This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [RFA] Fixing ARM libstdc++-v3 compilation test failures
- To: Benjamin Kosnik <bkoz at redhat dot com>
- Subject: Re: [RFA] Fixing ARM libstdc++-v3 compilation test failures
- From: Richard Henderson <rth at redhat dot com>
- Date: Fri, 1 Jun 2001 09:58:24 -0700
- Cc: gcc-patches at gcc dot gnu dot org, nickc at redhat dot com
- References: <Pine.SOL.3.91.1010601075750.10063A-100000@taarna.cygnus.com>
On Fri, Jun 01, 2001 at 07:59:10AM -0700, Benjamin Kosnik wrote:
> The first problem was a large number of undefined labels of the form
> LLSDA<number>. It turns out that these were because the code in
> dwarf2out_begin_prologue() was assuming that
> output_function_exception_table() would always use
> current_funcdef_number as the number in the LLSDA label. This is
> not true however, if USING_SJLJ_EXCEPTIONS is true, as instead it
> uses a staticly local value held in sjlj_funcdef_number.
The bug is that dwarf2 shouldn't be emitting lsda stuff if
USING_SJLJ_EXCEPTIONS. Or emitting .eh_frame at all for
that matter.
Why don't I see this on other targets with --enable-sjlj-exceptions?
> The second problem was that the functions _Unwind_GetTextRelBase()
> and _Unwind_GetDatRelBase() are not defined.
Your sources are at least 16 days out of date, Nick.
2001-05-15 Richard Henderson <rth@redhat.com>
* unwind-sjlj.c (_Unwind_GetDataRelBase): New.
(_Unwind_GetTextRelBase): New.
r~