OpenCloudOS-Kernel/fs/xfs/xfs_iomap.c

1273 lines
35 KiB
C
Raw Normal View History

// SPDX-License-Identifier: GPL-2.0
/*
* Copyright (c) 2000-2006 Silicon Graphics, Inc.
* Copyright (c) 2016-2018 Christoph Hellwig.
* All Rights Reserved.
*/
#include <linux/iomap.h>
#include "xfs.h"
#include "xfs_fs.h"
#include "xfs_shared.h"
#include "xfs_format.h"
#include "xfs_log_format.h"
#include "xfs_trans_resv.h"
#include "xfs_mount.h"
#include "xfs_defer.h"
#include "xfs_inode.h"
#include "xfs_btree.h"
#include "xfs_bmap_btree.h"
#include "xfs_bmap.h"
#include "xfs_bmap_util.h"
#include "xfs_errortag.h"
#include "xfs_error.h"
#include "xfs_trans.h"
#include "xfs_trans_space.h"
#include "xfs_inode_item.h"
#include "xfs_iomap.h"
xfs: event tracing support Convert the old xfs tracing support that could only be used with the out of tree kdb and xfsidbg patches to use the generic event tracer. To use it make sure CONFIG_EVENT_TRACING is enabled and then enable all xfs trace channels by: echo 1 > /sys/kernel/debug/tracing/events/xfs/enable or alternatively enable single events by just doing the same in one event subdirectory, e.g. echo 1 > /sys/kernel/debug/tracing/events/xfs/xfs_ihold/enable or set more complex filters, etc. In Documentation/trace/events.txt all this is desctribed in more detail. To reads the events do a cat /sys/kernel/debug/tracing/trace Compared to the last posting this patch converts the tracing mostly to the one tracepoint per callsite model that other users of the new tracing facility also employ. This allows a very fine-grained control of the tracing, a cleaner output of the traces and also enables the perf tool to use each tracepoint as a virtual performance counter, allowing us to e.g. count how often certain workloads git various spots in XFS. Take a look at http://lwn.net/Articles/346470/ for some examples. Also the btree tracing isn't included at all yet, as it will require additional core tracing features not in mainline yet, I plan to deliver it later. And the really nice thing about this patch is that it actually removes many lines of code while adding this nice functionality: fs/xfs/Makefile | 8 fs/xfs/linux-2.6/xfs_acl.c | 1 fs/xfs/linux-2.6/xfs_aops.c | 52 - fs/xfs/linux-2.6/xfs_aops.h | 2 fs/xfs/linux-2.6/xfs_buf.c | 117 +-- fs/xfs/linux-2.6/xfs_buf.h | 33 fs/xfs/linux-2.6/xfs_fs_subr.c | 3 fs/xfs/linux-2.6/xfs_ioctl.c | 1 fs/xfs/linux-2.6/xfs_ioctl32.c | 1 fs/xfs/linux-2.6/xfs_iops.c | 1 fs/xfs/linux-2.6/xfs_linux.h | 1 fs/xfs/linux-2.6/xfs_lrw.c | 87 -- fs/xfs/linux-2.6/xfs_lrw.h | 45 - fs/xfs/linux-2.6/xfs_super.c | 104 --- fs/xfs/linux-2.6/xfs_super.h | 7 fs/xfs/linux-2.6/xfs_sync.c | 1 fs/xfs/linux-2.6/xfs_trace.c | 75 ++ fs/xfs/linux-2.6/xfs_trace.h | 1369 +++++++++++++++++++++++++++++++++++++++++ fs/xfs/linux-2.6/xfs_vnode.h | 4 fs/xfs/quota/xfs_dquot.c | 110 --- fs/xfs/quota/xfs_dquot.h | 21 fs/xfs/quota/xfs_qm.c | 40 - fs/xfs/quota/xfs_qm_syscalls.c | 4 fs/xfs/support/ktrace.c | 323 --------- fs/xfs/support/ktrace.h | 85 -- fs/xfs/xfs.h | 16 fs/xfs/xfs_ag.h | 14 fs/xfs/xfs_alloc.c | 230 +----- fs/xfs/xfs_alloc.h | 27 fs/xfs/xfs_alloc_btree.c | 1 fs/xfs/xfs_attr.c | 107 --- fs/xfs/xfs_attr.h | 10 fs/xfs/xfs_attr_leaf.c | 14 fs/xfs/xfs_attr_sf.h | 40 - fs/xfs/xfs_bmap.c | 507 +++------------ fs/xfs/xfs_bmap.h | 49 - fs/xfs/xfs_bmap_btree.c | 6 fs/xfs/xfs_btree.c | 5 fs/xfs/xfs_btree_trace.h | 17 fs/xfs/xfs_buf_item.c | 87 -- fs/xfs/xfs_buf_item.h | 20 fs/xfs/xfs_da_btree.c | 3 fs/xfs/xfs_da_btree.h | 7 fs/xfs/xfs_dfrag.c | 2 fs/xfs/xfs_dir2.c | 8 fs/xfs/xfs_dir2_block.c | 20 fs/xfs/xfs_dir2_leaf.c | 21 fs/xfs/xfs_dir2_node.c | 27 fs/xfs/xfs_dir2_sf.c | 26 fs/xfs/xfs_dir2_trace.c | 216 ------ fs/xfs/xfs_dir2_trace.h | 72 -- fs/xfs/xfs_filestream.c | 8 fs/xfs/xfs_fsops.c | 2 fs/xfs/xfs_iget.c | 111 --- fs/xfs/xfs_inode.c | 67 -- fs/xfs/xfs_inode.h | 76 -- fs/xfs/xfs_inode_item.c | 5 fs/xfs/xfs_iomap.c | 85 -- fs/xfs/xfs_iomap.h | 8 fs/xfs/xfs_log.c | 181 +---- fs/xfs/xfs_log_priv.h | 20 fs/xfs/xfs_log_recover.c | 1 fs/xfs/xfs_mount.c | 2 fs/xfs/xfs_quota.h | 8 fs/xfs/xfs_rename.c | 1 fs/xfs/xfs_rtalloc.c | 1 fs/xfs/xfs_rw.c | 3 fs/xfs/xfs_trans.h | 47 + fs/xfs/xfs_trans_buf.c | 62 - fs/xfs/xfs_vnodeops.c | 8 70 files changed, 2151 insertions(+), 2592 deletions(-) Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Alex Elder <aelder@sgi.com>
2009-12-15 07:14:59 +08:00
#include "xfs_trace.h"
#include "xfs_icache.h"
#include "xfs_quota.h"
#include "xfs_dquot_item.h"
#include "xfs_dquot.h"
#include "xfs_reflink.h"
#define XFS_WRITEIO_ALIGN(mp,off) (((off) >> mp->m_writeio_log) \
<< mp->m_writeio_log)
static int
xfs_alert_fsblock_zero(
xfs_inode_t *ip,
xfs_bmbt_irec_t *imap)
{
xfs_alert_tag(ip->i_mount, XFS_PTAG_FSBLOCK_ZERO,
"Access to block zero in inode %llu "
"start_block: %llx start_off: %llx "
"blkcnt: %llx extent-state: %x",
(unsigned long long)ip->i_ino,
(unsigned long long)imap->br_startblock,
(unsigned long long)imap->br_startoff,
(unsigned long long)imap->br_blockcount,
imap->br_state);
return -EFSCORRUPTED;
}
int
xfs_bmbt_to_iomap(
struct xfs_inode *ip,
struct iomap *iomap,
struct xfs_bmbt_irec *imap,
bool shared)
{
struct xfs_mount *mp = ip->i_mount;
if (unlikely(!imap->br_startblock && !XFS_IS_REALTIME_INODE(ip)))
return xfs_alert_fsblock_zero(ip, imap);
if (imap->br_startblock == HOLESTARTBLOCK) {
iomap->addr = IOMAP_NULL_ADDR;
iomap->type = IOMAP_HOLE;
} else if (imap->br_startblock == DELAYSTARTBLOCK ||
isnullstartblock(imap->br_startblock)) {
iomap->addr = IOMAP_NULL_ADDR;
iomap->type = IOMAP_DELALLOC;
} else {
iomap->addr = BBTOB(xfs_fsb_to_db(ip, imap->br_startblock));
if (imap->br_state == XFS_EXT_UNWRITTEN)
iomap->type = IOMAP_UNWRITTEN;
else
iomap->type = IOMAP_MAPPED;
}
iomap->offset = XFS_FSB_TO_B(mp, imap->br_startoff);
iomap->length = XFS_FSB_TO_B(mp, imap->br_blockcount);
iomap->bdev = xfs_find_bdev_for_inode(VFS_I(ip));
iomap->dax_dev = xfs_find_daxdev_for_inode(VFS_I(ip));
if (xfs_ipincount(ip) &&
(ip->i_itemp->ili_fsync_fields & ~XFS_ILOG_TIMESTAMP))
iomap->flags |= IOMAP_F_DIRTY;
if (shared)
iomap->flags |= IOMAP_F_SHARED;
return 0;
}
static void
xfs_hole_to_iomap(
struct xfs_inode *ip,
struct iomap *iomap,
xfs_fileoff_t offset_fsb,
xfs_fileoff_t end_fsb)
{
iomap->addr = IOMAP_NULL_ADDR;
iomap->type = IOMAP_HOLE;
iomap->offset = XFS_FSB_TO_B(ip->i_mount, offset_fsb);
iomap->length = XFS_FSB_TO_B(ip->i_mount, end_fsb - offset_fsb);
iomap->bdev = xfs_find_bdev_for_inode(VFS_I(ip));
iomap->dax_dev = xfs_find_daxdev_for_inode(VFS_I(ip));
}
xfs_extlen_t
xfs_eof_alignment(
struct xfs_inode *ip,
xfs_extlen_t extsize)
{
struct xfs_mount *mp = ip->i_mount;
xfs_extlen_t align = 0;
if (!XFS_IS_REALTIME_INODE(ip)) {
/*
* Round up the allocation request to a stripe unit
* (m_dalign) boundary if the file size is >= stripe unit
* size, and we are allocating past the allocation eof.
*
* If mounted with the "-o swalloc" option the alignment is
* increased from the strip unit size to the stripe width.
*/
if (mp->m_swidth && (mp->m_flags & XFS_MOUNT_SWALLOC))
align = mp->m_swidth;
else if (mp->m_dalign)
align = mp->m_dalign;
if (align && XFS_ISIZE(ip) < XFS_FSB_TO_B(mp, align))
align = 0;
}
/*
* Always round up the allocation request to an extent boundary
* (when file on a real-time subvolume or has di_extsize hint).
*/
if (extsize) {
if (align)
align = roundup_64(align, extsize);
else
align = extsize;
}
return align;
}
STATIC int
xfs_iomap_eof_align_last_fsb(
struct xfs_inode *ip,
xfs_extlen_t extsize,
xfs_fileoff_t *last_fsb)
{
xfs_extlen_t align = xfs_eof_alignment(ip, extsize);
if (align) {
xfs_fileoff_t new_last_fsb = roundup_64(*last_fsb, align);
int eof, error;
error = xfs_bmap_eof(ip, new_last_fsb, XFS_DATA_FORK, &eof);
if (error)
return error;
if (eof)
*last_fsb = new_last_fsb;
}
return 0;
}
int
xfs_iomap_write_direct(
xfs_inode_t *ip,
xfs_off_t offset,
size_t count,
xfs_bmbt_irec_t *imap,
int nmaps)
{
xfs_mount_t *mp = ip->i_mount;
xfs_fileoff_t offset_fsb;
xfs_fileoff_t last_fsb;
xfs_filblks_t count_fsb, resaligned;
xfs_extlen_t extsz;
int nimaps;
int quota_flag;
int rt;
xfs_trans_t *tp;
uint qblocks, resblks, resrtextents;
int error;
xfs: add missing ilock around dio write last extent alignment The iomap codepath (via get_blocks()) acquires and release the inode lock in the case of a direct write that requires block allocation. This is because xfs_iomap_write_direct() allocates a transaction, which means the ilock must be dropped and reacquired after the transaction is allocated and reserved. xfs_iomap_write_direct() invokes xfs_iomap_eof_align_last_fsb() before the transaction is created and thus before the ilock is reacquired. This can lead to calls to xfs_iread_extents() and reads of the in-core extent list without any synchronization (via xfs_bmap_eof() and xfs_bmap_last_extent()). xfs_iread_extents() assert fails if the ilock is not held, but this is not currently seen in practice as the current callers had already invoked xfs_bmapi_read(). What has been seen in practice are reports of crashes down in the xfs_bmap_eof() codepath on direct writes due to seemingly bogus pointer references from xfs_iext_get_ext(). While an explicit reproducer is not currently available to confirm the cause of the problem, crash analysis and code inspection from David Jeffrey had identified the insufficient locking. xfs_iomap_eof_align_last_fsb() is called from other contexts with the inode lock already held, so we cannot acquire it therein. __xfs_get_blocks() acquires and drops the ilock with variable flags to cover the event that the extent list must be read in. The common case is that __xfs_get_blocks() acquires the shared ilock. To provide locking around the last extent alignment call without adding more lock cycles to the dio path, update xfs_iomap_write_direct() to expect the shared ilock held on entry and do the extent alignment under its protection. Demote the lock, if necessary, from __xfs_get_blocks() and push the xfs_qm_dqattach() call outside of the shared lock critical section. Also, add an assert to document that the extent list is always expected to be present in this path. Otherwise, we risk a call to xfs_iread_extents() while under the shared ilock. This is safe as all current callers have executed an xfs_bmapi_read() call under the current iolock context. Reported-by: David Jeffery <djeffery@redhat.com> Signed-off-by: Brian Foster <bfoster@redhat.com> Reviewed-by: Dave Chinner <dchinner@redhat.com> Signed-off-by: Dave Chinner <david@fromorbit.com>
2015-10-12 12:34:20 +08:00
int lockmode;
xfs: Don't use unwritten extents for DAX DAX has a page fault serialisation problem with block allocation. Because it allows concurrent page faults and does not have a page lock to serialise faults to the same page, it can get two concurrent faults to the page that race. When two read faults race, this isn't a huge problem as the data underlying the page is not changing and so "detect and drop" works just fine. The issues are to do with write faults. When two write faults occur, we serialise block allocation in get_blocks() so only one faul will allocate the extent. It will, however, be marked as an unwritten extent, and that is where the problem lies - the DAX fault code cannot differentiate between a block that was just allocated and a block that was preallocated and needs zeroing. The result is that both write faults end up zeroing the block and attempting to convert it back to written. The problem is that the first fault can zero and convert before the second fault starts zeroing, resulting in the zeroing for the second fault overwriting the data that the first fault wrote with zeros. The second fault then attempts to convert the unwritten extent, which is then a no-op because it's already written. Data loss occurs as a result of this race. Because there is no sane locking construct in the page fault code that we can use for serialisation across the page faults, we need to ensure block allocation and zeroing occurs atomically in the filesystem. This means we can still take concurrent page faults and the only time they will serialise is in the filesystem mapping/allocation callback. The page fault code will always see written, initialised extents, so we will be able to remove the unwritten extent handling from the DAX code when all filesystems are converted. Signed-off-by: Dave Chinner <dchinner@redhat.com> Reviewed-by: Brian Foster <bfoster@redhat.com> Signed-off-by: Dave Chinner <david@fromorbit.com>
2015-11-03 09:37:00 +08:00
int bmapi_flags = XFS_BMAPI_PREALLOC;
uint tflags = 0;
rt = XFS_IS_REALTIME_INODE(ip);
extsz = xfs_get_extsz_hint(ip);
xfs: add missing ilock around dio write last extent alignment The iomap codepath (via get_blocks()) acquires and release the inode lock in the case of a direct write that requires block allocation. This is because xfs_iomap_write_direct() allocates a transaction, which means the ilock must be dropped and reacquired after the transaction is allocated and reserved. xfs_iomap_write_direct() invokes xfs_iomap_eof_align_last_fsb() before the transaction is created and thus before the ilock is reacquired. This can lead to calls to xfs_iread_extents() and reads of the in-core extent list without any synchronization (via xfs_bmap_eof() and xfs_bmap_last_extent()). xfs_iread_extents() assert fails if the ilock is not held, but this is not currently seen in practice as the current callers had already invoked xfs_bmapi_read(). What has been seen in practice are reports of crashes down in the xfs_bmap_eof() codepath on direct writes due to seemingly bogus pointer references from xfs_iext_get_ext(). While an explicit reproducer is not currently available to confirm the cause of the problem, crash analysis and code inspection from David Jeffrey had identified the insufficient locking. xfs_iomap_eof_align_last_fsb() is called from other contexts with the inode lock already held, so we cannot acquire it therein. __xfs_get_blocks() acquires and drops the ilock with variable flags to cover the event that the extent list must be read in. The common case is that __xfs_get_blocks() acquires the shared ilock. To provide locking around the last extent alignment call without adding more lock cycles to the dio path, update xfs_iomap_write_direct() to expect the shared ilock held on entry and do the extent alignment under its protection. Demote the lock, if necessary, from __xfs_get_blocks() and push the xfs_qm_dqattach() call outside of the shared lock critical section. Also, add an assert to document that the extent list is always expected to be present in this path. Otherwise, we risk a call to xfs_iread_extents() while under the shared ilock. This is safe as all current callers have executed an xfs_bmapi_read() call under the current iolock context. Reported-by: David Jeffery <djeffery@redhat.com> Signed-off-by: Brian Foster <bfoster@redhat.com> Reviewed-by: Dave Chinner <dchinner@redhat.com> Signed-off-by: Dave Chinner <david@fromorbit.com>
2015-10-12 12:34:20 +08:00
lockmode = XFS_ILOCK_SHARED; /* locked by caller */
ASSERT(xfs_isilocked(ip, lockmode));
offset_fsb = XFS_B_TO_FSBT(mp, offset);
last_fsb = XFS_B_TO_FSB(mp, ((xfs_ufsize_t)(offset + count)));
if ((offset + count) > XFS_ISIZE(ip)) {
xfs: add missing ilock around dio write last extent alignment The iomap codepath (via get_blocks()) acquires and release the inode lock in the case of a direct write that requires block allocation. This is because xfs_iomap_write_direct() allocates a transaction, which means the ilock must be dropped and reacquired after the transaction is allocated and reserved. xfs_iomap_write_direct() invokes xfs_iomap_eof_align_last_fsb() before the transaction is created and thus before the ilock is reacquired. This can lead to calls to xfs_iread_extents() and reads of the in-core extent list without any synchronization (via xfs_bmap_eof() and xfs_bmap_last_extent()). xfs_iread_extents() assert fails if the ilock is not held, but this is not currently seen in practice as the current callers had already invoked xfs_bmapi_read(). What has been seen in practice are reports of crashes down in the xfs_bmap_eof() codepath on direct writes due to seemingly bogus pointer references from xfs_iext_get_ext(). While an explicit reproducer is not currently available to confirm the cause of the problem, crash analysis and code inspection from David Jeffrey had identified the insufficient locking. xfs_iomap_eof_align_last_fsb() is called from other contexts with the inode lock already held, so we cannot acquire it therein. __xfs_get_blocks() acquires and drops the ilock with variable flags to cover the event that the extent list must be read in. The common case is that __xfs_get_blocks() acquires the shared ilock. To provide locking around the last extent alignment call without adding more lock cycles to the dio path, update xfs_iomap_write_direct() to expect the shared ilock held on entry and do the extent alignment under its protection. Demote the lock, if necessary, from __xfs_get_blocks() and push the xfs_qm_dqattach() call outside of the shared lock critical section. Also, add an assert to document that the extent list is always expected to be present in this path. Otherwise, we risk a call to xfs_iread_extents() while under the shared ilock. This is safe as all current callers have executed an xfs_bmapi_read() call under the current iolock context. Reported-by: David Jeffery <djeffery@redhat.com> Signed-off-by: Brian Foster <bfoster@redhat.com> Reviewed-by: Dave Chinner <dchinner@redhat.com> Signed-off-by: Dave Chinner <david@fromorbit.com>
2015-10-12 12:34:20 +08:00
/*
* Assert that the in-core extent list is present since this can
* call xfs_iread_extents() and we only have the ilock shared.
* This should be safe because the lock was held around a bmapi
* call in the caller and we only need it to access the in-core
* list.
