msm-4.14/fs/binfmt_misc.c

993 lines
22 KiB
C
Raw Normal View History

/*
* binfmt_misc.c
*
* Copyright (C) 1997 Richard Günther
*
* binfmt_misc detects binaries via a magic or filename extension and invokes
* a specified wrapper. See Documentation/binfmt_misc.txt for more details.
*/
binfmt_misc: add comments & debug logs When trying to develop a custom format handler, the errors returned all effectively get bucketed as EINVAL with no kernel messages. The other errors (ENOMEM/EFAULT) are internal/obvious and basic. Thus any time a bad handler is rejected, the developer has to walk the dense code and try to guess where it went wrong. Needing to dive into kernel code is itself a fairly high barrier for a lot of people. To improve this situation, let's deploy extensive pr_debug markers at logical parse points, and add comments to the dense parsing logic. It let's you see exactly where the parsing aborts, the string the kernel received (useful when dealing with shell code), how it translated the buffers to binary data, and how it will apply the mask at runtime. Some example output: $ echo ':qemu-foo:M::\x7fELF\xAD\xAD\x01\x00:\xff\xff\xff\xff\xff\x00\xff\x00:/usr/bin/qemu-foo:POC' > register $ dmesg binfmt_misc: register: received 92 bytes binfmt_misc: register: delim: 0x3a {:} binfmt_misc: register: name: {qemu-foo} binfmt_misc: register: type: M (magic) binfmt_misc: register: offset: 0x0 binfmt_misc: register: magic[raw]: 5c 78 37 66 45 4c 46 5c 78 41 44 5c 78 41 44 5c \x7fELF\xAD\xAD\ binfmt_misc: register: magic[raw]: 78 30 31 5c 78 30 30 00 x01\x00. binfmt_misc: register: mask[raw]: 5c 78 66 66 5c 78 66 66 5c 78 66 66 5c 78 66 66 \xff\xff\xff\xff binfmt_misc: register: mask[raw]: 5c 78 66 66 5c 78 30 30 5c 78 66 66 5c 78 30 30 \xff\x00\xff\x00 binfmt_misc: register: mask[raw]: 00 . binfmt_misc: register: magic/mask length: 8 binfmt_misc: register: magic[decoded]: 7f 45 4c 46 ad ad 01 00 .ELF.... binfmt_misc: register: mask[decoded]: ff ff ff ff ff 00 ff 00 ........ binfmt_misc: register: magic[masked]: 7f 45 4c 46 ad 00 01 00 .ELF.... binfmt_misc: register: interpreter: {/usr/bin/qemu-foo} binfmt_misc: register: flag: P (preserve argv0) binfmt_misc: register: flag: O (open binary) binfmt_misc: register: flag: C (preserve creds) The [raw] lines show us exactly what was received from userspace. The lines after that show us how the kernel has decoded things. Signed-off-by: Mike Frysinger <vapier@gentoo.org> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Joe Perches <joe@perches.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-12-10 15:52:08 -08:00
#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
#include <linux/kernel.h>
#include <linux/module.h>
#include <linux/init.h>
#include <linux/sched/mm.h>
#include <linux/magic.h>
#include <linux/binfmts.h>
#include <linux/slab.h>
#include <linux/ctype.h>
#include <linux/string_helpers.h>
#include <linux/file.h>
#include <linux/pagemap.h>
#include <linux/namei.h>
#include <linux/mount.h>
#include <linux/syscalls.h>
#include <linux/fs.h>
binfmt_misc: add comments & debug logs When trying to develop a custom format handler, the errors returned all effectively get bucketed as EINVAL with no kernel messages. The other errors (ENOMEM/EFAULT) are internal/obvious and basic. Thus any time a bad handler is rejected, the developer has to walk the dense code and try to guess where it went wrong. Needing to dive into kernel code is itself a fairly high barrier for a lot of people. To improve this situation, let's deploy extensive pr_debug markers at logical parse points, and add comments to the dense parsing logic. It let's you see exactly where the parsing aborts, the string the kernel received (useful when dealing with shell code), how it translated the buffers to binary data, and how it will apply the mask at runtime. Some example output: $ echo ':qemu-foo:M::\x7fELF\xAD\xAD\x01\x00:\xff\xff\xff\xff\xff\x00\xff\x00:/usr/bin/qemu-foo:POC' > register $ dmesg binfmt_misc: register: received 92 bytes binfmt_misc: register: delim: 0x3a {:} binfmt_misc: register: name: {qemu-foo} binfmt_misc: register: type: M (magic) binfmt_misc: register: offset: 0x0 binfmt_misc: register: magic[raw]: 5c 78 37 66 45 4c 46 5c 78 41 44 5c 78 41 44 5c \x7fELF\xAD\xAD\ binfmt_misc: register: magic[raw]: 78 30 31 5c 78 30 30 00 x01\x00. binfmt_misc: register: mask[raw]: 5c 78 66 66 5c 78 66 66 5c 78 66 66 5c 78 66 66 \xff\xff\xff\xff binfmt_misc: register: mask[raw]: 5c 78 66 66 5c 78 30 30 5c 78 66 66 5c 78 30 30 \xff\x00\xff\x00 binfmt_misc: register: mask[raw]: 00 . binfmt_misc: register: magic/mask length: 8 binfmt_misc: register: magic[decoded]: 7f 45 4c 46 ad ad 01 00 .ELF.... binfmt_misc: register: mask[decoded]: ff ff ff ff ff 00 ff 00 ........ binfmt_misc: register: magic[masked]: 7f 45 4c 46 ad 00 01 00 .ELF.... binfmt_misc: register: interpreter: {/usr/bin/qemu-foo} binfmt_misc: register: flag: P (preserve argv0) binfmt_misc: register: flag: O (open binary) binfmt_misc: register: flag: C (preserve creds) The [raw] lines show us exactly what was received from userspace. The lines after that show us how the kernel has decoded things. Signed-off-by: Mike Frysinger <vapier@gentoo.org> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Joe Perches <joe@perches.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-12-10 15:52:08 -08:00
#include <linux/uaccess.h>
#include "internal.h"
binfmt_misc: add comments & debug logs When trying to develop a custom format handler, the errors returned all effectively get bucketed as EINVAL with no kernel messages. The other errors (ENOMEM/EFAULT) are internal/obvious and basic. Thus any time a bad handler is rejected, the developer has to walk the dense code and try to guess where it went wrong. Needing to dive into kernel code is itself a fairly high barrier for a lot of people. To improve this situation, let's deploy extensive pr_debug markers at logical parse points, and add comments to the dense parsing logic. It let's you see exactly where the parsing aborts, the string the kernel received (useful when dealing with shell code), how it translated the buffers to binary data, and how it will apply the mask at runtime. Some example output: $ echo ':qemu-foo:M::\x7fELF\xAD\xAD\x01\x00:\xff\xff\xff\xff\xff\x00\xff\x00:/usr/bin/qemu-foo:POC' > register $ dmesg binfmt_misc: register: received 92 bytes binfmt_misc: register: delim: 0x3a {:} binfmt_misc: register: name: {qemu-foo} binfmt_misc: register: type: M (magic) binfmt_misc: register: offset: 0x0 binfmt_misc: register: magic[raw]: 5c 78 37 66 45 4c 46 5c 78 41 44 5c 78 41 44 5c \x7fELF\xAD\xAD\ binfmt_misc: register: magic[raw]: 78 30 31 5c 78 30 30 00 x01\x00. binfmt_misc: register: mask[raw]: 5c 78 66 66 5c 78 66 66 5c 78 66 66 5c 78 66 66 \xff\xff\xff\xff binfmt_misc: register: mask[raw]: 5c 78 66 66 5c 78 30 30 5c 78 66 66 5c 78 30 30 \xff\x00\xff\x00 binfmt_misc: register: mask[raw]: 00 . binfmt_misc: register: magic/mask length: 8 binfmt_misc: register: magic[decoded]: 7f 45 4c 46 ad ad 01 00 .ELF.... binfmt_misc: register: mask[decoded]: ff ff ff ff ff 00 ff 00 ........ binfmt_misc: register: magic[masked]: 7f 45 4c 46 ad 00 01 00 .ELF.... binfmt_misc: register: interpreter: {/usr/bin/qemu-foo} binfmt_misc: register: flag: P (preserve argv0) binfmt_misc: register: flag: O (open binary) binfmt_misc: register: flag: C (preserve creds) The [raw] lines show us exactly what was received from userspace. The lines after that show us how the kernel has decoded things. Signed-off-by: Mike Frysinger <vapier@gentoo.org> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Joe Perches <joe@perches.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-12-10 15:52:08 -08:00
#ifdef DEBUG
# define USE_DEBUG 1
#else
# define USE_DEBUG 0
#endif
enum {
VERBOSE_STATUS = 1 /* make it zero to save 400 bytes kernel memory */
};
static LIST_HEAD(entries);
static int enabled = 1;
enum {Enabled, Magic};
#define MISC_FMT_PRESERVE_ARGV0 (1UL << 31)
#define MISC_FMT_OPEN_BINARY (1UL << 30)
#define MISC_FMT_CREDENTIALS (1UL << 29)
#define MISC_FMT_OPEN_FILE (1UL << 28)
typedef struct {
struct list_head list;
unsigned long flags; /* type, status, etc. */
int offset; /* offset of magic */
int size; /* size of magic/mask */
char *magic; /* magic or filename extension */
char *mask; /* mask, NULL for exact match */
const char *interpreter; /* filename of interpreter */
char *name;
struct dentry *dentry;
struct file *interp_file;
binfmt_misc: cleanup on filesystem umount [ Upstream commit 1c5976ef0f7ad76319df748ccb99a4c7ba2ba464 ] Currently, registering a new binary type pins the binfmt_misc filesystem. Specifically, this means that as long as there is at least one binary type registered the binfmt_misc filesystem survives all umounts, i.e. the superblock is not destroyed. Meaning that a umount followed by another mount will end up with the same superblock and the same binary type handlers. This is a behavior we tend to discourage for any new filesystems (apart from a few special filesystems such as e.g. configfs or debugfs). A umount operation without the filesystem being pinned - by e.g. someone holding a file descriptor to an open file - should usually result in the destruction of the superblock and all associated resources. This makes introspection easier and leads to clearly defined, simple and clean semantics. An administrator can rely on the fact that a umount will guarantee a clean slate making it possible to reinitialize a filesystem. Right now all binary types would need to be explicitly deleted before that can happen. This allows us to remove the heavy-handed calls to simple_pin_fs() and simple_release_fs() when creating and deleting binary types. This in turn allows us to replace the current brittle pinning mechanism abusing dget() which has caused a range of bugs judging from prior fixes in [2] and [3]. The additional dget() in load_misc_binary() pins the dentry but only does so for the sake to prevent ->evict_inode() from freeing the node when a user removes the binary type and kill_node() is run. Which would mean ->interpreter and ->interp_file would be freed causing a UAF. This isn't really nicely documented nor is it very clean because it relies on simple_pin_fs() pinning the filesystem as long as at least one binary type exists. Otherwise it would cause load_misc_binary() to hold on to a dentry belonging to a superblock that has been shutdown. Replace that implicit pinning with a clean and simple per-node refcount and get rid of the ugly dget() pinning. A similar mechanism exists for e.g. binderfs (cf. [4]). All the cleanup work can now be done in ->evict_inode(). In a follow-up patch we will make it possible to use binfmt_misc in sandboxes. We will use the cleaner semantics where a umount for the filesystem will cause the superblock and all resources to be deallocated. In preparation for this apply the same semantics to the initial binfmt_misc mount. Note, that this is a user-visible change and as such a uapi change but one that we can reasonably risk. We've discussed this in earlier versions of this patchset (cf. [1]). The main user and provider of binfmt_misc is systemd. Systemd provides binfmt_misc via autofs since it is configurable as a kernel module and is used by a few exotic packages and users. As such a binfmt_misc mount is triggered when /proc/sys/fs/binfmt_misc is accessed and is only provided on demand. Other autofs on demand filesystems include EFI ESP which systemd umounts if the mountpoint stays idle for a certain amount of time. This doesn't apply to the binfmt_misc autofs mount which isn't touched once it is mounted meaning this change can't accidently wipe binary type handlers without someone having explicitly unmounted binfmt_misc. After speaking to systemd folks they don't expect this change to affect them. In line with our general policy, if we see a regression for systemd or other users with this change we will switch back to the old behavior for the initial binfmt_misc mount and have binary types pin the filesystem again. But while we touch this code let's take the chance and let's improve on the status quo. [1]: https://lore.kernel.org/r/20191216091220.465626-2-laurent@vivier.eu [2]: commit 43a4f2619038 ("exec: binfmt_misc: fix race between load_misc_binary() and kill_node()" [3]: commit 83f918274e4b ("exec: binfmt_misc: shift filp_close(interp_file) from kill_node() to bm_evict_inode()") [4]: commit f0fe2c0f050d ("binder: prevent UAF for binderfs devices II") Link: https://lore.kernel.org/r/20211028103114.2849140-1-brauner@kernel.org (v1) Cc: Sargun Dhillon <sargun@sargun.me> Cc: Serge Hallyn <serge@hallyn.com> Cc: Jann Horn <jannh@google.com> Cc: Henning Schild <henning.schild@siemens.com> Cc: Andrei Vagin <avagin@gmail.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Laurent Vivier <laurent@vivier.eu> Cc: linux-fsdevel@vger.kernel.org Acked-by: Serge Hallyn <serge@hallyn.com> Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com> Signed-off-by: Christian Brauner <brauner@kernel.org> Signed-off-by: Kees Cook <keescook@chromium.org> Signed-off-by: Sasha Levin <sashal@kernel.org> (cherry picked from commit 263bcebf5c2ab1fe949517225157f34015124620) Signed-off-by: Vegard Nossum <vegard.nossum@oracle.com>
2021-10-28 12:31:13 +02:00
refcount_t users; /* sync removal with load_misc_binary() */
} Node;
static DEFINE_RWLOCK(entries_lock);
static struct file_system_type bm_fs_type;
binfmt_misc: expand the register format limit to 1920 bytes The current code places a 256 byte limit on the registration format. This ends up being fairly limited when you try to do matching against a binary format like ELF: - the magic & mask formats cannot have any embedded NUL chars (string_unescape_inplace halts at the first NUL) - each escape sequence quadruples the size: \x00 is needed for NUL - trying to match bytes at the start of the file as well as further on leads to a lot of \x00 sequences in the mask - magic & mask have to be the same length (when decoded) - still need bytes for the other fields - impossible! Let's look at a concrete (and common) example: using QEMU to run MIPS ELFs. The name field uses 11 bytes "qemu-mipsel". The interp uses 20 bytes "/usr/bin/qemu-mipsel". The type & flags takes up 4 bytes. We need 7 bytes for the delimiter (usually ":"). We can skip offset. So already we're down to 107 bytes to use with the magic/mask instead of the real limit of 128 (BINPRM_BUF_SIZE). If people use shell code to register (which they do the majority of the time), they're down to ~26 possible bytes since the escape sequence must be \x##. The ELF format looks like (both 32 & 64 bit): e_ident: 16 bytes e_type: 2 bytes e_machine: 2 bytes Those 20 bytes are enough for most architectures because they have so few formats in the first place, thus they can be uniquely identified. That also means for shell users, since 20 is smaller than 26, they can sanely register a handler. But for some targets (like MIPS), we need to poke further. The ELF fields continue on: e_entry: 4 or 8 bytes e_phoff: 4 or 8 bytes e_shoff: 4 or 8 bytes e_flags: 4 bytes We only care about e_flags here as that includes the bits to identify whether the ELF is O32/N32/N64. But now we have to consume another 16 bytes (for 32 bit ELFs) or 28 bytes (for 64 bit ELFs) just to match the flags. If every byte is escaped, we send 288 more bytes to the kernel ((20 {e_ident,e_type,e_machine} + 12 {e_entry,e_phoff,e_shoff} + 4 {e_flags}) * 2 {mask,magic} * 4 {escape}) and we've clearly blown our budget. Even if we try to be clever and do the decoding ourselves (rather than relying on the kernel to process \x##), we still can't hit the mark -- string_unescape_inplace treats mask & magic as C strings so NUL cannot be embedded. That leaves us with having to pass \x00 for the 12/24 entry/phoff/shoff bytes (as those will be completely random addresses), and that is a minimum requirement of 48/96 bytes for the mask alone. Add up the rest and we blow through it (this is for 64 bit ELFs): magic: 20 {e_ident,e_type,e_machine} + 24 {e_entry,e_phoff,e_shoff} + 4 {e_flags} = 48 # ^^ See note below. mask: 20 {e_ident,e_type,e_machine} + 96 {e_entry,e_phoff,e_shoff} + 4 {e_flags} = 120 Remember above we had 107 left over, and now we're at 168. This is of course the *best* case scenario -- you'll also want to have NUL bytes in the magic & mask too to match literal zeros. Note: the reason we can use 24 in the magic is that we can work off of the fact that for bytes the mask would clobber, we can stuff any value into magic that we want. So when mask is \x00, we don't need the magic to also be \x00, it can be an unescaped raw byte like '!'. This lets us handle more formats (barely) under the current 256 limit, but that's a pretty tall hoop to force people to jump through. With all that said, let's bump the limit from 256 bytes to 1920. This way we support escaping every byte of the mask & magic field (which is 1024 bytes by themselves -- 128 * 4 * 2), and we leave plenty of room for other fields. Like long paths to the interpreter (when you have source in your /really/long/homedir/qemu/foo). Since the current code stuffs more than one structure into the same buffer, we leave a bit of space to easily round up to 2k. 1920 is just as arbitrary as 256 ;). Signed-off-by: Mike Frysinger <vapier@gentoo.org> Cc: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-10-13 15:52:03 -07:00
/*
* Max length of the register string. Determined by:
* - 7 delimiters
* - name: ~50 bytes
* - type: 1 byte
* - offset: 3 bytes (has to be smaller than BINPRM_BUF_SIZE)
* - magic: 128 bytes (512 in escaped form)
* - mask: 128 bytes (512 in escaped form)
* - interp: ~50 bytes
* - flags: 5 bytes
* Round that up a bit, and then back off to hold the internal data
* (like struct Node).
*/
#define MAX_REGISTER_LENGTH 1920
binfmt_misc: cleanup on filesystem umount [ Upstream commit 1c5976ef0f7ad76319df748ccb99a4c7ba2ba464 ] Currently, registering a new binary type pins the binfmt_misc filesystem. Specifically, this means that as long as there is at least one binary type registered the binfmt_misc filesystem survives all umounts, i.e. the superblock is not destroyed. Meaning that a umount followed by another mount will end up with the same superblock and the same binary type handlers. This is a behavior we tend to discourage for any new filesystems (apart from a few special filesystems such as e.g. configfs or debugfs). A umount operation without the filesystem being pinned - by e.g. someone holding a file descriptor to an open file - should usually result in the destruction of the superblock and all associated resources. This makes introspection easier and leads to clearly defined, simple and clean semantics. An administrator can rely on the fact that a umount will guarantee a clean slate making it possible to reinitialize a filesystem. Right now all binary types would need to be explicitly deleted before that can happen. This allows us to remove the heavy-handed calls to simple_pin_fs() and simple_release_fs() when creating and deleting binary types. This in turn allows us to replace the current brittle pinning mechanism abusing dget() which has caused a range of bugs judging from prior fixes in [2] and [3]. The additional dget() in load_misc_binary() pins the dentry but only does so for the sake to prevent ->evict_inode() from freeing the node when a user removes the binary type and kill_node() is run. Which would mean ->interpreter and ->interp_file would be freed causing a UAF. This isn't really nicely documented nor is it very clean because it relies on simple_pin_fs() pinning the filesystem as long as at least one binary type exists. Otherwise it would cause load_misc_binary() to hold on to a dentry belonging to a superblock that has been shutdown. Replace that implicit pinning with a clean and simple per-node refcount and get rid of the ugly dget() pinning. A similar mechanism exists for e.g. binderfs (cf. [4]). All the cleanup work can now be done in ->evict_inode(). In a follow-up patch we will make it possible to use binfmt_misc in sandboxes. We will use the cleaner semantics where a umount for the filesystem will cause the superblock and all resources to be deallocated. In preparation for this apply the same semantics to the initial binfmt_misc mount. Note, that this is a user-visible change and as such a uapi change but one that we can reasonably risk. We've discussed this in earlier versions of this patchset (cf. [1]). The main user and provider of binfmt_misc is systemd. Systemd provides binfmt_misc via autofs since it is configurable as a kernel module and is used by a few exotic packages and users. As such a binfmt_misc mount is triggered when /proc/sys/fs/binfmt_misc is accessed and is only provided on demand. Other autofs on demand filesystems include EFI ESP which systemd umounts if the mountpoint stays idle for a certain amount of time. This doesn't apply to the binfmt_misc autofs mount which isn't touched once it is mounted meaning this change can't accidently wipe binary type handlers without someone having explicitly unmounted binfmt_misc. After speaking to systemd folks they don't expect this change to affect them. In line with our general policy, if we see a regression for systemd or other users with this change we will switch back to the old behavior for the initial binfmt_misc mount and have binary types pin the filesystem again. But while we touch this code let's take the chance and let's improve on the status quo. [1]: https://lore.kernel.org/r/20191216091220.465626-2-laurent@vivier.eu [2]: commit 43a4f2619038 ("exec: binfmt_misc: fix race between load_misc_binary() and kill_node()" [3]: commit 83f918274e4b ("exec: binfmt_misc: shift filp_close(interp_file) from kill_node() to bm_evict_inode()") [4]: commit f0fe2c0f050d ("binder: prevent UAF for binderfs devices II") Link: https://lore.kernel.org/r/20211028103114.2849140-1-brauner@kernel.org (v1) Cc: Sargun Dhillon <sargun@sargun.me> Cc: Serge Hallyn <serge@hallyn.com> Cc: Jann Horn <jannh@google.com> Cc: Henning Schild <henning.schild@siemens.com> Cc: Andrei Vagin <avagin@gmail.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Laurent Vivier <laurent@vivier.eu> Cc: linux-fsdevel@vger.kernel.org Acked-by: Serge Hallyn <serge@hallyn.com> Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com> Signed-off-by: Christian Brauner <brauner@kernel.org> Signed-off-by: Kees Cook <keescook@chromium.org> Signed-off-by: Sasha Levin <sashal@kernel.org> (cherry picked from commit 263bcebf5c2ab1fe949517225157f34015124620) Signed-off-by: Vegard Nossum <vegard.nossum@oracle.com>
2021-10-28 12:31:13 +02:00
/**
* search_binfmt_handler - search for a binary handler for @bprm
* @misc: handle to binfmt_misc instance
* @bprm: binary for which we are looking for a handler
*
* Search for a binary type handler for @bprm in the list of registered binary
* type handlers.
*
* Return: binary type list entry on success, NULL on failure
*/
binfmt_misc: cleanup on filesystem umount [ Upstream commit 1c5976ef0f7ad76319df748ccb99a4c7ba2ba464 ] Currently, registering a new binary type pins the binfmt_misc filesystem. Specifically, this means that as long as there is at least one binary type registered the binfmt_misc filesystem survives all umounts, i.e. the superblock is not destroyed. Meaning that a umount followed by another mount will end up with the same superblock and the same binary type handlers. This is a behavior we tend to discourage for any new filesystems (apart from a few special filesystems such as e.g. configfs or debugfs). A umount operation without the filesystem being pinned - by e.g. someone holding a file descriptor to an open file - should usually result in the destruction of the superblock and all associated resources. This makes introspection easier and leads to clearly defined, simple and clean semantics. An administrator can rely on the fact that a umount will guarantee a clean slate making it possible to reinitialize a filesystem. Right now all binary types would need to be explicitly deleted before that can happen. This allows us to remove the heavy-handed calls to simple_pin_fs() and simple_release_fs() when creating and deleting binary types. This in turn allows us to replace the current brittle pinning mechanism abusing dget() which has caused a range of bugs judging from prior fixes in [2] and [3]. The additional dget() in load_misc_binary() pins the dentry but only does so for the sake to prevent ->evict_inode() from freeing the node when a user removes the binary type and kill_node() is run. Which would mean ->interpreter and ->interp_file would be freed causing a UAF. This isn't really nicely documented nor is it very clean because it relies on simple_pin_fs() pinning the filesystem as long as at least one binary type exists. Otherwise it would cause load_misc_binary() to hold on to a dentry belonging to a superblock that has been shutdown. Replace that implicit pinning with a clean and simple per-node refcount and get rid of the ugly dget() pinning. A similar mechanism exists for e.g. binderfs (cf. [4]). All the cleanup work can now be done in ->evict_inode(). In a follow-up patch we will make it possible to use binfmt_misc in sandboxes. We will use the cleaner semantics where a umount for the filesystem will cause the superblock and all resources to be deallocated. In preparation for this apply the same semantics to the initial binfmt_misc mount. Note, that this is a user-visible change and as such a uapi change but one that we can reasonably risk. We've discussed this in earlier versions of this patchset (cf. [1]). The main user and provider of binfmt_misc is systemd. Systemd provides binfmt_misc via autofs since it is configurable as a kernel module and is used by a few exotic packages and users. As such a binfmt_misc mount is triggered when /proc/sys/fs/binfmt_misc is accessed and is only provided on demand. Other autofs on demand filesystems include EFI ESP which systemd umounts if the mountpoint stays idle for a certain amount of time. This doesn't apply to the binfmt_misc autofs mount which isn't touched once it is mounted meaning this change can't accidently wipe binary type handlers without someone having explicitly unmounted binfmt_misc. After speaking to systemd folks they don't expect this change to affect them. In line with our general policy, if we see a regression for systemd or other users with this change we will switch back to the old behavior for the initial binfmt_misc mount and have binary types pin the filesystem again. But while we touch this code let's take the chance and let's improve on the status quo. [1]: https://lore.kernel.org/r/20191216091220.465626-2-laurent@vivier.eu [2]: commit 43a4f2619038 ("exec: binfmt_misc: fix race between load_misc_binary() and kill_node()" [3]: commit 83f918274e4b ("exec: binfmt_misc: shift filp_close(interp_file) from kill_node() to bm_evict_inode()") [4]: commit f0fe2c0f050d ("binder: prevent UAF for binderfs devices II") Link: https://lore.kernel.org/r/20211028103114.2849140-1-brauner@kernel.org (v1) Cc: Sargun Dhillon <sargun@sargun.me> Cc: Serge Hallyn <serge@hallyn.com> Cc: Jann Horn <jannh@google.com> Cc: Henning Schild <henning.schild@siemens.com> Cc: Andrei Vagin <avagin@gmail.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Laurent Vivier <laurent@vivier.eu> Cc: linux-fsdevel@vger.kernel.org Acked-by: Serge Hallyn <serge@hallyn.com> Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com> Signed-off-by: Christian Brauner <brauner@kernel.org> Signed-off-by: Kees Cook <keescook@chromium.org> Signed-off-by: Sasha Levin <sashal@kernel.org> (cherry picked from commit 263bcebf5c2ab1fe949517225157f34015124620) Signed-off-by: Vegard Nossum <vegard.nossum@oracle.com>
2021-10-28 12:31:13 +02:00
static Node *search_binfmt_handler(struct linux_binprm *bprm)
{
char *p = strrchr(bprm->interp, '.');
binfmt_misc: cleanup on filesystem umount [ Upstream commit 1c5976ef0f7ad76319df748ccb99a4c7ba2ba464 ] Currently, registering a new binary type pins the binfmt_misc filesystem. Specifically, this means that as long as there is at least one binary type registered the binfmt_misc filesystem survives all umounts, i.e. the superblock is not destroyed. Meaning that a umount followed by another mount will end up with the same superblock and the same binary type handlers. This is a behavior we tend to discourage for any new filesystems (apart from a few special filesystems such as e.g. configfs or debugfs). A umount operation without the filesystem being pinned - by e.g. someone holding a file descriptor to an open file - should usually result in the destruction of the superblock and all associated resources. This makes introspection easier and leads to clearly defined, simple and clean semantics. An administrator can rely on the fact that a umount will guarantee a clean slate making it possible to reinitialize a filesystem. Right now all binary types would need to be explicitly deleted before that can happen. This allows us to remove the heavy-handed calls to simple_pin_fs() and simple_release_fs() when creating and deleting binary types. This in turn allows us to replace the current brittle pinning mechanism abusing dget() which has caused a range of bugs judging from prior fixes in [2] and [3]. The additional dget() in load_misc_binary() pins the dentry but only does so for the sake to prevent ->evict_inode() from freeing the node when a user removes the binary type and kill_node() is run. Which would mean ->interpreter and ->interp_file would be freed causing a UAF. This isn't really nicely documented nor is it very clean because it relies on simple_pin_fs() pinning the filesystem as long as at least one binary type exists. Otherwise it would cause load_misc_binary() to hold on to a dentry belonging to a superblock that has been shutdown. Replace that implicit pinning with a clean and simple per-node refcount and get rid of the ugly dget() pinning. A similar mechanism exists for e.g. binderfs (cf. [4]). All the cleanup work can now be done in ->evict_inode(). In a follow-up patch we will make it possible to use binfmt_misc in sandboxes. We will use the cleaner semantics where a umount for the filesystem will cause the superblock and all resources to be deallocated. In preparation for this apply the same semantics to the initial binfmt_misc mount. Note, that this is a user-visible change and as such a uapi change but one that we can reasonably risk. We've discussed this in earlier versions of this patchset (cf. [1]). The main user and provider of binfmt_misc is systemd. Systemd provides binfmt_misc via autofs since it is configurable as a kernel module and is used by a few exotic packages and users. As such a binfmt_misc mount is triggered when /proc/sys/fs/binfmt_misc is accessed and is only provided on demand. Other autofs on demand filesystems include EFI ESP which systemd umounts if the mountpoint stays idle for a certain amount of time. This doesn't apply to the binfmt_misc autofs mount which isn't touched once it is mounted meaning this change can't accidently wipe binary type handlers without someone having explicitly unmounted binfmt_misc. After speaking to systemd folks they don't expect this change to affect them. In line with our general policy, if we see a regression for systemd or other users with this change we will switch back to the old behavior for the initial binfmt_misc mount and have binary types pin the filesystem again. But while we touch this code let's take the chance and let's improve on the status quo. [1]: https://lore.kernel.org/r/20191216091220.465626-2-laurent@vivier.eu [2]: commit 43a4f2619038 ("exec: binfmt_misc: fix race between load_misc_binary() and kill_node()" [3]: commit 83f918274e4b ("exec: binfmt_misc: shift filp_close(interp_file) from kill_node() to bm_evict_inode()") [4]: commit f0fe2c0f050d ("binder: prevent UAF for binderfs devices II") Link: https://lore.kernel.org/r/20211028103114.2849140-1-brauner@kernel.org (v1) Cc: Sargun Dhillon <sargun@sargun.me> Cc: Serge Hallyn <serge@hallyn.com> Cc: Jann Horn <jannh@google.com> Cc: Henning Schild <henning.schild@siemens.com> Cc: Andrei Vagin <avagin@gmail.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Laurent Vivier <laurent@vivier.eu> Cc: linux-fsdevel@vger.kernel.org Acked-by: Serge Hallyn <serge@hallyn.com> Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com> Signed-off-by: Christian Brauner <brauner@kernel.org> Signed-off-by: Kees Cook <keescook@chromium.org> Signed-off-by: Sasha Levin <sashal@kernel.org> (cherry picked from commit 263bcebf5c2ab1fe949517225157f34015124620) Signed-off-by: Vegard Nossum <vegard.nossum@oracle.com>
2021-10-28 12:31:13 +02:00
Node *e;
binfmt_misc: add comments & debug logs When trying to develop a custom format handler, the errors returned all effectively get bucketed as EINVAL with no kernel messages. The other errors (ENOMEM/EFAULT) are internal/obvious and basic. Thus any time a bad handler is rejected, the developer has to walk the dense code and try to guess where it went wrong. Needing to dive into kernel code is itself a fairly high barrier for a lot of people. To improve this situation, let's deploy extensive pr_debug markers at logical parse points, and add comments to the dense parsing logic. It let's you see exactly where the parsing aborts, the string the kernel received (useful when dealing with shell code), how it translated the buffers to binary data, and how it will apply the mask at runtime. Some example output: $ echo ':qemu-foo:M::\x7fELF\xAD\xAD\x01\x00:\xff\xff\xff\xff\xff\x00\xff\x00:/usr/bin/qemu-foo:POC' > register $ dmesg binfmt_misc: register: received 92 bytes binfmt_misc: register: delim: 0x3a {:} binfmt_misc: register: name: {qemu-foo} binfmt_misc: register: type: M (magic) binfmt_misc: register: offset: 0x0 binfmt_misc: register: magic[raw]: 5c 78 37 66 45 4c 46 5c 78 41 44 5c 78 41 44 5c \x7fELF\xAD\xAD\ binfmt_misc: register: magic[raw]: 78 30 31 5c 78 30 30 00 x01\x00. binfmt_misc: register: mask[raw]: 5c 78 66 66 5c 78 66 66 5c 78 66 66 5c 78 66 66 \xff\xff\xff\xff binfmt_misc: register: mask[raw]: 5c 78 66 66 5c 78 30 30 5c 78 66 66 5c 78 30 30 \xff\x00\xff\x00 binfmt_misc: register: mask[raw]: 00 . binfmt_misc: register: magic/mask length: 8 binfmt_misc: register: magic[decoded]: 7f 45 4c 46 ad ad 01 00 .ELF.... binfmt_misc: register: mask[decoded]: ff ff ff ff ff 00 ff 00 ........ binfmt_misc: register: magic[masked]: 7f 45 4c 46 ad 00 01 00 .ELF.... binfmt_misc: register: interpreter: {/usr/bin/qemu-foo} binfmt_misc: register: flag: P (preserve argv0) binfmt_misc: register: flag: O (open binary) binfmt_misc: register: flag: C (preserve creds) The [raw] lines show us exactly what was received from userspace. The lines after that show us how the kernel has decoded things. Signed-off-by: Mike Frysinger <vapier@gentoo.org> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Joe Perches <joe@perches.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-12-10 15:52:08 -08:00
/* Walk all the registered handlers. */
binfmt_misc: cleanup on filesystem umount [ Upstream commit 1c5976ef0f7ad76319df748ccb99a4c7ba2ba464 ] Currently, registering a new binary type pins the binfmt_misc filesystem. Specifically, this means that as long as there is at least one binary type registered the binfmt_misc filesystem survives all umounts, i.e. the superblock is not destroyed. Meaning that a umount followed by another mount will end up with the same superblock and the same binary type handlers. This is a behavior we tend to discourage for any new filesystems (apart from a few special filesystems such as e.g. configfs or debugfs). A umount operation without the filesystem being pinned - by e.g. someone holding a file descriptor to an open file - should usually result in the destruction of the superblock and all associated resources. This makes introspection easier and leads to clearly defined, simple and clean semantics. An administrator can rely on the fact that a umount will guarantee a clean slate making it possible to reinitialize a filesystem. Right now all binary types would need to be explicitly deleted before that can happen. This allows us to remove the heavy-handed calls to simple_pin_fs() and simple_release_fs() when creating and deleting binary types. This in turn allows us to replace the current brittle pinning mechanism abusing dget() which has caused a range of bugs judging from prior fixes in [2] and [3]. The additional dget() in load_misc_binary() pins the dentry but only does so for the sake to prevent ->evict_inode() from freeing the node when a user removes the binary type and kill_node() is run. Which would mean ->interpreter and ->interp_file would be freed causing a UAF. This isn't really nicely documented nor is it very clean because it relies on simple_pin_fs() pinning the filesystem as long as at least one binary type exists. Otherwise it would cause load_misc_binary() to hold on to a dentry belonging to a superblock that has been shutdown. Replace that implicit pinning with a clean and simple per-node refcount and get rid of the ugly dget() pinning. A similar mechanism exists for e.g. binderfs (cf. [4]). All the cleanup work can now be done in ->evict_inode(). In a follow-up patch we will make it possible to use binfmt_misc in sandboxes. We will use the cleaner semantics where a umount for the filesystem will cause the superblock and all resources to be deallocated. In preparation for this apply the same semantics to the initial binfmt_misc mount. Note, that this is a user-visible change and as such a uapi change but one that we can reasonably risk. We've discussed this in earlier versions of this patchset (cf. [1]). The main user and provider of binfmt_misc is systemd. Systemd provides binfmt_misc via autofs since it is configurable as a kernel module and is used by a few exotic packages and users. As such a binfmt_misc mount is triggered when /proc/sys/fs/binfmt_misc is accessed and is only provided on demand. Other autofs on demand filesystems include EFI ESP which systemd umounts if the mountpoint stays idle for a certain amount of time. This doesn't apply to the binfmt_misc autofs mount which isn't touched once it is mounted meaning this change can't accidently wipe binary type handlers without someone having explicitly unmounted binfmt_misc. After speaking to systemd folks they don't expect this change to affect them. In line with our general policy, if we see a regression for systemd or other users with this change we will switch back to the old behavior for the initial binfmt_misc mount and have binary types pin the filesystem again. But while we touch this code let's take the chance and let's improve on the status quo. [1]: https://lore.kernel.org/r/20191216091220.465626-2-laurent@vivier.eu [2]: commit 43a4f2619038 ("exec: binfmt_misc: fix race between load_misc_binary() and kill_node()" [3]: commit 83f918274e4b ("exec: binfmt_misc: shift filp_close(interp_file) from kill_node() to bm_evict_inode()") [4]: commit f0fe2c0f050d ("binder: prevent UAF for binderfs devices II") Link: https://lore.kernel.org/r/20211028103114.2849140-1-brauner@kernel.org (v1) Cc: Sargun Dhillon <sargun@sargun.me> Cc: Serge Hallyn <serge@hallyn.com> Cc: Jann Horn <jannh@google.com> Cc: Henning Schild <henning.schild@siemens.com> Cc: Andrei Vagin <avagin@gmail.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Laurent Vivier <laurent@vivier.eu> Cc: linux-fsdevel@vger.kernel.org Acked-by: Serge Hallyn <serge@hallyn.com> Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com> Signed-off-by: Christian Brauner <brauner@kernel.org> Signed-off-by: Kees Cook <keescook@chromium.org> Signed-off-by: Sasha Levin <sashal@kernel.org> (cherry picked from commit 263bcebf5c2ab1fe949517225157f34015124620) Signed-off-by: Vegard Nossum <vegard.nossum@oracle.com>
2021-10-28 12:31:13 +02:00
list_for_each_entry(e, &entries, list) {
char *s;
int j;
binfmt_misc: add comments & debug logs When trying to develop a custom format handler, the errors returned all effectively get bucketed as EINVAL with no kernel messages. The other errors (ENOMEM/EFAULT) are internal/obvious and basic. Thus any time a bad handler is rejected, the developer has to walk the dense code and try to guess where it went wrong. Needing to dive into kernel code is itself a fairly high barrier for a lot of people. To improve this situation, let's deploy extensive pr_debug markers at logical parse points, and add comments to the dense parsing logic. It let's you see exactly where the parsing aborts, the string the kernel received (useful when dealing with shell code), how it translated the buffers to binary data, and how it will apply the mask at runtime. Some example output: $ echo ':qemu-foo:M::\x7fELF\xAD\xAD\x01\x00:\xff\xff\xff\xff\xff\x00\xff\x00:/usr/bin/qemu-foo:POC' > register $ dmesg binfmt_misc: register: received 92 bytes binfmt_misc: register: delim: 0x3a {:} binfmt_misc: register: name: {qemu-foo} binfmt_misc: register: type: M (magic) binfmt_misc: register: offset: 0x0 binfmt_misc: register: magic[raw]: 5c 78 37 66 45 4c 46 5c 78 41 44 5c 78 41 44 5c \x7fELF\xAD\xAD\ binfmt_misc: register: magic[raw]: 78 30 31 5c 78 30 30 00 x01\x00. binfmt_misc: register: mask[raw]: 5c 78 66 66 5c 78 66 66 5c 78 66 66 5c 78 66 66 \xff\xff\xff\xff binfmt_misc: register: mask[raw]: 5c 78 66 66 5c 78 30 30 5c 78 66 66 5c 78 30 30 \xff\x00\xff\x00 binfmt_misc: register: mask[raw]: 00 . binfmt_misc: register: magic/mask length: 8 binfmt_misc: register: magic[decoded]: 7f 45 4c 46 ad ad 01 00 .ELF.... binfmt_misc: register: mask[decoded]: ff ff ff ff ff 00 ff 00 ........ binfmt_misc: register: magic[masked]: 7f 45 4c 46 ad 00 01 00 .ELF.... binfmt_misc: register: interpreter: {/usr/bin/qemu-foo} binfmt_misc: register: flag: P (preserve argv0) binfmt_misc: register: flag: O (open binary) binfmt_misc: register: flag: C (preserve creds) The [raw] lines show us exactly what was received from userspace. The lines after that show us how the kernel has decoded things. Signed-off-by: Mike Frysinger <vapier@gentoo.org> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Joe Perches <joe@perches.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-12-10 15:52:08 -08:00
/* Make sure this one is currently enabled. */
if (!test_bit(Enabled, &e->flags))
continue;
binfmt_misc: add comments & debug logs When trying to develop a custom format handler, the errors returned all effectively get bucketed as EINVAL with no kernel messages. The other errors (ENOMEM/EFAULT) are internal/obvious and basic. Thus any time a bad handler is rejected, the developer has to walk the dense code and try to guess where it went wrong. Needing to dive into kernel code is itself a fairly high barrier for a lot of people. To improve this situation, let's deploy extensive pr_debug markers at logical parse points, and add comments to the dense parsing logic. It let's you see exactly where the parsing aborts, the string the kernel received (useful when dealing with shell code), how it translated the buffers to binary data, and how it will apply the mask at runtime. Some example output: $ echo ':qemu-foo:M::\x7fELF\xAD\xAD\x01\x00:\xff\xff\xff\xff\xff\x00\xff\x00:/usr/bin/qemu-foo:POC' > register $ dmesg binfmt_misc: register: received 92 bytes binfmt_misc: register: delim: 0x3a {:} binfmt_misc: register: name: {qemu-foo} binfmt_misc: register: type: M (magic) binfmt_misc: register: offset: 0x0 binfmt_misc: register: magic[raw]: 5c 78 37 66 45 4c 46 5c 78 41 44 5c 78 41 44 5c \x7fELF\xAD\xAD\ binfmt_misc: register: magic[raw]: 78 30 31 5c 78 30 30 00 x01\x00. binfmt_misc: register: mask[raw]: 5c 78 66 66 5c 78 66 66 5c 78 66 66 5c 78 66 66 \xff\xff\xff\xff binfmt_misc: register: mask[raw]: 5c 78 66 66 5c 78 30 30 5c 78 66 66 5c 78 30 30 \xff\x00\xff\x00 binfmt_misc: register: mask[raw]: 00 . binfmt_misc: register: magic/mask length: 8 binfmt_misc: register: magic[decoded]: 7f 45 4c 46 ad ad 01 00 .ELF.... binfmt_misc: register: mask[decoded]: ff ff ff ff ff 00 ff 00 ........ binfmt_misc: register: magic[masked]: 7f 45 4c 46 ad 00 01 00 .ELF.... binfmt_misc: register: interpreter: {/usr/bin/qemu-foo} binfmt_misc: register: flag: P (preserve argv0) binfmt_misc: register: flag: O (open binary) binfmt_misc: register: flag: C (preserve creds) The [raw] lines show us exactly what was received from userspace. The lines after that show us how the kernel has decoded things. Signed-off-by: Mike Frysinger <vapier@gentoo.org> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Joe Perches <joe@perches.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-12-10 15:52:08 -08:00
/* Do matching based on extension if applicable. */
if (!test_bit(Magic, &e->flags)) {
if (p && !strcmp(e->magic, p + 1))
return e;
continue;
}
binfmt_misc: add comments & debug logs When trying to develop a custom format handler, the errors returned all effectively get bucketed as EINVAL with no kernel messages. The other errors (ENOMEM/EFAULT) are internal/obvious and basic. Thus any time a bad handler is rejected, the developer has to walk the dense code and try to guess where it went wrong. Needing to dive into kernel code is itself a fairly high barrier for a lot of people. To improve this situation, let's deploy extensive pr_debug markers at logical parse points, and add comments to the dense parsing logic. It let's you see exactly where the parsing aborts, the string the kernel received (useful when dealing with shell code), how it translated the buffers to binary data, and how it will apply the mask at runtime. Some example output: $ echo ':qemu-foo:M::\x7fELF\xAD\xAD\x01\x00:\xff\xff\xff\xff\xff\x00\xff\x00:/usr/bin/qemu-foo:POC' > register $ dmesg binfmt_misc: register: received 92 bytes binfmt_misc: register: delim: 0x3a {:} binfmt_misc: register: name: {qemu-foo} binfmt_misc: register: type: M (magic) binfmt_misc: register: offset: 0x0 binfmt_misc: register: magic[raw]: 5c 78 37 66 45 4c 46 5c 78 41 44 5c 78 41 44 5c \x7fELF\xAD\xAD\ binfmt_misc: register: magic[raw]: 78 30 31 5c 78 30 30 00 x01\x00. binfmt_misc: register: mask[raw]: 5c 78 66 66 5c 78 66 66 5c 78 66 66 5c 78 66 66 \xff\xff\xff\xff binfmt_misc: register: mask[raw]: 5c 78 66 66 5c 78 30 30 5c 78 66 66 5c 78 30 30 \xff\x00\xff\x00 binfmt_misc: register: mask[raw]: 00 . binfmt_misc: register: magic/mask length: 8 binfmt_misc: register: magic[decoded]: 7f 45 4c 46 ad ad 01 00 .ELF.... binfmt_misc: register: mask[decoded]: ff ff ff ff ff 00 ff 00 ........ binfmt_misc: register: magic[masked]: 7f 45 4c 46 ad 00 01 00 .ELF.... binfmt_misc: register: interpreter: {/usr/bin/qemu-foo} binfmt_misc: register: flag: P (preserve argv0) binfmt_misc: register: flag: O (open binary) binfmt_misc: register: flag: C (preserve creds) The [raw] lines show us exactly what was received from userspace. The lines after that show us how the kernel has decoded things. Signed-off-by: Mike Frysinger <vapier@gentoo.org> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Joe Perches <joe@perches.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-12-10 15:52:08 -08:00
/* Do matching based on magic & mask. */
s = bprm->buf + e->offset;
if (e->mask) {
for (j = 0; j < e->size; j++)
if ((*s++ ^ e->magic[j]) & e->mask[j])
break;
} else {
for (j = 0; j < e->size; j++)
if ((*s++ ^ e->magic[j]))
break;
}
if (j == e->size)
return e;
}
binfmt_misc: cleanup on filesystem umount [ Upstream commit 1c5976ef0f7ad76319df748ccb99a4c7ba2ba464 ] Currently, registering a new binary type pins the binfmt_misc filesystem. Specifically, this means that as long as there is at least one binary type registered the binfmt_misc filesystem survives all umounts, i.e. the superblock is not destroyed. Meaning that a umount followed by another mount will end up with the same superblock and the same binary type handlers. This is a behavior we tend to discourage for any new filesystems (apart from a few special filesystems such as e.g. configfs or debugfs). A umount operation without the filesystem being pinned - by e.g. someone holding a file descriptor to an open file - should usually result in the destruction of the superblock and all associated resources. This makes introspection easier and leads to clearly defined, simple and clean semantics. An administrator can rely on the fact that a umount will guarantee a clean slate making it possible to reinitialize a filesystem. Right now all binary types would need to be explicitly deleted before that can happen. This allows us to remove the heavy-handed calls to simple_pin_fs() and simple_release_fs() when creating and deleting binary types. This in turn allows us to replace the current brittle pinning mechanism abusing dget() which has caused a range of bugs judging from prior fixes in [2] and [3]. The additional dget() in load_misc_binary() pins the dentry but only does so for the sake to prevent ->evict_inode() from freeing the node when a user removes the binary type and kill_node() is run. Which would mean ->interpreter and ->interp_file would be freed causing a UAF. This isn't really nicely documented nor is it very clean because it relies on simple_pin_fs() pinning the filesystem as long as at least one binary type exists. Otherwise it would cause load_misc_binary() to hold on to a dentry belonging to a superblock that has been shutdown. Replace that implicit pinning with a clean and simple per-node refcount and get rid of the ugly dget() pinning. A similar mechanism exists for e.g. binderfs (cf. [4]). All the cleanup work can now be done in ->evict_inode(). In a follow-up patch we will make it possible to use binfmt_misc in sandboxes. We will use the cleaner semantics where a umount for the filesystem will cause the superblock and all resources to be deallocated. In preparation for this apply the same semantics to the initial binfmt_misc mount. Note, that this is a user-visible change and as such a uapi change but one that we can reasonably risk. We've discussed this in earlier versions of this patchset (cf. [1]). The main user and provider of binfmt_misc is systemd. Systemd provides binfmt_misc via autofs since it is configurable as a kernel module and is used by a few exotic packages and users. As such a binfmt_misc mount is triggered when /proc/sys/fs/binfmt_misc is accessed and is only provided on demand. Other autofs on demand filesystems include EFI ESP which systemd umounts if the mountpoint stays idle for a certain amount of time. This doesn't apply to the binfmt_misc autofs mount which isn't touched once it is mounted meaning this change can't accidently wipe binary type handlers without someone having explicitly unmounted binfmt_misc. After speaking to systemd folks they don't expect this change to affect them. In line with our general policy, if we see a regression for systemd or other users with this change we will switch back to the old behavior for the initial binfmt_misc mount and have binary types pin the filesystem again. But while we touch this code let's take the chance and let's improve on the status quo. [1]: https://lore.kernel.org/r/20191216091220.465626-2-laurent@vivier.eu [2]: commit 43a4f2619038 ("exec: binfmt_misc: fix race between load_misc_binary() and kill_node()" [3]: commit 83f918274e4b ("exec: binfmt_misc: shift filp_close(interp_file) from kill_node() to bm_evict_inode()") [4]: commit f0fe2c0f050d ("binder: prevent UAF for binderfs devices II") Link: https://lore.kernel.org/r/20211028103114.2849140-1-brauner@kernel.org (v1) Cc: Sargun Dhillon <sargun@sargun.me> Cc: Serge Hallyn <serge@hallyn.com> Cc: Jann Horn <jannh@google.com> Cc: Henning Schild <henning.schild@siemens.com> Cc: Andrei Vagin <avagin@gmail.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Laurent Vivier <laurent@vivier.eu> Cc: linux-fsdevel@vger.kernel.org Acked-by: Serge Hallyn <serge@hallyn.com> Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com> Signed-off-by: Christian Brauner <brauner@kernel.org> Signed-off-by: Kees Cook <keescook@chromium.org> Signed-off-by: Sasha Levin <sashal@kernel.org> (cherry picked from commit 263bcebf5c2ab1fe949517225157f34015124620) Signed-off-by: Vegard Nossum <vegard.nossum@oracle.com>
2021-10-28 12:31:13 +02:00
return NULL;
}
binfmt_misc: cleanup on filesystem umount [ Upstream commit 1c5976ef0f7ad76319df748ccb99a4c7ba2ba464 ] Currently, registering a new binary type pins the binfmt_misc filesystem. Specifically, this means that as long as there is at least one binary type registered the binfmt_misc filesystem survives all umounts, i.e. the superblock is not destroyed. Meaning that a umount followed by another mount will end up with the same superblock and the same binary type handlers. This is a behavior we tend to discourage for any new filesystems (apart from a few special filesystems such as e.g. configfs or debugfs). A umount operation without the filesystem being pinned - by e.g. someone holding a file descriptor to an open file - should usually result in the destruction of the superblock and all associated resources. This makes introspection easier and leads to clearly defined, simple and clean semantics. An administrator can rely on the fact that a umount will guarantee a clean slate making it possible to reinitialize a filesystem. Right now all binary types would need to be explicitly deleted before that can happen. This allows us to remove the heavy-handed calls to simple_pin_fs() and simple_release_fs() when creating and deleting binary types. This in turn allows us to replace the current brittle pinning mechanism abusing dget() which has caused a range of bugs judging from prior fixes in [2] and [3]. The additional dget() in load_misc_binary() pins the dentry but only does so for the sake to prevent ->evict_inode() from freeing the node when a user removes the binary type and kill_node() is run. Which would mean ->interpreter and ->interp_file would be freed causing a UAF. This isn't really nicely documented nor is it very clean because it relies on simple_pin_fs() pinning the filesystem as long as at least one binary type exists. Otherwise it would cause load_misc_binary() to hold on to a dentry belonging to a superblock that has been shutdown. Replace that implicit pinning with a clean and simple per-node refcount and get rid of the ugly dget() pinning. A similar mechanism exists for e.g. binderfs (cf. [4]). All the cleanup work can now be done in ->evict_inode(). In a follow-up patch we will make it possible to use binfmt_misc in sandboxes. We will use the cleaner semantics where a umount for the filesystem will cause the superblock and all resources to be deallocated. In preparation for this apply the same semantics to the initial binfmt_misc mount. Note, that this is a user-visible change and as such a uapi change but one that we can reasonably risk. We've discussed this in earlier versions of this patchset (cf. [1]). The main user and provider of binfmt_misc is systemd. Systemd provides binfmt_misc via autofs since it is configurable as a kernel module and is used by a few exotic packages and users. As such a binfmt_misc mount is triggered when /proc/sys/fs/binfmt_misc is accessed and is only provided on demand. Other autofs on demand filesystems include EFI ESP which systemd umounts if the mountpoint stays idle for a certain amount of time. This doesn't apply to the binfmt_misc autofs mount which isn't touched once it is mounted meaning this change can't accidently wipe binary type handlers without someone having explicitly unmounted binfmt_misc. After speaking to systemd folks they don't expect this change to affect them. In line with our general policy, if we see a regression for systemd or other users with this change we will switch back to the old behavior for the initial binfmt_misc mount and have binary types pin the filesystem again. But while we touch this code let's take the chance and let's improve on the status quo. [1]: https://lore.kernel.org/r/20191216091220.465626-2-laurent@vivier.eu [2]: commit 43a4f2619038 ("exec: binfmt_misc: fix race between load_misc_binary() and kill_node()" [3]: commit 83f918274e4b ("exec: binfmt_misc: shift filp_close(interp_file) from kill_node() to bm_evict_inode()") [4]: commit f0fe2c0f050d ("binder: prevent UAF for binderfs devices II") Link: https://lore.kernel.org/r/20211028103114.2849140-1-brauner@kernel.org (v1) Cc: Sargun Dhillon <sargun@sargun.me> Cc: Serge Hallyn <serge@hallyn.com> Cc: Jann Horn <jannh@google.com> Cc: Henning Schild <henning.schild@siemens.com> Cc: Andrei Vagin <avagin@gmail.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Laurent Vivier <laurent@vivier.eu> Cc: linux-fsdevel@vger.kernel.org Acked-by: Serge Hallyn <serge@hallyn.com> Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com> Signed-off-by: Christian Brauner <brauner@kernel.org> Signed-off-by: Kees Cook <keescook@chromium.org> Signed-off-by: Sasha Levin <sashal@kernel.org> (cherry picked from commit 263bcebf5c2ab1fe949517225157f34015124620) Signed-off-by: Vegard Nossum <vegard.nossum@oracle.com>
2021-10-28 12:31:13 +02:00
/**
* get_binfmt_handler - try to find a binary type handler
* @misc: handle to binfmt_misc instance
* @bprm: binary for which we are looking for a handler
*
* Try to find a binfmt handler for the binary type. If one is found take a
* reference to protect against removal via bm_{entry,status}_write().
*
* Return: binary type list entry on success, NULL on failure
*/
static Node *get_binfmt_handler(struct linux_binprm *bprm)
{
Node *e;
read_lock(&entries_lock);
e = search_binfmt_handler(bprm);
if (e)
refcount_inc(&e->users);
read_unlock(&entries_lock);
return e;
}
/**
* put_binfmt_handler - put binary handler node
* @e: node to put
*
* Free node syncing with load_misc_binary() and defer final free to
* load_misc_binary() in case it is using the binary type handler we were
* requested to remove.
