This is a patch that provides behavior that is more consistent, and
probably less surprising to users. I consider the change optional, and
welcome opinions about whether it should be applied.
By default, pipes are created with a capacity of 64 kiB. However,
/proc/sys/fs/pipe-max-size may be set smaller than this value. In this
scenario, an unprivileged user could thus create a pipe whose initial
capacity exceeds the limit. Therefore, it seems logical to cap the
initial pipe capacity according to the value of pipe-max-size.
The test program shown earlier in this patch series can be used to
demonstrate the effect of the change brought about with this patch:
# cat /proc/sys/fs/pipe-max-size
1048576
# sudo -u mtk ./test_F_SETPIPE_SZ 1
Initial pipe capacity: 65536
# echo 10000 > /proc/sys/fs/pipe-max-size
# cat /proc/sys/fs/pipe-max-size
16384
# sudo -u mtk ./test_F_SETPIPE_SZ 1
Initial pipe capacity: 16384
# ./test_F_SETPIPE_SZ 1
Initial pipe capacity: 65536
The last two executions of 'test_F_SETPIPE_SZ' show that pipe-max-size
caps the initial allocation for a new pipe for unprivileged users, but
not for privileged users.
Link: http://lkml.kernel.org/r/31dc7064-2a17-9c5b-1df1-4e3012ee992c@gmail.com
Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>
Reviewed-by: Vegard Nossum <vegard.nossum@oracle.com>
Cc: Willy Tarreau <w@1wt.eu>
Cc: <socketpair@gmail.com>
Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Cc: Jens Axboe <axboe@fb.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This is an optional patch, to provide a small performance
improvement. Alter account_pipe_buffers() so that it returns the
new value in user->pipe_bufs. This means that we can refactor
too_many_pipe_buffers_soft() and too_many_pipe_buffers_hard() to
avoid the costs of repeated use of atomic_long_read() to get the
value user->pipe_bufs.
Link: http://lkml.kernel.org/r/93e5f193-1e5e-3e1f-3a20-eae79b7e1310@gmail.com
Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>
Reviewed-by: Vegard Nossum <vegard.nossum@oracle.com>
Cc: Willy Tarreau <w@1wt.eu>
Cc: <socketpair@gmail.com>
Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Cc: Jens Axboe <axboe@fb.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The limit checking in alloc_pipe_info() (used by pipe(2) and when
opening a FIFO) has the following problems:
(1) When checking capacity required for the new pipe, the checks against
the limit in /proc/sys/fs/pipe-user-pages-{soft,hard} are made
against existing consumption, and exclude the memory required for
the new pipe capacity. As a consequence: (1) the memory allocation
throttling provided by the soft limit does not kick in quite as
early as it should, and (2) the user can overrun the hard limit.
(2) As currently implemented, accounting and checking against the limits
is done as follows:
(a) Test whether the user has exceeded the limit.
(b) Make new pipe buffer allocation.
(c) Account new allocation against the limits.
This is racey. Multiple processes may pass point (a) simultaneously,
and then allocate pipe buffers that are accounted for only in step
(c). The race means that the user's pipe buffer allocation could be
pushed over the limit (by an arbitrary amount, depending on how
unlucky we were in the race). [Thanks to Vegard Nossum for spotting
this point, which I had missed.]
This patch addresses the above problems as follows:
* Alter the checks against limits to include the memory required for the
new pipe.
* Re-order the accounting step so that it precedes the buffer allocation.
If the accounting step determines that a limit has been reached, revert
the accounting and cause the operation to fail.
Link: http://lkml.kernel.org/r/8ff3e9f9-23f6-510c-644f-8e70cd1c0bd9@gmail.com
Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>
Reviewed-by: Vegard Nossum <vegard.nossum@oracle.com>
Cc: Willy Tarreau <w@1wt.eu>
Cc: <socketpair@gmail.com>
Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Cc: Jens Axboe <axboe@fb.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Replace an 'if' block that covers most of the code in this function
with a 'goto'. This makes the code a little simpler to read, and also
simplifies the next patch (fix limit checking in alloc_pipe_info())
Link: http://lkml.kernel.org/r/aef030c1-0257-98a9-4988-186efa48530c@gmail.com
Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>
Reviewed-by: Vegard Nossum <vegard.nossum@oracle.com>
Cc: Willy Tarreau <w@1wt.eu>
Cc: <socketpair@gmail.com>
Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Cc: Jens Axboe <axboe@fb.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The limit checking in pipe_set_size() (used by fcntl(F_SETPIPE_SZ))
has the following problems:
(1) When increasing the pipe capacity, the checks against the limits in
/proc/sys/fs/pipe-user-pages-{soft,hard} are made against existing
consumption, and exclude the memory required for the increased pipe
capacity. The new increase in pipe capacity can then push the total
memory used by the user for pipes (possibly far) over a limit. This
can also trigger the problem described next.
(2) The limit checks are performed even when the new pipe capacity is
less than the existing pipe capacity. This can lead to problems if a
user sets a large pipe capacity, and then the limits are lowered,
with the result that the user will no longer be able to decrease the
pipe capacity.
(3) As currently implemented, accounting and checking against the
limits is done as follows:
(a) Test whether the user has exceeded the limit.
(b) Make new pipe buffer allocation.
(c) Account new allocation against the limits.
This is racey. Multiple processes may pass point (a)
simultaneously, and then allocate pipe buffers that are accounted
for only in step (c). The race means that the user's pipe buffer
allocation could be pushed over the limit (by an arbitrary amount,
depending on how unlucky we were in the race). [Thanks to Vegard
Nossum for spotting this point, which I had missed.]
This patch addresses the above problems as follows:
* Perform checks against the limits only when increasing a pipe's
capacity; an unprivileged user can always decrease a pipe's capacity.
* Alter the checks against limits to include the memory required for
the new pipe capacity.
* Re-order the accounting step so that it precedes the buffer
allocation. If the accounting step determines that a limit has
been reached, revert the accounting and cause the operation to fail.
The program below can be used to demonstrate problems 1 and 2, and the
effect of the fix. The program takes one or more command-line arguments.
The first argument specifies the number of pipes that the program should
create. The remaining arguments are, alternately, pipe capacities that
should be set using fcntl(F_SETPIPE_SZ), and sleep intervals (in
seconds) between the fcntl() operations. (The sleep intervals allow the
possibility to change the limits between fcntl() operations.)
Problem 1
=========
Using the test program on an unpatched kernel, we first set some
limits:
# echo 0 > /proc/sys/fs/pipe-user-pages-soft
# echo 1000000000 > /proc/sys/fs/pipe-max-size
# echo 10000 > /proc/sys/fs/pipe-user-pages-hard # 40.96 MB
Then show that we can set a pipe with capacity (100MB) that is
over the hard limit
# sudo -u mtk ./test_F_SETPIPE_SZ 1 100000000
Initial pipe capacity: 65536
Loop 1: set pipe capacity to 100000000 bytes
F_SETPIPE_SZ returned 134217728
Now set the capacity to 100MB twice. The second call fails (which is
probably surprising to most users, since it seems like a no-op):
# sudo -u mtk ./test_F_SETPIPE_SZ 1 100000000 0 100000000
Initial pipe capacity: 65536
Loop 1: set pipe capacity to 100000000 bytes
F_SETPIPE_SZ returned 134217728
Loop 2: set pipe capacity to 100000000 bytes
Loop 2, pipe 0: F_SETPIPE_SZ failed: fcntl: Operation not permitted
With a patched kernel, setting a capacity over the limit fails at the
first attempt:
# echo 0 > /proc/sys/fs/pipe-user-pages-soft
# echo 1000000000 > /proc/sys/fs/pipe-max-size
# echo 10000 > /proc/sys/fs/pipe-user-pages-hard
# sudo -u mtk ./test_F_SETPIPE_SZ 1 100000000
Initial pipe capacity: 65536
Loop 1: set pipe capacity to 100000000 bytes
Loop 1, pipe 0: F_SETPIPE_SZ failed: fcntl: Operation not permitted
There is a small chance that the change to fix this problem could
break user-space, since there are cases where fcntl(F_SETPIPE_SZ)
calls that previously succeeded might fail. However, the chances are
small, since (a) the pipe-user-pages-{soft,hard} limits are new (in
4.5), and the default soft/hard limits are high/unlimited. Therefore,
it seems warranted to make these limits operate more precisely (and
behave more like what users probably expect).
Problem 2
=========
Running the test program on an unpatched kernel, we first set some limits:
# getconf PAGESIZE
4096
# echo 0 > /proc/sys/fs/pipe-user-pages-soft
# echo 1000000000 > /proc/sys/fs/pipe-max-size
# echo 10000 > /proc/sys/fs/pipe-user-pages-hard # 40.96 MB
Now perform two fcntl(F_SETPIPE_SZ) operations on a single pipe,
first setting a pipe capacity (10MB), sleeping for a few seconds,
during which time the hard limit is lowered, and then set pipe
capacity to a smaller amount (5MB):
# sudo -u mtk ./test_F_SETPIPE_SZ 1 10000000 15 5000000 &
[1] 748
# Initial pipe capacity: 65536
Loop 1: set pipe capacity to 10000000 bytes
F_SETPIPE_SZ returned 16777216
Sleeping 15 seconds
# echo 1000 > /proc/sys/fs/pipe-user-pages-hard # 4.096 MB
# Loop 2: set pipe capacity to 5000000 bytes
Loop 2, pipe 0: F_SETPIPE_SZ failed: fcntl: Operation not permitted
In this case, the user should be able to lower the limit.
With a kernel that has the patch below, the second fcntl()
succeeds:
# echo 0 > /proc/sys/fs/pipe-user-pages-soft
# echo 1000000000 > /proc/sys/fs/pipe-max-size
# echo 10000 > /proc/sys/fs/pipe-user-pages-hard
# sudo -u mtk ./test_F_SETPIPE_SZ 1 10000000 15 5000000 &
[1] 3215
# Initial pipe capacity: 65536
# Loop 1: set pipe capacity to 10000000 bytes
F_SETPIPE_SZ returned 16777216
Sleeping 15 seconds
# echo 1000 > /proc/sys/fs/pipe-user-pages-hard
# Loop 2: set pipe capacity to 5000000 bytes
F_SETPIPE_SZ returned 8388608
8x---8x---8x---8x---8x---8x---8x---8x---8x---8x---8x---8x---8x---8x---
/* test_F_SETPIPE_SZ.c
(C) 2016, Michael Kerrisk; licensed under GNU GPL version 2 or later
Test operation of fcntl(F_SETPIPE_SZ) for setting pipe capacity
and interactions with limits defined by /proc/sys/fs/pipe-* files.
*/
#define _GNU_SOURCE
#include <stdio.h>
#include <stdlib.h>
#include <fcntl.h>
#include <unistd.h>
int
main(int argc, char *argv[])
{
int (*pfd)[2];
int npipes;
int pcap, rcap;
int j, p, s, stime, loop;
if (argc < 2) {
fprintf(stderr, "Usage: %s num-pipes "
"[pipe-capacity sleep-time]...\n", argv[0]);
exit(EXIT_FAILURE);
}
npipes = atoi(argv[1]);
pfd = calloc(npipes, sizeof (int [2]));
if (pfd == NULL) {
perror("calloc");
exit(EXIT_FAILURE);
}
for (j = 0; j < npipes; j++) {
if (pipe(pfd[j]) == -1) {
fprintf(stderr, "Loop %d: pipe() failed: ", j);
perror("pipe");
exit(EXIT_FAILURE);
}
}
printf("Initial pipe capacity: %d\n", fcntl(pfd[0][0], F_GETPIPE_SZ));
for (j = 2; j < argc; j += 2 ) {
loop = j / 2;
pcap = atoi(argv[j]);
printf(" Loop %d: set pipe capacity to %d bytes\n", loop, pcap);
for (p = 0; p < npipes; p++) {
s = fcntl(pfd[p][0], F_SETPIPE_SZ, pcap);
if (s == -1) {
fprintf(stderr, " Loop %d, pipe %d: F_SETPIPE_SZ "
"failed: ", loop, p);
perror("fcntl");
exit(EXIT_FAILURE);
}
if (p == 0) {
printf(" F_SETPIPE_SZ returned %d\n", s);
rcap = s;
} else {
if (s != rcap) {
fprintf(stderr, " Loop %d, pipe %d: F_SETPIPE_SZ "
"unexpected return: %d\n", loop, p, s);
exit(EXIT_FAILURE);
}
}
stime = (j + 1 < argc) ? atoi(argv[j + 1]) : 0;
if (stime > 0) {
printf(" Sleeping %d seconds\n", stime);
sleep(stime);
}
}
}
exit(EXIT_SUCCESS);
}
8x---8x---8x---8x---8x---8x---8x---8x---8x---8x---8x---8x---8x---8x---
Patch history:
v2
* Switch order of test in 'if' statement to avoid function call
(to capability()) in normal path. [This is a fix to a preexisting
wart in the code. Thanks to Willy Tarreau]
* Perform (size > pipe_max_size) check before calling
account_pipe_buffers(). [Thanks to Vegard Nossum]
Quoting Vegard:
The potential problem happens if the user passes a very large number
which will overflow pipe->user->pipe_bufs.
