Bug 445674

Summary: Direct I/O cache invalidation after sync writes
Product: Red Hat Enterprise Linux 5 Reporter: David Mair <dmair>
Component: kernelAssignee: Jeff Moyer <jmoyer>
Status: CLOSED ERRATA QA Contact: Martin Jenner <mjenner>
Severity: high Docs Contact:
Priority: high    
Version: 5.1CC: donf, duck, dzickus, esandeen, jmoyer, richardl
Target Milestone: rcKeywords: Regression
Target Release: ---   
Hardware: All   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2009-01-20 19:57:24 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:    
Bug Blocks: 448732    
Attachments:
Description Flags
dio: fix cache invalidation after sync writes none

Description David Mair 2008-05-08 13:52:47 UTC 
Description of problem:
In BZ232715 a patch from upstream was committed to resolve double AIO completion
due to invalidate_inode_pages2 failure.  Unfortunately that patch introduced a
regression in the kernel that can cause a transient journal file corruption.  A
Red Hat ISV partner reported this problem to Red Hat very recently.

Version-Release number of selected component (if applicable):
RHEL5.1 (kernel-2.6.18-53)

How reproducible:
Can be reproduced

Steps to Reproduce:
1.
2.
3.
  
Actual results:


Expected results:


Additional info:
This problem was resolved in the upstream kernel with commit
bdb76ef5a4bc8676a81034a443f1eda450b4babb ("dio: fix cache invalidation after
sync writes")

A thread on lkml (including patch) can be found at the link below.

http://kerneltrap.org/mailarchive/linux-kernel/2007/10/26/360497
Comment 1 RHEL Program Management 2008-05-08 14:05:18 UTC 
This bugzilla has Keywords: Regression.  

Since no regressions are allowed between releases, 
it is also being proposed as a blocker for this release.  

Please resolve ASAP.
Comment 2 Jeff Moyer 2008-05-28 18:49:28 UTC 
We have a regression test for this (aio-dio-regress) which gets run through
RHTS.  I also modified the test to check the async DIO case, and there is an
additional failure there.  I have a patch that I'm running by our performance
team for that.  Once I have sign-off from them, I'll post the two patches here.

Again, test for this is covered by the aio-dio-regress regression tests that we
run via RHTS.
Comment 3 RHEL Program Management 2008-05-28 18:59:01 UTC 
This request was evaluated by Red Hat Product Management for inclusion in a Red
Hat Enterprise Linux maintenance release.  Product Management has requested
further review of this request by Red Hat Engineering, for potential
inclusion in a Red Hat Enterprise Linux Update release for currently deployed
products.  This request is not yet committed for inclusion in an Update
release.
Comment 5 Jeff Moyer 2008-07-03 15:49:38 UTC 
Created attachment 310935 [details]
dio: fix cache invalidation after sync writes

This patch fixes the aio-dio-regress test case.
Comment 6 Don Zickus 2008-07-23 18:55:40 UTC 
in kernel-2.6.18-99.el5
You can download this test kernel from http://people.redhat.com/dzickus/el5
Comment 11 errata-xmlrpc 2009-01-20 19:57:24 UTC 
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 therefore 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-2009-0225.html