xref: /kernel/linux/linux-5.10/fs/jfs/inode.c (revision 8c2ecf20) 
1// SPDX-License-Identifier: GPL-2.0-or-later
2/*
3 *   Copyright (C) International Business Machines Corp., 2000-2004
4 *   Portions Copyright (C) Christoph Hellwig, 2001-2002
5 */
6
7#include <linux/fs.h>
8#include <linux/mpage.h>
9#include <linux/buffer_head.h>
10#include <linux/pagemap.h>
11#include <linux/quotaops.h>
12#include <linux/uio.h>
13#include <linux/writeback.h>
14#include "jfs_incore.h"
15#include "jfs_inode.h"
16#include "jfs_filsys.h"
17#include "jfs_imap.h"
18#include "jfs_extent.h"
19#include "jfs_unicode.h"
20#include "jfs_debug.h"
21#include "jfs_dmap.h"
22
23
24struct inode *jfs_iget(struct super_block *sb, unsigned long ino)
25{
26	struct inode *inode;
27	int ret;
28
29	inode = iget_locked(sb, ino);
30	if (!inode)
31		return ERR_PTR(-ENOMEM);
32	if (!(inode->i_state & I_NEW))
33		return inode;
34
35	ret = diRead(inode);
36	if (ret < 0) {
37		iget_failed(inode);
38		return ERR_PTR(ret);
39	}
40
41	if (S_ISREG(inode->i_mode)) {
42		inode->i_op = &jfs_file_inode_operations;
43		inode->i_fop = &jfs_file_operations;
44		inode->i_mapping->a_ops = &jfs_aops;
45	} else if (S_ISDIR(inode->i_mode)) {
46		inode->i_op = &jfs_dir_inode_operations;
47		inode->i_fop = &jfs_dir_operations;
48	} else if (S_ISLNK(inode->i_mode)) {
49		if (inode->i_size >= IDATASIZE) {
50			inode->i_op = &page_symlink_inode_operations;
51			inode_nohighmem(inode);
52			inode->i_mapping->a_ops = &jfs_aops;
53		} else {
54			inode->i_op = &jfs_fast_symlink_inode_operations;
55			inode->i_link = JFS_IP(inode)->i_inline;
56			/*
57			 * The inline data should be null-terminated, but
58			 * don't let on-disk corruption crash the kernel
59			 */
60			inode->i_link[inode->i_size] = '\0';
61		}
62	} else {
63		inode->i_op = &jfs_file_inode_operations;
64		init_special_inode(inode, inode->i_mode, inode->i_rdev);
65	}
66	unlock_new_inode(inode);
67	return inode;
68}
69
70/*
71 * Workhorse of both fsync & write_inode
72 */
73int jfs_commit_inode(struct inode *inode, int wait)
74{
75	int rc = 0;
76	tid_t tid;
77	static int noisy = 5;
78
79	jfs_info("In jfs_commit_inode, inode = 0x%p", inode);
80
81	/*
82	 * Don't commit if inode has been committed since last being
83	 * marked dirty, or if it has been deleted.
84	 */
85	if (inode->i_nlink == 0 || !test_cflag(COMMIT_Dirty, inode))
86		return 0;
87
88	if (isReadOnly(inode)) {
89		/* kernel allows writes to devices on read-only
90		 * partitions and may think inode is dirty
91		 */
92		if (!special_file(inode->i_mode) && noisy) {
93			jfs_err("jfs_commit_inode(0x%p) called on read-only volume",
94				inode);
95			jfs_err("Is remount racy?");
96			noisy--;
97		}
98		return 0;
99	}
100
101	tid = txBegin(inode->i_sb, COMMIT_INODE);
102	mutex_lock(&JFS_IP(inode)->commit_mutex);
103
104	/*
105	 * Retest inode state after taking commit_mutex
106	 */
107	if (inode->i_nlink && test_cflag(COMMIT_Dirty, inode))
108		rc = txCommit(tid, 1, &inode, wait ? COMMIT_SYNC : 0);
109
110	txEnd(tid);
111	mutex_unlock(&JFS_IP(inode)->commit_mutex);
112	return rc;
113}
114
115int jfs_write_inode(struct inode *inode, struct writeback_control *wbc)
116{
117	int wait = wbc->sync_mode == WB_SYNC_ALL;
118
119	if (inode->i_nlink == 0)
120		return 0;
121	/*
122	 * If COMMIT_DIRTY is not set, the inode isn't really dirty.
123	 * It has been committed since the last change, but was still
124	 * on the dirty inode list.
125	 */
126	if (!test_cflag(COMMIT_Dirty, inode)) {
127		/* Make sure committed changes hit the disk */
128		jfs_flush_journal(JFS_SBI(inode->i_sb)->log, wait);
129		return 0;
130	}
131
132	if (jfs_commit_inode(inode, wait)) {
133		jfs_err("jfs_write_inode: jfs_commit_inode failed!");
134		return -EIO;
135	} else
136		return 0;
137}
138
139void jfs_evict_inode(struct inode *inode)
140{
141	struct jfs_inode_info *ji = JFS_IP(inode);
142
143	jfs_info("In jfs_evict_inode, inode = 0x%p", inode);
144
145	if (!inode->i_nlink && !is_bad_inode(inode)) {
146		dquot_initialize(inode);
147
148		if (JFS_IP(inode)->fileset == FILESYSTEM_I) {
149			struct inode *ipimap = JFS_SBI(inode->i_sb)->ipimap;
150			truncate_inode_pages_final(&inode->i_data);
151
152			if (test_cflag(COMMIT_Freewmap, inode))
153				jfs_free_zero_link(inode);
154
155			if (ipimap && JFS_IP(ipimap)->i_imap)
156				diFree(inode);
157
158			/*
159			 * Free the inode from the quota allocation.
160			 */
161			dquot_free_inode(inode);
162		}
163	} else {
164		truncate_inode_pages_final(&inode->i_data);
165	}
166	clear_inode(inode);
167	dquot_drop(inode);
168
169	BUG_ON(!list_empty(&ji->anon_inode_list));
170
171	spin_lock_irq(&ji->ag_lock);
172	if (ji->active_ag != -1) {
173		struct bmap *bmap = JFS_SBI(inode->i_sb)->bmap;
174		atomic_dec(&bmap->db_active[ji->active_ag]);
175		ji->active_ag = -1;
176	}
177	spin_unlock_irq(&ji->ag_lock);
178}
179
180void jfs_dirty_inode(struct inode *inode, int flags)
181{
182	static int noisy = 5;
183
184	if (isReadOnly(inode)) {
185		if (!special_file(inode->i_mode) && noisy) {
186			/* kernel allows writes to devices on read-only
187			 * partitions and may try to mark inode dirty
188			 */
189			jfs_err("jfs_dirty_inode called on read-only volume");
190			jfs_err("Is remount racy?");
191			noisy--;
192		}
193		return;
194	}
195
196	set_cflag(COMMIT_Dirty, inode);
197}
198
199int jfs_get_block(struct inode *ip, sector_t lblock,
200		  struct buffer_head *bh_result, int create)
201{
202	s64 lblock64 = lblock;
203	int rc = 0;
204	xad_t xad;
205	s64 xaddr;
206	int xflag;
207	s32 xlen = bh_result->b_size >> ip->i_blkbits;
208
209	/*
210	 * Take appropriate lock on inode
211	 */
212	if (create)
213		IWRITE_LOCK(ip, RDWRLOCK_NORMAL);
214	else
215		IREAD_LOCK(ip, RDWRLOCK_NORMAL);
216
217	if (((lblock64 << ip->i_sb->s_blocksize_bits) < ip->i_size) &&
218	    (!xtLookup(ip, lblock64, xlen, &xflag, &xaddr, &xlen, 0)) &&
219	    xaddr) {
220		if (xflag & XAD_NOTRECORDED) {
221			if (!create)
222				/*
223				 * Allocated but not recorded, read treats
224				 * this as a hole
225				 */
226				goto unlock;
227#ifdef _JFS_4K
228			XADoffset(&xad, lblock64);
229			XADlength(&xad, xlen);
230			XADaddress(&xad, xaddr);
231#else				/* _JFS_4K */
232			/*
233			 * As long as block size = 4K, this isn't a problem.
234			 * We should mark the whole page not ABNR, but how
235			 * will we know to mark the other blocks BH_New?
236			 */
237			BUG();
238#endif				/* _JFS_4K */
239			rc = extRecord(ip, &xad);
240			if (rc)
241				goto unlock;
242			set_buffer_new(bh_result);
243		}
244
245		map_bh(bh_result, ip->i_sb, xaddr);
246		bh_result->b_size = xlen << ip->i_blkbits;
247		goto unlock;
248	}
249	if (!create)
250		goto unlock;
251
252	/*
253	 * Allocate a new block
254	 */
255#ifdef _JFS_4K
256	if ((rc = extHint(ip, lblock64 << ip->i_sb->s_blocksize_bits, &xad)))
257		goto unlock;
258	rc = extAlloc(ip, xlen, lblock64, &xad, false);
259	if (rc)
260		goto unlock;
261
262	set_buffer_new(bh_result);
263	map_bh(bh_result, ip->i_sb, addressXAD(&xad));
264	bh_result->b_size = lengthXAD(&xad) << ip->i_blkbits;
265
266#else				/* _JFS_4K */
267	/*
268	 * We need to do whatever it takes to keep all but the last buffers
269	 * in 4K pages - see jfs_write.c
270	 */
271	BUG();
272#endif				/* _JFS_4K */
273
274      unlock:
275	/*
276	 * Release lock on inode
277	 */
278	if (create)
279		IWRITE_UNLOCK(ip);
280	else
281		IREAD_UNLOCK(ip);
282	return rc;
283}
284
285static int jfs_writepage(struct page *page, struct writeback_control *wbc)
286{
287	return block_write_full_page(page, jfs_get_block, wbc);
288}
289
290static int jfs_writepages(struct address_space *mapping,
291			struct writeback_control *wbc)
292{
293	return mpage_writepages(mapping, wbc, jfs_get_block);
294}
295
296static int jfs_readpage(struct file *file, struct page *page)
297{
298	return mpage_readpage(page, jfs_get_block);
299}
300
301static void jfs_readahead(struct readahead_control *rac)
302{
303	mpage_readahead(rac, jfs_get_block);
304}
305
306static void jfs_write_failed(struct address_space *mapping, loff_t to)
307{
308	struct inode *inode = mapping->host;
309
310	if (to > inode->i_size) {
311		truncate_pagecache(inode, inode->i_size);
312		jfs_truncate(inode);
313	}
314}
315
316static int jfs_write_begin(struct file *file, struct address_space *mapping,
317				loff_t pos, unsigned len, unsigned flags,
318				struct page **pagep, void **fsdata)
319{
320	int ret;
321
322	ret = nobh_write_begin(mapping, pos, len, flags, pagep, fsdata,
323				jfs_get_block);
324	if (unlikely(ret))
325		jfs_write_failed(mapping, pos + len);
326
327	return ret;
328}
329
330static sector_t jfs_bmap(struct address_space *mapping, sector_t block)
331{
332	return generic_block_bmap(mapping, block, jfs_get_block);
333}
334
335static ssize_t jfs_direct_IO(struct kiocb *iocb, struct iov_iter *iter)
336{
337	struct file *file = iocb->ki_filp;
338	struct address_space *mapping = file->f_mapping;
339	struct inode *inode = file->f_mapping->host;
340	size_t count = iov_iter_count(iter);
341	ssize_t ret;
342
343	ret = blockdev_direct_IO(iocb, inode, iter, jfs_get_block);
344
345	/*
346	 * In case of error extending write may have instantiated a few
347	 * blocks outside i_size. Trim these off again.
348	 */
349	if (unlikely(iov_iter_rw(iter) == WRITE && ret < 0)) {
350		loff_t isize = i_size_read(inode);
351		loff_t end = iocb->ki_pos + count;
352
353		if (end > isize)
354			jfs_write_failed(mapping, end);
355	}
356
357	return ret;
358}
359
360const struct address_space_operations jfs_aops = {
361	.readpage	= jfs_readpage,
362	.readahead	= jfs_readahead,
363	.writepage	= jfs_writepage,
364	.writepages	= jfs_writepages,
365	.write_begin	= jfs_write_begin,
366	.write_end	= nobh_write_end,
367	.bmap		= jfs_bmap,
368	.direct_IO	= jfs_direct_IO,
369};
370
371/*
372 * Guts of jfs_truncate.  Called with locks already held.  Can be called
373 * with directory for truncating directory index table.
374 */
375void jfs_truncate_nolock(struct inode *ip, loff_t length)
376{
377	loff_t newsize;
378	tid_t tid;
379
380	ASSERT(length >= 0);
381
382	if (test_cflag(COMMIT_Nolink, ip)) {
383		xtTruncate(0, ip, length, COMMIT_WMAP);
384		return;
385	}
386
387	do {
388		tid = txBegin(ip->i_sb, 0);
389
390		/*
391		 * The commit_mutex cannot be taken before txBegin.
392		 * txBegin may block and there is a chance the inode
393		 * could be marked dirty and need to be committed
394		 * before txBegin unblocks
395		 */
396		mutex_lock(&JFS_IP(ip)->commit_mutex);
397
398		newsize = xtTruncate(tid, ip, length,
399				     COMMIT_TRUNCATE | COMMIT_PWMAP);
400		if (newsize < 0) {
401			txEnd(tid);
402			mutex_unlock(&JFS_IP(ip)->commit_mutex);
403			break;
404		}
405
406		ip->i_mtime = ip->i_ctime = current_time(ip);
407		mark_inode_dirty(ip);
408
409		txCommit(tid, 1, &ip, 0);
410		txEnd(tid);
411		mutex_unlock(&JFS_IP(ip)->commit_mutex);
412	} while (newsize > length);	/* Truncate isn't always atomic */
413}
414
415void jfs_truncate(struct inode *ip)
416{
417	jfs_info("jfs_truncate: size = 0x%lx", (ulong) ip->i_size);
418
419	nobh_truncate_page(ip->i_mapping, ip->i_size, jfs_get_block);
420
421	IWRITE_LOCK(ip, RDWRLOCK_NORMAL);
422	jfs_truncate_nolock(ip, ip->i_size);
423	IWRITE_UNLOCK(ip);
424}
425