?  search:
summary | log | graphiclog1 | graphiclog2 | commit | commitdiff | tree | refs | edit | fork
blame | history | raw | HEAD
GCC: Fixes for gcc 14 on Fedora 40
[zfs.git] / module / zfs / vdev_raidz.c
blob34ef57a3548d6cede1017fe2ca97cda6a63fa4df
1 /*
2 * CDDL HEADER START
3 *
4 * The contents of this file are subject to the terms of the
5 * Common Development and Distribution License (the "License").
6 * You may not use this file except in compliance with the License.
7 *
8 * You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE
9 * or https://opensource.org/licenses/CDDL-1.0.
10 * See the License for the specific language governing permissions
11 * and limitations under the License.
12 *
13 * When distributing Covered Code, include this CDDL HEADER in each
14 * file and include the License file at usr/src/OPENSOLARIS.LICENSE.
15 * If applicable, add the following below this CDDL HEADER, with the
16 * fields enclosed by brackets "[]" replaced with your own identifying
17 * information: Portions Copyright [yyyy] [name of copyright owner]
18 *
19 * CDDL HEADER END
20 */
21
22 /*
23 * Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved.
24 * Copyright (c) 2012, 2020 by Delphix. All rights reserved.
25 * Copyright (c) 2016 Gvozden Nešković. All rights reserved.
26 */
27
28 #include <sys/zfs_context.h>
29 #include <sys/spa.h>
30 #include <sys/vdev_impl.h>
31 #include <sys/zio.h>
32 #include <sys/zio_checksum.h>
33 #include <sys/abd.h>
34 #include <sys/fs/zfs.h>
35 #include <sys/fm/fs/zfs.h>
36 #include <sys/vdev_raidz.h>
37 #include <sys/vdev_raidz_impl.h>
38 #include <sys/vdev_draid.h>
39
40 #ifdef ZFS_DEBUG
41 #include <sys/vdev.h> /* For vdev_xlate() in vdev_raidz_io_verify() */
42 #endif
43
44 /*
45 * Virtual device vector for RAID-Z.
46 *
47 * This vdev supports single, double, and triple parity. For single parity,
48 * we use a simple XOR of all the data columns. For double or triple parity,
49 * we use a special case of Reed-Solomon coding. This extends the
50 * technique described in "The mathematics of RAID-6" by H. Peter Anvin by
51 * drawing on the system described in "A Tutorial on Reed-Solomon Coding for
52 * Fault-Tolerance in RAID-like Systems" by James S. Plank on which the
53 * former is also based. The latter is designed to provide higher performance
54 * for writes.
55 *
56 * Note that the Plank paper claimed to support arbitrary N+M, but was then
57 * amended six years later identifying a critical flaw that invalidates its
58 * claims. Nevertheless, the technique can be adapted to work for up to
59 * triple parity. For additional parity, the amendment "Note: Correction to
60 * the 1997 Tutorial on Reed-Solomon Coding" by James S. Plank and Ying Ding
61 * is viable, but the additional complexity means that write performance will
62 * suffer.
63 *
64 * All of the methods above operate on a Galois field, defined over the
65 * integers mod 2^N. In our case we choose N=8 for GF(8) so that all elements
66 * can be expressed with a single byte. Briefly, the operations on the
67 * field are defined as follows:
68 *
69 * o addition (+) is represented by a bitwise XOR
70 * o subtraction (-) is therefore identical to addition: A + B = A - B
71 * o multiplication of A by 2 is defined by the following bitwise expression:
72 *
73 * (A * 2)_7 = A_6
74 * (A * 2)_6 = A_5
75 * (A * 2)_5 = A_4
76 * (A * 2)_4 = A_3 + A_7
77 * (A * 2)_3 = A_2 + A_7
78 * (A * 2)_2 = A_1 + A_7
79 * (A * 2)_1 = A_0
80 * (A * 2)_0 = A_7
81 *
82 * In C, multiplying by 2 is therefore ((a << 1) ^ ((a & 0x80) ? 0x1d : 0)).
83 * As an aside, this multiplication is derived from the error correcting
84 * primitive polynomial x^8 + x^4 + x^3 + x^2 + 1.
85 *
86 * Observe that any number in the field (except for 0) can be expressed as a
87 * power of 2 -- a generator for the field. We store a table of the powers of
88 * 2 and logs base 2 for quick look ups, and exploit the fact that A * B can
89 * be rewritten as 2^(log_2(A) + log_2(B)) (where '+' is normal addition rather
90 * than field addition). The inverse of a field element A (A^-1) is therefore
91 * A ^ (255 - 1) = A^254.
92 *
93 * The up-to-three parity columns, P, Q, R over several data columns,
94 * D_0, ... D_n-1, can be expressed by field operations:
95 *
96 * P = D_0 + D_1 + ... + D_n-2 + D_n-1
97 * Q = 2^n-1 * D_0 + 2^n-2 * D_1 + ... + 2^1 * D_n-2 + 2^0 * D_n-1
98 * = ((...((D_0) * 2 + D_1) * 2 + ...) * 2 + D_n-2) * 2 + D_n-1
99 * R = 4^n-1 * D_0 + 4^n-2 * D_1 + ... + 4^1 * D_n-2 + 4^0 * D_n-1
100 * = ((...((D_0) * 4 + D_1) * 4 + ...) * 4 + D_n-2) * 4 + D_n-1
101 *
102 * We chose 1, 2, and 4 as our generators because 1 corresponds to the trivial
103 * XOR operation, and 2 and 4 can be computed quickly and generate linearly-
104 * independent coefficients. (There are no additional coefficients that have
105 * this property which is why the uncorrected Plank method breaks down.)
106 *
107 * See the reconstruction code below for how P, Q and R can used individually
108 * or in concert to recover missing data columns.
109 */
110
111 #define VDEV_RAIDZ_P 0
112 #define VDEV_RAIDZ_Q 1
113 #define VDEV_RAIDZ_R 2
114
115 #define VDEV_RAIDZ_MUL_2(x) (((x) << 1) ^ (((x) & 0x80) ? 0x1d : 0))
116 #define VDEV_RAIDZ_MUL_4(x) (VDEV_RAIDZ_MUL_2(VDEV_RAIDZ_MUL_2(x)))
117
118 /*
119 * We provide a mechanism to perform the field multiplication operation on a
120 * 64-bit value all at once rather than a byte at a time. This works by
121 * creating a mask from the top bit in each byte and using that to
122 * conditionally apply the XOR of 0x1d.
123 */
124 #define VDEV_RAIDZ_64MUL_2(x, mask) \
125 { \
126 (mask) = (x) & 0x8080808080808080ULL; \
127 (mask) = ((mask) << 1) - ((mask) >> 7); \
128 (x) = (((x) << 1) & 0xfefefefefefefefeULL) ^ \
129 ((mask) & 0x1d1d1d1d1d1d1d1dULL); \
130 }
131
132 #define VDEV_RAIDZ_64MUL_4(x, mask) \
133 { \
134 VDEV_RAIDZ_64MUL_2((x), mask); \
135 VDEV_RAIDZ_64MUL_2((x), mask); \
136 }
137
138 static void
139 vdev_raidz_row_free(raidz_row_t *rr)
140 {
141 for (int c = 0; c < rr->rr_cols; c++) {
142 raidz_col_t *rc = &rr->rr_col[c];
143
144 if (rc->rc_size != 0)
145 abd_free(rc->rc_abd);
146 if (rc->rc_orig_data != NULL)
147 abd_free(rc->rc_orig_data);
148 }
149
150 if (rr->rr_abd_empty != NULL)
151 abd_free(rr->rr_abd_empty);
152
153 kmem_free(rr, offsetof(raidz_row_t, rr_col[rr->rr_scols]));
154 }
155
156 void
157 vdev_raidz_map_free(raidz_map_t *rm)
158 {
159 for (int i = 0; i < rm->rm_nrows; i++)
160 vdev_raidz_row_free(rm->rm_row[i]);
161
162 kmem_free(rm, offsetof(raidz_map_t, rm_row[rm->rm_nrows]));
163 }
164
165 static void
166 vdev_raidz_map_free_vsd(zio_t *zio)
167 {
168 raidz_map_t *rm = zio->io_vsd;
169
170 vdev_raidz_map_free(rm);
171 }
172
173 const zio_vsd_ops_t vdev_raidz_vsd_ops = {
174 .vsd_free = vdev_raidz_map_free_vsd,
175 };
176
177 static void
178 vdev_raidz_map_alloc_write(zio_t *zio, raidz_map_t *rm, uint64_t ashift)
179 {
180 int c;
181 int nwrapped = 0;
182 uint64_t off = 0;
183 raidz_row_t *rr = rm->rm_row[0];
184
185 ASSERT3U(zio->io_type, ==, ZIO_TYPE_WRITE);
186 ASSERT3U(rm->rm_nrows, ==, 1);
187
188 /*
189 * Pad any parity columns with additional space to account for skip
190 * sectors.
191 */
192 if (rm->rm_skipstart < rr->rr_firstdatacol) {
193 ASSERT0(rm->rm_skipstart);
194 nwrapped = rm->rm_nskip;
195 } else if (rr->rr_scols < (rm->rm_skipstart + rm->rm_nskip)) {
196 nwrapped =
197 (rm->rm_skipstart + rm->rm_nskip) % rr->rr_scols;
198 }
199
200 /*
201 * Optional single skip sectors (rc_size == 0) will be handled in
202 * vdev_raidz_io_start_write().
203 */
204 int skipped = rr->rr_scols - rr->rr_cols;
205
206 /* Allocate buffers for the parity columns */
207 for (c = 0; c < rr->rr_firstdatacol; c++) {
208 raidz_col_t *rc = &rr->rr_col[c];
209
210 /*
211 * Parity columns will pad out a linear ABD to account for
212 * the skip sector. A linear ABD is used here because
213 * parity calculations use the ABD buffer directly to calculate
214 * parity. This avoids doing a memcpy back to the ABD after the
215 * parity has been calculated. By issuing the parity column
216 * with the skip sector we can reduce contention on the child
217 * VDEV queue locks (vq_lock).
218 */
219 if (c < nwrapped) {
220 rc->rc_abd = abd_alloc_linear(
221 rc->rc_size + (1ULL << ashift), B_FALSE);
222 abd_zero_off(rc->rc_abd, rc->rc_size, 1ULL << ashift);
223 skipped++;
224 } else {
225 rc->rc_abd = abd_alloc_linear(rc->rc_size, B_FALSE);
226 }
227 }
228
229 for (off = 0; c < rr->rr_cols; c++) {
230 raidz_col_t *rc = &rr->rr_col[c];
231 abd_t *abd = abd_get_offset_struct(&rc->rc_abdstruct,
232 zio->io_abd, off, rc->rc_size);
233
234 /*
235 * Generate I/O for skip sectors to improve aggregation
236 * continuity. We will use gang ABD's to reduce contention
237 * on the child VDEV queue locks (vq_lock) by issuing
238 * a single I/O that contains the data and skip sector.
239 *
240 * It is important to make sure that rc_size is not updated
241 * even though we are adding a skip sector to the ABD. When
242 * calculating the parity in vdev_raidz_generate_parity_row()
243 * the rc_size is used to iterate through the ABD's. We can
244 * not have zero'd out skip sectors used for calculating
245 * parity for raidz, because those same sectors are not used
246 * during reconstruction.
247 */
248 if (c >= rm->rm_skipstart && skipped < rm->rm_nskip) {
249 rc->rc_abd = abd_alloc_gang();
250 abd_gang_add(rc->rc_abd, abd, B_TRUE);
251 abd_gang_add(rc->rc_abd,
252 abd_get_zeros(1ULL << ashift), B_TRUE);
253 skipped++;
254 } else {
255 rc->rc_abd = abd;
256 }
257 off += rc->rc_size;
258 }
259
260 ASSERT3U(off, ==, zio->io_size);
261 ASSERT3S(skipped, ==, rm->rm_nskip);
262 }
263
264 static void
265 vdev_raidz_map_alloc_read(zio_t *zio, raidz_map_t *rm)
266 {
267 int c;
268 raidz_row_t *rr = rm->rm_row[0];
269
270 ASSERT3U(rm->rm_nrows, ==, 1);
271
272 /* Allocate buffers for the parity columns */
273 for (c = 0; c < rr->rr_firstdatacol; c++)
274 rr->rr_col[c].rc_abd =
275 abd_alloc_linear(rr->rr_col[c].rc_size, B_FALSE);
276
277 for (uint64_t off = 0; c < rr->rr_cols; c++) {
278 raidz_col_t *rc = &rr->rr_col[c];
279 rc->rc_abd = abd_get_offset_struct(&rc->rc_abdstruct,
280 zio->io_abd, off, rc->rc_size);
281 off += rc->rc_size;
282 }
283 }
284
285 /*
286 * Divides the IO evenly across all child vdevs; usually, dcols is
287 * the number of children in the target vdev.
288 *
289 * Avoid inlining the function to keep vdev_raidz_io_start(), which
290 * is this functions only caller, as small as possible on the stack.
291 */
292 noinline raidz_map_t *
293 vdev_raidz_map_alloc(zio_t *zio, uint64_t ashift, uint64_t dcols,
294 uint64_t nparity)
295 {
296 raidz_row_t *rr;
297 /* The starting RAIDZ (parent) vdev sector of the block. */
298 uint64_t b = zio->io_offset >> ashift;
299 /* The zio's size in units of the vdev's minimum sector size. */
300 uint64_t s = zio->io_size >> ashift;
301 /* The first column for this stripe. */
302 uint64_t f = b % dcols;
303 /* The starting byte offset on each child vdev. */
304 uint64_t o = (b / dcols) << ashift;
305 uint64_t q, r, c, bc, col, acols, scols, coff, devidx, asize, tot;
306
307 raidz_map_t *rm =
308 kmem_zalloc(offsetof(raidz_map_t, rm_row[1]), KM_SLEEP);
309 rm->rm_nrows = 1;
310
311 /*
312 * "Quotient": The number of data sectors for this stripe on all but
313 * the "big column" child vdevs that also contain "remainder" data.
314 */
315 q = s / (dcols - nparity);
316
317 /*
318 * "Remainder": The number of partial stripe data sectors in this I/O.
319 * This will add a sector to some, but not all, child vdevs.
320 */
321 r = s - q * (dcols - nparity);
322
323 /* The number of "big columns" - those which contain remainder data. */
324 bc = (r == 0 ? 0 : r + nparity);
325
326 /*
327 * The total number of data and parity sectors associated with
328 * this I/O.
329 */
330 tot = s + nparity * (q + (r == 0 ? 0 : 1));
331
332 /*
333 * acols: The columns that will be accessed.
334 * scols: The columns that will be accessed or skipped.
335 */
336 if (q == 0) {
337 /* Our I/O request doesn't span all child vdevs. */
338 acols = bc;
339 scols = MIN(dcols, roundup(bc, nparity + 1));
340 } else {
341 acols = dcols;
342 scols = dcols;
343 }
344
345 ASSERT3U(acols, <=, scols);
346
347 rr = kmem_alloc(offsetof(raidz_row_t, rr_col[scols]), KM_SLEEP);
348 rm->rm_row[0] = rr;
349
350 rr->rr_cols = acols;
351 rr->rr_scols = scols;
352 rr->rr_bigcols = bc;
353 rr->rr_missingdata = 0;
354 rr->rr_missingparity = 0;
355 rr->rr_firstdatacol = nparity;
356 rr->rr_abd_empty = NULL;
357 rr->rr_nempty = 0;
358 #ifdef ZFS_DEBUG
359 rr->rr_offset = zio->io_offset;
360 rr->rr_size = zio->io_size;
361 #endif
362
363 asize = 0;
364
365 for (c = 0; c < scols; c++) {
366 raidz_col_t *rc = &rr->rr_col[c];
367 col = f + c;
368 coff = o;
369 if (col >= dcols) {
370 col -= dcols;
371 coff += 1ULL << ashift;
372 }
373 rc->rc_devidx = col;
374 rc->rc_offset = coff;
375 rc->rc_abd = NULL;
376 rc->rc_orig_data = NULL;
377 rc->rc_error = 0;
378 rc->rc_tried = 0;
379 rc->rc_skipped = 0;
380 rc->rc_force_repair = 0;
381 rc->rc_allow_repair = 1;
382 rc->rc_need_orig_restore = B_FALSE;
383
384 if (c >= acols)
385 rc->rc_size = 0;
386 else if (c < bc)
387 rc->rc_size = (q + 1) << ashift;
388 else
389 rc->rc_size = q << ashift;
390
391 asize += rc->rc_size;
392 }
393
394 ASSERT3U(asize, ==, tot << ashift);
395 rm->rm_nskip = roundup(tot, nparity + 1) - tot;
396 rm->rm_skipstart = bc;
397
398 /*
399 * If all data stored spans all columns, there's a danger that parity
400 * will always be on the same device and, since parity isn't read
401 * during normal operation, that device's I/O bandwidth won't be
402 * used effectively. We therefore switch the parity every 1MB.
403 *
404 * ... at least that was, ostensibly, the theory. As a practical
405 * matter unless we juggle the parity between all devices evenly, we
406 * won't see any benefit. Further, occasional writes that aren't a
407 * multiple of the LCM of the number of children and the minimum
408 * stripe width are sufficient to avoid pessimal behavior.
409 * Unfortunately, this decision created an implicit on-disk format
410 * requirement that we need to support for all eternity, but only
411 * for single-parity RAID-Z.
412 *
413 * If we intend to skip a sector in the zeroth column for padding
414 * we must make sure to note this swap. We will never intend to
415 * skip the first column since at least one data and one parity
416 * column must appear in each row.
417 */
418 ASSERT(rr->rr_cols >= 2);
419 ASSERT(rr->rr_col[0].rc_size == rr->rr_col[1].rc_size);
420
421 if (rr->rr_firstdatacol == 1 && (zio->io_offset & (1ULL << 20))) {
422 devidx = rr->rr_col[0].rc_devidx;
423 o = rr->rr_col[0].rc_offset;
424 rr->rr_col[0].rc_devidx = rr->rr_col[1].rc_devidx;
425 rr->rr_col[0].rc_offset = rr->rr_col[1].rc_offset;
426 rr->rr_col[1].rc_devidx = devidx;
427 rr->rr_col[1].rc_offset = o;
428
429 if (rm->rm_skipstart == 0)
430 rm->rm_skipstart = 1;
431 }
432
433 if (zio->io_type == ZIO_TYPE_WRITE) {
434 vdev_raidz_map_alloc_write(zio, rm, ashift);
435 } else {
436 vdev_raidz_map_alloc_read(zio, rm);
437 }
438
439 /* init RAIDZ parity ops */
440 rm->rm_ops = vdev_raidz_math_get_ops();
441
442 return (rm);
443 }
444
445 struct pqr_struct {
446 uint64_t *p;
447 uint64_t *q;
448 uint64_t *r;
449 };
450
451 static int
452 vdev_raidz_p_func(void *buf, size_t size, void *private)
453 {
454 struct pqr_struct *pqr = private;
455 const uint64_t *src = buf;
456 int i, cnt = size / sizeof (src[0]);
457
458 ASSERT(pqr->p && !pqr->q && !pqr->r);
459
460 for (i = 0; i < cnt; i++, src++, pqr->p++)
461 *pqr->p ^= *src;
462
463 return (0);
464 }
465
466 static int
467 vdev_raidz_pq_func(void *buf, size_t size, void *private)
468 {
469 struct pqr_struct *pqr = private;
470 const uint64_t *src = buf;
471 uint64_t mask;
472 int i, cnt = size / sizeof (src[0]);
473
474 ASSERT(pqr->p && pqr->q && !pqr->r);
475
476 for (i = 0; i < cnt; i++, src++, pqr->p++, pqr->q++) {
477 *pqr->p ^= *src;
478 VDEV_RAIDZ_64MUL_2(*pqr->q, mask);
479 *pqr->q ^= *src;
480 }
481
482 return (0);
483 }
484
485 static int
486 vdev_raidz_pqr_func(void *buf, size_t size, void *private)
487 {
488 struct pqr_struct *pqr = private;
489 const uint64_t *src = buf;
490 uint64_t mask;
491 int i, cnt = size / sizeof (src[0]);
492
493 ASSERT(pqr->p && pqr->q && pqr->r);
494
495 for (i = 0; i < cnt; i++, src++, pqr->p++, pqr->q++, pqr->r++) {
496 *pqr->p ^= *src;
497 VDEV_RAIDZ_64MUL_2(*pqr->q, mask);
498 *pqr->q ^= *src;
499 VDEV_RAIDZ_64MUL_4(*pqr->r, mask);
500 *pqr->r ^= *src;
501 }
502
503 return (0);
504 }
505
506 static void
507 vdev_raidz_generate_parity_p(raidz_row_t *rr)
508 {
509 uint64_t *p = abd_to_buf(rr->rr_col[VDEV_RAIDZ_P].rc_abd);
510
511 for (int c = rr->rr_firstdatacol; c < rr->rr_cols; c++) {
512 abd_t *src = rr->rr_col[c].rc_abd;
513
514 if (c == rr->rr_firstdatacol) {
515 abd_copy_to_buf(p, src, rr->rr_col[c].rc_size);
516 } else {
517 struct pqr_struct pqr = { p, NULL, NULL };
518 (void) abd_iterate_func(src, 0, rr->rr_col[c].rc_size,
519 vdev_raidz_p_func, &pqr);
520 }
521 }
522 }
523
524 static void
525 vdev_raidz_generate_parity_pq(raidz_row_t *rr)
526 {
527 uint64_t *p = abd_to_buf(rr->rr_col[VDEV_RAIDZ_P].rc_abd);
528 uint64_t *q = abd_to_buf(rr->rr_col[VDEV_RAIDZ_Q].rc_abd);
529 uint64_t pcnt = rr->rr_col[VDEV_RAIDZ_P].rc_size / sizeof (p[0]);
530 ASSERT(rr->rr_col[VDEV_RAIDZ_P].rc_size ==
531 rr->rr_col[VDEV_RAIDZ_Q].rc_size);
532
533 for (int c = rr->rr_firstdatacol; c < rr->rr_cols; c++) {
534 abd_t *src = rr->rr_col[c].rc_abd;
535
536 uint64_t ccnt = rr->rr_col[c].rc_size / sizeof (p[0]);
537
538 if (c == rr->rr_firstdatacol) {
539 ASSERT(ccnt == pcnt || ccnt == 0);
540 abd_copy_to_buf(p, src, rr->rr_col[c].rc_size);
541 (void) memcpy(q, p, rr->rr_col[c].rc_size);
542
543 for (uint64_t i = ccnt; i < pcnt; i++) {
544 p[i] = 0;
545 q[i] = 0;
546 }
547 } else {
548 struct pqr_struct pqr = { p, q, NULL };
549
550 ASSERT(ccnt <= pcnt);
551 (void) abd_iterate_func(src, 0, rr->rr_col[c].rc_size,
552 vdev_raidz_pq_func, &pqr);
553
554 /*
555 * Treat short columns as though they are full of 0s.
556 * Note that there's therefore nothing needed for P.
557 */
558 uint64_t mask;
559 for (uint64_t i = ccnt; i < pcnt; i++) {
560 VDEV_RAIDZ_64MUL_2(q[i], mask);
561 }
562 }
563 }
564 }
565
566 static void
567 vdev_raidz_generate_parity_pqr(raidz_row_t *rr)
568 {
569 uint64_t *p = abd_to_buf(rr->rr_col[VDEV_RAIDZ_P].rc_abd);
570 uint64_t *q = abd_to_buf(rr->rr_col[VDEV_RAIDZ_Q].rc_abd);
571 uint64_t *r = abd_to_buf(rr->rr_col[VDEV_RAIDZ_R].rc_abd);
572 uint64_t pcnt = rr->rr_col[VDEV_RAIDZ_P].rc_size / sizeof (p[0]);
573 ASSERT(rr->rr_col[VDEV_RAIDZ_P].rc_size ==
574 rr->rr_col[VDEV_RAIDZ_Q].rc_size);
575 ASSERT(rr->rr_col[VDEV_RAIDZ_P].rc_size ==
576 rr->rr_col[VDEV_RAIDZ_R].rc_size);
577
578 for (int c = rr->rr_firstdatacol; c < rr->rr_cols; c++) {
579 abd_t *src = rr->rr_col[c].rc_abd;
580
581 uint64_t ccnt = rr->rr_col[c].rc_size / sizeof (p[0]);
582
583 if (c == rr->rr_firstdatacol) {
584 ASSERT(ccnt == pcnt || ccnt == 0);
585 abd_copy_to_buf(p, src, rr->rr_col[c].rc_size);
586 (void) memcpy(q, p, rr->rr_col[c].rc_size);
587 (void) memcpy(r, p, rr->rr_col[c].rc_size);
588
589 for (uint64_t i = ccnt; i < pcnt; i++) {
590 p[i] = 0;
591 q[i] = 0;
592 r[i] = 0;
593 }
594 } else {
595 struct pqr_struct pqr = { p, q, r };
596
597 ASSERT(ccnt <= pcnt);
598 (void) abd_iterate_func(src, 0, rr->rr_col[c].rc_size,
599 vdev_raidz_pqr_func, &pqr);
600
601 /*
602 * Treat short columns as though they are full of 0s.
603 * Note that there's therefore nothing needed for P.
604 */
605 uint64_t mask;
606 for (uint64_t i = ccnt; i < pcnt; i++) {
607 VDEV_RAIDZ_64MUL_2(q[i], mask);
608 VDEV_RAIDZ_64MUL_4(r[i], mask);
609 }
610 }
611 }
612 }
613
614 /*
615 * Generate RAID parity in the first virtual columns according to the number of
616 * parity columns available.
617 */
618 void
619 vdev_raidz_generate_parity_row(raidz_map_t *rm, raidz_row_t *rr)
620 {
621 ASSERT3U(rr->rr_cols, !=, 0);
622
623 /* Generate using the new math implementation */
624 if (vdev_raidz_math_generate(rm, rr) != RAIDZ_ORIGINAL_IMPL)
625 return;
626
627 switch (rr->rr_firstdatacol) {
628 case 1:
629 vdev_raidz_generate_parity_p(rr);
630 break;
631 case 2:
632 vdev_raidz_generate_parity_pq(rr);
633 break;
634 case 3:
635 vdev_raidz_generate_parity_pqr(rr);
636 break;
637 default:
638 cmn_err(CE_PANIC, "invalid RAID-Z configuration");
639 }
640 }
641
642 void
643 vdev_raidz_generate_parity(raidz_map_t *rm)
644 {
645 for (int i = 0; i < rm->rm_nrows; i++) {
646 raidz_row_t *rr = rm->rm_row[i];
647 vdev_raidz_generate_parity_row(rm, rr);
648 }
649 }
650
651 static int
652 vdev_raidz_reconst_p_func(void *dbuf, void *sbuf, size_t size, void *private)
653 {
654 (void) private;
655 uint64_t *dst = dbuf;
656 uint64_t *src = sbuf;
657 int cnt = size / sizeof (src[0]);
658
659 for (int i = 0; i < cnt; i++) {
660 dst[i] ^= src[i];
661 }
662
663 return (0);
664 }
665
666 static int
667 vdev_raidz_reconst_q_pre_func(void *dbuf, void *sbuf, size_t size,
668 void *private)
669 {
670 (void) private;
671 uint64_t *dst = dbuf;
672 uint64_t *src = sbuf;
673 uint64_t mask;
674 int cnt = size / sizeof (dst[0]);
675
676 for (int i = 0; i < cnt; i++, dst++, src++) {
677 VDEV_RAIDZ_64MUL_2(*dst, mask);
678 *dst ^= *src;
679 }
680
681 return (0);
682 }
683
684 static int
685 vdev_raidz_reconst_q_pre_tail_func(void *buf, size_t size, void *private)
686 {
687 (void) private;
688 uint64_t *dst = buf;
689 uint64_t mask;
690 int cnt = size / sizeof (dst[0]);
691
692 for (int i = 0; i < cnt; i++, dst++) {
693 /* same operation as vdev_raidz_reconst_q_pre_func() on dst */
694 VDEV_RAIDZ_64MUL_2(*dst, mask);
695 }
696
697 return (0);
698 }
699
700 struct reconst_q_struct {
701 uint64_t *q;
702 int exp;
703 };
704
705 static int
706 vdev_raidz_reconst_q_post_func(void *buf, size_t size, void *private)
707 {
708 struct reconst_q_struct *rq = private;
709 uint64_t *dst = buf;
710 int cnt = size / sizeof (dst[0]);
711
712 for (int i = 0; i < cnt; i++, dst++, rq->q++) {
713 int j;
714 uint8_t *b;
715
716 *dst ^= *rq->q;
717 for (j = 0, b = (uint8_t *)dst; j < 8; j++, b++) {
718 *b = vdev_raidz_exp2(*b, rq->exp);
719 }
720 }
721
722 return (0);
723 }
724
725 struct reconst_pq_struct {
726 uint8_t *p;
727 uint8_t *q;
728 uint8_t *pxy;
729 uint8_t *qxy;
730 int aexp;
731 int bexp;
732 };
733
734 static int
735 vdev_raidz_reconst_pq_func(void *xbuf, void *ybuf, size_t size, void *private)
736 {
737 struct reconst_pq_struct *rpq = private;
738 uint8_t *xd = xbuf;
739 uint8_t *yd = ybuf;
740
741 for (int i = 0; i < size;
742 i++, rpq->p++, rpq->q++, rpq->pxy++, rpq->qxy++, xd++, yd++) {
743 *xd = vdev_raidz_exp2(*rpq->p ^ *rpq->pxy, rpq->aexp) ^
744 vdev_raidz_exp2(*rpq->q ^ *rpq->qxy, rpq->bexp);
745 *yd = *rpq->p ^ *rpq->pxy ^ *xd;
746 }
747
748 return (0);
749 }
750
751 static int
752 vdev_raidz_reconst_pq_tail_func(void *xbuf, size_t size, void *private)
753 {
754 struct reconst_pq_struct *rpq = private;
755 uint8_t *xd = xbuf;
756
757 for (int i = 0; i < size;
758 i++, rpq->p++, rpq->q++, rpq->pxy++, rpq->qxy++, xd++) {
759 /* same operation as vdev_raidz_reconst_pq_func() on xd */
760 *xd = vdev_raidz_exp2(*rpq->p ^ *rpq->pxy, rpq->aexp) ^
761 vdev_raidz_exp2(*rpq->q ^ *rpq->qxy, rpq->bexp);
762 }
763
764 return (0);
765 }
766
767 static void
768 vdev_raidz_reconstruct_p(raidz_row_t *rr, int *tgts, int ntgts)
769 {
770 int x = tgts[0];
771 abd_t *dst, *src;
772
773 ASSERT3U(ntgts, ==, 1);
774 ASSERT3U(x, >=, rr->rr_firstdatacol);
775 ASSERT3U(x, <, rr->rr_cols);
776
777 ASSERT3U(rr->rr_col[x].rc_size, <=, rr->rr_col[VDEV_RAIDZ_P].rc_size);
778
779 src = rr->rr_col[VDEV_RAIDZ_P].rc_abd;
780 dst = rr->rr_col[x].rc_abd;
781
782 abd_copy_from_buf(dst, abd_to_buf(src), rr->rr_col[x].rc_size);
783
784 for (int c = rr->rr_firstdatacol; c < rr->rr_cols; c++) {
785 uint64_t size = MIN(rr->rr_col[x].rc_size,
786 rr->rr_col[c].rc_size);
787
788 src = rr->rr_col[c].rc_abd;
789
790 if (c == x)
791 continue;
792
793 (void) abd_iterate_func2(dst, src, 0, 0, size,
794 vdev_raidz_reconst_p_func, NULL);
795 }
796 }
797
798 static void
799 vdev_raidz_reconstruct_q(raidz_row_t *rr, int *tgts, int ntgts)
800 {
801 int x = tgts[0];
802 int c, exp;
803 abd_t *dst, *src;
804
805 ASSERT(ntgts == 1);
806
807 ASSERT(rr->rr_col[x].rc_size <= rr->rr_col[VDEV_RAIDZ_Q].rc_size);
808
809 for (c = rr->rr_firstdatacol; c < rr->rr_cols; c++) {
810 uint64_t size = (c == x) ? 0 : MIN(rr->rr_col[x].rc_size,
811 rr->rr_col[c].rc_size);
812
813 src = rr->rr_col[c].rc_abd;
814 dst = rr->rr_col[x].rc_abd;
815
816 if (c == rr->rr_firstdatacol) {
817 abd_copy(dst, src, size);
818 if (rr->rr_col[x].rc_size > size) {
819 abd_zero_off(dst, size,
820 rr->rr_col[x].rc_size - size);
821 }
822 } else {
823 ASSERT3U(size, <=, rr->rr_col[x].rc_size);
824 (void) abd_iterate_func2(dst, src, 0, 0, size,
825 vdev_raidz_reconst_q_pre_func, NULL);
826 (void) abd_iterate_func(dst,
827 size, rr->rr_col[x].rc_size - size,
828 vdev_raidz_reconst_q_pre_tail_func, NULL);
829 }
830 }
831
832 src = rr->rr_col[VDEV_RAIDZ_Q].rc_abd;
833 dst = rr->rr_col[x].rc_abd;
834 exp = 255 - (rr->rr_cols - 1 - x);
835
836 struct reconst_q_struct rq = { abd_to_buf(src), exp };
837 (void) abd_iterate_func(dst, 0, rr->rr_col[x].rc_size,
838 vdev_raidz_reconst_q_post_func, &rq);
839 }
840
841 static void
842 vdev_raidz_reconstruct_pq(raidz_row_t *rr, int *tgts, int ntgts)
843 {
844 uint8_t *p, *q, *pxy, *qxy, tmp, a, b, aexp, bexp;
845 abd_t *pdata, *qdata;
846 uint64_t xsize, ysize;
847 int x = tgts[0];
848 int y = tgts[1];
849 abd_t *xd, *yd;
850
851 ASSERT(ntgts == 2);
852 ASSERT(x < y);
853 ASSERT(x >= rr->rr_firstdatacol);
854 ASSERT(y < rr->rr_cols);
855
856 ASSERT(rr->rr_col[x].rc_size >= rr->rr_col[y].rc_size);
857
858 /*
859 * Move the parity data aside -- we're going to compute parity as
860 * though columns x and y were full of zeros -- Pxy and Qxy. We want to
861 * reuse the parity generation mechanism without trashing the actual
862 * parity so we make those columns appear to be full of zeros by
863 * setting their lengths to zero.
864 */
865 pdata = rr->rr_col[VDEV_RAIDZ_P].rc_abd;
866 qdata = rr->rr_col[VDEV_RAIDZ_Q].rc_abd;
867 xsize = rr->rr_col[x].rc_size;
868 ysize = rr->rr_col[y].rc_size;
869
870 rr->rr_col[VDEV_RAIDZ_P].rc_abd =
871 abd_alloc_linear(rr->rr_col[VDEV_RAIDZ_P].rc_size, B_TRUE);
872 rr->rr_col[VDEV_RAIDZ_Q].rc_abd =
873 abd_alloc_linear(rr->rr_col[VDEV_RAIDZ_Q].rc_size, B_TRUE);
874 rr->rr_col[x].rc_size = 0;
875 rr->rr_col[y].rc_size = 0;
876
877 vdev_raidz_generate_parity_pq(rr);
878
879 rr->rr_col[x].rc_size = xsize;
880 rr->rr_col[y].rc_size = ysize;
881
882 p = abd_to_buf(pdata);
883 q = abd_to_buf(qdata);
884 pxy = abd_to_buf(rr->rr_col[VDEV_RAIDZ_P].rc_abd);
885 qxy = abd_to_buf(rr->rr_col[VDEV_RAIDZ_Q].rc_abd);
886 xd = rr->rr_col[x].rc_abd;
887 yd = rr->rr_col[y].rc_abd;
888
889 /*
890 * We now have:
891 * Pxy = P + D_x + D_y
892 * Qxy = Q + 2^(ndevs - 1 - x) * D_x + 2^(ndevs - 1 - y) * D_y
893 *
894 * We can then solve for D_x:
895 * D_x = A * (P + Pxy) + B * (Q + Qxy)
896 * where
897 * A = 2^(x - y) * (2^(x - y) + 1)^-1
898 * B = 2^(ndevs - 1 - x) * (2^(x - y) + 1)^-1
899 *
900 * With D_x in hand, we can easily solve for D_y:
901 * D_y = P + Pxy + D_x
902 */
903
904 a = vdev_raidz_pow2[255 + x - y];
905 b = vdev_raidz_pow2[255 - (rr->rr_cols - 1 - x)];
906 tmp = 255 - vdev_raidz_log2[a ^ 1];
907
908 aexp = vdev_raidz_log2[vdev_raidz_exp2(a, tmp)];
909 bexp = vdev_raidz_log2[vdev_raidz_exp2(b, tmp)];
910
911 ASSERT3U(xsize, >=, ysize);
912 struct reconst_pq_struct rpq = { p, q, pxy, qxy, aexp, bexp };
913
914 (void) abd_iterate_func2(xd, yd, 0, 0, ysize,
915 vdev_raidz_reconst_pq_func, &rpq);
916 (void) abd_iterate_func(xd, ysize, xsize - ysize,
917 vdev_raidz_reconst_pq_tail_func, &rpq);
918
919 abd_free(rr->rr_col[VDEV_RAIDZ_P].rc_abd);
920 abd_free(rr->rr_col[VDEV_RAIDZ_Q].rc_abd);
921
922 /*
923 * Restore the saved parity data.
924 */
925 rr->rr_col[VDEV_RAIDZ_P].rc_abd = pdata;
926 rr->rr_col[VDEV_RAIDZ_Q].rc_abd = qdata;
927 }
928
929 /*
930 * In the general case of reconstruction, we must solve the system of linear
931 * equations defined by the coefficients used to generate parity as well as
932 * the contents of the data and parity disks. This can be expressed with
933 * vectors for the original data (D) and the actual data (d) and parity (p)
934 * and a matrix composed of the identity matrix (I) and a dispersal matrix (V):
935 *
936 * __ __ __ __
937 * | | __ __ | p_0 |
938 * | V | | D_0 | | p_m-1 |
939 * | | x | : | = | d_0 |
940 * | I | | D_n-1 | | : |
941 * | | ~~ ~~ | d_n-1 |
942 * ~~ ~~ ~~ ~~
943 *
944 * I is simply a square identity matrix of size n, and V is a vandermonde
945 * matrix defined by the coefficients we chose for the various parity columns
946 * (1, 2, 4). Note that these values were chosen both for simplicity, speedy
947 * computation as well as linear separability.
948 *
949 * __ __ __ __
950 * | 1 .. 1 1 1 | | p_0 |
951 * | 2^n-1 .. 4 2 1 | __ __ | : |
952 * | 4^n-1 .. 16 4 1 | | D_0 | | p_m-1 |
953 * | 1 .. 0 0 0 | | D_1 | | d_0 |
954 * | 0 .. 0 0 0 | x | D_2 | = | d_1 |
955 * | : : : : | | : | | d_2 |
956 * | 0 .. 1 0 0 | | D_n-1 | | : |
957 * | 0 .. 0 1 0 | ~~ ~~ | : |
958 * | 0 .. 0 0 1 | | d_n-1 |
959 * ~~ ~~ ~~ ~~
960 *
961 * Note that I, V, d, and p are known. To compute D, we must invert the
962 * matrix and use the known data and parity values to reconstruct the unknown
963 * data values. We begin by removing the rows in V|I and d|p that correspond
964 * to failed or missing columns; we then make V|I square (n x n) and d|p
965 * sized n by removing rows corresponding to unused parity from the bottom up
966 * to generate (V|I)' and (d|p)'. We can then generate the inverse of (V|I)'
967 * using Gauss-Jordan elimination. In the example below we use m=3 parity
968 * columns, n=8 data columns, with errors in d_1, d_2, and p_1:
969 * __ __
970 * | 1 1 1 1 1 1 1 1 |
971 * | 128 64 32 16 8 4 2 1 | <-----+-+-- missing disks
972 * | 19 205 116 29 64 16 4 1 | / /
973 * | 1 0 0 0 0 0 0 0 | / /
974 * | 0 1 0 0 0 0 0 0 | <--' /
975 * (V|I) = | 0 0 1 0 0 0 0 0 | <---'
976 * | 0 0 0 1 0 0 0 0 |
977 * | 0 0 0 0 1 0 0 0 |
978 * | 0 0 0 0 0 1 0 0 |
979 * | 0 0 0 0 0 0 1 0 |
980 * | 0 0 0 0 0 0 0 1 |
981 * ~~ ~~
982 * __ __
983 * | 1 1 1 1 1 1 1 1 |
984 * | 128 64 32 16 8 4 2 1 |
985 * | 19 205 116 29 64 16 4 1 |
986 * | 1 0 0 0 0 0 0 0 |
987 * | 0 1 0 0 0 0 0 0 |
988 * (V|I)' = | 0 0 1 0 0 0 0 0 |
989 * | 0 0 0 1 0 0 0 0 |
990 * | 0 0 0 0 1 0 0 0 |
991 * | 0 0 0 0 0 1 0 0 |
992 * | 0 0 0 0 0 0 1 0 |
993 * | 0 0 0 0 0 0 0 1 |
994 * ~~ ~~
995 *
996 * Here we employ Gauss-Jordan elimination to find the inverse of (V|I)'. We
997 * have carefully chosen the seed values 1, 2, and 4 to ensure that this
998 * matrix is not singular.
999 * __ __
1000 * | 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 |
1001 * | 19 205 116 29 64 16 4 1 0 1 0 0 0 0 0 0 |
1002 * | 1 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 |
1003 * | 0 0 0 1 0 0 0 0 0 0 0 1 0 0 0 0 |
1004 * | 0 0 0 0 1 0 0 0 0 0 0 0 1 0 0 0 |
1005 * | 0 0 0 0 0 1 0 0 0 0 0 0 0 1 0 0 |
1006 * | 0 0 0 0 0 0 1 0 0 0 0 0 0 0 1 0 |
1007 * | 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 1 |
1008 * ~~ ~~
1009 * __ __
1010 * | 1 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 |
1011 * | 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 |
1012 * | 19 205 116 29 64 16 4 1 0 1 0 0 0 0 0 0 |
1013 * | 0 0 0 1 0 0 0 0 0 0 0 1 0 0 0 0 |
1014 * | 0 0 0 0 1 0 0 0 0 0 0 0 1 0 0 0 |
1015 * | 0 0 0 0 0 1 0 0 0 0 0 0 0 1 0 0 |
1016 * | 0 0 0 0 0 0 1 0 0 0 0 0 0 0 1 0 |
1017 * | 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 1 |
1018 * ~~ ~~
1019 * __ __
1020 * | 1 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 |
1021 * | 0 1 1 0 0 0 0 0 1 0 1 1 1 1 1 1 |
1022 * | 0 205 116 0 0 0 0 0 0 1 19 29 64 16 4 1 |
1023 * | 0 0 0 1 0 0 0 0 0 0 0 1 0 0 0 0 |
1024 * | 0 0 0 0 1 0 0 0 0 0 0 0 1 0 0 0 |
1025 * | 0 0 0 0 0 1 0 0 0 0 0 0 0 1 0 0 |
1026 * | 0 0 0 0 0 0 1 0 0 0 0 0 0 0 1 0 |
1027 * | 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 1 |
1028 * ~~ ~~
1029 * __ __
1030 * | 1 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 |
1031 * | 0 1 1 0 0 0 0 0 1 0 1 1 1 1 1 1 |
1032 * | 0 0 185 0 0 0 0 0 205 1 222 208 141 221 201 204 |
1033 * | 0 0 0 1 0 0 0 0 0 0 0 1 0 0 0 0 |
1034 * | 0 0 0 0 1 0 0 0 0 0 0 0 1 0 0 0 |
1035 * | 0 0 0 0 0 1 0 0 0 0 0 0 0 1 0 0 |
1036 * | 0 0 0 0 0 0 1 0 0 0 0 0 0 0 1 0 |
1037 * | 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 1 |
1038 * ~~ ~~
1039 * __ __
1040 * | 1 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 |
1041 * | 0 1 1 0 0 0 0 0 1 0 1 1 1 1 1 1 |
1042 * | 0 0 1 0 0 0 0 0 166 100 4 40 158 168 216 209 |
1043 * | 0 0 0 1 0 0 0 0 0 0 0 1 0 0 0 0 |
1044 * | 0 0 0 0 1 0 0 0 0 0 0 0 1 0 0 0 |
1045 * | 0 0 0 0 0 1 0 0 0 0 0 0 0 1 0 0 |
1046 * | 0 0 0 0 0 0 1 0 0 0 0 0 0 0 1 0 |
1047 * | 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 1 |
1048 * ~~ ~~
1049 * __ __
1050 * | 1 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 |
1051 * | 0 1 0 0 0 0 0 0 167 100 5 41 159 169 217 208 |
1052 * | 0 0 1 0 0 0 0 0 166 100 4 40 158 168 216 209 |
1053 * | 0 0 0 1 0 0 0 0 0 0 0 1 0 0 0 0 |
1054 * | 0 0 0 0 1 0 0 0 0 0 0 0 1 0 0 0 |
1055 * | 0 0 0 0 0 1 0 0 0 0 0 0 0 1 0 0 |
1056 * | 0 0 0 0 0 0 1 0 0 0 0 0 0 0 1 0 |
1057 * | 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 1 |
1058 * ~~ ~~
1059 * __ __
1060 * | 0 0 1 0 0 0 0 0 |
1061 * | 167 100 5 41 159 169 217 208 |
1062 * | 166 100 4 40 158 168 216 209 |
1063 * (V|I)'^-1 = | 0 0 0 1 0 0 0 0 |
1064 * | 0 0 0 0 1 0 0 0 |
1065 * | 0 0 0 0 0 1 0 0 |
1066 * | 0 0 0 0 0 0 1 0 |
1067 * | 0 0 0 0 0 0 0 1 |
1068 * ~~ ~~
1069 *
1070 * We can then simply compute D = (V|I)'^-1 x (d|p)' to discover the values
1071 * of the missing data.
1072 *
1073 * As is apparent from the example above, the only non-trivial rows in the
1074 * inverse matrix correspond to the data disks that we're trying to
1075 * reconstruct. Indeed, those are the only rows we need as the others would
1076 * only be useful for reconstructing data known or assumed to be valid. For
1077 * that reason, we only build the coefficients in the rows that correspond to
1078 * targeted columns.
1079 */
1080
1081 static void
1082 vdev_raidz_matrix_init(raidz_row_t *rr, int n, int nmap, int *map,
1083 uint8_t **rows)
1084 {
1085 int i, j;
1086 int pow;
1087
1088 ASSERT(n == rr->rr_cols - rr->rr_firstdatacol);
1089
1090 /*
1091 * Fill in the missing rows of interest.
1092 */
1093 for (i = 0; i < nmap; i++) {
1094 ASSERT3S(0, <=, map[i]);
1095 ASSERT3S(map[i], <=, 2);
1096
1097 pow = map[i] * n;
1098 if (pow > 255)
1099 pow -= 255;
1100 ASSERT(pow <= 255);
1101
1102 for (j = 0; j < n; j++) {
1103 pow -= map[i];
1104 if (pow < 0)
1105 pow += 255;
1106 rows[i][j] = vdev_raidz_pow2[pow];
1107 }
1108 }
1109 }
1110
1111 static void
1112 vdev_raidz_matrix_invert(raidz_row_t *rr, int n, int nmissing, int *missing,
1113 uint8_t **rows, uint8_t **invrows, const uint8_t *used)
1114 {
1115 int i, j, ii, jj;
1116 uint8_t log;
1117
1118 /*
1119 * Assert that the first nmissing entries from the array of used
1120 * columns correspond to parity columns and that subsequent entries
1121 * correspond to data columns.
1122 */
1123 for (i = 0; i < nmissing; i++) {
1124 ASSERT3S(used[i], <, rr->rr_firstdatacol);
1125 }
1126 for (; i < n; i++) {
1127 ASSERT3S(used[i], >=, rr->rr_firstdatacol);
1128 }
1129
1130 /*
1131 * First initialize the storage where we'll compute the inverse rows.
1132 */
1133 for (i = 0; i < nmissing; i++) {
1134 for (j = 0; j < n; j++) {
1135 invrows[i][j] = (i == j) ? 1 : 0;
1136 }
1137 }
1138
1139 /*
1140 * Subtract all trivial rows from the rows of consequence.
1141 */
1142 for (i = 0; i < nmissing; i++) {
1143 for (j = nmissing; j < n; j++) {
1144 ASSERT3U(used[j], >=, rr->rr_firstdatacol);
1145 jj = used[j] - rr->rr_firstdatacol;
1146 ASSERT3S(jj, <, n);
1147 invrows[i][j] = rows[i][jj];
1148 rows[i][jj] = 0;
1149 }
1150 }
1151
1152 /*
1153 * For each of the rows of interest, we must normalize it and subtract
1154 * a multiple of it from the other rows.
1155 */
1156 for (i = 0; i < nmissing; i++) {
1157 for (j = 0; j < missing[i]; j++) {
1158 ASSERT0(rows[i][j]);
1159 }
1160 ASSERT3U(rows[i][missing[i]], !=, 0);
1161
1162 /*
1163 * Compute the inverse of the first element and multiply each
1164 * element in the row by that value.
1165 */
1166 log = 255 - vdev_raidz_log2[rows[i][missing[i]]];
1167
1168 for (j = 0; j < n; j++) {
1169 rows[i][j] = vdev_raidz_exp2(rows[i][j], log);
1170 invrows[i][j] = vdev_raidz_exp2(invrows[i][j], log);
1171 }
1172
1173 for (ii = 0; ii < nmissing; ii++) {
1174 if (i == ii)
1175 continue;
1176
1177 ASSERT3U(rows[ii][missing[i]], !=, 0);
1178
1179 log = vdev_raidz_log2[rows[ii][missing[i]]];
1180
1181 for (j = 0; j < n; j++) {
1182 rows[ii][j] ^=
1183 vdev_raidz_exp2(rows[i][j], log);
1184 invrows[ii][j] ^=
1185 vdev_raidz_exp2(invrows[i][j], log);
1186 }
1187 }
1188 }
1189
1190 /*
1191 * Verify that the data that is left in the rows are properly part of
1192 * an identity matrix.
1193 */
1194 for (i = 0; i < nmissing; i++) {
1195 for (j = 0; j < n; j++) {
1196 if (j == missing[i]) {
1197 ASSERT3U(rows[i][j], ==, 1);
1198 } else {
1199 ASSERT0(rows[i][j]);
1200 }
1201 }
1202 }
1203 }
1204
1205 static void
1206 vdev_raidz_matrix_reconstruct(raidz_row_t *rr, int n, int nmissing,
1207 int *missing, uint8_t **invrows, const uint8_t *used)
1208 {
1209 int i, j, x, cc, c;
1210 uint8_t *src;
1211 uint64_t ccount;
1212 uint8_t *dst[VDEV_RAIDZ_MAXPARITY] = { NULL };
1213 uint64_t dcount[VDEV_RAIDZ_MAXPARITY] = { 0 };
1214 uint8_t log = 0;
1215 uint8_t val;
1216 int ll;
1217 uint8_t *invlog[VDEV_RAIDZ_MAXPARITY];
1218 uint8_t *p, *pp;
1219 size_t psize;
1220
1221 psize = sizeof (invlog[0][0]) * n * nmissing;
1222 p = kmem_alloc(psize, KM_SLEEP);
1223
1224 for (pp = p, i = 0; i < nmissing; i++) {
1225 invlog[i] = pp;
1226 pp += n;
1227 }
1228
1229 for (i = 0; i < nmissing; i++) {
1230 for (j = 0; j < n; j++) {
1231 ASSERT3U(invrows[i][j], !=, 0);
1232 invlog[i][j] = vdev_raidz_log2[invrows[i][j]];
1233 }
1234 }
1235
1236 for (i = 0; i < n; i++) {
1237 c = used[i];
1238 ASSERT3U(c, <, rr->rr_cols);
1239
1240 ccount = rr->rr_col[c].rc_size;
1241 ASSERT(ccount >= rr->rr_col[missing[0]].rc_size || i > 0);
1242 if (ccount == 0)
1243 continue;
1244 src = abd_to_buf(rr->rr_col[c].rc_abd);
1245 for (j = 0; j < nmissing; j++) {
1246 cc = missing[j] + rr->rr_firstdatacol;
1247 ASSERT3U(cc, >=, rr->rr_firstdatacol);
1248 ASSERT3U(cc, <, rr->rr_cols);
1249 ASSERT3U(cc, !=, c);
1250
1251 dcount[j] = rr->rr_col[cc].rc_size;
1252 if (dcount[j] != 0)
1253 dst[j] = abd_to_buf(rr->rr_col[cc].rc_abd);
1254 }
1255
1256 for (x = 0; x < ccount; x++, src++) {
1257 if (*src != 0)
1258 log = vdev_raidz_log2[*src];
1259
1260 for (cc = 0; cc < nmissing; cc++) {
1261 if (x >= dcount[cc])
1262 continue;
1263
1264 if (*src == 0) {
1265 val = 0;
1266 } else {
1267 if ((ll = log + invlog[cc][i]) >= 255)
1268 ll -= 255;
1269 val = vdev_raidz_pow2[ll];
1270 }
1271
1272 if (i == 0)
1273 dst[cc][x] = val;
1274 else
1275 dst[cc][x] ^= val;
1276 }
1277 }
1278 }
1279
1280 kmem_free(p, psize);
1281 }
1282
1283 static void
1284 vdev_raidz_reconstruct_general(raidz_row_t *rr, int *tgts, int ntgts)
1285 {
1286 int i, c, t, tt;
1287 unsigned int n;
1288 unsigned int nmissing_rows;
1289 int missing_rows[VDEV_RAIDZ_MAXPARITY];
1290 int parity_map[VDEV_RAIDZ_MAXPARITY];
1291 uint8_t *p, *pp;
1292 size_t psize;
1293 uint8_t *rows[VDEV_RAIDZ_MAXPARITY];
1294 uint8_t *invrows[VDEV_RAIDZ_MAXPARITY];
1295 uint8_t *used;
1296
1297 abd_t **bufs = NULL;
1298
1299 /*
1300 * Matrix reconstruction can't use scatter ABDs yet, so we allocate
1301 * temporary linear ABDs if any non-linear ABDs are found.
1302 */
1303 for (i = rr->rr_firstdatacol; i < rr->rr_cols; i++) {
1304 if (!abd_is_linear(rr->rr_col[i].rc_abd)) {
1305 bufs = kmem_alloc(rr->rr_cols * sizeof (abd_t *),
1306 KM_PUSHPAGE);
1307
1308 for (c = rr->rr_firstdatacol; c < rr->rr_cols; c++) {
1309 raidz_col_t *col = &rr->rr_col[c];
1310
1311 bufs[c] = col->rc_abd;
1312 if (bufs[c] != NULL) {
1313 col->rc_abd = abd_alloc_linear(
1314 col->rc_size, B_TRUE);
1315 abd_copy(col->rc_abd, bufs[c],
1316 col->rc_size);
1317 }
1318 }
1319
1320 break;
1321 }
1322 }
1323
1324 n = rr->rr_cols - rr->rr_firstdatacol;
1325
1326 /*
1327 * Figure out which data columns are missing.
1328 */
1329 nmissing_rows = 0;
1330 for (t = 0; t < ntgts; t++) {
1331 if (tgts[t] >= rr->rr_firstdatacol) {
1332 missing_rows[nmissing_rows++] =
1333 tgts[t] - rr->rr_firstdatacol;
1334 }
1335 }
1336
1337 /*
1338 * Figure out which parity columns to use to help generate the missing
1339 * data columns.
1340 */
1341 for (tt = 0, c = 0, i = 0; i < nmissing_rows; c++) {
1342 ASSERT(tt < ntgts);
1343 ASSERT(c < rr->rr_firstdatacol);
1344
1345 /*
1346 * Skip any targeted parity columns.
1347 */
1348 if (c == tgts[tt]) {
1349 tt++;
1350 continue;
1351 }
1352
1353 parity_map[i] = c;
1354 i++;
1355 }
1356
1357 psize = (sizeof (rows[0][0]) + sizeof (invrows[0][0])) *
1358 nmissing_rows * n + sizeof (used[0]) * n;
1359 p = kmem_alloc(psize, KM_SLEEP);
1360
1361 for (pp = p, i = 0; i < nmissing_rows; i++) {
1362 rows[i] = pp;
1363 pp += n;
1364 invrows[i] = pp;
1365 pp += n;
1366 }
1367 used = pp;
1368
1369 for (i = 0; i < nmissing_rows; i++) {
1370 used[i] = parity_map[i];
1371 }
1372
1373 for (tt = 0, c = rr->rr_firstdatacol; c < rr->rr_cols; c++) {
1374 if (tt < nmissing_rows &&
1375 c == missing_rows[tt] + rr->rr_firstdatacol) {
1376 tt++;
1377 continue;
1378 }
1379
1380 ASSERT3S(i, <, n);
1381 used[i] = c;
1382 i++;
1383 }
1384
1385 /*
1386 * Initialize the interesting rows of the matrix.
1387 */
1388 vdev_raidz_matrix_init(rr, n, nmissing_rows, parity_map, rows);
1389
1390 /*
1391 * Invert the matrix.
1392 */
1393 vdev_raidz_matrix_invert(rr, n, nmissing_rows, missing_rows, rows,
1394 invrows, used);
1395
1396 /*
1397 * Reconstruct the missing data using the generated matrix.
1398 */
1399 vdev_raidz_matrix_reconstruct(rr, n, nmissing_rows, missing_rows,
1400 invrows, used);
1401
1402 kmem_free(p, psize);
1403
1404 /*
1405 * copy back from temporary linear abds and free them
1406 */
1407 if (bufs) {
1408 for (c = rr->rr_firstdatacol; c < rr->rr_cols; c++) {
1409 raidz_col_t *col = &rr->rr_col[c];
1410
1411 if (bufs[c] != NULL) {
1412 abd_copy(bufs[c], col->rc_abd, col->rc_size);
1413 abd_free(col->rc_abd);
1414 }
1415 col->rc_abd = bufs[c];
1416 }
1417 kmem_free(bufs, rr->rr_cols * sizeof (abd_t *));
1418 }
1419 }
1420
1421 static void
1422 vdev_raidz_reconstruct_row(raidz_map_t *rm, raidz_row_t *rr,
1423 const int *t, int nt)
1424 {
1425 int tgts[VDEV_RAIDZ_MAXPARITY], *dt;
1426 int ntgts;
1427 int i, c, ret;
1428 int nbadparity, nbaddata;
1429 int parity_valid[VDEV_RAIDZ_MAXPARITY];
1430
1431 nbadparity = rr->rr_firstdatacol;
1432 nbaddata = rr->rr_cols - nbadparity;
1433 ntgts = 0;
1434 for (i = 0, c = 0; c < rr->rr_cols; c++) {
1435 if (c < rr->rr_firstdatacol)
1436 parity_valid[c] = B_FALSE;
1437
1438 if (i < nt && c == t[i]) {
1439 tgts[ntgts++] = c;
1440 i++;
1441 } else if (rr->rr_col[c].rc_error != 0) {
1442 tgts[ntgts++] = c;
1443 } else if (c >= rr->rr_firstdatacol) {
1444 nbaddata--;
1445 } else {
1446 parity_valid[c] = B_TRUE;
1447 nbadparity--;
1448 }
1449 }
1450
1451 ASSERT(ntgts >= nt);
1452 ASSERT(nbaddata >= 0);
1453 ASSERT(nbaddata + nbadparity == ntgts);
1454
1455 dt = &tgts[nbadparity];
1456
1457 /* Reconstruct using the new math implementation */
1458 ret = vdev_raidz_math_reconstruct(rm, rr, parity_valid, dt, nbaddata);
1459 if (ret != RAIDZ_ORIGINAL_IMPL)
1460 return;
1461
1462 /*
1463 * See if we can use any of our optimized reconstruction routines.
1464 */
1465 switch (nbaddata) {
1466 case 1:
1467 if (parity_valid[VDEV_RAIDZ_P]) {
1468 vdev_raidz_reconstruct_p(rr, dt, 1);
1469 return;
1470 }
1471
1472 ASSERT(rr->rr_firstdatacol > 1);
1473
1474 if (parity_valid[VDEV_RAIDZ_Q]) {
1475 vdev_raidz_reconstruct_q(rr, dt, 1);
1476 return;
1477 }
1478
1479 ASSERT(rr->rr_firstdatacol > 2);
1480 break;
1481
1482 case 2:
1483 ASSERT(rr->rr_firstdatacol > 1);
1484
1485 if (parity_valid[VDEV_RAIDZ_P] &&
1486 parity_valid[VDEV_RAIDZ_Q]) {
1487 vdev_raidz_reconstruct_pq(rr, dt, 2);
1488 return;
1489 }
1490
1491 ASSERT(rr->rr_firstdatacol > 2);
1492
1493 break;
1494 }
1495
1496 vdev_raidz_reconstruct_general(rr, tgts, ntgts);
1497 }
1498
1499 static int
1500 vdev_raidz_open(vdev_t *vd, uint64_t *asize, uint64_t *max_asize,
1501 uint64_t *logical_ashift, uint64_t *physical_ashift)
1502 {
1503 vdev_raidz_t *vdrz = vd->vdev_tsd;
1504 uint64_t nparity = vdrz->vd_nparity;
1505 int c;
1506 int lasterror = 0;
1507 int numerrors = 0;
1508
1509 ASSERT(nparity > 0);
1510
1511 if (nparity > VDEV_RAIDZ_MAXPARITY ||
1512 vd->vdev_children < nparity + 1) {
1513 vd->vdev_stat.vs_aux = VDEV_AUX_BAD_LABEL;
1514 return (SET_ERROR(EINVAL));
1515 }
1516
1517 vdev_open_children(vd);
1518
1519 for (c = 0; c < vd->vdev_children; c++) {
1520 vdev_t *cvd = vd->vdev_child[c];
1521
1522 if (cvd->vdev_open_error != 0) {
1523 lasterror = cvd->vdev_open_error;
1524 numerrors++;
1525 continue;
1526 }
1527
1528 *asize = MIN(*asize - 1, cvd->vdev_asize - 1) + 1;
1529 *max_asize = MIN(*max_asize - 1, cvd->vdev_max_asize - 1) + 1;
1530 *logical_ashift = MAX(*logical_ashift, cvd->vdev_ashift);
1531 }
1532 for (c = 0; c < vd->vdev_children; c++) {
1533 vdev_t *cvd = vd->vdev_child[c];
1534
1535 if (cvd->vdev_open_error != 0)
1536 continue;
1537 *physical_ashift = vdev_best_ashift(*logical_ashift,
1538 *physical_ashift, cvd->vdev_physical_ashift);
1539 }
1540
1541 *asize *= vd->vdev_children;
1542 *max_asize *= vd->vdev_children;
1543
1544 if (numerrors > nparity) {
1545 vd->vdev_stat.vs_aux = VDEV_AUX_NO_REPLICAS;
1546 return (lasterror);
1547 }
1548
1549 return (0);
1550 }
1551
1552 static void
1553 vdev_raidz_close(vdev_t *vd)
1554 {
1555 for (int c = 0; c < vd->vdev_children; c++) {
1556 if (vd->vdev_child[c] != NULL)
1557 vdev_close(vd->vdev_child[c]);
1558 }
1559 }
1560
1561 static uint64_t
1562 vdev_raidz_asize(vdev_t *vd, uint64_t psize)
1563 {
1564 vdev_raidz_t *vdrz = vd->vdev_tsd;
1565 uint64_t asize;
1566 uint64_t ashift = vd->vdev_top->vdev_ashift;
1567 uint64_t cols = vdrz->vd_logical_width;
1568 uint64_t nparity = vdrz->vd_nparity;
1569
1570 asize = ((psize - 1) >> ashift) + 1;
1571 asize += nparity * ((asize + cols - nparity - 1) / (cols - nparity));
1572 asize = roundup(asize, nparity + 1) << ashift;
1573
1574 return (asize);
1575 }
1576
1577 /*
1578 * The allocatable space for a raidz vdev is N * sizeof(smallest child)
1579 * so each child must provide at least 1/Nth of its asize.
1580 */
1581 static uint64_t
1582 vdev_raidz_min_asize(vdev_t *vd)
1583 {
1584 return ((vd->vdev_min_asize + vd->vdev_children - 1) /
1585 vd->vdev_children);
1586 }
1587
1588 void
1589 vdev_raidz_child_done(zio_t *zio)
1590 {
1591 raidz_col_t *rc = zio->io_private;
1592
1593 ASSERT3P(rc->rc_abd, !=, NULL);
1594 rc->rc_error = zio->io_error;
1595 rc->rc_tried = 1;
1596 rc->rc_skipped = 0;
1597 }
1598
1599 static void
1600 vdev_raidz_io_verify(vdev_t *vd, raidz_row_t *rr, int col)
1601 {
1602 #ifdef ZFS_DEBUG
1603 vdev_t *tvd = vd->vdev_top;
1604
1605 range_seg64_t logical_rs, physical_rs, remain_rs;
1606 logical_rs.rs_start = rr->rr_offset;
1607 logical_rs.rs_end = logical_rs.rs_start +
1608 vdev_raidz_asize(vd, rr->rr_size);
1609
1610 raidz_col_t *rc = &rr->rr_col[col];
1611 vdev_t *cvd = vd->vdev_child[rc->rc_devidx];
1612
1613 vdev_xlate(cvd, &logical_rs, &physical_rs, &remain_rs);
1614 ASSERT(vdev_xlate_is_empty(&remain_rs));
1615 ASSERT3U(rc->rc_offset, ==, physical_rs.rs_start);
1616 ASSERT3U(rc->rc_offset, <, physical_rs.rs_end);
1617 /*
1618 * It would be nice to assert that rs_end is equal
1619 * to rc_offset + rc_size but there might be an
1620 * optional I/O at the end that is not accounted in
1621 * rc_size.
1622 */
1623 if (physical_rs.rs_end > rc->rc_offset + rc->rc_size) {
1624 ASSERT3U(physical_rs.rs_end, ==, rc->rc_offset +
1625 rc->rc_size + (1 << tvd->vdev_ashift));
1626 } else {
1627 ASSERT3U(physical_rs.rs_end, ==, rc->rc_offset + rc->rc_size);
1628 }
1629 #endif
1630 }
1631
1632 static void
1633 vdev_raidz_io_start_write(zio_t *zio, raidz_row_t *rr, uint64_t ashift)
1634 {
1635 vdev_t *vd = zio->io_vd;
1636 raidz_map_t *rm = zio->io_vsd;
1637
1638 vdev_raidz_generate_parity_row(rm, rr);
1639
1640 for (int c = 0; c < rr->rr_scols; c++) {
1641 raidz_col_t *rc = &rr->rr_col[c];
1642 vdev_t *cvd = vd->vdev_child[rc->rc_devidx];
1643
1644 /* Verify physical to logical translation */
1645 vdev_raidz_io_verify(vd, rr, c);
1646
1647 if (rc->rc_size > 0) {
1648 ASSERT3P(rc->rc_abd, !=, NULL);
1649 zio_nowait(zio_vdev_child_io(zio, NULL, cvd,
1650 rc->rc_offset, rc->rc_abd,
1651 abd_get_size(rc->rc_abd), zio->io_type,
1652 zio->io_priority, 0, vdev_raidz_child_done, rc));
1653 } else {
1654 /*
1655 * Generate optional write for skip sector to improve
1656 * aggregation contiguity.
