commit 5cf02d09b5 upstream.
We've had some reports of a deadlock where rpciod ends up with a stack
trace like this:
PID: 2507 TASK: ffff88103691ab40 CPU: 14 COMMAND: "rpciod/14"
#0 [ffff8810343bf2f0] schedule at ffffffff814dabd9
#1 [ffff8810343bf3b8] nfs_wait_bit_killable at ffffffffa038fc04 [nfs]
#2 [ffff8810343bf3c8] __wait_on_bit at ffffffff814dbc2f
#3 [ffff8810343bf418] out_of_line_wait_on_bit at ffffffff814dbcd8
#4 [ffff8810343bf488] nfs_commit_inode at ffffffffa039e0c1 [nfs]
#5 [ffff8810343bf4f8] nfs_release_page at ffffffffa038bef6 [nfs]
#6 [ffff8810343bf528] try_to_release_page at ffffffff8110c670
#7 [ffff8810343bf538] shrink_page_list.clone.0 at ffffffff81126271
#8 [ffff8810343bf668] shrink_inactive_list at ffffffff81126638
#9 [ffff8810343bf818] shrink_zone at ffffffff8112788f
#10 [ffff8810343bf8c8] do_try_to_free_pages at ffffffff81127b1e
#11 [ffff8810343bf958] try_to_free_pages at ffffffff8112812f
#12 [ffff8810343bfa08] __alloc_pages_nodemask at ffffffff8111fdad
#13 [ffff8810343bfb28] kmem_getpages at ffffffff81159942
#14 [ffff8810343bfb58] fallback_alloc at ffffffff8115a55a
#15 [ffff8810343bfbd8] ____cache_alloc_node at ffffffff8115a2d9
#16 [ffff8810343bfc38] kmem_cache_alloc at ffffffff8115b09b
#17 [ffff8810343bfc78] sk_prot_alloc at ffffffff81411808
#18 [ffff8810343bfcb8] sk_alloc at ffffffff8141197c
#19 [ffff8810343bfce8] inet_create at ffffffff81483ba6
#20 [ffff8810343bfd38] __sock_create at ffffffff8140b4a7
#21 [ffff8810343bfd98] xs_create_sock at ffffffffa01f649b [sunrpc]
#22 [ffff8810343bfdd8] xs_tcp_setup_socket at ffffffffa01f6965 [sunrpc]
#23 [ffff8810343bfe38] worker_thread at ffffffff810887d0
#24 [ffff8810343bfee8] kthread at ffffffff8108dd96
#25 [ffff8810343bff48] kernel_thread at ffffffff8100c1ca
rpciod is trying to allocate memory for a new socket to talk to the
server. The VM ends up calling ->releasepage to get more memory, and it
tries to do a blocking commit. That commit can't succeed however without
a connected socket, so we deadlock.
Fix this by setting PF_FSTRANS on the workqueue task prior to doing the
socket allocation, and having nfs_release_page check for that flag when
deciding whether to do a commit call. Also, set PF_FSTRANS
unconditionally in rpc_async_schedule since that function can also do
allocations sometimes.
Signed-off-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 2930d381d2 upstream.
Actually, xfs and jfs can optionally be case insensitive; we'll handle
that case in later patches.
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit e9fbcb4220 upstream.
Each ordered operation has a free callback, and this was called with the
worker spinlock held. Josef made the free callback also call iput,
which we can't do with the spinlock.
This drops the spinlock for the free operation and grabs it again before
moving through the rest of the list. We'll circle back around to this
and find a cleaner way that doesn't bounce the lock around so much.
Signed-off-by: Chris Mason <chris.mason@fusionio.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 0ec4f431eb upstream.
The only checks of the long argument passed to fcntl(fd,F_SETLEASE,.)
are done after converting the long to an int. Thus some illegal values
may be let through and cause problems in later code.
[ They actually *don't* cause problems in mainline, as of Dave Jones's
commit 8d657eb3b4 "Remove easily user-triggerable BUG from
generic_setlease", but we should fix this anyway. And this patch will
be necessary to fix real bugs on earlier kernels. ]
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit a6bc32b899 upstream.
Stable note: Not tracked in Buzilla. This was part of a series that
reduced interactivity stalls experienced when THP was enabled.
These stalls were particularly noticable when copying data
to a USB stick but the experiences for users varied a lot.
This patch adds a lightweight sync migrate operation MIGRATE_SYNC_LIGHT
mode that avoids writing back pages to backing storage. Async compaction
maps to MIGRATE_ASYNC while sync compaction maps to MIGRATE_SYNC_LIGHT.
For other migrate_pages users such as memory hotplug, MIGRATE_SYNC is
used.
This avoids sync compaction stalling for an excessive length of time,
particularly when copying files to a USB stick where there might be a
large number of dirty pages backed by a filesystem that does not support
->writepages.
[aarcange@redhat.com: This patch is heavily based on Andrea's work]
[akpm@linux-foundation.org: fix fs/nfs/write.c build]
[akpm@linux-foundation.org: fix fs/btrfs/disk-io.c build]
Signed-off-by: Mel Gorman <mgorman@suse.de>
Reviewed-by: Rik van Riel <riel@redhat.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Minchan Kim <minchan.kim@gmail.com>
Cc: Dave Jones <davej@redhat.com>
Cc: Jan Kara <jack@suse.cz>
Cc: Andy Isaacson <adi@hexapodia.org>
Cc: Nai Xia <nai.xia@gmail.com>
Cc: Johannes Weiner <jweiner@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Mel Gorman <mgorman@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit b969c4ab9f upstream.
