Merge branch '9p-iov_iter' into for-next

fs/9p/v9fs_vfs.h

@@ -68,14 +68,10 @@ int v9fs_file_open(struct inode *inode, struct file *file);
 void v9fs_inode2stat(struct inode *inode, struct p9_wstat *stat);
 int v9fs_uflags2omode(int uflags, int extended);
 
-ssize_t v9fs_file_readn(struct file *, char *, char __user *, u32, u64);
-ssize_t v9fs_fid_readn(struct p9_fid *, char *, char __user *, u32, u64);
 void v9fs_blank_wstat(struct p9_wstat *wstat);
 int v9fs_vfs_setattr_dotl(struct dentry *, struct iattr *);
 int v9fs_file_fsync_dotl(struct file *filp, loff_t start, loff_t end,
 			 int datasync);
-ssize_t v9fs_file_write_internal(struct inode *, struct p9_fid *,
-				 const char __user *, size_t, loff_t *, int);
 int v9fs_refresh_inode(struct p9_fid *fid, struct inode *inode);
 int v9fs_refresh_inode_dotl(struct p9_fid *fid, struct inode *inode);
 static inline void v9fs_invalidate_inode_attr(struct inode *inode)

fs/9p/vfs_addr.c

@@ -51,12 +51,11 @@
  */
 static int v9fs_fid_readpage(struct p9_fid *fid, struct page *page)
 {
-	int retval;
-	loff_t offset;
-	char *buffer;
-	struct inode *inode;
+	struct inode *inode = page->mapping->host;
+	struct bio_vec bvec = {.bv_page = page, .bv_len = PAGE_SIZE};
+	struct iov_iter to;
+	int retval, err;
 
-	inode = page->mapping->host;
 	p9_debug(P9_DEBUG_VFS, "\n");
 
 	BUG_ON(!PageLocked(page));
@@ -65,16 +64,16 @@ static int v9fs_fid_readpage(struct p9_fid *fid, struct page *page)
 	if (retval == 0)
 		return retval;
 
-	buffer = kmap(page);
-	offset = page_offset(page);
+	iov_iter_bvec(&to, ITER_BVEC | READ, &bvec, 1, PAGE_SIZE);
 
-	retval = v9fs_fid_readn(fid, buffer, NULL, PAGE_CACHE_SIZE, offset);
-	if (retval < 0) {
+	retval = p9_client_read(fid, page_offset(page), &to, &err);
+	if (err) {
 		v9fs_uncache_page(inode, page);
+		retval = err;
 		goto done;
 	}
 
-	memset(buffer + retval, 0, PAGE_CACHE_SIZE - retval);
+	zero_user(page, retval, PAGE_SIZE - retval);
 	flush_dcache_page(page);
 	SetPageUptodate(page);
 
@@ -82,7 +81,6 @@ static int v9fs_fid_readpage(struct p9_fid *fid, struct page *page)
 	retval = 0;
 
 done:
-	kunmap(page);
 	unlock_page(page);
 	return retval;
 }
@@ -161,41 +159,32 @@ static void v9fs_invalidate_page(struct page *page, unsigned int offset,
 
 static int v9fs_vfs_writepage_locked(struct page *page)
 {
-	char *buffer;
-	int retval, len;
-	loff_t offset, size;
-	mm_segment_t old_fs;
-	struct v9fs_inode *v9inode;
 	struct inode *inode = page->mapping->host;
+	struct v9fs_inode *v9inode = V9FS_I(inode);
+	loff_t size = i_size_read(inode);
+	struct iov_iter from;
+	struct bio_vec bvec;
+	int err, len;
 
-	v9inode = V9FS_I(inode);
-	size = i_size_read(inode);
 	if (page->index == size >> PAGE_CACHE_SHIFT)
 		len = size & ~PAGE_CACHE_MASK;
 	else
 		len = PAGE_CACHE_SIZE;
 
-	set_page_writeback(page);
-
-	buffer = kmap(page);
-	offset = page_offset(page);
+	bvec.bv_page = page;
+	bvec.bv_offset = 0;
+	bvec.bv_len = len;
+	iov_iter_bvec(&from, ITER_BVEC | WRITE, &bvec, 1, len);
 
-	old_fs = get_fs();
-	set_fs(get_ds());
 	/* We should have writeback_fid always set */
 	BUG_ON(!v9inode->writeback_fid);
 
-	retval = v9fs_file_write_internal(inode,
-					  v9inode->writeback_fid,
-					  (__force const char __user *)buffer,
-					  len, &offset, 0);
-	if (retval > 0)
-		retval = 0;
+	set_page_writeback(page);
+
+	p9_client_write(v9inode->writeback_fid, page_offset(page), &from, &err);
 
-	set_fs(old_fs);
-	kunmap(page);
 	end_page_writeback(page);
-	return retval;
+	return err;
 }
 
 static int v9fs_vfs_writepage(struct page *page, struct writeback_control *wbc)
@@ -261,16 +250,21 @@ static int v9fs_launder_page(struct page *page)
 static ssize_t
 v9fs_direct_IO(int rw, struct kiocb *iocb, struct iov_iter *iter, loff_t pos)
 {
-	/*
-	 * FIXME
-	 * Now that we do caching with cache mode enabled, We need
-	 * to support direct IO
-	 */
-	p9_debug(P9_DEBUG_VFS, "v9fs_direct_IO: v9fs_direct_IO (%pD) off/no(%lld/%lu) EINVAL\n",
-		 iocb->ki_filp,
-		 (long long)pos, iter->nr_segs);
-
-	return -EINVAL;
+	struct file *file = iocb->ki_filp;
+	ssize_t n;
+	int err = 0;
+	if (rw & WRITE) {
+		n = p9_client_write(file->private_data, pos, iter, &err);
+		if (n) {
+			struct inode *inode = file_inode(file);
+			loff_t i_size = i_size_read(inode);
+			if (pos + n > i_size)
+				inode_add_bytes(inode, pos + n - i_size);
+		}
+	} else {
+		n = p9_client_read(file->private_data, pos, iter, &err);
+	}
+	return n ? n : err;
 }
 
 static int v9fs_write_begin(struct file *filp, struct address_space *mapping,

fs/9p/vfs_dir.c

@@ -33,6 +33,7 @@
 #include <linux/inet.h>
 #include <linux/idr.h>
 #include <linux/slab.h>
+#include <linux/uio.h>
 #include <net/9p/9p.h>
 #include <net/9p/client.h>
 
