From 402b08622d9ac6e32e25289573272e0f21bb58a7 Mon Sep 17 00:00:00 2001 From: Carsten Otte Date: Tue, 25 Mar 2008 18:47:10 +0100 Subject: s390: KVM preparation: provide hook to enable pgstes in user pagetable The SIE instruction on s390 uses the 2nd half of the page table page to virtualize the storage keys of a guest. This patch offers the s390_enable_sie function, which reorganizes the page tables of a single-threaded process to reserve space in the page table: s390_enable_sie makes sure that the process is single threaded and then uses dup_mm to create a new mm with reorganized page tables. The old mm is freed and the process has now a page status extended field after every page table. Code that wants to exploit pgstes should SELECT CONFIG_PGSTE. This patch has a small common code hit, namely making dup_mm non-static. Edit (Carsten): I've modified Martin's patch, following Jeremy Fitzhardinge's review feedback. Now we do have the prototype for dup_mm in include/linux/sched.h. Following Martin's suggestion, s390_enable_sie() does now call task_lock() to prevent race against ptrace modification of mm_users. Signed-off-by: Martin Schwidefsky Signed-off-by: Carsten Otte Acked-by: Andrew Morton Signed-off-by: Avi Kivity --- arch/s390/Kconfig | 4 +++ arch/s390/kernel/setup.c | 4 +++ arch/s390/mm/pgtable.c | 65 +++++++++++++++++++++++++++++++++++++++++++++--- 3 files changed, 70 insertions(+), 3 deletions(-) (limited to 'arch/s390') diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig index f6a68e178fc..513a0589e81 100644 --- a/arch/s390/Kconfig +++ b/arch/s390/Kconfig @@ -62,6 +62,10 @@ config GENERIC_LOCKBREAK default y depends on SMP && PREEMPT +config PGSTE + bool + default y if KVM + mainmenu "Linux Kernel Configuration" config S390 diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c index 7141147e6b6..2f35133ebc1 100644 --- a/arch/s390/kernel/setup.c +++ b/arch/s390/kernel/setup.c @@ -316,7 +316,11 @@ static int __init early_parse_ipldelay(char *p) early_param("ipldelay", early_parse_ipldelay); #ifdef CONFIG_S390_SWITCH_AMODE +#ifdef CONFIG_PGSTE +unsigned int switch_amode = 1; +#else unsigned int switch_amode = 0; +#endif EXPORT_SYMBOL_GPL(switch_amode); static void set_amode_and_uaccess(unsigned long user_amode, diff --git a/arch/s390/mm/pgtable.c b/arch/s390/mm/pgtable.c index fd072013f88..5c1aea97cd1 100644 --- a/arch/s390/mm/pgtable.c +++ b/arch/s390/mm/pgtable.c @@ -30,11 +30,27 @@ #define TABLES_PER_PAGE 4 #define FRAG_MASK 15UL #define SECOND_HALVES 10UL + +void clear_table_pgstes(unsigned long *table) +{ + clear_table(table, _PAGE_TYPE_EMPTY, PAGE_SIZE/4); + memset(table + 256, 0, PAGE_SIZE/4); + clear_table(table + 512, _PAGE_TYPE_EMPTY, PAGE_SIZE/4); + memset(table + 768, 0, PAGE_SIZE/4); +} + #else #define ALLOC_ORDER 2 #define TABLES_PER_PAGE 2 #define FRAG_MASK 3UL #define SECOND_HALVES 2UL + +void clear_table_pgstes(unsigned long *table) +{ + clear_table(table, _PAGE_TYPE_EMPTY, PAGE_SIZE/2); + memset(table + 256, 0, PAGE_SIZE/2); +} + #endif unsigned long *crst_table_alloc(struct mm_struct *mm, int noexec) @@ -153,7 +169,7 @@ unsigned long *page_table_alloc(struct mm_struct *mm) unsigned long *table; unsigned long bits; - bits = mm->context.noexec ? 3UL : 1UL; + bits = (mm->context.noexec || mm->context.pgstes) ? 3UL : 1UL; spin_lock(&mm->page_table_lock); page = NULL; if (!list_empty(&mm->context.pgtable_list)) { @@ -170,7 +186,10 @@ unsigned long *page_table_alloc(struct mm_struct *mm) pgtable_page_ctor(page); page->flags &= ~FRAG_MASK; table = (unsigned long *) page_to_phys(page); - clear_table(table, _PAGE_TYPE_EMPTY, PAGE_SIZE); + if (mm->context.pgstes) + clear_table_pgstes(table); + else + clear_table(table, _PAGE_TYPE_EMPTY, PAGE_SIZE); spin_lock(&mm->page_table_lock); list_add(&page->lru, &mm->context.pgtable_list); } @@ -191,7 +210,7 @@ void page_table_free(struct mm_struct *mm, unsigned long *table) struct page *page; unsigned long bits; - bits = mm->context.noexec ? 3UL : 1UL; + bits = (mm->context.noexec || mm->context.pgstes) ? 3UL : 1UL; bits <<= (__pa(table) & (PAGE_SIZE - 1)) / 256 / sizeof(unsigned long); page = pfn_to_page(__pa(table) >> PAGE_SHIFT); spin_lock(&mm->page_table_lock); @@ -228,3 +247,43 @@ void disable_noexec(struct mm_struct *mm, struct task_struct *tsk) mm->context.noexec = 0; update_mm(mm, tsk); } + +/* + * switch on pgstes for its userspace process (for kvm) + */ +int s390_enable_sie(void) +{ + struct task_struct *tsk = current; + struct mm_struct *mm; + int rc; + + task_lock(tsk); + + rc = 0; + if (tsk->mm->context.pgstes) + goto unlock; + + rc = -EINVAL; + if (!tsk->mm || atomic_read(&tsk->mm->mm_users) > 1 || + tsk->mm != tsk->active_mm || tsk->mm->ioctx_list) + goto unlock; + + tsk->mm->context.pgstes = 1; /* dirty little tricks .. */ + mm = dup_mm(tsk); + tsk->mm->context.pgstes = 0; + + rc = -ENOMEM; + if (!mm) + goto unlock; + mmput(tsk->mm); + tsk->mm = tsk->active_mm = mm; + preempt_disable(); + update_mm(mm, tsk); + cpu_set(smp_processor_id(), mm->cpu_vm_mask); + preempt_enable(); + rc = 0; +unlock: + task_unlock(tsk); + return rc; +} +EXPORT_SYMBOL_GPL(s390_enable_sie); -- cgit v1.2.3-70-g09d2 From b0c632db637d68ad39d9f97f452ce176253f5f4e Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Tue, 25 Mar 2008 18:47:20 +0100 Subject: KVM: s390: arch backend for the kvm kernel module This patch contains the port of Qumranet's kvm kernel module to IBM zSeries (aka s390x, mainframe) architecture. It uses the mainframe's virtualization instruction SIE to run virtual machines with up to 64 virtual CPUs each. This port is only usable on 64bit host kernels, and can only run 64bit guest kernels. However, running 31bit applications in guest userspace is possible. The following source files are introduced by this patch arch/s390/kvm/kvm-s390.c similar to arch/x86/kvm/x86.c, this implements all arch callbacks for kvm. __vcpu_run calls back into sie64a to enter the guest machine context arch/s390/kvm/sie64a.S assembler function sie64a, which enters guest context via SIE, and switches world before and after that include/asm-s390/kvm_host.h contains all vital data structures needed to run virtual machines on the mainframe include/asm-s390/kvm.h defines kvm_regs and friends for user access to guest register content arch/s390/kvm/gaccess.h functions similar to uaccess to access guest memory arch/s390/kvm/kvm-s390.h header file for kvm-s390 internals, extended by later patches Acked-by: Martin Schwidefsky Signed-off-by: Christian Borntraeger Signed-off-by: Heiko Carstens Signed-off-by: Carsten Otte Signed-off-by: Avi Kivity --- arch/s390/Makefile | 2 +- arch/s390/kernel/vtime.c | 1 + arch/s390/kvm/Makefile | 14 ++ arch/s390/kvm/gaccess.h | 274 +++++++++++++++++++++ arch/s390/kvm/kvm-s390.c | 566 ++++++++++++++++++++++++++++++++++++++++++++ arch/s390/kvm/kvm-s390.h | 29 +++ arch/s390/kvm/sie64a.S | 47 ++++ include/asm-s390/Kbuild | 1 + include/asm-s390/kvm.h | 41 +++- include/asm-s390/kvm_host.h | 127 ++++++++++ include/asm-s390/kvm_para.h | 30 +++ include/linux/kvm.h | 15 ++ 12 files changed, 1145 insertions(+), 2 deletions(-) create mode 100644 arch/s390/kvm/Makefile create mode 100644 arch/s390/kvm/gaccess.h create mode 100644 arch/s390/kvm/kvm-s390.c create mode 100644 arch/s390/kvm/kvm-s390.h create mode 100644 arch/s390/kvm/sie64a.S create mode 100644 include/asm-s390/kvm_host.h create mode 100644 include/asm-s390/kvm_para.h (limited to 'arch/s390') diff --git a/arch/s390/Makefile b/arch/s390/Makefile index f708be367b0..792a4e7743c 100644 --- a/arch/s390/Makefile +++ b/arch/s390/Makefile @@ -87,7 +87,7 @@ LDFLAGS_vmlinux := -e start head-y := arch/s390/kernel/head.o arch/s390/kernel/init_task.o core-y += arch/s390/mm/ arch/s390/kernel/ arch/s390/crypto/ \ - arch/s390/appldata/ arch/s390/hypfs/ + arch/s390/appldata/ arch/s390/hypfs/ arch/s390/kvm/ libs-y += arch/s390/lib/ drivers-y += drivers/s390/ drivers-$(CONFIG_MATHEMU) += arch/s390/math-emu/ diff --git a/arch/s390/kernel/vtime.c b/arch/s390/kernel/vtime.c index c5f05b3fb2c..ca90ee3f930 100644 --- a/arch/s390/kernel/vtime.c +++ b/arch/s390/kernel/vtime.c @@ -110,6 +110,7 @@ void account_system_vtime(struct task_struct *tsk) S390_lowcore.steal_clock -= cputime << 12; account_system_time(tsk, 0, cputime); } +EXPORT_SYMBOL_GPL(account_system_vtime); static inline void set_vtimer(__u64 expires) { diff --git a/arch/s390/kvm/Makefile b/arch/s390/kvm/Makefile new file mode 100644 index 00000000000..0d8d1135a27 --- /dev/null +++ b/arch/s390/kvm/Makefile @@ -0,0 +1,14 @@ +# Makefile for kernel virtual machines on s390 +# +# Copyright IBM Corp. 2008 +# +# This program is free software; you can redistribute it and/or modify +# it under the terms of the GNU General Public License (version 2 only) +# as published by the Free Software Foundation. + +common-objs = $(addprefix ../../../virt/kvm/, kvm_main.o) + +EXTRA_CFLAGS += -Ivirt/kvm -Iarch/s390/kvm + +kvm-objs := $(common-objs) kvm-s390.o sie64a.o +obj-$(CONFIG_KVM) += kvm.o diff --git a/arch/s390/kvm/gaccess.h b/arch/s390/kvm/gaccess.h new file mode 100644 index 00000000000..4e0633c413f --- /dev/null +++ b/arch/s390/kvm/gaccess.h @@ -0,0 +1,274 @@ +/* + * gaccess.h - access guest memory + * + * Copyright IBM Corp. 2008 + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License (version 2 only) + * as published by the Free Software Foundation. + * + * Author(s): Carsten Otte + */ + +#ifndef __KVM_S390_GACCESS_H +#define __KVM_S390_GACCESS_H + +#include +#include +#include + +static inline void __user *__guestaddr_to_user(struct kvm_vcpu *vcpu, + u64 guestaddr) +{ + u64 prefix = vcpu->arch.sie_block->prefix; + u64 origin = vcpu->kvm->arch.guest_origin; + u64 memsize = vcpu->kvm->arch.guest_memsize; + + if (guestaddr < 2 * PAGE_SIZE) + guestaddr += prefix; + else if ((guestaddr >= prefix) && (guestaddr < prefix + 2 * PAGE_SIZE)) + guestaddr -= prefix; + + if (guestaddr > memsize) + return (void __user __force *) ERR_PTR(-EFAULT); + + guestaddr += origin; + + return (void __user *) guestaddr; +} + +static inline int get_guest_u64(struct kvm_vcpu *vcpu, u64 guestaddr, + u64 *result) +{ + void __user *uptr = __guestaddr_to_user(vcpu, guestaddr); + + BUG_ON(guestaddr & 7); + + if (IS_ERR((void __force *) uptr)) + return PTR_ERR((void __force *) uptr); + + return get_user(*result, (u64 __user *) uptr); +} + +static inline int get_guest_u32(struct kvm_vcpu *vcpu, u64 guestaddr, + u32 *result) +{ + void __user *uptr = __guestaddr_to_user(vcpu, guestaddr); + + BUG_ON(guestaddr & 3); + + if (IS_ERR((void __force *) uptr)) + return PTR_ERR((void __force *) uptr); + + return get_user(*result, (u32 __user *) uptr); +} + +static inline int get_guest_u16(struct kvm_vcpu *vcpu, u64 guestaddr, + u16 *result) +{ + void __user *uptr = __guestaddr_to_user(vcpu, guestaddr); + + BUG_ON(guestaddr & 1); + + if (IS_ERR(uptr)) + return PTR_ERR(uptr); + + return get_user(*result, (u16 __user *) uptr); +} + +static inline int get_guest_u8(struct kvm_vcpu *vcpu, u64 guestaddr, + u8 *result) +{ + void __user *uptr = __guestaddr_to_user(vcpu, guestaddr); + + if (IS_ERR((void __force *) uptr)) + return PTR_ERR((void __force *) uptr); + + return get_user(*result, (u8 __user *) uptr); +} + +static inline int put_guest_u64(struct kvm_vcpu *vcpu, u64 guestaddr, + u64 value) +{ + void __user *uptr = __guestaddr_to_user(vcpu, guestaddr); + + BUG_ON(guestaddr & 7); + + if (IS_ERR((void __force *) uptr)) + return PTR_ERR((void __force *) uptr); + + return put_user(value, (u64 __user *) uptr); +} + +static inline int put_guest_u32(struct kvm_vcpu *vcpu, u64 guestaddr, + u32 value) +{ + void __user *uptr = __guestaddr_to_user(vcpu, guestaddr); + + BUG_ON(guestaddr & 3); + + if (IS_ERR((void __force *) uptr)) + return PTR_ERR((void __force *) uptr); + + return put_user(value, (u32 __user *) uptr); +} + +static inline int put_guest_u16(struct kvm_vcpu *vcpu, u64 guestaddr, + u16 value) +{ + void __user *uptr = __guestaddr_to_user(vcpu, guestaddr); + + BUG_ON(guestaddr & 1); + + if (IS_ERR((void __force *) uptr)) + return PTR_ERR((void __force *) uptr); + + return put_user(value, (u16 __user *) uptr); +} + +static inline int put_guest_u8(struct kvm_vcpu *vcpu, u64 guestaddr, + u8 value) +{ + void __user *uptr = __guestaddr_to_user(vcpu, guestaddr); + + if (IS_ERR((void __force *) uptr)) + return PTR_ERR((void __force *) uptr); + + return put_user(value, (u8 __user *) uptr); +} + + +static inline int __copy_to_guest_slow(struct kvm_vcpu *vcpu, u64 guestdest, + const void *from, unsigned long n) +{ + int rc; + unsigned long i; + const u8 *data = from; + + for (i = 0; i < n; i++) { + rc = put_guest_u8(vcpu, guestdest++, *(data++)); + if (rc < 0) + return rc; + } + return 0; +} + +static inline int copy_to_guest(struct kvm_vcpu *vcpu, u64 guestdest, + const void *from, unsigned long n) +{ + u64 prefix = vcpu->arch.sie_block->prefix; + u64 origin = vcpu->kvm->arch.guest_origin; + u64 memsize = vcpu->kvm->arch.guest_memsize; + + if ((guestdest < 2 * PAGE_SIZE) && (guestdest + n > 2 * PAGE_SIZE)) + goto slowpath; + + if ((guestdest < prefix) && (guestdest + n > prefix)) + goto slowpath; + + if ((guestdest < prefix + 2 * PAGE_SIZE) + && (guestdest + n > prefix + 2 * PAGE_SIZE)) + goto slowpath; + + if (guestdest < 2 * PAGE_SIZE) + guestdest += prefix; + else if ((guestdest >= prefix) && (guestdest < prefix + 2 * PAGE_SIZE)) + guestdest -= prefix; + + if (guestdest + n > memsize) + return -EFAULT; + + if (guestdest + n < guestdest) + return -EFAULT; + + guestdest += origin; + + return copy_to_user((void __user *) guestdest, from, n); +slowpath: + return __copy_to_guest_slow(vcpu, guestdest, from, n); +} + +static inline int __copy_from_guest_slow(struct kvm_vcpu *vcpu, void *to, + u64 guestsrc, unsigned long n) +{ + int rc; + unsigned long i; + u8 *data = to; + + for (i = 0; i < n; i++) { + rc = get_guest_u8(vcpu, guestsrc++, data++); + if (rc < 0) + return rc; + } + return 0; +} + +static inline int copy_from_guest(struct kvm_vcpu *vcpu, void *to, + u64 guestsrc, unsigned long n) +{ + u64 prefix = vcpu->arch.sie_block->prefix; + u64 origin = vcpu->kvm->arch.guest_origin; + u64 memsize = vcpu->kvm->arch.guest_memsize; + + if ((guestsrc < 2 * PAGE_SIZE) && (guestsrc + n > 2 * PAGE_SIZE)) + goto slowpath; + + if ((guestsrc < prefix) && (guestsrc + n > prefix)) + goto slowpath; + + if ((guestsrc < prefix + 2 * PAGE_SIZE) + && (guestsrc + n > prefix + 2 * PAGE_SIZE)) + goto slowpath; + + if (guestsrc < 2 * PAGE_SIZE) + guestsrc += prefix; + else if ((guestsrc >= prefix) && (guestsrc < prefix + 2 * PAGE_SIZE)) + guestsrc -= prefix; + + if (guestsrc + n > memsize) + return -EFAULT; + + if (guestsrc + n < guestsrc) + return -EFAULT; + + guestsrc += origin; + + return copy_from_user(to, (void __user *) guestsrc, n); +slowpath: + return __copy_from_guest_slow(vcpu, to, guestsrc, n); +} + +static inline int copy_to_guest_absolute(struct kvm_vcpu *vcpu, u64 guestdest, + const void *from, unsigned long n) +{ + u64 origin = vcpu->kvm->arch.guest_origin; + u64 memsize = vcpu->kvm->arch.guest_memsize; + + if (guestdest + n > memsize) + return -EFAULT; + + if (guestdest + n < guestdest) + return -EFAULT; + + guestdest += origin; + + return copy_to_user((void __user *) guestdest, from, n); +} + +static inline int copy_from_guest_absolute(struct kvm_vcpu *vcpu, void *to, + u64 guestsrc, unsigned long n) +{ + u64 origin = vcpu->kvm->arch.guest_origin; + u64 memsize = vcpu->kvm->arch.guest_memsize; + + if (guestsrc + n > memsize) + return -EFAULT; + + if (guestsrc + n < guestsrc) + return -EFAULT; + + guestsrc += origin; + + return copy_from_user(to, (void __user *) guestsrc, n); +} +#endif diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c new file mode 100644 index 00000000000..6e1e1d39ae1 --- /dev/null +++ b/arch/s390/kvm/kvm-s390.c @@ -0,0 +1,566 @@ +/* + * s390host.c -- hosting zSeries kernel virtual machines + * + * Copyright IBM Corp. 2008 + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License (version 2 only) + * as published by the Free Software Foundation. + * + * Author(s): Carsten Otte + * Christian Borntraeger + * Heiko Carstens + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include "gaccess.h" + +#define VCPU_STAT(x) offsetof(struct kvm_vcpu, stat.x), KVM_STAT_VCPU + +struct kvm_stats_debugfs_item debugfs_entries[] = { + { "userspace_handled", VCPU_STAT(exit_userspace) }, + { NULL } +}; + + +/* Section: not file related */ +void kvm_arch_hardware_enable(void *garbage) +{ + /* every s390 is virtualization enabled ;-) */ +} + +void kvm_arch_hardware_disable(void *garbage) +{ +} + +void decache_vcpus_on_cpu(int cpu) +{ +} + +int kvm_arch_hardware_setup(void) +{ + return 0; +} + +void kvm_arch_hardware_unsetup(void) +{ +} + +void kvm_arch_check_processor_compat(void *rtn) +{ +} + +int kvm_arch_init(void *opaque) +{ + return 0; +} + +void kvm_arch_exit(void) +{ +} + +/* Section: device related */ +long kvm_arch_dev_ioctl(struct file *filp, + unsigned int ioctl, unsigned long arg) +{ + if (ioctl == KVM_S390_ENABLE_SIE) + return s390_enable_sie(); + return -EINVAL; +} + +int kvm_dev_ioctl_check_extension(long ext) +{ + return 0; +} + +/* Section: vm related */ +/* + * Get (and clear) the dirty memory log for a memory slot. + */ +int kvm_vm_ioctl_get_dirty_log(struct kvm *kvm, + struct kvm_dirty_log *log) +{ + return 0; +} + +long kvm_arch_vm_ioctl(struct file *filp, + unsigned int ioctl, unsigned long arg) +{ + struct kvm *kvm = filp->private_data; + void __user *argp = (void __user *)arg; + int r; + + switch (ioctl) { + default: + r = -EINVAL; + } + + return r; +} + +struct kvm *kvm_arch_create_vm(void) +{ + struct kvm *kvm; + int rc; + char debug_name[16]; + + rc = s390_enable_sie(); + if (rc) + goto out_nokvm; + + rc = -ENOMEM; + kvm = kzalloc(sizeof(struct kvm), GFP_KERNEL); + if (!kvm) + goto out_nokvm; + + kvm->arch.sca = (struct sca_block *) get_zeroed_page(GFP_KERNEL); + if (!kvm->arch.sca) + goto out_nosca; + + sprintf(debug_name, "kvm-%u", current->pid); + + kvm->arch.dbf = debug_register(debug_name, 8, 2, 8 * sizeof(long)); + if (!kvm->arch.dbf) + goto out_nodbf; + + debug_register_view(kvm->arch.dbf, &debug_sprintf_view); + VM_EVENT(kvm, 3, "%s", "vm created"); + + try_module_get(THIS_MODULE); + + return kvm; +out_nodbf: + free_page((unsigned long)(kvm->arch.sca)); +out_nosca: + kfree(kvm); +out_nokvm: + return ERR_PTR(rc); +} + +void kvm_arch_destroy_vm(struct kvm *kvm) +{ + debug_unregister(kvm->arch.dbf); + free_page((unsigned long)(kvm->arch.sca)); + kfree(kvm); + module_put(THIS_MODULE); +} + +/* Section: vcpu related */ +int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu) +{ + return 0; +} + +void kvm_arch_vcpu_uninit(struct kvm_vcpu *vcpu) +{ + /* kvm common code refers to this, but does'nt call it */ + BUG(); +} + +void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu) +{ + save_fp_regs(&vcpu->arch.host_fpregs); + save_access_regs(vcpu->arch.host_acrs); + vcpu->arch.guest_fpregs.fpc &= FPC_VALID_MASK; + restore_fp_regs(&vcpu->arch.guest_fpregs); + restore_access_regs(vcpu->arch.guest_acrs); + + if (signal_pending(current)) + atomic_set_mask(CPUSTAT_STOP_INT, + &vcpu->arch.sie_block->cpuflags); +} + +void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu) +{ + save_fp_regs(&vcpu->arch.guest_fpregs); + save_access_regs(vcpu->arch.guest_acrs); + restore_fp_regs(&vcpu->arch.host_fpregs); + restore_access_regs(vcpu->arch.host_acrs); +} + +static void kvm_s390_vcpu_initial_reset(struct kvm_vcpu *vcpu) +{ + /* this equals initial cpu reset in pop, but we don't switch to ESA */ + vcpu->arch.sie_block->gpsw.mask = 0UL; + vcpu->arch.sie_block->gpsw.addr = 0UL; + vcpu->arch.sie_block->prefix = 0UL; + vcpu->arch.sie_block->ihcpu = 0xffff; + vcpu->arch.sie_block->cputm = 0UL; + vcpu->arch.sie_block->ckc = 0UL; + vcpu->arch.sie_block->todpr = 0; + memset(vcpu->arch.sie_block->gcr, 0, 16 * sizeof(__u64)); + vcpu->arch.sie_block->gcr[0] = 0xE0UL; + vcpu->arch.sie_block->gcr[14] = 0xC2000000UL; + vcpu->arch.guest_fpregs.fpc = 0; + asm volatile("lfpc %0" : : "Q" (vcpu->arch.guest_fpregs.fpc)); + vcpu->arch.sie_block->gbea = 1; +} + +int kvm_arch_vcpu_setup(struct kvm_vcpu *vcpu) +{ + atomic_set(&vcpu->arch.sie_block->cpuflags, CPUSTAT_ZARCH); + vcpu->arch.sie_block->gmslm = 0xffffffffffUL; + vcpu->arch.sie_block->gmsor = 0x000000000000; + vcpu->arch.sie_block->ecb = 2; + vcpu->arch.sie_block->eca = 0xC1002001U; + + return 0; +} + +struct kvm_vcpu *kvm_arch_vcpu_create(struct kvm *kvm, + unsigned int id) +{ + struct kvm_vcpu *vcpu = kzalloc(sizeof(struct kvm_vcpu), GFP_KERNEL); + int rc = -ENOMEM; + + if (!vcpu) + goto out_nomem; + + vcpu->arch.sie_block = (struct sie_block *) get_zeroed_page(GFP_KERNEL); + + if (!vcpu->arch.sie_block) + goto out_free_cpu; + + vcpu->arch.sie_block->icpua = id; + BUG_ON(!kvm->arch.sca); + BUG_ON(kvm->arch.sca->cpu[id].sda); + kvm->arch.sca->cpu[id].sda = (__u64) vcpu->arch.sie_block; + vcpu->arch.sie_block->scaoh = (__u32)(((__u64)kvm->arch.sca) >> 32); + vcpu->arch.sie_block->scaol = (__u32)(__u64)kvm->arch.sca; + + rc = kvm_vcpu_init(vcpu, kvm, id); + if (rc) + goto out_free_cpu; + VM_EVENT(kvm, 3, "create cpu %d at %p, sie block at %p", id, vcpu, + vcpu->arch.sie_block); + + try_module_get(THIS_MODULE); + + return vcpu; +out_free_cpu: + kfree(vcpu); +out_nomem: + return ERR_PTR(rc); +} + +void kvm_arch_vcpu_destroy(struct kvm_vcpu *vcpu) +{ + VCPU_EVENT(vcpu, 3, "%s", "destroy cpu"); + free_page((unsigned long)(vcpu->arch.sie_block)); + kfree(vcpu); + module_put(THIS_MODULE); +} + +int kvm_arch_vcpu_runnable(struct kvm_vcpu *vcpu) +{ + /* kvm common code refers to this, but never calls it */ + BUG(); + return 0; +} + +static int kvm_arch_vcpu_ioctl_initial_reset(struct kvm_vcpu *vcpu) +{ + vcpu_load(vcpu); + kvm_s390_vcpu_initial_reset(vcpu); + vcpu_put(vcpu); + return 0; +} + +int kvm_arch_vcpu_ioctl_set_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs) +{ + vcpu_load(vcpu); + memcpy(&vcpu->arch.guest_gprs, ®s->gprs, sizeof(regs->gprs)); + vcpu_put(vcpu); + return 0; +} + +int kvm_arch_vcpu_ioctl_get_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs) +{ + vcpu_load(vcpu); + memcpy(®s->gprs, &vcpu->arch.guest_gprs, sizeof(regs->gprs)); + vcpu_put(vcpu); + return 0; +} + +int kvm_arch_vcpu_ioctl_set_sregs(struct kvm_vcpu *vcpu, + struct kvm_sregs *sregs) +{ + vcpu_load(vcpu); + memcpy(&vcpu->arch.guest_acrs, &sregs->acrs, sizeof(sregs->acrs)); + memcpy(&vcpu->arch.sie_block->gcr, &sregs->crs, sizeof(sregs->crs)); + vcpu_put(vcpu); + return 0; +} + +int kvm_arch_vcpu_ioctl_get_sregs(struct kvm_vcpu *vcpu, + struct kvm_sregs *sregs) +{ + vcpu_load(vcpu); + memcpy(&sregs->acrs, &vcpu->arch.guest_acrs, sizeof(sregs->acrs)); + memcpy(&sregs->crs, &vcpu->arch.sie_block->gcr, sizeof(sregs->crs)); + vcpu_put(vcpu); + return 0; +} + +int kvm_arch_vcpu_ioctl_set_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu) +{ + vcpu_load(vcpu); + memcpy(&vcpu->arch.guest_fpregs.fprs, &fpu->fprs, sizeof(fpu->fprs)); + vcpu->arch.guest_fpregs.fpc = fpu->fpc; + vcpu_put(vcpu); + return 0; +} + +int kvm_arch_vcpu_ioctl_get_fpu(struct kvm_vcpu *vcpu, struct kvm_fpu *fpu) +{ + vcpu_load(vcpu); + memcpy(&fpu->fprs, &vcpu->arch.guest_fpregs.fprs, sizeof(fpu->fprs)); + fpu->fpc = vcpu->arch.guest_fpregs.fpc; + vcpu_put(vcpu); + return 0; +} + +static int kvm_arch_vcpu_ioctl_set_initial_psw(struct kvm_vcpu *vcpu, psw_t psw) +{ + int rc = 0; + + vcpu_load(vcpu); + if (atomic_read(&vcpu->arch.sie_block->cpuflags) & CPUSTAT_RUNNING) + rc = -EBUSY; + else + vcpu->arch.sie_block->gpsw = psw; + vcpu_put(vcpu); + return rc; +} + +int kvm_arch_vcpu_ioctl_translate(struct kvm_vcpu *vcpu, + struct kvm_translation *tr) +{ + return -EINVAL; /* not implemented yet */ +} + +int kvm_arch_vcpu_ioctl_debug_guest(struct kvm_vcpu *vcpu, + struct kvm_debug_guest *dbg) +{ + return -EINVAL; /* not implemented yet */ +} + +static void __vcpu_run(struct kvm_vcpu *vcpu) +{ + memcpy(&vcpu->arch.sie_block->gg14, &vcpu->arch.guest_gprs[14], 16); + + if (need_resched()) + schedule(); + + vcpu->arch.sie_block->icptcode = 0; + local_irq_disable(); + kvm_guest_enter(); + local_irq_enable(); + VCPU_EVENT(vcpu, 6, "entering sie flags %x", + atomic_read(&vcpu->arch.sie_block->cpuflags)); + sie64a(vcpu->arch.sie_block, vcpu->arch.guest_gprs); + VCPU_EVENT(vcpu, 6, "exit sie icptcode %d", + vcpu->arch.sie_block->icptcode); + local_irq_disable(); + kvm_guest_exit(); + local_irq_enable(); + + memcpy(&vcpu->arch.guest_gprs[14], &vcpu->arch.sie_block->gg14, 16); +} + +int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) +{ + sigset_t sigsaved; + + vcpu_load(vcpu); + + if (vcpu->sigset_active) + sigprocmask(SIG_SETMASK, &vcpu->sigset, &sigsaved); + + atomic_set_mask(CPUSTAT_RUNNING, &vcpu->arch.sie_block->cpuflags); + + __vcpu_run(vcpu); + + if (vcpu->sigset_active) + sigprocmask(SIG_SETMASK, &sigsaved, NULL); + + vcpu_put(vcpu); + + vcpu->stat.exit_userspace++; + return 0; +} + +static int __guestcopy(struct kvm_vcpu *vcpu, u64 guestdest, const void *from, + unsigned long n, int prefix) +{ + if (prefix) + return copy_to_guest(vcpu, guestdest, from, n); + else + return copy_to_guest_absolute(vcpu, guestdest, from, n); +} + +/* + * store status at address + * we use have two special cases: + * KVM_S390_STORE_STATUS_NOADDR: -> 0x1200 on 64 bit + * KVM_S390_STORE_STATUS_PREFIXED: -> prefix + */ +int __kvm_s390_vcpu_store_status(struct kvm_vcpu *vcpu, unsigned long addr) +{ + const unsigned char archmode = 1; + int prefix; + + if (addr == KVM_S390_STORE_STATUS_NOADDR) { + if (copy_to_guest_absolute(vcpu, 163ul, &archmode, 1)) + return -EFAULT; + addr = SAVE_AREA_BASE; + prefix = 0; + } else if (addr == KVM_S390_STORE_STATUS_PREFIXED) { + if (copy_to_guest(vcpu, 163ul, &archmode, 1)) + return -EFAULT; + addr = SAVE_AREA_BASE; + prefix = 1; + } else + prefix = 0; + + if (__guestcopy(vcpu, addr + offsetof(struct save_area_s390x, fp_regs), + vcpu->arch.guest_fpregs.fprs, 128, prefix)) + return -EFAULT; + + if (__guestcopy(vcpu, addr + offsetof(struct save_area_s390x, gp_regs), + vcpu->arch.guest_gprs, 128, prefix)) + return -EFAULT; + + if (__guestcopy(vcpu, addr + offsetof(struct save_area_s390x, psw), + &vcpu->arch.sie_block->gpsw, 16, prefix)) + return -EFAULT; + + if (__guestcopy(vcpu, addr + offsetof(struct save_area_s390x, pref_reg), + &vcpu->arch.sie_block->prefix, 4, prefix)) + return -EFAULT; + + if (__guestcopy(vcpu, + addr + offsetof(struct save_area_s390x, fp_ctrl_reg), + &vcpu->arch.guest_fpregs.fpc, 4, prefix)) + return -EFAULT; + + if (__guestcopy(vcpu, addr + offsetof(struct save_area_s390x, tod_reg), + &vcpu->arch.sie_block->todpr, 4, prefix)) + return -EFAULT; + + if (__guestcopy(vcpu, addr + offsetof(struct save_area_s390x, timer), + &vcpu->arch.sie_block->cputm, 8, prefix)) + return -EFAULT; + + if (__guestcopy(vcpu, addr + offsetof(struct save_area_s390x, clk_cmp), + &vcpu->arch.sie_block->ckc, 8, prefix)) + return -EFAULT; + + if (__guestcopy(vcpu, addr + offsetof(struct save_area_s390x, acc_regs), + &vcpu->arch.guest_acrs, 64, prefix)) + return -EFAULT; + + if (__guestcopy(vcpu, + addr + offsetof(struct save_area_s390x, ctrl_regs), + &vcpu->arch.sie_block->gcr, 128, prefix)) + return -EFAULT; + return 0; +} + +static int kvm_s390_vcpu_store_status(struct kvm_vcpu *vcpu, unsigned long addr) +{ + int rc; + + vcpu_load(vcpu); + rc = __kvm_s390_vcpu_store_status(vcpu, addr); + vcpu_put(vcpu); + return rc; +} + +long kvm_arch_vcpu_ioctl(struct file *filp, + unsigned int ioctl, unsigned long arg) +{ + struct kvm_vcpu *vcpu = filp->private_data; + void __user *argp = (void __user *)arg; + + switch (ioctl) { + case KVM_S390_STORE_STATUS: + return kvm_s390_vcpu_store_status(vcpu, arg); + case KVM_S390_SET_INITIAL_PSW: { + psw_t psw; + + if (copy_from_user(&psw, argp, sizeof(psw))) + return -EFAULT; + return kvm_arch_vcpu_ioctl_set_initial_psw(vcpu, psw); + } + case KVM_S390_INITIAL_RESET: + return kvm_arch_vcpu_ioctl_initial_reset(vcpu); + default: + ; + } + return -EINVAL; +} + +/* Section: memory related */ +int kvm_arch_set_memory_region(struct kvm *kvm, + struct kvm_userspace_memory_region *mem, + struct kvm_memory_slot old, + int user_alloc) +{ + /* A few sanity checks. We can have exactly one memory slot which has + to start at guest virtual zero and which has to be located at a + page boundary in userland and which has to end at a page boundary. + The memory in userland is ok to be fragmented into various different + vmas. It is okay to mmap() and munmap() stuff in this slot after + doing this call at any time */ + + if (mem->slot) + return -EINVAL; + + if (mem->guest_phys_addr) + return -EINVAL; + + if (mem->userspace_addr & (PAGE_SIZE - 1)) + return -EINVAL; + + if (mem->memory_size & (PAGE_SIZE - 1)) + return -EINVAL; + + kvm->arch.guest_origin = mem->userspace_addr; + kvm->arch.guest_memsize = mem->memory_size; + + /* FIXME: we do want to interrupt running CPUs and update their memory + configuration now to avoid race conditions. But hey, changing the + memory layout while virtual CPUs are running is usually bad + programming practice. */ + + return 0; +} + +gfn_t unalias_gfn(struct kvm *kvm, gfn_t gfn) +{ + return gfn; +} + +static int __init kvm_s390_init(void) +{ + return kvm_init(NULL, sizeof(struct kvm_vcpu), THIS_MODULE); +} + +static void __exit kvm_s390_exit(void) +{ + kvm_exit(); +} + +module_init(kvm_s390_init); +module_exit(kvm_s390_exit); diff --git a/arch/s390/kvm/kvm-s390.h b/arch/s390/kvm/kvm-s390.h new file mode 100644 index 00000000000..ed64a22ca86 --- /dev/null +++ b/arch/s390/kvm/kvm-s390.h @@ -0,0 +1,29 @@ +/* + * kvm_s390.h - definition for kvm on s390 + * + * Copyright IBM Corp. 2008 + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License (version 2 only) + * as published by the Free Software Foundation. + * + * Author(s): Carsten Otte + * Christian Borntraeger + */ + +#ifndef ARCH_S390_KVM_S390_H +#define ARCH_S390_KVM_S390_H +#define VM_EVENT(d_kvm, d_loglevel, d_string, d_args...)\ +do { \ + debug_sprintf_event(d_kvm->arch.dbf, d_loglevel, d_string "\n", \ + d_args); \ +} while (0) + +#define VCPU_EVENT(d_vcpu, d_loglevel, d_string, d_args...)