#include "journal_reclaim.h"
#include "journal_sb.h"
#include "journal_seq_blacklist.h"
-
-#include <trace/events/bcachefs.h>
-
-#define x(n) #n,
-static const char * const bch2_journal_watermarks[] = {
- JOURNAL_WATERMARKS()
- NULL
-};
+#include "trace.h"
static const char * const bch2_journal_errors[] = {
+#define x(n) #n,
JOURNAL_ERRORS()
+#undef x
NULL
};
-#undef x
static inline bool journal_seq_unwritten(struct journal *j, u64 seq)
{
static void journal_pin_list_init(struct journal_entry_pin_list *p, int count)
{
unsigned i;
+
for (i = 0; i < ARRAY_SIZE(p->list); i++)
INIT_LIST_HEAD(&p->list[i]);
INIT_LIST_HEAD(&p->flushed);
p->devs.nr = 0;
}
-/* journal entry close/open: */
+/*
+ * Detect stuck journal conditions and trigger shutdown. Technically the journal
+ * can end up stuck for a variety of reasons, such as a blocked I/O, journal
+ * reservation lockup, etc. Since this is a fatal error with potentially
+ * unpredictable characteristics, we want to be fairly conservative before we
+ * decide to shut things down.
+ *
+ * Consider the journal stuck when it appears full with no ability to commit
+ * btree transactions, to discard journal buckets, nor acquire priority
+ * (reserved watermark) reservation.
+ */
+static inline bool
+journal_error_check_stuck(struct journal *j, int error, unsigned flags)
+{
+ struct bch_fs *c = container_of(j, struct bch_fs, journal);
+ bool stuck = false;
+ struct printbuf buf = PRINTBUF;
+
+ if (!(error == JOURNAL_ERR_journal_full ||
+ error == JOURNAL_ERR_journal_pin_full) ||
+ nr_unwritten_journal_entries(j) ||
+ (flags & BCH_WATERMARK_MASK) != BCH_WATERMARK_reclaim)
+ return stuck;
+
+ spin_lock(&j->lock);
+
+ if (j->can_discard) {
+ spin_unlock(&j->lock);
+ return stuck;
+ }
+
+ stuck = true;
-void __bch2_journal_buf_put(struct journal *j)
+ /*
+ * The journal shutdown path will set ->err_seq, but do it here first to
+ * serialize against concurrent failures and avoid duplicate error
+ * reports.
+ */
+ if (j->err_seq) {
+ spin_unlock(&j->lock);
+ return stuck;
+ }
+ j->err_seq = journal_cur_seq(j);
+ spin_unlock(&j->lock);
+
+ bch_err(c, "Journal stuck! Hava a pre-reservation but journal full (error %s)",
+ bch2_journal_errors[error]);
+ bch2_journal_debug_to_text(&buf, j);
+ bch_err(c, "%s", buf.buf);
+
+ printbuf_reset(&buf);
+ bch2_journal_pins_to_text(&buf, j);
+ bch_err(c, "Journal pins:\n%s", buf.buf);
+ printbuf_exit(&buf);
+
+ bch2_fatal_error(c);
+ dump_stack();
+
+ return stuck;
+}
+
+/*
+ * Final processing when the last reference of a journal buffer has been
+ * dropped. Drop the pin list reference acquired at journal entry open and write
+ * the buffer, if requested.
