[PATCH v4 10/20] drm/xe/vm: Add an identifier for madvise in xe_vma_ops

Ghimiray, Himal Prasad himal.prasad.ghimiray at intel.com
Mon Jun 23 06:28:19 UTC 2025



On 23-06-2025 11:08, Matthew Brost wrote:
> On Fri, Jun 13, 2025 at 06:25:48PM +0530, Himal Prasad Ghimiray wrote:
>> The flag XE_VMA_OPS_FLAG_MADVISE is introduced to  determine whether
>> xe_vma_ops is initialized by madvise operation on not.
>>
> 
> I'd just squash this into patch #7.
> 
> Then move patch #11 directly after patch #7 to tell a coherent story.

Makes sense, will modify in next version.
Thanks for review.

> 
> Matt
> 
>> Suggested-by: Matthew Brost <matthew.brost at intel.com>
>> Signed-off-by: Himal Prasad Ghimiray <himal.prasad.ghimiray at intel.com>
>> ---
>>   drivers/gpu/drm/xe/xe_vm.c       | 1 +
>>   drivers/gpu/drm/xe/xe_vm_types.h | 1 +
>>   2 files changed, 2 insertions(+)
>>
>> diff --git a/drivers/gpu/drm/xe/xe_vm.c b/drivers/gpu/drm/xe/xe_vm.c
>> index f5f8de113eae..e059d9810d26 100644
>> --- a/drivers/gpu/drm/xe/xe_vm.c
>> +++ b/drivers/gpu/drm/xe/xe_vm.c
>> @@ -4237,6 +4237,7 @@ int xe_vm_alloc_madvise_vma(struct xe_vm *vm, uint64_t start, uint64_t range)
>>   	}
>>   
>>   	xe_vma_ops_init(&vops, vm, NULL, NULL, 0);
>> +	vops.flags |= XE_VMA_OPS_FLAG_MADVISE;
>>   	err = vm_bind_ioctl_ops_parse(vm, ops, &vops);
>>   	if (err)
>>   		goto unwind_ops;
>> diff --git a/drivers/gpu/drm/xe/xe_vm_types.h b/drivers/gpu/drm/xe/xe_vm_types.h
>> index cde2583da303..c7156f8e2ed2 100644
>> --- a/drivers/gpu/drm/xe/xe_vm_types.h
>> +++ b/drivers/gpu/drm/xe/xe_vm_types.h
>> @@ -489,6 +489,7 @@ struct xe_vma_ops {
>>   	struct xe_vm_pgtable_update_ops pt_update_ops[XE_MAX_TILES_PER_DEVICE];
>>   	/** @flag: signify the properties within xe_vma_ops*/
>>   #define XE_VMA_OPS_FLAG_HAS_SVM_PREFETCH BIT(0)
>> +#define XE_VMA_OPS_FLAG_MADVISE          BIT(1)
>>   	u32 flags;
>>   #ifdef TEST_VM_OPS_ERROR
>>   	/** @inject_error: inject error to test error handling */
>> -- 
>> 2.34.1
>>



More information about the Intel-xe mailing list