linux/fs/xfs/xfs_dquot.c
Dave Chinner b1c5ebb213 xfs: allocate log vector buffers outside CIL context lock
One of the problems we currently have with delayed logging is that
under serious memory pressure we can deadlock memory reclaim. THis
occurs when memory reclaim (such as run by kswapd) is reclaiming XFS
inodes and issues a log force to unpin inodes that are dirty in the
CIL.

The CIL is pushed, but this will only occur once it gets the CIL
context lock to ensure that all committing transactions are complete
and no new transactions start being committed to the CIL while the
push switches to a new context.

The deadlock occurs when the CIL context lock is held by a
committing process that is doing memory allocation for log vector
buffers, and that allocation is then blocked on memory reclaim
making progress. Memory reclaim, however, is blocked waiting for
a log force to make progress, and so we effectively deadlock at this
point.

To solve this problem, we have to move the CIL log vector buffer
allocation outside of the context lock so that memory reclaim can
always make progress when it needs to force the log. The problem
with doing this is that a CIL push can take place while we are
determining if we need to allocate a new log vector buffer for
an item and hence the current log vector may go away without
warning. That means we canot rely on the existing log vector being
present when we finally grab the context lock and so we must have a
replacement buffer ready to go at all times.

To ensure this, introduce a "shadow log vector" buffer that is
always guaranteed to be present when we gain the CIL context lock
and format the item. This shadow buffer may or may not be used
during the formatting, but if the log item does not have an existing
log vector buffer or that buffer is too small for the new
modifications, we swap it for the new shadow buffer and format
the modifications into that new log vector buffer.

The result of this is that for any object we modify more than once
in a given CIL checkpoint, we double the memory required
to track dirty regions in the log. For single modifications then
we consume the shadow log vectorwe allocate on commit, and that gets
consumed by the checkpoint. However, if we make multiple
modifications, then the second transaction commit will allocate a
shadow log vector and hence we will end up with double the memory
usage as only one of the log vectors is consumed by the CIL
checkpoint. The remaining shadow vector will be freed when th elog
item is freed.

This can probably be optimised in future - access to the shadow log
vector is serialised by the object lock (as opposited to the active
log vector, which is controlled by the CIL context lock) and so we
can probably free shadow log vector from some objects when the log
item is marked clean on removal from the AIL.

Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Brian Foster <bfoster@redhat.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>
2016-07-22 09:52:35 +10:00

1197 lines
30 KiB
C

/*
* Copyright (c) 2000-2003 Silicon Graphics, Inc.
* All Rights Reserved.
*
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU General Public License as
* published by the Free Software Foundation.
*
* This program is distributed in the hope that it would be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write the Free Software Foundation,
* Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
*/
#include "xfs.h"
#include "xfs_fs.h"
#include "xfs_format.h"
#include "xfs_log_format.h"
#include "xfs_shared.h"
#include "xfs_trans_resv.h"
#include "xfs_bit.h"
#include "xfs_mount.h"
#include "xfs_inode.h"
#include "xfs_bmap.h"
#include "xfs_bmap_util.h"
#include "xfs_alloc.h"
#include "xfs_quota.h"
#include "xfs_error.h"
#include "xfs_trans.h"
#include "xfs_buf_item.h"
#include "xfs_trans_space.h"
#include "xfs_trans_priv.h"
#include "xfs_qm.h"
#include "xfs_cksum.h"
#include "xfs_trace.h"
#include "xfs_log.h"
#include "xfs_bmap_btree.h"
/*
* Lock order:
*
* ip->i_lock
* qi->qi_tree_lock
* dquot->q_qlock (xfs_dqlock() and friends)
* dquot->q_flush (xfs_dqflock() and friends)
* qi->qi_lru_lock
*
* If two dquots need to be locked the order is user before group/project,
* otherwise by the lowest id first, see xfs_dqlock2.
*/
#ifdef DEBUG
xfs_buftarg_t *xfs_dqerror_target;
int xfs_do_dqerror;
int xfs_dqreq_num;
int xfs_dqerror_mod = 33;
#endif
struct kmem_zone *xfs_qm_dqtrxzone;
static struct kmem_zone *xfs_qm_dqzone;
static struct lock_class_key xfs_dquot_group_class;
static struct lock_class_key xfs_dquot_project_class;
/*
* This is called to free all the memory associated with a dquot
*/
void
xfs_qm_dqdestroy(
xfs_dquot_t *dqp)
{
ASSERT(list_empty(&dqp->q_lru));
kmem_free(dqp->q_logitem.qli_item.li_lv_shadow);
mutex_destroy(&dqp->q_qlock);
XFS_STATS_DEC(dqp->q_mount, xs_qm_dquot);
kmem_zone_free(xfs_qm_dqzone, dqp);
}
/*
* If default limits are in force, push them into the dquot now.
* We overwrite the dquot limits only if they are zero and this
* is not the root dquot.
*/
void
xfs_qm_adjust_dqlimits(
struct xfs_mount *mp,
struct xfs_dquot *dq)
{
struct xfs_quotainfo *q = mp->m_quotainfo;
struct xfs_disk_dquot *d = &dq->q_core;
struct xfs_def_quota *defq;
int prealloc = 0;
ASSERT(d->d_id);
defq = xfs_get_defquota(dq, q);
if (defq->bsoftlimit && !d->d_blk_softlimit) {
d->d_blk_softlimit = cpu_to_be64(defq->bsoftlimit);
prealloc = 1;
}
if (defq->bhardlimit && !d->d_blk_hardlimit) {
d->d_blk_hardlimit = cpu_to_be64(defq->bhardlimit);
prealloc = 1;
}
if (defq->isoftlimit && !d->d_ino_softlimit)
d->d_ino_softlimit = cpu_to_be64(defq->isoftlimit);
if (defq->ihardlimit && !d->d_ino_hardlimit)
d->d_ino_hardlimit = cpu_to_be64(defq->ihardlimit);
if (defq->rtbsoftlimit && !d->d_rtb_softlimit)
d->d_rtb_softlimit = cpu_to_be64(defq->rtbsoftlimit);
if (defq->rtbhardlimit && !d->d_rtb_hardlimit)
d->d_rtb_hardlimit = cpu_to_be64(defq->rtbhardlimit);
if (prealloc)
xfs_dquot_set_prealloc_limits(dq);
}
/*
* Check the limits and timers of a dquot and start or reset timers
* if necessary.
* This gets called even when quota enforcement is OFF, which makes our
* life a little less complicated. (We just don't reject any quota
* reservations in that case, when enforcement is off).
* We also return 0 as the values of the timers in Q_GETQUOTA calls, when
* enforcement's off.
* In contrast, warnings are a little different in that they don't
* 'automatically' get started when limits get exceeded. They do
* get reset to zero, however, when we find the count to be under
* the soft limit (they are only ever set non-zero via userspace).
*/
void
xfs_qm_adjust_dqtimers(
xfs_mount_t *mp,
xfs_disk_dquot_t *d)
{
ASSERT(d->d_id);
#ifdef DEBUG
if (d->d_blk_hardlimit)
ASSERT(be64_to_cpu(d->d_blk_softlimit) <=
be64_to_cpu(d->d_blk_hardlimit));
if (d->d_ino_hardlimit)
ASSERT(be64_to_cpu(d->d_ino_softlimit) <=
be64_to_cpu(d->d_ino_hardlimit));
if (d->d_rtb_hardlimit)
ASSERT(be64_to_cpu(d->d_rtb_softlimit) <=
be64_to_cpu(d->d_rtb_hardlimit));
#endif
if (!d->d_btimer) {
if ((d->d_blk_softlimit &&
(be64_to_cpu(d->d_bcount) >
be64_to_cpu(d->d_blk_softlimit))) ||
(d->d_blk_hardlimit &&
(be64_to_cpu(d->d_bcount) >
be64_to_cpu(d->d_blk_hardlimit)))) {
d->d_btimer = cpu_to_be32(get_seconds() +
mp->m_quotainfo->qi_btimelimit);
} else {
d->d_bwarns = 0;
}
} else {
if ((!d->d_blk_softlimit ||
(be64_to_cpu(d->d_bcount) <=
be64_to_cpu(d->d_blk_softlimit))) &&
(!d->d_blk_hardlimit ||
(be64_to_cpu(d->d_bcount) <=
be64_to_cpu(d->d_blk_hardlimit)))) {
d->d_btimer = 0;
}
}
if (!d->d_itimer) {
if ((d->d_ino_softlimit &&
(be64_to_cpu(d->d_icount) >
be64_to_cpu(d->d_ino_softlimit))) ||
(d->d_ino_hardlimit &&
(be64_to_cpu(d->d_icount) >
be64_to_cpu(d->d_ino_hardlimit)))) {
d->d_itimer = cpu_to_be32(get_seconds() +
mp->m_quotainfo->qi_itimelimit);
} else {
d->d_iwarns = 0;
}
} else {
if ((!d->d_ino_softlimit ||
(be64_to_cpu(d->d_icount) <=
be64_to_cpu(d->d_ino_softlimit))) &&
(!d->d_ino_hardlimit ||
(be64_to_cpu(d->d_icount) <=
be64_to_cpu(d->d_ino_hardlimit)))) {
d->d_itimer = 0;
}
}
if (!d->d_rtbtimer) {
if ((d->d_rtb_softlimit &&
(be64_to_cpu(d->d_rtbcount) >
be64_to_cpu(d->d_rtb_softlimit))) ||
(d->d_rtb_hardlimit &&
(be64_to_cpu(d->d_rtbcount) >
be64_to_cpu(d->d_rtb_hardlimit)))) {
d->d_rtbtimer = cpu_to_be32(get_seconds() +
mp->m_quotainfo->qi_rtbtimelimit);
} else {
d->d_rtbwarns = 0;
}
} else {
if ((!d->d_rtb_softlimit ||
(be64_to_cpu(d->d_rtbcount) <=
be64_to_cpu(d->d_rtb_softlimit))) &&
(!d->d_rtb_hardlimit ||
(be64_to_cpu(d->d_rtbcount) <=
be64_to_cpu(d->d_rtb_hardlimit)))) {
d->d_rtbtimer = 0;
}
}
}
/*
* initialize a buffer full of dquots and log the whole thing
*/
STATIC void
xfs_qm_init_dquot_blk(
xfs_trans_t *tp,
xfs_mount_t *mp,
xfs_dqid_t id,
uint type,
xfs_buf_t *bp)
{
struct xfs_quotainfo *q = mp->m_quotainfo;
xfs_dqblk_t *d;
xfs_dqid_t curid;
int i;
ASSERT(tp);
ASSERT(xfs_buf_islocked(bp));
d = bp->b_addr;
/*
* ID of the first dquot in the block - id's are zero based.
