All of lore.kernel.org
 help / color / mirror / Atom feed
From: Jens Axboe <axboe@kernel.dk>
To: linux-block@vger.kernel.org, osandov@osandov.com
Cc: Jens Axboe <axboe@kernel.dk>
Subject: [PATCH 1/3] sbitmap: ensure that sbitmap maps are properly aligned
Date: Thu, 29 Nov 2018 18:12:32 -0700	[thread overview]
Message-ID: <20181130011234.32674-2-axboe@kernel.dk> (raw)
In-Reply-To: <20181130011234.32674-1-axboe@kernel.dk>

We try to be careful with alignment for cache purposes, but all of
that is worthless if we don't actually align the maps themselves.

Signed-off-by: Jens Axboe <axboe@kernel.dk>
---
 include/linux/sbitmap.h | 11 ++++++++---
 lib/sbitmap.c           |  7 +++++--
 2 files changed, 13 insertions(+), 5 deletions(-)

diff --git a/include/linux/sbitmap.h b/include/linux/sbitmap.h
index 804a50983ec5..5cb1755d32da 100644
--- a/include/linux/sbitmap.h
+++ b/include/linux/sbitmap.h
@@ -63,9 +63,14 @@ struct sbitmap {
 	unsigned int map_nr;
 
 	/**
-	 * @map: Allocated bitmap.
+	 * @map: Aligned allocated bitmap.
 	 */
 	struct sbitmap_word *map;
+
+	/**
+	 * @map_ptr: Originally allocated map pointer
+	 */
+	void *map_ptr;
 };
 
 #define SBQ_WAIT_QUEUES 8
@@ -157,8 +162,8 @@ int sbitmap_init_node(struct sbitmap *sb, unsigned int depth, int shift,
  */
 static inline void sbitmap_free(struct sbitmap *sb)
 {
-	kfree(sb->map);
-	sb->map = NULL;
+	kfree(sb->map_ptr);
+	sb->map_ptr = sb->map = NULL;
 }
 
 /**
diff --git a/lib/sbitmap.c b/lib/sbitmap.c
index 45cab6bbc1c7..21e776e3128d 100644
--- a/lib/sbitmap.c
+++ b/lib/sbitmap.c
@@ -25,6 +25,7 @@ int sbitmap_init_node(struct sbitmap *sb, unsigned int depth, int shift,
 {
 	unsigned int bits_per_word;
 	unsigned int i;
+	size_t size;
 
 	if (shift < 0) {
 		shift = ilog2(BITS_PER_LONG);
@@ -52,9 +53,11 @@ int sbitmap_init_node(struct sbitmap *sb, unsigned int depth, int shift,
 		return 0;
 	}
 
-	sb->map = kcalloc_node(sb->map_nr, sizeof(*sb->map), flags, node);
-	if (!sb->map)
+	size = sb->map_nr * sizeof(*sb->map) + L1_CACHE_BYTES - 1;
+	sb->map_ptr = kzalloc_node(size, flags, node);
+	if (!sb->map_ptr)
 		return -ENOMEM;
+	sb->map = PTR_ALIGN(sb->map_ptr, L1_CACHE_BYTES);
 
 	for (i = 0; i < sb->map_nr; i++) {
 		sb->map[i].depth = min(depth, bits_per_word);
-- 
2.17.1


       reply	other threads:[~2018-11-30  1:12 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <20181130011234.32674-1-axboe@kernel.dk>
2018-11-30  1:12 ` Jens Axboe [this message]
2018-11-30  1:12 ` [PATCH 2/3] sbitmap: ammortize cost of clearing bits Jens Axboe
2018-12-09  5:51   ` Guenter Roeck
2018-12-09  6:19     ` Jens Axboe
2018-11-30  1:12 ` [PATCH 3/3] sbitmap: optimize wakeup check Jens Axboe
2018-11-30  2:09 ` 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=20181130011234.32674-2-axboe@kernel.dk \
    --to=axboe@kernel.dk \
    --cc=linux-block@vger.kernel.org \
    --cc=osandov@osandov.com \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.