On 32-bit, sizeof(int) == sizeof(long), so if they pass arg = INT_MAX
then round_pipe_size() returns INT_MAX. Although it's true that the
accounting is done in terms of pages and not bytes, so you'd need on
the order of (1 << 13) = 8192 processes hitting the limit at the same
time in order to make it overflow, which seems a bit unlikely.
(See https://lkml.org/lkml/2016/8/12/215 for another discussion on the
limit checking)
Link: http://lkml.kernel.org/r/1e464945-536b-2420-798b-e77b9c7e8593@gmail.com
Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>
Reviewed-by: Vegard Nossum <vegard.nossum@oracle.com>
Cc: Willy Tarreau <w@1wt.eu>
Cc: <socketpair@gmail.com>
Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Cc: Jens Axboe <axboe@fb.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This is a preparatory patch for following work. account_pipe_buffers()
performs accounting in the 'user_struct'. There is no need to pass a
pointer to a 'pipe_inode_info' struct (which is then dereferenced to
obtain a pointer to the 'user' field). Instead, pass a pointer directly
to the 'user_struct'. This change is needed in preparation for a
subsequent patch that the fixes the limit checking in alloc_pipe_info()
(and the resulting code is a little more logical).
Link: http://lkml.kernel.org/r/7277bf8c-a6fc-4a7d-659c-f5b145c981ab@gmail.com
Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>
Reviewed-by: Vegard Nossum <vegard.nossum@oracle.com>
Cc: Willy Tarreau <w@1wt.eu>
Cc: <socketpair@gmail.com>
Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Cc: Jens Axboe <axboe@fb.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This is a preparatory patch for following work. Move the F_SETPIPE_SZ
limit-checking logic from pipe_fcntl() into pipe_set_size(). This
simplifies the code a little, and allows for reworking required in
a later patch that fixes the limit checking in pipe_set_size()
Link: http://lkml.kernel.org/r/3701b2c5-2c52-2c3e-226d-29b9deb29b50@gmail.com
Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>
Reviewed-by: Vegard Nossum <vegard.nossum@oracle.com>
Cc: Willy Tarreau <w@1wt.eu>
Cc: <socketpair@gmail.com>
Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Cc: Jens Axboe <axboe@fb.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Patch series "pipe: fix limit handling", v2.
When changing a pipe's capacity with fcntl(F_SETPIPE_SZ), various limits
defined by /proc/sys/fs/pipe-* files are checked to see if unprivileged
users are exceeding limits on memory consumption.
While documenting and testing the operation of these limits I noticed
that, as currently implemented, these checks have a number of problems:
(1) When increasing the pipe capacity, the checks against the limits
in /proc/sys/fs/pipe-user-pages-{soft,hard} are made against
existing consumption, and exclude the memory required for the
increased pipe capacity. The new increase in pipe capacity can then
push the total memory used by the user for pipes (possibly far) over
a limit. This can also trigger the problem described next.
(2) The limit checks are performed even when the new pipe capacity
is less than the existing pipe capacity. This can lead to problems
if a user sets a large pipe capacity, and then the limits are
lowered, with the result that the user will no longer be able to
decrease the pipe capacity.
(3) As currently implemented, accounting and checking against the
limits is done as follows:
(a) Test whether the user has exceeded the limit.
(b) Make new pipe buffer allocation.
(c) Account new allocation against the limits.
This is racey. Multiple processes may pass point (a) simultaneously,
and then allocate pipe buffers that are accounted for only in step
(c). The race means that the user's pipe buffer allocation could be
pushed over the limit (by an arbitrary amount, depending on how
unlucky we were in the race). [Thanks to Vegard Nossum for spotting
this point, which I had missed.]
This patch series addresses these three problems.
This patch (of 8):
This is a minor preparatory patch. After subsequent patches,
round_pipe_size() will be called from pipe_set_size(), so place
round_pipe_size() above pipe_set_size().
Link: http://lkml.kernel.org/r/91a91fdb-a959-ba7f-b551-b62477cc98a1@gmail.com
Signed-off-by: Michael Kerrisk <mtk.manpages@gmail.com>
Reviewed-by: Vegard Nossum <vegard.nossum@oracle.com>
Cc: Willy Tarreau <w@1wt.eu>
Cc: <socketpair@gmail.com>
Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Cc: Jens Axboe <axboe@fb.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull more vfs updates from Al Viro:
">rename2() work from Miklos + current_time() from Deepa"
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
fs: Replace current_fs_time() with current_time()
fs: Replace CURRENT_TIME_SEC with current_time() for inode timestamps
fs: Replace CURRENT_TIME with current_time() for inode timestamps
fs: proc: Delete inode time initializations in proc_alloc_inode()
vfs: Add current_time() api
vfs: add note about i_op->rename changes to porting
fs: rename "rename2" i_op to "rename"
vfs: remove unused i_op->rename
fs: make remaining filesystems use .rename2
libfs: support RENAME_NOREPLACE in simple_rename()
fs: support RENAME_NOREPLACE for local filesystems
ncpfs: fix unused variable warning
CURRENT_TIME macro is not appropriate for filesystems as it
doesn't use the right granularity for filesystem timestamps.
