From patchwork Fri Mar 20 14:50:37 2015 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Josef Bacik X-Patchwork-Id: 6057401 Return-Path: X-Original-To: patchwork-linux-btrfs@patchwork.kernel.org Delivered-To: patchwork-parsemail@patchwork2.web.kernel.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.136]) by patchwork2.web.kernel.org (Postfix) with ESMTP id 1C582BF90F for ; Fri, 20 Mar 2015 14:51:03 +0000 (UTC) Received: from mail.kernel.org (localhost [127.0.0.1]) by mail.kernel.org (Postfix) with ESMTP id C781A20515 for ; Fri, 20 Mar 2015 14:51:00 +0000 (UTC) Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.kernel.org (Postfix) with ESMTP id 561AD20444 for ; Fri, 20 Mar 2015 14:50:58 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751699AbbCTOut (ORCPT ); Fri, 20 Mar 2015 10:50:49 -0400 Received: from mx0a-00082601.pphosted.com ([67.231.145.42]:49022 "EHLO mx0a-00082601.pphosted.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750823AbbCTOur (ORCPT ); Fri, 20 Mar 2015 10:50:47 -0400 Received: from pps.filterd (m0044010 [127.0.0.1]) by mx0a-00082601.pphosted.com (8.14.5/8.14.5) with SMTP id t2KEoGV9018006; Fri, 20 Mar 2015 07:50:41 -0700 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=fb.com; h=from : to : subject : date : message-id : in-reply-to : references : mime-version : content-type; s=facebook; bh=80+BXm6gxa9CLh8XU/9CjbfFIJHXj0KiD/G4/+UOX1E=; b=OZIGwSYQmhWMfvhGttJbN2V+9A4OR7GXuCT41g/melCfQ21+Y0WRrC0AZJfD5q3tTw07 zauoDFfEFiBejOsGBtgw2GW581FLUSfgvozvAaAIrfKuqStnDpTKEqNr4ryChgaYIQ6h TdfZGvhnKr4b4eXMZjf+3T1TP3YqVsQ8VSc= Received: from mail.thefacebook.com ([199.201.64.23]) by mx0a-00082601.pphosted.com with ESMTP id 1t8n9a81cg-1 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=NOT); Fri, 20 Mar 2015 07:50:41 -0700 Received: from localhost (192.168.54.13) by mail.TheFacebook.com (192.168.16.18) with Microsoft SMTP Server (TLS) id 14.3.195.1; Fri, 20 Mar 2015 07:50:39 -0700 From: Josef Bacik To: , , , , Subject: [PATCH 1/3] dm: log writes target V2 Date: Fri, 20 Mar 2015 10:50:37 -0400 Message-ID: <1426863037-5870-1-git-send-email-jbacik@fb.com> X-Mailer: git-send-email 1.8.3.1 In-Reply-To: <20150319231655.GA6751@lenny.home.zabbo.net> References: <20150319231655.GA6751@lenny.home.zabbo.net> MIME-Version: 1.0 X-Originating-IP: [192.168.54.13] X-Proofpoint-Virus-Version: vendor=fsecure engine=2.50.10432:5.13.68, 1.0.33, 0.0.0000 definitions=2015-03-20_05:2015-03-20, 2015-03-20, 1970-01-01 signatures=0 X-Proofpoint-Spam-Details: rule=fb_default_notspam policy=fb_default score=0 spamscore=0 suspectscore=1 phishscore=0 adultscore=0 bulkscore=0 classifier=spam adjust=0 reason=mlx scancount=1 engine=7.0.1-1402240000 definitions=main-1503200152 X-FB-Internal: deliver Sender: linux-btrfs-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-btrfs@vger.kernel.org X-Spam-Status: No, score=-6.8 required=5.0 tests=BAYES_00,DKIM_SIGNED, RCVD_IN_DNSWL_HI,T_DKIM_INVALID,T_RP_MATCHES_RCVD,UNPARSEABLE_RELAY autolearn=unavailable version=3.3.1 X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on mail.kernel.org X-Virus-Scanned: ClamAV using ClamSMTP This creates a new target that is meant for file system developers to test file system integrity at particular points in the life of a file system. We capture all write requests and the data and log the requests and the data to a separate device for later replay. There is a userspace utility to do this replay. The idea behind this is to give file system developers to verify that the file system is always consistent. Thanks, Signed-off-by: Josef Bacik Reviewed-by: Zach Brown --- V1->V2: fixed up stuff based on Zachs review. Documentation/device-mapper/dm-log-writes.txt | 136 +++++ drivers/md/Kconfig | 16 + drivers/md/Makefile | 1 + drivers/md/dm-log-writes.c | 826 ++++++++++++++++++++++++++ 4 files changed, 979 insertions(+) create mode 100644 Documentation/device-mapper/dm-log-writes.txt