From nobody Fri Dec 19 07:21:12 2025 Received: from mail.loongson.cn (mail.loongson.cn [114.242.206.163]) by smtp.subspace.kernel.org (Postfix) with ESMTP id 1F1731EDA2C; Wed, 17 Dec 2025 03:49:41 +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=1765943384; cv=none; b=X6QiKV37YacrnFDrX9Koff5vsMVa9rn9oLfB34ZmvqnSlHvUmlOC/DGErTLC4xLGXWQMlDQFpMQM/xjX5lD1FeTQ1oar6XmgJXovi+oqu6FNY1LXgLQxL7kUQoUqcL6wOLx2w0GE8V7vBgi/CdFh0xeH16iPTnBYxXuSoWyT5w8= ARC-Message-Signature: i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1765943384; c=relaxed/simple; bh=Wv0vfXgbERAhOsD6r93ZE1zOzlSN/9LFQMTNh54kg70=; h=From:To:Cc:Subject:Date:Message-Id:In-Reply-To:References: MIME-Version; b=GMi7kHBUJJkdESPiSDvtt3KPDkN2RA/BX9t83toVpVRNUeeoYksykiypMuEpWLCwtxKMsHg8hLvU4UOOst5rdzlrRh+KIf7554UfGkGdOelgzJFaRunuVGE/NeRm1ivhW8PIqs+F2SMlhCtyixKESFs+zy+JER2znehg/a+TZEM= 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 [10.2.5.185]) by gateway (Coremail) with SMTP id _____8BxcfBKKEJpygAAAA--.49S3; Wed, 17 Dec 2025 11:49:30 +0800 (CST) Received: from localhost.localdomain (unknown [10.2.5.185]) by front1 (Coremail) with SMTP id qMiowJCxPMJFKEJpksAAAA--.913S3; Wed, 17 Dec 2025 11:49:27 +0800 (CST) From: Xianglai Li To: loongarch@lists.linux.dev, linux-kernel@vger.kernel.org, kvm@vger.kernel.org, lixianglai@loongson.cn Cc: stable@vger.kernel.org, Huacai Chen , WANG Xuerui , Tianrui Zhao , Bibo Mao , Charlie Jenkins , Thomas Gleixner Subject: [PATCH 1/2] LoongArch: KVM: Compile the switch.S file directly into the kernel Date: Wed, 17 Dec 2025 11:24:49 +0800 Message-Id: <20251217032450.954344-2-lixianglai@loongson.cn> X-Mailer: git-send-email 2.39.1 In-Reply-To: <20251217032450.954344-1-lixianglai@loongson.cn> References: <20251217032450.954344-1-lixianglai@loongson.cn> Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-CM-TRANSID: qMiowJCxPMJFKEJpksAAAA--.913S3 X-CM-SenderInfo: 5ol0xt5qjotxo6or00hjvr0hdfq/ X-Coremail-Antispam: 1Uk129KBjDUn29KB7ZKAUJUUUUU529EdanIXcx71UUUUU7KY7 ZEXasCq-sGcSsGvfJ3UbIjqfuFe4nvWSU5nxnvy29KBjDU0xBIdaVrnUUvcSsGvfC2Kfnx nUUI43ZEXa7xR_UUUUUUUUU== Content-Type: text/plain; charset="utf-8" If we directly compile the switch.S file into the kernel, the address of the kvm_exc_entry function will definitely be within the DMW memory area. Therefore, we will no longer need to perform a copy relocation of kvm_exc_entry. Based on the above description, compile switch.S directly into the kernel, and then remove the copy relocation execution logic for the kvm_exc_entry function. Cc: stable@vger.kernel.org Signed-off-by: Xianglai Li --- Cc: Huacai Chen Cc: WANG Xuerui Cc: Tianrui Zhao Cc: Bibo Mao Cc: Charlie Jenkins Cc: Xianglai Li Cc: Thomas Gleixner arch/loongarch/Kbuild | 2 +- arch/loongarch/include/asm/asm-prototypes.h | 16 ++++++++++ arch/loongarch/include/asm/kvm_host.h | 5 +-- arch/loongarch/include/asm/kvm_vcpu.h | 20 ++++++------ arch/loongarch/kvm/Makefile | 2 +- arch/loongarch/kvm/main.c | 35 ++------------------- arch/loongarch/kvm/switch.S | 22 ++++++++++--- 7 files changed, 49 insertions(+), 53 deletions(-) diff --git a/arch/loongarch/Kbuild b/arch/loongarch/Kbuild index beb8499dd8ed..1c7a0dbe5e72 100644 --- a/arch/loongarch/Kbuild +++ b/arch/loongarch/Kbuild @@ -3,7 +3,7 @@ obj-y +=3D mm/ obj-y +=3D net/ obj-y +=3D vdso/ =20 -obj-$(CONFIG_KVM) +=3D kvm/ +obj-$(subst m,y,$(CONFIG_KVM)) +=3D kvm/ =20 # for cleaning subdir- +=3D boot diff --git a/arch/loongarch/include/asm/asm-prototypes.h b/arch/loongarch/i= nclude/asm/asm-prototypes.h index 704066b4f736..eb591276d191 100644 --- a/arch/loongarch/include/asm/asm-prototypes.h +++ b/arch/loongarch/include/asm/asm-prototypes.h @@ -20,3 +20,19 @@ asmlinkage void noinstr __no_stack_protector ret_from_ke= rnel_thread(struct task_ struct pt_regs *regs, int (*fn)(void *), void *fn_arg); + +void kvm_exc_entry(void); +int kvm_enter_guest(void *run, void *vcpu); + +#ifdef CONFIG_CPU_HAS_LSX +void kvm_save_lsx(void *fpu); +void kvm_restore_lsx(void *fpu); +#endif + +#ifdef CONFIG_CPU_HAS_LASX +void kvm_save_lasx(void *fpu); +void kvm_restore_lasx(void *fpu); +#endif + +void kvm_save_fpu(void *fpu); +void kvm_restore_fpu(void *fpu); diff --git a/arch/loongarch/include/asm/kvm_host.h b/arch/loongarch/include= /asm/kvm_host.h index e4fe5b8e8149..0aa7679536cc 100644 --- a/arch/loongarch/include/asm/kvm_host.h +++ b/arch/loongarch/include/asm/kvm_host.h @@ -85,7 +85,6 @@ struct kvm_context { struct kvm_world_switch { int (*exc_entry)(void); int (*enter_guest)(struct kvm_run *run, struct kvm_vcpu *vcpu); - unsigned long page_order; }; =20 #define MAX_PGTABLE_LEVELS 4 @@ -344,11 +343,9 @@ enum hrtimer_restart kvm_swtimer_wakeup(struct hrtimer= *timer); void kvm_arch_flush_remote_tlbs_memslot(struct kvm *kvm, const struct kvm_= memory_slot *memslot); void kvm_init_vmcs(struct kvm *kvm); void kvm_exc_entry(void); -int kvm_enter_guest(struct kvm_run *run, struct kvm_vcpu *vcpu); +int kvm_enter_guest(void *run, void *vcpu); =20 extern unsigned long vpid_mask; -extern const unsigned long kvm_exception_size; -extern const unsigned long kvm_enter_guest_size; extern struct kvm_world_switch *kvm_loongarch_ops; =20 #define SW_GCSR (1 << 0) diff --git a/arch/loongarch/include/asm/kvm_vcpu.h b/arch/loongarch/include= /asm/kvm_vcpu.h index 3784ab4ccdb5..8af98a3d7b0c 100644 --- a/arch/loongarch/include/asm/kvm_vcpu.h +++ b/arch/loongarch/include/asm/kvm_vcpu.h @@ -53,28 +53,28 @@ void kvm_deliver_exception(struct kvm_vcpu *vcpu); =20 void kvm_own_fpu(struct kvm_vcpu *vcpu); void kvm_lose_fpu(struct kvm_vcpu *vcpu); -void kvm_save_fpu(struct loongarch_fpu *fpu); -void kvm_restore_fpu(struct loongarch_fpu *fpu); +void kvm_save_fpu(void *fpu); +void kvm_restore_fpu(void *fpu); void kvm_restore_fcsr(struct loongarch_fpu *fpu); =20 #ifdef CONFIG_CPU_HAS_LSX int kvm_own_lsx(struct kvm_vcpu *vcpu); -void kvm_save_lsx(struct loongarch_fpu *fpu); -void kvm_restore_lsx(struct loongarch_fpu *fpu); +void kvm_save_lsx(void *fpu); +void kvm_restore_lsx(void *fpu); #else static inline int kvm_own_lsx(struct kvm_vcpu *vcpu) { return -EINVAL; } -static inline void kvm_save_lsx(struct loongarch_fpu *fpu) { } -static inline void kvm_restore_lsx(struct loongarch_fpu *fpu) { } +static inline void kvm_save_lsx(void *fpu) { } +static inline void kvm_restore_lsx(void *fpu) { } #endif =20 #ifdef CONFIG_CPU_HAS_LASX int kvm_own_lasx(struct kvm_vcpu *vcpu); -void kvm_save_lasx(struct loongarch_fpu *fpu); -void kvm_restore_lasx(struct loongarch_fpu *fpu); +void kvm_save_lasx(void *fpu); +void kvm_restore_lasx(void *fpu); #else static inline int kvm_own_lasx(struct kvm_vcpu *vcpu) { return -EINVAL; } -static inline void kvm_save_lasx(struct loongarch_fpu *fpu) { } -static inline void kvm_restore_lasx(struct loongarch_fpu *fpu) { } +static inline void kvm_save_lasx(void *fpu) { } +static inline void kvm_restore_lasx(void *fpu) { } #endif =20 #ifdef CONFIG_CPU_HAS_LBT diff --git a/arch/loongarch/kvm/Makefile b/arch/loongarch/kvm/Makefile index cb41d9265662..fe665054f824 100644 --- a/arch/loongarch/kvm/Makefile +++ b/arch/loongarch/kvm/Makefile @@ -11,7 +11,7 @@ kvm-y +=3D exit.o kvm-y +=3D interrupt.o kvm-y +=3D main.o kvm-y +=3D mmu.o -kvm-y +=3D switch.o +obj-y +=3D switch.o kvm-y +=3D timer.o kvm-y +=3D tlb.o kvm-y +=3D vcpu.o diff --git a/arch/loongarch/kvm/main.c b/arch/loongarch/kvm/main.c index 80ea63d465b8..67d234540ed4 100644 --- a/arch/loongarch/kvm/main.c +++ b/arch/loongarch/kvm/main.c @@ -340,8 +340,7 @@ void kvm_arch_disable_virtualization_cpu(void) =20 static int kvm_loongarch_env_init(void) { - int cpu, order, ret; - void *addr; + int cpu, ret; struct kvm_context *context; =20 vmcs =3D alloc_percpu(struct kvm_context); @@ -357,30 +356,8 @@ static int kvm_loongarch_env_init(void) return -ENOMEM; } =20 - /* - * PGD register is shared between root kernel and kvm hypervisor. - * So world switch entry should be in DMW area rather than TLB area - * to avoid page fault reenter. - * - * In future if hardware pagetable walking is supported, we won't - * need to copy world switch code to DMW area. - */ - order =3D get_order(kvm_exception_size + kvm_enter_guest_size); - addr =3D (void *)__get_free_pages(GFP_KERNEL, order); - if (!addr) { - free_percpu(vmcs); - vmcs =3D NULL; - kfree(kvm_loongarch_ops); - kvm_loongarch_ops =3D NULL; - return -ENOMEM; - } - - memcpy(addr, kvm_exc_entry, kvm_exception_size); - memcpy(addr + kvm_exception_size, kvm_enter_guest, kvm_enter_guest_size); - flush_icache_range((unsigned long)addr, (unsigned long)addr + kvm_excepti= on_size + kvm_enter_guest_size); - kvm_loongarch_ops->exc_entry =3D addr; - kvm_loongarch_ops->enter_guest =3D addr + kvm_exception_size; - kvm_loongarch_ops->page_order =3D order; + kvm_loongarch_ops->exc_entry =3D (void *)kvm_exc_entry; + kvm_loongarch_ops->enter_guest =3D (void *)kvm_enter_guest; =20 vpid_mask =3D read_csr_gstat(); vpid_mask =3D (vpid_mask & CSR_GSTAT_GIDBIT) >> CSR_GSTAT_GIDBIT_SHIFT; @@ -414,16 +391,10 @@ static int kvm_loongarch_env_init(void) =20 static void kvm_loongarch_env_exit(void) { - unsigned long addr; - if (vmcs) free_percpu(vmcs); =20 if (kvm_loongarch_ops) { - if (kvm_loongarch_ops->exc_entry) { - addr =3D (unsigned long)kvm_loongarch_ops->exc_entry; - free_pages(addr, kvm_loongarch_ops->page_order); - } kfree(kvm_loongarch_ops); } =20 diff --git a/arch/loongarch/kvm/switch.S b/arch/loongarch/kvm/switch.S index f1768b7a6194..93845ce53651 100644 --- a/arch/loongarch/kvm/switch.S +++ b/arch/loongarch/kvm/switch.S @@ -5,6 +5,7 @@ =20 #include #include +#include #include #include #include @@ -100,10 +101,18 @@ * - is still in guest mode, such as pgd table/vmid registers etc, * - will fix with hw page walk enabled in future * load kvm_vcpu from reserved CSR KVM_VCPU_KS, and save a2 to KVM_TEMP_KS + * + * PGD register is shared between root kernel and kvm hypervisor. + * So world switch entry should be in DMW area rather than TLB area + * to avoid page fault reenter. + * + * In future if hardware pagetable walking is supported, we won't + * need to copy world switch code to DMW area. */ .text .cfi_sections .debug_frame SYM_CODE_START(kvm_exc_entry) + .p2align PAGE_SHIFT UNWIND_HINT_UNDEFINED csrwr a2, KVM_TEMP_KS csrrd a2, KVM_VCPU_KS @@ -190,8 +199,8 @@ ret_to_host: kvm_restore_host_gpr a2 jr ra =20 -SYM_INNER_LABEL(kvm_exc_entry_end, SYM_L_LOCAL) SYM_CODE_END(kvm_exc_entry) +EXPORT_SYMBOL(kvm_exc_entry) =20 /* * int kvm_enter_guest(struct kvm_run *run, struct kvm_vcpu *vcpu) @@ -215,8 +224,8 @@ SYM_FUNC_START(kvm_enter_guest) /* Save kvm_vcpu to kscratch */ csrwr a1, KVM_VCPU_KS kvm_switch_to_guest -SYM_INNER_LABEL(kvm_enter_guest_end, SYM_L_LOCAL) SYM_FUNC_END(kvm_enter_guest) +EXPORT_SYMBOL(kvm_enter_guest) =20 SYM_FUNC_START(kvm_save_fpu) fpu_save_csr a0 t1 @@ -224,6 +233,7 @@ SYM_FUNC_START(kvm_save_fpu) fpu_save_cc a0 t1 t2 jr ra SYM_FUNC_END(kvm_save_fpu) +EXPORT_SYMBOL(kvm_save_fpu) =20 SYM_FUNC_START(kvm_restore_fpu) fpu_restore_double a0 t1 @@ -231,6 +241,7 @@ SYM_FUNC_START(kvm_restore_fpu) fpu_restore_cc a0 t1 t2 jr ra SYM_FUNC_END(kvm_restore_fpu) +EXPORT_SYMBOL(kvm_restore_fpu) =20 #ifdef CONFIG_CPU_HAS_LSX SYM_FUNC_START(kvm_save_lsx) @@ -239,6 +250,7 @@ SYM_FUNC_START(kvm_save_lsx) lsx_save_data a0 t1 jr ra SYM_FUNC_END(kvm_save_lsx) +EXPORT_SYMBOL(kvm_save_lsx) =20 SYM_FUNC_START(kvm_restore_lsx) lsx_restore_data a0 t1 @@ -246,6 +258,7 @@ SYM_FUNC_START(kvm_restore_lsx) fpu_restore_csr a0 t1 t2 jr ra SYM_FUNC_END(kvm_restore_lsx) +EXPORT_SYMBOL(kvm_restore_lsx) #endif =20 #ifdef CONFIG_CPU_HAS_LASX @@ -255,6 +268,7 @@ SYM_FUNC_START(kvm_save_lasx) lasx_save_data a0 t1 jr ra SYM_FUNC_END(kvm_save_lasx) +EXPORT_SYMBOL(kvm_save_lasx) =20 SYM_FUNC_START(kvm_restore_lasx) lasx_restore_data a0 t1 @@ -262,10 +276,8 @@ SYM_FUNC_START(kvm_restore_lasx) fpu_restore_csr a0 t1 t2 jr ra SYM_FUNC_END(kvm_restore_lasx) +EXPORT_SYMBOL(kvm_restore_lasx) #endif - .section ".rodata" -SYM_DATA(kvm_exception_size, .quad kvm_exc_entry_end - kvm_exc_entry) -SYM_DATA(kvm_enter_guest_size, .quad kvm_enter_guest_end - kvm_enter_guest) =20 #ifdef CONFIG_CPU_HAS_LBT STACK_FRAME_NON_STANDARD kvm_restore_fpu --=20 2.39.1