10:06 p.m.
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)?
--
; 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
7:44 p.m.
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?
Jim
8:54 p.m.
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
10:56 p.m.
> 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.
Thanks for the insight into the Intel provided driver, 1400x1050 would
be top resolution I'd try. This is the maximum choice for the OSS driver
also. I'm curious mainly in case I need to go the similar nv route. I
doubt that it will be needed.
Regarding the stolen memory issue. I have an excerpt from the working
Dell below. I guess it is recognized as an 865G. The full log is at the
link below the excerpt. I posted it for comparison between the different
cards. 815 and 865G. I remembered the stolen memory reference in the file.
(--) I810(0): Chipset: "865G"
(--) I810(0): Linear framebuffer at 0xE8000000
(--) I810(0): IO registers at addr 0xFEB80000
(II) I810(0): 1 display pipe available.
(II) I810(0): detected 8060 kB stolen memory.
(II) I810(0): I830CheckAvailableMemory: 441340 kB available
(II) I810(0): Will attempt to tell the BIOS that there is 12288 kB VideoRAM
(WW) I810(0): Extended BIOS function 0x5f11 not supported.
(II) Loading sub module "int10"
(II) LoadModule: "int10"
(II) Reloading /usr/X11R6/lib/modules/linux/libint10.a
(II) I810(0): initializing int10
(WW) I810(0): Bad V_BIOS checksum
(II) I810(0): Primary V_BIOS segment is: 0xc000
https://bugzilla.redhat.com/bugzilla/attachment.cgi?id=103809&action=...
About the rant. I've noticed a similar situation on sites which are
governed by the corporations sponsoring the product. Simple words like
petition should not be censored, but the main issue is for improvements
to the BIOS for all platforms should be followed. If they don't want to
do the work, they should open up enough information for an Optimized
BIOS for the chipset can be achieved for Linux operation.
Jim
8:03 a.m.
On Mon, Sep 27, 2004 at 11:56:49PM -0400, Jim Cornette wrote:
About the rant. I've noticed a similar situation on sites which
are
governed by the corporations sponsoring the product. Simple words like
Same with shopping sites, removing bad reviews of products and bending
ratings. Some at least seem to get the value of criticism but not many.
to the BIOS for all platforms should be followed. If they don't
want to
do the work, they should open up enough information for an Optimized
BIOS for the chipset can be achieved for Linux operation.
You assume they have the info themselves. That varies by vendor and machine,
and sometimes the reason we can't get info in the Linux community is that
the vendor would have confess to having bought the system in wholesale from
an external no-name builder.
Alan
8:34 p.m.
Alan Cox wrote:
On Mon, Sep 27, 2004 at 11:56:49PM -0400, Jim Cornette wrote:
>About the rant. I've noticed a similar situation on sites which are
>governed by the corporations sponsoring the product. Simple words like
Same with shopping sites, removing bad reviews of products and bending
ratings. Some at least seem to get the value of criticism but not many.
>to the BIOS for all platforms should be followed. If they don't want to
>do the work, they should open up enough information for an Optimized
>BIOS for the chipset can be achieved for Linux operation.
You assume they have the info themselves. That varies by vendor and machine,
and sometimes the reason we can't get info in the Linux community is that
the vendor would have confess to having bought the system in wholesale from
an external no-name builder.
Alan
Good point, outsourcing seems to be pretty strong of a reality. I guess
this idea is impractical then. I'll hold off on IBM's outsourcing blunder.
Jim
7157
days inactive
7160
days old
5 comments
3 participants
participants (3)
-
Alan Cox -
Jim Cornette -
Justin Georgeson