| Age | Commit message (Collapse) | Author |
|---|
|
The existing condition in fuse_uring_cmd() is there only to avoid
disabling io-uring for connections that already run with it, missing
was a condition to refuse any IORING_OP_URING_CMD if the
connection/channel didn't get enabled because of missing FUSE_INIT
reply flag FUSE_OVER_IO_URING. Without the reply flag the barrier in
fuse_uring_ready() doesn't work and IO could already be going on and
cause deadlock states (at a minimum one between fch->bg_lock and
queue->lock).
The change itself is trivial, but brings behavior change,
FUSE_OVER_IO_URING has to be set in the FUSE_INIT_REPLY by fuse servers
to accept any IORING_OP_URING_CMD. Libfuse does that and the only
non-libfuse implementation I found (fractal-fuse) also does it.
Qemu patches for fuse-io-uring are not merged yet, as far as I know.
Moved up is the smp_load_acquire(&fch->initialized) check, as a
fuse-server implementation might try to setup io-uring before FUSE_INIT
is processed and might have gotten -EOPNOTSUPP instead of -EAGAIN.
Also fixed is a stale comment that explains the handling of the
FUSE_OVER_IO_URING flag in early RFC versions.
If there should be a report from any library or application we
probably need to revert this commit.
Fixes: 3393ff964e0f ("fuse: block request allocation until io-uring init is complete")
Signed-off-by: Bernd Schubert <bernd@bsbernd.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
The fuse-io-uring transport copies req->in.h out to the ring in
fuse_uring_copy_to_ring() and req->out.h back in fuse_uring_commit().
Both headers live inside the fuse_request slab object, whose cache
(fuse_req_cachep) is created without a usercopy whitelist, so copying
them directly to/from userspace trips CONFIG_HARDENED_USERCOPY and
panics:
usercopy: Kernel memory exposure attempt detected from SLUB object
'fuse_request' (offset 56, size 40)!
kernel BUG at mm/usercopy.c:102!
Oops: invalid opcode: 0000 [#1] SMP KASAN NOPTI
RIP: 0010:usercopy_abort (mm/usercopy.c:90)
Call Trace:
__check_heap_object (mm/slub.c:8268)
__check_object_size (mm/usercopy.c:197 mm/usercopy.c:258 mm/usercopy.c:223)
copy_header_to_ring (fs/fuse/dev_uring.c:618)
fuse_uring_prepare_send (fs/fuse/dev_uring.c:776 fs/fuse/dev_uring.c:785)
fuse_uring_send_in_task (fs/fuse/dev_uring.c:1306)
tctx_task_work_run (io_uring/tw.c:96)
task_work_run (kernel/task_work.c:233)
io_run_task_work (io_uring/tw.h:84)
io_cqring_wait (io_uring/wait.c:278)
__do_sys_io_uring_enter (io_uring/io_uring.c:2685)
entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:121)
Bounce both headers through an on-stack copy so the usercopy touches
stack memory, not the slab object.
Fixes: c090c8abae4b ("fuse: Add io-uring sqe commit and fetch support")
Cc: stable@vger.kernel.org
Reported-by: Weiming Shi <bestswngs@gmail.com>
Assisted-by: Claude:claude-opus-4-8
Signed-off-by: Xiang Mei <xmei5@asu.edu>
Reviewed-by: Bernd Schubert <bernd@bsbernd.com>
Reviewed-by: Joanne Koong <joannelkoong@gmail.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
Implement zero-copy in fuse io-uring to eliminate memory copies between
the application, kernel, and server for read/write operations. The
server can directly access client pages or page cache folios without
copying data through an intermediary buffer. When a fuse request arrives,
the kernel registers the relevant pages into a sparse slot in the
server's io_uring registered buffer table. The server can then operate
on these pages directly using io-uring fixed buffer operations (eg
read_fixed/write_fixed) and the kernel unregisters these pages when the
request completes. Non-page-backed args (eg op out headers) will go
through the payload buffer as normal. The server can specify which open
files should have their reads/writes go through zero-copy, by setting
the FOPEN_IO_URING_ZERO_COPY flag when servicing opens.
This requires CAP_SYS_ADMIN and bufpools. This is gated behind
CAP_SYS_ADMIN because zero-copy allows the server direct access to the
client's underlying pages, rather than operating on an intermediary
buffer that the contents of the client's pages were copied into or on
page cache folios.
The request flow for the zero-copy direct-io write path (client writes
data, server reads it) is as follows:
=======================================================================
| Kernel | FUSE server
| |
| "write(fd, buf, 1MB)" |
| |
| >sys_write() |
| >fuse_file_write_iter() |
| >fuse_send_one() |
| [req->args->in_pages = true] |
| [folios hold client write data] |
| |
| >fuse_uring_copy_to_ring() |
| >copy_header_to_ring(IN_OUT) |
| [memcpy fuse_in_header] |
| >copy_header_to_ring(OP) |
| [memcpy write_in header] |
| |
| >fuse_uring_args_to_ring() |
| >setup_fuse_copy_state() |
| [skip_folio_copy = true] |
| |
| >fuse_uring_set_up_zero_copy() |
| [folio_get for each client folio] |
| [build bio_vec array from folios] |
| >io_buffer_register_bvec() |
| [register pages at
ent->zero_copy_index] |
| [ent->zero_copied = true] |
| |
| >fuse_copy_args() |
| [skip_folio_copy => return 0 |
| for page arg, skip data copy] |
| |
| >copy_header_to_ring(RING_ENT) |
| [memcpy ent_in_out] |
| >io_uring_cmd_done() |
| |
| | [CQE received]
| |
| | [issue io_uring READ at
| | ent->zero_copy_index]
| | [reads directly from
| |client's pages (ZERO_COPY)]
| |
| | [write data to backing
| | store]
| | [submit COMMIT AND FETCH]
| |
| >fuse_uring_commit_fetch() |
| >fuse_uring_commit() |
| >fuse_uring_copy_from_ring() |
| >fuse_uring_req_end() |
| >io_buffer_unregister(ent->zero_copy_index) |
| [unregister pages from index] |
| >fuse_zero_copy_release() |
| [folio_put for each folio] |
| [ent->zero_copied = false] |
| >fuse_request_end() |
| [wake up client] |
The zero-copy read path is analogous.
Some requests may have both page-backed args and non-page-backed args.
For these requests, the page-backed args are zero-copied while the
non-page-backed args are copied to the buffer selected from the buffer
pool:
zero-copy: pages registered via io_buffer_register_bvec()
non-page-backed: copied to payload buffer via fuse_copy_args()
For a request whose payload is zero-copied, the
registration/unregistration path looks like:
register: fuse_uring_set_up_zero_copy()
folio_get() for each folio
io_buffer_register_bvec(ent->zero_copy_index)
unregister: fuse_uring_req_end()
io_buffer_unregister(ent->zero_copy_index)
-> fuse_zero_copy_release() callback
folio_put() for each folio
Please note that on abort for in-flight zero-copied requests that have
been sent to userspace, the registered bvec slot remains occupied and
its folios remain pinned until the io-uring ring is destroyed, at which
point io-uring unregisters all buffers and the fuse_zero_copy_release()
callback drops the folio references. Unregistering at teardown would
require operating on the ring context directly, whose validity is hard
to ascertain; this is deemed not worth the complexity for the abort
race, since everything is freed when the ring is torn down.
The throughput improvement from zero-copy depends on how much of the
per-request latency is spent on data copying vs backing I/O. The gain
comes from eliminating the payload-buffer memcpy, but accessing the
zero-copied pages requires the server to issue the read/write as an
IORING_OP_READ/WRITE_FIXED operation. The benefit is largest when the
mempcy is a meaningful fraction of per-request latency while backing i/o
is still noticable enough that the extra io-uring op's overhead doesn't
dominate.
Benchmarked with passthrough_hp (--nopassthrough, q_depth=8) on a
2-socket Intel Xeon Gold 6138 (40 cores / 80 threads), using fio (sync
engine, bs=1M, O_DIRECT, numjobs=2, 30s run + 10s ramp, 3 runs) where
direct-I/O throughput is against a RAM-backed (tmpfs) source (backing
I/O is not the bottleneck):
baseline registered-buf zero-copy (zc vs base)
direct read ~5.1 GB/s ~5.4 GB/s ~8.9 GB/s (+75%)
direct write ~3.4 GB/s ~4.8 GB/s ~5.1 GB/s (+50%)
Reads end up higher than writes because the backing store reads faster
than it writes (the baseline shows the same read>write gap, and the raw
device does too). On a device-bound NVMe (~2 GB/s reads) the read gain
shrinks to ~10-16% (and no measurable gains for writes), as backing I/O
rather than the eliminated copy dominates latency.
The benefit overall scales with how much of the
per-request latency is the data copy versus backing I/O.
Signed-off-by: Joanne Koong <joannelkoong@gmail.com>
Reviewed-by: Bernd Schubert <bernd@bsbernd.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
Allow servers to use a buffer pool that is also registered through
io-uring. When the server registers a buffer pool with io-uring, the
pages backing the pool are pinned upfront. This eliminates the overhead
of pinning/unpinning user pages and translating virtual addresses per
i/o request. This also allows servers to use the same registered memory
for subsequent backing store I/O (eg read_fixed/write_fixed), keeping
data in the same pinned pages without additional pinning or mapping
overhead required.
To use this, the server needs to set the FUSE_URING_REGISTERED_BUFPOOL
flag when adding a bufpool through the FUSE_IO_URING_CMD_ADD_BUFPOOL
cmd. For every sqe submitted (including the one for adding the bufpool),
it should set sqe->uring_cmd_flags to include IORING_URING_CMD_FIXED,
and pass in the index where the registered bufpool resides to
sqe->buf_index.
Benchmarked with passthrough_hp (--nopassthrough, q_depth=8) on a
2-socket Intel Xeon Gold 6138 (40 cores / 80 threads), using fio (sync
engine, bs=1M, O_DIRECT, numjobs=2, 30s run + 10s ramp, 3 runs) where
direct-I/O throughput is against a RAM-backed (tmpfs) source (backing
I/O is not the bottleneck):
baseline registered buffers
direct read ~5.1 GB/s ~5.4 GB/s (+~5%)
direct write ~3.4 GB/s ~4.8 GB/s (+~45%)
Registered buffers bring up the write path speed up closer to speed of
reads. There isn't much improvement for reads because it is already fast
enough where it's at the copy-bound ceiling (surpassing that requires
doing zero-copy). On a device-bound NVMe though, the differences are
within noise, as backing I/O dominates per-request latency.
Signed-off-by: Joanne Koong <joannelkoong@gmail.com>
Reviewed-by: Bernd Schubert <bernd@bsbernd.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
Right now, ents and buffers are tightly coupled in fuse io-uring where
each entry has its own dedicated payload buffer, requiring N buffers for
N entries where each buffer must be large enough to accomodate the
maximum payload size. This is suboptimal as most request types (lookup,
open, release, getattr, etc) require vastly less bytes than the maximum
payload size and some requests (unlink, rmdir, fsync, flush, etc) do not
require payload buffers at all.
Instead of requiring a 1:1 coupling between ents and payload buffers,
allow the server to pass in a buffer pool (a contiguous chunk of memory)
that the kernel will use as it wishes for servicing ents/requests.
Entries only reserve a "buffer" from the pool while actively processing
a request that requires a payload buffer. This decoupling and letting
the kernel delegate memory from the pool for requests allows the kernel to
optimize memory usage and reduces the memory usage requirements needed
to use fuse-over-io-uring.
A pool is registered per queue with the new
FUSE_IO_URING_CMD_ADD_BUFPOOL command. The server passes the pool's base
address and length in fuse_uring_cmd_req.bufpool.{uaddr,len}.
Internally, the kernel splits the region into buffers of
ring->max_payload_sz bytes each (nr_bufs = pool len / max_payload_sz). A
queue commits to a payload mode on first use: registering an entry that
carries its own payload selects the legacy per-entry mode, while
ADD_BUFPOOL selects pool mode. The two are mutually exclusive, so
ADD_BUFPOOL must be issued before any payload-carrying entries are
registered on that queue. The queue must have been created before the
bufpool is added, through the FUSE_IO_URING_CMD_ADD_QUEUE command.
The kernel tracks free buffers with a bitmap (a set bit marks a free
buffer). On dispatch, a request that needs a payload claims a free
buffer (find_first_bit + clear). A request that needs none claims
nothing. The buffer's byte offset within the pool is reported to the
server in the new fuse_uring_ent_in_out.offset field so that the server
can locate the payload. On completion the buffer is returned to the pool
or reused directly if the next request on that entry also has a payload.
The FUSE_HAS_IO_URING_BUFPOOL flag advertises kernel support to the
server for bufpools.
Buffer pool request flow
~~~~~~~~~~~~~~~~~~~~~~~~
| Kernel | FUSE daemon
| |
| [request arrives] |
| [claim a free pool buffer] |
| >fuse_uring_select_buffer() |
| [copy headers to ring] |
| [copy payload to buffer] |
| [report buffer offset in ent_in_out] |
| >io_uring_cmd_done() |
| | [read headers]
| | [read/write payload at offset]
| | [process request]
| | >io_uring_submit()
| | COMMIT_AND_FETCH
| >fuse_uring_commit_fetch() |
| [copy reply from ring] |
| [return buffer to the pool] |
| >fuse_uring_recycle_buffer() |
Reviewed-by: Bernd Schubert <bernd@bsbernd.com>
Signed-off-by: Joanne Koong <joannelkoong@gmail.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
fuse-over-io-uring queues are currently created lazily, as a side effect
of the first FUSE_IO_URING_CMD_REGISTER command for a given qid. This
ties queue creation to entry registration.
Add a FUSE_IO_URING_CMD_ADD_QUEUE command so a server can create a queue
explicitly, decoupling queue setup from entry registration. This is
additionally a prerequisite for FUSE_IO_URING_CMD_ADD_BUFPOOL, which
attaches a buffer pool to an existing queue and therefore needs the
queue to have been created first.
Reviewed-by: Bernd Schubert <bernd@bsbernd.com>
Signed-off-by: Joanne Koong <joannelkoong@gmail.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
Currently, the connection's fuse_ring is created lazily on the first
FUSE_IO_URING_CMD_REGISTER command. A server registers entries from one
thread per queue (one per CPU) and those threads issue their first
REGISTER command concurrently. They then race to create the single
per-connection fuse_ring, which required open-coded handling in
fuse_uring_create() to detect and protect against concurrent creations.
Decouple fuse_ring creation from ent registration and move it to
FUSE_INIT reply processing after a server has negotiated and set
FUSE_OVER_IO_URING. The ring is published before the connection is
marked initialized. fuse_uring_register() no longer creates the ring and
it instead uses the ring set up at init time.
Reviewed-by: Bernd Schubert <bernd@bsbernd.com>
Signed-off-by: Joanne Koong <joannelkoong@gmail.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
fuse_uring_create_queue() initializes a fuse_ring_queue and then
publishes the pointer into ring->queues[qid] with WRITE_ONCE() under the
fch->lock. There are several readers that may concurrently be fetching
that pointer locklessly and then deferencing it.
WRITE_ONCE() doesn't ensure ordering of the queue's field
initialization before the ring->queues[qid] pointer assignment. The
queue must be published with smp_store_release() so the field
initialization is guaranteed to happen before.
Readers in paths where the read may happen concurrently with the store
need to use READ_ONCE() because any race involving a plain access is
undefined.
Fixes: 24fe962c86f5 ("fuse: {io-uring} Handle SQEs - register commands")
Cc: stable@vger.kernel.org
Reviewed-by: Bernd Schubert <bernd@bsbernd.com>
Signed-off-by: Joanne Koong <joannelkoong@gmail.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
fuse_chan_set_initialized() sets values for the connection state and
then sets fch->initialized to true, but lockless readers read
fch->initialized and if true, go to read the connection state values,
without using any barriers.
There are a few instances where this happens (fuse_uring_cmd() before
dispatching register / commit-and-fetch cmds, fuse_dev_do_wriite() for
handling notify retrieves, etc).
To make this as simple as possible, use release/acquire semantics for
writing/reading fch->initialized. Add the missing read barriers.
This is not marked for stable as these are not realistically reachable
on a well-behaved server, and buggy/malicious servers who trigger this
path fail benignly rather than crash or deadlock the kernel.
Signed-off-by: Joanne Koong <joannelkoong@gmail.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
fuse_uring_commit_fetch() error path called fuse_request_end(req) without
clearing ent->fuse_req when fuse_ring_ent_set_commit() fails. The
still-pending fuse_uring_send_in_task() task-work later dereferences the
dangling pointer through fuse_uring_prepare_send(), causing a
use-after-free.
End the request with fuse_uring_req_end(), which handles all conditions
already.
Annotation/edition by Bernd: The UAF should be fixed by other means already
and actually has to be avoided that way.
Just checking for ent->fuse_req == NULL in fuse_uring_send_in_task()
would be prone to race conditions, because if malicious userspace
would commit requests that have passed the NULL check, but are
in doing args copy, it would still trigger a use-after-free.
Setting ent->fuse_req = NULL in fuse_uring_commit_fetch() still
makes sense, though.
Reported-by: Shuvam Pandey <shuvampandey1@gmail.com>
Reported-by: Berkant Koc <me@berkoc.com>
Signed-off-by: Zhenghang Xiao <kipreyyy@gmail.com>
Signed-off-by: Bernd Schubert <bernd@bsbernd.com>
Reviewed-by: Joanne Koong <joannelkoong@gmail.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
Replace magic indices 0 and 1 for the iovec array with named constants
FUSE_URING_IOV_HEADERS and FUSE_URING_IOV_PAYLOAD. This makes the usages
self-documenting and prepares for buffer ring support which will also
reference these iovec slots by index.
Reviewed-by: Bernd Schubert <bernd@bsbernd.com>
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Reviewed-by: Baokun Li <libaokun@linux.alibaba.com>
Signed-off-by: Joanne Koong <joannelkoong@gmail.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
Add a new helper function setup_fuse_copy_state() to contain the logic
for setting up the copy state for payload copying.
Reviewed-by: Bernd Schubert <bschubert@ddn.com>
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Reviewed-by: Baokun Li <libaokun@linux.alibaba.com>
Signed-off-by: Joanne Koong <joannelkoong@gmail.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
Use enum types to identify which part of the header needs to be copied.
This improves the interface and will simplify both kernel-space and
user-space header addresses copying when buffer rings are added.
Reviewed-by: Bernd Schubert <bschubert@ddn.com>
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Reviewed-by: Baokun Li <libaokun@linux.alibaba.com>
Signed-off-by: Joanne Koong <joannelkoong@gmail.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
Move header copying from ring logic into a new copy_header_from_ring()
function. This makes the copy_from_user() logic more clear and
centralizes error handling / rate-limited logging.
Reviewed-by: Bernd Schubert <bschubert@ddn.com>
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Reviewed-by: Baokun Li <libaokun@linux.alibaba.com>
Signed-off-by: Joanne Koong <joannelkoong@gmail.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
Move header copying to ring logic into a new copy_header_to_ring()
function. This makes the copy_to_user() logic more clear and centralizes
error handling / rate-limited logging.
Reviewed-by: Bernd Schubert <bschubert@ddn.com>
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Reviewed-by: Baokun Li <libaokun@linux.alibaba.com>
Signed-off-by: Joanne Koong <joannelkoong@gmail.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
Simplify the logic for fetching + sending off the next request.
This gets rid of fuse_uring_send_next_to_ring() which contained
duplicated logic from fuse_uring_send(). This decouples request fetching
from the send operation, which makes the control flow clearer and
reduces unnecessary parameter passing.
Reviewed-by: Bernd Schubert <bschubert@ddn.com>
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Reviewed-by: Baokun Li <libaokun@linux.alibaba.com>
Signed-off-by: Joanne Koong <joannelkoong@gmail.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
This new tracepoint complements fuse_request_send (enqueue) and
fuse_request_end (completion). It fires after the request has been
successfully copied to the daemon's buffer, just before the daemon
can start to process it.
fuse_request_sent does not fire if the copy of the request fails.
It also does not fire for NOTIFY_REPLY, which fires the _end tracepoint
at the end of copy.
