6 files changed, 547 insertions, 25 deletions

diff --git a/drivers/xen/xenfs/Makefile b/drivers/xen/xenfs/Makefile
index 25275c3bbdff..4fde9440fe1f 100644
--- a/drivers/xen/xenfs/Makefile
+++ b/drivers/xen/xenfs/Makefile
@@ -1,3 +1,4 @@
 obj-$(CONFIG_XENFS) += xenfs.o
 
-xenfs-objs = super.o xenbus.o
\ No newline at end of file
+xenfs-y                   = super.o xenbus.o privcmd.o
+xenfs-$(CONFIG_XEN_DOM0) += xenstored.o
diff --git a/drivers/xen/xenfs/privcmd.c b/drivers/xen/xenfs/privcmd.c
new file mode 100644
index 000000000000..dbd3b16fd131
--- /dev/null
+++ b/drivers/xen/xenfs/privcmd.c
@@ -0,0 +1,400 @@
+/******************************************************************************
+ * privcmd.c
+ *
+ * Interface to privileged domain-0 commands.
+ *
+ * Copyright (c) 2002-2004, K A Fraser, B Dragovic
+ */
+
+#include <linux/kernel.h>
+#include <linux/sched.h>
+#include <linux/slab.h>
+#include <linux/string.h>
+#include <linux/errno.h>
+#include <linux/mm.h>
+#include <linux/mman.h>
+#include <linux/uaccess.h>
+#include <linux/swap.h>
+#include <linux/highmem.h>
+#include <linux/pagemap.h>
+#include <linux/seq_file.h>
+
+#include <asm/pgalloc.h>
+#include <asm/pgtable.h>
+#include <asm/tlb.h>
+#include <asm/xen/hypervisor.h>
+#include <asm/xen/hypercall.h>
+
+#include <xen/xen.h>
+#include <xen/privcmd.h>
+#include <xen/interface/xen.h>
+#include <xen/features.h>
+#include <xen/page.h>
+#include <xen/xen-ops.h>
+
+#ifndef HAVE_ARCH_PRIVCMD_MMAP
+static int privcmd_enforce_singleshot_mapping(struct vm_area_struct *vma);
+#endif
+
+static long privcmd_ioctl_hypercall(void __user *udata)
+{
+        struct privcmd_hypercall hypercall;
+        long ret;
+
+        if (copy_from_user(&hypercall, udata, sizeof(hypercall)))
+                return -EFAULT;
+
+        ret = privcmd_call(hypercall.op,
+                           hypercall.arg[0], hypercall.arg[1],
+                           hypercall.arg[2], hypercall.arg[3],
+                           hypercall.arg[4]);
+
+        return ret;
+}
+
+static void free_page_list(struct list_head *pages)
+{
+        struct page *p, *n;
+
+        list_for_each_entry_safe(p, n, pages, lru)
+                __free_page(p);
+
+        INIT_LIST_HEAD(pages);
+}
+
+/*
+ * Given an array of items in userspace, return a list of pages
+ * containing the data.  If copying fails, either because of memory
+ * allocation failure or a problem reading user memory, return an
+ * error code; its up to the caller to dispose of any partial list.
+ */
+static int gather_array(struct list_head *pagelist,
+                        unsigned nelem, size_t size,
+                        void __user *data)
+{
+        unsigned pageidx;
+        void *pagedata;
+        int ret;
+
+        if (size > PAGE_SIZE)
+                return 0;
+
+        pageidx = PAGE_SIZE;
+        pagedata = NULL;        /* quiet, gcc */
+        while (nelem--) {
+                if (pageidx > PAGE_SIZE-size) {
+                        struct page *page = alloc_page(GFP_KERNEL);
+
+                        ret = -ENOMEM;
+                        if (page == NULL)
+                                goto fail;
+
+                        pagedata = page_address(page);
+
+                        list_add_tail(&page->lru, pagelist);
+                        pageidx = 0;
+                }
+
+                ret = -EFAULT;
+                if (copy_from_user(pagedata + pageidx, data, size))
+                        goto fail;
+
+                data += size;
+                pageidx += size;
+        }
+
+        ret = 0;
+
+fail:
+        return ret;
+}
+
+/*
+ * Call function "fn" on each element of the array fragmented
+ * over a list of pages.
+ */
+static int traverse_pages(unsigned nelem, size_t size,
+                          struct list_head *pos,
+                          int (*fn)(void *data, void *state),
+                          void *state)
+{
+        void *pagedata;
+        unsigned pageidx;
+        int ret = 0;
+
+        BUG_ON(size > PAGE_SIZE);
+