Use current_time() instead.
CURRENT_TIME is also not y2038 safe.
This is also in preparation for the patch that transitions
vfs timestamps to use 64 bit time and hence make them
y2038 safe. As part of the effort current_time() will be
extended to do range checks. Hence, it is necessary for all
file system timestamps to use current_time(). Also,
current_time() will be transitioned along with vfs to be
y2038 safe.
Note that whenever a single call to current_time() is used
to change timestamps in different inodes, it is because they
share the same time granularity.
Signed-off-by: Deepa Dinamani <deepa.kernel@gmail.com>
Reviewed-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Felipe Balbi <balbi@kernel.org>
Acked-by: Steven Whitehouse <swhiteho@redhat.com>
Acked-by: Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp>
Acked-by: David Sterba <dsterba@suse.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
To distinguish non-slab pages charged to kmemcg we mark them PageKmemcg,
which sets page->_mapcount to -512. Currently, we set/clear PageKmemcg
in __alloc_pages_nodemask()/free_pages_prepare() for any page allocated
with __GFP_ACCOUNT, including those that aren't actually charged to any
cgroup, i.e. allocated from the root cgroup context. To avoid overhead
in case cgroups are not used, we only do that if memcg_kmem_enabled() is
true. The latter is set iff there are kmem-enabled memory cgroups
(online or offline). The root cgroup is not considered kmem-enabled.
As a result, if a page is allocated with __GFP_ACCOUNT for the root
cgroup when there are kmem-enabled memory cgroups and is freed after all
kmem-enabled memory cgroups were removed, e.g.
# no memory cgroups has been created yet, create one
mkdir /sys/fs/cgroup/memory/test
# run something allocating pages with __GFP_ACCOUNT, e.g.
# a program using pipe
dmesg | tail
# remove the memory cgroup
rmdir /sys/fs/cgroup/memory/test
we'll get bad page state bug complaining about page->_mapcount != -1:
BUG: Bad page state in process swapper/0 pfn:1fd945c
page:ffffea007f651700 count:0 mapcount:-511 mapping: (null) index:0x0
flags: 0x1000000000000000()
To avoid that, let's mark with PageKmemcg only those pages that are
actually charged to and hence pin a non-root memory cgroup.
Fixes: 4949148ad4 ("mm: charge/uncharge kmemcg from generic page allocator paths")
Reported-and-tested-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: Vladimir Davydov <vdavydov@virtuozzo.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pipes can consume a significant amount of system memory, hence they
should be accounted to kmemcg.
This patch marks pipe_inode_info and anonymous pipe buffer page
allocations as __GFP_ACCOUNT so that they would be charged to kmemcg.
Note, since a pipe buffer page can be "stolen" and get reused for other
purposes, including mapping to userspace, we clear PageKmemcg thus
resetting page->_mapcount and uncharge it in anon_pipe_buf_steal, which
is introduced by this patch.
A note regarding anon_pipe_buf_steal implementation. We allow to steal
the page if its ref count equals 1. It looks racy, but it is correct
for anonymous pipe buffer pages, because:
- We lock out all other pipe users, because ->steal is called with
pipe_lock held, so the page can't be spliced to another pipe from
under us.
- The page is not on LRU and it never was.
- Thus a parallel thread can access it only by PFN. Although this is
quite possible (e.g. see page_idle_get_page and balloon_page_isolate)
this is not dangerous, because all such functions do is increase page
ref count, check if the page is the one they are looking for, and
decrease ref count if it isn't. Since our page is clean except for
PageKmemcg mark, which doesn't conflict with other _mapcount users,
the worst that can happen is we see page_count > 2 due to a transient
ref, in which case we false-positively abort ->steal, which is still
fine, because ->steal is not guaranteed to succeed.
Link: http://lkml.kernel.org/r/20160527150313.GD26059@esperanza
Signed-off-by: Vladimir Davydov <vdavydov@virtuozzo.com>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Minchan Kim <minchan@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} macros were introduced *long* time
ago with promise that one day it will be possible to implement page
cache with bigger chunks than PAGE_SIZE.
