From: fdmanana@kernel.org
To: linux-btrfs@vger.kernel.org
Subject: [PATCH v2] Btrfs: avoid deadlock with memory reclaim due to allocation of devices
Date: Fri, 11 Jan 2019 17:17:59 +0000 [thread overview]
Message-ID: <20190111171759.19920-1-fdmanana@kernel.org> (raw)
In-Reply-To: <20181213211725.14832-1-fdmanana@kernel.org>
From: Filipe Manana <fdmanana@suse.com>
In a few places we are allocating a device using the GFP_KERNEL flag when
it is not safe to do so, because if reclaim is triggered it can cause a
transaction commit while we are holding the device list mutex. This mutex
is required in the transaction commit path (at write_all_supers() and
btrfs_update_commit_device_size()).
So fix this by setting up a nofs memory allocation context in those cases.
Fixes: 78f2c9e6dbb14 ("btrfs: device add and remove: use GFP_KERNEL")
Fixes: e0ae999414238 ("btrfs: preallocate device flush bio")
Signed-off-by: Filipe Manana <fdmanana@suse.com>
---
V2: Change the approach to fix the problem by setting up nofs contextes
where needed.
fs/btrfs/volumes.c | 33 ++++++++++++++++++++++++++++++---
1 file changed, 30 insertions(+), 3 deletions(-)
diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c
index 2576b1a379c9..663566baae78 100644
--- a/fs/btrfs/volumes.c
+++ b/fs/btrfs/volumes.c
@@ -14,6 +14,7 @@
#include <linux/semaphore.h>
#include <linux/uuid.h>
#include <linux/list_sort.h>
+#include <linux/sched/mm.h>
#include "ctree.h"
#include "extent_map.h"
#include "disk-io.h"
@@ -988,20 +989,29 @@ static noinline struct btrfs_device *device_list_add(const char *path,
}
if (!device) {
+ unsigned int nofs_flag;
+
if (fs_devices->opened) {
mutex_unlock(&fs_devices->device_list_mutex);
return ERR_PTR(-EBUSY);
}
+ /*
+ * Setup nofs context because we are holding the device list
+ * mutex, which is required for a transaction commit.
+ */
+ nofs_flag = memalloc_nofs_save();
device = btrfs_alloc_device(NULL, &devid,
disk_super->dev_item.uuid);
if (IS_ERR(device)) {
+ memalloc_nofs_restore(nofs_flag);
mutex_unlock(&fs_devices->device_list_mutex);
/* we can safely leave the fs_devices entry around */
return device;
}
- name = rcu_string_strdup(path, GFP_NOFS);
+ name = rcu_string_strdup(path, GFP_KERNEL);
+ memalloc_nofs_restore(nofs_flag);
if (!name) {
btrfs_free_device(device);
mutex_unlock(&fs_devices->device_list_mutex);
@@ -1137,11 +1147,19 @@ static struct btrfs_fs_devices *clone_fs_devices(struct btrfs_fs_devices *orig)
/* We have held the volume lock, it is safe to get the devices. */
list_for_each_entry(orig_dev, &orig->devices, dev_list) {
struct rcu_string *name;
+ unsigned int nofs_flag;
+ /*
+ * Setup nofs context because we are holding the device list
+ * mutex, which is required for a transaction commit.
+ */
+ nofs_flag = memalloc_nofs_save();
device = btrfs_alloc_device(NULL, &orig_dev->devid,
orig_dev->uuid);
- if (IS_ERR(device))
+ if (IS_ERR(device)) {
+ memalloc_nofs_restore(nofs_flag);
goto error;
+ }
/*
* This is ok to do without rcu read locked because we hold the
@@ -1151,12 +1169,14 @@ static struct btrfs_fs_devices *clone_fs_devices(struct btrfs_fs_devices *orig)
name = rcu_string_strdup(orig_dev->name->str,
GFP_KERNEL);
if (!name) {
+ memalloc_nofs_restore(nofs_flag);
btrfs_free_device(device);
goto error;
}
rcu_assign_pointer(device->name, name);
}
+ memalloc_nofs_restore(nofs_flag);
list_add(&device->dev_list, &fs_devices->devices);
device->fs_devices = fs_devices;
fs_devices->num_devices++;
@@ -1262,6 +1282,7 @@ static void btrfs_close_one_device(struct btrfs_device *device)
struct btrfs_fs_devices *fs_devices = device->fs_devices;
struct btrfs_device *new_device;
struct rcu_string *name;
+ unsigned int nofs_flag;
if (device->bdev)
fs_devices->open_devices--;
@@ -1277,17 +1298,23 @@ static void btrfs_close_one_device(struct btrfs_device *device)
btrfs_close_bdev(device);
+ /*
+ * Setup nofs context because we are holding the device list
+ * mutex, which is required for a transaction commit.
+ */
+ nofs_flag = memalloc_nofs_save();
new_device = btrfs_alloc_device(NULL, &device->devid,
device->uuid);
BUG_ON(IS_ERR(new_device)); /* -ENOMEM */
/* Safe because we are under uuid_mutex */
if (device->name) {
- name = rcu_string_strdup(device->name->str, GFP_NOFS);
+ name = rcu_string_strdup(device->name->str, GFP_KERNEL);
BUG_ON(!name); /* -ENOMEM */
rcu_assign_pointer(new_device->name, name);
}
+ memalloc_nofs_restore(nofs_flag);
list_replace_rcu(&device->dev_list, &new_device->dev_list);
new_device->fs_devices = device->fs_devices;
--
2.11.0
next prev parent reply other threads:[~2019-01-11 17:18 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-12-13 21:17 [PATCH] Btrfs: avoid deadlock with memory reclaim due to allocation of devices fdmanana
2018-12-14 7:27 ` Nikolay Borisov
2019-01-08 11:51 ` Filipe Manana
2019-01-09 18:26 ` David Sterba
2019-01-09 19:48 ` Filipe Manana
2019-01-10 7:32 ` Anand Jain
2019-01-10 7:03 ` Nikolay Borisov
2019-01-11 17:17 ` fdmanana [this message]
2019-01-14 8:21 ` [PATCH v2] " Anand Jain
2019-01-18 18:07 ` David Sterba
2019-01-25 2:56 ` Anand Jain
2019-01-25 3:40 ` Anand Jain
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=20190111171759.19920-1-fdmanana@kernel.org \
--to=fdmanana@kernel.org \
--cc=linux-btrfs@vger.kernel.org \
/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).