Syzbot reported a couple issues introduced by commit 44e602b4e52f ("binder_alloc: add missing mmap_lock calls when using the VMA"), in which we attempt to acquire the mmap_lock when alloc->vma_vm_mm has not been initialized yet.
This can happen if a binder_proc receives a transaction without having previously called mmap() to setup the binder_proc->alloc space in [1]. Also, a similar issue occurs via binder_alloc_print_pages() when we try to dump the debugfs binder stats file in [2].
Sample of syzbot's crash report: ================================================================== KASAN: null-ptr-deref in range [0x0000000000000128-0x000000000000012f] CPU: 0 PID: 3755 Comm: syz-executor229 Not tainted 6.0.0-rc1-next-20220819-syzkaller #0 syz-executor229[3755] cmdline: ./syz-executor2294415195 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022 RIP: 0010:__lock_acquire+0xd83/0x56d0 kernel/locking/lockdep.c:4923 [...] Call Trace: <TASK> lock_acquire kernel/locking/lockdep.c:5666 [inline] lock_acquire+0x1ab/0x570 kernel/locking/lockdep.c:5631 down_read+0x98/0x450 kernel/locking/rwsem.c:1499 mmap_read_lock include/linux/mmap_lock.h:117 [inline] binder_alloc_new_buf_locked drivers/android/binder_alloc.c:405 [inline] binder_alloc_new_buf+0xa5/0x19e0 drivers/android/binder_alloc.c:593 binder_transaction+0x242e/0x9a80 drivers/android/binder.c:3199 binder_thread_write+0x664/0x3220 drivers/android/binder.c:3986 binder_ioctl_write_read drivers/android/binder.c:5036 [inline] binder_ioctl+0x3470/0x6d00 drivers/android/binder.c:5323 vfs_ioctl fs/ioctl.c:51 [inline] __do_sys_ioctl fs/ioctl.c:870 [inline] __se_sys_ioctl fs/ioctl.c:856 [inline] __x64_sys_ioctl+0x193/0x200 fs/ioctl.c:856 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x35/0xb0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x63/0xcd [...] ==================================================================
Fix these issues by setting up alloc->vma_vm_mm pointer during open() and caching directly from current->mm. This guarantees we have a valid reference to take the mmap_lock during scenarios described above.
[1] https://syzkaller.appspot.com/bug?extid=f7dc54e5be28950ac459 [2] https://syzkaller.appspot.com/bug?extid=a75ebe0452711c9e56d9
Fixes: 44e602b4e52f ("binder_alloc: add missing mmap_lock calls when using the VMA") Reported-by: syzbot+f7dc54e5be28950ac459@syzkaller.appspotmail.com Reported-by: syzbot+a75ebe0452711c9e56d9@syzkaller.appspotmail.com Cc: stable@vger.kernel.org # v5.15+ Cc: Liam R. Howlett Liam.Howlett@oracle.com Signed-off-by: Carlos Llamas cmllamas@google.com --- drivers/android/binder_alloc.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/android/binder_alloc.c b/drivers/android/binder_alloc.c index 51f4e1c5cd01..9b1778c00610 100644 --- a/drivers/android/binder_alloc.c +++ b/drivers/android/binder_alloc.c @@ -322,7 +322,6 @@ static inline void binder_alloc_set_vma(struct binder_alloc *alloc, */ if (vma) { vm_start = vma->vm_start; - alloc->vma_vm_mm = vma->vm_mm; mmap_assert_write_locked(alloc->vma_vm_mm); } else { mmap_assert_locked(alloc->vma_vm_mm); @@ -795,7 +794,6 @@ int binder_alloc_mmap_handler(struct binder_alloc *alloc, binder_insert_free_buffer(alloc, buffer); alloc->free_async_space = alloc->buffer_size / 2; binder_alloc_set_vma(alloc, vma); - mmgrab(alloc->vma_vm_mm);
return 0;
@@ -1091,6 +1089,8 @@ static struct shrinker binder_shrinker = { void binder_alloc_init(struct binder_alloc *alloc) { alloc->pid = current->group_leader->pid; + alloc->vma_vm_mm = current->mm; + mmgrab(alloc->vma_vm_mm); mutex_init(&alloc->mutex); INIT_LIST_HEAD(&alloc->buffers); }
On Mon, 29 Aug 2022 20:12:48 +0000 Carlos Llamas cmllamas@google.com wrote:
Syzbot reported a couple issues introduced by commit 44e602b4e52f ("binder_alloc: add missing mmap_lock calls when using the VMA"), in which we attempt to acquire the mmap_lock when alloc->vma_vm_mm has not been initialized yet.
