}
}
-static void bch2_dev_usage_update(struct bch_fs *c, struct bch_dev *ca,
- struct bch_alloc_v4 old,
- struct bch_alloc_v4 new,
- u64 journal_seq, bool gc)
+void bch2_dev_usage_update(struct bch_fs *c, struct bch_dev *ca,
+ const struct bch_alloc_v4 *old,
+ const struct bch_alloc_v4 *new,
+ u64 journal_seq, bool gc)
{
struct bch_fs_usage *fs_usage;
struct bch_dev_usage *u;
preempt_disable();
fs_usage = fs_usage_ptr(c, journal_seq, gc);
- if (data_type_is_hidden(old.data_type))
+ if (data_type_is_hidden(old->data_type))
fs_usage->hidden -= ca->mi.bucket_size;
- if (data_type_is_hidden(new.data_type))
+ if (data_type_is_hidden(new->data_type))
fs_usage->hidden += ca->mi.bucket_size;
u = dev_usage_ptr(ca, journal_seq, gc);
- u->d[old.data_type].buckets--;
- u->d[new.data_type].buckets++;
+ u->d[old->data_type].buckets--;
+ u->d[new->data_type].buckets++;
- u->d[old.data_type].sectors -= bch2_bucket_sectors_dirty(old);
- u->d[new.data_type].sectors += bch2_bucket_sectors_dirty(new);
+ u->d[old->data_type].sectors -= bch2_bucket_sectors_dirty(*old);
+ u->d[new->data_type].sectors += bch2_bucket_sectors_dirty(*new);
- u->d[BCH_DATA_cached].sectors += new.cached_sectors;
- u->d[BCH_DATA_cached].sectors -= old.cached_sectors;
+ u->d[BCH_DATA_cached].sectors += new->cached_sectors;
+ u->d[BCH_DATA_cached].sectors -= old->cached_sectors;
- u->d[old.data_type].fragmented -= bch2_bucket_sectors_fragmented(ca, old);
- u->d[new.data_type].fragmented += bch2_bucket_sectors_fragmented(ca, new);
+ u->d[old->data_type].fragmented -= bch2_bucket_sectors_fragmented(ca, *old);
+ u->d[new->data_type].fragmented += bch2_bucket_sectors_fragmented(ca, *new);
preempt_enable();
}
};
}
-static void bch2_dev_usage_update_m(struct bch_fs *c, struct bch_dev *ca,
- struct bucket old, struct bucket new)
+void bch2_dev_usage_update_m(struct bch_fs *c, struct bch_dev *ca,
+ struct bucket *old, struct bucket *new)
{
- bch2_dev_usage_update(c, ca,
- bucket_m_to_alloc(old),
- bucket_m_to_alloc(new),
- 0, true);
+ struct bch_alloc_v4 old_a = bucket_m_to_alloc(*old);
+ struct bch_alloc_v4 new_a = bucket_m_to_alloc(*new);
+
+ bch2_dev_usage_update(c, ca, &old_a, &new_a, 0, true);
}
static inline int __update_replicas(struct bch_fs *c,
return 0;
}
-static inline int update_replicas(struct bch_fs *c, struct bkey_s_c k,
- struct bch_replicas_entry_v1 *r, s64 sectors,
- unsigned journal_seq, bool gc)
+int bch2_update_replicas(struct bch_fs *c, struct bkey_s_c k,
+ struct bch_replicas_entry_v1 *r, s64 sectors,
+ unsigned journal_seq, bool gc)
{
struct bch_fs_usage *fs_usage;
int idx, ret = 0;
bch2_replicas_entry_cached(&r.e, dev);
- return update_replicas(c, k, &r.e, sectors, journal_seq, gc);
+ return bch2_update_replicas(c, k, &r.e, sectors, journal_seq, gc);
}
static int __replicas_deltas_realloc(struct btree_trans *trans, unsigned more,
return bch2_update_replicas_list(trans, &r.e, sectors);
}
-int bch2_mark_alloc(struct btree_trans *trans,
- enum btree_id btree, unsigned level,
- struct bkey_s_c old, struct bkey_s_c new,
- unsigned flags)
-{
- bool gc = flags & BTREE_TRIGGER_GC;
- u64 journal_seq = trans->journal_res.seq;
- u64 bucket_journal_seq;
- struct bch_fs *c = trans->c;
- struct bch_alloc_v4 old_a_convert, new_a_convert;
- const struct bch_alloc_v4 *old_a, *new_a;
- struct bch_dev *ca;
- int ret = 0;
-
- /*
- * alloc btree is read in by bch2_alloc_read, not gc:
- */
- if ((flags & BTREE_TRIGGER_GC) &&
- !(flags & BTREE_TRIGGER_BUCKET_INVALIDATE))
- return 0;
-
- if (bch2_trans_inconsistent_on(!bch2_dev_bucket_exists(c, new.k->p), trans,
- "alloc key for invalid device or bucket"))
- return -EIO;
-
- ca = bch_dev_bkey_exists(c, new.k->p.inode);
-
- old_a = bch2_alloc_to_v4(old, &old_a_convert);
- new_a = bch2_alloc_to_v4(new, &new_a_convert);
-
- bucket_journal_seq = new_a->journal_seq;
-
- if ((flags & BTREE_TRIGGER_INSERT) &&
- data_type_is_empty(old_a->data_type) !=
- data_type_is_empty(new_a->data_type) &&
- new.k->type == KEY_TYPE_alloc_v4) {
- struct bch_alloc_v4 *v = (struct bch_alloc_v4 *) new.v;
-
- EBUG_ON(!journal_seq);
-
- /*
- * If the btree updates referring to a bucket weren't flushed
