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 C84483858D28; Tue, 21 Feb 2023 00:54:49 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org C84483858D28
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org;
	s=default; t=1676940889;
	bh=79M8y8UbzYMYCCcQXTn7hO891cHzBaqSiFL3r6bmNv4=;
	h=From:To:Subject:Date:In-Reply-To:References:From;
	b=bLJib4fw+yONfdfNYhvrcSMz5JBHeNIYOaQRWzWzL1P++9lxMgiXkbWTU8tzdAEhZ
	 XrEVidItVMV/zHee3jZSPl4uUiuaUTg1wfWOICzr9j5wQkeOhBuj9X/aQJyLweWohR
	 gkB2kMA5NTayJcmkgy2bnop91OHWL3WBzgZ9/HgE=
From: "crazylht at gmail dot com" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug target/108804] missed vectorization in presence of conversion
 from uint64_t to float
Date: Tue, 21 Feb 2023 00:54:47 +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: 12.2.1
X-Bugzilla-Keywords: missed-optimization
X-Bugzilla-Severity: enhancement
X-Bugzilla-Who: crazylht at gmail 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: ---
X-Bugzilla-Flags: 
X-Bugzilla-Changed-Fields: 
Message-ID: <bug-108804-4-4qQ5leQKR4@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-108804-4@http.gcc.gnu.org/bugzilla/>
References: <bug-108804-4@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=3D108804

--- Comment #3 from Hongtao.liu <crazylht at gmail dot com> ---
I think the point here is: although it's unit64_t -> float, but the range o=
f x
and y can be represent as int32(k & 0x007FFFFF) |  0x3F800000), so we can u=
se
int32 -> float instructions which are supported by the backend.

So it looks to me a middle-end issue.

A simple testcase clang generates vcvtdq2ps but gcc doesn't vectorize.

#include<stdint.h>

uint64_t d[512];
float f[1024];

void foo() {
    for (int i=3D0; i<512; ++i) {
        uint64_t k =3D d[i];
        f[i]=3D(k & 0x3F30FFFF);
    }
}

manually add convertion then gcc also can do vectorization.

#include<stdint.h>

uint64_t d[512];
float f[1024];

void foo() {
    for (int i=3D0; i<512; ++i) {
        uint64_t k =3D d[i];
        f[i]=3D(int)(k & 0x3F30FFFF);
    }
}=