7 files changed, 126 insertions, 53 deletions

diff --git a/arch/s390/include/asm/kvm_host.h b/arch/s390/include/asm/kvm_host.h
index 54ea39f96ecd..a27d0d5a6f86 100644
--- a/arch/s390/include/asm/kvm_host.h
+++ b/arch/s390/include/asm/kvm_host.h
@@ -13,6 +13,8 @@
 
 #ifndef ASM_KVM_HOST_H
 #define ASM_KVM_HOST_H
+#include <linux/hrtimer.h>
+#include <linux/interrupt.h>
 #include <linux/kvm_host.h>
 #include <asm/debug.h>
 #include <asm/cpuid.h>
@@ -210,7 +212,8 @@ struct kvm_vcpu_arch {
         s390_fp_regs      guest_fpregs;
         unsigned int      guest_acrs[NUM_ACRS];
         struct kvm_s390_local_interrupt local_int;
-        struct timer_list ckc_timer;
+        struct hrtimer    ckc_timer;
+        struct tasklet_struct tasklet;
         union  {
                 cpuid_t   cpu_id;
                 u64       stidp_data;
diff --git a/arch/s390/kvm/intercept.c b/arch/s390/kvm/intercept.c
index 9d19803111ba..98997ccba501 100644
--- a/arch/s390/kvm/intercept.c
+++ b/arch/s390/kvm/intercept.c
@@ -154,17 +154,25 @@ static int handle_stop(struct kvm_vcpu *vcpu)
 static int handle_validity(struct kvm_vcpu *vcpu)
 {
         int viwhy = vcpu->arch.sie_block->ipb >> 16;
+        int rc;
+
         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;
+        if ((viwhy == 0x37) && (vcpu->arch.sie_block->prefix
+                <= vcpu->kvm->arch.guest_memsize - 2*PAGE_SIZE)){
+                rc = fault_in_pages_writeable((char __user *)
+                         vcpu->kvm->arch.guest_origin +
+                         vcpu->arch.sie_block->prefix,
+                         2*PAGE_SIZE);
+                if (rc)
+                        /* user will receive sigsegv, exit to user */
+                        rc = -ENOTSUPP;
+        } else
+                rc = -ENOTSUPP;
+
+        if (rc)
+                VCPU_EVENT(vcpu, 2, "unhandled validity intercept code %d",
+                           viwhy);
+        return rc;
 }
 
 static int handle_instruction(struct kvm_vcpu *vcpu)
diff --git a/arch/s390/kvm/interrupt.c b/arch/s390/kvm/interrupt.c
index 0189356fe209..f04f5301b1b4 100644
--- a/arch/s390/kvm/interrupt.c
+++ b/arch/s390/kvm/interrupt.c
@@ -12,6 +12,8 @@
 
 #include <asm/lowcore.h>
 #include <asm/uaccess.h>
+#include <linux/hrtimer.h>
+#include <linux/interrupt.h>
 #include <linux/kvm_host.h>
 #include <linux/signal.h>
 #include "kvm-s390.h"
@@ -299,13 +301,13 @@ int kvm_cpu_has_interrupt(struct kvm_vcpu *vcpu)
         }
 
         if ((!rc) && atomic_read(&fi->active)) {
-                spin_lock_bh(&fi->lock);
+                spin_lock(&fi->lock);
                 list_for_each_entry(inti, &fi->list, list)
                         if (__interrupt_is_deliverable(vcpu, inti)) {
                                 rc = 1;
                                 break;
                         }
-                spin_unlock_bh(&fi->lock);
+                spin_unlock(&fi->lock);
         }
 
         if ((!rc) && (vcpu->arch.sie_block->ckc <
@@ -318,6 +320,12 @@ int kvm_cpu_has_interrupt(struct kvm_vcpu *vcpu)
         return rc;
 }
 
+int kvm_arch_interrupt_allowed(struct kvm_vcpu *vcpu)
+{
+        /* do real check here */
+        return 1;
+}
+
 int kvm_cpu_has_pending_timer(struct kvm_vcpu *vcpu)
 {
         return 0;
@@ -355,14 +363,12 @@ int kvm_s390_handle_wait(struct kvm_vcpu *vcpu)
                 return 0;
         }
 
-        sltime = (vcpu->arch.sie_block->ckc - now) / (0xf4240000ul / HZ) + 1;
+        sltime = ((vcpu->arch.sie_block->ckc - now)*125)>>9;
 
