binder: fix race between munmap() and direct reclaim
PATCH 1/1 binder: fix race between munmap() and direct reclaim
From: Todd Kjos <[email protected]>
commit 5cec2d2e5839f9c0fec319c523a911e0a7fd299f upstream.
An munmap() on a binder device causes binder_vma_close() to be called which clears the alloc->vma pointer.
If direct reclaim causes binder_alloc_free_page() to be called, there is a race where alloc->vma is read into a local vma pointer and then used later after the mm->mmap_sem is acquired. This can result in calling zap_page_range() with an invalid vma which manifests as a use-after-free in zap_page_range().
The fix is to check alloc->vma after acquiring the mmap_sem (which we were acquiring anyway) and skip zap_page_range() if it has changed to NULL.
---
Greg: This applies to 5.0, 4.19, 4.14. Not needed before 4.12.
drivers/android/binder_alloc.c | 18 ++++++++---------- 1 file changed, 8 insertions(+), 10 deletions(-)
drivers/android/binder_alloc.c +8 -10
@@ -959,14 +959,13 @@ enum lru_status binder_alloc_free_page(struct list_head *item, index = page - alloc->pages; page_addr = (uintptr_t )alloc->buffer + index *PAGE_SIZE ; + + mm = alloc->vma_vm_mm; +if (!mmget_not_zero (mm)) + goto err_mmget; +if (!down_write_trylock (&mm->mmap_sem)) + goto err_down_write_mmap_sem_failed; vma =binder_alloc_get_vma (alloc); -if (vma) { -if (!mmget_not_zero (alloc->vma_vm_mm)) - goto err_mmget; - mm = alloc->vma_vm_mm; -if (!down_write_trylock (&mm->mmap_sem)) - goto err_down_write_mmap_sem_failed; - }list_lru_isolate (lru, item);spin_unlock (lock);
@@ -979,10 +978,9 @@ enum lru_status binder_alloc_free_page(struct list_head *item, PAGE_SIZE);trace_binder_unmap_user_end (alloc, index); - -up_write (&mm->mmap_sem); -mmput (mm); } +up_write (&mm->mmap_sem); +mmput (mm);trace_binder_unmap_kernel_start (alloc, index);
Signature · 1 line
2.21.0.392.gf8f6787159e-goog
Greg KH
maintainerQuoted 17 lines · Todd Kjos
From: Todd Kjos <[email protected]>
commit
5cec2d2e5839f9c0fec319c523a911e0a7fd299fupstream.An
munmap()on a binder device causesbinder_vma_close()to be called which clears thealloc->vmapointer.If direct reclaim causes
binder_alloc_free_page()to be called, there is a race wherealloc->vmais read into a local vma pointer and then used later after themm->mmap_semis acquired. This can result in callingzap_page_range()with an invalid vma which manifests as a use-after-free inzap_page_range().The fix is to check
alloc->vmaafter acquiring themmap_sem(which we were acquiring anyway) and skipzap_page_range()if it has changed to NULL.
Quoted 6 lines
Signed-off-by Todd Kjos <[email protected]>Reviewed-by Joel Fernandes (Google) <[email protected]>Cc stable <[email protected]> # 5.0, 4.19, 4.14Signed-off-by Greg Kroah-Hartman <[email protected]>---
Greg: This applies to 5.0, 4.19, 4.14. Not needed before 4.12.
Thanks, now queued up.
greg k-h