userfaultfd: fix UFFDIO_MOVE livelock on large folio with extra references
Subsystems (3)
PATCH 1/1 userfaultfd: fix UFFDIO_MOVE livelock on large folio with extra references
To move one page out of a large folio, move_pages_ptes() first has to split the folio. __folio_split() refuses if the folio has any reference beyond its page table mappings and the caller's own, and it reports the refusal as -EAGAIN. move_pages() treats returned -EAGAIN as a transient race and keeps retrying the same address, holding the VMA read locks the whole time. This is incorrect since in some cases userspace can hold such a reference indefinitely, such as by registering the page as an io_uring buffer.
An unprivileged user can trigger this with three threads: first holds such a reference on a page of a THP, second calls UFFDIO_MOVE on another page of the same THP and spins in the kernel forever holding the VMA read locks, and third calls fork() or mprotect(), which takes mmap_lock for writing and then waits forever for those read locks. From then on every mmap_lock user of that mm hangs, including other processes inspecting it through /proc/<pid>/, until the process is SIGKILLed. The hung task detector reports the blocked tasks (third one), and with kernel.hung_task_panic=1 the machine panics:
INFO: task wedge:95 blocked for more than 122 seconds. task:wedge state:D stack:12528 pid:95 tgid:95 ppid:94 Call Trace: __schedule+0x596/0x1110 schedule+0x36/0xf0 __vma_start_exclude_readers+0x100/0x170 __vma_start_write+0x51/0xd0 __split_vma+0x208/0x3d0 vma_modify+0x6a8/0x840 vma_modify_flags+0xce/0x110 mprotect_fixup+0xbe/0x480 do_mprotect_pkey+0x2d9/0x580 __x64_sys_mprotect+0x1a/0x30
Fix by returning -EBUSY instead of -EAGAIN in move_pages_ptes(), the caller of split_folio(), when the split fails. Thus, move_pages() does not retry -EBUSY and the ioctl drops its locks and returns to userspace instead of spinning.
--- mm/userfaultfd.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-)
mm/userfaultfd.c +5 -1
@@ -1663,8 +1663,12 @@ static long move_pages_ptes(struct mm_struct *mm, pmd_t *dst_pmd, pmd_t *src_pmdpte_unmap (dst_pte); src_pte = dst_pte =NULL ; ret =split_folio (src_folio); -if (ret) +if (ret) { +/* do not retry, extra refs may be held forever */ +if (ret == -EAGAIN ) + ret = -EBUSY ; goto out; + }/* have to reacquire the folio after it got split */ folio_unlock (src_folio);folio_put (src_folio);
Signature · 1 line
2.55.0
Quoted 54 lines · Cen Zhang (Microsoft)
To move one page out of a large folio,
move_pages_ptes()first has to split the folio.__folio_split()refuses if the folio has any reference beyond its page table mappings and the caller's own, and it reports the refusal as-EAGAIN.move_pages()treats returned-EAGAINas a transient race and keeps retrying the same address, holding the VMA read locks the whole time. This is incorrect since in some cases userspace can hold such a reference indefinitely, such as by registering the page as anio_uringbuffer.An unprivileged user can trigger this with three threads: first holds such a reference on a page of a THP, second calls
UFFDIO_MOVEon another page of the same THP and spins in the kernel forever holding the VMA read locks, and third callsfork()ormprotect(), which takesmmap_lockfor writing and then waits forever for those read locks. From then on everymmap_lockuser of that mm hangs, including other processes inspecting it through /proc/<pid>/, until the process is SIGKILLed. The hung task detector reports the blocked tasks (third one), and with kernel.hung_task_panic=1 the machine panics:INFO: task wedge:95 blocked for more than 122 seconds. task:wedge state:D stack:12528 pid:95 tgid:95 ppid:94 Call Trace: __schedule+0x596/0x1110 schedule+0x36/0xf0 __vma_start_exclude_readers+0x100/0x170 __vma_start_write+0x51/0xd0 __split_vma+0x208/0x3d0 vma_modify+0x6a8/0x840 vma_modify_flags+0xce/0x110 mprotect_fixup+0xbe/0x480 do_mprotect_pkey+0x2d9/0x580 __x64_sys_mprotect+0x1a/0x30Fix by returning
-EBUSYinstead of-EAGAINinmove_pages_ptes(), the caller ofsplit_folio(), when the split fails. Thus,move_pages()does not retry-EBUSYand the ioctl drops its locks and returns to userspace instead of spinning.Fixes adef440691ba ("userfaultfd: UFFDIO_MOVE uABI")Reported-by [email protected]Assisted-by LLMSigned-off-by Cen Zhang (Microsoft) <[email protected]>--- mm/userfaultfd.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-)mm/userfaultfd.c +0
-0@@ -1663,8 +1663,12 @@ static long move_pages_ptes(struct mm_struct *mm, pmd_t *dst_pmd, pmd_t *src_pmdpte_unmap (dst_pte); src_pte = dst_pte =NULL ; ret =split_folio (src_folio);
mm/userfaultfd.c static long move_pages_ptes(struct mm_struct *mm, pmd_t *dst_pmd, pmd_t *src_pmd · line 1674
Quoted 10 lines
mm/userfaultfd.c +5
-1-if (ret) +if (ret) { +/* do not retry, extra refs may be held forever */ +if (ret == -EAGAIN ) + ret = -EBUSY ; goto out; + }/* have to reacquire the folio after it got split */ folio_unlock (src_folio);folio_put (src_folio);
Yeah, that makes sense to me.
Signature · 3 lines
Cheers, David