Search String: Display: Description: Sort:

Results:

References: [ +subject:/^(?:^\s*(re|sv|fwd|fw)[\[\]\d]*[:>-]+\s*)*review\:\s+don\'t\s+hold\s+ilock\s+when\s+calling\s+vn_iowait\s*$/: 12 ]

Total 12 documents matching your query.

1. review: don't hold ilock when calling vn_iowait (score: 1)
Author: David Chinner <dgc@xxxxxxx>
Date: Mon, 23 Apr 2007 09:03:03 +1000
Regression introduced by recent freezing fixes - we should not hold the ilock while waiting for I/O completion. Cheers, Dave. -- Dave Chinner Principal Engineer SGI Australian Software Group -- fs/xf
/archives/xfs/2007-04/msg00112.html (12,394 bytes)

2. Re: review: don't hold ilock when calling vn_iowait (score: 1)
Author: Christoph Hellwig <hch@xxxxxxxxxxxxx>
Date: Mon, 23 Apr 2007 22:43:38 +0100
Looks good, and actually simplies the twisted maze the xfs_sync_inodes is a little bit. And the missing IPOINTER_INSERT in the SYNC_CLOSE case looks like an actual bugfix. Of course in the end I'd st
/archives/xfs/2007-04/msg00118.html (8,479 bytes)

3. Re: review: don't hold ilock when calling vn_iowait (score: 1)
Author: David Chinner <dgc@xxxxxxx>
Date: Tue, 24 Apr 2007 09:17:06 +1000
I had to look closely at that IPOINTER_INSERT case with SYNC_CLOSE; it was actaully working properly because you'd always end up in the SYNC_CLOSE case having inserted a pointer earlier on in the flo
/archives/xfs/2007-04/msg00132.html (9,277 bytes)

4. Re: review: don't hold ilock when calling vn_iowait (score: 1)
Author: Timothy Shimmin <tes@xxxxxxx>
Date: Tue, 24 Apr 2007 12:00:38 +1000
Looks good, and actually simplies the twisted maze the xfs_sync_inodes is a little bit. And the missing IPOINTER_INSERT in the SYNC_CLOSE case looks like an actual bugfix. I had to look closely at th
/archives/xfs/2007-04/msg00136.html (10,630 bytes)

5. Re: review: don't hold ilock when calling vn_iowait (score: 1)
Author: David Chinner <dgc@xxxxxxx>
Date: Tue, 24 Apr 2007 13:08:26 +1000
Doesn't everyone? I confused that with the removal of the vp == NULL checks I removed. Too many things, so little time. So yes, this probably does fix a bug. Yes, we should be looking to rip all this
/archives/xfs/2007-04/msg00137.html (10,518 bytes)

6. Re: review: don't hold ilock when calling vn_iowait (score: 1)
Author: Christoph Hellwig <hch@xxxxxxxxxxxxx>
Date: Tue, 24 Apr 2007 10:10:00 +0100
In theory it does the same thing. The problem is that it's really hard to verify. Btw, before starting with this bit there's another item on my TODO list to simplify xfs_sync_inodes, and that's getti
/archives/xfs/2007-04/msg00142.html (9,515 bytes)

7. review: don't hold ilock when calling vn_iowait (score: 1)
Author: David Chinner <dgc@xxxxxxx>
Date: Mon, 23 Apr 2007 09:03:03 +1000
Regression introduced by recent freezing fixes - we should not hold the ilock while waiting for I/O completion. Cheers, Dave. -- Dave Chinner Principal Engineer SGI Australian Software Group -- fs/xf
/archives/xfs/2007-04/msg00317.html (12,394 bytes)

8. Re: review: don't hold ilock when calling vn_iowait (score: 1)
Author: Christoph Hellwig <hch@xxxxxxxxxxxxx>
Date: Mon, 23 Apr 2007 22:43:38 +0100
Looks good, and actually simplies the twisted maze the xfs_sync_inodes is a little bit. And the missing IPOINTER_INSERT in the SYNC_CLOSE case looks like an actual bugfix. Of course in the end I'd st
/archives/xfs/2007-04/msg00323.html (8,479 bytes)

9. Re: review: don't hold ilock when calling vn_iowait (score: 1)
Author: David Chinner <dgc@xxxxxxx>
Date: Tue, 24 Apr 2007 09:17:06 +1000
I had to look closely at that IPOINTER_INSERT case with SYNC_CLOSE; it was actaully working properly because you'd always end up in the SYNC_CLOSE case having inserted a pointer earlier on in the flo
/archives/xfs/2007-04/msg00337.html (9,277 bytes)

10. Re: review: don't hold ilock when calling vn_iowait (score: 1)
Author: Timothy Shimmin <tes@xxxxxxx>
Date: Tue, 24 Apr 2007 12:00:38 +1000
Looks good, and actually simplies the twisted maze the xfs_sync_inodes is a little bit. And the missing IPOINTER_INSERT in the SYNC_CLOSE case looks like an actual bugfix. I had to look closely at th
/archives/xfs/2007-04/msg00341.html (10,630 bytes)

11. Re: review: don't hold ilock when calling vn_iowait (score: 1)
Author: David Chinner <dgc@xxxxxxx>
Date: Tue, 24 Apr 2007 13:08:26 +1000
Doesn't everyone? I confused that with the removal of the vp == NULL checks I removed. Too many things, so little time. So yes, this probably does fix a bug. Yes, we should be looking to rip all this
/archives/xfs/2007-04/msg00342.html (10,518 bytes)

12. Re: review: don't hold ilock when calling vn_iowait (score: 1)
Author: Christoph Hellwig <hch@xxxxxxxxxxxxx>
Date: Tue, 24 Apr 2007 10:10:00 +0100
In theory it does the same thing. The problem is that it's really hard to verify. Btw, before starting with this bit there's another item on my TODO list to simplify xfs_sync_inodes, and that's getti
/archives/xfs/2007-04/msg00347.html (9,515 bytes)


This search system is powered by Namazu