Red Hat Bugzilla – Bug 160820
nVidia driver requires upstream page_attr patch
Last modified: 2007-11-30 17:07:07 EST
Description of problem:
Latest version of nVidia driver (7175 or better) requires the attached upstream
patch to run correctly. Without the patch, system hangs randomly and/or video
is corrupted. This is critical to DreamWorks rollout of new hardware. This is
also critical to HP's sales of xw9300 workstations.
The attached patch was downloaded from
which seems to have been removed since it was originally downloaded.
The patch appears to be a combination of the patches available from
Created attachment 115615 [details]
Patch to fix issues with nVidia driver
Additional Note: This is needed for both x86_64 and i386 architectures.
I have built test i386 and x86_64 (and ia32e) kernels containing the patch
in Comment #1.
In this location:
There are 6 kernel binary packages along with the associated kernel src.rpm:
At least one of the x86 and one of the x86_64 kernels need to be tested
with the nVidea driver.
DreamWorks has already tested the 2.4.21-32, 2.4.21-31, and 2.4.21-27 kernels
with the pageattr patch built into them. I built these for them just for
testing purposes, and they've said that all three kernels have worked well for
them. Would you like me to ask them to also test the 2.4.21-32.8.bz160820
kernels? Or are the previous tests good enough?
They haven't tested a 32-bit kernel yet, so I'll ask them to test that one.
DreamWorks says that they can test the 32-bit kernel but they need the
kernel-source package to test the nVidia driver with the kernel. Can that be
uploaded to people.redhat.com/anderson/.VZ_160820 as well?
Since we can't test them here, yes, please have them test
at least one of the x86 kernels and one of the x86_64 kernels
(x86_64 or ia32e). The kernel-source packages have been
Thanks. I've asked DreamWorks to test both x86_64 and x86.
Reverting to NEEDINFO state.
DreamWorks has said that they have tested both the 32- and 64-bit kernels. The
nVidia driver installs cleanly and everything worked fine. He used the system
all day with no incidents (which would have been highly improbably without the
Should we wait for a successful test report from HP as well?
I'm not Carnac, but I'll post the patch to rhkernel-list and
set this BZ on the RHEL3-U6 Proposed list.
A fix for this problem has just been committed to the RHEL3 U6
patch pool this evening (in kernel version 2.4.21-32.10.EL).
From User-Agent: XML-RPC
HOTFIX Requested http://seg.rdu.redhat.com/scripts/hotfix/edit.pl?id=396
From User-Agent: XML-RPC
HOTFIX Request has been released http://seg.rdu.redhat.com/scripts/hotfix/edit.pl?id=396
An advisory has been issued which should help the problem
described in this bug report. This report is therefore being
closed with a resolution of ERRATA. For more information
on the solution and/or where to find the updated files,
please follow the link below. You may reopen this bug report
if the solution does not work for you.