In case of fuse the MM subsystem doesn't guarantee that page writeback
completes by the time ->sync_fs() is called. This is because fuse
completes page writeback immediately to prevent DoS of memory reclaim by
the userspace file server.
This means that fuse itself must ensure that writes are synced before
sending the SYNCFS request to the server.
Introduce sync buckets, that hold a counter for the number of outstanding
write requests. On syncfs replace the current bucket with a new one and
wait until the old bucket's counter goes down to zero.
It is possible to have multiple syncfs calls in parallel, in which case
there could be more than one waited-on buckets. Descendant buckets must
not complete until the parent completes. Add a count to the child (new)
bucket until the (parent) old bucket completes.
Use RCU protection to dereference the current bucket and to wake up an
emptied bucket. Use fc->lock to protect against parallel assignments to
the current bucket.
This leaves just the counter to be a possible scalability issue. The
fc->num_waiting counter has a similar issue, so both should be addressed at
the same time.
Reported-by: Amir Goldstein <amir73il@gmail.com>
Fixes: 2d82ab251e ("virtiofs: propagate sync() to file server")
Cc: <stable@vger.kernel.org> # v5.14
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
-----BEGIN PGP SIGNATURE-----
iHUEABYIAB0WIQSQHSd0lITzzeNWNm3h3BK/laaZPAUCYTDKKAAKCRDh3BK/laaZ
PG9PAQCUF0fdBlCKudwSEt5PV5xemycL9OCAlYCd7d4XbBIe9wEA6sVJL9J+OwV2
aF0NomiXtJccE+S9+byjVCyqSzQJGQQ=
=6L2Y
-----END PGP SIGNATURE-----
Merge tag 'ovl-update-5.15' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/vfs
Pull overlayfs update from Miklos Szeredi:
- Copy up immutable/append/sync/noatime attributes (Amir Goldstein)
- Improve performance by enabling RCU lookup.
- Misc fixes and improvements
The reason this touches so many files is that the ->get_acl() method now
gets a "bool rcu" argument. The ->get_acl() API was updated based on
comments from Al and Linus:
Link: https://lore.kernel.org/linux-fsdevel/CAJfpeguQxpd6Wgc0Jd3ks77zcsAv_bn0q17L3VNnnmPKu11t8A@mail.gmail.com/
* tag 'ovl-update-5.15' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/vfs:
ovl: enable RCU'd ->get_acl()
vfs: add rcu argument to ->get_acl() callback
ovl: fix BUG_ON() in may_delete() when called from ovl_cleanup()
ovl: use kvalloc in xattr copy-up
ovl: update ctime when changing fileattr
ovl: skip checking lower file's i_writecount on truncate
ovl: relax lookup error on mismatch origin ftype
ovl: do not set overlay.opaque for new directories
ovl: add ovl_allow_offline_changes() helper
ovl: disable decoding null uuid with redirect_dir
ovl: consistent behavior for immutable/append-only inodes
ovl: copy up sync/noatime fileattr flags
ovl: pass ovl_fs to ovl_check_setxattr()
fs: add generic helper for filling statx attribute flags
Callers of fuse_writeback_range() assume that the file is ready for
modification by the server in the supplied byte range after the call
returns.
If there's a write that extends the file beyond the end of the supplied
range, then the file needs to be extended to at least the end of the range,
but currently that's not done.
There are at least two cases where this can cause problems:
- copy_file_range() will return short count if the file is not extended
up to end of the source range.
- FALLOC_FL_ZERO_RANGE | FALLOC_FL_KEEP_SIZE will not extend the file,
hence the region may not be fully allocated.
Fix by flushing writes from the start of the range up to the end of the
file. This could be optimized if the writes are non-extending, etc, but
it's probably not worth the trouble.
Fixes: a2bc923629 ("fuse: fix copy_file_range() in the writeback case")
Fixes: 6b1bdb56b1 ("fuse: allow fallocate(FALLOC_FL_ZERO_RANGE)")
Cc: <stable@vger.kernel.org> # v5.2
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
-----BEGIN PGP SIGNATURE-----
iQEzBAABCAAdFiEEq1nRK9aeMoq1VSgcnJ2qBz9kQNkFAmEmTZcACgkQnJ2qBz9k
QNkkmAgArW6XoF1CePds/ZaC9vfg/nk66/zVo0n+J8xXjMWAPxcKbWFfV0uWVixq
yk4lcLV47a2Mu/B/1oLNd3vrSmhwU+srWqNwOFn1nv+lP/6wJqr8oztRHn/0L9Q3
ZSRrukSejbQ6AvTL/WzTNnCjjCc2ne3Kyko6W41aU6uyJuzhSM32wbx7qlV6t54Z
iint9OrB4gM0avLohNafTUq6I+tEGzBMNwpCG/tqCmkcvDcv3rTDVAnPSCTm0Tx2
hdrYDcY/rLxo93pDBaW1rYA/fohR+mIVye6k2TjkPAL6T1x+rxeT5qnc+YijH5yF
sFPDhlD+ZsfOLi8stWXLOJ+8+gLODg==
=pDBR
-----END PGP SIGNATURE-----
Merge tag 'hole_punch_for_v5.15-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs
Pull fs hole punching vs cache filling race fixes from Jan Kara:
"Fix races leading to possible data corruption or stale data exposure
in multiple filesystems when hole punching races with operations such
as readahead.
This is the series I was sending for the last merge window but with
your objection fixed - now filemap_fault() has been modified to take
invalidate_lock only when we need to create new page in the page cache
and / or bring it uptodate"
* tag 'hole_punch_for_v5.15-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs:
filesystems/locking: fix Malformed table warning
cifs: Fix race between hole punch and page fault
ceph: Fix race between hole punch and page fault
fuse: Convert to using invalidate_lock
f2fs: Convert to using invalidate_lock
zonefs: Convert to using invalidate_lock
xfs: Convert double locking of MMAPLOCK to use VFS helpers
xfs: Convert to use invalidate_lock
xfs: Refactor xfs_isilocked()
ext2: Convert to using invalidate_lock
ext4: Convert to use mapping->invalidate_lock
mm: Add functions to lock invalidate_lock for two mappings
mm: Protect operations adding pages to page cache with invalidate_lock
documentation: Sync file_operations members with reality
mm: Fix comments mentioning i_mutex
Add a rcu argument to the ->get_acl() callback to allow
get_cached_acl_rcu() to call the ->get_acl() method in the next patch.
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
fuse_finish_open() will be called with FUSE_NOWRITE in case of atomic
O_TRUNC. This can deadlock with fuse_wait_on_page_writeback() in
fuse_launder_page() triggered by invalidate_inode_pages2().
Fix by replacing invalidate_inode_pages2() in fuse_finish_open() with a
truncate_pagecache() call. This makes sense regardless of FOPEN_KEEP_CACHE
or fc->writeback cache, so do it unconditionally.
Reported-by: Xie Yongji <xieyongji@bytedance.com>
Reported-and-tested-by: syzbot+bea44a5189836d956894@syzkaller.appspotmail.com
Fixes: e4648309b8 ("fuse: truncate pending writes on O_TRUNC")
Cc: <stable@vger.kernel.org>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Make it possible to create a new mount from a already working server.
Here's a detailed description of the problem from Jakob:
"The background for this question is occasional problems we see with our
fuse filesystem [1] and mount namespaces. On a usual client, we have
system-wide, autofs managed mountpoints. When a new mount namespace is
created (which can be done unprivileged in combination with user
namespaces), it can happen that a mountpoint is used inside the new
namespace but idle in the root mount namespace. So autofs unmounts the
parent, system-wide mountpoint. But the fuse module stays active and
still serves mountpoint in the child mount namespace. Because the fuse
daemon also blocks other system wide resources corresponding to the
mountpoint, this situation effectively prevents new mounts until the
child mount namespaces closes.
[1] https://github.com/cvmfs/cvmfs"
Reported-by: Jakob Blomer <jblomer@cern.ch>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Checking whether the "fd=", "rootmode=", "user_id=" and "group_id=" mount
options are present can be moved from fuse_get_tree() into
fuse_fill_super() where the value of the options are consumed.
This relaxes semantics of reusing a fuse blockdev mount using the device
name. Before this patch presence of these options were enforced but values
ignored, after this patch these options are completely ignored in this
case.
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
There is a potential race between fuse_read_interrupt() and
fuse_request_end().
TASK1
in fuse_read_interrupt(): delete req->intr_entry (while holding
fiq->lock)
TASK2
in fuse_request_end(): req->intr_entry is empty -> skip fiq->lock
wake up TASK3
TASK3
request is freed
TASK1
in fuse_read_interrupt(): dereference req->in.h.unique ***BAM***
Fix by always grabbing fiq->lock if the request was ever interrupted
(FR_INTERRUPTED set) thereby serializing with concurrent
fuse_read_interrupt() calls.
FR_INTERRUPTED is set before the request is queued on fiq->interrupts.
Dequeing the request is done with list_del_init() but FR_INTERRUPTED is not
cleared in this case.
Reported-by: lijiazi <lijiazi@xiaomi.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Use invalidate_lock instead of fuse's private i_mmap_sem. The intended
purpose is exactly the same. By this conversion we fix a long standing
race between hole punching and read(2) / readahead(2) paths that can
lead to stale page cache contents.
CC: Miklos Szeredi <miklos@szeredi.hu>
Reviewed-by: Miklos Szeredi <mszeredi@redhat.com>
Signed-off-by: Jan Kara <jack@suse.cz>
fuse_dax_mem_range_init() does not need the address or the pfn of the
memory requested in dax_direct_access(). It is only calling direct
access to get the number of pages.
Remove the unused variables and stop requesting the kaddr and pfn from
dax_direct_access().
Reviewed-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Ira Weiny <ira.weiny@intel.com>
Reviewed-by: Vivek Goyal <vgoyal@redhat.com>
Link: https://lore.kernel.org/r/20210525172428.3634316-2-ira.weiny@intel.com
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
-----BEGIN PGP SIGNATURE-----
iHUEABYIAB0WIQSQHSd0lITzzeNWNm3h3BK/laaZPAUCYORvYQAKCRDh3BK/laaZ
PCfvAQCbU+PW2RbwlqjZMet6w9qorh29XYe786P5pNRVbMYCygD+N45l66Sbd/Rz
7M7ioVDseyTW4dnLhb8SzSNB0zr6jQs=
=MDvD
-----END PGP SIGNATURE-----
Merge tag 'fuse-update-5.14' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/fuse
Pull fuse updates from Miklos Szeredi:
- Fixes for virtiofs submounts
- Misc fixes and cleanups
* tag 'fuse-update-5.14' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/fuse:
virtiofs: Fix spelling mistakes
fuse: use DIV_ROUND_UP helper macro for calculations
fuse: fix illegal access to inode with reused nodeid
fuse: allow fallocate(FALLOC_FL_ZERO_RANGE)
fuse: Make fuse_fill_super_submount() static
fuse: Switch to fc_mount() for submounts
fuse: Call vfs_get_tree() for submounts
fuse: add dedicated filesystem context ops for submounts
virtiofs: propagate sync() to file server
fuse: reject internal errno
fuse: check connected before queueing on fpq->io
fuse: ignore PG_workingset after stealing
fuse: Fix infinite loop in sget_fc()
fuse: Fix crash if superblock of submount gets killed early
fuse: Fix crash in fuse_dentry_automount() error path
Pull iov_iter updates from Al Viro:
"iov_iter cleanups and fixes.
There are followups, but this is what had sat in -next this cycle. IMO
the macro forest in there became much thinner and easier to follow..."
* 'work.iov_iter' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs: (37 commits)
csum_and_copy_to_pipe_iter(): leave handling of csum_state to caller
clean up copy_mc_pipe_to_iter()
pipe_zero(): we don't need no stinkin' kmap_atomic()...
iov_iter: clean csum_and_copy_...() primitives up a bit
copy_page_from_iter(): don't need kmap_atomic() for kvec/bvec cases
copy_page_to_iter(): don't bother with kmap_atomic() for bvec/kvec cases
iterate_xarray(): only of the first iteration we might get offset != 0
pull handling of ->iov_offset into iterate_{iovec,bvec,xarray}
iov_iter: make iterator callbacks use base and len instead of iovec
iov_iter: make the amount already copied available to iterator callbacks
iov_iter: get rid of separate bvec and xarray callbacks
iov_iter: teach iterate_{bvec,xarray}() about possible short copies
iterate_bvec(): expand bvec.h macro forest, massage a bit
iov_iter: unify iterate_iovec and iterate_kvec
iov_iter: massage iterate_iovec and iterate_kvec to logics similar to iterate_bvec
iterate_and_advance(): get rid of magic in case when n is 0
csum_and_copy_to_iter(): massage into form closer to csum_and_copy_from_iter()
iov_iter: replace iov_iter_copy_from_user_atomic() with iterator-advancing variant
[xarray] iov_iter_npages(): just use DIV_ROUND_UP()
iov_iter_npages(): don't bother with iterate_all_kinds()
...
These functions implement the address_space ->set_page_dirty operation and
should live in pagemap.h, not mm.h so that the rest of the kernel doesn't
get funny ideas about calling them directly.
Link: https://lkml.kernel.org/r/20210615162342.1669332-7-willy@infradead.org
Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Jan Kara <jack@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Use __set_page_dirty_no_writeback() instead. This will set the dirty bit
on the page, which will be used to avoid calling set_page_dirty() in the
future. It will have no effect on actually writing the page back, as the
pages are not on any LRU lists.
[akpm@linux-foundation.org: export __set_page_dirty_no_writeback() to modules]
Link: https://lkml.kernel.org/r/20210615162342.1669332-6-willy@infradead.org
Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Jan Kara <jack@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Replace open coded divisor calculations with the DIV_ROUND_UP kernel macro
for better readability.
Signed-off-by: Wu Bo <wubo40@huawei.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Server responds to LOOKUP and other ops (READDIRPLUS/CREATE/MKNOD/...)
with ourarg containing nodeid and generation.
If a fuse inode is found in inode cache with the same nodeid but different
generation, the existing fuse inode should be unhashed and marked "bad" and
a new inode with the new generation should be hashed instead.
This can happen, for example, with passhrough fuse filesystem that returns
the real filesystem ino/generation on lookup and where real inode numbers
can get recycled due to real files being unlinked not via the fuse
passthrough filesystem.
With current code, this situation will not be detected and an old fuse
dentry that used to point to an older generation real inode, can be used to
access a completely new inode, which should be accessed only via the new
dentry.
Note that because the FORGET message carries the nodeid w/o generation, the
server should wait to get FORGET counts for the nlookup counts of the old
and reused inodes combined, before it can free the resources associated to
that nodeid.
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
The current fuse module filters out fallocate(FALLOC_FL_ZERO_RANGE)
returning -EOPNOTSUPP. libnbd's nbdfuse would like to translate
FALLOC_FL_ZERO_RANGE requests into the NBD command
NBD_CMD_WRITE_ZEROES which allows NBD servers that support it to do
zeroing efficiently.
This commit treats this flag exactly like FALLOC_FL_PUNCH_HOLE.
A way to test this, requiring fuse >= 3, nbdkit >= 1.8 and the latest
nbdfuse from https://gitlab.com/nbdkit/libnbd/-/tree/master/fuse is to
create a file containing some data and "mirror" it to a fuse file:
$ dd if=/dev/urandom of=disk.img bs=1M count=1
$ nbdkit file disk.img
$ touch mirror.img
$ nbdfuse mirror.img nbd://localhost &
(mirror.img -> nbdfuse -> NBD over loopback -> nbdkit -> disk.img)
You can then run commands such as:
$ fallocate -z -o 1024 -l 1024 mirror.img
and check that the content of the original file ("disk.img") stays
synchronized. To show NBD commands, export LIBNBD_DEBUG=1 before
running nbdfuse. To clean up:
$ fusermount3 -u mirror.img
$ killall nbdkit
Signed-off-by: Richard W.M. Jones <rjones@redhat.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
This function used to be called from fuse_dentry_automount(). This code
was moved to fuse_get_tree_submount() in the same file since then. It
is unlikely there will ever be another user. No need to be extern in
this case.
Signed-off-by: Greg Kurz <groug@kaod.org>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
fc_mount() already handles the vfs_get_tree(), sb->s_umount
unlocking and vfs_create_mount() sequence. Using it greatly
simplifies fuse_dentry_automount().
Signed-off-by: Greg Kurz <groug@kaod.org>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
We recently fixed an infinite loop by setting the SB_BORN flag on
submounts along with the write barrier needed by super_cache_count().
This is the job of vfs_get_tree() and FUSE shouldn't have to care
about the barrier at all.
Split out some code from fuse_dentry_automount() to the dedicated
fuse_get_tree_submount() handler for submounts and call vfs_get_tree().
Signed-off-by: Greg Kurz <groug@kaod.org>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
The creation of a submount is open-coded in fuse_dentry_automount().
This brings a lot of complexity and we recently had to fix bugs
because we weren't setting SB_BORN or because we were unlocking
sb->s_umount before sb was fully configured. Most of these could
have been avoided by using the mount API instead of open-coding.
Basically, this means coming up with a proper ->get_tree()
implementation for submounts and call vfs_get_tree(), or better
fc_mount().
The creation of the superblock for submounts is quite different from
the root mount. Especially, it doesn't require to allocate a FUSE
filesystem context, nor to parse parameters.
Introduce a dedicated context ops for submounts to make this clear.
This is just a placeholder for now, fuse_get_tree_submount() will
be populated in a subsequent patch.
Only visible change is that we stop allocating/freeing a useless FUSE
filesystem context with submounts.
Signed-off-by: Greg Kurz <groug@kaod.org>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Even if POSIX doesn't mandate it, linux users legitimately expect sync() to
flush all data and metadata to physical storage when it is located on the
same system. This isn't happening with virtiofs though: sync() inside the
guest returns right away even though data still needs to be flushed from
the host page cache.
This is easily demonstrated by doing the following in the guest:
$ dd if=/dev/zero of=/mnt/foo bs=1M count=5K ; strace -T -e sync sync
5120+0 records in
5120+0 records out
5368709120 bytes (5.4 GB, 5.0 GiB) copied, 5.22224 s, 1.0 GB/s
sync() = 0 <0.024068>
and start the following in the host when the 'dd' command completes
in the guest:
$ strace -T -e fsync /usr/bin/sync virtiofs/foo
fsync(3) = 0 <10.371640>
There are no good reasons not to honor the expected behavior of sync()
actually: it gives an unrealistic impression that virtiofs is super fast
and that data has safely landed on HW, which isn't the case obviously.
Implement a ->sync_fs() superblock operation that sends a new FUSE_SYNCFS
request type for this purpose. Provision a 64-bit placeholder for possible
future extensions. Since the file server cannot handle the wait == 0 case,
we skip it to avoid a gratuitous roundtrip. Note that this is
per-superblock: a FUSE_SYNCFS is send for the root mount and for each
submount.
Like with FUSE_FSYNC and FUSE_FSYNCDIR, lack of support for FUSE_SYNCFS in
the file server is treated as permanent success. This ensures
compatibility with older file servers: the client will get the current
behavior of sync() not being propagated to the file server.
Note that such an operation allows the file server to DoS sync(). Since a
typical FUSE file server is an untrusted piece of software running in
userspace, this is disabled by default. Only enable it with virtiofs for
now since virtiofsd is supposedly trusted by the guest kernel.
Reported-by: Robert Krawitz <rlk@redhat.com>
Signed-off-by: Greg Kurz <groug@kaod.org>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
A request could end up on the fpq->io list after fuse_abort_conn() has
reset fpq->connected and aborted requests on that list:
Thread-1 Thread-2
======== ========
->fuse_simple_request() ->shutdown
->__fuse_request_send()
->queue_request() ->fuse_abort_conn()
->fuse_dev_do_read() ->acquire(fpq->lock)
->wait_for(fpq->lock) ->set err to all req's in fpq->io
->release(fpq->lock)
->acquire(fpq->lock)
->add req to fpq->io
After the userspace copy is done the request will be ended, but
req->out.h.error will remain uninitialized. Also the copy might block
despite being already aborted.
Fix both issues by not allowing the request to be queued on the fpq->io
list after fuse_abort_conn() has processed this list.
Reported-by: Pradeep P V K <pragalla@codeaurora.org>
Fixes: fd22d62ed0 ("fuse: no fc->lock for iqueue parts")
Cc: <stable@vger.kernel.org> # v4.2
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Fix the "fuse: trying to steal weird page" warning.
Description from Johannes Weiner:
"Think of it as similar to PG_active. It's just another usage/heat
indicator of file and anon pages on the reclaim LRU that, unlike
PG_active, persists across deactivation and even reclaim (we store it in
the page cache / swapper cache tree until the page refaults).
So if fuse accepts pages that can legally have PG_active set,
PG_workingset is fine too."
Reported-by: Thomas Lindroth <thomas.lindroth@gmail.com>
Fixes: 1899ad18c6 ("mm: workingset: tell cache transitions from workingset thrashing")
Cc: <stable@vger.kernel.org> # v4.20
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Replacement is called copy_page_from_iter_atomic(); unlike the old primitive the
callers do *not* need to do iov_iter_advance() after it. In case when they end
up consuming less than they'd been given they need to do iov_iter_revert() on
everything they had not consumed. That, however, needs to be done only on slow
paths.
All in-tree callers converted. And that kills the last user of iterate_all_kinds()
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
We don't set the SB_BORN flag on submounts. This is wrong as these
superblocks are then considered as partially constructed or dying
in the rest of the code and can break some assumptions.
One such case is when you have a virtiofs filesystem with submounts
and you try to mount it again : virtio_fs_get_tree() tries to obtain
a superblock with sget_fc(). The logic in sget_fc() is to loop until
it has either found an existing matching superblock with SB_BORN set
or to create a brand new one. It is assumed that a superblock without
SB_BORN is transient and the loop is restarted. Forgetting to set
SB_BORN on submounts hence causes sget_fc() to retry forever.
Setting SB_BORN requires special care, i.e. a write barrier for
super_cache_count() which can check SB_BORN without taking any lock.
We should call vfs_get_tree() to deal with that but this requires
to have a proper ->get_tree() implementation for submounts, which
is a bigger piece of work. Go for a simple bug fix in the meatime.
Fixes: bf109c6404 ("fuse: implement crossmounts")
Cc: stable@vger.kernel.org # v5.10+
Signed-off-by: Greg Kurz <groug@kaod.org>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
As soon as fuse_dentry_automount() does up_write(&sb->s_umount), the
superblock can theoretically be killed. If this happens before the
submount was added to the &fc->mounts list, fuse_mount_remove() later
crashes in list_del_init() because it assumes the submount to be
already there.
Add the submount before dropping sb->s_umount to fix the inconsistency.
It is okay to nest fc->killsb under sb->s_umount, we already do this
on the ->kill_sb() path.
Signed-off-by: Greg Kurz <groug@kaod.org>
Fixes: bf109c6404 ("fuse: implement crossmounts")
Cc: stable@vger.kernel.org # v5.10+
Reviewed-by: Max Reitz <mreitz@redhat.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
If fuse_fill_super_submount() returns an error, the error path
triggers a crash:
[ 26.206673] BUG: kernel NULL pointer dereference, address: 0000000000000000
[...]
[ 26.226362] RIP: 0010:__list_del_entry_valid+0x25/0x90
[...]
[ 26.247938] Call Trace:
[ 26.248300] fuse_mount_remove+0x2c/0x70 [fuse]
[ 26.248892] virtio_kill_sb+0x22/0x160 [virtiofs]
[ 26.249487] deactivate_locked_super+0x36/0xa0
[ 26.250077] fuse_dentry_automount+0x178/0x1a0 [fuse]
The crash happens because fuse_mount_remove() assumes that the FUSE
mount was already added to list under the FUSE connection, but this
only done after fuse_fill_super_submount() has returned success.
This means that until fuse_fill_super_submount() has returned success,
the FUSE mount isn't actually owned by the superblock. We should thus
reclaim ownership by clearing sb->s_fs_info, which will skip the call
to fuse_mount_remove(), and perform rollback, like virtio_fs_get_tree()
already does for the root sb.
Fixes: bf109c6404 ("fuse: implement crossmounts")
Cc: stable@vger.kernel.org # v5.10+
Signed-off-by: Greg Kurz <groug@kaod.org>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
another rudiment of fault-in originally having been limited to the
first segment, same as in generic_perform_write() and friends.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Pull misc vfs updates from Al Viro:
"Assorted stuff all over the place"
* 'work.misc' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
useful constants: struct qstr for ".."
hostfs_open(): don't open-code file_dentry()
whack-a-mole: kill strlen_user() (again)
autofs: should_expire() argument is guaranteed to be positive
apparmor:match_mn() - constify devpath argument
buffer: a small optimization in grow_buffers
get rid of autofs_getpath()
constify dentry argument of dentry_path()/dentry_path_raw()
Pull fileattr conversion updates from Miklos Szeredi via Al Viro:
"This splits the handling of FS_IOC_[GS]ETFLAGS from ->ioctl() into a
separate method.
The interface is reasonably uniform across the filesystems that
support it and gives nice boilerplate removal"
* 'miklos.fileattr' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs: (23 commits)
ovl: remove unneeded ioctls
fuse: convert to fileattr
fuse: add internal open/release helpers
fuse: unsigned open flags
fuse: move ioctl to separate source file
vfs: remove unused ioctl helpers
ubifs: convert to fileattr
reiserfs: convert to fileattr
ocfs2: convert to fileattr
nilfs2: convert to fileattr
jfs: convert to fileattr
hfsplus: convert to fileattr
efivars: convert to fileattr
xfs: convert to fileattr
orangefs: convert to fileattr
gfs2: convert to fileattr
f2fs: convert to fileattr
ext4: convert to fileattr
ext2: convert to fileattr
btrfs: convert to fileattr
...
Pull vfs inode type handling updates from Al Viro:
"We should never change the type bits of ->i_mode or the method tables
(->i_op and ->i_fop) of a live inode.
Unfortunately, not all filesystems took care to prevent that"
* 'work.inode-type-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
spufs: fix bogosity in S_ISGID handling
9p: missing chunk of "fs/9p: Don't update file type when updating file attributes"
openpromfs: don't do unlock_new_inode() until the new inode is set up
hostfs_mknod(): don't bother with init_special_inode()
cifs: have cifs_fattr_to_inode() refuse to change type on live inode
cifs: have ->mkdir() handle race with another client sanely
do_cifs_create(): don't set ->i_mode of something we had not created
gfs2: be careful with inode refresh
ocfs2_inode_lock_update(): make sure we don't change the type bits of i_mode
orangefs_inode_is_stale(): i_mode type bits do *not* form a bitmap...
vboxsf: don't allow to change the inode type
afs: Fix updating of i_mode due to 3rd party change
ceph: don't allow type or device number to change on non-I_NEW inodes
ceph: fix up error handling with snapdirs
new helper: inode_wrong_type()
For cloned connections cuse_channel_release() will be called more than
once, resulting in use after free.
Prevent device cloning for CUSE, which does not make sense at this point,
and highly unlikely to be used in real life.
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
get_user_ns() is done twice (once in virtio_fs_get_tree() and once in
fuse_conn_init()), resulting in a reference leak.
Also looks better to use fsc->user_ns (which *should* be the
current_user_ns() at this point).
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
If an incoming FUSE request can't fit on the virtqueue, the request is
placed onto a workqueue so a worker can try to resubmit it later where
there will (hopefully) be space for it next time.
This is fine for requests that aren't larger than a virtqueue's maximum
capacity. However, if a request's size exceeds the maximum capacity of the
virtqueue (even if the virtqueue is empty), it will be doomed to a life of
being placed on the workqueue, removed, discovered it won't fit, and placed
on the workqueue yet again.
Furthermore, from section 2.6.5.3.1 (Driver Requirements: Indirect
Descriptors) of the virtio spec:
"A driver MUST NOT create a descriptor chain longer than the Queue
Size of the device."
To fix this, limit the number of pages FUSE will use for an overall
request. This way, each request can realistically fit on the virtqueue
when it is decomposed into a scattergather list and avoid violating section
2.6.5.3.1 of the virtio spec.
Signed-off-by: Connor Kuehl <ckuehl@redhat.com>
Reviewed-by: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
In fuse when a direct/write-through write happens we invalidate attrs
because that might have updated mtime/ctime on server and cached
mtime/ctime will be stale.
What about page writeback path. Looks like we don't invalidate attrs
there. To be consistent, invalidate attrs in writeback path as well. Only
exception is when writeback_cache is enabled. In that case we strust local
mtime/ctime and there is no need to invalidate attrs.
Recently users started experiencing failure of xfstests generic/080,
geneirc/215 and generic/614 on virtiofs. This happened only newer "stat"
utility and not older one. This patch fixes the issue.
So what's the root cause of the issue. Here is detailed explanation.
generic/080 test does mmap write to a file, closes the file and then checks
if mtime has been updated or not. When file is closed, it leads to
flushing of dirty pages (and that should update mtime/ctime on server).
But we did not explicitly invalidate attrs after writeback finished. Still
generic/080 passed so far and reason being that we invalidated atime in
fuse_readpages_end(). This is called in fuse_readahead() path and always
seems to trigger before mmaped write.
So after mmaped write when lstat() is called, it sees that atleast one of
the fields being asked for is invalid (atime) and that results in
generating GETATTR to server and mtime/ctime also get updated and test
passes.
But newer /usr/bin/stat seems to have moved to using statx() syscall now
(instead of using lstat()). And statx() allows it to query only ctime or
mtime (and not rest of the basic stat fields). That means when querying
for mtime, fuse_update_get_attr() sees that mtime is not invalid (only
atime is invalid). So it does not generate a new GETATTR and fill stat
with cached mtime/ctime. And that means updated mtime is not seen by
xfstest and tests start failing.
Invalidating attrs after writeback completion should solve this problem in
a generic manner.
Signed-off-by: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
When posix access ACL is set, it can have an effect on file mode and it can
also need to clear SGID if.
- None of caller's group/supplementary groups match file owner group.
AND
- Caller is not priviliged (No CAP_FSETID).
As of now fuser server is responsible for changing the file mode as
well. But it does not know whether to clear SGID or not.
So add a flag FUSE_SETXATTR_ACL_KILL_SGID and send this info with SETXATTR
to let file server know that sgid needs to be cleared as well.
Reported-by: Luis Henriques <lhenriques@suse.de>
Signed-off-by: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Fuse client needs to send additional information to file server when it
calls SETXATTR(system.posix_acl_access), so add extra flags field to the
structure.
Signed-off-by: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
With commit f8425c9396 ("fuse: 32-bit user space ioctl compat for fuse
device") the matching constraints for the FUSE_DEV_IOC_CLONE ioctl command
are relaxed, limited to the testing of command type and number. As Arnd
noticed, this is wrong as it wouldn't ensure the correctness of the data
size or direction for the received FUSE device ioctl.
Fix by bringing back the comparison of the ioctl received by the FUSE
device to the originally generated FUSE_DEV_IOC_CLONE.
Fixes: f8425c9396 ("fuse: 32-bit user space ioctl compat for fuse device")
Reported-by: Arnd Bergmann <arnd@kernel.org>
Signed-off-by: Alessio Balsini <balsini@android.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
All callers of fuse_short_read already set the .page_zeroing flag, so no
need to do the tail zeroing again.
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
There are two modes for write(2) and friends in fuse:
a) write through (update page cache, send sync WRITE request to userspace)
b) buffered write (update page cache, async writeout later)
The write through method kept all the page cache pages locked that were
used for the request. Keeping more than one page locked is deadlock prone
and Qian Cai demonstrated this with trinity fuzzing.
The reason for keeping the pages locked is that concurrent mapped reads
shouldn't try to pull possibly stale data into the page cache.
For full page writes, the easy way to fix this is to make the cached page
be the authoritative source by marking the page PG_uptodate immediately.
After this the page can be safely unlocked, since mapped/cached reads will
take the written data from the cache.
Concurrent mapped writes will now cause data in the original WRITE request
to be updated; this however doesn't cause any data inconsistency and this
scenario should be exceedingly rare anyway.
If the WRITE request returns with an error in the above case, currently the
page is not marked uptodate; this means that a concurrent read will always
read consistent data. After this patch the page is uptodate between
writing to the cache and receiving the error: there's window where a cached
read will read the wrong data. While theoretically this could be a
regression, it is unlikely to be one in practice, since this is normal for
buffered writes.
In case of a partial page write to an already uptodate page the locking is
also unnecessary, with the above caveats.
Partial write of a not uptodate page still needs to be handled. One way
would be to read the complete page before doing the write. This is not
possible, since it might break filesystems that don't expect any READ
requests when the file was opened O_WRONLY.
The other solution is to serialize the synchronous write with reads from
the partial pages. The easiest way to do this is to keep the partial pages
locked. The problem is that a write() may involve two such pages (one head
and one tail). This patch fixes it by only locking the partial tail page.
If there's a partial head page as well, then split that off as a separate
WRITE request.
Reported-by: Qian Cai <cai@lca.pw>
Link: https://lore.kernel.org/linux-fsdevel/4794a3fa3742a5e84fb0f934944204b55730829b.camel@lca.pw/
Fixes: ea9b9907b8 ("fuse: implement perform_write")
Cc: <stable@vger.kernel.org> # v2.6.26
Signed-off-by: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Since fuse just passes ioctl args through to/from server, converting to the
fileattr API is more involved, than most other filesystems.
Both .fileattr_set() and .fileattr_get() need to obtain an open file to
operate on. The simplest way is with the following sequence:
FUSE_OPEN
FUSE_IOCTL
FUSE_RELEASE
If this turns out to be a performance problem, it could be optimized for
the case when there's already a file (any file) open for the inode.
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
With a 64-bit kernel build the FUSE device cannot handle ioctl requests
coming from 32-bit user space. This is due to the ioctl command
translation that generates different command identifiers that thus cannot
be used for direct comparisons without proper manipulation.
Explicitly extract type and number from the ioctl command to enable 32-bit
user space compatibility on 64-bit kernel builds.
Signed-off-by: Alessio Balsini <balsini@android.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
inode_wrong_type(inode, mode) returns true if setting inode->i_mode
to given value would've changed the inode type. We have enough of
those checks open-coded to make a helper worthwhile.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Right now "mount -t virtiofs -o dax myfs /mnt/virtiofs" succeeds even
if filesystem deivce does not have a cache window and hence DAX can't
be supported.
This gives a false sense to user that they are using DAX with virtiofs
but fact of the matter is that they are not.
Fix this by returning error if dax can't be supported and user has asked
for it.
Signed-off-by: Vivek Goyal <vgoyal@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Since commit 74d609585d ("page cache: Add and replace pages using the
XArray") was merged, the replace_page_cache_page() can not fail and always
return 0, we can remove the redundant return value and void it. Moreover
remove the unused gfp_mask.
Link: https://lkml.kernel.org/r/609c30e5274ba15d8b90c872fd0d8ac437a9b2bb.1610071401.git.baolin.wang@linux.alibaba.com
Signed-off-by: Baolin Wang <baolin.wang@linux.alibaba.com>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: Miklos Szeredi <miklos@szeredi.hu>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Extend some inode methods with an additional user namespace argument. A
filesystem that is aware of idmapped mounts will receive the user
namespace the mount has been marked with. This can be used for
additional permission checking and also to enable filesystems to
translate between uids and gids if they need to. We have implemented all
relevant helpers in earlier patches.
As requested we simply extend the exisiting inode method instead of
introducing new ones. This is a little more code churn but it's mostly
mechanical and doesnt't leave us with additional inode methods.
Link: https://lore.kernel.org/r/20210121131959.646623-25-christian.brauner@ubuntu.com
Cc: Christoph Hellwig <hch@lst.de>
Cc: David Howells <dhowells@redhat.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: linux-fsdevel@vger.kernel.org
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com>
The generic_fillattr() helper fills in the basic attributes associated
with an inode. Enable it to handle idmapped mounts. If the inode is
accessed through an idmapped mount map it into the mount's user
namespace before we store the uid and gid. If the initial user namespace
is passed nothing changes so non-idmapped mounts will see identical
behavior as before.
Link: https://lore.kernel.org/r/20210121131959.646623-12-christian.brauner@ubuntu.com
Cc: Christoph Hellwig <hch@lst.de>
Cc: David Howells <dhowells@redhat.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: linux-fsdevel@vger.kernel.org
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: James Morris <jamorris@linux.microsoft.com>
Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com>
The posix acl permission checking helpers determine whether a caller is
privileged over an inode according to the acls associated with the
inode. Add helpers that make it possible to handle acls on idmapped
mounts.
The vfs and the filesystems targeted by this first iteration make use of
posix_acl_fix_xattr_from_user() and posix_acl_fix_xattr_to_user() to
translate basic posix access and default permissions such as the
ACL_USER and ACL_GROUP type according to the initial user namespace (or
the superblock's user namespace) to and from the caller's current user
namespace. Adapt these two helpers to handle idmapped mounts whereby we
either map from or into the mount's user namespace depending on in which
direction we're translating.
Similarly, cap_convert_nscap() is used by the vfs to translate user
namespace and non-user namespace aware filesystem capabilities from the
superblock's user namespace to the caller's user namespace. Enable it to
handle idmapped mounts by accounting for the mount's user namespace.
In addition the fileystems targeted in the first iteration of this patch
series make use of the posix_acl_chmod() and, posix_acl_update_mode()
helpers. Both helpers perform permission checks on the target inode. Let
them handle idmapped mounts. These two helpers are called when posix
acls are set by the respective filesystems to handle this case we extend
the ->set() method to take an additional user namespace argument to pass
the mount's user namespace down.
Link: https://lore.kernel.org/r/20210121131959.646623-9-christian.brauner@ubuntu.com
Cc: Christoph Hellwig <hch@lst.de>
Cc: David Howells <dhowells@redhat.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: linux-fsdevel@vger.kernel.org
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com>
When file attributes are changed most filesystems rely on the
setattr_prepare(), setattr_copy(), and notify_change() helpers for
initialization and permission checking. Let them handle idmapped mounts.
If the inode is accessed through an idmapped mount map it into the
mount's user namespace. Afterwards the checks are identical to
non-idmapped mounts. If the initial user namespace is passed nothing
changes so non-idmapped mounts will see identical behavior as before.
Helpers that perform checks on the ia_uid and ia_gid fields in struct
iattr assume that ia_uid and ia_gid are intended values and have already
been mapped correctly at the userspace-kernelspace boundary as we
already do today. If the initial user namespace is passed nothing
changes so non-idmapped mounts will see identical behavior as before.
Link: https://lore.kernel.org/r/20210121131959.646623-8-christian.brauner@ubuntu.com
Cc: Christoph Hellwig <hch@lst.de>
Cc: David Howells <dhowells@redhat.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: linux-fsdevel@vger.kernel.org
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com>
The two helpers inode_permission() and generic_permission() are used by
the vfs to perform basic permission checking by verifying that the
caller is privileged over an inode. In order to handle idmapped mounts
we extend the two helpers with an additional user namespace argument.
On idmapped mounts the two helpers will make sure to map the inode
according to the mount's user namespace and then peform identical
permission checks to inode_permission() and generic_permission(). If the
initial user namespace is passed nothing changes so non-idmapped mounts
will see identical behavior as before.
Link: https://lore.kernel.org/r/20210121131959.646623-6-christian.brauner@ubuntu.com
Cc: Christoph Hellwig <hch@lst.de>
Cc: David Howells <dhowells@redhat.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: linux-fsdevel@vger.kernel.org
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: James Morris <jamorris@linux.microsoft.com>
Acked-by: Serge Hallyn <serge@hallyn.com>
Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com>
Jan Kara's analysis of the syzbot report (edited):
The reproducer opens a directory on FUSE filesystem, it then attaches
dnotify mark to the open directory. After that a fuse_do_getattr() call
finds that attributes returned by the server are inconsistent, and calls
make_bad_inode() which, among other things does:
inode->i_mode = S_IFREG;
This then confuses dnotify which doesn't tear down its structures
properly and eventually crashes.
Avoid calling make_bad_inode() on a live inode: switch to a private flag on
the fuse inode. Also add the test to ops which the bad_inode_ops would
have caught.
This bug goes back to the initial merge of fuse in 2.6.14...
Reported-by: syzbot+f427adf9324b92652ccc@syzkaller.appspotmail.com
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Tested-by: Jan Kara <jack@suse.cz>
Cc: <stable@vger.kernel.org>
Virtiofs can be slow with small writes if xattr are enabled and we are
doing cached writes (No direct I/O). Ganesh Mahalingam noticed this.
Some debugging showed that file_remove_privs() is called in cached write
path on every write. And everytime it calls security_inode_need_killpriv()
which results in call to __vfs_getxattr(XATTR_NAME_CAPS). And this goes to
file server to fetch xattr. This extra round trip for every write slows
down writes tremendously.
Normally to avoid paying this penalty on every write, vfs has the notion of
caching this information in inode (S_NOSEC). So vfs sets S_NOSEC, if
filesystem opted for it using super block flag SB_NOSEC. And S_NOSEC is
cleared when setuid/setgid bit is set or when security xattr is set on
inode so that next time a write happens, we check inode again for clearing
setuid/setgid bits as well clear any security.capability xattr.
This seems to work well for local file systems but for remote file systems
it is possible that VFS does not have full picture and a different client
sets setuid/setgid bit or security.capability xattr on file and that means
VFS information about S_NOSEC on another client will be stale. So for
remote filesystems SB_NOSEC was disabled by default.
Commit 9e1f1de02c ("more conservative S_NOSEC handling") mentioned that
these filesystems can still make use of SB_NOSEC as long as they clear
S_NOSEC when they are refreshing inode attriutes from server.
So this patch tries to enable SB_NOSEC on fuse (regular fuse as well as
virtiofs). And clear SB_NOSEC when we are refreshing inode attributes.
This is enabled only if server supports FUSE_HANDLE_KILLPRIV_V2. This says
that server will clear setuid/setgid/security.capability on
chown/truncate/write as apporpriate.
This should provide tighter coherency because now suid/sgid/
security.capability will be cleared even if fuse client cache has not seen
these attrs.
Basic idea is that fuse client will trigger suid/sgid/security.capability
clearing based on its attr cache. But even if cache has gone stale, it is
fine because FUSE_HANDLE_KILLPRIV_V2 will make sure WRITE clear
suid/sgid/security.capability.
We make this change only if server supports FUSE_HANDLE_KILLPRIV_V2. This
should make sure that existing filesystems which might be relying on
seucurity.capability always being queried from server are not impacted.
This tighter coherency relies on WRITE showing up on server (and not being
cached in guest). So writeback_cache mode will not provide that tight
coherency and it is not recommended to use two together. Having said that
it might work reasonably well for lot of use cases.
This change improves random write performance very significantly. Running
virtiofsd with cache=auto and following fio command:
fio --ioengine=libaio --direct=1 --name=test --filename=/mnt/virtiofs/random_read_write.fio --bs=4k --iodepth=64 --size=4G --readwrite=randwrite
Bandwidth increases from around 50MB/s to around 250MB/s as a result of
applying this patch. So improvement is very significant.
Link: https://github.com/kata-containers/runtime/issues/2815
Reported-by: "Mahalingam, Ganesh" <ganesh.mahalingam@intel.com>
Signed-off-by: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
With FUSE_HANDLE_KILLPRIV_V2 support, server will need to kill suid/sgid/
security.capability on open(O_TRUNC), if server supports
FUSE_ATOMIC_O_TRUNC.
But server needs to kill suid/sgid only if caller does not have CAP_FSETID.
Given server does not have this information, client needs to send this info
to server.
So add a flag FUSE_OPEN_KILL_SUIDGID to fuse_open_in request which tells
server to kill suid/sgid (only if group execute is set).
This flag is added to the FUSE_OPEN request, as well as the FUSE_CREATE
request if the create was non-exclusive, since that might result in an
existing file being opened/truncated.
Signed-off-by: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
If client does a write() on a suid/sgid file, VFS will first call
fuse_setattr() with ATTR_KILL_S[UG]ID set. This requires sending setattr
to file server with ATTR_MODE set to kill suid/sgid. But to do that client
needs to know latest mode otherwise it is racy.
To reduce the race window, current code first call fuse_do_getattr() to get
latest ->i_mode and then resets suid/sgid bits and sends rest to server
with setattr(ATTR_MODE). This does not reduce the race completely but
narrows race window significantly.
With fc->handle_killpriv_v2 enabled, it should be possible to remove this
race completely. Do not kill suid/sgid with ATTR_MODE at all. It will be
killed by server when WRITE request is sent to server soon. This is
similar to fc->handle_killpriv logic. V2 is just more refined version of
protocol. Hence this patch does not send ATTR_MODE to kill suid/sgid if
fc->handle_killpriv_v2 is enabled.
This creates an issue if fc->writeback_cache is enabled. In that case
WRITE can be cached in guest and server might not see WRITE request and
hence will not kill suid/sgid. Miklos suggested that in such cases, we
should fallback to a writethrough WRITE instead and that will generate
WRITE request and kill suid/sgid. This patch implements that too.
But this relies on client seeing the suid/sgid set. If another client sets
suid/sgid and this client does not see it immideately, then we will not
fallback to writethrough WRITE. So this is one limitation with both
fc->handle_killpriv_v2 and fc->writeback_cache enabled. Both the options
are not fully compatible. But might be good enough for many use cases.
Note: This patch is not checking whether security.capability is set or not
when falling back to writethrough path. If suid/sgid is not set and
only security.capability is set, that will be taken care of by
file_remove_privs() call in ->writeback_cache path.
Signed-off-by: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
If fc->handle_killpriv_v2 is enabled, we expect file server to clear
suid/sgid/security.capbility upon chown/truncate/write as appropriate.
Upon truncate (ATTR_SIZE), suid/sgid are cleared only if caller does not
have CAP_FSETID. File server does not know whether caller has CAP_FSETID
or not. Hence set FATTR_KILL_SUIDGID upon truncate to let file server know
that caller does not have CAP_FSETID and it should kill suid/sgid as
appropriate.
On chown (ATTR_UID/ATTR_GID) suid/sgid need to be cleared irrespective of
capabilities of calling process, so set FATTR_KILL_SUIDGID unconditionally
in that case.
Signed-off-by: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
With HANDLE_KILLPRIV_V2, server will need to kill suid/sgid if caller does
not have CAP_FSETID. We already have a flag FUSE_WRITE_KILL_SUIDGID in
WRITE request and we already set it in direct I/O path.
To make it work in cached write path also, start setting
FUSE_WRITE_KILL_SUIDGID in this path too.
Set it only if fc->handle_killpriv_v2 is set. Otherwise client is
responsible for kill suid/sgid.
In case of direct I/O we set FUSE_WRITE_KILL_SUIDGID unconditionally
because we don't call file_remove_privs() in that path (with cache=none
option).
Signed-off-by: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Kernel has:
ATTR_KILL_PRIV -> clear "security.capability"
ATTR_KILL_SUID -> clear S_ISUID
ATTR_KILL_SGID -> clear S_ISGID if executable
Fuse has:
FUSE_WRITE_KILL_PRIV -> clear S_ISUID and S_ISGID if executable
So FUSE_WRITE_KILL_PRIV implies the complement of ATTR_KILL_PRIV, which is
somewhat confusing. Also PRIV implies all privileges, including
"security.capability".
Change the name to FUSE_WRITE_KILL_SUIDGID and make FUSE_WRITE_KILL_PRIV an
alias to perserve API compatibility
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
We already have FUSE_HANDLE_KILLPRIV flag that says that file server will
remove suid/sgid/caps on truncate/chown/write. But that's little different
from what Linux VFS implements.
To be consistent with Linux VFS behavior what we want is.
- caps are always cleared on chown/write/truncate
- suid is always cleared on chown, while for truncate/write it is cleared
only if caller does not have CAP_FSETID.
- sgid is always cleared on chown, while for truncate/write it is cleared
only if caller does not have CAP_FSETID as well as file has group execute
permission.
As previous flag did not provide above semantics. Implement a V2 of the
protocol with above said constraints.
Server does not know if caller has CAP_FSETID or not. So for the case
of write()/truncate(), client will send information in special flag to
indicate whether to kill priviliges or not. These changes are in subsequent
patches.
FUSE_HANDLE_KILLPRIV_V2 relies on WRITE being sent to server to clear
suid/sgid/security.capability. But with ->writeback_cache, WRITES are
cached in guest. So it is not recommended to use FUSE_HANDLE_KILLPRIV_V2
and writeback_cache together. Though it probably might be good enough
for lot of use cases.
Signed-off-by: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Failure to do so may result in EEXIST even if the file only exists in the
cache and not in the filesystem.
The atomic nature of O_EXCL mandates that the cached state should be
ignored and existence verified anew.
Reported-by: Ken Schalk <kschalk@nvidia.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Fuse mount now only ever has a refcount of one (before being freed) so the
count field is unnecessary.
Remove the refcounting and fold fuse_mount_put() into callers. The only
caller of fuse_mount_put() where fm->fc was NULL is fuse_dentry_automount()
and here the fuse_conn_put() can simply be omitted.
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Currently when acquiring an sb for virtiofs fuse_mount_get() is being
called from virtio_fs_set_super() if a new sb is being filled and
fuse_mount_put() is called unconditionally after sget_fc() returns.
The exact same result can be obtained by checking whether
fs_contex->s_fs_info was set to NULL (ref trasferred to sb->s_fs_info) and
only calling fuse_mount_put() if the ref wasn't transferred (error or
matching sb found).
This allows getting rid of virtio_fs_set_super() and fuse_mount_get().
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
This can be triggered for example by adding the "-omand" mount option,
which will be rejected and virtio_fs_fill_super() will return an error.
In such a case the allocations for fuse_conn and fuse_mount will leak due
to s_root not yet being set and so ->put_super() not being called.
Fixes: a62a8ef9d9 ("virtio-fs: add virtiofs filesystem")
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Qian Cai reports that the WARNING in tree_insert() can be triggered by a
fuzzer with the following call chain:
invalidate_inode_pages2_range()
fuse_launder_page()
fuse_writepage_locked()
tree_insert()
The reason is that another write for the same page is already queued.
The simplest fix is to wait until the pending write is completed and only
after that queue the new write.
Since this case is very rare, the additional wait should not be a problem.
Reported-by: Qian Cai <cai@redhat.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
-----BEGIN PGP SIGNATURE-----
iHUEABYIAB0WIQSQHSd0lITzzeNWNm3h3BK/laaZPAUCX4n0/gAKCRDh3BK/laaZ
PM3jAP4xhaix0j/y3VyaxsUqWg6ZSrjq6X0o9clGMJv27IAtjgD/fJ7ZwzTldojD
qb7N3utjLiPVRjwFmvsZ8JZ7O7PbwQ0=
=oUbZ
-----END PGP SIGNATURE-----
Merge tag 'fuse-update-5.10' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/fuse
Pull fuse updates from Miklos Szeredi:
- Support directly accessing host page cache from virtiofs. This can
improve I/O performance for various workloads, as well as reducing
the memory requirement by eliminating double caching. Thanks to Vivek
Goyal for doing most of the work on this.
- Allow automatic submounting inside virtiofs. This allows unique
st_dev/ st_ino values to be assigned inside the guest to files
residing on different filesystems on the host. Thanks to Max Reitz
for the patches.
- Fix an old use after free bug found by Pradeep P V K.
* tag 'fuse-update-5.10' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/fuse: (25 commits)
virtiofs: calculate number of scatter-gather elements accurately
fuse: connection remove fix
fuse: implement crossmounts
fuse: Allow fuse_fill_super_common() for submounts
fuse: split fuse_mount off of fuse_conn
fuse: drop fuse_conn parameter where possible
fuse: store fuse_conn in fuse_req
fuse: add submount support to <uapi/linux/fuse.h>
fuse: fix page dereference after free
virtiofs: add logic to free up a memory range
virtiofs: maintain a list of busy elements
virtiofs: serialize truncate/punch_hole and dax fault path
virtiofs: define dax address space operations
virtiofs: add DAX mmap support
virtiofs: implement dax read/write operations
virtiofs: introduce setupmapping/removemapping commands
virtiofs: implement FUSE_INIT map_alignment field
virtiofs: keep a list of free dax memory ranges
virtiofs: add a mount option to enable dax
virtiofs: set up virtio_fs dax_device
...
virtiofs currently maps various buffers in scatter gather list and it looks
at number of pages (ap->pages) and assumes that same number of pages will
be used both for input and output (sg_count_fuse_req()), and calculates
total number of scatterlist elements accordingly.
But looks like this assumption is not valid in all the cases. For example,
Cai Qian reported that trinity, triggers warning with virtiofs sometimes.
A closer look revealed that if one calls ioctl(fd, 0x5a004000, buf), it
will trigger following warning.
WARN_ON(out_sgs + in_sgs != total_sgs)
In this case, total_sgs = 8, out_sgs=4, in_sgs=3. Number of pages is 2
(ap->pages), but out_sgs are using both the pages but in_sgs are using
only one page. In this case, fuse_do_ioctl() sets different size values
for input and output.
args->in_args[args->in_numargs - 1].size == 6656
args->out_args[args->out_numargs - 1].size == 4096
So current method of calculating how many scatter-gather list elements
will be used is not accurate. Make calculations more precise by parsing
size and ap->descs.
