This is the mail archive of the gcc-patches@gcc.gnu.org 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: 3.4/3.5 PATCH: Switch to DWARF-2 debugging on Solaris/x86 >= 7


> For the native as, the HAVE_AS_DWARF2_DEBUG_LINE test fails and the macro
> is undefined.  Btw., the failure happens both because as doesn't
> understand the `.file 1 "conftest.s"' (instead of `.file "conftest.s"')
> syntax and due to lack of .loc support.  Therefore, gcc emits the
> .debug_line section itself.

Redundancy can't hurt :-)

> With gas, the test succeeds and gcc let's gas emit that section.  The ld
> error linking libgcc_s.so.1 happens e.g. for _muldi3.o:

Aaargh!!  The test doesn't succeed for me, even with gas 2.14.  Let's take a 
look at configure.ac:

 gcc_GAS_CHECK_FEATURE([dwarf2 debug_line support],
  gcc_cv_as_dwarf2_debug_line,
  [elf,2,11,0],, [$conftest_s],
  [# ??? This fails with non-gnu grep.  Maybe use objdump?
   if grep debug_line conftest.o > /dev/null 2>&1; then
     gcc_cv_as_dwarf2_debug_line=yes
   fi])

Indeed, grep is Sun's on the box.

Now I'm a bit worried: does your problem occur on the 3.4 branch or does it 
somehow relate to the var-tracking stuff?  Could you file a PR for it?

Meanwhile, the ASM_DEBUG_SPEC is correctly emitted:

*asm_debug:
%{gstabs*:--gstabs}%{!gstabs*:%{g*:--gdwarf2}}

so I've checked in the patch.

Thanks.

-- 
Eric Botcazou


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