*/
curid = id - (id % q->qi_dqperchunk);
memset(d, 0, BBTOB(q->qi_dqchunklen));
for (i = 0; i < q->qi_dqperchunk; i++, d++, curid++) {
d->dd_diskdq.d_magic = cpu_to_be16(XFS_DQUOT_MAGIC);
d->dd_diskdq.d_version = XFS_DQUOT_VERSION;
d->dd_diskdq.d_id = cpu_to_be32(curid);
d->dd_diskdq.d_flags = type;
if (xfs_sb_version_hascrc(&mp->m_sb)) {
uuid_copy(&d->dd_uuid, &mp->m_sb.sb_meta_uuid);
xfs_update_cksum((char *)d, sizeof(struct xfs_dqblk),
XFS_DQUOT_CRC_OFF);
}
}
xfs_trans_dquot_buf(tp, bp,
(type & XFS_DQ_USER ? XFS_BLF_UDQUOT_BUF :
((type & XFS_DQ_PROJ) ? XFS_BLF_PDQUOT_BUF :
XFS_BLF_GDQUOT_BUF)));
xfs_trans_log_buf(tp, bp, 0, BBTOB(q->qi_dqchunklen) - 1);
}
/*
* Initialize the dynamic speculative preallocation thresholds. The lo/hi
* watermarks correspond to the soft and hard limits by default. If a soft limit
* is not specified, we use 95% of the hard limit.
*/
void
xfs_dquot_set_prealloc_limits(struct xfs_dquot *dqp)
{
__uint64_t space;
dqp->q_prealloc_hi_wmark = be64_to_cpu(dqp->q_core.d_blk_hardlimit);
dqp->q_prealloc_lo_wmark = be64_to_cpu(dqp->q_core.d_blk_softlimit);
if (!dqp->q_prealloc_lo_wmark) {
dqp->q_prealloc_lo_wmark = dqp->q_prealloc_hi_wmark;
do_div(dqp->q_prealloc_lo_wmark, 100);
dqp->q_prealloc_lo_wmark *= 95;
}
space = dqp->q_prealloc_hi_wmark;
do_div(space, 100);
dqp->q_low_space[XFS_QLOWSP_1_PCNT] = space;
dqp->q_low_space[XFS_QLOWSP_3_PCNT] = space * 3;
dqp->q_low_space[XFS_QLOWSP_5_PCNT] = space * 5;
}
/*
* Allocate a block and fill it with dquots.
* This is called when the bmapi finds a hole.
*/
STATIC int
xfs_qm_dqalloc(
xfs_trans_t **tpp,
xfs_mount_t *mp,
xfs_dquot_t *dqp,
xfs_inode_t *quotip,
xfs_fileoff_t offset_fsb,
xfs_buf_t **O_bpp)
{
xfs_fsblock_t firstblock;
xfs_bmap_free_t flist;
xfs_bmbt_irec_t map;
int nmaps, error;
xfs_buf_t *bp;
xfs_trans_t *tp = *tpp;
ASSERT(tp != NULL);
trace_xfs_dqalloc(dqp);
/*
* Initialize the bmap freelist prior to calling bmapi code.
*/
xfs_bmap_init(&flist, &firstblock);
xfs_ilock(quotip, XFS_ILOCK_EXCL);
/*
* Return if this type of quotas is turned off while we didn't
* have an inode lock
*/
if (!xfs_this_quota_on(dqp->q_mount, dqp->dq_flags)) {
xfs_iunlock(quotip, XFS_ILOCK_EXCL);
return -ESRCH;
}
xfs_trans_ijoin(tp, quotip, XFS_ILOCK_EXCL);
nmaps = 1;
error = xfs_bmapi_write(tp, quotip, offset_fsb,
XFS_DQUOT_CLUSTER_SIZE_FSB, XFS_BMAPI_METADATA,
&firstblock, XFS_QM_DQALLOC_SPACE_RES(mp),
&map, &nmaps, &flist);
if (error)
goto error0;
ASSERT(map.br_blockcount == XFS_DQUOT_CLUSTER_SIZE_FSB);
ASSERT(nmaps == 1);
ASSERT((map.br_startblock != DELAYSTARTBLOCK) &&
(map.br_startblock != HOLESTARTBLOCK));
/*
* Keep track of the blkno to save a lookup later
*/
dqp->q_blkno = XFS_FSB_TO_DADDR(mp, map.br_startblock);
/* now we can just get the buffer (there's nothing to read yet) */
bp = xfs_trans_get_buf(tp, mp->m_ddev_targp,
dqp->q_blkno,
mp->m_quotainfo->qi_dqchunklen,
0);
if (!bp) {
error = -ENOMEM;
goto error1;
}
bp->b_ops = &xfs_dquot_buf_ops;
/*
* Make a chunk of dquots out of this buffer and log
* the entire thing.