This is needed for tools tracking the in-flight state of user initiated
fuse requests.
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Reviewed-by: Joanne Koong <joannelkoong@gmail.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
Prior to this patchset, fuse_dev (containing fuse_pqueue) was allocated on
mount. But now fuse_dev is allocated when opening /dev/fuse, even though
the queues are not needed at that time.
Delay allocation of the pqueue (4k worth of list_head) just before mounting
or cloning a device.
Various distributions (e.g. Debian/Fedora) configure /dev/fuse as world
writable, so the pqueue allocation should be deferred to a privileged
operation (mount) to prevent unprivileged userspace from consuming pinned
kernel memory.
[Li Wang: fix kernel NULL pointer dereference in fuse_uring_add_to_pq()]
[Fix race in fuse_dev_release()]
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
Move a couple of function declarations from fuse_i.h to dev.h and
fuse_dev_i.h.
Add fuse_conn_get_id() helper that retrieves the connection ID (s_dev) from
fuse_conn.
With the exception of cuse.c, virtio_fs.c and trace.c source files now
either include fuse_i.h or fuse_dev_i/dev_uring_i.h but not both.
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
Store pointer to struct fuse_chan instead of struct fuse_conn in fuse_ring.
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
Set minor, max_write and max_pages in the fuse_chan. These match the same
fields in fuse_conn but are needed in both layers.
[Dongyang Jin: Pointers should use NULL instead of explicit '0']
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
Store pointer to struct fuse_chan instead of struct fuse_conn in fuse_dev.
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
Store a struct fuse_chan pointer in fuse_req instead of a struct fuse_mount
pointer.
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
- rename fuse_abort_conn() to fuse_chan_abort(), pass fuse_chan pointer
instead of fuse_conn
- pass an abort_with_err argument that tells fuse_dev_(read|write) to
return with ECONNABORTED instead of ENODEV
- move fc->aborted to fch->abort_with_err
- rename fuse_wait_aborted() to fuse_chan_wait_aborted()
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
Add helpers (get and set functions mainly) that cleanly separate the
layers.
Remove #include "fuse_dev_i.h" from:
- inode.c
- file.c
- control.c
Remove #include "dev_uring_i.h" from inode.c.
[Li Wang: drop redundant initializer in process_init_limits()]
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
Move:
- timeout
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
And document which members they protect.
end_polls() is called with both, outer fch->lock is probably unnecessary,
but doesn't hurt for now.
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
Move:
- io_uring
- ring
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
Move:
- initialized
- blocked
- blocked_waitq
- connected
- num_waiting
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
Move:
- max_background
- num_background
- active_background
- bg_queue
- bg_lock
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
Move the 'fiq' member from fuse_conn to fuse_chan.
Move iqueue related structure definitions and function declarations from
"fuse_i.h" to "fuse_dev_i.h".
Add a fuse_dev_chan_new() helper, that returns a fuse_chan initialized with
the fuse_dev_fiq_ops.
Add a fuse_chan_release() function, that calls fiq->ops->release().
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
When a background request completes via the io_uring path, the
background queue gets flushed to dispatch pending background requests,
but this is done before the connection-level background counters
(fc->num_background, fc->active_background) are properly accounted,
which may reduce effective queue depth to one.
The connection-level counters are decremented in fuse_request_end(), but
flush_bg_queue() flushes the /dev/fuse path queue (fc->bg_queue), not
the io_uring per-queue bg one, which means pending uring background
requests on the queue are never dispatched in this path.
Fix this by accounting the connection-level background counters first
before flushing the queue's background queue. Since
fuse_request_bg_finish() clears FR_BACKGROUND, fuse_request_end() will
skip the background cleanup branch entirely, which avoids any
double-decrements; it will call the wake_up(&req->waitq) branch but this
is effectively a no-op as background requests have no waiters on
req->waitq.
Reviewed-by: Bernd Schubert <bernd@bsbernd.com>
Fixes: 857b0263f30e ("fuse: Allow to queue bg requests through io-uring")
Cc: stable@vger.kernel.org
Signed-off-by: Joanne Koong <joannelkoong@gmail.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
If a copy into the userspace ring buffer fails, a request will be
terminated and fuse_uring_req_end() will set ent->fuse_req to NULL but
it will leave the entry on ent_w_req_queue in FRRS_FUSE_REQ state. This
can lead to a NULL deref if the request expiration logic scans
ent_w_req_queue in the window before the entry is moved off it.
Fix this by taking the entry off ent_w_req_queue and changing its state
from FRRS_FUSE_REQ to FRRS_INVALID before terminating the request.
Fixes: 4fea593e625c ("fuse: optimize over-io-uring request expiration check")
Cc: stable@kernel.org
Signed-off-by: Joanne Koong <joannelkoong@gmail.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
Bad userspace might try to trick us and send commit SQEs request
unique / commit-id of requests that are not even send to
fuse-server (io_uring_cmd_done() not called) yet.
fuse_uring_commit_fetch() ends the fuse request when the ring entry
has a wrong state, but that could have caused a use-after-free
with the memcpy operations in fuse_uring_send_in_task().
In order to avoid such races the call of fuse_uring_add_to_pq()
is moved after the copy operations and just before completing
the io-uring request - malicious userspace cannot find the request
anymore until all prepration work in fuse-client/kernel is completed.
This also moves fuse_uring_add_to_pq() a bit up in the code to
avoid a forward declaration. Also not with a preparation commit,
to make it easier to back port to older kernels.
Reported-by: xlabai <xlabai@tencent.com>
Reported-by: Berkant Koc <me@berkoc.com>
Fixes: c090c8abae4b6b ("fuse: Add io-uring sqe commit and fetch support")
Cc: stable@kernel.org # 6.14
Signed-off-by: Bernd Schubert <bernd@bsbernd.com>
Reviewed-by: Joanne Koong <joannelkoong@gmail.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
There are several readers of queue->stopped that check the value
under lock, but fuse_uring_commit_fetch() did not and actually
the value was not set under the lock in fuse_uring_abort_end_requests()
either. Especially in fuse_uring_commit_fetch it is important
to check under a lock, because due to races 'struct fuse_req'
might be freed with fuse_request_end, but another thread/cpu
might already do teardown work.
Cc: stable@kernel.org # 6.14
Fixes: 4a9bfb9b6850fec ("fuse: {io-uring} Handle teardown of ring entries")
Reported-by: Berkant Koc <me@berkoc.com>
Reported-by: xlabai <xlabai@tencent.com>
Signed-off-by: Bernd Schubert <bernd@bsbernd.com>
Reviewed-by: Joanne Koong <joannelkoong@gmail.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
fuse_uring_async_stop_queues() might run when the last reference
on ring->queue_refs was already dropped.
In order to avoid an early destruction a reference on struct fuse_conn
is now taken before starting fuse_uring_async_stop_queues() and that
reference is only released when that delayed work queue terminates.
Fixes: 4a9bfb9b6850 ("fuse: {io-uring} Handle teardown of ring entries")
Cc: stable@kernel.org # 6.14
Reported-by: Berkant Koc <me@berkoc.com>
Signed-off-by: Bernd Schubert <bernd@bsbernd.com>
Reviewed-by: Joanne Koong <joannelkoong@gmail.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
When io_uring delivers task work with tw.cancel set (PF_EXITING,
PF_KTHREAD fallback, or percpu_ref_is_dying on the ring context),
fuse_uring_send_in_task() takes the cancel branch, assigns
-ECANCELED, and falls through to fuse_uring_send(). That path only
flips the entry to FRRS_USERSPACE and completes the io_uring cmd;
it never discharges the ring entry's owning reference to the
fuse_req that fuse_uring_add_req_to_ring_ent() handed it at
dispatch time.
fuse_uring_send_in_task()
tw.cancel == true
err = -ECANCELED
fuse_uring_send(ent, cmd, err, issue_flags)
ent->state = FRRS_USERSPACE
list_move(&ent->list, &queue->ent_in_userspace)
ent->cmd = NULL
io_uring_cmd_done(-ECANCELED)
/* ent->fuse_req still set, req still hashed */
The fuse_req stays linked on fpq->processing[hash] and
fuse_request_end() is never invoked. The originating syscall
thread blocks in D-state in request_wait_answer() until
fuse_abort_conn() runs, which can be the entire connection
lifetime. For FR_BACKGROUND requests fc->num_background is never
decremented either, so repeated cancels inflate the counter until
max_background is hit and all later background ops stall. tw.cancel does
not imply a connection abort (e.g. a single io_uring worker thread exits
while the fuse connection stays up), so this cannot be left for
fuse_abort_conn() to clean up.
Ending the req but still routing the entry through fuse_uring_send()
is not enough: that leaves a req-less entry on ent_in_userspace, and
ent_list_request_expired() dereferences ent->fuse_req unconditionally
on the head of that list, which would then NULL-deref.
Fix the cancel branch to release the entry directly. Remove it from the
queue, complete the io_uring cmd, end the fuse_req, free the entry, and
drop its queue_refs (waking the teardown waiter if it was the last).
Fixes: c2c9af9a0b13 ("fuse: Allow to queue fg requests through io-uring")
Cc: stable@vger.kernel.org
Reviewed-by: Joanne Koong <joannelkoong@gmail.com>
Assisted-by: kres (claude-opus-4-7)
Signed-off-by: Chris Mason <clm@meta.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
fuse_uring_cancel() moves entries that are available (these have no reqs
attached) to the ent_in_userspace list. ent_list_request_expired()
checks the first entry on ent_in_userspace and dereferences
ent->fuse_req unconditionally, which will crash on a cancelled entry
that was moved to this list.
Fix this by freeing the entry and dropping queue_refs directly in
fuse_uring_cancel(). This is safe because cancel is the cancel handler
itself - after io_uring_cmd_done(), no more cancels will be dispatched
for this command, and teardown serializes with cancel via queue->lock.
Since cancel now decrements queue_refs, fuse_uring_abort() must no
longer gate fuse_uring_abort_end_requests() on queue_refs > 0, as
cancelled entries may have already dropped queue_refs while requests are
still queued. Remove the gate so abort always flushes requests and stops
queues.
