From patchwork Sun Jul 5 14:02:56 2020 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-Patchwork-Submitter: Thomas Huth X-Patchwork-Id: 11644483 Return-Path: Received: from mail.kernel.org (pdx-korg-mail-1.web.codeaurora.org [172.30.200.123]) by pdx-korg-patchwork-2.web.codeaurora.org (Postfix) with ESMTP id 74F4D739 for ; Sun, 5 Jul 2020 14:16:07 +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 458F22068F for ; Sun, 5 Jul 2020 14:16:07 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org 458F22068F Authentication-Results: mail.kernel.org; dmarc=none (p=none dis=none) header.from=tuxfamily.org Authentication-Results: mail.kernel.org; spf=pass smtp.mailfrom=qemu-devel-bounces+patchwork-qemu-devel=patchwork.kernel.org@nongnu.org Received: from localhost ([::1]:37422 helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1js5R8-0005R2-J4 for patchwork-qemu-devel@patchwork.kernel.org; Sun, 05 Jul 2020 10:16:06 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]:54122) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1js5F7-0007M9-7p for qemu-devel@nongnu.org; Sun, 05 Jul 2020 10:03:41 -0400 Received: from mail-ed1-f43.google.com ([209.85.208.43]:42810) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_128_GCM_SHA256:128) (Exim 4.90_1) (envelope-from ) id 1js5F4-0008JC-B4 for qemu-devel@nongnu.org; Sun, 05 Jul 2020 10:03:40 -0400 Received: by mail-ed1-f43.google.com with SMTP id z17so32269931edr.9 for ; Sun, 05 Jul 2020 07:03:37 -0700 (PDT) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:from:to:cc:subject:date:message-id:in-reply-to :references:mime-version:content-transfer-encoding; bh=sXQQoktsGrtJLUHF/7YnCLkoEinEFTJGZ5NeBg5xGHw=; b=Su9p69Amy4Xsm4b8fV6l2LmWpX40q6+QE0IHdFwRd5h00OOvd52x7Rab9GK1c7Uncz zNTxDqHoNjvsXP/MPuXd38UPgpqOjVzCR4ewfx/fDiU3dm2Cd7761dsPJtoY6M+h1DvA p0p/YvpAmik+7bxbI+Cq16maV8WS7DeoKtaYhOEI4XINBjKPkoAES9JG6AAkqCGw9yJA /v6XjZWcMSzMOi64PWLLAP64uZfM2p76YQrWx8B7S13L6Cm8Mc0d3TqSqXd993IvLYLo gFw1rSrVVZlUkNTuBZsSOK6ERSWSBimxyJ7/DsPwKgVBiiMHIKZOtl0mNua3v8UMUX7Z yn6g== X-Gm-Message-State: AOAM530M+viZSI56nM+n3Nviuh0Xck7vk+XcoWMFfJLHe7qhZKtsli+k zM1zi5Kc3OcokE1bzTsi1FS/jtg3 X-Google-Smtp-Source: ABdhPJxNxqq6wsha4hhNnAlxHpXt/qlRVAkJXa+xUWlCj8vkskY0r8y0OMSjMyKC2Sm04bkXtEhhng== X-Received: by 2002:aa7:c714:: with SMTP id i20mr51527625edq.215.1593957816422; Sun, 05 Jul 2020 07:03:36 -0700 (PDT) Received: from thl530.multi.box (pd9e83654.dip0.t-ipconnect.de. [217.232.54.84]) by smtp.gmail.com with ESMTPSA id bq8sm10941776ejb.103.2020.07.05.07.03.35 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Sun, 05 Jul 2020 07:03:36 -0700 (PDT) From: Thomas Huth To: qemu-devel@nongnu.org, Michael Rolnik , =?utf-8?q?Phi?= =?utf-8?q?lippe_Mathieu-Daud=C3=A9?= , Richard Henderson Subject: [PATCH rc6 11/30] target/avr: Add instruction translation - Arithmetic and Logic Instructions Date: Sun, 5 Jul 2020 16:02:56 +0200 Message-Id: <20200705140315.260514-12-huth@tuxfamily.org> X-Mailer: git-send-email 2.26.2 In-Reply-To: <20200705140315.260514-1-huth@tuxfamily.org> References: <20200705140315.260514-1-huth@tuxfamily.org> MIME-Version: 1.0 Received-SPF: pass client-ip=209.85.208.43; envelope-from=th.huth@gmail.com; helo=mail-ed1-f43.google.com X-detected-operating-system: by eggs.gnu.org: First seen = 2020/07/05 10:03:37 X-ACL-Warn: Detected OS = Linux 2.2.x-3.x [generic] [fuzzy] X-Spam_score_int: -8 X-Spam_score: -0.9 X-Spam_bar: / X-Spam_report: (-0.9 / 5.0 requ) BAYES_00=-1.9, FREEMAIL_FORGED_FROMDOMAIN=1, FREEMAIL_FROM=0.001, HEADER_FROM_DIFFERENT_DOMAINS=1, RCVD_IN_DNSWL_NONE=-0.0001, RCVD_IN_MSPIKE_H2=-1, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=_AUTOLEARN 