On Wed, Aug 23, 2017 at 02:14:08PM -0700, Eric Biggers wrote:
> From: Eric Biggers <ebiggers@xxxxxxxxxx>
>
> Commit 7c051267931a ("mm, fork: make dup_mmap wait for mmap_sem for
> write killable") made it possible to kill a forking task while it is
> waiting to acquire its ->mmap_sem for write, in dup_mmap(). However, it
> was overlooked that this introduced an new error path before a reference
> is taken on the mm_struct's ->exe_file. Since the ->exe_file of the new
> mm_struct was already set to the old ->exe_file by the memcpy() in
> dup_mm(), it was possible for the mmput() in the error path of dup_mm()
> to drop a reference to ->exe_file which was never taken. This caused
> the struct file to later be freed prematurely.
>
> Fix it by updating mm_init() to NULL out the ->exe_file, in the same
> place it clears other things like the list of mmaps.
> diff --git a/kernel/fork.c b/kernel/fork.c
> index e075b7780421..cbbea277b3fb 100644
> --- a/kernel/fork.c
> +++ b/kernel/fork.c
> @@ -806,6 +806,7 @@ static struct mm_struct *mm_init(struct mm_struct *mm, struct task_struct *p,
> mm_init_cpumask(mm);
> mm_init_aio(mm);
> mm_init_owner(mm, p);
> + RCU_INIT_POINTER(mm->exe_file, NULL);
> mmu_notifier_mm_init(mm);
> init_tlb_flush_pending(mm);
> #if defined(CONFIG_TRANSPARENT_HUGEPAGE) && !USE_SPLIT_PMD_PTLOCKS
I've been seeing similar issues on arm64 with use-after-free of a file
and other memory corruption [1].
This patch seems to fix that; a test that normally fired in a few
minutes has been happily running for hours with this applied.
Thanks,
Mark.
[1] https://lkml.kernel.org/r/20170824113743.GA14737@leverpostej