From patchwork Mon Oct 25 12:28:02 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Patchwork-Submitter: =?utf-8?b?RnLDqWTDqXJpYyBQw6l0cm90?= X-Patchwork-Id: 12581749 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 7950BC433EF for ; Mon, 25 Oct 2021 12:41:20 +0000 (UTC) Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.kernel.org (Postfix) with ESMTPS id 00B3F60EE9 for ; Mon, 25 Oct 2021 12:41:19 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.4.1 mail.kernel.org 00B3F60EE9 Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=univ-grenoble-alpes.fr Authentication-Results: mail.kernel.org; spf=pass smtp.mailfrom=nongnu.org Received: from localhost ([::1]:41314 helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1mezHz-0003xt-0Q for qemu-devel@archiver.kernel.org; Mon, 25 Oct 2021 08:41:19 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]:53000) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1mez5n-0005rc-9y; Mon, 25 Oct 2021 08:28:43 -0400 Received: from zm-mta-out-3.u-ga.fr ([152.77.200.56]:42688) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1mez5g-0007vq-Ck; Mon, 25 Oct 2021 08:28:42 -0400 Received: from mailhost.u-ga.fr (mailhost1.u-ga.fr [152.77.1.10]) by zm-mta-out-3.u-ga.fr (Postfix) with ESMTP id 1BC5D418E2; Mon, 25 Oct 2021 14:28:32 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=univ-grenoble-alpes.fr; s=2020; t=1635164912; bh=X2S7vw2TvJz5d5h8dXwq8RJHUnCtvB3uA/64/3lrp8o=; h=From:To:Cc:Subject:Date:In-Reply-To:References:From; b=rsgKXlagqFrN+JW/wyKRYUpezioeTrX1BPvJulCdraKoDkWA2j64dJOTD2Jw3OEtB mQG8nat2IYlao9JpjWF/yJyWYbCPXXMYIjd2wxbquxu4vX9NhbRZdl7kGlLGQDH5A1 TT+savmXXswwTcxFYGBKa7U/5sHpsyRJ2TlGzCJYPDTqpN1r5B+/g0gS9AEpyjZEcQ zohoPmhDaTH9sKQ+j9QGyEvkyZlCa6o1jMTCpfqfqMF0DH0AO9Mt8PuN6U0Qi6cNtg oBAxqSDVxMApQw24gWzSwbVLTMtnoSxFB31CgrjuDHPISYH21noqmbmH0TsMYWCgLA oH6LXXt+9pikg== Received: from smtps.univ-grenoble-alpes.fr (smtps2.u-ga.fr [152.77.18.2]) by mailhost.u-ga.fr (Postfix) with ESMTP id EAAA160067; Mon, 25 Oct 2021 14:28:31 +0200 (CEST) Received: from palmier.tima.u-ga.fr (35.201.90.79.rev.sfr.net [79.90.201.35]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (2048 bits) server-digest SHA256) (No client certificate requested) (Authenticated sender: petrotf@univ-grenoble-alpes.fr) by smtps.univ-grenoble-alpes.fr (Postfix) with ESMTPSA id 8E54614005D; Mon, 25 Oct 2021 14:28:31 +0200 (CEST) From: =?utf-8?b?RnLDqWTDqXJpYyBQw6l0cm90?= To: qemu-devel@nongnu.org, qemu-riscv@nongnu.org Subject: [PATCH v4 01/17] exec/memop: Rename MO_Q definition as MO_UQ and add MO_UO Date: Mon, 25 Oct 2021 14:28:02 +0200 Message-Id: <20211025122818.168890-2-frederic.petrot@univ-grenoble-alpes.fr> X-Mailer: git-send-email 2.33.0 In-Reply-To: <20211025122818.168890-1-frederic.petrot@univ-grenoble-alpes.fr> References: <20211025122818.168890-1-frederic.petrot@univ-grenoble-alpes.fr> MIME-Version: 1.0 X-Greylist: Whitelist-UGA SMTP Authentifie (petrotf@univ-grenoble-alpes.fr) via submission-587 ACL (41) X-Greylist: Whitelist-UGA MAILHOST (SMTP non authentifie) depuis 152.77.18.2 Received-SPF: pass client-ip=152.77.200.56; envelope-from=frederic.petrot@univ-grenoble-alpes.fr; helo=zm-mta-out-3.u-ga.fr X-Spam_score_int: -20 X-Spam_score: -2.1 X-Spam_bar: -- X-Spam_report: (-2.1 / 5.0 requ) BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_MSPIKE_H2=-0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: bin.meng@windriver.com, richard.henderson@linaro.org, =?utf-8?q?Philippe?= =?utf-8?q?_Mathieu-Daud=C3=A9?= , alistair.francis@wdc.com, fabien.portas@grenoble-inp.org, palmer@dabbelt.com, =?utf-8?b?RnLDqWTDqXJp?= =?utf-8?b?YyBQw6l0cm90?= , philmd@redhat.com Errors-To: qemu-devel-bounces+qemu-devel=archiver.kernel.org@nongnu.org Sender: "Qemu-devel" Introducing defines for unsigned quad, signed quad, and unsigned octo access types to handle load and store by 128-bit processors. Signed-off-by: Frédéric Pétrot Reviewed-by: Philippe Mathieu-Daudé --- include/exec/memop.h | 13 ++++++++---- target/arm/translate-a32.h | 4 ++-- target/arm/translate-a64.c | 8 ++++---- target/arm/translate-neon.c | 6 +++--- target/arm/translate-sve.c | 2 +- target/arm/translate-vfp.c | 8 ++++---- target/arm/translate.c | 2 +- target/ppc/translate.c | 24 +++++++++++----------- target/sparc/translate.c | 4 ++-- target/ppc/translate/fixedpoint-impl.c.inc | 20 +++++++++--------- target/ppc/translate/fp-impl.c.inc | 4 ++-- target/ppc/translate/vsx-impl.c.inc | 4 ++-- tcg/aarch64/tcg-target.c.inc | 2 +- tcg/arm/tcg-target.c.inc | 10 ++++----- tcg/i386/tcg-target.c.inc | 4 ++-- tcg/mips/tcg-target.c.inc | 4 ++-- tcg/ppc/tcg-target.c.inc | 8 ++++---- tcg/riscv/tcg-target.c.inc | 6 +++--- tcg/s390x/tcg-target.c.inc | 10 ++++----- 19 files changed, 74 insertions(+), 69 deletions(-) diff --git a/include/exec/memop.h b/include/exec/memop.h index 04264ffd6b..53a874e61e 100644 --- a/include/exec/memop.h +++ b/include/exec/memop.h @@ -85,29 +85,34 @@ typedef enum MemOp { MO_UB = MO_8, MO_UW = MO_16, MO_UL = MO_32, + MO_UQ = MO_64, MO_SB = MO_SIGN | MO_8, MO_SW = MO_SIGN | MO_16, MO_SL = MO_SIGN | MO_32, - MO_Q = MO_64, + MO_SQ = MO_SIGN | MO_64, + MO_UO = MO_128, MO_LEUW = MO_LE | MO_UW, MO_LEUL = MO_LE | MO_UL, MO_LESW = MO_LE | MO_SW, MO_LESL = MO_LE | MO_SL, - MO_LEQ = MO_LE | MO_Q, + MO_LEQ = MO_LE | MO_UQ, MO_BEUW = MO_BE | MO_UW, MO_BEUL = MO_BE | MO_UL, MO_BESW = MO_BE | MO_SW, MO_BESL = MO_BE | MO_SL, - MO_BEQ = MO_BE | MO_Q, + MO_BEQ = MO_BE | MO_UQ, #ifdef NEED_CPU_H MO_TEUW = MO_TE | MO_UW, MO_TEUL = MO_TE | MO_UL, + MO_TEUQ = MO_TE | MO_UQ, MO_TESW = MO_TE | MO_SW, MO_TESL = MO_TE | MO_SL, - MO_TEQ = MO_TE | MO_Q, + MO_TEQ = MO_TE | MO_UQ, + MO_TESQ = MO_TE | MO_SQ, + MO_TEO = MO_TE | MO_UO, #endif MO_SSIZE = MO_SIZE | MO_SIGN, diff --git a/target/arm/translate-a32.h b/target/arm/translate-a32.h index 88f15df60e..ec0330ea0f 100644 --- a/target/arm/translate-a32.h +++ b/target/arm/translate-a32.h @@ -114,13 +114,13 @@ void gen_aa32_st_i64(DisasContext *s, TCGv_i64 val, TCGv_i32 a32, static inline void gen_aa32_ld64(DisasContext *s, TCGv_i64 val, TCGv_i32 a32, int index) { - gen_aa32_ld_i64(s, val, a32, index, MO_Q); + gen_aa32_ld_i64(s, val, a32, index, MO_UQ); } static inline void gen_aa32_st64(DisasContext *s, TCGv_i64 val, TCGv_i32 a32, int index) { - gen_aa32_st_i64(s, val, a32, index, MO_Q); + gen_aa32_st_i64(s, val, a32, index, MO_UQ); } DO_GEN_LD(8u, MO_UB) diff --git a/target/arm/translate-a64.c b/target/arm/translate-a64.c index cec672f229..1411fdfb6f 100644 --- a/target/arm/translate-a64.c +++ b/target/arm/translate-a64.c @@ -973,7 +973,7 @@ static void do_fp_st(DisasContext *s, int srcidx, TCGv_i64 tcg_addr, int size) tcg_gen_ld_i64(tmphi, cpu_env, fp_reg_hi_offset(s, srcidx)); - mop = s->be_data | MO_Q; + mop = s->be_data | MO_UQ; tcg_gen_qemu_st_i64(be ? tmphi : tmplo, tcg_addr, get_mem_index(s), mop | (s->align_mem ? MO_ALIGN_16 : 0)); tcg_gen_addi_i64(tcg_hiaddr, tcg_addr, 8); @@ -1007,7 +1007,7 @@ static void do_fp_ld(DisasContext *s, int destidx, TCGv_i64 tcg_addr, int size) tmphi = tcg_temp_new_i64(); tcg_hiaddr = tcg_temp_new_i64(); - mop = s->be_data | MO_Q; + mop = s->be_data | MO_UQ; tcg_gen_qemu_ld_i64(be ? tmphi : tmplo, tcg_addr, get_mem_index(s), mop | (s->align_mem ? MO_ALIGN_16 : 0)); tcg_gen_addi_i64(tcg_hiaddr, tcg_addr, 8); @@ -4099,10 +4099,10 @@ static void disas_ldst_tag(DisasContext *s, uint32_t insn) int i, n = (1 + is_pair) << LOG2_TAG_GRANULE; tcg_gen_qemu_st_i64(tcg_zero, clean_addr, mem_index, - MO_Q | MO_ALIGN_16); + MO_UQ | MO_ALIGN_16); for (i = 8; i < n; i += 8) { tcg_gen_addi_i64(clean_addr, clean_addr, 8); - tcg_gen_qemu_st_i64(tcg_zero, clean_addr, mem_index, MO_Q); + tcg_gen_qemu_st_i64(tcg_zero, clean_addr, mem_index, MO_UQ); } tcg_temp_free_i64(tcg_zero); } diff --git a/target/arm/translate-neon.c b/target/arm/translate-neon.c index dd43de558e..3854dd3516 100644 --- a/target/arm/translate-neon.c +++ b/target/arm/translate-neon.c @@ -73,7 +73,7 @@ static void neon_load_element64(TCGv_i64 var, int reg, int ele, MemOp mop) case MO_UL: tcg_gen_ld32u_i64(var, cpu_env, offset); break; - case MO_Q: + case MO_UQ: tcg_gen_ld_i64(var, cpu_env, offset); break; default: @@ -1830,7 +1830,7 @@ static bool do_prewiden_3d(DisasContext *s, arg_3diff *a, return false; } - if ((a->vd & 1) || (src1_mop == MO_Q && (a->vn & 1))) { + if ((a->vd & 1) || (src1_mop == MO_UQ && (a->vn & 1))) { return false; } @@ -1910,7 +1910,7 @@ static bool do_prewiden_3d(DisasContext *s, arg_3diff *a, }; \ int narrow_mop = a->size == MO_32 ? MO_32 | SIGN : -1; \ return do_prewiden_3d(s, a, widenfn[a->size], addfn[a->size], \ - SRC1WIDE ? MO_Q : narrow_mop, \ + SRC1WIDE ? MO_UQ : narrow_mop, \ narrow_mop); \ } diff --git a/target/arm/translate-sve.c b/target/arm/translate-sve.c index bc91a64171..86104b857e 100644 --- a/target/arm/translate-sve.c +++ b/target/arm/translate-sve.c @@ -5284,7 +5284,7 @@ static const MemOp dtype_mop[16] = { MO_UB, MO_UB, MO_UB, MO_UB, MO_SL, MO_UW, MO_UW, MO_UW, MO_SW, MO_SW, MO_UL, MO_UL, - MO_SB, MO_SB, MO_SB, MO_Q + MO_SB, MO_SB, MO_SB, MO_UQ }; #define dtype_msz(x) (dtype_mop[x] & MO_SIZE) diff --git a/target/arm/translate-vfp.c b/target/arm/translate-vfp.c index 59bcaec5be..17f796e32a 100644 --- a/target/arm/translate-vfp.c +++ b/target/arm/translate-vfp.c @@ -1170,11 +1170,11 @@ static bool trans_VLDR_VSTR_dp(DisasContext *s, arg_VLDR_VSTR_dp *a) addr = add_reg_for_lit(s, a->rn, offset); tmp = tcg_temp_new_i64(); if (a->l) { - gen_aa32_ld_i64(s, tmp, addr, get_mem_index(s), MO_Q | MO_ALIGN_4); + gen_aa32_ld_i64(s, tmp, addr, get_mem_index(s), MO_UQ | MO_ALIGN_4); vfp_store_reg64(tmp, a->vd); } else { vfp_load_reg64(tmp, a->vd); - gen_aa32_st_i64(s, tmp, addr, get_mem_index(s), MO_Q | MO_ALIGN_4); + gen_aa32_st_i64(s, tmp, addr, get_mem_index(s), MO_UQ | MO_ALIGN_4); } tcg_temp_free_i64(tmp); tcg_temp_free_i32(addr); @@ -1322,12 +1322,12 @@ static bool trans_VLDM_VSTM_dp(DisasContext *s, arg_VLDM_VSTM_dp *a) for (i = 0; i < n; i++) { if (a->l) { /* load */ - gen_aa32_ld_i64(s, tmp, addr, get_mem_index(s), MO_Q | MO_ALIGN_4); + gen_aa32_ld_i64(s, tmp, addr, get_mem_index(s), MO_UQ | MO_ALIGN_4); vfp_store_reg64(tmp, a->vd + i); } else { /* store */ vfp_load_reg64(tmp, a->vd + i); - gen_aa32_st_i64(s, tmp, addr, get_mem_index(s), MO_Q | MO_ALIGN_4); + gen_aa32_st_i64(s, tmp, addr, get_mem_index(s), MO_UQ | MO_ALIGN_4); } tcg_gen_addi_i32(addr, addr, offset); } diff --git a/target/arm/translate.c b/target/arm/translate.c index d6af5b1b03..0390e9d48e 100644 --- a/target/arm/translate.c +++ b/target/arm/translate.c @@ -1220,7 +1220,7 @@ void read_neon_element64(TCGv_i64 dest, int reg, int ele, MemOp memop) case MO_UL: tcg_gen_ld32u_i64(dest, cpu_env, off); break; - case MO_Q: + case MO_UQ: tcg_gen_ld_i64(dest, cpu_env, off); break; default: diff --git a/target/ppc/translate.c b/target/ppc/translate.c index 518337bcb7..9fd8f5a7a9 100644 --- a/target/ppc/translate.c +++ b/target/ppc/translate.c @@ -3230,10 +3230,10 @@ GEN_QEMU_LOAD_64(ld8u, DEF_MEMOP(MO_UB)) GEN_QEMU_LOAD_64(ld16u, DEF_MEMOP(MO_UW)) GEN_QEMU_LOAD_64(ld32u, DEF_MEMOP(MO_UL)) GEN_QEMU_LOAD_64(ld32s, DEF_MEMOP(MO_SL)) -GEN_QEMU_LOAD_64(ld64, DEF_MEMOP(MO_Q)) +GEN_QEMU_LOAD_64(ld64, DEF_MEMOP(MO_UQ)) #if defined(TARGET_PPC64) -GEN_QEMU_LOAD_64(ld64ur, BSWAP_MEMOP(MO_Q)) +GEN_QEMU_LOAD_64(ld64ur, BSWAP_MEMOP(MO_UQ)) #endif #define GEN_QEMU_STORE_TL(stop, op) \ @@ -3264,10 +3264,10 @@ static void glue(gen_qemu_, glue(stop, _i64))(DisasContext *ctx, \ GEN_QEMU_STORE_64(st8, DEF_MEMOP(MO_UB)) GEN_QEMU_STORE_64(st16, DEF_MEMOP(MO_UW)) GEN_QEMU_STORE_64(st32, DEF_MEMOP(MO_UL)) -GEN_QEMU_STORE_64(st64, DEF_MEMOP(MO_Q)) +GEN_QEMU_STORE_64(st64, DEF_MEMOP(MO_UQ)) #if defined(TARGET_PPC64) -GEN_QEMU_STORE_64(st64r, BSWAP_MEMOP(MO_Q)) +GEN_QEMU_STORE_64(st64r, BSWAP_MEMOP(MO_UQ)) #endif #define GEN_LDX_E(name, ldop, opc2, opc3, type, type2, chk) \ @@ -3304,7 +3304,7 @@ GEN_LDEPX(lb, DEF_MEMOP(MO_UB), 0x1F, 0x02) GEN_LDEPX(lh, DEF_MEMOP(MO_UW), 0x1F, 0x08) GEN_LDEPX(lw, DEF_MEMOP(MO_UL), 0x1F, 0x00) #if defined(TARGET_PPC64) -GEN_LDEPX(ld, DEF_MEMOP(MO_Q), 0x1D, 0x00) +GEN_LDEPX(ld, DEF_MEMOP(MO_UQ), 0x1D, 0x00) #endif #if defined(TARGET_PPC64) @@ -3413,7 +3413,7 @@ GEN_STEPX(stb, DEF_MEMOP(MO_UB), 0x1F, 0x06) GEN_STEPX(sth, DEF_MEMOP(MO_UW), 0x1F, 0x0C) GEN_STEPX(stw, DEF_MEMOP(MO_UL), 0x1F, 0x04) #if defined(TARGET_PPC64) -GEN_STEPX(std, DEF_MEMOP(MO_Q), 0x1d, 0x04) +GEN_STEPX(std, DEF_MEMOP(MO_UQ), 0x1d, 0x04) #endif #if defined(TARGET_PPC64) @@ -3907,7 +3907,7 @@ static void gen_lwat(DisasContext *ctx) #ifdef TARGET_PPC64 static void gen_ldat(DisasContext *ctx) { - gen_ld_atomic(ctx, DEF_MEMOP(MO_Q)); + gen_ld_atomic(ctx, DEF_MEMOP(MO_UQ)); } #endif @@ -3990,7 +3990,7 @@ static void gen_stwat(DisasContext *ctx) #ifdef TARGET_PPC64 static void gen_stdat(DisasContext *ctx) { - gen_st_atomic(ctx, DEF_MEMOP(MO_Q)); + gen_st_atomic(ctx, DEF_MEMOP(MO_UQ)); } #endif @@ -4042,9 +4042,9 @@ STCX(stwcx_, DEF_MEMOP(MO_UL)) #if defined(TARGET_PPC64) /* ldarx */ -LARX(ldarx, DEF_MEMOP(MO_Q)) +LARX(ldarx, DEF_MEMOP(MO_UQ)) /* stdcx. */ -STCX(stdcx_, DEF_MEMOP(MO_Q)) +STCX(stdcx_, DEF_MEMOP(MO_UQ)) /* lqarx */ static void gen_lqarx(DisasContext *ctx) @@ -8055,7 +8055,7 @@ GEN_LDEPX(lb, DEF_MEMOP(MO_UB), 0x1F, 0x02) GEN_LDEPX(lh, DEF_MEMOP(MO_UW), 0x1F, 0x08) GEN_LDEPX(lw, DEF_MEMOP(MO_UL), 0x1F, 0x00) #if defined(TARGET_PPC64) -GEN_LDEPX(ld, DEF_MEMOP(MO_Q), 0x1D, 0x00) +GEN_LDEPX(ld, DEF_MEMOP(MO_UQ), 0x1D, 0x00) #endif #undef GEN_STX_E @@ -8081,7 +8081,7 @@ GEN_STEPX(stb, DEF_MEMOP(MO_UB), 0x1F, 0x06) GEN_STEPX(sth, DEF_MEMOP(MO_UW), 0x1F, 0x0C) GEN_STEPX(stw, DEF_MEMOP(MO_UL), 0x1F, 0x04) #if defined(TARGET_PPC64) -GEN_STEPX(std, DEF_MEMOP(MO_Q), 0x1D, 0x04) +GEN_STEPX(std, DEF_MEMOP(MO_UQ), 0x1D, 0x04) #endif #undef GEN_CRLOGIC diff --git a/target/sparc/translate.c b/target/sparc/translate.c index fdb8bbe5dc..7dfb33f867 100644 --- a/target/sparc/translate.c +++ b/target/sparc/translate.c @@ -2830,7 +2830,7 @@ static void gen_ldda_asi(DisasContext *dc, TCGv addr, int insn, int rd) default: { TCGv_i32 r_asi = tcg_const_i32(da.asi); - TCGv_i32 r_mop = tcg_const_i32(MO_Q); + TCGv_i32 r_mop = tcg_const_i32(MO_UQ); save_state(dc); gen_helper_ld_asi(t64, cpu_env, addr, r_asi, r_mop); @@ -2886,7 +2886,7 @@ static void gen_stda_asi(DisasContext *dc, TCGv hi, TCGv addr, default: { TCGv_i32 r_asi = tcg_const_i32(da.asi); - TCGv_i32 r_mop = tcg_const_i32(MO_Q); + TCGv_i32 r_mop = tcg_const_i32(MO_UQ); save_state(dc); gen_helper_st_asi(cpu_env, addr, t64, r_asi, r_mop); diff --git a/target/ppc/translate/fixedpoint-impl.c.inc b/target/ppc/translate/fixedpoint-impl.c.inc index 2e2518ee15..33ce041d0b 100644 --- a/target/ppc/translate/fixedpoint-impl.c.inc +++ b/target/ppc/translate/fixedpoint-impl.c.inc @@ -131,11 +131,11 @@ TRANS64(LWAUX, do_ldst_X, true, false, MO_SL) TRANS64(PLWA, do_ldst_PLS_D, false, false, MO_SL) /* Load Doubleword */ -TRANS64(LD, do_ldst_D, false, false, MO_Q) -TRANS64(LDX, do_ldst_X, false, false, MO_Q) -TRANS64(LDU, do_ldst_D, true, false, MO_Q) -TRANS64(LDUX, do_ldst_X, true, false, MO_Q) -TRANS64(PLD, do_ldst_PLS_D, false, false, MO_Q) +TRANS64(LD, do_ldst_D, false, false, MO_UQ) +TRANS64(LDX, do_ldst_X, false, false, MO_UQ) +TRANS64(LDU, do_ldst_D, true, false, MO_UQ) +TRANS64(LDUX, do_ldst_X, true, false, MO_UQ) +TRANS64(PLD, do_ldst_PLS_D, false, false, MO_UQ) /* Store Byte */ TRANS(STB, do_ldst_D, false, true, MO_UB) @@ -159,11 +159,11 @@ TRANS(STWUX, do_ldst_X, true, true, MO_UL) TRANS(PSTW, do_ldst_PLS_D, false, true, MO_UL) /* Store Doubleword */ -TRANS64(STD, do_ldst_D, false, true, MO_Q) -TRANS64(STDX, do_ldst_X, false, true, MO_Q) -TRANS64(STDU, do_ldst_D, true, true, MO_Q) -TRANS64(STDUX, do_ldst_X, true, true, MO_Q) -TRANS64(PSTD, do_ldst_PLS_D, false, true, MO_Q) +TRANS64(STD, do_ldst_D, false, true, MO_UQ) +TRANS64(STDX, do_ldst_X, false, true, MO_UQ) +TRANS64(STDU, do_ldst_D, true, true, MO_UQ) +TRANS64(STDUX, do_ldst_X, true, true, MO_UQ) +TRANS64(PSTD, do_ldst_PLS_D, false, true, MO_UQ) /* * Fixed-Point Compare Instructions diff --git a/target/ppc/translate/fp-impl.c.inc b/target/ppc/translate/fp-impl.c.inc index 9f7868ee28..01b5c53bf4 100644 --- a/target/ppc/translate/fp-impl.c.inc +++ b/target/ppc/translate/fp-impl.c.inc @@ -974,7 +974,7 @@ static void gen_lfdepx(DisasContext *ctx) EA = tcg_temp_new(); t0 = tcg_temp_new_i64(); gen_addr_reg_index(ctx, EA); - tcg_gen_qemu_ld_i64(t0, EA, PPC_TLB_EPID_LOAD, DEF_MEMOP(MO_Q)); + tcg_gen_qemu_ld_i64(t0, EA, PPC_TLB_EPID_LOAD, DEF_MEMOP(MO_UQ)); set_fpr(rD(ctx->opcode), t0); tcg_temp_free(EA); tcg_temp_free_i64(t0); @@ -1210,7 +1210,7 @@ static void gen_stfdepx(DisasContext *ctx) t0 = tcg_temp_new_i64(); gen_addr_reg_index(ctx, EA); get_fpr(t0, rD(ctx->opcode)); - tcg_gen_qemu_st_i64(t0, EA, PPC_TLB_EPID_STORE, DEF_MEMOP(MO_Q)); + tcg_gen_qemu_st_i64(t0, EA, PPC_TLB_EPID_STORE, DEF_MEMOP(MO_UQ)); tcg_temp_free(EA); tcg_temp_free_i64(t0); } diff --git a/target/ppc/translate/vsx-impl.c.inc b/target/ppc/translate/vsx-impl.c.inc index 57a7f73bba..c1b1dde01c 100644 --- a/target/ppc/translate/vsx-impl.c.inc +++ b/target/ppc/translate/vsx-impl.c.inc @@ -162,8 +162,8 @@ static void gen_lxvdsx(DisasContext *ctx) gen_addr_reg_index(ctx, EA); data = tcg_temp_new_i64(); - tcg_gen_qemu_ld_i64(data, EA, ctx->mem_idx, DEF_MEMOP(MO_Q)); - tcg_gen_gvec_dup_i64(MO_Q, vsr_full_offset(xT(ctx->opcode)), 16, 16, data); + tcg_gen_qemu_ld_i64(data, EA, ctx->mem_idx, DEF_MEMOP(MO_UQ)); + tcg_gen_gvec_dup_i64(MO_UQ, vsr_full_offset(xT(ctx->opcode)), 16, 16, data); tcg_temp_free(EA); tcg_temp_free_i64(data); diff --git a/tcg/aarch64/tcg-target.c.inc b/tcg/aarch64/tcg-target.c.inc index 5edca8d44d..a8db553287 100644 --- a/tcg/aarch64/tcg-target.c.inc +++ b/tcg/aarch64/tcg-target.c.inc @@ -1744,7 +1744,7 @@ static void tcg_out_qemu_ld_direct(TCGContext *s, MemOp memop, TCGType ext, case MO_SL: tcg_out_ldst_r(s, I3312_LDRSWX, data_r, addr_r, otype, off_r); break; - case MO_Q: + case MO_UQ: tcg_out_ldst_r(s, I3312_LDRX, data_r, addr_r, otype, off_r); break; default: diff --git a/tcg/arm/tcg-target.c.inc b/tcg/arm/tcg-target.c.inc index 633b8a37ba..e31f454695 100644 --- a/tcg/arm/tcg-target.c.inc +++ b/tcg/arm/tcg-target.c.inc @@ -1443,13 +1443,13 @@ static void * const qemu_ld_helpers[MO_SSIZE + 1] = { #ifdef HOST_WORDS_BIGENDIAN [MO_UW] = helper_be_lduw_mmu, [MO_UL] = helper_be_ldul_mmu, - [MO_Q] = helper_be_ldq_mmu, + [MO_UQ] = helper_be_ldq_mmu, [MO_SW] = helper_be_ldsw_mmu, [MO_SL] = helper_be_ldul_mmu, #else [MO_UW] = helper_le_lduw_mmu, [MO_UL] = helper_le_ldul_mmu, - [MO_Q] = helper_le_ldq_mmu, + [MO_UQ] = helper_le_ldq_mmu, [MO_SW] = helper_le_ldsw_mmu, [MO_SL] = helper_le_ldul_mmu, #endif @@ -1694,7 +1694,7 @@ static bool tcg_out_qemu_ld_slow_path(TCGContext *s, TCGLabelQemuLdst *lb) default: tcg_out_mov_reg(s, COND_AL, datalo, TCG_REG_R0); break; - case MO_Q: + case MO_UQ: if (datalo != TCG_REG_R1) { tcg_out_mov_reg(s, COND_AL, datalo, TCG_REG_R0); tcg_out_mov_reg(s, COND_AL, datahi, TCG_REG_R1); @@ -1781,7 +1781,7 @@ static void tcg_out_qemu_ld_index(TCGContext *s, MemOp opc, case MO_UL: tcg_out_ld32_r(s, COND_AL, datalo, addrlo, addend); break; - case MO_Q: + case MO_UQ: /* Avoid ldrd for user-only emulation, to handle unaligned. */ if (USING_SOFTMMU && use_armv6_instructions && (datalo & 1) == 0 && datahi == datalo + 1) { @@ -1824,7 +1824,7 @@ static void tcg_out_qemu_ld_direct(TCGContext *s, MemOp opc, TCGReg datalo, case MO_UL: tcg_out_ld32_12(s, COND_AL, datalo, addrlo, 0); break; - case MO_Q: + case MO_UQ: /* Avoid ldrd for user-only emulation, to handle unaligned. */ if (USING_SOFTMMU && use_armv6_instructions && (datalo & 1) == 0 && datahi == datalo + 1) { diff --git a/tcg/i386/tcg-target.c.inc b/tcg/i386/tcg-target.c.inc index 84b109bb84..0b5d385ad6 100644 --- a/tcg/i386/tcg-target.c.inc +++ b/tcg/i386/tcg-target.c.inc @@ -1827,7 +1827,7 @@ static bool tcg_out_qemu_ld_slow_path(TCGContext *s, TCGLabelQemuLdst *l) case MO_UL: tcg_out_mov(s, TCG_TYPE_I32, data_reg, TCG_REG_EAX); break; - case MO_Q: + case MO_UQ: if (TCG_TARGET_REG_BITS == 64) { tcg_out_mov(s, TCG_TYPE_I64, data_reg, TCG_REG_RAX); } else if (data_reg == TCG_REG_EDX) { @@ -2019,7 +2019,7 @@ static void tcg_out_qemu_ld_direct(TCGContext *s, TCGReg datalo, TCGReg datahi, } break; #endif - case MO_Q: + case MO_UQ: if (TCG_TARGET_REG_BITS == 64) { tcg_out_modrm_sib_offset(s, movop + P_REXW + seg, datalo, base, index, 0, ofs); diff --git a/tcg/mips/tcg-target.c.inc b/tcg/mips/tcg-target.c.inc index d8f6914f03..15704c84fa 100644 --- a/tcg/mips/tcg-target.c.inc +++ b/tcg/mips/tcg-target.c.inc @@ -1384,7 +1384,7 @@ static void tcg_out_qemu_ld_direct(TCGContext *s, TCGReg lo, TCGReg hi, case MO_SL: tcg_out_opc_imm(s, OPC_LW, lo, base, 0); break; - case MO_Q | MO_BSWAP: + case MO_UQ | MO_BSWAP: if (TCG_TARGET_REG_BITS == 64) { if (use_mips32r2_instructions) { tcg_out_opc_imm(s, OPC_LD, lo, base, 0); @@ -1413,7 +1413,7 @@ static void tcg_out_qemu_ld_direct(TCGContext *s, TCGReg lo, TCGReg hi, tcg_out_mov(s, TCG_TYPE_I32, MIPS_BE ? hi : lo, TCG_TMP3); } break; - case MO_Q: + case MO_UQ: /* Prefer to load from offset 0 first, but allow for overlap. */ if (TCG_TARGET_REG_BITS == 64) { tcg_out_opc_imm(s, OPC_LD, lo, base, 0); diff --git a/tcg/ppc/tcg-target.c.inc b/tcg/ppc/tcg-target.c.inc index 3e4ca2be88..6802cb06a3 100644 --- a/tcg/ppc/tcg-target.c.inc +++ b/tcg/ppc/tcg-target.c.inc @@ -1935,24 +1935,24 @@ static const uint32_t qemu_ldx_opc[(MO_SSIZE + MO_BSWAP) + 1] = { [MO_UB] = LBZX, [MO_UW] = LHZX, [MO_UL] = LWZX, - [MO_Q] = LDX, + [MO_UQ] = LDX, [MO_SW] = LHAX, [MO_SL] = LWAX, [MO_BSWAP | MO_UB] = LBZX, [MO_BSWAP | MO_UW] = LHBRX, [MO_BSWAP | MO_UL] = LWBRX, - [MO_BSWAP | MO_Q] = LDBRX, + [MO_BSWAP | MO_UQ] = LDBRX, }; static const uint32_t qemu_stx_opc[(MO_SIZE + MO_BSWAP) + 1] = { [MO_UB] = STBX, [MO_UW] = STHX, [MO_UL] = STWX, - [MO_Q] = STDX, + [MO_UQ] = STDX, [MO_BSWAP | MO_UB] = STBX, [MO_BSWAP | MO_UW] = STHBRX, [MO_BSWAP | MO_UL] = STWBRX, - [MO_BSWAP | MO_Q] = STDBRX, + [MO_BSWAP | MO_UQ] = STDBRX, }; static const uint32_t qemu_exts_opc[4] = { diff --git a/tcg/riscv/tcg-target.c.inc b/tcg/riscv/tcg-target.c.inc index 9b13a46fb4..b621694321 100644 --- a/tcg/riscv/tcg-target.c.inc +++ b/tcg/riscv/tcg-target.c.inc @@ -862,7 +862,7 @@ static void * const qemu_ld_helpers[MO_SSIZE + 1] = { #if TCG_TARGET_REG_BITS == 64 [MO_SL] = helper_be_ldsl_mmu, #endif - [MO_Q] = helper_be_ldq_mmu, + [MO_UQ] = helper_be_ldq_mmu, #else [MO_UW] = helper_le_lduw_mmu, [MO_SW] = helper_le_ldsw_mmu, @@ -870,7 +870,7 @@ static void * const qemu_ld_helpers[MO_SSIZE + 1] = { #if TCG_TARGET_REG_BITS == 64 [MO_SL] = helper_le_ldsl_mmu, #endif - [MO_Q] = helper_le_ldq_mmu, + [MO_UQ] = helper_le_ldq_mmu, #endif }; @@ -1083,7 +1083,7 @@ static void tcg_out_qemu_ld_direct(TCGContext *s, TCGReg lo, TCGReg hi, case MO_SL: tcg_out_opc_imm(s, OPC_LW, lo, base, 0); break; - case MO_Q: + case MO_UQ: /* Prefer to load from offset 0 first, but allow for overlap. */ if (TCG_TARGET_REG_BITS == 64) { tcg_out_opc_imm(s, OPC_LD, lo, base, 0); diff --git a/tcg/s390x/tcg-target.c.inc b/tcg/s390x/tcg-target.c.inc index 8938c446c8..61d6694268 100644 --- a/tcg/s390x/tcg-target.c.inc +++ b/tcg/s390x/tcg-target.c.inc @@ -1745,10 +1745,10 @@ static void tcg_out_qemu_ld_direct(TCGContext *s, MemOp opc, TCGReg data, tcg_out_insn(s, RXY, LGF, data, base, index, disp); break; - case MO_Q | MO_BSWAP: + case MO_UQ | MO_BSWAP: tcg_out_insn(s, RXY, LRVG, data, base, index, disp); break; - case MO_Q: + case MO_UQ: tcg_out_insn(s, RXY, LG, data, base, index, disp); break; @@ -1791,10 +1791,10 @@ static void tcg_out_qemu_st_direct(TCGContext *s, MemOp opc, TCGReg data, } break; - case MO_Q | MO_BSWAP: + case MO_UQ | MO_BSWAP: tcg_out_insn(s, RXY, STRVG, data, base, index, disp); break; - case MO_Q: + case MO_UQ: tcg_out_insn(s, RXY, STG, data, base, index, disp); break; @@ -1928,7 +1928,7 @@ static bool tcg_out_qemu_st_slow_path(TCGContext *s, TCGLabelQemuLdst *lb) case MO_UL: tgen_ext32u(s, TCG_REG_R4, data_reg); break; - case MO_Q: + case MO_UQ: tcg_out_mov(s, TCG_TYPE_I64, TCG_REG_R4, data_reg); break; default: From patchwork Mon Oct 25 12:28:03 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Patchwork-Submitter: =?utf-8?b?RnLDqWTDqXJpYyBQw6l0cm90?