This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: frame pointer elimination
- From: Aurelien Buhrig <aurelien dot buhrig dot gcc at gmail dot com>
- To: Eric Botcazou <ebotcazou at adacore dot com>
- Cc: gcc at gcc dot gnu dot org
- Date: Tue, 28 Jun 2016 12:01:03 +0200
- Subject: Re: frame pointer elimination
- Authentication-results: sourceware.org; auth=none
- References: <a29472d4-ccbe-d09a-4c9e-e3bb3fec7797 at gmail dot com> <1780439 dot 17DqMqHDWh at polaris>
>> I'm porting a private backend from an old 4.6 branch to the 6.1.0
>> release, and I have some troubles eliminating my frame pointer without
>> -fomit-frame-pointer option.
>> Elimination is correctly done with -fomit-frame-pointer.
> If the target's default is -fno-omit-frame-pointer, then it's as expected.
Ok, thanks for your reply. The 4.6 backend was testing if the hardware
FP was live, and did some optimization whatever -fomit-frame-pointer or
-fno-omit-frame-pointer. So it was not a valid optimization?
With the 6.1 branch, the hardware frame pointer is always live when
no-omit-frame-pointer, preventing such an optimization when the hardware
fp is not used...
Thanks,
Aurélien