[linux.git] / mm / filemap_xip.c

/*
 *	linux/mm/filemap_xip.c
 *
 * Copyright (C) 2005 IBM Corporation
 * Author: Carsten Otte <[email protected]>
 *
 * derived from linux/mm/filemap.c - Copyright (C) Linus Torvalds
 *
 */

#include <linux/fs.h>
#include <linux/pagemap.h>
#include <linux/export.h>
#include <linux/uio.h>
#include <linux/rmap.h>
#include <linux/mmu_notifier.h>
#include <linux/sched.h>
#include <linux/seqlock.h>
#include <linux/mutex.h>
#include <linux/gfp.h>
#include <asm/tlbflush.h>
#include <asm/io.h>

/*
 * We do use our own empty page to avoid interference with other users
 * of ZERO_PAGE(), such as /dev/zero
 */
static DEFINE_MUTEX(xip_sparse_mutex);
static seqcount_t xip_sparse_seq = SEQCNT_ZERO(xip_sparse_seq);
static struct page *__xip_sparse_page;

/* called under xip_sparse_mutex */
static struct page *xip_sparse_page(void)
{
	if (!__xip_sparse_page) {
		struct page *page = alloc_page(GFP_HIGHUSER | __GFP_ZERO);

		if (page)
			__xip_sparse_page = page;
	}
	return __xip_sparse_page;
}

/*
 * This is a file read routine for execute in place files, and uses
 * the mapping->a_ops->get_xip_mem() function for the actual low-level
 * stuff.
 *
 * Note the struct file* is not used at all.  It may be NULL.
 */
static ssize_t
do_xip_mapping_read(struct address_space *mapping,
		    struct file_ra_state *_ra,
		    struct file *filp,
		    char __user *buf,
		    size_t len,
		    loff_t *ppos)
{
	struct inode *inode = mapping->host;
	pgoff_t index, end_index;
	unsigned long offset;
	loff_t isize, pos;
	size_t copied = 0, error = 0;

	BUG_ON(!mapping->a_ops->get_xip_mem);

	pos = *ppos;
	index = pos >> PAGE_CACHE_SHIFT;
	offset = pos & ~PAGE_CACHE_MASK;

	isize = i_size_read(inode);
	if (!isize)
		goto out;

	end_index = (isize - 1) >> PAGE_CACHE_SHIFT;
	do {
		unsigned long nr, left;
		void *xip_mem;
		unsigned long xip_pfn;
		int zero = 0;

		/* nr is the maximum number of bytes to copy from this page */
		nr = PAGE_CACHE_SIZE;
		if (index >= end_index) {
			if (index > end_index)
				goto out;
			nr = ((isize - 1) & ~PAGE_CACHE_MASK) + 1;
			if (nr <= offset) {
				goto out;
			}
		}
		nr = nr - offset;
		if (nr > len - copied)
			nr = len - copied;

		error = mapping->a_ops->get_xip_mem(mapping, index, 0,
							&xip_mem, &xip_pfn);
		if (unlikely(error)) {
			if (error == -ENODATA) {
				/* sparse */
				zero = 1;
			} else
				goto out;
		}

		/* If users can be writing to this page using arbitrary
		 * virtual addresses, take care about potential aliasing
		 * before reading the page on the kernel side.
		 */
		if (mapping_writably_mapped(mapping))
			/* address based flush */ ;

		/*
		 * Ok, we have the mem, so now we can copy it to user space...
		 *
		 * The actor routine returns how many bytes were actually used..
		 * NOTE! This may not be the same as how much of a user buffer
		 * we filled up (we may be padding etc), so we can only update
		 * "pos" here (the actor routine has to update the user buffer
		 * pointers and the remaining count).
		 */
		if (!zero)
			left = __copy_to_user(buf+copied, xip_mem+offset, nr);
		else
			left = __clear_user(buf + copied, nr);

		if (left) {
			error = -EFAULT;
			goto out;
		}

		copied += (nr - left);
		offset += (nr - left);
		index += offset >> PAGE_CACHE_SHIFT;
		offset &= ~PAGE_CACHE_MASK;
	} while (copied < len);

out:
	*ppos = pos + copied;
	if (filp)
		file_accessed(filp);

	return (copied ? copied : error);
}

ssize_t
xip_file_read(struct file *filp, char __user *buf, size_t len, loff_t *ppos)
{
	if (!access_ok(VERIFY_WRITE, buf, len))
		return -EFAULT;

	return do_xip_mapping_read(filp->f_mapping, &filp->f_ra, filp,
			    buf, len, ppos);
}
EXPORT_SYMBOL_GPL(xip_file_read);

