i810 rehash and dell inspiron 2600

[Justin Georgeson]

Jim Cornette wrote:
> Justin Georgeson wrote:
> 
>> As an interesting note, Dell Inspiron 2600 laptops must stay at BIOS 
>> revision A08 or lower or this chipset will not work at all in any 
>> Linux distribution.
>>
>> What are the differences between the xorg i810 driver (is the upstream 
>> fixed one in the FC3 xorg 6.8.0 packages?) and the Intel one? The 
>> latest version from Intel was released in June of this year and the 
>> README tells how to configure dual-head support. Has anyone installed 
>> it (I haven't been able to build it so far on FC3T2 or Suse 9.1)?
>>
> 
> With a Dell computer that has an 828 656 Graphics controller, the driver 
> has changes that make the server solid vs. previous version of X. The 
> pre-CVS versions would totally crash X and leave whatever visual 
> representations were on the display present. X itself would not be in 
> the process list. I seem to recall that this GC identifies itself as an 
> 845 GC. I did a search for the specs for your laptop. It came back with 
> information that it uses the 830 chipset. Therefore, if it acts like the 
> 845, it should work better with the CVS version. If it acts more like 
> the 815, you are out of luck.
> 
> Regarding dual head support. I tried this for the 815 GC and had a bit 
> of trouble. This was with another troubled video card, so the results 
> were pretty useless. (Lockups and other things)
> 
> As a note on BIOS for the Dell, I had to upgrade BIOS from A03 to A04.
> 
> I have not tried the Intel driver version. Is this more capable than the 
> xorg-x11 version? Any different features besides dual-head?
> 

I haven't been able to get it to compile yet, I have an open ticket with 
their support department about it. I did find out that the Intel driver 
relies on the BIOS implementation for supported resolutions. So if the 
BIOS isn't set up to support 1450x resolution (I forget the other number 
in that res, just recall a bunch of people mentioning it) than the Intel 
driver won't either. So in that respect, the OSS driver seems to have 
the upper hand.

---- begin tangential rant which is not on-topic with the list
---- sorry

I made a post on Dell's forums (forums.us.dell.com) about having to 
revert my BIOS version to get the chip working. I relayed my complaint 
that Dell is intentionally impairing their BIOS implementation, as the 
Intel reference design, and indeed the BIOS on their own motherboards 
supports setting the stolen memory for proper linux support. The word 
petition was used, only to find out that Dell's forum considers that 
word to be profane. Another user apparently took that to heart and 
started a thread under a different topic (again on Dell's forums) about 
censoring the word petition. That thread has been summarily deleted due 
to violation of the TOS. That's Dell's right, but it certainly speaks 
volumes about how they respond to valid criticism. I hope others will 
join me in deciding to avoid Dell computers in the future due to their 
ambivalence towards customers.

-- 
; Justin Georgeson
; "Free the mallocs, delete the news"
; http://www.openoffice.org
; http://www.boycott-riaa.com
; http://www.freeiPods.com/default.aspx?referer=9559016