Merge page ref overflow branch.
Jann Horn reported that he can overflow the page ref count with
sufficient memory (and a filesystem that is intentionally extremely
slow).
Admittedly it's not exactly easy. To have more than four billion
references to a page requires a minimum of 32GB of kernel memory just
for the pointers to the pages, much less any metadata to keep track of
those pointers. Jann needed a total of 140GB of memory and a specially
crafted filesystem that leaves all reads pending (in order to not ever
free the page references and just keep adding more).
Still, we have a fairly straightforward way to limit the two obvious
user-controllable sources of page references: direct-IO like page
references gotten through get_user_pages(), and the splice pipe page
duplication. So let's just do that.
* branch page-refs:
fs: prevent page refcount overflow in pipe_buf_get
mm: prevent get_user_pages() from overflowing page refcount
mm: add 'try_get_page()' helper function
mm: make page ref count overflow check tighter and more explicit
Change pipe_buf_get() to return a bool indicating whether it succeeded
in raising the refcount of the page (if the thing in the pipe is a page).
This removes another mechanism for overflowing the page refcount. All
callers converted to handle a failure.
Reported-by: Jann Horn <jannh@google.com>
Signed-off-by: Matthew Wilcox <willy@infradead.org>
Cc: stable@kernel.org
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The only caller that needs fc->aborted set is fuse_conn_abort_write().
Setting fc->aborted is now racy (fuse_abort_conn() may already be in
progress or finished) but there's no reason to care.
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
This is rather natural action after previous patches, and it just decreases
load of fc->lock.
Signed-off-by: Kirill Tkhai <ktkhai@virtuozzo.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
When queue_interrupt() is called from fuse_dev_do_write(), it came from
userspace directly. Userspace may pass any request id, even the request's
we have not interrupted (or even background's request). This patch adds
sanity check to make kernel safe against that.
Signed-off-by: Kirill Tkhai <ktkhai@virtuozzo.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Currently, we wait on req->waitq in request_wait_answer() function only,
and it's never used for background requests. Since wake_up() is not a
light-weight macros, instead of this, it unfolds in really called function,
which makes locking operations taking some cpu cycles, let's avoid its call
for the case we definitely know it's completely useless.
Signed-off-by: Kirill Tkhai <ktkhai@virtuozzo.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
We take global fiq->waitq.lock every time, when we are in this function,
but interrupted requests are just small subset of all requests. This patch
optimizes request_end() and makes it to take the lock when it's really
needed.
queue_interrupt() needs small change for that. After req is linked to
interrupt list, we do smp_mb() and check for FR_FINISHED again. In case of
FR_FINISHED bit has appeared, we remove req and leave the function:
Signed-off-by: Kirill Tkhai <ktkhai@virtuozzo.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
We should sent signal only in case of interrupt is really queued. Not a
real problem, but this makes the code clearer and intuitive.
Signed-off-by: Kirill Tkhai <ktkhai@virtuozzo.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Some of the pipe_buf_release() handlers seem to assume that the pipe is
locked - in particular, anon_pipe_buf_release() accesses pipe->tmp_page
without taking any extra locks. From a glance through the callers of
pipe_buf_release(), it looks like FUSE is the only one that calls
pipe_buf_release() without having the pipe locked.
This bug should only lead to a memory leak, nothing terrible.
Fixes: dd3bb14f44 ("fuse: support splice() writing to fuse device")
Cc: stable@vger.kernel.org
Signed-off-by: Jann Horn <jannh@google.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
In current fuse_drop_waiting() implementation it's possible that
fuse_wait_aborted() will not be woken up in the unlikely case that
fuse_abort_conn() + fuse_wait_aborted() runs in between checking
fc->connected and calling atomic_dec(&fc->num_waiting).
Do the atomic_dec_and_test() unconditionally, which also provides the
necessary barrier against reordering with the fc->connected check.
The explicit smp_mb() in fuse_wait_aborted() is not actually needed, since
the spin_unlock() in fuse_abort_conn() provides the necessary RELEASE
barrier after resetting fc->connected. However, this is not a performance
sensitive path, and adding the explicit barrier makes it easier to
document.
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Fixes: b8f95e5d13 ("fuse: umount should wait for all requests")
Cc: <stable@vger.kernel.org> #v4.19
fuse_request_send_notify_reply() may fail if the connection was reset for
some reason (e.g. fs was unmounted). Don't leak request reference in this
case. Besides leaking memory, this resulted in fc->num_waiting not being
decremented and hence fuse_wait_aborted() left in a hanging and unkillable
state.
Fixes: 2d45ba381a ("fuse: add retrieve request")
Fixes: b8f95e5d13 ("fuse: umount should wait for all requests")
Reported-and-tested-by: syzbot+6339eda9cb4ebbc4c37b@syzkaller.appspotmail.com
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Cc: <stable@vger.kernel.org> #v2.6.36
Writeback caching currently allocates requests with the maximum number of
possible pages, while the actual number of pages per request depends on a
couple of factors that cannot be determined when the request is allocated
(whether page is already under writeback, whether page is contiguous with
previous pages already added to a request).
This patch allows such requests to start with no page allocation (all pages
inline) and grow the page array on demand.
If the max_pages tunable remains the default value, then this will mean
just one allocation that is the same size as before. If the tunable is
larger, then this adds at most 3 additional memory allocations (which is
generously compensated by the improved performance from the larger
request).
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Replace FUSE_MAX_PAGES_PER_REQ with the configurable parameter max_pages to
improve performance.
Old RFC with detailed description of the problem and many fixes by Mitsuo
Hayasaka (mitsuo.hayasaka.hu@hitachi.com):
- https://lkml.org/lkml/2012/7/5/136
We've encountered performance degradation and fixed it on a big and complex
virtual environment.
Environment to reproduce degradation and improvement:
1. Add lag to user mode FUSE
Add nanosleep(&(struct timespec){ 0, 1000 }, NULL); to xmp_write_buf in
passthrough_fh.c
2. patch UM fuse with configurable max_pages parameter. The patch will be
provided latter.
3. run test script and perform test on tmpfs
fuse_test()
{
cd /tmp
mkdir -p fusemnt
passthrough_fh -o max_pages=$1 /tmp/fusemnt
grep fuse /proc/self/mounts
dd conv=fdatasync oflag=dsync if=/dev/zero of=fusemnt/tmp/tmp \
count=1K bs=1M 2>&1 | grep -v records
rm fusemnt/tmp/tmp
killall passthrough_fh
}
Test results:
passthrough_fh /tmp/fusemnt fuse.passthrough_fh \
rw,nosuid,nodev,relatime,user_id=0,group_id=0 0 0
1073741824 bytes (1.1 GB) copied, 1.73867 s, 618 MB/s
passthrough_fh /tmp/fusemnt fuse.passthrough_fh \
rw,nosuid,nodev,relatime,user_id=0,group_id=0,max_pages=256 0 0
1073741824 bytes (1.1 GB) copied, 1.15643 s, 928 MB/s
Obviously with bigger lag the difference between 'before' and 'after'
will be more significant.
Mitsuo Hayasaka, in 2012 (https://lkml.org/lkml/2012/7/5/136),
observed improvement from 400-550 to 520-740.
Signed-off-by: Constantine Shulyupin <const@MakeLinux.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
When allocating page array for a request the array for the page pointers
and the array for page descriptors are allocated by two separate kmalloc()
calls. Merge these into one allocation.
Also instead of initializing the request and the page arrays with memset(),
use the zeroing allocation variants.
Reserved requests never carry pages (page array size is zero). Make that
explicit by initializing the page array pointers to NULL and make sure the
assumption remains true by adding a WARN_ON().
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
We noticed the performance bottleneck in FUSE running our Virtuozzo storage
over rdma. On some types of workload we observe 20% of times spent in
request_find() in profiler. This function is iterating over long requests
list, and it scales bad.
The patch introduces hash table to reduce the number of iterations, we do
in this function. Hash generating algorithm is taken from hash_add()
function, while 256 lines table is used to store pending requests. This
fixes problem and improves the performance.
Reported-by: Alexey Kuznetsov <kuznet@virtuozzo.com>
Signed-off-by: Kirill Tkhai <ktkhai@virtuozzo.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
This field is not needed after the previous patch, since we can easily
convert request ID to interrupt request ID and vice versa.
Signed-off-by: Kirill Tkhai <ktkhai@virtuozzo.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Using of two unconnected IDs req->in.h.unique and req->intr_unique does not
allow to link requests to a hash table. We need can't use none of them as a
key to calculate hash.
This patch changes the algorithm of allocation of IDs for a request. Plain
requests obtain even ID, while interrupt requests are encoded in the low
bit. So, in next patches we will be able to use the rest of ID bits to
calculate hash, and the hash will be the same for plain and interrupt
requests.
Signed-off-by: Kirill Tkhai <ktkhai@virtuozzo.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Currently, we take fc->lock there only to check for fc->connected.
But this flag is changed only on connection abort, which is very
rare operation.
So allow checking fc->connected under just fc->bg_lock and use this lock
(as well as fc->lock) when resetting fc->connected.
Signed-off-by: Kirill Tkhai <ktkhai@virtuozzo.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
To reduce contention of fc->lock, this patch introduces bg_lock for
protection of fields related to background queue. These are:
max_background, congestion_threshold, num_background, active_background,
bg_queue and blocked.
This allows next patch to make async reads not requiring fc->lock, so async
reads and writes will have better performance executed in parallel.
Signed-off-by: Kirill Tkhai <ktkhai@virtuozzo.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
This cleanup patch makes the function to use the primitive
instead of direct dereferencing.
Also, move fiq dereferencing out of cycle, since it's
always constant.
Signed-off-by: Kirill Tkhai <ktkhai@virtuozzo.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Using waitqueue_active() is racy. Make sure we issue a wake_up()
unconditionally after storing into fc->blocked. After that it's okay to
optimize with waitqueue_active() since the first wake up provides the
necessary barrier for all waiters, not the just the woken one.
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Fixes: 3c18ef8117 ("fuse: optimize wake_up")
Cc: <stable@vger.kernel.org> # v3.10