*/
ASSERT(XFS_IFORK_PTR(ip, XFS_DATA_FORK)->if_flags &
XFS_IFEXTENTS);
error = xfs_iomap_eof_align_last_fsb(ip, extsz, &last_fsb);
if (error)
xfs: add missing ilock around dio write last extent alignment The iomap codepath (via get_blocks()) acquires and release the inode lock in the case of a direct write that requires block allocation. This is because xfs_iomap_write_direct() allocates a transaction, which means the ilock must be dropped and reacquired after the transaction is allocated and reserved. xfs_iomap_write_direct() invokes xfs_iomap_eof_align_last_fsb() before the transaction is created and thus before the ilock is reacquired. This can lead to calls to xfs_iread_extents() and reads of the in-core extent list without any synchronization (via xfs_bmap_eof() and xfs_bmap_last_extent()). xfs_iread_extents() assert fails if the ilock is not held, but this is not currently seen in practice as the current callers had already invoked xfs_bmapi_read(). What has been seen in practice are reports of crashes down in the xfs_bmap_eof() codepath on direct writes due to seemingly bogus pointer references from xfs_iext_get_ext(). While an explicit reproducer is not currently available to confirm the cause of the problem, crash analysis and code inspection from David Jeffrey had identified the insufficient locking. xfs_iomap_eof_align_last_fsb() is called from other contexts with the inode lock already held, so we cannot acquire it therein. __xfs_get_blocks() acquires and drops the ilock with variable flags to cover the event that the extent list must be read in. The common case is that __xfs_get_blocks() acquires the shared ilock. To provide locking around the last extent alignment call without adding more lock cycles to the dio path, update xfs_iomap_write_direct() to expect the shared ilock held on entry and do the extent alignment under its protection. Demote the lock, if necessary, from __xfs_get_blocks() and push the xfs_qm_dqattach() call outside of the shared lock critical section. Also, add an assert to document that the extent list is always expected to be present in this path. Otherwise, we risk a call to xfs_iread_extents() while under the shared ilock. This is safe as all current callers have executed an xfs_bmapi_read() call under the current iolock context. Reported-by: David Jeffery <djeffery@redhat.com> Signed-off-by: Brian Foster <bfoster@redhat.com> Reviewed-by: Dave Chinner <dchinner@redhat.com> Signed-off-by: Dave Chinner <david@fromorbit.com>
2015-10-12 12:34:20 +08:00
goto out_unlock;
} else {
if (nmaps && (imap->br_startblock == HOLESTARTBLOCK))
last_fsb = min(last_fsb, (xfs_fileoff_t)
imap->br_blockcount +
imap->br_startoff);
}
count_fsb = last_fsb - offset_fsb;
ASSERT(count_fsb > 0);
resaligned = xfs_aligned_fsb_count(offset_fsb, count_fsb, extsz);
if (unlikely(rt)) {
resrtextents = qblocks = resaligned;
resrtextents /= mp->m_sb.sb_rextsize;
resblks = XFS_DIOSTRAT_SPACE_RES(mp, 0);
quota_flag = XFS_QMOPT_RES_RTBLKS;
} else {
resrtextents = 0;
resblks = qblocks = XFS_DIOSTRAT_SPACE_RES(mp, resaligned);
quota_flag = XFS_QMOPT_RES_REGBLKS;
}
xfs: add missing ilock around dio write last extent alignment The iomap codepath (via get_blocks()) acquires and release the inode lock in the case of a direct write that requires block allocation. This is because xfs_iomap_write_direct() allocates a transaction, which means the ilock must be dropped and reacquired after the transaction is allocated and reserved. xfs_iomap_write_direct() invokes xfs_iomap_eof_align_last_fsb() before the transaction is created and thus before the ilock is reacquired. This can lead to calls to xfs_iread_extents() and reads of the in-core extent list without any synchronization (via xfs_bmap_eof() and xfs_bmap_last_extent()). xfs_iread_extents() assert fails if the ilock is not held, but this is not currently seen in practice as the current callers had already invoked xfs_bmapi_read(). What has been seen in practice are reports of crashes down in the xfs_bmap_eof() codepath on direct writes due to seemingly bogus pointer references from xfs_iext_get_ext(). While an explicit reproducer is not currently available to confirm the cause of the problem, crash analysis and code inspection from David Jeffrey had identified the insufficient locking. xfs_iomap_eof_align_last_fsb() is called from other contexts with the inode lock already held, so we cannot acquire it therein. __xfs_get_blocks() acquires and drops the ilock with variable flags to cover the event that the extent list must be read in. The common case is that __xfs_get_blocks() acquires the shared ilock. To provide locking around the last extent alignment call without adding more lock cycles to the dio path, update xfs_iomap_write_direct() to expect the shared ilock held on entry and do the extent alignment under its protection. Demote the lock, if necessary, from __xfs_get_blocks() and push the xfs_qm_dqattach() call outside of the shared lock critical section. Also, add an assert to document that the extent list is always expected to be present in this path. Otherwise, we risk a call to xfs_iread_extents() while under the shared ilock. This is safe as all current callers have executed an xfs_bmapi_read() call under the current iolock context. Reported-by: David Jeffery <djeffery@redhat.com> Signed-off-by: Brian Foster <bfoster@redhat.com> Reviewed-by: Dave Chinner <dchinner@redhat.com> Signed-off-by: Dave Chinner <david@fromorbit.com>
2015-10-12 12:34:20 +08:00
/*
* Drop the shared lock acquired by the caller, attach the dquot if
* necessary and move on to transaction setup.
*/
xfs_iunlock(ip, lockmode);
error = xfs_qm_dqattach(ip);
xfs: add missing ilock around dio write last extent alignment The iomap codepath (via get_blocks()) acquires and release the inode lock in the case of a direct write that requires block allocation. This is because xfs_iomap_write_direct() allocates a transaction, which means the ilock must be dropped and reacquired after the transaction is allocated and reserved. xfs_iomap_write_direct() invokes xfs_iomap_eof_align_last_fsb() before the transaction is created and thus before the ilock is reacquired. This can lead to calls to xfs_iread_extents() and reads of the in-core extent list without any synchronization (via xfs_bmap_eof() and xfs_bmap_last_extent()). xfs_iread_extents() assert fails if the ilock is not held, but this is not currently seen in practice as the current callers had already invoked xfs_bmapi_read(). What has been seen in practice are reports of crashes down in the xfs_bmap_eof() codepath on direct writes due to seemingly bogus pointer references from xfs_iext_get_ext(). While an explicit reproducer is not currently available to confirm the cause of the problem, crash analysis and code inspection from David Jeffrey had identified the insufficient locking. xfs_iomap_eof_align_last_fsb() is called from other contexts with the inode lock already held, so we cannot acquire it therein. __xfs_get_blocks() acquires and drops the ilock with variable flags to cover the event that the extent list must be read in. The common case is that __xfs_get_blocks() acquires the shared ilock. To provide locking around the last extent alignment call without adding more lock cycles to the dio path, update xfs_iomap_write_direct() to expect the shared ilock held on entry and do the extent alignment under its protection. Demote the lock, if necessary, from __xfs_get_blocks() and push the xfs_qm_dqattach() call outside of the shared lock critical section. Also, add an assert to document that the extent list is always expected to be present in this path. Otherwise, we risk a call to xfs_iread_extents() while under the shared ilock. This is safe as all current callers have executed an xfs_bmapi_read() call under the current iolock context. Reported-by: David Jeffery <djeffery@redhat.com> Signed-off-by: Brian Foster <bfoster@redhat.com> Reviewed-by: Dave Chinner <dchinner@redhat.com> Signed-off-by: Dave Chinner <david@fromorbit.com>
2015-10-12 12:34:20 +08:00
if (error)
return error;
xfs: Don't use unwritten extents for DAX DAX has a page fault serialisation problem with block allocation. Because it allows concurrent page faults and does not have a page lock to serialise faults to the same page, it can get two concurrent faults to the page that race. When two read faults race, this isn't a huge problem as the data underlying the page is not changing and so "detect and drop" works just fine. The issues are to do with write faults. When two write faults occur, we serialise block allocation in get_blocks() so only one faul will allocate the extent. It will, however, be marked as an unwritten extent, and that is where the problem lies - the DAX fault code cannot differentiate between a block that was just allocated and a block that was preallocated and needs zeroing. The result is that both write faults end up zeroing the block and attempting to convert it back to written. The problem is that the first fault can zero and convert before the second fault starts zeroing, resulting in the zeroing for the second fault overwriting the data that the first fault wrote with zeros. The second fault then attempts to convert the unwritten extent, which is then a no-op because it's already written. Data loss occurs as a result of this race. Because there is no sane locking construct in the page fault code that we can use for serialisation across the page faults, we need to ensure block allocation and zeroing occurs atomically in the filesystem. This means we can still take concurrent page faults and the only time they will serialise is in the filesystem mapping/allocation callback. The page fault code will always see written, initialised extents, so we will be able to remove the unwritten extent handling from the DAX code when all filesystems are converted. Signed-off-by: Dave Chinner <dchinner@redhat.com> Reviewed-by: Brian Foster <bfoster@redhat.com> Signed-off-by: Dave Chinner <david@fromorbit.com>
2015-11-03 09:37:00 +08:00
/*
* For DAX, we do not allocate unwritten extents, but instead we zero
* the block before we commit the transaction. Ideally we'd like to do
* this outside the transaction context, but if we commit and then crash
* we may not have zeroed the blocks and this will be exposed on
* recovery of the allocation. Hence we must zero before commit.
*
xfs: Don't use unwritten extents for DAX DAX has a page fault serialisation problem with block allocation. Because it allows concurrent page faults and does not have a page lock to serialise faults to the same page, it can get two concurrent faults to the page that race. When two read faults race, this isn't a huge problem as the data underlying the page is not changing and so "detect and drop" works just fine. The issues are to do with write faults. When two write faults occur, we serialise block allocation in get_blocks() so only one faul will allocate the extent. It will, however, be marked as an unwritten extent, and that is where the problem lies - the DAX fault code cannot differentiate between a block that was just allocated and a block that was preallocated and needs zeroing. The result is that both write faults end up zeroing the block and attempting to convert it back to written. The problem is that the first fault can zero and convert before the second fault starts zeroing, resulting in the zeroing for the second fault overwriting the data that the first fault wrote with zeros. The second fault then attempts to convert the unwritten extent, which is then a no-op because it's already written. Data loss occurs as a result of this race. Because there is no sane locking construct in the page fault code that we can use for serialisation across the page faults, we need to ensure block allocation and zeroing occurs atomically in the filesystem. This means we can still take concurrent page faults and the only time they will serialise is in the filesystem mapping/allocation callback. The page fault code will always see written, initialised extents, so we will be able to remove the unwritten extent handling from the DAX code when all filesystems are converted. Signed-off-by: Dave Chinner <dchinner@redhat.com> Reviewed-by: Brian Foster <bfoster@redhat.com> Signed-off-by: Dave Chinner <david@fromorbit.com>
2015-11-03 09:37:00 +08:00
* Further, if we are mapping unwritten extents here, we need to zero
* and convert them to written so that we don't need an unwritten extent
* callback for DAX. This also means that we need to be able to dip into
* the reserve block pool for bmbt block allocation if there is no space
* left but we need to do unwritten extent conversion.
xfs: Don't use unwritten extents for DAX DAX has a page fault serialisation problem with block allocation. Because it allows concurrent page faults and does not have a page lock to serialise faults to the same page, it can get two concurrent faults to the page that race. When two read faults race, this isn't a huge problem as the data underlying the page is not changing and so "detect and drop" works just fine. The issues are to do with write faults. When two write faults occur, we serialise block allocation in get_blocks() so only one faul will allocate the extent. It will, however, be marked as an unwritten extent, and that is where the problem lies - the DAX fault code cannot differentiate between a block that was just allocated and a block that was preallocated and needs zeroing. The result is that both write faults end up zeroing the block and attempting to convert it back to written. The problem is that the first fault can zero and convert before the second fault starts zeroing, resulting in the zeroing for the second fault overwriting the data that the first fault wrote with zeros. The second fault then attempts to convert the unwritten extent, which is then a no-op because it's already written. Data loss occurs as a result of this race. Because there is no sane locking construct in the page fault code that we can use for serialisation across the page faults, we need to ensure block allocation and zeroing occurs atomically in the filesystem. This means we can still take concurrent page faults and the only time they will serialise is in the filesystem mapping/allocation callback. The page fault code will always see written, initialised extents, so we will be able to remove the unwritten extent handling from the DAX code when all filesystems are converted. Signed-off-by: Dave Chinner <dchinner@redhat.com> Reviewed-by: Brian Foster <bfoster@redhat.com> Signed-off-by: Dave Chinner <david@fromorbit.com>
2015-11-03 09:37:00 +08:00
*/
if (IS_DAX(VFS_I(ip))) {
bmapi_flags = XFS_BMAPI_CONVERT | XFS_BMAPI_ZERO;
if (imap->br_state == XFS_EXT_UNWRITTEN) {
tflags |= XFS_TRANS_RESERVE;
resblks = XFS_DIOSTRAT_SPACE_RES(mp, 0) << 1;
}
xfs: Don't use unwritten extents for DAX DAX has a page fault serialisation problem with block allocation. Because it allows concurrent page faults and does not have a page lock to serialise faults to the same page, it can get two concurrent faults to the page that race. When two read faults race, this isn't a huge problem as the data underlying the page is not changing and so "detect and drop" works just fine. The issues are to do with write faults. When two write faults occur, we serialise block allocation in get_blocks() so only one faul will allocate the extent. It will, however, be marked as an unwritten extent, and that is where the problem lies - the DAX fault code cannot differentiate between a block that was just allocated and a block that was preallocated and needs zeroing. The result is that both write faults end up zeroing the block and attempting to convert it back to written. The problem is that the first fault can zero and convert before the second fault starts zeroing, resulting in the zeroing for the second fault overwriting the data that the first fault wrote with zeros. The second fault then attempts to convert the unwritten extent, which is then a no-op because it's already written. Data loss occurs as a result of this race. Because there is no sane locking construct in the page fault code that we can use for serialisation across the page faults, we need to ensure block allocation and zeroing occurs atomically in the filesystem. This means we can still take concurrent page faults and the only time they will serialise is in the filesystem mapping/allocation callback. The page fault code will always see written, initialised extents, so we will be able to remove the unwritten extent handling from the DAX code when all filesystems are converted. Signed-off-by: Dave Chinner <dchinner@redhat.com> Reviewed-by: Brian Foster <bfoster@redhat.com> Signed-off-by: Dave Chinner <david@fromorbit.com>
2015-11-03 09:37:00 +08:00
}
error = xfs_trans_alloc(mp, &M_RES(mp)->tr_write, resblks, resrtextents,
tflags, &tp);
if (error)
return error;
xfs: add missing ilock around dio write last extent alignment The iomap codepath (via get_blocks()) acquires and release the inode lock in the case of a direct write that requires block allocation. This is because xfs_iomap_write_direct() allocates a transaction, which means the ilock must be dropped and reacquired after the transaction is allocated and reserved. xfs_iomap_write_direct() invokes xfs_iomap_eof_align_last_fsb() before the transaction is created and thus before the ilock is reacquired. This can lead to calls to xfs_iread_extents() and reads of the in-core extent list without any synchronization (via xfs_bmap_eof() and xfs_bmap_last_extent()). xfs_iread_extents() assert fails if the ilock is not held, but this is not currently seen in practice as the current callers had already invoked xfs_bmapi_read(). What has been seen in practice are reports of crashes down in the xfs_bmap_eof() codepath on direct writes due to seemingly bogus pointer references from xfs_iext_get_ext(). While an explicit reproducer is not currently available to confirm the cause of the problem, crash analysis and code inspection from David Jeffrey had identified the insufficient locking. xfs_iomap_eof_align_last_fsb() is called from other contexts with the inode lock already held, so we cannot acquire it therein. __xfs_get_blocks() acquires and drops the ilock with variable flags to cover the event that the extent list must be read in. The common case is that __xfs_get_blocks() acquires the shared ilock. To provide locking around the last extent alignment call without adding more lock cycles to the dio path, update xfs_iomap_write_direct() to expect the shared ilock held on entry and do the extent alignment under its protection. Demote the lock, if necessary, from __xfs_get_blocks() and push the xfs_qm_dqattach() call outside of the shared lock critical section. Also, add an assert to document that the extent list is always expected to be present in this path. Otherwise, we risk a call to xfs_iread_extents() while under the shared ilock. This is safe as all current callers have executed an xfs_bmapi_read() call under the current iolock context. Reported-by: David Jeffery <djeffery@redhat.com> Signed-off-by: Brian Foster <bfoster@redhat.com> Reviewed-by: Dave Chinner <dchinner@redhat.com> Signed-off-by: Dave Chinner <david@fromorbit.com>
2015-10-12 12:34:20 +08:00
lockmode = XFS_ILOCK_EXCL;
xfs_ilock(ip, lockmode);
error = xfs_trans_reserve_quota_nblks(tp, ip, qblocks, 0, quota_flag);
if (error)
goto out_trans_cancel;
xfs_trans_ijoin(tp, ip, 0);
/*
* From this point onwards we overwrite the imap pointer that the
* caller gave to us.
*/
nimaps = 1;
error = xfs_bmapi_write(tp, ip, offset_fsb, count_fsb,
bmapi_flags, resblks, imap, &nimaps);
if (error)
goto out_res_cancel;
/*
* Complete the transaction
*/
error = xfs_trans_commit(tp);
if (error)
goto out_unlock;
/*
* Copy any maps to caller's array and return any error.