- * before the bucket became empty again, then the we don't have
- * to wait on a journal flush before we can reuse the bucket:
- */
- v->journal_seq = bucket_journal_seq =
- data_type_is_empty(new_a->data_type) &&
- (journal_seq == v->journal_seq ||
- bch2_journal_noflush_seq(&c->journal, v->journal_seq))
- ? 0 : journal_seq;
- }
-
- if (!data_type_is_empty(old_a->data_type) &&
- data_type_is_empty(new_a->data_type) &&
- bucket_journal_seq) {
- ret = bch2_set_bucket_needs_journal_commit(&c->buckets_waiting_for_journal,
- c->journal.flushed_seq_ondisk,
- new.k->p.inode, new.k->p.offset,
- bucket_journal_seq);
- if (ret) {
- bch2_fs_fatal_error(c,
- "error setting bucket_needs_journal_commit: %i", ret);
- return ret;
- }
- }
-
- percpu_down_read(&c->mark_lock);
- if (!gc && new_a->gen != old_a->gen)
- *bucket_gen(ca, new.k->p.offset) = new_a->gen;
-
- bch2_dev_usage_update(c, ca, *old_a, *new_a, journal_seq, gc);
-
- if (gc) {
- struct bucket *g = gc_bucket(ca, new.k->p.offset);
-
- bucket_lock(g);
-
- g->gen_valid = 1;
- g->gen = new_a->gen;
- g->data_type = new_a->data_type;
- g->stripe = new_a->stripe;
- g->stripe_redundancy = new_a->stripe_redundancy;
- g->dirty_sectors = new_a->dirty_sectors;
- g->cached_sectors = new_a->cached_sectors;
-
- bucket_unlock(g);
- }
- percpu_up_read(&c->mark_lock);
-
- if (new_a->data_type == BCH_DATA_free &&
- (!new_a->journal_seq || new_a->journal_seq < c->journal.flushed_seq_ondisk))
- closure_wake_up(&c->freelist_wait);
-
- if (new_a->data_type == BCH_DATA_need_discard &&
- (!bucket_journal_seq || bucket_journal_seq < c->journal.flushed_seq_ondisk))
- bch2_do_discards(c);
-
- if (old_a->data_type != BCH_DATA_cached &&
- new_a->data_type == BCH_DATA_cached &&
- should_invalidate_buckets(ca, bch2_dev_usage_read(ca)))
- bch2_do_invalidates(c);
-
- if (new_a->data_type == BCH_DATA_need_gc_gens)
- bch2_do_gc_gens(c);
-
- return 0;
-}
-
int bch2_mark_metadata_bucket(struct bch_fs *c, struct bch_dev *ca,
size_t b, enum bch_data_type data_type,
unsigned sectors, struct gc_pos pos,
err:
bucket_unlock(g);
if (!ret)
- bch2_dev_usage_update_m(c, ca, old, new);
+ bch2_dev_usage_update_m(c, ca, &old, &new);
percpu_up_read(&c->mark_lock);
return ret;
}
-static int check_bucket_ref(struct btree_trans *trans,
- struct bkey_s_c k,
- const struct bch_extent_ptr *ptr,
- s64 sectors, enum bch_data_type ptr_data_type,
- u8 b_gen, u8 bucket_data_type,
- u32 bucket_sectors)
+int bch2_check_bucket_ref(struct btree_trans *trans,
+ struct bkey_s_c k,
+ const struct bch_extent_ptr *ptr,
+ s64 sectors, enum bch_data_type ptr_data_type,
+ u8 b_gen, u8 bucket_data_type,
+ u32 bucket_sectors)
{
struct bch_fs *c = trans->c;
struct bch_dev *ca = bch_dev_bkey_exists(c, ptr->dev);
goto out;
}
-static int mark_stripe_bucket(struct btree_trans *trans,
- struct bkey_s_c k,
- unsigned ptr_idx,
- unsigned flags)
-{
- struct bch_fs *c = trans->c;
- const struct bch_stripe *s = bkey_s_c_to_stripe(k).v;
- unsigned nr_data = s->nr_blocks - s->nr_redundant;
- bool parity = ptr_idx >= nr_data;
- enum bch_data_type data_type = parity ? BCH_DATA_parity : BCH_DATA_stripe;
- s64 sectors = parity ? le16_to_cpu(s->sectors) : 0;
- const struct bch_extent_ptr *ptr = s->ptrs + ptr_idx;
- struct bch_dev *ca = bch_dev_bkey_exists(c, ptr->dev);
- struct bucket old, new, *g;
- struct printbuf buf = PRINTBUF;
- int ret = 0;
-
- BUG_ON(!(flags & BTREE_TRIGGER_GC));
-
- /* * XXX doesn't handle deletion */
-
- percpu_down_read(&c->mark_lock);
- g = PTR_GC_BUCKET(ca, ptr);
-
- if (g->dirty_sectors ||
- (g->stripe && g->stripe != k.k->p.offset)) {
- bch2_fs_inconsistent(c,
- "bucket %u:%zu gen %u: multiple stripes using same bucket\n%s",
- ptr->dev, PTR_BUCKET_NR(ca, ptr), g->gen,
- (bch2_bkey_val_to_text(&buf, c, k), buf.buf));
- ret = -EINVAL;
- goto err;
- }
-
- bucket_lock(g);
- old = *g;
-
- ret = check_bucket_ref(trans, k, ptr, sectors, data_type,
- g->gen, g->data_type,
- g->dirty_sectors);
- if (ret)
- goto err;
-
- g->data_type = data_type;
- g->dirty_sectors += sectors;
-
- g->stripe = k.k->p.offset;
- g->stripe_redundancy = s->nr_redundant;
- new = *g;
-err:
- bucket_unlock(g);
- if (!ret)
- bch2_dev_usage_update_m(c, ca, old, new);
