1 /*- 2 * SPDX-License-Identifier: (BSD-2-Clause AND BSD-3-Clause) 3 * 4 * Copyright (c) 2003 Poul-Henning Kamp. 5 * Copyright (c) 1996, 1997 The NetBSD Foundation, Inc. 6 * All rights reserved. 7 * 8 * This code is derived from software contributed to The NetBSD Foundation 9 * by Jason R. Thorpe. 10 * 11 * Redistribution and use in source and binary forms, with or without 12 * modification, are permitted provided that the following conditions 13 * are met: 14 * 1. Redistributions of source code must retain the above copyright 15 * notice, this list of conditions and the following disclaimer. 16 * 2. Redistributions in binary form must reproduce the above copyright 17 * notice, this list of conditions and the following disclaimer in the 18 * documentation and/or other materials provided with the distribution. 19 * 20 * THIS SOFTWARE IS PROVIDED BY THE NETBSD FOUNDATION, INC. AND CONTRIBUTORS 21 * ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED 22 * TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR 23 * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE FOUNDATION OR CONTRIBUTORS 24 * BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR 25 * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF 26 * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS 27 * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN 28 * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) 29 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE 30 * POSSIBILITY OF SUCH DAMAGE. 31 * 32 * $NetBSD: ccd.c,v 1.22 1995/12/08 19:13:26 thorpej Exp $ 33 */ 34 35 /*- 36 * Copyright (c) 1988 University of Utah. 37 * Copyright (c) 1990, 1993 38 * The Regents of the University of California. All rights reserved. 39 * 40 * This code is derived from software contributed to Berkeley by 41 * the Systems Programming Group of the University of Utah Computer 42 * Science Department. 43 * 44 * Redistribution and use in source and binary forms, with or without 45 * modification, are permitted provided that the following conditions 46 * are met: 47 * 1. Redistributions of source code must retain the above copyright 48 * notice, this list of conditions and the following disclaimer. 49 * 2. Redistributions in binary form must reproduce the above copyright 50 * notice, this list of conditions and the following disclaimer in the 51 * documentation and/or other materials provided with the distribution. 52 * 3. Neither the name of the University nor the names of its contributors 53 * may be used to endorse or promote products derived from this software 54 * without specific prior written permission. 55 * 56 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND 57 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 58 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 59 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE 60 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 61 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 62 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 63 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 64 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 65 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 66 * SUCH DAMAGE. 67 * 68 * from: Utah $Hdr: cd.c 1.6 90/11/28$ 69 * 70 * @(#)cd.c 8.2 (Berkeley) 11/16/93 71 */ 72 73 /* 74 * Dynamic configuration and disklabel support by: 75 * Jason R. Thorpe <thorpej@nas.nasa.gov> 76 * Numerical Aerodynamic Simulation Facility 77 * Mail Stop 258-6 78 * NASA Ames Research Center 79 * Moffett Field, CA 94035 80 */ 81 82 #include <sys/cdefs.h> 83 __FBSDID("$FreeBSD$"); 84 85 #include <sys/param.h> 86 #include <sys/systm.h> 87 #include <sys/kernel.h> 88 #include <sys/module.h> 89 #include <sys/bio.h> 90 #include <sys/malloc.h> 91 #include <sys/sbuf.h> 92 #include <geom/geom.h> 93 94 /* 95 * Number of blocks to untouched in front of a component partition. 96 * This is to avoid violating its disklabel area when it starts at the 97 * beginning of the slice. 98 */ 99 #if !defined(CCD_OFFSET) 100 #define CCD_OFFSET 16 101 #endif 102 103 /* sc_flags */ 104 #define CCDF_UNIFORM 0x02 /* use LCCD of sizes for uniform interleave */ 105 #define CCDF_MIRROR 0x04 /* use mirroring */ 106 #define CCDF_NO_OFFSET 0x08 /* do not leave space in front */ 107 #define CCDF_LINUX 0x10 /* use Linux compatibility mode */ 108 109 /* Mask of user-settable ccd flags. */ 110 #define CCDF_USERMASK (CCDF_UNIFORM|CCDF_MIRROR) 111 112 /* 113 * Interleave description table. 114 * Computed at boot time to speed irregular-interleave lookups. 115 * The idea is that we interleave in "groups". First we interleave 116 * evenly over all component disks up to the size of the smallest 117 * component (the first group), then we interleave evenly over all 118 * remaining disks up to the size of the next-smallest (second group), 119 * and so on. 120 * 121 * Each table entry describes the interleave characteristics of one 122 * of these groups. For example if a concatenated disk consisted of 123 * three components of 5, 3, and 7 DEV_BSIZE blocks interleaved at 124 * DEV_BSIZE (1), the table would have three entries: 125 * 126 * ndisk startblk startoff dev 127 * 3 0 0 0, 1, 2 128 * 2 9 3 0, 2 129 * 1 13 5 2 130 * 0 - - - 131 * 132 * which says that the first nine blocks (0-8) are interleaved over 133 * 3 disks (0, 1, 2) starting at block offset 0 on any component disk, 134 * the next 4 blocks (9-12) are interleaved over 2 disks (0, 2) starting 135 * at component block 3, and the remaining blocks (13-14) are on disk 136 * 2 starting at offset 5. 137 */ 138 struct ccdiinfo { 139 int ii_ndisk; /* # of disks range is interleaved over */ 140 daddr_t ii_startblk; /* starting scaled block # for range */ 141 daddr_t ii_startoff; /* starting component offset (block #) */ 142 int *ii_index; /* ordered list of components in range */ 143 }; 144 145 /* 146 * Component info table. 147 * Describes a single component of a concatenated disk. 148 */ 149 struct ccdcinfo { 150 daddr_t ci_size; /* size */ 151 struct g_provider *ci_provider; /* provider */ 152 struct g_consumer *ci_consumer; /* consumer */ 153 }; 154 155 /* 156 * A concatenated disk is described by this structure. 