ATI 6.12.4 on ultraSparc runs slow

[jean-francois simon]

2010/3/24 Michel Dänzer <michel at daenzer.net>

>
> >
> > so i am assuming that most of the work is done by some X server
> > "painting code" to the shadowfb in system ram, which is then DMA'ed to
> > the scan out buffer in the GPU VRAM.
>
> It's only DMA'ed if you implemented that. ;)
>

so it is PIO not DMA? just curious, where is the code that is doing that
transfer?


> > since i see a big difference between solaris/sparc and linux/x86-64,
> > one idea that comes to mind is that there could some overhead due to
> > the difference in endianess between the solaris and the linux box.
>
> I don't think that matters here, we can use GPU facilities to handle
> endianness for direct CPU access to the framebuffer.
>
>
sorry i dont understand what you mean. are you saying i should no longer use
shadowfb?


> Have you tried profiling the X server in both setups to see where the
> cycles are being spent in each case? I'd guess that maybe the x86 CPU is
> faster and/or has higher write throughput to VRAM, e.g. due to write
> combining.
>
>
i'll try that. thx.
-jfs
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.x.org/archives/xorg-driver-ati/attachments/20100324/786320ff/attachment.html>