Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 14 Aug 1995 12:12:18 +0100 (BST)
From:      Paul Richards <paul@netcraft.co.uk>
To:        peter@haywire.DIALix.COM (Peter Wemm)
Cc:        bugs@freefall.FreeBSD.org
Subject:   Re: List of open Problem Reports
Message-ID:  <199508141112.MAA00702@server.netcraft.co.uk>
In-Reply-To: <199508121508.XAA17827@haywire.DIALix.COM> from "Peter Wemm" at Aug 12, 95 11:08:30 pm

next in thread | previous in thread | raw e-mail | index | archive | help
In reply to Peter Wemm who said
> 
> In freebsd.bugs, GNU GNATS writes:
> [...]
> 
> >[1995/06/12] bin/506       "Release" id generated by send-pr looks incorrect.
> >[1995/07/10] gnu/608       `send-pr' inserts wrong default for ">Release:" fiel
> >[1995/07/14] bin/613       send-pr reports wrong release
> 
> [...]
> 
> >[1995/07/27] bin/636       /usr/share/examples Makefiles missing
> >[1995/07/27] kern/637      If used heavily, ahc will crash, with disk timeouts
> >[PDT/07/29] kern/638      Transmitted packets not passed to bpf in if_le.c
> >[PDT/07/29] misc/641      
> >[PDT/08/01] docs/646      vmstat man page out of date
> >[PDT/08/01] kern/647      Sound cards fail to work
>   ^^^
> 
> Looks like the new (gnew? :-) GNATS is picking the wrong field out of
> a `date` somewhere.. 
> 

Hmm, yes, this is one my list of gnats problems. I don't think this
is serious enough at the moment drop back from the beta code but if
people think so I will. I'll manually change these to 1995 when I get the
chance.

> BTW: the make rules for send-pr are still broken.  obj/send-pr is
> getting left around for a *long* time and is not being rebuilt when it
> should.  IMHO, it'd be better (although still not right) to get it to
> read `uname -rsm` at runtime, rather than sed'ing it in at make time.
> 
> Perhaps another way would be to *force* send-pr to be rebuilt every
> single time, so there is never a stale copy in gnu/send-pr/obj.
> 
> Or.. perhaps send-pr needs to read `uname -rsm` for "kern" reports,
> but use the compiled in name for the other problem categories.. in
> case the kernel has been supped/rebuilt later, but the other binaries
> have been unmodified...
> 
> [[ It's this very same bug which caused 2.0.5-RELEASE to go out with a
> send-pr tag of "2.0-BUILT-nnnnnn".. There was a stale copy of send-pr
> left from before Jordan/Rod/whoever changes newvers.sh, and send-pr
> was not rebuilt, and was installed from the stale, several days old
> version. ]]

This all needs addressing but I'm not going to look at it until 3 things
happen:

1) The next release of GNATS is officially released. The repository has
   to change a lot for this since all of gnats will now need to be brought
   in so all the remote tools get distributed.
2) The cvs vendor branch issue gets resolved so I can put the next
   release on a venor branch.
3) We finally agree on how our version numbering is going to be handled.

-- 
  Paul Richards, Bluebird Computer Systems. FreeBSD core team member. 
  Internet: paul@FreeBSD.org, http://www.freebsd.org/~paul
  Phone: 0370 462071 (Mobile), +44 1222 457651 (home)



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199508141112.MAA00702>