+        pageidx = PAGE_SIZE;
+        pagedata = NULL;        /* hush, gcc */
+
+        while (nelem--) {
+                if (pageidx > PAGE_SIZE-size) {
+                        struct page *page;
+                        pos = pos->next;
+                        page = list_entry(pos, struct page, lru);
+                        pagedata = page_address(page);
+                        pageidx = 0;
+                }
+
+                ret = (*fn)(pagedata + pageidx, state);
+                if (ret)
+                        break;
+                pageidx += size;
+        }
+
+        return ret;
+}
+
+struct mmap_mfn_state {
+        unsigned long va;
+        struct vm_area_struct *vma;
+        domid_t domain;
+};
+
+static int mmap_mfn_range(void *data, void *state)
+{
+        struct privcmd_mmap_entry *msg = data;
+        struct mmap_mfn_state *st = state;
+        struct vm_area_struct *vma = st->vma;
+        int rc;
+
+        /* Do not allow range to wrap the address space. */
+        if ((msg->npages > (LONG_MAX >> PAGE_SHIFT)) ||
+            ((unsigned long)(msg->npages << PAGE_SHIFT) >= -st->va))
+                return -EINVAL;
+
+        /* Range chunks must be contiguous in va space. */
+        if ((msg->va != st->va) ||
+            ((msg->va+(msg->npages<<PAGE_SHIFT)) > vma->vm_end))
+                return -EINVAL;
+
+        rc = xen_remap_domain_mfn_range(vma,
+                                        msg->va & PAGE_MASK,
+                                        msg->mfn, msg->npages,
+                                        vma->vm_page_prot,
+                                        st->domain);
+        if (rc < 0)
+                return rc;
+
+        st->va += msg->npages << PAGE_SHIFT;
+
+        return 0;
+}
+
+static long privcmd_ioctl_mmap(void __user *udata)
+{
+        struct privcmd_mmap mmapcmd;
+        struct mm_struct *mm = current->mm;
+        struct vm_area_struct *vma;
+        int rc;
+        LIST_HEAD(pagelist);
+        struct mmap_mfn_state state;
+
+        if (!xen_initial_domain())
+                return -EPERM;
+
+        if (copy_from_user(&mmapcmd, udata, sizeof(mmapcmd)))
+                return -EFAULT;
+
+        rc = gather_array(&pagelist,
+                          mmapcmd.num, sizeof(struct privcmd_mmap_entry),
+                          mmapcmd.entry);
+
+        if (rc || list_empty(&pagelist))
+                goto out;
+
+        down_write(&mm->mmap_sem);
+
+        {
+                struct page *page = list_first_entry(&pagelist,
+                                                     struct page, lru);
+                struct privcmd_mmap_entry *msg = page_address(page);
+
+                vma = find_vma(mm, msg->va);
+                rc = -EINVAL;
+
+                if (!vma || (msg->va != vma->vm_start) ||
+                    !privcmd_enforce_singleshot_mapping(vma))
+                        goto out_up;
+        }
+
+        state.va = vma->vm_start;
+        state.vma = vma;
+        state.domain = mmapcmd.dom;
+
+        rc = traverse_pages(mmapcmd.num, sizeof(struct privcmd_mmap_entry),
+                            &pagelist,
+                            mmap_mfn_range, &state);
+
+
+out_up:
+        up_write(&mm->mmap_sem);
+
+out:
+        free_page_list(&pagelist);
+
+        return rc;
+}
+
+struct mmap_batch_state {
+        domid_t domain;
+        unsigned long va;
+        struct vm_area_struct *vma;
+        int err;
+
+        xen_pfn_t __user *user;
+};
+
+static int mmap_batch_fn(void *data, void *state)
+{
+        xen_pfn_t *mfnp = data;
+        struct mmap_batch_state *st = state;
+
+        if (xen_remap_domain_mfn_range(st->vma, st->va & PAGE_MASK, *mfnp, 1,
+                                       st->vma->vm_page_prot, st->domain) < 0) {
+                *mfnp |= 0xf0000000U;
+                st->err++;
+        }
+        st->va += PAGE_SIZE;
+
+        return 0;
+}
+
+static int mmap_return_errors(void *data, void *state)
+{
+        xen_pfn_t *mfnp = data;
+        struct mmap_batch_state *st = state;
+
+        return put_user(*mfnp, st->user++);
+}
+
+static struct vm_operations_struct privcmd_vm_ops;
+
+static long privcmd_ioctl_mmap_batch(void __user *udata)
+{
+        int ret;
+        struct privcmd_mmapbatch m;
+        struct mm_struct *mm = current->mm;
+        struct vm_area_struct *vma;
+        unsigned long nr_pages;
+        LIST_HEAD(pagelist);
+        struct mmap_batch_state state;
+
+        if (!xen_initial_domain())
+                return -EPERM;
+
+        if (copy_from_user(&m, udata, sizeof(m)))
+                return -EFAULT;
+
+        nr_pages = m.num;
+        if ((m.num <= 0) || (nr_pages > (LONG_MAX >> PAGE_SHIFT)))
+                return -EINVAL;
+
+        ret = gather_array(&pagelist, m.num, sizeof(xen_pfn_t),
+                           m.arr);
+
+        if (ret || list_empty(&pagelist))
+                goto out;
+
+        down_write(&mm->mmap_sem);
+
+        vma = find_vma(mm, m.addr);
+        ret = -EINVAL;
+        if (!vma ||
+            vma->vm_ops != &privcmd_vm_ops ||
+            (m.addr != vma->vm_start) ||
+            ((m.addr + (nr_pages << PAGE_SHIFT)) != vma->vm_end) ||
+            !privcmd_enforce_singleshot_mapping(vma)) {
+                up_write(&mm->mmap_sem);
+                goto out;
+        }
+
+        state.domain = m.dom;
+        state.vma = vma;
+        state.va = m.addr;
+        state.err = 0;
+
+        ret = traverse_pages(m.num, sizeof(xen_pfn_t),
+                             &pagelist, mmap_batch_fn, &state);
+
+        up_write(&mm->mmap_sem);
+
+        if (state.err > 0) {
+                state.user = m.arr;
+                ret = traverse_pages(m.num, sizeof(xen_pfn_t),
+                               &pagelist,
+                               mmap_return_errors, &state);
+        }
+
+out:
+        free_page_list(&pagelist);
+
+        return ret;
+}
+
+static long privcmd_ioctl(struct file *file,
+                          unsigned int cmd, unsigned long data)
+{
+        int ret = -ENOSYS;
+        void __user *udata = (void __user *) data;
+
+        switch (cmd) {
+        case IOCTL_PRIVCMD_HYPERCALL:
+                ret = privcmd_ioctl_hypercall(udata);
+                break;
+
+        case IOCTL_PRIVCMD_MMAP:
+                ret = privcmd_ioctl_mmap(udata);
+                break;
+
+        case IOCTL_PRIVCMD_MMAPBATCH:
+                ret = privcmd_ioctl_mmap_batch(udata);
+                break;
+
+        default:
+                ret = -EINVAL;
+                break;
+        }
+
+        return ret;
+}
+
+#ifndef HAVE_ARCH_PRIVCMD_MMAP
+static int privcmd_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
+{
+        printk(KERN_DEBUG "privcmd_fault: vma=%p %lx-%lx, pgoff=%lx, uv=%p\n",
+               vma, vma->vm_start, vma->vm_end,
+               vmf->pgoff, vmf->virtual_address);
+
+        return VM_FAULT_SIGBUS;
+}
+
+static struct vm_operations_struct privcmd_vm_ops = {
+        .fault = privcmd_fault
+};
+
+static int privcmd_mmap(struct file *file, struct vm_area_struct *vma)
+{
+        /* Unsupported for auto-translate guests. */
+        if (xen_feature(XENFEAT_auto_translated_physmap))
+                return -ENOSYS;
+
+        /* DONTCOPY is essential for Xen because copy_page_range doesn't know
+         * how to recreate these mappings */
+        vma->vm_flags |= VM_RESERVED | VM_IO | VM_DONTCOPY | VM_PFNMAP;
+        vma->vm_ops = &privcmd_vm_ops;
+        vma->vm_private_data = NULL;
+
+        return 0;
+}
+
+static int privcmd_enforce_singleshot_mapping(struct vm_area_struct *vma)
+{
+        return (xchg(&vma->vm_private_data, (void *)1) == NULL);
+}
+#endif
+
+const struct file_operations privcmd_file_ops = {
+        .unlocked_ioctl = privcmd_ioctl,
+        .mmap = privcmd_mmap,
+};
diff --git a/drivers/xen/xenfs/super.c b/drivers/xen/xenfs/super.c
index 78bfab0700ba..1aa389719846 100644
--- a/drivers/xen/xenfs/super.c
+++ b/drivers/xen/xenfs/super.c
@@ -22,6 +22,46 @@
 MODULE_DESCRIPTION("Xen filesystem");
 MODULE_LICENSE("GPL");
 
+static struct inode *xenfs_make_inode(struct super_block *sb, int mode)
+{
+        struct inode *ret = new_inode(sb);
+
+        if (ret) {
+                ret->i_mode = mode;
+                ret->i_uid = ret->i_gid = 0;
+                ret->i_blocks = 0;
+                ret->i_atime = ret->i_mtime = ret->i_ctime = CURRENT_TIME;
+        }
+        return ret;
+}
+
+static struct dentry *xenfs_create_file(struct super_block *sb,
+                                        struct dentry *parent,
+                                        const char *name,
+                                        const struct file_operations *fops,
+                                        void *data,
+                                        int mode)
+{
+        struct dentry *dentry;
+        struct inode *inode;
+
+        dentry = d_alloc_name(parent, name);
+        if (!dentry)
+                return NULL;
+
+        inode = xenfs_make_inode(sb, S_IFREG | mode);
+        if (!inode) {
+                dput(dentry);
+                return NULL;
+        }
+
+        inode->i_fop = fops;
+        inode->i_private = data;
+
+        d_add(dentry, inode);
+        return dentry;
+}
+
 static ssize_t capabilities_read(struct file *file, char __user *buf,
                                  size_t size, loff_t *off)
 {
@@ -35,6 +75,7 @@ static ssize_t capabilities_read(struct file *file, char __user *buf,
 
