[Intel-gfx] [PATCH v8 2/2] drm/i915: Allow user to set cache at BO creation

Yang, Fei fei.yang at intel.com
Mon May 15 20:28:47 UTC 2023


> Hi Fei,
>
> On Fri, May 12, 2023 at 04:28:25PM -0700, fei.yang at intel.com wrote:
>> From: Fei Yang <fei.yang at intel.com>
>>
>> To comply with the design that buffer objects shall have immutable
>> cache setting through out their life cycle, {set, get}_caching ioctl's
>> are no longer supported from MTL onward. With that change caching
>> policy can only be set at object creation time. The current code
>> applies a default (platform dependent) cache setting for all objects.
>> However this is not optimal for performance tuning. The patch extends
>> the existing gem_create uAPI to let user set PAT index for the object
>> at creation time.
>> The new extension is platform independent, so UMD's can switch to using
>> this extension for older platforms as well, while {set, get}_caching are
>> still supported on these legacy paltforms for compatibility reason.
>>
>> IGT posted at https://patchwork.freedesktop.org/series/117695/
>
> Test gem_create at create-ext-set-pat
>
>> Tested with https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/22878
>>
>> Tested-by: Jordan Justen <jordan.l.justen at intel.com>
>
> we need here an explicit ack to have the paper work in place. So
> that I still have to ask Jordan and Mesa folks to give an ack if
> things look right.

Will update once an a-b is in place.

> Thanks!
> Andi
>
>> Cc: Chris Wilson <chris.p.wilson at linux.intel.com>
>> Cc: Matt Roper <matthew.d.roper at intel.com>
>> Cc: Andi Shyti <andi.shyti at linux.intel.com>
>> Signed-off-by: Fei Yang <fei.yang at intel.com>
>> Reviewed-by: Andi Shyti <andi.shyti at linux.intel.com>
>
>PS:
>
> nitnitnitpick: the tags need to come in chronological order. So
> that:
>
>  - first you wrote it (Sob: Fei...)
>  - then you sent it (Cc: ...)
>  - then it has been reviewd (R-b)
>  - finally tested (T-b)
>
> I see that many people put the "Cc:" before the "Sob:" and I
> consider it a matter of taste (which might mean "I first prepare
> the mail (Cc:) and then I send it (Sob:)").
>
> But... don't mind too much at these things.
>
>Andi

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.freedesktop.org/archives/intel-gfx/attachments/20230515/98e52894/attachment-0001.htm>


More information about the Intel-gfx mailing list