Re: [PATCH] mm: get_user_pages(write,force) refuse to COW in shared areas

[Hugh Dickins <hughd@google.com>]

On Fri, 25 Apr 2014, Oleg Nesterov wrote:
> 
> And I forgot to mention, there is another reason why I would like to
> change uprobes to follow the same convention. I still think it would
> be better to kill __replace_page() and use gup(FOLL_WRITE | FORCE)
> in uprobe_write_opcode().

Oh, please please do!  I assumed there was a good atomicity reason
for doing it that way, but if you can do it with gup() please do so.

I went off into a little rant about __replace_page() in my reply to you;
but then felt that you had approached with an honest enquiry, and didn't
deserve a rant in response, so I deleted it.

And, of course, I'm conscious that I have from start to finish withheld
my involvement from uprobes, and refused to review that __replace_page()
(beyond insisting that it not be put in a common place for sharing with
KSM, because I just couldn't guarantee it for uprobes).  I'm afraid that
it's much like HWPoison to me, a complexity (nastiness?) way beyond what
I have time to support myself.

Hugh

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>