Re: [Intel-gfx] [PATCH libdrm 15/15] intel: Support passing of explicit fencing from execbuf

2016-10-01 Thread Chris Wilson
On Fri, Sep 30, 2016 at 01:53:06PM -0700, Rafael Antognolli wrote: > Hi Chris, > > On Thu, Aug 25, 2016 at 10:08:33AM +0100, Chris Wilson wrote: > > Allow the caller to pass in an fd to an array of fences to control > > serialisation of the execbuf in the kernel and on the GPU, and in return > >

Re: [Intel-gfx] [PATCH libdrm 15/15] intel: Support passing of explicit fencing from execbuf

2016-09-30 Thread Rafael Antognolli
Hi Chris, On Thu, Aug 25, 2016 at 10:08:33AM +0100, Chris Wilson wrote: > Allow the caller to pass in an fd to an array of fences to control > serialisation of the execbuf in the kernel and on the GPU, and in return > allow creation of a fence fd for signaling the completion (and flushing) > of

[Intel-gfx] [PATCH libdrm 15/15] intel: Support passing of explicit fencing from execbuf

2016-08-25 Thread Chris Wilson
Allow the caller to pass in an fd to an array of fences to control serialisation of the execbuf in the kernel and on the GPU, and in return allow creation of a fence fd for signaling the completion (and flushing) of the batch. When the returned fence is signaled, all writes to the buffers inside