#include "bcache.h"
#include "btree.h"
#include "debug.h"
#include "request.h"
#include "writeback.h"
#include <linux/module.h>
#include <linux/hash.h>
#include <linux/random.h>
#include <linux/backing-dev.h>
#include <trace/events/bcache.h>
#define CUTOFF_CACHE_ADD 95
#define CUTOFF_CACHE_READA 90
struct kmem_cache *bch_search_cache;
static void bch_data_insert_start(struct closure *cl);
static unsigned int cache_mode(struct cached_dev *dc)
{
return BDEV_CACHE_MODE(&dc->sb);
}
static bool verify(struct cached_dev *dc)
{
return dc->verify;
}
static void bio_csum(struct bio *bio, struct bkey *k)
{
struct bio_vec bv;
struct bvec_iter iter;
uint64_t csum = 0;
bio_for_each_segment(bv, bio, iter) {
void *d = bvec_kmap_local(&bv);
csum = crc64_be(csum, d, bv.bv_len);
kunmap_local(d);
}
k->ptr[KEY_PTRS(k)] = csum & (~0ULL >> 1);
}
static void bch_data_insert_keys(struct closure *cl)
{
struct data_insert_op *op = container_of(cl, struct data_insert_op, cl);
atomic_t *journal_ref = NULL;
struct bkey *replace_key = op->replace ? &op->replace_key : NULL;
int ret;
if (!op->replace)
journal_ref = bch_journal(op->c, &op->insert_keys,
op->flush_journal ? cl : NULL);
ret = bch_btree_insert(op->c, &op->insert_keys,
journal_ref, replace_key);
if (ret == -ESRCH) {
op->replace_collision = true;
} else if (ret) {
op->status = BLK_STS_RESOURCE;
op->insert_data_done = true;
}
if (journal_ref)
atomic_dec_bug(journal_ref);
if (!op->insert_data_done) {
continue_at(cl, bch_data_insert_start, op->wq);
return;
}
bch_keylist_free(&op->insert_keys);
closure_return(cl);
}
static int bch_keylist_realloc(struct keylist *l, unsigned int u64s,
struct cache_set *c)
{
size_t oldsize = bch_keylist_nkeys(l);
size_t newsize = oldsize + u64s;
if (newsize * sizeof(uint64_t) > block_bytes(c->cache) - sizeof(struct jset))
return -ENOMEM;
return __bch_keylist_realloc(l, u64s);
}
static void bch_data_invalidate(struct closure *cl)
{
struct data_insert_op *op = container_of(cl, struct data_insert_op, cl);
struct bio *bio = op->bio;
pr_debug("invalidating %i sectors from %llu\n",
bio_sectors(bio), (uint64_t) bio->bi_iter.bi_sector);
while (bio_sectors(bio)) {
unsigned int sectors = min(bio_sectors(bio),
1U << (KEY_SIZE_BITS - 1));
if (bch_keylist_realloc(&op->insert_keys, 2, op->c))
goto out;
bio->bi_iter.bi_sector += sectors;
bio->bi_iter.bi_size -= sectors << 9;
bch_keylist_add(&op->insert_keys,
&KEY(op->inode,
bio->bi_iter.bi_sector,
sectors));
}
op->insert_data_done = true;
bio_put(bio);
out:
continue_at(cl, bch_data_insert_keys, op->wq);
}
static void bch_data_insert_error(struct closure *cl)
{
struct data_insert_op *op = container_of(cl, struct data_insert_op, cl);
struct bkey *src = op->insert_keys.keys, *dst = op->insert_keys.keys;
while (src != op->insert_keys.top) {
struct bkey *n = bkey_next(src);
SET_KEY_PTRS(src, 0);
memmove(dst, src, bkey_bytes(src));
dst = bkey_next(dst);
src = n;
}
op->insert_keys.top = dst;
bch_data_insert_keys(cl);
}
static void bch_data_insert_endio(struct bio *bio)
{
struct closure *cl = bio->bi_private;
struct data_insert_op *op = container_of(cl, struct data_insert_op, cl);
if (bio->bi_status) {
