This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: output dwarf2 unwind info via asm directives
Richard Henderson wrote:
Andreas Tobler wrote:
for the record, I use a pretty old toolchain, with gcc 3.4.6 and
binutils 2.15.92 on linux x86. And it fails with unknown pseudo-op
.cfi_personality.
Ok, will update for that next.
Like so. I'll continue working on the C/C++ difference thing.
r~
* configure.ac (HAVE_GAS_CFI_DIRECTIVE): Check .cfi_personality.
* configure: Rebuild.
--- configure (revision 138743)
+++ configure (local)
@@ -20851,6 +20851,7 @@ fi
.cfi_same_value 1
.cfi_def_cfa 1, 2
.cfi_escape 1, 2, 3, 4, 5
+ .cfi_personality 0, symbol
.cfi_endproc' > conftest.s
if { ac_try='$gcc_cv_as -o conftest.o conftest.s >&5'
{ (eval echo "$as_me:$LINENO: \"$ac_try\"") >&5
--- configure.ac (revision 138743)
+++ configure.ac (local)
@@ -2189,6 +2189,7 @@ gcc_GAS_CHECK_FEATURE([cfi directives],
.cfi_same_value 1
.cfi_def_cfa 1, 2
.cfi_escape 1, 2, 3, 4, 5
+ .cfi_personality 0, symbol
.cfi_endproc])
AC_DEFINE_UNQUOTED(HAVE_GAS_CFI_DIRECTIVE,
[`if test $gcc_cv_as_cfi_directive = yes; then echo 1; else echo 0;
fi`],