Stable note: Not tracked in Bugzilla. A fix aimed at preserving page
aging information by reducing LRU list churning had the side-effect
of reducing THP allocation success rates. This was part of a series
to restore the success rates while preserving the reclaim fix.
Asynchronous compaction is used when allocating transparent hugepages to
avoid blocking for long periods of time. Due to reports of stalling,
there was a debate on disabling synchronous compaction but this severely
impacted allocation success rates. Part of the reason was that many dirty
pages are skipped in asynchronous compaction by the following check;
if (PageDirty(page) && !sync &&
mapping->a_ops->migratepage != migrate_page)
rc = -EBUSY;
This skips over all mapping aops using buffer_migrate_page() even though
it is possible to migrate some of these pages without blocking. This
patch updates the ->migratepage callback with a "sync" parameter. It is
the responsibility of the callback to fail gracefully if migration would
block.
Signed-off-by: Mel Gorman <mgorman@suse.de>
Reviewed-by: Rik van Riel <riel@redhat.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Minchan Kim <minchan.kim@gmail.com>
Cc: Dave Jones <davej@redhat.com>
Cc: Jan Kara <jack@suse.cz>
Cc: Andy Isaacson <adi@hexapodia.org>
Cc: Nai Xia <nai.xia@gmail.com>
Cc: Johannes Weiner <jweiner@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Mel Gorman <mgorman@suse.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit c6727932cf upstream.
UBIFS has a feature called "empty space fix-up" which is a quirk to work-around
limitations of dumb flasher programs. Namely, of those flashers that are unable
to skip NAND pages full of 0xFFs while flashing, resulting in empty space at
the end of half-filled eraseblocks to be unusable for UBIFS. This feature is
relatively new (introduced in v3.0).
The fix-up routine (fixup_free_space()) is executed only once at the very first
mount if the superblock has the 'space_fixup' flag set (can be done with -F
option of mkfs.ubifs). It basically reads all the UBIFS data and metadata and
writes it back to the same LEB. The routine assumes the image is pristine and
does not have anything in the journal.
There was a bug in 'fixup_free_space()' where it fixed up the log incorrectly.
All but one LEB of the log of a pristine file-system are empty. And one
contains just a commit start node. And 'fixup_free_space()' just unmapped this
LEB, which resulted in wiping the commit start node. As a result, some users
were unable to mount the file-system next time with the following symptom:
UBIFS error (pid 1): replay_log_leb: first log node at LEB 3:0 is not CS node
UBIFS error (pid 1): replay_log_leb: log error detected while replaying the log at LEB 3:0
The root-cause of this bug was that 'fixup_free_space()' wrongly assumed
that the beginning of empty space in the log head (c->lhead_offs) was known
on mount. However, it is not the case - it was always 0. UBIFS does not store
in it the master node and finds out by scanning the log on every mount.
The fix is simple - just pass commit start node size instead of 0 to
'fixup_leb()'.
Signed-off-by: Artem Bityutskiy <Artem.Bityutskiy@linux.intel.com>
Reported-by: Iwo Mergler <Iwo.Mergler@netcommwireless.com>
Tested-by: Iwo Mergler <Iwo.Mergler@netcommwireless.com>
Reported-by: James Nute <newten82@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit cd60042cc1 upstream.
When we get back a FIND_FIRST/NEXT result, we have some info about the
dentry that we use to instantiate a new inode. We were ignoring and
discarding that info when we had an existing dentry in the cache.
Fix this by updating the inode in place when we find an existing dentry
and the uniqueid is the same.
Reported-and-Tested-by: Andrew Bartlett <abartlet@samba.org>
Reported-by: Bill Robertson <bill_robertson@debortoli.com.au>
Reported-by: Dion Edwards <dion_edwards@debortoli.com.au>
Signed-off-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: Steve French <smfrench@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 05d290d66b upstream.
If a parent and child process open the two ends of a fifo, and the
child immediately exits, the parent may receive a SIGCHLD before its
open() returns. In that case, we need to make sure that open() will
return successfully after the SIGCHLD handler returns, instead of
throwing EINTR or being restarted. Otherwise, the restarted open()
would incorrectly wait for a second partner on the other end.
The following test demonstrates the EINTR that was wrongly thrown from
the parent’s open(). Change .sa_flags = 0 to .sa_flags = SA_RESTART
to see a deadlock instead, in which the restarted open() waits for a
second reader that will never come. (On my systems, this happens
pretty reliably within about 5 to 500 iterations. Others report that
it manages to loop ~forever sometimes; YMMV.)
#include <sys/stat.h>
#include <sys/types.h>
#include <sys/wait.h>
#include <fcntl.h>
#include <signal.h>
#include <stdio.h>
#include <stdlib.h>
#include <unistd.h>
#define CHECK(x) do if ((x) == -1) {perror(#x); abort();} while(0)
void handler(int signum) {}
int main()
{
struct sigaction act = {.sa_handler = handler, .sa_flags = 0};
CHECK(sigaction(SIGCHLD, &act, NULL));
CHECK(mknod("fifo", S_IFIFO | S_IRWXU, 0));
for (;;) {
int fd;
pid_t pid;
putc('.', stderr);
CHECK(pid = fork());
if (pid == 0) {
CHECK(fd = open("fifo", O_RDONLY));
_exit(0);
}
CHECK(fd = open("fifo", O_WRONLY));
CHECK(close(fd));
CHECK(waitpid(pid, NULL, 0));
}
}
This is what I suspect was causing the Git test suite to fail in
t9010-svn-fe.sh:
http://bugs.debian.org/678852
Signed-off-by: Anders Kaseorg <andersk@mit.edu>
Reviewed-by: Jonathan Nieder <jrnieder@gmail.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 91f68c89d8 upstream.