/*
 * __xip_unmap is invoked from xip_unmap and
 * xip_write
 *
 * This function walks all vmas of the address_space and unmaps the
 * __xip_sparse_page when found at pgoff.
 */
static void
__xip_unmap (struct address_space * mapping,
		     unsigned long pgoff)
{
	struct vm_area_struct *vma;
	struct mm_struct *mm;
	unsigned long address;
	pte_t *pte;
	pte_t pteval;
	spinlock_t *ptl;
	struct page *page;
	unsigned count;
	int locked = 0;

	count = read_seqcount_begin(&xip_sparse_seq);

	page = __xip_sparse_page;
	if (!page)
		return;

retry:
	mutex_lock(&mapping->i_mmap_mutex);
	vma_interval_tree_foreach(vma, &mapping->i_mmap, pgoff, pgoff) {
		mm = vma->vm_mm;
		address = vma->vm_start +
			((pgoff - vma->vm_pgoff) << PAGE_SHIFT);
		BUG_ON(address < vma->vm_start || address >= vma->vm_end);
		pte = page_check_address(page, mm, address, &ptl, 1);
		if (pte) {
			/* Nuke the page table entry. */
			flush_cache_page(vma, address, pte_pfn(*pte));
			pteval = ptep_clear_flush(vma, address, pte);
			page_remove_rmap(page);
			dec_mm_counter(mm, MM_FILEPAGES);
			BUG_ON(pte_dirty(pteval));
			pte_unmap_unlock(pte, ptl);
			/* must invalidate_page _before_ freeing the page */
			mmu_notifier_invalidate_page(mm, address);
			page_cache_release(page);
		}
	}
	mutex_unlock(&mapping->i_mmap_mutex);

	if (locked) {
		mutex_unlock(&xip_sparse_mutex);
	} else if (read_seqcount_retry(&xip_sparse_seq, count)) {
		mutex_lock(&xip_sparse_mutex);
		locked = 1;
		goto retry;
	}
}

/*
 * xip_fault() is invoked via the vma operations vector for a
 * mapped memory region to read in file data during a page fault.
 *
 * This function is derived from filemap_fault, but used for execute in place
 */
static int xip_file_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
{
	struct file *file = vma->vm_file;
	struct address_space *mapping = file->f_mapping;
	struct inode *inode = mapping->host;
	pgoff_t size;
	void *xip_mem;
	unsigned long xip_pfn;
	struct page *page;
	int error;

	/* XXX: are VM_FAULT_ codes OK? */
again:
	size = (i_size_read(inode) + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT;
	if (vmf->pgoff >= size)
		return VM_FAULT_SIGBUS;

	error = mapping->a_ops->get_xip_mem(mapping, vmf->pgoff, 0,
						&xip_mem, &xip_pfn);
	if (likely(!error))
		goto found;
	if (error != -ENODATA)
		return VM_FAULT_OOM;

	/* sparse block */
	if ((vma->vm_flags & (VM_WRITE | VM_MAYWRITE)) &&
	    (vma->vm_flags & (VM_SHARED | VM_MAYSHARE)) &&
	    (!(mapping->host->i_sb->s_flags & MS_RDONLY))) {
		int err;

		/* maybe shared writable, allocate new block */
		mutex_lock(&xip_sparse_mutex);
		error = mapping->a_ops->get_xip_mem(mapping, vmf->pgoff, 1,
							&xip_mem, &xip_pfn);
		mutex_unlock(&xip_sparse_mutex);
		if (error)
			return VM_FAULT_SIGBUS;
		/* unmap sparse mappings at pgoff from all other vmas */
		__xip_unmap(mapping, vmf->pgoff);

found:
		err = vm_insert_mixed(vma, (unsigned long)vmf->virtual_address,
							xip_pfn);
		if (err == -ENOMEM)
			return VM_FAULT_OOM;
		/*
		 * err == -EBUSY is fine, we've raced against another thread
		 * that faulted-in the same page
		 */
		if (err != -EBUSY)
			BUG_ON(err);
		return VM_FAULT_NOPAGE;
	} else {
		int err, ret = VM_FAULT_OOM;

		mutex_lock(&xip_sparse_mutex);
		write_seqcount_begin(&xip_sparse_seq);
		error = mapping->a_ops->get_xip_mem(mapping, vmf->pgoff, 0,
							&xip_mem, &xip_pfn);
		if (unlikely(!error)) {
			write_seqcount_end(&xip_sparse_seq);
			mutex_unlock(&xip_sparse_mutex);
			goto again;
		}
		if (error != -ENODATA)
			goto out;
		/* not shared and writable, use xip_sparse_page() */
		page = xip_sparse_page();
		if (!page)
			goto out;
		err = vm_insert_page(vma, (unsigned long)vmf->virtual_address,
							page);
		if (err == -ENOMEM)
			goto out;