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: Sarah Harris Errors-To: qemu-devel-bounces+patchwork-qemu-devel=patchwork.kernel.org@nongnu.org Sender: "Qemu-devel" From: Michael Rolnik This includes: - ADD, ADC, ADIW - SBIW, SUB, SUBI, SBC, SBCI - AND, ANDI - OR, ORI, EOR - COM, NEG - INC, DEC - MUL, MULS, MULSU - FMUL, FMULS, FMULSU - DES Signed-off-by: Michael Rolnik Signed-off-by: Richard Henderson Signed-off-by: Aleksandar Markovic Tested-by: Philippe Mathieu-Daudé Reviewed-by: Aleksandar Markovic Signed-off-by: Thomas Huth --- target/avr/insn.decode | 76 ++++ target/avr/translate.c | 823 +++++++++++++++++++++++++++++++++++++++++ 2 files changed, 899 insertions(+) create mode 100644 target/avr/insn.decode diff --git a/target/avr/insn.decode b/target/avr/insn.decode new file mode 100644 index 0000000000..43baf6d92f --- /dev/null +++ b/target/avr/insn.decode @@ -0,0 +1,76 @@ +# +# AVR instruction decode definitions. +# +# Copyright (c) 2019 Michael Rolnik +# +# This library is free software; you can redistribute it and/or +# modify it under the terms of the GNU Lesser General Public +# License as published by the Free Software Foundation; either +# version 2.1 of the License, or (at your option) any later version. +# +# This library is distributed in the hope that it will be useful, +# but WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU +# Lesser General Public License for more details. +# +# You should have received a copy of the GNU Lesser General Public +# License along with this library; if not, see . +# + +# +# regs_16_31_by_one = [16 .. 31] +# regs_16_23_by_one = [16 .. 23] +# regs_24_30_by_two = [24, 26, 28, 30] +# regs_00_30_by_two = [0, 2, 4, 6, 8, .. 30] + +%rd 4:5 +%rr 9:1 0:4 + +%rd_a 4:4 !function=to_regs_16_31_by_one +%rd_b 4:3 !function=to_regs_16_23_by_one +%rd_c 4:2 !function=to_regs_24_30_by_two +%rr_a 0:4 !function=to_regs_16_31_by_one +%rr_b 0:3 !function=to_regs_16_23_by_one + +%imm6 6:2 0:4 +%imm8 8:4 0:4 + +%io_imm 9:2 0:4 +%ldst_d_imm 13:1 10:2 0:3 + + +&rd_rr rd rr +&rd_imm rd imm + +@op_rd_rr .... .. . ..... .... &rd_rr rd=%rd rr=%rr +@op_rd_imm6 .... .... .. .. .... &rd_imm rd=%rd_c imm=%imm6 +@op_rd_imm8 .... .... .... .... &rd_imm rd=%rd_a imm=%imm8 +@fmul .... .... . ... . ... &rd_rr rd=%rd_b rr=%rr_b + +# +# Arithmetic Instructions +# +ADD 0000 11 . ..... .... @op_rd_rr +ADC 0001 11 . ..... .... @op_rd_rr +ADIW 1001 0110 .. .. .... @op_rd_imm6 +SUB 0001 10 . ..... .... @op_rd_rr +SUBI 0101 .... .... .... @op_rd_imm8 +SBC 0000 10 . ..... .... @op_rd_rr +SBCI 0100 .... .... .... @op_rd_imm8 +SBIW 1001 0111 .. .. .... @op_rd_imm6 +AND 0010 00 . ..... .... @op_rd_rr +ANDI 0111 .... .... .... @op_rd_imm8 +OR 0010 10 . ..... .... @op_rd_rr +ORI 0110 .... .... .... @op_rd_imm8 +EOR 0010 01 . ..... .... @op_rd_rr +COM 1001 010 rd:5 0000 +NEG 1001 010 rd:5 0001 +INC 1001 010 rd:5 0011 +DEC 1001 010 rd:5 1010 +MUL 1001 11 . ..... .... @op_rd_rr +MULS 0000 0010 .... .... &rd_rr rd=%rd_a rr=%rr_a +MULSU 0000 0011 0 ... 0 ... @fmul +FMUL 0000 0011 0 ... 1 ... @fmul +FMULS 0000 0011 1 ... 0 ... @fmul +FMULSU 0000 0011 1 ... 