[Bug 1769] sparc64 not using compiler optimizations

[bugzilla-daemon at rtems.org]

https://www.rtems.org/bugzilla/show_bug.cgi?id=1769

--- Comment #6 from Gedare <giddyup44 at yahoo.com> 2011-03-21 14:17:36 CDT ---
(In reply to comment #5)
> (In reply to comment #4)
> > (In reply to comment #3)
> > > (In reply to comment #2)
> > > > (In reply to comment #1)
> > > > > I am OK with the -O2 part, but I am not OK with -gstabs.
> > > > > 
> > > > > Why do you need -gstabs? -g is supposed to be sufficient.
> > > > 
> > > > With -gstabs, I get C source code line matching in my (non-GDB)
> > > non-GDB  == your personal use-case == not of relevance to the public.
> > > 
> > AFAIW, our research group is the current users of the sparc64 port. It runs on
> > the Simics simulator, which is not GDB and not free (we have academic licensing
> > to use it).
> Non-free source, not an RTEMS tool == irrelevant.
> 
Fair enough.

> > > > So it is mostly a matter of convenience.
> > > Please try using sparc-rtems-gdb (without -gstabs).
> > Please note, this patch is for the sparc64 target,
> I know, sorry - I was referring to sparc64.
> 
> > which is not equivalent to
> > sparc32==sparc.  I don't think there is a gdb that supports sparc64 binaries
> > that use the full (privileged) instruction set and RTEMS.
> AFAICT, sparc64-gcc is supposed to generate dwarf-2 debug infos. I don't know
> however if it actually works.
> 
I never had luck using sparc64-gdb, but I didn't spend much time trying either.
Supposedly gdb can be used for remote debugging with the M5 sim (which is open
source, and can run the niagara BSP).

> > Anyway, I don't care if the -gstabs is changed to -g or even removed entirely.
> > It is easy enough to change back by hand if debugging or source-code line
> > debugging are desired.
> I will remove the -gstabs.
OK.

-- 
Configure bugmail: https://www.rtems.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are watching all bug changes.