157 */ 158 159 struct ccd_s { 160 LIST_ENTRY(ccd_s) list; 161 162 int sc_unit; /* logical unit number */ 163 int sc_flags; /* flags */ 164 daddr_t sc_size; /* size of ccd */ 165 int sc_ileave; /* interleave */ 166 u_int sc_ndisks; /* number of components */ 167 struct ccdcinfo *sc_cinfo; /* component info */ 168 struct ccdiinfo *sc_itable; /* interleave table */ 169 uint32_t sc_secsize; /* # bytes per sector */ 170 int sc_pick; /* side of mirror picked */ 171 daddr_t sc_blk[2]; /* mirror localization */ 172 uint32_t sc_offset; /* actual offset used */ 173 }; 174 175 static g_start_t g_ccd_start; 176 static void ccdiodone(struct bio *bp); 177 static void ccdinterleave(struct ccd_s *); 178 static int ccdinit(struct gctl_req *req, struct ccd_s *); 179 static int ccdbuffer(struct bio **ret, struct ccd_s *, 180 struct bio *, daddr_t, caddr_t, long); 181 182 static void 183 g_ccd_orphan(struct g_consumer *cp) 184 { 185 /* 186 * XXX: We don't do anything here. It is not obvious 187 * XXX: what DTRT would be, so we do what the previous 188 * XXX: code did: ignore it and let the user cope. 189 */ 190 } 191 192 static int 193 g_ccd_access(struct g_provider *pp, int dr, int dw, int de) 194 { 195 struct g_geom *gp; 196 struct g_consumer *cp1, *cp2; 197 int error; 198 199 de += dr; 200 de += dw; 201 202 gp = pp->geom; 203 error = ENXIO; 204 LIST_FOREACH(cp1, &gp->consumer, consumer) { 205 error = g_access(cp1, dr, dw, de); 206 if (error) { 207 LIST_FOREACH(cp2, &gp->consumer, consumer) { 208 if (cp1 == cp2) 209 break; 210 g_access(cp2, -dr, -dw, -de); 211 } 212 break; 213 } 214 } 215 return (error); 216 } 217 218 /* 219 * Free the softc and its substructures. 220 */ 221 static void 222 g_ccd_freesc(struct ccd_s *sc) 223 { 224 struct ccdiinfo *ii; 225 226 g_free(sc->sc_cinfo); 227 if (sc->sc_itable != NULL) { 228 for (ii = sc->sc_itable; ii->ii_ndisk > 0; ii++) 229 g_free(ii->ii_index); 230 g_free(sc->sc_itable); 231 } 232 g_free(sc); 233 } 234 235 static int 236 ccdinit(struct gctl_req *req, struct ccd_s *cs) 237 { 238 struct ccdcinfo *ci; 239 daddr_t size; 240 int ix; 241 daddr_t minsize; 242 int maxsecsize; 243 off_t mediasize; 244 u_int sectorsize; 245 246 cs->sc_size = 0; 247 248 maxsecsize = 0; 249 minsize = 0; 250 251 if (cs->sc_flags & CCDF_LINUX) { 252 cs->sc_offset = 0; 253 cs->sc_ileave *= 2; 254 if (cs->sc_flags & CCDF_MIRROR && cs->sc_ndisks != 2) 255 gctl_error(req, "Mirror mode for Linux raids is " 256 "only supported with 2 devices"); 257 } else { 258 if (cs->sc_flags & CCDF_NO_OFFSET) 259 cs->sc_offset = 0; 260 else 261 cs->sc_offset = CCD_OFFSET; 262 } 263 for (ix = 0; ix < cs->sc_ndisks; ix++) { 264 ci = &cs->sc_cinfo[ix]; 265 266 mediasize = ci->ci_provider->mediasize; 267 sectorsize = ci->ci_provider->sectorsize; 268 if (sectorsize > maxsecsize) 269 maxsecsize = sectorsize; 270 size = mediasize / DEV_BSIZE - cs->sc_offset; 271 272 /* Truncate to interleave boundary */ 273 274 if (cs->sc_ileave > 1) 275 size -= size % cs->sc_ileave; 276 277 if (size == 0) { 278 gctl_error(req, "Component %s has effective size zero", 279 ci->ci_provider->name); 280 return(ENODEV); 281 } 282 283 if (minsize == 0 || size < minsize) 284 minsize = size; 285 ci->ci_size = size; 286 cs->sc_size += size; 287 } 288 289 /* 290 * Don't allow the interleave to be smaller than 291 * the biggest component sector. 