Bruce, On Thu, Feb 13, 2014 at 01:52:07AM +1100, Bruce Evans wrote: B> > --- head/lib/libkvm/kvm_getpcpu.3 Wed Feb 12 08:04:38 2014 (r261795) B> > +++ head/lib/libkvm/kvm_getpcpu.3 Wed Feb 12 09:41:17 2014 (r261796) B> > @@ -50,7 +50,7 @@ B> > .Ft void * B> > .Fn kvm_getpcpu "kvm_t *kd" "int cpu" B> > .Ft ssize_t B> > -.Fn kvm_read_zpcpu "kvm_t *kd" "void *buf" "u_long base" "size_t size" "int cpu" B> > +.Fn kvm_read_zpcpu "kvm_t *kd" "u_long base" "void *buf" "size_t size" "int cpu" B> B> This doesn't fix the documentation saying to use the namespace pollution for B> the changed function... B> B> > .Ft uint64_t B> > .Fn kvm_counter_u64_fetch "kvm_t *kd" "u_long base" B> B> ...or for other functions. B> B> This bug was missing in both the header and the man page for all of the B> older functions that use unsigned long (kvm_read(), kvm_uread() and B> kvm_write()). B> B> kvm.h otherwise depends on the full pollution of <sys/types.h> only for the B> declaration uint64_t. It should declare this itself like it does for all B> the other non-underscored typedefed types that it uses.
I'm sorry, Bruce, but my position is the same as before on this. I think that C99 types must be supported by compiler. While they are not, requiring a header that declares C99 types is okay. Typedefing a standard type manually looks for me more ugly than header requirement. -- Totus tuus, Glebius. _______________________________________________ svn-src-head@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/svn-src-head To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