create mode 100644 drivers/md/dm-log-writes.c diff --git a/Documentation/device-mapper/dm-log-writes.txt b/Documentation/device-mapper/dm-log-writes.txt new file mode 100644 index 0000000..f3a9fa2 --- /dev/null +++ b/Documentation/device-mapper/dm-log-writes.txt @@ -0,0 +1,136 @@ +dm-log-writes +============= + +This target takes 2 devices, one to pass all IO to normally, and one to log all +of the write operations to. This is intended for file system developers wishing +to verify the integrity of metadata or data as the file system is written to. +There is a log_writes_entry written for every WRITE request and the target is +able to take arbitrary data from userspace to insert into the log. The data +that is in the WRITE requests is copied into the log to make the replay happen +exactly as it happened originally. + +Log Ordering +============ + +We log things in order of completion once we are sure the write is no longer in +cache. This means that normal WRITE requests are not actually logged until the +next REQ_FLUSH request. This is to make it easier for userspace to replay the +log in a way that correlates to what is on disk and not what is in cache, to +make it easier to detect improper waiting/flushing. + +This works by attaching all WRITE requests to a list once the write completes. +Once we see a REQ_FLUSH request we splice this list onto the request and once +the FLUSH request completes we log all of the WRITE's and then the FLUSH. Only +completeled WRITEs at the time of the issue of the REQ_FLUSH are added in order +to simulate the worst case scenario with regard to power failures. Consider the +following example (W means write, C means complete) + +W1,W2,W3,C3,C2,Wflush,C1,Cflush + +The log would show the following + +W3,W2,flush,W1.... + +Again this is to simulate what is actually on disk, this allows us to detect +cases where a power failure at a particular point in time would create an +inconsistent file system. + +Any REQ_FUA requests bypass this flushing mechanism and are logged as soon as +they complete as those requests will obviously bypass the device cache. + +Any REQ_DISCARD requests are treated like WRITE requests. This is because +otherwise we would have all the DISCARD requests, and then the WRITE requests +and then the FLUSH request. Consider the following example + +WRITE block 1, DISCARD block 1, FLUSH + +If we logged DISCARD when it completed, the replay would look like this + +DISCARD 1, WRITE 1, FLUSH + +which isn't quite what happened and wouldn't be caught during the log replay. + +Marks +===== + +You can use dmsetup to set an arbitrary mark in a log. For example say you want +to fsck an file system after every write, but first you need to replay up to the +mkfs to make sure we're fsck'ing something reasonable, you would do something +like this + +mkfs.btrfs -f /dev/mapper/log +dmsetup message log 0 mark mkfs + + +This would allow you to replay the log up to the mkfs mark and then replay from +that point on doing the fsck check in the interval that you want. + +Every log has a mark at the end labeled "log-writes-end". + +Userspace component +=================== + +There is a userspace tool that will replay the log for you in various ways. +As of this writing the options are not well documented, they will be in the +future. It can be found here + +https://github.com/josefbacik/log-writes + +Example usage +============= + +Say you want to test fsync on your file system. You would do something like +this + +TABLE="0 $(blockdev --getsz /dev/sdb) log-writes /dev/sdb /dev/sdc" +dmsetup create log --table "$TABLE" +mkfs.btrfs -f /dev/mapper/log +dmsetup message log 0 mark mkfs + +mount /dev/mapper/log /mnt/btrfs-test + +dmsetup message log 0 mark fsync +md5sum /mnt/btrfs-test/foo +umount /mnt/btrfs-test + +dmsetup remove log +replay-log --log /dev/sdc --replay /dev/sdb --end-mark fsync +mount /dev/sdb /mnt/btrfs-test +md5sum /mnt/btrfs-test/foo + + +Another option is to do a complicated file system operation and verify the file +system is consistent during the entire operation. You could do this by doing + + +TABLE="0 $(blockdev --getsz /dev/sdb) log-writes /dev/sdb /dev/sdc" +dmsetup create log --table "$TABLE" +mkfs.btrfs -f /dev/mapper/log +dmsetup message log 0 mark mkfs + +mount /dev/mapper/log /mnt/btrfs-test + +btrfs filesystem balance /mnt/btrfs-test +umount /mnt/btrfs-test +dmsetup remove log + +replay-log --log /dev/sdc --replay /dev/sdb --end-mark mkfs +btrfsck /dev/sdb +replay-log --log /dev/sdc --replay /dev/sdb --start-mark mkfs \ + --fsck "btrfsck /dev/sdb" --check fua + +And that will replay the log until it sees a FUA request, run the fsck command +and if the fsck passes it will replay to the next FUA, until it is completed or +the fsck command exists abnormally. + +Table Parameters +---------------- + + +Mandatory parameters: + : Full pathname to the underlying block-device, or a "major:minor" + device-number. This device is the one that all of the IO will go + to normally, just think of it as a normal linear mapping. + : Same format as , this is the device where the + log entries are written to. + diff --git a/drivers/md/Kconfig b/drivers/md/Kconfig index 63e05e3..f928ad5 100644 --- a/drivers/md/Kconfig +++ b/drivers/md/Kconfig @@ -432,4 +432,20 @@ config DM_SWITCH If unsure, say N. +config DM_LOG_WRITES + tristate "Log writes target support" + depends on BLK_DEV_DM + ---help--- + This device-mapper target takes two devices, one device to use + normally, one to log all write operations done to the first device. + This is for use by file system developers wishing to verify that + their fs is writing a consitent file system at all times by allowing + them to replay the log in a variety of ways and to check the + contents. + + To compile this code as a module, choose M here: the module will + be called dm-log-writes. + + If unsure, say N. + endif # MD diff --git a/drivers/md/Makefile b/drivers/md/Makefile index a2da532..1863fea 100644 --- a/drivers/md/Makefile +++ b/drivers/md/Makefile @@ -55,6 +55,7 @@ obj-$(CONFIG_DM_CACHE) += dm-cache.o obj-$(CONFIG_DM_CACHE_MQ) += dm-cache-mq.o obj-$(CONFIG_DM_CACHE_CLEANER) += dm-cache-cleaner.o obj-$(CONFIG_DM_ERA) += dm-era.o +obj-$(CONFIG_DM_LOG_WRITES) += dm-log-writes.o ifeq ($(CONFIG_DM_UEVENT),y) dm-mod-objs += dm-uevent.o diff --git a/drivers/md/dm-log-writes.c b/drivers/md/dm-log-writes.c new file mode 100644 index 0000000..067cb07 --- /dev/null +++ b/drivers/md/dm-log-writes.c @@ -0,0 +1,826 @@ +/* + * Copyright (C) 2014 Facebook. All rights reserved. + * + * This file is released under the GPL. + */ + +#include + +#include +#include +#include +#include +#include +#include +#include + +#define DM_MSG_PREFIX "log-writes" + +/* + * This target will log sequentially all writes to the target device onto the + * log device. This is helpful for replaying writes to check for fs consitency + * at all times. This target provides a mechanism to mark specific events to + * check data at a later time. So for example you would + * + * write data + * fsync + * dmsetup message /dev/whatever mark mymark + * unmount /mnt/test + * + * Then replay the log up to mymark and check the contents of the replay to + * verify it matches what was written. + * + * We log writes only after they have been flushed, this makes the log describe + * close to the order in which the data hits the actual disk, not its cache. So + * for example the following sequence (W means write, C means complete) + * + * Wa,Wb,Wc,Cc,Ca,FLUSH,FUAd,Cb,CFLUSH,CFUAd + * + * Would result in the log looking like this + * + * c,a,flush,fuad,b,, + * + * This is meant to help expose problems where file systems do not properly wait + * on data being written before invoking a FLUSH. FUA bypasses cache so once it + * completes it is added to the log as it should be on disk. + * + * We