2
0
mirror of https://github.com/edk2-porting/linux-next.git synced 2024-12-28 15:13:55 +08:00
linux-next/fs/xfs/xfs_extfree_item.c

762 lines
21 KiB
C
Raw Normal View History

// SPDX-License-Identifier: GPL-2.0
/*
* Copyright (c) 2000-2001,2005 Silicon Graphics, Inc.
* All Rights Reserved.
*/
#include "xfs.h"
#include "xfs_fs.h"
#include "xfs_format.h"
#include "xfs_log_format.h"
#include "xfs_trans_resv.h"
#include "xfs_bit.h"
#include "xfs_shared.h"
#include "xfs_mount.h"
#include "xfs_defer.h"
#include "xfs_trans.h"
#include "xfs_trans_priv.h"
#include "xfs_extfree_item.h"
#include "xfs_log.h"
xfs: add owner field to extent allocation and freeing For the rmap btree to work, we have to feed the extent owner information to the the allocation and freeing functions. This information is what will end up in the rmap btree that tracks allocated extents. While we technically don't need the owner information when freeing extents, passing it allows us to validate that the extent we are removing from the rmap btree actually belonged to the owner we expected it to belong to. We also define a special set of owner values for internal metadata that would otherwise have no owner. This allows us to tell the difference between metadata owned by different per-ag btrees, as well as static fs metadata (e.g. AG headers) and internal journal blocks. There are also a couple of special cases we need to take care of - during EFI recovery, we don't actually know who the original owner was, so we need to pass a wildcard to indicate that we aren't checking the owner for validity. We also need special handling in growfs, as we "free" the space in the last AG when extending it, but because it's new space it has no actual owner... While touching the xfs_bmap_add_free() function, re-order the parameters to put the struct xfs_mount first. Extend the owner field to include both the owner type and some sort of index within the owner. The index field will be used to support reverse mappings when reflink is enabled. When we're freeing extents from an EFI, we don't have the owner information available (rmap updates have their own redo items). xfs_free_extent therefore doesn't need to do an rmap update. Make sure that the log replay code signals this correctly. This is based upon a patch originally from Dave Chinner. It has been extended to add more owner information with the intent of helping recovery operations when things go wrong (e.g. offset of user data block in a file). [dchinner: de-shout the xfs_rmap_*_owner helpers] [darrick: minor style fixes suggested by Christoph Hellwig] Signed-off-by: Dave Chinner <dchinner@redhat.com> Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com> Reviewed-by: Dave Chinner <dchinner@redhat.com> Signed-off-by: Dave Chinner <david@fromorbit.com>
2016-08-03 09:33:42 +08:00
#include "xfs_btree.h"
#include "xfs_rmap.h"
#include "xfs_alloc.h"
#include "xfs_bmap.h"
#include "xfs_trace.h"
#include "xfs_error.h"
#include "xfs_log_priv.h"
#include "xfs_log_recover.h"
kmem_zone_t *xfs_efi_zone;
kmem_zone_t *xfs_efd_zone;
static const struct xfs_item_ops xfs_efi_item_ops;
static inline struct xfs_efi_log_item *EFI_ITEM(struct xfs_log_item *lip)
{
return container_of(lip, struct xfs_efi_log_item, efi_item);
}
STATIC void
xfs_efi_item_free(
struct xfs_efi_log_item *efip)
{
xfs: allocate log vector buffers outside CIL context lock One of the problems we currently have with delayed logging is that under serious memory pressure we can deadlock memory reclaim. THis occurs when memory reclaim (such as run by kswapd) is reclaiming XFS inodes and issues a log force to unpin inodes that are dirty in the CIL. The CIL is pushed, but this will only occur once it gets the CIL context lock to ensure that all committing transactions are complete and no new transactions start being committed to the CIL while the push switches to a new context. The deadlock occurs when the CIL context lock is held by a committing process that is doing memory allocation for log vector buffers, and that allocation is then blocked on memory reclaim making progress. Memory reclaim, however, is blocked waiting for a log force to make progress, and so we effectively deadlock at this point. To solve this problem, we have to move the CIL log vector buffer allocation outside of the context lock so that memory reclaim can always make progress when it needs to force the log. The problem with doing this is that a CIL push can take place while we are determining if we need to allocate a new log vector buffer for an item and hence the current log vector may go away without warning. That means we canot rely on the existing log vector being present when we finally grab the context lock and so we must have a replacement buffer ready to go at all times. To ensure this, introduce a "shadow log vector" buffer that is always guaranteed to be present when we gain the CIL context lock and format the item. This shadow buffer may or may not be used during the formatting, but if the log item does not have an existing log vector buffer or that buffer is too small for the new modifications, we swap it for the new shadow buffer and format the modifications into that new log vector buffer. The result of this is that for any object we modify more than once in a given CIL checkpoint, we double the memory required to track dirty regions in the log. For single modifications then we consume the shadow log vectorwe allocate on commit, and that gets consumed by the checkpoint. However, if we make multiple modifications, then the second transaction commit will allocate a shadow log vector and hence we will end up with double the memory usage as only one of the log vectors is consumed by the CIL checkpoint. The remaining shadow vector will be freed when th elog item is freed. This can probably be optimised in future - access to the shadow log vector is serialised by the object lock (as opposited to the active log vector, which is controlled by the CIL context lock) and so we can probably free shadow log vector from some objects when the log item is marked clean on removal from the AIL. Signed-off-by: Dave Chinner <dchinner@redhat.com> Reviewed-by: Brian Foster <bfoster@redhat.com> Signed-off-by: Dave Chinner <david@fromorbit.com>
2016-07-22 07:52:35 +08:00
kmem_free(efip->efi_item.li_lv_shadow);
if (efip->efi_format.efi_nextents > XFS_EFI_MAX_FAST_EXTENTS)
kmem_free(efip);
else
kmem_cache_free(xfs_efi_zone, efip);
}
/*
* Freeing the efi requires that we remove it from the AIL if it has already
* been placed there. However, the EFI may not yet have been placed in the AIL
* when called by xfs_efi_release() from EFD processing due to the ordering of
* committed vs unpin operations in bulk insert operations. Hence the reference
* count to ensure only the last caller frees the EFI.
*/
STATIC void
xfs_efi_release(
struct xfs_efi_log_item *efip)
{
ASSERT(atomic_read(&efip->efi_refcount) > 0);
if (atomic_dec_and_test(&efip->efi_refcount)) {
xfs_trans_ail_delete(&efip->efi_item, SHUTDOWN_LOG_IO_ERROR);
xfs_efi_item_free(efip);
}
}
/*
* This returns the number of iovecs needed to log the given efi item.
* We only need 1 iovec for an efi item. It just logs the efi_log_format
* structure.
*/
static inline int
xfs_efi_item_sizeof(
struct xfs_efi_log_item *efip)
{
return sizeof(struct xfs_efi_log_format) +
(efip->efi_format.efi_nextents - 1) * sizeof(xfs_extent_t);
}
STATIC void
xfs_efi_item_size(
struct xfs_log_item *lip,
int *nvecs,
int *nbytes)
{
*nvecs += 1;
*nbytes += xfs_efi_item_sizeof(EFI_ITEM(lip));
}
/*
* This is called to fill in the vector of log iovecs for the
* given efi log item. We use only 1 iovec, and we point that
* at the efi_log_format structure embedded in the efi item.
* It is at this point that we assert that all of the extent
* slots in the efi item have been filled.
*/
STATIC void
xfs_efi_item_format(
struct xfs_log_item *lip,
struct xfs_log_vec *lv)
{
struct xfs_efi_log_item *efip = EFI_ITEM(lip);
struct xfs_log_iovec *vecp = NULL;
ASSERT(atomic_read(&efip->efi_next_extent) ==
efip->efi_format.efi_nextents);
efip->efi_format.efi_type = XFS_LI_EFI;
efip->efi_format.efi_size = 1;
xlog_copy_iovec(lv, &vecp, XLOG_REG_TYPE_EFI_FORMAT,
&efip->efi_format,
xfs_efi_item_sizeof(efip));
}
/*
xfs: fix efi/efd error handling to avoid fs shutdown hangs Freeing an extent in XFS involves logging an EFI (extent free intention), freeing the actual extent, and logging an EFD (extent free done). The EFI object is created with a reference count of 2: one for the current transaction and one for the subsequently created EFD. Under normal circumstances, the first reference is dropped when the EFI is unpinned and the second reference is dropped when the EFD is committed to the on-disk log. In event of errors or filesystem shutdown, there are various potential cleanup scenarios depending on the state of the EFI/EFD. The cleanup scenarios are confusing and racy, as demonstrated by the following test sequence: # mount $dev $mnt # fsstress -d $mnt -n 99999 -p 16 -z -f fallocate=1 \ -f punch=1 -f creat=1 -f unlink=1 & # sleep 5 # killall -9 fsstress; wait # godown -f $mnt # umount ... in which the final umount can hang due to the AIL being pinned indefinitely by one or more EFI items. This can occur due to several conditions. For example, if the shutdown occurs after the EFI is committed to the on-disk log and the EFD committed to the CIL, but before the EFD committed to the log, the EFD iop_committed() abort handler does not drop its reference to the EFI. Alternatively, manual error injection in the xfs_bmap_finish() codepath shows that if an error occurs after the EFI transaction is committed but before the EFD is constructed and logged, the EFI is never released from the AIL. Update the EFI/EFD item handling code to use a more straightforward and reliable approach to error handling. If an error occurs after the EFI transaction is committed and before the EFD is constructed, release the EFI explicitly from xfs_bmap_finish(). If the EFI transaction is cancelled, release the EFI in the unlock handler. Once the EFD is constructed, it is responsible for releasing the EFI under any circumstances (including whether the EFI item aborts due to log I/O error). Update the EFD item handlers to release the EFI if the transaction is cancelled or aborts due to log I/O error. Finally, update xfs_bmap_finish() to log at least one EFD extent to the transaction before xfs_free_extent() errors are handled to ensure the transaction is dirty and EFD item error handling is triggered. Signed-off-by: Brian Foster <bfoster@redhat.com> Reviewed-by: Dave Chinner <dchinner@redhat.com> Signed-off-by: Dave Chinner <david@fromorbit.com>
2015-08-19 07:51:16 +08:00
* The unpin operation is the last place an EFI is manipulated in the log. It is
* either inserted in the AIL or aborted in the event of a log I/O error. In
* either case, the EFI transaction has been successfully committed to make it
* this far. Therefore, we expect whoever committed the EFI to either construct
* and commit the EFD or drop the EFD's reference in the event of error. Simply
* drop the log's EFI reference now that the log is done with it.
*/
STATIC void
xfs_efi_item_unpin(
struct xfs_log_item *lip,
int remove)
{
struct xfs_efi_log_item *efip = EFI_ITEM(lip);
xfs_efi_release(efip);
}
xfs: fix efi/efd error handling to avoid fs shutdown hangs Freeing an extent in XFS involves logging an EFI (extent free intention), freeing the actual extent, and logging an EFD (extent free done). The EFI object is created with a reference count of 2: one for the current transaction and one for the subsequently created EFD. Under normal circumstances, the first reference is dropped when the EFI is unpinned and the second reference is dropped when the EFD is committed to the on-disk log. In event of errors or filesystem shutdown, there are various potential cleanup scenarios depending on the state of the EFI/EFD. The cleanup scenarios are confusing and racy, as demonstrated by the following test sequence: # mount $dev $mnt # fsstress -d $mnt -n 99999 -p 16 -z -f fallocate=1 \ -f punch=1 -f creat=1 -f unlink=1 & # sleep 5 # killall -9 fsstress; wait # godown -f $mnt # umount ... in which the final umount can hang due to the AIL being pinned indefinitely by one or more EFI items. This can occur due to several conditions. For example, if the shutdown occurs after the EFI is committed to the on-disk log and the EFD committed to the CIL, but before the EFD committed to the log, the EFD iop_committed() abort handler does not drop its reference to the EFI. Alternatively, manual error injection in the xfs_bmap_finish() codepath shows that if an error occurs after the EFI transaction is committed but before the EFD is constructed and logged, the EFI is never released from the AIL. Update the EFI/EFD item handling code to use a more straightforward and reliable approach to error handling. If an error occurs after the EFI transaction is committed and before the EFD is constructed, release the EFI explicitly from xfs_bmap_finish(). If the EFI transaction is cancelled, release the EFI in the unlock handler. Once the EFD is constructed, it is responsible for releasing the EFI under any circumstances (including whether the EFI item aborts due to log I/O error). Update the EFD item handlers to release the EFI if the transaction is cancelled or aborts due to log I/O error. Finally, update xfs_bmap_finish() to log at least one EFD extent to the transaction before xfs_free_extent() errors are handled to ensure the transaction is dirty and EFD item error handling is triggered. Signed-off-by: Brian Foster <bfoster@redhat.com> Reviewed-by: Dave Chinner <dchinner@redhat.com> Signed-off-by: Dave Chinner <david@fromorbit.com>
2015-08-19 07:51:16 +08:00
/*
* The EFI has been either committed or aborted if the transaction has been
* cancelled. If the transaction was cancelled, an EFD isn't going to be
* constructed and thus we free the EFI here directly.
*/
STATIC void
xfs: split iop_unlock The iop_unlock method is called when comitting or cancelling a transaction. In the latter case, the transaction may or may not be aborted. While there is no known problem with the current code in practice, this implementation is limited in that any log item implementation that might want to differentiate between a commit and a cancellation must rely on the aborted state. The aborted bit is only set when the cancelled transaction is dirty, however. This means that there is no way to distinguish between a commit and a clean transaction cancellation. For example, intent log items currently rely on this distinction. The log item is either transferred to the CIL on commit or released on transaction cancel. There is currently no possibility for a clean intent log item in a transaction, but if that state is ever introduced a cancel of such a transaction will immediately result in memory leaks of the associated log item(s). This is an interface deficiency and landmine. To clean this up, replace the iop_unlock method with an iop_release method that is specific to transaction cancel. The existing iop_committing method occurs at the same time as iop_unlock in the commit path and there is no need for two separate callbacks here. Overload the iop_committing method with the current commit time iop_unlock implementations to eliminate the need for the latter and further simplify the interface. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Brian Foster <bfoster@redhat.com> Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com> Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
2019-06-29 10:27:32 +08:00
xfs_efi_item_release(
struct xfs_log_item *lip)
{
xfs: split iop_unlock The iop_unlock method is called when comitting or cancelling a transaction. In the latter case, the transaction may or may not be aborted. While there is no known problem with the current code in practice, this implementation is limited in that any log item implementation that might want to differentiate between a commit and a cancellation must rely on the aborted state. The aborted bit is only set when the cancelled transaction is dirty, however. This means that there is no way to distinguish between a commit and a clean transaction cancellation. For example, intent log items currently rely on this distinction. The log item is either transferred to the CIL on commit or released on transaction cancel. There is currently no possibility for a clean intent log item in a transaction, but if that state is ever introduced a cancel of such a transaction will immediately result in memory leaks of the associated log item(s). This is an interface deficiency and landmine. To clean this up, replace the iop_unlock method with an iop_release method that is specific to transaction cancel. The existing iop_committing method occurs at the same time as iop_unlock in the commit path and there is no need for two separate callbacks here. Overload the iop_committing method with the current commit time iop_unlock implementations to eliminate the need for the latter and further simplify the interface. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Brian Foster <bfoster@redhat.com> Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com> Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
2019-06-29 10:27:32 +08:00
xfs_efi_release(EFI_ITEM(lip));
}
/*
* Allocate and initialize an efi item with the given number of extents.
*/
STATIC struct xfs_efi_log_item *
xfs_efi_init(
struct xfs_mount *mp,
uint nextents)
{
struct xfs_efi_log_item *efip;
uint size;
ASSERT(nextents > 0);
if (nextents > XFS_EFI_MAX_FAST_EXTENTS) {
size = (uint)(sizeof(struct xfs_efi_log_item) +
((nextents - 1) * sizeof(xfs_extent_t)));
efip = kmem_zalloc(size, 0);
} else {
efip = kmem_cache_zalloc(xfs_efi_zone,
GFP_KERNEL | __GFP_NOFAIL);
}
xfs_log_item_init(mp, &efip->efi_item, XFS_LI_EFI, &xfs_efi_item_ops);
efip->efi_format.efi_nextents = nextents;
efip->efi_format.efi_id = (uintptr_t)(void *)efip;
atomic_set(&efip->efi_next_extent, 0);
xfs: don't free EFIs before the EFDs are committed Filesystems are occasionally being shut down with this error: xfs_trans_ail_delete_bulk: attempting to delete a log item that is not in the AIL. It was diagnosed to be related to the EFI/EFD commit order when the EFI and EFD are in different checkpoints and the EFD is committed before the EFI here: http://oss.sgi.com/archives/xfs/2013-01/msg00082.html The real problem is that a single bit cannot fully describe the states that the EFI/EFD processing can be in. These completion states are: EFI EFI in AIL EFD Result committed/unpinned Yes committed OK committed/pinned No committed Shutdown uncommitted No committed Shutdown Note that the "result" field is what should happen, not what does happen. The current logic is broken and handles the first two cases correctly by luck. That is, the code will free the EFI if the XFS_EFI_COMMITTED bit is *not* set, rather than if it is set. The inverted logic "works" because if both EFI and EFD are committed, then the first __xfs_efi_release() call clears the XFS_EFI_COMMITTED bit, and the second frees the EFI item. Hence as long as xfs_efi_item_committed() has been called, everything appears to be fine. It is the third case where the logic fails - where xfs_efd_item_committed() is called before xfs_efi_item_committed(), and that results in the EFI being freed before it has been committed. That is the bug that triggered the shutdown, and hence keeping track of whether the EFI has been committed or not is insufficient to correctly order the EFI/EFD operations w.r.t. the AIL. What we really want is this: the EFI is always placed into the AIL before the last reference goes away. The only way to guarantee that is that the EFI is not freed until after it has been unpinned *and* the EFD has been committed. That is, restructure the logic so that the only case that can occur is the first case. This can be done easily by replacing the XFS_EFI_COMMITTED with an EFI reference count. The EFI is initialised with it's own count, and that is not released until it is unpinned. However, there is a complication to this method - the high level EFI/EFD code in xfs_bmap_finish() does not hold direct references to the EFI structure, and runs a transaction commit between the EFI and EFD processing. Hence the EFI can be freed even before the EFD is created using such a method. Further, log recovery uses the AIL for tracking EFI/EFDs that need to be recovered, but it uses the AIL *differently* to the EFI transaction commit. Hence log recovery never pins or unpins EFIs, so we can't drop the EFI reference count indirectly to free the EFI. However, this doesn't prevent us from using a reference count here. There is a 1:1 relationship between EFIs and EFDs, so when we initialise the EFI we can take a reference count for the EFD as well. This solves the xfs_bmap_finish() issue - the EFI will never be freed until the EFD is processed. In terms of log recovery, during the committing of the EFD we can look for the XFS_EFI_RECOVERED bit being set and drop the EFI reference as well, thereby ensuring everything works correctly there as well. Signed-off-by: Dave Chinner <dchinner@redhat.com> Reviewed-by: Mark Tinguely <tinguely@sgi.com> Signed-off-by: Ben Myers <bpm@sgi.com>
2013-04-03 11:09:21 +08:00
atomic_set(&efip->efi_refcount, 2);
return efip;
}
/*
* Copy an EFI format buffer from the given buf, and into the destination
* EFI format structure.
* The given buffer can be in 32 bit or 64 bit form (which has different padding),
* one of which will be the native format for this kernel.
* It will handle the conversion of formats if necessary.
