[libitm] Check if GCC uses assembler cfi support

The Solaris 11.2 x86 assembler is going to gain support for the cfi
directives.  To do things right this time, it will create read-only
.eh_frame sections, unlike what gas and gcc currently do.  The goal is
to let /bin/ld handle the required merging of read-only and read-write
.eh_frame sections, but that's not there yet and won't make the S11.2
release.  For the time being, gcc/configure.ac (gcc_cv_as_cfi_directive)
takes care of that and disables the support, but there's one issue left:
libitm does the cfi directive detection on its own, finds the assembler
support and uses it in config/x86/sjlj.S, which breaks linking the
32-bit libitm.so:

ld: fatal: file .libs/aatree.o; section [15].eh_frame and file .libs/sjlj.o; section [3].eh_frame have incompatibile attributes and cannot be merged into a single output section
collect2: error: ld returned 1 exit status
make[4]: *** [libitm.la] Error 1

To avoid this, libitm and gcc need to agree whether or not to use
assembler cfi support.

The following patch avoid this by checking __GCC_HAVE_DWARF2_CFI_ASM.

Alternatively, one could do away with
config/asmcfi.m4 (GCC_AS_CFI_PSEUDO_OP), and only check for
__GCC_HAVE_DWARF2_CFI_ASM under the assumption that libitm will only be
used in-tree with a matching gcc.

I've avoided changing config/asmcfi.m4 since that's also used by libffi,
which is external to the gcc tree and can be used with non-gcc
compilers.

The patch allowed i386-pc-solaris2.11 bootstraps to finish without
regressions on trunk and 4.8 branch, still need to test 4.7 branch.

Ok for trunk, 4.8 and 4.7 branches once testing completes?

	Rainer


2014-04-07  Rainer Orth  <ro@CeBiTec.Uni-Bielefeld.DE>

	* config/generic/asmcfi.h: Also check for
	__GCC_HAVE_DWARF2_CFI_ASM.

On 04/07/2014 04:33 AM, Rainer Orth wrote:
> The patch allowed i386-pc-solaris2.11 bootstraps to finish without
> regressions on trunk and 4.8 branch, still need to test 4.7 branch.
> 
> Ok for trunk, 4.8 and 4.7 branches once testing completes?
> 
> 	Rainer
> 
> 
> 2014-04-07  Rainer Orth  <ro@CeBiTec.Uni-Bielefeld.DE>
> 
> 	* config/generic/asmcfi.h: Also check for
> 	__GCC_HAVE_DWARF2_CFI_ASM.

Looks fine.


r~

Richard Henderson <rth@redhat.com> writes:

> On 04/07/2014 04:33 AM, Rainer Orth wrote:
>> The patch allowed i386-pc-solaris2.11 bootstraps to finish without
>> regressions on trunk and 4.8 branch, still need to test 4.7 branch.
>> 
>> Ok for trunk, 4.8 and 4.7 branches once testing completes?
>> 
>> 	Rainer
>> 
>> 
>> 2014-04-07  Rainer Orth  <ro@CeBiTec.Uni-Bielefeld.DE>
>> 
>> 	* config/generic/asmcfi.h: Also check for
>> 	__GCC_HAVE_DWARF2_CFI_ASM.
>
> Looks fine.

Testing was fine on i386-pc-solaris2.11 with /bin/as on trunk, 4.8 and
4.7 branches, as well as x86_64-unknown-linux-gnu.

However, i386-pc-solaris2.11 with gas was confusing at first: for the
32-bit PIC case, no .eh_frame was generated.  It turned out this happens
because in this case config/i386/sol2.h (ASM_PREFERRED_EH_DATA_FORMAT)
is a lie: it claims datarel encoding for the 32-bit PIC case.  This way,
dwarf2out_do_cfi_asm returns false and c-family/c-cppbuiltin.c
(c_cpp_builtins) doesn't define __GCC_HAVE_DWARF2_CFI_ASM.

In fact, gas with cfi directives uses pcrel encoding, thus
__GCC_HAVE_DWARF2_CFI_ASM should be defined.

ISTM that the ASM_PREFERRED_EH_DATA_FORMAT redefinition above should be
wrapped in HAVE_AS_IX86_DIFF_SECT_DELTA, but I'm quite reluctant to make
such a change so late in the 4.9 release cycle.

I'll certainly try it post-4.9, especially given that Solaris 9 support
will be removed by then and current Solaris 10 and 11 assemblers *can*
subtract symbols in different sections.

Unless I'm missing something, I'm thus going to check in the patch as is
within a day.  I believe that losing unwind info in a single case here
is worth avoiding the bootstrap failure.

	Rainer