treat DISCARDs as if they don't bypass cache so that they are logged in + * order of completion along with the normal writes. If we didn't do it this + * way we would process all the discards first and then write all the data, when + * in fact we want to do the data and the discard in the order that they + * completed. + */ +#define LOG_FLUSH_FLAG (1 << 0) +#define LOG_FUA_FLAG (1 << 1) +#define LOG_DISCARD_FLAG (1 << 2) +#define LOG_MARK_FLAG (1 << 3) + +#define WRITE_LOG_VERSION 1 +#define WRITE_LOG_MAGIC 0x6a736677736872 + +/* + * The disk format for this is braindead simple. + * + * At byte 0 we have our super, followed by the following sequence for + * nr_entries + * + * [ 1 sector ][ entry->nr_sectors ] + * [log_write_entry][ data written ] + * + * The log_write_entry takes up a full sector so we can have arbitrary length + * marks and it leaves us room for extra content in the future. + */ + +/* + * Basic info about the log for userspace. + */ +struct log_write_super { + __le64 magic; + __le64 version; + __le64 nr_entries; + __le32 sectorsize; +}; + +/* + * sector - the sector we wrote. + * nr_sectors - the number of sectors we wrote. + * flags - flags for this log entry. + * data_len - the size of the data in this log entry, this is for private log + * entry stuff, the MARK data provided by userspace for example. + */ +struct log_write_entry { + __le64 sector; + __le64 nr_sectors; + __le64 flags; + __le64 data_len; +}; + +struct log_writes_c { + struct dm_dev *dev; + struct dm_dev *logdev; + u64 logged_entries; + u32 sectorsize; + atomic_t io_blocks; + atomic_t pending_blocks; + sector_t next_sector; + sector_t end_sector; + bool logging_enabled; + bool device_supports_discard; + spinlock_t blocks_lock; + struct list_head unflushed_blocks; + struct list_head logging_blocks; + wait_queue_head_t wait; + struct task_struct *log_kthread; +}; + +struct pending_block { + int vec_cnt; + u64 flags; + sector_t sector; + sector_t nr_sectors; + char *data; + u32 datalen; + struct list_head list; + struct bio_vec vecs[0]; +}; + +struct per_bio_data { + struct pending_block *block; +}; + +static void put_pending_block(struct log_writes_c *lc) +{ + if (atomic_dec_and_test(&lc->pending_blocks)) { + smp_mb__after_atomic(); + if (waitqueue_active(&lc->wait)) + wake_up(&lc->wait); + } +} + +static void put_io_block(struct log_writes_c *lc) +{ + if (atomic_dec_and_test(&lc->io_blocks)) { + smp_mb__after_atomic(); + if (waitqueue_active(&lc->wait)) + wake_up(&lc->wait); + } +} + +static void log_end_io(struct bio *bio, int err) +{ + struct log_writes_c *lc = bio->bi_private; + struct bio_vec *bvec; + int i; + + if (err) { + unsigned long flags; + + DMERR("Error writing log block %d", err); + spin_lock_irqsave(&lc->blocks_lock, flags); + lc->logging_enabled = false; + spin_unlock_irqrestore(&lc->blocks_lock, flags); + } + + bio_for_each_segment_all(bvec, bio, i) + __free_page(bvec->bv_page); + + put_io_block(lc); + bio_put(bio); +} + +/* + * Meant to be called if there is an error, it will free all the pages + * associated with the block. + */ +static void free_pending_block(struct log_writes_c *lc, + struct pending_block *block) +{ + int i; + + for (i = 0; i < block->vec_cnt; i++) { + if (block->vecs[i].bv_page) + __free_page(block->vecs[i].bv_page); + } + kfree(block->data); + kfree(block); + put_pending_block(lc); +} + +static int write_metadata(struct log_writes_c *lc, void *entry, + size_t entrylen, void *data, size_t datalen, + sector_t sector) +{ + struct bio *bio; + struct page *page; + void *ptr; + size_t ret; + + bio = bio_alloc(GFP_KERNEL, 1); + if (!bio) { + DMERR("Couldn't alloc log bio"); + goto error; + } + bio->bi_iter.bi_size = 0; + bio->bi_iter.bi_sector = sector; + bio->bi_bdev = lc->logdev->bdev; + bio->bi_end_io = log_end_io; + bio->bi_private = lc; + set_bit(BIO_UPTODATE, &bio->bi_flags); + + page = alloc_page(GFP_KERNEL); + if (!page) { + DMERR("Couldn't alloc log page"); + bio_put(bio); + goto error; + } + + ptr = kmap_atomic(page); + memcpy(ptr, entry, entrylen); + if (datalen) + memcpy(ptr + entrylen, data, datalen); + memset(ptr + entrylen + datalen, 0, + lc->sectorsize - entrylen - datalen); + kunmap_atomic(ptr); + + ret = bio_add_page(bio, page, lc->sectorsize, 0); + if (ret != lc->sectorsize) { + DMERR("Couldn't add page to the log block"); + goto error_bio; + } + submit_bio(WRITE, bio); + return 0; +error_bio: + bio_put(bio); + __free_page(page); +error: + put_io_block(lc); + return -1; +} + +static int log_one_block(struct log_writes_c *lc, + struct pending_block *block, sector_t sector) +{ + struct bio *bio; + struct log_write_entry entry; + size_t ret; + int i; + + entry.sector = cpu_to_le64(block->sector); + entry.nr_sectors = cpu_to_le64(block->nr_sectors); + entry.flags = cpu_to_le64(block->flags); + entry.data_len = cpu_to_le64(block->datalen); + if (write_metadata(lc, &entry, sizeof(entry), block->data, + block->datalen, sector)) { + free_pending_block(lc, block); + return -1; + } + + if (!block->vec_cnt) + goto out; + sector++; + + bio = bio_alloc(GFP_KERNEL, block->vec_cnt); + if (!bio) { + DMERR("Couldn't alloc log bio"); + goto error; + } + atomic_inc(&lc->io_blocks); + bio->bi_iter.bi_size = 0; + bio->bi_iter.bi_sector = sector; + bio->bi_bdev = lc->logdev->bdev; + bio->bi_end_io = log_end_io; + bio->bi_private = lc; + set_bit(BIO_UPTODATE, &bio->bi_flags); + + for (i = 0; i < block->vec_cnt; i++) { + /* + * The page offset is always 0 because we allocate a new page + * for every bvec in the original bio for simplicity sake. + */ + ret = bio_add_page(bio, block->vecs[i].bv_page, + block->vecs[i].bv_len, 0); + if (ret != block->vecs[i].bv_len) { + atomic_inc(&lc->io_blocks); + submit_bio(WRITE, bio); + bio = bio_alloc(GFP_KERNEL, block->vec_cnt - i); + if (!bio) { + DMERR("Couldn't alloc log bio"); + goto error; + } + bio->bi_iter.bi_size = 0; + bio->bi_iter.bi_sector = sector; + bio->bi_bdev = lc->logdev->bdev; + bio->bi_end_io = log_end_io; + bio->bi_private = lc; + set_bit(BIO_UPTODATE, &bio->bi_flags); + + ret = bio_add_page(bio, block->vecs[i].bv_page, + block->vecs[i].bv_len, 0); + if (ret != block->vecs[i].bv_len) { + DMERR("Couldn't add page on new bio?"); + bio_put(bio); + goto error; + } + } + sector += block->vecs[i].bv_len >> SECTOR_SHIFT; + } + submit_bio(WRITE, bio); +out: + kfree(block->data); + kfree(block); + put_pending_block(lc); + return 0; +error: + free_pending_block(lc, block); + put_io_block(lc); + return -1; +} + +static int log_super(struct log_writes_c *lc) +{ + struct log_write_super super; + + super.magic = cpu_to_le64(WRITE_LOG_MAGIC); + super.version = cpu_to_le64(WRITE_LOG_VERSION); + super.nr_entries = cpu_to_le64(lc->logged_entries); + super.sectorsize = cpu_to_le32(lc->sectorsize); + + if (write_metadata(lc, &super, sizeof(super), NULL, 0, 0)) { + DMERR("Couldn't write super"); + return -1; + } + + return 0; +} + +static inline sector_t logdev_last_sector(struct log_writes_c *lc) +{ + return i_size_read(lc->logdev->bdev->bd_inode) >> SECTOR_SHIFT; +} + +static int log_writes_kthread(void *arg) +{ + struct log_writes_c *lc = (struct log_writes_c *)arg; + sector_t sector = 0; + + while (!kthread_should_stop()) { + bool super = false; + bool logging_enabled; + struct pending_block *block = NULL; + int ret; + + spin_lock_irq(&lc->blocks_lock); + if (!list_empty(&lc->logging_blocks)) { + block = list_first_entry(&lc->logging_blocks, + struct pending_block, list); + list_del_init(&block->list); + if (!lc->logging_enabled) + goto next; + + sector = lc->next_sector; + if (block->flags & LOG_DISCARD_FLAG) + lc->next_sector++; + else + lc->next_sector += block->nr_sectors + 1; + + /* + * Apparently the size of the device may not be known + * right away, so handle this properly. + */ + if (!lc->end_sector) + lc->end_sector = logdev_last_sector(lc); + if (lc->end_sector && + lc->next_sector >= lc->end_sector) { + DMERR("Ran out of space on the logdev"); + lc->logging_enabled = false; + goto next; + } + lc->logged_entries++; + atomic_inc(&lc->io_blocks); + + super = (block->flags & (LOG_FUA_FLAG | LOG_MARK_FLAG)); + if (super) + atomic_inc(&lc->io_blocks); + } +next: + logging_enabled = lc->logging_enabled; + spin_unlock_irq(&lc->blocks_lock); + if (block) { + if (logging_enabled) { + ret = log_one_block(lc, block, sector); + if (!ret && super) + ret = log_super(lc); + if (ret) { + spin_lock_irq(&lc->blocks_lock); + lc->logging_enabled = false; + spin_unlock_irq(&lc->blocks_lock); + } + } else + free_pending_block(lc, block); + continue; + } + + if (!try_to_freeze()) { + set_current_state(TASK_INTERRUPTIBLE); + if (!kthread_should_stop() && + !atomic_read(&lc->pending_blocks)) + schedule(); + __set_current_state(TASK_RUNNING); + } + } + return 0; +} + +/* + * Construct a log-writes mapping: + * log-writes + */ +static int log_writes_ctr(struct dm_target *ti, unsigned int argc, char **argv) +{ + struct log_writes_c *lc; + struct dm_arg_set as; + const char *devname, *logdevname; + + as.argc = argc; + as.argv = argv; + + if (argc < 2) { + ti->error = "Invalid argument count"; + return -EINVAL; + } + + lc = kzalloc(sizeof(struct log_writes_c), GFP_KERNEL); + if (!lc) { + ti->error = "Cannot allocate context"; + return -ENOMEM; + } + spin_lock_init(&lc->blocks_lock); + INIT_LIST_HEAD(&lc->unflushed_blocks); + INIT_LIST_HEAD(&lc->logging_blocks); + init_waitqueue_head(&lc->wait); + lc->sectorsize = 1 << SECTOR_SHIFT; + atomic_set(&lc->io_blocks, 0); + atomic_set(&lc->pending_blocks, 0); + + devname = dm_shift_arg(&as); + if (dm_get_device(ti, devname, dm_table_get_mode(ti->table), &lc->dev)) { + ti->error = "Device lookup failed"; + goto bad; + } + + logdevname = dm_shift_arg(&as); + if (dm_get_device(ti, logdevname, dm_table_get_mode(ti->table), &lc->logdev)) { + ti->error = "Log device lookup failed"; + dm_put_device(ti, lc->dev); + goto bad; + } + + lc->log_kthread = kthread_run(log_writes_kthread, lc, "log-write"); + if (!lc->log_kthread) { + ti->error = "Couldn't alloc kthread"; + dm_put_device(ti, lc->dev); + dm_put_device(ti, lc->logdev); + goto bad; + } + + /* We put the super at sector 0, start logging at sector 1 */ + lc->next_sector = 1; + lc->logging_enabled = true; + lc->end_sector = logdev_last_sector(lc); + lc->device_supports_discard = true; + + ti->num_flush_bios = 1; + ti->flush_supported = true; + ti->num_discard_bios = 1; + ti->discards_supported = true; + ti->per_bio_data_size = sizeof(struct per_bio_data); + ti->private = lc; + return 0; + +bad: + kfree(lc); + return -EINVAL; +} + +static int log_mark(struct log_writes_c *lc, char *data) +{ + struct pending_block *block; + size_t maxsize = lc->sectorsize - sizeof(struct log_write_entry); + + block = kzalloc(sizeof(struct pending_block), GFP_KERNEL); + if (!block) { + DMERR("Error allocating pending block"); + return -ENOMEM; + } + + block->data = kstrndup(data, maxsize, GFP_KERNEL); + if (!block->data) { + DMERR("Error copying mark data"); + kfree(block); + return -ENOMEM; + } + atomic_inc(&lc->pending_blocks); + block->datalen = strlen(block->data); + block->flags |= LOG_MARK_FLAG; + spin_lock_irq(&lc->blocks_lock); + list_add_tail(&block->list, &lc->logging_blocks); + spin_unlock_irq(&lc->blocks_lock); + wake_up_process(lc->log_kthread); + return 0; +} + +static void log_writes_dtr(struct dm_target *ti) +{ + struct log_writes_c *lc = ti->private; + + spin_lock_irq(&lc->blocks_lock); + list_splice_init(&lc->unflushed_blocks, &lc->logging_blocks); + spin_unlock_irq(&lc->blocks_lock); + + /* + * This is just nice to have since it'll update the super to include the + * unflushed blocks, if it fails we don't really care. + */ + log_mark(lc, "dm-log-writes-end"); + wake_up_process(lc->log_kthread); + wait_event(lc->wait, !atomic_read(&lc->io_blocks) && + !atomic_read(&lc->pending_blocks)); + kthread_stop(lc->log_kthread); + + WARN_ON(!list_empty(&lc->logging_blocks)); + WARN_ON(!list_empty(&lc->unflushed_blocks)); + dm_put_device(ti, lc->dev); + dm_put_device(ti, lc->logdev); + kfree(lc); +} + +static void normal_map_bio(struct dm_target *ti, struct bio *bio) +{ + struct log_writes_c *lc = ti->private; + + bio->bi_bdev = lc->dev->bdev; +} + +static int log_writes_map(struct dm_target *ti, struct bio *bio) +{ + struct log_writes_c *lc = ti->private; + struct per_bio_data *pb = dm_per_bio_data(bio, sizeof(struct per_bio_data)); + struct pending_block *block; + struct bvec_iter iter; + struct bio_vec bv; + size_t alloc_size; + int i = 0; + bool flush_bio = (bio->bi_rw & REQ_FLUSH); + bool fua_bio = (bio->bi_rw & REQ_FUA); + bool discard_bio = (bio->bi_rw & REQ_DISCARD); + + pb->block = NULL; + + /* Don't bother doing anything if logging has been disabled */ + if (!lc->logging_enabled) + goto map_bio; + + /* + * Map reads as normal. + */ + if (bio_data_dir(bio) == READ) + goto map_bio; + + /* No sectors and not a flush? Don't care */ + if (!bio_sectors(bio) && !flush_bio) + goto map_bio; + + /* + * Discards will have bi_size set but there's no actual data, so just + * allocate the size of the pending block. + */ + if (discard_bio) + alloc_size = sizeof(struct pending_block); + else + alloc_size = sizeof(struct pending_block) + sizeof(struct bio_vec) * bio_segments(bio); + + block = kzalloc(alloc_size, GFP_NOIO); + if (!block) { + DMERR("Error allocating pending block"); + spin_lock_irq(&lc->blocks_lock); + lc->logging_enabled = false; + spin_unlock_irq(&lc->blocks_lock); + return -ENOMEM; + } + INIT_LIST_HEAD(&block->list); + pb->block = block; + atomic_inc(&lc->pending_blocks); + + if (flush_bio) + block->flags |= LOG_FLUSH_FLAG; + if (fua_bio) + block->flags |= LOG_FUA_FLAG; + if (discard_bio) + block->flags |= LOG_DISCARD_FLAG; + + block->sector = bio->bi_iter.bi_sector; + block->nr_sectors = bio_sectors(bio); + + /* We don't need the data, just submit */ + if (discard_bio) { + WARN_ON(flush_bio || fua_bio); + if (lc->device_supports_discard) + goto map_bio; + bio_endio(bio, 0); + return DM_MAPIO_SUBMITTED; + } + + /* Flush bio, splice the unflushed blocks onto this list and submit */ + if (flush_bio && !bio_sectors(bio)) { + spin_lock_irq(&lc->blocks_lock); + list_splice_init(&lc->unflushed_blocks, &block->list); + spin_unlock_irq(&lc->blocks_lock); + goto map_bio; + } + + /* + * We will write this bio somewhere else way later so we need to copy + * the actual contents into new pages so we know the data will always be + * there. + * + * We do this because this could be a bio from O_DIRECT in which case we + * can't just hold onto the page until some later point, we have to + * manually copy the contents. + */ + bio_for_each_segment(bv, bio, iter) { + struct page *page; + void *src, *dst; + + page = alloc_page(GFP_NOIO); + if (!page) { + DMERR("Error allocing page"); + free_pending_block(lc, block); + spin_lock_irq(&lc->blocks_lock); + lc->logging_enabled = false; + spin_unlock_irq(&lc->blocks_lock); + return -ENOMEM; + } + + src = kmap_atomic(bv.bv_page); + dst = kmap_atomic(page); + memcpy(dst, src + bv.bv_offset, bv.bv_len); + kunmap_atomic(dst); + kunmap_atomic(src); + block->vecs[i].bv_page = page; + block->vecs[i].bv_len = bv.bv_len; + block->vec_cnt++; + i++; + } + + /* Had a flush with data in it, weird */ + if (flush_bio) { + spin_lock_irq(&lc->blocks_lock); + list_splice_init(&lc->unflushed_blocks, &block->list); + spin_unlock_irq(&lc->blocks_lock); + } +map_bio: + normal_map_bio(ti, bio); + return DM_MAPIO_REMAPPED; +} + +static int normal_end_io(struct dm_target *ti, struct bio *bio, int error) +{ + struct log_writes_c *lc = ti->private; + struct per_bio_data *pb = dm_per_bio_data(bio, sizeof(struct per_bio_data)); + + if (bio_data_dir(bio) == WRITE && pb->block) { + struct