[Openchrome-users] CN700 stability

[Christiaan van Dijk]

Here are my latest findings on my CN700 system:

I got an old WD400 drive (40gig/PATA/UDMA100) and reinstalled the 
complete system with Suse 10.2, added DRM and Openchrome (very easy with 
the new Xorg version). I also enabled DMA on all the devices. The system 
has now been running for weeks without a single problem! :-) The system 
can crash if any of the 3D hardware acceleration is used but this is a 
known issue. Suse for example disables hardware acceleration in MESA if 
a Via chipset is detected. AIGLX does use the 3D hardware and can cause 
problems (seen this with OpenGL screensaver). Have to figure out how to 
avoid this.

The system runs a Myth backend for testing, MPEG2 SDTV takes about 50% 
CPU load on a 1.2GHz board (connected to 1280*1024 display). Quite high 
compared to my CLE266 1.2GHz which takes around 20% (connected to TV). 
But still faster as full software. Have to try the CN700 on my TV, 
should be easy since the board has a VT1622A encoder.

Now I'm wondering what the exact cause of the problems with the 
320gig/PATA/UDMA133 drive are. It's a new drive so a faulty drive is a 
possibility. Stress testing the drive with software from Hitachi showed 
no problems however :-(  
These tests are not 100% since they're not using the UDMA modes (I 
think). Had other drives with failures which did not show up in any 
test. Really difficult to convince a supplier on this point. Think I 
just have to buy a new drive...

Christiaan.

Lasse Bigum wrote:
> On 19:03, Thu 18 Jan, Hindrik Hettema wrote:
>   
>> Christiaan van Dijk wrote:
>> Since no DMA was available (for example, I get the error message that 
>> hdparm is not able to set DMA), I now have to presume it is NOT related 
>> to DMA, only using DMA seems to increase the probability of failure.
>>
>> Does anybody still have any ideas, since I am running out of (and think 
>> to return my HUSH as a warranty claim) - the only thing I can think of 
>> is to install Windows to check if the behaviour also happens in that OS. 
>> If so, it is a VIA problem, if not -- it may still be a bug in the kernel.
>>     
>
> Have you tried what the Ubuntu help site has on this?
> https://help.ubuntu.com/community/OpenChrome#head-e440c4247a38516d69b8bc145c9604696e1a39be
>
> Might help?
>
> /Lasse
>   
> ------------------------------------------------------------------------
>
> _______________________________________________
> openchrome-users mailing list
> openchrome-users at openchrome.org
> http://wiki.openchrome.org/mailman/listinfo/openchrome-users
> Main page:
> http://www.openchrome.org
> Wiki:
> http://wiki.openchrome.org
> User Forum:
> http://wiki.openchrome.org/tikiwiki/tiki-view_forum.php?forumId=1