This promise never materialized. And unlikely will.
We have many places where PAGE_CACHE_SIZE assumed to be equal to
PAGE_SIZE. And it's constant source of confusion on whether
PAGE_CACHE_* or PAGE_* constant should be used in a particular case,
especially on the border between fs and mm.
Global switching to PAGE_CACHE_SIZE != PAGE_SIZE would cause to much
breakage to be doable.
Let's stop pretending that pages in page cache are special. They are
not.
The changes are pretty straight-forward:
- <foo> << (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>;
- <foo> >> (PAGE_CACHE_SHIFT - PAGE_SHIFT) -> <foo>;
- PAGE_CACHE_{SIZE,SHIFT,MASK,ALIGN} -> PAGE_{SIZE,SHIFT,MASK,ALIGN};
- page_cache_get() -> get_page();
- page_cache_release() -> put_page();
This patch contains automated changes generated with coccinelle using
script below. For some reason, coccinelle doesn't patch header files.
I've called spatch for them manually.
The only adjustment after coccinelle is revert of changes to
PAGE_CAHCE_ALIGN definition: we are going to drop it later.
There are few places in the code where coccinelle didn't reach. I'll
fix them manually in a separate patch. Comments and documentation also
will be addressed with the separate patch.
virtual patch
@@
expression E;
@@
- E << (PAGE_CACHE_SHIFT - PAGE_SHIFT)
+ E
@@
expression E;
@@
- E >> (PAGE_CACHE_SHIFT - PAGE_SHIFT)
+ E
@@
@@
- PAGE_CACHE_SHIFT
+ PAGE_SHIFT
@@
@@
- PAGE_CACHE_SIZE
+ PAGE_SIZE
@@
@@
- PAGE_CACHE_MASK
+ PAGE_MASK
@@
expression E;
@@
- PAGE_CACHE_ALIGN(E)
+ PAGE_ALIGN(E)
@@
expression E;
@@
- page_cache_get(E)
+ get_page(E)
@@
expression E;
@@
- page_cache_release(E)
+ put_page(E)
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
On no-so-small systems, it is possible for a single process to cause an
OOM condition by filling large pipes with data that are never read. A
typical process filling 4000 pipes with 1 MB of data will use 4 GB of
memory. On small systems it may be tricky to set the pipe max size to
prevent this from happening.
This patch makes it possible to enforce a per-user soft limit above
which new pipes will be limited to a single page, effectively limiting
them to 4 kB each, as well as a hard limit above which no new pipes may
be created for this user. This has the effect of protecting the system
against memory abuse without hurting other users, and still allowing
pipes to work correctly though with less data at once.
The limit are controlled by two new sysctls : pipe-user-pages-soft, and
pipe-user-pages-hard. Both may be disabled by setting them to zero. The
default soft limit allows the default number of FDs per process (1024)
to create pipes of the default size (64kB), thus reaching a limit of 64MB
before starting to create only smaller pipes. With 256 processes limited
to 1024 FDs each, this results in 1024*64kB + (256*1024 - 1024) * 4kB =
1084 MB of memory allocated for a user. The hard limit is disabled by
default to avoid breaking existing applications that make intensive use
of pipes (eg: for splicing).
Reported-by: socketpair@gmail.com
Reported-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Mitigates: CVE-2013-4312 (Linux 2.0+)
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Willy Tarreau <w@1wt.eu>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
pipe_write() would return 0 if it failed to merge the beginning of the
data to write with the last, partially filled pipe buffer. It should
return an error code instead. Userspace programs could be confused by
write() returning 0 when called with a nonzero 'count'.
The EFAULT error case was a regression from f0d1bec9d5 ("new helper:
copy_page_from_iter()"), while the ops->confirm() error case was a much
older bug.
Test program:
#include <assert.h>
#include <errno.h>
#include <unistd.h>
int main(void)
{
int fd[2];
char data[1] = {0};
assert(0 == pipe(fd));
assert(1 == write(fd[1], data, 1));
/* prior to this patch, write() returned 0 here */
assert(-1 == write(fd[1], NULL, 1));
assert(errno == EFAULT);
}
Cc: stable@vger.kernel.org # at least v3.15+
Signed-off-by: Eric Biggers <ebiggers3@gmail.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
If sys_pipe() was unable to allocate a 'struct file', it always failed
with ENFILE, which means "The number of simultaneously open files in the
system would exceed a system-imposed limit." However, alloc_file()
actually returns an ERR_PTR value and might fail with other error codes.
