ocfs2: fix the end cluster offset of FIEMAP
[linux-3.10.git] / fs / read_write.c
index b60324a..2cefa41 100644 (file)
 #include <linux/fcntl.h>
 #include <linux/file.h>
 #include <linux/uio.h>
-#include <linux/smp_lock.h>
+#include <linux/aio.h>
 #include <linux/fsnotify.h>
 #include <linux/security.h>
-#include <linux/module.h>
+#include <linux/export.h>
 #include <linux/syscalls.h>
+#include <linux/pagemap.h>
+#include <linux/splice.h>
+#include <linux/compat.h>
+#include "internal.h"
 
 #include <asm/uaccess.h>
 #include <asm/unistd.h>
 
-struct file_operations generic_ro_fops = {
+typedef ssize_t (*io_fn_t)(struct file *, char __user *, size_t, loff_t *);
+typedef ssize_t (*iov_fn_t)(struct kiocb *, const struct iovec *,
+               unsigned long, loff_t);
+
+const struct file_operations generic_ro_fops = {
        .llseek         = generic_file_llseek,
-       .read           = generic_file_read,
+       .read           = do_sync_read,
+       .aio_read       = generic_file_aio_read,
        .mmap           = generic_file_readonly_mmap,
-       .sendfile       = generic_file_sendfile,
+       .splice_read    = generic_file_splice_read,
 };
 
 EXPORT_SYMBOL(generic_ro_fops);
 
-loff_t generic_file_llseek(struct file *file, loff_t offset, int origin)
+static inline int unsigned_offsets(struct file *file)
 {
-       long long retval;
-       struct inode *inode = file->f_mapping->host;
+       return file->f_mode & FMODE_UNSIGNED_OFFSET;
+}
 
-       down(&inode->i_sem);
-       switch (origin) {
-               case 2:
-                       offset += inode->i_size;
-                       break;
-               case 1:
-                       offset += file->f_pos;
+static loff_t lseek_execute(struct file *file, struct inode *inode,
+               loff_t offset, loff_t maxsize)
+{
+       if (offset < 0 && !unsigned_offsets(file))
+               return -EINVAL;
+       if (offset > maxsize)
+               return -EINVAL;
+
+       if (offset != file->f_pos) {
+               file->f_pos = offset;
+               file->f_version = 0;
        }
-       retval = -EINVAL;
-       if (offset>=0 && offset<=inode->i_sb->s_maxbytes) {
-               if (offset != file->f_pos) {
-                       file->f_pos = offset;
-                       file->f_version = 0;
-               }
-               retval = offset;
+       return offset;
+}
+
+/**
+ * generic_file_llseek_size - generic llseek implementation for regular files
+ * @file:      file structure to seek on
+ * @offset:    file offset to seek to
+ * @whence:    type of seek
+ * @size:      max size of this file in file system
+ * @eof:       offset used for SEEK_END position
+ *
+ * This is a variant of generic_file_llseek that allows passing in a custom
+ * maximum file size and a custom EOF position, for e.g. hashed directories
+ *
+ * Synchronization:
+ * SEEK_SET and SEEK_END are unsynchronized (but atomic on 64bit platforms)
+ * SEEK_CUR is synchronized against other SEEK_CURs, but not read/writes.
+ * read/writes behave like SEEK_SET against seeks.
+ */
+loff_t
+generic_file_llseek_size(struct file *file, loff_t offset, int whence,
+               loff_t maxsize, loff_t eof)
+{
+       struct inode *inode = file->f_mapping->host;
+
+       switch (whence) {
+       case SEEK_END:
+               offset += eof;
+               break;
+       case SEEK_CUR:
+               /*
+                * Here we special-case the lseek(fd, 0, SEEK_CUR)
+                * position-querying operation.  Avoid rewriting the "same"
+                * f_pos value back to the file because a concurrent read(),
+                * write() or lseek() might have altered it
+                */
+               if (offset == 0)
+                       return file->f_pos;
+               /*
+                * f_lock protects against read/modify/write race with other
+                * SEEK_CURs. Note that parallel writes and reads behave
+                * like SEEK_SET.
+                */
+               spin_lock(&file->f_lock);
+               offset = lseek_execute(file, inode, file->f_pos + offset,
+                                      maxsize);
+               spin_unlock(&file->f_lock);
+               return offset;
+       case SEEK_DATA:
+               /*
+                * In the generic case the entire file is data, so as long as
+                * offset isn't at the end of the file then the offset is data.
+                */
+               if (offset >= eof)
+                       return -ENXIO;
+               break;
+       case SEEK_HOLE:
+               /*
+                * There is a virtual hole at the end of the file, so as long as
+                * offset isn't i_size or larger, return i_size.
+                */
+               if (offset >= eof)
+                       return -ENXIO;
+               offset = eof;
+               break;
        }
-       up(&inode->i_sem);
-       return retval;
+
+       return lseek_execute(file, inode, offset, maxsize);
 }
+EXPORT_SYMBOL(generic_file_llseek_size);
+
+/**
+ * generic_file_llseek - generic llseek implementation for regular files
+ * @file:      file structure to seek on
+ * @offset:    file offset to seek to
+ * @whence:    type of seek
+ *
+ * This is a generic implemenation of ->llseek useable for all normal local
+ * filesystems.  It just updates the file offset to the value specified by
+ * @offset and @whence.
+ */
+loff_t generic_file_llseek(struct file *file, loff_t offset, int whence)
+{
+       struct inode *inode = file->f_mapping->host;
 
+       return generic_file_llseek_size(file, offset, whence,
+                                       inode->i_sb->s_maxbytes,
+                                       i_size_read(inode));
+}
 EXPORT_SYMBOL(generic_file_llseek);
 
-loff_t remote_llseek(struct file *file, loff_t offset, int origin)
+/**
+ * noop_llseek - No Operation Performed llseek implementation
+ * @file:      file structure to seek on
+ * @offset:    file offset to seek to
+ * @whence:    type of seek
+ *
+ * This is an implementation of ->llseek useable for the rare special case when
+ * userspace expects the seek to succeed but the (device) file is actually not
+ * able to perform the seek. In this case you use noop_llseek() instead of
+ * falling back to the default implementation of ->llseek.
+ */
+loff_t noop_llseek(struct file *file, loff_t offset, int whence)
 {
-       long long retval;
-
-       lock_kernel();
-       switch (origin) {
-               case 2:
-                       offset += i_size_read(file->f_dentry->d_inode);
-                       break;
-               case 1:
-                       offset += file->f_pos;
-       }
-       retval = -EINVAL;
-       if (offset>=0 && offset<=file->f_dentry->d_inode->i_sb->s_maxbytes) {
-               if (offset != file->f_pos) {
-                       file->f_pos = offset;
-                       file->f_version = 0;
-               }
-               retval = offset;
-       }
-       unlock_kernel();
-       return retval;
+       return file->f_pos;
 }
-EXPORT_SYMBOL(remote_llseek);
+EXPORT_SYMBOL(noop_llseek);
 
-loff_t no_llseek(struct file *file, loff_t offset, int origin)
+loff_t no_llseek(struct file *file, loff_t offset, int whence)
 {
        return -ESPIPE;
 }
 EXPORT_SYMBOL(no_llseek);
 
-loff_t default_llseek(struct file *file, loff_t offset, int origin)
+loff_t default_llseek(struct file *file, loff_t offset, int whence)
 {
-       long long retval;
+       struct inode *inode = file_inode(file);
+       loff_t retval;
 
-       lock_kernel();
-       switch (origin) {
-               case 2:
-                       offset += i_size_read(file->f_dentry->d_inode);
+       mutex_lock(&inode->i_mutex);
+       switch (whence) {
+               case SEEK_END:
+                       offset += i_size_read(inode);
                        break;
-               case 1:
+               case SEEK_CUR:
+                       if (offset == 0) {
+                               retval = file->f_pos;
+                               goto out;
+                       }
                        offset += file->f_pos;
+                       break;
+               case SEEK_DATA:
+                       /*
+                        * In the generic case the entire file is data, so as
+                        * long as offset isn't at the end of the file then the
+                        * offset is data.
+                        */
+                       if (offset >= inode->i_size) {
+                               retval = -ENXIO;
+                               goto out;
+                       }
+                       break;
+               case SEEK_HOLE:
+                       /*
+                        * There is a virtual hole at the end of the file, so
+                        * as long as offset isn't i_size or larger, return
+                        * i_size.
+                        */
+                       if (offset >= inode->i_size) {
+                               retval = -ENXIO;
+                               goto out;
+                       }
+                       offset = inode->i_size;
+                       break;
        }
        retval = -EINVAL;
-       if (offset >= 0) {
+       if (offset >= 0 || unsigned_offsets(file)) {
                if (offset != file->f_pos) {
                        file->f_pos = offset;
                        file->f_version = 0;
                }
                retval = offset;
        }
-       unlock_kernel();
+out:
+       mutex_unlock(&inode->i_mutex);
        return retval;
 }
 EXPORT_SYMBOL(default_llseek);
 
-loff_t vfs_llseek(struct file *file, loff_t offset, int origin)
+loff_t vfs_llseek(struct file *file, loff_t offset, int whence)
 {
        loff_t (*fn)(struct file *, loff_t, int);
 
        fn = no_llseek;
        if (file->f_mode & FMODE_LSEEK) {
-               fn = default_llseek;
                if (file->f_op && file->f_op->llseek)
                        fn = file->f_op->llseek;
        }
-       return fn(file, offset, origin);
+       return fn(file, offset, whence);
 }
 EXPORT_SYMBOL(vfs_llseek);
 
-asmlinkage off_t sys_lseek(unsigned int fd, off_t offset, unsigned int origin)
+SYSCALL_DEFINE3(lseek, unsigned int, fd, off_t, offset, unsigned int, whence)
 {
        off_t retval;
-       struct file * file;
-       int fput_needed;
-
-       retval = -EBADF;
-       file = fget_light(fd, &fput_needed);
-       if (!file)
-               goto bad;
+       struct fd f = fdget(fd);
+       if (!f.file)
+               return -EBADF;
 
        retval = -EINVAL;
-       if (origin <= 2) {
-               loff_t res = vfs_llseek(file, offset, origin);
+       if (whence <= SEEK_MAX) {
+               loff_t res = vfs_llseek(f.file, offset, whence);
                retval = res;
                if (res != (loff_t)retval)
                        retval = -EOVERFLOW;    /* LFS: should only happen on 32 bit platforms */
        }
-       fput_light(file, fput_needed);
-bad:
+       fdput(f);
        return retval;
 }
 
+#ifdef CONFIG_COMPAT
+COMPAT_SYSCALL_DEFINE3(lseek, unsigned int, fd, compat_off_t, offset, unsigned int, whence)
+{
+       return sys_lseek(fd, offset, whence);
+}
+#endif
+
 #ifdef __ARCH_WANT_SYS_LLSEEK
-asmlinkage long sys_llseek(unsigned int fd, unsigned long offset_high,
-                          unsigned long offset_low, loff_t __user * result,
-                          unsigned int origin)
+SYSCALL_DEFINE5(llseek, unsigned int, fd, unsigned long, offset_high,
+               unsigned long, offset_low, loff_t __user *, result,
+               unsigned int, whence)
 {
        int retval;
-       struct file * file;
+       struct fd f = fdget(fd);
        loff_t offset;
-       int fput_needed;
 
-       retval = -EBADF;
-       file = fget_light(fd, &fput_needed);
-       if (!file)
-               goto bad;
+       if (!f.file)
+               return -EBADF;
 
        retval = -EINVAL;
-       if (origin > 2)
+       if (whence > SEEK_MAX)
                goto out_putf;
 
-       offset = vfs_llseek(file, ((loff_t) offset_high << 32) | offset_low,
-                       origin);
+       offset = vfs_llseek(f.file, ((loff_t) offset_high << 32) | offset_low,
+                       whence);
 
        retval = (int)offset;
        if (offset >= 0) {
@@ -176,54 +286,62 @@ asmlinkage long sys_llseek(unsigned int fd, unsigned long offset_high,
                        retval = 0;
        }
 out_putf:
-       fput_light(file, fput_needed);
-bad:
+       fdput(f);
        return retval;
 }
 #endif
 
-
+/*
+ * rw_verify_area doesn't like huge counts. We limit
+ * them to something that fits in "int" so that others
+ * won't have to do range checks all the time.
+ */
 int rw_verify_area(int read_write, struct file *file, loff_t *ppos, size_t count)
 {
        struct inode *inode;
        loff_t pos;
+       int retval = -EINVAL;
 
-       if (unlikely(count > INT_MAX))
-               goto Einval;
+       inode = file_inode(file);
+       if (unlikely((ssize_t) count < 0))
+               return retval;
        pos = *ppos;
-       if (unlikely((pos < 0) || (loff_t) (pos + count) < 0))
-               goto Einval;
-
-       inode = file->f_dentry->d_inode;
-       if (inode->i_flock && MANDATORY_LOCK(inode))
-               return locks_mandatory_area(read_write == READ ? FLOCK_VERIFY_READ : FLOCK_VERIFY_WRITE, inode, file, pos, count);
-       return 0;
-
-Einval:
-       return -EINVAL;
-}
+       if (unlikely(pos < 0)) {
+               if (!unsigned_offsets(file))
+                       return retval;
+               if (count >= -pos) /* both values are in 0..LLONG_MAX */
+                       return -EOVERFLOW;
+       } else if (unlikely((loff_t) (pos + count) < 0)) {
+               if (!unsigned_offsets(file))
+                       return retval;
+       }
 
-static void wait_on_retry_sync_kiocb(struct kiocb *iocb)
-{
-       set_current_state(TASK_UNINTERRUPTIBLE);
-       if (!kiocbIsKicked(iocb))
-               schedule();
-       else
-               kiocbClearKicked(iocb);
-       __set_current_state(TASK_RUNNING);
+       if (unlikely(inode->i_flock && mandatory_lock(inode))) {
+               retval = locks_mandatory_area(
+                       read_write == READ ? FLOCK_VERIFY_READ : FLOCK_VERIFY_WRITE,
+                       inode, file, pos, count);
+               if (retval < 0)
+                       return retval;
+       }
+       retval = security_file_permission(file,
+                               read_write == READ ? MAY_READ : MAY_WRITE);
+       if (retval)
+               return retval;
+       return count > MAX_RW_COUNT ? MAX_RW_COUNT : count;
 }
 
 ssize_t do_sync_read(struct file *filp, char __user *buf, size_t len, loff_t *ppos)
 {
+       struct iovec iov = { .iov_base = buf, .iov_len = len };
        struct kiocb kiocb;
        ssize_t ret;
 
        init_sync_kiocb(&kiocb, filp);
        kiocb.ki_pos = *ppos;
-       while (-EIOCBRETRY ==
-               (ret = filp->f_op->aio_read(&kiocb, buf, len, kiocb.ki_pos)))
-               wait_on_retry_sync_kiocb(&kiocb);
+       kiocb.ki_left = len;
+       kiocb.ki_nbytes = len;
 
+       ret = filp->f_op->aio_read(&kiocb, &iov, 1, kiocb.ki_pos);
        if (-EIOCBQUEUED == ret)
                ret = wait_on_sync_kiocb(&kiocb);
        *ppos = kiocb.ki_pos;
@@ -244,19 +362,17 @@ ssize_t vfs_read(struct file *file, char __user *buf, size_t count, loff_t *pos)
                return -EFAULT;
 
        ret = rw_verify_area(READ, file, pos, count);
-       if (!ret) {
-               ret = security_file_permission (file, MAY_READ);
-               if (!ret) {
-                       if (file->f_op->read)
-                               ret = file->f_op->read(file, buf, count, pos);
-                       else
-                               ret = do_sync_read(file, buf, count, pos);
-                       if (ret > 0) {
-                               fsnotify_access(file->f_dentry);
-                               current->rchar += ret;
-                       }
-                       current->syscr++;
+       if (ret >= 0) {
+               count = ret;
+               if (file->f_op->read)
+                       ret = file->f_op->read(file, buf, count, pos);
+               else
+                       ret = do_sync_read(file, buf, count, pos);
+               if (ret > 0) {
+                       fsnotify_access(file);
+                       add_rchar(current, ret);
                }
+               inc_syscr(current);
        }
 
        return ret;
@@ -266,15 +382,16 @@ EXPORT_SYMBOL(vfs_read);
 
 ssize_t do_sync_write(struct file *filp, const char __user *buf, size_t len, loff_t *ppos)
 {
+       struct iovec iov = { .iov_base = (void __user *)buf, .iov_len = len };
        struct kiocb kiocb;
        ssize_t ret;
 
        init_sync_kiocb(&kiocb, filp);
        kiocb.ki_pos = *ppos;
-       while (-EIOCBRETRY ==
-              (ret = filp->f_op->aio_write(&kiocb, buf, len, kiocb.ki_pos)))
-               wait_on_retry_sync_kiocb(&kiocb);
+       kiocb.ki_left = len;
+       kiocb.ki_nbytes = len;
 
+       ret = filp->f_op->aio_write(&kiocb, &iov, 1, kiocb.ki_pos);
        if (-EIOCBQUEUED == ret)
                ret = wait_on_sync_kiocb(&kiocb);
        *ppos = kiocb.ki_pos;
@@ -283,6 +400,33 @@ ssize_t do_sync_write(struct file *filp, const char __user *buf, size_t len, lof
 
 EXPORT_SYMBOL(do_sync_write);
 
+ssize_t __kernel_write(struct file *file, const char *buf, size_t count, loff_t *pos)
+{
+       mm_segment_t old_fs;
+       const char __user *p;
+       ssize_t ret;
+
+       if (!file->f_op || (!file->f_op->write && !file->f_op->aio_write))
+               return -EINVAL;
+
+       old_fs = get_fs();
+       set_fs(get_ds());
+       p = (__force const char __user *)buf;
+       if (count > MAX_RW_COUNT)
+               count =  MAX_RW_COUNT;
+       if (file->f_op->write)
+               ret = file->f_op->write(file, p, count, pos);
+       else
+               ret = do_sync_write(file, p, count, pos);
+       set_fs(old_fs);
+       if (ret > 0) {
+               fsnotify_modify(file);
+               add_wchar(current, ret);
+       }
+       inc_syscw(current);
+       return ret;
+}
+
 ssize_t vfs_write(struct file *file, const char __user *buf, size_t count, loff_t *pos)
 {
        ssize_t ret;
@@ -295,19 +439,19 @@ ssize_t vfs_write(struct file *file, const char __user *buf, size_t count, loff_
                return -EFAULT;
 
        ret = rw_verify_area(WRITE, file, pos, count);
-       if (!ret) {
-               ret = security_file_permission (file, MAY_WRITE);
-               if (!ret) {
-                       if (file->f_op->write)
-                               ret = file->f_op->write(file, buf, count, pos);
-                       else
-                               ret = do_sync_write(file, buf, count, pos);
-                       if (ret > 0) {
-                               fsnotify_modify(file->f_dentry);
-                               current->wchar += ret;
-                       }
-                       current->syscw++;
+       if (ret >= 0) {
+               count = ret;
+               file_start_write(file);
+               if (file->f_op->write)
+                       ret = file->f_op->write(file, buf, count, pos);
+               else
+                       ret = do_sync_write(file, buf, count, pos);
+               if (ret > 0) {
+                       fsnotify_modify(file);
+                       add_wchar(current, ret);
                }
+               inc_syscw(current);
+               file_end_write(file);
        }
 
        return ret;
@@ -325,78 +469,71 @@ static inline void file_pos_write(struct file *file, loff_t pos)
        file->f_pos = pos;
 }
 
-asmlinkage ssize_t sys_read(unsigned int fd, char __user * buf, size_t count)
+SYSCALL_DEFINE3(read, unsigned int, fd, char __user *, buf, size_t, count)
 {
-       struct file *file;
+       struct fd f = fdget(fd);
        ssize_t ret = -EBADF;
-       int fput_needed;
 
-       file = fget_light(fd, &fput_needed);
-       if (file) {
-               loff_t pos = file_pos_read(file);
-               ret = vfs_read(file, buf, count, &pos);
-               file_pos_write(file, pos);
-               fput_light(file, fput_needed);
+       if (f.file) {
+               loff_t pos = file_pos_read(f.file);
+               ret = vfs_read(f.file, buf, count, &pos);
+               file_pos_write(f.file, pos);
+               fdput(f);
        }
-
        return ret;
 }
-EXPORT_SYMBOL_GPL(sys_read);
 
-asmlinkage ssize_t sys_write(unsigned int fd, const char __user * buf, size_t count)
+SYSCALL_DEFINE3(write, unsigned int, fd, const char __user *, buf,
+               size_t, count)
 {
-       struct file *file;
+       struct fd f = fdget(fd);
        ssize_t ret = -EBADF;
-       int fput_needed;
 
-       file = fget_light(fd, &fput_needed);
-       if (file) {
-               loff_t pos = file_pos_read(file);
-               ret = vfs_write(file, buf, count, &pos);
-               file_pos_write(file, pos);
-               fput_light(file, fput_needed);
+       if (f.file) {
+               loff_t pos = file_pos_read(f.file);
+               ret = vfs_write(f.file, buf, count, &pos);
+               file_pos_write(f.file, pos);
+               fdput(f);
        }
 
        return ret;
 }
 
-asmlinkage ssize_t sys_pread64(unsigned int fd, char __user *buf,
-                            size_t count, loff_t pos)
+SYSCALL_DEFINE4(pread64, unsigned int, fd, char __user *, buf,
+                       size_t, count, loff_t, pos)
 {
-       struct file *file;
+       struct fd f;
        ssize_t ret = -EBADF;
-       int fput_needed;
 
        if (pos < 0)
                return -EINVAL;
 
-       file = fget_light(fd, &fput_needed);
-       if (file) {
+       f = fdget(fd);
+       if (f.file) {
                ret = -ESPIPE;
-               if (file->f_mode & FMODE_PREAD)
-                       ret = vfs_read(file, buf, count, &pos);
-               fput_light(file, fput_needed);
+               if (f.file->f_mode & FMODE_PREAD)
+                       ret = vfs_read(f.file, buf, count, &pos);
+               fdput(f);
        }
 
        return ret;
 }
 
-asmlinkage ssize_t sys_pwrite64(unsigned int fd, const char __user *buf,
-                             size_t count, loff_t pos)
+SYSCALL_DEFINE4(pwrite64, unsigned int, fd, const char __user *, buf,
+                        size_t, count, loff_t, pos)
 {
-       struct file *file;
+       struct fd f;
        ssize_t ret = -EBADF;
-       int fput_needed;
 
        if (pos < 0)
                return -EINVAL;
 
-       file = fget_light(fd, &fput_needed);
-       if (file) {
+       f = fdget(fd);
+       if (f.file) {
                ret = -ESPIPE;
-               if (file->f_mode & FMODE_PWRITE)  
-                       ret = vfs_write(file, buf, count, &pos);
-               fput_light(file, fput_needed);
+               if (f.file->f_mode & FMODE_PWRITE)  
+                       ret = vfs_write(f.file, buf, count, &pos);
+               fdput(f);
        }
 
        return ret;
@@ -421,134 +558,191 @@ unsigned long iov_shorten(struct iovec *iov, unsigned long nr_segs, size_t to)
        }
        return seg;
 }
-
 EXPORT_SYMBOL(iov_shorten);
 
+static ssize_t do_sync_readv_writev(struct file *filp, const struct iovec *iov,
+               unsigned long nr_segs, size_t len, loff_t *ppos, iov_fn_t fn)
+{
+       struct kiocb kiocb;
+       ssize_t ret;
+
+       init_sync_kiocb(&kiocb, filp);
+       kiocb.ki_pos = *ppos;
+       kiocb.ki_left = len;
+       kiocb.ki_nbytes = len;
+
+       ret = fn(&kiocb, iov, nr_segs, kiocb.ki_pos);
+       if (ret == -EIOCBQUEUED)
+               ret = wait_on_sync_kiocb(&kiocb);
+       *ppos = kiocb.ki_pos;
+       return ret;
+}
+
+/* Do it by hand, with file-ops */
+static ssize_t do_loop_readv_writev(struct file *filp, struct iovec *iov,
+               unsigned long nr_segs, loff_t *ppos, io_fn_t fn)
+{
+       struct iovec *vector = iov;
+       ssize_t ret = 0;
+
+       while (nr_segs > 0) {
+               void __user *base;
+               size_t len;
+               ssize_t nr;
+
+               base = vector->iov_base;
+               len = vector->iov_len;
+               vector++;
+               nr_segs--;
+
+               nr = fn(filp, base, len, ppos);
+
+               if (nr < 0) {
+                       if (!ret)
+                               ret = nr;
+                       break;
+               }
+               ret += nr;
+               if (nr != len)
+                       break;
+       }
+
+       return ret;
+}
+
 /* A write operation does a read from user space and vice versa */
 #define vrfy_dir(type) ((type) == READ ? VERIFY_WRITE : VERIFY_READ)
 
-static ssize_t do_readv_writev(int type, struct file *file,
-                              const struct iovec __user * uvector,
-                              unsigned long nr_segs, loff_t *pos)
+ssize_t rw_copy_check_uvector(int type, const struct iovec __user * uvector,
+                             unsigned long nr_segs, unsigned long fast_segs,
+                             struct iovec *fast_pointer,
+                             struct iovec **ret_pointer)
 {
-       typedef ssize_t (*io_fn_t)(struct file *, char __user *, size_t, loff_t *);
-       typedef ssize_t (*iov_fn_t)(struct file *, const struct iovec *, unsigned long, loff_t *);
-
-       size_t tot_len;
-       struct iovec iovstack[UIO_FASTIOV];
-       struct iovec *iov=iovstack, *vector;
+       unsigned long seg;
        ssize_t ret;
-       int seg;
-       io_fn_t fn;
-       iov_fn_t fnv;
+       struct iovec *iov = fast_pointer;
 
        /*
         * SuS says "The readv() function *may* fail if the iovcnt argument
         * was less than or equal to 0, or greater than {IOV_MAX}.  Linux has
         * traditionally returned zero for zero segments, so...
         */
-       ret = 0;
-       if (nr_segs == 0)
+       if (nr_segs == 0) {
+               ret = 0;
                goto out;
+       }
 
        /*
         * First get the "struct iovec" from user memory and
         * verify all the pointers
         */
-       ret = -EINVAL;
-       if ((nr_segs > UIO_MAXIOV) || (nr_segs <= 0))
+       if (nr_segs > UIO_MAXIOV) {
+               ret = -EINVAL;
                goto out;
-       if (!file->f_op)
-               goto out;
-       if (nr_segs > UIO_FASTIOV) {
-               ret = -ENOMEM;
+       }
+       if (nr_segs > fast_segs) {
                iov = kmalloc(nr_segs*sizeof(struct iovec), GFP_KERNEL);
-               if (!iov)
+               if (iov == NULL) {
+                       ret = -ENOMEM;
                        goto out;
+               }
        }
-       ret = -EFAULT;
-       if (copy_from_user(iov, uvector, nr_segs*sizeof(*uvector)))
+       if (copy_from_user(iov, uvector, nr_segs*sizeof(*uvector))) {
+               ret = -EFAULT;
                goto out;
+       }
 
        /*
-        * Single unix specification:
-        * We should -EINVAL if an element length is not >= 0 and fitting an
-        * ssize_t.  The total length is fitting an ssize_t
+        * According to the Single Unix Specification we should return EINVAL
+        * if an element length is < 0 when cast to ssize_t or if the
+        * total length would overflow the ssize_t return value of the
+        * system call.
         *
-        * Be careful here because iov_len is a size_t not an ssize_t
+        * Linux caps all read/write calls to MAX_RW_COUNT, and avoids the
+        * overflow case.
         */
-       tot_len = 0;
-       ret = -EINVAL;
+       ret = 0;
        for (seg = 0; seg < nr_segs; seg++) {
                void __user *buf = iov[seg].iov_base;
                ssize_t len = (ssize_t)iov[seg].iov_len;
 
-               if (len < 0)    /* size_t not fitting an ssize_t .. */
+               /* see if we we're about to use an invalid len or if
+                * it's about to overflow ssize_t */
+               if (len < 0) {
+                       ret = -EINVAL;
                        goto out;
-               if (unlikely(!access_ok(vrfy_dir(type), buf, len)))
-                       goto Efault;
-               tot_len += len;
-               if ((ssize_t)tot_len < 0) /* maths overflow on the ssize_t */
+               }
+               if (type >= 0
+                   && unlikely(!access_ok(vrfy_dir(type), buf, len))) {
+                       ret = -EFAULT;
                        goto out;
+               }
+               if (len > MAX_RW_COUNT - ret) {
+                       len = MAX_RW_COUNT - ret;
+                       iov[seg].iov_len = len;
+               }
+               ret += len;
        }
-       if (tot_len == 0) {
-               ret = 0;
+out:
+       *ret_pointer = iov;
+       return ret;
+}
+
+static ssize_t do_readv_writev(int type, struct file *file,
+                              const struct iovec __user * uvector,
+                              unsigned long nr_segs, loff_t *pos)
+{
+       size_t tot_len;
+       struct iovec iovstack[UIO_FASTIOV];
+       struct iovec *iov = iovstack;
+       ssize_t ret;
+       io_fn_t fn;
+       iov_fn_t fnv;
+
+       if (!file->f_op) {
+               ret = -EINVAL;
                goto out;
        }
 
+       ret = rw_copy_check_uvector(type, uvector, nr_segs,
+                                   ARRAY_SIZE(iovstack), iovstack, &iov);
+       if (ret <= 0)
+               goto out;
+
+       tot_len = ret;
        ret = rw_verify_area(type, file, pos, tot_len);
-       if (ret)
+       if (ret < 0)
                goto out;
 
        fnv = NULL;
        if (type == READ) {
                fn = file->f_op->read;
-               fnv = file->f_op->readv;
+               fnv = file->f_op->aio_read;
        } else {
                fn = (io_fn_t)file->f_op->write;
-               fnv = file->f_op->writev;
-       }
-       if (fnv) {
-               ret = fnv(file, iov, nr_segs, pos);
-               goto out;
+               fnv = file->f_op->aio_write;
+               file_start_write(file);
        }
 
-       /* Do it by hand, with file-ops */
-       ret = 0;
-       vector = iov;
-       while (nr_segs > 0) {
-               void __user * base;
-               size_t len;
-               ssize_t nr;
-
-               base = vector->iov_base;
-               len = vector->iov_len;
-               vector++;
-               nr_segs--;
+       if (fnv)
+               ret = do_sync_readv_writev(file, iov, nr_segs, tot_len,
+                                               pos, fnv);
+       else
+               ret = do_loop_readv_writev(file, iov, nr_segs, pos, fn);
 
-               nr = fn(file, base, len, pos);
+       if (type != READ)
+               file_end_write(file);
 
-               if (nr < 0) {
-                       if (!ret) ret = nr;
-                       break;
-               }
-               ret += nr;
-               if (nr != len)
-                       break;
-       }
 out:
        if (iov != iovstack)
                kfree(iov);
        if ((ret + (type == READ)) > 0) {
                if (type == READ)
-                       fsnotify_access(file->f_dentry);
+                       fsnotify_access(file);
                else
-                       fsnotify_modify(file->f_dentry);
+                       fsnotify_modify(file);
        }
        return ret;
-Efault:
-       ret = -EFAULT;
-       goto out;
 }
 
 ssize_t vfs_readv(struct file *file, const struct iovec __user *vec,
@@ -556,7 +750,7 @@ ssize_t vfs_readv(struct file *file, const struct iovec __user *vec,
 {
        if (!(file->f_mode & FMODE_READ))
                return -EBADF;
-       if (!file->f_op || (!file->f_op->readv && !file->f_op->read))
+       if (!file->f_op || (!file->f_op->aio_read && !file->f_op->read))
                return -EINVAL;
 
        return do_readv_writev(READ, file, vec, vlen, pos);
@@ -569,7 +763,7 @@ ssize_t vfs_writev(struct file *file, const struct iovec __user *vec,
 {
        if (!(file->f_mode & FMODE_WRITE))
                return -EBADF;
-       if (!file->f_op || (!file->f_op->writev && !file->f_op->write))
+       if (!file->f_op || (!file->f_op->aio_write && !file->f_op->write))
                return -EINVAL;
 
        return do_readv_writev(WRITE, file, vec, vlen, pos);
@@ -577,114 +771,346 @@ ssize_t vfs_writev(struct file *file, const struct iovec __user *vec,
 
 EXPORT_SYMBOL(vfs_writev);
 
-asmlinkage ssize_t
-sys_readv(unsigned long fd, const struct iovec __user *vec, unsigned long vlen)
+SYSCALL_DEFINE3(readv, unsigned long, fd, const struct iovec __user *, vec,
+               unsigned long, vlen)
 {
-       struct file *file;
+       struct fd f = fdget(fd);
        ssize_t ret = -EBADF;
-       int fput_needed;
 
-       file = fget_light(fd, &fput_needed);
-       if (file) {
-               loff_t pos = file_pos_read(file);
-               ret = vfs_readv(file, vec, vlen, &pos);
-               file_pos_write(file, pos);
-               fput_light(file, fput_needed);
+       if (f.file) {
+               loff_t pos = file_pos_read(f.file);
+               ret = vfs_readv(f.file, vec, vlen, &pos);
+               file_pos_write(f.file, pos);
+               fdput(f);
        }
 
        if (ret > 0)
-               current->rchar += ret;
-       current->syscr++;
+               add_rchar(current, ret);
+       inc_syscr(current);
        return ret;
 }
 
-asmlinkage ssize_t
-sys_writev(unsigned long fd, const struct iovec __user *vec, unsigned long vlen)
+SYSCALL_DEFINE3(writev, unsigned long, fd, const struct iovec __user *, vec,
+               unsigned long, vlen)
 {
-       struct file *file;
+       struct fd f = fdget(fd);
        ssize_t ret = -EBADF;
-       int fput_needed;
 
-       file = fget_light(fd, &fput_needed);
-       if (file) {
-               loff_t pos = file_pos_read(file);
-               ret = vfs_writev(file, vec, vlen, &pos);
-               file_pos_write(file, pos);
-               fput_light(file, fput_needed);
+       if (f.file) {
+               loff_t pos = file_pos_read(f.file);
+               ret = vfs_writev(f.file, vec, vlen, &pos);
+               file_pos_write(f.file, pos);
+               fdput(f);
        }
 
        if (ret > 0)
-               current->wchar += ret;
-       current->syscw++;
+               add_wchar(current, ret);
+       inc_syscw(current);
        return ret;
 }
 
+static inline loff_t pos_from_hilo(unsigned long high, unsigned long low)
+{
+#define HALF_LONG_BITS (BITS_PER_LONG / 2)
+       return (((loff_t)high << HALF_LONG_BITS) << HALF_LONG_BITS) | low;
+}
+
+SYSCALL_DEFINE5(preadv, unsigned long, fd, const struct iovec __user *, vec,
+               unsigned long, vlen, unsigned long, pos_l, unsigned long, pos_h)
+{
+       loff_t pos = pos_from_hilo(pos_h, pos_l);
+       struct fd f;
+       ssize_t ret = -EBADF;
+
+       if (pos < 0)
+               return -EINVAL;
+
+       f = fdget(fd);
+       if (f.file) {
+               ret = -ESPIPE;
+               if (f.file->f_mode & FMODE_PREAD)
+                       ret = vfs_readv(f.file, vec, vlen, &pos);
+               fdput(f);
+       }
+
+       if (ret > 0)
+               add_rchar(current, ret);
+       inc_syscr(current);
+       return ret;
+}
+
+SYSCALL_DEFINE5(pwritev, unsigned long, fd, const struct iovec __user *, vec,
+               unsigned long, vlen, unsigned long, pos_l, unsigned long, pos_h)
+{
+       loff_t pos = pos_from_hilo(pos_h, pos_l);
+       struct fd f;
+       ssize_t ret = -EBADF;
+
+       if (pos < 0)
+               return -EINVAL;
+
+       f = fdget(fd);
+       if (f.file) {
+               ret = -ESPIPE;
+               if (f.file->f_mode & FMODE_PWRITE)
+                       ret = vfs_writev(f.file, vec, vlen, &pos);
+               fdput(f);
+       }
+
+       if (ret > 0)
+               add_wchar(current, ret);
+       inc_syscw(current);
+       return ret;
+}
+
+#ifdef CONFIG_COMPAT
+
+static ssize_t compat_do_readv_writev(int type, struct file *file,
+                              const struct compat_iovec __user *uvector,
+                              unsigned long nr_segs, loff_t *pos)
+{
+       compat_ssize_t tot_len;
+       struct iovec iovstack[UIO_FASTIOV];
+       struct iovec *iov = iovstack;
+       ssize_t ret;
+       io_fn_t fn;
+       iov_fn_t fnv;
+
+       ret = -EINVAL;
+       if (!file->f_op)
+               goto out;
+
+       ret = -EFAULT;
+       if (!access_ok(VERIFY_READ, uvector, nr_segs*sizeof(*uvector)))
+               goto out;
+
+       ret = compat_rw_copy_check_uvector(type, uvector, nr_segs,
+                                              UIO_FASTIOV, iovstack, &iov);
+       if (ret <= 0)
+               goto out;
+
+       tot_len = ret;
+       ret = rw_verify_area(type, file, pos, tot_len);
+       if (ret < 0)
+               goto out;
+
+       fnv = NULL;
+       if (type == READ) {
+               fn = file->f_op->read;
+               fnv = file->f_op->aio_read;
+       } else {
+               fn = (io_fn_t)file->f_op->write;
+               fnv = file->f_op->aio_write;
+               file_start_write(file);
+       }
+
+       if (fnv)
+               ret = do_sync_readv_writev(file, iov, nr_segs, tot_len,
+                                               pos, fnv);
+       else
+               ret = do_loop_readv_writev(file, iov, nr_segs, pos, fn);
+
+       if (type != READ)
+               file_end_write(file);
+
+out:
+       if (iov != iovstack)
+               kfree(iov);
+       if ((ret + (type == READ)) > 0) {
+               if (type == READ)
+                       fsnotify_access(file);
+               else
+                       fsnotify_modify(file);
+       }
+       return ret;
+}
+
+static size_t compat_readv(struct file *file,
+                          const struct compat_iovec __user *vec,
+                          unsigned long vlen, loff_t *pos)
+{
+       ssize_t ret = -EBADF;
+
+       if (!(file->f_mode & FMODE_READ))
+               goto out;
+
+       ret = -EINVAL;
+       if (!file->f_op || (!file->f_op->aio_read && !file->f_op->read))
+               goto out;
+
+       ret = compat_do_readv_writev(READ, file, vec, vlen, pos);
+
+out:
+       if (ret > 0)
+               add_rchar(current, ret);
+       inc_syscr(current);
+       return ret;
+}
+
+COMPAT_SYSCALL_DEFINE3(readv, unsigned long, fd,
+               const struct compat_iovec __user *,vec,
+               unsigned long, vlen)
+{
+       struct fd f = fdget(fd);
+       ssize_t ret;
+       loff_t pos;
+
+       if (!f.file)
+               return -EBADF;
+       pos = f.file->f_pos;
+       ret = compat_readv(f.file, vec, vlen, &pos);
+       f.file->f_pos = pos;
+       fdput(f);
+       return ret;
+}
+
+COMPAT_SYSCALL_DEFINE4(preadv64, unsigned long, fd,
+               const struct compat_iovec __user *,vec,
+               unsigned long, vlen, loff_t, pos)
+{
+       struct fd f;
+       ssize_t ret;
+
+       if (pos < 0)
+               return -EINVAL;
+       f = fdget(fd);
+       if (!f.file)
+               return -EBADF;
+       ret = -ESPIPE;
+       if (f.file->f_mode & FMODE_PREAD)
+               ret = compat_readv(f.file, vec, vlen, &pos);
+       fdput(f);
+       return ret;
+}
+
+COMPAT_SYSCALL_DEFINE5(preadv, unsigned long, fd,
+               const struct compat_iovec __user *,vec,
+               unsigned long, vlen, u32, pos_low, u32, pos_high)
+{
+       loff_t pos = ((loff_t)pos_high << 32) | pos_low;
+       return compat_sys_preadv64(fd, vec, vlen, pos);
+}
+
+static size_t compat_writev(struct file *file,
+                           const struct compat_iovec __user *vec,
+                           unsigned long vlen, loff_t *pos)
+{
+       ssize_t ret = -EBADF;
+
+       if (!(file->f_mode & FMODE_WRITE))
+               goto out;
+
+       ret = -EINVAL;
+       if (!file->f_op || (!file->f_op->aio_write && !file->f_op->write))
+               goto out;
+
+       ret = compat_do_readv_writev(WRITE, file, vec, vlen, pos);
+
+out:
+       if (ret > 0)
+               add_wchar(current, ret);
+       inc_syscw(current);
+       return ret;
+}
+
+COMPAT_SYSCALL_DEFINE3(writev, unsigned long, fd,
+               const struct compat_iovec __user *, vec,
+               unsigned long, vlen)
+{
+       struct fd f = fdget(fd);
+       ssize_t ret;
+       loff_t pos;
+
+       if (!f.file)
+               return -EBADF;
+       pos = f.file->f_pos;
+       ret = compat_writev(f.file, vec, vlen, &pos);
+       f.file->f_pos = pos;
+       fdput(f);
+       return ret;
+}
+
+COMPAT_SYSCALL_DEFINE4(pwritev64, unsigned long, fd,
+               const struct compat_iovec __user *,vec,
+               unsigned long, vlen, loff_t, pos)
+{
+       struct fd f;
+       ssize_t ret;
+
+       if (pos < 0)
+               return -EINVAL;
+       f = fdget(fd);
+       if (!f.file)
+               return -EBADF;
+       ret = -ESPIPE;
+       if (f.file->f_mode & FMODE_PWRITE)
+               ret = compat_writev(f.file, vec, vlen, &pos);
+       fdput(f);
+       return ret;
+}
+
+COMPAT_SYSCALL_DEFINE5(pwritev, unsigned long, fd,
+               const struct compat_iovec __user *,vec,
+               unsigned long, vlen, u32, pos_low, u32, pos_high)
+{
+       loff_t pos = ((loff_t)pos_high << 32) | pos_low;
+       return compat_sys_pwritev64(fd, vec, vlen, pos);
+}
+#endif
+
 static ssize_t do_sendfile(int out_fd, int in_fd, loff_t *ppos,
-                          size_t count, loff_t max)
+                          size_t count, loff_t max)
 {
-       struct file * in_file, * out_file;
-       struct inode * in_inode, * out_inode;
+       struct fd in, out;
+       struct inode *in_inode, *out_inode;
        loff_t pos;
+       loff_t out_pos;
        ssize_t retval;
-       int fput_needed_in, fput_needed_out;
+       int fl;
 