Commit 080399aaaf ("block: don't mark buffers beyond end of disk as
mapped") exposed a bug in __getblk_slow that causes mount to hang as it
loops infinitely waiting for a buffer that lies beyond the end of the
disk to become uptodate.
The problem was initially reported by Torsten Hilbrich here:
https://lkml.org/lkml/2012/6/18/54
and also reported independently here:
http://www.sysresccd.org/forums/viewtopic.php?f=13&t=4511
and then Richard W.M. Jones and Marcos Mello noted a few separate
bugzillas also associated with the same issue. This patch has been
confirmed to fix:
https://bugzilla.redhat.com/show_bug.cgi?id=835019
The main problem is here, in __getblk_slow:
for (;;) {
struct buffer_head * bh;
int ret;
bh = __find_get_block(bdev, block, size);
if (bh)
return bh;
ret = grow_buffers(bdev, block, size);
if (ret < 0)
return NULL;
if (ret == 0)
free_more_memory();
}
__find_get_block does not find the block, since it will not be marked as
mapped, and so grow_buffers is called to fill in the buffers for the
associated page. I believe the for (;;) loop is there primarily to
retry in the case of memory pressure keeping grow_buffers from
succeeding. However, we also continue to loop for other cases, like the
block lying beond the end of the disk. So, the fix I came up with is to
only loop when grow_buffers fails due to memory allocation issues
(return value of 0).
The attached patch was tested by myself, Torsten, and Rich, and was
found to resolve the problem in call cases.
Signed-off-by: Jeff Moyer <jmoyer@redhat.com>
Reported-and-Tested-by: Torsten Hilbrich <torsten.hilbrich@secunet.com>
Tested-by: Richard W.M. Jones <rjones@redhat.com>
Reviewed-by: Josh Boyer <jwboyer@redhat.com>
[ Jens is on vacation, taking this directly - Linus ]
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit fea9f718b3 upstream.
There is a bug in the below scenario for !CONFIG_MMU:
1. create a new file
2. mmap the file and write to it
3. read the file can't get the correct value
Because
sys_read() -> generic_file_aio_read() -> simple_readpage() -> clear_page()
which causes the page to be zeroed.
Add SetPageUptodate() to ramfs_nommu_expand_for_mapping() so that
generic_file_aio_read() do not call simple_readpage().
Signed-off-by: Bob Liu <lliubbo@gmail.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Cc: Greg Ungerer <gerg@uclinux.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 9fe79d7600 upstream.
If the first attempt at opening the lower file read/write fails,
eCryptfs will retry using a privileged kthread. However, the privileged
retry should not happen if the lower file's inode is read-only because a
read/write open will still be unsuccessful.
The check for determining if the open should be retried was intended to
be based on the access mode of the lower file's open flags being
O_RDONLY, but the check was incorrectly performed. This would cause the
open to be retried by the privileged kthread, resulting in a second
failed open of the lower file. This patch corrects the check to
determine if the open request should be handled by the privileged
kthread.
Signed-off-by: Tyler Hicks <tyhicks@canonical.com>
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Acked-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 8dc6780587 upstream.
File operations on /dev/ecryptfs would BUG() when the operations were
performed by processes other than the process that originally opened the
file. This could happen with open files inherited after fork() or file
descriptors passed through IPC mechanisms. Rather than calling BUG(), an
error code can be safely returned in most situations.
In ecryptfs_miscdev_release(), eCryptfs still needs to handle the
release even if the last file reference is being held by a process that
didn't originally open the file. ecryptfs_find_daemon_by_euid() will not
be successful, so a pointer to the daemon is stored in the file's
private_data. The private_data pointer is initialized when the miscdev
file is opened and only used when the file is released.
https://launchpad.net/bugs/994247
Signed-off-by: Tyler Hicks <tyhicks@canonical.com>
Reported-by: Sasha Levin <levinsasha928@gmail.com>
Tested-by: Sasha Levin <levinsasha928@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 332a2e1244 upstream.
We already use them for openat() and friends, but fchdir() also wants to
be able to use O_PATH file descriptors. This should make it comparable
to the O_SEARCH of Solaris. In particular, O_PATH allows you to access
(not-quite-open) a directory you don't have read persmission to, only
execute permission.
Noticed during development of multithread support for ksh93.
Reported-by: ольга крыжановская <olga.kryzhanovska@gmail.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit b6305567e7 upstream.
While we are resolving directory modifications in the
tree log, we are triggering delayed metadata updates to
the filesystem btrees.
This commit forces the delayed updates to run so the
replay code can find any modifications done. It stops
us from crashing because the directory deleltion replay
expects items to be removed immediately from the tree.
Signed-off-by: Chris Mason <chris.mason@fusionio.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 1df2ae31c7 upstream.
Add sanity checks when loading sparing table from disk to avoid accessing
unallocated memory or writing to it.
Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit adee11b208 upstream.
Check provided length of partition table so that (possibly maliciously)
corrupted partition table cannot cause accessing data beyond current buffer.
Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit fbb24a3a91 upstream.
A gc-inode is a pseudo inode used to buffer the blocks to be moved by
garbage collection.
Block caches of gc-inodes must be cleared every time a garbage collection
function (nilfs_clean_segments) completes. Otherwise, stale blocks
buffered in the caches may be wrongly reused in successive calls of the GC
function.
For user files, this is not a problem because their gc-inodes are
distinguished by a checkpoint number as well as an inode number. They
never buffer different blocks if either an inode number, a checkpoint
number, or a block offset differs.
However, gc-inodes of sufile, cpfile and DAT file can store different data
for the same block offset. Thus, the nilfs_clean_segments function can
move incorrect block for these meta-data files if an old block is cached.
I found this is really causing meta-data corruption in nilfs.
This fixes the issue by ensuring cache clear of gc-inodes and resolves
reported GC problems including checkpoint file corruption, b-tree
corruption, and the following warning during GC.
nilfs_palloc_freev: entry number 307234 already freed.
...
Signed-off-by: Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp>
Tested-by: Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit a6dc8c0421 upstream.
The variable io_size was unsigned int, which caused the wrong sector number
to be calculated after aligning it. This then caused mount to fail with big
volumes, as backup volume header information was searched from a
wrong sector.
Signed-off-by: Janne Kalliomäki <janne@tuxera.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 45c72cd73c upstream.
Now we store attr->ino at inode->i_ino, return attr->ino at the
first time and then return inode->i_ino if the attribute timeout
isn't expired. That's wrong on 32 bit platforms because attr->ino
is 64 bit and inode->i_ino is 32 bit in this case.
Fix this by saving 64 bit ino in fuse_inode structure and returning
it every time we call getattr. Also squash attr->ino into inode->i_ino
explicitly.
Signed-off-by: Pavel Shilovsky <piastry@etersoft.ru>
Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit b22b1f178f upstream.
Commit 7990696 uses the ext4_{set,clear}_inode_flags() functions to
change the i_flags automatically but fails to remove the error setting
of i_flags. So we still have the problem of trashing state flags.
Fix this by removing the assignment.
Signed-off-by: Tao Ma <boyu.mt@taobao.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 95599968d1 upstream.
We can't have references held on pages in the s_buddy_cache while we are
trying to truncate its pages and put the inode. All the pages must be
gone before we reach clear_inode. This can only be gauranteed if we
can prevent new users from grabbing references to s_buddy_cache's pages.
The original bug can be reproduced and the bug fix can be verified by:
while true; do mount -t ext4 /dev/ram0 /export/hda3/ram0; \
umount /export/hda3/ram0; done &
while true; do cat /proc/fs/ext4/ram0/mb_groups; done
Signed-off-by: Salman Qazi <sqazi@google.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 02b7831019 upstream.
ext4_free_blocks fails to pair an ext4_mb_load_buddy with a matching
ext4_mb_unload_buddy when it fails a memory allocation.
Signed-off-by: Salman Qazi <sqazi@google.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 79906964a1 upstream.
In commit 353eb83c we removed i_state_flags with 64-bit longs, But
when handling the EXT4_IOC_SETFLAGS ioctl, we replace i_flags
directly, which trashes the state flags which are stored in the high
32-bits of i_flags on 64-bit platforms. So use the the
ext4_{set,clear}_inode_flags() functions which use atomic bit
manipulation functions instead.
Reported-by: Tao Ma <boyu.mt@taobao.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit f3fc0210c0 upstream.
The ext4_error() function is missing a call to save_error_info().
Since this is the function which marks the file system as containing
an error, this oversight (which was introduced in 2.6.36) is quite
significant, and should be backported to older stable kernels with
high urgency.
Reported-by: Ken Sumrall <ksumrall@google.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Cc: ksumrall@google.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 7e84b62164 upstream.
If ext4_setup_super() fails i.e. due to a too-high revision,
the error is logged in dmesg but the fs is not mounted RO as
indicated.
Tested by:
# mkfs.ext4 -r 4 /dev/sdb6
# mount /dev/sdb6 /mnt/test
# dmesg | grep "too high"
[164919.759248] EXT4-fs (sdb6): revision level too high, forcing read-only mode
# grep sdb6 /proc/mounts
/dev/sdb6 /mnt/test2 ext4 rw,seclabel,relatime,data=ordered 0 0
Reviewed-by: Andreas Dilger <adilger@whamcloud.com>
Signed-off-by: Eric Sandeen <sandeen@redhat.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 63d37a84ab upstream.
__mnt_make_shortterm() in there undoes the effect of __mnt_make_longterm()
we'd done back when we set ->mnt_ns non-NULL; it should not be done to
vfsmounts that had never gone through commit_tree() and friends. Kudos to
lczerner for catching that one...
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 2c0c2a08be upstream.
While traversing the linked list of open file handles, if the identfied
file handle is invalid, a reopen is attempted and if it fails, we
resume traversing where we stopped and cifs can oops while accessing
invalid next element, for list might have changed.
So mark the invalid file handle and attempt reopen if no
valid file handle is found in rest of the list.
If reopen fails, move the invalid file handle to the end of the list
and start traversing the list again from the begining.
Repeat this four times before giving up and returning an error if
file reopen keeps failing.
Signed-off-by: Shirish Pargaonkar <shirishpargaonkar@gmail.com>
Reviewed-by: Jeff Layton <jlayton@redhat.com>
Signed-off-by: Steve French <sfrench@us.ibm.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit a70b52ec1a upstream.
