Commit 31151cc3 authored by Dave Chinner's avatar Dave Chinner Committed by Dave Chinner

xfs: rework per-iclog header CIL reservation

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: default avatarDave Chinner <dchinner@redhat.com>
Reviewed-by: default avatarDarrick J. Wong <djwong@kernel.org>
parent 12380d23
...@@ -3406,7 +3406,8 @@ xfs_log_ticket_get( ...@@ -3406,7 +3406,8 @@ xfs_log_ticket_get(
static int static int
xlog_calc_unit_res( xlog_calc_unit_res(
struct xlog *log, struct xlog *log,
int unit_bytes) int unit_bytes,
int *niclogs)
{ {
int iclog_space; int iclog_space;
uint num_headers; uint num_headers;
...@@ -3486,6 +3487,8 @@ xlog_calc_unit_res( ...@@ -3486,6 +3487,8 @@ xlog_calc_unit_res(
/* roundoff padding for transaction data and one for commit record */ /* roundoff padding for transaction data and one for commit record */
unit_bytes += 2 * log->l_iclog_roundoff; unit_bytes += 2 * log->l_iclog_roundoff;
if (niclogs)
*niclogs = num_headers;
return unit_bytes; return unit_bytes;
} }
...@@ -3494,7 +3497,7 @@ xfs_log_calc_unit_res( ...@@ -3494,7 +3497,7 @@ xfs_log_calc_unit_res(
struct xfs_mount *mp, struct xfs_mount *mp,
int unit_bytes) int unit_bytes)
{ {
return xlog_calc_unit_res(mp->m_log, unit_bytes); return xlog_calc_unit_res(mp->m_log, unit_bytes, NULL);
} }
/* /*
...@@ -3512,7 +3515,7 @@ xlog_ticket_alloc( ...@@ -3512,7 +3515,7 @@ xlog_ticket_alloc(
tic = kmem_cache_zalloc(xfs_log_ticket_cache, GFP_NOFS | __GFP_NOFAIL); tic = kmem_cache_zalloc(xfs_log_ticket_cache, 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); atomic_set(&tic->t_ref, 1);
tic->t_task = current; tic->t_task = current;
......
...@@ -44,9 +44,20 @@ xlog_cil_ticket_alloc( ...@@ -44,9 +44,20 @@ xlog_cil_ticket_alloc(
* transaction overhead reservation from the first transaction commit. * transaction overhead reservation from the first transaction commit.
*/ */
tic->t_curr_res = 0; tic->t_curr_res = 0;
tic->t_iclog_hdrs = 0;
return tic; 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)));
}
/* /*
* Check if the current log item was first committed in this sequence. * Check if the current log item was first committed in this sequence.
* We can't rely on just the log item being in the CIL, we have to check * We can't rely on just the log item being in the CIL, we have to check
...@@ -102,6 +113,7 @@ xlog_cil_ctx_switch( ...@@ -102,6 +113,7 @@ xlog_cil_ctx_switch(
struct xfs_cil *cil, struct xfs_cil *cil,
struct xfs_cil_ctx *ctx) struct xfs_cil_ctx *ctx)
{ {
xlog_cil_set_iclog_hdr_count(cil);
set_bit(XLOG_CIL_EMPTY, &cil->xc_flags); set_bit(XLOG_CIL_EMPTY, &cil->xc_flags);
ctx->sequence = ++cil->xc_current_sequence; ctx->sequence = ++cil->xc_current_sequence;
ctx->cil = cil; ctx->cil = cil;
...@@ -124,6 +136,7 @@ xlog_cil_init_post_recovery( ...@@ -124,6 +136,7 @@ xlog_cil_init_post_recovery(
{ {
log->l_cilp->xc_ctx->ticket = xlog_cil_ticket_alloc(log); log->l_cilp->xc_ctx->ticket = xlog_cil_ticket_alloc(log);
log->l_cilp->xc_ctx->sequence = 1; log->l_cilp->xc_ctx->sequence = 1;
xlog_cil_set_iclog_hdr_count(log->l_cilp);
} }
static inline int static inline int
...@@ -451,7 +464,6 @@ xlog_cil_insert_items( ...@@ -451,7 +464,6 @@ xlog_cil_insert_items(
struct xfs_cil_ctx *ctx = cil->xc_ctx; struct xfs_cil_ctx *ctx = cil->xc_ctx;
struct xfs_log_item *lip; struct xfs_log_item *lip;
int len = 0; int len = 0;
int iclog_space;
int iovhdr_res = 0, split_res = 0, ctx_res = 0; int iovhdr_res = 0, split_res = 0, ctx_res = 0;
ASSERT(tp); ASSERT(tp);
...@@ -474,19 +486,36 @@ xlog_cil_insert_items( ...@@ -474,19 +486,36 @@ xlog_cil_insert_items(
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_res = ctx->ticket->t_unit_res;
spin_lock(&cil->xc_cil_lock); /*
* Check if we need to steal iclog headers. atomic_read() is not a
/* do we need space for more log record headers? */ * locked atomic operation, so we can check the value before we do any
iclog_space = log->l_iclog_size - log->l_iclog_hsize; * real atomic ops in the fast path. If we've already taken the CIL unit
if (len > 0 && (ctx->space_used / iclog_space != * reservation from this commit, we've already got one iclog header
(ctx->space_used + len) / iclog_space)) { * space reserved so we have to account for that otherwise we risk
split_res = (len + iclog_space - 1) / iclog_space; * overrunning the reservation on this ticket.
/* need to take into account split region headers, too */ *
split_res *= log->l_iclog_hsize + sizeof(struct xlog_op_header); * If the CIL is already at the hard limit, we might need more header
ctx->ticket->t_unit_res += split_res; * 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; ctx->space_used += len;
tp->t_ticket->t_curr_res += released_space; tp->t_ticket->t_curr_res += released_space;
......
...@@ -143,15 +143,16 @@ enum xlog_iclog_state { ...@@ -143,15 +143,16 @@ enum xlog_iclog_state {
#define XLOG_COVER_OPS 5 #define XLOG_COVER_OPS 5
typedef struct xlog_ticket { typedef struct xlog_ticket {
struct list_head t_queue; /* reserve/write queue */ struct list_head t_queue; /* reserve/write queue */
struct task_struct *t_task; /* task that owns this ticket */ struct task_struct *t_task; /* task that owns this ticket */
xlog_tid_t t_tid; /* transaction identifier : 4 */ xlog_tid_t t_tid; /* transaction identifier */
atomic_t t_ref; /* ticket reference count : 4 */ atomic_t t_ref; /* ticket reference count */
int t_curr_res; /* current reservation in bytes : 4 */ int t_curr_res; /* current reservation */
int t_unit_res; /* unit reservation in bytes : 4 */ int t_unit_res; /* unit reservation */
char t_ocnt; /* original count : 1 */ char t_ocnt; /* original unit count */
char t_cnt; /* current count : 1 */ char t_cnt; /* current unit count */
uint8_t t_flags; /* properties of reservation : 1 */ uint8_t t_flags; /* properties of reservation */
int t_iclog_hdrs; /* iclog hdrs in t_curr_res */
} xlog_ticket_t; } xlog_ticket_t;
/* /*
...@@ -249,6 +250,7 @@ struct xfs_cil_ctx { ...@@ -249,6 +250,7 @@ struct xfs_cil_ctx {
struct xfs_cil { struct xfs_cil {
struct xlog *xc_log; struct xlog *xc_log;
unsigned long xc_flags; unsigned long xc_flags;
atomic_t xc_iclog_hdrs;
struct list_head xc_cil; struct list_head xc_cil;
spinlock_t xc_cil_lock; spinlock_t xc_cil_lock;
struct workqueue_struct *xc_push_wq; struct workqueue_struct *xc_push_wq;
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment