This source file includes following definitions.
- ubd_test_bit
- ubd_set_bit
- make_proc_ide
- fake_ide_media_proc_show
- make_ide_entries
- fake_ide_setup
- parse_unit
- ubd_setup_common
- ubd_setup
- udb_setup
- bulk_req_safe_read
- ubd_handler
- ubd_intr
- kill_io_thread
- ubd_file_size
- read_cow_bitmap
- backing_file_mismatch
- path_requires_switch
- open_ubd_file
- create_cow_file
- ubd_close_dev
- ubd_open_dev
- ubd_device_release
- ubd_disk_register
- ubd_add
- ubd_config
- ubd_get_config
- ubd_id
- ubd_remove
- ubd_mc_init
- ubd0_init
- ubd_init
- ubd_driver_init
- ubd_open
- ubd_release
- cowify_bitmap
- cowify_req
- ubd_queue_one_vec
- queue_rw_req
- ubd_queue_rq
- ubd_getgeo
- ubd_ioctl
- map_error
- update_bitmap
- do_io
- io_thread
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22 #define UBD_SHIFT 4
23
24 #include <linux/module.h>
25 #include <linux/init.h>
26 #include <linux/blkdev.h>
27 #include <linux/blk-mq.h>
28 #include <linux/ata.h>
29 #include <linux/hdreg.h>
30 #include <linux/cdrom.h>
31 #include <linux/proc_fs.h>
32 #include <linux/seq_file.h>
33 #include <linux/ctype.h>
34 #include <linux/slab.h>
35 #include <linux/vmalloc.h>
36 #include <linux/platform_device.h>
37 #include <linux/scatterlist.h>
38 #include <asm/tlbflush.h>
39 #include <kern_util.h>
40 #include "mconsole_kern.h"
41 #include <init.h>
42 #include <irq_kern.h>
43 #include "ubd.h"
44 #include <os.h>
45 #include "cow.h"
46
47
48 #define UBD_MAX_REQUEST (8 * sizeof(long))
49
50 struct io_thread_req {
51 struct request *req;
52 int fds[2];
53 unsigned long offsets[2];
54 unsigned long long offset;
55 unsigned long length;
56 char *buffer;
57 int sectorsize;
58 unsigned long sector_mask;
59 unsigned long long cow_offset;
60 unsigned long bitmap_words[2];
61 int error;
62 };
63
64
65 static struct io_thread_req * (*irq_req_buffer)[];
66 static struct io_thread_req *irq_remainder;
67 static int irq_remainder_size;
68
69 static struct io_thread_req * (*io_req_buffer)[];
70 static struct io_thread_req *io_remainder;
71 static int io_remainder_size;
72
73
74
75 static inline int ubd_test_bit(__u64 bit, unsigned char *data)
76 {
77 __u64 n;
78 int bits, off;
79
80 bits = sizeof(data[0]) * 8;
81 n = bit / bits;
82 off = bit % bits;
83 return (data[n] & (1 << off)) != 0;
84 }
85
86 static inline void ubd_set_bit(__u64 bit, unsigned char *data)
87 {
88 __u64 n;
89 int bits, off;
90
91 bits = sizeof(data[0]) * 8;
92 n = bit / bits;
93 off = bit % bits;
94 data[n] |= (1 << off);
95 }
96
97
98 #define DRIVER_NAME "uml-blkdev"
99
100 static DEFINE_MUTEX(ubd_lock);
101 static DEFINE_MUTEX(ubd_mutex);
102
103 static int ubd_open(struct block_device *bdev, fmode_t mode);
104 static void ubd_release(struct gendisk *disk, fmode_t mode);
105 static int ubd_ioctl(struct block_device *bdev, fmode_t mode,
106 unsigned int cmd, unsigned long arg);
107 static int ubd_getgeo(struct block_device *bdev, struct hd_geometry *geo);
108
109 #define MAX_DEV (16)
110
111 static const struct block_device_operations ubd_blops = {
112 .owner = THIS_MODULE,
113 .open = ubd_open,
114 .release = ubd_release,
115 .ioctl = ubd_ioctl,
116 .getgeo = ubd_getgeo,
117 };
118
119
120 static int fake_major = UBD_MAJOR;
121 static struct gendisk *ubd_gendisk[MAX_DEV];
122 static struct gendisk *fake_gendisk[MAX_DEV];
123
124 #ifdef CONFIG_BLK_DEV_UBD_SYNC
125 #define OPEN_FLAGS ((struct openflags) { .r = 1, .w = 1, .s = 1, .c = 0, \
126 .cl = 1 })
127 #else
128 #define OPEN_FLAGS ((struct openflags) { .r = 1, .w = 1, .s = 0, .c = 0, \
129 .cl = 1 })
130 #endif
131 static struct openflags global_openflags = OPEN_FLAGS;
132
133 struct cow {
134
135 char *file;
136
137 int fd;
138 unsigned long *bitmap;
139 unsigned long bitmap_len;
140 int bitmap_offset;
141 int data_offset;
142 };
143
144 #define MAX_SG 64
145
146 struct ubd {
147
148
149 char *file;
150 int count;
151 int fd;
152 __u64 size;
153 struct openflags boot_openflags;
154 struct openflags openflags;
155 unsigned shared:1;
156 unsigned no_cow:1;
157 unsigned no_trim:1;
158 struct cow cow;
159 struct platform_device pdev;
160 struct request_queue *queue;
161 struct blk_mq_tag_set tag_set;
162 spinlock_t lock;
163 };
164
165 #define DEFAULT_COW { \
166 .file = NULL, \
167 .fd = -1, \
168 .bitmap = NULL, \
169 .bitmap_offset = 0, \
170 .data_offset = 0, \
171 }
172
173 #define DEFAULT_UBD { \
174 .file = NULL, \
175 .count = 0, \
176 .fd = -1, \
177 .size = -1, \
178 .boot_openflags = OPEN_FLAGS, \
179 .openflags = OPEN_FLAGS, \
180 .no_cow = 0, \
181 .no_trim = 0, \
182 .shared = 0, \
183 .cow = DEFAULT_COW, \
184 .lock = __SPIN_LOCK_UNLOCKED(ubd_devs.lock), \
185 }
186
187
188 static struct ubd ubd_devs[MAX_DEV] = { [0 ... MAX_DEV - 1] = DEFAULT_UBD };
189
190
191 static int fake_ide = 0;
192 static struct proc_dir_entry *proc_ide_root = NULL;
193 static struct proc_dir_entry *proc_ide = NULL;
194
195 static blk_status_t ubd_queue_rq(struct blk_mq_hw_ctx *hctx,
196 const struct blk_mq_queue_data *bd);
197
198 static void make_proc_ide(void)
199 {
200 proc_ide_root = proc_mkdir("ide", NULL);
201 proc_ide = proc_mkdir("ide0", proc_ide_root);
202 }
203
204 static int fake_ide_media_proc_show(struct seq_file *m, void *v)
205 {
206 seq_puts(m, "disk\n");
207 return 0;
208 }
209
210 static void make_ide_entries(const char *dev_name)
211 {
212 struct proc_dir_entry *dir, *ent;
213 char name[64];
214
215 if(proc_ide_root == NULL) make_proc_ide();
216
217 dir = proc_mkdir(dev_name, proc_ide);
218 if(!dir) return;
219
220 ent = proc_create_single("media", S_IRUGO, dir,
221 fake_ide_media_proc_show);
222 if(!ent) return;
223 snprintf(name, sizeof(name), "ide0/%s", dev_name);
224 proc_symlink(dev_name, proc_ide_root, name);
225 }
226
227 static int fake_ide_setup(char *str)
228 {
229 fake_ide = 1;
230 return 1;
231 }
232
233 __setup("fake_ide", fake_ide_setup);
234
235 __uml_help(fake_ide_setup,
236 "fake_ide\n"
237 " Create ide0 entries that map onto ubd devices.\n\n"
238 );
239
240 static int parse_unit(char **ptr)
241 {
242 char *str = *ptr, *end;
243 int n = -1;
244
245 if(isdigit(*str)) {
246 n = simple_strtoul(str, &end, 0);
247 if(end == str)
248 return -1;
249 *ptr = end;
250 }
251 else if (('a' <= *str) && (*str <= 'z')) {
252 n = *str - 'a';
253 str++;
254 *ptr = str;
255 }
256 return n;
257 }
258
259
260
261
262
263 static int ubd_setup_common(char *str, int *index_out, char **error_out)
264 {
265 struct ubd *ubd_dev;
266 struct openflags flags = global_openflags;
267 char *backing_file;
268 int n, err = 0, i;
269
270 if(index_out) *index_out = -1;
271 n = *str;
272 if(n == '='){
273 char *end;
274 int major;
275
276 str++;
277 if(!strcmp(str, "sync")){
278 global_openflags = of_sync(global_openflags);
279 return err;
280 }
281
282 err = -EINVAL;
283 major = simple_strtoul(str, &end, 0);
284 if((*end != '\0') || (end == str)){
285 *error_out = "Didn't parse major number";
286 return err;
287 }
288
289 mutex_lock(&ubd_lock);
290 if (fake_major != UBD_MAJOR) {
291 *error_out = "Can't assign a fake major twice";
292 goto out1;
293 }
294
295 fake_major = major;
296
297 printk(KERN_INFO "Setting extra ubd major number to %d\n",
298 major);
299 err = 0;
300 out1:
301 mutex_unlock(&ubd_lock);
302 return err;
303 }
304
305 n = parse_unit(&str);
306 if(n < 0){
307 *error_out = "Couldn't parse device number";
308 return -EINVAL;
309 }
310 if(n >= MAX_DEV){
311 *error_out = "Device number out of range";
312 return 1;
313 }
314
315 err = -EBUSY;
316 mutex_lock(&ubd_lock);
317
318 ubd_dev = &ubd_devs[n];
319 if(ubd_dev->file != NULL){
320 *error_out = "Device is already configured";
321 goto out;
322 }
323
324 if (index_out)
325 *index_out = n;
326
327 err = -EINVAL;
328 for (i = 0; i < sizeof("rscdt="); i++) {
329 switch (*str) {
330 case 'r':
331 flags.w = 0;
332 break;
333 case 's':
334 flags.s = 1;
335 break;
336 case 'd':
337 ubd_dev->no_cow = 1;
338 break;
339 case 'c':
340 ubd_dev->shared = 1;
341 break;
342 case 't':
343 ubd_dev->no_trim = 1;
344 break;
345 case '=':
346 str++;
347 goto break_loop;
348 default:
349 *error_out = "Expected '=' or flag letter "
350 "(r, s, c, t or d)";
351 goto out;
352 }
353 str++;
354 }
355
356 if (*str == '=')
357 *error_out = "Too many flags specified";
358 else
359 *error_out = "Missing '='";
360 goto out;
361
362 break_loop:
363 backing_file = strchr(str, ',');
364
365 if (backing_file == NULL)
366 backing_file = strchr(str, ':');
367
368 if(backing_file != NULL){
369 if(ubd_dev->no_cow){
370 *error_out = "Can't specify both 'd' and a cow file";
371 goto out;
372 }
373 else {
374 *backing_file = '\0';
375 backing_file++;
376 }
377 }
378 err = 0;
379 ubd_dev->file = str;
380 ubd_dev->cow.file = backing_file;
381 ubd_dev->boot_openflags = flags;
382 out:
383 mutex_unlock(&ubd_lock);
384 return err;
385 }
386
387 static int ubd_setup(char *str)
388 {
389 char *error;
390 int err;
391
392 err = ubd_setup_common(str, NULL, &error);
393 if(err)
394 printk(KERN_ERR "Failed to initialize device with \"%s\" : "
395 "%s\n", str, error);
396 return 1;
397 }
398
399 __setup("ubd", ubd_setup);
400 __uml_help(ubd_setup,
401 "ubd<n><flags>=<filename>[(:|,)<filename2>]\n"
402 " This is used to associate a device with a file in the underlying\n"
403 " filesystem. When specifying two filenames, the first one is the\n"
404 " COW name and the second is the backing file name. As separator you can\n"
405 " use either a ':' or a ',': the first one allows writing things like;\n"
406 " ubd0=~/Uml/root_cow:~/Uml/root_backing_file\n"
407 " while with a ',' the shell would not expand the 2nd '~'.\n"
408 " When using only one filename, UML will detect whether to treat it like\n"
409 " a COW file or a backing file. To override this detection, add the 'd'\n"
410 " flag:\n"
411 " ubd0d=BackingFile\n"
412 " Usually, there is a filesystem in the file, but \n"
413 " that's not required. Swap devices containing swap files can be\n"
414 " specified like this. Also, a file which doesn't contain a\n"
415 " filesystem can have its contents read in the virtual \n"
416 " machine by running 'dd' on the device. <n> must be in the range\n"
417 " 0 to 7. Appending an 'r' to the number will cause that device\n"
418 " to be mounted read-only. For example ubd1r=./ext_fs. Appending\n"
419 " an 's' will cause data to be written to disk on the host immediately.\n"
420 " 'c' will cause the device to be treated as being shared between multiple\n"
421 " UMLs and file locking will be turned off - this is appropriate for a\n"
422 " cluster filesystem and inappropriate at almost all other times.\n\n"
423 " 't' will disable trim/discard support on the device (enabled by default).\n\n"
424 );
425
426 static int udb_setup(char *str)
427 {
428 printk("udb%s specified on command line is almost certainly a ubd -> "
429 "udb TYPO\n", str);
430 return 1;
431 }
432
433 __setup("udb", udb_setup);
434 __uml_help(udb_setup,
435 "udb\n"
436 " This option is here solely to catch ubd -> udb typos, which can be\n"
437 " to impossible to catch visually unless you specifically look for\n"
438 " them. The only result of any option starting with 'udb' is an error\n"
439 " in the boot output.\n\n"
440 );
441
442
443 static int thread_fd = -1;
444
445
446
447
448
449 static int bulk_req_safe_read(
450 int fd,
451 struct io_thread_req * (*request_buffer)[],
452 struct io_thread_req **remainder,
453 int *remainder_size,
454 int max_recs
455 )
456 {
457 int n = 0;
458 int res = 0;
459
460 if (*remainder_size > 0) {
461 memmove(
462 (char *) request_buffer,
463 (char *) remainder, *remainder_size
464 );
465 n = *remainder_size;
466 }
467
468 res = os_read_file(
469 fd,
470 ((char *) request_buffer) + *remainder_size,
471 sizeof(struct io_thread_req *)*max_recs
472 - *remainder_size
473 );
474 if (res > 0) {
475 n += res;
476 if ((n % sizeof(struct io_thread_req *)) > 0) {
477
478
479
480
481
482 *remainder_size = n % sizeof(struct io_thread_req *);
483 WARN(*remainder_size > 0, "UBD IPC read returned a partial result");
484 memmove(
485 remainder,
486 ((char *) request_buffer) +
487 (n/sizeof(struct io_thread_req *))*sizeof(struct io_thread_req *),
488 *remainder_size
489 );
490 n = n - *remainder_size;
491 }
492 } else {
493 n = res;
494 }
495 return n;
496 }
497
498
499 static void ubd_handler(void)
500 {
501 int n;
502 int count;
503
504 while(1){
505 n = bulk_req_safe_read(
506 thread_fd,
507 irq_req_buffer,
508 &irq_remainder,
509 &irq_remainder_size,
510 UBD_REQ_BUFFER_SIZE
511 );
512 if (n < 0) {
513 if(n == -EAGAIN)
514 break;
515 printk(KERN_ERR "spurious interrupt in ubd_handler, "
516 "err = %d\n", -n);
517 return;
518 }
519 for (count = 0; count < n/sizeof(struct io_thread_req *); count++) {
520 struct io_thread_req *io_req = (*irq_req_buffer)[count];
521
522 if ((io_req->error == BLK_STS_NOTSUPP) && (req_op(io_req->req) == REQ_OP_DISCARD)) {
523 blk_queue_max_discard_sectors(io_req->req->q, 0);
524 blk_queue_max_write_zeroes_sectors(io_req->req->q, 0);
525 blk_queue_flag_clear(QUEUE_FLAG_DISCARD, io_req->req->q);
526 }
527 if ((io_req->error) || (io_req->buffer == NULL))
528 blk_mq_end_request(io_req->req, io_req->error);
529 else {
530 if (!blk_update_request(io_req->req, io_req->error, io_req->length))
531 __blk_mq_end_request(io_req->req, io_req->error);
532 }
533 kfree(io_req);
534 }
535 }
536 }
537
538 static irqreturn_t ubd_intr(int irq, void *dev)
539 {
540 ubd_handler();
541 return IRQ_HANDLED;
542 }
543
544
545 static int io_pid = -1;
546
547 static void kill_io_thread(void)
548 {
549 if(io_pid != -1)
550 os_kill_process(io_pid, 1);
551 }
552
553 __uml_exitcall(kill_io_thread);
554
555 static inline int ubd_file_size(struct ubd *ubd_dev, __u64 *size_out)
556 {
557 char *file;
558 int fd;
559 int err;
560
561 __u32 version;
562 __u32 align;
563 char *backing_file;
564 time_t mtime;
565 unsigned long long size;
566 int sector_size;
567 int bitmap_offset;
568
569 if (ubd_dev->file && ubd_dev->cow.file) {
570 file = ubd_dev->cow.file;
571
572 goto out;
573 }
574
575 fd = os_open_file(ubd_dev->file, of_read(OPENFLAGS()), 0);
576 if (fd < 0)
577 return fd;
578
579 err = read_cow_header(file_reader, &fd, &version, &backing_file, \
580 &mtime, &size, §or_size, &align, &bitmap_offset);
581 os_close_file(fd);
582
583 if(err == -EINVAL)
584 file = ubd_dev->file;
585 else
586 file = backing_file;
587
588 out:
589 return os_file_size(file, size_out);
590 }
591
592 static int read_cow_bitmap(int fd, void *buf, int offset, int len)
593 {
594 int err;
595
596 err = os_pread_file(fd, buf, len, offset);
597 if (err < 0)
598 return err;
599
600 return 0;
601 }
602
603 static int backing_file_mismatch(char *file, __u64 size, time_t mtime)
604 {
605 unsigned long modtime;
606 unsigned long long actual;
607 int err;
608
609 err = os_file_modtime(file, &modtime);
610 if (err < 0) {
611 printk(KERN_ERR "Failed to get modification time of backing "
612 "file \"%s\", err = %d\n", file, -err);
613 return err;
614 }
615
616 err = os_file_size(file, &actual);
617 if (err < 0) {
618 printk(KERN_ERR "Failed to get size of backing file \"%s\", "
619 "err = %d\n", file, -err);
620 return err;
621 }
622
623 if (actual != size) {
624
625
626 printk(KERN_ERR "Size mismatch (%llu vs %llu) of COW header "
627 "vs backing file\n", (unsigned long long) size, actual);
628 return -EINVAL;
629 }
630 if (modtime != mtime) {
631 printk(KERN_ERR "mtime mismatch (%ld vs %ld) of COW header vs "
632 "backing file\n", mtime, modtime);
633 return -EINVAL;
634 }
635 return 0;
636 }
637
638 static int path_requires_switch(char *from_cmdline, char *from_cow, char *cow)
639 {
640 struct uml_stat buf1, buf2;
641 int err;
642
643 if (from_cmdline == NULL)
644 return 0;
645 if (!strcmp(from_cmdline, from_cow))
646 return 0;
647
648 err = os_stat_file(from_cmdline, &buf1);
649 if (err < 0) {
650 printk(KERN_ERR "Couldn't stat '%s', err = %d\n", from_cmdline,
651 -err);
652 return 0;
653 }
654 err = os_stat_file(from_cow, &buf2);
655 if (err < 0) {
656 printk(KERN_ERR "Couldn't stat '%s', err = %d\n", from_cow,
657 -err);
658 return 1;
659 }
660 if ((buf1.ust_dev == buf2.ust_dev) && (buf1.ust_ino == buf2.ust_ino))
661 return 0;
662
663 printk(KERN_ERR "Backing file mismatch - \"%s\" requested, "
664 "\"%s\" specified in COW header of \"%s\"\n",
665 from_cmdline, from_cow, cow);
666 return 1;
667 }
668
669 static int open_ubd_file(char *file, struct openflags *openflags, int shared,
670 char **backing_file_out, int *bitmap_offset_out,
671 unsigned long *bitmap_len_out, int *data_offset_out,
672 int *create_cow_out)
673 {
674 time_t mtime;
675 unsigned long long size;
676 __u32 version, align;
677 char *backing_file;
678 int fd, err, sectorsize, asked_switch, mode = 0644;
679
680 fd = os_open_file(file, *openflags, mode);
681 if (fd < 0) {
682 if ((fd == -ENOENT) && (create_cow_out != NULL))
683 *create_cow_out = 1;
684 if (!openflags->w ||
685 ((fd != -EROFS) && (fd != -EACCES)))
686 return fd;
687 openflags->w = 0;
688 fd = os_open_file(file, *openflags, mode);
689 if (fd < 0)
690 return fd;
691 }
692
693 if (shared)
694 printk(KERN_INFO "Not locking \"%s\" on the host\n", file);
695 else {
696 err = os_lock_file(fd, openflags->w);
697 if (err < 0) {
698 printk(KERN_ERR "Failed to lock '%s', err = %d\n",
699 file, -err);
700 goto out_close;
701 }
702 }
703
704
705 if (backing_file_out == NULL)
706 return fd;
707
708 err = read_cow_header(file_reader, &fd, &version, &backing_file, &mtime,
709 &size, §orsize, &align, bitmap_offset_out);
710 if (err && (*backing_file_out != NULL)) {
711 printk(KERN_ERR "Failed to read COW header from COW file "
712 "\"%s\", errno = %d\n", file, -err);
713 goto out_close;
714 }
715 if (err)
716 return fd;
717
718 asked_switch = path_requires_switch(*backing_file_out, backing_file,
719 file);
720
721
722 if (asked_switch && !backing_file_mismatch(*backing_file_out, size,
723 mtime)) {
724 printk(KERN_ERR "Switching backing file to '%s'\n",
725 *backing_file_out);
726 err = write_cow_header(file, fd, *backing_file_out,
727 sectorsize, align, &size);
728 if (err) {
729 printk(KERN_ERR "Switch failed, errno = %d\n", -err);
730 goto out_close;
731 }
732 } else {
733 *backing_file_out = backing_file;
734 err = backing_file_mismatch(*backing_file_out, size, mtime);
735 if (err)
736 goto out_close;
737 }
738
739 cow_sizes(version, size, sectorsize, align, *bitmap_offset_out,
740 bitmap_len_out, data_offset_out);
741
742 return fd;
743 out_close:
744 os_close_file(fd);
745 return err;
746 }
747
748 static int create_cow_file(char *cow_file, char *backing_file,
749 struct openflags flags,
750 int sectorsize, int alignment, int *bitmap_offset_out,
751 unsigned long *bitmap_len_out, int *data_offset_out)
752 {
753 int err, fd;
754
755 flags.c = 1;
756 fd = open_ubd_file(cow_file, &flags, 0, NULL, NULL, NULL, NULL, NULL);
757 if (fd < 0) {
758 err = fd;
759 printk(KERN_ERR "Open of COW file '%s' failed, errno = %d\n",
760 cow_file, -err);
761 goto out;
762 }
763
764 err = init_cow_file(fd, cow_file, backing_file, sectorsize, alignment,
765 bitmap_offset_out, bitmap_len_out,
766 data_offset_out);
767 if (!err)
768 return fd;
769 os_close_file(fd);
770 out:
771 return err;
772 }
773
774 static void ubd_close_dev(struct ubd *ubd_dev)
775 {
776 os_close_file(ubd_dev->fd);
777 if(ubd_dev->cow.file == NULL)
778 return;
779
780 os_close_file(ubd_dev->cow.fd);
781 vfree(ubd_dev->cow.bitmap);
782 ubd_dev->cow.bitmap = NULL;
783 }
784
785 static int ubd_open_dev(struct ubd *ubd_dev)
786 {
787 struct openflags flags;
788 char **back_ptr;
789 int err, create_cow, *create_ptr;
790 int fd;
791
792 ubd_dev->openflags = ubd_dev->boot_openflags;
793 create_cow = 0;
794 create_ptr = (ubd_dev->cow.file != NULL) ? &create_cow : NULL;
795 back_ptr = ubd_dev->no_cow ? NULL : &ubd_dev->cow.file;
796
797 fd = open_ubd_file(ubd_dev->file, &ubd_dev->openflags, ubd_dev->shared,
798 back_ptr, &ubd_dev->cow.bitmap_offset,
799 &ubd_dev->cow.bitmap_len, &ubd_dev->cow.data_offset,
800 create_ptr);
801
802 if((fd == -ENOENT) && create_cow){
803 fd = create_cow_file(ubd_dev->file, ubd_dev->cow.file,
804 ubd_dev->openflags, SECTOR_SIZE, PAGE_SIZE,
805 &ubd_dev->cow.bitmap_offset,
806 &ubd_dev->cow.bitmap_len,
807 &ubd_dev->cow.data_offset);
808 if(fd >= 0){
809 printk(KERN_INFO "Creating \"%s\" as COW file for "
810 "\"%s\"\n", ubd_dev->file, ubd_dev->cow.file);
811 }
812 }
813
814 if(fd < 0){
815 printk("Failed to open '%s', errno = %d\n", ubd_dev->file,
816 -fd);
817 return fd;
818 }
819 ubd_dev->fd = fd;
820
821 if(ubd_dev->cow.file != NULL){
822 blk_queue_max_hw_sectors(ubd_dev->queue, 8 * sizeof(long));
823
824 err = -ENOMEM;
825 ubd_dev->cow.bitmap = vmalloc(ubd_dev->cow.bitmap_len);
826 if(ubd_dev->cow.bitmap == NULL){
827 printk(KERN_ERR "Failed to vmalloc COW bitmap\n");
828 goto error;
829 }
830 flush_tlb_kernel_vm();
831
832 err = read_cow_bitmap(ubd_dev->fd, ubd_dev->cow.bitmap,
833 ubd_dev->cow.bitmap_offset,
834 ubd_dev->cow.bitmap_len);
835 if(err < 0)
836 goto error;
837
838 flags = ubd_dev->openflags;
839 flags.w = 0;
840 err = open_ubd_file(ubd_dev->cow.file, &flags, ubd_dev->shared, NULL,
841 NULL, NULL, NULL, NULL);
842 if(err < 0) goto error;
843 ubd_dev->cow.fd = err;
844 }
845 if (ubd_dev->no_trim == 0) {
846 ubd_dev->queue->limits.discard_granularity = SECTOR_SIZE;
847 ubd_dev->queue->limits.discard_alignment = SECTOR_SIZE;
848 blk_queue_max_discard_sectors(ubd_dev->queue, UBD_MAX_REQUEST);
849 blk_queue_max_write_zeroes_sectors(ubd_dev->queue, UBD_MAX_REQUEST);
850 blk_queue_flag_set(QUEUE_FLAG_DISCARD, ubd_dev->queue);
851 }
852 blk_queue_flag_set(QUEUE_FLAG_NONROT, ubd_dev->queue);
853 return 0;
854 error:
855 os_close_file(ubd_dev->fd);
856 return err;
857 }
858
859 static void ubd_device_release(struct device *dev)
860 {
861 struct ubd *ubd_dev = dev_get_drvdata(dev);
862
863 blk_cleanup_queue(ubd_dev->queue);
864 blk_mq_free_tag_set(&ubd_dev->tag_set);
865 *ubd_dev = ((struct ubd) DEFAULT_UBD);
866 }
867
868 static int ubd_disk_register(int major, u64 size, int unit,
869 struct gendisk **disk_out)
870 {
871 struct device *parent = NULL;
872 struct gendisk *disk;
873
874 disk = alloc_disk(1 << UBD_SHIFT);
875 if(disk == NULL)
876 return -ENOMEM;
877
878 disk->major = major;
879 disk->first_minor = unit << UBD_SHIFT;
880 disk->fops = &ubd_blops;
881 set_capacity(disk, size / 512);
882 if (major == UBD_MAJOR)
883 sprintf(disk->disk_name, "ubd%c", 'a' + unit);
884 else
885 sprintf(disk->disk_name, "ubd_fake%d", unit);
886
887
888 if (major == UBD_MAJOR) {
889 ubd_devs[unit].pdev.id = unit;
890 ubd_devs[unit].pdev.name = DRIVER_NAME;
891 ubd_devs[unit].pdev.dev.release = ubd_device_release;
892 dev_set_drvdata(&ubd_devs[unit].pdev.dev, &ubd_devs[unit]);
893 platform_device_register(&ubd_devs[unit].pdev);
894 parent = &ubd_devs[unit].pdev.dev;
895 }
896
897 disk->private_data = &ubd_devs[unit];
898 disk->queue = ubd_devs[unit].queue;
899 device_add_disk(parent, disk, NULL);
900
901 *disk_out = disk;
902 return 0;
903 }
904
905 #define ROUND_BLOCK(n) ((n + (SECTOR_SIZE - 1)) & (-SECTOR_SIZE))
906
907 static const struct blk_mq_ops ubd_mq_ops = {
908 .queue_rq = ubd_queue_rq,
909 };
910
911 static int ubd_add(int n, char **error_out)
912 {
913 struct ubd *ubd_dev = &ubd_devs[n];
914 int err = 0;
915
916 if(ubd_dev->file == NULL)
917 goto out;
918
919 err = ubd_file_size(ubd_dev, &ubd_dev->size);
920 if(err < 0){
921 *error_out = "Couldn't determine size of device's file";
922 goto out;
923 }
924
925 ubd_dev->size = ROUND_BLOCK(ubd_dev->size);
926
927 ubd_dev->tag_set.ops = &ubd_mq_ops;
928 ubd_dev->tag_set.queue_depth = 64;
929 ubd_dev->tag_set.numa_node = NUMA_NO_NODE;
930 ubd_dev->tag_set.flags = BLK_MQ_F_SHOULD_MERGE;
931 ubd_dev->tag_set.driver_data = ubd_dev;
932 ubd_dev->tag_set.nr_hw_queues = 1;
933
934 err = blk_mq_alloc_tag_set(&ubd_dev->tag_set);
935 if (err)
936 goto out;
937
938 ubd_dev->queue = blk_mq_init_queue(&ubd_dev->tag_set);
939 if (IS_ERR(ubd_dev->queue)) {
940 err = PTR_ERR(ubd_dev->queue);
941 goto out_cleanup_tags;
942 }
943
944 ubd_dev->queue->queuedata = ubd_dev;
945 blk_queue_write_cache(ubd_dev->queue, true, false);
946
947 blk_queue_max_segments(ubd_dev->queue, MAX_SG);
948 err = ubd_disk_register(UBD_MAJOR, ubd_dev->size, n, &ubd_gendisk[n]);
949 if(err){
950 *error_out = "Failed to register device";
951 goto out_cleanup_tags;
952 }
953
954 if (fake_major != UBD_MAJOR)
955 ubd_disk_register(fake_major, ubd_dev->size, n,
956 &fake_gendisk[n]);
957
958
959
960
961
962 if (fake_ide)
963 make_ide_entries(ubd_gendisk[n]->disk_name);
964
965 err = 0;
966 out:
967 return err;
968
969 out_cleanup_tags:
970 blk_mq_free_tag_set(&ubd_dev->tag_set);
971 if (!(IS_ERR(ubd_dev->queue)))
972 blk_cleanup_queue(ubd_dev->queue);
973 goto out;
974 }
975
976 static int ubd_config(char *str, char **error_out)
977 {
978 int n, ret;
979
980
981
982
983
984 str = kstrdup(str, GFP_KERNEL);
985 if (str == NULL) {
986 *error_out = "Failed to allocate memory";
987 return -ENOMEM;
988 }
989
990 ret = ubd_setup_common(str, &n, error_out);
991 if (ret)
992 goto err_free;
993
994 if (n == -1) {
995 ret = 0;
996 goto err_free;
997 }
998
999 mutex_lock(&ubd_lock);
1000 ret = ubd_add(n, error_out);
1001 if (ret)
1002 ubd_devs[n].file = NULL;
1003 mutex_unlock(&ubd_lock);
1004
1005 out:
1006 return ret;
1007
1008 err_free:
1009 kfree(str);
1010 goto out;
1011 }
1012
1013 static int ubd_get_config(char *name, char *str, int size, char **error_out)
1014 {
1015 struct ubd *ubd_dev;
1016 int n, len = 0;
1017
1018 n = parse_unit(&name);
1019 if((n >= MAX_DEV) || (n < 0)){
1020 *error_out = "ubd_get_config : device number out of range";
1021 return -1;
1022 }
1023
1024 ubd_dev = &ubd_devs[n];
1025 mutex_lock(&ubd_lock);
1026
1027 if(ubd_dev->file == NULL){
1028 CONFIG_CHUNK(str, size, len, "", 1);
1029 goto out;
1030 }
1031
1032 CONFIG_CHUNK(str, size, len, ubd_dev->file, 0);
1033
1034 if(ubd_dev->cow.file != NULL){
1035 CONFIG_CHUNK(str, size, len, ",", 0);
1036 CONFIG_CHUNK(str, size, len, ubd_dev->cow.file, 1);
1037 }
1038 else CONFIG_CHUNK(str, size, len, "", 1);
1039
1040 out:
1041 mutex_unlock(&ubd_lock);
1042 return len;
1043 }
1044
1045 static int ubd_id(char **str, int *start_out, int *end_out)
1046 {
1047 int n;
1048
1049 n = parse_unit(str);
1050 *start_out = 0;
1051 *end_out = MAX_DEV - 1;
1052 return n;
1053 }
1054
1055 static int ubd_remove(int n, char **error_out)
1056 {
1057 struct gendisk *disk = ubd_gendisk[n];
1058 struct ubd *ubd_dev;
1059 int err = -ENODEV;
1060
1061 mutex_lock(&ubd_lock);
1062
1063 ubd_dev = &ubd_devs[n];
1064
1065 if(ubd_dev->file == NULL)
1066 goto out;
1067
1068
1069 err = -EBUSY;
1070 if(ubd_dev->count > 0)
1071 goto out;
1072
1073 ubd_gendisk[n] = NULL;
1074 if(disk != NULL){
1075 del_gendisk(disk);
1076 put_disk(disk);
1077 }
1078
1079 if(fake_gendisk[n] != NULL){
1080 del_gendisk(fake_gendisk[n]);
1081 put_disk(fake_gendisk[n]);
1082 fake_gendisk[n] = NULL;
1083 }
1084
1085 err = 0;
1086 platform_device_unregister(&ubd_dev->pdev);
1087 out:
1088 mutex_unlock(&ubd_lock);
1089 return err;
1090 }
1091
1092
1093
1094
1095 static struct mc_device ubd_mc = {
1096 .list = LIST_HEAD_INIT(ubd_mc.list),
1097 .name = "ubd",
1098 .config = ubd_config,
1099 .get_config = ubd_get_config,
1100 .id = ubd_id,
1101 .remove = ubd_remove,
1102 };
1103
1104 static int __init ubd_mc_init(void)
1105 {
1106 mconsole_register_dev(&ubd_mc);
1107 return 0;
1108 }
1109
1110 __initcall(ubd_mc_init);
1111
1112 static int __init ubd0_init(void)
1113 {
1114 struct ubd *ubd_dev = &ubd_devs[0];
1115
1116 mutex_lock(&ubd_lock);
1117 if(ubd_dev->file == NULL)
1118 ubd_dev->file = "root_fs";
1119 mutex_unlock(&ubd_lock);
1120
1121 return 0;
1122 }
1123
1124 __initcall(ubd0_init);
1125
1126
1127 static struct platform_driver ubd_driver = {
1128 .driver = {
1129 .name = DRIVER_NAME,
1130 },
1131 };
1132
1133 static int __init ubd_init(void)
1134 {
1135 char *error;
1136 int i, err;
1137
1138 if (register_blkdev(UBD_MAJOR, "ubd"))
1139 return -1;
1140
1141 if (fake_major != UBD_MAJOR) {
1142 char name[sizeof("ubd_nnn\0")];
1143
1144 snprintf(name, sizeof(name), "ubd_%d", fake_major);
1145 if (register_blkdev(fake_major, "ubd"))
1146 return -1;
1147 }
1148
1149 irq_req_buffer = kmalloc_array(UBD_REQ_BUFFER_SIZE,
1150 sizeof(struct io_thread_req *),
1151 GFP_KERNEL
1152 );
1153 irq_remainder = 0;
1154
1155 if (irq_req_buffer == NULL) {
1156 printk(KERN_ERR "Failed to initialize ubd buffering\n");
1157 return -1;
1158 }
1159 io_req_buffer = kmalloc_array(UBD_REQ_BUFFER_SIZE,
1160 sizeof(struct io_thread_req *),
1161 GFP_KERNEL
1162 );
1163
1164 io_remainder = 0;
1165
1166 if (io_req_buffer == NULL) {
1167 printk(KERN_ERR "Failed to initialize ubd buffering\n");
1168 return -1;
1169 }
1170 platform_driver_register(&ubd_driver);
1171 mutex_lock(&ubd_lock);
1172 for (i = 0; i < MAX_DEV; i++){
1173 err = ubd_add(i, &error);
1174 if(err)
1175 printk(KERN_ERR "Failed to initialize ubd device %d :"
1176 "%s\n", i, error);
1177 }
1178 mutex_unlock(&ubd_lock);
1179 return 0;
1180 }
1181
1182 late_initcall(ubd_init);
1183
1184 static int __init ubd_driver_init(void){
1185 unsigned long stack;
1186 int err;
1187
1188
1189 if(global_openflags.s){
1190 printk(KERN_INFO "ubd: Synchronous mode\n");
1191
1192
1193 }
1194 stack = alloc_stack(0, 0);
1195 io_pid = start_io_thread(stack + PAGE_SIZE - sizeof(void *),
1196 &thread_fd);
1197 if(io_pid < 0){
1198 printk(KERN_ERR
1199 "ubd : Failed to start I/O thread (errno = %d) - "
1200 "falling back to synchronous I/O\n", -io_pid);
1201 io_pid = -1;
1202 return 0;
1203 }
1204 err = um_request_irq(UBD_IRQ, thread_fd, IRQ_READ, ubd_intr,
1205 0, "ubd", ubd_devs);
1206 if(err != 0)
1207 printk(KERN_ERR "um_request_irq failed - errno = %d\n", -err);
1208 return 0;
1209 }
1210
1211 device_initcall(ubd_driver_init);
1212
1213 static int ubd_open(struct block_device *bdev, fmode_t mode)
1214 {
1215 struct gendisk *disk = bdev->bd_disk;
1216 struct ubd *ubd_dev = disk->private_data;
1217 int err = 0;
1218
1219 mutex_lock(&ubd_mutex);
1220 if(ubd_dev->count == 0){
1221 err = ubd_open_dev(ubd_dev);
1222 if(err){
1223 printk(KERN_ERR "%s: Can't open \"%s\": errno = %d\n",
1224 disk->disk_name, ubd_dev->file, -err);
1225 goto out;
1226 }
1227 }
1228 ubd_dev->count++;
1229 set_disk_ro(disk, !ubd_dev->openflags.w);
1230
1231
1232
1233
1234
1235
1236
1237 out:
1238 mutex_unlock(&ubd_mutex);
1239 return err;
1240 }
1241
1242 static void ubd_release(struct gendisk *disk, fmode_t mode)
1243 {
1244 struct ubd *ubd_dev = disk->private_data;
1245
1246 mutex_lock(&ubd_mutex);
1247 if(--ubd_dev->count == 0)
1248 ubd_close_dev(ubd_dev);
1249 mutex_unlock(&ubd_mutex);
1250 }
1251
1252 static void cowify_bitmap(__u64 io_offset, int length, unsigned long *cow_mask,
1253 __u64 *cow_offset, unsigned long *bitmap,
1254 __u64 bitmap_offset, unsigned long *bitmap_words,
1255 __u64 bitmap_len)
1256 {
1257 __u64 sector = io_offset >> SECTOR_SHIFT;
1258 int i, update_bitmap = 0;
1259
1260 for (i = 0; i < length >> SECTOR_SHIFT; i++) {
1261 if(cow_mask != NULL)
1262 ubd_set_bit(i, (unsigned char *) cow_mask);
1263 if(ubd_test_bit(sector + i, (unsigned char *) bitmap))
1264 continue;
1265
1266 update_bitmap = 1;
1267 ubd_set_bit(sector + i, (unsigned char *) bitmap);
1268 }
1269
1270 if(!update_bitmap)
1271 return;
1272
1273 *cow_offset = sector / (sizeof(unsigned long) * 8);
1274
1275
1276
1277
1278
1279
1280 if (*cow_offset == (DIV_ROUND_UP(bitmap_len,
1281 sizeof(unsigned long)) - 1))
1282 (*cow_offset)--;
1283
1284 bitmap_words[0] = bitmap[*cow_offset];
1285 bitmap_words[1] = bitmap[*cow_offset + 1];
1286
1287 *cow_offset *= sizeof(unsigned long);
1288 *cow_offset += bitmap_offset;
1289 }
1290
1291 static void cowify_req(struct io_thread_req *req, unsigned long *bitmap,
1292 __u64 bitmap_offset, __u64 bitmap_len)
1293 {
1294 __u64 sector = req->offset >> SECTOR_SHIFT;
1295 int i;
1296
1297 if (req->length > (sizeof(req->sector_mask) * 8) << SECTOR_SHIFT)
1298 panic("Operation too long");
1299
1300 if (req_op(req->req) == REQ_OP_READ) {
1301 for (i = 0; i < req->length >> SECTOR_SHIFT; i++) {
1302 if(ubd_test_bit(sector + i, (unsigned char *) bitmap))
1303 ubd_set_bit(i, (unsigned char *)
1304 &req->sector_mask);
1305 }
1306 }
1307 else cowify_bitmap(req->offset, req->length, &req->sector_mask,
1308 &req->cow_offset, bitmap, bitmap_offset,
1309 req->bitmap_words, bitmap_len);
1310 }
1311
1312 static int ubd_queue_one_vec(struct blk_mq_hw_ctx *hctx, struct request *req,
1313 u64 off, struct bio_vec *bvec)
1314 {
1315 struct ubd *dev = hctx->queue->queuedata;
1316 struct io_thread_req *io_req;
1317 int ret;
1318
1319 io_req = kmalloc(sizeof(struct io_thread_req), GFP_ATOMIC);
1320 if (!io_req)
1321 return -ENOMEM;
1322
1323 io_req->req = req;
1324 if (dev->cow.file)
1325 io_req->fds[0] = dev->cow.fd;
1326 else
1327 io_req->fds[0] = dev->fd;
1328 io_req->error = 0;
1329
1330 if (bvec != NULL) {
1331 io_req->buffer = page_address(bvec->bv_page) + bvec->bv_offset;
1332 io_req->length = bvec->bv_len;
1333 } else {
1334 io_req->buffer = NULL;
1335 io_req->length = blk_rq_bytes(req);
1336 }
1337
1338 io_req->sectorsize = SECTOR_SIZE;
1339 io_req->fds[1] = dev->fd;
1340 io_req->cow_offset = -1;
1341 io_req->offset = off;
1342 io_req->sector_mask = 0;
1343 io_req->offsets[0] = 0;
1344 io_req->offsets[1] = dev->cow.data_offset;
1345
1346 if (dev->cow.file)
1347 cowify_req(io_req, dev->cow.bitmap,
1348 dev->cow.bitmap_offset, dev->cow.bitmap_len);
1349
1350 ret = os_write_file(thread_fd, &io_req, sizeof(io_req));
1351 if (ret != sizeof(io_req)) {
1352 if (ret != -EAGAIN)
1353 pr_err("write to io thread failed: %d\n", -ret);
1354 kfree(io_req);
1355 }
1356 return ret;
1357 }
1358
1359 static int queue_rw_req(struct blk_mq_hw_ctx *hctx, struct request *req)
1360 {
1361 struct req_iterator iter;
1362 struct bio_vec bvec;
1363 int ret;
1364 u64 off = (u64)blk_rq_pos(req) << SECTOR_SHIFT;
1365
1366 rq_for_each_segment(bvec, req, iter) {
1367 ret = ubd_queue_one_vec(hctx, req, off, &bvec);
1368 if (ret < 0)
1369 return ret;
1370 off += bvec.bv_len;
1371 }
1372 return 0;
1373 }
1374
1375 static blk_status_t ubd_queue_rq(struct blk_mq_hw_ctx *hctx,
1376 const struct blk_mq_queue_data *bd)
1377 {
1378 struct ubd *ubd_dev = hctx->queue->queuedata;
1379 struct request *req = bd->rq;
1380 int ret = 0, res = BLK_STS_OK;
1381
1382 blk_mq_start_request(req);
1383
1384 spin_lock_irq(&ubd_dev->lock);
1385
1386 switch (req_op(req)) {
1387
1388 case REQ_OP_FLUSH:
1389 ret = ubd_queue_one_vec(hctx, req, 0, NULL);
1390 break;
1391 case REQ_OP_READ:
1392 case REQ_OP_WRITE:
1393 ret = queue_rw_req(hctx, req);
1394 break;
1395 case REQ_OP_DISCARD:
1396 case REQ_OP_WRITE_ZEROES:
1397 ret = ubd_queue_one_vec(hctx, req, (u64)blk_rq_pos(req) << 9, NULL);
1398 break;
1399 default:
1400 WARN_ON_ONCE(1);
1401 res = BLK_STS_NOTSUPP;
1402 }
1403
1404 spin_unlock_irq(&ubd_dev->lock);
1405
1406 if (ret < 0) {
1407 if (ret == -ENOMEM)
1408 res = BLK_STS_RESOURCE;
1409 else
1410 res = BLK_STS_DEV_RESOURCE;
1411 }
1412
1413 return res;
1414 }
1415
1416 static int ubd_getgeo(struct block_device *bdev, struct hd_geometry *geo)
1417 {
1418 struct ubd *ubd_dev = bdev->bd_disk->private_data;
1419
1420 geo->heads = 128;
1421 geo->sectors = 32;
1422 geo->cylinders = ubd_dev->size / (128 * 32 * 512);
1423 return 0;
1424 }
1425
1426 static int ubd_ioctl(struct block_device *bdev, fmode_t mode,
1427 unsigned int cmd, unsigned long arg)
1428 {
1429 struct ubd *ubd_dev = bdev->bd_disk->private_data;
1430 u16 ubd_id[ATA_ID_WORDS];
1431
1432 switch (cmd) {
1433 struct cdrom_volctrl volume;
1434 case HDIO_GET_IDENTITY:
1435 memset(&ubd_id, 0, ATA_ID_WORDS * 2);
1436 ubd_id[ATA_ID_CYLS] = ubd_dev->size / (128 * 32 * 512);
1437 ubd_id[ATA_ID_HEADS] = 128;
1438 ubd_id[ATA_ID_SECTORS] = 32;
1439 if(copy_to_user((char __user *) arg, (char *) &ubd_id,
1440 sizeof(ubd_id)))
1441 return -EFAULT;
1442 return 0;
1443
1444 case CDROMVOLREAD:
1445 if(copy_from_user(&volume, (char __user *) arg, sizeof(volume)))
1446 return -EFAULT;
1447 volume.channel0 = 255;
1448 volume.channel1 = 255;
1449 volume.channel2 = 255;
1450 volume.channel3 = 255;
1451 if(copy_to_user((char __user *) arg, &volume, sizeof(volume)))
1452 return -EFAULT;
1453 return 0;
1454 }
1455 return -EINVAL;
1456 }
1457
1458 static int map_error(int error_code)
1459 {
1460 switch (error_code) {
1461 case 0:
1462 return BLK_STS_OK;
1463 case ENOSYS:
1464 case EOPNOTSUPP:
1465 return BLK_STS_NOTSUPP;
1466 case ENOSPC:
1467 return BLK_STS_NOSPC;
1468 }
1469 return BLK_STS_IOERR;
1470 }
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485 static int update_bitmap(struct io_thread_req *req)
1486 {
1487 int n;
1488
1489 if(req->cow_offset == -1)
1490 return map_error(0);
1491
1492 n = os_pwrite_file(req->fds[1], &req->bitmap_words,
1493 sizeof(req->bitmap_words), req->cow_offset);
1494 if (n != sizeof(req->bitmap_words))
1495 return map_error(-n);
1496
1497 return map_error(0);
1498 }
1499
1500 static void do_io(struct io_thread_req *req)
1501 {
1502 char *buf = NULL;
1503 unsigned long len;
1504 int n, nsectors, start, end, bit;
1505 __u64 off;
1506
1507
1508
1509 if (req_op(req->req) == REQ_OP_FLUSH) {
1510
1511 req->error = map_error(-os_sync_file(req->fds[0]));
1512 return;
1513 }
1514
1515 nsectors = req->length / req->sectorsize;
1516 start = 0;
1517 do {
1518 bit = ubd_test_bit(start, (unsigned char *) &req->sector_mask);
1519 end = start;
1520 while((end < nsectors) &&
1521 (ubd_test_bit(end, (unsigned char *)
1522 &req->sector_mask) == bit))
1523 end++;
1524
1525 off = req->offset + req->offsets[bit] +
1526 start * req->sectorsize;
1527 len = (end - start) * req->sectorsize;
1528 if (req->buffer != NULL)
1529 buf = &req->buffer[start * req->sectorsize];
1530
1531 switch (req_op(req->req)) {
1532 case REQ_OP_READ:
1533 n = 0;
1534 do {
1535 buf = &buf[n];
1536 len -= n;
1537 n = os_pread_file(req->fds[bit], buf, len, off);
1538 if (n < 0) {
1539 req->error = map_error(-n);
1540 return;
1541 }
1542 } while((n < len) && (n != 0));
1543 if (n < len) memset(&buf[n], 0, len - n);
1544 break;
1545 case REQ_OP_WRITE:
1546 n = os_pwrite_file(req->fds[bit], buf, len, off);
1547 if(n != len){
1548 req->error = map_error(-n);
1549 return;
1550 }
1551 break;
1552 case REQ_OP_DISCARD:
1553 case REQ_OP_WRITE_ZEROES:
1554 n = os_falloc_punch(req->fds[bit], off, len);
1555 if (n) {
1556 req->error = map_error(-n);
1557 return;
1558 }
1559 break;
1560 default:
1561 WARN_ON_ONCE(1);
1562 req->error = BLK_STS_NOTSUPP;
1563 return;
1564 }
1565
1566 start = end;
1567 } while(start < nsectors);
1568
1569 req->error = update_bitmap(req);
1570 }
1571
1572
1573
1574
1575 int kernel_fd = -1;
1576
1577
1578 static int io_count = 0;
1579
1580 int io_thread(void *arg)
1581 {
1582 int n, count, written, res;
1583
1584 os_fix_helper_signals();
1585
1586 while(1){
1587 n = bulk_req_safe_read(
1588 kernel_fd,
1589 io_req_buffer,
1590 &io_remainder,
1591 &io_remainder_size,
1592 UBD_REQ_BUFFER_SIZE
1593 );
1594 if (n < 0) {
1595 if (n == -EAGAIN) {
1596 ubd_read_poll(-1);
1597 continue;
1598 }
1599 }
1600
1601 for (count = 0; count < n/sizeof(struct io_thread_req *); count++) {
1602 io_count++;
1603 do_io((*io_req_buffer)[count]);
1604 }
1605
1606 written = 0;
1607
1608 do {
1609 res = os_write_file(kernel_fd,
1610 ((char *) io_req_buffer) + written,
1611 n - written);
1612 if (res >= 0) {
1613 written += res;
1614 }
1615 if (written < n) {
1616 ubd_write_poll(-1);
1617 }
1618 } while (written < n);
1619 }
1620
1621 return 0;
1622 }