From: Dave Chinner <david@fromorbit.com>
To: linux-xfs@vger.kernel.org
Subject: [PATCH 34/45] xfs: rework per-iclog header CIL reservation
Date: Fri, 5 Mar 2021 16:11:32 +1100 [thread overview]
Message-ID: <20210305051143.182133-35-david@fromorbit.com> (raw)
In-Reply-To: <20210305051143.182133-1-david@fromorbit.com>
From: Dave Chinner <dchinner@redhat.com>
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.
Signed-off-by: Dave Chinner <dchinner@redhat.com>
---
fs/xfs/libxfs/xfs_log_rlimit.c | 2 +-
fs/xfs/libxfs/xfs_shared.h | 3 +-
fs/xfs/xfs_log.c | 12 +++++---
fs/xfs/xfs_log_cil.c | 55 ++++++++++++++++++++++++++--------
fs/xfs/xfs_log_priv.h | 20 +++++++------
5 files changed, 64 insertions(+), 28 deletions(-)
diff --git a/fs/xfs/libxfs/xfs_log_rlimit.c b/fs/xfs/libxfs/xfs_log_rlimit.c
index 7f55eb3f3653..75390134346d 100644
--- a/fs/xfs/libxfs/xfs_log_rlimit.c
+++ b/fs/xfs/libxfs/xfs_log_rlimit.c
@@ -88,7 +88,7 @@ xfs_log_calc_minimum_size(
xfs_log_get_max_trans_res(mp, &tres);
- max_logres = xfs_log_calc_unit_res(mp, tres.tr_logres);
+ max_logres = xfs_log_calc_unit_res(mp, tres.tr_logres, NULL);
if (tres.tr_logcount > 1)
max_logres *= tres.tr_logcount;
diff --git a/fs/xfs/libxfs/xfs_shared.h b/fs/xfs/libxfs/xfs_shared.h
index 8c61a461bf7b..b4791b817fe3 100644
--- a/fs/xfs/libxfs/xfs_shared.h
+++ b/fs/xfs/libxfs/xfs_shared.h
@@ -48,7 +48,8 @@ extern const struct xfs_buf_ops xfs_symlink_buf_ops;
extern const struct xfs_buf_ops xfs_rtbuf_ops;
/* log size calculation functions */
-int xfs_log_calc_unit_res(struct xfs_mount *mp, int unit_bytes);
+int xfs_log_calc_unit_res(struct xfs_mount *mp, int unit_bytes,
+ int *niclogs);
int xfs_log_calc_minimum_size(struct xfs_mount *);
struct xfs_trans_res;
diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c
index 8f4f7ae84358..46a006d41184 100644
--- a/fs/xfs/xfs_log.c
+++ b/fs/xfs/xfs_log.c
@@ -3312,7 +3312,8 @@ 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;
@@ -3392,15 +3393,18 @@ 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;
}
int
xfs_log_calc_unit_res(
struct xfs_mount *mp,
- int unit_bytes)
+ int unit_bytes,
+ int *niclogs)
{
- return xlog_calc_unit_res(mp->m_log, unit_bytes);
+ return xlog_calc_unit_res(mp->m_log, unit_bytes, niclogs);
}
/*
@@ -3418,7 +3422,7 @@ 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;
diff --git a/fs/xfs/xfs_log_cil.c b/fs/xfs/xfs_log_cil.c
index 50101336a7f4..f8fb2f59e24c 100644
--- a/fs/xfs/xfs_log_cil.c
+++ b/fs/xfs/xfs_log_cil.c
@@ -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;
/*
diff --git a/fs/xfs/xfs_log_priv.h b/fs/xfs/xfs_log_priv.h
index b0dc3bc9de59..e72d14c76e03 100644
--- a/fs/xfs/xfs_log_priv.h
+++ b/fs/xfs/xfs_log_priv.h
@@ -140,15 +140,16 @@ enum xlog_iclog_state {
#define XLOG_TIC_LEN_MAX 15
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;
/*
@@ -249,6 +250,7 @@ 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;
--
2.28.0
next prev parent reply other threads:[~2021-03-05 5:12 UTC|newest]
Thread overview: 145+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-03-05 5:10 [PATCH 00/45 v3] xfs: consolidated log and optimisation changes Dave Chinner
2021-03-05 5:10 ` [PATCH 01/45] xfs: initialise attr fork on inode create Dave Chinner
2021-03-08 22:20 ` Darrick J. Wong
2021-03-16 8:35 ` Christoph Hellwig
2021-03-05 5:11 ` [PATCH 02/45] xfs: log stripe roundoff is a property of the log Dave Chinner
2021-03-05 5:11 ` [PATCH 03/45] xfs: separate CIL commit record IO Dave Chinner
2021-03-08 8:34 ` Chandan Babu R
2021-03-15 14:40 ` Brian Foster
2021-03-16 8:40 ` Christoph Hellwig
2021-03-05 5:11 ` [PATCH 04/45] xfs: remove xfs_blkdev_issue_flush Dave Chinner
2021-03-08 9:31 ` Chandan Babu R
2021-03-08 22:21 ` Darrick J. Wong
2021-03-15 14:40 ` Brian Foster
2021-03-16 8:41 ` Christoph Hellwig
2021-03-05 5:11 ` [PATCH 05/45] xfs: async blkdev cache flush Dave Chinner
2021-03-08 9:48 ` Chandan Babu R
2021-03-08 22:24 ` Darrick J. Wong
2021-03-15 14:41 ` Brian Foster
2021-03-15 16:32 ` Darrick J. Wong
2021-03-16 8:43 ` Christoph Hellwig
2021-03-08 22:26 ` Darrick J. Wong
2021-03-15 14:42 ` Brian Foster
2021-03-05 5:11 ` [PATCH 06/45] xfs: CIL checkpoint flushes caches unconditionally Dave Chinner
2021-03-15 14:43 ` Brian Foster
2021-03-16 8:47 ` Christoph Hellwig
2021-03-05 5:11 ` [PATCH 07/45] xfs: remove need_start_rec parameter from xlog_write() Dave Chinner
2021-03-15 14:45 ` Brian Foster
2021-03-16 14:15 ` Christoph Hellwig
2021-03-05 5:11 ` [PATCH 08/45] xfs: journal IO cache flush reductions Dave Chinner
2021-03-08 10:49 ` Chandan Babu R
2021-03-08 12:25 ` Brian Foster
2021-03-09 1:13 ` Dave Chinner
2021-03-10 20:49 ` Brian Foster
2021-03-10 21:28 ` Dave Chinner
2021-03-05 5:11 ` [PATCH 09/45] xfs: Fix CIL throttle hang when CIL space used going backwards Dave Chinner
2021-03-05 5:11 ` [PATCH 10/45] xfs: reduce buffer log item shadow allocations Dave Chinner
2021-03-15 14:52 ` Brian Foster
2021-03-05 5:11 ` [PATCH 11/45] xfs: xfs_buf_item_size_segment() needs to pass segment offset Dave Chinner
2021-03-05 5:11 ` [PATCH 12/45] xfs: optimise xfs_buf_item_size/format for contiguous regions Dave Chinner
2021-03-05 5:11 ` [PATCH 13/45] xfs: xfs_log_force_lsn isn't passed a LSN Dave Chinner
2021-03-08 22:53 ` Darrick J. Wong
2021-03-11 0:26 ` Dave Chinner
2021-03-05 5:11 ` [PATCH 14/45] xfs: AIL needs asynchronous CIL forcing Dave Chinner
2021-03-08 23:45 ` Darrick J. Wong
2021-03-05 5:11 ` [PATCH 15/45] xfs: CIL work is serialised, not pipelined Dave Chinner
2021-03-08 23:14 ` Darrick J. Wong
2021-03-08 23:38 ` Dave Chinner
2021-03-09 1:55 ` Darrick J. Wong
2021-03-09 22:35 ` Andi Kleen
2021-03-10 6:11 ` Dave Chinner
2021-03-05 5:11 ` [PATCH 16/45] xfs: type verification is expensive Dave Chinner
2021-03-05 5:11 ` [PATCH 17/45] xfs: No need for inode number error injection in __xfs_dir3_data_check Dave Chinner
2021-03-05 5:11 ` [PATCH 18/45] xfs: reduce debug overhead of dir leaf/node checks Dave Chinner
2021-03-05 5:11 ` [PATCH 19/45] xfs: factor out the CIL transaction header building Dave Chinner
2021-03-08 23:47 ` Darrick J. Wong
2021-03-16 14:50 ` Brian Foster
2021-03-05 5:11 ` [PATCH 20/45] xfs: only CIL pushes require a start record Dave Chinner
2021-03-09 0:07 ` Darrick J. Wong
2021-03-16 14:51 ` Brian Foster
2021-03-05 5:11 ` [PATCH 21/45] xfs: embed the xlog_op_header in the unmount record Dave Chinner
2021-03-09 0:15 ` Darrick J. Wong
2021-03-11 2:54 ` Dave Chinner
2021-03-05 5:11 ` [PATCH 22/45] xfs: embed the xlog_op_header in the commit record Dave Chinner
2021-03-09 0:17 ` Darrick J. Wong
2021-03-05 5:11 ` [PATCH 23/45] xfs: log tickets don't need log client id Dave Chinner
2021-03-09 0:21 ` Darrick J. Wong
2021-03-09 1:19 ` Dave Chinner
2021-03-09 1:48 ` Darrick J. Wong
2021-03-11 3:01 ` Dave Chinner
2021-03-16 14:51 ` Brian Foster
2021-03-05 5:11 ` [PATCH 24/45] xfs: move log iovec alignment to preparation function Dave Chinner
2021-03-09 2:14 ` Darrick J. Wong
2021-03-16 14:51 ` Brian Foster
2021-03-05 5:11 ` [PATCH 25/45] xfs: reserve space and initialise xlog_op_header in item formatting Dave Chinner
2021-03-09 2:21 ` Darrick J. Wong
2021-03-11 3:29 ` Dave Chinner
2021-03-11 3:41 ` Darrick J. Wong
2021-03-16 14:54 ` Brian Foster
2021-03-16 14:53 ` Brian Foster
2021-05-19 3:18 ` Dave Chinner
2021-03-05 5:11 ` [PATCH 26/45] xfs: log ticket region debug is largely useless Dave Chinner
2021-03-09 2:31 ` Darrick J. Wong
2021-03-16 14:55 ` Brian Foster
2021-05-19 3:27 ` Dave Chinner
2021-03-05 5:11 ` [PATCH 27/45] xfs: pass lv chain length into xlog_write() Dave Chinner
2021-03-09 2:36 ` Darrick J. Wong
2021-03-11 3:37 ` Dave Chinner
2021-03-16 18:38 ` Brian Foster
2021-03-05 5:11 ` [PATCH 28/45] xfs: introduce xlog_write_single() Dave Chinner
2021-03-09 2:39 ` Darrick J. Wong
2021-03-11 4:19 ` Dave Chinner
2021-03-16 18:39 ` Brian Foster
2021-05-19 3:44 ` Dave Chinner
2021-03-05 5:11 ` [PATCH 29/45] xfs:_introduce xlog_write_partial() Dave Chinner
2021-03-09 2:59 ` Darrick J. Wong
2021-03-11 4:33 ` Dave Chinner
2021-03-18 13:22 ` Brian Foster
2021-05-19 4:49 ` Dave Chinner
2021-05-20 12:33 ` Brian Foster
2021-05-27 18:03 ` Darrick J. Wong
2021-03-05 5:11 ` [PATCH 30/45] xfs: xlog_write() no longer needs contwr state Dave Chinner
2021-03-09 3:01 ` Darrick J. Wong
2021-03-05 5:11 ` [PATCH 31/45] xfs: CIL context doesn't need to count iovecs Dave Chinner
2021-03-09 3:16 ` Darrick J. Wong
2021-03-11 5:03 ` Dave Chinner
2021-03-05 5:11 ` [PATCH 32/45] xfs: use the CIL space used counter for emptiness checks Dave Chinner
2021-03-10 23:01 ` Darrick J. Wong
2021-03-05 5:11 ` [PATCH 33/45] xfs: lift init CIL reservation out of xc_cil_lock Dave Chinner
2021-03-10 23:25 ` Darrick J. Wong
2021-03-11 5:42 ` Dave Chinner
2021-03-05 5:11 ` Dave Chinner [this message]
2021-03-11 0:03 ` [PATCH 34/45] xfs: rework per-iclog header CIL reservation Darrick J. Wong
2021-03-11 6:03 ` Dave Chinner
2021-03-05 5:11 ` [PATCH 35/45] xfs: introduce per-cpu CIL tracking sructure Dave Chinner
2021-03-11 0:11 ` Darrick J. Wong
2021-03-11 6:33 ` Dave Chinner
2021-03-11 6:42 ` Dave Chinner
2021-03-05 5:11 ` [PATCH 36/45] xfs: implement percpu cil space used calculation Dave Chinner
2021-03-11 0:20 ` Darrick J. Wong
2021-03-11 6:51 ` Dave Chinner
2021-03-05 5:11 ` [PATCH 37/45] xfs: track CIL ticket reservation in percpu structure Dave Chinner
2021-03-11 0:26 ` Darrick J. Wong
2021-03-12 0:47 ` Dave Chinner
2021-03-05 5:11 ` [PATCH 38/45] xfs: convert CIL busy extents to per-cpu Dave Chinner
2021-03-11 0:36 ` Darrick J. Wong
2021-03-12 1:15 ` Dave Chinner
2021-03-05 5:11 ` [PATCH 39/45] xfs: Add order IDs to log items in CIL Dave Chinner
2021-03-11 1:00 ` Darrick J. Wong
2021-03-05 5:11 ` [PATCH 40/45] xfs: convert CIL to unordered per cpu lists Dave Chinner
2021-03-11 1:15 ` Darrick J. Wong
2021-03-12 2:18 ` Dave Chinner
2021-03-05 5:11 ` [PATCH 41/45] xfs: move CIL ordering to the logvec chain Dave Chinner
2021-03-11 1:34 ` Darrick J. Wong
2021-03-12 2:29 ` Dave Chinner
2021-03-05 5:11 ` [PATCH 42/45] xfs: __percpu_counter_compare() inode count debug too expensive Dave Chinner
2021-03-11 1:36 ` Darrick J. Wong
2021-03-05 5:11 ` [PATCH 43/45] xfs: avoid cil push lock if possible Dave Chinner
2021-03-11 1:47 ` Darrick J. Wong
2021-03-12 2:36 ` Dave Chinner
2021-03-05 5:11 ` [PATCH 44/45] xfs: xlog_sync() manually adjusts grant head space Dave Chinner
2021-03-11 2:00 ` Darrick J. Wong
2021-03-16 3:04 ` Dave Chinner
2021-03-05 5:11 ` [PATCH 45/45] xfs: expanding delayed logging design with background material Dave Chinner
2021-03-11 2:30 ` Darrick J. Wong
2021-03-16 3:28 ` Dave Chinner
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20210305051143.182133-35-david@fromorbit.com \
--to=david@fromorbit.com \
--cc=linux-xfs@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox