commit 4df153652cc46545722879415937582028c18af5 upstream.
Binder objects are processed and copied individually into the target
buffer during transactions. Any raw data in-between these objects is
copied as well. However, this raw data copy lacks an out-of-bounds
check. If the raw data exceeds the data section size then the copy
overwrites the offsets section. This eventually triggers an error that
attempts to unwind the processed objects. However, at this point the
offsets used to index these objects are now corrupted.
Unwinding with corrupted offsets can result in decrements of arbitrary
nodes and lead to their premature release. Other users of such nodes are
left with a dangling pointer triggering a use-after-free. This issue is
made evident by the following KASAN report (trimmed):
==================================================================
BUG: KASAN: slab-use-after-free in _raw_spin_lock+0xe4/0x19c
Write of size 4 at addr ffff47fc91598f04 by task binder-util/743
CPU: 9 UID: 0 PID: 743 Comm: binder-util Not tainted 6.11.0-rc4 #1
Hardware name: linux,dummy-virt (DT)
Call trace:
_raw_spin_lock+0xe4/0x19c
binder_free_buf+0x128/0x434
binder_thread_write+0x8a4/0x3260
binder_ioctl+0x18f0/0x258c
[...]
Allocated by task 743:
__kmalloc_cache_noprof+0x110/0x270
binder_new_node+0x50/0x700
binder_transaction+0x413c/0x6da8
binder_thread_write+0x978/0x3260
binder_ioctl+0x18f0/0x258c
[...]
Freed by task 745:
kfree+0xbc/0x208
binder_thread_read+0x1c5c/0x37d4
binder_ioctl+0x16d8/0x258c
[...]
==================================================================
To avoid this issue, let's check that the raw data copy is within the
boundaries of the data section.
Fixes: 6d98eb95b4 ("binder: avoid potential data leakage when copying txn")
Cc: Todd Kjos <tkjos@google.com>
Cc: stable@vger.kernel.org
Signed-off-by: Carlos Llamas <cmllamas@google.com>
Link: https://lore.kernel.org/r/20240822182353.2129600-1-cmllamas@google.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 31643d84b8c3d9c846aa0e20bc033e46c68c7e7d upstream.
With the introduction of binder_available_for_proc_work_ilocked() in
commit 1b77e9dcc3 ("ANDROID: binder: remove proc waitqueue") a binder
thread can only "wait_for_proc_work" after its thread->looper has been
marked as BINDER_LOOPER_STATE_{ENTERED|REGISTERED}.
This means an unregistered reader risks waiting indefinitely for work
since it never gets added to the proc->waiting_threads. If there are no
further references to its waitqueue either the task will hang. The same
applies to readers using the (e)poll interface.
I couldn't find the rationale behind this restriction. So this patch
restores the previous behavior of allowing unregistered threads to
"wait_for_proc_work". Note that an error message for this scenario,
which had previously become unreachable, is now re-enabled.
Fixes: 1b77e9dcc3 ("ANDROID: binder: remove proc waitqueue")
Cc: stable@vger.kernel.org
Cc: Martijn Coenen <maco@google.com>
Cc: Arve Hjønnevåg <arve@google.com>
Signed-off-by: Carlos Llamas <cmllamas@google.com>
Link: https://lore.kernel.org/r/20240711201452.2017543-1-cmllamas@google.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 42316941335644a98335f209daafa4c122f28983 upstream.
The type defined for the BINDER_SET_MAX_THREADS ioctl was changed from
size_t to __u32 in order to avoid incompatibility issues between 32 and
64-bit kernels. However, the internal types used to copy from user and
store the value were never updated. Use u32 to fix the inconsistency.
Fixes: a9350fc859 ("staging: android: binder: fix BINDER_SET_MAX_THREADS declaration")
Reported-by: Arve Hjønnevåg <arve@android.com>
Cc: stable@vger.kernel.org
Signed-off-by: Carlos Llamas <cmllamas@google.com>
Reviewed-by: Alice Ryhl <aliceryhl@google.com>
Link: https://lore.kernel.org/r/20240421173750.3117808-1-cmllamas@google.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit aaef73821a3b0194a01bd23ca77774f704a04d40 upstream.
Commit 6d98eb95b4 ("binder: avoid potential data leakage when copying
txn") introduced changes to how binder objects are copied. In doing so,
it unintentionally removed an offset alignment check done through calls
to binder_alloc_copy_from_buffer() -> check_buffer().
These calls were replaced in binder_get_object() with copy_from_user(),
so now an explicit offset alignment check is needed here. This avoids
later complications when unwinding the objects gets harder.
It is worth noting this check existed prior to commit 7a67a39320
("binder: add function to copy binder object from buffer"), likely
removed due to redundancy at the time.
Fixes: 6d98eb95b4 ("binder: avoid potential data leakage when copying txn")
Cc: stable@vger.kernel.org
Signed-off-by: Carlos Llamas <cmllamas@google.com>
Acked-by: Todd Kjos <tkjos@google.com>
Link: https://lore.kernel.org/r/20240330190115.1877819-1-cmllamas@google.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 97830f3c3088638ff90b20dfba2eb4d487bf14d7 upstream.
In (e)poll mode, threads often depend on I/O events to determine when
data is ready for consumption. Within binder, a thread may initiate a
command via BINDER_WRITE_READ without a read buffer and then make use
of epoll_wait() or similar to consume any responses afterwards.
It is then crucial that epoll threads are signaled via wakeup when they
queue their own work. Otherwise, they risk waiting indefinitely for an
event leaving their work unhandled. What is worse, subsequent commands
won't trigger a wakeup either as the thread has pending work.
Fixes: 457b9a6f09 ("Staging: android: add binder driver")
Cc: Arve Hjønnevåg <arve@android.com>
Cc: Martijn Coenen <maco@android.com>
Cc: Alice Ryhl <aliceryhl@google.com>
Cc: Steven Moreland <smoreland@google.com>
Cc: stable@vger.kernel.org # v4.19+
Signed-off-by: Carlos Llamas <cmllamas@google.com>
Link: https://lore.kernel.org/r/20240131215347.1808751-1-cmllamas@google.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 9a9ab0d963621d9d12199df9817e66982582d5a5 upstream.
Task A calls binder_update_page_range() to allocate and insert pages on
a remote address space from Task B. For this, Task A pins the remote mm
via mmget_not_zero() first. This can race with Task B do_exit() and the
final mmput() refcount decrement will come from Task A.
Task A | Task B
------------------+------------------
mmget_not_zero() |
| do_exit()
| exit_mm()
| mmput()
mmput() |
exit_mmap() |
remove_vma() |
fput() |
In this case, the work of ____fput() from Task B is queued up in Task A
as TWA_RESUME. So in theory, Task A returns to userspace and the cleanup
work gets executed. However, Task A instead sleep, waiting for a reply
from Task B that never comes (it's dead).
This means the binder_deferred_release() is blocked until an unrelated
binder event forces Task A to go back to userspace. All the associated
death notifications will also be delayed until then.
In order to fix this use mmput_async() that will schedule the work in
the corresponding mm->async_put_work WQ instead of Task A.
Fixes: 457b9a6f09 ("Staging: android: add binder driver")
Reviewed-by: Alice Ryhl <aliceryhl@google.com>
Signed-off-by: Carlos Llamas <cmllamas@google.com>
Link: https://lore.kernel.org/r/20231201172212.1813387-4-cmllamas@google.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit c6d05e0762ab276102246d24affd1e116a46aa0c upstream.
Each transaction is associated with a 'struct binder_buffer' that stores
the metadata about its buffer area. Since commit 74310e06be ("android:
binder: Move buffer out of area shared with user space") this struct is
no longer embedded within the buffer itself but is instead allocated on
the heap to prevent userspace access to this driver-exclusive info.
Unfortunately, the space of this struct is still being accounted for in
the total buffer size calculation, specifically for async transactions.
This results in an additional 104 bytes added to every async buffer
request, and this area is never used.
This wasted space can be substantial. If we consider the maximum mmap
buffer space of SZ_4M, the driver will reserve half of it for async
transactions, or 0x200000. This area should, in theory, accommodate up
to 262,144 buffers of the minimum 8-byte size. However, after adding
the extra 'sizeof(struct binder_buffer)', the total number of buffers
drops to only 18,724, which is a sad 7.14% of the actual capacity.
This patch fixes the buffer size calculation to enable the utilization
of the entire async buffer space. This is expected to reduce the number
of -ENOSPC errors that are seen on the field.
Fixes: 74310e06be ("android: binder: Move buffer out of area shared with user space")
Signed-off-by: Carlos Llamas <cmllamas@google.com>
Reviewed-by: Alice Ryhl <aliceryhl@google.com>
Link: https://lore.kernel.org/r/20231201172212.1813387-6-cmllamas@google.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 3f489c2067c5824528212b0fc18b28d51332d906 upstream.
The mmap read lock is used during the shrinker's callback, which means
that using alloc->vma pointer isn't safe as it can race with munmap().
As of commit dd2283f260 ("mm: mmap: zap pages with read mmap_sem in
munmap") the mmap lock is downgraded after the vma has been isolated.
I was able to reproduce this issue by manually adding some delays and
triggering page reclaiming through the shrinker's debug sysfs. The
following KASAN report confirms the UAF:
==================================================================
BUG: KASAN: slab-use-after-free in zap_page_range_single+0x470/0x4b8
Read of size 8 at addr ffff356ed50e50f0 by task bash/478
CPU: 1 PID: 478 Comm: bash Not tainted 6.6.0-rc5-00055-g1c8b86a3799f-dirty #70
Hardware name: linux,dummy-virt (DT)
Call trace:
zap_page_range_single+0x470/0x4b8
binder_alloc_free_page+0x608/0xadc
__list_lru_walk_one+0x130/0x3b0
list_lru_walk_node+0xc4/0x22c
binder_shrink_scan+0x108/0x1dc
shrinker_debugfs_scan_write+0x2b4/0x500
full_proxy_write+0xd4/0x140
vfs_write+0x1ac/0x758
ksys_write+0xf0/0x1dc
__arm64_sys_write+0x6c/0x9c
Allocated by task 492:
kmem_cache_alloc+0x130/0x368
vm_area_alloc+0x2c/0x190
mmap_region+0x258/0x18bc
do_mmap+0x694/0xa60
vm_mmap_pgoff+0x170/0x29c
ksys_mmap_pgoff+0x290/0x3a0
__arm64_sys_mmap+0xcc/0x144
Freed by task 491:
kmem_cache_free+0x17c/0x3c8
vm_area_free_rcu_cb+0x74/0x98
rcu_core+0xa38/0x26d4
rcu_core_si+0x10/0x1c
__do_softirq+0x2fc/0xd24
Last potentially related work creation:
__call_rcu_common.constprop.0+0x6c/0xba0
call_rcu+0x10/0x1c
vm_area_free+0x18/0x24
remove_vma+0xe4/0x118
do_vmi_align_munmap.isra.0+0x718/0xb5c
do_vmi_munmap+0xdc/0x1fc
__vm_munmap+0x10c/0x278
__arm64_sys_munmap+0x58/0x7c
Fix this issue by performing instead a vma_lookup() which will fail to
find the vma that was isolated before the mmap lock downgrade. Note that
this option has better performance than upgrading to a mmap write lock
which would increase contention. Plus, mmap_write_trylock() has been
recently removed anyway.
Fixes: dd2283f260 ("mm: mmap: zap pages with read mmap_sem in munmap")
Cc: stable@vger.kernel.org
Cc: Liam Howlett <liam.howlett@oracle.com>
Cc: Minchan Kim <minchan@kernel.org>
Reviewed-by: Alice Ryhl <aliceryhl@google.com>
Signed-off-by: Carlos Llamas <cmllamas@google.com>
Link: https://lore.kernel.org/r/20231201172212.1813387-3-cmllamas@google.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 6ac061db9c58ca5b9270b1b3940d2464fb3ff183 upstream.
Use EPOLLERR instead of POLLERR to make sure it is cast to the correct
__poll_t type. This fixes the following sparse issue:
drivers/android/binder.c:5030:24: warning: incorrect type in return expression (different base types)
drivers/android/binder.c:5030:24: expected restricted __poll_t
drivers/android/binder.c:5030:24: got int
Fixes: f88982679f ("binder: check for binder_thread allocation failure in binder_poll()")
Cc: stable@vger.kernel.org
Cc: Eric Biggers <ebiggers@google.com>
Reviewed-by: Alice Ryhl <aliceryhl@google.com>
Signed-off-by: Carlos Llamas <cmllamas@google.com>
Link: https://lore.kernel.org/r/20231201172212.1813387-2-cmllamas@google.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 1aa3aaf8953c84bad398adf6c3cabc9d6685bf7d upstream
A transaction complete work is allocated and queued for each
transaction. Under certain conditions the work->type might be marked as
BINDER_WORK_TRANSACTION_ONEWAY_SPAM_SUSPECT to notify userspace about
potential spamming threads or as BINDER_WORK_TRANSACTION_PENDING when
the target is currently frozen.
However, these work types are not being handled in binder_release_work()
so they will leak during a cleanup. This was reported by syzkaller with
the following kmemleak dump:
BUG: memory leak
unreferenced object 0xffff88810e2d6de0 (size 32):
comm "syz-executor338", pid 5046, jiffies 4294968230 (age 13.590s)
hex dump (first 32 bytes):
e0 6d 2d 0e 81 88 ff ff e0 6d 2d 0e 81 88 ff ff .m-......m-.....
04 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
backtrace:
[<ffffffff81573b75>] kmalloc_trace+0x25/0x90 mm/slab_common.c:1114
[<ffffffff83d41873>] kmalloc include/linux/slab.h:599 [inline]
[<ffffffff83d41873>] kzalloc include/linux/slab.h:720 [inline]
[<ffffffff83d41873>] binder_transaction+0x573/0x4050 drivers/android/binder.c:3152
[<ffffffff83d45a05>] binder_thread_write+0x6b5/0x1860 drivers/android/binder.c:4010
[<ffffffff83d486dc>] binder_ioctl_write_read drivers/android/binder.c:5066 [inline]
[<ffffffff83d486dc>] binder_ioctl+0x1b2c/0x3cf0 drivers/android/binder.c:5352
[<ffffffff816b25f2>] vfs_ioctl fs/ioctl.c:51 [inline]
[<ffffffff816b25f2>] __do_sys_ioctl fs/ioctl.c:871 [inline]
[<ffffffff816b25f2>] __se_sys_ioctl fs/ioctl.c:857 [inline]
[<ffffffff816b25f2>] __x64_sys_ioctl+0xf2/0x140 fs/ioctl.c:857
[<ffffffff84b30008>] do_syscall_x64 arch/x86/entry/common.c:50 [inline]
[<ffffffff84b30008>] do_syscall_64+0x38/0xb0 arch/x86/entry/common.c:80
[<ffffffff84c0008b>] entry_SYSCALL_64_after_hwframe+0x63/0xcd
Fix the leaks by kfreeing these work types in binder_release_work() and
handle them as a BINDER_WORK_TRANSACTION_COMPLETE cleanup.
Cc: stable@vger.kernel.org
Fixes: a7dc1e6f99 ("binder: tell userspace to dump current backtrace when detected oneway spamming")
Reported-by: syzbot+7f10c1653e35933c0f1e@syzkaller.appspotmail.com
Closes: https://syzkaller.appspot.com/bug?extid=7f10c1653e35933c0f1e
Suggested-by: Alice Ryhl <aliceryhl@google.com>
Signed-off-by: Carlos Llamas <cmllamas@google.com>
Reviewed-by: Alice Ryhl <aliceryhl@google.com>
Acked-by: Todd Kjos <tkjos@google.com>
Link: https://lore.kernel.org/r/20230922175138.230331-1-cmllamas@google.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
[cmllamas: backport to v6.1 by dropping BINDER_WORK_TRANSACTION_PENDING
as commit 0567461a7a6e is not present. Remove fixes tag accordingly.]
Signed-off-by: Carlos Llamas <cmllamas@google.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
commit d1d8875c8c13517f6fd1ff8d4d3e1ac366a17e07 upstream.