1 ... @fmul +DES 1001 0100 imm:4 1011 diff --git a/target/avr/translate.c b/target/avr/translate.c index 44ee8849f0..d81bfaa04b 100644 --- a/target/avr/translate.c +++ b/target/avr/translate.c @@ -129,6 +129,22 @@ struct DisasContext { }; +static int to_regs_16_31_by_one(DisasContext *ctx, int indx) +{ + return 16 + (indx % 16); +} + +static int to_regs_16_23_by_one(DisasContext *ctx, int indx) +{ + return 16 + (indx % 8); +} + +static int to_regs_24_30_by_two(DisasContext *ctx, int indx) +{ + return 24 + (indx % 4) * 2; +} + + static bool avr_have_feature(DisasContext *ctx, int feature) { if (!avr_feature(ctx->env, feature)) { @@ -141,3 +157,810 @@ static bool avr_have_feature(DisasContext *ctx, int feature) static bool decode_insn(DisasContext *ctx, uint16_t insn); #include "decode_insn.inc.c" + +/* + * Arithmetic Instructions + */ + +/* + * Utility functions for updating status registers: + * + * - gen_add_CHf() + * - gen_add_Vf() + * - gen_sub_CHf() + * - gen_sub_Vf() + * - gen_NSf() + * - gen_ZNSf() + * + */ + +static void gen_add_CHf(TCGv R, TCGv Rd, TCGv Rr) +{ + TCGv t1 = tcg_temp_new_i32(); + TCGv t2 = tcg_temp_new_i32(); + TCGv t3 = tcg_temp_new_i32(); + + tcg_gen_and_tl(t1, Rd, Rr); /* t1 = Rd & Rr */ + tcg_gen_andc_tl(t2, Rd, R); /* t2 = Rd & ~R */ + tcg_gen_andc_tl(t3, Rr, R); /* t3 = Rr & ~R */ + tcg_gen_or_tl(t1, t1, t2); /* t1 = t1 | t2 | t3 */ + tcg_gen_or_tl(t1, t1, t3); + + tcg_gen_shri_tl(cpu_Cf, t1, 7); /* Cf = t1(7) */ + tcg_gen_shri_tl(cpu_Hf, t1, 3); /* Hf = t1(3) */ + tcg_gen_andi_tl(cpu_Hf, cpu_Hf, 1); + + tcg_temp_free_i32(t3); + tcg_temp_free_i32(t2); + tcg_temp_free_i32(t1); +} + + +static void gen_add_Vf(TCGv R, TCGv Rd, TCGv Rr) +{ + TCGv t1 = tcg_temp_new_i32(); + TCGv t2 = tcg_temp_new_i32(); + + /* t1 = Rd & Rr & ~R | ~Rd & ~Rr & R */ + /* = (Rd ^ R) & ~(Rd ^ Rr) */ + tcg_gen_xor_tl(t1, Rd, R); + tcg_gen_xor_tl(t2, Rd, Rr); + tcg_gen_andc_tl(t1, t1, t2); + + tcg_gen_shri_tl(cpu_Vf, t1, 7); /* Vf = t1(7) */ + + tcg_temp_free_i32(t2); + tcg_temp_free_i32(t1); +} + + +static void gen_sub_CHf(TCGv R, TCGv Rd, TCGv Rr) +{ + TCGv t1 = tcg_temp_new_i32(); + TCGv t2 = tcg_temp_new_i32(); + TCGv t3 = tcg_temp_new_i32(); + + tcg_gen_not_tl(t1, Rd); /* t1 = ~Rd */ + tcg_gen_and_tl(t2, t1, Rr); /* t2 = ~Rd & Rr */ + tcg_gen_or_tl(t3, t1, Rr); /* t3 = (~Rd | Rr) & R */ + tcg_gen_and_tl(t3, t3, R); + tcg_gen_or_tl(t2, t2, t3); /* t2 = ~Rd & Rr | ~Rd & R | R & Rr */ + + tcg_gen_shri_tl(cpu_Cf, t2, 7); /* Cf = t2(7) */ + tcg_gen_shri_tl(cpu_Hf, t2, 3); /* Hf = t2(3) */ + tcg_gen_andi_tl(cpu_Hf, cpu_Hf, 1); + + tcg_temp_free_i32(t3); + tcg_temp_free_i32(t2); + tcg_temp_free_i32(t1); +} + + +static void gen_sub_Vf(TCGv R, TCGv Rd, TCGv Rr) +{ + TCGv t1 = tcg_temp_new_i32(); + TCGv t2 = tcg_temp_new_i32(); + + /* t1 = Rd & ~Rr & ~R | ~Rd & Rr & R */ + /* = (Rd ^ R) & (Rd ^ R) */ + tcg_gen_xor_tl(t1, Rd, R); + tcg_gen_xor_tl(t2, Rd, Rr); + tcg_gen_and_tl(t1, t1, t2); + + tcg_gen_shri_tl(cpu_Vf, t1, 7); /* Vf = t1(7) */ + + tcg_temp_free_i32(t2); + tcg_temp_free_i32(t1); +} + + +static void gen_NSf(TCGv R) +{ + tcg_gen_shri_tl(cpu_Nf, R, 7); /* Nf = R(7) */ + tcg_gen_xor_tl(cpu_Sf, cpu_Nf, cpu_Vf); /* Sf = Nf ^ Vf */ +} + + +static void gen_ZNSf(TCGv R) +{ + tcg_gen_setcondi_tl(TCG_COND_EQ, cpu_Zf, R, 0); /* Zf = R == 0 */ + + /* update status register */ + tcg_gen_shri_tl(cpu_Nf, R, 7); /* Nf = R(7) */ + tcg_gen_xor_tl(cpu_Sf, cpu_Nf, cpu_Vf); /* Sf = Nf ^ Vf */ +} + +/* + * Adds two registers without the C Flag and places the result in the + * destination register Rd. + */ +static bool trans_ADD(DisasContext *ctx, arg_ADD *a) +{ + TCGv Rd = cpu_r[a->rd]; + TCGv Rr = cpu_r[a->rr]; + TCGv R = tcg_temp_new_i32(); + + tcg_gen_add_tl(R, Rd, Rr); /* Rd = Rd + Rr */ + tcg_gen_andi_tl(R, R, 0xff); /* make it 8 bits */ + + /* update status register */ + gen_add_CHf(R, Rd, Rr); + gen_add_Vf(R, Rd, Rr); + gen_ZNSf(R); + + /* update output registers */ + tcg_gen_mov_tl(Rd, R); + + tcg_temp_free_i32(R); + + return true; +} + +/* + * Adds two registers and the contents of the C Flag and places the result in + * the destination register Rd. + */ +static bool trans_ADC(DisasContext *ctx, arg_ADC *a) +{ + TCGv Rd = cpu_r[a->rd]; + TCGv Rr = cpu_r[a->rr]; + TCGv R = tcg_temp_new_i32(); + + tcg_gen_add_tl(R, Rd, Rr); /* R = Rd + Rr + Cf */ + tcg_gen_add_tl(R, R, cpu_Cf); + tcg_gen_andi_tl(R, R, 0xff); /* make it 8 bits */ + + /* update status register */ + gen_add_CHf(R, Rd, Rr); + gen_add_Vf(R, Rd, Rr); + gen_ZNSf(R); + + /* update output registers */ + tcg_gen_mov_tl(Rd, R); + + tcg_temp_free_i32(R); + + return true; +} + +/* + * Adds an immediate value (0 - 63) to a register pair and places the result + * in the register pair. This instruction operates on the upper four register + * pairs, and is well suited for operations on the pointer registers. This + * instruction is not available in all devices. Refer to the device specific + * instruction set summary. + */ +static bool trans_ADIW(DisasContext *ctx, arg_ADIW *a) +{ + if (!avr_have_feature(ctx, AVR_FEATURE_ADIW_SBIW)) { + return true; + } + + TCGv RdL = cpu_r[a->rd]; + TCGv RdH = cpu_r[a->rd + 1]; + int Imm = (a->imm); + TCGv R = tcg_temp_new_i32(); + TCGv Rd = tcg_temp_new_i32(); + + tcg_gen_deposit_tl(Rd, RdL, RdH, 8, 8); /* Rd = RdH:RdL */ + tcg_gen_addi_tl(R, Rd, Imm); /* R = Rd + Imm */ + tcg_gen_andi_tl(R, R, 0xffff); /* make it 16 bits */ + + /* update status register */ + tcg_gen_andc_tl(cpu_Cf, Rd, R); /* Cf = Rd & ~R */ + tcg_gen_shri_tl(cpu_Cf, cpu_Cf, 15); + tcg_gen_andc_tl(cpu_Vf, R, Rd); /* Vf = R & ~Rd */ + tcg_gen_shri_tl(cpu_Vf, cpu_Vf, 15); + tcg_gen_setcondi_tl(TCG_COND_EQ, cpu_Zf, R, 0); /* Zf = R == 0 */ + tcg_gen_shri_tl(cpu_Nf, R, 15); /* Nf = R(15) */ + tcg_gen_xor_tl(cpu_Sf, cpu_Nf, cpu_Vf);/* Sf = Nf ^ Vf */ + + /* update output registers */ + tcg_gen_andi_tl(RdL, R, 0xff); + tcg_gen_shri_tl(RdH, R, 8); + + tcg_temp_free_i32(Rd); + tcg_temp_free_i32(R); + + return true; +} + +/* + * Subtracts two registers and places the result in the destination + * register Rd. + */ +static bool trans_SUB(DisasContext *ctx, arg_SUB *a) +{ + TCGv Rd = cpu_r[a->rd]; + TCGv Rr = cpu_r[a->rr]; + TCGv R = tcg_temp_new_i32(); + + tcg_gen_sub_tl(R, Rd, Rr); /* R = Rd - Rr */ + tcg_gen_andi_tl(R, R, 0xff); /* make it 8 bits */ + + /* update status register */ + tcg_gen_andc_tl(cpu_Cf, Rd, R); /* Cf = Rd & ~R */ + gen_sub_CHf(R, Rd, Rr); + gen_sub_Vf(R, Rd, Rr); + gen_ZNSf(R); + + /* update output registers */ + tcg_gen_mov_tl(Rd, R); + + tcg_temp_free_i32(R); + + return true; +} + +/* + * Subtracts a register and a constant and places the result in the + * destination register Rd. This instruction is working on Register R16 to R31 + * and is very well suited for operations on the X, Y, and Z-pointers. + */ +static bool trans_SUBI(DisasContext *ctx, arg_SUBI *a) +{ + TCGv Rd = cpu_r[a->rd]; + TCGv Rr = tcg_const_i32(a->imm); + TCGv R = tcg_temp_new_i32(); + + tcg_gen_sub_tl(R, Rd, Rr); /* R = Rd - Imm */ + tcg_gen_andi_tl(R, R, 0xff); /* make it 8 bits */ + + /* update status register */ + gen_sub_CHf(R, Rd, Rr); + gen_sub_Vf(R, Rd, Rr); + gen_ZNSf(R); + + /* update output registers */ + tcg_gen_mov_tl(Rd, R); + + tcg_temp_free_i32(R); + tcg_temp_free_i32(Rr); + + return true; +} + +/* + * Subtracts two registers and subtracts with the C Flag and places the + * result in the destination register Rd. + */ +static bool trans_SBC(DisasContext *ctx, arg_SBC *a) +{ + TCGv Rd = cpu_r[a->rd]; + TCGv Rr = cpu_r[a->rr]; + TCGv R = tcg_temp_new_i32(); + TCGv zero = tcg_const_i32(0); + + tcg_gen_sub_tl(R, Rd, Rr); /* R = Rd - Rr - Cf */ + tcg_gen_sub_tl(R, R, cpu_Cf); + tcg_gen_andi_tl(R, R, 0xff); /* make it 8 bits */ + + /* update status register */ + gen_sub_CHf(R, Rd, Rr); + gen_sub_Vf(R, Rd, Rr); + gen_NSf(R); + + /* + * Previous value remains unchanged when the result is zero; + * cleared otherwise. + */ + tcg_gen_movcond_tl(TCG_COND_EQ, cpu_Zf, R, zero, cpu_Zf, zero); + + /* update output registers */ + tcg_gen_mov_tl(Rd, R); + + tcg_temp_free_i32(zero); + tcg_temp_free_i32(R); + + return true; +} + +/* + * SBCI -- Subtract Immediate with Carry + */ +static bool trans_SBCI(DisasContext *ctx, arg_SBCI *a) +{ + TCGv Rd = cpu_r[a->rd]; + TCGv Rr = tcg_const_i32(a->imm); + TCGv R = tcg_temp_new_i32(); + TCGv zero = tcg_const_i32(0); + + tcg_gen_sub_tl(R, Rd, Rr); /* R = Rd - Rr - Cf */ + tcg_gen_sub_tl(R, R, cpu_Cf); + tcg_gen_andi_tl(R, R, 0xff); /* make it 8 bits */ + + /* update status register */ + gen_sub_CHf(R, Rd, Rr); + gen_sub_Vf(R, Rd, Rr); + gen_NSf(R); + + /* + * Previous value remains unchanged when the result is zero; + * cleared otherwise. + */ + tcg_gen_movcond_tl(TCG_COND_EQ, cpu_Zf, R, zero, cpu_Zf, zero); + + /* update output registers */ + tcg_gen_mov_tl(Rd, R); + + tcg_temp_free_i32(zero); + tcg_temp_free_i32(R); + tcg_temp_free_i32(Rr); + + return true; +} + +/* + * Subtracts an immediate value (0-63) from a register pair and places the + * result in the register pair. This instruction operates on the upper four + * register pairs, and is well suited for operations on the Pointer Registers. + * This instruction is not available in all devices. Refer to the device + * specific instruction set summary. + */ +static bool trans_SBIW(DisasContext *ctx, arg_SBIW *a) +{ + if (!avr_have_feature(ctx, AVR_FEATURE_ADIW_SBIW)) { + return true; + } + + TCGv RdL = cpu_r[a->rd]; + TCGv RdH = cpu_r[a->rd + 1]; + int Imm = (a->imm); + TCGv R = tcg_temp_new_i32(); + TCGv Rd = tcg_temp_new_i32(); + + tcg_gen_deposit_tl(Rd, RdL, RdH, 8, 8); /* Rd = RdH:RdL */ + tcg_gen_subi_tl(R, Rd, Imm); /* R = Rd - Imm */ + tcg_gen_andi_tl(R, R, 0xffff); /* make it 16 bits */ + + /* update status register */ + tcg_gen_andc_tl(cpu_Cf, R, Rd); + tcg_gen_shri_tl(cpu_Cf, cpu_Cf, 15); /* Cf = R & ~Rd */ + tcg_gen_andc_tl(cpu_Vf, Rd, R); + tcg_gen_shri_tl(cpu_Vf, cpu_Vf, 15); /* Vf = Rd & ~R */ + tcg_gen_setcondi_tl(TCG_COND_EQ, cpu_Zf, R, 0); /* Zf = R == 0 */ + tcg_gen_shri_tl(cpu_Nf, R, 15); /* Nf = R(15) */ + tcg_gen_xor_tl(cpu_Sf, cpu_Nf, cpu_Vf); /* Sf = Nf ^ Vf */ + + /* update output registers */ + tcg_gen_andi_tl(RdL, R, 0xff); + tcg_gen_shri_tl(RdH, R, 8); + + tcg_temp_free_i32(Rd); + tcg_temp_free_i32(R); + + return true; +} + +/* + * Performs the logical AND between the contents of register Rd and register + * Rr and places the result in the destination register Rd. + */ +static bool trans_AND(DisasContext *ctx, arg_AND *a) +{ + TCGv Rd = cpu_r[a->rd]; + TCGv Rr = cpu_r[a->rr]; + TCGv R = tcg_temp_new_i32(); + + tcg_gen_and_tl(R, Rd, Rr); /* Rd = Rd and Rr */ + + /* update status register */ + tcg_gen_movi_tl(cpu_Vf, 0); /* Vf = 0 */ + tcg_gen_setcondi_tl(TCG_COND_EQ, cpu_Zf, R, 0); /* Zf = R == 0 */ + gen_ZNSf(R); + + /* update output registers */ + tcg_gen_mov_tl(Rd, R); + + tcg_temp_free_i32(R); + + return true; +} + +/* + * Performs the logical AND between the contents of register Rd and a constant + * and places the result in the destination register Rd. + */ +static bool trans_ANDI(DisasContext *ctx, arg_ANDI *a) +{ + TCGv Rd = cpu_r[a->rd]; + int Imm = (a->imm); + + tcg_gen_andi_tl(Rd, Rd, Imm); /* Rd = Rd & Imm */ + + /* update status register */ + tcg_gen_movi_tl(cpu_Vf, 0x00); /* Vf = 0 */ + gen_ZNSf(Rd); + + return true; +} + +/* + * Performs the logical OR between the contents of register Rd and register + * Rr and places the result in the destination register Rd. + */ +static bool trans_OR(DisasContext *ctx, arg_OR *a) +{ + TCGv Rd = cpu_r[a->rd]; + TCGv Rr = cpu_r[a->rr]; + TCGv R = tcg_temp_new_i32(); + + tcg_gen_or_tl(R, Rd, Rr); + + /* update status register */ + tcg_gen_movi_tl(cpu_Vf, 0); + gen_ZNSf(R); + + /* update output registers */ + tcg_gen_mov_tl(Rd, R); + + tcg_temp_free_i32(R); + + return true; +} + +/* + * Performs the logical OR between the contents of register Rd and a + * constant and places the result in the destination register Rd. + */ +static bool trans_ORI(DisasContext *ctx, arg_ORI *a) +{ + TCGv Rd = cpu_r[a->rd]; + int Imm = (a->imm); + + tcg_gen_ori_tl(Rd, Rd, Imm); /* Rd = Rd | Imm */ + + /* update status register */ + tcg_gen_movi_tl(cpu_Vf, 0x00); /* Vf = 0 */ + gen_ZNSf(Rd); + + return true; +} + +/* + * Performs the logical EOR between the contents of register Rd and + * register Rr and places the result in the destination register Rd. + */ +static bool trans_EOR(DisasContext *ctx, arg_EOR *a) +{ + TCGv Rd = cpu_r[a->rd]; + TCGv Rr = cpu_r[a->rr]; + + tcg_gen_xor_tl(Rd, Rd, Rr); + + /* update status register */ + tcg_gen_movi_tl(cpu_Vf, 0); + gen_ZNSf(Rd); + + return true; +} + +/* + * Clears the specified bits in register Rd. Performs the logical AND + * between the contents of register Rd and the complement of the constant mask + * K. The result will be placed in register Rd. + */ +static bool trans_COM(DisasContext *ctx, arg_COM *a) +{ + TCGv Rd = cpu_r[a->rd]; + TCGv R = tcg_temp_new_i32(); + + tcg_gen_xori_tl(Rd, Rd, 0xff); + + /* update status register */ + tcg_gen_movi_tl(cpu_Cf, 1); /* Cf = 1 */ + tcg_gen_movi_tl(cpu_Vf, 0); /* Vf = 0 */ + gen_ZNSf(Rd); + + tcg_temp_free_i32(R); + + return true; +} + +/* + * Replaces the contents of register Rd with its two's complement; the + * value $80 is left unchanged. + */ +static bool trans_NEG(DisasContext *ctx, arg_NEG *a) +{ + TCGv Rd = cpu_r[a->rd]; + TCGv t0 = tcg_const_i32(0); + TCGv R = tcg_temp_new_i32(); + + tcg_gen_sub_tl(R, t0, Rd); /* R = 0 - Rd */ + tcg_gen_andi_tl(R, R, 0xff); /* make it 8 bits */ + + /* update status register */ + gen_sub_CHf(R, t0, Rd); + gen_sub_Vf(R, t0, Rd); + gen_ZNSf(R); + + /* update output registers */ + tcg_gen_mov_tl(Rd, R); + + tcg_temp_free_i32(t0); + tcg_temp_free_i32(R); + + return true; +} + +/* + * Adds one -1- to the contents of register Rd and places the result in the + * destination register Rd. The C Flag in SREG is not affected by the + * operation, thus allowing the INC instruction to be used on a loop counter in + * multiple-precision computations. When operating on unsigned numbers, only + * BREQ and BRNE branches can be expected to perform consistently. When + * operating on two's complement values, all signed branches are available. + */ +static bool trans_INC(DisasContext *ctx, arg_INC *a) +{ + TCGv Rd = cpu_r[a->rd]; + + tcg_gen_addi_tl(Rd, Rd, 1); + tcg_gen_andi_tl(Rd, Rd, 0xff); + + /* update status register */ + tcg_gen_setcondi_tl(TCG_COND_EQ, cpu_Vf, Rd, 0x80); /* Vf = Rd == 0x80 */ + gen_ZNSf(Rd); + + return true; +} + +/* + * Subtracts one -1- from the contents of register Rd and places the result + * in the destination register Rd. The C Flag in SREG is not affected by the + * operation, thus allowing the DEC instruction to be used on a loop counter in + * multiple-precision computations. When operating on unsigned values, only + * BREQ and BRNE branches can be expected to perform consistently. When + * operating on two's complement values, all signed branches are available. + */ +static bool trans_DEC(DisasContext *ctx, arg_DEC *a) +{ + TCGv Rd = cpu_r[a->rd]; + + tcg_gen_subi_tl(Rd, Rd, 1); /* Rd = Rd - 1 */ + tcg_gen_andi_tl(Rd, Rd, 0xff); /* make it 8 bits */ + + /* update status register */ + tcg_gen_setcondi_tl(TCG_COND_EQ, cpu_Vf, Rd, 0x7f); /* Vf = Rd == 0x7f */ + gen_ZNSf(Rd); + + return true; +} + +/* + * This instruction performs 8-bit x 8-bit -> 16-bit unsigned multiplication. + */ +static bool trans_MUL(DisasContext *ctx, arg_MUL *a) +{ + if (!avr_have_feature(ctx, AVR_FEATURE_MUL)) { + return true; + } + + TCGv R0 = cpu_r[0]; + TCGv R1 = cpu_r[1]; + TCGv Rd = cpu_r[a->rd]; + TCGv Rr = cpu_r[a->rr]; + TCGv R = tcg_temp_new_i32(); + + tcg_gen_mul_tl(R, Rd, Rr); /* R = Rd * Rr */ + tcg_gen_andi_tl(R0, R, 0xff); + tcg_gen_shri_tl(R1, R, 8); + + /* update status register */ + tcg_gen_shri_tl(cpu_Cf, R, 15); /* Cf = R(15) */ + tcg_gen_setcondi_tl(TCG_COND_EQ, cpu_Zf, R, 0); /* Zf = R == 0 */ + + tcg_temp_free_i32(R); + + return true; +} + +/* + * This instruction performs 8-bit x 8-bit -> 16-bit signed multiplication. + */ +static bool trans_MULS(DisasContext *ctx, arg_MULS *a) +{ + if (!avr_have_feature(ctx, AVR_FEATURE_MUL)) { + return true; + } + + TCGv R0 = cpu_r[0]; + TCGv R1 = cpu_r[1]; + TCGv Rd = cpu_r[a->rd]; + TCGv Rr = cpu_r[a->rr]; + TCGv R = tcg_temp_new_i32(); + TCGv t0 = tcg_temp_new_i32(); + TCGv t1 = tcg_temp_new_i32(); + + tcg_gen_ext8s_tl(t0, Rd); /* make Rd full 32 bit signed */ + tcg_gen_ext8s_tl(t1, Rr); /* make Rr full 32 bit signed */ + tcg_gen_mul_tl(R, t0, t1); /* R = Rd * Rr */ + tcg_gen_andi_tl(R, R, 0xffff); /* make it 16 bits */ + tcg_gen_andi_tl(R0, R, 0xff); + tcg_gen_shri_tl(R1, R, 8); + + /* update status register */ + tcg_gen_shri_tl(cpu_Cf, R, 15); /* Cf = R(15) */ + tcg_gen_setcondi_tl(TCG_COND_EQ, cpu_Zf, R, 0); /* Zf = R == 0 */ + + tcg_temp_free_i32(t1); + tcg_temp_free_i32(t0); + tcg_temp_free_i32(R); + + return true; +} + +/* + * This instruction performs 8-bit x 8-bit -> 16-bit multiplication of a + * signed and an unsigned number. + */ +static bool trans_MULSU(DisasContext *ctx, arg_MULSU *a) +{ + if (!avr_have_feature(ctx, AVR_FEATURE_MUL)) { + return true; + } + + TCGv R0 = cpu_r[0]; + TCGv R1 = cpu_r[1]; + TCGv Rd = cpu_r[a->rd]; + TCGv Rr = cpu_r[a->rr]; + TCGv R = tcg_temp_new_i32(); + TCGv t0 = tcg_temp_new_i32(); + + tcg_gen_ext8s_tl(t0, Rd); /* make Rd full 32 bit signed */ + tcg_gen_mul_tl(R, t0, Rr); /* R = Rd * Rr */ + tcg_gen_andi_tl(R, R, 0xffff); /* make R 16 bits */ + tcg_gen_andi_tl(R0, R, 0xff); + tcg_gen_shri_tl(R1, R, 8); + + /* update status register */ + tcg_gen_shri_tl(cpu_Cf, R, 15); /* Cf = R(15) */ + tcg_gen_setcondi_tl(TCG_COND_EQ, cpu_Zf, R, 0); /* Zf = R == 0 */ + + tcg_temp_free_i32(t0); + tcg_temp_free_i32(R); + + return true; +} + +/* + * This instruction performs 8-bit x 8-bit -> 16-bit unsigned + * multiplication and shifts the result one bit left. + */ +static bool trans_FMUL(DisasContext *ctx, arg_FMUL *a) +{ + if (!avr_have_feature(ctx, AVR_FEATURE_MUL)) { + return true; + } + + TCGv R0 = cpu_r[0]; + TCGv R1 = cpu_r[1]; + TCGv Rd = cpu_r[a->rd]; + TCGv Rr = cpu_r[a->rr]; + TCGv R = tcg_temp_new_i32(); + + tcg_gen_mul_tl(R, Rd, Rr); /* R = Rd * Rr */ + + /* update status register */ + tcg_gen_shri_tl(cpu_Cf, R, 15); /* Cf = R(15) */ + tcg_gen_setcondi_tl(TCG_COND_EQ, cpu_Zf, R, 0); /* Zf = R == 0 */ + + /* update output registers */ + tcg_gen_shli_tl(R, R, 1); + tcg_gen_andi_tl(R0, R, 0xff); + tcg_gen_shri_tl(R1, R, 8); + tcg_gen_andi_tl(R1, R1, 