Escalated to Bugzilla from IssueTracker
Created attachment 120327 [details] Candidate #1 - undo wrong changes, add spinlock
I am very sorry to report, but this appears to be a regression in the fix I made for the bug 135583. It is obvious that we have to undo those. To compensate, I added the spinlock around the call to remove_wait_queue.
A test kernel is available at this URL: ftp://people.redhat.com/zaitcev/171129 I need this tested on rx5670.
Pete, Can you also post a kernel-source RPM for the test kernel? The kernel-source RPM is required to use the nvidia driver. Thanks, Lonni
SRPM and kernel-source are uploaded.
Thanks, could you also post the kernel-source RPM for x86_64? That's where I'm seeing this problem.
Nevermind, I see that the i386 kernel-source RPM works just as well in x86_64. Thanks.
Actually, I take that back, I can't build/install the nvidia driver in an x86_64 environment if I install the i386 kernel source RPM. Pete, can you please post the x86_64 kernel-source RPM as well? thanks!
In my testing this kernel resolves the USB crash. Thanks.
A fix for this problem has just been committed to the RHEL3 U7 patch pool this evening (in kernel version 2.4.21-37.12.EL).
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. http://rhn.redhat.com/errata/RHSA-2006-0144.html
*** Bug 181627 has been marked as a duplicate of this bug. ***