From patchwork Thu Dec 21 17:51:31 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Daniel Henrique Barboza X-Patchwork-Id: 1879375 Return-Path: X-Original-To: incoming@patchwork.ozlabs.org Delivered-To: patchwork-incoming@legolas.ozlabs.org Authentication-Results: legolas.ozlabs.org; dkim=pass (2048-bit key; unprotected) header.d=ventanamicro.com header.i=@ventanamicro.com header.a=rsa-sha256 header.s=google header.b=Bp0BMkCd; dkim-atps=neutral Authentication-Results: legolas.ozlabs.org; spf=pass (sender SPF authorized) smtp.mailfrom=nongnu.org (client-ip=209.51.188.17; helo=lists.gnu.org; envelope-from=qemu-devel-bounces+incoming=patchwork.ozlabs.org@nongnu.org; receiver=patchwork.ozlabs.org) Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) (using TLSv1.2 with cipher ECDHE-ECDSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by legolas.ozlabs.org (Postfix) with ESMTPS id 4SwyjK1yKRz1ySd for ; Fri, 22 Dec 2023 04:53:57 +1100 (AEDT) Received: from localhost ([::1] helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1rGNDl-00037l-Dp; Thu, 21 Dec 2023 12:52:33 -0500 Received: from eggs.gnu.org ([2001:470:142:3::10]) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1rGNDZ-0002cV-7C for qemu-devel@nongnu.org; Thu, 21 Dec 2023 12:52:21 -0500 Received: from mail-io1-xd33.google.com ([2607:f8b0:4864:20::d33]) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_128_GCM_SHA256:128) (Exim 4.90_1) (envelope-from ) id 1rGNDU-000807-GW for qemu-devel@nongnu.org; Thu, 21 Dec 2023 12:52:20 -0500 Received: by mail-io1-xd33.google.com with SMTP id ca18e2360f4ac-7b3b819f8a3so52811739f.1 for ; Thu, 21 Dec 2023 09:52:16 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=ventanamicro.com; s=google; t=1703181134; x=1703785934; 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=hSBipYs1ja9hQ+IdxMnaSg+JbEymd7tED/lgmeFP+2A=; b=Bp0BMkCdYA51wJoOD8R+U613fi9EwRoFovnn6x/ArylilaJY+9xTg0mAX4AdnVZU0N 28biYe6i/dTpkZO3tE2ILJyRArA5mtrRZlr10TwTfukhSsx9hiu++fUgBIpqQ0fB6C5H 9ssQrh6ma9L0QrRw17OUpZLpquJJLM1emj5O/zZuR13Q+AXSaYjyrSArnnXICR1Br39P +HZfmq/uhJLdfwiQv5NZ9gUU/K3woxWu1uNC2EhNGKB0r8U4rQQS275i/0VLum2J6uVj QUtX6bXbdiNHys0lRvIxNIrQn8EG/kJ9+sh/wLK4aztDPTLuWp9KdmnYhSiTpCz8n8j9 eUiw== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20230601; t=1703181134; x=1703785934; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-message-state:from:to:cc :subject:date:message-id:reply-to; bh=hSBipYs1ja9hQ+IdxMnaSg+JbEymd7tED/lgmeFP+2A=; b=bNaNVjwzQa99VEUPHmdrteuyexheVqNve/UvsFu9wvezxr39j0z83vC1nHKY5ivDxM DYSkRw0jew5y9388wu7hutlnLQZ+J3HmmCbPIl0PifJt9Fbo1jP88+2OSc0BqJ6/PHLX 5usOx524tIA6Qjm2EVxJ+ybJvtaK8zqFXM1J7pa8QEjhhCfid6X0ammJAF3mQrwn+194 4bwZ35Iu7lPNNSIxFbpqIxtIamoUocCr1FM3ZdFDZh0wYv32CHpF3zrionZfPbJQ+ABf o4gKDpNiHQm3uUC8eBThzVBqvZfzwLntCCQf3bAzyCO4u3967c+EauFmTQ67T8T8fmww lO4Q== X-Gm-Message-State: AOJu0YzyGaNS5Ovqf2TxrWaDm8z3m0rW85CGSytad9tsFiLV5jJfosBv ZXNN6B7X/+sbZeRe0aSUqSN4OxgILgB7wLDe0ju9Wg== X-Google-Smtp-Source: AGHT+IF2eaThLi06BT+Oyuavd5o69LImy0603i33yeqSCaUxLqAKkfKqV1cT/VGwCGpAULgo/A1lNg== X-Received: by 2002:a05:6e02:1bc6:b0:35d:59a2:bb1 with SMTP id x6-20020a056e021bc600b0035d59a20bb1mr21687163ilv.71.1703181134446; Thu, 21 Dec 2023 09:52:14 -0800 (PST) Received: from grind.dc1.ventanamicro.com ([179.93.21.205]) by smtp.gmail.com with ESMTPSA id w10-20020a63f50a000000b005b18c53d73csm1806261pgh.16.2023.12.21.09.52.11 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Thu, 21 Dec 2023 09:52:14 -0800 (PST) From: Daniel Henrique Barboza To: qemu-devel@nongnu.org Cc: qemu-riscv@nongnu.org, alistair.francis@wdc.com, bmeng@tinylab.org, liwei1518@gmail.com, zhiwei_liu@linux.alibaba.com, palmer@rivosinc.com, ajones@ventanamicro.com, Daniel Henrique Barboza Subject: [PATCH 10/16] target/riscv: create finalize_features() for KVM Date: Thu, 21 Dec 2023 14:51:31 -0300 Message-ID: <20231221175137.497379-11-dbarboza@ventanamicro.com> X-Mailer: git-send-email 2.43.0 In-Reply-To: <20231221175137.497379-1-dbarboza@ventanamicro.com> References: <20231221175137.497379-1-dbarboza@ventanamicro.com> MIME-Version: 1.0 Received-SPF: pass client-ip=2607:f8b0:4864:20::d33; envelope-from=dbarboza@ventanamicro.com; helo=mail-io1-xd33.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, RCVD_IN_DNSWL_NONE=-0.0001, SPF_HELO_NONE=0.001, SPF_PASS=-0.001, T_SCC_BODY_TEXT_LINE=-0.01 autolearn=unavailable autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: qemu-devel-bounces+incoming=patchwork.ozlabs.org@nongnu.org Sender: qemu-devel-bounces+incoming=patchwork.ozlabs.org@nongnu.org To turn cbom_blocksize and cboz_blocksize into class properties we need KVM specific changes. KVM is creating its own version of these options with a customized setter() that prevents users from picking an invalid value during init() time. This comes at the cost of duplicating each option that KVM supports. This will keep happening for each new shared option KVM implements in the future. We can avoid that by using the same property TCG uses and adding specific KVM handling during finalize() time, like TCG already does with riscv_tcg_cpu_finalize_features(). To do that, the common CPU property offers a way of knowing if an option was user set or not, sparing us from doing unneeded syscalls. riscv_kvm_cpu_finalize_features() is then created using the same KVMScratch CPU we already use during init() time, since finalize() time is still too early to use the official KVM CPU for it. cbom_blocksize and cboz_blocksize are then handled during finalize() in the same way they're handled by their KVM specific setter. With this change we can proceed with the blocksize changes in the common code without breaking the KVM driver. Signed-off-by: Daniel Henrique Barboza --- target/riscv/cpu.c | 16 +++++++--- target/riscv/cpu.h | 1 + target/riscv/kvm/kvm-cpu.c | 59 ++++++++++++++++++++++++++++++++++++ target/riscv/kvm/kvm_riscv.h | 1 + 4 files changed, 72 insertions(+), 5 deletions(-) diff --git a/target/riscv/cpu.c b/target/riscv/cpu.c index aec1968cad..d7e74b3428 100644 --- a/target/riscv/cpu.c +++ b/target/riscv/cpu.c @@ -63,6 +63,11 @@ static void cpu_option_add_user_setting(const char *optname, uint32_t value) GUINT_TO_POINTER(value)); } +bool riscv_cpu_option_set(const char *optname) +{ + return g_hash_table_contains(general_user_opts, optname); +} + #define ISA_EXT_DATA_ENTRY(_name, _min_ver, _prop) \ {#_name, _min_ver, CPU_CFG_OFFSET(_prop)} @@ -1056,17 +1061,18 @@ void riscv_cpu_finalize_features(RISCVCPU *cpu, Error **errp) { Error *local_err = NULL; - /* - * KVM accel does not have a specialized finalize() - * callback because its extensions are validated - * in the get()/set() callbacks of each property. - */ if (tcg_enabled()) { riscv_tcg_cpu_finalize_features(cpu, &local_err); if (local_err != NULL) { error_propagate(errp, local_err); return; } + } else if (kvm_enabled()) { + riscv_kvm_cpu_finalize_features(cpu, &local_err); + if (local_err != NULL) { + error_propagate(errp, local_err); + return; + } } #ifndef CONFIG_USER_ONLY diff --git a/target/riscv/cpu.h b/target/riscv/cpu.h index 53101b82c5..988471c7ba 100644 --- a/target/riscv/cpu.h +++ b/target/riscv/cpu.h @@ -495,6 +495,7 @@ bool riscv_cpu_tlb_fill(CPUState *cs, vaddr address, int size, bool probe, uintptr_t retaddr); char *riscv_isa_string(RISCVCPU *cpu); void riscv_cpu_list(void); +bool riscv_cpu_option_set(const char *optname); #define cpu_list riscv_cpu_list #define cpu_mmu_index riscv_cpu_mmu_index diff --git a/target/riscv/kvm/kvm-cpu.c b/target/riscv/kvm/kvm-cpu.c index 62a1e51f0a..70fb075846 100644 --- a/target/riscv/kvm/kvm-cpu.c +++ b/target/riscv/kvm/kvm-cpu.c @@ -1490,6 +1490,65 @@ static void kvm_cpu_instance_init(CPUState *cs) } } +void riscv_kvm_cpu_finalize_features(RISCVCPU *cpu, Error **errp) +{ + CPURISCVState *env = &cpu->env; + KVMScratchCPU kvmcpu; + struct kvm_one_reg reg; + uint64_t val; + int ret; + + /* short-circuit without spinning the scratch CPU */ + if (!cpu->cfg.ext_zicbom && !cpu->cfg.ext_zicboz) { + return; + } + + if (!kvm_riscv_create_scratch_vcpu(&kvmcpu)) { + error_setg(errp, "Unable to create scratch KVM cpu"); + return; + } + + if (cpu->cfg.ext_zicbom && + riscv_cpu_option_set(kvm_cbom_blocksize.name)) { + + reg.id = kvm_riscv_reg_id_ulong(env, KVM_REG_RISCV_CONFIG, + kvm_cbom_blocksize.kvm_reg_id); + reg.addr = (uint64_t)&val; + ret = ioctl(kvmcpu.cpufd, KVM_GET_ONE_REG, ®); + if (ret != 0) { + error_setg(errp, "Unable to read cbom_blocksize, error %d", errno); + return; + } + + if (cpu->cfg.cbom_blocksize != val) { + error_setg(errp, "Unable to set cbom_blocksize to a different " + "value than the host (%lu)", val); + return; + } + } + + if (cpu->cfg.ext_zicboz && + riscv_cpu_option_set(kvm_cboz_blocksize.name)) { + + reg.id = kvm_riscv_reg_id_ulong(env, KVM_REG_RISCV_CONFIG, + kvm_cboz_blocksize.kvm_reg_id); + reg.addr = (uint64_t)&val; + ret = ioctl(kvmcpu.cpufd, KVM_GET_ONE_REG, ®); + if (ret != 0) { + error_setg(errp, "Unable to read cbom_blocksize, error %d", errno); + return; + } + + if (cpu->cfg.cboz_blocksize != val) { + error_setg(errp, "Unable to set cboz_blocksize to a different " + "value than the host (%lu)", val); + return; + } + } + + kvm_riscv_destroy_scratch_vcpu(&kvmcpu); +} + static void kvm_cpu_accel_class_init(ObjectClass *oc, void *data) { AccelCPUClass *acc = ACCEL_CPU_CLASS(oc); diff --git a/target/riscv/kvm/kvm_riscv.h b/target/riscv/kvm/kvm_riscv.h index 8329cfab82..4bd98fddc7 100644 --- a/target/riscv/kvm/kvm_riscv.h +++ b/target/riscv/kvm/kvm_riscv.h @@ -27,5 +27,6 @@ void kvm_riscv_aia_create(MachineState *machine, uint64_t group_shift, uint64_t guest_num); void riscv_kvm_aplic_request(void *opaque, int irq, int level); int kvm_riscv_sync_mpstate_to_kvm(RISCVCPU *cpu, int state); +void riscv_kvm_cpu_finalize_features(RISCVCPU *cpu, Error **errp); #endif