@@ -115,6 +116,7 @@ static int v9fs_dir_readdir(struct file *file, struct dir_context *ctx)
 	int buflen;
 	int reclen = 0;
 	struct p9_rdir *rdir;
+	struct kvec kvec;
 
 	p9_debug(P9_DEBUG_VFS, "name %pD\n", file);
 	fid = file->private_data;
@@ -124,16 +126,21 @@ static int v9fs_dir_readdir(struct file *file, struct dir_context *ctx)
 	rdir = v9fs_alloc_rdir_buf(file, buflen);
 	if (!rdir)
 		return -ENOMEM;
+	kvec.iov_base = rdir->buf;
+	kvec.iov_len = buflen;
 
 	while (1) {
 		if (rdir->tail == rdir->head) {
-			err = v9fs_file_readn(file, rdir->buf, NULL,
-							buflen, ctx->pos);
-			if (err <= 0)
+			struct iov_iter to;
+			int n;
+			iov_iter_kvec(&to, READ | ITER_KVEC, &kvec, 1, buflen);
+			n = p9_client_read(file->private_data, ctx->pos, &to,
+					   &err);
+			if (err)
 				return err;
 
 			rdir->head = 0;
-			rdir->tail = err;
+			rdir->tail = n;
 		}
 		while (rdir->head < rdir->tail) {
 			p9stat_init(&st);

fs/9p/vfs_file.c

@@ -36,6 +36,8 @@
 #include <linux/utsname.h>
 #include <asm/uaccess.h>
 #include <linux/idr.h>
+#include <linux/uio.h>
+#include <linux/slab.h>
 #include <net/9p/9p.h>
 #include <net/9p/client.h>
 
@@ -285,6 +287,7 @@ static int v9fs_file_getlock(struct file *filp, struct file_lock *fl)
 			fl->fl_end = glock.start + glock.length - 1;
 		fl->fl_pid = glock.proc_id;
 	}
+	kfree(glock.client_id);
 	return res;
 }
 
@@ -363,63 +366,6 @@ out_err:
 	return ret;
 }
 
-/**
- * v9fs_fid_readn - read from a fid
- * @fid: fid to read
- * @data: data buffer to read data into
- * @udata: user data buffer to read data into
- * @count: size of buffer
- * @offset: offset at which to read data
- *
- */
-ssize_t
-v9fs_fid_readn(struct p9_fid *fid, char *data, char __user *udata, u32 count,
-	       u64 offset)
-{
-	int n, total, size;
-
-	p9_debug(P9_DEBUG_VFS, "fid %d offset %llu count %d\n",
-		 fid->fid, (long long unsigned)offset, count);
-	n = 0;
-	total = 0;
-	size = fid->iounit ? fid->iounit : fid->clnt->msize - P9_IOHDRSZ;
-	do {
-		n = p9_client_read(fid, data, udata, offset, count);
-		if (n <= 0)
-			break;
-
-		if (data)
-			data += n;
-		if (udata)
-			udata += n;
-
-		offset += n;
-		count -= n;
-		total += n;
-	} while (count > 0 && n == size);
-
-	if (n < 0)
-		total = n;
-
-	return total;
-}
-
-/**
- * v9fs_file_readn - read from a file
- * @filp: file pointer to read
- * @data: data buffer to read data into
- * @udata: user data buffer to read data into
- * @count: size of buffer
- * @offset: offset at which to read data
- *
- */
-ssize_t
-v9fs_file_readn(struct file *filp, char *data, char __user *udata, u32 count,
-	       u64 offset)
-{
-	return v9fs_fid_readn(filp->private_data, data, udata, count, offset);
-}
-
 /**
  * v9fs_file_read - read from a file
  * @filp: file pointer to read
@@ -430,69 +376,22 @@ v9fs_file_readn(struct file *filp, char *data, char __user *udata, u32 count,
  */
 
 static ssize_t
-v9fs_file_read(struct file *filp, char __user *udata, size_t count,
-	       loff_t * offset)
+v9fs_file_read_iter(struct kiocb *iocb, struct iov_iter *to)
 {
-	int ret;
-	struct p9_fid *fid;
-	size_t size;
+	struct p9_fid *fid = iocb->ki_filp->private_data;
+	int ret, err;
 
-	p9_debug(P9_DEBUG_VFS, "count %zu offset %lld\n", count, *offset);
-	fid = filp->private_data;
+	p9_debug(P9_DEBUG_VFS, "count %zu offset %lld\n",
+		 iov_iter_count(to), iocb->ki_pos);
 
-	size = fid->iounit ? fid->iounit : fid->clnt->msize - P9_IOHDRSZ;
-	if (count > size)
-		ret = v9fs_file_readn(filp, NULL, udata, count, *offset);
-	else
-		ret = p9_client_read(fid, NULL, udata, *offset, count);
-
-	if (ret > 0)
-		*offset += ret;
+	ret = p9_client_read(fid, iocb->ki_pos, to, &err);
+	if (!ret)
+		return err;
 
+	iocb->ki_pos += ret;
 	return ret;
 }
 
-ssize_t
-v9fs_file_write_internal(struct inode *inode, struct p9_fid *fid,
-			 const char __user *data, size_t count,
-			 loff_t *offset, int invalidate)
-{
-	int n;
-	loff_t i_size;
-	size_t total = 0;
-	loff_t origin = *offset;
-	unsigned long pg_start, pg_end;
-
-	p9_debug(P9_DEBUG_VFS, "data %p count %d offset %x\n",
-		 data, (int)count, (int)*offset);
-
-	do {
-		n = p9_client_write(fid, NULL, data+total, origin+total, count);
-		if (n <= 0)
-			break;
-		count -= n;
-		total += n;
-	} while (count > 0);
-
-	if (invalidate && (total > 0)) {
-		pg_start = origin >> PAGE_CACHE_SHIFT;
-		pg_end = (origin + total - 1) >> PAGE_CACHE_SHIFT;
-		if (inode->i_mapping && inode->i_mapping->nrpages)
-			invalidate_inode_pages2_range(inode->i_mapping,
-						      pg_start, pg_end);
-		*offset += total;
-		i_size = i_size_read(inode);
-		if (*offset > i_size) {
-			inode_add_bytes(inode, *offset - i_size);
-			i_size_write(inode, *offset);
-		}
-	}
-	if (n < 0)
-		return n;
-
-	return total;
-}
-
 /**
  * v9fs_file_write - write to a file
  * @filp: file pointer to write
@@ -502,35 +401,45 @@ v9fs_file_write_internal(struct inode *inode, struct p9_fid *fid,
  *
  */
 static ssize_t
-v9fs_file_write(struct file *filp, const char __user * data,
-		size_t count, loff_t *offset)
+v9fs_file_write_iter(struct kiocb *iocb, struct iov_iter *from)
 {
+	struct file *file = iocb->ki_filp;
 	ssize_t retval = 0;
-	loff_t origin = *offset;
+	loff_t origin = iocb->ki_pos;
+	size_t count = iov_iter_count(from);
+	int err = 0;
 
-
-	retval = generic_write_checks(filp, &origin, &count, 0);
+	retval = generic_write_checks(file, &origin, &count, 0);
 	if (retval)
-		goto out;
+		return retval;
+
+	iov_iter_truncate(from, count);
 
-	retval = -EINVAL;
-	if ((ssize_t) count < 0)
-		goto out;
-	retval = 0;
 	if (!count)
-		goto out;
+		return 0;
 
