On Friday, April 13, 2018 2:37:05 PM PDT James Xiong wrote: > On Fri, 13 Apr 2018 14:33:09 -0700 > Kenneth Graunke <kenn...@whitecape.org> wrote: > > > On Friday, April 13, 2018 2:08:40 PM PDT James Xiong wrote: > > > On Fri, 13 Apr 2018 13:51:02 -0700 > > > Kenneth Graunke <kenn...@whitecape.org> wrote: > > > > > > > On Friday, April 13, 2018 1:35:45 PM PDT Kenneth Graunke wrote: > > > > > I've sent a simpler patch which I think should hopefully fix > > > > > your bug: https://patchwork.freedesktop.org/patch/217107/ > > > > > > > > Lionel noticed that I botched that patch. Here's an actual one: > > > > > > > > https://patchwork.freedesktop.org/patch/217108/ > > > Yes it will fix the existing bug. However the assumption here is > > > that the init allocation size will NOT be rounded up as it happens > > > to be the bucket size. > > > I am working on an optimization to improve memory usage(that's how > > > I find out this bug), this assumption is no longer true. > > > Essentially the bufmgr could return a buffer with the same or > > > larger size whether it is same as the bucket's or not. Anyway I > > > guess I can send the fix later along with the optimization > > > patches. > > > > Ah, that's a good point. Your patch also tries to use the BO size > > for the initial malloc as well, which is a good idea... > So what do you want? you want me to change this patch and sent for > review or take yours for now.
Hi James, I pushed my version of the fix since and Cc'd stable since it seemed like the simplest change that would fix the issue. I sent out a second patch to fix the initial size issue: https://patchwork.freedesktop.org/patch/217768/ --Ken
signature.asc
Description: This is a digitally signed message part.
_______________________________________________ mesa-dev mailing list mesa-dev@lists.freedesktop.org https://lists.freedesktop.org/mailman/listinfo/mesa-dev