 static const struct file_operations capabilities_file_ops = {
         .read = capabilities_read,
+        .llseek = default_llseek,
 };
 
 static int xenfs_fill_super(struct super_block *sb, void *data, int silent)
@@ -43,23 +84,36 @@ static int xenfs_fill_super(struct super_block *sb, void *data, int silent)
                 [1] = {},
                 { "xenbus", &xenbus_file_ops, S_IRUSR|S_IWUSR },
                 { "capabilities", &capabilities_file_ops, S_IRUGO },
+                { "privcmd", &privcmd_file_ops, S_IRUSR|S_IWUSR },
                 {""},
         };
+        int rc;
+
+        rc = simple_fill_super(sb, XENFS_SUPER_MAGIC, xenfs_files);
+        if (rc < 0)
+                return rc;
+
+        if (xen_initial_domain()) {
+                xenfs_create_file(sb, sb->s_root, "xsd_kva",
+                                  &xsd_kva_file_ops, NULL, S_IRUSR|S_IWUSR);
+                xenfs_create_file(sb, sb->s_root, "xsd_port",
+                                  &xsd_port_file_ops, NULL, S_IRUSR|S_IWUSR);
+        }
 
-        return simple_fill_super(sb, XENFS_SUPER_MAGIC, xenfs_files);
+        return rc;
 }
 
-static int xenfs_get_sb(struct file_system_type *fs_type,
-                        int flags, const char *dev_name,
-                        void *data, struct vfsmount *mnt)
+static struct dentry *xenfs_mount(struct file_system_type *fs_type,
+                                  int flags, const char *dev_name,
+                                  void *data)
 {
-        return get_sb_single(fs_type, flags, data, xenfs_fill_super, mnt);
+        return mount_single(fs_type, flags, data, xenfs_fill_super);
 }
 
 static struct file_system_type xenfs_type = {
         .owner =        THIS_MODULE,
         .name =         "xenfs",
-        .get_sb =       xenfs_get_sb,
+        .mount =        xenfs_mount,
         .kill_sb =      kill_litter_super,
 };
 
diff --git a/drivers/xen/xenfs/xenbus.c b/drivers/xen/xenfs/xenbus.c
index 3b39c3752e21..bbd000f88af7 100644
--- a/drivers/xen/xenfs/xenbus.c
+++ b/drivers/xen/xenfs/xenbus.c
@@ -122,6 +122,7 @@ static ssize_t xenbus_file_read(struct file *filp,
         int ret;
 
         mutex_lock(&u->reply_mutex);
+again:
         while (list_empty(&u->read_buffers)) {
                 mutex_unlock(&u->reply_mutex);
                 if (filp->f_flags & O_NONBLOCK)
@@ -144,7 +145,7 @@ static ssize_t xenbus_file_read(struct file *filp,
                 i += sz - ret;
                 rb->cons += sz - ret;
 
-                if (ret != sz) {
+                if (ret != 0) {
                         if (i == 0)
                                 i = -EFAULT;
                         goto out;
@@ -160,6 +161,8 @@ static ssize_t xenbus_file_read(struct file *filp,
                                         struct read_buffer, list);
                 }
         }
+        if (i == 0)
+                goto again;
 
 out:
         mutex_unlock(&u->reply_mutex);
@@ -407,6 +410,7 @@ static int xenbus_write_watch(unsigned msg_type, struct xenbus_file_priv *u)
 
                 mutex_lock(&u->reply_mutex);
                 rc = queue_reply(&u->read_buffers, &reply, sizeof(reply));
+                wake_up(&u->read_waitq);
                 mutex_unlock(&u->reply_mutex);
         }
 
@@ -455,7 +459,7 @@ static ssize_t xenbus_file_write(struct file *filp,
 
         ret = copy_from_user(u->u.buffer + u->len, ubuf, len);
 
-        if (ret == len) {
+        if (ret != 0) {
                 rc = -EFAULT;
                 goto out;
         }
@@ -488,21 +492,6 @@ static ssize_t xenbus_file_write(struct file *filp,
         msg_type = u->u.msg.type;
 
         switch (msg_type) {
-        case XS_TRANSACTION_START:
-        case XS_TRANSACTION_END:
-        case XS_DIRECTORY:
-        case XS_READ:
-        case XS_GET_PERMS:
-        case XS_RELEASE:
-        case XS_GET_DOMAIN_PATH:
-        case XS_WRITE:
-        case XS_MKDIR:
-        case XS_RM:
-        case XS_SET_PERMS:
-                /* Send out a transaction */
-                ret = xenbus_write_transaction(msg_type, u);
-                break;
-
         case XS_WATCH:
         case XS_UNWATCH:
                 /* (Un)Ask for some path to be watched for changes */
@@ -510,7 +499,8 @@ static ssize_t xenbus_file_write(struct file *filp,
                 break;
 
         default:
-                ret = -EINVAL;
+                /* Send out a transaction */
+                ret = xenbus_write_transaction(msg_type, u);
                 break;
         }
         if (ret != 0)
@@ -555,6 +545,7 @@ static int xenbus_file_release(struct inode *inode, struct file *filp)
         struct xenbus_file_priv *u = filp->private_data;
         struct xenbus_transaction_holder *trans, *tmp;
         struct watch_adapter *watch, *tmp_watch;
+        struct read_buffer *rb, *tmp_rb;
 
