Re: [Mesa-dev] [PATCHv3] r600g: Use a fake reloc to sleep for fences

2012-02-10 Thread Simon Farnsworth
On Thursday 9 February 2012 20:31:18 Marek Olšák wrote: > 2012/2/9 Simon Farnsworth : > > On Wednesday 8 February 2012 18:28:05 Michel Dänzer wrote: > >> On Fre, 2012-02-03 at 17:32 +, Simon Farnsworth wrote: > >> > --- a/src/gallium/drivers/r600/r600_hw_context.c > >> > +++ b/src/gallium/drive

Re: [Mesa-dev] [PATCHv3] r600g: Use a fake reloc to sleep for fences

2012-02-09 Thread Marek Olšák
2012/2/9 Simon Farnsworth : > On Wednesday 8 February 2012 18:28:05 Michel Dänzer wrote: >> On Fre, 2012-02-03 at 17:32 +, Simon Farnsworth wrote: >> > r300g is able to sleep until a fence completes rather than busywait >> > because it creates a special buffer object and relocation that stays >

Re: [Mesa-dev] [PATCHv3] r600g: Use a fake reloc to sleep for fences

2012-02-09 Thread Simon Farnsworth
On Wednesday 8 February 2012 18:28:05 Michel Dänzer wrote: > On Fre, 2012-02-03 at 17:32 +, Simon Farnsworth wrote: > > r300g is able to sleep until a fence completes rather than busywait > > because it creates a special buffer object and relocation that stays > > busy until the CS containing t

Re: [Mesa-dev] [PATCHv3] r600g: Use a fake reloc to sleep for fences

2012-02-08 Thread Michel Dänzer
On Fre, 2012-02-03 at 17:32 +, Simon Farnsworth wrote: > r300g is able to sleep until a fence completes rather than busywait because > it creates a special buffer object and relocation that stays busy until the > CS containing the fence is finished. > > Copy the idea into r600g, and use it to

[Mesa-dev] [PATCHv3] r600g: Use a fake reloc to sleep for fences

2012-02-03 Thread Simon Farnsworth
r300g is able to sleep until a fence completes rather than busywait because it creates a special buffer object and relocation that stays busy until the CS containing the fence is finished. Copy the idea into r600g, and use it to sleep if the user asked for an infinite wait, falling back to busywai