- percpu_up_read(&c->mark_lock);
- printbuf_exit(&buf);
- return ret;
-}
-
-static int __mark_pointer(struct btree_trans *trans,
- struct bkey_s_c k,
- const struct bch_extent_ptr *ptr,
- s64 sectors, enum bch_data_type ptr_data_type,
- u8 bucket_gen, u8 *bucket_data_type,
- u32 *dirty_sectors, u32 *cached_sectors)
-{
- u32 *dst_sectors = !ptr->cached
- ? dirty_sectors
- : cached_sectors;
- int ret = check_bucket_ref(trans, k, ptr, sectors, ptr_data_type,
- bucket_gen, *bucket_data_type, *dst_sectors);
-
- if (ret)
- return ret;
-
- *dst_sectors += sectors;
-
- if (!*dirty_sectors && !*cached_sectors)
- *bucket_data_type = 0;
- else if (*bucket_data_type != BCH_DATA_stripe)
- *bucket_data_type = ptr_data_type;
-
- return 0;
-}
-
-static int bch2_mark_pointer(struct btree_trans *trans,
- enum btree_id btree_id, unsigned level,
- struct bkey_s_c k,
- struct extent_ptr_decoded p,
- s64 sectors,
- unsigned flags)
-{
- struct bch_fs *c = trans->c;
- struct bch_dev *ca = bch_dev_bkey_exists(c, p.ptr.dev);
- struct bucket old, new, *g;
- enum bch_data_type data_type = bkey_ptr_data_type(btree_id, level, k, p);
- u8 bucket_data_type;
- int ret = 0;
-
- BUG_ON(!(flags & BTREE_TRIGGER_GC));
-
- percpu_down_read(&c->mark_lock);
- g = PTR_GC_BUCKET(ca, &p.ptr);
- bucket_lock(g);
- old = *g;
-
- bucket_data_type = g->data_type;
- ret = __mark_pointer(trans, k, &p.ptr, sectors,
- data_type, g->gen,
- &bucket_data_type,
- &g->dirty_sectors,
- &g->cached_sectors);
- if (!ret)
- g->data_type = bucket_data_type;
-
- new = *g;
- bucket_unlock(g);
- if (!ret)
- bch2_dev_usage_update_m(c, ca, old, new);
- percpu_up_read(&c->mark_lock);
-
- return ret;
-}
-
-static int bch2_mark_stripe_ptr(struct btree_trans *trans,
- struct bkey_s_c k,
- struct bch_extent_stripe_ptr p,
- enum bch_data_type data_type,
- s64 sectors,
- unsigned flags)
-{
- struct bch_fs *c = trans->c;
- struct bch_replicas_padded r;
- struct gc_stripe *m;
-
- BUG_ON(!(flags & BTREE_TRIGGER_GC));
-
- m = genradix_ptr_alloc(&c->gc_stripes, p.idx, GFP_KERNEL);
- if (!m) {
- bch_err(c, "error allocating memory for gc_stripes, idx %llu",
- (u64) p.idx);
- return -BCH_ERR_ENOMEM_mark_stripe_ptr;
- }
-
- mutex_lock(&c->ec_stripes_heap_lock);
-
- if (!m || !m->alive) {
- mutex_unlock(&c->ec_stripes_heap_lock);
- bch_err_ratelimited(c, "pointer to nonexistent stripe %llu",
- (u64) p.idx);
- bch2_inconsistent_error(c);
- return -EIO;
- }
-
- m->block_sectors[p.block] += sectors;
-
- r = m->r;
- mutex_unlock(&c->ec_stripes_heap_lock);
-
- r.e.data_type = data_type;
- update_replicas(c, k, &r.e, sectors, trans->journal_res.seq, true);
-
- return 0;
-}
-
-static int __mark_extent(struct btree_trans *trans,
- enum btree_id btree_id, unsigned level,
- struct bkey_s_c k, unsigned flags)
-{
- u64 journal_seq = trans->journal_res.seq;
- struct bch_fs *c = trans->c;
- struct bkey_ptrs_c ptrs = bch2_bkey_ptrs_c(k);
- const union bch_extent_entry *entry;
- struct extent_ptr_decoded p;
- struct bch_replicas_padded r;
- enum bch_data_type data_type = bkey_is_btree_ptr(k.k)
- ? BCH_DATA_btree
- : BCH_DATA_user;
- s64 sectors = bkey_is_btree_ptr(k.k)
- ? btree_sectors(c)
- : k.k->size;
- s64 dirty_sectors = 0;
- bool stale;
- int ret;
-
- BUG_ON(!(flags & BTREE_TRIGGER_GC));
-
- r.e.data_type = data_type;
- r.e.nr_devs = 0;
- r.e.nr_required = 1;
-
- bkey_for_each_ptr_decode(k.k, ptrs, p, entry) {
- s64 disk_sectors = ptr_disk_sectors(sectors, p);
-
- if (flags & BTREE_TRIGGER_OVERWRITE)
- disk_sectors = -disk_sectors;
-
- ret = bch2_mark_pointer(trans, btree_id, level, k, p, disk_sectors, flags);
- if (ret < 0)
- return ret;
-
- stale = ret > 0;
-
- if (p.ptr.cached) {
- if (!stale) {
- ret = update_cached_sectors(c, k, p.ptr.dev,
- disk_sectors, journal_seq, true);
- if (ret) {
- bch2_fs_fatal_error(c, "%s(): no replicas entry while updating cached sectors",
- __func__);
- return ret;
- }
- }
- } else if (!p.has_ec) {
- dirty_sectors += disk_sectors;
- r.e.devs[r.e.nr_devs++] = p.ptr.dev;
- } else {
- ret = bch2_mark_stripe_ptr(trans, k, p.ec, data_type,
- disk_sectors, flags);
- if (ret)
- return ret;
-
- /*
- * There may be other dirty pointers in this extent, but
