Messages in this thread | | | Date | Fri, 9 Mar 2012 12:41:50 -0800 | From | Greg KH <> | Subject | Re: [ 17/68] regset: Return -EFAULT, not -EIO, on host-side memory fault |
| |
On Fri, Mar 09, 2012 at 02:34:46PM -0600, Jonathan Nieder wrote: > Greg KH wrote: > > > 3.0-stable review patch. If anyone has any objections, please let me know. > > > > ------------------ > > > > From: "H. Peter Anvin" <hpa@zytor.com> > > > > commit 5189fa19a4b2b4c3bec37c3a019d446148827717 upstream. > > > > There is only one error code to return for a bad user-space buffer > > pointer passed to a system call in the same address space as the > > system call is executed, and that is EFAULT. > > I don't think this has the potential to cause regressions, and it > certainly makes things saner, so from that point of view it looks > good.
"making things sane" is good enough for me :)
> Part of the reason I am asking is to figure out whether the patch > ought to be applied to 2.6.32.y, too.
Why do you think it should not be?
thanks,
greg k-h
| |