*/
if (nimaps == 0) {
error = -ENOSPC;
goto out_unlock;
}
if (!(imap->br_startblock || XFS_IS_REALTIME_INODE(ip)))
error = xfs_alert_fsblock_zero(ip, imap);
out_unlock:
xfs: add missing ilock around dio write last extent alignment The iomap codepath (via get_blocks()) acquires and release the inode lock in the case of a direct write that requires block allocation. This is because xfs_iomap_write_direct() allocates a transaction, which means the ilock must be dropped and reacquired after the transaction is allocated and reserved. xfs_iomap_write_direct() invokes xfs_iomap_eof_align_last_fsb() before the transaction is created and thus before the ilock is reacquired. This can lead to calls to xfs_iread_extents() and reads of the in-core extent list without any synchronization (via xfs_bmap_eof() and xfs_bmap_last_extent()). xfs_iread_extents() assert fails if the ilock is not held, but this is not currently seen in practice as the current callers had already invoked xfs_bmapi_read(). What has been seen in practice are reports of crashes down in the xfs_bmap_eof() codepath on direct writes due to seemingly bogus pointer references from xfs_iext_get_ext(). While an explicit reproducer is not currently available to confirm the cause of the problem, crash analysis and code inspection from David Jeffrey had identified the insufficient locking. xfs_iomap_eof_align_last_fsb() is called from other contexts with the inode lock already held, so we cannot acquire it therein. __xfs_get_blocks() acquires and drops the ilock with variable flags to cover the event that the extent list must be read in. The common case is that __xfs_get_blocks() acquires the shared ilock. To provide locking around the last extent alignment call without adding more lock cycles to the dio path, update xfs_iomap_write_direct() to expect the shared ilock held on entry and do the extent alignment under its protection. Demote the lock, if necessary, from __xfs_get_blocks() and push the xfs_qm_dqattach() call outside of the shared lock critical section. Also, add an assert to document that the extent list is always expected to be present in this path. Otherwise, we risk a call to xfs_iread_extents() while under the shared ilock. This is safe as all current callers have executed an xfs_bmapi_read() call under the current iolock context. Reported-by: David Jeffery <djeffery@redhat.com> Signed-off-by: Brian Foster <bfoster@redhat.com> Reviewed-by: Dave Chinner <dchinner@redhat.com> Signed-off-by: Dave Chinner <david@fromorbit.com>
2015-10-12 12:34:20 +08:00
xfs_iunlock(ip, lockmode);
return error;
out_res_cancel:
xfs_trans_unreserve_quota_nblks(tp, ip, (long)qblocks, 0, quota_flag);
out_trans_cancel:
xfs_trans_cancel(tp);
goto out_unlock;
}
STATIC bool
xfs_quota_need_throttle(
struct xfs_inode *ip,
int type,
xfs_fsblock_t alloc_blocks)
{
struct xfs_dquot *dq = xfs_inode_dquot(ip, type);
if (!dq || !xfs_this_quota_on(ip->i_mount, type))
return false;
/* no hi watermark, no throttle */
if (!dq->q_prealloc_hi_wmark)
return false;
/* under the lo watermark, no throttle */
if (dq->q_res_bcount + alloc_blocks < dq->q_prealloc_lo_wmark)
return false;
return true;
}
STATIC void
xfs_quota_calc_throttle(
struct xfs_inode *ip,
int type,
xfs_fsblock_t *qblocks,
int *qshift,
int64_t *qfreesp)
{
int64_t freesp;
int shift = 0;
struct xfs_dquot *dq = xfs_inode_dquot(ip, type);
/* no dq, or over hi wmark, squash the prealloc completely */
if (!dq || dq->q_res_bcount >= dq->q_prealloc_hi_wmark) {
*qblocks = 0;
*qfreesp = 0;
return;
}
freesp = dq->q_prealloc_hi_wmark - dq->q_res_bcount;
if (freesp < dq->q_low_space[XFS_QLOWSP_5_PCNT]) {
shift = 2;
if (freesp < dq->q_low_space[XFS_QLOWSP_3_PCNT])
shift += 2;
if (freesp < dq->q_low_space[XFS_QLOWSP_1_PCNT])
shift += 2;
}
if (freesp < *qfreesp)
*qfreesp = freesp;
/* only overwrite the throttle values if we are more aggressive */
if ((freesp >> shift) < (*qblocks >> *qshift)) {
*qblocks = freesp;
*qshift = shift;
}
}
xfs: dynamic speculative EOF preallocation Currently the size of the speculative preallocation during delayed allocation is fixed by either the allocsize mount option of a default size. We are seeing a lot of cases where we need to recommend using the allocsize mount option to prevent fragmentation when buffered writes land in the same AG. Rather than using a fixed preallocation size by default (up to 64k), make it dynamic by basing it on the current inode size. That way the EOF preallocation will increase as the file size increases. Hence for streaming writes we are much more likely to get large preallocations exactly when we need it to reduce fragementation. For default settings, the size of the initial extents is determined by the number of parallel writers and the amount of memory in the machine. For 4GB RAM and 4 concurrent 32GB file writes: EXT: FILE-OFFSET BLOCK-RANGE AG AG-OFFSET TOTAL 0: [0..1048575]: 1048672..2097247 0 (1048672..2097247) 1048576 1: [1048576..2097151]: 5242976..6291551 0 (5242976..6291551) 1048576 2: [2097152..4194303]: 12583008..14680159 0 (12583008..14680159) 2097152 3: [4194304..8388607]: 25165920..29360223 0 (25165920..29360223) 4194304 4: [8388608..16777215]: 58720352..67108959 0 (58720352..67108959) 8388608 5: [16777216..33554423]: 117440584..134217791 0 (117440584..134217791) 16777208 6: [33554424..50331511]: 184549056..201326143 0 (184549056..201326143) 16777088 7: [50331512..67108599]: 251657408..268434495 0 (251657408..268434495) 16777088 and for 16 concurrent 16GB file writes: EXT: FILE-OFFSET BLOCK-RANGE AG AG-OFFSET TOTAL 0: [0..262143]: 2490472..2752615 0 (2490472..2752615) 262144 1: [262144..524287]: 6291560..6553703 0 (6291560..6553703) 262144 2: [524288..1048575]: 13631592..14155879 0 (13631592..14155879) 524288 3: [1048576..2097151]: 30408808..31457383 0 (30408808..31457383) 1048576 4: [2097152..4194303]: 52428904..54526055 0 (52428904..54526055) 2097152 5: [4194304..8388607]: 104857704..109052007 0 (104857704..109052007) 4194304 6: [8388608..16777215]: 209715304..218103911 0 (209715304..218103911) 8388608 7: [16777216..33554423]: 452984848..469762055 0 (452984848..469762055) 16777208 Because it is hard to take back specualtive preallocation, cases where there are large slow growing log files on a nearly full filesystem may cause premature ENOSPC. Hence as the filesystem nears full, the maximum dynamic prealloc size іs reduced according to this table (based on 4k block size): freespace max prealloc size >5% full extent (8GB) 4-5% 2GB (8GB >> 2) 3-4% 1GB (8GB >> 3) 2-3% 512MB (8GB >> 4) 1-2% 256MB (8GB >> 5) <1% 128MB (8GB >> 6) This should reduce the amount of space held in speculative preallocation for such cases. The allocsize mount option turns off the dynamic behaviour and fixes the prealloc size to whatever the mount option specifies. i.e. the behaviour is unchanged. Signed-off-by: Dave Chinner <dchinner@redhat.com>
2011-01-04 08:35:03 +08:00
/*
* If we are doing a write at the end of the file and there are no allocations
* past this one, then extend the allocation out to the file system's write
* iosize.
*
xfs: dynamic speculative EOF preallocation Currently the size of the speculative preallocation during delayed allocation is fixed by either the allocsize mount option of a default size. We are seeing a lot of cases where we need to recommend using the allocsize mount option to prevent fragmentation when buffered writes land in the same AG. Rather than using a fixed preallocation size by default (up to 64k), make it dynamic by basing it on the current inode size. That way the EOF preallocation will increase as the file size increases. Hence for streaming writes we are much more likely to get large preallocations exactly when we need it to reduce fragementation. For default settings, the size of the initial extents is determined by the number of parallel writers and the amount of memory in the machine. For 4GB RAM and 4 concurrent 32GB file writes: EXT: FILE-OFFSET BLOCK-RANGE AG AG-OFFSET TOTAL 0: [0..1048575]: 1048672..2097247 0 (1048672..2097247) 1048576 1: [1048576..2097151]: 5242976..6291551 0 (5242976..6291551) 1048576 2: [2097152..4194303]: 12583008..14680159 0 (12583008..14680159) 2097152 3: [4194304..8388607]: 25165920..29360223 0 (25165920..29360223) 4194304 4: [8388608..16777215]: 58720352..67108959 0 (58720352..67108959) 8388608 5: [16777216..33554423]: 117440584..134217791 0 (117440584..134217791) 16777208 6: [33554424..50331511]: 184549056..201326143 0 (184549056..201326143) 16777088 7: [50331512..67108599]: 251657408..268434495 0 (251657408..268434495) 16777088 and for 16 concurrent 16GB file writes: EXT: FILE-OFFSET BLOCK-RANGE AG AG-OFFSET TOTAL 0: [0..262143]: 2490472..2752615 0 (2490472..2752615) 262144 1: [262144..524287]: 6291560..6553703 0 (6291560..6553703) 262144 2: [524288..1048575]: 13631592..14155879 0 (13631592..14155879) 524288 3: [1048576..2097151]: 30408808..31457383 0 (30408808..31457383) 1048576 4: [2097152..4194303]: 52428904..54526055 0 (52428904..54526055) 2097152 5: [4194304..8388607]: 104857704..109052007 0 (104857704..109052007) 4194304 6: [8388608..16777215]: 209715304..218103911 0 (209715304..218103911) 8388608 7: [16777216..33554423]: 452984848..469762055 0 (452984848..469762055) 16777208 Because it is hard to take back specualtive preallocation, cases where there are large slow growing log files on a nearly full filesystem may cause premature ENOSPC. Hence as the filesystem nears full, the maximum dynamic prealloc size іs reduced according to this table (based on 4k block size): freespace max prealloc size >5% full extent (8GB) 4-5% 2GB (8GB >> 2) 3-4% 1GB (8GB >> 3) 2-3% 512MB (8GB >> 4) 1-2% 256MB (8GB >> 5) <1% 128MB (8GB >> 6) This should reduce the amount of space held in speculative preallocation for such cases. The allocsize mount option turns off the dynamic behaviour and fixes the prealloc size to whatever the mount option specifies. i.e. the behaviour is unchanged. Signed-off-by: Dave Chinner <dchinner@redhat.com>
2011-01-04 08:35:03 +08:00
* If we don't have a user specified preallocation size, dynamically increase
* the preallocation size as the size of the file grows. Cap the maximum size
xfs: dynamic speculative EOF preallocation Currently the size of the speculative preallocation during delayed allocation is fixed by either the allocsize mount option of a default size. We are seeing a lot of cases where we need to recommend using the allocsize mount option to prevent fragmentation when buffered writes land in the same AG. Rather than using a fixed preallocation size by default (up to 64k), make it dynamic by basing it on the current inode size. That way the EOF preallocation will increase as the file size increases. Hence for streaming writes we are much more likely to get large preallocations exactly when we need it to reduce fragementation. For default settings, the size of the initial extents is determined by the number of parallel writers and the amount of memory in the machine. For 4GB RAM and 4 concurrent 32GB file writes: EXT: FILE-OFFSET BLOCK-RANGE AG AG-OFFSET TOTAL 0: [0..1048575]: 1048672..2097247 0 (1048672..2097247) 1048576 1: [1048576..2097151]: 5242976..6291551 0 (5242976..6291551) 1048576 2: [2097152..4194303]: 12583008..14680159 0 (12583008..14680159) 2097152 3: [4194304..8388607]: 25165920..29360223 0 (25165920..29360223) 4194304 4: [8388608..16777215]: 58720352..67108959 0 (58720352..67108959) 8388608 5: [16777216..33554423]: 117440584..134217791 0 (117440584..134217791) 16777208 6: [33554424..50331511]: 184549056..201326143 0 (184549056..201326143) 16777088 7: [50331512..67108599]: 251657408..268434495 0 (251657408..268434495) 16777088 and for 16 concurrent 16GB file writes: EXT: FILE-OFFSET BLOCK-RANGE AG AG-OFFSET TOTAL 0: [0..262143]: 2490472..2752615 0 (2490472..2752615) 262144 1: [262144..524287]: 6291560..6553703 0 (6291560..6553703) 262144 2: [524288..1048575]: 13631592..14155879 0 (13631592..14155879) 524288 3: [1048576..2097151]: 30408808..31457383 0 (30408808..31457383) 1048576 4: [2097152..4194303]: 52428904..54526055 0 (52428904..54526055) 2097152 5: [4194304..8388607]: 104857704..109052007 0 (104857704..109052007) 4194304 6: [8388608..16777215]: 209715304..218103911 0 (209715304..218103911) 8388608 7: [16777216..33554423]: 452984848..469762055 0 (452984848..469762055) 16777208 Because it is hard to take back specualtive preallocation, cases where there are large slow growing log files on a nearly full filesystem may cause premature ENOSPC. Hence as the filesystem nears full, the maximum dynamic prealloc size іs reduced according to this table (based on 4k block size): freespace max prealloc size >5% full extent (8GB) 4-5% 2GB (8GB >> 2) 3-4% 1GB (8GB >> 3) 2-3% 512MB (8GB >> 4) 1-2% 256MB (8GB >> 5) <1% 128MB (8GB >> 6) This should reduce the amount of space held in speculative preallocation for such cases. The allocsize mount option turns off the dynamic behaviour and fixes the prealloc size to whatever the mount option specifies. i.e. the behaviour is unchanged. Signed-off-by: Dave Chinner <dchinner@redhat.com>
2011-01-04 08:35:03 +08:00
* at a single extent or less if the filesystem is near full. The closer the
* filesystem is to full, the smaller the maximum prealocation.
*
* As an exception we don't do any preallocation at all if the file is smaller
* than the minimum preallocation and we are using the default dynamic
* preallocation scheme, as it is likely this is the only write to the file that
* is going to be done.
*
* We clean up any extra space left over when the file is closed in
* xfs_inactive().
xfs: dynamic speculative EOF preallocation Currently the size of the speculative preallocation during delayed allocation is fixed by either the allocsize mount option of a default size. We are seeing a lot of cases where we need to recommend using the allocsize mount option to prevent fragmentation when buffered writes land in the same AG. Rather than using a fixed preallocation size by default (up to 64k), make it dynamic by basing it on the current inode size. That way the EOF preallocation will increase as the file size increases. Hence for streaming writes we are much more likely to get large preallocations exactly when we need it to reduce fragementation. For default settings, the size of the initial extents is determined by the number of parallel writers and the amount of memory in the machine. For 4GB RAM and 4 concurrent 32GB file writes: EXT: FILE-OFFSET BLOCK-RANGE AG AG-OFFSET TOTAL 0: [0..1048575]: 1048672..2097247 0 (1048672..2097247) 1048576 1: [1048576..2097151]: 5242976..6291551 0 (5242976..6291551) 1048576 2: [2097152..4194303]: 12583008..14680159 0 (12583008..14680159) 2097152 3: [4194304..8388607]: 25165920..29360223 0 (25165920..29360223) 4194304 4: [8388608..16777215]: 58720352..67108959 0 (58720352..67108959) 8388608 5: [16777216..33554423]: 117440584..134217791 0 (117440584..134217791) 16777208 6: [33554424..50331511]: 184549056..201326143 0 (184549056..201326143) 16777088 7: [50331512..67108599]: 251657408..268434495 0 (251657408..268434495) 16777088 and for 16 concurrent 16GB file writes: EXT: FILE-OFFSET BLOCK-RANGE AG AG-OFFSET TOTAL 0: [0..262143]: 2490472..2752615 0 (2490472..2752615) 262144 1: [262144..524287]: 6291560..6553703 0 (6291560..6553703) 262144 2: [524288..1048575]: 13631592..14155879 0 (13631592..14155879) 524288 3: [1048576..2097151]: 30408808..31457383 0 (30408808..31457383) 1048576 4: [2097152..4194303]: 52428904..54526055 0 (52428904..54526055) 2097152 5: [4194304..8388607]: 104857704..109052007 0 (104857704..109052007) 4194304 6: [8388608..16777215]: 209715304..218103911 0 (209715304..218103911) 8388608 7: [16777216..33554423]: 452984848..469762055 0 (452984848..469762055) 16777208 Because it is hard to take back specualtive preallocation, cases where there are large slow growing log files on a nearly full filesystem may cause premature ENOSPC. Hence as the filesystem nears full, the maximum dynamic prealloc size іs reduced according to this table (based on 4k block size): freespace max prealloc size >5% full extent (8GB) 4-5% 2GB (8GB >> 2) 3-4% 1GB (8GB >> 3) 2-3% 512MB (8GB >> 4) 1-2% 256MB (8GB >> 5) <1% 128MB (8GB >> 6) This should reduce the amount of space held in speculative preallocation for such cases. The allocsize mount option turns off the dynamic behaviour and fixes the prealloc size to whatever the mount option specifies. i.e. the behaviour is unchanged. Signed-off-by: Dave Chinner <dchinner@redhat.com>
2011-01-04 08:35:03 +08:00
*/
STATIC xfs_fsblock_t
xfs_iomap_prealloc_size(
xfs: limit speculative prealloc size on sparse files Speculative preallocation based on the current file size works well for contiguous files, but is sub-optimal for sparse files where the EOF preallocation can fill holes and result in large amounts of zeros being written when it is not necessary. The algorithm is modified to prevent EOF speculative preallocation from triggering larger allocations on IO patterns of truncate--to-zero-seek-write-seek-write-.... which results in non-sparse files for large files. This, unfortunately, is the way cp now behaves when copying sparse files and so needs to be fixed. What this code does is that it looks at the existing extent adjacent to the current EOF and if it determines that it is a hole we disable speculative preallocation altogether. To avoid the next write from doing a large prealloc, it takes the size of subsequent preallocations from the current size of the existing EOF extent. IOWs, if you leave a hole in the file, it resets preallocation behaviour to the same as if it was a zero size file. Example new behaviour: $ xfs_io -f -c "pwrite 0 31m" \ -c "pwrite 33m 1m" \ -c "pwrite 128m 1m" \ -c "fiemap -v" /mnt/scratch/blah wrote 32505856/32505856 bytes at offset 0 31 MiB, 7936 ops; 0.0000 sec (1.608 GiB/sec and 421432.7439 ops/sec) wrote 1048576/1048576 bytes at offset 34603008 1 MiB, 256 ops; 0.0000 sec (1.462 GiB/sec and 383233.5329 ops/sec) wrote 1048576/1048576 bytes at offset 134217728 1 MiB, 256 ops; 0.0000 sec (1.719 GiB/sec and 450704.2254 ops/sec) /mnt/scratch/blah: EXT: FILE-OFFSET BLOCK-RANGE TOTAL FLAGS 0: [0..65535]: 96..65631 65536 0x0 1: [65536..67583]: hole 2048 2: [67584..69631]: 67680..69727 2048 0x0 3: [69632..262143]: hole 192512 4: [262144..264191]: 262240..264287 2048 0x1 Signed-off-by: Dave Chinner <dchinner@redhat.com> Reviewed-by: Mark Tinguely <tinguely@sgi.com> Reviewed-by: Brian Foster <bfoster@redhat.com> Signed-off-by: Ben Myers <bpm@sgi.com>
2013-02-11 13:05:01 +08:00
struct xfs_inode *ip,
xfs: introduce an always_cow mode Add a mode where XFS never overwrites existing blocks in place. This is to aid debugging our COW code, and also put infatructure in place for things like possible future support for zoned block devices, which can't support overwrites. This mode is enabled globally by doing a: echo 1 > /sys/fs/xfs/debug/always_cow Note that the parameter is global to allow running all tests in xfstests easily in this mode, which would not easily be possible with a per-fs sysfs file. In always_cow mode persistent preallocations are disabled, and fallocate will fail when called with a 0 mode (with our without FALLOC_FL_KEEP_SIZE), and not create unwritten extent for zeroed space when called with FALLOC_FL_ZERO_RANGE or FALLOC_FL_UNSHARE_RANGE. There are a few interesting xfstests failures when run in always_cow mode: - generic/392 fails because the bytes used in the file used to test hole punch recovery are less after the log replay. This is because the blocks written and then punched out are only freed with a delay due to the logging mechanism. - xfs/170 will fail as the already fragile file streams mechanism doesn't seem to interact well with the COW allocator - xfs/180 xfs/182 xfs/192 xfs/198 xfs/204 and xfs/208 will claim the file system is badly fragmented, but there is not much we can do to avoid that when always writing out of place - xfs/205 fails because overwriting a file in always_cow mode will require new space allocation and the assumption in the test thus don't work anymore. - xfs/326 fails to modify the file at all in always_cow mode after injecting the refcount error, leading to an unexpected md5sum after the remount, but that again is expected Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com> Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
2019-02-19 01:38:49 +08:00
int whichfork,
loff_t offset,
loff_t count,
struct xfs_iext_cursor *icur)
xfs: dynamic speculative EOF preallocation Currently the size of the speculative preallocation during delayed allocation is fixed by either the allocsize mount option of a default size. We are seeing a lot of cases where we need to recommend using the allocsize mount option to prevent fragmentation when buffered writes land in the same AG. Rather than using a fixed preallocation size by default (up to 64k), make it dynamic by basing it on the current inode size. That way the EOF preallocation will increase as the file size increases. Hence for streaming writes we are much more likely to get large preallocations exactly when we need it to reduce fragementation. For default settings, the size of the initial extents is determined by the number of parallel writers and the amount of memory in the machine. For 4GB RAM and 4 concurrent 32GB file writes: EXT: FILE-OFFSET BLOCK-RANGE AG AG-OFFSET TOTAL 0: [0..1048575]: 1048672..2097247 0 (1048672..2097247) 1048576 1: [1048576..2097151]: 5242976..6291551 0 (5242976..6291551) 1048576 2: [2097152..4194303]: 12583008..14680159 0 (12583008..14680159) 2097152 3: [4194304..8388607]: 25165920..29360223 0 (25165920..29360223) 4194304 4: [8388608..16777215]: 58720352..67108959 0 (58720352..67108959) 8388608 5: [16777216..33554423]: 117440584..134217791 0 (117440584..134217791) 16777208 6: [33554424..50331511]: 184549056..201326143 0 (184549056..201326143) 16777088 7: [50331512..67108599]: 251657408..268434495 0 (251657408..268434495) 16777088 and for 16 concurrent 16GB file writes: EXT: FILE-OFFSET BLOCK-RANGE AG AG-OFFSET TOTAL 0: [0..262143]: 2490472..2752615 0 (2490472..2752615) 262144 1: [262144..524287]: 6291560..6553703 0 (6291560..6553703) 262144 2: [524288..1048575]: 13631592..14155879 0 (13631592..14155879) 524288 3: [1048576..2097151]: 30408808..31457383 0 (30408808..31457383) 1048576 4: [2097152..4194303]: 52428904..54526055 0 (52428904..54526055) 2097152 5: [4194304..8388607]: 104857704..109052007 0 (104857704..109052007) 4194304 6: [8388608..16777215]: 209715304..218103911 0 (209715304..218103911) 8388608 7: [16777216..33554423]: 452984848..469762055 0 (452984848..469762055) 16777208 Because it is hard to take back specualtive preallocation, cases where there are large slow growing log files on a nearly full filesystem may cause premature ENOSPC. Hence as the filesystem nears full, the maximum dynamic prealloc size іs reduced according to this table (based on 4k block size): freespace max prealloc size >5% full extent (8GB) 4-5% 2GB (8GB >> 2) 3-4% 1GB (8GB >> 3) 2-3% 512MB (8GB >> 4) 1-2% 256MB (8GB >> 5) <1% 128MB (8GB >> 6) This should reduce the amount of space held in speculative preallocation for such cases. The allocsize mount option turns off the dynamic behaviour and fixes the prealloc size to whatever the mount option specifies. i.e. the behaviour is unchanged. Signed-off-by: Dave Chinner <dchinner@redhat.com>
2011-01-04 08:35:03 +08:00
{
struct xfs_mount *mp = ip->i_mount;