+ */
+void bch2_journal_buf_put_final(struct journal *j, u64 seq, bool write)
{
struct bch_fs *c = container_of(j, struct bch_fs, journal);
- closure_call(&j->io, bch2_journal_write, c->io_complete_wq, NULL);
+ lockdep_assert_held(&j->lock);
+
+ if (__bch2_journal_pin_put(j, seq))
+ bch2_journal_reclaim_fast(j);
+ if (write)
+ closure_call(&j->io, bch2_journal_write, c->io_complete_wq, NULL);
}
/*
/* Close out old buffer: */
buf->data->u64s = cpu_to_le32(old.cur_entry_offset);
+ trace_journal_entry_close(c, vstruct_bytes(buf->data));
+
sectors = vstruct_blocks_plus(buf->data, c->block_bits,
buf->u64s_reserved) << c->block_bits;
BUG_ON(sectors > buf->sectors);
buf->data->last_seq = cpu_to_le64(buf->last_seq);
BUG_ON(buf->last_seq > le64_to_cpu(buf->data->seq));
- __bch2_journal_pin_put(j, le64_to_cpu(buf->data->seq));
-
cancel_delayed_work(&j->write_work);
bch2_journal_space_available(j);
- bch2_journal_buf_put(j, old.idx);
+ __bch2_journal_buf_put(j, old.idx, le64_to_cpu(buf->data->seq));
}
void bch2_journal_halt(struct journal *j)
__journal_entry_close(j, JOURNAL_ENTRY_ERROR_VAL);
if (!j->err_seq)
j->err_seq = journal_cur_seq(j);
+ journal_wake(j);
spin_unlock(&j->lock);
}
} while ((v = atomic64_cmpxchg(&j->reservations.counter,
old.v, new.v)) != old.v);
- if (j->res_get_blocked_start)
- bch2_time_stats_update(j->blocked_time,
- j->res_get_blocked_start);
- j->res_get_blocked_start = 0;
-
mod_delayed_work(c->io_complete_wq,
&j->write_work,
msecs_to_jiffies(c->opts.journal_flush_delay));
spin_lock(&j->lock);
+ /* check once more in case somebody else shut things down... */
+ if (bch2_journal_error(j)) {
+ spin_unlock(&j->lock);
+ return -BCH_ERR_erofs_journal_err;
+ }
+
/*
* Recheck after taking the lock, so we don't race with another thread
* that just did journal_entry_open() and call journal_entry_close()
return 0;
}
- if ((flags & JOURNAL_WATERMARK_MASK) < j->watermark) {
+ if ((flags & BCH_WATERMARK_MASK) < j->watermark) {
/*
* Don't want to close current journal entry, just need to
* invoke reclaim:
__journal_entry_close(j, JOURNAL_ENTRY_CLOSED_VAL);
ret = journal_entry_open(j);
- if (ret == JOURNAL_ERR_max_in_flight)
+ if (ret == JOURNAL_ERR_max_in_flight) {
+ track_event_change(&c->times[BCH_TIME_blocked_journal_max_in_flight],
+ &j->max_in_flight_start, true);
trace_and_count(c, journal_entry_full, c);
-unlock:
- if ((ret && ret != JOURNAL_ERR_insufficient_devices) &&
- !j->res_get_blocked_start) {
- j->res_get_blocked_start = local_clock() ?: 1;
- trace_and_count(c, journal_full, c);
}
-
+unlock:
can_discard = j->can_discard;
spin_unlock(&j->lock);
if (!ret)
goto retry;
-
- if ((ret == JOURNAL_ERR_journal_full ||
- ret == JOURNAL_ERR_journal_pin_full) &&
- !can_discard &&
- !nr_unwritten_journal_entries(j) &&
- (flags & JOURNAL_WATERMARK_MASK) == JOURNAL_WATERMARK_reserved) {
- struct printbuf buf = PRINTBUF;
-
- bch_err(c, "Journal stuck! Hava a pre-reservation but journal full (ret %s)",
- bch2_journal_errors[ret]);
-
- bch2_journal_debug_to_text(&buf, j);
- bch_err(c, "%s", buf.buf);
-
- printbuf_reset(&buf);
- bch2_journal_pins_to_text(&buf, j);
- bch_err(c, "Journal pins:\n%s", buf.buf);
-
- printbuf_exit(&buf);
- bch2_fatal_error(c);
- dump_stack();
- }
+ if (journal_error_check_stuck(j, ret, flags))
+ ret = -BCH_ERR_journal_res_get_blocked;
/*
* Journal is full - can't rely on reclaim from work item due to
}
return ret == JOURNAL_ERR_insufficient_devices
- ? -EROFS
+ ? -BCH_ERR_erofs_journal_err
: -BCH_ERR_journal_res_get_blocked;
}
int ret;
closure_wait_event(&j->async_wait,
- (ret = __journal_res_get(j, res, flags)) !=
- -BCH_ERR_journal_res_get_blocked||
+ (ret = __journal_res_get(j, res, flags)) != -BCH_ERR_journal_res_get_blocked ||
(flags & JOURNAL_RES_GET_NONBLOCK));
return ret;
}
-/* journal_preres: */
-
-static bool journal_preres_available(struct journal *j,
- struct journal_preres *res,
- unsigned new_u64s,
- unsigned flags)
-{
- bool ret = bch2_journal_preres_get_fast(j, res, new_u64s, flags, true);
-
- if (!ret && mutex_trylock(&j->reclaim_lock)) {
- bch2_journal_reclaim(j);
- mutex_unlock(&j->reclaim_lock);
- }
-
- return ret;
-}
-
-int __bch2_journal_preres_get(struct journal *j,
- struct journal_preres *res,
- unsigned new_u64s,
- unsigned flags)
-{
- int ret;
-
- closure_wait_event(&j->preres_wait,
- (ret = bch2_journal_error(j)) ||
- journal_preres_available(j, res, new_u64s, flags));
- return ret;
-}
-
/* journal_entry_res: */
void bch2_journal_entry_res_resize(struct journal *j,
/**
* bch2_journal_flush_seq_async - wait for a journal entry to be written
+ * @j: journal object
+ * @seq: seq to flush
+ * @parent: closure object to wait with
+ * Returns: 1 if @seq has already been flushed, 0 if @seq is being flushed,
+ * -EIO if @seq will never be flushed
*
- * like bch2_journal_wait_on_seq, except that it triggers a write immediately if
+ * Like bch2_journal_wait_on_seq, except that it triggers a write immediately if
* necessary
*/
int bch2_journal_flush_seq_async(struct journal *j, u64 seq,
break;
}
} else {
- ob[nr_got] = bch2_bucket_alloc(c, ca, RESERVE_none, cl);
+ ob[nr_got] = bch2_bucket_alloc(c, ca, BCH_WATERMARK_normal, cl);
ret = PTR_ERR_OR_ZERO(ob[nr_got]);
if (ret)
break;
ret = bch2_trans_run(c,
- bch2_trans_mark_metadata_bucket(&trans, ca,
+ bch2_trans_mark_metadata_bucket(trans, ca,
ob[nr_got]->bucket, BCH_DATA_journal,
ca->mi.bucket_size));
if (ret) {
bch2_open_bucket_put(c, ob[nr_got]);
- bch_err(c, "error marking new journal buckets: %s", bch2_err_str(ret));
+ bch_err_msg(c, ret, "marking new journal buckets");
break;
}
if (ret && !new_fs)
for (i = 0; i < nr_got; i++)
bch2_trans_run(c,
- bch2_trans_mark_metadata_bucket(&trans, ca,
+ bch2_trans_mark_metadata_bucket(trans, ca,
bu[i], BCH_DATA_free, 0));
err_free:
if (!new_fs)
goto unlock;
while (ja->nr < nr) {
- struct disk_reservation disk_res = { 0, 0 };
+ struct disk_reservation disk_res = { 0, 0, 0 };
/*
* note: journal buckets aren't really counted as _sectors_ used yet, so
}
if (ret)
- bch_err(c, "%s: err %s", __func__, bch2_err_str(ret));
+ bch_err_fn(c, ret);
unlock:
up_write(&c->state_lock);
return ret;
int bch2_dev_journal_alloc(struct bch_dev *ca)
{
unsigned nr;
+ int ret;
- if (dynamic_fault("bcachefs:add:journal_alloc"))
- return -BCH_ERR_ENOMEM_set_nr_journal_buckets;
+ if (dynamic_fault("bcachefs:add:journal_alloc")) {
+ ret = -BCH_ERR_ENOMEM_set_nr_journal_buckets;
+ goto err;
+ }
/* 1/128th of the device by default: */
nr = ca->mi.nbuckets >> 7;
min(1 << 13,
(1 << 24) / ca->mi.bucket_size));
- return __bch2_set_nr_journal_buckets(ca, nr, true, NULL);
+ ret = __bch2_set_nr_journal_buckets(ca, nr, true, NULL);
+err:
+ if (ret)
+ bch_err_fn(ca, ret);
+ return ret;
+}
+
+int bch2_fs_journal_alloc(struct bch_fs *c)
+{
+ struct bch_dev *ca;
+ unsigned i;
+
+ for_each_online_member(ca, c, i) {
+ if (ca->journal.nr)
+ continue;
+
+ int ret = bch2_dev_journal_alloc(ca);
+ if (ret) {
+ percpu_ref_put(&ca->io_ref);
+ return ret;
+ }
+ }
+
+ return 0;
}
/* startup/shutdown: */
{
struct journal_device *ja = &ca->journal;
struct bch_sb_field_journal *journal_buckets =
- bch2_sb_get_journal(sb);
+ bch2_sb_field_get(sb, journal);
struct bch_sb_field_journal_v2 *journal_buckets_v2 =
- bch2_sb_get_journal_v2(sb);
+ bch2_sb_field_get(sb, journal_v2);
unsigned i, nr_bvecs;
ja->nr = 0;
int bch2_fs_journal_init(struct journal *j)
{
- struct bch_fs *c = container_of(j, struct bch_fs, journal);
static struct lock_class_key res_key;
unsigned i;
- int ret = 0;
-
- pr_verbose_init(c->opts, "");
spin_lock_init(&j->lock);
spin_lock_init(&j->err_lock);
((union journal_res_state)
{ .cur_entry_offset = JOURNAL_ENTRY_CLOSED_VAL }).v);
- if (!(init_fifo(&j->pin, JOURNAL_PIN, GFP_KERNEL))) {
- ret = -BCH_ERR_ENOMEM_journal_pin_fifo;
- goto out;
- }
+ if (!(init_fifo(&j->pin, JOURNAL_PIN, GFP_KERNEL)))
+ return -BCH_ERR_ENOMEM_journal_pin_fifo;
for (i = 0; i < ARRAY_SIZE(j->buf); i++) {
j->buf[i].buf_size = JOURNAL_ENTRY_SIZE_MIN;
j->buf[i].data = kvpmalloc(j->buf[i].buf_size, GFP_KERNEL);
- if (!j->buf[i].data) {
- ret = -BCH_ERR_ENOMEM_journal_buf;
- goto out;
- }
+ if (!j->buf[i].data)
+ return -BCH_ERR_ENOMEM_journal_buf;
}
j->pin.front = j->pin.back = 1;
-out:
- pr_verbose_init(c->opts, "ret %i", ret);
- return ret;
+ return 0;
}
/* debug: */
union journal_res_state s;
struct bch_dev *ca;
unsigned long now = jiffies;
+ u64 nr_writes = j->nr_flush_writes + j->nr_noflush_writes;
u64 seq;
unsigned i;
prt_printf(out, "dirty journal entries:\t%llu/%llu\n", fifo_used(&j->pin), j->pin.size);
prt_printf(out, "seq:\t\t\t%llu\n", journal_cur_seq(j));
prt_printf(out, "seq_ondisk:\t\t%llu\n", j->seq_ondisk);
- prt_printf(out, "last_seq:\t\t%llu\n", journal_last_seq(j));
+ prt_printf(out, "last_seq:\t\t%llu\n", journal_last_seq(j));
prt_printf(out, "last_seq_ondisk:\t%llu\n", j->last_seq_ondisk);
- prt_printf(out, "flushed_seq_ondisk:\t%llu\n", j->flushed_seq_ondisk);
- prt_printf(out, "prereserved:\t\t%u/%u\n", j->prereserved.reserved, j->prereserved.remaining);
- prt_printf(out, "watermark:\t\t%s\n", bch2_journal_watermarks[j->watermark]);
- prt_printf(out, "each entry reserved:\t%u\n", j->entry_u64s_reserved);
+ prt_printf(out, "flushed_seq_ondisk:\t%llu\n", j->flushed_seq_ondisk);
+ prt_printf(out, "watermark:\t\t%s\n", bch2_watermarks[j->watermark]);
+ prt_printf(out, "each entry reserved:\t%u\n", j->entry_u64s_reserved);
prt_printf(out, "nr flush writes:\t%llu\n", j->nr_flush_writes);
- prt_printf(out, "nr noflush writes:\t%llu\n", j->nr_noflush_writes);
- prt_printf(out, "nr direct reclaim:\t%llu\n", j->nr_direct_reclaim);
+ prt_printf(out, "nr noflush writes:\t%llu\n", j->nr_noflush_writes);
+ prt_printf(out, "average write size:\t");
+ prt_human_readable_u64(out, nr_writes ? div64_u64(j->entry_bytes_written, nr_writes) : 0);
+ prt_newline(out);
+ prt_printf(out, "nr direct reclaim:\t%llu\n", j->nr_direct_reclaim);
prt_printf(out, "nr background reclaim:\t%llu\n", j->nr_background_reclaim);
prt_printf(out, "reclaim kicked:\t\t%u\n", j->reclaim_kicked);
- prt_printf(out, "reclaim runs in:\t%u ms\n", time_after(j->next_reclaim, now)
+ prt_printf(out, "reclaim runs in:\t%u ms\n", time_after(j->next_reclaim, now)
? jiffies_to_msecs(j->next_reclaim - jiffies) : 0);
- prt_printf(out, "current entry sectors:\t%u\n", j->cur_entry_sectors);
- prt_printf(out, "current entry error:\t%s\n", bch2_journal_errors[j->cur_entry_error]);
+ prt_printf(out, "current entry sectors:\t%u\n", j->cur_entry_sectors);
+ prt_printf(out, "current entry error:\t%s\n", bch2_journal_errors[j->cur_entry_error]);
prt_printf(out, "current entry:\t\t");
switch (s.cur_entry_offset) {