Re: [PATCH v4 3/4] btrfs: lzo: Add header length check to avoid slab out of bounds access

Reviewed-by: Misono Tomohiro Thanks, Tomohiro Misono On 2018/05/30 13:58, Qu Wenruo wrote: > James Harvey reported that some corrupted compressed extent data can > lead to various kernel memory corruption. > > Such corrupted extent data belongs to inode with NODATASUM flags, thus > data csum

[PATCH v4 3/4] btrfs: lzo: Add header length check to avoid slab out of bounds access

James Harvey reported that some corrupted compressed extent data can lead to various kernel memory corruption. Such corrupted extent data belongs to inode with NODATASUM flags, thus data csum won't help us detecting such bug. If lucky enough, kasan could catch it like:

[PATCH 2/2] btrfs-progs: ins: dump-tree: Print tree name for tree block backref

Print tree name instead of number to make output more readable. Signed-off-by: Misono Tomohiro --- print-tree.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/print-tree.c b/print-tree.c index 90173c2b..8d551d09 100644 --- a/print-tree.c +++ b/print-tree.c @@ -470,8

[PATCH 0/2] btrfs-progs: ins: dump-tree: Small updates for extent tree output

Two small updates for extent tree to make output more human readable. Misono Tomohiro (2): btrfs-progs: ins: dump-tree: Use %lld for extent data backref offset btrfs-progs: ins: dump-tree: Print tree name for tree block backref print-tree.c | 9 + 1 file changed, 5 insertions(+), 4

[PATCH 1/2] btrfs-progs: ins: dump-tree: Use %lld for extent data backref offset

Offset of extent data backref in extent item can be negative. So, let's use %lld insteand of %llu to output readable value. Signed-off-by: Misono Tomohiro --- print-tree.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/print-tree.c b/print-tree.c index

[PATCH REBASED 0/2] btrfs: Delay block group auto removal to avoid interfering qgroups

The patchset can be fetched from github: https://github.com/adam900710/linux/tree/delayed_bg_removal It's based on David's misc-next branch, HEAD is: commit c77a75861d82a497bd1a26054588cd9af59eb5db (david/misc-next) Author: Ethan Lien Date: Mon May 28 13:48:20 2018 +0800 btrfs: balance

[PATCH REBASED 1/2] btrfs: Use btrfs_mark_bg_unused() to replace open code

Introduce a small helper, btrfs_mark_bg_unused(), to accquire needed locks and add a block group to unused_bgs list. No functional modification, and only 3 callers are involved. Signed-off-by: Qu Wenruo --- fs/btrfs/ctree.h | 1 + fs/btrfs/extent-tree.c | 36

[PATCH REBASED 2/2] btrfs: Delayed empty block group auto removal to next transaction

Under certain KVM load and LTP tests, we are possible to hit the following calltrace if quota is enabled: -- BTRFS critical (device vda2): unable to find logical 8820195328 length 4096 BTRFS critical (device vda2): unable to find logical 8820195328 length 4096 [ cut here

[PATCH 2/2] btrfs: drop useless member qgroup_reserved of btrfs_pending_snapshot

From: Gu JinXiang Since there is no more use of qgroup_reserved member in struct btrfs_pending_snapshot, remove it. Signed-off-by: Gu JinXiang --- fs/btrfs/transaction.h | 1 - 1 file changed, 1 deletion(-) diff --git a/fs/btrfs/transaction.h b/fs/btrfs/transaction.h index

[PATCH 1/2] btrfs: drop unused parameter qgroup_reserved

From: Gu JinXiang Since commit 7775c8184ec0 ("btrfs: remove unused parameter from btrfs_subvolume_release_metadata") parameter qgroup_reserved is not used by caller of function btrfs_subvolume_reserve_metadata. So remove it. Signed-off-by: Gu JinXiang --- fs/btrfs/ctree.h | 3 +--

Re: [PATCH 1/2] btrfs: returns -ENOMEM if path allocation failed in btrfs_cross_ref_exist()

Sorry, the [PATCH 2/2] is abandoned. So the subject should be [PATCH]. On 05/30/2018 10:07 AM, Su Yue wrote: > Signed-off-by: Su Yue > --- > fs/btrfs/extent-tree.c | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c > index

[PATCH 1/2] btrfs: returns -ENOMEM if path allocation failed in btrfs_cross_ref_exist()