xfs: introduce an always_cow mode Add a mode where XFS never overwrites existing blocks in place. This is to aid debugging our COW code, and also put infatructure in place for things like possible future support for zoned block devices, which can't support overwrites. This mode is enabled globally by doing a: echo 1 > /sys/fs/xfs/debug/always_cow Note that the parameter is global to allow running all tests in xfstests easily in this mode, which would not easily be possible with a per-fs sysfs file. In always_cow mode persistent preallocations are disabled, and fallocate will fail when called with a 0 mode (with our without FALLOC_FL_KEEP_SIZE), and not create unwritten extent for zeroed space when called with FALLOC_FL_ZERO_RANGE or FALLOC_FL_UNSHARE_RANGE. There are a few interesting xfstests failures when run in always_cow mode: - generic/392 fails because the bytes used in the file used to test hole punch recovery are less after the log replay. This is because the blocks written and then punched out are only freed with a delay due to the logging mechanism. - xfs/170 will fail as the already fragile file streams mechanism doesn't seem to interact well with the COW allocator - xfs/180 xfs/182 xfs/192 xfs/198 xfs/204 and xfs/208 will claim the file system is badly fragmented, but there is not much we can do to avoid that when always writing out of place - xfs/205 fails because overwriting a file in always_cow mode will require new space allocation and the assumption in the test thus don't work anymore. - xfs/326 fails to modify the file at all in always_cow mode after injecting the refcount error, leading to an unexpected md5sum after the remount, but that again is expected Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com> Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
2019-02-19 01:38:49 +08:00
struct xfs_ifork *ifp = XFS_IFORK_PTR(ip, whichfork);
xfs_fileoff_t offset_fsb = XFS_B_TO_FSBT(mp, offset);
struct xfs_bmbt_irec prev;
int shift = 0;
int64_t freesp;
xfs_fsblock_t qblocks;
int qshift = 0;
xfs_fsblock_t alloc_blocks = 0;
if (offset + count <= XFS_ISIZE(ip))
return 0;
if (!(mp->m_flags & XFS_MOUNT_DFLT_IOSIZE) &&
(XFS_ISIZE(ip) < XFS_FSB_TO_B(mp, mp->m_writeio_blocks)))
return 0;
/*
* If an explicit allocsize is set, the file is small, or we
* are writing behind a hole, then use the minimum prealloc:
*/
if ((mp->m_flags & XFS_MOUNT_DFLT_IOSIZE) ||
XFS_ISIZE(ip) < XFS_FSB_TO_B(mp, mp->m_dalign) ||
!xfs_iext_peek_prev_extent(ifp, icur, &prev) ||
prev.br_startoff + prev.br_blockcount < offset_fsb)
return mp->m_writeio_blocks;
xfs: dynamic speculative EOF preallocation Currently the size of the speculative preallocation during delayed allocation is fixed by either the allocsize mount option of a default size. We are seeing a lot of cases where we need to recommend using the allocsize mount option to prevent fragmentation when buffered writes land in the same AG. Rather than using a fixed preallocation size by default (up to 64k), make it dynamic by basing it on the current inode size. That way the EOF preallocation will increase as the file size increases. Hence for streaming writes we are much more likely to get large preallocations exactly when we need it to reduce fragementation. For default settings, the size of the initial extents is determined by the number of parallel writers and the amount of memory in the machine. For 4GB RAM and 4 concurrent 32GB file writes: EXT: FILE-OFFSET BLOCK-RANGE AG AG-OFFSET TOTAL 0: [0..1048575]: 1048672..2097247 0 (1048672..2097247) 1048576 1: [1048576..2097151]: 5242976..6291551 0 (5242976..6291551) 1048576 2: [2097152..4194303]: 12583008..14680159 0 (12583008..14680159) 2097152 3: [4194304..8388607]: 25165920..29360223 0 (25165920..29360223) 4194304 4: [8388608..16777215]: 58720352..67108959 0 (58720352..67108959) 8388608 5: [16777216..33554423]: 117440584..134217791 0 (117440584..134217791) 16777208 6: [33554424..50331511]: 184549056..201326143 0 (184549056..201326143) 16777088 7: [50331512..67108599]: 251657408..268434495 0 (251657408..268434495) 16777088 and for 16 concurrent 16GB file writes: EXT: FILE-OFFSET BLOCK-RANGE AG AG-OFFSET TOTAL 0: [0..262143]: 2490472..2752615 0 (2490472..2752615) 262144 1: [262144..524287]: 6291560..6553703 0 (6291560..6553703) 262144 2: [524288..1048575]: 13631592..14155879 0 (13631592..14155879) 524288 3: [1048576..2097151]: 30408808..31457383 0 (30408808..31457383) 1048576 4: [2097152..4194303]: 52428904..54526055 0 (52428904..54526055) 2097152 5: [4194304..8388607]: 104857704..109052007 0 (104857704..109052007) 4194304 6: [8388608..16777215]: 209715304..218103911 0 (209715304..218103911) 8388608 7: [16777216..33554423]: 452984848..469762055 0 (452984848..469762055) 16777208 Because it is hard to take back specualtive preallocation, cases where there are large slow growing log files on a nearly full filesystem may cause premature ENOSPC. Hence as the filesystem nears full, the maximum dynamic prealloc size іs reduced according to this table (based on 4k block size): freespace max prealloc size >5% full extent (8GB) 4-5% 2GB (8GB >> 2) 3-4% 1GB (8GB >> 3) 2-3% 512MB (8GB >> 4) 1-2% 256MB (8GB >> 5) <1% 128MB (8GB >> 6) This should reduce the amount of space held in speculative preallocation for such cases. The allocsize mount option turns off the dynamic behaviour and fixes the prealloc size to whatever the mount option specifies. i.e. the behaviour is unchanged. Signed-off-by: Dave Chinner <dchinner@redhat.com>
2011-01-04 08:35:03 +08:00
/*
* Determine the initial size of the preallocation. We are beyond the
* current EOF here, but we need to take into account whether this is
* a sparse write or an extending write when determining the
* preallocation size. Hence we need to look up the extent that ends
* at the current write offset and use the result to determine the
* preallocation size.
*
* If the extent is a hole, then preallocation is essentially disabled.
* Otherwise we take the size of the preceding data extent as the basis
* for the preallocation size. If the size of the extent is greater than
* half the maximum extent length, then use the current offset as the
* basis. This ensures that for large files the preallocation size
* always extends to MAXEXTLEN rather than falling short due to things
* like stripe unit/width alignment of real extents.
*/
if (prev.br_blockcount <= (MAXEXTLEN >> 1))
alloc_blocks = prev.br_blockcount << 1;
else
alloc_blocks = XFS_B_TO_FSB(mp, offset);
if (!alloc_blocks)
goto check_writeio;
qblocks = alloc_blocks;
/*
* MAXEXTLEN is not a power of two value but we round the prealloc down
* to the nearest power of two value after throttling. To prevent the
* round down from unconditionally reducing the maximum supported prealloc
* size, we round up first, apply appropriate throttling, round down and
* cap the value to MAXEXTLEN.
*/
alloc_blocks = XFS_FILEOFF_MIN(roundup_pow_of_two(MAXEXTLEN),
alloc_blocks);
freesp = percpu_counter_read_positive(&mp->m_fdblocks);
if (freesp < mp->m_low_space[XFS_LOWSP_5_PCNT]) {
shift = 2;
if (freesp < mp->m_low_space[XFS_LOWSP_4_PCNT])
shift++;
if (freesp < mp->m_low_space[XFS_LOWSP_3_PCNT])
shift++;
if (freesp < mp->m_low_space[XFS_LOWSP_2_PCNT])
shift++;
if (freesp < mp->m_low_space[XFS_LOWSP_1_PCNT])
shift++;
xfs: dynamic speculative EOF preallocation Currently the size of the speculative preallocation during delayed allocation is fixed by either the allocsize mount option of a default size. We are seeing a lot of cases where we need to recommend using the allocsize mount option to prevent fragmentation when buffered writes land in the same AG. Rather than using a fixed preallocation size by default (up to 64k), make it dynamic by basing it on the current inode size. That way the EOF preallocation will increase as the file size increases. Hence for streaming writes we are much more likely to get large preallocations exactly when we need it to reduce fragementation. For default settings, the size of the initial extents is determined by the number of parallel writers and the amount of memory in the machine. For 4GB RAM and 4 concurrent 32GB file writes: EXT: FILE-OFFSET BLOCK-RANGE AG AG-OFFSET TOTAL 0: [0..1048575]: 1048672..2097247 0 (1048672..2097247) 1048576 1: [1048576..2097151]: 5242976..6291551 0 (5242976..6291551) 1048576 2: [2097152..4194303]: 12583008..14680159 0 (12583008..14680159) 2097152 3: [4194304..8388607]: 25165920..29360223 0 (25165920..29360223) 4194304 4: [8388608..16777215]: 58720352..67108959 0 (58720352..67108959) 8388608 5: [16777216..33554423]: 117440584..134217791 0 (117440584..134217791) 16777208 6: [33554424..50331511]: 184549056..201326143 0 (184549056..201326143) 16777088 7: [50331512..67108599]: 251657408..268434495 0 (251657408..268434495) 16777088 and for 16 concurrent 16GB file writes: EXT: FILE-OFFSET BLOCK-RANGE AG AG-OFFSET TOTAL 0: [0..262143]: 2490472..2752615 0 (2490472..2752615) 262144 1: [262144..524287]: 6291560..6553703 0 (6291560..6553703) 262144 2: [524288..1048575]: 13631592..14155879 0 (13631592..14155879) 524288 3: [1048576..2097151]: 30408808..31457383 0 (30408808..31457383) 1048576 4: [2097152..4194303]: 52428904..54526055 0 (52428904..54526055) 2097152 5: [4194304..8388607]: 104857704..109052007 0 (104857704..109052007) 4194304 6: [8388608..16777215]: 209715304..218103911 0 (209715304..218103911) 8388608 7: [16777216..33554423]: 452984848..469762055 0 (452984848..469762055) 16777208 Because it is hard to take back specualtive preallocation, cases where there are large slow growing log files on a nearly full filesystem may cause premature ENOSPC. Hence as the filesystem nears full, the maximum dynamic prealloc size іs reduced according to this table (based on 4k block size): freespace max prealloc size >5% full extent (8GB) 4-5% 2GB (8GB >> 2) 3-4% 1GB (8GB >> 3) 2-3% 512MB (8GB >> 4) 1-2% 256MB (8GB >> 5) <1% 128MB (8GB >> 6) This should reduce the amount of space held in speculative preallocation for such cases. The allocsize mount option turns off the dynamic behaviour and fixes the prealloc size to whatever the mount option specifies. i.e. the behaviour is unchanged. Signed-off-by: Dave Chinner <dchinner@redhat.com>
2011-01-04 08:35:03 +08:00
}
/*
* Check each quota to cap the prealloc size, provide a shift value to
* throttle with and adjust amount of available space.
*/
if (xfs_quota_need_throttle(ip, XFS_DQ_USER, alloc_blocks))
xfs_quota_calc_throttle(ip, XFS_DQ_USER, &qblocks, &qshift,
&freesp);
if (xfs_quota_need_throttle(ip, XFS_DQ_GROUP, alloc_blocks))
xfs_quota_calc_throttle(ip, XFS_DQ_GROUP, &qblocks, &qshift,
&freesp);
if (xfs_quota_need_throttle(ip, XFS_DQ_PROJ, alloc_blocks))
xfs_quota_calc_throttle(ip, XFS_DQ_PROJ, &qblocks, &qshift,
&freesp);
/*
* The final prealloc size is set to the minimum of free space available
* in each of the quotas and the overall filesystem.
*
* The shift throttle value is set to the maximum value as determined by
* the global low free space values and per-quota low free space values.
*/
alloc_blocks = min(alloc_blocks, qblocks);
shift = max(shift, qshift);
if (shift)
alloc_blocks >>= shift;
/*
* rounddown_pow_of_two() returns an undefined result if we pass in
* alloc_blocks = 0.
*/
if (alloc_blocks)
alloc_blocks = rounddown_pow_of_two(alloc_blocks);
if (alloc_blocks > MAXEXTLEN)
alloc_blocks = MAXEXTLEN;
/*
* If we are still trying to allocate more space than is
* available, squash the prealloc hard. This can happen if we
* have a large file on a small filesystem and the above
* lowspace thresholds are smaller than MAXEXTLEN.
*/
while (alloc_blocks && alloc_blocks >= freesp)
alloc_blocks >>= 4;
check_writeio:
xfs: dynamic speculative EOF preallocation Currently the size of the speculative preallocation during delayed allocation is fixed by either the allocsize mount option of a default size. We are seeing a lot of cases where we need to recommend using the allocsize mount option to prevent fragmentation when buffered writes land in the same AG. Rather than using a fixed preallocation size by default (up to 64k), make it dynamic by basing it on the current inode size. That way the EOF preallocation will increase as the file size increases. Hence for streaming writes we are much more likely to get large preallocations exactly when we need it to reduce fragementation. For default settings, the size of the initial extents is determined by the number of parallel writers and the amount of memory in the machine. For 4GB RAM and 4 concurrent 32GB file writes: EXT: FILE-OFFSET BLOCK-RANGE AG AG-OFFSET TOTAL 0: [0..1048575]: 1048672..2097247 0 (1048672..2097247) 1048576 1: [1048576..2097151]: 5242976..6291551 0 (5242976..6291551) 1048576 2: [2097152..4194303]: 12583008..14680159 0 (12583008..14680159) 2097152 3: [4194304..8388607]: 25165920..29360223 0 (25165920..29360223) 4194304 4: [8388608..16777215]: 58720352..67108959 0 (58720352..67108959) 8388608 5: [16777216..33554423]: 117440584..134217791 0 (117440584..134217791) 16777208 6: [33554424..50331511]: 184549056..201326143 0 (184549056..201326143) 16777088 7: [50331512..67108599]: 251657408..268434495 0 (251657408..268434495) 16777088 and for 16 concurrent 16GB file writes: EXT: FILE-OFFSET BLOCK-RANGE AG AG-OFFSET TOTAL 0: [0..262143]: 2490472..2752615 0 (2490472..2752615) 262144 1: [262144..524287]: 6291560..6553703 0 (6291560..6553703) 262144 2: [524288..1048575]: 13631592..14155879 0 (13631592..14155879) 524288 3: [1048576..2097151]: 30408808..31457383 0 (30408808..31457383) 1048576 4: [2097152..4194303]: 52428904..54526055 0 (52428904..54526055) 2097152 5: [4194304..8388607]: 104857704..109052007 0 (104857704..109052007) 4194304 6: [8388608..16777215]: 209715304..218103911 0 (209715304..218103911) 8388608 7: [16777216..33554423]: 452984848..469762055 0 (452984848..469762055) 16777208 Because it is hard to take back specualtive preallocation, cases where there are large slow growing log files on a nearly full filesystem may cause premature ENOSPC. Hence as the filesystem nears full, the maximum dynamic prealloc size іs reduced according to this table (based on 4k block size): freespace max prealloc size >5% full extent (8GB) 4-5% 2GB (8GB >> 2) 3-4% 1GB (8GB >> 3) 2-3% 512MB (8GB >> 4) 1-2% 256MB (8GB >> 5) <1% 128MB (8GB >> 6) This should reduce the amount of space held in speculative preallocation for such cases. The allocsize mount option turns off the dynamic behaviour and fixes the prealloc size to whatever the mount option specifies. i.e. the behaviour is unchanged. Signed-off-by: Dave Chinner <dchinner@redhat.com>
2011-01-04 08:35:03 +08:00
if (alloc_blocks < mp->m_writeio_blocks)
alloc_blocks = mp->m_writeio_blocks;
trace_xfs_iomap_prealloc_size(ip, alloc_blocks, shift,
mp->m_writeio_blocks);
xfs: dynamic speculative EOF preallocation Currently the size of the speculative preallocation during delayed allocation is fixed by either the allocsize mount option of a default size. We are seeing a lot of cases where we need to recommend using the allocsize mount option to prevent fragmentation when buffered writes land in the same AG. Rather than using a fixed preallocation size by default (up to 64k), make it dynamic by basing it on the current inode size. That way the EOF preallocation will increase as the file size increases. Hence for streaming writes we are much more likely to get large preallocations exactly when we need it to reduce fragementation. For default settings, the size of the initial extents is determined by the number of parallel writers and the amount of memory in the machine. For 4GB RAM and 4 concurrent 32GB file writes: EXT: FILE-OFFSET BLOCK-RANGE AG AG-OFFSET TOTAL 0: [0..1048575]: 1048672..2097247 0 (1048672..2097247) 1048576 1: [1048576..2097151]: 5242976..6291551 0 (5242976..6291551) 1048576 2: [2097152..4194303]: 12583008..14680159 0 (12583008..14680159) 2097152 3: [4194304..8388607]: 25165920..29360223 0 (25165920..29360223) 4194304 4: [8388608..16777215]: 58720352..67108959 0 (58720352..67108959) 8388608 5: [16777216..33554423]: 117440584..134217791 0 (117440584..134217791) 16777208 6: [33554424..50331511]: 184549056..201326143 0 (184549056..201326143) 16777088 7: [50331512..67108599]: 251657408..268434495 0 (251657408..268434495) 16777088 and for 16 concurrent 16GB file writes: EXT: FILE-OFFSET BLOCK-RANGE AG AG-OFFSET TOTAL 0: [0..262143]: 2490472..2752615 0 (2490472..2752615) 262144 1: [262144..524287]: 6291560..6553703 0 (6291560..6553703) 262144 2: [524288..1048575]: 13631592..14155879 0 (13631592..14155879) 524288 3: [1048576..2097151]: 30408808..31457383 0 (30408808..31457383) 1048576 4: [2097152..4194303]: 52428904..54526055 0 (52428904..54526055) 2097152 5: [4194304..8388607]: 104857704..109052007 0 (104857704..109052007) 4194304 6: [8388608..16777215]: 209715304..218103911 0 (209715304..218103911) 8388608 7: [16777216..33554423]: 452984848..469762055 0 (452984848..469762055) 16777208 Because it is hard to take back specualtive preallocation, cases where there are large slow growing log files on a nearly full filesystem may cause premature ENOSPC. Hence as the filesystem nears full, the maximum dynamic prealloc size іs reduced according to this table (based on 4k block size): freespace max prealloc size >5% full extent (8GB) 4-5% 2GB (8GB >> 2) 3-4% 1GB (8GB >> 3) 2-3% 512MB (8GB >> 4) 1-2% 256MB (8GB >> 5) <1% 128MB (8GB >> 6) This should reduce the amount of space held in speculative preallocation for such cases. The allocsize mount option turns off the dynamic behaviour and fixes the prealloc size to whatever the mount option specifies. i.e. the behaviour is unchanged. Signed-off-by: Dave Chinner <dchinner@redhat.com>
2011-01-04 08:35:03 +08:00
return alloc_blocks;
}
static int
xfs_file_iomap_begin_delay(
struct inode *inode,
loff_t offset,
loff_t count,
unsigned flags,
struct iomap *iomap)
{
struct xfs_inode *ip = XFS_I(inode);
struct xfs_mount *mp = ip->i_mount;
xfs_fileoff_t offset_fsb = XFS_B_TO_FSBT(mp, offset);
xfs_fileoff_t maxbytes_fsb =
XFS_B_TO_FSB(mp, mp->m_super->s_maxbytes);
xfs_fileoff_t end_fsb;
struct xfs_bmbt_irec imap, cmap;
struct xfs_iext_cursor icur, ccur;
xfs_fsblock_t prealloc_blocks = 0;
bool eof = false, cow_eof = false, shared = false;
int whichfork = XFS_DATA_FORK;
int error = 0;
ASSERT(!XFS_IS_REALTIME_INODE(ip));
ASSERT(!xfs_get_extsz_hint(ip));
xfs_ilock(ip, XFS_ILOCK_EXCL);
if (unlikely(XFS_TEST_ERROR(
(XFS_IFORK_FORMAT(ip, XFS_DATA_FORK) != XFS_DINODE_FMT_EXTENTS &&
XFS_IFORK_FORMAT(ip, XFS_DATA_FORK) != XFS_DINODE_FMT_BTREE),
mp, XFS_ERRTAG_BMAPIFORMAT))) {
XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, mp);
error = -EFSCORRUPTED;
goto out_unlock;
}
xfs: limit speculative prealloc size on sparse files Speculative preallocation based on the current file size works well for contiguous files, but is sub-optimal for sparse files where the EOF preallocation can fill holes and result in large amounts of zeros being written when it is not necessary. The algorithm is modified to prevent EOF speculative preallocation from triggering larger allocations on IO patterns of truncate--to-zero-seek-write-seek-write-.... which results in non-sparse files for large files. This, unfortunately, is the way cp now behaves when copying sparse files and so needs to be fixed. What this code does is that it looks at the existing extent adjacent to the current EOF and if it determines that it is a hole we disable speculative preallocation altogether. To avoid the next write from doing a large prealloc, it takes the size of subsequent preallocations from the current size of the existing EOF extent. IOWs, if you leave a hole in the file, it resets preallocation behaviour to the same as if it was a zero size file. Example new behaviour: $ xfs_io -f -c "pwrite 0 31m" \ -c "pwrite 33m 1m" \ -c "pwrite 128m 1m" \ -c "fiemap -v" /mnt/scratch/blah wrote 32505856/32505856 bytes at offset 0 31 MiB, 7936 ops; 0.0000 sec (1.608 GiB/sec and 421432.7439 ops/sec) wrote 1048576/1048576 bytes at offset 34603008 1 MiB, 256 ops; 0.0000 sec (1.462 GiB/sec and 383233.5329 ops/sec) wrote 1048576/1048576 bytes at offset 134217728 1 MiB, 256 ops; 0.0000 sec (1.719 GiB/sec and 450704.2254 ops/sec) /mnt/scratch/blah: EXT: FILE-OFFSET BLOCK-RANGE TOTAL FLAGS 0: [0..65535]: 96..65631 65536 0x0 1: [65536..67583]: hole 2048 2: [67584..69631]: 67680..69727 2048 0x0 3: [69632..262143]: hole 192512 4: [262144..264191]: 262240..264287 2048 0x1 Signed-off-by: Dave Chinner <dchinner@redhat.com> Reviewed-by: Mark Tinguely <tinguely@sgi.com> Reviewed-by: Brian Foster <bfoster@redhat.com> Signed-off-by: Ben Myers <bpm@sgi.com>
2013-02-11 13:05:01 +08:00
XFS_STATS_INC(mp, xs_blk_mapw);
xfs: dynamic speculative EOF preallocation Currently the size of the speculative preallocation during delayed allocation is fixed by either the allocsize mount option of a default size. We are seeing a lot of cases where we need to recommend using the allocsize mount option to prevent fragmentation when buffered writes land in the same AG. Rather than using a fixed preallocation size by default (up to 64k), make it dynamic by basing it on the current inode size. That way the EOF preallocation will increase as the file size increases. Hence for streaming writes we are much more likely to get large preallocations exactly when we need it to reduce fragementation. For default settings, the size of the initial extents is determined by the number of parallel writers and the amount of memory in the machine. For 4GB RAM and 4 concurrent 32GB file writes: EXT: FILE-OFFSET BLOCK-RANGE AG AG-OFFSET TOTAL 0: [0..1048575]: 1048672..2097247 0 (1048672..2097247) 1048576 1: [1048576..2097151]: 5242976..6291551 0 (5242976..6291551) 1048576 2: [2097152..4194303]: 12583008..14680159 0 (12583008..14680159) 2097152 3: [4194304..8388607]: 25165920..29360223 0 (25165920..29360223) 4194304 4: [8388608..16777215]: 58720352..67108959 0 (58720352..67108959) 8388608 5: [16777216..33554423]: 117440584..134217791 0 (117440584..134217791) 16777208 6: [33554424..50331511]: 184549056..201326143 0 (184549056..201326143) 16777088 7: [50331512..67108599]: 251657408..268434495 0 (251657408..268434495) 16777088 and for 16 concurrent 16GB file writes: EXT: FILE-OFFSET BLOCK-RANGE AG AG-OFFSET TOTAL 0: [0..262143]: 2490472..2752615 0 (2490472..2752615) 262144 1: [262144..524287]: 6291560..6553703 0 (6291560..6553703) 262144 2: [524288..1048575]: 13631592..14155879 0 (13631592..14155879) 524288 3: [1048576..2097151]: 30408808..31457383 0 (30408808..31457383) 1048576 4: [2097152..4194303]: 52428904..54526055 0 (52428904..54526055) 2097152 5: [4194304..8388607]: 104857704..109052007 0 (104857704..109052007) 4194304 6: [8388608..16777215]: 209715304..218103911 0 (209715304..218103911) 8388608 7: [16777216..33554423]: 452984848..469762055 0 (452984848..469762055) 16777208 Because it is hard to take back specualtive preallocation, cases where there are large slow growing log files on a nearly full filesystem may cause premature ENOSPC. Hence as the filesystem nears full, the maximum dynamic prealloc size іs reduced according to this table (based on 4k block size): freespace max prealloc size >5% full extent (8GB) 4-5% 2GB (8GB >> 2) 3-4% 1GB (8GB >> 3) 2-3% 512MB (8GB >> 4) 1-2% 256MB (8GB >> 5) <1% 128MB (8GB >> 6) This should reduce the amount of space held in speculative preallocation for such cases. The allocsize mount option turns off the dynamic behaviour and fixes the prealloc size to whatever the mount option specifies. i.e. the behaviour is unchanged. Signed-off-by: Dave Chinner <dchinner@redhat.com>
2011-01-04 08:35:03 +08:00
if (!(ip->i_df.if_flags & XFS_IFEXTENTS)) {
error = xfs_iread_extents(NULL, ip, XFS_DATA_FORK);
if (error)
goto out_unlock;
}
end_fsb = min(XFS_B_TO_FSB(mp, offset + count), maxbytes_fsb);
/*
* Search the data fork fork first to look up our source mapping. We
* always need the data fork map, as we have to return it to the
* iomap code so that the higher level write code can read data in to
* perform read-modify-write cycles for unaligned writes.
