 	if (!amdgpu_ras_mgr_is_ready(adev)) {
 		RAS_DEV_ERR(adev, "Invalid ras suspend!\n");
 		return -EPERM;
@@ -653,6 +656,9 @@ int amdgpu_ras_mgr_pre_reset(struct amdgpu_device *adev)
 
 int amdgpu_ras_mgr_post_reset(struct amdgpu_device *adev)
 {
+	if (amdgpu_sriov_vf(adev))
+		return amdgpu_virt_ras_post_reset(adev);
+
 	if (!amdgpu_ras_mgr_is_ready(adev)) {
 		RAS_DEV_ERR(adev, "Invalid ras resume!\n");
 		return -EPERM;
diff --git a/drivers/gpu/drm/amd/ras/ras_mgr/amdgpu_virt_ras_cmd.c b/drivers/gpu/drm/amd/ras/ras_mgr/amdgpu_virt_ras_cmd.c
index b8d5482e704f..895b68785849 100644
--- a/drivers/gpu/drm/amd/ras/ras_mgr/amdgpu_virt_ras_cmd.c
+++ b/drivers/gpu/drm/amd/ras/ras_mgr/amdgpu_virt_ras_cmd.c
@@ -413,3 +413,18 @@ int amdgpu_virt_ras_hw_fini(struct amdgpu_device *adev)
 
 	return 0;
 }
+
+int amdgpu_virt_ras_pre_reset(struct amdgpu_device *adev)
+{
+	struct amdgpu_ras_mgr *ras_mgr = amdgpu_ras_mgr_get_context(adev);
+	struct amdgpu_virt_ras_cmd *virt_ras =
+		(struct amdgpu_virt_ras_cmd *)ras_mgr->virt_ras_cmd;
+
+	virt_ras->blocks_ecc.auto_update_actived = false;
+	return 0;
+}
+
+int amdgpu_virt_ras_post_reset(struct amdgpu_device *adev)
+{
+	return 0;
+}
diff --git a/drivers/gpu/drm/amd/ras/ras_mgr/amdgpu_virt_ras_cmd.h b/drivers/gpu/drm/amd/ras/ras_mgr/amdgpu_virt_ras_cmd.h
index ae7bf67b3a3b..03c3cf8363ca 100644
--- a/drivers/gpu/drm/amd/ras/ras_mgr/amdgpu_virt_ras_cmd.h
+++ b/drivers/gpu/drm/amd/ras/ras_mgr/amdgpu_virt_ras_cmd.h
@@ -49,5 +49,6 @@ int amdgpu_virt_ras_hw_init(struct amdgpu_device *adev);
 int amdgpu_virt_ras_hw_fini(struct amdgpu_device *adev);
 int amdgpu_virt_ras_handle_cmd(struct ras_core_context *ras_core,
 		struct ras_cmd_ctx *cmd);
-
+int amdgpu_virt_ras_pre_reset(struct amdgpu_device *adev);
+int amdgpu_virt_ras_post_reset(struct amdgpu_device *adev);
 #endif