-	retval = v9fs_file_write_internal(file_inode(filp),
-					filp->private_data,
-					data, count, &origin, 1);
-	/* update offset on successful write */
-	if (retval > 0)
-		*offset = origin;
-out:
-	return retval;
+	retval = p9_client_write(file->private_data, origin, from, &err);
+	if (retval > 0) {
+		struct inode *inode = file_inode(file);
+		loff_t i_size;
+		unsigned long pg_start, pg_end;
+		pg_start = origin >> PAGE_CACHE_SHIFT;
+		pg_end = (origin + retval - 1) >> PAGE_CACHE_SHIFT;
+		if (inode->i_mapping && inode->i_mapping->nrpages)
+			invalidate_inode_pages2_range(inode->i_mapping,
+						      pg_start, pg_end);
+		origin += retval;
+		i_size = i_size_read(inode);
+		iocb->ki_pos = origin;
+		if (origin > i_size) {
+			inode_add_bytes(inode, origin - i_size);
+			i_size_write(inode, origin);
+		}
+		return retval;
+	}
+	return err;
 }
 
-
 static int v9fs_file_fsync(struct file *filp, loff_t start, loff_t end,
 			   int datasync)
 {
@@ -657,44 +566,6 @@ out_unlock:
 	return VM_FAULT_NOPAGE;
 }
 
-static ssize_t
-v9fs_direct_read(struct file *filp, char __user *udata, size_t count,
-		 loff_t *offsetp)
-{
-	loff_t size, offset;
-	struct inode *inode;
-	struct address_space *mapping;
-
-	offset = *offsetp;
-	mapping = filp->f_mapping;
-	inode = mapping->host;
-	if (!count)
-		return 0;
-	size = i_size_read(inode);
-	if (offset < size)
-		filemap_write_and_wait_range(mapping, offset,
-					     offset + count - 1);
-
-	return v9fs_file_read(filp, udata, count, offsetp);
-}
-
-/**
- * v9fs_cached_file_read - read from a file
- * @filp: file pointer to read
- * @data: user data buffer to read data into
- * @count: size of buffer
- * @offset: offset at which to read data
- *
- */
-static ssize_t
-v9fs_cached_file_read(struct file *filp, char __user *data, size_t count,
-		      loff_t *offset)
-{
-	if (filp->f_flags & O_DIRECT)
-		return v9fs_direct_read(filp, data, count, offset);
-	return new_sync_read(filp, data, count, offset);
-}
-
 /**
  * v9fs_mmap_file_read - read from a file
  * @filp: file pointer to read
@@ -704,84 +575,12 @@ v9fs_cached_file_read(struct file *filp, char __user *data, size_t count,
  *
  */
 static ssize_t
-v9fs_mmap_file_read(struct file *filp, char __user *data, size_t count,
-		      loff_t *offset)
+v9fs_mmap_file_read_iter(struct kiocb *iocb, struct iov_iter *to)
 {
 	/* TODO: Check if there are dirty pages */
-	return v9fs_file_read(filp, data, count, offset);
+	return v9fs_file_read_iter(iocb, to);
 }
 
-static ssize_t
-v9fs_direct_write(struct file *filp, const char __user * data,
-		  size_t count, loff_t *offsetp)
-{
-	loff_t offset;
-	ssize_t retval;
-	struct inode *inode;
-	struct address_space *mapping;
-
-	offset = *offsetp;
-	mapping = filp->f_mapping;
-	inode = mapping->host;
-	if (!count)
-		return 0;
-
-	mutex_lock(&inode->i_mutex);
-	retval = filemap_write_and_wait_range(mapping, offset,
-					      offset + count - 1);
-	if (retval)
-		goto err_out;
-	/*
-	 * After a write we want buffered reads to be sure to go to disk to get
-	 * the new data.  We invalidate clean cached page from the region we're
-	 * about to write.  We do this *before* the write so that if we fail
-	 * here we fall back to buffered write
-	 */
-	if (mapping->nrpages) {
-		pgoff_t pg_start = offset >> PAGE_CACHE_SHIFT;
-		pgoff_t pg_end   = (offset + count - 1) >> PAGE_CACHE_SHIFT;
-
-		retval = invalidate_inode_pages2_range(mapping,
-							pg_start, pg_end);
-		/*
-		 * If a page can not be invalidated, fall back
-		 * to buffered write.
-		 */
-		if (retval) {
-			if (retval == -EBUSY)
-				goto buff_write;
-			goto err_out;
-		}
-	}
-	retval = v9fs_file_write(filp, data, count, offsetp);
-err_out:
-	mutex_unlock(&inode->i_mutex);
-	return retval;
-
-buff_write:
-	mutex_unlock(&inode->i_mutex);
-	return new_sync_write(filp, data, count, offsetp);
-}
-
-/**
- * v9fs_cached_file_write - write to a file
- * @filp: file pointer to write
- * @data: data buffer to write data from
- * @count: size of buffer
- * @offset: offset at which to write data
- *
- */
-static ssize_t
-v9fs_cached_file_write(struct file *filp, const char __user * data,
-		       size_t count, loff_t *offset)
-{
-
-	if (filp->f_flags & O_DIRECT)
-		return v9fs_direct_write(filp, data, count, offset);
-	return new_sync_write(filp, data, count, offset);
-}
-
-
 /**
  * v9fs_mmap_file_write - write to a file
  * @filp: file pointer to write
@@ -791,14 +590,13 @@ v9fs_cached_file_write(struct file *filp, const char __user * data,
  *
  */
 static ssize_t
-v9fs_mmap_file_write(struct file *filp, const char __user *data,
-		       size_t count, loff_t *offset)
+v9fs_mmap_file_write_iter(struct kiocb *iocb, struct iov_iter *from)
 {
 	/*
 	 * TODO: invalidate mmaps on filp's inode between
 	 * offset and offset+count
 	 */
-	return v9fs_file_write(filp, data, count, offset);
+	return v9fs_file_write_iter(iocb, from);
 }
 
 static void v9fs_mmap_vm_close(struct vm_area_struct *vma)
@@ -843,8 +641,8 @@ static const struct vm_operations_struct v9fs_mmap_file_vm_ops = {
 
 const struct file_operations v9fs_cached_file_operations = {
 	.llseek = generic_file_llseek,
-	.read = v9fs_cached_file_read,
-	.write = v9fs_cached_file_write,
+	.read = new_sync_read,
+	.write = new_sync_write,
 	.read_iter = generic_file_read_iter,
 	.write_iter = generic_file_write_iter,
 	.open = v9fs_file_open,
@@ -856,8 +654,8 @@ const struct file_operations v9fs_cached_file_operations = {
 
 const struct file_operations v9fs_cached_file_operations_dotl = {
 	.llseek = generic_file_llseek,
-	.read = v9fs_cached_file_read,
-	.write = v9fs_cached_file_write,
+	.read = new_sync_read,
+	.write = new_sync_write,
 	.read_iter = generic_file_read_iter,
 	.write_iter = generic_file_write_iter,
 	.open = v9fs_file_open,
@@ -870,8 +668,10 @@ const struct file_operations v9fs_cached_file_operations_dotl = {
 
