Re: [BUGS] solaris non gcc compiler debug options - Mailing list pgsql-patches

From Kris Jurka
Subject Re: [BUGS] solaris non gcc compiler debug options
Date
Msg-id Pine.BSO.4.56.0411281645120.27887@leary.csoft.net
Whole thread Raw
In response to Re: [BUGS] solaris non gcc compiler debug options  (Bruce Momjian <pgman@candle.pha.pa.us>)
Responses Re: [BUGS] solaris non gcc compiler debug options  (Bruce Momjian <pgman@candle.pha.pa.us>)
List pgsql-patches

On Sat, 27 Nov 2004, Bruce Momjian wrote:

> Kris Jurka wrote:
> >
> > Compiling on solaris with a non gcc compiler does not correctly enable
> > debugging when --enable-debug is specified.  src/template/solaris is
> > specifying CFLAGS="-O -v" and -O overrides the -g that --enable-debug
> > adds.
>
> Good point.  Our template files should be _adding_ to CFLAGS and
> CPPFLAGS, not overriding them.  Later optimization flags on the
> command line override earlier ones, so this patch should fix the
> problem, and it exists in a few platforms.  Applied.
>

I don't believe you've fixed an actual problem.  Manually set CFLAGS are
already accounted for in configure.in (line 234) here:

# CFLAGS are selected so:
# If the user specifies something in the environment, that is used.
# else:  If the template file set something, that is used.
# else:  If the compiler is GCC, then we use -O2.
# else:  If the compiler is something else, then we use -0.

if test "$ac_env_CFLAGS_set" = set; then
  CFLAGS=$ac_env_CFLAGS_value
elif test "${CFLAGS+set}" = set; then
  : # (keep what template set)
elif test "$GCC" = yes; then
  CFLAGS="-O2"
else
  # if the user selected debug mode, don't use -O
  if test "$enable_debug" != yes; then
    CFLAGS="-O"
  fi
fi

It seems the test at the end of this (enable_debug != yes) needs to be
added to the solaris template instead of unconditionally adding -O to
CFLAGS. I was trying to debug a problem and --enable-debug wasn't really
helping until I took -O out of this src/template/solaris line.  The man
page isn't particularly clear on what "best-effort" is, but it wasn't good
enough for me.

          If you issue -g, and the optimization level is -x03 or
          lower, the compiler provides best-effort symbolic
          information with almost full optimization. Tail-call
          optimization and back-end inlining are disabled.

Kris Jurka


pgsql-patches by date:

Previous
From: Simon Riggs
Date:
Subject: Re: [DOCS] SQL conformance related patch
Next
From: Tom Lane
Date:
Subject: Re: pg_dumplo schema support