- * if so they're not required for mounting if we have an
- * erasure coded pointer in this extent:
- */
- r.e.nr_required = 0;
- }
- }
-
- if (r.e.nr_devs) {
- ret = update_replicas(c, k, &r.e, dirty_sectors, journal_seq, true);
- if (ret) {
- struct printbuf buf = PRINTBUF;
-
- bch2_bkey_val_to_text(&buf, c, k);
- bch2_fs_fatal_error(c, "%s(): no replicas entry for %s", __func__, buf.buf);
- printbuf_exit(&buf);
- return ret;
- }
- }
-
- return 0;
-}
-
-int bch2_mark_extent(struct btree_trans *trans,
- enum btree_id btree_id, unsigned level,
- struct bkey_s_c old, struct bkey_s_c new,
- unsigned flags)
-{
- return mem_trigger_run_overwrite_then_insert(__mark_extent, trans, btree_id, level, old, new, flags);
-}
-
-int bch2_mark_stripe(struct btree_trans *trans,
- enum btree_id btree_id, unsigned level,
- struct bkey_s_c old, struct bkey_s_c new,
- unsigned flags)
-{
- bool gc = flags & BTREE_TRIGGER_GC;
- u64 journal_seq = trans->journal_res.seq;
- struct bch_fs *c = trans->c;
- u64 idx = new.k->p.offset;
- const struct bch_stripe *old_s = old.k->type == KEY_TYPE_stripe
- ? bkey_s_c_to_stripe(old).v : NULL;
- const struct bch_stripe *new_s = new.k->type == KEY_TYPE_stripe
- ? bkey_s_c_to_stripe(new).v : NULL;
- unsigned i;
- int ret;
-
- BUG_ON(gc && old_s);
-
- if (!gc) {
- struct stripe *m = genradix_ptr(&c->stripes, idx);
-
- if (!m) {
- struct printbuf buf1 = PRINTBUF;
- struct printbuf buf2 = PRINTBUF;
-
- bch2_bkey_val_to_text(&buf1, c, old);
- bch2_bkey_val_to_text(&buf2, c, new);
- bch_err_ratelimited(c, "error marking nonexistent stripe %llu while marking\n"
- "old %s\n"
- "new %s", idx, buf1.buf, buf2.buf);
- printbuf_exit(&buf2);
- printbuf_exit(&buf1);
- bch2_inconsistent_error(c);
- return -1;
- }
-
- if (!new_s) {
- bch2_stripes_heap_del(c, m, idx);
-
- memset(m, 0, sizeof(*m));
- } else {
- m->sectors = le16_to_cpu(new_s->sectors);
- m->algorithm = new_s->algorithm;
- m->nr_blocks = new_s->nr_blocks;
- m->nr_redundant = new_s->nr_redundant;
- m->blocks_nonempty = 0;
-
- for (i = 0; i < new_s->nr_blocks; i++)
- m->blocks_nonempty += !!stripe_blockcount_get(new_s, i);
-
- if (!old_s)
- bch2_stripes_heap_insert(c, m, idx);
- else
- bch2_stripes_heap_update(c, m, idx);
- }
- } else {
- struct gc_stripe *m =
- genradix_ptr_alloc(&c->gc_stripes, idx, GFP_KERNEL);
-
- if (!m) {
- bch_err(c, "error allocating memory for gc_stripes, idx %llu",
- idx);
- return -BCH_ERR_ENOMEM_mark_stripe;
- }
- /*
- * This will be wrong when we bring back runtime gc: we should
- * be unmarking the old key and then marking the new key
- */
- m->alive = true;
- m->sectors = le16_to_cpu(new_s->sectors);
- m->nr_blocks = new_s->nr_blocks;
- m->nr_redundant = new_s->nr_redundant;
-
- for (i = 0; i < new_s->nr_blocks; i++)
- m->ptrs[i] = new_s->ptrs[i];
-
- bch2_bkey_to_replicas(&m->r.e, new);
-
- /*
- * gc recalculates this field from stripe ptr
- * references:
- */
- memset(m->block_sectors, 0, sizeof(m->block_sectors));
-
- for (i = 0; i < new_s->nr_blocks; i++) {
- ret = mark_stripe_bucket(trans, new, i, flags);
- if (ret)
- return ret;
- }
-
- ret = update_replicas(c, new, &m->r.e,
- ((s64) m->sectors * m->nr_redundant),
- journal_seq, gc);
- if (ret) {
- struct printbuf buf = PRINTBUF;
-
- bch2_bkey_val_to_text(&buf, c, new);
- bch2_fs_fatal_error(c, "no replicas entry for %s", buf.buf);
- printbuf_exit(&buf);
- return ret;
- }
- }
-
- return 0;
-}
-
-static int __mark_reservation(struct btree_trans *trans,
- enum btree_id btree_id, unsigned level,
- struct bkey_s_c k, unsigned flags)
-{
- struct bch_fs *c = trans->c;
- struct bch_fs_usage *fs_usage;
- unsigned replicas = bkey_s_c_to_reservation(k).v->nr_replicas;
- s64 sectors = (s64) k.k->size;
-
- BUG_ON(!(flags & BTREE_TRIGGER_GC));
-
- if (flags & BTREE_TRIGGER_OVERWRITE)
- sectors = -sectors;
- sectors *= replicas;
-
- percpu_down_read(&c->mark_lock);
- preempt_disable();
-
- fs_usage = fs_usage_ptr(c, trans->journal_res.seq, flags & BTREE_TRIGGER_GC);
- replicas = clamp_t(unsigned, replicas, 1,
- ARRAY_SIZE(fs_usage->persistent_reserved));
-
- fs_usage->reserved += sectors;
- fs_usage->persistent_reserved[replicas - 1] += sectors;
