1 files changed, 92 insertions, 55 deletions

diff --git a/drivers/lguest/lguest_user.c b/drivers/lguest/lguest_user.c
index 3b92a61ba8d2..85d42d3d01a9 100644
--- a/drivers/lguest/lguest_user.c
+++ b/drivers/lguest/lguest_user.c
@@ -6,6 +6,7 @@
 #include <linux/uaccess.h>
 #include <linux/miscdevice.h>
 #include <linux/fs.h>
+#include <linux/sched.h>
 #include "lg.h"
 
 /*L:055 When something happens, the Waker process needs a way to stop the
@@ -13,7 +14,7 @@
  * LHREQ_BREAK and the value "1" to /dev/lguest to do this.  Once the Launcher
  * has done whatever needs attention, it writes LHREQ_BREAK and "0" to release
  * the Waker. */
-static int break_guest_out(struct lguest *lg, const unsigned long __user *input)
+static int break_guest_out(struct lg_cpu *cpu, const unsigned long __user*input)
 {
         unsigned long on;
 
@@ -22,21 +23,21 @@ static int break_guest_out(struct lguest *lg, const unsigned long __user *input)
                 return -EFAULT;
 
         if (on) {
-                lg->break_out = 1;
+                cpu->break_out = 1;
                 /* Pop it out of the Guest (may be running on different CPU) */
-                wake_up_process(lg->tsk);
+                wake_up_process(cpu->tsk);
                 /* Wait for them to reset it */
-                return wait_event_interruptible(lg->break_wq, !lg->break_out);
+                return wait_event_interruptible(cpu->break_wq, !cpu->break_out);
         } else {
-                lg->break_out = 0;
-                wake_up(&lg->break_wq);
+                cpu->break_out = 0;
+                wake_up(&cpu->break_wq);
                 return 0;
         }
 }
 
 /*L:050 Sending an interrupt is done by writing LHREQ_IRQ and an interrupt
  * number to /dev/lguest. */
-static int user_send_irq(struct lguest *lg, const unsigned long __user *input)
+static int user_send_irq(struct lg_cpu *cpu, const unsigned long __user *input)
 {
         unsigned long irq;
 
@@ -46,7 +47,7 @@ static int user_send_irq(struct lguest *lg, const unsigned long __user *input)
                 return -EINVAL;
         /* Next time the Guest runs, the core code will see if it can deliver
          * this interrupt. */
-        set_bit(irq, lg->irqs_pending);
+        set_bit(irq, cpu->irqs_pending);
         return 0;
 }
 
@@ -55,13 +56,21 @@ static int user_send_irq(struct lguest *lg, const unsigned long __user *input)
 static ssize_t read(struct file *file, char __user *user, size_t size,loff_t*o)
 {
         struct lguest *lg = file->private_data;
+        struct lg_cpu *cpu;
+        unsigned int cpu_id = *o;
 
         /* You must write LHREQ_INITIALIZE first! */
         if (!lg)
                 return -EINVAL;
 
+        /* Watch out for arbitrary vcpu indexes! */
+        if (cpu_id >= lg->nr_cpus)
+                return -EINVAL;
+
+        cpu = &lg->cpus[cpu_id];
+
         /* If you're not the task which owns the Guest, go away. */
-        if (current != lg->tsk)
+        if (current != cpu->tsk)
                 return -EPERM;
 
         /* If the guest is already dead, we indicate why */
@@ -81,11 +90,53 @@ static ssize_t read(struct file *file, char __user *user, size_t size,loff_t*o)
 
         /* If we returned from read() last time because the Guest notified,
          * clear the flag. */
-        if (lg->pending_notify)
-                lg->pending_notify = 0;
+        if (cpu->pending_notify)
+                cpu->pending_notify = 0;
 
         /* Run the Guest until something interesting happens. */
-        return run_guest(lg, (unsigned long __user *)user);
+        return run_guest(cpu, (unsigned long __user *)user);
+}
+
+static int lg_cpu_start(struct lg_cpu *cpu, unsigned id, unsigned long start_ip)
+{
+        if (id >= NR_CPUS)
+                return -EINVAL;
+
+        cpu->id = id;
+        cpu->lg = container_of((cpu - id), struct lguest, cpus[0]);
+        cpu->lg->nr_cpus++;
+        init_clockdev(cpu);
+
+        /* We need a complete page for the Guest registers: they are accessible
+         * to the Guest and we can only grant it access to whole pages. */
+        cpu->regs_page = get_zeroed_page(GFP_KERNEL);
+        if (!cpu->regs_page)
+                return -ENOMEM;
+
+        /* We actually put the registers at the bottom of the page. */
+        cpu->regs = (void *)cpu->regs_page + PAGE_SIZE - sizeof(*cpu->regs);
+
+        /* Now we initialize the Guest's registers, handing it the start
+         * address. */
+        lguest_arch_setup_regs(cpu, start_ip);
+
+        /* Initialize the queue for the waker to wait on */
+        init_waitqueue_head(&cpu->break_wq);
+
+        /* We keep a pointer to the Launcher task (ie. current task) for when
+         * other Guests want to wake this one (inter-Guest I/O). */
+        cpu->tsk = current;
+
+        /* We need to keep a pointer to the Launcher's memory map, because if
+         * the Launcher dies we need to clean it up.  If we don't keep a
+         * reference, it is destroyed before close() is called. */
+        cpu->mm = get_task_mm(cpu->tsk);
+
+        /* We remember which CPU's pages this Guest used last, for optimization
+         * when the same Guest runs on the same CPU twice. */
+        cpu->last_pages = NULL;
+
+        return 0;
 }
 
 /*L:020 The initialization write supplies 4 pointer sized (32 or 64 bit)
@@ -134,15 +185,10 @@ static int initialize(struct file *file, const unsigned long __user *input)
         lg->mem_base = (void __user *)(long)args[0];
         lg->pfn_limit = args[1];
 
-        /* We need a complete page for the Guest registers: they are accessible
-         * to the Guest and we can only grant it access to whole pages. */
-        lg->regs_page = get_zeroed_page(GFP_KERNEL);
-        if (!lg->regs_page) {
-                err = -ENOMEM;
+        /* This is the first cpu */
+        err = lg_cpu_start(&lg->cpus[0], 0, args[3]);
+        if (err)
                 goto release_guest;
-        }
-        /* We actually put the registers at the bottom of the page. */
-        lg->regs = (void *)lg->regs_page + PAGE_SIZE - sizeof(*lg->regs);
 
         /* Initialize the Guest's shadow page tables, using the toplevel
          * address the Launcher gave us.  This allocates memory, so can
@@ -151,28 +197,6 @@ static int initialize(struct file *file, const unsigned long __user *input)
         if (err)
                 goto free_regs;
 
-        /* Now we initialize the Guest's registers, handing it the start
-         * address. */
-        lguest_arch_setup_regs(lg, args[3]);
-
-        /* The timer for lguest's clock needs initialization. */
-        init_clockdev(lg);
-
-        /* We keep a pointer to the Launcher task (ie. current task) for when
-         * other Guests want to wake this one (inter-Guest I/O). */
-        lg->tsk = current;
-        /* We need to keep a pointer to the Launcher's memory map, because if
-         * the Launcher dies we need to clean it up.  If we don't keep a
-         * reference, it is destroyed before close() is called. */
-        lg->mm = get_task_mm(lg->tsk);
-
-        /* Initialize the queue for the waker to wait on */
-        init_waitqueue_head(&lg->break_wq);
-
-        /* We remember which CPU's pages this Guest used last, for optimization
-         * when the same Guest runs on the same CPU twice. */
-        lg->last_pages = NULL;
-
         /* We keep our "struct lguest" in the file's private_data. */
         file->private_data = lg;
 