This can happen if a binder_proc receives a transaction without having previously called mmap() to setup the binder_proc->alloc space in [1]. Also, a similar issue occurs via binder_alloc_print_pages() when we try to dump the debugfs binder stats file in [2].
Thanks. I assume you'll be merging all these into mainline?
Fixes: 44e602b4e52f ("binder_alloc: add missing mmap_lock calls when using the VMA") Reported-by: syzbot+f7dc54e5be28950ac459@syzkaller.appspotmail.com Reported-by: syzbot+a75ebe0452711c9e56d9@syzkaller.appspotmail.com Cc: stable@vger.kernel.org # v5.15+
44e602b4e52f is only present in 6.0-rcX?
On Mon, Aug 29, 2022 at 01:34:52PM -0700, Andrew Morton wrote:
On Mon, 29 Aug 2022 20:12:48 +0000 Carlos Llamas cmllamas@google.com wrote:
Syzbot reported a couple issues introduced by commit 44e602b4e52f ("binder_alloc: add missing mmap_lock calls when using the VMA"), in which we attempt to acquire the mmap_lock when alloc->vma_vm_mm has not been initialized yet.
This can happen if a binder_proc receives a transaction without having previously called mmap() to setup the binder_proc->alloc space in [1]. Also, a similar issue occurs via binder_alloc_print_pages() when we try to dump the debugfs binder stats file in [2].
Thanks. I assume you'll be merging all these into mainline?
Yes, I believe Greg will pick up these patches into his char-misc tree.
Fixes: 44e602b4e52f ("binder_alloc: add missing mmap_lock calls when using the VMA") Reported-by: syzbot+f7dc54e5be28950ac459@syzkaller.appspotmail.com Reported-by: syzbot+a75ebe0452711c9e56d9@syzkaller.appspotmail.com Cc: stable@vger.kernel.org # v5.15+
44e602b4e52f is only present in 6.0-rcX?
Right, it was just added to the stable queue earlier today: https://lore.kernel.org/all/20220829105814.857786586@linuxfoundation.org/ https://lore.kernel.org/all/20220829105809.855177179@linuxfoundation.org/
* Carlos Llamas cmllamas@google.com [220829 16:13]:
Syzbot reported a couple issues introduced by commit 44e602b4e52f ("binder_alloc: add missing mmap_lock calls when using the VMA"), in which we attempt to acquire the mmap_lock when alloc->vma_vm_mm has not been initialized yet.
This can happen if a binder_proc receives a transaction without having previously called mmap() to setup the binder_proc->alloc space in [1]. Also, a similar issue occurs via binder_alloc_print_pages() when we try to dump the debugfs binder stats file in [2].
Sample of syzbot's crash report:
KASAN: null-ptr-deref in range [0x0000000000000128-0x000000000000012f] CPU: 0 PID: 3755 Comm: syz-executor229 Not tainted 6.0.0-rc1-next-20220819-syzkaller #0 syz-executor229[3755] cmdline: ./syz-executor2294415195 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022 RIP: 0010:__lock_acquire+0xd83/0x56d0 kernel/locking/lockdep.c:4923 [...] Call Trace:
<TASK> lock_acquire kernel/locking/lockdep.c:5666 [inline] lock_acquire+0x1ab/0x570 kernel/locking/lockdep.c:5631 down_read+0x98/0x450 kernel/locking/rwsem.c:1499 mmap_read_lock include/linux/mmap_lock.h:117 [inline] binder_alloc_new_buf_locked drivers/android/binder_alloc.c:405 [inline] binder_alloc_new_buf+0xa5/0x19e0 drivers/android/binder_alloc.c:593 binder_transaction+0x242e/0x9a80 drivers/android/binder.c:3199 binder_thread_write+0x664/0x3220 drivers/android/binder.c:3986 binder_ioctl_write_read drivers/android/binder.c:5036 [inline] binder_ioctl+0x3470/0x6d00 drivers/android/binder.c:5323 vfs_ioctl fs/ioctl.c:51 [inline] __do_sys_ioctl fs/ioctl.c:870 [inline] __se_sys_ioctl fs/ioctl.c:856 [inline] __x64_sys_ioctl+0x193/0x200 fs/ioctl.c:856 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x35/0xb0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x63/0xcd [...] ==================================================================
Fix these issues by setting up alloc->vma_vm_mm pointer during open() and caching directly from current->mm. This guarantees we have a valid reference to take the mmap_lock during scenarios described above.
[1] https://syzkaller.appspot.com/bug?extid=f7dc54e5be28950ac459 [2] https://syzkaller.appspot.com/bug?extid=a75ebe0452711c9e56d9
Fixes: 44e602b4e52f ("binder_alloc: add missing mmap_lock calls when using the VMA") Reported-by: syzbot+f7dc54e5be28950ac459@syzkaller.appspotmail.com Reported-by: syzbot+a75ebe0452711c9e56d9@syzkaller.appspotmail.com Cc: stable@vger.kernel.org # v5.15+ Cc: Liam R. Howlett Liam.Howlett@oracle.com Signed-off-by: Carlos Llamas cmllamas@google.com
drivers/android/binder_alloc.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/android/binder_alloc.c b/drivers/android/binder_alloc.c index 51f4e1c5cd01..9b1778c00610 100644 --- a/drivers/android/binder_alloc.c +++ b/drivers/android/binder_alloc.c @@ -322,7 +322,6 @@ static inline void binder_alloc_set_vma(struct binder_alloc *alloc, */ if (vma) { vm_start = vma->vm_start;
alloc->vma_vm_mm = vma->vm_mm;
Is this really the null pointer dereference? We check for vma above..?
mmap_assert_write_locked(alloc->vma_vm_mm);} else { mmap_assert_locked(alloc->vma_vm_mm); @@ -795,7 +794,6 @@ int binder_alloc_mmap_handler(struct binder_alloc *alloc, binder_insert_free_buffer(alloc, buffer); alloc->free_async_space = alloc->buffer_size / 2; binder_alloc_set_vma(alloc, vma);
- mmgrab(alloc->vma_vm_mm);
return 0; @@ -1091,6 +1089,8 @@ static struct shrinker binder_shrinker = { void binder_alloc_init(struct binder_alloc *alloc) { alloc->pid = current->group_leader->pid;
- alloc->vma_vm_mm = current->mm;
- mmgrab(alloc->vma_vm_mm); mutex_init(&alloc->mutex); INIT_LIST_HEAD(&alloc->buffers);
}
2.37.2.672.g94769d06f0-goog
On Tue, Aug 30, 2022 at 07:06:37PM +0000, Liam Howlett wrote:
diff --git a/drivers/android/binder_alloc.c b/drivers/android/binder_alloc.c index 51f4e1c5cd01..9b1778c00610 100644 --- a/drivers/android/binder_alloc.c +++ b/drivers/android/binder_alloc.c @@ -322,7 +322,6 @@ static inline void binder_alloc_set_vma(struct binder_alloc *alloc, */ if (vma) { vm_start = vma->vm_start;
alloc->vma_vm_mm = vma->vm_mm;Is this really the null pointer dereference? We check for vma above..?
Not here. The sequence leading to the null-ptr-deref happens when we try to take alloc->vma_vm_mm->mmap_lock in binder_alloc_new_buf_locked() and in binder_alloc_print_pages() without initializing alloc->vma_vm_mm first (e.g. mmap() was never called). These sequences are described in the commit message but basically they translate to mmap_read_lock(NULL) calls.
* Carlos Llamas cmllamas@google.com [220830 15:41]:
On Tue, Aug 30, 2022 at 07:06:37PM +0000, Liam Howlett wrote:
diff --git a/drivers/android/binder_alloc.c b/drivers/android/binder_alloc.c index 51f4e1c5cd01..9b1778c00610 100644 --- a/drivers/android/binder_alloc.c +++ b/drivers/android/binder_alloc.c @@ -322,7 +322,6 @@ static inline void binder_alloc_set_vma(struct binder_alloc *alloc, */ if (vma) { vm_start = vma->vm_start;
alloc->vma_vm_mm = vma->vm_mm;Is this really the null pointer dereference? We check for vma above..?
Not here. The sequence leading to the null-ptr-deref happens when we try to take alloc->vma_vm_mm->mmap_lock in binder_alloc_new_buf_locked() and in binder_alloc_print_pages() without initializing alloc->vma_vm_mm first (e.g. mmap() was never called). These sequences are described in the commit message but basically they translate to mmap_read_lock(NULL) calls.
Ah, this is unnecessary with the rest of the change.
Feel free to add my reviewed-by if you want.
Reviewed-by: Liam R. Howlett Liam.Howlett@oracle.com
On Mon, Aug 29, 2022 at 1:13 PM 'Carlos Llamas' via kernel-team kernel-team@android.com wrote:
Syzbot reported a couple issues introduced by commit 44e602b4e52f ("binder_alloc: add missing mmap_lock calls when using the VMA"), in which we attempt to acquire the mmap_lock when alloc->vma_vm_mm has not been initialized yet.