if (op->writeback)
op->status = bio->bi_status;
else if (!op->replace)
set_closure_fn(cl, bch_data_insert_error, op->wq);
else
set_closure_fn(cl, NULL, NULL);
}
bch_bbio_endio(op->c, bio, bio->bi_status, "writing data to cache");
}
static void bch_data_insert_start(struct closure *cl)
{
struct data_insert_op *op = container_of(cl, struct data_insert_op, cl);
struct bio *bio = op->bio, *n;
if (op->bypass)
return bch_data_invalidate(cl);
if (atomic_sub_return(bio_sectors(bio), &op->c->sectors_to_gc) < 0)
wake_up_gc(op->c);
bio->bi_opf &= ~(REQ_PREFLUSH|REQ_FUA);
do {
unsigned int i;
struct bkey *k;
struct bio_set *split = &op->c->bio_split;
if (bch_keylist_realloc(&op->insert_keys,
3 + (op->csum ? 1 : 0),
op->c)) {
continue_at(cl, bch_data_insert_keys, op->wq);
return;
}
k = op->insert_keys.top;
bkey_init(k);
SET_KEY_INODE(k, op->inode);
SET_KEY_OFFSET(k, bio->bi_iter.bi_sector);
if (!bch_alloc_sectors(op->c, k, bio_sectors(bio),
op->write_point, op->write_prio,
op->writeback))
goto err;
n = bio_next_split(bio, KEY_SIZE(k), GFP_NOIO, split);
n->bi_end_io = bch_data_insert_endio;
n->bi_private = cl;
if (op->writeback) {
SET_KEY_DIRTY(k, true);
for (i = 0; i < KEY_PTRS(k); i++)
SET_GC_MARK(PTR_BUCKET(op->c, k, i),
GC_MARK_DIRTY);
}
SET_KEY_CSUM(k, op->csum);
if (KEY_CSUM(k))
bio_csum(n, k);
trace_bcache_cache_insert(k);
bch_keylist_push(&op->insert_keys);
n->bi_opf = REQ_OP_WRITE;
bch_submit_bbio(n, op->c, k, 0);
} while (n != bio);
op->insert_data_done = true;
continue_at(cl, bch_data_insert_keys, op->wq);
return;
err:
BUG_ON(op->writeback);
if (!op->replace) {
op->bypass = true;
return bch_data_invalidate(cl);
} else {
op->insert_data_done = true;
bio_put(bio);
if (!bch_keylist_empty(&op->insert_keys))
continue_at(cl, bch_data_insert_keys, op->wq);
else
closure_return(cl);
}
}
void bch_data_insert(struct closure *cl)
{
struct data_insert_op *op = container_of(cl, struct data_insert_op, cl);
trace_bcache_write(op->c, op->inode, op->bio,
op->writeback, op->bypass);
bch_keylist_init(&op->insert_keys);
bio_get(op->bio);
bch_data_insert_start(cl);
}
unsigned int bch_get_congested(const struct cache_set *c)
{
int i;
if (!c->congested_read_threshold_us &&
!c->congested_write_threshold_us)
return 0;
i = (local_clock_us() - c->congested_last_us) / 1024;
if (i < 0)
return 0;
i += atomic_read(&c->congested);
if (i >= 0)
return 0;
i += CONGESTED_MAX;
if (i > 0)
i = fract_exp_two(i, 6);
i -= hweight32(get_random_u32());
return i > 0 ? i : 1;
}
static void add_sequential(struct task_struct *t)
{
ewma_add(t->sequential_io_avg,
t->sequential_io, 8, 0);
t->sequential_io = 0;
}
static struct hlist_head *iohash(struct cached_dev *dc, uint64_t k)
{
return &dc->io_hash[hash_64(k, RECENT_IO_BITS)];
}
static bool check_should_bypass(struct cached_dev *dc, struct bio *bio)
{
struct cache_set *c = dc->disk.c;
unsigned int mode = cache_mode(dc);
unsigned int sectors, congested;
struct task_struct *task = current;
struct io *i;
if (test_bit(BCACHE_DEV_DETACHING, &dc->disk.flags) ||
c->gc_stats.in_use > CUTOFF_CACHE_ADD ||
(bio_op(bio) == REQ_OP_DISCARD))
goto skip;
if (mode == CACHE_MODE_NONE ||
(mode == CACHE_MODE_WRITEAROUND &&