292 */ 293 if ((cs->sc_ileave > 0) && 294 (cs->sc_ileave < (maxsecsize / DEV_BSIZE))) { 295 gctl_error(req, "Interleave to small for sector size"); 296 return(EINVAL); 297 } 298 299 /* 300 * If uniform interleave is desired set all sizes to that of 301 * the smallest component. This will guarantee that a single 302 * interleave table is generated. 303 * 304 * Lost space must be taken into account when calculating the 305 * overall size. Half the space is lost when CCDF_MIRROR is 306 * specified. 307 */ 308 if (cs->sc_flags & CCDF_UNIFORM) { 309 for (ix = 0; ix < cs->sc_ndisks; ix++) { 310 ci = &cs->sc_cinfo[ix]; 311 ci->ci_size = minsize; 312 } 313 cs->sc_size = cs->sc_ndisks * minsize; 314 } 315 316 if (cs->sc_flags & CCDF_MIRROR) { 317 /* 318 * Check to see if an even number of components 319 * have been specified. The interleave must also 320 * be non-zero in order for us to be able to 321 * guarantee the topology. 322 */ 323 if (cs->sc_ndisks % 2) { 324 gctl_error(req, 325 "Mirroring requires an even number of disks"); 326 return(EINVAL); 327 } 328 if (cs->sc_ileave == 0) { 329 gctl_error(req, 330 "An interleave must be specified when mirroring"); 331 return(EINVAL); 332 } 333 cs->sc_size = (cs->sc_ndisks/2) * minsize; 334 } 335 336 /* 337 * Construct the interleave table. 338 */ 339 ccdinterleave(cs); 340 341 /* 342 * Create pseudo-geometry based on 1MB cylinders. It's 343 * pretty close. 344 */ 345 cs->sc_secsize = maxsecsize; 346 347 return (0); 348 } 349 350 static void 351 ccdinterleave(struct ccd_s *cs) 352 { 353 struct ccdcinfo *ci, *smallci; 354 struct ccdiinfo *ii; 355 daddr_t bn, lbn; 356 int ix; 357 daddr_t size; 358 359 /* 360 * Allocate an interleave table. The worst case occurs when each 361 * of N disks is of a different size, resulting in N interleave 362 * tables. 363 * 364 * Chances are this is too big, but we don't care. 365 */ 366 size = (cs->sc_ndisks + 1) * sizeof(struct ccdiinfo); 367 cs->sc_itable = g_malloc(size, M_WAITOK | M_ZERO); 368 369 /* 370 * Trivial case: no interleave (actually interleave of disk size). 371 * Each table entry represents a single component in its entirety. 372 * 373 * An interleave of 0 may not be used with a mirror setup. 374 */ 375 if (cs->sc_ileave == 0) { 376 bn = 0; 377 ii = cs->sc_itable; 378 379 for (ix = 0; ix < cs->sc_ndisks; ix++) { 380 /* Allocate space for ii_index. */ 381 ii->ii_index = g_malloc(sizeof(int), M_WAITOK); 382 ii->ii_ndisk = 1; 383 ii->ii_startblk = bn; 384 ii->ii_startoff = 0; 385 ii->ii_index[0] = ix; 386 bn += cs->sc_cinfo[ix].ci_size; 387 ii++; 388 } 389 ii->ii_ndisk = 0; 390 return; 391 } 392 393 /* 394 * The following isn't fast or pretty; it doesn't have to be. 395 */ 396 size = 0; 397 bn = lbn = 0; 398 for (ii = cs->sc_itable; ; ii++) { 399 /* 400 * Allocate space for ii_index. We might allocate more then 401 * we use. 402 */ 403 ii->ii_index = g_malloc((sizeof(int) * cs->sc_ndisks), 404 M_WAITOK); 405 406 /* 407 * Locate the smallest of the remaining components 408 */ 409 smallci = NULL; 410 for (ci = cs->sc_cinfo; ci < &cs->sc_cinfo[cs->sc_ndisks]; 411 ci++) { 412 if (ci->ci_size > size && 413 (smallci == NULL || 414 ci->ci_size < smallci->ci_size)) { 415 smallci = ci; 416 } 417 } 418 419 /* 420 * Nobody left, all done 421 */ 422 if (smallci == NULL) { 423 ii->ii_ndisk = 0; 424 g_free(ii->ii_index); 425 ii->ii_index = NULL; 426 break; 427 } 428 429 /* 430 * Record starting logical block using an sc_ileave blocksize. 