| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
In the current FUSE writeback design (see commit 3be5a52b30aa
("fuse: support writable mmap")), a temp page is allocated for every
dirty page to be written back, the contents of the dirty page are copied
over to the temp page, and the temp page gets handed to the server to
write back.
This is done so that writeback may be immediately cleared on the dirty
page, and this in turn is done in order to mitigate the following
deadlock scenario that may arise if reclaim waits on writeback on the
dirty page to complete:
* single-threaded FUSE server is in the middle of handling a request
that needs a memory allocation
* memory allocation triggers direct reclaim
* direct reclaim waits on a folio under writeback
* the FUSE server can't write back the folio since it's stuck in
direct reclaim
With a recent change that added AS_WRITEBACK_MAY_DEADLOCK_ON_RECLAIM and
mitigates the situation described above, FUSE writeback does not need
to use temp pages if it sets AS_WRITEBACK_MAY_DEADLOCK_ON_RECLAIM on its
inode mappings.
This commit sets AS_WRITEBACK_MAY_DEADLOCK_ON_RECLAIM on the inode
mappings and removes the temporary pages + extra copying and the internal
rb tree.
fio benchmarks --
(using averages observed from 10 runs, throwing away outliers)
Setup:
sudo mount -t tmpfs -o size=30G tmpfs ~/tmp_mount
./libfuse/build/example/passthrough_ll -o writeback -o max_threads=4 -o source=~/tmp_mount ~/fuse_mount
fio --name=writeback --ioengine=sync --rw=write --bs={1k,4k,1M} --size=2G
--numjobs=2 --ramp_time=30 --group_reporting=1 --directory=/root/fuse_mount
bs = 1k 4k 1M
Before 351 MiB/s 1818 MiB/s 1851 MiB/s
After 341 MiB/s 2246 MiB/s 2685 MiB/s
% diff -3% 23% 45%
Signed-off-by: Joanne Koong <joannelkoong@gmail.com>
Reviewed-by: Jingbo Xu <jefflexu@linux.alibaba.com>
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
Currently userspace is able to notify the kernel to invalidate the cache
for an inode. This means that, if all the inodes in a filesystem need to
be invalidated, then userspace needs to iterate through all of them and do
this kernel notification separately.
This patch adds the concept of 'epoch': each fuse connection will have the
current epoch initialized and every new dentry will have it's d_time set to
the current epoch value. A new operation will then allow userspace to
increment the epoch value. Every time a dentry is d_revalidate()'ed, it's
epoch is compared with the current connection epoch and invalidated if it's
value is different.
Signed-off-by: Luis Henriques <luis@igalia.com>
Tested-by: Laura Promberger <laura.promberger@cern.ch>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
Prefer 'unsigned int' to bare 'unsigned', as reported by checkpatch.pl:
WARNING: Prefer 'unsigned int' to bare use of 'unsigned'.
Signed-off-by: Jiale Yang <295107659@qq.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
Our file system has a translation capability for S3-to-posix.
The current value of 1kiB is enough to cover S3 keys, but
does not allow encoding of %xx escape characters.
The limit is increased to (PATH_MAX - 1), as we need
3 x 1024 and that is close to PATH_MAX (4kB) already.
-1 is used as the terminating null is not included in the
length calculation.
Testing large file names was hard with libfuse/example file systems,
so I created a new memfs that does not have a 255 file name length
limitation.
https://github.com/libfuse/libfuse/pull/1077
The connection is initialized with FUSE_NAME_LOW_MAX, which
is set to the previous value of FUSE_NAME_MAX of 1024. With
FUSE_MIN_READ_BUFFER of 8192 that is enough for two file names
+ fuse headers.
When FUSE_INIT reply sets max_pages to a value > 1 we know
that fuse daemon supports request buffers of at least 2 pages
(+ header) and can therefore hold 2 x PATH_MAX file names - operations
like rename or link that need two file names are no issue then.
Signed-off-by: Bernd Schubert <bschubert@ddn.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
Introduce two new sysctls, "default_request_timeout" and
"max_request_timeout". These control how long (in seconds) a server can
take to reply to a request. If the server does not reply by the timeout,
then the connection will be aborted. The upper bound on these sysctl
values is 65535.
"default_request_timeout" sets the default timeout if no timeout is
specified by the fuse server on mount. 0 (default) indicates no default
timeout should be enforced. If the server did specify a timeout, then
default_request_timeout will be ignored.
