The include list still reflects code that left this file years ago:
nothing here uses sched/mm.h, pagemap.h, namei.h, syscalls.h or
anything from fs/internal.h anymore, mount.h and the bm_fs_type
forward declaration lost their last user when the pinned bm_mnt
machinery was removed. Drop all of that and instead spell out the
headers the file actually relies on but so far pulled in
transitively: bitops, bits, bug, cleanup, cred, kstrtox, printk,
refcount, string and user_namespace. With that nothing needs the
kernel.h grab bag anymore, so it goes too, and the list is sorted
alphabetically.
Link: https://patch.msgid.link/20260710-work-binfmt_misc-locking-v3-23-a162f7cb58d6@kernel.org
Signed-off-by: Christian Brauner (Amutable) <brauner@kernel.org>
Use umode_t for the mode argument of bm_get_inode(), constify the
fixed status strings in bm_status_read(), give the super_operations
the bm_ prefix everything else in this file uses, replace the stale
scanarg() comment which still described parameters and an err
variable it lost decades ago and fix the delimiter typo plus a
missing space nearby.
No functional change.
Link: https://patch.msgid.link/20260710-work-binfmt_misc-locking-v3-22-a162f7cb58d6@kernel.org
Reviewed-by: Jori Koolstra <jkoolstra@xs4all.nl>
Signed-off-by: Christian Brauner (Amutable) <brauner@kernel.org>
bm_register_write() has to free the entry it got from create_entry()
on every failure until add_entry() has linked it into the filesystem
and made the inode its owner. Arm the entry with __free(kfree) so the
error branches can simply return and disarm it via
retain_and_null_ptr() once ownership has been handed to the inode.
The interpreter file keeps its manual error cleanup as freeing the
entry would not close it.
No functional change.
Link: https://patch.msgid.link/20260710-work-binfmt_misc-locking-v3-21-a162f7cb58d6@kernel.org
Reviewed-by: Jori Koolstra <jkoolstra@xs4all.nl>
Signed-off-by: Christian Brauner (Amutable) <brauner@kernel.org>
create_entry() is a two hundred line parser with the M and E field
handling inlined as the two arms of its largest branch. Move them
into parse_magic_fields() and parse_extension_fields() which return
the new parse position or NULL so create_entry() itself reads like
the register string grammar again.
The offset parsing loses a provably dead check on the way: after
*s = '\0' and p = s the subsequent if (*p++) always reads the just
written NUL byte and can never fail, it only obscured that the code
simply advances past the delimiter.
With the field parsing gone every remaining failure unwinds the same
way, so hand the entry to __free(kfree), return errors directly and
pass ownership out via no_free_ptr() on success instead of routing
every exit through goto tails.
Link: https://patch.msgid.link/20260710-work-binfmt_misc-locking-v3-20-a162f7cb58d6@kernel.org
Signed-off-by: Christian Brauner (Amutable) <brauner@kernel.org>
create_entry() allocates the entry and the register string it parses
into in one chunk and finds the string part again through manual
pointer arithmetic behind a cast. Make the layout explicit with a
flexible array member and struct_size(), and give the magic pad of
trailing delimiters a name while at it.
No functional change.
Link: https://patch.msgid.link/20260710-work-binfmt_misc-locking-v3-19-a162f7cb58d6@kernel.org
Reviewed-by: Jori Koolstra <jkoolstra@xs4all.nl>
Signed-off-by: Christian Brauner (Amutable) <brauner@kernel.org>
Both write handlers open-code the same removal dance - grab the root
inode lock, unlink, unlock - each carrying a verbatim copy of the
same eleven-line locking comment, and bm_entry_write() reuses its
inode variable for the root inode halfway through to pull it off.
Move the dance into bm_remove_entry() and bm_remove_all_entries()
and the locking rules into the kernel-doc of remove_binfmt_handler()
which both helpers wrap.
No functional change.
Link: https://patch.msgid.link/20260710-work-binfmt_misc-locking-v3-17-a162f7cb58d6@kernel.org
Reviewed-by: Jori Koolstra <jkoolstra@xs4all.nl>
Signed-off-by: Christian Brauner (Amutable) <brauner@kernel.org>
load_misc_binary() seeds retval with the error for checks that
happen further down, reassigns it along the way and funnels every
exit through a ret label whose only job is dropping the entry
reference, so figuring out what an early return actually returns
means replaying the assignment history. Give put_binfmt_handler() a
cleanup class and take the reference with __free() so every failure
can return its error right where the condition is checked. The
comment at the label restated what the put_binfmt_handler()
kernel-doc already explains, it goes with the label. Drop the dead
NULL initialization of interp_file which is assigned on all paths
before use.
No functional change.
Link: https://patch.msgid.link/20260710-work-binfmt_misc-locking-v3-15-a162f7cb58d6@kernel.org
Reviewed-by: Jori Koolstra <jkoolstra@xs4all.nl>
Signed-off-by: Christian Brauner (Amutable) <brauner@kernel.org>
load_binfmt_misc() is one word swap away from load_misc_binary(),
the binfmt loader it serves. It doesn't load anything, it looks up
the binfmt_misc instance of the caller's user namespace, so name it
after what it returns in the style of current_user_ns() and friends.
Tighten the parent walk into a for loop and fix the stale wording
and typos in the kernel-doc while at it.
Link: https://patch.msgid.link/20260710-work-binfmt_misc-locking-v3-14-a162f7cb58d6@kernel.org
Reviewed-by: Jori Koolstra <jkoolstra@xs4all.nl>
Signed-off-by: Christian Brauner (Amutable) <brauner@kernel.org>
search_binfmt_handler() open-codes both match types in one loop body
with the maskless magic comparison spelled as a manual xor loop that
is just memcmp() in disguise. Move the extension and magic checks
into helpers so the walk reads as policy - skip disabled entries,
match by entry type - and the maskless case actually uses memcmp().
No functional change.