1657 */
1658 ASSERT3P(rc->rc_abd, ==, NULL);
1659 zio_nowait(zio_vdev_child_io(zio, NULL, cvd,
1660 rc->rc_offset, NULL, 1ULL << ashift,
1661 zio->io_type, zio->io_priority,
1662 ZIO_FLAG_NODATA | ZIO_FLAG_OPTIONAL, NULL,
1663 NULL));
1664 }
1665 }
1666 }
1667
1668 static void
1669 vdev_raidz_io_start_read(zio_t *zio, raidz_row_t *rr)
1670 {
1671 vdev_t *vd = zio->io_vd;
1672
1673 /*
1674 * Iterate over the columns in reverse order so that we hit the parity
1675 * last -- any errors along the way will force us to read the parity.
1676 */
1677 for (int c = rr->rr_cols - 1; c >= 0; c--) {
1678 raidz_col_t *rc = &rr->rr_col[c];
1679 if (rc->rc_size == 0)
1680 continue;
1681 vdev_t *cvd = vd->vdev_child[rc->rc_devidx];
1682 if (!vdev_readable(cvd)) {
1683 if (c >= rr->rr_firstdatacol)
1684 rr->rr_missingdata++;
1685 else
1686 rr->rr_missingparity++;
1687 rc->rc_error = SET_ERROR(ENXIO);
1688 rc->rc_tried = 1; /* don't even try */
1689 rc->rc_skipped = 1;
1690 continue;
1691 }
1692 if (vdev_dtl_contains(cvd, DTL_MISSING, zio->io_txg, 1)) {
1693 if (c >= rr->rr_firstdatacol)
1694 rr->rr_missingdata++;
1695 else
1696 rr->rr_missingparity++;
1697 rc->rc_error = SET_ERROR(ESTALE);
1698 rc->rc_skipped = 1;
1699 continue;
1700 }
1701 if (c >= rr->rr_firstdatacol || rr->rr_missingdata > 0 ||
1702 (zio->io_flags & (ZIO_FLAG_SCRUB | ZIO_FLAG_RESILVER))) {
1703 zio_nowait(zio_vdev_child_io(zio, NULL, cvd,
1704 rc->rc_offset, rc->rc_abd, rc->rc_size,
1705 zio->io_type, zio->io_priority, 0,
1706 vdev_raidz_child_done, rc));
1707 }
1708 }
1709 }
1710
1711 /*
1712 * Start an IO operation on a RAIDZ VDev
1713 *
1714 * Outline:
1715 * - For write operations:
1716 * 1. Generate the parity data
1717 * 2. Create child zio write operations to each column's vdev, for both
1718 * data and parity.
1719 * 3. If the column skips any sectors for padding, create optional dummy
1720 * write zio children for those areas to improve aggregation continuity.
1721 * - For read operations:
1722 * 1. Create child zio read operations to each data column's vdev to read
1723 * the range of data required for zio.
1724 * 2. If this is a scrub or resilver operation, or if any of the data
1725 * vdevs have had errors, then create zio read operations to the parity
1726 * columns' VDevs as well.
1727 */
1728 static void
1729 vdev_raidz_io_start(zio_t *zio)
1730 {
1731 vdev_t *vd = zio->io_vd;
1732 vdev_t *tvd = vd->vdev_top;
1733 vdev_raidz_t *vdrz = vd->vdev_tsd;
1734
1735 raidz_map_t *rm = vdev_raidz_map_alloc(zio, tvd->vdev_ashift,
1736 vdrz->vd_logical_width, vdrz->vd_nparity);
1737 zio->io_vsd = rm;
1738 zio->io_vsd_ops = &vdev_raidz_vsd_ops;
1739
1740 /*
1741 * Until raidz expansion is implemented all maps for a raidz vdev
1742 * contain a single row.
1743 */
1744 ASSERT3U(rm->rm_nrows, ==, 1);
1745 raidz_row_t *rr = rm->rm_row[0];
1746
1747 if (zio->io_type == ZIO_TYPE_WRITE) {
1748 vdev_raidz_io_start_write(zio, rr, tvd->vdev_ashift);
1749 } else {
1750 ASSERT(zio->io_type == ZIO_TYPE_READ);
1751 vdev_raidz_io_start_read(zio, rr);
1752 }
1753
1754 zio_execute(zio);
1755 }
1756
1757 /*
1758 * Report a checksum error for a child of a RAID-Z device.
1759 */
1760 void
1761 vdev_raidz_checksum_error(zio_t *zio, raidz_col_t *rc, abd_t *bad_data)
1762 {
1763 vdev_t *vd = zio->io_vd->vdev_child[rc->rc_devidx];
1764
1765 if (!(zio->io_flags & ZIO_FLAG_SPECULATIVE) &&
1766 zio->io_priority != ZIO_PRIORITY_REBUILD) {
1767 zio_bad_cksum_t zbc;
1768 raidz_map_t *rm = zio->io_vsd;
1769
1770 zbc.zbc_has_cksum = 0;
1771 zbc.zbc_injected = rm->rm_ecksuminjected;
1772
1773 mutex_enter(&vd->vdev_stat_lock);
1774 vd->vdev_stat.vs_checksum_errors++;
1775 mutex_exit(&vd->vdev_stat_lock);
1776 (void) zfs_ereport_post_checksum(zio->io_spa, vd,
1777 &zio->io_bookmark, zio, rc->rc_offset, rc->rc_size,
1778 rc->rc_abd, bad_data, &zbc);
1779 }
1780 }
1781
1782 /*
1783 * We keep track of whether or not there were any injected errors, so that
1784 * any ereports we generate can note it.
1785 */
1786 static int
1787 raidz_checksum_verify(zio_t *zio)
1788 {
1789 zio_bad_cksum_t zbc = {0};
1790 raidz_map_t *rm = zio->io_vsd;
1791
1792 int ret = zio_checksum_error(zio, &zbc);
1793 if (ret != 0 && zbc.zbc_injected != 0)
1794 rm->rm_ecksuminjected = 1;
1795
1796 return (ret);
1797 }
1798
1799 /*
1800 * Generate the parity from the data columns. If we tried and were able to
1801 * read the parity without error, verify that the generated parity matches the
1802 * data we read. If it doesn't, we fire off a checksum error. Return the
1803 * number of such failures.
1804 */
1805 static int
1806 raidz_parity_verify(zio_t *zio, raidz_row_t *rr)
1807 {
1808 abd_t *orig[VDEV_RAIDZ_MAXPARITY];
1809 int c, ret = 0;
1810 raidz_map_t *rm = zio->io_vsd;
1811 raidz_col_t *rc;
1812
1813 blkptr_t *bp = zio->io_bp;
1814 enum zio_checksum checksum = (bp == NULL ? zio->io_prop.zp_checksum :
1815 (BP_IS_GANG(bp) ? ZIO_CHECKSUM_GANG_HEADER : BP_GET_CHECKSUM(bp)));
1816
1817 if (checksum == ZIO_CHECKSUM_NOPARITY)
1818 return (ret);
1819
1820 for (c = 0; c < rr->rr_firstdatacol; c++) {
1821 rc = &rr->rr_col[c];
1822 if (!rc->rc_tried || rc->rc_error != 0)
1823 continue;
1824
1825 orig[c] = rc->rc_abd;
1826 ASSERT3U(abd_get_size(rc->rc_abd), ==, rc->rc_size);
1827 rc->rc_abd = abd_alloc_linear(rc->rc_size, B_FALSE);
1828 }
1829
1830 /*
1831 * Verify any empty sectors are zero filled to ensure the parity
1832 * is calculated correctly even if these non-data sectors are damaged.
1833 */
1834 if (rr->rr_nempty && rr->rr_abd_empty != NULL)
1835 ret += vdev_draid_map_verify_empty(zio, rr);
1836
1837 /*
1838 * Regenerates parity even for !tried||rc_error!=0 columns. This
1839 * isn't harmful but it does have the side effect of fixing stuff
1840 * we didn't realize was necessary (i.e. even if we return 0).
1841 */
1842 vdev_raidz_generate_parity_row(rm, rr);
1843
1844 for (c = 0; c < rr->rr_firstdatacol; c++) {
1845 rc = &rr->rr_col[c];
1846
1847 if (!rc->rc_tried || rc->rc_error != 0)
1848 continue;
1849
1850 if (abd_cmp(orig[c], rc->rc_abd) != 0) {
1851 vdev_raidz_checksum_error(zio, rc, orig[c]);
1852 rc->rc_error = SET_ERROR(ECKSUM);
1853 ret++;
1854 }
1855 abd_free(orig[c]);
1856 }
1857
1858 return (ret);
1859 }
1860
1861 static int
1862 vdev_raidz_worst_error(raidz_row_t *rr)
1863 {
1864 int error = 0;
1865
1866 for (int c = 0; c < rr->rr_cols; c++)
1867 error = zio_worst_error(error, rr->rr_col[c].rc_error);
1868
1869 return (error);
1870 }
1871
1872 static void
1873 vdev_raidz_io_done_verified(zio_t *zio, raidz_row_t *rr)
1874 {
1875 int unexpected_errors = 0;
1876 int parity_errors = 0;
1877 int parity_untried = 0;
1878 int data_errors = 0;
1879
1880 ASSERT3U(zio->io_type, ==, ZIO_TYPE_READ);
1881
1882 for (int c = 0; c < rr->rr_cols; c++) {
1883 raidz_col_t *rc = &rr->rr_col[c];
1884
1885 if (rc->rc_error) {
1886 if (c < rr->rr_firstdatacol)
1887 parity_errors++;
1888 else
1889 data_errors++;
1890
1891 if (!rc->rc_skipped)
1892 unexpected_errors++;
1893 } else if (c < rr->rr_firstdatacol && !rc->rc_tried) {
1894 parity_untried++;
1895 }
1896
1897 if (rc->rc_force_repair)
1898 unexpected_errors++;
1899 }
1900
1901 /*
1902 * If we read more parity disks than were used for
1903 * reconstruction, confirm that the other parity disks produced
1904 * correct data.
1905 *
1906 * Note that we also regenerate parity when resilvering so we
1907 * can write it out to failed devices later.
1908 */
1909 if (parity_errors + parity_untried <
1910 rr->rr_firstdatacol - data_errors ||
1911 (zio->io_flags & ZIO_FLAG_RESILVER)) {
1912 int n = raidz_parity_verify(zio, rr);
1913 unexpected_errors += n;
1914 }
1915
1916 if (zio->io_error == 0 && spa_writeable(zio->io_spa) &&
1917 (unexpected_errors > 0 || (zio->io_flags & ZIO_FLAG_RESILVER))) {
1918 /*
1919 * Use the good data we have in hand to repair damaged children.
1920 */
1921 for (int c = 0; c < rr->rr_cols; c++) {
1922 raidz_col_t *rc = &rr->rr_col[c];
1923 vdev_t *vd = zio->io_vd;
1924 vdev_t *cvd = vd->vdev_child[rc->rc_devidx];
1925
1926 if (!rc->rc_allow_repair) {
1927 continue;
1928 } else if (!rc->rc_force_repair &&
1929 (rc->rc_error == 0 || rc->rc_size == 0)) {
1930 continue;
1931 }
1932
1933 zio_nowait(zio_vdev_child_io(zio, NULL, cvd,
1934 rc->rc_offset, rc->rc_abd, rc->rc_size,
1935 ZIO_TYPE_WRITE,
1936 zio->io_priority == ZIO_PRIORITY_REBUILD ?
1937 ZIO_PRIORITY_REBUILD : ZIO_PRIORITY_ASYNC_WRITE,
1938 ZIO_FLAG_IO_REPAIR | (unexpected_errors ?
1939 ZIO_FLAG_SELF_HEAL : 0), NULL, NULL));
1940 }
1941 }
1942 }
1943
1944 static void
1945 raidz_restore_orig_data(raidz_map_t *rm)
1946 {
1947 for (int i = 0; i < rm->rm_nrows; i++) {
1948 raidz_row_t *rr = rm->rm_row[i];
1949 for (int c = 0; c < rr->rr_cols; c++) {
1950 raidz_col_t *rc = &rr->rr_col[c];
1951 if (rc->rc_need_orig_restore) {
1952 abd_copy(rc->rc_abd,
1953 rc->rc_orig_data, rc->rc_size);
1954 rc->rc_need_orig_restore = B_FALSE;
1955 }
1956 }
1957 }
1958 }
1959
1960 /*
1961 * returns EINVAL if reconstruction of the block will not be possible
1962 * returns ECKSUM if this specific reconstruction failed
1963 * returns 0 on successful reconstruction
1964 */
1965 static int
1966 raidz_reconstruct(zio_t *zio, int *ltgts, int ntgts, int nparity)
1967 {
1968 raidz_map_t *rm = zio->io_vsd;
1969
1970 /* Reconstruct each row */
1971 for (int r = 0; r < rm->rm_nrows; r++) {
1972 raidz_row_t *rr = rm->rm_row[r];
1973 int my_tgts[VDEV_RAIDZ_MAXPARITY]; /* value is child id */
1974 int t = 0;
1975 int dead = 0;
1976 int dead_data = 0;
1977
1978 for (int c = 0; c < rr->rr_cols; c++) {
1979 raidz_col_t *rc = &rr->rr_col[c];
1980 ASSERT0(rc->rc_need_orig_restore);
1981 if (rc->rc_error != 0) {
1982 dead++;
1983 if (c >= nparity)
1984 dead_data++;
1985 continue;
1986 }
1987 if (rc->rc_size == 0)
1988 continue;
1989 for (int lt = 0; lt < ntgts; lt++) {
1990 if (rc->rc_devidx == ltgts[lt]) {
1991 if (rc->rc_orig_data == NULL) {
1992 rc->rc_orig_data =
1993 abd_alloc_linear(
1994 rc->rc_size, B_TRUE);
1995 abd_copy(rc->rc_orig_data,
1996 rc->rc_abd, rc->rc_size);
1997 }
1998 rc->rc_need_orig_restore = B_TRUE;
1999
2000 dead++;
2001 if (c >= nparity)
2002 dead_data++;
2003 my_tgts[t++] = c;
2004 break;
2005 }
2006 }
2007 }
2008 if (dead > nparity) {
2009 /* reconstruction not possible */
2010 raidz_restore_orig_data(rm);
2011 return (EINVAL);
2012 }
2013 if (dead_data > 0)
2014 vdev_raidz_reconstruct_row(rm, rr, my_tgts, t);
2015 }
2016
2017 /* Check for success */
2018 if (raidz_checksum_verify(zio) == 0) {
2019
2020 /* Reconstruction succeeded - report errors */
2021 for (int i = 0; i < rm->rm_nrows; i++) {
2022 raidz_row_t *rr = rm->rm_row[i];
2023
2024 for (int c = 0; c < rr->rr_cols; c++) {
2025 raidz_col_t *rc = &rr->rr_col[c];
2026 if (rc->rc_need_orig_restore) {
2027 /*
2028 * Note: if this is a parity column,
2029 * we don't really know if it's wrong.
