public inbox for linux-xfs@vger.kernel.org
 help / color / mirror / Atom feed
* [PATCH] Remove XFS_BUF_SHUT() and friends
@ 2008-12-04  7:27 Lachlan McIlroy
  2008-12-04 13:27 ` Christoph Hellwig
  0 siblings, 1 reply; 4+ messages in thread
From: Lachlan McIlroy @ 2008-12-04  7:27 UTC (permalink / raw)
  To: xfs-oss

Code does nothing so remove it.

--- a/fs/xfs/linux-2.6/xfs_buf.h	2008-12-04 14:24:12.000000000 +1100
+++ b/fs/xfs/linux-2.6/xfs_buf.h	2008-12-04 14:24:22.000000000 +1100
@@ -311,10 +311,6 @@ extern void xfs_buf_trace(xfs_buf_t *, c
  #define XFS_BUF_UNORDERED(bp)	((bp)->b_flags &= ~XBF_ORDERED)
  #define XFS_BUF_ISORDERED(bp)	((bp)->b_flags & XBF_ORDERED)

-#define XFS_BUF_SHUT(bp)	do { } while (0)
-#define XFS_BUF_UNSHUT(bp)	do { } while (0)
-#define XFS_BUF_ISSHUT(bp)	(0)
-
  #define XFS_BUF_HOLD(bp)	xfs_buf_hold(bp)
  #define XFS_BUF_READ(bp)	((bp)->b_flags |= XBF_READ)
  #define XFS_BUF_UNREAD(bp)	((bp)->b_flags &= ~XBF_READ)
--- a/fs/xfs/xfs_buf_item.c	2008-12-04 14:24:12.000000000 +1100
+++ b/fs/xfs/xfs_buf_item.c	2008-12-04 14:13:32.000000000 +1100
@@ -998,21 +1000,7 @@ xfs_buf_iodone_callbacks(
  			xfs_buf_do_callbacks(bp, lip);
  			XFS_BUF_SET_FSPRIVATE(bp, NULL);
  			XFS_BUF_CLR_IODONE_FUNC(bp);
-
-			/*
-			 * XFS_SHUT flag gets set when we go thru the
-			 * entire buffer cache and deliberately start
-			 * throwing away delayed write buffers.
-			 * Since there's no biowait done on those,
-			 * we should just brelse them.
-			 */
-			if (XFS_BUF_ISSHUT(bp)) {
-			    XFS_BUF_UNSHUT(bp);
-				xfs_buf_relse(bp);
-			} else {
-				xfs_biodone(bp);
-			}
-
+			xfs_biodone(bp);
  			return;
  		}

--- a/fs/xfs/xfs_inode.c	2008-12-04 14:24:12.000000000 +1100
+++ b/fs/xfs/xfs_inode.c	2008-12-04 14:21:54.000000000 +1100
@@ -3089,7 +3089,6 @@ cluster_corrupt_out:
  			XFS_BUF_CLR_BDSTRAT_FUNC(bp);
  			XFS_BUF_UNDONE(bp);
  			XFS_BUF_STALE(bp);
-			XFS_BUF_SHUT(bp);
  			XFS_BUF_ERROR(bp,EIO);
  			xfs_biodone(bp);
  		} else {

_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [PATCH] Remove XFS_BUF_SHUT() and friends
  2008-12-04  7:27 [PATCH] Remove XFS_BUF_SHUT() and friends Lachlan McIlroy
@ 2008-12-04 13:27 ` Christoph Hellwig
  2008-12-05  3:52   ` Lachlan McIlroy
  0 siblings, 1 reply; 4+ messages in thread
From: Christoph Hellwig @ 2008-12-04 13:27 UTC (permalink / raw)
  To: Lachlan McIlroy; +Cc: xfs-oss

On Thu, Dec 04, 2008 at 06:27:16PM +1100, Lachlan McIlroy wrote:
> Code does nothing so remove it.

Looks good to me.  Does anyone remember why this flag was set on IRIX?

_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [PATCH] Remove XFS_BUF_SHUT() and friends
  2008-12-04 13:27 ` Christoph Hellwig
@ 2008-12-05  3:52   ` Lachlan McIlroy
  2008-12-17 14:20     ` Christoph Hellwig
  0 siblings, 1 reply; 4+ messages in thread
From: Lachlan McIlroy @ 2008-12-05  3:52 UTC (permalink / raw)
  To: Christoph Hellwig; +Cc: xfs-oss

Christoph Hellwig wrote:
> On Thu, Dec 04, 2008 at 06:27:16PM +1100, Lachlan McIlroy wrote:
>> Code does nothing so remove it.
> 
> Looks good to me.  Does anyone remember why this flag was set on IRIX?

During a forced shutdown we would go through all incore buffers and mark
them B_STALE/B_ERROR/B_SHUT and then call biodone() on them.  The code in
xfs_buf_iodone_callbacks() that checks for B_SHUT is to avoid a double
biodone() I think.  We don't have the code in the Linux port that scans
through all the incore buffers on a forced shutdown but we still have one
case in xfs_iflush_cluster().  In any case it's not doing anything.

_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [PATCH] Remove XFS_BUF_SHUT() and friends
  2008-12-05  3:52   ` Lachlan McIlroy
@ 2008-12-17 14:20     ` Christoph Hellwig
  0 siblings, 0 replies; 4+ messages in thread
From: Christoph Hellwig @ 2008-12-17 14:20 UTC (permalink / raw)
  To: Lachlan McIlroy; +Cc: Christoph Hellwig, xfs-oss

On Fri, Dec 05, 2008 at 02:52:47PM +1100, Lachlan McIlroy wrote:
> Christoph Hellwig wrote:
> > On Thu, Dec 04, 2008 at 06:27:16PM +1100, Lachlan McIlroy wrote:
> >> Code does nothing so remove it.
> > 
> > Looks good to me.  Does anyone remember why this flag was set on IRIX?
> 
> During a forced shutdown we would go through all incore buffers and mark
> them B_STALE/B_ERROR/B_SHUT and then call biodone() on them.  The code in
> xfs_buf_iodone_callbacks() that checks for B_SHUT is to avoid a double
> biodone() I think.  We don't have the code in the Linux port that scans
> through all the incore buffers on a forced shutdown but we still have one
> case in xfs_iflush_cluster().  In any case it's not doing anything.

Ok.

_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs

^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2008-12-17 14:20 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2008-12-04  7:27 [PATCH] Remove XFS_BUF_SHUT() and friends Lachlan McIlroy
2008-12-04 13:27 ` Christoph Hellwig
2008-12-05  3:52   ` Lachlan McIlroy
2008-12-17 14:20     ` Christoph Hellwig

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox