Re: [PATCH v3 2/7] drm/xe/hw_error: Initialize hw_error_work per tile

["Tauro, Riana" <riana.tauro@intel.com>]

On 04-11-2026 15:01, Mallesh, Koujalagi wrote:
>
>
> On 03-11-2026 11:39 am, Riana Tauro wrote:
>> Since hw_error_work is now being reused generically, initialize it per tile
>> so that we don't hit uninitialized work for non-root tile. While at it,
>> disable hw_error_work on remove to prevent UAF during unwind.
>>
>> Signed-off-by: Riana Tauro<riana.tauro@intel.com>
>> ---
>> v2: move disable work to xe_device_remove
>>      add dgfx and sriov check (Sashiko)
>> ---
>>   drivers/gpu/drm/xe/xe_device.c   |  3 +++
>>   drivers/gpu/drm/xe/xe_hw_error.c | 24 ++++++++++++++++++++++--
>>   drivers/gpu/drm/xe/xe_hw_error.h |  1 +
>>   3 files changed, 26 insertions(+), 2 deletions(-)
>>
>> diff --git a/drivers/gpu/drm/xe/xe_device.c b/drivers/gpu/drm/xe/xe_device.c
>> index 07a7ab6ac083..8f498e3e3e58 100644
>> --- a/drivers/gpu/drm/xe/xe_device.c
>> +++ b/drivers/gpu/drm/xe/xe_device.c
>> @@ -45,6 +45,7 @@
>>   #include "xe_guc.h"
>>   #include "xe_guc_pc.h"
>>   #include "xe_hw_engine_group.h"
>> +#include "xe_hw_error.h"
>>   #include "xe_hwmon.h"
>>   #include "xe_i2c.h"
>>   #include "xe_irq.h"
>> @@ -1118,6 +1119,8 @@ int xe_device_probe(struct xe_device *xe)
>>   
>>   void xe_device_remove(struct xe_device *xe)
>>   {
>> +	xe_hw_error_work_disable(xe);
>> +
>>   	xe_display_unregister(xe);
>>   
>>   	drm_dev_unplug(&xe->drm);
>> diff --git a/drivers/gpu/drm/xe/xe_hw_error.c b/drivers/gpu/drm/xe/xe_hw_error.c
>> index a934b649575c..aaa2437d40bf 100644
>> --- a/drivers/gpu/drm/xe/xe_hw_error.c
>> +++ b/drivers/gpu/drm/xe/xe_hw_error.c
>> @@ -537,6 +537,24 @@ static void process_hw_errors(struct xe_device *xe)
>>   	}
>>   }
>>   
>> +/**
>> + * xe_hw_error_work_disable - Disable hw error work
>> + * @xe: xe device instance
>> + *
>> + * This function disables work for hardware error handling for all tiles.
>> + */
>> +void xe_hw_error_work_disable(struct xe_device *xe)
>> +{
>> +	struct xe_tile *tile;
>> +	u8 id;
>> +
>> +	if (!IS_DGFX(xe) || IS_SRIOV_VF(xe))
>> +		return;
>> +
>> +	for_each_tile(tile, xe, id)
>> +		disable_work_sync(&tile->hw_error_work);
>> +}
>> +
>>   /**
>>    * xe_hw_error_init - Initialize hw errors
>>    * @xe: xe device instance
>> @@ -546,12 +564,14 @@ static void process_hw_errors(struct xe_device *xe)
>>    */
>>   void xe_hw_error_init(struct xe_device *xe)
>>   {
>> -	struct xe_tile *tile = xe_device_get_root_tile(xe);
>> +	struct xe_tile *tile;
>> +	u8 id;
>>   
>>   	if (!IS_DGFX(xe) || IS_SRIOV_VF(xe))
>>   		return;
>>   
>> -	INIT_WORK(&tile->hw_error_work, hw_error_work);
>> +	for_each_tile(tile, xe, id)
>> +		INIT_WORK(&tile->hw_error_work, hw_error_work);
>
> What will happen when probe failure occurred after xe_hw_error_ini(), 
> how are tile workers cancel?
>

Makes sense. Will add it in xe_device_probe error path

Thanks
Riana

> Thanks,
>
> -/Mallesh
>
>>   	process_hw_errors(xe);
>>   }
>> diff --git a/drivers/gpu/drm/xe/xe_hw_error.h b/drivers/gpu/drm/xe/xe_hw_error.h
>> index 5e3a11424108..944699331a93 100644
>> --- a/drivers/gpu/drm/xe/xe_hw_error.h
>> +++ b/drivers/gpu/drm/xe/xe_hw_error.h
>> @@ -12,4 +12,5 @@ struct xe_device;
>>   
>>   void xe_hw_error_irq_handler(struct xe_tile *tile, const u32 master_ctl);
>>   void xe_hw_error_init(struct xe_device *xe);
>> +void xe_hw_error_work_disable(struct xe_device *xe);
>>   #endif