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 92EA43858D38; Wed, 22 Nov 2023 01:54:24 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 92EA43858D38
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org;
	s=default; t=1700618064;
	bh=jz+HnsuYRQyJaII0mvjDkb80rFAKjifEOVXuBDberpA=;
	h=From:To:Subject:Date:In-Reply-To:References:From;
	b=ry/LCaUlQ0dk3Y7DhEaO0sdLqq1DlZWDAElBv/wCXk0APVBsQfmprEJ5+xtqZNLEs
	 cU8cqCdP1GeiC1IHIpQJMY/nMuwft6OeRnJK2uz863ZCN7Jfensk7rChNvUuPzHrls
	 62hfOmo7/fb+yet+X+yprYSweXf/Ub1xoqjiYNtM=
From: "haochen.jiang at intel dot com" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug target/112643] [14 regression] including x86intrin.h is broken
 for -march=native (which adds -mno-avx10.1-256 )
Date: Wed, 22 Nov 2023 01:54:22 +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: 14.0
X-Bugzilla-Keywords: build, rejects-valid
X-Bugzilla-Severity: normal
X-Bugzilla-Who: haochen.jiang at intel dot com
X-Bugzilla-Status: NEW
X-Bugzilla-Resolution: 
X-Bugzilla-Priority: P3
X-Bugzilla-Assigned-To: unassigned at gcc dot gnu.org
X-Bugzilla-Target-Milestone: 14.0
X-Bugzilla-Flags: 
X-Bugzilla-Changed-Fields: 
Message-ID: <bug-112643-4-xbVE6MuYNX@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-112643-4@http.gcc.gnu.org/bugzilla/>
References: <bug-112643-4@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=3D112643

--- Comment #21 from Haochen Jiang <haochen.jiang at intel dot com> ---
(In reply to Andrew Pinski from comment #20)
> The use of __builtin_ia32_2intersectd128 in avx512vp2intersectvlintrin.h =
has:
> #pragma GCC target("avx512vp2intersect,avx512vl,no-evex512")
>=20
> While i386-builtin.def does:
> BDESC (0, OPTION_MASK_ISA2_AVX512VP2INTERSECT, CODE_FOR_nothing,
> "__builtin_ia32_2intersectd128", IX86_BUILTIN_2INTERSECTD128, UNKNOWN, (i=
nt)
> VOID_FTYPE_PUQI_PUQI_V4SI_V4SI)

This is a known issue I figured out yesterday but should not cause the prob=
lem
since it actually relaxed conditions. It will cause ICE when caliing builti=
ns
directly.

The reason why I am not reproducing the regression seems mainly caused by t=
he
machine I built with all have AVX512, but all the CPUs mentioned here did n=
ot,
which will lead to different behavior on march and option override.

Rebuilding on AVX only machine to reproduce.=