\ +do { \ + debug_sprintf_event(d_vcpu->kvm->arch.dbf, d_loglevel, \ + "%02d[%016lx-%016lx]: " d_string "\n", d_vcpu->vcpu_id, \ + d_vcpu->arch.sie_block->gpsw.mask, d_vcpu->arch.sie_block->gpsw.addr,\ + d_args); \ +} while (0) +#endif diff --git a/arch/s390/kvm/sie64a.S b/arch/s390/kvm/sie64a.S new file mode 100644 index 00000000000..934fd6a885f --- /dev/null +++ b/arch/s390/kvm/sie64a.S @@ -0,0 +1,47 @@ +/* + * sie64a.S - low level sie call + * + * Copyright IBM Corp. 2008 + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License (version 2 only) + * as published by the Free Software Foundation. + * + * Author(s): Heiko Carstens + */ + +#include +#include + +SP_R5 = 5 * 8 # offset into stackframe +SP_R6 = 6 * 8 + +/* + * sie64a calling convention: + * %r2 pointer to sie control block + * %r3 guest register save area + */ + .globl sie64a +sie64a: + lgr %r5,%r3 + stmg %r5,%r14,SP_R5(%r15) # save register on entry + lgr %r14,%r2 # pointer to sie control block + lmg %r0,%r13,0(%r3) # load guest gprs 0-13 +sie_inst: + sie 0(%r14) + lg %r14,SP_R5(%r15) + stmg %r0,%r13,0(%r14) # save guest gprs 0-13 + lghi %r2,0 + lmg %r6,%r14,SP_R6(%r15) + br %r14 + +sie_err: + lg %r14,SP_R5(%r15) + stmg %r0,%r13,0(%r14) # save guest gprs 0-13 + lghi %r2,-EFAULT + lmg %r6,%r14,SP_R6(%r15) + br %r14 + + .section __ex_table,"a" + .quad sie_inst,sie_err + .previous diff --git a/include/asm-s390/Kbuild b/include/asm-s390/Kbuild index e92b429d2be..13c9805349f 100644 --- a/include/asm-s390/Kbuild +++ b/include/asm-s390/Kbuild @@ -7,6 +7,7 @@ header-y += tape390.h header-y += ucontext.h header-y += vtoc.h header-y += zcrypt.h +header-y += kvm.h unifdef-y += cmb.h unifdef-y += debug.h diff --git a/include/asm-s390/kvm.h b/include/asm-s390/kvm.h index 573f2a35138..d74002f9579 100644 --- a/include/asm-s390/kvm.h +++ b/include/asm-s390/kvm.h @@ -1,6 +1,45 @@ #ifndef __LINUX_KVM_S390_H #define __LINUX_KVM_S390_H -/* s390 does not support KVM */ +/* + * asm-s390/kvm.h - KVM s390 specific structures and definitions + * + * Copyright IBM Corp. 2008 + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License (version 2 only) + * as published by the Free Software Foundation. + * + * Author(s): Carsten Otte + * Christian Borntraeger + */ +#include + +/* for KVM_GET_IRQCHIP and KVM_SET_IRQCHIP */ +struct kvm_pic_state { + /* no PIC for s390 */ +}; + +struct kvm_ioapic_state { + /* no IOAPIC for s390 */ +}; + +/* for KVM_GET_REGS and KVM_SET_REGS */ +struct kvm_regs { + /* general purpose regs for s390 */ + __u64 gprs[16]; +}; + +/* for KVM_GET_SREGS and KVM_SET_SREGS */ +struct kvm_sregs { + __u32 acrs[16]; + __u64 crs[16]; +}; + +/* for KVM_GET_FPU and KVM_SET_FPU */ +struct kvm_fpu { + __u32 fpc; + __u64 fprs[16]; +}; #endif diff --git a/include/asm-s390/kvm_host.h b/include/asm-s390/kvm_host.h new file mode 100644 index 00000000000..c9d65333310 --- /dev/null +++ b/include/asm-s390/kvm_host.h @@ -0,0 +1,127 @@ +/* + * asm-s390/kvm_host.h - definition for kernel virtual machines on s390 + * + * Copyright IBM Corp. 2008 + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License (version 2 only) + * as published by the Free Software Foundation. + * + * Author(s): Carsten Otte + */ + + +#ifndef ASM_KVM_HOST_H +#define ASM_KVM_HOST_H +#include +#include + +#define KVM_MAX_VCPUS 64 +#define KVM_MEMORY_SLOTS 32 +/* memory slots that does not exposed to userspace */ +#define KVM_PRIVATE_MEM_SLOTS 4 + +struct kvm_guest_debug { +}; + +struct sca_entry { + atomic_t scn; + __u64 reserved; + __u64 sda; + __u64 reserved2[2]; +} __attribute__((packed)); + + +struct sca_block { + __u64 ipte_control; + __u64 reserved[5]; + __u64 mcn; + __u64 reserved2; + struct sca_entry cpu[64]; +} __attribute__((packed)); + +#define KVM_PAGES_PER_HPAGE 256 + +#define CPUSTAT_HOST 0x80000000 +#define CPUSTAT_WAIT 0x10000000 +#define CPUSTAT_ECALL_PEND 0x08000000 +#define CPUSTAT_STOP_INT 0x04000000 +#define CPUSTAT_IO_INT 0x02000000 +#define CPUSTAT_EXT_INT 0x01000000 +#define CPUSTAT_RUNNING 0x00800000 +#define CPUSTAT_RETAINED 0x00400000 +#define CPUSTAT_TIMING_SUB 0x00020000 +#define CPUSTAT_SIE_SUB 0x00010000 +#define CPUSTAT_RRF 0x00008000 +#define CPUSTAT_SLSV 0x00004000 +#define CPUSTAT_SLSR 0x00002000 +#define CPUSTAT_ZARCH 0x00000800 +#define CPUSTAT_MCDS 0x00000100 +#define CPUSTAT_SM 0x00000080 +#define CPUSTAT_G 0x00000008 +#define CPUSTAT_J 0x00000002 +#define CPUSTAT_P 0x00000001 + +struct sie_block { + atomic_t cpuflags; /* 0x0000 */ + __u32 prefix; /* 0x0004 */ + __u8 reserved8[32]; /* 0x0008 */ + __u64 cputm; /* 0x0028 */ + __u64 ckc; /* 0x0030 */ + __u64 epoch; /* 0x0038 */ + __u8 reserved40[4]; /* 0x0040 */ + __u16 lctl; /* 0x0044 */ + __s16 icpua; /* 0x0046 */ + __u32 ictl; /* 0x0048 */ + __u32 eca; /* 0x004c */ + __u8 icptcode; /* 0x0050 */ + __u8 reserved51; /* 0x0051 */ + __u16 ihcpu; /* 0x0052 */ + __u8 reserved54[2]; /* 0x0054 */ + __u16 ipa; /* 0x0056 */ + __u32 ipb; /* 0x0058 */ + __u32 scaoh; /* 0x005c */ + __u8 reserved60; /* 0x0060 */ + __u8 ecb; /* 0x0061 */ + __u8 reserved62[2]; /* 0x0062 */ + __u32 scaol; /* 0x0064 */ + __u8 reserved68[4]; /* 0x0068 */ + __u32 todpr; /* 0x006c */ + __u8 reserved70[16]; /* 0x0070 */ + __u64 gmsor; /* 0x0080 */ + __u64 gmslm; /* 0x0088 */ + psw_t gpsw; /* 0x0090 */ + __u64 gg14; /* 0x00a0 */ + __u64 gg15; /* 0x00a8 */ + __u8 reservedb0[80]; /* 0x00b0 */ + __u64 gcr[16]; /* 0x0100 */ + __u64 gbea; /* 0x0180 */ + __u8 reserved188[120]; /* 0x0188 */ +} __attribute__((packed)); + +struct kvm_vcpu_stat { + u32 exit_userspace; +}; + +struct kvm_vcpu_arch { + struct sie_block *sie_block; + unsigned long guest_gprs[16]; + s390_fp_regs host_fpregs; + unsigned int host_acrs[NUM_ACRS]; + s390_fp_regs guest_fpregs; + unsigned int guest_acrs[NUM_ACRS]; +}; + +struct kvm_vm_stat { + u32 remote_tlb_flush; +}; + +struct kvm_arch{ + unsigned long guest_origin; + unsigned long guest_memsize; + struct sca_block *sca; + debug_info_t *dbf; +}; + +extern int sie64a(struct sie_block *, __u64 *); +#endif diff --git a/include/asm-s390/kvm_para.h b/include/asm-s390/kvm_para.h new file mode 100644 index 00000000000..e9bd3fbe878 --- /dev/null +++ b/include/asm-s390/kvm_para.h @@ -0,0 +1,30 @@ +/* + * asm-s390/kvm_para.h - definition for paravirtual devices on s390 + * + * Copyright IBM Corp. 2008 + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License (version 2 only) + * as published by the Free Software Foundation. + * + * Author(s): Christian Borntraeger + */ + +#ifndef __S390_KVM_PARA_H +#define __S390_KVM_PARA_H + +/* + * No hypercalls for KVM on s390 + */ + +static inline int kvm_para_available(void) +{ + return 0; +} + +static inline unsigned int kvm_arch_para_features(void) +{ + return 0; +} + +#endif /* __S390_KVM_PARA_H */ diff --git a/include/linux/kvm.h b/include/linux/kvm.h index 3bd38284bfe..2367ff0c5dd 100644 --- a/include/linux/kvm.h +++ b/include/linux/kvm.h @@ -205,6 +205,11 @@ struct kvm_vapic_addr { __u64 vapic_addr; }; +struct kvm_s390_psw { + __u64 mask; + __u64 addr; +}; + #define KVMIO 0xAE /* @@ -213,6 +218,8 @@ struct kvm_vapic_addr { #define KVM_GET_API_VERSION _IO(KVMIO, 0x00) #define KVM_CREATE_VM _IO(KVMIO, 0x01) /* returns a VM fd */ #define KVM_GET_MSR_INDEX_LIST _IOWR(KVMIO, 0x02, struct kvm_msr_list) + +#define KVM_S390_ENABLE_SIE _IO(KVMIO, 0x06) /* * Check if a kvm extension is available. Argument is extension number, * return is 1 (yes) or 0 (no, sorry). @@ -291,5 +298,13 @@ struct kvm_vapic_addr { #define KVM_TPR_ACCESS_REPORTING _IOWR(KVMIO, 0x92, struct kvm_tpr_access_ctl) /* Available with KVM_CAP_VAPIC */ #define KVM_SET_VAPIC_ADDR _IOW(KVMIO, 0x93, struct kvm_vapic_addr) +/* store status for s390 */ +#define KVM_S390_STORE_STATUS_NOADDR (-1ul) +#define KVM_S390_STORE_STATUS_PREFIXED (-2ul) +#define KVM_S390_STORE_STATUS _IOW(KVMIO, 0x95, unsigned long) +/* initial ipl psw for s390 */ +#define KVM_S390_SET_INITIAL_PSW _IOW(KVMIO, 0x96, struct kvm_s390_psw) +/* initial reset for s390 */ +#define KVM_S390_INITIAL_RESET _IO(KVMIO, 0x97) #endif -- cgit v1.2.3-70-g09d2 From 8f2abe6a1e525e878bdf58f68ccd146d543fde84 Mon Sep 17 00:00:00 2001 From: Christian Borntraeger Date: Tue, 25 Mar 2008 18:47:23 +0100 Subject: KVM: s390: sie intercept handling This path introduces handling of sie intercepts in three flavors: Intercepts are either handled completely in-kernel by kvm_handle_sie_intercept(), or passed to userspace with corresponding data in struct kvm_run in case kvm_handle_sie_intercept() returns -ENOTSUPP. In case of partial execution in kernel with the need of userspace support, kvm_handle_sie_intercept() may choose to set up struct kvm_run and return -EREMOTE. The trivial intercept reasons are handled in this patch: handle_noop() just does nothing for intercepts that don't require our support at all handle_stop() is called when a cpu enters stopped state, and it drops out to userland after updating our vcpu state handle_validity() faults in the cpu lowcore if needed, or passes the request to userland Acked-by: Martin Schwidefsky Signed-off-by: Christian Borntraeger Signed-off-by: Carsten Otte Signed-off-by: Avi Kivity --- arch/s390/kvm/Makefile | 2 +- arch/s390/kvm/intercept.c | 80 +++++++++++++++++++++++++++++++++++++++++++++ arch/s390/kvm/kvm-s390.c | 46 +++++++++++++++++++++++++- arch/s390/kvm/kvm-s390.h | 7 ++++ include/asm-s390/kvm_host.h | 4 +++ include/linux/kvm.h | 9 +++++ 6 files changed, 146 insertions(+), 2 deletions(-) create mode 100644 arch/s390/kvm/intercept.c (limited to 'arch/s390') diff --git a/arch/s390/kvm/Makefile b/arch/s390/kvm/Makefile index 0d8d1135a27..27882b35ef0 100644 --- a/arch/s390/kvm/Makefile +++ b/arch/s390/kvm/Makefile @@ -10,5 +10,5 @@ common-objs = $(addprefix ../../../virt/kvm/, kvm_main.o) EXTRA_CFLAGS += -Ivirt/kvm -Iarch/s390/kvm -kvm-objs := $(common-objs) kvm-s390.o sie64a.o +kvm-objs := $(common-objs) kvm-s390.o sie64a.o intercept.o obj-$(CONFIG_KVM) += kvm.o diff --git a/arch/s390/kvm/intercept.c b/arch/s390/kvm/intercept.c new file mode 100644 index 00000000000..e757230b982 --- /dev/null +++ b/arch/s390/kvm/intercept.c @@ -0,0 +1,80 @@ +/* + * intercept.c - in-kernel handling for sie intercepts + * + * Copyright IBM Corp. 2008 + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License (version 2 only) + * as published by the Free Software Foundation. + * + * Author(s): Carsten Otte + * Christian Borntraeger + */ + +#include +#include +#include + +#include + +#include "kvm-s390.h" + +static int handle_noop(struct kvm_vcpu *vcpu) +{ + switch (vcpu->arch.sie_block->icptcode) { + case 0x10: + vcpu->stat.exit_external_request++; + break; + case 0x14: + vcpu->stat.exit_external_interrupt++; + break; + default: + break; /* nothing */ + } + return 0; +} + +static int handle_stop(struct kvm_vcpu *vcpu) +{ + vcpu->stat.exit_stop_request++; + VCPU_EVENT(vcpu, 3, "%s", "cpu stopped"); + atomic_clear_mask(CPUSTAT_RUNNING, &vcpu->arch.sie_block->cpuflags); + return -ENOTSUPP; +} + +static int handle_validity(struct kvm_vcpu *vcpu) +{ + int viwhy = vcpu->arch.sie_block->ipb >> 16; + vcpu->stat.exit_validity++; + if (viwhy == 0x37) { + fault_in_pages_writeable((char __user *) + vcpu->kvm->arch.guest_origin + + vcpu->arch.sie_block->prefix, + PAGE_SIZE); + return 0; + } + VCPU_EVENT(vcpu, 2, "unhandled validity intercept code %d", + viwhy); + return -ENOTSUPP; +} + +static const intercept_handler_t intercept_funcs[0x48 >> 2] = { + [0x00 >> 2] = handle_noop, + [0x10 >> 2] = handle_noop, + [0x14 >> 2] = handle_noop, + [0x20 >> 2] = handle_validity, + [0x28 >> 2] = handle_stop, +}; + +int kvm_handle_sie_intercept(struct kvm_vcpu *vcpu) +{ + intercept_handler_t func; + u8 code = vcpu->arch.sie_block->icptcode; + + if (code & 3 || code > 0x48) + return -ENOTSUPP; + func = intercept_funcs[code >> 2]; + if (func) + return func(vcpu); + return -ENOTSUPP; +} diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c index 6e1e1d39ae1..a906499214b 100644 --- a/arch/s390/kvm/kvm-s390.c +++ b/arch/s390/kvm/kvm-s390.c @@ -23,12 +23,17 @@ #include #include +#include "kvm-s390.h" #include "gaccess.h" #define VCPU_STAT(x) offsetof(struct kvm_vcpu, stat.x), KVM_STAT_VCPU struct kvm_stats_debugfs_item debugfs_entries[] = { { "userspace_handled", VCPU_STAT(exit_userspace) }, + { "exit_validity", VCPU_STAT(exit_validity) }, + { "exit_stop_request", VCPU_STAT(exit_stop_request) }, + { "exit_external_request", VCPU_STAT(exit_external_request) }, + { "exit_external_interrupt", VCPU_STAT(exit_external_interrupt) }, { NULL } }; @@ -380,6 +385,7 @@ static void __vcpu_run(struct kvm_vcpu *vcpu) int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) { + int rc; sigset_t sigsaved; vcpu_load(vcpu); @@ -389,7 +395,45 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) atomic_set_mask(CPUSTAT_RUNNING, &vcpu->arch.sie_block->cpuflags); - __vcpu_run(vcpu); + switch (kvm_run->exit_reason) { + case KVM_EXIT_S390_SIEIC: + vcpu->arch.sie_block->gpsw.mask = kvm_run->s390_sieic.mask; + vcpu->arch.sie_block->gpsw.addr = kvm_run->s390_sieic.addr; + break; + case KVM_EXIT_UNKNOWN: + case KVM_EXIT_S390_RESET: + break; + default: + BUG(); + } + + might_sleep(); + + do { + __vcpu_run(vcpu); + + rc = kvm_handle_sie_intercept(vcpu); + } while (!signal_pending(current) && !rc); + + if (signal_pending(current) && !rc) + rc = -EINTR; + + if (rc == -ENOTSUPP) { + /* intercept cannot be handled in-kernel, prepare kvm-run */ + kvm_run->exit_reason = KVM_EXIT_S390_SIEIC; + kvm_run->s390_sieic.icptcode = vcpu->arch.sie_block->icptcode; + kvm_run->s390_sieic.mask = vcpu->arch.sie_block->gpsw.mask; + kvm_run->s390_sieic.addr = vcpu->arch.sie_block->gpsw.addr; + kvm_run->s390_sieic.ipa = vcpu->arch.sie_block->ipa; + kvm_run->s390_sieic.ipb = vcpu->arch.sie_block->ipb; + rc = 0; + } + + if (rc == -EREMOTE) { + /* intercept was handled, but userspace support is needed + * kvm_run has been prepared by the handler */ + rc = 0; + } if (vcpu->sigset_active) sigprocmask(SIG_SETMASK, &sigsaved, NULL); diff --git a/arch/s390/kvm/kvm-s390.h b/arch/s390/kvm/kvm-s390.h index ed64a22ca86..5b82527b7f8 100644 --- a/arch/s390/kvm/kvm-s390.h +++ b/arch/s390/kvm/kvm-s390.h @@ -13,6 +13,13 @@ #ifndef ARCH_S390_KVM_S390_H #define ARCH_S390_KVM_S390_H + +#include + +typedef int (*intercept_handler_t)(struct kvm_vcpu *vcpu); + +int kvm_handle_sie_intercept(struct kvm_vcpu *vcpu); + #define VM_EVENT(d_kvm, d_loglevel, d_string, d_args...)\ do { \ debug_sprintf_event(d_kvm->arch.dbf, d_loglevel, d_string "\n", \ diff --git a/include/asm-s390/kvm_host.h b/include/asm-s390/kvm_host.h index c9d65333310..8965b38d0a3 100644 --- a/include/asm-s390/kvm_host.h +++ b/include/asm-s390/kvm_host.h @@ -101,6 +101,10 @@ struct sie_block { struct kvm_vcpu_stat { u32 exit_userspace; + u32 exit_external_request; + u32 exit_external_interrupt; + u32 exit_stop_request; + u32 exit_validity; }; struct kvm_vcpu_arch { diff --git a/include/linux/kvm.h b/include/linux/kvm.h index 2367ff0c5dd..f2acd6b9ab4 100644 --- a/include/linux/kvm.h +++ b/include/linux/kvm.h @@ -74,6 +74,7 @@ struct kvm_irqchip { #define KVM_EXIT_INTR 10 #define KVM_EXIT_SET_TPR 11 #define KVM_EXIT_TPR_ACCESS 12 +#define KVM_EXIT_S390_SIEIC 13 /* for KVM_RUN, returned by mmap(vcpu_fd, offset=0) */ struct kvm_run { @@ -138,6 +139,14 @@ struct kvm_run { __u32 is_write; __u32 pad; } tpr_access; + /* KVM_EXIT_S390_SIEIC */ + struct { + __u8 icptcode; + __u64 mask; /* psw upper half */ + __u64 addr; /* psw lower half */ + __u16 ipa; + __u32 ipb; + } s390_sieic; /* Fix the size of the union. */ char padding[256]; }; -- cgit v1.2.3-70-g09d2 From ba5c1e9b6ceebdc39343cc03eb39f077abd3c571 Mon Sep 17 00:00:00 2001 From: Carsten Otte Date: Tue, 25 Mar 2008 18:47:26 +0100 Subject: KVM: s390: interrupt subsystem, cpu timer, waitpsw This patch contains the s390 interrupt subsystem (similar to in kernel apic) including timer interrupts (similar to in-kernel-pit) and enabled wait (similar to in kernel hlt). In order to achieve that, this patch also introduces intercept handling for instruction intercepts, and it implements load control instructions. This patch introduces an ioctl KVM_S390_INTERRUPT which is valid for both the vm file descriptors and the vcpu file descriptors. In case this ioctl is issued against a vm file descriptor, the interrupt is considered floating. Floating interrupts may be delivered to any virtual cpu in the configuration. The following interrupts are supported: SIGP STOP - interprocessor signal that stops a remote cpu SIGP SET PREFIX - interprocessor signal that sets the prefix register of a (stopped) remote cpu INT EMERGENCY - interprocessor interrupt, usually used to signal need_reshed and for smp_call_function() in the guest. PROGRAM INT - exception during program execution such as page fault, illegal instruction and friends RESTART - interprocessor signal that starts a stopped cpu INT VIRTIO - floating interrupt for virtio signalisation INT SERVICE - floating interrupt for signalisations from the system service processor struct kvm_s390_interrupt, which is submitted as ioctl parameter when injecting an interrupt, also carrys parameter data for interrupts along with the interrupt type. Interrupts on s390 usually have a state that represents the current operation, or identifies which device has caused the interruption on s390. kvm_s390_handle_wait() does handle waitpsw in two flavors: in case of a disabled wait (that is, disabled for interrupts), we exit to userspace. In case of an enabled wait we set up a timer that equals the cpu clock comparator value and sleep on a wait queue. [christian: change virtio interrupt to 0x2603] Acked-by: Martin Schwidefsky Signed-off-by: Heiko Carstens Signed-off-by: Carsten Otte Signed-off-by: Christian Borntraeger Signed-off-by: Avi Kivity --- arch/s390/kvm/Makefile | 2 +- arch/s390/kvm/intercept.c | 116 +++++++++ arch/s390/kvm/interrupt.c | 587 ++++++++++++++++++++++++++++++++++++++++++++ arch/s390/kvm/kvm-s390.c | 48 +++- arch/s390/kvm/kvm-s390.h | 15 ++ include/asm-s390/kvm_host.h | 75 ++++++ include/linux/kvm.h | 17 ++ 7 files changed, 857 insertions(+), 3 deletions(-) create mode 100644 arch/s390/kvm/interrupt.c (limited to 'arch/s390') diff --git a/arch/s390/kvm/Makefile b/arch/s390/kvm/Makefile index 27882b35ef0..7275a1aa4ee 100644 --- a/arch/s390/kvm/Makefile +++ b/arch/s390/kvm/Makefile @@ -10,5 +10,5 @@ common-objs = $(addprefix ../../../virt/kvm/, kvm_main.o) EXTRA_CFLAGS += -Ivirt/kvm -Iarch/s390/kvm -kvm-objs := $(common-objs) kvm-s390.o sie64a.o intercept.o +kvm-objs := $(common-objs) kvm-s390.o sie64a.o intercept.o interrupt.o obj-$(CONFIG_KVM) += kvm.o diff --git a/arch/s390/kvm/intercept.c b/arch/s390/kvm/intercept.c index e757230b982..7f7347b5f34 100644 --- a/arch/s390/kvm/intercept.c +++ b/arch/s390/kvm/intercept.c @@ -18,6 +18,86 @@ #include #include "kvm-s390.h" +#include "gaccess.h" + +static int handle_lctg(struct kvm_vcpu *vcpu) +{ + int reg1 = (vcpu->arch.sie_block->ipa & 0x00f0) >> 4; + int reg3 = vcpu->arch.sie_block->ipa & 0x000f; + int base2 = vcpu->arch.sie_block->ipb >> 28; + int disp2 = ((vcpu->arch.sie_block->ipb & 0x0fff0000) >> 16) + + ((vcpu->arch.sie_block->ipb & 0xff00) << 4); + u64 useraddr; + int reg, rc; + + vcpu->stat.instruction_lctg++; + if ((vcpu->arch.sie_block->ipb & 0xff) != 0x2f) + return -ENOTSUPP; + + useraddr = disp2; + if (base2) + useraddr += vcpu->arch.guest_gprs[base2]; + + reg = reg1; + + VCPU_EVENT(vcpu, 5, "lctg r1:%x, r3:%x,b2:%x,d2:%x", reg1, reg3, base2, + disp2); + + do { + rc = get_guest_u64(vcpu, useraddr, + &vcpu->arch.sie_block->gcr[reg]); + if (rc == -EFAULT) { + kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING); + break; + } + useraddr += 8; + if (reg == reg3) + break; + reg = (reg + 1) % 16; + } while (1); + return 0; +} + +static int handle_lctl(struct kvm_vcpu *vcpu) +{ + int reg1 = (vcpu->arch.sie_block->ipa & 0x00f0) >> 4; + int reg3 = vcpu->arch.sie_block->ipa & 0x000f; + int base2 = vcpu->arch.sie_block->ipb >> 28; + int disp2 = ((vcpu->arch.sie_block->ipb & 0x0fff0000) >> 16); + u64 useraddr; + u32 val = 0; + int reg, rc; + + vcpu->stat.instruction_lctl++; + + useraddr = disp2; + if (base2) + useraddr += vcpu->arch.guest_gprs[base2]; + + VCPU_EVENT(vcpu, 5, "lctl r1:%x, r3:%x,b2:%x,d2:%x", reg1, reg3, base2, + disp2); + + reg = reg1; + do { + rc = get_guest_u32(vcpu, useraddr, &val); + if (rc == -EFAULT) { + kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING); + break; + } + vcpu->arch.sie_block->gcr[reg] &= 0xffffffff00000000ul; + vcpu->arch.sie_block->gcr[reg] |= val; + useraddr += 4; + if (reg == reg3) + break; + reg = (reg + 1) % 16; + } while (1); + return 0; +} + +static intercept_handler_t instruction_handlers[256] = { + [0xb7] = handle_lctl, + [0xeb] = handle_lctg, +}; static int handle_noop(struct kvm_vcpu *vcpu) { @@ -58,10 +138,46 @@ static int handle_validity(struct kvm_vcpu *vcpu) return -ENOTSUPP; } +static int handle_instruction(struct kvm_vcpu *vcpu) +{ + intercept_handler_t handler; + + vcpu->stat.exit_instruction++; + handler = instruction_handlers[vcpu->arch.sie_block->ipa >> 8]; + if (handler) + return handler(vcpu); + return -ENOTSUPP; +} + +static int handle_prog(struct kvm_vcpu *vcpu) +{ + vcpu->stat.exit_program_interruption++; + return kvm_s390_inject_program_int(vcpu, vcpu->arch.sie_block->iprcc); +} + +static int handle_instruction_and_prog(struct kvm_vcpu *vcpu) +{ + int rc, rc2; + + vcpu->stat.exit_instr_and_program++; + rc = handle_instruction(vcpu); + rc2 = handle_prog(vcpu); + + if (rc == -ENOTSUPP) + vcpu->arch.sie_block->icptcode = 0x04; + if (rc) + return rc; + return rc2; +} + static const intercept_handler_t intercept_funcs[0x48 >> 2] = { [0x00 >> 2] = handle_noop, + [0x04 >> 2] = handle_instruction, + [0x08 >> 2] = handle_prog, + [0x0C >> 2] = handle_instruction_and_prog, [0x10 >> 2] = handle_noop, [0x14 >> 2] = handle_noop, + [0x1C >> 2] = kvm_s390_handle_wait, [0x20 >> 2] = handle_validity, [0x28 >> 2] = handle_stop, }; diff --git a/arch/s390/kvm/interrupt.c b/arch/s390/kvm/interrupt.c new file mode 100644 index 00000000000..f62588cb75f --- /dev/null +++ b/arch/s390/kvm/interrupt.c @@ -0,0 +1,587 @@ +/* + * interrupt.c - handling kvm guest interrupts + * + * Copyright IBM Corp. 2008 + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License (version 2 only) + * as published by the Free Software Foundation. + * + * Author(s): Carsten Otte + */ + +#include +#include +#include +#include "kvm-s390.h" +#include "gaccess.h" + +static int psw_extint_disabled(struct kvm_vcpu *vcpu) +{ + return !(vcpu->arch.sie_block->gpsw.mask & PSW_MASK_EXT); +} + +static int psw_interrupts_disabled(struct kvm_vcpu *vcpu) +{ + if ((vcpu->arch.sie_block->gpsw.mask & PSW_MASK_PER) || + (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_IO) || + (vcpu->arch.sie_block->gpsw.mask & PSW_MASK_EXT)) + return 0; + return 1; +} + +static int __interrupt_is_deliverable(struct kvm_vcpu *vcpu, + struct interrupt_info *inti) +{ + switch (inti->type) { + case KVM_S390_INT_EMERGENCY: + if (psw_extint_disabled(vcpu)) + return 0; + if (vcpu->arch.sie_block->gcr[0] & 0x4000ul) + return 1; + return 0; + case KVM_S390_INT_SERVICE: + if (psw_extint_disabled(vcpu)) + return 0; + if (vcpu->arch.sie_block->gcr[0] & 0x200ul) + return 1; + return 0; + case KVM_S390_INT_VIRTIO: + if (psw_extint_disabled(vcpu)) + return 0; + if (vcpu->arch.sie_block->gcr[0] & 0x200ul) + return 1; + return 0; + case KVM_S390_PROGRAM_INT: + case KVM_S390_SIGP_STOP: + case KVM_S390_SIGP_SET_PREFIX: + case KVM_S390_RESTART: + return 1; + default: + BUG(); + } + return 0; +} + +static void __set_cpu_idle(struct kvm_vcpu *vcpu) +{ + BUG_ON(vcpu->vcpu_id > KVM_MAX_VCPUS - 1); + atomic_set_mask(CPUSTAT_WAIT, &vcpu->arch.sie_block->cpuflags); + set_bit(vcpu->vcpu_id, vcpu->arch.local_int.float_int->idle_mask); +} + +static void __unset_cpu_idle(struct kvm_vcpu *vcpu) +{ + BUG_ON(vcpu->vcpu_id > KVM_MAX_VCPUS - 1); + atomic_clear_mask(CPUSTAT_WAIT, &vcpu->arch.sie_block->cpuflags); + clear_bit(vcpu->vcpu_id, vcpu->arch.local_int.float_int->idle_mask); +} + +static void __reset_intercept_indicators(struct kvm_vcpu *vcpu) +{ + atomic_clear_mask(CPUSTAT_ECALL_PEND | + CPUSTAT_IO_INT | CPUSTAT_EXT_INT | CPUSTAT_STOP_INT, + &vcpu->arch.sie_block->cpuflags); + vcpu->arch.sie_block->lctl = 0x0000; +} + +static void __set_cpuflag(struct kvm_vcpu *vcpu, u32 flag) +{ + atomic_set_mask(flag, &vcpu->arch.sie_block->cpuflags); +} + +static void __set_intercept_indicator(struct kvm_vcpu *vcpu, + struct interrupt_info *inti) +{ + switch (inti->type) { + case KVM_S390_INT_EMERGENCY: + case KVM_S390_INT_SERVICE: + case KVM_S390_INT_VIRTIO: + if (psw_extint_disabled(vcpu)) + __set_cpuflag(vcpu, CPUSTAT_EXT_INT); + else + vcpu->arch.sie_block->lctl |= LCTL_CR0; + break; + case KVM_S390_SIGP_STOP: + __set_cpuflag(vcpu, CPUSTAT_STOP_INT); + break; + default: + BUG(); + } +} + +static void __do_deliver_interrupt(struct kvm_vcpu *vcpu, + struct interrupt_info *inti) +{ + const unsigned short table[] = { 2, 4, 4, 6 }; + int rc, exception = 0; + + switch (inti->type) { + case KVM_S390_INT_EMERGENCY: + VCPU_EVENT(vcpu, 4, "%s", "interrupt: sigp emerg"); + vcpu->stat.deliver_emergency_signal++; + rc = put_guest_u16(vcpu, __LC_EXT_INT_CODE, 0x1201); + if (rc == -EFAULT) + exception = 1; + + rc = copy_to_guest(vcpu, __LC_EXT_OLD_PSW, + &vcpu->arch.sie_block->gpsw, sizeof(psw_t)); + if (rc == -EFAULT) + exception = 1; + + rc = copy_from_guest(vcpu, &vcpu->arch.sie_block->gpsw, + __LC_EXT_NEW_PSW, sizeof(psw_t)); + if (rc == -EFAULT) + exception = 1; + break; + + case KVM_S390_INT_SERVICE: + VCPU_EVENT(vcpu, 4, "interrupt: sclp parm:%x", + inti->ext.ext_params); + vcpu->stat.deliver_service_signal++; + rc = put_guest_u16(vcpu, __LC_EXT_INT_CODE, 0x2401); + if (rc == -EFAULT) + exception = 1; + + rc = copy_to_guest(vcpu, __LC_EXT_OLD_PSW, + &vcpu->arch.sie_block->gpsw, sizeof(psw_t)); + if (rc == -EFAULT) + exception = 1; + + rc = copy_from_guest(vcpu, &vcpu->arch.sie_block->gpsw, + __LC_EXT_NEW_PSW, sizeof(psw_t)); + if (rc == -EFAULT) + exception = 1; + + rc = put_guest_u32(vcpu, __LC_EXT_PARAMS, inti->ext.ext_params); + if (rc == -EFAULT) + exception = 1; + break; + + case KVM_S390_INT_VIRTIO: + VCPU_EVENT(vcpu, 4, "interrupt: virtio parm:%x,parm64:%lx", + inti->ext.ext_params, inti->ext.ext_params2); + vcpu->stat.deliver_virtio_interrupt++; + rc = put_guest_u16(vcpu, __LC_EXT_INT_CODE, 0x2603); + if (rc == -EFAULT) + exception = 1; + + rc = put_guest_u16(vcpu, __LC_CPU_ADDRESS, 0x0d00); + if (rc == -EFAULT) + exception = 1; + + rc = copy_to_guest(vcpu, __LC_EXT_OLD_PSW, + &vcpu->arch.sie_block->gpsw, sizeof(psw_t)); + if (rc == -EFAULT) + exception = 1; + + rc = copy_from_guest(vcpu, &vcpu->arch.sie_block->gpsw, + __LC_EXT_NEW_PSW, sizeof(psw_t)); + if (rc == -EFAULT) + exception = 1; + + rc = put_guest_u32(vcpu, __LC_EXT_PARAMS, inti->ext.ext_params); + if (rc == -EFAULT) + exception = 1; + + rc = put_guest_u64(vcpu, __LC_PFAULT_INTPARM, + inti->ext.ext_params2); + if (rc == -EFAULT) + exception = 1; + break; + + case KVM_S390_SIGP_STOP: + VCPU_EVENT(vcpu, 4, "%s", "interrupt: cpu stop"); + vcpu->stat.deliver_stop_signal++; + __set_intercept_indicator(vcpu, inti); + break; + + case KVM_S390_SIGP_SET_PREFIX: + VCPU_EVENT(vcpu, 4, "interrupt: set prefix to %x", + inti->prefix.address); + vcpu->stat.deliver_prefix_signal++; + vcpu->arch.sie_block->prefix = inti->prefix.address; + vcpu->arch.sie_block->ihcpu = 0xffff; + break; + + case KVM_S390_RESTART: + VCPU_EVENT(vcpu, 4, "%s", "interrupt: cpu restart"); + vcpu->stat.deliver_restart_signal++; + rc = copy_to_guest(vcpu, offsetof(struct _lowcore, + restart_old_psw), &vcpu->arch.sie_block->gpsw, sizeof(psw_t)); + if (rc == -EFAULT) + exception = 1; + + rc = copy_from_guest(vcpu, &vcpu->arch.sie_block->gpsw, + offsetof(struct _lowcore, restart_psw), sizeof(psw_t)); + if (rc == -EFAULT) + exception = 1; + break; + + case KVM_S390_PROGRAM_INT: + VCPU_EVENT(vcpu, 4, "interrupt: pgm check code:%x, ilc:%x", + inti->pgm.code, + table[vcpu->arch.sie_block->ipa >> 14]); + vcpu->stat.deliver_program_int++; + rc = put_guest_u16(vcpu, __LC_PGM_INT_CODE, inti->pgm.code); + if (rc == -EFAULT) + exception = 1; + + rc = put_guest_u16(vcpu, __LC_PGM_ILC, + table[vcpu->arch.sie_block->ipa >> 14]); + if (rc == -EFAULT) + exception = 1; + + rc = copy_to_guest(vcpu, __LC_PGM_OLD_PSW, + &vcpu->arch.sie_block->gpsw, sizeof(psw_t)); + if (rc == -EFAULT) + exception = 1; + + rc = copy_from_guest(vcpu, &vcpu->arch.sie_block->gpsw, + __LC_PGM_NEW_PSW, sizeof(psw_t)); + if (rc == -EFAULT) + exception = 1; + break; + + default: + BUG(); + } + + if (exception) { + VCPU_EVENT(vcpu, 1, "%s", "program exception while delivering" + " interrupt"); + kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING); + if (inti->type == KVM_S390_PROGRAM_INT) { + printk(KERN_WARNING "kvm: recursive program check\n"); + BUG(); + } + } +} + +static int __try_deliver_ckc_interrupt(struct kvm_vcpu *vcpu) +{ + int rc, exception = 0; + + if (psw_extint_disabled(vcpu)) + return 0; + if (!(vcpu->arch.sie_block->gcr[0] & 0x800ul)) + return 0; + rc = put_guest_u16(vcpu, __LC_EXT_INT_CODE, 0x1004); + if (rc == -EFAULT) + exception = 1; + rc = copy_to_guest(vcpu, __LC_EXT_OLD_PSW, + &vcpu->arch.sie_block->gpsw, sizeof(psw_t)); + if (rc == -EFAULT) + exception = 1; + rc = copy_from_guest(vcpu, &vcpu->arch.sie_block->gpsw, + __LC_EXT_NEW_PSW, sizeof(psw_t)); + if (rc == -EFAULT) + exception = 1; + + if (exception) { + VCPU_EVENT(vcpu, 1, "%s", "program exception while delivering" \ + " ckc interrupt"); + kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING); + return 0; + } + + return 1; +} + +int kvm_cpu_has_interrupt(struct kvm_vcpu *vcpu) +{ + struct local_interrupt *li = &vcpu->arch.local_int; + struct float_interrupt *fi = vcpu->arch.local_int.float_int; + struct interrupt_info *inti; + int rc = 0; + + if (atomic_read(&li->active)) { + spin_lock_bh(&li->lock); + list_for_each_entry(inti, &li->list, list) + if (__interrupt_is_deliverable(vcpu, inti)) { + rc = 1; + break; + } + spin_unlock_bh(&li->lock); + } + + if ((!rc) && atomic_read(&fi->active)) { + spin_lock_bh(&fi->lock); + list_for_each_entry(inti, &fi->list, list) + if (__interrupt_is_deliverable(vcpu, inti)) { + rc = 1; + break; + } + spin_unlock_bh(&fi->lock); + } + + if ((!rc) && (vcpu->arch.sie_block->ckc < + get_clock() + vcpu->arch.sie_block->epoch)) { + if ((!psw_extint_disabled(vcpu)) && + (vcpu->arch.sie_block->gcr[0] & 0x800ul)) + rc = 1; + } + + return rc; +} + +int kvm_s390_handle_wait(struct kvm_vcpu *vcpu) +{ + u64 now, sltime; + DECLARE_WAITQUEUE(wait, current); + + vcpu->stat.exit_wait_state++; + if (kvm_cpu_has_interrupt(vcpu)) + return 0; + + if (psw_interrupts_disabled(vcpu)) { + VCPU_EVENT(vcpu, 3, "%s", "disabled wait"); + __unset_cpu_idle(vcpu); + return -ENOTSUPP; /* disabled wait */ + } + + if (psw_extint_disabled(vcpu) || + (!(vcpu->arch.sie_block->gcr[0] & 0x800ul))) { + VCPU_EVENT(vcpu, 3, "%s", "enabled wait w/o timer"); + goto no_timer; + } + + now = get_clock() + vcpu->arch.sie_block->epoch; + if (vcpu->arch.sie_block->ckc < now) { + __unset_cpu_idle(vcpu); + return 0; + } + + sltime = (vcpu->arch.sie_block->ckc - now) / (0xf4240000ul / HZ) + 1; + + vcpu->arch.ckc_timer.expires = jiffies + sltime; + + add_timer(&vcpu->arch.ckc_timer); + VCPU_EVENT(vcpu, 5, "enabled wait timer:%lx jiffies", sltime); +no_timer: + spin_lock_bh(&vcpu->arch.local_int.float_int->lock); + spin_lock_bh(&vcpu->arch.local_int.lock); + __set_cpu_idle(vcpu); + vcpu->arch.local_int.timer_due = 0; + add_wait_queue(&vcpu->arch.local_int.wq, &wait); + while (list_empty(&vcpu->arch.local_int.list) && + list_empty(&vcpu->arch.local_int.float_int->list) && + (!vcpu->arch.local_int.timer_due) && + !signal_pending(current)) { + set_current_state(TASK_INTERRUPTIBLE); + spin_unlock_bh(&vcpu->arch.local_int.lock); + spin_unlock_bh(&vcpu->arch.local_int.float_int->lock); + vcpu_put(vcpu); + schedule(); + vcpu_load(vcpu); + spin_lock_bh(&vcpu->arch.local_int.float_int->lock); + spin_lock_bh(&vcpu->arch.local_int.lock); + } + __unset_cpu_idle(vcpu); + __set_current_state(TASK_RUNNING); + remove_wait_queue(&vcpu->wq, &wait); + spin_unlock_bh(&vcpu->arch.local_int.lock); + spin_unlock_bh(&vcpu->arch.local_int.float_int->lock); + del_timer(&vcpu->arch.ckc_timer); + return 0; +} + +void kvm_s390_idle_wakeup(unsigned long data) +{ + struct kvm_vcpu *vcpu = (struct kvm_vcpu *)data; + + spin_lock_bh(&vcpu->arch.local_int.lock); + vcpu->arch.local_int.timer_due = 1; + if (waitqueue_active(&vcpu->arch.local_int.wq)) + wake_up_interruptible(&vcpu->arch.local_int.wq); + spin_unlock_bh(&vcpu->arch.local_int.lock); +} + + +void kvm_s390_deliver_pending_interrupts(struct kvm_vcpu *vcpu) +{ + struct local_interrupt *li = &vcpu->arch.local_int; + struct float_interrupt *fi = vcpu->arch.local_int.float_int; + struct interrupt_info *n, *inti = NULL; + int deliver; + + __reset_intercept_indicators(vcpu); + if (atomic_read(&li->active)) { + do { + deliver = 0; + spin_lock_bh(&li->lock); + list_for_each_entry_safe(inti, n, &li->list, list) { + if (__interrupt_is_deliverable(vcpu, inti)) { + list_del(&inti->list); + deliver = 1; + break; + } + __set_intercept_indicator(vcpu, inti); + } + if (list_empty(&li->list)) + atomic_set(&li->active, 0); + spin_unlock_bh(&li->lock); + if (deliver) { + __do_deliver_interrupt(vcpu, inti); + kfree(inti); + } + } while (deliver); + } + + if ((vcpu->arch.sie_block->ckc < + get_clock() + vcpu->arch.sie_block->epoch)) + __try_deliver_ckc_interrupt(vcpu); + + if (atomic_read(&fi->active)) { + do { + deliver = 0; + spin_lock_bh(&fi->lock); + list_for_each_entry_safe(inti, n, &fi->list, list) { + if (__interrupt_is_deliverable(vcpu, inti)) { + list_del(&inti->list); + deliver = 1; + break; + } + __set_intercept_indicator(vcpu, inti); + } + if (list_empty(&fi->list)) + atomic_set(&fi->active, 0); + spin_unlock_bh(&fi->lock); + if (deliver) { + __do_deliver_interrupt(vcpu, inti); + kfree(inti); + } + } while (deliver); + } +} + +int kvm_s390_inject_program_int(struct kvm_vcpu *vcpu, u16 code) +{ + struct local_interrupt *li = &vcpu->arch.local_int; + struct interrupt_info *inti; + + inti = kzalloc(sizeof(*inti), GFP_KERNEL); + if (!inti) + return -ENOMEM; + + inti->type = KVM_S390_PROGRAM_INT;; + inti->pgm.code = code; + + VCPU_EVENT(vcpu, 3, "inject: program check %d (from kernel)", code); + spin_lock_bh(&li->lock); + list_add(&inti->list, &li->list); + atomic_set(&li->active, 1); + BUG_ON(waitqueue_active(&li->wq)); + spin_unlock_bh(&li->lock); + return 0; +} + +int kvm_s390_inject_vm(struct kvm *kvm, + struct kvm_s390_interrupt *s390int) +{ + struct local_interrupt *li; + struct float_interrupt *fi; + struct interrupt_info *inti; + int sigcpu; + + inti = kzalloc(sizeof(*inti), GFP_KERNEL); + if (!inti) + return -ENOMEM; + + switch (s390int->type) { + case KVM_S390_INT_VIRTIO: + VM_EVENT(kvm, 5, "inject: virtio parm:%x,parm64:%lx", + s390int->parm, s390int->parm64); + inti->type = s390int->type; + inti->ext.ext_params = s390int->parm; + inti->ext.ext_params2 = s390int->parm64; + break; + case KVM_S390_INT_SERVICE: + VM_EVENT(kvm, 5, "inject: sclp parm:%x", s390int->parm); + inti->type = s390int->type; + inti->ext.ext_params = s390int->parm; + break; + case KVM_S390_PROGRAM_INT: + case KVM_S390_SIGP_STOP: + case KVM_S390_INT_EMERGENCY: + default: + kfree(inti); + return -EINVAL; + } + + mutex_lock(&kvm->lock); + fi = &kvm->arch.float_int; + spin_lock_bh(&fi->lock); + list_add_tail(&inti->list, &fi->list); + atomic_set(&fi->active, 1); + sigcpu = find_first_bit(fi->idle_mask, KVM_MAX_VCPUS); + if (sigcpu == KVM_MAX_VCPUS) { + do { + sigcpu = fi->next_rr_cpu++; + if (sigcpu == KVM_MAX_VCPUS) + sigcpu = fi->next_rr_cpu = 0; + } while (fi->local_int[sigcpu] == NULL); + } + li = fi->local_int[sigcpu]; + spin_lock_bh(&li->lock); + atomic_set_mask(CPUSTAT_EXT_INT, li->cpuflags); + if (waitqueue_active(&li->wq)) + wake_up_interruptible(&li->wq); + spin_unlock_bh(&li->lock); + spin_unlock_bh(&fi->lock); + mutex_unlock(&kvm->lock); + return 0; +} + +int kvm_s390_inject_vcpu(struct kvm_vcpu *vcpu, + struct kvm_s390_interrupt *s390int) +{ + struct local_interrupt *li; + struct interrupt_info *inti; + + inti = kzalloc(sizeof(*inti), GFP_KERNEL); + if (!inti) + return -ENOMEM; + + switch (s390int->type) { + case KVM_S390_PROGRAM_INT: + if (s390int->parm & 0xffff0000) { + kfree(inti); + return -EINVAL; + } + inti->type = s390int->type; + inti->pgm.code = s390int->parm; + VCPU_EVENT(vcpu, 3, "inject: program check %d (from user)", + s390int->parm); + break; + case KVM_S390_SIGP_STOP: + case KVM_S390_RESTART: + case KVM_S390_SIGP_SET_PREFIX: + case KVM_S390_INT_EMERGENCY: + VCPU_EVENT(vcpu, 3, "inject: type %x", s390int->type); + inti->type = s390int->type; + break; + case KVM_S390_INT_VIRTIO: + case KVM_S390_INT_SERVICE: + default: + kfree(inti); + return -EINVAL; + } + + mutex_lock(&vcpu->kvm->lock); + li = &vcpu->arch.local_int; + spin_lock_bh(&li->lock); + if (inti->type == KVM_S390_PROGRAM_INT) + list_add(&inti->list, &li->list); + else + list_add_tail(&inti->list, &li->list); + atomic_set(&li->active, 1); + if (inti->type == KVM_S390_SIGP_STOP) + li->action_bits |= ACTION_STOP_ON_STOP; + atomic_set_mask(CPUSTAT_EXT_INT, li->cpuflags); + if (waitqueue_active(&li->wq)) + wake_up_interruptible(&vcpu->arch.local_int.wq); + spin_unlock_bh(&li->lock); + mutex_unlock(&vcpu->kvm->lock); + return 0; +} diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c index a906499214b..5e3473c9a63 100644 --- a/arch/s390/kvm/kvm-s390.c +++ b/arch/s390/kvm/kvm-s390.c @@ -20,6 +20,7 @@ #include #include #include +#include #include #include @@ -34,6 +35,19 @@ struct kvm_stats_debugfs_item debugfs_entries[] = { { "exit_stop_request", VCPU_STAT(exit_stop_request) }, { "exit_external_request", VCPU_STAT(exit_external_request) }, { "exit_external_interrupt", VCPU_STAT(exit_external_interrupt) }, + { "exit_instruction", VCPU_STAT(exit_instruction) }, + { "exit_program_interruption", VCPU_STAT(exit_program_interruption) }, + { "exit_instr_and_program_int", VCPU_STAT(exit_instr_and_program) }, + { "instruction_lctg", VCPU_STAT(instruction_lctg) }, + { "instruction_lctl", VCPU_STAT(instruction_lctl) }, + { "deliver_emergency_signal", VCPU_STAT(deliver_emergency_signal) }, + { "deliver_service_signal", VCPU_STAT(deliver_service_signal) }, + { "deliver_virtio_interrupt", VCPU_STAT(deliver_virtio_interrupt) }, + { "deliver_stop_signal", VCPU_STAT(deliver_stop_signal) }, + { "deliver_prefix_signal", VCPU_STAT(deliver_prefix_signal) }, + { "deliver_restart_signal", VCPU_STAT(deliver_restart_signal) }, + { "deliver_program_interruption", VCPU_STAT(deliver_program_int) }, + { "exit_wait_state", VCPU_STAT(exit_wait_state) }, { NULL } }; @@ -106,6 +120,15 @@ long kvm_arch_vm_ioctl(struct file *filp, int r; switch (ioctl) { + case KVM_S390_INTERRUPT: { + struct kvm_s390_interrupt s390int; + + r = -EFAULT; + if (copy_from_user(&s390int, argp, sizeof(s390int))) + break; + r = kvm_s390_inject_vm(kvm, &s390int); + break; + } default: r = -EINVAL; } @@ -138,6 +161,9 @@ struct kvm *kvm_arch_create_vm(void) if (!kvm->arch.dbf) goto out_nodbf; + spin_lock_init(&kvm->arch.float_int.lock); + INIT_LIST_HEAD(&kvm->arch.float_int.list); + debug_register_view(kvm->arch.dbf, &debug_sprintf_view); VM_EVENT(kvm, 3, "%s", "vm created"); @@ -218,7 +244,8 @@ int kvm_arch_vcpu_setup(struct kvm_vcpu *vcpu) vcpu->arch.sie_block->gmsor = 0x000000000000; vcpu->arch.sie_block->ecb = 2; vcpu->arch.sie_block->eca = 0xC1002001U; - + setup_timer(&vcpu->arch.ckc_timer, kvm_s390_idle_wakeup, + (unsigned long) vcpu); return 0; } @@ -243,6 +270,14 @@ struct kvm_vcpu *kvm_arch_vcpu_create(struct kvm *kvm, vcpu->arch.sie_block->scaoh = (__u32)(((__u64)kvm->arch.sca) >> 32); vcpu->arch.sie_block->scaol = (__u32)(__u64)kvm->arch.sca; + spin_lock_init(&vcpu->arch.local_int.lock); + INIT_LIST_HEAD(&vcpu->arch.local_int.list); + vcpu->arch.local_int.float_int = &kvm->arch.float_int; + spin_lock_bh(&kvm->arch.float_int.lock); + kvm->arch.float_int.local_int[id] = &vcpu->arch.local_int; + init_waitqueue_head(&vcpu->arch.local_int.wq); + spin_unlock_bh(&kvm->arch.float_int.lock); + rc = kvm_vcpu_init(vcpu, kvm, id); if (rc) goto out_free_cpu; @@ -395,6 +430,8 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) atomic_set_mask(CPUSTAT_RUNNING, &vcpu->arch.sie_block->cpuflags); + BUG_ON(vcpu->kvm->arch.float_int.local_int[vcpu->vcpu_id] == NULL); + switch (kvm_run->exit_reason) { case KVM_EXIT_S390_SIEIC: vcpu->arch.sie_block->gpsw.mask = kvm_run->s390_sieic.mask; @@ -410,8 +447,8 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) might_sleep(); do { + kvm_s390_deliver_pending_interrupts(vcpu); __vcpu_run(vcpu); - rc = kvm_handle_sie_intercept(vcpu); } while (!signal_pending(current) && !rc); @@ -538,6 +575,13 @@ long kvm_arch_vcpu_ioctl(struct file *filp, void __user *argp = (void __user *)arg; switch (ioctl) { + case KVM_S390_INTERRUPT: { + struct kvm_s390_interrupt s390int; + + if (copy_from_user(&s390int, argp, sizeof(s390int))) + return -EFAULT; + return kvm_s390_inject_vcpu(vcpu, &s390int); + } case KVM_S390_STORE_STATUS: return kvm_s390_vcpu_store_status(vcpu, arg); case KVM_S390_SET_INITIAL_PSW: { diff --git a/arch/s390/kvm/kvm-s390.h b/arch/s390/kvm/kvm-s390.h index 5b82527b7f8..8df745bc08d 100644 --- a/arch/s390/kvm/kvm-s390.h +++ b/arch/s390/kvm/kvm-s390.h @@ -14,6 +14,7 @@ #ifndef ARCH_S390_KVM_S390_H #define ARCH_S390_KVM_S390_H +#include #include typedef int (*intercept_handler_t)(struct kvm_vcpu *vcpu); @@ -33,4 +34,18 @@ do { \ d_vcpu->arch.sie_block->gpsw.mask, d_vcpu->arch.sie_block->gpsw.addr,\ d_args); \ } while (0) + +static inline int __cpu_is_stopped(struct kvm_vcpu *vcpu) +{ + return atomic_read(&vcpu->arch.sie_block->cpuflags) & CPUSTAT_STOP_INT; +} + +int kvm_s390_handle_wait(struct kvm_vcpu *vcpu); +void kvm_s390_idle_wakeup(unsigned long data); +void kvm_s390_deliver_pending_interrupts(struct kvm_vcpu *vcpu); +int kvm_s390_inject_vm(struct kvm *kvm, + struct kvm_s390_interrupt *s390int); +int kvm_s390_inject_vcpu(struct kvm_vcpu *vcpu, + struct kvm_s390_interrupt *s390int); +int kvm_s390_inject_program_int(struct kvm_vcpu *vcpu, u16 code); #endif diff --git a/include/asm-s390/kvm_host.h b/include/asm-s390/kvm_host.h index 8965b38d0a3..4fe19305888 100644 --- a/include/asm-s390/kvm_host.h +++ b/include/asm-s390/kvm_host.h @@ -70,6 +70,7 @@ struct sie_block { __u64 ckc; /* 0x0030 */ __u64 epoch; /* 0x0038 */ __u8 reserved40[4]; /* 0x0040 */ +#define LCTL_CR0 0x8000 __u16 lctl; /* 0x0044 */ __s16 icpua; /* 0x0046 */ __u32 ictl; /* 0x0048 */ @@ -105,8 +106,79 @@ struct kvm_vcpu_stat { u32 exit_external_interrupt; u32 exit_stop_request; u32 exit_validity; + u32 exit_instruction; + u32 instruction_lctl; + u32 instruction_lctg; + u32 exit_program_interruption; + u32 exit_instr_and_program; + u32 deliver_emergency_signal; + u32 deliver_service_signal; + u32 deliver_virtio_interrupt; + u32 deliver_stop_signal; + u32 deliver_prefix_signal; + u32 deliver_restart_signal; + u32 deliver_program_int; + u32 exit_wait_state; }; +struct io_info { + __u16 subchannel_id; /* 0x0b8 */ + __u16 subchannel_nr; /* 0x0ba */ + __u32 io_int_parm; /* 0x0bc */ + __u32 io_int_word; /* 0x0c0 */ +}; + +struct ext_info { + __u32 ext_params; + __u64 ext_params2; +}; + +#define PGM_OPERATION 0x01 +#define PGM_PRIVILEGED_OPERATION 0x02 +#define PGM_EXECUTE 0x03 +#define PGM_PROTECTION 0x04 +#define PGM_ADDRESSING 0x05 +#define PGM_SPECIFICATION 0x06 +#define PGM_DATA 0x07 + +struct pgm_info { + __u16 code; +}; + +struct prefix_info { + __u32 address; +}; + +struct interrupt_info { + struct list_head list; + u64 type; + union { + struct io_info io; + struct ext_info ext; + struct pgm_info pgm; + struct prefix_info prefix; + }; +}; + +struct local_interrupt { + spinlock_t lock; + struct list_head list; + atomic_t active; + struct float_interrupt *float_int; + int timer_due; /* event indicator for waitqueue below */ + wait_queue_head_t wq; +}; + +struct float_interrupt { + spinlock_t lock; + struct list_head list; + atomic_t active; + int next_rr_cpu; + unsigned long idle_mask [(64 + sizeof(long) - 1) / sizeof(long)]; + struct local_interrupt *local_int[64]; +}; + + struct kvm_vcpu_arch { struct sie_block *sie_block; unsigned long guest_gprs[16]; @@ -114,6 +186,8 @@ struct kvm_vcpu_arch { unsigned int host_acrs[NUM_ACRS]; s390_fp_regs guest_fpregs; unsigned int guest_acrs[NUM_ACRS]; + struct local_interrupt local_int; + struct timer_list ckc_timer; }; struct kvm_vm_stat { @@ -125,6 +199,7 @@ struct kvm_arch{ unsigned long guest_memsize; struct sca_block *sca; debug_info_t *dbf; + struct float_interrupt float_int; }; extern int sie64a(struct sie_block *, __u64 *); diff --git a/include/linux/kvm.h b/include/linux/kvm.h index f2acd6b9ab4..029f0284a2f 100644 --- a/include/linux/kvm.h +++ b/include/linux/kvm.h @@ -219,6 +219,21 @@ struct kvm_s390_psw { __u64 addr; }; +/* valid values for type in kvm_s390_interrupt */ +#define KVM_S390_SIGP_STOP 0xfffe0000u +#define KVM_S390_PROGRAM_INT 0xfffe0001u +#define KVM_S390_SIGP_SET_PREFIX 0xfffe0002u +#define KVM_S390_RESTART 0xfffe0003u +#define KVM_S390_INT_VIRTIO 0xffff2603u +#define KVM_S390_INT_SERVICE 0xffff2401u +#define KVM_S390_INT_EMERGENCY 0xffff1201u + +struct kvm_s390_interrupt { + __u32 type; + __u32 parm; + __u64 parm64; +}; + #define KVMIO 0xAE /* @@ -307,6 +322,8 @@ struct kvm_s390_psw { #define KVM_TPR_ACCESS_REPORTING _IOWR(KVMIO, 0x92, struct kvm_tpr_access_ctl) /* Available with KVM_CAP_VAPIC */ #define KVM_SET_VAPIC_ADDR _IOW(KVMIO, 0x93, struct kvm_vapic_addr) +/* valid for virtual machine (for floating interrupt)_and_ vcpu */ +#define KVM_S390_INTERRUPT _IOW(KVMIO, 0x94, struct kvm_s390_interrupt) /* store status for s390 */ #define KVM_S390_STORE_STATUS_NOADDR (-1ul) #define KVM_S390_STORE_STATUS_PREFIXED (-2ul) -- cgit v1.2.3-70-g09d2 From 453423dce2785b8e22077e3b3eeecb4f60fe3470 Mon Sep 17 00:00:00 2001 From: Christian Borntraeger Date: Tue, 25 Mar 2008 18:47:29 +0100 Subject: KVM: s390: intercepts for privileged instructions This patch introduces in-kernel handling of some intercepts for privileged instructions: handle_set_prefix() sets the prefix register of the local cpu handle_store_prefix() stores the content of the prefix register to memory handle_store_cpu_address() stores the cpu number of the current cpu to memory handle_skey() just decrements the instruction address and retries handle_stsch() delivers condition code 3 "operation not supported" handle_chsc() same here handle_stfl() stores the facility list which contains the capabilities of the cpu handle_stidp() stores cpu type/model/revision and such handle_stsi() stores information about the system topology Acked-by: Martin Schwidefsky Signed-off-by: Christian Borntraeger Signed-off-by: Heiko Carstens Signed-off-by: Carsten Otte Signed-off-by: Avi Kivity --- arch/s390/kvm/Makefile | 2 +- arch/s390/kvm/intercept.c | 1 + arch/s390/kvm/kvm-s390.c | 11 ++ arch/s390/kvm/kvm-s390.h | 3 + arch/s390/kvm/priv.c | 323 ++++++++++++++++++++++++++++++++++++++++++++ include/asm-s390/kvm_host.h | 13 ++ 6 files changed, 352 insertions(+), 1 deletion(-) create mode 100644 arch/s390/kvm/priv.c (limited to 'arch/s390') diff --git a/arch/s390/kvm/Makefile b/arch/s390/kvm/Makefile index 7275a1aa4ee..82dde1ef4d4 100644 --- a/arch/s390/kvm/Makefile +++ b/arch/s390/kvm/Makefile @@ -10,5 +10,5 @@ common-objs = $(addprefix ../../../virt/kvm/, kvm_main.o) EXTRA_CFLAGS += -Ivirt/kvm -Iarch/s390/kvm -kvm-objs := $(common-objs) kvm-s390.o sie64a.o intercept.o interrupt.o +kvm-objs := $(common-objs) kvm-s390.o sie64a.o intercept.o interrupt.o priv.o obj-$(CONFIG_KVM) += kvm.o diff --git a/arch/s390/kvm/intercept.c b/arch/s390/kvm/intercept.c index 7f7347b5f34..7a20d63a2eb 100644 --- a/arch/s390/kvm/intercept.c +++ b/arch/s390/kvm/intercept.c @@ -95,6 +95,7 @@ static int handle_lctl(struct kvm_vcpu *vcpu) } static intercept_handler_t instruction_handlers[256] = { + [0xb2] = kvm_s390_handle_priv, [0xb7] = handle_lctl, [0xeb] = handle_lctg, }; diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c index 5e3473c9a63..5a17176fb64 100644 --- a/arch/s390/kvm/kvm-s390.c +++ b/arch/s390/kvm/kvm-s390.c @@ -48,6 +48,15 @@ struct kvm_stats_debugfs_item debugfs_entries[] = { { "deliver_restart_signal", VCPU_STAT(deliver_restart_signal) }, { "deliver_program_interruption", VCPU_STAT(deliver_program_int) }, { "exit_wait_state", VCPU_STAT(exit_wait_state) }, + { "instruction_stidp", VCPU_STAT(instruction_stidp) }, + { "instruction_spx", VCPU_STAT(instruction_spx) }, + { "instruction_stpx", VCPU_STAT(instruction_stpx) }, + { "instruction_stap", VCPU_STAT(instruction_stap) }, + { "instruction_storage_key", VCPU_STAT(instruction_storage_key) }, + { "instruction_stsch", VCPU_STAT(instruction_stsch) }, + { "instruction_chsc", VCPU_STAT(instruction_chsc) }, + { "instruction_stsi", VCPU_STAT(instruction_stsi) }, + { "instruction_stfl", VCPU_STAT(instruction_stfl) }, { NULL } }; @@ -246,6 +255,8 @@ int kvm_arch_vcpu_setup(struct kvm_vcpu *vcpu) vcpu->arch.sie_block->eca = 0xC1002001U; setup_timer(&vcpu->arch.ckc_timer, kvm_s390_idle_wakeup, (unsigned long) vcpu); + get_cpu_id(&vcpu->arch.cpu_id); + vcpu->arch.cpu_id.version = 0xfe; return 0; } diff --git a/arch/s390/kvm/kvm-s390.h b/arch/s390/kvm/kvm-s390.h index 8df745bc08d..50f96b3bcf6 100644 --- a/arch/s390/kvm/kvm-s390.h +++ b/arch/s390/kvm/kvm-s390.h @@ -48,4 +48,7 @@ int kvm_s390_inject_vm(struct kvm *kvm, int kvm_s390_inject_vcpu(struct kvm_vcpu *vcpu, struct kvm_s390_interrupt *s390int); int kvm_s390_inject_program_int(struct kvm_vcpu *vcpu, u16 code); + +/* implemented in priv.c */ +int kvm_s390_handle_priv(struct kvm_vcpu *vcpu); #endif diff --git a/arch/s390/kvm/priv.c b/arch/s390/kvm/priv.c new file mode 100644 index 00000000000..c97e904ca3b --- /dev/null +++ b/arch/s390/kvm/priv.c @@ -0,0 +1,323 @@ +/* + * priv.c - handling privileged instructions + * + * Copyright IBM Corp. 2008 + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License (version 2 only) + * as published by the Free Software Foundation. + * + * Author(s): Carsten Otte + * Christian Borntraeger + */ + +#include +#include +#include +#include +#include +#include +#include "gaccess.h" +#include "kvm-s390.h" + +static int handle_set_prefix(struct kvm_vcpu *vcpu) +{ + int base2 = vcpu->arch.sie_block->ipb >> 28; + int disp2 = ((vcpu->arch.sie_block->ipb & 0x0fff0000) >> 16); + u64 operand2; + u32 address = 0; + u8 tmp; + + vcpu->stat.instruction_spx++; + + operand2 = disp2; + if (base2) + operand2 += vcpu->arch.guest_gprs[base2]; + + /* must be word boundary */ + if (operand2 & 3) { + kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); + goto out; + } + + /* get the value */ + if (get_guest_u32(vcpu, operand2, &address)) { + kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING); + goto out; + } + + address = address & 0x7fffe000u; + + /* make sure that the new value is valid memory */ + if (copy_from_guest_absolute(vcpu, &tmp, address, 1) || + (copy_from_guest_absolute(vcpu, &tmp, address + PAGE_SIZE, 1))) { + kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING); + goto out; + } + + vcpu->arch.sie_block->prefix = address; + vcpu->arch.sie_block->ihcpu = 0xffff; + + VCPU_EVENT(vcpu, 5, "setting prefix to %x", address); +out: + return 0; +} + +static int handle_store_prefix(struct kvm_vcpu *vcpu) +{ + int base2 = vcpu->arch.sie_block->ipb >> 28; + int disp2 = ((vcpu->arch.sie_block->ipb & 0x0fff0000) >> 16); + u64 operand2; + u32 address; + + vcpu->stat.instruction_stpx++; + operand2 = disp2; + if (base2) + operand2 += vcpu->arch.guest_gprs[base2]; + + /* must be word boundary */ + if (operand2 & 3) { + kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); + goto out; + } + + address = vcpu->arch.sie_block->prefix; + address = address & 0x7fffe000u; + + /* get the value */ + if (put_guest_u32(vcpu, operand2, address)) { + kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING); + goto out; + } + + VCPU_EVENT(vcpu, 5, "storing prefix to %x", address); +out: + return 0; +} + +static int handle_store_cpu_address(struct kvm_vcpu *vcpu) +{ + int base2 = vcpu->arch.sie_block->ipb >> 28; + int disp2 = ((vcpu->arch.sie_block->ipb & 0x0fff0000) >> 16); + u64 useraddr; + int rc; + + vcpu->stat.instruction_stap++; + useraddr = disp2; + if (base2) + useraddr += vcpu->arch.guest_gprs[base2]; + + if (useraddr & 1) { + kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); + goto out; + } + + rc = put_guest_u16(vcpu, useraddr, vcpu->vcpu_id); + if (rc == -EFAULT) { + kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING); + goto out; + } + + VCPU_EVENT(vcpu, 5, "storing cpu address to %lx", useraddr); +out: + return 0; +} + +static int handle_skey(struct kvm_vcpu *vcpu) +{ + vcpu->stat.instruction_storage_key++; + vcpu->arch.sie_block->gpsw.addr -= 4; + VCPU_EVENT(vcpu, 4, "%s", "retrying storage key operation"); + return 0; +} + +static int handle_stsch(struct kvm_vcpu *vcpu) +{ + vcpu->stat.instruction_stsch++; + VCPU_EVENT(vcpu, 4, "%s", "store subchannel - CC3"); + /* condition code 3 */ + vcpu->arch.sie_block->gpsw.mask &= ~(3ul << 44); + vcpu->arch.sie_block->gpsw.mask |= (3 & 3ul) << 44; + return 0; +} + +static int handle_chsc(struct kvm_vcpu *vcpu) +{ + vcpu->stat.instruction_chsc++; + VCPU_EVENT(vcpu, 4, "%s", "channel subsystem call - CC3"); + /* condition code 3 */ + vcpu->arch.sie_block->gpsw.mask &= ~(3ul << 44); + vcpu->arch.sie_block->gpsw.mask |= (3 & 3ul) << 44; + return 0; +} + +static unsigned int stfl(void) +{ + asm volatile( + " .insn s,0xb2b10000,0(0)\n" /* stfl */ + "0:\n" + EX_TABLE(0b, 0b)); + return S390_lowcore.stfl_fac_list; +} + +static int handle_stfl(struct kvm_vcpu *vcpu) +{ + unsigned int facility_list = stfl(); + int rc; + + vcpu->stat.instruction_stfl++; + facility_list &= ~(1UL<<24); /* no stfle */ + + rc = copy_to_guest(vcpu, offsetof(struct _lowcore, stfl_fac_list), + &facility_list, sizeof(facility_list)); + if (rc == -EFAULT) + kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING); + else + VCPU_EVENT(vcpu, 5, "store facility list value %x", + facility_list); + return 0; +} + +static int handle_stidp(struct kvm_vcpu *vcpu) +{ + int base2 = vcpu->arch.sie_block->ipb >> 28; + int disp2 = ((vcpu->arch.sie_block->ipb & 0x0fff0000) >> 16); + u64 operand2; + int rc; + + vcpu->stat.instruction_stidp++; + operand2 = disp2; + if (base2) + operand2 += vcpu->arch.guest_gprs[base2]; + + if (operand2 & 7) { + kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); + goto out; + } + + rc = put_guest_u64(vcpu, operand2, vcpu->arch.stidp_data); + if (rc == -EFAULT) { + kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING); + goto out; + } + + VCPU_EVENT(vcpu, 5, "%s", "store cpu id"); +out: + return 0; +} + +static void handle_stsi_3_2_2(struct kvm_vcpu *vcpu, struct sysinfo_3_2_2 *mem) +{ + struct float_interrupt *fi = &vcpu->kvm->arch.float_int; + int cpus = 0; + int n; + + spin_lock_bh(&fi->lock); + for (n = 0; n < KVM_MAX_VCPUS; n++) + if (fi->local_int[n]) + cpus++; + spin_unlock_bh(&fi->lock); + + /* deal with other level 3 hypervisors */ + if (stsi(mem, 3, 2, 2) == -ENOSYS) + mem->count = 0; + if (mem->count < 8) + mem->count++; + for (n = mem->count - 1; n > 0 ; n--) + memcpy(&mem->vm[n], &mem->vm[n - 1], sizeof(mem->vm[0])); + + mem->vm[0].cpus_total = cpus; + mem->vm[0].cpus_configured = cpus; + mem->vm[0].cpus_standby = 0; + mem->vm[0].cpus_reserved = 0; + mem->vm[0].caf = 1000; + memcpy(mem->vm[0].name, "KVMguest", 8); + ASCEBC(mem->vm[0].name, 8); + memcpy(mem->vm[0].cpi, "KVM/Linux ", 16); + ASCEBC(mem->vm[0].cpi, 16); +} + +static int handle_stsi(struct kvm_vcpu *vcpu) +{ + int fc = (vcpu->arch.guest_gprs[0] & 0xf0000000) >> 28; + int sel1 = vcpu->arch.guest_gprs[0] & 0xff; + int sel2 = vcpu->arch.guest_gprs[1] & 0xffff; + int base2 = vcpu->arch.sie_block->ipb >> 28; + int disp2 = ((vcpu->arch.sie_block->ipb & 0x0fff0000) >> 16); + u64 operand2; + unsigned long mem; + + vcpu->stat.instruction_stsi++; + VCPU_EVENT(vcpu, 4, "stsi: fc: %x sel1: %x sel2: %x", fc, sel1, sel2); + + operand2 = disp2; + if (base2) + operand2 += vcpu->arch.guest_gprs[base2]; + + if (operand2 & 0xfff && fc > 0) + return kvm_s390_inject_program_int(vcpu, PGM_SPECIFICATION); + + switch (fc) { + case 0: + vcpu->arch.guest_gprs[0] = 3 << 28; + vcpu->arch.sie_block->gpsw.mask &= ~(3ul << 44); + return 0; + case 1: /* same handling for 1 and 2 */ + case 2: + mem = get_zeroed_page(GFP_KERNEL); + if (!mem) + goto out_fail; + if (stsi((void *) mem, fc, sel1, sel2) == -ENOSYS) + goto out_mem; + break; + case 3: + if (sel1 != 2 || sel2 != 2) + goto out_fail; + mem = get_zeroed_page(GFP_KERNEL); + if (!mem) + goto out_fail; + handle_stsi_3_2_2(vcpu, (void *) mem); + break; + default: + goto out_fail; + } + + if (copy_to_guest_absolute(vcpu, operand2, (void *) mem, PAGE_SIZE)) { + kvm_s390_inject_program_int(vcpu, PGM_ADDRESSING); + goto out_mem; + } + free_page(mem); + vcpu->arch.sie_block->gpsw.mask &= ~(3ul << 44); + vcpu->arch.guest_gprs[0] = 0; + return 0; +out_mem: + free_page(mem); +out_fail: + /* condition code 3 */ + vcpu->arch.sie_block->gpsw.mask |= 3ul << 44; + return 0; +} + +static intercept_handler_t priv_handlers[256] = { + [0x02] = handle_stidp, + [0x10] = handle_set_prefix, + [0x11] = handle_store_prefix, + [0x12] = handle_store_cpu_address, + [0x29] = handle_skey, + [0x2a] = handle_skey, + [0x2b] = handle_skey, + [0x34] = handle_stsch, + [0x5f] = handle_chsc, + [0x7d] = handle_stsi, + [0xb1] = handle_stfl, +}; + +int kvm_s390_handle_priv(struct kvm_vcpu *vcpu) +{ + intercept_handler_t handler; + + handler = priv_handlers[vcpu->arch.sie_block->ipa & 0x00ff]; + if (handler) + return handler(vcpu); + return -ENOTSUPP; +} diff --git a/include/asm-s390/kvm_host.h b/include/asm-s390/kvm_host.h index 4fe19305888..2eaf6fede02 100644 --- a/include/asm-s390/kvm_host.h +++ b/include/asm-s390/kvm_host.h @@ -119,6 +119,15 @@ struct kvm_vcpu_stat { u32 deliver_restart_signal; u32 deliver_program_int; u32 exit_wait_state; + u32 instruction_stidp; + u32 instruction_spx; + u32 instruction_stpx; + u32 instruction_stap; + u32 instruction_storage_key; + u32 instruction_stsch; + u32 instruction_chsc; + u32 instruction_stsi; + u32 instruction_stfl; }; struct io_info { @@ -188,6 +197,10 @@ struct kvm_vcpu_arch { unsigned int guest_acrs[NUM_ACRS]; struct local_interrupt local_int; struct timer_list ckc_timer; + union { + cpuid_t cpu_id; + u64 stidp_data; + }; }; struct kvm_vm_stat { -- cgit v1.2.3-70-g09d2 From 5288fbf0ef041ba0e8b4dcb2df4536b5e3a48b32 Mon Sep 17 00:00:00 2001 From: Christian Borntraeger Date: Tue, 25 Mar 2008 18:47:31 +0100 Subject: KVM: s390: interprocessor communication via sigp This patch introduces in-kernel handling of _some_ sigp interprocessor signals (similar to ipi). kvm_s390_handle_sigp() decodes the sigp instruction and calls individual handlers depending on the operation requested: - sigp sense tries to retrieve information such as existence or running state of the remote cpu - sigp emergency sends an external interrupt to the remove cpu - sigp stop stops a remove cpu - sigp stop store status stops a remote cpu, and stores its entire internal state to the cpus lowcore - sigp set arch sets the architecture mode of the remote cpu. setting to ESAME (s390x 64bit) is accepted, setting to ESA/S390 (s390, 31 or 24 bit) is denied, all others are passed to userland - sigp set prefix sets the prefix register of a remote cpu For implementation of this, the stop intercept indication starts to get reused on purpose: a set of action bits defines what to do once a cpu gets stopped: ACTION_STOP_ON_STOP really stops the cpu when a stop intercept is recognized ACTION_STORE_ON_STOP stores the cpu status to lowcore when a stop intercept is recognized Acked-by: Martin Schwidefsky Signed-off-by: Christian Borntraeger Signed-off-by: Carsten Otte Signed-off-by: Carsten Otte Signed-off-by: Avi Kivity --- arch/s390/kvm/Makefile | 2 +- arch/s390/kvm/intercept.c | 22 +++- arch/s390/kvm/kvm-s390.c | 7 ++ arch/s390/kvm/kvm-s390.h | 7 ++ arch/s390/kvm/sigp.c | 288 ++++++++++++++++++++++++++++++++++++++++++++ include/asm-s390/kvm_host.h | 12 ++ 6 files changed, 335 insertions(+), 3 deletions(-) create mode 100644 arch/s390/kvm/sigp.c (limited to 'arch/s390') diff --git a/arch/s390/kvm/Makefile b/arch/s390/kvm/Makefile index 82dde1ef4d4..f3bf11a88bc 100644 --- a/arch/s390/kvm/Makefile +++ b/arch/s390/kvm/Makefile @@ -10,5 +10,5 @@ common-objs = $(addprefix ../../../virt/kvm/, kvm_main.o) EXTRA_CFLAGS += -Ivirt/kvm -Iarch/s390/kvm -kvm-objs := $(common-objs) kvm-s390.o sie64a.o intercept.o interrupt.o priv.o +kvm-objs := $(common-objs) kvm-s390.o sie64a.o intercept.o interrupt.o priv.o sigp.o obj-$(CONFIG_KVM) += kvm.o diff --git a/arch/s390/kvm/intercept.c b/arch/s390/kvm/intercept.c index 7a20d63a2eb..9f0d8b23943 100644 --- a/arch/s390/kvm/intercept.c +++ b/arch/s390/kvm/intercept.c @@ -95,6 +95,7 @@ static int handle_lctl(struct kvm_vcpu *vcpu) } static intercept_handler_t instruction_handlers[256] = { + [0xae] = kvm_s390_handle_sigp, [0xb2] = kvm_s390_handle_priv, [0xb7] = handle_lctl, [0xeb] = handle_lctg, @@ -117,10 +118,27 @@ static int handle_noop(struct kvm_vcpu *vcpu) static int handle_stop(struct kvm_vcpu *vcpu) { + int rc; + vcpu->stat.exit_stop_request++; - VCPU_EVENT(vcpu, 3, "%s", "cpu stopped"); atomic_clear_mask(CPUSTAT_RUNNING, &vcpu->arch.sie_block->cpuflags); - return -ENOTSUPP; + spin_lock_bh(&vcpu->arch.local_int.lock); + if (vcpu->arch.local_int.action_bits & ACTION_STORE_ON_STOP) { + vcpu->arch.local_int.action_bits &= ~ACTION_STORE_ON_STOP; + rc = __kvm_s390_vcpu_store_status(vcpu, + KVM_S390_STORE_STATUS_NOADDR); + if (rc >= 0) + rc = -ENOTSUPP; + } + + if (vcpu->arch.local_int.action_bits & ACTION_STOP_ON_STOP) { + vcpu->arch.local_int.action_bits &= ~ACTION_STOP_ON_STOP; + VCPU_EVENT(vcpu, 3, "%s", "cpu stopped"); + rc = -ENOTSUPP; + } else + rc = 0; + spin_unlock_bh(&vcpu->arch.local_int.lock); + return rc; } static int handle_validity(struct kvm_vcpu *vcpu) diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c index 5a17176fb64..c632180739e 100644 --- a/arch/s390/kvm/kvm-s390.c +++ b/arch/s390/kvm/kvm-s390.c @@ -57,6 +57,12 @@ struct kvm_stats_debugfs_item debugfs_entries[] = { { "instruction_chsc", VCPU_STAT(instruction_chsc) }, { "instruction_stsi", VCPU_STAT(instruction_stsi) }, { "instruction_stfl", VCPU_STAT(instruction_stfl) }, + { "instruction_sigp_sense", VCPU_STAT(instruction_sigp_sense) }, + { "instruction_sigp_emergency", VCPU_STAT(instruction_sigp_emergency) }, + { "instruction_sigp_stop", VCPU_STAT(instruction_sigp_stop) }, + { "instruction_sigp_set_arch", VCPU_STAT(instruction_sigp_arch) }, + { "instruction_sigp_set_prefix", VCPU_STAT(instruction_sigp_prefix) }, + { "instruction_sigp_restart", VCPU_STAT(instruction_sigp_restart) }, { NULL } }; @@ -287,6 +293,7 @@ struct kvm_vcpu *kvm_arch_vcpu_create(struct kvm *kvm, spin_lock_bh(&kvm->arch.float_int.lock); kvm->arch.float_int.local_int[id] = &vcpu->arch.local_int; init_waitqueue_head(&vcpu->arch.local_int.wq); + vcpu->arch.local_int.cpuflags = &vcpu->arch.sie_block->cpuflags; spin_unlock_bh(&kvm->arch.float_int.lock); rc = kvm_vcpu_init(vcpu, kvm, id); diff --git a/arch/s390/kvm/kvm-s390.h b/arch/s390/kvm/kvm-s390.h index 50f96b3bcf6..e6e5756a0e0 100644 --- a/arch/s390/kvm/kvm-s390.h +++ b/arch/s390/kvm/kvm-s390.h @@ -51,4 +51,11 @@ int kvm_s390_inject_program_int(struct kvm_vcpu *vcpu, u16 code); /* implemented in priv.c */ int kvm_s390_handle_priv(struct kvm_vcpu *vcpu); + +/* implemented in sigp.c */ +int kvm_s390_handle_sigp(struct kvm_vcpu *vcpu); + +/* implemented in kvm-s390.c */ +int __kvm_s390_vcpu_store_status(struct kvm_vcpu *vcpu, + unsigned long addr); #endif diff --git a/arch/s390/kvm/sigp.c b/arch/s390/kvm/sigp.c new file mode 100644 index 00000000000..0a236acfb5f --- /dev/null +++ b/arch/s390/kvm/sigp.c @@ -0,0 +1,288 @@ +/* + * sigp.c - handlinge interprocessor communication + * + * Copyright IBM Corp. 2008 + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License (version 2 only) + * as published by the Free Software Foundation. + * + * Author(s): Carsten Otte + * Christian Borntraeger + */ + +#include +#include +#include "gaccess.h" +#include "kvm-s390.h" + +/* sigp order codes */ +#define SIGP_SENSE 0x01 +#define SIGP_EXTERNAL_CALL 0x02 +#define SIGP_EMERGENCY 0x03 +#define SIGP_START 0x04 +#define SIGP_STOP 0x05 +#define SIGP_RESTART 0x06 +#define SIGP_STOP_STORE_STATUS 0x09 +#define SIGP_INITIAL_CPU_RESET 0x0b +#define SIGP_CPU_RESET 0x0c +#define SIGP_SET_PREFIX 0x0d +#define SIGP_STORE_STATUS_ADDR 0x0e +#define SIGP_SET_ARCH 0x12 + +/* cpu status bits */ +#define SIGP_STAT_EQUIPMENT_CHECK 0x80000000UL +#define SIGP_STAT_INCORRECT_STATE 0x00000200UL +#define SIGP_STAT_INVALID_PARAMETER 0x00000100UL +#define SIGP_STAT_EXT_CALL_PENDING 0x00000080UL +#define SIGP_STAT_STOPPED 0x00000040UL +#define SIGP_STAT_OPERATOR_INTERV 0x00000020UL +#define SIGP_STAT_CHECK_STOP 0x00000010UL +#define SIGP_STAT_INOPERATIVE 0x00000004UL +#define SIGP_STAT_INVALID_ORDER 0x00000002UL +#define SIGP_STAT_RECEIVER_CHECK 0x00000001UL + + +static int __sigp_sense(struct kvm_vcpu *vcpu, u16 cpu_addr, u64 *reg) +{ + struct float_interrupt *fi = &vcpu->kvm->arch.float_int; + int rc; + + if (cpu_addr >= KVM_MAX_VCPUS) + return 3; /* not operational */ + + spin_lock_bh(&fi->lock); + if (fi->local_int[cpu_addr] == NULL) + rc = 3; /* not operational */ + else if (atomic_read(fi->local_int[cpu_addr]->cpuflags) + & CPUSTAT_RUNNING) { + *reg &= 0xffffffff00000000UL; + rc = 1; /* status stored */ + } else { + *reg &= 0xffffffff00000000UL; + *reg |= SIGP_STAT_STOPPED; + rc = 1; /* status stored */ + } + spin_unlock_bh(&fi->lock); + + VCPU_EVENT(vcpu, 4, "sensed status of cpu %x rc %x", cpu_addr, rc); + return rc; +} + +static int __sigp_emergency(struct kvm_vcpu *vcpu, u16 cpu_addr) +{ + struct float_interrupt *fi = &vcpu->kvm->arch.float_int; + struct local_interrupt *li; + struct interrupt_info *inti; + int rc; + + if (cpu_addr >= KVM_MAX_VCPUS) + return 3; /* not operational */ + + inti = kzalloc(sizeof(*inti), GFP_KERNEL); + if (!inti) + return -ENOMEM; + + inti->type = KVM_S390_INT_EMERGENCY; + + spin_lock_bh(&fi->lock); + li = fi->local_int[cpu_addr]; + if (li == NULL) { + rc = 3; /* not operational */ + kfree(inti); + goto unlock; + } + spin_lock_bh(&li->lock); + list_add_tail(&inti->list, &li->list); + atomic_set(&li->active, 1); + atomic_set_mask(CPUSTAT_EXT_INT, li->cpuflags); + if (waitqueue_active(&li->wq)) + wake_up_interruptible(&li->wq); + spin_unlock_bh(&li->lock); + rc = 0; /* order accepted */ +unlock: + spin_unlock_bh(&fi->lock); + VCPU_EVENT(vcpu, 4, "sent sigp emerg to cpu %x", cpu_addr); + return rc; +} + +static int __sigp_stop(struct kvm_vcpu *vcpu, u16 cpu_addr, int store) +{ + struct float_interrupt *fi = &vcpu->kvm->arch.float_int; + struct local_interrupt *li; + struct interrupt_info *inti; + int rc; + + if (cpu_addr >= KVM_MAX_VCPUS) + return 3; /* not operational */ + + inti = kzalloc(sizeof(*inti), GFP_KERNEL); + if (!inti) + return -ENOMEM; + + inti->type = KVM_S390_SIGP_STOP; + + spin_lock_bh(&fi->lock); + li = fi->local_int[cpu_addr]; + if (li == NULL) { + rc = 3; /* not operational */ + kfree(inti); + goto unlock; + } + spin_lock_bh(&li->lock); + list_add_tail(&inti->list, &li->list); + atomic_set(&li->active, 1); + atomic_set_mask(CPUSTAT_STOP_INT, li->cpuflags); + if (store) + li->action_bits |= ACTION_STORE_ON_STOP; + li->action_bits |= ACTION_STOP_ON_STOP; + if (waitqueue_active(&li->wq)) + wake_up_interruptible(&li->wq); + spin_unlock_bh(&li->lock); + rc = 0; /* order accepted */ +unlock: + spin_unlock_bh(&fi->lock); + VCPU_EVENT(vcpu, 4, "sent sigp stop to cpu %x", cpu_addr); + return rc; +} + +static int __sigp_set_arch(struct kvm_vcpu *vcpu, u32 parameter) +{ + int rc; + + switch (parameter & 0xff) { + case 0: + printk(KERN_WARNING "kvm: request to switch to ESA/390 mode" + " not supported"); + rc = 3; /* not operational */ + break; + case 1: + case 2: + rc = 0; /* order accepted */ + break; + default: + rc = -ENOTSUPP; + } + return rc; +} + +static int __sigp_set_prefix(struct kvm_vcpu *vcpu, u16 cpu_addr, u32 address, + u64 *reg) +{ + struct float_interrupt *fi = &vcpu->kvm->arch.float_int; + struct local_interrupt *li; + struct interrupt_info *inti; + int rc; + u8 tmp; + + /* make sure that the new value is valid memory */ + address = address & 0x7fffe000u; + if ((copy_from_guest(vcpu, &tmp, + (u64) (address + vcpu->kvm->arch.guest_origin) , 1)) || + (copy_from_guest(vcpu, &tmp, (u64) (address + + vcpu->kvm->arch.guest_origin + PAGE_SIZE), 1))) { + *reg |= SIGP_STAT_INVALID_PARAMETER; + return 1; /* invalid parameter */ + } + + inti = kzalloc(sizeof(*inti), GFP_KERNEL); + if (!inti) + return 2; /* busy */ + + spin_lock_bh(&fi->lock); + li = fi->local_int[cpu_addr]; + + if ((cpu_addr >= KVM_MAX_VCPUS) || (li == NULL)) { + rc = 1; /* incorrect state */ + *reg &= SIGP_STAT_INCORRECT_STATE; + kfree(inti); + goto out_fi; + } + + spin_lock_bh(&li->lock); + /* cpu must be in stopped state */ + if (atomic_read(li->cpuflags) & CPUSTAT_RUNNING) { + rc = 1; /* incorrect state */ + *reg &= SIGP_STAT_INCORRECT_STATE; + kfree(inti); + goto out_li; + } + + inti->type = KVM_S390_SIGP_SET_PREFIX; + inti->prefix.address = address; + + list_add_tail(&inti->list, &li->list); + atomic_set(&li->active, 1); + if (waitqueue_active(&li->wq)) + wake_up_interruptible(&li->wq); + rc = 0; /* order accepted */ + + VCPU_EVENT(vcpu, 4, "set prefix of cpu %02x to %x", cpu_addr, address); +out_li: + spin_unlock_bh(&li->lock); +out_fi: + spin_unlock_bh(&fi->lock); + return rc; +} + +int kvm_s390_handle_sigp(struct kvm_vcpu *vcpu) +{ + int r1 = (vcpu->arch.sie_block->ipa & 0x00f0) >> 4; + int r3 = vcpu->arch.sie_block->ipa & 0x000f; + int base2 = vcpu->arch.sie_block->ipb >> 28; + int disp2 = ((vcpu->arch.sie_block->ipb & 0x0fff0000) >> 16); + u32 parameter; + u16 cpu_addr = vcpu->arch.guest_gprs[r3]; + u8 order_code; + int rc; + + order_code = disp2; + if (base2) + order_code += vcpu->arch.guest_gprs[base2]; + + if (r1 % 2) + parameter = vcpu->arch.guest_gprs[r1]; + else + parameter = vcpu->arch.guest_gprs[r1 + 1]; + + switch (order_code) { + case SIGP_SENSE: + vcpu->stat.instruction_sigp_sense++; + rc = __sigp_sense(vcpu, cpu_addr, + &vcpu->arch.guest_gprs[r1]); + break; + case SIGP_EMERGENCY: + vcpu->stat.instruction_sigp_emergency++; + rc = __sigp_emergency(vcpu, cpu_addr); + break; + case SIGP_STOP: + vcpu->stat.instruction_sigp_stop++; + rc = __sigp_stop(vcpu, cpu_addr, 0); + break; + case SIGP_STOP_STORE_STATUS: + vcpu->stat.instruction_sigp_stop++; + rc = __sigp_stop(vcpu, cpu_addr, 1); + break; + case SIGP_SET_ARCH: + vcpu->stat.instruction_sigp_arch++; + rc = __sigp_set_arch(vcpu, parameter); + break; + case SIGP_SET_PREFIX: + vcpu->stat.instruction_sigp_prefix++; + rc = __sigp_set_prefix(vcpu, cpu_addr, parameter, + &vcpu->arch.guest_gprs[r1]); + break; + case SIGP_RESTART: + vcpu->stat.instruction_sigp_restart++; + /* user space must know about restart */ + default: + return -ENOTSUPP; + } + + if (rc < 0) + return rc; + + vcpu->arch.sie_block->gpsw.mask &= ~(3ul << 44); + vcpu->arch.sie_block->gpsw.mask |= (rc & 3ul) << 44; + return 0; +} diff --git a/include/asm-s390/kvm_host.h b/include/asm-s390/kvm_host.h index 2eaf6fede02..1c829bdf588 100644 --- a/include/asm-s390/kvm_host.h +++ b/include/asm-s390/kvm_host.h @@ -128,6 +128,12 @@ struct kvm_vcpu_stat { u32 instruction_chsc; u32 instruction_stsi; u32 instruction_stfl; + u32 instruction_sigp_sense; + u32 instruction_sigp_emergency; + u32 instruction_sigp_stop; + u32 instruction_sigp_arch; + u32 instruction_sigp_prefix; + u32 instruction_sigp_restart; }; struct io_info { @@ -169,6 +175,10 @@ struct interrupt_info { }; }; +/* for local_interrupt.action_flags */ +#define ACTION_STORE_ON_STOP 1 +#define ACTION_STOP_ON_STOP 2 + struct local_interrupt { spinlock_t lock; struct list_head list; @@ -176,6 +186,8 @@ struct local_interrupt { struct float_interrupt *float_int; int timer_due; /* event indicator for waitqueue below */ wait_queue_head_t wq; + atomic_t *cpuflags; + unsigned int action_bits; }; struct float_interrupt { -- cgit v1.2.3-70-g09d2 From e28acfea5dd9dbc67c2594cbefc140129dbd0e3f Mon Sep 17 00:00:00 2001 From: Christian Borntraeger Date: Tue, 25 Mar 2008 18:47:34 +0100 Subject: KVM: s390: intercepts for diagnose instructions This patch introduces interpretation of some diagnose instruction intercepts. Diagnose is our classic architected way of doing a hypercall. This patch features the following diagnose codes: - vm storage size, that tells the guest about its memory layout - time slice end, which is used by the guest to indicate that it waits for a lock and thus cannot use up its time slice in a useful way - ipl functions, which a guest can use to reset and reboot itself In order to implement ipl functions, we also introduce an exit reason that causes userspace to perform various resets on the virtual machine. All resets are described in the principles of operation book, except KVM_S390_RESET_IPL which causes a reboot of the machine. Acked-by: Martin Schwidefsky Signed-off-by: Christian Borntraeger Signed-off-by: Carsten Otte Signed-off-by: Avi Kivity --- arch/s390/kvm/Makefile | 2 +- arch/s390/kvm/diag.c | 67 +++++++++++++++++++++++++++++++++++++++++++++ arch/s390/kvm/intercept.c | 1 + arch/s390/kvm/kvm-s390.c | 1 + arch/s390/kvm/kvm-s390.h | 3 ++ include/asm-s390/kvm_host.h | 5 +++- include/linux/kvm.h | 8 ++++++ 7 files changed, 85 insertions(+), 2 deletions(-) create mode 100644 arch/s390/kvm/diag.c (limited to 'arch/s390') diff --git a/arch/s390/kvm/Makefile b/arch/s390/kvm/Makefile index f3bf11a88bc..e5221ec0b8e 100644 --- a/arch/s390/kvm/Makefile +++ b/arch/s390/kvm/Makefile @@ -10,5 +10,5 @@ common-objs = $(addprefix ../../../virt/kvm/, kvm_main.o) EXTRA_CFLAGS += -Ivirt/kvm -Iarch/s390/kvm -kvm-objs := $(common-objs) kvm-s390.o sie64a.o intercept.o interrupt.o priv.o sigp.o +kvm-objs := $(common-objs) kvm-s390.o sie64a.o intercept.o interrupt.o priv.o sigp.o diag.o obj-$(CONFIG_KVM) += kvm.o diff --git a/arch/s390/kvm/diag.c b/arch/s390/kvm/diag.c new file mode 100644 index 00000000000..f639a152869 --- /dev/null +++ b/arch/s390/kvm/diag.c @@ -0,0 +1,67 @@ +/* + * diag.c - handling diagnose instructions + * + * Copyright IBM Corp. 2008 + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License (version 2 only) + * as published by the Free Software Foundation. + * + * Author(s): Carsten Otte + * Christian Borntraeger + */ + +#include +#include +#include "kvm-s390.h" + +static int __diag_time_slice_end(struct kvm_vcpu *vcpu) +{ + VCPU_EVENT(vcpu, 5, "%s", "diag time slice end"); + vcpu->stat.diagnose_44++; + vcpu_put(vcpu); + schedule(); + vcpu_load(vcpu); + return 0; +} + +static int __diag_ipl_functions(struct kvm_vcpu *vcpu) +{ + unsigned int reg = vcpu->arch.sie_block->ipa & 0xf; + unsigned long subcode = vcpu->arch.guest_gprs[reg] & 0xffff; + + VCPU_EVENT(vcpu, 5, "diag ipl functions, subcode %lx", subcode); + switch (subcode) { + case 3: + vcpu->run->s390_reset_flags = KVM_S390_RESET_CLEAR; + break; + case 4: + vcpu->run->s390_reset_flags = 0; + break; + default: + return -ENOTSUPP; + } + + atomic_clear_mask(CPUSTAT_RUNNING, &vcpu->arch.sie_block->cpuflags); + vcpu->run->s390_reset_flags |= KVM_S390_RESET_SUBSYSTEM; + vcpu->run->s390_reset_flags |= KVM_S390_RESET_IPL; + vcpu->run->s390_reset_flags |= KVM_S390_RESET_CPU_INIT; + vcpu->run->exit_reason = KVM_EXIT_S390_RESET; + VCPU_EVENT(vcpu, 3, "requesting userspace resets %lx", + vcpu->run->s390_reset_flags); + return -EREMOTE; +} + +int kvm_s390_handle_diag(struct kvm_vcpu *vcpu) +{ + int code = (vcpu->arch.sie_block->ipb & 0xfff0000) >> 16; + + switch (code) { + case 0x44: + return __diag_time_slice_end(vcpu); + case 0x308: + return __diag_ipl_functions(vcpu); + default: + return -ENOTSUPP; + } +} diff --git a/arch/s390/kvm/intercept.c b/arch/s390/kvm/intercept.c index 9f0d8b23943..349581a2610 100644 --- a/arch/s390/kvm/intercept.c +++ b/arch/s390/kvm/intercept.c @@ -95,6 +95,7 @@ static int handle_lctl(struct kvm_vcpu *vcpu) } static intercept_handler_t instruction_handlers[256] = { + [0x83] = kvm_s390_handle_diag, [0xae] = kvm_s390_handle_sigp, [0xb2] = kvm_s390_handle_priv, [0xb7] = handle_lctl, diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c index c632180739e..d3b1de83678 100644 --- a/arch/s390/kvm/kvm-s390.c +++ b/arch/s390/kvm/kvm-s390.c @@ -63,6 +63,7 @@ struct kvm_stats_debugfs_item debugfs_entries[] = { { "instruction_sigp_set_arch", VCPU_STAT(instruction_sigp_arch) }, { "instruction_sigp_set_prefix", VCPU_STAT(instruction_sigp_prefix) }, { "instruction_sigp_restart", VCPU_STAT(instruction_sigp_restart) }, + { "diagnose_44", VCPU_STAT(diagnose_44) }, { NULL } }; diff --git a/arch/s390/kvm/kvm-s390.h b/arch/s390/kvm/kvm-s390.h index e6e5756a0e0..3893cf12eac 100644 --- a/arch/s390/kvm/kvm-s390.h +++ b/arch/s390/kvm/kvm-s390.h @@ -58,4 +58,7 @@ int kvm_s390_handle_sigp(struct kvm_vcpu *vcpu); /* implemented in kvm-s390.c */ int __kvm_s390_vcpu_store_status(struct kvm_vcpu *vcpu, unsigned long addr); +/* implemented in diag.c */ +int kvm_s390_handle_diag(struct kvm_vcpu *vcpu); + #endif diff --git a/include/asm-s390/kvm_host.h b/include/asm-s390/kvm_host.h index 1c829bdf588..f8204a4f2e0 100644 --- a/include/asm-s390/kvm_host.h +++ b/include/asm-s390/kvm_host.h @@ -94,7 +94,9 @@ struct sie_block { psw_t gpsw; /* 0x0090 */ __u64 gg14; /* 0x00a0 */ __u64 gg15; /* 0x00a8 */ - __u8 reservedb0[80]; /* 0x00b0 */ + __u8 reservedb0[30]; /* 0x00b0 */ + __u16 iprcc; /* 0x00ce */ + __u8 reservedd0[48]; /* 0x00d0 */ __u64 gcr[16]; /* 0x0100 */ __u64 gbea; /* 0x0180 */ __u8 reserved188[120]; /* 0x0188 */ @@ -134,6 +136,7 @@ struct kvm_vcpu_stat { u32 instruction_sigp_arch; u32 instruction_sigp_prefix; u32 instruction_sigp_restart; + u32 diagnose_44; }; struct io_info { diff --git a/include/linux/kvm.h b/include/linux/kvm.h index 029f0284a2f..f04bb426618 100644 --- a/include/linux/kvm.h +++ b/include/linux/kvm.h @@ -75,6 +75,7 @@ struct kvm_irqchip { #define KVM_EXIT_SET_TPR 11 #define KVM_EXIT_TPR_ACCESS 12 #define KVM_EXIT_S390_SIEIC 13 +#define KVM_EXIT_S390_RESET 14 /* for KVM_RUN, returned by mmap(vcpu_fd, offset=0) */ struct kvm_run { @@ -147,6 +148,13 @@ struct kvm_run { __u16 ipa; __u32 ipb; } s390_sieic; + /* KVM_EXIT_S390_RESET */ +#define KVM_S390_RESET_POR 1 +#define KVM_S390_RESET_CLEAR 2 +#define KVM_S390_RESET_SUBSYSTEM 4 +#define KVM_S390_RESET_CPU_INIT 8 +#define KVM_S390_RESET_IPL 16 + __u64 s390_reset_flags; /* Fix the size of the union. */ char padding[256]; }; -- cgit v1.2.3-70-g09d2 From 77b455f1bcfa0fddb31b8e6f9f2adc246acb4216 Mon Sep 17 00:00:00 2001 From: Christian Borntraeger Date: Tue, 25 Mar 2008 18:47:36 +0100 Subject: KVM: s390: add kvm to kconfig on s390 This patch adds the virtualization submenu and the kvm option to the kernel config. It also defines HAVE_KVM for 64bit kernels. Acked-by: Martin Schwidefsky Signed-off-by: Christian Borntraeger Signed-off-by: Carsten Otte Signed-off-by: Avi Kivity --- arch/s390/Kconfig | 3 +++ arch/s390/kvm/Kconfig | 43 +++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 46 insertions(+) create mode 100644 arch/s390/kvm/Kconfig (limited to 'arch/s390') diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig index 513a0589e81..9a742ce9cc9 100644 --- a/arch/s390/Kconfig +++ b/arch/s390/Kconfig @@ -73,6 +73,7 @@ config S390 select HAVE_OPROFILE select HAVE_KPROBES select HAVE_KRETPROBES + select HAVE_KVM if 64BIT source "init/Kconfig" @@ -540,3 +541,5 @@ source "security/Kconfig" source "crypto/Kconfig" source "lib/Kconfig" + +source "arch/s390/kvm/Kconfig" diff --git a/arch/s390/kvm/Kconfig b/arch/s390/kvm/Kconfig new file mode 100644 index 00000000000..2489b341a00 --- /dev/null +++ b/arch/s390/kvm/Kconfig @@ -0,0 +1,43 @@ +# +# KVM configuration +# +config HAVE_KVM + bool + +menuconfig VIRTUALIZATION + bool "Virtualization" + default y + ---help--- + Say Y here to get to see options for using your Linux host to run other + operating systems inside virtual machines (guests). + This option alone does not add any kernel code. + + If you say N, all options in this submenu will be skipped and disabled. + +if VIRTUALIZATION + +config KVM + tristate "Kernel-based Virtual Machine (KVM) support" + depends on HAVE_KVM && EXPERIMENTAL + select PREEMPT_NOTIFIERS + select ANON_INODES + select S390_SWITCH_AMODE + select PREEMPT + ---help--- + Support hosting paravirtualized guest machines using the SIE + virtualization capability on the mainframe. This should work + on any 64bit machine. + + This module provides access to the hardware capabilities through + a character device node named /dev/kvm. + + To compile this as a module, choose M here: the module + will be called kvm. + + If unsure, say N. + +# OK, it's a little counter-intuitive to do this, but it puts it neatly under +# the virtualization menu. +source drivers/virtio/Kconfig + +endif # VIRTUALIZATION -- cgit v1.2.3-70-g09d2 From fa5877439d5a062d91c3abd5a690483bbdb4268e Mon Sep 17 00:00:00 2001 From: Carsten Otte Date: Tue, 25 Mar 2008 18:47:44 +0100 Subject: s390: KVM guest: detect when running on kvm This patch adds functionality to detect if the kernel runs under the KVM hypervisor. A macro MACHINE_IS_KVM is exported for device drivers. This allows