From patchwork Thu Apr 11 10:08:10 2019 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: David Hildenbrand X-Patchwork-Id: 1083891 Return-Path: X-Original-To: incoming@patchwork.ozlabs.org Delivered-To: patchwork-incoming@bilbo.ozlabs.org Authentication-Results: ozlabs.org; spf=pass (mailfrom) smtp.mailfrom=nongnu.org (client-ip=209.51.188.17; helo=lists.gnu.org; envelope-from=qemu-devel-bounces+incoming=patchwork.ozlabs.org@nongnu.org; receiver=) Authentication-Results: ozlabs.org; dmarc=fail (p=none dis=none) header.from=redhat.com Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by ozlabs.org (Postfix) with ESMTPS id 44fyRL3JJCz9s9y for ; Thu, 11 Apr 2019 20:47:58 +1000 (AEST) Received: from localhost ([127.0.0.1]:46070 helo=lists.gnu.org) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1hEXFM-0007Uv-AB for incoming@patchwork.ozlabs.org; Thu, 11 Apr 2019 06:47:56 -0400 Received: from eggs.gnu.org ([209.51.188.92]:54895) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1hEWsa-00039p-6A for qemu-devel@nongnu.org; Thu, 11 Apr 2019 06:24:26 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1hEWdr-0007HB-86 for qemu-devel@nongnu.org; Thu, 11 Apr 2019 06:09:12 -0400 Received: from mx1.redhat.com ([209.132.183.28]:57654) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1hEWdq-0007Gp-So; Thu, 11 Apr 2019 06:09:11 -0400 Received: from smtp.corp.redhat.com (int-mx07.intmail.prod.int.phx2.redhat.com [10.5.11.22]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mx1.redhat.com (Postfix) with ESMTPS id 261C33680A; Thu, 11 Apr 2019 10:09:10 +0000 (UTC) Received: from t460s.redhat.com (unknown [10.36.118.43]) by smtp.corp.redhat.com (Postfix) with ESMTP id C511F1001E71; Thu, 11 Apr 2019 10:09:08 +0000 (UTC) From: David Hildenbrand To: qemu-devel@nongnu.org Date: Thu, 11 Apr 2019 12:08:10 +0200 Message-Id: <20190411100836.646-16-david@redhat.com> In-Reply-To: <20190411100836.646-1-david@redhat.com> References: <20190411100836.646-1-david@redhat.com> MIME-Version: 1.0 X-Scanned-By: MIMEDefang 2.84 on 10.5.11.22 X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.5.16 (mx1.redhat.com [10.5.110.30]); Thu, 11 Apr 2019 10:09:10 +0000 (UTC) X-detected-operating-system: by eggs.gnu.org: GNU/Linux 2.2.x-3.x [generic] X-Received-From: 209.132.183.28 Subject: [Qemu-devel] [PATCH v1 15/41] s390x/tcg: Implement VECTOR GALOIS FIELD MULTIPLY SUM (AND ACCUMULATE) X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.21 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: qemu-s390x@nongnu.org, Cornelia Huck , David Hildenbrand , Thomas Huth , Richard Henderson Errors-To: qemu-devel-bounces+incoming=patchwork.ozlabs.org@nongnu.org Sender: "Qemu-devel" A galois field multiplication in field 2 is like binary multiplication, however instead of doing ordinary binary additions, xor's are performed. So