= X-Patchwork-Id: 12581753 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 47E1DC433F5 for ; Mon, 25 Oct 2021 12:41:43 +0000 (UTC) Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.kernel.org (Postfix) with ESMTPS id E4A2660EE9 for ; Mon, 25 Oct 2021 12:41:42 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.4.1 mail.kernel.org E4A2660EE9 Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=univ-grenoble-alpes.fr Authentication-Results: mail.kernel.org; spf=pass smtp.mailfrom=nongnu.org Received: from localhost ([::1]:43100 helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1mezIM-0005Ev-1c for qemu-devel@archiver.kernel.org; Mon, 25 Oct 2021 08:41:42 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]:52974) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1mez5m-0005rb-7H; Mon, 25 Oct 2021 08:28:43 -0400 Received: from zm-mta-out-3.u-ga.fr ([152.77.200.56]:42712) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1mez5g-0007wE-Ba; Mon, 25 Oct 2021 08:28:41 -0400 Received: from mailhost.u-ga.fr (mailhost1.u-ga.fr [152.77.1.10]) by zm-mta-out-3.u-ga.fr (Postfix) with ESMTP id 5910941580; Mon, 25 Oct 2021 14:28:32 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=univ-grenoble-alpes.fr; s=2020; t=1635164912; bh=6FD2b+85OO7OuXJRNyfVEJmlhNDt7L2ii5QvFT7brIE=; h=From:To:Cc:Subject:Date:In-Reply-To:References:From; b=G34aFOJPA3tYuPXA37aCJxrY36/6xhfHJHNdPFekNl26l3otTHK0WaSrJlEjNLxm3 uICKpjMJneo3Zdi9FoY4a+Z8TW0E6co+/PwZQVZnacED1wF9ibELlP7UVHlRFyJ8et CZdMFhAl1mygfxFUayTP7Kc1WZy/CafS0ugytq16kJvupUbUuE1sy3LJ+nJXDzgaj5 OWAtnFV2w5Yjgr32Y+gBSI3MOMrFJmvbcohHbe/BP2d/89WEP7fA7k/1kg7tzfuXEC le8MCbtbMgNPprxcg04fLFrD1UXXwAUO3Ah+rewMHdJ7lFjNaI02ywvJO2kYE2FXIq pG3J91CdZ3RdA== Received: from smtps.univ-grenoble-alpes.fr (smtps2.u-ga.fr [152.77.18.2]) by mailhost.u-ga.fr (Postfix) with ESMTP id 3DD9260067; Mon, 25 Oct 2021 14:28:32 +0200 (CEST) Received: from palmier.tima.u-ga.fr (35.201.90.79.rev.sfr.net [79.90.201.35]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (2048 bits) server-digest SHA256) (No client certificate requested) (Authenticated sender: petrotf@univ-grenoble-alpes.fr) by smtps.univ-grenoble-alpes.fr (Postfix) with ESMTPSA id F196214005A; Mon, 25 Oct 2021 14:28:31 +0200 (CEST) From: =?utf-8?b?RnLDqWTDqXJpYyBQw6l0cm90?= To: qemu-devel@nongnu.org, qemu-riscv@nongnu.org Subject: [PATCH v4 02/17] qemu/int128: addition of a few 128-bit operations Date: Mon, 25 Oct 2021 14:28:03 +0200 Message-Id: <20211025122818.168890-3-frederic.petrot@univ-grenoble-alpes.fr> X-Mailer: git-send-email 2.33.0 In-Reply-To: <20211025122818.168890-1-frederic.petrot@univ-grenoble-alpes.fr> References: <20211025122818.168890-1-frederic.petrot@univ-grenoble-alpes.fr> MIME-Version: 1.0 X-Greylist: Whitelist-UGA SMTP Authentifie (petrotf@univ-grenoble-alpes.fr) via submission-587 ACL (41) X-Greylist: Whitelist-UGA MAILHOST (SMTP non authentifie) depuis 152.77.18.2 Received-SPF: pass client-ip=152.77.200.56; envelope-from=frederic.petrot@univ-grenoble-alpes.fr; helo=zm-mta-out-3.u-ga.fr X-Spam_score_int: -20 X-Spam_score: -2.1 X-Spam_bar: -- X-Spam_report: (-2.1 / 5.0 requ) BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_MSPIKE_H2=-0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: bin.meng@windriver.com, richard.henderson@linaro.org, alistair.francis@wdc.com, fabien.portas@grenoble-inp.org, palmer@dabbelt.com, =?utf-8?b?RnLDqWTDqXJpYyBQw6l0cm90?= , philmd@redhat.com Errors-To: qemu-devel-bounces+qemu-devel=archiver.kernel.org@nongnu.org Sender: "Qemu-devel" Addition of not, xor, div and rem on 128-bit integers, used in particular within div/rem and csr helpers for computations on 128-bit registers in the 128-bit riscv target. Signed-off-by: Frédéric Pétrot Co-authored-by: Fabien Portas --- include/qemu/int128.h | 26 +++++ util/int128.c | 218 ++++++++++++++++++++++++++++++++++++++++++ util/meson.build | 1 + 3 files changed, 245 insertions(+) create mode 100644 util/int128.c diff --git a/include/qemu/int128.h b/include/qemu/int128.h index 2ac0746426..ef41892dac 100644 --- a/include/qemu/int128.h +++ b/include/qemu/int128.h @@ -58,6 +58,11 @@ static inline Int128 int128_exts64(int64_t a) return a; } +static inline Int128 int128_not(Int128 a) +{ + return ~a; +} + static inline Int128 int128_and(Int128 a, Int128 b) { return a & b; @@ -68,6 +73,11 @@ static inline Int128 int128_or(Int128 a, Int128 b) return a | b; } +static inline Int128 int128_xor(Int128 a, Int128 b) +{ + return a ^ b; +} + static inline Int128 int128_rshift(Int128 a, int n) { return a >> n; @@ -235,6 +245,11 @@ static inline Int128 int128_exts64(int64_t a) return int128_make128(a, (a < 0) ? -1 : 0); } +static inline Int128 int128_not(Int128 a) +{ + return int128_make128(~a.lo, ~a.hi); +} + static inline Int128 int128_and(Int128 a, Int128 b) { return int128_make128(a.lo & b.lo, a.hi & b.hi); @@ -245,6 +260,11 @@ static inline Int128 int128_or(Int128 a, Int128 b) return int128_make128(a.lo | b.lo, a.hi | b.hi); } +static inline Int128 int128_xor(Int128 a, Int128 b) +{ + return int128_make128(a.lo ^ b.lo, a.hi ^ b.hi); +} + static inline Int128 int128_rshift(Int128 a, int n) { int64_t h; @@ -366,4 +386,10 @@ static inline void bswap128s(Int128 *s) *s = bswap128(*s); } +#define UINT128_MAX int128_make128(~0LL, ~0LL) +Int128 int128_divu(Int128, Int128); +Int128 int128_remu(Int128, Int128); +Int128 int128_divs(Int128, Int128); +Int128 int128_rems(Int128, Int128); + #endif /* INT128_H */ diff --git a/util/int128.c b/util/int128.c new file mode 100644 index 0000000000..035ef936f8 --- /dev/null +++ b/util/int128.c @@ -0,0 +1,218 @@ +#include "qemu/osdep.h" +#include "qemu/host-utils.h" +#include "qemu/int128.h" + +#ifndef CONFIG_INT128 + +Int128 int128_divu(Int128 a, Int128 b) +{ + return (__uint128_t)a / (__uint128_t)b; +} + +Int128 int128_remu(Int128 a, Int128 b) +{ + return (__uint128_t)a % (__uint128_t)b; +} + +Int128 int128_divs(Int128 a, Int128 b) +{ + return a / b; +} + +Int128 int128_rems(Int128 a, Int128 b) +{ + return a % b; +} + +#else +/* + * Division and remainder algorithms for 128-bit. + * Naïve implementation of Knuth Algorithm D, can be optimized quite a bit if + * it becomes a bootleneck. + * Precondition: function should never be called with v equals to 0, it has to + * be dealt with beforehand. + */ + +static inline void int128_to_uint32(uint32_t a[4], Int128 i) +{ + a[0] = int128_getlo(i) & 0xffffffff; + a[1] = int128_getlo(i) >> 32; + a[2] = int128_gethi(i) & 0xffffffff; + a[3] = int128_gethi(i) >> 32; +} + +static inline Int128 int128_from_uint32(uint32_t a[4]) +{ + return int128_make128(a[0] | (((uint64_t)a[1]) << 32), + a[2] | (((uint64_t)a[3]) << 32)); +} + +static void divrem128(Int128 uu, Int128 vv, Int128 *qq, Int128 *rr) +{ + const uint64_t b = ((uint64_t) 1) << 32; + const int m = 4; + uint64_t qhat, rhat, p; + int n = 0, s = 0, i; + int64_t j, t, k; + + /* Build arrays of 32-bit words for u and v */ + uint32_t u[5] = {[4] = 0}; + uint32_t v[4]; + + int128_to_uint32(u, uu); + int128_to_uint32(v, vv); + + uint32_t q[4] = {0}; + + if (v[3]) { + n = 4; + } else if (v[2]) { + n = 3; + } else if (v[1]) { + n = 2; + } else if (v[0]) { + n = 1; + } else { + /* function should not be called with zero as divisor */ + g_assert_not_reached(); + } + + if (n == 1) { + /* Take care of the case of a single-digit divisor here */ + k = 0; + for (j = m - 1; j >= 0; j--) { + q[j] = (k * b + u[j]) / v[0]; + k = (k * b + u[j]) - q[j] * v[0]; + } + u[0] = k; + u[1] = u[2] = u[3] = u[4] = 0; + } else { + Int128 ss; + s = clz32(v[n - 1]); /* 0 <= s <= 32 */ + if (s != 0) { + ss = int128_lshift(int128_from_uint32(v), s); + int128_to_uint32(v, ss); + ss = int128_lshift(int128_from_uint32(u), s); + /* Keep otherwise shifted out most significant byte */ + u[4] = u[3] >> (32 - s); + int128_to_uint32(u, ss); + } + + /* Step D2 : loop on j */ + for (j = m - n; j >= 0; j--) { /* Main loop */ + /* Step D3 : Compute estimate qhat of q[j] */ + qhat = (u[j + n] * b + u[j + n - 1]) / v[n - 1]; + /* Optimized mod v[n -1 ] */ + rhat = (u[j + n] * b + u[j + n - 1]) - qhat * v[n - 1]; + + while (true) { + if (qhat == b || qhat * v[n - 2] > b * rhat + u[j + n - 2]) { + qhat = qhat - 1; + rhat = rhat + v[n - 1]; + if (rhat < b) { + continue; + } + } + break; + } + + /* Step D4 : Multiply and subtract */ + k = 0; + for (i = 0; i < n; i++) { + p = qhat * v[i]; + t = u[i + j] - k - (p & 0xffffffff); + u[i + j] = t; + k = (p >> 32) - (t >> 32); + } + t = u[j + n] - k; + u[j + n] = t; + + /* Step D5 */ + q[j] = qhat; /* Store quotient digit */ + /* Step D6 */ + if (t < 0) { /* If we subtracted too much, add back */ + q[j] = q[j] - 1; + k = 0; + for (i = 0; i < n; i++) { + t = u[i + j] + v[i] + k; + u[i + j] = t; + k = t >> 32; + } + u[j + n] = u[j + n] + k; + } + } /* D7 Loop */ + } + + if (qq) { + *qq = int128_from_uint32(q); + } + + if (rr) { + /* Step D8 : Unnormalize */ + *rr = s && n != 1 + ? int128_rshift(int128_from_uint32(u), s) + : int128_from_uint32(u); + } +} + +Int128 int128_divu(Int128 a, Int128 b) +{ + Int128 q; + divrem128(a, b, &q, NULL); + return q; +} + +Int128 int128_remu(Int128 a, Int128 b) +{ + Int128 r; + divrem128(a, b, NULL, &r); + return r; +} + +Int128 int128_divs(Int128 a, Int128 b) +{ + Int128 q; + bool sgna = !int128_nonneg(a), + sgnb = !int128_nonneg(b); + + if (sgna) { + a = int128_neg(a); + } + + if (sgnb) { + b = int128_neg(b); + } + + divrem128(a, b, &q, NULL); + + if (sgna != sgnb) { + q = int128_neg(q); + } + + return q; +} + +Int128 int128_rems(Int128 a, Int128 b) +{ + Int128 r; + bool sgna = !int128_nonneg(a), + sgnb = !int128_nonneg(b); + + if (sgna) { + a = int128_neg(a); + } + + if (sgnb) { + b = int128_neg(b); + } + + divrem128(a, b, NULL, &r); + + if (sgna) { + r = int128_neg(r); + } + + return r; +} + +#endif diff --git a/util/meson.build b/util/meson.build index 05b593055a..e676b2f6c6 100644 --- a/util/meson.build +++ b/util/meson.build @@ -48,6 +48,7 @@ util_ss.add(files('transactions.c')) util_ss.add(when: 'CONFIG_POSIX', if_true: files('drm.c')) util_ss.add(files('guest-random.c')) util_ss.add(files('yank.c')) +util_ss.add(files('int128.c')) if have_user util_ss.add(files('selfmap.c')) From patchwork Mon Oct 25 12:28:04 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Patchwork-Submitter: =?utf-8?b?RnLDqWTDqXJpYyBQw6l0cm90?= X-Patchwork-Id: 12581727 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 27DC4C433EF for ; Mon, 25 Oct 2021 12:32:16 +0000 (UTC) Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.kernel.org (Postfix) with ESMTPS id 937FC6023F for ; Mon, 25 Oct 2021 12:32:15 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.4.1 mail.kernel.org 937FC6023F Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=univ-grenoble-alpes.fr Authentication-Results: mail.kernel.org; spf=pass smtp.mailfrom=nongnu.org Received: from localhost ([::1]:52380 helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1mez9C-0000Ri-M2 for qemu-devel@archiver.kernel.org; Mon, 25 Oct 2021 08:32:14 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]:52920) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1mez5k-0005qp-55; Mon, 25 Oct 2021 08:28:41 -0400 Received: from zm-mta-out-3.u-ga.fr ([152.77.200.56]:42730) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1mez5g-0007wX-DL; Mon, 25 Oct 2021 08:28:39 -0400 Received: from mailhost.u-ga.fr (mailhost2.u-ga.fr [129.88.177.242]) by zm-mta-out-3.u-ga.fr (Postfix) with ESMTP id 93DED41933; Mon, 25 Oct 2021 14:28:32 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=univ-grenoble-alpes.fr; s=2020; t=1635164912; bh=LPOi91CFE5+F/viUbvte0LfmcN76fCWGoe18/0IG7d0=; h=From:To:Cc:Subject:Date:In-Reply-To:References:From; b=0Tb6UZ/huAtfmEBs3aupjPopOee2U1hm8e4bEACYPzMRdXcDovj2zmiumWC5QDSBM K3IP9q35erhw0owZGNBhqjnsNi1Zew6/wMNNPCocPLi+roOyBqRaBv6qS7iAyZUaHt HpSzhKZNPs2zhueUru4V8XAgqIYFpO18/E6c6f5DzE8LX6pEqd1ZMYitn7zll4jGfp CBRAmex+t1geb7ajOMddP4cTm2YlKWlne08Mtvf6KeOwB6nR/QH4zzO3pOPFYrc0OI cnBox3UyiRj6uZWqOATga1uRB2Jtr1lPiwQ0bo4LKuG+lIeHnQ5FZ9MJsNJfh87/tQ VhOqV+8XhxWJw== Received: from smtps.univ-grenoble-alpes.fr (smtps2.u-ga.fr [152.77.18.2]) by mailhost.u-ga.fr (Postfix) with ESMTP id 796F060066; Mon, 25 Oct 2021 14:28:32 +0200 (CEST) Received: from palmier.tima.u-ga.fr (35.201.90.79.rev.sfr.net [79.90.201.35]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (2048 bits) server-digest SHA256) (No client certificate requested) (Authenticated sender: petrotf@univ-grenoble-alpes.fr) by smtps.univ-grenoble-alpes.fr (Postfix) with ESMTPSA id 4470314005D; Mon, 25 Oct 2021 14:28:32 +0200 (CEST) From: =?utf-8?b?RnLDqWTDqXJpYyBQw6l0cm90?= To: qemu-devel@nongnu.org, qemu-riscv@nongnu.org Subject: [PATCH v4 03/17] target/riscv: additional macros to check instruction support Date: Mon, 25 Oct 2021 14:28:04 +0200 Message-Id: <20211025122818.168890-4-frederic.petrot@univ-grenoble-alpes.fr> X-Mailer: git-send-email 2.33.0 In-Reply-To: <20211025122818.168890-1-frederic.petrot@univ-grenoble-alpes.fr> References: <20211025122818.168890-1-frederic.petrot@univ-grenoble-alpes.fr> MIME-Version: 1.0 X-Greylist: Whitelist-UGA SMTP Authentifie (petrotf@univ-grenoble-alpes.fr) via submission-587 ACL (41) X-Greylist: Whitelist-UGA MAILHOST (SMTP non authentifie) depuis 152.77.18.2 Received-SPF: pass client-ip=152.77.200.56; envelope-from=frederic.petrot@univ-grenoble-alpes.fr; helo=zm-mta-out-3.u-ga.fr X-Spam_score_int: -20 X-Spam_score: -2.1 X-Spam_bar: -- X-Spam_report: (-2.1 / 5.0 requ) BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_MSPIKE_H2=-0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: bin.meng@windriver.com, richard.henderson@linaro.org, alistair.francis@wdc.com, fabien.portas@grenoble-inp.org, palmer@dabbelt.com, =?utf-8?b?RnLDqWTDqXJpYyBQw6l0cm90?= , philmd@redhat.com Errors-To: qemu-devel-bounces+qemu-devel=archiver.kernel.org@nongnu.org Sender: "Qemu-devel" Given that the 128-bit version of the riscv spec adds new instructions, and that some instructions that were previously only available in 64-bit mode are now available for both 64-bit and 128-bit, we added new macros to check for the processor mode during translation. Although RV128 is a superset of RV64, we keep for now the RV64 only tests for extensions other than RVI and RVM. Signed-off-by: Frédéric Pétrot Co-authored-by: Fabien Portas Reviewed-by: Richard Henderson --- target/riscv/translate.c | 20 ++++++++++++++++---- 1 file changed, 16 insertions(+), 4 deletions(-) diff --git a/target/riscv/translate.c b/target/riscv/translate.c index d38f87d718..e10c8769b3 100644 --- a/target/riscv/translate.c +++ b/target/riscv/translate.c @@ -344,10 +344,22 @@ EX_SH(12) } \ } while (0) -#define REQUIRE_64BIT(ctx) do { \ - if (get_xl(ctx) < MXL_RV64) { \ - return false; \ - } \ +#define REQUIRE_64BIT(ctx) do { \ + if (get_xl(ctx) != MXL_RV64) { \ + return false; \ + } \ +} while (0) + +#define REQUIRE_128BIT(ctx) do { \ + if (get_xl(ctx) != MXL_RV128) { \ + return false; \ + } \ +} while (0) + +#define REQUIRE_64_OR_128BIT(ctx) do { \ + if (get_xl(ctx) == MXL_RV32) { \ + return false; \ + } \ } while (0) static int ex_rvc_register(DisasContext *ctx, int reg) From patchwork Mon Oct 25 12:28:05 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Patchwork-Submitter: =?utf-8?b?RnLDqWTDqXJpYyBQw6l0cm90?= X-Patchwork-Id: 12581735 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 962C3C433F5 for ; Mon, 25 Oct 2021 12:37:22 +0000 (UTC) Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.kernel.org (Postfix) with ESMTPS id 20BF9604E9 for ; Mon, 25 Oct 2021 12:37:22 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.4.1 mail.kernel.org 20BF9604E9 Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=univ-grenoble-alpes.fr Authentication-Results: mail.kernel.org; spf=pass smtp.mailfrom=nongnu.org Received: from localhost ([::1]:60936 helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1mezE8-0006c1-08 for qemu-devel@archiver.kernel.org; Mon, 25 Oct 2021 08:37:20 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]:52924) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1mez5k-0005r3-O7; Mon, 25 Oct 2021 08:28:41 -0400 Received: from zm-mta-out-3.u-ga.fr ([152.77.200.56]:42748) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1mez5g-0007wz-B8; Mon, 25 Oct 2021 08:28:40 -0400 Received: from mailhost.u-ga.fr (mailhost2.u-ga.fr [129.88.177.242]) by zm-mta-out-3.u-ga.fr (Postfix) with ESMTP id D735841866; Mon, 25 Oct 2021 14:28:32 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=univ-grenoble-alpes.fr; s=2020; t=1635164912; bh=xaFt3bSIeDUxsG9cz5UtJ0cc7VYxozTNCEYroWi2KgQ=; h=From:To:Cc:Subject:Date:In-Reply-To:References:From; b=0bXNkR1YWHJzXPVZq+2PohOSRqH++7ZJUJyYyncMnOwTb/xIF/4h3l3u/MK30c2MZ mi83QnjqGVw0oHDL2fOSxjFCWoVkxBkRsrDEPwh4ph84bcFcaT+WWQOJqkI87DF4LX 5yL2afvGE+Ry1t8XMeEpaVabwK2QpSOwW2xKRLPL4Iqg951VgnNrlLxyNqWtH8f2zb jziQQogcfDmEYIUslTZc9ty7HwXyn/yYbgo6EfPX4F9ToAvi+vAEZadHGX8MNGxoda vBhWlp7mImpAiCfQ/CUc8YhuJ/Ikl3oDx8HLqn/pgV3o8kZWgb1hwhXqck/Pk/a+H3 rNhkt2V99o3aw== Received: from smtps.univ-grenoble-alpes.fr (smtps2.u-ga.fr [152.77.18.2]) by mailhost.u-ga.fr (Postfix) with ESMTP id B589960066; Mon, 25 Oct 2021 14:28:32 +0200 (CEST) Received: from palmier.tima.u-ga.fr (35.201.90.79.rev.sfr.net [79.90.201.35]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (2048 bits) server-digest SHA256) (No client certificate requested) (Authenticated sender: petrotf@univ-grenoble-alpes.fr) by smtps.univ-grenoble-alpes.fr (Postfix) with ESMTPSA id 805D214005A; Mon, 25 Oct 2021 14:28:32 +0200 (CEST) From: =?utf-8?b?RnLDqWTDqXJpYyBQw6l0cm90?= To: qemu-devel@nongnu.org, qemu-riscv@nongnu.org Subject: [PATCH v4 04/17] target/riscv: separation of bitwise logic and aritmetic helpers Date: Mon, 25 Oct 2021 14:28:05 +0200 Message-Id: <20211025122818.168890-5-frederic.petrot@univ-grenoble-alpes.fr> X-Mailer: git-send-email 2.33.0 In-Reply-To: <20211025122818.168890-1-frederic.petrot@univ-grenoble-alpes.fr> References: <20211025122818.168890-1-frederic.petrot@univ-grenoble-alpes.fr> MIME-Version: 1.0 X-Greylist: Whitelist-UGA SMTP Authentifie (petrotf@univ-grenoble-alpes.fr) via submission-587 ACL (41) X-Greylist: Whitelist-UGA MAILHOST (SMTP non authentifie) depuis 152.77.18.2 Received-SPF: pass client-ip=152.77.200.56; envelope-from=frederic.petrot@univ-grenoble-alpes.fr; helo=zm-mta-out-3.u-ga.fr X-Spam_score_int: -20 X-Spam_score: -2.1 X-Spam_bar: -- X-Spam_report: (-2.1 / 5.0 requ) BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_MSPIKE_H2=-0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: bin.meng@windriver.com, richard.henderson@linaro.org, alistair.francis@wdc.com, fabien.portas@grenoble-inp.org, palmer@dabbelt.com, =?utf-8?b?RnLDqWTDqXJpYyBQw6l0cm90?= , philmd@redhat.com Errors-To: qemu-devel-bounces+qemu-devel=archiver.kernel.org@nongnu.org Sender: "Qemu-devel" Introduction of a gen_logic function for bitwise logic to implement instructions in which not propagation of information occurs between bits and use of this function on the bitwise instructions. Signed-off-by: Frédéric Pétrot Co-authored-by: Fabien Portas Reviewed-by: Richard Henderson Reviewed-by: Philippe Mathieu-Daudé --- target/riscv/translate.c | 27 +++++++++++++++++++++++++ target/riscv/insn_trans/trans_rvb.c.inc | 6 +++--- target/riscv/insn_trans/trans_rvi.c.inc | 12 +++++------ 3 files changed, 36 insertions(+), 9 deletions(-) diff --git a/target/riscv/translate.c b/target/riscv/translate.c index e10c8769b3..5c7971b189 100644 --- a/target/riscv/translate.c +++ b/target/riscv/translate.c @@ -376,6 +376,33 @@ static int ex_rvc_shifti(DisasContext *ctx, int imm) /* Include the auto-generated decoder for 32 bit insn */ #include "decode-insn32.c.inc" +static bool gen_logic_imm_fn(DisasContext *ctx, arg_i *a, + void (*func)(TCGv, TCGv, target_long)) +{ + TCGv dest = dest_gpr(ctx, a->rd); + TCGv src1 = get_gpr(ctx, a->rs1, EXT_NONE); + + func(dest, src1, a->imm); + + gen_set_gpr(ctx, a->rd, dest); + + return true; +} + +static bool gen_logic(DisasContext *ctx, arg_r *a, DisasExtend ext, + void (*func)(TCGv, TCGv, TCGv)) +{ + TCGv dest = dest_gpr(ctx, a->rd); + TCGv src1 = get_gpr(ctx, a->rs1, ext); + TCGv src2 = get_gpr(ctx, a->rs2, ext); + + func(dest, src1, src2); + + gen_set_gpr(ctx, a->rd, dest); + + return true; +} + static bool gen_arith_imm_fn(DisasContext *ctx, arg_i *a, DisasExtend ext, void (*func)(TCGv, TCGv, target_long)) { diff --git a/target/riscv/insn_trans/trans_rvb.c.inc b/target/riscv/insn_trans/trans_rvb.c.inc index c8d31907c5..acdcfdb178 100644 --- a/target/riscv/insn_trans/trans_rvb.c.inc +++ b/target/riscv/insn_trans/trans_rvb.c.inc @@ -86,19 +86,19 @@ static bool trans_cpop(DisasContext *ctx, arg_cpop *a) static bool trans_andn(DisasContext *ctx, arg_andn *a) { REQUIRE_ZBB(ctx); - return gen_arith(ctx, a, EXT_NONE, tcg_gen_andc_tl); + return gen_logic(ctx, a, EXT_NONE, tcg_gen_andc_tl); } static bool trans_orn(DisasContext *ctx, arg_orn *a) { REQUIRE_ZBB(ctx); - return gen_arith(ctx, a, EXT_NONE, tcg_gen_orc_tl); + return gen_logic(ctx, a, EXT_NONE, tcg_gen_orc_tl); } static bool trans_xnor(DisasContext *ctx, arg_xnor *a) { REQUIRE_ZBB(ctx); - return gen_arith(ctx, a, EXT_NONE, tcg_gen_eqv_tl); + return gen_logic(ctx, a, EXT_NONE, tcg_gen_eqv_tl); } static bool trans_min(DisasContext *ctx, arg_min *a) diff --git a/target/riscv/insn_trans/trans_rvi.c.inc b/target/riscv/insn_trans/trans_rvi.c.inc index 91dc438a3a..fae82bd946 100644 --- a/target/riscv/insn_trans/trans_rvi.c.inc +++ b/target/riscv/insn_trans/trans_rvi.c.inc @@ -250,17 +250,17 @@ static bool trans_sltiu(DisasContext *ctx, arg_sltiu *a) static bool trans_xori(DisasContext *ctx, arg_xori *a) { - return gen_arith_imm_fn(ctx, a, EXT_NONE, tcg_gen_xori_tl); + return gen_logic_imm_fn(ctx, a, tcg_gen_xori_tl); } static bool trans_ori(DisasContext *ctx, arg_ori *a) { - return gen_arith_imm_fn(ctx, a, EXT_NONE, tcg_gen_ori_tl); + return gen_logic_imm_fn(ctx, a, tcg_gen_ori_tl); } static bool trans_andi(DisasContext *ctx, arg_andi *a) { - return gen_arith_imm_fn(ctx, a, EXT_NONE, tcg_gen_andi_tl); + return gen_logic_imm_fn(ctx, a, tcg_gen_andi_tl); } static bool trans_slli(DisasContext *ctx, arg_slli *a) @@ -317,7 +317,7 @@ static bool trans_sltu(DisasContext *ctx, arg_sltu *a) static bool trans_xor(DisasContext *ctx, arg_xor *a) { - return gen_arith(ctx, a, EXT_NONE, tcg_gen_xor_tl); + return gen_logic(ctx, a, EXT_NONE, tcg_gen_xor_tl); } static bool trans_srl(DisasContext *ctx, arg_srl *a) @@ -332,12 +332,12 @@ static bool trans_sra(DisasContext *ctx, arg_sra *a) static bool trans_or(DisasContext *ctx, arg_or *a) { - return gen_arith(ctx, a, EXT_NONE, tcg_gen_or_tl); + return gen_logic(ctx, a, EXT_NONE, tcg_gen_or_tl); } static bool trans_and(DisasContext *ctx, arg_and *a) { - return gen_arith(ctx, a, EXT_NONE, tcg_gen_and_tl); + return gen_logic(ctx, a, EXT_NONE, tcg_gen_and_tl); } static bool trans_addiw(DisasContext *ctx, arg_addiw *a) From patchwork Mon Oct 25 12:28:06 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Patchwork-Submitter: =?utf-8?b?RnLDqWTDqXJpYyBQw6l0cm90?= X-Patchwork-Id: 12581739 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 77D2AC433F5 for ; Mon, 25 Oct 2021 12:37:57 +0000 (UTC) Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.kernel.org (Postfix) with ESMTPS id ED4BB60E08 for ; Mon, 25 Oct 2021 12:37:56 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.4.1 mail.kernel.org ED4BB60E08 Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=univ-grenoble-alpes.fr Authentication-Results: mail.kernel.org; spf=pass smtp.mailfrom=nongnu.org Received: from localhost ([::1]:34324 helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1mezEh-0007hR-VI for qemu-devel@archiver.kernel.org; Mon, 25 Oct 2021 08:37:55 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]:53010) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1mez5n-0005re-Jb; Mon, 25 Oct 2021 08:28:43 -0400 Received: from zm-mta-out-3.u-ga.fr ([152.77.200.56]:43062) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1mez5l-00083W-Cs; Mon, 25 Oct 2021 08:28:43 -0400 Received: from mailhost.u-ga.fr (mailhost1.u-ga.fr [152.77.1.10]) by zm-mta-out-3.u-ga.fr (Postfix) with ESMTP id 1BAEE41965; Mon, 25 Oct 2021 14:28:33 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=univ-grenoble-alpes.fr; s=2020; t=1635164913; bh=CrGAg0XluHvjypIQoChkRG8OCZ7bzXGXawDAStTZlxg=; h=From:To:Cc:Subject:Date:In-Reply-To:References:From; b=coNik2FD19DztaLtpbXiVwyVfxkAAALUMI7dc1HiqPJlgnV6lL1kHH3SBN1uzGI8o Zl0chhs2S0yPpoLatA6JPZsN4xzZm6rOrWrKiChTHob2tYD7bhQbvk4WdJJL/Wi0nF cvPrjeN8ral84QUI9snKpedvqrHN7Jrkg62DIQEpHosEG9L53acmIoCpT6vt3+HhnL cJsV6XFZ0UNeZqhRRIL1NfHURf91RhNxQ24IlO3CVzqjM7/KlwOuSDAEVgY9Jzi24M 5vDSFItq6O6HaL4Ly8yr0LKujgyJaqweVhCXYtx2lsfPIqMeyMGgO7II5lBNwA4Ob8 PFhuRZx6x/O/Q== Received: from smtps.univ-grenoble-alpes.fr (smtps2.u-ga.fr [152.77.18.2]) by mailhost.u-ga.fr (Postfix) with ESMTP id EE2DA60070; Mon, 25 Oct 2021 14:28:32 +0200 (CEST) Received: from palmier.tima.u-ga.fr (35.201.90.79.rev.sfr.net [79.90.201.35]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (2048 bits) server-digest SHA256) (No client certificate requested) (Authenticated sender: petrotf@univ-grenoble-alpes.fr) by smtps.univ-grenoble-alpes.fr (Postfix) with ESMTPSA id BBD4A14005D; Mon, 25 Oct 2021 14:28:32 +0200 (CEST) From: =?utf-8?b?RnLDqWTDqXJpYyBQw6l0cm90?= To: qemu-devel@nongnu.org, qemu-riscv@nongnu.org Subject: [PATCH v4 05/17] target/riscv: array for the 64 upper bits of 128-bit registers Date: Mon, 25 Oct 2021 14:28:06 +0200 Message-Id: <20211025122818.168890-6-frederic.petrot@univ-grenoble-alpes.fr> X-Mailer: git-send-email 2.33.0 In-Reply-To: <20211025122818.168890-1-frederic.petrot@univ-grenoble-alpes.fr> References: <20211025122818.168890-1-frederic.petrot@univ-grenoble-alpes.fr> MIME-Version: 1.0 X-Greylist: Whitelist-UGA SMTP Authentifie (petrotf@univ-grenoble-alpes.fr) via submission-587 ACL (41) X-Greylist: Whitelist-UGA MAILHOST (SMTP non authentifie) depuis 152.77.18.2 Received-SPF: pass client-ip=152.77.200.56; envelope-from=frederic.petrot@univ-grenoble-alpes.fr; helo=zm-mta-out-3.u-ga.fr X-Spam_score_int: -20 X-Spam_score: -2.1 X-Spam_bar: -- X-Spam_report: (-2.1 / 5.0 requ) BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_MSPIKE_H2=-0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: bin.meng@windriver.com, richard.henderson@linaro.org, alistair.francis@wdc.com, fabien.portas@grenoble-inp.org, palmer@dabbelt.com, =?utf-8?b?RnLDqWTDqXJpYyBQw6l0cm90?