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 lists.gnu.org (lists1p.gnu.org [209.51.188.17]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.lore.kernel.org (Postfix) with ESMTPS id 60028F364DB for ; Fri, 10 Apr 2026 11:10:28 +0000 (UTC) Received: from localhost ([::1] helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1wB9k3-00079W-8Y; Fri, 10 Apr 2026 07:09:39 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]) by lists1p.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1wB9k2-000797-0d for qemu-devel@nongnu.org; Fri, 10 Apr 2026 07:09:38 -0400 Received: from mail-wm1-x335.google.com ([2a00:1450:4864:20::335]) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_128_GCM_SHA256:128) (Exim 4.90_1) (envelope-from ) id 1wB9k0-0003Tt-18 for qemu-devel@nongnu.org; Fri, 10 Apr 2026 07:09:37 -0400 Received: by mail-wm1-x335.google.com with SMTP id 5b1f17b1804b1-488a4bc360bso10696925e9.0 for ; Fri, 10 Apr 2026 04:09:35 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20251104; t=1775819374; x=1776424174; darn=nongnu.org; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:from:to:cc:subject:date :message-id:reply-to; bh=wL2RjIiakxvUiKqLallUtvD80SyqMUagJMAW4JWo6rY=; b=oG8uh/8O41McF7zJS5zlKIGuxIvm8Hr5gM41BQWyLqgE2cfPXDHTCBv4iFv/mmWSZ1 Fr1o/adNHC6U+++fEaK0reZmsefDLR+edZ59ZEcCVbxQOBuB4C7iRfoEjKf9guGXGj3e m1pC4b1dexAMlOPGfls1O/y9UL3qRY8Hp1lE5KKbNV/EoN44ZsSv3/MFWZsaN50y9cKL l15dZ/Mb8drtWWsrikwLeclzlCnvVyAsmzdHH1lwzsQq+YUksMlhfVa5KXj1lYdLeLWD c9t5rXragZM7M8stZECCUjyRkD4fRLHUOYcVyC2EW+HQDeKT1ttzNcOPHHXw27H+XoL9 J5JQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20251104; t=1775819374; x=1776424174; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-gg:x-gm-message-state:from :to:cc:subject:date:message-id:reply-to; bh=wL2RjIiakxvUiKqLallUtvD80SyqMUagJMAW4JWo6rY=; b=BdHDzQkUFyTk8ZXbkig2Y9tSjPECWHEQhFGGxD6BMhMLT9ziAEZTX0pkZSB6ndj9OV hOGj0395IqnZsoAIa6tVSiifrCEdYI9pOehwhGdDq9oEruzOMsUtTrwQDWuwx7x+13Gu PBCHB02xbRg/hULN/kp9HUoAaOsQKgY6QYCy0DMT5SAzknwXid2qmtQoritmoZf8VaYH 3PV1o3jsy/9pRdZIx6qjXZKin+w+HGuHtwk+xMCmPuv6jMCzba9cQDHkuOCmxPkiI6UI ePZgejcxVi+pRaCgIiklG/lsoXHBSNoSi23SyusyQ0XxeJMHzAecLc0eTYNTLTPdFv3x 7aAA== X-Gm-Message-State: AOJu0YxhjVawLmtFic7hfX23rFunaQWqHVHnJZrqfJxU9R1ABqxhL4cs JfYHQm/t32pmj47xju19zSMlrfBWWduPPWok0IUdr3YWyxOU6xseNt9zfV66Gl/h X-Gm-Gg: AeBDievC7RZmsyvPXF08qAVe/XToBJZb16HO4xQ02K7w4aZirfadBl9JhXySwWIGfWG 7A7uEl4THBlaCPP66qieL7dRSL/3h64oT+a+I1Ea6BFteB3Ye2xF+yduRaBTQzn7VvzJ1EFAzPc mDp3T6EJIck6z0qMdFGSf9aQSpRww6wPsgIQMQXgo+joLgPsatDZcIJle5/cxK5s6G6Yh0OcNLr NpWMAglKRBs7xzKvds1fMrbbmGSEhOM4pxmn2rnDdLX3V37eVfCG+busCWU78HNgtM10NFUW/FV W8xFzwFTqDN0nIPMaD4oyCTD2gGlN0iUCwaHiuPVi6ckE1pbqVKZJLtItXVloxCLoR/hYf33x++ avbyV1kzBQ8H+umwasshBjPm0pMv5P67SeOvwiaRR5Xprjs4MVOt0FSoqxBPo4ALAkn1P01xmX5 v7GijperKH+yRn1FsBWv4HEBeOBh5w2KfPl7kQZulgH/VQsfXWPh57y26iyEk98s3AnarQ X-Received: by 2002:a05:600c:a105:b0:485:3af5:7e53 with SMTP id 5b1f17b1804b1-488d6839821mr23465265e9.19.1775819373838; Fri, 10 Apr 2026 04:09:33 -0700 (PDT) Received: from ninolomata-AERO-15-KC.. (89-181-36-85.net.novis.pt. [89.181.36.85]) by smtp.gmail.com with ESMTPSA id 5b1f17b1804b1-488d5dc7070sm19311605e9.10.2026.04.10.04.09.32 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Fri, 10 Apr 2026 04:09:33 -0700 (PDT) From: Bruno Sa To: qemu-devel@nongnu.org Cc: qemu-riscv@nongnu.org, palmer@dabbelt.com, alistair.francis@wdc.com, liwei1518@gmail.com, dbarboza@ventanamicro.com, zhiwei_liu@linux.alibaba.com, Bruno Sa Subject: [PATCH v2] target/riscv: fix RV32 stateen CSR handling Date: Fri, 10 Apr 2026 12:08:42 +0100 Message-ID: <20260410110928.1014170-1-bruno.vilaca.sa@gmail.com> X-Mailer: git-send-email 2.43.0 In-Reply-To: <20260409155344.2849233-1-bruno.vilaca.sa@gmail.com> References: <20260409155344.2849233-1-bruno.vilaca.sa@gmail.com> MIME-Version: 1.0 Content-Transfer-Encoding: 8bit Received-SPF: pass client-ip=2a00:1450:4864:20::335; envelope-from=bruno.vilaca.sa@gmail.com; helo=mail-wm1-x335.google.com 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, FREEMAIL_FROM=0.001, RCVD_IN_DNSWL_NONE=-0.0001, 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: 2.1.29 Precedence: list List-Id: qemu development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: qemu-devel-bounces+qemu-devel=archiver.kernel.org@nongnu.org Sender: qemu-devel-bounces+qemu-devel=archiver.kernel.org@nongnu.org The RV32 stateen CSRs are split between the low-half CSR and the corresponding xH CSR, but the current implementation still handles some upper-half bits through the low-half write paths and also accepts the xH CSRs on RV64. Fix this by: - rejecting mstateen*h and hstateen*h accesses on RV64 - keeping the RV64-only writable bits in the low-half write paths - handling the RV32 upper-half writable bits in write_mstateen0h() and write_hstateen0h() - dropping unsupported writable bits from write_sstateen0() Signed-off-by: Bruno Sa --- v2: - rebase on riscv-to-apply.next - resend only patch 2 after patch 1 was applied - wrap the AIA comment text to keep checkpatch clean target/riscv/csr.c | 117 ++++++++++++++++++++++++++++++++------------- 1 file changed, 83 insertions(+), 34 deletions(-) diff --git a/target/riscv/csr.c b/target/riscv/csr.c index cfd076b368..80727aa81e 100644 --- a/target/riscv/csr.c +++ b/target/riscv/csr.c @@ -502,6 +502,15 @@ static RISCVException mstateen(CPURISCVState *env, int csrno) return any(env, csrno); } +static RISCVException mstateen_32(CPURISCVState *env, int csrno) +{ + if (riscv_cpu_mxl(env) != MXL_RV32) { + return RISCV_EXCP_ILLEGAL_INST; + } + + return mstateen(env, csrno); +} + static RISCVException hstateen_pred(CPURISCVState *env, int csrno, int base) { if (!riscv_cpu_cfg(env)->ext_smstateen) { @@ -533,6 +542,10 @@ static RISCVException hstateen(CPURISCVState *env, int csrno) static RISCVException hstateenh(CPURISCVState *env, int csrno) { + if (riscv_cpu_mxl(env) != MXL_RV32) { + return RISCV_EXCP_ILLEGAL_INST; + } + return hstateen_pred(env, csrno, CSR_HSTATEEN0H); } @@ -3447,25 +3460,29 @@ static RISCVException write_mstateen0(CPURISCVState *env, int csrno, wr_mask |= SMSTATEEN0_FCSR; } - if (env->priv_ver >= PRIV_VERSION_1_13_0) { - wr_mask |= SMSTATEEN0_P1P13; - } + if (riscv_cpu_mxl(env) == MXL_RV64) { + if (env->priv_ver >= PRIV_VERSION_1_13_0) { + wr_mask |= SMSTATEEN0_P1P13; + } - if (riscv_cpu_cfg(env)->ext_smaia || riscv_cpu_cfg(env)->ext_smcsrind) { - wr_mask |= SMSTATEEN0_SVSLCT; - } + if (riscv_cpu_cfg(env)->ext_smaia || + riscv_cpu_cfg(env)->ext_smcsrind) { + wr_mask |= SMSTATEEN0_SVSLCT; + } - /* - * As per the AIA specification, SMSTATEEN0_IMSIC is valid only if IMSIC is - * implemented. However, that information is with MachineState and we can't - * figure that out in csr.c. Just enable if Smaia is available. - */ - if (riscv_cpu_cfg(env)->ext_smaia) { - wr_mask |= (SMSTATEEN0_AIA | SMSTATEEN0_IMSIC); - } + /* + * As per the AIA specification, SMSTATEEN0_IMSIC is valid + * only if IMSIC is implemented. However, that information is + * with MachineState and we can't figure that out in csr.c. + * Just enable if Smaia is available. + */ + if (riscv_cpu_cfg(env)->ext_smaia) { + wr_mask |= (SMSTATEEN0_AIA | SMSTATEEN0_IMSIC); + } - if (riscv_cpu_cfg(env)->ext_ssctr) { - wr_mask |= SMSTATEEN0_CTR; + if (riscv_cpu_cfg(env)->ext_ssctr) { + wr_mask |= SMSTATEEN0_CTR; + } } return write_mstateen(env, csrno, wr_mask, new_val); @@ -3507,6 +3524,20 @@ static RISCVException write_mstateen0h(CPURISCVState *env, int csrno, wr_mask |= SMSTATEEN0_P1P13; } + if (riscv_cpu_cfg(env)->ext_smaia || riscv_cpu_cfg(env)->ext_smcsrind) { + wr_mask |= SMSTATEEN0_SVSLCT; + } + + /* + * As per the AIA specification, SMSTATEEN0_IMSIC is valid only if + * IMSIC is implemented. However, that information is with + * MachineState and we can't figure that out in csr.c. Just enable + * if Smaia is available. + */ + if (riscv_cpu_cfg(env)->ext_smaia) { + wr_mask |= (SMSTATEEN0_AIA | SMSTATEEN0_IMSIC); + } + if (riscv_cpu_cfg(env)->ext_ssctr) { wr_mask |= SMSTATEEN0_CTR; } @@ -3552,21 +3583,25 @@ static RISCVException write_hstateen0(CPURISCVState *env, int csrno, wr_mask |= SMSTATEEN0_FCSR; } - if (riscv_cpu_cfg(env)->ext_ssaia || riscv_cpu_cfg(env)->ext_sscsrind) { - wr_mask |= SMSTATEEN0_SVSLCT; - } + if (riscv_cpu_mxl(env) == MXL_RV64) { + if (riscv_cpu_cfg(env)->ext_ssaia || + riscv_cpu_cfg(env)->ext_sscsrind) { + wr_mask |= SMSTATEEN0_SVSLCT; + } - /* - * As per the AIA specification, SMSTATEEN0_IMSIC is valid only if IMSIC is - * implemented. However, that information is with MachineState and we can't - * figure that out in csr.c. Just enable if Ssaia is available. - */ - if (riscv_cpu_cfg(env)->ext_ssaia) { - wr_mask |= (SMSTATEEN0_AIA | SMSTATEEN0_IMSIC); - } + /* + * As per the AIA specification, SMSTATEEN0_IMSIC is valid + * only if IMSIC is implemented. However, that information is + * with MachineState and we can't figure that out in csr.c. + * Just enable if Ssaia is available. + */ + if (riscv_cpu_cfg(env)->ext_ssaia) { + wr_mask |= (SMSTATEEN0_AIA | SMSTATEEN0_IMSIC); + } - if (riscv_cpu_cfg(env)->ext_ssctr) { - wr_mask |= SMSTATEEN0_CTR; + if (riscv_cpu_cfg(env)->ext_ssctr) { + wr_mask |= SMSTATEEN0_CTR; + } } return write_hstateen(env, csrno, wr_mask, new_val); @@ -3608,6 +3643,20 @@ static RISCVException write_hstateen0h(CPURISCVState *env, int csrno, { uint64_t wr_mask = SMSTATEEN_STATEEN | SMSTATEEN0_HSENVCFG; + if (riscv_cpu_cfg(env)->ext_ssaia || riscv_cpu_cfg(env)->ext_sscsrind) { + wr_mask |= SMSTATEEN0_SVSLCT; + } + + /* + * As per the AIA specification, SMSTATEEN0_IMSIC is valid only if + * IMSIC is implemented. However, that information is with + * MachineState and we can't figure that out in csr.c. Just enable + * if Ssaia is available. + */ + if (riscv_cpu_cfg(env)->ext_ssaia) { + wr_mask |= (SMSTATEEN0_AIA | SMSTATEEN0_IMSIC); + } + if (riscv_cpu_cfg(env)->ext_ssctr) { wr_mask |= SMSTATEEN0_CTR; } @@ -3657,7 +3706,7 @@ static RISCVException write_sstateen(CPURISCVState *env, int csrno, static RISCVException write_sstateen0(CPURISCVState *env, int csrno, target_ulong new_val, uintptr_t ra) { - uint64_t wr_mask = SMSTATEEN_STATEEN | SMSTATEEN0_HSENVCFG; + uint64_t wr_mask = 0; if (!riscv_has_ext(env, RVF)) { wr_mask |= SMSTATEEN0_FCSR; @@ -5937,25 +5986,25 @@ riscv_csr_operations csr_ops[CSR_TABLE_SIZE] = { /* Smstateen extension CSRs */ [CSR_MSTATEEN0] = { "mstateen0", mstateen, read_mstateen, write_mstateen0, .min_priv_ver = PRIV_VERSION_1_12_0 }, - [CSR_MSTATEEN0H] = { "mstateen0h", mstateen, read_mstateenh, + [CSR_MSTATEEN0H] = { "mstateen0h", mstateen_32, read_mstateenh, write_mstateen0h, .min_priv_ver = PRIV_VERSION_1_12_0 }, [CSR_MSTATEEN1] = { "mstateen1", mstateen, read_mstateen, write_mstateen_1_3, .min_priv_ver = PRIV_VERSION_1_12_0 }, - [CSR_MSTATEEN1H] = { "mstateen1h", mstateen, read_mstateenh, + [CSR_MSTATEEN1H] = { "mstateen1h", mstateen_32, read_mstateenh, write_mstateenh_1_3, .min_priv_ver = PRIV_VERSION_1_12_0 }, [CSR_MSTATEEN2] = { "mstateen2", mstateen, read_mstateen, write_mstateen_1_3, .min_priv_ver = PRIV_VERSION_1_12_0 }, - [CSR_MSTATEEN2H] = { "mstateen2h", mstateen, read_mstateenh, + [CSR_MSTATEEN2H] = { "mstateen2h", mstateen_32, read_mstateenh, write_mstateenh_1_3, .min_priv_ver = PRIV_VERSION_1_12_0 }, [CSR_MSTATEEN3] = { "mstateen3", mstateen, read_mstateen, write_mstateen_1_3, .min_priv_ver = PRIV_VERSION_1_12_0 }, - [CSR_MSTATEEN3H] = { "mstateen3h", mstateen, read_mstateenh, + [CSR_MSTATEEN3H] = { "mstateen3h", mstateen_32, read_mstateenh, write_mstateenh_1_3, .min_priv_ver = PRIV_VERSION_1_12_0 }, [CSR_HSTATEEN0] = { "hstateen0", hstateen, read_hstateen, write_hstateen0, -- 2.43.0 Rebased on riscv-to-apply.next as requested. Patch 1 was already applied, so this v2 resends only patch 2. Thanks, Bruno