-        vcpu->arch.ckc_timer.expires = jiffies + sltime;
-
-        add_timer(&vcpu->arch.ckc_timer);
-        VCPU_EVENT(vcpu, 5, "enabled wait timer:%llx jiffies", sltime);
+        hrtimer_start(&vcpu->arch.ckc_timer, ktime_set (0, sltime) , HRTIMER_MODE_REL);
+        VCPU_EVENT(vcpu, 5, "enabled wait via clock comparator: %llx ns", sltime);
 no_timer:
-        spin_lock_bh(&vcpu->arch.local_int.float_int->lock);
+        spin_lock(&vcpu->arch.local_int.float_int->lock);
         spin_lock_bh(&vcpu->arch.local_int.lock);
         add_wait_queue(&vcpu->arch.local_int.wq, &wait);
         while (list_empty(&vcpu->arch.local_int.list) &&
@@ -371,33 +377,46 @@ no_timer:
                 !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);
+                spin_unlock(&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(&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);
+        spin_unlock(&vcpu->arch.local_int.float_int->lock);
+        hrtimer_try_to_cancel(&vcpu->arch.ckc_timer);
         return 0;
 }
 
-void kvm_s390_idle_wakeup(unsigned long data)
+void kvm_s390_tasklet(unsigned long parm)
 {
-        struct kvm_vcpu *vcpu = (struct kvm_vcpu *)data;
+        struct kvm_vcpu *vcpu = (struct kvm_vcpu *) parm;
 
-        spin_lock_bh(&vcpu->arch.local_int.lock);
+        spin_lock(&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);
+        spin_unlock(&vcpu->arch.local_int.lock);
 }
 
+/*
+ * low level hrtimer wake routine. Because this runs in hardirq context
+ * we schedule a tasklet to do the real work.
+ */
+enum hrtimer_restart kvm_s390_idle_wakeup(struct hrtimer *timer)
+{
+        struct kvm_vcpu *vcpu;
+
+        vcpu = container_of(timer, struct kvm_vcpu, arch.ckc_timer);
+        tasklet_schedule(&vcpu->arch.tasklet);
+
+        return HRTIMER_NORESTART;
+}
 
 void kvm_s390_deliver_pending_interrupts(struct kvm_vcpu *vcpu)
 {
@@ -436,7 +455,7 @@ void kvm_s390_deliver_pending_interrupts(struct kvm_vcpu *vcpu)
         if (atomic_read(&fi->active)) {
                 do {
                         deliver = 0;
-                        spin_lock_bh(&fi->lock);
+                        spin_lock(&fi->lock);
                         list_for_each_entry_safe(inti, n, &fi->list, list) {
                                 if (__interrupt_is_deliverable(vcpu, inti)) {
                                         list_del(&inti->list);
@@ -447,7 +466,7 @@ void kvm_s390_deliver_pending_interrupts(struct kvm_vcpu *vcpu)
                         }
                         if (list_empty(&fi->list))
                                 atomic_set(&fi->active, 0);
-                        spin_unlock_bh(&fi->lock);
+                        spin_unlock(&fi->lock);
                         if (deliver) {
                                 __do_deliver_interrupt(vcpu, inti);
                                 kfree(inti);
@@ -512,7 +531,7 @@ int kvm_s390_inject_vm(struct kvm *kvm,
 
         mutex_lock(&kvm->lock);
         fi = &kvm->arch.float_int;
-        spin_lock_bh(&fi->lock);
+        spin_lock(&fi->lock);
         list_add_tail(&inti->list, &fi->list);
         atomic_set(&fi->active, 1);
         sigcpu = find_first_bit(fi->idle_mask, KVM_MAX_VCPUS);
@@ -529,7 +548,7 @@ int kvm_s390_inject_vm(struct kvm *kvm,
         if (waitqueue_active(&li->wq))
                 wake_up_interruptible(&li->wq);
         spin_unlock_bh(&li->lock);
-        spin_unlock_bh(&fi->lock);
+        spin_unlock(&fi->lock);
         mutex_unlock(&kvm->lock);
         return 0;
 }
diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c
index f4d56e9939c9..10bccd1f8aee 100644
--- a/arch/s390/kvm/kvm-s390.c
+++ b/arch/s390/kvm/kvm-s390.c
@@ -15,6 +15,7 @@
 #include <linux/compiler.h>
 #include <linux/err.h>
 #include <linux/fs.h>
+#include <linux/hrtimer.h>
 #include <linux/init.h>
 #include <linux/kvm.h>
 #include <linux/kvm_host.h>
@@ -195,6 +196,10 @@ out_nokvm:
 void kvm_arch_vcpu_destroy(struct kvm_vcpu *vcpu)
 {
         VCPU_EVENT(vcpu, 3, "%s", "free cpu");
+        if (vcpu->kvm->arch.sca->cpu[vcpu->vcpu_id].sda ==
+                (__u64) vcpu->arch.sie_block)
+                vcpu->kvm->arch.sca->cpu[vcpu->vcpu_id].sda = 0;
+        smp_mb();
         free_page((unsigned long)(vcpu->arch.sie_block));
         kvm_vcpu_uninit(vcpu);
         kfree(vcpu);
@@ -283,8 +288,10 @@ int kvm_arch_vcpu_setup(struct kvm_vcpu *vcpu)
         vcpu->arch.sie_block->gmsor = vcpu->kvm->arch.guest_origin;
         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);
+        hrtimer_init(&vcpu->arch.ckc_timer, CLOCK_REALTIME, HRTIMER_MODE_ABS);
+        tasklet_init(&vcpu->arch.tasklet, kvm_s390_tasklet,
+                     (unsigned long) vcpu);
+        vcpu->arch.ckc_timer.function = kvm_s390_idle_wakeup;
         get_cpu_id(&vcpu->arch.cpu_id);
         vcpu->arch.cpu_id.version = 0xff;
         return 0;
@@ -307,19 +314,21 @@ struct kvm_vcpu *kvm_arch_vcpu_create(struct kvm *kvm,
 
