lkml.org 
[lkml]   [2022]   [May]   [13]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
From
SubjectRe: [PATCH] mm/mmap: Fix potential leak on do_mas_align_munmap()
Date
* Liam Howlett <liam.howlett@oracle.com> [220513 10:16]:
> There is a leak when the system is low on resources and fails to
> allocate enough memory to complete the munmap task. Fix this by adding
> the necessary free operations in the unwinding.
>
> Fixes: a760774e7b7b (mm: start tracking VMAs with maple tree)

This is the wrong fixes. It fixes "67/69 mm: remove the vma
linked list".


> Signed-off-by: Liam R. Howlett <Liam.Howlett@oracle.com>
> ---
> mm/mmap.c | 4 +++-
> 1 file changed, 3 insertions(+), 1 deletion(-)
>
> diff --git a/mm/mmap.c b/mm/mmap.c
> index ebfd71a79814..9d41df730b22 100644
> --- a/mm/mmap.c
> +++ b/mm/mmap.c
> @@ -2365,7 +2365,7 @@ do_mas_align_munmap(struct ma_state *mas, struct vm_area_struct *vma,
> return -ENOMEM;
>
> if (mas_preallocate(&mas_detach, vma, GFP_KERNEL))
> - return -ENOMEM;
> + goto detach_alloc_fail;
>
> mas->last = end - 1;
> /*
> @@ -2504,6 +2504,8 @@ do_mas_align_munmap(struct ma_state *mas, struct vm_area_struct *vma,
> map_count_exceeded:
> split_failed:
> userfaultfd_error:
> + mas_destroy(&mas_detach);
> +detach_alloc_fail:
> mas_destroy(mas);
> return error;
> }
> --
> 2.35.1
>
\
 
 \ /
  Last update: 2022-05-13 16:39    [W:0.053 / U:0.188 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site