0xff); + + + tcg_temp_free_i32(R); + + return true; +} + +/* + * This instruction performs 8-bit x 8-bit -> 16-bit signed multiplication + * and shifts the result one bit left. + */ +static bool trans_FMULS(DisasContext *ctx, arg_FMULS *a) +{ + if (!avr_have_feature(ctx, AVR_FEATURE_MUL)) { + return true; + } + + TCGv R0 = cpu_r[0]; + TCGv R1 = cpu_r[1]; + TCGv Rd = cpu_r[a->rd]; + TCGv Rr = cpu_r[a->rr]; + TCGv R = tcg_temp_new_i32(); + TCGv t0 = tcg_temp_new_i32(); + TCGv t1 = tcg_temp_new_i32(); + + tcg_gen_ext8s_tl(t0, Rd); /* make Rd full 32 bit signed */ + tcg_gen_ext8s_tl(t1, Rr); /* make Rr full 32 bit signed */ + tcg_gen_mul_tl(R, t0, t1); /* R = Rd * Rr */ + tcg_gen_andi_tl(R, R, 0xffff); /* make it 16 bits */ + + /* update status register */ + tcg_gen_shri_tl(cpu_Cf, R, 15); /* Cf = R(15) */ + tcg_gen_setcondi_tl(TCG_COND_EQ, cpu_Zf, R, 0); /* Zf = R == 0 */ + + /* update output registers */ + tcg_gen_shli_tl(R, R, 1); + tcg_gen_andi_tl(R0, R, 0xff); + tcg_gen_shri_tl(R1, R, 8); + tcg_gen_andi_tl(R1, R1, 0xff); + + tcg_temp_free_i32(t1); + tcg_temp_free_i32(t0); + tcg_temp_free_i32(R); + + return true; +} + +/* + * This instruction performs 8-bit x 8-bit -> 16-bit signed multiplication + * and shifts the result one bit left. + */ +static bool trans_FMULSU(DisasContext *ctx, arg_FMULSU *a) +{ + if (!avr_have_feature(ctx, AVR_FEATURE_MUL)) { + return true; + } + + TCGv R0 = cpu_r[0]; + TCGv R1 = cpu_r[1]; + TCGv Rd = cpu_r[a->rd]; + TCGv Rr = cpu_r[a->rr]; + TCGv R = tcg_temp_new_i32(); + TCGv t0 = tcg_temp_new_i32(); + + tcg_gen_ext8s_tl(t0, Rd); /* make Rd full 32 bit signed */ + tcg_gen_mul_tl(R, t0, Rr); /* R = Rd * Rr */ + tcg_gen_andi_tl(R, R, 0xffff); /* make it 16 bits */ + + /* update status register */ + tcg_gen_shri_tl(cpu_Cf, R, 15); /* Cf = R(15) */ + tcg_gen_setcondi_tl(TCG_COND_EQ, cpu_Zf, R, 0); /* Zf = R == 0 */ + + /* update output registers */ + tcg_gen_shli_tl(R, R, 1); + tcg_gen_andi_tl(R0, R, 0xff); + tcg_gen_shri_tl(R1, R, 8); + tcg_gen_andi_tl(R1, R1, 0xff); + + tcg_temp_free_i32(t0); + tcg_temp_free_i32(R); + + return true; +} + +/* + * The module is an instruction set extension to the AVR CPU, performing + * DES iterations. The 64-bit data block (plaintext or ciphertext) is placed in + * the CPU register file, registers R0-R7, where LSB of data is placed in LSB + * of R0 and MSB of data is placed in MSB of R7. The full 64-bit key (including + * parity bits) is placed in registers R8- R15, organized in the register file + * with LSB of key in LSB of R8 and MSB of key in MSB of R15. Executing one DES + * instruction performs one round in the DES algorithm. Sixteen rounds must be + * executed in increasing order to form the correct DES ciphertext or + * plaintext. Intermediate results are stored in the register file (R0-R15) + * after each DES instruction. The instruction's operand (K) determines which + * round is executed, and the half carry flag (H) determines whether encryption + * or decryption is performed. The DES algorithm is described in + * "Specifications for the Data Encryption Standard" (Federal Information + * Processing Standards Publication 46). Intermediate results in this + * implementation differ from the standard because the initial permutation and + * the inverse initial permutation are performed each iteration. This does not + * affect the result in the final ciphertext or plaintext, but reduces + * execution time. + */ +static bool trans_DES(DisasContext *ctx, arg_DES *a) +{ + /* TODO */ + if (!avr_have_feature(ctx, AVR_FEATURE_DES)) { + return true; + } + + return true; +}