From patchwork Wed Oct 31 13:20:11 2018 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Bastian Koppelmann X-Patchwork-Id: 10662737 Return-Path: Received: from mail.wl.linuxfoundation.org (pdx-wl-mail.web.codeaurora.org [172.30.200.125]) by pdx-korg-patchwork-2.web.codeaurora.org (Postfix) with ESMTP id 6611C14BD for ; Wed, 31 Oct 2018 13:47:52 +0000 (UTC) Received: from mail.wl.linuxfoundation.org (localhost [127.0.0.1]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id 516A72A400 for ; Wed, 31 Oct 2018 13:47:52 +0000 (UTC) Received: by mail.wl.linuxfoundation.org (Postfix, from userid 486) id 43F432A405; Wed, 31 Oct 2018 13:47:52 +0000 (UTC) X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on pdx-wl-mail.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-7.9 required=2.0 tests=BAYES_00,MAILING_LIST_MULTI, RCVD_IN_DNSWL_HI autolearn=ham version=3.3.1 Received: from lists.gnu.org (lists.gnu.org [208.118.235.17]) (using TLSv1 with cipher AES256-SHA (256/256 bits)) (No client certificate requested) by mail.wl.linuxfoundation.org (Postfix) with ESMTPS id 7152F2A400 for ; Wed, 31 Oct 2018 13:47:51 +0000 (UTC) Received: from localhost ([::1]:59615 helo=lists.gnu.org) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1gHqqc-0008DZ-Od for patchwork-qemu-devel@patchwork.kernel.org; Wed, 31 Oct 2018 09:47:50 -0400 Received: from eggs.gnu.org ([2001:4830:134:3::10]:37502) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1gHqRb-0004n2-Jy for qemu-devel@nongnu.org; Wed, 31 Oct 2018 09:22:07 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1gHqRX-00034w-Q8 for qemu-devel@nongnu.org; Wed, 31 Oct 2018 09:21:59 -0400 Received: from mail.uni-paderborn.de ([131.234.142.9]:57796) by eggs.gnu.org with esmtps (TLS1.0:RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1gHqRX-0002Cz-8g; Wed, 31 Oct 2018 09:21:55 -0400 Received: from wormulon.uni-paderborn.de ([131.234.189.22] helo=localhost.localdomain) by mail.uni-paderborn.de with esmtp (Exim 4.89 zuban) id 1gHqQa-0008LW-Jh; Wed, 31 Oct 2018 14:20:56 +0100 Received: from mail.uni-paderborn.de by wormulon with queue id 2965938-5; Wed, 31 Oct 2018 13:20:55 GMT X-Envelope-From: Received: from [131.234.87.238] (helo=mustique.hni.uni-paderborn.de) by mail.uni-paderborn.de with esmtpsa (TLS1.2:ECDHE_RSA_AES_128_GCM_SHA256:128) (Exim 4.89 nylar) id 1gHqQZ-0005dh-P9; Wed, 31 Oct 2018 14:20:55 +0100 From: Bastian Koppelmann To: mjc@sifive.com, sagark@eecs.berkeley.edu, palmer@sifive.com, kbastian@mail.uni-paderborn.de, Alistair.Francis@wdc.com Date: Wed, 31 Oct 2018 14:20:11 +0100 Message-Id: <20181031132029.4887-18-kbastian@mail.uni-paderborn.de> X-Mailer: git-send-email 2.19.1 In-Reply-To: <20181031132029.4887-1-kbastian@mail.uni-paderborn.de> References: <20181031132029.4887-1-kbastian@mail.uni-paderborn.de> MIME-Version: 1.0 X-PMX-Version: 6.4.5.2775670, Antispam-Engine: 2.7.2.2107409, Antispam-Data: 2018.10.31.131216, AntiVirus-Engine: 5.53.0, AntiVirus-Data: 2018.10.10.5530001 X-IMT-Spam-Score: 0.0 () X-IMT-Authenticated-Sender: X-detected-operating-system: by eggs.gnu.org: Genre and OS details not recognized. X-Received-From: 131.234.142.9 Subject: [Qemu-devel] [PATCH v3 17/35] target/riscv: Convert quadrant 0 of RVXC insns to decodetree X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.21 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: qemu-riscv@nongnu.org, peer.adelt@hni.uni-paderborn.de, richard.henderson@linaro.org, qemu-devel@nongnu.org Errors-To: qemu-devel-bounces+patchwork-qemu-devel=patchwork.kernel.org@nongnu.org Sender: "Qemu-devel" X-Virus-Scanned: ClamAV using ClamSMTP Reviewed-by: Richard Henderson Signed-off-by: Bastian Koppelmann Signed-off-by: Peer Adelt --- v2 -> v3: - trans_c_flw_ld and trans_c_fsw_sd now return false for now, as we cannot use the insn argument anymore for manual decoding. We fix this in a later patch when rvc is properly split up into insn16-32.decode and insn16-64.decode. - special case of trans_c_addi4spn() returns false in this patch - dropped insn argument of trans_foo functions target/riscv/Makefile.objs | 9 ++- target/riscv/insn16.decode | 55 ++++++++++++++++++ target/riscv/insn_trans/trans_rvc.inc.c | 75 +++++++++++++++++++++++++ target/riscv/translate.c | 53 ++++++----------- 4 files changed, 154 insertions(+), 38 deletions(-) create mode 100644 target/riscv/insn16.decode create mode 100644 target/riscv/insn_trans/trans_rvc.inc.c diff --git a/target/riscv/Makefile.objs b/target/riscv/Makefile.objs index b9b8152cc2..ba661b69e3 100644 --- a/target/riscv/Makefile.objs +++ b/target/riscv/Makefile.objs @@ -10,4 +10,11 @@ target/riscv/decode_insn32.inc.c: $(decode32-y) $(DECODETREE) $(PYTHON) $(DECODETREE) -o $@ --decode decode_insn32 $(decode32-y), \ "GEN", $(TARGET_DIR)$@) -target/riscv/translate.o: target/riscv/decode_insn32.inc.c +target/riscv/decode_insn16.inc.c: \ + $(SRC_PATH)/target/riscv/insn16.decode $(DECODETREE) + $(call quiet-command, \ + $(PYTHON) $(DECODETREE) -o $@ --decode decode_insn16 --insnwidth 16 $<, \ + "GEN", $(TARGET_DIR)$@) + +target/riscv/translate.o: target/riscv/decode_insn32.inc.c \ + target/riscv/decode_insn16.inc.c diff --git a/target/riscv/insn16.decode b/target/riscv/insn16.decode new file mode 100644 index 0000000000..558c0c41f0 --- /dev/null +++ b/target/riscv/insn16.decode @@ -0,0 +1,55 @@ +# +# RISC-V translation routines for the RVXI Base Integer Instruction Set. +# +# Copyright (c) 2018 Peer Adelt, peer.adelt@hni.uni-paderborn.de +# Bastian Koppelmann, kbastian@mail.uni-paderborn.de +# +# 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 . + +# Fields: +%rd 7:5 +%rs1_3 7:3 !function=ex_rvc_register +%rs2_3 2:3 !function=ex_rvc_register + +# Immediates: +%nzuimm_ciw 7:4 11:2 5:1 6:1 !function=ex_shift_2 +%uimm_cl_d 5:2 10:3 !function=ex_shift_3 +%uimm_cl_w 5:1 10:3 6:1 !function=ex_shift_2 + + +# Argument sets: +&cl rs1 rd +&cl_dw uimm rs1 rd +&ciw nzuimm rd +&cs rs1 rs2 +&cs_dw uimm rs1 rs2 + + +# Formats 16: +@ciw ... ........ ... .. &ciw nzuimm=%nzuimm_ciw rd=%rs2_3 +@cl_d ... ... ... .. ... .. &cl_dw uimm=%uimm_cl_d rs1=%rs1_3 rd=%rs2_3 +@cl_w ... ... ... .. ... .. &cl_dw uimm=%uimm_cl_w rs1=%rs1_3 rd=%rs2_3 +@cl ... ... ... .. ... .. &cl rs1=%rs1_3 rd=%rs2_3 +@cs ... ... ... .. ... .. &cs rs1=%rs1_3 rs2=%rs2_3 +@cs_d ... ... ... .. ... .. &cs_dw uimm=%uimm_cl_d rs1=%rs1_3 rs2=%rs2_3 +@cs_w ... ... ... .. ... .. &cs_dw uimm=%uimm_cl_w rs1=%rs1_3 rs2=%rs2_3 + + +# *** RV64C Standard Extension (Quadrant 0) *** +c_addi4spn 000 ........ ... 00 @ciw +c_fld 001 ... ... .. ... 00 @cl_d +c_lw 010 ... ... .. ... 00 @cl_w +c_flw_ld 011 --- ... -- ... 00 @cl #Note: Must parse uimm manually +c_fsd 101 ... ... .. ... 00 @cs_d +c_sw 110 ... ... .. ... 00 @cs_w +c_fsw_sd 111 --- ... -- ... 00 @cs #Note: Must parse uimm manually diff --git a/target/riscv/insn_trans/trans_rvc.inc.c b/target/riscv/insn_trans/trans_rvc.inc.c new file mode 100644 index 0000000000..93ec8aa30b --- /dev/null +++ b/target/riscv/insn_trans/trans_rvc.inc.c @@ -0,0 +1,75 @@ +/* + * RISC-V translation routines for the RVC Compressed Instruction Set. + * + * Copyright (c) 2016-2017 Sagar Karandikar, sagark@eecs.berkeley.edu + * Copyright (c) 2018 Peer Adelt, peer.adelt@hni.uni-paderborn.de + * Bastian Koppelmann, kbastian@mail.uni-paderborn.de + * + * 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 . + */ + +static bool trans_c_addi4spn(DisasContext *ctx, arg_c_addi4spn *a) +{ + if (a->nzuimm == 0) { + /* Reserved in ISA */ + return false; + } + arg_addi arg = { .rd = a->rd, .rs1 = 2, .imm = a->nzuimm }; + return trans_addi(ctx, &arg); +} + +static bool trans_c_fld(DisasContext *ctx, arg_c_fld *a) +{ + arg_fld arg = { .rd = a->rd, .rs1 = a->rs1, .imm = a->uimm }; + return trans_fld(ctx, &arg); +} + +static bool trans_c_lw(DisasContext *ctx, arg_c_lw *a) +{ + arg_lw arg = { .rd = a->rd, .rs1 = a->rs1, .imm = a->uimm }; + return trans_lw(ctx, &arg); +} + +static bool trans_c_flw_ld(DisasContext *ctx, arg_c_flw_ld *a) +{ +#ifdef TARGET_RISCV32 + /* C.FLW ( RV32FC-only ) */ + return false; +#else + /* C.LD ( RV64C/RV128C-only ) */ + return false; +#endif +} + +static bool trans_c_fsd(DisasContext *ctx, arg_c_fsd *a) +{ + arg_fsd arg = { .rs1 = a->rs1, .rs2 = a->rs2, .imm = a->uimm }; + return trans_fsd(ctx, &arg); +} + +static bool trans_c_sw(DisasContext *ctx, arg_c_sw *a) +{ + arg_sw arg = { .rs1 = a->rs1, .rs2 = a->rs2, .imm = a->uimm }; + return trans_sw(ctx, &arg); +} + +static bool trans_c_fsw_sd(DisasContext *ctx, arg_c_fsw_sd *a) +{ +#ifdef TARGET_RISCV32 + /* C.FSW ( RV32FC-only ) */ + return false; +#else + /* C.SD ( RV64C/RV128C-only ) */ + return false; +#endif +} diff --git a/target/riscv/translate.c b/target/riscv/translate.c index 885e48881c..8bddebb35e 100644 --- a/target/riscv/translate.c +++ b/target/riscv/translate.c @@ -721,27 +721,6 @@ static void decode_RV32_64C0(DisasContext *ctx) uint8_t rs1s = GET_C_RS1S(ctx->opcode); switch (funct3) { - case 0: - /* illegal */ - if (ctx->opcode == 0) { - gen_exception_illegal(ctx); - } else { - /* C.ADDI4SPN -> addi rd', x2, zimm[9:2]*/ - gen_arith_imm(ctx, OPC_RISC_ADDI, rd_rs2, 2, - GET_C_ADDI4SPN_IMM(ctx->opcode)); - } - break; - case 1: - /* C.FLD -> fld rd', offset[7:3](rs1')*/ - gen_fp_load(ctx, OPC_RISC_FLD, rd_rs2, rs1s, - GET_C_LD_IMM(ctx->opcode)); - /* C.LQ(RV128) */ - break; - case 2: - /* C.LW -> lw rd', offset[6:2](rs1') */ - gen_load(ctx, OPC_RISC_LW, rd_rs2, rs1s, - GET_C_LW_IMM(ctx->opcode)); - break; case 3: #if defined(TARGET_RISCV64) /* C.LD(RV64/128) -> ld rd', offset[7:3](rs1')*/ @@ -753,21 +732,6 @@ static void decode_RV32_64C0(DisasContext *ctx) GET_C_LW_IMM(ctx->opcode)); #endif break; - case 4: - /* reserved */ - gen_exception_illegal(ctx); - break; - case 5: - /* C.FSD(RV32/64) -> fsd rs2', offset[7:3](rs1') */ - gen_fp_store(ctx, OPC_RISC_FSD, rs1s, rd_rs2, - GET_C_LD_IMM(ctx->opcode)); - /* C.SQ (RV128) */ - break; - case 6: - /* C.SW -> sw rs2', offset[6:2](rs1')*/ - gen_store(ctx, OPC_RISC_SW, rs1s, rd_rs2, - GET_C_LW_IMM(ctx->opcode)); - break; case 7: #if defined(TARGET_RISCV64) /* C.SD (RV64/128) -> sd rs2', offset[7:3](rs1')*/ @@ -997,8 +961,15 @@ static void decode_RV32_64C(CPURISCVState *env, DisasContext *ctx) return imm << amount; \ } EX_SH(1) +EX_SH(2) +EX_SH(3) EX_SH(12) +static int ex_rvc_register(int reg) +{ + return 8 + reg; +} + bool decode_insn32(DisasContext *ctx, uint32_t insn); /* Include the auto-generated decoder for 32 bit insn */ #include "decode_insn32.inc.c" @@ -1010,6 +981,11 @@ bool decode_insn32(DisasContext *ctx, uint32_t insn); #include "insn_trans/trans_rvd.inc.c" #include "insn_trans/trans_privileged.inc.c" +bool decode_insn16(DisasContext *ctx, uint16_t insn); +/* auto-generated decoder*/ +#include "decode_insn16.inc.c" +#include "insn_trans/trans_rvc.inc.c" + static void decode_RV32_64G(CPURISCVState *env, DisasContext *ctx) { int rs1, rd; @@ -1043,7 +1019,10 @@ static void decode_opc(DisasContext *ctx) gen_exception_illegal(ctx); } else { ctx->pc_succ_insn = ctx->base.pc_next + 2; - decode_RV32_64C(ctx->env, ctx); + if (!decode_insn16(ctx, ctx->opcode)) { + /* fall back to old decoder */ + decode_RV32_64C(ctx->env, ctx); + } } } else { ctx->pc_succ_insn = ctx->base.pc_next + 4;