public inbox for linux-bcache@vger.kernel.org
 help / color / mirror / Atom feed
From: colyli@suse.de
To: axboe@kernel.dk
Cc: linux-bcache@vger.kernel.org, linux-block@vger.kernel.org,
	Coly Li <colyli@suse.de>
Subject: [PATCH 15/17] bcache: remove member accessed from struct btree
Date: Fri, 24 Jan 2020 01:01:40 +0800	[thread overview]
Message-ID: <20200123170142.98974-16-colyli@suse.de> (raw)
In-Reply-To: <20200123170142.98974-1-colyli@suse.de>

From: Coly Li <colyli@suse.de>

The member 'accessed' of struct btree is used in bch_mca_scan() when
shrinking btree node caches. The original idea is, if b->accessed is
set, clean it and look at next btree node cache from c->btree_cache
list, and only shrink the caches whose b->accessed is cleaned. Then
only cold btree node cache will be shrunk.

But when I/O pressure is high, it is very probably that b->accessed
of a btree node cache will be set again in bch_btree_node_get()
before bch_mca_scan() selects it again. Then there is no chance for
bch_mca_scan() to shrink enough memory back to slub or slab system.

This patch removes member accessed from struct btree, then once a
btree node ache is selected, it will be immediately shunk. By this
change, bch_mca_scan() may release btree node cahce more efficiently.

Signed-off-by: Coly Li <colyli@suse.de>
---
 drivers/md/bcache/btree.c | 8 ++------
 drivers/md/bcache/btree.h | 2 --
 2 files changed, 2 insertions(+), 8 deletions(-)

diff --git a/drivers/md/bcache/btree.c b/drivers/md/bcache/btree.c
index 14d6c33b0957..357535a5c89c 100644
--- a/drivers/md/bcache/btree.c
+++ b/drivers/md/bcache/btree.c
@@ -754,14 +754,12 @@ static unsigned long bch_mca_scan(struct shrinker *shrink,
 		b = list_first_entry(&c->btree_cache, struct btree, list);
 		list_rotate_left(&c->btree_cache);
 
-		if (!b->accessed &&
-		    !mca_reap(b, 0, false)) {
+		if (!mca_reap(b, 0, false)) {
 			mca_bucket_free(b);
 			mca_data_free(b);
 			rw_unlock(true, b);
 			freed++;
-		} else
-			b->accessed = 0;
+		}
 	}
 out:
 	mutex_unlock(&c->bucket_lock);
@@ -1069,7 +1067,6 @@ struct btree *bch_btree_node_get(struct cache_set *c, struct btree_op *op,
 	BUG_ON(!b->written);
 
 	b->parent = parent;
-	b->accessed = 1;
 
 	for (; i <= b->keys.nsets && b->keys.set[i].size; i++) {
 		prefetch(b->keys.set[i].tree);
@@ -1160,7 +1157,6 @@ struct btree *__bch_btree_node_alloc(struct cache_set *c, struct btree_op *op,
 		goto retry;
 	}
 
-	b->accessed = 1;
 	b->parent = parent;
 	bch_bset_init_next(&b->keys, b->keys.set->data, bset_magic(&b->c->sb));
 
diff --git a/drivers/md/bcache/btree.h b/drivers/md/bcache/btree.h
index 76cfd121a486..f4dcca449391 100644
--- a/drivers/md/bcache/btree.h
+++ b/drivers/md/bcache/btree.h
@@ -121,8 +121,6 @@ struct btree {
 	/* Key/pointer for this btree node */
 	BKEY_PADDED(key);
 
-	/* Single bit - set when accessed, cleared by shrinker */
-	unsigned long		accessed;
 	unsigned long		seq;
 	struct rw_semaphore	lock;
 	struct cache_set	*c;
-- 
2.16.4

  parent reply	other threads:[~2020-01-23 17:01 UTC|newest]

Thread overview: 24+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-01-23 17:01 [PATCH 00/17] bcache patches for Linux v5.6 colyli
2020-01-23 17:01 ` [PATCH 01/17] bcache: cached_dev_free needs to put the sb page colyli
2020-01-23 17:01 ` [PATCH 02/17] bcache: use a separate data structure for the on-disk super block colyli
2020-01-23 17:01 ` [PATCH 03/17] bcache: rework error unwinding in register_bcache colyli
2020-01-23 17:01 ` [PATCH 04/17] bcache: properly initialize 'path' and 'err' in register_bcache() colyli
2020-01-23 17:01 ` [PATCH 05/17] bcache: fix use-after-free " colyli
2020-01-23 17:01 ` [PATCH 06/17] bcache: transfer the sb_page reference to register_{bdev,cache} colyli
2020-01-23 17:01 ` [PATCH 07/17] bcache: return a pointer to the on-disk sb from read_super colyli
2020-01-23 17:01 ` [PATCH 08/17] bcache: store a pointer to the on-disk sb in the cache and cached_dev structures colyli
2020-01-23 17:01 ` [PATCH 09/17] bcache: use read_cache_page_gfp to read the superblock colyli
2020-01-23 17:01 ` [PATCH 10/17] lib: crc64: include <linux/crc64.h> for 'crc64_be' colyli
2020-01-23 17:01 ` [PATCH 11/17] bcache: add code comments for state->pool in __btree_sort() colyli
2020-01-23 17:01 ` [PATCH 12/17] bcache: avoid unnecessary btree nodes flushing in btree_flush_write() colyli
2020-01-23 17:01 ` [PATCH 13/17] bcache: print written and keys in trace_bcache_btree_write colyli
2020-01-23 17:01 ` [PATCH 14/17] bcache: back to cache all readahead I/Os colyli
2020-01-23 17:19   ` Michael Lyle
2020-01-23 17:27     ` Coly Li
2020-01-23 18:31       ` Jens Axboe
2020-01-24  0:49         ` Coly Li
2020-01-24  1:14           ` Jens Axboe
2020-01-24 16:48       ` Michael Lyle
2020-01-23 17:01 ` colyli [this message]
2020-01-23 17:01 ` [PATCH 16/17] bcache: reap c->btree_cache_freeable from the tail in bch_mca_scan() colyli
2020-01-23 17:01 ` [PATCH 17/17] bcache: reap from tail of c->btree_cache " colyli

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=20200123170142.98974-16-colyli@suse.de \
    --to=colyli@suse.de \
    --cc=axboe@kernel.dk \
    --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