From: Philip Yang <Philip.Yang@amd.com>
Date: Mon, 15 Sep 2025 19:57:32 +0000 (-0400)
Subject: drm/amdkfd: Fix mmap write lock not release
X-Git-Url: https://www.infradead.org/git/?a=commitdiff_plain;h=7574f30337e19045f03126b4c51f525b84e5049e;p=users%2Fhch%2Fmisc.git

drm/amdkfd: Fix mmap write lock not release

If mmap write lock is taken while draining retry fault, mmap write lock
is not released because svm_range_restore_pages calls mmap_read_unlock
then returns. This causes deadlock and system hangs later because mmap
read or write lock cannot be taken.

Downgrade mmap write lock to read lock if draining retry fault fix this
bug.

Signed-off-by: Philip Yang <Philip.Yang@amd.com>
Reviewed-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
index 273f42e3afdd..9d72411c3379 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
@@ -3045,6 +3045,8 @@ retry_write_locked:
 	if (svms->checkpoint_ts[gpuidx] != 0) {
 		if (amdgpu_ih_ts_after_or_equal(ts,  svms->checkpoint_ts[gpuidx])) {
 			pr_debug("draining retry fault, drop fault 0x%llx\n", addr);
+			if (write_locked)
+				mmap_write_downgrade(mm);
 			r = -EAGAIN;
 			goto out_unlock_svms;
 		} else {