Re: [PATCH] drm/xe/sa: Drop hardcoded 4K guard in sub-allocator

[Matthew Auld <matthew.auld@intel.com>]

On 17/12/2024 22:39, Matthew Brost wrote:
> On Tue, Dec 17, 2024 at 11:22:46PM +0100, Michal Wajdeczko wrote:
>> Any required prefetch guards are added during batch buffer
>> allocations anyway.
>>
> 
> This should work but I think we actually want to do the opposite of
> this - drop the prefetch pad in BB allocation. This would enable a more
> optimial usage of each suballocation. I think that would work unless we
> have an odd caching issue - if caching is a problem then maybe the BB is
> a cacheline.

Also would be good to update bb_prefetch(), since current prefetch value 
is too small for xe2+ on some engines, so the hardcoded 4K here was 
maybe saving the day.

> 
> I haven't had time to try to out yet but I think we explore the above
> option first. If I'm missing something and the above does not work, then
> agree with this patch.
> 
> Matt
> 
>> Suggested-by: Matthew Brost <matthew.brost@intel.com>
>> Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
>> Cc: Matthew Brost <matthew.brost@intel.com>
>> ---
>>   drivers/gpu/drm/xe/xe_sa.c | 5 ++---
>>   1 file changed, 2 insertions(+), 3 deletions(-)
>>
>> diff --git a/drivers/gpu/drm/xe/xe_sa.c b/drivers/gpu/drm/xe/xe_sa.c
>> index e055bed7ae55..2f69277b1a50 100644
>> --- a/drivers/gpu/drm/xe/xe_sa.c
>> +++ b/drivers/gpu/drm/xe/xe_sa.c
>> @@ -34,7 +34,6 @@ static void xe_sa_bo_manager_fini(struct drm_device *drm, void *arg)
>>   struct xe_sa_manager *xe_sa_bo_manager_init(struct xe_tile *tile, u32 size, u32 align)
>>   {
>>   	struct xe_device *xe = tile_to_xe(tile);
>> -	u32 managed_size = size - SZ_4K;
>>   	struct xe_bo *bo;
>>   	int ret;
>>   
>> @@ -58,11 +57,11 @@ struct xe_sa_manager *xe_sa_bo_manager_init(struct xe_tile *tile, u32 size, u32
>>   	sa_manager->bo = bo;
>>   	sa_manager->is_iomem = bo->vmap.is_iomem;
>>   
>> -	drm_suballoc_manager_init(&sa_manager->base, managed_size, align);
>> +	drm_suballoc_manager_init(&sa_manager->base, size, align);
>>   	sa_manager->gpu_addr = xe_bo_ggtt_addr(bo);
>>   
>>   	if (bo->vmap.is_iomem) {
>> -		sa_manager->cpu_ptr = kvzalloc(managed_size, GFP_KERNEL);
>> +		sa_manager->cpu_ptr = kvzalloc(size, GFP_KERNEL);
>>   		if (!sa_manager->cpu_ptr) {
>>   			sa_manager->bo = NULL;
>>   			return ERR_PTR(-ENOMEM);
>> -- 
>> 2.47.1
>>