1 // SPDX-License-Identifier: GPL-2.0-or-later
5 * Buffer cache handling
7 * Copyright (C) 2002, 2004 Oracle. All rights reserved.
11 #include <linux/types.h>
12 #include <linux/highmem.h>
13 #include <linux/bio.h>
15 #include <cluster/masklog.h>
23 #include "buffer_head_io.h"
24 #include "ocfs2_trace.h"
27 * Bits on bh->b_state used by ocfs2.
29 * These MUST be after the JBD2 bits. Hence, we use BH_JBDPrivateStart.
31 enum ocfs2_state_bits
{
32 BH_NeedsValidate
= BH_JBDPrivateStart
,
35 /* Expand the magic b_state functions */
36 BUFFER_FNS(NeedsValidate
, needs_validate
);
38 int ocfs2_write_block(struct ocfs2_super
*osb
, struct buffer_head
*bh
,
39 struct ocfs2_caching_info
*ci
)
43 trace_ocfs2_write_block((unsigned long long)bh
->b_blocknr
, ci
);
45 BUG_ON(bh
->b_blocknr
< OCFS2_SUPER_BLOCK_BLKNO
);
46 BUG_ON(buffer_jbd(bh
));
48 /* No need to check for a soft readonly file system here. non
49 * journalled writes are only ever done on system files which
50 * can get modified during recovery even if read-only. */
51 if (ocfs2_is_hard_readonly(osb
)) {
57 ocfs2_metadata_cache_io_lock(ci
);
60 set_buffer_uptodate(bh
);
62 /* remove from dirty list before I/O. */
63 clear_buffer_dirty(bh
);
65 get_bh(bh
); /* for end_buffer_write_sync() */
66 bh
->b_end_io
= end_buffer_write_sync
;
67 submit_bh(REQ_OP_WRITE
, bh
);
71 if (buffer_uptodate(bh
)) {
72 ocfs2_set_buffer_uptodate(ci
, bh
);
74 /* We don't need to remove the clustered uptodate
75 * information for this bh as it's not marked locally
81 ocfs2_metadata_cache_io_unlock(ci
);
86 /* Caller must provide a bhs[] with all NULL or non-NULL entries, so it
87 * will be easier to handle read failure.
89 int ocfs2_read_blocks_sync(struct ocfs2_super
*osb
, u64 block
,
90 unsigned int nr
, struct buffer_head
*bhs
[])
94 struct buffer_head
*bh
;
97 trace_ocfs2_read_blocks_sync((unsigned long long)block
, nr
);
102 /* Don't put buffer head and re-assign it to NULL if it is allocated
103 * outside since the caller can't be aware of this alternation!
105 new_bh
= (bhs
[0] == NULL
);
107 for (i
= 0 ; i
< nr
; i
++) {
108 if (bhs
[i
] == NULL
) {
109 bhs
[i
] = sb_getblk(osb
->sb
, block
++);
110 if (bhs
[i
] == NULL
) {
118 if (buffer_jbd(bh
)) {
119 trace_ocfs2_read_blocks_sync_jbd(
120 (unsigned long long)bh
->b_blocknr
);
124 if (buffer_dirty(bh
)) {
125 /* This should probably be a BUG, or
126 * at least return an error. */
128 "trying to sync read a dirty "
129 "buffer! (blocknr = %llu), skipping\n",
130 (unsigned long long)bh
->b_blocknr
);
135 if (buffer_jbd(bh
)) {
136 #ifdef CATCH_BH_JBD_RACES
138 "block %llu had the JBD bit set "
139 "while I was in lock_buffer!",
140 (unsigned long long)bh
->b_blocknr
);
148 get_bh(bh
); /* for end_buffer_read_sync() */
149 bh
->b_end_io
= end_buffer_read_sync
;
150 submit_bh(REQ_OP_READ
, bh
);
154 for (i
= nr
; i
> 0; i
--) {
157 if (unlikely(status
)) {
159 /* If middle bh fails, let previous bh
160 * finish its read and then put it to
167 } else if (bh
&& buffer_uptodate(bh
)) {
168 clear_buffer_uptodate(bh
);
173 /* No need to wait on the buffer if it's managed by JBD. */
177 if (!buffer_uptodate(bh
)) {
178 /* Status won't be cleared from here on out,
179 * so we can safely record this and loop back
180 * to cleanup the other buffers. */
190 /* Caller must provide a bhs[] with all NULL or non-NULL entries, so it
191 * will be easier to handle read failure.
193 int ocfs2_read_blocks(struct ocfs2_caching_info
*ci
, u64 block
, int nr
,
194 struct buffer_head
*bhs
[], int flags
,
195 int (*validate
)(struct super_block
*sb
,
196 struct buffer_head
*bh
))
199 int i
, ignore_cache
= 0;
200 struct buffer_head
*bh
;
201 struct super_block
*sb
= ocfs2_metadata_cache_get_super(ci
);
204 trace_ocfs2_read_blocks_begin(ci
, (unsigned long long)block
, nr
, flags
);
207 BUG_ON((flags
& OCFS2_BH_READAHEAD
) &&
208 (flags
& OCFS2_BH_IGNORE_CACHE
));
217 mlog(ML_ERROR
, "asked to read %d blocks!\n", nr
);
228 /* Don't put buffer head and re-assign it to NULL if it is allocated
229 * outside since the caller can't be aware of this alternation!
231 new_bh
= (bhs
[0] == NULL
);
233 ocfs2_metadata_cache_io_lock(ci
);
234 for (i
= 0 ; i
< nr
; i
++) {
235 if (bhs
[i
] == NULL
) {
236 bhs
[i
] = sb_getblk(sb
, block
++);
237 if (bhs
[i
] == NULL
) {
240 /* Don't forget to put previous bh! */
245 ignore_cache
= (flags
& OCFS2_BH_IGNORE_CACHE
);
247 /* There are three read-ahead cases here which we need to
248 * be concerned with. All three assume a buffer has
249 * previously been submitted with OCFS2_BH_READAHEAD
250 * and it hasn't yet completed I/O.
252 * 1) The current request is sync to disk. This rarely
253 * happens these days, and never when performance
254 * matters - the code can just wait on the buffer
255 * lock and re-submit.
257 * 2) The current request is cached, but not
258 * readahead. ocfs2_buffer_uptodate() will return
259 * false anyway, so we'll wind up waiting on the
260 * buffer lock to do I/O. We re-check the request
261 * with after getting the lock to avoid a re-submit.
263 * 3) The current request is readahead (and so must
264 * also be a caching one). We short circuit if the
265 * buffer is locked (under I/O) and if it's in the
266 * uptodate cache. The re-check from #2 catches the
267 * case that the previous read-ahead completes just
268 * before our is-it-in-flight check.
271 if (!ignore_cache
&& !ocfs2_buffer_uptodate(ci
, bh
)) {
272 trace_ocfs2_read_blocks_from_disk(
273 (unsigned long long)bh
->b_blocknr
,
274 (unsigned long long)ocfs2_metadata_cache_owner(ci
));
275 /* We're using ignore_cache here to say
280 trace_ocfs2_read_blocks_bh((unsigned long long)bh
->b_blocknr
,
281 ignore_cache
, buffer_jbd(bh
), buffer_dirty(bh
));
283 if (buffer_jbd(bh
)) {
288 if (buffer_dirty(bh
)) {
289 /* This should probably be a BUG, or
290 * at least return an error. */
294 /* A read-ahead request was made - if the
295 * buffer is already under read-ahead from a
296 * previously submitted request than we are
298 if ((flags
& OCFS2_BH_READAHEAD
)
299 && ocfs2_buffer_read_ahead(ci
, bh
))
303 if (buffer_jbd(bh
)) {
304 #ifdef CATCH_BH_JBD_RACES
305 mlog(ML_ERROR
, "block %llu had the JBD bit set "
306 "while I was in lock_buffer!",
307 (unsigned long long)bh
->b_blocknr
);
315 /* Re-check ocfs2_buffer_uptodate() as a
316 * previously read-ahead buffer may have
317 * completed I/O while we were waiting for the
319 if (!(flags
& OCFS2_BH_IGNORE_CACHE
)
320 && !(flags
& OCFS2_BH_READAHEAD
)
321 && ocfs2_buffer_uptodate(ci
, bh
)) {
326 get_bh(bh
); /* for end_buffer_read_sync() */
328 set_buffer_needs_validate(bh
);
329 bh
->b_end_io
= end_buffer_read_sync
;
330 submit_bh(REQ_OP_READ
, bh
);
336 for (i
= (nr
- 1); i
>= 0; i
--) {
339 if (!(flags
& OCFS2_BH_READAHEAD
)) {
340 if (unlikely(status
)) {
341 /* Clear the buffers on error including those
342 * ever succeeded in reading
345 /* If middle bh fails, let previous bh
346 * finish its read and then put it to
353 } else if (bh
&& buffer_uptodate(bh
)) {
354 clear_buffer_uptodate(bh
);
358 /* We know this can't have changed as we hold the
359 * owner sem. Avoid doing any work on the bh if the
364 if (!buffer_uptodate(bh
)) {
365 /* Status won't be cleared from here on out,
366 * so we can safely record this and loop back
367 * to cleanup the other buffers. Don't need to
368 * remove the clustered uptodate information
369 * for this bh as it's not marked locally
372 clear_buffer_needs_validate(bh
);
376 if (buffer_needs_validate(bh
)) {
377 /* We never set NeedsValidate if the
378 * buffer was held by the journal, so
379 * that better not have changed */
380 BUG_ON(buffer_jbd(bh
));
381 clear_buffer_needs_validate(bh
);
382 status
= validate(sb
, bh
);
388 /* Always set the buffer in the cache, even if it was
389 * a forced read, or read-ahead which hasn't yet
392 ocfs2_set_buffer_uptodate(ci
, bh
);
394 ocfs2_metadata_cache_io_unlock(ci
);
396 trace_ocfs2_read_blocks_end((unsigned long long)block
, nr
,
397 flags
, ignore_cache
);
404 /* Check whether the blkno is the super block or one of the backups. */
405 static void ocfs2_check_super_or_backup(struct super_block
*sb
,
411 if (blkno
== OCFS2_SUPER_BLOCK_BLKNO
)
414 for (i
= 0; i
< OCFS2_MAX_BACKUP_SUPERBLOCKS
; i
++) {
415 backup_blkno
= ocfs2_backup_super_blkno(sb
, i
);
416 if (backup_blkno
== blkno
)
424 * Write super block and backups doesn't need to collaborate with journal,
425 * so we don't need to lock ip_io_mutex and ci doesn't need to bea passed
426 * into this function.
428 int ocfs2_write_super_or_backup(struct ocfs2_super
*osb
,
429 struct buffer_head
*bh
)
432 struct ocfs2_dinode
*di
= (struct ocfs2_dinode
*)bh
->b_data
;
434 BUG_ON(buffer_jbd(bh
));
435 ocfs2_check_super_or_backup(osb
->sb
, bh
->b_blocknr
);
437 if (ocfs2_is_hard_readonly(osb
) || ocfs2_is_soft_readonly(osb
)) {
444 set_buffer_uptodate(bh
);
446 /* remove from dirty list before I/O. */
447 clear_buffer_dirty(bh
);
449 get_bh(bh
); /* for end_buffer_write_sync() */
450 bh
->b_end_io
= end_buffer_write_sync
;
451 ocfs2_compute_meta_ecc(osb
->sb
, bh
->b_data
, &di
->i_check
);
452 submit_bh(REQ_OP_WRITE
, bh
);
456 if (!buffer_uptodate(bh
)) {