*/
static void put_binfmt_handler(Node *e)
{
if (refcount_dec_and_test(&e->users)) {
if (e->flags & MISC_FMT_OPEN_FILE)
filp_close(e->interp_file, NULL);
kfree(e);
}
}
/*
* the loader itself
*/
static int load_misc_binary(struct linux_binprm *bprm)
{
Node *fmt;
struct file *interp_file = NULL;
int retval;
int fd_binary = -1;
retval = -ENOEXEC;
if (!enabled)
return retval;
binfmt_misc: cleanup on filesystem umount [ Upstream commit 1c5976ef0f7ad76319df748ccb99a4c7ba2ba464 ] Currently, registering a new binary type pins the binfmt_misc filesystem. Specifically, this means that as long as there is at least one binary type registered the binfmt_misc filesystem survives all umounts, i.e. the superblock is not destroyed. Meaning that a umount followed by another mount will end up with the same superblock and the same binary type handlers. This is a behavior we tend to discourage for any new filesystems (apart from a few special filesystems such as e.g. configfs or debugfs). A umount operation without the filesystem being pinned - by e.g. someone holding a file descriptor to an open file - should usually result in the destruction of the superblock and all associated resources. This makes introspection easier and leads to clearly defined, simple and clean semantics. An administrator can rely on the fact that a umount will guarantee a clean slate making it possible to reinitialize a filesystem. Right now all binary types would need to be explicitly deleted before that can happen. This allows us to remove the heavy-handed calls to simple_pin_fs() and simple_release_fs() when creating and deleting binary types. This in turn allows us to replace the current brittle pinning mechanism abusing dget() which has caused a range of bugs judging from prior fixes in [2] and [3]. The additional dget() in load_misc_binary() pins the dentry but only does so for the sake to prevent ->evict_inode() from freeing the node when a user removes the binary type and kill_node() is run. Which would mean ->interpreter and ->interp_file would be freed causing a UAF. This isn't really nicely documented nor is it very clean because it relies on simple_pin_fs() pinning the filesystem as long as at least one binary type exists. Otherwise it would cause load_misc_binary() to hold on to a dentry belonging to a superblock that has been shutdown. Replace that implicit pinning with a clean and simple per-node refcount and get rid of the ugly dget() pinning. A similar mechanism exists for e.g. binderfs (cf. [4]). All the cleanup work can now be done in ->evict_inode(). In a follow-up patch we will make it possible to use binfmt_misc in sandboxes. We will use the cleaner semantics where a umount for the filesystem will cause the superblock and all resources to be deallocated. In preparation for this apply the same semantics to the initial binfmt_misc mount. Note, that this is a user-visible change and as such a uapi change but one that we can reasonably risk. We've discussed this in earlier versions of this patchset (cf. [1]). The main user and provider of binfmt_misc is systemd. Systemd provides binfmt_misc via autofs since it is configurable as a kernel module and is used by a few exotic packages and users. As such a binfmt_misc mount is triggered when /proc/sys/fs/binfmt_misc is accessed and is only provided on demand. Other autofs on demand filesystems include EFI ESP which systemd umounts if the mountpoint stays idle for a certain amount of time. This doesn't apply to the binfmt_misc autofs mount which isn't touched once it is mounted meaning this change can't accidently wipe binary type handlers without someone having explicitly unmounted binfmt_misc. After speaking to systemd folks they don't expect this change to affect them. In line with our general policy, if we see a regression for systemd or other users with this change we will switch back to the old behavior for the initial binfmt_misc mount and have binary types pin the filesystem again. But while we touch this code let's take the chance and let's improve on the status quo. [1]: https://lore.kernel.org/r/20191216091220.465626-2-laurent@vivier.eu [2]: commit 43a4f2619038 ("exec: binfmt_misc: fix race between load_misc_binary() and kill_node()" [3]: commit 83f918274e4b ("exec: binfmt_misc: shift filp_close(interp_file) from kill_node() to bm_evict_inode()") [4]: commit f0fe2c0f050d ("binder: prevent UAF for binderfs devices II") Link: https://lore.kernel.org/r/20211028103114.2849140-1-brauner@kernel.org (v1) Cc: Sargun Dhillon <sargun@sargun.me> Cc: Serge Hallyn <serge@hallyn.com> Cc: Jann Horn <jannh@google.com> Cc: Henning Schild <henning.schild@siemens.com> Cc: Andrei Vagin <avagin@gmail.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Laurent Vivier <laurent@vivier.eu> Cc: linux-fsdevel@vger.kernel.org Acked-by: Serge Hallyn <serge@hallyn.com> Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com> Signed-off-by: Christian Brauner <brauner@kernel.org> Signed-off-by: Kees Cook <keescook@chromium.org> Signed-off-by: Sasha Levin <sashal@kernel.org> (cherry picked from commit 263bcebf5c2ab1fe949517225157f34015124620) Signed-off-by: Vegard Nossum <vegard.nossum@oracle.com>
2021-10-28 12:31:13 +02:00
fmt = get_binfmt_handler(bprm);
if (!fmt)
return retval;
syscalls: implement execveat() system call This patchset adds execveat(2) for x86, and is derived from Meredydd Luff's patch from Sept 2012 (https://lkml.org/lkml/2012/9/11/528). The primary aim of adding an execveat syscall is to allow an implementation of fexecve(3) that does not rely on the /proc filesystem, at least for executables (rather than scripts). The current glibc version of fexecve(3) is implemented via /proc, which causes problems in sandboxed or otherwise restricted environments. Given the desire for a /proc-free fexecve() implementation, HPA suggested (https://lkml.org/lkml/2006/7/11/556) that an execveat(2) syscall would be an appropriate generalization. Also, having a new syscall means that it can take a flags argument without back-compatibility concerns. The current implementation just defines the AT_EMPTY_PATH and AT_SYMLINK_NOFOLLOW flags, but other flags could be added in future -- for example, flags for new namespaces (as suggested at https://lkml.org/lkml/2006/7/11/474). Related history: - https://lkml.org/lkml/2006/12/27/123 is an example of someone realizing that fexecve() is likely to fail in a chroot environment. - http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=514043 covered documenting the /proc requirement of fexecve(3) in its manpage, to "prevent other people from wasting their time". - https://bugzilla.redhat.com/show_bug.cgi?id=241609 described a problem where a process that did setuid() could not fexecve() because it no longer had access to /proc/self/fd; this has since been fixed. This patch (of 4): Add a new execveat(2) system call. execveat() is to execve() as openat() is to open(): it takes a file descriptor that refers to a directory, and resolves the filename relative to that. In addition, if the filename is empty and AT_EMPTY_PATH is specified, execveat() executes the file to which the file descriptor refers. This replicates the functionality of fexecve(), which is a system call in other UNIXen, but in Linux glibc it depends on opening "/proc/self/fd/<fd>" (and so relies on /proc being mounted). The filename fed to the executed program as argv[0] (or the name of the script fed to a script interpreter) will be of the form "/dev/fd/<fd>" (for an empty filename) or "/dev/fd/<fd>/<filename>", effectively reflecting how the executable was found. This does however mean that execution of a script in a /proc-less environment won't work; also, script execution via an O_CLOEXEC file descriptor fails (as the file will not be accessible after exec). Based on patches by Meredydd Luff. Signed-off-by: David Drysdale <drysdale@google.com> Cc: Meredydd Luff <meredydd@senatehouse.org> Cc: Shuah Khan <shuah.kh@samsung.com> Cc: "Eric W. Biederman" <ebiederm@xmission.com> Cc: Andy Lutomirski <luto@amacapital.net> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ingo Molnar <mingo@redhat.com> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: Kees Cook <keescook@chromium.org> Cc: Arnd Bergmann <arnd@arndb.de> Cc: Rich Felker <dalias@aerifal.cx> Cc: Christoph Hellwig <hch@infradead.org> Cc: Michael Kerrisk <mtk.manpages@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-12-12 16:57:29 -08:00
/* Need to be able to load the file after exec */
retval = -ENOENT;
syscalls: implement execveat() system call This patchset adds execveat(2) for x86, and is derived from Meredydd Luff's patch from Sept 2012 (https://lkml.org/lkml/2012/9/11/528). The primary aim of adding an execveat syscall is to allow an implementation of fexecve(3) that does not rely on the /proc filesystem, at least for executables (rather than scripts). The current glibc version of fexecve(3) is implemented via /proc, which causes problems in sandboxed or otherwise restricted environments. Given the desire for a /proc-free fexecve() implementation, HPA suggested (https://lkml.org/lkml/2006/7/11/556) that an execveat(2) syscall would be an appropriate generalization. Also, having a new syscall means that it can take a flags argument without back-compatibility concerns. The current implementation just defines the AT_EMPTY_PATH and AT_SYMLINK_NOFOLLOW flags, but other flags could be added in future -- for example, flags for new namespaces (as suggested at https://lkml.org/lkml/2006/7/11/474). Related history: - https://lkml.org/lkml/2006/12/27/123 is an example of someone realizing that fexecve() is likely to fail in a chroot environment. - http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=514043 covered documenting the /proc requirement of fexecve(3) in its manpage, to "prevent other people from wasting their time". - https://bugzilla.redhat.com/show_bug.cgi?id=241609 described a problem where a process that did setuid() could not fexecve() because it no longer had access to /proc/self/fd; this has since been fixed. This patch (of 4): Add a new execveat(2) system call. execveat() is to execve() as openat() is to open(): it takes a file descriptor that refers to a directory, and resolves the filename relative to that. In addition, if the filename is empty and AT_EMPTY_PATH is specified, execveat() executes the file to which the file descriptor refers. This replicates the functionality of fexecve(), which is a system call in other UNIXen, but in Linux glibc it depends on opening "/proc/self/fd/<fd>" (and so relies on /proc being mounted). The filename fed to the executed program as argv[0] (or the name of the script fed to a script interpreter) will be of the form "/dev/fd/<fd>" (for an empty filename) or "/dev/fd/<fd>/<filename>", effectively reflecting how the executable was found. This does however mean that execution of a script in a /proc-less environment won't work; also, script execution via an O_CLOEXEC file descriptor fails (as the file will not be accessible after exec). Based on patches by Meredydd Luff. Signed-off-by: David Drysdale <drysdale@google.com> Cc: Meredydd Luff <meredydd@senatehouse.org> Cc: Shuah Khan <shuah.kh@samsung.com> Cc: "Eric W. Biederman" <ebiederm@xmission.com> Cc: Andy Lutomirski <luto@amacapital.net> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ingo Molnar <mingo@redhat.com> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: Kees Cook <keescook@chromium.org> Cc: Arnd Bergmann <arnd@arndb.de> Cc: Rich Felker <dalias@aerifal.cx> Cc: Christoph Hellwig <hch@infradead.org> Cc: Michael Kerrisk <mtk.manpages@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-12-12 16:57:29 -08:00
if (bprm->interp_flags & BINPRM_FLAGS_PATH_INACCESSIBLE)
goto ret;
syscalls: implement execveat() system call This patchset adds execveat(2) for x86, and is derived from Meredydd Luff's patch from Sept 2012 (https://lkml.org/lkml/2012/9/11/528). The primary aim of adding an execveat syscall is to allow an implementation of fexecve(3) that does not rely on the /proc filesystem, at least for executables (rather than scripts). The current glibc version of fexecve(3) is implemented via /proc, which causes problems in sandboxed or otherwise restricted environments. Given the desire for a /proc-free fexecve() implementation, HPA suggested (https://lkml.org/lkml/2006/7/11/556) that an execveat(2) syscall would be an appropriate generalization. Also, having a new syscall means that it can take a flags argument without back-compatibility concerns. The current implementation just defines the AT_EMPTY_PATH and AT_SYMLINK_NOFOLLOW flags, but other flags could be added in future -- for example, flags for new namespaces (as suggested at https://lkml.org/lkml/2006/7/11/474). Related history: - https://lkml.org/lkml/2006/12/27/123 is an example of someone realizing that fexecve() is likely to fail in a chroot environment. - http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=514043 covered documenting the /proc requirement of fexecve(3) in its manpage, to "prevent other people from wasting their time". - https://bugzilla.redhat.com/show_bug.cgi?id=241609 described a problem where a process that did setuid() could not fexecve() because it no longer had access to /proc/self/fd; this has since been fixed. This patch (of 4): Add a new execveat(2) system call. execveat() is to execve() as openat() is to open(): it takes a file descriptor that refers to a directory, and resolves the filename relative to that. In addition, if the filename is empty and AT_EMPTY_PATH is specified, execveat() executes the file to which the file descriptor refers. This replicates the functionality of fexecve(), which is a system call in other UNIXen, but in Linux glibc it depends on opening "/proc/self/fd/<fd>" (and so relies on /proc being mounted). The filename fed to the executed program as argv[0] (or the name of the script fed to a script interpreter) will be of the form "/dev/fd/<fd>" (for an empty filename) or "/dev/fd/<fd>/<filename>", effectively reflecting how the executable was found. This does however mean that execution of a script in a /proc-less environment won't work; also, script execution via an O_CLOEXEC file descriptor fails (as the file will not be accessible after exec). Based on patches by Meredydd Luff. Signed-off-by: David Drysdale <drysdale@google.com> Cc: Meredydd Luff <meredydd@senatehouse.org> Cc: Shuah Khan <shuah.kh@samsung.com> Cc: "Eric W. Biederman" <ebiederm@xmission.com> Cc: Andy Lutomirski <luto@amacapital.net> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ingo Molnar <mingo@redhat.com> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: Kees Cook <keescook@chromium.org> Cc: Arnd Bergmann <arnd@arndb.de> Cc: Rich Felker <dalias@aerifal.cx> Cc: Christoph Hellwig <hch@infradead.org> Cc: Michael Kerrisk <mtk.manpages@gmail.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-12-12 16:57:29 -08:00
if (!(fmt->flags & MISC_FMT_PRESERVE_ARGV0)) {
retval = remove_arg_zero(bprm);
if (retval)
goto ret;
}
if (fmt->flags & MISC_FMT_OPEN_BINARY) {
/* if the binary should be opened on behalf of the
* interpreter than keep it open and assign descriptor
* to it
*/
fd_binary = get_unused_fd_flags(0);
if (fd_binary < 0) {
retval = fd_binary;
goto ret;
}
fd_install(fd_binary, bprm->file);
/* if the binary is not readable than enforce mm->dumpable=0
regardless of the interpreter's permissions */
would_dump(bprm, bprm->file);
allow_write_access(bprm->file);
bprm->file = NULL;
/* mark the bprm that fd should be passed to interp */
bprm->interp_flags |= BINPRM_FLAGS_EXECFD;
bprm->interp_data = fd_binary;
} else {
allow_write_access(bprm->file);
fput(bprm->file);
bprm->file = NULL;
}
/* make argv[1] be the path to the binary */
retval = copy_strings_kernel(1, &bprm->interp, bprm);
if (retval < 0)
goto error;
bprm->argc++;
/* add the interp as argv[0] */
retval = copy_strings_kernel(1, &fmt->interpreter, bprm);
if (retval < 0)
goto error;
bprm->argc++;
exec: do not leave bprm->interp on stack If a series of scripts are executed, each triggering module loading via unprintable bytes in the script header, kernel stack contents can leak into the command line. Normally execution of binfmt_script and binfmt_misc happens recursively. However, when modules are enabled, and unprintable bytes exist in the bprm->buf, execution will restart after attempting to load matching binfmt modules. Unfortunately, the logic in binfmt_script and binfmt_misc does not expect to get restarted. They leave bprm->interp pointing to their local stack. This means on restart bprm->interp is left pointing into unused stack memory which can then be copied into the userspace argv areas. After additional study, it seems that both recursion and restart remains the desirable way to handle exec with scripts, misc, and modules. As such, we need to protect the changes to interp. This changes the logic to require allocation for any changes to the bprm->interp. To avoid adding a new kmalloc to every exec, the default value is left as-is. Only when passing through binfmt_script or binfmt_misc does an allocation take place. For a proof of concept, see DoTest.sh from: http://www.halfdog.net/Security/2012/LinuxKernelBinfmtScriptStackDataDisclosure/ Signed-off-by: Kees Cook <keescook@chromium.org> Cc: halfdog <me@halfdog.net> Cc: P J P <ppandit@redhat.com> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-12-20 15:05:16 -08:00
/* Update interp in case binfmt_script needs it. */
retval = bprm_change_interp(fmt->interpreter, bprm);
exec: do not leave bprm->interp on stack If a series of scripts are executed, each triggering module loading via unprintable bytes in the script header, kernel stack contents can leak into the command line. Normally execution of binfmt_script and binfmt_misc happens recursively. However, when modules are enabled, and unprintable bytes exist in the bprm->buf, execution will restart after attempting to load matching binfmt modules. Unfortunately, the logic in binfmt_script and binfmt_misc does not expect to get restarted. They leave bprm->interp pointing to their local stack. This means on restart bprm->interp is left pointing into unused stack memory which can then be copied into the userspace argv areas. After additional study, it seems that both recursion and restart remains the desirable way to handle exec with scripts, misc, and modules. As such, we need to protect the changes to interp. This changes the logic to require allocation for any changes to the bprm->interp. To avoid adding a new kmalloc to every exec, the default value is left as-is. Only when passing through binfmt_script or binfmt_misc does an allocation take place. For a proof of concept, see DoTest.sh from: http://www.halfdog.net/Security/2012/LinuxKernelBinfmtScriptStackDataDisclosure/ Signed-off-by: Kees Cook <keescook@chromium.org> Cc: halfdog <me@halfdog.net> Cc: P J P <ppandit@redhat.com> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-12-20 15:05:16 -08:00
if (retval < 0)
goto error;
if (fmt->flags & MISC_FMT_OPEN_FILE) {
interp_file = filp_clone_open(fmt->interp_file);
if (!IS_ERR(interp_file))
deny_write_access(interp_file);
} else {
interp_file = open_exec(fmt->interpreter);
}
retval = PTR_ERR(interp_file);
if (IS_ERR(interp_file))
goto error;
bprm->file = interp_file;
if (fmt->flags & MISC_FMT_CREDENTIALS) {
loff_t pos = 0;
/*
* No need to call prepare_binprm(), it's already been
* done. bprm->buf is stale, update from interp_file.
*/
memset(bprm->buf, 0, BINPRM_BUF_SIZE);
retval = kernel_read(bprm->file, bprm->buf, BINPRM_BUF_SIZE,
&pos);
} else
retval = prepare_binprm(bprm);
if (retval < 0)
goto error;
retval = search_binary_handler(bprm);
if (retval < 0)
goto error;
ret:
binfmt_misc: cleanup on filesystem umount [ Upstream commit 1c5976ef0f7ad76319df748ccb99a4c7ba2ba464 ] Currently, registering a new binary type pins the binfmt_misc filesystem. Specifically, this means that as long as there is at least one binary type registered the binfmt_misc filesystem survives all umounts, i.e. the superblock is not destroyed. Meaning that a umount followed by another mount will end up with the same superblock and the same binary type handlers. This is a behavior we tend to discourage for any new filesystems (apart from a few special filesystems such as e.g. configfs or debugfs). A umount operation without the filesystem being pinned - by e.g. someone holding a file descriptor to an open file - should usually result in the destruction of the superblock and all associated resources. This makes introspection easier and leads to clearly defined, simple and clean semantics. An administrator can rely on the fact that a umount will guarantee a clean slate making it possible to reinitialize a filesystem. Right now all binary types would need to be explicitly deleted before that can happen. This allows us to remove the heavy-handed calls to simple_pin_fs() and simple_release_fs() when creating and deleting binary types. This in turn allows us to replace the current brittle pinning mechanism abusing dget() which has caused a range of bugs judging from prior fixes in [2] and [3]. The additional dget() in load_misc_binary() pins the dentry but only does so for the sake to prevent ->evict_inode() from freeing the node when a user removes the binary type and kill_node() is run. Which would mean ->interpreter and ->interp_file would be freed causing a UAF. This isn't really nicely documented nor is it very clean because it relies on simple_pin_fs() pinning the filesystem as long as at least one binary type exists. Otherwise it would cause load_misc_binary() to hold on to a dentry belonging to a superblock that has been shutdown. Replace that implicit pinning with a clean and simple per-node refcount and get rid of the ugly dget() pinning. A similar mechanism exists for e.g. binderfs (cf. [4]). All the cleanup work can now be done in ->evict_inode(). In a follow-up patch we will make it possible to use binfmt_misc in sandboxes. We will use the cleaner semantics where a umount for the filesystem will cause the superblock and all resources to be deallocated. In preparation for this apply the same semantics to the initial binfmt_misc mount. Note, that this is a user-visible change and as such a uapi change but one that we can reasonably risk. We've discussed this in earlier versions of this patchset (cf. [1]). The main user and provider of binfmt_misc is systemd. Systemd provides binfmt_misc via autofs since it is configurable as a kernel module and is used by a few exotic packages and users. As such a binfmt_misc mount is triggered when /proc/sys/fs/binfmt_misc is accessed and is only provided on demand. Other autofs on demand filesystems include EFI ESP which systemd umounts if the mountpoint stays idle for a certain amount of time. This doesn't apply to the binfmt_misc autofs mount which isn't touched once it is mounted meaning this change can't accidently wipe binary type handlers without someone having explicitly unmounted binfmt_misc. After speaking to systemd folks they don't expect this change to affect them. In line with our general policy, if we see a regression for systemd or other users with this change we will switch back to the old behavior for the initial binfmt_misc mount and have binary types pin the filesystem again. But while we touch this code let's take the chance and let's improve on the status quo. [1]: https://lore.kernel.org/r/20191216091220.465626-2-laurent@vivier.eu [2]: commit 43a4f2619038 ("exec: binfmt_misc: fix race between load_misc_binary() and kill_node()" [3]: commit 83f918274e4b ("exec: binfmt_misc: shift filp_close(interp_file) from kill_node() to bm_evict_inode()") [4]: commit f0fe2c0f050d ("binder: prevent UAF for binderfs devices II") Link: https://lore.kernel.org/r/20211028103114.2849140-1-brauner@kernel.org (v1) Cc: Sargun Dhillon <sargun@sargun.me> Cc: Serge Hallyn <serge@hallyn.com> Cc: Jann Horn <jannh@google.com> Cc: Henning Schild <henning.schild@siemens.com> Cc: Andrei Vagin <avagin@gmail.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Laurent Vivier <laurent@vivier.eu> Cc: linux-fsdevel@vger.kernel.org Acked-by: Serge Hallyn <serge@hallyn.com> Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com> Signed-off-by: Christian Brauner <brauner@kernel.org> Signed-off-by: Kees Cook <keescook@chromium.org> Signed-off-by: Sasha Levin <sashal@kernel.org> (cherry picked from commit 263bcebf5c2ab1fe949517225157f34015124620) Signed-off-by: Vegard Nossum <vegard.nossum@oracle.com>
2021-10-28 12:31:13 +02:00
/*
* If we actually put the node here all concurrent calls to
* load_misc_binary() will have finished. We also know
* that for the refcount to be zero ->evict_inode() must have removed
* the node to be deleted from the list. All that is left for us is to
* close and free.
*/
put_binfmt_handler(fmt);
return retval;
error:
if (fd_binary > 0)
sys_close(fd_binary);
bprm->interp_flags = 0;
bprm->interp_data = 0;
goto ret;
}
/* Command parsers */
/*
* parses and copies one argument enclosed in del from *sp to *dp,
* recognising the \x special.
* returns pointer to the copied argument or NULL in case of an
* error (and sets err) or null argument length.
*/
static char *scanarg(char *s, char del)
{
char c;
while ((c = *s++) != del) {
if (c == '\\' && *s == 'x') {
s++;
if (!isxdigit(*s++))
return NULL;
if (!isxdigit(*s++))
return NULL;
}
}
s[-1] ='\0';
return s;
}
static char *check_special_flags(char *sfs, Node *e)
{
char *p = sfs;
int cont = 1;
/* special flags */
while (cont) {
switch (*p) {
case 'P':
pr_debug("register: flag: P (preserve argv0)\n");
p++;
e->flags |= MISC_FMT_PRESERVE_ARGV0;
break;
case 'O':
pr_debug("register: flag: O (open binary)\n");
p++;
e->flags |= MISC_FMT_OPEN_BINARY;
break;
case 'C':
pr_debug("register: flag: C (preserve creds)\n");
p++;
/* this flags also implies the
open-binary flag */
e->flags |= (MISC_FMT_CREDENTIALS |
MISC_FMT_OPEN_BINARY);
break;
case 'F':
pr_debug("register: flag: F: open interpreter file now\n");
p++;
e->flags |= MISC_FMT_OPEN_FILE;
break;
default:
cont = 0;
}
}
return p;
}
/*
* This registers a new binary format, it recognises the syntax
* ':name:type:offset:magic:mask:interpreter:flags'
* where the ':' is the IFS, that can be chosen with the first char
*/
static Node *create_entry(const char __user *buffer, size_t count)
{
Node *e;
int memsize, err;
char *buf, *p;
char del;
binfmt_misc: add comments & debug logs When trying to develop a custom format handler, the errors returned all effectively get bucketed as EINVAL with no kernel messages. The other errors (ENOMEM/EFAULT) are internal/obvious and basic. Thus any time a bad handler is rejected, the developer has to walk the dense code and try to guess where it went wrong. Needing to dive into kernel code is itself a fairly high barrier for a lot of people. To improve this situation, let's deploy extensive pr_debug markers at logical parse points, and add comments to the dense parsing logic. It let's you see exactly where the parsing aborts, the string the kernel received (useful when dealing with shell code), how it translated the buffers to binary data, and how it will apply the mask at runtime. Some example output: $ echo ':qemu-foo:M::\x7fELF\xAD\xAD\x01\x00:\xff\xff\xff\xff\xff\x00\xff\x00:/usr/bin/qemu-foo:POC' > register $ dmesg binfmt_misc: register: received 92 bytes binfmt_misc: register: delim: 0x3a {:} binfmt_misc: register: name: {qemu-foo} binfmt_misc: register: type: M (magic) binfmt_misc: register: offset: 0x0 binfmt_misc: register: magic[raw]: 5c 78 37 66 45 4c 46 5c 78 41 44 5c 78 41 44 5c \x7fELF\xAD\xAD\ binfmt_misc: register: magic[raw]: 78 30 31 5c 78 30 30 00 x01\x00. binfmt_misc: register: mask[raw]: 5c 78 66 66 5c 78 66 66 5c 78 66 66 5c 78 66 66 \xff\xff\xff\xff binfmt_misc: register: mask[raw]: 5c 78 66 66 5c 78 30 30 5c 78 66 66 5c 78 30 30 \xff\x00\xff\x00 binfmt_misc: register: mask[raw]: 00 . binfmt_misc: register: magic/mask length: 8 binfmt_misc: register: magic[decoded]: 7f 45 4c 46 ad ad 01 00 .ELF.... binfmt_misc: register: mask[decoded]: ff ff ff ff ff 00 ff 00 ........ binfmt_misc: register: magic[masked]: 7f 45 4c 46 ad 00 01 00 .ELF.... binfmt_misc: register: interpreter: {/usr/bin/qemu-foo} binfmt_misc: register: flag: P (preserve argv0) binfmt_misc: register: flag: O (open binary) binfmt_misc: register: flag: C (preserve creds) The [raw] lines show us exactly what was received from userspace. The lines after that show us how the kernel has decoded things. Signed-off-by: Mike Frysinger <vapier@gentoo.org> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Joe Perches <joe@perches.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-12-10 15:52:08 -08:00
pr_debug("register: received %zu bytes\n", count);
/* some sanity checks */
err = -EINVAL;
binfmt_misc: expand the register format limit to 1920 bytes The current code places a 256 byte limit on the registration format. This ends up being fairly limited when you try to do matching against a binary format like ELF: - the magic & mask formats cannot have any embedded NUL chars (string_unescape_inplace halts at the first NUL) - each escape sequence quadruples the size: \x00 is needed for NUL - trying to match bytes at the start of the file as well as further on leads to a lot of \x00 sequences in the mask - magic & mask have to be the same length (when decoded) - still need bytes for the other fields - impossible! Let's look at a concrete (and common) example: using QEMU to run MIPS ELFs. The name field uses 11 bytes "qemu-mipsel". The interp uses 20 bytes "/usr/bin/qemu-mipsel". The type & flags takes up 4 bytes. We need 7 bytes for the delimiter (usually ":"). We can skip offset. So already we're down to 107 bytes to use with the magic/mask instead of the real limit of 128 (BINPRM_BUF_SIZE). If people use shell code to register (which they do the majority of the time), they're down to ~26 possible bytes since the escape sequence must be \x##. The ELF format looks like (both 32 & 64 bit): e_ident: 16 bytes e_type: 2 bytes e_machine: 2 bytes Those 20 bytes are enough for most architectures because they have so few formats in the first place, thus they can be uniquely identified. That also means for shell users, since 20 is smaller than 26, they can sanely register a handler. But for some targets (like MIPS), we need to poke further. The ELF fields continue on: e_entry: 4 or 8 bytes e_phoff: 4 or 8 bytes e_shoff: 4 or 8 bytes e_flags: 4 bytes We only care about e_flags here as that includes the bits to identify whether the ELF is O32/N32/N64. But now we have to consume another 16 bytes (for 32 bit ELFs) or 28 bytes (for 64 bit ELFs) just to match the flags. If every byte is escaped, we send 288 more bytes to the kernel ((20 {e_ident,e_type,e_machine} + 12 {e_entry,e_phoff,e_shoff} + 4 {e_flags}) * 2 {mask,magic} * 4 {escape}) and we've clearly blown our budget. Even if we try to be clever and do the decoding ourselves (rather than relying on the kernel to process \x##), we still can't hit the mark -- string_unescape_inplace treats mask & magic as C strings so NUL cannot be embedded. That leaves us with having to pass \x00 for the 12/24 entry/phoff/shoff bytes (as those will be completely random addresses), and that is a minimum requirement of 48/96 bytes for the mask alone. Add up the rest and we blow through it (this is for 64 bit ELFs): magic: 20 {e_ident,e_type,e_machine} + 24 {e_entry,e_phoff,e_shoff} + 4 {e_flags} = 48 # ^^ See note below. mask: 20 {e_ident,e_type,e_machine} + 96 {e_entry,e_phoff,e_shoff} + 4 {e_flags} = 120 Remember above we had 107 left over, and now we're at 168. This is of course the *best* case scenario -- you'll also want to have NUL bytes in the magic & mask too to match literal zeros. Note: the reason we can use 24 in the magic is that we can work off of the fact that for bytes the mask would clobber, we can stuff any value into magic that we want. So when mask is \x00, we don't need the magic to also be \x00, it can be an unescaped raw byte like '!'. This lets us handle more formats (barely) under the current 256 limit, but that's a pretty tall hoop to force people to jump through. With all that said, let's bump the limit from 256 bytes to 1920. This way we support escaping every byte of the mask & magic field (which is 1024 bytes by themselves -- 128 * 4 * 2), and we leave plenty of room for other fields. Like long paths to the interpreter (when you have source in your /really/long/homedir/qemu/foo). Since the current code stuffs more than one structure into the same buffer, we leave a bit of space to easily round up to 2k. 1920 is just as arbitrary as 256 ;). Signed-off-by: Mike Frysinger <vapier@gentoo.org> Cc: Al Viro <viro@zeniv.linux.org.uk> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-10-13 15:52:03 -07:00
if ((count < 11) || (count > MAX_REGISTER_LENGTH))
goto out;
err = -ENOMEM;
memsize = sizeof(Node) + count + 8;
e = kmalloc(memsize, GFP_KERNEL);
if (!e)
goto out;
p = buf = (char *)e + sizeof(Node);
memset(e, 0, sizeof(Node));
if (copy_from_user(buf, buffer, count))
goto efault;
del = *p++; /* delimeter */
binfmt_misc: add comments & debug logs When trying to develop a custom format handler, the errors returned all effectively get bucketed as EINVAL with no kernel messages. The other errors (ENOMEM/EFAULT) are internal/obvious and basic. Thus any time a bad handler is rejected, the developer has to walk the dense code and try to guess where it went wrong. Needing to dive into kernel code is itself a fairly high barrier for a lot of people. To improve this situation, let's deploy extensive pr_debug markers at logical parse points, and add comments to the dense parsing logic. It let's you see exactly where the parsing aborts, the string the kernel received (useful when dealing with shell code), how it translated the buffers to binary data, and how it will apply the mask at runtime. Some example output: $ echo ':qemu-foo:M::\x7fELF\xAD\xAD\x01\x00:\xff\xff\xff\xff\xff\x00\xff\x00:/usr/bin/qemu-foo:POC' > register $ dmesg binfmt_misc: register: received 92 bytes binfmt_misc: register: delim: 0x3a {:} binfmt_misc: register: name: {qemu-foo} binfmt_misc: register: type: M (magic) binfmt_misc: register: offset: 0x0 binfmt_misc: register: magic[raw]: 5c 78 37 66 45 4c 46 5c 78 41 44 5c 78 41 44 5c \x7fELF\xAD\xAD\ binfmt_misc: register: magic[raw]: 78 30 31 5c 78 30 30 00 x01\x00. binfmt_misc: register: mask[raw]: 5c 78 66 66 5c 78 66 66 5c 78 66 66 5c 78 66 66 \xff\xff\xff\xff binfmt_misc: register: mask[raw]: 5c 78 66 66 5c 78 30 30 5c 78 66 66 5c 78 30 30 \xff\x00\xff\x00 binfmt_misc: register: mask[raw]: 00 . binfmt_misc: register: magic/mask length: 8 binfmt_misc: register: magic[decoded]: 7f 45 4c 46 ad ad 01 00 .ELF.... binfmt_misc: register: mask[decoded]: ff ff ff ff ff 00 ff 00 ........ binfmt_misc: register: magic[masked]: 7f 45 4c 46 ad 00 01 00 .ELF.... binfmt_misc: register: interpreter: {/usr/bin/qemu-foo} binfmt_misc: register: flag: P (preserve argv0) binfmt_misc: register: flag: O (open binary) binfmt_misc: register: flag: C (preserve creds) The [raw] lines show us exactly what was received from userspace. The lines after that show us how the kernel has decoded things. Signed-off-by: Mike Frysinger <vapier@gentoo.org> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Joe Perches <joe@perches.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-12-10 15:52:08 -08:00
pr_debug("register: delim: %#x {%c}\n", del, del);
/* Pad the buffer with the delim to simplify parsing below. */
memset(buf + count, del, 8);
binfmt_misc: add comments & debug logs When trying to develop a custom format handler, the errors returned all effectively get bucketed as EINVAL with no kernel messages. The other errors (ENOMEM/EFAULT) are internal/obvious and basic. Thus any time a bad handler is rejected, the developer has to walk the dense code and try to guess where it went wrong. Needing to dive into kernel code is itself a fairly high barrier for a lot of people. To improve this situation, let's deploy extensive pr_debug markers at logical parse points, and add comments to the dense parsing logic. It let's you see exactly where the parsing aborts, the string the kernel received (useful when dealing with shell code), how it translated the buffers to binary data, and how it will apply the mask at runtime. Some example output: $ echo ':qemu-foo:M::\x7fELF\xAD\xAD\x01\x00:\xff\xff\xff\xff\xff\x00\xff\x00:/usr/bin/qemu-foo:POC' > register $ dmesg binfmt_misc: register: received 92 bytes binfmt_misc: register: delim: 0x3a {:} binfmt_misc: register: name: {qemu-foo} binfmt_misc: register: type: M (magic) binfmt_misc: register: offset: 0x0 binfmt_misc: register: magic[raw]: 5c 78 37 66 45 4c 46 5c 78 41 44 5c 78 41 44 5c \x7fELF\xAD\xAD\ binfmt_misc: register: magic[raw]: 78 30 31 5c 78 30 30 00 x01\x00. binfmt_misc: register: mask[raw]: 5c 78 66 66 5c 78 66 66 5c 78 66 66 5c 78 66 66 \xff\xff\xff\xff binfmt_misc: register: mask[raw]: 5c 78 66 66 5c 78 30 30 5c 78 66 66 5c 78 30 30 \xff\x00\xff\x00 binfmt_misc: register: mask[raw]: 00 . binfmt_misc: register: magic/mask length: 8 binfmt_misc: register: magic[decoded]: 7f 45 4c 46 ad ad 01 00 .ELF.... binfmt_misc: register: mask[decoded]: ff ff ff ff ff 00 ff 00 ........ binfmt_misc: register: magic[masked]: 7f 45 4c 46 ad 00 01 00 .ELF.... binfmt_misc: register: interpreter: {/usr/bin/qemu-foo} binfmt_misc: register: flag: P (preserve argv0) binfmt_misc: register: flag: O (open binary) binfmt_misc: register: flag: C (preserve creds) The [raw] lines show us exactly what was received from userspace. The lines after that show us how the kernel has decoded things. Signed-off-by: Mike Frysinger <vapier@gentoo.org> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Joe Perches <joe@perches.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-12-10 15:52:08 -08:00
/* Parse the 'name' field. */
e->name = p;
p = strchr(p, del);
if (!p)
goto einval;
*p++ = '\0';
if (!e->name[0] ||
!strcmp(e->name, ".") ||
!strcmp(e->name, "..") ||
strchr(e->name, '/'))
goto einval;
binfmt_misc: add comments & debug logs When trying to develop a custom format handler, the errors returned all effectively get bucketed as EINVAL with no kernel messages. The other errors (ENOMEM/EFAULT) are internal/obvious and basic. Thus any time a bad handler is rejected, the developer has to walk the dense code and try to guess where it went wrong. Needing to dive into kernel code is itself a fairly high barrier for a lot of people. To improve this situation, let's deploy extensive pr_debug markers at logical parse points, and add comments to the dense parsing logic. It let's you see exactly where the parsing aborts, the string the kernel received (useful when dealing with shell code), how it translated the buffers to binary data, and how it will apply the mask at runtime. Some example output: $ echo ':qemu-foo:M::\x7fELF\xAD\xAD\x01\x00:\xff\xff\xff\xff\xff\x00\xff\x00:/usr/bin/qemu-foo:POC' > register $ dmesg binfmt_misc: register: received 92 bytes binfmt_misc: register: delim: 0x3a {:} binfmt_misc: register: name: {qemu-foo} binfmt_misc: register: type: M (magic) binfmt_misc: register: offset: 0x0 binfmt_misc: register: magic[raw]: 5c 78 37 66 45 4c 46 5c 78 41 44 5c 78 41 44 5c \x7fELF\xAD\xAD\ binfmt_misc: register: magic[raw]: 78 30 31 5c 78 30 30 00 x01\x00. binfmt_misc: register: mask[raw]: 5c 78 66 66 5c 78 66 66 5c 78 66 66 5c 78 66 66 \xff\xff\xff\xff binfmt_misc: register: mask[raw]: 5c 78 66 66 5c 78 30 30 5c 78 66 66 5c 78 30 30 \xff\x00\xff\x00 binfmt_misc: register: mask[raw]: 00 . binfmt_misc: register: magic/mask length: 8 binfmt_misc: register: magic[decoded]: 7f 45 4c 46 ad ad 01 00 .ELF.... binfmt_misc: register: mask[decoded]: ff ff ff ff ff 00 ff 00 ........ binfmt_misc: register: magic[masked]: 7f 45 4c 46 ad 00 01 00 .ELF.... binfmt_misc: register: interpreter: {/usr/bin/qemu-foo} binfmt_misc: register: flag: P (preserve argv0) binfmt_misc: register: flag: O (open binary) binfmt_misc: register: flag: C (preserve creds) The [raw] lines show us exactly what was received from userspace. The lines after that show us how the kernel has decoded things. Signed-off-by: Mike Frysinger <vapier@gentoo.org> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Joe Perches <joe@perches.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-12-10 15:52:08 -08:00
pr_debug("register: name: {%s}\n", e->name);
/* Parse the 'type' field. */
switch (*p++) {
binfmt_misc: add comments & debug logs When trying to develop a custom format handler, the errors returned all effectively get bucketed as EINVAL with no kernel messages. The other errors (ENOMEM/EFAULT) are internal/obvious and basic. Thus any time a bad handler is rejected, the developer has to walk the dense code and try to guess where it went wrong. Needing to dive into kernel code is itself a fairly high barrier for a lot of people. To improve this situation, let's deploy extensive pr_debug markers at logical parse points, and add comments to the dense parsing logic. It let's you see exactly where the parsing aborts, the string the kernel received (useful when dealing with shell code), how it translated the buffers to binary data, and how it will apply the mask at runtime. Some example output: $ echo ':qemu-foo:M::\x7fELF\xAD\xAD\x01\x00:\xff\xff\xff\xff\xff\x00\xff\x00:/usr/bin/qemu-foo:POC' > register $ dmesg binfmt_misc: register: received 92 bytes binfmt_misc: register: delim: 0x3a {:} binfmt_misc: register: name: {qemu-foo} binfmt_misc: register: type: M (magic) binfmt_misc: register: offset: 0x0 binfmt_misc: register: magic[raw]: 5c 78 37 66 45 4c 46 5c 78 41 44 5c 78 41 44 5c \x7fELF\xAD\xAD\ binfmt_misc: register: magic[raw]: 78 30 31 5c 78 30 30 00 x01\x00. binfmt_misc: register: mask[raw]: 5c 78 66 66 5c 78 66 66 5c 78 66 66 5c 78 66 66 \xff\xff\xff\xff binfmt_misc: register: mask[raw]: 5c 78 66 66 5c 78 30 30 5c 78 66 66 5c 78 30 30 \xff\x00\xff\x00 binfmt_misc: register: mask[raw]: 00 . binfmt_misc: register: magic/mask length: 8 binfmt_misc: register: magic[decoded]: 7f 45 4c 46 ad ad 01 00 .ELF.... binfmt_misc: register: mask[decoded]: ff ff ff ff ff 00 ff 00 ........ binfmt_misc: register: magic[masked]: 7f 45 4c 46 ad 00 01 00 .ELF.... binfmt_misc: register: interpreter: {/usr/bin/qemu-foo} binfmt_misc: register: flag: P (preserve argv0) binfmt_misc: register: flag: O (open binary) binfmt_misc: register: flag: C (preserve creds) The [raw] lines show us exactly what was received from userspace. The lines after that show us how the kernel has decoded things. Signed-off-by: Mike Frysinger <vapier@gentoo.org> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Joe Perches <joe@perches.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-12-10 15:52:08 -08:00
case 'E':
pr_debug("register: type: E (extension)\n");
e->flags = 1 << Enabled;
break;
case 'M':
pr_debug("register: type: M (magic)\n");
e->flags = (1 << Enabled) | (1 << Magic);
break;
default:
goto einval;
}
if (*p++ != del)
goto einval;
binfmt_misc: add comments & debug logs When trying to develop a custom format handler, the errors returned all effectively get bucketed as EINVAL with no kernel messages. The other errors (ENOMEM/EFAULT) are internal/obvious and basic. Thus any time a bad handler is rejected, the developer has to walk the dense code and try to guess where it went wrong. Needing to dive into kernel code is itself a fairly high barrier for a lot of people. To improve this situation, let's deploy extensive pr_debug markers at logical parse points, and add comments to the dense parsing logic. It let's you see exactly where the parsing aborts, the string the kernel received (useful when dealing with shell code), how it translated the buffers to binary data, and how it will apply the mask at runtime. Some example output: $ echo ':qemu-foo:M::\x7fELF\xAD\xAD\x01\x00:\xff\xff\xff\xff\xff\x00\xff\x00:/usr/bin/qemu-foo:POC' > register $ dmesg binfmt_misc: register: received 92 bytes binfmt_misc: register: delim: 0x3a {:} binfmt_misc: register: name: {qemu-foo} binfmt_misc: register: type: M (magic) binfmt_misc: register: offset: 0x0 binfmt_misc: register: magic[raw]: 5c 78 37 66 45 4c 46 5c 78 41 44 5c 78 41 44 5c \x7fELF\xAD\xAD\ binfmt_misc: register: magic[raw]: 78 30 31 5c 78 30 30 00 x01\x00. binfmt_misc: register: mask[raw]: 5c 78 66 66 5c 78 66 66 5c 78 66 66 5c 78 66 66 \xff\xff\xff\xff binfmt_misc: register: mask[raw]: 5c 78 66 66 5c 78 30 30 5c 78 66 66 5c 78 30 30 \xff\x00\xff\x00 binfmt_misc: register: mask[raw]: 00 . binfmt_misc: register: magic/mask length: 8 binfmt_misc: register: magic[decoded]: 7f 45 4c 46 ad ad 01 00 .ELF.... binfmt_misc: register: mask[decoded]: ff ff ff ff ff 00 ff 00 ........ binfmt_misc: register: magic[masked]: 7f 45 4c 46 ad 00 01 00 .ELF.... binfmt_misc: register: interpreter: {/usr/bin/qemu-foo} binfmt_misc: register: flag: P (preserve argv0) binfmt_misc: register: flag: O (open binary) binfmt_misc: register: flag: C (preserve creds) The [raw] lines show us exactly what was received from userspace. The lines after that show us how the kernel has decoded things. Signed-off-by: Mike Frysinger <vapier@gentoo.org> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Joe Perches <joe@perches.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-12-10 15:52:08 -08:00
if (test_bit(Magic, &e->flags)) {
binfmt_misc: add comments & debug logs When trying to develop a custom format handler, the errors returned all effectively get bucketed as EINVAL with no kernel messages. The other errors (ENOMEM/EFAULT) are internal/obvious and basic. Thus any time a bad handler is rejected, the developer has to walk the dense code and try to guess where it went wrong. Needing to dive into kernel code is itself a fairly high barrier for a lot of people. To improve this situation, let's deploy extensive pr_debug markers at logical parse points, and add comments to the dense parsing logic. It let's you see exactly where the parsing aborts, the string the kernel received (useful when dealing with shell code), how it translated the buffers to binary data, and how it will apply the mask at runtime. Some example output: $ echo ':qemu-foo:M::\x7fELF\xAD\xAD\x01\x00:\xff\xff\xff\xff\xff\x00\xff\x00:/usr/bin/qemu-foo:POC' > register $ dmesg binfmt_misc: register: received 92 bytes binfmt_misc: register: delim: 0x3a {:} binfmt_misc: register: name: {qemu-foo} binfmt_misc: register: type: M (magic) binfmt_misc: register: offset: 0x0 binfmt_misc: register: magic[raw]: 5c 78 37 66 45 4c 46 5c 78 41 44 5c 78 41 44 5c \x7fELF\xAD\xAD\ binfmt_misc: register: magic[raw]: 78 30 31 5c 78 30 30 00 x01\x00. binfmt_misc: register: mask[raw]: 5c 78 66 66 5c 78 66 66 5c 78 66 66 5c 78 66 66 \xff\xff\xff\xff binfmt_misc: register: mask[raw]: 5c 78 66 66 5c 78 30 30 5c 78 66 66 5c 78 30 30 \xff\x00\xff\x00 binfmt_misc: register: mask[raw]: 00 . binfmt_misc: register: magic/mask length: 8 binfmt_misc: register: magic[decoded]: 7f 45 4c 46 ad ad 01 00 .ELF.... binfmt_misc: register: mask[decoded]: ff ff ff ff ff 00 ff 00 ........ binfmt_misc: register: magic[masked]: 7f 45 4c 46 ad 00 01 00 .ELF.... binfmt_misc: register: interpreter: {/usr/bin/qemu-foo} binfmt_misc: register: flag: P (preserve argv0) binfmt_misc: register: flag: O (open binary) binfmt_misc: register: flag: C (preserve creds) The [raw] lines show us exactly what was received from userspace. The lines after that show us how the kernel has decoded things. Signed-off-by: Mike Frysinger <vapier@gentoo.org> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Joe Perches <joe@perches.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-12-10 15:52:08 -08:00
/* Handle the 'M' (magic) format. */
char *s;
/* Parse the 'offset' field. */
s = strchr(p, del);
if (!s)
goto einval;
fs/binfmt_misc.c: do not allow offset overflow commit 5cc41e099504b77014358b58567c5ea6293dd220 upstream. WHen registering a new binfmt_misc handler, it is possible to overflow the offset to get a negative value, which might crash the system, or possibly leak kernel data. Here is a crash log when 2500000000 was used as an offset: BUG: unable to handle kernel paging request at ffff989cfd6edca0 IP: load_misc_binary+0x22b/0x470 [binfmt_misc] PGD 1ef3e067 P4D 1ef3e067 PUD 0 Oops: 0000 [#1] SMP NOPTI Modules linked in: binfmt_misc kvm_intel ppdev kvm irqbypass joydev input_leds serio_raw mac_hid parport_pc qemu_fw_cfg parpy CPU: 0 PID: 2499 Comm: bash Not tainted 4.15.0-22-generic #24-Ubuntu Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.11.1-1 04/01/2014 RIP: 0010:load_misc_binary+0x22b/0x470 [binfmt_misc] Call Trace: search_binary_handler+0x97/0x1d0 do_execveat_common.isra.34+0x667/0x810 SyS_execve+0x31/0x40 do_syscall_64+0x73/0x130 entry_SYSCALL_64_after_hwframe+0x3d/0xa2 Use kstrtoint instead of simple_strtoul. It will work as the code already set the delimiter byte to '\0' and we only do it when the field is not empty. Tested with offsets -1, 2500000000, UINT_MAX and INT_MAX. Also tested with examples documented at Documentation/admin-guide/binfmt-misc.rst and other registrations from packages on Ubuntu. Link: http://lkml.kernel.org/r/20180529135648.14254-1-cascardo@canonical.com Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Signed-off-by: Thadeu Lima de Souza Cascardo <cascardo@canonical.com> Reviewed-by: Andrew Morton <akpm@linux-foundation.org> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2018-06-07 17:11:01 -07:00
*s = '\0';
if (p != s) {
int r = kstrtoint(p, 10, &e->offset);
if (r != 0 || e->offset < 0)
goto einval;
}
p = s;
if (*p++)
goto einval;
binfmt_misc: add comments & debug logs When trying to develop a custom format handler, the errors returned all effectively get bucketed as EINVAL with no kernel messages. The other errors (ENOMEM/EFAULT) are internal/obvious and basic. Thus any time a bad handler is rejected, the developer has to walk the dense code and try to guess where it went wrong. Needing to dive into kernel code is itself a fairly high barrier for a lot of people. To improve this situation, let's deploy extensive pr_debug markers at logical parse points, and add comments to the dense parsing logic. It let's you see exactly where the parsing aborts, the string the kernel received (useful when dealing with shell code), how it translated the buffers to binary data, and how it will apply the mask at runtime. Some example output: $ echo ':qemu-foo:M::\x7fELF\xAD\xAD\x01\x00:\xff\xff\xff\xff\xff\x00\xff\x00:/usr/bin/qemu-foo:POC' > register $ dmesg binfmt_misc: register: received 92 bytes binfmt_misc: register: delim: 0x3a {:} binfmt_misc: register: name: {qemu-foo} binfmt_misc: register: type: M (magic) binfmt_misc: register: offset: 0x0 binfmt_misc: register: magic[raw]: 5c 78 37 66 45 4c 46 5c 78 41 44 5c 78 41 44 5c \x7fELF\xAD\xAD\ binfmt_misc: register: magic[raw]: 78 30 31 5c 78 30 30 00 x01\x00. binfmt_misc: register: mask[raw]: 5c 78 66 66 5c 78 66 66 5c 78 66 66 5c 78 66 66 \xff\xff\xff\xff binfmt_misc: register: mask[raw]: 5c 78 66 66 5c 78 30 30 5c 78 66 66 5c 78 30 30 \xff\x00\xff\x00 binfmt_misc: register: mask[raw]: 00 . binfmt_misc: register: magic/mask length: 8 binfmt_misc: register: magic[decoded]: 7f 45 4c 46 ad ad 01 00 .ELF.... binfmt_misc: register: mask[decoded]: ff ff ff ff ff 00 ff 00 ........ binfmt_misc: register: magic[masked]: 7f 45 4c 46 ad 00 01 00 .ELF.... binfmt_misc: register: interpreter: {/usr/bin/qemu-foo} binfmt_misc: register: flag: P (preserve argv0) binfmt_misc: register: flag: O (open binary) binfmt_misc: register: flag: C (preserve creds) The [raw] lines show us exactly what was received from userspace. The lines after that show us how the kernel has decoded things. Signed-off-by: Mike Frysinger <vapier@gentoo.org> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Joe Perches <joe@perches.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-12-10 15:52:08 -08:00
pr_debug("register: offset: %#x\n", e->offset);
/* Parse the 'magic' field. */
e->magic = p;
p = scanarg(p, del);
if (!p)
goto einval;
if (!e->magic[0])
goto einval;
binfmt_misc: add comments & debug logs When trying to develop a custom format handler, the errors returned all effectively get bucketed as EINVAL with no kernel messages. The other errors (ENOMEM/EFAULT) are internal/obvious and basic. Thus any time a bad handler is rejected, the developer has to walk the dense code and try to guess where it went wrong. Needing to dive into kernel code is itself a fairly high barrier for a lot of people. To improve this situation, let's deploy extensive pr_debug markers at logical parse points, and add comments to the dense parsing logic. It let's you see exactly where the parsing aborts, the string the kernel received (useful when dealing with shell code), how it translated the buffers to binary data, and how it will apply the mask at runtime. Some example output: $ echo ':qemu-foo:M::\x7fELF\xAD\xAD\x01\x00:\xff\xff\xff\xff\xff\x00\xff\x00:/usr/bin/qemu-foo:POC' > register $ dmesg binfmt_misc: register: received 92 bytes binfmt_misc: register: delim: 0x3a {:} binfmt_misc: register: name: {qemu-foo} binfmt_misc: register: type: M (magic) binfmt_misc: register: offset: 0x0 binfmt_misc: register: magic[raw]: 5c 78 37 66 45 4c 46 5c 78 41 44 5c 78 41 44 5c \x7fELF\xAD\xAD\ binfmt_misc: register: magic[raw]: 78 30 31 5c 78 30 30 00 x01\x00. binfmt_misc: register: mask[raw]: 5c 78 66 66 5c 78 66 66 5c 78 66 66 5c 78 66 66 \xff\xff\xff\xff binfmt_misc: register: mask[raw]: 5c 78 66 66 5c 78 30 30 5c 78 66 66 5c 78 30 30 \xff\x00\xff\x00 binfmt_misc: register: mask[raw]: 00 . binfmt_misc: register: magic/mask length: 8 binfmt_misc: register: magic[decoded]: 7f 45 4c 46 ad ad 01 00 .ELF.... binfmt_misc: register: mask[decoded]: ff ff ff ff ff 00 ff 00 ........ binfmt_misc: register: magic[masked]: 7f 45 4c 46 ad 00 01 00 .ELF.... binfmt_misc: register: interpreter: {/usr/bin/qemu-foo} binfmt_misc: register: flag: P (preserve argv0) binfmt_misc: register: flag: O (open binary) binfmt_misc: register: flag: C (preserve creds) The [raw] lines show us exactly what was received from userspace. The lines after that show us how the kernel has decoded things. Signed-off-by: Mike Frysinger <vapier@gentoo.org> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Joe Perches <joe@perches.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-12-10 15:52:08 -08:00
if (USE_DEBUG)
print_hex_dump_bytes(
KBUILD_MODNAME ": register: magic[raw]: ",
DUMP_PREFIX_NONE, e->magic, p - e->magic);
/* Parse the 'mask' field. */
e->mask = p;
p = scanarg(p, del);
if (!p)
goto einval;
if (!e->mask[0]) {
e->mask = NULL;
binfmt_misc: add comments & debug logs When trying to develop a custom format handler, the errors returned all effectively get bucketed as EINVAL with no kernel messages. The other errors (ENOMEM/EFAULT) are internal/obvious and basic. Thus any time a bad handler is rejected, the developer has to walk the dense code and try to guess where it went wrong. Needing to dive into kernel code is itself a fairly high barrier for a lot of people. To improve this situation, let's deploy extensive pr_debug markers at logical parse points, and add comments to the dense parsing logic. It let's you see exactly where the parsing aborts, the string the kernel received (useful when dealing with shell code), how it translated the buffers to binary data, and how it will apply the mask at runtime. Some example output: $ echo ':qemu-foo:M::\x7fELF\xAD\xAD\x01\x00:\xff\xff\xff\xff\xff\x00\xff\x00:/usr/bin/qemu-foo:POC' > register $ dmesg binfmt_misc: register: received 92 bytes binfmt_misc: register: delim: 0x3a {:} binfmt_misc: register: name: {qemu-foo} binfmt_misc: register: type: M (magic) binfmt_misc: register: offset: 0x0 binfmt_misc: register: magic[raw]: 5c 78 37 66 45 4c 46 5c 78 41 44 5c 78 41 44 5c \x7fELF\xAD\xAD\ binfmt_misc: register: magic[raw]: 78 30 31 5c 78 30 30 00 x01\x00. binfmt_misc: register: mask[raw]: 5c 78 66 66 5c 78 66 66 5c 78 66 66 5c 78 66 66 \xff\xff\xff\xff binfmt_misc: register: mask[raw]: 5c 78 66 66 5c 78 30 30 5c 78 66 66 5c 78 30 30 \xff\x00\xff\x00 binfmt_misc: register: mask[raw]: 00 . binfmt_misc: register: magic/mask length: 8 binfmt_misc: register: magic[decoded]: 7f 45 4c 46 ad ad 01 00 .ELF.... binfmt_misc: register: mask[decoded]: ff ff ff ff ff 00 ff 00 ........ binfmt_misc: register: magic[masked]: 7f 45 4c 46 ad 00 01 00 .ELF.... binfmt_misc: register: interpreter: {/usr/bin/qemu-foo} binfmt_misc: register: flag: P (preserve argv0) binfmt_misc: register: flag: O (open binary) binfmt_misc: register: flag: C (preserve creds) The [raw] lines show us exactly what was received from userspace. The lines after that show us how the kernel has decoded things. Signed-off-by: Mike Frysinger <vapier@gentoo.org> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Joe Perches <joe@perches.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-12-10 15:52:08 -08:00
pr_debug("register: mask[raw]: none\n");
} else if (USE_DEBUG)
print_hex_dump_bytes(
KBUILD_MODNAME ": register: mask[raw]: ",
DUMP_PREFIX_NONE, e->mask, p - e->mask);
/*
* Decode the magic & mask fields.
* Note: while we might have accepted embedded NUL bytes from
* above, the unescape helpers here will stop at the first one
* it encounters.
*/
e->size = string_unescape_inplace(e->magic, UNESCAPE_HEX);
if (e->mask &&
string_unescape_inplace(e->mask, UNESCAPE_HEX) != e->size)
goto einval;
fs/binfmt_misc.c: do not allow offset overflow commit 5cc41e099504b77014358b58567c5ea6293dd220 upstream. WHen registering a new binfmt_misc handler, it is possible to overflow the offset to get a negative value, which might crash the system, or possibly leak kernel data. Here is a crash log when 2500000000 was used as an offset: BUG: unable to handle kernel paging request at ffff989cfd6edca0 IP: load_misc_binary+0x22b/0x470 [binfmt_misc] PGD 1ef3e067 P4D 1ef3e067 PUD 0 Oops: 0000 [#1] SMP NOPTI Modules linked in: binfmt_misc kvm_intel ppdev kvm irqbypass joydev input_leds serio_raw mac_hid parport_pc qemu_fw_cfg parpy CPU: 0 PID: 2499 Comm: bash Not tainted 4.15.0-22-generic #24-Ubuntu Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.11.1-1 04/01/2014 RIP: 0010:load_misc_binary+0x22b/0x470 [binfmt_misc] Call Trace: search_binary_handler+0x97/0x1d0 do_execveat_common.isra.34+0x667/0x810 SyS_execve+0x31/0x40 do_syscall_64+0x73/0x130 entry_SYSCALL_64_after_hwframe+0x3d/0xa2 Use kstrtoint instead of simple_strtoul. It will work as the code already set the delimiter byte to '\0' and we only do it when the field is not empty. Tested with offsets -1, 2500000000, UINT_MAX and INT_MAX. Also tested with examples documented at Documentation/admin-guide/binfmt-misc.rst and other registrations from packages on Ubuntu. Link: http://lkml.kernel.org/r/20180529135648.14254-1-cascardo@canonical.com Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2") Signed-off-by: Thadeu Lima de Souza Cascardo <cascardo@canonical.com> Reviewed-by: Andrew Morton <akpm@linux-foundation.org> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2018-06-07 17:11:01 -07:00
if (e->size > BINPRM_BUF_SIZE ||
BINPRM_BUF_SIZE - e->size < e->offset)
goto einval;
binfmt_misc: add comments & debug logs When trying to develop a custom format handler, the errors returned all effectively get bucketed as EINVAL with no kernel messages. The other errors (ENOMEM/EFAULT) are internal/obvious and basic. Thus any time a bad handler is rejected, the developer has to walk the dense code and try to guess where it went wrong. Needing to dive into kernel code is itself a fairly high barrier for a lot of people. To improve this situation, let's deploy extensive pr_debug markers at logical parse points, and add comments to the dense parsing logic. It let's you see exactly where the parsing aborts, the string the kernel received (useful when dealing with shell code), how it translated the buffers to binary data, and how it will apply the mask at runtime. Some example output: $ echo ':qemu-foo:M::\x7fELF\xAD\xAD\x01\x00:\xff\xff\xff\xff\xff\x00\xff\x00:/usr/bin/qemu-foo:POC' > register $ dmesg binfmt_misc: register: received 92 bytes binfmt_misc: register: delim: 0x3a {:} binfmt_misc: register: name: {qemu-foo} binfmt_misc: register: type: M (magic) binfmt_misc: register: offset: 0x0 binfmt_misc: register: magic[raw]: 5c 78 37 66 45 4c 46 5c 78 41 44 5c 78 41 44 5c \x7fELF\xAD\xAD\ binfmt_misc: register: magic[raw]: 78 30 31 5c 78 30 30 00 x01\x00. binfmt_misc: register: mask[raw]: 5c 78 66 66 5c 78 66 66 5c 78 66 66 5c 78 66 66 \xff\xff\xff\xff binfmt_misc: register: mask[raw]: 5c 78 66 66 5c 78 30 30 5c 78 66 66 5c 78 30 30 \xff\x00\xff\x00 binfmt_misc: register: mask[raw]: 00 . binfmt_misc: register: magic/mask length: 8 binfmt_misc: register: magic[decoded]: 7f 45 4c 46 ad ad 01 00 .ELF.... binfmt_misc: register: mask[decoded]: ff ff ff ff ff 00 ff 00 ........ binfmt_misc: register: magic[masked]: 7f 45 4c 46 ad 00 01 00 .ELF.... binfmt_misc: register: interpreter: {/usr/bin/qemu-foo} binfmt_misc: register: flag: P (preserve argv0) binfmt_misc: register: flag: O (open binary) binfmt_misc: register: flag: C (preserve creds) The [raw] lines show us exactly what was received from userspace. The lines after that show us how the kernel has decoded things. Signed-off-by: Mike Frysinger <vapier@gentoo.org> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Joe Perches <joe@perches.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-12-10 15:52:08 -08:00
pr_debug("register: magic/mask length: %i\n", e->size);
if (USE_DEBUG) {
print_hex_dump_bytes(
KBUILD_MODNAME ": register: magic[decoded]: ",
DUMP_PREFIX_NONE, e->magic, e->size);
if (e->mask) {
int i;
char *masked = kmalloc(e->size, GFP_KERNEL);
binfmt_misc: add comments & debug logs When trying to develop a custom format handler, the errors returned all effectively get bucketed as EINVAL with no kernel messages. The other errors (ENOMEM/EFAULT) are internal/obvious and basic. Thus any time a bad handler is rejected, the developer has to walk the dense code and try to guess where it went wrong. Needing to dive into kernel code is itself a fairly high barrier for a lot of people. To improve this situation, let's deploy extensive pr_debug markers at logical parse points, and add comments to the dense parsing logic. It let's you see exactly where the parsing aborts, the string the kernel received (useful when dealing with shell code), how it translated the buffers to binary data, and how it will apply the mask at runtime. Some example output: $ echo ':qemu-foo:M::\x7fELF\xAD\xAD\x01\x00:\xff\xff\xff\xff\xff\x00\xff\x00:/usr/bin/qemu-foo:POC' > register $ dmesg binfmt_misc: register: received 92 bytes binfmt_misc: register: delim: 0x3a {:} binfmt_misc: register: name: {qemu-foo} binfmt_misc: register: type: M (magic) binfmt_misc: register: offset: 0x0 binfmt_misc: register: magic[raw]: 5c 78 37 66 45 4c 46 5c 78 41 44 5c 78 41 44 5c \x7fELF\xAD\xAD\ binfmt_misc: register: magic[raw]: 78 30 31 5c 78 30 30 00 x01\x00. binfmt_misc: register: mask[raw]: 5c 78 66 66 5c 78 66 66 5c 78 66 66 5c 78 66 66 \xff\xff\xff\xff binfmt_misc: register: mask[raw]: 5c 78 66 66 5c 78 30 30 5c 78 66 66 5c 78 30 30 \xff\x00\xff\x00 binfmt_misc: register: mask[raw]: 00 . binfmt_misc: register: magic/mask length: 8 binfmt_misc: register: magic[decoded]: 7f 45 4c 46 ad ad 01 00 .ELF.... binfmt_misc: register: mask[decoded]: ff ff ff ff ff 00 ff 00 ........ binfmt_misc: register: magic[masked]: 7f 45 4c 46 ad 00 01 00 .ELF.... binfmt_misc: register: interpreter: {/usr/bin/qemu-foo} binfmt_misc: register: flag: P (preserve argv0) binfmt_misc: register: flag: O (open binary) binfmt_misc: register: flag: C (preserve creds) The [raw] lines show us exactly what was received from userspace. The lines after that show us how the kernel has decoded things. Signed-off-by: Mike Frysinger <vapier@gentoo.org> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Joe Perches <joe@perches.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-12-10 15:52:08 -08:00
print_hex_dump_bytes(
KBUILD_MODNAME ": register: mask[decoded]: ",
DUMP_PREFIX_NONE, e->mask, e->size);
if (masked) {
for (i = 0; i < e->size; ++i)
masked[i] = e->magic[i] & e->mask[i];
print_hex_dump_bytes(
KBUILD_MODNAME ": register: magic[masked]: ",
DUMP_PREFIX_NONE, masked, e->size);
kfree(masked);
}
}
}
} else {
binfmt_misc: add comments & debug logs When trying to develop a custom format handler, the errors returned all effectively get bucketed as EINVAL with no kernel messages. The other errors (ENOMEM/EFAULT) are internal/obvious and basic. Thus any time a bad handler is rejected, the developer has to walk the dense code and try to guess where it went wrong. Needing to dive into kernel code is itself a fairly high barrier for a lot of people. To improve this situation, let's deploy extensive pr_debug markers at logical parse points, and add comments to the dense parsing logic. It let's you see exactly where the parsing aborts, the string the kernel received (useful when dealing with shell code), how it translated the buffers to binary data, and how it will apply the mask at runtime. Some example output: $ echo ':qemu-foo:M::\x7fELF\xAD\xAD\x01\x00:\xff\xff\xff\xff\xff\x00\xff\x00:/usr/bin/qemu-foo:POC' > register $ dmesg binfmt_misc: register: received 92 bytes binfmt_misc: register: delim: 0x3a {:} binfmt_misc: register: name: {qemu-foo} binfmt_misc: register: type: M (magic) binfmt_misc: register: offset: 0x0 binfmt_misc: register: magic[raw]: 5c 78 37 66 45 4c 46 5c 78 41 44 5c 78 41 44 5c \x7fELF\xAD\xAD\ binfmt_misc: register: magic[raw]: 78 30 31 5c 78 30 30 00 x01\x00. binfmt_misc: register: mask[raw]: 5c 78 66 66 5c 78 66 66 5c 78 66 66 5c 78 66 66 \xff\xff\xff\xff binfmt_misc: register: mask[raw]: 5c 78 66 66 5c 78 30 30 5c 78 66 66 5c 78 30 30 \xff\x00\xff\x00 binfmt_misc: register: mask[raw]: 00 . binfmt_misc: register: magic/mask length: 8 binfmt_misc: register: magic[decoded]: 7f 45 4c 46 ad ad 01 00 .ELF.... binfmt_misc: register: mask[decoded]: ff ff ff ff ff 00 ff 00 ........ binfmt_misc: register: magic[masked]: 7f 45 4c 46 ad 00 01 00 .ELF.... binfmt_misc: register: interpreter: {/usr/bin/qemu-foo} binfmt_misc: register: flag: P (preserve argv0) binfmt_misc: register: flag: O (open binary) binfmt_misc: register: flag: C (preserve creds) The [raw] lines show us exactly what was received from userspace. The lines after that show us how the kernel has decoded things. Signed-off-by: Mike Frysinger <vapier@gentoo.org> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Joe Perches <joe@perches.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-12-10 15:52:08 -08:00
/* Handle the 'E' (extension) format. */
/* Skip the 'offset' field. */
p = strchr(p, del);
if (!p)
goto einval;
*p++ = '\0';
binfmt_misc: add comments & debug logs When trying to develop a custom format handler, the errors returned all effectively get bucketed as EINVAL with no kernel messages. The other errors (ENOMEM/EFAULT) are internal/obvious and basic. Thus any time a bad handler is rejected, the developer has to walk the dense code and try to guess where it went wrong. Needing to dive into kernel code is itself a fairly high barrier for a lot of people. To improve this situation, let's deploy extensive pr_debug markers at logical parse points, and add comments to the dense parsing logic. It let's you see exactly where the parsing aborts, the string the kernel received (useful when dealing with shell code), how it translated the buffers to binary data, and how it will apply the mask at runtime. Some example output: $ echo ':qemu-foo:M::\x7fELF\xAD\xAD\x01\x00:\xff\xff\xff\xff\xff\x00\xff\x00:/usr/bin/qemu-foo:POC' > register $ dmesg binfmt_misc: register: received 92 bytes binfmt_misc: register: delim: 0x3a {:} binfmt_misc: register: name: {qemu-foo} binfmt_misc: register: type: M (magic) binfmt_misc: register: offset: 0x0 binfmt_misc: register: magic[raw]: 5c 78 37 66 45 4c 46 5c 78 41 44 5c 78 41 44 5c \x7fELF\xAD\xAD\ binfmt_misc: register: magic[raw]: 78 30 31 5c 78 30 30 00 x01\x00. binfmt_misc: register: mask[raw]: 5c 78 66 66 5c 78 66 66 5c 78 66 66 5c 78 66 66 \xff\xff\xff\xff binfmt_misc: register: mask[raw]: 5c 78 66 66 5c 78 30 30 5c 78 66 66 5c 78 30 30 \xff\x00\xff\x00 binfmt_misc: register: mask[raw]: 00 . binfmt_misc: register: magic/mask length: 8 binfmt_misc: register: magic[decoded]: 7f 45 4c 46 ad ad 01 00 .ELF.... binfmt_misc: register: mask[decoded]: ff ff ff ff ff 00 ff 00 ........ binfmt_misc: register: magic[masked]: 7f 45 4c 46 ad 00 01 00 .ELF.... binfmt_misc: register: interpreter: {/usr/bin/qemu-foo} binfmt_misc: register: flag: P (preserve argv0) binfmt_misc: register: flag: O (open binary) binfmt_misc: register: flag: C (preserve creds) The [raw] lines show us exactly what was received from userspace. The lines after that show us how the kernel has decoded things. Signed-off-by: Mike Frysinger <vapier@gentoo.org> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Joe Perches <joe@perches.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-12-10 15:52:08 -08:00
/* Parse the 'magic' field. */
e->magic = p;
p = strchr(p, del);
if (!p)
goto einval;
*p++ = '\0';
if (!e->magic[0] || strchr(e->magic, '/'))
goto einval;
binfmt_misc: add comments & debug logs When trying to develop a custom format handler, the errors returned all effectively get bucketed as EINVAL with no kernel messages. The other errors (ENOMEM/EFAULT) are internal/obvious and basic. Thus any time a bad handler is rejected, the developer has to walk the dense code and try to guess where it went wrong. Needing to dive into kernel code is itself a fairly high barrier for a lot of people. To improve this situation, let's deploy extensive pr_debug markers at logical parse points, and add comments to the dense parsing logic. It let's you see exactly where the parsing aborts, the string the kernel received (useful when dealing with shell code), how it translated the buffers to binary data, and how it will apply the mask at runtime. Some example output: $ echo ':qemu-foo:M::\x7fELF\xAD\xAD\x01\x00:\xff\xff\xff\xff\xff\x00\xff\x00:/usr/bin/qemu-foo:POC' > register $ dmesg binfmt_misc: register: received 92 bytes binfmt_misc: register: delim: 0x3a {:} binfmt_misc: register: name: {qemu-foo} binfmt_misc: register: type: M (magic) binfmt_misc: register: offset: 0x0 binfmt_misc: register: magic[raw]: 5c 78 37 66 45 4c 46 5c 78 41 44 5c 78 41 44 5c \x7fELF\xAD\xAD\ binfmt_misc: register: magic[raw]: 78 30 31 5c 78 30 30 00 x01\x00. binfmt_misc: register: mask[raw]: 5c 78 66 66 5c 78 66 66 5c 78 66 66 5c 78 66 66 \xff\xff\xff\xff binfmt_misc: register: mask[raw]: 5c 78 66 66 5c 78 30 30 5c 78 66 66 5c 78 30 30 \xff\x00\xff\x00 binfmt_misc: register: mask[raw]: 00 . binfmt_misc: register: magic/mask length: 8 binfmt_misc: register: magic[decoded]: 7f 45 4c 46 ad ad 01 00 .ELF.... binfmt_misc: register: mask[decoded]: ff ff ff ff ff 00 ff 00 ........ binfmt_misc: register: magic[masked]: 7f 45 4c 46 ad 00 01 00 .ELF.... binfmt_misc: register: interpreter: {/usr/bin/qemu-foo} binfmt_misc: register: flag: P (preserve argv0) binfmt_misc: register: flag: O (open binary) binfmt_misc: register: flag: C (preserve creds) The [raw] lines show us exactly what was received from userspace. The lines after that show us how the kernel has decoded things. Signed-off-by: Mike Frysinger <vapier@gentoo.org> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Joe Perches <joe@perches.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-12-10 15:52:08 -08:00
pr_debug("register: extension: {%s}\n", e->magic);
/* Skip the 'mask' field. */
p = strchr(p, del);
if (!p)
goto einval;
*p++ = '\0';
}
binfmt_misc: add comments & debug logs When trying to develop a custom format handler, the errors returned all effectively get bucketed as EINVAL with no kernel messages. The other errors (ENOMEM/EFAULT) are internal/obvious and basic. Thus any time a bad handler is rejected, the developer has to walk the dense code and try to guess where it went wrong. Needing to dive into kernel code is itself a fairly high barrier for a lot of people. To improve this situation, let's deploy extensive pr_debug markers at logical parse points, and add comments to the dense parsing logic. It let's you see exactly where the parsing aborts, the string the kernel received (useful when dealing with shell code), how it translated the buffers to binary data, and how it will apply the mask at runtime. Some example output: $ echo ':qemu-foo:M::\x7fELF\xAD\xAD\x01\x00:\xff\xff\xff\xff\xff\x00\xff\x00:/usr/bin/qemu-foo:POC' > register $ dmesg binfmt_misc: register: received 92 bytes binfmt_misc: register: delim: 0x3a {:} binfmt_misc: register: name: {qemu-foo} binfmt_misc: register: type: M (magic) binfmt_misc: register: offset: 0x0 binfmt_misc: register: magic[raw]: 5c 78 37 66 45 4c 46 5c 78 41 44 5c 78 41 44 5c \x7fELF\xAD\xAD\ binfmt_misc: register: magic[raw]: 78 30 31 5c 78 30 30 00 x01\x00. binfmt_misc: register: mask[raw]: 5c 78 66 66 5c 78 66 66 5c 78 66 66 5c 78 66 66 \xff\xff\xff\xff binfmt_misc: register: mask[raw]: 5c 78 66 66 5c 78 30 30 5c 78 66 66 5c 78 30 30 \xff\x00\xff\x00 binfmt_misc: register: mask[raw]: 00 . binfmt_misc: register: magic/mask length: 8 binfmt_misc: register: magic[decoded]: 7f 45 4c 46 ad ad 01 00 .ELF.... binfmt_misc: register: mask[decoded]: ff ff ff ff ff 00 ff 00 ........ binfmt_misc: register: magic[masked]: 7f 45 4c 46 ad 00 01 00 .ELF.... binfmt_misc: register: interpreter: {/usr/bin/qemu-foo} binfmt_misc: register: flag: P (preserve argv0) binfmt_misc: register: flag: O (open binary) binfmt_misc: register: flag: C (preserve creds) The [raw] lines show us exactly what was received from userspace. The lines after that show us how the kernel has decoded things. Signed-off-by: Mike Frysinger <vapier@gentoo.org> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Joe Perches <joe@perches.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-12-10 15:52:08 -08:00
/* Parse the 'interpreter' field. */
e->interpreter = p;
p = strchr(p, del);
if (!p)
goto einval;
*p++ = '\0';
if (!e->interpreter[0])
goto einval;
binfmt_misc: add comments & debug logs When trying to develop a custom format handler, the errors returned all effectively get bucketed as EINVAL with no kernel messages. The other errors (ENOMEM/EFAULT) are internal/obvious and basic. Thus any time a bad handler is rejected, the developer has to walk the dense code and try to guess where it went wrong. Needing to dive into kernel code is itself a fairly high barrier for a lot of people. To improve this situation, let's deploy extensive pr_debug markers at logical parse points, and add comments to the dense parsing logic. It let's you see exactly where the parsing aborts, the string the kernel received (useful when dealing with shell code), how it translated the buffers to binary data, and how it will apply the mask at runtime. Some example output: $ echo ':qemu-foo:M::\x7fELF\xAD\xAD\x01\x00:\xff\xff\xff\xff\xff\x00\xff\x00:/usr/bin/qemu-foo:POC' > register $ dmesg binfmt_misc: register: received 92 bytes binfmt_misc: register: delim: 0x3a {:} binfmt_misc: register: name: {qemu-foo} binfmt_misc: register: type: M (magic) binfmt_misc: register: offset: 0x0 binfmt_misc: register: magic[raw]: 5c 78 37 66 45 4c 46 5c 78 41 44 5c 78 41 44 5c \x7fELF\xAD\xAD\ binfmt_misc: register: magic[raw]: 78 30 31 5c 78 30 30 00 x01\x00. binfmt_misc: register: mask[raw]: 5c 78 66 66 5c 78 66 66 5c 78 66 66 5c 78 66 66 \xff\xff\xff\xff binfmt_misc: register: mask[raw]: 5c 78 66 66 5c 78 30 30 5c 78 66 66 5c 78 30 30 \xff\x00\xff\x00 binfmt_misc: register: mask[raw]: 00 . binfmt_misc: register: magic/mask length: 8 binfmt_misc: register: magic[decoded]: 7f 45 4c 46 ad ad 01 00 .ELF.... binfmt_misc: register: mask[decoded]: ff ff ff ff ff 00 ff 00 ........ binfmt_misc: register: magic[masked]: 7f 45 4c 46 ad 00 01 00 .ELF.... binfmt_misc: register: interpreter: {/usr/bin/qemu-foo} binfmt_misc: register: flag: P (preserve argv0) binfmt_misc: register: flag: O (open binary) binfmt_misc: register: flag: C (preserve creds) The [raw] lines show us exactly what was received from userspace. The lines after that show us how the kernel has decoded things. Signed-off-by: Mike Frysinger <vapier@gentoo.org> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Joe Perches <joe@perches.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-12-10 15:52:08 -08:00
pr_debug("register: interpreter: {%s}\n", e->interpreter);
binfmt_misc: add comments & debug logs When trying to develop a custom format handler, the errors returned all effectively get bucketed as EINVAL with no kernel messages. The other errors (ENOMEM/EFAULT) are internal/obvious and basic. Thus any time a bad handler is rejected, the developer has to walk the dense code and try to guess where it went wrong. Needing to dive into kernel code is itself a fairly high barrier for a lot of people. To improve this situation, let's deploy extensive pr_debug markers at logical parse points, and add comments to the dense parsing logic. It let's you see exactly where the parsing aborts, the string the kernel received (useful when dealing with shell code), how it translated the buffers to binary data, and how it will apply the mask at runtime. Some example output: $ echo ':qemu-foo:M::\x7fELF\xAD\xAD\x01\x00:\xff\xff\xff\xff\xff\x00\xff\x00:/usr/bin/qemu-foo:POC' > register $ dmesg binfmt_misc: register: received 92 bytes binfmt_misc: register: delim: 0x3a {:} binfmt_misc: register: name: {qemu-foo} binfmt_misc: register: type: M (magic) binfmt_misc: register: offset: 0x0 binfmt_misc: register: magic[raw]: 5c 78 37 66 45 4c 46 5c 78 41 44 5c 78 41 44 5c \x7fELF\xAD\xAD\ binfmt_misc: register: magic[raw]: 78 30 31 5c 78 30 30 00 x01\x00. binfmt_misc: register: mask[raw]: 5c 78 66 66 5c 78 66 66 5c 78 66 66 5c 78 66 66 \xff\xff\xff\xff binfmt_misc: register: mask[raw]: 5c 78 66 66 5c 78 30 30 5c 78 66 66 5c 78 30 30 \xff\x00\xff\x00 binfmt_misc: register: mask[raw]: 00 . binfmt_misc: register: magic/mask length: 8 binfmt_misc: register: magic[decoded]: 7f 45 4c 46 ad ad 01 00 .ELF.... binfmt_misc: register: mask[decoded]: ff ff ff ff ff 00 ff 00 ........ binfmt_misc: register: magic[masked]: 7f 45 4c 46 ad 00 01 00 .ELF.... binfmt_misc: register: interpreter: {/usr/bin/qemu-foo} binfmt_misc: register: flag: P (preserve argv0) binfmt_misc: register: flag: O (open binary) binfmt_misc: register: flag: C (preserve creds) The [raw] lines show us exactly what was received from userspace. The lines after that show us how the kernel has decoded things. Signed-off-by: Mike Frysinger <vapier@gentoo.org> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Joe Perches <joe@perches.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-12-10 15:52:08 -08:00
/* Parse the 'flags' field. */
p = check_special_flags(p, e);
if (*p == '\n')
p++;
if (p != buf + count)
goto einval;
return e;
out:
return ERR_PTR(err);
efault:
kfree(e);
return ERR_PTR(-EFAULT);
einval:
kfree(e);
return ERR_PTR(-EINVAL);
}
/*
* Set status of entry/binfmt_misc:
* '1' enables, '0' disables and '-1' clears entry/binfmt_misc
*/
static int parse_command(const char __user *buffer, size_t count)
{
char s[4];
if (count > 3)
return -EINVAL;
if (copy_from_user(s, buffer, count))
return -EFAULT;
if (!count)
return 0;
if (s[count - 1] == '\n')
count--;
if (count == 1 && s[0] == '0')
return 1;
if (count == 1 && s[0] == '1')
return 2;
if (count == 2 && s[0] == '-' && s[1] == '1')
return 3;
return -EINVAL;
}
/* generic stuff */
static void entry_status(Node *e, char *page)
{
char *dp = page;
const char *status = "disabled";
if (test_bit(Enabled, &e->flags))
status = "enabled";
if (!VERBOSE_STATUS) {
sprintf(page, "%s\n", status);
return;
}
dp += sprintf(dp, "%s\ninterpreter %s\n", status, e->interpreter);
/* print the special flags */
dp += sprintf(dp, "flags: ");
if (e->flags & MISC_FMT_PRESERVE_ARGV0)
*dp++ = 'P';
if (e->flags & MISC_FMT_OPEN_BINARY)
*dp++ = 'O';
if (e->flags & MISC_FMT_CREDENTIALS)
*dp++ = 'C';
if (e->flags & MISC_FMT_OPEN_FILE)
*dp++ = 'F';
*dp++ = '\n';
if (!test_bit(Magic, &e->flags)) {
sprintf(dp, "extension .%s\n", e->magic);
} else {
dp += sprintf(dp, "offset %i\nmagic ", e->offset);
dp = bin2hex(dp, e->magic, e->size);
if (e->mask) {
dp += sprintf(dp, "\nmask ");
dp = bin2hex(dp, e->mask, e->size);
}
*dp++ = '\n';
*dp = '\0';
}
}
static struct inode *bm_get_inode(struct super_block *sb, int mode)
{
struct inode *inode = new_inode(sb);
if (inode) {
inode->i_ino = get_next_ino();
inode->i_mode = mode;
inode->i_atime = inode->i_mtime = inode->i_ctime =
current_time(inode);
}
return inode;
}
binfmt_misc: cleanup on filesystem umount [ Upstream commit 1c5976ef0f7ad76319df748ccb99a4c7ba2ba464 ] Currently, registering a new binary type pins the binfmt_misc filesystem. Specifically, this means that as long as there is at least one binary type registered the binfmt_misc filesystem survives all umounts, i.e. the superblock is not destroyed. Meaning that a umount followed by another mount will end up with the same superblock and the same binary type handlers. This is a behavior we tend to discourage for any new filesystems (apart from a few special filesystems such as e.g. configfs or debugfs). A umount operation without the filesystem being pinned - by e.g. someone holding a file descriptor to an open file - should usually result in the destruction of the superblock and all associated resources. This makes introspection easier and leads to clearly defined, simple and clean semantics. An administrator can rely on the fact that a umount will guarantee a clean slate making it possible to reinitialize a filesystem. Right now all binary types would need to be explicitly deleted before that can happen. This allows us to remove the heavy-handed calls to simple_pin_fs() and simple_release_fs() when creating and deleting binary types. This in turn allows us to replace the current brittle pinning mechanism abusing dget() which has caused a range of bugs judging from prior fixes in [2] and [3]. The additional dget() in load_misc_binary() pins the dentry but only does so for the sake to prevent ->evict_inode() from freeing the node when a user removes the binary type and kill_node() is run. Which would mean ->interpreter and ->interp_file would be freed causing a UAF. This isn't really nicely documented nor is it very clean because it relies on simple_pin_fs() pinning the filesystem as long as at least one binary type exists. Otherwise it would cause load_misc_binary() to hold on to a dentry belonging to a superblock that has been shutdown. Replace that implicit pinning with a clean and simple per-node refcount and get rid of the ugly dget() pinning. A similar mechanism exists for e.g. binderfs (cf. [4]). All the cleanup work can now be done in ->evict_inode(). In a follow-up patch we will make it possible to use binfmt_misc in sandboxes. We will use the cleaner semantics where a umount for the filesystem will cause the superblock and all resources to be deallocated. In preparation for this apply the same semantics to the initial binfmt_misc mount. Note, that this is a user-visible change and as such a uapi change but one that we can reasonably risk. We've discussed this in earlier versions of this patchset (cf. [1]). The main user and provider of binfmt_misc is systemd. Systemd provides binfmt_misc via autofs since it is configurable as a kernel module and is used by a few exotic packages and users. As such a binfmt_misc mount is triggered when /proc/sys/fs/binfmt_misc is accessed and is only provided on demand. Other autofs on demand filesystems include EFI ESP which systemd umounts if the mountpoint stays idle for a certain amount of time. This doesn't apply to the binfmt_misc autofs mount which isn't touched once it is mounted meaning this change can't accidently wipe binary type handlers without someone having explicitly unmounted binfmt_misc. After speaking to systemd folks they don't expect this change to affect them. In line with our general policy, if we see a regression for systemd or other users with this change we will switch back to the old behavior for the initial binfmt_misc mount and have binary types pin the filesystem again. But while we touch this code let's take the chance and let's improve on the status quo. [1]: https://lore.kernel.org/r/20191216091220.465626-2-laurent@vivier.eu [2]: commit 43a4f2619038 ("exec: binfmt_misc: fix race between load_misc_binary() and kill_node()" [3]: commit 83f918274e4b ("exec: binfmt_misc: shift filp_close(interp_file) from kill_node() to bm_evict_inode()") [4]: commit f0fe2c0f050d ("binder: prevent UAF for binderfs devices II") Link: https://lore.kernel.org/r/20211028103114.2849140-1-brauner@kernel.org (v1) Cc: Sargun Dhillon <sargun@sargun.me> Cc: Serge Hallyn <serge@hallyn.com> Cc: Jann Horn <jannh@google.com> Cc: Henning Schild <henning.schild@siemens.com> Cc: Andrei Vagin <avagin@gmail.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Laurent Vivier <laurent@vivier.eu> Cc: linux-fsdevel@vger.kernel.org Acked-by: Serge Hallyn <serge@hallyn.com> Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com> Signed-off-by: Christian Brauner <brauner@kernel.org> Signed-off-by: Kees Cook <keescook@chromium.org> Signed-off-by: Sasha Levin <sashal@kernel.org> (cherry picked from commit 263bcebf5c2ab1fe949517225157f34015124620) Signed-off-by: Vegard Nossum <vegard.nossum@oracle.com>
2021-10-28 12:31:13 +02:00
/**
* bm_evict_inode - cleanup data associated with @inode
* @inode: inode to which the data is attached
*
* Cleanup the binary type handler data associated with @inode if a binary type
* entry is removed or the filesystem is unmounted and the super block is
* shutdown.
*
* If the ->evict call was not caused by a super block shutdown but by a write
* to remove the entry or all entries via bm_{entry,status}_write() the entry
* will have already been removed from the list. We keep the list_empty() check
* to make that explicit.
*/
static void bm_evict_inode(struct inode *inode)
{
Node *e = inode->i_private;
clear_inode(inode);
binfmt_misc: cleanup on filesystem umount [ Upstream commit 1c5976ef0f7ad76319df748ccb99a4c7ba2ba464 ] Currently, registering a new binary type pins the binfmt_misc filesystem. Specifically, this means that as long as there is at least one binary type registered the binfmt_misc filesystem survives all umounts, i.e. the superblock is not destroyed. Meaning that a umount followed by another mount will end up with the same superblock and the same binary type handlers. This is a behavior we tend to discourage for any new filesystems (apart from a few special filesystems such as e.g. configfs or debugfs). A umount operation without the filesystem being pinned - by e.g. someone holding a file descriptor to an open file - should usually result in the destruction of the superblock and all associated resources. This makes introspection easier and leads to clearly defined, simple and clean semantics. An administrator can rely on the fact that a umount will guarantee a clean slate making it possible to reinitialize a filesystem. Right now all binary types would need to be explicitly deleted before that can happen. This allows us to remove the heavy-handed calls to simple_pin_fs() and simple_release_fs() when creating and deleting binary types. This in turn allows us to replace the current brittle pinning mechanism abusing dget() which has caused a range of bugs judging from prior fixes in [2] and [3]. The additional dget() in load_misc_binary() pins the dentry but only does so for the sake to prevent ->evict_inode() from freeing the node when a user removes the binary type and kill_node() is run. Which would mean ->interpreter and ->interp_file would be freed causing a UAF. This isn't really nicely documented nor is it very clean because it relies on simple_pin_fs() pinning the filesystem as long as at least one binary type exists. Otherwise it would cause load_misc_binary() to hold on to a dentry belonging to a superblock that has been shutdown. Replace that implicit pinning with a clean and simple per-node refcount and get rid of the ugly dget() pinning. A similar mechanism exists for e.g. binderfs (cf. [4]). All the cleanup work can now be done in ->evict_inode(). In a follow-up patch we will make it possible to use binfmt_misc in sandboxes. We will use the cleaner semantics where a umount for the filesystem will cause the superblock and all resources to be deallocated. In preparation for this apply the same semantics to the initial binfmt_misc mount. Note, that this is a user-visible change and as such a uapi change but one that we can reasonably risk. We've discussed this in earlier versions of this patchset (cf. [1]). The main user and provider of binfmt_misc is systemd. Systemd provides binfmt_misc via autofs since it is configurable as a kernel module and is used by a few exotic packages and users. As such a binfmt_misc mount is triggered when /proc/sys/fs/binfmt_misc is accessed and is only provided on demand. Other autofs on demand filesystems include EFI ESP which systemd umounts if the mountpoint stays idle for a certain amount of time. This doesn't apply to the binfmt_misc autofs mount which isn't touched once it is mounted meaning this change can't accidently wipe binary type handlers without someone having explicitly unmounted binfmt_misc. After speaking to systemd folks they don't expect this change to affect them. In line with our general policy, if we see a regression for systemd or other users with this change we will switch back to the old behavior for the initial binfmt_misc mount and have binary types pin the filesystem again. But while we touch this code let's take the chance and let's improve on the status quo. [1]: https://lore.kernel.org/r/20191216091220.465626-2-laurent@vivier.eu [2]: commit 43a4f2619038 ("exec: binfmt_misc: fix race between load_misc_binary() and kill_node()" [3]: commit 83f918274e4b ("exec: binfmt_misc: shift filp_close(interp_file) from kill_node() to bm_evict_inode()") [4]: commit f0fe2c0f050d ("binder: prevent UAF for binderfs devices II") Link: https://lore.kernel.org/r/20211028103114.2849140-1-brauner@kernel.org (v1) Cc: Sargun Dhillon <sargun@sargun.me> Cc: Serge Hallyn <serge@hallyn.com> Cc: Jann Horn <jannh@google.com> Cc: Henning Schild <henning.schild@siemens.com> Cc: Andrei Vagin <avagin@gmail.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Laurent Vivier <laurent@vivier.eu> Cc: linux-fsdevel@vger.kernel.org Acked-by: Serge Hallyn <serge@hallyn.com> Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com> Signed-off-by: Christian Brauner <brauner@kernel.org> Signed-off-by: Kees Cook <keescook@chromium.org> Signed-off-by: Sasha Levin <sashal@kernel.org> (cherry picked from commit 263bcebf5c2ab1fe949517225157f34015124620) Signed-off-by: Vegard Nossum <vegard.nossum@oracle.com>
2021-10-28 12:31:13 +02:00
if (e) {
write_lock(&entries_lock);
if (!list_empty(&e->list))
list_del_init(&e->list);
write_unlock(&entries_lock);
put_binfmt_handler(e);
}
}
binfmt_misc: cleanup on filesystem umount [ Upstream commit 1c5976ef0f7ad76319df748ccb99a4c7ba2ba464 ] Currently, registering a new binary type pins the binfmt_misc filesystem. Specifically, this means that as long as there is at least one binary type registered the binfmt_misc filesystem survives all umounts, i.e. the superblock is not destroyed. Meaning that a umount followed by another mount will end up with the same superblock and the same binary type handlers. This is a behavior we tend to discourage for any new filesystems (apart from a few special filesystems such as e.g. configfs or debugfs). A umount operation without the filesystem being pinned - by e.g. someone holding a file descriptor to an open file - should usually result in the destruction of the superblock and all associated resources. This makes introspection easier and leads to clearly defined, simple and clean semantics. An administrator can rely on the fact that a umount will guarantee a clean slate making it possible to reinitialize a filesystem. Right now all binary types would need to be explicitly deleted before that can happen. This allows us to remove the heavy-handed calls to simple_pin_fs() and simple_release_fs() when creating and deleting binary types. This in turn allows us to replace the current brittle pinning mechanism abusing dget() which has caused a range of bugs judging from prior fixes in [2] and [3]. The additional dget() in load_misc_binary() pins the dentry but only does so for the sake to prevent ->evict_inode() from freeing the node when a user removes the binary type and kill_node() is run. Which would mean ->interpreter and ->interp_file would be freed causing a UAF. This isn't really nicely documented nor is it very clean because it relies on simple_pin_fs() pinning the filesystem as long as at least one binary type exists. Otherwise it would cause load_misc_binary() to hold on to a dentry belonging to a superblock that has been shutdown. Replace that implicit pinning with a clean and simple per-node refcount and get rid of the ugly dget() pinning. A similar mechanism exists for e.g. binderfs (cf. [4]). All the cleanup work can now be done in ->evict_inode(). In a follow-up patch we will make it possible to use binfmt_misc in sandboxes. We will use the cleaner semantics where a umount for the filesystem will cause the superblock and all resources to be deallocated. In preparation for this apply the same semantics to the initial binfmt_misc mount. Note, that this is a user-visible change and as such a uapi change but one that we can reasonably risk. We've discussed this in earlier versions of this patchset (cf. [1]). The main user and provider of binfmt_misc is systemd. Systemd provides binfmt_misc via autofs since it is configurable as a kernel module and is used by a few exotic packages and users. As such a binfmt_misc mount is triggered when /proc/sys/fs/binfmt_misc is accessed and is only provided on demand. Other autofs on demand filesystems include EFI ESP which systemd umounts if the mountpoint stays idle for a certain amount of time. This doesn't apply to the binfmt_misc autofs mount which isn't touched once it is mounted meaning this change can't accidently wipe binary type handlers without someone having explicitly unmounted binfmt_misc. After speaking to systemd folks they don't expect this change to affect them. In line with our general policy, if we see a regression for systemd or other users with this change we will switch back to the old behavior for the initial binfmt_misc mount and have binary types pin the filesystem again. But while we touch this code let's take the chance and let's improve on the status quo. [1]: https://lore.kernel.org/r/20191216091220.465626-2-laurent@vivier.eu [2]: commit 43a4f2619038 ("exec: binfmt_misc: fix race between load_misc_binary() and kill_node()" [3]: commit 83f918274e4b ("exec: binfmt_misc: shift filp_close(interp_file) from kill_node() to bm_evict_inode()") [4]: commit f0fe2c0f050d ("binder: prevent UAF for binderfs devices II") Link: https://lore.kernel.org/r/20211028103114.2849140-1-brauner@kernel.org (v1) Cc: Sargun Dhillon <sargun@sargun.me> Cc: Serge Hallyn <serge@hallyn.com> Cc: Jann Horn <jannh@google.com> Cc: Henning Schild <henning.schild@siemens.com> Cc: Andrei Vagin <avagin@gmail.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Laurent Vivier <laurent@vivier.eu> Cc: linux-fsdevel@vger.kernel.org Acked-by: Serge Hallyn <serge@hallyn.com> Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com> Signed-off-by: Christian Brauner <brauner@kernel.org> Signed-off-by: Kees Cook <keescook@chromium.org> Signed-off-by: Sasha Levin <sashal@kernel.org> (cherry picked from commit 263bcebf5c2ab1fe949517225157f34015124620) Signed-off-by: Vegard Nossum <vegard.nossum@oracle.com>
2021-10-28 12:31:13 +02:00
/**
* unlink_binfmt_dentry - remove the dentry for the binary type handler
* @dentry: dentry associated with the binary type handler
*
* Do the actual filesystem work to remove a dentry for a registered binary
* type handler. Since binfmt_misc only allows simple files to be created
* directly under the root dentry of the filesystem we ensure that we are
* indeed passed a dentry directly beneath the root dentry, that the inode
* associated with the root dentry is locked, and that it is a regular file we
* are asked to remove.
*/
static void unlink_binfmt_dentry(struct dentry *dentry)
{
binfmt_misc: cleanup on filesystem umount [ Upstream commit 1c5976ef0f7ad76319df748ccb99a4c7ba2ba464 ] Currently, registering a new binary type pins the binfmt_misc filesystem. Specifically, this means that as long as there is at least one binary type registered the binfmt_misc filesystem survives all umounts, i.e. the superblock is not destroyed. Meaning that a umount followed by another mount will end up with the same superblock and the same binary type handlers. This is a behavior we tend to discourage for any new filesystems (apart from a few special filesystems such as e.g. configfs or debugfs). A umount operation without the filesystem being pinned - by e.g. someone holding a file descriptor to an open file - should usually result in the destruction of the superblock and all associated resources. This makes introspection easier and leads to clearly defined, simple and clean semantics. An administrator can rely on the fact that a umount will guarantee a clean slate making it possible to reinitialize a filesystem. Right now all binary types would need to be explicitly deleted before that can happen. This allows us to remove the heavy-handed calls to simple_pin_fs() and simple_release_fs() when creating and deleting binary types. This in turn allows us to replace the current brittle pinning mechanism abusing dget() which has caused a range of bugs judging from prior fixes in [2] and [3]. The additional dget() in load_misc_binary() pins the dentry but only does so for the sake to prevent ->evict_inode() from freeing the node when a user removes the binary type and kill_node() is run. Which would mean ->interpreter and ->interp_file would be freed causing a UAF. This isn't really nicely documented nor is it very clean because it relies on simple_pin_fs() pinning the filesystem as long as at least one binary type exists. Otherwise it would cause load_misc_binary() to hold on to a dentry belonging to a superblock that has been shutdown. Replace that implicit pinning with a clean and simple per-node refcount and get rid of the ugly dget() pinning. A similar mechanism exists for e.g. binderfs (cf. [4]). All the cleanup work can now be done in ->evict_inode(). In a follow-up patch we will make it possible to use binfmt_misc in sandboxes. We will use the cleaner semantics where a umount for the filesystem will cause the superblock and all resources to be deallocated. In preparation for this apply the same semantics to the initial binfmt_misc mount. Note, that this is a user-visible change and as such a uapi change but one that we can reasonably risk. We've discussed this in earlier versions of this patchset (cf. [1]). The main user and provider of binfmt_misc is systemd. Systemd provides binfmt_misc via autofs since it is configurable as a kernel module and is used by a few exotic packages and users. As such a binfmt_misc mount is triggered when /proc/sys/fs/binfmt_misc is accessed and is only provided on demand. Other autofs on demand filesystems include EFI ESP which systemd umounts if the mountpoint stays idle for a certain amount of time. This doesn't apply to the binfmt_misc autofs mount which isn't touched once it is mounted meaning this change can't accidently wipe binary type handlers without someone having explicitly unmounted binfmt_misc. After speaking to systemd folks they don't expect this change to affect them. In line with our general policy, if we see a regression for systemd or other users with this change we will switch back to the old behavior for the initial binfmt_misc mount and have binary types pin the filesystem again. But while we touch this code let's take the chance and let's improve on the status quo. [1]: https://lore.kernel.org/r/20191216091220.465626-2-laurent@vivier.eu [2]: commit 43a4f2619038 ("exec: binfmt_misc: fix race between load_misc_binary() and kill_node()" [3]: commit 83f918274e4b ("exec: binfmt_misc: shift filp_close(interp_file) from kill_node() to bm_evict_inode()") [4]: commit f0fe2c0f050d ("binder: prevent UAF for binderfs devices II") Link: https://lore.kernel.org/r/20211028103114.2849140-1-brauner@kernel.org (v1) Cc: Sargun Dhillon <sargun@sargun.me> Cc: Serge Hallyn <serge@hallyn.com> Cc: Jann Horn <jannh@google.com> Cc: Henning Schild <henning.schild@siemens.com> Cc: Andrei Vagin <avagin@gmail.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Laurent Vivier <laurent@vivier.eu> Cc: linux-fsdevel@vger.kernel.org Acked-by: Serge Hallyn <serge@hallyn.com> Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com> Signed-off-by: Christian Brauner <brauner@kernel.org> Signed-off-by: Kees Cook <keescook@chromium.org> Signed-off-by: Sasha Levin <sashal@kernel.org> (cherry picked from commit 263bcebf5c2ab1fe949517225157f34015124620) Signed-off-by: Vegard Nossum <vegard.nossum@oracle.com>
2021-10-28 12:31:13 +02:00
struct dentry *parent = dentry->d_parent;
struct inode *inode, *parent_inode;
/* All entries are immediate descendants of the root dentry. */
if (WARN_ON_ONCE(dentry->d_sb->s_root != parent))
return;
binfmt_misc: cleanup on filesystem umount [ Upstream commit 1c5976ef0f7ad76319df748ccb99a4c7ba2ba464 ] Currently, registering a new binary type pins the binfmt_misc filesystem. Specifically, this means that as long as there is at least one binary type registered the binfmt_misc filesystem survives all umounts, i.e. the superblock is not destroyed. Meaning that a umount followed by another mount will end up with the same superblock and the same binary type handlers. This is a behavior we tend to discourage for any new filesystems (apart from a few special filesystems such as e.g. configfs or debugfs). A umount operation without the filesystem being pinned - by e.g. someone holding a file descriptor to an open file - should usually result in the destruction of the superblock and all associated resources. This makes introspection easier and leads to clearly defined, simple and clean semantics. An administrator can rely on the fact that a umount will guarantee a clean slate making it possible to reinitialize a filesystem. Right now all binary types would need to be explicitly deleted before that can happen. This allows us to remove the heavy-handed calls to simple_pin_fs() and simple_release_fs() when creating and deleting binary types. This in turn allows us to replace the current brittle pinning mechanism abusing dget() which has caused a range of bugs judging from prior fixes in [2] and [3]. The additional dget() in load_misc_binary() pins the dentry but only does so for the sake to prevent ->evict_inode() from freeing the node when a user removes the binary type and kill_node() is run. Which would mean ->interpreter and ->interp_file would be freed causing a UAF. This isn't really nicely documented nor is it very clean because it relies on simple_pin_fs() pinning the filesystem as long as at least one binary type exists. Otherwise it would cause load_misc_binary() to hold on to a dentry belonging to a superblock that has been shutdown. Replace that implicit pinning with a clean and simple per-node refcount and get rid of the ugly dget() pinning. A similar mechanism exists for e.g. binderfs (cf. [4]). All the cleanup work can now be done in ->evict_inode(). In a follow-up patch we will make it possible to use binfmt_misc in sandboxes. We will use the cleaner semantics where a umount for the filesystem will cause the superblock and all resources to be deallocated. In preparation for this apply the same semantics to the initial binfmt_misc mount. Note, that this is a user-visible change and as such a uapi change but one that we can reasonably risk. We've discussed this in earlier versions of this patchset (cf. [1]). The main user and provider of binfmt_misc is systemd. Systemd provides binfmt_misc via autofs since it is configurable as a kernel module and is used by a few exotic packages and users. As such a binfmt_misc mount is triggered when /proc/sys/fs/binfmt_misc is accessed and is only provided on demand. Other autofs on demand filesystems include EFI ESP which systemd umounts if the mountpoint stays idle for a certain amount of time. This doesn't apply to the binfmt_misc autofs mount which isn't touched once it is mounted meaning this change can't accidently wipe binary type handlers without someone having explicitly unmounted binfmt_misc. After speaking to systemd folks they don't expect this change to affect them. In line with our general policy, if we see a regression for systemd or other users with this change we will switch back to the old behavior for the initial binfmt_misc mount and have binary types pin the filesystem again. But while we touch this code let's take the chance and let's improve on the status quo. [1]: https://lore.kernel.org/r/20191216091220.465626-2-laurent@vivier.eu [2]: commit 43a4f2619038 ("exec: binfmt_misc: fix race between load_misc_binary() and kill_node()" [3]: commit 83f918274e4b ("exec: binfmt_misc: shift filp_close(interp_file) from kill_node() to bm_evict_inode()") [4]: commit f0fe2c0f050d ("binder: prevent UAF for binderfs devices II") Link: https://lore.kernel.org/r/20211028103114.2849140-1-brauner@kernel.org (v1) Cc: Sargun Dhillon <sargun@sargun.me> Cc: Serge Hallyn <serge@hallyn.com> Cc: Jann Horn <jannh@google.com> Cc: Henning Schild <henning.schild@siemens.com> Cc: Andrei Vagin <avagin@gmail.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Laurent Vivier <laurent@vivier.eu> Cc: linux-fsdevel@vger.kernel.org Acked-by: Serge Hallyn <serge@hallyn.com> Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com> Signed-off-by: Christian Brauner <brauner@kernel.org> Signed-off-by: Kees Cook <keescook@chromium.org> Signed-off-by: Sasha Levin <sashal@kernel.org> (cherry picked from commit 263bcebf5c2ab1fe949517225157f34015124620) Signed-off-by: Vegard Nossum <vegard.nossum@oracle.com>
2021-10-28 12:31:13 +02:00
/* We only expect to be called on regular files. */
inode = d_inode(dentry);
if (WARN_ON_ONCE(!S_ISREG(inode->i_mode)))
return;
/* The parent inode must be locked. */
parent_inode = d_inode(parent);
if (WARN_ON_ONCE(!inode_is_locked(parent_inode)))
return;
if (simple_positive(dentry)) {
dget(dentry);
simple_unlink(parent_inode, dentry);
d_delete(dentry);
dput(dentry);
}
}
/**
* remove_binfmt_handler - remove a binary type handler
* @misc: handle to binfmt_misc instance
* @e: binary type handler to remove
*
* Remove a binary type handler from the list of binary type handlers and
* remove its associated dentry. This is called from
* binfmt_{entry,status}_write(). In the future, we might want to think about
* adding a proper ->unlink() method to binfmt_misc instead of forcing caller's
* to use writes to files in order to delete binary type handlers. But it has
* worked for so long that it's not a pressing issue.
*/
static void remove_binfmt_handler(Node *e)
{
write_lock(&entries_lock);
list_del_init(&e->list);
write_unlock(&entries_lock);
binfmt_misc: cleanup on filesystem umount [ Upstream commit 1c5976ef0f7ad76319df748ccb99a4c7ba2ba464 ] Currently, registering a new binary type pins the binfmt_misc filesystem. Specifically, this means that as long as there is at least one binary type registered the binfmt_misc filesystem survives all umounts, i.e. the superblock is not destroyed. Meaning that a umount followed by another mount will end up with the same superblock and the same binary type handlers. This is a behavior we tend to discourage for any new filesystems (apart from a few special filesystems such as e.g. configfs or debugfs). A umount operation without the filesystem being pinned - by e.g. someone holding a file descriptor to an open file - should usually result in the destruction of the superblock and all associated resources. This makes introspection easier and leads to clearly defined, simple and clean semantics. An administrator can rely on the fact that a umount will guarantee a clean slate making it possible to reinitialize a filesystem. Right now all binary types would need to be explicitly deleted before that can happen. This allows us to remove the heavy-handed calls to simple_pin_fs() and simple_release_fs() when creating and deleting binary types. This in turn allows us to replace the current brittle pinning mechanism abusing dget() which has caused a range of bugs judging from prior fixes in [2] and [3]. The additional dget() in load_misc_binary() pins the dentry but only does so for the sake to prevent ->evict_inode() from freeing the node when a user removes the binary type and kill_node() is run. Which would mean ->interpreter and ->interp_file would be freed causing a UAF. This isn't really nicely documented nor is it very clean because it relies on simple_pin_fs() pinning the filesystem as long as at least one binary type exists. Otherwise it would cause load_misc_binary() to hold on to a dentry belonging to a superblock that has been shutdown. Replace that implicit pinning with a clean and simple per-node refcount and get rid of the ugly dget() pinning. A similar mechanism exists for e.g. binderfs (cf. [4]). All the cleanup work can now be done in ->evict_inode(). In a follow-up patch we will make it possible to use binfmt_misc in sandboxes. We will use the cleaner semantics where a umount for the filesystem will cause the superblock and all resources to be deallocated. In preparation for this apply the same semantics to the initial binfmt_misc mount. Note, that this is a user-visible change and as such a uapi change but one that we can reasonably risk. We've discussed this in earlier versions of this patchset (cf. [1]). The main user and provider of binfmt_misc is systemd. Systemd provides binfmt_misc via autofs since it is configurable as a kernel module and is used by a few exotic packages and users. As such a binfmt_misc mount is triggered when /proc/sys/fs/binfmt_misc is accessed and is only provided on demand. Other autofs on demand filesystems include EFI ESP which systemd umounts if the mountpoint stays idle for a certain amount of time. This doesn't apply to the binfmt_misc autofs mount which isn't touched once it is mounted meaning this change can't accidently wipe binary type handlers without someone having explicitly unmounted binfmt_misc. After speaking to systemd folks they don't expect this change to affect them. In line with our general policy, if we see a regression for systemd or other users with this change we will switch back to the old behavior for the initial binfmt_misc mount and have binary types pin the filesystem again. But while we touch this code let's take the chance and let's improve on the status quo. [1]: https://lore.kernel.org/r/20191216091220.465626-2-laurent@vivier.eu [2]: commit 43a4f2619038 ("exec: binfmt_misc: fix race between load_misc_binary() and kill_node()" [3]: commit 83f918274e4b ("exec: binfmt_misc: shift filp_close(interp_file) from kill_node() to bm_evict_inode()") [4]: commit f0fe2c0f050d ("binder: prevent UAF for binderfs devices II") Link: https://lore.kernel.org/r/20211028103114.2849140-1-brauner@kernel.org (v1) Cc: Sargun Dhillon <sargun@sargun.me> Cc: Serge Hallyn <serge@hallyn.com> Cc: Jann Horn <jannh@google.com> Cc: Henning Schild <henning.schild@siemens.com> Cc: Andrei Vagin <avagin@gmail.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Laurent Vivier <laurent@vivier.eu> Cc: linux-fsdevel@vger.kernel.org Acked-by: Serge Hallyn <serge@hallyn.com> Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com> Signed-off-by: Christian Brauner <brauner@kernel.org> Signed-off-by: Kees Cook <keescook@chromium.org> Signed-off-by: Sasha Levin <sashal@kernel.org> (cherry picked from commit 263bcebf5c2ab1fe949517225157f34015124620) Signed-off-by: Vegard Nossum <vegard.nossum@oracle.com>
2021-10-28 12:31:13 +02:00
unlink_binfmt_dentry(e->dentry);
}
/* /<entry> */
static ssize_t
bm_entry_read(struct file *file, char __user *buf, size_t nbytes, loff_t *ppos)
{
Node *e = file_inode(file)->i_private;
ssize_t res;
char *page;
page = (char *) __get_free_page(GFP_KERNEL);
if (!page)
return -ENOMEM;
entry_status(e, page);
res = simple_read_from_buffer(buf, nbytes, ppos, page, strlen(page));
free_page((unsigned long) page);
return res;
}
static ssize_t bm_entry_write(struct file *file, const char __user *buffer,
size_t count, loff_t *ppos)
{
binfmt_misc: cleanup on filesystem umount [ Upstream commit 1c5976ef0f7ad76319df748ccb99a4c7ba2ba464 ] Currently, registering a new binary type pins the binfmt_misc filesystem. Specifically, this means that as long as there is at least one binary type registered the binfmt_misc filesystem survives all umounts, i.e. the superblock is not destroyed. Meaning that a umount followed by another mount will end up with the same superblock and the same binary type handlers. This is a behavior we tend to discourage for any new filesystems (apart from a few special filesystems such as e.g. configfs or debugfs). A umount operation without the filesystem being pinned - by e.g. someone holding a file descriptor to an open file - should usually result in the destruction of the superblock and all associated resources. This makes introspection easier and leads to clearly defined, simple and clean semantics. An administrator can rely on the fact that a umount will guarantee a clean slate making it possible to reinitialize a filesystem. Right now all binary types would need to be explicitly deleted before that can happen. This allows us to remove the heavy-handed calls to simple_pin_fs() and simple_release_fs() when creating and deleting binary types. This in turn allows us to replace the current brittle pinning mechanism abusing dget() which has caused a range of bugs judging from prior fixes in [2] and [3]. The additional dget() in load_misc_binary() pins the dentry but only does so for the sake to prevent ->evict_inode() from freeing the node when a user removes the binary type and kill_node() is run. Which would mean ->interpreter and ->interp_file would be freed causing a UAF. This isn't really nicely documented nor is it very clean because it relies on simple_pin_fs() pinning the filesystem as long as at least one binary type exists. Otherwise it would cause load_misc_binary() to hold on to a dentry belonging to a superblock that has been shutdown. Replace that implicit pinning with a clean and simple per-node refcount and get rid of the ugly dget() pinning. A similar mechanism exists for e.g. binderfs (cf. [4]). All the cleanup work can now be done in ->evict_inode(). In a follow-up patch we will make it possible to use binfmt_misc in sandboxes. We will use the cleaner semantics where a umount for the filesystem will cause the superblock and all resources to be deallocated. In preparation for this apply the same semantics to the initial binfmt_misc mount. Note, that this is a user-visible change and as such a uapi change but one that we can reasonably risk. We've discussed this in earlier versions of this patchset (cf. [1]). The main user and provider of binfmt_misc is systemd. Systemd provides binfmt_misc via autofs since it is configurable as a kernel module and is used by a few exotic packages and users. As such a binfmt_misc mount is triggered when /proc/sys/fs/binfmt_misc is accessed and is only provided on demand. Other autofs on demand filesystems include EFI ESP which systemd umounts if the mountpoint stays idle for a certain amount of time. This doesn't apply to the binfmt_misc autofs mount which isn't touched once it is mounted meaning this change can't accidently wipe binary type handlers without someone having explicitly unmounted binfmt_misc. After speaking to systemd folks they don't expect this change to affect them. In line with our general policy, if we see a regression for systemd or other users with this change we will switch back to the old behavior for the initial binfmt_misc mount and have binary types pin the filesystem again. But while we touch this code let's take the chance and let's improve on the status quo. [1]: https://lore.kernel.org/r/20191216091220.465626-2-laurent@vivier.eu [2]: commit 43a4f2619038 ("exec: binfmt_misc: fix race between load_misc_binary() and kill_node()" [3]: commit 83f918274e4b ("exec: binfmt_misc: shift filp_close(interp_file) from kill_node() to bm_evict_inode()") [4]: commit f0fe2c0f050d ("binder: prevent UAF for binderfs devices II") Link: https://lore.kernel.org/r/20211028103114.2849140-1-brauner@kernel.org (v1) Cc: Sargun Dhillon <sargun@sargun.me> Cc: Serge Hallyn <serge@hallyn.com> Cc: Jann Horn <jannh@google.com> Cc: Henning Schild <henning.schild@siemens.com> Cc: Andrei Vagin <avagin@gmail.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Laurent Vivier <laurent@vivier.eu> Cc: linux-fsdevel@vger.kernel.org Acked-by: Serge Hallyn <serge@hallyn.com> Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com> Signed-off-by: Christian Brauner <brauner@kernel.org> Signed-off-by: Kees Cook <keescook@chromium.org> Signed-off-by: Sasha Levin <sashal@kernel.org> (cherry picked from commit 263bcebf5c2ab1fe949517225157f34015124620) Signed-off-by: Vegard Nossum <vegard.nossum@oracle.com>
2021-10-28 12:31:13 +02:00
struct inode *inode = file_inode(file);
Node *e = inode->i_private;
int res = parse_command(buffer, count);
switch (res) {
case 1:
/* Disable this handler. */
clear_bit(Enabled, &e->flags);
break;
case 2:
/* Enable this handler. */
set_bit(Enabled, &e->flags);
break;
case 3:
/* Delete this handler. */
binfmt_misc: cleanup on filesystem umount [ Upstream commit 1c5976ef0f7ad76319df748ccb99a4c7ba2ba464 ] Currently, registering a new binary type pins the binfmt_misc filesystem. Specifically, this means that as long as there is at least one binary type registered the binfmt_misc filesystem survives all umounts, i.e. the superblock is not destroyed. Meaning that a umount followed by another mount will end up with the same superblock and the same binary type handlers. This is a behavior we tend to discourage for any new filesystems (apart from a few special filesystems such as e.g. configfs or debugfs). A umount operation without the filesystem being pinned - by e.g. someone holding a file descriptor to an open file - should usually result in the destruction of the superblock and all associated resources. This makes introspection easier and leads to clearly defined, simple and clean semantics. An administrator can rely on the fact that a umount will guarantee a clean slate making it possible to reinitialize a filesystem. Right now all binary types would need to be explicitly deleted before that can happen. This allows us to remove the heavy-handed calls to simple_pin_fs() and simple_release_fs() when creating and deleting binary types. This in turn allows us to replace the current brittle pinning mechanism abusing dget() which has caused a range of bugs judging from prior fixes in [2] and [3]. The additional dget() in load_misc_binary() pins the dentry but only does so for the sake to prevent ->evict_inode() from freeing the node when a user removes the binary type and kill_node() is run. Which would mean ->interpreter and ->interp_file would be freed causing a UAF. This isn't really nicely documented nor is it very clean because it relies on simple_pin_fs() pinning the filesystem as long as at least one binary type exists. Otherwise it would cause load_misc_binary() to hold on to a dentry belonging to a superblock that has been shutdown. Replace that implicit pinning with a clean and simple per-node refcount and get rid of the ugly dget() pinning. A similar mechanism exists for e.g. binderfs (cf. [4]). All the cleanup work can now be done in ->evict_inode(). In a follow-up patch we will make it possible to use binfmt_misc in sandboxes. We will use the cleaner semantics where a umount for the filesystem will cause the superblock and all resources to be deallocated. In preparation for this apply the same semantics to the initial binfmt_misc mount. Note, that this is a user-visible change and as such a uapi change but one that we can reasonably risk. We've discussed this in earlier versions of this patchset (cf. [1]). The main user and provider of binfmt_misc is systemd. Systemd provides binfmt_misc via autofs since it is configurable as a kernel module and is used by a few exotic packages and users. As such a binfmt_misc mount is triggered when /proc/sys/fs/binfmt_misc is accessed and is only provided on demand. Other autofs on demand filesystems include EFI ESP which systemd umounts if the mountpoint stays idle for a certain amount of time. This doesn't apply to the binfmt_misc autofs mount which isn't touched once it is mounted meaning this change can't accidently wipe binary type handlers without someone having explicitly unmounted binfmt_misc. After speaking to systemd folks they don't expect this change to affect them. In line with our general policy, if we see a regression for systemd or other users with this change we will switch back to the old behavior for the initial binfmt_misc mount and have binary types pin the filesystem again. But while we touch this code let's take the chance and let's improve on the status quo. [1]: https://lore.kernel.org/r/20191216091220.465626-2-laurent@vivier.eu [2]: commit 43a4f2619038 ("exec: binfmt_misc: fix race between load_misc_binary() and kill_node()" [3]: commit 83f918274e4b ("exec: binfmt_misc: shift filp_close(interp_file) from kill_node() to bm_evict_inode()") [4]: commit f0fe2c0f050d ("binder: prevent UAF for binderfs devices II") Link: https://lore.kernel.org/r/20211028103114.2849140-1-brauner@kernel.org (v1) Cc: Sargun Dhillon <sargun@sargun.me> Cc: Serge Hallyn <serge@hallyn.com> Cc: Jann Horn <jannh@google.com> Cc: Henning Schild <henning.schild@siemens.com> Cc: Andrei Vagin <avagin@gmail.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Laurent Vivier <laurent@vivier.eu> Cc: linux-fsdevel@vger.kernel.org Acked-by: Serge Hallyn <serge@hallyn.com> Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com> Signed-off-by: Christian Brauner <brauner@kernel.org> Signed-off-by: Kees Cook <keescook@chromium.org> Signed-off-by: Sasha Levin <sashal@kernel.org> (cherry picked from commit 263bcebf5c2ab1fe949517225157f34015124620) Signed-off-by: Vegard Nossum <vegard.nossum@oracle.com>
2021-10-28 12:31:13 +02:00
inode = d_inode(inode->i_sb->s_root);
inode_lock(inode);
binfmt_misc: cleanup on filesystem umount [ Upstream commit 1c5976ef0f7ad76319df748ccb99a4c7ba2ba464 ] Currently, registering a new binary type pins the binfmt_misc filesystem. Specifically, this means that as long as there is at least one binary type registered the binfmt_misc filesystem survives all umounts, i.e. the superblock is not destroyed. Meaning that a umount followed by another mount will end up with the same superblock and the same binary type handlers. This is a behavior we tend to discourage for any new filesystems (apart from a few special filesystems such as e.g. configfs or debugfs). A umount operation without the filesystem being pinned - by e.g. someone holding a file descriptor to an open file - should usually result in the destruction of the superblock and all associated resources. This makes introspection easier and leads to clearly defined, simple and clean semantics. An administrator can rely on the fact that a umount will guarantee a clean slate making it possible to reinitialize a filesystem. Right now all binary types would need to be explicitly deleted before that can happen. This allows us to remove the heavy-handed calls to simple_pin_fs() and simple_release_fs() when creating and deleting binary types. This in turn allows us to replace the current brittle pinning mechanism abusing dget() which has caused a range of bugs judging from prior fixes in [2] and [3]. The additional dget() in load_misc_binary() pins the dentry but only does so for the sake to prevent ->evict_inode() from freeing the node when a user removes the binary type and kill_node() is run. Which would mean ->interpreter and ->interp_file would be freed causing a UAF. This isn't really nicely documented nor is it very clean because it relies on simple_pin_fs() pinning the filesystem as long as at least one binary type exists. Otherwise it would cause load_misc_binary() to hold on to a dentry belonging to a superblock that has been shutdown. Replace that implicit pinning with a clean and simple per-node refcount and get rid of the ugly dget() pinning. A similar mechanism exists for e.g. binderfs (cf. [4]). All the cleanup work can now be done in ->evict_inode(). In a follow-up patch we will make it possible to use binfmt_misc in sandboxes. We will use the cleaner semantics where a umount for the filesystem will cause the superblock and all resources to be deallocated. In preparation for this apply the same semantics to the initial binfmt_misc mount. Note, that this is a user-visible change and as such a uapi change but one that we can reasonably risk. We've discussed this in earlier versions of this patchset (cf. [1]). The main user and provider of binfmt_misc is systemd. Systemd provides binfmt_misc via autofs since it is configurable as a kernel module and is used by a few exotic packages and users. As such a binfmt_misc mount is triggered when /proc/sys/fs/binfmt_misc is accessed and is only provided on demand. Other autofs on demand filesystems include EFI ESP which systemd umounts if the mountpoint stays idle for a certain amount of time. This doesn't apply to the binfmt_misc autofs mount which isn't touched once it is mounted meaning this change can't accidently wipe binary type handlers without someone having explicitly unmounted binfmt_misc. After speaking to systemd folks they don't expect this change to affect them. In line with our general policy, if we see a regression for systemd or other users with this change we will switch back to the old behavior for the initial binfmt_misc mount and have binary types pin the filesystem again. But while we touch this code let's take the chance and let's improve on the status quo. [1]: https://lore.kernel.org/r/20191216091220.465626-2-laurent@vivier.eu [2]: commit 43a4f2619038 ("exec: binfmt_misc: fix race between load_misc_binary() and kill_node()" [3]: commit 83f918274e4b ("exec: binfmt_misc: shift filp_close(interp_file) from kill_node() to bm_evict_inode()") [4]: commit f0fe2c0f050d ("binder: prevent UAF for binderfs devices II") Link: https://lore.kernel.org/r/20211028103114.2849140-1-brauner@kernel.org (v1) Cc: Sargun Dhillon <sargun@sargun.me> Cc: Serge Hallyn <serge@hallyn.com> Cc: Jann Horn <jannh@google.com> Cc: Henning Schild <henning.schild@siemens.com> Cc: Andrei Vagin <avagin@gmail.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Laurent Vivier <laurent@vivier.eu> Cc: linux-fsdevel@vger.kernel.org Acked-by: Serge Hallyn <serge@hallyn.com> Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com> Signed-off-by: Christian Brauner <brauner@kernel.org> Signed-off-by: Kees Cook <keescook@chromium.org> Signed-off-by: Sasha Levin <sashal@kernel.org> (cherry picked from commit 263bcebf5c2ab1fe949517225157f34015124620) Signed-off-by: Vegard Nossum <vegard.nossum@oracle.com>
2021-10-28 12:31:13 +02:00
/*
* In order to add new element or remove elements from the list
* via bm_{entry,register,status}_write() inode_lock() on the
* root inode must be held.
* The lock is exclusive ensuring that the list can't be
* modified. Only load_misc_binary() can access but does so
* read-only. So we only need to take the write lock when we
* actually remove the entry from the list.
*/
if (!list_empty(&e->list))
binfmt_misc: cleanup on filesystem umount [ Upstream commit 1c5976ef0f7ad76319df748ccb99a4c7ba2ba464 ] Currently, registering a new binary type pins the binfmt_misc filesystem. Specifically, this means that as long as there is at least one binary type registered the binfmt_misc filesystem survives all umounts, i.e. the superblock is not destroyed. Meaning that a umount followed by another mount will end up with the same superblock and the same binary type handlers. This is a behavior we tend to discourage for any new filesystems (apart from a few special filesystems such as e.g. configfs or debugfs). A umount operation without the filesystem being pinned - by e.g. someone holding a file descriptor to an open file - should usually result in the destruction of the superblock and all associated resources. This makes introspection easier and leads to clearly defined, simple and clean semantics. An administrator can rely on the fact that a umount will guarantee a clean slate making it possible to reinitialize a filesystem. Right now all binary types would need to be explicitly deleted before that can happen. This allows us to remove the heavy-handed calls to simple_pin_fs() and simple_release_fs() when creating and deleting binary types. This in turn allows us to replace the current brittle pinning mechanism abusing dget() which has caused a range of bugs judging from prior fixes in [2] and [3]. The additional dget() in load_misc_binary() pins the dentry but only does so for the sake to prevent ->evict_inode() from freeing the node when a user removes the binary type and kill_node() is run. Which would mean ->interpreter and ->interp_file would be freed causing a UAF. This isn't really nicely documented nor is it very clean because it relies on simple_pin_fs() pinning the filesystem as long as at least one binary type exists. Otherwise it would cause load_misc_binary() to hold on to a dentry belonging to a superblock that has been shutdown. Replace that implicit pinning with a clean and simple per-node refcount and get rid of the ugly dget() pinning. A similar mechanism exists for e.g. binderfs (cf. [4]). All the cleanup work can now be done in ->evict_inode(). In a follow-up patch we will make it possible to use binfmt_misc in sandboxes. We will use the cleaner semantics where a umount for the filesystem will cause the superblock and all resources to be deallocated. In preparation for this apply the same semantics to the initial binfmt_misc mount. Note, that this is a user-visible change and as such a uapi change but one that we can reasonably risk. We've discussed this in earlier versions of this patchset (cf. [1]). The main user and provider of binfmt_misc is systemd. Systemd provides binfmt_misc via autofs since it is configurable as a kernel module and is used by a few exotic packages and users. As such a binfmt_misc mount is triggered when /proc/sys/fs/binfmt_misc is accessed and is only provided on demand. Other autofs on demand filesystems include EFI ESP which systemd umounts if the mountpoint stays idle for a certain amount of time. This doesn't apply to the binfmt_misc autofs mount which isn't touched once it is mounted meaning this change can't accidently wipe binary type handlers without someone having explicitly unmounted binfmt_misc. After speaking to systemd folks they don't expect this change to affect them. In line with our general policy, if we see a regression for systemd or other users with this change we will switch back to the old behavior for the initial binfmt_misc mount and have binary types pin the filesystem again. But while we touch this code let's take the chance and let's improve on the status quo. [1]: https://lore.kernel.org/r/20191216091220.465626-2-laurent@vivier.eu [2]: commit 43a4f2619038 ("exec: binfmt_misc: fix race between load_misc_binary() and kill_node()" [3]: commit 83f918274e4b ("exec: binfmt_misc: shift filp_close(interp_file) from kill_node() to bm_evict_inode()") [4]: commit f0fe2c0f050d ("binder: prevent UAF for binderfs devices II") Link: https://lore.kernel.org/r/20211028103114.2849140-1-brauner@kernel.org (v1) Cc: Sargun Dhillon <sargun@sargun.me> Cc: Serge Hallyn <serge@hallyn.com> Cc: Jann Horn <jannh@google.com> Cc: Henning Schild <henning.schild@siemens.com> Cc: Andrei Vagin <avagin@gmail.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Laurent Vivier <laurent@vivier.eu> Cc: linux-fsdevel@vger.kernel.org Acked-by: Serge Hallyn <serge@hallyn.com> Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com> Signed-off-by: Christian Brauner <brauner@kernel.org> Signed-off-by: Kees Cook <keescook@chromium.org> Signed-off-by: Sasha Levin <sashal@kernel.org> (cherry picked from commit 263bcebf5c2ab1fe949517225157f34015124620) Signed-off-by: Vegard Nossum <vegard.nossum@oracle.com>
2021-10-28 12:31:13 +02:00
remove_binfmt_handler(e);
binfmt_misc: cleanup on filesystem umount [ Upstream commit 1c5976ef0f7ad76319df748ccb99a4c7ba2ba464 ] Currently, registering a new binary type pins the binfmt_misc filesystem. Specifically, this means that as long as there is at least one binary type registered the binfmt_misc filesystem survives all umounts, i.e. the superblock is not destroyed. Meaning that a umount followed by another mount will end up with the same superblock and the same binary type handlers. This is a behavior we tend to discourage for any new filesystems (apart from a few special filesystems such as e.g. configfs or debugfs). A umount operation without the filesystem being pinned - by e.g. someone holding a file descriptor to an open file - should usually result in the destruction of the superblock and all associated resources. This makes introspection easier and leads to clearly defined, simple and clean semantics. An administrator can rely on the fact that a umount will guarantee a clean slate making it possible to reinitialize a filesystem. Right now all binary types would need to be explicitly deleted before that can happen. This allows us to remove the heavy-handed calls to simple_pin_fs() and simple_release_fs() when creating and deleting binary types. This in turn allows us to replace the current brittle pinning mechanism abusing dget() which has caused a range of bugs judging from prior fixes in [2] and [3]. The additional dget() in load_misc_binary() pins the dentry but only does so for the sake to prevent ->evict_inode() from freeing the node when a user removes the binary type and kill_node() is run. Which would mean ->interpreter and ->interp_file would be freed causing a UAF. This isn't really nicely documented nor is it very clean because it relies on simple_pin_fs() pinning the filesystem as long as at least one binary type exists. Otherwise it would cause load_misc_binary() to hold on to a dentry belonging to a superblock that has been shutdown. Replace that implicit pinning with a clean and simple per-node refcount and get rid of the ugly dget() pinning. A similar mechanism exists for e.g. binderfs (cf. [4]). All the cleanup work can now be done in ->evict_inode(). In a follow-up patch we will make it possible to use binfmt_misc in sandboxes. We will use the cleaner semantics where a umount for the filesystem will cause the superblock and all resources to be deallocated. In preparation for this apply the same semantics to the initial binfmt_misc mount. Note, that this is a user-visible change and as such a uapi change but one that we can reasonably risk. We've discussed this in earlier versions of this patchset (cf. [1]). The main user and provider of binfmt_misc is systemd. Systemd provides binfmt_misc via autofs since it is configurable as a kernel module and is used by a few exotic packages and users. As such a binfmt_misc mount is triggered when /proc/sys/fs/binfmt_misc is accessed and is only provided on demand. Other autofs on demand filesystems include EFI ESP which systemd umounts if the mountpoint stays idle for a certain amount of time. This doesn't apply to the binfmt_misc autofs mount which isn't touched once it is mounted meaning this change can't accidently wipe binary type handlers without someone having explicitly unmounted binfmt_misc. After speaking to systemd folks they don't expect this change to affect them. In line with our general policy, if we see a regression for systemd or other users with this change we will switch back to the old behavior for the initial binfmt_misc mount and have binary types pin the filesystem again. But while we touch this code let's take the chance and let's improve on the status quo. [1]: https://lore.kernel.org/r/20191216091220.465626-2-laurent@vivier.eu [2]: commit 43a4f2619038 ("exec: binfmt_misc: fix race between load_misc_binary() and kill_node()" [3]: commit 83f918274e4b ("exec: binfmt_misc: shift filp_close(interp_file) from kill_node() to bm_evict_inode()") [4]: commit f0fe2c0f050d ("binder: prevent UAF for binderfs devices II") Link: https://lore.kernel.org/r/20211028103114.2849140-1-brauner@kernel.org (v1) Cc: Sargun Dhillon <sargun@sargun.me> Cc: Serge Hallyn <serge@hallyn.com> Cc: Jann Horn <jannh@google.com> Cc: Henning Schild <henning.schild@siemens.com> Cc: Andrei Vagin <avagin@gmail.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Laurent Vivier <laurent@vivier.eu> Cc: linux-fsdevel@vger.kernel.org Acked-by: Serge Hallyn <serge@hallyn.com> Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com> Signed-off-by: Christian Brauner <brauner@kernel.org> Signed-off-by: Kees Cook <keescook@chromium.org> Signed-off-by: Sasha Levin <sashal@kernel.org> (cherry picked from commit 263bcebf5c2ab1fe949517225157f34015124620) Signed-off-by: Vegard Nossum <vegard.nossum@oracle.com>
2021-10-28 12:31:13 +02:00
inode_unlock(inode);
break;
default:
return res;
}
return count;
}
static const struct file_operations bm_entry_operations = {
.read = bm_entry_read,
.write = bm_entry_write,
llseek: automatically add .llseek fop All file_operations should get a .llseek operation so we can make nonseekable_open the default for future file operations without a .llseek pointer. The three cases that we can automatically detect are no_llseek, seq_lseek and default_llseek. For cases where we can we can automatically prove that the file offset is always ignored, we use noop_llseek, which maintains the current behavior of not returning an error from a seek. New drivers should normally not use noop_llseek but instead use no_llseek and call nonseekable_open at open time. Existing drivers can be converted to do the same when the maintainer knows for certain that no user code relies on calling seek on the device file. The generated code is often incorrectly indented and right now contains comments that clarify for each added line why a specific variant was chosen. In the version that gets submitted upstream, the comments will be gone and I will manually fix the indentation, because there does not seem to be a way to do that using coccinelle. Some amount of new code is currently sitting in linux-next that should get the same modifications, which I will do at the end of the merge window. Many thanks to Julia Lawall for helping me learn to write a semantic patch that does all this. ===== begin semantic patch ===== // This adds an llseek= method to all file operations, // as a preparation for making no_llseek the default. // // The rules are // - use no_llseek explicitly if we do nonseekable_open // - use seq_lseek for sequential files // - use default_llseek if we know we access f_pos // - use noop_llseek if we know we don't access f_pos, // but we still want to allow users to call lseek // @ open1 exists @ identifier nested_open; @@ nested_open(...) { <+... nonseekable_open(...) ...+> } @ open exists@ identifier open_f; identifier i, f; identifier open1.nested_open; @@ int open_f(struct inode *i, struct file *f) { <+... ( nonseekable_open(...) | nested_open(...) ) ...+> } @ read disable optional_qualifier exists @ identifier read_f; identifier f, p, s, off; type ssize_t, size_t, loff_t; expression E; identifier func; @@ ssize_t read_f(struct file *f, char *p, size_t s, loff_t *off) { <+... ( *off = E | *off += E | func(..., off, ...) | E = *off ) ...+> } @ read_no_fpos disable optional_qualifier exists @ identifier read_f; identifier f, p, s, off; type ssize_t, size_t, loff_t; @@ ssize_t read_f(struct file *f, char *p, size_t s, loff_t *off) { ... when != off } @ write @ identifier write_f; identifier f, p, s, off; type ssize_t, size_t, loff_t; expression E; identifier func; @@ ssize_t write_f(struct file *f, const char *p, size_t s, loff_t *off) { <+... ( *off = E | *off += E | func(..., off, ...) | E = *off ) ...+> } @ write_no_fpos @ identifier write_f; identifier f, p, s, off; type ssize_t, size_t, loff_t; @@ ssize_t write_f(struct file *f, const char *p, size_t s, loff_t *off) { ... when != off } @ fops0 @ identifier fops; @@ struct file_operations fops = { ... }; @ has_llseek depends on fops0 @ identifier fops0.fops; identifier llseek_f; @@ struct file_operations fops = { ... .llseek = llseek_f, ... }; @ has_read depends on fops0 @ identifier fops0.fops; identifier read_f; @@ struct file_operations fops = { ... .read = read_f, ... }; @ has_write depends on fops0 @ identifier fops0.fops; identifier write_f; @@ struct file_operations fops = { ... .write = write_f, ... }; @ has_open depends on fops0 @ identifier fops0.fops; identifier open_f; @@ struct file_operations fops = { ... .open = open_f, ... }; // use no_llseek if we call nonseekable_open //////////////////////////////////////////// @ nonseekable1 depends on !has_llseek && has_open @ identifier fops0.fops; identifier nso ~= "nonseekable_open"; @@ struct file_operations fops = { ... .open = nso, ... +.llseek = no_llseek, /* nonseekable */ }; @ nonseekable2 depends on !has_llseek @ identifier fops0.fops; identifier open.open_f; @@ struct file_operations fops = { ... .open = open_f, ... +.llseek = no_llseek, /* open uses nonseekable */ }; // use seq_lseek for sequential files ///////////////////////////////////// @ seq depends on !has_llseek @ identifier fops0.fops; identifier sr ~= "seq_read"; @@ struct file_operations fops = { ... .read = sr, ... +.llseek = seq_lseek, /* we have seq_read */ }; // use default_llseek if there is a readdir /////////////////////////////////////////// @ fops1 depends on !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier readdir_e; @@ // any other fop is used that changes pos struct file_operations fops = { ... .readdir = readdir_e, ... +.llseek = default_llseek, /* readdir is present */ }; // use default_llseek if at least one of read/write touches f_pos ///////////////////////////////////////////////////////////////// @ fops2 depends on !fops1 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier read.read_f; @@ // read fops use offset struct file_operations fops = { ... .read = read_f, ... +.llseek = default_llseek, /* read accesses f_pos */ }; @ fops3 depends on !fops1 && !fops2 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier write.write_f; @@ // write fops use offset struct file_operations fops = { ... .write = write_f, ... + .llseek = default_llseek, /* write accesses f_pos */ }; // Use noop_llseek if neither read nor write accesses f_pos /////////////////////////////////////////////////////////// @ fops4 depends on !fops1 && !fops2 && !fops3 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier read_no_fpos.read_f; identifier write_no_fpos.write_f; @@ // write fops use offset struct file_operations fops = { ... .write = write_f, .read = read_f, ... +.llseek = noop_llseek, /* read and write both use no f_pos */ }; @ depends on has_write && !has_read && !fops1 && !fops2 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier write_no_fpos.write_f; @@ struct file_operations fops = { ... .write = write_f, ... +.llseek = noop_llseek, /* write uses no f_pos */ }; @ depends on has_read && !has_write && !fops1 && !fops2 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier read_no_fpos.read_f; @@ struct file_operations fops = { ... .read = read_f, ... +.llseek = noop_llseek, /* read uses no f_pos */ }; @ depends on !has_read && !has_write && !fops1 && !fops2 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; @@ struct file_operations fops = { ... +.llseek = noop_llseek, /* no read or write fn */ }; ===== End semantic patch ===== Signed-off-by: Arnd Bergmann <arnd@arndb.de> Cc: Julia Lawall <julia@diku.dk> Cc: Christoph Hellwig <hch@infradead.org>
2010-08-15 18:52:59 +02:00
.llseek = default_llseek,
};
/* /register */
static ssize_t bm_register_write(struct file *file, const char __user *buffer,
size_t count, loff_t *ppos)
{
Node *e;
struct inode *inode;
struct super_block *sb = file_inode(file)->i_sb;
struct dentry *root = sb->s_root, *dentry;
int err = 0;
binfmt_misc: fix possible deadlock in bm_register_write commit e7850f4d844e0acfac7e570af611d89deade3146 upstream. There is a deadlock in bm_register_write: First, in the begining of the function, a lock is taken on the binfmt_misc root inode with inode_lock(d_inode(root)). Then, if the user used the MISC_FMT_OPEN_FILE flag, the function will call open_exec on the user-provided interpreter. open_exec will call a path lookup, and if the path lookup process includes the root of binfmt_misc, it will try to take a shared lock on its inode again, but it is already locked, and the code will get stuck in a deadlock To reproduce the bug: $ echo ":iiiii:E::ii::/proc/sys/fs/binfmt_misc/bla:F" > /proc/sys/fs/binfmt_misc/register backtrace of where the lock occurs (#5): 0 schedule () at ./arch/x86/include/asm/current.h:15 1 0xffffffff81b51237 in rwsem_down_read_slowpath (sem=0xffff888003b202e0, count=<optimized out>, state=state@entry=2) at kernel/locking/rwsem.c:992 2 0xffffffff81b5150a in __down_read_common (state=2, sem=<optimized out>) at kernel/locking/rwsem.c:1213 3 __down_read (sem=<optimized out>) at kernel/locking/rwsem.c:1222 4 down_read (sem=<optimized out>) at kernel/locking/rwsem.c:1355 5 0xffffffff811ee22a in inode_lock_shared (inode=<optimized out>) at ./include/linux/fs.h:783 6 open_last_lookups (op=0xffffc9000022fe34, file=0xffff888004098600, nd=0xffffc9000022fd10) at fs/namei.c:3177 7 path_openat (nd=nd@entry=0xffffc9000022fd10, op=op@entry=0xffffc9000022fe34, flags=flags@entry=65) at fs/namei.c:3366 8 0xffffffff811efe1c in do_filp_open (dfd=<optimized out>, pathname=pathname@entry=0xffff8880031b9000, op=op@entry=0xffffc9000022fe34) at fs/namei.c:3396 9 0xffffffff811e493f in do_open_execat (fd=fd@entry=-100, name=name@entry=0xffff8880031b9000, flags=<optimized out>, flags@entry=0) at fs/exec.c:913 10 0xffffffff811e4a92 in open_exec (name=<optimized out>) at fs/exec.c:948 11 0xffffffff8124aa84 in bm_register_write (file=<optimized out>, buffer=<optimized out>, count=19, ppos=<optimized out>) at fs/binfmt_misc.c:682 12 0xffffffff811decd2 in vfs_write (file=file@entry=0xffff888004098500, buf=buf@entry=0xa758d0 ":iiiii:E::ii::i:CF ", count=count@entry=19, pos=pos@entry=0xffffc9000022ff10) at fs/read_write.c:603 13 0xffffffff811defda in ksys_write (fd=<optimized out>, buf=0xa758d0 ":iiiii:E::ii::i:CF ", count=19) at fs/read_write.c:658 14 0xffffffff81b49813 in do_syscall_64 (nr=<optimized out>, regs=0xffffc9000022ff58) at arch/x86/entry/common.c:46 15 0xffffffff81c0007c in entry_SYSCALL_64 () at arch/x86/entry/entry_64.S:120 To solve the issue, the open_exec call is moved to before the write lock is taken by bm_register_write Link: https://lkml.kernel.org/r/20210228224414.95962-1-liorribak@gmail.com Fixes: 948b701a607f1 ("binfmt_misc: add persistent opened binary handler for containers") Signed-off-by: Lior Ribak <liorribak@gmail.com> Acked-by: Helge Deller <deller@gmx.de> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2021-03-12 21:07:41 -08:00
struct file *f = NULL;
e = create_entry(buffer, count);
if (IS_ERR(e))
return PTR_ERR(e);
binfmt_misc: fix possible deadlock in bm_register_write commit e7850f4d844e0acfac7e570af611d89deade3146 upstream. There is a deadlock in bm_register_write: First, in the begining of the function, a lock is taken on the binfmt_misc root inode with inode_lock(d_inode(root)). Then, if the user used the MISC_FMT_OPEN_FILE flag, the function will call open_exec on the user-provided interpreter. open_exec will call a path lookup, and if the path lookup process includes the root of binfmt_misc, it will try to take a shared lock on its inode again, but it is already locked, and the code will get stuck in a deadlock To reproduce the bug: $ echo ":iiiii:E::ii::/proc/sys/fs/binfmt_misc/bla:F" > /proc/sys/fs/binfmt_misc/register backtrace of where the lock occurs (#5): 0 schedule () at ./arch/x86/include/asm/current.h:15 1 0xffffffff81b51237 in rwsem_down_read_slowpath (sem=0xffff888003b202e0, count=<optimized out>, state=state@entry=2) at kernel/locking/rwsem.c:992 2 0xffffffff81b5150a in __down_read_common (state=2, sem=<optimized out>) at kernel/locking/rwsem.c:1213 3 __down_read (sem=<optimized out>) at kernel/locking/rwsem.c:1222 4 down_read (sem=<optimized out>) at kernel/locking/rwsem.c:1355 5 0xffffffff811ee22a in inode_lock_shared (inode=<optimized out>) at ./include/linux/fs.h:783 6 open_last_lookups (op=0xffffc9000022fe34, file=0xffff888004098600, nd=0xffffc9000022fd10) at fs/namei.c:3177 7 path_openat (nd=nd@entry=0xffffc9000022fd10, op=op@entry=0xffffc9000022fe34, flags=flags@entry=65) at fs/namei.c:3366 8 0xffffffff811efe1c in do_filp_open (dfd=<optimized out>, pathname=pathname@entry=0xffff8880031b9000, op=op@entry=0xffffc9000022fe34) at fs/namei.c:3396 9 0xffffffff811e493f in do_open_execat (fd=fd@entry=-100, name=name@entry=0xffff8880031b9000, flags=<optimized out>, flags@entry=0) at fs/exec.c:913 10 0xffffffff811e4a92 in open_exec (name=<optimized out>) at fs/exec.c:948 11 0xffffffff8124aa84 in bm_register_write (file=<optimized out>, buffer=<optimized out>, count=19, ppos=<optimized out>) at fs/binfmt_misc.c:682 12 0xffffffff811decd2 in vfs_write (file=file@entry=0xffff888004098500, buf=buf@entry=0xa758d0 ":iiiii:E::ii::i:CF ", count=count@entry=19, pos=pos@entry=0xffffc9000022ff10) at fs/read_write.c:603 13 0xffffffff811defda in ksys_write (fd=<optimized out>, buf=0xa758d0 ":iiiii:E::ii::i:CF ", count=19) at fs/read_write.c:658 14 0xffffffff81b49813 in do_syscall_64 (nr=<optimized out>, regs=0xffffc9000022ff58) at arch/x86/entry/common.c:46 15 0xffffffff81c0007c in entry_SYSCALL_64 () at arch/x86/entry/entry_64.S:120 To solve the issue, the open_exec call is moved to before the write lock is taken by bm_register_write Link: https://lkml.kernel.org/r/20210228224414.95962-1-liorribak@gmail.com Fixes: 948b701a607f1 ("binfmt_misc: add persistent opened binary handler for containers") Signed-off-by: Lior Ribak <liorribak@gmail.com> Acked-by: Helge Deller <deller@gmx.de> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2021-03-12 21:07:41 -08:00
if (e->flags & MISC_FMT_OPEN_FILE) {
f = open_exec(e->interpreter);
if (IS_ERR(f)) {
pr_notice("register: failed to install interpreter file %s\n",
e->interpreter);
kfree(e);
return PTR_ERR(f);
}
e->interp_file = f;
}
inode_lock(d_inode(root));
dentry = lookup_one_len(e->name, root, strlen(e->name));
err = PTR_ERR(dentry);
if (IS_ERR(dentry))
goto out;
err = -EEXIST;
if (d_really_is_positive(dentry))
goto out2;
inode = bm_get_inode(sb, S_IFREG | 0644);
err = -ENOMEM;
if (!inode)
goto out2;
binfmt_misc: cleanup on filesystem umount [ Upstream commit 1c5976ef0f7ad76319df748ccb99a4c7ba2ba464 ] Currently, registering a new binary type pins the binfmt_misc filesystem. Specifically, this means that as long as there is at least one binary type registered the binfmt_misc filesystem survives all umounts, i.e. the superblock is not destroyed. Meaning that a umount followed by another mount will end up with the same superblock and the same binary type handlers. This is a behavior we tend to discourage for any new filesystems (apart from a few special filesystems such as e.g. configfs or debugfs). A umount operation without the filesystem being pinned - by e.g. someone holding a file descriptor to an open file - should usually result in the destruction of the superblock and all associated resources. This makes introspection easier and leads to clearly defined, simple and clean semantics. An administrator can rely on the fact that a umount will guarantee a clean slate making it possible to reinitialize a filesystem. Right now all binary types would need to be explicitly deleted before that can happen. This allows us to remove the heavy-handed calls to simple_pin_fs() and simple_release_fs() when creating and deleting binary types. This in turn allows us to replace the current brittle pinning mechanism abusing dget() which has caused a range of bugs judging from prior fixes in [2] and [3]. The additional dget() in load_misc_binary() pins the dentry but only does so for the sake to prevent ->evict_inode() from freeing the node when a user removes the binary type and kill_node() is run. Which would mean ->interpreter and ->interp_file would be freed causing a UAF. This isn't really nicely documented nor is it very clean because it relies on simple_pin_fs() pinning the filesystem as long as at least one binary type exists. Otherwise it would cause load_misc_binary() to hold on to a dentry belonging to a superblock that has been shutdown. Replace that implicit pinning with a clean and simple per-node refcount and get rid of the ugly dget() pinning. A similar mechanism exists for e.g. binderfs (cf. [4]). All the cleanup work can now be done in ->evict_inode(). In a follow-up patch we will make it possible to use binfmt_misc in sandboxes. We will use the cleaner semantics where a umount for the filesystem will cause the superblock and all resources to be deallocated. In preparation for this apply the same semantics to the initial binfmt_misc mount. Note, that this is a user-visible change and as such a uapi change but one that we can reasonably risk. We've discussed this in earlier versions of this patchset (cf. [1]). The main user and provider of binfmt_misc is systemd. Systemd provides binfmt_misc via autofs since it is configurable as a kernel module and is used by a few exotic packages and users. As such a binfmt_misc mount is triggered when /proc/sys/fs/binfmt_misc is accessed and is only provided on demand. Other autofs on demand filesystems include EFI ESP which systemd umounts if the mountpoint stays idle for a certain amount of time. This doesn't apply to the binfmt_misc autofs mount which isn't touched once it is mounted meaning this change can't accidently wipe binary type handlers without someone having explicitly unmounted binfmt_misc. After speaking to systemd folks they don't expect this change to affect them. In line with our general policy, if we see a regression for systemd or other users with this change we will switch back to the old behavior for the initial binfmt_misc mount and have binary types pin the filesystem again. But while we touch this code let's take the chance and let's improve on the status quo. [1]: https://lore.kernel.org/r/20191216091220.465626-2-laurent@vivier.eu [2]: commit 43a4f2619038 ("exec: binfmt_misc: fix race between load_misc_binary() and kill_node()" [3]: commit 83f918274e4b ("exec: binfmt_misc: shift filp_close(interp_file) from kill_node() to bm_evict_inode()") [4]: commit f0fe2c0f050d ("binder: prevent UAF for binderfs devices II") Link: https://lore.kernel.org/r/20211028103114.2849140-1-brauner@kernel.org (v1) Cc: Sargun Dhillon <sargun@sargun.me> Cc: Serge Hallyn <serge@hallyn.com> Cc: Jann Horn <jannh@google.com> Cc: Henning Schild <henning.schild@siemens.com> Cc: Andrei Vagin <avagin@gmail.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Laurent Vivier <laurent@vivier.eu> Cc: linux-fsdevel@vger.kernel.org Acked-by: Serge Hallyn <serge@hallyn.com> Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com> Signed-off-by: Christian Brauner <brauner@kernel.org> Signed-off-by: Kees Cook <keescook@chromium.org> Signed-off-by: Sasha Levin <sashal@kernel.org> (cherry picked from commit 263bcebf5c2ab1fe949517225157f34015124620) Signed-off-by: Vegard Nossum <vegard.nossum@oracle.com>
2021-10-28 12:31:13 +02:00
refcount_set(&e->users, 1);
e->dentry = dget(dentry);
inode->i_private = e;
inode->i_fop = &bm_entry_operations;
d_instantiate(dentry, inode);
write_lock(&entries_lock);
list_add(&e->list, &entries);
write_unlock(&entries_lock);
err = 0;
out2:
dput(dentry);
out:
inode_unlock(d_inode(root));
if (err) {
binfmt_misc: fix possible deadlock in bm_register_write commit e7850f4d844e0acfac7e570af611d89deade3146 upstream. There is a deadlock in bm_register_write: First, in the begining of the function, a lock is taken on the binfmt_misc root inode with inode_lock(d_inode(root)). Then, if the user used the MISC_FMT_OPEN_FILE flag, the function will call open_exec on the user-provided interpreter. open_exec will call a path lookup, and if the path lookup process includes the root of binfmt_misc, it will try to take a shared lock on its inode again, but it is already locked, and the code will get stuck in a deadlock To reproduce the bug: $ echo ":iiiii:E::ii::/proc/sys/fs/binfmt_misc/bla:F" > /proc/sys/fs/binfmt_misc/register backtrace of where the lock occurs (#5): 0 schedule () at ./arch/x86/include/asm/current.h:15 1 0xffffffff81b51237 in rwsem_down_read_slowpath (sem=0xffff888003b202e0, count=<optimized out>, state=state@entry=2) at kernel/locking/rwsem.c:992 2 0xffffffff81b5150a in __down_read_common (state=2, sem=<optimized out>) at kernel/locking/rwsem.c:1213 3 __down_read (sem=<optimized out>) at kernel/locking/rwsem.c:1222 4 down_read (sem=<optimized out>) at kernel/locking/rwsem.c:1355 5 0xffffffff811ee22a in inode_lock_shared (inode=<optimized out>) at ./include/linux/fs.h:783 6 open_last_lookups (op=0xffffc9000022fe34, file=0xffff888004098600, nd=0xffffc9000022fd10) at fs/namei.c:3177 7 path_openat (nd=nd@entry=0xffffc9000022fd10, op=op@entry=0xffffc9000022fe34, flags=flags@entry=65) at fs/namei.c:3366 8 0xffffffff811efe1c in do_filp_open (dfd=<optimized out>, pathname=pathname@entry=0xffff8880031b9000, op=op@entry=0xffffc9000022fe34) at fs/namei.c:3396 9 0xffffffff811e493f in do_open_execat (fd=fd@entry=-100, name=name@entry=0xffff8880031b9000, flags=<optimized out>, flags@entry=0) at fs/exec.c:913 10 0xffffffff811e4a92 in open_exec (name=<optimized out>) at fs/exec.c:948 11 0xffffffff8124aa84 in bm_register_write (file=<optimized out>, buffer=<optimized out>, count=19, ppos=<optimized out>) at fs/binfmt_misc.c:682 12 0xffffffff811decd2 in vfs_write (file=file@entry=0xffff888004098500, buf=buf@entry=0xa758d0 ":iiiii:E::ii::i:CF ", count=count@entry=19, pos=pos@entry=0xffffc9000022ff10) at fs/read_write.c:603 13 0xffffffff811defda in ksys_write (fd=<optimized out>, buf=0xa758d0 ":iiiii:E::ii::i:CF ", count=19) at fs/read_write.c:658 14 0xffffffff81b49813 in do_syscall_64 (nr=<optimized out>, regs=0xffffc9000022ff58) at arch/x86/entry/common.c:46 15 0xffffffff81c0007c in entry_SYSCALL_64 () at arch/x86/entry/entry_64.S:120 To solve the issue, the open_exec call is moved to before the write lock is taken by bm_register_write Link: https://lkml.kernel.org/r/20210228224414.95962-1-liorribak@gmail.com Fixes: 948b701a607f1 ("binfmt_misc: add persistent opened binary handler for containers") Signed-off-by: Lior Ribak <liorribak@gmail.com> Acked-by: Helge Deller <deller@gmx.de> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2021-03-12 21:07:41 -08:00
if (f)
filp_close(f, NULL);
kfree(e);
return err;
}
return count;
}
static const struct file_operations bm_register_operations = {
.write = bm_register_write,