"max_request_timeout" sets the max amount of time the server may take to
reply to a request. 0 (default) indicates no maximum timeout. If
max_request_timeout is set and the fuse server attempts to set a
timeout greater than max_request_timeout, the system will use
max_request_timeout as the timeout. Similarly, if default_request_timeout
is greater than max_request_timeout, the system will use
max_request_timeout as the timeout. If the server does not request a
timeout and default_request_timeout is set to 0 but max_request_timeout
is set, then the timeout will be max_request_timeout.
Please note that these timeouts are not 100% precise. The request may
take roughly an extra FUSE_TIMEOUT_TIMER_FREQ seconds beyond the set max
timeout due to how it's internally implemented.
$ sysctl -a | grep fuse.default_request_timeout
fs.fuse.default_request_timeout = 0
$ echo 65536 | sudo tee /proc/sys/fs/fuse/default_request_timeout
tee: /proc/sys/fs/fuse/default_request_timeout: Invalid argument
$ echo 65535 | sudo tee /proc/sys/fs/fuse/default_request_timeout
65535
$ sysctl -a | grep fuse.default_request_timeout
fs.fuse.default_request_timeout = 65535
$ echo 0 | sudo tee /proc/sys/fs/fuse/default_request_timeout
0
$ sysctl -a | grep fuse.default_request_timeout
fs.fuse.default_request_timeout = 0
[Luis Henriques: Limit the timeout to the range [FUSE_TIMEOUT_TIMER_FREQ,
fuse_max_req_timeout]]
Signed-off-by: Joanne Koong <joannelkoong@gmail.com>
Reviewed-by: Bernd Schubert <bschubert@ddn.com>
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Reviewed-by: Sergey Senozhatsky <senozhatsky@chromium.org>
Reviewed-by: Luis Henriques <luis@igalia.com>
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
There are situations where fuse servers can become unresponsive or
stuck, for example if the server is deadlocked. Currently, there's no
good way to detect if a server is stuck and needs to be killed manually.
This commit adds an option for enforcing a timeout (in seconds) for
requests where if the timeout elapses without the server responding to
the request, the connection will be automatically aborted.
Please note that these timeouts are not 100% precise. For example, the
request may take roughly an extra FUSE_TIMEOUT_TIMER_FREQ seconds beyond
the requested timeout due to internal implementation, in order to
mitigate overhead.
[SzM: Bump the API version number]
Signed-off-by: Joanne Koong <joannelkoong@gmail.com>
Reviewed-by: Jeff Layton <jlayton@kernel.org>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
If filesystem doesn't support FUSE_LINK (i.e. returns -ENOSYS), then
remember this and next time return immediately, without incurring the
overhead of a round trip to the server.
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
task-A (application) might be in request_wait_answer and
try to remove the request when it has FR_PENDING set.
task-B (a fuse-server io-uring task) might handle this
request with FUSE_IO_URING_CMD_COMMIT_AND_FETCH, when
fetching the next request and accessed the req from
the pending list in fuse_uring_ent_assign_req().
That code path was not protected by fiq->lock and so
might race with task-A.
For scaling reasons we better don't use fiq->lock, but
add a handler to remove canceled requests from the queue.
This also removes usage of fiq->lock from
fuse_uring_add_req_to_ring_ent() altogether, as it was
there just to protect against this race and incomplete.
Also added is a comment why FR_PENDING is not cleared.
Fixes: c090c8abae4b ("fuse: Add io-uring sqe commit and fetch support")
Cc: <stable@vger.kernel.org> # v6.14
Reported-by: Joanne Koong <joannelkoong@gmail.com>
Closes: https://lore.kernel.org/all/CAJnrk1ZgHNb78dz-yfNTpxmW7wtT88A=m-zF0ZoLXKLUHRjNTw@mail.gmail.com/
Signed-off-by: Bernd Schubert <bschubert@ddn.com>
Reviewed-by: Joanne Koong <joannelkoong@gmail.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
Avoid races and block request allocation until io-uring
queues are ready.
This is a especially important for background requests,
as bg request completion might cause lock order inversion
of the typical queue->lock and then fc->bg_lock
fuse_request_end
spin_lock(&fc->bg_lock);
flush_bg_queue
fuse_send_one
fuse_uring_queue_fuse_req
spin_lock(&queue->lock);
Signed-off-by: Bernd Schubert <bernd@bsbernd.com>
Reviewed-by: Luis Henriques <luis@igalia.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
This adds support for fuse request completion through ring SQEs
(FUSE_URING_CMD_COMMIT_AND_FETCH handling). After committing
the ring entry it becomes available for new fuse requests.
Handling of requests through the ring (SQE/CQE handling)
is complete now.
Fuse request data are copied through the mmaped ring buffer,
there is no support for any zero copy yet.
