Bug 492116
Summary: | iwl3945 causes panic with 2.6.27.9-159.fc10.x86_64 | ||||||||
---|---|---|---|---|---|---|---|---|---|
Product: | [Fedora] Fedora | Reporter: | Mike Harvey <mharvey> | ||||||
Component: | kernel | Assignee: | John W. Linville <linville> | ||||||
Status: | CLOSED INSUFFICIENT_DATA | QA Contact: | Fedora Extras Quality Assurance <extras-qa> | ||||||
Severity: | medium | Docs Contact: | |||||||
Priority: | low | ||||||||
Version: | 10 | CC: | jlaska, kernel-maint, quintela, reinette.chatre | ||||||
Target Milestone: | --- | ||||||||
Target Release: | --- | ||||||||
Hardware: | All | ||||||||
OS: | Linux | ||||||||
Whiteboard: | |||||||||
Fixed In Version: | Doc Type: | Bug Fix | |||||||
Doc Text: | Story Points: | --- | |||||||
Clone Of: | 480701 | Environment: | |||||||
Last Closed: | 2009-05-21 01:24:36 UTC | Type: | --- | ||||||
Regression: | --- | Mount Type: | --- | ||||||
Documentation: | --- | CRM: | |||||||
Verified Versions: | Category: | --- | |||||||
oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |||||||
Cloudforms Team: | --- | Target Upstream Version: | |||||||
Embargoed: | |||||||||
Bug Depends On: | 480701 | ||||||||
Bug Blocks: | |||||||||
Attachments: |
|
Description
Mike Harvey
2009-03-25 13:16:49 UTC
iwl3945_mac_conf_tx wraps mutex_lock(&priv->mutex) around iwl3945_activate_qos iwlagn just uses priv->lock Proper usage of priv->mutex versus priv->lock in iwl3945 is unclear to me, so I'm making Intel team aware of the issue. Created attachment 336678 [details]
change locking
Could you please test with attached patch?
Sorry. I rebuilt my O/S from scratch yesterday. This was my primary use system and I was getting Kernel panics in the middle of work and loosing work. I am running the Latest Fedora 10 and I have not seen the Kernel panic since I reinstalled the O/S. [mharvey@dhcp231-215 ~]$ uname -a Linux dhcp231-215.rdu.redhat.com 2.6.27.19-170.2.35.fc10.i686 #1 SMP Mon Feb 23 13:21:22 EST 2009 i686 i686 i386 GNU/Linux That patch certainly matches the iwlagn version. Are you confident that it complies with the other usage of priv->mutex and priv->lock? Protecting qos_data with priv->lock is the right thing to do and I am thus confident about this particular case. wrt the other usages of priv->mutex and priv->lock we may have a few cases that are still inconsistent/incorrect even after this fix. Kernel w/ patch from comment 2 is building here: http://koji.fedoraproject.org/koji/buildinfo?buildID=96422 Please give that a try when the RPMs become available and post the results here...thanks! In NEEDINFO state for 45 days, closing... |