Re: [Patch V2] BaseTools: support the NOOPT target with the GCC tool chains

[Ard Biesheuvel <ard.biesheuvel@linaro.org>]

On 6 October 2016 at 09:19, Laszlo Ersek <lersek@redhat.com> wrote:
> On 10/06/16 00:39, Bruce Cran wrote:
>> On 10/04/2016 07:30 PM, Yonghong Zhu wrote:
>>
>>> Update the tools_def.template to add NOOPT support with GCC tool chains.
>>>
>>> Cc: Liming Gao <liming.gao@intel.com>
>>> Cc: Laszlo Ersek <lersek@redhat.com>
>>> Contributed-under: TianoCore Contribution Agreement 1.0
>>> Signed-off-by: Yonghong Zhu <yonghong.zhu@intel.com>
>>
>> Reviewed-by: Bruce Cran <bruce.cran@gmail.com>
>> Tested-by: Bruce Cran <bruce.cran@gmail.com>
>>
>> Tested with both GCC49 and GCC5 toolchain settings (using gcc 6.2.1 for
>> both) and verified that both have functional debugging, though gdb skips
>> around with GCC5 as expected due to LTCG.
>>
>
> Right; as I mentioned up-thread, I wonder if we should disable -flto and -Os in NOOPT_GCC5_IA32_DLINK_FLAGS and NOOPT_GCC5_X64_DLINK_FLAGS.
>
> How about we add the following to Yonghong's v2?
> - first, move "-flto" from GCC5_IA32_X64_DLINK_FLAGS and GCC5_X64_DLINK_FLAGS to the users of those macros (both GCC5- and CLANG38-related users exist),
> - second, for the GCC5 users of these macros: split them into DEBUG/RELEASE/NOOPT, and remove -Os from NOOPT.
>
> The end result is that none of the earlier macro values change, except for NOOPT_GCC5_(IA32|X64)_DLINK_FLAGS; those two lose both -flto and -Os, which is our purpose. This would eliminate the skipping around that you mention.
>
> What do you guys think? See the patch below (again, to be applied on top of Yonghong's v2, or to be squashed into it).
>

I agree. If we are going through the trouble of having a separate
NOOPT flavor, it should at least do what it says on the tin. If it
skips around, using it has no advantage over using DEBUG