cvs commit: src/sys/alpha/alpha support.s src/sys/i386/i386 swtch.s src/sys/kern kern_shutdown.c src/sys/sys systm.h

[Nate Lawson]

On Tue, 17 Feb 2004, Jun Kuriyama wrote:
> At Sun, 15 Feb 2004 09:53:25 +1100 (EST),
> Bruce Evans wrote:
> > > I also think that commit is not so useful, but there are some
> > > supporters for his modification.  Do you want to back out this
> > > completely?   Or fixing just style bugs are acceptable?
> >
> > Fixing the style bugs and other bugs, and adding an option (off by
> > default) to control it would be acceptable.  Backing it out would
> > be easier.
>
> I think an option which is off by default would not be accepted by
> Poul-Henning.
>
> I re-read all of this thread and collect opinions.
>
> pls back out:	bde,nate,wpaul,danfe at nsu.ru
> good:		phk,kientzle,grog
> positive:	silby
> negative:	jhb,kuriyama
>
> This does not include private emails to Poul-Henning and I don't think
> we should decide something with above.  That's just FYI.
>
> But I personally feel this should be back out after re-reading and we
> should tackle traceback() improvements.
>
> Sorry my stupid, but it seems everyone seems happy if traceback()
> improvement happens.  Is that really too hard to go forward?

Thanks for the summary.  I already committed the backtrace() patch on Jan.
27th that makes it use printf when not in the debugger.  It was by Ian
Dowse.

http://docs.freebsd.org/cgi/getmsg.cgi?fetch=588156+0+archive/2004/cvs-all/20040208.cvs-all

-Nate