public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
* [PATCH] msync shouldn't go over bss sections
@ 2004-05-24 14:14 Alexander Nyberg
  2004-05-24 20:43 ` Andrew Morton
  0 siblings, 1 reply; 2+ messages in thread
From: Alexander Nyberg @ 2004-05-24 14:14 UTC (permalink / raw)
  To: akpm; +Cc: linux-kernel

This changes the behaviour of msync_interval() to make it impossible to
try to sys_msync() anything not file mapped.


--- mm/msync_orig.c     2004-05-23 21:31:32.000000000 +0200
+++ mm/msync.c  2004-05-24 16:10:24.000000000 +0200
@@ -137,7 +137,7 @@ static int filemap_sync(struct vm_area_s
 static int msync_interval(struct vm_area_struct * vma,
        unsigned long start, unsigned long end, int flags)
 {
-       int ret = 0;
+       int ret = -ENOMEM;
        struct file * file = vma->vm_file;
  
        if ((flags & MS_INVALIDATE) && (vma->vm_flags & VM_LOCKED))


^ permalink raw reply	[flat|nested] 2+ messages in thread

* Re: [PATCH] msync shouldn't go over bss sections
  2004-05-24 14:14 [PATCH] msync shouldn't go over bss sections Alexander Nyberg
@ 2004-05-24 20:43 ` Andrew Morton
  0 siblings, 0 replies; 2+ messages in thread
From: Andrew Morton @ 2004-05-24 20:43 UTC (permalink / raw)
  To: Alexander Nyberg; +Cc: linux-kernel

Alexander Nyberg <alexn@telia.com> wrote:
>
> This changes the behaviour of msync_interval() to make it impossible to
> try to sys_msync() anything not file mapped.
> 

Well the patch doesn't "make it impossible".  It makes it return -ENOMEM.

>From my reading of the spec your patch converts correct behaviour to
incorrect behaviour, and even if that's untrue, I think we're stuck with
the current behaviour - this change can break current applications.


> --- mm/msync_orig.c     2004-05-23 21:31:32.000000000 +0200
> +++ mm/msync.c  2004-05-24 16:10:24.000000000 +0200
> @@ -137,7 +137,7 @@ static int filemap_sync(struct vm_area_s
>  static int msync_interval(struct vm_area_struct * vma,
>         unsigned long start, unsigned long end, int flags)
>  {
> -       int ret = 0;
> +       int ret = -ENOMEM;
>         struct file * file = vma->vm_file;
>   
>         if ((flags & MS_INVALIDATE) && (vma->vm_flags & VM_LOCKED))

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2004-05-24 20:40 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2004-05-24 14:14 [PATCH] msync shouldn't go over bss sections Alexander Nyberg
2004-05-24 20:43 ` Andrew Morton

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox