diff options
| author | Yisheng Xie <xieyisheng1@huawei.com> | 2018-02-28 14:59:22 +0800 |
|---|---|---|
| committer | Greg Kroah-Hartman <gregkh@linuxfoundation.org> | 2018-03-24 11:02:43 +0100 |
| commit | bcaf449b5809846b10474127d267502b2a241bb1 (patch) | |
| tree | 58815cd0caab79cb4b1e2553470415185da4f27b /drivers | |
| parent | 1d60b78021042f79c01958720a52d4a3914bb955 (diff) | |
staging: android: ashmem: Fix possible deadlock in ashmem_ioctl
commit 740a5759bf222332fbb5eda42f89aa25ba38f9b2 upstream.
ashmem_mutex may create a chain of dependencies like:
CPU0 CPU1
mmap syscall ioctl syscall
-> mmap_sem (acquired) -> ashmem_ioctl
-> ashmem_mmap -> ashmem_mutex (acquired)
-> ashmem_mutex (try to acquire) -> copy_from_user
-> mmap_sem (try to acquire)
There is a lock odering problem between mmap_sem and ashmem_mutex causing
a lockdep splat[1] during a syzcaller test. This patch fixes the problem
by move copy_from_user out of ashmem_mutex.
[1] https://www.spinics.net/lists/kernel/msg2733200.html
Fixes: ce8a3a9e76d0 (staging: android: ashmem: Fix a race condition in pin ioctls)
Reported-by: syzbot+d7a918a7a8e1c952bc36@syzkaller.appspotmail.com
Signed-off-by: Yisheng Xie <xieyisheng1@huawei.com>
Cc: "Joel Fernandes (Google)" <joel.opensrc@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Diffstat (limited to 'drivers')
| -rw-r--r-- | drivers/staging/android/ashmem.c | 8 |
1 files changed, 3 insertions, 5 deletions
diff --git a/drivers/staging/android/ashmem.c b/drivers/staging/android/ashmem.c index d9941b0c468d..893b2836089c 100644 --- a/drivers/staging/android/ashmem.c +++ b/drivers/staging/android/ashmem.c @@ -709,16 +709,14 @@ static int ashmem_pin_unpin(struct ashmem_area *asma, unsigned long cmd, size_t pgstart, pgend; int ret = -EINVAL; + if (unlikely(copy_from_user(&pin, p, sizeof(pin)))) + return -EFAULT; + mutex_lock(&ashmem_mutex); if (unlikely(!asma->file)) goto out_unlock; - if (unlikely(copy_from_user(&pin, p, sizeof(pin)))) { - ret = -EFAULT; - goto out_unlock; - } - /* per custom, you can pass zero for len to mean "everything onward" */ if (!pin.len) pin.len = PAGE_ALIGN(asma->size) - pin.offset; |