From patchwork Fri May 24 07:38:11 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Bibo Mao X-Patchwork-Id: 13672779 Received: from mail.loongson.cn (mail.loongson.cn [114.242.206.163]) by smtp.subspace.kernel.org (Postfix) with ESMTP id 0B5C78625C; Fri, 24 May 2024 07:38:22 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=114.242.206.163 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1716536306; cv=none; b=pH/n2Z4u0Vtp+1W+05BoTg3TWlYqFWn5NLYrtJt2cMrmwkaDKy1cwCuvpmtvmnfSaE0JL9HYbhLlIa/e4ewQ5OVQAcdmwvTBLqDbyftmJ72Bh8i8dIHLcTDfNQ0JKXV83xODBdUfltjFm3hdaGZdZXaDR/52Iao8p44FpwGLr2E= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1716536306; c=relaxed/simple; bh=LoepqsLKDA1y97qbi8piNPlunT2bkfj1In/oCaUmwOo=; h=From:To:Cc:Subject:Date:Message-Id:In-Reply-To:References: MIME-Version; b=OyM+6HTglAwFeOIg3J0uWTPn4PtsT6ISOJ+FgrofNlng1ahMlQpjsrrA74vkwWIX/7zRPpcs0/Q0snEZGZw1t7EXxcnh0tdEk+0lni8IpTQQetchPN3J/1Ms5py6oOf7BeKKWsE4PQQ/f7SxzauQowtfhND/zT0qctpMkaZCRBM= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=none (p=none dis=none) header.from=loongson.cn; spf=pass smtp.mailfrom=loongson.cn; arc=none smtp.client-ip=114.242.206.163 Authentication-Results: smtp.subspace.kernel.org; dmarc=none (p=none dis=none) header.from=loongson.cn Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=loongson.cn Received: from loongson.cn (unknown [10.2.5.213]) by gateway (Coremail) with SMTP id _____8CxaermQ1Bm3EsDAA--.4003S3; Fri, 24 May 2024 15:38:14 +0800 (CST) Received: from localhost.localdomain (unknown [10.2.5.213]) by localhost.localdomain (Coremail) with SMTP id AQAAf8CxZcXlQ1BmScUHAA--.10650S3; Fri, 24 May 2024 15:38:13 +0800 (CST) From: Bibo Mao To: Tianrui Zhao , Huacai Chen Cc: Juergen Gross , kvm@vger.kernel.org, loongarch@lists.linux.dev, linux-kernel@vger.kernel.org, x86@kernel.org, virtualization@lists.linux.dev Subject: [PATCH v4 1/2] LoongArch: KVM: Add steal time support in kvm side Date: Fri, 24 May 2024 15:38:11 +0800 Message-Id: <20240524073812.731032-2-maobibo@loongson.cn> X-Mailer: git-send-email 2.39.3 In-Reply-To: <20240524073812.731032-1-maobibo@loongson.cn> References: <20240524073812.731032-1-maobibo@loongson.cn> Precedence: bulk X-Mailing-List: kvm@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 X-CM-TRANSID: AQAAf8CxZcXlQ1BmScUHAA--.10650S3 X-CM-SenderInfo: xpdruxter6z05rqj20fqof0/ X-Coremail-Antispam: 1Uk129KBjDUn29KB7ZKAUJUUUUU529EdanIXcx71UUUUU7KY7 ZEXasCq-sGcSsGvfJ3UbIjqfuFe4nvWSU5nxnvy29KBjDU0xBIdaVrnUUvcSsGvfC2Kfnx nUUI43ZEXa7xR_UUUUUUUUU== Steal time feature is added here in kvm side, VM can search supported features provided by KVM hypervisor, feature KVM_FEATURE_STEAL_TIME is added here. Like x86, steal time structure is saved in guest memory, one hypercall function KVM_HCALL_FUNC_NOTIFY is added to notify KVM to enable the feature. One cpu attr ioctl command