[Openchrome-users] Slow performance with Flash player#

[Brian Candler]

On Tue, Jun 26, 2007 at 11:51:35AM +0100, John Robinson wrote:
> >
> >(II) VIA(0): Benchmarking video copy. Less is better.
> >(--) VIA(0): Timed   libc YUV420 copy... 15725702. Throughput: 37.7 MiB/s.
> >(--) VIA(0): Timed kernel YUV420 copy... 15610756. Throughput: 38.0 MiB/s.
> >(--) VIA(0): Timed    SSE YUV420 copy... 8894087. Throughput: 66.7 MiB/s.
> >(--) VIA(0): Timed    MMX YUV420 copy... 9003383. Throughput: 65.9 MiB/s.
> >(--) VIA(0): Ditch 3DNow! YUV420 copy... Not supported by CPU.
> >(--) VIA(0): Timed   MMX2 YUV420 copy... 8775754. Throughput: 67.6 MiB/s.
> 
> Hmm, that really doesn't look good. On my EPIA M-II 10000, which is 
> pretty similar to your EPIA M 10000, I get:
> (II) VIA(0): Benchmarking video copy. Less is better.
> (--) VIA(0): Timed   libc YUV420 copy... 7644387. Throughput: 77.6 MiB/s.
> (--) VIA(0): Timed kernel YUV420 copy... 7638392. Throughput: 77.7 MiB/s.
> (--) VIA(0): Timed    SSE YUV420 copy... 2826591. Throughput: 209.8 MiB/s.
> (--) VIA(0): Timed    MMX YUV420 copy... 4419723. Throughput: 134.2 MiB/s.
> (--) VIA(0): Ditch 3DNow! YUV420 copy... Not supported by CPU.
> (--) VIA(0): Timed   MMX2 YUV420 copy... 3888107. Throughput: 152.6 MiB/s.

I have a single 512MB DIMM, but I think the M-II is the same there. I guess
I could try tweaking its timing in the BIOS.

I had a browse through http://www.gentoo.org/doc/en/dri-howto.xml, it looks
like I have all the necessary bits already loaded:

$ lsmod | egrep 'drm|agp'
drm                    73236  2 via
via_agp                 9856  1
agpgart                34888  2 drm,via_agp

Looking at the capabilities of my board, it says FW- (no fast writes):

# lspci -vv
...
0000:01:00.0 VGA compatible controller: VIA Technologies, Inc. VT8623 [Apollo CLE266] integrated CastleRock graphics (rev 03) (prog-if 00 [VGA])
        Subsystem: VIA Technologies, Inc. VT8623 [Apollo CLE266] integrated CastleRock graphics
        Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B-
        Status: Cap+ 66MHz+ UDF- FastB2B- ParErr- DEVSEL=medium >TAbort- <TAbort- <MAbort- >SERR- <PERR-
        Latency: 32 (500ns min)
        Interrupt: pin A routed to IRQ 11
        Region 0: Memory at d8000000 (32-bit, prefetchable) [size=64M]
        Region 1: Memory at dc000000 (32-bit, non-prefetchable) [size=16M]
        Expansion ROM at dd000000 [disabled] [size=64K]
        Capabilities: [60] Power Management version 2
                Flags: PMEClk- DSI+ D1+ D2+ AuxCurrent=0mA PME(D0-,D1-,D2-,D3hot-,D3cold-)
                Status: D0 PME-Enable- DSel=0 DScale=0 PME-
        Capabilities: [70] AGP version 2.0
                Status: RQ=32 Iso- ArqSz=0 Cal=0 SBA+ ITACoh- GART64- HTrans- 64bit- FW- AGP3- Rate=x1,x2,x4
                Command: RQ=32 ArqSz=0 Cal=0 SBA+ AGP+ GART64- 64bit- FW- Rate=x4

Not sure where else I should look.

> And I'm running some very old stuff:
> 
> Jun  8 03:07:59 out kernel: [drm] Initialized drm 1.0.1 20051102
> Jun  8 03:07:59 out kernel: [drm] Initialized via 2.9.1 20060111 on 
> minor 0: VIA Technologies, Inc. VT8623 [Apollo CLE266] integrated 
> CastleRock graphics

But not quite as old as mine :-(

$ dmesg | grep drm
[17179632.232000] [drm] Initialized drm 1.0.1 20051102
[17179632.268000] [drm] Initialized via 2.7.4 20051116 on minor 0

I guess that's a consequence of Ubuntu being Debian-derived.

Regards,

Brian.