		ret = VM_FAULT_NOPAGE;
out:
		write_seqcount_end(&xip_sparse_seq);
		mutex_unlock(&xip_sparse_mutex);

		return ret;
	}
}

static const struct vm_operations_struct xip_file_vm_ops = {
	.fault	= xip_file_fault,
	.page_mkwrite	= filemap_page_mkwrite,
	.remap_pages = generic_file_remap_pages,
};

int xip_file_mmap(struct file * file, struct vm_area_struct * vma)
{
	BUG_ON(!file->f_mapping->a_ops->get_xip_mem);

	file_accessed(file);
	vma->vm_ops = &xip_file_vm_ops;
	vma->vm_flags |= VM_MIXEDMAP;
	return 0;
}
EXPORT_SYMBOL_GPL(xip_file_mmap);

static ssize_t
__xip_file_write(struct file *filp, const char __user *buf,
		  size_t count, loff_t pos, loff_t *ppos)
{
	struct address_space * mapping = filp->f_mapping;
	const struct address_space_operations *a_ops = mapping->a_ops;
	struct inode 	*inode = mapping->host;
	long		status = 0;
	size_t		bytes;
	ssize_t		written = 0;

	BUG_ON(!mapping->a_ops->get_xip_mem);

	do {
		unsigned long index;
		unsigned long offset;
		size_t copied;
		void *xip_mem;
		unsigned long xip_pfn;

		offset = (pos & (PAGE_CACHE_SIZE -1)); /* Within page */
		index = pos >> PAGE_CACHE_SHIFT;
		bytes = PAGE_CACHE_SIZE - offset;
		if (bytes > count)
			bytes = count;

		status = a_ops->get_xip_mem(mapping, index, 0,
						&xip_mem, &xip_pfn);
		if (status == -ENODATA) {
			/* we allocate a new page unmap it */
			mutex_lock(&xip_sparse_mutex);
			status = a_ops->get_xip_mem(mapping, index, 1,
							&xip_mem, &xip_pfn);
			mutex_unlock(&xip_sparse_mutex);
			if (!status)
				/* unmap page at pgoff from all other vmas */
				__xip_unmap(mapping, index);
		}

		if (status)
			break;

		copied = bytes -
			__copy_from_user_nocache(xip_mem + offset, buf, bytes);

		if (likely(copied > 0)) {
			status = copied;

			if (status >= 0) {
				written += status;
				count -= status;
				pos += status;
				buf += status;
			}
		}
		if (unlikely(copied != bytes))
			if (status >= 0)
				status = -EFAULT;
		if (status < 0)
			break;
	} while (count);
	*ppos = pos;
	/*
	 * No need to use i_size_read() here, the i_size
	 * cannot change under us because we hold i_mutex.
	 */
	if (pos > inode->i_size) {
		i_size_write(inode, pos);
		mark_inode_dirty(inode);
	}

	return written ? written : status;
}

ssize_t
xip_file_write(struct file *filp, const char __user *buf, size_t len,
	       loff_t *ppos)
{
	struct address_space *mapping = filp->f_mapping;
	struct inode *inode = mapping->host;
	size_t count;
	loff_t pos;
	ssize_t ret;

	mutex_lock(&inode->i_mutex);

	if (!access_ok(VERIFY_READ, buf, len)) {
		ret=-EFAULT;
		goto out_up;
	}

	pos = *ppos;
	count = len;

	/* We can write back this queue in page reclaim */
	current->backing_dev_info = mapping->backing_dev_info;

	ret = generic_write_checks(filp, &pos, &count, S_ISBLK(inode->i_mode));
	if (ret)
		goto out_backing;
	if (count == 0)
		goto out_backing;

	ret = file_remove_suid(filp);
	if (ret)
		goto out_backing;

	ret = file_update_time(filp);
	if (ret)
		goto out_backing;

	ret = __xip_file_write (filp, buf, count, pos, ppos);

 out_backing:
	current->backing_dev_info = NULL;
 out_up:
	mutex_unlock(&inode->i_mutex);
	return ret;
}
EXPORT_SYMBOL_GPL(xip_file_write);

/*
 * truncate a page used for execute in place
 * functionality is analog to block_truncate_page but does use get_xip_mem
 * to get the page instead of page cache
 */
int
xip_truncate_page(struct address_space *mapping, loff_t from)
{
	pgoff_t index = from >> PAGE_CACHE_SHIFT;
	unsigned offset = from & (PAGE_CACHE_SIZE-1);
	unsigned blocksize;
	unsigned length;
	void *xip_mem;
	unsigned long xip_pfn;
	int err;

	BUG_ON(!mapping->a_ops->get_xip_mem);

	blocksize = 1 << mapping->host->i_blkbits;
	length = offset & (blocksize - 1);

	/* Block boundary? Nothing to do */
	if (!length)
		return 0;

	length = blocksize - length;

	err = mapping->a_ops->get_xip_mem(mapping, index, 0,
						&xip_mem, &xip_pfn);
	if (unlikely(err)) {
		if (err == -ENODATA)
			/* Hole? No need to truncate */
			return 0;
		else
			return err;
	}
	memset(xip_mem + offset, 0, length);
	return 0;
}
EXPORT_SYMBOL_GPL(xip_truncate_page);
Commit	Line	Data
ceffc078 CO	1	/*
	2	* linux/mm/filemap_xip.c
	3	*
	4	* Copyright (C) 2005 IBM Corporation
	5	* Author: Carsten Otte <[email protected]>
	6	*
	7	* derived from linux/mm/filemap.c - Copyright (C) Linus Torvalds
	8	*
	9	*/
	10
	11	#include <linux/fs.h>
	12	#include <linux/pagemap.h>
b95f1b31	13	#include <linux/export.h>
ceffc078 CO	14	#include <linux/uio.h>
ceffc078 CO	15	#include <linux/rmap.h>
cddb8a5c	16	#include <linux/mmu_notifier.h>
e8edc6e0	17	#include <linux/sched.h>
538f8ea6 NP	18	#include <linux/seqlock.h>
538f8ea6 NP	19	#include <linux/mutex.h>
5a0e3ad6	20	#include <linux/gfp.h>
ceffc078	21	#include <asm/tlbflush.h>
70688e4d	22	#include <asm/io.h>
ceffc078	23
a76c0b97 CO	24	/*
	25	* We do use our own empty page to avoid interference with other users
	26	* of ZERO_PAGE(), such as /dev/zero
	27	*/
538f8ea6	28	static DEFINE_MUTEX(xip_sparse_mutex);
1ca7d67c	29	static seqcount_t xip_sparse_seq = SEQCNT_ZERO(xip_sparse_seq);
a76c0b97 CO	30	static struct page *__xip_sparse_page;
a76c0b97 CO	31
538f8ea6	32	/* called under xip_sparse_mutex */
a76c0b97 CO	33	static struct page *xip_sparse_page(void)
	34	{
	35	if (!__xip_sparse_page) {
c51b1a16 AM	36	struct page *page = alloc_page(GFP_HIGHUSER \| __GFP_ZERO);
c51b1a16 AM	37
538f8ea6 NP	38	if (page)
538f8ea6 NP	39	__xip_sparse_page = page;
a76c0b97 CO	40	}
	41	return __xip_sparse_page;
	42	}
	43
ceffc078 CO	44	/*
ceffc078 CO	45	* This is a file read routine for execute in place files, and uses
70688e4d	46	* the mapping->a_ops->get_xip_mem() function for the actual low-level
ceffc078 CO	47	* stuff.
	48	*
	49	* Note the struct file* is not used at all. It may be NULL.
	50	*/
70688e4d	51	static ssize_t
ceffc078 CO	52	do_xip_mapping_read(struct address_space *mapping,
	53	struct file_ra_state *_ra,
	54	struct file *filp,
70688e4d NP	55	char __user *buf,
	56	size_t len,
	57	loff_t *ppos)
ceffc078 CO	58	{
ceffc078 CO	59	struct inode *inode = mapping->host;
2004dc8e JK	60	pgoff_t index, end_index;
2004dc8e JK	61	unsigned long offset;
70688e4d NP	62	loff_t isize, pos;
70688e4d NP	63	size_t copied = 0, error = 0;
ceffc078	64
70688e4d	65	BUG_ON(!mapping->a_ops->get_xip_mem);
ceffc078	66
70688e4d NP	67	pos = *ppos;
	68	index = pos >> PAGE_CACHE_SHIFT;
	69	offset = pos & ~PAGE_CACHE_MASK;
ceffc078 CO	70
	71	isize = i_size_read(inode);
	72	if (!isize)
	73	goto out;
	74
	75	end_index = (isize - 1) >> PAGE_CACHE_SHIFT;
70688e4d NP	76	do {
	77	unsigned long nr, left;
	78	void *xip_mem;
	79	unsigned long xip_pfn;
	80	int zero = 0;
ceffc078 CO	81
	82	/* nr is the maximum number of bytes to copy from this page */
	83	nr = PAGE_CACHE_SIZE;
	84	if (index >= end_index) {
	85	if (index > end_index)
	86	goto out;
	87	nr = ((isize - 1) & ~PAGE_CACHE_MASK) + 1;
	88	if (nr <= offset) {
	89	goto out;
	90	}
	91	}
	92	nr = nr - offset;
58984ce2 MS	93	if (nr > len - copied)
58984ce2 MS	94	nr = len - copied;
ceffc078	95
70688e4d NP	96	error = mapping->a_ops->get_xip_mem(mapping, index, 0,
	97	&xip_mem, &xip_pfn);
	98	if (unlikely(error)) {
	99	if (error == -ENODATA) {
ceffc078	100	/* sparse */
70688e4d NP	101	zero = 1;
70688e4d NP	102	} else
ceffc078	103	goto out;
afa597ba	104	}
ceffc078 CO	105
	106	/* If users can be writing to this page using arbitrary
	107	* virtual addresses, take care about potential aliasing
	108	* before reading the page on the kernel side.
	109	*/
	110	if (mapping_writably_mapped(mapping))
70688e4d	111	/* address based flush */ ;
ceffc078 CO	112
ceffc078 CO	113	/*
70688e4d	114	* Ok, we have the mem, so now we can copy it to user space...
ceffc078 CO	115	*
	116	* The actor routine returns how many bytes were actually used..
	117	* NOTE! This may not be the same as how much of a user buffer
	118	* we filled up (we may be padding etc), so we can only update
	119	* "pos" here (the actor routine has to update the user buffer
	120	* pointers and the remaining count).
	121	*/
70688e4d NP	122	if (!zero)
	123	left = __copy_to_user(buf+copied, xip_mem+offset, nr);
	124	else
	125	left = __clear_user(buf + copied, nr);
ceffc078	126
70688e4d NP	127	if (left) {
	128	error = -EFAULT;
	129	goto out;
	130	}
ceffc078	131
70688e4d NP	132	copied += (nr - left);
	133	offset += (nr - left);
	134	index += offset >> PAGE_CACHE_SHIFT;
	135	offset &= ~PAGE_CACHE_MASK;
	136	} while (copied < len);
ceffc078 CO	137
ceffc078 CO	138	out:
70688e4d	139	*ppos = pos + copied;
ceffc078 CO	140	if (filp)
ceffc078 CO	141	file_accessed(filp);
70688e4d NP	142
70688e4d NP	143	return (copied ? copied : error);
ceffc078 CO	144	}
ceffc078 CO	145
ceffc078	146	ssize_t
eb6fe0c3	147	xip_file_read(struct file filp, char __user buf, size_t len, loff_t *ppos)
ceffc078	148	{
eb6fe0c3 CO	149	if (!access_ok(VERIFY_WRITE, buf, len))
eb6fe0c3 CO	150	return -EFAULT;
ceffc078	151
70688e4d NP	152	return do_xip_mapping_read(filp->f_mapping, &filp->f_ra, filp,
70688e4d NP	153	buf, len, ppos);
ceffc078	154	}
eb6fe0c3	155	EXPORT_SYMBOL_GPL(xip_file_read);
ceffc078	156
ceffc078 CO	157	/*
	158	* __xip_unmap is invoked from xip_unmap and
	159	* xip_write
	160	*
	161	* This function walks all vmas of the address_space and unmaps the
a76c0b97	162	* __xip_sparse_page when found at pgoff.
ceffc078 CO	163	*/
	164	static void
	165	__xip_unmap (struct address_space * mapping,
	166	unsigned long pgoff)
	167	{
	168	struct vm_area_struct *vma;
	169	struct mm_struct *mm;
ceffc078 CO	170	unsigned long address;
	171	pte_t *pte;
	172	pte_t pteval;
c0718806	173	spinlock_t *ptl;
67b02f11	174	struct page *page;
538f8ea6 NP	175	unsigned count;
	176	int locked = 0;
	177
	178	count = read_seqcount_begin(&xip_sparse_seq);
ceffc078	179
a76c0b97 CO	180	page = __xip_sparse_page;
	181	if (!page)
	182	return;
	183
538f8ea6	184	retry:
3d48ae45	185	mutex_lock(&mapping->i_mmap_mutex);
6b2dbba8	186	vma_interval_tree_foreach(vma, &mapping->i_mmap, pgoff, pgoff) {
ceffc078 CO	187	mm = vma->vm_mm;
	188	address = vma->vm_start +
	189	((pgoff - vma->vm_pgoff) << PAGE_SHIFT);
	190	BUG_ON(address < vma->vm_start \|\| address >= vma->vm_end);
479db0bf	191	pte = page_check_address(page, mm, address, &ptl, 1);
c0718806	192	if (pte) {
ceffc078	193	/* Nuke the page table entry. */
082ff0a9	194	flush_cache_page(vma, address, pte_pfn(*pte));
2ec74c3e	195	pteval = ptep_clear_flush(vma, address, pte);
edc315fd	196	page_remove_rmap(page);
d559db08	197	dec_mm_counter(mm, MM_FILEPAGES);
ceffc078	198	BUG_ON(pte_dirty(pteval));
c0718806	199	pte_unmap_unlock(pte, ptl);
2ec74c3e SG	200	/* must invalidate_page _before_ freeing the page */
2ec74c3e SG	201	mmu_notifier_invalidate_page(mm, address);
b5810039	202	page_cache_release(page);
ceffc078 CO	203	}
ceffc078 CO	204	}
3d48ae45	205	mutex_unlock(&mapping->i_mmap_mutex);
538f8ea6 NP	206
	207	if (locked) {
	208	mutex_unlock(&xip_sparse_mutex);
	209	} else if (read_seqcount_retry(&xip_sparse_seq, count)) {
	210	mutex_lock(&xip_sparse_mutex);
	211	locked = 1;
	212	goto retry;
	213	}
ceffc078 CO	214	}
	215
	216	/*
54cb8821	217	* xip_fault() is invoked via the vma operations vector for a
ceffc078 CO	218	* mapped memory region to read in file data during a page fault.
ceffc078 CO	219	*
54cb8821	220	* This function is derived from filemap_fault, but used for execute in place
ceffc078	221	*/
70688e4d	222	static int xip_file_fault(struct vm_area_struct vma, struct vm_fault vmf)
ceffc078	223	{
70688e4d	224	struct file *file = vma->vm_file;
ceffc078 CO	225	struct address_space *mapping = file->f_mapping;
ceffc078 CO	226	struct inode *inode = mapping->host;
54cb8821	227	pgoff_t size;
70688e4d NP	228	void *xip_mem;
	229	unsigned long xip_pfn;
	230	struct page *page;
	231	int error;
ceffc078	232
54cb8821	233	/* XXX: are VM_FAULT_ codes OK? */
538f8ea6	234	again:
ceffc078	235	size = (i_size_read(inode) + PAGE_CACHE_SIZE - 1) >> PAGE_CACHE_SHIFT;
d0217ac0 NP	236	if (vmf->pgoff >= size)
d0217ac0 NP	237	return VM_FAULT_SIGBUS;
ceffc078	238
70688e4d NP	239	error = mapping->a_ops->get_xip_mem(mapping, vmf->pgoff, 0,
	240	&xip_mem, &xip_pfn);
	241	if (likely(!error))
	242	goto found;
	243	if (error != -ENODATA)
d0217ac0	244	return VM_FAULT_OOM;
ceffc078 CO	245
ceffc078 CO	246	/* sparse block */
70688e4d NP	247	if ((vma->vm_flags & (VM_WRITE \| VM_MAYWRITE)) &&
70688e4d NP	248	(vma->vm_flags & (VM_SHARED \| VM_MAYSHARE)) &&
ceffc078	249	(!(mapping->host->i_sb->s_flags & MS_RDONLY))) {
70688e4d NP	250	int err;
70688e4d NP	251
ceffc078	252	/* maybe shared writable, allocate new block */
14bac5ac	253	mutex_lock(&xip_sparse_mutex);
70688e4d NP	254	error = mapping->a_ops->get_xip_mem(mapping, vmf->pgoff, 1,
70688e4d NP	255	&xip_mem, &xip_pfn);
14bac5ac	256	mutex_unlock(&xip_sparse_mutex);
70688e4d	257	if (error)
d0217ac0	258	return VM_FAULT_SIGBUS;
70688e4d	259	/* unmap sparse mappings at pgoff from all other vmas */
d0217ac0	260	__xip_unmap(mapping, vmf->pgoff);
70688e4d NP	261
	262	found:
	263	err = vm_insert_mixed(vma, (unsigned long)vmf->virtual_address,
	264	xip_pfn);
	265	if (err == -ENOMEM)
	266	return VM_FAULT_OOM;
99f02ef1 CO	267	/*
	268	* err == -EBUSY is fine, we've raced against another thread
	269	* that faulted-in the same page
	270	*/
	271	if (err != -EBUSY)
	272	BUG_ON(err);
70688e4d	273	return VM_FAULT_NOPAGE;
ceffc078	274	} else {
538f8ea6 NP	275	int err, ret = VM_FAULT_OOM;
	276
	277	mutex_lock(&xip_sparse_mutex);
	278	write_seqcount_begin(&xip_sparse_seq);
	279	error = mapping->a_ops->get_xip_mem(mapping, vmf->pgoff, 0,
	280	&xip_mem, &xip_pfn);
	281	if (unlikely(!error)) {
	282	write_seqcount_end(&xip_sparse_seq);
	283	mutex_unlock(&xip_sparse_mutex);
	284	goto again;
	285	}
	286	if (error != -ENODATA)
	287	goto out;
a76c0b97 CO	288	/* not shared and writable, use xip_sparse_page() */
a76c0b97 CO	289	page = xip_sparse_page();
d0217ac0	290	if (!page)
538f8ea6 NP	291	goto out;
	292	err = vm_insert_page(vma, (unsigned long)vmf->virtual_address,
	293	page);
	294	if (err == -ENOMEM)
	295	goto out;
	296
	297	ret = VM_FAULT_NOPAGE;
	298	out:
	299	write_seqcount_end(&xip_sparse_seq);
	300	mutex_unlock(&xip_sparse_mutex);
ceffc078	301
538f8ea6	302	return ret;
70688e4d	303	}
ceffc078 CO	304	}
ceffc078 CO	305
f0f37e2f	306	static const struct vm_operations_struct xip_file_vm_ops = {
54cb8821	307	.fault = xip_file_fault,
4fcf1c62	308	.page_mkwrite = filemap_page_mkwrite,
0b173bc4	309	.remap_pages = generic_file_remap_pages,
ceffc078 CO	310	};
	311
	312	int xip_file_mmap(struct file * file, struct vm_area_struct * vma)
	313	{
70688e4d	314	BUG_ON(!file->f_mapping->a_ops->get_xip_mem);
ceffc078 CO	315
	316	file_accessed(file);
	317	vma->vm_ops = &xip_file_vm_ops;
0b173bc4	318	vma->vm_flags \|= VM_MIXEDMAP;
ceffc078 CO	319	return 0;
	320	}
	321	EXPORT_SYMBOL_GPL(xip_file_mmap);
	322
	323	static ssize_t
eb6fe0c3 CO	324	__xip_file_write(struct file filp, const char __user buf,
eb6fe0c3 CO	325	size_t count, loff_t pos, loff_t *ppos)
ceffc078	326	{
eb6fe0c3	327	struct address_space * mapping = filp->f_mapping;
f5e54d6e	328	const struct address_space_operations *a_ops = mapping->a_ops;
ceffc078 CO	329	struct inode *inode = mapping->host;
ceffc078 CO	330	long status = 0;
ceffc078	331	size_t bytes;
ceffc078 CO	332	ssize_t written = 0;
ceffc078 CO	333
70688e4d	334	BUG_ON(!mapping->a_ops->get_xip_mem);
ceffc078	335
ceffc078 CO	336	do {
	337	unsigned long index;
	338	unsigned long offset;
	339	size_t copied;
70688e4d NP	340	void *xip_mem;
70688e4d NP	341	unsigned long xip_pfn;
ceffc078 CO	342
	343	offset = (pos & (PAGE_CACHE_SIZE -1)); /* Within page */
	344	index = pos >> PAGE_CACHE_SHIFT;
	345	bytes = PAGE_CACHE_SIZE - offset;
	346	if (bytes > count)
	347	bytes = count;
	348
70688e4d NP	349	status = a_ops->get_xip_mem(mapping, index, 0,
	350	&xip_mem, &xip_pfn);
	351	if (status == -ENODATA) {
ceffc078	352	/* we allocate a new page unmap it */
14bac5ac	353	mutex_lock(&xip_sparse_mutex);
70688e4d NP	354	status = a_ops->get_xip_mem(mapping, index, 1,
70688e4d NP	355	&xip_mem, &xip_pfn);
14bac5ac	356	mutex_unlock(&xip_sparse_mutex);
70688e4d	357	if (!status)
eb6fe0c3 CO	358	/* unmap page at pgoff from all other vmas */
eb6fe0c3 CO	359	__xip_unmap(mapping, index);
ceffc078 CO	360	}
ceffc078 CO	361
70688e4d	362	if (status)
ceffc078	363	break;
ceffc078	364
4a9e5ef1	365	copied = bytes -
70688e4d	366	__copy_from_user_nocache(xip_mem + offset, buf, bytes);
4a9e5ef1	367
ceffc078 CO	368	if (likely(copied > 0)) {
	369	status = copied;
	370
	371	if (status >= 0) {