Signed-off-by: Bernd Schubert <bschubert@ddn.com>
Reviewed-by: Pavel Begunkov <asml.silence@gmail.com> # io_uring
Reviewed-by: Luis Henriques <luis@igalia.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
fuse-over-io-uring uses existing functions to find requests based
on their unique id - make these functions non-static.
Signed-off-by: Bernd Schubert <bschubert@ddn.com>
Reviewed-by: Joanne Koong <joannelkoong@gmail.com>
Reviewed-by: Luis Henriques <luis@igalia.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
This adds basic support for ring SQEs (with opcode=IORING_OP_URING_CMD).
For now only FUSE_IO_URING_CMD_REGISTER is handled to register queue
entries.
Signed-off-by: Bernd Schubert <bschubert@ddn.com>
Reviewed-by: Pavel Begunkov <asml.silence@gmail.com> # io_uring
Reviewed-by: Luis Henriques <luis@igalia.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
This change sets up FUSE operations to always have headers in
args.in_args[0], even for opcodes without an actual header.
This step prepares for a clean separation of payload from headers,
initially it is used by fuse-over-io-uring.
For opcodes without a header, we use a zero-sized struct as a
placeholder. This approach:
- Keeps things consistent across all FUSE operations
- Will help with payload alignment later
- Avoids future issues when header sizes change
Op codes that already have an op code specific header do not
need modification.
Op codes that have neither payload nor op code headers
are not modified either (FUSE_READLINK and FUSE_DESTROY).
FUSE_BATCH_FORGET already has the header in the right place,
but is not using fuse_copy_args - as -over-uring is currently
not handling forgets it does not matter for now, but header
separation will later need special attention for that op code.
Correct the struct fuse_args->in_args array max size.
Signed-off-by: Bernd Schubert <bschubert@ddn.com>
Reviewed-by: Joanne Koong <joannelkoong@gmail.com>
Reviewed-by: Luis Henriques <luis@igalia.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
Function fuse_direntplus_link() might call fuse_iget() to initialize a new
fuse_inode and change its attributes. If fi->attr_version is always
initialized with 0, even if the attributes returned by the FUSE_READDIR
request is staled, as the new fi->attr_version is 0, fuse_change_attributes
will still set the staled attributes to inode. This wrong behaviour may
cause file size inconsistency even when there is no changes from
server-side.
To reproduce the issue, consider the following 2 programs (A and B) are
running concurrently,
A B
---------------------------------- --------------------------------
{ /fusemnt/dir/f is a file path in a fuse mount, the size of f is 0. }
readdir(/fusemnt/dir) start
//Daemon set size 0 to f direntry
fallocate(f, 1024)
stat(f) // B see size 1024
echo 2 > /proc/sys/vm/drop_caches
readdir(/fusemnt/dir) reply to kernel
Kernel set 0 to the I_NEW inode
stat(f) // B see size 0
In the above case, only program B is modifying the file size, however, B
observes file size changing between the 2 'readonly' stat() calls. To fix
this issue, we should make sure readdirplus still follows the rule of
attr_version staleness checking even if the fi->attr_version is lost due to
inode eviction.
To identify this situation, the new fc->evict_ctr is used to record whether
the eviction of inodes occurs during the readdirplus request processing.
If it does, the result of readdirplus may be inaccurate; otherwise, the
result of readdirplus can be trusted. Although this may still lead to
incorrect invalidation, considering the relatively low frequency of
evict occurrences, it should be acceptable.
Link: https://lore.kernel.org/lkml/20230711043405.66256-2-zhangjiachen.jaycee@bytedance.com/
Link: https://lore.kernel.org/lkml/20241114070905.48901-1-zhangtianci.1997@bytedance.com/
Reported-by: Jiachen Zhang <zhangjiachen.jaycee@bytedance.com>
Suggested-by: Miklos Szeredi <miklos@szeredi.hu>
Signed-off-by: Zhang Tianci <zhangtianci.1997@bytedance.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
All fuse requests use folios instead of pages for transferring data.
Remove pages from the requests and exclusively use folios.
No functional changes.
[SzM: rename back folio_descs -> descs, etc.]
Signed-off-by: Joanne Koong <joannelkoong@gmail.com>
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
Convert direct io requests to use folios instead of pages.
No functional changes.
Signed-off-by: Joanne Koong <joannelkoong@gmail.com>
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
Convert ioctl requests to use folios instead of pages.
No functional changes.
Signed-off-by: Joanne Koong <joannelkoong@gmail.com>
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
Convert non-writeback write requests to use folios instead of pages.
No functional changes.
Signed-off-by: Joanne Koong <joannelkoong@gmail.com>
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
Convert read requests to use folios instead of pages.
No functional changes.
Signed-off-by: Joanne Koong <joannelkoong@gmail.com>
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
This adds support in struct fuse_args_pages and fuse_copy_pages() for
using folios instead of pages for transferring data. Both folios and
pages must be supported right now in struct fuse_args_pages and
fuse_copy_pages() until all request types have been converted to use
folios. Once all have been converted, then
struct fuse_args_pages and fuse_copy_pages() will only support folios.
Right now in fuse, all folios are one page (large folios are not yet
supported). As such, copying folio->page is sufficient for copying
the entire folio in fuse_copy_pages().
No functional changes.
Signed-off-by: Joanne Koong <joannelkoong@gmail.com>
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
When trying to insert a 10MB kernel module kept in a virtio-fs with cache
disabled, the following warning was reported:
------------[ cut here ]------------
WARNING: CPU: 1 PID: 404 at mm/page_alloc.c:4551 ......
Modules linked in:
CPU: 1 PID: 404 Comm: insmod Not tainted 6.9.0-rc5+ #123
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996) ......
RIP: 0010:__alloc_pages+0x2bf/0x380
......
Call Trace:
<TASK>
? __warn+0x8e/0x150
? __alloc_pages+0x2bf/0x380
__kmalloc_large_node+0x86/0x160
__kmalloc+0x33c/0x480
virtio_fs_enqueue_req+0x240/0x6d0
virtio_fs_wake_pending_and_unlock+0x7f/0x190
queue_request_and_unlock+0x55/0x60
fuse_simple_request+0x152/0x2b0
fuse_direct_io+0x5d2/0x8c0
fuse_file_read_iter+0x121/0x160
__kernel_read+0x151/0x2d0
kernel_read+0x45/0x50
kernel_read_file+0x1a9/0x2a0
init_module_from_file+0x6a/0xe0
idempotent_init_module+0x175/0x230
__x64_sys_finit_module+0x5d/0xb0
x64_sys_call+0x1c3/0x9e0
do_syscall_64+0x3d/0xc0
entry_SYSCALL_64_after_hwframe+0x4b/0x53
......
</TASK>
---[ end trace 0000000000000000 ]---
The warning is triggered as follows:
1) syscall finit_module() handles the module insertion and it invokes
kernel_read_file() to read the content of the module first.
2) kernel_read_file() allocates a 10MB buffer by using vmalloc() and
passes it to kernel_read(). kernel_read() constructs a kvec iter by
using iov_iter_kvec() and passes it to fuse_file_read_iter().
3) virtio-fs disables the cache, so fuse_file_read_iter() invokes
fuse_direct_io(). As for now, the maximal read size for kvec iter is
only limited by fc->max_read. For virtio-fs, max_read is UINT_MAX, so
fuse_direct_io() doesn't split the 10MB buffer. It saves the address and
the size of the 10MB-sized buffer in out_args[0] of a fuse request and
passes the fuse request to virtio_fs_wake_pending_and_unlock().
4) virtio_fs_wake_pending_and_unlock() uses virtio_fs_enqueue_req() to
queue the request. Because virtiofs need DMA-able address, so
virtio_fs_enqueue_req() uses kmalloc() to allocate a bounce buffer for
all fuse args, copies these args into the bounce buffer and passed the
physical address of the bounce buffer to virtiofsd. The total length of
these fuse args for the passed fuse request is about 10MB, so
copy_args_to_argbuf() invokes kmalloc() with a 10MB size parameter and
it triggers the warning in __alloc_pages():
if (WARN_ON_ONCE_GFP(order > MAX_PAGE_ORDER, gfp))
return NULL;
5) virtio_fs_enqueue_req() will retry the memory allocation in a
kworker, but it won't help, because kmalloc() will always return NULL
due to the abnormal size and finit_module() will hang forever.
A feasible solution is to limit the value of max_read for virtio-fs, so
the length passed to kmalloc() will be limited. However it will affect
the maximal read size for normal read. And for virtio-fs write initiated
from kernel, it has the similar problem but now there is no way to limit
fc->max_write in kernel.
So instead of limiting both the values of max_read and max_write in
kernel, introducing use_pages_for_kvec_io in fuse_conn and setting it as
true in virtiofs. When use_pages_for_kvec_io is enabled, fuse will use
pages instead of pointer to pass the KVEC_IO data.
After switching to pages for KVEC_IO data, these pages will be used for
DMA through virtio-fs. If these pages are backed by vmalloc(),
{flush|invalidate}_kernel_vmap_range() are necessary to flush or
invalidate the cache before the DMA operation. So add two new fields in
fuse_args_pages to record the base address of vmalloc area and the
condition indicating whether invalidation is needed. Perform the flush
in fuse_get_user_pages() for write operations and the invalidation in
fuse_release_user_pages() for read operations.
It may seem necessary to introduce another field in fuse_conn to
indicate that these KVEC_IO pages are used for DMA, However, considering
that virtio-fs is currently the only user of use_pages_for_kvec_io, just
reuse use_pages_for_kvec_io to indicate that these pages will be used
for DMA.
Fixes: a62a8ef9d97d ("virtio-fs: add virtiofs filesystem")
Signed-off-by: Hou Tao <houtao1@huawei.com>
Tested-by: Jingbo Xu <jefflexu@linux.alibaba.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
Introduce the capability to dynamically configure the max pages limit
(FUSE_MAX_MAX_PAGES) through a sysctl. This allows system administrators
to dynamically set the maximum number of pages that can be used for
servicing requests in fuse.
Previously, this is gated by FUSE_MAX_MAX_PAGES which is statically set
to 256 pages. One result of this is that the buffer size for a write
request is limited to 1 MiB on a 4k-page system.
The default value for this sysctl is the original limit (256 pages).
$ sysctl -a | grep max_pages_limit
fs.fuse.max_pages_limit = 256
$ sysctl -n fs.fuse.max_pages_limit
256
$ echo 1024 | sudo tee /proc/sys/fs/fuse/max_pages_limit
1024
$ sysctl -n fs.fuse.max_pages_limit
1024
$ echo 65536 | sudo tee /proc/sys/fs/fuse/max_pages_limit
tee: /proc/sys/fs/fuse/max_pages_limit: Invalid argument
$ echo 0 | sudo tee /proc/sys/fs/fuse/max_pages_limit
tee: /proc/sys/fs/fuse/max_pages_limit: Invalid argument
$ echo 65535 | sudo tee /proc/sys/fs/fuse/max_pages_limit
65535
$ sysctl -n fs.fuse.max_pages_limit
65535
Signed-off-by: Joanne Koong <joannelkoong@gmail.com>
Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Reviewed-by: Sweet Tea Dorminy <sweettea-kernel@dorminy.me>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
We should convert fs/fuse code to use a newly introduced
invalid_mnt_idmap instead of passing a NULL as idmap pointer.
Suggested-by: Christian Brauner <brauner@kernel.org>
Signed-off-by: Alexander Mikhalitsyn <aleksandr.mikhalitsyn@canonical.com>
Reviewed-by: Christian Brauner <brauner@kernel.org>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
Let's convert all existing callers properly.
No functional changes intended.
Suggested-by: Christian Brauner <brauner@kernel.org>
Signed-off-by: Alexander Mikhalitsyn <aleksandr.mikhalitsyn@canonical.com>
Reviewed-by: Christian Brauner <brauner@kernel.org>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
Need to translate uid and gid in case of chown(2).
Signed-off-by: Alexander Mikhalitsyn <aleksandr.mikhalitsyn@canonical.com>
Reviewed-by: Christian Brauner <brauner@kernel.org>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
If idmap == NULL *and* filesystem daemon declared idmapped mounts
support, then uid/gid values in a fuse header will be -1.
No functional changes intended.
Signed-off-by: Alexander Mikhalitsyn <aleksandr.mikhalitsyn@canonical.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
fuse_mount_list doesn't exist, use fuse_conn_list.
Signed-off-by: Aurelien Aptel <aaptel@nvidia.com>
Reviewed-by: Bernd Schubert <bschubert@ddn.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
Virtiofs has its own queuing mechanism, but still requests are first queued
on fiq->pending to be immediately dequeued and queued onto the virtio
queue.
The queuing on fiq->pending is unnecessary and might even have some
performance impact due to being a contention point.
Forget requests are handled similarly.
Move the queuing of requests and forgets into the fiq->ops->*.
fuse_iqueue_ops are renamed to reflect the new semantics.
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Fixed-by: Jingbo Xu <jefflexu@linux.alibaba.com>
Reviewed-by: Jingbo Xu <jefflexu@linux.alibaba.com>
Tested-by: Peter-Jan Gootzen <pgootzen@nvidia.com>
Reviewed-by: Peter-Jan Gootzen <pgootzen@nvidia.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
There is a confusion with fuse_file_uncached_io_{start,end} interface.
These helpers do two things when called from passthrough open()/release():
1. Take/drop negative refcount of fi->iocachectr (inode uncached io mode)
2. State change ff->iomode IOM_NONE <-> IOM_UNCACHED (file uncached open)
The calls from parallel dio write path need to take a reference on
fi->iocachectr, but they should not be changing ff->iomode state, because
in this case, the fi->iocachectr reference does not stick around until file
release().
Factor out helpers fuse_inode_uncached_io_{start,end}, to be used from
parallel dio write path and rename fuse_file_*cached_io_{start,end} helpers
to fuse_file_*cached_io_{open,release} to clarify the difference.
Fixes: 205c1d802683 ("fuse: allow parallel dio writes with FUSE_DIRECT_IO_ALLOW_MMAP")
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/fuse
Pull fuse updates from Miklos Szeredi:
- Add passthrough mode for regular file I/O.
This allows performing read and write (also via memory maps) on a
backing file without incurring the overhead of roundtrips to
userspace. For now this is only allowed to privileged servers, but
this limitation will go away in the future (Amir Goldstein)
- Fix interaction of direct I/O mode with memory maps (Bernd Schubert)
- Export filesystem tags through sysfs for virtiofs (Stefan Hajnoczi)
- Allow resending queued requests for server crash recovery (Zhao Chen)
- Misc fixes and cleanups
* tag 'fuse-update-6.9' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/fuse: (38 commits)
fuse: get rid of ff->readdir.lock
fuse: remove unneeded lock which protecting update of congestion_threshold
fuse: Fix missing FOLL_PIN for direct-io
fuse: remove an unnecessary if statement
fuse: Track process write operations in both direct and writethrough modes
fuse: Use the high bit of request ID for indicating resend requests
fuse: Introduce a new notification type for resend pending requests
fuse: add support for explicit export disabling
fuse: __kuid_val/__kgid_val helpers in fuse_fill_attr_from_inode()
fuse: fix typo for fuse_permission comment
fuse: Convert fuse_writepage_locked to take a folio
fuse: Remove fuse_writepage
virtio_fs: remove duplicate check if queue is broken
fuse: use FUSE_ROOT_ID in fuse_get_root_inode()
fuse: don't unhash root
fuse: fix root lookup with nonzero generation
fuse: replace remaining make_bad_inode() with fuse_make_bad()
virtiofs: drop __exit from virtio_fs_sysfs_exit()
fuse: implement passthrough for mmap
fuse: implement splice read/write passthrough
...
|
|
The same protection is provided by file->f_pos_lock.
Note, this relies on the fact that file->f_mode has FMODE_ATOMIC_POS.
This flag is cleared by stream_open(), which would prevent locking of
f_pos_lock.
Prior to commit 7de64d521bf9 ("fuse: break up fuse_open_common()")
FOPEN_STREAM on a directory would cause stream_open() to be called.
After this commit this is not done anymore, so f_pos_lock will always
be locked.
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
Our user space filesystem relies on fuse to provide POSIX interface.
In our test, a known string is written into a file and the content
is read back later to verify correct data returned. We observed wrong
data returned in read buffer in rare cases although correct data are
stored in our filesystem.
Fuse kernel module calls iov_iter_get_pages2() to get the physical
pages of the user-space read buffer passed in read(). The pages are
not pinned to avoid page migration. When page migration occurs, the
consequence are two-folds.
1) Applications do not receive correct data in read buffer.
2) fuse kernel writes data into a wrong place.
Using iov_iter_extract_pages() to pin pages fixes the issue in our
test.
An auxiliary variable "struct page **pt_pages" is used in the patch
to prepare the 2nd parameter for iov_iter_extract_pages() since
iov_iter_get_pages2() uses a different type for the 2nd parameter.
[SzM] add iov_iter_extract_will_pin(ii) and unpin only if true.
Signed-off-by: Lei Huang <lei.huang@linux.intel.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
The root inode is assumed to be always hashed. Do not unhash the root
inode even if it is marked BAD.
Fixes: 5d069dbe8aaf ("fuse: fix bad inode")
Cc: <stable@vger.kernel.org> # v5.11
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
An mmap request for a file open in passthrough mode, maps the memory
directly to the backing file.
An mmap of a file in direct io mode, usually uses cached mmap and puts
the inode in caching io mode, which denies new passthrough opens of that
inode, because caching io mode is conflicting with passthrough io mode.
For the same reason, trying to mmap a direct io file, while there is
a passthrough file open on the same inode will fail with -ENODEV.
An mmap of a file in direct io mode, also needs to wait for parallel
dio writes in-progress to complete.
If a passthrough file is opened, while an mmap of another direct io
file is waiting for parallel dio writes to complete, the wait is aborted
and mmap fails with -ENODEV.