         /*
          * No need for locking here because there are no other users,
@@ -573,6 +564,10 @@ static int xenbus_file_release(struct inode *inode, struct file *filp)
                 free_watch_adapter(watch);
         }
 
+        list_for_each_entry_safe(rb, tmp_rb, &u->read_buffers, list) {
+                list_del(&rb->list);
+                kfree(rb);
+        }
         kfree(u);
 
         return 0;
@@ -594,4 +589,5 @@ const struct file_operations xenbus_file_ops = {
         .open = xenbus_file_open,
         .release = xenbus_file_release,
         .poll = xenbus_file_poll,
+        .llseek = no_llseek,
 };
diff --git a/drivers/xen/xenfs/xenfs.h b/drivers/xen/xenfs/xenfs.h
index 51f08b2d0bf1..b68aa6200003 100644
--- a/drivers/xen/xenfs/xenfs.h
+++ b/drivers/xen/xenfs/xenfs.h
@@ -2,5 +2,8 @@
 #define _XENFS_XENBUS_H
 
 extern const struct file_operations xenbus_file_ops;
+extern const struct file_operations privcmd_file_ops;
+extern const struct file_operations xsd_kva_file_ops;
+extern const struct file_operations xsd_port_file_ops;
 
 #endif  /* _XENFS_XENBUS_H */
diff --git a/drivers/xen/xenfs/xenstored.c b/drivers/xen/xenfs/xenstored.c
new file mode 100644
index 000000000000..fef20dbc6a5c
--- /dev/null
+++ b/drivers/xen/xenfs/xenstored.c
@@ -0,0 +1,68 @@
+#include <linux/slab.h>
+#include <linux/types.h>
+#include <linux/mm.h>
+#include <linux/fs.h>
+
+#include <xen/page.h>
+
+#include "xenfs.h"
+#include "../xenbus/xenbus_comms.h"
+
+static ssize_t xsd_read(struct file *file, char __user *buf,
+                            size_t size, loff_t *off)
+{
+        const char *str = (const char *)file->private_data;
+        return simple_read_from_buffer(buf, size, off, str, strlen(str));
+}
+
+static int xsd_release(struct inode *inode, struct file *file)
+{
+        kfree(file->private_data);
+        return 0;
+}
+
+static int xsd_kva_open(struct inode *inode, struct file *file)
+{
+        file->private_data = (void *)kasprintf(GFP_KERNEL, "0x%p",
+                                               xen_store_interface);
+        if (!file->private_data)
+                return -ENOMEM;
+        return 0;
+}
+
+static int xsd_kva_mmap(struct file *file, struct vm_area_struct *vma)
+{
+        size_t size = vma->vm_end - vma->vm_start;
+
+        if ((size > PAGE_SIZE) || (vma->vm_pgoff != 0))
+                return -EINVAL;
+
+        if (remap_pfn_range(vma, vma->vm_start,
+                            virt_to_pfn(xen_store_interface),
+                            size, vma->vm_page_prot))
+                return -EAGAIN;
+
+        return 0;
+}
+
+const struct file_operations xsd_kva_file_ops = {
+        .open = xsd_kva_open,
+        .mmap = xsd_kva_mmap,
+        .read = xsd_read,
+        .release = xsd_release,
+};
+
+static int xsd_port_open(struct inode *inode, struct file *file)
+{
+        file->private_data = (void *)kasprintf(GFP_KERNEL, "%d",
+                                               xen_store_evtchn);
+        if (!file->private_data)
+                return -ENOMEM;
+        return 0;
+}
+
+const struct file_operations xsd_port_file_ops = {
+        .open = xsd_port_open,
+        .read = xsd_read,
+        .release = xsd_release,
+};