As discussed at Linux Plumbers Conference 2018 in Vancouver [1] this is the
implementation of binderfs.
/* Abstract */
binderfs is a backwards-compatible filesystem for Android's binder ipc
mechanism. Each ipc namespace will mount a new binderfs instance. Mounting
binderfs multiple times at different locations in the same ipc namespace
will not cause a new super block to be allocated and hence it will be the
same filesystem instance.
Each new binderfs mount will have its own set of binder devices only
visible in the ipc namespace it has been mounted in. All devices in a new
binderfs mount will follow the scheme binder%d and numbering will always
start at 0.
/* Backwards compatibility */
Devices requested in the Kconfig via CONFIG_ANDROID_BINDER_DEVICES for the
initial ipc namespace will work as before. They will be registered via
misc_register() and appear in the devtmpfs mount. Specifically, the
standard devices binder, hwbinder, and vndbinder will all appear in their
standard locations in /dev. Mounting or unmounting the binderfs mount in
the initial ipc namespace will have no effect on these devices, i.e. they
will neither show up in the binderfs mount nor will they disappear when the
binderfs mount is gone.
/* binder-control */
Each new binderfs instance comes with a binder-control device. No other
devices will be present at first. The binder-control device can be used to
dynamically allocate binder devices. All requests operate on the binderfs
mount the binder-control device resides in.
Assuming a new instance of binderfs has been mounted at /dev/binderfs
via mount -t binderfs binderfs /dev/binderfs. Then a request to create a
new binder device can be made as illustrated in [2].
Binderfs devices can simply be removed via unlink().
/* Implementation details */
- dynamic major number allocation:
When binderfs is registered as a new filesystem it will dynamically
allocate a new major number. The allocated major number will be returned
in struct binderfs_device when a new binder device is allocated.
- global minor number tracking:
Minor are tracked in a global idr struct that is capped at
BINDERFS_MAX_MINOR. The minor number tracker is protected by a global
mutex. This is the only point of contention between binderfs mounts.
- struct binderfs_info:
Each binderfs super block has its own struct binderfs_info that tracks
specific details about a binderfs instance:
- ipc namespace
- dentry of the binder-control device
- root uid and root gid of the user namespace the binderfs instance
was mounted in
- mountable by user namespace root:
binderfs can be mounted by user namespace root in a non-initial user
namespace. The devices will be owned by user namespace root.
- binderfs binder devices without misc infrastructure:
New binder devices associated with a binderfs mount do not use the
full misc_register() infrastructure.
The misc_register() infrastructure can only create new devices in the
host's devtmpfs mount. binderfs does however only make devices appear
under its own mountpoint and thus allocates new character device nodes
from the inode of the root dentry of the super block. This will have
the side-effect that binderfs specific device nodes do not appear in
sysfs. This behavior is similar to devpts allocated pts devices and
has no effect on the functionality of the ipc mechanism itself.
[1]: https://goo.gl/JL2tfX
[2]: program to allocate a new binderfs binder device:
#define _GNU_SOURCE
#include <errno.h>
#include <fcntl.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <sys/ioctl.h>
#include <sys/stat.h>
#include <sys/types.h>
#include <unistd.h>
#include <linux/android/binder_ctl.h>
int main(int argc, char *argv[])
{
int fd, ret, saved_errno;
size_t len;
struct binderfs_device device = { 0 };
if (argc < 2)
exit(EXIT_FAILURE);
len = strlen(argv[1]);
if (len > BINDERFS_MAX_NAME)
exit(EXIT_FAILURE);
memcpy(device.name, argv[1], len);
fd = open("/dev/binderfs/binder-control", O_RDONLY | O_CLOEXEC);
if (fd < 0) {
printf("%s - Failed to open binder-control device\n",
strerror(errno));
exit(EXIT_FAILURE);
}
ret = ioctl(fd, BINDER_CTL_ADD, &device);
saved_errno = errno;
close(fd);
errno = saved_errno;
if (ret < 0) {
printf("%s - Failed to allocate new binder device\n",
strerror(errno));
exit(EXIT_FAILURE);
}
printf("Allocated new binder device with major %d, minor %d, and "
"name %s\n", device.major, device.minor,
device.name);
exit(EXIT_SUCCESS);
}
Cc: Martijn Coenen <maco@android.com>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com>
Acked-by: Todd Kjos <tkjos@google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
44d8047f1d ("binder: use standard functions to allocate fds")
exposed a pre-existing issue in the binder driver.
fdget() is used in ksys_ioctl() as a performance optimization.
One of the rules associated with fdget() is that ksys_close() must
not be called between the fdget() and the fdput(). There is a case
where this requirement is not met in the binder driver which results
in the reference count dropping to 0 when the device is still in
use. This can result in use-after-free or other issues.
If userpace has passed a file-descriptor for the binder driver using
a BINDER_TYPE_FDA object, then kys_close() is called on it when
handling a binder_ioctl(BC_FREE_BUFFER) command. This violates
the assumptions for using fdget().
The problem is fixed by deferring the close using task_work_add(). A
new variant of __close_fd() was created that returns a struct file
with a reference. The fput() is deferred instead of using ksys_close().
Fixes: 44d8047f1d ("binder: use standard functions to allocate fds")
Suggested-by: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Todd Kjos <tkjos@google.com>
Cc: stable <stable@vger.kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When dumping out binder transactions via a debug node,
the output is too verbose if a process has many nodes.
Change the output for transaction dumps to only display
nodes with pending async transactions.
Signed-off-by: Todd Kjos <tkjos@google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Add __acquire()/__release() annnotations to fix warnings
in sparse context checking
There is one case where the warning was due to a lack of
a "default:" case in a switch statement where a lock was
being released in each of the cases, so the default
case was added.
Signed-off-by: Todd Kjos <tkjos@google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Malicious code can attempt to free buffers using the BC_FREE_BUFFER
ioctl to binder. There are protections against a user freeing a buffer
while in use by the kernel, however there was a window where
BC_FREE_BUFFER could be used to free a recently allocated buffer that
was not completely initialized. This resulted in a use-after-free
detected by KASAN with a malicious test program.
This window is closed by setting the buffer's allow_user_free attribute
to 0 when the buffer is allocated or when the user has previously freed
it instead of waiting for the caller to set it. The problem was that
when the struct buffer was recycled, allow_user_free was stale and set
to 1 allowing a free to go through.
Signed-off-by: Todd Kjos <tkjos@google.com>
Acked-by: Arve Hjønnevåg <arve@android.com>
Cc: stable <stable@vger.kernel.org> # 4.14
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Fixes the following sparse warning:
drivers/android/binder.c:3312:1: warning:
symbol 'binder_free_buf' was not declared. Should it be static?
Signed-off-by: Wei Yongjun <weiyongjun1@huawei.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This allows the context manager to retrieve information about nodes
that it holds a reference to, such as the current number of
references to those nodes.
Such information can for example be used to determine whether the
servicemanager is the only process holding a reference to a node.
This information can then be passed on to the process holding the
node, which can in turn decide whether it wants to shut down to
reduce resource usage.
Signed-off-by: Martijn Coenen <maco@android.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Binder uses internal fs interfaces to allocate and install fds:
__alloc_fd
__fd_install
__close_fd
get_files_struct
put_files_struct
These were used to support the passing of fds between processes
as part of a transaction. The actual allocation and installation
of the fds in the target process was handled by the sending
process so the standard functions, alloc_fd() and fd_install()
which assume task==current couldn't be used.
This patch refactors this mechanism so that the fds are
allocated and installed by the target process allowing the
standard functions to be used.
The sender now creates a list of fd fixups that contains the
struct *file and the address to fixup with the new fd once
it is allocated. This list is processed by the target process
when the transaction is dequeued.
A new error case is introduced by this change. If an async
transaction with file descriptors cannot allocate new
fds in the target (probably due to out of file descriptors),
the transaction is discarded with a log message. In the old
implementation this would have been detected in the sender
context and failed prior to sending.
Signed-off-by: Todd Kjos <tkjos@google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When a process dies, failed reply is sent to the sender of any transaction
queued on a dead thread's todo list. The sender asserts that the
received failed reply corresponds to the head of the transaction stack.
This assert can fail if the dead thread is allowed to send outgoing
transactions when there is already a transaction on its todo list,
because this new transaction can end up on the transaction stack of the
original sender. The following steps illustrate how this assertion can
fail.
1. Thread1 sends txn19 to Thread2
(T1->transaction_stack=txn19, T2->todo+=txn19)
2. Without processing todo list, Thread2 sends txn20 to Thread1
(T1->todo+=txn20, T2->transaction_stack=txn20)
3. T1 processes txn20 on its todo list
(T1->transaction_stack=txn20->txn19, T1->todo=<empty>)
4. T2 dies, T2->todo cleanup attempts to send failed reply for txn19, but
T1->transaction_stack points to txn20 -- assertion failes
Step 2. is the incorrect behavior. When there is a transaction on a
thread's todo list, this thread should not be able to send any outgoing
synchronous transactions. Only the head of the todo list needs to be
checked because only threads that are waiting for proc work can directly
receive work from another thread, and no work is allowed to be queued
on such a thread without waking up the thread. This patch also enforces
that a thread is not waiting for proc work when a work is directly
enqueued to its todo list.
Acked-by: Arve Hjønnevåg <arve@android.com>
Signed-off-by: Sherry Yang <sherryy@android.com>
Reviewed-by: Martijn Coenen <maco@android.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
There is RaceFuzzer report like below because we have no lock to close
below the race between binder_mmap and binder_alloc_new_buf_locked.
To close the race, let's use memory barrier so that if someone see
alloc->vma is not NULL, alloc->vma_vm_mm should be never NULL.
(I didn't add stable mark intentionallybecause standard android
userspace libraries that interact with binder (libbinder & libhwbinder)
prevent the mmap/ioctl race. - from Todd)
"
Thread interleaving:
CPU0 (binder_alloc_mmap_handler) CPU1 (binder_alloc_new_buf_locked)
===== =====
// drivers/android/binder_alloc.c
// #L718 (v4.18-rc3)
alloc->vma = vma;
// drivers/android/binder_alloc.c
// #L346 (v4.18-rc3)
if (alloc->vma == NULL) {
...
// alloc->vma is not NULL at this point
return ERR_PTR(-ESRCH);
}
...
// #L438
binder_update_page_range(alloc, 0,
(void *)PAGE_ALIGN((uintptr_t)buffer->data),
end_page_addr);
// In binder_update_page_range() #L218
// But still alloc->vma_vm_mm is NULL here
if (need_mm && mmget_not_zero(alloc->vma_vm_mm))
alloc->vma_vm_mm = vma->vm_mm;
Crash Log:
==================================================================
BUG: KASAN: null-ptr-deref in __atomic_add_unless include/asm-generic/atomic-instrumented.h:89 [inline]
BUG: KASAN: null-ptr-deref in atomic_add_unless include/linux/atomic.h:533 [inline]
BUG: KASAN: null-ptr-deref in mmget_not_zero include/linux/sched/mm.h:75 [inline]
BUG: KASAN: null-ptr-deref in binder_update_page_range+0xece/0x18e0 drivers/android/binder_alloc.c:218
Write of size 4 at addr 0000000000000058 by task syz-executor0/11184
CPU: 1 PID: 11184 Comm: syz-executor0 Not tainted 4.18.0-rc3 #1
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.8.2-0-g33fbe13 by qemu-project.org 04/01/2014
Call Trace:
__dump_stack lib/dump_stack.c:77 [inline]
dump_stack+0x16e/0x22c lib/dump_stack.c:113
kasan_report_error mm/kasan/report.c:352 [inline]
kasan_report+0x163/0x380 mm/kasan/report.c:412
check_memory_region_inline mm/kasan/kasan.c:260 [inline]
check_memory_region+0x140/0x1a0 mm/kasan/kasan.c:267
kasan_check_write+0x14/0x20 mm/kasan/kasan.c:278
__atomic_add_unless include/asm-generic/atomic-instrumented.h:89 [inline]
atomic_add_unless include/linux/atomic.h:533 [inline]
mmget_not_zero include/linux/sched/mm.h:75 [inline]
binder_update_page_range+0xece/0x18e0 drivers/android/binder_alloc.c:218
binder_alloc_new_buf_locked drivers/android/binder_alloc.c:443 [inline]
binder_alloc_new_buf+0x467/0xc30 drivers/android/binder_alloc.c:513
binder_transaction+0x125b/0x4fb0 drivers/android/binder.c:2957
binder_thread_write+0xc08/0x2770 drivers/android/binder.c:3528
binder_ioctl_write_read.isra.39+0x24f/0x8e0 drivers/android/binder.c:4456
binder_ioctl+0xa86/0xf34 drivers/android/binder.c:4596
vfs_ioctl fs/ioctl.c:46 [inline]
do_vfs_ioctl+0x154/0xd40 fs/ioctl.c:686
ksys_ioctl+0x94/0xb0 fs/ioctl.c:701
__do_sys_ioctl fs/ioctl.c:708 [inline]
__se_sys_ioctl fs/ioctl.c:706 [inline]
__x64_sys_ioctl+0x43/0x50 fs/ioctl.c:706
do_syscall_64+0x167/0x4b0 arch/x86/entry/common.c:290
entry_SYSCALL_64_after_hwframe+0x49/0xbe
"
Signed-off-by: Todd Kjos <tkjos@google.com>
Signed-off-by: Minchan Kim <minchan@kernel.org>
Reviewed-by: Martijn Coenen <maco@android.com>
Cc: stable <stable@vger.kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
If asm/cacheflush.h is included first, the following build warnings are
seen with sparc32 builds.
In file included from arch/sparc/include/asm/cacheflush.h:11:0,
from drivers/android/binder.c:54:
arch/sparc/include/asm/cacheflush_32.h:40:37: warning:
'struct page' declared inside parameter list will not be visible
outside of this definition or declaration
Moving the asm/ include after linux/ includes solves the problem.
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
If asm/cacheflush.h is included first, the following build warnings are
seen with sparc32 builds.
In file included from ./arch/sparc/include/asm/cacheflush.h:11:0,
from drivers/android/binder_alloc.c:20:
./arch/sparc/include/asm/cacheflush_32.h:40:37: warning:
'struct page' declared inside parameter list
Moving the asm/ include after linux/ includes fixes the problem.
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Guenter Roeck <linux@roeck-us.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
As of commit 7124330dab ("m68k/uaccess: Revive 64-bit
get_user()"), the 64-bit Android binder interface builds fine on m68k.
Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Use new return type vm_fault_t for fault handler in
struct vm_operations_struct. For now, this is just
documenting that the function returns a VM_FAULT
value rather than an errno. Once all instances are
converted, vm_fault_t will become a distinct type.
Reference id -> 1c8f422059 ("mm: change return type
to vm_fault_t")
Signed-off-by: Souptick Joarder <jrdr.linux@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
binder_update_page_range needs down_write of mmap_sem because
vm_insert_page need to change vma->vm_flags to VM_MIXEDMAP unless
it is set. However, when I profile binder working, it seems
every binder buffers should be mapped in advance by binder_mmap.
It means we could set VM_MIXEDMAP in binder_mmap time which is
already hold a mmap_sem as down_write so binder_update_page_range
doesn't need to hold a mmap_sem as down_write.
Please use proper API down_read. It would help mmap_sem contention
problem as well as fixing down_write abuse.
Ganesh Mahendran tested app launching and binder throughput test
and he said he couldn't find any problem and I did binder latency
test per Greg KH request(Thanks Martijn to teach me how I can do)
I cannot find any problem, too.
Cc: Ganesh Mahendran <opensource.ganesh@gmail.com>
Cc: Joe Perches <joe@perches.com>
Cc: Arve Hjønnevåg <arve@android.com>
Cc: Todd Kjos <tkjos@google.com>
Reviewed-by: Martijn Coenen <maco@android.com>
Signed-off-by: Minchan Kim <minchan@kernel.org>
Reviewed-by: Joel Fernandes (Google) <joel@joelfernandes.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When to execute binder_stat_br the e->cmd has been modifying as BR_OK
instead of the original return error cmd, in fact we want to know the
original return error, such as BR_DEAD_REPLY or BR_FAILED_REPLY, etc.
instead of always BR_OK, in order to avoid the value of the e->cmd is
always BR_OK, so we need assign the value of the e->cmd to cmd before
e->cmd = BR_OK.
Signed-off-by: songjinshi <songjinshi@xiaomi.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
New devices launching with Android P need to use the 64-bit
binder interface, even on 32-bit SoCs [0].
This change removes the Kconfig option to select the 32-bit
binder interface. We don't think this will affect existing
userspace for the following reasons:
1) The latest Android common tree is 4.14, so we don't
believe any Android devices are on kernels >4.14.
2) Android devices launch on an LTS release and stick with
it, so we wouldn't expect devices running on <= 4.14 now
to upgrade to 4.17 or later. But even if they did, they'd
rebuild the world (kernel + userspace) anyway.
3) Other userspaces like 'anbox' are already using the
64-bit interface.
Note that this change doesn't remove the 32-bit UAPI
itself; the reason for that is that Android userspace
always uses the latest UAPI headers from upstream, and
userspace retains 32-bit support for devices that are
upgrading. This will be removed as well in 2-3 years,
at which point we can remove the code from the UAPI
as well.
Finally, this change introduces build errors on archs where
64-bit get_user/put_user is not supported, so make binder
unavailable on m68k (which wouldn't want it anyway).
[0]: https://android-review.googlesource.com/c/platform/build/+/595193
Signed-off-by: Martijn Coenen <maco@android.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