*/
xfs_qm_init_dquot_blk(tp, mp, be32_to_cpu(dqp->q_core.d_id),
dqp->dq_flags & XFS_DQ_ALLTYPES, bp);
/*
* xfs_bmap_finish() may commit the current transaction and
* start a second transaction if the freelist is not empty.
*
* Since we still want to modify this buffer, we need to
* ensure that the buffer is not released on commit of
* the first transaction and ensure the buffer is added to the
* second transaction.
*
* If there is only one transaction then don't stop the buffer
* from being released when it commits later on.
*/
xfs_trans_bhold(tp, bp);
error = xfs_bmap_finish(tpp, &flist, NULL);
if (error)
goto error1;
/* Transaction was committed? */
if (*tpp != tp) {
tp = *tpp;
xfs_trans_bjoin(tp, bp);
} else {
xfs_trans_bhold_release(tp, bp);
}
*O_bpp = bp;
return 0;
error1:
xfs_bmap_cancel(&flist);
error0:
xfs_iunlock(quotip, XFS_ILOCK_EXCL);
return error;
}
STATIC int
xfs_qm_dqrepair(
struct xfs_mount *mp,
struct xfs_trans *tp,
struct xfs_dquot *dqp,
xfs_dqid_t firstid,
struct xfs_buf **bpp)
{
int error;
struct xfs_disk_dquot *ddq;
struct xfs_dqblk *d;
int i;
/*
* Read the buffer without verification so we get the corrupted
* buffer returned to us. make sure we verify it on write, though.
*/
error = xfs_trans_read_buf(mp, tp, mp->m_ddev_targp, dqp->q_blkno,
mp->m_quotainfo->qi_dqchunklen,
0, bpp, NULL);
if (error) {
ASSERT(*bpp == NULL);
return error;
}
(*bpp)->b_ops = &xfs_dquot_buf_ops;
ASSERT(xfs_buf_islocked(*bpp));
d = (struct xfs_dqblk *)(*bpp)->b_addr;
/* Do the actual repair of dquots in this buffer */
for (i = 0; i < mp->m_quotainfo->qi_dqperchunk; i++) {
ddq = &d[i].dd_diskdq;
error = xfs_dqcheck(mp, ddq, firstid + i,
dqp->dq_flags & XFS_DQ_ALLTYPES,
XFS_QMOPT_DQREPAIR, "xfs_qm_dqrepair");
if (error) {
/* repair failed, we're screwed */
xfs_trans_brelse(tp, *bpp);
return -EIO;
}
}
return 0;
}
/*
* Maps a dquot to the buffer containing its on-disk version.
* This returns a ptr to the buffer containing the on-disk dquot
* in the bpp param, and a ptr to the on-disk dquot within that buffer
*/
STATIC int
xfs_qm_dqtobp(
xfs_trans_t **tpp,
xfs_dquot_t *dqp,
xfs_disk_dquot_t **O_ddpp,
xfs_buf_t **O_bpp,
uint flags)
{
struct xfs_bmbt_irec map;
int nmaps = 1, error;
struct xfs_buf *bp;
struct xfs_inode *quotip;
struct xfs_mount *mp = dqp->q_mount;
xfs_dqid_t id = be32_to_cpu(dqp->q_core.d_id);
struct xfs_trans *tp = (tpp ? *tpp : NULL);
uint lock_mode;
quotip = xfs_quota_inode(dqp->q_mount, dqp->dq_flags);
dqp->q_fileoffset = (xfs_fileoff_t)id / mp->m_quotainfo->qi_dqperchunk;
lock_mode = xfs_ilock_data_map_shared(quotip);
if (!xfs_this_quota_on(dqp->q_mount, dqp->dq_flags)) {
/*
* Return if this type of quotas is turned off while we
* didn't have the quota inode lock.
*/
xfs_iunlock(quotip, lock_mode);
return -ESRCH;
}
/*
* Find the block map; no allocations yet
*/
error = xfs_bmapi_read(quotip, dqp->q_fileoffset,
XFS_DQUOT_CLUSTER_SIZE_FSB, &map, &nmaps, 0);
xfs_iunlock(quotip, lock_mode);
if (error)
return error;
ASSERT(nmaps == 1);
ASSERT(map.br_blockcount == 1);
/*
* Offset of dquot in the (fixed sized) dquot chunk.
*/
dqp->q_bufoffset = (id % mp->m_quotainfo->qi_dqperchunk) *
sizeof(xfs_dqblk_t);
ASSERT(map.br_startblock != DELAYSTARTBLOCK);
if (map.br_startblock == HOLESTARTBLOCK) {
/*
* We don't allocate unless we're asked to
*/
if (!(flags & XFS_QMOPT_DQALLOC))
return -ENOENT;
ASSERT(tp);
error = xfs_qm_dqalloc(tpp, mp, dqp, quotip,
dqp->q_fileoffset, &bp);
if (error)
return error;
tp = *tpp;
} else {
trace_xfs_dqtobp_read(dqp);
/*
* store the blkno etc so that we don't have to do the
* mapping all the time
*/
dqp->q_blkno = XFS_FSB_TO_DADDR(mp, map.br_startblock);
error = xfs_trans_read_buf(mp, tp, mp->m_ddev_targp,
dqp->q_blkno,
mp->m_quotainfo->qi_dqchunklen,
0, &bp, &xfs_dquot_buf_ops);
if (error == -EFSCORRUPTED && (flags & XFS_QMOPT_DQREPAIR)) {
xfs_dqid_t firstid = (xfs_dqid_t)map.br_startoff *
mp->m_quotainfo->qi_dqperchunk;
ASSERT(bp == NULL);
error = xfs_qm_dqrepair(mp, tp, dqp, firstid, &bp);
}
if (error) {
ASSERT(bp == NULL);
return error;
}
}
ASSERT(xfs_buf_islocked(bp));
*O_bpp = bp;
*O_ddpp = bp->b_addr + dqp->q_bufoffset;
return 0;
}
/*
* Read in the ondisk dquot using dqtobp() then copy it to an incore version,
* and release the buffer immediately.
