From patchwork Fri Mar 25 06:39:20 2022 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Christoph Hellwig X-Patchwork-Id: 12791324 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 49DDEC433EF for ; Fri, 25 Mar 2022 06:40:02 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S233618AbiCYGle (ORCPT ); Fri, 25 Mar 2022 02:41:34 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:41068 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1358473AbiCYGld (ORCPT ); Fri, 25 Mar 2022 02:41:33 -0400 Received: from bombadil.infradead.org (bombadil.infradead.org [IPv6:2607:7c80:54:e::133]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id D08725A149 for ; Thu, 24 Mar 2022 23:39:59 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=infradead.org; s=bombadil.20210309; h=Content-Transfer-Encoding: MIME-Version:References:In-Reply-To:Message-Id:Date:Subject:Cc:To:From:Sender :Reply-To:Content-Type:Content-ID:Content-Description; bh=aYm7YxqfHnFVDKJqQbeIB0HPIdRThjCeZtVtfwljj/A=; b=eZ/7//0XIPwvhr3Qdf4iXMs6FQ 8VdPNy7agDLUIA4r7ExrVZlsS9zsEI6DrRKVme637SpNtZ/TljeHFscwbJSyyefAzFoTUNErtESYK b2DSJiBDc44JP/SmbLztNrCrFsd7lVYEkBgg6yGfXKeEkPebUlV76fcDLCTQ+iXd+jK27w8ZGmjpG VLdZ9lLPYmagarJzY7I17BjQS7mMxqnaaPM9hCdzJB98CkZn8SkkvmhaVvChgBvnKnM8qjfYzXzBQ htIVGXYLIUoWrbsy8rA13sldgwQgyREZcLozIMfVoPW5X+SrCxL+K08mA916h6mjBNezEt11hi784 +wUiy3CQ==; Received: from 089144194144.atnat0003.highway.a1.net ([89.144.194.144] helo=localhost) by bombadil.infradead.org with esmtpsa (Exim 4.94.2 #2 (Red Hat Linux)) id 1nXdc0-001HJM-Fs; Fri, 25 Mar 2022 06:39:53 +0000 From: Christoph Hellwig To: Jens Axboe , Josef Bacik , Minchan Kim , Nitin Gupta Cc: Tetsuo Handa , Jan Kara , "Darrick J . Wong" , Ming Lei , Matteo Croce , linux-block@vger.kernel.org, nbd@other.debian.org Subject: [PATCH 05/14] block: turn bdev->bd_openers into an atomic_t Date: Fri, 25 Mar 2022 07:39:20 +0100 Message-Id: <20220325063929.1773899-6-hch@lst.de> X-Mailer: git-send-email 2.30.2 In-Reply-To: <20220325063929.1773899-1-hch@lst.de> References: <20220325063929.1773899-1-hch@lst.de> MIME-Version: 1.0 X-SRS-Rewrite: SMTP reverse-path rewritten from by bombadil.infradead.org. See http://www.infradead.org/rpr.html Precedence: bulk List-ID: X-Mailing-List: linux-block@vger.kernel.org All manipulation of bd_openers is under disk->open_mutex and will remain so for the foreseeable future. But at least one place reads it without the lock (blkdev_get) and there are more to be added. So make sure the compiler does not do turn the increments and decrements into non-atomic sequences by using an atomic_t. Signed-off-by: Christoph Hellwig Reviewed-by: Jan Kara --- block/bdev.c | 16 ++++++++-------- block/partitions/core.c | 2 +- include/linux/blk_types.h | 2 +- include/linux/blkdev.h | 2 +- 4 files changed, 11 insertions(+), 11 deletions(-) diff --git a/block/bdev.c b/block/bdev.c index 13de871fa8169..7bf88e591aaf3 100644 --- a/block/bdev.c +++ b/block/bdev.c @@ -673,17 +673,17 @@ static int blkdev_get_whole(struct block_device *bdev, fmode_t mode) } } - if (!bdev->bd_openers) + if (!atomic_read(&bdev->bd_openers)) set_init_blocksize(bdev); if (test_bit(GD_NEED_PART_SCAN, &disk->state)) bdev_disk_changed(disk, false); - bdev->bd_openers++; + atomic_inc(&bdev->bd_openers); return 0; } static void blkdev_put_whole(struct block_device *bdev, fmode_t mode) { - if (!--bdev->bd_openers) + if (atomic_dec_and_test(&bdev->bd_openers)) blkdev_flush_mapping(bdev); if (bdev->bd_disk->fops->release) bdev->bd_disk->fops->release(bdev->bd_disk, mode); @@ -694,7 +694,7 @@ static int blkdev_get_part(struct block_device *part, fmode_t mode) struct gendisk *disk = part->bd_disk; int ret; - if (part->bd_openers) + if (atomic_read(&part->bd_openers)) goto done; ret = blkdev_get_whole(bdev_whole(part), mode); @@ -708,7 +708,7 @@ static int blkdev_get_part(struct block_device *part, fmode_t mode) disk->open_partitions++; set_init_blocksize(part); done: - part->bd_openers++; + atomic_inc(&part->bd_openers); return 0; out_blkdev_put: @@ -720,7 +720,7 @@ static void blkdev_put_part(struct block_device *part, fmode_t mode) { struct block_device *whole = bdev_whole(part); - if (--part->bd_openers) + if (!atomic_dec_and_test(&part->bd_openers)) return; blkdev_flush_mapping(part); whole->bd_disk->open_partitions--; @@ -899,7 +899,7 @@ void blkdev_put(struct block_device *bdev, fmode_t mode) * of the world and we want to avoid long (could be several minute) * syncs while holding the mutex. */ - if (bdev->bd_openers == 1) + if (atomic_read(&bdev->bd_openers) == 1) sync_blockdev(bdev); mutex_lock(&disk->open_mutex); @@ -1044,7 +1044,7 @@ void sync_bdevs(bool wait) bdev = I_BDEV(inode); mutex_lock(&bdev->bd_disk->open_mutex); - if (!bdev->bd_openers) { + if (!atomic_read(&bdev->bd_openers)) { ; /* skip */ } else if (wait) { /* diff --git a/block/partitions/core.c b/block/partitions/core.c index 2ef8dfa1e5c85..373ed748dcf26 100644 --- a/block/partitions/core.c +++ b/block/partitions/core.c @@ -486,7 +486,7 @@ int bdev_del_partition(struct gendisk *disk, int partno) goto out_unlock; ret = -EBUSY; - if (part->bd_openers) + if (atomic_read(&part->bd_openers)) goto out_unlock; delete_partition(part); diff --git a/include/linux/blk_types.h b/include/linux/blk_types.h index 0c3563b45fe90..b1ced43ed0d3f 100644 --- a/include/linux/blk_types.h +++ b/include/linux/blk_types.h @@ -44,7 +44,7 @@ struct block_device { unsigned long bd_stamp; bool bd_read_only; /* read-only policy */ dev_t bd_dev; - int bd_openers; + atomic_t bd_openers; struct inode * bd_inode; /* will die */ struct super_block * bd_super; void * bd_claiming; diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 9824ebc9b4d31..6b7c5af1d01df 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -188,7 +188,7 @@ static inline bool disk_live(struct gendisk *disk) */ static inline unsigned int disk_openers(struct gendisk *disk) { - return disk->part0->bd_openers; + return atomic_read(&disk->part0->bd_openers); } /*