[ltp] Looking for X21 and R50 (some) owners - Kernel bug 3022 - Subsystem vendor/device IDs
Giorgio Lando
linux-thinkpad@linux-thinkpad.org
Wed, 3 May 2006 17:32:11 +0200
In the message sent about ten days ago to linux-thinkpad mailing list I forward below, we were looking for
owners of certain thinkpad models, asking them to send the output of a couple
of commands. Thanks for the many answers sent. However, we still miss the
data concerning two models:
Thinkpad X21
Thinkpad R50 (only if equipped with an Ati Mobility Radeon 7500)
It is enough if you tell us which one you own and the output of:
lspci -d "1002:*" -vn | grep Subsystem
Please send this information to patroclo7@gmail.org.
I send this email also to the general thinkpad mailing list. In fact, the data we need (the SUBSYSTEM vendor/device IDs of the graphic chipset - NOT the mere PCI vendor device IDs of the grphic chipset) can be grasped also if you happen to own one of these two laptops and to run windows on it. In this case, just drop me a line and I will try to hep you to get the data we need.
I thank you warmly in advance
Giorgio Lando
----- Forwarded message from Giorgio Lando <patroclo7@gmail.com> -----
Date: Thu, 20 Apr 2006 15:10:51 +0200
From: Giorgio Lando <patroclo7@gmail.com>
To: linux-thinkpad@linux-thinkpad.org
Subject: Help Needed - Kernel bug 3022 - Subsystem vendor/device IDs
User-Agent: mutt-ng/devel-r802 (Linux)
Hi. We are working on a well known kernel bug, concerning a powerdrain issue
on some thinkpad models equipped with radeon graphic chipsets. The problem
is labelled as kernel bug 3022 in the kernel bug tracker and is also documented in the thinkwiki website:
www.thinkwiki.org/wiki/Problem_with_high_power_drain_in_ACPI_sleep
The problem seems to be solved on many models through a patch for the radeon
framebuffer driver, which
activates suspension-to-D2 for the video card. However, suspension-to-D2 for
radeon cards on x86 machines is in general not well documented, so we need to
exactly identify the affected thinkpad models and to whitelist them into the
patch itself. In the actual version of the patch, the whitelist includes the
dmi version strings, but this system generates a whitelist too long and
difficult to maintain/expand. Thus, in order to have this patch included in
the mainline kernel, we have been asked to try to whitelist the subsystem
vendor/device ids: in this way, the exact association between the mobo and
the graphic chipset is identified. Please note that the subsystem
vendor/device IDs we need are not the pci vendor/device IDs of the graphic
chipset.
Actually we need to translate the old whitelist in the new one, and we have
been up to now unable to get the subsystem vendor/device IDs for certain
mobo<->graphic chipsets association. Here you can help us: if you have one of
the below listed thinkpad, you should send me (send an email to
patroclo7@gmail.com, do not reply to the mailing list, if you do not want to
tell something of general interest) the output of the following
command:
lspci -d "1002:*" -vn | grep Subsystem
This is the list of the affected thinkpad models (also with the graphic card
specified, when the model is sold with different graphic cards) for
which we do not know the subsystem vendor/device Ids and we need the output
of the command above.
R32 (all of them)
R40 (only if equipped with an ATI Mobility Radeon 7000, PCI ID 1002:4c59, NOT
if equipped with an ATI Mobility Radeon 7500, PCI ID 1002:4c57)
R50 (only if equipped with an ATI Mobility Radeon 7500, PCI ID 1002:4c57, NOT
if equipped with an ATI Mobility Radeon 9000, PCI ID 1002:4c66)
T40 (only if equipped with an ATI Mobility Radeon 9000, PCI ID 1002:4c66, NOT
if equipped with an ATI Mobility Radeon 7500, PCI ID 1002:4c57)
X21 (all of them)
X31 (all of them)
X32 (all of them)
I thank you warmly in advance
Giorgio Lando
----- End forwarded message -----