= , philmd@redhat.com Errors-To: qemu-devel-bounces+qemu-devel=archiver.kernel.org@nongnu.org Sender: "Qemu-devel" The upper 64-bit of the 128-bit registers have now a place inside the cpu state structure, and are created as globals for future use. Signed-off-by: Frédéric Pétrot Co-authored-by: Fabien Portas Reviewed-by: Richard Henderson --- target/riscv/cpu.h | 2 ++ target/riscv/cpu.c | 9 +++++++++ target/riscv/machine.c | 20 ++++++++++++++++++++ target/riscv/translate.c | 5 ++++- 4 files changed, 35 insertions(+), 1 deletion(-) diff --git a/target/riscv/cpu.h b/target/riscv/cpu.h index a33dc30be8..f6d7a1793d 100644 --- a/target/riscv/cpu.h +++ b/target/riscv/cpu.h @@ -109,6 +109,7 @@ FIELD(VTYPE, VILL, sizeof(target_ulong) * 8 - 1, 1) struct CPURISCVState { target_ulong gpr[32]; + target_ulong gprh[32]; /* 64 top bits of the 128-bit registers */ uint64_t fpr[32]; /* assume both F and D extensions */ /* vector coprocessor state. */ @@ -326,6 +327,7 @@ static inline bool riscv_feature(CPURISCVState *env, int feature) #include "cpu_user.h" extern const char * const riscv_int_regnames[]; +extern const char * const riscv_int_regnamesh[]; extern const char * const riscv_fpr_regnames[]; const char *riscv_cpu_get_trap_name(target_ulong cause, bool async); diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c index 788fa0b11c..8c1cda35b4 100644 --- a/target/riscv/cpu.c +++ b/target/riscv/cpu.c @@ -42,6 +42,15 @@ const char * const riscv_int_regnames[] = { "x28/t3", "x29/t4", "x30/t5", "x31/t6" }; +const char * const riscv_int_regnamesh[] = { + "x0h/zeroh", "x1h/rah", "x2h/sph", "x3h/gph", "x4h/tph", "x5h/t0h", + "x6h/t1h", "x7h/t2h", "x8h/s0h", "x9h/s1h", "x10h/a0h", "x11h/a1h", + "x12h/a2h", "x13h/a3h", "x14h/a4h", "x15h/a5h", "x16h/a6h", "x17h/a7h", + "x18h/s2h", "x19h/s3h", "x20h/s4h", "x21h/s5h", "x22h/s6h", "x23h/s7h", + "x24h/s8h", "x25h/s9h", "x26h/s10h", "x27h/s11h", "x28h/t3h", "x29h/t4h", + "x30h/t5h", "x31h/t6h" +}; + const char * const riscv_fpr_regnames[] = { "f0/ft0", "f1/ft1", "f2/ft2", "f3/ft3", "f4/ft4", "f5/ft5", "f6/ft6", "f7/ft7", "f8/fs0", "f9/fs1", "f10/fa0", "f11/fa1", diff --git a/target/riscv/machine.c b/target/riscv/machine.c index f64b2a96c1..78636ec96d 100644 --- a/target/riscv/machine.c +++ b/target/riscv/machine.c @@ -84,6 +84,14 @@ static bool vector_needed(void *opaque) return riscv_has_ext(env, RVV); } +static bool rv128_needed(void *opaque) +{ + RISCVCPU *cpu = opaque; + CPURISCVState *env = &cpu->env; + + return env->misa_mxl_max == MXL_RV128; +} + static const VMStateDescription vmstate_vector = { .name = "cpu/vector", .version_id = 1, @@ -138,6 +146,17 @@ static const VMStateDescription vmstate_hyper = { } }; +static const VMStateDescription vmstate_rv128 = { + .name = "cpu/rv128", + .version_id = 1, + .minimum_version_id = 1, + .needed = rv128_needed, + .fields = (VMStateField[]) { + VMSTATE_UINTTL_ARRAY(env.gprh, RISCVCPU, 32), + VMSTATE_END_OF_LIST() + } +}; + const VMStateDescription vmstate_riscv_cpu = { .name = "cpu", .version_id = 3, @@ -191,6 +210,7 @@ const VMStateDescription vmstate_riscv_cpu = { &vmstate_pmp, &vmstate_hyper, &vmstate_vector, + &vmstate_rv128, NULL } }; diff --git a/target/riscv/translate.c b/target/riscv/translate.c index 5c7971b189..63a52bb707 100644 --- a/target/riscv/translate.c +++ b/target/riscv/translate.c @@ -32,7 +32,7 @@ #include "instmap.h" /* global register indices */ -static TCGv cpu_gpr[32], cpu_pc, cpu_vl; +static TCGv cpu_gpr[32], cpu_gprh[32], cpu_pc, cpu_vl; static TCGv_i64 cpu_fpr[32]; /* assume F and D extensions */ static TCGv load_res; static TCGv load_val; @@ -749,10 +749,13 @@ void riscv_translate_init(void) * unless you specifically block reads/writes to reg 0. */ cpu_gpr[0] = NULL; + cpu_gprh[0] = NULL; for (i = 1; i < 32; i++) { cpu_gpr[i] = tcg_global_mem_new(cpu_env, offsetof(CPURISCVState, gpr[i]), riscv_int_regnames[i]); + cpu_gprh[i] = tcg_global_mem_new(cpu_env, + offsetof(CPURISCVState, gprh[i]), riscv_int_regnamesh[i]); } for (i = 0; i < 32; i++) { From patchwork Mon Oct 25 12:28:07 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Patchwork-Submitter: =?utf-8?b?RnLDqWTDqXJpYyBQw6l0cm90?= X-Patchwork-Id: 12581757 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id AF62EC433EF for ; Mon, 25 Oct 2021 12:44:53 +0000 (UTC) Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.kernel.org (Postfix) with ESMTPS id 44D0260EE9 for ; Mon, 25 Oct 2021 12:44:53 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.4.1 mail.kernel.org 44D0260EE9 Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=univ-grenoble-alpes.fr Authentication-Results: mail.kernel.org; spf=pass smtp.mailfrom=nongnu.org Received: from localhost ([::1]:52122 helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1mezLQ-00033c-D9 for qemu-devel@archiver.kernel.org; Mon, 25 Oct 2021 08:44:52 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]:53024) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1mez5o-0005sc-3m; Mon, 25 Oct 2021 08:28:45 -0400 Received: from zm-mta-out-3.u-ga.fr ([152.77.200.56]:43064) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1mez5l-00083V-F4; Mon, 25 Oct 2021 08:28:43 -0400 Received: from mailhost.u-ga.fr (mailhost2.u-ga.fr [129.88.177.242]) by zm-mta-out-3.u-ga.fr (Postfix) with ESMTP id 5778D4194E; Mon, 25 Oct 2021 14:28:33 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=univ-grenoble-alpes.fr; s=2020; t=1635164913; bh=uT6vnRpemLtf724ZUcTdhjnsd29ROhVIsC3XLY4F2pw=; h=From:To:Cc:Subject:Date:In-Reply-To:References:From; b=lzcFO6tclepRwm0bP3ryVOcqkxx5TX6rwJwUzV3Wgc6G4BnHMl7VP50wWgAqacFut pAedVkCnPqpWMb4XR4nkXjVuRh56+44DvNAft1wwCs0zHqond73cR88ErJrrN2rrLq 9ZH1+wi4MshtJQepKV7etlzRQXwNL+qxYxiaDV7SiFzx9sf6pWmRgFOcMimOz+mUmD UgfaR+/yj24MekeMtbLWcHpiH3+xuI03LUkmXt/u+Hp5u5Uko1GWeIA0TII5vaTDz4 iO/z8Gi+ofi6EWrde5JEgtjGp+3jLQzOOL3igobaIGtP0PAEuPzYZIVkh5ghIn2T/8 EWo5pK9rQJR2g== Received: from smtps.univ-grenoble-alpes.fr (smtps2.u-ga.fr [152.77.18.2]) by mailhost.u-ga.fr (Postfix) with ESMTP id 3B94160066; Mon, 25 Oct 2021 14:28:33 +0200 (CEST) Received: from palmier.tima.u-ga.fr (35.201.90.79.rev.sfr.net [79.90.201.35]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (2048 bits) server-digest SHA256) (No client certificate requested) (Authenticated sender: petrotf@univ-grenoble-alpes.fr) by smtps.univ-grenoble-alpes.fr (Postfix) with ESMTPSA id 02DE614005A; Mon, 25 Oct 2021 14:28:32 +0200 (CEST) From: =?utf-8?b?RnLDqWTDqXJpYyBQw6l0cm90?= To: qemu-devel@nongnu.org, qemu-riscv@nongnu.org Subject: [PATCH v4 06/17] target/riscv: setup everything so that riscv128-softmmu compiles Date: Mon, 25 Oct 2021 14:28:07 +0200 Message-Id: <20211025122818.168890-7-frederic.petrot@univ-grenoble-alpes.fr> X-Mailer: git-send-email 2.33.0 In-Reply-To: <20211025122818.168890-1-frederic.petrot@univ-grenoble-alpes.fr> References: <20211025122818.168890-1-frederic.petrot@univ-grenoble-alpes.fr> MIME-Version: 1.0 X-Greylist: Whitelist-UGA SMTP Authentifie (petrotf@univ-grenoble-alpes.fr) via submission-587 ACL (41) X-Greylist: Whitelist-UGA MAILHOST (SMTP non authentifie) depuis 152.77.18.2 Received-SPF: pass client-ip=152.77.200.56; envelope-from=frederic.petrot@univ-grenoble-alpes.fr; helo=zm-mta-out-3.u-ga.fr X-Spam_score_int: -20 X-Spam_score: -2.1 X-Spam_bar: -- X-Spam_report: (-2.1 / 5.0 requ) BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_MSPIKE_H2=-0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: bin.meng@windriver.com, richard.henderson@linaro.org, alistair.francis@wdc.com, fabien.portas@grenoble-inp.org, palmer@dabbelt.com, =?utf-8?b?RnLDqWTDqXJpYyBQw6l0cm90?= , philmd@redhat.com Errors-To: qemu-devel-bounces+qemu-devel=archiver.kernel.org@nongnu.org Sender: "Qemu-devel" This patch is kind of a mess because several files have to be slightly modified to allow for a new target. In the current status, we have done our best to have RV64 and RV128 under the same RV64 umbrella, but there is still work to do to have a single executable for both. Once this patch applied, adding risc128-sofmmu to --target-list produces a (no so useful yet) executable. Signed-off-by: Frédéric Pétrot Co-authored-by: Fabien Portas --- configs/devices/riscv128-softmmu/default.mak | 17 +++++++++++++++ configs/targets/riscv128-softmmu.mak | 5 +++++ include/disas/dis-asm.h | 1 + include/hw/riscv/sifive_cpu.h | 3 +++ target/riscv/cpu-param.h | 5 +++++ target/riscv/cpu.h | 3 +++ disas/riscv.c | 5 +++++ target/riscv/cpu.c | 22 ++++++++++++++++++-- target/riscv/gdbstub.c | 8 +++++++ target/riscv/insn_trans/trans_rvd.c.inc | 12 +++++------ target/riscv/insn_trans/trans_rvf.c.inc | 6 +++--- target/riscv/Kconfig | 3 +++ 12 files changed, 79 insertions(+), 11 deletions(-) create mode 100644 configs/devices/riscv128-softmmu/default.mak create mode 100644 configs/targets/riscv128-softmmu.mak diff --git a/configs/devices/riscv128-softmmu/default.mak b/configs/devices/riscv128-softmmu/default.mak new file mode 100644 index 0000000000..e838f35785 --- /dev/null +++ b/configs/devices/riscv128-softmmu/default.mak @@ -0,0 +1,17 @@ +# Default configuration for riscv128-softmmu + +# Uncomment the following lines to disable these optional devices: +# +#CONFIG_PCI_DEVICES=n +# No does not seem to be an option for these two parameters +CONFIG_SEMIHOSTING=y +CONFIG_ARM_COMPATIBLE_SEMIHOSTING=y + +# Boards: +# +CONFIG_SPIKE=n +CONFIG_SIFIVE_E=n +CONFIG_SIFIVE_U=n +CONFIG_RISCV_VIRT=y +CONFIG_MICROCHIP_PFSOC=n +CONFIG_SHAKTI_C=n diff --git a/configs/targets/riscv128-softmmu.mak b/configs/targets/riscv128-softmmu.mak new file mode 100644 index 0000000000..a9343d57d3 --- /dev/null +++ b/configs/targets/riscv128-softmmu.mak @@ -0,0 +1,5 @@ +TARGET_ARCH=riscv128 +TARGET_BASE_ARCH=riscv +TARGET_SUPPORTS_MTTCG=y +TARGET_XML_FILES= gdb-xml/riscv-64bit-cpu.xml gdb-xml/riscv-32bit-fpu.xml gdb-xml/riscv-64bit-fpu.xml gdb-xml/riscv-64bit-virtual.xml +TARGET_NEED_FDT=y diff --git a/include/disas/dis-asm.h b/include/disas/dis-asm.h index 08e1beec85..102a1e7f50 100644 --- a/include/disas/dis-asm.h +++ b/include/disas/dis-asm.h @@ -459,6 +459,7 @@ int print_insn_nios2(bfd_vma, disassemble_info*); int print_insn_xtensa (bfd_vma, disassemble_info*); int print_insn_riscv32 (bfd_vma, disassemble_info*); int print_insn_riscv64 (bfd_vma, disassemble_info*); +int print_insn_riscv128 (bfd_vma, disassemble_info*); int print_insn_rx(bfd_vma, disassemble_info *); int print_insn_hexagon(bfd_vma, disassemble_info *); diff --git a/include/hw/riscv/sifive_cpu.h b/include/hw/riscv/sifive_cpu.h index 136799633a..64078feba8 100644 --- a/include/hw/riscv/sifive_cpu.h +++ b/include/hw/riscv/sifive_cpu.h @@ -26,6 +26,9 @@ #elif defined(TARGET_RISCV64) #define SIFIVE_E_CPU TYPE_RISCV_CPU_SIFIVE_E51 #define SIFIVE_U_CPU TYPE_RISCV_CPU_SIFIVE_U54 +#else +#define SIFIVE_E_CPU TYPE_RISCV_CPU_SIFIVE_E51 +#define SIFIVE_U_CPU TYPE_RISCV_CPU_SIFIVE_U54 #endif #endif /* HW_SIFIVE_CPU_H */ diff --git a/target/riscv/cpu-param.h b/target/riscv/cpu-param.h index 80eb615f93..c10459b56f 100644 --- a/target/riscv/cpu-param.h +++ b/target/riscv/cpu-param.h @@ -16,6 +16,11 @@ # define TARGET_LONG_BITS 32 # define TARGET_PHYS_ADDR_SPACE_BITS 34 /* 22-bit PPN */ # define TARGET_VIRT_ADDR_SPACE_BITS 32 /* sv32 */ +#else +/* 64-bit target, since QEMU isn't built to have TARGET_LONG_BITS over 64 */ +# define TARGET_LONG_BITS 64 +# define TARGET_PHYS_ADDR_SPACE_BITS 56 /* 44-bit PPN */ +# define TARGET_VIRT_ADDR_SPACE_BITS 48 /* sv48 */ #endif #define TARGET_PAGE_BITS 12 /* 4 KiB Pages */ /* diff --git a/target/riscv/cpu.h b/target/riscv/cpu.h index f6d7a1793d..6c15dcacfc 100644 --- a/target/riscv/cpu.h +++ b/target/riscv/cpu.h @@ -38,6 +38,7 @@ #define TYPE_RISCV_CPU_ANY RISCV_CPU_TYPE_NAME("any") #define TYPE_RISCV_CPU_BASE32 RISCV_CPU_TYPE_NAME("rv32") #define TYPE_RISCV_CPU_BASE64 RISCV_CPU_TYPE_NAME("rv64") +#define TYPE_RISCV_CPU_BASE128 RISCV_CPU_TYPE_NAME("rv128") #define TYPE_RISCV_CPU_IBEX RISCV_CPU_TYPE_NAME("lowrisc-ibex") #define TYPE_RISCV_CPU_SHAKTI_C RISCV_CPU_TYPE_NAME("shakti-c") #define TYPE_RISCV_CPU_SIFIVE_E31 RISCV_CPU_TYPE_NAME("sifive-e31") @@ -50,6 +51,8 @@ # define TYPE_RISCV_CPU_BASE TYPE_RISCV_CPU_BASE32 #elif defined(TARGET_RISCV64) # define TYPE_RISCV_CPU_BASE TYPE_RISCV_CPU_BASE64 +#else +# define TYPE_RISCV_CPU_BASE TYPE_RISCV_CPU_BASE128 #endif #define RV(x) ((target_ulong)1 << (x - 'A')) diff --git a/disas/riscv.c b/disas/riscv.c index 793ad14c27..03c8dc9961 100644 --- a/disas/riscv.c +++ b/disas/riscv.c @@ -3090,3 +3090,8 @@ int print_insn_riscv64(bfd_vma memaddr, struct disassemble_info *info) { return print_insn_riscv(memaddr, info, rv64); } + +int print_insn_riscv128(bfd_vma memaddr, struct disassemble_info *info) +{ + return print_insn_riscv(memaddr, info, rv128); +} diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c index 8c1cda35b4..141b5a329f 100644 --- a/target/riscv/cpu.c +++ b/target/riscv/cpu.c @@ -152,6 +152,8 @@ static void riscv_any_cpu_init(Object *obj) set_misa(env, MXL_RV32, RVI | RVM | RVA | RVF | RVD | RVC | RVU); #elif defined(TARGET_RISCV64) set_misa(env, MXL_RV64, RVI | RVM | RVA | RVF | RVD | RVC | RVU); +#else + set_misa(env, MXL_RV128, RVI | RVM | RVA | RVF | RVD | RVC | RVU); #endif set_priv_version(env, PRIV_VERSION_1_11_0); } @@ -178,7 +180,7 @@ static void rv64_sifive_e_cpu_init(Object *obj) set_priv_version(env, PRIV_VERSION_1_10_0); qdev_prop_set_bit(DEVICE(obj), "mmu", false); } -#else +#elif defined(TARGET_RISCV32) static void rv32_base_cpu_init(Object *obj) { CPURISCVState *env = &RISCV_CPU(obj)->env; @@ -218,6 +220,13 @@ static void rv32_imafcu_nommu_cpu_init(Object *obj) set_resetvec(env, DEFAULT_RSTVEC); qdev_prop_set_bit(DEVICE(obj), "mmu", false); } +#else +static void rv128_base_cpu_init(Object *obj) +{ + CPURISCVState *env = &RISCV_CPU(obj)->env; + /* We set this in the realise function */ + set_misa(env, MXL_RV128, 0); +} #endif static ObjectClass *riscv_cpu_class_by_name(const char *cpu_model) @@ -393,6 +402,9 @@ static void riscv_cpu_disas_set_info(CPUState *s, disassemble_info *info) case MXL_RV64: info->print_insn = print_insn_riscv64; break; + case MXL_RV128: + info->print_insn = print_insn_riscv128; + break; default: g_assert_not_reached(); } @@ -455,6 +467,9 @@ static void riscv_cpu_realize(DeviceState *dev, Error **errp) #ifdef TARGET_RISCV64 case MXL_RV64: break; +#elif !defined(TARGET_RISCV32) + case MXL_RV128: + break; #endif case MXL_RV32: break; @@ -657,6 +672,7 @@ static gchar *riscv_gdb_arch_name(CPUState *cs) case MXL_RV32: return g_strdup("riscv:rv32"); case MXL_RV64: + case MXL_RV128: return g_strdup("riscv:rv64"); default: g_assert_not_reached(); @@ -720,7 +736,7 @@ static void riscv_cpu_class_init(ObjectClass *c, void *data) cc->gdb_num_core_regs = 33; #if defined(TARGET_RISCV32) cc->gdb_core_xml_file = "riscv-32bit-cpu.xml"; -#elif defined(TARGET_RISCV64) +#else cc->gdb_core_xml_file = "riscv-64bit-cpu.xml"; #endif cc->gdb_stop_before_watchpoint = true; @@ -809,6 +825,8 @@ static const TypeInfo riscv_cpu_type_infos[] = { DEFINE_CPU(TYPE_RISCV_CPU_SIFIVE_E51, rv64_sifive_e_cpu_init), DEFINE_CPU(TYPE_RISCV_CPU_SIFIVE_U54, rv64_sifive_u_cpu_init), DEFINE_CPU(TYPE_RISCV_CPU_SHAKTI_C, rv64_sifive_u_cpu_init), +#else + DEFINE_CPU(TYPE_RISCV_CPU_BASE128, rv128_base_cpu_init), #endif }; diff --git a/target/riscv/gdbstub.c b/target/riscv/gdbstub.c index 23429179e2..7672ae123f 100644 --- a/target/riscv/gdbstub.c +++ b/target/riscv/gdbstub.c @@ -164,6 +164,11 @@ static int riscv_gen_dynamic_csr_xml(CPUState *cs, int base_reg) int bitsize = 16 << env->misa_mxl_max; int i; + /* Until gdb knows about 128-bit registers */ + if (bitsize > 64) { + bitsize = 64; + } + g_string_printf(s, ""); g_string_append_printf(s, ""); g_string_append_printf(s, ""); @@ -204,6 +209,9 @@ void riscv_cpu_register_gdb_regs_for_features(CPUState *cs) #elif defined(TARGET_RISCV64) gdb_register_coprocessor(cs, riscv_gdb_get_virtual, riscv_gdb_set_virtual, 1, "riscv-64bit-virtual.xml", 0); +#else + gdb_register_coprocessor(cs, riscv_gdb_get_virtual, riscv_gdb_set_virtual, + 1, "riscv-64bit-virtual.xml", 0); #endif gdb_register_coprocessor(cs, riscv_gdb_get_csr, riscv_gdb_set_csr, diff --git a/target/riscv/insn_trans/trans_rvd.c.inc b/target/riscv/insn_trans/trans_rvd.c.inc index db9ae15755..41da696ec4 100644 --- a/target/riscv/insn_trans/trans_rvd.c.inc +++ b/target/riscv/insn_trans/trans_rvd.c.inc @@ -393,11 +393,11 @@ static bool trans_fmv_x_d(DisasContext *ctx, arg_fmv_x_d *a) REQUIRE_FPU; REQUIRE_EXT(ctx, RVD); -#ifdef TARGET_RISCV64 +#ifdef TARGET_RISCV32 + qemu_build_not_reached(); +#else gen_set_gpr(ctx, a->rd, cpu_fpr[a->rs1]); return true; -#else - qemu_build_not_reached(); #endif } @@ -437,11 +437,11 @@ static bool trans_fmv_d_x(DisasContext *ctx, arg_fmv_d_x *a) REQUIRE_FPU; REQUIRE_EXT(ctx, RVD); -#ifdef TARGET_RISCV64 +#ifdef TARGET_RISCV32 + qemu_build_not_reached(); +#else tcg_gen_mov_tl(cpu_fpr[a->rd], get_gpr(ctx, a->rs1, EXT_NONE)); mark_fs_dirty(ctx); return true; -#else - qemu_build_not_reached(); #endif } diff --git a/target/riscv/insn_trans/trans_rvf.c.inc b/target/riscv/insn_trans/trans_rvf.c.inc index bddbd418d9..90cc51e5d6 100644 --- a/target/riscv/insn_trans/trans_rvf.c.inc +++ b/target/riscv/insn_trans/trans_rvf.c.inc @@ -311,10 +311,10 @@ static bool trans_fmv_x_w(DisasContext *ctx, arg_fmv_x_w *a) TCGv dest = dest_gpr(ctx, a->rd); -#if defined(TARGET_RISCV64) - tcg_gen_ext32s_tl(dest, cpu_fpr[a->rs1]); -#else +#if defined(TARGET_RISCV32) tcg_gen_extrl_i64_i32(dest, cpu_fpr[a->rs1]); +#else + tcg_gen_ext32s_tl(dest, cpu_fpr[a->rs1]); #endif gen_set_gpr(ctx, a->rd, dest); diff --git a/target/riscv/Kconfig b/target/riscv/Kconfig index b9e5932f13..f9ea52a59a 100644 --- a/target/riscv/Kconfig +++ b/target/riscv/Kconfig @@ -3,3 +3,6 @@ config RISCV32 config RISCV64 bool + +config RISCV128 + bool From patchwork Mon Oct 25 12:28:08 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Patchwork-Submitter: =?utf-8?b?RnLDqWTDqXJpYyBQw6l0cm90?= X-Patchwork-Id: 12581767 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 8DE5DC433F5 for ; Mon, 25 Oct 2021 12:49:34 +0000 (UTC) Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.kernel.org (Postfix) with ESMTPS id 314EA60F22 for ; Mon, 25 Oct 2021 12:49:34 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.4.1 mail.kernel.org 314EA60F22 Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=univ-grenoble-alpes.fr Authentication-Results: mail.kernel.org; spf=pass smtp.mailfrom=nongnu.org Received: from localhost ([::1]:60952 helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1mezPx-0000wb-16 for qemu-devel@archiver.kernel.org; Mon, 25 Oct 2021 08:49:33 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]:53052) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1mez5p-0005sd-IO; Mon, 25 Oct 2021 08:28:45 -0400 Received: from zm-mta-out-3.u-ga.fr ([152.77.200.56]:43074) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1mez5m-00083e-55; Mon, 25 Oct 2021 08:28:45 -0400 Received: from mailhost.u-ga.fr (mailhost1.u-ga.fr [152.77.1.10]) by zm-mta-out-3.u-ga.fr (Postfix) with ESMTP id 911274196A; Mon, 25 Oct 2021 14:28:33 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=univ-grenoble-alpes.fr; s=2020; t=1635164913; bh=rnzny5SBNcDev6OxmQUYKCH0VIP3Mz0joQDG3oeRr0o=; h=From:To:Cc:Subject:Date:In-Reply-To:References:From; b=R6dL7A6Iw4Oczoo0kA0uObZBNiafRqcj41v0u0whMI5Rl2dDWSYMwFOZscHWxiRL1 f7yWRZ5X6VgFNpp59x0QGVLXFB8s7IIOeK4r4y24SHsssti0N0o65OfJjHKMCzoLNb gaNeMqIt3whoxq0ohI8lW47T2VR2AmZhQ3qRUeAdcrBshqTfttZebeMsNfNduCG9Jq 1e8WNhWtYgjuUX+x5/eYlf+5o9M7FmaYyBqzNP16JLb0UUiR2GYaD/dtkADMC/BM8+ X6/Eyf+IM6GEa/U/kfbVXvVUWf8A1W45Yhom2n3KYx2Pu87fskZvhDk3hYqDDbm65W IDDfle7659ghA== Received: from smtps.univ-grenoble-alpes.fr (smtps2.u-ga.fr [152.77.18.2]) by mailhost.u-ga.fr (Postfix) with ESMTP id 748AD60067; Mon, 25 Oct 2021 14:28:33 +0200 (CEST) Received: from palmier.tima.u-ga.fr (35.201.90.79.rev.sfr.net [79.90.201.35]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (2048 bits) server-digest SHA256) (No client certificate requested) (Authenticated sender: petrotf@univ-grenoble-alpes.fr) by smtps.univ-grenoble-alpes.fr (Postfix) with ESMTPSA id 41A4714005D; Mon, 25 Oct 2021 14:28:33 +0200 (CEST) From: =?utf-8?b?RnLDqWTDqXJpYyBQw6l0cm90?= To: qemu-devel@nongnu.org, qemu-riscv@nongnu.org Subject: [PATCH v4 07/17] target/riscv: moving some insns close to similar insns Date: Mon, 25 Oct 2021 14:28:08 +0200 Message-Id: <20211025122818.168890-8-frederic.petrot@univ-grenoble-alpes.fr> X-Mailer: git-send-email 2.33.0 In-Reply-To: <20211025122818.168890-1-frederic.petrot@univ-grenoble-alpes.fr> References: <20211025122818.168890-1-frederic.petrot@univ-grenoble-alpes.fr> MIME-Version: 1.0 X-Greylist: Whitelist-UGA SMTP Authentifie (petrotf@univ-grenoble-alpes.fr) via submission-587 ACL (41) X-Greylist: Whitelist-UGA MAILHOST (SMTP non authentifie) depuis 152.77.18.2 Received-SPF: pass client-ip=152.77.200.56; envelope-from=frederic.petrot@univ-grenoble-alpes.fr; helo=zm-mta-out-3.u-ga.fr X-Spam_score_int: -20 X-Spam_score: -2.1 X-Spam_bar: -- X-Spam_report: (-2.1 / 5.0 requ) BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_MSPIKE_H2=-0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: bin.meng@windriver.com, richard.henderson@linaro.org, alistair.francis@wdc.com, fabien.portas@grenoble-inp.org, palmer@dabbelt.com, =?utf-8?b?RnLDqWTDqXJpYyBQw6l0cm90?= , philmd@redhat.com Errors-To: qemu-devel-bounces+qemu-devel=archiver.kernel.org@nongnu.org Sender: "Qemu-devel" lwu and ld are functionally close to the other loads, but were after the stores in the source file. Similarly, xor was away from or and and by two arithmetic functions, while the immediate versions were nicely put together. This patch moves the aforementioned loads after lhu, and xor above or, where they more logically belong. Signed-off-by: Frédéric Pétrot Co-authored-by: Fabien Portas Reviewed-by: Richard Henderson Reviewed-by: Philippe Mathieu-Daudé --- target/riscv/insn_trans/trans_rvi.c.inc | 34 ++++++++++++------------- 1 file changed, 17 insertions(+), 17 deletions(-) diff --git a/target/riscv/insn_trans/trans_rvi.c.inc b/target/riscv/insn_trans/trans_rvi.c.inc index fae82bd946..34164494b9 100644 --- a/target/riscv/insn_trans/trans_rvi.c.inc +++ b/target/riscv/insn_trans/trans_rvi.c.inc @@ -175,6 +175,18 @@ static bool trans_lhu(DisasContext *ctx, arg_lhu *a) return gen_load(ctx, a, MO_TEUW); } +static bool trans_lwu(DisasContext *ctx, arg_lwu *a) +{ + REQUIRE_64BIT(ctx); + return gen_load(ctx, a, MO_TEUL); +} + +static bool trans_ld(DisasContext *ctx, arg_ld *a) +{ + REQUIRE_64BIT(ctx); + return gen_load(ctx, a, MO_TEQ); +} + static bool gen_store(DisasContext *ctx, arg_sb *a, MemOp memop) { TCGv addr = get_gpr(ctx, a->rs1, EXT_NONE); @@ -205,18 +217,6 @@ static bool trans_sw(DisasContext *ctx, arg_sw *a) return gen_store(ctx, a, MO_TESL); } -static bool trans_lwu(DisasContext *ctx, arg_lwu *a) -{ - REQUIRE_64BIT(ctx); - return gen_load(ctx, a, MO_TEUL); -} - -static bool trans_ld(DisasContext *ctx, arg_ld *a) -{ - REQUIRE_64BIT(ctx); - return gen_load(ctx, a, MO_TEQ); -} - static bool trans_sd(DisasContext *ctx, arg_sd *a) { REQUIRE_64BIT(ctx); @@ -315,11 +315,6 @@ static bool trans_sltu(DisasContext *ctx, arg_sltu *a) return gen_arith(ctx, a, EXT_SIGN, gen_sltu); } -static bool trans_xor(DisasContext *ctx, arg_xor *a) -{ - return gen_logic(ctx, a, EXT_NONE, tcg_gen_xor_tl); -} - static bool trans_srl(DisasContext *ctx, arg_srl *a) { return gen_shift(ctx, a, EXT_ZERO, tcg_gen_shr_tl); @@ -330,6 +325,11 @@ static bool trans_sra(DisasContext *ctx, arg_sra *a) return gen_shift(ctx, a, EXT_SIGN, tcg_gen_sar_tl); } +static bool trans_xor(DisasContext *ctx, arg_xor *a) +{ + return gen_logic(ctx, a, EXT_NONE, tcg_gen_xor_tl); +} + static bool trans_or(DisasContext *ctx, arg_or *a) { return gen_logic(ctx, a, EXT_NONE, tcg_gen_or_tl); From patchwork Mon Oct 25 12:28:09 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Patchwork-Submitter: =?utf-8?b?RnLDqWTDqXJpYyBQw6l0cm90?= X-Patchwork-Id: 12581783 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id E000DC433F5 for ; Mon, 25 Oct 2021 12:55:42 +0000 (UTC) Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.kernel.org (Postfix) with ESMTPS id 5730860FDC for ; Mon, 25 Oct 2021 12:55:42 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.4.1 mail.kernel.org 5730860FDC Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=univ-grenoble-alpes.fr Authentication-Results: mail.kernel.org; spf=pass smtp.mailfrom=nongnu.org Received: from localhost ([::1]:50824 helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1mezVt-0004gt-FA for qemu-devel@archiver.kernel.org; Mon, 25 Oct 2021 08:55:41 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]:53060) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1mez5r-0005tc-4p; Mon, 25 Oct 2021 08:28:47 -0400 Received: from zm-mta-out-3.u-ga.fr ([152.77.200.56]:43076) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1mez5m-00083f-5K; Mon, 25 Oct 2021 08:28:46 -0400 Received: from mailhost.u-ga.fr (mailhost2.u-ga.fr [129.88.177.242]) by zm-mta-out-3.u-ga.fr (Postfix) with ESMTP id D535A4196E; Mon, 25 Oct 2021 14:28:33 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=univ-grenoble-alpes.fr; s=2020; t=1635164913; bh=t7CZ/ykazyQByTEQK67mp0FWUl11/p/ftC8xmY3hk8I=; h=From:To:Cc:Subject:Date:In-Reply-To:References:From; b=R00xFpDZ9q90P0Oz/gAsnUyQtXB6nq+8n2FSi5EOgteBp8Lcucs6B8sUTWnv5HqB7 0j/sXy5QxdT+hPGYfCpKANrGGmZGhdUMED8sZzVDmxZ0Jm8VGiMW2h6QiTVB5C+cGZ yo9dVWCVauiHpQIwu6QYbGTGBHvQJ5BVMCqaVIYdGpvpj8w6TnRADiSBayEfFnv7lB GuyMd6AIWOXTzivOVXFM/rigkq1XPOefMJyl2Y0N/6VhEWSvbU2qBlD9euSrPyeroK qMGhqR533g83blJKluaLY13OiYTE93DXan6FNCeiN2ZdN52VOKDwbqsf8fYQQekOjS CYeK4DpaDYdjg== Received: from smtps.univ-grenoble-alpes.fr (smtps2.u-ga.fr [152.77.18.2]) by mailhost.u-ga.fr (Postfix) with ESMTP id B954960066; Mon, 25 Oct 2021 14:28:33 +0200 (CEST) Received: from palmier.tima.u-ga.fr (35.201.90.79.rev.sfr.net [79.90.201.35]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (2048 bits) server-digest SHA256) (No client certificate requested) (Authenticated sender: petrotf@univ-grenoble-alpes.fr) by smtps.univ-grenoble-alpes.fr (Postfix) with ESMTPSA id 7931B14005A; Mon, 25 Oct 2021 14:28:33 +0200 (CEST) From: =?utf-8?b?RnLDqWTDqXJpYyBQw6l0cm90?= To: qemu-devel@nongnu.org, qemu-riscv@nongnu.org Subject: [PATCH v4 08/17] target/riscv: accessors to registers upper part and 128-bit load/store Date: Mon, 25 Oct 2021 14:28:09 +0200 Message-Id: <20211025122818.168890-9-frederic.petrot@univ-grenoble-alpes.fr> X-Mailer: git-send-email 2.33.0 In-Reply-To: <20211025122818.168890-1-frederic.petrot@univ-grenoble-alpes.fr> References: <20211025122818.168890-1-frederic.petrot@univ-grenoble-alpes.fr> MIME-Version: 1.0 X-Greylist: Whitelist-UGA SMTP Authentifie (petrotf@univ-grenoble-alpes.fr) via submission-587 ACL (41) X-Greylist: Whitelist-UGA MAILHOST (SMTP non authentifie) depuis 152.77.18.2 Received-SPF: pass client-ip=152.77.200.56; envelope-from=frederic.petrot@univ-grenoble-alpes.fr; helo=zm-mta-out-3.u-ga.fr X-Spam_score_int: -20 X-Spam_score: -2.1 X-Spam_bar: -- X-Spam_report: (-2.1 / 5.0 requ) BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_MSPIKE_H2=-0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: bin.meng@windriver.com, richard.henderson@linaro.org, alistair.francis@wdc.com, fabien.portas@grenoble-inp.org, palmer@dabbelt.com, =?utf-8?b?RnLDqWTDqXJpYyBQw6l0cm90?= , philmd@redhat.com Errors-To: qemu-devel-bounces+qemu-devel=archiver.kernel.org@nongnu.org Sender: "Qemu-devel" Get function to retrieve the 64 top bits of a register, stored in the gprh field of the cpu state. Set function that writes the 128-bit value at once. The access to the gprh field can not be protected at compile time to make sure it is accessed only in the 128-bit version of the processor because we have no way to indicate that the misa_mxl_max field is const. The 128-bit ISA adds ldu, lq and sq. We provide support for these instructions. Note that we compte only 64-bit addresses to actually access memory, cowardly utilizing the existing address translation mechanism of QEMU. Signed-off-by: Frédéric Pétrot Co-authored-by: Fabien Portas --- configs/targets/riscv128-softmmu.mak | 3 +- target/riscv/insn16.decode | 27 +++++- target/riscv/insn32.decode | 5 ++ target/riscv/translate.c | 43 +++++++++ target/riscv/insn_trans/trans_rvi.c.inc | 112 ++++++++++++++++++++++-- 5 files changed, 178 insertions(+), 12 deletions(-) diff --git a/configs/targets/riscv128-softmmu.mak b/configs/targets/riscv128-softmmu.mak index a9343d57d3..41daab1cd3 100644 --- a/configs/targets/riscv128-softmmu.mak +++ b/configs/targets/riscv128-softmmu.mak @@ -1,5 +1,6 @@ TARGET_ARCH=riscv128 TARGET_BASE_ARCH=riscv -TARGET_SUPPORTS_MTTCG=y +# As long as we have no atomic accesses for aligned 128-bit addresses +TARGET_SUPPORTS_MTTCG=n TARGET_XML_FILES= gdb-xml/riscv-64bit-cpu.xml gdb-xml/riscv-32bit-fpu.xml gdb-xml/riscv-64bit-fpu.xml gdb-xml/riscv-64bit-virtual.xml TARGET_NEED_FDT=y diff --git a/target/riscv/insn16.decode b/target/riscv/insn16.decode index 2e9212663c..02c8f61b48 100644 --- a/target/riscv/insn16.decode +++ b/target/riscv/insn16.decode @@ -25,14 +25,17 @@ # Immediates: %imm_ci 12:s1 2:5 %nzuimm_ciw 7:4 11:2 5:1 6:1 !function=ex_shift_2 +%uimm_cl_q 10:1 5:2 11:2 !function=ex_shift_4 %uimm_cl_d 5:2 10:3 !function=ex_shift_3 %uimm_cl_w 5:1 10:3 6:1 !function=ex_shift_2 %imm_cb 12:s1 5:2 2:1 10:2 3:2 !function=ex_shift_1 %imm_cj 12:s1 8:1 9:2 6:1 7:1 2:1 11:1 3:3 !function=ex_shift_1 %shimm_6bit 12:1 2:5 !function=ex_rvc_shifti +%uimm_6bit_lq 2:4 12:1 6:1 !function=ex_shift_4 %uimm_6bit_ld 2:3 12:1 5:2 !function=ex_shift_3 %uimm_6bit_lw 2:2 12:1 4:3 !function=ex_shift_2 +%uimm_6bit_sq 7:4 11:2 !function=ex_shift_4 %uimm_6bit_sd 7:3 10:3 !function=ex_shift_3 %uimm_6bit_sw 7:2 9:4 !function=ex_shift_2 @@ -54,16 +57,20 @@ # Formats 16: @cr .... ..... ..... .. &r rs2=%rs2_5 rs1=%rd %rd @ci ... . ..... ..... .. &i imm=%imm_ci rs1=%rd %rd +@cl_q ... . ..... ..... .. &i imm=%uimm_cl_q rs1=%rs1_3 rd=%rs2_3 @cl_d ... ... ... .. ... .. &i imm=%uimm_cl_d rs1=%rs1_3 rd=%rs2_3 @cl_w ... ... ... .. ... .. &i imm=%uimm_cl_w rs1=%rs1_3 rd=%rs2_3 @cs_2 ... ... ... .. ... .. &r rs2=%rs2_3 rs1=%rs1_3 rd=%rs1_3 +@cs_q ... ... ... .. ... .. &s imm=%uimm_cl_q rs1=%rs1_3 rs2=%rs2_3 @cs_d ... ... ... .. ... .. &s imm=%uimm_cl_d rs1=%rs1_3 rs2=%rs2_3 @cs_w ... ... ... .. ... .. &s imm=%uimm_cl_w rs1=%rs1_3 rs2=%rs2_3 @cj ... ........... .. &j imm=%imm_cj @cb_z ... ... ... .. ... .. &b imm=%imm_cb rs1=%rs1_3 rs2=0 +@c_lqsp ... . ..... ..... .. &i imm=%uimm_6bit_lq rs1=2 %rd @c_ldsp ... . ..... ..... .. &i imm=%uimm_6bit_ld rs1=2 %rd @c_lwsp ... . ..... ..... .. &i imm=%uimm_6bit_lw rs1=2 %rd +@c_sqsp ... . ..... ..... .. &s imm=%uimm_6bit_sq rs1=2 rs2=%rs2_5 @c_sdsp ... . ..... ..... .. &s imm=%uimm_6bit_sd rs1=2 rs2=%rs2_5 @c_swsp ... . ..... ..... .. &s imm=%uimm_6bit_sw rs1=2 rs2=%rs2_5 @c_li ... . ..... ..... .. &i imm=%imm_ci rs1=0 %rd @@ -87,9 +94,15 @@ illegal 000 000 000 00 --- 00 addi 000 ... ... .. ... 00 @c_addi4spn } -fld 001 ... ... .. ... 00 @cl_d +{ + lq 001 ... ... .. ... 00 @cl_q + fld 001 ... ... .. ... 00 @cl_d +} lw 010 ... ... .. ... 00 @cl_w -fsd 101 ... ... .. ... 00 @cs_d +{ + sq 101 ... ... .. ... 00 @cs_q + fsd 101 ... ... .. ... 00 @cs_d +} sw 110 ... ... .. ... 00 @cs_w # *** RV32C and RV64C specific Standard Extension (Quadrant 0) *** @@ -132,7 +145,10 @@ addw 100 1 11 ... 01 ... 01 @cs_2 # *** RV32/64C Standard Extension (Quadrant 2) *** slli 000 . ..... ..... 10 @c_shift2 -fld 001 . ..... ..... 10 @c_ldsp +{ + lq 001 ... ... .. ... 10 @c_lqsp + fld 001 . ..... ..... 10 @c_ldsp +} { illegal 010 - 00000 ----- 10 # c.lwsp, RES rd=0 lw 010 . ..... ..... 10 @c_lwsp @@ -147,7 +163,10 @@ fld 001 . ..... ..... 10 @c_ldsp jalr 100 1 ..... 00000 10 @c_jalr rd=1 # C.JALR add 100 1 ..... ..... 10 @cr } -fsd 101 ...... ..... 10 @c_sdsp +{ + sq 101 ... ... .. ... 10 @c_sqsp + fsd 101 ...... ..... 10 @c_sdsp +} sw 110 . ..... ..... 10 @c_swsp # *** RV32C and RV64C specific Standard Extension (Quadrant 2) *** diff --git a/target/riscv/insn32.decode b/target/riscv/insn32.decode index 2f251dac1b..02889c6082 100644 --- a/target/riscv/insn32.decode +++ b/target/riscv/insn32.decode @@ -163,6 +163,11 @@ sllw 0000000 ..... ..... 001 ..... 0111011 @r srlw 0000000 ..... ..... 101 ..... 0111011 @r sraw 0100000 ..... ..... 101 ..... 0111011 @r +# *** RV128I Base Instruction Set (in addition to RV64I) *** +ldu ............ ..... 111 ..... 0000011 @i +lq ............ ..... 010 ..... 0001111 @i +sq ............ ..... 100 ..... 0100011 @s + # *** RV32M Standard Extension *** mul 0000001 ..... ..... 000 ..... 0110011 @r mulh 0000001 ..... ..... 001 ..... 0110011 @r diff --git a/target/riscv/translate.c b/target/riscv/translate.c index 63a52bb707..070db77ca5 100644 --- a/target/riscv/translate.c +++ b/target/riscv/translate.c @@ -55,6 +55,7 @@ typedef struct DisasContext { /* pc_succ_insn points to the instruction following base.pc_next */ target_ulong pc_succ_insn; target_ulong priv_ver; + RISCVMXL misa_mxl_max; RISCVMXL xl; uint32_t misa_ext; uint32_t opcode; @@ -116,6 +117,13 @@ static inline int get_olen(DisasContext *ctx) return 16 << get_ol(ctx); } +/* The maximum register length */ +#ifdef TARGET_RISCV32 +#define get_xl_max(ctx) MXL_RV32 +#else +#define get_xl_max(ctx) ((ctx)->misa_mxl_max) +#endif + /* * RISC-V requires NaN-boxing of narrower width floating point values. * This applies when a 32-bit value is assigned to a 64-bit FP register. @@ -220,6 +228,7 @@ static TCGv get_gpr(DisasContext *ctx, int reg_num, DisasExtend ext) } break; case MXL_RV64: + case MXL_RV128: break; default: g_assert_not_reached(); @@ -227,6 +236,15 @@ static TCGv get_gpr(DisasContext *ctx, int reg_num, DisasExtend ext) return cpu_gpr[reg_num]; } +static TCGv get_gprh(DisasContext *ctx, int reg_num) +{ + assert(get_xl(ctx) == MXL_RV128); + if (reg_num == 0) { + return ctx->zero; + } + return cpu_gprh[reg_num]; +} + static TCGv dest_gpr(DisasContext *ctx, int reg_num) { if (reg_num == 0 || get_olen(ctx) < TARGET_LONG_BITS) { @@ -235,6 +253,14 @@ static TCGv dest_gpr(DisasContext *ctx, int reg_num) return cpu_gpr[reg_num]; } +static TCGv dest_gprh(DisasContext *ctx, int reg_num) +{ + if (reg_num == 0 || get_ol(ctx) < MXL_RV128) { + return temp_new(ctx); + } + return cpu_gprh[reg_num]; +} + static void gen_set_gpr(DisasContext *ctx, int reg_num, TCGv t) { if (reg_num != 0) { @@ -243,11 +269,27 @@ static void gen_set_gpr(DisasContext *ctx, int reg_num, TCGv t) tcg_gen_ext32s_tl(cpu_gpr[reg_num], t); break; case MXL_RV64: + case MXL_RV128: tcg_gen_mov_tl(cpu_gpr[reg_num], t); break; default: g_assert_not_reached(); } + + if (get_xl_max(ctx) == MXL_RV128) { + tcg_gen_sari_tl(cpu_gprh[reg_num], cpu_gpr[reg_num], 63); + } + } +} + +static void gen_set_gpr128(DisasContext *ctx, int reg_num, TCGv rl, TCGv rh) +{ + if (get_ol(ctx) != MXL_RV128) { + g_assert_not_reached(); + } + if (reg_num != 0) { + tcg_gen_mov_tl(cpu_gpr[reg_num], rl); + tcg_gen_mov_tl(cpu_gprh[reg_num], rh); } } @@ -649,6 +691,7 @@ static void riscv_tr_init_disas_context(DisasContextBase *dcbase, CPUState *cs) ctx->lmul = FIELD_EX32(tb_flags, TB_FLAGS, LMUL); ctx->mlen = 1 << (ctx->sew + 3 - ctx->lmul); ctx->vl_eq_vlmax = FIELD_EX32(tb_flags, TB_FLAGS, VL_EQ_VLMAX); + ctx->misa_mxl_max = env->misa_mxl_max; ctx->xl = FIELD_EX32(tb_flags, TB_FLAGS, XL); ctx->cs = cs; ctx->ntemp = 0; diff --git a/target/riscv/insn_trans/trans_rvi.c.inc b/target/riscv/insn_trans/trans_rvi.c.inc index 34164494b9..d17bde6a3a 100644 --- a/target/riscv/insn_trans/trans_rvi.c.inc +++ b/target/riscv/insn_trans/trans_rvi.c.inc @@ -134,7 +134,7 @@ static bool trans_bgeu(DisasContext *ctx, arg_bgeu *a) return gen_branch(ctx, a, TCG_COND_GEU); } -static bool gen_load(DisasContext *ctx, arg_lb *a, MemOp memop) +static bool gen_load_tl(DisasContext *ctx, arg_lb *a, MemOp memop) { TCGv dest = dest_gpr(ctx, a->rd); TCGv addr = get_gpr(ctx, a->rs1, EXT_NONE); @@ -150,6 +150,50 @@ static bool gen_load(DisasContext *ctx, arg_lb *a, MemOp memop) return true; } +/* Compute only 64-bit addresses to use the address translation mechanism */ +static bool gen_load_i128(DisasContext *ctx, arg_lb *a, MemOp memop) +{ + TCGv src1l = get_gpr(ctx, a->rs1, EXT_NONE); + TCGv destl = dest_gpr(ctx, a->rd); + TCGv desth = dest_gprh(ctx, a->rd); + TCGv addrl = tcg_temp_new(); + + if (a->imm != 0) { + tcg_gen_addi_tl(addrl, src1l, a->imm); + } else { + tcg_gen_mov_tl(addrl, src1l); + } + + if (memop != MO_TEO) { + tcg_gen_qemu_ld_tl(destl, addrl, ctx->mem_idx, memop); + if (memop & MO_SIGN) { + tcg_gen_sari_tl(desth, destl, 63); + } else { + tcg_gen_movi_tl(desth, 0); + } + } else { + tcg_gen_qemu_ld_tl(memop & MO_BSWAP ? desth : destl, addrl, + ctx->mem_idx, MO_TEQ); + tcg_gen_addi_tl(addrl, addrl, 8); + tcg_gen_qemu_ld_tl(memop & MO_BSWAP ? destl : desth, addrl, + ctx->mem_idx, MO_TEQ); + } + + gen_set_gpr128(ctx, a->rd, destl, desth); + + tcg_temp_free(addrl); + return true; +} + +static bool gen_load(DisasContext *ctx, arg_lb *a, MemOp memop) +{ + if (get_xl(ctx) == MXL_RV128) { + return gen_load_i128(ctx, a, memop); + } else { + return gen_load_tl(ctx, a, memop); + } +} + static bool trans_lb(DisasContext *ctx, arg_lb *a) { return gen_load(ctx, a, MO_SB); @@ -165,6 +209,18 @@ static bool trans_lw(DisasContext *ctx, arg_lw *a) return gen_load(ctx, a, MO_TESL); } +static bool trans_ld(DisasContext *ctx, arg_ld *a) +{ + REQUIRE_64_OR_128BIT(ctx); + return gen_load(ctx, a, MO_TESQ); +} + +static bool trans_lq(DisasContext *ctx, arg_lq *a) +{ + REQUIRE_128BIT(ctx); + return gen_load(ctx, a, MO_TEO); +} + static bool trans_lbu(DisasContext *ctx, arg_lbu *a) { return gen_load(ctx, a, MO_UB); @@ -177,17 +233,17 @@ static bool trans_lhu(DisasContext *ctx, arg_lhu *a) static bool trans_lwu(DisasContext *ctx, arg_lwu *a) { - REQUIRE_64BIT(ctx); + REQUIRE_64_OR_128BIT(ctx); return gen_load(ctx, a, MO_TEUL); } -static bool trans_ld(DisasContext *ctx, arg_ld *a) +static bool trans_ldu(DisasContext *ctx, arg_ldu *a) { - REQUIRE_64BIT(ctx); - return gen_load(ctx, a, MO_TEQ); + REQUIRE_128BIT(ctx); + return gen_load(ctx, a, MO_TEUQ); } -static bool gen_store(DisasContext *ctx, arg_sb *a, MemOp memop) +static bool gen_store_tl(DisasContext *ctx, arg_sb *a, MemOp memop) { TCGv addr = get_gpr(ctx, a->rs1, EXT_NONE); TCGv data = get_gpr(ctx, a->rs2, EXT_NONE); @@ -202,6 +258,42 @@ static bool gen_store(DisasContext *ctx, arg_sb *a, MemOp memop) return true; } +static bool gen_store_i128(DisasContext *ctx, arg_sb *a, MemOp memop) +{ + TCGv src1l = get_gpr(ctx, a->rs1, EXT_NONE); + TCGv src2l = get_gpr(ctx, a->rs2, EXT_NONE); + TCGv src2h = get_gprh(ctx, a->rs2); + TCGv addrl = tcg_temp_new(); + + if (a->imm != 0) { + tcg_gen_addi_tl(addrl, src1l, a->imm); + } else { + tcg_gen_mov_tl(addrl, src1l); + } + + if (memop != MO_TEO) { + tcg_gen_qemu_st_tl(src2l, addrl, ctx->mem_idx, memop); + } else { + tcg_gen_qemu_st_tl(memop & MO_BSWAP ? src2h : src2l, addrl, + ctx->mem_idx, MO_TEQ); + tcg_gen_addi_tl(addrl, addrl, 8); + tcg_gen_qemu_st_tl(memop & MO_BSWAP ? src2l : src2h, addrl, + ctx->mem_idx, MO_TEQ); + } + + tcg_temp_free(addrl); + return true; +} + +static bool gen_store(DisasContext *ctx, arg_sb *a, MemOp memop) +{ + if (get_xl(ctx) == MXL_RV128) { + return gen_store_i128(ctx, a, memop); + } else { + return gen_store_tl(ctx, a, memop); + } +} + static bool trans_sb(DisasContext *ctx, arg_sb *a) { return gen_store(ctx, a, MO_SB); @@ -219,10 +311,16 @@ static bool trans_sw(DisasContext *ctx, arg_sw *a) static bool trans_sd(DisasContext *ctx, arg_sd *a) { - REQUIRE_64BIT(ctx); + REQUIRE_64_OR_128BIT(ctx); return gen_store(ctx, a, MO_TEQ); } +static bool trans_sq(DisasContext *ctx, arg_sq *a) +{ + REQUIRE_128BIT(ctx); + return gen_store(ctx, a, MO_TEO); +} + static bool trans_addi(DisasContext *ctx, arg_addi *a) { return gen_arith_imm_fn(ctx, a, EXT_NONE, tcg_gen_addi_tl); From patchwork Mon Oct 25 12:28:10 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Patchwork-Submitter: =?utf-8?b?RnLDqWTDqXJpYyBQw6l0cm90?= X-Patchwork-Id: 12581769 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 2C7B9C433EF for ; Mon, 25 Oct 2021 12:52:56 +0000 (UTC) Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.kernel.org (Postfix) with ESMTPS id DDE5060F22 for ; Mon, 25 Oct 2021 12:52:55 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.4.1 mail.kernel.org DDE5060F22 Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=univ-grenoble-alpes.fr Authentication-Results: mail.kernel.org; spf=pass smtp.mailfrom=nongnu.org Received: from localhost ([::1]:42122 helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1mezTC-0007KE-QV for qemu-devel@archiver.kernel.org; Mon, 25 Oct 2021 08:52:54 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]:53054) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1mez5q-0005tP-0B; Mon, 25 Oct 2021 08:28:47 -0400 Received: from zm-mta-out-3.u-ga.fr ([152.77.200.56]:43092) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1mez5n-00085P-Jq; Mon, 25 Oct 2021 08:28:45 -0400 Received: from mailhost.u-ga.fr (mailhost2.u-ga.fr [129.88.177.242]) by zm-mta-out-3.u-ga.fr (Postfix) with ESMTP id 1A4C141991; Mon, 25 Oct 2021 14:28:34 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=univ-grenoble-alpes.fr; s=2020; t=1635164914; bh=/bNWm2CFEsWTQuQiFZGa3cZy0Xiybgs7M/tdpEV7Gfk=; h=From:To:Cc:Subject:Date:In-Reply-To:References:From; b=EEz5ZtKorV25bZmhLmhJuCYeegAukgyug9yXX5Z8opKqpHc5FVJONYhZO7P5SA8iP clUMU5wIhMN2cDJEMTQOX5opnJ5/1y9UWW8JDpZdkB8lZ2k76zqG5h1zVEZLBdDKzQ hQyudLm249U6SvO7olutwTc2bCXOSkpS0nfrc1NK2Du05lZTPZb5GW93zYIkMudeKt B37H7iOMu6Lb+hvLDEgWpY1dcChZcNaTvMl3sSblnlhw9GaLB8IMpegjMx+/KNJeyO dZZ0+MUrZx4f3zKw73+p2EA0NdVgDOP2U+v9A6sfL11AsV+Q3fqVPXi6HZsmX9nrsa LHJ1vmNqPr2cQ== Received: from smtps.univ-grenoble-alpes.fr (smtps2.u-ga.fr [152.77.18.2]) by mailhost.u-ga.fr (Postfix) with ESMTP id F22BD60066; Mon, 25 Oct 2021 14:28:33 +0200 (CEST) Received: from palmier.tima.u-ga.fr (35.201.90.79.rev.sfr.net [79.90.201.35]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (2048 bits) server-digest SHA256) (No client certificate requested) (Authenticated sender: petrotf@univ-grenoble-alpes.fr) by smtps.univ-grenoble-alpes.fr (Postfix) with ESMTPSA id C02A114005D; Mon, 25 Oct 2021 14:28:33 +0200 (CEST) From: =?utf-8?b?RnLDqWTDqXJpYyBQw6l0cm90?= To: qemu-devel@nongnu.org, qemu-riscv@nongnu.org Subject: [PATCH v4 09/17] target/riscv: support for 128-bit bitwise instructions Date: Mon, 25 Oct 2021 14:28:10 +0200 Message-Id: <20211025122818.168890-10-frederic.petrot@univ-grenoble-alpes.fr> X-Mailer: git-send-email 2.33.0 In-Reply-To: <20211025122818.168890-1-frederic.petrot@univ-grenoble-alpes.fr> References: <20211025122818.168890-1-frederic.petrot@univ-grenoble-alpes.fr> MIME-Version: 1.0 X-Greylist: Whitelist-UGA SMTP Authentifie (petrotf@univ-grenoble-alpes.fr) via submission-587 ACL (41) X-Greylist: Whitelist-UGA MAILHOST (SMTP non authentifie) depuis 152.77.18.2 Received-SPF: pass client-ip=152.77.200.56; envelope-from=frederic.petrot@univ-grenoble-alpes.fr; helo=zm-mta-out-3.u-ga.fr X-Spam_score_int: -20 X-Spam_score: -2.1 X-Spam_bar: -- X-Spam_report: (-2.1 / 5.0 requ) BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_MSPIKE_H2=-0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: bin.meng@windriver.com, richard.henderson@linaro.org, alistair.francis@wdc.com, fabien.portas@grenoble-inp.org, palmer@dabbelt.com, =?utf-8?b?RnLDqWTDqXJpYyBQw6l0cm90?= , philmd@redhat.com Errors-To: qemu-devel-bounces+qemu-devel=archiver.kernel.org@nongnu.org Sender: "Qemu-devel" The 128-bit bitwise instructions do not need any function prototype change as the functions can be applied independently on the lower and upper part of the registers. Signed-off-by: Frédéric Pétrot Co-authored-by: Fabien Portas Reviewed-by: Richard Henderson --- target/riscv/translate.c | 21 +++++++++++++++++++-- 1 file changed, 19 insertions(+), 2 deletions(-) diff --git a/target/riscv/translate.c b/target/riscv/translate.c index 070db77ca5..b183ad2b6e 100644 --- a/target/riscv/translate.c +++ b/target/riscv/translate.c @@ -426,7 +426,15 @@ static bool gen_logic_imm_fn(DisasContext *ctx, arg_i *a, func(dest, src1, a->imm); - gen_set_gpr(ctx, a->rd, dest); + if (get_xl(ctx) == MXL_RV128) { + TCGv src1h = get_gprh(ctx, a->rs1); + TCGv desth = dest_gprh(ctx, a->rd); + + func(desth, src1h, -(a->imm < 0)); + gen_set_gpr128(ctx, a->rd, dest, desth); + } else { + gen_set_gpr(ctx, a->rd, dest); + } return true; } @@ -440,7 +448,16 @@ static bool gen_logic(DisasContext *ctx, arg_r *a, DisasExtend ext, func(dest, src1, src2); - gen_set_gpr(ctx, a->rd, dest); + if (get_xl(ctx) == MXL_RV128) { + TCGv src1h = get_gprh(ctx, a->rs1); + TCGv src2h = get_gprh(ctx, a->rs2); + TCGv desth = dest_gprh(ctx, a->rd); + + func(desth, src1h, src2h); + gen_set_gpr128(ctx, a->rd, dest, desth); + } else { + gen_set_gpr(ctx, a->rd, dest); + } return true; } From patchwork Mon Oct 25 12:28:11 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Patchwork-Submitter: =?utf-8?b?RnLDqWTDqXJpYyBQw6l0cm90?= X-Patchwork-Id: 12581747 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id B3CB1C433F5 for ; Mon, 25 Oct 2021 12:39:28 +0000 (UTC) Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.kernel.org (Postfix) with ESMTPS id 53B5260EFE for ; Mon, 25 Oct 2021 12:39:28 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.4.1 mail.kernel.org 53B5260EFE Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=univ-grenoble-alpes.fr Authentication-Results: mail.kernel.org; spf=pass smtp.mailfrom=nongnu.org Received: from localhost ([::1]:38090 helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1mezGB-0001mw-Ed for qemu-devel@archiver.kernel.org; Mon, 25 Oct 2021 08:39:27 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]:53164) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1mez6A-0006Pc-1S; Mon, 25 Oct 2021 08:29:06 -0400 Received: from zm-mta-out-3.u-ga.fr ([152.77.200.56]:43090) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1mez67-00085R-L9; Mon, 25 Oct 2021 08:29:05 -0400 Received: from mailhost.u-ga.fr (mailhost1.u-ga.fr [152.77.1.10]) by zm-mta-out-3.u-ga.fr (Postfix) with ESMTP id 5891941A83; Mon, 25 Oct 2021 14:28:34 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=univ-grenoble-alpes.fr; s=2020; t=1635164914; bh=CDLOH5KLJqjjJyhk64e79ajNgoRDyTEZjw1WAtSm7f8=; h=From:To:Cc:Subject:Date:In-Reply-To:References:From; b=P9Kr2IIZCX7IUzHiCWlsrcHHBiX142Yg/9BpINC3Ny6aC44UTqbd0i2CZYgETjPdk 6H4guU3zperEsHsCEmNLbLWHoJhPLZF5CVHVhbQu7q6DmxeCKgRQxh1wKYPagnjz4R 656L0lhLG87Cfqmui5zWOhwIMI6EFXfQVLq4UXkF7dudTDq/mywYRRxseCkULEuUaz iklFh+s4qVE29qf+9+LzMofl/iQF120w6QPD3o/kTEE4lq2of8U9LYwgKuZUbPHEEh zkVlmSdIjets42brdP6zHNPKI/cDJZ0d3d6bcPT2vKDEE6MTOUdr0GXaimyOkx7Yzv OQ+odGfW5tBLg== Received: from smtps.univ-grenoble-alpes.fr (smtps2.u-ga.fr [152.77.18.2]) by mailhost.u-ga.fr (Postfix) with ESMTP id 3600A60070; Mon, 25 Oct 2021 14:28:34 +0200 (CEST) Received: from palmier.tima.u-ga.fr (35.201.90.79.rev.sfr.net [79.90.201.35]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (2048 bits) server-digest SHA256) (No client certificate requested) (Authenticated sender: petrotf@univ-grenoble-alpes.fr) by smtps.univ-grenoble-alpes.fr (Postfix) with ESMTPSA id 01DA414005A; Mon, 25 Oct 2021 14:28:33 +0200 (CEST) From: =?utf-8?b?RnLDqWTDqXJpYyBQw6l0cm90?= To: qemu-devel@nongnu.org, qemu-riscv@nongnu.org Subject: [PATCH v4 10/17] target/riscv: support for 128-bit U-type instructions Date: Mon, 25 Oct 2021 14:28:11 +0200 Message-Id: <20211025122818.168890-11-frederic.petrot@univ-grenoble-alpes.fr> X-Mailer: git-send-email 2.33.0 In-Reply-To: <20211025122818.168890-1-frederic.petrot@univ-grenoble-alpes.fr> References: <20211025122818.168890-1-frederic.petrot@univ-grenoble-alpes.fr> MIME-Version: 1.0 X-Greylist: Whitelist-UGA SMTP Authentifie (petrotf@univ-grenoble-alpes.fr) via submission-587 ACL (41) X-Greylist: Whitelist-UGA MAILHOST (SMTP non authentifie) depuis 152.77.18.2 Received-SPF: pass client-ip=152.77.200.56; envelope-from=frederic.petrot@univ-grenoble-alpes.fr; helo=zm-mta-out-3.u-ga.fr X-Spam_score_int: -20 X-Spam_score: -2.1 X-Spam_bar: -- X-Spam_report: (-2.1 / 5.0 requ) BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_MSPIKE_H2=-0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: bin.meng@windriver.com, richard.henderson@linaro.org, alistair.francis@wdc.com, fabien.portas@grenoble-inp.org, palmer@dabbelt.com, =?utf-8?b?RnLDqWTDqXJpYyBQw6l0cm90?= , philmd@redhat.com Errors-To: qemu-devel-bounces+qemu-devel=archiver.kernel.org@nongnu.org Sender: "Qemu-devel" Adding the 128-bit version of lui and auipc, and introducing to that end a set register with immediat function to handle extension on 128 bits. Signed-off-by: Frédéric Pétrot Co-authored-by: Fabien Portas Reviewed-by: Richard Henderson --- target/riscv/translate.c | 22 ++++++++++++++++++++++ target/riscv/insn_trans/trans_rvi.c.inc | 8 ++++---- 2 files changed, 26 insertions(+), 4 deletions(-) diff --git a/target/riscv/translate.c b/target/riscv/translate.c index b183ad2b6e..ab706d799a 100644 --- a/target/riscv/translate.c +++ b/target/riscv/translate.c @@ -282,6 +282,28 @@ static void gen_set_gpr(DisasContext *ctx, int reg_num, TCGv t) } } +static void gen_set_gpri(DisasContext *ctx, int reg_num, target_long imm) +{ + if (reg_num != 0) { + switch (get_ol(ctx)) { + case MXL_RV32: + tcg_gen_movi_tl(cpu_gpr[reg_num], imm); + tcg_gen_ext32s_tl(cpu_gpr[reg_num], cpu_gpr[reg_num]); + break; + case MXL_RV64: + case MXL_RV128: + tcg_gen_movi_tl(cpu_gpr[reg_num], imm); + break; + default: + g_assert_not_reached(); + } + + if (get_xl_max(ctx) == MXL_RV128) { + tcg_gen_movi_tl(cpu_gprh[reg_num], -(imm < 0)); + } + } +} + static void gen_set_gpr128(DisasContext *ctx, int reg_num, TCGv rl, TCGv rh) { if (get_ol(ctx) != MXL_RV128) { diff --git a/target/riscv/insn_trans/trans_rvi.c.inc b/target/riscv/insn_trans/trans_rvi.c.inc index d17bde6a3a..3222a45d72 100644 --- a/target/riscv/insn_trans/trans_rvi.c.inc +++ b/target/riscv/insn_trans/trans_rvi.c.inc @@ -26,14 +26,14 @@ static bool trans_illegal(DisasContext *ctx, arg_empty *a) static bool trans_c64_illegal(DisasContext *ctx, arg_empty *a) { - REQUIRE_64BIT(ctx); - return trans_illegal(ctx, a); + REQUIRE_64_OR_128BIT(ctx); + return trans_illegal(ctx, a); } static bool trans_lui(DisasContext *ctx, arg_lui *a) { if (a->rd != 0) { - tcg_gen_movi_tl(cpu_gpr[a->rd], a->imm); + gen_set_gpri(ctx, a->rd, a->imm); } return true; } @@ -41,7 +41,7 @@ static bool trans_lui(DisasContext *ctx, arg_lui *a) static bool trans_auipc(DisasContext *ctx, arg_auipc *a) { if (a->rd != 0) { - tcg_gen_movi_tl(cpu_gpr[a->rd], a->imm + ctx->base.pc_next); + gen_set_gpri(ctx, a->rd, a->imm + ctx->base.pc_next); } return true; } From patchwork Mon Oct 25 12:28:12 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Patchwork-Submitter: =?utf-8?b?RnLDqWTDqXJpYyBQw6l0cm90?