A FUSE server that uses passthrough and direct io opens on the same inode
that may also be mmaped, is advised to provide a backing fd also for the
files that are open in direct io mode (i.e. use the flags combination
FOPEN_DIRECT_IO | FOPEN_PASSTHROUGH), so that mmap will always use the
backing file, even if read/write do not passthrough.
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
This allows passing fstests generic/249 and generic/591.
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
Use the backing file read/write helpers to implement read/write
passthrough to a backing file.
After read/write, we invalidate a/c/mtime/size attributes.
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
After getting a backing file id with FUSE_DEV_IOC_BACKING_OPEN ioctl,
a FUSE server can reply to an OPEN request with flag FOPEN_PASSTHROUGH
and the backing file id.
The FUSE server should reuse the same backing file id for all the open
replies of the same FUSE inode and open will fail (with -EIO) if a the
server attempts to open the same inode with conflicting io modes or to
setup passthrough to two different backing files for the same FUSE inode.
Using the same backing file id for several different inodes is allowed.
Opening a new file with FOPEN_DIRECT_IO for an inode that is already
open for passthrough is allowed, but only if the FOPEN_PASSTHROUGH flag
and correct backing file id are specified as well.
The read/write IO of such files will not use passthrough operations to
the backing file, but mmap, which does not support direct_io, will use
the backing file insead of using the page cache as it always did.
Even though all FUSE passthrough files of the same inode use the same
backing file as a backing inode reference, each FUSE file opens a unique
instance of a backing_file object to store the FUSE path that was used
to open the inode and the open flags of the specific open file.
The per-file, backing_file object is released along with the FUSE file.
The inode associated fuse_backing object is released when the last FUSE
passthrough file of that inode is released AND when the backing file id
is closed by the server using the FUSE_DEV_IOC_BACKING_CLOSE ioctl.
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
In preparation for opening file in passthrough mode, store the
fuse_open_out argument in ff->args to be passed into fuse_file_io_open()
with the optional backing_id member.
This will be used for setting up passthrough to backing file on open
reply with FOPEN_PASSTHROUGH flag and a valid backing_id.
Opening a file in passthrough mode may fail for several reasons, such as
missing capability, conflicting open flags or inode in caching mode.
Return EIO from fuse_file_io_open() in those cases.
The combination of FOPEN_PASSTHROUGH and FOPEN_DIRECT_IO is allowed -
it mean that read/write operations will go directly to the server,
but mmap will be done to the backing file.
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
FUSE server calls the FUSE_DEV_IOC_BACKING_OPEN ioctl with a backing file
descriptor. If the call succeeds, a backing file identifier is returned.
A later change will be using this backing file id in a reply to OPEN
request with the flag FOPEN_PASSTHROUGH to setup passthrough of file
operations on the open FUSE file to the backing file.
The FUSE server should call FUSE_DEV_IOC_BACKING_CLOSE ioctl to close the
backing file by its id.
This can be done at any time, but if an open reply with FOPEN_PASSTHROUGH
flag is still in progress, the open may fail if the backing file is
closed before the fuse file was opened.
Setting up backing files requires a server with CAP_SYS_ADMIN privileges.
For the backing file to be successfully setup, the backing file must
implement both read_iter and write_iter file operations.
The limitation on the level of filesystem stacking allowed for the
backing file is enforced before setting up the backing file.
Signed-off-by: Alessio Balsini <balsini@android.com>
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
->permission(), ->get_link() and ->inode_get_acl() might dereference
->s_fs_info (and, in case of ->permission(), ->s_fs_info->fc->user_ns
as well) when called from rcu pathwalk.
Freeing ->s_fs_info->fc is rcu-delayed; we need to make freeing ->s_fs_info
and dropping ->user_ns rcu-delayed too.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
|
|
FUSE_PASSTHROUGH capability to passthrough FUSE operations to backing
files will be made available with kernel config CONFIG_FUSE_PASSTHROUGH.
When requesting FUSE_PASSTHROUGH, userspace needs to specify the
max_stack_depth that is allowed for FUSE on top of backing files.
Introduce the flag FOPEN_PASSTHROUGH and backing_id to fuse_open_out
argument that can be used when replying to OPEN request, to setup
passthrough of io operations on the fuse inode to a backing file.
Introduce a refcounted fuse_backing object that will be used to
associate an open backing file with a fuse inode.
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
Instead of denying caching mode on parallel dio open, deny caching
open only while parallel dio are in-progress and wait for in-progress
parallel dio writes before entering inode caching io mode.
This allows executing parallel dio when inode is not in caching mode
even if shared mmap is allowed, but no mmaps have been performed on
the inode in question.
