From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <gcc-bugzilla@gcc.gnu.org>
Received: by sourceware.org (Postfix, from userid 48)
	id C6C0F3858436; Wed, 21 Dec 2022 09:47:46 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org C6C0F3858436
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org;
	s=default; t=1671616066;
	bh=O1zw8KgGNN7kJry1bs4JU0Yy5HtzWwqK7/0iLfEBwxY=;
	h=From:To:Subject:Date:In-Reply-To:References:From;
	b=WqIQtxek1AozFiJV4ZjvKJwVO6yHe5IFTUqKyMYjmPAFec4SXH/sWSHJkTpnVSCxg
	 A+/EeHLpzZUGGp8fEmILrZ4yBWd5fgpWAggUhMbzTInta8RX3xkrr9DK6WjU36jjWQ
	 x2aHrquf6rthebUEV9lRjORyQSTFXBzPRjW7yYBs=
From: "jakub at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug target/108191] Add support to usage of *intrin.h without
 -mavx512f -mavx512cd
Date: Wed, 21 Dec 2022 09:47:46 +0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: changed
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: target
X-Bugzilla-Version: 13.0
X-Bugzilla-Keywords: 
X-Bugzilla-Severity: enhancement
X-Bugzilla-Who: jakub at gcc dot gnu.org
X-Bugzilla-Status: RESOLVED
X-Bugzilla-Resolution: WONTFIX
X-Bugzilla-Priority: P3
X-Bugzilla-Assigned-To: unassigned at gcc dot gnu.org
X-Bugzilla-Target-Milestone: ---
X-Bugzilla-Flags: 
X-Bugzilla-Changed-Fields: 
Message-ID: <bug-108191-4-z6i7yCsiNj@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-108191-4@http.gcc.gnu.org/bugzilla/>
References: <bug-108191-4@http.gcc.gnu.org/bugzilla/>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
X-Bugzilla-URL: http://gcc.gnu.org/bugzilla/
Auto-Submitted: auto-generated
MIME-Version: 1.0
List-Id: <gcc-bugs.sourceware.org>

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=3D108191

--- Comment #7 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
(In reply to =E7=BD=97=E5=8B=87=E5=88=9A(Yonggang Luo) from comment #6)
> Is the following command are valid usage? It's compiled properly

No, this is invalid.
>=20
> ```
>=20
> // compile args:  -fPIC -O2 -D__SSE3__=3D1 -D__SSSE3__=3D1 -D__SSE4_1__=
=3D1
> -D__SSE4_2__=3D1 -D__SSE4A__=3D1 -D__POPCNT__=3D1 -D__XSAVE__=3D1 -D__CRC=
32__=3D1
> -D__AVX__=3D1 -D__AVX2__=3D1 -D__FP_FAST_FMAF32=3D1 -D__FP_FAST_FMAF64=3D1
> -D__FP_FAST_FMAF=3D1 -D__FP_FAST_FMAF32x=3D1 -D__AVX512F__=3D1 -D__AVX512=
CD__=3D1

Only -fPIC -O2 here, none of the -D arguments, all of them are internal
GCC macros that shouldn't be redefined by users.
Plus it isn't needed.

> #include <math.h>
>=20
> #pragma GCC push_options
> #pragma GCC target("avx512f")
> #pragma GCC target("avx512cd")
> #pragma GCC target("sse4a")
>=20
> #if defined(_MSC_VER)
> #include <intrin.h>
> #else
> #include <x86intrin.h>
> #endif
>=20
> #pragma GCC pop_options

You can do it, but for GCC it is completely useless, you can just
#include <x86intrin.h> without anything further.

> #pragma GCC push_options
> #pragma GCC target("avx512f")
> #pragma GCC target("avx512cd")
> #pragma GCC target("sse4a")

This is certainly fine, but avx512f in there isn't needed, that is implied =
by
avx512cd.
Though, I don't see anything avx512cd nor sse4a-ish in there.
>=20
> void util_fadd_512(float *a, float *b, float *c) {
>     /* a =3D b + c */
>     __m512 av =3D _mm512_load_ps(a);
>     __m512 bv =3D _mm512_load_ps(b);
>     __m512 cv =3D _mm512_add_ps(av, bv);
>     _mm512_store_ps(c, cv);
> }
> static inline int
> util_iround(float f)
> {
>    __m128 m =3D _mm_set_ss(f);
>    return _mm_cvtss_i32(m);
> }
>=20
> #pragma GCC pop_options
>=20
> int util_iround_outside(int x, float y) {
>     return x + util_iround(y);
> }
> float util_fadd(float a, float b) {
>    return a + b;
> }
> ```

That said, code with avx512cd etc. target won't inline into code without it=
.=