This is the mail archive of the
gdb-patches@sourceware.org
mailing list for the GDB project.
Re: [PATCH 1/2] Use -fno-asynchronous-unwind-tables if C program is compiled without debug info on x86
- From: Pedro Alves <palves at redhat dot com>
- To: Yao Qi <qiyaoltc at gmail dot com>, gdb-patches at sourceware dot org
- Date: Tue, 19 Apr 2016 14:35:33 +0100
- Subject: Re: [PATCH 1/2] Use -fno-asynchronous-unwind-tables if C program is compiled without debug info on x86
- Authentication-results: sourceware.org; auth=none
- References: <1461052220-10149-1-git-send-email-yao dot qi at linaro dot org> <1461052220-10149-2-git-send-email-yao dot qi at linaro dot org>
I think the subject would be clearer if it mentioned this is about gdb tests. E.g.:
Use -fno-asynchronous-unwind-tables on x86 for gdb C tests without debug info
On 04/19/2016 08:50 AM, Yao Qi wrote:
> .eh_frame is added in default on x86 and x86_64 so that DWARF unwinder
> is always used, but prologue unwinders won't be used at all. IOW,
> prologue unwinders are not tested by regression tests. If the test is
> intended to compiled *without* debug info, we shouldn't generate any
> debug info (.eh_frame for example) at all.
>
> This patch is to disable async-unwind-tables generation if the C program
> is intended to compile without debug info on x86-like target.
No sure about this. This is an ABI change on x86_64 -- the x86_64 ABI
requires eh_frame.
Should we instead add a new "nounwind" option, and a few
prologue-unwinder-specific tests?
Thanks,
Pedro Alves