= X-Patchwork-Id: 12581791 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 0497DC433F5 for ; Mon, 25 Oct 2021 12:58:53 +0000 (UTC) Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.kernel.org (Postfix) with ESMTPS id 74D0561027 for ; Mon, 25 Oct 2021 12:58:52 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.4.1 mail.kernel.org 74D0561027 Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=univ-grenoble-alpes.fr Authentication-Results: mail.kernel.org; spf=pass smtp.mailfrom=nongnu.org Received: from localhost ([::1]:60546 helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1mezYx-0002kc-Fw for qemu-devel@archiver.kernel.org; Mon, 25 Oct 2021 08:58:51 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]:53214) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1mez6C-0006Td-QZ; Mon, 25 Oct 2021 08:29:09 -0400 Received: from zm-mta-out-3.u-ga.fr ([152.77.200.56]:43094) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1mez68-000875-PJ; Mon, 25 Oct 2021 08:29:08 -0400 Received: from mailhost.u-ga.fr (mailhost2.u-ga.fr [129.88.177.242]) by zm-mta-out-3.u-ga.fr (Postfix) with ESMTP id 931F641ACF; Mon, 25 Oct 2021 14:28:34 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=univ-grenoble-alpes.fr; s=2020; t=1635164914; bh=oCR/VksJfPKNYaM1ZADue40F7V/0yfLsjDceAhJ5gmE=; h=From:To:Cc:Subject:Date:In-Reply-To:References:From; b=rNfpXnnYsqJhprB/nvrusNDIVGqfinYtd83xNb3NY5hNw2cpbgpiwbldc8kxObqfm bBAJQ8D9uwaZT9PhtDQaiU5edYbW1SNoZZTVd5FXPFj+lDy1Xm5KnJnZFGZI6JiOf1 p9fQ5KMtLRbwKsNh3DbEmNpn8j0FSIpn7DI/h4AP/AmcvpxgCGP4xq1r3NJCDRCkuC B93cwK21qifpCRMu4xT6Bv7wBv06Uq9GvdxVJBWVtb1ZGq+4vjpaTJjqGkPKROuVyx j4tu1OsYRry0qToVhOSRY3NzqOzccEgvo1WOFLWhPThT7Ip0eanYTcirI95i/7Jfpf fQtywJaZCmJGA== Received: from smtps.univ-grenoble-alpes.fr (smtps2.u-ga.fr [152.77.18.2]) by mailhost.u-ga.fr (Postfix) with ESMTP id 7749860066; Mon, 25 Oct 2021 14:28:34 +0200 (CEST) Received: from palmier.tima.u-ga.fr (35.201.90.79.rev.sfr.net [79.90.201.35]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (2048 bits) server-digest SHA256) (No client certificate requested) (Authenticated sender: petrotf@univ-grenoble-alpes.fr) by smtps.univ-grenoble-alpes.fr (Postfix) with ESMTPSA id 3E6D914005D; Mon, 25 Oct 2021 14:28:34 +0200 (CEST) From: =?utf-8?b?RnLDqWTDqXJpYyBQw6l0cm90?= To: qemu-devel@nongnu.org, qemu-riscv@nongnu.org Subject: [PATCH v4 11/17] target/riscv: support for 128-bit shift instructions Date: Mon, 25 Oct 2021 14:28:12 +0200 Message-Id: <20211025122818.168890-12-frederic.petrot@univ-grenoble-alpes.fr> X-Mailer: git-send-email 2.33.0 In-Reply-To: <20211025122818.168890-1-frederic.petrot@univ-grenoble-alpes.fr> References: <20211025122818.168890-1-frederic.petrot@univ-grenoble-alpes.fr> MIME-Version: 1.0 X-Greylist: Whitelist-UGA SMTP Authentifie (petrotf@univ-grenoble-alpes.fr) via submission-587 ACL (41) X-Greylist: Whitelist-UGA MAILHOST (SMTP non authentifie) depuis 152.77.18.2 Received-SPF: pass client-ip=152.77.200.56; envelope-from=frederic.petrot@univ-grenoble-alpes.fr; helo=zm-mta-out-3.u-ga.fr X-Spam_score_int: -1 X-Spam_score: -0.2 X-Spam_bar: / X-Spam_report: (-0.2 / 5.0 requ) DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_MSPIKE_H2=-0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: bin.meng@windriver.com, richard.henderson@linaro.org, alistair.francis@wdc.com, fabien.portas@grenoble-inp.org, palmer@dabbelt.com, =?utf-8?b?RnLDqWTDqXJpYyBQw6l0cm90?= , philmd@redhat.com Errors-To: qemu-devel-bounces+qemu-devel=archiver.kernel.org@nongnu.org Sender: "Qemu-devel" Handling shifts for 32, 64 and 128 operation length for RV128, following the general framework for handling various olens proposed by Richard. Signed-off-by: Frédéric Pétrot Co-authored-by: Fabien Portas --- target/riscv/insn32.decode | 10 ++ target/riscv/translate.c | 55 ++++-- target/riscv/insn_trans/trans_rvb.c.inc | 22 +-- target/riscv/insn_trans/trans_rvi.c.inc | 225 ++++++++++++++++++++++-- 4 files changed, 269 insertions(+), 43 deletions(-) diff --git a/target/riscv/insn32.decode b/target/riscv/insn32.decode index 02889c6082..e338a803a0 100644 --- a/target/riscv/insn32.decode +++ b/target/riscv/insn32.decode @@ -22,6 +22,7 @@ %rs1 15:5 %rd 7:5 %sh5 20:5 +%sh6 20:6 %sh7 20:7 %csr 20:12 @@ -92,6 +93,9 @@ # Formats 64: @sh5 ....... ..... ..... ... ..... ....... &shift shamt=%sh5 %rs1 %rd +# Formats 128: +@sh6 ...... ...... ..... ... ..... ....... &shift shamt=%sh6 %rs1 %rd + # *** Privileged Instructions *** ecall 000000000000 00000 000 00000 1110011 ebreak 000000000001 00000 000 00000 1110011 @@ -167,6 +171,12 @@ sraw 0100000 ..... ..... 101 ..... 0111011 @r ldu ............ ..... 111 ..... 0000011 @i lq ............ ..... 010 ..... 0001111 @i sq ............ ..... 100 ..... 0100011 @s +sllid 000000 ...... ..... 001 ..... 1011011 @sh6 +srlid 000000 ...... ..... 101 ..... 1011011 @sh6 +sraid 010000 ...... ..... 101 ..... 1011011 @sh6 +slld 0000000 ..... ..... 001 ..... 1111011 @r +srld 0000000 ..... ..... 101 ..... 1111011 @r +srad 0100000 ..... ..... 101 ..... 1111011 @r # *** RV32M Standard Extension *** mul 0000001 ..... ..... 000 ..... 0110011 @r diff --git a/target/riscv/translate.c b/target/riscv/translate.c index ab706d799a..36f62651ef 100644 --- a/target/riscv/translate.c +++ b/target/riscv/translate.c @@ -539,7 +539,8 @@ static bool gen_arith_per_ol(DisasContext *ctx, arg_r *a, DisasExtend ext, } static bool gen_shift_imm_fn(DisasContext *ctx, arg_shift *a, DisasExtend ext, - void (*func)(TCGv, TCGv, target_long)) + void (*func)(TCGv, TCGv, target_long), + void (*f128)(TCGv, TCGv, TCGv, TCGv, target_long)) { TCGv dest, src1; int max_len = get_olen(ctx); @@ -551,26 +552,38 @@ static bool gen_shift_imm_fn(DisasContext *ctx, arg_shift *a, DisasExtend ext, dest = dest_gpr(ctx, a->rd); src1 = get_gpr(ctx, a->rs1, ext); - func(dest, src1, a->shamt); + if (get_xl(ctx) < MXL_RV128 || max_len < 128) { + func(dest, src1, a->shamt); + gen_set_gpr(ctx, a->rd, dest); + } else { + TCGv src1h = get_gprh(ctx, a->rs1); + TCGv desth = dest_gprh(ctx, a->rd); - gen_set_gpr(ctx, a->rd, dest); + if (f128 == NULL) { + return false; + } + f128(dest, desth, src1, src1h, a->shamt); + gen_set_gpr128(ctx, a->rd, dest, desth); + } return true; } static bool gen_shift_imm_fn_per_ol(DisasContext *ctx, arg_shift *a, DisasExtend ext, void (*f_tl)(TCGv, TCGv, target_long), - void (*f_32)(TCGv, TCGv, target_long)) + void (*f_32)(TCGv, TCGv, target_long), + void (*f_128)(TCGv, TCGv, TCGv, TCGv, + target_long)) { int olen = get_olen(ctx); if (olen != TARGET_LONG_BITS) { if (olen == 32) { f_tl = f_32; - } else { + } else if (olen != 128) { g_assert_not_reached(); } } - return gen_shift_imm_fn(ctx, a, ext, f_tl); + return gen_shift_imm_fn(ctx, a, ext, f_tl, f_128); } static bool gen_shift_imm_tl(DisasContext *ctx, arg_shift *a, DisasExtend ext, @@ -594,34 +607,48 @@ static bool gen_shift_imm_tl(DisasContext *ctx, arg_shift *a, DisasExtend ext, } static bool gen_shift(DisasContext *ctx, arg_r *a, DisasExtend ext, - void (*func)(TCGv, TCGv, TCGv)) + void (*func)(TCGv, TCGv, TCGv), + void (*f128)(TCGv, TCGv, TCGv, TCGv, TCGv)) { - TCGv dest = dest_gpr(ctx, a->rd); - TCGv src1 = get_gpr(ctx, a->rs1, ext); TCGv src2 = get_gpr(ctx, a->rs2, EXT_NONE); TCGv ext2 = tcg_temp_new(); tcg_gen_andi_tl(ext2, src2, get_olen(ctx) - 1); - func(dest, src1, ext2); - gen_set_gpr(ctx, a->rd, dest); + TCGv dest = dest_gpr(ctx, a->rd); + TCGv src1 = get_gpr(ctx, a->rs1, ext); + + if (get_xl(ctx) < MXL_RV128 || get_ol(ctx) < MXL_RV128) { + func(dest, src1, ext2); + gen_set_gpr(ctx, a->rd, dest); + } else { + TCGv src1h = get_gprh(ctx, a->rs1); + TCGv desth = dest_gprh(ctx, a->rd); + + if (f128 == NULL) { + return false; + } + f128(dest, desth, src1, src1h, ext2); + gen_set_gpr128(ctx, a->rd, dest, desth); + } tcg_temp_free(ext2); return true; } static bool gen_shift_per_ol(DisasContext *ctx, arg_r *a, DisasExtend ext, void (*f_tl)(TCGv, TCGv, TCGv), - void (*f_32)(TCGv, TCGv, TCGv)) + void (*f_32)(TCGv, TCGv, TCGv), + void (*f_128)(TCGv, TCGv, TCGv, TCGv, TCGv)) { int olen = get_olen(ctx); if (olen != TARGET_LONG_BITS) { if (olen == 32) { f_tl = f_32; - } else { + } else if (olen != 128) { g_assert_not_reached(); } } - return gen_shift(ctx, a, ext, f_tl); + return gen_shift(ctx, a, ext, f_tl, f_128); } static bool gen_unary(DisasContext *ctx, arg_r2 *a, DisasExtend ext, diff --git a/target/riscv/insn_trans/trans_rvb.c.inc b/target/riscv/insn_trans/trans_rvb.c.inc index acdcfdb178..7b1686022a 100644 --- a/target/riscv/insn_trans/trans_rvb.c.inc +++ b/target/riscv/insn_trans/trans_rvb.c.inc @@ -156,7 +156,7 @@ static void gen_bset(TCGv ret, TCGv arg1, TCGv shamt) static bool trans_bset(DisasContext *ctx, arg_bset *a) { REQUIRE_ZBS(ctx); - return gen_shift(ctx, a, EXT_NONE, gen_bset); + return gen_shift(ctx, a, EXT_NONE, gen_bset, NULL); } static bool trans_bseti(DisasContext *ctx, arg_bseti *a) @@ -178,7 +178,7 @@ static void gen_bclr(TCGv ret, TCGv arg1, TCGv shamt) static bool trans_bclr(DisasContext *ctx, arg_bclr *a) { REQUIRE_ZBS(ctx); - return gen_shift(ctx, a, EXT_NONE, gen_bclr); + return gen_shift(ctx, a, EXT_NONE, gen_bclr, NULL); } static bool trans_bclri(DisasContext *ctx, arg_bclri *a) @@ -200,7 +200,7 @@ static void gen_binv(TCGv ret, TCGv arg1, TCGv shamt) static bool trans_binv(DisasContext *ctx, arg_binv *a) { REQUIRE_ZBS(ctx); - return gen_shift(ctx, a, EXT_NONE, gen_binv); + return gen_shift(ctx, a, EXT_NONE, gen_binv, NULL); } static bool trans_binvi(DisasContext *ctx, arg_binvi *a) @@ -218,7 +218,7 @@ static void gen_bext(TCGv ret, TCGv arg1, TCGv shamt) static bool trans_bext(DisasContext *ctx, arg_bext *a) { REQUIRE_ZBS(ctx); - return gen_shift(ctx, a, EXT_NONE, gen_bext); + return gen_shift(ctx, a, EXT_NONE, gen_bext, NULL); } static bool trans_bexti(DisasContext *ctx, arg_bexti *a) @@ -248,7 +248,7 @@ static void gen_rorw(TCGv ret, TCGv arg1, TCGv arg2) static bool trans_ror(DisasContext *ctx, arg_ror *a) { REQUIRE_ZBB(ctx); - return gen_shift_per_ol(ctx, a, EXT_NONE, tcg_gen_rotr_tl, gen_rorw); + return gen_shift_per_ol(ctx, a, EXT_NONE, tcg_gen_rotr_tl, gen_rorw, NULL); } static void gen_roriw(TCGv ret, TCGv arg1, target_long shamt) @@ -266,7 +266,7 @@ static bool trans_rori(DisasContext *ctx, arg_rori *a) { REQUIRE_ZBB(ctx); return gen_shift_imm_fn_per_ol(ctx, a, EXT_NONE, - tcg_gen_rotri_tl, gen_roriw); + tcg_gen_rotri_tl, gen_roriw, NULL); } static void gen_rolw(TCGv ret, TCGv arg1, TCGv arg2) @@ -290,7 +290,7 @@ static void gen_rolw(TCGv ret, TCGv arg1, TCGv arg2) static bool trans_rol(DisasContext *ctx, arg_rol *a) { REQUIRE_ZBB(ctx); - return gen_shift_per_ol(ctx, a, EXT_NONE, tcg_gen_rotl_tl, gen_rolw); + return gen_shift_per_ol(ctx, a, EXT_NONE, tcg_gen_rotl_tl, gen_rolw, NULL); } static void gen_rev8_32(TCGv ret, TCGv src1) @@ -405,7 +405,7 @@ static bool trans_rorw(DisasContext *ctx, arg_rorw *a) REQUIRE_64BIT(ctx); REQUIRE_ZBB(ctx); ctx->ol = MXL_RV32; - return gen_shift(ctx, a, EXT_NONE, gen_rorw); + return gen_shift(ctx, a, EXT_NONE, gen_rorw, NULL); } static bool trans_roriw(DisasContext *ctx, arg_roriw *a) @@ -413,7 +413,7 @@ static bool trans_roriw(DisasContext *ctx, arg_roriw *a) REQUIRE_64BIT(ctx); REQUIRE_ZBB(ctx); ctx->ol = MXL_RV32; - return gen_shift_imm_fn(ctx, a, EXT_NONE, gen_roriw); + return gen_shift_imm_fn(ctx, a, EXT_NONE, gen_roriw, NULL); } static bool trans_rolw(DisasContext *ctx, arg_rolw *a) @@ -421,7 +421,7 @@ static bool trans_rolw(DisasContext *ctx, arg_rolw *a) REQUIRE_64BIT(ctx); REQUIRE_ZBB(ctx); ctx->ol = MXL_RV32; - return gen_shift(ctx, a, EXT_NONE, gen_rolw); + return gen_shift(ctx, a, EXT_NONE, gen_rolw, NULL); } #define GEN_SHADD_UW(SHAMT) \ @@ -478,7 +478,7 @@ static bool trans_slli_uw(DisasContext *ctx, arg_slli_uw *a) { REQUIRE_64BIT(ctx); REQUIRE_ZBA(ctx); - return gen_shift_imm_fn(ctx, a, EXT_NONE, gen_slli_uw); + return gen_shift_imm_fn(ctx, a, EXT_NONE, gen_slli_uw, NULL); } static bool trans_clmul(DisasContext *ctx, arg_clmul *a) diff --git a/target/riscv/insn_trans/trans_rvi.c.inc b/target/riscv/insn_trans/trans_rvi.c.inc index 3222a45d72..1cb4356b96 100644 --- a/target/riscv/insn_trans/trans_rvi.c.inc +++ b/target/riscv/insn_trans/trans_rvi.c.inc @@ -361,9 +361,22 @@ static bool trans_andi(DisasContext *ctx, arg_andi *a) return gen_logic_imm_fn(ctx, a, tcg_gen_andi_tl); } +static void gen_slli_i128(TCGv retl, TCGv reth, + TCGv src1l, TCGv src1h, + target_long shamt) +{ + if (shamt >= 64) { + tcg_gen_shli_tl(reth, src1l, shamt - 64); + tcg_gen_movi_tl(retl, 0); + } else { + tcg_gen_extract2_tl(reth, src1l, src1h, 64 - shamt); + tcg_gen_shli_tl(retl, src1l, shamt); + } +} + static bool trans_slli(DisasContext *ctx, arg_slli *a) { - return gen_shift_imm_fn(ctx, a, EXT_NONE, tcg_gen_shli_tl); + return gen_shift_imm_fn(ctx, a, EXT_NONE, tcg_gen_shli_tl, gen_slli_i128); } static void gen_srliw(TCGv dst, TCGv src, target_long shamt) @@ -371,10 +384,23 @@ static void gen_srliw(TCGv dst, TCGv src, target_long shamt) tcg_gen_extract_tl(dst, src, shamt, 32 - shamt); } +static void gen_srli_i128(TCGv retl, TCGv reth, + TCGv src1l, TCGv src1h, + target_long shamt) +{ + if (shamt >= 64) { + tcg_gen_shri_tl(retl, src1h, shamt - 64); + tcg_gen_movi_tl(reth, 0); + } else { + tcg_gen_extract2_tl(retl, src1l, src1h, shamt); + tcg_gen_shri_tl(reth, src1h, shamt); + } +} + static bool trans_srli(DisasContext *ctx, arg_srli *a) { return gen_shift_imm_fn_per_ol(ctx, a, EXT_NONE, - tcg_gen_shri_tl, gen_srliw); + tcg_gen_shri_tl, gen_srliw, gen_srli_i128); } static void gen_sraiw(TCGv dst, TCGv src, target_long shamt) @@ -382,10 +408,23 @@ static void gen_sraiw(TCGv dst, TCGv src, target_long shamt) tcg_gen_sextract_tl(dst, src, shamt, 32 - shamt); } +static void gen_srai_i128(TCGv retl, TCGv reth, + TCGv src1l, TCGv src1h, + target_long shamt) +{ + if (shamt >= 64) { + tcg_gen_sari_tl(retl, src1h, shamt - 64); + tcg_gen_sari_tl(reth, src1h, 63); + } else { + tcg_gen_extract2_tl(retl, src1l, src1h, shamt); + tcg_gen_sari_tl(reth, src1h, shamt); + } +} + static bool trans_srai(DisasContext *ctx, arg_srai *a) { return gen_shift_imm_fn_per_ol(ctx, a, EXT_NONE, - tcg_gen_sari_tl, gen_sraiw); + tcg_gen_sari_tl, gen_sraiw, gen_srai_i128); } static bool trans_add(DisasContext *ctx, arg_add *a) @@ -398,9 +437,44 @@ static bool trans_sub(DisasContext *ctx, arg_sub *a) return gen_arith(ctx, a, EXT_NONE, tcg_gen_sub_tl); } +static void gen_sll_i128(TCGv destl, TCGv desth, + TCGv src1l, TCGv src1h, TCGv shamt) +{ + TCGv ls = tcg_temp_new(); + TCGv rs = tcg_temp_new(); + TCGv hs = tcg_temp_new(); + TCGv ll = tcg_temp_new(); + TCGv lr = tcg_temp_new(); + TCGv h0 = tcg_temp_new(); + TCGv h1 = tcg_temp_new(); + TCGv zero = tcg_constant_tl(0); + + tcg_gen_andi_tl(hs, shamt, 64); + tcg_gen_andi_tl(ls, shamt, 63); + tcg_gen_neg_tl(shamt, shamt); + tcg_gen_andi_tl(rs, shamt, 63); + + tcg_gen_shl_tl(ll, src1l, ls); + tcg_gen_shl_tl(h0, src1h, ls); + tcg_gen_shr_tl(lr, src1l, rs); + tcg_gen_movcond_tl(TCG_COND_NE, lr, shamt, zero, lr, zero); + tcg_gen_or_tl(h1, h0, lr); + + tcg_gen_movcond_tl(TCG_COND_NE, destl, hs, zero, zero, ll); + tcg_gen_movcond_tl(TCG_COND_NE, desth, hs, zero, ll, h1); + + tcg_temp_free(ls); + tcg_temp_free(rs); + tcg_temp_free(hs); + tcg_temp_free(ll); + tcg_temp_free(lr); + tcg_temp_free(h0); + tcg_temp_free(h1); +} + static bool trans_sll(DisasContext *ctx, arg_sll *a) { - return gen_shift(ctx, a, EXT_NONE, tcg_gen_shl_tl); + return gen_shift(ctx, a, EXT_NONE, tcg_gen_shl_tl, gen_sll_i128); } static bool trans_slt(DisasContext *ctx, arg_slt *a) @@ -413,14 +487,85 @@ static bool trans_sltu(DisasContext *ctx, arg_sltu *a) return gen_arith(ctx, a, EXT_SIGN, gen_sltu); } +static void gen_srl_i128(TCGv destl, TCGv desth, + TCGv src1l, TCGv src1h, TCGv shamt) +{ + TCGv ls = tcg_temp_new(); + TCGv rs = tcg_temp_new(); + TCGv hs = tcg_temp_new(); + TCGv ll = tcg_temp_new(); + TCGv lr = tcg_temp_new(); + TCGv h0 = tcg_temp_new(); + TCGv h1 = tcg_temp_new(); + TCGv zero = tcg_constant_tl(0); + + tcg_gen_andi_tl(hs, shamt, 64); + tcg_gen_andi_tl(rs, shamt, 63); + tcg_gen_neg_tl(shamt, shamt); + tcg_gen_andi_tl(ls, shamt, 63); + + tcg_gen_shr_tl(lr, src1l, rs); + tcg_gen_shr_tl(h1, src1h, rs); + tcg_gen_shl_tl(ll, src1h, ls); + tcg_gen_movcond_tl(TCG_COND_NE, ll, shamt, zero, ll, zero); + tcg_gen_or_tl(h0, ll, lr); + + tcg_gen_movcond_tl(TCG_COND_NE, destl, hs, zero, h1, h0); + tcg_gen_movcond_tl(TCG_COND_NE, desth, hs, zero, zero, h1); + + tcg_temp_free(ls); + tcg_temp_free(rs); + tcg_temp_free(hs); + tcg_temp_free(ll); + tcg_temp_free(lr); + tcg_temp_free(h0); + tcg_temp_free(h1); +} + static bool trans_srl(DisasContext *ctx, arg_srl *a) { - return gen_shift(ctx, a, EXT_ZERO, tcg_gen_shr_tl); + return gen_shift(ctx, a, EXT_ZERO, tcg_gen_shr_tl, gen_srl_i128); +} + +static void gen_sra_i128(TCGv destl, TCGv desth, + TCGv src1l, TCGv src1h, TCGv shamt) +{ + TCGv ls = tcg_temp_new(); + TCGv rs = tcg_temp_new(); + TCGv hs = tcg_temp_new(); + TCGv ll = tcg_temp_new(); + TCGv lr = tcg_temp_new(); + TCGv h0 = tcg_temp_new(); + TCGv h1 = tcg_temp_new(); + TCGv zero = tcg_constant_tl(0); + + tcg_gen_andi_tl(hs, shamt, 64); + tcg_gen_andi_tl(rs, shamt, 63); + tcg_gen_neg_tl(shamt, shamt); + tcg_gen_andi_tl(ls, shamt, 63); + + tcg_gen_shr_tl(lr, src1l, rs); + tcg_gen_sar_tl(h1, src1h, rs); + tcg_gen_shl_tl(ll, src1h, ls); + tcg_gen_movcond_tl(TCG_COND_NE, ll, shamt, zero, ll, zero); + tcg_gen_or_tl(h0, ll, lr); + tcg_gen_sari_tl(lr, src1h, 63); + + tcg_gen_movcond_tl(TCG_COND_NE, destl, hs, zero, h1, h0); + tcg_gen_movcond_tl(TCG_COND_NE, desth, hs, zero, lr, h1); + + tcg_temp_free(ls); + tcg_temp_free(rs); + tcg_temp_free(hs); + tcg_temp_free(ll); + tcg_temp_free(lr); + tcg_temp_free(h0); + tcg_temp_free(h1); } static bool trans_sra(DisasContext *ctx, arg_sra *a) { - return gen_shift(ctx, a, EXT_SIGN, tcg_gen_sar_tl); + return gen_shift(ctx, a, EXT_SIGN, tcg_gen_sar_tl, gen_sra_i128); } static bool trans_xor(DisasContext *ctx, arg_xor *a) @@ -447,25 +592,47 @@ static bool trans_addiw(DisasContext *ctx, arg_addiw *a) static bool trans_slliw(DisasContext *ctx, arg_slliw *a) { - REQUIRE_64BIT(ctx); + REQUIRE_64_OR_128BIT(ctx); ctx->ol = MXL_RV32; - return gen_shift_imm_fn(ctx, a, EXT_NONE, tcg_gen_shli_tl); + return gen_shift_imm_fn(ctx, a, EXT_NONE, tcg_gen_shli_tl, NULL); } static bool trans_srliw(DisasContext *ctx, arg_srliw *a) { - REQUIRE_64BIT(ctx); + REQUIRE_64_OR_128BIT(ctx); ctx->ol = MXL_RV32; - return gen_shift_imm_fn(ctx, a, EXT_NONE, gen_srliw); + return gen_shift_imm_fn(ctx, a, EXT_NONE, gen_srliw, NULL); } static bool trans_sraiw(DisasContext *ctx, arg_sraiw *a) { - REQUIRE_64BIT(ctx); + REQUIRE_64_OR_128BIT(ctx); ctx->ol = MXL_RV32; - return gen_shift_imm_fn(ctx, a, EXT_NONE, gen_sraiw); + return gen_shift_imm_fn(ctx, a, EXT_NONE, gen_sraiw, NULL); } +static bool trans_sllid(DisasContext *ctx, arg_sllid *a) +{ + REQUIRE_128BIT(ctx); + ctx->ol = MXL_RV64; + return gen_shift_imm_fn(ctx, a, EXT_NONE, tcg_gen_shli_tl, NULL); +} + +static bool trans_srlid(DisasContext *ctx, arg_srlid *a) +{ + REQUIRE_128BIT(ctx); + ctx->ol = MXL_RV64; + return gen_shift_imm_fn(ctx, a, EXT_NONE, tcg_gen_shri_tl, NULL); +} + +static bool trans_sraid(DisasContext *ctx, arg_sraid *a) +{ + REQUIRE_128BIT(ctx); + ctx->ol = MXL_RV64; + return gen_shift_imm_fn(ctx, a, EXT_NONE, tcg_gen_sari_tl, NULL); +} + + static bool trans_addw(DisasContext *ctx, arg_addw *a) { REQUIRE_64BIT(ctx); @@ -482,25 +649,47 @@ static bool trans_subw(DisasContext *ctx, arg_subw *a) static bool trans_sllw(DisasContext *ctx, arg_sllw *a) { - REQUIRE_64BIT(ctx); + REQUIRE_64_OR_128BIT(ctx); ctx->ol = MXL_RV32; - return gen_shift(ctx, a, EXT_NONE, tcg_gen_shl_tl); + return gen_shift(ctx, a, EXT_NONE, tcg_gen_shl_tl, NULL); } static bool trans_srlw(DisasContext *ctx, arg_srlw *a) { - REQUIRE_64BIT(ctx); + REQUIRE_64_OR_128BIT(ctx); ctx->ol = MXL_RV32; - return gen_shift(ctx, a, EXT_ZERO, tcg_gen_shr_tl); + return gen_shift(ctx, a, EXT_ZERO, tcg_gen_shr_tl, NULL); } static bool trans_sraw(DisasContext *ctx, arg_sraw *a) { - REQUIRE_64BIT(ctx); + REQUIRE_64_OR_128BIT(ctx); ctx->ol = MXL_RV32; - return gen_shift(ctx, a, EXT_SIGN, tcg_gen_sar_tl); + return gen_shift(ctx, a, EXT_SIGN, tcg_gen_sar_tl, NULL); } +static bool trans_slld(DisasContext *ctx, arg_slld *a) +{ + REQUIRE_128BIT(ctx); + ctx->ol = MXL_RV64; + return gen_shift(ctx, a, EXT_NONE, tcg_gen_shl_tl, NULL); +} + +static bool trans_srld(DisasContext *ctx, arg_srld *a) +{ + REQUIRE_128BIT(ctx); + ctx->ol = MXL_RV64; + return gen_shift(ctx, a, EXT_ZERO, tcg_gen_shr_tl, NULL); +} + +static bool trans_srad(DisasContext *ctx, arg_srad *a) +{ + REQUIRE_128BIT(ctx); + ctx->ol = MXL_RV64; + return gen_shift(ctx, a, EXT_SIGN, tcg_gen_sar_tl, NULL); +} + + static bool trans_fence(DisasContext *ctx, arg_fence *a) { /* FENCE is a full memory barrier. */ From patchwork Mon Oct 25 12:28:13 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Patchwork-Submitter: =?utf-8?b?RnLDqWTDqXJpYyBQw6l0cm90?= X-Patchwork-Id: 12581751 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 71782C433F5 for ; Mon, 25 Oct 2021 12:41:37 +0000 (UTC) Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.kernel.org (Postfix) with ESMTPS id E746B60EE9 for ; Mon, 25 Oct 2021 12:41:36 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.4.1 mail.kernel.org E746B60EE9 Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=univ-grenoble-alpes.fr Authentication-Results: mail.kernel.org; spf=pass smtp.mailfrom=nongnu.org Received: from localhost ([::1]:42572 helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1mezIF-0004q3-Vk for qemu-devel@archiver.kernel.org; Mon, 25 Oct 2021 08:41:36 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]:53256) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1mez6F-0006Vg-0M; Mon, 25 Oct 2021 08:29:11 -0400 Received: from zm-mta-out-3.u-ga.fr ([152.77.200.56]:43096) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1mez68-000874-O9; Mon, 25 Oct 2021 08:29:10 -0400 Received: from mailhost.u-ga.fr (mailhost2.u-ga.fr [129.88.177.242]) by zm-mta-out-3.u-ga.fr (Postfix) with ESMTP id D7B5B41AD8; Mon, 25 Oct 2021 14:28:34 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=univ-grenoble-alpes.fr; s=2020; t=1635164914; bh=A5DVOw50wz+F4DcRjFIk6oDffH5tJ4lKPvuZpm3QBYQ=; h=From:To:Cc:Subject:Date:In-Reply-To:References:From; b=A9Yh2lAhl5673TIeibpvHMu+ZbAqpBeBgnVmcrjhYnWFo9rqw1zcUUleIPxRyMYbt Db9kOTqHX02M2vzani4hzaLL3ewB7jVRT7GTXEaYNloVO+ldQgOBgjjq17SOLZe6Vu 9LeAjarl7w2fpw2dkLa+XNZBsf3ko5L5hsJlWcAILqN3dV/BxzcB1+4NRB5y5imY7T EFaKnmsAAt2SJIpEJGj9cywr3rxdg3OjttjUPOFCeBVM+w4hZ35nuVkkO9HyGE3uQi swZOrG2i0fmPqGqksKlXY66lmsBRaUq8vJyb/lNKopFZvYYAqlFb8GtrCk4wtgmvzT vbrthQhs3Cg5w== Received: from smtps.univ-grenoble-alpes.fr (smtps2.u-ga.fr [152.77.18.2]) by mailhost.u-ga.fr (Postfix) with ESMTP id BA70C60066; Mon, 25 Oct 2021 14:28:34 +0200 (CEST) Received: from palmier.tima.u-ga.fr (35.201.90.79.rev.sfr.net [79.90.201.35]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (2048 bits) server-digest SHA256) (No client certificate requested) (Authenticated sender: petrotf@univ-grenoble-alpes.fr) by smtps.univ-grenoble-alpes.fr (Postfix) with ESMTPSA id 7EE6514005A; Mon, 25 Oct 2021 14:28:34 +0200 (CEST) From: =?utf-8?b?RnLDqWTDqXJpYyBQw6l0cm90?= To: qemu-devel@nongnu.org, qemu-riscv@nongnu.org Subject: [PATCH v4 12/17] target/riscv: support for 128-bit arithmetic instructions Date: Mon, 25 Oct 2021 14:28:13 +0200 Message-Id: <20211025122818.168890-13-frederic.petrot@univ-grenoble-alpes.fr> X-Mailer: git-send-email 2.33.0 In-Reply-To: <20211025122818.168890-1-frederic.petrot@univ-grenoble-alpes.fr> References: <20211025122818.168890-1-frederic.petrot@univ-grenoble-alpes.fr> MIME-Version: 1.0 X-Greylist: Whitelist-UGA SMTP Authentifie (petrotf@univ-grenoble-alpes.fr) via submission-587 ACL (41) X-Greylist: Whitelist-UGA MAILHOST (SMTP non authentifie) depuis 152.77.18.2 Received-SPF: pass client-ip=152.77.200.56; envelope-from=frederic.petrot@univ-grenoble-alpes.fr; helo=zm-mta-out-3.u-ga.fr X-Spam_score_int: -20 X-Spam_score: -2.1 X-Spam_bar: -- X-Spam_report: (-2.1 / 5.0 requ) BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_MSPIKE_H2=-0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: bin.meng@windriver.com, richard.henderson@linaro.org, alistair.francis@wdc.com, fabien.portas@grenoble-inp.org, palmer@dabbelt.com, =?utf-8?b?RnLDqWTDqXJpYyBQw6l0cm90?