431 */ 432 ii->ii_startblk = bn / cs->sc_ileave; 433 434 /* 435 * Record starting component block using an sc_ileave 436 * blocksize. This value is relative to the beginning of 437 * a component disk. 438 */ 439 ii->ii_startoff = lbn; 440 441 /* 442 * Determine how many disks take part in this interleave 443 * and record their indices. 444 */ 445 ix = 0; 446 for (ci = cs->sc_cinfo; 447 ci < &cs->sc_cinfo[cs->sc_ndisks]; ci++) { 448 if (ci->ci_size >= smallci->ci_size) { 449 ii->ii_index[ix++] = ci - cs->sc_cinfo; 450 } 451 } 452 ii->ii_ndisk = ix; 453 bn += ix * (smallci->ci_size - size); 454 lbn = smallci->ci_size / cs->sc_ileave; 455 size = smallci->ci_size; 456 } 457 } 458 459 static void 460 g_ccd_start(struct bio *bp) 461 { 462 long bcount, rcount; 463 struct bio *cbp[2]; 464 caddr_t addr; 465 daddr_t bn; 466 int err; 467 struct ccd_s *cs; 468 469 cs = bp->bio_to->geom->softc; 470 471 /* 472 * Block all GETATTR requests, we wouldn't know which of our 473 * subdevices we should ship it off to. 474 * XXX: this may not be the right policy. 475 */ 476 if(bp->bio_cmd == BIO_GETATTR) { 477 g_io_deliver(bp, EINVAL); 478 return; 479 } 480 481 /* 482 * Translate the partition-relative block number to an absolute. 483 */ 484 bn = bp->bio_offset / cs->sc_secsize; 485 486 /* 487 * Allocate component buffers and fire off the requests 488 */ 489 addr = bp->bio_data; 490 for (bcount = bp->bio_length; bcount > 0; bcount -= rcount) { 491 err = ccdbuffer(cbp, cs, bp, bn, addr, bcount); 492 if (err) { 493 bp->bio_completed += bcount; 494 if (bp->bio_error == 0) 495 bp->bio_error = err; 496 if (bp->bio_completed == bp->bio_length) 497 g_io_deliver(bp, bp->bio_error); 498 return; 499 } 500 rcount = cbp[0]->bio_length; 501 502 if (cs->sc_flags & CCDF_MIRROR) { 503 /* 504 * Mirroring. Writes go to both disks, reads are 505 * taken from whichever disk seems most appropriate. 506 * 507 * We attempt to localize reads to the disk whos arm 508 * is nearest the read request. We ignore seeks due 509 * to writes when making this determination and we 510 * also try to avoid hogging. 511 */ 512 if (cbp[0]->bio_cmd != BIO_READ) { 513 g_io_request(cbp[0], cbp[0]->bio_from); 514 g_io_request(cbp[1], cbp[1]->bio_from); 515 } else { 516 int pick = cs->sc_pick; 517 daddr_t range = cs->sc_size / 16; 518 519 if (bn < cs->sc_blk[pick] - range || 520 bn > cs->sc_blk[pick] + range 521 ) { 522 cs->sc_pick = pick = 1 - pick; 523 } 524 cs->sc_blk[pick] = bn + btodb(rcount); 525 g_io_request(cbp[pick], cbp[pick]->bio_from); 526 } 527 } else { 528 /* 529 * Not mirroring 530 */ 531 g_io_request(cbp[0], cbp[0]->bio_from); 532 } 533 bn += btodb(rcount); 534 addr += rcount; 535 } 536 } 537 538 /* 539 * Build a component buffer header. 540 */ 541 static int 542 ccdbuffer(struct bio **cb, struct ccd_s *cs, struct bio *bp, daddr_t bn, caddr_t addr, long bcount) 543 { 544 struct ccdcinfo *ci, *ci2 = NULL; 545 struct bio *cbp; 546 daddr_t cbn, cboff; 547 off_t cbc; 548 549 /* 550 * Determine which component bn falls in. 551 */ 552 cbn = bn; 553 cboff = 0; 554 555 if (cs->sc_ileave == 0) { 556 /* 557 * Serially concatenated and neither a mirror nor a parity 558 * config. This is a special case. 559 */ 560 daddr_t sblk; 561 562 sblk = 0; 563 for (ci = cs->sc_cinfo; cbn >= sblk + ci->ci_size; ci++) 564 sblk += ci->ci_size; 565 cbn -= sblk; 566 } else { 567 struct ccdiinfo *ii; 568 int ccdisk, off; 569 570 /* 571 * Calculate cbn, the logical superblock (sc_ileave chunks), 572 * and cboff, a normal block offset (DEV_BSIZE chunks) relative 573 * to cbn. 574 */ 575 cboff = cbn % cs->sc_ileave; /* DEV_BSIZE gran */ 576 cbn = cbn / cs->sc_ileave; /* DEV_BSIZE * ileave gran */ 577 578 /* 579 * Figure out which interleave table to use. 580 */ 581 for (ii = cs->sc_itable; ii->ii_ndisk; ii++) { 582 if (ii->ii_startblk > cbn) 583 break; 584 } 585 ii--; 586 587 /* 588 * off is the logical superblock relative to the beginning 589 * of this interleave block. 590 */ 591 off = cbn - ii->ii_startblk; 592 593 /* 594 * We must calculate which disk component to use (ccdisk), 595 * and recalculate cbn to be the superblock relative to 596 * the beginning of the component. This is typically done by 597 * adding 'off' and ii->ii_startoff together. However, 'off' 598 * must typically be divided by the number of components in 599 * this interleave array to be properly convert it from a 600 * CCD-relative logical superblock number to a 601 * component-relative superblock number. 602 */ 603 if (ii->ii_ndisk == 1) { 604 /* 605 * When we have just one disk, it can't be a mirror 606 * or a parity config. 607 */ 608 ccdisk = ii->ii_index[0]; 609 cbn = ii->ii_startoff + off; 610 } else { 611 if (cs->sc_flags & CCDF_MIRROR) { 612 /* 613 * We have forced a uniform mapping, resulting 614 * in a single interleave array. We double 615 * up on the first half of the available 616 * components and our mirror is in the second 617 * half. This only works with a single 618 * interleave array because doubling up 619 * doubles the number of sectors, so there 620 * cannot be another interleave array because 621 * the next interleave array's calculations 622 * would be off. 623 */ 624 int ndisk2 = ii->ii_ndisk / 2; 625 ccdisk = ii->ii_index[off % ndisk2]; 626 cbn = ii->ii_startoff + off / ndisk2; 627 ci2 = &cs->sc_cinfo[ccdisk + ndisk2]; 628 } else { 629 ccdisk = ii->ii_index[off % ii->ii_ndisk]; 630 cbn = ii->ii_startoff + off / ii->ii_ndisk; 631 } 632 } 633 634 ci = &cs->sc_cinfo[ccdisk]; 635 636 /* 637 * Convert cbn from a superblock to a normal block so it 638 * can be used to calculate (along with cboff) the normal 639 * block index into this particular disk. 640 */ 641 cbn *= cs->sc_ileave; 642 } 643 644 /* 645 * Fill in the component buf structure. 