-
- preempt_enable();
- percpu_up_read(&c->mark_lock);
-
- return 0;
-}
-
-int bch2_mark_reservation(struct btree_trans *trans,
- enum btree_id btree_id, unsigned level,
- struct bkey_s_c old, struct bkey_s_c new,
- unsigned flags)
-{
- return mem_trigger_run_overwrite_then_insert(__mark_reservation, trans, btree_id, level, old, new, flags);
-}
-
void bch2_trans_fs_usage_revert(struct btree_trans *trans,
struct replicas_delta_list *deltas)
{
return -1;
}
-/* trans_mark: */
+/* KEY_TYPE_extent: */
+
+static int __mark_pointer(struct btree_trans *trans,
+ struct bkey_s_c k,
+ const struct bch_extent_ptr *ptr,
+ s64 sectors, enum bch_data_type ptr_data_type,
+ u8 bucket_gen, u8 *bucket_data_type,
+ u32 *dirty_sectors, u32 *cached_sectors)
+{
+ u32 *dst_sectors = !ptr->cached
+ ? dirty_sectors
+ : cached_sectors;
+ int ret = bch2_check_bucket_ref(trans, k, ptr, sectors, ptr_data_type,
+ bucket_gen, *bucket_data_type, *dst_sectors);
+
+ if (ret)
+ return ret;
+
+ *dst_sectors += sectors;
+
+ if (!*dirty_sectors && !*cached_sectors)
+ *bucket_data_type = 0;
+ else if (*bucket_data_type != BCH_DATA_stripe)
+ *bucket_data_type = ptr_data_type;
+
+ return 0;
+}
-static inline int bch2_trans_mark_pointer(struct btree_trans *trans,
- enum btree_id btree_id, unsigned level,
- struct bkey_s_c k, struct extent_ptr_decoded p,
- unsigned flags)
+static int bch2_trigger_pointer(struct btree_trans *trans,
+ enum btree_id btree_id, unsigned level,
+ struct bkey_s_c k, struct extent_ptr_decoded p,
+ s64 *sectors,
+ unsigned flags)
{
bool insert = !(flags & BTREE_TRIGGER_OVERWRITE);
- struct btree_iter iter;
- struct bkey_i_alloc_v4 *a;
struct bpos bucket;
struct bch_backpointer bp;
- s64 sectors;
- int ret;
bch2_extent_ptr_to_bp(trans->c, btree_id, level, k, p, &bucket, &bp);
- sectors = bp.bucket_len;
- if (!insert)
- sectors = -sectors;
-
- a = bch2_trans_start_alloc_update(trans, &iter, bucket);
- if (IS_ERR(a))
- return PTR_ERR(a);
+ *sectors = insert ? bp.bucket_len : -((s64) bp.bucket_len);
- ret = __mark_pointer(trans, k, &p.ptr, sectors, bp.data_type,
- a->v.gen, &a->v.data_type,
- &a->v.dirty_sectors, &a->v.cached_sectors) ?:
- bch2_trans_update(trans, &iter, &a->k_i, 0);
- bch2_trans_iter_exit(trans, &iter);
+ if (flags & BTREE_TRIGGER_TRANSACTIONAL) {
+ struct btree_iter iter;
+ struct bkey_i_alloc_v4 *a = bch2_trans_start_alloc_update(trans, &iter, bucket);
+ int ret = PTR_ERR_OR_ZERO(a);
+ if (ret)
+ return ret;
- if (ret)
- return ret;
+ ret = __mark_pointer(trans, k, &p.ptr, *sectors, bp.data_type,
+ a->v.gen, &a->v.data_type,
+ &a->v.dirty_sectors, &a->v.cached_sectors) ?:
+ bch2_trans_update(trans, &iter, &a->k_i, 0);
+ bch2_trans_iter_exit(trans, &iter);
- if (!p.ptr.cached) {
- ret = bch2_bucket_backpointer_mod(trans, bucket, bp, k, insert);
if (ret)
return ret;
+
+ if (!p.ptr.cached) {
+ ret = bch2_bucket_backpointer_mod(trans, bucket, bp, k, insert);
+ if (ret)
+ return ret;
+ }
+ }
+
+ if (flags & BTREE_TRIGGER_GC) {
+ struct bch_fs *c = trans->c;
+ struct bch_dev *ca = bch_dev_bkey_exists(c, p.ptr.dev);
+ enum bch_data_type data_type = bkey_ptr_data_type(btree_id, level, k, p);
+
+ percpu_down_read(&c->mark_lock);
+ struct bucket *g = PTR_GC_BUCKET(ca, &p.ptr);
+ bucket_lock(g);
+ struct bucket old = *g;
+
+ u8 bucket_data_type = g->data_type;
+ int ret = __mark_pointer(trans, k, &p.ptr, *sectors,
+ data_type, g->gen,
+ &bucket_data_type,
+ &g->dirty_sectors,
+ &g->cached_sectors);
+ if (ret) {
+ bucket_unlock(g);
+ percpu_up_read(&c->mark_lock);
+ return ret;
+ }
+
+ g->data_type = bucket_data_type;
+ struct bucket new = *g;
+ bucket_unlock(g);
+ bch2_dev_usage_update_m(c, ca, &old, &new);
+ percpu_up_read(&c->mark_lock);
}
return 0;
}
-static int bch2_trans_mark_stripe_ptr(struct btree_trans *trans,
- struct extent_ptr_decoded p,
- s64 sectors, enum bch_data_type data_type)
+static int bch2_trigger_stripe_ptr(struct btree_trans *trans,
+ struct bkey_s_c k,
+ struct extent_ptr_decoded p,
+ enum bch_data_type data_type,
+ s64 sectors, unsigned flags)
{
- struct btree_iter iter;
- struct bkey_i_stripe *s;
- struct bch_replicas_padded r;
- int ret = 0;
+ if (flags & BTREE_TRIGGER_TRANSACTIONAL) {