= , philmd@redhat.com Errors-To: qemu-devel-bounces+qemu-devel=archiver.kernel.org@nongnu.org Sender: "Qemu-devel" Addition of 128-bit adds and subs in their various sizes. Signed-off-by: Frédéric Pétrot Co-authored-by: Fabien Portas --- target/riscv/insn32.decode | 3 + target/riscv/translate.c | 63 ++++++-- target/riscv/insn_trans/trans_rvb.c.inc | 20 +-- target/riscv/insn_trans/trans_rvi.c.inc | 190 ++++++++++++++++++++++-- target/riscv/insn_trans/trans_rvm.c.inc | 26 ++-- 5 files changed, 253 insertions(+), 49 deletions(-) diff --git a/target/riscv/insn32.decode b/target/riscv/insn32.decode index e338a803a0..afaf243b4e 100644 --- a/target/riscv/insn32.decode +++ b/target/riscv/insn32.decode @@ -171,9 +171,12 @@ sraw 0100000 ..... ..... 101 ..... 0111011 @r ldu ............ ..... 111 ..... 0000011 @i lq ............ ..... 010 ..... 0001111 @i sq ............ ..... 100 ..... 0100011 @s +addid ............ ..... 000 ..... 1011011 @i sllid 000000 ...... ..... 001 ..... 1011011 @sh6 srlid 000000 ...... ..... 101 ..... 1011011 @sh6 sraid 010000 ...... ..... 101 ..... 1011011 @sh6 +addd 0000000 ..... ..... 000 ..... 1111011 @r +subd 0100000 ..... ..... 000 ..... 1111011 @r slld 0000000 ..... ..... 001 ..... 1111011 @r srld 0000000 ..... ..... 101 ..... 1111011 @r srad 0100000 ..... ..... 101 ..... 1111011 @r diff --git a/target/riscv/translate.c b/target/riscv/translate.c index 36f62651ef..26cca144f5 100644 --- a/target/riscv/translate.c +++ b/target/riscv/translate.c @@ -485,57 +485,96 @@ static bool gen_logic(DisasContext *ctx, arg_r *a, DisasExtend ext, } static bool gen_arith_imm_fn(DisasContext *ctx, arg_i *a, DisasExtend ext, - void (*func)(TCGv, TCGv, target_long)) + void (*func)(TCGv, TCGv, target_long), + void (*f128)(TCGv, TCGv, TCGv, TCGv, target_long)) { TCGv dest = dest_gpr(ctx, a->rd); TCGv src1 = get_gpr(ctx, a->rs1, ext); - func(dest, src1, a->imm); + if (get_xl(ctx) < MXL_RV128 || get_ol(ctx) < MXL_RV128) { + func(dest, src1, a->imm); + gen_set_gpr(ctx, a->rd, dest); + } else { + if (f128 == NULL) { + return false; + } - gen_set_gpr(ctx, a->rd, dest); + TCGv src1h = get_gprh(ctx, a->rs1); + TCGv desth = dest_gprh(ctx, a->rd); + + f128(dest, desth, src1, src1h, a->imm); + gen_set_gpr128(ctx, a->rd, dest, desth); + } return true; } static bool gen_arith_imm_tl(DisasContext *ctx, arg_i *a, DisasExtend ext, - void (*func)(TCGv, TCGv, TCGv)) + void (*func)(TCGv, TCGv, TCGv), + void (*f128)(TCGv, TCGv, TCGv, TCGv, TCGv, TCGv)) { TCGv dest = dest_gpr(ctx, a->rd); TCGv src1 = get_gpr(ctx, a->rs1, ext); TCGv src2 = tcg_constant_tl(a->imm); - func(dest, src1, src2); + if (get_xl(ctx) < MXL_RV128 || get_ol(ctx) < MXL_RV128) { + func(dest, src1, src2); + gen_set_gpr(ctx, a->rd, dest); + } else { + if (f128 == NULL) { + return false; + } - gen_set_gpr(ctx, a->rd, dest); + TCGv src1h = get_gprh(ctx, a->rs1); + TCGv src2h = tcg_constant_tl(-(a->imm < 0)); + TCGv desth = dest_gprh(ctx, a->rd); + + f128(dest, desth, src1, src1h, src2, src2h); + gen_set_gpr128(ctx, a->rd, dest, desth); + } return true; } static bool gen_arith(DisasContext *ctx, arg_r *a, DisasExtend ext, - void (*func)(TCGv, TCGv, TCGv)) + void (*func)(TCGv, TCGv, TCGv), + void (*f128)(TCGv, TCGv, TCGv, TCGv, TCGv, TCGv)) { TCGv dest = dest_gpr(ctx, a->rd); TCGv src1 = get_gpr(ctx, a->rs1, ext); TCGv src2 = get_gpr(ctx, a->rs2, ext); - func(dest, src1, src2); + if (get_xl(ctx) < MXL_RV128 || get_ol(ctx) < MXL_RV128) { + func(dest, src1, src2); + gen_set_gpr(ctx, a->rd, dest); + } else { + if (f128 == NULL) { + return false; + } - gen_set_gpr(ctx, a->rd, dest); + TCGv src1h = get_gprh(ctx, a->rs1); + TCGv src2h = get_gprh(ctx, a->rs2); + TCGv desth = dest_gprh(ctx, a->rd); + + f128(dest, desth, src1, src1h, src2, src2h); + gen_set_gpr128(ctx, a->rd, dest, desth); + } return true; } static bool gen_arith_per_ol(DisasContext *ctx, arg_r *a, DisasExtend ext, void (*f_tl)(TCGv, TCGv, TCGv), - void (*f_32)(TCGv, TCGv, TCGv)) + void (*f_32)(TCGv, TCGv, TCGv), + void (*f_128)(TCGv, TCGv, TCGv, TCGv, TCGv, TCGv)) { int olen = get_olen(ctx); if (olen != TARGET_LONG_BITS) { if (olen == 32) { f_tl = f_32; - } else { + } else if (olen != 128) { g_assert_not_reached(); } } - return gen_arith(ctx, a, ext, f_tl); + return gen_arith(ctx, a, ext, f_tl, f_128); } static bool gen_shift_imm_fn(DisasContext *ctx, arg_shift *a, DisasExtend ext, diff --git a/target/riscv/insn_trans/trans_rvb.c.inc b/target/riscv/insn_trans/trans_rvb.c.inc index 7b1686022a..c368afbd82 100644 --- a/target/riscv/insn_trans/trans_rvb.c.inc +++ b/target/riscv/insn_trans/trans_rvb.c.inc @@ -104,25 +104,25 @@ static bool trans_xnor(DisasContext *ctx, arg_xnor *a) static bool trans_min(DisasContext *ctx, arg_min *a) { REQUIRE_ZBB(ctx); - return gen_arith(ctx, a, EXT_SIGN, tcg_gen_smin_tl); + return gen_arith(ctx, a, EXT_SIGN, tcg_gen_smin_tl, NULL); } static bool trans_max(DisasContext *ctx, arg_max *a) { REQUIRE_ZBB(ctx); - return gen_arith(ctx, a, EXT_SIGN, tcg_gen_smax_tl); + return gen_arith(ctx, a, EXT_SIGN, tcg_gen_smax_tl, NULL); } static bool trans_minu(DisasContext *ctx, arg_minu *a) { REQUIRE_ZBB(ctx); - return gen_arith(ctx, a, EXT_SIGN, tcg_gen_umin_tl); + return gen_arith(ctx, a, EXT_SIGN, tcg_gen_umin_tl, NULL); } static bool trans_maxu(DisasContext *ctx, arg_maxu *a) { REQUIRE_ZBB(ctx); - return gen_arith(ctx, a, EXT_SIGN, tcg_gen_umax_tl); + return gen_arith(ctx, a, EXT_SIGN, tcg_gen_umax_tl, NULL); } static bool trans_sext_b(DisasContext *ctx, arg_sext_b *a) @@ -357,7 +357,7 @@ GEN_SHADD(3) static bool trans_sh##SHAMT##add(DisasContext *ctx, arg_sh##SHAMT##add *a) \ { \ REQUIRE_ZBA(ctx); \ - return gen_arith(ctx, a, EXT_NONE, gen_sh##SHAMT##add); \ + return gen_arith(ctx, a, EXT_NONE, gen_sh##SHAMT##add, NULL); \ } GEN_TRANS_SHADD(1) @@ -447,7 +447,7 @@ static bool trans_sh##SHAMT##add_uw(DisasContext *ctx, \ { \ REQUIRE_64BIT(ctx); \ REQUIRE_ZBA(ctx); \ - return gen_arith(ctx, a, EXT_NONE, gen_sh##SHAMT##add_uw); \ + return gen_arith(ctx, a, EXT_NONE, gen_sh##SHAMT##add_uw, NULL); \ } GEN_TRANS_SHADD_UW(1) @@ -466,7 +466,7 @@ static bool trans_add_uw(DisasContext *ctx, arg_add_uw *a) { REQUIRE_64BIT(ctx); REQUIRE_ZBA(ctx); - return gen_arith(ctx, a, EXT_NONE, gen_add_uw); + return gen_arith(ctx, a, EXT_NONE, gen_add_uw, NULL); } static void gen_slli_uw(TCGv dest, TCGv src, target_long shamt) @@ -484,7 +484,7 @@ static bool trans_slli_uw(DisasContext *ctx, arg_slli_uw *a) static bool trans_clmul(DisasContext *ctx, arg_clmul *a) { REQUIRE_ZBC(ctx); - return gen_arith(ctx, a, EXT_NONE, gen_helper_clmul); + return gen_arith(ctx, a, EXT_NONE, gen_helper_clmul, NULL); } static void gen_clmulh(TCGv dst, TCGv src1, TCGv src2) @@ -496,11 +496,11 @@ static void gen_clmulh(TCGv dst, TCGv src1, TCGv src2) static bool trans_clmulh(DisasContext *ctx, arg_clmulr *a) { REQUIRE_ZBC(ctx); - return gen_arith(ctx, a, EXT_NONE, gen_clmulh); + return gen_arith(ctx, a, EXT_NONE, gen_clmulh, NULL); } static bool trans_clmulr(DisasContext *ctx, arg_clmulh *a) { REQUIRE_ZBC(ctx); - return gen_arith(ctx, a, EXT_NONE, gen_helper_clmulr); + return gen_arith(ctx, a, EXT_NONE, gen_helper_clmulr, NULL); } diff --git a/target/riscv/insn_trans/trans_rvi.c.inc b/target/riscv/insn_trans/trans_rvi.c.inc index 1cb4356b96..254c335b53 100644 --- a/target/riscv/insn_trans/trans_rvi.c.inc +++ b/target/riscv/insn_trans/trans_rvi.c.inc @@ -82,13 +82,134 @@ static bool trans_jalr(DisasContext *ctx, arg_jalr *a) return true; } +/* + * Comparison predicates using bitwise logic taken from Hacker's Delight, 2.12 + * We are just interested in the sign bit, so rl is not used but for subtracting + */ +static bool gen_setcond_i128(TCGv rl, TCGv rh, + TCGv al, TCGv ah, + TCGv bl, TCGv bh, + TCGCond cond) +{ + switch (cond) { + case TCG_COND_EQ: + tcg_gen_xor_tl(rl, al, bl); + tcg_gen_xor_tl(rh, ah, bh); + tcg_gen_or_tl(rh, rl, rh); + tcg_gen_setcondi_tl(TCG_COND_EQ, rl, rh, 0); + break; + + case TCG_COND_NE: + tcg_gen_xor_tl(rl, al, bl); + tcg_gen_xor_tl(rh, ah, bh); + tcg_gen_or_tl(rh, rl, rh); + tcg_gen_setcondi_tl(TCG_COND_NE, rl, rh, 0); + break; + + case TCG_COND_LT: + { + TCGv tmp1 = tcg_temp_new(); + TCGv tmp2 = tcg_temp_new(); + + tcg_gen_sub2_tl(rl, rh, al, ah, bl, bh); + tcg_gen_xor_tl(tmp1, rh, ah); + tcg_gen_xor_tl(tmp2, ah, bh); + tcg_gen_and_tl(tmp1, tmp1, tmp2); + tcg_gen_xor_tl(tmp1, rh, tmp1); + tcg_gen_shri_tl(rl, tmp1, 63); + + tcg_temp_free(tmp1); + tcg_temp_free(tmp2); + break; + } + + case TCG_COND_GE: + /* Invert result of TCG_COND_LT */ + gen_setcond_i128(rl, rh, al, ah, bl, bh, TCG_COND_LT); + tcg_gen_xori_tl(rl, rl, 1); + break; + + case TCG_COND_LTU: + gen_setcond_i128(rl, rh, al, ah, bl, bh, TCG_COND_GEU); + tcg_gen_xori_tl(rl, rl, 1); + break; + + case TCG_COND_GEU: + { + TCGv tmp1 = tcg_temp_new(); + TCGv tmp2 = tcg_temp_new(); + TCGv zero = tcg_constant_tl(0); + TCGv one = tcg_constant_tl(1); + /* borrow in to second word */ + tcg_gen_setcond_tl(TCG_COND_LTU, tmp1, al, bl); + /* seed third word with 1, which will be result */ + tcg_gen_sub2_tl(tmp1, tmp2, ah, one, tmp1, zero); + tcg_gen_sub2_tl(tmp1, rl, tmp1, tmp2, bh, zero); + + tcg_temp_free(tmp1); + tcg_temp_free(tmp2); + break; + } + + default: + return false; + } + tcg_gen_movi_tl(rh, 0); + return true; +} + static bool gen_branch(DisasContext *ctx, arg_b *a, TCGCond cond) { TCGLabel *l = gen_new_label(); TCGv src1 = get_gpr(ctx, a->rs1, EXT_SIGN); TCGv src2 = get_gpr(ctx, a->rs2, EXT_SIGN); - tcg_gen_brcond_tl(cond, src1, src2, l); + if (get_xl(ctx) == MXL_RV128) { + TCGv src1h = get_gprh(ctx, a->rs1); + TCGv src2h = get_gprh(ctx, a->rs2); + TCGv tmpl = tcg_temp_new(); + TCGv tmph = tcg_temp_new(); + + /* + * Do not use gen_setcond_i128 for EQ and NE as these conditions are + * often met and can be more efficiently implemented. + * Some comparisons with zero are also simpler, let's do them. + */ + if (cond == TCG_COND_EQ || cond == TCG_COND_NE) { + /* + * bnez and beqz being used quite often too, lets optimize them, + * although QEMU's tcg optimizer handles these cases nicely + */ + if (a->rs2 == 0) { + tcg_gen_or_tl(tmpl, src1, src1h); + tcg_gen_brcondi_tl(cond, tmpl, 0, l); + } else { + tcg_gen_xor_tl(tmpl, src1, src2); + tcg_gen_xor_tl(tmph, src1h, src2h); + tcg_gen_or_tl(tmpl, tmpl, tmph); + tcg_gen_brcondi_tl(cond, tmpl, 0, l); + } + } else if (a->rs2 == 0 + && (cond == TCG_COND_LT || cond == TCG_COND_GE)) { + tcg_gen_shri_tl(tmpl, src1h, 63); + /* hack: transform LT in EQ and GE in NE */ + tcg_gen_brcondi_tl((cond & 13) | 8, tmpl, 1, l); + } else { + if (cond == TCG_COND_GE || cond == TCG_COND_LTU) { + gen_setcond_i128(tmpl, tmph, src1, src1h, src2, src2h, + tcg_invert_cond(cond)); + tcg_gen_brcondi_tl(TCG_COND_EQ, tmpl, 0, l); + } else { + gen_setcond_i128(tmpl, tmph, src1, src1h, src2, src2h, cond); + tcg_gen_brcondi_tl(TCG_COND_NE, tmpl, 0, l); + } + } + + tcg_temp_free(tmph); + tcg_temp_free(tmpl); + } else { + tcg_gen_brcond_tl(cond, src1, src2, l); + } gen_goto_tb(ctx, 1, ctx->pc_succ_insn); gen_set_label(l); /* branch taken */ @@ -321,9 +442,38 @@ static bool trans_sq(DisasContext *ctx, arg_sq *a) return gen_store(ctx, a, MO_TEO); } +static bool trans_addd(DisasContext *ctx, arg_addd *a) +{ + REQUIRE_128BIT(ctx); + ctx->ol = MXL_RV64; + return gen_arith(ctx, a, EXT_NONE, tcg_gen_add_tl, NULL); +} + +static bool trans_addid(DisasContext *ctx, arg_addid *a) +{ + REQUIRE_128BIT(ctx); + ctx->ol = MXL_RV64; + return gen_arith_imm_fn(ctx, a, EXT_NONE, tcg_gen_addi_tl, NULL); +} + +static bool trans_subd(DisasContext *ctx, arg_subd *a) +{ + REQUIRE_128BIT(ctx); + ctx->ol = MXL_RV64; + return gen_arith(ctx, a, EXT_NONE, tcg_gen_sub_tl, NULL); +} + +static void gen_addi2_i128(TCGv retl, TCGv reth, + TCGv srcl, TCGv srch, target_long imm) +{ + TCGv imml = tcg_constant_tl(imm); + TCGv immh = tcg_constant_tl(-(imm < 0)); + tcg_gen_add2_tl(retl, reth, srcl, srch, imml, immh); +} + static bool trans_addi(DisasContext *ctx, arg_addi *a) { - return gen_arith_imm_fn(ctx, a, EXT_NONE, tcg_gen_addi_tl); + return gen_arith_imm_fn(ctx, a, EXT_NONE, tcg_gen_addi_tl, gen_addi2_i128); } static void gen_slt(TCGv ret, TCGv s1, TCGv s2) @@ -331,19 +481,31 @@ static void gen_slt(TCGv ret, TCGv s1, TCGv s2) tcg_gen_setcond_tl(TCG_COND_LT, ret, s1, s2); } +static void gen_slt_i128(TCGv retl, TCGv reth, + TCGv s1l, TCGv s1h, TCGv s2l, TCGv s2h) +{ + gen_setcond_i128(retl, reth, s1l, s1h, s2l, s2h, TCG_COND_LT); +} + static void gen_sltu(TCGv ret, TCGv s1, TCGv s2) { tcg_gen_setcond_tl(TCG_COND_LTU, ret, s1, s2); } +static void gen_sltu_i128(TCGv retl, TCGv reth, + TCGv s1l, TCGv s1h, TCGv s2l, TCGv s2h) +{ + gen_setcond_i128(retl, reth, s1l, s1h, s2l, s2h, TCG_COND_LTU); +} + static bool trans_slti(DisasContext *ctx, arg_slti *a) { - return gen_arith_imm_tl(ctx, a, EXT_SIGN, gen_slt); + return gen_arith_imm_tl(ctx, a, EXT_SIGN, gen_slt, gen_slt_i128); } static bool trans_sltiu(DisasContext *ctx, arg_sltiu *a) { - return gen_arith_imm_tl(ctx, a, EXT_SIGN, gen_sltu); + return gen_arith_imm_tl(ctx, a, EXT_SIGN, gen_sltu, gen_sltu_i128); } static bool trans_xori(DisasContext *ctx, arg_xori *a) @@ -429,12 +591,12 @@ static bool trans_srai(DisasContext *ctx, arg_srai *a) static bool trans_add(DisasContext *ctx, arg_add *a) { - return gen_arith(ctx, a, EXT_NONE, tcg_gen_add_tl); + return gen_arith(ctx, a, EXT_NONE, tcg_gen_add_tl, tcg_gen_add2_tl); } static bool trans_sub(DisasContext *ctx, arg_sub *a) { - return gen_arith(ctx, a, EXT_NONE, tcg_gen_sub_tl); + return gen_arith(ctx, a, EXT_NONE, tcg_gen_sub_tl, tcg_gen_sub2_tl); } static void gen_sll_i128(TCGv destl, TCGv desth, @@ -479,12 +641,12 @@ static bool trans_sll(DisasContext *ctx, arg_sll *a) static bool trans_slt(DisasContext *ctx, arg_slt *a) { - return gen_arith(ctx, a, EXT_SIGN, gen_slt); + return gen_arith(ctx, a, EXT_SIGN, gen_slt, gen_slt_i128); } static bool trans_sltu(DisasContext *ctx, arg_sltu *a) { - return gen_arith(ctx, a, EXT_SIGN, gen_sltu); + return gen_arith(ctx, a, EXT_SIGN, gen_sltu, gen_sltu_i128); } static void gen_srl_i128(TCGv destl, TCGv desth, @@ -585,9 +747,9 @@ static bool trans_and(DisasContext *ctx, arg_and *a) static bool trans_addiw(DisasContext *ctx, arg_addiw *a) { - REQUIRE_64BIT(ctx); + REQUIRE_64_OR_128BIT(ctx); ctx->ol = MXL_RV32; - return gen_arith_imm_fn(ctx, a, EXT_NONE, tcg_gen_addi_tl); + return gen_arith_imm_fn(ctx, a, EXT_NONE, tcg_gen_addi_tl, NULL); } static bool trans_slliw(DisasContext *ctx, arg_slliw *a) @@ -635,16 +797,16 @@ static bool trans_sraid(DisasContext *ctx, arg_sraid *a) static bool trans_addw(DisasContext *ctx, arg_addw *a) { - REQUIRE_64BIT(ctx); + REQUIRE_64_OR_128BIT(ctx); ctx->ol = MXL_RV32; - return gen_arith(ctx, a, EXT_NONE, tcg_gen_add_tl); + return gen_arith(ctx, a, EXT_NONE, tcg_gen_add_tl, NULL); } static bool trans_subw(DisasContext *ctx, arg_subw *a) { - REQUIRE_64BIT(ctx); + REQUIRE_64_OR_128BIT(ctx); ctx->ol = MXL_RV32; - return gen_arith(ctx, a, EXT_NONE, tcg_gen_sub_tl); + return gen_arith(ctx, a, EXT_NONE, tcg_gen_sub_tl, NULL); } static bool trans_sllw(DisasContext *ctx, arg_sllw *a) diff --git a/target/riscv/insn_trans/trans_rvm.c.inc b/target/riscv/insn_trans/trans_rvm.c.inc index 2af0e5c139..efe25dfc11 100644 --- a/target/riscv/insn_trans/trans_rvm.c.inc +++ b/target/riscv/insn_trans/trans_rvm.c.inc @@ -22,7 +22,7 @@ static bool trans_mul(DisasContext *ctx, arg_mul *a) { REQUIRE_EXT(ctx, RVM); - return gen_arith(ctx, a, EXT_NONE, tcg_gen_mul_tl); + return gen_arith(ctx, a, EXT_NONE, tcg_gen_mul_tl, NULL); } static void gen_mulh(TCGv ret, TCGv s1, TCGv s2) @@ -42,7 +42,7 @@ static void gen_mulh_w(TCGv ret, TCGv s1, TCGv s2) static bool trans_mulh(DisasContext *ctx, arg_mulh *a) { REQUIRE_EXT(ctx, RVM); - return gen_arith_per_ol(ctx, a, EXT_SIGN, gen_mulh, gen_mulh_w); + return gen_arith_per_ol(ctx, a, EXT_SIGN, gen_mulh, gen_mulh_w, NULL); } static void gen_mulhsu(TCGv ret, TCGv arg1, TCGv arg2) @@ -76,7 +76,7 @@ static void gen_mulhsu_w(TCGv ret, TCGv arg1, TCGv arg2) static bool trans_mulhsu(DisasContext *ctx, arg_mulhsu *a) { REQUIRE_EXT(ctx, RVM); - return gen_arith_per_ol(ctx, a, EXT_NONE, gen_mulhsu, gen_mulhsu_w); + return gen_arith_per_ol(ctx, a, EXT_NONE, gen_mulhsu, gen_mulhsu_w, NULL); } static void gen_mulhu(TCGv ret, TCGv s1, TCGv s2) @@ -91,7 +91,7 @@ static bool trans_mulhu(DisasContext *ctx, arg_mulhu *a) { REQUIRE_EXT(ctx, RVM); /* gen_mulh_w works for either sign as input. */ - return gen_arith_per_ol(ctx, a, EXT_ZERO, gen_mulhu, gen_mulh_w); + return gen_arith_per_ol(ctx, a, EXT_ZERO, gen_mulhu, gen_mulh_w, NULL); } static void gen_div(TCGv ret, TCGv source1, TCGv source2) @@ -130,7 +130,7 @@ static void gen_div(TCGv ret, TCGv source1, TCGv source2) static bool trans_div(DisasContext *ctx, arg_div *a) { REQUIRE_EXT(ctx, RVM); - return gen_arith(ctx, a, EXT_SIGN, gen_div); + return gen_arith(ctx, a, EXT_SIGN, gen_div, NULL); } static void gen_divu(TCGv ret, TCGv source1, TCGv source2) @@ -158,7 +158,7 @@ static void gen_divu(TCGv ret, TCGv source1, TCGv source2) static bool trans_divu(DisasContext *ctx, arg_divu *a) { REQUIRE_EXT(ctx, RVM); - return gen_arith(ctx, a, EXT_ZERO, gen_divu); + return gen_arith(ctx, a, EXT_ZERO, gen_divu, NULL); } static void gen_rem(TCGv ret, TCGv source1, TCGv source2) @@ -199,7 +199,7 @@ static void gen_rem(TCGv ret, TCGv source1, TCGv source2) static bool trans_rem(DisasContext *ctx, arg_rem *a) { REQUIRE_EXT(ctx, RVM); - return gen_arith(ctx, a, EXT_SIGN, gen_rem); + return gen_arith(ctx, a, EXT_SIGN, gen_rem, NULL); } static void gen_remu(TCGv ret, TCGv source1, TCGv source2) @@ -227,7 +227,7 @@ static void gen_remu(TCGv ret, TCGv source1, TCGv source2) static bool trans_remu(DisasContext *ctx, arg_remu *a) { REQUIRE_EXT(ctx, RVM); - return gen_arith(ctx, a, EXT_ZERO, gen_remu); + return gen_arith(ctx, a, EXT_ZERO, gen_remu, NULL); } static bool trans_mulw(DisasContext *ctx, arg_mulw *a) @@ -235,7 +235,7 @@ static bool trans_mulw(DisasContext *ctx, arg_mulw *a) REQUIRE_64BIT(ctx); REQUIRE_EXT(ctx, RVM); ctx->ol = MXL_RV32; - return gen_arith(ctx, a, EXT_NONE, tcg_gen_mul_tl); + return gen_arith(ctx, a, EXT_NONE, tcg_gen_mul_tl, NULL); } static bool trans_divw(DisasContext *ctx, arg_divw *a) @@ -243,7 +243,7 @@ static bool trans_divw(DisasContext *ctx, arg_divw *a) REQUIRE_64BIT(ctx); REQUIRE_EXT(ctx, RVM); ctx->ol = MXL_RV32; - return gen_arith(ctx, a, EXT_SIGN, gen_div); + return gen_arith(ctx, a, EXT_SIGN, gen_div, NULL); } static bool trans_divuw(DisasContext *ctx, arg_divuw *a) @@ -251,7 +251,7 @@ static bool trans_divuw(DisasContext *ctx, arg_divuw *a) REQUIRE_64BIT(ctx); REQUIRE_EXT(ctx, RVM); ctx->ol = MXL_RV32; - return gen_arith(ctx, a, EXT_ZERO, gen_divu); + return gen_arith(ctx, a, EXT_ZERO, gen_divu, NULL); } static bool trans_remw(DisasContext *ctx, arg_remw *a) @@ -259,7 +259,7 @@ static bool trans_remw(DisasContext *ctx, arg_remw *a) REQUIRE_64BIT(ctx); REQUIRE_EXT(ctx, RVM); ctx->ol = MXL_RV32; - return gen_arith(ctx, a, EXT_SIGN, gen_rem); + return gen_arith(ctx, a, EXT_SIGN, gen_rem, NULL); } static bool trans_remuw(DisasContext *ctx, arg_remuw *a) @@ -267,5 +267,5 @@ static bool trans_remuw(DisasContext *ctx, arg_remuw *a) REQUIRE_64BIT(ctx); REQUIRE_EXT(ctx, RVM); ctx->ol = MXL_RV32; - return gen_arith(ctx, a, EXT_ZERO, gen_remu); + return gen_arith(ctx, a, EXT_ZERO, gen_remu, NULL); } From patchwork Mon Oct 25 12:28:14 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Patchwork-Submitter: =?utf-8?b?RnLDqWTDqXJpYyBQw6l0cm90?= X-Patchwork-Id: 12581759 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 6DA19C433F5 for ; Mon, 25 Oct 2021 12:45:01 +0000 (UTC) Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.kernel.org (Postfix) with ESMTPS id E772A60EFF for ; Mon, 25 Oct 2021 12:45:00 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.4.1 mail.kernel.org E772A60EFF Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=univ-grenoble-alpes.fr Authentication-Results: mail.kernel.org; spf=pass smtp.mailfrom=nongnu.org Received: from localhost ([::1]:52572 helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1mezLY-0003Lo-11 for qemu-devel@archiver.kernel.org; Mon, 25 Oct 2021 08:45:00 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]:53276) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1mez6G-0006XJ-Qi; Mon, 25 Oct 2021 08:29:13 -0400 Received: from zm-mta-out-3.u-ga.fr ([152.77.200.56]:43108) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1mez69-00087I-OV; Mon, 25 Oct 2021 08:29:12 -0400 Received: from mailhost.u-ga.fr (mailhost2.u-ga.fr [129.88.177.242]) by zm-mta-out-3.u-ga.fr (Postfix) with ESMTP id 2BDC241AD3; Mon, 25 Oct 2021 14:28:35 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=univ-grenoble-alpes.fr; s=2020; t=1635164915; bh=nkFBR9auNMSKxc3iqVenpM4PKPHQfrjxoH2oE1AFNjc=; h=From:To:Cc:Subject:Date:In-Reply-To:References:From; b=RkrW87/BFyzjzFRxErZhBVKy+0it7Va5RYKRFlHYUsTycOwMUy+lzSsY4MJtMnpvm uyRSKisN2krreaxEFnWLe6++Y7IIXtV9KSbVfkpzviOf39ObN/C5ptcSf70SAgrf0r r1IXc2freEPVMJKHXhUEIMg15uZgAU5mdFne7KGRgVxOiUpAynwItNxzaPYQH6RkAg 42Pc8PyCifte99WCy4IZZzblUAo3P+5KexUiEVJaa06aj4glzHPO7biJh+LFPhMGSp ehS9zRVjrN8eJvN0Pp2aIPwm2gxKKCmLJJNnnm1uclA7sKOlhA+M6o7U+jf5Ikca/M qFGyQOc5MJn1Q== Received: from smtps.univ-grenoble-alpes.fr (smtps2.u-ga.fr [152.77.18.2]) by mailhost.u-ga.fr (Postfix) with ESMTP id 04B0560066; Mon, 25 Oct 2021 14:28:35 +0200 (CEST) Received: from palmier.tima.u-ga.fr (35.201.90.79.rev.sfr.net [79.90.201.35]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (2048 bits) server-digest SHA256) (No client certificate requested) (Authenticated sender: petrotf@univ-grenoble-alpes.fr) by smtps.univ-grenoble-alpes.fr (Postfix) with ESMTPSA id BEA6614005D; Mon, 25 Oct 2021 14:28:34 +0200 (CEST) From: =?utf-8?b?RnLDqWTDqXJpYyBQw6l0cm90?= To: qemu-devel@nongnu.org, qemu-riscv@nongnu.org Subject: [PATCH v4 13/17] target/riscv: support for 128-bit M extension Date: Mon, 25 Oct 2021 14:28:14 +0200 Message-Id: <20211025122818.168890-14-frederic.petrot@univ-grenoble-alpes.fr> X-Mailer: git-send-email 2.33.0 In-Reply-To: <20211025122818.168890-1-frederic.petrot@univ-grenoble-alpes.fr> References: <20211025122818.168890-1-frederic.petrot@univ-grenoble-alpes.fr> MIME-Version: 1.0 X-Greylist: Whitelist-UGA SMTP Authentifie (petrotf@univ-grenoble-alpes.fr) via submission-587 ACL (41) X-Greylist: Whitelist-UGA MAILHOST (SMTP non authentifie) depuis 152.77.18.2 Received-SPF: pass client-ip=152.77.200.56; envelope-from=frederic.petrot@univ-grenoble-alpes.fr; helo=zm-mta-out-3.u-ga.fr X-Spam_score_int: -20 X-Spam_score: -2.1 X-Spam_bar: -- X-Spam_report: (-2.1 / 5.0 requ) BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_MSPIKE_H2=-0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: bin.meng@windriver.com, richard.henderson@linaro.org, alistair.francis@wdc.com, fabien.portas@grenoble-inp.org, palmer@dabbelt.com, =?utf-8?b?RnLDqWTDqXJpYyBQw6l0cm90?= , philmd@redhat.com Errors-To: qemu-devel-bounces+qemu-devel=archiver.kernel.org@nongnu.org Sender: "Qemu-devel" Given the complexity of the implementation of these instructions, we call helpers to produce their behavior. From an implementation standpoint, we ended up by adding two more tcg globals to return the 128-bit result in a wrapper that itself is called by gen_arith. The sub 128-bit insns are now handled through the existing generation functions. Signed-off-by: Frédéric Pétrot Co-authored-by: Fabien Portas Reviewed-by: Richard Henderson --- target/riscv/cpu.h | 3 + target/riscv/helper.h | 6 + target/riscv/insn32.decode | 7 + target/riscv/m128_helper.c | 109 ++++++++++++++ target/riscv/insn_trans/trans_rvm.c.inc | 187 ++++++++++++++++++++++-- target/riscv/meson.build | 1 + 6 files changed, 300 insertions(+), 13 deletions(-) create mode 100644 target/riscv/m128_helper.c diff --git a/target/riscv/cpu.h b/target/riscv/cpu.h index 6c15dcacfc..923ef01583 100644 --- a/target/riscv/cpu.h +++ b/target/riscv/cpu.h @@ -142,6 +142,9 @@ struct CPURISCVState { uint32_t misa_ext; /* current extensions */ uint32_t misa_ext_mask; /* max ext for this cpu */ + /* 128-bit helpers upper part return value */ + target_ulong retxh; + uint32_t features; #ifdef CONFIG_USER_ONLY diff --git a/target/riscv/helper.h b/target/riscv/helper.h index c7a5376227..957ae8a2a2 100644 --- a/target/riscv/helper.h +++ b/target/riscv/helper.h @@ -1147,3 +1147,9 @@ DEF_HELPER_6(vcompress_vm_b, void, ptr, ptr, ptr, ptr, env, i32) DEF_HELPER_6(vcompress_vm_h, void, ptr, ptr, ptr, ptr, env, i32) DEF_HELPER_6(vcompress_vm_w, void, ptr, ptr, ptr, ptr, env, i32) DEF_HELPER_6(vcompress_vm_d, void, ptr, ptr, ptr, ptr, env, i32) + +/* 128-bit integer multiplication and division */ +DEF_HELPER_5(divu_i128, i64, env, i64, i64, i64, i64) +DEF_HELPER_5(divs_i128, i64, env, i64, i64, i64, i64) +DEF_HELPER_5(remu_i128, i64, env, i64, i64, i64, i64) +DEF_HELPER_5(rems_i128, i64, env, i64, i64, i64, i64) diff --git a/target/riscv/insn32.decode b/target/riscv/insn32.decode index afaf243b4e..16d40362e6 100644 --- a/target/riscv/insn32.decode +++ b/target/riscv/insn32.decode @@ -198,6 +198,13 @@ divuw 0000001 ..... ..... 