2030 * We need to let
2031 * vdev_raidz_io_done_verified() check
2032 * it, and if we set rc_error, it will
2033 * think that it is a "known" error
2034 * that doesn't need to be checked
2035 * or corrected.
2036 */
2037 if (rc->rc_error == 0 &&
2038 c >= rr->rr_firstdatacol) {
2039 vdev_raidz_checksum_error(zio,
2040 rc, rc->rc_orig_data);
2041 rc->rc_error =
2042 SET_ERROR(ECKSUM);
2043 }
2044 rc->rc_need_orig_restore = B_FALSE;
2045 }
2046 }
2047
2048 vdev_raidz_io_done_verified(zio, rr);
2049 }
2050
2051 zio_checksum_verified(zio);
2052
2053 return (0);
2054 }
2055
2056 /* Reconstruction failed - restore original data */
2057 raidz_restore_orig_data(rm);
2058 return (ECKSUM);
2059 }
2060
2061 /*
2062 * Iterate over all combinations of N bad vdevs and attempt a reconstruction.
2063 * Note that the algorithm below is non-optimal because it doesn't take into
2064 * account how reconstruction is actually performed. For example, with
2065 * triple-parity RAID-Z the reconstruction procedure is the same if column 4
2066 * is targeted as invalid as if columns 1 and 4 are targeted since in both
2067 * cases we'd only use parity information in column 0.
2068 *
2069 * The order that we find the various possible combinations of failed
2070 * disks is dictated by these rules:
2071 * - Examine each "slot" (the "i" in tgts[i])
2072 * - Try to increment this slot (tgts[i] = tgts[i] + 1)
2073 * - if we can't increment because it runs into the next slot,
2074 * reset our slot to the minimum, and examine the next slot
2075 *
2076 * For example, with a 6-wide RAIDZ3, and no known errors (so we have to choose
2077 * 3 columns to reconstruct), we will generate the following sequence:
2078 *
2079 * STATE ACTION
2080 * 0 1 2 special case: skip since these are all parity
2081 * 0 1 3 first slot: reset to 0; middle slot: increment to 2
2082 * 0 2 3 first slot: increment to 1
2083 * 1 2 3 first: reset to 0; middle: reset to 1; last: increment to 4
2084 * 0 1 4 first: reset to 0; middle: increment to 2
2085 * 0 2 4 first: increment to 1
2086 * 1 2 4 first: reset to 0; middle: increment to 3
2087 * 0 3 4 first: increment to 1
2088 * 1 3 4 first: increment to 2
2089 * 2 3 4 first: reset to 0; middle: reset to 1; last: increment to 5
2090 * 0 1 5 first: reset to 0; middle: increment to 2
2091 * 0 2 5 first: increment to 1
2092 * 1 2 5 first: reset to 0; middle: increment to 3
2093 * 0 3 5 first: increment to 1
2094 * 1 3 5 first: increment to 2
2095 * 2 3 5 first: reset to 0; middle: increment to 4
2096 * 0 4 5 first: increment to 1
2097 * 1 4 5 first: increment to 2
2098 * 2 4 5 first: increment to 3
2099 * 3 4 5 done
2100 *
2101 * This strategy works for dRAID but is less efficient when there are a large
2102 * number of child vdevs and therefore permutations to check. Furthermore,
2103 * since the raidz_map_t rows likely do not overlap reconstruction would be
2104 * possible as long as there are no more than nparity data errors per row.
2105 * These additional permutations are not currently checked but could be as
2106 * a future improvement.
2107 */
2108 static int
2109 vdev_raidz_combrec(zio_t *zio)
2110 {
2111 int nparity = vdev_get_nparity(zio->io_vd);
2112 raidz_map_t *rm = zio->io_vsd;
2113
2114 /* Check if there's enough data to attempt reconstrution. */
2115 for (int i = 0; i < rm->rm_nrows; i++) {
2116 raidz_row_t *rr = rm->rm_row[i];
2117 int total_errors = 0;
2118
2119 for (int c = 0; c < rr->rr_cols; c++) {
2120 if (rr->rr_col[c].rc_error)
2121 total_errors++;
2122 }
2123
2124 if (total_errors > nparity)
2125 return (vdev_raidz_worst_error(rr));
2126 }
2127
2128 for (int num_failures = 1; num_failures <= nparity; num_failures++) {
2129 int tstore[VDEV_RAIDZ_MAXPARITY + 2];
2130 int *ltgts = &tstore[1]; /* value is logical child ID */
2131
2132 /* Determine number of logical children, n */
2133 int n = zio->io_vd->vdev_children;
2134
2135 ASSERT3U(num_failures, <=, nparity);
2136 ASSERT3U(num_failures, <=, VDEV_RAIDZ_MAXPARITY);
2137
2138 /* Handle corner cases in combrec logic */
2139 ltgts[-1] = -1;
2140 for (int i = 0; i < num_failures; i++) {
2141 ltgts[i] = i;
2142 }
2143 ltgts[num_failures] = n;
2144
2145 for (;;) {
2146 int err = raidz_reconstruct(zio, ltgts, num_failures,
2147 nparity);
2148 if (err == EINVAL) {
2149 /*
2150 * Reconstruction not possible with this #
2151 * failures; try more failures.
2152 */
2153 break;
2154 } else if (err == 0)
2155 return (0);
2156
2157 /* Compute next targets to try */
2158 for (int t = 0; ; t++) {
2159 ASSERT3U(t, <, num_failures);
2160 ltgts[t]++;
2161 if (ltgts[t] == n) {
2162 /* try more failures */
2163 ASSERT3U(t, ==, num_failures - 1);
2164 break;
2165 }
2166
2167 ASSERT3U(ltgts[t], <, n);
2168 ASSERT3U(ltgts[t], <=, ltgts[t + 1]);
2169
2170 /*
2171 * If that spot is available, we're done here.
2172 * Try the next combination.
2173 */
2174 if (ltgts[t] != ltgts[t + 1])
2175 break;
2176
2177 /*
2178 * Otherwise, reset this tgt to the minimum,
2179 * and move on to the next tgt.
2180 */
2181 ltgts[t] = ltgts[t - 1] + 1;
2182 ASSERT3U(ltgts[t], ==, t);
2183 }
2184
2185 /* Increase the number of failures and keep trying. */
2186 if (ltgts[num_failures - 1] == n)
2187 break;
2188 }
2189 }
2190
2191 return (ECKSUM);
2192 }
2193
2194 void
2195 vdev_raidz_reconstruct(raidz_map_t *rm, const int *t, int nt)
2196 {
2197 for (uint64_t row = 0; row < rm->rm_nrows; row++) {
2198 raidz_row_t *rr = rm->rm_row[row];
2199 vdev_raidz_reconstruct_row(rm, rr, t, nt);
2200 }
2201 }
2202
2203 /*
2204 * Complete a write IO operation on a RAIDZ VDev
2205 *
2206 * Outline:
2207 * 1. Check for errors on the child IOs.
2208 * 2. Return, setting an error code if too few child VDevs were written
2209 * to reconstruct the data later. Note that partial writes are
2210 * considered successful if they can be reconstructed at all.
2211 */
2212 static void
2213 vdev_raidz_io_done_write_impl(zio_t *zio, raidz_row_t *rr)
2214 {
2215 int total_errors = 0;
2216
2217 ASSERT3U(rr->rr_missingparity, <=, rr->rr_firstdatacol);
2218 ASSERT3U(rr->rr_missingdata, <=, rr->rr_cols - rr->rr_firstdatacol);
2219 ASSERT3U(zio->io_type, ==, ZIO_TYPE_WRITE);
2220
2221 for (int c = 0; c < rr->rr_cols; c++) {
2222 raidz_col_t *rc = &rr->rr_col[c];
2223
2224 if (rc->rc_error) {
2225 ASSERT(rc->rc_error != ECKSUM); /* child has no bp */
2226
2227 total_errors++;
2228 }
2229 }
2230
2231 /*
2232 * Treat partial writes as a success. If we couldn't write enough
2233 * columns to reconstruct the data, the I/O failed. Otherwise,
2234 * good enough.
2235 *
2236 * Now that we support write reallocation, it would be better
2237 * to treat partial failure as real failure unless there are
2238 * no non-degraded top-level vdevs left, and not update DTLs
2239 * if we intend to reallocate.
2240 */
2241 if (total_errors > rr->rr_firstdatacol) {
2242 zio->io_error = zio_worst_error(zio->io_error,
2243 vdev_raidz_worst_error(rr));
2244 }
2245 }
2246
2247 static void
2248 vdev_raidz_io_done_reconstruct_known_missing(zio_t *zio, raidz_map_t *rm,
2249 raidz_row_t *rr)
2250 {
2251 int parity_errors = 0;
2252 int parity_untried = 0;
2253 int data_errors = 0;
2254 int total_errors = 0;
2255
2256 ASSERT3U(rr->rr_missingparity, <=, rr->rr_firstdatacol);
2257 ASSERT3U(rr->rr_missingdata, <=, rr->rr_cols - rr->rr_firstdatacol);
2258 ASSERT3U(zio->io_type, ==, ZIO_TYPE_READ);
2259
2260 for (int c = 0; c < rr->rr_cols; c++) {
2261 raidz_col_t *rc = &rr->rr_col[c];
2262
2263 /*
2264 * If scrubbing and a replacing/sparing child vdev determined
2265 * that not all of its children have an identical copy of the
2266 * data, then clear the error so the column is treated like
2267 * any other read and force a repair to correct the damage.
2268 */
2269 if (rc->rc_error == ECKSUM) {
2270 ASSERT(zio->io_flags & ZIO_FLAG_SCRUB);
2271 vdev_raidz_checksum_error(zio, rc, rc->rc_abd);
2272 rc->rc_force_repair = 1;
2273 rc->rc_error = 0;
2274 }
2275
2276 if (rc->rc_error) {
2277 if (c < rr->rr_firstdatacol)
2278 parity_errors++;
2279 else
2280 data_errors++;
2281
2282 total_errors++;
2283 } else if (c < rr->rr_firstdatacol && !rc->rc_tried) {
2284 parity_untried++;
2285 }
2286 }
2287
2288 /*
2289 * If there were data errors and the number of errors we saw was
2290 * correctable -- less than or equal to the number of parity disks read
2291 * -- reconstruct based on the missing data.
2292 */
2293 if (data_errors != 0 &&
2294 total_errors <= rr->rr_firstdatacol - parity_untried) {
2295 /*
2296 * We either attempt to read all the parity columns or
2297 * none of them. If we didn't try to read parity, we
2298 * wouldn't be here in the correctable case. There must
2299 * also have been fewer parity errors than parity
2300 * columns or, again, we wouldn't be in this code path.
2301 */
2302 ASSERT(parity_untried == 0);
2303 ASSERT(parity_errors < rr->rr_firstdatacol);
2304
2305 /*
2306 * Identify the data columns that reported an error.
2307 */
2308 int n = 0;
2309 int tgts[VDEV_RAIDZ_MAXPARITY];
2310 for (int c = rr->rr_firstdatacol; c < rr->rr_cols; c++) {
2311 raidz_col_t *rc = &rr->rr_col[c];
2312 if (rc->rc_error != 0) {
2313 ASSERT(n < VDEV_RAIDZ_MAXPARITY);
2314 tgts[n++] = c;
2315 }
2316 }
2317
2318 ASSERT(rr->rr_firstdatacol >= n);
2319
2320 vdev_raidz_reconstruct_row(rm, rr, tgts, n);
2321 }
2322 }
2323
2324 /*
2325 * Return the number of reads issued.
2326 */
2327 static int
2328 vdev_raidz_read_all(zio_t *zio, raidz_row_t *rr)
2329 {
2330 vdev_t *vd = zio->io_vd;
2331 int nread = 0;
2332
2333 rr->rr_missingdata = 0;
2334 rr->rr_missingparity = 0;
2335
2336 /*
2337 * If this rows contains empty sectors which are not required
2338 * for a normal read then allocate an ABD for them now so they
2339 * may be read, verified, and any needed repairs performed.
2340 */
2341 if (rr->rr_nempty && rr->rr_abd_empty == NULL)
2342 vdev_draid_map_alloc_empty(zio, rr);
2343
2344 for (int c = 0; c < rr->rr_cols; c++) {
2345 raidz_col_t *rc = &rr->rr_col[c];
2346 if (rc->rc_tried || rc->rc_size == 0)
2347 continue;
2348
2349 zio_nowait(zio_vdev_child_io(zio, NULL,
2350 vd->vdev_child[rc->rc_devidx],
2351 rc->rc_offset, rc->rc_abd, rc->rc_size,
2352 zio->io_type, zio->io_priority, 0,
2353 vdev_raidz_child_done, rc));
2354 nread++;
2355 }
2356 return (nread);
2357 }
2358
2359 /*
2360 * We're here because either there were too many errors to even attempt
2361 * reconstruction (total_errors == rm_first_datacol), or vdev_*_combrec()
2362 * failed. In either case, there is enough bad data to prevent reconstruction.
