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 829D93858D37; Mon, 20 May 2024 06:48:00 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 829D93858D37
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org;
	s=default; t=1716187680;
	bh=G6aim6XnlLJwEKvIHdO/kUrJ46BXHvwB1frCUDkVtDc=;
	h=From:To:Subject:Date:In-Reply-To:References:From;
	b=vNlz81pq3EpcmAHOVlvrqJI4dQrExuYScOG1OeELka99+aCnoH3dujfoWv/JufbW2
	 8JfWsmKp7m/nl3Pd1MZDcy8UKOYLEb7lpf//f4uvNocxIvQm1aqMAA6LuvdxgqACjr
	 blI2F/zqSx1WFiyp3YMpx6A0aAcng9nwflbL5J5U=
From: "ubizjak at gmail dot com" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug target/115069] [14/15 regression] 8 bit integer vector
 performance regression, x86, between gcc-14 and gcc-13 using avx2 target
 clones on skylake platform
Date: Mon, 20 May 2024 06:47:59 +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: missed-optimization
X-Bugzilla-Severity: normal
X-Bugzilla-Who: ubizjak 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: 14.2
X-Bugzilla-Flags: 
X-Bugzilla-Changed-Fields: 
Message-ID: <bug-115069-4-tUHoBsPnxF@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-115069-4@http.gcc.gnu.org/bugzilla/>
References: <bug-115069-4@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=3D115069

--- Comment #17 from Uro=C5=A1 Bizjak <ubizjak at gmail dot com> ---
(In reply to Haochen Jiang from comment #15)
> I am doing like this way. Suppose should be same as Comment 8.

Yes, but IMO the patch in Comment #8 better describes where the problem is.

Please note that without VPMOVWB we fall-back to the original
ix86_expand_vecop_qihi, where the expansion is implemented in a different w=
ay.

(BTW: If there is a better way to emulate VPMOVWB, it should be implemented=
 in
vec-perm routines, it will universally benefit this permutation. In this ca=
se,
early exit, as introduced in the mentioned patch, could be removed.)=