*/
eof = !xfs_iext_lookup_extent(ip, &ip->i_df, offset_fsb, &icur, &imap);
if (eof)
imap.br_startoff = end_fsb; /* fake hole until the end */
/* We never need to allocate blocks for zeroing a hole. */
if ((flags & IOMAP_ZERO) && imap.br_startoff > offset_fsb) {
xfs_hole_to_iomap(ip, iomap, offset_fsb, imap.br_startoff);
goto out_unlock;
}
/*
* Search the COW fork extent list even if we did not find a data fork
* extent. This serves two purposes: first this implements the
* speculative preallocation using cowextsize, so that we also unshare
* block adjacent to shared blocks instead of just the shared blocks
* themselves. Second the lookup in the extent list is generally faster
* than going out to the shared extent tree.
*/
xfs: introduce an always_cow mode Add a mode where XFS never overwrites existing blocks in place. This is to aid debugging our COW code, and also put infatructure in place for things like possible future support for zoned block devices, which can't support overwrites. This mode is enabled globally by doing a: echo 1 > /sys/fs/xfs/debug/always_cow Note that the parameter is global to allow running all tests in xfstests easily in this mode, which would not easily be possible with a per-fs sysfs file. In always_cow mode persistent preallocations are disabled, and fallocate will fail when called with a 0 mode (with our without FALLOC_FL_KEEP_SIZE), and not create unwritten extent for zeroed space when called with FALLOC_FL_ZERO_RANGE or FALLOC_FL_UNSHARE_RANGE. There are a few interesting xfstests failures when run in always_cow mode: - generic/392 fails because the bytes used in the file used to test hole punch recovery are less after the log replay. This is because the blocks written and then punched out are only freed with a delay due to the logging mechanism. - xfs/170 will fail as the already fragile file streams mechanism doesn't seem to interact well with the COW allocator - xfs/180 xfs/182 xfs/192 xfs/198 xfs/204 and xfs/208 will claim the file system is badly fragmented, but there is not much we can do to avoid that when always writing out of place - xfs/205 fails because overwriting a file in always_cow mode will require new space allocation and the assumption in the test thus don't work anymore. - xfs/326 fails to modify the file at all in always_cow mode after injecting the refcount error, leading to an unexpected md5sum after the remount, but that again is expected Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com> Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
2019-02-19 01:38:49 +08:00
if (xfs_is_cow_inode(ip)) {
if (!ip->i_cowfp) {
ASSERT(!xfs_is_reflink_inode(ip));
xfs_ifork_init_cow(ip);
}
cow_eof = !xfs_iext_lookup_extent(ip, ip->i_cowfp, offset_fsb,
&ccur, &cmap);
if (!cow_eof && cmap.br_startoff <= offset_fsb) {
trace_xfs_reflink_cow_found(ip, &cmap);
whichfork = XFS_COW_FORK;
goto done;
}
}
if (imap.br_startoff <= offset_fsb) {
/*
* For reflink files we may need a delalloc reservation when
* overwriting shared extents. This includes zeroing of
* existing extents that contain data.
*/
xfs: introduce an always_cow mode Add a mode where XFS never overwrites existing blocks in place. This is to aid debugging our COW code, and also put infatructure in place for things like possible future support for zoned block devices, which can't support overwrites. This mode is enabled globally by doing a: echo 1 > /sys/fs/xfs/debug/always_cow Note that the parameter is global to allow running all tests in xfstests easily in this mode, which would not easily be possible with a per-fs sysfs file. In always_cow mode persistent preallocations are disabled, and fallocate will fail when called with a 0 mode (with our without FALLOC_FL_KEEP_SIZE), and not create unwritten extent for zeroed space when called with FALLOC_FL_ZERO_RANGE or FALLOC_FL_UNSHARE_RANGE. There are a few interesting xfstests failures when run in always_cow mode: - generic/392 fails because the bytes used in the file used to test hole punch recovery are less after the log replay. This is because the blocks written and then punched out are only freed with a delay due to the logging mechanism. - xfs/170 will fail as the already fragile file streams mechanism doesn't seem to interact well with the COW allocator - xfs/180 xfs/182 xfs/192 xfs/198 xfs/204 and xfs/208 will claim the file system is badly fragmented, but there is not much we can do to avoid that when always writing out of place - xfs/205 fails because overwriting a file in always_cow mode will require new space allocation and the assumption in the test thus don't work anymore. - xfs/326 fails to modify the file at all in always_cow mode after injecting the refcount error, leading to an unexpected md5sum after the remount, but that again is expected Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com> Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
2019-02-19 01:38:49 +08:00
if (!xfs_is_cow_inode(ip) ||
((flags & IOMAP_ZERO) && imap.br_state != XFS_EXT_NORM)) {
trace_xfs_iomap_found(ip, offset, count, XFS_DATA_FORK,
&imap);
goto done;
}
xfs_trim_extent(&imap, offset_fsb, end_fsb - offset_fsb);
/* Trim the mapping to the nearest shared extent boundary. */
xfs: introduce an always_cow mode Add a mode where XFS never overwrites existing blocks in place. This is to aid debugging our COW code, and also put infatructure in place for things like possible future support for zoned block devices, which can't support overwrites. This mode is enabled globally by doing a: echo 1 > /sys/fs/xfs/debug/always_cow Note that the parameter is global to allow running all tests in xfstests easily in this mode, which would not easily be possible with a per-fs sysfs file. In always_cow mode persistent preallocations are disabled, and fallocate will fail when called with a 0 mode (with our without FALLOC_FL_KEEP_SIZE), and not create unwritten extent for zeroed space when called with FALLOC_FL_ZERO_RANGE or FALLOC_FL_UNSHARE_RANGE. There are a few interesting xfstests failures when run in always_cow mode: - generic/392 fails because the bytes used in the file used to test hole punch recovery are less after the log replay. This is because the blocks written and then punched out are only freed with a delay due to the logging mechanism. - xfs/170 will fail as the already fragile file streams mechanism doesn't seem to interact well with the COW allocator - xfs/180 xfs/182 xfs/192 xfs/198 xfs/204 and xfs/208 will claim the file system is badly fragmented, but there is not much we can do to avoid that when always writing out of place - xfs/205 fails because overwriting a file in always_cow mode will require new space allocation and the assumption in the test thus don't work anymore. - xfs/326 fails to modify the file at all in always_cow mode after injecting the refcount error, leading to an unexpected md5sum after the remount, but that again is expected Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com> Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
2019-02-19 01:38:49 +08:00
error = xfs_inode_need_cow(ip, &imap, &shared);
if (error)
goto out_unlock;
/* Not shared? Just report the (potentially capped) extent. */
if (!shared) {
trace_xfs_iomap_found(ip, offset, count, XFS_DATA_FORK,
&imap);
goto done;
}
/*
* Fork all the shared blocks from our write offset until the
* end of the extent.
*/
whichfork = XFS_COW_FORK;
end_fsb = imap.br_startoff + imap.br_blockcount;
} else {
/*
* We cap the maximum length we map here to MAX_WRITEBACK_PAGES
* pages to keep the chunks of work done where somewhat
* symmetric with the work writeback does. This is a completely
* arbitrary number pulled out of thin air.
*
* Note that the values needs to be less than 32-bits wide until
* the lower level functions are updated.
*/
count = min_t(loff_t, count, 1024 * PAGE_SIZE);
end_fsb = min(XFS_B_TO_FSB(mp, offset + count), maxbytes_fsb);
xfs: introduce an always_cow mode Add a mode where XFS never overwrites existing blocks in place. This is to aid debugging our COW code, and also put infatructure in place for things like possible future support for zoned block devices, which can't support overwrites. This mode is enabled globally by doing a: echo 1 > /sys/fs/xfs/debug/always_cow Note that the parameter is global to allow running all tests in xfstests easily in this mode, which would not easily be possible with a per-fs sysfs file. In always_cow mode persistent preallocations are disabled, and fallocate will fail when called with a 0 mode (with our without FALLOC_FL_KEEP_SIZE), and not create unwritten extent for zeroed space when called with FALLOC_FL_ZERO_RANGE or FALLOC_FL_UNSHARE_RANGE. There are a few interesting xfstests failures when run in always_cow mode: - generic/392 fails because the bytes used in the file used to test hole punch recovery are less after the log replay. This is because the blocks written and then punched out are only freed with a delay due to the logging mechanism. - xfs/170 will fail as the already fragile file streams mechanism doesn't seem to interact well with the COW allocator - xfs/180 xfs/182 xfs/192 xfs/198 xfs/204 and xfs/208 will claim the file system is badly fragmented, but there is not much we can do to avoid that when always writing out of place - xfs/205 fails because overwriting a file in always_cow mode will require new space allocation and the assumption in the test thus don't work anymore. - xfs/326 fails to modify the file at all in always_cow mode after injecting the refcount error, leading to an unexpected md5sum after the remount, but that again is expected Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com> Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
2019-02-19 01:38:49 +08:00
if (xfs_is_always_cow_inode(ip))
whichfork = XFS_COW_FORK;
}
error = xfs_qm_dqattach_locked(ip, false);
if (error)
goto out_unlock;
xfs: introduce an always_cow mode Add a mode where XFS never overwrites existing blocks in place. This is to aid debugging our COW code, and also put infatructure in place for things like possible future support for zoned block devices, which can't support overwrites. This mode is enabled globally by doing a: echo 1 > /sys/fs/xfs/debug/always_cow Note that the parameter is global to allow running all tests in xfstests easily in this mode, which would not easily be possible with a per-fs sysfs file. In always_cow mode persistent preallocations are disabled, and fallocate will fail when called with a 0 mode (with our without FALLOC_FL_KEEP_SIZE), and not create unwritten extent for zeroed space when called with FALLOC_FL_ZERO_RANGE or FALLOC_FL_UNSHARE_RANGE. There are a few interesting xfstests failures when run in always_cow mode: - generic/392 fails because the bytes used in the file used to test hole punch recovery are less after the log replay. This is because the blocks written and then punched out are only freed with a delay due to the logging mechanism. - xfs/170 will fail as the already fragile file streams mechanism doesn't seem to interact well with the COW allocator - xfs/180 xfs/182 xfs/192 xfs/198 xfs/204 and xfs/208 will claim the file system is badly fragmented, but there is not much we can do to avoid that when always writing out of place - xfs/205 fails because overwriting a file in always_cow mode will require new space allocation and the assumption in the test thus don't work anymore. - xfs/326 fails to modify the file at all in always_cow mode after injecting the refcount error, leading to an unexpected md5sum after the remount, but that again is expected Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com> Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
2019-02-19 01:38:49 +08:00
if (eof) {
prealloc_blocks = xfs_iomap_prealloc_size(ip, whichfork, offset,
count, &icur);
if (prealloc_blocks) {
xfs_extlen_t align;
xfs_off_t end_offset;
xfs_fileoff_t p_end_fsb;
end_offset = XFS_WRITEIO_ALIGN(mp, offset + count - 1);
p_end_fsb = XFS_B_TO_FSBT(mp, end_offset) +
prealloc_blocks;
align = xfs_eof_alignment(ip, 0);
if (align)
p_end_fsb = roundup_64(p_end_fsb, align);
p_end_fsb = min(p_end_fsb, maxbytes_fsb);
ASSERT(p_end_fsb > offset_fsb);
prealloc_blocks = p_end_fsb - end_fsb;
}
}
retry:
error = xfs_bmapi_reserve_delalloc(ip, whichfork, offset_fsb,
end_fsb - offset_fsb, prealloc_blocks,
whichfork == XFS_DATA_FORK ? &imap : &cmap,
whichfork == XFS_DATA_FORK ? &icur : &ccur,
whichfork == XFS_DATA_FORK ? eof : cow_eof);
xfs: dynamic speculative EOF preallocation Currently the size of the speculative preallocation during delayed allocation is fixed by either the allocsize mount option of a default size. We are seeing a lot of cases where we need to recommend using the allocsize mount option to prevent fragmentation when buffered writes land in the same AG. Rather than using a fixed preallocation size by default (up to 64k), make it dynamic by basing it on the current inode size. That way the EOF preallocation will increase as the file size increases. Hence for streaming writes we are much more likely to get large preallocations exactly when we need it to reduce fragementation. For default settings, the size of the initial extents is determined by the number of parallel writers and the amount of memory in the machine. For 4GB RAM and 4 concurrent 32GB file writes: EXT: FILE-OFFSET BLOCK-RANGE AG AG-OFFSET TOTAL 0: [0..1048575]: 1048672..2097247 0 (1048672..2097247) 1048576 1: [1048576..2097151]: 5242976..6291551 0 (5242976..6291551) 1048576 2: [2097152..4194303]: 12583008..14680159 0 (12583008..14680159) 2097152 3: [4194304..8388607]: 25165920..29360223 0 (25165920..29360223) 4194304 4: [8388608..16777215]: 58720352..67108959 0 (58720352..67108959) 8388608 5: [16777216..33554423]: 117440584..134217791 0 (117440584..134217791) 16777208 6: [33554424..50331511]: 184549056..201326143 0 (184549056..201326143) 16777088 7: [50331512..67108599]: 251657408..268434495 0 (251657408..268434495) 16777088 and for 16 concurrent 16GB file writes: EXT: FILE-OFFSET BLOCK-RANGE AG AG-OFFSET TOTAL 0: [0..262143]: 2490472..2752615 0 (2490472..2752615) 262144 1: [262144..524287]: 6291560..6553703 0 (6291560..6553703) 262144 2: [524288..1048575]: 13631592..14155879 0 (13631592..14155879) 524288 3: [1048576..2097151]: 30408808..31457383 0 (30408808..31457383) 1048576 4: [2097152..4194303]: 52428904..54526055 0 (52428904..54526055) 2097152 5: [4194304..8388607]: 104857704..109052007 0 (104857704..109052007) 4194304 6: [8388608..16777215]: 209715304..218103911 0 (209715304..218103911) 8388608 7: [16777216..33554423]: 452984848..469762055 0 (452984848..469762055) 16777208 Because it is hard to take back specualtive preallocation, cases where there are large slow growing log files on a nearly full filesystem may cause premature ENOSPC. Hence as the filesystem nears full, the maximum dynamic prealloc size іs reduced according to this table (based on 4k block size): freespace max prealloc size >5% full extent (8GB) 4-5% 2GB (8GB >> 2) 3-4% 1GB (8GB >> 3) 2-3% 512MB (8GB >> 4) 1-2% 256MB (8GB >> 5) <1% 128MB (8GB >> 6) This should reduce the amount of space held in speculative preallocation for such cases. The allocsize mount option turns off the dynamic behaviour and fixes the prealloc size to whatever the mount option specifies. i.e. the behaviour is unchanged. Signed-off-by: Dave Chinner <dchinner@redhat.com>
2011-01-04 08:35:03 +08:00
switch (error) {
case 0:
break;
case -ENOSPC:
case -EDQUOT:
/* retry without any preallocation */