Link: https://patch.msgid.link/20260710-work-binfmt_misc-locking-v3-13-a162f7cb58d6@kernel.org
Reviewed-by: Jori Koolstra <jkoolstra@xs4all.nl>
Signed-off-by: Christian Brauner (Amutable) <brauner@kernel.org>
Reading an entry file allocates a whole page and formats the status
into it with a chain of manually advanced sprintf() calls, silently
relying on MAX_REGISTER_LENGTH plus the hex-expanded magic and mask
always staying below PAGE_SIZE. Convert the read side to seq_file
which sizes its buffer as needed and gets rid of the open-coded
pointer arithmetic including the last bin2hex() user in the file.
The output is byte for byte identical.
seq_open() clears FMODE_PWRITE for historical reasons and would
silently turn pwrite() on entry files into -ESPIPE even though
bm_entry_write() accepts writes at any offset. Restore the flag in
bm_entry_open() the same way kernfs does for its seq_file backed
files so pwrite() keeps working.
The only user-visible difference is that seeking is now bound by
seq_lseek() instead of default_llseek(), i.e. SEEK_END stops working
on entry files, which nothing can sensibly use anyway.
The status file keeps its simple_read_from_buffer() as it only ever
returns one of two fixed strings.
Link: https://patch.msgid.link/20260710-work-binfmt_misc-locking-v3-12-a162f7cb58d6@kernel.org
Reviewed-by: Jori Koolstra <jkoolstra@xs4all.nl>
Signed-off-by: Christian Brauner (Amutable) <brauner@kernel.org>
The hex dumps in create_entry() are compiled out unless someone edits
the file to define DEBUG while the pr_debug() calls right next to
them are dynamic-debug aware. Switch the dumps to
print_hex_dump_debug() which follows the same rules as pr_debug() so
the register parsing debug output is uniformly controlled through
dynamic debug, and remove the USE_DEBUG machinery.
Drop the magic[masked] dump instead of converting it: it printed the
bitwise AND of two buffers dumped right above it and required a
temporary allocation on every registration just to recompute what
the reader can derive from the magic and mask dumps directly.
Link: https://patch.msgid.link/20260710-work-binfmt_misc-locking-v3-11-a162f7cb58d6@kernel.org
Reviewed-by: Jori Koolstra <jkoolstra@xs4all.nl>
Signed-off-by: Christian Brauner (Amutable) <brauner@kernel.org>
VERBOSE_STATUS is a compile-time constant that has been fixed to 1
for as long as git history reaches. Turning it off requires editing
the source and yields entry files that only ever report
"enabled"/"disabled", a format nothing has ever seen in the wild.
Remove the pretend knob and the dead branch it guards.
Link: https://patch.msgid.link/20260710-work-binfmt_misc-locking-v3-10-a162f7cb58d6@kernel.org
Reviewed-by: Jori Koolstra <jkoolstra@xs4all.nl>
Signed-off-by: Christian Brauner (Amutable) <brauner@kernel.org>
The CamelCase Node typedef is a 1997 leftover and hides that this is
a plain struct. Call it what it is: struct binfmt_misc_entry, matching
struct binfmt_misc that it hangs off of and the entry bit and flag
enums. Drop the typedef, switch the size computations in
create_entry() to sizeof(*e) and adjust the comments that still
referred to the old name.
No functional change.
Link: https://patch.msgid.link/20260710-work-binfmt_misc-locking-v3-9-a162f7cb58d6@kernel.org
Reviewed-by: Jori Koolstra <jkoolstra@xs4all.nl>
Signed-off-by: Christian Brauner (Amutable) <brauner@kernel.org>
The MISC_FMT_* behavior flags are macros using unsigned long literals
while the entry bit numbers right above them are now a proper enum.
Move the flags into an enum as well so every flags word constant is
declared in one form and shows up in debuginfo. (1U << N) keeps the
enumerators within unsigned int range which is well-defined for enum
constants and the values are unchanged when promoted to the unsigned
long flags word.
No functional change.
Link: https://patch.msgid.link/20260710-work-binfmt_misc-locking-v3-8-a162f7cb58d6@kernel.org
Reviewed-by: Jori Koolstra <jkoolstra@xs4all.nl>
Signed-off-by: Christian Brauner (Amutable) <brauner@kernel.org>
Enabled and Magic are bit numbers in the flags word of an entry but
are declared as bare, unprefixed enumerators with implicit values in
a style that predates the git history. Give the enum a name, explicit
bit numbers and namespaced names and use BIT() instead of open-coding
the shifts when building the initial flags word in create_entry().
No functional change.
Link: https://patch.msgid.link/20260710-work-binfmt_misc-locking-v3-7-a162f7cb58d6@kernel.org
Reviewed-by: Jori Koolstra <jkoolstra@xs4all.nl>
Signed-off-by: Christian Brauner (Amutable) <brauner@kernel.org>
->enabled has always been read and written locklessly: every exec
reads it in load_misc_binary() while bm_status_write() or a concurrent
remount via bm_fill_super() may flip it. That is fine as it is an
independent boolean toggle but the accesses should be marked
accordingly for KCSAN. Annotate them with READ_ONCE()/WRITE_ONCE().
Link: https://patch.msgid.link/20260710-work-binfmt_misc-locking-v3-6-a162f7cb58d6@kernel.org
Reviewed-by: Jori Koolstra <jkoolstra@xs4all.nl>
Signed-off-by: Christian Brauner (Amutable) <brauner@kernel.org>
Once binfmt_misc is loaded load_misc_binary() runs for every execve()
on the system since binfmt_misc registers at the head of the formats
list. Every exec therefore performs read_lock() and read_unlock() on
the entries_lock of the relevant binfmt_misc instance, i.e., two
atomic read-modify-writes on a shared cacheline. User namespaces
without their own binfmt_misc mount fall back to an ancestor's
instance so on container-heavy systems every exec on the machine
typically ends up hammering the cacheline of init_binfmt_misc. On
PREEMPT_RT the rwlock additionally turns the handler lookup into a
sleeping lock on the exec fast path.
The lock protects very little. Entries are immutable after publication
except for the Enabled bit which is already toggled locklessly via
set_bit()/clear_bit() and entry lifetime is already handled by the
users refcount via get_binfmt_handler()/put_binfmt_handler(). The read
lock's only remaining job is to make "the entry is still linked" and
"take a reference" atomic with respect to the unlink sites.
Switch the lookup to an RCU walk:
* Lookup walks the entry list under rcu_read_lock() and acquires a
reference via refcount_inc_not_zero(). The refcount can only drop to
zero after an entry has been unlinked so a failed increment means
the walk raced with an unlink. Restarting the search is bounded
because an unlinked entry cannot be found again.
* The unlink sites use hlist_del_init_rcu() which keeps the forward
pointer intact for concurrent walkers and preserves hlist_unhashed()
as the protection against double removal.
* The final put frees the entry via kfree_rcu() as a concurrent walker
may still dereference its flags, magic, mask, and inline strings.
They all live in the entry allocation itself and thus stay valid
until a grace period has elapsed. Closing the interpreter file stays
synchronous. It is only used with a reference already held and all
final puts run in process context.
* Writers remain serialized by the inode lock of the root dentry with
one exception. bm_evict_inode() called from generic_shutdown_super()
during umount unlinks entries without holding it. Keep a spinlock
around the unlink sites instead of relying on superblock lifetime
rules to make that exclusion implicit.
Handler removal semantics are unchanged. An exec that acquired a
reference just before its handler was unregistered already completes
with the removed handler today. The read lock never protected against
that, it only made the window smaller.
With this an exec that matches no binfmt_misc entry, the common case,
no longer writes to any shared cacheline at all.
Link: https://patch.msgid.link/20260710-work-binfmt_misc-locking-v3-5-a162f7cb58d6@kernel.org
Reviewed-by: Jori Koolstra <jkoolstra@xs4all.nl>
Signed-off-by: Christian Brauner (Amutable) <brauner@kernel.org>
The upcoming conversion of the handler lookup to RCU walks cannot use
list_del_init(): reinitializing the forward pointer of a removed entry
would make a concurrent lockless walker standing on that entry loop
back onto it indefinitely. The removal paths do rely on
reinitialization though because bm_{entry,status}_write() and
bm_evict_inode() need to detect whether an entry has already been
unlinked.
hlists support exactly this pattern: hlist_del_init_rcu() keeps the
forward pointer of the removed entry intact for concurrent walkers and
only zeroes ->pprev with hlist_unhashed() serving as the linked test.
Convert the entry list to an hlist now while keeping the rwlock so the
subsequent RCU conversion is a pure locking change. hlist_add_head()
inserts at the head just as list_add() did so lookup precedence
between registered handlers is unchanged.
Link: https://patch.msgid.link/20260710-work-binfmt_misc-locking-v3-4-a162f7cb58d6@kernel.org
Reviewed-by: Jori Koolstra <jkoolstra@xs4all.nl>
Signed-off-by: Christian Brauner (Amutable) <brauner@kernel.org>
bm_get_tree() takes a reference to the user namespace and hands it to
get_tree_keyed() as the sget key. sget_fc() moves that reference into
sb->s_fs_info and clears fc->s_fs_info, so from that point on the
superblock owns it and bm_free() doesn't see it anymore.
The superblock drops it in ->put_super(). But generic_shutdown_super()
only calls ->put_super() from inside the if (sb->s_root) branch, so
nothing releases it when bm_fill_super() fails:
- The kzalloc_obj() failure leaves s_root NULL and the whole branch is
skipped.
- A simple_fill_super() failure in the file loop leaves s_root set, but
s_op still points at simple_super_operations, which has no
->put_super(). bm_fill_super() installs s_ops only once
simple_fill_super() returned success, and installing it earlier
wouldn't help either because simple_fill_super() overwrites s_op.
Either way vfs_get_super() calls deactivate_locked_super() and the
reference is gone for good. binfmt_misc mounts are available in a user
namespace and both the inode and the dentry cache are SLAB_ACCOUNT, so
an unprivileged caller under a tight memory cgroup can fail
simple_fill_super() on demand and leak one user namespace per attempt.
Drop the reference in ->kill_sb() instead, which runs unconditionally,
the same way nfsd and rpc_pipefs release their keyed s_fs_info.
That also stops ->put_super() from clearing s_fs_info while the
superblock is still on @fs_supers. generic_shutdown_super() leaves it
there on purpose so that sget_fc() keeps finding it until kill_sb() has
run, but a NULL s_fs_info makes test_keyed_super() miss it, so a
concurrent mount for the same user namespace skips the grab_super()
wait and creates a second superblock for a namespace that is still
being torn down.
Link: https://patch.msgid.link/20260728-work-binfmt_misc-usernsleak-v1-1-dbd8d5e626e7@kernel.org
Fixes: 21ca59b365 ("binfmt_misc: enable sandboxed mounts")
Cc: stable@vger.kernel.org
Signed-off-by: Christian Brauner (Amutable) <brauner@kernel.org>
The registration string starts with a user chosen delimiter that
separates the individual fields. So that the field parsers terminate
even on a truncated string create_entry() pads the buffer with that
same delimiter:
memset(buf + count, del, 8);
Most fields are scanned for the delimiter with strchr()/scanarg() and
happily stop on the padding. The flags field is different: instead of
scanning for the delimiter check_special_flags() consumes the flag
characters 'P', 'O', 'C' and 'F' and stops at the first byte that is
none of them, relying on the trailing delimiter to end the scan.
If the delimiter is itself a flag character the padding no longer acts
as a terminator. The scan swallows all eight padding bytes and keeps
reading past the end of the allocation until it hits a byte that is
not a flag character. For example registering
PaPEPPxPPiP
with 'P' as the delimiter (name "a", type extension, magic "x",
interpreter "i", empty flags) leaves the flag scan running off the end
of the buffer. The registration is rejected in the end because the
parser does not stop exactly at buf + count, but only after the out of
bounds read has already happened. With an unlucky allocation layout the
scan can walk into an unmapped page; under KASAN it is reported as a
slab out of bounds read. binfmt_misc mounts are available to
unprivileged users in a user namespace so the read is reachable without
privileges.
Reject a delimiter that is one of the flag characters up front. Such a
registration was always rejected anyway, only after the out of bounds
read, so no valid registration string changes meaning.
Link: https://patch.msgid.link/20260710-work-binfmt_misc-locking-v3-3-a162f7cb58d6@kernel.org
Fixes: 1da177e4c3 ("Linux-2.6.12-rc2")
Cc: stable@vger.kernel.org
Signed-off-by: Christian Brauner (Amutable) <brauner@kernel.org>
For MISC_FMT_OPEN_FILE entries load_misc_binary() clones the
registered interpreter file and denies write access to the clone via
plain deny_write_access(). The clone is installed as
bprm->interpreter and later released by the exec machinery through
exe_file_allow_write_access() which skips the i_writecount increment
for files with FMODE_FSNOTIFY_HSM set.
The deny and allow side can therefore come to different conclusions
when pre-content watches are in play: if a pre-content watch is added
to the interpreter after registration every subsequent exec through
that entry takes a write denial on the clone that is never paired
with a write allowance, driving the interpreter inode's i_writecount
further down with each exec and leaving the interpreter unwritable
even after the entry and all its users are gone.
Take the write denial via exe_file_deny_write_access() so both sides
of the pairing base their decision on the same file mode, and
propagate failure instead of silently ignoring it: an interpreter
that is concurrently open for writing now fails the exec with
ETXTBSY, exactly like an interpreter freshly opened via open_exec()
would.
Link: https://patch.msgid.link/20260710-work-binfmt_misc-locking-v3-2-a162f7cb58d6@kernel.org
Fixes: 0357ef03c9 ("fs: don't block write during exec on pre-content watched files")
Cc: stable@vger.kernel.org
Signed-off-by: Christian Brauner (Amutable) <brauner@kernel.org>
Registering an entry with the MISC_FMT_OPEN_FILE flag opens the
interpreter via open_exec() which denies write access to it for as
long as the entry exists. Removing the entry closes the interpreter
file via filp_close() but never restores write access, leaving the
inode's i_writecount permanently negative. Opening the interpreter
for writing keeps failing with ETXTBSY long after the entry is gone
until the inode is evicted from the inode cache.
Commit 90f601b497 ("binfmt_misc: restore write access before
closing files opened by open_exec()") fixed the same imbalance in the
error path of bm_register_write() but the actual removal path has
been leaking the write denial since the introduction of the flag.
Restore write access in put_binfmt_handler() before closing the
interpreter file.
Link: https://patch.msgid.link/20260710-work-binfmt_misc-locking-v3-1-a162f7cb58d6@kernel.org
Fixes: 948b701a60 ("binfmt_misc: add persistent opened binary handler for containers")
Cc: stable@vger.kernel.org
Signed-off-by: Christian Brauner (Amutable) <brauner@kernel.org>
Christian Brauner <brauner@kernel.org> says:
An entry registered with 'F' opens its interpreter at registration time
and holds that file until the entry is freed. Any entry nobody removes
by hand only gets closed once the binfmt_misc superblock is shut down.
If the interpreter lives on a mount that keeps that superblock alive the
two pin each other:
binfmt_misc sb -> inode -> entry -> interp_file -> vfsmount -> binfmt_misc sb
TL;DR the file is never closed. Once the mount namespace is gone there
is nothing left to unregister through either.
There are two ways to trigger this bug:
- Point the interpreter at the instance itself. Its files are regular
files owned by the mounter and both bm_get_inode() and
simple_fill_super() leave i_op at empty_iops. So notify_change() falls
back to simple_setattr() and chmod +x works. We never set SB_I_NOEXEC
and so open_exec() accepts it.
- Use the instance as an overlayfs lower layer. The overlay superblock
holds a clone_private_mount() of every layer until it is destroyed and
that clone is in no namespace. So umount_tree() never reaches it.
That's a DoS. And it isn't only the superblock that leaks. It pins the
user namespace it was mounted in, so every iteration permanently eats
one of the caller's user namespace charges.
So let's just do the sane thing. SB_I_NOEXEC makes open_exec() fail on
the instance's own files and s_stack_depth makes overlayfs reject the
layer before it ever takes a clone. That also covers the ecryptfs and
fuse passthrough variants. What 'F' promises is unchanged.
The stable tag is narrower than the Fixes tags on purpose. Before
sandboxed mounts this needed global root against the single instance
everyone shares, and the change doesn't apply to those trees anyway.
* patches from https://patch.msgid.link/20260728-work-binfmt_misc-selfpin-v1-0-74df5daeca5b@kernel.org:
binfmt_misc: don't let an 'F' entry pin its own instance
Link: https://patch.msgid.link/20260728-work-binfmt_misc-selfpin-v1-0-74df5daeca5b@kernel.org
Signed-off-by: Christian Brauner (Amutable) <brauner@kernel.org>
An entry registered with 'F' opens its interpreter at registration time
and holds that file until the entry is freed. Any entry nobody removes
by hand only gets closed once the binfmt_misc superblock is shut down.
If the interpreter lives on a mount that keeps that superblock alive the
two pin each other:
binfmt_misc sb -> inode -> entry -> interp_file -> vfsmount -> binfmt_misc sb
TL;DR the file is never closed. Once the mount namespace is gone there
is nothing left to unregister through either.
There are two ways to trigger this bug:
- Point the interpreter at the instance itself. Its files are regular
files owned by the mounter and both bm_get_inode() and
simple_fill_super() leave i_op at empty_iops. So notify_change() falls
back to simple_setattr() and chmod +x works. We never set SB_I_NOEXEC
and so open_exec() accepts it.
- Use the instance as an overlayfs lower layer. The overlay superblock
holds a clone_private_mount() of every layer until it is destroyed and
that clone is in no namespace. So umount_tree() never reaches it.
That's a DoS. And it isn't only the superblock that leaks. It pins the
user namespace it was mounted in, so every iteration permanently eats
one of the caller's user namespace charges.
So let's just do the sane thing. SB_I_NOEXEC makes open_exec() fail on
the instance's own files and s_stack_depth makes overlayfs reject the
layer before it ever takes a clone. That also covers the ecryptfs and
fuse passthrough variants. What 'F' promises is unchanged.
The stable tag is narrower than the Fixes tags on purpose. Before
sandboxed mounts this needed global root against the single instance
everyone shares, and the change doesn't apply to those trees anyway.
Note that SB_I_NODEV is implicitly raised for userns mounts but raise it
explicitly here as well.
Link: https://patch.msgid.link/20260728-work-binfmt_misc-selfpin-v1-1-74df5daeca5b@kernel.org
Fixes: 948b701a60 ("binfmt_misc: add persistent opened binary handler for containers")
Fixes: 21ca59b365 ("binfmt_misc: enable sandboxed mounts")
Cc: stable@vger.kernel.org # v6.7+
Signed-off-by: Christian Brauner (Amutable) <brauner@kernel.org>
David Howells <dhowells@redhat.com> says:
Here are some miscellaneous fixes for netfslib.
(1) Clear PG_private_2 on copy-to-cache append failure.
(2) Fix handling of rolling buffer allocation failure in single-object
writeback. This is probably unnecessary with (4), but if we're only
writing to the cache, we can skip the write.
(3) Fix cleanup of readeahead folios if iterator preparation fails.
(4) Fix folio_queue allocation failure in writeback by adding a mempool.
This also improves request and subrequest allocation.
* patches from https://patch.msgid.link/20260727130716.1099906-1-dhowells@redhat.com:
netfs: Fix folio_queue ENOMEM in writeback by adding a mempool
netfs: release readahead folios on iterator preparation failure
netfs: handle single writeback rolling buffer allocation failure
netfs: clear PG_private_2 on copy-to-cache append failure
Link: https://patch.msgid.link/20260727130716.1099906-1-dhowells@redhat.com
Signed-off-by: Christian Brauner (Amutable) <brauner@kernel.org>
In afs_make_call(), there's a race with async call reception and
destruction. If a call is dispatched that doesn't have call->write_iter
set (used to specify the data content for FS.StoreData), then the first
rxrpc_kernel_send_data() will not set MSG_MORE in the msghdr.
Once rxrpc_send_data() queues the last request packet, the response could
come in at any time and cause the call to be completed and put. However,
afs_make_call() will look at the call again to see it ->write_iter should
be handled - something it's only allowed to do if it has its own ref on the
call. Whilst this is the case for synchronous calls, it isn't true for
async calls such as FS.FetchData.
There's also a potential UAF in afs_make_call() in the event that an
asynchronous call is being sent, but the call fails in some way (e.g. it
gets aborted from the server). The problem there is that afs_make_call()
tries to abort a call if the rxrpc send fails, but the asynchronous
notification from rxrpc may have caused the afs_call to be torn down.
generic/650 plays games with randomly taking CPUs offline, and can
interject a significant delay such that the call is deallocated before
afs_make_call() gets to check call->write_iter - and a UAF ensues (caught
by KASAN).
BUG: KASAN: slab-use-after-free in afs_make_call+0x1c90/0x2210 [kafs]
Read of size 8 at addr ffff888035e050e8 by task fsstress/1409
Fix this by making afs_make_op_call() give the op->call its own ref rather
than transferring the caller's ref to it and then dropping the ref when
afs_make_call() returns.
This also means that the afs_make_call() func never loses its ref on the
call now.
Fixes: eddf51f2bb ("afs: Make {Y,}FS.FetchData an asynchronous operation")
Fixes: e49c7b2f6d ("afs: Build an abstraction around an "operation" concept")
Link: https://sashiko.dev/#/patchset/20260702144919.172295-1-dhowells%40redhat.com
Reported-by: Marc Dionne <marc.dionne@auristor.com>
Signed-off-by: David Howells <dhowells@redhat.com>
Link: https://patch.msgid.link/20260723113452.566619-4-dhowells@redhat.com
cc: Jeffrey Altman <jaltman@auristor.com>
cc: linux-afs@lists.infradead.org
cc: stable@kernel.org
Signed-off-by: Christian Brauner (Amutable) <brauner@kernel.org>
Pull misc fixes from Andrew Morton:
"13 hotfixes. All are cc:stable. 11 are for MM. All are singletons -
please see the changelogs for details"
* tag 'mm-hotfixes-stable-2026-07-27-14-18' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm:
fs/proc/task_mmu: fix PAGEMAP_SCAN written state for PMD holes
mm/hugetlb: fix list corruption in allocate_file_region_entries()
mm: mglru: fix stale batch updates after memcg reparenting
selftest: fix headers in fclog.c
ocfs2: fix boundary check in ocfs2_check_dir_entry() to use buffer offset
mm/percpu-km: fix bitmap overflow and accounting in pcpu_create_chunk()
mm/util: don't read __page_2 for order-1 folios in snapshot_page()
mm/hugetlb: fix swap entry corruption when clearing uffd-wp at fork()
mm: migrate_device: fix pte_pfn/pte_dirty called on non-present PTE
fs/proc/task_mmu: fix PAGEMAP_SCAN written state for unpopulated ptes
userfaultfd: wait on source PMD during UFFDIO_MOVE
lib: test_hmm: use device devt for coherent device range selection
mm/vmstat: fold stranded per-cpu node stats when a node comes online
Pull keys fixes from Jarkko Sakkinen:
- An unprivileged keyring whose keys collide through the
description-chunk path can drive assoc_array node splitting
into an out-of-bounds slot write. Fix it.
- Fix the DCP trusted keys backend
* tag 'for-next-keys-7.2-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/jarkko/linux-tpmdd:
assoc_array: trim the final shortcut word using the current chunk end
keys: make keyring key-chunk byte order agree with keyring_diff_objects()
keys: fix out-of-bounds read in keyring_get_key_chunk()
KEYS: trusted: dcp: fix key_len validation and calc_blob_len() return type
Pull erofs fixes from Gao Xiang:
"Fix a regression in page cache sharing which can cause a NULL pointer
dereference, and limit LZMA stream memory usage on systems with many
CPUs.
- Keep a valid f_path for page cache sharing to fix a recent
mincore() NULL pointer dereference
- Limit LZMA stream pool size when too many processors are available
- Sync up with Hongbo Li's latest email address"
* tag 'erofs-for-7.2-rc6-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/xiang/erofs:
erofs: cap LZMA stream pool size
erofs: ensure valid f_path for page cache sharing
MAINTAINERS: update Hongbo Li's email address
Pull pin control fixes from Linus Walleij:
"The most interesting commit is the S4 fix for AMD, which probably is
helpful to a whole bunch of important machines.
- Wakeup nits on the Qualcomm SC8280XP
- Double-free issues on the device tree parsing error path
- Fixup of the S4 sleep state handling on AMD pin control
- Missing Kconfig select REGMAP_MMIO for the Microchip driver leading
to compile stalls
- Missing Kconfig select GENERIC_PINCONF for the Bitmain BM1880
leading to compile stalls"
* tag 'pinctrl-v7.2-2' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-pinctrl:
pinctrl: bm1880: add missing select GENERIC_PINCONF
pinctrl-amd: Don't clear S4 wake bits at probe
pinctrl: microchip-sgpio: add missing select REGMAP_MMIO
pinctrl: devicetree: don't free uninitialized dev_name on error path
pinctrl: qcom: sc8280xp: Add missing wakeup entries for GPIO143/151
pinctrl: qcom: Unconditionally mark gpio as wakeup enable
fs/erofs/decompressor_lzma.c sizes the module-global MicroLZMA stream
pool from num_possible_cpus() when the lzma_streams module parameter is
unset, then z_erofs_load_lzma_config() preallocates one image-supplied
dictionary per stream, accepting dictionaries up to 8 MiB. On high-CPU
systems, a small EROFS image can pin hundreds of MiB of vmalloc-backed
decoder state until the erofs module is unloaded.
Impact: An EROFS image mounted by the system can pin up to 8 MiB of
vmalloc memory per LZMA stream, either as intended or unexpectedly.
Bound the default stream count by a new
CONFIG_EROFS_FS_ZIP_LZMA_DEFAULT_MAX_STREAMS option, default 16, so the
worst-case default preallocation is 128 MiB if the number of CPUs is no
less than 16 while preserving the existing per-image dictionary limit.
An explicit lzma_streams module parameter is still honoured as-is, so
administrators who deliberately size the pool are not affected.
Fixes: 622ceaddb7 ("erofs: lzma compression support")
Cc: stable@vger.kernel.org
Assisted-by: Claude:claude-opus-4-8
Signed-off-by: Michael Bommarito <michael.bommarito@gmail.com>
Reviewed-by: Gao Xiang <hsiangkao@linux.alibaba.com>
Signed-off-by: Gao Xiang <hsiangkao@linux.alibaba.com>
Previously, backing files for page cache sharing were set up with
f_path left as NULL (only f_inode was valid). It worked, but a recent
mincore fix relies on f_path.mnt and crashes (found by "erofs/028" on
7.2-rc4):
BUG: kernel NULL pointer dereference, address: 0000000000000018
#PF: supervisor read access in kernel mode
#PF: error_code(0x0000) - not-present page
PGD 0 P4D 0
Oops: Oops: 0000 [#1] SMP PTI
CPU: 3 UID: 0 PID: 675528 Comm: fincore Not tainted 7.2.0-rc4-00002-g[]-dirty #1 PREEMPT(lazy)
Hardware name: Red Hat KVM, BIOS 1.16.0-4.al8 04/01/2014
RIP: 0010:__do_sys_mincore+0xc0/0x2c0
...
Specify valid paths using valid disconnected dentries together with
erofs_ishare_mnt instead of leaving f_path empty, so they are more
like real backing files in a pseudo filesystem and standard
backing_file_open() can be used directly.
Fixes: e187bc02f8 ("mm: do file ownership checks with the proper mount idmap")
Acked-by: Hongbo Li <hongbohbli@tencent.com>
Signed-off-by: Gao Xiang <xiang@kernel.org>
Pull vfs fixes from Christian Brauner:
- vfs: Preserve the ACL_DONT_CACHE state in forget_cached_acl().
ACL_DONT_CACHE is meant to be a permanent opt-out from ACL caching
which FUSE relies on for servers that don't negotiate FUSE_POSIX_ACL.
The helper replaced it with ACL_NOT_CACHED, silently re-enabling the
cache, and as fuse doesn't invalidate the cache for such servers a
properly timed get_acl() returned stale ACLs. Comes with a fuse
selftest reproducing this.
- pidfs:
- Preserve PIDFD_THREAD when a thread pidfd is reopened via
open_by_handle_at(). PIDFD_THREAD shares the O_EXCL bit which
do_dentry_open() strips after the flags have been validated, so
the reopened pidfd silently became a process pidfd. Comes with a
selftest.
- Add a pidfs_dentry_open() helper so the regular pidfd allocation
path and the file handle path share the code that forces O_RDWR
and reapplies the pidfd flags that do_dentry_open() strips.
- Handle FS_IOC32_GETVERSION in the compat ioctl path.
- Make pidfs_ino_lock static.
- iomap:
- Fix the block range calculation in ifs_clear_range_dirty() so a
partial clear doesn't drop the dirty state of blocks the range
only partially covers.
- Support invalidating partial folios so a partial truncate or hole
punch with blocksize < foliosize doesn't leave stale dirty bits
behind.
- Only set did_zero when iomap_zero_iter() actually zeroed
something.
- Guard ifs_set_range_dirty() and ifs_set_range_uptodate() against
zero-length ranges where the unsigned last-block calculation
underflows and bitmap_set() writes far beyond the ifs->state
allocation.
- Don't merge ioends with different io_private values as the merge
could leak or corrupt the private data of the individual ioends.
- exec:
- Raise bprm->have_execfd only once the binfmt_misc interpreter has
actually been opened. The flag was set as soon as a matching 'O'
or 'C' entry was found. If the interpreter open failed with
ENOEXEC the exec fell through to the next binary format with
have_execfd raised but no executable staged and begin_new_exec()
NULL derefed past the point of no return.
- Fix an unsigned loop counter wrap in transfer_args_to_stack() on
nommu. An overlong argument or environment string pushes bprm->p
below PAGE_SIZE, the stop index becomes zero, and the loop never
terminates, wrapping its counter and copying garbage from in
front of the page array into the new process stack.
- Make binfmt_elf_fdpic only honour the first PT_INTERP like
binfmt_elf does. Each additional PT_INTERP overwrote the previous
interpreter, leaking the name allocation and the interpreter file
reference together with the write denial open_exec() took,
leaving the file unwritable for as long as the system runs.
- overlayfs:
- Compare the full escaped xattr prefix including the trailing dot.
An xattr like "trusted.overlay.overlayfoo" was misclassified as
an escaped overlay xattr.
- Check read access to the copy_file_range() source with the
source's mounter credentials.
- super: Thawing a filesystem whose block device was frozen with
bdev_freeze() deadlocked. Dropping the last block layer freeze
reference from under s_umount ends up in fs_bdev_thaw() which
reacquires s_umount on the same task. Pin the superblock with an
active reference instead and call bdev_thaw() without holding
s_umount.
- procfs: Return EACCES instead of success when the ptrace access check
for namespace links fails.
- afs: Use afs_dir_get_block() rather than afs_dir_find_block() for
block 0 in afs_edit_dir_remove(), matching afs_edit_dir_add().
- Push the memcg gating of ->nr_cached_objects() down into the btrfs
and shmem callbacks instead of skipping every callback during
non-root memcg reclaim. The blanket check short-circuited XFS whose
inode reclaim hook is intentionally driven from per-memcg contexts to
free memcg-charged slab.
- eventpoll: Pin files while checking reverse paths.
Since struct file became SLAB_TYPESAFE_BY_RCU a concurrent close
could free and recycle the file under the check which then took and
dropped the f_lock of whatever live file now occupies that slot.
* tag 'vfs-7.2-rc5.fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs: (24 commits)
super: fix emergency thaw deadlock on frozen block devices
pidfs: make pidfs_ino_lock static
eventpoll: pin files while checking reverse paths
fs: push nr_cached_objects memcg gating into individual filesystems
afs: Fix afs_edit_dir_remove() to get, not find, block 0
iomap: prevent ioend merge when io_private differs
iomap: add comments for ifs_clear/set_range_dirty()
iomap: fix out-of-bounds bitmap_set() with zero-length range
iomap: fix incorrect did_zero setting in iomap_zero_iter()
iomap: support invalidating partial folios
iomap: correct the range of a partial dirty clear
fs/super: fix emergency thaw double-unlock of s_umount
pidfs: handle FS_IOC32_GETVERSION in compat ioctl
ovl: check access to copy_file_range source with src mounter creds
proc: Fix broken error paths for namespace links
pidfs: add pidfs_dentry_open() helper
selftests/pidfd: check PIDFD_THREAD survives open_by_handle_at()
pidfs: preserve thread pidfds reopened by file handle
ovl: fix trusted xattr escape prefix matching
selftests/fuse: add ACL_DONT_CACHE regression test
...
Pull spi fixes from Mark Brown:
"Just a couple of small bits for the SpacemiT driver - one small fix,
and a new compatible in the DT binding"
* tag 'spi-fix-v7.2-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/spi:
spi: dt-bindings: spacemit: add K3 SPI compatible
spi: spacemit: Correct TX FIFO slot calculation
Pull regulator fixes from Mark Brown:
"One driver specific fix where one of the MediaTek drivers duplicated
some core code buggily, and a core fix for an ordering issue on
startup where we could end up configuring a voltage outside of
constraints due to the order in which we applied constraints"
* tag 'regulator-fix-v7.2-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/regulator:
regulator: core: clamp voltage constraints before applying apply_uV
regulator: mt6358: use regmap helper to read fixed LDO calibration
Pull char/misc driver fixes from Greg KH:
"Here are a number of small char/misc/etc driver fixes for 7.2-rc5 that
resolve a bunch of different reported issues. Included in here are:
- rust_binder error message reporting fix
- stratix10-svc firmware driver fixes
- mei driver fix
- intel_th hardware tracing driver fix
- comedi driver fix
- uio_hv_generic driver fix
- ntsync selftest fix
- nsm misc driver fix
- some MAINTAINER file updates
All of these have been in linux-next for over a week with no reported
issues"
* tag 'char-misc-7.2-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc:
MAINTAINERS: Update wine-devel list address
rust_binder: only print failure if error has source
intel_th: fix MSC output device reference leak
misc: nsm: pin the module while the device is open
mei: bus: access mei_device under device_lock on cleanup
misc: nsm: only unlock nsm_dev on post-lock error paths
selftests: ntsync: correct CONFIG_NTSYNC name
comedi: comedi_parport: deal with premature interrupt
uio_hv_generic: Bind to FCopy device by default
MAINTAINERS: Add Greg Kroah-Hartman to GPIB
firmware: stratix10-svc: fix teardown order in remove to prevent race
firmware: stratix10-svc: handle NO_RESPONSE in async poll
firmware: stratix10-svc: fix FCS SMC call kernel-doc
firmware: stratix10-svc: fix memory leaks and list corruption bugs
Pull staging driver fixes from Greg KH:
"Here are two small staging driver fixes for 7.2-rc5. They both resolve
some reported bugs in the rtl8723bs staging driver and have been in
linux-next for over a week with no reported issues"
* tag 'staging-7.2-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging:
staging: rtl8723bs: fix OOB reads in rtw_get_wps_ie()
staging: rtl8723bs: fix inverted HT40 secondary channel offset
Pull serial driver fixes from Greg KH:
"Here are two small serial driver fixes for 7.2-rc5. They are:
- sc16is7xx get_direction() callback fix, which resolves a
user-triggerable warning in the driver
- NULL pointer dereference on some platforms using the 8250_mid
serial driver
Both have been in linux-next for over a week with no reported issues"
* tag 'tty-7.2-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/tty:
serial: sc16is7xx: implement gpio get_direction() callback
serial: 8250_mid: Fix NULL function pointer dereference on DNV/ICX-D/SNR platforms
Pull USB fixes from Greg KH:
"Here are some small USB fixes and new device quirks and ids:
- usb storage quirk added
- new usb serial device ids added
- usb-serial device name leak and other bug fixes
- small xhci driver fixes
- normal batch of typec driver fixes for reported issues
- usb-atm much-reported-by-syzbot fix for firmware download races
- sysfs BOS device removal race fix
- lots of usb gadget driver fixes for reported issues
- other small USB driver fixes for other reported problems
All of these have been in linux-next this past week, many of them much
longer"
* tag 'usb-7.2-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb: (30 commits)
usb: typec: ucsi: Correct teardown ordering in ucsi_init() error path
USB: serial: io_edgeport: cap received transmit credits
USB: serial: option: add TDTECH MT5710-CN
USB: serial: io_ti: reject oversized boot-mode firmware
USB: serial: mxuport: validate firmware header size
usb: atm: ueagle-atm: reject descriptors that confuse probe and disconnect
usb: typec: ucsi: yoga_c630: Remove redundant duplicate altmode handling
usb: typec: ucsi: Add duplicate detection to nvidia registration path
usb: typec: ucsi: Detect and skip duplicate altmodes from buggy firmware
usb: gadget: dummy_hcd: prevent fifo_req reuse during giveback
usb: chipidea: fix usage_count leak when autosuspend_delay is negative
usb: core: sysfs: add lock to bos_descriptors_read()
usb: musb: omap2430: Do not put borrowed of_node in probe
usb: core: port: Deattach Type-C connector on component unbind
USB: storage: add NO_ATA_1X quirk for Longmai USB Key
USB: serial: ftdi_sio: add support for E+H FXA291
USB: serial: keyspan_pda: fix data loss on receive throttling
usb: gadget: printer: fix infinite loop in printer_read()
usb: gadget: f_midi: cancel pending IN work before freeing the midi object
usb: gadget: udc: bdc: free IRQ and drain func_wake_notify before teardown
...
Pull tracing fixes from Steven Rostedt:
- Move rb_desc->nr_page_va before updating dynamic array
The rb_descr->page_va is a dynamic array counted by nr_page_va. But
the updating of the page_va[] is done before the nr_page_va is
incremented causing a build with CONFIG_UBSAN_BOUNDS to flag it as an
overflow.
Move the increment of the counted by value before the array element
is updated.
- Propagate errors from remote event bulk updates
The return value of trace_remote_enable_event() was not being checked
by remote_events_dir_enable_write() where it would silently fail.
Have it check the return value and propagate that back up to user
space.
- Fix resource leak on mmiotrace trace_pipe close
The mmiotrace tracer was created in 2008 before the trace_pipe had a
close callback to allow tracers to do clean up from trace_pipe open.
The trace_pipe close cleanup callback was added in 2009 but the
mmiotrace tracer was not updated. It had a hack to do the cleanup in
the read call, where it may leak if user space did not read the
entire buffer.
Add a callback to mmiotrace trace_pipe close do to the cleanup
properly.
- Fix a possible NULL pointer dereference in the mmiotrace tracer
If the mmio_pipe_open() fails to find a PCI device, it will set the
hiter->dev pointer to NULL. The read function will blindly
dereference that pointer. Fix the read call to check to see if that
pointer is populated before dereferencing it.
- Fix union collision of module and refcnt for dynamic events
In 'struct trace_event_call', the 'module' pointer and the 'refcnt'
atomic variable share the same memory space in a union. The filter on
module logic only checked if the 'module' was set to determine if the
event belonged to the module. As dynamic events are always builtin,
it doesn't need the 'module' field of the structure and used a
refcount. But the module filtering logic would then mistaken these
dynamic events as a module and call module_name(event->module) on it.
Add a check to see if the event is a dynamic event and if so, do not
check it for being part of the given module.
- Reset the top level buffer in selftests before running instances
The ftracetest selftest initializes each instance before executing
the tests. But it does not reset the top level buffer. Dynamic events
are only added and removed by the top level so any left over dynamic
events will not be removed by the reset in the instances.
Left over dynamic events can cause the tests to incorrectly fail.
Reset the top level buffer before running the instances.
- Make the context_switch counter 64 bit
The code to read user space for a system call trace event or for a
trace_marker will disable migration, enable preemption, read user
space into a per CPU buffer, disable preemption and enable migration
again. It checks if the per CPU context switch counter to see if it
changed, and if it did not, it would know that the per CPU buffer was
not touched by another task.
But the save counter was 32 bit and it would compare it to the 64 bit
context_switch variable. A long running system could have the
context_switch variable greater that 1<<32 in which case the compare
will always fail. The compare will promote the 32 bit int saved value
to 64 bit and compare it to the full 64 bit counter. Since the top 32
bits of the saved value was zero, it would never match.
- Fix a use-after-free of the event_enable trigger
The event_enable trigger allows for enabling one event when another
event is triggered. When the trigger is removed, it must go through a
synchronization phase to make sure it is not triggered again. The
trigger itself is delayed by the "bulk delay" logic that was recently
added. But the code that frees the event_enable data used to rely on
the trigger code to do the synchronization. Now that the code uses
the call RCU functions (and a workqueue), that delay no longer is
there.
Add a callback private_data_free() function that allows triggers to
clean up data after the synchronization phase has completed.
- Move the module_ref counter into the delay callback
Since an event of the event_enable trigger can enable an event for a
module, it ups the module ref count for that event's module. This
prevents the event from trying to enable an event that no longer
exists and cause a use-after-free bug.
The ref counter was set back down when the trigger was removed but
not after thy synchronization phase. This could lead to the module
data being accessed after module was unloaded.
Move the module ref decrement into the private_data_free() callback
of the event_enable trigger.
- Add mutex to protect parser in ftrace filtering
The set_ftrace_filter file uses a parsing descriptor that is
allocated at open and modified by writes. If multiple threads were to
write to the descriptor at the same time, it can corrupt the parser.
Add a mutex around the modifications of the parser descriptor.
- Fix possible corruption in perf syscall tracing
The perf system call trace events can now read user space. To do so,
the reads of user space enable preemption and disables it again.
During this time that preemption is enabled, the task can migrate.
The perf event list head is assigned via a per CPU pointer. It is
done before the user space part is called. If the user space reading
migrates the task to another CPU, then the head pointer is no longer
valid.
Re-assign the head pointer after the reading of user space to keep it
using the correct data.
* tag 'trace-v7.2-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace:
tracing: perf: Fix stale head for perf syscall tracing
ftrace: Add global mutex to serialize trace_parser access
tracing: Delay module ref count for "enable_event" trigger
tracing: Fix use-after-free freeing trigger private data
tracing: Fix context switch counter truncation
selftests/ftrace: Reset triggers at top level before instance loop
tracing: Fix union collision of module and refcnt for dynamic events
tracing: Fix mmiotrace possible NULL dereferencing of hiter->dev
tracing: Fix resource leak on mmiotrace trace_pipe close
tracing: Propagate errors from remote event bulk updates
tracing/remotes: Fix page_va[] access before counter update in trace_remote_alloc_buffer()
