From: YiPeng Chai <YiPeng.Chai@amd.com>
Date: Mon, 21 Jul 2025 07:14:03 +0000 (+0800)
Subject: drm/amdgpu: Improve ras fatal error handling function
X-Git-Url: http://git.ipfire.org/gitweb.cgi?a=commitdiff_plain;h=408bd841ad24fd9891a1cd9ca11cf4f48d9667dc;p=thirdparty%2Fkernel%2Flinux.git

drm/amdgpu: Improve ras fatal error handling function

In multi-gpu case, a fatal error will generate several
fatal error interrupts. After improving this function,
the ras module can reuse this function to only
handle the first interrupt.

V3:
  Initialize event_id using RAS_EVENT_INVALID_ID.

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/amdgpu/amdgpu_ras.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
index 9aa4b93ac6afd..9e632adc7746c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.c
@@ -4650,20 +4650,18 @@ u64 amdgpu_ras_acquire_event_id(struct amdgpu_device *adev, enum ras_event_type
 	return id;
 }
 
-void amdgpu_ras_global_ras_isr(struct amdgpu_device *adev)
+int amdgpu_ras_global_ras_isr(struct amdgpu_device *adev)
 {
 	if (atomic_cmpxchg(&amdgpu_ras_in_intr, 0, 1) == 0) {
 		struct amdgpu_ras *ras = amdgpu_ras_get_context(adev);
 		enum ras_event_type type = RAS_EVENT_TYPE_FATAL;
-		u64 event_id;
+		u64 event_id = RAS_EVENT_INVALID_ID;
 
-		if (amdgpu_ras_mark_ras_event(adev, type)) {
-			dev_err(adev->dev,
-				"uncorrectable hardware error (ERREVENT_ATHUB_INTERRUPT) detected!\n");
-			return;
-		}
+		if (amdgpu_uniras_enabled(adev))
+			return 0;
 
-		event_id = amdgpu_ras_acquire_event_id(adev, type);
+		if (!amdgpu_ras_mark_ras_event(adev, type))
+			event_id = amdgpu_ras_acquire_event_id(adev, type);
 
 		RAS_EVENT_LOG(adev, event_id, "uncorrectable hardware error"
 			      "(ERREVENT_ATHUB_INTERRUPT) detected!\n");
@@ -4672,6 +4670,8 @@ void amdgpu_ras_global_ras_isr(struct amdgpu_device *adev)
 		ras->gpu_reset_flags |= AMDGPU_RAS_GPU_RESET_MODE1_RESET;
 		amdgpu_ras_reset_gpu(adev);
 	}
+
+	return -EBUSY;
 }
 
 bool amdgpu_ras_need_emergency_restart(struct amdgpu_device *adev)
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
index 9f21b6cf87245..556cf4d7b5ef8 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_ras.h
@@ -910,7 +910,7 @@ static inline void amdgpu_ras_intr_cleared(void)
 	atomic_set(&amdgpu_ras_in_intr, 0);
 }
 
-void amdgpu_ras_global_ras_isr(struct amdgpu_device *adev);
+int amdgpu_ras_global_ras_isr(struct amdgpu_device *adev);
 
 void amdgpu_ras_set_error_query_ready(struct amdgpu_device *adev, bool ready);
 
diff --git a/drivers/gpu/drm/amd/ras/ras_mgr/amdgpu_ras_sys.c b/drivers/gpu/drm/amd/ras/ras_mgr/amdgpu_ras_sys.c
index 40071b8763336..f21cd55a25be9 100644
--- a/drivers/gpu/drm/amd/ras/ras_mgr/amdgpu_ras_sys.c
+++ b/drivers/gpu/drm/amd/ras/ras_mgr/amdgpu_ras_sys.c
@@ -29,8 +29,13 @@
 static int amdgpu_ras_sys_detect_fatal_event(struct ras_core_context *ras_core, void *data)
 {
 	struct amdgpu_device *adev = (struct amdgpu_device *)ras_core->dev;
+	int ret;
 	uint64_t seq_no;
 
+	ret = amdgpu_ras_global_ras_isr(adev);
+	if (ret)
+		return ret;
+
 	seq_no = amdgpu_ras_mgr_gen_ras_event_seqno(adev, RAS_SEQNO_TYPE_UE);
 	RAS_DEV_INFO(adev,
 		"{%llu} Uncorrectable hardware error(ERREVENT_ATHUB_INTERRUPT) detected!\n",