 const struct file_operations v9fs_file_operations = {
 	.llseek = generic_file_llseek,
-	.read = v9fs_file_read,
-	.write = v9fs_file_write,
+	.read = new_sync_read,
+	.write = new_sync_write,
+	.read_iter = v9fs_file_read_iter,
+	.write_iter = v9fs_file_write_iter,
 	.open = v9fs_file_open,
 	.release = v9fs_dir_release,
 	.lock = v9fs_file_lock,
@@ -881,8 +681,10 @@ const struct file_operations v9fs_file_operations = {
 
 const struct file_operations v9fs_file_operations_dotl = {
 	.llseek = generic_file_llseek,
-	.read = v9fs_file_read,
-	.write = v9fs_file_write,
+	.read = new_sync_read,
+	.write = new_sync_write,
+	.read_iter = v9fs_file_read_iter,
+	.write_iter = v9fs_file_write_iter,
 	.open = v9fs_file_open,
 	.release = v9fs_dir_release,
 	.lock = v9fs_file_lock_dotl,
@@ -893,8 +695,10 @@ const struct file_operations v9fs_file_operations_dotl = {
 
 const struct file_operations v9fs_mmap_file_operations = {
 	.llseek = generic_file_llseek,
-	.read = v9fs_mmap_file_read,
-	.write = v9fs_mmap_file_write,
+	.read = new_sync_read,
+	.write = new_sync_write,
+	.read_iter = v9fs_mmap_file_read_iter,
+	.write_iter = v9fs_mmap_file_write_iter,
 	.open = v9fs_file_open,
 	.release = v9fs_dir_release,
 	.lock = v9fs_file_lock,
@@ -904,8 +708,10 @@ const struct file_operations v9fs_mmap_file_operations = {
 
 const struct file_operations v9fs_mmap_file_operations_dotl = {
 	.llseek = generic_file_llseek,
-	.read = v9fs_mmap_file_read,
-	.write = v9fs_mmap_file_write,
+	.read = new_sync_read,
+	.write = new_sync_write,
+	.read_iter = v9fs_mmap_file_read_iter,
+	.write_iter = v9fs_mmap_file_write_iter,
 	.open = v9fs_file_open,
 	.release = v9fs_dir_release,
 	.lock = v9fs_file_lock_dotl,

fs/9p/xattr.c

@@ -15,6 +15,7 @@
 #include <linux/module.h>
 #include <linux/fs.h>
 #include <linux/sched.h>
+#include <linux/uio.h>
 #include <net/9p/9p.h>
 #include <net/9p/client.h>
 
@@ -25,50 +26,34 @@ ssize_t v9fs_fid_xattr_get(struct p9_fid *fid, const char *name,
 			   void *buffer, size_t buffer_size)
 {
 	ssize_t retval;
-	int msize, read_count;
-	u64 offset = 0, attr_size;
+	u64 attr_size;
 	struct p9_fid *attr_fid;
+	struct kvec kvec = {.iov_base = buffer, .iov_len = buffer_size};
+	struct iov_iter to;
+	int err;
+
+	iov_iter_kvec(&to, READ | ITER_KVEC, &kvec, 1, buffer_size);
 
 	attr_fid = p9_client_xattrwalk(fid, name, &attr_size);
 	if (IS_ERR(attr_fid)) {
 		retval = PTR_ERR(attr_fid);
 		p9_debug(P9_DEBUG_VFS, "p9_client_attrwalk failed %zd\n",
 			 retval);
-		attr_fid = NULL;
-		goto error;
-	}
-	if (!buffer_size) {
-		/* request to get the attr_size */
-		retval = attr_size;
-		goto error;
+		return retval;
 	}
 	if (attr_size > buffer_size) {
-		retval = -ERANGE;
-		goto error;
-	}
-	msize = attr_fid->clnt->msize;
-	while (attr_size) {
-		if (attr_size > (msize - P9_IOHDRSZ))
-			read_count = msize - P9_IOHDRSZ;
+		if (!buffer_size) /* request to get the attr_size */
+			retval = attr_size;
 		else
-			read_count = attr_size;
-		read_count = p9_client_read(attr_fid, ((char *)buffer)+offset,
-					NULL, offset, read_count);
-		if (read_count < 0) {
-			/* error in xattr read */
-			retval = read_count;
-			goto error;
-		}
-		offset += read_count;
-		attr_size -= read_count;
+			retval = -ERANGE;
+	} else {
+		iov_iter_truncate(&to, attr_size);
+		retval = p9_client_read(attr_fid, 0, &to, &err);
+		if (err)
+			retval = err;
 	}
-	/* Total read xattr bytes */
-	retval = offset;
-error:
-	if (attr_fid)
-		p9_client_clunk(attr_fid);
+	p9_client_clunk(attr_fid);
 	return retval;
-
 }
 
 
@@ -120,8 +105,11 @@ int v9fs_xattr_set(struct dentry *dentry, const char *name,
 int v9fs_fid_xattr_set(struct p9_fid *fid, const char *name,
 		   const void *value, size_t value_len, int flags)
 {
-	u64 offset = 0;
-	int retval, msize, write_count;
+	struct kvec kvec = {.iov_base = (void *)value, .iov_len = value_len};
+	struct iov_iter from;
+	int retval;
+
+	iov_iter_kvec(&from, WRITE | ITER_KVEC, &kvec, 1, value_len);
 
 	p9_debug(P9_DEBUG_VFS, "name = %s value_len = %zu flags = %d\n",
 		 name, value_len, flags);
@@ -135,29 +123,11 @@ int v9fs_fid_xattr_set(struct p9_fid *fid, const char *name,
 	 * On success fid points to xattr
 	 */
 	retval = p9_client_xattrcreate(fid, name, value_len, flags);
-	if (retval < 0) {
+	if (retval < 0)
 		p9_debug(P9_DEBUG_VFS, "p9_client_xattrcreate failed %d\n",
 			 retval);
-		goto err;
-	}
-	msize = fid->clnt->msize;
-	while (value_len) {
-		if (value_len > (msize - P9_IOHDRSZ))
-			write_count = msize - P9_IOHDRSZ;
-		else
-			write_count = value_len;
-		write_count = p9_client_write(fid, ((char *)value)+offset,
-					NULL, offset, write_count);
-		if (write_count < 0) {
-			/* error in xattr write */
-			retval = write_count;
-			goto err;
-		}
-		offset += write_count;
-		value_len -= write_count;
-	}
-	retval = 0;
-err:
+	else
+		p9_client_write(fid, 0, &from, &retval);
 	p9_client_clunk(fid);
 	return retval;
 }

include/net/9p/client.h

@@ -211,6 +211,8 @@ struct p9_dirent {
 	char d_name[256];
 };
 