xfs: event tracing support Convert the old xfs tracing support that could only be used with the out of tree kdb and xfsidbg patches to use the generic event tracer. To use it make sure CONFIG_EVENT_TRACING is enabled and then enable all xfs trace channels by: echo 1 > /sys/kernel/debug/tracing/events/xfs/enable or alternatively enable single events by just doing the same in one event subdirectory, e.g. echo 1 > /sys/kernel/debug/tracing/events/xfs/xfs_ihold/enable or set more complex filters, etc. In Documentation/trace/events.txt all this is desctribed in more detail. To reads the events do a cat /sys/kernel/debug/tracing/trace Compared to the last posting this patch converts the tracing mostly to the one tracepoint per callsite model that other users of the new tracing facility also employ. This allows a very fine-grained control of the tracing, a cleaner output of the traces and also enables the perf tool to use each tracepoint as a virtual performance counter, allowing us to e.g. count how often certain workloads git various spots in XFS. Take a look at http://lwn.net/Articles/346470/ for some examples. Also the btree tracing isn't included at all yet, as it will require additional core tracing features not in mainline yet, I plan to deliver it later. And the really nice thing about this patch is that it actually removes many lines of code while adding this nice functionality: fs/xfs/Makefile | 8 fs/xfs/linux-2.6/xfs_acl.c | 1 fs/xfs/linux-2.6/xfs_aops.c | 52 - fs/xfs/linux-2.6/xfs_aops.h | 2 fs/xfs/linux-2.6/xfs_buf.c | 117 +-- fs/xfs/linux-2.6/xfs_buf.h | 33 fs/xfs/linux-2.6/xfs_fs_subr.c | 3 fs/xfs/linux-2.6/xfs_ioctl.c | 1 fs/xfs/linux-2.6/xfs_ioctl32.c | 1 fs/xfs/linux-2.6/xfs_iops.c | 1 fs/xfs/linux-2.6/xfs_linux.h | 1 fs/xfs/linux-2.6/xfs_lrw.c | 87 -- fs/xfs/linux-2.6/xfs_lrw.h | 45 - fs/xfs/linux-2.6/xfs_super.c | 104 --- fs/xfs/linux-2.6/xfs_super.h | 7 fs/xfs/linux-2.6/xfs_sync.c | 1 fs/xfs/linux-2.6/xfs_trace.c | 75 ++ fs/xfs/linux-2.6/xfs_trace.h | 1369 +++++++++++++++++++++++++++++++++++++++++ fs/xfs/linux-2.6/xfs_vnode.h | 4 fs/xfs/quota/xfs_dquot.c | 110 --- fs/xfs/quota/xfs_dquot.h | 21 fs/xfs/quota/xfs_qm.c | 40 - fs/xfs/quota/xfs_qm_syscalls.c | 4 fs/xfs/support/ktrace.c | 323 --------- fs/xfs/support/ktrace.h | 85 -- fs/xfs/xfs.h | 16 fs/xfs/xfs_ag.h | 14 fs/xfs/xfs_alloc.c | 230 +----- fs/xfs/xfs_alloc.h | 27 fs/xfs/xfs_alloc_btree.c | 1 fs/xfs/xfs_attr.c | 107 --- fs/xfs/xfs_attr.h | 10 fs/xfs/xfs_attr_leaf.c | 14 fs/xfs/xfs_attr_sf.h | 40 - fs/xfs/xfs_bmap.c | 507 +++------------ fs/xfs/xfs_bmap.h | 49 - fs/xfs/xfs_bmap_btree.c | 6 fs/xfs/xfs_btree.c | 5 fs/xfs/xfs_btree_trace.h | 17 fs/xfs/xfs_buf_item.c | 87 -- fs/xfs/xfs_buf_item.h | 20 fs/xfs/xfs_da_btree.c | 3 fs/xfs/xfs_da_btree.h | 7 fs/xfs/xfs_dfrag.c | 2 fs/xfs/xfs_dir2.c | 8 fs/xfs/xfs_dir2_block.c | 20 fs/xfs/xfs_dir2_leaf.c | 21 fs/xfs/xfs_dir2_node.c | 27 fs/xfs/xfs_dir2_sf.c | 26 fs/xfs/xfs_dir2_trace.c | 216 ------ fs/xfs/xfs_dir2_trace.h | 72 -- fs/xfs/xfs_filestream.c | 8 fs/xfs/xfs_fsops.c | 2 fs/xfs/xfs_iget.c | 111 --- fs/xfs/xfs_inode.c | 67 -- fs/xfs/xfs_inode.h | 76 -- fs/xfs/xfs_inode_item.c | 5 fs/xfs/xfs_iomap.c | 85 -- fs/xfs/xfs_iomap.h | 8 fs/xfs/xfs_log.c | 181 +---- fs/xfs/xfs_log_priv.h | 20 fs/xfs/xfs_log_recover.c | 1 fs/xfs/xfs_mount.c | 2 fs/xfs/xfs_quota.h | 8 fs/xfs/xfs_rename.c | 1 fs/xfs/xfs_rtalloc.c | 1 fs/xfs/xfs_rw.c | 3 fs/xfs/xfs_trans.h | 47 + fs/xfs/xfs_trans_buf.c | 62 - fs/xfs/xfs_vnodeops.c | 8 70 files changed, 2151 insertions(+), 2592 deletions(-) Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Alex Elder <aelder@sgi.com>
2009-12-15 07:14:59 +08:00
trace_xfs_delalloc_enospc(ip, offset, count);
if (prealloc_blocks) {
prealloc_blocks = 0;
xfs: xfs_sync_data is redundant. We don't do any data writeback from XFS any more - the VFS is completely responsible for that, including for freeze. We can replace the remaining caller with a VFS level function that achieves the same thing, but without conflicting with current writeback work. This means we can remove the flush_work and xfs_flush_inodes() - the VFS functionality completely replaces the internal flush queue for doing this writeback work in a separate context to avoid stack overruns. This does have one complication - it cannot be called with page locks held. Hence move the flushing of delalloc space when ENOSPC occurs back up into xfs_file_aio_buffered_write when we don't hold any locks that will stall writeback. Unfortunately, writeback_inodes_sb_if_idle() is not sufficient to trigger delalloc conversion fast enough to prevent spurious ENOSPC whent here are hundreds of writers, thousands of small files and GBs of free RAM. Hence we need to use sync_sb_inodes() to block callers while we wait for writeback like the previous xfs_flush_inodes implementation did. That means we have to hold the s_umount lock here, but because this call can nest inside i_mutex (the parent directory in the create case, held by the VFS), we have to use down_read_trylock() to avoid potential deadlocks. In practice, this trylock will succeed on almost every attempt as unmount/remount type operations are exceedingly rare. Note: we always need to pass a count of zero to generic_file_buffered_write() as the previously written byte count. We only do this by accident before this patch by the virtue of ret always being zero when there are no errors. Make this explicit rather than needing to specifically zero ret in the ENOSPC retry case. Signed-off-by: Dave Chinner <dchinner@redhat.com> Tested-by: Brian Foster <bfoster@redhat.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Ben Myers <bpm@sgi.com>
2012-10-08 18:56:04 +08:00
goto retry;
xfs: dynamic speculative EOF preallocation Currently the size of the speculative preallocation during delayed allocation is fixed by either the allocsize mount option of a default size. We are seeing a lot of cases where we need to recommend using the allocsize mount option to prevent fragmentation when buffered writes land in the same AG. Rather than using a fixed preallocation size by default (up to 64k), make it dynamic by basing it on the current inode size. That way the EOF preallocation will increase as the file size increases. Hence for streaming writes we are much more likely to get large preallocations exactly when we need it to reduce fragementation. For default settings, the size of the initial extents is determined by the number of parallel writers and the amount of memory in the machine. For 4GB RAM and 4 concurrent 32GB file writes: EXT: FILE-OFFSET BLOCK-RANGE AG AG-OFFSET TOTAL 0: [0..1048575]: 1048672..2097247 0 (1048672..2097247) 1048576 1: [1048576..2097151]: 5242976..6291551 0 (5242976..6291551) 1048576 2: [2097152..4194303]: 12583008..14680159 0 (12583008..14680159) 2097152 3: [4194304..8388607]: 25165920..29360223 0 (25165920..29360223) 4194304 4: [8388608..16777215]: 58720352..67108959 0 (58720352..67108959) 8388608 5: [16777216..33554423]: 117440584..134217791 0 (117440584..134217791) 16777208 6: [33554424..50331511]: 184549056..201326143 0 (184549056..201326143) 16777088 7: [50331512..67108599]: 251657408..268434495 0 (251657408..268434495) 16777088 and for 16 concurrent 16GB file writes: EXT: FILE-OFFSET BLOCK-RANGE AG AG-OFFSET TOTAL 0: [0..262143]: 2490472..2752615 0 (2490472..2752615) 262144 1: [262144..524287]: 6291560..6553703 0 (6291560..6553703) 262144 2: [524288..1048575]: 13631592..14155879 0 (13631592..14155879) 524288 3: [1048576..2097151]: 30408808..31457383 0 (30408808..31457383) 1048576 4: [2097152..4194303]: 52428904..54526055 0 (52428904..54526055) 2097152 5: [4194304..8388607]: 104857704..109052007 0 (104857704..109052007) 4194304 6: [8388608..16777215]: 209715304..218103911 0 (209715304..218103911) 8388608 7: [16777216..33554423]: 452984848..469762055 0 (452984848..469762055) 16777208 Because it is hard to take back specualtive preallocation, cases where there are large slow growing log files on a nearly full filesystem may cause premature ENOSPC. Hence as the filesystem nears full, the maximum dynamic prealloc size іs reduced according to this table (based on 4k block size): freespace max prealloc size >5% full extent (8GB) 4-5% 2GB (8GB >> 2) 3-4% 1GB (8GB >> 3) 2-3% 512MB (8GB >> 4) 1-2% 256MB (8GB >> 5) <1% 128MB (8GB >> 6) This should reduce the amount of space held in speculative preallocation for such cases. The allocsize mount option turns off the dynamic behaviour and fixes the prealloc size to whatever the mount option specifies. i.e. the behaviour is unchanged. Signed-off-by: Dave Chinner <dchinner@redhat.com>
2011-01-04 08:35:03 +08:00
}
/*FALLTHRU*/
default:
goto out_unlock;
}
xfs: use iomap new flag for newly allocated delalloc blocks Commit fa7f138 ("xfs: clear delalloc and cache on buffered write failure") fixed one regression in the iomap error handling code and exposed another. The fundamental problem is that if a buffered write is a rewrite of preexisting delalloc blocks and the write fails, the failure handling code can punch out preexisting blocks with valid file data. This was reproduced directly by sub-block writes in the LTP kernel/syscalls/write/write03 test. A first 100 byte write allocates a single block in a file. A subsequent 100 byte write fails and punches out the block, including the data successfully written by the previous write. To address this problem, update the ->iomap_begin() handler to distinguish newly allocated delalloc blocks from preexisting delalloc blocks via the IOMAP_F_NEW flag. Use this flag in the ->iomap_end() handler to decide when a failed or short write should punch out delalloc blocks. This introduces the subtle requirement that ->iomap_begin() should never combine newly allocated delalloc blocks with existing blocks in the resulting iomap descriptor. This can occur when a new delalloc reservation merges with a neighboring extent that is part of the current write, for example. Therefore, drop the post-allocation extent lookup from xfs_bmapi_reserve_delalloc() and just return the record inserted into the fork. This ensures only new blocks are returned and thus that preexisting delalloc blocks are always handled as "found" blocks and not punched out on a failed rewrite. Reported-by: Xiong Zhou <xzhou@redhat.com> Signed-off-by: Brian Foster <bfoster@redhat.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com> Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
2017-03-09 01:58:08 +08:00
/*
* Flag newly allocated delalloc blocks with IOMAP_F_NEW so we punch
* them out if the write happens to fail.
*/
iomap->flags |= IOMAP_F_NEW;
trace_xfs_iomap_alloc(ip, offset, count, whichfork,
whichfork == XFS_DATA_FORK ? &imap : &cmap);
done:
if (whichfork == XFS_COW_FORK) {
if (imap.br_startoff > offset_fsb) {
xfs_trim_extent(&cmap, offset_fsb,
imap.br_startoff - offset_fsb);
error = xfs_bmbt_to_iomap(ip, iomap, &cmap, true);
goto out_unlock;
}
/* ensure we only report blocks we have a reservation for */
xfs_trim_extent(&imap, cmap.br_startoff, cmap.br_blockcount);
shared = true;
}
error = xfs_bmbt_to_iomap(ip, iomap, &imap, shared);
out_unlock:
xfs_iunlock(ip, XFS_ILOCK_EXCL);
return error;
}
int
xfs_iomap_write_unwritten(
xfs_inode_t *ip,
xfs_off_t offset,
xfs_off_t count,
bool update_isize)
{
xfs_mount_t *mp = ip->i_mount;
xfs_fileoff_t offset_fsb;
xfs_filblks_t count_fsb;
xfs_filblks_t numblks_fsb;
int nimaps;
xfs_trans_t *tp;
xfs_bmbt_irec_t imap;
struct inode *inode = VFS_I(ip);
xfs_fsize_t i_size;
uint resblks;
int error;
xfs: event tracing support Convert the old xfs tracing support that could only be used with the out of tree kdb and xfsidbg patches to use the generic event tracer. To use it make sure CONFIG_EVENT_TRACING is enabled and then enable all xfs trace channels by: echo 1 > /sys/kernel/debug/tracing/events/xfs/enable or alternatively enable single events by just doing the same in one event subdirectory, e.g. echo 1 > /sys/kernel/debug/tracing/events/xfs/xfs_ihold/enable or set more complex filters, etc. In Documentation/trace/events.txt all this is desctribed in more detail. To reads the events do a cat /sys/kernel/debug/tracing/trace Compared to the last posting this patch converts the tracing mostly to the one tracepoint per callsite model that other users of the new tracing facility also employ. This allows a very fine-grained control of the tracing, a cleaner output of the traces and also enables the perf tool to use each tracepoint as a virtual performance counter, allowing us to e.g. count how often certain workloads git various spots in XFS. Take a look at http://lwn.net/Articles/346470/ for some examples. Also the btree tracing isn't included at all yet, as it will require additional core tracing features not in mainline yet, I plan to deliver it later. And the really nice thing about this patch is that it actually removes many lines of code while adding this nice functionality: fs/xfs/Makefile | 8 fs/xfs/linux-2.6/xfs_acl.c | 1 fs/xfs/linux-2.6/xfs_aops.c | 52 - fs/xfs/linux-2.6/xfs_aops.h | 2 fs/xfs/linux-2.6/xfs_buf.c | 117 +-- fs/xfs/linux-2.6/xfs_buf.h | 33 fs/xfs/linux-2.6/xfs_fs_subr.c | 3 fs/xfs/linux-2.6/xfs_ioctl.c | 1 fs/xfs/linux-2.6/xfs_ioctl32.c | 1 fs/xfs/linux-2.6/xfs_iops.c | 1 fs/xfs/linux-2.6/xfs_linux.h | 1 fs/xfs/linux-2.6/xfs_lrw.c | 87 -- fs/xfs/linux-2.6/xfs_lrw.h | 45 - fs/xfs/linux-2.6/xfs_super.c | 104 --- fs/xfs/linux-2.6/xfs_super.h | 7 fs/xfs/linux-2.6/xfs_sync.c | 1 fs/xfs/linux-2.6/xfs_trace.c | 75 ++ fs/xfs/linux-2.6/xfs_trace.h | 1369 +++++++++++++++++++++++++++++++++++++++++ fs/xfs/linux-2.6/xfs_vnode.h | 4 fs/xfs/quota/xfs_dquot.c | 110 --- fs/xfs/quota/xfs_dquot.h | 21 fs/xfs/quota/xfs_qm.c | 40 - fs/xfs/quota/xfs_qm_syscalls.c | 4 fs/xfs/support/ktrace.c | 323 --------- fs/xfs/support/ktrace.h | 85 -- fs/xfs/xfs.h | 16 fs/xfs/xfs_ag.h | 14 fs/xfs/xfs_alloc.c | 230 +----- fs/xfs/xfs_alloc.h | 27 fs/xfs/xfs_alloc_btree.c | 1 fs/xfs/xfs_attr.c | 107 --- fs/xfs/xfs_attr.h | 10 fs/xfs/xfs_attr_leaf.c | 14 fs/xfs/xfs_attr_sf.h | 40 - fs/xfs/xfs_bmap.c | 507 +++------------ fs/xfs/xfs_bmap.h | 49 - fs/xfs/xfs_bmap_btree.c | 6 fs/xfs/xfs_btree.c | 5 fs/xfs/xfs_btree_trace.h | 17 fs/xfs/xfs_buf_item.c | 87 -- fs/xfs/xfs_buf_item.h | 20 fs/xfs/xfs_da_btree.c | 3 fs/xfs/xfs_da_btree.h | 7 fs/xfs/xfs_dfrag.c | 2 fs/xfs/xfs_dir2.c | 8 fs/xfs/xfs_dir2_block.c | 20 fs/xfs/xfs_dir2_leaf.c | 21 fs/xfs/xfs_dir2_node.c | 27 fs/xfs/xfs_dir2_sf.c | 26 fs/xfs/xfs_dir2_trace.c | 216 ------ fs/xfs/xfs_dir2_trace.h | 72 -- fs/xfs/xfs_filestream.c | 8 fs/xfs/xfs_fsops.c | 2 fs/xfs/xfs_iget.c | 111 --- fs/xfs/xfs_inode.c | 67 -- fs/xfs/xfs_inode.h | 76 -- fs/xfs/xfs_inode_item.c | 5 fs/xfs/xfs_iomap.c | 85 -- fs/xfs/xfs_iomap.h | 8 fs/xfs/xfs_log.c | 181 +---- fs/xfs/xfs_log_priv.h | 20 fs/xfs/xfs_log_recover.c | 1 fs/xfs/xfs_mount.c | 2 fs/xfs/xfs_quota.h | 8 fs/xfs/xfs_rename.c | 1 fs/xfs/xfs_rtalloc.c | 1 fs/xfs/xfs_rw.c | 3 fs/xfs/xfs_trans.h | 47 + fs/xfs/xfs_trans_buf.c | 62 - fs/xfs/xfs_vnodeops.c | 8 70 files changed, 2151 insertions(+), 2592 deletions(-) Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Alex Elder <aelder@sgi.com>
2009-12-15 07:14:59 +08:00
trace_xfs_unwritten_convert(ip, offset, count);
offset_fsb = XFS_B_TO_FSBT(mp, offset);
count_fsb = XFS_B_TO_FSB(mp, (xfs_ufsize_t)offset + count);
count_fsb = (xfs_filblks_t)(count_fsb - offset_fsb);
/*
* Reserve enough blocks in this transaction for two complete extent
* btree splits. We may be converting the middle part of an unwritten
* extent and in this case we will insert two new extents in the btree
* each of which could cause a full split.
*
* This reservation amount will be used in the first call to
* xfs_bmbt_split() to select an AG with enough space to satisfy the
* rest of the operation.
*/
resblks = XFS_DIOSTRAT_SPACE_RES(mp, 0) << 1;
do {
/*
* Set up a transaction to convert the range of extents
* from unwritten to real. Do allocations in a loop until
* we have covered the range passed in.
*
* Note that we can't risk to recursing back into the filesystem
* here as we might be asked to write out the same inode that we
* complete here and might deadlock on the iolock.
*/
error = xfs_trans_alloc(mp, &M_RES(mp)->tr_write, resblks, 0,
XFS_TRANS_RESERVE | XFS_TRANS_NOFS, &tp);
if (error)
return error;
xfs_ilock(ip, XFS_ILOCK_EXCL);
xfs_trans_ijoin(tp, ip, 0);
/*
* Modify the unwritten extent state of the buffer.
*/
nimaps = 1;
error = xfs_bmapi_write(tp, ip, offset_fsb, count_fsb,
XFS_BMAPI_CONVERT, resblks, &imap,
&nimaps);
if (error)
goto error_on_bmapi_transaction;
/*
* Log the updated inode size as we go. We have to be careful
* to only log it up to the actual write offset if it is
* halfway into a block.