[ cmllamas: clean forward port from commit 015ac18be7 ("binder: fix
UAF of alloc->vma in race with munmap()") in 5.10 stable. It is needed
in mainline after the revert of commit a43cfc87ca ("android: binder:
stop saving a pointer to the VMA") as pointed out by Liam. The commit
log and tags have been tweaked to reflect this. ]
In commit 720c241924 ("ANDROID: binder: change down_write to
down_read") binder assumed the mmap read lock is sufficient to protect
alloc->vma inside binder_update_page_range(). This used to be accurate
until commit dd2283f260 ("mm: mmap: zap pages with read mmap_sem in
munmap"), which now downgrades the mmap_lock after detaching the vma
from the rbtree in munmap(). Then it proceeds to teardown and free the
vma with only the read lock held.
This means that accesses to alloc->vma in binder_update_page_range() now
will race with vm_area_free() in munmap() and can cause a UAF as shown
in the following KASAN trace:
==================================================================
BUG: KASAN: use-after-free in vm_insert_page+0x7c/0x1f0
Read of size 8 at addr ffff16204ad00600 by task server/558
CPU: 3 PID: 558 Comm: server Not tainted 5.10.150-00001-gdc8dcf942daa #1
Hardware name: linux,dummy-virt (DT)
Call trace:
dump_backtrace+0x0/0x2a0
show_stack+0x18/0x2c
dump_stack+0xf8/0x164
print_address_description.constprop.0+0x9c/0x538
kasan_report+0x120/0x200
__asan_load8+0xa0/0xc4
vm_insert_page+0x7c/0x1f0
binder_update_page_range+0x278/0x50c
binder_alloc_new_buf+0x3f0/0xba0
binder_transaction+0x64c/0x3040
binder_thread_write+0x924/0x2020
binder_ioctl+0x1610/0x2e5c
__arm64_sys_ioctl+0xd4/0x120
el0_svc_common.constprop.0+0xac/0x270
do_el0_svc+0x38/0xa0
el0_svc+0x1c/0x2c
el0_sync_handler+0xe8/0x114
el0_sync+0x180/0x1c0
Allocated by task 559:
kasan_save_stack+0x38/0x6c
__kasan_kmalloc.constprop.0+0xe4/0xf0
kasan_slab_alloc+0x18/0x2c
kmem_cache_alloc+0x1b0/0x2d0
vm_area_alloc+0x28/0x94
mmap_region+0x378/0x920
do_mmap+0x3f0/0x600
vm_mmap_pgoff+0x150/0x17c
ksys_mmap_pgoff+0x284/0x2dc
__arm64_sys_mmap+0x84/0xa4
el0_svc_common.constprop.0+0xac/0x270
do_el0_svc+0x38/0xa0
el0_svc+0x1c/0x2c
el0_sync_handler+0xe8/0x114
el0_sync+0x180/0x1c0
Freed by task 560:
kasan_save_stack+0x38/0x6c
kasan_set_track+0x28/0x40
kasan_set_free_info+0x24/0x4c
__kasan_slab_free+0x100/0x164
kasan_slab_free+0x14/0x20
kmem_cache_free+0xc4/0x34c
vm_area_free+0x1c/0x2c
remove_vma+0x7c/0x94
__do_munmap+0x358/0x710
__vm_munmap+0xbc/0x130
__arm64_sys_munmap+0x4c/0x64
el0_svc_common.constprop.0+0xac/0x270
do_el0_svc+0x38/0xa0
el0_svc+0x1c/0x2c
el0_sync_handler+0xe8/0x114
el0_sync+0x180/0x1c0
[...]
==================================================================
To prevent the race above, revert back to taking the mmap write lock
inside binder_update_page_range(). One might expect an increase of mmap
lock contention. However, binder already serializes these calls via top
level alloc->mutex. Also, there was no performance impact shown when
running the binder benchmark tests.
Fixes: c0fd2101781e ("Revert "android: binder: stop saving a pointer to the VMA"")
Fixes: dd2283f260 ("mm: mmap: zap pages with read mmap_sem in munmap")
Reported-by: Jann Horn <jannh@google.com>
Closes: https://lore.kernel.org/all/20230518144052.xkj6vmddccq4v66b@revolver
Cc: <stable@vger.kernel.org>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Yang Shi <yang.shi@linux.alibaba.com>
Cc: Liam Howlett <liam.howlett@oracle.com>
Signed-off-by: Carlos Llamas <cmllamas@google.com>
Acked-by: Todd Kjos <tkjos@google.com>
Link: https://lore.kernel.org/r/20230519195950.1775656-1-cmllamas@google.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit bdc1c5fac982845a58d28690cdb56db8c88a530d upstream.
In binder_transaction_buffer_release() the 'failed_at' offset indicates
the number of objects to clean up. However, this function was changed by
commit 44d8047f1d ("binder: use standard functions to allocate fds"),
to release all the objects in the buffer when 'failed_at' is zero.
This introduced an issue when a transaction buffer is released without
any objects having been processed so far. In this case, 'failed_at' is
indeed zero yet it is misinterpreted as releasing the entire buffer.
This leads to use-after-free errors where nodes are incorrectly freed
and subsequently accessed. Such is the case in the following KASAN
report:
==================================================================
BUG: KASAN: slab-use-after-free in binder_thread_read+0xc40/0x1f30
Read of size 8 at addr ffff4faf037cfc58 by task poc/474
CPU: 6 PID: 474 Comm: poc Not tainted 6.3.0-12570-g7df047b3f0aa #5
Hardware name: linux,dummy-virt (DT)
Call trace:
dump_backtrace+0x94/0xec
show_stack+0x18/0x24
dump_stack_lvl+0x48/0x60
print_report+0xf8/0x5b8
kasan_report+0xb8/0xfc
__asan_load8+0x9c/0xb8
binder_thread_read+0xc40/0x1f30
binder_ioctl+0xd9c/0x1768
__arm64_sys_ioctl+0xd4/0x118
invoke_syscall+0x60/0x188
[...]
Allocated by task 474:
kasan_save_stack+0x3c/0x64
kasan_set_track+0x2c/0x40
kasan_save_alloc_info+0x24/0x34
__kasan_kmalloc+0xb8/0xbc
kmalloc_trace+0x48/0x5c
binder_new_node+0x3c/0x3a4
binder_transaction+0x2b58/0x36f0
binder_thread_write+0x8e0/0x1b78
binder_ioctl+0x14a0/0x1768
__arm64_sys_ioctl+0xd4/0x118
invoke_syscall+0x60/0x188
[...]
Freed by task 475:
kasan_save_stack+0x3c/0x64
kasan_set_track+0x2c/0x40
kasan_save_free_info+0x38/0x5c
__kasan_slab_free+0xe8/0x154
__kmem_cache_free+0x128/0x2bc
kfree+0x58/0x70
binder_dec_node_tmpref+0x178/0x1fc
binder_transaction_buffer_release+0x430/0x628
binder_transaction+0x1954/0x36f0
binder_thread_write+0x8e0/0x1b78
binder_ioctl+0x14a0/0x1768
__arm64_sys_ioctl+0xd4/0x118
invoke_syscall+0x60/0x188
[...]
==================================================================
In order to avoid these issues, let's always calculate the intended
'failed_at' offset beforehand. This is renamed and wrapped in a helper
function to make it clear and convenient.
Fixes: 32e9f56a96 ("binder: don't detect sender/target during buffer cleanup")
Reported-by: Zi Fan Tan <zifantan@google.com>
Cc: stable@vger.kernel.org
Signed-off-by: Carlos Llamas <cmllamas@google.com>
Acked-by: Todd Kjos <tkjos@google.com>
Link: https://lore.kernel.org/r/20230505203020.4101154-1-cmllamas@google.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit 0fa53349c3acba0239369ba4cd133740a408d246 upstream.
Bring back the original lockless design in binder_alloc to determine
whether the buffer setup has been completed by the ->mmap() handler.
However, this time use smp_load_acquire() and smp_store_release() to
wrap all the ordering in a single macro call.
Also, add comments to make it evident that binder uses alloc->vma to
determine when the binder_alloc has been fully initialized. In these
scenarios acquiring the mmap_lock is not required.
Fixes: a43cfc87ca ("android: binder: stop saving a pointer to the VMA")
Cc: Liam Howlett <liam.howlett@oracle.com>
Cc: Suren Baghdasaryan <surenb@google.com>
Cc: stable@vger.kernel.org
Signed-off-by: Carlos Llamas <cmllamas@google.com>
Link: https://lore.kernel.org/r/20230502201220.1756319-3-cmllamas@google.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit c0fd2101781ef761b636769b2f445351f71c3626 upstream.
This reverts commit a43cfc87ca.
This patch fixed an issue reported by syzkaller in [1]. However, this
turned out to be only a band-aid in binder. The root cause, as bisected
by syzkaller, was fixed by commit 5789151e48 ("mm/mmap: undo ->mmap()
when mas_preallocate() fails"). We no longer need the patch for binder.
Reverting such patch allows us to have a lockless access to alloc->vma
in specific cases where the mmap_lock is not required. This approach
avoids the contention that caused a performance regression.
[1] https://lore.kernel.org/all/0000000000004a0dbe05e1d749e0@google.com
[cmllamas: resolved conflicts with rework of alloc->mm and removal of
binder_alloc_set_vma() also fixed comment section]
Fixes: a43cfc87ca ("android: binder: stop saving a pointer to the VMA")
Cc: Liam Howlett <liam.howlett@oracle.com>
Cc: Suren Baghdasaryan <surenb@google.com>
Cc: stable@vger.kernel.org
Signed-off-by: Carlos Llamas <cmllamas@google.com>
Link: https://lore.kernel.org/r/20230502201220.1756319-2-cmllamas@google.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
commit b15655b12ddca7ade09807f790bafb6fab61b50a upstream.
This reverts commit 44e602b4e5.
This caused a performance regression particularly when pages are getting
reclaimed. We don't need to acquire the mmap_lock to determine when the
binder buffer has been fully initialized. A subsequent patch will bring
back the lockless approach for this.
[cmllamas: resolved trivial conflicts with renaming of alloc->mm]
Fixes: 44e602b4e5 ("binder_alloc: add missing mmap_lock calls when using the VMA")
Cc: Liam Howlett <liam.howlett@oracle.com>
Cc: Suren Baghdasaryan <surenb@google.com>
Cc: stable@vger.kernel.org
Signed-off-by: Carlos Llamas <cmllamas@google.com>
Link: https://lore.kernel.org/r/20230502201220.1756319-1-cmllamas@google.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Since commit 1da52815d5 ("binder: fix alloc->vma_vm_mm null-ptr
dereference") binder caches a pointer to the current->mm during open().
This fixes a null-ptr dereference reported by syzkaller. Unfortunately,
it also opens the door for a process to update its mm after the open(),
(e.g. via execve) making the cached alloc->mm pointer invalid.
Things get worse when the process continues to mmap() a vma. From this
point forward, binder will attempt to find this vma using an obsolete
alloc->mm reference. Such as in binder_update_page_range(), where the
wrong vma is obtained via vma_lookup(), yet binder proceeds to happily
insert new pages into it.
To avoid this issue fail the ->mmap() callback if we detect a mismatch
between the vma->vm_mm and the original alloc->mm pointer. This prevents
alloc->vm_addr from getting set, so that any subsequent vma_lookup()
calls fail as expected.
Fixes: 1da52815d5 ("binder: fix alloc->vma_vm_mm null-ptr dereference")
Reported-by: Jann Horn <jannh@google.com>
Cc: <stable@vger.kernel.org> # 5.15+
Signed-off-by: Carlos Llamas <cmllamas@google.com>
Acked-by: Todd Kjos <tkjos@google.com>
Link: https://lore.kernel.org/r/20221104231235.348958-1-cmllamas@google.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Pull scheduler updates from Ingo Molnar:
"Debuggability:
- Change most occurances of BUG_ON() to WARN_ON_ONCE()
- Reorganize & fix TASK_ state comparisons, turn it into a bitmap
- Update/fix misc scheduler debugging facilities
Load-balancing & regular scheduling:
- Improve the behavior of the scheduler in presence of lot of
SCHED_IDLE tasks - in particular they should not impact other
scheduling classes.
