Front page | perl.perl5.porters |
Postings from April 2012
On Tue, Apr 24, 2012 at 15:22, Nicholas Clark <nick@ccl4.org> wrote: > On Mon, Apr 23, 2012 at 01:39:49PM +0200, Leon Timmermans wrote: >> On Sun, Apr 22, 2012 at 12:05 PM, H.Merijn Brand <h.m.brand@xs4all.nl> wrote: >> > I never thought of additional digits when allowing that. I could extend >> > the current behavior to do so if p5p thinks that is sane. >> >> If you do that, don't forget to allow -ggdb3 and similar too too. > > Seems sane to me. At least, I can't think of anything it might be confused > with. > > Is there a "list" of "similar too"? Or do we assume that anything > /^-g[a-z0-9][-+a-z0-9]*$/ is valid? I don't think so, but as I recall it's possible that you need -gdwarf-2 -g3. Also, primary problem with instructions in INSTALL is that those doesn't cover all possible cases. I wanted to compile variant with -DDEBUGGING set and -gLevel adjusted to get macros expansion in GDB. Even if you extended -DDEBUGGING option with -gXXX support then it's still open question on how to set -DDEBUGGING for perl itself. May be the following is better: -Aoptimize=-g3 -DEBUGGING It probably works and you don't have to repeate -O3 flag in optimize. If optimize option is properly described then there is no need in -DEBUGGING=-g and =both variants. For backwards compatibility -DDEBUGGING may add -g to optimize unless there -g* and -UDEBUGGING or =none variant can delete all -g* flags from optimize. > (ie is there a counter-example of a compiler flag starting -g that isn't > about debugging, and would be confusing to add here) > > Nicholas Clark -- Best regards, Ruslan.Thread Previous