This is the mail archive of the
mailing list for the GCC project.
Re: RFC: Turn on -fomit-frame-pointer and -fasynchronous-unwind-tables for Linux/i386
- From: "H.J. Lu" <hjl dot tools at gmail dot com>
- To: Chris Lattner <clattner at apple dot com>
- Cc: gcc-patches at gcc dot gnu dot org, Uros Bizjak <ubizjak at gmail dot com>
- Date: Thu, 15 Jul 2010 11:16:19 -0700
- Subject: Re: RFC: Turn on -fomit-frame-pointer and -fasynchronous-unwind-tables for Linux/i386
- References: <20100715174246.GA20583@intel.com> <B1EFE802-2B3F-4B20-B313-19C49C4D496D@apple.com>
On Thu, Jul 15, 2010 at 11:08 AM, Chris Lattner <firstname.lastname@example.org> wrote:
> On Jul 15, 2010, at 10:42 AM, H.J. Lu wrote:
>> This patch turns on -fomit-frame-pointer and -fasynchronous-unwind-tables
>> for Linux/i386. ?Tested on Linux/ia32 and Linux/x86-64. ?I am also
>> enclosing a spreadsheet of comparison of shared library segment sizes
>> in gcc. ?Segment size differences range from -1% to 30%. The smaller
>> the DSO size is, the bigger its size increases. Overall, the size
>> difference is close to 0%. This comparison may not be typical since
>> C++ and Java libraries have .eh_frame sections anyway. ?Any comments?
> Does glibc backtrace() look at eh_frame on i386? ?I suspect that most unwinders don't. ?This is a borderline ABI change.