An mmap on direct_io file now waits for all in-progress parallel dio
writes to complete, so parallel dio writes together with
FUSE_DIRECT_IO_ALLOW_MMAP is enabled by this commit.
Signed-off-by: Bernd Schubert <bschubert@ddn.com>
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
The fuse inode io mode is determined by the mode of its open files/mmaps
and parallel dio opens and expressed in the value of fi->iocachectr:
> 0 - caching io: files open in caching mode or mmap on direct_io file
< 0 - parallel dio: direct io mode with parallel dio writes enabled
== 0 - direct io: no files open in caching mode and no files mmaped
Note that iocachectr value of 0 might become positive or negative,
while non-parallel dio is getting processed.
direct_io mmap uses page cache, so first mmap will mark the file as
ff->io_opened and increment fi->iocachectr to enter the caching io mode.
If the server opens the file in caching mode while it is already open
for parallel dio or vice versa the open fails.
This allows executing parallel dio when inode is not in caching mode
and no mmaps have been performed on the inode in question.
Signed-off-by: Bernd Schubert <bschubert@ddn.com>
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
In preparation for inode io modes, a server open response could fail due to
conflicting inode io modes.
Allow returning an error from fuse_finish_open() and handle the error in
the callers.
fuse_finish_open() is used as the callback of finish_open(), so that
FMODE_OPENED will not be set if fuse_finish_open() fails.
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
fuse_open_common() has a lot of code relevant only for regular files and
O_TRUNC in particular.
Copy the little bit of remaining code into fuse_dir_open() and stop using
this common helper for directory open.
Also split out fuse_dir_finish_open() from fuse_finish_open() before we add
inode io modes to fuse_finish_open().
Suggested-by: Miklos Szeredi <miklos@szeredi.hu>
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
This removed the need to pass isdir argument to fuse_put_file().
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
Fuse submounts do not perform a lookup for the nodeid that they inherit
from their parent. Instead, the code decrements the nlookup on the
submount's fuse_inode when it is instantiated, and no forget is
performed when a submount root is evicted.
Trouble arises when the submount's parent is evicted despite the
submount itself being in use. In this author's case, the submount was
in a container and deatched from the initial mount namespace via a
MNT_DEATCH operation. When memory pressure triggered the shrinker, the
inode from the parent was evicted, which triggered enough forgets to
render the submount's nodeid invalid.
Since submounts should still function, even if their parent goes away,
solve this problem by sharing refcounted state between the parent and
its submount. When all of the references on this shared state reach
zero, it's safe to forget the final lookup of the fuse nodeid.
Signed-off-by: Krister Johansen <kjlx@templeofstupid.com>
Cc: stable@vger.kernel.org
Fixes: 1866d779d5d2 ("fuse: Allow fuse_fill_super_common() for submounts")
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
Although DIRECT_IO_RELAX's initial usage is to allow shared mmap, its
description indicates a purpose of reducing memory footprint. This
may imply that it could be further used to relax other DIRECT_IO
operations in the future.
Replace it with a flag DIRECT_IO_ALLOW_MMAP which does only one thing,
allow shared mmap of DIRECT_IO files while still bypassing the cache
on regular reads and writes.
[Miklos] Also Keep DIRECT_IO_RELAX definition for backward compatibility.
Signed-off-by: Tyler Fanelli <tfanelli@redhat.com>
Fixes: e78662e818f9 ("fuse: add a new fuse init flag to relax restrictions in no cache mode")
Cc: <stable@vger.kernel.org> # v6.6
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
|
This makes it harder for accidental or malicious changes to
fuse_xattr_handlers at runtime.
Cc: Miklos Szeredi <miklos@szeredi.hu>
Cc: linux-fsdevel@vger.kernel.org
Signed-off-by: Wedson Almeida Filho <walmeida@microsoft.com>
Link: https://lore.kernel.org/r/20230930050033.41174-12-wedsonaf@gmail.com
Signed-off-by: Christian Brauner <brauner@kernel.org>
|
|
Not all inode attributes are supported by all filesystems, but for the
basic stats (which are returned by stat(2) and friends) all of them will
have some value, even if that doesn't reflect a real attribute of the file.
Btime is different, in that filesystems are free to report or not report a
value in statx. If the value is available, then STATX_BTIME bit is set in
stx_mask.
When caching the value of btime, remember the availability of the attribute
as well as the value (if available). This is done by using the
FUSE_I_BTIME bit in fuse_inode->state to indicate availability, while using
fuse_inode->inval_mask & STATX_BTIME to indicate the state of the cache
itself (i.e. set if cache is invalid, and cleared if cache is valid).
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
|
