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>
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>