Currently, in addition to ENFILE, it can fail with ENOMEM, potentially
when there are few open files in the system. Update sys_pipe() to
preserve this error code.
In a prior submission of a similar patch (1) some concern was raised
about introducing a new error code for sys_pipe(). However, for most
system calls, programs cannot assume that new error codes will never be
introduced. In addition, ENOMEM was, in fact, already a possible error
code for sys_pipe(), in the case where the file descriptor table could
not be expanded due to insufficient memory.
(1) http://comments.gmane.org/gmane.linux.kernel/1357942
Signed-off-by: Eric Biggers <ebiggers3@gmail.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
All places outside of core VFS that checked ->read and ->write for being NULL or
called the methods directly are gone now, so NULL {read,write} with non-NULL
{read,write}_iter will do the right thing in all cases.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
struct kiocb now is a generic I/O container, so move it to fs.h.
Also do a #include diet for aio.h while we're at it.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
For now, just use the same thing we pass to ->direct_IO() - it's all
iovec-based at the moment. Pass it explicitly to iov_iter_init() and
account for kvec vs. iovec in there, by the same kludge NFS ->direct_IO()
uses.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Pipe has no data associated with fs so it is not good idea to block
pipe_write() if FS is frozen, but we can not update file's time on such
filesystem. Let's use same idea as we use in touch_time().
Addresses https://bugzilla.kernel.org/show_bug.cgi?id=65701
Signed-off-by: Dmitry Monakhov <dmonakhov@openvz.org>
Reviewed-by: Jan Kara <jack@suse.cz>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The pipe code was trying (and failing) to be very careful about freeing
the pipe info only after the last access, with a pattern like:
spin_lock(&inode->i_lock);
if (!--pipe->files) {
inode->i_pipe = NULL;
kill = 1;
}
spin_unlock(&inode->i_lock);
__pipe_unlock(pipe);
if (kill)
free_pipe_info(pipe);
where the final freeing is done last.
HOWEVER. The above is actually broken, because while the freeing is
done at the end, if we have two racing processes releasing the pipe
inode info, the one that *doesn't* free it will decrement the ->files
count, and unlock the inode i_lock, but then still use the
"pipe_inode_info" afterwards when it does the "__pipe_unlock(pipe)".
This is *very* hard to trigger in practice, since the race window is
very small, and adding debug options seems to just hide it by slowing
things down.
Simon originally reported this way back in July as an Oops in
kmem_cache_allocate due to a single bit corruption (due to the final
"spin_unlock(pipe->mutex.wait_lock)" incrementing a field in a different
allocation that had re-used the free'd pipe-info), it's taken this long
to figure out.
Since the 'pipe->files' accesses aren't even protected by the pipe lock
(we very much use the inode lock for that), the simple solution is to
just drop the pipe lock early. And since there were two users of this
pattern, create a helper function for it.
Introduced commit ba5bb14733 ("pipe: take allocation and freeing of
pipe_inode_info out of ->i_mutex").
Reported-by: Simon Kirby <sim@hostway.ca>
Reported-by: Ian Applegate <ia@cloudflare.com>
Acked-by: Al Viro <viro@zeniv.linux.org.uk>
Cc: stable@kernel.org # v3.10+
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
it's used only as a flag to distinguish normal pipes/FIFOs from the
internal per-task one used by file-to-file splice. And pipe->files
would work just as well for that purpose...
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
fs/pipe.c file_operations methods *know* that pipe is not an internal one;
no need to check pipe->inode for those callers.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
* new field - pipe->files; number of struct file over that pipe (all
sharing the same inode, of course); protected by inode->i_lock.
* pipe_release() decrements pipe->files, clears inode->i_pipe when
if the counter has reached 0 (all under ->i_lock) and, in that case,
frees pipe after having done pipe_unlock()
* fifo_open() starts with grabbing ->i_lock, and either bumps pipe->files
if ->i_pipe was non-NULL or allocates a new pipe (dropping and regaining
->i_lock) and rechecks ->i_pipe; if it's still NULL, inserts new pipe
there, otherwise bumps ->i_pipe->files and frees the one we'd allocated.
At that point we know that ->i_pipe is non-NULL and won't go away, so
we can do pipe_lock() on it and proceed as we used to. If we end up
failing, decrement pipe->files and if it reaches 0 clear ->i_pipe and
free the sucker after pipe_unlock().
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
* use the fact that file_inode(file)->i_pipe doesn't change
while the file is opened - no locks needed to access that.