+ struct btree_iter iter;
+ struct bkey_i_stripe *s = bch2_bkey_get_mut_typed(trans, &iter,
+ BTREE_ID_stripes, POS(0, p.ec.idx),
+ BTREE_ITER_WITH_UPDATES, stripe);
+ int ret = PTR_ERR_OR_ZERO(s);
+ if (unlikely(ret)) {
+ bch2_trans_inconsistent_on(bch2_err_matches(ret, ENOENT), trans,
+ "pointer to nonexistent stripe %llu",
+ (u64) p.ec.idx);
+ goto err;
+ }
- s = bch2_bkey_get_mut_typed(trans, &iter,
- BTREE_ID_stripes, POS(0, p.ec.idx),
- BTREE_ITER_WITH_UPDATES, stripe);
- ret = PTR_ERR_OR_ZERO(s);
- if (unlikely(ret)) {
- bch2_trans_inconsistent_on(bch2_err_matches(ret, ENOENT), trans,
- "pointer to nonexistent stripe %llu",
- (u64) p.ec.idx);
- goto err;
- }
+ if (!bch2_ptr_matches_stripe(&s->v, p)) {
+ bch2_trans_inconsistent(trans,
+ "stripe pointer doesn't match stripe %llu",
+ (u64) p.ec.idx);
+ ret = -EIO;
+ goto err;
+ }
- if (!bch2_ptr_matches_stripe(&s->v, p)) {
- bch2_trans_inconsistent(trans,
- "stripe pointer doesn't match stripe %llu",
- (u64) p.ec.idx);
- ret = -EIO;
- goto err;
+ stripe_blockcount_set(&s->v, p.ec.block,
+ stripe_blockcount_get(&s->v, p.ec.block) +
+ sectors);
+
+ struct bch_replicas_padded r;
+ bch2_bkey_to_replicas(&r.e, bkey_i_to_s_c(&s->k_i));
+ r.e.data_type = data_type;
+ ret = bch2_update_replicas_list(trans, &r.e, sectors);
+err:
+ bch2_trans_iter_exit(trans, &iter);
+ return ret;
}
- stripe_blockcount_set(&s->v, p.ec.block,
- stripe_blockcount_get(&s->v, p.ec.block) +
- sectors);
+ if (flags & BTREE_TRIGGER_GC) {
+ struct bch_fs *c = trans->c;
- bch2_bkey_to_replicas(&r.e, bkey_i_to_s_c(&s->k_i));
- r.e.data_type = data_type;
- ret = bch2_update_replicas_list(trans, &r.e, sectors);
-err:
- bch2_trans_iter_exit(trans, &iter);
- return ret;
+ BUG_ON(!(flags & BTREE_TRIGGER_GC));
+
+ struct gc_stripe *m = genradix_ptr_alloc(&c->gc_stripes, p.ec.idx, GFP_KERNEL);
+ if (!m) {
+ bch_err(c, "error allocating memory for gc_stripes, idx %llu",
+ (u64) p.ec.idx);
+ return -BCH_ERR_ENOMEM_mark_stripe_ptr;
+ }
+
+ mutex_lock(&c->ec_stripes_heap_lock);
+
+ if (!m || !m->alive) {
+ mutex_unlock(&c->ec_stripes_heap_lock);
+ struct printbuf buf = PRINTBUF;
+ bch2_bkey_val_to_text(&buf, c, k);
+ bch_err_ratelimited(c, "pointer to nonexistent stripe %llu\n while marking %s",
+ (u64) p.ec.idx, buf.buf);
+ printbuf_exit(&buf);
+ bch2_inconsistent_error(c);
+ return -EIO;
+ }
+
+ m->block_sectors[p.ec.block] += sectors;
+
+ struct bch_replicas_padded r = m->r;
+ mutex_unlock(&c->ec_stripes_heap_lock);
+
+ r.e.data_type = data_type;
+ bch2_update_replicas(c, k, &r.e, sectors, trans->journal_res.seq, true);
+ }
+
+ return 0;
}
-static int __trans_mark_extent(struct btree_trans *trans,
- enum btree_id btree_id, unsigned level,
- struct bkey_s_c k, unsigned flags)
+static int __trigger_extent(struct btree_trans *trans,
+ enum btree_id btree_id, unsigned level,
+ struct bkey_s_c k, unsigned flags)
{
+ bool gc = flags & BTREE_TRIGGER_GC;
struct bch_fs *c = trans->c;
struct bkey_ptrs_c ptrs = bch2_bkey_ptrs_c(k);
const union bch_extent_entry *entry;
enum bch_data_type data_type = bkey_is_btree_ptr(k.k)
? BCH_DATA_btree
: BCH_DATA_user;
- s64 sectors = bkey_is_btree_ptr(k.k)
- ? btree_sectors(c)
- : k.k->size;
s64 dirty_sectors = 0;
- bool stale;
int ret = 0;
r.e.data_type = data_type;
r.e.nr_required = 1;
bkey_for_each_ptr_decode(k.k, ptrs, p, entry) {
- s64 disk_sectors = ptr_disk_sectors(sectors, p);
-
- if (flags & BTREE_TRIGGER_OVERWRITE)
- disk_sectors = -disk_sectors;
-
- ret = bch2_trans_mark_pointer(trans, btree_id, level, k, p, flags);
+ s64 disk_sectors;
+ ret = bch2_trigger_pointer(trans, btree_id, level, k, p, &disk_sectors, flags);
if (ret < 0)
return ret;
- stale = ret > 0;
+ bool stale = ret > 0;
if (p.ptr.cached) {
if (!stale) {
- ret = bch2_update_cached_sectors_list(trans, p.ptr.dev,
- disk_sectors);
+ ret = !gc
+ ? bch2_update_cached_sectors_list(trans, p.ptr.dev, disk_sectors)
+ : update_cached_sectors(c, k, p.ptr.dev, disk_sectors, 0, true);
+ bch2_fs_fatal_err_on(ret && gc, c, "%s(): no replicas entry while updating cached sectors",
+ __func__);
if (ret)
return ret;
}
dirty_sectors += disk_sectors;
r.e.devs[r.e.nr_devs++] = p.ptr.dev;
