linux-fsdevel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Al Viro <viro@zeniv.linux.org.uk>
To: Christoph Hellwig <hch@lst.de>
Cc: linux-fsdevel@vger.kernel.org, Yu Kuai <yukuai1@huaweicloud.com>,
	linux-block@vger.kernel.org,
	Christian Brauner <brauner@kernel.org>,
	Jens Axboe <axboe@kernel.dk>
Subject: Re: [PATCHES][RFC] packing struct block_device flags
Date: Mon, 29 Apr 2024 18:02:09 +0100	[thread overview]
Message-ID: <20240429170209.GA2118490@ZenIV> (raw)
In-Reply-To: <20240429073107.GZ2118490@ZenIV>

On Mon, Apr 29, 2024 at 08:31:07AM +0100, Al Viro wrote:

> FWIW, we could go for atomic_t there and use
> 	atomic_read() & 0xff
> for partno, with atomic_or()/atomic_and() for set/clear and
> atomic_read() & constant for test.  That might slightly optimize
> set/clear on some architectures, but setting/clearing flags is
> nowhere near hot enough for that to make a difference.

Incremental for that (would be folded into 3/8 if we went that way)
is below; again, I'm not at all sure it's idiomatic enough to bother
with, but that should at least show what's going on:

diff --git a/block/bdev.c b/block/bdev.c
index 9aa23620fe92..fae30eae7741 100644
--- a/block/bdev.c
+++ b/block/bdev.c
@@ -411,7 +411,7 @@ struct block_device *bdev_alloc(struct gendisk *disk, u8 partno)
 	mutex_init(&bdev->bd_fsfreeze_mutex);
 	spin_lock_init(&bdev->bd_size_lock);
 	mutex_init(&bdev->bd_holder_lock);
-	bdev->__bd_flags = partno;
+	atomic_set(&bdev->__bd_flags, partno);
 	bdev->bd_inode = inode;
 	bdev->bd_queue = disk->queue;
 	if (partno && bdev_test_flag(disk->part0, BD_HAS_SUBMIT_BIO))
diff --git a/include/linux/blk_types.h b/include/linux/blk_types.h
index 98e1c2d28d60..a822911e28e5 100644
--- a/include/linux/blk_types.h
+++ b/include/linux/blk_types.h
@@ -45,7 +45,7 @@ struct block_device {
 	struct request_queue *	bd_queue;
 	struct disk_stats __percpu *bd_stats;
 	unsigned long		bd_stamp;
-	u32			__bd_flags;	// partition number + flags
+	atomic_t		__bd_flags;	// partition number + flags
 	dev_t			bd_dev;
 	struct inode		*bd_inode;	/* will die */
 
diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index d556cec9224b..a8271497ac62 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -722,38 +722,22 @@ void disk_uevent(struct gendisk *disk, enum kobject_action action);
 
 static inline u8 bdev_partno(const struct block_device *bdev)
 {
-	return bdev->__bd_flags & 0xff;
+	return atomic_read(&bdev->__bd_flags) & 0xff;
 }
 
 static inline bool bdev_test_flag(const struct block_device *bdev, int flag)
 {
-	return bdev->__bd_flags & (1 << (flag + 8));
+	return atomic_read(&bdev->__bd_flags) & (1 << (flag + 8));
 }
 
 static inline void bdev_set_flag(struct block_device *bdev, int flag)
 {
-	u32 v = bdev->__bd_flags;
-
-	for (;;) {
-		u32 w = cmpxchg(&bdev->__bd_flags, v, v | (1 << (flag + 8)));
-
-		if (v == w)
-			return;
-		v = w;
-	}
+	atomic_or(1 << (flag + 8), &bdev->__bd_flags);
 }
 
 static inline void bdev_clear_flag(struct block_device *bdev, int flag)
 {
-	u32 v = bdev->__bd_flags;
-
-	for (;;) {
-		u32 w = cmpxchg(&bdev->__bd_flags, v, v & ~(1 << (flag + 8)));
-
-		if (v == w)
-			return;
-		v = w;
-	}
+	atomic_and(~(1 << (flag + 8)), &bdev->__bd_flags);
 }
 
 static inline int get_disk_ro(struct gendisk *disk)

  reply	other threads:[~2024-04-29 17:02 UTC|newest]

Thread overview: 24+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-04-28  5:12 [PATCHES][RFC] packing struct block_device flags Al Viro
2024-04-28  5:14 ` [PATCH 1/8] Use bdev_is_paritition() instead of open-coding it Al Viro
2024-04-29  5:19   ` Christoph Hellwig
2024-04-28  5:15 ` [PATCH 2/8] wrapper for access to ->bd_partno Al Viro
2024-04-28  5:16 ` [PATCH 3/8] bdev: infrastructure for flags Al Viro
2024-04-28  5:17 ` [PATCH 4/8] bdev: move ->bd_read_only to ->__bd_flags Al Viro
2024-04-28  5:18 ` [PATCH 5/8] bdev: move ->bd_write_holder into ->__bd_flags Al Viro
2024-04-28  5:19 ` [PATCH 6/8] bdev: move ->bd_has_subit_bio to ->__bd_flags Al Viro
2024-04-28  5:19 ` [PATCH 7/8] bdev: move ->bd_ro_warned " Al Viro
2024-04-28  5:21 ` [PATCH 8/8] bdev: move ->bd_make_it_fail " Al Viro
2024-04-29  5:23 ` [PATCHES][RFC] packing struct block_device flags Christoph Hellwig
2024-04-29  7:31   ` Al Viro
2024-04-29 17:02     ` Al Viro [this message]
2024-04-29 18:13       ` Al Viro
2024-04-29 18:30         ` Al Viro
2024-05-03  0:06           ` Al Viro
2024-05-03  0:07             ` [PATCH 1/8] Use bdev_is_paritition() instead of open-coding it Al Viro
2024-05-03  0:07             ` [PATCH 2/8] wrapper for access to ->bd_partno Al Viro
2024-05-03  0:08             ` [PATCH v2 3/8] bdev: infrastructure for flags Al Viro
2024-05-03  0:09             ` [PATCH v2 4/8] bdev: move ->bd_read_only to ->__bd_flags Al Viro
2024-05-03  0:09             ` [PATCH v2 5/8] bdev: move ->bd_write_holder into ->__bd_flags Al Viro
2024-05-03  0:10             ` [PATCH v2 6/8] bdev: move ->bd_has_subit_bio to ->__bd_flags Al Viro
2024-05-03  0:10             ` [PATCH v2 7/8] bdev: move ->bd_ro_warned " Al Viro
2024-05-03  0:11             ` [PATCH v2 8/8] bdev: move ->bd_make_it_fail " Al Viro

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=20240429170209.GA2118490@ZenIV \
    --to=viro@zeniv.linux.org.uk \
    --cc=axboe@kernel.dk \
    --cc=brauner@kernel.org \
    --cc=hch@lst.de \
    --cc=linux-block@vger.kernel.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=yukuai1@huaweicloud.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 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).