Reported-by: Qian Cai <cai@redhat.com>
Signed-off-by: Vivek Goyal <vgoyal@redhat.com>
Link: https://lore.kernel.org/linux-fsdevel/5ea77e9f6cb8c2db43b09fbd4158ab2d8c066a0a.camel@redhat.com/
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
-----BEGIN PGP SIGNATURE-----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=LzyE
-----END PGP SIGNATURE-----
Merge tag 'block-5.10-2020-10-12' of git://git.kernel.dk/linux-block
Pull block updates from Jens Axboe:
- Series of merge handling cleanups (Baolin, Christoph)
- Series of blk-throttle fixes and cleanups (Baolin)
- Series cleaning up BDI, seperating the block device from the
backing_dev_info (Christoph)
- Removal of bdget() as a generic API (Christoph)
- Removal of blkdev_get() as a generic API (Christoph)
- Cleanup of is-partition checks (Christoph)
- Series reworking disk revalidation (Christoph)
- Series cleaning up bio flags (Christoph)
- bio crypt fixes (Eric)
- IO stats inflight tweak (Gabriel)
- blk-mq tags fixes (Hannes)
- Buffer invalidation fixes (Jan)
- Allow soft limits for zone append (Johannes)
- Shared tag set improvements (John, Kashyap)
- Allow IOPRIO_CLASS_RT for CAP_SYS_NICE (Khazhismel)
- DM no-wait support (Mike, Konstantin)
- Request allocation improvements (Ming)
- Allow md/dm/bcache to use IO stat helpers (Song)
- Series improving blk-iocost (Tejun)
- Various cleanups (Geert, Damien, Danny, Julia, Tetsuo, Tian, Wang,
Xianting, Yang, Yufen, yangerkun)
* tag 'block-5.10-2020-10-12' of git://git.kernel.dk/linux-block: (191 commits)
block: fix uapi blkzoned.h comments
blk-mq: move cancel of hctx->run_work to the front of blk_exit_queue
blk-mq: get rid of the dead flush handle code path
block: get rid of unnecessary local variable
block: fix comment and add lockdep assert
blk-mq: use helper function to test hw stopped
block: use helper function to test queue register
block: remove redundant mq check
block: invoke blk_mq_exit_sched no matter whether have .exit_sched
percpu_ref: don't refer to ref->data if it isn't allocated
block: ratelimit handle_bad_sector() message
blk-throttle: Re-use the throtl_set_slice_end()
blk-throttle: Open code __throtl_de/enqueue_tg()
blk-throttle: Move service tree validation out of the throtl_rb_first()
blk-throttle: Move the list operation after list validation
blk-throttle: Fix IO hang for a corner case
blk-throttle: Avoid tracking latency if low limit is invalid
blk-throttle: Avoid getting the current time if tg->last_finish_time is 0
blk-throttle: Remove a meaningless parameter for throtl_downgrade_state()
block: Remove redundant 'return' statement
...
Re-add lost removal of fc from fuse_conn_list and the control filesystem.
Reported-by: kernel test robot <rong.a.chen@intel.com>
Fixes: fcee216beb ("fuse: split fuse_mount off of fuse_conn")
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
FUSE servers can indicate crossmount points by setting FUSE_ATTR_SUBMOUNT
in fuse_attr.flags. The inode will then be marked as S_AUTOMOUNT, and the
.d_automount implementation creates a new submount at that location, so
that the submount gets a distinct st_dev value.
Note that all submounts get a distinct superblock and a distinct st_dev
value, so for virtio-fs, even if the same filesystem is mounted more than
once on the host, none of its mount points will have the same st_dev. We
need distinct superblocks because the superblock points to the root node,
but the different host mounts may show different trees (e.g. due to
submounts in some of them, but not in others).
Right now, this behavior is only enabled when fuse_conn.auto_submounts is
set, which is the case only for virtio-fs.
Signed-off-by: Max Reitz <mreitz@redhat.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Replace BDI_CAP_NO_ACCT_WB with a positive BDI_CAP_WRITEBACK_ACCT to
make the checks more obvious. Also remove the pointless
bdi_cap_account_writeback wrapper that just obsfucates the check.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Jan Kara <jack@suse.cz>
Reviewed-by: Johannes Thumshirn <johannes.thumshirn@wdc.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Set up a readahead size by default, as very few users have a good
reason to change it. This means code, ecryptfs, and orangefs now
set up the values while they were previously missing it, while ubifs,
mtd and vboxsf manually set it to 0 to avoid readahead.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Jan Kara <jack@suse.cz>
Acked-by: David Sterba <dsterba@suse.com> [btrfs]
Acked-by: Richard Weinberger <richard@nod.at> [ubifs, mtd]
Signed-off-by: Jens Axboe <axboe@kernel.dk>
Submounts have their own superblock, which needs to be initialized.
However, they do not have a fuse_fs_context associated with them, and
the root node's attributes should be taken from the mountpoint's node.
Extend fuse_fill_super_common() to work for submounts by making the @ctx
parameter optional, and by adding a @submount_finode parameter.
(There is a plain "unsigned" in an existing code block that is being
indented by this commit. Extend it to "unsigned int" so checkpatch does
not complain.)
Signed-off-by: Max Reitz <mreitz@redhat.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
We want to allow submounts for the same fuse_conn, but with different
superblocks so that each of the submounts has its own device ID. To do
so, we need to split all mount-specific information off of fuse_conn
into a new fuse_mount structure, so that multiple mounts can share a
single fuse_conn.
We need to take care only to perform connection-level actions once (i.e.
when the fuse_conn and thus the first fuse_mount are established, or
when the last fuse_mount and thus the fuse_conn are destroyed). For
example, fuse_sb_destroy() must invoke fuse_send_destroy() until the
last superblock is released.
To do so, we keep track of which fuse_mount is the root mount and
perform all fuse_conn-level actions only when this fuse_mount is
involved.
Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
With the last commit, all functions that handle some existing fuse_req
no longer need to be given the associated fuse_conn, because they can
get it from the fuse_req object.
Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Every fuse_req belongs to a fuse_conn. Right now, we always know which
fuse_conn that is based on the respective device, but we want to allow
multiple (sub)mounts per single connection, and then the corresponding
filesystem is not going to be so trivial to obtain.
Storing a pointer to the associated fuse_conn in every fuse_req will
allow us to trivially find any request's superblock (and thus
filesystem) even then.
Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
After unlock_request() pages from the ap->pages[] array may be put (e.g. by
aborting the connection) and the pages can be freed.
Prevent use after free by grabbing a reference to the page before calling
unlock_request().
The original patch was created by Pradeep P V K.
Reported-by: Pradeep P V K <ppvk@codeaurora.org>
Cc: <stable@vger.kernel.org>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
the callers rely upon having any iov_iter_truncate() done inside
->direct_IO() countered by iov_iter_reexpand().
Reported-by: Qian Cai <cai@redhat.com>
Tested-by: Qian Cai <cai@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Add logic to free up a busy memory range. Freed memory range will be
returned to free pool. Add a worker which can be started to select
and free some busy memory ranges.
Process can also steal one of its busy dax ranges if free range is not
available. I will refer it to as direct reclaim.
If free range is not available and nothing can't be stolen from same
inode, caller waits on a waitq for free range to become available.
For reclaiming a range, as of now we need to hold following locks in
specified order.
down_write(&fi->i_mmap_sem);
down_write(&fi->dax->sem);
We look for a free range in following order.
A. Try to get a free range.
B. If not, try direct reclaim.
C. If not, wait for a memory range to become free
Signed-off-by: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Liu Bo <bo.liu@linux.alibaba.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
This list will be used selecting fuse_dax_mapping to free when number of
free mappings drops below a threshold.
Signed-off-by: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