} else {
- ret = bch2_trans_mark_stripe_ptr(trans, p,
- disk_sectors, data_type);
+ ret = bch2_trigger_stripe_ptr(trans, k, p, data_type, disk_sectors, flags);
if (ret)
return ret;
+ /*
+ * There may be other dirty pointers in this extent, but
+ * if so they're not required for mounting if we have an
+ * erasure coded pointer in this extent:
+ */
r.e.nr_required = 0;
}
}
- if (r.e.nr_devs)
- ret = bch2_update_replicas_list(trans, &r.e, dirty_sectors);
-
- return ret;
-}
-
-int bch2_trans_mark_extent(struct btree_trans *trans,
- enum btree_id btree_id, unsigned level,
- struct bkey_s_c old, struct bkey_i *new,
- unsigned flags)
-{
- struct bch_fs *c = trans->c;
- int mod = (int) bch2_bkey_needs_rebalance(c, bkey_i_to_s_c(new)) -
- (int) bch2_bkey_needs_rebalance(c, old);
+ if (r.e.nr_devs) {
+ ret = !gc
+ ? bch2_update_replicas_list(trans, &r.e, dirty_sectors)
+ : bch2_update_replicas(c, k, &r.e, dirty_sectors, 0, true);
+ if (unlikely(ret && gc)) {
+ struct printbuf buf = PRINTBUF;
- if (mod) {
- int ret = bch2_btree_bit_mod(trans, BTREE_ID_rebalance_work, new->k.p, mod > 0);
+ bch2_bkey_val_to_text(&buf, c, k);
+ bch2_fs_fatal_error(c, "%s(): no replicas entry for %s", __func__, buf.buf);
+ printbuf_exit(&buf);
+ }
if (ret)
return ret;
}
- return trigger_run_overwrite_then_insert(__trans_mark_extent, trans, btree_id, level, old, new, flags);
+ return 0;
}
-static int bch2_trans_mark_stripe_bucket(struct btree_trans *trans,
- struct bkey_s_c_stripe s,
- unsigned idx, bool deleting)
+int bch2_trigger_extent(struct btree_trans *trans,
+ enum btree_id btree_id, unsigned level,
+ struct bkey_s_c old, struct bkey_s new,
+ unsigned flags)
{
- struct bch_fs *c = trans->c;
- const struct bch_extent_ptr *ptr = &s.v->ptrs[idx];
- struct btree_iter iter;
- struct bkey_i_alloc_v4 *a;
- enum bch_data_type data_type = idx >= s.v->nr_blocks - s.v->nr_redundant
- ? BCH_DATA_parity : 0;
- s64 sectors = data_type ? le16_to_cpu(s.v->sectors) : 0;
- int ret = 0;
-
- if (deleting)
- sectors = -sectors;
-
- a = bch2_trans_start_alloc_update(trans, &iter, PTR_BUCKET_POS(c, ptr));
- if (IS_ERR(a))
- return PTR_ERR(a);
-
- ret = check_bucket_ref(trans, s.s_c, ptr, sectors, data_type,
- a->v.gen, a->v.data_type,
- a->v.dirty_sectors);
- if (ret)
- goto err;
-
- if (!deleting) {
- if (bch2_trans_inconsistent_on(a->v.stripe ||
- a->v.stripe_redundancy, trans,
- "bucket %llu:%llu gen %u data type %s dirty_sectors %u: multiple stripes using same bucket (%u, %llu)",
- iter.pos.inode, iter.pos.offset, a->v.gen,
- bch2_data_types[a->v.data_type],
- a->v.dirty_sectors,
- a->v.stripe, s.k->p.offset)) {
- ret = -EIO;
- goto err;
- }
-
- if (bch2_trans_inconsistent_on(data_type && a->v.dirty_sectors, trans,
- "bucket %llu:%llu gen %u data type %s dirty_sectors %u: data already in stripe bucket %llu",
- iter.pos.inode, iter.pos.offset, a->v.gen,
- bch2_data_types[a->v.data_type],
- a->v.dirty_sectors,
- s.k->p.offset)) {
- ret = -EIO;
- goto err;
- }
+ if (flags & BTREE_TRIGGER_TRANSACTIONAL) {
+ struct bch_fs *c = trans->c;
+ int mod = (int) bch2_bkey_needs_rebalance(c, new.s_c) -
+ (int) bch2_bkey_needs_rebalance(c, old);
- a->v.stripe = s.k->p.offset;
- a->v.stripe_redundancy = s.v->nr_redundant;
- a->v.data_type = BCH_DATA_stripe;
- } else {
- if (bch2_trans_inconsistent_on(a->v.stripe != s.k->p.offset ||
- a->v.stripe_redundancy != s.v->nr_redundant, trans,
- "bucket %llu:%llu gen %u: not marked as stripe when deleting stripe %llu (got %u)",
- iter.pos.inode, iter.pos.offset, a->v.gen,
- s.k->p.offset, a->v.stripe)) {
- ret = -EIO;
- goto err;
+ if (mod) {
+ int ret = bch2_btree_bit_mod(trans, BTREE_ID_rebalance_work, new.k->p, mod > 0);
+ if (ret)
+ return ret;
}
-
- a->v.stripe = 0;
- a->v.stripe_redundancy = 0;
- a->v.data_type = alloc_data_type(a->v, BCH_DATA_user);
}
- a->v.dirty_sectors += sectors;
- if (data_type)
- a->v.data_type = !deleting ? data_type : 0;
+ if (flags & (BTREE_TRIGGER_TRANSACTIONAL|BTREE_TRIGGER_GC))
+ return trigger_run_overwrite_then_insert(__trigger_extent, trans, btree_id, level, old, new, flags);
- ret = bch2_trans_update(trans, &iter, &a->k_i, 0);
- if (ret)
- goto err;
-err:
- bch2_trans_iter_exit(trans, &iter);
- return ret;
+ return 0;
}
-int bch2_trans_mark_stripe(struct btree_trans *trans,
- enum btree_id btree_id, unsigned level,
- struct bkey_s_c old, struct bkey_i *new,
- unsigned flags)
-{
- const struct bch_stripe *old_s = NULL;
- struct bch_stripe *new_s = NULL;
- struct bch_replicas_padded r;
- unsigned i, nr_blocks;
- int ret = 0;
-
- if (old.k->type == KEY_TYPE_stripe)
- old_s = bkey_s_c_to_stripe(old).v;
- if (new->k.type == KEY_TYPE_stripe)
- new_s = &bkey_i_to_stripe(new)->v;
-
- /*
- * If the pointers aren't changing, we don't need to do anything:
- */
- if (new_s && old_s &&
- new_s->nr_blocks == old_s->nr_blocks &&
- new_s->nr_redundant == old_s->nr_redundant &&
- !memcmp(old_s->ptrs, new_s->ptrs,
- new_s->nr_blocks * sizeof(struct bch_extent_ptr)))
- return 0;
+/* KEY_TYPE_reservation */
- BUG_ON(new_s && old_s &&
- (new_s->nr_blocks != old_s->nr_blocks ||
- new_s->nr_redundant != old_s->nr_redundant));
-
- nr_blocks = new_s ? new_s->nr_blocks : old_s->nr_blocks;
-
- if (new_s) {
- s64 sectors = le16_to_cpu(new_s->sectors);
-
- bch2_bkey_to_replicas(&r.e, bkey_i_to_s_c(new));
- ret = bch2_update_replicas_list(trans, &r.e, sectors * new_s->nr_redundant);
- if (ret)
- return ret;
- }
+static int __trigger_reservation(struct btree_trans *trans,
+ enum btree_id btree_id, unsigned level,
+ struct bkey_s_c k, unsigned flags)
+{
+ struct bch_fs *c = trans->c;
+ unsigned replicas = bkey_s_c_to_reservation(k).v->nr_replicas;
+ s64 sectors = (s64) k.k->size * replicas;
- if (old_s) {
- s64 sectors = -((s64) le16_to_cpu(old_s->sectors));
+ if (flags & BTREE_TRIGGER_OVERWRITE)
+ sectors = -sectors;
- bch2_bkey_to_replicas(&r.e, old);
- ret = bch2_update_replicas_list(trans, &r.e, sectors * old_s->nr_redundant);
+ if (flags & BTREE_TRIGGER_TRANSACTIONAL) {
+ int ret = bch2_replicas_deltas_realloc(trans, 0);
if (ret)
return ret;
- }
-
- for (i = 0; i < nr_blocks; i++) {
- if (new_s && old_s &&
- !memcmp(&new_s->ptrs[i],
- &old_s->ptrs[i],
- sizeof(new_s->ptrs[i])))
- continue;
- if (new_s) {
- ret = bch2_trans_mark_stripe_bucket(trans,
- bkey_i_to_s_c_stripe(new), i, false);
- if (ret)
- break;
- }
+ struct replicas_delta_list *d = trans->fs_usage_deltas;
+ replicas = min(replicas, ARRAY_SIZE(d->persistent_reserved));
- if (old_s) {
- ret = bch2_trans_mark_stripe_bucket(trans,
- bkey_s_c_to_stripe(old), i, true);
- if (ret)
- break;
- }
+ d->persistent_reserved[replicas - 1] += sectors;
}
- return ret;
-}
-
-static int __trans_mark_reservation(struct btree_trans *trans,
- enum btree_id btree_id, unsigned level,
- struct bkey_s_c k, unsigned flags)
-{
- unsigned replicas = bkey_s_c_to_reservation(k).v->nr_replicas;
- s64 sectors = (s64) k.k->size;
- struct replicas_delta_list *d;
- int ret;
+ if (flags & BTREE_TRIGGER_GC) {
+ percpu_down_read(&c->mark_lock);
+ preempt_disable();
- if (flags & BTREE_TRIGGER_OVERWRITE)
- sectors = -sectors;
- sectors *= replicas;
+ struct bch_fs_usage *fs_usage = this_cpu_ptr(c->usage_gc);
- ret = bch2_replicas_deltas_realloc(trans, 0);
- if (ret)
- return ret;
+ replicas = min(replicas, ARRAY_SIZE(fs_usage->persistent_reserved));
+ fs_usage->reserved += sectors;
+ fs_usage->persistent_reserved[replicas - 1] += sectors;
- d = trans->fs_usage_deltas;
- replicas = clamp_t(unsigned, replicas, 1,
- ARRAY_SIZE(d->persistent_reserved));
+ preempt_enable();
+ percpu_up_read(&c->mark_lock);
+ }
- d->persistent_reserved[replicas - 1] += sectors;
return 0;
}
-int bch2_trans_mark_reservation(struct btree_trans *trans,
- enum btree_id btree_id, unsigned level,
- struct bkey_s_c old,
- struct bkey_i *new,
- unsigned flags)
+int bch2_trigger_reservation(struct btree_trans *trans,
+ enum btree_id btree_id, unsigned level,
+ struct bkey_s_c old, struct bkey_s new,
+ unsigned flags)
{
- return trigger_run_overwrite_then_insert(__trans_mark_reservation, trans, btree_id, level, old, new, flags);
+ return trigger_run_overwrite_then_insert(__trigger_reservation, trans, btree_id, level, old, new, flags);
}
+/* Mark superblocks: */
+
static int __bch2_trans_mark_metadata_bucket(struct btree_trans *trans,
struct bch_dev *ca, size_t b,
enum bch_data_type type,