Re: [RFA] convert blocks to dictionaries, phase 1, main part

[Jim Blandy <jimb@redhat.com>]

Andrew Cagney <ac131313@ges.redhat.com> writes:
> >>>> - you'll want to start compiling all targets
> >
> >>> Sure.  How do I go about doing this?  I don't have access to a wide
> >>> range of machines; are there machines at Red Hat that I can use?
> >
> >> target's not hosts :-) There is some sed in MAINTAINERS for
> >> generating the target list.  Richard E, I believe, posted a script
> >> for doing all the config's and builds.
> > Oh, right.  Thanks, that script looks useful.
> > I'm curious exactly how long it will take my machine to do all those
> > builds - it's a 4-year-old 333 MHz PII, and I don't think it would be
> > proper for me to ask the math department to buy a new machine just for
> > working on GDB - but they might all finish overnight.  If not, they'll
> > certainly finish over the weekend.
> 
> There are now only 16(?) targets.  I was originally building 50
> targets (gdb not gdb+dejagnu -- dejagnu is what takes the time) using
> a P200 and it took ~8 hours.

What's the value of building across all target architectures?  It's
not clear to me that different targets will exercise different parts
of the symtab code; making sure all the debug formats are covered
should be sufficient.

It seems to me that architecture-specific issues in the symtab code
will be pretty rare, and easily picked off as they are noticed.  It's
very hard for me to imagine broad target coverage revealing problems
in the design.  For regcache work, sure --- but for the symtab, I
don't see it.