[AMD Official Use Only - General]
Reviewed-by: Jesse Zhang <Jesse.Zhang@xxxxxxx>
-----Original Message-----
From: Xiaogang.Chen <xiaogang.chen@xxxxxxx>
Sent: Wednesday, September 27, 2023 11:00 AM
To: amd-gfx@xxxxxxxxxxxxxxxxxxxxx
Cc: Yang, Philip <Philip.Yang@xxxxxxx>; Kuehling, Felix <Felix.Kuehling@xxxxxxx>; Zhang, Jesse(Jie) <Jesse.Zhang@xxxxxxx>; Sierra Guiza, Alejandro (Alex) <Alex.Sierra@xxxxxxx>; Chen, Xiaogang <Xiaogang.Chen@xxxxxxx>; Chen, Xiaogang <Xiaogang.Chen@xxxxxxx>
Subject: [PATCH] drm/amdkfd: Fix a race condition of vram buffer unref in svm code
From: Xiaogang Chen <xiaogang.chen@xxxxxxx>
prange->svm_bo unref can happen in both mmu callback and a callback
prange->after
migrate to system ram. Both are async call in different tasks. Sync svm_bo unref operation to avoid random "use-after-free".
Signed-off-by: Xiaogang.Chen <Xiaogang.Chen@xxxxxxx>
---
drivers/gpu/drm/amd/amdkfd/kfd_svm.c | 9 +++++++++
1 file changed, 9 insertions(+)
diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
index 70aa882636ab..8e246e848018 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_svm.c
@@ -637,6 +637,15 @@ void svm_range_vram_node_free(struct svm_range *prange) {
svm_range_bo_unref(prange->svm_bo);
prange->ttm_res = NULL;
+ /* serialize prange->svm_bo unref */
+ mutex_lock(&prange->lock);
+ /* prange->svm_bo has not been unref */
+ if (prange->ttm_res) {
+ prange->ttm_res = NULL;
+ mutex_unlock(&prange->lock);
+ svm_range_bo_unref(prange->svm_bo);
+ } else
+ mutex_unlock(&prange->lock);
}
struct kfd_node *
--
2.25.1