Signed-off-by: Su Yue --- fs/btrfs/extent-tree.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 44093f92a532..99df6199dffc 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -3311,7 +3311,7 @@ int

Re: [PATCH] btrfs-progs: check: Also compare data between mirrors to detect corruption for NODATASUM extents

On Tue, May 29, 2018 at 3:45 AM, Qu Wenruo wrote: > As the lzo corruption reported by James Harvey, for data extents without > checksum, neither btrfs check nor kernel scrub could detect anything > wrong. > > However if our profile supports duplication, we still have a chance to > detect such

Re: [PATCH] Btrfs: clean up error handling in btrfs_truncate()

On Tue, May 22, 2018 at 09:59:50AM -0700, Omar Sandoval wrote: > From: Omar Sandoval > > btrfs_truncate() uses two variables for error handling, ret and err (if > this sounds familiar, it's because btrfs_truncate_inode_items() did > something similar). This is error prone, as was made evident by

Re: [PATCH 1/2] btrfs: kill btrfs_write_inode

On Mon, May 28, 2018 at 06:57:59PM +0200, David Sterba wrote: > On Tue, May 22, 2018 at 01:47:22PM -0400, Josef Bacik wrote: > > From: Josef Bacik > > > > We don't actually need this. It used to be in place for O_SYNC writes, > > but we've used the normal fsync() path for that for years now.

Re: off-by-one uncompressed invalid ram_bytes corruptions

On 05/21/2018 03:07 AM, Qu Wenruo wrote: > > [...] > > And the problem of wrong "root=" output also makes me pretty curious. Yeah, there a bunch of error messages in the kernel with always say root=1, regardless of the actual root the tree block is in. I think I once tried to find out why, but

Re: [PATCH 0/3] btrfs: Delay block group auto removal to avoid interfering qgroups

On Mon, May 28, 2018 at 05:20:06PM +0800, Qu Wenruo wrote: > The patchset can be fetched from github: > https://github.com/adam900710/linux/tree/delayed_bg_removal > > It's based on v4.17-rc5 branch. Please refresh the patches on top of today's misc-next. The patch 2/3 has been there already and

Re: [PATCH] btrfs: set device state reappearing missing device

On Tue, May 29, 2018 at 02:50:09PM +0800, Gu Jinxiang wrote: > From: Gu JinXiang > > patch for commit 921518924084 > ("btrfs: handle dynamically reappearing missing device"). The commit is not in any of the development branches and I only found the patch in the mailinglist, so this is for

Re: [PATCH v2] btrfs: balance dirty metadata pages in btrfs_finish_ordered_io

On Mon, May 28, 2018 at 01:48:20PM +0800, Ethan Lien wrote: > [Problem description and how we fix it] > We should balance dirty metadata pages at the end of > btrfs_finish_ordered_io, since a small, unmergeable random write can > potentially produce dirty metadata which is multiple times larger

Re: off-by-one uncompressed invalid ram_bytes corruptions

Qu Wenruo writes: > On 2018年05月28日 11:47, Steve Leung wrote: >> On 05/26/2018 06:57 PM, Qu Wenruo wrote: >>> >>> >>> On 2018年05月26日 22:06, Steve Leung wrote: On 05/20/2018 07:07 PM, Qu Wenruo wrote: > > > On 2018年05月21日 04:43, Steve Leung wrote: >> On 05/19/2018 07:02 PM, Qu

Re: [PATCH v3] Btrfs: remove unused check of skip_locking

On Tue, May 29, 2018 at 09:27:06PM +0800, Liu Bo wrote: > The check is superfluous since all of callers who set search_for_commit > also have skip_locking set. > > ASSERT() is put in place to ensure skip_locking is set by callers. > > Reviewed-by: Qu Wenruo > Signed-off-by: Liu Bo Added to

Re: csum failed root raveled during balance

On 2018-05-29 10:02, ein wrote: On 05/29/2018 02:12 PM, Austin S. Hemmelgarn wrote: On 2018-05-28 13:10, ein wrote: On 05/23/2018 01:03 PM, Austin S. Hemmelgarn wrote: On 2018-05-23 06:09, ein wrote: On 05/23/2018 11:09 AM, Duncan wrote: ein posted on Wed, 23 May 2018 10:03:52 +0200 as

Re: [PATCH 2/4] Btrfs: make should_defrag_range() understood compressed extents

вт, 19 дек. 2017 г. в 13:02, Timofey Titovets : > Both, defrag ioctl and autodefrag - call btrfs_defrag_file() > for file defragmentation. > Kernel default target extent size - 256KiB. > Btrfs progs default - 32MiB. > Both bigger then maximum size of compressed extent - 128KiB. >

Re: csum failed root raveled during balance

On 05/29/2018 02:12 PM, Austin S. Hemmelgarn wrote: > On 2018-05-28 13:10, ein wrote: >> On 05/23/2018 01:03 PM, Austin S. Hemmelgarn wrote: >>> On 2018-05-23 06:09, ein wrote: On 05/23/2018 11:09 AM, Duncan wrote: > ein posted on Wed, 23 May 2018 10:03:52 +0200 as excerpted: >

Re: [PATCH v2 1/4] fstests: btrfs: add seed sprout functionality test

On Mon, May 28, 2018 at 05:51:45PM +0800, Anand Jain wrote: > Create a seed device and add the sprout device to it. > > Signed-off-by: Anand Jain This series looks fine to me from fstests' point of view, there're just some really minor common issues. But I'd like some reviews from other btrfs

Possible deadlock in btrfs_alloc_data_chunk_ondemand

Hi, I am fuzzing 4.17-rc4 with BTRFS using Syzkaller and found a possible deadlock. This crash has happened 3 times in an hour. Stack Trace: https://pastebin.com/1mjA9fzv Kernel Configs: https://pastebin.com/6J1KMczH I don't have a reproducer yet, unfortunately. Regards, Shankara Pailoor

[PATCH v3] Btrfs: remove unused check of skip_locking

The check is superfluous since all of callers who set search_for_commit also have skip_locking set. ASSERT() is put in place to ensure skip_locking is set by callers. Reviewed-by: Qu Wenruo Signed-off-by: Liu Bo --- v3: Add assertion and comments to ensure skip_locking is not forgot by

Re: csum failed root raveled during balance

On 2018-05-28 13:10, ein wrote: On 05/23/2018 01:03 PM, Austin S. Hemmelgarn wrote: On 2018-05-23 06:09, ein wrote: On 05/23/2018 11:09 AM, Duncan wrote: ein posted on Wed, 23 May 2018 10:03:52 +0200 as excerpted: IMHO the best course of action would be to disable checksumming for you vm

Re: [PATCH 2nd try v4] btrfs: drop uuid_mutex in btrfs_free_extra_devids()

On Tue, May 29, 2018 at 01:40:34PM +0800, Anand Jain wrote: > >>> v3->v4: As we traverse through the seed device, fs_device gets > >>> updated with > >>> the child seed fs_devices, so make sure we use the same fs_devices > >>> pointer for the mutex_unlock as used for the mutex_lock. > >>

Re: [PATCH] btrfs-progs: check: Also compare data between mirrors to detect corruption for NODATASUM extents

On Tue, May 29, 2018 at 03:45:27PM +0800, Qu Wenruo wrote: >As the lzo corruption reported by James Harvey, for data extents without >checksum, neither btrfs check nor kernel scrub could detect anything >wrong. > >However if our profile supports duplication, we still have a chance to >detect such

Re: [PATCH 0/2] UUID tree fs_info cleanup

On Tue, May 29, 2018 at 03:01:52PM +0800, Lu Fengqi wrote: > Just remove the redundant fs_info argument from uuid_tree functions > because these functions have already taken transaction which has a > reference to the fs_info. Thanks. I've also renamed the function btrfs_uuid_tree_rem to

Re: [PATCH] btrfs-progs: check: Also compare data between mirrors to detect corruption for NODATASUM extents

On 05/29/2018 03:45 PM, Qu Wenruo wrote: > As the lzo corruption reported by James Harvey, for data extents without > checksum, neither btrfs check nor kernel scrub could detect anything > wrong. > > However if our profile supports duplication, we still have a chance to > detect such

Re: [PATCH v3 3/4] btrfs: lzo: Add header length check to avoid slab out of bounds access

On 2018年05月29日 16:30, Misono Tomohiro wrote: > On 2018/05/23 17:23, Qu Wenruo wrote: >> James Harvey reported that some corrupted compressed extent data can >> lead to various kernel memory corruption. >> >> Such corrupted extent data belongs to inode with NODATASUM flags, thus >> data csum

Re: [PATCH v3 3/4] btrfs: lzo: Add header length check to avoid slab out of bounds access

On 2018/05/23 17:23, Qu Wenruo wrote: > James Harvey reported that some corrupted compressed extent data can > lead to various kernel memory corruption. > > Such corrupted extent data belongs to inode with NODATASUM flags, thus > data csum won't help us detecting such bug. > > If lucky enough,

Re: [PATCH v2 0/6] btrfs_search_slot cleanups

On Mon, May 28, 2018 at 06:17:18PM +0200, David Sterba wrote: > On Mon, May 28, 2018 at 04:40:28PM +0200, David Sterba wrote: > > On Fri, May 18, 2018 at 11:00:18AM +0800, Liu Bo wrote: > > > Here are a collection of patches I did for btrfs_search_slot(). > > > > > > v2: more explicit commit log

[PATCH] btrfs-progs: check: Also compare data between mirrors to detect corruption for NODATASUM extents

As the lzo corruption reported by James Harvey, for data extents without checksum, neither btrfs check nor kernel scrub could detect anything wrong. However if our profile supports duplication, we still have a chance to detect such corruption, as in that case the mirrors will mismatch with each

Re: [LKP] [lkp-robot] [mm] 9092c71bb7: blogbench.write_score -12.3% regression

Hi, Josef, Do you have time to take a look at the regression? kernel test robot writes: > Greeting, > > FYI, we noticed a -12.3% regression of blogbench.write_score and a +9.6% > improvement > of blogbench.read_score due to commit: > > > commit: 9092c71bb724dba2ecba849eae69e5c9d39bd3d2 ("mm:

Re: [PATCH 0/2] UUID tree fs_info cleanup

On 29.05.2018 10:01, Lu Fengqi wrote: > Just remove the redundant fs_info argument from uuid_tree functions > because these functions have already taken transaction which has a > reference to the fs_info. > > No functional change. > > Lu Fengqi (2): > btrfs: Remove fs_info argument from

[PATCH 1/2] btrfs: Remove fs_info argument from btrfs_uuid_tree_add

This function always takes a transaction handle which contains a reference to the fs_info. Use that and remove the extra argument. Signed-off-by: Lu Fengqi --- fs/btrfs/ctree.h | 3 +-- fs/btrfs/ioctl.c | 4 ++-- fs/btrfs/transaction.c | 7 +++ fs/btrfs/uuid-tree.c | 4 ++--

[PATCH 2/2] btrfs: Remove fs_info argument from btrfs_uuid_tree_rem

This function always takes a transaction handle which contains a reference to the fs_info. Use that and remove the extra argument. Signed-off-by: Lu Fengqi --- fs/btrfs/ctree.h | 3 +-- fs/btrfs/inode.c | 5 ++--- fs/btrfs/ioctl.c | 5 ++--- fs/btrfs/uuid-tree.c | 6 +++--- 4 files

[PATCH 0/2] UUID tree fs_info cleanup

Just remove the redundant fs_info argument from uuid_tree functions because these functions have already taken transaction which has a reference to the fs_info. No functional change. Lu Fengqi (2): btrfs: Remove fs_info argument from btrfs_uuid_tree_add btrfs: Remove fs_info argument from

[PATCH] btrfs: set device state reappearing missing device

From: Gu JinXiang patch for commit 921518924084 ("btrfs: handle dynamically reappearing missing device"). Since BTRFS_DEV_STATE_IN_FS_METADATA will be clear in btrfs_open_one_device, and when alloc a new chunk in __btrfs_alloc_chunk, device with BTRFS_DEV_STATE_IN_FS_METADATA not be set will be

Re: [PATCH 3/3] btrfs: Delayed empty block group auto removal to next transaction

On 2018年05月29日 14:20, Nikolay Borisov wrote: > > > On 28.05.2018 12:20, Qu Wenruo wrote: >> Under certain KVM load and LTP tests, we are possible to hit the >> following calltrace if quota is enabled: >> -- >> BTRFS critical (device vda2): unable to find logical 8820195328 length 4096 >>

Re: [PATCH 3/3] btrfs: Delayed empty block group auto removal to next transaction

On 28.05.2018 12:20, Qu Wenruo wrote: > Under certain KVM load and LTP tests, we are possible to hit the > following calltrace if quota is enabled: > -- > BTRFS critical (device vda2): unable to find logical 8820195328 length 4096 > BTRFS critical (device vda2): unable to find logical