Greetings,
I was hit by an infamous hardware incompatibility between VIA VT6420/1 sata card (sata_via) and newer Western Digital drives.
The bug has been worked around in mainline kernel at 2.6.32.7, * https://bugzilla.kernel.org/show_bug.cgi?id=15173 * http://article.gmane.org/gmane.linux.ide/46352 * http://thread.gmane.org/gmane.linux.kernel/1062139
The sata_via patch from bug#15173 applies cleanly to sata_via module of 2.6.18-238 and now I'm running using a 'fixed' module with no problem.
Unfortunately the patched module uses symbols not from kABI whitelist so universal replacement kernel module could not be made, but the patch itself could be integrated to upstream's tree or into the CentOS-Plus kernel. As I'm not a RHEL subscriber, I can only ask for the latter.
Any ideas?
Best regards, Dmitry Mikhailov
On Wed, May 4, 2011 at 12:31 AM, Dmitry E. Mikhailov d.mikhailov@infocommunications.ru wrote:
Greetings,
I was hit by an infamous hardware incompatibility between VIA VT6420/1 sata card (sata_via) and newer Western Digital drives.
The bug has been worked around in mainline kernel at 2.6.32.7, * https://bugzilla.kernel.org/show_bug.cgi?id=15173 * http://article.gmane.org/gmane.linux.ide/46352 * http://thread.gmane.org/gmane.linux.kernel/1062139
The sata_via patch from bug#15173 applies cleanly to sata_via module of 2.6.18-238 and now I'm running using a 'fixed' module with no problem.
Unfortunately the patched module uses symbols not from kABI whitelist so universal replacement kernel module could not be made, but the patch itself could be integrated to upstream's tree or into the CentOS-Plus kernel. As I'm not a RHEL subscriber, I can only ask for the latter.
Any ideas?
Getting the patch into the centosplus kernel is as easy as filing a request at http://bugs.centos.org . :)
Regarding the upstream kernel, you don't have to be a customer to report a bug at http://bugzilla.redhat.com . In this case, because the patch is already in mainline kernel, there is some good chance that it is picked up by Red Hat.
Akemi
On 04/05/11 08:31, Dmitry E. Mikhailov wrote:
Greetings,
I was hit by an infamous hardware incompatibility between VIA VT6420/1 sata card (sata_via) and newer Western Digital drives.
The bug has been worked around in mainline kernel at 2.6.32.7, * https://bugzilla.kernel.org/show_bug.cgi?id=15173 * http://article.gmane.org/gmane.linux.ide/46352 * http://thread.gmane.org/gmane.linux.kernel/1062139
The sata_via patch from bug#15173 applies cleanly to sata_via module of 2.6.18-238 and now I'm running using a 'fixed' module with no problem.
Unfortunately the patched module uses symbols not from kABI whitelist so universal replacement kernel module could not be made, but the patch itself could be integrated to upstream's tree or into the CentOS-Plus kernel. As I'm not a RHEL subscriber, I can only ask for the latter.
Any ideas?
Best regards, Dmitry Mikhailov
Hi Dmitry,
There are actually 3 relevant patches in the latest upstream kernel (2.6.38.5) for this issue:
2010-11-19 Tejun Heo sata_via: apply magic FIFO fix to vt6420 too 2010-06-02 Tejun Heo sata_via: magic vt6421 fix for transmission problems... 2010-06-07 Tejun Heo sata_via: explain the magic fix
ELRepo has updated drivers for sata_via available that fix this issue for both EL5 and EL6:
http://elrepo.org http://elrepo.org/tiki/kmod-sata_via http://lists.elrepo.org/pipermail/elrepo/2011-May/000660.html
As this isn't really a CentOS Development issue, can I ask that we take any discussion of the above to the elrepo mailing lists and/or bug tracker:
http://lists.elrepo.org/mailman/listinfo/elrepo http://elrepo.org/bugs/
Thanks.
-
Akemi Yagi -
Dmitry E. Mikhailov -
Ned Slider