KVM_LOONGARCH_VCPU_PVTIME_CTRL is added to save and restore base address of steal time structure when VM is migrated. Signed-off-by: Bibo Mao --- arch/loongarch/include/asm/kvm_host.h | 7 ++ arch/loongarch/include/asm/kvm_para.h | 10 ++ arch/loongarch/include/asm/kvm_vcpu.h | 4 + arch/loongarch/include/asm/loongarch.h | 1 + arch/loongarch/include/uapi/asm/kvm.h | 4 + arch/loongarch/kvm/Kconfig | 1 + arch/loongarch/kvm/exit.c | 38 +++++++- arch/loongarch/kvm/vcpu.c | 124 +++++++++++++++++++++++++ 8 files changed, 187 insertions(+), 2 deletions(-) diff --git a/arch/loongarch/include/asm/kvm_host.h b/arch/loongarch/include/asm/kvm_host.h index c87b6ea0ec47..2eb2f7572023 100644 --- a/arch/loongarch/include/asm/kvm_host.h +++ b/arch/loongarch/include/asm/kvm_host.h @@ -30,6 +30,7 @@ #define KVM_PRIVATE_MEM_SLOTS 0 #define KVM_HALT_POLL_NS_DEFAULT 500000 +#define KVM_REQ_STEAL_UPDATE KVM_ARCH_REQ(1) #define KVM_GUESTDBG_SW_BP_MASK \ (KVM_GUESTDBG_ENABLE | KVM_GUESTDBG_USE_SW_BP) @@ -201,6 +202,12 @@ struct kvm_vcpu_arch { struct kvm_mp_state mp_state; /* cpucfg */ u32 cpucfg[KVM_MAX_CPUCFG_REGS]; + /* paravirt steal time */ + struct { + u64 guest_addr; + u64 last_steal; + struct gfn_to_hva_cache cache; + } st; }; static inline unsigned long readl_sw_gcsr(struct loongarch_csrs *csr, int reg) diff --git a/arch/loongarch/include/asm/kvm_para.h b/arch/loongarch/include/asm/kvm_para.h index 4ba2312e5f8c..a9ba8185d4af 100644 --- a/arch/loongarch/include/asm/kvm_para.h +++ b/arch/loongarch/include/asm/kvm_para.h @@ -14,6 +14,7 @@ #define KVM_HCALL_SERVICE HYPERCALL_ENCODE(HYPERVISOR_KVM, KVM_HCALL_CODE_SERVICE) #define KVM_HCALL_FUNC_IPI 1 +#define KVM_HCALL_FUNC_NOTIFY 2 #define KVM_HCALL_SWDBG HYPERCALL_ENCODE(HYPERVISOR_KVM, KVM_HCALL_CODE_SWDBG) @@ -24,6 +25,15 @@ #define KVM_HCALL_INVALID_CODE -1UL #define KVM_HCALL_INVALID_PARAMETER -2UL +#define KVM_STEAL_PHYS_VALID BIT_ULL(0) +#define KVM_STEAL_PHYS_MASK GENMASK_ULL(63, 6) +struct kvm_steal_time { + __u64 steal; + __u32 version; + __u32 flags; + __u32 pad[12]; +}; + /* * Hypercall interface for KVM hypervisor * diff --git a/arch/loongarch/include/asm/kvm_vcpu.h b/arch/loongarch/include/asm/kvm_vcpu.h index 590a92cb5416..d7e51300a89f 100644 --- a/arch/loongarch/include/asm/kvm_vcpu.h +++ b/arch/loongarch/include/asm/kvm_vcpu.h @@ -120,4 +120,8 @@ static inline void kvm_write_reg(struct kvm_vcpu *vcpu, int num, unsigned long v vcpu->arch.gprs[num] = val; } +static inline bool kvm_pvtime_supported(void) +{ + return !!sched_info_on(); +} #endif /* __ASM_LOONGARCH_KVM_VCPU_H__ */ diff --git a/arch/loongarch/include/asm/loongarch.h b/arch/loongarch/include/asm/loongarch.h index eb09adda54b7..7a4633ef284b 100644 --- a/arch/loongarch/include/asm/loongarch.h +++ b/arch/loongarch/include/asm/loongarch.h @@ -169,6 +169,7 @@ #define KVM_SIGNATURE "KVM\0" #define CPUCFG_KVM_FEATURE (CPUCFG_KVM_BASE + 4) #define KVM_FEATURE_IPI BIT(1) +#define KVM_FEATURE_STEAL_TIME BIT(2) #ifndef __ASSEMBLY__ diff --git a/arch/loongarch/include/uapi/asm/kvm.h b/arch/loongarch/include/uapi/asm/kvm.h index f9abef382317..ddc5cab0ffd0 100644 --- a/arch/loongarch/include/uapi/asm/kvm.h +++ b/arch/loongarch/include/uapi/asm/kvm.h @@ -81,7 +81,11 @@ struct kvm_fpu { #define LOONGARCH_REG_64(TYPE, REG) (TYPE | KVM_REG_SIZE_U64 | (REG << LOONGARCH_REG_SHIFT)) #define KVM_IOC_CSRID(REG) LOONGARCH_REG_64(KVM_REG_LOONGARCH_CSR, REG) #define KVM_IOC_CPUCFG(REG) LOONGARCH_REG_64(KVM_REG_LOONGARCH_CPUCFG, REG) + +/* Device Control API on vcpu fd */ #define KVM_LOONGARCH_VCPU_CPUCFG 0 +#define KVM_LOONGARCH_VCPU_PVTIME_CTRL 1 +#define KVM_LOONGARCH_VCPU_PVTIME_GPA 0 struct kvm_debug_exit_arch { }; diff --git a/arch/loongarch/kvm/Kconfig b/arch/loongarch/kvm/Kconfig index c4ef2b4d9797..248744b4d086 100644 --- a/arch/loongarch/kvm/Kconfig +++ b/arch/loongarch/kvm/Kconfig @@ -29,6 +29,7 @@ config KVM select KVM_MMIO select HAVE_KVM_READONLY_MEM select KVM_XFER_TO_GUEST_WORK + select SCHED_INFO help Support hosting virtualized guest machines using hardware virtualization extensions. You will need diff --git a/arch/loongarch/kvm/exit.c b/arch/loongarch/kvm/exit.c index c86e099af5ca..e2abd97fb13f 100644 --- a/arch/loongarch/kvm/exit.c +++ b/arch/loongarch/kvm/exit.c @@ -24,7 +24,7 @@ static int kvm_emu_cpucfg(struct kvm_vcpu *vcpu, larch_inst inst) { int rd, rj; - unsigned int index; + unsigned int index, ret; if (inst.reg2_format.opcode != cpucfg_op) return EMULATE_FAIL; @@ -50,7 +50,10 @@ static int kvm_emu_cpucfg(struct kvm_vcpu *vcpu, larch_inst inst) vcpu->arch.gprs[rd] = *(unsigned int *)KVM_SIGNATURE; break; case CPUCFG_KVM_FEATURE: - vcpu->arch.gprs[rd] = KVM_FEATURE_IPI; + ret = KVM_FEATURE_IPI; + if (sched_info_on()) + ret |= KVM_FEATURE_STEAL_TIME; + vcpu->arch.gprs[rd] = ret; break; default: vcpu->arch.gprs[rd] = 0; @@ -687,6 +690,34 @@ static int kvm_handle_fpu_disabled(struct kvm_vcpu *vcpu) return RESUME_GUEST; } +static long kvm_save_notify(struct kvm_vcpu *vcpu) +{ + unsigned long id, data; + + id = kvm_read_reg(vcpu, LOONGARCH_GPR_A1); + data = kvm_read_reg(vcpu, LOONGARCH_GPR_A2); + switch (id) { + case KVM_FEATURE_STEAL_TIME: + if (!kvm_pvtime_supported()) + return KVM_HCALL_INVALID_CODE; + + if (data & ~(KVM_STEAL_PHYS_MASK | KVM_STEAL_PHYS_VALID)) + return KVM_HCALL_INVALID_PARAMETER; + + vcpu->arch.st.guest_addr = data; + if (!(data & KVM_STEAL_PHYS_VALID)) + break; + + vcpu->arch.st.last_steal = current->sched_info.run_delay; + kvm_make_request(KVM_REQ_STEAL_UPDATE, vcpu); + break; + default: + break; + }; + + return 0; +}; + /* * kvm_handle_lsx_disabled() - Guest used LSX while disabled in root. * @vcpu: Virtual CPU context. @@ -758,6 +789,9 @@ static void kvm_handle_service(struct kvm_vcpu *vcpu) kvm_send_pv_ipi(vcpu); ret = KVM_HCALL_SUCCESS; break; + case KVM_HCALL_FUNC_NOTIFY: + ret = kvm_save_notify(vcpu); + break; default: ret = KVM_HCALL_INVALID_CODE; break; diff --git a/arch/loongarch/kvm/vcpu.c b/arch/loongarch/kvm/vcpu.c index 9e8030d45129..382796f1d3e6 100644 --- a/arch/loongarch/kvm/vcpu.c +++ b/arch/loongarch/kvm/vcpu.c @@ -31,6 +31,117 @@ const struct kvm_stats_header kvm_vcpu_stats_header = { sizeof(kvm_vcpu_stats_desc), }; +static void kvm_update_stolen_time(struct kvm_vcpu *vcpu) +{ + struct kvm_steal_time __user *st; + struct gfn_to_hva_cache *ghc; + struct kvm_memslots *slots; + gpa_t gpa; + u64 steal; + u32 version; + + ghc = &vcpu->arch.st.cache; + gpa = vcpu->arch.st.guest_addr; + if (!(gpa & KVM_STEAL_PHYS_VALID)) + return; + + gpa &= KVM_STEAL_PHYS_MASK; + slots = kvm_memslots(vcpu->kvm); + if (slots->generation != ghc->generation || gpa != ghc->gpa) { + if (kvm_gfn_to_hva_cache_init(vcpu->kvm, ghc, gpa, + sizeof(*st))) { + ghc->gpa = INVALID_GPA; + return; + } + } + + st = (struct kvm_steal_time __user *)ghc->hva; + unsafe_get_user(version, &st->version, out); + if (version & 1) + version += 1; + version += 1; + unsafe_put_user(version, &st->version, out); + smp_wmb(); + + unsafe_get_user(steal, &st->steal, out); + steal += current->sched_info.run_delay - + vcpu->arch.st.last_steal; + vcpu->arch.st.last_steal = current->sched_info.run_delay; + unsafe_put_user(steal, &st->steal, out); + + smp_wmb(); + version += 1; + unsafe_put_user(version, &st->version, out); +out: + mark_page_dirty_in_slot(vcpu->kvm, ghc->memslot, gpa_to_gfn(ghc->gpa)); +} + +static int kvm_loongarch_pvtime_has_attr(struct kvm_vcpu *vcpu, + struct kvm_device_attr *attr) +{ + if (!kvm_pvtime_supported() || + attr->attr != KVM_LOONGARCH_VCPU_PVTIME_GPA) + return -ENXIO; + + return 0; +} + +static int kvm_loongarch_pvtime_get_attr(struct kvm_vcpu *vcpu, + struct kvm_device_attr *attr) +{ + u64 __user *user = (u64 __user *)attr->addr; + u64 gpa; + + if (!kvm_pvtime_supported() || + attr->attr != KVM_LOONGARCH_VCPU_PVTIME_GPA) + return -ENXIO; + + gpa = vcpu->arch.st.guest_addr; + if (put_user(gpa, user)) + return -EFAULT; + + return 0; +} + +static int kvm_loongarch_pvtime_set_attr(struct kvm_vcpu *vcpu, + struct kvm_device_attr *attr) +{ + u64 __user *user = (u64 __user *)attr->addr; + struct kvm *kvm = vcpu->kvm; + u64 gpa; + int ret = 0; + int idx; + + if (!kvm_pvtime_supported() || + attr->attr != KVM_LOONGARCH_VCPU_PVTIME_GPA) + return -ENXIO; + + if (get_user(gpa, user)) + return -EFAULT; + + if (gpa & ~(KVM_STEAL_PHYS_MASK | KVM_STEAL_PHYS_VALID)) + return -EINVAL; + + if (!(gpa & KVM_STEAL_PHYS_VALID)) { + vcpu->arch.st.guest_addr = gpa; + return 0; + } + + /* Check the address is in a valid memslot */ + idx = srcu_read_lock(&kvm->srcu); + if (kvm_is_error_hva(gfn_to_hva(kvm, gpa >> PAGE_SHIFT))) + ret = -EINVAL; + srcu_read_unlock(&kvm->srcu, idx); + + if (!ret) { + vcpu->arch.st.guest_addr = gpa; + vcpu->arch.st.last_steal = current->sched_info.run_delay; + kvm_make_request(KVM_REQ_STEAL_UPDATE, vcpu); + } + + return ret; +} + /* * kvm_check_requests - check and handle pending vCPU requests * @@ -48,6 +159,9 @@ static int kvm_check_requests(struct kvm_vcpu *vcpu) if (kvm_dirty_ring_check_request(vcpu)) return RESUME_HOST; + if (kvm_check_request(KVM_REQ_STEAL_UPDATE, vcpu)) + kvm_update_stolen_time(vcpu); + return RESUME_GUEST; } @@ -671,6 +785,9 @@ static int kvm_loongarch_vcpu_has_attr(struct kvm_vcpu *vcpu, case KVM_LOONGARCH_VCPU_CPUCFG: ret = kvm_loongarch_cpucfg_has_attr(vcpu, attr); break; + case KVM_LOONGARCH_VCPU_PVTIME_CTRL: + ret = kvm_loongarch_pvtime_has_attr(vcpu, attr); + break; default: break; } @@ -703,6 +820,9 @@ static int kvm_loongarch_vcpu_get_attr(struct kvm_vcpu *vcpu, case KVM_LOONGARCH_VCPU_CPUCFG: ret = kvm_loongarch_get_cpucfg_attr(vcpu, attr); break; + case KVM_LOONGARCH_VCPU_PVTIME_CTRL: + ret = kvm_loongarch_pvtime_get_attr(vcpu, attr); + break; default: break; } @@ -725,6 +845,9 @@ static int kvm_loongarch_vcpu_set_attr(struct kvm_vcpu *vcpu, case KVM_LOONGARCH_VCPU_CPUCFG: ret = kvm_loongarch_cpucfg_set_attr(vcpu, attr); break; + case KVM_LOONGARCH_VCPU_PVTIME_CTRL: + ret = kvm_loongarch_pvtime_set_attr(vcpu, attr); + break; default: break; } @@ -1084,6 +1207,7 @@ static int _kvm_vcpu_load(struct kvm_vcpu *vcpu, int cpu) /* Control guest page CCA attribute */ change_csr_gcfg(CSR_GCFG_MATC_MASK, CSR_GCFG_MATC_ROOT); + kvm_make_request(KVM_REQ_STEAL_UPDATE, vcpu); /* Don't bother restoring registers multiple times unless necessary */ if (vcpu->arch.aux_inuse & KVM_LARCH_HWCSR_USABLE) From patchwork Fri May 24 07:38:12 2024 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Bibo Mao X-Patchwork-Id: 13672778 Received: from mail.loongson.cn (mail.loongson.cn [114.242.206.163]) by smtp.subspace.kernel.org (Postfix) with ESMTP id 0B4EE38FA3; Fri, 24 May 2024 07:38:22 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=114.242.206.163 ARC-Seal: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1716536305; cv=none; b=b5v0FpNzmSXZFM7nZ24PfwVqBCYLjDXskVGz1SaKeK/yONWNJCpVyXoVAPapeV3y+58VxIiyI0NZ7N2rF/ulT/cQHfQAzouL0Gy3rqzdVq28yahEEfCMwjk9M6Fn1/co4ElZtLW6yrFKDBG2l3f/mB/RgK2vILiceRk41Bts/eU= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1716536305; c=relaxed/simple; bh=P4S+dNcHYnfYJmNly4QLN7+vGJFStlg4TJNdtPEEaYk=; h=From:To:Cc:Subject:Date:Message-Id:In-Reply-To:References: MIME-Version; b=WlDGtFJjs5jLWo/cEl74938MTCOVsQUHOjldx9wbcOuEOpfjDwRl6CllERpqYAL5jg8vtq3iSqqt0GbpGj9ZLTGZVhmELXSAKJiaWb8SRt/Q8YVq5+buJGCFDy+YFHyAanZlI6rC+EaOnlrulXq4PfMRW3ioW3vVqk/eOEH9g6o= ARC-Authentication-Results: i=1; smtp.subspace.kernel.org; dmarc=none (p=none dis=none) header.from=loongson.cn; spf=pass smtp.mailfrom=loongson.cn; arc=none smtp.client-ip=114.242.206.163 Authentication-Results: smtp.subspace.kernel.org; dmarc=none (p=none dis=none) header.from=loongson.cn Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=loongson.cn Received: from loongson.cn (unknown [10.2.5.213]) by gateway (Coremail) with SMTP id _____8BxGvDoQ1Bm4EsDAA--.9022S3; Fri, 24 May 2024 15:38:16 +0800 (CST) Received: from localhost.localdomain (unknown [10.2.5.213]) by localhost.localdomain (Coremail) with SMTP id AQAAf8CxZcXlQ1BmScUHAA--.10650S4; Fri, 24 May 2024 15:38:14 +0800 (CST) From: Bibo Mao To: Tianrui Zhao , Huacai Chen Cc: Juergen Gross , kvm@vger.kernel.org, loongarch@lists.linux.dev, linux-kernel@vger.kernel.org, x86@kernel.org, virtualization@lists.linux.dev Subject: [PATCH v4 2/2] LoongArch: Add steal time support in guest side Date: Fri, 24 May 2024 15:38:12 +0800 Message-Id: <20240524073812.731032-3-maobibo@loongson.cn> X-Mailer: git-send-email 2.39.3 In-Reply-To: <20240524073812.731032-1-maobibo@loongson.cn> References: <20240524073812.731032-1-maobibo@loongson.cn> Precedence: bulk X-Mailing-List: kvm@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 X-CM-TRANSID: AQAAf8CxZcXlQ1BmScUHAA--.10650S4 X-CM-SenderInfo: xpdruxter6z05rqj20fqof0/ X-Coremail-Antispam: 1Uk129KBjDUn29KB7ZKAUJUUUUU529EdanIXcx71UUUUU7KY7 ZEXasCq-sGcSsGvfJ3UbIjqfuFe4nvWSU5nxnvy29KBjDU0xBIdaVrnUUvcSsGvfC2Kfnx nUUI43ZEXa7xR_UUUUUUUUU== Percpu struct kvm_steal_time is added here, its size is 64 bytes and also defined as 64 bytes, so that the whole structure is in one physical page. When vcpu is onlined, function pv_enable_steal_time() is called. This function will pass guest physical address of struct kvm_steal_time and tells hypervisor to enable steal time. When vcpu is offline, physical address is set as 0 and tells hypervisor to disable steal time. Here is output of vmstat on guest when there is workload on both host and guest. It shows steal time stat information. procs -----------memory---------- -----io---- -system-- ------cpu----- r b swpd free inact active bi bo in cs us sy id wa st 15 1 0 7583616 184112 72208 20 0 162 52 31 6 43 0 20 17 0 0 7583616 184704 72192 0 0 6318 6885 5 60 8 5 22 16 0 0 7583616 185392 72144 0 0 1766 1081 0 49 0 1 50 16 0 0 7583616 184816 72304 0 0 6300 6166 4 62 12 2 20 18 0 0 7583632 184480 72240 0 0 2814 1754 2 58 4 1 35 Signed-off-by: Bibo Mao --- .../admin-guide/kernel-parameters.txt | 2 +- arch/loongarch/Kconfig | 11 ++ arch/loongarch/include/asm/paravirt.h | 5 + arch/loongarch/kernel/paravirt.c | 133 ++++++++++++++++++ arch/loongarch/kernel/time.c | 2 + 5 files changed, 152 insertions(+), 1 deletion(-) diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt index ef25e06ec08c..3435eaab392b 100644 --- a/Documentation/admin-guide/kernel-parameters.txt +++ b/Documentation/admin-guide/kernel-parameters.txt @@ -4057,7 +4057,7 @@ prediction) vulnerability. System may allow data leaks with this option. - no-steal-acc [X86,PV_OPS,ARM64,PPC/PSERIES,RISCV,EARLY] Disable + no-steal-acc [X86,PV_OPS,ARM64,PPC/PSERIES,RISCV,LOONGARCH,EARLY] Disable paravirtualized steal time accounting. steal time is computed, but won't influence scheduler behaviour diff --git a/arch/loongarch/Kconfig b/arch/loongarch/Kconfig index 4bda59525a13..a1c4b0080039 100644 --- a/arch/loongarch/Kconfig +++ b/arch/loongarch/Kconfig @@ -641,6 +641,17 @@ config PARAVIRT over full virtualization. However, when run without a hypervisor the kernel is theoretically slower and slightly larger. +config PARAVIRT_TIME_ACCOUNTING + bool "Paravirtual steal time accounting" + select PARAVIRT + help + Select this option to enable fine granularity task steal time + accounting. Time spent executing other tasks in parallel with + the current vCPU is discounted from the vCPU power. To account for + that, there can be a small performance impact. + + If in doubt, say N here. + endmenu config ARCH_SELECT_MEMORY_MODEL diff --git a/arch/loongarch/include/asm/paravirt.h b/arch/loongarch/include/asm/paravirt.h index 0965710f47f2..dddec49671ae 100644 --- a/arch/loongarch/include/asm/paravirt.h +++ b/arch/loongarch/include/asm/paravirt.h @@ -18,6 +18,7 @@ static inline u64 paravirt_steal_clock(int cpu) } int __init pv_ipi_init(void); +int __init pv_time_init(void); #else @@ -26,5 +27,9 @@ static inline int pv_ipi_init(void) return 0; } +static inline int pv_time_init(void) +{ + return 0; +} #endif // CONFIG_PARAVIRT #endif diff --git a/arch/loongarch/kernel/paravirt.c b/arch/loongarch/kernel/paravirt.c index 1633ed4f692f..68f3cd75862a 100644 --- a/arch/loongarch/kernel/paravirt.c +++ b/arch/loongarch/kernel/paravirt.c @@ -4,11 +4,14 @@ #include #include #include +#include #include #include struct static_key paravirt_steal_enabled; struct static_key paravirt_steal_rq_enabled; +static DEFINE_PER_CPU(struct kvm_steal_time, steal_time) __aligned(64); +static int has_steal_clock; static u64 native_steal_clock(int cpu) { @@ -17,6 +20,57 @@ static u64 native_steal_clock(int cpu) DEFINE_STATIC_CALL(pv_steal_clock, native_steal_clock); +static bool steal_acc = true; +static int __init parse_no_stealacc(char *arg) +{ + steal_acc = false; + return 0; +} +early_param("no-steal-acc", parse_no_stealacc); + +static u64 para_steal_clock(int cpu) +{ + u64 steal; + struct kvm_steal_time *src; + int version; + + src = &per_cpu(steal_time, cpu); + do { + + version = src->version; + /* Make sure that the version is read before the steal */ + virt_rmb(); + steal = src->steal; + /* Make sure that the steal is read before the next version */ + virt_rmb(); + + } while ((version & 1) || (version != src->version)); + return steal; +} + +static int pv_enable_steal_time(void) +{ + int