op_is_write(bio_op(bio))))
goto skip;
if ((bio->bi_opf & (REQ_RAHEAD|REQ_BACKGROUND))) {
if (!(bio->bi_opf & (REQ_META|REQ_PRIO)) &&
(dc->cache_readahead_policy != BCH_CACHE_READA_ALL))
goto skip;
}
if (bio->bi_iter.bi_sector & (c->cache->sb.block_size - 1) ||
bio_sectors(bio) & (c->cache->sb.block_size - 1)) {
pr_debug("skipping unaligned io\n");
goto skip;
}
if (bypass_torture_test(dc)) {
if (get_random_u32_below(4) == 3)
goto skip;
else
goto rescale;
}
congested = bch_get_congested(c);
if (!congested && !dc->sequential_cutoff)
goto rescale;
spin_lock(&dc->io_lock);
hlist_for_each_entry(i, iohash(dc, bio->bi_iter.bi_sector), hash)
if (i->last == bio->bi_iter.bi_sector &&
time_before(jiffies, i->jiffies))
goto found;
i = list_first_entry(&dc->io_lru, struct io, lru);
add_sequential(task);
i->sequential = 0;
found:
if (i->sequential + bio->bi_iter.bi_size > i->sequential)
i->sequential += bio->bi_iter.bi_size;
i->last = bio_end_sector(bio);
i->jiffies = jiffies + msecs_to_jiffies(5000);
task->sequential_io = i->sequential;
hlist_del(&i->hash);
hlist_add_head(&i->hash, iohash(dc, i->last));
list_move_tail(&i->lru, &dc->io_lru);
spin_unlock(&dc->io_lock);
sectors = max(task->sequential_io,
task->sequential_io_avg) >> 9;
if (dc->sequential_cutoff &&
sectors >= dc->sequential_cutoff >> 9) {
trace_bcache_bypass_sequential(bio);
goto skip;
}
if (congested && sectors >= congested) {
trace_bcache_bypass_congested(bio);
goto skip;
}
rescale:
bch_rescale_priorities(c, bio_sectors(bio));
return false;
skip:
bch_mark_sectors_bypassed(c, dc, bio_sectors(bio));
return true;
}
struct search {
struct closure cl;
struct bbio bio;
struct bio *orig_bio;
struct bio *cache_miss;
struct bcache_device *d;
unsigned int insert_bio_sectors;
unsigned int recoverable:1;
unsigned int write:1;
unsigned int read_dirty_data:1;
unsigned int cache_missed:1;
struct block_device *orig_bdev;
unsigned long start_time;
struct btree_op op;
struct data_insert_op iop;
};
static void bch_cache_read_endio(struct bio *bio)
{
struct bbio *b = container_of(bio, struct bbio, bio);
struct closure *cl = bio->bi_private;
struct search *s = container_of(cl, struct search, cl);
if (bio->bi_status)
s->iop.status = bio->bi_status;
else if (!KEY_DIRTY(&b->key) &&
ptr_stale(s->iop.c, &b->key, 0)) {
atomic_long_inc(&s->iop.c->cache_read_races);
s->iop.status = BLK_STS_IOERR;
}
bch_bbio_endio(s->iop.c, bio, bio->bi_status, "reading from cache");
}
static int cache_lookup_fn(struct btree_op *op, struct btree *b, struct bkey *k)
{
struct search *s = container_of(op, struct search, op);
struct bio *n, *bio = &s->bio.bio;
struct bkey *bio_key;
unsigned int ptr;
if (bkey_cmp(k, &KEY(s->iop.inode, bio->bi_iter.bi_sector, 0)) <= 0)
return MAP_CONTINUE;
if (KEY_INODE(k) != s->iop.inode ||
KEY_START(k) > bio->bi_iter.bi_sector) {
unsigned int bio_sectors = bio_sectors(bio);
unsigned int sectors = KEY_INODE(k) == s->iop.inode
? min_t(uint64_t, INT_MAX,
KEY_START(k) - bio->bi_iter.bi_sector)
: INT_MAX;
int ret = s->d->cache_miss(b, s, bio, sectors);
if (ret != MAP_CONTINUE)
return ret;
BUG_ON(bio_sectors <= sectors);
}
if (!KEY_SIZE(k))
return MAP_CONTINUE;
ptr = 0;
PTR_BUCKET(b->c, k, ptr)->prio = INITIAL_PRIO;
if (KEY_DIRTY(k))
s->read_dirty_data = true;
n = bio_next_split(bio, min_t(uint64_t, INT_MAX,
KEY_OFFSET(k) - bio->bi_iter.bi_sector),
GFP_NOIO, &s->d->bio_split);
bio_key = &container_of(n, struct bbio, bio)->key;
bch_bkey_copy_single_ptr(bio_key, k, ptr);
bch_cut_front(&KEY(s->iop.inode, n->bi_iter.bi_sector, 0), bio_key);
bch_cut_back(&KEY(s->iop.inode, bio_end_sector(n), 0), bio_key);
n->bi_end_io = bch_cache_read_endio;
n->bi_private = &s->cl;
__bch_submit_bbio(n, b->c);
return n == bio ? MAP_DONE : MAP_CONTINUE;
}
static void cache_lookup(struct closure *cl)
{
struct search *s = container_of(cl, struct search, iop.cl);
struct bio *bio = &s->bio.bio;
struct cached_dev *dc;
int ret;
bch_btree_op_init(&s->op, -1);
ret = bch_btree_map_keys(&s->op, s->iop.c,
&KEY(s->iop.inode, bio->bi_iter.bi_sector, 0),
cache_lookup_fn, MAP_END_KEY);
if (ret == -EAGAIN) {
continue_at(cl, cache_lookup, bcache_wq);
return;
}
if (ret < 0) {
BUG_ON(ret == -EINTR);
if (s->d && s->d->c &&
!UUID_FLASH_ONLY(&s->d->c->uuids[s->d->id])) {
dc = container_of(s->d, struct cached_dev, disk);
if (dc && atomic_read(&dc->has_dirty))
s->recoverable = false;
}
if (!s->iop.status)
s->iop.status = BLK_STS_IOERR;
}
closure_return(cl);
}
static void request_endio(struct bio *bio)
{
struct closure *cl = bio->bi_private;
if (bio->bi_status) {
struct search *s = container_of(cl, struct search, cl);
s->iop.status = bio->bi_status;
s->recoverable = false;
}
bio_put(bio);
closure_put(cl);
}
static void backing_request_endio(struct bio *bio)
{
struct closure *cl = bio->bi_private;
if (bio->bi_status) {
struct search *s = container_of(cl, struct search, cl);
struct cached_dev *dc = container_of(s->d,
struct cached_dev, disk);
if (unlikely(s->iop.writeback &&
bio->bi_opf & REQ_PREFLUSH)) {
pr_err("Can't flush %pg: returned bi_status %i\n",
dc->bdev, bio->bi_status);
} else {
s->iop.status = bio->bi_status;
}
s->recoverable = false;
bch_count_backing_io_errors(dc, bio);
}
bio_put(bio);
closure_put(cl);
}
static void bio_complete(struct search *s)
{
if (s->orig_bio) {
bio_end_io_acct_remapped(s->orig_bio, s->start_time,
s->orig_bdev);
trace_bcache_request_end(s->d, s->orig_bio);
s->orig_bio->bi_status = s->iop.status;
bio_endio(s->orig_bio);
s->orig_bio = NULL;
}
}
static void do_bio_hook(struct search *s,
struct bio *orig_bio,
bio_end_io_t *end_io_fn)
{
struct bio *bio = &s->bio.bio;
bio_init_clone(orig_bio->bi_bdev, bio, orig_bio, GFP_NOIO);
bio->bi_end_io = end_io_fn;
bio->bi_private = &s->cl;
bio_cnt_set(bio, 3);
}
static void search_free(struct closure *cl)
{
struct search *s = container_of(cl, struct search, cl);
atomic_dec(&s->iop.c->search_inflight);
if (s->iop.bio)
bio_put(s->iop.bio);
bio_complete(s);
closure_debug_destroy(cl);
mempool_free(s, &s->iop.c->search);
}
static inline struct search *search_alloc(struct bio *bio,
struct bcache_device *d, struct block_device *orig_bdev,
unsigned long start_time)
{
struct search *s;
s = mempool_alloc(&d->c->search, GFP_NOIO);
closure_init(&s->cl, NULL);
do_bio_hook(s, bio, request_endio);
atomic_inc(&d->c->search_inflight);
