Running jobs on a hardware context while it is in the process of
releasing resources can lead to use-after-free and crashes.

Fix this by stopping job scheduling before calling
aie2_release_resource() and restarting it after the release completes.
Additionally, aie2_sched_job_run() now checks whether the hardware
context is still active.

Fixes: 4fd6ca90fc7f ("accel/amdxdna: Refactor hardware context destroy routine")
Signed-off-by: Lizhi Hou <lizhi.hou@amd.com>
---
 drivers/accel/amdxdna/aie2_ctx.c | 6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/drivers/accel/amdxdna/aie2_ctx.c b/drivers/accel/amdxdna/aie2_ctx.c
index 5511ab2ef242..c4a58c00e442 100644
--- a/drivers/accel/amdxdna/aie2_ctx.c
+++ b/drivers/accel/amdxdna/aie2_ctx.c
@@ -315,6 +315,9 @@ aie2_sched_job_run(struct drm_sched_job *sched_job)
 	struct dma_fence *fence;
 	int ret;
 
+	if (hwctx->status != HWCTX_STAT_READY)
+		return NULL;
+
 	if (!mmget_not_zero(job->mm))
 		return ERR_PTR(-ESRCH);
 
@@ -705,7 +708,10 @@ void aie2_hwctx_fini(struct amdxdna_hwctx *hwctx)
 	aie2_hwctx_wait_for_idle(hwctx);
 
 	/* Request fw to destroy hwctx and cancel the rest pending requests */
+	drm_sched_stop(&hwctx->priv->sched, NULL);
 	aie2_release_resource(hwctx);
+	hwctx->status = HWCTX_STAT_STOP;
+	drm_sched_start(&hwctx->priv->sched, 0);
 
 	mutex_unlock(&xdna->dev_lock);
 	drm_sched_entity_destroy(&hwctx->priv->entity);
-- 
2.34.1

On 1/29/26 6:32 PM, Lizhi Hou wrote:
> Running jobs on a hardware context while it is in the process of
> releasing resources can lead to use-after-free and crashes.
> 
> Fix this by stopping job scheduling before calling
> aie2_release_resource() and restarting it after the release completes.
> Additionally, aie2_sched_job_run() now checks whether the hardware
> context is still active.
> 
> Fixes: 4fd6ca90fc7f ("accel/amdxdna: Refactor hardware context destroy routine")
> Signed-off-by: Lizhi Hou <lizhi.hou@amd.com>
Reviewed-by: Mario Limonciello (AMD) <superm1@kernel.org>
> ---
>   drivers/accel/amdxdna/aie2_ctx.c | 6 ++++++
>   1 file changed, 6 insertions(+)
> 
> diff --git a/drivers/accel/amdxdna/aie2_ctx.c b/drivers/accel/amdxdna/aie2_ctx.c
> index 5511ab2ef242..c4a58c00e442 100644
> --- a/drivers/accel/amdxdna/aie2_ctx.c
> +++ b/drivers/accel/amdxdna/aie2_ctx.c
> @@ -315,6 +315,9 @@ aie2_sched_job_run(struct drm_sched_job *sched_job)
>   	struct dma_fence *fence;
>   	int ret;
>   
> +	if (hwctx->status != HWCTX_STAT_READY)
> +		return NULL;
> +
>   	if (!mmget_not_zero(job->mm))
>   		return ERR_PTR(-ESRCH);
>   
> @@ -705,7 +708,10 @@ void aie2_hwctx_fini(struct amdxdna_hwctx *hwctx)
>   	aie2_hwctx_wait_for_idle(hwctx);
>   
>   	/* Request fw to destroy hwctx and cancel the rest pending requests */
> +	drm_sched_stop(&hwctx->priv->sched, NULL);
>   	aie2_release_resource(hwctx);
> +	hwctx->status = HWCTX_STAT_STOP;
> +	drm_sched_start(&hwctx->priv->sched, 0);
>   
>   	mutex_unlock(&xdna->dev_lock);
>   	drm_sched_entity_destroy(&hwctx->priv->entity);

Applied to drm-misc-next-fixes

On 1/30/26 11:24, Mario Limonciello wrote:
> On 1/29/26 6:32 PM, Lizhi Hou wrote:
>> Running jobs on a hardware context while it is in the process of
>> releasing resources can lead to use-after-free and crashes.
>>
>> Fix this by stopping job scheduling before calling
>> aie2_release_resource() and restarting it after the release completes.
>> Additionally, aie2_sched_job_run() now checks whether the hardware
>> context is still active.
>>
>> Fixes: 4fd6ca90fc7f ("accel/amdxdna: Refactor hardware context 
>> destroy routine")
>> Signed-off-by: Lizhi Hou <lizhi.hou@amd.com>
> Reviewed-by: Mario Limonciello (AMD) <superm1@kernel.org>
>> ---
>>   drivers/accel/amdxdna/aie2_ctx.c | 6 ++++++
>>   1 file changed, 6 insertions(+)
>>
>> diff --git a/drivers/accel/amdxdna/aie2_ctx.c 
>> b/drivers/accel/amdxdna/aie2_ctx.c
>> index 5511ab2ef242..c4a58c00e442 100644
>> --- a/drivers/accel/amdxdna/aie2_ctx.c
>> +++ b/drivers/accel/amdxdna/aie2_ctx.c
>> @@ -315,6 +315,9 @@ aie2_sched_job_run(struct drm_sched_job *sched_job)
>>       struct dma_fence *fence;
>>       int ret;
>>   +    if (hwctx->status != HWCTX_STAT_READY)
>> +        return NULL;
>> +
>>       if (!mmget_not_zero(job->mm))
>>           return ERR_PTR(-ESRCH);
>>   @@ -705,7 +708,10 @@ void aie2_hwctx_fini(struct amdxdna_hwctx *hwctx)
>>       aie2_hwctx_wait_for_idle(hwctx);
>>         /* Request fw to destroy hwctx and cancel the rest pending 
>> requests */
>> +    drm_sched_stop(&hwctx->priv->sched, NULL);
>>       aie2_release_resource(hwctx);
>> +    hwctx->status = HWCTX_STAT_STOP;
>> +    drm_sched_start(&hwctx->priv->sched, 0);
>>         mutex_unlock(&xdna->dev_lock);
>>       drm_sched_entity_destroy(&hwctx->priv->entity);
>