	372	written += status;
	373	count -= status;
	374	pos += status;
	375	buf += status;
ceffc078 CO	376	}
	377	}
	378	if (unlikely(copied != bytes))
	379	if (status >= 0)
	380	status = -EFAULT;
	381	if (status < 0)
	382	break;
	383	} while (count);
	384	*ppos = pos;
	385	/*
	386	* No need to use i_size_read() here, the i_size
1b1dcc1b	387	* cannot change under us because we hold i_mutex.
ceffc078 CO	388	*/
	389	if (pos > inode->i_size) {
	390	i_size_write(inode, pos);
	391	mark_inode_dirty(inode);
	392	}
	393
	394	return written ? written : status;
	395	}
	396
eb6fe0c3 CO	397	ssize_t
	398	xip_file_write(struct file filp, const char __user buf, size_t len,
	399	loff_t *ppos)
ceffc078	400	{
eb6fe0c3 CO	401	struct address_space *mapping = filp->f_mapping;
	402	struct inode *inode = mapping->host;
	403	size_t count;
	404	loff_t pos;
	405	ssize_t ret;
ceffc078	406
1b1dcc1b	407	mutex_lock(&inode->i_mutex);
ceffc078	408
eb6fe0c3 CO	409	if (!access_ok(VERIFY_READ, buf, len)) {
	410	ret=-EFAULT;
	411	goto out_up;
ceffc078 CO	412	}
ceffc078 CO	413
ceffc078	414	pos = *ppos;
eb6fe0c3	415	count = len;
ceffc078	416
eb6fe0c3 CO	417	/* We can write back this queue in page reclaim */
eb6fe0c3 CO	418	current->backing_dev_info = mapping->backing_dev_info;
ceffc078	419
eb6fe0c3 CO	420	ret = generic_write_checks(filp, &pos, &count, S_ISBLK(inode->i_mode));
	421	if (ret)
	422	goto out_backing;
ceffc078	423	if (count == 0)
eb6fe0c3	424	goto out_backing;
ceffc078	425
2f1936b8	426	ret = file_remove_suid(filp);
eb6fe0c3 CO	427	if (ret)
eb6fe0c3 CO	428	goto out_backing;
ceffc078	429
c3b2da31 JB	430	ret = file_update_time(filp);
	431	if (ret)
	432	goto out_backing;
ceffc078	433
eb6fe0c3	434	ret = __xip_file_write (filp, buf, count, pos, ppos);
ceffc078	435
eb6fe0c3 CO	436	out_backing:
	437	current->backing_dev_info = NULL;
	438	out_up:
1b1dcc1b	439	mutex_unlock(&inode->i_mutex);
ceffc078 CO	440	return ret;
ceffc078 CO	441	}
eb6fe0c3	442	EXPORT_SYMBOL_GPL(xip_file_write);
ceffc078 CO	443
	444	/*
	445	* truncate a page used for execute in place
70688e4d	446	* functionality is analog to block_truncate_page but does use get_xip_mem
ceffc078 CO	447	* to get the page instead of page cache
	448	*/
	449	int
	450	xip_truncate_page(struct address_space *mapping, loff_t from)
	451	{
	452	pgoff_t index = from >> PAGE_CACHE_SHIFT;
	453	unsigned offset = from & (PAGE_CACHE_SIZE-1);
	454	unsigned blocksize;
	455	unsigned length;
70688e4d NP	456	void *xip_mem;
	457	unsigned long xip_pfn;
	458	int err;
ceffc078	459
70688e4d	460	BUG_ON(!mapping->a_ops->get_xip_mem);
ceffc078 CO	461
	462	blocksize = 1 << mapping->host->i_blkbits;
	463	length = offset & (blocksize - 1);
	464
	465	/* Block boundary? Nothing to do */
	466	if (!length)
	467	return 0;
	468
	469	length = blocksize - length;
	470
70688e4d NP	471	err = mapping->a_ops->get_xip_mem(mapping, index, 0,
	472	&xip_mem, &xip_pfn);
	473	if (unlikely(err)) {
	474	if (err == -ENODATA)
ceffc078 CO	475	/* Hole? No need to truncate */
ceffc078 CO	476	return 0;
eb6fe0c3	477	else
70688e4d	478	return err;
afa597ba	479	}
70688e4d	480	memset(xip_mem + offset, 0, length);
eb6fe0c3	481	return 0;
ceffc078 CO	482	}
ceffc078 CO	483	EXPORT_SYMBOL_GPL(xip_truncate_page);