s->orig_bio = bio;
s->cache_miss = NULL;
s->cache_missed = 0;
s->d = d;
s->recoverable = 1;
s->write = op_is_write(bio_op(bio));
s->read_dirty_data = 0;
s->orig_bdev = orig_bdev;
s->start_time = start_time;
s->iop.c = d->c;
s->iop.bio = NULL;
s->iop.inode = d->id;
s->iop.write_point = hash_long((unsigned long) current, 16);
s->iop.write_prio = 0;
s->iop.status = 0;
s->iop.flags = 0;
s->iop.flush_journal = op_is_flush(bio->bi_opf);
s->iop.wq = bcache_wq;
return s;
}
static void cached_dev_bio_complete(struct closure *cl)
{
struct search *s = container_of(cl, struct search, cl);
struct cached_dev *dc = container_of(s->d, struct cached_dev, disk);
cached_dev_put(dc);
search_free(cl);
}
static void cached_dev_read_error_done(struct closure *cl)
{
struct search *s = container_of(cl, struct search, cl);
if (s->iop.replace_collision)
bch_mark_cache_miss_collision(s->iop.c, s->d);
if (s->iop.bio)
bio_free_pages(s->iop.bio);
cached_dev_bio_complete(cl);
}
static void cached_dev_read_error(struct closure *cl)
{
struct search *s = container_of(cl, struct search, cl);
struct bio *bio = &s->bio.bio;
if (s->recoverable && !s->read_dirty_data) {
trace_bcache_read_retry(s->orig_bio);
s->iop.status = 0;
do_bio_hook(s, s->orig_bio, backing_request_endio);
closure_bio_submit(s->iop.c, bio, cl);
}
continue_at(cl, cached_dev_read_error_done, NULL);
}
static void cached_dev_cache_miss_done(struct closure *cl)
{
struct search *s = container_of(cl, struct search, cl);
struct bcache_device *d = s->d;
if (s->iop.replace_collision)
bch_mark_cache_miss_collision(s->iop.c, s->d);
if (s->iop.bio)
bio_free_pages(s->iop.bio);
cached_dev_bio_complete(cl);
closure_put(&d->cl);
}
static void cached_dev_read_done(struct closure *cl)
{
struct search *s = container_of(cl, struct search, cl);
struct cached_dev *dc = container_of(s->d, struct cached_dev, disk);
if (s->iop.bio) {
bio_reset(s->iop.bio, s->cache_miss->bi_bdev, REQ_OP_READ);
s->iop.bio->bi_iter.bi_sector =
s->cache_miss->bi_iter.bi_sector;
s->iop.bio->bi_iter.bi_size = s->insert_bio_sectors << 9;
bio_clone_blkg_association(s->iop.bio, s->cache_miss);
bch_bio_map(s->iop.bio, NULL);
bio_copy_data(s->cache_miss, s->iop.bio);
bio_put(s->cache_miss);
s->cache_miss = NULL;
}
if (verify(dc) && s->recoverable && !s->read_dirty_data)
bch_data_verify(dc, s->orig_bio);
closure_get(&dc->disk.cl);
bio_complete(s);
if (s->iop.bio &&
!test_bit(CACHE_SET_STOPPING, &s->iop.c->flags)) {
BUG_ON(!s->iop.replace);
closure_call(&s->iop.cl, bch_data_insert, NULL, cl);
}
continue_at(cl, cached_dev_cache_miss_done, NULL);
}
static void cached_dev_read_done_bh(struct closure *cl)
{
struct search *s = container_of(cl, struct search, cl);
struct cached_dev *dc = container_of(s->d, struct cached_dev, disk);
bch_mark_cache_accounting(s->iop.c, s->d,
!s->cache_missed, s->iop.bypass);
trace_bcache_read(s->orig_bio, !s->cache_missed, s->iop.bypass);
if (s->iop.status)
continue_at_nobarrier(cl, cached_dev_read_error, bcache_wq);
else if (s->iop.bio || verify(dc))
continue_at_nobarrier(cl, cached_dev_read_done, bcache_wq);
else
continue_at_nobarrier(cl, cached_dev_bio_complete, NULL);
}