*
* If XFS_QMOPT_DQALLOC is set, allocate a dquot on disk if it needed.
*/
int
xfs_qm_dqread(
struct xfs_mount *mp,
xfs_dqid_t id,
uint type,
uint flags,
struct xfs_dquot **O_dqpp)
{
struct xfs_dquot *dqp;
struct xfs_disk_dquot *ddqp;
struct xfs_buf *bp;
struct xfs_trans *tp = NULL;
int error;
dqp = kmem_zone_zalloc(xfs_qm_dqzone, KM_SLEEP);
dqp->dq_flags = type;
dqp->q_core.d_id = cpu_to_be32(id);
dqp->q_mount = mp;
INIT_LIST_HEAD(&dqp->q_lru);
mutex_init(&dqp->q_qlock);
init_waitqueue_head(&dqp->q_pinwait);
/*
* Because we want to use a counting completion, complete
* the flush completion once to allow a single access to
* the flush completion without blocking.
*/
init_completion(&dqp->q_flush);
complete(&dqp->q_flush);
/*
* Make sure group quotas have a different lock class than user
* quotas.
*/
switch (type) {
case XFS_DQ_USER:
/* uses the default lock class */
break;
case XFS_DQ_GROUP:
lockdep_set_class(&dqp->q_qlock, &xfs_dquot_group_class);
break;
case XFS_DQ_PROJ:
lockdep_set_class(&dqp->q_qlock, &xfs_dquot_project_class);
break;
default:
ASSERT(0);
break;
}
XFS_STATS_INC(mp, xs_qm_dquot);
trace_xfs_dqread(dqp);
if (flags & XFS_QMOPT_DQALLOC) {
error = xfs_trans_alloc(mp, &M_RES(mp)->tr_qm_dqalloc,
XFS_QM_DQALLOC_SPACE_RES(mp), 0, 0, &tp);
if (error)
goto error0;
}
/*
* get a pointer to the on-disk dquot and the buffer containing it
* dqp already knows its own type (GROUP/USER).
*/
error = xfs_qm_dqtobp(&tp, dqp, &ddqp, &bp, flags);
if (error) {
/*
* This can happen if quotas got turned off (ESRCH),
* or if the dquot didn't exist on disk and we ask to
* allocate (ENOENT).
*/
trace_xfs_dqread_fail(dqp);
goto error1;
}
/* copy everything from disk dquot to the incore dquot */
memcpy(&dqp->q_core, ddqp, sizeof(xfs_disk_dquot_t));
xfs_qm_dquot_logitem_init(dqp);
/*
* Reservation counters are defined as reservation plus current usage
* to avoid having to add every time.
*/
dqp->q_res_bcount = be64_to_cpu(ddqp->d_bcount);
dqp->q_res_icount = be64_to_cpu(ddqp->d_icount);
dqp->q_res_rtbcount = be64_to_cpu(ddqp->d_rtbcount);
/* initialize the dquot speculative prealloc thresholds */
xfs_dquot_set_prealloc_limits(dqp);
/* Mark the buf so that this will stay incore a little longer */
xfs_buf_set_ref(bp, XFS_DQUOT_REF);
/*
* We got the buffer with a xfs_trans_read_buf() (in dqtobp())
* So we need to release with xfs_trans_brelse().
* The strategy here is identical to that of inodes; we lock
* the dquot in xfs_qm_dqget() before making it accessible to
* others. This is because dquots, like inodes, need a good level of
* concurrency, and we don't want to take locks on the entire buffers
* for dquot accesses.
* Note also that the dquot buffer may even be dirty at this point, if
* this particular dquot was repaired. We still aren't afraid to
* brelse it because we have the changes incore.
*/
ASSERT(xfs_buf_islocked(bp));
xfs_trans_brelse(tp, bp);
if (tp) {
error = xfs_trans_commit(tp);
if (error)
goto error0;
}
*O_dqpp = dqp;
return error;
error1:
if (tp)
xfs_trans_cancel(tp);
error0:
xfs_qm_dqdestroy(dqp);
*O_dqpp = NULL;
return error;
}
/*
* Advance to the next id in the current chunk, or if at the
* end of the chunk, skip ahead to first id in next allocated chunk
* using the SEEK_DATA interface.
