Re: [RFC 02/13] drm/i915/svm: Define SVM UAPI

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Quoting Niranjana Vishwanathapura (2019-11-22 22:57:23)
> Define UAPI for Shared Virtual Memory (SVM) fucntionality including
> SVM capability and configurability.
> When there is no GPU page fault support available, add ioctls
> to explicitly bind and migrate virtual address ranges.
> 
> Cc: Joonas Lahtinen <joonas.lahtinen@xxxxxxxxxxxxxxx>
> Cc: Jon Bloomfield <jon.bloomfield@xxxxxxxxx>
> Cc: Daniel Vetter <daniel.vetter@xxxxxxxxx>
> Cc: Sudeep Dutt <sudeep.dutt@xxxxxxxxx>
> Signed-off-by: Niranjana Vishwanathapura <niranjana.vishwanathapura@xxxxxxxxx>
> Signed-off-by: Venkata Sandeep Dhanalakota <venkata.s.dhanalakota@xxxxxxxxx>

Having this as a separate commit ahead of the functionality breaks
bisecting.

The uAPI should be exposed in similar sized chunks as it's implemented.
If it has to go in as single patch, that should be after all the
plumbing is already in place.

That also gives a clear indication to anybody backporting that you
need to backport until the uAPI patch.

Regards, Joonas
_______________________________________________
Intel-gfx mailing list
Intel-gfx@xxxxxxxxxxxxxxxxxxxxx
https://lists.freedesktop.org/mailman/listinfo/intel-gfx




[Index of Archives]     [AMD Graphics]     [Linux USB Devel]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]

  Powered by Linux