drivers to skip device detection if the systems runs non-virtualized. We also define a preferred console to avoid having the ttyS0, which is a line mode only console. Signed-off-by: Christian Borntraeger Acked-by: Martin Schwidefsky Signed-off-by: Carsten Otte Signed-off-by: Avi Kivity --- arch/s390/Kconfig | 7 +++++++ arch/s390/kernel/early.c | 4 ++++ arch/s390/kernel/setup.c | 10 +++++++--- include/asm-s390/setup.h | 1 + 4 files changed, 19 insertions(+), 3 deletions(-) (limited to 'arch/s390') diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig index 9a742ce9cc9..8f5f02160ff 100644 --- a/arch/s390/Kconfig +++ b/arch/s390/Kconfig @@ -520,6 +520,13 @@ config ZFCPDUMP Select this option if you want to build an zfcpdump enabled kernel. Refer to for more details on this. +config S390_GUEST +bool "s390 guest support (EXPERIMENTAL)" + depends on 64BIT && EXPERIMENTAL + select VIRTIO + select VIRTIO_RING + help + Select this option if you want to run the kernel under s390 linux endmenu source "net/Kconfig" diff --git a/arch/s390/kernel/early.c b/arch/s390/kernel/early.c index 540a67f979b..68ec4083bf7 100644 --- a/arch/s390/kernel/early.c +++ b/arch/s390/kernel/early.c @@ -144,6 +144,10 @@ static noinline __init void detect_machine_type(void) /* Running on a P/390 ? */ if (cpuinfo->cpu_id.machine == 0x7490) machine_flags |= 4; + + /* Running under KVM ? */ + if (cpuinfo->cpu_id.version == 0xfe) + machine_flags |= 64; } #ifdef CONFIG_64BIT diff --git a/arch/s390/kernel/setup.c b/arch/s390/kernel/setup.c index 2f35133ebc1..a9d18aafa5f 100644 --- a/arch/s390/kernel/setup.c +++ b/arch/s390/kernel/setup.c @@ -801,9 +801,13 @@ setup_arch(char **cmdline_p) "This machine has an IEEE fpu\n" : "This machine has no IEEE fpu\n"); #else /* CONFIG_64BIT */ - printk((MACHINE_IS_VM) ? - "We are running under VM (64 bit mode)\n" : - "We are running native (64 bit mode)\n"); + if (MACHINE_IS_VM) + printk("We are running under VM (64 bit mode)\n"); + else if (MACHINE_IS_KVM) { + printk("We are running under KVM (64 bit mode)\n"); + add_preferred_console("ttyS", 1, NULL); + } else + printk("We are running native (64 bit mode)\n"); #endif /* CONFIG_64BIT */ /* Save unparsed command line copy for /proc/cmdline */ diff --git a/include/asm-s390/setup.h b/include/asm-s390/setup.h index a76a6b8fd88..aaf4b518b94 100644 --- a/include/asm-s390/setup.h +++ b/include/asm-s390/setup.h @@ -62,6 +62,7 @@ extern unsigned long machine_flags; #define MACHINE_IS_VM (machine_flags & 1) #define MACHINE_IS_P390 (machine_flags & 4) #define MACHINE_HAS_MVPG (machine_flags & 16) +#define MACHINE_IS_KVM (machine_flags & 64) #define MACHINE_HAS_IDTE (machine_flags & 128) #define MACHINE_HAS_DIAG9C (machine_flags & 256) -- cgit v1.2.3-70-g09d2 From 7e8e6ab48d78147f69c1ba2d6a362f8d33254468 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Fri, 4 Apr 2008 15:12:35 +0200 Subject: KVM: s390: Fix incorrect return value kvm_arch_vcpu_ioctl_run currently incorrectly always returns 0. Signed-off-by: Heiko Carstens Signed-off-by: Carsten Otte Signed-off-by: Avi Kivity --- arch/s390/kvm/kvm-s390.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/s390') diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c index d3b1de83678..d9661376597 100644 --- a/arch/s390/kvm/kvm-s390.c +++ b/arch/s390/kvm/kvm-s390.c @@ -497,7 +497,7 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) vcpu_put(vcpu); vcpu->stat.exit_userspace++; - return 0; + return rc; } static int __guestcopy(struct kvm_vcpu *vcpu, u64 guestdest, const void *from, -- cgit v1.2.3-70-g09d2 From f603f0731f43421403160f5f8b12e90f2e51f064 Mon Sep 17 00:00:00 2001 From: Heiko Carstens Date: Fri, 4 Apr 2008 15:12:40 +0200 Subject: KVM: s390: rename stfl to kvm_stfl Temporarily rename this function to avoid merge conflicts and/or dependencies. This function will be removed as soon as git-s390 and kvm.git are finally upstream. Signed-off-by: Heiko Carstens Signed-off-by: Carsten Otte Signed-off-by: Avi Kivity --- arch/s390/kvm/priv.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'arch/s390') diff --git a/arch/s390/kvm/priv.c b/arch/s390/kvm/priv.c index c97e904ca3b..1465946325c 100644 --- a/arch/s390/kvm/priv.c +++ b/arch/s390/kvm/priv.c @@ -151,7 +151,7 @@ static int handle_chsc(struct kvm_vcpu *vcpu) return 0; } -static unsigned int stfl(void) +static unsigned int kvm_stfl(void) { asm volatile( " .insn s,0xb2b10000,0(0)\n" /* stfl */ @@ -162,7 +162,7 @@ static unsigned int stfl(void) static int handle_stfl(struct kvm_vcpu *vcpu) { - unsigned int facility_list = stfl(); + unsigned int facility_list = kvm_stfl(); int rc; vcpu->stat.instruction_stfl++; -- cgit v1.2.3-70-g09d2 From 7732a8d19bdc6ae18f68f9adb47d11c82a3a86cd Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Fri, 11 Apr 2008 02:50:40 +0300 Subject: KVM: s390: Stub out kvmtrace Signed-off-by: Avi Kivity --- arch/s390/kvm/Kconfig | 3 +++ 1 file changed, 3 insertions(+) (limited to 'arch/s390') diff --git a/arch/s390/kvm/Kconfig b/arch/s390/kvm/Kconfig index 2489b341a00..1761b74d639 100644 --- a/arch/s390/kvm/Kconfig +++ b/arch/s390/kvm/Kconfig @@ -36,6 +36,9 @@ config KVM If unsure, say N. +config KVM_TRACE + bool + # OK, it's a little counter-intuitive to do this, but it puts it neatly under # the virtualization menu. source drivers/virtio/Kconfig -- cgit v1.2.3-70-g09d2 From 3d80840d96127401ba6aeadd813c3a15b84e70fe Mon Sep 17 00:00:00 2001 From: Marcelo Tosatti Date: Fri, 11 Apr 2008 14:53:26 -0300 Subject: KVM: hlt emulation should take in-kernel APIC/PIT timers into account Timers that fire between guest hlt and vcpu_block's add_wait_queue() are ignored, possibly resulting in hangs. Also make sure that atomic_inc and waitqueue_active tests happen in the specified order, otherwise the following race is open: CPU0 CPU1 if (waitqueue_active(wq)) add_wait_queue() if (!atomic_read(pit_timer->pending)) schedule() atomic_inc(pit_timer->pending) Signed-off-by: Marcelo Tosatti Signed-off-by: Avi Kivity --- arch/ia64/kvm/kvm-ia64.c | 5 +++++ arch/s390/kvm/interrupt.c | 5 +++++ arch/x86/kvm/i8254.c | 10 ++++++++++ arch/x86/kvm/irq.c | 15 +++++++++++++++ arch/x86/kvm/irq.h | 3 +++ arch/x86/kvm/lapic.c | 10 ++++++++++ include/linux/kvm_host.h | 1 + virt/kvm/kvm_main.c | 1 + 8 files changed, 50 insertions(+) (limited to 'arch/s390') diff --git a/arch/ia64/kvm/kvm-ia64.c b/arch/ia64/kvm/kvm-ia64.c index 9c56b6429cb..ca1cfb124d4 100644 --- a/arch/ia64/kvm/kvm-ia64.c +++ b/arch/ia64/kvm/kvm-ia64.c @@ -1778,6 +1778,11 @@ int kvm_cpu_has_interrupt(struct kvm_vcpu *vcpu) return 0; } +int kvm_cpu_has_pending_timer(struct kvm_vcpu *vcpu) +{ + return 0; +} + gfn_t unalias_gfn(struct kvm *kvm, gfn_t gfn) { return gfn; diff --git a/arch/s390/kvm/interrupt.c b/arch/s390/kvm/interrupt.c index f62588cb75f..fcd1ed8015c 100644 --- a/arch/s390/kvm/interrupt.c +++ b/arch/s390/kvm/interrupt.c @@ -325,6 +325,11 @@ int kvm_cpu_has_interrupt(struct kvm_vcpu *vcpu) return rc; } +int kvm_cpu_has_pending_timer(struct kvm_vcpu *vcpu) +{ + return 0; +} + int kvm_s390_handle_wait(struct kvm_vcpu *vcpu) { u64 now, sltime; diff --git a/arch/x86/kvm/i8254.c b/arch/x86/kvm/i8254.c index 9f118e2f350..ed1af80432b 100644 --- a/arch/x86/kvm/i8254.c +++ b/arch/x86/kvm/i8254.c @@ -212,6 +212,16 @@ int __pit_timer_fn(struct kvm_kpit_state *ps) return (pt->period == 0 ? 0 : 1); } +int pit_has_pending_timer(struct kvm_vcpu *vcpu) +{ + struct kvm_pit *pit = vcpu->kvm->arch.vpit; + + if (pit && vcpu->vcpu_id == 0) + return atomic_read(&pit->pit_state.pit_timer.pending); + + return 0; +} + static enum hrtimer_restart pit_timer_fn(struct hrtimer *data) { struct kvm_kpit_state *ps; diff --git a/arch/x86/kvm/irq.c b/arch/x86/kvm/irq.c index dbfe21c99c4..ce1f583459b 100644 --- a/arch/x86/kvm/irq.c +++ b/arch/x86/kvm/irq.c @@ -25,6 +25,21 @@ #include "irq.h" #include "i8254.h" +/* + * check if there are pending timer events + * to be processed. + */ +int kvm_cpu_has_pending_timer(struct kvm_vcpu *vcpu) +{ + int ret; + + ret = pit_has_pending_timer(vcpu); + ret |= apic_has_pending_timer(vcpu); + + return ret; +} +EXPORT_SYMBOL(kvm_cpu_has_pending_timer); + /* * check if there is pending interrupt without * intack. diff --git a/arch/x86/kvm/irq.h b/arch/x86/kvm/irq.h index fa5ed5d59b5..1802134b836 100644 --- a/arch/x86/kvm/irq.h +++ b/arch/x86/kvm/irq.h @@ -85,4 +85,7 @@ void kvm_inject_pending_timer_irqs(struct kvm_vcpu *vcpu); void kvm_inject_apic_timer_irqs(struct kvm_vcpu *vcpu); void __kvm_migrate_apic_timer(struct kvm_vcpu *vcpu); +int pit_has_pending_timer(struct kvm_vcpu *vcpu); +int apic_has_pending_timer(struct kvm_vcpu *vcpu); + #endif diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c index 31280df7d2e..debf58211bd 100644 --- a/arch/x86/kvm/lapic.c +++ b/arch/x86/kvm/lapic.c @@ -952,6 +952,16 @@ static int __apic_timer_fn(struct kvm_lapic *apic) return result; } +int apic_has_pending_timer(struct kvm_vcpu *vcpu) +{ + struct kvm_lapic *lapic = vcpu->arch.apic; + + if (lapic) + return atomic_read(&lapic->timer.pending); + + return 0; +} + static int __inject_apic_timer_irq(struct kvm_lapic *apic) { int vector; diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h index bd0c2d2d840..0bc400387ca 100644 --- a/include/linux/kvm_host.h +++ b/include/linux/kvm_host.h @@ -269,6 +269,7 @@ void kvm_arch_destroy_vm(struct kvm *kvm); int kvm_cpu_get_interrupt(struct kvm_vcpu *v); int kvm_cpu_has_interrupt(struct kvm_vcpu *v); +int kvm_cpu_has_pending_timer(struct kvm_vcpu *vcpu); void kvm_vcpu_kick(struct kvm_vcpu *vcpu); static inline void kvm_guest_enter(void) diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index d5911d9895c..47cbc6e3faf 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -765,6 +765,7 @@ void kvm_vcpu_block(struct kvm_vcpu *vcpu) * We will block until either an interrupt or a signal wakes us up */ while (!kvm_cpu_has_interrupt(vcpu) + && !kvm_cpu_has_pending_timer(vcpu) && !signal_pending(current) && !kvm_arch_vcpu_runnable(vcpu)) { set_current_state(TASK_INTERRUPTIBLE); -- cgit v1.2.3-70-g09d2 From 62d9f0dbc92d7e398fde53fc6021338393522e68 Mon Sep 17 00:00:00 2001 From: Marcelo Tosatti Date: Fri, 11 Apr 2008 13:24:45 -0300 Subject: KVM: add ioctls to save/store mpstate So userspace can save/restore the mpstate during migration. [avi: export the #define constants describing the value] [christian: add s390 stubs] [avi: ditto for ia64] Signed-off-by: Marcelo Tosatti Signed-off-by: Christian Borntraeger Signed-off-by: Carsten Otte Signed-off-by: Avi Kivity --- arch/ia64/kvm/kvm-ia64.c | 12 ++++++++++++ arch/s390/kvm/kvm-s390.c | 12 ++++++++++++ arch/x86/kvm/x86.c | 19 +++++++++++++++++++ include/asm-x86/kvm_host.h | 5 ----- include/linux/kvm.h | 15 +++++++++++++++ include/linux/kvm_host.h | 4 ++++ virt/kvm/kvm_main.c | 24 ++++++++++++++++++++++++ 7 files changed, 86 insertions(+), 5 deletions(-) (limited to 'arch/s390') diff --git a/arch/ia64/kvm/kvm-ia64.c b/arch/ia64/kvm/kvm-ia64.c index f7589dba75a..6df07324013 100644 --- a/arch/ia64/kvm/kvm-ia64.c +++ b/arch/ia64/kvm/kvm-ia64.c @@ -1792,3 +1792,15 @@ int kvm_arch_vcpu_runnable(struct kvm_vcpu *vcpu) { return vcpu->arch.mp_state == KVM_MP_STATE_RUNNABLE; } + +int kvm_arch_vcpu_ioctl_get_mpstate(struct kvm_vcpu *vcpu, + struct kvm_mp_state *mp_state) +{ + return -EINVAL; +} + +int kvm_arch_vcpu_ioctl_set_mpstate(struct kvm_vcpu *vcpu, + struct kvm_mp_state *mp_state) +{ + return -EINVAL; +} diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c index d9661376597..98d1e73e01f 100644 --- a/arch/s390/kvm/kvm-s390.c +++ b/arch/s390/kvm/kvm-s390.c @@ -414,6 +414,18 @@ int kvm_arch_vcpu_ioctl_debug_guest(struct kvm_vcpu *vcpu, return -EINVAL; /* not implemented yet */ } +int kvm_arch_vcpu_ioctl_get_mpstate(struct kvm_vcpu *vcpu, + struct kvm_mp_state *mp_state) +{ + return -EINVAL; /* not implemented yet */ +} + +int kvm_arch_vcpu_ioctl_set_mpstate(struct kvm_vcpu *vcpu, + struct kvm_mp_state *mp_state) +{ + return -EINVAL; /* not implemented yet */ +} + static void __vcpu_run(struct kvm_vcpu *vcpu) { memcpy(&vcpu->arch.sie_block->gg14, &vcpu->arch.guest_gprs[14], 16); diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index b364d192896..5c3c9d38c78 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -817,6 +817,7 @@ int kvm_dev_ioctl_check_extension(long ext) case KVM_CAP_CLOCKSOURCE: case KVM_CAP_PIT: case KVM_CAP_NOP_IO_DELAY: + case KVM_CAP_MP_STATE: r = 1; break; case KVM_CAP_VAPIC: @@ -3083,6 +3084,24 @@ int kvm_arch_vcpu_ioctl_get_sregs(struct kvm_vcpu *vcpu, return 0; } +int kvm_arch_vcpu_ioctl_get_mpstate(struct kvm_vcpu *vcpu, + struct kvm_mp_state *mp_state) +{ + vcpu_load(vcpu); + mp_state->mp_state = vcpu->arch.mp_state; + vcpu_put(vcpu); + return 0; +} + +int kvm_arch_vcpu_ioctl_set_mpstate(struct kvm_vcpu *vcpu, + struct kvm_mp_state *mp_state) +{ + vcpu_load(vcpu); + vcpu->arch.mp_state = mp_state->mp_state; + vcpu_put(vcpu); + return 0; +} + static void set_segment(struct kvm_vcpu *vcpu, struct kvm_segment *var, int seg) { diff --git a/include/asm-x86/kvm_host.h b/include/asm-x86/kvm_host.h index f35a6ad43c0..9d963cd6533 100644 --- a/include/asm-x86/kvm_host.h +++ b/include/asm-x86/kvm_host.h @@ -227,11 +227,6 @@ struct kvm_vcpu_arch { u64 shadow_efer; u64 apic_base; struct kvm_lapic *apic; /* kernel irqchip context */ -#define KVM_MP_STATE_RUNNABLE 0 -#define KVM_MP_STATE_UNINITIALIZED 1 -#define KVM_MP_STATE_INIT_RECEIVED 2 -#define KVM_MP_STATE_SIPI_RECEIVED 3 -#define KVM_MP_STATE_HALTED 4 int mp_state; int sipi_vector; u64 ia32_misc_enable_msr; diff --git a/include/linux/kvm.h b/include/linux/kvm.h index d302d63517e..f8e211d8ce8 100644 --- a/include/linux/kvm.h +++ b/include/linux/kvm.h @@ -228,6 +228,18 @@ struct kvm_vapic_addr { __u64 vapic_addr; }; +/* for KVM_SET_MPSTATE */ + +#define KVM_MP_STATE_RUNNABLE 0 +#define KVM_MP_STATE_UNINITIALIZED 1 +#define KVM_MP_STATE_INIT_RECEIVED 2 +#define KVM_MP_STATE_HALTED 3 +#define KVM_MP_STATE_SIPI_RECEIVED 4 + +struct kvm_mp_state { + __u32 mp_state; +}; + struct kvm_s390_psw { __u64 mask; __u64 addr; @@ -326,6 +338,7 @@ struct kvm_trace_rec { #define KVM_CAP_PIT 11 #define KVM_CAP_NOP_IO_DELAY 12 #define KVM_CAP_PV_MMU 13 +#define KVM_CAP_MP_STATE 14 /* * ioctls for VM fds @@ -387,5 +400,7 @@ struct kvm_trace_rec { #define KVM_S390_SET_INITIAL_PSW _IOW(KVMIO, 0x96, struct kvm_s390_psw) /* initial reset for s390 */ #define KVM_S390_INITIAL_RESET _IO(KVMIO, 0x97) +#define KVM_GET_MP_STATE _IOR(KVMIO, 0x98, struct kvm_mp_state) +#define KVM_SET_MP_STATE _IOW(KVMIO, 0x99, struct kvm_mp_state) #endif diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h index 0bc400387ca..81d4c3305a2 100644 --- a/include/linux/kvm_host.h +++ b/include/linux/kvm_host.h @@ -237,6 +237,10 @@ int kvm_arch_vcpu_ioctl_get_sregs(struct kvm_vcpu *vcpu, struct kvm_sregs *sregs); int kvm_arch_vcpu_ioctl_set_sregs(struct kvm_vcpu *vcpu, struct kvm_sregs *sregs); +int kvm_arch_vcpu_ioctl_get_mpstate(struct kvm_vcpu *vcpu, + struct kvm_mp_state *mp_state); +int kvm_arch_vcpu_ioctl_set_mpstate(struct kvm_vcpu *vcpu, + struct kvm_mp_state *mp_state); int kvm_arch_vcpu_ioctl_debug_guest(struct kvm_vcpu *vcpu, struct kvm_debug_guest *dbg); int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run); diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index 47cbc6e3faf..09984557490 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -979,6 +979,30 @@ out_free2: r = 0; break; } + case KVM_GET_MP_STATE: { + struct kvm_mp_state mp_state; + + r = kvm_arch_vcpu_ioctl_get_mpstate(vcpu, &mp_state); + if (r) + goto out; + r = -EFAULT; + if (copy_to_user(argp, &mp_state, sizeof mp_state)) + goto out; + r = 0; + break; + } + case KVM_SET_MP_STATE: { + struct kvm_mp_state mp_state; + + r = -EFAULT; + if (copy_from_user(&mp_state, argp, sizeof mp_state)) + goto out; + r = kvm_arch_vcpu_ioctl_set_mpstate(vcpu, &mp_state); + if (r) + goto out; + r = 0; + break; + } case KVM_TRANSLATE: { struct kvm_translation tr; -- cgit v1.2.3-70-g09d2