svn commit: r351729 - in head: lib/libc/gen lib/libc/sys sys/compat/freebsd32 sys/kern sys/sys

[Justin Hibbits]

On Tue, 3 Sep 2019 10:20:35 -0400
Shawn Webb <shawn.webb at hardenedbsd.org> wrote:

> On Tue, Sep 03, 2019 at 07:47:40AM -0400, Shawn Webb wrote:
> > On Tue, Sep 03, 2019 at 11:45:23AM +0000, Brooks Davis wrote:  
> > > On Tue, Sep 03, 2019 at 07:35:05AM -0400, Shawn Webb wrote:  
> > > > Hey Mateusz,
> > > > 
> > > > On Tue, Sep 03, 2019 at 04:16:31AM +0000, Mateusz Guzik wrote:  
> > > > > Author: mjg
> > > > > Date: Tue Sep  3 04:16:30 2019
> > > > > New Revision: 351729
> > > > > URL: https://svnweb.freebsd.org/changeset/base/351729
> > > > > 
> > > > > Log:
> > > > >   Add sysctlbyname system call
> > > > >   
> > > > >   Previously userspace would issue one syscall to resolve the
> > > > > sysctl and then another one to actually use it. Do it all in
> > > > > one trip. 
> > > > >   Fallback is provided in case newer libc happens to be
> > > > > running on an older kernel.
> > > > >   
> > > > >   Submitted by:	Pawel Biernacki
> > > > >   Reported by:	kib, brooks
> > > > >   Differential Revision:
> > > > > https://reviews.freebsd.org/D17282
> > > > > 
> > > > > Modified:  
> > > > ... snip ...  
> > > > >   head/sys/sys/param.h  
> > > > 
> > > > ... snip ...
> > > >   
> > > > > 
> > > > > Modified: head/sys/sys/param.h
> > > > > ==============================================================================
> > > > > --- head/sys/sys/param.h	Mon Sep  2 21:57:57
> > > > > 2019	(r351728) +++ head/sys/sys/param.h	Tue Sep
> > > > >  3 04:16:30 2019	(r351729) @@ -60,7 +60,7 @@
> > > > >   *		in the range 5 to 9.
> > > > >   */
> > > > >  #undef __FreeBSD_version
> > > > > -#define __FreeBSD_version 1300044	/* Master,
> > > > > propagated to newvers */ +#define __FreeBSD_version
> > > > > 1300045	/* Master, propagated to newvers */  
> > > > 
> > > > To an outsider, it seems that __FreeBSD_version tends to be
> > > > bumped in a separate commit. Am I remembering that right?  
> > > 
> > > It should be bumped in the same commit, but people forget or the
> > > bump they have in their review turns into a no-op because someone
> > > else does a bump in the interim (the latter has bit me several
> > > times).  
> > 
> > Interesting. Thanks for the clarification!  
> 
> One thought for making the version bump a seperate commit is if the
> original commit needed to be reverted, the commit can be reverted
> wholesale (well, from the perspective of __FreeBSD_version) without
> worry of accidentally decrementing the version number to a prior
> value.
> 
> That's my "need-more-caffeine" verbose way of saying "separating the
> version bump from the actual work allows for easier maintenance of the
> version number, helping ensure an always-increasing number."
> 
> Sorry if I sound dry here. My ten-month-old puppy is tiring me out way
> faster than I can tire him out.
> 
> Thanks,
> 

I always thought convention was separate commits to ease MFCs.  The few
times I've bumped __FreeBSD_version I've done it explicitly as a
followup commit for that reason.  Guess I now know better :)

- Justin