Reported-by: Heechan Kang <gganji11@naver.com>
Tested-by: Heechan Kang <gganji11@naver.com>
Reviewed-by: Bernd Schubert <bernd@bsbernd.com>
Fixes: 4fea593e625c ("fuse: optimize over-io-uring request expiration check")
Cc: stable@vger.kernel.org
Suggested-by: Jian Huang Li <ali@ddn.com>
Suggested-by: Horst Birthelmer <horst@birthelmer.de>
Signed-off-by: Joanne Koong <joannelkoong@gmail.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
Check fch->connected under fch->lock in fuse_uring_create() before
attaching a new ring. Without this, a race between fuse_uring_create()
and fuse_chan_abort() can result in the ring, queue, and fpq.processing
table being created after fuse_uring_abort() has already run, leading
to unnecessary allocation and teardown. These are eventually cleaned up
by fuse_uring_destruct() but will linger until the process exits, even
with the connection aborted.
Reviewed-by: Bernd Schubert <bernd@bsbernd.com>
Signed-off-by: Joanne Koong <joannelkoong@gmail.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
This fixes this race:
- thread a: io_uring_enter -> register sqe ->
fuse_uring_create_ring_ent -> allocate ent but doesn't grab queue_ref
yet
- thread b: fuse_conn_destroy() -> fuse_chan_abort() ->
fuse_uring_abort() is a no-op due to queue ref being 0
- thread a: grabs the queue_ref, queue_ref is now 1, rest of
fuse_uring_do_register() logic executes
- thread b: fuse_chan_abort() returns, fuse_chan_wait_aborted() now runs
and calls
"wait_event(ring->stop_waitq, atomic_read(&ring->queue_refs) == 0);"
The abort/unmount thread will hang indefinitely in unkillable state as
nothing will decrement queue_refs or wake stop_waitq, and the ring,
queue, and ent are leaked.
Fix this by checking fch->connected under fch->lock after the created
ent has grabbed a ref count on the queue. This ensures that in the
scenario above, it is guaranteed that we either release the queue ref
and wake up stop_waitq (in case fuse_chan_wait_aborted() is already
waiting) in fuse_uring_do_register() when we detect !fch->connected, or
if the connection is aborted after the check, it is guaranteed that the
async teardown worker will be running in the background cleaning up ents
and decrementing the ent's ref on the queue, which will unblock the
eventual queue and ring teardown.
Fixes: 24fe962c86f5 ("fuse: {io-uring} Handle SQEs - register commands")
Cc: stable@vger.kernel.org
Reviewed-by: Bernd Schubert <bernd@bsbernd.com>
Signed-off-by: Joanne Koong <joannelkoong@gmail.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
On weakly-ordered architectures, the store to fiq->ops can be
reordered past the store to ring->ready, allowing a CPU that sees
ring->ready == true via fuse_uring_ready() to dispatch requests
through a stale fiq->ops pointer. Upgrade the store to
smp_store_release() and the load in fuse_uring_ready() to
smp_load_acquire() so that the preceding WRITE_ONCE(fiq->ops, ...)
is visible to any CPU that observes ring->ready == true.
Additionally, fuse_uring_do_register() publishes ring->ready with
WRITE_ONCE() but the fast-path check reads it with a plain load.
This is a marked-vs-unmarked access that KCSAN will flag. Wrap it in
READ_ONCE() to mark it without adding unnecessary ordering.
Also wrap the fc->ring load in fuse_uring_ready() in READ_ONCE() to
prevent the compiler from reloading it between the NULL check and the
dereference.
Fixes: c2c9af9a0b13 ("fuse: Allow to queue fg requests through io-uring")
Cc: stable@vger.kernel.org
Reviewed-by: Joanne Koong <joannelkoong@gmail.com>
Assisted-by: kres (claude-opus-4-7)
Signed-off-by: Chris Mason <clm@meta.com>
Reviewed-by: Bernd Schubert <bernd@bsbernd.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
copy_from_user() returns the number of bytes not copied as an unsigned
residual on failure (1..sizeof(struct fuse_out_header)). fuse_uring_commit
stores that residual in ssize_t err, sets req->out.h.error to -EFAULT,
then jumps to out: with err still holding the positive residual.
err = copy_from_user(&req->out.h, &ent->headers->in_out,
sizeof(req->out.h));
if (err) {
req->out.h.error = -EFAULT;
goto out; /* err is the positive residual */
}
...
out:
fuse_uring_req_end(ent, req, err);
fuse_uring_req_end() then runs
if (error)
req->out.h.error = error;
which overwrites the just-assigned -EFAULT with the positive residual.
FUSE callers such as fuse_simple_request() test err < 0 to detect
failure, so the positive value is interpreted as success and the
caller proceeds with an uninitialised or partial req->out.args.
Fix by assigning err = -EFAULT in the failure branch before jumping
to out, so fuse_uring_req_end() receives a negative errno and sets
req->out.h.error to -EFAULT.
Fixes: c090c8abae4b ("fuse: Add io-uring sqe commit and fetch support")
Cc: stable@vger.kernel.org
Reviewed-by: Joanne Koong <joannelkoong@gmail.com>
Assisted-by: kres (claude-opus-4-7)
Signed-off-by: Chris Mason <clm@meta.com>
Reviewed-by: Bernd Schubert <bernd@bsbernd.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
Extract common queue iteration and teardown logic into
fuse_uring_teardown_all_queues() helper function to eliminate code
duplication between fuse_uring_async_stop_queues() and
fuse_uring_stop_queues().