+struct iov_iter;
+
 int p9_client_statfs(struct p9_fid *fid, struct p9_rstatfs *sb);
 int p9_client_rename(struct p9_fid *fid, struct p9_fid *newdirfid,
 		     const char *name);
@@ -236,10 +238,8 @@ int p9_client_clunk(struct p9_fid *fid);
 int p9_client_fsync(struct p9_fid *fid, int datasync);
 int p9_client_remove(struct p9_fid *fid);
 int p9_client_unlinkat(struct p9_fid *dfid, const char *name, int flags);
-int p9_client_read(struct p9_fid *fid, char *data, char __user *udata,
-							u64 offset, u32 count);
-int p9_client_write(struct p9_fid *fid, char *data, const char __user *udata,
-							u64 offset, u32 count);
+int p9_client_read(struct p9_fid *fid, u64 offset, struct iov_iter *to, int *err);
+int p9_client_write(struct p9_fid *fid, u64 offset, struct iov_iter *from, int *err);
 int p9_client_readdir(struct p9_fid *fid, char *data, u32 count, u64 offset);
 int p9dirent_read(struct p9_client *clnt, char *buf, int len,
 		  struct p9_dirent *dirent);

include/net/9p/transport.h

@@ -61,7 +61,7 @@ struct p9_trans_module {
 	int (*cancel) (struct p9_client *, struct p9_req_t *req);
 	int (*cancelled)(struct p9_client *, struct p9_req_t *req);
 	int (*zc_request)(struct p9_client *, struct p9_req_t *,
-			  char *, char *, int , int, int, int);
+			  struct iov_iter *, struct iov_iter *, int , int, int);
 };
 
 void v9fs_register_trans(struct p9_trans_module *m);

net/9p/client.c

@@ -34,6 +34,7 @@
 #include <linux/slab.h>
 #include <linux/sched.h>
 #include <linux/uaccess.h>
+#include <linux/uio.h>
 #include <net/9p/9p.h>
 #include <linux/parser.h>
 #include <net/9p/client.h>
@@ -555,7 +556,7 @@ out_err:
  */
 
 static int p9_check_zc_errors(struct p9_client *c, struct p9_req_t *req,
-			      char *uidata, int in_hdrlen, int kern_buf)
+			      struct iov_iter *uidata, int in_hdrlen)
 {
 	int err;
 	int ecode;
@@ -591,16 +592,11 @@ static int p9_check_zc_errors(struct p9_client *c, struct p9_req_t *req,
 		ename = &req->rc->sdata[req->rc->offset];
 		if (len > inline_len) {
 			/* We have error in external buffer */
-			if (kern_buf) {
-				memcpy(ename + inline_len, uidata,
-				       len - inline_len);
-			} else {
-				err = copy_from_user(ename + inline_len,
-						     uidata, len - inline_len);
-				if (err) {
-					err = -EFAULT;
-					goto out_err;
-				}
+			err = copy_from_iter(ename + inline_len,
+					     len - inline_len, uidata);
+			if (err != len - inline_len) {
+				err = -EFAULT;
+				goto out_err;
 			}
 		}
 		ename = NULL;
@@ -806,8 +802,8 @@ reterr:
  * p9_client_zc_rpc - issue a request and wait for a response
  * @c: client session
  * @type: type of request
- * @uidata: user bffer that should be ued for zero copy read
- * @uodata: user buffer that shoud be user for zero copy write
+ * @uidata: destination for zero copy read
+ * @uodata: source for zero copy write
  * @inlen: read buffer size
  * @olen: write buffer size
  * @hdrlen: reader header size, This is the size of response protocol data
@@ -816,9 +812,10 @@ reterr:
  * Returns request structure (which client must free using p9_free_req)
  */
 static struct p9_req_t *p9_client_zc_rpc(struct p9_client *c, int8_t type,
-					 char *uidata, char *uodata,
+					 struct iov_iter *uidata,
+					 struct iov_iter *uodata,
 					 int inlen, int olen, int in_hdrlen,
-					 int kern_buf, const char *fmt, ...)
+					 const char *fmt, ...)
 {
 	va_list ap;
 	int sigpending, err;
@@ -841,12 +838,8 @@ static struct p9_req_t *p9_client_zc_rpc(struct p9_client *c, int8_t type,
 	} else
 		sigpending = 0;
 
-	/* If we are called with KERNEL_DS force kern_buf */
-	if (segment_eq(get_fs(), KERNEL_DS))
-		kern_buf = 1;
-
 	err = c->trans_mod->zc_request(c, req, uidata, uodata,
-				       inlen, olen, in_hdrlen, kern_buf);
+				       inlen, olen, in_hdrlen);
 	if (err < 0) {
 		if (err == -EIO)
 			c->status = Disconnected;
@@ -876,7 +869,7 @@ static struct p9_req_t *p9_client_zc_rpc(struct p9_client *c, int8_t type,
 	if (err < 0)
 		goto reterr;
 
-	err = p9_check_zc_errors(c, req, uidata, in_hdrlen, kern_buf);
+	err = p9_check_zc_errors(c, req, uidata, in_hdrlen);
 	trace_9p_client_res(c, type, req->rc->tag, err);
 	if (!err)
 		return req;
@@ -1123,6 +1116,7 @@ struct p9_fid *p9_client_attach(struct p9_client *clnt, struct p9_fid *afid,
 		fid = NULL;
 		goto error;
 	}
+	fid->uid = n_uname;
 
 	req = p9_client_rpc(clnt, P9_TATTACH, "ddss?u", fid->fid,
 			afid ? afid->fid : P9_NOFID, uname, aname, n_uname);
@@ -1541,142 +1535,128 @@ error:
 EXPORT_SYMBOL(p9_client_unlinkat);
 
 int
-p9_client_read(struct p9_fid *fid, char *data, char __user *udata, u64 offset,
-								u32 count)
+p9_client_read(struct p9_fid *fid, u64 offset, struct iov_iter *to, int *err)
 {
-	char *dataptr;
-	int kernel_buf = 0;
+	struct p9_client *clnt = fid->clnt;
 	struct p9_req_t *req;
-	struct p9_client *clnt;
-	int err, rsize, non_zc = 0;
-
+	int total = 0;
 
 	p9_debug(P9_DEBUG_9P, ">>> TREAD fid %d offset %llu %d\n",
-		   fid->fid, (unsigned long long) offset, count);
-	err = 0;
-	clnt = fid->clnt;
-
-	rsize = fid->iounit;
-	if (!rsize || rsize > clnt->msize-P9_IOHDRSZ)
-		rsize = clnt->msize - P9_IOHDRSZ;
-
-	if (count < rsize)
-		rsize = count;
-
-	/* Don't bother zerocopy for small IO (< 1024) */
-	if (clnt->trans_mod->zc_request && rsize > 1024) {
-		char *indata;
-		if (data) {
-			kernel_buf = 1;
-			indata = data;
-		} else
-			indata = (__force char *)udata;
-		/*
-		 * response header len is 11
-		 * PDU Header(7) + IO Size (4)
-		 */
-		req = p9_client_zc_rpc(clnt, P9_TREAD, indata, NULL, rsize, 0,
-				       11, kernel_buf, "dqd", fid->fid,
-				       offset, rsize);
-	} else {
-		non_zc = 1;
-		req = p9_client_rpc(clnt, P9_TREAD, "dqd", fid->fid, offset,
-				    rsize);
-	}
-	if (IS_ERR(req)) {
-		err = PTR_ERR(req);
-		goto error;
-	}
+		   fid->fid, (unsigned long long) offset, (int)iov_iter_count(to));
+
+	while (iov_iter_count(to)) {
+		int count = iov_iter_count(to);
+		int rsize, non_zc = 0;
+		char *dataptr;
+			
+		rsize = fid->iounit;
+		if (!rsize || rsize > clnt->msize-P9_IOHDRSZ)
+			rsize = clnt->msize - P9_IOHDRSZ;
+
+		if (count < rsize)
+			rsize = count;
+
+		/* Don't bother zerocopy for small IO (< 1024) */
+		if (clnt->trans_mod->zc_request && rsize > 1024) {
+			/*
+			 * response header len is 11
+			 * PDU Header(7) + IO Size (4)
+			 */
+			req = p9_client_zc_rpc(clnt, P9_TREAD, to, NULL, rsize,
+					       0, 11, "dqd", fid->fid,
+					       offset, rsize);
+		} else {
+			non_zc = 1;
+			req = p9_client_rpc(clnt, P9_TREAD, "dqd", fid->fid, offset,
+					    rsize);
+		}
+		if (IS_ERR(req)) {
+			*err = PTR_ERR(req);
+			break;
+		}
 
