From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: by sourceware.org (Postfix, from userid 48) id EFB123858C52; Fri, 26 Jan 2024 03:17:38 +0000 (GMT) DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org EFB123858C52 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org; s=default; t=1706239058; bh=RKP6eMC22dAzLBXNfBwH+rkqq5S/EL8YfOCiysH66S4=; h=From:To:Subject:Date:From; b=NoJ5EtOvf75sstJbBGDGtUx55aCk65lnOPxFH5hPif8k/R5gF7Ys2wAtS0+BYtlFE H4ZSOPMasn8ska+/rA0pDDgn6feW82fLDbzgjLZ3SM41u1G1VKhqRPHObRsyQi2o/7 OQuGoo5i3/3E7qi+KzWBzteLjAkPUjV2Pl2txpic= From: "liuhongt at gcc dot gnu.org" To: gcc-bugs@gcc.gnu.org Subject: [Bug target/113609] New: EQ/NE comparison between avx512 kmask and -1 can be optimized with kxortest with checking CF. Date: Fri, 26 Jan 2024 03:17:38 +0000 X-Bugzilla-Reason: CC X-Bugzilla-Type: new X-Bugzilla-Watch-Reason: None X-Bugzilla-Product: gcc X-Bugzilla-Component: target X-Bugzilla-Version: 14.0 X-Bugzilla-Keywords: X-Bugzilla-Severity: normal X-Bugzilla-Who: liuhongt at gcc dot gnu.org X-Bugzilla-Status: UNCONFIRMED 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: bug_id short_desc product version bug_status bug_severity priority component assigned_to reporter target_milestone Message-ID: 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: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=3D113609 Bug ID: 113609 Summary: EQ/NE comparison between avx512 kmask and -1 can be optimized with kxortest with checking CF. Product: gcc Version: 14.0 Status: UNCONFIRMED Severity: normal Priority: P3 Component: target Assignee: unassigned at gcc dot gnu.org Reporter: liuhongt at gcc dot gnu.org Target Milestone: --- It's from PR113576, there's code like=20 35 kmovb %k0, %edx 36 cmpb $-1, %dl 37 jne .L21 The original codegen is buggy, but still exposed a optimization issue. Compare 8/16/32/64-bit kmask to -1 and check equal or not can be optimized = with kortest by check CF. KORTESTW =C2=B6 TMP[15:0] :=3D DEST[15:0] BITWISE OR SRC[15:0] IF(TMP[15:0]=3D0) THEN ZF :=3D 1 ELSE ZF :=3D 0 FI; IF(TMP[15:0]=3DFFFFh) THEN CF :=3D 1 ELSE CF :=3D 0 FI; KORTESTB =C2=B6 TMP[7:0] :=3D DEST[7:0] BITWISE OR SRC[7:0] IF(TMP[7:0]=3D0) THEN ZF :=3D 1 ELSE ZF :=3D 0 FI; IF(TMP[7:0]=3D=3DFFh) THEN CF :=3D 1 ELSE CF :=3D 0 FI; KORTESTQ =C2=B6 TMP[63:0] :=3D DEST[63:0] BITWISE OR SRC[63:0] IF(TMP[63:0]=3D0) THEN ZF :=3D 1 ELSE ZF :=3D 0 FI; IF(TMP[63:0]=3D=3DFFFFFFFF_FFFFFFFFh) THEN CF :=3D 1 ELSE CF :=3D 0 FI; KORTESTD =C2=B6 TMP[31:0] :=3D DEST[31:0] BITWISE OR SRC[31:0] IF(TMP[31:0]=3D0) THEN ZF :=3D 1 ELSE ZF :=3D 0 FI; IF(TMP[31:0]=3DFFFFFFFFh) THEN CF :=3D 1 ELSE CF :=3D 0 FI;=