101 ..... 0111011 @r remw 0000001 ..... ..... 110 ..... 0111011 @r remuw 0000001 ..... ..... 111 ..... 0111011 @r +# *** RV128M Standard Extension (in addition to RV64M) *** +muld 0000001 ..... ..... 000 ..... 1111011 @r +divd 0000001 ..... ..... 100 ..... 1111011 @r +divud 0000001 ..... ..... 101 ..... 1111011 @r +remd 0000001 ..... ..... 110 ..... 1111011 @r +remud 0000001 ..... ..... 111 ..... 1111011 @r + # *** RV32A Standard Extension *** lr_w 00010 . . 00000 ..... 010 ..... 0101111 @atom_ld sc_w 00011 . . ..... ..... 010 ..... 0101111 @atom_st diff --git a/target/riscv/m128_helper.c b/target/riscv/m128_helper.c new file mode 100644 index 0000000000..bd8a37afe7 --- /dev/null +++ b/target/riscv/m128_helper.c @@ -0,0 +1,109 @@ +/* + * RISC-V Emulation Helpers for QEMU. + * + * Copyright (c) 2016-2017 Sagar Karandikar, sagark@eecs.berkeley.edu + * Copyright (c) 2017-2018 SiFive, Inc. + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2 or later, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + * You should have received a copy of the GNU General Public License along with + * this program. If not, see . + */ + +#include "qemu/osdep.h" +#include "cpu.h" +#include "qemu/main-loop.h" +#include "exec/exec-all.h" +#include "exec/helper-proto.h" + +uint64_t HELPER(divu_i128)(CPURISCVState *env, + uint64_t ul, uint64_t uh, + uint64_t vl, uint64_t vh) +{ + uint64_t ql, qh; + Int128 q; + + if (vl == 0 && vh == 0) { /* Handle special behavior on div by zero */ + ql = ~0x0; + qh = ~0x0; + } else { + q = int128_divu(int128_make128(ul, uh), int128_make128(vl, vh)); + ql = int128_getlo(q); + qh = int128_gethi(q); + } + + env->retxh = qh; + return ql; +} + +uint64_t HELPER(remu_i128)(CPURISCVState *env, + uint64_t ul, uint64_t uh, + uint64_t vl, uint64_t vh) +{ + uint64_t rl, rh; + Int128 r; + + if (vl == 0 && vh == 0) { + rl = ul; + rh = uh; + } else { + r = int128_remu(int128_make128(ul, uh), int128_make128(vl, vh)); + rl = int128_getlo(r); + rh = int128_gethi(r); + } + + env->retxh = rh; + return rl; +} + +uint64_t HELPER(divs_i128)(CPURISCVState *env, + uint64_t ul, uint64_t uh, + uint64_t vl, uint64_t vh) +{ + uint64_t qh, ql; + Int128 q; + + if (vl == 0 && vh == 0) { /* Div by zero check */ + ql = ~0x0; + qh = ~0x0; + } else if (uh == 0x8000000000000000 && ul == 0 && + vh == ~0x0 && vl == ~0x0) { + /* Signed div overflow check (-2**127 / -1) */ + ql = ul; + qh = uh; + } else { + q = int128_divs(int128_make128(ul, uh), int128_make128(vl, vh)); + ql = int128_getlo(q); + qh = int128_gethi(q); + } + + env->retxh = qh; + return ql; +} + +uint64_t HELPER(rems_i128)(CPURISCVState *env, + uint64_t ul, uint64_t uh, + uint64_t vl, uint64_t vh) +{ + uint64_t rh, rl; + Int128 r; + + if (vl == 0 && vh == 0) { + rl = ul; + rh = uh; + } else { + r = int128_rems(int128_make128(ul, uh), int128_make128(vl, vh)); + rl = int128_getlo(r); + rh = int128_gethi(r); + } + + env->retxh = rh; + return rl; +} diff --git a/target/riscv/insn_trans/trans_rvm.c.inc b/target/riscv/insn_trans/trans_rvm.c.inc index efe25dfc11..1a2d6da42f 100644 --- a/target/riscv/insn_trans/trans_rvm.c.inc +++ b/target/riscv/insn_trans/trans_rvm.c.inc @@ -18,11 +18,80 @@ * this program. If not, see . */ +static void gen_mulu2_i128(TCGv r2, TCGv r3, TCGv al, TCGv ah, TCGv bl, TCGv bh) +{ + TCGv tmpl = tcg_temp_new(); + TCGv tmph = tcg_temp_new(); + TCGv r0 = tcg_temp_new(); + TCGv r1 = tcg_temp_new(); + TCGv zero = tcg_constant_tl(0); + + tcg_gen_mulu2_tl(r0, r1, al, bl); + + tcg_gen_mulu2_tl(tmpl, tmph, al, bh); + tcg_gen_add2_tl(r1, r2, r1, zero, tmpl, tmph); + tcg_gen_mulu2_tl(tmpl, tmph, ah, bl); + tcg_gen_add2_tl(r1, tmph, r1, r2, tmpl, tmph); + /* Overflow detection into r3 */ + tcg_gen_setcond_tl(TCG_COND_LTU, r3, tmph, r2); + + tcg_gen_mov_tl(r2, tmph); + + tcg_gen_mulu2_tl(tmpl, tmph, ah, bh); + tcg_gen_add2_tl(r2, r3, r2, r3, tmpl, tmph); + + tcg_temp_free(tmpl); + tcg_temp_free(tmph); +} + +static void gen_mul_i128(TCGv rl, TCGv rh, + TCGv rs1l, TCGv rs1h, TCGv rs2l, TCGv rs2h) +{ + TCGv tmpl = tcg_temp_new(); + TCGv tmph = tcg_temp_new(); + TCGv tmpx = tcg_temp_new(); + TCGv zero = tcg_constant_tl(0); + + tcg_gen_mulu2_tl(rl, rh, rs1l, rs2l); + tcg_gen_mulu2_tl(tmpl, tmph, rs1l, rs2h); + tcg_gen_add2_tl(rh, tmpx, rh, zero, tmpl, tmph); + tcg_gen_mulu2_tl(tmpl, tmph, rs1h, rs2l); + tcg_gen_add2_tl(rh, tmph, rh, tmpx, tmpl, tmph); + + tcg_temp_free(tmpl); + tcg_temp_free(tmph); + tcg_temp_free(tmpx); +} + static bool trans_mul(DisasContext *ctx, arg_mul *a) { REQUIRE_EXT(ctx, RVM); - return gen_arith(ctx, a, EXT_NONE, tcg_gen_mul_tl, NULL); + return gen_arith(ctx, a, EXT_NONE, tcg_gen_mul_tl, gen_mul_i128); +} + +static void gen_mulh_i128(TCGv rl, TCGv rh, + TCGv rs1l, TCGv rs1h, TCGv rs2l, TCGv rs2h) +{ + TCGv t0l = tcg_temp_new(); + TCGv t0h = tcg_temp_new(); + TCGv t1l = tcg_temp_new(); + TCGv t1h = tcg_temp_new(); + + gen_mulu2_i128(rl, rh, rs1l, rs1h, rs2l, rs2h); + tcg_gen_sari_tl(t0h, rs1h, 63); + tcg_gen_and_tl(t0l, t0h, rs2l); + tcg_gen_and_tl(t0h, t0h, rs2h); + tcg_gen_sari_tl(t1h, rs2h, 63); + tcg_gen_and_tl(t1l, t1h, rs1l); + tcg_gen_and_tl(t1h, t1h, rs1h); + tcg_gen_sub2_tl(t0l, t0h, rl, rh, t0l, t0h); + tcg_gen_sub2_tl(rl, rh, t0l, t0h, t1l, t1h); + + tcg_temp_free(t0l); + tcg_temp_free(t0h); + tcg_temp_free(t1l); + tcg_temp_free(t1h); } static void gen_mulh(TCGv ret, TCGv s1, TCGv s2) @@ -42,7 +111,25 @@ static void gen_mulh_w(TCGv ret, TCGv s1, TCGv s2) static bool trans_mulh(DisasContext *ctx, arg_mulh *a) { REQUIRE_EXT(ctx, RVM); - return gen_arith_per_ol(ctx, a, EXT_SIGN, gen_mulh, gen_mulh_w, NULL); + return gen_arith_per_ol(ctx, a, EXT_SIGN, gen_mulh, gen_mulh_w, + gen_mulh_i128); +} + +static void gen_mulhsu_i128(TCGv rl, TCGv rh, + TCGv rs1l, TCGv rs1h, TCGv rs2l, TCGv rs2h) +{ + + TCGv t0l = tcg_temp_new(); + TCGv t0h = tcg_temp_new(); + + gen_mulu2_i128(rl, rh, rs1l, rs1h, rs2l, rs2h); + tcg_gen_sari_tl(t0h, rs1h, 63); + tcg_gen_and_tl(t0l, t0h, rs2l); + tcg_gen_and_tl(t0h, t0h, rs2h); + tcg_gen_sub2_tl(rl, rh, rl, rh, t0l, t0h); + + tcg_temp_free(t0l); + tcg_temp_free(t0h); } static void gen_mulhsu(TCGv ret, TCGv arg1, TCGv arg2) @@ -76,7 +163,8 @@ static void gen_mulhsu_w(TCGv ret, TCGv arg1, TCGv arg2) static bool trans_mulhsu(DisasContext *ctx, arg_mulhsu *a) { REQUIRE_EXT(ctx, RVM); - return gen_arith_per_ol(ctx, a, EXT_NONE, gen_mulhsu, gen_mulhsu_w, NULL); + return gen_arith_per_ol(ctx, a, EXT_NONE, gen_mulhsu, gen_mulhsu_w, + gen_mulhsu_i128); } static void gen_mulhu(TCGv ret, TCGv s1, TCGv s2) @@ -91,7 +179,16 @@ static bool trans_mulhu(DisasContext *ctx, arg_mulhu *a) { REQUIRE_EXT(ctx, RVM); /* gen_mulh_w works for either sign as input. */ - return gen_arith_per_ol(ctx, a, EXT_ZERO, gen_mulhu, gen_mulh_w, NULL); + return gen_arith_per_ol(ctx, a, EXT_ZERO, gen_mulhu, gen_mulh_w, + gen_mulu2_i128); +} + +static void gen_div_i128(TCGv rdl, TCGv rdh, + TCGv rs1l, TCGv rs1h, TCGv rs2l, TCGv rs2h) +{ + gen_helper_divs_i128((TCGv_i64)rdl, cpu_env, (TCGv_i64)rs1l, (TCGv_i64)rs1h, + (TCGv_i64)rs2l, (TCGv_i64)rs2h); + tcg_gen_ld_tl(rdh, cpu_env, offsetof(CPURISCVState, retxh)); } static void gen_div(TCGv ret, TCGv source1, TCGv source2) @@ -130,7 +227,15 @@ static void gen_div(TCGv ret, TCGv source1, TCGv source2) static bool trans_div(DisasContext *ctx, arg_div *a) { REQUIRE_EXT(ctx, RVM); - return gen_arith(ctx, a, EXT_SIGN, gen_div, NULL); + return gen_arith(ctx, a, EXT_SIGN, gen_div, gen_div_i128); +} + +static void gen_divu_i128(TCGv rdl, TCGv rdh, + TCGv rs1l, TCGv rs1h, TCGv rs2l, TCGv rs2h) +{ + gen_helper_divu_i128((TCGv_i64)rdl, cpu_env, (TCGv_i64)rs1l, (TCGv_i64)rs1h, + (TCGv_i64)rs2l, (TCGv_i64)rs2h); + tcg_gen_ld_tl(rdh, cpu_env, offsetof(CPURISCVState, retxh)); } static void gen_divu(TCGv ret, TCGv source1, TCGv source2) @@ -158,7 +263,15 @@ static void gen_divu(TCGv ret, TCGv source1, TCGv source2) static bool trans_divu(DisasContext *ctx, arg_divu *a) { REQUIRE_EXT(ctx, RVM); - return gen_arith(ctx, a, EXT_ZERO, gen_divu, NULL); + return gen_arith(ctx, a, EXT_ZERO, gen_divu, gen_divu_i128); +} + +static void gen_rem_i128(TCGv rdl, TCGv rdh, + TCGv rs1l, TCGv rs1h, TCGv rs2l, TCGv rs2h) +{ + gen_helper_rems_i128((TCGv_i64)rdl, cpu_env, (TCGv_i64)rs1l, (TCGv_i64)rs1h, + (TCGv_i64)rs2l, (TCGv_i64)rs2h); + tcg_gen_ld_tl(rdh, cpu_env, offsetof(CPURISCVState, retxh)); } static void gen_rem(TCGv ret, TCGv source1, TCGv source2) @@ -199,7 +312,15 @@ static void gen_rem(TCGv ret, TCGv source1, TCGv source2) static bool trans_rem(DisasContext *ctx, arg_rem *a) { REQUIRE_EXT(ctx, RVM); - return gen_arith(ctx, a, EXT_SIGN, gen_rem, NULL); + return gen_arith(ctx, a, EXT_SIGN, gen_rem, gen_rem_i128); +} + +static void gen_remu_i128(TCGv rdl, TCGv rdh, + TCGv rs1l, TCGv rs1h, TCGv rs2l, TCGv rs2h) +{ + gen_helper_remu_i128((TCGv_i64)rdl, cpu_env, (TCGv_i64)rs1l, (TCGv_i64)rs1h, + (TCGv_i64)rs2l, (TCGv_i64)rs2h); + tcg_gen_ld_tl(rdh, cpu_env, offsetof(CPURISCVState, retxh)); } static void gen_remu(TCGv ret, TCGv source1, TCGv source2) @@ -227,12 +348,12 @@ static void gen_remu(TCGv ret, TCGv source1, TCGv source2) static bool trans_remu(DisasContext *ctx, arg_remu *a) { REQUIRE_EXT(ctx, RVM); - return gen_arith(ctx, a, EXT_ZERO, gen_remu, NULL); + return gen_arith(ctx, a, EXT_ZERO, gen_remu, gen_remu_i128); } static bool trans_mulw(DisasContext *ctx, arg_mulw *a) { - REQUIRE_64BIT(ctx); + REQUIRE_64_OR_128BIT(ctx); REQUIRE_EXT(ctx, RVM); ctx->ol = MXL_RV32; return gen_arith(ctx, a, EXT_NONE, tcg_gen_mul_tl, NULL); @@ -240,7 +361,7 @@ static bool trans_mulw(DisasContext *ctx, arg_mulw *a) static bool trans_divw(DisasContext *ctx, arg_divw *a) { - REQUIRE_64BIT(ctx); + REQUIRE_64_OR_128BIT(ctx); REQUIRE_EXT(ctx, RVM); ctx->ol = MXL_RV32; return gen_arith(ctx, a, EXT_SIGN, gen_div, NULL); @@ -248,7 +369,7 @@ static bool trans_divw(DisasContext *ctx, arg_divw *a) static bool trans_divuw(DisasContext *ctx, arg_divuw *a) { - REQUIRE_64BIT(ctx); + REQUIRE_64_OR_128BIT(ctx); REQUIRE_EXT(ctx, RVM); ctx->ol = MXL_RV32; return gen_arith(ctx, a, EXT_ZERO, gen_divu, NULL); @@ -256,7 +377,7 @@ static bool trans_divuw(DisasContext *ctx, arg_divuw *a) static bool trans_remw(DisasContext *ctx, arg_remw *a) { - REQUIRE_64BIT(ctx); + REQUIRE_64_OR_128BIT(ctx); REQUIRE_EXT(ctx, RVM); ctx->ol = MXL_RV32; return gen_arith(ctx, a, EXT_SIGN, gen_rem, NULL); @@ -264,8 +385,48 @@ static bool trans_remw(DisasContext *ctx, arg_remw *a) static bool trans_remuw(DisasContext *ctx, arg_remuw *a) { - REQUIRE_64BIT(ctx); + REQUIRE_64_OR_128BIT(ctx); REQUIRE_EXT(ctx, RVM); ctx->ol = MXL_RV32; return gen_arith(ctx, a, EXT_ZERO, gen_remu, NULL); } + +static bool trans_muld(DisasContext *ctx, arg_muld *a) +{ + REQUIRE_128BIT(ctx); + REQUIRE_EXT(ctx, RVM); + ctx->ol = MXL_RV64; + return gen_arith(ctx, a, EXT_SIGN, tcg_gen_mul_tl, NULL); +} + +static bool trans_divd(DisasContext *ctx, arg_divd *a) +{ + REQUIRE_128BIT(ctx); + REQUIRE_EXT(ctx, RVM); + ctx->ol = MXL_RV64; + return gen_arith(ctx, a, EXT_SIGN, gen_div, NULL); +} + +static bool trans_divud(DisasContext *ctx, arg_divud *a) +{ + REQUIRE_128BIT(ctx); + REQUIRE_EXT(ctx, RVM); + ctx->ol = MXL_RV64; + return gen_arith(ctx, a, EXT_ZERO, gen_divu, NULL); +} + +static bool trans_remd(DisasContext *ctx, arg_remd *a) +{ + REQUIRE_128BIT(ctx); + REQUIRE_EXT(ctx, RVM); + ctx->ol = MXL_RV64; + return gen_arith(ctx, a, EXT_SIGN, gen_rem, NULL); +} + +static bool trans_remud(DisasContext *ctx, arg_remud *a) +{ + REQUIRE_128BIT(ctx); + REQUIRE_EXT(ctx, RVM); + ctx->ol = MXL_RV64; + return gen_arith(ctx, a, EXT_ZERO, gen_remu, NULL); +} diff --git a/target/riscv/meson.build b/target/riscv/meson.build index d5e0bc93ea..a32158da93 100644 --- a/target/riscv/meson.build +++ b/target/riscv/meson.build @@ -18,6 +18,7 @@ riscv_ss.add(files( 'vector_helper.c', 'bitmanip_helper.c', 'translate.c', + 'm128_helper.c' )) riscv_softmmu_ss = ss.source_set() From patchwork Mon Oct 25 12:28:15 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Patchwork-Submitter: =?utf-8?b?RnLDqWTDqXJpYyBQw6l0cm90?= X-Patchwork-Id: 12581737 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 3EE65C433EF for ; Mon, 25 Oct 2021 12:37:41 +0000 (UTC) Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.kernel.org (Postfix) with ESMTPS id D37C560EE9 for ; Mon, 25 Oct 2021 12:37:40 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.4.1 mail.kernel.org D37C560EE9 Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=univ-grenoble-alpes.fr Authentication-Results: mail.kernel.org; spf=pass smtp.mailfrom=nongnu.org Received: from localhost ([::1]:33326 helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1mezER-00070t-Vj for qemu-devel@archiver.kernel.org; Mon, 25 Oct 2021 08:37:40 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]:53216) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1mez6C-0006Te-Sw; Mon, 25 Oct 2021 08:29:09 -0400 Received: from zm-mta-out-3.u-ga.fr ([152.77.200.56]:43110) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1mez69-00087J-OH; Mon, 25 Oct 2021 08:29:08 -0400 Received: from mailhost.u-ga.fr (mailhost1.u-ga.fr [152.77.1.10]) by zm-mta-out-3.u-ga.fr (Postfix) with ESMTP id 58D9F41AEB; Mon, 25 Oct 2021 14:28:35 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=univ-grenoble-alpes.fr; s=2020; t=1635164915; bh=Jsfdnn9fmQX25DCB9x7cNC8krS3rLPaQfYC9SyINQKw=; h=From:To:Cc:Subject:Date:In-Reply-To:References:From; b=JI3/Pq67YkXXjNpO06ju55niFbaQUezz/YxOJ8F8kz7/v0ezIPu+A/1U3csEaUN3j KjIOBHjPI9Ak5E3X58WEqMBPdWkZij1QUKrVxKpwGv9SqQkNvqXrC9CDtpr1y5ngD3 0bp2JsoTFKCMWUMdcCA/fu/EBhrFwH2KgFIl+xh6zOi5p4+5sfyaMv9zUdc3J2ipIC bR910HimSdW35EYmiFREuPYe867CQeRQkxukVAIZBeiw64uKxe7DVW6DkAhAPGgU+6 eebFFEFCcnWr/98P86cLP+gF1mJzz8RsqVIGKc+46BDXcFk318/jObANlLQrf0FU4v ZZMM7aTBskGXw== Received: from smtps.univ-grenoble-alpes.fr (smtps2.u-ga.fr [152.77.18.2]) by mailhost.u-ga.fr (Postfix) with ESMTP id 41D2460067; Mon, 25 Oct 2021 14:28:35 +0200 (CEST) Received: from palmier.tima.u-ga.fr (35.201.90.79.rev.sfr.net [79.90.201.35]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (2048 bits) server-digest SHA256) (No client certificate requested) (Authenticated sender: petrotf@univ-grenoble-alpes.fr) by smtps.univ-grenoble-alpes.fr (Postfix) with ESMTPSA id 09B8A14005A; Mon, 25 Oct 2021 14:28:35 +0200 (CEST) From: =?utf-8?b?RnLDqWTDqXJpYyBQw6l0cm90?= To: qemu-devel@nongnu.org, qemu-riscv@nongnu.org Subject: [PATCH v4 14/17] target/riscv: adding high part of some csrs Date: Mon, 25 Oct 2021 14:28:15 +0200 Message-Id: <20211025122818.168890-15-frederic.petrot@univ-grenoble-alpes.fr> X-Mailer: git-send-email 2.33.0 In-Reply-To: <20211025122818.168890-1-frederic.petrot@univ-grenoble-alpes.fr> References: <20211025122818.168890-1-frederic.petrot@univ-grenoble-alpes.fr> MIME-Version: 1.0 X-Greylist: Whitelist-UGA SMTP Authentifie (petrotf@univ-grenoble-alpes.fr) via submission-587 ACL (41) X-Greylist: Whitelist-UGA MAILHOST (SMTP non authentifie) depuis 152.77.18.2 Received-SPF: pass client-ip=152.77.200.56; envelope-from=frederic.petrot@univ-grenoble-alpes.fr; helo=zm-mta-out-3.u-ga.fr X-Spam_score_int: -20 X-Spam_score: -2.1 X-Spam_bar: -- X-Spam_report: (-2.1 / 5.0 requ) BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_MSPIKE_H2=-0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: bin.meng@windriver.com, richard.henderson@linaro.org, alistair.francis@wdc.com, fabien.portas@grenoble-inp.org, palmer@dabbelt.com, =?utf-8?b?RnLDqWTDqXJpYyBQw6l0cm90?= , philmd@redhat.com Errors-To: qemu-devel-bounces+qemu-devel=archiver.kernel.org@nongnu.org Sender: "Qemu-devel" Adding the high part of a very minimal set of csr. Signed-off-by: Frédéric Pétrot Co-authored-by: Fabien Portas Reviewed-by: Richard Henderson --- target/riscv/cpu.h | 4 ++++ target/riscv/machine.c | 2 ++ 2 files changed, 6 insertions(+) diff --git a/target/riscv/cpu.h b/target/riscv/cpu.h index 923ef01583..d8093be74d 100644 --- a/target/riscv/cpu.h +++ b/target/riscv/cpu.h @@ -194,6 +194,10 @@ struct CPURISCVState { target_ulong hgatp; uint64_t htimedelta; + /* Upper 64-bits of 128-bit CSRs */ + uint64_t mscratchh; + uint64_t sscratchh; + /* Virtual CSRs */ /* * For RV32 this is 32-bit vsstatus and 32-bit vsstatush. diff --git a/target/riscv/machine.c b/target/riscv/machine.c index 78636ec96d..df4e667c59 100644 --- a/target/riscv/machine.c +++ b/target/riscv/machine.c @@ -153,6 +153,8 @@ static const VMStateDescription vmstate_rv128 = { .needed = rv128_needed, .fields = (VMStateField[]) { VMSTATE_UINTTL_ARRAY(env.gprh, RISCVCPU, 32), + VMSTATE_UINT64(env.mscratchh, RISCVCPU), + VMSTATE_UINT64(env.sscratchh, RISCVCPU), VMSTATE_END_OF_LIST() } }; From patchwork Mon Oct 25 12:28:16 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Patchwork-Submitter: =?utf-8?b?RnLDqWTDqXJpYyBQw6l0cm90?= X-Patchwork-Id: 12581755 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id F2384C433EF for ; Mon, 25 Oct 2021 12:43:24 +0000 (UTC) Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.kernel.org (Postfix) with ESMTPS id 7180060EFF for ; Mon, 25 Oct 2021 12:43:24 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.4.1 mail.kernel.org 7180060EFF Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=univ-grenoble-alpes.fr Authentication-Results: mail.kernel.org; spf=pass smtp.mailfrom=nongnu.org Received: from localhost ([::1]:47082 helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1mezJz-000891-91 for qemu-devel@archiver.kernel.org; Mon, 25 Oct 2021 08:43:23 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]:53246) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1mez6E-0006VW-AR; Mon, 25 Oct 2021 08:29:11 -0400 Received: from zm-mta-out-3.u-ga.fr ([152.77.200.56]:43116) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1mez69-00087U-Ox; Mon, 25 Oct 2021 08:29:10 -0400 Received: from mailhost.u-ga.fr (mailhost2.u-ga.fr [129.88.177.242]) by zm-mta-out-3.u-ga.fr (Postfix) with ESMTP id DA0E541AEF; Mon, 25 Oct 2021 14:28:35 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=univ-grenoble-alpes.fr; s=2020; t=1635164915; bh=Yd4JUM3tPb7y6mATLzioGQ04byH/nS4PUed+b3PWtqA=; h=From:To:Cc:Subject:Date:In-Reply-To:References:From; b=0GKxb5a292f5vrpkgOYmQ8vkj9eM0WE10Om0ajS+Fuqar22pDs5PB1pZUu0tkW/hA eYno5YOnSnZw7LmQ2uAbPhjzvL1zbnNyi2217MkkJBbakf62TQpMNAjGN9/MtgJt+g 9y3h4cRJdvUy8ZMY5yk+O1nDYQIz9MVeLUg2b78rwrgWq1S4E4UFskxpYH0uEuoLdB Nr/Qdcy48lSpduVZl+6O4vMNxzB8XUmMUhgZXbPr9FYkqW4IGYAqKSsa7u6ppTDMM/ I1yex7dan5nWsYxtPacwkyWvD0txcwsd7NcTXKzC1M0xtTGmQvfTxjtkHrYn5j7ku2 23+LQWI83AS+Q== Received: from smtps.univ-grenoble-alpes.fr (smtps2.u-ga.fr [152.77.18.2]) by mailhost.u-ga.fr (Postfix) with ESMTP id BC4CC60066; Mon, 25 Oct 2021 14:28:35 +0200 (CEST) Received: from palmier.tima.u-ga.fr (35.201.90.79.rev.sfr.net [79.90.201.35]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (2048 bits) server-digest SHA256) (No client certificate requested) (Authenticated sender: petrotf@univ-grenoble-alpes.fr) by smtps.univ-grenoble-alpes.fr (Postfix) with ESMTPSA id 69C9F14005A; Mon, 25 Oct 2021 14:28:35 +0200 (CEST) From: =?utf-8?b?RnLDqWTDqXJpYyBQw6l0cm90?= To: qemu-devel@nongnu.org, qemu-riscv@nongnu.org Subject: [PATCH v4 15/17] target/riscv: helper functions to wrap calls to 128-bit csr insns Date: Mon, 25 Oct 2021 14:28:16 +0200 Message-Id: <20211025122818.168890-16-frederic.petrot@univ-grenoble-alpes.fr> X-Mailer: git-send-email 2.33.0 In-Reply-To: <20211025122818.168890-1-frederic.petrot@univ-grenoble-alpes.fr> References: <20211025122818.168890-1-frederic.petrot@univ-grenoble-alpes.fr> MIME-Version: 1.0 X-Greylist: Whitelist-UGA SMTP Authentifie (petrotf@univ-grenoble-alpes.fr) via submission-587 ACL (41) X-Greylist: Whitelist-UGA MAILHOST (SMTP non authentifie) depuis 152.77.18.2 Received-SPF: pass client-ip=152.77.200.56; envelope-from=frederic.petrot@univ-grenoble-alpes.fr; helo=zm-mta-out-3.u-ga.fr X-Spam_score_int: -20 X-Spam_score: -2.1 X-Spam_bar: -- X-Spam_report: (-2.1 / 5.0 requ) BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_MSPIKE_H2=-0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: bin.meng@windriver.com, richard.henderson@linaro.org, alistair.francis@wdc.com, fabien.portas@grenoble-inp.org, palmer@dabbelt.com, =?utf-8?b?RnLDqWTDqXJpYyBQw6l0cm90?= , philmd@redhat.com Errors-To: qemu-devel-bounces+qemu-devel=archiver.kernel.org@nongnu.org Sender: "Qemu-devel" Given the side effects they have, the csr instructions are realized as helpers. We extend this existing infrastructure for 128-bit sized csr. We return 128-bit values using the same approach as for div/rem. Theses helpers all call a unique function that is currently a fallback on the 64-bit version. The trans_csrxx functions supporting 128-bit are yet to be implemented. Signed-off-by: Frédéric Pétrot Co-authored-by: Fabien Portas Reviewed-by: Richard Henderson --- target/riscv/cpu.h | 4 ++++ target/riscv/helper.h | 3 +++ target/riscv/csr.c | 17 ++++++++++++++++ target/riscv/op_helper.c | 44 ++++++++++++++++++++++++++++++++++++++++ 4 files changed, 68 insertions(+) diff --git a/target/riscv/cpu.h b/target/riscv/cpu.h index d8093be74d..328ba42a30 100644 --- a/target/riscv/cpu.h +++ b/target/riscv/cpu.h @@ -470,6 +470,10 @@ typedef RISCVException (*riscv_csr_op_fn)(CPURISCVState *env, int csrno, target_ulong new_value, target_ulong write_mask); +RISCVException riscv_csrrw_i128(CPURISCVState *env, int csrno, + Int128 *ret_value, + Int128 new_value, Int128 write_mask); + typedef struct { const char *name; riscv_csr_predicate_fn predicate; diff --git a/target/riscv/helper.h b/target/riscv/helper.h index 957ae8a2a2..1ef43e0daf 100644 --- a/target/riscv/helper.h +++ b/target/riscv/helper.h @@ -66,6 +66,9 @@ DEF_HELPER_FLAGS_2(clmulr, TCG_CALL_NO_RWG_SE, tl, tl, tl) DEF_HELPER_2(csrr, tl, env, int) DEF_HELPER_3(csrw, void, env, int, tl) DEF_HELPER_4(csrrw, tl, env, int, tl, tl) +DEF_HELPER_2(csrr_i128, tl, env, int) +DEF_HELPER_4(csrw_i128, void, env, int, tl, tl) +DEF_HELPER_6(csrrw_i128, tl, env, int, tl, tl, tl, tl) #ifndef CONFIG_USER_ONLY DEF_HELPER_2(sret, tl, env, tl) DEF_HELPER_2(mret, tl, env, tl) diff --git a/target/riscv/csr.c b/target/riscv/csr.c index 69e4d65fcd..8a791d6b7c 100644 --- a/target/riscv/csr.c +++ b/target/riscv/csr.c @@ -1516,6 +1516,23 @@ RISCVException riscv_csrrw(CPURISCVState *env, int csrno, return RISCV_EXCP_NONE; } +RISCVException riscv_csrrw_i128(CPURISCVState *env, int csrno, + Int128 *ret_value, + Int128 new_value, Int128 write_mask) +{ + /* fall back to 64-bit version for now */ + target_ulong ret_64; + RISCVException ret = riscv_csrrw(env, csrno, &ret_64, + int128_getlo(new_value), + int128_getlo(write_mask)); + + if (ret_value) { + *ret_value = int128_make64(ret_64); + } + + return ret; +} + /* * Debugger support. If not in user mode, set env->debugger before the * riscv_csrrw call and clear it after the call. diff --git a/target/riscv/op_helper.c b/target/riscv/op_helper.c index ee7c24efe7..f4cf9c4698 100644 --- a/target/riscv/op_helper.c +++ b/target/riscv/op_helper.c @@ -69,6 +69,50 @@ target_ulong helper_csrrw(CPURISCVState *env, int csr, return val; } +target_ulong helper_csrr_i128(CPURISCVState *env, int csr) +{ + Int128 rv = int128_zero(); + RISCVException ret = riscv_csrrw_i128(env, csr, &rv, + int128_zero(), + int128_zero()); + + if (ret != RISCV_EXCP_NONE) { + riscv_raise_exception(env, ret, GETPC()); + } + + env->retxh = int128_gethi(rv); + return int128_getlo(rv); +} + +void helper_csrw_i128(CPURISCVState *env, int csr, + target_ulong srcl, target_ulong srch) +{ + RISCVException ret = riscv_csrrw_i128(env, csr, NULL, + int128_make128(srcl, srch), + UINT128_MAX); + + if (ret != RISCV_EXCP_NONE) { + riscv_raise_exception(env, ret, GETPC()); + } +} + +target_ulong helper_csrrw_i128(CPURISCVState *env, int csr, + target_ulong srcl, target_ulong srch, + target_ulong maskl, target_ulong maskh) +{ + Int128 rv = int128_zero(); + RISCVException ret = riscv_csrrw_i128(env, csr, &rv, + int128_make128(srcl, srch), + int128_make128(maskl, maskh)); + + if (ret != RISCV_EXCP_NONE) { + riscv_raise_exception(env, ret, GETPC()); + } + + env->retxh = int128_gethi(rv); + return int128_getlo(rv); +} + #ifndef CONFIG_USER_ONLY target_ulong helper_sret(CPURISCVState *env, target_ulong cpu_pc_deb) From patchwork Mon Oct 25 12:28:17 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Patchwork-Submitter: =?utf-8?b?RnLDqWTDqXJpYyBQw6l0cm90?