*/
i_size = XFS_FSB_TO_B(mp, offset_fsb + count_fsb);
if (i_size > offset + count)
i_size = offset + count;
if (update_isize && i_size > i_size_read(inode))
i_size_write(inode, i_size);
i_size = xfs_new_eof(ip, i_size);
if (i_size) {
ip->i_d.di_size = i_size;
xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE);
}
error = xfs_trans_commit(tp);
xfs_iunlock(ip, XFS_ILOCK_EXCL);
if (error)
return error;
if (!(imap.br_startblock || XFS_IS_REALTIME_INODE(ip)))
return xfs_alert_fsblock_zero(ip, &imap);
if ((numblks_fsb = imap.br_blockcount) == 0) {
/*
* The numblks_fsb value should always get
* smaller, otherwise the loop is stuck.
*/
ASSERT(imap.br_blockcount);
break;
}
offset_fsb += numblks_fsb;
count_fsb -= numblks_fsb;
} while (count_fsb > 0);
return 0;
error_on_bmapi_transaction:
xfs_trans_cancel(tp);
xfs_iunlock(ip, XFS_ILOCK_EXCL);
return error;
}
static inline bool
imap_needs_alloc(
struct inode *inode,
struct xfs_bmbt_irec *imap,
int nimaps)
{
return !nimaps ||
imap->br_startblock == HOLESTARTBLOCK ||
imap->br_startblock == DELAYSTARTBLOCK ||
(IS_DAX(inode) && imap->br_state == XFS_EXT_UNWRITTEN);
}
static inline bool
needs_cow_for_zeroing(
struct xfs_bmbt_irec *imap,
int nimaps)
{
return nimaps &&
imap->br_startblock != HOLESTARTBLOCK &&
imap->br_state != XFS_EXT_UNWRITTEN;
}
static int
xfs_ilock_for_iomap(
struct xfs_inode *ip,
unsigned flags,
unsigned *lockmode)
{
unsigned mode = XFS_ILOCK_SHARED;
xfs: recheck reflink state after grabbing ILOCK_SHARED for a write The reflink iflag could have changed since the earlier unlocked check, so if we got ILOCK_SHARED for a write and but we're now a reflink inode we have to switch to ILOCK_EXCL and relock. This helps us avoid blowing lock assertions in things like generic/166: XFS: Assertion failed: xfs_isilocked(ip, XFS_ILOCK_EXCL), file: fs/xfs/xfs_reflink.c, line: 383 WARNING: CPU: 1 PID: 24707 at fs/xfs/xfs_message.c:104 assfail+0x25/0x30 [xfs] Modules linked in: deadline_iosched dm_snapshot dm_bufio ext4 mbcache jbd2 dm_flakey xfs libcrc32c dax_pmem device_dax nd_pmem sch_fq_codel af_packet [last unloaded: scsi_debug] CPU: 1 PID: 24707 Comm: xfs_io Not tainted 4.18.0-rc1-djw #1 Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.10.2-1ubuntu1 04/01/2014 RIP: 0010:assfail+0x25/0x30 [xfs] Code: ff 0f 0b c3 90 66 66 66 66 90 48 89 f1 41 89 d0 48 c7 c6 e8 ef 1b a0 48 89 fa 31 ff e8 54 f9 ff ff 80 3d fd ba 0f 00 00 75 03 <0f> 0b c3 0f 0b 66 0f 1f 44 00 00 66 66 66 66 90 48 63 f6 49 89 f9 RSP: 0018:ffffc90006423ad8 EFLAGS: 00010246 RAX: 0000000000000000 RBX: ffff880030b65e80 RCX: 0000000000000000 RDX: 00000000ffffffc0 RSI: 000000000000000a RDI: ffffffffa01b0447 RBP: ffffc90006423c10 R08: 0000000000000000 R09: 0000000000000000 R10: ffff88003d43fc30 R11: f000000000000000 R12: ffff880077cda000 R13: 0000000000000000 R14: ffffc90006423c30 R15: ffffc90006423bf9 FS: 00007feba8986800(0000) GS:ffff88003ec00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 000000000138ab58 CR3: 000000003d40a000 CR4: 00000000000006a0 Call Trace: xfs_reflink_allocate_cow+0x24c/0x3d0 [xfs] xfs_file_iomap_begin+0x6d2/0xeb0 [xfs] ? iomap_to_fiemap+0x80/0x80 iomap_apply+0x5e/0x130 iomap_dio_rw+0x2e0/0x400 ? iomap_to_fiemap+0x80/0x80 ? xfs_file_dio_aio_write+0x133/0x4a0 [xfs] xfs_file_dio_aio_write+0x133/0x4a0 [xfs] xfs_file_write_iter+0x7b/0xb0 [xfs] __vfs_write+0x16f/0x1f0 vfs_write+0xc8/0x1c0 ksys_pwrite64+0x74/0x90 do_syscall_64+0x56/0x180 entry_SYSCALL_64_after_hwframe+0x49/0xbe Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2018-06-22 14:26:57 +08:00
bool is_write = flags & (IOMAP_WRITE | IOMAP_ZERO);
/*
* COW writes may allocate delalloc space or convert unwritten COW
* extents, so we need to make sure to take the lock exclusively here.
*/
xfs: introduce an always_cow mode Add a mode where XFS never overwrites existing blocks in place. This is to aid debugging our COW code, and also put infatructure in place for things like possible future support for zoned block devices, which can't support overwrites. This mode is enabled globally by doing a: echo 1 > /sys/fs/xfs/debug/always_cow Note that the parameter is global to allow running all tests in xfstests easily in this mode, which would not easily be possible with a per-fs sysfs file. In always_cow mode persistent preallocations are disabled, and fallocate will fail when called with a 0 mode (with our without FALLOC_FL_KEEP_SIZE), and not create unwritten extent for zeroed space when called with FALLOC_FL_ZERO_RANGE or FALLOC_FL_UNSHARE_RANGE. There are a few interesting xfstests failures when run in always_cow mode: - generic/392 fails because the bytes used in the file used to test hole punch recovery are less after the log replay. This is because the blocks written and then punched out are only freed with a delay due to the logging mechanism. - xfs/170 will fail as the already fragile file streams mechanism doesn't seem to interact well with the COW allocator - xfs/180 xfs/182 xfs/192 xfs/198 xfs/204 and xfs/208 will claim the file system is badly fragmented, but there is not much we can do to avoid that when always writing out of place - xfs/205 fails because overwriting a file in always_cow mode will require new space allocation and the assumption in the test thus don't work anymore. - xfs/326 fails to modify the file at all in always_cow mode after injecting the refcount error, leading to an unexpected md5sum after the remount, but that again is expected Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com> Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
2019-02-19 01:38:49 +08:00
if (xfs_is_cow_inode(ip) && is_write) {
/*
* FIXME: It could still overwrite on unshared extents and not
* need allocation.
*/
if (flags & IOMAP_NOWAIT)
return -EAGAIN;
mode = XFS_ILOCK_EXCL;
}
/*
* Extents not yet cached requires exclusive access, don't block. This
* is an opencoded xfs_ilock_data_map_shared() call but with
* non-blocking behaviour.
*/
if (!(ip->i_df.if_flags & XFS_IFEXTENTS)) {
if (flags & IOMAP_NOWAIT)
return -EAGAIN;
mode = XFS_ILOCK_EXCL;
}
xfs: recheck reflink state after grabbing ILOCK_SHARED for a write The reflink iflag could have changed since the earlier unlocked check, so if we got ILOCK_SHARED for a write and but we're now a reflink inode we have to switch to ILOCK_EXCL and relock. This helps us avoid blowing lock assertions in things like generic/166: XFS: Assertion failed: xfs_isilocked(ip, XFS_ILOCK_EXCL), file: fs/xfs/xfs_reflink.c, line: 383 WARNING: CPU: 1 PID: 24707 at fs/xfs/xfs_message.c:104 assfail+0x25/0x30 [xfs] Modules linked in: deadline_iosched dm_snapshot dm_bufio ext4 mbcache jbd2 dm_flakey xfs libcrc32c dax_pmem device_dax nd_pmem sch_fq_codel af_packet [last unloaded: scsi_debug] CPU: 1 PID: 24707 Comm: xfs_io Not tainted 4.18.0-rc1-djw #1 Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.10.2-1ubuntu1 04/01/2014 RIP: 0010:assfail+0x25/0x30 [xfs] Code: ff 0f 0b c3 90 66 66 66 66 90 48 89 f1 41 89 d0 48 c7 c6 e8 ef 1b a0 48 89 fa 31 ff e8 54 f9 ff ff 80 3d fd ba 0f 00 00 75 03 <0f> 0b c3 0f 0b 66 0f 1f 44 00 00 66 66 66 66 90 48 63 f6 49 89 f9 RSP: 0018:ffffc90006423ad8 EFLAGS: 00010246 RAX: 0000000000000000 RBX: ffff880030b65e80 RCX: 0000000000000000 RDX: 00000000ffffffc0 RSI: 000000000000000a RDI: ffffffffa01b0447 RBP: ffffc90006423c10 R08: 0000000000000000 R09: 0000000000000000 R10: ffff88003d43fc30 R11: f000000000000000 R12: ffff880077cda000 R13: 0000000000000000 R14: ffffc90006423c30 R15: ffffc90006423bf9 FS: 00007feba8986800(0000) GS:ffff88003ec00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 000000000138ab58 CR3: 000000003d40a000 CR4: 00000000000006a0 Call Trace: xfs_reflink_allocate_cow+0x24c/0x3d0 [xfs] xfs_file_iomap_begin+0x6d2/0xeb0 [xfs] ? iomap_to_fiemap+0x80/0x80 iomap_apply+0x5e/0x130 iomap_dio_rw+0x2e0/0x400 ? iomap_to_fiemap+0x80/0x80 ? xfs_file_dio_aio_write+0x133/0x4a0 [xfs] xfs_file_dio_aio_write+0x133/0x4a0 [xfs] xfs_file_write_iter+0x7b/0xb0 [xfs] __vfs_write+0x16f/0x1f0 vfs_write+0xc8/0x1c0 ksys_pwrite64+0x74/0x90 do_syscall_64+0x56/0x180 entry_SYSCALL_64_after_hwframe+0x49/0xbe Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2018-06-22 14:26:57 +08:00
relock:
if (flags & IOMAP_NOWAIT) {
if (!xfs_ilock_nowait(ip, mode))
return -EAGAIN;
} else {
xfs_ilock(ip, mode);
}
xfs: recheck reflink state after grabbing ILOCK_SHARED for a write The reflink iflag could have changed since the earlier unlocked check, so if we got ILOCK_SHARED for a write and but we're now a reflink inode we have to switch to ILOCK_EXCL and relock. This helps us avoid blowing lock assertions in things like generic/166: XFS: Assertion failed: xfs_isilocked(ip, XFS_ILOCK_EXCL), file: fs/xfs/xfs_reflink.c, line: 383 WARNING: CPU: 1 PID: 24707 at fs/xfs/xfs_message.c:104 assfail+0x25/0x30 [xfs] Modules linked in: deadline_iosched dm_snapshot dm_bufio ext4 mbcache jbd2 dm_flakey xfs libcrc32c dax_pmem device_dax nd_pmem sch_fq_codel af_packet [last unloaded: scsi_debug] CPU: 1 PID: 24707 Comm: xfs_io Not tainted 4.18.0-rc1-djw #1 Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.10.2-1ubuntu1 04/01/2014 RIP: 0010:assfail+0x25/0x30 [xfs] Code: ff 0f 0b c3 90 66 66 66 66 90 48 89 f1 41 89 d0 48 c7 c6 e8 ef 1b a0 48 89 fa 31 ff e8 54 f9 ff ff 80 3d fd ba 0f 00 00 75 03 <0f> 0b c3 0f 0b 66 0f 1f 44 00 00 66 66 66 66 90 48 63 f6 49 89 f9 RSP: 0018:ffffc90006423ad8 EFLAGS: 00010246 RAX: 0000000000000000 RBX: ffff880030b65e80 RCX: 0000000000000000 RDX: 00000000ffffffc0 RSI: 000000000000000a RDI: ffffffffa01b0447 RBP: ffffc90006423c10 R08: 0000000000000000 R09: 0000000000000000 R10: ffff88003d43fc30 R11: f000000000000000 R12: ffff880077cda000 R13: 0000000000000000 R14: ffffc90006423c30 R15: ffffc90006423bf9 FS: 00007feba8986800(0000) GS:ffff88003ec00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 000000000138ab58 CR3: 000000003d40a000 CR4: 00000000000006a0 Call Trace: xfs_reflink_allocate_cow+0x24c/0x3d0 [xfs] xfs_file_iomap_begin+0x6d2/0xeb0 [xfs] ? iomap_to_fiemap+0x80/0x80 iomap_apply+0x5e/0x130 iomap_dio_rw+0x2e0/0x400 ? iomap_to_fiemap+0x80/0x80 ? xfs_file_dio_aio_write+0x133/0x4a0 [xfs] xfs_file_dio_aio_write+0x133/0x4a0 [xfs] xfs_file_write_iter+0x7b/0xb0 [xfs] __vfs_write+0x16f/0x1f0 vfs_write+0xc8/0x1c0 ksys_pwrite64+0x74/0x90 do_syscall_64+0x56/0x180 entry_SYSCALL_64_after_hwframe+0x49/0xbe Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2018-06-22 14:26:57 +08:00
/*
* The reflink iflag could have changed since the earlier unlocked
* check, so if we got ILOCK_SHARED for a write and but we're now a
* reflink inode we have to switch to ILOCK_EXCL and relock.
*/
xfs: introduce an always_cow mode Add a mode where XFS never overwrites existing blocks in place. This is to aid debugging our COW code, and also put infatructure in place for things like possible future support for zoned block devices, which can't support overwrites. This mode is enabled globally by doing a: echo 1 > /sys/fs/xfs/debug/always_cow Note that the parameter is global to allow running all tests in xfstests easily in this mode, which would not easily be possible with a per-fs sysfs file. In always_cow mode persistent preallocations are disabled, and fallocate will fail when called with a 0 mode (with our without FALLOC_FL_KEEP_SIZE), and not create unwritten extent for zeroed space when called with FALLOC_FL_ZERO_RANGE or FALLOC_FL_UNSHARE_RANGE. There are a few interesting xfstests failures when run in always_cow mode: - generic/392 fails because the bytes used in the file used to test hole punch recovery are less after the log replay. This is because the blocks written and then punched out are only freed with a delay due to the logging mechanism. - xfs/170 will fail as the already fragile file streams mechanism doesn't seem to interact well with the COW allocator - xfs/180 xfs/182 xfs/192 xfs/198 xfs/204 and xfs/208 will claim the file system is badly fragmented, but there is not much we can do to avoid that when always writing out of place - xfs/205 fails because overwriting a file in always_cow mode will require new space allocation and the assumption in the test thus don't work anymore. - xfs/326 fails to modify the file at all in always_cow mode after injecting the refcount error, leading to an unexpected md5sum after the remount, but that again is expected Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com> Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
2019-02-19 01:38:49 +08:00
if (mode == XFS_ILOCK_SHARED && is_write && xfs_is_cow_inode(ip)) {
xfs: recheck reflink state after grabbing ILOCK_SHARED for a write The reflink iflag could have changed since the earlier unlocked check, so if we got ILOCK_SHARED for a write and but we're now a reflink inode we have to switch to ILOCK_EXCL and relock. This helps us avoid blowing lock assertions in things like generic/166: XFS: Assertion failed: xfs_isilocked(ip, XFS_ILOCK_EXCL), file: fs/xfs/xfs_reflink.c, line: 383 WARNING: CPU: 1 PID: 24707 at fs/xfs/xfs_message.c:104 assfail+0x25/0x30 [xfs] Modules linked in: deadline_iosched dm_snapshot dm_bufio ext4 mbcache jbd2 dm_flakey xfs libcrc32c dax_pmem device_dax nd_pmem sch_fq_codel af_packet [last unloaded: scsi_debug] CPU: 1 PID: 24707 Comm: xfs_io Not tainted 4.18.0-rc1-djw #1 Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.10.2-1ubuntu1 04/01/2014 RIP: 0010:assfail+0x25/0x30 [xfs] Code: ff 0f 0b c3 90 66 66 66 66 90 48 89 f1 41 89 d0 48 c7 c6 e8 ef 1b a0 48 89 fa 31 ff e8 54 f9 ff ff 80 3d fd ba 0f 00 00 75 03 <0f> 0b c3 0f 0b 66 0f 1f 44 00 00 66 66 66 66 90 48 63 f6 49 89 f9 RSP: 0018:ffffc90006423ad8 EFLAGS: 00010246 RAX: 0000000000000000 RBX: ffff880030b65e80 RCX: 0000000000000000 RDX: 00000000ffffffc0 RSI: 000000000000000a RDI: ffffffffa01b0447 RBP: ffffc90006423c10 R08: 0000000000000000 R09: 0000000000000000 R10: ffff88003d43fc30 R11: f000000000000000 R12: ffff880077cda000 R13: 0000000000000000 R14: ffffc90006423c30 R15: ffffc90006423bf9 FS: 00007feba8986800(0000) GS:ffff88003ec00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 000000000138ab58 CR3: 000000003d40a000 CR4: 00000000000006a0 Call Trace: xfs_reflink_allocate_cow+0x24c/0x3d0 [xfs] xfs_file_iomap_begin+0x6d2/0xeb0 [xfs] ? iomap_to_fiemap+0x80/0x80 iomap_apply+0x5e/0x130 iomap_dio_rw+0x2e0/0x400 ? iomap_to_fiemap+0x80/0x80 ? xfs_file_dio_aio_write+0x133/0x4a0 [xfs] xfs_file_dio_aio_write+0x133/0x4a0 [xfs] xfs_file_write_iter+0x7b/0xb0 [xfs] __vfs_write+0x16f/0x1f0 vfs_write+0xc8/0x1c0 ksys_pwrite64+0x74/0x90 do_syscall_64+0x56/0x180 entry_SYSCALL_64_after_hwframe+0x49/0xbe Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2018-06-22 14:26:57 +08:00
xfs_iunlock(ip, mode);
mode = XFS_ILOCK_EXCL;
goto relock;
}
*lockmode = mode;
return 0;
}
static int
xfs_file_iomap_begin(
struct inode *inode,
loff_t offset,
loff_t length,
unsigned flags,
struct iomap *iomap)
{
struct xfs_inode *ip = XFS_I(inode);
struct xfs_mount *mp = ip->i_mount;
struct xfs_bmbt_irec imap;
xfs_fileoff_t offset_fsb, end_fsb;
int nimaps = 1, error = 0;
bool shared = false;
unsigned lockmode;
if (XFS_FORCED_SHUTDOWN(mp))
return -EIO;
if ((flags & (IOMAP_WRITE | IOMAP_ZERO)) && !(flags & IOMAP_DIRECT) &&
!IS_DAX(inode) && !xfs_get_extsz_hint(ip)) {
/* Reserve delalloc blocks for regular writeback. */
return xfs_file_iomap_begin_delay(inode, offset, length, flags,
iomap);
}
/*
* Lock the inode in the manner required for the specified operation and
* check for as many conditions that would result in blocking as
* possible. This removes most of the non-blocking checks from the
* mapping code below.
*/
error = xfs_ilock_for_iomap(ip, flags, &lockmode);
if (error)
return error;
ASSERT(offset <= mp->m_super->s_maxbytes);
if (offset > mp->m_super->s_maxbytes - length)
length = mp->m_super->s_maxbytes - offset;
offset_fsb = XFS_B_TO_FSBT(mp, offset);
end_fsb = XFS_B_TO_FSB(mp, offset + length);
error = xfs_bmapi_read(ip, offset_fsb, end_fsb - offset_fsb, &imap,
&nimaps, 0);
if (error)
goto out_unlock;
if (flags & IOMAP_REPORT) {
/* Trim the mapping to the nearest shared extent boundary. */
error = xfs_reflink_trim_around_shared(ip, &imap, &shared);
if (error)
goto out_unlock;
}
/* Non-modifying mapping requested, so we are done */
if (!(flags & (IOMAP_WRITE | IOMAP_ZERO)))
goto out_found;
/*
* Break shared extents if necessary. Checks for non-blocking IO have
* been done up front, so we don't need to do them here.
