Re: Many bugfixes in staging

Etienne Gagnon wrote:
> I would like to release 1.11 tomorrow, if possible, as the new code
> solves important problems that have been annoying our users for a long
> time.  I just don't want it to show important regressions, so your
> cooperation to test it today is highly appreciated.

I can't test anything before this weekend.  It would be interesting to 
see if this thread initialization stuff fixed the SMP problems, but I 
don't see it as urgently needing to be done for 1.11.

 > Chris, I am willing to appoint you as "backport maintainer", if you
 > accept the job.  (I personally have no time to dedicate to this.)

Rather than accepting some new hat, how about this: if I do end up 
starting to grab only bugfixes for my own needs, after I update to 1.11 
and get that all sorted out, rather than just merging them into my 
sandbox, I will create a shared branch of 1.11 and merge them into that 
first.  However, as you know, I am also away the entire summer...

> PS:  It's not fun debugging when gdb (or binutils, or something else)
> fails to show a full back trace.  It seems there's something broken on
> Debian-PowerPC. :(

gdb doesn't find line number info for me either, although it partially 
depends on configure options and architecture ... I don't think I've 
ever seen a backtrace on x86_64, and I wasted a large number of hours 
trying to get one last week.  I was told on #debian that libc6-dbg might 
help if debugging is not working.

Chris