llseek: automatically add .llseek fop All file_operations should get a .llseek operation so we can make nonseekable_open the default for future file operations without a .llseek pointer. The three cases that we can automatically detect are no_llseek, seq_lseek and default_llseek. For cases where we can we can automatically prove that the file offset is always ignored, we use noop_llseek, which maintains the current behavior of not returning an error from a seek. New drivers should normally not use noop_llseek but instead use no_llseek and call nonseekable_open at open time. Existing drivers can be converted to do the same when the maintainer knows for certain that no user code relies on calling seek on the device file. The generated code is often incorrectly indented and right now contains comments that clarify for each added line why a specific variant was chosen. In the version that gets submitted upstream, the comments will be gone and I will manually fix the indentation, because there does not seem to be a way to do that using coccinelle. Some amount of new code is currently sitting in linux-next that should get the same modifications, which I will do at the end of the merge window. Many thanks to Julia Lawall for helping me learn to write a semantic patch that does all this. ===== begin semantic patch ===== // This adds an llseek= method to all file operations, // as a preparation for making no_llseek the default. // // The rules are // - use no_llseek explicitly if we do nonseekable_open // - use seq_lseek for sequential files // - use default_llseek if we know we access f_pos // - use noop_llseek if we know we don't access f_pos, // but we still want to allow users to call lseek // @ open1 exists @ identifier nested_open; @@ nested_open(...) { <+... nonseekable_open(...) ...+> } @ open exists@ identifier open_f; identifier i, f; identifier open1.nested_open; @@ int open_f(struct inode *i, struct file *f) { <+... ( nonseekable_open(...) | nested_open(...) ) ...+> } @ read disable optional_qualifier exists @ identifier read_f; identifier f, p, s, off; type ssize_t, size_t, loff_t; expression E; identifier func; @@ ssize_t read_f(struct file *f, char *p, size_t s, loff_t *off) { <+... ( *off = E | *off += E | func(..., off, ...) | E = *off ) ...+> } @ read_no_fpos disable optional_qualifier exists @ identifier read_f; identifier f, p, s, off; type ssize_t, size_t, loff_t; @@ ssize_t read_f(struct file *f, char *p, size_t s, loff_t *off) { ... when != off } @ write @ identifier write_f; identifier f, p, s, off; type ssize_t, size_t, loff_t; expression E; identifier func; @@ ssize_t write_f(struct file *f, const char *p, size_t s, loff_t *off) { <+... ( *off = E | *off += E | func(..., off, ...) | E = *off ) ...+> } @ write_no_fpos @ identifier write_f; identifier f, p, s, off; type ssize_t, size_t, loff_t; @@ ssize_t write_f(struct file *f, const char *p, size_t s, loff_t *off) { ... when != off } @ fops0 @ identifier fops; @@ struct file_operations fops = { ... }; @ has_llseek depends on fops0 @ identifier fops0.fops; identifier llseek_f; @@ struct file_operations fops = { ... .llseek = llseek_f, ... }; @ has_read depends on fops0 @ identifier fops0.fops; identifier read_f; @@ struct file_operations fops = { ... .read = read_f, ... }; @ has_write depends on fops0 @ identifier fops0.fops; identifier write_f; @@ struct file_operations fops = { ... .write = write_f, ... }; @ has_open depends on fops0 @ identifier fops0.fops; identifier open_f; @@ struct file_operations fops = { ... .open = open_f, ... }; // use no_llseek if we call nonseekable_open //////////////////////////////////////////// @ nonseekable1 depends on !has_llseek && has_open @ identifier fops0.fops; identifier nso ~= "nonseekable_open"; @@ struct file_operations fops = { ... .open = nso, ... +.llseek = no_llseek, /* nonseekable */ }; @ nonseekable2 depends on !has_llseek @ identifier fops0.fops; identifier open.open_f; @@ struct file_operations fops = { ... .open = open_f, ... +.llseek = no_llseek, /* open uses nonseekable */ }; // use seq_lseek for sequential files ///////////////////////////////////// @ seq depends on !has_llseek @ identifier fops0.fops; identifier sr ~= "seq_read"; @@ struct file_operations fops = { ... .read = sr, ... +.llseek = seq_lseek, /* we have seq_read */ }; // use default_llseek if there is a readdir /////////////////////////////////////////// @ fops1 depends on !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier readdir_e; @@ // any other fop is used that changes pos struct file_operations fops = { ... .readdir = readdir_e, ... +.llseek = default_llseek, /* readdir is present */ }; // use default_llseek if at least one of read/write touches f_pos ///////////////////////////////////////////////////////////////// @ fops2 depends on !fops1 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier read.read_f; @@ // read fops use offset struct file_operations fops = { ... .read = read_f, ... +.llseek = default_llseek, /* read accesses f_pos */ }; @ fops3 depends on !fops1 && !fops2 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier write.write_f; @@ // write fops use offset struct file_operations fops = { ... .write = write_f, ... + .llseek = default_llseek, /* write accesses f_pos */ }; // Use noop_llseek if neither read nor write accesses f_pos /////////////////////////////////////////////////////////// @ fops4 depends on !fops1 && !fops2 && !fops3 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier read_no_fpos.read_f; identifier write_no_fpos.write_f; @@ // write fops use offset struct file_operations fops = { ... .write = write_f, .read = read_f, ... +.llseek = noop_llseek, /* read and write both use no f_pos */ }; @ depends on has_write && !has_read && !fops1 && !fops2 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier write_no_fpos.write_f; @@ struct file_operations fops = { ... .write = write_f, ... +.llseek = noop_llseek, /* write uses no f_pos */ }; @ depends on has_read && !has_write && !fops1 && !fops2 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier read_no_fpos.read_f; @@ struct file_operations fops = { ... .read = read_f, ... +.llseek = noop_llseek, /* read uses no f_pos */ }; @ depends on !has_read && !has_write && !fops1 && !fops2 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; @@ struct file_operations fops = { ... +.llseek = noop_llseek, /* no read or write fn */ }; ===== End semantic patch ===== Signed-off-by: Arnd Bergmann <arnd@arndb.de> Cc: Julia Lawall <julia@diku.dk> Cc: Christoph Hellwig <hch@infradead.org>
2010-08-15 18:52:59 +02:00
.llseek = noop_llseek,
};
/* /status */
static ssize_t
bm_status_read(struct file *file, char __user *buf, size_t nbytes, loff_t *ppos)
{
char *s = enabled ? "enabled\n" : "disabled\n";
return simple_read_from_buffer(buf, nbytes, ppos, s, strlen(s));
}
static ssize_t bm_status_write(struct file *file, const char __user *buffer,
size_t count, loff_t *ppos)
{
int res = parse_command(buffer, count);
binfmt_misc: cleanup on filesystem umount [ Upstream commit 1c5976ef0f7ad76319df748ccb99a4c7ba2ba464 ] Currently, registering a new binary type pins the binfmt_misc filesystem. Specifically, this means that as long as there is at least one binary type registered the binfmt_misc filesystem survives all umounts, i.e. the superblock is not destroyed. Meaning that a umount followed by another mount will end up with the same superblock and the same binary type handlers. This is a behavior we tend to discourage for any new filesystems (apart from a few special filesystems such as e.g. configfs or debugfs). A umount operation without the filesystem being pinned - by e.g. someone holding a file descriptor to an open file - should usually result in the destruction of the superblock and all associated resources. This makes introspection easier and leads to clearly defined, simple and clean semantics. An administrator can rely on the fact that a umount will guarantee a clean slate making it possible to reinitialize a filesystem. Right now all binary types would need to be explicitly deleted before that can happen. This allows us to remove the heavy-handed calls to simple_pin_fs() and simple_release_fs() when creating and deleting binary types. This in turn allows us to replace the current brittle pinning mechanism abusing dget() which has caused a range of bugs judging from prior fixes in [2] and [3]. The additional dget() in load_misc_binary() pins the dentry but only does so for the sake to prevent ->evict_inode() from freeing the node when a user removes the binary type and kill_node() is run. Which would mean ->interpreter and ->interp_file would be freed causing a UAF. This isn't really nicely documented nor is it very clean because it relies on simple_pin_fs() pinning the filesystem as long as at least one binary type exists. Otherwise it would cause load_misc_binary() to hold on to a dentry belonging to a superblock that has been shutdown. Replace that implicit pinning with a clean and simple per-node refcount and get rid of the ugly dget() pinning. A similar mechanism exists for e.g. binderfs (cf. [4]). All the cleanup work can now be done in ->evict_inode(). In a follow-up patch we will make it possible to use binfmt_misc in sandboxes. We will use the cleaner semantics where a umount for the filesystem will cause the superblock and all resources to be deallocated. In preparation for this apply the same semantics to the initial binfmt_misc mount. Note, that this is a user-visible change and as such a uapi change but one that we can reasonably risk. We've discussed this in earlier versions of this patchset (cf. [1]). The main user and provider of binfmt_misc is systemd. Systemd provides binfmt_misc via autofs since it is configurable as a kernel module and is used by a few exotic packages and users. As such a binfmt_misc mount is triggered when /proc/sys/fs/binfmt_misc is accessed and is only provided on demand. Other autofs on demand filesystems include EFI ESP which systemd umounts if the mountpoint stays idle for a certain amount of time. This doesn't apply to the binfmt_misc autofs mount which isn't touched once it is mounted meaning this change can't accidently wipe binary type handlers without someone having explicitly unmounted binfmt_misc. After speaking to systemd folks they don't expect this change to affect them. In line with our general policy, if we see a regression for systemd or other users with this change we will switch back to the old behavior for the initial binfmt_misc mount and have binary types pin the filesystem again. But while we touch this code let's take the chance and let's improve on the status quo. [1]: https://lore.kernel.org/r/20191216091220.465626-2-laurent@vivier.eu [2]: commit 43a4f2619038 ("exec: binfmt_misc: fix race between load_misc_binary() and kill_node()" [3]: commit 83f918274e4b ("exec: binfmt_misc: shift filp_close(interp_file) from kill_node() to bm_evict_inode()") [4]: commit f0fe2c0f050d ("binder: prevent UAF for binderfs devices II") Link: https://lore.kernel.org/r/20211028103114.2849140-1-brauner@kernel.org (v1) Cc: Sargun Dhillon <sargun@sargun.me> Cc: Serge Hallyn <serge@hallyn.com> Cc: Jann Horn <jannh@google.com> Cc: Henning Schild <henning.schild@siemens.com> Cc: Andrei Vagin <avagin@gmail.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Laurent Vivier <laurent@vivier.eu> Cc: linux-fsdevel@vger.kernel.org Acked-by: Serge Hallyn <serge@hallyn.com> Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com> Signed-off-by: Christian Brauner <brauner@kernel.org> Signed-off-by: Kees Cook <keescook@chromium.org> Signed-off-by: Sasha Levin <sashal@kernel.org> (cherry picked from commit 263bcebf5c2ab1fe949517225157f34015124620) Signed-off-by: Vegard Nossum <vegard.nossum@oracle.com>
2021-10-28 12:31:13 +02:00
Node *e, *next;
struct inode *inode;
switch (res) {
case 1:
/* Disable all handlers. */
enabled = 0;
break;
case 2:
/* Enable all handlers. */
enabled = 1;
break;
case 3:
/* Delete all handlers. */
binfmt_misc: cleanup on filesystem umount [ Upstream commit 1c5976ef0f7ad76319df748ccb99a4c7ba2ba464 ] Currently, registering a new binary type pins the binfmt_misc filesystem. Specifically, this means that as long as there is at least one binary type registered the binfmt_misc filesystem survives all umounts, i.e. the superblock is not destroyed. Meaning that a umount followed by another mount will end up with the same superblock and the same binary type handlers. This is a behavior we tend to discourage for any new filesystems (apart from a few special filesystems such as e.g. configfs or debugfs). A umount operation without the filesystem being pinned - by e.g. someone holding a file descriptor to an open file - should usually result in the destruction of the superblock and all associated resources. This makes introspection easier and leads to clearly defined, simple and clean semantics. An administrator can rely on the fact that a umount will guarantee a clean slate making it possible to reinitialize a filesystem. Right now all binary types would need to be explicitly deleted before that can happen. This allows us to remove the heavy-handed calls to simple_pin_fs() and simple_release_fs() when creating and deleting binary types. This in turn allows us to replace the current brittle pinning mechanism abusing dget() which has caused a range of bugs judging from prior fixes in [2] and [3]. The additional dget() in load_misc_binary() pins the dentry but only does so for the sake to prevent ->evict_inode() from freeing the node when a user removes the binary type and kill_node() is run. Which would mean ->interpreter and ->interp_file would be freed causing a UAF. This isn't really nicely documented nor is it very clean because it relies on simple_pin_fs() pinning the filesystem as long as at least one binary type exists. Otherwise it would cause load_misc_binary() to hold on to a dentry belonging to a superblock that has been shutdown. Replace that implicit pinning with a clean and simple per-node refcount and get rid of the ugly dget() pinning. A similar mechanism exists for e.g. binderfs (cf. [4]). All the cleanup work can now be done in ->evict_inode(). In a follow-up patch we will make it possible to use binfmt_misc in sandboxes. We will use the cleaner semantics where a umount for the filesystem will cause the superblock and all resources to be deallocated. In preparation for this apply the same semantics to the initial binfmt_misc mount. Note, that this is a user-visible change and as such a uapi change but one that we can reasonably risk. We've discussed this in earlier versions of this patchset (cf. [1]). The main user and provider of binfmt_misc is systemd. Systemd provides binfmt_misc via autofs since it is configurable as a kernel module and is used by a few exotic packages and users. As such a binfmt_misc mount is triggered when /proc/sys/fs/binfmt_misc is accessed and is only provided on demand. Other autofs on demand filesystems include EFI ESP which systemd umounts if the mountpoint stays idle for a certain amount of time. This doesn't apply to the binfmt_misc autofs mount which isn't touched once it is mounted meaning this change can't accidently wipe binary type handlers without someone having explicitly unmounted binfmt_misc. After speaking to systemd folks they don't expect this change to affect them. In line with our general policy, if we see a regression for systemd or other users with this change we will switch back to the old behavior for the initial binfmt_misc mount and have binary types pin the filesystem again. But while we touch this code let's take the chance and let's improve on the status quo. [1]: https://lore.kernel.org/r/20191216091220.465626-2-laurent@vivier.eu [2]: commit 43a4f2619038 ("exec: binfmt_misc: fix race between load_misc_binary() and kill_node()" [3]: commit 83f918274e4b ("exec: binfmt_misc: shift filp_close(interp_file) from kill_node() to bm_evict_inode()") [4]: commit f0fe2c0f050d ("binder: prevent UAF for binderfs devices II") Link: https://lore.kernel.org/r/20211028103114.2849140-1-brauner@kernel.org (v1) Cc: Sargun Dhillon <sargun@sargun.me> Cc: Serge Hallyn <serge@hallyn.com> Cc: Jann Horn <jannh@google.com> Cc: Henning Schild <henning.schild@siemens.com> Cc: Andrei Vagin <avagin@gmail.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Laurent Vivier <laurent@vivier.eu> Cc: linux-fsdevel@vger.kernel.org Acked-by: Serge Hallyn <serge@hallyn.com> Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com> Signed-off-by: Christian Brauner <brauner@kernel.org> Signed-off-by: Kees Cook <keescook@chromium.org> Signed-off-by: Sasha Levin <sashal@kernel.org> (cherry picked from commit 263bcebf5c2ab1fe949517225157f34015124620) Signed-off-by: Vegard Nossum <vegard.nossum@oracle.com>
2021-10-28 12:31:13 +02:00
inode = d_inode(file_inode(file)->i_sb->s_root);
inode_lock(inode);
binfmt_misc: cleanup on filesystem umount [ Upstream commit 1c5976ef0f7ad76319df748ccb99a4c7ba2ba464 ] Currently, registering a new binary type pins the binfmt_misc filesystem. Specifically, this means that as long as there is at least one binary type registered the binfmt_misc filesystem survives all umounts, i.e. the superblock is not destroyed. Meaning that a umount followed by another mount will end up with the same superblock and the same binary type handlers. This is a behavior we tend to discourage for any new filesystems (apart from a few special filesystems such as e.g. configfs or debugfs). A umount operation without the filesystem being pinned - by e.g. someone holding a file descriptor to an open file - should usually result in the destruction of the superblock and all associated resources. This makes introspection easier and leads to clearly defined, simple and clean semantics. An administrator can rely on the fact that a umount will guarantee a clean slate making it possible to reinitialize a filesystem. Right now all binary types would need to be explicitly deleted before that can happen. This allows us to remove the heavy-handed calls to simple_pin_fs() and simple_release_fs() when creating and deleting binary types. This in turn allows us to replace the current brittle pinning mechanism abusing dget() which has caused a range of bugs judging from prior fixes in [2] and [3]. The additional dget() in load_misc_binary() pins the dentry but only does so for the sake to prevent ->evict_inode() from freeing the node when a user removes the binary type and kill_node() is run. Which would mean ->interpreter and ->interp_file would be freed causing a UAF. This isn't really nicely documented nor is it very clean because it relies on simple_pin_fs() pinning the filesystem as long as at least one binary type exists. Otherwise it would cause load_misc_binary() to hold on to a dentry belonging to a superblock that has been shutdown. Replace that implicit pinning with a clean and simple per-node refcount and get rid of the ugly dget() pinning. A similar mechanism exists for e.g. binderfs (cf. [4]). All the cleanup work can now be done in ->evict_inode(). In a follow-up patch we will make it possible to use binfmt_misc in sandboxes. We will use the cleaner semantics where a umount for the filesystem will cause the superblock and all resources to be deallocated. In preparation for this apply the same semantics to the initial binfmt_misc mount. Note, that this is a user-visible change and as such a uapi change but one that we can reasonably risk. We've discussed this in earlier versions of this patchset (cf. [1]). The main user and provider of binfmt_misc is systemd. Systemd provides binfmt_misc via autofs since it is configurable as a kernel module and is used by a few exotic packages and users. As such a binfmt_misc mount is triggered when /proc/sys/fs/binfmt_misc is accessed and is only provided on demand. Other autofs on demand filesystems include EFI ESP which systemd umounts if the mountpoint stays idle for a certain amount of time. This doesn't apply to the binfmt_misc autofs mount which isn't touched once it is mounted meaning this change can't accidently wipe binary type handlers without someone having explicitly unmounted binfmt_misc. After speaking to systemd folks they don't expect this change to affect them. In line with our general policy, if we see a regression for systemd or other users with this change we will switch back to the old behavior for the initial binfmt_misc mount and have binary types pin the filesystem again. But while we touch this code let's take the chance and let's improve on the status quo. [1]: https://lore.kernel.org/r/20191216091220.465626-2-laurent@vivier.eu [2]: commit 43a4f2619038 ("exec: binfmt_misc: fix race between load_misc_binary() and kill_node()" [3]: commit 83f918274e4b ("exec: binfmt_misc: shift filp_close(interp_file) from kill_node() to bm_evict_inode()") [4]: commit f0fe2c0f050d ("binder: prevent UAF for binderfs devices II") Link: https://lore.kernel.org/r/20211028103114.2849140-1-brauner@kernel.org (v1) Cc: Sargun Dhillon <sargun@sargun.me> Cc: Serge Hallyn <serge@hallyn.com> Cc: Jann Horn <jannh@google.com> Cc: Henning Schild <henning.schild@siemens.com> Cc: Andrei Vagin <avagin@gmail.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Laurent Vivier <laurent@vivier.eu> Cc: linux-fsdevel@vger.kernel.org Acked-by: Serge Hallyn <serge@hallyn.com> Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com> Signed-off-by: Christian Brauner <brauner@kernel.org> Signed-off-by: Kees Cook <keescook@chromium.org> Signed-off-by: Sasha Levin <sashal@kernel.org> (cherry picked from commit 263bcebf5c2ab1fe949517225157f34015124620) Signed-off-by: Vegard Nossum <vegard.nossum@oracle.com>
2021-10-28 12:31:13 +02:00
/*
* In order to add new element or remove elements from the list
* via bm_{entry,register,status}_write() inode_lock() on the
* root inode must be held.
* The lock is exclusive ensuring that the list can't be
* modified. Only load_misc_binary() can access but does so
* read-only. So we only need to take the write lock when we
* actually remove the entry from the list.
*/
list_for_each_entry_safe(e, next, &entries, list)
remove_binfmt_handler(e);
binfmt_misc: cleanup on filesystem umount [ Upstream commit 1c5976ef0f7ad76319df748ccb99a4c7ba2ba464 ] Currently, registering a new binary type pins the binfmt_misc filesystem. Specifically, this means that as long as there is at least one binary type registered the binfmt_misc filesystem survives all umounts, i.e. the superblock is not destroyed. Meaning that a umount followed by another mount will end up with the same superblock and the same binary type handlers. This is a behavior we tend to discourage for any new filesystems (apart from a few special filesystems such as e.g. configfs or debugfs). A umount operation without the filesystem being pinned - by e.g. someone holding a file descriptor to an open file - should usually result in the destruction of the superblock and all associated resources. This makes introspection easier and leads to clearly defined, simple and clean semantics. An administrator can rely on the fact that a umount will guarantee a clean slate making it possible to reinitialize a filesystem. Right now all binary types would need to be explicitly deleted before that can happen. This allows us to remove the heavy-handed calls to simple_pin_fs() and simple_release_fs() when creating and deleting binary types. This in turn allows us to replace the current brittle pinning mechanism abusing dget() which has caused a range of bugs judging from prior fixes in [2] and [3]. The additional dget() in load_misc_binary() pins the dentry but only does so for the sake to prevent ->evict_inode() from freeing the node when a user removes the binary type and kill_node() is run. Which would mean ->interpreter and ->interp_file would be freed causing a UAF. This isn't really nicely documented nor is it very clean because it relies on simple_pin_fs() pinning the filesystem as long as at least one binary type exists. Otherwise it would cause load_misc_binary() to hold on to a dentry belonging to a superblock that has been shutdown. Replace that implicit pinning with a clean and simple per-node refcount and get rid of the ugly dget() pinning. A similar mechanism exists for e.g. binderfs (cf. [4]). All the cleanup work can now be done in ->evict_inode(). In a follow-up patch we will make it possible to use binfmt_misc in sandboxes. We will use the cleaner semantics where a umount for the filesystem will cause the superblock and all resources to be deallocated. In preparation for this apply the same semantics to the initial binfmt_misc mount. Note, that this is a user-visible change and as such a uapi change but one that we can reasonably risk. We've discussed this in earlier versions of this patchset (cf. [1]). The main user and provider of binfmt_misc is systemd. Systemd provides binfmt_misc via autofs since it is configurable as a kernel module and is used by a few exotic packages and users. As such a binfmt_misc mount is triggered when /proc/sys/fs/binfmt_misc is accessed and is only provided on demand. Other autofs on demand filesystems include EFI ESP which systemd umounts if the mountpoint stays idle for a certain amount of time. This doesn't apply to the binfmt_misc autofs mount which isn't touched once it is mounted meaning this change can't accidently wipe binary type handlers without someone having explicitly unmounted binfmt_misc. After speaking to systemd folks they don't expect this change to affect them. In line with our general policy, if we see a regression for systemd or other users with this change we will switch back to the old behavior for the initial binfmt_misc mount and have binary types pin the filesystem again. But while we touch this code let's take the chance and let's improve on the status quo. [1]: https://lore.kernel.org/r/20191216091220.465626-2-laurent@vivier.eu [2]: commit 43a4f2619038 ("exec: binfmt_misc: fix race between load_misc_binary() and kill_node()" [3]: commit 83f918274e4b ("exec: binfmt_misc: shift filp_close(interp_file) from kill_node() to bm_evict_inode()") [4]: commit f0fe2c0f050d ("binder: prevent UAF for binderfs devices II") Link: https://lore.kernel.org/r/20211028103114.2849140-1-brauner@kernel.org (v1) Cc: Sargun Dhillon <sargun@sargun.me> Cc: Serge Hallyn <serge@hallyn.com> Cc: Jann Horn <jannh@google.com> Cc: Henning Schild <henning.schild@siemens.com> Cc: Andrei Vagin <avagin@gmail.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Laurent Vivier <laurent@vivier.eu> Cc: linux-fsdevel@vger.kernel.org Acked-by: Serge Hallyn <serge@hallyn.com> Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com> Signed-off-by: Christian Brauner <brauner@kernel.org> Signed-off-by: Kees Cook <keescook@chromium.org> Signed-off-by: Sasha Levin <sashal@kernel.org> (cherry picked from commit 263bcebf5c2ab1fe949517225157f34015124620) Signed-off-by: Vegard Nossum <vegard.nossum@oracle.com>
2021-10-28 12:31:13 +02:00
inode_unlock(inode);
break;
default:
return res;
}
return count;
}
static const struct file_operations bm_status_operations = {
.read = bm_status_read,
.write = bm_status_write,
llseek: automatically add .llseek fop All file_operations should get a .llseek operation so we can make nonseekable_open the default for future file operations without a .llseek pointer. The three cases that we can automatically detect are no_llseek, seq_lseek and default_llseek. For cases where we can we can automatically prove that the file offset is always ignored, we use noop_llseek, which maintains the current behavior of not returning an error from a seek. New drivers should normally not use noop_llseek but instead use no_llseek and call nonseekable_open at open time. Existing drivers can be converted to do the same when the maintainer knows for certain that no user code relies on calling seek on the device file. The generated code is often incorrectly indented and right now contains comments that clarify for each added line why a specific variant was chosen. In the version that gets submitted upstream, the comments will be gone and I will manually fix the indentation, because there does not seem to be a way to do that using coccinelle. Some amount of new code is currently sitting in linux-next that should get the same modifications, which I will do at the end of the merge window. Many thanks to Julia Lawall for helping me learn to write a semantic patch that does all this. ===== begin semantic patch ===== // This adds an llseek= method to all file operations, // as a preparation for making no_llseek the default. // // The rules are // - use no_llseek explicitly if we do nonseekable_open // - use seq_lseek for sequential files // - use default_llseek if we know we access f_pos // - use noop_llseek if we know we don't access f_pos, // but we still want to allow users to call lseek // @ open1 exists @ identifier nested_open; @@ nested_open(...) { <+... nonseekable_open(...) ...+> } @ open exists@ identifier open_f; identifier i, f; identifier open1.nested_open; @@ int open_f(struct inode *i, struct file *f) { <+... ( nonseekable_open(...) | nested_open(...) ) ...+> } @ read disable optional_qualifier exists @ identifier read_f; identifier f, p, s, off; type ssize_t, size_t, loff_t; expression E; identifier func; @@ ssize_t read_f(struct file *f, char *p, size_t s, loff_t *off) { <+... ( *off = E | *off += E | func(..., off, ...) | E = *off ) ...+> } @ read_no_fpos disable optional_qualifier exists @ identifier read_f; identifier f, p, s, off; type ssize_t, size_t, loff_t; @@ ssize_t read_f(struct file *f, char *p, size_t s, loff_t *off) { ... when != off } @ write @ identifier write_f; identifier f, p, s, off; type ssize_t, size_t, loff_t; expression E; identifier func; @@ ssize_t write_f(struct file *f, const char *p, size_t s, loff_t *off) { <+... ( *off = E | *off += E | func(..., off, ...) | E = *off ) ...+> } @ write_no_fpos @ identifier write_f; identifier f, p, s, off; type ssize_t, size_t, loff_t; @@ ssize_t write_f(struct file *f, const char *p, size_t s, loff_t *off) { ... when != off } @ fops0 @ identifier fops; @@ struct file_operations fops = { ... }; @ has_llseek depends on fops0 @ identifier fops0.fops; identifier llseek_f; @@ struct file_operations fops = { ... .llseek = llseek_f, ... }; @ has_read depends on fops0 @ identifier fops0.fops; identifier read_f; @@ struct file_operations fops = { ... .read = read_f, ... }; @ has_write depends on fops0 @ identifier fops0.fops; identifier write_f; @@ struct file_operations fops = { ... .write = write_f, ... }; @ has_open depends on fops0 @ identifier fops0.fops; identifier open_f; @@ struct file_operations fops = { ... .open = open_f, ... }; // use no_llseek if we call nonseekable_open //////////////////////////////////////////// @ nonseekable1 depends on !has_llseek && has_open @ identifier fops0.fops; identifier nso ~= "nonseekable_open"; @@ struct file_operations fops = { ... .open = nso, ... +.llseek = no_llseek, /* nonseekable */ }; @ nonseekable2 depends on !has_llseek @ identifier fops0.fops; identifier open.open_f; @@ struct file_operations fops = { ... .open = open_f, ... +.llseek = no_llseek, /* open uses nonseekable */ }; // use seq_lseek for sequential files ///////////////////////////////////// @ seq depends on !has_llseek @ identifier fops0.fops; identifier sr ~= "seq_read"; @@ struct file_operations fops = { ... .read = sr, ... +.llseek = seq_lseek, /* we have seq_read */ }; // use default_llseek if there is a readdir /////////////////////////////////////////// @ fops1 depends on !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier readdir_e; @@ // any other fop is used that changes pos struct file_operations fops = { ... .readdir = readdir_e, ... +.llseek = default_llseek, /* readdir is present */ }; // use default_llseek if at least one of read/write touches f_pos ///////////////////////////////////////////////////////////////// @ fops2 depends on !fops1 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier read.read_f; @@ // read fops use offset struct file_operations fops = { ... .read = read_f, ... +.llseek = default_llseek, /* read accesses f_pos */ }; @ fops3 depends on !fops1 && !fops2 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier write.write_f; @@ // write fops use offset struct file_operations fops = { ... .write = write_f, ... + .llseek = default_llseek, /* write accesses f_pos */ }; // Use noop_llseek if neither read nor write accesses f_pos /////////////////////////////////////////////////////////// @ fops4 depends on !fops1 && !fops2 && !fops3 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier read_no_fpos.read_f; identifier write_no_fpos.write_f; @@ // write fops use offset struct file_operations fops = { ... .write = write_f, .read = read_f, ... +.llseek = noop_llseek, /* read and write both use no f_pos */ }; @ depends on has_write && !has_read && !fops1 && !fops2 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier write_no_fpos.write_f; @@ struct file_operations fops = { ... .write = write_f, ... +.llseek = noop_llseek, /* write uses no f_pos */ }; @ depends on has_read && !has_write && !fops1 && !fops2 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; identifier read_no_fpos.read_f; @@ struct file_operations fops = { ... .read = read_f, ... +.llseek = noop_llseek, /* read uses no f_pos */ }; @ depends on !has_read && !has_write && !fops1 && !fops2 && !has_llseek && !nonseekable1 && !nonseekable2 && !seq @ identifier fops0.fops; @@ struct file_operations fops = { ... +.llseek = noop_llseek, /* no read or write fn */ }; ===== End semantic patch ===== Signed-off-by: Arnd Bergmann <arnd@arndb.de> Cc: Julia Lawall <julia@diku.dk> Cc: Christoph Hellwig <hch@infradead.org>
2010-08-15 18:52:59 +02:00
.llseek = default_llseek,
};
/* Superblock handling */
static const struct super_operations s_ops = {
.statfs = simple_statfs,
.evict_inode = bm_evict_inode,
};
static int bm_fill_super(struct super_block *sb, void *data, int silent)
{
int err;
static const struct tree_descr bm_files[] = {
[2] = {"status", &bm_status_operations, S_IWUSR|S_IRUGO},
[3] = {"register", &bm_register_operations, S_IWUSR},
/* last one */ {""}
};
err = simple_fill_super(sb, BINFMTFS_MAGIC, bm_files);
if (!err)
sb->s_op = &s_ops;
return err;
}
static struct dentry *bm_mount(struct file_system_type *fs_type,
int flags, const char *dev_name, void *data)
{
return mount_single(fs_type, flags, data, bm_fill_super);
}
static struct linux_binfmt misc_format = {
.module = THIS_MODULE,
.load_binary = load_misc_binary,
};
static struct file_system_type bm_fs_type = {
.owner = THIS_MODULE,
.name = "binfmt_misc",
.mount = bm_mount,
.kill_sb = kill_litter_super,
};
fs: Limit sys_mount to only request filesystem modules. Modify the request_module to prefix the file system type with "fs-" and add aliases to all of the filesystems that can be built as modules to match. A common practice is to build all of the kernel code and leave code that is not commonly needed as modules, with the result that many users are exposed to any bug anywhere in the kernel. Looking for filesystems with a fs- prefix limits the pool of possible modules that can be loaded by mount to just filesystems trivially making things safer with no real cost. Using aliases means user space can control the policy of which filesystem modules are auto-loaded by editing /etc/modprobe.d/*.conf with blacklist and alias directives. Allowing simple, safe, well understood work-arounds to known problematic software. This also addresses a rare but unfortunate problem where the filesystem name is not the same as it's module name and module auto-loading would not work. While writing this patch I saw a handful of such cases. The most significant being autofs that lives in the module autofs4. This is relevant to user namespaces because we can reach the request module in get_fs_type() without having any special permissions, and people get uncomfortable when a user specified string (in this case the filesystem type) goes all of the way to request_module. After having looked at this issue I don't think there is any particular reason to perform any filtering or permission checks beyond making it clear in the module request that we want a filesystem module. The common pattern in the kernel is to call request_module() without regards to the users permissions. In general all a filesystem module does once loaded is call register_filesystem() and go to sleep. Which means there is not much attack surface exposed by loading a filesytem module unless the filesystem is mounted. In a user namespace filesystems are not mounted unless .fs_flags = FS_USERNS_MOUNT, which most filesystems do not set today. Acked-by: Serge Hallyn <serge.hallyn@canonical.com> Acked-by: Kees Cook <keescook@chromium.org> Reported-by: Kees Cook <keescook@google.com> Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
2013-03-02 19:39:14 -08:00
MODULE_ALIAS_FS("binfmt_misc");
static int __init init_misc_binfmt(void)
{
int err = register_filesystem(&bm_fs_type);
if (!err)
insert_binfmt(&misc_format);
return err;
}
static void __exit exit_misc_binfmt(void)
{
unregister_binfmt(&misc_format);
unregister_filesystem(&bm_fs_type);
}
core_initcall(init_misc_binfmt);
module_exit(exit_misc_binfmt);
MODULE_LICENSE("GPL");