From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from relay.sgi.com (relay2.corp.sgi.com [137.38.102.29]) by oss.sgi.com (Postfix) with ESMTP id 096D229E05 for ; Mon, 29 Feb 2016 08:29:37 -0600 (CST) Received: from cuda.sgi.com (cuda3.sgi.com [192.48.176.15]) by relay2.corp.sgi.com (Postfix) with ESMTP id CFD06304043 for ; Mon, 29 Feb 2016 06:29:36 -0800 (PST) Received: from mx1.redhat.com (mx1.redhat.com [209.132.183.28]) by cuda.sgi.com with ESMTP id VYFz5MDD9AlNtx6i (version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256 verify=NO) for ; Mon, 29 Feb 2016 06:29:32 -0800 (PST) Received: from int-mx14.intmail.prod.int.phx2.redhat.com (int-mx14.intmail.prod.int.phx2.redhat.com [10.5.11.27]) by mx1.redhat.com (Postfix) with ESMTPS id 1AE91804E2 for ; Mon, 29 Feb 2016 14:29:32 +0000 (UTC) Received: from bfoster.bfoster (dhcp-41-22.bos.redhat.com [10.18.41.22]) by int-mx14.intmail.prod.int.phx2.redhat.com (8.14.4/8.14.4) with ESMTP id u1TETVdK001951 for ; Mon, 29 Feb 2016 09:29:31 -0500 From: Brian Foster Subject: [PATCH 2/3] xfs: update icsb freeblocks counter after extent deletion Date: Mon, 29 Feb 2016 09:29:29 -0500 Message-Id: <1456756170-51248-3-git-send-email-bfoster@redhat.com> In-Reply-To: <1456756170-51248-1-git-send-email-bfoster@redhat.com> References: <1456756170-51248-1-git-send-email-bfoster@redhat.com> List-Id: XFS Filesystem from SGI List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Errors-To: xfs-bounces@oss.sgi.com Sender: xfs-bounces@oss.sgi.com To: xfs@oss.sgi.com xfs_bunmapi() currently updates the icsb fdblocks counter, unreserves quota, etc. before the extent is deleted by xfs_bmap_del_extent(). The function has problems dividing up the indirect reserved blocks for scenarios where a single delalloc extent is split in two. Particularly, there aren't always enough blocks reserved for multiple extents in a single extent reservation. The solution to this problem is to allow the extent removal code to steal from the deleted extent to meet indirect reservation requirements. Move the block of code in xfs_bmapi() that updates the icsb fdblocks counter to after the call to xfs_bmap_del_extent() to allow the codepath to update the extent record before the free blocks are accounted. Also, reshuffle the code slightly so the delalloc accounting occurs near the xfs_bmap_del_extent() call to provide context for the comments. Signed-off-by: Brian Foster --- fs/xfs/libxfs/xfs_bmap.c | 58 ++++++++++++++++++++++++++++-------------------- 1 file changed, 34 insertions(+), 24 deletions(-) diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index 6a05166..7180104 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -5296,31 +5296,7 @@ xfs_bunmapi( goto nodelete; } } - if (wasdel) { - ASSERT(startblockval(del.br_startblock) > 0); - /* Update realtime/data freespace, unreserve quota */ - if (isrt) { - xfs_filblks_t rtexts; - rtexts = XFS_FSB_TO_B(mp, del.br_blockcount); - do_div(rtexts, mp->m_sb.sb_rextsize); - xfs_mod_frextents(mp, (int64_t)rtexts); - (void)xfs_trans_reserve_quota_nblks(NULL, - ip, -((long)del.br_blockcount), 0, - XFS_QMOPT_RES_RTBLKS); - } else { - xfs_mod_fdblocks(mp, (int64_t)del.br_blockcount, - false); - (void)xfs_trans_reserve_quota_nblks(NULL, - ip, -((long)del.br_blockcount), 0, - XFS_QMOPT_RES_REGBLKS); - } - ip->i_delayed_blks -= del.br_blockcount; - if (cur) - cur->bc_private.b.flags |= - XFS_BTCUR_BPRV_WASDEL; - } else if (cur) - cur->bc_private.b.flags &= ~XFS_BTCUR_BPRV_WASDEL; /* * If it's the case where the directory code is running * with no block reservation, and the deleted block is in @@ -5342,11 +5318,45 @@ xfs_bunmapi( error = -ENOSPC; goto error0; } + + /* + * Unreserve quota and update realtime free space, if + * appropriate. If delayed allocation, update the inode delalloc + * counter now and wait to update the sb counters as + * xfs_bmap_del_extent() might need to borrow some blocks. + */ + if (wasdel) { + ASSERT(startblockval(del.br_startblock) > 0); + if (isrt) { + xfs_filblks_t rtexts; + + rtexts = XFS_FSB_TO_B(mp, del.br_blockcount); + do_div(rtexts, mp->m_sb.sb_rextsize); + xfs_mod_frextents(mp, (int64_t)rtexts); + (void)xfs_trans_reserve_quota_nblks(NULL, + ip, -((long)del.br_blockcount), 0, + XFS_QMOPT_RES_RTBLKS); + } else { + (void)xfs_trans_reserve_quota_nblks(NULL, + ip, -((long)del.br_blockcount), 0, + XFS_QMOPT_RES_REGBLKS); + } + ip->i_delayed_blks -= del.br_blockcount; + if (cur) + cur->bc_private.b.flags |= + XFS_BTCUR_BPRV_WASDEL; + } else if (cur) + cur->bc_private.b.flags &= ~XFS_BTCUR_BPRV_WASDEL; + error = xfs_bmap_del_extent(ip, tp, &lastx, flist, cur, &del, &tmp_logflags, whichfork); logflags |= tmp_logflags; if (error) goto error0; + + if (!isrt && wasdel) + xfs_mod_fdblocks(mp, (int64_t)del.br_blockcount, false); + bno = del.br_startoff - 1; nodelete: /* -- 2.4.3 _______________________________________________ xfs mailing list xfs@oss.sgi.com http://oss.sgi.com/mailman/listinfo/xfs