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 E8BF23858C98; Thu,  4 Apr 2024 19:03:50 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org E8BF23858C98
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org;
	s=default; t=1712257430;
	bh=FnuJtl9CtVhwKTwCTm85anN/dMTN77yr40NwF/IZcDc=;
	h=From:To:Subject:Date:In-Reply-To:References:From;
	b=sLVgNAAjz/JdcGTJqoUW3gGB/EHTM5xc6LaHt1mrI8g2Rj+6Zvyj6W3utWdbjNn58
	 cWPENliEeDnn3/jEPuUR1uyZDSOf3L/F6FRZaR/MxQbbEpKz5me7XLRoJrR7+KNPY7
	 FuTIB4GRRJwv4Ovrx0eQDYeR7bGI7cG3P6u7QZ2U=
From: "pinskia at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug target/114591] rtl-reload introduce an extra load operation
 since gcc-12
Date: Thu, 04 Apr 2024 19:03:50 +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: 13.2.0
X-Bugzilla-Keywords: missed-optimization, ra
X-Bugzilla-Severity: normal
X-Bugzilla-Who: pinskia at gcc dot gnu.org
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: cf_reconfirmed_on cf_gcctarget everconfirmed
 bug_status component
Message-ID: <bug-114591-4-o2PD0FPjDD@http.gcc.gnu.org/bugzilla/>
In-Reply-To: <bug-114591-4@http.gcc.gnu.org/bugzilla/>
References: <bug-114591-4@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=3D114591

Andrew Pinski <pinskia at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Last reconfirmed|                            |2024-04-04
             Target|                            |x86_64
     Ever confirmed|0                           |1
             Status|UNCONFIRMED                 |NEW
          Component|rtl-optimization            |target

--- Comment #1 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
IRA in GCC 12+ has:
```
  Loop 0 (parent -1, header bb2, depth 0)
    bbs: 2
    all: 0r85 1r82
    modified regnos: 82 85
    border:
    Pressure: GENERAL_REGS=3D1
    Hard reg set forest:
      0:( 0-6 8-15 20-51)@0
        1:( 0-6 36-43)@26000
      Spill a1(r82,l0)
      Allocno a0r85 of GENERAL_REGS(15) has 15 avail. regs  0-6 36-43, node=
:=20
0-6 36-43 (confl regs =3D  7-35 44-75)
      Forming thread from colorable bucket:
        Forming thread by copy 0:a0r85-a1r82 (freq=3D1000):
          Result (freq=3D5000): a0r85(2000) a1r82(3000)
      Pushing a0(r85,l0)(cost 0)
      Popping a0(r85,l0)  --         assign reg 0
Disposition:
    1:r82  l0   mem    0:r85  l0     0
New iteration of spill/restore move
+++Costs: overall 1000, reg -1000, mem 2000, ld 0, st 0, move 0
+++       move loops 0, new jumps 0
```

While before it was:
```
  Loop 0 (parent -1, header bb2, depth 0)
    bbs: 2
    all: 0r85 1r82
    modified regnos: 82 85
    border:
    Pressure: GENERAL_REGS=3D1
    Hard reg set forest:
      0:( 0-6 8-15 20-51)@0
        1:( 0-6 36-43)@46000
      Allocno a0r85 of GENERAL_REGS(15) has 15 avail. regs  0-6 36-43, node=
:=20
0-6 36-43 (confl regs =3D  7-35 44-75)
      Allocno a1r82 of GENERAL_REGS(15) has 15 avail. regs  0-6 36-43, node=
:=20
0-6 36-43 (confl regs =3D  7-35 44-75)
      Forming thread from colorable bucket:
        Forming thread by copy 0:a0r85-a1r82 (freq=3D1000):
          Result (freq=3D5000): a0r85(2000) a1r82(3000)
      Pushing a0(r85,l0)(cost 0)
      Pushing a1(r82,l0)(cost 0)
      Popping a1(r82,l0)  --         assign reg 0
      Popping a0(r85,l0)  --         assign reg 0
Disposition:
    1:r82  l0     0    0:r85  l0     0
New iteration of spill/restore move
+++Costs: overall 5000, reg 5000, mem 0, ld 0, st 0, move 0
+++       move loops 0, new jumps 0
```

Notice: `      Spill a1(r82,l0)` in GCC 12+=