*/
STATIC int
xfs_efi_copy_format(xfs_log_iovec_t *buf, xfs_efi_log_format_t *dst_efi_fmt)
{
xfs_efi_log_format_t *src_efi_fmt = buf->i_addr;
uint i;
uint len = sizeof(xfs_efi_log_format_t) +
(src_efi_fmt->efi_nextents - 1) * sizeof(xfs_extent_t);
uint len32 = sizeof(xfs_efi_log_format_32_t) +
(src_efi_fmt->efi_nextents - 1) * sizeof(xfs_extent_32_t);
uint len64 = sizeof(xfs_efi_log_format_64_t) +
(src_efi_fmt->efi_nextents - 1) * sizeof(xfs_extent_64_t);
if (buf->i_len == len) {
memcpy((char *)dst_efi_fmt, (char*)src_efi_fmt, len);
return 0;
} else if (buf->i_len == len32) {
xfs_efi_log_format_32_t *src_efi_fmt_32 = buf->i_addr;
dst_efi_fmt->efi_type = src_efi_fmt_32->efi_type;
dst_efi_fmt->efi_size = src_efi_fmt_32->efi_size;
dst_efi_fmt->efi_nextents = src_efi_fmt_32->efi_nextents;
dst_efi_fmt->efi_id = src_efi_fmt_32->efi_id;
for (i = 0; i < dst_efi_fmt->efi_nextents; i++) {
dst_efi_fmt->efi_extents[i].ext_start =
src_efi_fmt_32->efi_extents[i].ext_start;
dst_efi_fmt->efi_extents[i].ext_len =
src_efi_fmt_32->efi_extents[i].ext_len;
}
return 0;
} else if (buf->i_len == len64) {
xfs_efi_log_format_64_t *src_efi_fmt_64 = buf->i_addr;
dst_efi_fmt->efi_type = src_efi_fmt_64->efi_type;
dst_efi_fmt->efi_size = src_efi_fmt_64->efi_size;
dst_efi_fmt->efi_nextents = src_efi_fmt_64->efi_nextents;
dst_efi_fmt->efi_id = src_efi_fmt_64->efi_id;
for (i = 0; i < dst_efi_fmt->efi_nextents; i++) {
dst_efi_fmt->efi_extents[i].ext_start =
src_efi_fmt_64->efi_extents[i].ext_start;
dst_efi_fmt->efi_extents[i].ext_len =
src_efi_fmt_64->efi_extents[i].ext_len;
}
return 0;
}
XFS_ERROR_REPORT(__func__, XFS_ERRLEVEL_LOW, NULL);
return -EFSCORRUPTED;
}
static inline struct xfs_efd_log_item *EFD_ITEM(struct xfs_log_item *lip)
{
return container_of(lip, struct xfs_efd_log_item, efd_item);
}
STATIC void
xfs_efd_item_free(struct xfs_efd_log_item *efdp)
{
xfs: allocate log vector buffers outside CIL context lock One of the problems we currently have with delayed logging is that under serious memory pressure we can deadlock memory reclaim. THis occurs when memory reclaim (such as run by kswapd) is reclaiming XFS inodes and issues a log force to unpin inodes that are dirty in the CIL. The CIL is pushed, but this will only occur once it gets the CIL context lock to ensure that all committing transactions are complete and no new transactions start being committed to the CIL while the push switches to a new context. The deadlock occurs when the CIL context lock is held by a committing process that is doing memory allocation for log vector buffers, and that allocation is then blocked on memory reclaim making progress. Memory reclaim, however, is blocked waiting for a log force to make progress, and so we effectively deadlock at this point. To solve this problem, we have to move the CIL log vector buffer allocation outside of the context lock so that memory reclaim can always make progress when it needs to force the log. The problem with doing this is that a CIL push can take place while we are determining if we need to allocate a new log vector buffer for an item and hence the current log vector may go away without warning. That means we canot rely on the existing log vector being present when we finally grab the context lock and so we must have a replacement buffer ready to go at all times. To ensure this, introduce a "shadow log vector" buffer that is always guaranteed to be present when we gain the CIL context lock and format the item. This shadow buffer may or may not be used during the formatting, but if the log item does not have an existing log vector buffer or that buffer is too small for the new modifications, we swap it for the new shadow buffer and format the modifications into that new log vector buffer. The result of this is that for any object we modify more than once in a given CIL checkpoint, we double the memory required to track dirty regions in the log. For single modifications then we consume the shadow log vectorwe allocate on commit, and that gets consumed by the checkpoint. However, if we make multiple modifications, then the second transaction commit will allocate a shadow log vector and hence we will end up with double the memory usage as only one of the log vectors is consumed by the CIL checkpoint. The remaining shadow vector will be freed when th elog item is freed. This can probably be optimised in future - access to the shadow log vector is serialised by the object lock (as opposited to the active log vector, which is controlled by the CIL context lock) and so we can probably free shadow log vector from some objects when the log item is marked clean on removal from the AIL. Signed-off-by: Dave Chinner <dchinner@redhat.com> Reviewed-by: Brian Foster <bfoster@redhat.com> Signed-off-by: Dave Chinner <david@fromorbit.com>
2016-07-22 07:52:35 +08:00
kmem_free(efdp->efd_item.li_lv_shadow);
if (efdp->efd_format.efd_nextents > XFS_EFD_MAX_FAST_EXTENTS)
kmem_free(efdp);
else
kmem_cache_free(xfs_efd_zone, efdp);
}
/*
* This returns the number of iovecs needed to log the given efd item.
* We only need 1 iovec for an efd item. It just logs the efd_log_format
* structure.
*/
static inline int
xfs_efd_item_sizeof(
struct xfs_efd_log_item *efdp)
{
return sizeof(xfs_efd_log_format_t) +
(efdp->efd_format.efd_nextents - 1) * sizeof(xfs_extent_t);
}
STATIC void
xfs_efd_item_size(
struct xfs_log_item *lip,
int *nvecs,
int *nbytes)
{
*nvecs += 1;
*nbytes += xfs_efd_item_sizeof(EFD_ITEM(lip));
}
/*
* This is called to fill in the vector of log iovecs for the
* given efd log item. We use only 1 iovec, and we point that
* at the efd_log_format structure embedded in the efd item.
* It is at this point that we assert that all of the extent
* slots in the efd item have been filled.
*/
STATIC void
xfs_efd_item_format(
struct xfs_log_item *lip,
struct xfs_log_vec *lv)
{
struct xfs_efd_log_item *efdp = EFD_ITEM(lip);
struct xfs_log_iovec *vecp = NULL;
ASSERT(efdp->efd_next_extent == efdp->efd_format.efd_nextents);
efdp->efd_format.efd_type = XFS_LI_EFD;
efdp->efd_format.efd_size = 1;
xlog_copy_iovec(lv, &vecp, XLOG_REG_TYPE_EFD_FORMAT,
&efdp->efd_format,
xfs_efd_item_sizeof(efdp));
}
xfs: fix efi/efd error handling to avoid fs shutdown hangs Freeing an extent in XFS involves logging an EFI (extent free intention), freeing the actual extent, and logging an EFD (extent free done). The EFI object is created with a reference count of 2: one for the current transaction and one for the subsequently created EFD. Under normal circumstances, the first reference is dropped when the EFI is unpinned and the second reference is dropped when the EFD is committed to the on-disk log. In event of errors or filesystem shutdown, there are various potential cleanup scenarios depending on the state of the EFI/EFD. The cleanup scenarios are confusing and racy, as demonstrated by the following test sequence: # mount $dev $mnt # fsstress -d $mnt -n 99999 -p 16 -z -f fallocate=1 \ -f punch=1 -f creat=1 -f unlink=1 & # sleep 5 # killall -9 fsstress; wait # godown -f $mnt # umount ... in which the final umount can hang due to the AIL being pinned indefinitely by one or more EFI items. This can occur due to several conditions. For example, if the shutdown occurs after the EFI is committed to the on-disk log and the EFD committed to the CIL, but before the EFD committed to the log, the EFD iop_committed() abort handler does not drop its reference to the EFI. Alternatively, manual error injection in the xfs_bmap_finish() codepath shows that if an error occurs after the EFI transaction is committed but before the EFD is constructed and logged, the EFI is never released from the AIL. Update the EFI/EFD item handling code to use a more straightforward and reliable approach to error handling. If an error occurs after the EFI transaction is committed and before the EFD is constructed, release the EFI explicitly from xfs_bmap_finish(). If the EFI transaction is cancelled, release the EFI in the unlock handler. Once the EFD is constructed, it is responsible for releasing the EFI under any circumstances (including whether the EFI item aborts due to log I/O error). Update the EFD item handlers to release the EFI if the transaction is cancelled or aborts due to log I/O error. Finally, update xfs_bmap_finish() to log at least one EFD extent to the transaction before xfs_free_extent() errors are handled to ensure the transaction is dirty and EFD item error handling is triggered. Signed-off-by: Brian Foster <bfoster@redhat.com> Reviewed-by: Dave Chinner <dchinner@redhat.com> Signed-off-by: Dave Chinner <david@fromorbit.com>
2015-08-19 07:51:16 +08:00
/*
* The EFD is either committed or aborted if the transaction is cancelled. If
* the transaction is cancelled, drop our reference to the EFI and free the EFD.
