xfs: CIL and log optimisations
Subsystems (3)
v4 cover xfs: CIL and log optimisations
Hi folks,
This is an update of the consolidated log scalability patchset I've been working on. Version 3 was posted here:
https://lore.kernel.org/linux-xfs/[email protected]/
This version addresses many of the review comments, fixes a couple of bugs in the CIL scalability series found by shutdown/log recovery tests, and has a heap more testing time run on it.
Performance improvements are largely documented in the change logs of the individual patches. Headline numbers are an increase in transaction rate from 700k commits/s to 1.7M commits/s, and a reduction in fua/flush operations by 2-3 orders of magnitude on metadata heavy workloads that don't use fsync.
Summary of series:
Patches Modifications
------- -------------
1-7: log write FUA/FLUSH optimisations
8: bug fix
9-11: Async CIL pushes
12-25: xlog_write() rework
26-39: CIL commit scalability
Change log is below.
Cheers,
Dave.
Version 4:
- rebase on 5.13-rc2+
- fixed completion logic for async cache flush
- trimmed superflous comments about not requiring REQ_PREFLUSH for iclog IO anymore.
- ensure that setting/clearing XLOG_ICL_NEED_FLUSH is atomic (i.e. only modified while holding the icloglock)
- ensure callers only add iclog flush/fua flags appropriately before releasing the iclog so that multiple independent writes to an iclog doesn't clear flags other writes into the iclog depend on.
- buffer log item dirty tracking patches merged so removed from series
- replaced XFS_LSN_CMP() checks with direct lsn1 == lsn2 comparisons to simplify the code
- changed "push_async" to "push_commit_stable" to indicate that the push caller wants the entire CIL checkpoint and commit record to be on stable storage when it completes.
- updated comment to indicate that iclog sync state is set according to the caller's desire for a stable checkpoint to be performed.
- Added comment explaining why the CIL workqueue is limited to 4 concurrent works per filesystem.
- debug overhead reduction patches merged so removed from series
- cleaned up a couple of typedef uses.
- updated pahole output for checkpoint header in commit message
- Added BUILD_BUG_ON() to check the size of unmount records.
- got rid of XFS_VOLUME define.
- got rid of XLOG_TIC_LEN_MAX define.
- cleaned up extra blank lines.
- fixed double initialisation of lv in xlog_write_single().
- removed the unnecessary change for reserved iclog space in xlog_state_get_iclog_space().
- no need to check for XLOG_CONTINUE_TRANS in xlog_write_partial() as it will always be set.
- added a patch for removing the optype parameter from xlog_write()
- removed unused nvecs from struct xfs_cil_ctx
- fixed whitespace damage in xlog_cil_pcp_dead()
- added missing cpu dead accounting transfer in xlog_cil_pcp_dead().
- factored out CIL push percpu structure aggregation into xlog_cil_pcp_aggregate()
- added the ctx->ticket->t_unit_res update back into the code even though it is largely unnecessary.
- cleaned up the pcp, cilpcp, pcptr mess in xlog_cil_pcp_alloc() and elsewhere to use variable names consistently.
- simplified the CIL sort comparison functions to a single comparison operation
- fixed percpu CIL item list sort order where items in the same transaction (order id) were reversed, leading to intents being replayed in the wrong order.
- split out log vector chain conversion to list_head into separate patch
- Updated documentation with all the fixes and suggestions made.
Version 3:
- rebase onto 5.12-rc1+
- aggregate many small dependent patchsets in one large one.
- simplify xlog_wait_on_iclog_lsn() back to just a call to xlog_wait_on_iclog()
- remove xfs_blkdev_issue_flush() instead of moving and renaming it.
- pass bio to xfs_flush_bdev_async() so it doesn't need allocation.
- skip cache flush in xfs_flush_bdev_async() if the underlying queue does not require it.
- fixed whitespace in xfs_flush_bdev_async()
- remove the implicit external log's data device cache flush code and replace it with an explicit flush in the unmount record write so that it works the same as the new CIL checkpoint cache pre-flush mechanism. This mechanism now guarantees metadata vs journal ordering for both internal and external logs.
- updated various commit messages
- fixed incorrect/unintended changes to xfs_log_force() behaviour
- typedef uint64_t xfs_csn_t; and conversion.
- removed stray trace_printk()s that were used for debugging.
- fixed minor formatting details.
- uninlined xlog_prepare_iovec()
- fixed up "lv chain vector and size calculation" commit message to reflect we are only calculating and passin gin the vector byte count.
- reworked the loop in xlog_write_single() based on Christoph's suggestion. Much cleaner!
- added patch to pass log ticket down to xlog_sync() so that it accounts the roundoff to the log ticket rather than directly modifying grant heads. Grant heads are hot, so every little bit helps.
- added patch to update delayed logging design doc with background material on how transactions and log space accounting works in XFS.
Version 2:
- fix ticket reservation roundoff to include 2 roundoffs
- removed stale copied comment from roundoff initialisation.
- clarified "separation" to mean "separation for ordering purposes" in commit message.
- added comment that newly activated, clean, empty iclogs have a LSN of 0 so are captured by the "iclog lsn < start_lsn" case that avoids needing to wait before releasing the commit iclog to be written.
- added async cache flush infrastructure
- convert CIL checkpoint push work it issue an unconditional metadata device cache flush rather than asking the first iclog write to issue it via REQ_PREFLUSH.
- cleaned up xlog_write() to remove a redundant parameter and prepare the logic for setting flags on the iclog based on the type of operational data is being written to the log.
- added XLOG_ICL_NEED_FUA flag to complement the NEED_FLUSH flag, allowing callers to issue explicit flushes and clear the NEED_FLUSH flag before the iclog is written without dropping the REQ_FUA requirement in /dev/null...
- added CIL commit-in-start-iclog optimisation that clears the NEED_FLUSH flag to avoid an unnecessary cache flush when issuing the iclog.
- fixed typo in CIL throttle bugfix comment.
- fixed trailing whitespace in commit message.
PATCH v4 1/39 xfs: log stripe roundoff is a property of the log
From: Dave Chinner <[email protected]>
We don't need to look at the xfs_mount and superblock every time we need to do an iclog roundoff calculation. The property is fixed for the life of the log, so store the roundoff in the log at mount time and use that everywhere.
On a debug build:
$ size fs/xfs/xfs_log.o.* text data bss dec hex filename 27360 560 8 27928 6d18 fs/xfs/xfs_log.o.orig 27219 560 8 27787 6c8b fs/xfs/xfs_log.o.patched
--- fs/xfs/libxfs/xfs_log_format.h | 3 -- fs/xfs/xfs_log.c | 59 ++++++++++++++-------------------- fs/xfs/xfs_log_priv.h | 2 ++ 3 files changed, 27 insertions(+), 37 deletions(-)
fs/xfs/libxfs/xfs_log_format.h +0 -3
@@ -34,9 +34,6 @@ typedef uint32_t xlog_tid_t;#define XLOG_MIN_RECORD_BSHIFT 14/* 16384 == 1 << 14 */ #define XLOG_BIG_RECORD_BSHIFT 15/* 32k == 1 << 15 */ #define XLOG_MAX_RECORD_BSHIFT 18/* 256k == 1 << 18 */ -#define XLOG_BTOLSUNIT (log, b) (((b)+(log)->l_mp->m_sb.sb_logsunit-1) / \ - (log)->l_mp->m_sb.sb_logsunit) -#define XLOG_LSUNITTOB (log, su) ((su) * (log)->l_mp->m_sb.sb_logsunit)#define XLOG_HEADER_SIZE 512
fs/xfs/xfs_log.c +25 -34
107 lines not on this page · raw message
fs/xfs/xfs_log_priv.h +2 -0
@@ -436,6 +436,8 @@ struct xlog {#endif /* log recovery lsn tracking (for buffer submission */ xfs_lsn_t l_recovery_lsn; + +uint32_t l_iclog_roundoff;/* padding roundoff */ };#define XLOG_BUF_CANCEL_BUCKET (log, blkno) \
Signature · 1 line
2.31.1
Re: [PATCH 01/39] xfs: log stripe roundoff is a property of the log
Quoted 14 lines · Dave Chinner
From: Dave Chinner <[email protected]>
We don't need to look at the
xfs_mountand superblock every time we need to do an iclog roundoff calculation. The property is fixed for the life of the log, so store the roundoff in the log at mount time and use that everywhere.On a debug build:
$ size fs/xfs/xfs_log.o.* text data bss dec hex filename 27360 560 8 27928 6d18 fs/xfs/xfs_log.o.orig 27219 560 8 27787 6c8b fs/xfs/xfs_log.o.patched
Quoted 4 lines
Signed-off-by Dave Chinner <[email protected]>Reviewed-by Chandan Babu R <[email protected]>Reviewed-by Darrick J. Wong <[email protected]>Reviewed-by Christoph Hellwig <[email protected]>
Ok makes sense
Quoted 151 lines
--- fs/xfs/libxfs/xfs_log_format.h | 3 -- fs/xfs/xfs_log.c | 59 ++++++++++++++-------------------- fs/xfs/xfs_log_priv.h | 2 ++ 3 files changed, 27 insertions(+), 37 deletions(-)fs/xfs/libxfs/xfs_log_format.h +0
-3@@ -34,9 +34,6 @@ typedef uint32_t xlog_tid_t;#define XLOG_MIN_RECORD_BSHIFT 14/* 16384 == 1 << 14 */ #define XLOG_BIG_RECORD_BSHIFT 15/* 32k == 1 << 15 */ #define XLOG_MAX_RECORD_BSHIFT 18/* 256k == 1 << 18 */ -#define XLOG_BTOLSUNIT (log, b) (((b)+(log)->l_mp->m_sb.sb_logsunit-1) / \ - (log)->l_mp->m_sb.sb_logsunit) -#define XLOG_LSUNITTOB (log, su) ((su) * (log)->l_mp->m_sb.sb_logsunit)#define XLOG_HEADER_SIZE 512fs/xfs/xfs_log.c +25
-34@@ -1401,6 +1401,11 @@ xlog_alloc_log(xlog_assign_atomic_lsn (&log->l_last_sync_lsn,1 ,0 ); log->l_curr_cycle =1 ;/* 0 is bad since this is initial value */ +if (xfs_sb_version_haslogv2 (&mp->m_sb) && mp->m_sb.sb_logsunit >1 ) + log->l_iclog_roundoff = mp->m_sb.sb_logsunit; +else + log->l_iclog_roundoff =BBSIZE ; +xlog_grant_head_init (&log->l_reserve_head);xlog_grant_head_init (&log->l_write_head);@@ -1854,29 +1859,15 @@ xlog_calc_iclog_size( uint32_t *roundoff) {uint32_t count_init, count; -bool use_lsunit; - - use_lsunit =xfs_sb_version_haslogv2 (&log->l_mp->m_sb) && - log->l_mp->m_sb.sb_logsunit >1 ;/* Add for LR header */ count_init = log->l_iclog_hsize + iclog->ic_offset; + count =roundup (count_init, log->l_iclog_roundoff); -/* Round out the log write size */ -if (use_lsunit) { -/* we have a v2 stripe unit to use */ - count =XLOG_LSUNITTOB (log,XLOG_BTOLSUNIT (log, count_init)); - }else { - count =BBTOB (BTOBB (count_init)); - } - -ASSERT (count >= count_init); *roundoff = count - count_init; -if (use_lsunit) -ASSERT (*roundoff < log->l_mp->m_sb.sb_logsunit); -else -ASSERT (*roundoff <BBTOB (1 )); +ASSERT (count >= count_init); +ASSERT (*roundoff < log->l_iclog_roundoff);return count; }@@ -3151,10 +3142,9 @@ xlog_state_switch_iclogs( log->l_curr_block +=BTOBB (eventual_size)+BTOBB (log->l_iclog_hsize);/* Round up to next log-sunit */ -if (xfs_sb_version_haslogv2 (&log->l_mp->m_sb) && - log->l_mp->m_sb.sb_logsunit >1 ) { -uint32_t sunit_bb =BTOBB (log->l_mp->m_sb.sb_logsunit); - log->l_curr_block =roundup (log->l_curr_block, sunit_bb); +if (log->l_iclog_roundoff >BBSIZE ) { + log->l_curr_block =roundup (log->l_curr_block, +BTOBB (log->l_iclog_roundoff)); }if (log->l_curr_block >= log->l_logBBsize) {@@ -3406,12 +3396,11 @@ xfs_log_ticket_get(* Figure out the total log space unit (in bytes) that would be * required for a log ticket. */ -int -xfs_log_calc_unit_res ( -struct xfs_mount *mp, +static int +xlog_calc_unit_res ( +struct xlog *log,int unit_bytes) { -struct xlog *log = mp->m_log;int iclog_space;uint num_headers;@@ -3487,18 +3476,20 @@ xfs_log_calc_unit_res(/* for commit-rec LR header - note: padding will subsume the ophdr */ unit_bytes += log->l_iclog_hsize; -/* for roundoff padding for transaction data and one for commit record */ -if (xfs_sb_version_haslogv2 (&mp->m_sb) && mp->m_sb.sb_logsunit >1 ) { -/* log su roundoff */ - unit_bytes +=2 * mp->m_sb.sb_logsunit; - }else { -/* BB roundoff */ - unit_bytes +=2 *BBSIZE ; - } +/* roundoff padding for transaction data and one for commit record */ + unit_bytes +=2 * log->l_iclog_roundoff;return unit_bytes; } +int +xfs_log_calc_unit_res ( +struct xfs_mount *mp, +int unit_bytes) +{ +return xlog_calc_unit_res (mp->m_log, unit_bytes); +} +/* * Allocate and initialise a new log ticket. */ @@ -3515,7 +3506,7 @@ xlog_ticket_alloc( tic =kmem_cache_zalloc (xfs_log_ticket_zone,GFP_NOFS | __GFP_NOFAIL); - unit_res =xfs_log_calc_unit_res (log->l_mp, unit_bytes); + unit_res =xlog_calc_unit_res (log, unit_bytes);atomic_set (&tic->t_ref,1 ); tic->t_task = current;fs/xfs/xfs_log_priv.h +2
-0@@ -436,6 +436,8 @@ struct xlog {#endif /* log recovery lsn tracking (for buffer submission */ xfs_lsn_t l_recovery_lsn; + +uint32_t l_iclog_roundoff;/* padding roundoff */ };#define XLOG_BUF_CANCEL_BUCKET (log, blkno) \
PATCH v4 2/39 xfs: separate CIL commit record IO
In reply to Dave ChinnerFrom: Dave Chinner <[email protected]>
To allow for iclog IO device cache flush behaviour to be optimised, we first need to separate out the commit record iclog IO from the rest of the checkpoint so we can wait for the checkpoint IO to complete before we issue the commit record.
This separation is only necessary if the commit record is being written into a different iclog to the start of the checkpoint as the upcoming cache flushing changes requires completion ordering against the other iclogs submitted by the checkpoint.
If the entire checkpoint and commit is in the one iclog, then they are both covered by the one set of cache flush primitives on the iclog and hence there is no need to separate them for ordering.
Otherwise, we need to wait for all the previous iclogs to complete so they are ordered correctly and made stable by the REQ_PREFLUSH that the commit record iclog IO issues. This guarantees that if a reader sees the commit record in the journal, they will also see the entire checkpoint that commit record closes off.
This also provides the guarantee that when the commit record IO completes, we can safely unpin all the log items in the checkpoint so they can be written back because the entire checkpoint is stable in the journal.
--- fs/xfs/xfs_log.c | 8 +++++--- fs/xfs/xfs_log_cil.c | 9 +++++++++ fs/xfs/xfs_log_priv.h | 2 ++ 3 files changed, 16 insertions(+), 3 deletions(-)
fs/xfs/xfs_log.c +5 -3
@@ -786,10 +786,12 @@ xfs_log_mount_cancel( }/* -* Wait for the iclog to be written disk, or return an error if the log has been -* shut down. +* Wait for the iclog and all prior iclogs to be written disk as required by the +* log force state machine. Waiting on ic_force_wait ensures iclog completions +* have been ordered and callbacks run before we are woken here, hence +* guaranteeing that all the iclogs up to this one are on stable storage. */ -static int +int xlog_wait_on_iclog( struct xlog_in_core *iclog) __releases(iclog->ic_log->l_icloglock)
fs/xfs/xfs_log_cil.c +9 -0
@@ -870,6 +870,15 @@ xlog_cil_push_work(wake_up_all (&cil->xc_commit_wait);spin_unlock (&cil->xc_push_lock); +/* +* If the checkpoint spans multiple iclogs, wait for all previous +* iclogs to complete before we submit the commit_iclog. +*/ +if (ctx->start_lsn != commit_lsn) { +spin_lock (&log->l_icloglock); +xlog_wait_on_iclog (commit_iclog->ic_prev); + } +/* release the hounds! */ xfs_log_release_iclog (commit_iclog);return ;
fs/xfs/xfs_log_priv.h +2 -0
@@ -584,6 +584,8 @@ xlog_wait(remove_wait_queue (wq, &wait); } +int xlog_wait_on_iclog (struct xlog_in_core *iclog); +/* * The LSN is valid so long as it is behind the current LSN. If it isn't, this * means that the next log record that includes this metadata could have a
Signature · 1 line
2.31.1
Re: [PATCH 02/39] xfs: separate CIL commit record IO
Quoted 27 lines · Dave Chinner
From: Dave Chinner <[email protected]>
To allow for iclog IO device cache flush behaviour to be optimised, we first need to separate out the commit record iclog IO from the rest of the checkpoint so we can wait for the checkpoint IO to complete before we issue the commit record.
This separation is only necessary if the commit record is being written into a different iclog to the start of the checkpoint as the upcoming cache flushing changes requires completion ordering against the other iclogs submitted by the checkpoint.
If the entire checkpoint and commit is in the one iclog, then they are both covered by the one set of cache flush primitives on the iclog and hence there is no need to separate them for ordering.
Otherwise, we need to wait for all the previous iclogs to complete so they are ordered correctly and made stable by the
REQ_PREFLUSHthat the commit record iclog IO issues. This guarantees that if a reader sees the commit record in the journal, they will also see the entire checkpoint that commit record closes off.This also provides the guarantee that when the commit record IO completes, we can safely unpin all the log items in the checkpoint so they can be written back because the entire checkpoint is stable in the journal.
Quoted 4 lines
Signed-off-by Dave Chinner <[email protected]>Reviewed-by Darrick J. Wong <[email protected]>Reviewed-by Chandan Babu R <[email protected]>Reviewed-by Brian Foster <[email protected]>
Ok, makes sense
Quoted 60 lines
--- fs/xfs/xfs_log.c | 8 +++++--- fs/xfs/xfs_log_cil.c | 9 +++++++++ fs/xfs/xfs_log_priv.h | 2 ++ 3 files changed, 16 insertions(+), 3 deletions(-)fs/xfs/xfs_log.c +5
-3@@ -786,10 +786,12 @@ xfs_log_mount_cancel( }/* -* Wait for the iclog to be written disk, or return an error if the log has been -* shut down. +* Wait for the iclog and all prior iclogs to be written disk as required by the +* log force state machine. Waiting on ic_force_wait ensures iclog completions +* have been ordered and callbacks run before we are woken here, hence +* guaranteeing that all the iclogs up to this one are on stable storage. */ -static int +int xlog_wait_on_iclog( struct xlog_in_core *iclog) __releases(iclog->ic_log->l_icloglock)fs/xfs/xfs_log_cil.c +9
-0@@ -870,6 +870,15 @@ xlog_cil_push_work(wake_up_all (&cil->xc_commit_wait);spin_unlock (&cil->xc_push_lock); +/* +* If the checkpoint spans multiple iclogs, wait for all previous +* iclogs to complete before we submit the commit_iclog. +*/ +if (ctx->start_lsn != commit_lsn) { +spin_lock (&log->l_icloglock); +xlog_wait_on_iclog (commit_iclog->ic_prev); + } +/* release the hounds! */ xfs_log_release_iclog (commit_iclog);return ;fs/xfs/xfs_log_priv.h +2
-0@@ -584,6 +584,8 @@ xlog_wait(remove_wait_queue (wq, &wait); } +int xlog_wait_on_iclog (struct xlog_in_core *iclog); +/* * The LSN is valid so long as it is behind the current LSN. If it isn't, this * means that the next log record that includes this metadata could have a
PATCH v4 3/39 xfs: remove xfs_blkdev_issue_flush
In reply to Dave ChinnerFrom: Dave Chinner <[email protected]>
It's a one line wrapper around blkdev_issue_flush(). Just replace it with direct calls to blkdev_issue_flush().
--- fs/xfs/xfs_buf.c | 2 +- fs/xfs/xfs_file.c | 6 +++--- fs/xfs/xfs_log.c | 2 +- fs/xfs/xfs_super.c | 7 ------- fs/xfs/xfs_super.h | 1 - 5 files changed, 5 insertions(+), 13 deletions(-)
fs/xfs/xfs_buf.c +1 -1
@@ -1945,7 +1945,7 @@ xfs_free_buftarg(percpu_counter_destroy (&btp->bt_io_count);list_lru_destroy (&btp->bt_lru); -xfs_blkdev_issue_flush (btp); +blkdev_issue_flush (btp->bt_bdev);kmem_free (btp); }
fs/xfs/xfs_file.c +3 -3
@@ -197,9 +197,9 @@ xfs_file_fsync(* inode size in case of an extending write. */ if (XFS_IS_REALTIME_INODE (ip)) -xfs_blkdev_issue_flush (mp->m_rtdev_targp); +blkdev_issue_flush (mp->m_rtdev_targp->bt_bdev);else if (mp->m_logdev_targp != mp->m_ddev_targp) -xfs_blkdev_issue_flush (mp->m_ddev_targp); +blkdev_issue_flush (mp->m_ddev_targp->bt_bdev);/* * Any inode that has dirty modifications in the log is pinned. The
@@ -219,7 +219,7 @@ xfs_file_fsync(*/ if (!log_flushed && !XFS_IS_REALTIME_INODE (ip) && mp->m_logdev_targp == mp->m_ddev_targp) -xfs_blkdev_issue_flush (mp->m_ddev_targp); +blkdev_issue_flush (mp->m_ddev_targp->bt_bdev);return error; }
fs/xfs/xfs_log.c +1 -1
@@ -1964,7 +1964,7 @@ xlog_sync(* layer state machine for preflushes. */ if (log->l_targ != log->l_mp->m_ddev_targp || split) { -xfs_blkdev_issue_flush (log->l_mp->m_ddev_targp); +blkdev_issue_flush (log->l_mp->m_ddev_targp->bt_bdev); need_flush = false; }
fs/xfs/xfs_super.c +0 -7
@@ -340,13 +340,6 @@ xfs_blkdev_put(blkdev_put (bdev,FMODE_READ |FMODE_WRITE |FMODE_EXCL ); } -void -xfs_blkdev_issue_flush ( -xfs_buftarg_t *buftarg) -{ -blkdev_issue_flush (buftarg->bt_bdev); -} - STATIC voidxfs_close_devices (struct xfs_mount *mp)
fs/xfs/xfs_super.h +0 -1
@@ -87,7 +87,6 @@ struct xfs_buftarg;struct block_device ;extern void xfs_flush_inodes (struct xfs_mount *mp); -extern void xfs_blkdev_issue_flush (struct xfs_buftarg *);extern xfs_agnumber_t xfs_set_inode_alloc (struct xfs_mount *,xfs_agnumber_t agcount);
Signature · 1 line
2.31.1
Re: [PATCH 03/39] xfs: remove xfs_blkdev_issue_flush
Quoted 10 lines · Dave Chinner
From: Dave Chinner <[email protected]>
It's a one line wrapper around
blkdev_issue_flush(). Just replace it with direct calls toblkdev_issue_flush().Signed-off-by Dave Chinner <[email protected]>Reviewed-by Chandan Babu R <[email protected]>Reviewed-by Darrick J. Wong <[email protected]>Reviewed-by Christoph Hellwig <[email protected]>Reviewed-by Brian Foster <[email protected]>
Looks fine
Quoted 90 lines
--- fs/xfs/xfs_buf.c | 2 +- fs/xfs/xfs_file.c | 6 +++--- fs/xfs/xfs_log.c | 2 +- fs/xfs/xfs_super.c | 7 ------- fs/xfs/xfs_super.h | 1 - 5 files changed, 5 insertions(+), 13 deletions(-)fs/xfs/xfs_buf.c +1
-1@@ -1945,7 +1945,7 @@ xfs_free_buftarg(percpu_counter_destroy (&btp->bt_io_count);list_lru_destroy (&btp->bt_lru); -xfs_blkdev_issue_flush (btp); +blkdev_issue_flush (btp->bt_bdev);kmem_free (btp); }fs/xfs/xfs_file.c +3
-3@@ -197,9 +197,9 @@ xfs_file_fsync(* inode size in case of an extending write. */ if (XFS_IS_REALTIME_INODE (ip)) -xfs_blkdev_issue_flush (mp->m_rtdev_targp); +blkdev_issue_flush (mp->m_rtdev_targp->bt_bdev);else if (mp->m_logdev_targp != mp->m_ddev_targp) -xfs_blkdev_issue_flush (mp->m_ddev_targp); +blkdev_issue_flush (mp->m_ddev_targp->bt_bdev);/* * Any inode that has dirty modifications in the log is pinned. The @@ -219,7 +219,7 @@ xfs_file_fsync(*/ if (!log_flushed && !XFS_IS_REALTIME_INODE (ip) && mp->m_logdev_targp == mp->m_ddev_targp) -xfs_blkdev_issue_flush (mp->m_ddev_targp); +blkdev_issue_flush (mp->m_ddev_targp->bt_bdev);return error; }fs/xfs/xfs_log.c +1
-1@@ -1964,7 +1964,7 @@ xlog_sync(* layer state machine for preflushes. */ if (log->l_targ != log->l_mp->m_ddev_targp || split) { -xfs_blkdev_issue_flush (log->l_mp->m_ddev_targp); +blkdev_issue_flush (log->l_mp->m_ddev_targp->bt_bdev); need_flush = false; }fs/xfs/xfs_super.c +0
-7@@ -340,13 +340,6 @@ xfs_blkdev_put(blkdev_put (bdev,FMODE_READ |FMODE_WRITE |FMODE_EXCL ); } -void -xfs_blkdev_issue_flush ( -xfs_buftarg_t *buftarg) -{ -blkdev_issue_flush (buftarg->bt_bdev); -} - STATIC voidxfs_close_devices (struct xfs_mount *mp)fs/xfs/xfs_super.h +0
-1@@ -87,7 +87,6 @@ struct xfs_buftarg;struct block_device ;extern void xfs_flush_inodes (struct xfs_mount *mp); -extern void xfs_blkdev_issue_flush (struct xfs_buftarg *);extern xfs_agnumber_t xfs_set_inode_alloc (struct xfs_mount *,xfs_agnumber_t agcount);
PATCH v4 4/39 xfs: async blkdev cache flush
In reply to Dave ChinnerFrom: Dave Chinner <[email protected]>
The new checkpoint cache flush mechanism requires us to issue an unconditional cache flush before we start a new checkpoint. We don't want to block for this if we can help it, and we have a fair chunk of CPU work to do between starting the checkpoint and issuing the first journal IO.
Hence it makes sense to amortise the latency cost of the cache flush by issuing it asynchronously and then waiting for it only when we need to issue the first IO in the transaction.
To do this, we need async cache flush primitives to submit the cache flush bio and to wait on it. The block layer has no such primitives for filesystems, so roll our own for the moment.
--- fs/xfs/xfs_bio_io.c | 35 +++++++++++++++++++++++++++++++++++ fs/xfs/xfs_linux.h | 2 ++ 2 files changed, 37 insertions(+)
fs/xfs/xfs_bio_io.c +35 -0
@@ -9,6 +9,41 @@ static inline unsigned int bio_max_vecs(unsigned int count)return bio_max_segs (howmany (count,PAGE_SIZE )); } +static void +xfs_flush_bdev_async_endio ( +struct bio *bio) +{ +complete (bio->bi_private); +} + +/* +* Submit a request for an async cache flush to run. If the request queue does +* not require flush operations, just skip it altogether. If the caller needsi +* to wait for the flush completion at a later point in time, they must supply a +* valid completion. This will be signalled when the flush completes. The +* caller never sees the bio that is issued here. +*/ +void +xfs_flush_bdev_async ( +struct bio *bio, +struct block_device *bdev, +struct completion *done) +{ +struct request_queue *q = bdev->bd_disk->queue; + +if (!test_bit (QUEUE_FLAG_WC , &q->queue_flags)) { +complete (done); +return ; + } + +bio_init (bio,NULL ,0 ); +bio_set_dev (bio, bdev); + bio->bi_opf =REQ_OP_WRITE |REQ_PREFLUSH |REQ_SYNC ; + bio->bi_private = done; + bio->bi_end_io = xfs_flush_bdev_async_endio; + +submit_bio (bio); +} int xfs_rw_bdev( struct block_device *bdev,
fs/xfs/xfs_linux.h +2 -0
@@ -196,6 +196,8 @@ static inline uint64_t howmany_64(uint64_t x, uint32_t y)int xfs_rw_bdev (struct block_device *bdev,sector_t sector,unsigned int count,char *data,unsigned int op); +void xfs_flush_bdev_async (struct bio *bio,struct block_device *bdev, +struct completion *done);#define ASSERT_ALWAYS (expr) \ (likely(expr) ? (void)0 : assfail(NULL, #expr, __FILE__, __LINE__))
Signature · 1 line
2.31.1
Re: [PATCH 04/39] xfs: async blkdev cache flush
Quoted 12 lines · Dave Chinner
From: Dave Chinner <[email protected]>
The new checkpoint cache flush mechanism requires us to issue an unconditional cache flush before we start a new checkpoint. We don't want to block for this if we can help it, and we have a fair chunk of CPU work to do between starting the checkpoint and issuing the first journal IO.
Hence it makes sense to amortise the latency cost of the cache flush by issuing it asynchronously and then waiting for it only when we need to issue the first IO in the transaction.
Quoted 6 lines
To do this, we need async cache flush primitives to submit the cache flush bio and to wait on it. The block layer has no such primitives for filesystems, so roll our own for the moment.
Signed-off-by Dave Chinner <[email protected]>Reviewed-by Brian Foster <[email protected]>
Looks good to me,
--D
Quoted 67 lines
--- fs/xfs/xfs_bio_io.c | 35 +++++++++++++++++++++++++++++++++++ fs/xfs/xfs_linux.h | 2 ++ 2 files changed, 37 insertions(+)fs/xfs/xfs_bio_io.c +35
-0@@ -9,6 +9,41 @@ static inline unsigned int bio_max_vecs(unsigned int count)return bio_max_segs (howmany (count,PAGE_SIZE )); } +static void +xfs_flush_bdev_async_endio ( +struct bio *bio) +{ +complete (bio->bi_private); +} + +/* +* Submit a request for an async cache flush to run. If the request queue does +* not require flush operations, just skip it altogether. If the caller needsi +* to wait for the flush completion at a later point in time, they must supply a +* valid completion. This will be signalled when the flush completes. The +* caller never sees the bio that is issued here. +*/ +void +xfs_flush_bdev_async ( +struct bio *bio, +struct block_device *bdev, +struct completion *done) +{ +struct request_queue *q = bdev->bd_disk->queue; + +if (!test_bit (QUEUE_FLAG_WC , &q->queue_flags)) { +complete (done); +return ; + } + +bio_init (bio,NULL ,0 ); +bio_set_dev (bio, bdev); + bio->bi_opf =REQ_OP_WRITE |REQ_PREFLUSH |REQ_SYNC ; + bio->bi_private = done; + bio->bi_end_io = xfs_flush_bdev_async_endio; + +submit_bio (bio); +} int xfs_rw_bdev( struct block_device *bdev,fs/xfs/xfs_linux.h +2
-0@@ -196,6 +196,8 @@ static inline uint64_t howmany_64(uint64_t x, uint32_t y)int xfs_rw_bdev (struct block_device *bdev,sector_t sector,unsigned int count,char *data,unsigned int op); +void xfs_flush_bdev_async (struct bio *bio,struct block_device *bdev, +struct completion *done);#define ASSERT_ALWAYS (expr) \ (likely(expr) ? (void)0 : assfail(NULL, #expr, __FILE__, __LINE__))Signature · 1 line
2.31.1
Re: [PATCH 04/39] xfs: async blkdev cache flush
In reply to Dave ChinnerQuoted 31 lines · Dave Chinner
From: Dave Chinner <[email protected]>
The new checkpoint cache flush mechanism requires us to issue an unconditional cache flush before we start a new checkpoint. We don't want to block for this if we can help it, and we have a fair chunk of CPU work to do between starting the checkpoint and issuing the first journal IO.
Hence it makes sense to amortise the latency cost of the cache flush by issuing it asynchronously and then waiting for it only when we need to issue the first IO in the transaction.
To do this, we need async cache flush primitives to submit the cache flush bio and to wait on it. The block layer has no such primitives for filesystems, so roll our own for the moment.
Signed-off-by Dave Chinner <[email protected]>Reviewed-by Brian Foster <[email protected]>--- fs/xfs/xfs_bio_io.c | 35 +++++++++++++++++++++++++++++++++++ fs/xfs/xfs_linux.h | 2 ++ 2 files changed, 37 insertions(+)fs/xfs/xfs_bio_io.c +0
-0@@ -9,6 +9,41 @@ static inline unsigned int bio_max_vecs(unsigned int count)return bio_max_segs (howmany (count,PAGE_SIZE )); }
fs/xfs/xfs_bio_io.c static inline unsigned int bio_max_vecs(unsigned int count) · line 21
Quoted 10 lines
fs/xfs/xfs_bio_io.c +10
-0+static void +xfs_flush_bdev_async_endio ( +struct bio *bio) +{ +complete (bio->bi_private); +} + +/* +* Submit a request for an async cache flush to run. If the request queue does +* not require flush operations, just skip it altogether. If the caller needsi
typo nit: needs
Otherwise looks fine
Quoted 42 lines
fs/xfs/xfs_bio_io.c +25
-0+* to wait for the flush completion at a later point in time, they must supply a +* valid completion. This will be signalled when the flush completes. The +* caller never sees the bio that is issued here. +*/ +void +xfs_flush_bdev_async ( +struct bio *bio, +struct block_device *bdev, +struct completion *done) +{ +struct request_queue *q = bdev->bd_disk->queue; + +if (!test_bit (QUEUE_FLAG_WC , &q->queue_flags)) { +complete (done); +return ; + } + +bio_init (bio,NULL ,0 ); +bio_set_dev (bio, bdev); + bio->bi_opf =REQ_OP_WRITE |REQ_PREFLUSH |REQ_SYNC ; + bio->bi_private = done; + bio->bi_end_io = xfs_flush_bdev_async_endio; + +submit_bio (bio); +} int xfs_rw_bdev( struct block_device *bdev,fs/xfs/xfs_linux.h +2
-0@@ -196,6 +196,8 @@ static inline uint64_t howmany_64(uint64_t x, uint32_t y)int xfs_rw_bdev (struct block_device *bdev,sector_t sector,unsigned int count,char *data,unsigned int op); +void xfs_flush_bdev_async (struct bio *bio,struct block_device *bdev, +struct completion *done);#define ASSERT_ALWAYS (expr) \ (likely(expr) ? (void)0 : assfail(NULL, #expr, __FILE__, __LINE__))
PATCH v4 5/39 xfs: CIL checkpoint flushes caches unconditionally
In reply to Dave ChinnerFrom: Dave Chinner <[email protected]>
Currently every journal IO is issued as REQ_PREFLUSH | REQ_FUA to guarantee the ordering requirements the journal has w.r.t. metadata writeback. THe two ordering constraints are:
1. we cannot overwrite metadata in the journal until we guarantee that the dirty metadata has been written back in place and is stable.
2. we cannot write back dirty metadata until it has been written to the journal and guaranteed to be stable (and hence recoverable) in the journal.
These rules apply to the atomic transactions recorded in the journal, not to the journal IO itself. Hence we need to ensure metadata is stable before we start writing a new transaction to the journal (guarantee #1), and we need to ensure the entire transaction is stable in the journal before we start metadata writeback (guarantee #2).
The ordering guarantees of #1 are currently provided by REQ_PREFLUSH being added to every iclog IO. This causes the journal IO to issue a cache flush and wait for it to complete before issuing the write IO to the journal. Hence all completed metadata IO is guaranteed to be stable before the journal overwrites the old metadata.
However, for long running CIL checkpoints that might do a thousand journal IOs, we don't need every single one of these iclog IOs to issue a cache flush - the cache flush done before the first iclog is submitted is sufficient to cover the entire range in the log that the checkpoint will overwrite because the CIL space reservation guarantees the tail of the log (completed metadata) is already beyond the range of the checkpoint write.
Hence we only need a full cache flush between closing off the CIL checkpoint context (i.e. when the push switches it out) and issuing the first journal IO. Rather than plumbing this through to the journal IO, we can start this cache flush the moment the CIL context is owned exclusively by the push worker. The cache flush can be in progress while we process the CIL ready for writing, hence reducing the latency of the initial iclog write. This is especially true for large checkpoints, where we might have to process hundreds of thousands of log vectors before we issue the first iclog write. In these cases, it is likely the cache flush has already been completed by the time we have built the CIL log vector chain.
--- fs/xfs/xfs_log_cil.c | 25 +++++++++++++++++++++---- 1 file changed, 21 insertions(+), 4 deletions(-)
fs/xfs/xfs_log_cil.c +21 -4
@@ -656,6 +656,8 @@ xlog_cil_push_work(struct xfs_log_vec lvhdr = {NULL };xfs_lsn_t commit_lsn;xfs_lsn_t push_seq; +struct bio bio; +DECLARE_COMPLETION_ONSTACK (bdev_flush); new_ctx =kmem_zalloc (sizeof (*new_ctx),KM_NOFS ); new_ctx->ticket =xlog_cil_ticket_alloc (log);
@@ -719,10 +721,19 @@ xlog_cil_push_work(spin_unlock (&cil->xc_push_lock);/* -* pull all the log vectors off the items in the CIL, and -* remove the items from the CIL. We don't need the CIL lock -* here because it's only needed on the transaction commit -* side which is currently locked out by the flush lock. +* The CIL is stable at this point - nothing new will be added to it +* because we hold the flush lock exclusively. Hence we can now issue +* a cache flush to ensure all the completed metadata in the journal we +* are about to overwrite is on stable storage. +*/ +xfs_flush_bdev_async (&bio, log->l_mp->m_ddev_targp->bt_bdev, + &bdev_flush); + +/* +* Pull all the log vectors off the items in the CIL, and remove the +* items from the CIL. We don't need the CIL lock here because it's only +* needed on the transaction commit side which is currently locked out +* by the flush lock. */ lv =NULL ; num_iovecs =0 ;
@@ -806,6 +817,12 @@ xlog_cil_push_work( lvhdr.lv_iovecp = &lhdr; lvhdr.lv_next = ctx->lv_chain; +/* +* Before we format and submit the first iclog, we have to ensure that +* the metadata writeback ordering cache flush is complete. +*/ +wait_for_completion (&bdev_flush); + error =xlog_write (log, &lvhdr, tic, &ctx->start_lsn,NULL ,0 , true);if (error) goto out_abort_free_ticket;
Signature · 1 line
2.31.1
Re: [PATCH 05/39] xfs: CIL checkpoint flushes caches unconditionally
Quoted 47 lines · Dave Chinner
From: Dave Chinner <[email protected]>
Currently every journal IO is issued as
REQ_PREFLUSH|REQ_FUAto guarantee the ordering requirements the journal has w.r.t. metadata writeback. THe two ordering constraints are:1. we cannot overwrite metadata in the journal until we guarantee that the dirty metadata has been written back in place and is stable.
2. we cannot write back dirty metadata until it has been written to the journal and guaranteed to be stable (and hence recoverable) in the journal.
These rules apply to the atomic transactions recorded in the journal, not to the journal IO itself. Hence we need to ensure metadata is stable before we start writing a new transaction to the journal (guarantee #1), and we need to ensure the entire transaction is stable in the journal before we start metadata writeback (guarantee #2).
The ordering guarantees of #1 are currently provided by
REQ_PREFLUSHbeing added to every iclog IO. This causes the journal IO to issue a cache flush and wait for it to complete before issuing the write IO to the journal. Hence all completed metadata IO is guaranteed to be stable before the journal overwrites the old metadata.However, for long running CIL checkpoints that might do a thousand journal IOs, we don't need every single one of these iclog IOs to issue a cache flush - the cache flush done before the first iclog is submitted is sufficient to cover the entire range in the log that the checkpoint will overwrite because the CIL space reservation guarantees the tail of the log (completed metadata) is already beyond the range of the checkpoint write.
Hence we only need a full cache flush between closing off the CIL checkpoint context (i.e. when the push switches it out) and issuing the first journal IO. Rather than plumbing this through to the journal IO, we can start this cache flush the moment the CIL context is owned exclusively by the push worker. The cache flush can be in progress while we process the CIL ready for writing, hence reducing the latency of the initial iclog write. This is especially true for large checkpoints, where we might have to process hundreds of thousands of log vectors before we issue the first iclog write. In these cases, it is likely the cache flush has already been completed by the time we have built the CIL log vector chain.
Quoted 4 lines
Signed-off-by Dave Chinner <[email protected]>Reviewed-by Chandan Babu R <[email protected]>Reviewed-by Darrick J. Wong <[email protected]>Reviewed-by Brian Foster <[email protected]>
Ok, makes sense
Quoted 55 lines
--- fs/xfs/xfs_log_cil.c | 25 +++++++++++++++++++++---- 1 file changed, 21 insertions(+), 4 deletions(-)fs/xfs/xfs_log_cil.c +21
-4@@ -656,6 +656,8 @@ xlog_cil_push_work(struct xfs_log_vec lvhdr = {NULL };xfs_lsn_t commit_lsn;xfs_lsn_t push_seq; +struct bio bio; +DECLARE_COMPLETION_ONSTACK (bdev_flush); new_ctx =kmem_zalloc (sizeof (*new_ctx),KM_NOFS ); new_ctx->ticket =xlog_cil_ticket_alloc (log);@@ -719,10 +721,19 @@ xlog_cil_push_work(spin_unlock (&cil->xc_push_lock);/* -* pull all the log vectors off the items in the CIL, and -* remove the items from the CIL. We don't need the CIL lock -* here because it's only needed on the transaction commit -* side which is currently locked out by the flush lock. +* The CIL is stable at this point - nothing new will be added to it +* because we hold the flush lock exclusively. Hence we can now issue +* a cache flush to ensure all the completed metadata in the journal we +* are about to overwrite is on stable storage. +*/ +xfs_flush_bdev_async (&bio, log->l_mp->m_ddev_targp->bt_bdev, + &bdev_flush); + +/* +* Pull all the log vectors off the items in the CIL, and remove the +* items from the CIL. We don't need the CIL lock here because it's only +* needed on the transaction commit side which is currently locked out +* by the flush lock. */ lv =NULL ; num_iovecs =0 ;@@ -806,6 +817,12 @@ xlog_cil_push_work( lvhdr.lv_iovecp = &lhdr; lvhdr.lv_next = ctx->lv_chain; +/* +* Before we format and submit the first iclog, we have to ensure that +* the metadata writeback ordering cache flush is complete. +*/ +wait_for_completion (&bdev_flush); + error =xlog_write (log, &lvhdr, tic, &ctx->start_lsn,NULL ,0 , true);if (error) goto out_abort_free_ticket;
PATCH v4 6/39 xfs: remove need_start_rec parameter from xlog_write()
In reply to Dave ChinnerFrom: Dave Chinner <[email protected]>
The CIL push is the only call to xlog_write that sets this variable to true. The other callers don't need a start rec, and they tell xlog_write what to do by passing the type of ophdr they need written in the flags field. The need_start_rec parameter essentially tells xlog_write to to write an extra ophdr with a XLOG_START_TRANS type, so get rid of the variable to do this and pass XLOG_START_TRANS as the flag value into xlog_write() from the CIL push.
$ size fs/xfs/xfs_log.o* text data bss dec hex filename 27595 560 8 28163 6e03 fs/xfs/xfs_log.o.orig 27454 560 8 28022 6d76 fs/xfs/xfs_log.o.patched
--- fs/xfs/xfs_log.c | 44 +++++++++++++++++++++---------------------- fs/xfs/xfs_log_cil.c | 3 ++- fs/xfs/xfs_log_priv.h | 3 +-- 3 files changed, 25 insertions(+), 25 deletions(-)
fs/xfs/xfs_log.c +22 -22
@@ -820,9 +820,7 @@ xlog_wait_on_iclog(static int xlog_write_unmount_record (struct xlog *log, -struct xlog_ticket *ticket, -xfs_lsn_t *lsn, -uint flags) +struct xlog_ticket *ticket) {struct xfs_unmount_log_format ulf = { .magic =XLOG_UNMOUNT_TYPE ,
@@ -839,7 +837,7 @@ xlog_write_unmount_record(/* account for space used by record data */ ticket->t_curr_res -=sizeof (ulf); -return xlog_write (log, &vec, ticket, lsn,NULL , flags, false); +return xlog_write (log, &vec, ticket,NULL ,NULL ,XLOG_UNMOUNT_TRANS ); }/*
@@ -853,15 +851,13 @@ xlog_unmount_write(struct xfs_mount *mp = log->l_mp;struct xlog_in_core *iclog;struct xlog_ticket *tic =NULL ; -xfs_lsn_t lsn; -uint flags =XLOG_UNMOUNT_TRANS ;int error; error =xfs_log_reserve (mp,600 ,1 , &tic,XFS_LOG ,0 );if (error) goto out_err; - error =xlog_write_unmount_record (log, tic, &lsn, flags); + error =xlog_write_unmount_record (log, tic);/* * At this point, we're umounting anyway, so there's no point in * transitioning log state to IOERROR. Just continue...
@@ -1553,8 +1549,7 @@ xlog_commit_record(if (XLOG_FORCED_SHUTDOWN (log))return -EIO ; - error =xlog_write (log, &vec, ticket, lsn, iclog,XLOG_COMMIT_TRANS , - false); + error =xlog_write (log, &vec, ticket, lsn, iclog,XLOG_COMMIT_TRANS );if (error)xfs_force_shutdown (log->l_mp,SHUTDOWN_LOG_IO_ERROR );return error;
@@ -2151,13 +2146,16 @@ static intxlog_write_calc_vec_length ( struct xlog_ticket *ticket, struct xfs_log_vec *log_vector, - bool need_start_rec) + uint optype) { struct xfs_log_vec *lv; -int headers = need_start_rec ?1 :0 ; +int headers =0 ;int len =0 ;int i; +if (optype &XLOG_START_TRANS ) + headers++; +for (lv = log_vector; lv; lv = lv->lv_next) {/* we don't write ordered log vectors */ if (lv->lv_buf_len ==XFS_LOG_VEC_ORDERED )
@@ -2377,8 +2375,7 @@ xlog_write(struct xlog_ticket *ticket, xfs_lsn_t *start_lsn, struct xlog_in_core **commit_iclog, - uint flags, - bool need_start_rec) + uint optype) { struct xlog_in_core *iclog = NULL;struct xfs_log_vec *lv = log_vector;
@@ -2406,8 +2403,9 @@ xlog_write(xfs_force_shutdown (log->l_mp,SHUTDOWN_LOG_IO_ERROR ); } - len =xlog_write_calc_vec_length (ticket, log_vector, need_start_rec); - *start_lsn =0 ; + len =xlog_write_calc_vec_length (ticket, log_vector, optype); +if (start_lsn) + *start_lsn =0 ;while (lv && (!lv->lv_niovecs || index < lv->lv_niovecs)) {void *ptr;int log_offset;
@@ -2421,7 +2419,7 @@ xlog_write( ptr = iclog->ic_datap + log_offset;/* start_lsn is the first lsn written to. That's all we need. */ -if (!*start_lsn) +if (start_lsn && !*start_lsn) *start_lsn =be64_to_cpu (iclog->ic_header.h_lsn);/*
@@ -2434,6 +2432,7 @@ xlog_write(int copy_len;int copy_off;bool ordered = false; +bool wrote_start_rec = false;/* ordered log vectors have no regions to write */ if (lv->lv_buf_len ==XFS_LOG_VEC_ORDERED ) {
@@ -2451,13 +2450,15 @@ xlog_write(* write a start record. Only do this for the first * iclog we write to. */ -if (need_start_rec) { +if (optype &XLOG_START_TRANS ) {xlog_write_start_rec (ptr, ticket);xlog_write_adv_cnt (&ptr, &len, &log_offset,sizeof (struct xlog_op_header )); + optype &= ~XLOG_START_TRANS ; + wrote_start_rec = true; } - ophdr =xlog_write_setup_ophdr (log, ptr, ticket, flags); + ophdr =xlog_write_setup_ophdr (log, ptr, ticket, optype);if (!ophdr)return -EIO ;
@@ -2488,14 +2489,13 @@ xlog_write( } copy_len +=sizeof (struct xlog_op_header ); record_cnt++; -if (need_start_rec) { +if (wrote_start_rec) { copy_len +=sizeof (struct xlog_op_header ); record_cnt++; - need_start_rec = false; } data_cnt += contwr ? copy_len :0 ; - error = xlog_write_copy_finish(log, iclog, flags, + error = xlog_write_copy_finish(log, iclog, optype, &record_cnt, &data_cnt, &partial_copy, &partial_copy_len,
@@ -2539,7 +2539,7 @@ xlog_write(spin_lock (&log->l_icloglock);xlog_state_finish_copy (log, iclog, record_cnt, data_cnt);if (commit_iclog) { -ASSERT (flags &XLOG_COMMIT_TRANS ); +ASSERT (optype &XLOG_COMMIT_TRANS ); *commit_iclog = iclog; }else { error =xlog_state_release_iclog (log, iclog);
fs/xfs/xfs_log_cil.c +2 -1
@@ -823,7 +823,8 @@ xlog_cil_push_work(*/ wait_for_completion (&bdev_flush); - error =xlog_write (log, &lvhdr, tic, &ctx->start_lsn,NULL ,0 , true); + error =xlog_write (log, &lvhdr, tic, &ctx->start_lsn,NULL , +XLOG_START_TRANS );if (error) goto out_abort_free_ticket;
fs/xfs/xfs_log_priv.h +1 -2
@@ -480,8 +480,7 @@ void xlog_print_tic_res(struct xfs_mount *mp, struct xlog_ticket *ticket);void xlog_print_trans (struct xfs_trans *);int xlog_write (struct xlog *log,struct xfs_log_vec *log_vector,struct xlog_ticket *tic,xfs_lsn_t *start_lsn, -struct xlog_in_core **commit_iclog,uint flags, -bool need_start_rec); +struct xlog_in_core **commit_iclog,uint optype);int xlog_commit_record (struct xlog *log,struct xlog_ticket *ticket,struct xlog_in_core **iclog,xfs_lsn_t *lsn);void xfs_log_ticket_ungrant (struct xlog *log,struct xlog_ticket *ticket);
Signature · 1 line
2.31.1
PATCH v4 7/39 xfs: journal IO cache flush reductions
In reply to Dave ChinnerFrom: Dave Chinner <[email protected]>
Currently every journal IO is issued as REQ_PREFLUSH | REQ_FUA to guarantee the ordering requirements the journal has w.r.t. metadata writeback. THe two ordering constraints are:
1. we cannot overwrite metadata in the journal until we guarantee that the dirty metadata has been written back in place and is stable.
2. we cannot write back dirty metadata until it has been written to the journal and guaranteed to be stable (and hence recoverable) in the journal.
The ordering guarantees of #1 are provided by REQ_PREFLUSH. This causes the journal IO to issue a cache flush and wait for it to complete before issuing the write IO to the journal. Hence all completed metadata IO is guaranteed to be stable before the journal overwrites the old metadata.
The ordering guarantees of #2 are provided by the REQ_FUA, which ensures the journal writes do not complete until they are on stable storage. Hence by the time the last journal IO in a checkpoint completes, we know that the entire checkpoint is on stable storage and we can unpin the dirty metadata and allow it to be written back.
This is the mechanism by which ordering was first implemented in XFS way back in 2002 by commit 95d97c36e5155075ba2eb22b17562cfcc53fcf96 ("Add support for drive write cache flushing") in the xfs-archive tree.
A lot has changed since then, most notably we now use delayed logging to checkpoint the filesystem to the journal rather than write each individual transaction to the journal. Cache flushes on journal IO are necessary when individual transactions are wholly contained within a single iclog. However, CIL checkpoints are single transactions that typically span hundreds to thousands of individual journal writes, and so the requirements for device cache flushing have changed.
That is, the ordering rules I state above apply to ordering of atomic transactions recorded in the journal, not to the journal IO itself. Hence we need to ensure metadata is stable before we start writing a new transaction to the journal (guarantee #1), and we need to ensure the entire transaction is stable in the journal before we start metadata writeback (guarantee #2).
Hence we only need a REQ_PREFLUSH on the journal IO that starts a new journal transaction to provide #1, and it is not on any other journal IO done within the context of that journal transaction.
The CIL checkpoint already issues a cache flush before it starts writing to the log, so we no longer need the iclog IO to issue a REQ_REFLUSH for us. Hence if XLOG_START_TRANS is passed to xlog_write(), we no longer need to mark the first iclog in the log write with REQ_PREFLUSH for this case. As an added bonus, this ordering mechanism works for both internal and external logs, meaning we can remove the explicit data device cache flushes from the iclog write code when using external logs.
Given the new ordering semantics of commit records for the CIL, we need iclogs containing commit records to issue a REQ_PREFLUSH. We also require unmount records to do this. Hence for both XLOG_COMMIT_TRANS and XLOG_UNMOUNT_TRANS xlog_write() calls we need to mark the first iclog being written with REQ_PREFLUSH.
For both commit records and unmount records, we also want them immediately on stable storage, so we want to also mark the iclogs that contain these records to be marked REQ_FUA. That means if a record is split across multiple iclogs, they are all marked REQ_FUA and not just the last one so that when the transaction is completed all the parts of the record are on stable storage.
And for external logs, unmount records need a pre-write data device cache flush similar to the CIL checkpoint cache pre-flush as the internal iclog write code does not do this implicitly anymore.
As an optimisation, when the commit record lands in the same iclog as the journal transaction starts, we don't need to wait for anything and can simply use REQ_FUA to provide guarantee #2. This means that for fsync() heavy workloads, the cache flush behaviour is completely unchanged and there is no degradation in performance as a result of optimise the multi-IO transaction case.
The most notable sign that there is less IO latency on my test machine (nvme SSDs) is that the "noiclogs" rate has dropped substantially. This metric indicates that the CIL push is blocking in xlog_get_iclog_space() waiting for iclog IO completion to occur. With 8 iclogs of 256kB, the rate is appoximately 1 noiclog event to every 4 iclog writes. IOWs, every 4th call to xlog_get_iclog_space() is blocking waiting for log IO. With the changes in this patch, this drops to 1 noiclog event for every 100 iclog writes. Hence it is clear that log IO is completing much faster than it was previously, but it is also clear that for large iclog sizes, this isn't the performance limiting factor on this hardware.
With smaller iclogs (32kB), however, there is a sustantial difference. With the cache flush modifications, the journal is now running at over 4000 write IOPS, and the journal throughput is largely identical to the 256kB iclogs and the noiclog event rate stays low at about 1:50 iclog writes. The existing code tops out at about 2500 IOPS as the number of cache flushes dominate performance and latency. The noiclog event rate is about 1:4, and the performance variance is quite large as the journal throughput can fall to less than half the peak sustained rate when the cache flush rate prevents metadata writeback from keeping up and the log runs out of space and throttles reservations.
As a result:
logbsize fsmark create rate rm -rf
before 32kb 152851+/-5.3e+04 5m28s
patched 32kb 221533+/-1.1e+04 5m24s
before 256kb 220239+/-6.2e+03 4m58s
patched 256kb 228286+/-9.2e+03 5m06s
The rm -rf times are included because I ran them, but the differences are largely noise. This workload is largely metadata read IO latency bound and the changes to the journal cache flushing doesn't really make any noticable difference to behaviour apart from a reduction in noiclog events from background CIL pushing.
--- fs/xfs/xfs_log.c | 66 +++++++++++++++---------------------------- fs/xfs/xfs_log.h | 1 - fs/xfs/xfs_log_cil.c | 18 +++++++++--- fs/xfs/xfs_log_priv.h | 6 ++++ 4 files changed, 43 insertions(+), 48 deletions(-)
fs/xfs/xfs_log.c +23 -43
@@ -513,7 +513,7 @@ __xlog_state_release_iclog(* Flush iclog to disk if this is the last reference to the given iclog and the * it is in the WANT_SYNC state. */ -static int +int xlog_state_release_iclog (struct xlog *log,struct xlog_in_core *iclog)
@@ -533,23 +533,6 @@ xlog_state_release_iclog(return 0 ; } -void -xfs_log_release_iclog ( -struct xlog_in_core *iclog) -{ -struct xlog *log = iclog->ic_log; -bool sync = false; - -if (atomic_dec_and_lock (&iclog->ic_refcnt, &log->l_icloglock)) { -if (iclog->ic_state !=XLOG_STATE_IOERROR ) - sync =__xlog_state_release_iclog (log, iclog); -spin_unlock (&log->l_icloglock); - } - -if (sync) -xlog_sync (log, iclog); -} -/* * Mount a log filesystem *
@@ -837,6 +820,14 @@ xlog_write_unmount_record(/* account for space used by record data */ ticket->t_curr_res -=sizeof (ulf); + +/* +* For external log devices, we need to flush the data device cache +* first to ensure all metadata writeback is on stable storage before we +* stamp the tail LSN into the unmount record. +*/ +if (log->l_targ != log->l_mp->m_ddev_targp) +blkdev_issue_flush (log->l_targ->bt_bdev);return xlog_write (log, &vec, ticket,NULL ,NULL ,XLOG_UNMOUNT_TRANS ); }
@@ -874,6 +865,11 @@ xlog_unmount_write(else ASSERT (iclog->ic_state ==XLOG_STATE_WANT_SYNC || iclog->ic_state ==XLOG_STATE_IOERROR ); +/* +* Ensure the journal is fully flushed and on stable storage once the +* iclog containing the unmount record is written. +*/ + iclog->ic_flags |= (XLOG_ICL_NEED_FLUSH |XLOG_ICL_NEED_FUA ); error =xlog_state_release_iclog (log, iclog);xlog_wait_on_iclog (iclog);
@@ -1755,8 +1751,7 @@ xlog_write_iclog(struct xlog *log, struct xlog_in_core *iclog, uint64_t bno, - unsigned int count, - bool need_flush) + unsigned int count) { ASSERT(bno < log->l_logBBsize);
@@ -1794,10 +1789,12 @@ xlog_write_iclog(* writeback throttle from throttling log writes behind background * metadata writeback and causing priority inversions. */ - iclog->ic_bio.bi_opf =REQ_OP_WRITE |REQ_META |REQ_SYNC | -REQ_IDLE |REQ_FUA ; -if (need_flush) + iclog->ic_bio.bi_opf =REQ_OP_WRITE |REQ_META |REQ_SYNC |REQ_IDLE ; +if (iclog->ic_flags &XLOG_ICL_NEED_FLUSH ) iclog->ic_bio.bi_opf |=REQ_PREFLUSH ; +if (iclog->ic_flags &XLOG_ICL_NEED_FUA ) + iclog->ic_bio.bi_opf |=REQ_FUA ; + iclog->ic_flags &= ~(XLOG_ICL_NEED_FLUSH |XLOG_ICL_NEED_FUA );if (xlog_map_iclog_data (&iclog->ic_bio, iclog->ic_data, count)) {xfs_force_shutdown (log->l_mp,SHUTDOWN_LOG_IO_ERROR );
@@ -1900,7 +1897,6 @@ xlog_sync(unsigned int roundoff;/* roundoff to BB or stripe */ uint64_t bno;unsigned int size; -bool need_flush = true, split = false;ASSERT (atomic_read (&iclog->ic_refcnt) ==0 );
@@ -1925,10 +1921,8 @@ xlog_sync( bno =BLOCK_LSN (be64_to_cpu (iclog->ic_header.h_lsn));/* Do we need to split this write into 2 parts? */ -if (bno +BTOBB (count) > log->l_logBBsize) { +if (bno +BTOBB (count) > log->l_logBBsize)xlog_split_iclog (log, &iclog->ic_header, bno, count); - split = true; - }/* calculcate the checksum */ iclog->ic_header.h_crc = xlog_cksum(log, &iclog->ic_header,
@@ -1949,22 +1943,8 @@ xlog_sync( be64_to_cpu(iclog->ic_header.h_lsn)); }#endif - -/* -* Flush the data device before flushing the log to make sure all meta -* data written back from the AIL actually made it to disk before -* stamping the new log tail LSN into the log buffer. For an external -* log we need to issue the flush explicitly, and unfortunately -* synchronously here; for an internal log we can simply use the block -* layer state machine for preflushes. -*/ -if (log->l_targ != log->l_mp->m_ddev_targp || split) { -blkdev_issue_flush (log->l_mp->m_ddev_targp->bt_bdev); - need_flush = false; - } -xlog_verify_iclog (log, iclog, count); -xlog_write_iclog (log, iclog, bno, count, need_flush); +xlog_write_iclog (log, iclog, bno, count); }/*
@@ -2418,7 +2398,7 @@ xlog_write(ASSERT (log_offset <= iclog->ic_size -1 ); ptr = iclog->ic_datap + log_offset; -/* start_lsn is the first lsn written to. That's all we need. */ +/* Start_lsn is the first lsn written to. */ if (start_lsn && !*start_lsn) *start_lsn =be64_to_cpu (iclog->ic_header.h_lsn);
fs/xfs/xfs_log.h +0 -1
@@ -117,7 +117,6 @@ void xfs_log_mount_cancel(struct xfs_mount *);xfs_lsn_t xlog_assign_tail_lsn (struct xfs_mount *mp);xfs_lsn_t xlog_assign_tail_lsn_locked (struct xfs_mount *mp);void xfs_log_space_wake (struct xfs_mount *mp); -void xfs_log_release_iclog (struct xlog_in_core *iclog); int xfs_log_reserve(struct xfs_mount *mp, int length, int count,
fs/xfs/xfs_log_cil.c +14 -4
@@ -890,15 +890,25 @@ xlog_cil_push_work(/* * If the checkpoint spans multiple iclogs, wait for all previous -* iclogs to complete before we submit the commit_iclog. +* iclogs to complete before we submit the commit_iclog. In this case, +* the commit_iclog write needs to issue a pre-flush so that the +* ordering is correctly preserved down to stable storage. */ +spin_lock (&log->l_icloglock);if (ctx->start_lsn != commit_lsn) { -spin_lock (&log->l_icloglock);xlog_wait_on_iclog (commit_iclog->ic_prev); +spin_lock (&log->l_icloglock); + commit_iclog->ic_flags |=XLOG_ICL_NEED_FLUSH ; } -/* release the hounds! */ -xfs_log_release_iclog (commit_iclog); +/* +* The commit iclog must be written to stable storage to guarantee +* journal IO vs metadata writeback IO is correctly ordered on stable +* storage. +*/ + commit_iclog->ic_flags |=XLOG_ICL_NEED_FUA ; +xlog_state_release_iclog (log, commit_iclog); +spin_unlock (&log->l_icloglock);return ; out_skip:
fs/xfs/xfs_log_priv.h +6 -0
@@ -133,6 +133,9 @@ enum xlog_iclog_state {#define XLOG_COVER_OPS 5 +#define XLOG_ICL_NEED_FLUSH (1 << 0)/* iclog needs REQ_PREFLUSH */ +#define XLOG_ICL_NEED_FUA (1 << 1)/* iclog needs REQ_FUA */ +/* Ticket reservation region accounting */ #define XLOG_TIC_LEN_MAX 15
@@ -201,6 +204,7 @@ typedef struct xlog_in_core {u32 ic_size;u32 ic_offset;enum xlog_iclog_state ic_state; +unsigned int ic_flags;char *ic_datap;/* pointer to iclog data */ /* Callback structures need their own cacheline */
@@ -486,6 +490,8 @@ int xlog_commit_record(struct xlog *log, struct xlog_ticket *ticket,void xfs_log_ticket_ungrant (struct xlog *log,struct xlog_ticket *ticket);void xfs_log_ticket_regrant (struct xlog *log,struct xlog_ticket *ticket); +int xlog_state_release_iclog (struct xlog *log,struct xlog_in_core *iclog); +/* * When we crack an atomic LSN, we sample it first so that the value will not * change while we are cracking it into the component values. This means we
Signature · 1 line
2.31.1
Re: [PATCH 07/39] xfs: journal IO cache flush reductions
Quoted 96 lines · Dave Chinner
From: Dave Chinner <[email protected]>
Currently every journal IO is issued as
REQ_PREFLUSH|REQ_FUAto guarantee the ordering requirements the journal has w.r.t. metadata writeback. THe two ordering constraints are:1. we cannot overwrite metadata in the journal until we guarantee that the dirty metadata has been written back in place and is stable.
