public inbox for linux-xfs@vger.kernel.org
 help / color / mirror / Atom feed
From: Dave Chinner <david@fromorbit.com>
To: xfs@oss.sgi.com
Subject: [PATCH 05/16] xfs: rename xfs_buf_get_nodaddr to be more appropriate
Date: Wed, 22 Sep 2010 16:44:18 +1000	[thread overview]
Message-ID: <1285137869-10310-6-git-send-email-david@fromorbit.com> (raw)
In-Reply-To: <1285137869-10310-1-git-send-email-david@fromorbit.com>

From: Dave Chinner <dchinner@redhat.com>

xfs_buf_get_nodaddr() is really used to allocate a buffer that is
uncached. While it is not directly assigned a disk address, the fact
that they are not cached is a more important distinction. With the
upcoming uncached buffer read primitive, we should be consistent
with this disctinction.

While there, make page allocation in xfs_buf_get_nodaddr() safe
against memory reclaim re-entrancy into the filesystem by changing
the allocation to GFP_NOFS.

Signed-off-by: Dave Chinner <dchinner@redhat.com>
---
 fs/xfs/linux-2.6/xfs_buf.c   |    7 ++++---
 fs/xfs/linux-2.6/xfs_buf.h   |    2 +-
 fs/xfs/linux-2.6/xfs_trace.h |    2 +-
 fs/xfs/xfs_log.c             |    3 ++-
 fs/xfs/xfs_log_recover.c     |    3 ++-
 fs/xfs/xfs_vnodeops.c        |    2 +-
 6 files changed, 11 insertions(+), 8 deletions(-)

diff --git a/fs/xfs/linux-2.6/xfs_buf.c b/fs/xfs/linux-2.6/xfs_buf.c
index 286e36e..fbbc6d3 100644
--- a/fs/xfs/linux-2.6/xfs_buf.c
+++ b/fs/xfs/linux-2.6/xfs_buf.c
@@ -707,8 +707,9 @@ xfs_buf_associate_memory(
 }
 
 xfs_buf_t *
-xfs_buf_get_noaddr(
+xfs_buf_get_uncached(
 	size_t			len,
+	int			flags,
 	xfs_buftarg_t		*target)
 {
 	unsigned long		page_count = PAGE_ALIGN(len) >> PAGE_SHIFT;
@@ -725,7 +726,7 @@ xfs_buf_get_noaddr(
 		goto fail_free_buf;
 
 	for (i = 0; i < page_count; i++) {
-		bp->b_pages[i] = alloc_page(GFP_KERNEL);
+		bp->b_pages[i] = alloc_page(xb_to_gfp(flags));
 		if (!bp->b_pages[i])
 			goto fail_free_mem;
 	}
@@ -740,7 +741,7 @@ xfs_buf_get_noaddr(
 
 	xfs_buf_unlock(bp);
 
-	trace_xfs_buf_get_noaddr(bp, _RET_IP_);
+	trace_xfs_buf_get_uncached(bp, _RET_IP_);
 	return bp;
 
  fail_free_mem:
diff --git a/fs/xfs/linux-2.6/xfs_buf.h b/fs/xfs/linux-2.6/xfs_buf.h
index 2a05614..fbcc77b 100644
--- a/fs/xfs/linux-2.6/xfs_buf.h
+++ b/fs/xfs/linux-2.6/xfs_buf.h
@@ -213,7 +213,7 @@ extern xfs_buf_t *xfs_buf_read(xfs_buftarg_t *, xfs_off_t, size_t,
 				xfs_buf_flags_t);
 
 extern xfs_buf_t *xfs_buf_get_empty(size_t, xfs_buftarg_t *);
-extern xfs_buf_t *xfs_buf_get_noaddr(size_t, xfs_buftarg_t *);
+extern xfs_buf_t *xfs_buf_get_uncached(size_t, int, xfs_buftarg_t *);
 extern int xfs_buf_associate_memory(xfs_buf_t *, void *, size_t);
 extern void xfs_buf_hold(xfs_buf_t *);
 extern void xfs_buf_readahead(xfs_buftarg_t *, xfs_off_t, size_t,
diff --git a/fs/xfs/linux-2.6/xfs_trace.h b/fs/xfs/linux-2.6/xfs_trace.h
index be5dffd..2a1d4fb 100644
--- a/fs/xfs/linux-2.6/xfs_trace.h
+++ b/fs/xfs/linux-2.6/xfs_trace.h
@@ -331,7 +331,7 @@ DEFINE_BUF_EVENT(xfs_buf_iowait_done);
 DEFINE_BUF_EVENT(xfs_buf_delwri_queue);
 DEFINE_BUF_EVENT(xfs_buf_delwri_dequeue);
 DEFINE_BUF_EVENT(xfs_buf_delwri_split);
-DEFINE_BUF_EVENT(xfs_buf_get_noaddr);
+DEFINE_BUF_EVENT(xfs_buf_get_uncached);
 DEFINE_BUF_EVENT(xfs_bdstrat_shut);
 DEFINE_BUF_EVENT(xfs_buf_item_relse);
 DEFINE_BUF_EVENT(xfs_buf_item_iodone);
diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c
index 33f718f..6119a9e 100644
--- a/fs/xfs/xfs_log.c
+++ b/fs/xfs/xfs_log.c
@@ -1131,7 +1131,8 @@ xlog_alloc_log(xfs_mount_t	*mp,
 		iclog->ic_prev = prev_iclog;
 		prev_iclog = iclog;
 