static int cached_dev_cache_miss(struct btree *b, struct search *s,
struct bio *bio, unsigned int sectors)
{
int ret = MAP_CONTINUE;
struct cached_dev *dc = container_of(s->d, struct cached_dev, disk);
struct bio *miss, *cache_bio;
unsigned int size_limit;
s->cache_missed = 1;
if (s->cache_miss || s->iop.bypass) {
miss = bio_next_split(bio, sectors, GFP_NOIO, &s->d->bio_split);
ret = miss == bio ? MAP_DONE : MAP_CONTINUE;
goto out_submit;
}
size_limit = min_t(unsigned int, BIO_MAX_VECS * PAGE_SECTORS,
(1 << KEY_SIZE_BITS) - 1);
s->insert_bio_sectors = min3(size_limit, sectors, bio_sectors(bio));
s->iop.replace_key = KEY(s->iop.inode,
bio->bi_iter.bi_sector + s->insert_bio_sectors,
s->insert_bio_sectors);
ret = bch_btree_insert_check_key(b, &s->op, &s->iop.replace_key);
if (ret)
return ret;
s->iop.replace = true;
miss = bio_next_split(bio, s->insert_bio_sectors, GFP_NOIO,
&s->d->bio_split);
ret = miss == bio ? MAP_DONE : -EINTR;
cache_bio = bio_alloc_bioset(miss->bi_bdev,
DIV_ROUND_UP(s->insert_bio_sectors, PAGE_SECTORS),
0, GFP_NOWAIT, &dc->disk.bio_split);
if (!cache_bio)
goto out_submit;
cache_bio->bi_iter.bi_sector = miss->bi_iter.bi_sector;
cache_bio->bi_iter.bi_size = s->insert_bio_sectors << 9;
cache_bio->bi_end_io = backing_request_endio;
cache_bio->bi_private = &s->cl;
bch_bio_map(cache_bio, NULL);
if (bch_bio_alloc_pages(cache_bio, __GFP_NOWARN|GFP_NOIO))
goto out_put;
s->cache_miss = miss;
s->iop.bio = cache_bio;
bio_get(cache_bio);
closure_bio_submit(s->iop.c, cache_bio, &s->cl);
return ret;
out_put:
bio_put(cache_bio);
out_submit:
miss->bi_end_io = backing_request_endio;
miss->bi_private = &s->cl;
closure_bio_submit(s->iop.c, miss, &s->cl);
return ret;
}
static void cached_dev_read(struct cached_dev *dc, struct search *s)
{
struct closure *cl = &s->cl;
closure_call(&s->iop.cl, cache_lookup, NULL, cl);
continue_at(cl, cached_dev_read_done_bh, NULL);
}
static void cached_dev_write_complete(struct closure *cl)
{
struct search *s = container_of(cl, struct search, cl);
struct cached_dev *dc = container_of(s->d, struct cached_dev, disk);
up_read_non_owner(&dc->writeback_lock);
cached_dev_bio_complete(cl);
}
static void cached_dev_write(struct cached_dev *dc, struct search *s)
{
struct closure *cl = &s->cl;
struct bio *bio = &s->bio.bio;
struct bkey start = KEY(dc->disk.id, bio->bi_iter.bi_sector, 0);
struct bkey end = KEY(dc->disk.id, bio_end_sector(bio), 0);
bch_keybuf_check_overlapping(&s->iop.c->moving_gc_keys, &start, &end);
down_read_non_owner(&dc->writeback_lock);
if (bch_keybuf_check_overlapping(&dc->writeback_keys, &start, &end)) {
s->iop.bypass = false;
s->iop.writeback = true;
}
if (bio_op(bio) == REQ_OP_DISCARD)
s->iop.bypass = true;
if (should_writeback(dc, s->orig_bio,
cache_mode(dc),
s->iop.bypass)) {
s->iop.bypass = false;
s->iop.writeback = true;
}
if (s->iop.bypass) {
s->iop.bio = s->orig_bio;
bio_get(s->iop.bio);
if (bio_op(bio) == REQ_OP_DISCARD &&
!bdev_max_discard_sectors(dc->bdev))
goto insert_data;
bio->bi_end_io = backing_request_endio;
closure_bio_submit(s->iop.c, bio, cl);
} else if (s->iop.writeback) {
bch_writeback_add(dc);
s->iop.bio = bio;
if (bio->bi_opf & REQ_PREFLUSH) {