         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;
+        if (!kvm->arch.sca->cpu[id].sda)
+                kvm->arch.sca->cpu[id].sda = (__u64) vcpu->arch.sie_block;
+        else
+                BUG_ON(!kvm->vcpus[id]); /* vcpu does already exist */
         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);
+        spin_lock(&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);
+        spin_unlock(&kvm->arch.float_int.lock);
 
         rc = kvm_vcpu_init(vcpu, kvm, id);
         if (rc)
@@ -478,6 +487,12 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
 
         vcpu_load(vcpu);
 
+        /* verify, that memory has been registered */
+        if (!vcpu->kvm->arch.guest_memsize) {
+                vcpu_put(vcpu);
+                return -EINVAL;
+        }
+
         if (vcpu->sigset_active)
                 sigprocmask(SIG_SETMASK, &vcpu->sigset, &sigsaved);
 
@@ -657,6 +672,8 @@ int kvm_arch_set_memory_region(struct kvm *kvm,
                                 struct kvm_memory_slot old,
                                 int user_alloc)
 {
+        int i;
+
         /* 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.
@@ -664,7 +681,7 @@ int kvm_arch_set_memory_region(struct kvm *kvm,
            vmas. It is okay to mmap() and munmap() stuff in this slot after
            doing this call at any time */
 
-        if (mem->slot)
+        if (mem->slot || kvm->arch.guest_memsize)
                 return -EINVAL;
 
         if (mem->guest_phys_addr)
@@ -676,15 +693,39 @@ int kvm_arch_set_memory_region(struct kvm *kvm,
         if (mem->memory_size & (PAGE_SIZE - 1))
                 return -EINVAL;
 
+        if (!user_alloc)
+                return -EINVAL;
+
+        /* lock all vcpus */
+        for (i = 0; i < KVM_MAX_VCPUS; ++i) {
+                if (!kvm->vcpus[i])
+                        continue;
+                if (!mutex_trylock(&kvm->vcpus[i]->mutex))
+                        goto fail_out;
+        }
+
         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. */
+        /* update sie control blocks, and unlock all vcpus */
+        for (i = 0; i < KVM_MAX_VCPUS; ++i) {
+                if (kvm->vcpus[i]) {
+                        kvm->vcpus[i]->arch.sie_block->gmsor =
+                                kvm->arch.guest_origin;
+                        kvm->vcpus[i]->arch.sie_block->gmslm =
+                                kvm->arch.guest_memsize +
+                                kvm->arch.guest_origin +
+                                VIRTIODESCSPACE - 1ul;
+                        mutex_unlock(&kvm->vcpus[i]->mutex);
+                }
+        }
 
         return 0;
+
+fail_out:
+        for (; i >= 0; i--)
+                mutex_unlock(&kvm->vcpus[i]->mutex);
+        return -EINVAL;
 }
 
 void kvm_arch_flush_shadow(struct kvm *kvm)
diff --git a/arch/s390/kvm/kvm-s390.h b/arch/s390/kvm/kvm-s390.h
index 00bbe69b78da..748fee872323 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 <linux/hrtimer.h>
 #include <linux/kvm.h>
 #include <linux/kvm_host.h>
 
