Re: Fix 70278 (LRA split_regs followup patch)

[Christophe Lyon <christophe.lyon@linaro.org>]

On 22 March 2016 at 13:14, Bernd Schmidt <bschmidt@redhat.com> wrote:
> On 03/22/2016 10:24 AM, Christophe Lyon wrote:
>>
>>
>> The ARM test isn't sufficiently protected against non-compliant
>> configurations,
>> and fails if GCC is configured for arm*linux-gnueabihf for instance
>> (see
>> http://people.linaro.org/~christophe.lyon/cross-validation/gcc/trunk/234342/report-build-info.html)
>>
>> The attached small patch fixes that by requiring arm_arch_v4t_multilib
>> effective target.
>>
>> I used arm_arch_v4t_multilib instead of arm_arch_v4t because, as I
>> reported a long time ago
>> the later does not complain in some unsupported configuration because
>> the sample effective
>> target test does not contain actual code. In particular it's not
>> sufficient to reject thumb-1 with
>> hard-float.
>>
>> OK?
>
>
> No objections from me, but I copied all this from the existing testcase
> ftest-armv4t-thumb.c, so I'm puzzled why that one doesn't fail.
>

It's similar to why I tried to explain above: ftest-armv4t-thumb.c contains
only preprocessor tests, no actual code.

When the program contains code (or even a single global variable definition),
the compiler complains that" Thumb-1 hard-float VFP ABI" is not
implemented.

A long time ago, I submitted a patch to add some code to the
arm_arch_FUNC_ok effective target, but it was not accepted.

Christophe.

>
> Bernd