From: Coly Li <colyli@suse.de>
To: axboe@kernel.dk
Cc: linux-block@vger.kernel.org, linux-bcache@vger.kernel.org,
Coly Li <colyli@suse.de>, Hannes Reinecke <hare@suse.de>
Subject: [PATCH 20/25] bcache: handle btree node memory allocation properly for bucket size > 8MB
Date: Sat, 25 Jul 2020 20:00:34 +0800 [thread overview]
Message-ID: <20200725120039.91071-21-colyli@suse.de> (raw)
In-Reply-To: <20200725120039.91071-1-colyli@suse.de>
Currently the bcache internal btree node occupies a whole bucket. When
loading the btree node from cache device into memory, mca_data_alloc()
will call bch_btree_keys_alloc() to allocate memory for the whole bucket
size, ilog2(b->c->btree_pages) is send to bch_btree_keys_alloc() as the
parameter 'page_order'.
c->btree_pages is set as bucket_pages() in bch_cache_set_alloc(), for
bucket size > 8MB, ilog2(b->c->btree_pages) is 12 for 4KB page size. By
default the maximum page order __get_free_pages() accepts is MAX_ORDER
(11), in this condition bch_btree_keys_alloc() will always fail.
Because of other over-page-order allocation failure fails the cache
device registration, such btree node allocation failure wasn't observed
during runtime. After other blocking page allocation failures for bucket
size > 8MB, this btree node allocation issue may trigger potentical risk
e.g. infinite dead-loop to retry btree node allocation after failure.
This patch fixes the potential problem by setting c->btree_pages to
meta_bucket_pages() in bch_cache_set_alloc(). In the condition that
bucket size > 8MB, meta_bucket_pages() will always return a number which
won't exceed the maximum page order of the buddy allocator.
Signed-off-by: Coly Li <colyli@suse.de>
Reviewed-by: Hannes Reinecke <hare@suse.de>
---
drivers/md/bcache/super.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/md/bcache/super.c b/drivers/md/bcache/super.c
index 1c4a4c3557f7..172246ff5ec2 100644
--- a/drivers/md/bcache/super.c
+++ b/drivers/md/bcache/super.c
@@ -1867,7 +1867,7 @@ struct cache_set *bch_cache_set_alloc(struct cache_sb *sb)
c->nr_uuids = meta_bucket_bytes(&c->sb) / sizeof(struct uuid_entry);
c->devices_max_used = 0;
atomic_set(&c->attached_dev_nr, 0);
- c->btree_pages = bucket_pages(c);
+ c->btree_pages = meta_bucket_pages(&c->sb);
if (c->btree_pages > BTREE_MAX_PAGES)
c->btree_pages = max_t(int, c->btree_pages / 4,
BTREE_MAX_PAGES);
--
2.26.2
next prev parent reply other threads:[~2020-07-25 12:03 UTC|newest]
Thread overview: 34+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-07-25 12:00 [PATCH 00/25] bcache patches for Linux v5.9 Coly Li
2020-07-25 12:00 ` [PATCH 01/25] bcache: Fix typo in Kconfig name Coly Li
2020-07-25 12:00 ` [PATCH 02/25] bcache: allocate meta data pages as compound pages Coly Li
2020-07-25 12:00 ` [PATCH 03/25] bcache: journel: use for_each_clear_bit() to simplify the code Coly Li
2020-07-25 12:00 ` [PATCH 04/25] bcache: writeback: Remove unneeded variable i Coly Li
2020-07-25 12:00 ` [PATCH 05/25] bcache: movinggc: Use struct_size() helper in kzalloc() Coly Li
2020-07-25 12:00 ` [PATCH 06/25] bcache: Use struct_size() " Coly Li
2020-07-25 12:00 ` [PATCH 07/25] bcache: avoid nr_stripes overflow in bcache_device_init() Coly Li
2020-07-27 21:24 ` Sasha Levin
2020-07-25 12:00 ` [PATCH 08/25] bcache: fix overflow in offset_to_stripe() Coly Li
2020-07-27 21:24 ` Sasha Levin
2020-07-25 12:00 ` [PATCH 09/25] bcache: add read_super_common() to read major part of super block Coly Li
2020-07-25 12:00 ` [PATCH 10/25] bcache: add more accurate error information in read_super_common() Coly Li
2020-07-25 12:00 ` [PATCH 11/25] bcache: disassemble the big if() checks in bch_cache_set_alloc() Coly Li
2020-07-25 12:00 ` [PATCH 12/25] bcache: fix super block seq numbers comparision in register_cache_set() Coly Li
2020-07-25 12:00 ` [PATCH 13/25] bcache: increase super block version for cache device and backing device Coly Li
2020-07-25 12:00 ` [PATCH 14/25] bcache: move bucket related code into read_super_common() Coly Li
2020-07-25 12:00 ` [PATCH 15/25] bcache: struct cache_sb is only for in-memory super block now Coly Li
2020-07-25 12:00 ` [PATCH 16/25] bcache: introduce meta_bucket_pages() related helper routines Coly Li
2020-07-25 12:00 ` [PATCH 17/25] bcache: handle c->uuids properly for bucket size > 8MB Coly Li
2020-07-25 12:00 ` [PATCH 18/25] bcache: handle cache prio_buckets and disk_buckets " Coly Li
2020-07-25 12:00 ` [PATCH 19/25] bcache: handle cache set verify_ondisk " Coly Li
2020-07-25 12:00 ` Coly Li [this message]
2020-07-25 12:00 ` [PATCH 21/25] bcache: add bucket_size_hi into struct cache_sb_disk for large bucket Coly Li
2020-07-25 12:00 ` [PATCH 22/25] bcache: add sysfs file to display feature sets information of cache set Coly Li
2020-07-25 12:00 ` [PATCH 23/25] bcache: avoid extra memory allocation from mempool c->fill_iter Coly Li
2020-07-25 12:00 ` [PATCH 24/25] bcache: avoid extra memory consumption in struct bbio for large bucket size Coly Li
2020-07-25 12:00 ` [PATCH 25/25] bcache: fix bio_{start,end}_io_acct with proper device Coly Li
2020-07-26 15:07 ` Christoph Hellwig
2020-07-25 13:39 ` [PATCH 00/25] bcache patches for Linux v5.9 Jens Axboe
2020-07-28 12:14 ` Christoph Hellwig
2020-07-28 12:40 ` Coly Li
2020-07-28 12:41 ` Christoph Hellwig
2020-07-28 15:13 ` Jens Axboe
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=20200725120039.91071-21-colyli@suse.de \
--to=colyli@suse.de \
--cc=axboe@kernel.dk \
--cc=hare@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).