@@ -41,7 +42,8 @@ static inline int __cpu_is_stopped(struct kvm_vcpu *vcpu)
 }
 
 int kvm_s390_handle_wait(struct kvm_vcpu *vcpu);
-void kvm_s390_idle_wakeup(unsigned long data);
+enum hrtimer_restart kvm_s390_idle_wakeup(struct hrtimer *timer);
+void kvm_s390_tasklet(unsigned long parm);
 void kvm_s390_deliver_pending_interrupts(struct kvm_vcpu *vcpu);
 int kvm_s390_inject_vm(struct kvm *kvm,
                 struct kvm_s390_interrupt *s390int);
diff --git a/arch/s390/kvm/priv.c b/arch/s390/kvm/priv.c
index 4b88834b8dd8..93ecd06e1a74 100644
--- a/arch/s390/kvm/priv.c
+++ b/arch/s390/kvm/priv.c
@@ -204,11 +204,11 @@ static void handle_stsi_3_2_2(struct kvm_vcpu *vcpu, struct sysinfo_3_2_2 *mem)
         int cpus = 0;
         int n;
 
-        spin_lock_bh(&fi->lock);
+        spin_lock(&fi->lock);
         for (n = 0; n < KVM_MAX_VCPUS; n++)
                 if (fi->local_int[n])
                         cpus++;
-        spin_unlock_bh(&fi->lock);
+        spin_unlock(&fi->lock);
 
         /* deal with other level 3 hypervisors */
         if (stsi(mem, 3, 2, 2) == -ENOSYS)
diff --git a/arch/s390/kvm/sigp.c b/arch/s390/kvm/sigp.c
index f27dbedf0866..36678835034d 100644
--- a/arch/s390/kvm/sigp.c
+++ b/arch/s390/kvm/sigp.c
@@ -52,7 +52,7 @@ static int __sigp_sense(struct kvm_vcpu *vcpu, u16 cpu_addr,
         if (cpu_addr >= KVM_MAX_VCPUS)
                 return 3; /* not operational */
 
-        spin_lock_bh(&fi->lock);
+        spin_lock(&fi->lock);
         if (fi->local_int[cpu_addr] == NULL)
                 rc = 3; /* not operational */
         else if (atomic_read(fi->local_int[cpu_addr]->cpuflags)
@@ -64,7 +64,7 @@ static int __sigp_sense(struct kvm_vcpu *vcpu, u16 cpu_addr,
                 *reg |= SIGP_STAT_STOPPED;
                 rc = 1; /* status stored */
         }
-        spin_unlock_bh(&fi->lock);
+        spin_unlock(&fi->lock);
 
         VCPU_EVENT(vcpu, 4, "sensed status of cpu %x rc %x", cpu_addr, rc);
         return rc;
@@ -86,7 +86,7 @@ static int __sigp_emergency(struct kvm_vcpu *vcpu, u16 cpu_addr)
 
         inti->type = KVM_S390_INT_EMERGENCY;
 
-        spin_lock_bh(&fi->lock);
+        spin_lock(&fi->lock);
         li = fi->local_int[cpu_addr];
         if (li == NULL) {
                 rc = 3; /* not operational */
@@ -102,7 +102,7 @@ static int __sigp_emergency(struct kvm_vcpu *vcpu, u16 cpu_addr)
         spin_unlock_bh(&li->lock);
         rc = 0; /* order accepted */
 unlock:
-        spin_unlock_bh(&fi->lock);
+        spin_unlock(&fi->lock);
         VCPU_EVENT(vcpu, 4, "sent sigp emerg to cpu %x", cpu_addr);
         return rc;
 }
@@ -123,7 +123,7 @@ static int __sigp_stop(struct kvm_vcpu *vcpu, u16 cpu_addr, int store)
 
         inti->type = KVM_S390_SIGP_STOP;
 
-        spin_lock_bh(&fi->lock);
+        spin_lock(&fi->lock);
         li = fi->local_int[cpu_addr];
         if (li == NULL) {
                 rc = 3; /* not operational */
@@ -142,7 +142,7 @@ static int __sigp_stop(struct kvm_vcpu *vcpu, u16 cpu_addr, int store)
         spin_unlock_bh(&li->lock);
         rc = 0; /* order accepted */
 unlock:
-        spin_unlock_bh(&fi->lock);
+        spin_unlock(&fi->lock);
         VCPU_EVENT(vcpu, 4, "sent sigp stop to cpu %x", cpu_addr);
         return rc;
 }
@@ -188,7 +188,7 @@ static int __sigp_set_prefix(struct kvm_vcpu *vcpu, u16 cpu_addr, u32 address,
         if (!inti)
                 return 2; /* busy */
 
-        spin_lock_bh(&fi->lock);
+        spin_lock(&fi->lock);
         li = fi->local_int[cpu_addr];
 
         if ((cpu_addr >= KVM_MAX_VCPUS) || (li == NULL)) {
@@ -220,7 +220,7 @@ static int __sigp_set_prefix(struct kvm_vcpu *vcpu, u16 cpu_addr, u32 address,
 out_li:
         spin_unlock_bh(&li->lock);
 out_fi:
-        spin_unlock_bh(&fi->lock);
+        spin_unlock(&fi->lock);
         return rc;
 }