Re: [PATCH] Fix mlockall for PROT_NONE mappings

[Muli Ben-Yehuda <mulix@mulix.org>]

[-- Attachment #1: Type: text/plain, Size: 1224 bytes --]

On Thu, Oct 09, 2003 at 01:22:45PM +0200, Andi Kleen wrote:
> On Thu, Oct 09, 2003 at 12:49:18PM +0200, Muli Ben-Yehuda wrote:
> > On Thu, Oct 09, 2003 at 12:42:18PM +0200, Andi Kleen wrote:
> > 
> > > I added a new argument force==2 to get_user_pages that means to ignore
> > > SIGBUS or unaccessible pages errors. MAY_* is still checked for like
> > > with the old force ==1, it just doesn't error out now for SIGBUS
> > > errors on handle_mm_fault. 
> > 
> > How about making it an enum or define for code readability? I'd much
> > rather see an IGNORE_BAD_PAGES or some such than a cryptic '2' in the
> > code. I can send a patch to that effect if you'd like? 
> 
> Doesn't look essential. You could submit it as a follow-up patch as soon
> as Linus merged this version, but I'm not sure it satisfies the current
> "no more cleanups" rule because it isn't a bugfix.

That's exactly the reason I wrote to you in the first place - if your
patch gets merged *with* the readability cleanups, it is irrelevant
whether the a follow up patch is acceptible or not.  

I agree that improved readability is not "essential". Do you agree
that it's preferable?  
-- 
Muli Ben-Yehuda
http://www.mulix.org


[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 189 bytes --]