cpu = smp_processor_id(); + struct kvm_steal_time *st; + unsigned long addr; + + if (!has_steal_clock) + return -EPERM; + + st = &per_cpu(steal_time, cpu); + addr = per_cpu_ptr_to_phys(st); + + /* The whole structure kvm_steal_time should be one page */ + if (PFN_DOWN(addr) != PFN_DOWN(addr + sizeof(*st))) { + pr_warn("Illegal PV steal time addr %lx\n", addr); + return -EFAULT; + } + + addr |= KVM_STEAL_PHYS_VALID; + kvm_hypercall2(KVM_HCALL_FUNC_NOTIFY, KVM_FEATURE_STEAL_TIME, addr); + return 0; +} + #ifdef CONFIG_SMP static void pv_send_ipi_single(int cpu, unsigned int action) { @@ -112,6 +166,32 @@ static void pv_init_ipi(void) if (r < 0) panic("SWI0 IRQ request failed\n"); } + +static void pv_disable_steal_time(void) +{ + if (has_steal_clock) + kvm_hypercall2(KVM_HCALL_FUNC_NOTIFY, KVM_FEATURE_STEAL_TIME, 0); +} + +static int pv_time_cpu_online(unsigned int cpu) +{ + unsigned long flags; + + local_irq_save(flags); + pv_enable_steal_time(); + local_irq_restore(flags); + return 0; +} + +static int pv_time_cpu_down_prepare(unsigned int cpu) +{ + unsigned long flags; + + local_irq_save(flags); + pv_disable_steal_time(); + local_irq_restore(flags); + return 0; +} #endif static bool kvm_para_available(void) @@ -149,3 +229,56 @@ int __init pv_ipi_init(void) return 0; } + +static void pv_cpu_reboot(void *unused) +{ + pv_disable_steal_time(); +} + +static int pv_reboot_notify(struct notifier_block *nb, unsigned long code, + void *unused) +{ + on_each_cpu(pv_cpu_reboot, NULL, 1); + return NOTIFY_DONE; +} + +static struct notifier_block pv_reboot_nb = { + .notifier_call = pv_reboot_notify, +}; + +int __init pv_time_init(void) +{ + int feature; + + if (!cpu_has_hypervisor) + return 0; + if (!kvm_para_available()) + return 0; + + feature = read_cpucfg(CPUCFG_KVM_FEATURE); + if (!(feature & KVM_FEATURE_STEAL_TIME)) + return 0; + + has_steal_clock = 1; + if (pv_enable_steal_time()) { + has_steal_clock = 0; + return 0; + } + + register_reboot_notifier(&pv_reboot_nb); + static_call_update(pv_steal_clock, para_steal_clock); + static_key_slow_inc(¶virt_steal_enabled); +#ifdef CONFIG_PARAVIRT_TIME_ACCOUNTING + if (steal_acc) + static_key_slow_inc(¶virt_steal_rq_enabled); +#endif + +#ifdef CONFIG_SMP + if (cpuhp_setup_state_nocalls(CPUHP_AP_ONLINE_DYN, + "loongarch/pvi_time:online", + pv_time_cpu_online, pv_time_cpu_down_prepare) < 0) + pr_err("Failed to install cpu hotplug callbacks\n"); +#endif + pr_info("Using stolen time PV\n"); + return 0; +} diff --git a/arch/loongarch/kernel/time.c b/arch/loongarch/kernel/time.c index fd5354f9be7c..46d7d40c87e3 100644 --- a/arch/loongarch/kernel/time.c +++ b/arch/loongarch/kernel/time.c @@ -15,6 +15,7 @@ #include #include +#include #include u64 cpu_clock_freq; @@ -214,4 +215,5 @@ void __init time_init(void) constant_clockevent_init(); constant_clocksource_init(); + pv_time_init(); }