-	err = p9pdu_readf(req->rc, clnt->proto_version, "D", &count, &dataptr);
-	if (err) {
-		trace_9p_protocol_dump(clnt, req->rc);
-		goto free_and_error;
-	}
+		*err = p9pdu_readf(req->rc, clnt->proto_version,
+				   "D", &count, &dataptr);
+		if (*err) {
+			trace_9p_protocol_dump(clnt, req->rc);
+			p9_free_req(clnt, req);
+			break;
+		}
 
-	p9_debug(P9_DEBUG_9P, "<<< RREAD count %d\n", count);
+		p9_debug(P9_DEBUG_9P, "<<< RREAD count %d\n", count);
+		if (!count) {
+			p9_free_req(clnt, req);
+			break;
+		}
 
-	if (non_zc) {
-		if (data) {
-			memmove(data, dataptr, count);
-		} else {
-			err = copy_to_user(udata, dataptr, count);
-			if (err) {
-				err = -EFAULT;
-				goto free_and_error;
+		if (non_zc) {
+			int n = copy_to_iter(dataptr, count, to);
+			total += n;
+			offset += n;
+			if (n != count) {
+				*err = -EFAULT;
+				p9_free_req(clnt, req);
+				break;
 			}
+		} else {
+			iov_iter_advance(to, count);
+			total += count;
+			offset += count;
 		}
+		p9_free_req(clnt, req);
 	}
-	p9_free_req(clnt, req);
-	return count;
-
-free_and_error:
-	p9_free_req(clnt, req);
-error:
-	return err;
+	return total;
 }
 EXPORT_SYMBOL(p9_client_read);
 
 int
-p9_client_write(struct p9_fid *fid, char *data, const char __user *udata,
-							u64 offset, u32 count)
+p9_client_write(struct p9_fid *fid, u64 offset, struct iov_iter *from, int *err)
 {
-	int err, rsize;
-	int kernel_buf = 0;
-	struct p9_client *clnt;
+	struct p9_client *clnt = fid->clnt;
 	struct p9_req_t *req;
+	int total = 0;
+
+	p9_debug(P9_DEBUG_9P, ">>> TWRITE fid %d offset %llu count %zd\n",
+				fid->fid, (unsigned long long) offset,
+				iov_iter_count(from));
+
+	while (iov_iter_count(from)) {
+		int count = iov_iter_count(from);
+		int rsize = fid->iounit;
+		if (!rsize || rsize > clnt->msize-P9_IOHDRSZ)
+			rsize = clnt->msize - P9_IOHDRSZ;
+
+		if (count < rsize)
+			rsize = count;
+
+		/* Don't bother zerocopy for small IO (< 1024) */
+		if (clnt->trans_mod->zc_request && rsize > 1024) {
+			req = p9_client_zc_rpc(clnt, P9_TWRITE, NULL, from, 0,
+					       rsize, P9_ZC_HDR_SZ, "dqd",
+					       fid->fid, offset, rsize);
+		} else {
+			req = p9_client_rpc(clnt, P9_TWRITE, "dqV", fid->fid,
+						    offset, rsize, from);
+		}
+		if (IS_ERR(req)) {
+			*err = PTR_ERR(req);
+			break;
+		}
 
-	p9_debug(P9_DEBUG_9P, ">>> TWRITE fid %d offset %llu count %d\n",
-				fid->fid, (unsigned long long) offset, count);
-	err = 0;
-	clnt = fid->clnt;
-
-	rsize = fid->iounit;
-	if (!rsize || rsize > clnt->msize-P9_IOHDRSZ)
-		rsize = clnt->msize - P9_IOHDRSZ;
+		*err = p9pdu_readf(req->rc, clnt->proto_version, "d", &count);
+		if (*err) {
+			trace_9p_protocol_dump(clnt, req->rc);
+			p9_free_req(clnt, req);
+		}
 
-	if (count < rsize)
-		rsize = count;
+		p9_debug(P9_DEBUG_9P, "<<< RWRITE count %d\n", count);
 
-	/* Don't bother zerocopy for small IO (< 1024) */
-	if (clnt->trans_mod->zc_request && rsize > 1024) {
-		char *odata;
-		if (data) {
-			kernel_buf = 1;
-			odata = data;
-		} else
-			odata = (char *)udata;
-		req = p9_client_zc_rpc(clnt, P9_TWRITE, NULL, odata, 0, rsize,
-				       P9_ZC_HDR_SZ, kernel_buf, "dqd",
-				       fid->fid, offset, rsize);
-	} else {
-		if (data)
-			req = p9_client_rpc(clnt, P9_TWRITE, "dqD", fid->fid,
-					    offset, rsize, data);
-		else
-			req = p9_client_rpc(clnt, P9_TWRITE, "dqU", fid->fid,
-					    offset, rsize, udata);
-	}
-	if (IS_ERR(req)) {
-		err = PTR_ERR(req);
-		goto error;
-	}
-
-	err = p9pdu_readf(req->rc, clnt->proto_version, "d", &count);
-	if (err) {
-		trace_9p_protocol_dump(clnt, req->rc);
-		goto free_and_error;
+		p9_free_req(clnt, req);
+		iov_iter_advance(from, count);
+		total += count;
+		offset += count;
 	}
-
-	p9_debug(P9_DEBUG_9P, "<<< RWRITE count %d\n", count);
-
-	p9_free_req(clnt, req);
-	return count;
-
-free_and_error:
-	p9_free_req(clnt, req);
-error:
-	return err;
+	return total;
 }
 EXPORT_SYMBOL(p9_client_write);
 
