This is the mail archive of the
gcc-help@gcc.gnu.org
mailing list for the GCC project.
Re: Is "-march=atom" deprecated?
- From: Mason <slash dot tmp at free dot fr>
- To: waltdnes at waltdnes dot org
- Cc: GCC help <gcc-help at gcc dot gnu dot org>
- Date: Mon, 3 Oct 2016 11:06:13 +0200
- Subject: Re: Is "-march=atom" deprecated?
- Authentication-results: sourceware.org; auth=none
- References: <20160928231454.GA21018@waltdnes.org> <8610e4a4-0e60-1633-f245-3f8051473487@redhat.com> <20160930030411.GA31976@waltdnes.org>
On 30/09/2016 05:04, waltdnes@waltdnes.org wrote:
> On Thu, Sep 29, 2016 at 10:21:46AM +0100, Andrew Haley wrote
>
>> That sounds fine. Intel have used the "Atom" trademark to refer to
>> Bonnell, Silvermont, and Goldmont processors. We like to be more
>> specific.
>
> Thanks. For future reference, what does "-march=atom" do? Do I get
> a "bonnell" build, or generic, or something else?
As far as I can tell, the -march=atom option was documented until gcc 4.8
https://gcc.gnu.org/onlinedocs/gcc-4.8.4/gcc/i386-and-x86-64-Options.html
'atom'
Intel Atom CPU with 64-bit extensions, MOVBE, MMX, SSE, SSE2, SSE3 and SSSE3 instruction set support.
Then they changed the doc for gcc 4.9
'bonnell'
Intel Bonnell CPU with 64-bit extensions, MOVBE, MMX, SSE, SSE2, SSE3 and SSSE3 instruction set support.
'silvermont'
Intel Silvermont CPU with 64-bit extensions, MOVBE, MMX, SSE, SSE2, SSE3, SSSE3, SSE4.1, SSE4.2, POPCNT, AES, PCLMUL and RDRND instruction set support.
It does look like 'atom' and 'bonnell' are similar.
I don't have 4.9 handy, try this:
echo | gcc -xc -S -fverbose-asm -march=atom -o v1.s -
echo | gcc -xc -S -fverbose-asm -march=bonnell -o v2.s -
diff -u v1.s v2.s
Maybe they are identical?
Regards.