From 34d0640e268923bb1fb9d244a047cdfcd3f77909 Mon Sep 17 00:00:00 2001
From: Al Viro <viro@zeniv.linux.org.uk>
Date: Fri, 3 Apr 2015 11:31:35 -0400
Subject: [PATCH] switch hugetlbfs to ->read_iter()

... and fix the case when the area we are asked to read crosses
a hugepage boundary

Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
---
 fs/hugetlbfs/inode.c | 92 ++++++++++++++++----------------------------
 1 file changed, 34 insertions(+), 58 deletions(-)

diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c
index c274aca8e8dc..5291c14ee6b8 100644
--- a/fs/hugetlbfs/inode.c
+++ b/fs/hugetlbfs/inode.c
@@ -34,6 +34,7 @@
 #include <linux/security.h>
 #include <linux/magic.h>
 #include <linux/migrate.h>
+#include <linux/uio.h>
 
 #include <asm/uaccess.h>
 
@@ -179,42 +180,33 @@ hugetlb_get_unmapped_area(struct file *file, unsigned long addr,
 }
 #endif
 
-static int
+static size_t
 hugetlbfs_read_actor(struct page *page, unsigned long offset,
-			char __user *buf, unsigned long count,
-			unsigned long size)
+			struct iov_iter *to, unsigned long size)
 {
-	char *kaddr;
-	unsigned long left, copied = 0;
+	size_t copied = 0;
 	int i, chunksize;
 
-	if (size > count)
-		size = count;
-
 	/* Find which 4k chunk and offset with in that chunk */
 	i = offset >> PAGE_CACHE_SHIFT;
 	offset = offset & ~PAGE_CACHE_MASK;
 
 	while (size) {
+		size_t n;
 		chunksize = PAGE_CACHE_SIZE;
 		if (offset)
 			chunksize -= offset;
 		if (chunksize > size)
 			chunksize = size;
-		kaddr = kmap(&page[i]);
-		left = __copy_to_user(buf, kaddr + offset, chunksize);
-		kunmap(&page[i]);
-		if (left) {
-			copied += (chunksize - left);
-			break;
-		}
+		n = copy_page_to_iter(&page[i], offset, chunksize, to);
+		copied += n;
+		if (n != chunksize)
+			return copied;
 		offset = 0;
 		size -= chunksize;
-		buf += chunksize;
-		copied += chunksize;
 		i++;
 	}
-	return copied ? copied : -EFAULT;
+	return copied;
 }
 
 /*
@@ -222,39 +214,34 @@ hugetlbfs_read_actor(struct page *page, unsigned long offset,
  * data. Its *very* similar to do_generic_mapping_read(), we can't use that
  * since it has PAGE_CACHE_SIZE assumptions.
  */
-static ssize_t hugetlbfs_read(struct file *filp, char __user *buf,
-			      size_t len, loff_t *ppos)
+static ssize_t hugetlbfs_read_iter(struct kiocb *iocb, struct iov_iter *to)
 {
-	struct hstate *h = hstate_file(filp);
-	struct address_space *mapping = filp->f_mapping;
+	struct file *file = iocb->ki_filp;
+	struct hstate *h = hstate_file(file);
+	struct address_space *mapping = file->f_mapping;
 	struct inode *inode = mapping->host;
-	unsigned long index = *ppos >> huge_page_shift(h);
-	unsigned long offset = *ppos & ~huge_page_mask(h);
+	unsigned long index = iocb->ki_pos >> huge_page_shift(h);
+	unsigned long offset = iocb->ki_pos & ~huge_page_mask(h);
 	unsigned long end_index;
 	loff_t isize;
 	ssize_t retval = 0;
 
-	/* validate length */
-	if (len == 0)
-		goto out;
-
-	for (;;) {
+	while (iov_iter_count(to)) {
 		struct page *page;
-		unsigned long nr, ret;
-		int ra;
+		size_t nr, copied;
 
 		/* nr is the maximum number of bytes to copy from this page */
 		nr = huge_page_size(h);
 		isize = i_size_read(inode);
 		if (!isize)
-			goto out;
+			break;
 		end_index = (isize - 1) >> huge_page_shift(h);
-		if (index >= end_index) {
-			if (index > end_index)
-				goto out;
+		if (index > end_index)
+			break;
+		if (index == end_index) {
 			nr = ((isize - 1) & ~huge_page_mask(h)) + 1;
 			if (nr <= offset)
-				goto out;
+				break;
 		}
 		nr = nr - offset;
 
@@ -265,39 +252,27 @@ static ssize_t hugetlbfs_read(struct file *filp, char __user *buf,
 			 * We have a HOLE, zero out the user-buffer for the
 			 * length of the hole or request.
 			 */
-			ret = len < nr ? len : nr;
-			if (clear_user(buf, ret))
-				ra = -EFAULT;
-			else
-				ra = 0;
+			copied = iov_iter_zero(nr, to);
 		} else {
 			unlock_page(page);
 
 			/*
 			 * We have the page, copy it to user space buffer.
 			 */
-			ra = hugetlbfs_read_actor(page, offset, buf, len, nr);
-			ret = ra;
+			copied = hugetlbfs_read_actor(page, offset, to, nr);
 			page_cache_release(page);
 		}
-		if (ra < 0) {
-			if (retval == 0)
-				retval = ra;
-			goto out;
+		offset += copied;
+		retval += copied;
+		if (copied != nr && iov_iter_count(to)) {
+			if (!retval)
+				retval = -EFAULT;
+			break;
 		}
-
-		offset += ret;
-		retval += ret;
-		len -= ret;
 		index += offset >> huge_page_shift(h);
 		offset &= ~huge_page_mask(h);
-
-		/* short read or no more work */
-		if ((ret != nr) || (len == 0))
-			break;
 	}
-out:
-	*ppos = ((loff_t)index << huge_page_shift(h)) + offset;
+	iocb->ki_pos = ((loff_t)index << huge_page_shift(h)) + offset;
 	return retval;
 }
 
@@ -721,7 +696,8 @@ static void init_once(void *foo)
 }
 
 const struct file_operations hugetlbfs_file_operations = {
-	.read			= hugetlbfs_read,
+	.read			= new_sync_read,
+	.read_iter		= hugetlbfs_read_iter,
 	.mmap			= hugetlbfs_file_mmap,
 	.fsync			= noop_fsync,
 	.get_unmapped_area	= hugetlb_get_unmapped_area,
-- 
2.30.2