lkml.org 
[lkml]   [2005]   [Nov]   [10]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: Nick's core remove PageReserved broke vmware...
Quoting Hugh Dickins <hugh@veritas.com>:
> Subject: Re: Nick's core remove PageReserved broke vmware...
>
> On Tue, 8 Nov 2005, Michael S. Tsirkin wrote:
> >
> > Hugh, did you have something like the following in mind
> > (this is only boot-tested and only on x86-64)?
>
> Yes, that looks pretty good to me, a few comments below.
> Only another twenty or so architectures to go ;)

Yea, thats easy :).

> (I had been imagining VM_DONTCOPY plus another flag to say set by the
> user: better your way, we would like to merge these vmas, and
> VM_DONTCOPY is in that peculiar list of special flags that prevent merging.)
>
> > Hmm, maybe MADV_INHERIT and MADV_DONT_INHERIT would be better names,
>
> You're right, and it would be a good choice, except that MAP_INHERIT on
> some OSes has a particular meaning (about inheriting across an exec),
> so I think avoid confusion with that. MADV_DONTFORK and MADV_DOFORK?
> Accompanied by VM_DONTFORK?

Its actually similiar.
Maybe MADV_FORK_INHERIT/MADV_FORK_DONT_INHERIT then?
I find using "COPY" there confusing, since the copy is only done on write ...

> > Index: linux-2.6.14-dontcopy/include/asm-x86_64/mman.h
> > ===================================================================
> > --- linux-2.6.14-dontcopy.orig/include/asm-x86_64/mman.h
> 2005-11-08 23:19:35.000000000 +0200
> > +++ linux-2.6.14-dontcopy/include/asm-x86_64/mman.h 2005-11-08
> 23:19:46.000000000 +0200
> > @@ -36,6 +36,8 @@
> > #define MADV_SEQUENTIAL 0x2 /* read-ahead aggressively */
> > #define MADV_WILLNEED 0x3 /* pre-fault pages */
> > #define MADV_DONTNEED 0x4 /* discard these pages
> */
> > +#define MADV_DONTCOPY 0x30 /* dont inherit across fork */
> > +#define MADV_DOCOPY 0x31 /* do inherit across fork */
> >
> > /* compatibility flags */
> > #define MAP_ANON MAP_ANONYMOUS
>
> I think that's probably a good idea, to choose a range away from the rest.
> But I'm not quite sure: anyone familiar with adding APIs listening?
>
> Hugh
>

My reason was to make it possible to have identical values for all
architectures, making userspace portability easier.

--
MST
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2005-11-10 14:36    [W:0.128 / U:0.136 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site