pending_block *block = pb->block; + unsigned long flags; + + spin_lock_irqsave(&lc->blocks_lock, flags); + if (block->flags & LOG_FLUSH_FLAG) { + list_splice_tail_init(&block->list, &lc->logging_blocks); + list_add_tail(&block->list, &lc->logging_blocks); + wake_up_process(lc->log_kthread); + } else if (block->flags & LOG_FUA_FLAG) { + list_add_tail(&block->list, &lc->logging_blocks); + wake_up_process(lc->log_kthread); + } else + list_add_tail(&block->list, &lc->unflushed_blocks); + spin_unlock_irqrestore(&lc->blocks_lock, flags); + } + + return error; +} + +/* + * INFO format: + */ +static void log_writes_status(struct dm_target *ti, status_type_t type, + unsigned status_flags, char *result, + unsigned maxlen) +{ + unsigned sz = 0; + struct log_writes_c *lc = ti->private; + + switch (type) { + case STATUSTYPE_INFO: + DMEMIT("%llu %llu", lc->logged_entries, + (unsigned long long)lc->next_sector - 1); + if (!lc->logging_enabled) + DMEMIT(" logging_disabled"); + break; + + case STATUSTYPE_TABLE: + DMEMIT("%s %s", lc->dev->name, lc->logdev->name); + break; + } +} + +static int log_writes_ioctl(struct dm_target *ti, unsigned int cmd, + unsigned long arg) +{ + struct log_writes_c *lc = ti->private; + struct dm_dev *dev = lc->dev; + int r = 0; + + /* + * Only pass ioctls through if the device sizes match exactly. + */ + if (ti->len != i_size_read(dev->bdev->bd_inode) >> SECTOR_SHIFT) + r = scsi_verify_blk_ioctl(NULL, cmd); + + return r ? : __blkdev_driver_ioctl(dev->bdev, dev->mode, cmd, arg); +} + +static int log_writes_merge(struct dm_target *ti, struct bvec_merge_data *bvm, + struct bio_vec *biovec, int max_size) +{ + struct log_writes_c *lc = ti->private; + struct request_queue *q = bdev_get_queue(lc->dev->bdev); + + if (!q->merge_bvec_fn) + return max_size; + + bvm->bi_bdev = lc->dev->bdev; + bvm->bi_sector = dm_target_offset(ti, bvm->bi_sector); + + return min(max_size, q->merge_bvec_fn(q, bvm, biovec)); +} + +static int log_writes_iterate_devices(struct dm_target *ti, + iterate_devices_callout_fn fn, + void *data) +{ + struct log_writes_c *lc = ti->private; + + return fn(ti, lc->dev, 0, ti->len, data); +} + +/* + * Messages supported: + * mark - specify the marked data. + */ +static int log_writes_message(struct dm_target *ti, unsigned argc, char **argv) +{ + int r = -EINVAL; + struct log_writes_c *lc = ti->private; + + if (argc != 2) { + DMWARN("Invalid log-writes message arguments, expect 2 arguments, got %d", argc); + return r; + } + + if (!strcasecmp(argv[0], "mark")) + r = log_mark(lc, argv[1]); + else + DMWARN("Unrecognised log writes target message received: %s", argv[0]); + + return r; +} + +static void log_writes_io_hints(struct dm_target *ti, struct queue_limits *limits) +{ + struct log_writes_c *lc = ti->private; + struct request_queue *q = bdev_get_queue(lc->dev->bdev); + + if (!q || !blk_queue_discard(q)) { + lc->device_supports_discard = false; + limits->discard_granularity = 1 << SECTOR_SHIFT; + limits->max_discard_sectors = (UINT_MAX >> SECTOR_SHIFT); + } +} + +static struct target_type log_writes_target = { + .name = "log-writes", + .version = {1, 0, 0}, + .module = THIS_MODULE, + .ctr = log_writes_ctr, + .dtr = log_writes_dtr, + .map = log_writes_map, + .end_io = normal_end_io, + .status = log_writes_status, + .ioctl = log_writes_ioctl, + .merge = log_writes_merge, + .message = log_writes_message, + .iterate_devices = log_writes_iterate_devices, + .io_hints = log_writes_io_hints, +}; + +static int __init dm_log_writes_init(void) +{ + int r = dm_register_target(&log_writes_target); + + if (r < 0) + DMERR("register failed %d", r); + + return r; +} + +static void __exit dm_log_writes_exit(void) +{ + dm_unregister_target(&log_writes_target); +} + +/* Module hooks */ +module_init(dm_log_writes_init); +module_exit(dm_log_writes_exit); + +MODULE_DESCRIPTION(DM_NAME " log writes target"); +MODULE_AUTHOR("Josef Bacik "); +MODULE_LICENSE("GPL");