        /*
         * Get input file, and verify that it is ok..
         */
        retval = -EBADF;
-       in_file = fget_light(in_fd, &fput_needed_in);
-       if (!in_file)
+       in = fdget(in_fd);
+       if (!in.file)
                goto out;
-       if (!(in_file->f_mode & FMODE_READ))
-               goto fput_in;
-       retval = -EINVAL;
-       in_inode = in_file->f_dentry->d_inode;
-       if (!in_inode)
-               goto fput_in;
-       if (!in_file->f_op || !in_file->f_op->sendfile)
+       if (!(in.file->f_mode & FMODE_READ))
                goto fput_in;
        retval = -ESPIPE;
-       if (!ppos)
-               ppos = &in_file->f_pos;
-       else
-               if (!(in_file->f_mode & FMODE_PREAD))
+       if (!ppos) {
+               pos = in.file->f_pos;
+       } else {
+               pos = *ppos;
+               if (!(in.file->f_mode & FMODE_PREAD))
                        goto fput_in;
-       retval = rw_verify_area(READ, in_file, ppos, count);
-       if (retval)
-               goto fput_in;
-
-       retval = security_file_permission (in_file, MAY_READ);
-       if (retval)
+       }
+       retval = rw_verify_area(READ, in.file, &pos, count);
+       if (retval < 0)
                goto fput_in;
+       count = retval;
 
        /*
         * Get output file, and verify that it is ok..
         */
        retval = -EBADF;
-       out_file = fget_light(out_fd, &fput_needed_out);
-       if (!out_file)
+       out = fdget(out_fd);
+       if (!out.file)
                goto fput_in;
-       if (!(out_file->f_mode & FMODE_WRITE))
+       if (!(out.file->f_mode & FMODE_WRITE))
                goto fput_out;
        retval = -EINVAL;
-       if (!out_file->f_op || !out_file->f_op->sendpage)
-               goto fput_out;
-       out_inode = out_file->f_dentry->d_inode;
-       retval = rw_verify_area(WRITE, out_file, &out_file->f_pos, count);
-       if (retval)
-               goto fput_out;
-
-       retval = security_file_permission (out_file, MAY_WRITE);
-       if (retval)
+       in_inode = file_inode(in.file);
+       out_inode = file_inode(out.file);
+       out_pos = out.file->f_pos;
+       retval = rw_verify_area(WRITE, out.file, &out_pos, count);
+       if (retval < 0)
                goto fput_out;
+       count = retval;
 
        if (!max)
                max = min(in_inode->i_sb->s_maxbytes, out_inode->i_sb->s_maxbytes);
 
-       pos = *ppos;
-       retval = -EINVAL;
-       if (unlikely(pos < 0))
-               goto fput_out;
        if (unlikely(pos + count > max)) {
                retval = -EOVERFLOW;
                if (pos >= max)
@@ -692,27 +1118,45 @@ static ssize_t do_sendfile(int out_fd, int in_fd, loff_t *ppos,
                count = max - pos;
        }
 
-       retval = in_file->f_op->sendfile(in_file, ppos, count, file_send_actor, out_file);
+       fl = 0;
+#if 0
+       /*
+        * We need to debate whether we can enable this or not. The
+        * man page documents EAGAIN return for the output at least,
+        * and the application is arguably buggy if it doesn't expect
+        * EAGAIN on a non-blocking file descriptor.
+        */
+       if (in.file->f_flags & O_NONBLOCK)
+               fl = SPLICE_F_NONBLOCK;
+#endif
+       retval = do_splice_direct(in.file, &pos, out.file, &out_pos, count, fl);
 
        if (retval > 0) {
-               current->rchar += retval;
-               current->wchar += retval;
+               add_rchar(current, retval);
+               add_wchar(current, retval);
+               fsnotify_access(in.file);
+               fsnotify_modify(out.file);
+               out.file->f_pos = out_pos;
+               if (ppos)
+                       *ppos = pos;
+               else
+                       in.file->f_pos = pos;
        }
-       current->syscr++;
-       current->syscw++;
 
-       if (*ppos > max)
+       inc_syscr(current);
+       inc_syscw(current);
+       if (pos > max)
                retval = -EOVERFLOW;
 
 fput_out:
-       fput_light(out_file, fput_needed_out);
+       fdput(out);
 fput_in:
-       fput_light(in_file, fput_needed_in);
+       fdput(in);
 out:
        return retval;
 }
 
-asmlinkage ssize_t sys_sendfile(int out_fd, int in_fd, off_t __user *offset, size_t count)
+SYSCALL_DEFINE4(sendfile, int, out_fd, int, in_fd, off_t __user *, offset, size_t, count)
 {
        loff_t pos;
        off_t off;
@@ -731,7 +1175,7 @@ asmlinkage ssize_t sys_sendfile(int out_fd, int in_fd, off_t __user *offset, siz
        return do_sendfile(out_fd, in_fd, NULL, count, 0);
 }
 
-asmlinkage ssize_t sys_sendfile64(int out_fd, int in_fd, loff_t __user *offset, size_t count)
+SYSCALL_DEFINE4(sendfile64, int, out_fd, int, in_fd, loff_t __user *, offset, size_t, count)
 {
        loff_t pos;
        ssize_t ret;
@@ -747,3 +1191,43 @@ asmlinkage ssize_t sys_sendfile64(int out_fd, int in_fd, loff_t __user *offset,
 
        return do_sendfile(out_fd, in_fd, NULL, count, 0);
 }
+
+#ifdef CONFIG_COMPAT
+COMPAT_SYSCALL_DEFINE4(sendfile, int, out_fd, int, in_fd,
+               compat_off_t __user *, offset, compat_size_t, count)
+{
+       loff_t pos;
+       off_t off;
+       ssize_t ret;
+
+       if (offset) {
+               if (unlikely(get_user(off, offset)))
+                       return -EFAULT;
+               pos = off;
+               ret = do_sendfile(out_fd, in_fd, &pos, count, MAX_NON_LFS);
+               if (unlikely(put_user(pos, offset)))
+                       return -EFAULT;
+               return ret;
+       }
+
+       return do_sendfile(out_fd, in_fd, NULL, count, 0);
+}
+
+COMPAT_SYSCALL_DEFINE4(sendfile64, int, out_fd, int, in_fd,
+               compat_loff_t __user *, offset, compat_size_t, count)
+{
+       loff_t pos;
+       ssize_t ret;
+
+       if (offset) {
+               if (unlikely(copy_from_user(&pos, offset, sizeof(loff_t))))
+                       return -EFAULT;
+               ret = do_sendfile(out_fd, in_fd, &pos, count, 0);
+               if (unlikely(put_user(pos, offset)))
+                       return -EFAULT;
+               return ret;
+       }
+
+       return do_sendfile(out_fd, in_fd, NULL, count, 0);
+}
+#endif