[PATCH 08/13] drm/xe/sa: Allow creating suballocator with custom guard size
Matthew Brost
matthew.brost at intel.com
Thu Dec 12 22:48:03 UTC 2024
On Thu, Dec 12, 2024 at 10:57:31PM +0100, Michal Wajdeczko wrote:
>
>
> On 12.12.2024 04:23, Matthew Brost wrote:
> > On Thu, Dec 12, 2024 at 02:01:36AM +0100, Michal Wajdeczko wrote:
>
> ...
>
> >>
> >> +static inline struct xe_sa_manager *xe_sa_bo_manager_init(struct xe_tile *tile, u32 size, u32 align)
> >> +{
> >> + return __xe_sa_bo_manager_init(tile, size, SZ_4K, align);
> >
> > So why is this guard needed? I'm reasoning this was for CS prefetches
> > but all current uses of the SA (xe_bb.c) do suballocations with CS
> > prefetch padding. We likely don't need to do this in both places.
> >
> > I'd say drop prefetch padding from xe_bb.c and move the prefetch
> > calculation size to here.
> >
> > Can be done in a follow up but before RBing let's make sure we
> > understand exactly why this is needed and develop a follow up plan as
> > the code as is doesn't look right / necessary.
> >
>
> since none of SA functions were documented, same with BB code, I just
> wanted to preserve existing usage without going too deep into a "why"
>
> to get right answers we need original author(s), but since code was part
> of the initial submission, I can't tell who is it
>
I think Maarten was the original author here - will follow up with him.
Agree this out of scope for this series and not a blocker but we
definitly need to follow up on this - having things the driver which are
not understood is not recipe for longterm maintainability.
Anyways since this is not changing any existing behavior:
Reviewed-by: Matthew Brost <matthew.brost at intel.com>
More information about the Intel-xe
mailing list