Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > linux.kernel > #1333855
| From | Greg Kroah-Hartman <gregkh@linuxfoundation.org> |
|---|---|
| Newsgroups | linux.kernel |
| Subject | [PATCH 3.10 18/64] jbd2: Fix unreclaimed pages after truncate in data=journal mode |
| Date | 2016-02-15 00:50 +0100 |
| Message-ID | <r2eAr-1fL-25@gated-at.bofh.it> (permalink) |
| References | <r2dO1-Ir-9@gated-at.bofh.it> |
| Organization | linux.* mail to news gateway |
3.10-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jan Kara <jack@suse.cz>
commit bc23f0c8d7ccd8d924c4e70ce311288cb3e61ea8 upstream.
Ted and Namjae have reported that truncated pages don't get timely
reclaimed after being truncated in data=journal mode. The following test
triggers the issue easily:
for (i = 0; i < 1000; i++) {
pwrite(fd, buf, 1024*1024, 0);
fsync(fd);
fsync(fd);
ftruncate(fd, 0);
}
The reason is that journal_unmap_buffer() finds that truncated buffers
are not journalled (jh->b_transaction == NULL), they are part of
checkpoint list of a transaction (jh->b_cp_transaction != NULL) and have
been already written out (!buffer_dirty(bh)). We clean such buffers but
we leave them in the checkpoint list. Since checkpoint transaction holds
a reference to the journal head, these buffers cannot be released until
the checkpoint transaction is cleaned up. And at that point we don't
call release_buffer_page() anymore so pages detached from mapping are
lingering in the system waiting for reclaim to find them and free them.
Fix the problem by removing buffers from transaction checkpoint lists
when journal_unmap_buffer() finds out they don't have to be there
anymore.
Reported-and-tested-by: Namjae Jeon <namjae.jeon@samsung.com>
Fixes: de1b794130b130e77ffa975bb58cb843744f9ae5
Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Theodore Ts'o <tytso@mit.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
fs/jbd2/transaction.c | 2 ++
1 file changed, 2 insertions(+)
--- a/fs/jbd2/transaction.c
+++ b/fs/jbd2/transaction.c
@@ -1936,6 +1936,7 @@ static int journal_unmap_buffer(journal_
if (!buffer_dirty(bh)) {
/* bdflush has written it. We can drop it now */
+ __jbd2_journal_remove_checkpoint(jh);
goto zap_buffer;
}
@@ -1965,6 +1966,7 @@ static int journal_unmap_buffer(journal_
/* The orphan record's transaction has
* committed. We can cleanse this buffer */
clear_buffer_jbddirty(bh);
+ __jbd2_journal_remove_checkpoint(jh);
goto zap_buffer;
}
}
Back to linux.kernel | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread
[PATCH 3.10 00/64] 3.10.97-stable review Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-02-15 00:40 +0100
[PATCH 3.10 03/64] crypto: algif_hash - Only export and import on sockets with data Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-02-15 00:40 +0100
[PATCH 3.10 31/64] ALSA: rawmidi: Remove kernel WARNING for NULL user-space buffer check Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-02-15 00:40 +0100
[PATCH 3.10 55/64] crypto: af_alg - Disallow bind/setkey/... after accept(2) Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-02-15 00:40 +0100
[PATCH 3.10 25/64] fix sysvfs symlinks Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-02-15 00:40 +0100
[PATCH 3.10 01/64] recordmcount: Fix endianness handling bug for nop_mcount Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-02-15 00:40 +0100
[PATCH 3.10 37/64] ALSA: seq: Fix lockdep warnings due to double mutex locks Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-02-15 00:40 +0100
[PATCH 3.10 60/64] crypto: algif_hash - wait for crypto_ahash_init() to complete Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-02-15 00:40 +0100
[PATCH 3.10 59/64] ahci: Intel DNV device IDs SATA Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-02-15 00:40 +0100
[PATCH 3.10 07/64] kernel/signal.c: unexport sigsuspend() Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-02-15 00:40 +0100
[PATCH 3.10 24/64] [media] media: vb2 dma-contig: Fully cache synchronise buffers in prepare and finish Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-02-15 00:40 +0100
[PATCH 3.10 39/64] ALSA: timer: Fix leftover link at closing Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-02-15 00:40 +0100
[PATCH 3.10 28/64] ALSA: compress: Disable GET_CODEC_CAPS ioctl for some architectures Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-02-15 00:40 +0100
[PATCH 3.10 23/64] [media] v4l2-compat-ioctl32: fix alignment for ARM64 Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-02-15 00:40 +0100
[PATCH 3.10 05/64] HID: usbhid: fix recursive deadlock Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-02-15 00:40 +0100
[PATCH 3.10 57/64] AHCI: Fix softreset failed issue of Port Multiplier Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-02-15 00:40 +0100
[PATCH 3.10 04/64] dm btree: fix leak of bufio-backed block in btree_split_sibling error path Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-02-15 00:40 +0100
[PATCH 3.10 16/64] vTPM: fix memory allocation flag for rtce buffer at kernel boot Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-02-15 00:50 +0100
[PATCH 3.10 14/64] wlcore/wl12xx: spi: fix oops on firmware load Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-02-15 00:50 +0100
[PATCH 3.10 15/64] wlcore/wl12xx: spi: fix NULL pointer dereference (Oops) Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-02-15 00:50 +0100
[PATCH 3.10 22/64] parisc: Fix __ARCH_SI_PREAMBLE_SIZE Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-02-15 00:50 +0100
[PATCH 3.10 19/64] [PATCH] fix calculation of meta_bg descriptor backups Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-02-15 00:50 +0100
[PATCH 3.10 18/64] jbd2: Fix unreclaimed pages after truncate in data=journal mode Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-02-15 00:50 +0100
[PATCH 3.10 17/64] mtd: mtdpart: fix add_mtd_partitions error path Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-02-15 00:50 +0100
[PATCH 3.10 12/64] spi: atmel: Fix DMA-setup for transfers with more than 8 bits per word Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-02-15 00:50 +0100
[PATCH 3.10 21/64] parisc: Fix syscall restarts Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-02-15 00:50 +0100
[PATCH 3.10 11/64] Revert "dm mpath: fix stalls when handling invalid ioctls" Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-02-15 00:50 +0100
[PATCH 3.10 13/64] spi: fix parent-device reference leak Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-02-15 00:50 +0100
[PATCH 3.10 02/64] xhci: fix placement of call to usb_disabled() Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-02-15 00:50 +0100
Re: [PATCH 3.10 00/64] 3.10.97-stable review Guenter Roeck <linux@roeck-us.net> - 2016-02-15 06:30 +0100
Re: [PATCH 3.10 00/64] 3.10.97-stable review Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-02-15 19:20 +0100
Re: [PATCH 3.10 00/64] 3.10.97-stable review Guenter Roeck <linux@roeck-us.net> - 2016-02-15 16:50 +0100
Re: [PATCH 3.10 00/64] 3.10.97-stable review Shuah Khan <shuahkh@osg.samsung.com> - 2016-02-15 18:20 +0100
Re: [PATCH 3.10 00/64] 3.10.97-stable review Shuah Khan <shuahkh@osg.samsung.com> - 2016-02-17 21:40 +0100
Re: [PATCH 3.10 00/64] 3.10.97-stable review Willy Tarreau <w@1wt.eu> - 2016-02-18 00:30 +0100
Re: [PATCH 3.10 00/64] 3.10.97-stable review Shuah Khan <shuahkh@osg.samsung.com> - 2016-02-18 00:40 +0100
Re: [PATCH 3.10 00/64] 3.10.97-stable review Shuah Khan <shuahkh@osg.samsung.com> - 2016-02-18 04:30 +0100
Re: [PATCH 3.10 00/64] 3.10.97-stable review Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-02-19 23:30 +0100
Re: [PATCH 3.10 00/64] 3.10.97-stable review Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2016-02-17 21:40 +0100
csiph-web