All of lore.kernel.org
 help / color / mirror / Atom feed
From: Christoph Hellwig <hch@lst.de>
To: linux-xfs@vger.kernel.org
Cc: eguan@redhat.com, darrick.wong@oracle.com
Subject: [PATCH 2/4] xfs: calculate minleft correctly for bmap allocations
Date: Tue, 13 Dec 2016 16:59:25 +0100	[thread overview]
Message-ID: <1481644767-9098-3-git-send-email-hch@lst.de> (raw)
In-Reply-To: <1481644767-9098-1-git-send-email-hch@lst.de>

We need to take into account the worst possible number of bmap and rmap
btree block allocations.  Use the helper we have to calculate it instead
of doing an overly optimistic estimate.  Also move the call to calculate
it into the function that sets the allocation parameters, as we have
more accurate information about the allocation there.

Signed-off-by: Christoph Hellwig <hch@lst.de>
---
 fs/xfs/libxfs/xfs_bmap.c | 16 ++++++----------
 fs/xfs/libxfs/xfs_bmap.h |  1 -
 2 files changed, 6 insertions(+), 11 deletions(-)

diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c
index 44773c9..2126ecc 100644
--- a/fs/xfs/libxfs/xfs_bmap.c
+++ b/fs/xfs/libxfs/xfs_bmap.c
@@ -3764,7 +3764,12 @@ xfs_bmap_btalloc(
 		args.alignment = 1;
 		args.minalignslop = 0;
 	}
-	args.minleft = ap->minleft;
+
+	/*
+	 * Ensure we have enough blocks left in the AG to satisfy all possible
+	 * bmap and rmap tree block allocations.
+	 */
+	args.minleft = xfs_bmap_worst_indlen(ap->ip, args.maxlen);
 	args.wasdel = ap->wasdel;
 	args.resv = XFS_AG_RESV_NONE;
 	args.datatype = ap->datatype;
@@ -4572,15 +4577,6 @@ xfs_bmapi_write(
 
 	XFS_STATS_INC(mp, xs_blk_mapw);
 
-	if (*firstblock == NULLFSBLOCK) {
-		if (XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_BTREE)
-			bma.minleft = be16_to_cpu(ifp->if_broot->bb_level) + 1;
-		else
-			bma.minleft = 1;
-	} else {
-		bma.minleft = 0;
-	}
-
 	if (!(ifp->if_flags & XFS_IFEXTENTS)) {
 		error = xfs_iread_extents(tp, ip, whichfork);
 		if (error)
diff --git a/fs/xfs/libxfs/xfs_bmap.h b/fs/xfs/libxfs/xfs_bmap.h
index cecd094..92e6755 100644
--- a/fs/xfs/libxfs/xfs_bmap.h
+++ b/fs/xfs/libxfs/xfs_bmap.h
@@ -49,7 +49,6 @@ struct xfs_bmalloca {
 
 	xfs_extlen_t		total;	/* total blocks needed for xaction */
 	xfs_extlen_t		minlen;	/* minimum allocation size (blocks) */
-	xfs_extlen_t		minleft; /* amount must be left after alloc */
 	bool			eof;	/* set if allocating past last extent */
 	bool			wasdel;	/* replacing a delayed allocation */
 	bool			aeof;	/* allocated space at eof */
-- 
2.1.4


  parent reply	other threads:[~2016-12-13 15:59 UTC|newest]

Thread overview: 31+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-12-13 15:59 minleft fixes Christoph Hellwig
2016-12-13 15:59 ` [PATCH 1/4] xfs: fix bogus minleft manipulations Christoph Hellwig
2016-12-14 17:35   ` Brian Foster
2016-12-14 19:36     ` Christoph Hellwig
2016-12-14 21:51       ` Brian Foster
     [not found]         ` <20161215143430.GB29477@bfoster.bfoster>
2016-12-16  8:21           ` Christoph Hellwig
2016-12-19 11:38           ` Christoph Hellwig
2016-12-20 14:17             ` Brian Foster
2016-12-20 21:45               ` Dave Chinner
2016-12-15 22:09   ` Dave Chinner
2016-12-16  8:20     ` Christoph Hellwig
2017-01-04  6:32       ` Darrick J. Wong
2017-01-04  8:50         ` Christoph Hellwig
2016-12-13 15:59 ` Christoph Hellwig [this message]
2016-12-14 18:24   ` [PATCH 2/4] xfs: calculate minleft correctly for bmap allocations Brian Foster
2016-12-13 15:59 ` [PATCH 3/4] xfs: adjust allocation length in xfs_alloc_space_available Christoph Hellwig
2016-12-14 18:24   ` Brian Foster
2016-12-14 19:37     ` Christoph Hellwig
2016-12-15 20:41   ` Libor Klepáč
2016-12-16  8:20     ` Christoph Hellwig
2016-12-16  0:28   ` Dave Chinner
2016-12-16  8:25     ` Christoph Hellwig
2016-12-18 23:55       ` Dave Chinner
2016-12-13 15:59 ` [PATCH 4/4] xfs: don't rely on ->total " Christoph Hellwig
2016-12-14 18:30   ` Brian Foster
2016-12-14 19:38     ` Christoph Hellwig
2016-12-14 21:51       ` Brian Foster
2016-12-15  8:55         ` Christoph Hellwig
2016-12-15 12:00           ` Brian Foster
2016-12-14 10:51 ` minleft fixes Eryu Guan
2016-12-15 10:24   ` Eryu Guan

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=1481644767-9098-3-git-send-email-hch@lst.de \
    --to=hch@lst.de \
    --cc=darrick.wong@oracle.com \
    --cc=eguan@redhat.com \
    --cc=linux-xfs@vger.kernel.org \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.