<div dir="ltr">I ran some benchmarks and things seem to be running about the same.<div>No one on our graphics team seemed concerned about the change.<br></div><div><br></div><div>The only concern I heard was about the increased complexity of the new sync code as opposed to the old sync framework which tried to keep things straightforward.</div>
</div><div class="gmail_extra"><br><br><div class="gmail_quote">On Fri, Nov 8, 2013 at 3:43 AM, Maarten Lankhorst <span dir="ltr"><<a href="mailto:maarten.lankhorst@canonical.com" target="_blank">maarten.lankhorst@canonical.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div class="im">op 07-11-13 22:11, Rom Lemarchand schreef:<br>
</div><div class="im">> Hi Maarten, I tested your changes and needed the attached patch: behavior<br>
> now seems equivalent as android sync. I haven't tested performance.<br>
><br>
> The issue resolved by this patch happens when i_b < b->num_fences and i_a<br>
>> = a->num_fences (or vice versa). Then, pt_a is invalid and so<br>
> dereferencing pt_a->context causes a crash.<br>
><br>
</div>Yeah, I pushed my original fix. I intended to keep android userspace behavior the same, and I tried to keep the kernelspace the api same as much as I could. If peformance is the same, or not noticeably worse, would there be any objections on the android side about renaming dma-fence to syncpoint, and getting it in mainline?<br>
<span class="HOEnZb"><font color="#888888"><br>
~Maarten<br>
</font></span></blockquote></div><br></div>