Bug 170545

Summary: kernel_lock() problem through NFS mount
Product: Red Hat Enterprise Linux 4 Reporter: Linda Wang <lwang>
Component: kernelAssignee: Steve Dickson <steved>
Status: CLOSED DUPLICATE QA Contact: Brian Brock <bbrock>
Severity: medium Docs Contact:
Priority: medium    
Version: 4.0CC: jbaron, lwoodman, staubach
Target Milestone: ---   
Target Release: ---   
Hardware: All   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2005-10-20 14:24:24 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:
Attachments:
Description Flags
Backport of the entire NFS locking changes from 2.6.11 none

Description Linda Wang 2005-10-12 18:19:26 UTC
From Bugzilla Helper:
User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7.8) Gecko/20050512 Red Hat/1.0.4-1.4.1 Firefox/1.0.4

Description of problem:
We have discovered a problem where the kernel lock is still being held by a process
when it returns to user mode.  This results in other processes that are in the kernel
and want the kernel lock spinning inside of lock_kernel() and reacquire_kernel_lock()
even though the real lock holder is no longer in the kernel.

Version-Release number of selected component (if applicable):
kernel-2.6.9

How reproducible:
Always

Steps to Reproduce:
1.install RHEL4
2.mount a nfs filesystem
3.unlock a nfs mounted file
  

Actual Results:  all sorts of weird behavior happens with kernel lock held and going to user space.. espeically on SMP systems, you can notice the slowness of the system since it renders other cpus useless during swapping to user space.

Expected Results:  On smp kernel, there shouldn't be any slowness when a cpu goes into user space.

Additional info:

Comment 1 Linda Wang 2005-10-12 19:00:32 UTC
*** Bug 170546 has been marked as a duplicate of this bug. ***

Comment 2 Larry Woodman 2005-10-13 02:24:32 UTC
Created attachment 119871 [details]
Backport of the entire NFS locking changes from 2.6.11

I think we should include the entire NFS locking code from 2.6.11.  This fixes
several known problems including exiting the kernel with the kernel_lock still
held.

Larry Woodman

Comment 4 Steve Dickson 2005-10-20 14:24:24 UTC

*** This bug has been marked as a duplicate of 167192 ***