struct bio *flush;
flush = bio_alloc_bioset(bio->bi_bdev, 0,
REQ_OP_WRITE | REQ_PREFLUSH,
GFP_NOIO, &dc->disk.bio_split);
if (!flush) {
s->iop.status = BLK_STS_RESOURCE;
goto insert_data;
}
flush->bi_end_io = backing_request_endio;
flush->bi_private = cl;
closure_bio_submit(s->iop.c, flush, cl);
}
} else {
s->iop.bio = bio_alloc_clone(bio->bi_bdev, bio, GFP_NOIO,
&dc->disk.bio_split);
bio->bi_end_io = backing_request_endio;
closure_bio_submit(s->iop.c, bio, cl);
}
insert_data:
closure_call(&s->iop.cl, bch_data_insert, NULL, cl);
continue_at(cl, cached_dev_write_complete, NULL);
}
static void cached_dev_nodata(struct closure *cl)
{
struct search *s = container_of(cl, struct search, cl);
struct bio *bio = &s->bio.bio;
if (s->iop.flush_journal)
bch_journal_meta(s->iop.c, cl);
bio->bi_end_io = backing_request_endio;
closure_bio_submit(s->iop.c, bio, cl);
continue_at(cl, cached_dev_bio_complete, NULL);
}
struct detached_dev_io_private {
struct bcache_device *d;
unsigned long start_time;
bio_end_io_t *bi_end_io;
void *bi_private;
struct block_device *orig_bdev;
};
static void detached_dev_end_io(struct bio *bio)
{
struct detached_dev_io_private *ddip;
ddip = bio->bi_private;
bio->bi_end_io = ddip->bi_end_io;
bio->bi_private = ddip->bi_private;
bio_end_io_acct_remapped(bio, ddip->start_time, ddip->orig_bdev);
if (bio->bi_status) {
struct cached_dev *dc = container_of(ddip->d,
struct cached_dev, disk);
bch_count_backing_io_errors(dc, bio);
}
kfree(ddip);
bio->bi_end_io(bio);
}
static void detached_dev_do_request(struct bcache_device *d, struct bio *bio,
struct block_device *orig_bdev, unsigned long start_time)
{
struct detached_dev_io_private *ddip;
struct cached_dev *dc = container_of(d, struct cached_dev, disk);
ddip = kzalloc(sizeof(struct detached_dev_io_private), GFP_NOIO);
if (!ddip) {
bio->bi_status = BLK_STS_RESOURCE;
bio->bi_end_io(bio);
return;
}
ddip->d = d;
ddip->orig_bdev = orig_bdev;
ddip->start_time = start_time;
ddip->bi_end_io = bio->bi_end_io;
ddip->bi_private = bio->bi_private;
bio->bi_end_io = detached_dev_end_io;
bio->bi_private = ddip;
if ((bio_op(bio) == REQ_OP_DISCARD) &&
!bdev_max_discard_sectors(dc->bdev))
bio->bi_end_io(bio);
else
submit_bio_noacct(bio);
}
static void quit_max_writeback_rate(struct cache_set *c,
struct cached_dev *this_dc)
{
int i;
struct bcache_device *d;
struct cached_dev *dc;
if (mutex_trylock(&bch_register_lock)) {
for (i = 0; i < c->devices_max_used; i++) {
if (!c->devices[i])
continue;
if (UUID_FLASH_ONLY(&c->uuids[i]))
continue;
d = c->devices[i];
dc = container_of(d, struct cached_dev, disk);
atomic_long_set(&dc->writeback_rate.rate, 1);
}
mutex_unlock(&bch_register_lock);
} else
atomic_long_set(&this_dc->writeback_rate.rate, 1);
}
void cached_dev_submit_bio(struct bio *bio)
{
struct search *s;
struct block_device *orig_bdev = bio->bi_bdev;
struct bcache_device *d = orig_bdev->bd_disk->private_data;
struct cached_dev *dc = container_of(d, struct cached_dev, disk);
unsigned long start_time;
int rw = bio_data_dir(bio);
if (unlikely((d->c && test_bit(CACHE_SET_IO_DISABLE, &d->c->flags)) ||
dc->io_disable)) {
bio->bi_status = BLK_STS_IOERR;
bio_endio(bio);
