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 907343858D39; Tue, 16 Apr 2024 21:54:17 +0000 (GMT)
DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 907343858D39
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gcc.gnu.org;
	s=default; t=1713304457;
	bh=ySkHkrNORkXmEpsrlKqbBvIV6TF78gIedQGnjsA+9N4=;
	h=From:To:Subject:Date:From;
	b=nQPdge7eqDLqXfiBduJRAtDHi00ZEqYSL2gvEZ+uvZVS+kDgsrOlN6vG3nFsbCJ9+
	 HuInfWkK7UippR0L9j2zZGO8if0XpINPoS+ntpHPFlt6TL1APh4LJ6g9Yf/vs0XZhp
	 QcbgQmVz9hi8S8L2CVJDiNmhtXe+a+0+thjSJe8Y=
From: "wojciech_mula at poczta dot onet.pl" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug c++/114747] New: [RISC-V RVV] Wrong SEW set for mixed-size
 intrinsics
Date: Tue, 16 Apr 2024 21:54:17 +0000
X-Bugzilla-Reason: CC
X-Bugzilla-Type: new
X-Bugzilla-Watch-Reason: None
X-Bugzilla-Product: gcc
X-Bugzilla-Component: c++
X-Bugzilla-Version: 13.2.0
X-Bugzilla-Keywords: 
X-Bugzilla-Severity: normal
X-Bugzilla-Who: wojciech_mula at poczta dot onet.pl
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: <bug-114747-4@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=3D114747

            Bug ID: 114747
           Summary: [RISC-V RVV] Wrong SEW set for mixed-size intrinsics
           Product: gcc
           Version: 13.2.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: c++
          Assignee: unassigned at gcc dot gnu.org
          Reporter: wojciech_mula at poczta dot onet.pl
  Target Milestone: ---

This is a distilled procedure from simdutf project:

---
#include <cstdint>
#include <cstdlib>
#include <riscv_vector.h>

size_t convert_latin1_to_utf16le(const char *src, size_t len, char16_t *dst=
) {
  char16_t *beg =3D dst;
  for (size_t vl; len > 0; len -=3D vl, src +=3D vl, dst +=3D vl) {
    vl =3D __riscv_vsetvl_e8m4(len);
    vuint8m4_t v =3D __riscv_vle8_v_u8m4((uint8_t*)src, vl);
    __riscv_vse16_v_u16m8((uint16_t*)dst, __riscv_vzext_vf2_u16m8(v, vl), v=
l);
  }
  return dst - beg;
}
---

When compiled with gcc 13.2.0 with flags "-march=3Drv64gcv -O2" it sets a w=
rong
SEW:

---
convert_latin1_to_utf16le(char const*, unsigned long, char16_t*):
        beq     a1,zero,.L4
        mv      a4,a2
.L3:
        vsetvli a5,a1,e8,m4,ta,ma  # set SEW=3D8
        vle8.v  v8,0(a0)
        slli    a3,a5,1
        vzext.vf2       v24,v8     # illegal instruction, as SEW/2 < 8
        sub     a1,a1,a5
        vse16.v v24,0(a4)
        add     a0,a0,a5
        add     a4,a4,a3
        bne     a1,zero,.L3
        sub     a0,a4,a2
        srai    a0,a0,1
        ret
.L4:
        li      a0,0
        ret
---

The trunk available on godbold.org (riscv64-unknown-linux-gnu-g++ 14.0.1
20240415) emits vsetvli with e16 argument, which seems to be fine.=