svn commit: r253662 - in head: lib/libc/gen sys/vm tools/tools/sysdoc

[Juli Mallett]

On Fri, Jul 26, 2013 at 2:26 PM, Andrey Zonov <zont at freebsd.org> wrote:

> On 7/26/13 8:09 AM, Jilles Tjoelker wrote:
> > On Fri, Jul 26, 2013 at 02:00:07AM +0000, Andrey Zonov wrote:
> >> Author: zont
> >> Date: Fri Jul 26 02:00:06 2013
> >> New Revision: 253662
> >> URL: http://svnweb.freebsd.org/changeset/base/253662
> >
> >> Log:
> >>   Remove define and documentation for vm_pageout_algorithm missed in
> r253587
> >
> >> Modified:
> >>   head/lib/libc/gen/sysctl.3
> >>   head/sys/vm/vm_param.h
> >>   head/tools/tools/sysdoc/tunables.mdoc
> >
> >> Modified: head/sys/vm/vm_param.h
> >>
> ==============================================================================
> >> --- head/sys/vm/vm_param.h   Fri Jul 26 00:28:19 2013        (r253661)
> >> +++ head/sys/vm/vm_param.h   Fri Jul 26 02:00:06 2013        (r253662)
> >> @@ -82,9 +82,8 @@
> >>  #define     VM_V_CACHE_MIN          7       /* cnt.v_cache_min */
> >>  #define     VM_V_CACHE_MAX          8       /* cnt.v_cache_max */
> >>  #define VM_V_PAGEOUT_FREE_MIN       9       /* cnt.v_pageout_free_min
> */
> >> -#define     VM_PAGEOUT_ALGORITHM    10      /* pageout algorithm */
> >> -#define VM_SWAPPING_ENABLED 11      /* swapping enabled */
> >> -#define     VM_MAXID                12      /* number of valid vm ids
> */
> >> +#define VM_SWAPPING_ENABLED 10      /* swapping enabled */
> >> +#define     VM_MAXID                11      /* number of valid vm ids
> */
> >
> > As noted in mail from Bruce Evans, please preserve the ABI of
> > VM_SWAPPING_ENABLED here.
> >
>
> I don't think it's a big problem.  Internally we don't use it.  Good
> code uses sysctlbyname() instead of sysctl().  It doesn't seem to me
> this is very popular sysctl.  It's also CURRENT without MFC.


That doesn't seem like a great argument.

Why should we break the ABI for bad code that used that sysctl and all
following?  It seems pretty trivial to preserve, and no-cost.  Also, the
fact that it's in current misses the point completely: it breaks ABI with
older binaries, namely older binaries running on -CURRENT.  And not just
for that field.