- Optimize task load tracking, cleanups & fixes
- Clean up & simplify misc load-balancing code
Freezer:
- Rewrite the core freezer to behave better wrt thawing and be
simpler in general, by replacing PF_FROZEN with TASK_FROZEN &
fixing/adjusting all the fallout.
Deadline scheduler:
- Fix the DL capacity-aware code
- Factor out dl_task_is_earliest_deadline() &
replenish_dl_new_period()
- Relax/optimize locking in task_non_contending()
Cleanups:
- Factor out the update_current_exec_runtime() helper
- Various cleanups, simplifications"
* tag 'sched-core-2022-10-07' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (41 commits)
sched: Fix more TASK_state comparisons
sched: Fix TASK_state comparisons
sched/fair: Move call to list_last_entry() in detach_tasks
sched/fair: Cleanup loop_max and loop_break
sched/fair: Make sure to try to detach at least one movable task
sched: Show PF_flag holes
freezer,sched: Rewrite core freezer logic
sched: Widen TAKS_state literals
sched/wait: Add wait_event_state()
sched/completion: Add wait_for_completion_state()
sched: Add TASK_ANY for wait_task_inactive()
sched: Change wait_task_inactive()s match_state
freezer,umh: Clean up freezer/initrd interaction
freezer: Have {,un}lock_system_sleep() save/restore flags
sched: Rename task_running() to task_on_cpu()
sched/fair: Cleanup for SIS_PROP
sched/fair: Default to false in test_idle_cores()
sched/fair: Remove useless check in select_idle_core()
sched/fair: Avoid double search on same cpu
sched/fair: Remove redundant check in select_idle_smt()
...
Rewrite the core freezer to behave better wrt thawing and be simpler
in general.
By replacing PF_FROZEN with TASK_FROZEN, a special block state, it is
ensured frozen tasks stay frozen until thawed and don't randomly wake
up early, as is currently possible.
As such, it does away with PF_FROZEN and PF_FREEZER_SKIP, freeing up
two PF_flags (yay!).
Specifically; the current scheme works a little like:
freezer_do_not_count();
schedule();
freezer_count();
And either the task is blocked, or it lands in try_to_freezer()
through freezer_count(). Now, when it is blocked, the freezer
considers it frozen and continues.
However, on thawing, once pm_freezing is cleared, freezer_count()
stops working, and any random/spurious wakeup will let a task run
before its time.
That is, thawing tries to thaw things in explicit order; kernel
threads and workqueues before doing bringing SMP back before userspace
etc.. However due to the above mentioned races it is entirely possible
for userspace tasks to thaw (by accident) before SMP is back.
This can be a fatal problem in asymmetric ISA architectures (eg ARMv9)
where the userspace task requires a special CPU to run.
As said; replace this with a special task state TASK_FROZEN and add
the following state transitions:
TASK_FREEZABLE -> TASK_FROZEN
__TASK_STOPPED -> TASK_FROZEN
__TASK_TRACED -> TASK_FROZEN
The new TASK_FREEZABLE can be set on any state part of TASK_NORMAL
(IOW. TASK_INTERRUPTIBLE and TASK_UNINTERRUPTIBLE) -- any such state
is already required to deal with spurious wakeups and the freezer
causes one such when thawing the task (since the original state is
lost).
The special __TASK_{STOPPED,TRACED} states *can* be restored since
their canonical state is in ->jobctl.
With this, frozen tasks need an explicit TASK_FROZEN wakeup and are
free of undue (early / spurious) wakeups.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
Link: https://lore.kernel.org/r/20220822114649.055452969@infradead.org
Update the kernel-doc section of struct binder_alloc to fix the
following warnings reported by ./scripts/kernel-doc:
warning: Function parameter or member 'mutex' not described in 'binder_alloc'
warning: Function parameter or member 'vma_addr' not described in 'binder_alloc'
No functional changes in this patch.
Reviewed-by: Christian Brauner (Microsoft) <brauner@kernel.org>
Acked-by: Todd Kjos <tkjos@google.com>
Signed-off-by: Carlos Llamas <cmllamas@google.com>
Link: https://lore.kernel.org/r/20220906135948.3048225-4-cmllamas@google.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
