From patchwork Thu Nov 1 07:58:13 2012 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Darrick Wong X-Patchwork-Id: 196092 Return-Path: X-Original-To: incoming@patchwork.ozlabs.org Delivered-To: patchwork-incoming@bilbo.ozlabs.org Received: from merlin.infradead.org (merlin.infradead.org [IPv6:2001:4978:20e::2]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (Client did not present a certificate) by ozlabs.org (Postfix) with ESMTPS id DD68B2C00A3 for ; Thu, 1 Nov 2012 19:23:41 +1100 (EST) Received: from localhost ([::1] helo=merlin.infradead.org) by merlin.infradead.org with esmtp (Exim 4.76 #1 (Red Hat Linux)) id 1TTq2o-0007ni-1Z; Thu, 01 Nov 2012 08:22:30 +0000 Received: from aserp1050.oracle.com ([141.146.126.70]) by merlin.infradead.org with esmtps (Exim 4.76 #1 (Red Hat Linux)) id 1TTq2k-0007nM-AS; Thu, 01 Nov 2012 08:22:28 +0000 Received: from aserp1040.oracle.com (aserp1040.oracle.com [141.146.126.69]) by aserp1050.oracle.com (Sentrion-MTA-4.2.2/Sentrion-MTA-4.2.2) with ESMTP id qA17wnuk005176 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK); Thu, 1 Nov 2012 07:58:49 GMT Received: from ucsinet21.oracle.com (ucsinet21.oracle.com [156.151.31.93]) by aserp1040.oracle.com (Sentrion-MTA-4.2.2/Sentrion-MTA-4.2.2) with ESMTP id qA17wJd1014018 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK); Thu, 1 Nov 2012 07:58:20 GMT Received: from acsmt358.oracle.com (acsmt358.oracle.com [141.146.40.158]) by ucsinet21.oracle.com (8.14.4+Sun/8.14.4) with ESMTP id qA17wHL4023016 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NO); Thu, 1 Nov 2012 07:58:17 GMT Received: from abhmt102.oracle.com (abhmt102.oracle.com [141.146.116.54]) by acsmt358.oracle.com (8.12.11.20060308/8.12.11) with ESMTP id qA17wGSO023163; Thu, 1 Nov 2012 02:58:16 -0500 Received: from localhost (/67.171.138.228) by default (Oracle Beehive Gateway v4.0) with ESMTP ; Thu, 01 Nov 2012 00:58:16 -0700 Subject: [PATCH 1/3] bdi: Track users that require stable page writes To: axboe@kernel.dk, lucho@ionkov.net, tytso@mit.edu, sage@inktank.com, darrick.wong@oracle.com, ericvh@gmail.com, mfasheh@suse.com, dedekind1@gmail.com, adrian.hunter@intel.com, dhowells@redhat.com, sfrench@samba.org, jlbec@evilplan.org, rminnich@sandia.gov From: "Darrick J. Wong" Date: Thu, 01 Nov 2012 00:58:13 -0700 Message-ID: <20121101075813.16153.94581.stgit@blackbox.djwong.org> In-Reply-To: <20121101075805.16153.64714.stgit@blackbox.djwong.org> References: <20121101075805.16153.64714.stgit@blackbox.djwong.org> User-Agent: StGit/0.15 MIME-Version: 1.0 X-Source-IP: aserp1040.oracle.com [141.146.126.69] X-CRM114-Version: 20100106-BlameMichelson ( TRE 0.8.0 (BSD) ) MR-646709E3 X-CRM114-CacheID: sfid-20121101_042226_526572_09146772 X-CRM114-Status: GOOD ( 16.07 ) X-Spam-Score: 0.4 (/) X-Spam-Report: SpamAssassin version 3.3.2 on merlin.infradead.org summary: Content analysis details: (0.4 points) pts rule name description ---- ---------------------- -------------------------------------------------- 3.0 KHOP_BIG_TO_CC Sent to 10+ recipients instaed of Bcc or a list -0.0 SPF_PASS SPF: sender matches SPF record -0.7 RP_MATCHES_RCVD Envelope sender domain matches handover relay domain -1.9 BAYES_00 BODY: Bayes spam probability is 0 to 1% [score: 0.0000] 0.0 UNPARSEABLE_RELAY Informational: message has unparseable relay lines Cc: linux-cifs@vger.kernel.org, jack@suse.cz, martin.petersen@oracle.com, neilb@suse.de, david@fromorbit.com, linux-kernel@vger.kernel.org, linux-mm@kvack.org, linux-mtd@lists.infradead.org, bharrosh@panasas.com, linux-fsdevel@vger.kernel.org, v9fs-developer@lists.sourceforge.net, ceph-devel@vger.kernel.org, linux-ext4@vger.kernel.org, linux-afs@lists.infradead.org, ocfs2-devel@oss.oracle.com X-BeenThere: linux-mtd@lists.infradead.org X-Mailman-Version: 2.1.14 Precedence: list List-Id: Linux MTD discussion mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: linux-mtd-bounces@lists.infradead.org Errors-To: linux-mtd-bounces+incoming=patchwork.ozlabs.org@lists.infradead.org This creates a per-backing-device counter that tracks the number of users which require pages to be held immutable during writeout. Eventually it will be used to waive wait_for_page_writeback() if nobody requires stable pages. Signed-off-by: Darrick J. Wong --- Documentation/ABI/testing/sysfs-class-bdi | 7 +++++ block/blk-integrity.c | 4 +++ include/linux/backing-dev.h | 16 ++++++++++++ include/linux/blkdev.h | 10 ++++++++ mm/backing-dev.c | 38 +++++++++++++++++++++++++++++ 5 files changed, 75 insertions(+) diff --git a/Documentation/ABI/testing/sysfs-class-bdi b/Documentation/ABI/testing/sysfs-class-bdi index 5f50097..218a618 100644 --- a/Documentation/ABI/testing/sysfs-class-bdi +++ b/Documentation/ABI/testing/sysfs-class-bdi @@ -48,3 +48,10 @@ max_ratio (read-write) most of the write-back cache. For example in case of an NFS mount that is prone to get stuck, or a FUSE mount which cannot be trusted to play fair. + +stable_pages_required (read-write) + + If set, the backing device requires that all pages comprising a write + request must not be changed until writeout is complete. The system + administrator can turn this on if the hardware does not do so already. + However, once enabled, this flag cannot be disabled. diff --git a/block/blk-integrity.c b/block/blk-integrity.c index da2a818..cf2dd95 100644 --- a/block/blk-integrity.c +++ b/block/blk-integrity.c @@ -420,6 +420,8 @@ int blk_integrity_register(struct gendisk *disk, struct blk_integrity *template) } else bi->name = bi_unsupported_name; + queue_require_stable_pages(disk->queue); + return 0; } EXPORT_SYMBOL(blk_integrity_register); @@ -438,6 +440,8 @@ void blk_integrity_unregister(struct gendisk *disk) if (!disk || !disk->integrity) return; + queue_unrequire_stable_pages(disk->queue); + bi = disk->integrity; kobject_uevent(&bi->kobj, KOBJ_REMOVE); diff --git a/include/linux/backing-dev.h b/include/linux/backing-dev.h index 2a9a9ab..0554f5d 100644 --- a/include/linux/backing-dev.h +++ b/include/linux/backing-dev.h @@ -109,6 +109,7 @@ struct backing_dev_info { struct dentry *debug_dir; struct dentry *debug_stats; #endif + atomic_t stable_page_users; }; int bdi_init(struct backing_dev_info *bdi); @@ -307,6 +308,21 @@ long wait_iff_congested(struct zone *zone, int sync, long timeout); int pdflush_proc_obsolete(struct ctl_table *table, int write, void __user *buffer, size_t *lenp, loff_t *ppos); +static inline void bdi_require_stable_pages(struct backing_dev_info *bdi) +{ + atomic_inc(&bdi->stable_page_users); +} + +static inline void bdi_unrequire_stable_pages(struct backing_dev_info *bdi) +{ + atomic_dec(&bdi->stable_page_users); +} + +static inline bool bdi_cap_stable_pages_required(struct backing_dev_info *bdi) +{ + return atomic_read(&bdi->stable_page_users) > 0; +} + static inline bool bdi_cap_writeback_dirty(struct backing_dev_info *bdi) { return !(bdi->capabilities & BDI_CAP_NO_WRITEBACK); diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index 1756001..bf927c0 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -458,6 +458,16 @@ struct request_queue { (1 << QUEUE_FLAG_SAME_COMP) | \ (1 << QUEUE_FLAG_ADD_RANDOM)) +static inline void queue_require_stable_pages(struct request_queue *q) +{ + bdi_require_stable_pages(&q->backing_dev_info); +} + +static inline void queue_unrequire_stable_pages(struct request_queue *q) +{ + bdi_unrequire_stable_pages(&q->backing_dev_info); +} + static inline void queue_lockdep_assert_held(struct request_queue *q) { if (q->queue_lock) diff --git a/mm/backing-dev.c b/mm/backing-dev.c index d3ca2b3..dd9f5ed 100644 --- a/mm/backing-dev.c +++ b/mm/backing-dev.c @@ -221,12 +221,48 @@ static ssize_t max_ratio_store(struct device *dev, } BDI_SHOW(max_ratio, bdi->max_ratio) +static ssize_t stable_pages_required_store(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + struct backing_dev_info *bdi = dev_get_drvdata(dev); + unsigned int spw; + ssize_t ret; + + ret = kstrtouint(buf, 10, &spw); + if (ret < 0) + return ret; + + /* + * SPW could be enabled due to hw requirement, so don't + * let users disable it. + */ + if (bdi_cap_stable_pages_required(bdi) && spw == 0) + return -EINVAL; + + if (spw != 0) + atomic_inc(&bdi->stable_page_users); + + return count; +} + +static ssize_t stable_pages_required_show(struct device *dev, + struct device_attribute *attr, + char *page) +{ + struct backing_dev_info *bdi = dev_get_drvdata(dev); + + return snprintf(page, PAGE_SIZE-1, "%d\n", + bdi_cap_stable_pages_required(bdi) ? 1 : 0); +} + #define __ATTR_RW(attr) __ATTR(attr, 0644, attr##_show, attr##_store) static struct device_attribute bdi_dev_attrs[] = { __ATTR_RW(read_ahead_kb), __ATTR_RW(min_ratio), __ATTR_RW(max_ratio), + __ATTR_RW(stable_pages_required), __ATTR_NULL, }; @@ -650,6 +686,8 @@ int bdi_init(struct backing_dev_info *bdi) bdi->write_bandwidth = INIT_BW; bdi->avg_write_bandwidth = INIT_BW; + atomic_set(&bdi->stable_page_users, 0); + err = fprop_local_init_percpu(&bdi->completions); if (err) {