*/
static int
xfs_dq_get_next_id(
xfs_mount_t *mp,
uint type,
xfs_dqid_t *id,
loff_t eof)
{
struct xfs_inode *quotip;
xfs_fsblock_t start;
loff_t offset;
uint lock;
xfs_dqid_t next_id;
int error = 0;
/* Simple advance */
next_id = *id + 1;
/* If new ID is within the current chunk, advancing it sufficed */
if (next_id % mp->m_quotainfo->qi_dqperchunk) {
*id = next_id;
return 0;
}
/* Nope, next_id is now past the current chunk, so find the next one */
start = (xfs_fsblock_t)next_id / mp->m_quotainfo->qi_dqperchunk;
quotip = xfs_quota_inode(mp, type);
lock = xfs_ilock_data_map_shared(quotip);
offset = __xfs_seek_hole_data(VFS_I(quotip), XFS_FSB_TO_B(mp, start),
eof, SEEK_DATA);
if (offset < 0)
error = offset;
xfs_iunlock(quotip, lock);
/* -ENXIO is essentially "no more data" */
if (error)
return (error == -ENXIO ? -ENOENT: error);
/* Convert next data offset back to a quota id */
*id = XFS_B_TO_FSB(mp, offset) * mp->m_quotainfo->qi_dqperchunk;
return 0;
}
/*
* Given the file system, inode OR id, and type (UDQUOT/GDQUOT), return a
* a locked dquot, doing an allocation (if requested) as needed.
* When both an inode and an id are given, the inode's id takes precedence.
* That is, if the id changes while we don't hold the ilock inside this
* function, the new dquot is returned, not necessarily the one requested
* in the id argument.
*/
int
xfs_qm_dqget(
xfs_mount_t *mp,
xfs_inode_t *ip, /* locked inode (optional) */
xfs_dqid_t id, /* uid/projid/gid depending on type */
uint type, /* XFS_DQ_USER/XFS_DQ_PROJ/XFS_DQ_GROUP */
uint flags, /* DQALLOC, DQSUSER, DQREPAIR, DOWARN */
xfs_dquot_t **O_dqpp) /* OUT : locked incore dquot */
{
struct xfs_quotainfo *qi = mp->m_quotainfo;
struct radix_tree_root *tree = xfs_dquot_tree(qi, type);
struct xfs_dquot *dqp;
loff_t eof = 0;
int error;
ASSERT(XFS_IS_QUOTA_RUNNING(mp));
if ((! XFS_IS_UQUOTA_ON(mp) && type == XFS_DQ_USER) ||
(! XFS_IS_PQUOTA_ON(mp) && type == XFS_DQ_PROJ) ||
(! XFS_IS_GQUOTA_ON(mp) && type == XFS_DQ_GROUP)) {
return -ESRCH;
}
#ifdef DEBUG
if (xfs_do_dqerror) {
if ((xfs_dqerror_target == mp->m_ddev_targp) &&
(xfs_dqreq_num++ % xfs_dqerror_mod) == 0) {
xfs_debug(mp, "Returning error in dqget");
return -EIO;
}
}
ASSERT(type == XFS_DQ_USER ||
type == XFS_DQ_PROJ ||
type == XFS_DQ_GROUP);
if (ip) {
ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL));
ASSERT(xfs_inode_dquot(ip, type) == NULL);
}
#endif
/* Get the end of the quota file if we need it */
if (flags & XFS_QMOPT_DQNEXT) {
struct xfs_inode *quotip;
xfs_fileoff_t last;
uint lock_mode;
quotip = xfs_quota_inode(mp, type);
lock_mode = xfs_ilock_data_map_shared(quotip);
error = xfs_bmap_last_offset(quotip, &last, XFS_DATA_FORK);
xfs_iunlock(quotip, lock_mode);
if (error)
return error;
eof = XFS_FSB_TO_B(mp, last);
}
restart:
mutex_lock(&qi->qi_tree_lock);
dqp = radix_tree_lookup(tree, id);
if (dqp) {
xfs_dqlock(dqp);
if (dqp->dq_flags & XFS_DQ_FREEING) {
xfs_dqunlock(dqp);
mutex_unlock(&qi->qi_tree_lock);
trace_xfs_dqget_freeing(dqp);
delay(1);
goto restart;
}
/* uninit / unused quota found in radix tree, keep looking */
if (flags & XFS_QMOPT_DQNEXT) {
if (XFS_IS_DQUOT_UNINITIALIZED(dqp)) {
xfs_dqunlock(dqp);
mutex_unlock(&qi->qi_tree_lock);
error = xfs_dq_get_next_id(mp, type, &id, eof);
if (error)
return error;
goto restart;
}
}
dqp->q_nrefs++;
mutex_unlock(&qi->qi_tree_lock);
trace_xfs_dqget_hit(dqp);
XFS_STATS_INC(mp, xs_qm_dqcachehits);
*O_dqpp = dqp;
return 0;
}
mutex_unlock(&qi->qi_tree_lock);
XFS_STATS_INC(mp, xs_qm_dqcachemisses);
/*
* Dquot cache miss. We don't want to keep the inode lock across
* a (potential) disk read. Also we don't want to deal with the lock
* ordering between quotainode and this inode. OTOH, dropping the inode
* lock here means dealing with a chown that can happen before
* we re-acquire the lock.
*/
if (ip)
xfs_iunlock(ip, XFS_ILOCK_EXCL);
error = xfs_qm_dqread(mp, id, type, flags, &dqp);
if (ip)
xfs_ilock(ip, XFS_ILOCK_EXCL);
/* If we are asked to find next active id, keep looking */
if (error == -ENOENT && (flags & XFS_QMOPT_DQNEXT)) {
error = xfs_dq_get_next_id(mp, type, &id, eof);
if (!error)
goto restart;
}
if (error)
return error;
if (ip) {
/*
* A dquot could be attached to this inode by now, since
* we had dropped the ilock.