2. we cannot write back dirty metadata until it has been written to the journal and guaranteed to be stable (and hence recoverable) in the journal.
The ordering guarantees of #1 are provided by
REQ_PREFLUSH. This causes the journal IO to issue a cache flush and wait for it to complete before issuing the write IO to the journal. Hence all completed metadata IO is guaranteed to be stable before the journal overwrites the old metadata.The ordering guarantees of #2 are provided by the
REQ_FUA, which ensures the journal writes do not complete until they are on stable storage. Hence by the time the last journal IO in a checkpoint completes, we know that the entire checkpoint is on stable storage and we can unpin the dirty metadata and allow it to be written back.This is the mechanism by which ordering was first implemented in XFS way back in 2002 by commit
95d97c36e5155075ba2eb22b17562cfcc53fcf96("Add support for drive write cache flushing") in the xfs-archive tree.A lot has changed since then, most notably we now use delayed logging to checkpoint the filesystem to the journal rather than write each individual transaction to the journal. Cache flushes on journal IO are necessary when individual transactions are wholly contained within a single iclog. However, CIL checkpoints are single transactions that typically span hundreds to thousands of individual journal writes, and so the requirements for device cache flushing have changed.
That is, the ordering rules I state above apply to ordering of atomic transactions recorded in the journal, not to the journal IO itself. Hence we need to ensure metadata is stable before we start writing a new transaction to the journal (guarantee #1), and we need to ensure the entire transaction is stable in the journal before we start metadata writeback (guarantee #2).
Hence we only need a
REQ_PREFLUSHon the journal IO that starts a new journal transaction to provide #1, and it is not on any other journal IO done within the context of that journal transaction.The CIL checkpoint already issues a cache flush before it starts writing to the log, so we no longer need the iclog IO to issue a
REQ_REFLUSHfor us. Hence ifXLOG_START_TRANSis passed toxlog_write(), we no longer need to mark the first iclog in the log write withREQ_PREFLUSHfor this case. As an added bonus, this ordering mechanism works for both internal and external logs, meaning we can remove the explicit data device cache flushes from the iclog write code when using external logs.Given the new ordering semantics of commit records for the CIL, we need iclogs containing commit records to issue a
REQ_PREFLUSH. We also require unmount records to do this. Hence for bothXLOG_COMMIT_TRANSandXLOG_UNMOUNT_TRANSxlog_write()calls we need to mark the first iclog being written withREQ_PREFLUSH.For both commit records and unmount records, we also want them immediately on stable storage, so we want to also mark the iclogs that contain these records to be marked
REQ_FUA. That means if a record is split across multiple iclogs, they are all markedREQ_FUAand not just the last one so that when the transaction is completed all the parts of the record are on stable storage.And for external logs, unmount records need a pre-write data device cache flush similar to the CIL checkpoint cache pre-flush as the internal iclog write code does not do this implicitly anymore.
As an optimisation, when the commit record lands in the same iclog as the journal transaction starts, we don't need to wait for anything and can simply use
REQ_FUAto provide guarantee #2. This means that forfsync()heavy workloads, the cache flush behaviour is completely unchanged and there is no degradation in performance as a result of optimise the multi-IO transaction case.The most notable sign that there is less IO latency on my test machine (nvme SSDs) is that the "noiclogs" rate has dropped substantially. This metric indicates that the CIL push is blocking in
xlog_get_iclog_space()waiting for iclog IO completion to occur. With 8 iclogs of 256kB, the rate is appoximately 1 noiclog event to every 4 iclog writes. IOWs, every 4th call toxlog_get_iclog_space()is blocking waiting for log IO. With the changes in this patch, this drops to 1 noiclog event for every 100 iclog writes. Hence it is clear that log IO is completing much faster than it was previously, but it is also clear that for large iclog sizes, this isn't the performance limiting factor on this hardware.
Quoted 1 line
With smaller iclogs (32kB), however, there is a sustantial
s/sustantial/substantial/
Quoted 20 lines
difference. With the cache flush modifications, the journal is now running at over 4000 write IOPS, and the journal throughput is largely identical to the 256kB iclogs and the noiclog event rate stays low at about 1:50 iclog writes. The existing code tops out at about 2500 IOPS as the number of cache flushes dominate performance and latency. The noiclog event rate is about 1:4, and the performance variance is quite large as the journal throughput can fall to less than half the peak sustained rate when the cache flush rate prevents metadata writeback from keeping up and the log runs out of space and throttles reservations.
As a result:
logbsize fsmark create rate rm -rf
before 32kb 152851+/-5.3e+04 5m28s
patched 32kb 221533+/-1.1e+04 5m24sbefore 256kb 220239+/-6.2e+03 4m58s
patched 256kb 228286+/-9.2e+03 5m06s
Quoted 8 lines
The rm -rf times are included because I ran them, but the differences are largely noise. This workload is largely metadata read IO latency bound and the changes to the journal cache flushing doesn't really make any noticable difference to behaviour apart from a reduction in noiclog events from background CIL pushing.
Signed-off-by Dave Chinner <[email protected]>Reviewed-by Chandan Babu R <[email protected]>
With that fixed,
--D
Quoted 230 lines
--- fs/xfs/xfs_log.c | 66 +++++++++++++++---------------------------- fs/xfs/xfs_log.h | 1 - fs/xfs/xfs_log_cil.c | 18 +++++++++--- fs/xfs/xfs_log_priv.h | 6 ++++ 4 files changed, 43 insertions(+), 48 deletions(-)fs/xfs/xfs_log.c +23
-43@@ -513,7 +513,7 @@ __xlog_state_release_iclog(* Flush iclog to disk if this is the last reference to the given iclog and the * it is in the WANT_SYNC state. */ -static int +int xlog_state_release_iclog (struct xlog *log,struct xlog_in_core *iclog)@@ -533,23 +533,6 @@ xlog_state_release_iclog(return 0 ; } -void -xfs_log_release_iclog ( -struct xlog_in_core *iclog) -{ -struct xlog *log = iclog->ic_log; -bool sync = false; - -if (atomic_dec_and_lock (&iclog->ic_refcnt, &log->l_icloglock)) { -if (iclog->ic_state !=XLOG_STATE_IOERROR ) - sync =__xlog_state_release_iclog (log, iclog); -spin_unlock (&log->l_icloglock); - } - -if (sync) -xlog_sync (log, iclog); -} -/* * Mount a log filesystem * @@ -837,6 +820,14 @@ xlog_write_unmount_record(/* account for space used by record data */ ticket->t_curr_res -=sizeof (ulf); + +/* +* For external log devices, we need to flush the data device cache +* first to ensure all metadata writeback is on stable storage before we +* stamp the tail LSN into the unmount record. +*/ +if (log->l_targ != log->l_mp->m_ddev_targp) +blkdev_issue_flush (log->l_targ->bt_bdev);return xlog_write (log, &vec, ticket,NULL ,NULL ,XLOG_UNMOUNT_TRANS ); }@@ -874,6 +865,11 @@ xlog_unmount_write(else ASSERT (iclog->ic_state ==XLOG_STATE_WANT_SYNC || iclog->ic_state ==XLOG_STATE_IOERROR ); +/* +* Ensure the journal is fully flushed and on stable storage once the +* iclog containing the unmount record is written. +*/ + iclog->ic_flags |= (XLOG_ICL_NEED_FLUSH |XLOG_ICL_NEED_FUA ); error =xlog_state_release_iclog (log, iclog);xlog_wait_on_iclog (iclog);@@ -1755,8 +1751,7 @@ xlog_write_iclog(struct xlog *log, struct xlog_in_core *iclog, uint64_t bno, - unsigned int count, - bool need_flush) + unsigned int count) { ASSERT(bno < log->l_logBBsize);@@ -1794,10 +1789,12 @@ xlog_write_iclog(* writeback throttle from throttling log writes behind background * metadata writeback and causing priority inversions. */ - iclog->ic_bio.bi_opf =REQ_OP_WRITE |REQ_META |REQ_SYNC | -REQ_IDLE |REQ_FUA ; -if (need_flush) + iclog->ic_bio.bi_opf =REQ_OP_WRITE |REQ_META |REQ_SYNC |REQ_IDLE ; +if (iclog->ic_flags &XLOG_ICL_NEED_FLUSH ) iclog->ic_bio.bi_opf |=REQ_PREFLUSH ; +if (iclog->ic_flags &XLOG_ICL_NEED_FUA ) + iclog->ic_bio.bi_opf |=REQ_FUA ; + iclog->ic_flags &= ~(XLOG_ICL_NEED_FLUSH |XLOG_ICL_NEED_FUA );if (xlog_map_iclog_data (&iclog->ic_bio, iclog->ic_data, count)) {xfs_force_shutdown (log->l_mp,SHUTDOWN_LOG_IO_ERROR );@@ -1900,7 +1897,6 @@ xlog_sync(unsigned int roundoff;/* roundoff to BB or stripe */ uint64_t bno;unsigned int size; -bool need_flush = true, split = false;ASSERT (atomic_read (&iclog->ic_refcnt) ==0 );@@ -1925,10 +1921,8 @@ xlog_sync( bno =BLOCK_LSN (be64_to_cpu (iclog->ic_header.h_lsn));/* Do we need to split this write into 2 parts? */ -if (bno +BTOBB (count) > log->l_logBBsize) { +if (bno +BTOBB (count) > log->l_logBBsize)xlog_split_iclog (log, &iclog->ic_header, bno, count); - split = true; - }/* calculcate the checksum */ iclog->ic_header.h_crc = xlog_cksum(log, &iclog->ic_header,@@ -1949,22 +1943,8 @@ xlog_sync( be64_to_cpu(iclog->ic_header.h_lsn)); }#endif - -/* -* Flush the data device before flushing the log to make sure all meta -* data written back from the AIL actually made it to disk before -* stamping the new log tail LSN into the log buffer. For an external -* log we need to issue the flush explicitly, and unfortunately -* synchronously here; for an internal log we can simply use the block -* layer state machine for preflushes. -*/ -if (log->l_targ != log->l_mp->m_ddev_targp || split) { -blkdev_issue_flush (log->l_mp->m_ddev_targp->bt_bdev); - need_flush = false; - } -xlog_verify_iclog (log, iclog, count); -xlog_write_iclog (log, iclog, bno, count, need_flush); +xlog_write_iclog (log, iclog, bno, count); }/* @@ -2418,7 +2398,7 @@ xlog_write(ASSERT (log_offset <= iclog->ic_size -1 ); ptr = iclog->ic_datap + log_offset; -/* start_lsn is the first lsn written to. That's all we need. */ +/* Start_lsn is the first lsn written to. */ if (start_lsn && !*start_lsn) *start_lsn =be64_to_cpu (iclog->ic_header.h_lsn);fs/xfs/xfs_log.h +0
-1@@ -117,7 +117,6 @@ void xfs_log_mount_cancel(struct xfs_mount *);xfs_lsn_t xlog_assign_tail_lsn (struct xfs_mount *mp);xfs_lsn_t xlog_assign_tail_lsn_locked (struct xfs_mount *mp);void xfs_log_space_wake (struct xfs_mount *mp); -void xfs_log_release_iclog (struct xlog_in_core *iclog); int xfs_log_reserve(struct xfs_mount *mp, int length, int count,fs/xfs/xfs_log_cil.c +14
-4@@ -890,15 +890,25 @@ xlog_cil_push_work(/* * If the checkpoint spans multiple iclogs, wait for all previous -* iclogs to complete before we submit the commit_iclog. +* iclogs to complete before we submit the commit_iclog. In this case, +* the commit_iclog write needs to issue a pre-flush so that the +* ordering is correctly preserved down to stable storage. */ +spin_lock (&log->l_icloglock);if (ctx->start_lsn != commit_lsn) { -spin_lock (&log->l_icloglock);xlog_wait_on_iclog (commit_iclog->ic_prev); +spin_lock (&log->l_icloglock); + commit_iclog->ic_flags |=XLOG_ICL_NEED_FLUSH ; } -/* release the hounds! */ -xfs_log_release_iclog (commit_iclog); +/* +* The commit iclog must be written to stable storage to guarantee +* journal IO vs metadata writeback IO is correctly ordered on stable +* storage. +*/ + commit_iclog->ic_flags |=XLOG_ICL_NEED_FUA ; +xlog_state_release_iclog (log, commit_iclog); +spin_unlock (&log->l_icloglock);return ; out_skip:fs/xfs/xfs_log_priv.h +6
-0@@ -133,6 +133,9 @@ enum xlog_iclog_state {#define XLOG_COVER_OPS 5 +#define XLOG_ICL_NEED_FLUSH (1 << 0)/* iclog needs REQ_PREFLUSH */ +#define XLOG_ICL_NEED_FUA (1 << 1)/* iclog needs REQ_FUA */ +/* Ticket reservation region accounting */ #define XLOG_TIC_LEN_MAX 15@@ -201,6 +204,7 @@ typedef struct xlog_in_core {u32 ic_size;u32 ic_offset;enum xlog_iclog_state ic_state; +unsigned int ic_flags;char *ic_datap;/* pointer to iclog data */ /* Callback structures need their own cacheline */ @@ -486,6 +490,8 @@ int xlog_commit_record(struct xlog *log, struct xlog_ticket *ticket,void xfs_log_ticket_ungrant (struct xlog *log,struct xlog_ticket *ticket);void xfs_log_ticket_regrant (struct xlog *log,struct xlog_ticket *ticket); +int xlog_state_release_iclog (struct xlog *log,struct xlog_in_core *iclog); +/* * When we crack an atomic LSN, we sample it first so that the value will not * change while we are cracking it into the component values. This means we Signature · 1 line
2.31.1
PATCH v4 8/39 xfs: Fix CIL throttle hang when CIL space used going backwards
In reply to Dave ChinnerFrom: Dave Chinner <[email protected]>
A hang with tasks stuck on the CIL hard throttle was reported and largely diagnosed by Donald Buczek, who discovered that it was a result of the CIL context space usage decrementing in committed transactions once the hard throttle limit had been hit and processes were already blocked. This resulted in the CIL push not waking up those waiters because the CIL context was no longer over the hard throttle limit.
The surprising aspect of this was the CIL space usage going backwards regularly enough to trigger this situation. Assumptions had been made in design that the relogging process would only increase the size of the objects in the CIL, and so that space would only increase.
This change and commit message fixes the issue and documents the result of an audit of the triggers that can cause the CIL space to go backwards, how large the backwards steps tend to be, the frequency in which they occur, and what the impact on the CIL accounting code is.
Even though the CIL ctx->space_used can go backwards, it will only do so if the log item is already logged to the CIL and contains a space reservation for it's entire logged state. This is tracked by the shadow buffer state on the log item. If the item is not previously logged in the CIL it has no shadow buffer nor log vector, and hence the entire size of the logged item copied to the log vector is accounted to the CIL space usage. i.e. it will always go up in this case.
If the item has a log vector (i.e. already in the CIL) and the size decreases, then the existing log vector will be overwritten and the space usage will go down. This is the only condition where the space usage reduces, and it can only occur when an item is already tracked in the CIL. Hence we are safe from CIL space usage underruns as a result of log items decreasing in size when they are relogged.
Typically this reduction in CIL usage occurs from metadata blocks being free, such as when a btree block merge occurs or a directory enter/xattr entry is removed and the da-tree is reduced in size. This generally results in a reduction in size of around a single block in the CIL, but also tends to increase the number of log vectors because the parent and sibling nodes in the tree needs to be updated when a btree block is removed. If a multi-level merge occurs, then we see reduction in size of 2+ blocks, but again the log vector count goes up.
The other vector is inode fork size changes, which only log the current size of the fork and ignore the previously logged size when the fork is relogged. Hence if we are removing items from the inode fork (dir/xattr removal in shortform, extent record removal in extent form, etc) the relogged size of the inode for can decrease.
No other log items can decrease in size either because they are a fixed size (e.g. dquots) or they cannot be relogged (e.g. relogging an intent actually creates a new intent log item and doesn't relog the old item at all.) Hence the only two vectors for CIL context size reduction are relogging inode forks and marking buffers active in the CIL as stale.
Long story short: the majority of the code does the right thing and handles the reduction in log item size correctly, and only the CIL hard throttle implementation is problematic and needs fixing. This patch makes that fix, as well as adds comments in the log item code that result in items shrinking in size when they are relogged as a clear reminder that this can and does happen frequently.
The throttle fix is based upon the change Donald proposed, though it goes further to ensure that once the throttle is activated, it captures all tasks until the CIL push issues a wakeup, regardless of whether the CIL space used has gone back under the throttle threshold.
This ensures that we prevent tasks reducing the CIL slightly under the throttle threshold and then making more changes that push it well over the throttle limit. This is acheived by checking if the throttle wait queue is already active as a condition of throttling. Hence once we start throttling, we continue to apply the throttle until the CIL context push wakes everything on the wait queue.
We can use waitqueue_active() for the waitqueue manipulations and checks as they are all done under the ctx->xc_push_lock. Hence the waitqueue has external serialisation and we can safely peek inside the wait queue without holding the internal waitqueue locks.
Many thanks to Donald for his diagnostic and analysis work to isolate the cause of this hang.
--- fs/xfs/xfs_buf_item.c | 37 ++++++++++++++++++------------------- fs/xfs/xfs_inode_item.c | 14 ++++++++++++++ fs/xfs/xfs_log_cil.c | 22 +++++++++++++++++----- 3 files changed, 49 insertions(+), 24 deletions(-)
fs/xfs/xfs_buf_item.c +18 -19
@@ -74,14 +74,12 @@ xfs_buf_item_straddle( }/* -* This returns the number of log iovecs needed to log the -* given buf log item. +* Return the number of log iovecs and space needed to log the given buf log +* item segment. * -* It calculates this as 1 iovec for the buf log format structure -* and 1 for each stretch of non-contiguous chunks to be logged. -* Contiguous chunks are logged in a single iovec. -* -* If the XFS_BLI_STALE flag has been set, then log nothing. +* It calculates this as 1 iovec for the buf log format structure and 1 for each +* stretch of non-contiguous chunks to be logged. Contiguous chunks are logged +* in a single iovec. */ STATIC void xfs_buf_item_size_segment(
@@ -168,11 +166,8 @@ xfs_buf_item_size_segment( }/* -* This returns the number of log iovecs needed to log the given buf log item. -* -* It calculates this as 1 iovec for the buf log format structure and 1 for each -* stretch of non-contiguous chunks to be logged. Contiguous chunks are logged -* in a single iovec. +* Return the number of log iovecs and space needed to log the given buf log +* item. * * Discontiguous buffers need a format structure per region that is being * logged. This makes the changes in the buffer appear to log recovery as though
@@ -182,7 +177,11 @@ xfs_buf_item_size_segment(* what ends up on disk. * * If the XFS_BLI_STALE flag has been set, then log nothing but the buf log -* format structures. +* format structures. If the item has previously been logged and has dirty +* regions, we do not relog them in stale buffers. This has the effect of +* reducing the size of the relogged item by the amount of dirty data tracked +* by the log item. This can result in the committing transaction reducing the +* amount of space being consumed by the CIL. */ STATIC void xfs_buf_item_size(
@@ -199,9 +198,9 @@ xfs_buf_item_size(ASSERT (atomic_read (&bip->bli_refcount) >0 );if (bip->bli_flags &XFS_BLI_STALE ) {/* -* The buffer is stale, so all we need to log -* is the buf log format structure with the -* cancel flag in it. +* The buffer is stale, so all we need to log is the buf log +* format structure with the cancel flag in it as we are never +* going to replay the changes tracked in the log item. */ trace_xfs_buf_item_size_stale (bip);ASSERT (bip->__bli_format.blf_flags &XFS_BLF_CANCEL );
@@ -216,9 +215,9 @@ xfs_buf_item_size(if (bip->bli_flags &XFS_BLI_ORDERED ) {/* -* The buffer has been logged just to order it. -* It is not being included in the transaction -* commit, so no vectors are used at all. +* The buffer has been logged just to order it. It is not being +* included in the transaction commit, so no vectors are used at +* all. */ trace_xfs_buf_item_size_ordered (bip); *nvecs =XFS_LOG_VEC_ORDERED ;
fs/xfs/xfs_inode_item.c +14 -0
@@ -28,6 +28,20 @@ static inline struct xfs_inode_log_item *INODE_ITEM(struct xfs_log_item *lip) return container_of(lip, struct xfs_inode_log_item, ili_item); } +/* +* The logged size of an inode fork is always the current size of the inode +* fork. This means that when an inode fork is relogged, the size of the logged +* region is determined by the current state, not the combination of the +* previously logged state + the current state. This is different relogging +* behaviour to most other log items which will retain the size of the +* previously logged changes when smaller regions are relogged. +* +* Hence operations that remove data from the inode fork (e.g. shortform +* dir/attr remove, extent form extent removal, etc), the size of the relogged +* inode gets -smaller- rather than stays the same size as the previously logged +* size and this can result in the committing transaction reducing the amount of +* space being consumed by the CIL. +*/ STATIC void xfs_inode_item_data_fork_size( struct xfs_inode_log_item *iip,
fs/xfs/xfs_log_cil.c +17 -5
@@ -670,9 +670,14 @@ xlog_cil_push_work(ASSERT (push_seq <= ctx->sequence);/* -* Wake up any background push waiters now this context is being pushed. +* As we are about to switch to a new, empty CIL context, we no longer +* need to throttle tasks on CIL space overruns. Wake any waiters that +* the hard push throttle may have caught so they can start committing +* to the new context. The ctx->xc_push_lock provides the serialisation +* necessary for safely using the lockless waitqueue_active() check in +* this context. */ -if (ctx->space_used >=XLOG_CIL_BLOCKING_SPACE_LIMIT (log)) +if (waitqueue_active (&cil->xc_push_wait))wake_up_all (&cil->xc_push_wait);/*
@@ -944,7 +949,7 @@ xlog_cil_push_background(ASSERT (!list_empty (&cil->xc_cil));/* -* don't do a background push if we haven't used up all the +* Don't do a background push if we haven't used up all the * space available yet. */ if (cil->xc_ctx->space_used <XLOG_CIL_SPACE_LIMIT (log)) {
@@ -968,9 +973,16 @@ xlog_cil_push_background(/* * If we are well over the space limit, throttle the work that is being -* done until the push work on this context has begun. +* done until the push work on this context has begun. Enforce the hard +* throttle on all transaction commits once it has been activated, even +* if the committing transactions have resulted in the space usage +* dipping back down under the hard limit. +* +* The ctx->xc_push_lock provides the serialisation necessary for safely +* using the lockless waitqueue_active() check in this context. */ -if (cil->xc_ctx->space_used >=XLOG_CIL_BLOCKING_SPACE_LIMIT (log)) { +if (cil->xc_ctx->space_used >=XLOG_CIL_BLOCKING_SPACE_LIMIT (log) || +waitqueue_active (&cil->xc_push_wait)) {trace_xfs_log_cil_wait (log, cil->xc_ctx->ticket);ASSERT (cil->xc_ctx->space_used < log->l_logsize);xlog_wait (&cil->xc_push_wait, &cil->xc_push_lock);
Signature · 1 line
2.31.1
PATCH v4 9/39 xfs: xfs_log_force_lsn isn't passed a LSN
In reply to Dave ChinnerFrom: Dave Chinner <[email protected]>
In doing an investigation into AIL push stalls, I was looking at the log force code to see if an async CIL push could be done instead. This lead me to xfs_log_force_lsn() and looking at how it works.
xfs_log_force_lsn() is only called from inode synchronisation contexts such as fsync(), and it takes the ip->i_itemp->ili_last_lsn value as the LSN to sync the log to. This gets passed to xlog_cil_force_lsn() via xfs_log_force_lsn() to flush the CIL to the journal, and then used by xfs_log_force_lsn() to flush the iclogs to the journal.
The problem is that ip->i_itemp->ili_last_lsn does not store a log sequence number. What it stores is passed to it from the ->iop_committing method, which is called by xfs_log_commit_cil(). The value this passes to the iop_committing method is the CIL context sequence number that the item was committed to.
As it turns out, xlog_cil_force_lsn() converts the sequence to an actual commit LSN for the related context and returns that to xfs_log_force_lsn(). xfs_log_force_lsn() overwrites it's "lsn" variable that contained a sequence with an actual LSN and then uses that to sync the iclogs.
This caused me some confusion for a while, even though I originally wrote all this code a decade ago. ->iop_committing is only used by a couple of log item types, and only inode items use the sequence number it is passed.
Let's clean up the API, CIL structures and inode log item to call it a sequence number, and make it clear that the high level code is using CIL sequence numbers and not on-disk LSNs for integrity synchronisation purposes.
--- fs/xfs/libxfs/xfs_types.h | 1 + fs/xfs/xfs_buf_item.c | 2 +- fs/xfs/xfs_dquot_item.c | 2 +- fs/xfs/xfs_file.c | 14 +++++++------- fs/xfs/xfs_inode.c | 10 +++++----- fs/xfs/xfs_inode_item.c | 4 ++-- fs/xfs/xfs_inode_item.h | 2 +- fs/xfs/xfs_log.c | 27 ++++++++++++++------------- fs/xfs/xfs_log.h | 4 +--- fs/xfs/xfs_log_cil.c | 30 +++++++++++------------------- fs/xfs/xfs_log_priv.h | 15 +++++++-------- fs/xfs/xfs_trans.c | 6 +++--- fs/xfs/xfs_trans.h | 4 ++-- 13 files changed, 56 insertions(+), 65 deletions(-)
fs/xfs/libxfs/xfs_types.h +1 -0
@@ -21,6 +21,7 @@ typedef int32_t xfs_suminfo_t; /* type of bitmap summary info */typedef uint32_t xfs_rtword_t ;/* word type for bitmap manipulations */ typedef int64_t xfs_lsn_t ;/* log sequence number */ +typedef int64_t xfs_csn_t ;/* CIL sequence number */ typedef uint32_t xfs_dablk_t ;/* dir/attr block number (in file) */ typedef uint32_t xfs_dahash_t ;/* dir/attr hash value */
fs/xfs/xfs_buf_item.c +1 -1
@@ -713,7 +713,7 @@ xfs_buf_item_release( STATIC voidxfs_buf_item_committing (struct xfs_log_item *lip, -xfs_lsn_t commit_lsn) +xfs_csn_t seq) {return xfs_buf_item_release (lip); }
fs/xfs/xfs_dquot_item.c +1 -1
@@ -188,7 +188,7 @@ xfs_qm_dquot_logitem_release( STATIC voidxfs_qm_dquot_logitem_committing (struct xfs_log_item *lip, -xfs_lsn_t commit_lsn) +xfs_csn_t seq) {return xfs_qm_dquot_logitem_release (lip); }
fs/xfs/xfs_file.c +7 -7
@@ -119,8 +119,8 @@ xfs_dir_fsync(return xfs_log_force_inode (ip); } -static xfs_lsn_t -xfs_fsync_lsn ( +static xfs_csn_t +xfs_fsync_seq (struct xfs_inode *ip,bool datasync) {
@@ -128,7 +128,7 @@ xfs_fsync_lsn(return 0 ;if (datasync && !(ip->i_itemp->ili_fsync_fields & ~XFS_ILOG_TIMESTAMP ))return 0 ; -return ip->i_itemp->ili_last_lsn; +return ip->i_itemp->ili_commit_seq; }/*
@@ -151,12 +151,12 @@ xfs_fsync_flush_log( int *log_flushed) {int error =0 ; -xfs_lsn_t lsn; +xfs_csn_t seq;xfs_ilock (ip,XFS_ILOCK_SHARED ); - lsn =xfs_fsync_lsn (ip, datasync); -if (lsn) { - error =xfs_log_force_lsn (ip->i_mount, lsn,XFS_LOG_SYNC , + seq =xfs_fsync_seq (ip, datasync); +if (seq) { + error =xfs_log_force_seq (ip->i_mount, seq,XFS_LOG_SYNC , log_flushed);spin_lock (&ip->i_itemp->ili_lock);
fs/xfs/xfs_inode.c +5 -5
@@ -2604,7 +2604,7 @@ xfs_iunpin(trace_xfs_inode_unpin_nowait (ip, _RET_IP_);/* Give the log a push to start the unpinning I/O */ -xfs_log_force_lsn (ip->i_mount, ip->i_itemp->ili_last_lsn,0 ,NULL ); +xfs_log_force_seq (ip->i_mount, ip->i_itemp->ili_commit_seq,0 ,NULL ); }
@@ -3618,16 +3618,16 @@ intxfs_log_force_inode ( struct xfs_inode *ip) { -xfs_lsn_t lsn =0 ; +xfs_csn_t seq =0 ;xfs_ilock (ip,XFS_ILOCK_SHARED );if (xfs_ipincount (ip)) - lsn = ip->i_itemp->ili_last_lsn; + seq = ip->i_itemp->ili_commit_seq;xfs_iunlock (ip,XFS_ILOCK_SHARED ); -if (!lsn) +if (!seq)return 0 ; -return xfs_log_force_lsn (ip->i_mount, lsn,XFS_LOG_SYNC ,NULL ); +return xfs_log_force_seq (ip->i_mount, seq,XFS_LOG_SYNC ,NULL ); }/*
fs/xfs/xfs_inode_item.c +2 -2
@@ -643,9 +643,9 @@ xfs_inode_item_committed( STATIC voidxfs_inode_item_committing (struct xfs_log_item *lip, -xfs_lsn_t commit_lsn) +xfs_csn_t seq) { -INODE_ITEM (lip)->ili_last_lsn = commit_lsn; +INODE_ITEM (lip)->ili_commit_seq = seq;return xfs_inode_item_release (lip); }
fs/xfs/xfs_inode_item.h +1 -1
@@ -33,7 +33,7 @@ struct xfs_inode_log_item {unsigned int ili_fields;/* fields to be logged */ unsigned int ili_fsync_fields;/* logged since last fsync */ xfs_lsn_t ili_flush_lsn;/* lsn at last flush */ -xfs_lsn_t ili_last_lsn;/* lsn at last transaction */ +xfs_csn_t ili_commit_seq;/* last transaction commit */ };static inline int xfs_inode_clean (struct xfs_inode *ip)
fs/xfs/xfs_log.c +14 -13
@@ -3252,14 +3252,13 @@ xfs_log_force( }static int -__xfs_log_force_lsn ( -struct xfs_mount *mp, +xlog_force_lsn ( +struct xlog *log,xfs_lsn_t lsn,uint flags,int *log_flushed,bool already_slept) { -struct xlog *log = mp->m_log;struct xlog_in_core *iclog;spin_lock (&log->l_icloglock);
@@ -3292,8 +3291,6 @@ __xfs_log_force_lsn(if (!already_slept && (iclog->ic_prev->ic_state ==XLOG_STATE_WANT_SYNC || iclog->ic_prev->ic_state ==XLOG_STATE_SYNCING )) { -XFS_STATS_INC (mp, xs_log_force_sleep); -xlog_wait (&iclog->ic_prev->ic_write_wait, &log->l_icloglock);return -EAGAIN ;
@@ -3331,25 +3328,29 @@ __xfs_log_force_lsn(* to disk, that thread will wake up all threads waiting on the queue. */ int -xfs_log_force_lsn ( +xfs_log_force_seq (struct xfs_mount *mp, -xfs_lsn_t lsn, +xfs_csn_t seq,uint flags,int *log_flushed) { +struct xlog *log = mp->m_log; +xfs_lsn_t lsn;int ret; -ASSERT (lsn !=0 ); +ASSERT (seq !=0 );XFS_STATS_INC (mp, xs_log_force); -trace_xfs_log_force (mp, lsn, _RET_IP_); +trace_xfs_log_force (mp, seq, _RET_IP_); - lsn =xlog_cil_force_lsn (mp->m_log, lsn); + lsn =xlog_cil_force_seq (log, seq);if (lsn ==NULLCOMMITLSN )return 0 ; - ret =__xfs_log_force_lsn (mp, lsn, flags, log_flushed, false); -if (ret == -EAGAIN ) - ret =__xfs_log_force_lsn (mp, lsn, flags, log_flushed, true); + ret =xlog_force_lsn (log, lsn, flags, log_flushed, false); +if (ret == -EAGAIN ) { +XFS_STATS_INC (mp, xs_log_force_sleep); + ret =xlog_force_lsn (log, lsn, flags, log_flushed, true); + }return ret; }
fs/xfs/xfs_log.h +1 -3
@@ -106,7 +106,7 @@ struct xfs_item_ops;struct xfs_trans ;int xfs_log_force (struct xfs_mount *mp,uint flags); -int xfs_log_force_lsn (struct xfs_mount *mp,xfs_lsn_t lsn,uint flags, +int xfs_log_force_seq (struct xfs_mount *mp,xfs_csn_t seq,uint flags,int *log_forced); int xfs_log_mount(struct xfs_mount *mp, struct xfs_buftarg *log_target,
@@ -131,8 +131,6 @@ bool xfs_log_writable(struct xfs_mount *mp);struct xlog_ticket *xfs_log_ticket_get (struct xlog_ticket *ticket);void xfs_log_ticket_put (struct xlog_ticket *ticket); -void xfs_log_commit_cil (struct xfs_mount *mp,struct xfs_trans *tp, -xfs_lsn_t *commit_lsn,bool regrant);void xlog_cil_process_committed (struct list_head *list);bool xfs_log_item_in_current_chkpt (struct xfs_log_item *lip);
fs/xfs/xfs_log_cil.c +11 -19
@@ -788,7 +788,7 @@ xlog_cil_push_work(* that higher sequences will wait for us to write out a commit record * before they do. * -* xfs_log_force_lsn requires us to mirror the new sequence into the cil +* xfs_log_force_seq requires us to mirror the new sequence into the cil * structure atomically with the addition of this sequence to the * committing list. This also ensures that we can do unlocked checks * against the current sequence in log forces without risking
@@ -1057,16 +1057,14 @@ xlog_cil_empty(* allowed again. */ void -xfs_log_commit_cil ( -struct xfs_mount *mp, +xlog_cil_commit ( +struct xlog *log,struct xfs_trans *tp, -xfs_lsn_t *commit_lsn, +xfs_csn_t *commit_seq,bool regrant) { -struct xlog *log = mp->m_log;struct xfs_cil *cil = log->l_cilp;struct xfs_log_item *lip, *next; -xfs_lsn_t xc_commit_lsn;/* * Do all necessary memory allocation before we lock the CIL.
@@ -1080,10 +1078,6 @@ xfs_log_commit_cil(xlog_cil_insert_items (log, tp); - xc_commit_lsn = cil->xc_ctx->sequence; -if (commit_lsn) - *commit_lsn = xc_commit_lsn; -if (regrant && !XLOG_FORCED_SHUTDOWN (log))xfs_log_ticket_regrant (log, tp->t_ticket);else
@@ -1106,8 +1100,10 @@ xfs_log_commit_cil(list_for_each_entry_safe (lip, next, &tp->t_items, li_trans) {xfs_trans_del_item (lip);if (lip->li_ops->iop_committing) - lip->li_ops->iop_committing (lip, xc_commit_lsn); + lip->li_ops->iop_committing (lip, cil->xc_ctx->sequence); } +if (commit_seq) + *commit_seq = cil->xc_ctx->sequence;/* xlog_cil_push_background() releases cil->xc_ctx_lock */ xlog_cil_push_background (log);
@@ -1124,9 +1120,9 @@ xfs_log_commit_cil(* iclog flush is necessary following this call. */ xfs_lsn_t -xlog_cil_force_lsn ( +xlog_cil_force_seq (struct xlog *log, -xfs_lsn_t sequence) +xfs_csn_t sequence) {struct xfs_cil *cil = log->l_cilp;struct xfs_cil_ctx *ctx;
@@ -1222,21 +1218,17 @@ boolxfs_log_item_in_current_chkpt ( struct xfs_log_item *lip) { -struct xfs_cil_ctx *ctx; +struct xfs_cil_ctx *ctx = lip->li_mountp->m_log->l_cilp->xc_ctx;if (list_empty (&lip->li_cil))return false; - ctx = lip->li_mountp->m_log->l_cilp->xc_ctx; -/* * li_seq is written on the first commit of a log item to record the * first checkpoint it is written to. Hence if it is different to the * current sequence, we're in a new checkpoint. */ -if (XFS_LSN_CMP (lip->li_seq, ctx->sequence) !=0 ) -return false; -return true; +return lip->li_seq == ctx->sequence; }/*
fs/xfs/xfs_log_priv.h +7 -8
@@ -234,7 +234,7 @@ struct xfs_cil;struct xfs_cil_ctx {struct xfs_cil *cil; -xfs_lsn_t sequence;/* chkpt sequence # */ +xfs_csn_t sequence;/* chkpt sequence # */ xfs_lsn_t start_lsn;/* first LSN of chkpt commit */ xfs_lsn_t commit_lsn;/* chkpt commit record lsn */ struct xlog_ticket *ticket;/* chkpt ticket */
@@ -272,10 +272,10 @@ struct xfs_cil {struct xfs_cil_ctx *xc_ctx;spinlock_t xc_push_lock____cacheline_aligned_in_smp ; -xfs_lsn_t xc_push_seq; +xfs_csn_t xc_push_seq;struct list_head xc_committing;wait_queue_head_t xc_commit_wait; -xfs_lsn_t xc_current_sequence; +xfs_csn_t xc_current_sequence;struct work_struct xc_push_work;wait_queue_head_t xc_push_wait;/* background push throttle */ }____cacheline_aligned_in_smp ;
@@ -554,19 +554,18 @@ int xlog_cil_init(struct xlog *log);void xlog_cil_init_post_recovery (struct xlog *log);void xlog_cil_destroy (struct xlog *log);bool xlog_cil_empty (struct xlog *log); +void xlog_cil_commit (struct xlog *log,struct xfs_trans *tp, +xfs_csn_t *commit_seq,bool regrant);/* * CIL force routines */ -xfs_lsn_t -xlog_cil_force_lsn ( -struct xlog *log, -xfs_lsn_t sequence); +xfs_lsn_t xlog_cil_force_seq (struct xlog *log,xfs_csn_t sequence);static inline void xlog_cil_force (struct xlog *log) { -xlog_cil_force_lsn (log, log->l_cilp->xc_current_sequence); +xlog_cil_force_seq (log, log->l_cilp->xc_current_sequence); }/*
fs/xfs/xfs_trans.c +3 -3
@@ -839,7 +839,7 @@ __xfs_trans_commit( bool regrant) {struct xfs_mount *mp = tp->t_mountp; -xfs_lsn_t commit_lsn =-1 ; +xfs_csn_t commit_seq =0 ;int error =0 ;int sync = tp->t_flags &XFS_TRANS_SYNC ;
@@ -881,7 +881,7 @@ __xfs_trans_commit(xfs_trans_apply_sb_deltas (tp);xfs_trans_apply_dquot_deltas (tp); -xfs_log_commit_cil (mp, tp, &commit_lsn, regrant); +xlog_cil_commit (mp->m_log, tp, &commit_seq, regrant);xfs_trans_free (tp);
@@ -890,7 +890,7 @@ __xfs_trans_commit(* log out now and wait for it. */ if (sync) { - error =xfs_log_force_lsn (mp, commit_lsn,XFS_LOG_SYNC ,NULL ); + error =xfs_log_force_seq (mp, commit_seq,XFS_LOG_SYNC ,NULL );XFS_STATS_INC (mp, xs_trans_sync); }else {XFS_STATS_INC (mp, xs_trans_async);
fs/xfs/xfs_trans.h +2 -2
@@ -43,7 +43,7 @@ struct xfs_log_item {struct list_head li_cil;/* CIL pointers */ struct xfs_log_vec *li_lv;/* active log vector */ struct xfs_log_vec *li_lv_shadow;/* standby vector */ -xfs_lsn_t li_seq;/* CIL commit seq */ +xfs_csn_t li_seq;/* CIL commit seq */ };/*
@@ -69,7 +69,7 @@ struct xfs_item_ops {void (*iop_pin)(struct xfs_log_item *);void (*iop_unpin)(struct xfs_log_item *,int remove);uint (*iop_push)(struct xfs_log_item *,struct list_head *); -void (*iop_committing)(struct xfs_log_item *,xfs_lsn_t commit_lsn); +void (*iop_committing)(struct xfs_log_item *lip,xfs_csn_t seq);void (*iop_release)(struct xfs_log_item *);xfs_lsn_t (*iop_committed)(struct xfs_log_item *,xfs_lsn_t ); int (*iop_recover)(struct xfs_log_item *lip,
Signature · 1 line
2.31.1
Re: [PATCH 09/39] xfs: xfs_log_force_lsn isn't passed a LSN
Quoted 30 lines · Dave Chinner
From: Dave Chinner <[email protected]>
In doing an investigation into AIL push stalls, I was looking at the log force code to see if an async CIL push could be done instead. This lead me to
xfs_log_force_lsn()and looking at how it works.
xfs_log_force_lsn()is only called from inode synchronisation contexts such asfsync(), and it takes theip->i_itemp->ili_last_lsnvalue as the LSN to sync the log to. This gets passed toxlog_cil_force_lsn()viaxfs_log_force_lsn()to flush the CIL to the journal, and then used byxfs_log_force_lsn()to flush the iclogs to the journal.The problem is that
ip->i_itemp->ili_last_lsndoes not store a log sequence number. What it stores is passed to it from the ->iop_committing method, which is called byxfs_log_commit_cil(). The value this passes to theiop_committingmethod is the CIL context sequence number that the item was committed to.As it turns out,
xlog_cil_force_lsn()converts the sequence to an actual commit LSN for the related context and returns that toxfs_log_force_lsn().xfs_log_force_lsn()overwrites it's "lsn" variable that contained a sequence with an actual LSN and then uses that to sync the iclogs.This caused me some confusion for a while, even though I originally wrote all this code a decade ago. ->iop_committing is only used by a couple of log item types, and only inode items use the sequence number it is passed.
Quoted 7 lines
Let's clean up the API, CIL structures and inode log item to call it a sequence number, and make it clear that the high level code is using CIL sequence numbers and not on-disk LSNs for integrity synchronisation purposes.
Signed-off-by Dave Chinner <[email protected]>Reviewed-by Brian Foster <[email protected]>
Looks good this time,
--D
Quoted 446 lines
--- fs/xfs/libxfs/xfs_types.h | 1 + fs/xfs/xfs_buf_item.c | 2 +- fs/xfs/xfs_dquot_item.c | 2 +- fs/xfs/xfs_file.c | 14 +++++++------- fs/xfs/xfs_inode.c | 10 +++++----- fs/xfs/xfs_inode_item.c | 4 ++-- fs/xfs/xfs_inode_item.h | 2 +- fs/xfs/xfs_log.c | 27 ++++++++++++++------------- fs/xfs/xfs_log.h | 4 +--- fs/xfs/xfs_log_cil.c | 30 +++++++++++------------------- fs/xfs/xfs_log_priv.h | 15 +++++++-------- fs/xfs/xfs_trans.c | 6 +++--- fs/xfs/xfs_trans.h | 4 ++-- 13 files changed, 56 insertions(+), 65 deletions(-)fs/xfs/libxfs/xfs_types.h +1
-0@@ -21,6 +21,7 @@ typedef int32_t xfs_suminfo_t; /* type of bitmap summary info */typedef uint32_t xfs_rtword_t ;/* word type for bitmap manipulations */ typedef int64_t xfs_lsn_t ;/* log sequence number */ +typedef int64_t xfs_csn_t ;/* CIL sequence number */ typedef uint32_t xfs_dablk_t ;/* dir/attr block number (in file) */ typedef uint32_t xfs_dahash_t ;/* dir/attr hash value */ fs/xfs/xfs_buf_item.c +1
-1@@ -713,7 +713,7 @@ xfs_buf_item_release( STATIC voidxfs_buf_item_committing (struct xfs_log_item *lip, -xfs_lsn_t commit_lsn) +xfs_csn_t seq) {return xfs_buf_item_release (lip); }fs/xfs/xfs_dquot_item.c +1
-1@@ -188,7 +188,7 @@ xfs_qm_dquot_logitem_release( STATIC voidxfs_qm_dquot_logitem_committing (struct xfs_log_item *lip, -xfs_lsn_t commit_lsn) +xfs_csn_t seq) {return xfs_qm_dquot_logitem_release (lip); }fs/xfs/xfs_file.c +7
-7@@ -119,8 +119,8 @@ xfs_dir_fsync(return xfs_log_force_inode (ip); } -static xfs_lsn_t -xfs_fsync_lsn ( +static xfs_csn_t +xfs_fsync_seq (struct xfs_inode *ip,bool datasync) {@@ -128,7 +128,7 @@ xfs_fsync_lsn(return 0 ;if (datasync && !(ip->i_itemp->ili_fsync_fields & ~XFS_ILOG_TIMESTAMP ))return 0 ; -return ip->i_itemp->ili_last_lsn; +return ip->i_itemp->ili_commit_seq; }/* @@ -151,12 +151,12 @@ xfs_fsync_flush_log( int *log_flushed) {int error =0 ; -xfs_lsn_t lsn; +xfs_csn_t seq;xfs_ilock (ip,XFS_ILOCK_SHARED ); - lsn =xfs_fsync_lsn (ip, datasync); -if (lsn) { - error =xfs_log_force_lsn (ip->i_mount, lsn,XFS_LOG_SYNC , + seq =xfs_fsync_seq (ip, datasync); +if (seq) { + error =xfs_log_force_seq (ip->i_mount, seq,XFS_LOG_SYNC , log_flushed);spin_lock (&ip->i_itemp->ili_lock);fs/xfs/xfs_inode.c +5
-5@@ -2604,7 +2604,7 @@ xfs_iunpin(trace_xfs_inode_unpin_nowait (ip, _RET_IP_);/* Give the log a push to start the unpinning I/O */ -xfs_log_force_lsn (ip->i_mount, ip->i_itemp->ili_last_lsn,0 ,NULL ); +xfs_log_force_seq (ip->i_mount, ip->i_itemp->ili_commit_seq,0 ,NULL ); }@@ -3618,16 +3618,16 @@ intxfs_log_force_inode ( struct xfs_inode *ip) { -xfs_lsn_t lsn =0 ; +xfs_csn_t seq =0 ;xfs_ilock (ip,XFS_ILOCK_SHARED );if (xfs_ipincount (ip)) - lsn = ip->i_itemp->ili_last_lsn; + seq = ip->i_itemp->ili_commit_seq;xfs_iunlock (ip,XFS_ILOCK_SHARED ); -if (!lsn) +if (!seq)return 0 ; -return xfs_log_force_lsn (ip->i_mount, lsn,XFS_LOG_SYNC ,NULL ); +return xfs_log_force_seq (ip->i_mount, seq,XFS_LOG_SYNC ,NULL ); }/* fs/xfs/xfs_inode_item.c +2
-2@@ -643,9 +643,9 @@ xfs_inode_item_committed( STATIC voidxfs_inode_item_committing (struct xfs_log_item *lip, -xfs_lsn_t commit_lsn) +xfs_csn_t seq) { -INODE_ITEM (lip)->ili_last_lsn = commit_lsn; +INODE_ITEM (lip)->ili_commit_seq = seq;return xfs_inode_item_release (lip); }fs/xfs/xfs_inode_item.h +1
-1@@ -33,7 +33,7 @@ struct xfs_inode_log_item {unsigned int ili_fields;/* fields to be logged */ unsigned int ili_fsync_fields;/* logged since last fsync */ xfs_lsn_t ili_flush_lsn;/* lsn at last flush */ -xfs_lsn_t ili_last_lsn;/* lsn at last transaction */ +xfs_csn_t ili_commit_seq;/* last transaction commit */ };static inline int xfs_inode_clean (struct xfs_inode *ip)fs/xfs/xfs_log.c +14
-13@@ -3252,14 +3252,13 @@ xfs_log_force( }static int -__xfs_log_force_lsn ( -struct xfs_mount *mp, +xlog_force_lsn ( +struct xlog *log,xfs_lsn_t lsn,uint flags,int *log_flushed,bool already_slept) { -struct xlog *log = mp->m_log;struct xlog_in_core *iclog;spin_lock (&log->l_icloglock);@@ -3292,8 +3291,6 @@ __xfs_log_force_lsn(if (!already_slept && (iclog->ic_prev->ic_state ==XLOG_STATE_WANT_SYNC || iclog->ic_prev->ic_state ==XLOG_STATE_SYNCING )) { -XFS_STATS_INC (mp, xs_log_force_sleep); -xlog_wait (&iclog->ic_prev->ic_write_wait, &log->l_icloglock);return -EAGAIN ;@@ -3331,25 +3328,29 @@ __xfs_log_force_lsn(* to disk, that thread will wake up all threads waiting on the queue. */ int -xfs_log_force_lsn ( +xfs_log_force_seq (struct xfs_mount *mp, -xfs_lsn_t lsn, +xfs_csn_t seq,uint flags,int *log_flushed) { +struct xlog *log = mp->m_log; +xfs_lsn_t lsn;int ret; -ASSERT (lsn !=0 ); +ASSERT (seq !=0 );XFS_STATS_INC (mp, xs_log_force); -trace_xfs_log_force (mp, lsn, _RET_IP_); +trace_xfs_log_force (mp, seq, _RET_IP_); - lsn =xlog_cil_force_lsn (mp->m_log, lsn); + lsn =xlog_cil_force_seq (log, seq);if (lsn ==NULLCOMMITLSN )return 0 ; - ret =__xfs_log_force_lsn (mp, lsn, flags, log_flushed, false); -if (ret == -EAGAIN ) - ret =__xfs_log_force_lsn (mp, lsn, flags, log_flushed, true); + ret =xlog_force_lsn (log, lsn, flags, log_flushed, false); +if (ret == -EAGAIN ) { +XFS_STATS_INC (mp, xs_log_force_sleep); + ret =xlog_force_lsn (log, lsn, flags, log_flushed, true); + }return ret; }fs/xfs/xfs_log.h +1
-3@@ -106,7 +106,7 @@ struct xfs_item_ops;struct xfs_trans ;int xfs_log_force (struct xfs_mount *mp,uint flags); -int xfs_log_force_lsn (struct xfs_mount *mp,xfs_lsn_t lsn,uint flags, +int xfs_log_force_seq (struct xfs_mount *mp,xfs_csn_t seq,uint flags,int *log_forced); int xfs_log_mount(struct xfs_mount *mp, struct xfs_buftarg *log_target,@@ -131,8 +131,6 @@ bool xfs_log_writable(struct xfs_mount *mp);struct xlog_ticket *xfs_log_ticket_get (struct xlog_ticket *ticket);void xfs_log_ticket_put (struct xlog_ticket *ticket); -void xfs_log_commit_cil (struct xfs_mount *mp,struct xfs_trans *tp, -xfs_lsn_t *commit_lsn,bool regrant);void xlog_cil_process_committed (struct list_head *list);bool xfs_log_item_in_current_chkpt (struct xfs_log_item *lip);fs/xfs/xfs_log_cil.c +11
-19@@ -788,7 +788,7 @@ xlog_cil_push_work(* that higher sequences will wait for us to write out a commit record * before they do. * -* xfs_log_force_lsn requires us to mirror the new sequence into the cil +* xfs_log_force_seq requires us to mirror the new sequence into the cil * structure atomically with the addition of this sequence to the * committing list. This also ensures that we can do unlocked checks * against the current sequence in log forces without risking @@ -1057,16 +1057,14 @@ xlog_cil_empty(* allowed again. */ void -xfs_log_commit_cil ( -struct xfs_mount *mp, +xlog_cil_commit ( +struct xlog *log,struct xfs_trans *tp, -xfs_lsn_t *commit_lsn, +xfs_csn_t *commit_seq,bool regrant) { -struct xlog *log = mp->m_log;struct xfs_cil *cil = log->l_cilp;struct xfs_log_item *lip, *next; -xfs_lsn_t xc_commit_lsn;/* * Do all necessary memory allocation before we lock the CIL. @@ -1080,10 +1078,6 @@ xfs_log_commit_cil(xlog_cil_insert_items (log, tp); - xc_commit_lsn = cil->xc_ctx->sequence; -if (commit_lsn) - *commit_lsn = xc_commit_lsn; -if (regrant && !XLOG_FORCED_SHUTDOWN (log))xfs_log_ticket_regrant (log, tp->t_ticket);else @@ -1106,8 +1100,10 @@ xfs_log_commit_cil(list_for_each_entry_safe (lip, next, &tp->t_items, li_trans) {xfs_trans_del_item (lip);if (lip->li_ops->iop_committing) - lip->li_ops->iop_committing (lip, xc_commit_lsn); + lip->li_ops->iop_committing (lip, cil->xc_ctx->sequence); } +if (commit_seq) + *commit_seq = cil->xc_ctx->sequence;/* xlog_cil_push_background() releases cil->xc_ctx_lock */ xlog_cil_push_background (log);@@ -1124,9 +1120,9 @@ xfs_log_commit_cil(* iclog flush is necessary following this call. */ xfs_lsn_t -xlog_cil_force_lsn ( +xlog_cil_force_seq (struct xlog *log, -xfs_lsn_t sequence) +xfs_csn_t sequence) {struct xfs_cil *cil = log->l_cilp;struct xfs_cil_ctx *ctx;@@ -1222,21 +1218,17 @@ boolxfs_log_item_in_current_chkpt ( struct xfs_log_item *lip) { -struct xfs_cil_ctx *ctx; +struct xfs_cil_ctx *ctx = lip->li_mountp->m_log->l_cilp->xc_ctx;if (list_empty (&lip->li_cil))return false; - ctx = lip->li_mountp->m_log->l_cilp->xc_ctx; -/* * li_seq is written on the first commit of a log item to record the * first checkpoint it is written to. Hence if it is different to the * current sequence, we're in a new checkpoint. */ -if (XFS_LSN_CMP (lip->li_seq, ctx->sequence) !=0 ) -return false; -return true; +return lip->li_seq == ctx->sequence; }/* fs/xfs/xfs_log_priv.h +7
-8@@ -234,7 +234,7 @@ struct xfs_cil;struct xfs_cil_ctx {struct xfs_cil *cil; -xfs_lsn_t sequence;/* chkpt sequence # */ +xfs_csn_t sequence;/* chkpt sequence # */ xfs_lsn_t start_lsn;/* first LSN of chkpt commit */ xfs_lsn_t commit_lsn;/* chkpt commit record lsn */ struct xlog_ticket *ticket;/* chkpt ticket */ @@ -272,10 +272,10 @@ struct xfs_cil {struct xfs_cil_ctx *xc_ctx;spinlock_t xc_push_lock____cacheline_aligned_in_smp ; -xfs_lsn_t xc_push_seq; +xfs_csn_t xc_push_seq;struct list_head xc_committing;wait_queue_head_t xc_commit_wait; -xfs_lsn_t xc_current_sequence; +xfs_csn_t xc_current_sequence;struct work_struct xc_push_work;wait_queue_head_t xc_push_wait;/* background push throttle */ }____cacheline_aligned_in_smp ;@@ -554,19 +554,18 @@ int xlog_cil_init(struct xlog *log);void xlog_cil_init_post_recovery (struct xlog *log);void xlog_cil_destroy (struct xlog *log);bool xlog_cil_empty (struct xlog *log); +void xlog_cil_commit (struct xlog *log,struct xfs_trans *tp, +xfs_csn_t *commit_seq,bool regrant);/* * CIL force routines */ -xfs_lsn_t -xlog_cil_force_lsn ( -struct xlog *log, -xfs_lsn_t sequence); +xfs_lsn_t xlog_cil_force_seq (struct xlog *log,xfs_csn_t sequence);static inline void xlog_cil_force (struct xlog *log) { -xlog_cil_force_lsn (log, log->l_cilp->xc_current_sequence); +xlog_cil_force_seq (log, log->l_cilp->xc_current_sequence); }/* fs/xfs/xfs_trans.c +3
-3@@ -839,7 +839,7 @@ __xfs_trans_commit( bool regrant) {struct xfs_mount *mp = tp->t_mountp; -xfs_lsn_t commit_lsn =-1 ; +xfs_csn_t commit_seq =0 ;int error =0 ;int sync = tp->t_flags &XFS_TRANS_SYNC ;@@ -881,7 +881,7 @@ __xfs_trans_commit(xfs_trans_apply_sb_deltas (tp);xfs_trans_apply_dquot_deltas (tp); -xfs_log_commit_cil (mp, tp, &commit_lsn, regrant); +xlog_cil_commit (mp->m_log, tp, &commit_seq, regrant);xfs_trans_free (tp);@@ -890,7 +890,7 @@ __xfs_trans_commit(* log out now and wait for it. */ if (sync) { - error =xfs_log_force_lsn (mp, commit_lsn,XFS_LOG_SYNC ,NULL ); + error =xfs_log_force_seq (mp, commit_seq,XFS_LOG_SYNC ,NULL );XFS_STATS_INC (mp, xs_trans_sync); }else {XFS_STATS_INC (mp, xs_trans_async);fs/xfs/xfs_trans.h +2
-2@@ -43,7 +43,7 @@ struct xfs_log_item {struct list_head li_cil;/* CIL pointers */ struct xfs_log_vec *li_lv;/* active log vector */ struct xfs_log_vec *li_lv_shadow;/* standby vector */ -xfs_lsn_t li_seq;/* CIL commit seq */ +xfs_csn_t li_seq;/* CIL commit seq */ };/* @@ -69,7 +69,7 @@ struct xfs_item_ops {void (*iop_pin)(struct xfs_log_item *);void (*iop_unpin)(struct xfs_log_item *,int remove);uint (*iop_push)(struct xfs_log_item *,struct list_head *); -void (*iop_committing)(struct xfs_log_item *,xfs_lsn_t commit_lsn); +void (*iop_committing)(struct xfs_log_item *lip,xfs_csn_t seq);void (*iop_release)(struct xfs_log_item *);xfs_lsn_t (*iop_committed)(struct xfs_log_item *,xfs_lsn_t ); int (*iop_recover)(struct xfs_log_item *lip,Signature · 1 line
2.31.1
PATCH v4 10/39 xfs: AIL needs asynchronous CIL forcing
In reply to Dave ChinnerFrom: Dave Chinner <[email protected]>
The AIL pushing is stalling on log forces when it comes across pinned items. This is happening on removal workloads where the AIL is dominated by stale items that are removed from AIL when the checkpoint that marks the items stale is committed to the journal. This results is relatively few items in the AIL, but those that are are often pinned as directories items are being removed from are still being logged.
As a result, many push cycles through the CIL will first issue a blocking log force to unpin the items. This can take some time to complete, with tracing regularly showing push delays of half a second and sometimes up into the range of several seconds. Sequences like this aren't uncommon:
....
399.829437: xfsaild: last lsn 0x11002dd000 count 101 stuck 101 flushing 0 tout 20
<wanted 20ms, got 270ms delay>
400.099622: xfsaild: target 0x11002f3600, prev 0x11002f3600, last lsn 0x0 400.099623: xfsaild: first lsn 0x11002f3600 400.099679: xfsaild: last lsn 0x1100305000 count 16 stuck 11 flushing 0 tout 50 <wanted 50ms, got 500ms delay> 400.589348: xfsaild: target 0x110032e600, prev 0x11002f3600, last lsn 0x0 400.589349: xfsaild: first lsn 0x1100305000 400.589595: xfsaild: last lsn 0x110032e600 count 156 stuck 101 flushing 30 tout 50 <wanted 50ms, got 460ms delay> 400.950341: xfsaild: target 0x1100353000, prev 0x110032e600, last lsn 0x0 400.950343: xfsaild: first lsn 0x1100317c00 400.950436: xfsaild: last lsn 0x110033d200 count 105 stuck 101 flushing 0 tout 20 <wanted 20ms, got 200ms delay> 401.142333: xfsaild: target 0x1100361600, prev 0x1100353000, last lsn 0x0 401.142334: xfsaild: first lsn 0x110032e600 401.142535: xfsaild: last lsn 0x1100353000 count 122 stuck 101 flushing 8 tout 10 <wanted 10ms, got 10ms delay> 401.154323: xfsaild: target 0x1100361600, prev 0x1100361600, last lsn 0x1100353000 401.154328: xfsaild: first lsn 0x1100353000 401.154389: xfsaild: last lsn 0x1100353000 count 101 stuck 101 flushing 0 tout 20 <wanted 20ms, got 300ms delay> 401.451525: xfsaild: target 0x1100361600, prev 0x1100361600, last lsn 0x0 401.451526: xfsaild: first lsn 0x1100353000 401.451804: xfsaild: last lsn 0x1100377200 count 170 stuck 22 flushing 122 tout 50
<wanted 50ms, got 500ms delay>
401.933581: xfsaild: target 0x1100361600, prev 0x1100361600, last lsn 0x0
....
In each of these cases, every AIL pass saw 101 log items stuck on the AIL (pinned) with very few other items being found. Each pass, a log force was issued, and delay between last/first is the sleep time
+ the sync log force time.
Some of these 101 items pinned the tail of the log. The tail of the log does slowly creep forward (first lsn), but the problem is that the log is actually out of reservation space because it's been running so many transactions that stale items that never reach the AIL but consume log space. Hence we have a largely empty AIL, with long term pins on items that pin the tail of the log that don't get pushed frequently enough to keep log space available.
The problem is the hundreds of milliseconds that we block in the log force pushing the CIL out to disk. The AIL should not be stalled like this - it needs to run and flush items that are at the tail of the log with minimal latency. What we really need to do is trigger a log flush, but then not wait for it at all - we've already done our waiting for stuff to complete when we backed off prior to the log force being issued.
Even if we remove the XFS_LOG_SYNC from the xfs_log_force() call, we still do a blocking flush of the CIL and that is what is causing the issue. Hence we need a new interface for the CIL to trigger an immediate background push of the CIL to get it moving faster but not to wait on that to occur. While the CIL is pushing, the AIL can also be pushing.
We already have an internal interface to do this - xlog_cil_push_now() - but we need a wrapper for it to be used externally. xlog_cil_force_seq() can easily be extended to do what we need as it already implements the synchronous CIL push via xlog_cil_push_now(). Add the necessary flags and "push current sequence" semantics to xlog_cil_force_seq() and convert the AIL pushing to use it.
One of the complexities here is that the CIL push does not guarantee that the commit record for the CIL checkpoint is written to disk. The current log force ensures this by submitting the current ACTIVE iclog that the commit record was written to. We need the CIL to actually write this commit record to disk for an async push to ensure that the checkpoint actually makes it to disk and unpins the pinned items in the checkpoint on completion. Hence we need to pass down to the CIL push that we are doing an async flush so that it can switch out the commit_iclog if necessary to get written to disk when the commit iclog is finally released.
--- fs/xfs/xfs_log.c | 38 ++++++++++++++------------ fs/xfs/xfs_log.h | 1 + fs/xfs/xfs_log_cil.c | 61 ++++++++++++++++++++++++++++++++++++------ fs/xfs/xfs_log_priv.h | 5 ++++ fs/xfs/xfs_sysfs.c | 1 + fs/xfs/xfs_trace.c | 1 + fs/xfs/xfs_trans.c | 2 +- fs/xfs/xfs_trans_ail.c | 11 +++++--- 8 files changed, 91 insertions(+), 29 deletions(-)
fs/xfs/xfs_log.c +21 -17
@@ -50,11 +50,6 @@ xlog_state_get_iclog_space(int *continued_write, int *logoffsetp); STATIC void -xlog_state_switch_iclogs ( -struct xlog *log, -struct xlog_in_core *iclog, -int eventual_size); -STATIC voidxlog_grant_push_ail (struct xlog *log,int need_bytes);
@@ -3104,7 +3099,7 @@ xfs_log_ticket_ungrant(* This routine will mark the current iclog in the ring as WANT_SYNC and move * the current iclog pointer to the next iclog in the ring. */ -STATIC void +void xlog_state_switch_iclogs( struct xlog *log, struct xlog_in_core *iclog,
@@ -3251,6 +3246,20 @@ xfs_log_force(return -EIO ; } +/* +* Force the log to a specific LSN. +* +* If an iclog with that lsn can be found: +* If it is in the DIRTY state, just return. +* If it is in the ACTIVE state, move the in-core log into the WANT_SYNC +* state and go to sleep or return. +* If it is in any other state, go to sleep or return. +* +* Synchronous forces are implemented with a wait queue. All callers trying +* to force a given lsn to disk must wait on the queue attached to the +* specific in-core log. When given in-core log finally completes its write +* to disk, that thread will wake up all threads waiting on the queue. +*/ static int xlog_force_lsn( struct xlog *log,
@@ -3314,18 +3323,13 @@ xlog_force_lsn( }/* -* Force the in-core log to disk for a specific LSN. +* Force the log to a specific checkpoint sequence. * -* Find in-core log with lsn. -* If it is in the DIRTY state, just return. -* If it is in the ACTIVE state, move the in-core log into the WANT_SYNC -* state and go to sleep or return. -* If it is in any other state, go to sleep or return. -* -* Synchronous forces are implemented with a wait queue. All callers trying -* to force a given lsn to disk must wait on the queue attached to the -* specific in-core log. When given in-core log finally completes its write -* to disk, that thread will wake up all threads waiting on the queue. +* First force the CIL so that all the required changes have been flushed to the +* iclogs. If the CIL force completed it will return a commit LSN that indicates +* the iclog that needs to be flushed to stable storage. If the caller needs +* a synchronous log force, we will wait on the iclog with the LSN returned by +* xlog_cil_force_seq() to be completed. */ int xfs_log_force_seq(
fs/xfs/xfs_log.h +1 -0
@@ -104,6 +104,7 @@ struct xlog_ticket;struct xfs_log_item ;struct xfs_item_ops ;struct xfs_trans ; +struct xlog ;int xfs_log_force (struct xfs_mount *mp,uint flags); int xfs_log_force_seq(struct xfs_mount *mp, xfs_csn_t seq, uint flags,
fs/xfs/xfs_log_cil.c +53 -8
130 lines not on this page · raw message
fs/xfs/xfs_log_priv.h +5 -0
@@ -273,6 +273,7 @@ struct xfs_cil {spinlock_t xc_push_lock____cacheline_aligned_in_smp ;xfs_csn_t xc_push_seq; +bool xc_push_commit_stable;struct list_head xc_committing;wait_queue_head_t xc_commit_wait;xfs_csn_t xc_current_sequence;
@@ -487,9 +488,12 @@ int xlog_write(struct xlog *log, struct xfs_log_vec *log_vector, struct xlog_in_core **commit_iclog, uint optype);int xlog_commit_record (struct xlog *log,struct xlog_ticket *ticket,struct xlog_in_core **iclog,xfs_lsn_t *lsn); +void xfs_log_ticket_ungrant (struct xlog *log,struct xlog_ticket *ticket);void xfs_log_ticket_regrant (struct xlog *log,struct xlog_ticket *ticket); +void xlog_state_switch_iclogs (struct xlog *log,struct xlog_in_core *iclog, +int eventual_size);int xlog_state_release_iclog (struct xlog *log,struct xlog_in_core *iclog);/*
@@ -560,6 +564,7 @@ void xlog_cil_commit(struct xlog *log, struct xfs_trans *tp,/* * CIL force routines */ +void xlog_cil_flush (struct xlog *log);xfs_lsn_t xlog_cil_force_seq (struct xlog *log,xfs_csn_t sequence); static inline void
fs/xfs/xfs_sysfs.c +1 -0
@@ -10,6 +10,7 @@#include "xfs_log_format.h" #include "xfs_trans_resv.h" #include "xfs_sysfs.h" +#include "xfs_log.h" #include "xfs_log_priv.h" #include "xfs_mount.h"
fs/xfs/xfs_trace.c +1 -0
@@ -20,6 +20,7 @@#include "xfs_bmap.h" #include "xfs_attr.h" #include "xfs_trans.h" +#include "xfs_log.h" #include "xfs_log_priv.h" #include "xfs_buf_item.h" #include "xfs_quota.h"
fs/xfs/xfs_trans.c +1 -1
@@ -9,7 +9,6 @@#include "xfs_shared.h" #include "xfs_format.h" #include "xfs_log_format.h" -#include "xfs_log_priv.h" #include "xfs_trans_resv.h" #include "xfs_mount.h" #include "xfs_extent_busy.h"
@@ -17,6 +16,7 @@#include "xfs_trans.h" #include "xfs_trans_priv.h" #include "xfs_log.h" +#include "xfs_log_priv.h" #include "xfs_trace.h" #include "xfs_error.h" #include "xfs_defer.h"
fs/xfs/xfs_trans_ail.c +8 -3
@@ -17,6 +17,7 @@#include "xfs_errortag.h" #include "xfs_error.h" #include "xfs_log.h" +#include "xfs_log_priv.h" #ifdef DEBUG /*
@@ -429,8 +430,12 @@ xfsaild_push(/* * If we encountered pinned items or did not finish writing out all -* buffers the last time we ran, force the log first and wait for it -* before pushing again. +* buffers the last time we ran, force a background CIL push to get the +* items unpinned in the near future. We do not wait on the CIL push as +* that could stall us for seconds if there is enough background IO +* load. Stalling for that long when the tail of the log is pinned and +* needs flushing will hard stop the transaction subsystem when log +* space runs out. */ if (ailp->ail_log_flush && ailp->ail_last_pushed_lsn == 0 && (!list_empty_careful(&ailp->ail_buf_list) ||
@@ -438,7 +443,7 @@ xfsaild_push( ailp->ail_log_flush =0 ;XFS_STATS_INC (mp, xs_push_ail_flush); -xfs_log_force (mp,XFS_LOG_SYNC ); +xlog_cil_flush (mp->m_log); }spin_lock (&ailp->ail_lock);
Signature · 1 line
2.31.1
Re: [PATCH 10/39] xfs: AIL needs asynchronous CIL forcing
Quoted 93 lines · Dave Chinner
From: Dave Chinner <[email protected]>
The AIL pushing is stalling on log forces when it comes across pinned items. This is happening on removal workloads where the AIL is dominated by stale items that are removed from AIL when the checkpoint that marks the items stale is committed to the journal. This results is relatively few items in the AIL, but those that are are often pinned as directories items are being removed from are still being logged.
As a result, many push cycles through the CIL will first issue a blocking log force to unpin the items. This can take some time to complete, with tracing regularly showing push delays of half a second and sometimes up into the range of several seconds. Sequences like this aren't uncommon:
....
399.829437: xfsaild: last lsn0x11002dd000count 101 stuck 101 flushing 0 tout 20
<wanted 20ms, got 270ms delay>400.099622: xfsaild: target 0x11002f3600, prev 0x11002f3600, last lsn 0x0 400.099623: xfsaild: first lsn 0x11002f3600 400.099679: xfsaild: last lsn 0x1100305000 count 16 stuck 11 flushing 0 tout 50 <wanted 50ms, got 500ms delay> 400.589348: xfsaild: target 0x110032e600, prev 0x11002f3600, last lsn 0x0 400.589349: xfsaild: first lsn 0x1100305000 400.589595: xfsaild: last lsn 0x110032e600 count 156 stuck 101 flushing 30 tout 50 <wanted 50ms, got 460ms delay> 400.950341: xfsaild: target 0x1100353000, prev 0x110032e600, last lsn 0x0 400.950343: xfsaild: first lsn 0x1100317c00 400.950436: xfsaild: last lsn 0x110033d200 count 105 stuck 101 flushing 0 tout 20 <wanted 20ms, got 200ms delay> 401.142333: xfsaild: target 0x1100361600, prev 0x1100353000, last lsn 0x0 401.142334: xfsaild: first lsn 0x110032e600 401.142535: xfsaild: last lsn 0x1100353000 count 122 stuck 101 flushing 8 tout 10 <wanted 10ms, got 10ms delay> 401.154323: xfsaild: target 0x1100361600, prev 0x1100361600, last lsn 0x1100353000 401.154328: xfsaild: first lsn 0x1100353000 401.154389: xfsaild: last lsn 0x1100353000 count 101 stuck 101 flushing 0 tout 20 <wanted 20ms, got 300ms delay> 401.451525: xfsaild: target 0x1100361600, prev 0x1100361600, last lsn 0x0 401.451526: xfsaild: first lsn 0x1100353000 401.451804: xfsaild: last lsn 0x1100377200 count 170 stuck 22 flushing 122 tout 50<wanted 50ms, got 500ms delay>
401.933581: xfsaild: target0x1100361600, prev0x1100361600, last lsn0x0
....In each of these cases, every AIL pass saw 101 log items stuck on the AIL (pinned) with very few other items being found. Each pass, a log force was issued, and delay between last/first is the sleep time
+ the sync log force time.Some of these 101 items pinned the tail of the log. The tail of the log does slowly creep forward (first lsn), but the problem is that the log is actually out of reservation space because it's been running so many transactions that stale items that never reach the AIL but consume log space. Hence we have a largely empty AIL, with long term pins on items that pin the tail of the log that don't get pushed frequently enough to keep log space available.
The problem is the hundreds of milliseconds that we block in the log force pushing the CIL out to disk. The AIL should not be stalled like this - it needs to run and flush items that are at the tail of the log with minimal latency. What we really need to do is trigger a log flush, but then not wait for it at all - we've already done our waiting for stuff to complete when we backed off prior to the log force being issued.
Even if we remove the
XFS_LOG_SYNCfrom thexfs_log_force()call, we still do a blocking flush of the CIL and that is what is causing the issue. Hence we need a new interface for the CIL to trigger an immediate background push of the CIL to get it moving faster but not to wait on that to occur. While the CIL is pushing, the AIL can also be pushing.We already have an internal interface to do this -
xlog_cil_push_now()- but we need a wrapper for it to be used externally.xlog_cil_force_seq()can easily be extended to do what we need as it already implements the synchronous CIL push viaxlog_cil_push_now(). Add the necessary flags and "push current sequence" semantics toxlog_cil_force_seq()and convert the AIL pushing to use it.One of the complexities here is that the CIL push does not guarantee that the commit record for the CIL checkpoint is written to disk. The current log force ensures this by submitting the current ACTIVE iclog that the commit record was written to. We need the CIL to actually write this commit record to disk for an async push to ensure that the checkpoint actually makes it to disk and unpins the pinned items in the checkpoint on completion. Hence we need to pass down to the CIL push that we are doing an async flush so that it can switch out the
commit_iclogif necessary to get written to disk when the commit iclog is finally released.
Quoted 1 line
Signed-off-by Dave Chinner <[email protected]>
Not sure if I like this new push_commit_stable name, but at least it doesn't have foo_sync = bar_async like last time.
--D
Quoted 353 lines
--- fs/xfs/xfs_log.c | 38 ++++++++++++++------------ fs/xfs/xfs_log.h | 1 + fs/xfs/xfs_log_cil.c | 61 ++++++++++++++++++++++++++++++++++++------ fs/xfs/xfs_log_priv.h | 5 ++++ fs/xfs/xfs_sysfs.c | 1 + fs/xfs/xfs_trace.c | 1 + fs/xfs/xfs_trans.c | 2 +- fs/xfs/xfs_trans_ail.c | 11 +++++--- 8 files changed, 91 insertions(+), 29 deletions(-)fs/xfs/xfs_log.c +21
-17@@ -50,11 +50,6 @@ xlog_state_get_iclog_space(int *continued_write, int *logoffsetp); STATIC void -xlog_state_switch_iclogs ( -struct xlog *log, -struct xlog_in_core *iclog, -int eventual_size); -STATIC voidxlog_grant_push_ail (struct xlog *log,int need_bytes);@@ -3104,7 +3099,7 @@ xfs_log_ticket_ungrant(* This routine will mark the current iclog in the ring as WANT_SYNC and move * the current iclog pointer to the next iclog in the ring. */ -STATIC void +void xlog_state_switch_iclogs( struct xlog *log, struct xlog_in_core *iclog,@@ -3251,6 +3246,20 @@ xfs_log_force(return -EIO ; } +/* +* Force the log to a specific LSN. +* +* If an iclog with that lsn can be found: +* If it is in the DIRTY state, just return. +* If it is in the ACTIVE state, move the in-core log into the WANT_SYNC +* state and go to sleep or return. +* If it is in any other state, go to sleep or return. +* +* Synchronous forces are implemented with a wait queue. All callers trying +* to force a given lsn to disk must wait on the queue attached to the +* specific in-core log. When given in-core log finally completes its write +* to disk, that thread will wake up all threads waiting on the queue. +*/ static int xlog_force_lsn( struct xlog *log,@@ -3314,18 +3323,13 @@ xlog_force_lsn( }/* -* Force the in-core log to disk for a specific LSN. +* Force the log to a specific checkpoint sequence. * -* Find in-core log with lsn. -* If it is in the DIRTY state, just return. -* If it is in the ACTIVE state, move the in-core log into the WANT_SYNC -* state and go to sleep or return. -* If it is in any other state, go to sleep or return. -* -* Synchronous forces are implemented with a wait queue. All callers trying -* to force a given lsn to disk must wait on the queue attached to the -* specific in-core log. When given in-core log finally completes its write -* to disk, that thread will wake up all threads waiting on the queue. +* First force the CIL so that all the required changes have been flushed to the +* iclogs. If the CIL force completed it will return a commit LSN that indicates +* the iclog that needs to be flushed to stable storage. If the caller needs +* a synchronous log force, we will wait on the iclog with the LSN returned by +* xlog_cil_force_seq() to be completed. */ int xfs_log_force_seq(fs/xfs/xfs_log.h +1
-0@@ -104,6 +104,7 @@ struct xlog_ticket;struct xfs_log_item ;struct xfs_item_ops ;struct xfs_trans ; +struct xlog ;int xfs_log_force (struct xfs_mount *mp,uint flags); int xfs_log_force_seq(struct xfs_mount *mp, xfs_csn_t seq, uint flags,fs/xfs/xfs_log_cil.c +53
-8@@ -658,6 +658,7 @@ xlog_cil_push_work(xfs_lsn_t push_seq;struct bio bio;DECLARE_COMPLETION_ONSTACK (bdev_flush); +bool push_commit_stable; new_ctx =kmem_zalloc (sizeof (*new_ctx),KM_NOFS ); new_ctx->ticket =xlog_cil_ticket_alloc (log);@@ -668,6 +669,8 @@ xlog_cil_push_work(spin_lock (&cil->xc_push_lock); push_seq = cil->xc_push_seq;ASSERT (push_seq <= ctx->sequence); + push_commit_stable = cil->xc_push_commit_stable; + cil->xc_push_commit_stable = false;/* * As we are about to switch to a new, empty CIL context, we no longer @@ -910,8 +913,15 @@ xlog_cil_push_work(* The commit iclog must be written to stable storage to guarantee * journal IO vs metadata writeback IO is correctly ordered on stable * storage. +* +* If the push caller needs the commit to be immediately stable and the +* commit_iclog is not yet marked as XLOG_STATE_WANT_SYNC to indicate it +* will be written when released, switch it's state to WANT_SYNC right +* now. */ commit_iclog->ic_flags |=XLOG_ICL_NEED_FUA ; +if (push_commit_stable && commit_iclog->ic_state ==XLOG_STATE_ACTIVE ) +xlog_state_switch_iclogs (log, commit_iclog,0 );xlog_state_release_iclog (log, commit_iclog);spin_unlock (&log->l_icloglock);return ;@@ -996,13 +1006,26 @@ xlog_cil_push_background(/* * xlog_cil_push_now() is used to trigger an immediate CIL push to the sequence * number that is passed. When it returns, the work will be queued for -* @push_seq, but it won't be completed. The caller is expected to do any -* waiting for push_seq to complete if it is required. +* @push_seq, but it won't be completed. +* +* If the caller is performing a synchronous force, we will flush the workqueue +* to get previously queued work moving to minimise the wait time they will +* undergo waiting for all outstanding pushes to complete. The caller is +* expected to do the required waiting for push_seq to complete. +* +* If the caller is performing an async push, we need to ensure that the +* checkpoint is fully flushed out of the iclogs when we finish the push. If we +* don't do this, then the commit record may remain sitting in memory in an +* ACTIVE iclog. This then requires another full log force to push to disk, +* which defeats the purpose of having an async, non-blocking CIL force +* mechanism. Hence in this case we need to pass a flag to the push work to +* indicate it needs to flush the commit record itself. */ static void xlog_cil_push_now (struct xlog *log, -xfs_lsn_t push_seq) +xfs_lsn_t push_seq, +bool async) {struct xfs_cil *cil = log->l_cilp;@@ -1012,7 +1035,8 @@ xlog_cil_push_now(ASSERT (push_seq && push_seq <= cil->xc_current_sequence);/* start on any pending background push to minimise wait time on it */ -flush_work (&cil->xc_push_work); +if (!async) +flush_work (&cil->xc_push_work);/* * If the CIL is empty or we've already pushed the sequence then @@ -1025,6 +1049,7 @@ xlog_cil_push_now( } cil->xc_push_seq = push_seq; + cil->xc_push_commit_stable = async;queue_work (log->l_mp->m_cil_workqueue, &cil->xc_push_work);spin_unlock (&cil->xc_push_lock); }@@ -1109,12 +1134,27 @@ xlog_cil_commit(xlog_cil_push_background (log); } +/* +* Flush the CIL to stable storage but don't wait for it to complete. This +* requires the CIL push to ensure the commit record for the push hits the disk, +* but otherwise is no different to a push done from a log force. +*/ +void +xlog_cil_flush ( +struct xlog *log) +{ +xfs_csn_t seq = log->l_cilp->xc_current_sequence; + +trace_xfs_log_force (log->l_mp, seq, _RET_IP_); +xlog_cil_push_now (log, seq, true); +} +/* * Conditionally push the CIL based on the sequence passed in. * -* We only need to push if we haven't already pushed the sequence -* number given. Hence the only time we will trigger a push here is -* if the push sequence is the same as the current context. +* We only need to push if we haven't already pushed the sequence number given. +* Hence the only time we will trigger a push here is if the push sequence is +* the same as the current context. * * We return the current commit lsn to allow the callers to determine if a * iclog flush is necessary following this call. @@ -1130,13 +1170,17 @@ xlog_cil_force_seq(ASSERT (sequence <= cil->xc_current_sequence); +if (!sequence) + sequence = cil->xc_current_sequence; +trace_xfs_log_force (log->l_mp, sequence, _RET_IP_); +/* * check to see if we need to force out the current context. * xlog_cil_push() handles racing pushes for the same sequence, * so no need to deal with it here. */ restart: -xlog_cil_push_now (log, sequence); +xlog_cil_push_now (log, sequence, false);/* * See if we can find a previous sequence still committing. @@ -1160,6 +1204,7 @@ xlog_cil_force_seq(* It is still being pushed! Wait for the push to * complete, then start again from the beginning. */ +XFS_STATS_INC (log->l_mp, xs_log_force_sleep);xlog_wait (&cil->xc_commit_wait, &cil->xc_push_lock); goto restart; }fs/xfs/xfs_log_priv.h +5
-0@@ -273,6 +273,7 @@ struct xfs_cil {spinlock_t xc_push_lock____cacheline_aligned_in_smp ;xfs_csn_t xc_push_seq; +bool xc_push_commit_stable;struct list_head xc_committing;wait_queue_head_t xc_commit_wait;xfs_csn_t xc_current_sequence;@@ -487,9 +488,12 @@ int xlog_write(struct xlog *log, struct xfs_log_vec *log_vector, struct xlog_in_core **commit_iclog, uint optype);int xlog_commit_record (struct xlog *log,struct xlog_ticket *ticket,struct xlog_in_core **iclog,xfs_lsn_t *lsn); +void xfs_log_ticket_ungrant (struct xlog *log,struct xlog_ticket *ticket);void xfs_log_ticket_regrant (struct xlog *log,struct xlog_ticket *ticket); +void xlog_state_switch_iclogs (struct xlog *log,struct xlog_in_core *iclog, +int eventual_size);int xlog_state_release_iclog (struct xlog *log,struct xlog_in_core *iclog);/* @@ -560,6 +564,7 @@ void xlog_cil_commit(struct xlog *log, struct xfs_trans *tp,/* * CIL force routines */ +void xlog_cil_flush (struct xlog *log);xfs_lsn_t xlog_cil_force_seq (struct xlog *log,xfs_csn_t sequence); static inline voidfs/xfs/xfs_sysfs.c +1
-0@@ -10,6 +10,7 @@#include "xfs_log_format.h" #include "xfs_trans_resv.h" #include "xfs_sysfs.h" +#include "xfs_log.h" #include "xfs_log_priv.h" #include "xfs_mount.h" fs/xfs/xfs_trace.c +1
-0@@ -20,6 +20,7 @@#include "xfs_bmap.h" #include "xfs_attr.h" #include "xfs_trans.h" +#include "xfs_log.h" #include "xfs_log_priv.h" #include "xfs_buf_item.h" #include "xfs_quota.h" fs/xfs/xfs_trans.c +1
-1@@ -9,7 +9,6 @@#include "xfs_shared.h" #include "xfs_format.h" #include "xfs_log_format.h" -#include "xfs_log_priv.h" #include "xfs_trans_resv.h" #include "xfs_mount.h" #include "xfs_extent_busy.h" @@ -17,6 +16,7 @@#include "xfs_trans.h" #include "xfs_trans_priv.h" #include "xfs_log.h" +#include "xfs_log_priv.h" #include "xfs_trace.h" #include "xfs_error.h" #include "xfs_defer.h" fs/xfs/xfs_trans_ail.c +8
-3@@ -17,6 +17,7 @@#include "xfs_errortag.h" #include "xfs_error.h" #include "xfs_log.h" +#include "xfs_log_priv.h" #ifdef DEBUG /* @@ -429,8 +430,12 @@ xfsaild_push(/* * If we encountered pinned items or did not finish writing out all -* buffers the last time we ran, force the log first and wait for it -* before pushing again. +* buffers the last time we ran, force a background CIL push to get the +* items unpinned in the near future. We do not wait on the CIL push as +* that could stall us for seconds if there is enough background IO +* load. Stalling for that long when the tail of the log is pinned and +* needs flushing will hard stop the transaction subsystem when log +* space runs out. */ if (ailp->ail_log_flush && ailp->ail_last_pushed_lsn == 0 && (!list_empty_careful(&ailp->ail_buf_list) ||@@ -438,7 +443,7 @@ xfsaild_push( ailp->ail_log_flush =0 ;XFS_STATS_INC (mp, xs_push_ail_flush); -xfs_log_force (mp,XFS_LOG_SYNC ); +xlog_cil_flush (mp->m_log); }spin_lock (&ailp->ail_lock);Signature · 1 line
2.31.1
PATCH v4 11/39 xfs: CIL work is serialised, not pipelined
In reply to Dave ChinnerFrom: Dave Chinner <[email protected]>
Because we use a single work structure attached to the CIL rather than the CIL context, we can only queue a single work item at a time. This results in the CIL being single threaded and limits performance when it becomes CPU bound.
The design of the CIL is that it is pipelined and multiple commits can be running concurrently, but the way the work is currently implemented means that it is not pipelining as it was intended. The critical work to switch the CIL context can take a few milliseconds to run, but the rest of the CIL context flush can take hundreds of milliseconds to complete. The context switching is the serialisation point of the CIL, once the context has been switched the rest of the context push can run asynchrnously with all other context pushes.
Hence we can move the work to the CIL context so that we can run multiple CIL pushes at the same time and spread the majority of the work out over multiple CPUs. We can keep the per-cpu CIL commit state on the CIL rather than the context, because the context is pinned to the CIL until the switch is done and we aggregate and drain the per-cpu state held on the CIL during the context switch.
However, because we no longer serialise the CIL work, we can have effectively unlimited CIL pushes in progress. We don't want to do this - not only does it create contention on the iclogs and the state machine locks, we can run the log right out of space with outstanding pushes. Instead, limit the work concurrency to 4 concurrent works being processed at a time. THis is enough concurrency to remove the CIL from being a CPU bound bottleneck but not enough to create new contention points or unbound concurrency issues.
--- fs/xfs/xfs_log_cil.c | 80 +++++++++++++++++++++++-------------------- fs/xfs/xfs_log_priv.h | 2 +- fs/xfs/xfs_super.c | 6 +++- 3 files changed, 48 insertions(+), 40 deletions(-)
fs/xfs/xfs_log_cil.c +42 -38
@@ -47,6 +47,34 @@ xlog_cil_ticket_alloc(return tic; } +/* +* Unavoidable forward declaration - xlog_cil_push_work() calls +* xlog_cil_ctx_alloc() itself. +*/ +static void xlog_cil_push_work (struct work_struct *work); + +static struct xfs_cil_ctx * +xlog_cil_ctx_alloc (void ) +{ +struct xfs_cil_ctx *ctx; + + ctx =kmem_zalloc (sizeof (*ctx),KM_NOFS ); +INIT_LIST_HEAD (&ctx->committing); +INIT_LIST_HEAD (&ctx->busy_extents); +INIT_WORK (&ctx->push_work, xlog_cil_push_work); +return ctx; +} + +static void +xlog_cil_ctx_switch ( +struct xfs_cil *cil, +struct xfs_cil_ctx *ctx) +{ + ctx->sequence = ++cil->xc_current_sequence; + ctx->cil = cil; + cil->xc_ctx = ctx; +} +/* * After the first stage of log recovery is done, we know where the head and * tail of the log are. We need this log initialisation done before we can
@@ -641,11 +669,11 @@ static voidxlog_cil_push_work ( struct work_struct *work) { -struct xfs_cil *cil = - container_of(work, struct xfs_cil, xc_push_work); +struct xfs_cil_ctx *ctx = + container_of(work, struct xfs_cil_ctx, push_work); +struct xfs_cil *cil = ctx->cil;struct xlog *log = cil->xc_log;struct xfs_log_vec *lv; -struct xfs_cil_ctx *ctx;struct xfs_cil_ctx *new_ctx;struct xlog_in_core *commit_iclog;struct xlog_ticket *tic;
@@ -660,11 +688,10 @@ xlog_cil_push_work(DECLARE_COMPLETION_ONSTACK (bdev_flush);bool push_commit_stable; - new_ctx =kmem_zalloc (sizeof (*new_ctx),KM_NOFS ); + new_ctx =xlog_cil_ctx_alloc (); new_ctx->ticket =xlog_cil_ticket_alloc (log);down_write (&cil->xc_ctx_lock); - ctx = cil->xc_ctx;spin_lock (&cil->xc_push_lock); push_seq = cil->xc_push_seq;
@@ -696,7 +723,7 @@ xlog_cil_push_work(/* check for a previously pushed sequence */ -if (push_seq < cil->xc_ctx->sequence) { +if (push_seq < ctx->sequence) {spin_unlock (&cil->xc_push_lock); goto out_skip; }
@@ -761,19 +788,7 @@ xlog_cil_push_work( }/* -* initialise the new context and attach it to the CIL. Then attach -* the current context to the CIL committing list so it can be found -* during log forces to extract the commit lsn of the sequence that -* needs to be forced. -*/ -INIT_LIST_HEAD (&new_ctx->committing); -INIT_LIST_HEAD (&new_ctx->busy_extents); - new_ctx->sequence = ctx->sequence +1 ; - new_ctx->cil = cil; - cil->xc_ctx = new_ctx; - -/* -* The switch is now done, so we can drop the context lock and move out +* Switch the contexts so we can drop the context lock and move out * of a shared context. We can't just go straight to the commit record, * though - we need to synchronise with previous and future commits so * that the commit records are correctly ordered in the log to ensure
@@ -798,7 +813,7 @@ xlog_cil_push_work(* deferencing a freed context pointer. */ spin_lock (&cil->xc_push_lock); - cil->xc_current_sequence = new_ctx->sequence; +xlog_cil_ctx_switch (cil, new_ctx);spin_unlock (&cil->xc_push_lock);up_write (&cil->xc_ctx_lock);
@@ -970,7 +985,7 @@ xlog_cil_push_background(spin_lock (&cil->xc_push_lock);if (cil->xc_push_seq < cil->xc_current_sequence) { cil->xc_push_seq = cil->xc_current_sequence; -queue_work (log->l_mp->m_cil_workqueue, &cil->xc_push_work); +queue_work (log->l_mp->m_cil_workqueue, &cil->xc_ctx->push_work); }/*
@@ -1036,7 +1051,7 @@ xlog_cil_push_now(/* start on any pending background push to minimise wait time on it */ if (!async) -flush_work (&cil->xc_push_work); +flush_workqueue (log->l_mp->m_cil_workqueue);/* * If the CIL is empty or we've already pushed the sequence then
@@ -1050,7 +1065,7 @@ xlog_cil_push_now( cil->xc_push_seq = push_seq; cil->xc_push_commit_stable = async; -queue_work (log->l_mp->m_cil_workqueue, &cil->xc_push_work); +queue_work (log->l_mp->m_cil_workqueue, &cil->xc_ctx->push_work);spin_unlock (&cil->xc_push_lock); }
@@ -1290,13 +1305,6 @@ xlog_cil_init(if (!cil)return -ENOMEM ; - ctx =kmem_zalloc (sizeof (*ctx),KM_MAYFAIL ); -if (!ctx) { -kmem_free (cil); -return -ENOMEM ; - } - -INIT_WORK (&cil->xc_push_work, xlog_cil_push_work);INIT_LIST_HEAD (&cil->xc_cil);INIT_LIST_HEAD (&cil->xc_committing);spin_lock_init (&cil->xc_cil_lock);
@@ -1304,16 +1312,12 @@ xlog_cil_init(init_waitqueue_head (&cil->xc_push_wait);init_rwsem (&cil->xc_ctx_lock);init_waitqueue_head (&cil->xc_commit_wait); - -INIT_LIST_HEAD (&ctx->committing); -INIT_LIST_HEAD (&ctx->busy_extents); - ctx->sequence =1 ; - ctx->cil = cil; - cil->xc_ctx = ctx; - cil->xc_current_sequence = ctx->sequence; - cil->xc_log = log; log->l_cilp = cil; + + ctx =xlog_cil_ctx_alloc (); +xlog_cil_ctx_switch (cil, ctx); +return 0 ; }
fs/xfs/xfs_log_priv.h +1 -1
@@ -245,6 +245,7 @@ struct xfs_cil_ctx {struct list_head iclog_entry;struct list_head committing;/* ctx committing list */ struct work_struct discard_endio_work; +struct work_struct push_work; };/*
@@ -277,7 +278,6 @@ struct xfs_cil {struct list_head xc_committing;wait_queue_head_t xc_commit_wait;xfs_csn_t xc_current_sequence; -struct work_struct xc_push_work;wait_queue_head_t xc_push_wait;/* background push throttle */ }____cacheline_aligned_in_smp ;
fs/xfs/xfs_super.c +5 -1
@@ -501,9 +501,13 @@ xfs_init_mount_workqueues(if (!mp->m_unwritten_workqueue) goto out_destroy_buf; +/* +* Limit the CIL pipeline depth to 4 concurrent works to bound the +* concurrency the log spinlocks will be exposed to. +*/ mp->m_cil_workqueue =alloc_workqueue ("xfs-cil/%s" ,XFS_WQFLAGS (WQ_FREEZABLE |WQ_MEM_RECLAIM |WQ_UNBOUND ), -0 , mp->m_super->s_id); +4 , mp->m_super->s_id);if (!mp->m_cil_workqueue) goto out_destroy_unwritten;
Signature · 1 line
2.31.1
Re: [PATCH 11/39] xfs: CIL work is serialised, not pipelined
Quoted 33 lines · Dave Chinner
From: Dave Chinner <[email protected]>
Because we use a single work structure attached to the CIL rather than the CIL context, we can only queue a single work item at a time. This results in the CIL being single threaded and limits performance when it becomes CPU bound.
The design of the CIL is that it is pipelined and multiple commits can be running concurrently, but the way the work is currently implemented means that it is not pipelining as it was intended. The critical work to switch the CIL context can take a few milliseconds to run, but the rest of the CIL context flush can take hundreds of milliseconds to complete. The context switching is the serialisation point of the CIL, once the context has been switched the rest of the context push can run asynchrnously with all other context pushes.
Hence we can move the work to the CIL context so that we can run multiple CIL pushes at the same time and spread the majority of the work out over multiple CPUs. We can keep the per-cpu CIL commit state on the CIL rather than the context, because the context is pinned to the CIL until the switch is done and we aggregate and drain the per-cpu state held on the CIL during the context switch.
However, because we no longer serialise the CIL work, we can have effectively unlimited CIL pushes in progress. We don't want to do this - not only does it create contention on the iclogs and the state machine locks, we can run the log right out of space with outstanding pushes. Instead, limit the work concurrency to 4 concurrent works being processed at a time. THis is enough concurrency to remove the CIL from being a CPU bound bottleneck but not enough to create new contention points or unbound concurrency issues.
Quoted 1 line
Signed-off-by Dave Chinner <[email protected]>
Heh woo :)
--D
Quoted 216 lines
--- fs/xfs/xfs_log_cil.c | 80 +++++++++++++++++++++++-------------------- fs/xfs/xfs_log_priv.h | 2 +- fs/xfs/xfs_super.c | 6 +++- 3 files changed, 48 insertions(+), 40 deletions(-)fs/xfs/xfs_log_cil.c +42
-38@@ -47,6 +47,34 @@ xlog_cil_ticket_alloc(return tic; } +/* +* Unavoidable forward declaration - xlog_cil_push_work() calls +* xlog_cil_ctx_alloc() itself. +*/ +static void xlog_cil_push_work (struct work_struct *work); + +static struct xfs_cil_ctx * +xlog_cil_ctx_alloc (void ) +{ +struct xfs_cil_ctx *ctx; + + ctx =kmem_zalloc (sizeof (*ctx),KM_NOFS ); +INIT_LIST_HEAD (&ctx->committing); +INIT_LIST_HEAD (&ctx->busy_extents); +INIT_WORK (&ctx->push_work, xlog_cil_push_work); +return ctx; +} + +static void +xlog_cil_ctx_switch ( +struct xfs_cil *cil, +struct xfs_cil_ctx *ctx) +{ + ctx->sequence = ++cil->xc_current_sequence; + ctx->cil = cil; + cil->xc_ctx = ctx; +} +/* * After the first stage of log recovery is done, we know where the head and * tail of the log are. We need this log initialisation done before we can @@ -641,11 +669,11 @@ static voidxlog_cil_push_work ( struct work_struct *work) { -struct xfs_cil *cil = - container_of(work, struct xfs_cil, xc_push_work); +struct xfs_cil_ctx *ctx = + container_of(work, struct xfs_cil_ctx, push_work); +struct xfs_cil *cil = ctx->cil;struct xlog *log = cil->xc_log;struct xfs_log_vec *lv; -struct xfs_cil_ctx *ctx;struct xfs_cil_ctx *new_ctx;struct xlog_in_core *commit_iclog;struct xlog_ticket *tic;@@ -660,11 +688,10 @@ xlog_cil_push_work(DECLARE_COMPLETION_ONSTACK (bdev_flush);bool push_commit_stable; - new_ctx =kmem_zalloc (sizeof (*new_ctx),KM_NOFS ); + new_ctx =xlog_cil_ctx_alloc (); new_ctx->ticket =xlog_cil_ticket_alloc (log);down_write (&cil->xc_ctx_lock); - ctx = cil->xc_ctx;spin_lock (&cil->xc_push_lock); push_seq = cil->xc_push_seq;@@ -696,7 +723,7 @@ xlog_cil_push_work(/* check for a previously pushed sequence */ -if (push_seq < cil->xc_ctx->sequence) { +if (push_seq < ctx->sequence) {spin_unlock (&cil->xc_push_lock); goto out_skip; }@@ -761,19 +788,7 @@ xlog_cil_push_work( }/* -* initialise the new context and attach it to the CIL. Then attach -* the current context to the CIL committing list so it can be found -* during log forces to extract the commit lsn of the sequence that -* needs to be forced. -*/ -INIT_LIST_HEAD (&new_ctx->committing); -INIT_LIST_HEAD (&new_ctx->busy_extents); - new_ctx->sequence = ctx->sequence +1 ; - new_ctx->cil = cil; - cil->xc_ctx = new_ctx; - -/* -* The switch is now done, so we can drop the context lock and move out +* Switch the contexts so we can drop the context lock and move out * of a shared context. We can't just go straight to the commit record, * though - we need to synchronise with previous and future commits so * that the commit records are correctly ordered in the log to ensure @@ -798,7 +813,7 @@ xlog_cil_push_work(* deferencing a freed context pointer. */ spin_lock (&cil->xc_push_lock); - cil->xc_current_sequence = new_ctx->sequence; +xlog_cil_ctx_switch (cil, new_ctx);spin_unlock (&cil->xc_push_lock);up_write (&cil->xc_ctx_lock);@@ -970,7 +985,7 @@ xlog_cil_push_background(spin_lock (&cil->xc_push_lock);if (cil->xc_push_seq < cil->xc_current_sequence) { cil->xc_push_seq = cil->xc_current_sequence; -queue_work (log->l_mp->m_cil_workqueue, &cil->xc_push_work); +queue_work (log->l_mp->m_cil_workqueue, &cil->xc_ctx->push_work); }/* @@ -1036,7 +1051,7 @@ xlog_cil_push_now(/* start on any pending background push to minimise wait time on it */ if (!async) -flush_work (&cil->xc_push_work); +flush_workqueue (log->l_mp->m_cil_workqueue);/* * If the CIL is empty or we've already pushed the sequence then @@ -1050,7 +1065,7 @@ xlog_cil_push_now( cil->xc_push_seq = push_seq; cil->xc_push_commit_stable = async; -queue_work (log->l_mp->m_cil_workqueue, &cil->xc_push_work); +queue_work (log->l_mp->m_cil_workqueue, &cil->xc_ctx->push_work);spin_unlock (&cil->xc_push_lock); }@@ -1290,13 +1305,6 @@ xlog_cil_init(if (!cil)return -ENOMEM ; - ctx =kmem_zalloc (sizeof (*ctx),KM_MAYFAIL ); -if (!ctx) { -kmem_free (cil); -return -ENOMEM ; - } - -INIT_WORK (&cil->xc_push_work, xlog_cil_push_work);INIT_LIST_HEAD (&cil->xc_cil);INIT_LIST_HEAD (&cil->xc_committing);spin_lock_init (&cil->xc_cil_lock);@@ -1304,16 +1312,12 @@ xlog_cil_init(init_waitqueue_head (&cil->xc_push_wait);init_rwsem (&cil->xc_ctx_lock);init_waitqueue_head (&cil->xc_commit_wait); - -INIT_LIST_HEAD (&ctx->committing); -INIT_LIST_HEAD (&ctx->busy_extents); - ctx->sequence =1 ; - ctx->cil = cil; - cil->xc_ctx = ctx; - cil->xc_current_sequence = ctx->sequence; - cil->xc_log = log; log->l_cilp = cil; + + ctx =xlog_cil_ctx_alloc (); +xlog_cil_ctx_switch (cil, ctx); +return 0 ; }fs/xfs/xfs_log_priv.h +1
-1@@ -245,6 +245,7 @@ struct xfs_cil_ctx {struct list_head iclog_entry;struct list_head committing;/* ctx committing list */ struct work_struct discard_endio_work; +struct work_struct push_work; };/* @@ -277,7 +278,6 @@ struct xfs_cil {struct list_head xc_committing;wait_queue_head_t xc_commit_wait;xfs_csn_t xc_current_sequence; -struct work_struct xc_push_work;wait_queue_head_t xc_push_wait;/* background push throttle */ }____cacheline_aligned_in_smp ;fs/xfs/xfs_super.c +5
-1@@ -501,9 +501,13 @@ xfs_init_mount_workqueues(if (!mp->m_unwritten_workqueue) goto out_destroy_buf; +/* +* Limit the CIL pipeline depth to 4 concurrent works to bound the +* concurrency the log spinlocks will be exposed to. +*/ mp->m_cil_workqueue =alloc_workqueue ("xfs-cil/%s" ,XFS_WQFLAGS (WQ_FREEZABLE |WQ_MEM_RECLAIM |WQ_UNBOUND ), -0 , mp->m_super->s_id); +4 , mp->m_super->s_id);if (!mp->m_cil_workqueue) goto out_destroy_unwritten;Signature · 1 line
2.31.1
PATCH v4 12/39 xfs: factor out the CIL transaction header building
In reply to Dave ChinnerFrom: Dave Chinner <[email protected]>
It is static code deep in the middle of the CIL push logic. Factor it out into a helper so that it is clear and easy to modify separately.
--- fs/xfs/xfs_log_cil.c | 71 +++++++++++++++++++++++++++++--------------- 1 file changed, 47 insertions(+), 24 deletions(-)
fs/xfs/xfs_log_cil.c +47 -24
@@ -651,6 +651,41 @@ xlog_cil_process_committed( } } +struct xlog_cil_trans_hdr { +struct xfs_trans_header thdr; +struct xfs_log_iovec lhdr; +}; + +/* +* Build a checkpoint transaction header to begin the journal transaction. We +* need to account for the space used by the transaction header here as it is +* not accounted for in xlog_write(). +*/ +static void +xlog_cil_build_trans_hdr ( +struct xfs_cil_ctx *ctx, +struct xlog_cil_trans_hdr *hdr, +struct xfs_log_vec *lvhdr, +int num_iovecs) +{ +struct xlog_ticket *tic = ctx->ticket; + +memset (hdr,0 ,sizeof (*hdr)); + + hdr->thdr.th_magic =XFS_TRANS_HEADER_MAGIC ; + hdr->thdr.th_type =XFS_TRANS_CHECKPOINT ; + hdr->thdr.th_tid = tic->t_tid; + hdr->thdr.th_num_items = num_iovecs; + hdr->lhdr.i_addr = &hdr->thdr; + hdr->lhdr.i_len =sizeof (xfs_trans_header_t); + hdr->lhdr.i_type =XLOG_REG_TYPE_TRANSHDR ; + tic->t_curr_res -= hdr->lhdr.i_len +sizeof (struct xlog_op_header ); + + lvhdr->lv_niovecs =1 ; + lvhdr->lv_iovecp = &hdr->lhdr; + lvhdr->lv_next = ctx->lv_chain; +} +/* * Push the Committed Item List to the log. *
@@ -676,11 +711,9 @@ xlog_cil_push_work(struct xfs_log_vec *lv;struct xfs_cil_ctx *new_ctx;struct xlog_in_core *commit_iclog; -struct xlog_ticket *tic;int num_iovecs;int error =0 ; -struct xfs_trans_header thdr; -struct xfs_log_iovec lhdr; +struct xlog_cil_trans_hdr thdr;struct xfs_log_vec lvhdr = {NULL };xfs_lsn_t commit_lsn;xfs_lsn_t push_seq;
@@ -821,24 +854,8 @@ xlog_cil_push_work(* Build a checkpoint transaction header and write it to the log to * begin the transaction. We need to account for the space used by the * transaction header here as it is not accounted for in xlog_write(). -* -* The LSN we need to pass to the log items on transaction commit is -* the LSN reported by the first log vector write. If we use the commit -* record lsn then we can move the tail beyond the grant write head. */ - tic = ctx->ticket; - thdr.th_magic =XFS_TRANS_HEADER_MAGIC ; - thdr.th_type =XFS_TRANS_CHECKPOINT ; - thdr.th_tid = tic->t_tid; - thdr.th_num_items = num_iovecs; - lhdr.i_addr = &thdr; - lhdr.i_len =sizeof (xfs_trans_header_t); - lhdr.i_type =XLOG_REG_TYPE_TRANSHDR ; - tic->t_curr_res -= lhdr.i_len +sizeof (xlog_op_header_t); - - lvhdr.lv_niovecs =1 ; - lvhdr.lv_iovecp = &lhdr; - lvhdr.lv_next = ctx->lv_chain; +xlog_cil_build_trans_hdr (ctx, &thdr, &lvhdr, num_iovecs);/* * Before we format and submit the first iclog, we have to ensure that
@@ -846,7 +863,13 @@ xlog_cil_push_work(*/ wait_for_completion (&bdev_flush); - error =xlog_write (log, &lvhdr, tic, &ctx->start_lsn,NULL , +/* +* The LSN we need to pass to the log items on transaction commit is the +* LSN reported by the first log vector write, not the commit lsn. If we +* use the commit record lsn then we can move the tail beyond the grant +* write head. +*/ + error =xlog_write (log, &lvhdr, ctx->ticket, &ctx->start_lsn,NULL ,XLOG_START_TRANS );if (error) goto out_abort_free_ticket;
@@ -885,11 +908,11 @@ xlog_cil_push_work( }spin_unlock (&cil->xc_push_lock); - error =xlog_commit_record (log, tic, &commit_iclog, &commit_lsn); + error =xlog_commit_record (log, ctx->ticket, &commit_iclog, &commit_lsn);if (error) goto out_abort_free_ticket; -xfs_log_ticket_ungrant (log, tic); +xfs_log_ticket_ungrant (log, ctx->ticket);spin_lock (&commit_iclog->ic_callback_lock);if (commit_iclog->ic_state ==XLOG_STATE_IOERROR ) {
@@ -948,7 +971,7 @@ xlog_cil_push_work(return ; out_abort_free_ticket: -xfs_log_ticket_ungrant (log, tic); +xfs_log_ticket_ungrant (log, ctx->ticket); out_abort:ASSERT (XLOG_FORCED_SHUTDOWN (log));xlog_cil_committed (ctx);
Signature · 1 line
2.31.1
PATCH v4 13/39 xfs: only CIL pushes require a start record
In reply to Dave ChinnerFrom: Dave Chinner <[email protected]>
So move the one-off start record writing in xlog_write() out into the static header that the CIL push builds to write into the log initially. This simplifes the xlog_write() logic a lot.
pahole on x86-64 confirms that the xlog_cil_trans_hdr is correctly 32 bit aligned and packed for copying the log op and transaction headers directly into the log as a single log region copy.
struct xlog_cil_trans_hdr {
struct xlog_op_header oph[2]; /* 0 24 */
struct xfs_trans_header thdr; /* 24 16 */
struct xfs_log_iovec lhdr[2]; /* 40 32 */ /* size: 72, cachelines: 2, members: 3 */
/* last cacheline: 8 bytes */
};A wart is needed to handle the fact that length of the region the opheader points to doesn't include the opheader length. hence if we embed the opheader, we have to substract the opheader length from the length written into the opheader by the generic copying code. This will eventually go away when everything is converted to embedded opheaders.
--- fs/xfs/xfs_log.c | 90 ++++++++++++++++++++++---------------------- fs/xfs/xfs_log_cil.c | 44 ++++++++++++++++++---- 2 files changed, 81 insertions(+), 53 deletions(-)
fs/xfs/xfs_log.c +45 -45
@@ -2113,9 +2113,9 @@ xlog_print_trans( }/* -* Calculate the potential space needed by the log vector. We may need a start -* record, and each region gets its own struct xlog_op_header and may need to be -* double word aligned. +* Calculate the potential space needed by the log vector. If this is a start +* transaction, the caller has already accounted for both opheaders in the start +* transaction, so we don't need to account for them here. */ static int xlog_write_calc_vec_length(
@@ -2128,9 +2128,6 @@ xlog_write_calc_vec_length(int len =0 ;int i; -if (optype &XLOG_START_TRANS ) - headers++; -for (lv = log_vector; lv; lv = lv->lv_next) {/* we don't write ordered log vectors */ if (lv->lv_buf_len ==XFS_LOG_VEC_ORDERED )
@@ -2146,24 +2143,20 @@ xlog_write_calc_vec_length( } } +/* Don't account for regions with embedded ophdrs */ +if (optype && headers >0 ) { +if (optype &XLOG_START_TRANS ) { +ASSERT (headers >=2 ); + headers -=2 ; + } + } + ticket->t_res_num_ophdrs += headers; len += headers *sizeof (struct xlog_op_header );return len; } -static void -xlog_write_start_rec ( -struct xlog_op_header *ophdr, -struct xlog_ticket *ticket) -{ - ophdr->oh_tid =cpu_to_be32 (ticket->t_tid); - ophdr->oh_clientid = ticket->t_clientid; - ophdr->oh_len =0 ; - ophdr->oh_flags =XLOG_START_TRANS ; - ophdr->oh_res2 =0 ; -} -static xlog_op_header_t * xlog_write_setup_ophdr( struct xlog *log,
@@ -2368,9 +2361,11 @@ xlog_write(* If this is a commit or unmount transaction, we don't need a start * record to be written. We do, however, have to account for the * commit or unmount header that gets written. Hence we always have -* to account for an extra xlog_op_header here. +* to account for an extra xlog_op_header here for commit and unmount +* records. */ - ticket->t_curr_res -=sizeof (struct xlog_op_header ); +if (optype & (XLOG_COMMIT_TRANS |XLOG_UNMOUNT_TRANS )) + ticket->t_curr_res -=sizeof (struct xlog_op_header );if (ticket->t_curr_res <0 ) {xfs_alert_tag (log->l_mp,XFS_PTAG_LOGRES ,"ctx ticket reservation ran out. Need to up reservation" );
@@ -2407,7 +2402,7 @@ xlog_write(int copy_len;int copy_off;bool ordered = false; -bool wrote_start_rec = false; +bool added_ophdr = false;/* ordered log vectors have no regions to write */ if (lv->lv_buf_len ==XFS_LOG_VEC_ORDERED ) {
@@ -2421,25 +2416,24 @@ xlog_write(ASSERT ((unsigned long )ptr %sizeof (int32_t ) ==0 );/* -* Before we start formatting log vectors, we need to -* write a start record. Only do this for the first -* iclog we write to. +* The XLOG_START_TRANS has embedded ophdrs for the +* start record and transaction header. They will always +* be the first two regions in the lv chain. */ if (optype &XLOG_START_TRANS ) { -xlog_write_start_rec (ptr, ticket); -xlog_write_adv_cnt (&ptr, &len, &log_offset, -sizeof (struct xlog_op_header )); - optype &= ~XLOG_START_TRANS ; - wrote_start_rec = true; - } - - ophdr =xlog_write_setup_ophdr (log, ptr, ticket, optype); -if (!ophdr) -return -EIO ; + ophdr = reg->i_addr; +if (index) + optype &= ~XLOG_START_TRANS ; + }else { + ophdr =xlog_write_setup_ophdr (log, ptr, + ticket, optype); +if (!ophdr) +return -EIO ; -xlog_write_adv_cnt (&ptr, &len, &log_offset, +xlog_write_adv_cnt (&ptr, &len, &log_offset,sizeof (struct xlog_op_header )); - + added_ophdr = true; + } len += xlog_write_setup_copy(ticket, ophdr, iclog->ic_size-log_offset, reg->i_len,
@@ -2448,13 +2442,22 @@ xlog_write( &partial_copy_len);xlog_verify_dest_ptr (log, ptr); + +/* +* Wart: need to update length in embedded ophdr not +* to include it's own length. +*/ +if (!added_ophdr) { + ophdr->oh_len =cpu_to_be32 (copy_len - +sizeof (struct xlog_op_header )); + }/* * Copy region. * -* Unmount records just log an opheader, so can have -* empty payloads with no data region to copy. Hence we -* only copy the payload if the vector says it has data -* to copy. +* Commit and unmount records just log an opheader, so +* we can have empty payloads with no data region to +* copy. Hence we only copy the payload if the vector +* says it has data to copy. */ ASSERT (copy_len >=0 );if (copy_len >0 ) {
@@ -2462,12 +2465,9 @@ xlog_write(xlog_write_adv_cnt (&ptr, &len, &log_offset, copy_len); } - copy_len +=sizeof (struct xlog_op_header ); - record_cnt++; -if (wrote_start_rec) { +if (added_ophdr) copy_len +=sizeof (struct xlog_op_header ); - record_cnt++; - } + record_cnt++; data_cnt += contwr ? copy_len :0 ; error = xlog_write_copy_finish(log, iclog, optype,
fs/xfs/xfs_log_cil.c +36 -8
70 lines not on this page · raw message
Signature · 1 line
2.31.1
PATCH v4 14/39 xfs: embed the xlog_op_header in the unmount record
In reply to Dave ChinnerFrom: Dave Chinner <[email protected]>
Remove another case where xlog_write() has to prepend an opheader to a log transaction. The unmount record + ophdr is smaller than the minimum amount of space guaranteed to be free in an iclog (2 * sizeof(ophdr)) and so we don't have to care about an unmount record being split across 2 iclogs.
--- fs/xfs/xfs_log.c | 39 ++++++++++++++++++++++++++++----------- 1 file changed, 28 insertions(+), 11 deletions(-)
fs/xfs/xfs_log.c +28 -11
@@ -800,12 +800,22 @@ xlog_write_unmount_record( struct xlog *log, struct xlog_ticket *ticket) { - struct xfs_unmount_log_format ulf = { - .magic =XLOG_UNMOUNT_TYPE , +struct { +struct xlog_op_header ophdr; +struct xfs_unmount_log_format ulf; + } unmount_rec = { + .ophdr = { + .oh_clientid =XFS_LOG , + .oh_tid =cpu_to_be32 (ticket->t_tid), + .oh_flags =XLOG_UNMOUNT_TRANS , + }, + .ulf = { + .magic =XLOG_UNMOUNT_TYPE , + }, };struct xfs_log_iovec reg = { - .i_addr = &ulf, - .i_len =sizeof (ulf), + .i_addr = &unmount_rec, + .i_len =sizeof (unmount_rec), .i_type =XLOG_REG_TYPE_UNMOUNT , };struct xfs_log_vec vec = {
@@ -813,8 +823,12 @@ xlog_write_unmount_record( .lv_iovecp = ®, }; +BUILD_BUG_ON ((sizeof (struct xlog_op_header ) + +sizeof (struct xfs_unmount_log_format )) != +sizeof (unmount_rec)); +/* account for space used by record data */ - ticket->t_curr_res -=sizeof (ulf); + ticket->t_curr_res -=sizeof (unmount_rec);/* * For external log devices, we need to flush the data device cache
@@ -2145,6 +2159,8 @@ xlog_write_calc_vec_length(/* Don't account for regions with embedded ophdrs */ if (optype && headers >0 ) { +if (optype &XLOG_UNMOUNT_TRANS ) + headers--;if (optype &XLOG_START_TRANS ) {ASSERT (headers >=2 ); headers -=2 ;
@@ -2359,12 +2375,11 @@ xlog_write(/* * If this is a commit or unmount transaction, we don't need a start -* record to be written. We do, however, have to account for the -* commit or unmount header that gets written. Hence we always have -* to account for an extra xlog_op_header here for commit and unmount -* records. +* record to be written. We do, however, have to account for the commit +* header that gets written. Hence we always have to account for an +* extra xlog_op_header here for commit records. */ -if (optype & (XLOG_COMMIT_TRANS |XLOG_UNMOUNT_TRANS )) +if (optype &XLOG_COMMIT_TRANS ) ticket->t_curr_res -=sizeof (struct xlog_op_header );if (ticket->t_curr_res <0 ) { xfs_alert_tag(log->l_mp, XFS_PTAG_LOGRES,
@@ -2424,6 +2439,8 @@ xlog_write( ophdr = reg->i_addr;if (index) optype &= ~XLOG_START_TRANS ; + }else if (optype &XLOG_UNMOUNT_TRANS ) { + ophdr = reg->i_addr; }else { ophdr =xlog_write_setup_ophdr (log, ptr, ticket, optype);
@@ -2454,7 +2471,7 @@ xlog_write(/* * Copy region. * -* Commit and unmount records just log an opheader, so +* Commit records just log an opheader, so * we can have empty payloads with no data region to * copy. Hence we only copy the payload if the vector * says it has data to copy.
Signature · 1 line
2.31.1
Re: [PATCH 14/39] xfs: embed the xlog_op_header in the unmount record
Quoted 10 lines · Dave Chinner
From: Dave Chinner <[email protected]>
Remove another case where
xlog_write()has to prepend an opheader to a log transaction. The unmount record + ophdr is smaller than the minimum amount of space guaranteed to be free in an iclog (2 * sizeof(ophdr)) and so we don't have to care about an unmount record being split across 2 iclogs.Signed-off-by Dave Chinner <[email protected]>Reviewed-by Christoph Hellwig <[email protected]>
I love me some BUILD_BUG_ON, so
--D
Quoted 96 lines
--- fs/xfs/xfs_log.c | 39 ++++++++++++++++++++++++++++----------- 1 file changed, 28 insertions(+), 11 deletions(-)fs/xfs/xfs_log.c +28
-11@@ -800,12 +800,22 @@ xlog_write_unmount_record( struct xlog *log, struct xlog_ticket *ticket) { - struct xfs_unmount_log_format ulf = { - .magic =XLOG_UNMOUNT_TYPE , +struct { +struct xlog_op_header ophdr; +struct xfs_unmount_log_format ulf; + } unmount_rec = { + .ophdr = { + .oh_clientid =XFS_LOG , + .oh_tid =cpu_to_be32 (ticket->t_tid), + .oh_flags =XLOG_UNMOUNT_TRANS , + }, + .ulf = { + .magic =XLOG_UNMOUNT_TYPE , + }, };struct xfs_log_iovec reg = { - .i_addr = &ulf, - .i_len =sizeof (ulf), + .i_addr = &unmount_rec, + .i_len =sizeof (unmount_rec), .i_type =XLOG_REG_TYPE_UNMOUNT , };struct xfs_log_vec vec = {@@ -813,8 +823,12 @@ xlog_write_unmount_record( .lv_iovecp = ®, }; +BUILD_BUG_ON ((sizeof (struct xlog_op_header ) + +sizeof (struct xfs_unmount_log_format )) != +sizeof (unmount_rec)); +/* account for space used by record data */ - ticket->t_curr_res -=sizeof (ulf); + ticket->t_curr_res -=sizeof (unmount_rec);/* * For external log devices, we need to flush the data device cache @@ -2145,6 +2159,8 @@ xlog_write_calc_vec_length(/* Don't account for regions with embedded ophdrs */ if (optype && headers >0 ) { +if (optype &XLOG_UNMOUNT_TRANS ) + headers--;if (optype &XLOG_START_TRANS ) {ASSERT (headers >=2 ); headers -=2 ;@@ -2359,12 +2375,11 @@ xlog_write(/* * If this is a commit or unmount transaction, we don't need a start -* record to be written. We do, however, have to account for the -* commit or unmount header that gets written. Hence we always have -* to account for an extra xlog_op_header here for commit and unmount -* records. +* record to be written. We do, however, have to account for the commit +* header that gets written. Hence we always have to account for an +* extra xlog_op_header here for commit records. */ -if (optype & (XLOG_COMMIT_TRANS |XLOG_UNMOUNT_TRANS )) +if (optype &XLOG_COMMIT_TRANS ) ticket->t_curr_res -=sizeof (struct xlog_op_header );if (ticket->t_curr_res <0 ) { xfs_alert_tag(log->l_mp, XFS_PTAG_LOGRES,@@ -2424,6 +2439,8 @@ xlog_write( ophdr = reg->i_addr;if (index) optype &= ~XLOG_START_TRANS ; + }else if (optype &XLOG_UNMOUNT_TRANS ) { + ophdr = reg->i_addr; }else { ophdr =xlog_write_setup_ophdr (log, ptr, ticket, optype);@@ -2454,7 +2471,7 @@ xlog_write(/* * Copy region. * -* Commit and unmount records just log an opheader, so +* Commit records just log an opheader, so * we can have empty payloads with no data region to * copy. Hence we only copy the payload if the vector * says it has data to copy. Signature · 1 line
2.31.1
PATCH v4 15/39 xfs: embed the xlog_op_header in the commit record
In reply to Dave ChinnerRemove the final case where xlog_write() has to prepend an opheader to a log transaction. Similar to the start record, the commit record is just an empty opheader with a XLOG_COMMIT_TRANS type, so we can just make this the payload for the region being passed to xlog_write() and remove the special handling in xlog_write() for the commit record.
--- fs/xfs/xfs_log.c | 33 +++++++++++++++------------------ 1 file changed, 15 insertions(+), 18 deletions(-)
fs/xfs/xfs_log.c +15 -18
@@ -1540,9 +1540,14 @@ xlog_commit_record(struct xlog_in_core **iclog, xfs_lsn_t *lsn) { + struct xlog_op_header ophdr = { + .oh_clientid =XFS_TRANSACTION , + .oh_tid =cpu_to_be32 (ticket->t_tid), + .oh_flags =XLOG_COMMIT_TRANS , + };struct xfs_log_iovec reg = { - .i_addr =NULL , - .i_len =0 , + .i_addr = &ophdr, + .i_len =sizeof (struct xlog_op_header ), .i_type =XLOG_REG_TYPE_COMMIT , }; struct xfs_log_vec vec = {
@@ -1554,6 +1559,8 @@ xlog_commit_record(if (XLOG_FORCED_SHUTDOWN (log))return -EIO ; +/* account for space used by record data */ + ticket->t_curr_res -= reg.i_len; error =xlog_write (log, &vec, ticket, lsn, iclog,XLOG_COMMIT_TRANS );if (error)xfs_force_shutdown (log->l_mp,SHUTDOWN_LOG_IO_ERROR );
@@ -2159,11 +2166,10 @@ xlog_write_calc_vec_length(/* Don't account for regions with embedded ophdrs */ if (optype && headers >0 ) { -if (optype &XLOG_UNMOUNT_TRANS ) - headers--; + headers--;if (optype &XLOG_START_TRANS ) { -ASSERT (headers >=2 ); - headers -=2 ; +ASSERT (headers >=1 ); + headers--; } }
@@ -2373,14 +2379,6 @@ xlog_write(int data_cnt =0 ;int error =0 ; -/* -* If this is a commit or unmount transaction, we don't need a start -* record to be written. We do, however, have to account for the commit -* header that gets written. Hence we always have to account for an -* extra xlog_op_header here for commit records. -*/ -if (optype &XLOG_COMMIT_TRANS ) - ticket->t_curr_res -=sizeof (struct xlog_op_header );if (ticket->t_curr_res <0 ) {xfs_alert_tag (log->l_mp,XFS_PTAG_LOGRES ,"ctx ticket reservation ran out. Need to up reservation" );
@@ -2433,14 +2431,13 @@ xlog_write(/* * The XLOG_START_TRANS has embedded ophdrs for the * start record and transaction header. They will always -* be the first two regions in the lv chain. +* be the first two regions in the lv chain. Commit and +* unmount records also have embedded ophdrs. */ -if (optype &XLOG_START_TRANS ) { +if (optype) { ophdr = reg->i_addr;if (index) optype &= ~XLOG_START_TRANS ; - }else if (optype &XLOG_UNMOUNT_TRANS ) { - ophdr = reg->i_addr; }else { ophdr =xlog_write_setup_ophdr (log, ptr, ticket, optype);
Signature · 1 line
2.31.1
PATCH v4 16/39 xfs: log tickets don't need log client id
In reply to Dave ChinnerFrom: Dave Chinner <[email protected]>
We currently set the log ticket client ID when we reserve a transaction. This client ID is only ever written to the log by a CIL checkpoint or unmount records, and so anything using a high level transaction allocated through xfs_trans_alloc() does not need a log ticket client ID to be set.
For the CIL checkpoint, the client ID written to the journal is always XFS_TRANSACTION, and for the unmount record it is always XFS_LOG, and nothing else writes to the log. All of these operations tell xlog_write() exactly what they need to write to the log (the optype) and build their own opheaders for start, commit and unmount records. Hence we no longer need to set the client id in either the log ticket or the xfs_trans.
--- fs/xfs/libxfs/xfs_log_format.h | 1 - fs/xfs/xfs_log.c | 47 ++++++---------------------------- fs/xfs/xfs_log.h | 16 +++++------- fs/xfs/xfs_log_cil.c | 2 +- fs/xfs/xfs_log_priv.h | 10 ++------ fs/xfs/xfs_trans.c | 6 ++--- 6 files changed, 19 insertions(+), 63 deletions(-)
fs/xfs/libxfs/xfs_log_format.h +0 -1
@@ -69,7 +69,6 @@ static inline uint xlog_get_cycle(char *ptr)/* Log Clients */ #define XFS_TRANSACTION 0x69 -#define XFS_VOLUME 0x2#define XFS_LOG 0xaa#define XLOG_UNMOUNT_TYPE 0x556e/* Un for Unmount */
fs/xfs/xfs_log.c +8 -39
100 lines not on this page · raw message
fs/xfs/xfs_log.h +6 -10
@@ -117,16 +117,12 @@ int xfs_log_mount_finish(struct xfs_mount *mp);void xfs_log_mount_cancel (struct xfs_mount *);xfs_lsn_t xlog_assign_tail_lsn (struct xfs_mount *mp);xfs_lsn_t xlog_assign_tail_lsn_locked (struct xfs_mount *mp); -void xfs_log_space_wake (struct xfs_mount *mp); -int xfs_log_reserve (struct xfs_mount *mp, -int length, -int count, -struct xlog_ticket **ticket, -uint8_t clientid, -bool permanent); -int xfs_log_regrant (struct xfs_mount *mp,struct xlog_ticket *tic); -void xfs_log_unmount (struct xfs_mount *mp); -int xfs_log_force_umount (struct xfs_mount *mp,int logerror); +void xfs_log_space_wake (struct xfs_mount *mp); +int xfs_log_reserve (struct xfs_mount *mp,int length,int count, +struct xlog_ticket **ticket,bool permanent); +int xfs_log_regrant (struct xfs_mount *mp,struct xlog_ticket *tic); +void xfs_log_unmount (struct xfs_mount *mp); +int xfs_log_force_umount (struct xfs_mount *mp,int logerror);bool xfs_log_writable (struct xfs_mount *mp);struct xlog_ticket *xfs_log_ticket_get (struct xlog_ticket *ticket);
fs/xfs/xfs_log_cil.c +1 -1
@@ -37,7 +37,7 @@ xlog_cil_ticket_alloc( {struct xlog_ticket *tic; - tic =xlog_ticket_alloc (log,0 ,1 ,XFS_TRANSACTION ,0 ); + tic =xlog_ticket_alloc (log,0 ,1 ,0 );/* * set the current reservation to zero so we know to steal the basic
fs/xfs/xfs_log_priv.h +2 -8
@@ -158,7 +158,6 @@ typedef struct xlog_ticket {int t_unit_res;/* unit reservation in bytes : 4 */ char t_ocnt;/* original count : 1 */ char t_cnt;/* current count : 1 */ -char t_clientid;/* who does this belong to; : 1 */ char t_flags;/* properties of reservation : 1 */ /* reservation array fields */
@@ -465,13 +464,8 @@ extern __le32 xlog_cksum(struct xlog *log, struct xlog_rec_header *rhead, char *dp, int size);extern kmem_zone_t *xfs_log_ticket_zone; -struct xlog_ticket * -xlog_ticket_alloc ( -struct xlog *log, -int unit_bytes, -int count, -char client, -bool permanent); +struct xlog_ticket *xlog_ticket_alloc (struct xlog *log,int unit_bytes, +int count,bool permanent);static inline void xlog_write_adv_cnt (void **ptr,int *len,int *off,size_t bytes)
fs/xfs/xfs_trans.c +2 -4
@@ -194,11 +194,9 @@ xfs_trans_reserve(ASSERT (resp->tr_logflags &XFS_TRANS_PERM_LOG_RES ); error =xfs_log_regrant (mp, tp->t_ticket); }else { - error =xfs_log_reserve (mp, - resp->tr_logres, + error =xfs_log_reserve (mp, resp->tr_logres, resp->tr_logcount, - &tp->t_ticket,XFS_TRANSACTION , - permanent); + &tp->t_ticket, permanent); }if (error)
Signature · 1 line
2.31.1
Re: [PATCH 16/39] xfs: log tickets don't need log client id
Quoted 16 lines · Dave Chinner
From: Dave Chinner <[email protected]>
We currently set the log ticket client ID when we reserve a transaction. This client ID is only ever written to the log by a CIL checkpoint or unmount records, and so anything using a high level transaction allocated through
xfs_trans_alloc()does not need a log ticket client ID to be set.For the CIL checkpoint, the client ID written to the journal is always
XFS_TRANSACTION, and for the unmount record it is alwaysXFS_LOG, and nothing else writes to the log. All of these operations tellxlog_write()exactly what they need to write to the log (the optype) and build their own opheaders for start, commit and unmount records. Hence we no longer need to set the client id in either the log ticket or thexfs_trans.
Quoted 3 lines
Signed-off-by Dave Chinner <[email protected]>Reviewed-by Christoph Hellwig <[email protected]>Reviewed-by Brian Foster <[email protected]>
Yay for removing never-used macros,
--D
Quoted 221 lines
--- fs/xfs/libxfs/xfs_log_format.h | 1 - fs/xfs/xfs_log.c | 47 ++++++---------------------------- fs/xfs/xfs_log.h | 16 +++++------- fs/xfs/xfs_log_cil.c | 2 +- fs/xfs/xfs_log_priv.h | 10 ++------ fs/xfs/xfs_trans.c | 6 ++--- 6 files changed, 19 insertions(+), 63 deletions(-)fs/xfs/libxfs/xfs_log_format.h +0
-1@@ -69,7 +69,6 @@ static inline uint xlog_get_cycle(char *ptr)/* Log Clients */ #define XFS_TRANSACTION 0x69 -#define XFS_VOLUME 0x2#define XFS_LOG 0xaa#define XLOG_UNMOUNT_TYPE 0x556e/* Un for Unmount */ fs/xfs/xfs_log.c +8
-39@@ -433,10 +433,9 @@ xfs_log_regrant(int xfs_log_reserve (struct xfs_mount *mp, -int unit_bytes, -int cnt, +int unit_bytes, +int cnt,struct xlog_ticket **ticp, -uint8_t client,bool permanent) {struct xlog *log = mp->m_log;@@ -444,15 +443,13 @@ xfs_log_reserve(int need_bytes;int error =0 ; -ASSERT (client ==XFS_TRANSACTION || client ==XFS_LOG ); -if (XLOG_FORCED_SHUTDOWN (log))return -EIO ;XFS_STATS_INC (mp, xs_try_logspace);ASSERT (*ticp ==NULL ); - tic =xlog_ticket_alloc (log, unit_bytes, cnt, client, permanent); + tic =xlog_ticket_alloc (log, unit_bytes, cnt, permanent); *ticp = tic; xlog_grant_push_ail(log, tic->t_cnt ? tic->t_unit_res * tic->t_cnt@@ -853,7 +850,7 @@ xlog_unmount_write(struct xlog_ticket *tic =NULL ;int error; - error =xfs_log_reserve (mp,600 ,1 , &tic,XFS_LOG ,0 ); + error =xfs_log_reserve (mp,600 ,1 , &tic,0 );if (error) goto out_err;@@ -2181,35 +2178,13 @@ xlog_write_calc_vec_length(static xlog_op_header_t *xlog_write_setup_ophdr ( -struct xlog *log,struct xlog_op_header *ophdr, -struct xlog_ticket *ticket, -uint flags) +struct xlog_ticket *ticket) { ophdr->oh_tid =cpu_to_be32 (ticket->t_tid); - ophdr->oh_clientid = ticket->t_clientid; + ophdr->oh_clientid =XFS_TRANSACTION ; ophdr->oh_res2 =0 ; - -/* are we copying a commit or unmount record? */ - ophdr->oh_flags = flags; - -/* -* We've seen logs corrupted with bad transaction client ids. This -* makes sure that XFS doesn't generate them on. Turn this into an EIO -* and shut down the filesystem. -*/ -switch (ophdr->oh_clientid) { -case XFS_TRANSACTION : -case XFS_VOLUME : -case XFS_LOG : -break ; -default : -xfs_warn (log->l_mp, -"Bad XFS transaction clientid 0x%x in ticket " PTR_FMT , - ophdr->oh_clientid, ticket); -return NULL ; - } - + ophdr->oh_flags =0 ;return ophdr; }@@ -2439,11 +2414,7 @@ xlog_write(if (index) optype &= ~XLOG_START_TRANS ; }else { - ophdr =xlog_write_setup_ophdr (log, ptr, - ticket, optype); -if (!ophdr) -return -EIO ; - + ophdr =xlog_write_setup_ophdr (ptr, ticket);xlog_write_adv_cnt (&ptr, &len, &log_offset,sizeof (struct xlog_op_header )); added_ophdr = true;@@ -3499,7 +3470,6 @@ xlog_ticket_alloc(struct xlog *log, int unit_bytes, int cnt, - char client, bool permanent) { struct xlog_ticket *tic;@@ -3517,7 +3487,6 @@ xlog_ticket_alloc( tic->t_cnt = cnt; tic->t_ocnt = cnt; tic->t_tid =prandom_u32 (); - tic->t_clientid = client;if (permanent) tic->t_flags |=XLOG_TIC_PERM_RESERV ;fs/xfs/xfs_log.h +6
-10@@ -117,16 +117,12 @@ int xfs_log_mount_finish(struct xfs_mount *mp);void xfs_log_mount_cancel (struct xfs_mount *);xfs_lsn_t xlog_assign_tail_lsn (struct xfs_mount *mp);xfs_lsn_t xlog_assign_tail_lsn_locked (struct xfs_mount *mp); -void xfs_log_space_wake (struct xfs_mount *mp); -int xfs_log_reserve (struct xfs_mount *mp, -int length, -int count, -struct xlog_ticket **ticket, -uint8_t clientid, -bool permanent); -int xfs_log_regrant (struct xfs_mount *mp,struct xlog_ticket *tic); -void xfs_log_unmount (struct xfs_mount *mp); -int xfs_log_force_umount (struct xfs_mount *mp,int logerror); +void xfs_log_space_wake (struct xfs_mount *mp); +int xfs_log_reserve (struct xfs_mount *mp,int length,int count, +struct xlog_ticket **ticket,bool permanent); +int xfs_log_regrant (struct xfs_mount *mp,struct xlog_ticket *tic); +void xfs_log_unmount (struct xfs_mount *mp); +int xfs_log_force_umount (struct xfs_mount *mp,int logerror);bool xfs_log_writable (struct xfs_mount *mp);struct xlog_ticket *xfs_log_ticket_get (struct xlog_ticket *ticket);fs/xfs/xfs_log_cil.c +1
-1@@ -37,7 +37,7 @@ xlog_cil_ticket_alloc( {struct xlog_ticket *tic; - tic =xlog_ticket_alloc (log,0 ,1 ,XFS_TRANSACTION ,0 ); + tic =xlog_ticket_alloc (log,0 ,1 ,0 );/* * set the current reservation to zero so we know to steal the basic fs/xfs/xfs_log_priv.h +2
-8@@ -158,7 +158,6 @@ typedef struct xlog_ticket {int t_unit_res;/* unit reservation in bytes : 4 */ char t_ocnt;/* original count : 1 */ char t_cnt;/* current count : 1 */ -char t_clientid;/* who does this belong to; : 1 */ char t_flags;/* properties of reservation : 1 */ /* reservation array fields */ @@ -465,13 +464,8 @@ extern __le32 xlog_cksum(struct xlog *log, struct xlog_rec_header *rhead, char *dp, int size);extern kmem_zone_t *xfs_log_ticket_zone; -struct xlog_ticket * -xlog_ticket_alloc ( -struct xlog *log, -int unit_bytes, -int count, -char client, -bool permanent); +struct xlog_ticket *xlog_ticket_alloc (struct xlog *log,int unit_bytes, +int count,bool permanent);static inline void xlog_write_adv_cnt (void **ptr,int *len,int *off,size_t bytes)fs/xfs/xfs_trans.c +2
-4@@ -194,11 +194,9 @@ xfs_trans_reserve(ASSERT (resp->tr_logflags &XFS_TRANS_PERM_LOG_RES ); error =xfs_log_regrant (mp, tp->t_ticket); }else { - error =xfs_log_reserve (mp, - resp->tr_logres, + error =xfs_log_reserve (mp, resp->tr_logres, resp->tr_logcount, - &tp->t_ticket,XFS_TRANSACTION , - permanent); + &tp->t_ticket, permanent); }if (error)Signature · 1 line
2.31.1
PATCH v4 17/39 xfs: move log iovec alignment to preparation function
In reply to Dave ChinnerFrom: Dave Chinner <[email protected]>
To include log op headers directly into the log iovec regions that the ophdrs wrap, we need to move the buffer alignment code from xlog_finish_iovec() to xlog_prepare_iovec(). This is because the xlog_op_header is only 12 bytes long, and we need the buffer that the caller formats their data into to be 8 byte aligned.
Hence once we start prepending the ophdr in xlog_prepare_iovec(), we are going to need to manage the padding directly to ensure that the buffer pointer returned is correctly aligned.
--- fs/xfs/xfs_log.h | 25 ++++++++++++++----------- 1 file changed, 14 insertions(+), 11 deletions(-)
fs/xfs/xfs_log.h +14 -11
@@ -21,6 +21,16 @@ struct xfs_log_vec {#define XFS_LOG_VEC_ORDERED (-1) +/* +* We need to make sure the buffer pointer returned is naturally aligned for the +* biggest basic data type we put into it. We have already accounted for this +* padding when sizing the buffer. +* +* However, this padding does not get written into the log, and hence we have to +* track the space used by the log vectors separately to prevent log space hangs +* due to inaccurate accounting (i.e. a leak) of the used log space through the +* CIL context ticket. +*/ static inline void *xlog_prepare_iovec (struct xfs_log_vec *lv,struct xfs_log_iovec **vecp,uint type)
@@ -34,6 +44,9 @@ xlog_prepare_iovec(struct xfs_log_vec *lv, struct xfs_log_iovec **vecp, vec = &lv->lv_iovecp[0 ]; } +if (!IS_ALIGNED (lv->lv_buf_len,sizeof (uint64_t ))) + lv->lv_buf_len =round_up (lv->lv_buf_len,sizeof (uint64_t )); + vec->i_type = type; vec->i_addr = lv->lv_buf + lv->lv_buf_len;
@@ -43,20 +56,10 @@ xlog_prepare_iovec(struct xfs_log_vec *lv, struct xfs_log_iovec **vecp,return vec->i_addr; } -/* -* We need to make sure the next buffer is naturally aligned for the biggest -* basic data type we put into it. We already accounted for this padding when -* sizing the buffer. -* -* However, this padding does not get written into the log, and hence we have to -* track the space used by the log vectors separately to prevent log space hangs -* due to inaccurate accounting (i.e. a leak) of the used log space through the -* CIL context ticket. -*/ static inline void xlog_finish_iovec (struct xfs_log_vec *lv,struct xfs_log_iovec *vec,int len) { - lv->lv_buf_len +=round_up (len,sizeof (uint64_t )); + lv->lv_buf_len += len; lv->lv_bytes += len; vec->i_len = len; }
Signature · 1 line
2.31.1
PATCH v4 18/39 xfs: reserve space and initialise xlog_op_header in item formatting
In reply to Dave ChinnerFrom: Dave Chinner <[email protected]>
Current xlog_write() adds op headers to the log manually for every log item region that is in the vector passed to it. While xlog_write() needs to stamp the transaction ID into the ophdr, we already know it's length, flags, clientid, etc at CIL commit time.
This means the only time that xlog write really needs to format and reserve space for a new ophdr is when a region is split across two iclogs. Adding the opheader and accounting for it as part of the normal formatted item region means we simplify the accounting of space used by a transaction and we don't have to special case reserving of space in for the ophdrs in xlog_write(). It also means we can largely initialise the ophdr in transaction commit instead of xlog_write, making the xlog_write formatting inner loop much tighter.
xlog_prepare_iovec() is now too large to stay as an inline function, so we move it out of line and into xfs_log.c.
Object sizes:
text data bss dec hex filename
1125934 305951 484 1432369 15db31 fs/xfs/built-in.a.before
1123360 305951 484 1429795 15d123 fs/xfs/built-in.a.after
So the code is a roughly 2.5kB smaller with xlog_prepare_iovec() now out of line, even though it grew in size itself.
--- fs/xfs/xfs_log.c | 115 +++++++++++++++++++++++++++++-------------- fs/xfs/xfs_log.h | 42 +++------------- fs/xfs/xfs_log_cil.c | 25 +++++----- 3 files changed, 99 insertions(+), 83 deletions(-)
fs/xfs/xfs_log.c +77 -38
@@ -89,6 +89,62 @@ xlog_iclogs_empty(static int xfs_log_cover (struct xfs_mount *); +/* +* We need to make sure the buffer pointer returned is naturally aligned for the +* biggest basic data type we put into it. We have already accounted for this +* padding when sizing the buffer. +* +* However, this padding does not get written into the log, and hence we have to +* track the space used by the log vectors separately to prevent log space hangs +* due to inaccurate accounting (i.e. a leak) of the used log space through the +* CIL context ticket. +* +* We also add space for the xlog_op_header that describes this region in the +* log. This prepends the data region we return to the caller to copy their data +* into, so do all the static initialisation of the ophdr now. Because the ophdr +* is not 8 byte aligned, we have to be careful to ensure that we align the +* start of the buffer such that the region we return to the call is 8 byte +* aligned and packed against the tail of the ophdr. +*/ +void * +xlog_prepare_iovec ( +struct xfs_log_vec *lv, +struct xfs_log_iovec **vecp, +uint type) +{ +struct xfs_log_iovec *vec = *vecp; +struct xlog_op_header *oph; +uint32_t len; +void *buf; + +if (vec) { +ASSERT (vec - lv->lv_iovecp < lv->lv_niovecs); + vec++; + }else { + vec = &lv->lv_iovecp[0 ]; + } + + len = lv->lv_buf_len +sizeof (struct xlog_op_header ); +if (!IS_ALIGNED (len,sizeof (uint64_t ))) { + lv->lv_buf_len =round_up (len,sizeof (uint64_t )) - +sizeof (struct xlog_op_header ); + } + + vec->i_type = type; + vec->i_addr = lv->lv_buf + lv->lv_buf_len; + + oph = vec->i_addr; + oph->oh_clientid =XFS_TRANSACTION ; + oph->oh_res2 =0 ; + oph->oh_flags =0 ; + + buf = vec->i_addr +sizeof (struct xlog_op_header ); +ASSERT (IS_ALIGNED ((unsigned long )buf,sizeof (uint64_t ))); + + *vecp = vec; +return buf; +} + static void xlog_grant_sub_space( struct xlog *log,
@@ -2131,9 +2187,9 @@ xlog_print_trans( }/* -* Calculate the potential space needed by the log vector. If this is a start -* transaction, the caller has already accounted for both opheaders in the start -* transaction, so we don't need to account for them here. +* Calculate the potential space needed by the log vector. All regions contain +* their own opheaders and they are accounted for in region space so we don't +* need to add them to the vector length here. */ static int xlog_write_calc_vec_length(
@@ -2160,18 +2216,7 @@ xlog_write_calc_vec_length(xlog_tic_add_region (ticket, vecp->i_len, vecp->i_type); } } - -/* Don't account for regions with embedded ophdrs */ -if (optype && headers >0 ) { - headers--; -if (optype &XLOG_START_TRANS ) { -ASSERT (headers >=1 ); - headers--; - } - } - ticket->t_res_num_ophdrs += headers; - len += headers *sizeof (struct xlog_op_header );return len; }
@@ -2181,7 +2226,6 @@ xlog_write_setup_ophdr(struct xlog_op_header *ophdr, struct xlog_ticket *ticket) { - ophdr->oh_tid = cpu_to_be32(ticket->t_tid); ophdr->oh_clientid = XFS_TRANSACTION; ophdr->oh_res2 =0 ; ophdr->oh_flags =0 ;
@@ -2404,21 +2448,25 @@ xlog_write(ASSERT ((unsigned long )ptr %sizeof (int32_t ) ==0 );/* -* The XLOG_START_TRANS has embedded ophdrs for the -* start record and transaction header. They will always -* be the first two regions in the lv chain. Commit and -* unmount records also have embedded ophdrs. +* Regions always have their ophdr at the start of the +* region, except for: +* - a transaction start which has a start record ophdr +* before the first region ophdr; and +* - the previous region didn't fully fit into an iclog +* so needs a continuation ophdr to prepend the region +* in this new iclog. */ -if (optype) { - ophdr = reg->i_addr; -if (index) - optype &= ~XLOG_START_TRANS ; - }else { + ophdr = reg->i_addr; +if (optype && index) { + optype &= ~XLOG_START_TRANS ; + }else if (partial_copy) { ophdr =xlog_write_setup_ophdr (ptr, ticket);xlog_write_adv_cnt (&ptr, &len, &log_offset,sizeof (struct xlog_op_header )); added_ophdr = true; } + ophdr->oh_tid =cpu_to_be32 (ticket->t_tid); + len += xlog_write_setup_copy(ticket, ophdr, iclog->ic_size-log_offset, reg->i_len,
@@ -2436,20 +2484,11 @@ xlog_write( ophdr->oh_len =cpu_to_be32 (copy_len -sizeof (struct xlog_op_header )); } -/* -* Copy region. -* -* Commit records just log an opheader, so -* we can have empty payloads with no data region to -* copy. Hence we only copy the payload if the vector -* says it has data to copy. -*/ -ASSERT (copy_len >=0 ); -if (copy_len >0 ) { -memcpy (ptr, reg->i_addr + copy_off, copy_len); -xlog_write_adv_cnt (&ptr, &len, &log_offset, - copy_len); - } + +ASSERT (copy_len >0 ); +memcpy (ptr, reg->i_addr + copy_off, copy_len); +xlog_write_adv_cnt (&ptr, &len, &log_offset, copy_len); +if (added_ophdr) copy_len +=sizeof (struct xlog_op_header ); record_cnt++;
fs/xfs/xfs_log.h +8 -34
52 lines not on this page · raw message
fs/xfs/xfs_log_cil.c +14 -11
@@ -181,13 +181,20 @@ xlog_cil_alloc_shadow_bufs( }/* -* We 64-bit align the length of each iovec so that the start -* of the next one is naturally aligned. We'll need to -* account for that slack space here. Then round nbytes up -* to 64-bit alignment so that the initial buffer alignment is -* easy to calculate and verify. +* We 64-bit align the length of each iovec so that the start of +* the next one is naturally aligned. We'll need to account for +* that slack space here. +* +* We also add the xlog_op_header to each region when +* formatting, but that's not accounted to the size of the item +* at this point. Hence we'll need an addition number of bytes +* for each vector to hold an opheader. +* +* Then round nbytes up to 64-bit alignment so that the initial +* buffer alignment is easy to calculate and verify. */ - nbytes += niovecs *sizeof (uint64_t ); + nbytes += niovecs * + (sizeof (uint64_t ) +sizeof (struct xlog_op_header )); nbytes =round_up (nbytes,sizeof (uint64_t ));/*
@@ -433,11 +440,6 @@ xlog_cil_insert_items(spin_lock (&cil->xc_cil_lock); -/* account for space used by new iovec headers */ - iovhdr_res = diff_iovecs *sizeof (xlog_op_header_t); - len += iovhdr_res; - ctx->nvecs += diff_iovecs; -/* attach the transaction to the CIL if it has any busy extents */ if (!list_empty (&tp->t_busy))list_splice_init (&tp->t_busy, &ctx->busy_extents);
@@ -469,6 +471,7 @@ xlog_cil_insert_items( } tp->t_ticket->t_curr_res -= len; ctx->space_used += len; + ctx->nvecs += diff_iovecs;/* * If we've overrun the reservation, dump the tx details before we move
Signature · 1 line
2.31.1
PATCH v4 19/39 xfs: log ticket region debug is largely useless
In reply to Dave ChinnerFrom: Dave Chinner <[email protected]>
xlog_tic_add_region() is used to trace the regions being added to a log ticket to provide information in the situation where a ticket reservation overrun occurs. The information gathered is stored int the ticket, and dumped if xlog_print_tic_res() is called.
For a front end struct xfs_trans overrun, the ticket only contains reservation tracking information - the ticket is never handed to the log so has no regions attached to it. The overrun debug information in this case comes from xlog_print_trans(), which walks the items attached to the transaction and dumps their attached formatted log vectors directly. It also dumps the ticket state, but that only contains reservation accounting and nothing else. Hence xlog_print_tic_res() never dumps region or overrun information from this path.
xlog_tic_add_region() is actually called from xlog_write(), which means it is being used to track the regions seen in a CIL checkpoint log vector chain. In looking at CIL behaviour recently, I've seen 32MB checkpoints regularly exceed 250,000 regions in the LV chain. The log ticket debug code can track *15* regions. IOWs, if there is a ticket overrun in the CIL code, the ticket region tracking code is going to be completely useless for determining what went wrong. The only thing it can tell us is how much of an overrun occurred, and we really don't need extra debug information in the log ticket to tell us that.
Indeed, the main place we call xlog_tic_add_region() is also adding up the number of regions and the space used so that xlog_write() knows how much will be written to the log. This is exactly the same information that log ticket is storing once we take away the useless region tracking array. Hence xlog_tic_add_region() is not useful, but can be called 250,000 times a CIL push...
Just strip all that debug "information" out of the of the log ticket and only have it report reservation space information when an overrun occurs. This also reduces the size of a log ticket down by about 150 bytes...
--- fs/xfs/xfs_log.c | 107 +++--------------------------------------- fs/xfs/xfs_log_priv.h | 20 -------- 2 files changed, 6 insertions(+), 121 deletions(-)
fs/xfs/xfs_log.c +6 -101
@@ -377,30 +377,6 @@ xlog_grant_head_check(return error; } -static void -xlog_tic_reset_res (xlog_ticket_t *tic) -{ - tic->t_res_num =0 ; - tic->t_res_arr_sum =0 ; - tic->t_res_num_ophdrs =0 ; -} - -static void -xlog_tic_add_region (xlog_ticket_t *tic,uint len,uint type) -{ -if (tic->t_res_num ==XLOG_TIC_LEN_MAX ) { -/* add to overflow and start again */ - tic->t_res_o_flow += tic->t_res_arr_sum; - tic->t_res_num =0 ; - tic->t_res_arr_sum =0 ; - } - - tic->t_res_arr[tic->t_res_num].r_len = len; - tic->t_res_arr[tic->t_res_num].r_type = type; - tic->t_res_arr_sum += len; - tic->t_res_num++; -} -bool xfs_log_writable (struct xfs_mount *mp)
@@ -450,8 +426,6 @@ xfs_log_regrant(xlog_grant_push_ail (log, tic->t_unit_res); tic->t_curr_res = tic->t_unit_res; -xlog_tic_reset_res (tic); -if (tic->t_cnt >0 )return 0 ;
@@ -2077,63 +2051,11 @@ xlog_print_tic_res(struct xfs_mount *mp, struct xlog_ticket *ticket) { - uint i; -uint ophdr_spc = ticket->t_res_num_ophdrs * (uint )sizeof (xlog_op_header_t); - -/* match with XLOG_REG_TYPE_* in xfs_log.h */ -#define REG_TYPE_STR (type, str) [XLOG_REG_TYPE_##type] = str -static char *res_type_str[] = { -REG_TYPE_STR (BFORMAT ,"bformat" ), -REG_TYPE_STR (BCHUNK ,"bchunk" ), -REG_TYPE_STR (EFI_FORMAT ,"efi_format" ), -REG_TYPE_STR (EFD_FORMAT ,"efd_format" ), -REG_TYPE_STR (IFORMAT ,"iformat" ), -REG_TYPE_STR (ICORE ,"icore" ), -REG_TYPE_STR (IEXT ,"iext" ), -REG_TYPE_STR (IBROOT ,"ibroot" ), -REG_TYPE_STR (ILOCAL ,"ilocal" ), -REG_TYPE_STR (IATTR_EXT ,"iattr_ext" ), -REG_TYPE_STR (IATTR_BROOT ,"iattr_broot" ), -REG_TYPE_STR (IATTR_LOCAL ,"iattr_local" ), -REG_TYPE_STR (QFORMAT ,"qformat" ), -REG_TYPE_STR (DQUOT ,"dquot" ), -REG_TYPE_STR (QUOTAOFF ,"quotaoff" ), -REG_TYPE_STR (LRHEADER ,"LR header" ), -REG_TYPE_STR (UNMOUNT ,"unmount" ), -REG_TYPE_STR (COMMIT ,"commit" ), -REG_TYPE_STR (TRANSHDR ,"trans header" ), -REG_TYPE_STR (ICREATE ,"inode create" ), -REG_TYPE_STR (RUI_FORMAT ,"rui_format" ), -REG_TYPE_STR (RUD_FORMAT ,"rud_format" ), -REG_TYPE_STR (CUI_FORMAT ,"cui_format" ), -REG_TYPE_STR (CUD_FORMAT ,"cud_format" ), -REG_TYPE_STR (BUI_FORMAT ,"bui_format" ), -REG_TYPE_STR (BUD_FORMAT ,"bud_format" ), - }; -BUILD_BUG_ON (ARRAY_SIZE (res_type_str) !=XLOG_REG_TYPE_MAX +1 ); -#undef REG_TYPE_STR - xfs_warn(mp, "ticket reservation summary:"); -xfs_warn (mp," unit res = %d bytes" , - ticket->t_unit_res); -xfs_warn (mp," current res = %d bytes" , - ticket->t_curr_res); -xfs_warn (mp," total reg = %u bytes (o/flow = %u bytes)" , - ticket->t_res_arr_sum, ticket->t_res_o_flow); -xfs_warn (mp," ophdrs = %u (ophdr space = %u bytes)" , - ticket->t_res_num_ophdrs, ophdr_spc); -xfs_warn (mp," ophdr + reg = %u bytes" , - ticket->t_res_arr_sum + ticket->t_res_o_flow + ophdr_spc); -xfs_warn (mp," num regions = %u" , - ticket->t_res_num); - -for (i =0 ; i < ticket->t_res_num; i++) { -uint r_type = ticket->t_res_arr[i].r_type; -xfs_warn (mp,"region[%u]: %s - %u bytes" , i, - ((r_type <=0 || r_type >XLOG_REG_TYPE_MAX ) ? -"bad-rtype" : res_type_str[r_type]), - ticket->t_res_arr[i].r_len); - } +xfs_warn (mp," unit res = %d bytes" , ticket->t_unit_res); +xfs_warn (mp," current res = %d bytes" , ticket->t_curr_res); +xfs_warn (mp," original count = %d" , ticket->t_ocnt); +xfs_warn (mp," remaining count = %d" , ticket->t_cnt); }/*
@@ -2198,7 +2120,6 @@ xlog_write_calc_vec_length( uint optype) {struct xfs_log_vec *lv; -int headers =0 ;int len =0 ;int i;
@@ -2207,17 +2128,9 @@ xlog_write_calc_vec_length(if (lv->lv_buf_len ==XFS_LOG_VEC_ORDERED )continue ; - headers += lv->lv_niovecs; - -for (i =0 ; i < lv->lv_niovecs; i++) { -struct xfs_log_iovec *vecp = &lv->lv_iovecp[i]; - - len += vecp->i_len; -xlog_tic_add_region (ticket, vecp->i_len, vecp->i_type); - } +for (i =0 ; i < lv->lv_niovecs; i++) + len += lv->lv_iovecp[i].i_len; } - ticket->t_res_num_ophdrs += headers; -return len; }
@@ -2276,7 +2189,6 @@ xlog_write_setup_copy(/* account for new log op header */ ticket->t_curr_res -=sizeof (struct xlog_op_header ); - ticket->t_res_num_ophdrs++;return sizeof (struct xlog_op_header ); }
@@ -2973,9 +2885,6 @@ xlog_state_get_iclog_space(*/ if (log_offset ==0 ) { ticket->t_curr_res -= log->l_iclog_hsize; -xlog_tic_add_region (ticket, - log->l_iclog_hsize, -XLOG_REG_TYPE_LRHEADER ); head->h_cycle =cpu_to_be32 (log->l_curr_cycle); head->h_lsn =cpu_to_be64 (xlog_assign_lsn (log->l_curr_cycle, log->l_curr_block));
@@ -3055,7 +2964,6 @@ xfs_log_ticket_regrant(xlog_grant_sub_space (log, &log->l_write_head.grant, ticket->t_curr_res); ticket->t_curr_res = ticket->t_unit_res; -xlog_tic_reset_res (ticket);trace_xfs_log_ticket_regrant_sub (log, ticket);
@@ -3066,7 +2974,6 @@ xfs_log_ticket_regrant(trace_xfs_log_ticket_regrant_exit (log, ticket); ticket->t_curr_res = ticket->t_unit_res; -xlog_tic_reset_res (ticket); }xfs_log_ticket_put (ticket);
@@ -3529,8 +3436,6 @@ xlog_ticket_alloc(if (permanent) tic->t_flags |=XLOG_TIC_PERM_RESERV ; -xlog_tic_reset_res (tic); -return tic; }
fs/xfs/xfs_log_priv.h +0 -20
@@ -136,19 +136,6 @@ enum xlog_iclog_state {#define XLOG_ICL_NEED_FLUSH (1 << 0)/* iclog needs REQ_PREFLUSH */ #define XLOG_ICL_NEED_FUA (1 << 1)/* iclog needs REQ_FUA */ -/* Ticket reservation region accounting */ -#define XLOG_TIC_LEN_MAX 15 - -/* -* Reservation region -* As would be stored in xfs_log_iovec but without the i_addr which -* we don't care about. -*/ -typedef struct xlog_res { -uint r_len;/* region length :4 */ -uint r_type;/* region's transaction type :4 */ -}xlog_res_t ; - typedef struct xlog_ticket { struct list_head t_queue; /* reserve/write queue */ struct task_struct *t_task; /* task that owns this ticket */
@@ -159,13 +146,6 @@ typedef struct xlog_ticket {char t_ocnt;/* original count : 1 */ char t_cnt;/* current count : 1 */ char t_flags;/* properties of reservation : 1 */ - -/* reservation array fields */ -uint t_res_num;/* num in array : 4 */ -uint t_res_num_ophdrs;/* num op hdrs : 4 */ -uint t_res_arr_sum;/* array sum : 4 */ -uint t_res_o_flow;/* sum overflow : 4 */ -xlog_res_t t_res_arr[XLOG_TIC_LEN_MAX ];/* array of res : 8 * 15 */ } xlog_ticket_t;/*
Signature · 1 line
2.31.1
PATCH v4 20/39 xfs: pass lv chain length into xlog_write()
In reply to Dave ChinnerFrom: Dave Chinner <[email protected]>
The caller of xlog_write() usually has a close accounting of the aggregated vector length contained in the log vector chain passed to xlog_write(). There is no need to iterate the chain to calculate he length of the data in xlog_write_calculate_len() if the caller is already iterating that chain to build it.
Passing in the vector length avoids doing an extra chain iteration, which can be a significant amount of work given that large CIL commits can have hundreds of thousands of vectors attached to the chain.
--- fs/xfs/xfs_log.c | 37 ++++++------------------------------- fs/xfs/xfs_log_cil.c | 17 ++++++++++++----- fs/xfs/xfs_log_priv.h | 2 +- 3 files changed, 19 insertions(+), 37 deletions(-)
fs/xfs/xfs_log.c +6 -31
@@ -864,7 +864,8 @@ xlog_write_unmount_record(*/ if (log->l_targ != log->l_mp->m_ddev_targp)blkdev_issue_flush (log->l_targ->bt_bdev); -return xlog_write (log, &vec, ticket,NULL ,NULL ,XLOG_UNMOUNT_TRANS ); +return xlog_write (log, &vec, ticket,NULL ,NULL ,XLOG_UNMOUNT_TRANS , + reg.i_len); }/*
@@ -1588,7 +1589,8 @@ xlog_commit_record(/* account for space used by record data */ ticket->t_curr_res -= reg.i_len; - error =xlog_write (log, &vec, ticket, lsn, iclog,XLOG_COMMIT_TRANS ); + error =xlog_write (log, &vec, ticket, lsn, iclog,XLOG_COMMIT_TRANS , + reg.i_len);if (error)xfs_force_shutdown (log->l_mp,SHUTDOWN_LOG_IO_ERROR );return error;
@@ -2108,32 +2110,6 @@ xlog_print_trans( } } -/* -* Calculate the potential space needed by the log vector. All regions contain -* their own opheaders and they are accounted for in region space so we don't -* need to add them to the vector length here. -*/ -static int -xlog_write_calc_vec_length ( -struct xlog_ticket *ticket, -struct xfs_log_vec *log_vector, -uint optype) -{ -struct xfs_log_vec *lv; -int len =0 ; -int i; - -for (lv = log_vector; lv; lv = lv->lv_next) { -/* we don't write ordered log vectors */ -if (lv->lv_buf_len ==XFS_LOG_VEC_ORDERED ) -continue ; - -for (i =0 ; i < lv->lv_niovecs; i++) - len += lv->lv_iovecp[i].i_len; - } -return len; -} -static xlog_op_header_t * xlog_write_setup_ophdr( struct xlog_op_header *ophdr,
@@ -2296,13 +2272,13 @@ xlog_write(struct xlog_ticket *ticket, xfs_lsn_t *start_lsn, struct xlog_in_core **commit_iclog, - uint optype) + uint optype, + uint32_t len) { struct xlog_in_core *iclog = NULL;struct xfs_log_vec *lv = log_vector;struct xfs_log_iovec *vecp = lv->lv_iovecp;int index =0 ; -int len;int partial_copy =0 ;int partial_copy_len =0 ;int contwr =0 ;
@@ -2317,7 +2293,6 @@ xlog_write(xfs_force_shutdown (log->l_mp,SHUTDOWN_LOG_IO_ERROR ); } - len =xlog_write_calc_vec_length (ticket, log_vector, optype);if (start_lsn) *start_lsn =0 ;while (lv && (!lv->lv_niovecs || index < lv->lv_niovecs)) {
fs/xfs/xfs_log_cil.c +12 -5
@@ -710,11 +710,12 @@ xlog_cil_build_trans_hdr(sizeof (struct xfs_trans_header ); hdr->lhdr[1 ].i_type =XLOG_REG_TYPE_TRANSHDR ; - tic->t_curr_res -= hdr->lhdr[0 ].i_len + hdr->lhdr[1 ].i_len; - lvhdr->lv_niovecs =2 ; lvhdr->lv_iovecp = &hdr->lhdr[0 ]; + lvhdr->lv_bytes = hdr->lhdr[0 ].i_len + hdr->lhdr[1 ].i_len; lvhdr->lv_next = ctx->lv_chain; + + tic->t_curr_res -= lvhdr->lv_bytes; }/*
@@ -742,7 +743,8 @@ xlog_cil_push_work(struct xfs_log_vec *lv;struct xfs_cil_ctx *new_ctx;struct xlog_in_core *commit_iclog; -int num_iovecs; +int num_iovecs =0 ; +int num_bytes =0 ;int error =0 ;struct xlog_cil_trans_hdr thdr;struct xfs_log_vec lvhdr = {NULL };
@@ -835,7 +837,6 @@ xlog_cil_push_work(* by the flush lock. */ lv =NULL ; - num_iovecs =0 ;while (!list_empty (&cil->xc_cil)) {struct xfs_log_item *item;
@@ -849,6 +850,10 @@ xlog_cil_push_work( lv = item->li_lv; item->li_lv =NULL ; num_iovecs += lv->lv_niovecs; + +/* we don't write ordered log vectors */ +if (lv->lv_buf_len !=XFS_LOG_VEC_ORDERED ) + num_bytes += lv->lv_bytes; }/*
@@ -887,6 +892,8 @@ xlog_cil_push_work(* transaction header here as it is not accounted for in xlog_write(). */ xlog_cil_build_trans_hdr (ctx, &thdr, &lvhdr, num_iovecs); + num_iovecs += lvhdr.lv_niovecs; + num_bytes += lvhdr.lv_bytes;/* * Before we format and submit the first iclog, we have to ensure that
@@ -901,7 +908,7 @@ xlog_cil_push_work(* write head. */ error =xlog_write (log, &lvhdr, ctx->ticket, &ctx->start_lsn,NULL , -XLOG_START_TRANS ); +XLOG_START_TRANS , num_bytes);if (error) goto out_abort_free_ticket;
fs/xfs/xfs_log_priv.h +1 -1
@@ -459,7 +459,7 @@ void xlog_print_tic_res(struct xfs_mount *mp, struct xlog_ticket *ticket);void xlog_print_trans (struct xfs_trans *);int xlog_write (struct xlog *log,struct xfs_log_vec *log_vector,struct xlog_ticket *tic,xfs_lsn_t *start_lsn, -struct xlog_in_core **commit_iclog,uint optype); +struct xlog_in_core **commit_iclog,uint optype,uint32_t len);int xlog_commit_record (struct xlog *log,struct xlog_ticket *ticket,struct xlog_in_core **iclog,xfs_lsn_t *lsn);
Signature · 1 line
2.31.1
Re: [PATCH 20/39] xfs: pass lv chain length into xlog_write()
Quoted 150 lines · Dave Chinner
From: Dave Chinner <[email protected]>
The caller of
xlog_write()usually has a close accounting of the aggregated vector length contained in the log vector chain passed toxlog_write(). There is no need to iterate the chain to calculate he length of the data inxlog_write_calculate_len()if the caller is already iterating that chain to build it.Passing in the vector length avoids doing an extra chain iteration, which can be a significant amount of work given that large CIL commits can have hundreds of thousands of vectors attached to the chain.
Signed-off-by Dave Chinner <[email protected]>--- fs/xfs/xfs_log.c | 37 ++++++------------------------------- fs/xfs/xfs_log_cil.c | 17 ++++++++++++----- fs/xfs/xfs_log_priv.h | 2 +- 3 files changed, 19 insertions(+), 37 deletions(-)fs/xfs/xfs_log.c +6
-31@@ -864,7 +864,8 @@ xlog_write_unmount_record(*/ if (log->l_targ != log->l_mp->m_ddev_targp)blkdev_issue_flush (log->l_targ->bt_bdev); -return xlog_write (log, &vec, ticket,NULL ,NULL ,XLOG_UNMOUNT_TRANS ); +return xlog_write (log, &vec, ticket,NULL ,NULL ,XLOG_UNMOUNT_TRANS , + reg.i_len); }/* @@ -1588,7 +1589,8 @@ xlog_commit_record(/* account for space used by record data */ ticket->t_curr_res -= reg.i_len; - error =xlog_write (log, &vec, ticket, lsn, iclog,XLOG_COMMIT_TRANS ); + error =xlog_write (log, &vec, ticket, lsn, iclog,XLOG_COMMIT_TRANS , + reg.i_len);if (error)xfs_force_shutdown (log->l_mp,SHUTDOWN_LOG_IO_ERROR );return error;@@ -2108,32 +2110,6 @@ xlog_print_trans( } } -/* -* Calculate the potential space needed by the log vector. All regions contain -* their own opheaders and they are accounted for in region space so we don't -* need to add them to the vector length here. -*/ -static int -xlog_write_calc_vec_length ( -struct xlog_ticket *ticket, -struct xfs_log_vec *log_vector, -uint optype) -{ -struct xfs_log_vec *lv; -int len =0 ; -int i; - -for (lv = log_vector; lv; lv = lv->lv_next) { -/* we don't write ordered log vectors */ -if (lv->lv_buf_len ==XFS_LOG_VEC_ORDERED ) -continue ; - -for (i =0 ; i < lv->lv_niovecs; i++) - len += lv->lv_iovecp[i].i_len; - } -return len; -} -static xlog_op_header_t * xlog_write_setup_ophdr( struct xlog_op_header *ophdr,@@ -2296,13 +2272,13 @@ xlog_write(struct xlog_ticket *ticket, xfs_lsn_t *start_lsn, struct xlog_in_core **commit_iclog, - uint optype) + uint optype, + uint32_t len) { struct xlog_in_core *iclog = NULL;struct xfs_log_vec *lv = log_vector;struct xfs_log_iovec *vecp = lv->lv_iovecp;int index =0 ; -int len;int partial_copy =0 ;int partial_copy_len =0 ;int contwr =0 ;@@ -2317,7 +2293,6 @@ xlog_write(xfs_force_shutdown (log->l_mp,SHUTDOWN_LOG_IO_ERROR ); } - len =xlog_write_calc_vec_length (ticket, log_vector, optype);if (start_lsn) *start_lsn =0 ;while (lv && (!lv->lv_niovecs || index < lv->lv_niovecs)) {fs/xfs/xfs_log_cil.c +9
-4@@ -710,11 +710,12 @@ xlog_cil_build_trans_hdr(sizeof (struct xfs_trans_header ); hdr->lhdr[1 ].i_type =XLOG_REG_TYPE_TRANSHDR ; - tic->t_curr_res -= hdr->lhdr[0 ].i_len + hdr->lhdr[1 ].i_len; - lvhdr->lv_niovecs =2 ; lvhdr->lv_iovecp = &hdr->lhdr[0 ]; + lvhdr->lv_bytes = hdr->lhdr[0 ].i_len + hdr->lhdr[1 ].i_len; lvhdr->lv_next = ctx->lv_chain; + + tic->t_curr_res -= lvhdr->lv_bytes; }/* @@ -742,7 +743,8 @@ xlog_cil_push_work(struct xfs_log_vec *lv;struct xfs_cil_ctx *new_ctx;struct xlog_in_core *commit_iclog; -int num_iovecs; +int num_iovecs =0 ; +int num_bytes =0 ;int error =0 ;struct xlog_cil_trans_hdr thdr;struct xfs_log_vec lvhdr = {NULL };@@ -835,7 +837,6 @@ xlog_cil_push_work(* by the flush lock. */ lv =NULL ; - num_iovecs =0 ;while (!list_empty (&cil->xc_cil)) {struct xfs_log_item *item;@@ -849,6 +850,10 @@ xlog_cil_push_work( lv = item->li_lv; item->li_lv =NULL ; num_iovecs += lv->lv_niovecs; + +/* we don't write ordered log vectors */ +if (lv->lv_buf_len !=XFS_LOG_VEC_ORDERED ) + num_bytes += lv->lv_bytes; }/*
fs/xfs/xfs_log_cil.c xlog_cil_push_work( · line 895
Quoted 4 lines
fs/xfs/xfs_log_cil.c +1
-0@@ -887,6 +892,8 @@ xlog_cil_push_work(* transaction header here as it is not accounted for in xlog_write(). */ xlog_cil_build_trans_hdr (ctx, &thdr, &lvhdr, num_iovecs); + num_iovecs += lvhdr.lv_niovecs;
I have the same question that Brian had last time, which is: What's the point of updating num_iovecs here? It's not used after xlog_cil_build_trans_hdr, either here or at the end of the patchset.
Is the idea that num_{iovecs,bytes} will always reflect everything in the cil context chain that's about to be passed to xlog_write?
--D
Quoted 29 lines
fs/xfs/xfs_log_cil.c +2
-1+ num_bytes += lvhdr.lv_bytes;/* * Before we format and submit the first iclog, we have to ensure that @@ -901,7 +908,7 @@ xlog_cil_push_work(* write head. */ error =xlog_write (log, &lvhdr, ctx->ticket, &ctx->start_lsn,NULL , -XLOG_START_TRANS ); +XLOG_START_TRANS , num_bytes);if (error) goto out_abort_free_ticket;fs/xfs/xfs_log_priv.h +1
-1@@ -459,7 +459,7 @@ void xlog_print_tic_res(struct xfs_mount *mp, struct xlog_ticket *ticket);void xlog_print_trans (struct xfs_trans *);int xlog_write (struct xlog *log,struct xfs_log_vec *log_vector,struct xlog_ticket *tic,xfs_lsn_t *start_lsn, -struct xlog_in_core **commit_iclog,uint optype); +struct xlog_in_core **commit_iclog,uint optype,uint32_t len);int xlog_commit_record (struct xlog *log,struct xlog_ticket *ticket,struct xlog_in_core **iclog,xfs_lsn_t *lsn);Signature · 1 line
2.31.1
Re: [PATCH 20/39] xfs: pass lv chain length into xlog_write()
Quoted 15 lines · Darrick J. Wong
Quoted 14 lines · Dave Chinner
From: Dave Chinner <[email protected]>
The caller of
xlog_write()usually has a close accounting of the aggregated vector length contained in the log vector chain passed toxlog_write(). There is no need to iterate the chain to calculate he length of the data inxlog_write_calculate_len()if the caller is already iterating that chain to build it.Passing in the vector length avoids doing an extra chain iteration, which can be a significant amount of work given that large CIL commits can have hundreds of thousands of vectors attached to the chain.
Signed-off-by Dave Chinner <[email protected]>
....
Quoted 21 lines
Quoted 12 lines
fs/xfs/xfs_log_cil.c +4
-0@@ -849,6 +850,10 @@ xlog_cil_push_work( lv = item->li_lv; item->li_lv =NULL ; num_iovecs += lv->lv_niovecs; + +/* we don't write ordered log vectors */ +if (lv->lv_buf_len !=XFS_LOG_VEC_ORDERED ) + num_bytes += lv->lv_bytes; }/* fs/xfs/xfs_log_cil.c xlog_cil_push_work( · line 895
Quoted 4 lines
fs/xfs/xfs_log_cil.c +1
-0@@ -887,6 +892,8 @@ xlog_cil_push_work(* transaction header here as it is not accounted for in xlog_write(). */ xlog_cil_build_trans_hdr (ctx, &thdr, &lvhdr, num_iovecs); + num_iovecs += lvhdr.lv_niovecs;I have the same question that Brian had last time, which is: What's the point of updating
num_iovecshere? It's not used afterxlog_cil_build_trans_hdr, either here or at the end of the patchset.
Quoted 2 lines
Is the idea that num_{iovecs,bytes} will always reflect everything in the cil context chain that's about to be passed to
xlog_write?
I left it there because I did want to keep the two variables up to date for future use. i.e. I didn't want to leave a landmine later down the track if I need to use num_iovecs in future changes. I've also used it a few times for temporary debugging code, so I'd prefer to keep it even though it isn't used.
But if "not used" is the only reason for people not giving rvbs, then I can remove it...
Cheers,
Dave.
Signature · 2 lines
Dave Chinner [email protected]
Re: [PATCH 20/39] xfs: pass lv chain length into xlog_write()
Quoted 41 lines · Dave Chinner
Quoted 15 lines · Darrick J. Wong
Quoted 14 lines · Dave Chinner
From: Dave Chinner <[email protected]>
The caller of
xlog_write()usually has a close accounting of the aggregated vector length contained in the log vector chain passed toxlog_write(). There is no need to iterate the chain to calculate he length of the data inxlog_write_calculate_len()if the caller is already iterating that chain to build it.Passing in the vector length avoids doing an extra chain iteration, which can be a significant amount of work given that large CIL commits can have hundreds of thousands of vectors attached to the chain.
Signed-off-by Dave Chinner <[email protected]>....
Quoted 21 lines
Quoted 12 lines
fs/xfs/xfs_log_cil.c +4
-0@@ -849,6 +850,10 @@ xlog_cil_push_work( lv = item->li_lv; item->li_lv =NULL ; num_iovecs += lv->lv_niovecs; + +/* we don't write ordered log vectors */ +if (lv->lv_buf_len !=XFS_LOG_VEC_ORDERED ) + num_bytes += lv->lv_bytes; }/* fs/xfs/xfs_log_cil.c xlog_cil_push_work( · line 895
Quoted 4 lines
fs/xfs/xfs_log_cil.c +1
-0@@ -887,6 +892,8 @@ xlog_cil_push_work(* transaction header here as it is not accounted for in xlog_write(). */ xlog_cil_build_trans_hdr (ctx, &thdr, &lvhdr, num_iovecs); + num_iovecs += lvhdr.lv_niovecs;I have the same question that Brian had last time, which is: What's the point of updating
num_iovecshere? It's not used afterxlog_cil_build_trans_hdr, either here or at the end of the patchset.Quoted 2 lines
Is the idea that num_{iovecs,bytes} will always reflect everything in the cil context chain that's about to be passed to
xlog_write?
Quoted 8 lines
I left it there because I did want to keep the two variables up to date for future use. i.e. I didn't want to leave a landmine later down the track if I need to use
num_iovecsin future changes. I've also used it a few times for temporary debugging code, so I'd prefer to keep it even though it isn't used.But if "not used" is the only reason for people not giving rvbs, then I can remove it...
...or feed it to a tracepoint, if you find it useful for debugging the size of log writes? <shrug>
--D
Quoted 7 lines
Cheers,
Dave.
Signature · 2 lines
Dave Chinner [email protected]
PATCH v2 20/39 xfs: pass lv chain length into xlog_write()
From: Dave Chinner <[email protected]>
The caller of xlog_write() usually has a close accounting of the aggregated vector length contained in the log vector chain passed to xlog_write(). There is no need to iterate the chain to calculate he length of the data in xlog_write_calculate_len() if the caller is already iterating that chain to build it.
Passing in the vector length avoids doing an extra chain iteration, which can be a significant amount of work given that large CIL commits can have hundreds of thousands of vectors attached to the chain.
---
V2: removed unnecessary update of num_iovecs.
fs/xfs/xfs_log.c | 37 ++++++------------------------------- fs/xfs/xfs_log_cil.c | 16 +++++++++++----- fs/xfs/xfs_log_priv.h | 2 +- 3 files changed, 18 insertions(+), 37 deletions(-)
fs/xfs/xfs_log.c +6 -31
@@ -864,7 +864,8 @@ xlog_write_unmount_record(*/ if (log->l_targ != log->l_mp->m_ddev_targp)blkdev_issue_flush (log->l_targ->bt_bdev); -return xlog_write (log, &vec, ticket,NULL ,NULL ,XLOG_UNMOUNT_TRANS ); +return xlog_write (log, &vec, ticket,NULL ,NULL ,XLOG_UNMOUNT_TRANS , + reg.i_len); }/*
@@ -1588,7 +1589,8 @@ xlog_commit_record(/* account for space used by record data */ ticket->t_curr_res -= reg.i_len; - error =xlog_write (log, &vec, ticket, lsn, iclog,XLOG_COMMIT_TRANS ); + error =xlog_write (log, &vec, ticket, lsn, iclog,XLOG_COMMIT_TRANS , + reg.i_len);if (error)xfs_force_shutdown (log->l_mp,SHUTDOWN_LOG_IO_ERROR );return error;
@@ -2108,32 +2110,6 @@ xlog_print_trans( } } -/* -* Calculate the potential space needed by the log vector. All regions contain -* their own opheaders and they are accounted for in region space so we don't -* need to add them to the vector length here. -*/ -static int -xlog_write_calc_vec_length ( -struct xlog_ticket *ticket, -struct xfs_log_vec *log_vector, -uint optype) -{ -struct xfs_log_vec *lv; -int len =0 ; -int i; - -for (lv = log_vector; lv; lv = lv->lv_next) { -/* we don't write ordered log vectors */ -if (lv->lv_buf_len ==XFS_LOG_VEC_ORDERED ) -continue ; - -for (i =0 ; i < lv->lv_niovecs; i++) - len += lv->lv_iovecp[i].i_len; - } -return len; -} -static xlog_op_header_t * xlog_write_setup_ophdr( struct xlog_op_header *ophdr,
@@ -2296,13 +2272,13 @@ xlog_write(struct xlog_ticket *ticket, xfs_lsn_t *start_lsn, struct xlog_in_core **commit_iclog, - uint optype) + uint optype, + uint32_t len) { struct xlog_in_core *iclog = NULL;struct xfs_log_vec *lv = log_vector;struct xfs_log_iovec *vecp = lv->lv_iovecp;int index =0 ; -int len;int partial_copy =0 ;int partial_copy_len =0 ;int contwr =0 ;
@@ -2317,7 +2293,6 @@ xlog_write(xfs_force_shutdown (log->l_mp,SHUTDOWN_LOG_IO_ERROR ); } - len =xlog_write_calc_vec_length (ticket, log_vector, optype);if (start_lsn) *start_lsn =0 ;while (lv && (!lv->lv_niovecs || index < lv->lv_niovecs)) {
fs/xfs/xfs_log_cil.c +11 -5
@@ -710,11 +710,12 @@ xlog_cil_build_trans_hdr(sizeof (struct xfs_trans_header ); hdr->lhdr[1 ].i_type =XLOG_REG_TYPE_TRANSHDR ; - tic->t_curr_res -= hdr->lhdr[0 ].i_len + hdr->lhdr[1 ].i_len; - lvhdr->lv_niovecs =2 ; lvhdr->lv_iovecp = &hdr->lhdr[0 ]; + lvhdr->lv_bytes = hdr->lhdr[0 ].i_len + hdr->lhdr[1 ].i_len; lvhdr->lv_next = ctx->lv_chain; + + tic->t_curr_res -= lvhdr->lv_bytes; }/*
@@ -742,7 +743,8 @@ xlog_cil_push_work(struct xfs_log_vec *lv;struct xfs_cil_ctx *new_ctx;struct xlog_in_core *commit_iclog; -int num_iovecs; +int num_iovecs =0 ; +int num_bytes =0 ;int error =0 ;struct xlog_cil_trans_hdr thdr;struct xfs_log_vec lvhdr = {NULL };
@@ -835,7 +837,6 @@ xlog_cil_push_work(* by the flush lock. */ lv =NULL ; - num_iovecs =0 ;while (!list_empty (&cil->xc_cil)) {struct xfs_log_item *item;
@@ -849,6 +850,10 @@ xlog_cil_push_work( lv = item->li_lv; item->li_lv =NULL ; num_iovecs += lv->lv_niovecs; + +/* we don't write ordered log vectors */ +if (lv->lv_buf_len !=XFS_LOG_VEC_ORDERED ) + num_bytes += lv->lv_bytes; }/*
@@ -887,6 +892,7 @@ xlog_cil_push_work(* transaction header here as it is not accounted for in xlog_write(). */ xlog_cil_build_trans_hdr (ctx, &thdr, &lvhdr, num_iovecs); + num_bytes += lvhdr.lv_bytes;/* * Before we format and submit the first iclog, we have to ensure that
@@ -901,7 +907,7 @@ xlog_cil_push_work(* write head. */ error =xlog_write (log, &lvhdr, ctx->ticket, &ctx->start_lsn,NULL , -XLOG_START_TRANS ); +XLOG_START_TRANS , num_bytes);if (error) goto out_abort_free_ticket;
fs/xfs/xfs_log_priv.h +1 -1
@@ -459,7 +459,7 @@ void xlog_print_tic_res(struct xfs_mount *mp, struct xlog_ticket *ticket);void xlog_print_trans (struct xfs_trans *);int xlog_write (struct xlog *log,struct xfs_log_vec *log_vector,struct xlog_ticket *tic,xfs_lsn_t *start_lsn, -struct xlog_in_core **commit_iclog,uint optype); +struct xlog_in_core **commit_iclog,uint optype,uint32_t len);int xlog_commit_record (struct xlog *log,struct xlog_ticket *ticket,struct xlog_in_core **iclog,xfs_lsn_t *lsn);
Signature · 2 lines
Dave Chinner [email protected]
Re: [PATCH 20/39 V2] xfs: pass lv chain length into xlog_write()
Quoted 8 lines · Dave Chinner
From: Dave Chinner <[email protected]>
The caller of
xlog_write()usually has a close accounting of the aggregated vector length contained in the log vector chain passed toxlog_write(). There is no need to iterate the chain to calculate he length of the data inxlog_write_calculate_len()if the caller is already iterating that chain to build it.
Quoted 8 lines
Passing in the vector length avoids doing an extra chain iteration, which can be a significant amount of work given that large CIL commits can have hundreds of thousands of vectors attached to the chain.
Signed-off-by Dave Chinner <[email protected]>---
V2: removed unnecessary update ofnum_iovecs.
Heh, nice refactor :)
--D
Quoted 168 lines
fs/xfs/xfs_log.c | 37 ++++++------------------------------- fs/xfs/xfs_log_cil.c | 16 +++++++++++----- fs/xfs/xfs_log_priv.h | 2 +- 3 files changed, 18 insertions(+), 37 deletions(-)fs/xfs/xfs_log.c +6
-31@@ -864,7 +864,8 @@ xlog_write_unmount_record(*/ if (log->l_targ != log->l_mp->m_ddev_targp)blkdev_issue_flush (log->l_targ->bt_bdev); -return xlog_write (log, &vec, ticket,NULL ,NULL ,XLOG_UNMOUNT_TRANS ); +return xlog_write (log, &vec, ticket,NULL ,NULL ,XLOG_UNMOUNT_TRANS , + reg.i_len); }/* @@ -1588,7 +1589,8 @@ xlog_commit_record(/* account for space used by record data */ ticket->t_curr_res -= reg.i_len; - error =xlog_write (log, &vec, ticket, lsn, iclog,XLOG_COMMIT_TRANS ); + error =xlog_write (log, &vec, ticket, lsn, iclog,XLOG_COMMIT_TRANS , + reg.i_len);if (error)xfs_force_shutdown (log->l_mp,SHUTDOWN_LOG_IO_ERROR );return error;@@ -2108,32 +2110,6 @@ xlog_print_trans( } } -/* -* Calculate the potential space needed by the log vector. All regions contain -* their own opheaders and they are accounted for in region space so we don't -* need to add them to the vector length here. -*/ -static int -xlog_write_calc_vec_length ( -struct xlog_ticket *ticket, -struct xfs_log_vec *log_vector, -uint optype) -{ -struct xfs_log_vec *lv; -int len =0 ; -int i; - -for (lv = log_vector; lv; lv = lv->lv_next) { -/* we don't write ordered log vectors */ -if (lv->lv_buf_len ==XFS_LOG_VEC_ORDERED ) -continue ; - -for (i =0 ; i < lv->lv_niovecs; i++) - len += lv->lv_iovecp[i].i_len; - } -return len; -} -static xlog_op_header_t * xlog_write_setup_ophdr( struct xlog_op_header *ophdr,@@ -2296,13 +2272,13 @@ xlog_write(struct xlog_ticket *ticket, xfs_lsn_t *start_lsn, struct xlog_in_core **commit_iclog, - uint optype) + uint optype, + uint32_t len) { struct xlog_in_core *iclog = NULL;struct xfs_log_vec *lv = log_vector;struct xfs_log_iovec *vecp = lv->lv_iovecp;int index =0 ; -int len;int partial_copy =0 ;int partial_copy_len =0 ;int contwr =0 ;@@ -2317,7 +2293,6 @@ xlog_write(xfs_force_shutdown (log->l_mp,SHUTDOWN_LOG_IO_ERROR ); } - len =xlog_write_calc_vec_length (ticket, log_vector, optype);if (start_lsn) *start_lsn =0 ;while (lv && (!lv->lv_niovecs || index < lv->lv_niovecs)) {fs/xfs/xfs_log_cil.c +11
-5@@ -710,11 +710,12 @@ xlog_cil_build_trans_hdr(sizeof (struct xfs_trans_header ); hdr->lhdr[1 ].i_type =XLOG_REG_TYPE_TRANSHDR ; - tic->t_curr_res -= hdr->lhdr[0 ].i_len + hdr->lhdr[1 ].i_len; - lvhdr->lv_niovecs =2 ; lvhdr->lv_iovecp = &hdr->lhdr[0 ]; + lvhdr->lv_bytes = hdr->lhdr[0 ].i_len + hdr->lhdr[1 ].i_len; lvhdr->lv_next = ctx->lv_chain; + + tic->t_curr_res -= lvhdr->lv_bytes; }/* @@ -742,7 +743,8 @@ xlog_cil_push_work(struct xfs_log_vec *lv;struct xfs_cil_ctx *new_ctx;struct xlog_in_core *commit_iclog; -int num_iovecs; +int num_iovecs =0 ; +int num_bytes =0 ;int error =0 ;struct xlog_cil_trans_hdr thdr;struct xfs_log_vec lvhdr = {NULL };@@ -835,7 +837,6 @@ xlog_cil_push_work(* by the flush lock. */ lv =NULL ; - num_iovecs =0 ;while (!list_empty (&cil->xc_cil)) {struct xfs_log_item *item;@@ -849,6 +850,10 @@ xlog_cil_push_work( lv = item->li_lv; item->li_lv =NULL ; num_iovecs += lv->lv_niovecs; + +/* we don't write ordered log vectors */ +if (lv->lv_buf_len !=XFS_LOG_VEC_ORDERED ) + num_bytes += lv->lv_bytes; }/* @@ -887,6 +892,7 @@ xlog_cil_push_work(* transaction header here as it is not accounted for in xlog_write(). */ xlog_cil_build_trans_hdr (ctx, &thdr, &lvhdr, num_iovecs); + num_bytes += lvhdr.lv_bytes;/* * Before we format and submit the first iclog, we have to ensure that @@ -901,7 +907,7 @@ xlog_cil_push_work(* write head. */ error =xlog_write (log, &lvhdr, ctx->ticket, &ctx->start_lsn,NULL , -XLOG_START_TRANS ); +XLOG_START_TRANS , num_bytes);if (error) goto out_abort_free_ticket;fs/xfs/xfs_log_priv.h +1
-1@@ -459,7 +459,7 @@ void xlog_print_tic_res(struct xfs_mount *mp, struct xlog_ticket *ticket);void xlog_print_trans (struct xfs_trans *);int xlog_write (struct xlog *log,struct xfs_log_vec *log_vector,struct xlog_ticket *tic,xfs_lsn_t *start_lsn, -struct xlog_in_core **commit_iclog,uint optype); +struct xlog_in_core **commit_iclog,uint optype,uint32_t len);int xlog_commit_record (struct xlog *log,struct xlog_ticket *ticket,struct xlog_in_core **iclog,xfs_lsn_t *lsn);Signature · 2 lines
Dave Chinner [email protected]
PATCH v4 21/39 xfs: introduce xlog_write_single()
In reply to Dave ChinnerFrom: Dave Chinner <[email protected]>
Introduce an optimised version of xlog_write() that is used when the entire write will fit in a single iclog. This greatly simplifies the implementation of writing a log vector chain into an iclog, and sets the ground work for a much more understandable xlog_write() implementation.
--- fs/xfs/xfs_log.c | 57 +++++++++++++++++++++++++++++++++++++++++++++++- 1 file changed, 56 insertions(+), 1 deletion(-)
fs/xfs/xfs_log.c +56 -1
@@ -2225,6 +2225,52 @@ xlog_write_copy_finish(return error; } +/* +* Write log vectors into a single iclog which is guaranteed by the caller +* to have enough space to write the entire log vector into. Return the number +* of log vectors written into the iclog. +*/ +static int +xlog_write_single ( +struct xfs_log_vec *log_vector, +struct xlog_ticket *ticket, +struct xlog_in_core *iclog, +uint32_t log_offset, +uint32_t len) +{ +struct xfs_log_vec *lv; +void *ptr; +int index =0 ; +int record_cnt =0 ; + +ASSERT (log_offset + len <= iclog->ic_size); + + ptr = iclog->ic_datap + log_offset; +for (lv = log_vector; lv; lv = lv->lv_next) { +/* +* Ordered log vectors have no regions to write so this +* loop will naturally skip them. +*/ +for (index =0 ; index < lv->lv_niovecs; index++) { +struct xfs_log_iovec *reg = &lv->lv_iovecp[index]; +struct xlog_op_header *ophdr = reg->i_addr; + +ASSERT (reg->i_len %sizeof (int32_t ) ==0 ); +ASSERT ((unsigned long )ptr %sizeof (int32_t ) ==0 ); + + ophdr->oh_tid =cpu_to_be32 (ticket->t_tid); + ophdr->oh_len =cpu_to_be32 (reg->i_len - +sizeof (struct xlog_op_header )); +memcpy (ptr, reg->i_addr, reg->i_len); +xlog_write_adv_cnt (&ptr, &len, &log_offset, reg->i_len); + record_cnt++; + } + } +ASSERT (len ==0 ); +return record_cnt; +} + +/* * Write some region out to in-core log *
@@ -2305,16 +2351,25 @@ xlog_write(return error;ASSERT (log_offset <= iclog->ic_size -1 ); - ptr = iclog->ic_datap + log_offset;/* Start_lsn is the first lsn written to. */ if (start_lsn && !*start_lsn) *start_lsn =be64_to_cpu (iclog->ic_header.h_lsn); +/* If this is a single iclog write, go fast... */ +if (!contwr && lv == log_vector) { + record_cnt =xlog_write_single (lv, ticket, iclog, + log_offset, len); + len =0 ; + data_cnt = len; +break ; + } +/* * This loop writes out as many regions as can fit in the amount * of space which was allocated by xlog_state_get_iclog_space(). */ + ptr = iclog->ic_datap + log_offset;while (lv && (!lv->lv_niovecs || index < lv->lv_niovecs)) {struct xfs_log_iovec *reg;struct xlog_op_header *ophdr;
Signature · 1 line
2.31.1
Re: [PATCH 21/39] xfs: introduce xlog_write_single()
Quoted 9 lines · Dave Chinner
From: Dave Chinner <[email protected]>
Introduce an optimised version of
xlog_write()that is used when the entire write will fit in a single iclog. This greatly simplifies the implementation of writing a log vector chain into an iclog, and sets the ground work for a much more understandablexlog_write()implementation.Signed-off-by Dave Chinner <[email protected]>
Looks like a fairly simple hoist. I might have more comments once I wade through the next patch.
--D
Quoted 91 lines
--- fs/xfs/xfs_log.c | 57 +++++++++++++++++++++++++++++++++++++++++++++++- 1 file changed, 56 insertions(+), 1 deletion(-)fs/xfs/xfs_log.c +56
-1@@ -2225,6 +2225,52 @@ xlog_write_copy_finish(return error; } +/* +* Write log vectors into a single iclog which is guaranteed by the caller +* to have enough space to write the entire log vector into. Return the number +* of log vectors written into the iclog. +*/ +static int +xlog_write_single ( +struct xfs_log_vec *log_vector, +struct xlog_ticket *ticket, +struct xlog_in_core *iclog, +uint32_t log_offset, +uint32_t len) +{ +struct xfs_log_vec *lv; +void *ptr; +int index =0 ; +int record_cnt =0 ; + +ASSERT (log_offset + len <= iclog->ic_size); + + ptr = iclog->ic_datap + log_offset; +for (lv = log_vector; lv; lv = lv->lv_next) { +/* +* Ordered log vectors have no regions to write so this +* loop will naturally skip them. +*/ +for (index =0 ; index < lv->lv_niovecs; index++) { +struct xfs_log_iovec *reg = &lv->lv_iovecp[index]; +struct xlog_op_header *ophdr = reg->i_addr; + +ASSERT (reg->i_len %sizeof (int32_t ) ==0 ); +ASSERT ((unsigned long )ptr %sizeof (int32_t ) ==0 ); + + ophdr->oh_tid =cpu_to_be32 (ticket->t_tid); + ophdr->oh_len =cpu_to_be32 (reg->i_len - +sizeof (struct xlog_op_header )); +memcpy (ptr, reg->i_addr, reg->i_len); +xlog_write_adv_cnt (&ptr, &len, &log_offset, reg->i_len); + record_cnt++; + } + } +ASSERT (len ==0 ); +return record_cnt; +} + +/* * Write some region out to in-core log * @@ -2305,16 +2351,25 @@ xlog_write(return error;ASSERT (log_offset <= iclog->ic_size -1 ); - ptr = iclog->ic_datap + log_offset;/* Start_lsn is the first lsn written to. */ if (start_lsn && !*start_lsn) *start_lsn =be64_to_cpu (iclog->ic_header.h_lsn); +/* If this is a single iclog write, go fast... */ +if (!contwr && lv == log_vector) { + record_cnt =xlog_write_single (lv, ticket, iclog, + log_offset, len); + len =0 ; + data_cnt = len; +break ; + } +/* * This loop writes out as many regions as can fit in the amount * of space which was allocated by xlog_state_get_iclog_space(). */ + ptr = iclog->ic_datap + log_offset;while (lv && (!lv->lv_niovecs || index < lv->lv_niovecs)) {struct xfs_log_iovec *reg;struct xlog_op_header *ophdr;Signature · 1 line
2.31.1
PATCH v4 22/39 xfs:_introduce xlog_write_partial()
In reply to Dave ChinnerFrom: Dave Chinner <[email protected]>
Handle writing of a logvec chain into an iclog that doesn't have enough space to fit it all. The iclog has already been changed to WANT_SYNC by xlog_get_iclog_space(), so the entire remaining space in the iclog is exclusively owned by this logvec chain.
The difference between the single and partial cases is that we end up with partial iovec writes in the iclog and have to split a log vec regions across two iclogs. The state handling for this is currently awful and so we're building up the pieces needed to handle this more cleanly one at a time.
--- fs/xfs/xfs_log.c | 504 ++++++++++++++++++++++------------------------- 1 file changed, 240 insertions(+), 264 deletions(-)
fs/xfs/xfs_log.c +240 -264
@@ -2110,166 +2110,250 @@ xlog_print_trans( } } -static xlog_op_header_t * -xlog_write_setup_ophdr ( -struct xlog_op_header *ophdr, -struct xlog_ticket *ticket) -{ - ophdr->oh_clientid =XFS_TRANSACTION ; - ophdr->oh_res2 =0 ; - ophdr->oh_flags =0 ; -return ophdr; -} -/* -* Set up the parameters of the region copy into the log. This has -* to handle region write split across multiple log buffers - this -* state is kept external to this function so that this code can -* be written in an obvious, self documenting manner. +* Write whole log vectors into a single iclog which is guaranteed to have +* either sufficient space for the entire log vector chain to be written or +* exclusive access to the remaining space in the iclog. +* +* Return the number of iovecs and data written into the iclog, as well as +* a pointer to the logvec that doesn't fit in the log (or NULL if we hit the +* end of the chain. */ -static int -xlog_write_setup_copy ( +static struct xfs_log_vec * +xlog_write_single ( +struct xfs_log_vec *log_vector,struct xlog_ticket *ticket, -struct xlog_op_header *ophdr, -int space_available, -int space_required, -int *copy_off, -int *copy_len, -int *last_was_partial_copy, -int *bytes_consumed) -{ -int still_to_copy; - - still_to_copy = space_required - *bytes_consumed; - *copy_off = *bytes_consumed; - -if (still_to_copy <= space_available) { -/* write of region completes here */ - *copy_len = still_to_copy; - ophdr->oh_len =cpu_to_be32 (*copy_len); -if (*last_was_partial_copy) - ophdr->oh_flags |= (XLOG_END_TRANS |XLOG_WAS_CONT_TRANS ); - *last_was_partial_copy =0 ; - *bytes_consumed =0 ; -return 0 ; - } - -/* partial write of region, needs extra log op header reservation */ - *copy_len = space_available; - ophdr->oh_len =cpu_to_be32 (*copy_len); - ophdr->oh_flags |=XLOG_CONTINUE_TRANS ; -if (*last_was_partial_copy) - ophdr->oh_flags |=XLOG_WAS_CONT_TRANS ; - *bytes_consumed += *copy_len; - (*last_was_partial_copy)++; - -/* account for new log op header */ - ticket->t_curr_res -=sizeof (struct xlog_op_header ); - -return sizeof (struct xlog_op_header ); -} - -static int -xlog_write_copy_finish ( -struct xlog *log,struct xlog_in_core *iclog, -uint flags, -int *record_cnt, -int *data_cnt, -int *partial_copy, -int *partial_copy_len, -int log_offset, -struct xlog_in_core **commit_iclog) +uint32_t *log_offset, +uint32_t *len, +uint32_t *record_cnt, +uint32_t *data_cnt) { -int error; +struct xfs_log_vec *lv; +void *ptr; +int index; + +ASSERT (*log_offset + *len <= iclog->ic_size || + iclog->ic_state ==XLOG_STATE_WANT_SYNC ); -if (*partial_copy) { + ptr = iclog->ic_datap + *log_offset; +for (lv = log_vector; lv; lv = lv->lv_next) {/* -* This iclog has already been marked WANT_SYNC by -* xlog_state_get_iclog_space. +* If the entire log vec does not fit in the iclog, punt it to +* the partial copy loop which can handle this case. */ -spin_lock (&log->l_icloglock); -xlog_state_finish_copy (log, iclog, *record_cnt, *data_cnt); - *record_cnt =0 ; - *data_cnt =0 ; - goto release_iclog; - } +if (lv->lv_niovecs && + lv->lv_bytes > iclog->ic_size - *log_offset) +break ; - *partial_copy =0 ; - *partial_copy_len =0 ; +/* +* Ordered log vectors have no regions to write so this +* loop will naturally skip them. +*/ +for (index =0 ; index < lv->lv_niovecs; index++) { +struct xfs_log_iovec *reg = &lv->lv_iovecp[index]; +struct xlog_op_header *ophdr = reg->i_addr; -if (iclog->ic_size - log_offset <=sizeof (xlog_op_header_t)) { -/* no more space in this iclog - push it. */ -spin_lock (&log->l_icloglock); -xlog_state_finish_copy (log, iclog, *record_cnt, *data_cnt); - *record_cnt =0 ; - *data_cnt =0 ; +ASSERT (reg->i_len %sizeof (int32_t ) ==0 ); +ASSERT ((unsigned long )ptr %sizeof (int32_t ) ==0 ); -if (iclog->ic_state ==XLOG_STATE_ACTIVE ) -xlog_state_switch_iclogs (log, iclog,0 ); -else -ASSERT (iclog->ic_state ==XLOG_STATE_WANT_SYNC || - iclog->ic_state ==XLOG_STATE_IOERROR ); -if (!commit_iclog) - goto release_iclog; -spin_unlock (&log->l_icloglock); -ASSERT (flags &XLOG_COMMIT_TRANS ); - *commit_iclog = iclog; + ophdr->oh_tid =cpu_to_be32 (ticket->t_tid); + ophdr->oh_len =cpu_to_be32 (reg->i_len - +sizeof (struct xlog_op_header )); +memcpy (ptr, reg->i_addr, reg->i_len); +xlog_write_adv_cnt (&ptr, len, log_offset, reg->i_len); + (*record_cnt)++; + *data_cnt += reg->i_len; + } } +ASSERT (*len ==0 || lv); +return lv; +} -return 0 ; +static int +xlog_write_get_more_iclog_space ( +struct xlog *log, +struct xlog_ticket *ticket, +struct xlog_in_core **iclogp, +uint32_t *log_offset, +uint32_t len, +uint32_t *record_cnt, +uint32_t *data_cnt, +int *contwr) +{ +struct xlog_in_core *iclog = *iclogp; +int error; -release_iclog: +spin_lock (&log->l_icloglock); +xlog_state_finish_copy (log, iclog, *record_cnt, *data_cnt); +ASSERT (iclog->ic_state ==XLOG_STATE_WANT_SYNC || + iclog->ic_state ==XLOG_STATE_IOERROR ); error =xlog_state_release_iclog (log, iclog);spin_unlock (&log->l_icloglock); -return error; +if (error) +return error; + + error =xlog_state_get_iclog_space (log, len, &iclog, + ticket, contwr, log_offset); +if (error) +return error; + *record_cnt =0 ; + *data_cnt =0 ; + *iclogp = iclog; +return 0 ; }/* -* Write log vectors into a single iclog which is guaranteed by the caller -* to have enough space to write the entire log vector into. Return the number -* of log vectors written into the iclog. +* Write log vectors into a single iclog which is smaller than the current chain +* length. We write until we cannot fit a full record into the remaining space +* and then stop. We return the log vector that is to be written that cannot +* wholly fit in the iclog. */ -static int -xlog_write_single ( +static struct xfs_log_vec * +xlog_write_partial ( +struct xlog *log,struct xfs_log_vec *log_vector,struct xlog_ticket *ticket, -struct xlog_in_core *iclog, -uint32_t log_offset, -uint32_t len) +struct xlog_in_core **iclogp, +uint32_t *log_offset, +uint32_t *len, +uint32_t *record_cnt, +uint32_t *data_cnt, +int *contwr) { -struct xfs_log_vec *lv; +struct xlog_in_core *iclog = *iclogp; +struct xfs_log_vec *lv = log_vector; +struct xfs_log_iovec *reg; +struct xlog_op_header *ophdr;void *ptr;int index =0 ; -int record_cnt =0 ; +uint32_t rlen; +int error; -ASSERT (log_offset + len <= iclog->ic_size); +/* walk the logvec, copying until we run out of space in the iclog */ + ptr = iclog->ic_datap + *log_offset; +for (index =0 ; index < lv->lv_niovecs; index++) { +uint32_t reg_offset =0 ; + + reg = &lv->lv_iovecp[index]; +ASSERT (reg->i_len %sizeof (int32_t ) ==0 ); - ptr = iclog->ic_datap + log_offset; -for (lv = log_vector; lv; lv = lv->lv_next) {/* -* Ordered log vectors have no regions to write so this -* loop will naturally skip them. +* The first region of a continuation must have a non-zero +* length otherwise log recovery will just skip over it and +* start recovering from the next opheader it finds. Because we +* mark the next opheader as a continuation, recovery will then +* incorrectly add the continuation to the previous region and +* that breaks stuff. +* +* Hence if there isn't space for region data after the +* opheader, then we need to start afresh with a new iclog. */ -for (index =0 ; index < lv->lv_niovecs; index++) { -struct xfs_log_iovec *reg = &lv->lv_iovecp[index]; -struct xlog_op_header *ophdr = reg->i_addr; +if (iclog->ic_size - *log_offset <= +sizeof (struct xlog_op_header )) { + error =xlog_write_get_more_iclog_space (log, ticket, + &iclog, log_offset, *len, record_cnt, + data_cnt, contwr); +if (error) +return ERR_PTR (error); + ptr = iclog->ic_datap + *log_offset; + } -ASSERT (reg->i_len %sizeof (int32_t ) ==0 ); -ASSERT ((unsigned long )ptr %sizeof (int32_t ) ==0 ); + ophdr = reg->i_addr; + rlen =min_t (uint32_t, reg->i_len, iclog->ic_size - *log_offset); + + ophdr->oh_tid =cpu_to_be32 (ticket->t_tid); + ophdr->oh_len =cpu_to_be32 (rlen -sizeof (struct xlog_op_header )); +if (rlen != reg->i_len) + ophdr->oh_flags |=XLOG_CONTINUE_TRANS ; + +ASSERT ((unsigned long )ptr %sizeof (int32_t ) ==0 ); +xlog_verify_dest_ptr (log, ptr); +memcpy (ptr, reg->i_addr, rlen); +xlog_write_adv_cnt (&ptr, len, log_offset, rlen); + (*record_cnt)++; + *data_cnt += rlen; + +/* If we wrote the whole region, move to the next. */ +if (rlen == reg->i_len) +continue ; + +/* +* We now have a partially written iovec, but it can span +* multiple iclogs so we loop here. First we release the iclog +* we currently have, then we get a new iclog and add a new +* opheader. Then we continue copying from where we were until +* we either complete the iovec or fill the iclog. If we +* complete the iovec, then we increment the index and go right +* back to the top of the outer loop. if we fill the iclog, we +* run the inner loop again. +* +* This is complicated by the tail of a region using all the +* space in an iclog and hence requiring us to release the iclog +* and get a new one before returning to the outer loop. We must +* always guarantee that we exit this inner loop with at least +* space for log transaction opheaders left in the current +* iclog, hence we cannot just terminate the loop at the end +* of the of the continuation. So we loop while there is no +* space left in the current iclog, and check for the end of the +* continuation after getting a new iclog. +*/ +do { +/* +* Account for the continuation opheader before we get +* a new iclog. This is necessary so that we reserve +* space in the iclog for it. +*/ + *len +=sizeof (struct xlog_op_header ); + ticket->t_curr_res -=sizeof (struct xlog_op_header ); + + error =xlog_write_get_more_iclog_space (log, ticket, + &iclog, log_offset, *len, record_cnt, + data_cnt, contwr); +if (error) +return ERR_PTR (error); + ptr = iclog->ic_datap + *log_offset; + ophdr = ptr; ophdr->oh_tid =cpu_to_be32 (ticket->t_tid); - ophdr->oh_len =cpu_to_be32 (reg->i_len - + ophdr->oh_clientid =XFS_TRANSACTION ; + ophdr->oh_res2 =0 ; + ophdr->oh_flags =XLOG_WAS_CONT_TRANS ; + +xlog_write_adv_cnt (&ptr, len, log_offset,sizeof (struct xlog_op_header )); -memcpy (ptr, reg->i_addr, reg->i_len); -xlog_write_adv_cnt (&ptr, &len, &log_offset, reg->i_len); - record_cnt++; - } + *data_cnt +=sizeof (struct xlog_op_header ); + +/* +* If rlen fits in the iclog, then end the region +* continuation. Otherwise we're going around again. +*/ + reg_offset += rlen; + rlen = reg->i_len - reg_offset; +if (rlen <= iclog->ic_size - *log_offset) + ophdr->oh_flags |=XLOG_END_TRANS ; +else + ophdr->oh_flags |=XLOG_CONTINUE_TRANS ; + + rlen =min_t (uint32_t, rlen, iclog->ic_size - *log_offset); + ophdr->oh_len =cpu_to_be32 (rlen); + +xlog_verify_dest_ptr (log, ptr); +memcpy (ptr, reg->i_addr + reg_offset, rlen); +xlog_write_adv_cnt (&ptr, len, log_offset, rlen); + (*record_cnt)++; + *data_cnt += rlen; + + }while (ophdr->oh_flags &XLOG_CONTINUE_TRANS ); } -ASSERT (len ==0 ); -return record_cnt; -} +/* +* No more iovecs remain in this logvec so return the next log vec to +* the caller so it can go back to fast path copying. +*/ + *iclogp = iclog; +return lv->lv_next; +}/* * Write some region out to in-core log
@@ -2323,14 +2407,11 @@ xlog_write( {struct xlog_in_core *iclog =NULL ;struct xfs_log_vec *lv = log_vector; -struct xfs_log_iovec *vecp = lv->lv_iovecp; -int index =0 ; -int partial_copy =0 ; -int partial_copy_len =0 ;int contwr =0 ;int record_cnt =0 ;int data_cnt =0 ;int error =0 ; +int log_offset;if (ticket->t_curr_res <0 ) { xfs_alert_tag(log->l_mp, XFS_PTAG_LOGRES,
@@ -2339,146 +2420,40 @@ xlog_write(xfs_force_shutdown (log->l_mp,SHUTDOWN_LOG_IO_ERROR ); } -if (start_lsn) - *start_lsn =0 ; -while (lv && (!lv->lv_niovecs || index < lv->lv_niovecs)) { -void *ptr; -int log_offset; - - error =xlog_state_get_iclog_space (log, len, &iclog, ticket, - &contwr, &log_offset); -if (error) -return error; - -ASSERT (log_offset <= iclog->ic_size -1 ); + error =xlog_state_get_iclog_space (log, len, &iclog, ticket, + &contwr, &log_offset); +if (error) +return error; -/* Start_lsn is the first lsn written to. */ -if (start_lsn && !*start_lsn) - *start_lsn =be64_to_cpu (iclog->ic_header.h_lsn); +/* start_lsn is the LSN of the first iclog written to. */ +if (start_lsn) + *start_lsn =be64_to_cpu (iclog->ic_header.h_lsn); -/* If this is a single iclog write, go fast... */ -if (!contwr && lv == log_vector) { - record_cnt =xlog_write_single (lv, ticket, iclog, - log_offset, len); - len =0 ; - data_cnt = len; +while (lv) { + lv =xlog_write_single (lv, ticket, iclog, &log_offset, + &len, &record_cnt, &data_cnt); +if (!lv)break ; - } - -/* -* This loop writes out as many regions as can fit in the amount -* of space which was allocated by xlog_state_get_iclog_space(). -*/ - ptr = iclog->ic_datap + log_offset; -while (lv && (!lv->lv_niovecs || index < lv->lv_niovecs)) { -struct xfs_log_iovec *reg; -struct xlog_op_header *ophdr; -int copy_len; -int copy_off; -bool ordered = false; -bool added_ophdr = false; - -/* ordered log vectors have no regions to write */ -if (lv->lv_buf_len ==XFS_LOG_VEC_ORDERED ) { -ASSERT (lv->lv_niovecs ==0 ); - ordered = true; - goto next_lv; - } - - reg = &vecp[index]; -ASSERT (reg->i_len %sizeof (int32_t ) ==0 ); -ASSERT ((unsigned long )ptr %sizeof (int32_t ) ==0 ); - -/* -* Regions always have their ophdr at the start of the -* region, except for: -* - a transaction start which has a start record ophdr -* before the first region ophdr; and -* - the previous region didn't fully fit into an iclog -* so needs a continuation ophdr to prepend the region -* in this new iclog. -*/ - ophdr = reg->i_addr; -if (optype && index) { - optype &= ~XLOG_START_TRANS ; - }else if (partial_copy) { - ophdr =xlog_write_setup_ophdr (ptr, ticket); -xlog_write_adv_cnt (&ptr, &len, &log_offset, -sizeof (struct xlog_op_header )); - added_ophdr = true; - } - ophdr->oh_tid =cpu_to_be32 (ticket->t_tid); - - len +=xlog_write_setup_copy (ticket, ophdr, - iclog->ic_size-log_offset, - reg->i_len, - ©_off, ©_len, - &partial_copy, - &partial_copy_len); -xlog_verify_dest_ptr (log, ptr); - -/* -* Wart: need to update length in embedded ophdr not -* to include it's own length. -*/ -if (!added_ophdr) { - ophdr->oh_len =cpu_to_be32 (copy_len - -sizeof (struct xlog_op_header )); - } - -ASSERT (copy_len >0 ); -memcpy (ptr, reg->i_addr + copy_off, copy_len); -xlog_write_adv_cnt (&ptr, &len, &log_offset, copy_len); - -if (added_ophdr) - copy_len +=sizeof (struct xlog_op_header ); - record_cnt++; - data_cnt += contwr ? copy_len :0 ; - - error =xlog_write_copy_finish (log, iclog, optype, - &record_cnt, &data_cnt, - &partial_copy, - &partial_copy_len, - log_offset, - commit_iclog); -if (error) -return error; - -/* -* if we had a partial copy, we need to get more iclog -* space but we don't want to increment the region -* index because there is still more is this region to -* write. -* -* If we completed writing this region, and we flushed -* the iclog (indicated by resetting of the record -* count), then we also need to get more log space. If -* this was the last record, though, we are done and -* can just return. -*/ -if (partial_copy) -break ; - -if (++index == lv->lv_niovecs) { -next_lv: - lv = lv->lv_next; - index =0 ; -if (lv) - vecp = lv->lv_iovecp; - } -if (record_cnt ==0 && !ordered) { -if (!lv) -return 0 ; -break ; - } +ASSERT (!(optype & (XLOG_COMMIT_TRANS |XLOG_UNMOUNT_TRANS ))); + lv =xlog_write_partial (log, lv, ticket, &iclog, &log_offset, + &len, &record_cnt, &data_cnt, &contwr); +if (IS_ERR_OR_NULL (lv)) { + error =PTR_ERR_OR_ZERO (lv); +break ; } } +ASSERT ((len ==0 && !lv) || error); -ASSERT (len ==0 ); - +/* +* We've already been guaranteed that the last writes will fit inside +* the current iclog, and hence it will already have the space used by +* those writes accounted to it. Hence we do not need to update the +* iclog with the number of bytes written here. +*/ +ASSERT (!contwr ||XLOG_FORCED_SHUTDOWN (log));spin_lock (&log->l_icloglock); -xlog_state_finish_copy (log, iclog, record_cnt, data_cnt); +xlog_state_finish_copy (log, iclog, record_cnt,0 );if (commit_iclog) {ASSERT (optype &XLOG_COMMIT_TRANS ); *commit_iclog = iclog;
@@ -3633,11 +3608,12 @@ xlog_verify_iclog( iclog->ic_header.h_cycle_data[idx]); } } -if (clientid !=XFS_TRANSACTION && clientid !=XFS_LOG ) +if (clientid !=XFS_TRANSACTION && clientid !=XFS_LOG ) {xfs_warn (log->l_mp, -"%s: invalid clientid %d op " PTR_FMT " offset 0x%lx" , - __func__, clientid, ophead, +"%s: op %d invalid clientid %d op " PTR_FMT " offset 0x%lx" , + __func__, i, clientid, ophead, (unsigned long )field_offset); + }/* check length */ p = &ophead->oh_len;
Signature · 1 line
2.31.1
Re: [PATCH 22/39] xfs:_introduce xlog_write_partial()
Quoted 7 lines · Dave Chinner
From: Dave Chinner <[email protected]>
Handle writing of a logvec chain into an iclog that doesn't have enough space to fit it all. The iclog has already been changed to
WANT_SYNCbyxlog_get_iclog_space(), so the entire remaining space in the iclog is exclusively owned by this logvec chain.
Quoted 7 lines
The difference between the single and partial cases is that we end up with partial iovec writes in the iclog and have to split a log vec regions across two iclogs. The state handling for this is currently awful and so we're building up the pieces needed to handle this more cleanly one at a time.
Signed-off-by Dave Chinner <[email protected]>
Egad this diff is hard to read. Brian's right, the patience diff is easier to understand and shorter to boot.
That said, I actually understand what the new code does now, so:
Might be nice to hoist:
memcpy(ptr, reg->i_addr + reg_offset, rlen); xlog_write_adv_cnt(&ptr, len, log_offset, rlen); (*record_cnt)++; *data_cnt += rlen;
into a helper but it's only four lines so I'm not gonna fuss any further.
--D
Quoted 592 lines
--- fs/xfs/xfs_log.c | 504 ++++++++++++++++++++++------------------------- 1 file changed, 240 insertions(+), 264 deletions(-)fs/xfs/xfs_log.c +240
-264@@ -2110,166 +2110,250 @@ xlog_print_trans( } } -static xlog_op_header_t * -xlog_write_setup_ophdr ( -struct xlog_op_header *ophdr, -struct xlog_ticket *ticket) -{ - ophdr->oh_clientid =XFS_TRANSACTION ; - ophdr->oh_res2 =0 ; - ophdr->oh_flags =0 ; -return ophdr; -} -/* -* Set up the parameters of the region copy into the log. This has -* to handle region write split across multiple log buffers - this -* state is kept external to this function so that this code can -* be written in an obvious, self documenting manner. +* Write whole log vectors into a single iclog which is guaranteed to have +* either sufficient space for the entire log vector chain to be written or +* exclusive access to the remaining space in the iclog. +* +* Return the number of iovecs and data written into the iclog, as well as +* a pointer to the logvec that doesn't fit in the log (or NULL if we hit the +* end of the chain. */ -static int -xlog_write_setup_copy ( +static struct xfs_log_vec * +xlog_write_single ( +struct xfs_log_vec *log_vector,struct xlog_ticket *ticket, -struct xlog_op_header *ophdr, -int space_available, -int space_required, -int *copy_off, -int *copy_len, -int *last_was_partial_copy, -int *bytes_consumed) -{ -int still_to_copy; - - still_to_copy = space_required - *bytes_consumed; - *copy_off = *bytes_consumed; - -if (still_to_copy <= space_available) { -/* write of region completes here */ - *copy_len = still_to_copy; - ophdr->oh_len =cpu_to_be32 (*copy_len); -if (*last_was_partial_copy) - ophdr->oh_flags |= (XLOG_END_TRANS |XLOG_WAS_CONT_TRANS ); - *last_was_partial_copy =0 ; - *bytes_consumed =0 ; -return 0 ; - } - -/* partial write of region, needs extra log op header reservation */ - *copy_len = space_available; - ophdr->oh_len =cpu_to_be32 (*copy_len); - ophdr->oh_flags |=XLOG_CONTINUE_TRANS ; -if (*last_was_partial_copy) - ophdr->oh_flags |=XLOG_WAS_CONT_TRANS ; - *bytes_consumed += *copy_len; - (*last_was_partial_copy)++; - -/* account for new log op header */ - ticket->t_curr_res -=sizeof (struct xlog_op_header ); - -return sizeof (struct xlog_op_header ); -} - -static int -xlog_write_copy_finish ( -struct xlog *log,struct xlog_in_core *iclog, -uint flags, -int *record_cnt, -int *data_cnt, -int *partial_copy, -int *partial_copy_len, -int log_offset, -struct xlog_in_core **commit_iclog) +uint32_t *log_offset, +uint32_t *len, +uint32_t *record_cnt, +uint32_t *data_cnt) { -int error; +struct xfs_log_vec *lv; +void *ptr; +int index; + +ASSERT (*log_offset + *len <= iclog->ic_size || + iclog->ic_state ==XLOG_STATE_WANT_SYNC ); -if (*partial_copy) { + ptr = iclog->ic_datap + *log_offset; +for (lv = log_vector; lv; lv = lv->lv_next) {/* -* This iclog has already been marked WANT_SYNC by -* xlog_state_get_iclog_space. +* If the entire log vec does not fit in the iclog, punt it to +* the partial copy loop which can handle this case. */ -spin_lock (&log->l_icloglock); -xlog_state_finish_copy (log, iclog, *record_cnt, *data_cnt); - *record_cnt =0 ; - *data_cnt =0 ; - goto release_iclog; - } +if (lv->lv_niovecs && + lv->lv_bytes > iclog->ic_size - *log_offset) +break ; - *partial_copy =0 ; - *partial_copy_len =0 ; +/* +* Ordered log vectors have no regions to write so this +* loop will naturally skip them. +*/ +for (index =0 ; index < lv->lv_niovecs; index++) { +struct xfs_log_iovec *reg = &lv->lv_iovecp[index]; +struct xlog_op_header *ophdr = reg->i_addr; -if (iclog->ic_size - log_offset <=sizeof (xlog_op_header_t)) { -/* no more space in this iclog - push it. */ -spin_lock (&log->l_icloglock); -xlog_state_finish_copy (log, iclog, *record_cnt, *data_cnt); - *record_cnt =0 ; - *data_cnt =0 ; +ASSERT (reg->i_len %sizeof (int32_t ) ==0 ); +ASSERT ((unsigned long )ptr %sizeof (int32_t ) ==0 ); -if (iclog->ic_state ==XLOG_STATE_ACTIVE ) -xlog_state_switch_iclogs (log, iclog,0 ); -else -ASSERT (iclog->ic_state ==XLOG_STATE_WANT_SYNC || - iclog->ic_state ==XLOG_STATE_IOERROR ); -if (!commit_iclog) - goto release_iclog; -spin_unlock (&log->l_icloglock); -ASSERT (flags &XLOG_COMMIT_TRANS ); - *commit_iclog = iclog; + ophdr->oh_tid =cpu_to_be32 (ticket->t_tid); + ophdr->oh_len =cpu_to_be32 (reg->i_len - +sizeof (struct xlog_op_header )); +memcpy (ptr, reg->i_addr, reg->i_len); +xlog_write_adv_cnt (&ptr, len, log_offset, reg->i_len); + (*record_cnt)++; + *data_cnt += reg->i_len; + } } +ASSERT (*len ==0 || lv); +return lv; +} -return 0 ; +static int +xlog_write_get_more_iclog_space ( +struct xlog *log, +struct xlog_ticket *ticket, +struct xlog_in_core **iclogp, +uint32_t *log_offset, +uint32_t len, +uint32_t *record_cnt, +uint32_t *data_cnt, +int *contwr) +{ +struct xlog_in_core *iclog = *iclogp; +int error; -release_iclog: +spin_lock (&log->l_icloglock); +xlog_state_finish_copy (log, iclog, *record_cnt, *data_cnt); +ASSERT (iclog->ic_state ==XLOG_STATE_WANT_SYNC || + iclog->ic_state ==XLOG_STATE_IOERROR ); error =xlog_state_release_iclog (log, iclog);spin_unlock (&log->l_icloglock); -return error; +if (error) +return error; + + error =xlog_state_get_iclog_space (log, len, &iclog, + ticket, contwr, log_offset); +if (error) +return error; + *record_cnt =0 ; + *data_cnt =0 ; + *iclogp = iclog; +return 0 ; }/* -* Write log vectors into a single iclog which is guaranteed by the caller -* to have enough space to write the entire log vector into. Return the number -* of log vectors written into the iclog. +* Write log vectors into a single iclog which is smaller than the current chain +* length. We write until we cannot fit a full record into the remaining space +* and then stop. We return the log vector that is to be written that cannot +* wholly fit in the iclog. */ -static int -xlog_write_single ( +static struct xfs_log_vec * +xlog_write_partial ( +struct xlog *log,struct xfs_log_vec *log_vector,struct xlog_ticket *ticket, -struct xlog_in_core *iclog, -uint32_t log_offset, -uint32_t len) +struct xlog_in_core **iclogp, +uint32_t *log_offset, +uint32_t *len, +uint32_t *record_cnt, +uint32_t *data_cnt, +int *contwr) { -struct xfs_log_vec *lv; +struct xlog_in_core *iclog = *iclogp; +struct xfs_log_vec *lv = log_vector; +struct xfs_log_iovec *reg; +struct xlog_op_header *ophdr;void *ptr;int index =0 ; -int record_cnt =0 ; +uint32_t rlen; +int error; -ASSERT (log_offset + len <= iclog->ic_size); +/* walk the logvec, copying until we run out of space in the iclog */ + ptr = iclog->ic_datap + *log_offset; +for (index =0 ; index < lv->lv_niovecs; index++) { +uint32_t reg_offset =0 ; + + reg = &lv->lv_iovecp[index]; +ASSERT (reg->i_len %sizeof (int32_t ) ==0 ); - ptr = iclog->ic_datap + log_offset; -for (lv = log_vector; lv; lv = lv->lv_next) {/* -* Ordered log vectors have no regions to write so this -* loop will naturally skip them. +* The first region of a continuation must have a non-zero +* length otherwise log recovery will just skip over it and +* start recovering from the next opheader it finds. Because we +* mark the next opheader as a continuation, recovery will then +* incorrectly add the continuation to the previous region and +* that breaks stuff. +* +* Hence if there isn't space for region data after the +* opheader, then we need to start afresh with a new iclog. */ -for (index =0 ; index < lv->lv_niovecs; index++) { -struct xfs_log_iovec *reg = &lv->lv_iovecp[index]; -struct xlog_op_header *ophdr = reg->i_addr; +if (iclog->ic_size - *log_offset <= +sizeof (struct xlog_op_header )) { + error =xlog_write_get_more_iclog_space (log, ticket, + &iclog, log_offset, *len, record_cnt, + data_cnt, contwr); +if (error) +return ERR_PTR (error); + ptr = iclog->ic_datap + *log_offset; + } -ASSERT (reg->i_len %sizeof (int32_t ) ==0 ); -ASSERT ((unsigned long )ptr %sizeof (int32_t ) ==0 ); + ophdr = reg->i_addr; + rlen =min_t (uint32_t, reg->i_len, iclog->ic_size - *log_offset); + + ophdr->oh_tid =cpu_to_be32 (ticket->t_tid); + ophdr->oh_len =cpu_to_be32 (rlen -sizeof (struct xlog_op_header )); +if (rlen != reg->i_len) + ophdr->oh_flags |=XLOG_CONTINUE_TRANS ; + +ASSERT ((unsigned long )ptr %sizeof (int32_t ) ==0 ); +xlog_verify_dest_ptr (log, ptr); +memcpy (ptr, reg->i_addr, rlen); +xlog_write_adv_cnt (&ptr, len, log_offset, rlen); + (*record_cnt)++; + *data_cnt += rlen; + +/* If we wrote the whole region, move to the next. */ +if (rlen == reg->i_len) +continue ; + +/* +* We now have a partially written iovec, but it can span +* multiple iclogs so we loop here. First we release the iclog +* we currently have, then we get a new iclog and add a new +* opheader. Then we continue copying from where we were until +* we either complete the iovec or fill the iclog. If we +* complete the iovec, then we increment the index and go right +* back to the top of the outer loop. if we fill the iclog, we +* run the inner loop again. +* +* This is complicated by the tail of a region using all the +* space in an iclog and hence requiring us to release the iclog +* and get a new one before returning to the outer loop. We must +* always guarantee that we exit this inner loop with at least +* space for log transaction opheaders left in the current +* iclog, hence we cannot just terminate the loop at the end +* of the of the continuation. So we loop while there is no +* space left in the current iclog, and check for the end of the +* continuation after getting a new iclog. +*/ +do { +/* +* Account for the continuation opheader before we get +* a new iclog. This is necessary so that we reserve +* space in the iclog for it. +*/ + *len +=sizeof (struct xlog_op_header ); + ticket->t_curr_res -=sizeof (struct xlog_op_header ); + + error =xlog_write_get_more_iclog_space (log, ticket, + &iclog, log_offset, *len, record_cnt, + data_cnt, contwr); +if (error) +return ERR_PTR (error); + ptr = iclog->ic_datap + *log_offset; + ophdr = ptr; ophdr->oh_tid =cpu_to_be32 (ticket->t_tid); - ophdr->oh_len =cpu_to_be32 (reg->i_len - + ophdr->oh_clientid =XFS_TRANSACTION ; + ophdr->oh_res2 =0 ; + ophdr->oh_flags =XLOG_WAS_CONT_TRANS ; + +xlog_write_adv_cnt (&ptr, len, log_offset,sizeof (struct xlog_op_header )); -memcpy (ptr, reg->i_addr, reg->i_len); -xlog_write_adv_cnt (&ptr, &len, &log_offset, reg->i_len); - record_cnt++; - } + *data_cnt +=sizeof (struct xlog_op_header ); + +/* +* If rlen fits in the iclog, then end the region +* continuation. Otherwise we're going around again. +*/ + reg_offset += rlen; + rlen = reg->i_len - reg_offset; +if (rlen <= iclog->ic_size - *log_offset) + ophdr->oh_flags |=XLOG_END_TRANS ; +else + ophdr->oh_flags |=XLOG_CONTINUE_TRANS ; + + rlen =min_t (uint32_t, rlen, iclog->ic_size - *log_offset); + ophdr->oh_len =cpu_to_be32 (rlen); + +xlog_verify_dest_ptr (log, ptr); +memcpy (ptr, reg->i_addr + reg_offset, rlen); +xlog_write_adv_cnt (&ptr, len, log_offset, rlen); + (*record_cnt)++; + *data_cnt += rlen; + + }while (ophdr->oh_flags &XLOG_CONTINUE_TRANS ); } -ASSERT (len ==0 ); -return record_cnt; -} +/* +* No more iovecs remain in this logvec so return the next log vec to +* the caller so it can go back to fast path copying. +*/ + *iclogp = iclog; +return lv->lv_next; +}/* * Write some region out to in-core log @@ -2323,14 +2407,11 @@ xlog_write( {struct xlog_in_core *iclog =NULL ;struct xfs_log_vec *lv = log_vector; -struct xfs_log_iovec *vecp = lv->lv_iovecp; -int index =0 ; -int partial_copy =0 ; -int partial_copy_len =0 ;int contwr =0 ;int record_cnt =0 ;int data_cnt =0 ;int error =0 ; +int log_offset;if (ticket->t_curr_res <0 ) { xfs_alert_tag(log->l_mp, XFS_PTAG_LOGRES,@@ -2339,146 +2420,40 @@ xlog_write(xfs_force_shutdown (log->l_mp,SHUTDOWN_LOG_IO_ERROR ); } -if (start_lsn) - *start_lsn =0 ; -while (lv && (!lv->lv_niovecs || index < lv->lv_niovecs)) { -void *ptr; -int log_offset; - - error =xlog_state_get_iclog_space (log, len, &iclog, ticket, - &contwr, &log_offset); -if (error) -return error; - -ASSERT (log_offset <= iclog->ic_size -1 ); + error =xlog_state_get_iclog_space (log, len, &iclog, ticket, + &contwr, &log_offset); +if (error) +return error; -/* Start_lsn is the first lsn written to. */ -if (start_lsn && !*start_lsn) - *start_lsn =be64_to_cpu (iclog->ic_header.h_lsn); +/* start_lsn is the LSN of the first iclog written to. */ +if (start_lsn) + *start_lsn =be64_to_cpu (iclog->ic_header.h_lsn); -/* If this is a single iclog write, go fast... */ -if (!contwr && lv == log_vector) { - record_cnt =xlog_write_single (lv, ticket, iclog, - log_offset, len); - len =0 ; - data_cnt = len; +while (lv) { + lv =xlog_write_single (lv, ticket, iclog, &log_offset, + &len, &record_cnt, &data_cnt); +if (!lv)break ; - } - -/* -* This loop writes out as many regions as can fit in the amount -* of space which was allocated by xlog_state_get_iclog_space(). -*/ - ptr = iclog->ic_datap + log_offset; -while (lv && (!lv->lv_niovecs || index < lv->lv_niovecs)) { -struct xfs_log_iovec *reg; -struct xlog_op_header *ophdr; -int copy_len; -int copy_off; -bool ordered = false; -bool added_ophdr = false; - -/* ordered log vectors have no regions to write */ -if (lv->lv_buf_len ==XFS_LOG_VEC_ORDERED ) { -ASSERT (lv->lv_niovecs ==0 ); - ordered = true; - goto next_lv; - } - - reg = &vecp[index]; -ASSERT (reg->i_len %sizeof (int32_t ) ==0 ); -ASSERT ((unsigned long )ptr %sizeof (int32_t ) ==0 ); - -/* -* Regions always have their ophdr at the start of the -* region, except for: -* - a transaction start which has a start record ophdr -* before the first region ophdr; and -* - the previous region didn't fully fit into an iclog -* so needs a continuation ophdr to prepend the region -* in this new iclog. -*/ - ophdr = reg->i_addr; -if (optype && index) { - optype &= ~XLOG_START_TRANS ; - }else if (partial_copy) { - ophdr =xlog_write_setup_ophdr (ptr, ticket); -xlog_write_adv_cnt (&ptr, &len, &log_offset, -sizeof (struct xlog_op_header )); - added_ophdr = true; - } - ophdr->oh_tid =cpu_to_be32 (ticket->t_tid); - - len +=xlog_write_setup_copy (ticket, ophdr, - iclog->ic_size-log_offset, - reg->i_len, - ©_off, ©_len, - &partial_copy, - &partial_copy_len); -xlog_verify_dest_ptr (log, ptr); - -/* -* Wart: need to update length in embedded ophdr not -* to include it's own length. -*/ -if (!added_ophdr) { - ophdr->oh_len =cpu_to_be32 (copy_len - -sizeof (struct xlog_op_header )); - } - -ASSERT (copy_len >0 ); -memcpy (ptr, reg->i_addr + copy_off, copy_len); -xlog_write_adv_cnt (&ptr, &len, &log_offset, copy_len); - -if (added_ophdr) - copy_len +=sizeof (struct xlog_op_header ); - record_cnt++; - data_cnt += contwr ? copy_len :0 ; - - error =xlog_write_copy_finish (log, iclog, optype, - &record_cnt, &data_cnt, - &partial_copy, - &partial_copy_len, - log_offset, - commit_iclog); -if (error) -return error; - -/* -* if we had a partial copy, we need to get more iclog -* space but we don't want to increment the region -* index because there is still more is this region to -* write. -* -* If we completed writing this region, and we flushed -* the iclog (indicated by resetting of the record -* count), then we also need to get more log space. If -* this was the last record, though, we are done and -* can just return. -*/ -if (partial_copy) -break ; - -if (++index == lv->lv_niovecs) { -next_lv: - lv = lv->lv_next; - index =0 ; -if (lv) - vecp = lv->lv_iovecp; - } -if (record_cnt ==0 && !ordered) { -if (!lv) -return 0 ; -break ; - } +ASSERT (!(optype & (XLOG_COMMIT_TRANS |XLOG_UNMOUNT_TRANS ))); + lv =xlog_write_partial (log, lv, ticket, &iclog, &log_offset, + &len, &record_cnt, &data_cnt, &contwr); +if (IS_ERR_OR_NULL (lv)) { + error =PTR_ERR_OR_ZERO (lv); +break ; } } +ASSERT ((len ==0 && !lv) || error); -ASSERT (len ==0 ); - +/* +* We've already been guaranteed that the last writes will fit inside +* the current iclog, and hence it will already have the space used by +* those writes accounted to it. Hence we do not need to update the +* iclog with the number of bytes written here. +*/ +ASSERT (!contwr ||XLOG_FORCED_SHUTDOWN (log));spin_lock (&log->l_icloglock); -xlog_state_finish_copy (log, iclog, record_cnt, data_cnt); +xlog_state_finish_copy (log, iclog, record_cnt,0 );if (commit_iclog) {ASSERT (optype &XLOG_COMMIT_TRANS ); *commit_iclog = iclog;@@ -3633,11 +3608,12 @@ xlog_verify_iclog( iclog->ic_header.h_cycle_data[idx]); } } -if (clientid !=XFS_TRANSACTION && clientid !=XFS_LOG ) +if (clientid !=XFS_TRANSACTION && clientid !=XFS_LOG ) {xfs_warn (log->l_mp, -"%s: invalid clientid %d op " PTR_FMT " offset 0x%lx" , - __func__, clientid, ophead, +"%s: op %d invalid clientid %d op " PTR_FMT " offset 0x%lx" , + __func__, i, clientid, ophead, (unsigned long )field_offset); + }/* check length */ p = &ophead->oh_len;Signature · 1 line
2.31.1
Re: [PATCH 22/39] xfs:_introduce xlog_write_partial()
Quoted 14 lines · Darrick J. Wong
Quoted 13 lines · Dave Chinner
From: Dave Chinner <[email protected]>
Handle writing of a logvec chain into an iclog that doesn't have enough space to fit it all. The iclog has already been changed to
WANT_SYNCbyxlog_get_iclog_space(), so the entire remaining space in the iclog is exclusively owned by this logvec chain.The difference between the single and partial cases is that we end up with partial iovec writes in the iclog and have to split a log vec regions across two iclogs. The state handling for this is currently awful and so we're building up the pieces needed to handle this more cleanly one at a time.
Quoted 8 lines
Quoted 1 line
Signed-off-by Dave Chinner <[email protected]>Egad this diff is hard to read. Brian's right, the patience diff is easier to understand and shorter to boot.
That said, I actually understand what the new code does now, so:
Reviewed-by Darrick J. Wong <[email protected]>
Thx!
Quoted 9 lines
Might be nice to hoist:
memcpy(ptr, reg->i_addr + reg_offset, rlen); xlog_write_adv_cnt(&ptr, len, log_offset, rlen); (*record_cnt)++; *data_cnt += rlen;into a helper but it's only four lines so I'm not gonna fuss any further.
Agreed, there are opportunities for further factoring and simplification of this code, but I'll leave that for another patchset rather than risking destabilisation at this late point.
-Dave.
Signature · 2 lines
Dave Chinner [email protected]
PATCH v4 23/39 xfs: xlog_write() no longer needs contwr state
In reply to Dave ChinnerFrom: Dave Chinner <[email protected]>
The rework of xlog_write() no longer requires xlog_get_iclog_state() to tell it about internal iclog space reservation state to direct it on what to do. Remove this parameter.
$ size fs/xfs/xfs_log.o.* text data bss dec hex filename 26520 560 8 27088 69d0 fs/xfs/xfs_log.o.orig 26384 560 8 26952 6948 fs/xfs/xfs_log.o.patched
--- fs/xfs/xfs_log.c | 29 ++++++++++------------------- 1 file changed, 10 insertions(+), 19 deletions(-)
fs/xfs/xfs_log.c +10 -19
@@ -47,7 +47,6 @@ xlog_state_get_iclog_space(int len, struct xlog_in_core **iclog, struct xlog_ticket *ticket, - int *continued_write, int *logoffsetp); STATIC void xlog_grant_push_ail(
@@ -2178,8 +2177,7 @@ xlog_write_get_more_iclog_space(uint32_t *log_offset, uint32_t len, uint32_t *record_cnt, - uint32_t *data_cnt, - int *contwr) + uint32_t *data_cnt) { struct xlog_in_core *iclog = *iclogp;int error;
@@ -2193,8 +2191,8 @@ xlog_write_get_more_iclog_space(if (error)return error; - error =xlog_state_get_iclog_space (log, len, &iclog, - ticket, contwr, log_offset); + error =xlog_state_get_iclog_space (log, len, &iclog, ticket, + log_offset);if (error)return error; *record_cnt =0 ;
@@ -2218,8 +2216,7 @@ xlog_write_partial(uint32_t *log_offset, uint32_t *len, uint32_t *record_cnt, - uint32_t *data_cnt, - int *contwr) + uint32_t *data_cnt) { struct xlog_in_core *iclog = *iclogp;struct xfs_log_vec *lv = log_vector;
@@ -2253,7 +2250,7 @@ xlog_write_partial( sizeof(struct xlog_op_header)) { error =xlog_write_get_more_iclog_space (log, ticket, &iclog, log_offset, *len, record_cnt, - data_cnt, contwr); + data_cnt);if (error)return ERR_PTR (error); ptr = iclog->ic_datap + *log_offset;
@@ -2309,7 +2306,7 @@ xlog_write_partial( error =xlog_write_get_more_iclog_space (log, ticket, &iclog, log_offset, *len, record_cnt, - data_cnt, contwr); + data_cnt);if (error)return ERR_PTR (error); ptr = iclog->ic_datap + *log_offset;
@@ -2407,7 +2404,6 @@ xlog_write( {struct xlog_in_core *iclog =NULL ;struct xfs_log_vec *lv = log_vector; -int contwr =0 ;int record_cnt =0 ;int data_cnt =0 ;int error =0 ;
@@ -2421,7 +2417,7 @@ xlog_write( } error =xlog_state_get_iclog_space (log, len, &iclog, ticket, - &contwr, &log_offset); + &log_offset);if (error)return error;
@@ -2437,7 +2433,7 @@ xlog_write(ASSERT (!(optype & (XLOG_COMMIT_TRANS |XLOG_UNMOUNT_TRANS ))); lv =xlog_write_partial (log, lv, ticket, &iclog, &log_offset, - &len, &record_cnt, &data_cnt, &contwr); + &len, &record_cnt, &data_cnt);if (IS_ERR_OR_NULL (lv)) { error =PTR_ERR_OR_ZERO (lv);break ;
@@ -2451,7 +2447,6 @@ xlog_write(* those writes accounted to it. Hence we do not need to update the * iclog with the number of bytes written here. */ -ASSERT (!contwr ||XLOG_FORCED_SHUTDOWN (log));spin_lock (&log->l_icloglock);xlog_state_finish_copy (log, iclog, record_cnt,0 );if (commit_iclog) {
@@ -2855,7 +2850,6 @@ xlog_state_get_iclog_space(int len, struct xlog_in_core **iclogp, struct xlog_ticket *ticket, - int *continued_write, int *logoffsetp) { int log_offset;
@@ -2931,13 +2925,10 @@ xlog_state_get_iclog_space(* iclogs (to mark it taken), this particular iclog will release/sync * to disk in xlog_write(). */ -if (len <= iclog->ic_size - iclog->ic_offset) { - *continued_write =0 ; +if (len <= iclog->ic_size - iclog->ic_offset) iclog->ic_offset += len; - }else { - *continued_write =1 ; +else xlog_state_switch_iclogs (log, iclog, iclog->ic_size); - } *iclogp = iclog;ASSERT (iclog->ic_offset <= iclog->ic_size);
Signature · 1 line
2.31.1
PATCH v4 24/39 xfs: xlog_write() doesn't need optype anymore
In reply to Dave ChinnerFrom: Dave Chinner <[email protected]>
So remove it from the interface and callers.
--- fs/xfs/xfs_log.c | 14 ++++---------- fs/xfs/xfs_log_cil.c | 2 +- fs/xfs/xfs_log_priv.h | 2 +- 3 files changed, 6 insertions(+), 12 deletions(-)
fs/xfs/xfs_log.c +4 -10
@@ -863,8 +863,7 @@ xlog_write_unmount_record(*/ if (log->l_targ != log->l_mp->m_ddev_targp)blkdev_issue_flush (log->l_targ->bt_bdev); -return xlog_write (log, &vec, ticket,NULL ,NULL ,XLOG_UNMOUNT_TRANS , - reg.i_len); +return xlog_write (log, &vec, ticket,NULL ,NULL , reg.i_len); }/*
@@ -1588,8 +1587,7 @@ xlog_commit_record(/* account for space used by record data */ ticket->t_curr_res -= reg.i_len; - error =xlog_write (log, &vec, ticket, lsn, iclog,XLOG_COMMIT_TRANS , - reg.i_len); + error =xlog_write (log, &vec, ticket, lsn, iclog, reg.i_len);if (error)xfs_force_shutdown (log->l_mp,SHUTDOWN_LOG_IO_ERROR );return error;
@@ -2399,7 +2397,6 @@ xlog_write(struct xlog_ticket *ticket, xfs_lsn_t *start_lsn, struct xlog_in_core **commit_iclog, - uint optype, uint32_t len) { struct xlog_in_core *iclog = NULL;
@@ -2431,7 +2428,6 @@ xlog_write(if (!lv)break ; -ASSERT (!(optype & (XLOG_COMMIT_TRANS |XLOG_UNMOUNT_TRANS ))); lv =xlog_write_partial (log, lv, ticket, &iclog, &log_offset, &len, &record_cnt, &data_cnt);if (IS_ERR_OR_NULL (lv)) {
@@ -2449,12 +2445,10 @@ xlog_write(*/ spin_lock (&log->l_icloglock);xlog_state_finish_copy (log, iclog, record_cnt,0 ); -if (commit_iclog) { -ASSERT (optype &XLOG_COMMIT_TRANS ); +if (commit_iclog) *commit_iclog = iclog; - }else { +else error =xlog_state_release_iclog (log, iclog); - }spin_unlock (&log->l_icloglock);return error;
fs/xfs/xfs_log_cil.c +1 -1
@@ -908,7 +908,7 @@ xlog_cil_push_work(* write head. */ error =xlog_write (log, &lvhdr, ctx->ticket, &ctx->start_lsn,NULL , -XLOG_START_TRANS , num_bytes); + num_bytes);if (error) goto out_abort_free_ticket;
fs/xfs/xfs_log_priv.h +1 -1
@@ -459,7 +459,7 @@ void xlog_print_tic_res(struct xfs_mount *mp, struct xlog_ticket *ticket);void xlog_print_trans (struct xfs_trans *);int xlog_write (struct xlog *log,struct xfs_log_vec *log_vector,struct xlog_ticket *tic,xfs_lsn_t *start_lsn, -struct xlog_in_core **commit_iclog,uint optype,uint32_t len); +struct xlog_in_core **commit_iclog,uint32_t len);int xlog_commit_record (struct xlog *log,struct xlog_ticket *ticket,struct xlog_in_core **iclog,xfs_lsn_t *lsn);
Signature · 1 line
2.31.1
Re: [PATCH 24/39] xfs: xlog_write() doesn't need optype anymore
Quoted 5 lines · Dave Chinner
From: Dave Chinner <[email protected]>
So remove it from the interface and callers.
Signed-off-by Dave Chinner <[email protected]>
Looks pretty straightforward.
--D
Quoted 90 lines
--- fs/xfs/xfs_log.c | 14 ++++---------- fs/xfs/xfs_log_cil.c | 2 +- fs/xfs/xfs_log_priv.h | 2 +- 3 files changed, 6 insertions(+), 12 deletions(-)fs/xfs/xfs_log.c +4
-10@@ -863,8 +863,7 @@ xlog_write_unmount_record(*/ if (log->l_targ != log->l_mp->m_ddev_targp)blkdev_issue_flush (log->l_targ->bt_bdev); -return xlog_write (log, &vec, ticket,NULL ,NULL ,XLOG_UNMOUNT_TRANS , - reg.i_len); +return xlog_write (log, &vec, ticket,NULL ,NULL , reg.i_len); }/* @@ -1588,8 +1587,7 @@ xlog_commit_record(/* account for space used by record data */ ticket->t_curr_res -= reg.i_len; - error =xlog_write (log, &vec, ticket, lsn, iclog,XLOG_COMMIT_TRANS , - reg.i_len); + error =xlog_write (log, &vec, ticket, lsn, iclog, reg.i_len);if (error)xfs_force_shutdown (log->l_mp,SHUTDOWN_LOG_IO_ERROR );return error;@@ -2399,7 +2397,6 @@ xlog_write(struct xlog_ticket *ticket, xfs_lsn_t *start_lsn, struct xlog_in_core **commit_iclog, - uint optype, uint32_t len) { struct xlog_in_core *iclog = NULL;@@ -2431,7 +2428,6 @@ xlog_write(if (!lv)break ; -ASSERT (!(optype & (XLOG_COMMIT_TRANS |XLOG_UNMOUNT_TRANS ))); lv =xlog_write_partial (log, lv, ticket, &iclog, &log_offset, &len, &record_cnt, &data_cnt);if (IS_ERR_OR_NULL (lv)) {@@ -2449,12 +2445,10 @@ xlog_write(*/ spin_lock (&log->l_icloglock);xlog_state_finish_copy (log, iclog, record_cnt,0 ); -if (commit_iclog) { -ASSERT (optype &XLOG_COMMIT_TRANS ); +if (commit_iclog) *commit_iclog = iclog; - }else { +else error =xlog_state_release_iclog (log, iclog); - }spin_unlock (&log->l_icloglock);return error;fs/xfs/xfs_log_cil.c +1
-1@@ -908,7 +908,7 @@ xlog_cil_push_work(* write head. */ error =xlog_write (log, &lvhdr, ctx->ticket, &ctx->start_lsn,NULL , -XLOG_START_TRANS , num_bytes); + num_bytes);if (error) goto out_abort_free_ticket;fs/xfs/xfs_log_priv.h +1
-1@@ -459,7 +459,7 @@ void xlog_print_tic_res(struct xfs_mount *mp, struct xlog_ticket *ticket);void xlog_print_trans (struct xfs_trans *);int xlog_write (struct xlog *log,struct xfs_log_vec *log_vector,struct xlog_ticket *tic,xfs_lsn_t *start_lsn, -struct xlog_in_core **commit_iclog,uint optype,uint32_t len); +struct xlog_in_core **commit_iclog,uint32_t len);int xlog_commit_record (struct xlog *log,struct xlog_ticket *ticket,struct xlog_in_core **iclog,xfs_lsn_t *lsn);Signature · 1 line
2.31.1
PATCH v4 25/39 xfs: CIL context doesn't need to count iovecs
In reply to Dave ChinnerFrom: Dave Chinner <[email protected]>
Now that we account for log opheaders in the log item formatting code, we don't actually use the aggregated count of log iovecs in the CIL for anything. Remove it and the tracking code that calculates it.
--- fs/xfs/xfs_log_cil.c | 22 ++++++---------------- fs/xfs/xfs_log_priv.h | 1 - 2 files changed, 6 insertions(+), 17 deletions(-)
fs/xfs/xfs_log_cil.c +6 -16
@@ -252,22 +252,18 @@ xlog_cil_alloc_shadow_bufs(/* * Prepare the log item for insertion into the CIL. Calculate the difference in -* log space and vectors it will consume, and if it is a new item pin it as -* well. +* log space it will consume, and if it is a new item pin it as well. */ STATIC voidxfs_cil_prepare_item (struct xlog *log,struct xfs_log_vec *lv,struct xfs_log_vec *old_lv, -int *diff_len, -int *diff_iovecs) +int *diff_len) {/* Account for the new LV being passed in */ -if (lv->lv_buf_len !=XFS_LOG_VEC_ORDERED ) { +if (lv->lv_buf_len !=XFS_LOG_VEC_ORDERED ) *diff_len += lv->lv_bytes; - *diff_iovecs += lv->lv_niovecs; - }/* * If there is no old LV, this is the first time we've seen the item in
@@ -284,7 +280,6 @@ xfs_cil_prepare_item(ASSERT (lv->lv_buf_len !=XFS_LOG_VEC_ORDERED ); *diff_len -= old_lv->lv_bytes; - *diff_iovecs -= old_lv->lv_niovecs; lv->lv_item->li_lv_shadow = old_lv; }
@@ -333,12 +328,10 @@ static voidxlog_cil_insert_format_items ( struct xlog *log, struct xfs_trans *tp, - int *diff_len, - int *diff_iovecs) + int *diff_len) { struct xfs_log_item *lip; -/* Bail out if we didn't find a log item. */ if (list_empty (&tp->t_items)) {ASSERT (0 );
@@ -381,7 +374,6 @@ xlog_cil_insert_format_items(* set the item up as though it is a new insertion so * that the space reservation accounting is correct. */ - *diff_iovecs -= lv->lv_niovecs; *diff_len -= lv->lv_bytes;/* Ensure the lv is set up according to ->iop_size */
@@ -406,7 +398,7 @@ xlog_cil_insert_format_items(ASSERT (IS_ALIGNED ((unsigned long )lv->lv_buf,sizeof (uint64_t ))); lip->li_ops->iop_format (lip, lv); insert: -xfs_cil_prepare_item (log, lv, old_lv, diff_len, diff_iovecs); +xfs_cil_prepare_item (log, lv, old_lv, diff_len); } }
@@ -426,7 +418,6 @@ xlog_cil_insert_items(struct xfs_cil_ctx *ctx = cil->xc_ctx;struct xfs_log_item *lip;int len =0 ; -int diff_iovecs =0 ;int iclog_space;int iovhdr_res =0 , split_res =0 , ctx_res =0 ;
@@ -436,7 +427,7 @@ xlog_cil_insert_items(* We can do this safely because the context can't checkpoint until we * are done so it doesn't matter exactly how we update the CIL. */ -xlog_cil_insert_format_items (log, tp, &len, &diff_iovecs); +xlog_cil_insert_format_items (log, tp, &len);spin_lock (&cil->xc_cil_lock);
@@ -471,7 +462,6 @@ xlog_cil_insert_items( } tp->t_ticket->t_curr_res -= len; ctx->space_used += len; - ctx->nvecs += diff_iovecs;/* * If we've overrun the reservation, dump the tx details before we move
fs/xfs/xfs_log_priv.h +0 -1
@@ -217,7 +217,6 @@ struct xfs_cil_ctx {xfs_lsn_t start_lsn;/* first LSN of chkpt commit */ xfs_lsn_t commit_lsn;/* chkpt commit record lsn */ struct xlog_ticket *ticket;/* chkpt ticket */ -int nvecs;/* number of regions */ int space_used;/* aggregate size of regions */ struct list_head busy_extents;/* busy extents in chkpt */ struct xfs_log_vec *lv_chain;/* logvecs being pushed */
Signature · 1 line
2.31.1
Re: [PATCH 25/39] xfs: CIL context doesn't need to count iovecs
Quoted 8 lines · Dave Chinner
From: Dave Chinner <[email protected]>
Now that we account for log opheaders in the log item formatting code, we don't actually use the aggregated count of log iovecs in the CIL for anything. Remove it and the tracking code that calculates it.
Signed-off-by Dave Chinner <[email protected]>
Looks good this time around,
--D
Quoted 114 lines
--- fs/xfs/xfs_log_cil.c | 22 ++++++---------------- fs/xfs/xfs_log_priv.h | 1 - 2 files changed, 6 insertions(+), 17 deletions(-)fs/xfs/xfs_log_cil.c +6
-16@@ -252,22 +252,18 @@ xlog_cil_alloc_shadow_bufs(/* * Prepare the log item for insertion into the CIL. Calculate the difference in -* log space and vectors it will consume, and if it is a new item pin it as -* well. +* log space it will consume, and if it is a new item pin it as well. */ STATIC voidxfs_cil_prepare_item (struct xlog *log,struct xfs_log_vec *lv,struct xfs_log_vec *old_lv, -int *diff_len, -int *diff_iovecs) +int *diff_len) {/* Account for the new LV being passed in */ -if (lv->lv_buf_len !=XFS_LOG_VEC_ORDERED ) { +if (lv->lv_buf_len !=XFS_LOG_VEC_ORDERED ) *diff_len += lv->lv_bytes; - *diff_iovecs += lv->lv_niovecs; - }/* * If there is no old LV, this is the first time we've seen the item in @@ -284,7 +280,6 @@ xfs_cil_prepare_item(ASSERT (lv->lv_buf_len !=XFS_LOG_VEC_ORDERED ); *diff_len -= old_lv->lv_bytes; - *diff_iovecs -= old_lv->lv_niovecs; lv->lv_item->li_lv_shadow = old_lv; }@@ -333,12 +328,10 @@ static voidxlog_cil_insert_format_items ( struct xlog *log, struct xfs_trans *tp, - int *diff_len, - int *diff_iovecs) + int *diff_len) { struct xfs_log_item *lip; -/* Bail out if we didn't find a log item. */ if (list_empty (&tp->t_items)) {ASSERT (0 );@@ -381,7 +374,6 @@ xlog_cil_insert_format_items(* set the item up as though it is a new insertion so * that the space reservation accounting is correct. */ - *diff_iovecs -= lv->lv_niovecs; *diff_len -= lv->lv_bytes;/* Ensure the lv is set up according to ->iop_size */ @@ -406,7 +398,7 @@ xlog_cil_insert_format_items(ASSERT (IS_ALIGNED ((unsigned long )lv->lv_buf,sizeof (uint64_t ))); lip->li_ops->iop_format (lip, lv); insert: -xfs_cil_prepare_item (log, lv, old_lv, diff_len, diff_iovecs); +xfs_cil_prepare_item (log, lv, old_lv, diff_len); } }@@ -426,7 +418,6 @@ xlog_cil_insert_items(struct xfs_cil_ctx *ctx = cil->xc_ctx;struct xfs_log_item *lip;int len =0 ; -int diff_iovecs =0 ;int iclog_space;int iovhdr_res =0 , split_res =0 , ctx_res =0 ;@@ -436,7 +427,7 @@ xlog_cil_insert_items(* We can do this safely because the context can't checkpoint until we * are done so it doesn't matter exactly how we update the CIL. */ -xlog_cil_insert_format_items (log, tp, &len, &diff_iovecs); +xlog_cil_insert_format_items (log, tp, &len);spin_lock (&cil->xc_cil_lock);@@ -471,7 +462,6 @@ xlog_cil_insert_items( } tp->t_ticket->t_curr_res -= len; ctx->space_used += len; - ctx->nvecs += diff_iovecs;/* * If we've overrun the reservation, dump the tx details before we move fs/xfs/xfs_log_priv.h +0
-1@@ -217,7 +217,6 @@ struct xfs_cil_ctx {xfs_lsn_t start_lsn;/* first LSN of chkpt commit */ xfs_lsn_t commit_lsn;/* chkpt commit record lsn */ struct xlog_ticket *ticket;/* chkpt ticket */ -int nvecs;/* number of regions */ int space_used;/* aggregate size of regions */ struct list_head busy_extents;/* busy extents in chkpt */ struct xfs_log_vec *lv_chain;/* logvecs being pushed */ Signature · 1 line
2.31.1
PATCH v4 26/39 xfs: use the CIL space used counter for emptiness checks
In reply to Dave ChinnerFrom: Dave Chinner <[email protected]>
In the next patches we are going to make the CIL list itself per-cpu, and so we cannot use list_empty() to check is the list is empty. Replace the list_empty() checks with a flag in the CIL to indicate we have committed at least one transaction to the CIL and hence the CIL is not empty.
We need this flag to be an atomic so that we can clear it without holding any locks in the commit fast path, but we also need to be careful to avoid atomic operations in the fast path. Hence we use the fact that test_bit() is not an atomic op to first check if the flag is set and then run the atomic test_and_clear_bit() operation to clear it and steal the initial unit reservation for the CIL context checkpoint.
When we are switching to a new context in a push, we place the setting of the XLOG_CIL_EMPTY flag under the xc_push_lock. THis allows all the other places that need to check whether the CIL is empty to use test_bit() and still be serialised correctly with the CIL context swaps that set the bit.
--- fs/xfs/xfs_log_cil.c | 45 ++++++++++++++++++++++++------------------- fs/xfs/xfs_log_priv.h | 4 ++++ 2 files changed, 29 insertions(+), 20 deletions(-)
fs/xfs/xfs_log_cil.c +25 -20
@@ -70,6 +70,7 @@ xlog_cil_ctx_switch(struct xfs_cil *cil, struct xfs_cil_ctx *ctx) { + set_bit(XLOG_CIL_EMPTY, &cil->xc_flags); ctx->sequence = ++cil->xc_current_sequence; ctx->cil = cil; cil->xc_ctx = ctx;
@@ -436,13 +437,12 @@ xlog_cil_insert_items(list_splice_init (&tp->t_busy, &ctx->busy_extents);/* -* Now transfer enough transaction reservation to the context ticket -* for the checkpoint. The context ticket is special - the unit -* reservation has to grow as well as the current reservation as we -* steal from tickets so we can correctly determine the space used -* during the transaction commit. +* We need to take the CIL checkpoint unit reservation on the first +* commit into the CIL. Test the XLOG_CIL_EMPTY bit first so we don't +* unnecessarily do an atomic op in the fast path here. */ -if (ctx->ticket->t_curr_res ==0 ) { +if (test_bit (XLOG_CIL_EMPTY , &cil->xc_flags) && +test_and_clear_bit (XLOG_CIL_EMPTY , &cil->xc_flags)) { ctx_res = ctx->ticket->t_unit_res; ctx->ticket->t_curr_res = ctx_res; tp->t_ticket->t_curr_res -= ctx_res;
@@ -771,7 +771,7 @@ xlog_cil_push_work(* move on to a new sequence number and so we have to be able to push * this sequence again later. */ -if (list_empty (&cil->xc_cil)) { +if (test_bit (XLOG_CIL_EMPTY , &cil->xc_flags)) { cil->xc_push_seq =0 ;spin_unlock (&cil->xc_push_lock); goto out_skip;
@@ -1020,9 +1020,10 @@ xlog_cil_push_background(/* * The cil won't be empty because we are called while holding the -* context lock so whatever we added to the CIL will still be there +* context lock so whatever we added to the CIL will still be there. */ ASSERT (!list_empty (&cil->xc_cil)); +ASSERT (!test_bit (XLOG_CIL_EMPTY , &cil->xc_flags));/* * Don't do a background push if we haven't used up all the
@@ -1109,7 +1110,8 @@ xlog_cil_push_now(* there's no work we need to do. */ spin_lock (&cil->xc_push_lock); -if (list_empty (&cil->xc_cil) || push_seq <= cil->xc_push_seq) { +if (test_bit (XLOG_CIL_EMPTY , &cil->xc_flags) || + push_seq <= cil->xc_push_seq) {spin_unlock (&cil->xc_push_lock);return ; }
@@ -1128,7 +1130,7 @@ xlog_cil_empty(bool empty = false;spin_lock (&cil->xc_push_lock); -if (list_empty (&cil->xc_cil)) +if (test_bit (XLOG_CIL_EMPTY , &cil->xc_flags)) empty = true;spin_unlock (&cil->xc_push_lock);return empty;
@@ -1296,7 +1298,7 @@ xlog_cil_force_seq(* we would have found the context on the committing list. */ if (sequence == cil->xc_current_sequence && - !list_empty (&cil->xc_cil)) { + !test_bit (XLOG_CIL_EMPTY , &cil->xc_flags)) {spin_unlock (&cil->xc_push_lock); goto restart; }
@@ -1329,9 +1331,9 @@ boolxfs_log_item_in_current_chkpt ( struct xfs_log_item *lip) { -struct xfs_cil_ctx *ctx = lip->li_mountp->m_log->l_cilp->xc_ctx; +struct xfs_cil *cil = lip->li_mountp->m_log->l_cilp; -if (list_empty (&lip->li_cil)) +if (test_bit (XLOG_CIL_EMPTY , &cil->xc_flags))return false;/*
@@ -1339,7 +1341,7 @@ xfs_log_item_in_current_chkpt(* first checkpoint it is written to. Hence if it is different to the * current sequence, we're in a new checkpoint. */ -return lip->li_seq == ctx->sequence; +return lip->li_seq == cil->xc_ctx->sequence; }/*
@@ -1376,13 +1378,16 @@ voidxlog_cil_destroy ( struct xlog *log) { -if (log->l_cilp->xc_ctx) { -if (log->l_cilp->xc_ctx->ticket) -xfs_log_ticket_put (log->l_cilp->xc_ctx->ticket); -kmem_free (log->l_cilp->xc_ctx); +struct xfs_cil *cil = log->l_cilp; + +if (cil->xc_ctx) { +if (cil->xc_ctx->ticket) +xfs_log_ticket_put (cil->xc_ctx->ticket); +kmem_free (cil->xc_ctx); } -ASSERT (list_empty (&log->l_cilp->xc_cil)); -kmem_free (log->l_cilp); +ASSERT (list_empty (&cil->xc_cil)); +ASSERT (test_bit (XLOG_CIL_EMPTY , &cil->xc_flags)); +kmem_free (cil); }
fs/xfs/xfs_log_priv.h +4 -0
@@ -244,6 +244,7 @@ struct xfs_cil_ctx {*/ struct xfs_cil {struct xlog *xc_log; +unsigned long xc_flags;struct list_head xc_cil;spinlock_t xc_cil_lock;
@@ -259,6 +260,9 @@ struct xfs_cil {wait_queue_head_t xc_push_wait;/* background push throttle */ }____cacheline_aligned_in_smp ; +/* xc_flags bit values */ +#define XLOG_CIL_EMPTY 1 +/* * The amount of log space we allow the CIL to aggregate is difficult to size. * Whatever we choose, we have to make sure we can get a reservation for the
Signature · 1 line
2.31.1
PATCH v4 27/39 xfs: lift init CIL reservation out of xc_cil_lock
In reply to Dave ChinnerFrom: Dave Chinner <[email protected]>
The xc_cil_lock is the most highly contended lock in XFS now. To start the process of getting rid of it, lift the initial reservation of the CIL log space out from under the xc_cil_lock.
--- fs/xfs/xfs_log_cil.c | 27 ++++++++++++--------------- 1 file changed, 12 insertions(+), 15 deletions(-)
fs/xfs/xfs_log_cil.c +12 -15
@@ -430,23 +430,19 @@ xlog_cil_insert_items(*/ xlog_cil_insert_format_items (log, tp, &len); -spin_lock (&cil->xc_cil_lock); - -/* attach the transaction to the CIL if it has any busy extents */ -if (!list_empty (&tp->t_busy)) -list_splice_init (&tp->t_busy, &ctx->busy_extents); -/* * We need to take the CIL checkpoint unit reservation on the first * commit into the CIL. Test the XLOG_CIL_EMPTY bit first so we don't -* unnecessarily do an atomic op in the fast path here. +* unnecessarily do an atomic op in the fast path here. We don't need to +* hold the xc_cil_lock here to clear the XLOG_CIL_EMPTY bit as we are +* under the xc_ctx_lock here and that needs to be held exclusively to +* reset the XLOG_CIL_EMPTY bit. */ if (test_bit (XLOG_CIL_EMPTY , &cil->xc_flags) && -test_and_clear_bit (XLOG_CIL_EMPTY , &cil->xc_flags)) { +test_and_clear_bit (XLOG_CIL_EMPTY , &cil->xc_flags)) ctx_res = ctx->ticket->t_unit_res; - ctx->ticket->t_curr_res = ctx_res; - tp->t_ticket->t_curr_res -= ctx_res; - } + +spin_lock (&cil->xc_cil_lock);/* do we need space for more log record headers? */ iclog_space = log->l_iclog_size - log->l_iclog_hsize;
@@ -456,11 +452,9 @@ xlog_cil_insert_items(/* need to take into account split region headers, too */ split_res *= log->l_iclog_hsize +sizeof (struct xlog_op_header ); ctx->ticket->t_unit_res += split_res; - ctx->ticket->t_curr_res += split_res; - tp->t_ticket->t_curr_res -= split_res; -ASSERT (tp->t_ticket->t_curr_res >= len); } - tp->t_ticket->t_curr_res -= len; + tp->t_ticket->t_curr_res -= split_res + ctx_res + len; + ctx->ticket->t_curr_res += split_res + ctx_res; ctx->space_used += len;/*
@@ -498,6 +492,9 @@ xlog_cil_insert_items(list_move_tail (&lip->li_cil, &cil->xc_cil); } +/* attach the transaction to the CIL if it has any busy extents */ +if (!list_empty (&tp->t_busy)) +list_splice_init (&tp->t_busy, &ctx->busy_extents);spin_unlock (&cil->xc_cil_lock);if (tp->t_ticket->t_curr_res <0 )
Signature · 1 line
2.31.1
PATCH v4 28/39 xfs: rework per-iclog header CIL reservation
In reply to Dave ChinnerFrom: Dave Chinner <[email protected]>
For every iclog that a CIL push will use up, we need to ensure we have space reserved for the iclog header in each iclog. It is extremely difficult to do this accurately with a per-cpu counter without expensive summing of the counter in every commit. However, we know what the maximum CIL size is going to be because of the hard space limit we have, and hence we know exactly how many iclogs we are going to need to write out the CIL.
We are constrained by the requirement that small transactions only have reservation space for a single iclog header built into them. At commit time we don't know how much of the current transaction reservation is made up of iclog header reservations as calculated by xfs_log_calc_unit_res() when the ticket was reserved. As larger reservations have multiple header spaces reserved, we can steal more than one iclog header reservation at a time, but we only steal the exact number needed for the given log vector size delta.
As a result, we don't know exactly when we are going to steal iclog header reservations, nor do we know exactly how many we are going to need for a given CIL.
To make things simple, start by calculating the worst case number of iclog headers a full CIL push will require. Record this into an atomic variable in the CIL. Then add a byte counter to the log ticket that records exactly how much iclog header space has been reserved in this ticket by xfs_log_calc_unit_res(). This tells us exactly how much space we can steal from the ticket at transaction commit time.
Now, at transaction commit time, we can check if the CIL has a full iclog header reservation and, if not, steal the entire reservation the current ticket holds for iclog headers. This minimises the number of times we need to do atomic operations in the fast path, but still guarantees we get all the reservations we need.
--- fs/xfs/xfs_log.c | 9 ++++--- fs/xfs/xfs_log_cil.c | 55 +++++++++++++++++++++++++++++++++---------- fs/xfs/xfs_log_priv.h | 20 +++++++++------- 3 files changed, 59 insertions(+), 25 deletions(-)
fs/xfs/xfs_log.c +6 -3
@@ -3307,7 +3307,8 @@ xfs_log_ticket_get(static int xlog_calc_unit_res (struct xlog *log, -int unit_bytes) +int unit_bytes, +int *niclogs) {int iclog_space;uint num_headers;
@@ -3387,6 +3388,8 @@ xlog_calc_unit_res(/* roundoff padding for transaction data and one for commit record */ unit_bytes +=2 * log->l_iclog_roundoff; +if (niclogs) + *niclogs = num_headers;return unit_bytes; }
@@ -3395,7 +3398,7 @@ xfs_log_calc_unit_res( struct xfs_mount *mp, int unit_bytes) { -return xlog_calc_unit_res (mp->m_log, unit_bytes); +return xlog_calc_unit_res (mp->m_log, unit_bytes,NULL ); }/*
@@ -3413,7 +3416,7 @@ xlog_ticket_alloc( tic =kmem_cache_zalloc (xfs_log_ticket_zone,GFP_NOFS | __GFP_NOFAIL); - unit_res =xlog_calc_unit_res (log, unit_bytes); + unit_res =xlog_calc_unit_res (log, unit_bytes, &tic->t_iclog_hdrs);atomic_set (&tic->t_ref,1 ); tic->t_task = current;
fs/xfs/xfs_log_cil.c +42 -13
89 lines not on this page · raw message
fs/xfs/xfs_log_priv.h +11 -9
@@ -137,15 +137,16 @@ enum xlog_iclog_state {#define XLOG_ICL_NEED_FUA (1 << 1)/* iclog needs REQ_FUA */ typedef struct xlog_ticket { -struct list_head t_queue;/* reserve/write queue */ -struct task_struct *t_task;/* task that owns this ticket */ -xlog_tid_t t_tid;/* transaction identifier : 4 */ -atomic_t t_ref;/* ticket reference count : 4 */ -int t_curr_res;/* current reservation in bytes : 4 */ -int t_unit_res;/* unit reservation in bytes : 4 */ -char t_ocnt;/* original count : 1 */ -char t_cnt;/* current count : 1 */ -char t_flags;/* properties of reservation : 1 */ +struct list_head t_queue;/* reserve/write queue */ +struct task_struct *t_task;/* task that owns this ticket */ +xlog_tid_t t_tid;/* transaction identifier */ +atomic_t t_ref;/* ticket reference count */ +int t_curr_res;/* current reservation */ +int t_unit_res;/* unit reservation */ +char t_ocnt;/* original count */ +char t_cnt;/* current count */ +char t_flags;/* properties of reservation */ +int t_iclog_hdrs;/* iclog hdrs in t_curr_res */ }xlog_ticket_t ;/*
@@ -245,6 +246,7 @@ struct xfs_cil_ctx {struct xfs_cil {struct xlog *xc_log;unsigned long xc_flags; +atomic_t xc_iclog_hdrs;struct list_head xc_cil;spinlock_t xc_cil_lock;
Signature · 1 line
2.31.1
Re: [PATCH 28/39] xfs: rework per-iclog header CIL reservation
Quoted 31 lines · Dave Chinner
From: Dave Chinner <[email protected]>
For every iclog that a CIL push will use up, we need to ensure we have space reserved for the iclog header in each iclog. It is extremely difficult to do this accurately with a per-cpu counter without expensive summing of the counter in every commit. However, we know what the maximum CIL size is going to be because of the hard space limit we have, and hence we know exactly how many iclogs we are going to need to write out the CIL.
We are constrained by the requirement that small transactions only have reservation space for a single iclog header built into them. At commit time we don't know how much of the current transaction reservation is made up of iclog header reservations as calculated by
xfs_log_calc_unit_res()when the ticket was reserved. As larger reservations have multiple header spaces reserved, we can steal more than one iclog header reservation at a time, but we only steal the exact number needed for the given log vector size delta.As a result, we don't know exactly when we are going to steal iclog header reservations, nor do we know exactly how many we are going to need for a given CIL.
To make things simple, start by calculating the worst case number of iclog headers a full CIL push will require. Record this into an atomic variable in the CIL. Then add a byte counter to the log ticket that records exactly how much iclog header space has been reserved in this ticket by
xfs_log_calc_unit_res(). This tells us exactly how much space we can steal from the ticket at transaction commit time.
Quoted 7 lines
Now, at transaction commit time, we can check if the CIL has a full iclog header reservation and, if not, steal the entire reservation the current ticket holds for iclog headers. This minimises the number of times we need to do atomic operations in the fast path, but still guarantees we get all the reservations we need.
Signed-off-by Dave Chinner <[email protected]>
No new questions since last time, so:
--D
Quoted 186 lines
--- fs/xfs/xfs_log.c | 9 ++++--- fs/xfs/xfs_log_cil.c | 55 +++++++++++++++++++++++++++++++++---------- fs/xfs/xfs_log_priv.h | 20 +++++++++------- 3 files changed, 59 insertions(+), 25 deletions(-)fs/xfs/xfs_log.c +6
-3@@ -3307,7 +3307,8 @@ xfs_log_ticket_get(static int xlog_calc_unit_res (struct xlog *log, -int unit_bytes) +int unit_bytes, +int *niclogs) {int iclog_space;uint num_headers;@@ -3387,6 +3388,8 @@ xlog_calc_unit_res(/* roundoff padding for transaction data and one for commit record */ unit_bytes +=2 * log->l_iclog_roundoff; +if (niclogs) + *niclogs = num_headers;return unit_bytes; }@@ -3395,7 +3398,7 @@ xfs_log_calc_unit_res( struct xfs_mount *mp, int unit_bytes) { -return xlog_calc_unit_res (mp->m_log, unit_bytes); +return xlog_calc_unit_res (mp->m_log, unit_bytes,NULL ); }/* @@ -3413,7 +3416,7 @@ xlog_ticket_alloc( tic =kmem_cache_zalloc (xfs_log_ticket_zone,GFP_NOFS | __GFP_NOFAIL); - unit_res =xlog_calc_unit_res (log, unit_bytes); + unit_res =xlog_calc_unit_res (log, unit_bytes, &tic->t_iclog_hdrs);atomic_set (&tic->t_ref,1 ); tic->t_task = current;fs/xfs/xfs_log_cil.c +42
-13@@ -44,9 +44,20 @@ xlog_cil_ticket_alloc(* transaction overhead reservation from the first transaction commit. */ tic->t_curr_res =0 ; + tic->t_iclog_hdrs =0 ;return tic; } +static inline void +xlog_cil_set_iclog_hdr_count (struct xfs_cil *cil) +{ +struct xlog *log = cil->xc_log; + +atomic_set (&cil->xc_iclog_hdrs, + (XLOG_CIL_BLOCKING_SPACE_LIMIT (log) / + (log->l_iclog_size - log->l_iclog_hsize))); +} +/* * Unavoidable forward declaration - xlog_cil_push_work() calls * xlog_cil_ctx_alloc() itself. @@ -70,6 +81,7 @@ xlog_cil_ctx_switch(struct xfs_cil *cil, struct xfs_cil_ctx *ctx) { +xlog_cil_set_iclog_hdr_count (cil );set_bit (XLOG_CIL_EMPTY , &cil->xc_flags); ctx->sequence = ++cil->xc_current_sequence; ctx->cil = cil;@@ -92,6 +104,7 @@ xlog_cil_init_post_recovery( { log->l_cilp->xc_ctx->ticket =xlog_cil_ticket_alloc (log); log->l_cilp->xc_ctx->sequence =1 ; +xlog_cil_set_iclog_hdr_count (log->l_cilp); } static inline int@@ -419,7 +432,6 @@ xlog_cil_insert_items(struct xfs_cil_ctx *ctx = cil->xc_ctx;struct xfs_log_item *lip;int len =0 ; -int iclog_space;int iovhdr_res =0 , split_res =0 , ctx_res =0 ;ASSERT (tp);@@ -442,19 +454,36 @@ xlog_cil_insert_items( test_and_clear_bit(XLOG_CIL_EMPTY, &cil->xc_flags)) ctx_res = ctx->ticket->t_unit_res; -spin_lock (&cil->xc_cil_lock); - -/* do we need space for more log record headers? */ - iclog_space = log->l_iclog_size - log->l_iclog_hsize; -if (len >0 && (ctx->space_used / iclog_space != - (ctx->space_used + len) / iclog_space)) { - split_res = (len + iclog_space -1 ) / iclog_space; -/* need to take into account split region headers, too */ - split_res *= log->l_iclog_hsize +sizeof (struct xlog_op_header ); - ctx->ticket->t_unit_res += split_res; +/* +* Check if we need to steal iclog headers. atomic_read() is not a +* locked atomic operation, so we can check the value before we do any +* real atomic ops in the fast path. If we've already taken the CIL unit +* reservation from this commit, we've already got one iclog header +* space reserved so we have to account for that otherwise we risk +* overrunning the reservation on this ticket. +* +* If the CIL is already at the hard limit, we might need more header +* space that originally reserved. So steal more header space from every +* commit that occurs once we are over the hard limit to ensure the CIL +* push won't run out of reservation space. +* +* This can steal more than we need, but that's OK. +*/ +if (atomic_read (&cil->xc_iclog_hdrs) >0 || + ctx->space_used + len >=XLOG_CIL_BLOCKING_SPACE_LIMIT (log)) { +int split_res = log->l_iclog_hsize + +sizeof (struct xlog_op_header ); +if (ctx_res) + ctx_res += split_res * (tp->t_ticket->t_iclog_hdrs -1 ); +else + ctx_res = split_res * tp->t_ticket->t_iclog_hdrs; +atomic_sub (tp->t_ticket->t_iclog_hdrs, &cil->xc_iclog_hdrs); } - tp->t_ticket->t_curr_res -= split_res + ctx_res + len; - ctx->ticket->t_curr_res += split_res + ctx_res; + +spin_lock (&cil->xc_cil_lock); + tp->t_ticket->t_curr_res -= ctx_res + len; + ctx->ticket->t_unit_res += ctx_res; + ctx->ticket->t_curr_res += ctx_res; ctx->space_used += len;/* fs/xfs/xfs_log_priv.h +11
-9@@ -137,15 +137,16 @@ enum xlog_iclog_state {#define XLOG_ICL_NEED_FUA (1 << 1)/* iclog needs REQ_FUA */ typedef struct xlog_ticket { -struct list_head t_queue;/* reserve/write queue */ -struct task_struct *t_task;/* task that owns this ticket */ -xlog_tid_t t_tid;/* transaction identifier : 4 */ -atomic_t t_ref;/* ticket reference count : 4 */ -int t_curr_res;/* current reservation in bytes : 4 */ -int t_unit_res;/* unit reservation in bytes : 4 */ -char t_ocnt;/* original count : 1 */ -char t_cnt;/* current count : 1 */ -char t_flags;/* properties of reservation : 1 */ +struct list_head t_queue;/* reserve/write queue */ +struct task_struct *t_task;/* task that owns this ticket */ +xlog_tid_t t_tid;/* transaction identifier */ +atomic_t t_ref;/* ticket reference count */ +int t_curr_res;/* current reservation */ +int t_unit_res;/* unit reservation */ +char t_ocnt;/* original count */ +char t_cnt;/* current count */ +char t_flags;/* properties of reservation */ +int t_iclog_hdrs;/* iclog hdrs in t_curr_res */ }xlog_ticket_t ;/* @@ -245,6 +246,7 @@ struct xfs_cil_ctx {struct xfs_cil {struct xlog *xc_log;unsigned long xc_flags; +atomic_t xc_iclog_hdrs;struct list_head xc_cil;spinlock_t xc_cil_lock;Signature · 1 line
2.31.1
PATCH v4 29/39 xfs: introduce per-cpu CIL tracking structure
In reply to Dave ChinnerFrom: Dave Chinner <[email protected]>
The CIL push lock is highly contended on larger machines, becoming a hard bottleneck that about 700,000 transaction commits/s on >16p machines. To address this, start moving the CIL tracking infrastructure to utilise per-CPU structures.
We need to track the space used, the amount of log reservation space reserved to write the CIL, the log items in the CIL and the busy extents that need to be completed by the CIL commit. This requires a couple of per-cpu counters, an unordered per-cpu list and a globally ordered per-cpu list.
Create a per-cpu structure to hold these and all the management interfaces needed, as well as the hooks to handle hotplug CPUs.
--- fs/xfs/xfs_log_cil.c | 106 +++++++++++++++++++++++++++++++++++++ fs/xfs/xfs_log_priv.h | 15 ++++++ include/linux/cpuhotplug.h | 1 + 3 files changed, 122 insertions(+)
fs/xfs/xfs_log_cil.c +106 -0
@@ -1370,6 +1370,105 @@ xfs_log_item_in_current_chkpt(return lip->li_seq == cil->xc_ctx->sequence; } +#ifdef CONFIG_HOTPLUG_CPU +static LIST_HEAD (xlog_cil_pcp_list); +static DEFINE_SPINLOCK (xlog_cil_pcp_lock); +static bool xlog_cil_pcp_init; + +/* +* Move dead percpu state to the relevant CIL context structures. +* +* We have to lock the CIL context here to ensure that nothing is modifying +* the percpu state, either addition or removal. Both of these are done under +* the CIL context lock, so grabbing that exclusively here will ensure we can +* safely drain the cilpcp for the CPU that is dying. +*/ +static int +xlog_cil_pcp_dead ( +unsigned int cpu) +{ +struct xfs_cil *cil, *n; + +spin_lock (&xlog_cil_pcp_lock); +list_for_each_entry_safe (cil, n, &xlog_cil_pcp_list, xc_pcp_list) { +spin_unlock (&xlog_cil_pcp_lock); +down_write (&cil->xc_ctx_lock); +/* move stuff on dead CPU to context */ +up_write (&cil->xc_ctx_lock); +spin_lock (&xlog_cil_pcp_lock); + } +spin_unlock (&xlog_cil_pcp_lock); +return 0 ; +} + +static int +xlog_cil_pcp_hpadd ( +struct xfs_cil *cil) +{ +if (!xlog_cil_pcp_init) { +int ret; + ret =cpuhp_setup_state_nocalls (CPUHP_XFS_CIL_DEAD , +"xfs/cil_pcp:dead" ,NULL , + xlog_cil_pcp_dead); +if (ret <0 ) { +xfs_warn (cil->xc_log->l_mp, +"Failed to initialise CIL hotplug, error %d. XFS is non-functional." , + ret); +ASSERT (0 ); +return -ENOMEM ; + } + xlog_cil_pcp_init = true; + } + +INIT_LIST_HEAD (&cil->xc_pcp_list); +spin_lock (&xlog_cil_pcp_lock); +list_add (&cil->xc_pcp_list, &xlog_cil_pcp_list); +spin_unlock (&xlog_cil_pcp_lock); +return 0 ; +} + +static void +xlog_cil_pcp_hpremove ( +struct xfs_cil *cil) +{ +spin_lock (&xlog_cil_pcp_lock); +list_del (&cil->xc_pcp_list); +spin_unlock (&xlog_cil_pcp_lock); +} + +#else /* !CONFIG_HOTPLUG_CPU */ +static inline void xlog_cil_pcp_hpadd (struct xfs_cil *cil) {} +static inline void xlog_cil_pcp_hpremove (struct xfs_cil *cil) {} +#endif + +static void __percpu * +xlog_cil_pcp_alloc ( +struct xfs_cil *cil) +{ +void __percpu *pcp; + + pcp = alloc_percpu(struct xlog_cil_pcp); +if (!pcp) +return NULL ; + +if (xlog_cil_pcp_hpadd (cil) <0 ) { +free_percpu (pcp); +return NULL ; + } +return pcp; +} + +static void +xlog_cil_pcp_free ( +struct xfs_cil *cil, +void __percpu *pcp) +{ +if (!pcp) +return ; +xlog_cil_pcp_hpremove (cil); +free_percpu (pcp); +} +/* * Perform initial CIL structure initialisation. */
@@ -1384,6 +1483,12 @@ xlog_cil_init(if (!cil)return -ENOMEM ; + cil->xc_pcp =xlog_cil_pcp_alloc (cil); +if (!cil->xc_pcp) { +kmem_free (cil); +return -ENOMEM ; + } +INIT_LIST_HEAD (&cil->xc_cil);INIT_LIST_HEAD (&cil->xc_committing);spin_lock_init (&cil->xc_cil_lock);
@@ -1414,6 +1519,7 @@ xlog_cil_destroy(ASSERT (list_empty (&cil->xc_cil));ASSERT (test_bit (XLOG_CIL_EMPTY , &cil->xc_flags)); +xlog_cil_pcp_free (cil, cil->xc_pcp);kmem_free (cil); }
fs/xfs/xfs_log_priv.h +15 -0
@@ -227,6 +227,16 @@ struct xfs_cil_ctx {struct work_struct push_work; }; +/* +* Per-cpu CIL tracking items +*/ +struct xlog_cil_pcp { +uint32_t space_used; +uint32_t curr_res; +struct list_head busy_extents; +struct list_head log_items; +}; +/* * Committed Item List structure *
@@ -260,6 +270,11 @@ struct xfs_cil {wait_queue_head_t xc_commit_wait;xfs_csn_t xc_current_sequence;wait_queue_head_t xc_push_wait;/* background push throttle */ + +void __percpu *xc_pcp;/* percpu CIL structures */ +#ifdef CONFIG_HOTPLUG_CPU +struct list_head xc_pcp_list; +#endif }____cacheline_aligned_in_smp ;/* xc_flags bit values */
include/linux/cpuhotplug.h +1 -0
@@ -52,6 +52,7 @@ enum cpuhp_state {CPUHP_FS_BUFF_DEAD ,CPUHP_PRINTK_DEAD ,CPUHP_MM_MEMCQ_DEAD , +CPUHP_XFS_CIL_DEAD ,CPUHP_PERCPU_CNT_DEAD ,CPUHP_RADIX_DEAD , CPUHP_PAGE_ALLOC_DEAD,
Signature · 1 line
2.31.1
Re: [PATCH 29/39] xfs: introduce per-cpu CIL tracking structure
Quoted 59 lines · Dave Chinner
From: Dave Chinner <[email protected]>
The CIL push lock is highly contended on larger machines, becoming a hard bottleneck that about 700,000 transaction commits/s on >16p machines. To address this, start moving the CIL tracking infrastructure to utilise per-CPU structures.
We need to track the space used, the amount of log reservation space reserved to write the CIL, the log items in the CIL and the busy extents that need to be completed by the CIL commit. This requires a couple of per-cpu counters, an unordered per-cpu list and a globally ordered per-cpu list.
Create a per-cpu structure to hold these and all the management interfaces needed, as well as the hooks to handle hotplug CPUs.
Signed-off-by Dave Chinner <[email protected]>--- fs/xfs/xfs_log_cil.c | 106 +++++++++++++++++++++++++++++++++++++ fs/xfs/xfs_log_priv.h | 15 ++++++ include/linux/cpuhotplug.h | 1 + 3 files changed, 122 insertions(+)fs/xfs/xfs_log_cil.c +28
-0@@ -1370,6 +1370,105 @@ xfs_log_item_in_current_chkpt(return lip->li_seq == cil->xc_ctx->sequence; } +#ifdef CONFIG_HOTPLUG_CPU +static LIST_HEAD (xlog_cil_pcp_list); +static DEFINE_SPINLOCK (xlog_cil_pcp_lock); +static bool xlog_cil_pcp_init; + +/* +* Move dead percpu state to the relevant CIL context structures. +* +* We have to lock the CIL context here to ensure that nothing is modifying +* the percpu state, either addition or removal. Both of these are done under +* the CIL context lock, so grabbing that exclusively here will ensure we can +* safely drain the cilpcp for the CPU that is dying. +*/ +static int +xlog_cil_pcp_dead ( +unsigned int cpu) +{ +struct xfs_cil *cil, *n; + +spin_lock (&xlog_cil_pcp_lock); +list_for_each_entry_safe (cil, n, &xlog_cil_pcp_list, xc_pcp_list) { +spin_unlock (&xlog_cil_pcp_lock); +down_write (&cil->xc_ctx_lock); +/* move stuff on dead CPU to context */ +up_write (&cil->xc_ctx_lock); +spin_lock (&xlog_cil_pcp_lock); + } +spin_unlock (&xlog_cil_pcp_lock);
fs/xfs/xfs_log_cil.c xfs_log_item_in_current_chkpt( · line 1410
Quoted 10 lines
fs/xfs/xfs_log_cil.c +10
-0+return 0 ; +} + +static int +xlog_cil_pcp_hpadd ( +struct xfs_cil *cil) +{ +if (!xlog_cil_pcp_init) { +int ret; + ret = cpuhp_setup_state_nocalls(CPUHP_XFS_CIL_DEAD,
Nit: blank line between variable declarations and code.
Quoted 90 lines
fs/xfs/xfs_log_cil.c +68
-0+"xfs/cil_pcp:dead" ,NULL , + xlog_cil_pcp_dead); +if (ret <0 ) { +xfs_warn (cil->xc_log->l_mp, +"Failed to initialise CIL hotplug, error %d. XFS is non-functional." , + ret); +ASSERT (0 ); +return -ENOMEM ; + } + xlog_cil_pcp_init = true; + } + +INIT_LIST_HEAD (&cil->xc_pcp_list); +spin_lock (&xlog_cil_pcp_lock); +list_add (&cil->xc_pcp_list, &xlog_cil_pcp_list); +spin_unlock (&xlog_cil_pcp_lock); +return 0 ; +} + +static void +xlog_cil_pcp_hpremove ( +struct xfs_cil *cil) +{ +spin_lock (&xlog_cil_pcp_lock); +list_del (&cil->xc_pcp_list); +spin_unlock (&xlog_cil_pcp_lock); +} + +#else /* !CONFIG_HOTPLUG_CPU */ +static inline void xlog_cil_pcp_hpadd (struct xfs_cil *cil) {} +static inline void xlog_cil_pcp_hpremove (struct xfs_cil *cil) {} +#endif + +static void __percpu * +xlog_cil_pcp_alloc ( +struct xfs_cil *cil) +{ +void __percpu *pcp; + + pcp = alloc_percpu(struct xlog_cil_pcp); +if (!pcp) +return NULL ; + +if (xlog_cil_pcp_hpadd (cil) <0 ) { +free_percpu (pcp); +return NULL ; + } +return pcp; +} + +static void +xlog_cil_pcp_free ( +struct xfs_cil *cil, +void __percpu *pcp) +{ +if (!pcp) +return ; +xlog_cil_pcp_hpremove (cil); +free_percpu (pcp); +} +/* * Perform initial CIL structure initialisation. */ @@ -1384,6 +1483,12 @@ xlog_cil_init(if (!cil)return -ENOMEM ; + cil->xc_pcp =xlog_cil_pcp_alloc (cil); +if (!cil->xc_pcp) { +kmem_free (cil); +return -ENOMEM ; + } +INIT_LIST_HEAD (&cil->xc_cil);INIT_LIST_HEAD (&cil->xc_committing);spin_lock_init (&cil->xc_cil_lock);@@ -1414,6 +1519,7 @@ xlog_cil_destroy(ASSERT (list_empty (&cil->xc_cil));ASSERT (test_bit (XLOG_CIL_EMPTY , &cil->xc_flags)); +xlog_cil_pcp_free (cil, cil->xc_pcp);kmem_free (cil); }fs/xfs/xfs_log_priv.h +0
-0
fs/xfs/xfs_log_priv.h struct xfs_cil_ctx { · line 235
Quoted 9 lines
fs/xfs/xfs_log_priv.h +6
-0@@ -227,6 +227,16 @@ struct xfs_cil_ctx {struct work_struct push_work; }; +/* +* Per-cpu CIL tracking items +*/ +struct xlog_cil_pcp { +uint32_t space_used; +uint32_t curr_res;
I don't think these fields need to be in this patch.
Especially since you rename one of them in patch 31.
I think this looks like the skeleton of adding per-cpu structures to the log, though I'll read the next few patches to see if the comments make more sense once you actually starting using the pcpu data.
--D
Quoted 34 lines
fs/xfs/xfs_log_priv.h +9
-0+struct list_head busy_extents; +struct list_head log_items; +}; +/* * Committed Item List structure * @@ -260,6 +270,11 @@ struct xfs_cil {wait_queue_head_t xc_commit_wait;xfs_csn_t xc_current_sequence;wait_queue_head_t xc_push_wait;/* background push throttle */ + +void __percpu *xc_pcp;/* percpu CIL structures */ +#ifdef CONFIG_HOTPLUG_CPU +struct list_head xc_pcp_list; +#endif }____cacheline_aligned_in_smp ;/* xc_flags bit values */ include/linux/cpuhotplug.h +1
-0@@ -52,6 +52,7 @@ enum cpuhp_state {CPUHP_FS_BUFF_DEAD ,CPUHP_PRINTK_DEAD ,CPUHP_MM_MEMCQ_DEAD , +CPUHP_XFS_CIL_DEAD ,CPUHP_PERCPU_CNT_DEAD ,CPUHP_RADIX_DEAD , CPUHP_PAGE_ALLOC_DEAD,Signature · 1 line
2.31.1
PATCH v4 30/39 xfs: implement percpu cil space used calculation
In reply to Dave ChinnerFrom: Dave Chinner <[email protected]>
Now that we have the CIL percpu structures in place, implement the space used counter with a fast sum check similar to the percpu_counter infrastructure.
--- fs/xfs/xfs_log_cil.c | 61 ++++++++++++++++++++++++++++++++++++++----- fs/xfs/xfs_log_priv.h | 2 +- 2 files changed, 55 insertions(+), 8 deletions(-)
fs/xfs/xfs_log_cil.c +54 -7
@@ -76,6 +76,24 @@ xlog_cil_ctx_alloc(void)return ctx; } +/* +* Aggregate the CIL per cpu structures into global counts, lists, etc and +* clear the percpu state ready for the next context to use. +*/ +static void +xlog_cil_pcp_aggregate ( +struct xfs_cil *cil, +struct xfs_cil_ctx *ctx) +{ +struct xlog_cil_pcp *cilpcp; +int cpu; + +for_each_online_cpu (cpu) { + cilpcp =per_cpu_ptr (cil->xc_pcp, cpu); + cilpcp->space_used =0 ; + } +} + static void xlog_cil_ctx_switch( struct xfs_cil *cil,
@@ -433,6 +451,8 @@ xlog_cil_insert_items(struct xfs_log_item *lip;int len =0 ;int iovhdr_res =0 , split_res =0 , ctx_res =0 ; +int space_used; +struct xlog_cil_pcp *cilpcp;ASSERT (tp);
@@ -469,8 +489,9 @@ xlog_cil_insert_items(* * This can steal more than we need, but that's OK. */ + space_used =atomic_read (&ctx->space_used);if (atomic_read (&cil->xc_iclog_hdrs) >0 || - ctx->space_used + len >=XLOG_CIL_BLOCKING_SPACE_LIMIT (log)) { + space_used + len >=XLOG_CIL_BLOCKING_SPACE_LIMIT (log)) {int split_res = log->l_iclog_hsize +sizeof (struct xlog_op_header );if (ctx_res)
@@ -480,16 +501,34 @@ xlog_cil_insert_items(atomic_sub (tp->t_ticket->t_iclog_hdrs, &cil->xc_iclog_hdrs); } +/* +* Update the CIL percpu pointer. This updates the global counter when +* over the percpu batch size or when the CIL is over the space limit. +* This means low lock overhead for normal updates, and when over the +* limit the space used is immediately accounted. This makes enforcing +* the hard limit much more accurate. The per cpu fold threshold is +* based on how close we are to the hard limit. +*/ + cilpcp =get_cpu_ptr (cil->xc_pcp); + cilpcp->space_used += len; +if (space_used >=XLOG_CIL_SPACE_LIMIT (log) || + cilpcp->space_used > + ((XLOG_CIL_BLOCKING_SPACE_LIMIT (log) - space_used) / +num_online_cpus ())) { +atomic_add (cilpcp->space_used, &ctx->space_used); + cilpcp->space_used =0 ; + } +put_cpu_ptr (cilpcp); +spin_lock (&cil->xc_cil_lock); - tp->t_ticket->t_curr_res -= ctx_res + len; ctx->ticket->t_unit_res += ctx_res; ctx->ticket->t_curr_res += ctx_res; - ctx->space_used += len;/* * If we've overrun the reservation, dump the tx details before we move * the log items. Shutdown is imminent... */ + tp->t_ticket->t_curr_res -= ctx_res + len;if (WARN_ON (tp->t_ticket->t_curr_res <0 )) {xfs_warn (log->l_mp,"Transaction log reservation overrun:" ); xfs_warn(log->l_mp,
@@ -846,6 +885,8 @@ xlog_cil_push_work(xfs_flush_bdev_async (&bio, log->l_mp->m_ddev_targp->bt_bdev, &bdev_flush); +xlog_cil_pcp_aggregate (cil, ctx); +/* * Pull all the log vectors off the items in the CIL, and remove the * items from the CIL. We don't need the CIL lock here because it's only
@@ -1043,6 +1084,7 @@ xlog_cil_push_background( struct xlog *log) __releases(cil->xc_ctx_lock) {struct xfs_cil *cil = log->l_cilp; +int space_used =atomic_read (&cil->xc_ctx->space_used);/* * The cil won't be empty because we are called while holding the
@@ -1055,7 +1097,7 @@ xlog_cil_push_background(* Don't do a background push if we haven't used up all the * space available yet. */ -if (cil->xc_ctx->space_used <XLOG_CIL_SPACE_LIMIT (log)) { +if (space_used <XLOG_CIL_SPACE_LIMIT (log)) {up_read (&cil->xc_ctx_lock);return ; }
@@ -1084,10 +1126,10 @@ xlog_cil_push_background(* The ctx->xc_push_lock provides the serialisation necessary for safely * using the lockless waitqueue_active() check in this context. */ -if (cil->xc_ctx->space_used >=XLOG_CIL_BLOCKING_SPACE_LIMIT (log) || +if (space_used >=XLOG_CIL_BLOCKING_SPACE_LIMIT (log) ||waitqueue_active (&cil->xc_push_wait)) {trace_xfs_log_cil_wait (log, cil->xc_ctx->ticket); -ASSERT (cil->xc_ctx->space_used < log->l_logsize); +ASSERT (space_used < log->l_logsize);xlog_wait (&cil->xc_push_wait, &cil->xc_push_lock);return ; }
@@ -1391,9 +1433,14 @@ xlog_cil_pcp_dead(spin_lock (&xlog_cil_pcp_lock);list_for_each_entry_safe (cil, n, &xlog_cil_pcp_list, xc_pcp_list) { +struct xlog_cil_pcp *cilpcp =per_cpu_ptr (cil->xc_pcp, cpu); +spin_unlock (&xlog_cil_pcp_lock);down_write (&cil->xc_ctx_lock); -/* move stuff on dead CPU to context */ + +atomic_add (cilpcp->space_used, &cil->xc_ctx->space_used); + cilpcp->space_used =0 ; +up_write (&cil->xc_ctx_lock);spin_lock (&xlog_cil_pcp_lock); }
fs/xfs/xfs_log_priv.h +1 -1
@@ -218,7 +218,7 @@ struct xfs_cil_ctx {xfs_lsn_t start_lsn;/* first LSN of chkpt commit */ xfs_lsn_t commit_lsn;/* chkpt commit record lsn */ struct xlog_ticket *ticket;/* chkpt ticket */ -int space_used;/* aggregate size of regions */ +atomic_t space_used;/* aggregate size of regions */ struct list_head busy_extents;/* busy extents in chkpt */ struct xfs_log_vec *lv_chain;/* logvecs being pushed */ struct list_head iclog_entry;
Signature · 1 line
2.31.1
Re: [PATCH 30/39] xfs: implement percpu cil space used calculation
Quoted 25 lines · Dave Chinner
From: Dave Chinner <[email protected]>
Now that we have the CIL percpu structures in place, implement the space used counter with a fast sum check similar to the
percpu_counterinfrastructure.Signed-off-by Dave Chinner <[email protected]>--- fs/xfs/xfs_log_cil.c | 61 ++++++++++++++++++++++++++++++++++++++----- fs/xfs/xfs_log_priv.h | 2 +- 2 files changed, 55 insertions(+), 8 deletions(-)fs/xfs/xfs_log_cil.c +5
-0@@ -76,6 +76,24 @@ xlog_cil_ctx_alloc(void)return ctx; } +/* +* Aggregate the CIL per cpu structures into global counts, lists, etc and +* clear the percpu state ready for the next context to use. +*/ +static void
fs/xfs/xfs_log_cil.c xlog_cil_ctx_alloc(void) · line 93
Quoted 10 lines
fs/xfs/xfs_log_cil.c +10
-0+xlog_cil_pcp_aggregate ( + struct xfs_cil *cil, + struct xfs_cil_ctx *ctx) +{ + struct xlog_cil_pcp *cilpcp; +int cpu; + +for_each_online_cpu (cpu) { + cilpcp =per_cpu_ptr (cil->xc_pcp, cpu); + cilpcp->space_used =0 ;
How does this aggregate anything? All I see here is zeroing a counter? I see that we /can/ add the percpu space_used counter to the cil context if we're over the space limits, but I don't actually see where...
Quoted 44 lines
fs/xfs/xfs_log_cil.c +21
-1+ } +} + static void xlog_cil_ctx_switch( struct xfs_cil *cil,@@ -433,6 +451,8 @@ xlog_cil_insert_items(struct xfs_log_item *lip;int len =0 ;int iovhdr_res =0 , split_res =0 , ctx_res =0 ; +int space_used; +struct xlog_cil_pcp *cilpcp;ASSERT (tp);@@ -469,8 +489,9 @@ xlog_cil_insert_items(* * This can steal more than we need, but that's OK. */ + space_used =atomic_read (&ctx->space_used);if (atomic_read (&cil->xc_iclog_hdrs) >0 || - ctx->space_used + len >=XLOG_CIL_BLOCKING_SPACE_LIMIT (log)) { + space_used + len >=XLOG_CIL_BLOCKING_SPACE_LIMIT (log)) {int split_res = log->l_iclog_hsize +sizeof (struct xlog_op_header );if (ctx_res)@@ -480,16 +501,34 @@ xlog_cil_insert_items(atomic_sub (tp->t_ticket->t_iclog_hdrs, &cil->xc_iclog_hdrs); } +/* +* Update the CIL percpu pointer. This updates the global counter when +* over the percpu batch size or when the CIL is over the space limit. +* This means low lock overhead for normal updates, and when over the +* limit the space used is immediately accounted. This makes enforcing +* the hard limit much more accurate. The per cpu fold threshold is +* based on how close we are to the hard limit. +*/ + cilpcp =get_cpu_ptr (cil->xc_pcp); + cilpcp->space_used += len; +if (space_used >=XLOG_CIL_SPACE_LIMIT (log) || + cilpcp->space_used > + ((XLOG_CIL_BLOCKING_SPACE_LIMIT (log) - space_used) / +num_online_cpus ())) {
fs/xfs/xfs_log_cil.c xlog_cil_insert_items( · line 487
Quoted 10 lines
fs/xfs/xfs_log_cil.c +5
-2+atomic_add (cilpcp->space_used, &ctx->space_used); + cilpcp->space_used =0 ; + } +put_cpu_ptr (cilpcp); +spin_lock (&cil->xc_cil_lock); - tp->t_ticket->t_curr_res -= ctx_res + len; ctx->ticket->t_unit_res += ctx_res; ctx->ticket->t_curr_res += ctx_res; - ctx->space_used += len;
...this update happens if we're not over the space limit?
--D
Quoted 80 lines
fs/xfs/xfs_log_cil.c +13
-4/* * If we've overrun the reservation, dump the tx details before we move * the log items. Shutdown is imminent... */ + tp->t_ticket->t_curr_res -= ctx_res + len;if (WARN_ON (tp->t_ticket->t_curr_res <0 )) {xfs_warn (log->l_mp,"Transaction log reservation overrun:" ); xfs_warn(log->l_mp,@@ -846,6 +885,8 @@ xlog_cil_push_work(xfs_flush_bdev_async (&bio, log->l_mp->m_ddev_targp->bt_bdev, &bdev_flush); +xlog_cil_pcp_aggregate (cil, ctx); +/* * Pull all the log vectors off the items in the CIL, and remove the * items from the CIL. We don't need the CIL lock here because it's only @@ -1043,6 +1084,7 @@ xlog_cil_push_background( struct xlog *log) __releases(cil->xc_ctx_lock) {struct xfs_cil *cil = log->l_cilp; +int space_used =atomic_read (&cil->xc_ctx->space_used);/* * The cil won't be empty because we are called while holding the @@ -1055,7 +1097,7 @@ xlog_cil_push_background(* Don't do a background push if we haven't used up all the * space available yet. */ -if (cil->xc_ctx->space_used <XLOG_CIL_SPACE_LIMIT (log)) { +if (space_used <XLOG_CIL_SPACE_LIMIT (log)) {up_read (&cil->xc_ctx_lock);return ; }@@ -1084,10 +1126,10 @@ xlog_cil_push_background(* The ctx->xc_push_lock provides the serialisation necessary for safely * using the lockless waitqueue_active() check in this context. */ -if (cil->xc_ctx->space_used >=XLOG_CIL_BLOCKING_SPACE_LIMIT (log) || +if (space_used >=XLOG_CIL_BLOCKING_SPACE_LIMIT (log) ||waitqueue_active (&cil->xc_push_wait)) {trace_xfs_log_cil_wait (log, cil->xc_ctx->ticket); -ASSERT (cil->xc_ctx->space_used < log->l_logsize); +ASSERT (space_used < log->l_logsize);xlog_wait (&cil->xc_push_wait, &cil->xc_push_lock);return ; }@@ -1391,9 +1433,14 @@ xlog_cil_pcp_dead(spin_lock (&xlog_cil_pcp_lock);list_for_each_entry_safe (cil, n, &xlog_cil_pcp_list, xc_pcp_list) { +struct xlog_cil_pcp *cilpcp =per_cpu_ptr (cil->xc_pcp, cpu); +spin_unlock (&xlog_cil_pcp_lock);down_write (&cil->xc_ctx_lock); -/* move stuff on dead CPU to context */ + +atomic_add (cilpcp->space_used, &cil->xc_ctx->space_used); + cilpcp->space_used =0 ; +up_write (&cil->xc_ctx_lock);spin_lock (&xlog_cil_pcp_lock); }fs/xfs/xfs_log_priv.h +1
-1@@ -218,7 +218,7 @@ struct xfs_cil_ctx {xfs_lsn_t start_lsn;/* first LSN of chkpt commit */ xfs_lsn_t commit_lsn;/* chkpt commit record lsn */ struct xlog_ticket *ticket;/* chkpt ticket */ -int space_used;/* aggregate size of regions */ +atomic_t space_used;/* aggregate size of regions */ struct list_head busy_extents;/* busy extents in chkpt */ struct xfs_log_vec *lv_chain;/* logvecs being pushed */ struct list_head iclog_entry;Signature · 1 line
2.31.1
Re: [PATCH 30/39] xfs: implement percpu cil space used calculation
Quoted 30 lines · Darrick J. Wong
Quoted 29 lines · Dave Chinner
From: Dave Chinner <[email protected]>
Now that we have the CIL percpu structures in place, implement the space used counter with a fast sum check similar to the
percpu_counterinfrastructure.Signed-off-by Dave Chinner <[email protected]>--- fs/xfs/xfs_log_cil.c | 61 ++++++++++++++++++++++++++++++++++++++----- fs/xfs/xfs_log_priv.h | 2 +- 2 files changed, 55 insertions(+), 8 deletions(-)fs/xfs/xfs_log_cil.c +9
-0@@ -76,6 +76,24 @@ xlog_cil_ctx_alloc(void)return ctx; } +/* +* Aggregate the CIL per cpu structures into global counts, lists, etc and +* clear the percpu state ready for the next context to use. +*/ +static void +xlog_cil_pcp_aggregate ( +struct xfs_cil *cil, +struct xfs_cil_ctx *ctx) +{
Quoted 8 lines
fs/xfs/xfs_log_cil.c xlog_cil_ctx_alloc(void) · line 93
Quoted 6 lines
fs/xfs/xfs_log_cil.c +6
-0+struct xlog_cil_pcp *cilpcp; +int cpu; + +for_each_online_cpu (cpu) { + cilpcp =per_cpu_ptr (cil->xc_pcp, cpu); + cilpcp->space_used =0 ;How does this aggregate anything? All I see here is zeroing a counter?
Yup, zeroing all the percpu counters is an aggregation function....
By definition "aggregate != sum".
An aggregate is formed by the collection of discrete units into a larger whole; the collective definition involves manipulating all discrete units as a single whole entity. e.g. a percpu counter is an aggregate of percpu variables that, via aggregation, can sum the discrete variables into a single value. IOWs, percpu_counter_sum() is an aggregation function that sums...
Quoted 2 lines
I see that we /can/ add the percpu
space_usedcounter to the cil context if we're over the space limits, but I don't actually see where...
In this case, the global CIL space used counter is summed by the per-cpu counter update context and not an aggregation context. For it to work as a global counter since a distinct point in time, it needs an aggregation operation that zeros all the discrete units of the counter at a single point in time. IOWs, the aggregation function of this counter is a zeroing operation, not a summing operation. This is what xlog_cil_pcp_aggregate() is doing here.
Put simply, an aggregation function is not a summing function, but a function that operates on all the discrete units of the aggregate so that it can operate correctly as a single unit....
I don't know of a better way of describing what this function does. At the end of the series, this function will zero some units. In other cases it will sum units. In some cases it will do both. Not to mention that it will merge discrete lists into a global list. And so on. The only common thing between these operations is that they are all aggregation functions that allow the CIL context to operate as a whole unit...
If you've got a better name, then I'm all ears :)
....
Quoted 22 lines
Quoted 22 lines
fs/xfs/xfs_log_cil.c +18
-0@@ -480,16 +501,34 @@ xlog_cil_insert_items(atomic_sub (tp->t_ticket->t_iclog_hdrs, &cil->xc_iclog_hdrs); } +/* +* Update the CIL percpu pointer. This updates the global counter when +* over the percpu batch size or when the CIL is over the space limit. +* This means low lock overhead for normal updates, and when over the +* limit the space used is immediately accounted. This makes enforcing +* the hard limit much more accurate. The per cpu fold threshold is +* based on how close we are to the hard limit. +*/ + cilpcp =get_cpu_ptr (cil->xc_pcp); + cilpcp->space_used += len; +if (space_used >=XLOG_CIL_SPACE_LIMIT (log) || + cilpcp->space_used > + ((XLOG_CIL_BLOCKING_SPACE_LIMIT (log) - space_used) / +num_online_cpus ())) { +atomic_add (cilpcp->space_used, &ctx->space_used); + cilpcp->space_used =0 ; + } +put_cpu_ptr (cilpcp);
Quoted 8 lines
fs/xfs/xfs_log_cil.c xlog_cil_insert_items( · line 487
Quoted 6 lines
fs/xfs/xfs_log_cil.c +1
-2+spin_lock (&cil->xc_cil_lock); - tp->t_ticket->t_curr_res -= ctx_res + len; ctx->ticket->t_unit_res += ctx_res; ctx->ticket->t_curr_res += ctx_res; - ctx->space_used += len;...this update happens if we're not over the space limit?
It's the second case in the above if statement. As the space used in the percpu pointer goes over it's fraction of the remaining space limit (limit remaining / num_cpus_online), then it adds the pcp counter back into the global counter. Essentially it is:
if (over push threshold ||
Quoted 1 line
fs/xfs/xfs_log_cil.c xlog_cil_insert_items( · line 526
Quoted 1 line
fs/xfs/xfs_log_cil.c +0
-0>>>> pcp->used > ((hard limit - ctx->space_used) / cpus)) {
ctx->space_used += pcp->used; pcp->used = 0; }
Hence, to begin with, the percpu counter is allowed to sum a large chunk of space before it trips the per CPU summing threshold. When summing occurs, the per-cpu threshold goes down, meaning there pcp counters will trip sooner in the next cycle.
IOWs, the summing threshold gets closer to zero the closer the global count gets to the hard limit. Hence when there's lots of space available, we have little summing contention, but when we are close to the blocking limit we essentially update the global counter on every modification.
As such, we get scalability when the CIL is empty by trading off accuracy, but we get accuracy when it is nearing full by trading off scalability. We might need to tweak it for really large CPU counts (maybe use log2(num_online_cpus()), but fundamentally the algorithm is designed to scale according to how close we are to the push thresholds....
Cheers,
Dave.
Signature · 2 lines
Dave Chinner [email protected]
Re: [PATCH 30/39] xfs: implement percpu cil space used calculation
Quoted 62 lines · Dave Chinner
Quoted 30 lines · Darrick J. Wong
Quoted 29 lines · Dave Chinner
From: Dave Chinner <[email protected]>
Now that we have the CIL percpu structures in place, implement the space used counter with a fast sum check similar to the
percpu_counterinfrastructure.Signed-off-by Dave Chinner <[email protected]>--- fs/xfs/xfs_log_cil.c | 61 ++++++++++++++++++++++++++++++++++++++----- fs/xfs/xfs_log_priv.h | 2 +- 2 files changed, 55 insertions(+), 8 deletions(-)fs/xfs/xfs_log_cil.c +9
-0@@ -76,6 +76,24 @@ xlog_cil_ctx_alloc(void)return ctx; } +/* +* Aggregate the CIL per cpu structures into global counts, lists, etc and +* clear the percpu state ready for the next context to use. +*/ +static void +xlog_cil_pcp_aggregate ( +struct xfs_cil *cil, +struct xfs_cil_ctx *ctx) +{Quoted 8 lines
fs/xfs/xfs_log_cil.c xlog_cil_ctx_alloc(void) · line 93
Quoted 6 lines
fs/xfs/xfs_log_cil.c +6
-0+struct xlog_cil_pcp *cilpcp; +int cpu; + +for_each_online_cpu (cpu) { + cilpcp =per_cpu_ptr (cil->xc_pcp, cpu); + cilpcp->space_used =0 ;How does this aggregate anything? All I see here is zeroing a counter?
Yup, zeroing all the percpu counters is an aggregation function....
By definition "aggregate != sum".
An aggregate is formed by the collection of discrete units into a larger whole; the collective definition involves manipulating all discrete units as a single whole entity. e.g. a percpu counter is an aggregate of percpu variables that, via aggregation, can sum the discrete variables into a single value. IOWs,
percpu_counter_sum()is an aggregation function that sums...Quoted 2 lines
I see that we /can/ add the percpu
space_usedcounter to the cil context if we're over the space limits, but I don't actually see where...In this case, the global CIL space used counter is summed by the per-cpu counter update context and not an aggregation context. For it to work as a global counter since a distinct point in time, it needs an aggregation operation that zeros all the discrete units of the counter at a single point in time. IOWs, the aggregation function of this counter is a zeroing operation, not a summing operation. This is what
xlog_cil_pcp_aggregate()is doing here.
Quoted 3 lines
Put simply, an aggregation function is not a summing function, but a function that operates on all the discrete units of the aggregate so that it can operate correctly as a single unit....
<nod> I grok what 'aggregate' means as a general term, though perhaps I was too quick to associate it with 'sum' here.
Quoted 7 lines
I don't know of a better way of describing what this function does. At the end of the series, this function will zero some units. In other cases it will sum units. In some cases it will do both. Not to mention that it will merge discrete lists into a global list. And so on. The only common thing between these operations is that they are all aggregation functions that allow the CIL context to operate as a whole unit...
*Oh* I think I realized where my understanding gap lies.
space_used isn't part of some global space accounting scheme that has to be kept accurate. It's a per-cil-context variable that we used to throttle incoming commits when the current context starts to get full. That's why _cil_insert_items only bothers to add the per-cpu space_used to the ctx space_used if either (a) the current cpu has used up more than its slice of space or (b) enough cpus have hit (a) to push the ctx's space_used above the XLOG_CIL_SPACE_LIMIT. If no cpus have hit
(a) then the current cil context still has plenty of space and there isn't any need to throttle the frontend.
By the time we get to the aggregation step in _cil_push_work we've already decided to install a new context and write the current context to disk. We don't care about throttling the frontend on this (closed) context and we hold xc_ctx_lock so nobody can see the new context. That's why the aggregate function zeroes the per-cpu space_used.
Quoted 1 line
If you've got a better name, then I'm all ears :)
Not really.
Quoted 49 lines
....
Quoted 22 lines
Quoted 22 lines
fs/xfs/xfs_log_cil.c +18
-0@@ -480,16 +501,34 @@ xlog_cil_insert_items(atomic_sub (tp->t_ticket->t_iclog_hdrs, &cil->xc_iclog_hdrs); } +/* +* Update the CIL percpu pointer. This updates the global counter when +* over the percpu batch size or when the CIL is over the space limit. +* This means low lock overhead for normal updates, and when over the +* limit the space used is immediately accounted. This makes enforcing +* the hard limit much more accurate. The per cpu fold threshold is +* based on how close we are to the hard limit. +*/ + cilpcp =get_cpu_ptr (cil->xc_pcp); + cilpcp->space_used += len; +if (space_used >=XLOG_CIL_SPACE_LIMIT (log) || + cilpcp->space_used > + ((XLOG_CIL_BLOCKING_SPACE_LIMIT (log) - space_used) / +num_online_cpus ())) { +atomic_add (cilpcp->space_used, &ctx->space_used); + cilpcp->space_used =0 ; + } +put_cpu_ptr (cilpcp);Quoted 8 lines
fs/xfs/xfs_log_cil.c xlog_cil_insert_items( · line 487
Quoted 6 lines
fs/xfs/xfs_log_cil.c +1
-2+spin_lock (&cil->xc_cil_lock); - tp->t_ticket->t_curr_res -= ctx_res + len; ctx->ticket->t_unit_res += ctx_res; ctx->ticket->t_curr_res += ctx_res; - ctx->space_used += len;...this update happens if we're not over the space limit?
It's the second case in the above if statement. As the space used in the percpu pointer goes over it's fraction of the remaining space limit (limit remaining /
num_cpus_online), then it adds the pcp counter back into the global counter. Essentially it is:if (over push threshold ||Quoted 1 line
fs/xfs/xfs_log_cil.c xlog_cil_insert_items( · line 526
Quoted 1 line
fs/xfs/xfs_log_cil.c +0
-0>>>> pcp->used > ((hard limit - ctx->space_used) / cpus)) {ctx->space_used += pcp->used; pcp->used = 0; }Hence, to begin with, the percpu counter is allowed to sum a large chunk of space before it trips the per CPU summing threshold. When summing occurs, the per-cpu threshold goes down, meaning there pcp counters will trip sooner in the next cycle.
Quoted 5 lines
IOWs, the summing threshold gets closer to zero the closer the global count gets to the hard limit. Hence when there's lots of space available, we have little summing contention, but when we are close to the blocking limit we essentially update the global counter on every modification.
Ok, I think I get it now. The statement "ctx->space_used = 0" is part of clearing the percpu state and is not part of aggregating the CIL per cpu structure into the context.
So assuming that I grokked it all on the second try, maybe a comment is in order for the aggregate function?
/* * We're in the middle of switching cil contexts. Reset the * counter we use to detect when the current context is nearing * full. */ ctx->space_used = 0;
Quoted 6 lines
As such, we get scalability when the CIL is empty by trading off accuracy, but we get accuracy when it is nearing full by trading off scalability. We might need to tweak it for really large CPU counts (maybe use log2(num_online_cpus()), but fundamentally the algorithm is designed to scale according to how close we are to the push thresholds....
<nod> I suppose if you had a very large number of CPUs and a very small log then the slivers could be zero ... which just means we'd just lose performance due to overly-careful accounting. I wonder if you want to leave a breadcrumb to warn if:
XLOG_CIL_BLOCKING_SPACE_LIMIT() / num_online_cpus() == 0
just in case someone ever wanders in with such a configuration?
--D
Quoted 7 lines
Cheers,
Dave.
Signature · 2 lines
Dave Chinner [email protected]
Re: [PATCH 30/39] xfs: implement percpu cil space used calculation
Quoted 90 lines · Darrick J. Wong
Quoted 62 lines · Dave Chinner
Quoted 30 lines · Darrick J. Wong
Quoted 29 lines · Dave Chinner
From: Dave Chinner <[email protected]>
Now that we have the CIL percpu structures in place, implement the space used counter with a fast sum check similar to the
percpu_counterinfrastructure.Signed-off-by Dave Chinner <[email protected]>--- fs/xfs/xfs_log_cil.c | 61 ++++++++++++++++++++++++++++++++++++++----- fs/xfs/xfs_log_priv.h | 2 +- 2 files changed, 55 insertions(+), 8 deletions(-)fs/xfs/xfs_log_cil.c +9
-0@@ -76,6 +76,24 @@ xlog_cil_ctx_alloc(void)return ctx; } +/* +* Aggregate the CIL per cpu structures into global counts, lists, etc and +* clear the percpu state ready for the next context to use. +*/ +static void +xlog_cil_pcp_aggregate ( +struct xfs_cil *cil, +struct xfs_cil_ctx *ctx) +{Quoted 8 lines
fs/xfs/xfs_log_cil.c xlog_cil_ctx_alloc(void) · line 93
Quoted 6 lines
fs/xfs/xfs_log_cil.c +6
-0+struct xlog_cil_pcp *cilpcp; +int cpu; + +for_each_online_cpu (cpu) { + cilpcp =per_cpu_ptr (cil->xc_pcp, cpu); + cilpcp->space_used =0 ;How does this aggregate anything? All I see here is zeroing a counter?
Yup, zeroing all the percpu counters is an aggregation function....
By definition "aggregate != sum".
An aggregate is formed by the collection of discrete units into a larger whole; the collective definition involves manipulating all discrete units as a single whole entity. e.g. a percpu counter is an aggregate of percpu variables that, via aggregation, can sum the discrete variables into a single value. IOWs,
percpu_counter_sum()is an aggregation function that sums...Quoted 2 lines
I see that we /can/ add the percpu
space_usedcounter to the cil context if we're over the space limits, but I don't actually see where...In this case, the global CIL space used counter is summed by the per-cpu counter update context and not an aggregation context. For it to work as a global counter since a distinct point in time, it needs an aggregation operation that zeros all the discrete units of the counter at a single point in time. IOWs, the aggregation function of this counter is a zeroing operation, not a summing operation. This is what
xlog_cil_pcp_aggregate()is doing here.Quoted 3 lines
Put simply, an aggregation function is not a summing function, but a function that operates on all the discrete units of the aggregate so that it can operate correctly as a single unit....
<nod> I grok what 'aggregate' means as a general term, though perhaps I was too quick to associate it with 'sum' here.
Quoted 7 lines
I don't know of a better way of describing what this function does. At the end of the series, this function will zero some units. In other cases it will sum units. In some cases it will do both. Not to mention that it will merge discrete lists into a global list. And so on. The only common thing between these operations is that they are all aggregation functions that allow the CIL context to operate as a whole unit...
*Oh* I think I realized where my understanding gap lies.
space_usedisn't part of some global space accounting scheme that has to be kept accurate. It's a per-cil-context variable that we used to throttle incoming commits when the current context starts to get full. That's why_cil_insert_itemsonly bothers to add the per-cpuspace_usedto the ctxspace_usedif either (a) the current cpu has used up more than its slice of space or (b) enough cpus have hit (a) to push the ctx'sspace_usedabove theXLOG_CIL_SPACE_LIMIT. If no cpus have hit
(a) then the current cil context still has plenty of space and there isn't any need to throttle the frontend.
Quoted 5 lines
By the time we get to the aggregation step in
_cil_push_workwe've already decided to install a new context and write the current context to disk. We don't care about throttling the frontend on this (closed) context and we holdxc_ctx_lockso nobody can see the new context. That's why the aggregate function zeroes the per-cpuspace_used.
Exactly.
Quoted 49 lines
Quoted 43 lines
Quoted 18 lines
Quoted 18 lines
fs/xfs/xfs_log_cil.c +18
-0+/* +* Update the CIL percpu pointer. This updates the global counter when +* over the percpu batch size or when the CIL is over the space limit. +* This means low lock overhead for normal updates, and when over the +* limit the space used is immediately accounted. This makes enforcing +* the hard limit much more accurate. The per cpu fold threshold is +* based on how close we are to the hard limit. +*/ + cilpcp =get_cpu_ptr (cil->xc_pcp); + cilpcp->space_used += len; +if (space_used >=XLOG_CIL_SPACE_LIMIT (log) || + cilpcp->space_used > + ((XLOG_CIL_BLOCKING_SPACE_LIMIT (log) - space_used) / +num_online_cpus ())) { +atomic_add (cilpcp->space_used, &ctx->space_used); + cilpcp->space_used =0 ; + } +put_cpu_ptr (cilpcp);Quoted 8 lines
fs/xfs/xfs_log_cil.c xlog_cil_insert_items( · line 487
Quoted 6 lines
fs/xfs/xfs_log_cil.c +1
-2+spin_lock (&cil->xc_cil_lock); - tp->t_ticket->t_curr_res -= ctx_res + len; ctx->ticket->t_unit_res += ctx_res; ctx->ticket->t_curr_res += ctx_res; - ctx->space_used += len;...this update happens if we're not over the space limit?
It's the second case in the above if statement. As the space used in the percpu pointer goes over it's fraction of the remaining space limit (limit remaining /
num_cpus_online), then it adds the pcp counter back into the global counter. Essentially it is:if (over push threshold ||Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
pcp->used > ((hard limit - ctx->space_used) / cpus)) {ctx->space_used += pcp->used; pcp->used = 0; }Hence, to begin with, the percpu counter is allowed to sum a large chunk of space before it trips the per CPU summing threshold. When summing occurs, the per-cpu threshold goes down, meaning there pcp counters will trip sooner in the next cycle.
Quoted 5 lines
IOWs, the summing threshold gets closer to zero the closer the global count gets to the hard limit. Hence when there's lots of space available, we have little summing contention, but when we are close to the blocking limit we essentially update the global counter on every modification.
Quoted 3 lines
Ok, I think I get it now. The statement "
ctx->space_used= 0" is part of clearing the percpu state and is not part of aggregating the CIL per cpu structure into the context.
If you mean the 'cilpcp->space_used = 0' statement in the _aggregate() function, then yes.
We don't actually ever zero the ctx->space_used because it always is initialised to zero by allocation of a new context and switching to it...
Quoted 9 lines
So assuming that I grokked it all on the second try, maybe a comment is in order for the aggregate function?
/* * We're in the middle of switching cil contexts. Reset the * counter we use to detect when the current context is nearing * full. */ ctx->space_used = 0;
Hmmmm - I'm not sure where you are asking I put this comment...
Quoted 7 lines
Quoted 6 lines
As such, we get scalability when the CIL is empty by trading off accuracy, but we get accuracy when it is nearing full by trading off scalability. We might need to tweak it for really large CPU counts (maybe use log2(num_online_cpus()), but fundamentally the algorithm is designed to scale according to how close we are to the push thresholds....
Quoted 8 lines
<nod> I suppose if you had a very large number of CPUs and a very small log then the slivers could be zero ... which just means we'd just lose performance due to overly-careful accounting. I wonder if you want to leave a breadcrumb to warn if:
XLOG_CIL_BLOCKING_SPACE_LIMIT()/num_online_cpus()== 0just in case someone ever wanders in with such a configuration?
I don't think that can happen. The smallest blocking size limit will be a quarter of the smallest log size which puts us around 800KB as the blocking limit. I can't see us supporting 800k CPUs any time soon, and even if you have a couple of hundred thousand CPUs, you're going to have lots of other basic problems trying to do concurrent operations on such a tiny log before we even consider this threshold...
I'm not too worried about scalability and performance on tiny logs - if they result in an atomic update every transaction, it'll still be sufficient to scale to the concurrency the transaction reservation code will let into the commit path in the first place...
Cheers,
Dave.
Signature · 2 lines
Dave Chinner [email protected]
Re: [PATCH 30/39] xfs: implement percpu cil space used calculation
Quoted 91 lines · Dave Chinner
Quoted 90 lines · Darrick J. Wong
Quoted 62 lines · Dave Chinner
Quoted 30 lines · Darrick J. Wong
Quoted 29 lines · Dave Chinner
From: Dave Chinner <[email protected]>
Now that we have the CIL percpu structures in place, implement the space used counter with a fast sum check similar to the
percpu_counterinfrastructure.Signed-off-by Dave Chinner <[email protected]>--- fs/xfs/xfs_log_cil.c | 61 ++++++++++++++++++++++++++++++++++++++----- fs/xfs/xfs_log_priv.h | 2 +- 2 files changed, 55 insertions(+), 8 deletions(-)fs/xfs/xfs_log_cil.c +9
-0@@ -76,6 +76,24 @@ xlog_cil_ctx_alloc(void)return ctx; } +/* +* Aggregate the CIL per cpu structures into global counts, lists, etc and +* clear the percpu state ready for the next context to use. +*/ +static void +xlog_cil_pcp_aggregate ( +struct xfs_cil *cil, +struct xfs_cil_ctx *ctx) +{Quoted 8 lines
fs/xfs/xfs_log_cil.c xlog_cil_ctx_alloc(void) · line 93
Quoted 6 lines
fs/xfs/xfs_log_cil.c +6
-0+struct xlog_cil_pcp *cilpcp; +int cpu; + +for_each_online_cpu (cpu) { + cilpcp =per_cpu_ptr (cil->xc_pcp, cpu); + cilpcp->space_used =0 ;How does this aggregate anything? All I see here is zeroing a counter?
Yup, zeroing all the percpu counters is an aggregation function....
By definition "aggregate != sum".
An aggregate is formed by the collection of discrete units into a larger whole; the collective definition involves manipulating all discrete units as a single whole entity. e.g. a percpu counter is an aggregate of percpu variables that, via aggregation, can sum the discrete variables into a single value. IOWs,
percpu_counter_sum()is an aggregation function that sums...Quoted 2 lines
I see that we /can/ add the percpu
space_usedcounter to the cil context if we're over the space limits, but I don't actually see where...In this case, the global CIL space used counter is summed by the per-cpu counter update context and not an aggregation context. For it to work as a global counter since a distinct point in time, it needs an aggregation operation that zeros all the discrete units of the counter at a single point in time. IOWs, the aggregation function of this counter is a zeroing operation, not a summing operation. This is what
xlog_cil_pcp_aggregate()is doing here.Quoted 3 lines
Put simply, an aggregation function is not a summing function, but a function that operates on all the discrete units of the aggregate so that it can operate correctly as a single unit....
<nod> I grok what 'aggregate' means as a general term, though perhaps I was too quick to associate it with 'sum' here.
Quoted 7 lines
I don't know of a better way of describing what this function does. At the end of the series, this function will zero some units. In other cases it will sum units. In some cases it will do both. Not to mention that it will merge discrete lists into a global list. And so on. The only common thing between these operations is that they are all aggregation functions that allow the CIL context to operate as a whole unit...
*Oh* I think I realized where my understanding gap lies.
space_usedisn't part of some global space accounting scheme that has to be kept accurate. It's a per-cil-context variable that we used to throttle incoming commits when the current context starts to get full. That's why_cil_insert_itemsonly bothers to add the per-cpuspace_usedto the ctxspace_usedif either (a) the current cpu has used up more than its slice of space or (b) enough cpus have hit (a) to push the ctx'sspace_usedabove theXLOG_CIL_SPACE_LIMIT. If no cpus have hit
(a) then the current cil context still has plenty of space and there isn't any need to throttle the frontend.
Quoted 7 lines
Quoted 5 lines
By the time we get to the aggregation step in
_cil_push_workwe've already decided to install a new context and write the current context to disk. We don't care about throttling the frontend on this (closed) context and we holdxc_ctx_lockso nobody can see the new context. That's why the aggregate function zeroes the per-cpuspace_used.Exactly.
oh good!
Quoted 49 lines
Quoted 49 lines
Quoted 48 lines
Quoted 18 lines
Quoted 18 lines
fs/xfs/xfs_log_cil.c +18
-0+/* +* Update the CIL percpu pointer. This updates the global counter when +* over the percpu batch size or when the CIL is over the space limit. +* This means low lock overhead for normal updates, and when over the +* limit the space used is immediately accounted. This makes enforcing +* the hard limit much more accurate. The per cpu fold threshold is +* based on how close we are to the hard limit. +*/ + cilpcp =get_cpu_ptr (cil->xc_pcp); + cilpcp->space_used += len; +if (space_used >=XLOG_CIL_SPACE_LIMIT (log) || + cilpcp->space_used > + ((XLOG_CIL_BLOCKING_SPACE_LIMIT (log) - space_used) / +num_online_cpus ())) { +atomic_add (cilpcp->space_used, &ctx->space_used); + cilpcp->space_used =0 ; + } +put_cpu_ptr (cilpcp);Quoted 8 lines
fs/xfs/xfs_log_cil.c xlog_cil_insert_items( · line 487
Quoted 6 lines
fs/xfs/xfs_log_cil.c +1
-2+spin_lock (&cil->xc_cil_lock); - tp->t_ticket->t_curr_res -= ctx_res + len; ctx->ticket->t_unit_res += ctx_res; ctx->ticket->t_curr_res += ctx_res; - ctx->space_used += len;...this update happens if we're not over the space limit?
It's the second case in the above if statement. As the space used in the percpu pointer goes over it's fraction of the remaining space limit (limit remaining /
num_cpus_online), then it adds the pcp counter back into the global counter. Essentially it is:if (over push threshold ||Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
pcp->used > ((hard limit - ctx->space_used) / cpus)) {ctx->space_used += pcp->used; pcp->used = 0; }Hence, to begin with, the percpu counter is allowed to sum a large chunk of space before it trips the per CPU summing threshold. When summing occurs, the per-cpu threshold goes down, meaning there pcp counters will trip sooner in the next cycle.
IOWs, the summing threshold gets closer to zero the closer the global count gets to the hard limit. Hence when there's lots of space available, we have little summing contention, but when we are close to the blocking limit we essentially update the global counter on every modification.
Quoted 6 lines
Quoted 3 lines
Ok, I think I get it now. The statement "
ctx->space_used= 0" is part of clearing the percpu state and is not part of aggregating the CIL per cpu structure into the context.If you mean the '
cilpcp->space_used= 0' statement in the_aggregate()function, then yes.
Yes. My brain is tired.
(Also it's really hot here.)
Quoted 7 lines
We don't actually ever zero the
ctx->space_usedbecause it always is initialised to zero by allocation of a new context and switching to it...Quoted 3 lines
So assuming that I grokked it all on the second try, maybe a comment is in order for the aggregate function?
Quoted 8 lines
Quoted 6 lines
/* * We're in the middle of switching cil contexts. Reset the * counter we use to detect when the current context is nearing * full. */ ctx->space_used = 0;Hmmmm - I'm not sure where you are asking I put this comment...
Sorry, I meant the comment to be placed above the "cilpcp->space_used = 0;" in the _aggregate function.
Quoted 16 lines
Quoted 7 lines
Quoted 6 lines
As such, we get scalability when the CIL is empty by trading off accuracy, but we get accuracy when it is nearing full by trading off scalability. We might need to tweak it for really large CPU counts (maybe use log2(num_online_cpus()), but fundamentally the algorithm is designed to scale according to how close we are to the push thresholds....
Quoted 8 lines
<nod> I suppose if you had a very large number of CPUs and a very small log then the slivers could be zero ... which just means we'd just lose performance due to overly-careful accounting. I wonder if you want to leave a breadcrumb to warn if:
XLOG_CIL_BLOCKING_SPACE_LIMIT()/num_online_cpus()== 0just in case someone ever wanders in with such a configuration?
Quoted 7 lines
I don't think that can happen. The smallest blocking size limit will be a quarter of the smallest log size which puts us around 800KB as the blocking limit. I can't see us supporting 800k CPUs any time soon, and even if you have a couple of hundred thousand CPUs, you're going to have lots of other basic problems trying to do concurrent operations on such a tiny log before we even consider this threshold...
Just think of what IPIs look like! :D
Quoted 5 lines
I'm not too worried about scalability and performance on tiny logs - if they result in an atomic update every transaction, it'll still be sufficient to scale to the concurrency the transaction reservation code will let into the commit path in the first place...
Ok.
--D
Quoted 7 lines
Cheers,
Dave.
Signature · 2 lines
Dave Chinner [email protected]
Re: [PATCH 30/39] xfs: implement percpu cil space used calculation
Quoted 12 lines · Darrick J. Wong
Quoted 11 lines · Dave Chinner
Quoted 9 lines · Darrick J. Wong
So assuming that I grokked it all on the second try, maybe a comment is in order for the aggregate function?
/* * We're in the middle of switching cil contexts. Reset the * counter we use to detect when the current context is nearing * full. */ ctx->space_used = 0;
Quoted 4 lines
Quoted 1 line
Hmmmm - I'm not sure where you are asking I put this comment...
Sorry, I meant the comment to be placed above the "
cilpcp->space_used= 0;" in the_aggregatefunction.
Ok. That'll cause rejects on all the subsequent patches so there's not much point in posting just an update to this patch with that done. I guess I'm sending out another 39 patches before the end of the day.... :/
Cheers,
Dave.
Signature · 2 lines
Dave Chinner [email protected]
PATCH v4 31/39 xfs: track CIL ticket reservation in percpu structure
In reply to Dave ChinnerFrom: Dave Chinner <[email protected]>
To get it out from under the cil spinlock.
--- fs/xfs/xfs_log_cil.c | 20 +++++++++++++++----- fs/xfs/xfs_log_priv.h | 2 +- 2 files changed, 16 insertions(+), 6 deletions(-)
fs/xfs/xfs_log_cil.c +15 -5
@@ -90,6 +90,10 @@ xlog_cil_pcp_aggregate(for_each_online_cpu (cpu) { cilpcp =per_cpu_ptr (cil->xc_pcp, cpu); + + ctx->ticket->t_curr_res += cilpcp->space_reserved; + ctx->ticket->t_unit_res += cilpcp->space_reserved; + cilpcp->space_reserved =0 ; cilpcp->space_used =0 ; } }
@@ -510,6 +514,7 @@ xlog_cil_insert_items(* based on how close we are to the hard limit. */ cilpcp =get_cpu_ptr (cil->xc_pcp); + cilpcp->space_reserved += ctx_res; cilpcp->space_used += len;if (space_used >=XLOG_CIL_SPACE_LIMIT (log) || cilpcp->space_used >
@@ -520,10 +525,6 @@ xlog_cil_insert_items( }put_cpu_ptr (cilpcp); -spin_lock (&cil->xc_cil_lock); - ctx->ticket->t_unit_res += ctx_res; - ctx->ticket->t_curr_res += ctx_res; -/* * If we've overrun the reservation, dump the tx details before we move * the log items. Shutdown is imminent...
@@ -545,6 +546,7 @@ xlog_cil_insert_items(* We do this here so we only need to take the CIL lock once during * the transaction commit. */ +spin_lock (&cil->xc_cil_lock);list_for_each_entry (lip, &tp->t_items, li_trans) {/* Skip items which aren't dirty in this transaction. */
@@ -1434,12 +1436,20 @@ xlog_cil_pcp_dead(spin_lock (&xlog_cil_pcp_lock);list_for_each_entry_safe (cil, n, &xlog_cil_pcp_list, xc_pcp_list) {struct xlog_cil_pcp *cilpcp =per_cpu_ptr (cil->xc_pcp, cpu); +struct xfs_cil_ctx *ctx;spin_unlock (&xlog_cil_pcp_lock);down_write (&cil->xc_ctx_lock); + ctx = cil->xc_ctx; + +atomic_add (cilpcp->space_used, &ctx->space_used); +if (ctx->ticket) { + ctx->ticket->t_curr_res += cilpcp->space_reserved; + ctx->ticket->t_unit_res += cilpcp->space_reserved; + } -atomic_add (cilpcp->space_used, &cil->xc_ctx->space_used); cilpcp->space_used =0 ; + cilpcp->space_reserved =0 ;up_write (&cil->xc_ctx_lock);spin_lock (&xlog_cil_pcp_lock);
fs/xfs/xfs_log_priv.h +1 -1
@@ -232,7 +232,7 @@ struct xfs_cil_ctx {*/ struct xlog_cil_pcp {uint32_t space_used; -uint32_t curr_res; +uint32_t space_reserved;struct list_head busy_extents;struct list_head log_items; };
Signature · 1 line
2.31.1
Re: [PATCH 31/39] xfs: track CIL ticket reservation in percpu structure
Quoted 5 lines · Dave Chinner
From: Dave Chinner <[email protected]>
To get it out from under the cil spinlock.
Signed-off-by Dave Chinner <[email protected]>
Looks ... straightforward enough for a percpu thing ;)
--D
Quoted 85 lines
--- fs/xfs/xfs_log_cil.c | 20 +++++++++++++++----- fs/xfs/xfs_log_priv.h | 2 +- 2 files changed, 16 insertions(+), 6 deletions(-)fs/xfs/xfs_log_cil.c +15
-5@@ -90,6 +90,10 @@ xlog_cil_pcp_aggregate(for_each_online_cpu (cpu) { cilpcp =per_cpu_ptr (cil->xc_pcp, cpu); + + ctx->ticket->t_curr_res += cilpcp->space_reserved; + ctx->ticket->t_unit_res += cilpcp->space_reserved; + cilpcp->space_reserved =0 ; cilpcp->space_used =0 ; } }@@ -510,6 +514,7 @@ xlog_cil_insert_items(* based on how close we are to the hard limit. */ cilpcp =get_cpu_ptr (cil->xc_pcp); + cilpcp->space_reserved += ctx_res; cilpcp->space_used += len;if (space_used >=XLOG_CIL_SPACE_LIMIT (log) || cilpcp->space_used >@@ -520,10 +525,6 @@ xlog_cil_insert_items( }put_cpu_ptr (cilpcp); -spin_lock (&cil->xc_cil_lock); - ctx->ticket->t_unit_res += ctx_res; - ctx->ticket->t_curr_res += ctx_res; -/* * If we've overrun the reservation, dump the tx details before we move * the log items. Shutdown is imminent... @@ -545,6 +546,7 @@ xlog_cil_insert_items(* We do this here so we only need to take the CIL lock once during * the transaction commit. */ +spin_lock (&cil->xc_cil_lock);list_for_each_entry (lip, &tp->t_items, li_trans) {/* Skip items which aren't dirty in this transaction. */ @@ -1434,12 +1436,20 @@ xlog_cil_pcp_dead(spin_lock (&xlog_cil_pcp_lock);list_for_each_entry_safe (cil, n, &xlog_cil_pcp_list, xc_pcp_list) {struct xlog_cil_pcp *cilpcp =per_cpu_ptr (cil->xc_pcp, cpu); +struct xfs_cil_ctx *ctx;spin_unlock (&xlog_cil_pcp_lock);down_write (&cil->xc_ctx_lock); + ctx = cil->xc_ctx; + +atomic_add (cilpcp->space_used, &ctx->space_used); +if (ctx->ticket) { + ctx->ticket->t_curr_res += cilpcp->space_reserved; + ctx->ticket->t_unit_res += cilpcp->space_reserved; + } -atomic_add (cilpcp->space_used, &cil->xc_ctx->space_used); cilpcp->space_used =0 ; + cilpcp->space_reserved =0 ;up_write (&cil->xc_ctx_lock);spin_lock (&xlog_cil_pcp_lock);fs/xfs/xfs_log_priv.h +1
-1@@ -232,7 +232,7 @@ struct xfs_cil_ctx {*/ struct xlog_cil_pcp {uint32_t space_used; -uint32_t curr_res; +uint32_t space_reserved;struct list_head busy_extents;struct list_head log_items; };Signature · 1 line
2.31.1
PATCH v4 32/39 xfs: convert CIL busy extents to per-cpu
In reply to Dave ChinnerFrom: Dave Chinner <[email protected]>
To get them out from under the CIL lock.
This is an unordered list, so we can simply punt it to per-cpu lists during transaction commits and reaggregate it back into a single list during the CIL push work.
--- fs/xfs/xfs_log_cil.c | 22 +++++++++++++++++++--- 1 file changed, 19 insertions(+), 3 deletions(-)
fs/xfs/xfs_log_cil.c +19 -3
@@ -93,6 +93,11 @@ xlog_cil_pcp_aggregate( ctx->ticket->t_curr_res += cilpcp->space_reserved; ctx->ticket->t_unit_res += cilpcp->space_reserved; +if (!list_empty (&cilpcp->busy_extents)) { +list_splice_init (&cilpcp->busy_extents, + &ctx->busy_extents); + } + cilpcp->space_reserved =0 ; cilpcp->space_used =0 ; }
@@ -523,6 +528,9 @@ xlog_cil_insert_items(atomic_add (cilpcp->space_used, &ctx->space_used); cilpcp->space_used =0 ; } +/* attach the transaction to the CIL if it has any busy extents */ +if (!list_empty (&tp->t_busy)) +list_splice_init (&tp->t_busy, &cilpcp->busy_extents);put_cpu_ptr (cilpcp);/*
@@ -562,9 +570,6 @@ xlog_cil_insert_items(list_move_tail (&lip->li_cil, &cil->xc_cil); } -/* attach the transaction to the CIL if it has any busy extents */ -if (!list_empty (&tp->t_busy)) -list_splice_init (&tp->t_busy, &ctx->busy_extents);spin_unlock (&cil->xc_cil_lock);if (tp->t_ticket->t_curr_res <0 )
@@ -1447,6 +1452,10 @@ xlog_cil_pcp_dead( ctx->ticket->t_curr_res += cilpcp->space_reserved; ctx->ticket->t_unit_res += cilpcp->space_reserved; } +if (!list_empty (&cilpcp->busy_extents)) { +list_splice_init (&cilpcp->busy_extents, + &ctx->busy_extents); + } cilpcp->space_used =0 ; cilpcp->space_reserved =0 ;
@@ -1502,7 +1511,9 @@ static void __percpu *xlog_cil_pcp_alloc ( struct xfs_cil *cil) { +struct xlog_cil_pcp *cilpcp;void __percpu *pcp; +int cpu; pcp = alloc_percpu(struct xlog_cil_pcp);if (!pcp)
@@ -1512,6 +1523,11 @@ xlog_cil_pcp_alloc(free_percpu (pcp);return NULL ; } + +for_each_possible_cpu (cpu) { + cilpcp =per_cpu_ptr (pcp, cpu); +INIT_LIST_HEAD (&cilpcp->busy_extents); + }return pcp; }
Signature · 1 line
2.31.1
Re: [PATCH 32/39] xfs: convert CIL busy extents to per-cpu
Quoted 9 lines · Dave Chinner
From: Dave Chinner <[email protected]>
To get them out from under the CIL lock.
This is an unordered list, so we can simply punt it to per-cpu lists during transaction commits and reaggregate it back into a single list during the CIL push work.
Signed-off-by Dave Chinner <[email protected]>
--D
Quoted 76 lines
--- fs/xfs/xfs_log_cil.c | 22 +++++++++++++++++++--- 1 file changed, 19 insertions(+), 3 deletions(-)fs/xfs/xfs_log_cil.c +19
-3@@ -93,6 +93,11 @@ xlog_cil_pcp_aggregate( ctx->ticket->t_curr_res += cilpcp->space_reserved; ctx->ticket->t_unit_res += cilpcp->space_reserved; +if (!list_empty (&cilpcp->busy_extents)) { +list_splice_init (&cilpcp->busy_extents, + &ctx->busy_extents); + } + cilpcp->space_reserved =0 ; cilpcp->space_used =0 ; }@@ -523,6 +528,9 @@ xlog_cil_insert_items(atomic_add (cilpcp->space_used, &ctx->space_used); cilpcp->space_used =0 ; } +/* attach the transaction to the CIL if it has any busy extents */ +if (!list_empty (&tp->t_busy)) +list_splice_init (&tp->t_busy, &cilpcp->busy_extents);put_cpu_ptr (cilpcp);/* @@ -562,9 +570,6 @@ xlog_cil_insert_items(list_move_tail (&lip->li_cil, &cil->xc_cil); } -/* attach the transaction to the CIL if it has any busy extents */ -if (!list_empty (&tp->t_busy)) -list_splice_init (&tp->t_busy, &ctx->busy_extents);spin_unlock (&cil->xc_cil_lock);if (tp->t_ticket->t_curr_res <0 )@@ -1447,6 +1452,10 @@ xlog_cil_pcp_dead( ctx->ticket->t_curr_res += cilpcp->space_reserved; ctx->ticket->t_unit_res += cilpcp->space_reserved; } +if (!list_empty (&cilpcp->busy_extents)) { +list_splice_init (&cilpcp->busy_extents, + &ctx->busy_extents); + } cilpcp->space_used =0 ; cilpcp->space_reserved =0 ;@@ -1502,7 +1511,9 @@ static void __percpu *xlog_cil_pcp_alloc ( struct xfs_cil *cil) { +struct xlog_cil_pcp *cilpcp;void __percpu *pcp; +int cpu; pcp = alloc_percpu(struct xlog_cil_pcp);if (!pcp)@@ -1512,6 +1523,11 @@ xlog_cil_pcp_alloc(free_percpu (pcp);return NULL ; } + +for_each_possible_cpu (cpu) { + cilpcp =per_cpu_ptr (pcp, cpu); +INIT_LIST_HEAD (&cilpcp->busy_extents); + }return pcp; }Signature · 1 line
2.31.1
PATCH v4 33/39 xfs: Add order IDs to log items in CIL
In reply to Dave ChinnerFrom: Dave Chinner <[email protected]>
Before we split the ordered CIL up into per cpu lists, we need a mechanism to track the order of the items in the CIL. We need to do this because there are rules around the order in which related items must physically appear in the log even inside a single checkpoint transaction.
An example of this is intents - an intent must appear in the log before it's intent done record so taht log recovery can cancel the intent correctly. If we have these two records misordered in the CIL, then they will not be recovered correctly by journal replay.
We also will not be able to move items to the tail of the CIL list when they are relogged, hence the log items will need some mechanism to allow the correct log item order to be recreated before we write log items to the hournal.
Hence we need to have a mechanism for recording global order of transactions in the log items so that we can recover that order from un-ordered per-cpu lists.
Do this with a simple monotonic increasing commit counter in the CIL context. Each log item in the transaction gets stamped with the current commit order ID before it is added to the CIL. If the item is already in the CIL, leave it where it is instead of moving it to the tail of the list and instead sort the list before we start the push work.
XXX: list_sort() under the cil_ctx_lock held exclusive starts hurting that >16 threads. Front end commits are waiting on the push to switch contexts much longer. The item order id should likely be moved into the logvecs when they are detacted from the items, then the sort can be done on the logvec after the cil_ctx_lock has been released. logvecs will need to use a list_head for this rather than a single linked list like they do now....
--- fs/xfs/xfs_log_cil.c | 38 ++++++++++++++++++++++++++++++-------- fs/xfs/xfs_log_priv.h | 1 + fs/xfs/xfs_trans.h | 1 + 3 files changed, 32 insertions(+), 8 deletions(-)
fs/xfs/xfs_log_cil.c +30 -8
@@ -461,6 +461,7 @@ xlog_cil_insert_items(int len =0 ;int iovhdr_res =0 , split_res =0 , ctx_res =0 ;int space_used; +int order;struct xlog_cil_pcp *cilpcp;ASSERT (tp);
@@ -550,10 +551,12 @@ xlog_cil_insert_items( }/* -* Now (re-)position everything modified at the tail of the CIL. +* Now update the order of everything modified in the transaction +* and insert items into the CIL if they aren't already there. * We do this here so we only need to take the CIL lock once during * the transaction commit. */ + order =atomic_inc_return (&ctx->order_id);spin_lock (&cil->xc_cil_lock);list_for_each_entry (lip, &tp->t_items, li_trans) {
@@ -561,13 +564,10 @@ xlog_cil_insert_items(if (!test_bit (XFS_LI_DIRTY , &lip->li_flags))continue ; -/* -* Only move the item if it isn't already at the tail. This is -* to prevent a transient list_empty() state when reinserting -* an item that is already the only item in the CIL. -*/ -if (!list_is_last (&lip->li_cil, &cil->xc_cil)) -list_move_tail (&lip->li_cil, &cil->xc_cil); + lip->li_order_id = order; +if (!list_empty (&lip->li_cil)) +continue ; +list_add_tail (&lip->li_cil, &cil->xc_cil); }spin_unlock (&cil->xc_cil_lock);
@@ -780,6 +780,26 @@ xlog_cil_build_trans_hdr( tic->t_curr_res -= lvhdr->lv_bytes; } +/* +* CIL item reordering compare function. We want to order in ascending ID order, +* but we want to leave items with the same ID in the order they were added to +* the list. This is important for operations like reflink where we log 4 order +* dependent intents in a single transaction when we overwrite an existing +* shared extent with a new shared extent. i.e. BUI(unmap), CUI(drop), +* CUI (inc), BUI(remap)... +*/ +static int +xlog_cil_order_cmp ( +void *priv, +const struct list_head *a, +const struct list_head *b) +{ + struct xfs_log_item *l1 = container_of(a, struct xfs_log_item, li_cil); + struct xfs_log_item *l2 = container_of(b, struct xfs_log_item, li_cil); + +return l1->li_order_id > l2->li_order_id; +} +/* * Push the Committed Item List to the log. *
@@ -900,6 +920,7 @@ xlog_cil_push_work(* needed on the transaction commit side which is currently locked out * by the flush lock. */ +list_sort (NULL , &cil->xc_cil, xlog_cil_order_cmp); lv =NULL ;while (!list_empty (&cil->xc_cil)) {struct xfs_log_item *item;
@@ -907,6 +928,7 @@ xlog_cil_push_work( item =list_first_entry (&cil->xc_cil, struct xfs_log_item, li_cil);list_del_init (&item->li_cil); + item->li_order_id =0 ;if (!ctx->lv_chain) ctx->lv_chain = item->li_lv;else
fs/xfs/xfs_log_priv.h +1 -0
@@ -225,6 +225,7 @@ struct xfs_cil_ctx {struct list_head committing;/* ctx committing list */ struct work_struct discard_endio_work;struct work_struct push_work; +atomic_t order_id; };/*
fs/xfs/xfs_trans.h +1 -0
@@ -44,6 +44,7 @@ struct xfs_log_item {struct xfs_log_vec *li_lv;/* active log vector */ struct xfs_log_vec *li_lv_shadow;/* standby vector */ xfs_csn_t li_seq;/* CIL commit seq */ +uint32_t li_order_id;/* CIL commit order */ };/*
Signature · 1 line
2.31.1
Re: [PATCH 33/39] xfs: Add order IDs to log items in CIL
Quoted 10 lines · Dave Chinner
From: Dave Chinner <[email protected]>
Before we split the ordered CIL up into per cpu lists, we need a mechanism to track the order of the items in the CIL. We need to do this because there are rules around the order in which related items must physically appear in the log even inside a single checkpoint transaction.
An example of this is intents - an intent must appear in the log before it's intent done record so taht log recovery can cancel the
s/taht/that/
Quoted 19 lines
intent correctly. If we have these two records misordered in the CIL, then they will not be recovered correctly by journal replay.
We also will not be able to move items to the tail of the CIL list when they are relogged, hence the log items will need some mechanism to allow the correct log item order to be recreated before we write log items to the hournal.
Hence we need to have a mechanism for recording global order of transactions in the log items so that we can recover that order from un-ordered per-cpu lists.
Do this with a simple monotonic increasing commit counter in the CIL context. Each log item in the transaction gets stamped with the current commit order ID before it is added to the CIL. If the item is already in the CIL, leave it where it is instead of moving it to the tail of the list and instead sort the list before we start the push work.
Quoted 7 lines
XXX:
list_sort()under thecil_ctx_lockheld exclusive starts hurting that >16 threads. Front end commits are waiting on the push to switch contexts much longer. The item order id should likely be moved into the logvecs when they are detacted from the items, then the sort can be done on the logvec after thecil_ctx_lockhas been released. logvecs will need to use alist_headfor this rather than a single linked list like they do now....
...which I guess happens in patch 35 now?
Quoted 52 lines
Signed-off-by Dave Chinner <[email protected]>--- fs/xfs/xfs_log_cil.c | 38 ++++++++++++++++++++++++++++++-------- fs/xfs/xfs_log_priv.h | 1 + fs/xfs/xfs_trans.h | 1 + 3 files changed, 32 insertions(+), 8 deletions(-)fs/xfs/xfs_log_cil.c +8
-8@@ -461,6 +461,7 @@ xlog_cil_insert_items(int len =0 ;int iovhdr_res =0 , split_res =0 , ctx_res =0 ;int space_used; +int order;struct xlog_cil_pcp *cilpcp;ASSERT (tp);@@ -550,10 +551,12 @@ xlog_cil_insert_items( }/* -* Now (re-)position everything modified at the tail of the CIL. +* Now update the order of everything modified in the transaction +* and insert items into the CIL if they aren't already there. * We do this here so we only need to take the CIL lock once during * the transaction commit. */ + order =atomic_inc_return (&ctx->order_id);spin_lock (&cil->xc_cil_lock);list_for_each_entry (lip, &tp->t_items, li_trans) {@@ -561,13 +564,10 @@ xlog_cil_insert_items(if (!test_bit (XFS_LI_DIRTY , &lip->li_flags))continue ; -/* -* Only move the item if it isn't already at the tail. This is -* to prevent a transient list_empty() state when reinserting -* an item that is already the only item in the CIL. -*/ -if (!list_is_last (&lip->li_cil, &cil->xc_cil)) -list_move_tail (&lip->li_cil, &cil->xc_cil); + lip->li_order_id = order; +if (!list_empty (&lip->li_cil)) +continue ; +list_add_tail (&lip->li_cil, &cil->xc_cil); }spin_unlock (&cil->xc_cil_lock);
fs/xfs/xfs_log_cil.c xlog_cil_build_trans_hdr( · line 785
Quoted 6 lines
fs/xfs/xfs_log_cil.c +3
-0@@ -780,6 +780,26 @@ xlog_cil_build_trans_hdr( tic->t_curr_res -= lvhdr->lv_bytes; } +/* +* CIL item reordering compare function. We want to order in ascending ID order, +* but we want to leave items with the same ID in the order they were added to
When do we have items with the same id?
I guess that happens if we have multiple transactions adding items to the cil at the same time? I guess that's not a big deal since each of those threads will hold a disjoint set of locks, so even if the order ids are the same for a bunch of items, they're never going to be touching the same AG/inode/metadata object, right?
If that's correct, then:
--D
Quoted 63 lines
fs/xfs/xfs_log_cil.c +19
-0+* the list. This is important for operations like reflink where we log 4 order +* dependent intents in a single transaction when we overwrite an existing +* shared extent with a new shared extent. i.e. BUI(unmap), CUI(drop), +* CUI (inc), BUI(remap)... +*/ +static int +xlog_cil_order_cmp ( +void *priv, +const struct list_head *a, +const struct list_head *b) +{ + struct xfs_log_item *l1 = container_of(a, struct xfs_log_item, li_cil); + struct xfs_log_item *l2 = container_of(b, struct xfs_log_item, li_cil); + +return l1->li_order_id > l2->li_order_id; +} +/* * Push the Committed Item List to the log. * @@ -900,6 +920,7 @@ xlog_cil_push_work(* needed on the transaction commit side which is currently locked out * by the flush lock. */ +list_sort (NULL , &cil->xc_cil, xlog_cil_order_cmp); lv =NULL ;while (!list_empty (&cil->xc_cil)) {struct xfs_log_item *item;@@ -907,6 +928,7 @@ xlog_cil_push_work( item =list_first_entry (&cil->xc_cil, struct xfs_log_item, li_cil);list_del_init (&item->li_cil); + item->li_order_id =0 ;if (!ctx->lv_chain) ctx->lv_chain = item->li_lv;else fs/xfs/xfs_log_priv.h +1
-0@@ -225,6 +225,7 @@ struct xfs_cil_ctx {struct list_head committing;/* ctx committing list */ struct work_struct discard_endio_work;struct work_struct push_work; +atomic_t order_id; };/* fs/xfs/xfs_trans.h +1
-0@@ -44,6 +44,7 @@ struct xfs_log_item {struct xfs_log_vec *li_lv;/* active log vector */ struct xfs_log_vec *li_lv_shadow;/* standby vector */ xfs_csn_t li_seq;/* CIL commit seq */ +uint32_t li_order_id;/* CIL commit order */ };/* Signature · 1 line
2.31.1
Re: [PATCH 33/39] xfs: Add order IDs to log items in CIL
Quoted 41 lines · Darrick J. Wong
Quoted 10 lines · Dave Chinner
From: Dave Chinner <[email protected]>
Before we split the ordered CIL up into per cpu lists, we need a mechanism to track the order of the items in the CIL. We need to do this because there are rules around the order in which related items must physically appear in the log even inside a single checkpoint transaction.
An example of this is intents - an intent must appear in the log before it's intent done record so taht log recovery can cancel the
s/taht/that/
Quoted 19 lines
intent correctly. If we have these two records misordered in the CIL, then they will not be recovered correctly by journal replay.
We also will not be able to move items to the tail of the CIL list when they are relogged, hence the log items will need some mechanism to allow the correct log item order to be recreated before we write log items to the hournal.
Hence we need to have a mechanism for recording global order of transactions in the log items so that we can recover that order from un-ordered per-cpu lists.
Do this with a simple monotonic increasing commit counter in the CIL context. Each log item in the transaction gets stamped with the current commit order ID before it is added to the CIL. If the item is already in the CIL, leave it where it is instead of moving it to the tail of the list and instead sort the list before we start the push work.
Quoted 7 lines
XXX:
list_sort()under thecil_ctx_lockheld exclusive starts hurting that >16 threads. Front end commits are waiting on the push to switch contexts much longer. The item order id should likely be moved into the logvecs when they are detacted from the items, then the sort can be done on the logvec after thecil_ctx_lockhas been released. logvecs will need to use alist_headfor this rather than a single linked list like they do now....
Quoted 1 line
...which I guess happens in patch 35 now?
Right. I'll just remove this from the commit message.
Quoted 9 lines
fs/xfs/xfs_log_cil.c xlog_cil_build_trans_hdr( · line 785
Quoted 7 lines
fs/xfs/xfs_log_cil.c +3
-0@@ -780,6 +780,26 @@ xlog_cil_build_trans_hdr( tic->t_curr_res -= lvhdr->lv_bytes; } +/* +* CIL item reordering compare function. We want to order in ascending ID order, +* but we want to leave items with the same ID in the order they were added to When do we have items with the same id?
All the items in a single transaction have the same id. The order id increments before we tag all the items in the transaction and insert them into the CIL.
Quoted 8 lines
I guess that happens if we have multiple transactions adding items to the cil at the same time? I guess that's not a big deal since each of those threads will hold a disjoint set of locks, so even if the order ids are the same for a bunch of items, they're never going to be touching the same AG/inode/metadata object, right?
If that's correct, then:
Reviewed-by Darrick J. Wong <[email protected]>
While true, it's not the way this works so I won't immediately accept your RVB. The reason for not changing the ordering within a single transaction is actually intent logging. i.e. this:
Quoted 4 lines
fs/xfs/xfs_log_cil.c xlog_cil_build_trans_hdr( · line 789
Quoted 4 lines
fs/xfs/xfs_log_cil.c +4
-0+* the list. This is important for operations like reflink where we log 4 order +* dependent intents in a single transaction when we overwrite an existing +* shared extent with a new shared extent. i.e. BUI(unmap), CUI(drop), +* CUI (inc), BUI(remap)...
There's a specific order of operations that recovery must run these intents in, and so if we re-order them here in the CIL they'll be out of order in the log and recovery will replay the intents in the wrong order. Replaying the intents in the wrong order results in corruption warnings and assert failures during log recovery, hence the constraint of not re-ordering items within the same transaction.
Cheers,
Dave.
Signature · 2 lines
Dave Chinner [email protected]
Re: [PATCH 33/39] xfs: Add order IDs to log items in CIL
Quoted 79 lines · Dave Chinner
Quoted 41 lines · Darrick J. Wong
Quoted 10 lines · Dave Chinner
From: Dave Chinner <[email protected]>
Before we split the ordered CIL up into per cpu lists, we need a mechanism to track the order of the items in the CIL. We need to do this because there are rules around the order in which related items must physically appear in the log even inside a single checkpoint transaction.
An example of this is intents - an intent must appear in the log before it's intent done record so taht log recovery can cancel the
s/taht/that/
Quoted 19 lines
intent correctly. If we have these two records misordered in the CIL, then they will not be recovered correctly by journal replay.
We also will not be able to move items to the tail of the CIL list when they are relogged, hence the log items will need some mechanism to allow the correct log item order to be recreated before we write log items to the hournal.
Hence we need to have a mechanism for recording global order of transactions in the log items so that we can recover that order from un-ordered per-cpu lists.
Do this with a simple monotonic increasing commit counter in the CIL context. Each log item in the transaction gets stamped with the current commit order ID before it is added to the CIL. If the item is already in the CIL, leave it where it is instead of moving it to the tail of the list and instead sort the list before we start the push work.
Quoted 7 lines
XXX:
list_sort()under thecil_ctx_lockheld exclusive starts hurting that >16 threads. Front end commits are waiting on the push to switch contexts much longer. The item order id should likely be moved into the logvecs when they are detacted from the items, then the sort can be done on the logvec after thecil_ctx_lockhas been released. logvecs will need to use alist_headfor this rather than a single linked list like they do now....Quoted 1 line
...which I guess happens in patch 35 now?
Right. I'll just remove this from the commit message.
Quoted 9 lines
fs/xfs/xfs_log_cil.c xlog_cil_build_trans_hdr( · line 785
Quoted 7 lines
fs/xfs/xfs_log_cil.c +3
-0@@ -780,6 +780,26 @@ xlog_cil_build_trans_hdr( tic->t_curr_res -= lvhdr->lv_bytes; } +/* +* CIL item reordering compare function. We want to order in ascending ID order, +* but we want to leave items with the same ID in the order they were added to When do we have items with the same id?
All the items in a single transaction have the same id. The order id increments before we tag all the items in the transaction and insert them into the CIL.
Quoted 8 lines
I guess that happens if we have multiple transactions adding items to the cil at the same time? I guess that's not a big deal since each of those threads will hold a disjoint set of locks, so even if the order ids are the same for a bunch of items, they're never going to be touching the same AG/inode/metadata object, right?
If that's correct, then:
Reviewed-by Darrick J. Wong <[email protected]>While true, it's not the way this works so I won't immediately accept your RVB. The reason for not changing the ordering within a single transaction is actually intent logging. i.e. this:
Quoted 4 lines
fs/xfs/xfs_log_cil.c xlog_cil_build_trans_hdr( · line 789
Quoted 4 lines
fs/xfs/xfs_log_cil.c +4
-0+* the list. This is important for operations like reflink where we log 4 order +* dependent intents in a single transaction when we overwrite an existing +* shared extent with a new shared extent. i.e. BUI(unmap), CUI(drop), +* CUI (inc), BUI(remap)...
Quoted 6 lines
There's a specific order of operations that recovery must run these intents in, and so if we re-order them here in the CIL they'll be out of order in the log and recovery will replay the intents in the wrong order. Replaying the intents in the wrong order results in corruption warnings and assert failures during log recovery, hence the constraint of not re-ordering items within the same transaction.
<ding> lightbulb comes on. I think I understood this better the last time I read all these patches. :/
Basically, for each item that can be attached to a transaction, you're assigning it an "order id" that is a monotonically increasing counter that (roughly) records the last time the item was committed. Certain items (like inodes) can be relogged and committed multiple times in rapid fire succession, in which case the order_id will get bumped forward.
In the /next/ patch you'll change the cil item list to be per-cpu and only splice the mess together at cil push time. For that to work properly, you have to re-sort that resulting list in commit order (aka the order_id) to keep the items in order of commit.
For items *within* a transaction, you take advantage of the property of list_sort that it won't reorder items with cmp(a, b) == 0, which means that all the intents logged to a transaction will maintain the same order that the author of higher level code wrote into the software.
Question: xlog_cil_push_work zeroes the order_id of pushed log items. Is there any potential problem here when ctx->order_id wraps around to zero? I think the answer is that we'll move on to a new cil context long before we hit 2^32-1 transactions?
--D
Quoted 7 lines
Cheers,
Dave.
Signature · 2 lines
Dave Chinner [email protected]
Re: [PATCH 33/39] xfs: Add order IDs to log items in CIL
Quoted 90 lines · Darrick J. Wong
Quoted 79 lines · Dave Chinner
Quoted 41 lines · Darrick J. Wong
Quoted 10 lines · Dave Chinner
From: Dave Chinner <[email protected]>
Before we split the ordered CIL up into per cpu lists, we need a mechanism to track the order of the items in the CIL. We need to do this because there are rules around the order in which related items must physically appear in the log even inside a single checkpoint transaction.
An example of this is intents - an intent must appear in the log before it's intent done record so taht log recovery can cancel the
s/taht/that/
Quoted 19 lines
intent correctly. If we have these two records misordered in the CIL, then they will not be recovered correctly by journal replay.
We also will not be able to move items to the tail of the CIL list when they are relogged, hence the log items will need some mechanism to allow the correct log item order to be recreated before we write log items to the hournal.
Hence we need to have a mechanism for recording global order of transactions in the log items so that we can recover that order from un-ordered per-cpu lists.
Do this with a simple monotonic increasing commit counter in the CIL context. Each log item in the transaction gets stamped with the current commit order ID before it is added to the CIL. If the item is already in the CIL, leave it where it is instead of moving it to the tail of the list and instead sort the list before we start the push work.
Quoted 7 lines
XXX:
list_sort()under thecil_ctx_lockheld exclusive starts hurting that >16 threads. Front end commits are waiting on the push to switch contexts much longer. The item order id should likely be moved into the logvecs when they are detacted from the items, then the sort can be done on the logvec after thecil_ctx_lockhas been released. logvecs will need to use alist_headfor this rather than a single linked list like they do now....Quoted 1 line
...which I guess happens in patch 35 now?
Right. I'll just remove this from the commit message.
Quoted 9 lines
fs/xfs/xfs_log_cil.c xlog_cil_build_trans_hdr( · line 785
Quoted 7 lines
fs/xfs/xfs_log_cil.c +3
-0@@ -780,6 +780,26 @@ xlog_cil_build_trans_hdr( tic->t_curr_res -= lvhdr->lv_bytes; } +/* +* CIL item reordering compare function. We want to order in ascending ID order, +* but we want to leave items with the same ID in the order they were added to When do we have items with the same id?
All the items in a single transaction have the same id. The order id increments before we tag all the items in the transaction and insert them into the CIL.
Quoted 8 lines
I guess that happens if we have multiple transactions adding items to the cil at the same time? I guess that's not a big deal since each of those threads will hold a disjoint set of locks, so even if the order ids are the same for a bunch of items, they're never going to be touching the same AG/inode/metadata object, right?
If that's correct, then:
Reviewed-by Darrick J. Wong <[email protected]>While true, it's not the way this works so I won't immediately accept your RVB. The reason for not changing the ordering within a single transaction is actually intent logging. i.e. this:
Quoted 4 lines
fs/xfs/xfs_log_cil.c xlog_cil_build_trans_hdr( · line 789
Quoted 4 lines
fs/xfs/xfs_log_cil.c +4
-0+* the list. This is important for operations like reflink where we log 4 order +* dependent intents in a single transaction when we overwrite an existing +* shared extent with a new shared extent. i.e. BUI(unmap), CUI(drop), +* CUI (inc), BUI(remap)... Quoted 6 lines
There's a specific order of operations that recovery must run these intents in, and so if we re-order them here in the CIL they'll be out of order in the log and recovery will replay the intents in the wrong order. Replaying the intents in the wrong order results in corruption warnings and assert failures during log recovery, hence the constraint of not re-ordering items within the same transaction.
<ding> lightbulb comes on. I think I understood this better the last time I read all these patches. :/
Quoted 6 lines
Basically, for each item that can be attached to a transaction, you're assigning it an "order id" that is a monotonically increasing counter that (roughly) records the last time the item was committed. Certain items (like inodes) can be relogged and committed multiple times in rapid fire succession, in which case the
order_idwill get bumped forward.
Effectively, yes.
Quoted 9 lines
In the /next/ patch you'll change the cil item list to be per-cpu and only splice the mess together at cil push time. For that to work properly, you have to re-sort that resulting list in commit order (aka the
order_id) to keep the items in order of commit.For items *within* a transaction, you take advantage of the property of
list_sortthat it won't reorder items withcmp(a, b)== 0, which means that all the intents logged to a transaction will maintain the same order that the author of higher level code wrote into the software.
Correct.
Quoted 4 lines
Question:
xlog_cil_push_workzeroes theorder_idof pushed log items. Is there any potential problem here whenctx->order_idwraps around to zero? I think the answer is that we'll move on to a new cil context long before we hit 2^32-1 transactions?
Yes. At the moment, the max transaction rate is about 800k/s, which means it'd take a couple of hours to run 4 billion transactions. So we're in no danger of overruning the number of transactions in a CIL commit any time soon. And if we ever get near that, we can just bump the counter to a 64 bit value...
Cheers,
Dave.
Signature · 2 lines
Dave Chinner [email protected]
Darrick J. Wong
Reviewed-byRe: [PATCH 33/39] xfs: Add order IDs to log items in CIL
Quoted 105 lines · Dave Chinner
Quoted 90 lines · Darrick J. Wong
Quoted 79 lines · Dave Chinner
Quoted 41 lines · Darrick J. Wong
Quoted 10 lines · Dave Chinner
From: Dave Chinner <[email protected]>
Before we split the ordered CIL up into per cpu lists, we need a mechanism to track the order of the items in the CIL. We need to do this because there are rules around the order in which related items must physically appear in the log even inside a single checkpoint transaction.
An example of this is intents - an intent must appear in the log before it's intent done record so taht log recovery can cancel the
s/taht/that/
Quoted 19 lines
intent correctly. If we have these two records misordered in the CIL, then they will not be recovered correctly by journal replay.
We also will not be able to move items to the tail of the CIL list when they are relogged, hence the log items will need some mechanism to allow the correct log item order to be recreated before we write log items to the hournal.
Hence we need to have a mechanism for recording global order of transactions in the log items so that we can recover that order from un-ordered per-cpu lists.
Do this with a simple monotonic increasing commit counter in the CIL context. Each log item in the transaction gets stamped with the current commit order ID before it is added to the CIL. If the item is already in the CIL, leave it where it is instead of moving it to the tail of the list and instead sort the list before we start the push work.
Quoted 7 lines
XXX:
list_sort()under thecil_ctx_lockheld exclusive starts hurting that >16 threads. Front end commits are waiting on the push to switch contexts much longer. The item order id should likely be moved into the logvecs when they are detacted from the items, then the sort can be done on the logvec after thecil_ctx_lockhas been released. logvecs will need to use alist_headfor this rather than a single linked list like they do now....Quoted 1 line
...which I guess happens in patch 35 now?
Right. I'll just remove this from the commit message.
Quoted 9 lines
fs/xfs/xfs_log_cil.c xlog_cil_build_trans_hdr( · line 785
Quoted 7 lines
fs/xfs/xfs_log_cil.c +3
-0@@ -780,6 +780,26 @@ xlog_cil_build_trans_hdr( tic->t_curr_res -= lvhdr->lv_bytes; } +/* +* CIL item reordering compare function. We want to order in ascending ID order, +* but we want to leave items with the same ID in the order they were added to When do we have items with the same id?
All the items in a single transaction have the same id. The order id increments before we tag all the items in the transaction and insert them into the CIL.
Quoted 8 lines
I guess that happens if we have multiple transactions adding items to the cil at the same time? I guess that's not a big deal since each of those threads will hold a disjoint set of locks, so even if the order ids are the same for a bunch of items, they're never going to be touching the same AG/inode/metadata object, right?
If that's correct, then:
Reviewed-by Darrick J. Wong <[email protected]>While true, it's not the way this works so I won't immediately accept your RVB. The reason for not changing the ordering within a single transaction is actually intent logging. i.e. this:
Quoted 4 lines
fs/xfs/xfs_log_cil.c xlog_cil_build_trans_hdr( · line 789
Quoted 4 lines
fs/xfs/xfs_log_cil.c +4
-0+* the list. This is important for operations like reflink where we log 4 order +* dependent intents in a single transaction when we overwrite an existing +* shared extent with a new shared extent. i.e. BUI(unmap), CUI(drop), +* CUI (inc), BUI(remap)... Quoted 6 lines
There's a specific order of operations that recovery must run these intents in, and so if we re-order them here in the CIL they'll be out of order in the log and recovery will replay the intents in the wrong order. Replaying the intents in the wrong order results in corruption warnings and assert failures during log recovery, hence the constraint of not re-ordering items within the same transaction.
<ding> lightbulb comes on. I think I understood this better the last time I read all these patches. :/
Quoted 6 lines
Basically, for each item that can be attached to a transaction, you're assigning it an "order id" that is a monotonically increasing counter that (roughly) records the last time the item was committed. Certain items (like inodes) can be relogged and committed multiple times in rapid fire succession, in which case the
order_idwill get bumped forward.Effectively, yes.
Quoted 5 lines
In the /next/ patch you'll change the cil item list to be per-cpu and only splice the mess together at cil push time. For that to work properly, you have to re-sort that resulting list in commit order (aka the
order_id) to keep the items in order of commit.
Quoted 6 lines
Quoted 4 lines
For items *within* a transaction, you take advantage of the property of
list_sortthat it won't reorder items withcmp(a, b)== 0, which means that all the intents logged to a transaction will maintain the same order that the author of higher level code wrote into the software.Correct.
Ok, good.
Quoted 10 lines
Quoted 4 lines
Question:
xlog_cil_push_workzeroes theorder_idof pushed log items. Is there any potential problem here whenctx->order_idwraps around to zero? I think the answer is that we'll move on to a new cil context long before we hit 2^32-1 transactions?Yes. At the moment, the max transaction rate is about 800k/s, which means it'd take a couple of hours to run 4 billion transactions. So we're in no danger of overruning the number of transactions in a CIL commit any time soon. And if we ever get near that, we can just bump the counter to a 64 bit value...
Ok.
With the "taht" in the commit message fixed,
--D
Quoted 7 lines
Cheers,
Dave.
Signature · 2 lines
Dave Chinner [email protected]
PATCH v4 34/39 xfs: convert CIL to unordered per cpu lists
In reply to Dave ChinnerFrom: Dave Chinner <[email protected]>
So that we can remove the cil_lock which is a global serialisation point. We've already got ordering sorted, so all we need to do is treat the CIL list like the busy extent list and reconstruct it before the push starts.
This is what we're trying to avoid:
- 75.35% 1.83% [kernel] [k] xfs_log_commit_cil
- 46.35% xfs_log_commit_cil
- 41.54% _raw_spin_lock
- 67.30% do_raw_spin_lock
66.96% __pv_queued_spin_lock_slowpath
Which happens on a 32p system when running a 32-way 'rm -rf' workload. After this patch:
- 20.90% 3.23% [kernel] [k] xfs_log_commit_cil
- 17.67% xfs_log_commit_cil
- 6.51% xfs_log_ticket_ungrant
1.40% xfs_log_space_wake
2.32% memcpy_erms
- 2.18% xfs_buf_item_committing
- 2.12% xfs_buf_item_release
- 1.03% xfs_buf_unlock
0.96% up
0.72% xfs_buf_rele
1.33% xfs_inode_item_format
1.19% down_read
0.91% up_read
0.76% xfs_buf_item_format
- 0.68% kmem_alloc_large
- 0.67% kmem_alloc
0.64% __kmalloc
0.50% xfs_buf_item_sizeIt kinda looks like the workload is running out of log space all the time. But all the spinlock contention is gone and the transaction commit rate has gone from 800k/s to 1.3M/s so the amount of real work being done has gone up a *lot*.
--- fs/xfs/xfs_log_cil.c | 69 +++++++++++++++++++------------------------ fs/xfs/xfs_log_priv.h | 3 +- 2 files changed, 31 insertions(+), 41 deletions(-)
fs/xfs/xfs_log_cil.c +30 -39
@@ -72,6 +72,7 @@ xlog_cil_ctx_alloc(void) ctx =kmem_zalloc (sizeof (*ctx),KM_NOFS );INIT_LIST_HEAD (&ctx->committing);INIT_LIST_HEAD (&ctx->busy_extents); +INIT_LIST_HEAD (&ctx->log_items);INIT_WORK (&ctx->push_work, xlog_cil_push_work);return ctx; }
@@ -97,6 +98,8 @@ xlog_cil_pcp_aggregate(list_splice_init (&cilpcp->busy_extents, &ctx->busy_extents); } +if (!list_empty (&cilpcp->log_items)) +list_splice_init (&cilpcp->log_items, &ctx->log_items); cilpcp->space_reserved =0 ; cilpcp->space_used =0 ;
@@ -475,10 +478,9 @@ xlog_cil_insert_items(/* * We need to take the CIL checkpoint unit reservation on the first * commit into the CIL. Test the XLOG_CIL_EMPTY bit first so we don't -* unnecessarily do an atomic op in the fast path here. We don't need to -* hold the xc_cil_lock here to clear the XLOG_CIL_EMPTY bit as we are -* under the xc_ctx_lock here and that needs to be held exclusively to -* reset the XLOG_CIL_EMPTY bit. +* unnecessarily do an atomic op in the fast path here. We can clear the +* XLOG_CIL_EMPTY bit as we are under the xc_ctx_lock here and that +* needs to be held exclusively to reset the XLOG_CIL_EMPTY bit. */ if (test_bit (XLOG_CIL_EMPTY , &cil->xc_flags) &&test_and_clear_bit (XLOG_CIL_EMPTY , &cil->xc_flags))
@@ -532,24 +534,6 @@ xlog_cil_insert_items(/* attach the transaction to the CIL if it has any busy extents */ if (!list_empty (&tp->t_busy))list_splice_init (&tp->t_busy, &cilpcp->busy_extents); -put_cpu_ptr (cilpcp); - -/* -* If we've overrun the reservation, dump the tx details before we move -* the log items. Shutdown is imminent... -*/ - tp->t_ticket->t_curr_res -= ctx_res + len; -if (WARN_ON (tp->t_ticket->t_curr_res <0 )) { -xfs_warn (log->l_mp,"Transaction log reservation overrun:" ); -xfs_warn (log->l_mp, -" log items: %d bytes (iov hdrs: %d bytes)" , - len, iovhdr_res); -xfs_warn (log->l_mp," split region headers: %d bytes" , - split_res); -xfs_warn (log->l_mp," ctx ticket: %d bytes" , ctx_res); -xlog_print_trans (tp); - } -/* * Now update the order of everything modified in the transaction * and insert items into the CIL if they aren't already there.
@@ -557,7 +541,6 @@ xlog_cil_insert_items(* the transaction commit. */ order =atomic_inc_return (&ctx->order_id); -spin_lock (&cil->xc_cil_lock);list_for_each_entry (lip, &tp->t_items, li_trans) {/* Skip items which aren't dirty in this transaction. */
@@ -567,10 +550,25 @@ xlog_cil_insert_items( lip->li_order_id = order;if (!list_empty (&lip->li_cil))continue ; -list_add_tail (&lip->li_cil, &cil->xc_cil); +list_add_tail (&lip->li_cil, &cilpcp->log_items); } +put_cpu_ptr (cilpcp); -spin_unlock (&cil->xc_cil_lock); +/* +* If we've overrun the reservation, dump the tx details before we move +* the log items. Shutdown is imminent... +*/ + tp->t_ticket->t_curr_res -= ctx_res + len; +if (WARN_ON (tp->t_ticket->t_curr_res <0 )) { +xfs_warn (log->l_mp,"Transaction log reservation overrun:" ); +xfs_warn (log->l_mp, +" log items: %d bytes (iov hdrs: %d bytes)" , + len, iovhdr_res); +xfs_warn (log->l_mp," split region headers: %d bytes" , + split_res); +xfs_warn (log->l_mp," ctx ticket: %d bytes" , ctx_res); +xlog_print_trans (tp); + }if (tp->t_ticket->t_curr_res <0 )xfs_force_shutdown (log->l_mp,SHUTDOWN_LOG_IO_ERROR );
@@ -914,18 +912,12 @@ xlog_cil_push_work(xlog_cil_pcp_aggregate (cil, ctx); -/* -* Pull all the log vectors off the items in the CIL, and remove the -* items from the CIL. We don't need the CIL lock here because it's only -* needed on the transaction commit side which is currently locked out -* by the flush lock. -*/ -list_sort (NULL , &cil->xc_cil, xlog_cil_order_cmp); - lv =NULL ; -while (!list_empty (&cil->xc_cil)) { +list_sort (NULL , &ctx->log_items, xlog_cil_order_cmp); + +while (!list_empty (&ctx->log_items)) {struct xfs_log_item *item; - item =list_first_entry (&cil->xc_cil, + item =list_first_entry (&ctx->log_items, struct xfs_log_item, li_cil);list_del_init (&item->li_cil); item->li_order_id =0 ;
@@ -1119,7 +1111,6 @@ xlog_cil_push_background(* The cil won't be empty because we are called while holding the * context lock so whatever we added to the CIL will still be there. */ -ASSERT (!list_empty (&cil->xc_cil));ASSERT (!test_bit (XLOG_CIL_EMPTY , &cil->xc_flags));/*
@@ -1478,6 +1469,8 @@ xlog_cil_pcp_dead(list_splice_init (&cilpcp->busy_extents, &ctx->busy_extents); } +if (!list_empty (&cilpcp->log_items)) +list_splice_init (&cilpcp->log_items, &ctx->log_items); cilpcp->space_used =0 ; cilpcp->space_reserved =0 ;
@@ -1549,6 +1542,7 @@ xlog_cil_pcp_alloc(for_each_possible_cpu (cpu) { cilpcp =per_cpu_ptr (pcp, cpu);INIT_LIST_HEAD (&cilpcp->busy_extents); +INIT_LIST_HEAD (&cilpcp->log_items); }return pcp; }
@@ -1584,9 +1578,7 @@ xlog_cil_init(return -ENOMEM ; } -INIT_LIST_HEAD (&cil->xc_cil);INIT_LIST_HEAD (&cil->xc_committing); -spin_lock_init (&cil->xc_cil_lock);spin_lock_init (&cil->xc_push_lock);init_waitqueue_head (&cil->xc_push_wait);init_rwsem (&cil->xc_ctx_lock);
@@ -1612,7 +1604,6 @@ xlog_cil_destroy(kmem_free (cil->xc_ctx); } -ASSERT (list_empty (&cil->xc_cil));ASSERT (test_bit (XLOG_CIL_EMPTY , &cil->xc_flags));xlog_cil_pcp_free (cil, cil->xc_pcp);kmem_free (cil);
fs/xfs/xfs_log_priv.h +1 -2
@@ -220,6 +220,7 @@ struct xfs_cil_ctx {struct xlog_ticket *ticket;/* chkpt ticket */ atomic_t space_used;/* aggregate size of regions */ struct list_head busy_extents;/* busy extents in chkpt */ +struct list_head log_items;/* log items in chkpt */ struct xfs_log_vec *lv_chain;/* logvecs being pushed */ struct list_head iclog_entry;struct list_head committing;/* ctx committing list */
@@ -258,8 +259,6 @@ struct xfs_cil {struct xlog *xc_log;unsigned long xc_flags;atomic_t xc_iclog_hdrs; -struct list_head xc_cil; -spinlock_t xc_cil_lock;struct rw_semaphore xc_ctx_lock____cacheline_aligned_in_smp ;struct xfs_cil_ctx *xc_ctx;
Signature · 1 line
2.31.1
Re: [PATCH 34/39] xfs: convert CIL to unordered per cpu lists
Quoted 53 lines · Dave Chinner
From: Dave Chinner <[email protected]>
So that we can remove the
cil_lockwhich is a global serialisation point. We've already got ordering sorted, so all we need to do is treat the CIL list like the busy extent list and reconstruct it before the push starts.This is what we're trying to avoid:
- 75.35% 1.83% [kernel] [k]
xfs_log_commit_cil
- 46.35%xfs_log_commit_cil
- 41.54%_raw_spin_lock
- 67.30%do_raw_spin_lock
66.96%__pv_queued_spin_lock_slowpathWhich happens on a 32p system when running a 32-way 'rm -rf' workload. After this patch:
- 20.90% 3.23% [kernel] [k]
xfs_log_commit_cil
- 17.67%xfs_log_commit_cil- 6.51% xfs_log_ticket_ungrant 1.40% xfs_log_space_wake 2.32% memcpy_erms - 2.18% xfs_buf_item_committing - 2.12% xfs_buf_item_release - 1.03% xfs_buf_unlock 0.96% up 0.72% xfs_buf_rele 1.33% xfs_inode_item_format 1.19% down_read 0.91% up_read 0.76% xfs_buf_item_format - 0.68% kmem_alloc_large - 0.67% kmem_alloc 0.64% __kmalloc 0.50% xfs_buf_item_sizeIt kinda looks like the workload is running out of log space all the time. But all the spinlock contention is gone and the transaction commit rate has gone from 800k/s to 1.3M/s so the amount of real work being done has gone up a *lot*.
Signed-off-by Dave Chinner <[email protected]>--- fs/xfs/xfs_log_cil.c | 69 +++++++++++++++++++------------------------ fs/xfs/xfs_log_priv.h | 3 +- 2 files changed, 31 insertions(+), 41 deletions(-)fs/xfs/xfs_log_cil.c +0
-0
fs/xfs/xfs_log_cil.c xlog_cil_ctx_alloc(void) · line 75
Quoted 4 lines
fs/xfs/xfs_log_cil.c +1
-0@@ -72,6 +72,7 @@ xlog_cil_ctx_alloc(void) ctx =kmem_zalloc (sizeof (*ctx),KM_NOFS );INIT_LIST_HEAD (&ctx->committing);INIT_LIST_HEAD (&ctx->busy_extents); +INIT_LIST_HEAD (&ctx->log_items);
I see you moved the log item list to the cil ctx for benefit of _pcp_dead, correct?
If so, then this isn't especially different from the last version.
Yay for shortening lock critical sections,
--D
Quoted 177 lines
fs/xfs/xfs_log_cil.c +29
-39INIT_WORK (&ctx->push_work, xlog_cil_push_work);return ctx; }@@ -97,6 +98,8 @@ xlog_cil_pcp_aggregate(list_splice_init (&cilpcp->busy_extents, &ctx->busy_extents); } +if (!list_empty (&cilpcp->log_items)) +list_splice_init (&cilpcp->log_items, &ctx->log_items); cilpcp->space_reserved =0 ; cilpcp->space_used =0 ;@@ -475,10 +478,9 @@ xlog_cil_insert_items(/* * We need to take the CIL checkpoint unit reservation on the first * commit into the CIL. Test the XLOG_CIL_EMPTY bit first so we don't -* unnecessarily do an atomic op in the fast path here. We don't need to -* hold the xc_cil_lock here to clear the XLOG_CIL_EMPTY bit as we are -* under the xc_ctx_lock here and that needs to be held exclusively to -* reset the XLOG_CIL_EMPTY bit. +* unnecessarily do an atomic op in the fast path here. We can clear the +* XLOG_CIL_EMPTY bit as we are under the xc_ctx_lock here and that +* needs to be held exclusively to reset the XLOG_CIL_EMPTY bit. */ if (test_bit (XLOG_CIL_EMPTY , &cil->xc_flags) &&test_and_clear_bit (XLOG_CIL_EMPTY , &cil->xc_flags))@@ -532,24 +534,6 @@ xlog_cil_insert_items(/* attach the transaction to the CIL if it has any busy extents */ if (!list_empty (&tp->t_busy))list_splice_init (&tp->t_busy, &cilpcp->busy_extents); -put_cpu_ptr (cilpcp); - -/* -* If we've overrun the reservation, dump the tx details before we move -* the log items. Shutdown is imminent... -*/ - tp->t_ticket->t_curr_res -= ctx_res + len; -if (WARN_ON (tp->t_ticket->t_curr_res <0 )) { -xfs_warn (log->l_mp,"Transaction log reservation overrun:" ); -xfs_warn (log->l_mp, -" log items: %d bytes (iov hdrs: %d bytes)" , - len, iovhdr_res); -xfs_warn (log->l_mp," split region headers: %d bytes" , - split_res); -xfs_warn (log->l_mp," ctx ticket: %d bytes" , ctx_res); -xlog_print_trans (tp); - } -/* * Now update the order of everything modified in the transaction * and insert items into the CIL if they aren't already there. @@ -557,7 +541,6 @@ xlog_cil_insert_items(* the transaction commit. */ order =atomic_inc_return (&ctx->order_id); -spin_lock (&cil->xc_cil_lock);list_for_each_entry (lip, &tp->t_items, li_trans) {/* Skip items which aren't dirty in this transaction. */ @@ -567,10 +550,25 @@ xlog_cil_insert_items( lip->li_order_id = order;if (!list_empty (&lip->li_cil))continue ; -list_add_tail (&lip->li_cil, &cil->xc_cil); +list_add_tail (&lip->li_cil, &cilpcp->log_items); } +put_cpu_ptr (cilpcp); -spin_unlock (&cil->xc_cil_lock); +/* +* If we've overrun the reservation, dump the tx details before we move +* the log items. Shutdown is imminent... +*/ + tp->t_ticket->t_curr_res -= ctx_res + len; +if (WARN_ON (tp->t_ticket->t_curr_res <0 )) { +xfs_warn (log->l_mp,"Transaction log reservation overrun:" ); +xfs_warn (log->l_mp, +" log items: %d bytes (iov hdrs: %d bytes)" , + len, iovhdr_res); +xfs_warn (log->l_mp," split region headers: %d bytes" , + split_res); +xfs_warn (log->l_mp," ctx ticket: %d bytes" , ctx_res); +xlog_print_trans (tp); + }if (tp->t_ticket->t_curr_res <0 )xfs_force_shutdown (log->l_mp,SHUTDOWN_LOG_IO_ERROR );@@ -914,18 +912,12 @@ xlog_cil_push_work(xlog_cil_pcp_aggregate (cil, ctx); -/* -* Pull all the log vectors off the items in the CIL, and remove the -* items from the CIL. We don't need the CIL lock here because it's only -* needed on the transaction commit side which is currently locked out -* by the flush lock. -*/ -list_sort (NULL , &cil->xc_cil, xlog_cil_order_cmp); - lv =NULL ; -while (!list_empty (&cil->xc_cil)) { +list_sort (NULL , &ctx->log_items, xlog_cil_order_cmp); + +while (!list_empty (&ctx->log_items)) {struct xfs_log_item *item; - item =list_first_entry (&cil->xc_cil, + item =list_first_entry (&ctx->log_items, struct xfs_log_item, li_cil);list_del_init (&item->li_cil); item->li_order_id =0 ;@@ -1119,7 +1111,6 @@ xlog_cil_push_background(* The cil won't be empty because we are called while holding the * context lock so whatever we added to the CIL will still be there. */ -ASSERT (!list_empty (&cil->xc_cil));ASSERT (!test_bit (XLOG_CIL_EMPTY , &cil->xc_flags));/* @@ -1478,6 +1469,8 @@ xlog_cil_pcp_dead(list_splice_init (&cilpcp->busy_extents, &ctx->busy_extents); } +if (!list_empty (&cilpcp->log_items)) +list_splice_init (&cilpcp->log_items, &ctx->log_items); cilpcp->space_used =0 ; cilpcp->space_reserved =0 ;@@ -1549,6 +1542,7 @@ xlog_cil_pcp_alloc(for_each_possible_cpu (cpu) { cilpcp =per_cpu_ptr (pcp, cpu);INIT_LIST_HEAD (&cilpcp->busy_extents); +INIT_LIST_HEAD (&cilpcp->log_items); }return pcp; }@@ -1584,9 +1578,7 @@ xlog_cil_init(return -ENOMEM ; } -INIT_LIST_HEAD (&cil->xc_cil);INIT_LIST_HEAD (&cil->xc_committing); -spin_lock_init (&cil->xc_cil_lock);spin_lock_init (&cil->xc_push_lock);init_waitqueue_head (&cil->xc_push_wait);init_rwsem (&cil->xc_ctx_lock);@@ -1612,7 +1604,6 @@ xlog_cil_destroy(kmem_free (cil->xc_ctx); } -ASSERT (list_empty (&cil->xc_cil));ASSERT (test_bit (XLOG_CIL_EMPTY , &cil->xc_flags));xlog_cil_pcp_free (cil, cil->xc_pcp);kmem_free (cil);fs/xfs/xfs_log_priv.h +1
-2@@ -220,6 +220,7 @@ struct xfs_cil_ctx {struct xlog_ticket *ticket;/* chkpt ticket */ atomic_t space_used;/* aggregate size of regions */ struct list_head busy_extents;/* busy extents in chkpt */ +struct list_head log_items;/* log items in chkpt */ struct xfs_log_vec *lv_chain;/* logvecs being pushed */ struct list_head iclog_entry;struct list_head committing;/* ctx committing list */ @@ -258,8 +259,6 @@ struct xfs_cil {struct xlog *xc_log;unsigned long xc_flags;atomic_t xc_iclog_hdrs; -struct list_head xc_cil; -spinlock_t xc_cil_lock;struct rw_semaphore xc_ctx_lock____cacheline_aligned_in_smp ;struct xfs_cil_ctx *xc_ctx;Signature · 1 line
2.31.1
Re: [PATCH 34/39] xfs: convert CIL to unordered per cpu lists
Quoted 49 lines · Darrick J. Wong
Quoted 48 lines · Dave Chinner
From: Dave Chinner <[email protected]>
So that we can remove the
cil_lockwhich is a global serialisation point. We've already got ordering sorted, so all we need to do is treat the CIL list like the busy extent list and reconstruct it before the push starts.This is what we're trying to avoid:
- 75.35% 1.83% [kernel] [k]
xfs_log_commit_cil
- 46.35%xfs_log_commit_cil
- 41.54%_raw_spin_lock
- 67.30%do_raw_spin_lock
66.96%__pv_queued_spin_lock_slowpathWhich happens on a 32p system when running a 32-way 'rm -rf' workload. After this patch:
- 20.90% 3.23% [kernel] [k]
xfs_log_commit_cil
- 17.67%xfs_log_commit_cil- 6.51% xfs_log_ticket_ungrant 1.40% xfs_log_space_wake 2.32% memcpy_erms - 2.18% xfs_buf_item_committing - 2.12% xfs_buf_item_release - 1.03% xfs_buf_unlock 0.96% up 0.72% xfs_buf_rele 1.33% xfs_inode_item_format 1.19% down_read 0.91% up_read 0.76% xfs_buf_item_format - 0.68% kmem_alloc_large - 0.67% kmem_alloc 0.64% __kmalloc 0.50% xfs_buf_item_sizeIt kinda looks like the workload is running out of log space all the time. But all the spinlock contention is gone and the transaction commit rate has gone from 800k/s to 1.3M/s so the amount of real work being done has gone up a *lot*.
Signed-off-by Dave Chinner <[email protected]>--- fs/xfs/xfs_log_cil.c | 69 +++++++++++++++++++------------------------ fs/xfs/xfs_log_priv.h | 3 +- 2 files changed, 31 insertions(+), 41 deletions(-)
Quoted 12 lines
fs/xfs/xfs_log_cil.c xlog_cil_ctx_alloc(void) · line 75
Quoted 9 lines
fs/xfs/xfs_log_cil.c +1
-0@@ -72,6 +72,7 @@ xlog_cil_ctx_alloc(void) ctx =kmem_zalloc (sizeof (*ctx),KM_NOFS );INIT_LIST_HEAD (&ctx->committing);INIT_LIST_HEAD (&ctx->busy_extents); +INIT_LIST_HEAD (&ctx->log_items);I see you moved the log item list to the cil ctx for benefit of
_pcp_dead, correct?
Largely, yes. It also helps to have the item push list rooted in the structure that holds all of the push specific state (i.e. the CIL ctx) once we detatch that from the CIL itself.
Quoted 1 line
If so, then this isn't especially different from the last version.
*nod*
Quoted 2 lines
Yay for shortening lock critical sections,
Reviewed-by Darrick J. Wong <[email protected]>
Ta.
-Dave.
Signature · 2 lines
Dave Chinner [email protected]
PATCH v4 35/39 xfs: convert log vector chain to use list heads
In reply to Dave ChinnerFrom: Dave Chinner <[email protected]>
Because the next change is going to require sorting log vectors, and that requires arbitrary rearrangement of the list which cannot be done easily with a single linked list.
--- fs/xfs/xfs_log.c | 35 +++++++++++++++++++++++++--------- fs/xfs/xfs_log.h | 2 +- fs/xfs/xfs_log_cil.c | 42 +++++++++++++++++++++++------------------ fs/xfs/xfs_log_priv.h | 4 ++-- fs/xfs/xfs_trans.c | 4 ++-- fs/xfs/xfs_trans_priv.h | 3 ++- 6 files changed, 57 insertions(+), 33 deletions(-)
fs/xfs/xfs_log.c +26 -9
@@ -848,6 +848,9 @@ xlog_write_unmount_record( .lv_niovecs = 1, .lv_iovecp = ®, }; +LIST_HEAD (lv_chain); +INIT_LIST_HEAD (&vec.lv_list); +list_add (&vec.lv_list, &lv_chain); BUILD_BUG_ON((sizeof(struct xlog_op_header) + sizeof(struct xfs_unmount_log_format)) !=
@@ -863,7 +866,7 @@ xlog_write_unmount_record(*/ if (log->l_targ != log->l_mp->m_ddev_targp)blkdev_issue_flush (log->l_targ->bt_bdev); -return xlog_write (log, &vec, ticket,NULL ,NULL , reg.i_len); +return xlog_write (log, &lv_chain, ticket,NULL ,NULL , reg.i_len); }/*
@@ -1581,13 +1584,16 @@ xlog_commit_record( .lv_iovecp = ®, };int error; +LIST_HEAD (lv_chain); +INIT_LIST_HEAD (&vec.lv_list); +list_add (&vec.lv_list, &lv_chain);if (XLOG_FORCED_SHUTDOWN (log))return -EIO ;/* account for space used by record data */ ticket->t_curr_res -= reg.i_len; - error =xlog_write (log, &vec, ticket, lsn, iclog, reg.i_len); + error =xlog_write (log, &lv_chain, ticket, lsn, iclog, reg.i_len);if (error)xfs_force_shutdown (log->l_mp,SHUTDOWN_LOG_IO_ERROR );return error;
@@ -2118,6 +2124,7 @@ xlog_print_trans(*/ static struct xfs_log_vec * xlog_write_single( + struct list_head *lv_chain, struct xfs_log_vec *log_vector, struct xlog_ticket *ticket, struct xlog_in_core *iclog,
@@ -2134,7 +2141,9 @@ xlog_write_single( iclog->ic_state == XLOG_STATE_WANT_SYNC); ptr = iclog->ic_datap + *log_offset; -for (lv = log_vector; lv; lv = lv->lv_next) { +for (lv = log_vector; + !list_entry_is_head (lv, lv_chain, lv_list); + lv =list_next_entry (lv, lv_list)) {/* * If the entire log vec does not fit in the iclog, punt it to * the partial copy loop which can handle this case.
@@ -2163,6 +2172,8 @@ xlog_write_single( *data_cnt += reg->i_len; } } +if (list_entry_is_head (lv, lv_chain, lv_list)) + lv =NULL ;ASSERT (*len ==0 || lv);return lv; }
@@ -2208,6 +2219,7 @@ xlog_write_get_more_iclog_space(static struct xfs_log_vec * xlog_write_partial( struct xlog *log, + struct list_head *lv_chain, struct xfs_log_vec *log_vector, struct xlog_ticket *ticket, struct xlog_in_core **iclogp,
@@ -2347,7 +2359,10 @@ xlog_write_partial(* the caller so it can go back to fast path copying. */ *iclogp = iclog; -return lv->lv_next; + lv =list_next_entry (lv, lv_list); +if (list_entry_is_head (lv, lv_chain, lv_list)) +return NULL ; +return lv; }/*
@@ -2393,14 +2408,14 @@ xlog_write_partial(int xlog_write (struct xlog *log, -struct xfs_log_vec *log_vector, +struct list_head *lv_chain,struct xlog_ticket *ticket,xfs_lsn_t *start_lsn,struct xlog_in_core **commit_iclog,uint32_t len) {struct xlog_in_core *iclog =NULL ; -struct xfs_log_vec *lv = log_vector; +struct xfs_log_vec *lv;int record_cnt =0 ;int data_cnt =0 ;int error =0 ;
@@ -2422,14 +2437,16 @@ xlog_write(if (start_lsn) *start_lsn =be64_to_cpu (iclog->ic_header.h_lsn); + lv = list_first_entry_or_null(lv_chain, struct xfs_log_vec, lv_list);while (lv) { - lv =xlog_write_single (lv, ticket, iclog, &log_offset, + lv =xlog_write_single (lv_chain, lv, ticket, iclog, &log_offset, &len, &record_cnt, &data_cnt);if (!lv)break ; - lv =xlog_write_partial (log, lv, ticket, &iclog, &log_offset, - &len, &record_cnt, &data_cnt); + lv =xlog_write_partial (log, lv_chain, lv, ticket, &iclog, + &log_offset, &len, &record_cnt, + &data_cnt);if (IS_ERR_OR_NULL (lv)) { error =PTR_ERR_OR_ZERO (lv);break ;
fs/xfs/xfs_log.h +1 -1
@@ -9,7 +9,7 @@struct xfs_cil_ctx ;struct xfs_log_vec { -struct xfs_log_vec *lv_next;/* next lv in build list */ +struct list_head lv_list;/* CIL lv chain ptrs */ int lv_niovecs;/* number of iovecs in lv */ struct xfs_log_iovec *lv_iovecp;/* iovec array */ struct xfs_log_item *lv_item;/* owner */
fs/xfs/xfs_log_cil.c +24 -18
128 lines not on this page · raw message
fs/xfs/xfs_log_priv.h +2 -2
@@ -221,7 +221,7 @@ struct xfs_cil_ctx {atomic_t space_used;/* aggregate size of regions */ struct list_head busy_extents;/* busy extents in chkpt */ struct list_head log_items;/* log items in chkpt */ -struct xfs_log_vec *lv_chain;/* logvecs being pushed */ +struct list_head lv_chain;/* logvecs being pushed */ struct list_head iclog_entry;struct list_head committing;/* ctx committing list */ struct work_struct discard_endio_work;
@@ -477,7 +477,7 @@ xlog_write_adv_cnt(void **ptr, int *len, int *off, size_t bytes)void xlog_print_tic_res (struct xfs_mount *mp,struct xlog_ticket *ticket);void xlog_print_trans (struct xfs_trans *); -int xlog_write (struct xlog *log,struct xfs_log_vec *log_vector, +int xlog_write (struct xlog *log,struct list_head *lv_chain,struct xlog_ticket *tic,xfs_lsn_t *start_lsn,struct xlog_in_core **commit_iclog,uint32_t len); int xlog_commit_record(struct xlog *log, struct xlog_ticket *ticket,
fs/xfs/xfs_trans.c +2 -2
@@ -735,7 +735,7 @@ xfs_log_item_batch_insert(void xfs_trans_committed_bulk (struct xfs_ail *ailp, -struct xfs_log_vec *log_vector, +struct list_head *lv_chain,xfs_lsn_t commit_lsn,bool aborted) {
@@ -750,7 +750,7 @@ xfs_trans_committed_bulk(spin_unlock (&ailp->ail_lock);/* unpin all the log items */ -for (lv = log_vector; lv; lv = lv->lv_next ) { +list_for_each_entry (lv, lv_chain, lv_list) {struct xfs_log_item *lip = lv->lv_item;xfs_lsn_t item_lsn;
fs/xfs/xfs_trans_priv.h +2 -1
@@ -18,7 +18,8 @@ void xfs_trans_add_item(struct xfs_trans *, struct xfs_log_item *);void xfs_trans_del_item (struct xfs_log_item *);void xfs_trans_unreserve_and_mod_sb (struct xfs_trans *tp); -void xfs_trans_committed_bulk (struct xfs_ail *ailp,struct xfs_log_vec *lv, +void xfs_trans_committed_bulk (struct xfs_ail *ailp, +struct list_head *lv_chain,xfs_lsn_t commit_lsn,bool aborted);/* * AIL traversal cursor.
Signature · 1 line
2.31.1
Re: [PATCH 35/39] xfs: convert log vector chain to use list heads
Quoted 193 lines · Dave Chinner
From: Dave Chinner <[email protected]>
Because the next change is going to require sorting log vectors, and that requires arbitrary rearrangement of the list which cannot be done easily with a single linked list.
Signed-off-by Dave Chinner <[email protected]>--- fs/xfs/xfs_log.c | 35 +++++++++++++++++++++++++--------- fs/xfs/xfs_log.h | 2 +- fs/xfs/xfs_log_cil.c | 42 +++++++++++++++++++++++------------------ fs/xfs/xfs_log_priv.h | 4 ++-- fs/xfs/xfs_trans.c | 4 ++-- fs/xfs/xfs_trans_priv.h | 3 ++- 6 files changed, 57 insertions(+), 33 deletions(-)fs/xfs/xfs_log.c +26
-9@@ -848,6 +848,9 @@ xlog_write_unmount_record( .lv_niovecs = 1, .lv_iovecp = ®, }; +LIST_HEAD (lv_chain); +INIT_LIST_HEAD (&vec.lv_list); +list_add (&vec.lv_list, &lv_chain); BUILD_BUG_ON((sizeof(struct xlog_op_header) + sizeof(struct xfs_unmount_log_format)) !=@@ -863,7 +866,7 @@ xlog_write_unmount_record(*/ if (log->l_targ != log->l_mp->m_ddev_targp)blkdev_issue_flush (log->l_targ->bt_bdev); -return xlog_write (log, &vec, ticket,NULL ,NULL , reg.i_len); +return xlog_write (log, &lv_chain, ticket,NULL ,NULL , reg.i_len); }/* @@ -1581,13 +1584,16 @@ xlog_commit_record( .lv_iovecp = ®, };int error; +LIST_HEAD (lv_chain); +INIT_LIST_HEAD (&vec.lv_list); +list_add (&vec.lv_list, &lv_chain);if (XLOG_FORCED_SHUTDOWN (log))return -EIO ;/* account for space used by record data */ ticket->t_curr_res -= reg.i_len; - error =xlog_write (log, &vec, ticket, lsn, iclog, reg.i_len); + error =xlog_write (log, &lv_chain, ticket, lsn, iclog, reg.i_len);if (error)xfs_force_shutdown (log->l_mp,SHUTDOWN_LOG_IO_ERROR );return error;@@ -2118,6 +2124,7 @@ xlog_print_trans(*/ static struct xfs_log_vec * xlog_write_single( + struct list_head *lv_chain, struct xfs_log_vec *log_vector, struct xlog_ticket *ticket, struct xlog_in_core *iclog,@@ -2134,7 +2141,9 @@ xlog_write_single( iclog->ic_state == XLOG_STATE_WANT_SYNC); ptr = iclog->ic_datap + *log_offset; -for (lv = log_vector; lv; lv = lv->lv_next) { +for (lv = log_vector; + !list_entry_is_head (lv, lv_chain, lv_list); + lv =list_next_entry (lv, lv_list)) {/* * If the entire log vec does not fit in the iclog, punt it to * the partial copy loop which can handle this case. @@ -2163,6 +2172,8 @@ xlog_write_single( *data_cnt += reg->i_len; } } +if (list_entry_is_head (lv, lv_chain, lv_list)) + lv =NULL ;ASSERT (*len ==0 || lv);return lv; }@@ -2208,6 +2219,7 @@ xlog_write_get_more_iclog_space(static struct xfs_log_vec * xlog_write_partial( struct xlog *log, + struct list_head *lv_chain, struct xfs_log_vec *log_vector, struct xlog_ticket *ticket, struct xlog_in_core **iclogp,@@ -2347,7 +2359,10 @@ xlog_write_partial(* the caller so it can go back to fast path copying. */ *iclogp = iclog; -return lv->lv_next; + lv =list_next_entry (lv, lv_list); +if (list_entry_is_head (lv, lv_chain, lv_list)) +return NULL ; +return lv; }/* @@ -2393,14 +2408,14 @@ xlog_write_partial(int xlog_write (struct xlog *log, -struct xfs_log_vec *log_vector, +struct list_head *lv_chain,struct xlog_ticket *ticket,xfs_lsn_t *start_lsn,struct xlog_in_core **commit_iclog,uint32_t len) {struct xlog_in_core *iclog =NULL ; -struct xfs_log_vec *lv = log_vector; +struct xfs_log_vec *lv;int record_cnt =0 ;int data_cnt =0 ;int error =0 ;@@ -2422,14 +2437,16 @@ xlog_write(if (start_lsn) *start_lsn =be64_to_cpu (iclog->ic_header.h_lsn); + lv = list_first_entry_or_null(lv_chain, struct xfs_log_vec, lv_list);while (lv) { - lv =xlog_write_single (lv, ticket, iclog, &log_offset, + lv =xlog_write_single (lv_chain, lv, ticket, iclog, &log_offset, &len, &record_cnt, &data_cnt);if (!lv)break ; - lv =xlog_write_partial (log, lv, ticket, &iclog, &log_offset, - &len, &record_cnt, &data_cnt); + lv =xlog_write_partial (log, lv_chain, lv, ticket, &iclog, + &log_offset, &len, &record_cnt, + &data_cnt);if (IS_ERR_OR_NULL (lv)) { error =PTR_ERR_OR_ZERO (lv);break ;fs/xfs/xfs_log.h +1
-1@@ -9,7 +9,7 @@struct xfs_cil_ctx ;struct xfs_log_vec { -struct xfs_log_vec *lv_next;/* next lv in build list */ +struct list_head lv_list;/* CIL lv chain ptrs */ int lv_niovecs;/* number of iovecs in lv */ struct xfs_log_iovec *lv_iovecp;/* iovec array */ struct xfs_log_item *lv_item;/* owner */ fs/xfs/xfs_log_cil.c +2
-2@@ -73,6 +73,7 @@ xlog_cil_ctx_alloc(void)INIT_LIST_HEAD (&ctx->committing);INIT_LIST_HEAD (&ctx->busy_extents);INIT_LIST_HEAD (&ctx->log_items); +INIT_LIST_HEAD (&ctx->lv_chain);INIT_WORK (&ctx->push_work, xlog_cil_push_work);return ctx; }@@ -267,6 +268,7 @@ xlog_cil_alloc_shadow_bufs( lv =kmem_alloc_large (buf_size,KM_NOFS );memset (lv,0 ,xlog_cil_iovec_space (niovecs)); +INIT_LIST_HEAD (&lv->lv_list); lv->lv_item = lip; lv->lv_size = buf_size;if (ordered)@@ -282,7 +284,6 @@ xlog_cil_alloc_shadow_bufs(else lv->lv_buf_len =0 ; lv->lv_bytes =0 ; - lv->lv_next =NULL ; }/* Ensure the lv is set up according to ->iop_size */ @@ -409,7 +410,6 @@ xlog_cil_insert_format_items(if (lip->li_lv && shadow->lv_size <= lip->li_lv->lv_size) {/* same or smaller, optimise common overwrite case */ lv = lip->li_lv; - lv->lv_next =NULL ;if (ordered) goto insert;@@ -576,14 +576,14 @@ xlog_cil_insert_items(
fs/xfs/xfs_log_cil.c xlog_cil_insert_items( · line 583
Quoted 10 lines
fs/xfs/xfs_log_cil.c +2
-3static void xlog_cil_free_logvec ( -struct xfs_log_vec *log_vector) +struct list_head *lv_chain) {struct xfs_log_vec *lv; -for (lv = log_vector; lv; ) { -struct xfs_log_vec *next = lv->lv_next; +while (!list_empty (lv_chain)) {
Nit: space after 'while'.
Quoted 45 lines
fs/xfs/xfs_log_cil.c +5
-5+ lv = list_first_entry(lv_chain, struct xfs_log_vec, lv_list); +list_del_init (&lv->lv_list);kmem_free (lv); - lv = next; } }@@ -682,7 +682,7 @@ xlog_cil_committed(spin_unlock (&ctx->cil->xc_push_lock); } -xfs_trans_committed_bulk (ctx->cil->xc_log->l_ailp, ctx->lv_chain, +xfs_trans_committed_bulk (ctx->cil->xc_log->l_ailp, &ctx->lv_chain, ctx->start_lsn, abort);xfs_extent_busy_sort (&ctx->busy_extents);@@ -693,7 +693,7 @@ xlog_cil_committed(list_del (&ctx->committing);spin_unlock (&ctx->cil->xc_push_lock); -xlog_cil_free_logvec (ctx->lv_chain); +xlog_cil_free_logvec (&ctx->lv_chain);if (!list_empty (&ctx->busy_extents))xlog_discard_busy_extents (mp, ctx);@@ -773,7 +773,6 @@ xlog_cil_build_trans_hdr( lvhdr->lv_niovecs =2 ; lvhdr->lv_iovecp = &hdr->lhdr[0 ]; lvhdr->lv_bytes = hdr->lhdr[0 ].i_len + hdr->lhdr[1 ].i_len; - lvhdr->lv_next = ctx->lv_chain; tic->t_curr_res -= lvhdr->lv_bytes; }@@ -913,25 +912,23 @@ xlog_cil_push_work(xlog_cil_pcp_aggregate (cil, ctx);list_sort (NULL , &ctx->log_items, xlog_cil_order_cmp); -while (!list_empty (&ctx->log_items)) {struct xfs_log_item *item; item =list_first_entry (&ctx->log_items, struct xfs_log_item, li_cil); + lv = item->li_lv;list_del_init (&item->li_cil);
fs/xfs/xfs_log_cil.c xlog_cil_push_work( · line 925
Quoted 10 lines
fs/xfs/xfs_log_cil.c +1
-6item->li_order_id =0 ; -if (!ctx->lv_chain) - ctx->lv_chain = item->li_lv; -else - lv->lv_next = item->li_lv; - lv = item->li_lv; item->li_lv =NULL ; - num_iovecs += lv->lv_niovecs; + num_iovecs += lv->lv_niovecs;
Not sure why "lv = item->li_lv" needed to move up?
I think the only change needed here is replacing the lv_chain/lv_next business with the list_add_tail?
Quoted 27 lines
fs/xfs/xfs_log_cil.c +6
-0/* we don't write ordered log vectors */ if (lv->lv_buf_len !=XFS_LOG_VEC_ORDERED ) num_bytes += lv->lv_bytes; + +list_add_tail (&lv->lv_list, &ctx->lv_chain); + }/* @@ -968,10 +965,13 @@ xlog_cil_push_work(* Build a checkpoint transaction header and write it to the log to * begin the transaction. We need to account for the space used by the * transaction header here as it is not accounted for in xlog_write(). +* Add the lvhdr to the head of the lv chain we pass to xlog_write() so +* it gets written into the iclog first. */ xlog_cil_build_trans_hdr (ctx, &thdr, &lvhdr, num_iovecs); num_iovecs += lvhdr.lv_niovecs; num_bytes += lvhdr.lv_bytes; +list_add (&lvhdr.lv_list, &ctx->lv_chain);/* * Before we format and submit the first iclog, we have to ensure that @@ -985,8 +985,14 @@ xlog_cil_push_work(* use the commit record lsn then we can move the tail beyond the grant * write head. */
fs/xfs/xfs_log_cil.c xlog_cil_push_work( · line 995
Quoted 10 lines
fs/xfs/xfs_log_cil.c +8
-2- error =xlog_write (log, &lvhdr, ctx->ticket, &ctx->start_lsn,NULL , - num_bytes); + error =xlog_write (log, &ctx->lv_chain, ctx->ticket, &ctx->start_lsn, +NULL , num_bytes); + +/* +* Take the lvhdr back off the lv_chain as it should not be passed +* to log IO completion. +*/ +list_del (&lvhdr.lv_list);
Seems a little clunky, but I guess I see why it's needed.
I /think/ I don't see any place where the onstack lvhdr can escape out of the chain after _push_work returns, so this is safe enough.
--D
Quoted 64 lines
fs/xfs/xfs_log_cil.c +0
-0if (error) goto out_abort_free_ticket;fs/xfs/xfs_log_priv.h +2
-2@@ -221,7 +221,7 @@ struct xfs_cil_ctx {atomic_t space_used;/* aggregate size of regions */ struct list_head busy_extents;/* busy extents in chkpt */ struct list_head log_items;/* log items in chkpt */ -struct xfs_log_vec *lv_chain;/* logvecs being pushed */ +struct list_head lv_chain;/* logvecs being pushed */ struct list_head iclog_entry;struct list_head committing;/* ctx committing list */ struct work_struct discard_endio_work;@@ -477,7 +477,7 @@ xlog_write_adv_cnt(void **ptr, int *len, int *off, size_t bytes)void xlog_print_tic_res (struct xfs_mount *mp,struct xlog_ticket *ticket);void xlog_print_trans (struct xfs_trans *); -int xlog_write (struct xlog *log,struct xfs_log_vec *log_vector, +int xlog_write (struct xlog *log,struct list_head *lv_chain,struct xlog_ticket *tic,xfs_lsn_t *start_lsn,struct xlog_in_core **commit_iclog,uint32_t len); int xlog_commit_record(struct xlog *log, struct xlog_ticket *ticket,fs/xfs/xfs_trans.c +2
-2@@ -735,7 +735,7 @@ xfs_log_item_batch_insert(void xfs_trans_committed_bulk (struct xfs_ail *ailp, -struct xfs_log_vec *log_vector, +struct list_head *lv_chain,xfs_lsn_t commit_lsn,bool aborted) {@@ -750,7 +750,7 @@ xfs_trans_committed_bulk(spin_unlock (&ailp->ail_lock);/* unpin all the log items */ -for (lv = log_vector; lv; lv = lv->lv_next ) { +list_for_each_entry (lv, lv_chain, lv_list) {struct xfs_log_item *lip = lv->lv_item;xfs_lsn_t item_lsn;fs/xfs/xfs_trans_priv.h +2
-1@@ -18,7 +18,8 @@ void xfs_trans_add_item(struct xfs_trans *, struct xfs_log_item *);void xfs_trans_del_item (struct xfs_log_item *);void xfs_trans_unreserve_and_mod_sb (struct xfs_trans *tp); -void xfs_trans_committed_bulk (struct xfs_ail *ailp,struct xfs_log_vec *lv, +void xfs_trans_committed_bulk (struct xfs_ail *ailp, +struct list_head *lv_chain,xfs_lsn_t commit_lsn,bool aborted);/* * AIL traversal cursor. Signature · 1 line
2.31.1
Re: [PATCH 35/39] xfs: convert log vector chain to use list heads
Quoted 1 line · Darrick J. Wong
On Wed, May 19, 2021 at 10:13:13PM +1000, Dave Chinner wrote:
.....
Quoted 21 lines
Quoted 21 lines
fs/xfs/xfs_log_cil.c +1
-7@@ -913,25 +912,23 @@ xlog_cil_push_work(xlog_cil_pcp_aggregate (cil, ctx);list_sort (NULL , &ctx->log_items, xlog_cil_order_cmp); -while (!list_empty (&ctx->log_items)) {struct xfs_log_item *item; item =list_first_entry (&ctx->log_items, struct xfs_log_item, li_cil); + lv = item->li_lv;list_del_init (&item->li_cil); item->li_order_id =0 ; -if (!ctx->lv_chain) - ctx->lv_chain = item->li_lv; -else - lv->lv_next = item->li_lv; - lv = item->li_lv; item->li_lv =NULL ; - num_iovecs += lv->lv_niovecs;
Quoted 6 lines
fs/xfs/xfs_log_cil.c xlog_cil_push_work( · line 925
Quoted 1 line
fs/xfs/xfs_log_cil.c +1
-0+ num_iovecs += lv->lv_niovecs;Not sure why "lv =
item->li_lv" needed to move up?I think the only change needed here is replacing the lv_chain/lv_next business with the
list_add_tail?
Yes, but someone complained about the awful diff in the next patch, so moving the "lv = item->li_lv" made the diff in the next patch much, much cleaner...
<shrug>
I can move it back to the next patch if you really want, but it's really just shuffling deck chairs at this point...
Quoted 8 lines
Quoted 8 lines
fs/xfs/xfs_log_cil.c +2
-2@@ -985,8 +985,14 @@ xlog_cil_push_work(* use the commit record lsn then we can move the tail beyond the grant * write head. */ - error =xlog_write (log, &lvhdr, ctx->ticket, &ctx->start_lsn,NULL , - num_bytes); + error =xlog_write (log, &ctx->lv_chain, ctx->ticket, &ctx->start_lsn, +NULL , num_bytes);
Quoted 8 lines
fs/xfs/xfs_log_cil.c xlog_cil_push_work( · line 995
Quoted 6 lines
fs/xfs/xfs_log_cil.c +6
-0+ +/* +* Take the lvhdr back off the lv_chain as it should not be passed +* to log IO completion. +*/ +list_del (&lvhdr.lv_list);Seems a little clunky, but I guess I see why it's needed.
I could replace the stack structure with a memory allocation and then we wouldn't need to care, but I'm trying to keep memory allocation out of this fast path as much as possible....
Quoted 2 lines
I /think/ I don't see any place where the onstack lvhdr can escape out of the chain after
_push_workreturns, so this is safe enough.
It can't, because we own the chain here and are completely responsible for cleaning it up on failure.
Cheers,
Dave.
Signature · 2 lines
Dave Chinner [email protected]
Darrick J. Wong
Reviewed-byRe: [PATCH 35/39] xfs: convert log vector chain to use list heads
Quoted 31 lines · Dave Chinner
Quoted 1 line · Darrick J. Wong
On Wed, May 19, 2021 at 10:13:13PM +1000, Dave Chinner wrote:
.....
Quoted 21 lines
Quoted 21 lines
fs/xfs/xfs_log_cil.c +1
-7@@ -913,25 +912,23 @@ xlog_cil_push_work(xlog_cil_pcp_aggregate (cil, ctx);list_sort (NULL , &ctx->log_items, xlog_cil_order_cmp); -while (!list_empty (&ctx->log_items)) {struct xfs_log_item *item; item =list_first_entry (&ctx->log_items, struct xfs_log_item, li_cil); + lv = item->li_lv;list_del_init (&item->li_cil); item->li_order_id =0 ; -if (!ctx->lv_chain) - ctx->lv_chain = item->li_lv; -else - lv->lv_next = item->li_lv; - lv = item->li_lv; item->li_lv =NULL ; - num_iovecs += lv->lv_niovecs;Quoted 6 lines
fs/xfs/xfs_log_cil.c xlog_cil_push_work( · line 925
Quoted 1 line
fs/xfs/xfs_log_cil.c +1
-0+ num_iovecs += lv->lv_niovecs;Not sure why "lv =
item->li_lv" needed to move up?I think the only change needed here is replacing the lv_chain/lv_next business with the
list_add_tail?
Quoted 8 lines
Yes, but someone complained about the awful diff in the next patch, so moving the "lv =
item->li_lv" made the diff in the next patch much, much cleaner...<shrug>
I can move it back to the next patch if you really want, but it's really just shuffling deck chairs at this point...
Nope, don't care that much.
Quoted 17 lines
Quoted 8 lines
Quoted 8 lines
fs/xfs/xfs_log_cil.c +2
-2@@ -985,8 +985,14 @@ xlog_cil_push_work(* use the commit record lsn then we can move the tail beyond the grant * write head. */ - error =xlog_write (log, &lvhdr, ctx->ticket, &ctx->start_lsn,NULL , - num_bytes); + error =xlog_write (log, &ctx->lv_chain, ctx->ticket, &ctx->start_lsn, +NULL , num_bytes);Quoted 8 lines
fs/xfs/xfs_log_cil.c xlog_cil_push_work( · line 995
Quoted 6 lines
fs/xfs/xfs_log_cil.c +6
-0+ +/* +* Take the lvhdr back off the lv_chain as it should not be passed +* to log IO completion. +*/ +list_del (&lvhdr.lv_list);Seems a little clunky, but I guess I see why it's needed.
Quoted 3 lines
I could replace the stack structure with a memory allocation and then we wouldn't need to care, but I'm trying to keep memory allocation out of this fast path as much as possible....
Oh, that's much worse.
Quoted 5 lines
Quoted 2 lines
I /think/ I don't see any place where the onstack lvhdr can escape out of the chain after
_push_workreturns, so this is safe enough.It can't, because we own the chain here and are completely responsible for cleaning it up on failure.
Ok. I think I'm satisfied now:
--D
Quoted 7 lines
Cheers,
Dave.
Signature · 2 lines
Dave Chinner [email protected]
PATCH v4 36/39 xfs: move CIL ordering to the logvec chain
In reply to Dave ChinnerFrom: Dave Chinner <[email protected]>
Adding a list_sort() call to the CIL push work while the xc_ctx_lock is held exclusively has resulted in fairly long lock hold times and that stops all front end transaction commits from making progress.
We can move the sorting out of the xc_ctx_lock if we can transfer the ordering information to the log vectors as they are detached from the log items and then we can sort the log vectors. With these changes, we can move the list_sort() call to just before we call xlog_write() when we aren't holding any locks at all.
--- fs/xfs/xfs_log.h | 1 + fs/xfs/xfs_log_cil.c | 23 ++++++++++++++--------- 2 files changed, 15 insertions(+), 9 deletions(-)
fs/xfs/xfs_log.h +1 -0
@@ -10,6 +10,7 @@ struct xfs_cil_ctx;struct xfs_log_vec {struct list_head lv_list;/* CIL lv chain ptrs */ +uint32_t lv_order_id;/* chain ordering info */ int lv_niovecs;/* number of iovecs in lv */ struct xfs_log_iovec *lv_iovecp;/* iovec array */ struct xfs_log_item *lv_item;/* owner */
fs/xfs/xfs_log_cil.c +14 -9
@@ -791,10 +791,10 @@ xlog_cil_order_cmp(const struct list_head *a, const struct list_head *b) { - struct xfs_log_item *l1 = container_of(a, struct xfs_log_item, li_cil); - struct xfs_log_item *l2 = container_of(b, struct xfs_log_item, li_cil); + struct xfs_log_vec *l1 = container_of(a, struct xfs_log_vec, lv_list); + struct xfs_log_vec *l2 = container_of(b, struct xfs_log_vec, lv_list); -return l1->li_order_id > l2->li_order_id; +return l1->lv_order_id > l2->lv_order_id; }/*
@@ -911,24 +911,22 @@ xlog_cil_push_work(xlog_cil_pcp_aggregate (cil, ctx); -list_sort (NULL , &ctx->log_items, xlog_cil_order_cmp);while (!list_empty (&ctx->log_items)) {struct xfs_log_item *item; item =list_first_entry (&ctx->log_items, struct xfs_log_item, li_cil); lv = item->li_lv; -list_del_init (&item->li_cil); - item->li_order_id =0 ; - item->li_lv =NULL ; - + lv->lv_order_id = item->li_order_id; num_iovecs += lv->lv_niovecs;/* we don't write ordered log vectors */ if (lv->lv_buf_len !=XFS_LOG_VEC_ORDERED ) num_bytes += lv->lv_bytes;list_add_tail (&lv->lv_list, &ctx->lv_chain); - +list_del_init (&item->li_cil); + item->li_order_id =0 ; + item->li_lv =NULL ; }/*
@@ -961,6 +959,13 @@ xlog_cil_push_work(spin_unlock (&cil->xc_push_lock);up_write (&cil->xc_ctx_lock); +/* +* Sort the log vector chain before we add the transaction headers. +* This ensures we always have the transaction headers at the start +* of the chain. +*/ +list_sort (NULL , &ctx->lv_chain, xlog_cil_order_cmp); +/* * Build a checkpoint transaction header and write it to the log to * begin the transaction. We need to account for the space used by the
Signature · 1 line
2.31.1
Re: [PATCH 36/39] xfs: move CIL ordering to the logvec chain
Quoted 6 lines · Dave Chinner
From: Dave Chinner <[email protected]>
Adding a
list_sort()call to the CIL push work while thexc_ctx_lockis held exclusively has resulted in fairly long lock hold times and that stops all front end transaction commits from making progress.
Quoted 7 lines
We can move the sorting out of the
xc_ctx_lockif we can transfer the ordering information to the log vectors as they are detached from the log items and then we can sort the log vectors. With these changes, we can move thelist_sort()call to just before we callxlog_write()when we aren't holding any locks at all.Signed-off-by Dave Chinner <[email protected]>
Looks good to me,
--D
Quoted 81 lines
--- fs/xfs/xfs_log.h | 1 + fs/xfs/xfs_log_cil.c | 23 ++++++++++++++--------- 2 files changed, 15 insertions(+), 9 deletions(-)fs/xfs/xfs_log.h +1
-0@@ -10,6 +10,7 @@ struct xfs_cil_ctx;struct xfs_log_vec {struct list_head lv_list;/* CIL lv chain ptrs */ +uint32_t lv_order_id;/* chain ordering info */ int lv_niovecs;/* number of iovecs in lv */ struct xfs_log_iovec *lv_iovecp;/* iovec array */ struct xfs_log_item *lv_item;/* owner */ fs/xfs/xfs_log_cil.c +14
-9@@ -791,10 +791,10 @@ xlog_cil_order_cmp(const struct list_head *a, const struct list_head *b) { - struct xfs_log_item *l1 = container_of(a, struct xfs_log_item, li_cil); - struct xfs_log_item *l2 = container_of(b, struct xfs_log_item, li_cil); + struct xfs_log_vec *l1 = container_of(a, struct xfs_log_vec, lv_list); + struct xfs_log_vec *l2 = container_of(b, struct xfs_log_vec, lv_list); -return l1->li_order_id > l2->li_order_id; +return l1->lv_order_id > l2->lv_order_id; }/* @@ -911,24 +911,22 @@ xlog_cil_push_work(xlog_cil_pcp_aggregate (cil, ctx); -list_sort (NULL , &ctx->log_items, xlog_cil_order_cmp);while (!list_empty (&ctx->log_items)) {struct xfs_log_item *item; item =list_first_entry (&ctx->log_items, struct xfs_log_item, li_cil); lv = item->li_lv; -list_del_init (&item->li_cil); - item->li_order_id =0 ; - item->li_lv =NULL ; - + lv->lv_order_id = item->li_order_id; num_iovecs += lv->lv_niovecs;/* we don't write ordered log vectors */ if (lv->lv_buf_len !=XFS_LOG_VEC_ORDERED ) num_bytes += lv->lv_bytes;list_add_tail (&lv->lv_list, &ctx->lv_chain); - +list_del_init (&item->li_cil); + item->li_order_id =0 ; + item->li_lv =NULL ; }/* @@ -961,6 +959,13 @@ xlog_cil_push_work(spin_unlock (&cil->xc_push_lock);up_write (&cil->xc_ctx_lock); +/* +* Sort the log vector chain before we add the transaction headers. +* This ensures we always have the transaction headers at the start +* of the chain. +*/ +list_sort (NULL , &ctx->lv_chain, xlog_cil_order_cmp); +/* * Build a checkpoint transaction header and write it to the log to * begin the transaction. We need to account for the space used by the Signature · 1 line
2.31.1
PATCH v4 37/39 xfs: avoid cil push lock if possible
In reply to Dave ChinnerFrom: Dave Chinner <[email protected]>
Because now it hurts when the CIL fills up.
- 37.20% __xfs_trans_commit
- 35.84% xfs_log_commit_cil
- 19.34% _raw_spin_lock
- do_raw_spin_lock
19.01% __pv_queued_spin_lock_slowpath
- 4.20% xfs_log_ticket_ungrant
0.90% xfs_log_space_wake
--- fs/xfs/xfs_log_cil.c | 14 +++++++++++--- 1 file changed, 11 insertions(+), 3 deletions(-)
fs/xfs/xfs_log_cil.c +11 -3
@@ -1125,10 +1125,18 @@ xlog_cil_push_background(ASSERT (!test_bit (XLOG_CIL_EMPTY , &cil->xc_flags));/* -* Don't do a background push if we haven't used up all the -* space available yet. +* We are done if: +* - we haven't used up all the space available yet; or +* - we've already queued up a push; and +* - we're not over the hard limit; and +* - nothing has been over the hard limit. +* +* If so, we don't need to take the push lock as there's nothing to do. */ -if (space_used <XLOG_CIL_SPACE_LIMIT (log)) { +if (space_used <XLOG_CIL_SPACE_LIMIT (log) || + (cil->xc_push_seq == cil->xc_current_sequence && + space_used <XLOG_CIL_BLOCKING_SPACE_LIMIT (log) && + !waitqueue_active (&cil->xc_push_wait))) {up_read (&cil->xc_ctx_lock);return ; }
Signature · 1 line
2.31.1
Re: [PATCH 37/39] xfs: avoid cil push lock if possible
Quoted 13 lines · Dave Chinner
From: Dave Chinner <[email protected]>
Because now it hurts when the CIL fills up.
- 37.20%
__xfs_trans_commit
- 35.84%xfs_log_commit_cil
- 19.34%_raw_spin_lock
-do_raw_spin_lock
19.01%__pv_queued_spin_lock_slowpath
- 4.20%xfs_log_ticket_ungrant
0.90%xfs_log_space_wake
Quoted 1 line
Signed-off-by Dave Chinner <[email protected]>
Looks good,
--D
Quoted 33 lines
--- fs/xfs/xfs_log_cil.c | 14 +++++++++++--- 1 file changed, 11 insertions(+), 3 deletions(-)fs/xfs/xfs_log_cil.c +11
-3@@ -1125,10 +1125,18 @@ xlog_cil_push_background(ASSERT (!test_bit (XLOG_CIL_EMPTY , &cil->xc_flags));/* -* Don't do a background push if we haven't used up all the -* space available yet. +* We are done if: +* - we haven't used up all the space available yet; or +* - we've already queued up a push; and +* - we're not over the hard limit; and +* - nothing has been over the hard limit. +* +* If so, we don't need to take the push lock as there's nothing to do. */ -if (space_used <XLOG_CIL_SPACE_LIMIT (log)) { +if (space_used <XLOG_CIL_SPACE_LIMIT (log) || + (cil->xc_push_seq == cil->xc_current_sequence && + space_used <XLOG_CIL_BLOCKING_SPACE_LIMIT (log) && + !waitqueue_active (&cil->xc_push_wait))) {up_read (&cil->xc_ctx_lock);return ; }Signature · 1 line
2.31.1
PATCH v4 38/39 xfs: xlog_sync() manually adjusts grant head space
In reply to Dave ChinnerFrom: Dave Chinner <[email protected]>
When xlog_sync() rounds off the tail the iclog that is being flushed, it manually subtracts that space from the grant heads. This space is actually reserved by the transaction ticket that covers the xlog_sync() call from xlog_write(), but we don't plumb the ticket down far enough for it to account for the space consumed in the current log ticket.
The grant heads are hot, so we really should be accounting this to the ticket is we can, rather than adding thousands of extra grant head updates every CIL commit.
Interestingly, this actually indicates a potential log space overrun can occur when we force the log. By the time that xfs_log_force() pushes out an active iclog and consumes the roundoff space, the reservation for that roundoff space has been returned to the grant heads and is no longer covered by a reservation. In theory the roundoff added to log force on an already full log could push the write head past the tail. In practice, the CIL commit that writes to the log and needs the iclog pushed will have reserved space for roundoff, so when it releases the ticket there will still be physical space for the roundoff to be committed to the log, even though it is no longer reserved. This roundoff won't be enough space to allow a transaction to be woken if the log is full, so overruns should not actually occur in practice.
That said, it indicates that we should not release the CIL context log ticket until after we've released the commit iclog. It also means that xlog_sync() still needs the direct grant head manipulation if we don't provide it with a ticket. Log forces are rare when we are in fast paths running 1.5 million transactions/s that make the grant heads hot, so let's optimise the hot case and pass CIL log tickets down to the xlog_sync() code.
--- fs/xfs/xfs_log.c | 39 +++++++++++++++++++++++++-------------- fs/xfs/xfs_log_cil.c | 19 ++++++++++++++----- fs/xfs/xfs_log_priv.h | 3 ++- 3 files changed, 41 insertions(+), 20 deletions(-)
fs/xfs/xfs_log.c +25 -14
@@ -55,7 +55,8 @@ xlog_grant_push_ail( STATIC voidxlog_sync (struct xlog *log, -struct xlog_in_core *iclog); +struct xlog_in_core *iclog, +struct xlog_ticket *ticket);#if defined(DEBUG ) STATIC void xlog_verify_dest_ptr(
@@ -537,7 +538,8 @@ __xlog_state_release_iclog(int xlog_state_release_iclog (struct xlog *log, -struct xlog_in_core *iclog) +struct xlog_in_core *iclog, +struct xlog_ticket *ticket) {lockdep_assert_held (&log->l_icloglock);
@@ -547,7 +549,7 @@ xlog_state_release_iclog(if (atomic_dec_and_test (&iclog->ic_refcnt) &&__xlog_state_release_iclog (log, iclog)) {spin_unlock (&log->l_icloglock); -xlog_sync (log, iclog); +xlog_sync (log, iclog, ticket);spin_lock (&log->l_icloglock); }
@@ -908,7 +910,7 @@ xlog_unmount_write(* iclog containing the unmount record is written. */ iclog->ic_flags |= (XLOG_ICL_NEED_FLUSH |XLOG_ICL_NEED_FUA ); - error =xlog_state_release_iclog (log, iclog); + error =xlog_state_release_iclog (log, iclog, tic);xlog_wait_on_iclog (iclog);if (tic) {
@@ -1939,7 +1941,8 @@ xlog_calc_iclog_size( STATIC voidxlog_sync (struct xlog *log, -struct xlog_in_core *iclog) +struct xlog_in_core *iclog, +struct xlog_ticket *ticket) {unsigned int count;/* byte count of bwrite */ unsigned int roundoff;/* roundoff to BB or stripe */
@@ -1950,12 +1953,20 @@ xlog_sync( count =xlog_calc_iclog_size (log, iclog, &roundoff); -/* move grant heads by roundoff in sync */ -xlog_grant_add_space (log, &log->l_reserve_head.grant, roundoff); -xlog_grant_add_space (log, &log->l_write_head.grant, roundoff); +/* +* If we have a ticket, account for the roundoff via the ticket +* reservation to avoid touching the hot grant heads needlessly. +* Otherwise, we have to move grant heads directly. +*/ +if (ticket) { + ticket->t_curr_res -= roundoff; + }else { +xlog_grant_add_space (log, &log->l_reserve_head.grant, roundoff); +xlog_grant_add_space (log, &log->l_write_head.grant, roundoff); + }/* put cycle number in every block */ -xlog_pack_data (log, iclog, roundoff); +xlog_pack_data (log, iclog, roundoff);/* real byte length */ size = iclog->ic_offset;
@@ -2195,7 +2206,7 @@ xlog_write_get_more_iclog_space(xlog_state_finish_copy (log, iclog, *record_cnt, *data_cnt);ASSERT (iclog->ic_state ==XLOG_STATE_WANT_SYNC || iclog->ic_state ==XLOG_STATE_IOERROR ); - error =xlog_state_release_iclog (log, iclog); + error =xlog_state_release_iclog (log, iclog, ticket);spin_unlock (&log->l_icloglock);if (error)return error;
@@ -2465,7 +2476,7 @@ xlog_write(if (commit_iclog) *commit_iclog = iclog;else - error =xlog_state_release_iclog (log, iclog); + error =xlog_state_release_iclog (log, iclog, ticket);spin_unlock (&log->l_icloglock);return error;
@@ -2923,7 +2934,7 @@ xlog_state_get_iclog_space(* reference to the iclog. */ if (!atomic_add_unless (&iclog->ic_refcnt,-1 ,1 )) - error =xlog_state_release_iclog (log, iclog); + error =xlog_state_release_iclog (log, iclog, ticket);spin_unlock (&log->l_icloglock);if (error)return error;
@@ -3151,7 +3162,7 @@ xfs_log_force(atomic_inc (&iclog->ic_refcnt); lsn =be64_to_cpu (iclog->ic_header.h_lsn);xlog_state_switch_iclogs (log, iclog,0 ); -if (xlog_state_release_iclog (log, iclog)) +if (xlog_state_release_iclog (log, iclog,NULL )) goto out_error;if (be64_to_cpu (iclog->ic_header.h_lsn) != lsn)
@@ -3244,7 +3255,7 @@ xlog_force_lsn( }atomic_inc (&iclog->ic_refcnt);xlog_state_switch_iclogs (log, iclog,0 ); -if (xlog_state_release_iclog (log, iclog)) +if (xlog_state_release_iclog (log, iclog,NULL )) goto out_error;if (log_flushed) *log_flushed =1 ;
fs/xfs/xfs_log_cil.c +14 -5
@@ -832,6 +832,7 @@ xlog_cil_push_work(struct bio bio;DECLARE_COMPLETION_ONSTACK (bdev_flush);bool push_commit_stable; +struct xlog_ticket *ticket; new_ctx =xlog_cil_ctx_alloc (); new_ctx->ticket =xlog_cil_ticket_alloc (log);
@@ -1039,12 +1040,10 @@ xlog_cil_push_work(if (error) goto out_abort_free_ticket; -xfs_log_ticket_ungrant (log, ctx->ticket); -spin_lock (&commit_iclog->ic_callback_lock);if (commit_iclog->ic_state ==XLOG_STATE_IOERROR ) {spin_unlock (&commit_iclog->ic_callback_lock); - goto out_abort; + goto out_abort_free_ticket; }ASSERT_ALWAYS (commit_iclog->ic_state ==XLOG_STATE_ACTIVE || commit_iclog->ic_state ==XLOG_STATE_WANT_SYNC );
@@ -1061,6 +1060,15 @@ xlog_cil_push_work(wake_up_all (&cil->xc_commit_wait);spin_unlock (&cil->xc_push_lock); +/* +* Pull the ticket off the ctx so we can ungrant it after releasing the +* commit_iclog. The ctx may be freed by the time we return from +* releasing the commit_iclog (i.e. checkpoint has been completed and +* callback run) so we can't reference the ctx after the call to +* xlog_state_release_iclog(). +*/ + ticket = ctx->ticket; +/* * If the checkpoint spans multiple iclogs, wait for all previous * iclogs to complete before we submit the commit_iclog. In this case,
@@ -1087,8 +1095,10 @@ xlog_cil_push_work( commit_iclog->ic_flags |=XLOG_ICL_NEED_FUA ;if (push_commit_stable && commit_iclog->ic_state ==XLOG_STATE_ACTIVE )xlog_state_switch_iclogs (log, commit_iclog,0 ); -xlog_state_release_iclog (log, commit_iclog); +xlog_state_release_iclog (log, commit_iclog, ticket);spin_unlock (&log->l_icloglock); + +xfs_log_ticket_ungrant (log, ticket);return ; out_skip:
@@ -1099,7 +1109,6 @@ xlog_cil_push_work( out_abort_free_ticket:xfs_log_ticket_ungrant (log, ctx->ticket); -out_abort:ASSERT (XLOG_FORCED_SHUTDOWN (log));xlog_cil_committed (ctx); }
fs/xfs/xfs_log_priv.h +2 -1
@@ -488,7 +488,8 @@ void xfs_log_ticket_regrant(struct xlog *log, struct xlog_ticket *ticket);void xlog_state_switch_iclogs (struct xlog *log,struct xlog_in_core *iclog,int eventual_size); -int xlog_state_release_iclog (struct xlog *log,struct xlog_in_core *iclog); +int xlog_state_release_iclog (struct xlog *xlog,struct xlog_in_core *iclog, +struct xlog_ticket *ticket);/* * When we crack an atomic LSN, we sample it first so that the value will not
Signature · 1 line
2.31.1