return;
}
if (likely(d->c)) {
if (atomic_read(&d->c->idle_counter))
atomic_set(&d->c->idle_counter, 0);
if (unlikely(atomic_read(&d->c->at_max_writeback_rate) == 1)) {
atomic_set(&d->c->at_max_writeback_rate, 0);
quit_max_writeback_rate(d->c, dc);
}
}
start_time = bio_start_io_acct(bio);
bio_set_dev(bio, dc->bdev);
bio->bi_iter.bi_sector += dc->sb.data_offset;
if (cached_dev_get(dc)) {
s = search_alloc(bio, d, orig_bdev, start_time);
trace_bcache_request_start(s->d, bio);
if (!bio->bi_iter.bi_size) {
continue_at_nobarrier(&s->cl,
cached_dev_nodata,
bcache_wq);
} else {
s->iop.bypass = check_should_bypass(dc, bio);
if (rw)
cached_dev_write(dc, s);
else
cached_dev_read(dc, s);
}
} else
detached_dev_do_request(d, bio, orig_bdev, start_time);
}
static int cached_dev_ioctl(struct bcache_device *d, blk_mode_t mode,
unsigned int cmd, unsigned long arg)
{
struct cached_dev *dc = container_of(d, struct cached_dev, disk);
if (dc->io_disable)
return -EIO;
if (!dc->bdev->bd_disk->fops->ioctl)
return -ENOTTY;
return dc->bdev->bd_disk->fops->ioctl(dc->bdev, mode, cmd, arg);
}
void bch_cached_dev_request_init(struct cached_dev *dc)
{
dc->disk.cache_miss = cached_dev_cache_miss;
dc->disk.ioctl = cached_dev_ioctl;
}
static int flash_dev_cache_miss(struct btree *b, struct search *s,
struct bio *bio, unsigned int sectors)
{
unsigned int bytes = min(sectors, bio_sectors(bio)) << 9;
swap(bio->bi_iter.bi_size, bytes);
zero_fill_bio(bio);
swap(bio->bi_iter.bi_size, bytes);
bio_advance(bio, bytes);
if (!bio->bi_iter.bi_size)
return MAP_DONE;
return MAP_CONTINUE;
}
static void flash_dev_nodata(struct closure *cl)
{
struct search *s = container_of(cl, struct search, cl);
if (s->iop.flush_journal)
bch_journal_meta(s->iop.c, cl);
continue_at(cl, search_free, NULL);
}
void flash_dev_submit_bio(struct bio *bio)
{
struct search *s;
struct closure *cl;
struct bcache_device *d = bio->bi_bdev->bd_disk->private_data;
if (unlikely(d->c && test_bit(CACHE_SET_IO_DISABLE, &d->c->flags))) {
bio->bi_status = BLK_STS_IOERR;
bio_endio(bio);
return;
}
s = search_alloc(bio, d, bio->bi_bdev, bio_start_io_acct(bio));
cl = &s->cl;
bio = &s->bio.bio;
trace_bcache_request_start(s->d, bio);
if (!bio->bi_iter.bi_size) {
continue_at_nobarrier(&s->cl,
flash_dev_nodata,
bcache_wq);
return;
} else if (bio_data_dir(bio)) {
bch_keybuf_check_overlapping(&s->iop.c->moving_gc_keys,
&KEY(d->id, bio->bi_iter.bi_sector, 0),
&KEY(d->id, bio_end_sector(bio), 0));
s->iop.bypass = (bio_op(bio) == REQ_OP_DISCARD) != 0;
s->iop.writeback = true;
s->iop.bio = bio;
closure_call(&s->iop.cl, bch_data_insert, NULL, cl);
} else {
closure_call(&s->iop.cl, cache_lookup, NULL, cl);
}
continue_at(cl, search_free, NULL);
}
static int flash_dev_ioctl(struct bcache_device *d, blk_mode_t mode,
unsigned int cmd, unsigned long arg)
{
return -ENOTTY;
}
void bch_flash_dev_request_init(struct bcache_device *d)
{
d->cache_miss = flash_dev_cache_miss;
d->ioctl = flash_dev_ioctl;
}
void bch_request_exit(void)
{
kmem_cache_destroy(bch_search_cache);
}
int __init bch_request_init(void)
{
bch_search_cache = KMEM_CACHE(search, 0);
if (!bch_search_cache)
return -ENOMEM;
return 0;
}