*/
if (xfs_this_quota_on(mp, type)) {
struct xfs_dquot *dqp1;
dqp1 = xfs_inode_dquot(ip, type);
if (dqp1) {
xfs_qm_dqdestroy(dqp);
dqp = dqp1;
xfs_dqlock(dqp);
goto dqret;
}
} else {
/* inode stays locked on return */
xfs_qm_dqdestroy(dqp);
return -ESRCH;
}
}
mutex_lock(&qi->qi_tree_lock);
error = radix_tree_insert(tree, id, dqp);
if (unlikely(error)) {
WARN_ON(error != -EEXIST);
/*
* Duplicate found. Just throw away the new dquot and start
* over.
*/
mutex_unlock(&qi->qi_tree_lock);
trace_xfs_dqget_dup(dqp);
xfs_qm_dqdestroy(dqp);
XFS_STATS_INC(mp, xs_qm_dquot_dups);
goto restart;
}
/*
* We return a locked dquot to the caller, with a reference taken
*/
xfs_dqlock(dqp);
dqp->q_nrefs = 1;
qi->qi_dquots++;
mutex_unlock(&qi->qi_tree_lock);
/* If we are asked to find next active id, keep looking */
if (flags & XFS_QMOPT_DQNEXT) {
if (XFS_IS_DQUOT_UNINITIALIZED(dqp)) {
xfs_qm_dqput(dqp);
error = xfs_dq_get_next_id(mp, type, &id, eof);
if (error)
return error;
goto restart;
}
}
dqret:
ASSERT((ip == NULL) || xfs_isilocked(ip, XFS_ILOCK_EXCL));
trace_xfs_dqget_miss(dqp);
*O_dqpp = dqp;
return 0;
}
/*
* Release a reference to the dquot (decrement ref-count) and unlock it.
*
* If there is a group quota attached to this dquot, carefully release that
* too without tripping over deadlocks'n'stuff.
*/
void
xfs_qm_dqput(
struct xfs_dquot *dqp)
{
ASSERT(dqp->q_nrefs > 0);
ASSERT(XFS_DQ_IS_LOCKED(dqp));
trace_xfs_dqput(dqp);
if (--dqp->q_nrefs == 0) {
struct xfs_quotainfo *qi = dqp->q_mount->m_quotainfo;
trace_xfs_dqput_free(dqp);
if (list_lru_add(&qi->qi_lru, &dqp->q_lru))
XFS_STATS_INC(dqp->q_mount, xs_qm_dquot_unused);
}
xfs_dqunlock(dqp);
}
/*
* Release a dquot. Flush it if dirty, then dqput() it.
* dquot must not be locked.
*/
void
xfs_qm_dqrele(
xfs_dquot_t *dqp)
{
if (!dqp)
return;
trace_xfs_dqrele(dqp);
xfs_dqlock(dqp);
/*
* We don't care to flush it if the dquot is dirty here.
* That will create stutters that we want to avoid.
* Instead we do a delayed write when we try to reclaim
* a dirty dquot. Also xfs_sync will take part of the burden...
*/
xfs_qm_dqput(dqp);
}
/*
* This is the dquot flushing I/O completion routine. It is called
* from interrupt level when the buffer containing the dquot is
* flushed to disk. It is responsible for removing the dquot logitem
* from the AIL if it has not been re-logged, and unlocking the dquot's
* flush lock. This behavior is very similar to that of inodes..
*/
STATIC void
xfs_qm_dqflush_done(
struct xfs_buf *bp,
struct xfs_log_item *lip)
{
xfs_dq_logitem_t *qip = (struct xfs_dq_logitem *)lip;
xfs_dquot_t *dqp = qip->qli_dquot;
struct xfs_ail *ailp = lip->li_ailp;
/*
* We only want to pull the item from the AIL if its
* location in the log has not changed since we started the flush.
* Thus, we only bother if the dquot's lsn has
* not changed. First we check the lsn outside the lock
* since it's cheaper, and then we recheck while
* holding the lock before removing the dquot from the AIL.
*/
if ((lip->li_flags & XFS_LI_IN_AIL) &&
lip->li_lsn == qip->qli_flush_lsn) {
/* xfs_trans_ail_delete() drops the AIL lock. */
spin_lock(&ailp->xa_lock);
if (lip->li_lsn == qip->qli_flush_lsn)
xfs_trans_ail_delete(ailp, lip, SHUTDOWN_CORRUPT_INCORE);
else
spin_unlock(&ailp->xa_lock);
}
/*
* Release the dq's flush lock since we're done with it.
*/
xfs_dqfunlock(dqp);
}
/*
* Write a modified dquot to disk.
* The dquot must be locked and the flush lock too taken by caller.
* The flush lock will not be unlocked until the dquot reaches the disk,
* but the dquot is free to be unlocked and modified by the caller
* in the interim. Dquot is still locked on return. This behavior is
* identical to that of inodes.
*/
int
xfs_qm_dqflush(
struct xfs_dquot *dqp,
struct xfs_buf **bpp)
{
struct xfs_mount *mp = dqp->q_mount;
struct xfs_buf *bp;
struct xfs_disk_dquot *ddqp;
int error;
ASSERT(XFS_DQ_IS_LOCKED(dqp));
ASSERT(!completion_done(&dqp->q_flush));
trace_xfs_dqflush(dqp);
*bpp = NULL;
xfs_qm_dqunpin_wait(dqp);
/*
* This may have been unpinned because the filesystem is shutting
* down forcibly. If that's the case we must not write this dquot
* to disk, because the log record didn't make it to disk.
*
* We also have to remove the log item from the AIL in this case,
* as we wait for an emptry AIL as part of the unmount process.
*/
if (XFS_FORCED_SHUTDOWN(mp)) {
struct xfs_log_item *lip = &dqp->q_logitem.qli_item;
dqp->dq_flags &= ~XFS_DQ_DIRTY;
xfs_trans_ail_remove(lip, SHUTDOWN_CORRUPT_INCORE);
error = -EIO;
goto out_unlock;
}
/*
* Get the buffer containing the on-disk dquot
*/
error = xfs_trans_read_buf(mp, NULL, mp->m_ddev_targp, dqp->q_blkno,
mp->m_quotainfo->qi_dqchunklen, 0, &bp,
&xfs_dquot_buf_ops);
if (error)
goto out_unlock;
/*
* Calculate the location of the dquot inside the buffer.
*/
ddqp = bp->b_addr + dqp->q_bufoffset;
/*
* A simple sanity check in case we got a corrupted dquot..
*/
error = xfs_dqcheck(mp, &dqp->q_core, be32_to_cpu(ddqp->d_id), 0,
XFS_QMOPT_DOWARN, "dqflush (incore copy)");
if (error) {
xfs_buf_relse(bp);
xfs_dqfunlock(dqp);
xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE);
return -EIO;
}
/* This is the only portion of data that needs to persist */
memcpy(ddqp, &dqp->q_core, sizeof(xfs_disk_dquot_t));
/*
* Clear the dirty field and remember the flush lsn for later use.
*/
dqp->dq_flags &= ~XFS_DQ_DIRTY;
xfs_trans_ail_copy_lsn(mp->m_ail, &dqp->q_logitem.qli_flush_lsn,
&dqp->q_logitem.qli_item.li_lsn);
/*
* copy the lsn into the on-disk dquot now while we have the in memory
* dquot here. This can't be done later in the write verifier as we
* can't get access to the log item at that point in time.
*
* We also calculate the CRC here so that the on-disk dquot in the
* buffer always has a valid CRC. This ensures there is no possibility
* of a dquot without an up-to-date CRC getting to disk.
*/
if (xfs_sb_version_hascrc(&mp->m_sb)) {
struct xfs_dqblk *dqb = (struct xfs_dqblk *)ddqp;
dqb->dd_lsn = cpu_to_be64(dqp->q_logitem.qli_item.li_lsn);
xfs_update_cksum((char *)dqb, sizeof(struct xfs_dqblk),
XFS_DQUOT_CRC_OFF);
}
/*
* Attach an iodone routine so that we can remove this dquot from the
* AIL and release the flush lock once the dquot is synced to disk.
*/
xfs_buf_attach_iodone(bp, xfs_qm_dqflush_done,
&dqp->q_logitem.qli_item);
/*
* If the buffer is pinned then push on the log so we won't
* get stuck waiting in the write for too long.
*/
if (xfs_buf_ispinned(bp)) {
trace_xfs_dqflush_force(dqp);
xfs_log_force(mp, 0);
}
trace_xfs_dqflush_done(dqp);
*bpp = bp;
return 0;
out_unlock:
xfs_dqfunlock(dqp);
return -EIO;
}
/*
* Lock two xfs_dquot structures.
*
* To avoid deadlocks we always lock the quota structure with
* the lowerd id first.
*/
void
xfs_dqlock2(
xfs_dquot_t *d1,
xfs_dquot_t *d2)
{
if (d1 && d2) {
ASSERT(d1 != d2);
if (be32_to_cpu(d1->q_core.d_id) >
be32_to_cpu(d2->q_core.d_id)) {
mutex_lock(&d2->q_qlock);
mutex_lock_nested(&d1->q_qlock, XFS_QLOCK_NESTED);
} else {
mutex_lock(&d1->q_qlock);
mutex_lock_nested(&d2->q_qlock, XFS_QLOCK_NESTED);
}
} else if (d1) {
mutex_lock(&d1->q_qlock);
} else if (d2) {
mutex_lock(&d2->q_qlock);
}
}
int __init
xfs_qm_init(void)
{
xfs_qm_dqzone =
kmem_zone_init(sizeof(struct xfs_dquot), "xfs_dquot");
if (!xfs_qm_dqzone)
goto out;
xfs_qm_dqtrxzone =
kmem_zone_init(sizeof(struct xfs_dquot_acct), "xfs_dqtrx");
if (!xfs_qm_dqtrxzone)
goto out_free_dqzone;
return 0;
out_free_dqzone:
kmem_zone_destroy(xfs_qm_dqzone);
out:
return -ENOMEM;
}
void
xfs_qm_exit(void)
{
kmem_zone_destroy(xfs_qm_dqtrxzone);
kmem_zone_destroy(xfs_qm_dqzone);
}