no carries are considered. Implement all variants via helpers. s390_vec_sar() and s390_vec_shr() will be reused later on. Signed-off-by: David Hildenbrand Reviewed-by: Richard Henderson --- target/s390x/helper.h | 8 ++ target/s390x/insn-data.def | 4 + target/s390x/translate_vx.inc.c | 38 ++++++++ target/s390x/vec_int_helper.c | 168 ++++++++++++++++++++++++++++++++ 4 files changed, 218 insertions(+) diff --git a/target/s390x/helper.h b/target/s390x/helper.h index 83e5070821..18a3df6b07 100644 --- a/target/s390x/helper.h +++ b/target/s390x/helper.h @@ -156,6 +156,14 @@ DEF_HELPER_FLAGS_3(gvec_vclz8, TCG_CALL_NO_RWG, void, ptr, cptr, i32) DEF_HELPER_FLAGS_3(gvec_vclz16, TCG_CALL_NO_RWG, void, ptr, cptr, i32) DEF_HELPER_FLAGS_3(gvec_vctz8, TCG_CALL_NO_RWG, void, ptr, cptr, i32) DEF_HELPER_FLAGS_3(gvec_vctz16, TCG_CALL_NO_RWG, void, ptr, cptr, i32) +DEF_HELPER_FLAGS_4(gvec_vgfm8, TCG_CALL_NO_RWG, void, ptr, cptr, cptr, i32) +DEF_HELPER_FLAGS_4(gvec_vgfm16, TCG_CALL_NO_RWG, void, ptr, cptr, cptr, i32) +DEF_HELPER_FLAGS_4(gvec_vgfm32, TCG_CALL_NO_RWG, void, ptr, cptr, cptr, i32) +DEF_HELPER_FLAGS_4(gvec_vgfm64, TCG_CALL_NO_RWG, void, ptr, cptr, cptr, i32) +DEF_HELPER_FLAGS_5(gvec_vgfma8, TCG_CALL_NO_RWG, void, ptr, cptr, cptr, cptr, i32) +DEF_HELPER_FLAGS_5(gvec_vgfma16, TCG_CALL_NO_RWG, void, ptr, cptr, cptr, cptr, i32) +DEF_HELPER_FLAGS_5(gvec_vgfma32, TCG_CALL_NO_RWG, void, ptr, cptr, cptr, cptr, i32) +DEF_HELPER_FLAGS_5(gvec_vgfma64, TCG_CALL_NO_RWG, void, ptr, cptr, cptr, cptr, i32) #ifndef CONFIG_USER_ONLY DEF_HELPER_3(servc, i32, env, i64, i64) diff --git a/target/s390x/insn-data.def b/target/s390x/insn-data.def index b8400c191a..add174b793 100644 --- a/target/s390x/insn-data.def +++ b/target/s390x/insn-data.def @@ -1090,6 +1090,10 @@ F(0xe752, VCTZ, VRR_a, V, 0, 0, 0, 0, vctz, 0, IF_VEC) /* VECTOR EXCLUSIVE OR */ F(0xe76d, VX, VRR_c, V, 0, 0, 0, 0, vx, 0, IF_VEC) +/* VECTOR GALOIS FIELD MULTIPLY SUM */ + F(0xe7b4, VGFM, VRR_c, V, 0, 0, 0, 0, vgfm, 0, IF_VEC) +/* VECTOR GALOIS FIELD MULTIPLY SUM AND ACCUMULATE */ + F(0xe7bc, VGFMA, VRR_d, V, 0, 0, 0, 0, vgfma, 0, IF_VEC) #ifndef CONFIG_USER_ONLY /* COMPARE AND SWAP AND PURGE */ diff --git a/target/s390x/translate_vx.inc.c b/target/s390x/translate_vx.inc.c index 85a0f20b90..fe36d027b6 100644 --- a/target/s390x/translate_vx.inc.c +++ b/target/s390x/translate_vx.inc.c @@ -1441,3 +1441,41 @@ static DisasJumpType op_vx(DisasContext *s, DisasOps *o) get_field(s->fields, v3)); return DISAS_NEXT; } + +static DisasJumpType op_vgfm(DisasContext *s, DisasOps *o) +{ + const uint8_t es = get_field(s->fields, m4); + static const GVecGen3 g[4] = { + { .fno = gen_helper_gvec_vgfm8, }, + { .fno = gen_helper_gvec_vgfm16, }, + { .fno = gen_helper_gvec_vgfm32, }, + { .fno = gen_helper_gvec_vgfm64, }, + }; + + if (es > ES_64) { + gen_program_exception(s, PGM_SPECIFICATION); + return