We had for some reason overlooked the AIO interface, and it didn't use
the proper rw_verify_area() helper function that checks (for example)
mandatory locking on the file, and that the size of the access doesn't
cause us to overflow the provided offset limits etc.
Instead, AIO did just the security_file_permission() thing (that
rw_verify_area() also does) directly.
This fixes it to do all the proper helper functions, which not only
means that now mandatory file locking works with AIO too, we can
actually remove lines of code.
Reported-by: Manish Honap <manish_honap_vit@yahoo.co.in>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 080399aaaf upstream.
Hi,
We have a bug report open where a squashfs image mounted on ppc64 would
exhibit errors due to trying to read beyond the end of the disk. It can
easily be reproduced by doing the following:
[root@ibm-p750e-02-lp3 ~]# ls -l install.img
-rw-r--r-- 1 root root 142032896 Apr 30 16:46 install.img
[root@ibm-p750e-02-lp3 ~]# mount -o loop ./install.img /mnt/test
[root@ibm-p750e-02-lp3 ~]# dd if=/dev/loop0 of=/dev/null
dd: reading `/dev/loop0': Input/output error
277376+0 records in
277376+0 records out
142016512 bytes (142 MB) copied, 0.9465 s, 150 MB/s
In dmesg, you'll find the following:
squashfs: version 4.0 (2009/01/31) Phillip Lougher
[ 43.106012] attempt to access beyond end of device
[ 43.106029] loop0: rw=0, want=277410, limit=277408
[ 43.106039] Buffer I/O error on device loop0, logical block 138704
[ 43.106053] attempt to access beyond end of device
[ 43.106057] loop0: rw=0, want=277412, limit=277408
[ 43.106061] Buffer I/O error on device loop0, logical block 138705
[ 43.106066] attempt to access beyond end of device
[ 43.106070] loop0: rw=0, want=277414, limit=277408
[ 43.106073] Buffer I/O error on device loop0, logical block 138706
[ 43.106078] attempt to access beyond end of device
[ 43.106081] loop0: rw=0, want=277416, limit=277408
[ 43.106085] Buffer I/O error on device loop0, logical block 138707
[ 43.106089] attempt to access beyond end of device
[ 43.106093] loop0: rw=0, want=277418, limit=277408
[ 43.106096] Buffer I/O error on device loop0, logical block 138708
[ 43.106101] attempt to access beyond end of device
[ 43.106104] loop0: rw=0, want=277420, limit=277408
[ 43.106108] Buffer I/O error on device loop0, logical block 138709
[ 43.106112] attempt to access beyond end of device
[ 43.106116] loop0: rw=0, want=277422, limit=277408
[ 43.106120] Buffer I/O error on device loop0, logical block 138710
[ 43.106124] attempt to access beyond end of device
[ 43.106128] loop0: rw=0, want=277424, limit=277408
[ 43.106131] Buffer I/O error on device loop0, logical block 138711
[ 43.106135] attempt to access beyond end of device
[ 43.106139] loop0: rw=0, want=277426, limit=277408
[ 43.106143] Buffer I/O error on device loop0, logical block 138712
[ 43.106147] attempt to access beyond end of device
[ 43.106151] loop0: rw=0, want=277428, limit=277408
[ 43.106154] Buffer I/O error on device loop0, logical block 138713
[ 43.106158] attempt to access beyond end of device
[ 43.106162] loop0: rw=0, want=277430, limit=277408
[ 43.106166] attempt to access beyond end of device
[ 43.106169] loop0: rw=0, want=277432, limit=277408
...
[ 43.106307] attempt to access beyond end of device
[ 43.106311] loop0: rw=0, want=277470, limit=2774
Squashfs manages to read in the end block(s) of the disk during the
mount operation. Then, when dd reads the block device, it leads to
block_read_full_page being called with buffers that are beyond end of
disk, but are marked as mapped. Thus, it would end up submitting read
I/O against them, resulting in the errors mentioned above. I fixed the
problem by modifying init_page_buffers to only set the buffer mapped if
it fell inside of i_size.
Cheers,
Jeff
Signed-off-by: Jeff Moyer <jmoyer@redhat.com>
Acked-by: Nick Piggin <npiggin@kernel.dk>
--
Changes from v1->v2: re-used max_block, as suggested by Nick Piggin.
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit e1616300a2 upstream.
dd slept infinitely when fsfeeze failed because of EIO.
To fix this problem, if ->freeze_fs fails, freeze_super() wakes up
the tasks waiting for the filesystem to become unfrozen.
When s_frozen isn't SB_UNFROZEN in __generic_file_aio_write(),
the function sleeps until FITHAW ioctl wakes up s_wait_unfrozen.
However, if ->freeze_fs fails, s_frozen is set to SB_UNFROZEN and then
freeze_super() returns an error number. In this case, FITHAW ioctl returns
EINVAL because s_frozen is already SB_UNFROZEN. There is no way to wake up
s_wait_unfrozen, so __generic_file_aio_write() sleeps infinitely.
Signed-off-by: Kazuya Mio <k-mio@sx.jp.nec.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit acd6ad8351 upstream.
When insert_inode_locked() fails in ext4_new_inode() it most likely means inode
bitmap got corrupted and we allocated again inode which is already in use. Also
doing unlock_new_inode() during error recovery is wrong since the inode does
not have I_NEW set. Fix the problem by jumping to fail: (instead of fail_drop:)
which declares filesystem error and does not call unlock_new_inode().
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>
commit 1415dd8705 upstream.
When insert_inode_locked() fails in ext3_new_inode() it most likely
means inode bitmap got corrupted and we allocated again inode which
is already in use. Also doing unlock_new_inode() during error recovery
is wrong since inode does not have I_NEW set. Fix the problem by jumping
to fail: (instead of fail_drop:) which declares filesystem error and
does not call unlock_new_inode().
Reviewed-by: Eric Sandeen <sandeen@redhat.com>
Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This is a shorter (and more appropriate for stable kernels) analog to
the following upstream commit:
commit 6926afd192
Author: Trond Myklebust <Trond.Myklebust@netapp.com>
Date: Sat Jan 7 13:22:46 2012 -0500
NFSv4: Save the owner/group name string when doing open
...so that we can do the uid/gid mapping outside the asynchronous RPC
context.
This fixes a bug in the current NFSv4 atomic open code where the client
isn't able to determine what the true uid/gid fields of the file are,
(because the asynchronous nature of the OPEN call denies it the ability
to do an upcall) and so fills them with default values, marking the
inode as needing revalidation.
Unfortunately, in some cases, the VFS will do some additional sanity
checks on the file, and may override the server's decision to allow
the open because it sees the wrong owner/group fields.
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Without this patch, logging into two different machines with home
directories mounted over NFS4 and then running "vim" and typing ":q"
in each reliably produces the following error on the second machine:
E137: Viminfo file is not writable: /users/system/rtheys/.viminfo
This regression was introduced by 80e52aced1 ("NFSv4: Don't do
idmapper upcalls for asynchronous RPC calls", merged during the 2.6.32
cycle) --- after the OPEN call, .viminfo has the default values for
st_uid and st_gid (0xfffffffe) cached because we do not want to let
rpciod wait for an idmapper upcall to fill them in.
The fix used in mainline is to save the owner and group as strings and
perform the upcall in _nfs4_proc_open outside the rpciod context,
which takes about 600 lines. For stable, we can do something similar
with a one-liner: make open check for the stale fields and make a
(synchronous) GETATTR call to fill them when needed.
Trond dictated the patch, I typed it in, and Rik tested it.
Addresses http://bugs.debian.org/659111 and
https://bugzilla.redhat.com/789298
Reported-by: Rik Theys <Rik.Theys@esat.kuleuven.be>
Explained-by: David Flyn <davidf@rd.bbc.co.uk>
Signed-off-by: Jonathan Nieder <jrnieder@gmail.com>
Tested-by: Rik Theys <Rik.Theys@esat.kuleuven.be>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit c1bb05a657 upstream.
Processes hang forever on a sync-mounted ext2 file system that
is mounted with the ext4 module (default in Fedora 16).
I can reproduce this reliably by mounting an ext2 partition with
"-o sync" and opening a new file an that partition with vim. vim
will hang in "D" state forever. The same happens on ext4 without
a journal.
I am attaching a small patch here that solves this issue for me.
In the sync mounted case without a journal,
ext4_handle_dirty_metadata() may call sync_dirty_buffer(), which
can't be called with buffer lock held.
Also move mb_cache_entry_release inside lock to avoid race
fixed previously by 8a2bfdcb ext[34]: EA block reference count racing fix
Note too that ext2 fixed this same problem in 2006 with
b2f49033 [PATCH] fix deadlock in ext2
Signed-off-by: Martin.Wilck@ts.fujitsu.com
[sandeen@redhat.com: move mb_cache_entry_release before unlock, edit commit msg]
Signed-off-by: Eric Sandeen <sandeen@redhat.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 226bb7df3d upstream.
The locking policy is such that the erase_complete_block spinlock is
nested within the alloc_sem mutex. This fixes a case in which the
acquisition order was erroneously reversed. This issue was caught by
the following lockdep splat:
=======================================================
[ INFO: possible circular locking dependency detected ]
3.0.5 #1
-------------------------------------------------------
jffs2_gcd_mtd6/299 is trying to acquire lock:
(&c->alloc_sem){+.+.+.}, at: [<c01f7714>] jffs2_garbage_collect_pass+0x314/0x890
but task is already holding lock:
(&(&c->erase_completion_lock)->rlock){+.+...}, at: [<c01f7708>] jffs2_garbage_collect_pass+0x308/0x890
which lock already depends on the new lock.
the existing dependency chain (in reverse order) is:
-> #1 (&(&c->erase_completion_lock)->rlock){+.+...}:
[<c008bec4>] validate_chain+0xe6c/0x10bc
[<c008c660>] __lock_acquire+0x54c/0xba4
[<c008d240>] lock_acquire+0xa4/0x114
[<c046780c>] _raw_spin_lock+0x3c/0x4c
[<c01f744c>] jffs2_garbage_collect_pass+0x4c/0x890
[<c01f937c>] jffs2_garbage_collect_thread+0x1b4/0x1cc
[<c0071a68>] kthread+0x98/0xa0
[<c000f264>] kernel_thread_exit+0x0/0x8
-> #0 (&c->alloc_sem){+.+.+.}:
[<c008ad2c>] print_circular_bug+0x70/0x2c4
[<c008c08c>] validate_chain+0x1034/0x10bc
[<c008c660>] __lock_acquire+0x54c/0xba4
[<c008d240>] lock_acquire+0xa4/0x114
[<c0466628>] mutex_lock_nested+0x74/0x33c
[<c01f7714>] jffs2_garbage_collect_pass+0x314/0x890
[<c01f937c>] jffs2_garbage_collect_thread+0x1b4/0x1cc
[<c0071a68>] kthread+0x98/0xa0
[<c000f264>] kernel_thread_exit+0x0/0x8
other info that might help us debug this:
Possible unsafe locking scenario:
CPU0 CPU1
---- ----
lock(&(&c->erase_completion_lock)->rlock);
lock(&c->alloc_sem);
lock(&(&c->erase_completion_lock)->rlock);
lock(&c->alloc_sem);
*** DEADLOCK ***
1 lock held by jffs2_gcd_mtd6/299:
#0: (&(&c->erase_completion_lock)->rlock){+.+...}, at: [<c01f7708>] jffs2_garbage_collect_pass+0x308/0x890
stack backtrace:
[<c00155dc>] (unwind_backtrace+0x0/0x100) from [<c0463dc0>] (dump_stack+0x20/0x24)
[<c0463dc0>] (dump_stack+0x20/0x24) from [<c008ae84>] (print_circular_bug+0x1c8/0x2c4)
[<c008ae84>] (print_circular_bug+0x1c8/0x2c4) from [<c008c08c>] (validate_chain+0x1034/0x10bc)
[<c008c08c>] (validate_chain+0x1034/0x10bc) from [<c008c660>] (__lock_acquire+0x54c/0xba4)
[<c008c660>] (__lock_acquire+0x54c/0xba4) from [<c008d240>] (lock_acquire+0xa4/0x114)
[<c008d240>] (lock_acquire+0xa4/0x114) from [<c0466628>] (mutex_lock_nested+0x74/0x33c)
[<c0466628>] (mutex_lock_nested+0x74/0x33c) from [<c01f7714>] (jffs2_garbage_collect_pass+0x314/0x890)
[<c01f7714>] (jffs2_garbage_collect_pass+0x314/0x890) from [<c01f937c>] (jffs2_garbage_collect_thread+0x1b4/0x1cc)
[<c01f937c>] (jffs2_garbage_collect_thread+0x1b4/0x1cc) from [<c0071a68>] (kthread+0x98/0xa0)
[<c0071a68>] (kthread+0x98/0xa0) from [<c000f264>] (kernel_thread_exit+0x0/0x8)
This was introduce in '81cfc9f jffs2: Fix serious write stall due to erase'.
Signed-off-by: Josh Cartwright <joshc@linux.com>
Signed-off-by: Artem Bityutskiy <artem.bityutskiy@linux.intel.com>
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 64f371bc31 upstream.
The autofs packet size has had a very unfortunate size problem on x86:
because the alignment of 'u64' differs in 32-bit and 64-bit modes, and
because the packet data was not 8-byte aligned, the size of the autofsv5
packet structure differed between 32-bit and 64-bit modes despite
looking otherwise identical (300 vs 304 bytes respectively).
We first fixed that up by making the 64-bit compat mode know about this
problem in commit a32744d4ab ("autofs: work around unhappy compat
problem on x86-64"), and that made a 32-bit 'systemd' work happily on a
64-bit kernel because everything then worked the same way as on a 32-bit
kernel.
But it turned out that 'automount' had actually known and worked around
this problem in user space, so fixing the kernel to do the proper 32-bit
compatibility handling actually *broke* 32-bit automount on a 64-bit
kernel, because it knew that the packet sizes were wrong and expected
those incorrect sizes.
As a result, we ended up reverting that compatibility mode fix, and
thus breaking systemd again, in commit fcbf94b9de.
With both automount and systemd doing a single read() system call, and
verifying that they get *exactly* the size they expect but using
different sizes, it seemed that fixing one of them inevitably seemed to
break the other. At one point, a patch I seriously considered applying
from Michael Tokarev did a "strcmp()" to see if it was automount that
was doing the operation. Ugly, ugly.
However, a prettier solution exists now thanks to the packetized pipe
mode. By marking the communication pipe as being packetized (by simply
setting the O_DIRECT flag), we can always just write the bigger packet
size, and if user-space does a smaller read, it will just get that
partial end result and the extra alignment padding will simply be thrown
away.
This makes both automount and systemd happy, since they now get the size
they asked for, and the kernel side of autofs simply no longer needs to
care - it could pad out the packet arbitrarily.
Of course, if there is some *other* user of autofs (please, please,
please tell me it ain't so - and we haven't heard of any) that tries to
read the packets with multiple writes, that other user will now be
broken - the whole point of the packetized mode is that one system call
gets exactly one packet, and you cannot read a packet in pieces.
Tested-by: Michael Tokarev <mjt@tls.msk.ru>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Cc: David Miller <davem@davemloft.net>
Cc: Ian Kent <raven@themaw.net>
Cc: Thomas Meyer <thomas@m3y3r.de>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 9883035ae7 upstream.
The actual internal pipe implementation is already really about
individual packets (called "pipe buffers"), and this simply exposes that
as a special packetized mode.
When we are in the packetized mode (marked by O_DIRECT as suggested by
Alan Cox), a write() on a pipe will not merge the new data with previous
writes, so each write will get a pipe buffer of its own. The pipe
buffer is then marked with the PIPE_BUF_FLAG_PACKET flag, which in turn
will tell the reader side to break the read at that boundary (and throw
away any partial packet contents that do not fit in the read buffer).