-		bp = xfs_buf_get_noaddr(log->l_iclog_size, mp->m_logdev_targp);
+		bp = xfs_buf_get_uncached(log->l_iclog_size, 0,
+						mp->m_logdev_targp);
 		if (!bp)
 			goto out_free_iclog;
 		if (!XFS_BUF_CPSEMA(bp))
diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c
index 6f3f5fa..2f57be0 100644
--- a/fs/xfs/xfs_log_recover.c
+++ b/fs/xfs/xfs_log_recover.c
@@ -107,7 +107,8 @@ xlog_get_bp(
 		nbblks += log->l_sectBBsize;
 	nbblks = round_up(nbblks, log->l_sectBBsize);
 
-	return xfs_buf_get_noaddr(BBTOB(nbblks), log->l_mp->m_logdev_targp);
+	return xfs_buf_get_uncached(BBTOB(nbblks), 0,
+					log->l_mp->m_logdev_targp);
 }
 
 STATIC void
diff --git a/fs/xfs/xfs_vnodeops.c b/fs/xfs/xfs_vnodeops.c
index 7413a02..d0dc6d0 100644
--- a/fs/xfs/xfs_vnodeops.c
+++ b/fs/xfs/xfs_vnodeops.c
@@ -2431,7 +2431,7 @@ xfs_zero_remaining_bytes(
 	if (endoff > ip->i_size)
 		endoff = ip->i_size;
 
-	bp = xfs_buf_get_noaddr(mp->m_sb.sb_blocksize,
+	bp = xfs_buf_get_uncached(mp->m_sb.sb_blocksize, XBF_DONT_BLOCK,
 				XFS_IS_REALTIME_INODE(ip) ?
 				mp->m_rtdev_targp : mp->m_ddev_targp);
 	if (!bp)
-- 
1.7.1

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

  parent reply	other threads:[~2010-09-22  6:43 UTC|newest]

Thread overview: 39+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-09-22  6:44 [PATCH 0/16] xfs: metadata scalability V2 Dave Chinner
2010-09-22  6:44 ` [PATCH 01/16] xfs: reduce the number of CIL lock round trips during commit Dave Chinner
2010-09-22 16:51   ` Christoph Hellwig
2010-09-22 19:57   ` Alex Elder
2010-09-22  6:44 ` [PATCH 02/16] xfs: remove debug assert for per-ag reference counting Dave Chinner
2010-09-22  6:44 ` [PATCH 03/16] xfs: lockless per-ag lookups Dave Chinner
2010-09-22  6:44 ` [PATCH 04/16] xfs: don't use vfs writeback for pure metadata modifications Dave Chinner
2010-09-22 17:24   ` Christoph Hellwig
2010-09-23  0:36     ` Dave Chinner
2010-09-23 16:19   ` Alex Elder
2010-09-22  6:44 ` Dave Chinner [this message]
2010-09-22 17:25   ` [PATCH 05/16] xfs: rename xfs_buf_get_nodaddr to be more appropriate Christoph Hellwig
2010-09-23  0:37     ` Dave Chinner
2010-09-23 16:22   ` Alex Elder
2010-09-22  6:44 ` [PATCH 06/16] xfs: introduced uncached buffer read primitve Dave Chinner
2010-09-22  6:44 ` [PATCH 07/16] xfs: store xfs_mount in the buftarg instead of in the xfs_buf Dave Chinner
2010-09-22  6:44 ` [PATCH 08/16] xfs: kill XBF_FS_MANAGED buffers Dave Chinner
2010-09-22  6:44 ` [PATCH 09/16] xfs: use unhashed buffers for size checks Dave Chinner
2010-09-22  6:44 ` [PATCH 10/16] xfs: remove buftarg hash for external devices Dave Chinner
2010-09-22  6:44 ` [PATCH 11/16] xfs: split inode AG walking into separate code for reclaim Dave Chinner
2010-09-22 17:28   ` Christoph Hellwig
2010-09-23 16:45   ` Alex Elder
2010-09-22  6:44 ` [PATCH 12/16] xfs: implement batched inode lookups for AG walking Dave Chinner
2010-09-22 17:33   ` Christoph Hellwig
2010-09-23  0:40     ` Dave Chinner
2010-09-23 17:17   ` Alex Elder
2010-09-24  9:15     ` Dave Chinner
2010-09-27 16:05       ` Alex Elder
2010-09-27 17:43       ` Alex Elder
2010-09-22  6:44 ` [PATCH 13/16] xfs: batch inode reclaim lookup Dave Chinner
2010-09-22 17:34   ` Christoph Hellwig
2010-09-23  0:43     ` Dave Chinner
2010-09-23 17:39   ` Alex Elder
2010-09-22  6:44 ` [PATCH 14/16] xfs: serialise inode reclaim within an AG Dave Chinner
2010-09-23 17:50   ` Alex Elder
2010-09-22  6:44 ` [PATCH 16/16] xfs; pack xfs_buf structure more tightly Dave Chinner
2010-09-22 14:53 ` [PATCH 0/16] xfs: metadata scalability V2 Christoph Hellwig
2010-09-22 20:55 ` Alex Elder
2010-09-23  0:46   ` [PATCH 15/16] xfs: convert buffer cache hash to rbtree Dave Chinner

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1285137869-10310-6-git-send-email-david@fromorbit.com \
    --to=david@fromorbit.com \
    --cc=xfs@oss.sgi.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox