[PATCH] Don't build 32-bit libatomic with -march=i486 on x86-64

[H.J. Lu]

On Tue, Apr 19, 2016 at 11:30 AM, Uros Bizjak <ubizjak@gmail.com> wrote:
> On Tue, Apr 19, 2016 at 8:24 PM, H.J. Lu <hjl.tools@gmail.com> wrote:
>> On Tue, Apr 19, 2016 at 11:18 AM, Uros Bizjak <ubizjak@gmail.com> wrote:
>>> On Tue, Apr 19, 2016 at 8:08 PM, H.J. Lu <hjl.tools@gmail.com> wrote:
>>>> On Tue, Apr 19, 2016 at 8:45 AM, Uros Bizjak <ubizjak@gmail.com> wrote:
>>>>> On Tue, Apr 19, 2016 at 5:07 PM, H.J. Lu <hongjiu.lu@intel.com> wrote:
>>>>>> Gcc uses the same -march= for both -m32 and -m64 on x86-64 unless
>>>>>> --with-arch-32= is used.  There is no need for -march=i486 to compile
>>>>>> 32-bit libatomic on x86-64.
>>>>>>
>>>>>> Tested on x86-64.  OK for trunk?
>>>>>>
>>>>>> H.J.
>>>>>> ---
>>>>>>         PR target/70454
>>>>>>         * configure.tgt (XCFLAGS): Don't add -march=i486 to compile
>>>>>>         32-bit x86 target library on x86-64.
>>>>>> ---
>>>>>>  libatomic/configure.tgt | 10 ++--------
>>>>>>  1 file changed, 2 insertions(+), 8 deletions(-)
>>>>>>
>>>>>> diff --git a/libatomic/configure.tgt b/libatomic/configure.tgt
>>>>>> index c5470d7..bbb93fc 100644
>>>>>> --- a/libatomic/configure.tgt
>>>>>> +++ b/libatomic/configure.tgt
>>>>>> @@ -81,14 +81,8 @@ case "${target_cpu}" in
>>>>>>         try_ifunc=yes
>>>>>>         ;;
>>>>>>    x86_64)
>>>>>> -       case " ${CC} ${CFLAGS} " in
>>>>>> -         *" -m32 "*)
>>>>>> -           XCFLAGS="${XCFLAGS} -march=i486 -mtune=generic"
>>>>>> -           XCFLAGS="${XCFLAGS} -fomit-frame-pointer"
>>>>>> -           ;;
>>>>>> -         *)
>>>>>> -           ;;
>>>>>> -       esac
>>>>>> +       # Since 64-bit arch > i486, we can use the same -march= to build
>>>>>> +       # both 32-bit and 64-bit target libraries.
>>>>>>         ARCH=x86
>>>>>>         # ??? Detect when -mcx16 is already enabled.
>>>>>>         try_ifunc=yes
>>>>>> --
>>>>>> 2.5.5
>>>>>>
>>>>>
>>>>> No, this is wrong. My build with default options defaults to i386. So,
>>>>> the difference between
>>>>>
>>>>
>>>> How was your GCC configured?  Did you use
>>>>
>>>> --with-arch_32=i386
>>>
>>> Nope, just:
>>>
>>> ~/gcc-svn/trunk/configure
>>>
>>
>> $ /ssd/uros/gcc-build/gcc/cc1 -E -dM -m32 hello.c > aaa
>>
>> I don't think cc1 is supposed to be used directly.  Can you use gcc
>> driver instead, like
>>
>> $ /ssd/uros/gcc-build/gcc/xgcc -B/ssd/uros/gcc-build/gcc/  -E -dM -m32 hello.c
>
> This works, since the driver passes:
>
> COLLECT_GCC_OPTIONS='-B' '/ssd/uros/gcc-build/gcc/' '-E' '-dM' '-m32'
> '-mtune=generic' '-march=x86-64'
>

That is why I submitted my patches.  Since -m32 passes -march=x86-64
to cc1 on x86-64,  we shouldn't pass -march=i486 to cc1.  It is undesirable
especially when --with-arch= is used.  I noticed the issue when 32-bit
libatomic/libgomp/libitm weren't optimized with -march=haswell when GCC
was configured with --with-arch=haswell


-- 
H.J.