*/
STATIC void
xfs: split iop_unlock The iop_unlock method is called when comitting or cancelling a transaction. In the latter case, the transaction may or may not be aborted. While there is no known problem with the current code in practice, this implementation is limited in that any log item implementation that might want to differentiate between a commit and a cancellation must rely on the aborted state. The aborted bit is only set when the cancelled transaction is dirty, however. This means that there is no way to distinguish between a commit and a clean transaction cancellation. For example, intent log items currently rely on this distinction. The log item is either transferred to the CIL on commit or released on transaction cancel. There is currently no possibility for a clean intent log item in a transaction, but if that state is ever introduced a cancel of such a transaction will immediately result in memory leaks of the associated log item(s). This is an interface deficiency and landmine. To clean this up, replace the iop_unlock method with an iop_release method that is specific to transaction cancel. The existing iop_committing method occurs at the same time as iop_unlock in the commit path and there is no need for two separate callbacks here. Overload the iop_committing method with the current commit time iop_unlock implementations to eliminate the need for the latter and further simplify the interface. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Brian Foster <bfoster@redhat.com> Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com> Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
2019-06-29 10:27:32 +08:00
xfs_efd_item_release(
struct xfs_log_item *lip)
{
xfs: fix efi/efd error handling to avoid fs shutdown hangs Freeing an extent in XFS involves logging an EFI (extent free intention), freeing the actual extent, and logging an EFD (extent free done). The EFI object is created with a reference count of 2: one for the current transaction and one for the subsequently created EFD. Under normal circumstances, the first reference is dropped when the EFI is unpinned and the second reference is dropped when the EFD is committed to the on-disk log. In event of errors or filesystem shutdown, there are various potential cleanup scenarios depending on the state of the EFI/EFD. The cleanup scenarios are confusing and racy, as demonstrated by the following test sequence: # mount $dev $mnt # fsstress -d $mnt -n 99999 -p 16 -z -f fallocate=1 \ -f punch=1 -f creat=1 -f unlink=1 & # sleep 5 # killall -9 fsstress; wait # godown -f $mnt # umount ... in which the final umount can hang due to the AIL being pinned indefinitely by one or more EFI items. This can occur due to several conditions. For example, if the shutdown occurs after the EFI is committed to the on-disk log and the EFD committed to the CIL, but before the EFD committed to the log, the EFD iop_committed() abort handler does not drop its reference to the EFI. Alternatively, manual error injection in the xfs_bmap_finish() codepath shows that if an error occurs after the EFI transaction is committed but before the EFD is constructed and logged, the EFI is never released from the AIL. Update the EFI/EFD item handling code to use a more straightforward and reliable approach to error handling. If an error occurs after the EFI transaction is committed and before the EFD is constructed, release the EFI explicitly from xfs_bmap_finish(). If the EFI transaction is cancelled, release the EFI in the unlock handler. Once the EFD is constructed, it is responsible for releasing the EFI under any circumstances (including whether the EFI item aborts due to log I/O error). Update the EFD item handlers to release the EFI if the transaction is cancelled or aborts due to log I/O error. Finally, update xfs_bmap_finish() to log at least one EFD extent to the transaction before xfs_free_extent() errors are handled to ensure the transaction is dirty and EFD item error handling is triggered. Signed-off-by: Brian Foster <bfoster@redhat.com> Reviewed-by: Dave Chinner <dchinner@redhat.com> Signed-off-by: Dave Chinner <david@fromorbit.com>
2015-08-19 07:51:16 +08:00
struct xfs_efd_log_item *efdp = EFD_ITEM(lip);
xfs: split iop_unlock The iop_unlock method is called when comitting or cancelling a transaction. In the latter case, the transaction may or may not be aborted. While there is no known problem with the current code in practice, this implementation is limited in that any log item implementation that might want to differentiate between a commit and a cancellation must rely on the aborted state. The aborted bit is only set when the cancelled transaction is dirty, however. This means that there is no way to distinguish between a commit and a clean transaction cancellation. For example, intent log items currently rely on this distinction. The log item is either transferred to the CIL on commit or released on transaction cancel. There is currently no possibility for a clean intent log item in a transaction, but if that state is ever introduced a cancel of such a transaction will immediately result in memory leaks of the associated log item(s). This is an interface deficiency and landmine. To clean this up, replace the iop_unlock method with an iop_release method that is specific to transaction cancel. The existing iop_committing method occurs at the same time as iop_unlock in the commit path and there is no need for two separate callbacks here. Overload the iop_committing method with the current commit time iop_unlock implementations to eliminate the need for the latter and further simplify the interface. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Brian Foster <bfoster@redhat.com> Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com> Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
2019-06-29 10:27:32 +08:00
xfs_efi_release(efdp->efd_efip);
xfs_efd_item_free(efdp);
}
static const struct xfs_item_ops xfs_efd_item_ops = {
.flags = XFS_ITEM_RELEASE_WHEN_COMMITTED,
.iop_size = xfs_efd_item_size,
.iop_format = xfs_efd_item_format,
xfs: split iop_unlock The iop_unlock method is called when comitting or cancelling a transaction. In the latter case, the transaction may or may not be aborted. While there is no known problem with the current code in practice, this implementation is limited in that any log item implementation that might want to differentiate between a commit and a cancellation must rely on the aborted state. The aborted bit is only set when the cancelled transaction is dirty, however. This means that there is no way to distinguish between a commit and a clean transaction cancellation. For example, intent log items currently rely on this distinction. The log item is either transferred to the CIL on commit or released on transaction cancel. There is currently no possibility for a clean intent log item in a transaction, but if that state is ever introduced a cancel of such a transaction will immediately result in memory leaks of the associated log item(s). This is an interface deficiency and landmine. To clean this up, replace the iop_unlock method with an iop_release method that is specific to transaction cancel. The existing iop_committing method occurs at the same time as iop_unlock in the commit path and there is no need for two separate callbacks here. Overload the iop_committing method with the current commit time iop_unlock implementations to eliminate the need for the latter and further simplify the interface. Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Brian Foster <bfoster@redhat.com> Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com> Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
2019-06-29 10:27:32 +08:00
.iop_release = xfs_efd_item_release,
};
/*
* Allocate an "extent free done" log item that will hold nextents worth of
* extents. The caller must use all nextents extents, because we are not
* flexible about this at all.
