This is the mail archive of the mailing list for the GCC project.

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [PATCH] Fix recent debug.exp failures

> Date: Fri, 12 Jun 2009 20:12:23 +0200
> From: Jakub Jelinek <>

> On Fri, Jun 12, 2009 at 06:38:20PM +0200, Hans-Peter Nilsson wrote:
> > A -gdwarf-21 is currently not the same as -gdwarf-2 -g1, as a
> > regtest of e.g. r148404 (before your change) together with
> > <> will
> > show.  For example just C:
> That's because -gdwarf-2 -g1 as separate option no longer matches the
> "*1" check on the first argument.

Ah I completely missed the "set doit 0" pruning.  ...which also
explains why it sort-of-worked when swapping the order to -g1
-gdwarf-2. :)

> 	PR testsuite/40426
> 	* lib/gcc-dg.exp (gcc-dg-debug-runtest): For type -gdwarf-2 and
> 	level != "" use separate -gdwarf-2 -g${level} options instead of
> 	-gdwarf-2${level}.
> 	* lib/gfortran-dg.exp (gfortran-dg-debug-runtest): Likewise.
> 	* gfortran.dg/debug/pr37738.f: Also skip if -gdwarf-2 -g1.
> 	* gfortran.dg/debug/pr35154-dwarf2.f: Likewise.

Looks good to me, not that I can approve it but I hope someone
does, soon.

brgds, H-P

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]