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 7CF503857C4B; Wed,  6 Mar 2024 14:58:00 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 7CF503857C4B
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org;
	s=default; t=1709737080;
	bh=zwGU0Y1glHXrL50UqtSTT37Sg0c9+eNMNLcQzf093fQ=;
	h=From:To:Subject:Date:In-Reply-To:References:From;
	b=RAT8Ii2XB8oUW+vXFtX75dyh4F6ov69CqM7pwoZuQIsSUtfGAX2yrKZnj+P3U62b8
	 yvuiy/1l0ekZ0wPK6xLyu0tlufUcz0dBNJgLrGozKst3pgWgfDZab3zkfQPDRnXwdC
	 +VCI7XpJITivBEEbJQ9RlXUV9dycdSfrgLWnCeJM=
From: "amacleod at redhat dot com" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug tree-optimization/114151] [14 Regression] weird and inefficient
 codegen and addressing modes since r14-9193
Date: Wed, 06 Mar 2024 14:57:58 +0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: changed
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: tree-optimization
X-Bugzilla-Version: 14.0
X-Bugzilla-Keywords: missed-optimization
X-Bugzilla-Severity: normal
X-Bugzilla-Who: amacleod at redhat 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-114151-4-57r6N3kd9V@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-114151-4@http.gcc.gnu.org/bugzilla/>
References: <bug-114151-4@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=3D114151

--- Comment #12 from Andrew Macleod <amacleod at redhat dot com> ---
(In reply to Richard Biener from comment #11)
> (In reply to Richard Biener from comment #10)
> > (In reply to Andrew Macleod from comment #9)
> > > Created attachment 57620 [details]
> > > proposed patch
> > >=20
> > > Does this solve your problem if there is an active ranger?  it bootst=
raps
> > > with no regressions
> >=20
> > I'll check what it does.
>=20

>=20
> So the important part is that it got the fact that _12 is positive.  As
> analyzed in earlier comments I think that's all we can do, we don't know
> anything about the other variable involved and thus can't avoid the
> unsigned punning during SCEV analysis.

Yeah, I wouldn't want to invoke any dynamic lookup changes at this point. t=
hat
would be too hard to predict or contain.    I will continue poking at what =
is
triggering the loop issues because I think its a good longer term solution =
to
have range_of_expr with no context to invoke range_of_stmt if the DEF is in=
 the
IL and has not been processed.=20

>=20
> I think it's a good change, let's keep it queued for stage1 at this point
> unless we really know a case it helps to avoid a regression with
> r14-9193-ga0b1798042d033
>=20
> For testing, what's the "easiest" pass/thing to do to recompute global
> ranges now?  In the past I'd schedule EVRP but is there now a ranger
> API to do this?  Just to see if full global range compute before IVOPTs
> would help.

all VRP passes are the same now. so just schedule EVRP.   in theory, you co=
uld
schedule the fast vrp pass I added, but its not heavily tested... but you c=
ould
try it.  It doesnt do any back edges or switches (iirc), but does basic
calculations in DOM order and exports/updates globals.

NEXT_PASS (pass_fast_vrp)=