2363 * Start checksum ereports for all children which haven't failed.
2364 */
2365 static void
2366 vdev_raidz_io_done_unrecoverable(zio_t *zio)
2367 {
2368 raidz_map_t *rm = zio->io_vsd;
2369
2370 for (int i = 0; i < rm->rm_nrows; i++) {
2371 raidz_row_t *rr = rm->rm_row[i];
2372
2373 for (int c = 0; c < rr->rr_cols; c++) {
2374 raidz_col_t *rc = &rr->rr_col[c];
2375 vdev_t *cvd = zio->io_vd->vdev_child[rc->rc_devidx];
2376
2377 if (rc->rc_error != 0)
2378 continue;
2379
2380 zio_bad_cksum_t zbc;
2381 zbc.zbc_has_cksum = 0;
2382 zbc.zbc_injected = rm->rm_ecksuminjected;
2383
2384 mutex_enter(&cvd->vdev_stat_lock);
2385 cvd->vdev_stat.vs_checksum_errors++;
2386 mutex_exit(&cvd->vdev_stat_lock);
2387 (void) zfs_ereport_start_checksum(zio->io_spa,
2388 cvd, &zio->io_bookmark, zio, rc->rc_offset,
2389 rc->rc_size, &zbc);
2390 }
2391 }
2392 }
2393
2394 void
2395 vdev_raidz_io_done(zio_t *zio)
2396 {
2397 raidz_map_t *rm = zio->io_vsd;
2398
2399 if (zio->io_type == ZIO_TYPE_WRITE) {
2400 for (int i = 0; i < rm->rm_nrows; i++) {
2401 vdev_raidz_io_done_write_impl(zio, rm->rm_row[i]);
2402 }
2403 } else {
2404 for (int i = 0; i < rm->rm_nrows; i++) {
2405 raidz_row_t *rr = rm->rm_row[i];
2406 vdev_raidz_io_done_reconstruct_known_missing(zio,
2407 rm, rr);
2408 }
2409
2410 if (raidz_checksum_verify(zio) == 0) {
2411 for (int i = 0; i < rm->rm_nrows; i++) {
2412 raidz_row_t *rr = rm->rm_row[i];
2413 vdev_raidz_io_done_verified(zio, rr);
2414 }
2415 zio_checksum_verified(zio);
2416 } else {
2417 /*
2418 * A sequential resilver has no checksum which makes
2419 * combinatoral reconstruction impossible. This code
2420 * path is unreachable since raidz_checksum_verify()
2421 * has no checksum to verify and must succeed.
2422 */
2423 ASSERT3U(zio->io_priority, !=, ZIO_PRIORITY_REBUILD);
2424
2425 /*
2426 * This isn't a typical situation -- either we got a
2427 * read error or a child silently returned bad data.
2428 * Read every block so we can try again with as much
2429 * data and parity as we can track down. If we've
2430 * already been through once before, all children will
2431 * be marked as tried so we'll proceed to combinatorial
2432 * reconstruction.
2433 */
2434 int nread = 0;
2435 for (int i = 0; i < rm->rm_nrows; i++) {
2436 nread += vdev_raidz_read_all(zio,
2437 rm->rm_row[i]);
2438 }
2439 if (nread != 0) {
2440 /*
2441 * Normally our stage is VDEV_IO_DONE, but if
2442 * we've already called redone(), it will have
2443 * changed to VDEV_IO_START, in which case we
2444 * don't want to call redone() again.
2445 */
2446 if (zio->io_stage != ZIO_STAGE_VDEV_IO_START)
2447 zio_vdev_io_redone(zio);
2448 return;
2449 }
2450
2451 zio->io_error = vdev_raidz_combrec(zio);
2452 if (zio->io_error == ECKSUM &&
2453 !(zio->io_flags & ZIO_FLAG_SPECULATIVE)) {
2454 vdev_raidz_io_done_unrecoverable(zio);
2455 }
2456 }
2457 }
2458 }
2459
2460 static void
2461 vdev_raidz_state_change(vdev_t *vd, int faulted, int degraded)
2462 {
2463 vdev_raidz_t *vdrz = vd->vdev_tsd;
2464 if (faulted > vdrz->vd_nparity)
2465 vdev_set_state(vd, B_FALSE, VDEV_STATE_CANT_OPEN,
2466 VDEV_AUX_NO_REPLICAS);
2467 else if (degraded + faulted != 0)
2468 vdev_set_state(vd, B_FALSE, VDEV_STATE_DEGRADED, VDEV_AUX_NONE);
2469 else
2470 vdev_set_state(vd, B_FALSE, VDEV_STATE_HEALTHY, VDEV_AUX_NONE);
2471 }
2472
2473 /*
2474 * Determine if any portion of the provided block resides on a child vdev
2475 * with a dirty DTL and therefore needs to be resilvered. The function
2476 * assumes that at least one DTL is dirty which implies that full stripe
2477 * width blocks must be resilvered.
2478 */
2479 static boolean_t
2480 vdev_raidz_need_resilver(vdev_t *vd, const dva_t *dva, size_t psize,
2481 uint64_t phys_birth)
2482 {
2483 vdev_raidz_t *vdrz = vd->vdev_tsd;
2484 uint64_t dcols = vd->vdev_children;
2485 uint64_t nparity = vdrz->vd_nparity;
2486 uint64_t ashift = vd->vdev_top->vdev_ashift;
2487 /* The starting RAIDZ (parent) vdev sector of the block. */
2488 uint64_t b = DVA_GET_OFFSET(dva) >> ashift;
2489 /* The zio's size in units of the vdev's minimum sector size. */
2490 uint64_t s = ((psize - 1) >> ashift) + 1;
2491 /* The first column for this stripe. */
2492 uint64_t f = b % dcols;
2493
2494 /* Unreachable by sequential resilver. */
2495 ASSERT3U(phys_birth, !=, TXG_UNKNOWN);
2496
2497 if (!vdev_dtl_contains(vd, DTL_PARTIAL, phys_birth, 1))
2498 return (B_FALSE);
2499
2500 if (s + nparity >= dcols)
2501 return (B_TRUE);
2502
2503 for (uint64_t c = 0; c < s + nparity; c++) {
2504 uint64_t devidx = (f + c) % dcols;
2505 vdev_t *cvd = vd->vdev_child[devidx];
2506
2507 /*
2508 * dsl_scan_need_resilver() already checked vd with
2509 * vdev_dtl_contains(). So here just check cvd with
2510 * vdev_dtl_empty(), cheaper and a good approximation.
2511 */
2512 if (!vdev_dtl_empty(cvd, DTL_PARTIAL))
2513 return (B_TRUE);
2514 }
2515
2516 return (B_FALSE);
2517 }
2518
2519 static void
2520 vdev_raidz_xlate(vdev_t *cvd, const range_seg64_t *logical_rs,
2521 range_seg64_t *physical_rs, range_seg64_t *remain_rs)
2522 {
2523 (void) remain_rs;
2524
2525 vdev_t *raidvd = cvd->vdev_parent;
2526 ASSERT(raidvd->vdev_ops == &vdev_raidz_ops);
2527
2528 uint64_t width = raidvd->vdev_children;
2529 uint64_t tgt_col = cvd->vdev_id;
2530 uint64_t ashift = raidvd->vdev_top->vdev_ashift;
2531
2532 /* make sure the offsets are block-aligned */
2533 ASSERT0(logical_rs->rs_start % (1 << ashift));
2534 ASSERT0(logical_rs->rs_end % (1 << ashift));
2535 uint64_t b_start = logical_rs->rs_start >> ashift;
2536 uint64_t b_end = logical_rs->rs_end >> ashift;
2537
2538 uint64_t start_row = 0;
2539 if (b_start > tgt_col) /* avoid underflow */
2540 start_row = ((b_start - tgt_col - 1) / width) + 1;
2541
2542 uint64_t end_row = 0;
2543 if (b_end > tgt_col)
2544 end_row = ((b_end - tgt_col - 1) / width) + 1;
2545
2546 physical_rs->rs_start = start_row << ashift;
2547 physical_rs->rs_end = end_row << ashift;
2548
2549 ASSERT3U(physical_rs->rs_start, <=, logical_rs->rs_start);
2550 ASSERT3U(physical_rs->rs_end - physical_rs->rs_start, <=,
2551 logical_rs->rs_end - logical_rs->rs_start);
2552 }
2553
2554 /*
2555 * Initialize private RAIDZ specific fields from the nvlist.
2556 */
2557 static int
2558 vdev_raidz_init(spa_t *spa, nvlist_t *nv, void **tsd)
2559 {
2560 vdev_raidz_t *vdrz;
2561 uint64_t nparity;
2562
2563 uint_t children;
2564 nvlist_t **child;
2565 int error = nvlist_lookup_nvlist_array(nv,
2566 ZPOOL_CONFIG_CHILDREN, &child, &children);
2567 if (error != 0)
2568 return (SET_ERROR(EINVAL));
2569
2570 if (nvlist_lookup_uint64(nv, ZPOOL_CONFIG_NPARITY, &nparity) == 0) {
2571 if (nparity == 0 || nparity > VDEV_RAIDZ_MAXPARITY)
2572 return (SET_ERROR(EINVAL));
2573
2574 /*
2575 * Previous versions could only support 1 or 2 parity
2576 * device.
2577 */
2578 if (nparity > 1 && spa_version(spa) < SPA_VERSION_RAIDZ2)
2579 return (SET_ERROR(EINVAL));
2580 else if (nparity > 2 && spa_version(spa) < SPA_VERSION_RAIDZ3)
2581 return (SET_ERROR(EINVAL));
2582 } else {
2583 /*
2584 * We require the parity to be specified for SPAs that
2585 * support multiple parity levels.
2586 */
2587 if (spa_version(spa) >= SPA_VERSION_RAIDZ2)
2588 return (SET_ERROR(EINVAL));
2589
2590 /*
2591 * Otherwise, we default to 1 parity device for RAID-Z.
2592 */
2593 nparity = 1;
2594 }
2595
2596 vdrz = kmem_zalloc(sizeof (*vdrz), KM_SLEEP);
2597 vdrz->vd_logical_width = children;
2598 vdrz->vd_nparity = nparity;
2599
2600 *tsd = vdrz;
2601
2602 return (0);
2603 }
2604
2605 static void
2606 vdev_raidz_fini(vdev_t *vd)
2607 {
2608 kmem_free(vd->vdev_tsd, sizeof (vdev_raidz_t));
2609 }
2610
2611 /*
2612 * Add RAIDZ specific fields to the config nvlist.
2613 */
2614 static void
2615 vdev_raidz_config_generate(vdev_t *vd, nvlist_t *nv)
2616 {
2617 ASSERT3P(vd->vdev_ops, ==, &vdev_raidz_ops);
2618 vdev_raidz_t *vdrz = vd->vdev_tsd;
2619
2620 /*
2621 * Make sure someone hasn't managed to sneak a fancy new vdev
2622 * into a crufty old storage pool.
2623 */
2624 ASSERT(vdrz->vd_nparity == 1 ||
2625 (vdrz->vd_nparity <= 2 &&
2626 spa_version(vd->vdev_spa) >= SPA_VERSION_RAIDZ2) ||
2627 (vdrz->vd_nparity <= 3 &&
2628 spa_version(vd->vdev_spa) >= SPA_VERSION_RAIDZ3));
2629
2630 /*
2631 * Note that we'll add these even on storage pools where they
2632 * aren't strictly required -- older software will just ignore
2633 * it.
2634 */
2635 fnvlist_add_uint64(nv, ZPOOL_CONFIG_NPARITY, vdrz->vd_nparity);
2636 }
2637
2638 static uint64_t
2639 vdev_raidz_nparity(vdev_t *vd)
2640 {
2641 vdev_raidz_t *vdrz = vd->vdev_tsd;
2642 return (vdrz->vd_nparity);
2643 }
2644
2645 static uint64_t
2646 vdev_raidz_ndisks(vdev_t *vd)
2647 {
2648 return (vd->vdev_children);
2649 }
2650
2651 vdev_ops_t vdev_raidz_ops = {
2652 .vdev_op_init = vdev_raidz_init,
2653 .vdev_op_fini = vdev_raidz_fini,
2654 .vdev_op_open = vdev_raidz_open,
2655 .vdev_op_close = vdev_raidz_close,
2656 .vdev_op_asize = vdev_raidz_asize,
2657 .vdev_op_min_asize = vdev_raidz_min_asize,
2658 .vdev_op_min_alloc = NULL,
2659 .vdev_op_io_start = vdev_raidz_io_start,
2660 .vdev_op_io_done = vdev_raidz_io_done,
2661 .vdev_op_state_change = vdev_raidz_state_change,
2662 .vdev_op_need_resilver = vdev_raidz_need_resilver,
2663 .vdev_op_hold = NULL,
2664 .vdev_op_rele = NULL,
2665 .vdev_op_remap = NULL,
2666 .vdev_op_xlate = vdev_raidz_xlate,
2667 .vdev_op_rebuild_asize = NULL,
2668 .vdev_op_metaslab_init = NULL,
2669 .vdev_op_config_generate = vdev_raidz_config_generate,
2670 .vdev_op_nparity = vdev_raidz_nparity,
2671 .vdev_op_ndisks = vdev_raidz_ndisks,
2672 .vdev_op_type = VDEV_TYPE_RAIDZ, /* name of this vdev type */
2673 .vdev_op_leaf = B_FALSE /* not a leaf vdev */
2674 };
OpenZFS on Linux and FreeBSD