*/
xfs: introduce an always_cow mode Add a mode where XFS never overwrites existing blocks in place. This is to aid debugging our COW code, and also put infatructure in place for things like possible future support for zoned block devices, which can't support overwrites. This mode is enabled globally by doing a: echo 1 > /sys/fs/xfs/debug/always_cow Note that the parameter is global to allow running all tests in xfstests easily in this mode, which would not easily be possible with a per-fs sysfs file. In always_cow mode persistent preallocations are disabled, and fallocate will fail when called with a 0 mode (with our without FALLOC_FL_KEEP_SIZE), and not create unwritten extent for zeroed space when called with FALLOC_FL_ZERO_RANGE or FALLOC_FL_UNSHARE_RANGE. There are a few interesting xfstests failures when run in always_cow mode: - generic/392 fails because the bytes used in the file used to test hole punch recovery are less after the log replay. This is because the blocks written and then punched out are only freed with a delay due to the logging mechanism. - xfs/170 will fail as the already fragile file streams mechanism doesn't seem to interact well with the COW allocator - xfs/180 xfs/182 xfs/192 xfs/198 xfs/204 and xfs/208 will claim the file system is badly fragmented, but there is not much we can do to avoid that when always writing out of place - xfs/205 fails because overwriting a file in always_cow mode will require new space allocation and the assumption in the test thus don't work anymore. - xfs/326 fails to modify the file at all in always_cow mode after injecting the refcount error, leading to an unexpected md5sum after the remount, but that again is expected Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com> Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
2019-02-19 01:38:49 +08:00
if (xfs_is_cow_inode(ip)) {
struct xfs_bmbt_irec cmap;
bool directio = (flags & IOMAP_DIRECT);
/* if zeroing doesn't need COW allocation, then we are done. */
if ((flags & IOMAP_ZERO) &&
!needs_cow_for_zeroing(&imap, nimaps))
goto out_found;
/* may drop and re-acquire the ilock */
cmap = imap;
error = xfs_reflink_allocate_cow(ip, &cmap, &shared, &lockmode,
directio);
if (error)
goto out_unlock;
/*
* For buffered writes we need to report the address of the
* previous block (if there was any) so that the higher level
* write code can perform read-modify-write operations; we
* won't need the CoW fork mapping until writeback. For direct
* I/O, which must be block aligned, we need to report the
* newly allocated address. If the data fork has a hole, copy
* the COW fork mapping to avoid allocating to the data fork.
*/
if (directio || imap.br_startblock == HOLESTARTBLOCK)
imap = cmap;
end_fsb = imap.br_startoff + imap.br_blockcount;
length = XFS_FSB_TO_B(mp, end_fsb) - offset;
}
/* Don't need to allocate over holes when doing zeroing operations. */
if (flags & IOMAP_ZERO)
goto out_found;
if (!imap_needs_alloc(inode, &imap, nimaps))
goto out_found;
/* If nowait is set bail since we are going to make allocations. */
if (flags & IOMAP_NOWAIT) {
error = -EAGAIN;
goto out_unlock;
}
/*
* We cap the maximum length we map to a sane size to keep the chunks
* of work done where somewhat symmetric with the work writeback does.
* This is a completely arbitrary number pulled out of thin air as a
* best guess for initial testing.
*
* Note that the values needs to be less than 32-bits wide until the
* lower level functions are updated.
*/
length = min_t(loff_t, length, 1024 * PAGE_SIZE);
/*
* xfs_iomap_write_direct() expects the shared lock. It is unlocked on
* return.
*/
if (lockmode == XFS_ILOCK_EXCL)
xfs_ilock_demote(ip, lockmode);
error = xfs_iomap_write_direct(ip, offset, length, &imap,
nimaps);
if (error)
return error;
iomap->flags |= IOMAP_F_NEW;
trace_xfs_iomap_alloc(ip, offset, length, XFS_DATA_FORK, &imap);
out_finish:
return xfs_bmbt_to_iomap(ip, iomap, &imap, shared);
out_found:
ASSERT(nimaps);
xfs_iunlock(ip, lockmode);
trace_xfs_iomap_found(ip, offset, length, XFS_DATA_FORK, &imap);
goto out_finish;
out_unlock:
xfs_iunlock(ip, lockmode);
return error;
}
static int
xfs_file_iomap_end_delalloc(
struct xfs_inode *ip,
loff_t offset,
loff_t length,
xfs: use iomap new flag for newly allocated delalloc blocks Commit fa7f138 ("xfs: clear delalloc and cache on buffered write failure") fixed one regression in the iomap error handling code and exposed another. The fundamental problem is that if a buffered write is a rewrite of preexisting delalloc blocks and the write fails, the failure handling code can punch out preexisting blocks with valid file data. This was reproduced directly by sub-block writes in the LTP kernel/syscalls/write/write03 test. A first 100 byte write allocates a single block in a file. A subsequent 100 byte write fails and punches out the block, including the data successfully written by the previous write. To address this problem, update the ->iomap_begin() handler to distinguish newly allocated delalloc blocks from preexisting delalloc blocks via the IOMAP_F_NEW flag. Use this flag in the ->iomap_end() handler to decide when a failed or short write should punch out delalloc blocks. This introduces the subtle requirement that ->iomap_begin() should never combine newly allocated delalloc blocks with existing blocks in the resulting iomap descriptor. This can occur when a new delalloc reservation merges with a neighboring extent that is part of the current write, for example. Therefore, drop the post-allocation extent lookup from xfs_bmapi_reserve_delalloc() and just return the record inserted into the fork. This ensures only new blocks are returned and thus that preexisting delalloc blocks are always handled as "found" blocks and not punched out on a failed rewrite. Reported-by: Xiong Zhou <xzhou@redhat.com> Signed-off-by: Brian Foster <bfoster@redhat.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com> Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
2017-03-09 01:58:08 +08:00
ssize_t written,
struct iomap *iomap)
{
struct xfs_mount *mp = ip->i_mount;
xfs_fileoff_t start_fsb;
xfs_fileoff_t end_fsb;
int error = 0;
xfs: use iomap new flag for newly allocated delalloc blocks Commit fa7f138 ("xfs: clear delalloc and cache on buffered write failure") fixed one regression in the iomap error handling code and exposed another. The fundamental problem is that if a buffered write is a rewrite of preexisting delalloc blocks and the write fails, the failure handling code can punch out preexisting blocks with valid file data. This was reproduced directly by sub-block writes in the LTP kernel/syscalls/write/write03 test. A first 100 byte write allocates a single block in a file. A subsequent 100 byte write fails and punches out the block, including the data successfully written by the previous write. To address this problem, update the ->iomap_begin() handler to distinguish newly allocated delalloc blocks from preexisting delalloc blocks via the IOMAP_F_NEW flag. Use this flag in the ->iomap_end() handler to decide when a failed or short write should punch out delalloc blocks. This introduces the subtle requirement that ->iomap_begin() should never combine newly allocated delalloc blocks with existing blocks in the resulting iomap descriptor. This can occur when a new delalloc reservation merges with a neighboring extent that is part of the current write, for example. Therefore, drop the post-allocation extent lookup from xfs_bmapi_reserve_delalloc() and just return the record inserted into the fork. This ensures only new blocks are returned and thus that preexisting delalloc blocks are always handled as "found" blocks and not punched out on a failed rewrite. Reported-by: Xiong Zhou <xzhou@redhat.com> Signed-off-by: Brian Foster <bfoster@redhat.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com> Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
2017-03-09 01:58:08 +08:00
/*
* Behave as if the write failed if drop writes is enabled. Set the NEW
* flag to force delalloc cleanup.
*/
if (XFS_TEST_ERROR(false, mp, XFS_ERRTAG_DROP_WRITES)) {
xfs: use iomap new flag for newly allocated delalloc blocks Commit fa7f138 ("xfs: clear delalloc and cache on buffered write failure") fixed one regression in the iomap error handling code and exposed another. The fundamental problem is that if a buffered write is a rewrite of preexisting delalloc blocks and the write fails, the failure handling code can punch out preexisting blocks with valid file data. This was reproduced directly by sub-block writes in the LTP kernel/syscalls/write/write03 test. A first 100 byte write allocates a single block in a file. A subsequent 100 byte write fails and punches out the block, including the data successfully written by the previous write. To address this problem, update the ->iomap_begin() handler to distinguish newly allocated delalloc blocks from preexisting delalloc blocks via the IOMAP_F_NEW flag. Use this flag in the ->iomap_end() handler to decide when a failed or short write should punch out delalloc blocks. This introduces the subtle requirement that ->iomap_begin() should never combine newly allocated delalloc blocks with existing blocks in the resulting iomap descriptor. This can occur when a new delalloc reservation merges with a neighboring extent that is part of the current write, for example. Therefore, drop the post-allocation extent lookup from xfs_bmapi_reserve_delalloc() and just return the record inserted into the fork. This ensures only new blocks are returned and thus that preexisting delalloc blocks are always handled as "found" blocks and not punched out on a failed rewrite. Reported-by: Xiong Zhou <xzhou@redhat.com> Signed-off-by: Brian Foster <bfoster@redhat.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com> Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
2017-03-09 01:58:08 +08:00
iomap->flags |= IOMAP_F_NEW;
written = 0;
xfs: use iomap new flag for newly allocated delalloc blocks Commit fa7f138 ("xfs: clear delalloc and cache on buffered write failure") fixed one regression in the iomap error handling code and exposed another. The fundamental problem is that if a buffered write is a rewrite of preexisting delalloc blocks and the write fails, the failure handling code can punch out preexisting blocks with valid file data. This was reproduced directly by sub-block writes in the LTP kernel/syscalls/write/write03 test. A first 100 byte write allocates a single block in a file. A subsequent 100 byte write fails and punches out the block, including the data successfully written by the previous write. To address this problem, update the ->iomap_begin() handler to distinguish newly allocated delalloc blocks from preexisting delalloc blocks via the IOMAP_F_NEW flag. Use this flag in the ->iomap_end() handler to decide when a failed or short write should punch out delalloc blocks. This introduces the subtle requirement that ->iomap_begin() should never combine newly allocated delalloc blocks with existing blocks in the resulting iomap descriptor. This can occur when a new delalloc reservation merges with a neighboring extent that is part of the current write, for example. Therefore, drop the post-allocation extent lookup from xfs_bmapi_reserve_delalloc() and just return the record inserted into the fork. This ensures only new blocks are returned and thus that preexisting delalloc blocks are always handled as "found" blocks and not punched out on a failed rewrite. Reported-by: Xiong Zhou <xzhou@redhat.com> Signed-off-by: Brian Foster <bfoster@redhat.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com> Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
2017-03-09 01:58:08 +08:00
}
/*
* start_fsb refers to the first unused block after a short write. If
* nothing was written, round offset down to point at the first block in
* the range.
*/
if (unlikely(!written))
start_fsb = XFS_B_TO_FSBT(mp, offset);
else
start_fsb = XFS_B_TO_FSB(mp, offset + written);
end_fsb = XFS_B_TO_FSB(mp, offset + length);
/*
xfs: use iomap new flag for newly allocated delalloc blocks Commit fa7f138 ("xfs: clear delalloc and cache on buffered write failure") fixed one regression in the iomap error handling code and exposed another. The fundamental problem is that if a buffered write is a rewrite of preexisting delalloc blocks and the write fails, the failure handling code can punch out preexisting blocks with valid file data. This was reproduced directly by sub-block writes in the LTP kernel/syscalls/write/write03 test. A first 100 byte write allocates a single block in a file. A subsequent 100 byte write fails and punches out the block, including the data successfully written by the previous write. To address this problem, update the ->iomap_begin() handler to distinguish newly allocated delalloc blocks from preexisting delalloc blocks via the IOMAP_F_NEW flag. Use this flag in the ->iomap_end() handler to decide when a failed or short write should punch out delalloc blocks. This introduces the subtle requirement that ->iomap_begin() should never combine newly allocated delalloc blocks with existing blocks in the resulting iomap descriptor. This can occur when a new delalloc reservation merges with a neighboring extent that is part of the current write, for example. Therefore, drop the post-allocation extent lookup from xfs_bmapi_reserve_delalloc() and just return the record inserted into the fork. This ensures only new blocks are returned and thus that preexisting delalloc blocks are always handled as "found" blocks and not punched out on a failed rewrite. Reported-by: Xiong Zhou <xzhou@redhat.com> Signed-off-by: Brian Foster <bfoster@redhat.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com> Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
2017-03-09 01:58:08 +08:00
* Trim delalloc blocks if they were allocated by this write and we
* didn't manage to write the whole range.
*
* We don't need to care about racing delalloc as we hold i_mutex
* across the reserve/allocate/unreserve calls. If there are delalloc
* blocks in the range, they are ours.
*/
xfs: use iomap new flag for newly allocated delalloc blocks Commit fa7f138 ("xfs: clear delalloc and cache on buffered write failure") fixed one regression in the iomap error handling code and exposed another. The fundamental problem is that if a buffered write is a rewrite of preexisting delalloc blocks and the write fails, the failure handling code can punch out preexisting blocks with valid file data. This was reproduced directly by sub-block writes in the LTP kernel/syscalls/write/write03 test. A first 100 byte write allocates a single block in a file. A subsequent 100 byte write fails and punches out the block, including the data successfully written by the previous write. To address this problem, update the ->iomap_begin() handler to distinguish newly allocated delalloc blocks from preexisting delalloc blocks via the IOMAP_F_NEW flag. Use this flag in the ->iomap_end() handler to decide when a failed or short write should punch out delalloc blocks. This introduces the subtle requirement that ->iomap_begin() should never combine newly allocated delalloc blocks with existing blocks in the resulting iomap descriptor. This can occur when a new delalloc reservation merges with a neighboring extent that is part of the current write, for example. Therefore, drop the post-allocation extent lookup from xfs_bmapi_reserve_delalloc() and just return the record inserted into the fork. This ensures only new blocks are returned and thus that preexisting delalloc blocks are always handled as "found" blocks and not punched out on a failed rewrite. Reported-by: Xiong Zhou <xzhou@redhat.com> Signed-off-by: Brian Foster <bfoster@redhat.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com> Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
2017-03-09 01:58:08 +08:00
if ((iomap->flags & IOMAP_F_NEW) && start_fsb < end_fsb) {
truncate_pagecache_range(VFS_I(ip), XFS_FSB_TO_B(mp, start_fsb),
XFS_FSB_TO_B(mp, end_fsb) - 1);
error = xfs_bmap_punch_delalloc_range(ip, start_fsb,
end_fsb - start_fsb);
if (error && !XFS_FORCED_SHUTDOWN(mp)) {
xfs_alert(mp, "%s: unable to clean up ino %lld",
__func__, ip->i_ino);
return error;
}
}
return 0;
}
static int
xfs_file_iomap_end(
struct inode *inode,
loff_t offset,
loff_t length,
ssize_t written,
unsigned flags,
struct iomap *iomap)
{
if ((flags & IOMAP_WRITE) && iomap->type == IOMAP_DELALLOC)
return xfs_file_iomap_end_delalloc(XFS_I(inode), offset,
xfs: use iomap new flag for newly allocated delalloc blocks Commit fa7f138 ("xfs: clear delalloc and cache on buffered write failure") fixed one regression in the iomap error handling code and exposed another. The fundamental problem is that if a buffered write is a rewrite of preexisting delalloc blocks and the write fails, the failure handling code can punch out preexisting blocks with valid file data. This was reproduced directly by sub-block writes in the LTP kernel/syscalls/write/write03 test. A first 100 byte write allocates a single block in a file. A subsequent 100 byte write fails and punches out the block, including the data successfully written by the previous write. To address this problem, update the ->iomap_begin() handler to distinguish newly allocated delalloc blocks from preexisting delalloc blocks via the IOMAP_F_NEW flag. Use this flag in the ->iomap_end() handler to decide when a failed or short write should punch out delalloc blocks. This introduces the subtle requirement that ->iomap_begin() should never combine newly allocated delalloc blocks with existing blocks in the resulting iomap descriptor. This can occur when a new delalloc reservation merges with a neighboring extent that is part of the current write, for example. Therefore, drop the post-allocation extent lookup from xfs_bmapi_reserve_delalloc() and just return the record inserted into the fork. This ensures only new blocks are returned and thus that preexisting delalloc blocks are always handled as "found" blocks and not punched out on a failed rewrite. Reported-by: Xiong Zhou <xzhou@redhat.com> Signed-off-by: Brian Foster <bfoster@redhat.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com> Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
2017-03-09 01:58:08 +08:00
length, written, iomap);
return 0;
}
const struct iomap_ops xfs_iomap_ops = {
.iomap_begin = xfs_file_iomap_begin,
.iomap_end = xfs_file_iomap_end,
};
static int
xfs_seek_iomap_begin(
struct inode *inode,
loff_t offset,
loff_t length,
unsigned flags,
struct iomap *iomap)
{
struct xfs_inode *ip = XFS_I(inode);
struct xfs_mount *mp = ip->i_mount;
xfs_fileoff_t offset_fsb = XFS_B_TO_FSBT(mp, offset);
xfs_fileoff_t end_fsb = XFS_B_TO_FSB(mp, offset + length);
xfs_fileoff_t cow_fsb = NULLFILEOFF, data_fsb = NULLFILEOFF;
struct xfs_iext_cursor icur;
struct xfs_bmbt_irec imap, cmap;
int error = 0;
unsigned lockmode;
if (XFS_FORCED_SHUTDOWN(mp))
return -EIO;
lockmode = xfs_ilock_data_map_shared(ip);
if (!(ip->i_df.if_flags & XFS_IFEXTENTS)) {
error = xfs_iread_extents(NULL, ip, XFS_DATA_FORK);
if (error)
goto out_unlock;
}
if (xfs_iext_lookup_extent(ip, &ip->i_df, offset_fsb, &icur, &imap)) {
/*
* If we found a data extent we are done.
*/
if (imap.br_startoff <= offset_fsb)
goto done;
data_fsb = imap.br_startoff;
} else {
/*
* Fake a hole until the end of the file.
*/
data_fsb = min(XFS_B_TO_FSB(mp, offset + length),
XFS_B_TO_FSB(mp, mp->m_super->s_maxbytes));
}
/*
* If a COW fork extent covers the hole, report it - capped to the next
* data fork extent:
*/
if (xfs_inode_has_cow_data(ip) &&
xfs_iext_lookup_extent(ip, ip->i_cowfp, offset_fsb, &icur, &cmap))
cow_fsb = cmap.br_startoff;
if (cow_fsb != NULLFILEOFF && cow_fsb <= offset_fsb) {
if (data_fsb < cow_fsb + cmap.br_blockcount)
end_fsb = min(end_fsb, data_fsb);
xfs_trim_extent(&cmap, offset_fsb, end_fsb);
error = xfs_bmbt_to_iomap(ip, iomap, &cmap, true);
/*
* This is a COW extent, so we must probe the page cache
* because there could be dirty page cache being backed
* by this extent.
*/
iomap->type = IOMAP_UNWRITTEN;
goto out_unlock;
}
/*
* Else report a hole, capped to the next found data or COW extent.
*/
if (cow_fsb != NULLFILEOFF && cow_fsb < data_fsb)
imap.br_blockcount = cow_fsb - offset_fsb;
else
imap.br_blockcount = data_fsb - offset_fsb;
imap.br_startoff = offset_fsb;
imap.br_startblock = HOLESTARTBLOCK;
imap.br_state = XFS_EXT_NORM;
done:
xfs_trim_extent(&imap, offset_fsb, end_fsb);
error = xfs_bmbt_to_iomap(ip, iomap, &imap, false);
out_unlock:
xfs_iunlock(ip, lockmode);
return error;
}
const struct iomap_ops xfs_seek_iomap_ops = {
.iomap_begin = xfs_seek_iomap_begin,
};
static int
xfs_xattr_iomap_begin(
struct inode *inode,
loff_t offset,
loff_t length,
unsigned flags,
struct iomap *iomap)
{
struct xfs_inode *ip = XFS_I(inode);
struct xfs_mount *mp = ip->i_mount;
xfs_fileoff_t offset_fsb = XFS_B_TO_FSBT(mp, offset);
xfs_fileoff_t end_fsb = XFS_B_TO_FSB(mp, offset + length);
struct xfs_bmbt_irec imap;
int nimaps = 1, error = 0;
unsigned lockmode;
if (XFS_FORCED_SHUTDOWN(mp))
return -EIO;
lockmode = xfs_ilock_attr_map_shared(ip);
/* if there are no attribute fork or extents, return ENOENT */
if (!XFS_IFORK_Q(ip) || !ip->i_d.di_anextents) {
error = -ENOENT;
goto out_unlock;
}
ASSERT(ip->i_d.di_aformat != XFS_DINODE_FMT_LOCAL);
error = xfs_bmapi_read(ip, offset_fsb, end_fsb - offset_fsb, &imap,
&nimaps, XFS_BMAPI_ATTRFORK);
out_unlock:
xfs_iunlock(ip, lockmode);
if (error)
return error;
ASSERT(nimaps);
return xfs_bmbt_to_iomap(ip, iomap, &imap, false);
}
const struct iomap_ops xfs_xattr_iomap_ops = {
.iomap_begin = xfs_xattr_iomap_begin,
};