From: Emily Deng <Emily.Deng@amd.com>
Date: Thu, 6 Mar 2025 00:40:01 +0000 (+0800)
Subject: drm/amdgpu: Fix the race condition for draining retry fault
X-Git-Url: http://review.tizen.org/git/?a=commitdiff_plain;h=e64be12f8401819662e608efa247638b61d023cd;p=platform%2Fkernel%2Flinux-rpi.git

drm/amdgpu: Fix the race condition for draining retry fault

[ Upstream commit f844732e3ad9c4b78df7436232949b8d2096d1a6 ]

Issue:
In the scenario where svm_range_restore_pages is called, but
svm->checkpoint_ts has not been set and the retry fault has not been
drained, svm_range_unmap_from_cpu is triggered and calls svm_range_free.
Meanwhile, svm_range_restore_pages continues execution and reaches
svm_range_from_addr. This results in a "failed to find prange..." error,
 causing the page recovery to fail.

How to fix:
Move the timestamp check code under the protection of svm->lock.

v2:
Make sure all right locks are released before go out.

v3:
Directly goto out_unlock_svms, and return -EAGAIN.

v4:
Refine code.

Signed-off-by: Emily Deng <Emily.Deng@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
index 8c61dee5ca0d..b50283864dcd 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
@@ -2992,19 +2992,6 @@ svm_range_restore_pages(struct amdgpu_device *adev, unsigned int pasid,
 		goto out;
 	}
 
-	/* check if this page fault time stamp is before svms->checkpoint_ts */
-	if (svms->checkpoint_ts[gpuidx] != 0) {
-		if (amdgpu_ih_ts_after(ts,  svms->checkpoint_ts[gpuidx])) {
-			pr_debug("draining retry fault, drop fault 0x%llx\n", addr);
-			r = 0;
-			goto out;
-		} else
-			/* ts is after svms->checkpoint_ts now, reset svms->checkpoint_ts
-			 * to zero to avoid following ts wrap around give wrong comparing
-			 */
-			svms->checkpoint_ts[gpuidx] = 0;
-	}
-
 	if (!p->xnack_enabled) {
 		pr_debug("XNACK not enabled for pasid 0x%x\n", pasid);
 		r = -EFAULT;
@@ -3024,6 +3011,21 @@ svm_range_restore_pages(struct amdgpu_device *adev, unsigned int pasid,
 	mmap_read_lock(mm);
 retry_write_locked:
 	mutex_lock(&svms->lock);
+
+	/* check if this page fault time stamp is before svms->checkpoint_ts */
+	if (svms->checkpoint_ts[gpuidx] != 0) {
+		if (amdgpu_ih_ts_after(ts,  svms->checkpoint_ts[gpuidx])) {
+			pr_debug("draining retry fault, drop fault 0x%llx\n", addr);
+			r = -EAGAIN;
+			goto out_unlock_svms;
+		} else {
+			/* ts is after svms->checkpoint_ts now, reset svms->checkpoint_ts
+			 * to zero to avoid following ts wrap around give wrong comparing
+			 */
+			svms->checkpoint_ts[gpuidx] = 0;
+		}
+	}
+
 	prange = svm_range_from_addr(svms, addr, NULL);
 	if (!prange) {
 		pr_debug("failed to find prange svms 0x%p address [0x%llx]\n",
@@ -3148,7 +3150,8 @@ out_unlock_svms:
 	mutex_unlock(&svms->lock);
 	mmap_read_unlock(mm);
 
-	svm_range_count_fault(node, p, gpuidx);
+	if (r != -EAGAIN)
+		svm_range_count_fault(node, p, gpuidx);
 
 	mmput(mm);
 out: