From 749cf76c5a363e1383108a914ea09530bfa0bd43 Mon Sep 17 00:00:00 2001 From: Christoffer Dall Date: Sun, 20 Jan 2013 18:28:06 -0500 Subject: KVM: ARM: Initial skeleton to compile KVM support Targets KVM support for Cortex A-15 processors. Contains all the framework components, make files, header files, some tracing functionality, and basic user space API. Only supported core is Cortex-A15 for now. Most functionality is in arch/arm/kvm/* or arch/arm/include/asm/kvm_*.h. Reviewed-by: Will Deacon Reviewed-by: Marcelo Tosatti Signed-off-by: Rusty Russell Signed-off-by: Marc Zyngier Signed-off-by: Christoffer Dall --- arch/arm/include/asm/kvm_asm.h | 58 ++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 58 insertions(+) create mode 100644 arch/arm/include/asm/kvm_asm.h (limited to 'arch/arm/include/asm/kvm_asm.h') diff --git a/arch/arm/include/asm/kvm_asm.h b/arch/arm/include/asm/kvm_asm.h new file mode 100644 index 000000000000..f9993e5fb695 --- /dev/null +++ b/arch/arm/include/asm/kvm_asm.h @@ -0,0 +1,58 @@ +/* + * Copyright (C) 2012 - Virtual Open Systems and Columbia University + * Author: Christoffer Dall + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License, version 2, as + * published by the Free Software Foundation. + * + * This program 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 General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. + */ + +#ifndef __ARM_KVM_ASM_H__ +#define __ARM_KVM_ASM_H__ + +/* 0 is reserved as an invalid value. */ +#define c0_MPIDR 1 /* MultiProcessor ID Register */ +#define c0_CSSELR 2 /* Cache Size Selection Register */ +#define c1_SCTLR 3 /* System Control Register */ +#define c1_ACTLR 4 /* Auxilliary Control Register */ +#define c1_CPACR 5 /* Coprocessor Access Control */ +#define c2_TTBR0 6 /* Translation Table Base Register 0 */ +#define c2_TTBR0_high 7 /* TTBR0 top 32 bits */ +#define c2_TTBR1 8 /* Translation Table Base Register 1 */ +#define c2_TTBR1_high 9 /* TTBR1 top 32 bits */ +#define c2_TTBCR 10 /* Translation Table Base Control R. */ +#define c3_DACR 11 /* Domain Access Control Register */ +#define c5_DFSR 12 /* Data Fault Status Register */ +#define c5_IFSR 13 /* Instruction Fault Status Register */ +#define c5_ADFSR 14 /* Auxilary Data Fault Status R */ +#define c5_AIFSR 15 /* Auxilary Instrunction Fault Status R */ +#define c6_DFAR 16 /* Data Fault Address Register */ +#define c6_IFAR 17 /* Instruction Fault Address Register */ +#define c9_L2CTLR 18 /* Cortex A15 L2 Control Register */ +#define c10_PRRR 19 /* Primary Region Remap Register */ +#define c10_NMRR 20 /* Normal Memory Remap Register */ +#define c12_VBAR 21 /* Vector Base Address Register */ +#define c13_CID 22 /* Context ID Register */ +#define c13_TID_URW 23 /* Thread ID, User R/W */ +#define c13_TID_URO 24 /* Thread ID, User R/O */ +#define c13_TID_PRIV 25 /* Thread ID, Privileged */ +#define NR_CP15_REGS 26 /* Number of regs (incl. invalid) */ + +#define ARM_EXCEPTION_RESET 0 +#define ARM_EXCEPTION_UNDEFINED 1 +#define ARM_EXCEPTION_SOFTWARE 2 +#define ARM_EXCEPTION_PREF_ABORT 3 +#define ARM_EXCEPTION_DATA_ABORT 4 +#define ARM_EXCEPTION_IRQ 5 +#define ARM_EXCEPTION_FIQ 6 + +#endif /* __ARM_KVM_ASM_H__ */ -- cgit v1.2.3 From 342cd0ab0e6ca3fe7c88a78890352748b8e894a9 Mon Sep 17 00:00:00 2001 From: Christoffer Dall Date: Sun, 20 Jan 2013 18:28:06 -0500 Subject: KVM: ARM: Hypervisor initialization Sets up KVM code to handle all exceptions taken to Hyp mode. When the kernel is booted in Hyp mode, calling an hvc instruction with r0 pointing to the new vectors, the HVBAR is changed to the the vector pointers. This allows subsystems (like KVM here) to execute code in Hyp-mode with the MMU disabled. We initialize other Hyp-mode registers and enables the MMU for Hyp-mode from the id-mapped hyp initialization code. Afterwards, the HVBAR is changed to point to KVM Hyp vectors used to catch guest faults and to switch to Hyp mode to perform a world-switch into a KVM guest. Also provides memory mapping code to map required code pages, data structures, and I/O regions accessed in Hyp mode at the same virtual address as the host kernel virtual addresses, but which conforms to the architectural requirements for translations in Hyp mode. This interface is added in arch/arm/kvm/arm_mmu.c and comprises: - create_hyp_mappings(from, to); - create_hyp_io_mappings(from, to, phys_addr); - free_hyp_pmds(); Reviewed-by: Will Deacon Reviewed-by: Marcelo Tosatti Signed-off-by: Marc Zyngier Signed-off-by: Christoffer Dall --- arch/arm/include/asm/kvm_arm.h | 124 ++++++++++++++ arch/arm/include/asm/kvm_asm.h | 20 +++ arch/arm/include/asm/kvm_host.h | 1 + arch/arm/include/asm/kvm_mmu.h | 29 ++++ arch/arm/include/asm/pgtable-3level-hwdef.h | 4 + arch/arm/kvm/arm.c | 172 +++++++++++++++++++ arch/arm/kvm/init.S | 95 +++++++++++ arch/arm/kvm/interrupts.S | 62 +++++++ arch/arm/kvm/mmu.c | 248 ++++++++++++++++++++++++++++ 9 files changed, 755 insertions(+) create mode 100644 arch/arm/include/asm/kvm_mmu.h (limited to 'arch/arm/include/asm/kvm_asm.h') diff --git a/arch/arm/include/asm/kvm_arm.h b/arch/arm/include/asm/kvm_arm.h index dc678e193417..8875b3f605a7 100644 --- a/arch/arm/include/asm/kvm_arm.h +++ b/arch/arm/include/asm/kvm_arm.h @@ -21,4 +21,128 @@ #include +/* Hyp Configuration Register (HCR) bits */ +#define HCR_TGE (1 << 27) +#define HCR_TVM (1 << 26) +#define HCR_TTLB (1 << 25) +#define HCR_TPU (1 << 24) +#define HCR_TPC (1 << 23) +#define HCR_TSW (1 << 22) +#define HCR_TAC (1 << 21) +#define HCR_TIDCP (1 << 20) +#define HCR_TSC (1 << 19) +#define HCR_TID3 (1 << 18) +#define HCR_TID2 (1 << 17) +#define HCR_TID1 (1 << 16) +#define HCR_TID0 (1 << 15) +#define HCR_TWE (1 << 14) +#define HCR_TWI (1 << 13) +#define HCR_DC (1 << 12) +#define HCR_BSU (3 << 10) +#define HCR_BSU_IS (1 << 10) +#define HCR_FB (1 << 9) +#define HCR_VA (1 << 8) +#define HCR_VI (1 << 7) +#define HCR_VF (1 << 6) +#define HCR_AMO (1 << 5) +#define HCR_IMO (1 << 4) +#define HCR_FMO (1 << 3) +#define HCR_PTW (1 << 2) +#define HCR_SWIO (1 << 1) +#define HCR_VM 1 + +/* + * The bits we set in HCR: + * TAC: Trap ACTLR + * TSC: Trap SMC + * TSW: Trap cache operations by set/way + * TWI: Trap WFI + * TIDCP: Trap L2CTLR/L2ECTLR + * BSU_IS: Upgrade barriers to the inner shareable domain + * FB: Force broadcast of all maintainance operations + * AMO: Override CPSR.A and enable signaling with VA + * IMO: Override CPSR.I and enable signaling with VI + * FMO: Override CPSR.F and enable signaling with VF + * SWIO: Turn set/way invalidates into set/way clean+invalidate + */ +#define HCR_GUEST_MASK (HCR_TSC | HCR_TSW | HCR_TWI | HCR_VM | HCR_BSU_IS | \ + HCR_FB | HCR_TAC | HCR_AMO | HCR_IMO | HCR_FMO | \ + HCR_SWIO | HCR_TIDCP) + +/* Hyp System Control Register (HSCTLR) bits */ +#define HSCTLR_TE (1 << 30) +#define HSCTLR_EE (1 << 25) +#define HSCTLR_FI (1 << 21) +#define HSCTLR_WXN (1 << 19) +#define HSCTLR_I (1 << 12) +#define HSCTLR_C (1 << 2) +#define HSCTLR_A (1 << 1) +#define HSCTLR_M 1 +#define HSCTLR_MASK (HSCTLR_M | HSCTLR_A | HSCTLR_C | HSCTLR_I | \ + HSCTLR_WXN | HSCTLR_FI | HSCTLR_EE | HSCTLR_TE) + +/* TTBCR and HTCR Registers bits */ +#define TTBCR_EAE (1 << 31) +#define TTBCR_IMP (1 << 30) +#define TTBCR_SH1 (3 << 28) +#define TTBCR_ORGN1 (3 << 26) +#define TTBCR_IRGN1 (3 << 24) +#define TTBCR_EPD1 (1 << 23) +#define TTBCR_A1 (1 << 22) +#define TTBCR_T1SZ (3 << 16) +#define TTBCR_SH0 (3 << 12) +#define TTBCR_ORGN0 (3 << 10) +#define TTBCR_IRGN0 (3 << 8) +#define TTBCR_EPD0 (1 << 7) +#define TTBCR_T0SZ 3 +#define HTCR_MASK (TTBCR_T0SZ | TTBCR_IRGN0 | TTBCR_ORGN0 | TTBCR_SH0) + +/* Hyp Debug Configuration Register bits */ +#define HDCR_TDRA (1 << 11) +#define HDCR_TDOSA (1 << 10) +#define HDCR_TDA (1 << 9) +#define HDCR_TDE (1 << 8) +#define HDCR_HPME (1 << 7) +#define HDCR_TPM (1 << 6) +#define HDCR_TPMCR (1 << 5) +#define HDCR_HPMN_MASK (0x1F) + +/* + * The architecture supports 40-bit IPA as input to the 2nd stage translations + * and PTRS_PER_S2_PGD becomes 1024, because each entry covers 1GB of address + * space. + */ +#define KVM_PHYS_SHIFT (40) +#define KVM_PHYS_SIZE (1ULL << KVM_PHYS_SHIFT) +#define KVM_PHYS_MASK (KVM_PHYS_SIZE - 1ULL) +#define PTRS_PER_S2_PGD (1ULL << (KVM_PHYS_SHIFT - 30)) +#define S2_PGD_ORDER get_order(PTRS_PER_S2_PGD * sizeof(pgd_t)) +#define S2_PGD_SIZE (1 << S2_PGD_ORDER) + +/* Virtualization Translation Control Register (VTCR) bits */ +#define VTCR_SH0 (3 << 12) +#define VTCR_ORGN0 (3 << 10) +#define VTCR_IRGN0 (3 << 8) +#define VTCR_SL0 (3 << 6) +#define VTCR_S (1 << 4) +#define VTCR_T0SZ (0xf) +#define VTCR_MASK (VTCR_SH0 | VTCR_ORGN0 | VTCR_IRGN0 | VTCR_SL0 | \ + VTCR_S | VTCR_T0SZ) +#define VTCR_HTCR_SH (VTCR_SH0 | VTCR_ORGN0 | VTCR_IRGN0) +#define VTCR_SL_L2 (0 << 6) /* Starting-level: 2 */ +#define VTCR_SL_L1 (1 << 6) /* Starting-level: 1 */ +#define KVM_VTCR_SL0 VTCR_SL_L1 +/* stage-2 input address range defined as 2^(32-T0SZ) */ +#define KVM_T0SZ (32 - KVM_PHYS_SHIFT) +#define KVM_VTCR_T0SZ (KVM_T0SZ & VTCR_T0SZ) +#define KVM_VTCR_S ((KVM_VTCR_T0SZ << 1) & VTCR_S) + +/* Virtualization Translation Table Base Register (VTTBR) bits */ +#if KVM_VTCR_SL0 == VTCR_SL_L2 /* see ARM DDI 0406C: B4-1720 */ +#define VTTBR_X (14 - KVM_T0SZ) +#else +#define VTTBR_X (5 - KVM_T0SZ) +#endif + + #endif /* __ARM_KVM_ARM_H__ */ diff --git a/arch/arm/include/asm/kvm_asm.h b/arch/arm/include/asm/kvm_asm.h index f9993e5fb695..81324e2eb3f9 100644 --- a/arch/arm/include/asm/kvm_asm.h +++ b/arch/arm/include/asm/kvm_asm.h @@ -54,5 +54,25 @@ #define ARM_EXCEPTION_DATA_ABORT 4 #define ARM_EXCEPTION_IRQ 5 #define ARM_EXCEPTION_FIQ 6 +#define ARM_EXCEPTION_HVC 7 + +#ifndef __ASSEMBLY__ +struct kvm_vcpu; + +extern char __kvm_hyp_init[]; +extern char __kvm_hyp_init_end[]; + +extern char __kvm_hyp_exit[]; +extern char __kvm_hyp_exit_end[]; + +extern char __kvm_hyp_vector[]; + +extern char __kvm_hyp_code_start[]; +extern char __kvm_hyp_code_end[]; + +extern void __kvm_flush_vm_context(void); + +extern int __kvm_vcpu_run(struct kvm_vcpu *vcpu); +#endif #endif /* __ARM_KVM_ASM_H__ */ diff --git a/arch/arm/include/asm/kvm_host.h b/arch/arm/include/asm/kvm_host.h index 0d9938a20751..067ef2898c26 100644 --- a/arch/arm/include/asm/kvm_host.h +++ b/arch/arm/include/asm/kvm_host.h @@ -111,4 +111,5 @@ int kvm_arm_copy_reg_indices(struct kvm_vcpu *vcpu, u64 __user *indices); struct kvm_one_reg; int kvm_arm_get_reg(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg); int kvm_arm_set_reg(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg); +u64 kvm_call_hyp(void *hypfn, ...); #endif /* __ARM_KVM_HOST_H__ */ diff --git a/arch/arm/include/asm/kvm_mmu.h b/arch/arm/include/asm/kvm_mmu.h new file mode 100644 index 000000000000..e8679b317b0f --- /dev/null +++ b/arch/arm/include/asm/kvm_mmu.h @@ -0,0 +1,29 @@ +/* + * Copyright (C) 2012 - Virtual Open Systems and Columbia University + * Author: Christoffer Dall + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License, version 2, as + * published by the Free Software Foundation. + * + * This program 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 General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. + */ + +#ifndef __ARM_KVM_MMU_H__ +#define __ARM_KVM_MMU_H__ + +int create_hyp_mappings(void *from, void *to); +int create_hyp_io_mappings(void *from, void *to, phys_addr_t); +void free_hyp_pmds(void); + +phys_addr_t kvm_mmu_get_httbr(void); +int kvm_mmu_init(void); +void kvm_clear_hyp_idmap(void); +#endif /* __ARM_KVM_MMU_H__ */ diff --git a/arch/arm/include/asm/pgtable-3level-hwdef.h b/arch/arm/include/asm/pgtable-3level-hwdef.h index a2d404ed1ade..18f5cef82ad5 100644 --- a/arch/arm/include/asm/pgtable-3level-hwdef.h +++ b/arch/arm/include/asm/pgtable-3level-hwdef.h @@ -32,6 +32,9 @@ #define PMD_TYPE_SECT (_AT(pmdval_t, 1) << 0) #define PMD_BIT4 (_AT(pmdval_t, 0)) #define PMD_DOMAIN(x) (_AT(pmdval_t, 0)) +#define PMD_APTABLE_SHIFT (61) +#define PMD_APTABLE (_AT(pgdval_t, 3) << PGD_APTABLE_SHIFT) +#define PMD_PXNTABLE (_AT(pgdval_t, 1) << 59) /* * - section @@ -41,6 +44,7 @@ #define PMD_SECT_S (_AT(pmdval_t, 3) << 8) #define PMD_SECT_AF (_AT(pmdval_t, 1) << 10) #define PMD_SECT_nG (_AT(pmdval_t, 1) << 11) +#define PMD_SECT_PXN (_AT(pmdval_t, 1) << 53) #define PMD_SECT_XN (_AT(pmdval_t, 1) << 54) #define PMD_SECT_AP_WRITE (_AT(pmdval_t, 0)) #define PMD_SECT_AP_READ (_AT(pmdval_t, 0)) diff --git a/arch/arm/kvm/arm.c b/arch/arm/kvm/arm.c index d3506b4001aa..2c6b780e78a7 100644 --- a/arch/arm/kvm/arm.c +++ b/arch/arm/kvm/arm.c @@ -34,11 +34,21 @@ #include #include #include +#include +#include +#include +#include +#include #ifdef REQUIRES_VIRT __asm__(".arch_extension virt"); #endif +static DEFINE_PER_CPU(unsigned long, kvm_arm_hyp_stack_page); +static struct vfp_hard_struct __percpu *kvm_host_vfp_state; +static unsigned long hyp_default_vectors; + + int kvm_arch_hardware_enable(void *garbage) { return 0; @@ -331,9 +341,171 @@ long kvm_arch_vm_ioctl(struct file *filp, return -EINVAL; } +static void cpu_init_hyp_mode(void *vector) +{ + unsigned long long pgd_ptr; + unsigned long pgd_low, pgd_high; + unsigned long hyp_stack_ptr; + unsigned long stack_page; + unsigned long vector_ptr; + + /* Switch from the HYP stub to our own HYP init vector */ + __hyp_set_vectors((unsigned long)vector); + + pgd_ptr = (unsigned long long)kvm_mmu_get_httbr(); + pgd_low = (pgd_ptr & ((1ULL << 32) - 1)); + pgd_high = (pgd_ptr >> 32ULL); + stack_page = __get_cpu_var(kvm_arm_hyp_stack_page); + hyp_stack_ptr = stack_page + PAGE_SIZE; + vector_ptr = (unsigned long)__kvm_hyp_vector; + + /* + * Call initialization code, and switch to the full blown + * HYP code. The init code doesn't need to preserve these registers as + * r1-r3 and r12 are already callee save according to the AAPCS. + * Note that we slightly misuse the prototype by casing the pgd_low to + * a void *. + */ + kvm_call_hyp((void *)pgd_low, pgd_high, hyp_stack_ptr, vector_ptr); +} + +/** + * Inits Hyp-mode on all online CPUs + */ +static int init_hyp_mode(void) +{ + phys_addr_t init_phys_addr; + int cpu; + int err = 0; + + /* + * Allocate Hyp PGD and setup Hyp identity mapping + */ + err = kvm_mmu_init(); + if (err) + goto out_err; + + /* + * It is probably enough to obtain the default on one + * CPU. It's unlikely to be different on the others. + */ + hyp_default_vectors = __hyp_get_vectors(); + + /* + * Allocate stack pages for Hypervisor-mode + */ + for_each_possible_cpu(cpu) { + unsigned long stack_page; + + stack_page = __get_free_page(GFP_KERNEL); + if (!stack_page) { + err = -ENOMEM; + goto out_free_stack_pages; + } + + per_cpu(kvm_arm_hyp_stack_page, cpu) = stack_page; + } + + /* + * Execute the init code on each CPU. + * + * Note: The stack is not mapped yet, so don't do anything else than + * initializing the hypervisor mode on each CPU using a local stack + * space for temporary storage. + */ + init_phys_addr = virt_to_phys(__kvm_hyp_init); + for_each_online_cpu(cpu) { + smp_call_function_single(cpu, cpu_init_hyp_mode, + (void *)(long)init_phys_addr, 1); + } + + /* + * Unmap the identity mapping + */ + kvm_clear_hyp_idmap(); + + /* + * Map the Hyp-code called directly from the host + */ + err = create_hyp_mappings(__kvm_hyp_code_start, __kvm_hyp_code_end); + if (err) { + kvm_err("Cannot map world-switch code\n"); + goto out_free_mappings; + } + + /* + * Map the Hyp stack pages + */ + for_each_possible_cpu(cpu) { + char *stack_page = (char *)per_cpu(kvm_arm_hyp_stack_page, cpu); + err = create_hyp_mappings(stack_page, stack_page + PAGE_SIZE); + + if (err) { + kvm_err("Cannot map hyp stack\n"); + goto out_free_mappings; + } + } + + /* + * Map the host VFP structures + */ + kvm_host_vfp_state = alloc_percpu(struct vfp_hard_struct); + if (!kvm_host_vfp_state) { + err = -ENOMEM; + kvm_err("Cannot allocate host VFP state\n"); + goto out_free_mappings; + } + + for_each_possible_cpu(cpu) { + struct vfp_hard_struct *vfp; + + vfp = per_cpu_ptr(kvm_host_vfp_state, cpu); + err = create_hyp_mappings(vfp, vfp + 1); + + if (err) { + kvm_err("Cannot map host VFP state: %d\n", err); + goto out_free_vfp; + } + } + + kvm_info("Hyp mode initialized successfully\n"); + return 0; +out_free_vfp: + free_percpu(kvm_host_vfp_state); +out_free_mappings: + free_hyp_pmds(); +out_free_stack_pages: + for_each_possible_cpu(cpu) + free_page(per_cpu(kvm_arm_hyp_stack_page, cpu)); +out_err: + kvm_err("error initializing Hyp mode: %d\n", err); + return err; +} + +/** + * Initialize Hyp-mode and memory mappings on all CPUs. + */ int kvm_arch_init(void *opaque) { + int err; + + if (!is_hyp_mode_available()) { + kvm_err("HYP mode not available\n"); + return -ENODEV; + } + + if (kvm_target_cpu() < 0) { + kvm_err("Target CPU not supported!\n"); + return -ENODEV; + } + + err = init_hyp_mode(); + if (err) + goto out_err; + return 0; +out_err: + return err; } /* NOP: Compiling as a module not supported */ diff --git a/arch/arm/kvm/init.S b/arch/arm/kvm/init.S index 1dc8926e26d2..9f37a79b880b 100644 --- a/arch/arm/kvm/init.S +++ b/arch/arm/kvm/init.S @@ -15,5 +15,100 @@ * along with this program; if not, write to the Free Software * Foundation, 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#include +#include #include #include +#include + +/******************************************************************** + * Hypervisor initialization + * - should be called with: + * r0,r1 = Hypervisor pgd pointer + * r2 = top of Hyp stack (kernel VA) + * r3 = pointer to hyp vectors + */ + + .text + .pushsection .hyp.idmap.text,"ax" + .align 5 +__kvm_hyp_init: + .globl __kvm_hyp_init + + @ Hyp-mode exception vector + W(b) . + W(b) . + W(b) . + W(b) . + W(b) . + W(b) __do_hyp_init + W(b) . + W(b) . + +__do_hyp_init: + @ Set the HTTBR to point to the hypervisor PGD pointer passed + mcrr p15, 4, r0, r1, c2 + + @ Set the HTCR and VTCR to the same shareability and cacheability + @ settings as the non-secure TTBCR and with T0SZ == 0. + mrc p15, 4, r0, c2, c0, 2 @ HTCR + ldr r12, =HTCR_MASK + bic r0, r0, r12 + mrc p15, 0, r1, c2, c0, 2 @ TTBCR + and r1, r1, #(HTCR_MASK & ~TTBCR_T0SZ) + orr r0, r0, r1 + mcr p15, 4, r0, c2, c0, 2 @ HTCR + + mrc p15, 4, r1, c2, c1, 2 @ VTCR + ldr r12, =VTCR_MASK + bic r1, r1, r12 + bic r0, r0, #(~VTCR_HTCR_SH) @ clear non-reusable HTCR bits + orr r1, r0, r1 + orr r1, r1, #(KVM_VTCR_SL0 | KVM_VTCR_T0SZ | KVM_VTCR_S) + mcr p15, 4, r1, c2, c1, 2 @ VTCR + + @ Use the same memory attributes for hyp. accesses as the kernel + @ (copy MAIRx ro HMAIRx). + mrc p15, 0, r0, c10, c2, 0 + mcr p15, 4, r0, c10, c2, 0 + mrc p15, 0, r0, c10, c2, 1 + mcr p15, 4, r0, c10, c2, 1 + + @ Set the HSCTLR to: + @ - ARM/THUMB exceptions: Kernel config (Thumb-2 kernel) + @ - Endianness: Kernel config + @ - Fast Interrupt Features: Kernel config + @ - Write permission implies XN: disabled + @ - Instruction cache: enabled + @ - Data/Unified cache: enabled + @ - Memory alignment checks: enabled + @ - MMU: enabled (this code must be run from an identity mapping) + mrc p15, 4, r0, c1, c0, 0 @ HSCR + ldr r12, =HSCTLR_MASK + bic r0, r0, r12 + mrc p15, 0, r1, c1, c0, 0 @ SCTLR + ldr r12, =(HSCTLR_EE | HSCTLR_FI | HSCTLR_I | HSCTLR_C) + and r1, r1, r12 + ARM( ldr r12, =(HSCTLR_M | HSCTLR_A) ) + THUMB( ldr r12, =(HSCTLR_M | HSCTLR_A | HSCTLR_TE) ) + orr r1, r1, r12 + orr r0, r0, r1 + isb + mcr p15, 4, r0, c1, c0, 0 @ HSCR + isb + + @ Set stack pointer and return to the kernel + mov sp, r2 + + @ Set HVBAR to point to the HYP vectors + mcr p15, 4, r3, c12, c0, 0 @ HVBAR + + eret + + .ltorg + + .globl __kvm_hyp_init_end +__kvm_hyp_init_end: + + .popsection diff --git a/arch/arm/kvm/interrupts.S b/arch/arm/kvm/interrupts.S index 1dc8926e26d2..d10a8075409a 100644 --- a/arch/arm/kvm/interrupts.S +++ b/arch/arm/kvm/interrupts.S @@ -15,5 +15,67 @@ * along with this program; if not, write to the Free Software * Foundation, 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#include +#include +#include +#include #include #include +#include + + .text + +__kvm_hyp_code_start: + .globl __kvm_hyp_code_start + +/******************************************************************** + * Flush per-VMID TLBs + */ +ENTRY(__kvm_flush_vm_context) + bx lr +ENDPROC(__kvm_flush_vm_context) + +/******************************************************************** + * Hypervisor world-switch code + */ +ENTRY(__kvm_vcpu_run) + bx lr + +/******************************************************************** + * Call function in Hyp mode + * + * + * u64 kvm_call_hyp(void *hypfn, ...); + * + * This is not really a variadic function in the classic C-way and care must + * be taken when calling this to ensure parameters are passed in registers + * only, since the stack will change between the caller and the callee. + * + * Call the function with the first argument containing a pointer to the + * function you wish to call in Hyp mode, and subsequent arguments will be + * passed as r0, r1, and r2 (a maximum of 3 arguments in addition to the + * function pointer can be passed). The function being called must be mapped + * in Hyp mode (see init_hyp_mode in arch/arm/kvm/arm.c). Return values are + * passed in r0 and r1. + * + * The calling convention follows the standard AAPCS: + * r0 - r3: caller save + * r12: caller save + * rest: callee save + */ +ENTRY(kvm_call_hyp) + hvc #0 + bx lr + +/******************************************************************** + * Hypervisor exception vector and handlers + */ + + .align 5 +__kvm_hyp_vector: + .globl __kvm_hyp_vector + nop + +__kvm_hyp_code_end: + .globl __kvm_hyp_code_end diff --git a/arch/arm/kvm/mmu.c b/arch/arm/kvm/mmu.c index 10ed4643269f..4decdb618019 100644 --- a/arch/arm/kvm/mmu.c +++ b/arch/arm/kvm/mmu.c @@ -15,3 +15,251 @@ * along with this program; if not, write to the Free Software * Foundation, 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ + +#include +#include +#include +#include +#include +#include +#include +#include + +extern char __hyp_idmap_text_start[], __hyp_idmap_text_end[]; + +static DEFINE_MUTEX(kvm_hyp_pgd_mutex); + +static void kvm_set_pte(pte_t *pte, pte_t new_pte) +{ + pte_val(*pte) = new_pte; + /* + * flush_pmd_entry just takes a void pointer and cleans the necessary + * cache entries, so we can reuse the function for ptes. + */ + flush_pmd_entry(pte); +} + +static void free_ptes(pmd_t *pmd, unsigned long addr) +{ + pte_t *pte; + unsigned int i; + + for (i = 0; i < PTRS_PER_PMD; i++, addr += PMD_SIZE) { + if (!pmd_none(*pmd) && pmd_table(*pmd)) { + pte = pte_offset_kernel(pmd, addr); + pte_free_kernel(NULL, pte); + } + pmd++; + } +} + +/** + * free_hyp_pmds - free a Hyp-mode level-2 tables and child level-3 tables + * + * Assumes this is a page table used strictly in Hyp-mode and therefore contains + * only mappings in the kernel memory area, which is above PAGE_OFFSET. + */ +void free_hyp_pmds(void) +{ + pgd_t *pgd; + pud_t *pud; + pmd_t *pmd; + unsigned long addr; + + mutex_lock(&kvm_hyp_pgd_mutex); + for (addr = PAGE_OFFSET; addr != 0; addr += PGDIR_SIZE) { + pgd = hyp_pgd + pgd_index(addr); + pud = pud_offset(pgd, addr); + + if (pud_none(*pud)) + continue; + BUG_ON(pud_bad(*pud)); + + pmd = pmd_offset(pud, addr); + free_ptes(pmd, addr); + pmd_free(NULL, pmd); + pud_clear(pud); + } + mutex_unlock(&kvm_hyp_pgd_mutex); +} + +static void create_hyp_pte_mappings(pmd_t *pmd, unsigned long start, + unsigned long end) +{ + pte_t *pte; + unsigned long addr; + struct page *page; + + for (addr = start & PAGE_MASK; addr < end; addr += PAGE_SIZE) { + pte = pte_offset_kernel(pmd, addr); + BUG_ON(!virt_addr_valid(addr)); + page = virt_to_page(addr); + kvm_set_pte(pte, mk_pte(page, PAGE_HYP)); + } +} + +static void create_hyp_io_pte_mappings(pmd_t *pmd, unsigned long start, + unsigned long end, + unsigned long *pfn_base) +{ + pte_t *pte; + unsigned long addr; + + for (addr = start & PAGE_MASK; addr < end; addr += PAGE_SIZE) { + pte = pte_offset_kernel(pmd, addr); + BUG_ON(pfn_valid(*pfn_base)); + kvm_set_pte(pte, pfn_pte(*pfn_base, PAGE_HYP_DEVICE)); + (*pfn_base)++; + } +} + +static int create_hyp_pmd_mappings(pud_t *pud, unsigned long start, + unsigned long end, unsigned long *pfn_base) +{ + pmd_t *pmd; + pte_t *pte; + unsigned long addr, next; + + for (addr = start; addr < end; addr = next) { + pmd = pmd_offset(pud, addr); + + BUG_ON(pmd_sect(*pmd)); + + if (pmd_none(*pmd)) { + pte = pte_alloc_one_kernel(NULL, addr); + if (!pte) { + kvm_err("Cannot allocate Hyp pte\n"); + return -ENOMEM; + } + pmd_populate_kernel(NULL, pmd, pte); + } + + next = pmd_addr_end(addr, end); + + /* + * If pfn_base is NULL, we map kernel pages into HYP with the + * virtual address. Otherwise, this is considered an I/O + * mapping and we map the physical region starting at + * *pfn_base to [start, end[. + */ + if (!pfn_base) + create_hyp_pte_mappings(pmd, addr, next); + else + create_hyp_io_pte_mappings(pmd, addr, next, pfn_base); + } + + return 0; +} + +static int __create_hyp_mappings(void *from, void *to, unsigned long *pfn_base) +{ + unsigned long start = (unsigned long)from; + unsigned long end = (unsigned long)to; + pgd_t *pgd; + pud_t *pud; + pmd_t *pmd; + unsigned long addr, next; + int err = 0; + + BUG_ON(start > end); + if (start < PAGE_OFFSET) + return -EINVAL; + + mutex_lock(&kvm_hyp_pgd_mutex); + for (addr = start; addr < end; addr = next) { + pgd = hyp_pgd + pgd_index(addr); + pud = pud_offset(pgd, addr); + + if (pud_none_or_clear_bad(pud)) { + pmd = pmd_alloc_one(NULL, addr); + if (!pmd) { + kvm_err("Cannot allocate Hyp pmd\n"); + err = -ENOMEM; + goto out; + } + pud_populate(NULL, pud, pmd); + } + + next = pgd_addr_end(addr, end); + err = create_hyp_pmd_mappings(pud, addr, next, pfn_base); + if (err) + goto out; + } +out: + mutex_unlock(&kvm_hyp_pgd_mutex); + return err; +} + +/** + * create_hyp_mappings - map a kernel virtual address range in Hyp mode + * @from: The virtual kernel start address of the range + * @to: The virtual kernel end address of the range (exclusive) + * + * The same virtual address as the kernel virtual address is also used in + * Hyp-mode mapping to the same underlying physical pages. + * + * Note: Wrapping around zero in the "to" address is not supported. + */ +int create_hyp_mappings(void *from, void *to) +{ + return __create_hyp_mappings(from, to, NULL); +} + +/** + * create_hyp_io_mappings - map a physical IO range in Hyp mode + * @from: The virtual HYP start address of the range + * @to: The virtual HYP end address of the range (exclusive) + * @addr: The physical start address which gets mapped + */ +int create_hyp_io_mappings(void *from, void *to, phys_addr_t addr) +{ + unsigned long pfn = __phys_to_pfn(addr); + return __create_hyp_mappings(from, to, &pfn); +} + +int kvm_handle_guest_abort(struct kvm_vcpu *vcpu, struct kvm_run *run) +{ + return -EINVAL; +} + +phys_addr_t kvm_mmu_get_httbr(void) +{ + VM_BUG_ON(!virt_addr_valid(hyp_pgd)); + return virt_to_phys(hyp_pgd); +} + +int kvm_mmu_init(void) +{ + return hyp_pgd ? 0 : -ENOMEM; +} + +/** + * kvm_clear_idmap - remove all idmaps from the hyp pgd + * + * Free the underlying pmds for all pgds in range and clear the pgds (but + * don't free them) afterwards. + */ +void kvm_clear_hyp_idmap(void) +{ + unsigned long addr, end; + unsigned long next; + pgd_t *pgd = hyp_pgd; + pud_t *pud; + pmd_t *pmd; + + addr = virt_to_phys(__hyp_idmap_text_start); + end = virt_to_phys(__hyp_idmap_text_end); + + pgd += pgd_index(addr); + do { + next = pgd_addr_end(addr, end); + if (pgd_none_or_clear_bad(pgd)) + continue; + pud = pud_offset(pgd, addr); + pmd = pmd_offset(pud, addr); + + pud_clear(pud); + clean_pmd_entry(pmd); + pmd_free(NULL, (pmd_t *)((unsigned long)pmd & PAGE_MASK)); + } while (pgd++, addr = next, addr < end); +} -- cgit v1.2.3 From d5d8184d35c990b1324d9b30bcd0e4e8aa08f56d Mon Sep 17 00:00:00 2001 From: Christoffer Dall Date: Sun, 20 Jan 2013 18:28:07 -0500 Subject: KVM: ARM: Memory virtualization setup This commit introduces the framework for guest memory management through the use of 2nd stage translation. Each VM has a pointer to a level-1 table (the pgd field in struct kvm_arch) which is used for the 2nd stage translations. Entries are added when handling guest faults (later patch) and the table itself can be allocated and freed through the following functions implemented in arch/arm/kvm/arm_mmu.c: - kvm_alloc_stage2_pgd(struct kvm *kvm); - kvm_free_stage2_pgd(struct kvm *kvm); Each entry in TLBs and caches are tagged with a VMID identifier in addition to ASIDs. The VMIDs are assigned consecutively to VMs in the order that VMs are executed, and caches and tlbs are invalidated when the VMID space has been used to allow for more than 255 simultaenously running guests. The 2nd stage pgd is allocated in kvm_arch_init_vm(). The table is freed in kvm_arch_destroy_vm(). Both functions are called from the main KVM code. We pre-allocate page table memory to be able to synchronize using a spinlock and be called under rcu_read_lock from the MMU notifiers. We steal the mmu_memory_cache implementation from x86 and adapt for our specific usage. We support MMU notifiers (thanks to Marc Zyngier) through kvm_unmap_hva and kvm_set_spte_hva. Finally, define kvm_phys_addr_ioremap() to map a device at a guest IPA, which is used by VGIC support to map the virtual CPU interface registers to the guest. This support is added by Marc Zyngier. Reviewed-by: Will Deacon Reviewed-by: Marcelo Tosatti Signed-off-by: Marc Zyngier Signed-off-by: Christoffer Dall --- arch/arm/include/asm/kvm_asm.h | 2 + arch/arm/include/asm/kvm_host.h | 18 ++ arch/arm/include/asm/kvm_mmu.h | 9 + arch/arm/kvm/Kconfig | 1 + arch/arm/kvm/arm.c | 37 +++- arch/arm/kvm/interrupts.S | 7 + arch/arm/kvm/mmu.c | 370 +++++++++++++++++++++++++++++++++++++++- arch/arm/kvm/trace.h | 46 +++++ 8 files changed, 488 insertions(+), 2 deletions(-) (limited to 'arch/arm/include/asm/kvm_asm.h') diff --git a/arch/arm/include/asm/kvm_asm.h b/arch/arm/include/asm/kvm_asm.h index 81324e2eb3f9..f6652f6c5d84 100644 --- a/arch/arm/include/asm/kvm_asm.h +++ b/arch/arm/include/asm/kvm_asm.h @@ -57,6 +57,7 @@ #define ARM_EXCEPTION_HVC 7 #ifndef __ASSEMBLY__ +struct kvm; struct kvm_vcpu; extern char __kvm_hyp_init[]; @@ -71,6 +72,7 @@ extern char __kvm_hyp_code_start[]; extern char __kvm_hyp_code_end[]; extern void __kvm_flush_vm_context(void); +extern void __kvm_tlb_flush_vmid(struct kvm *kvm); extern int __kvm_vcpu_run(struct kvm_vcpu *vcpu); #endif diff --git a/arch/arm/include/asm/kvm_host.h b/arch/arm/include/asm/kvm_host.h index 067ef2898c26..3636c7ea4eb2 100644 --- a/arch/arm/include/asm/kvm_host.h +++ b/arch/arm/include/asm/kvm_host.h @@ -112,4 +112,22 @@ struct kvm_one_reg; int kvm_arm_get_reg(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg); int kvm_arm_set_reg(struct kvm_vcpu *vcpu, const struct kvm_one_reg *reg); u64 kvm_call_hyp(void *hypfn, ...); + +#define KVM_ARCH_WANT_MMU_NOTIFIER +struct kvm; +int kvm_unmap_hva(struct kvm *kvm, unsigned long hva); +int kvm_unmap_hva_range(struct kvm *kvm, + unsigned long start, unsigned long end); +void kvm_set_spte_hva(struct kvm *kvm, unsigned long hva, pte_t pte); + +/* We do not have shadow page tables, hence the empty hooks */ +static inline int kvm_age_hva(struct kvm *kvm, unsigned long hva) +{ + return 0; +} + +static inline int kvm_test_age_hva(struct kvm *kvm, unsigned long hva) +{ + return 0; +} #endif /* __ARM_KVM_HOST_H__ */ diff --git a/arch/arm/include/asm/kvm_mmu.h b/arch/arm/include/asm/kvm_mmu.h index e8679b317b0f..499e7b0925ff 100644 --- a/arch/arm/include/asm/kvm_mmu.h +++ b/arch/arm/include/asm/kvm_mmu.h @@ -23,6 +23,15 @@ int create_hyp_mappings(void *from, void *to); int create_hyp_io_mappings(void *from, void *to, phys_addr_t); void free_hyp_pmds(void); +int kvm_alloc_stage2_pgd(struct kvm *kvm); +void kvm_free_stage2_pgd(struct kvm *kvm); +int kvm_phys_addr_ioremap(struct kvm *kvm, phys_addr_t guest_ipa, + phys_addr_t pa, unsigned long size); + +int kvm_handle_guest_abort(struct kvm_vcpu *vcpu, struct kvm_run *run); + +void kvm_mmu_free_memory_caches(struct kvm_vcpu *vcpu); + phys_addr_t kvm_mmu_get_httbr(void); int kvm_mmu_init(void); void kvm_clear_hyp_idmap(void); diff --git a/arch/arm/kvm/Kconfig b/arch/arm/kvm/Kconfig index 4a01b6fbf380..05227cb57a7b 100644 --- a/arch/arm/kvm/Kconfig +++ b/arch/arm/kvm/Kconfig @@ -36,6 +36,7 @@ config KVM_ARM_HOST bool "KVM host support for ARM cpus." depends on KVM depends on MMU + select MMU_NOTIFIER ---help--- Provides host support for ARM processors. diff --git a/arch/arm/kvm/arm.c b/arch/arm/kvm/arm.c index 2c6b780e78a7..d810afb6cb84 100644 --- a/arch/arm/kvm/arm.c +++ b/arch/arm/kvm/arm.c @@ -81,12 +81,33 @@ void kvm_arch_sync_events(struct kvm *kvm) { } +/** + * kvm_arch_init_vm - initializes a VM data structure + * @kvm: pointer to the KVM struct + */ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type) { + int ret = 0; + if (type) return -EINVAL; - return 0; + ret = kvm_alloc_stage2_pgd(kvm); + if (ret) + goto out_fail_alloc; + + ret = create_hyp_mappings(kvm, kvm + 1); + if (ret) + goto out_free_stage2_pgd; + + /* Mark the initial VMID generation invalid */ + kvm->arch.vmid_gen = 0; + + return ret; +out_free_stage2_pgd: + kvm_free_stage2_pgd(kvm); +out_fail_alloc: + return ret; } int kvm_arch_vcpu_fault(struct kvm_vcpu *vcpu, struct vm_fault *vmf) @@ -104,10 +125,16 @@ int kvm_arch_create_memslot(struct kvm_memory_slot *slot, unsigned long npages) return 0; } +/** + * kvm_arch_destroy_vm - destroy the VM data structure + * @kvm: pointer to the KVM struct + */ void kvm_arch_destroy_vm(struct kvm *kvm) { int i; + kvm_free_stage2_pgd(kvm); + for (i = 0; i < KVM_MAX_VCPUS; ++i) { if (kvm->vcpus[i]) { kvm_arch_vcpu_free(kvm->vcpus[i]); @@ -196,7 +223,13 @@ struct kvm_vcpu *kvm_arch_vcpu_create(struct kvm *kvm, unsigned int id) if (err) goto free_vcpu; + err = create_hyp_mappings(vcpu, vcpu + 1); + if (err) + goto vcpu_uninit; + return vcpu; +vcpu_uninit: + kvm_vcpu_uninit(vcpu); free_vcpu: kmem_cache_free(kvm_vcpu_cache, vcpu); out: @@ -210,6 +243,8 @@ int kvm_arch_vcpu_postcreate(struct kvm_vcpu *vcpu) void kvm_arch_vcpu_free(struct kvm_vcpu *vcpu) { + kvm_mmu_free_memory_caches(vcpu); + kmem_cache_free(kvm_vcpu_cache, vcpu); } void kvm_arch_vcpu_destroy(struct kvm_vcpu *vcpu) diff --git a/arch/arm/kvm/interrupts.S b/arch/arm/kvm/interrupts.S index d10a8075409a..f701aff31e44 100644 --- a/arch/arm/kvm/interrupts.S +++ b/arch/arm/kvm/interrupts.S @@ -32,6 +32,13 @@ __kvm_hyp_code_start: /******************************************************************** * Flush per-VMID TLBs */ +ENTRY(__kvm_tlb_flush_vmid) + bx lr +ENDPROC(__kvm_tlb_flush_vmid) + +/******************************************************************** + * Flush TLBs and instruction caches of current CPU for all VMIDs + */ ENTRY(__kvm_flush_vm_context) bx lr ENDPROC(__kvm_flush_vm_context) diff --git a/arch/arm/kvm/mmu.c b/arch/arm/kvm/mmu.c index 4decdb618019..4347d68f052f 100644 --- a/arch/arm/kvm/mmu.c +++ b/arch/arm/kvm/mmu.c @@ -23,12 +23,21 @@ #include #include #include +#include #include +#include + +#include "trace.h" extern char __hyp_idmap_text_start[], __hyp_idmap_text_end[]; static DEFINE_MUTEX(kvm_hyp_pgd_mutex); +static void kvm_tlb_flush_vmid(struct kvm *kvm) +{ + kvm_call_hyp(__kvm_tlb_flush_vmid, kvm); +} + static void kvm_set_pte(pte_t *pte, pte_t new_pte) { pte_val(*pte) = new_pte; @@ -39,6 +48,38 @@ static void kvm_set_pte(pte_t *pte, pte_t new_pte) flush_pmd_entry(pte); } +static int mmu_topup_memory_cache(struct kvm_mmu_memory_cache *cache, + int min, int max) +{ + void *page; + + BUG_ON(max > KVM_NR_MEM_OBJS); + if (cache->nobjs >= min) + return 0; + while (cache->nobjs < max) { + page = (void *)__get_free_page(PGALLOC_GFP); + if (!page) + return -ENOMEM; + cache->objects[cache->nobjs++] = page; + } + return 0; +} + +static void mmu_free_memory_cache(struct kvm_mmu_memory_cache *mc) +{ + while (mc->nobjs) + free_page((unsigned long)mc->objects[--mc->nobjs]); +} + +static void *mmu_memory_cache_alloc(struct kvm_mmu_memory_cache *mc) +{ + void *p; + + BUG_ON(!mc || !mc->nobjs); + p = mc->objects[--mc->nobjs]; + return p; +} + static void free_ptes(pmd_t *pmd, unsigned long addr) { pte_t *pte; @@ -217,11 +258,333 @@ int create_hyp_io_mappings(void *from, void *to, phys_addr_t addr) return __create_hyp_mappings(from, to, &pfn); } +/** + * kvm_alloc_stage2_pgd - allocate level-1 table for stage-2 translation. + * @kvm: The KVM struct pointer for the VM. + * + * Allocates the 1st level table only of size defined by S2_PGD_ORDER (can + * support either full 40-bit input addresses or limited to 32-bit input + * addresses). Clears the allocated pages. + * + * Note we don't need locking here as this is only called when the VM is + * created, which can only be done once. + */ +int kvm_alloc_stage2_pgd(struct kvm *kvm) +{ + pgd_t *pgd; + + if (kvm->arch.pgd != NULL) { + kvm_err("kvm_arch already initialized?\n"); + return -EINVAL; + } + + pgd = (pgd_t *)__get_free_pages(GFP_KERNEL, S2_PGD_ORDER); + if (!pgd) + return -ENOMEM; + + /* stage-2 pgd must be aligned to its size */ + VM_BUG_ON((unsigned long)pgd & (S2_PGD_SIZE - 1)); + + memset(pgd, 0, PTRS_PER_S2_PGD * sizeof(pgd_t)); + clean_dcache_area(pgd, PTRS_PER_S2_PGD * sizeof(pgd_t)); + kvm->arch.pgd = pgd; + + return 0; +} + +static void clear_pud_entry(pud_t *pud) +{ + pmd_t *pmd_table = pmd_offset(pud, 0); + pud_clear(pud); + pmd_free(NULL, pmd_table); + put_page(virt_to_page(pud)); +} + +static void clear_pmd_entry(pmd_t *pmd) +{ + pte_t *pte_table = pte_offset_kernel(pmd, 0); + pmd_clear(pmd); + pte_free_kernel(NULL, pte_table); + put_page(virt_to_page(pmd)); +} + +static bool pmd_empty(pmd_t *pmd) +{ + struct page *pmd_page = virt_to_page(pmd); + return page_count(pmd_page) == 1; +} + +static void clear_pte_entry(pte_t *pte) +{ + if (pte_present(*pte)) { + kvm_set_pte(pte, __pte(0)); + put_page(virt_to_page(pte)); + } +} + +static bool pte_empty(pte_t *pte) +{ + struct page *pte_page = virt_to_page(pte); + return page_count(pte_page) == 1; +} + +/** + * unmap_stage2_range -- Clear stage2 page table entries to unmap a range + * @kvm: The VM pointer + * @start: The intermediate physical base address of the range to unmap + * @size: The size of the area to unmap + * + * Clear a range of stage-2 mappings, lowering the various ref-counts. Must + * be called while holding mmu_lock (unless for freeing the stage2 pgd before + * destroying the VM), otherwise another faulting VCPU may come in and mess + * with things behind our backs. + */ +static void unmap_stage2_range(struct kvm *kvm, phys_addr_t start, u64 size) +{ + pgd_t *pgd; + pud_t *pud; + pmd_t *pmd; + pte_t *pte; + phys_addr_t addr = start, end = start + size; + u64 range; + + while (addr < end) { + pgd = kvm->arch.pgd + pgd_index(addr); + pud = pud_offset(pgd, addr); + if (pud_none(*pud)) { + addr += PUD_SIZE; + continue; + } + + pmd = pmd_offset(pud, addr); + if (pmd_none(*pmd)) { + addr += PMD_SIZE; + continue; + } + + pte = pte_offset_kernel(pmd, addr); + clear_pte_entry(pte); + range = PAGE_SIZE; + + /* If we emptied the pte, walk back up the ladder */ + if (pte_empty(pte)) { + clear_pmd_entry(pmd); + range = PMD_SIZE; + if (pmd_empty(pmd)) { + clear_pud_entry(pud); + range = PUD_SIZE; + } + } + + addr += range; + } +} + +/** + * kvm_free_stage2_pgd - free all stage-2 tables + * @kvm: The KVM struct pointer for the VM. + * + * Walks the level-1 page table pointed to by kvm->arch.pgd and frees all + * underlying level-2 and level-3 tables before freeing the actual level-1 table + * and setting the struct pointer to NULL. + * + * Note we don't need locking here as this is only called when the VM is + * destroyed, which can only be done once. + */ +void kvm_free_stage2_pgd(struct kvm *kvm) +{ + if (kvm->arch.pgd == NULL) + return; + + unmap_stage2_range(kvm, 0, KVM_PHYS_SIZE); + free_pages((unsigned long)kvm->arch.pgd, S2_PGD_ORDER); + kvm->arch.pgd = NULL; +} + + +static int stage2_set_pte(struct kvm *kvm, struct kvm_mmu_memory_cache *cache, + phys_addr_t addr, const pte_t *new_pte, bool iomap) +{ + pgd_t *pgd; + pud_t *pud; + pmd_t *pmd; + pte_t *pte, old_pte; + + /* Create 2nd stage page table mapping - Level 1 */ + pgd = kvm->arch.pgd + pgd_index(addr); + pud = pud_offset(pgd, addr); + if (pud_none(*pud)) { + if (!cache) + return 0; /* ignore calls from kvm_set_spte_hva */ + pmd = mmu_memory_cache_alloc(cache); + pud_populate(NULL, pud, pmd); + pmd += pmd_index(addr); + get_page(virt_to_page(pud)); + } else + pmd = pmd_offset(pud, addr); + + /* Create 2nd stage page table mapping - Level 2 */ + if (pmd_none(*pmd)) { + if (!cache) + return 0; /* ignore calls from kvm_set_spte_hva */ + pte = mmu_memory_cache_alloc(cache); + clean_pte_table(pte); + pmd_populate_kernel(NULL, pmd, pte); + pte += pte_index(addr); + get_page(virt_to_page(pmd)); + } else + pte = pte_offset_kernel(pmd, addr); + + if (iomap && pte_present(*pte)) + return -EFAULT; + + /* Create 2nd stage page table mapping - Level 3 */ + old_pte = *pte; + kvm_set_pte(pte, *new_pte); + if (pte_present(old_pte)) + kvm_tlb_flush_vmid(kvm); + else + get_page(virt_to_page(pte)); + + return 0; +} + +/** + * kvm_phys_addr_ioremap - map a device range to guest IPA + * + * @kvm: The KVM pointer + * @guest_ipa: The IPA at which to insert the mapping + * @pa: The physical address of the device + * @size: The size of the mapping + */ +int kvm_phys_addr_ioremap(struct kvm *kvm, phys_addr_t guest_ipa, + phys_addr_t pa, unsigned long size) +{ + phys_addr_t addr, end; + int ret = 0; + unsigned long pfn; + struct kvm_mmu_memory_cache cache = { 0, }; + + end = (guest_ipa + size + PAGE_SIZE - 1) & PAGE_MASK; + pfn = __phys_to_pfn(pa); + + for (addr = guest_ipa; addr < end; addr += PAGE_SIZE) { + pte_t pte = pfn_pte(pfn, PAGE_S2_DEVICE | L_PTE_S2_RDWR); + + ret = mmu_topup_memory_cache(&cache, 2, 2); + if (ret) + goto out; + spin_lock(&kvm->mmu_lock); + ret = stage2_set_pte(kvm, &cache, addr, &pte, true); + spin_unlock(&kvm->mmu_lock); + if (ret) + goto out; + + pfn++; + } + +out: + mmu_free_memory_cache(&cache); + return ret; +} + int kvm_handle_guest_abort(struct kvm_vcpu *vcpu, struct kvm_run *run) { return -EINVAL; } +static void handle_hva_to_gpa(struct kvm *kvm, + unsigned long start, + unsigned long end, + void (*handler)(struct kvm *kvm, + gpa_t gpa, void *data), + void *data) +{ + struct kvm_memslots *slots; + struct kvm_memory_slot *memslot; + + slots = kvm_memslots(kvm); + + /* we only care about the pages that the guest sees */ + kvm_for_each_memslot(memslot, slots) { + unsigned long hva_start, hva_end; + gfn_t gfn, gfn_end; + + hva_start = max(start, memslot->userspace_addr); + hva_end = min(end, memslot->userspace_addr + + (memslot->npages << PAGE_SHIFT)); + if (hva_start >= hva_end) + continue; + + /* + * {gfn(page) | page intersects with [hva_start, hva_end)} = + * {gfn_start, gfn_start+1, ..., gfn_end-1}. + */ + gfn = hva_to_gfn_memslot(hva_start, memslot); + gfn_end = hva_to_gfn_memslot(hva_end + PAGE_SIZE - 1, memslot); + + for (; gfn < gfn_end; ++gfn) { + gpa_t gpa = gfn << PAGE_SHIFT; + handler(kvm, gpa, data); + } + } +} + +static void kvm_unmap_hva_handler(struct kvm *kvm, gpa_t gpa, void *data) +{ + unmap_stage2_range(kvm, gpa, PAGE_SIZE); + kvm_tlb_flush_vmid(kvm); +} + +int kvm_unmap_hva(struct kvm *kvm, unsigned long hva) +{ + unsigned long end = hva + PAGE_SIZE; + + if (!kvm->arch.pgd) + return 0; + + trace_kvm_unmap_hva(hva); + handle_hva_to_gpa(kvm, hva, end, &kvm_unmap_hva_handler, NULL); + return 0; +} + +int kvm_unmap_hva_range(struct kvm *kvm, + unsigned long start, unsigned long end) +{ + if (!kvm->arch.pgd) + return 0; + + trace_kvm_unmap_hva_range(start, end); + handle_hva_to_gpa(kvm, start, end, &kvm_unmap_hva_handler, NULL); + return 0; +} + +static void kvm_set_spte_handler(struct kvm *kvm, gpa_t gpa, void *data) +{ + pte_t *pte = (pte_t *)data; + + stage2_set_pte(kvm, NULL, gpa, pte, false); +} + + +void kvm_set_spte_hva(struct kvm *kvm, unsigned long hva, pte_t pte) +{ + unsigned long end = hva + PAGE_SIZE; + pte_t stage2_pte; + + if (!kvm->arch.pgd) + return; + + trace_kvm_set_spte_hva(hva); + stage2_pte = pfn_pte(pte_pfn(pte), PAGE_S2); + handle_hva_to_gpa(kvm, hva, end, &kvm_set_spte_handler, &stage2_pte); +} + +void kvm_mmu_free_memory_caches(struct kvm_vcpu *vcpu) +{ + mmu_free_memory_cache(&vcpu->arch.mmu_page_cache); +} + phys_addr_t kvm_mmu_get_httbr(void) { VM_BUG_ON(!virt_addr_valid(hyp_pgd)); @@ -230,7 +593,12 @@ phys_addr_t kvm_mmu_get_httbr(void) int kvm_mmu_init(void) { - return hyp_pgd ? 0 : -ENOMEM; + if (!hyp_pgd) { + kvm_err("Hyp mode PGD not allocated\n"); + return -ENOMEM; + } + + return 0; } /** diff --git a/arch/arm/kvm/trace.h b/arch/arm/kvm/trace.h index f8869c19c0a3..862b2cc12fbe 100644 --- a/arch/arm/kvm/trace.h +++ b/arch/arm/kvm/trace.h @@ -39,7 +39,53 @@ TRACE_EVENT(kvm_exit, TP_printk("PC: 0x%08lx", __entry->vcpu_pc) ); +TRACE_EVENT(kvm_unmap_hva, + TP_PROTO(unsigned long hva), + TP_ARGS(hva), + TP_STRUCT__entry( + __field( unsigned long, hva ) + ), + + TP_fast_assign( + __entry->hva = hva; + ), + + TP_printk("mmu notifier unmap hva: %#08lx", __entry->hva) +); + +TRACE_EVENT(kvm_unmap_hva_range, + TP_PROTO(unsigned long start, unsigned long end), + TP_ARGS(start, end), + + TP_STRUCT__entry( + __field( unsigned long, start ) + __field( unsigned long, end ) + ), + + TP_fast_assign( + __entry->start = start; + __entry->end = end; + ), + + TP_printk("mmu notifier unmap range: %#08lx -- %#08lx", + __entry->start, __entry->end) +); + +TRACE_EVENT(kvm_set_spte_hva, + TP_PROTO(unsigned long hva), + TP_ARGS(hva), + + TP_STRUCT__entry( + __field( unsigned long, hva ) + ), + + TP_fast_assign( + __entry->hva = hva; + ), + + TP_printk("mmu notifier set pte hva: %#08lx", __entry->hva) +); #endif /* _TRACE_KVM_H */ -- cgit v1.2.3 From 94f8e6418d3915dbefbb5d66b63146f1df12b0c0 Mon Sep 17 00:00:00 2001 From: Christoffer Dall Date: Sun, 20 Jan 2013 18:28:12 -0500 Subject: KVM: ARM: Handle guest faults in KVM Handles the guest faults in KVM by mapping in corresponding user pages in the 2nd stage page tables. We invalidate the instruction cache by MVA whenever we map a page to the guest (no, we cannot only do it when we have an iabt because the guest may happily read/write a page before hitting the icache) if the hardware uses VIPT or PIPT. In the latter case, we can invalidate only that physical page. In the first case, all bets are off and we simply must invalidate the whole affair. Not that VIVT icaches are tagged with vmids, and we are out of the woods on that one. Alexander Graf was nice enough to remind us of this massive pain. Reviewed-by: Will Deacon Reviewed-by: Marcelo Tosatti Signed-off-by: Marc Zyngier Signed-off-by: Christoffer Dall --- arch/arm/include/asm/kvm_asm.h | 2 + arch/arm/include/asm/kvm_mmu.h | 12 ++++ arch/arm/kvm/mmu.c | 153 ++++++++++++++++++++++++++++++++++++++++- arch/arm/kvm/trace.h | 26 +++++++ 4 files changed, 192 insertions(+), 1 deletion(-) (limited to 'arch/arm/include/asm/kvm_asm.h') diff --git a/arch/arm/include/asm/kvm_asm.h b/arch/arm/include/asm/kvm_asm.h index f6652f6c5d84..5e06e8177784 100644 --- a/arch/arm/include/asm/kvm_asm.h +++ b/arch/arm/include/asm/kvm_asm.h @@ -71,6 +71,8 @@ extern char __kvm_hyp_vector[]; extern char __kvm_hyp_code_start[]; extern char __kvm_hyp_code_end[]; +extern void __kvm_tlb_flush_vmid(struct kvm *kvm); + extern void __kvm_flush_vm_context(void); extern void __kvm_tlb_flush_vmid(struct kvm *kvm); diff --git a/arch/arm/include/asm/kvm_mmu.h b/arch/arm/include/asm/kvm_mmu.h index 499e7b0925ff..421a20b34874 100644 --- a/arch/arm/include/asm/kvm_mmu.h +++ b/arch/arm/include/asm/kvm_mmu.h @@ -35,4 +35,16 @@ void kvm_mmu_free_memory_caches(struct kvm_vcpu *vcpu); phys_addr_t kvm_mmu_get_httbr(void); int kvm_mmu_init(void); void kvm_clear_hyp_idmap(void); + +static inline bool kvm_is_write_fault(unsigned long hsr) +{ + unsigned long hsr_ec = hsr >> HSR_EC_SHIFT; + if (hsr_ec == HSR_EC_IABT) + return false; + else if ((hsr & HSR_ISV) && !