James Carlson wrote:
> Roland Mainz writes:
> > > So, create your own project-related environment variable and use that
> > > in the makefiles.
> >
> > Umpf... that's exactly what I'd like to avoid: Create yet another
> > project-specific debugging variable. IMHO it may be nice to have one
> > switch for all projects within OS/Net...
[snip]
> I haven't really encountered a need for build-time enabling of special
> debug features that have a more substantial or system-wide support
> infrastructure than that.  I guess your mileage will vary, so if you
> see an obvious gap here, feel free to propose something more concrete.

AFAIK the "obvious gap" is a lack of a global switch which says "compile
everything with -g". I've spend a couple of days now with finding a
problem with |posix_spawn()| and this was a great sightseeing-tour
through libshell, libast, libc (needed to BFU the system), libnsl
(needed to BFU the system), libsecdb.so (skipped BFU and only use
tar|tar to clobber the file) and finally libsocket (ahhgrrlll...
xx@@@!!... ;-( ).

After THAT experience I really wish I were smarter and just compiled
everything in the tree with "-g". That would have saved me lots of pain
(the actual problem is a race-condition when a shell creates a pipe
chain and uses |posix_spawn()| to create the process group leader
(short: Debugging-horror)).

Therefore it would be nice to have a general purpose "catch all" switch
in "bldenv" which compiles the entire tree in a way that debuggers like
dbx allow source-level debugging (which doesn't work for optimized code
(well, it would be possible as seen in Unicos where the debugger allows
source-level debugging of optimized code)), e.g. use "-g" (or even "-g
-xs") and disable "-xO[0-5]" ...

----

Bye,
Roland

-- 
  __ .  . __
 (o.\ \/ /.o) [EMAIL PROTECTED]
  \__\/\/__/  MPEG specialist, C&&JAVA&&Sun&&Unix programmer
  /O /==\ O\  TEL +49 641 7950090
 (;O/ \/ \O;)
_______________________________________________
opensolaris-code mailing list
opensolaris-code@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/opensolaris-code

Reply via email to