From: Coly Li <colyli@suse.de>
To: linux-bcache@vger.kernel.org
Cc: linux-block@vger.kernel.org, Coly Li <colyli@suse.de>
Subject: [PATCH v2 11/17] bcache: handle cache prio_buckets and disk_buckets properly for bucket size > 8MB
Date: Wed, 15 Jul 2020 13:46:06 +0800 [thread overview]
Message-ID: <20200715054612.6349-12-colyli@suse.de> (raw)
In-Reply-To: <20200715054612.6349-1-colyli@suse.de>
Similar to c->uuids, struct cache's prio_buckets and disk_buckets also
have the potential memory allocation failure during cache registration
if the bucket size > 8MB.
ca->prio_buckets can be stored on cache device in multiple buckets, its
in-memory space is allocated by kzalloc() interface but normally
allocated by alloc_pages() because the size > KMALLOC_MAX_CACHE_SIZE.
So allocation of ca->prio_buckets has the MAX_ORDER restriction too. If
the bucket size > 8MB, by default the page allocator will fail because
the page order > 11 (default MAX_ORDER value). ca->prio_buckets should
also use meta_bucket_bytes(), meta_bucket_pages() to decide its memory
size and use alloc_meta_bucket_pages() to allocate pages, to avoid the
allocation failure during cache set registration when bucket size > 8MB.
ca->disk_buckets is a single bucket size memory buffer, it is used to
iterate each bucket of ca->prio_buckets, and compose the bio based on
memory of ca->disk_buckets, then write ca->disk_buckets memory to cache
disk one-by-one for each bucket of ca->prio_buckets. ca->disk_buckets
should have in-memory size exact to the meta_bucket_pages(), this is the
size that ca->prio_buckets will be stored into each on-disk bucket.
This patch fixes the above issues and handle cache's prio_buckets and
disk_buckets properly for bucket size larger than 8MB.
Signed-off-by: Coly Li <colyli@suse.de>
---
drivers/md/bcache/bcache.h | 9 +++++----
drivers/md/bcache/super.c | 10 +++++-----
2 files changed, 10 insertions(+), 9 deletions(-)
diff --git a/drivers/md/bcache/bcache.h b/drivers/md/bcache/bcache.h
index 972f1aff0f70..0ebfda284866 100644
--- a/drivers/md/bcache/bcache.h
+++ b/drivers/md/bcache/bcache.h
@@ -782,11 +782,12 @@ static inline unsigned int meta_bucket_bytes(struct cache_sb *sb)
return meta_bucket_pages(sb) << PAGE_SHIFT;
}
-#define prios_per_bucket(c) \
- ((bucket_bytes(c) - sizeof(struct prio_set)) / \
+#define prios_per_bucket(ca) \
+ ((meta_bucket_bytes(&(ca)->sb) - sizeof(struct prio_set)) / \
sizeof(struct bucket_disk))
-#define prio_buckets(c) \
- DIV_ROUND_UP((size_t) (c)->sb.nbuckets, prios_per_bucket(c))
+
+#define prio_buckets(ca) \
+ DIV_ROUND_UP((size_t) (ca)->sb.nbuckets, prios_per_bucket(ca))
static inline size_t sector_to_bucket(struct cache_set *c, sector_t s)
{
diff --git a/drivers/md/bcache/super.c b/drivers/md/bcache/super.c
index 9e1f2f529fc3..e2c28aa5531c 100644
--- a/drivers/md/bcache/super.c
+++ b/drivers/md/bcache/super.c
@@ -563,7 +563,7 @@ static void prio_io(struct cache *ca, uint64_t bucket, int op,
bio->bi_iter.bi_sector = bucket * ca->sb.bucket_size;
bio_set_dev(bio, ca->bdev);
- bio->bi_iter.bi_size = bucket_bytes(ca);
+ bio->bi_iter.bi_size = meta_bucket_bytes(&ca->sb);
bio->bi_end_io = prio_endio;
bio->bi_private = ca;
@@ -621,7 +621,7 @@ int bch_prio_write(struct cache *ca, bool wait)
p->next_bucket = ca->prio_buckets[i + 1];
p->magic = pset_magic(&ca->sb);
- p->csum = bch_crc64(&p->magic, bucket_bytes(ca) - 8);
+ p->csum = bch_crc64(&p->magic, meta_bucket_bytes(&ca->sb) - 8);
bucket = bch_bucket_alloc(ca, RESERVE_PRIO, wait);
BUG_ON(bucket == -1);
@@ -674,7 +674,7 @@ static int prio_read(struct cache *ca, uint64_t bucket)
prio_io(ca, bucket, REQ_OP_READ, 0);
if (p->csum !=
- bch_crc64(&p->magic, bucket_bytes(ca) - 8)) {
+ bch_crc64(&p->magic, meta_bucket_bytes(&ca->sb) - 8)) {
pr_warn("bad csum reading priorities\n");
goto out;
}
@@ -2222,7 +2222,7 @@ void bch_cache_release(struct kobject *kobj)
ca->set->cache[ca->sb.nr_this_dev] = NULL;
}
- free_pages((unsigned long) ca->disk_buckets, ilog2(bucket_pages(ca)));
+ free_pages((unsigned long) ca->disk_buckets, ilog2(meta_bucket_pages(&ca->sb)));
kfree(ca->prio_buckets);
vfree(ca->buckets);
@@ -2319,7 +2319,7 @@ static int cache_alloc(struct cache *ca)
goto err_prio_buckets_alloc;
}
- ca->disk_buckets = alloc_bucket_pages(GFP_KERNEL, ca);
+ ca->disk_buckets = alloc_meta_bucket_pages(GFP_KERNEL, &ca->sb);
if (!ca->disk_buckets) {
err = "ca->disk_buckets alloc failed";
goto err_disk_buckets_alloc;
--
2.26.2
next prev parent reply other threads:[~2020-07-15 5:46 UTC|newest]
Thread overview: 29+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-07-15 5:45 [PATCH v2 00/17] bcache: extend bucket size to 32bit width Coly Li
2020-07-15 5:45 ` [PATCH v2 01/17] bcache: add comments to mark member offset of struct cache_sb_disk Coly Li
2020-07-15 6:02 ` Hannes Reinecke
2020-07-15 9:03 ` Coly Li
2020-07-15 9:08 ` Johannes Thumshirn
2020-07-15 10:35 ` Coly Li
2020-07-15 5:45 ` [PATCH v2 02/17] bcache: add read_super_basic() to read major part of super block Coly Li
2020-07-15 6:03 ` Hannes Reinecke
2020-07-15 5:45 ` [PATCH v2 03/17] bcache: add more accurate error information in read_super_basic() Coly Li
2020-07-15 6:04 ` Hannes Reinecke
2020-07-15 5:45 ` [PATCH v2 04/17] bcache: disassemble the big if() checks in bch_cache_set_alloc() Coly Li
2020-07-15 6:08 ` Hannes Reinecke
2020-07-15 5:46 ` [PATCH v2 05/17] bcache: fix super block seq numbers comparision in register_cache_set() Coly Li
2020-07-15 6:10 ` Hannes Reinecke
2020-07-15 5:46 ` [PATCH v2 06/17] bcache: increase super block version for cache device and backing device Coly Li
2020-07-15 6:11 ` Hannes Reinecke
2020-07-15 5:46 ` [PATCH v2 07/17] bcache: move bucket related code into read_super_basic() Coly Li
2020-07-15 6:44 ` Hannes Reinecke
2020-07-15 5:46 ` [PATCH v2 08/17] bcache: struct cache_sb is only for in-memory super block now Coly Li
2020-07-15 6:45 ` Hannes Reinecke
2020-07-15 5:46 ` [PATCH v2 09/17] bcache: introduce meta_bucket_pages() related helper routines Coly Li
2020-07-15 5:46 ` [PATCH v2 10/17] bcache: handle c->uuids properly for bucket size > 8MB Coly Li
2020-07-15 5:46 ` Coly Li [this message]
2020-07-15 5:46 ` [PATCH v2 12/17] bcache: handle cache set verify_ondisk " Coly Li
2020-07-15 5:46 ` [PATCH v2 13/17] bcache: handle btree node memory allocation " Coly Li
2020-07-15 5:46 ` [PATCH v2 14/17] bcache: add bucket_size_hi into struct cache_sb_disk for large bucket Coly Li
2020-07-15 5:46 ` [PATCH v2 15/17] bcache: add sysfs file to display feature sets information of cache set Coly Li
2020-07-15 5:46 ` [PATCH v2 16/17] bcache: avoid extra memory allocation from mempool c->fill_iter Coly Li
2020-07-15 5:46 ` [PATCH v2 17/17] bcache: avoid extra memory consumption in struct bbio for large bucket size Coly Li
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=20200715054612.6349-12-colyli@suse.de \
--to=colyli@suse.de \
--cc=linux-bcache@vger.kernel.org \
--cc=linux-block@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;
as well as URLs for NNTP newsgroup(s).