svn commit: r254309 - in head: share/man/man9 sys/cddl/contrib/opensolaris/uts/common/dtrace sys/cddl/dev/dtrace sys/cddl/dev/sdt sys/kern sys/sys

Mark Johnston markj at freebsd.org
Fri Aug 16 17:40:49 UTC 2013


On Fri, Aug 16, 2013 at 07:13:16PM +0200, Davide Italiano wrote:
> [trim old mails]
> 
> > diff --git a/sys/sys/pmckern.h b/sys/sys/pmckern.h
> > index e3e18a6..90585de 100644
> > --- a/sys/sys/pmckern.h
> > +++ b/sys/sys/pmckern.h
> > @@ -51,13 +51,11 @@
> >  #define        PMC_FN_CSW_IN                   2
> >  #define        PMC_FN_CSW_OUT                  3
> >  #define        PMC_FN_DO_SAMPLES               4
> > -#define        PMC_FN_KLD_LOAD                 5
> > -#define        PMC_FN_KLD_UNLOAD               6
> > -#define        PMC_FN_MMAP                     7
> > -#define        PMC_FN_MUNMAP                   8
> > -#define        PMC_FN_USER_CALLCHAIN           9
> > -#define        PMC_FN_USER_CALLCHAIN_SOFT      10
> > -#define        PMC_FN_SOFT_SAMPLING            11
> > +#define        PMC_FN_MMAP                     5
> > +#define        PMC_FN_MUNMAP                   6
> > +#define        PMC_FN_USER_CALLCHAIN           7
> > +#define        PMC_FN_USER_CALLCHAIN_SOFT      8
> > +#define        PMC_FN_SOFT_SAMPLING            9
> >
> 
> I've skimmed over your patch quickly so I could miss something, but I
> worry about this change breaking the KBI.
> Does this make sense for you?

I think you're right. I considered this last night, but it didn't occur
to me that external modules might try to invoke these hooks. I'm not
sure if such modules exist, but it's better to be safe. I updated the
patch here:

http://people.freebsd.org/~markj/patches/hwpmc-eh/hwpmc-eh-3.diff

Thanks!
-Mark


More information about the svn-src-head mailing list