@@ -2068,6 +2048,10 @@ int p9_client_readdir(struct p9_fid *fid, char *data, u32 count, u64 offset)
 	struct p9_client *clnt;
 	struct p9_req_t *req;
 	char *dataptr;
+	struct kvec kv = {.iov_base = data, .iov_len = count};
+	struct iov_iter to;
+
+	iov_iter_kvec(&to, READ | ITER_KVEC, &kv, 1, count);
 
 	p9_debug(P9_DEBUG_9P, ">>> TREADDIR fid %d offset %llu count %d\n",
 				fid->fid, (unsigned long long) offset, count);
@@ -2088,8 +2072,8 @@ int p9_client_readdir(struct p9_fid *fid, char *data, u32 count, u64 offset)
 		 * response header len is 11
 		 * PDU Header(7) + IO Size (4)
 		 */
-		req = p9_client_zc_rpc(clnt, P9_TREADDIR, data, NULL, rsize, 0,
-				       11, 1, "dqd", fid->fid, offset, rsize);
+		req = p9_client_zc_rpc(clnt, P9_TREADDIR, &to, NULL, rsize, 0,
+				       11, "dqd", fid->fid, offset, rsize);
 	} else {
 		non_zc = 1;
 		req = p9_client_rpc(clnt, P9_TREADDIR, "dqd", fid->fid,

net/9p/protocol.c

@@ -33,6 +33,7 @@
 #include <linux/sched.h>
 #include <linux/stddef.h>
 #include <linux/types.h>
+#include <linux/uio.h>
 #include <net/9p/9p.h>
 #include <net/9p/client.h>
 #include "protocol.h"
@@ -69,10 +70,11 @@ static size_t pdu_write(struct p9_fcall *pdu, const void *data, size_t size)
 }
 
 static size_t
-pdu_write_u(struct p9_fcall *pdu, const char __user *udata, size_t size)
+pdu_write_u(struct p9_fcall *pdu, struct iov_iter *from, size_t size)
 {
 	size_t len = min(pdu->capacity - pdu->size, size);
-	if (copy_from_user(&pdu->sdata[pdu->size], udata, len))
+	struct iov_iter i = *from;
+	if (copy_from_iter(&pdu->sdata[pdu->size], len, &i) != len)
 		len = 0;
 
 	pdu->size += len;
@@ -437,23 +439,13 @@ p9pdu_vwritef(struct p9_fcall *pdu, int proto_version, const char *fmt,
 						 stbuf->extension, stbuf->n_uid,
 						 stbuf->n_gid, stbuf->n_muid);
 			} break;
-		case 'D':{
-				uint32_t count = va_arg(ap, uint32_t);
-				const void *data = va_arg(ap, const void *);
-
-				errcode = p9pdu_writef(pdu, proto_version, "d",
-									count);
-				if (!errcode && pdu_write(pdu, data, count))
-					errcode = -EFAULT;
-			}
-			break;
-		case 'U':{
+		case 'V':{
 				int32_t count = va_arg(ap, int32_t);
-				const char __user *udata =
-						va_arg(ap, const void __user *);
+				struct iov_iter *from =
+						va_arg(ap, struct iov_iter *);
 				errcode = p9pdu_writef(pdu, proto_version, "d",
 									count);
-				if (!errcode && pdu_write_u(pdu, udata, count))
+				if (!errcode && pdu_write_u(pdu, from, count))
 					errcode = -EFAULT;
 			}
 			break;

net/9p/trans_common.c

@@ -12,12 +12,8 @@
  *
  */
 
-#include <linux/slab.h>
+#include <linux/mm.h>
 #include <linux/module.h>
-#include <net/9p/9p.h>
-#include <net/9p/client.h>
-#include <linux/scatterlist.h>
-#include "trans_common.h"
 
 /**
  *  p9_release_req_pages - Release pages after the transaction.
@@ -31,39 +27,3 @@ void p9_release_pages(struct page **pages, int nr_pages)
 			put_page(pages[i]);
 }
 EXPORT_SYMBOL(p9_release_pages);
-
-/**
- * p9_nr_pages - Return number of pages needed to accommodate the payload.
- */
-int p9_nr_pages(char *data, int len)
-{
-	unsigned long start_page, end_page;
-	start_page =  (unsigned long)data >> PAGE_SHIFT;
-	end_page = ((unsigned long)data + len + PAGE_SIZE - 1) >> PAGE_SHIFT;
-	return end_page - start_page;
-}
-EXPORT_SYMBOL(p9_nr_pages);
-
-/**
- * payload_gup - Translates user buffer into kernel pages and
- * pins them either for read/write through get_user_pages_fast().
- * @req: Request to be sent to server.
- * @pdata_off: data offset into the first page after translation (gup).
- * @pdata_len: Total length of the IO. gup may not return requested # of pages.
- * @nr_pages: number of pages to accommodate the payload
- * @rw: Indicates if the pages are for read or write.
- */
-
-int p9_payload_gup(char *data, int *nr_pages, struct page **pages, int write)
-{
-	int nr_mapped_pages;
-
-	nr_mapped_pages = get_user_pages_fast((unsigned long)data,
-					      *nr_pages, write, pages);
-	if (nr_mapped_pages <= 0)
-		return nr_mapped_pages;
-
-	*nr_pages = nr_mapped_pages;
-	return 0;
-}
-EXPORT_SYMBOL(p9_payload_gup);

net/9p/trans_common.h

@@ -13,5 +13,3 @@
  */
 
 void p9_release_pages(struct page **, int);
-int p9_payload_gup(char *, int *, struct page **, int);
-int p9_nr_pages(char *, int);

net/9p/trans_virtio.c

@@ -217,15 +217,15 @@ static int p9_virtio_cancel(struct p9_client *client, struct p9_req_t *req)
  * @start: which segment of the sg_list to start at
  * @pdata: a list of pages to add into sg.
  * @nr_pages: number of pages to pack into the scatter/gather list
- * @data: data to pack into scatter/gather list
+ * @offs: amount of data in the beginning of first page _not_ to pack
  * @count: amount of data to pack into the scatter/gather list
  */
 static int
 pack_sg_list_p(struct scatterlist *sg, int start, int limit,
-	       struct page **pdata, int nr_pages, char *data, int count)
+	       struct page **pdata, int nr_pages, size_t offs, int count)
 {
 	int i = 0, s;
-	int data_off;
+	int data_off = offs;
 	int index = start;
 