DISAS_NORETURN; + } + gen_gvec_3(get_field(s->fields, v1), get_field(s->fields, v2), + get_field(s->fields, v3), &g[es]); + return DISAS_NEXT; +} + +static DisasJumpType op_vgfma(DisasContext *s, DisasOps *o) +{ + const uint8_t es = get_field(s->fields, m5); + static const GVecGen4 g[4] = { + { .fno = gen_helper_gvec_vgfma8, }, + { .fno = gen_helper_gvec_vgfma16, }, + { .fno = gen_helper_gvec_vgfma32, }, + { .fno = gen_helper_gvec_vgfma64, }, + }; + + if (es > ES_64) { + gen_program_exception(s, PGM_SPECIFICATION); + return DISAS_NORETURN; + } + gen_gvec_4(get_field(s->fields, v1), get_field(s->fields, v2), + get_field(s->fields, v3), get_field(s->fields, v4), &g[es]); + return DISAS_NEXT; +} diff --git a/target/s390x/vec_int_helper.c b/target/s390x/vec_int_helper.c index c589f92765..99dd0653f6 100644 --- a/target/s390x/vec_int_helper.c +++ b/target/s390x/vec_int_helper.c @@ -37,6 +37,60 @@ static bool s390_vec_add(S390Vector *d, const S390Vector *a, return high_carry; } +static bool s390_vec_is_zero(const S390Vector *v) +{ + return !v->doubleword[0] && !v->doubleword[1]; +} + +static void s390_vec_xor(S390Vector *res, const S390Vector *a, + const S390Vector *b) +{ + res->doubleword[0] = a->doubleword[0] ^ b->doubleword[0]; + res->doubleword[1] = a->doubleword[1] ^ b->doubleword[1]; +} + +static void s390_vec_shl(S390Vector *d, const S390Vector *a, uint64_t count) +{ + uint64_t tmp; + + g_assert(count < 128); + if (count == 0) { + d->doubleword[0] = a->doubleword[0]; + d->doubleword[1] = a->doubleword[1]; + } else if (count == 64) { + d->doubleword[0] = a->doubleword[1]; + d->doubleword[1] = 0; + } else if (count < 64) { + tmp = extract64(a->doubleword[1], 64 - count, count); + d->doubleword[1] = a->doubleword[1] << count; + d->doubleword[0] = (a->doubleword[0] << count) | tmp; + } else { + d->doubleword[0] = a->doubleword[1] << (count - 64); + d->doubleword[1] = 0; + } +} + +static void s390_vec_shr(S390Vector *d, const S390Vector *a, uint64_t count) +{ + uint64_t tmp; + + g_assert(count < 128); + if (count == 0) { + d->doubleword[0] = a->doubleword[0]; + d->doubleword[1] = a->doubleword[1]; + } else if (count == 64) { + d->doubleword[1] = a->doubleword[0]; + d->doubleword[0] = 0; + } else if (count < 64) { + tmp = a->doubleword[1] >> count; + d->doubleword[1] = deposit64(tmp, 64 - count, count, a->doubleword[0]); + d->doubleword[0] = a->doubleword[0] >> count; + } else { + d->doubleword[1] = a->doubleword[0] >> (count - 64); + d->doubleword[0] = 0; + } +} + void HELPER(gvec_vacc128)(void *v1, const void *v2, const void *v3, uint32_t desc) { @@ -121,3 +175,117 @@ void HELPER(gvec_vctz##BITS)(void *v1, const void *v2, uint32_t desc) \ } DEF_VCTZ(8) DEF_VCTZ(16) + +/* like binary multiplication, but XOR instead of addition */ +#define DEF_GALOIS_MULTIPLY(BITS, TBITS) \ +static uint##TBITS##_t galois_multiply##BITS(uint##TBITS##_t a, \ + uint##TBITS##_t b) \ +{ \ + uint##TBITS##_t res = 0; \ + \ + while (b) { \ + if (b & 0x1) { \ + res = res ^ a; \ + } \ + a = a << 1; \ + b = b >> 1; \ + } \ + return res; \ +} +DEF_GALOIS_MULTIPLY(8, 16) +DEF_GALOIS_MULTIPLY(16, 32) +DEF_GALOIS_MULTIPLY(32, 64) + +static S390Vector galois_multiply64(uint64_t a, uint64_t b) +{ + S390Vector res = {}; + S390Vector va = { + .doubleword[1] = a, + }; + S390Vector vb = { + .doubleword[1] = b, + }; + + while (!s390_vec_is_zero(&vb)) { + if (vb.doubleword[1] & 0x1) { + s390_vec_xor(&res, &res, &va); + } + s390_vec_shl(&va, &va, 1); + s390_vec_shr(&vb, &vb, 1); + } + return res; +} + +#define DEF_VGFM(BITS, TBITS) \ +void HELPER(gvec_vgfm##BITS)(void *v1, const void *v2, const void *v3, \ + uint32_t desc) \ +{ \ + int i; \ + \ + for (i = 0; i < (128 / TBITS); i++) { \ + uint##BITS##_t a = s390_vec_read_element##BITS(v2, i * 2); \ + uint##BITS##_t b = s390_vec_read_element##BITS(v3, i * 2); \ + uint##TBITS##_t d = galois_multiply##BITS(a, b); \ + \ + a = s390_vec_read_element##BITS(v2, i * 2 + 1); \ + b = s390_vec_read_element##BITS(v3, i * 2 + 1); \ + d = d ^ galois_multiply32(a, b); \ + s390_vec_write_element##TBITS(v1, i, d); \ + } \ +} +DEF_VGFM(8, 16) +DEF_VGFM(16, 32) +DEF_VGFM(32, 64) + +void HELPER(gvec_vgfm64)(void *v1, const void *v2, const void *v3, + uint32_t desc) +{ + S390Vector tmp1, tmp2; + uint64_t a, b; + + a = s390_vec_read_element64(v2, 0); + b = s390_vec_read_element64(v3, 0); + tmp1 = galois_multiply64(a, b); + a = s390_vec_read_element64(v2, 1); + b = s390_vec_read_element64(v3, 1); + tmp2 = galois_multiply64(a, b); + s390_vec_xor(v1, &tmp1, &tmp2); +} + +#define DEF_VGFMA(BITS, TBITS) \ +void HELPER(gvec_vgfma##BITS)(void *v1, const void *v2, const void *v3, \ + const void *v4, uint32_t desc) \ +{ \ + int i; \ + \ + for (i = 0; i < (128 / TBITS); i++) { \ + uint##BITS##_t a = s390_vec_read_element##BITS(v2, i * 2); \ + uint##BITS##_t b = s390_vec_read_element##BITS(v3, i * 2); \ + uint##TBITS##_t d = galois_multiply##BITS(a, b); \ + \ + a = s390_vec_read_element##BITS(v2, i * 2 + 1); \ + b = s390_vec_read_element##BITS(v3, i * 2 + 1); \ + d = d ^ galois_multiply32(a, b); \ + d = d ^ s390_vec_read_element##TBITS(v4, i); \ + s390_vec_write_element##TBITS(v1, i, d); \ + } \ +} +DEF_VGFMA(8, 16) +DEF_VGFMA(16, 32) +DEF_VGFMA(32, 64) + +void HELPER(gvec_vgfma64)(void *v1, const void *v2, const void *v3, + const void *v4, uint32_t desc) +{ + S390Vector tmp1, tmp2; + uint64_t a, b; + + a = s390_vec_read_element64(v2, 0); + b = s390_vec_read_element64(v3, 0); + tmp1 = galois_multiply64(a, b); + a = s390_vec_read_element64(v2, 1); + b = s390_vec_read_element64(v3, 1); + tmp2 = galois_multiply64(a, b); + s390_vec_xor(&tmp1, &tmp1, &tmp2); + s390_vec_xor(v1, &tmp1, v4); +}