@@ -182,7 +206,8 @@ static int initialize(struct file *file, const unsigned long __user *input)
         return sizeof(args);
 
 free_regs:
-        free_page(lg->regs_page);
+        /* FIXME: This should be in free_vcpu */
+        free_page(lg->cpus[0].regs_page);
 release_guest:
         kfree(lg);
 unlock:
@@ -202,30 +227,37 @@ static ssize_t write(struct file *file, const char __user *in,
         struct lguest *lg = file->private_data;
         const unsigned long __user *input = (const unsigned long __user *)in;
         unsigned long req;
+        struct lg_cpu *uninitialized_var(cpu);
+        unsigned int cpu_id = *off;
 
         if (get_user(req, input) != 0)
                 return -EFAULT;
         input++;
 
         /* If you haven't initialized, you must do that first. */
-        if (req != LHREQ_INITIALIZE && !lg)
-                return -EINVAL;
+        if (req != LHREQ_INITIALIZE) {
+                if (!lg || (cpu_id >= lg->nr_cpus))
+                        return -EINVAL;
+                cpu = &lg->cpus[cpu_id];
+                if (!cpu)
+                        return -EINVAL;
+        }
 
         /* Once the Guest is dead, all you can do is read() why it died. */
         if (lg && lg->dead)
                 return -ENOENT;
 
         /* If you're not the task which owns the Guest, you can only break */
-        if (lg && current != lg->tsk && req != LHREQ_BREAK)
+        if (lg && current != cpu->tsk && req != LHREQ_BREAK)
                 return -EPERM;
 
         switch (req) {
         case LHREQ_INITIALIZE:
                 return initialize(file, input);
         case LHREQ_IRQ:
-                return user_send_irq(lg, input);
+                return user_send_irq(cpu, input);
         case LHREQ_BREAK:
-                return break_guest_out(lg, input);
+                return break_guest_out(cpu, input);
         default:
                 return -EINVAL;
         }
@@ -241,6 +273,7 @@ static ssize_t write(struct file *file, const char __user *in,
 static int close(struct inode *inode, struct file *file)
 {
         struct lguest *lg = file->private_data;
+        unsigned int i;
 
         /* If we never successfully initialized, there's nothing to clean up */
         if (!lg)
@@ -249,19 +282,23 @@ static int close(struct inode *inode, struct file *file)
         /* We need the big lock, to protect from inter-guest I/O and other
          * Launchers initializing guests. */
         mutex_lock(&lguest_lock);
-        /* Cancels the hrtimer set via LHCALL_SET_CLOCKEVENT. */
-        hrtimer_cancel(&lg->hrt);
+
         /* Free up the shadow page tables for the Guest. */
         free_guest_pagetable(lg);
-        /* Now all the memory cleanups are done, it's safe to release the
-         * Launcher's memory management structure. */
-        mmput(lg->mm);
+
+        for (i = 0; i < lg->nr_cpus; i++) {
+                /* Cancels the hrtimer set via LHCALL_SET_CLOCKEVENT. */
+                hrtimer_cancel(&lg->cpus[i].hrt);
+                /* We can free up the register page we allocated. */
+                free_page(lg->cpus[i].regs_page);
+                /* Now all the memory cleanups are done, it's safe to release
+                 * the Launcher's memory management structure. */
+                mmput(lg->cpus[i].mm);
+        }
         /* If lg->dead doesn't contain an error code it will be NULL or a
          * kmalloc()ed string, either of which is ok to hand to kfree(). */
         if (!IS_ERR(lg->dead))
                 kfree(lg->dead);
-        /* We can free up the register page we allocated. */
-        free_page(lg->regs_page);
         /* We clear the entire structure, which also marks it as free for the
          * next user. */
         memset(lg, 0, sizeof(*lg));