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 D76D83858D32; Thu, 28 Dec 2023 04:51:27 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org D76D83858D32
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org;
	s=default; t=1703739087;
	bh=J1kZ0C0+GU2JeUHqGQFmKrkJQMyata4cLEEPSm3pQJs=;
	h=From:To:Subject:Date:In-Reply-To:References:From;
	b=iqUxnccGY9ZOqB0c5Ezv6LhqajvfAWvdCcZzdm4JEmoLM501IUrGIKvIDjyBxIHA7
	 ltViKFLGT+Fr0NP75qOX0Q6JS+JROcy18MpqGisVwEwX+gOhqjdExy34hK8hvsPBWw
	 chifF/xIT1AJ1oJ+YvQj9776F2FjEaB45fIoUQeo=
From: "juzhe.zhong at rivai dot ai" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug c/113134] gcc does not version loops with early break
 conditions that don't have side-effects
Date: Thu, 28 Dec 2023 04:51:27 +0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: changed
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: c
X-Bugzilla-Version: 14.0
X-Bugzilla-Keywords: missed-optimization
X-Bugzilla-Severity: normal
X-Bugzilla-Who: juzhe.zhong at rivai dot ai
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-113134-4-WrhdShNtUk@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-113134-4@http.gcc.gnu.org/bugzilla/>
References: <bug-113134-4@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=3D113134

--- Comment #14 from JuzheZhong <juzhe.zhong at rivai dot ai> ---
(In reply to Tamar Christina from comment #13)
> (In reply to JuzheZhong from comment #12)
> > (In reply to Tamar Christina from comment #11)
> > > (In reply to JuzheZhong from comment #10)
> > > > (In reply to Tamar Christina from comment #9)
> > > > > (In reply to JuzheZhong from comment #8)
> > > > > > Suppose the loop mask is generated by whilelo instruction of AR=
M SVE.
> > > > > >=20
> > > > > > Suppose we have 8 elements in a single whole vector.
> > > > > >=20
> > > > > > mask =3D whilo (0, res) if res =3D 6, then mask =3D 11111000.
> > > > > > data =3D 12345678
> > > > > >=20
> > > > > > Then if it is early break. You are reversing both data and mask=
 as follows:
> > > > > >=20
> > > > > > new_mask =3D 00011111
> > > > > > new_data =3D 87654321
> > > > > >=20
> > > > > > Then use the EXTRACT_LAST, we will get value =3D 1 for early br=
eak.
> > > > > >=20
> > > > > > Am I right ?
> > > > >=20
> > > > > Yeah, the idea being the scalar loop will then run from 1 to 6 to=
 do any
> > > > > side effects that we couldn't apply.
> > > > >=20
> > > > > We went with this approach first because it works for non-masked
> > > > > architectures too. In GCC-15 we'll try to implement staying entir=
ely inside
> > > > > a vector loop by splitting the mask in elements until first activ=
e and
> > > > > element from first active so we can correctly mask the operations.
> > > >=20
> > > > Ok. For the current approach. Isn't it the first element is always =
element 0
> > > > ?
> > > >=20
> > > > Since for ARM SVE loop mask is generated by whilelo instructions, i=
t always
> > > > set
> > > > mask bit from 0 to the last active element - 1.
> > >=20
> > > sure, but you can't use BIT_FIELD_REF on VLA vectors.
> >=20
> > So, for length partial vector. We can use VEC_EXTRACT with index =3D 0 =
since
> > VEC_EXTRACT optab allows VLA vectors now for length target.
>=20
> Sounds good :)

I wonder whether ARM SVE can also use this approach VEC_EXTRACT with index =
=3D 0.

I guess the only issue is that when mask =3D all zero. That is, there is no
active elements, What behavior should be here for early break ?=