Re: [PATCH v9 3/7] drm/xe: Trigger queue cleanup if not in wedged mode 2

["Dong, Zhanjun" <zhanjun.dong@intel.com>]

LGTM

Reviewed-by: Zhanjun Dong <zhanjun.dong@intel.com>


On 2026-03-13 2:49 p.m., Matthew Brost wrote:
> On Wed, Mar 11, 2026 at 12:33:54PM -0400, Dong, Zhanjun wrote:
>>
>>
>> On 2026-03-10 6:50 p.m., Zhanjun Dong wrote:
>>> The intent of wedging a device is to allow queues to continue running
>>> only in wedged mode 2. In other modes, queues should initiate cleanup
>>> and signal all remaining fences. Fix xe_guc_submit_wedge to correctly
>>> clean up queues when wedge mode != 2.
>>>
>>> Fixes: 7dbe8af13c18 ("drm/xe: Wedge the entire device")
>>> Cc: stable@vger.kernel.org
>>> Signed-off-by: Matthew Brost <matthew.brost@intel.com>
>>> ---
>>>    drivers/gpu/drm/xe/xe_guc_submit.c | 35 +++++++++++++++++++-----------
>>>    1 file changed, 22 insertions(+), 13 deletions(-)
>>>
>>> diff --git a/drivers/gpu/drm/xe/xe_guc_submit.c b/drivers/gpu/drm/xe/xe_guc_submit.c
>>> index 8afd424b27fb..cb32053d57ec 100644
>>> --- a/drivers/gpu/drm/xe/xe_guc_submit.c
>>> +++ b/drivers/gpu/drm/xe/xe_guc_submit.c
>>> @@ -1319,6 +1319,7 @@ static void disable_scheduling_deregister(struct xe_guc *guc,
>>>     */
>>>    void xe_guc_submit_wedge(struct xe_guc *guc)
>>>    {
>>> +	struct xe_device *xe = guc_to_xe(guc);
>>>    	struct xe_gt *gt = guc_to_gt(guc);
>>>    	struct xe_exec_queue *q;
>>>    	unsigned long index;
>>> @@ -1333,20 +1334,28 @@ void xe_guc_submit_wedge(struct xe_guc *guc)
>>>    	if (!guc->submission_state.initialized)
>>>    		return;
>>> -	err = devm_add_action_or_reset(guc_to_xe(guc)->drm.dev,
>>> -				       guc_submit_wedged_fini, guc);
>>> -	if (err) {
>>> -		xe_gt_err(gt, "Failed to register clean-up in wedged.mode=%s; "
>>> -			  "Although device is wedged.\n",
>>> -			  xe_wedged_mode_to_string(XE_WEDGED_MODE_UPON_ANY_HANG_NO_RESET));
>>> -		return;
>>> -	}
>>> +	if (xe->wedged.mode == 2) {
>>> +		err = devm_add_action_or_reset(guc_to_xe(guc)->drm.dev,
>>> +					       guc_submit_wedged_fini, guc);
>>> +		if (err) {
>>> +			xe_gt_err(gt, "Failed to register clean-up on wedged.mode=2; "
>>> +				  "Although device is wedged.\n");
>>> +			return;
>>> +		}
>>> -	mutex_lock(&guc->submission_state.lock);
>>> -	xa_for_each(&guc->submission_state.exec_queue_lookup, index, q)
>>> -		if (xe_exec_queue_get_unless_zero(q))
>>> -			set_exec_queue_wedged(q);
>>> -	mutex_unlock(&guc->submission_state.lock);
>>> +		mutex_lock(&guc->submission_state.lock);
>>> +		xa_for_each(&guc->submission_state.exec_queue_lookup, index, q)
>>> +			if (xe_exec_queue_get_unless_zero(q))
>>> +				set_exec_queue_wedged(q);
>>> +		mutex_unlock(&guc->submission_state.lock);
>>> +	} else {
>>> +		/* Forcefully kill any remaining exec queues, signal fences */
>> Q: Shall we do VF bypass here?
>>
> 
> Same answer as last patch - no.
> 
> Matt
> 
>> Regards,
>> Zhanjun Dong
>>> +		guc_submit_reset_prepare(guc);
>>> +		xe_guc_submit_stop(guc);
>>> +		xe_guc_softreset(guc);
>>> +		xe_uc_fw_sanitize(&guc->fw);
>>> +		xe_guc_submit_pause_abort(guc);
>>> +	}
>>>    }
>>>    static bool guc_submit_hint_wedged(struct xe_guc *guc)
>>