[Intel-gfx] [PATCH v4] dma-buf: Add ioctls to allow userspace to flush

[Thomas Hellstrom]

On 08/26/2015 04:32 PM, Tiago Vignatti wrote:
> On 08/26/2015 09:58 AM, Daniel Vetter wrote:
>> The other is that right now there's no user nor implementation in sight
>> which actually does range-based flush optimizations, so I'm pretty much
>> expecting we'll get it wrong. Maybe instead we should go one step
>> further
>> and remove the range from the internal dma-buf interface and also
>> drop it
>> from the ioctl? With the flags we can always add something later on once
>> we have a real user with a clear need for it. But afaik cros only
>> wants to
>> shuffle around entire tiles and has a buffer-per-tile approach.
>
> Thomas, I think Daniel has a point here and also, I wouldn't mind
> removing all range control from the dma-buf ioctl either.
>
> In this last patch we can see that it's not complicated to add the
> 2d-sync if we eventually decides to want it. But right now there's no
> way we can test it. Therefore in that case I'm all in favour of doing
> this work gradually, adding the basics first.
>

Sure. Unfortunately I wasn't completely clear about the use-case here.
IMO adding a user-space sync functionality should be purely for performance.

/Thomas