This is a pure refactoring with no functional changes, intended to
improve maintainability.
Signed-off-by: Yuto Ohnuki <ytohnuki@amazon.com>
Reviewed-by: Bernd Schubert <bernd@bsbernd.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
This was done entirely with mindless brute force, using
git grep -l '\<k[vmz]*alloc_objs*(.*, GFP_KERNEL)' |
xargs sed -i 's/\(alloc_objs*(.*\), GFP_KERNEL)/\1)/'
to convert the new alloc_obj() users that had a simple GFP_KERNEL
argument to just drop that argument.
Note that due to the extreme simplicity of the scripting, any slightly
more complex cases spread over multiple lines would not be triggered:
they definitely exist, but this covers the vast bulk of the cases, and
the resulting diff is also then easier to check automatically.
For the same reason the 'flex' versions will be done as a separate
conversion.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/kees/linux
Pull kmalloc_obj conversion from Kees Cook:
"This does the tree-wide conversion to kmalloc_obj() and friends using
coccinelle, with a subsequent small manual cleanup of whitespace
alignment that coccinelle does not handle.
This uncovered a clang bug in __builtin_counted_by_ref(), so the
conversion is preceded by disabling that for current versions of
clang. The imminent clang 22.1 release has the fix.
I've done allmodconfig build tests for x86_64, arm64, i386, and arm. I
did defconfig builds for alpha, m68k, mips, parisc, powerpc, riscv,
s390, sparc, sh, arc, csky, xtensa, hexagon, and openrisc"
* tag 'kmalloc_obj-treewide-v7.0-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/kees/linux:
kmalloc_obj: Clean up after treewide replacements
treewide: Replace kmalloc with kmalloc_obj for non-scalar types
compiler_types: Disable __builtin_counted_by_ref for Clang
|
|
This is the result of running the Coccinelle script from
scripts/coccinelle/api/kmalloc_objs.cocci. The script is designed to
avoid scalar types (which need careful case-by-case checking), and
instead replace kmalloc-family calls that allocate struct or union
object instances:
Single allocations: kmalloc(sizeof(TYPE), ...)
are replaced with: kmalloc_obj(TYPE, ...)
Array allocations: kmalloc_array(COUNT, sizeof(TYPE), ...)
are replaced with: kmalloc_objs(TYPE, COUNT, ...)
Flex array allocations: kmalloc(struct_size(PTR, FAM, COUNT), ...)
are replaced with: kmalloc_flex(*PTR, FAM, COUNT, ...)
(where TYPE may also be *VAR)
The resulting allocations no longer return "void *", instead returning
"TYPE *".
Signed-off-by: Kees Cook <kees@kernel.org>
|
|
For SQE128, sqe->cmd provides 80 bytes for uring_cmd. Add macro to
check if size of user struct does not exceed 80 bytes at compile time.
User doesn't have to track this manually during development.
Replace io_uring_sqe_cmd() inline func with macro and add
io_uring_sqe128_cmd() which checks struct
size for 16 bytes cmd and 80 bytes cmd respectively.
Signed-off-by: Govindarajulu Varadarajan <govind.varadar@gmail.com>
Reviewed-by: Caleb Sander Mateos <csander@purestorage.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/fuse
Pull fuse updates from Miklos Szeredi:
- Add mechanism for cleaning out unused, stale dentries; controlled via
a module option (Luis Henriques)
- Fix various bugs
- Cleanups
* tag 'fuse-update-6.19' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/fuse:
fuse: Uninitialized variable in fuse_epoch_work()
fuse: fix io-uring list corruption for terminated non-committed requests
fuse: signal that a fuse inode should exhibit local fs behaviors
fuse: Always flush the page cache before FOPEN_DIRECT_IO write
fuse: Invalidate the page cache after FOPEN_DIRECT_IO write
fuse: rename 'namelen' to 'namesize'
fuse: use strscpy instead of strcpy
fuse: refactor fuse_conn_put() to remove negative logic.
fuse: new work queue to invalidate dentries from old epochs
fuse: new work queue to periodically invalidate expired dentries
dcache: export shrink_dentry_list() and add new helper d_dispose_if_unused()
fuse: add WARN_ON and comment for RCU revalidate
fuse: Fix whitespace for fuse_uring_args_to_ring() comment
fuse: missing copy_finish in fuse-over-io-uring argument copies
fuse: fix readahead reclaim deadlock
|
|
When a request is terminated before it has been committed, the request
is not removed from the queue's list. This leaves a dangling list entry
that leads to list corruption and use-after-free issues.
Remove the request from the queue's list for terminated non-committed
requests.
Signed-off-by: Joanne Koong <joannelkoong@gmail.com>
Fixes: c090c8abae4b ("fuse: Add io-uring sqe commit and fetch support")
Cc: stable@vger.kernel.org
Reviewed-by: Bernd Schubert <bschubert@ddn.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
The function comment accidentally got wrong indentation.
Signed-off-by: Bernd Schubert <bschubert@ddn.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