This can happen if a binder_proc receives a transaction without having previously called mmap() to setup the binder_proc->alloc space in [1]. Also, a similar issue occurs via binder_alloc_print_pages() when we try to dump the debugfs binder stats file in [2].
Sample of syzbot's crash report:
KASAN: null-ptr-deref in range [0x0000000000000128-0x000000000000012f] CPU: 0 PID: 3755 Comm: syz-executor229 Not tainted 6.0.0-rc1-next-20220819-syzkaller #0 syz-executor229[3755] cmdline: ./syz-executor2294415195 Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 07/22/2022 RIP: 0010:__lock_acquire+0xd83/0x56d0 kernel/locking/lockdep.c:4923 [...] Call Trace:
<TASK> lock_acquire kernel/locking/lockdep.c:5666 [inline] lock_acquire+0x1ab/0x570 kernel/locking/lockdep.c:5631 down_read+0x98/0x450 kernel/locking/rwsem.c:1499 mmap_read_lock include/linux/mmap_lock.h:117 [inline] binder_alloc_new_buf_locked drivers/android/binder_alloc.c:405 [inline] binder_alloc_new_buf+0xa5/0x19e0 drivers/android/binder_alloc.c:593 binder_transaction+0x242e/0x9a80 drivers/android/binder.c:3199 binder_thread_write+0x664/0x3220 drivers/android/binder.c:3986 binder_ioctl_write_read drivers/android/binder.c:5036 [inline] binder_ioctl+0x3470/0x6d00 drivers/android/binder.c:5323 vfs_ioctl fs/ioctl.c:51 [inline] __do_sys_ioctl fs/ioctl.c:870 [inline] __se_sys_ioctl fs/ioctl.c:856 [inline] __x64_sys_ioctl+0x193/0x200 fs/ioctl.c:856 do_syscall_x64 arch/x86/entry/common.c:50 [inline] do_syscall_64+0x35/0xb0 arch/x86/entry/common.c:80 entry_SYSCALL_64_after_hwframe+0x63/0xcd [...] ==================================================================
Fix these issues by setting up alloc->vma_vm_mm pointer during open() and caching directly from current->mm. This guarantees we have a valid reference to take the mmap_lock during scenarios described above.
[1] https://syzkaller.appspot.com/bug?extid=f7dc54e5be28950ac459 [2] https://syzkaller.appspot.com/bug?extid=a75ebe0452711c9e56d9
Fixes: 44e602b4e52f ("binder_alloc: add missing mmap_lock calls when using the VMA") Reported-by: syzbot+f7dc54e5be28950ac459@syzkaller.appspotmail.com Reported-by: syzbot+a75ebe0452711c9e56d9@syzkaller.appspotmail.com Cc: stable@vger.kernel.org # v5.15+ Cc: Liam R. Howlett Liam.Howlett@oracle.com Signed-off-by: Carlos Llamas cmllamas@google.com
Acked-by: Todd Kjos tkjos@google.com
drivers/android/binder_alloc.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/android/binder_alloc.c b/drivers/android/binder_alloc.c index 51f4e1c5cd01..9b1778c00610 100644 --- a/drivers/android/binder_alloc.c +++ b/drivers/android/binder_alloc.c @@ -322,7 +322,6 @@ static inline void binder_alloc_set_vma(struct binder_alloc *alloc, */ if (vma) { vm_start = vma->vm_start;
alloc->vma_vm_mm = vma->vm_mm; mmap_assert_write_locked(alloc->vma_vm_mm); } else { mmap_assert_locked(alloc->vma_vm_mm);@@ -795,7 +794,6 @@ int binder_alloc_mmap_handler(struct binder_alloc *alloc, binder_insert_free_buffer(alloc, buffer); alloc->free_async_space = alloc->buffer_size / 2; binder_alloc_set_vma(alloc, vma);
mmgrab(alloc->vma_vm_mm); return 0;@@ -1091,6 +1089,8 @@ static struct shrinker binder_shrinker = { void binder_alloc_init(struct binder_alloc *alloc) { alloc->pid = current->group_leader->pid;
alloc->vma_vm_mm = current->mm;mmgrab(alloc->vma_vm_mm); mutex_init(&alloc->mutex); INIT_LIST_HEAD(&alloc->buffers);}
2.37.2.672.g94769d06f0-goog
-- To unsubscribe from this group and stop receiving emails from it, send an email to kernel-team+unsubscribe@android.com.
linux-stable-mirror@lists.linaro.org
-
Andrew Morton -
Carlos Llamas -
Liam Howlett -
Todd Kjos