= X-Patchwork-Id: 12581819 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 94A59C433EF for ; Mon, 25 Oct 2021 13:04:02 +0000 (UTC) Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.kernel.org (Postfix) with ESMTPS id 0599460F0F for ; Mon, 25 Oct 2021 13:04:01 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.4.1 mail.kernel.org 0599460F0F Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=univ-grenoble-alpes.fr Authentication-Results: mail.kernel.org; spf=pass smtp.mailfrom=nongnu.org Received: from localhost ([::1]:41280 helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1mezdw-0000OR-Ld for qemu-devel@archiver.kernel.org; Mon, 25 Oct 2021 09:04:00 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]:53220) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1mez6D-0006Tf-1H; Mon, 25 Oct 2021 08:29:09 -0400 Received: from zm-mta-out-3.u-ga.fr ([152.77.200.56]:43118) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1mez69-00087W-Oa; Mon, 25 Oct 2021 08:29:08 -0400 Received: from mailhost.u-ga.fr (mailhost2.u-ga.fr [129.88.177.242]) by zm-mta-out-3.u-ga.fr (Postfix) with ESMTP id 233FC41AD9; Mon, 25 Oct 2021 14:28:36 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=univ-grenoble-alpes.fr; s=2020; t=1635164916; bh=gEnSLFooE3arbYQ+L2dczgE9gCPMYLtCW0YoQJJlKlg=; h=From:To:Cc:Subject:Date:In-Reply-To:References:From; b=cbmtA4/2ifA9uWa+ZNHuV6gBHwIMaZ7FqQIIrPgsBsj6MyDUcPOAYJkgbVtcifqXI tJ2oxGq1ptVkhn9kmR1PEWdb0+wOnQhyBut8GQZXVtf6mpArA3YKMaYAGakd93cdVf x9Cw6v6+51Xxpmxj4bSw4DSt7231GJjE/qIptJz+2SItrImmCQKQ3ktSHdj9LfbLEH 2GzhOqtVYrXu6NriKQhvU7EeAzUl9etNnHpQph/GyFPWterijSrez4RZwZob+sbWwz TAl8EHByuLJcicthpeKsrnlsaqbACGkzIqBQl2D2Q72hgr/5LIz+BF0PtRxqWQMbRw Y7F6IuDPeZOsQ== Received: from smtps.univ-grenoble-alpes.fr (smtps2.u-ga.fr [152.77.18.2]) by mailhost.u-ga.fr (Postfix) with ESMTP id 054CE60066; Mon, 25 Oct 2021 14:28:36 +0200 (CEST) Received: from palmier.tima.u-ga.fr (35.201.90.79.rev.sfr.net [79.90.201.35]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (2048 bits) server-digest SHA256) (No client certificate requested) (Authenticated sender: petrotf@univ-grenoble-alpes.fr) by smtps.univ-grenoble-alpes.fr (Postfix) with ESMTPSA id C379F14005D; Mon, 25 Oct 2021 14:28:35 +0200 (CEST) From: =?utf-8?b?RnLDqWTDqXJpYyBQw6l0cm90?= To: qemu-devel@nongnu.org, qemu-riscv@nongnu.org Subject: [PATCH v4 16/17] target/riscv: modification of the trans_csrxx for 128-bit support Date: Mon, 25 Oct 2021 14:28:17 +0200 Message-Id: <20211025122818.168890-17-frederic.petrot@univ-grenoble-alpes.fr> X-Mailer: git-send-email 2.33.0 In-Reply-To: <20211025122818.168890-1-frederic.petrot@univ-grenoble-alpes.fr> References: <20211025122818.168890-1-frederic.petrot@univ-grenoble-alpes.fr> MIME-Version: 1.0 X-Greylist: Whitelist-UGA SMTP Authentifie (petrotf@univ-grenoble-alpes.fr) via submission-587 ACL (41) X-Greylist: Whitelist-UGA MAILHOST (SMTP non authentifie) depuis 152.77.18.2 Received-SPF: pass client-ip=152.77.200.56; envelope-from=frederic.petrot@univ-grenoble-alpes.fr; helo=zm-mta-out-3.u-ga.fr X-Spam_score_int: -20 X-Spam_score: -2.1 X-Spam_bar: -- X-Spam_report: (-2.1 / 5.0 requ) BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_MSPIKE_H2=-0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: bin.meng@windriver.com, richard.henderson@linaro.org, alistair.francis@wdc.com, fabien.portas@grenoble-inp.org, palmer@dabbelt.com, =?utf-8?b?RnLDqWTDqXJpYyBQw6l0cm90?= , philmd@redhat.com Errors-To: qemu-devel-bounces+qemu-devel=archiver.kernel.org@nongnu.org Sender: "Qemu-devel" As opposed to the gen_arith and gen_shift generation helpers, the csr insns do not have a common prototype, so the choice to generate 32/64 or 128-bit helper calls is done in the trans_csrxx functions. Signed-off-by: Frédéric Pétrot Co-authored-by: Fabien Portas Reviewed-by: Richard Henderson --- target/riscv/insn_trans/trans_rvi.c.inc | 205 ++++++++++++++++++------ 1 file changed, 160 insertions(+), 45 deletions(-) diff --git a/target/riscv/insn_trans/trans_rvi.c.inc b/target/riscv/insn_trans/trans_rvi.c.inc index 254c335b53..1b4e76ef24 100644 --- a/target/riscv/insn_trans/trans_rvi.c.inc +++ b/target/riscv/insn_trans/trans_rvi.c.inc @@ -921,20 +921,78 @@ static bool do_csrrw(DisasContext *ctx, int rd, int rc, TCGv src, TCGv mask) return do_csr_post(ctx); } +static bool do_csrr_i128(DisasContext *ctx, int rd, int rc) +{ + TCGv destl = dest_gpr(ctx, rd); + TCGv desth = dest_gprh(ctx, rd); + TCGv_i32 csr = tcg_constant_i32(rc); + + if (tb_cflags(ctx->base.tb) & CF_USE_ICOUNT) { + gen_io_start(); + } + gen_helper_csrr_i128(destl, cpu_env, csr); + tcg_gen_ld_tl(desth, cpu_env, offsetof(CPURISCVState, retxh)); + gen_set_gpr128(ctx, rd, destl, desth); + return do_csr_post(ctx); +} + +static bool do_csrw_i128(DisasContext *ctx, int rc, TCGv srcl, TCGv srch) +{ + TCGv_i32 csr = tcg_constant_i32(rc); + + if (tb_cflags(ctx->base.tb) & CF_USE_ICOUNT) { + gen_io_start(); + } + gen_helper_csrw_i128(cpu_env, csr, srcl, srch); + return do_csr_post(ctx); +} + +static bool do_csrrw_i128(DisasContext *ctx, int rd, int rc, + TCGv srcl, TCGv srch, TCGv maskl, TCGv maskh) +{ + TCGv destl = dest_gpr(ctx, rd); + TCGv desth = dest_gprh(ctx, rd); + TCGv_i32 csr = tcg_constant_i32(rc); + + if (tb_cflags(ctx->base.tb) & CF_USE_ICOUNT) { + gen_io_start(); + } + gen_helper_csrrw_i128(destl, cpu_env, csr, srcl, srch, maskl, maskh); + tcg_gen_ld_tl(desth, cpu_env, offsetof(CPURISCVState, retxh)); + gen_set_gpr128(ctx, rd, destl, desth); + return do_csr_post(ctx); +} + static bool trans_csrrw(DisasContext *ctx, arg_csrrw *a) { - TCGv src = get_gpr(ctx, a->rs1, EXT_NONE); - - /* - * If rd == 0, the insn shall not read the csr, nor cause any of the - * side effects that might occur on a csr read. - */ - if (a->rd == 0) { - return do_csrw(ctx, a->csr, src); + if (get_xl(ctx) < MXL_RV128) { + TCGv src = get_gpr(ctx, a->rs1, EXT_NONE); + + /* + * If rd == 0, the insn shall not read the csr, nor cause any of the + * side effects that might occur on a csr read. + */ + if (a->rd == 0) { + return do_csrw(ctx, a->csr, src); + } + + TCGv mask = tcg_constant_tl(-1); + return do_csrrw(ctx, a->rd, a->csr, src, mask); + } else { + TCGv srcl = get_gpr(ctx, a->rs1, EXT_NONE); + TCGv srch = get_gprh(ctx, a->rs1); + + /* + * If rd == 0, the insn shall not read the csr, nor cause any of the + * side effects that might occur on a csr read. + */ + if (a->rd == 0) { + return do_csrw_i128(ctx, a->csr, srcl, srch); + } + + TCGv mask = tcg_constant_tl(-1); + return do_csrrw_i128(ctx, a->rd, a->csr, srcl, srch, mask, mask); } - - TCGv mask = tcg_constant_tl(-1); - return do_csrrw(ctx, a->rd, a->csr, src, mask); } static bool trans_csrrs(DisasContext *ctx, arg_csrrs *a) @@ -946,13 +1004,24 @@ static bool trans_csrrs(DisasContext *ctx, arg_csrrs *a) * a zero value, the instruction will still attempt to write the * unmodified value back to the csr and will cause side effects. */ - if (a->rs1 == 0) { - return do_csrr(ctx, a->rd, a->csr); + if (get_xl(ctx) < MXL_RV128) { + if (a->rs1 == 0) { + return do_csrr(ctx, a->rd, a->csr); + } + + TCGv ones = tcg_constant_tl(-1); + TCGv mask = get_gpr(ctx, a->rs1, EXT_ZERO); + return do_csrrw(ctx, a->rd, a->csr, ones, mask); + } else { + if (a->rs1 == 0) { + return do_csrr_i128(ctx, a->rd, a->csr); + } + + TCGv ones = tcg_constant_tl(-1); + TCGv maskl = get_gpr(ctx, a->rs1, EXT_ZERO); + TCGv maskh = get_gprh(ctx, a->rs1); + return do_csrrw_i128(ctx, a->rd, a->csr, ones, ones, maskl, maskh); } - - TCGv ones = tcg_constant_tl(-1); - TCGv mask = get_gpr(ctx, a->rs1, EXT_ZERO); - return do_csrrw(ctx, a->rd, a->csr, ones, mask); } static bool trans_csrrc(DisasContext *ctx, arg_csrrc *a) @@ -964,28 +1033,54 @@ static bool trans_csrrc(DisasContext *ctx, arg_csrrc *a) * a zero value, the instruction will still attempt to write the * unmodified value back to the csr and will cause side effects. */ - if (a->rs1 == 0) { - return do_csrr(ctx, a->rd, a->csr); + if (get_xl(ctx) < MXL_RV128) { + if (a->rs1 == 0) { + return do_csrr(ctx, a->rd, a->csr); + } + + TCGv mask = get_gpr(ctx, a->rs1, EXT_ZERO); + return do_csrrw(ctx, a->rd, a->csr, ctx->zero, mask); + } else { + if (a->rs1 == 0) { + return do_csrr_i128(ctx, a->rd, a->csr); + } + + TCGv maskl = get_gpr(ctx, a->rs1, EXT_ZERO); + TCGv maskh = get_gprh(ctx, a->rs1); + return do_csrrw_i128(ctx, a->rd, a->csr, + ctx->zero, ctx->zero, maskl, maskh); } - - TCGv mask = get_gpr(ctx, a->rs1, EXT_ZERO); - return do_csrrw(ctx, a->rd, a->csr, ctx->zero, mask); } static bool trans_csrrwi(DisasContext *ctx, arg_csrrwi *a) { - TCGv src = tcg_constant_tl(a->rs1); - - /* - * If rd == 0, the insn shall not read the csr, nor cause any of the - * side effects that might occur on a csr read. - */ - if (a->rd == 0) { - return do_csrw(ctx, a->csr, src); + if (get_xl(ctx) < MXL_RV128) { + TCGv src = tcg_constant_tl(a->rs1); + + /* + * If rd == 0, the insn shall not read the csr, nor cause any of the + * side effects that might occur on a csr read. + */ + if (a->rd == 0) { + return do_csrw(ctx, a->csr, src); + } + + TCGv mask = tcg_constant_tl(-1); + return do_csrrw(ctx, a->rd, a->csr, src, mask); + } else { + TCGv src = tcg_constant_tl(a->rs1); + + /* + * If rd == 0, the insn shall not read the csr, nor cause any of the + * side effects that might occur on a csr read. + */ + if (a->rd == 0) { + return do_csrw_i128(ctx, a->csr, src, ctx->zero); + } + + TCGv mask = tcg_constant_tl(-1); + return do_csrrw_i128(ctx, a->rd, a->csr, src, ctx->zero, mask, mask); } - - TCGv mask = tcg_constant_tl(-1); - return do_csrrw(ctx, a->rd, a->csr, src, mask); } static bool trans_csrrsi(DisasContext *ctx, arg_csrrsi *a) @@ -997,16 +1092,26 @@ static bool trans_csrrsi(DisasContext *ctx, arg_csrrsi *a) * a zero value, the instruction will still attempt to write the * unmodified value back to the csr and will cause side effects. */ - if (a->rs1 == 0) { - return do_csrr(ctx, a->rd, a->csr); + if (get_xl(ctx) < MXL_RV128) { + if (a->rs1 == 0) { + return do_csrr(ctx, a->rd, a->csr); + } + + TCGv ones = tcg_constant_tl(-1); + TCGv mask = tcg_constant_tl(a->rs1); + return do_csrrw(ctx, a->rd, a->csr, ones, mask); + } else { + if (a->rs1 == 0) { + return do_csrr_i128(ctx, a->rd, a->csr); + } + + TCGv ones = tcg_constant_tl(-1); + TCGv mask = tcg_constant_tl(a->rs1); + return do_csrrw_i128(ctx, a->rd, a->csr, ones, ones, mask, ctx->zero); } - - TCGv ones = tcg_constant_tl(-1); - TCGv mask = tcg_constant_tl(a->rs1); - return do_csrrw(ctx, a->rd, a->csr, ones, mask); } -static bool trans_csrrci(DisasContext *ctx, arg_csrrci *a) +static bool trans_csrrci(DisasContext *ctx, arg_csrrci * a) { /* * If rs1 == 0, the insn shall not write to the csr at all, nor @@ -1015,10 +1120,20 @@ static bool trans_csrrci(DisasContext *ctx, arg_csrrci *a) * a zero value, the instruction will still attempt to write the * unmodified value back to the csr and will cause side effects. */ - if (a->rs1 == 0) { - return do_csrr(ctx, a->rd, a->csr); + if (get_xl(ctx) < MXL_RV128) { + if (a->rs1 == 0) { + return do_csrr(ctx, a->rd, a->csr); + } + + TCGv mask = tcg_constant_tl(a->rs1); + return do_csrrw(ctx, a->rd, a->csr, ctx->zero, mask); + } else { + if (a->rs1 == 0) { + return do_csrr_i128(ctx, a->rd, a->csr); + } + + TCGv mask = tcg_constant_tl(a->rs1); + return do_csrrw_i128(ctx, a->rd, a->csr, + ctx->zero, ctx->zero, mask, ctx->zero); } - - TCGv mask = tcg_constant_tl(a->rs1); - return do_csrrw(ctx, a->rd, a->csr, ctx->zero, mask); } From patchwork Mon Oct 25 12:28:18 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Patchwork-Submitter: =?utf-8?b?RnLDqWTDqXJpYyBQw6l0cm90?= X-Patchwork-Id: 12581821 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 3D604C433F5 for ; Mon, 25 Oct 2021 13:08:13 +0000 (UTC) Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.kernel.org (Postfix) with ESMTPS id B954260EB1 for ; Mon, 25 Oct 2021 13:08:12 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.4.1 mail.kernel.org B954260EB1 Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=univ-grenoble-alpes.fr Authentication-Results: mail.kernel.org; spf=pass smtp.mailfrom=nongnu.org Received: from localhost ([::1]:50614 helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1mezhz-0006vR-Oa for qemu-devel@archiver.kernel.org; Mon, 25 Oct 2021 09:08:11 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]:53266) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1mez6F-0006Vh-86; Mon, 25 Oct 2021 08:29:11 -0400 Received: from zm-mta-out-3.u-ga.fr ([152.77.200.56]:43124) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1mez69-00087v-P2; Mon, 25 Oct 2021 08:29:10 -0400 Received: from mailhost.u-ga.fr (mailhost2.u-ga.fr [129.88.177.242]) by zm-mta-out-3.u-ga.fr (Postfix) with ESMTP id 6036841AF5; Mon, 25 Oct 2021 14:28:36 +0200 (CEST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=univ-grenoble-alpes.fr; s=2020; t=1635164916; bh=LHXXR6Q50Mnx7d6EjTMhVtLQyiFgFNK/2y3j4fnKcAo=; h=From:To:Cc:Subject:Date:In-Reply-To:References:From; b=Fros4MempNgsmbyn2ZIDFRdz0qG3hv0Si0iP40SwAZqUeyQ9o3pTqpC1Y58Q5r4YH Td6guK88Gs0djPH0GJq3znvASAJhsOpT7gOwILxRIpogx2+Igumk2+ubdrlDf68tDp +rBdeQyW9kvZF5PpF6eiCrqTZ00sKLfnntqoFKJ65vEI8QGitGF9FdJCaYu8NwSqHe BXl4Ka/IftsQ4fI7OE5QirgqrIkAWNICjVxnuEIjz3kaq0YFJUND2+uBDwsWaaYli7 q4ShF4I52KBYNSHNHBjw+4e0wTaxPnwIxiMk9p7+j0FRrn+VzxOtuuEUecfs0MqiU0 hfxybG3mCPkjg== Received: from smtps.univ-grenoble-alpes.fr (smtps2.u-ga.fr [152.77.18.2]) by mailhost.u-ga.fr (Postfix) with ESMTP id 477FB60067; Mon, 25 Oct 2021 14:28:36 +0200 (CEST) Received: from palmier.tima.u-ga.fr (35.201.90.79.rev.sfr.net [79.90.201.35]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (2048 bits) server-digest SHA256) (No client certificate requested) (Authenticated sender: petrotf@univ-grenoble-alpes.fr) by smtps.univ-grenoble-alpes.fr (Postfix) with ESMTPSA id 0CE3414005A; Mon, 25 Oct 2021 14:28:36 +0200 (CEST) From: =?utf-8?b?RnLDqWTDqXJpYyBQw6l0cm90?= To: qemu-devel@nongnu.org, qemu-riscv@nongnu.org Subject: [PATCH v4 17/17] target/riscv: actual functions to realize crs 128-bit insns Date: Mon, 25 Oct 2021 14:28:18 +0200 Message-Id: <20211025122818.168890-18-frederic.petrot@univ-grenoble-alpes.fr> X-Mailer: git-send-email 2.33.0 In-Reply-To: <20211025122818.168890-1-frederic.petrot@univ-grenoble-alpes.fr> References: <20211025122818.168890-1-frederic.petrot@univ-grenoble-alpes.fr> MIME-Version: 1.0 X-Greylist: Whitelist-UGA SMTP Authentifie (petrotf@univ-grenoble-alpes.fr) via submission-587 ACL (41) X-Greylist: Whitelist-UGA MAILHOST (SMTP non authentifie) depuis 152.77.18.2 Received-SPF: pass client-ip=152.77.200.56; envelope-from=frederic.petrot@univ-grenoble-alpes.fr; helo=zm-mta-out-3.u-ga.fr X-Spam_score_int: -20 X-Spam_score: -2.1 X-Spam_bar: -- X-Spam_report: (-2.1 / 5.0 requ) BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, DKIM_VALID_EF=-0.1, RCVD_IN_MSPIKE_H2=-0.001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: bin.meng@windriver.com, richard.henderson@linaro.org, alistair.francis@wdc.com, fabien.portas@grenoble-inp.org, palmer@dabbelt.com, =?utf-8?b?RnLDqWTDqXJpYyBQw6l0cm90?= , philmd@redhat.com Errors-To: qemu-devel-bounces+qemu-devel=archiver.kernel.org@nongnu.org Sender: "Qemu-devel" The csrs are accessed through function pointers: we add read operations in the table for two 128-bit accesses (writes fallback to the 64-bit version): - misa, as mxl is needed for proper operation, - mstatus, as sd needs to be known also In addition, we also add read and write accesses to the machine and supervisor scratch registers. Signed-off-by: Frédéric Pétrot Co-authored-by: Fabien Portas --- target/riscv/cpu.h | 12 +++ target/riscv/cpu_bits.h | 1 + target/riscv/csr.c | 192 +++++++++++++++++++++++++++++++++------- 3 files changed, 172 insertions(+), 33 deletions(-) diff --git a/target/riscv/cpu.h b/target/riscv/cpu.h index 328ba42a30..d1af4a9ba3 100644 --- a/target/riscv/cpu.h +++ b/target/riscv/cpu.h @@ -474,12 +474,24 @@ RISCVException riscv_csrrw_i128(CPURISCVState *env, int csrno, Int128 *ret_value, Int128 new_value, Int128 write_mask); +typedef RISCVException (*riscv_csr_read128_fn)(CPURISCVState *env, int csrno, + Int128 *ret_value); +typedef RISCVException (*riscv_csr_write128_fn)(CPURISCVState *env, int csrno, + Int128 new_value); +typedef RISCVException (*riscv_csr_op128_fn)(CPURISCVState *env, int csrno, + Int128 *ret_value, + Int128 new_value, + Int128 write_mask); + typedef struct { const char *name; riscv_csr_predicate_fn predicate; riscv_csr_read_fn read; riscv_csr_write_fn write; riscv_csr_op_fn op; + riscv_csr_read128_fn read128; + riscv_csr_write128_fn write128; + riscv_csr_op128_fn op128; } riscv_csr_operations; /* CSR function table constants */ diff --git a/target/riscv/cpu_bits.h b/target/riscv/cpu_bits.h index cffcd3a5df..1f8fda2d83 100644 --- a/target/riscv/cpu_bits.h +++ b/target/riscv/cpu_bits.h @@ -360,6 +360,7 @@ #define MSTATUS32_SD 0x80000000 #define MSTATUS64_SD 0x8000000000000000ULL +#define MSTATUSH128_SD 0x8000000000000000ULL #define MISA32_MXL 0xC0000000 #define MISA64_MXL 0xC000000000000000ULL diff --git a/target/riscv/csr.c b/target/riscv/csr.c index 8a791d6b7c..68cf74a92c 100644 --- a/target/riscv/csr.c +++ b/target/riscv/csr.c @@ -488,6 +488,8 @@ static uint64_t add_status_sd(RISCVMXL xl, uint64_t status) return status | MSTATUS32_SD; case MXL_RV64: return status | MSTATUS64_SD; + case MXL_RV128: + return MSTATUSH128_SD; default: g_assert_not_reached(); } @@ -537,10 +539,11 @@ static RISCVException write_mstatus(CPURISCVState *env, int csrno, mstatus = (mstatus & ~mask) | (val & mask); - if (riscv_cpu_mxl(env) == MXL_RV64) { + RISCVMXL xl = riscv_cpu_mxl(env); + if (xl > MXL_RV32) { /* SXL and UXL fields are for now read only */ - mstatus = set_field(mstatus, MSTATUS64_SXL, MXL_RV64); - mstatus = set_field(mstatus, MSTATUS64_UXL, MXL_RV64); + mstatus = set_field(mstatus, MSTATUS64_SXL, xl); + mstatus = set_field(mstatus, MSTATUS64_UXL, xl); } env->mstatus = mstatus; @@ -569,6 +572,20 @@ static RISCVException write_mstatush(CPURISCVState *env, int csrno, return RISCV_EXCP_NONE; } +static RISCVException read_mstatus_i128(CPURISCVState *env, int csrno, + Int128 *val) +{ + *val = int128_make128(env->mstatus, add_status_sd(riscv_cpu_mxl(env), 0)); + return RISCV_EXCP_NONE; +} + +static RISCVException read_misa_i128(CPURISCVState *env, int csrno, + Int128 *val) +{ + *val = int128_make128(env->misa_ext, (uint64_t)MXL_RV128 << 62); + return RISCV_EXCP_NONE; +} + static RISCVException read_misa(CPURISCVState *env, int csrno, target_ulong *val) { @@ -726,6 +743,21 @@ static RISCVException write_mcounteren(CPURISCVState *env, int csrno, } /* Machine Trap Handling */ +static RISCVException read_mscratch_i128(CPURISCVState *env, int csrno, + Int128 *val) +{ + *val = int128_make128(env->mscratch, env->mscratchh); + return RISCV_EXCP_NONE; +} + +static RISCVException write_mscratch_i128(CPURISCVState *env, int csrno, + Int128 val) +{ + env->mscratch = int128_getlo(val); + env->mscratchh = int128_gethi(val); + return RISCV_EXCP_NONE; +} + static RISCVException read_mscratch(CPURISCVState *env, int csrno, target_ulong *val) { @@ -898,6 +930,21 @@ static RISCVException write_scounteren(CPURISCVState *env, int csrno, } /* Supervisor Trap Handling */ +static RISCVException read_sscratch_i128(CPURISCVState *env, int csrno, + Int128 *val) +{ + *val = int128_make128(env->sscratch, env->sscratchh); + return RISCV_EXCP_NONE; +} + +static RISCVException write_sscratch_i128(CPURISCVState *env, int csrno, + Int128 val) +{ + env->sscratch = int128_getlo(val); + env->sscratchh = int128_gethi(val); + return RISCV_EXCP_NONE; +} + static RISCVException read_sscratch(CPURISCVState *env, int csrno, target_ulong *val) { @@ -1436,18 +1483,15 @@ static RISCVException write_pmpaddr(CPURISCVState *env, int csrno, * csrrc <-> riscv_csrrw(env, csrno, ret_value, 0, value); */ -RISCVException riscv_csrrw(CPURISCVState *env, int csrno, - target_ulong *ret_value, - target_ulong new_value, target_ulong write_mask) +static inline RISCVException riscv_csrrw_check(CPURISCVState *env, + int csrno, + bool write_mask, + RISCVCPU *cpu) { - RISCVException ret; - target_ulong old_value; - RISCVCPU *cpu = env_archcpu(env); - int read_only = get_field(csrno, 0xC00) == 3; - - /* check privileges and return RISCV_EXCP_ILLEGAL_INST if check fails */ + /* check privileges and return -1 if check fails */ #if !defined(CONFIG_USER_ONLY) int effective_priv = env->priv; + int read_only = get_field(csrno, 0xc00) == 3; if (riscv_has_ext(env, RVH) && env->priv == PRV_S && @@ -1460,13 +1504,11 @@ RISCVException riscv_csrrw(CPURISCVState *env, int csrno, effective_priv++; } - if (!env->debugger && (effective_priv < get_field(csrno, 0x300))) { + if ((write_mask && read_only) || + (!env->debugger && (effective_priv < get_field(csrno, 0x300)))) { return RISCV_EXCP_ILLEGAL_INST; } #endif - if (write_mask && read_only) { - return RISCV_EXCP_ILLEGAL_INST; - } /* ensure the CSR extension is enabled. */ if (!cpu->cfg.ext_icsr) { @@ -1477,10 +1519,17 @@ RISCVException riscv_csrrw(CPURISCVState *env, int csrno, if (!csr_ops[csrno].predicate) { return RISCV_EXCP_ILLEGAL_INST; } - ret = csr_ops[csrno].predicate(env, csrno); - if (ret != RISCV_EXCP_NONE) { - return ret; - } + + return csr_ops[csrno].predicate(env, csrno); +} + +static RISCVException riscv_csrrw_do64(CPURISCVState *env, int csrno, + target_ulong *ret_value, + target_ulong new_value, + target_ulong write_mask) +{ + RISCVException ret; + target_ulong old_value; /* execute combined read/write operation if it exists */ if (csr_ops[csrno].op) { @@ -1516,20 +1565,93 @@ RISCVException riscv_csrrw(CPURISCVState *env, int csrno, return RISCV_EXCP_NONE; } +RISCVException riscv_csrrw(CPURISCVState *env, int csrno, + target_ulong *ret_value, + target_ulong new_value, target_ulong write_mask) +{ + RISCVCPU *cpu = env_archcpu(env); + + RISCVException ret = riscv_csrrw_check(env, csrno, write_mask, cpu); + if (ret != RISCV_EXCP_NONE) { + return ret; + } + + return riscv_csrrw_do64(env, csrno, ret_value, new_value, write_mask); +} + +static RISCVException riscv_csrrw_do128(CPURISCVState *env, int csrno, + Int128 *ret_value, + Int128 new_value, + Int128 write_mask) +{ + RISCVException ret; + Int128 old_value; + + /* execute combined read/write operation if it exists */ + if (csr_ops[csrno].op128) { + return csr_ops[csrno].op128(env, csrno, ret_value, + new_value, write_mask); + } + + /* if no accessor exists then return failure */ + if (!csr_ops[csrno].read128) { + return RISCV_EXCP_ILLEGAL_INST; + } + /* read old value */ + ret = csr_ops[csrno].read128(env, csrno, &old_value); + if (ret != RISCV_EXCP_NONE) { + return ret; + } + + /* write value if writable and write mask set, otherwise drop writes */ + if (int128_nz(write_mask)) { + new_value = int128_or(int128_and(old_value, int128_not(write_mask)), + int128_and(new_value, write_mask)); + if (csr_ops[csrno].write128) { + ret = csr_ops[csrno].write128(env, csrno, new_value); + if (ret != RISCV_EXCP_NONE) { + return ret; + } + } + } + + /* return old value */ + if (ret_value) { + *ret_value = old_value; + } + + return RISCV_EXCP_NONE; +} + RISCVException riscv_csrrw_i128(CPURISCVState *env, int csrno, - Int128 *ret_value, - Int128 new_value, Int128 write_mask) + Int128 *ret_value, + Int128 new_value, Int128 write_mask) { - /* fall back to 64-bit version for now */ - target_ulong ret_64; - RISCVException ret = riscv_csrrw(env, csrno, &ret_64, - int128_getlo(new_value), - int128_getlo(write_mask)); + RISCVException ret; + RISCVCPU *cpu = env_archcpu(env); - if (ret_value) { - *ret_value = int128_make64(ret_64); + ret = riscv_csrrw_check(env, csrno, int128_nz(write_mask), cpu); + if (ret != RISCV_EXCP_NONE) { + return ret; } + if (csr_ops[csrno].read128 || csr_ops[csrno].op128) { + return riscv_csrrw_do128(env, csrno, ret_value, new_value, write_mask); + } + + /* + * Fall back to 64-bit version for now, if the 128-bit + * alternative isn't defined. + * Note, some CSRs don't extend to MXLEN, for those, + * this fallback is correctly handling the read/write. + */ + target_ulong old_value; + ret = riscv_csrrw_do64(env, csrno, &old_value, + int128_getlo(new_value), + int128_getlo(write_mask)); + if (ret == RISCV_EXCP_NONE) { + *ret_value = int128_make64(old_value); + } return ret; } @@ -1592,8 +1714,10 @@ riscv_csr_operations csr_ops[CSR_TABLE_SIZE] = { [CSR_MHARTID] = { "mhartid", any, read_mhartid }, /* Machine Trap Setup */ - [CSR_MSTATUS] = { "mstatus", any, read_mstatus, write_mstatus }, - [CSR_MISA] = { "misa", any, read_misa, write_misa }, + [CSR_MSTATUS] = { "mstatus", any, read_mstatus, write_mstatus, NULL, + read_mstatus_i128 }, + [CSR_MISA] = { "misa", any, read_misa, write_misa, NULL, + read_misa_i128 }, [CSR_MIDELEG] = { "mideleg", any, read_mideleg, write_mideleg }, [CSR_MEDELEG] = { "medeleg", any, read_medeleg, write_medeleg }, [CSR_MIE] = { "mie", any, read_mie, write_mie }, @@ -1603,7 +1727,8 @@ riscv_csr_operations csr_ops[CSR_TABLE_SIZE] = { [CSR_MSTATUSH] = { "mstatush", any32, read_mstatush, write_mstatush }, /* Machine Trap Handling */ - [CSR_MSCRATCH] = { "mscratch", any, read_mscratch, write_mscratch }, + [CSR_MSCRATCH] = { "mscratch", any, read_mscratch, write_mscratch, NULL, + read_mscratch_i128, write_mscratch_i128 }, [CSR_MEPC] = { "mepc", any, read_mepc, write_mepc }, [CSR_MCAUSE] = { "mcause", any, read_mcause, write_mcause }, [CSR_MTVAL] = { "mtval", any, read_mtval, write_mtval }, @@ -1616,7 +1741,8 @@ riscv_csr_operations csr_ops[CSR_TABLE_SIZE] = { [CSR_SCOUNTEREN] = { "scounteren", smode, read_scounteren, write_scounteren }, /* Supervisor Trap Handling */ - [CSR_SSCRATCH] = { "sscratch", smode, read_sscratch, write_sscratch }, + [CSR_SSCRATCH] = { "sscratch", smode, read_sscratch, write_sscratch, NULL, + read_sscratch_i128, write_sscratch_i128 }, [CSR_SEPC] = { "sepc", smode, read_sepc, write_sepc }, [CSR_SCAUSE] = { "scause", smode, read_scause, write_scause }, [CSR_STVAL] = { "stval", smode, read_stval, write_stval },