*/
static struct xfs_efd_log_item *
xfs_trans_get_efd(
struct xfs_trans *tp,
struct xfs_efi_log_item *efip,
unsigned int nextents)
{
struct xfs_efd_log_item *efdp;
ASSERT(nextents > 0);
if (nextents > XFS_EFD_MAX_FAST_EXTENTS) {
efdp = kmem_zalloc(sizeof(struct xfs_efd_log_item) +
(nextents - 1) * sizeof(struct xfs_extent),
0);
} else {
efdp = kmem_cache_zalloc(xfs_efd_zone,
GFP_KERNEL | __GFP_NOFAIL);
}
xfs_log_item_init(tp->t_mountp, &efdp->efd_item, XFS_LI_EFD,
&xfs_efd_item_ops);
efdp->efd_efip = efip;
efdp->efd_format.efd_nextents = nextents;
efdp->efd_format.efd_efi_id = efip->efi_format.efi_id;
xfs_trans_add_item(tp, &efdp->efd_item);
return efdp;
}
/*
* Free an extent and log it to the EFD. Note that the transaction is marked
* dirty regardless of whether the extent free succeeds or fails to support the
* EFI/EFD lifecycle rules.
*/
static int
xfs_trans_free_extent(
struct xfs_trans *tp,
struct xfs_efd_log_item *efdp,
xfs_fsblock_t start_block,
xfs_extlen_t ext_len,
const struct xfs_owner_info *oinfo,
bool skip_discard)
{
struct xfs_mount *mp = tp->t_mountp;
struct xfs_extent *extp;
uint next_extent;
xfs_agnumber_t agno = XFS_FSB_TO_AGNO(mp, start_block);
xfs_agblock_t agbno = XFS_FSB_TO_AGBNO(mp,
start_block);
int error;
trace_xfs_bmap_free_deferred(tp->t_mountp, agno, 0, agbno, ext_len);
error = __xfs_free_extent(tp, start_block, ext_len,
oinfo, XFS_AG_RESV_NONE, skip_discard);
/*
* Mark the transaction dirty, even on error. This ensures the
* transaction is aborted, which:
*
* 1.) releases the EFI and frees the EFD
* 2.) shuts down the filesystem
*/
tp->t_flags |= XFS_TRANS_DIRTY;
set_bit(XFS_LI_DIRTY, &efdp->efd_item.li_flags);
next_extent = efdp->efd_next_extent;
ASSERT(next_extent < efdp->efd_format.efd_nextents);
extp = &(efdp->efd_format.efd_extents[next_extent]);
extp->ext_start = start_block;
extp->ext_len = ext_len;
efdp->efd_next_extent++;
return error;
}
/* Sort bmap items by AG. */
static int
xfs_extent_free_diff_items(
void *priv,
const struct list_head *a,
const struct list_head *b)
{
struct xfs_mount *mp = priv;
struct xfs_extent_free_item *ra;
struct xfs_extent_free_item *rb;
ra = container_of(a, struct xfs_extent_free_item, xefi_list);
rb = container_of(b, struct xfs_extent_free_item, xefi_list);
return XFS_FSB_TO_AGNO(mp, ra->xefi_startblock) -
XFS_FSB_TO_AGNO(mp, rb->xefi_startblock);
}
/* Log a free extent to the intent item. */
STATIC void
xfs_extent_free_log_item(
struct xfs_trans *tp,
struct xfs_efi_log_item *efip,
struct xfs_extent_free_item *free)
{
uint next_extent;
struct xfs_extent *extp;
tp->t_flags |= XFS_TRANS_DIRTY;
set_bit(XFS_LI_DIRTY, &efip->efi_item.li_flags);
/*
* atomic_inc_return gives us the value after the increment;
* we want to use it as an array index so we need to subtract 1 from
* it.
*/
next_extent = atomic_inc_return(&efip->efi_next_extent) - 1;
ASSERT(next_extent < efip->efi_format.efi_nextents);
extp = &efip->efi_format.efi_extents[next_extent];
extp->ext_start = free->xefi_startblock;
extp->ext_len = free->xefi_blockcount;
}
static struct xfs_log_item *
xfs_extent_free_create_intent(
struct xfs_trans *tp,
struct list_head *items,
unsigned int count,
bool sort)
{
struct xfs_mount *mp = tp->t_mountp;
struct xfs_efi_log_item *efip = xfs_efi_init(mp, count);
struct xfs_extent_free_item *free;
ASSERT(count > 0);
xfs_trans_add_item(tp, &efip->efi_item);
if (sort)
list_sort(mp, items, xfs_extent_free_diff_items);
list_for_each_entry(free, items, xefi_list)
xfs_extent_free_log_item(tp, efip, free);
return &efip->efi_item;
}
/* Get an EFD so we can process all the free extents. */
static struct xfs_log_item *
xfs_extent_free_create_done(
struct xfs_trans *tp,
struct xfs_log_item *intent,
unsigned int count)
{
return &xfs_trans_get_efd(tp, EFI_ITEM(intent), count)->efd_item;
}
/* Process a free extent. */
STATIC int
xfs_extent_free_finish_item(
struct xfs_trans *tp,
struct xfs_log_item *done,
struct list_head *item,
struct xfs_btree_cur **state)
{
struct xfs_extent_free_item *free;
int error;
free = container_of(item, struct xfs_extent_free_item, xefi_list);
error = xfs_trans_free_extent(tp, EFD_ITEM(done),
free->xefi_startblock,
free->xefi_blockcount,
&free->xefi_oinfo, free->xefi_skip_discard);
kmem_free(free);
return error;
}
/* Abort all pending EFIs. */
STATIC void
xfs_extent_free_abort_intent(
struct xfs_log_item *intent)
{
xfs_efi_release(EFI_ITEM(intent));
}
/* Cancel a free extent. */
STATIC void
xfs_extent_free_cancel_item(
struct list_head *item)
{
struct xfs_extent_free_item *free;
free = container_of(item, struct xfs_extent_free_item, xefi_list);
kmem_free(free);
}
const struct xfs_defer_op_type xfs_extent_free_defer_type = {
.max_items = XFS_EFI_MAX_FAST_EXTENTS,
.create_intent = xfs_extent_free_create_intent,
.abort_intent = xfs_extent_free_abort_intent,
.create_done = xfs_extent_free_create_done,
.finish_item = xfs_extent_free_finish_item,
.cancel_item = xfs_extent_free_cancel_item,
};
/*
* AGFL blocks are accounted differently in the reserve pools and are not
* inserted into the busy extent list.
*/
STATIC int
xfs_agfl_free_finish_item(
struct xfs_trans *tp,
struct xfs_log_item *done,
struct list_head *item,
struct xfs_btree_cur **state)
{
struct xfs_mount *mp = tp->t_mountp;
struct xfs_efd_log_item *efdp = EFD_ITEM(done);
struct xfs_extent_free_item *free;
struct xfs_extent *extp;
struct xfs_buf *agbp;
int error;
xfs_agnumber_t agno;
xfs_agblock_t agbno;
uint next_extent;
free = container_of(item, struct xfs_extent_free_item, xefi_list);
ASSERT(free->xefi_blockcount == 1);
agno = XFS_FSB_TO_AGNO(mp, free->xefi_startblock);
agbno = XFS_FSB_TO_AGBNO(mp, free->xefi_startblock);
trace_xfs_agfl_free_deferred(mp, agno, 0, agbno, free->xefi_blockcount);
error = xfs_alloc_read_agf(mp, tp, agno, 0, &agbp);
if (!error)
error = xfs_free_agfl_block(tp, agno, agbno, agbp,
&free->xefi_oinfo);
/*
* Mark the transaction dirty, even on error. This ensures the
* transaction is aborted, which:
*
* 1.) releases the EFI and frees the EFD
* 2.) shuts down the filesystem
*/
tp->t_flags |= XFS_TRANS_DIRTY;
set_bit(XFS_LI_DIRTY, &efdp->efd_item.li_flags);
next_extent = efdp->efd_next_extent;
ASSERT(next_extent < efdp->efd_format.efd_nextents);
extp = &(efdp->efd_format.efd_extents[next_extent]);
extp->ext_start = free->xefi_startblock;
extp->ext_len = free->xefi_blockcount;
efdp->efd_next_extent++;
kmem_free(free);
return error;
}
/* sub-type with special handling for AGFL deferred frees */
const struct xfs_defer_op_type xfs_agfl_free_defer_type = {
.max_items = XFS_EFI_MAX_FAST_EXTENTS,
.create_intent = xfs_extent_free_create_intent,
.abort_intent = xfs_extent_free_abort_intent,
.create_done = xfs_extent_free_create_done,
.finish_item = xfs_agfl_free_finish_item,
.cancel_item = xfs_extent_free_cancel_item,
};
/* Is this recovered EFI ok? */
static inline bool
xfs_efi_validate_ext(
struct xfs_mount *mp,
struct xfs_extent *extp)
{
return xfs_verify_fsbext(mp, extp->ext_start, extp->ext_len);
}
/*
* Process an extent free intent item that was recovered from
* the log. We need to free the extents that it describes.
*/
STATIC int
xfs_efi_item_recover(
struct xfs_log_item *lip,
xfs: proper replay of deferred ops queued during log recovery When we replay unfinished intent items that have been recovered from the log, it's possible that the replay will cause the creation of more deferred work items. As outlined in commit 509955823cc9c ("xfs: log recovery should replay deferred ops in order"), later work items have an implicit ordering dependency on earlier work items. Therefore, recovery must replay the items (both recovered and created) in the same order that they would have been during normal operation. For log recovery, we enforce this ordering by using an empty transaction to collect deferred ops that get created in the process of recovering a log intent item to prevent them from being committed before the rest of the recovered intent items. After we finish committing all the recovered log items, we allocate a transaction with an enormous block reservation, splice our huge list of created deferred ops into that transaction, and commit it, thereby finishing all those ops. This is /really/ hokey -- it's the one place in XFS where we allow nested transactions; the splicing of the defer ops list is is inelegant and has to be done twice per recovery function; and the broken way we handle inode pointers and block reservations cause subtle use-after-free and allocator problems that will be fixed by this patch and the two patches after it. Therefore, replace the hokey empty transaction with a structure designed to capture each chain of deferred ops that are created as part of recovering a single unfinished log intent. Finally, refactor the loop that replays those chains to do so using one transaction per chain. Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com> Reviewed-by: Brian Foster <bfoster@redhat.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2020-09-26 08:39:37 +08:00
struct list_head *capture_list)
{
struct xfs_efi_log_item *efip = EFI_ITEM(lip);
xfs: proper replay of deferred ops queued during log recovery When we replay unfinished intent items that have been recovered from the log, it's possible that the replay will cause the creation of more deferred work items. As outlined in commit 509955823cc9c ("xfs: log recovery should replay deferred ops in order"), later work items have an implicit ordering dependency on earlier work items. Therefore, recovery must replay the items (both recovered and created) in the same order that they would have been during normal operation. For log recovery, we enforce this ordering by using an empty transaction to collect deferred ops that get created in the process of recovering a log intent item to prevent them from being committed before the rest of the recovered intent items. After we finish committing all the recovered log items, we allocate a transaction with an enormous block reservation, splice our huge list of created deferred ops into that transaction, and commit it, thereby finishing all those ops. This is /really/ hokey -- it's the one place in XFS where we allow nested transactions; the splicing of the defer ops list is is inelegant and has to be done twice per recovery function; and the broken way we handle inode pointers and block reservations cause subtle use-after-free and allocator problems that will be fixed by this patch and the two patches after it. Therefore, replace the hokey empty transaction with a structure designed to capture each chain of deferred ops that are created as part of recovering a single unfinished log intent. Finally, refactor the loop that replays those chains to do so using one transaction per chain. Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com> Reviewed-by: Brian Foster <bfoster@redhat.com> Reviewed-by: Christoph Hellwig <hch@lst.de>
2020-09-26 08:39:37 +08:00
struct xfs_mount *mp = lip->li_mountp;
struct xfs_efd_log_item *efdp;
struct xfs_trans *tp;
struct xfs_extent *extp;
int i;
int error = 0;
/*
* First check the validity of the extents described by the
* EFI. If any are bad, then assume that all are bad and
* just toss the EFI.
*/
for (i = 0; i < efip->efi_format.efi_nextents; i++) {
if (!xfs_efi_validate_ext(mp,
&efip->efi_format.efi_extents[i])) {
XFS_CORRUPTION_ERROR(__func__, XFS_ERRLEVEL_LOW, mp,
&efip->efi_format,
sizeof(efip->efi_format));
return -EFSCORRUPTED;
}
}
error = xfs_trans_alloc(mp, &M_RES(mp)->tr_itruncate, 0, 0, 0, &tp);
if (error)
return error;
efdp = xfs_trans_get_efd(tp, efip, efip->efi_format.efi_nextents);
for (i = 0; i < efip->efi_format.efi_nextents; i++) {
extp = &efip->efi_format.efi_extents[i];
error = xfs_trans_free_extent(tp, efdp, extp->ext_start,
extp->ext_len,
&XFS_RMAP_OINFO_ANY_OWNER, false);
if (error)
goto abort_error;
}
return xfs_defer_ops_capture_and_commit(tp, NULL, capture_list);
abort_error:
xfs_trans_cancel(tp);
return error;
}
STATIC bool
xfs_efi_item_match(
struct xfs_log_item *lip,
uint64_t intent_id)
{
return EFI_ITEM(lip)->efi_format.efi_id == intent_id;
}
xfs: periodically relog deferred intent items There's a subtle design flaw in the deferred log item code that can lead to pinning the log tail. Taking up the defer ops chain examples from the previous commit, we can get trapped in sequences like this: Caller hands us a transaction t0 with D0-D3 attached. The defer ops chain will look like the following if the transaction rolls succeed: t1: D0(t0), D1(t0), D2(t0), D3(t0) t2: d4(t1), d5(t1), D1(t0), D2(t0), D3(t0) t3: d5(t1), D1(t0), D2(t0), D3(t0) ... t9: d9(t7), D3(t0) t10: D3(t0) t11: d10(t10), d11(t10) t12: d11(t10) In transaction 9, we finish d9 and try to roll to t10 while holding onto an intent item for D3 that we logged in t0. The previous commit changed the order in which we place new defer ops in the defer ops processing chain to reduce the maximum chain length. Now make xfs_defer_finish_noroll capable of relogging the entire chain periodically so that we can always move the log tail forward. Most chains will never get relogged, except for operations that generate very long chains (large extents containing many blocks with different sharing levels) or are on filesystems with small logs and a lot of ongoing metadata updates. Callers are now required to ensure that the transaction reservation is large enough to handle logging done items and new intent items for the maximum possible chain length. Most callers are careful to keep the chain lengths low, so the overhead should be minimal. The decision to relog an intent item is made based on whether the intent was logged in a previous checkpoint, since there's no point in relogging an intent into the same checkpoint. Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com> Reviewed-by: Brian Foster <bfoster@redhat.com>
2020-09-28 07:18:13 +08:00
/* Relog an intent item to push the log tail forward. */
static struct xfs_log_item *
xfs_efi_item_relog(
struct xfs_log_item *intent,
struct xfs_trans *tp)
{
struct xfs_efd_log_item *efdp;
struct xfs_efi_log_item *efip;
struct xfs_extent *extp;
unsigned int count;
count = EFI_ITEM(intent)->efi_format.efi_nextents;
extp = EFI_ITEM(intent)->efi_format.efi_extents;
tp->t_flags |= XFS_TRANS_DIRTY;
efdp = xfs_trans_get_efd(tp, EFI_ITEM(intent), count);
efdp->efd_next_extent = count;
memcpy(efdp->efd_format.efd_extents, extp, count * sizeof(*extp));
set_bit(XFS_LI_DIRTY, &efdp->efd_item.li_flags);
efip = xfs_efi_init(tp->t_mountp, count);
memcpy(efip->efi_format.efi_extents, extp, count * sizeof(*extp));
atomic_set(&efip->efi_next_extent, count);
xfs_trans_add_item(tp, &efip->efi_item);
set_bit(XFS_LI_DIRTY, &efip->efi_item.li_flags);
return &efip->efi_item;
}
static const struct xfs_item_ops xfs_efi_item_ops = {
.iop_size = xfs_efi_item_size,
.iop_format = xfs_efi_item_format,
.iop_unpin = xfs_efi_item_unpin,
.iop_release = xfs_efi_item_release,
.iop_recover = xfs_efi_item_recover,
.iop_match = xfs_efi_item_match,
xfs: periodically relog deferred intent items There's a subtle design flaw in the deferred log item code that can lead to pinning the log tail. Taking up the defer ops chain examples from the previous commit, we can get trapped in sequences like this: Caller hands us a transaction t0 with D0-D3 attached. The defer ops chain will look like the following if the transaction rolls succeed: t1: D0(t0), D1(t0), D2(t0), D3(t0) t2: d4(t1), d5(t1), D1(t0), D2(t0), D3(t0) t3: d5(t1), D1(t0), D2(t0), D3(t0) ... t9: d9(t7), D3(t0) t10: D3(t0) t11: d10(t10), d11(t10) t12: d11(t10) In transaction 9, we finish d9 and try to roll to t10 while holding onto an intent item for D3 that we logged in t0. The previous commit changed the order in which we place new defer ops in the defer ops processing chain to reduce the maximum chain length. Now make xfs_defer_finish_noroll capable of relogging the entire chain periodically so that we can always move the log tail forward. Most chains will never get relogged, except for operations that generate very long chains (large extents containing many blocks with different sharing levels) or are on filesystems with small logs and a lot of ongoing metadata updates. Callers are now required to ensure that the transaction reservation is large enough to handle logging done items and new intent items for the maximum possible chain length. Most callers are careful to keep the chain lengths low, so the overhead should be minimal. The decision to relog an intent item is made based on whether the intent was logged in a previous checkpoint, since there's no point in relogging an intent into the same checkpoint. Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com> Reviewed-by: Brian Foster <bfoster@redhat.com>
2020-09-28 07:18:13 +08:00
.iop_relog = xfs_efi_item_relog,
};
/*
* This routine is called to create an in-core extent free intent
* item from the efi format structure which was logged on disk.
* It allocates an in-core efi, copies the extents from the format
* structure into it, and adds the efi to the AIL with the given
* LSN.
*/
STATIC int
xlog_recover_efi_commit_pass2(
struct xlog *log,
struct list_head *buffer_list,
struct xlog_recover_item *item,
xfs_lsn_t lsn)
{
struct xfs_mount *mp = log->l_mp;
struct xfs_efi_log_item *efip;
struct xfs_efi_log_format *efi_formatp;
int error;
efi_formatp = item->ri_buf[0].i_addr;
efip = xfs_efi_init(mp, efi_formatp->efi_nextents);
error = xfs_efi_copy_format(&item->ri_buf[0], &efip->efi_format);
if (error) {
xfs_efi_item_free(efip);
return error;
}
atomic_set(&efip->efi_next_extent, efi_formatp->efi_nextents);
/*
* Insert the intent into the AIL directly and drop one reference so
* that finishing or canceling the work will drop the other.
*/
xfs_trans_ail_insert(log->l_ailp, &efip->efi_item, lsn);
xfs_efi_release(efip);
return 0;
}
const struct xlog_recover_item_ops xlog_efi_item_ops = {
.item_type = XFS_LI_EFI,
.commit_pass2 = xlog_recover_efi_commit_pass2,
};
/*
* This routine is called when an EFD format structure is found in a committed
* transaction in the log. Its purpose is to cancel the corresponding EFI if it
* was still in the log. To do this it searches the AIL for the EFI with an id
* equal to that in the EFD format structure. If we find it we drop the EFD
* reference, which removes the EFI from the AIL and frees it.
*/
STATIC int
xlog_recover_efd_commit_pass2(
struct xlog *log,
struct list_head *buffer_list,
struct xlog_recover_item *item,
xfs_lsn_t lsn)
{
struct xfs_efd_log_format *efd_formatp;
efd_formatp = item->ri_buf[0].i_addr;
ASSERT((item->ri_buf[0].i_len == (sizeof(xfs_efd_log_format_32_t) +
((efd_formatp->efd_nextents - 1) * sizeof(xfs_extent_32_t)))) ||
(item->ri_buf[0].i_len == (sizeof(xfs_efd_log_format_64_t) +
((efd_formatp->efd_nextents - 1) * sizeof(xfs_extent_64_t)))));
xlog_recover_release_intent(log, XFS_LI_EFI, efd_formatp->efd_efi_id);
return 0;
}
const struct xlog_recover_item_ops xlog_efd_item_ops = {
.item_type = XFS_LI_EFD,
.commit_pass2 = xlog_recover_efd_commit_pass2,
};