On Thu, 2008-05-22 at 01:40 +1000, Daniel X Moore wrote:
> nscott@xxxxxxxxxx wrote:
> >> >> I came across an oddity in pmnscomp.c (the binary PMNS
> >> >> compiler) - the -v option contains conditional code on
> >> >> the BUILDTOOL macro, which conditional makes version 2
> >> >> available. This is not enabled anywhere, probably was
> >> >> on IRIX. Version 2 looks to be similar to version 1,
> >> >> except has added a checksum.
> >> >>
> >> >> Anyone know what the history is here? Why aren't we
> >> >> enabling version 2 everywhere?
> >> Wasn't it something Mr. X Moore did as his first project?
> >
> > CC'd - Daniel?
>
> Sheesh - talk about stretching my memory!
>
> IIRC, I wrote a modified version of the archive format... we either
> didn't check it in or didn't enable it because despite being more
> compact, it took a bit more CPU and Ken didn't want to make the trade-off.
>
Hmm, did you add a checksum on the binary PMNS at the same time
as archives? (this is pmnscomp.c & libpcp/src/pmns.c that I'm
banging on about here) ... not sure a PMNS checksum would add
much CPU overhead, as its changed relatively infrequently.
I hate to ask, after the (ongoing) pain the last such request
seems to have invoked, but can someone "over there" do a p_rlog
on pmnscomp.c (and/or src/libpcp/src/pmns.c) and see when the
v2 PMNS checksum code was added (and maybe fwd on the checkin
message?).
thanks!
--
Nathan
|