2015-07-05 21:53:18

by Theodore Ts'o

[permalink] [raw]
Subject: [GIT PULL] ext4 bug fixes for 4.2-rc2

The following changes since commit a2fd66d069d86d793e9d39d4079b96f46d13f237:

ext4: set lazytime on remount if MS_LAZYTIME is set by mount (2015-06-23 11:03:54 -0400)

are available in the git repository at:

git://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4.git tags/ext4_for_linus_stable

for you to fetch changes up to 7444a072c387a93ebee7066e8aee776954ab0e41:

ext4: replace open coded nofail allocation in ext4_free_blocks() (2015-07-05 12:33:44 -0400)

----------------------------------------------------------------
Bug fixes (all for stable kernels) for ext4:
* address corner cases for indirect blocks->extent migration
* fix reserved block accounting invalidate_page when
page_size != block_size (i.e., ppc or 1k block size file systems)
* fix deadlocks when a memcg is under heavy memory pressure
* fix fencepost error in lazytime optimization

----------------------------------------------------------------
Eryu Guan (2):
ext4: be more strict when migrating to non-extent based file
ext4: correctly migrate a file with a hole at the beginning

Lukas Czerner (1):
ext4: fix reservation release on invalidatepage for delalloc fs

Michal Hocko (1):
ext4: replace open coded nofail allocation in ext4_free_blocks()

Nikolay Borisov (2):
bufferhead: Add _gfp version for sb_getblk()
ext4: avoid deadlocks in the writeback path by using sb_getblk_gfp

Theodore Ts'o (1):
ext4: fix fencepost error in lazytime optimization

fs/ext4/extents.c | 6 +++---
fs/ext4/inode.c | 22 ++++++++++++++++++----
fs/ext4/mballoc.c | 16 +++++-----------
fs/ext4/migrate.c | 17 ++++++++++++++---
include/linux/buffer_head.h | 7 +++++++
5 files changed, 47 insertions(+), 21 deletions(-)


2015-07-11 12:13:04

by Jörg-Volker Peetz

[permalink] [raw]
Subject: Re: [GIT PULL] ext4 bug fixes for 4.2-rc2

Theodore Ts'o wrote on 07/05/2015 23:53:
> The following changes since commit a2fd66d069d86d793e9d39d4079b96f46d13f237:
>
> ext4: set lazytime on remount if MS_LAZYTIME is set by mount (2015-06-23 11:03:54 -0400)
>
<snip>

Hi Ted,

I'm testing this patch and "ext4: set lazytime on remount if MS_LAZYTIME is set
by mount" on top of stable kernel 4.1.2.

So far no problems with lazytime enabled. Thanks for that :-)

But remounting with mount version 2.26.2 sucks. As promised, lazytime can be
enabled. But disabling it or trying to switch between atime, relatime, and
noatime doesn't work.

Starting from

# grep sda /proc/mounts
/dev/sda2 /home ext4 rw,lazytime,noatime,nobarrier,errors=remount-ro 0 0

some strace output:

# strace -o /tmp/st mount -o remount,nolazytime /home
# grep ^mount /tmp/st
mount("/dev/sda2", "/home", 0x2152780, MS_MGC_VAL|MS_REMOUNT|MS_NOATIME,
"nobarrier,errors=remount-ro") = 0

# strace -o /tmp/st mount -o remount,relatime /home
# grep ^mount /tmp/st
mount("/dev/sda2", "/home", 0x1543780, MS_REMOUNT|MS_RELATIME|MS_NOATIME,
"nobarrier,errors=remount-ro") = 0

# strace -o /tmp/st mount -o remount,atime /home
# grep ^mount /tmp/st
mount("/dev/sda2", "/home", 0x1534780, MS_MGC_VAL|MS_REMOUNT,
"nobarrier,errors=remount-ro") = 0

The contents of /proc/mounts isn't changed by all this :-(

Or should I contact the util-linux developers?

--
Regards,
J?rg.