 	BUG_ON(nr_pages > (limit - start));
@@ -233,16 +233,14 @@ pack_sg_list_p(struct scatterlist *sg, int start, int limit,
 	 * if the first page doesn't start at
 	 * page boundary find the offset
 	 */
-	data_off = offset_in_page(data);
 	while (nr_pages) {
-		s = rest_of_page(data);
+		s = PAGE_SIZE - data_off;
 		if (s > count)
 			s = count;
 		/* Make sure we don't terminate early. */
 		sg_unmark_end(&sg[index]);
 		sg_set_page(&sg[index++], pdata[i++], s, data_off);
 		data_off = 0;
-		data += s;
 		count -= s;
 		nr_pages--;
 	}
@@ -314,11 +312,20 @@ req_retry:
 }
 
 static int p9_get_mapped_pages(struct virtio_chan *chan,
-			       struct page **pages, char *data,
-			       int nr_pages, int write, int kern_buf)
+			       struct page ***pages,
+			       struct iov_iter *data,
+			       int count,
+			       size_t *offs,
+			       int *need_drop)
 {
+	int nr_pages;
 	int err;
-	if (!kern_buf) {
+
+	if (!iov_iter_count(data))
+		return 0;
+
+	if (!(data->type & ITER_KVEC)) {
+		int n;
 		/*
 		 * We allow only p9_max_pages pinned. We wait for the
 		 * Other zc request to finish here
@@ -329,26 +336,49 @@ static int p9_get_mapped_pages(struct virtio_chan *chan,
 			if (err == -ERESTARTSYS)
 				return err;
 		}
-		err = p9_payload_gup(data, &nr_pages, pages, write);
-		if (err < 0)
-			return err;
+		n = iov_iter_get_pages_alloc(data, pages, count, offs);
+		if (n < 0)
+			return n;
+		*need_drop = 1;
+		nr_pages = DIV_ROUND_UP(n + *offs, PAGE_SIZE);
 		atomic_add(nr_pages, &vp_pinned);
+		return n;
 	} else {
 		/* kernel buffer, no need to pin pages */
-		int s, index = 0;
-		int count = nr_pages;
-		while (nr_pages) {
-			s = rest_of_page(data);
-			if (is_vmalloc_addr(data))
-				pages[index++] = vmalloc_to_page(data);
+		int index;
+		size_t len;
+		void *p;
+
+		/* we'd already checked that it's non-empty */
+		while (1) {
+			len = iov_iter_single_seg_count(data);
+			if (likely(len)) {
+				p = data->kvec->iov_base + data->iov_offset;
+				break;
+			}
+			iov_iter_advance(data, 0);
+		}
+		if (len > count)
+			len = count;
+
+		nr_pages = DIV_ROUND_UP((unsigned long)p + len, PAGE_SIZE) -
+			   (unsigned long)p / PAGE_SIZE;
+
+		*pages = kmalloc(sizeof(struct page *) * nr_pages, GFP_NOFS);
+		if (!*pages)
+			return -ENOMEM;
+
+		*need_drop = 0;
+		p -= (*offs = (unsigned long)p % PAGE_SIZE);
+		for (index = 0; index < nr_pages; index++) {
+			if (is_vmalloc_addr(p))
+				(*pages)[index] = vmalloc_to_page(p);
 			else
-				pages[index++] = kmap_to_page(data);
-			data += s;
-			nr_pages--;
+				(*pages)[index] = kmap_to_page(p);
+			p += PAGE_SIZE;
 		}
-		nr_pages = count;
+		return len;
 	}
-	return nr_pages;
 }
 
 /**
@@ -364,8 +394,8 @@ static int p9_get_mapped_pages(struct virtio_chan *chan,
  */
 static int
 p9_virtio_zc_request(struct p9_client *client, struct p9_req_t *req,
-		     char *uidata, char *uodata, int inlen,
-		     int outlen, int in_hdr_len, int kern_buf)
+		     struct iov_iter *uidata, struct iov_iter *uodata,
+		     int inlen, int outlen, int in_hdr_len)
 {
 	int in, out, err, out_sgs, in_sgs;
 	unsigned long flags;
@@ -373,41 +403,32 @@ p9_virtio_zc_request(struct p9_client *client, struct p9_req_t *req,
 	struct page **in_pages = NULL, **out_pages = NULL;
 	struct virtio_chan *chan = client->trans;
 	struct scatterlist *sgs[4];
+	size_t offs;
+	int need_drop = 0;
 
 	p9_debug(P9_DEBUG_TRANS, "virtio request\n");
 
 	if (uodata) {
-		out_nr_pages = p9_nr_pages(uodata, outlen);
-		out_pages = kmalloc(sizeof(struct page *) * out_nr_pages,
-				    GFP_NOFS);
-		if (!out_pages) {
-			err = -ENOMEM;
-			goto err_out;
-		}
-		out_nr_pages = p9_get_mapped_pages(chan, out_pages, uodata,
-						   out_nr_pages, 0, kern_buf);
-		if (out_nr_pages < 0) {
-			err = out_nr_pages;
-			kfree(out_pages);
-			out_pages = NULL;
-			goto err_out;
+		int n = p9_get_mapped_pages(chan, &out_pages, uodata,
+					    outlen, &offs, &need_drop);
+		if (n < 0)
+			return n;
+		out_nr_pages = DIV_ROUND_UP(n + offs, PAGE_SIZE);
+		if (n != outlen) {
+			__le32 v = cpu_to_le32(n);
+			memcpy(&req->tc->sdata[req->tc->size - 4], &v, 4);
+			outlen = n;
 		}
-	}
-	if (uidata) {
-		in_nr_pages = p9_nr_pages(uidata, inlen);
-		in_pages = kmalloc(sizeof(struct page *) * in_nr_pages,
-				   GFP_NOFS);
-		if (!in_pages) {
-			err = -ENOMEM;
-			goto err_out;
-		}
-		in_nr_pages = p9_get_mapped_pages(chan, in_pages, uidata,
-						  in_nr_pages, 1, kern_buf);
-		if (in_nr_pages < 0) {
-			err = in_nr_pages;
-			kfree(in_pages);
-			in_pages = NULL;
-			goto err_out;
+	} else if (uidata) {
+		int n = p9_get_mapped_pages(chan, &in_pages, uidata,
+					    inlen, &offs, &need_drop);
+		if (n < 0)
+			return n;
+		in_nr_pages = DIV_ROUND_UP(n + offs, PAGE_SIZE);
+		if (n != inlen) {
+			__le32 v = cpu_to_le32(n);
+			memcpy(&req->tc->sdata[req->tc->size - 4], &v, 4);
+			inlen = n;
 		}
 	}
 	req->status = REQ_STATUS_SENT;
@@ -426,7 +447,7 @@ req_retry_pinned:
 	if (out_pages) {
 		sgs[out_sgs++] = chan->sg + out;
 		out += pack_sg_list_p(chan->sg, out, VIRTQUEUE_NUM,
-				      out_pages, out_nr_pages, uodata, outlen);
+				      out_pages, out_nr_pages, offs, outlen);
 	}
 		
 	/*
@@ -444,7 +465,7 @@ req_retry_pinned:
 	if (in_pages) {
 		sgs[out_sgs + in_sgs++] = chan->sg + out + in;
 		in += pack_sg_list_p(chan->sg, out + in, VIRTQUEUE_NUM,
-				     in_pages, in_nr_pages, uidata, inlen);
+				     in_pages, in_nr_pages, offs, inlen);
 	}
 
 	BUG_ON(out_sgs + in_sgs > ARRAY_SIZE(sgs));
@@ -478,7 +499,7 @@ req_retry_pinned:
 	 * Non kernel buffers are pinned, unpin them
 	 */
 err_out:
-	if (!kern_buf) {
+	if (need_drop) {
 		if (in_pages) {
 			p9_release_pages(in_pages, in_nr_pages);
 			atomic_sub(in_nr_pages, &vp_pinned);