All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Darrick J. Wong" <darrick.wong@oracle.com>
To: darrick.wong@oracle.com
Cc: linux-xfs@vger.kernel.org
Subject: [PATCH 2/8] xfs: fix fdblocks accounting w/ RMAPBT per-AG reservation
Date: Thu, 21 Jun 2018 11:31:16 -0700	[thread overview]
Message-ID: <152960587647.26246.17630479835058118272.stgit@magnolia> (raw)
In-Reply-To: <152960586416.26246.8634761888260524091.stgit@magnolia>

From: Darrick J. Wong <darrick.wong@oracle.com>

In __xfs_ag_resv_init we incorrectly calculate the amount by which to
decrease fdblocks when reserving blocks for the rmapbt.  Because rmapbt
allocations do not decrease fdblocks, we must decrease fdblocks by the
entire size of the requested reservation in order to achieve our goal of
always having enough free blocks to satisfy an rmapbt expansion.

This is in contrast to the refcountbt/finobt, which /do/ subtract from
fdblocks whenever they allocate a block.  For this allocation type we
preserve the existing behavior where we decrease fdblocks only by the
requested reservation minus the size of the existing tree.

This fixes the problem where the available block counts reported by
statfs change across a remount if there had been an rmapbt size change
since mount time.

Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
---
 fs/xfs/libxfs/xfs_ag_resv.c |   31 +++++++++++++++++++++++++++----
 1 file changed, 27 insertions(+), 4 deletions(-)


diff --git a/fs/xfs/libxfs/xfs_ag_resv.c b/fs/xfs/libxfs/xfs_ag_resv.c
index 84db76e0e3e3..fecd187fcf2c 100644
--- a/fs/xfs/libxfs/xfs_ag_resv.c
+++ b/fs/xfs/libxfs/xfs_ag_resv.c
@@ -157,6 +157,7 @@ __xfs_ag_resv_free(
 	error = xfs_mod_fdblocks(pag->pag_mount, oldresv, true);
 	resv->ar_reserved = 0;
 	resv->ar_asked = 0;
+	resv->ar_orig_reserved = 0;
 
 	if (error)
 		trace_xfs_ag_resv_free_error(pag->pag_mount, pag->pag_agno,
@@ -189,13 +190,34 @@ __xfs_ag_resv_init(
 	struct xfs_mount		*mp = pag->pag_mount;
 	struct xfs_ag_resv		*resv;
 	int				error;
-	xfs_extlen_t			reserved;
+	xfs_extlen_t			hidden_space;
 
 	if (used > ask)
 		ask = used;
-	reserved = ask - used;
 
-	error = xfs_mod_fdblocks(mp, -(int64_t)reserved, true);
+	switch (type) {
+	case XFS_AG_RESV_RMAPBT:
+		/*
+		 * Space taken by the rmapbt is not subtracted from fdblocks
+		 * because the rmapbt lives in the free space.  Here we must
+		 * subtract the entire reservation from fdblocks so that we
+		 * always have blocks available for rmapbt expansion.
+		 */
+		hidden_space = ask;
+		break;
+	case XFS_AG_RESV_METADATA:
+		/*
+		 * Space taken by all other metadata btrees are accounted
+		 * on-disk as used space.  We therefore only hide the space
+		 * that is reserved but not used by the trees.
+		 */
+		hidden_space = ask - used;
+		break;
+	default:
+		ASSERT(0);
+		return -EINVAL;
+	}
+	error = xfs_mod_fdblocks(mp, -(int64_t)hidden_space, true);
 	if (error) {
 		trace_xfs_ag_resv_init_error(pag->pag_mount, pag->pag_agno,
 				error, _RET_IP_);
@@ -216,7 +238,8 @@ __xfs_ag_resv_init(
 
 	resv = xfs_perag_resv(pag, type);
 	resv->ar_asked = ask;
-	resv->ar_reserved = resv->ar_orig_reserved = reserved;
+	resv->ar_orig_reserved = hidden_space;
+	resv->ar_reserved = ask - used;
 
 	trace_xfs_ag_resv_init(pag, type, ask);
 	return 0;


  parent reply	other threads:[~2018-06-21 18:31 UTC|newest]

Thread overview: 31+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-06-21 18:31 [PATCH 0/8] xfs-4.18: various fixes Darrick J. Wong
2018-06-21 18:31 ` [PATCH 1/8] xfs: allow empty transactions while frozen Darrick J. Wong
2018-06-22  6:18   ` Allison Henderson
2018-06-22  6:31   ` Christoph Hellwig
2018-06-21 18:31 ` Darrick J. Wong [this message]
2018-06-22  6:21   ` [PATCH 2/8] xfs: fix fdblocks accounting w/ RMAPBT per-AG reservation Allison Henderson
2018-06-21 18:31 ` [PATCH 3/8] xfs: don't trip over negative free space in xfs_reserve_blocks Darrick J. Wong
2018-06-22  6:31   ` Christoph Hellwig
2018-06-22  6:32   ` Allison Henderson
2018-06-21 18:31 ` [PATCH 4/8] xfs: don't allow insert-range to shift extents past the maximum offset Darrick J. Wong
2018-06-22  6:33   ` Allison Henderson
2018-06-22  6:39   ` Christoph Hellwig
2018-06-22  6:47     ` Darrick J. Wong
2018-06-22 21:21   ` [PATCH v2 " Darrick J. Wong
2018-06-23  6:47     ` Christoph Hellwig
2018-06-21 18:31 ` [PATCH 5/8] xfs: recheck reflink state after grabbing ILOCK_SHARED for a write Darrick J. Wong
2018-06-22  6:40   ` Christoph Hellwig
2018-06-21 18:31 ` [PATCH 6/8] xfs: fix uninitialized field in rtbitmap fsmap backend Darrick J. Wong
2018-06-22  6:37   ` Allison Henderson
2018-06-22  6:41   ` Christoph Hellwig
2018-06-22  6:47     ` Darrick J. Wong
2018-06-22 21:22   ` [PATCH v2 " Darrick J. Wong
2018-06-23  6:47     ` Christoph Hellwig
2018-06-21 18:31 ` [PATCH 7/8] xfs: fix off-by-one error in xfs_rtalloc_query_range Darrick J. Wong
2018-06-22  6:41   ` Christoph Hellwig
2018-06-22  6:41   ` Allison Henderson
2018-06-21 18:31 ` [PATCH 8/8] xfs: ensure post-EOF zeroing happens after zeroing part of a file Darrick J. Wong
2018-06-22  6:29   ` Christoph Hellwig
2018-06-22  6:49     ` Darrick J. Wong
2018-06-22  6:42   ` Allison Henderson
2018-06-22  6:21 ` [PATCH 9/8] xfs: check leaf attribute block freemap in verifier Darrick J. Wong

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=152960587647.26246.17630479835058118272.stgit@magnolia \
    --to=darrick.wong@oracle.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.