[Openchrome-devel] release 0.2.902 this week-end ?

[Xavier Bachelot]

Benno Schulenberg wrote:
> Xavier Bachelot wrote:
>> Benno Schulenberg wrote:
>>> http://www.openchrome.org/trac/ticket/145 ,
>> Another commenter said it was ok for him with 2.6.20, but not ok
>> anymore with 2.6.23.
> 
> An anonymous commenter -- we can't ask him to prove it or give more 
> information.  He might not know what he's doing, or he might even 
> be pissing us off on purpose.
> 
Following ticket seems to support the theory that kernel is the variable
http://www.openchrome.org/trac/ticket/164

So :
2.6.18 : ok
2.6.19 : ? (I'd say ok, but no specific bug report)
2.6.20 : ok
2.6.21 : ?
2.6.22 : bug
2.6.23 : bug
2.6.24 : bug

I have asked the reporter to test 2.6.21.

>>> The disabling does not do any harm, and it does appear to
>>> prevent screen corruption and lockups on the affected chipsets.
>> There are some performances penalties.
> 
> Performance comes a distant third after: 1) it _must boot, and 2) it 
> must not corrupt a normal desktop.
> 
>> imho, this is just sweeping bugs under the carpet.
> 
> Yes, it's swept under the carpet.  But why make users fall over the 
> bug and get incredibly frustrated?  They tend to not file carefully 
> thought-out bug reports but instead shout angrily about why people 
> leave such a damn block of stone in the middle of the room.
> 
>> And there are not even our bugs...
> 
> Unsure.  It may be changes in X, in drm, or in the kernel that 
> trigger an openchrome driver bug.
> 
>> If no one see the bug, there's no chance for it to be fixed.
> 
> We know about the bug.  More people experiencing it and suffering 
> from it won't increase the chances of it getting fixed.  Only 
> people that read the manual and the mailing lists, and then try to 
> enable AGP DMA, and then experience the problem, are likely to come 
> up with useful information.
> 
> Regards,
> 
> Benno