1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467
|
/* nbdkit
* Copyright Red Hat
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions are
* met:
*
* * Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
*
* * Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
*
* * Neither the name of Red Hat nor the names of its contributors may be
* used to endorse or promote products derived from this software without
* specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY RED HAT AND CONTRIBUTORS ''AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO,
* THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
* PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL RED HAT OR
* CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
* SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
* LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF
* USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
* ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
* OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT
* OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*/
#include <config.h>
#include <stdio.h>
#include <stdlib.h>
#include <stdint.h>
#include <inttypes.h>
#include <string.h>
#include <unistd.h>
#include <fcntl.h>
#include <errno.h>
#include <sys/types.h>
#include <stdbool.h>
#include <pthread.h>
#include <nbdkit-plugin.h>
#include "cleanup.h"
#include "pread.h"
#include "pwrite.h"
#include "string-vector.h"
#include "utils.h"
#include "windows-compat.h"
/* The files. */
static string_vector filenames = empty_vector;
#ifdef SEEK_HOLE
/* Any callbacks using lseek must be protected by this lock. */
static pthread_mutex_t lseek_lock = PTHREAD_MUTEX_INITIALIZER;
#endif
static void
split_unload (void)
{
string_vector_empty (&filenames);
}
static int
split_config (const char *key, const char *value)
{
char *s;
if (strcmp (key, "file") == 0) {
s = nbdkit_realpath (value);
if (s == NULL)
return -1;
if (string_vector_append (&filenames, s) == -1) {
nbdkit_error ("realloc: %m");
return -1;
}
}
else {
nbdkit_error ("unknown parameter '%s'", key);
return -1;
}
return 0;
}
#define split_config_help \
"file=<FILENAME> (required) File(s) to serve."
/* The per-connection handle. */
struct handle {
struct file *files;
uint64_t size; /* Total concatenated size. */
};
struct file {
uint64_t offset, size;
int fd;
bool can_extents;
};
/* Create the per-connection handle. */
static void *
split_open (int readonly)
{
struct handle *h;
int flags;
size_t i;
uint64_t offset;
#ifdef SEEK_HOLE
off_t r;
#endif
h = malloc (sizeof *h);
if (h == NULL) {
nbdkit_error ("malloc: %m");
return NULL;
}
h->files = malloc (filenames.len * sizeof (struct file));
if (h->files == NULL) {
nbdkit_error ("malloc: %m");
free (h);
return NULL;
}
for (i = 0; i < filenames.len; ++i)
h->files[i].fd = -1;
/* Open the files. */
flags = O_CLOEXEC|O_NOCTTY;
if (readonly)
flags |= O_RDONLY;
else
flags |= O_RDWR;
for (i = 0; i < filenames.len; ++i) {
h->files[i].fd = open (filenames.ptr[i], flags);
if (h->files[i].fd == -1) {
nbdkit_error ("open: %s: %m", filenames.ptr[i]);
goto err;
}
}
offset = 0;
for (i = 0; i < filenames.len; ++i) {
h->files[i].offset = offset;
h->files[i].size = device_size (h->files[i].fd, NULL);
if (h->files[i].size == -1) {
nbdkit_error ("%s: device_size: %m", filenames.ptr[i]);
goto err;
}
offset += h->files[i].size;
nbdkit_debug ("file[%zu]=%s: offset=%" PRIu64 ", size=%" PRIu64,
i, filenames.ptr[i], h->files[i].offset, h->files[i].size);
#ifdef SEEK_HOLE
/* Test if this file supports extents. */
ACQUIRE_LOCK_FOR_CURRENT_SCOPE (&lseek_lock);
r = lseek (h->files[i].fd, 0, SEEK_DATA);
if (r == -1 && errno != ENXIO) {
nbdkit_debug ("disabling extents: lseek on %s: %m", filenames.ptr[i]);
h->files[i].can_extents = false;
}
else
h->files[i].can_extents = true;
#else
h->files[i].can_extents = false;
#endif
}
h->size = offset;
nbdkit_debug ("total size=%" PRIu64, h->size);
return h;
err:
for (i = 0; i < filenames.len; ++i) {
if (h->files[i].fd >= 0)
close (h->files[i].fd);
}
free (h->files);
free (h);
return NULL;
}
/* Free up the per-connection handle. */
static void
split_close (void *handle)
{
struct handle *h = handle;
size_t i;
for (i = 0; i < filenames.len; ++i)
close (h->files[i].fd);
free (h->files);
free (h);
}
#define THREAD_MODEL NBDKIT_THREAD_MODEL_SERIALIZE_REQUESTS
/* Get the disk size. */
static int64_t
split_get_size (void *handle)
{
struct handle *h = handle;
return (int64_t) h->size;
}
static int
split_can_cache (void *handle)
{
/* Prefer posix_fadvise(), but letting nbdkit call .pread on our
* behalf also tends to work well for the local file system
* cache.
*/
#if HAVE_POSIX_FADVISE
return NBDKIT_FUA_NATIVE;
#else
return NBDKIT_FUA_EMULATE;
#endif
}
/* Helper function to map the offset to the correct file. */
static int
compare_offset (const void *offsetp, const void *filep)
{
const uint64_t offset = *(uint64_t *)offsetp;
const struct file *file = (struct file *)filep;
if (offset < file->offset) return -1;
if (offset >= file->offset + file->size) return 1;
return 0;
}
static struct file *
get_file (struct handle *h, uint64_t offset)
{
return bsearch (&offset, h->files,
filenames.len, sizeof (struct file),
compare_offset);
}
/* Read data. */
static int
split_pread (void *handle, void *buf, uint32_t count, uint64_t offset)
{
struct handle *h = handle;
while (count > 0) {
struct file *file = get_file (h, offset);
uint64_t foffs = offset - file->offset;
uint64_t max;
ssize_t r;
max = file->size - foffs;
if (max > count)
max = count;
r = pread (file->fd, buf, max, foffs);
if (r == -1) {
nbdkit_error ("pread: %m");
return -1;
}
if (r == 0) {
nbdkit_error ("pread: unexpected end of file");
return -1;
}
buf += r;
count -= r;
offset += r;
}
return 0;
}
/* Write data to the file. */
static int
split_pwrite (void *handle, const void *buf, uint32_t count, uint64_t offset)
{
struct handle *h = handle;
while (count > 0) {
struct file *file = get_file (h, offset);
uint64_t foffs = offset - file->offset;
uint64_t max;
ssize_t r;
max = file->size - foffs;
if (max > count)
max = count;
r = pwrite (file->fd, buf, max, offset);
if (r == -1) {
nbdkit_error ("pwrite: %m");
return -1;
}
buf += r;
count -= r;
offset += r;
}
return 0;
}
#if HAVE_POSIX_FADVISE
/* Caching. */
static int
split_cache (void *handle, uint32_t count, uint64_t offset, uint32_t flags)
{
struct handle *h = handle;
/* Cache is advisory, we don't care if this fails */
while (count > 0) {
struct file *file = get_file (h, offset);
uint64_t foffs = offset - file->offset;
uint64_t max;
int r;
max = file->size - foffs;
if (max > count)
max = count;
r = posix_fadvise (file->fd, offset, max, POSIX_FADV_WILLNEED);
if (r) {
errno = r;
nbdkit_error ("posix_fadvise: %m");
return -1;
}
count -= r;
offset += r;
}
return 0;
}
#endif /* HAVE_POSIX_FADVISE */
#ifdef SEEK_HOLE
static int64_t
do_extents (struct file *file, uint32_t count, uint64_t offset,
bool req_one, struct nbdkit_extents *extents)
{
int64_t r = 0;
uint64_t end = offset + count;
do {
off_t pos;
pos = lseek (file->fd, offset, SEEK_DATA);
if (pos == -1) {
if (errno == ENXIO) {
/* The current man page does not describe this situation well,
* but a proposed change to POSIX adds these words for ENXIO:
* "or the whence argument is SEEK_DATA and the offset falls
* within the final hole of the file."
*/
pos = end;
}
else {
nbdkit_error ("lseek: SEEK_DATA: %" PRIu64 ": %m", offset);
return -1;
}
}
/* We know there is a hole from offset to pos-1. */
if (pos > offset) {
if (nbdkit_add_extent (extents, offset + file->offset, pos - offset,
NBDKIT_EXTENT_HOLE | NBDKIT_EXTENT_ZERO) == -1)
return -1;
r += pos - offset;
if (req_one)
break;
}
offset = pos;
if (offset >= end)
break;
pos = lseek (file->fd, offset, SEEK_HOLE);
if (pos == -1) {
nbdkit_error ("lseek: SEEK_HOLE: %" PRIu64 ": %m", offset);
return -1;
}
/* We know there is data from offset to pos-1. */
if (pos > offset) {
if (nbdkit_add_extent (extents, offset + file->offset, pos - offset,
0 /* allocated data */) == -1)
return -1;
r += pos - offset;
if (req_one)
break;
}
offset = pos;
} while (offset < end);
return r;
}
static int
split_extents (void *handle, uint32_t count, uint64_t offset,
uint32_t flags, struct nbdkit_extents *extents)
{
struct handle *h = handle;
const bool req_one = flags & NBDKIT_FLAG_REQ_ONE;
while (count > 0) {
struct file *file = get_file (h, offset);
uint64_t foffs = offset - file->offset;
uint64_t max;
int64_t r;
max = file->size - foffs;
if (max > count)
max = count;
if (file->can_extents) {
ACQUIRE_LOCK_FOR_CURRENT_SCOPE (&lseek_lock);
max = r = do_extents (file, max, foffs, req_one, extents);
}
else
r = nbdkit_add_extent (extents, offset, max, 0 /* allocated data */);
if (r == -1)
return -1;
count -= max;
offset += max;
if (req_one)
break;
}
return 0;
}
#endif /* SEEK_HOLE */
static struct nbdkit_plugin plugin = {
.name = "split",
.version = PACKAGE_VERSION,
.unload = split_unload,
.config = split_config,
.config_help = split_config_help,
.magic_config_key = "file",
.open = split_open,
.close = split_close,
.get_size = split_get_size,
.can_cache = split_can_cache,
.pread = split_pread,
.pwrite = split_pwrite,
#if HAVE_POSIX_FADVISE
.cache = split_cache,
#endif
#ifdef SEEK_HOLE
.extents = split_extents,
#endif
/* In this plugin, errno is preserved properly along error return
* paths from failed system calls.
*/
.errno_is_preserved = 1,
};
NBDKIT_REGISTER_PLUGIN (plugin)
|