(hsr & HSR_WNR)) + return false; + else + return true; +} + #endif /* __ARM_KVM_MMU_H__ */ diff --git a/arch/arm/kvm/mmu.c b/arch/arm/kvm/mmu.c index 4347d68f052f..a4b7b0f900e5 100644 --- a/arch/arm/kvm/mmu.c +++ b/arch/arm/kvm/mmu.c @@ -21,9 +21,11 @@ #include #include #include +#include #include #include #include +#include #include #include @@ -488,9 +490,158 @@ out: return ret; } +static void coherent_icache_guest_page(struct kvm *kvm, gfn_t gfn) +{ + /* + * If we are going to insert an instruction page and the icache is + * either VIPT or PIPT, there is a potential problem where the host + * (or another VM) may have used the same page as this guest, and we + * read incorrect data from the icache. If we're using a PIPT cache, + * we can invalidate just that page, but if we are using a VIPT cache + * we need to invalidate the entire icache - damn shame - as written + * in the ARM ARM (DDI 0406C.b - Page B3-1393). + * + * VIVT caches are tagged using both the ASID and the VMID and doesn't + * need any kind of flushing (DDI 0406C.b - Page B3-1392). + */ + if (icache_is_pipt()) { + unsigned long hva = gfn_to_hva(kvm, gfn); + __cpuc_coherent_user_range(hva, hva + PAGE_SIZE); + } else if (!icache_is_vivt_asid_tagged()) { + /* any kind of VIPT cache */ + __flush_icache_all(); + } +} + +static int user_mem_abort(struct kvm_vcpu *vcpu, phys_addr_t fault_ipa, + gfn_t gfn, struct kvm_memory_slot *memslot, + unsigned long fault_status) +{ + pte_t new_pte; + pfn_t pfn; + int ret; + bool write_fault, writable; + unsigned long mmu_seq; + struct kvm_mmu_memory_cache *memcache = &vcpu->arch.mmu_page_cache; + + write_fault = kvm_is_write_fault(vcpu->arch.hsr); + if (fault_status == FSC_PERM && !write_fault) { + kvm_err("Unexpected L2 read permission error\n"); + return -EFAULT; + } + + /* We need minimum second+third level pages */ + ret = mmu_topup_memory_cache(memcache, 2, KVM_NR_MEM_OBJS); + if (ret) + return ret; + + mmu_seq = vcpu->kvm->mmu_notifier_seq; + /* + * Ensure the read of mmu_notifier_seq happens before we call + * gfn_to_pfn_prot (which calls get_user_pages), so that we don't risk + * the page we just got a reference to gets unmapped before we have a + * chance to grab the mmu_lock, which ensure that if the page gets + * unmapped afterwards, the call to kvm_unmap_hva will take it away + * from us again properly. This smp_rmb() interacts with the smp_wmb() + * in kvm_mmu_notifier_invalidate_. + */ + smp_rmb(); + + pfn = gfn_to_pfn_prot(vcpu->kvm, gfn, write_fault, &writable); + if (is_error_pfn(pfn)) + return -EFAULT; + + new_pte = pfn_pte(pfn, PAGE_S2); + coherent_icache_guest_page(vcpu->kvm, gfn); + + spin_lock(&vcpu->kvm->mmu_lock); + if (mmu_notifier_retry(vcpu->kvm, mmu_seq)) + goto out_unlock; + if (writable) { + pte_val(new_pte) |= L_PTE_S2_RDWR; + kvm_set_pfn_dirty(pfn); + } + stage2_set_pte(vcpu->kvm, memcache, fault_ipa, &new_pte, false); + +out_unlock: + spin_unlock(&vcpu->kvm->mmu_lock); + kvm_release_pfn_clean(pfn); + return 0; +} + +/** + * kvm_handle_guest_abort - handles all 2nd stage aborts + * @vcpu: the VCPU pointer + * @run: the kvm_run structure + * + * Any abort that gets to the host is almost guaranteed to be caused by a + * missing second stage translation table entry, which can mean that either the + * guest simply needs more memory and we must allocate an appropriate page or it + * can mean that the guest tried to access I/O memory, which is emulated by user + * space. The distinction is based on the IPA causing the fault and whether this + * memory region has been registered as standard RAM by user space. + */ int kvm_handle_guest_abort(struct kvm_vcpu *vcpu, struct kvm_run *run) { - return -EINVAL; + unsigned long hsr_ec; + unsigned long fault_status; + phys_addr_t fault_ipa; + struct kvm_memory_slot *memslot; + bool is_iabt; + gfn_t gfn; + int ret, idx; + + hsr_ec = vcpu->arch.hsr >> HSR_EC_SHIFT; + is_iabt = (hsr_ec == HSR_EC_IABT); + fault_ipa = ((phys_addr_t)vcpu->arch.hpfar & HPFAR_MASK) << 8; + + trace_kvm_guest_fault(*vcpu_pc(vcpu), vcpu->arch.hsr, + vcpu->arch.hxfar, fault_ipa); + + /* Check the stage-2 fault is trans. fault or write fault */ + fault_status = (vcpu->arch.hsr & HSR_FSC_TYPE); + if (fault_status != FSC_FAULT && fault_status != FSC_PERM) { + kvm_err("Unsupported fault status: EC=%#lx DFCS=%#lx\n", + hsr_ec, fault_status); + return -EFAULT; + } + + idx = srcu_read_lock(&vcpu->kvm->srcu); + + gfn = fault_ipa >> PAGE_SHIFT; + if (!kvm_is_visible_gfn(vcpu->kvm, gfn)) { + if (is_iabt) { + /* Prefetch Abort on I/O address */ + kvm_inject_pabt(vcpu, vcpu->arch.hxfar); + ret = 1; + goto out_unlock; + } + + if (fault_status != FSC_FAULT) { + kvm_err("Unsupported fault status on io memory: %#lx\n", + fault_status); + ret = -EFAULT; + goto out_unlock; + } + + kvm_pr_unimpl("I/O address abort..."); + ret = 0; + goto out_unlock; + } + + memslot = gfn_to_memslot(vcpu->kvm, gfn); + if (!memslot->user_alloc) { + kvm_err("non user-alloc memslots not supported\n"); + ret = -EINVAL; + goto out_unlock; + } + + ret = user_mem_abort(vcpu, fault_ipa, gfn, memslot, fault_status); + if (ret == 0) + ret = 1; +out_unlock: + srcu_read_unlock(&vcpu->kvm->srcu, idx); + return ret; } static void handle_hva_to_gpa(struct kvm *kvm, diff --git a/arch/arm/kvm/trace.h b/arch/arm/kvm/trace.h index 022305b38c27..624b5a4e8fad 100644 --- a/arch/arm/kvm/trace.h +++ b/arch/arm/kvm/trace.h @@ -39,6 +39,32 @@ TRACE_EVENT(kvm_exit, TP_printk("PC: 0x%08lx", __entry->vcpu_pc) ); +TRACE_EVENT(kvm_guest_fault, + TP_PROTO(unsigned long vcpu_pc, unsigned long hsr, + unsigned long hxfar, + unsigned long long ipa), + TP_ARGS(vcpu_pc, hsr, hxfar, ipa), + + TP_STRUCT__entry( + __field( unsigned long, vcpu_pc ) + __field( unsigned long, hsr ) + __field( unsigned long, hxfar ) + __field( unsigned long long, ipa ) + ), + + TP_fast_assign( + __entry->vcpu_pc = vcpu_pc; + __entry->hsr = hsr; + __entry->hxfar = hxfar; + __entry->ipa = ipa; + ), + + TP_printk("guest fault at PC %#08lx (hxfar %#08lx, " + "ipa %#16llx, hsr %#08lx", + __entry->vcpu_pc, __entry->hxfar, + __entry->ipa, __entry->hsr) +); + TRACE_EVENT(kvm_irq_line, TP_PROTO(unsigned int type, int vcpu_idx, int irq_num, int level), TP_ARGS(type, vcpu_idx, irq_num, level), -- cgit v1.2.3