On Fri, Jan 10, 2014 at 10:46 AM, Thomas Hellstrom
wrote:
> On 01/10/2014 04:23 PM, Rob Clark wrote:
>> On Tue, Jan 7, 2014 at 1:37 PM, Thomas Hellstrom
>> wrote:
>>> Conclusion: Avoid using dma-buf mmap() outside of drivers that know
>>> exactly what they are
>>> doing, and avoid it at all cost
On 01/10/2014 04:23 PM, Rob Clark wrote:
> On Tue, Jan 7, 2014 at 1:37 PM, Thomas Hellstrom
> wrote:
>> Conclusion: Avoid using dma-buf mmap() outside of drivers that know
>> exactly what they are
>> doing, and avoid it at all cost.
>
> well, to be fair, if you are using a gpu on the client side
On Tue, Jan 7, 2014 at 1:37 PM, Thomas Hellstrom wrote:
> Conclusion: Avoid using dma-buf mmap() outside of drivers that know
> exactly what they are
> doing, and avoid it at all cost.
well, to be fair, if you are using a gpu on the client side and pixman
backend on the server side, you probably
On Tue, Jan 07, 2014 at 07:37:11PM +0100, Thomas Hellstrom wrote:
>
> I think dma-buf mmap() used without care like this should be avoided at
> all cost,
> because it will make people happily start using it without thinking
> about non-coherent
> architectures where mmap would be painfully ineffic
I think dma-buf mmap() used without care like this should be avoided at
all cost,
because it will make people happily start using it without thinking
about non-coherent
architectures where mmap would be painfully inefficient: This is because
when the accelerator
uses the dma-buf it has no idea wha
From: Benjamin Gaignard
Make drm compositor aware of the wl_dmabuf protocol if pixman is used.
Add functions to have a wl_dmabuf server inside drm compositor.
Change pixman to let it know how use a wl_dmabuf buffer.
Signed-off-by: Benjamin Gaignard
---
src/compositor-drm.c | 83