End result: as long as you do writes less than PIPE_BUF in size (so that
the pipe doesn't have to split them up), you can now treat the pipe as a
packet interface, where each read() system call will read one packet at
a time. You can just use a sufficiently big read buffer (PIPE_BUF is
sufficient, since bigger than that doesn't guarantee atomicity anyway),
and the return value of the read() will naturally give you the size of
the packet.
NOTE! We do not support zero-sized packets, and zero-sized reads and
writes to a pipe continue to be no-ops. Also note that big packets will
currently be split at write time, but that the size at which that
happens is not really specified (except that it's bigger than PIPE_BUF).
Currently that limit is the system page size, but we might want to
explicitly support bigger packets some day.
The main user for this is going to be the autofs packet interface,
allowing us to stop having to care so deeply about exact packet sizes
(which have had bugs with 32/64-bit compatibility modes). But user
space can create packetized pipes with "pipe2(fd, O_DIRECT)", which will
fail with an EINVAL on kernels that do not support this interface.
Tested-by: Michael Tokarev <mjt@tls.msk.ru>
Cc: Alan Cox <alan@lxorguk.ukuu.org.uk>
Cc: David Miller <davem@davemloft.net>
Cc: Ian Kent <raven@themaw.net>
Cc: Thomas Meyer <thomas@m3y3r.de>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 04da6e9d63 upstream.
nfsd_open() already returns an NFS error value; only vfs_test_lock()
result needs to be fed through nfserrno(). Broken by commit 55ef12
(nfsd: Ensure nfsv4 calls the underlying filesystem on LOCKT)
three years ago...
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Jonathan Nieder <jrnieder@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit fcbf94b9de upstream.
This reverts commit a32744d4ab.
While that commit was technically the right thing to do, and made the
x86-64 compat mode work identically to native 32-bit mode (and thus
fixing the problem with a 32-bit systemd install on a 64-bit kernel), it
turns out that the automount binaries had workarounds for this compat
problem.
Now, the workarounds are disgusting: doing an "uname()" to find out the
architecture of the kernel, and then comparing it for the 64-bit cases
and fixing up the size of the read() in automount for those. And they
were confused: it's not actually a generic 64-bit issue at all, it's
very much tied to just x86-64, which has different alignment for an
'u64' in 64-bit mode than in 32-bit mode.
But the end result is that fixing the compat layer actually breaks the
case of a 32-bit automount on a x86-64 kernel.
There are various approaches to fix this (including just doing a
"strcmp()" on current->comm and comparing it to "automount"), but I
think that I will do the one that teaches pipes about a special "packet
mode", which will allow user space to not have to care too deeply about
the padding at the end of the autofs packet.
That change will make the compat workaround unnecessary, so let's revert
it first, and get automount working again in compat mode. The
packetized pipes will then fix autofs for systemd.
Reported-and-requested-by: Michael Tokarev <mjt@tls.msk.ru>
Cc: Ian Kent <raven@themaw.net>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 55725513b5 upstream.
Since we may be simulating flock() locks using NFS byte range locks,
we can't rely on the VFS having checked the file open mode for us.
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 05ffe24f52 upstream.
All callers of nfs4_handle_exception() that need to handle
NFS4ERR_OPENMODE correctly should set exception->inode
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 98a2139f4f upstream.
When hostname contains colon (e.g. when it is an IPv6 address) it needs
to be enclosed in brackets to make parsing of NFS device string possible.
Fix nfs_do_root_mount() to enclose hostname properly when needed. NFS code
actually does not need this as it does not parse the string passed by
nfs_do_root_mount() but the device string is exposed to userspace in
/proc/mounts.
CC: Josh Boyer <jwboyer@redhat.com>
CC: Trond Myklebust <Trond.Myklebust@netapp.com>
Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[ This combines upstream commit
2f53384424 and the follow-on bug fix
commit 35f9c09fe9 ]
vmsplice()/splice(pipe, socket) call do_tcp_sendpages() one page at a
time, adding at most 4096 bytes to an skb. (assuming PAGE_SIZE=4096)
The call to tcp_push() at the end of do_tcp_sendpages() forces an
immediate xmit when pipe is not already filled, and tso_fragment() try
to split these skb to MSS multiples.
4096 bytes are usually split in a skb with 2 MSS, and a remaining
sub-mss skb (assuming MTU=1500)
This makes slow start suboptimal because many small frames are sent to
qdisc/driver layers instead of big ones (constrained by cwnd and packets
in flight of course)
In fact, applications using sendmsg() (adding an additional memory copy)
instead of vmsplice()/splice()/sendfile() are a bit faster because of
this anomaly, especially if serving small files in environments with
large initial [c]wnd.
Call tcp_push() only if MSG_MORE is not set in the flags parameter.
This bit is automatically provided by splice() internals but for the
last page, or on all pages if user specified SPLICE_F_MORE splice()
flag.
In some workloads, this can reduce number of sent logical packets by an
order of magnitude, making zero-copy TCP actually faster than
one-copy :)
Reported-by: Tom Herbert <therbert@google.com>
Cc: Nandita Dukkipati <nanditad@google.com>
Cc: Neal Cardwell <ncardwell@google.com>
Cc: Tom Herbert <therbert@google.com>
Cc: Yuchung Cheng <ycheng@google.com>
Cc: H.K. Jerry Chu <hkchu@google.com>
Cc: Maciej Żenczykowski <maze@google.com>
Cc: Mahesh Bandewar <maheshb@google.com>
Cc: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi>
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