* switch to pipe_lock/pipe_unlock where it's easy to do
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
If you open a pipe for neither read nor write, the pipe code will not
add any usage counters to the pipe, causing the 'struct pipe_inode_info"
to be potentially released early.
That doesn't normally matter, since you cannot actually use the pipe,
but the pipe release code - particularly fasync handling - still expects
the actual pipe infrastructure to all be there. And rather than adding
NULL pointer checks, let's just disallow this case, the same way we
already do for the named pipe ("fifo") case.
This is ancient going back to pre-2.4 days, and until trinity, nobody
naver noticed.
Reported-by: Dave Jones <davej@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Allocating a file structure in function get_empty_filp() might fail because
of several reasons:
- not enough memory for file structures
- operation is not allowed
- user is over its limit
Currently the function returns NULL in all cases and we loose the exact
reason of the error. All callers of get_empty_filp() assume that the function
can fail with ENFILE only.
Return error through pointer. Change all callers to preserve this error code.
[AV: cleaned up a bit, carved the get_empty_filp() part out into a separate commit
(things remaining here deal with alloc_file()), removed pipe(2) behaviour change]
Signed-off-by: Anatol Pomozov <anatol.pomozov@gmail.com>
Reviewed-by: "Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
don't mess with sys_close() if copy_to_user() fails; just postpone
fd_install() until we know it hasn't.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Pull second vfs pile from Al Viro:
"The stuff in there: fsfreeze deadlock fixes by Jan (essentially, the
deadlock reproduced by xfstests 068), symlink and hardlink restriction
patches, plus assorted cleanups and fixes.
Note that another fsfreeze deadlock (emergency thaw one) is *not*
dealt with - the series by Fernando conflicts a lot with Jan's, breaks
userland ABI (FIFREEZE semantics gets changed) and trades the deadlock
for massive vfsmount leak; this is going to be handled next cycle.
There probably will be another pull request, but that stuff won't be
in it."
Fix up trivial conflicts due to unrelated changes next to each other in
drivers/{staging/gdm72xx/usb_boot.c, usb/gadget/storage_common.c}
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs: (54 commits)
delousing target_core_file a bit
Documentation: Correct s_umount state for freeze_fs/unfreeze_fs
fs: Remove old freezing mechanism
ext2: Implement freezing
btrfs: Convert to new freezing mechanism
nilfs2: Convert to new freezing mechanism
ntfs: Convert to new freezing mechanism
fuse: Convert to new freezing mechanism
gfs2: Convert to new freezing mechanism
ocfs2: Convert to new freezing mechanism
xfs: Convert to new freezing code
ext4: Convert to new freezing mechanism
fs: Protect write paths by sb_start_write - sb_end_write
fs: Skip atime update on frozen filesystem
fs: Add freezing handling to mnt_want_write() / mnt_drop_write()
fs: Improve filesystem freezing handling
switch the protection of percpu_counter list to spinlock
nfsd: Push mnt_want_write() outside of i_mutex
btrfs: Push mnt_want_write() outside of i_mutex
fat: Push mnt_want_write() outside of i_mutex
...
Btrfs has to make sure we have space to allocate new blocks in order to modify
the inode, so updating time can fail. We've gotten around this by having our
own file_update_time but this is kind of a pain, and Christoph has indicated he
would like to make xfs do something different with atime updates. So introduce
->update_time, where we will deal with i_version an a/m/c time updates and
indicate which changes need to be made. The normal version just does what it
has always done, updates the time and marks the inode dirty, and then
filesystems can choose to do something different.
I've gone through all of the users of file_update_time and made them check for
errors with the exception of the fault code since it's complicated and I wasn't
quite sure what to do there, also Jan is going to be pushing the file time
updates into page_mkwrite for those who have it so that should satisfy btrfs and
make it not a big deal to check the file_update_time() return code in the
generic fault path. Thanks,
Signed-off-by: Josef Bacik <josef@redhat.com>
As described in commit 07d106d0a ("vfs: fix up ENOIOCTLCMD error
handling"), drivers should return -ENOIOCTLCMD if they receive an ioctl
command which they don't understand. Doing so will result in -ENOTTY
being returned to userspace, which matches the behaviour of the compat
layer if it fails to translate an ioctl command.
This patch fixes the pipe ioctl to return -ENOIOCTLCMD instead of
-EINVAL when passed an unknown ioctl command.
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
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>
Cc: stable@kernel.org # needed for systemd/autofs interaction fix
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Michael Kerrisk (man-pages)