646 */ 647 cbp = g_clone_bio(bp); 648 if (cbp == NULL) 649 return (ENOMEM); 650 cbp->bio_done = g_std_done; 651 cbp->bio_offset = dbtob(cbn + cboff + cs->sc_offset); 652 cbp->bio_data = addr; 653 if (cs->sc_ileave == 0) 654 cbc = dbtob((off_t)(ci->ci_size - cbn)); 655 else 656 cbc = dbtob((off_t)(cs->sc_ileave - cboff)); 657 cbp->bio_length = (cbc < bcount) ? cbc : bcount; 658 659 cbp->bio_from = ci->ci_consumer; 660 cb[0] = cbp; 661 662 if (cs->sc_flags & CCDF_MIRROR) { 663 cbp = g_clone_bio(bp); 664 if (cbp == NULL) 665 return (ENOMEM); 666 cbp->bio_done = cb[0]->bio_done = ccdiodone; 667 cbp->bio_offset = cb[0]->bio_offset; 668 cbp->bio_data = cb[0]->bio_data; 669 cbp->bio_length = cb[0]->bio_length; 670 cbp->bio_from = ci2->ci_consumer; 671 cbp->bio_caller1 = cb[0]; 672 cb[0]->bio_caller1 = cbp; 673 cb[1] = cbp; 674 } 675 return (0); 676 } 677 678 /* 679 * Called only for mirrored operations. 680 */ 681 static void 682 ccdiodone(struct bio *cbp) 683 { 684 struct bio *mbp, *pbp; 685 686 mbp = cbp->bio_caller1; 687 pbp = cbp->bio_parent; 688 689 if (pbp->bio_cmd == BIO_READ) { 690 if (cbp->bio_error == 0) { 691 /* We will not be needing the partner bio */ 692 if (mbp != NULL) { 693 pbp->bio_inbed++; 694 g_destroy_bio(mbp); 695 } 696 g_std_done(cbp); 697 return; 698 } 699 if (mbp != NULL) { 700 /* Try partner the bio instead */ 701 mbp->bio_caller1 = NULL; 702 pbp->bio_inbed++; 703 g_destroy_bio(cbp); 704 g_io_request(mbp, mbp->bio_from); 705 /* 706 * XXX: If this comes back OK, we should actually 707 * try to write the good data on the failed mirror 708 */ 709 return; 710 } 711 g_std_done(cbp); 712 return; 713 } 714 if (mbp != NULL) { 715 mbp->bio_caller1 = NULL; 716 pbp->bio_inbed++; 717 if (cbp->bio_error != 0 && pbp->bio_error == 0) 718 pbp->bio_error = cbp->bio_error; 719 g_destroy_bio(cbp); 720 return; 721 } 722 g_std_done(cbp); 723 } 724 725 static void 726 g_ccd_create(struct gctl_req *req, struct g_class *mp) 727 { 728 int *unit, *ileave, *nprovider; 729 struct g_geom *gp; 730 struct g_consumer *cp; 731 struct g_provider *pp; 732 struct ccd_s *sc; 733 struct sbuf *sb; 734 char buf[20]; 735 int i, error; 736 737 g_topology_assert(); 738 unit = gctl_get_paraml(req, "unit", sizeof (*unit)); 739 if (unit == NULL) { 740 gctl_error(req, "unit parameter not given"); 741 return; 742 } 743 ileave = gctl_get_paraml(req, "ileave", sizeof (*ileave)); 744 if (ileave == NULL) { 745 gctl_error(req, "ileave parameter not given"); 746 return; 747 } 748 nprovider = gctl_get_paraml(req, "nprovider", sizeof (*nprovider)); 749 if (nprovider == NULL) { 750 gctl_error(req, "nprovider parameter not given"); 751 return; 752 } 753 754 /* Check for duplicate unit */ 755 LIST_FOREACH(gp, &mp->geom, geom) { 756 sc = gp->softc; 757 if (sc != NULL && sc->sc_unit == *unit) { 758 gctl_error(req, "Unit %d already configured", *unit); 759 return; 760 } 761 } 762 763 if (*nprovider <= 0) { 764 gctl_error(req, "Bogus nprovider argument (= %d)", *nprovider); 765 return; 766 } 767 768 /* Check all providers are valid */ 769 for (i = 0; i < *nprovider; i++) { 770 snprintf(buf, sizeof(buf), "provider%d", i); 771 pp = gctl_get_provider(req, buf); 772 if (pp == NULL) 773 return; 774 } 775 776 gp = g_new_geomf(mp, "ccd%d", *unit); 777 sc = g_malloc(sizeof *sc, M_WAITOK | M_ZERO); 778 gp->softc = sc; 779 sc->sc_ndisks = *nprovider; 780 781 /* Allocate space for the component info. */ 782 sc->sc_cinfo = g_malloc(sc->sc_ndisks * sizeof(struct ccdcinfo), 783 M_WAITOK | M_ZERO); 784 785 /* Create consumers and attach to all providers */ 786 for (i = 0; i < *nprovider; i++) { 787 snprintf(buf, sizeof(buf), "provider%d", i); 788 pp = gctl_get_provider(req, buf); 789 cp = g_new_consumer(gp); 790 error = g_attach(cp, pp); 791 KASSERT(error == 0, ("attach to %s failed", pp->name)); 792 sc->sc_cinfo[i].ci_consumer = cp; 793 sc->sc_cinfo[i].ci_provider = pp; 794 } 795 796 sc->sc_unit = *unit; 797 sc->sc_ileave = *ileave; 798 799 if (gctl_get_param(req, "no_offset", NULL)) 800 sc->sc_flags |= CCDF_NO_OFFSET; 801 if (gctl_get_param(req, "linux", NULL)) 802 sc->sc_flags |= CCDF_LINUX; 803 804 if (gctl_get_param(req, "uniform", NULL)) 805 sc->sc_flags |= CCDF_UNIFORM; 806 if (gctl_get_param(req, "mirror", NULL)) 807 sc->sc_flags |= CCDF_MIRROR; 808 809 if (sc->sc_ileave == 0 && (sc->sc_flags & CCDF_MIRROR)) { 810 printf("%s: disabling mirror, interleave is 0\n", gp->name); 811 sc->sc_flags &= ~(CCDF_MIRROR); 812 } 813 814 if ((sc->sc_flags & CCDF_MIRROR) && !(sc->sc_flags & CCDF_UNIFORM)) { 815 printf("%s: mirror/parity forces uniform flag\n", gp->name); 816 sc->sc_flags |= CCDF_UNIFORM; 817 } 818 819 error = ccdinit(req, sc); 820 if (error != 0) { 821 g_ccd_freesc(sc); 822 gp->softc = NULL; 823 g_wither_geom(gp, ENXIO); 824 return; 825 } 826 827 pp = g_new_providerf(gp, "%s", gp->name); 828 pp->mediasize = sc->sc_size * (off_t)sc->sc_secsize; 829 pp->sectorsize = sc->sc_secsize; 830 g_error_provider(pp, 0); 831 832 sb = sbuf_new_auto(); 833 sbuf_printf(sb, "ccd%d: %d components ", sc->sc_unit, *nprovider); 834 for (i = 0; i < *nprovider; i++) { 835 sbuf_printf(sb, "%s%s", 836 i == 0 ? "(" : ", ", 837 sc->sc_cinfo[i].ci_provider->name); 838 } 839 sbuf_printf(sb, "), %jd blocks ", (off_t)pp->mediasize / DEV_BSIZE); 840 if (sc->sc_ileave != 0) 841 sbuf_printf(sb, "interleaved at %d blocks\n", 842 sc->sc_ileave); 843 else 844 sbuf_printf(sb, "concatenated\n"); 845 sbuf_finish(sb); 846 gctl_set_param_err(req, "output", sbuf_data(sb), sbuf_len(sb) + 1); 847 sbuf_delete(sb); 848 } 849 850 static int 851 g_ccd_destroy_geom(struct gctl_req *req, struct g_class *mp, struct g_geom *gp) 852 { 853 struct g_provider *pp; 854 struct ccd_s *sc; 855 856 g_topology_assert(); 857 sc = gp->softc; 858 pp = LIST_FIRST(&gp->provider); 859 if (sc == NULL || pp == NULL) 860 return (EBUSY); 861 if (pp->acr != 0 || pp->acw != 0 || pp->ace != 0) { 862 gctl_error(req, "%s is open(r%dw%de%d)", gp->name, 863 pp->acr, pp->acw, pp->ace); 864 return (EBUSY); 865 } 866 g_ccd_freesc(sc); 867 gp->softc = NULL; 868 g_wither_geom(gp, ENXIO); 869 return (0); 870 } 871 872 static void 873 g_ccd_list(struct gctl_req *req, struct g_class *mp) 874 { 875 struct sbuf *sb; 876 struct ccd_s *cs; 877 struct g_geom *gp; 878 int i, unit, *up; 879 880 up = gctl_get_paraml(req, "unit", sizeof (*up)); 881 if (up == NULL) { 882 gctl_error(req, "unit parameter not given"); 883 return; 884 } 885 unit = *up; 886 sb = sbuf_new_auto(); 887 LIST_FOREACH(gp, &mp->geom, geom) { 888 cs = gp->softc; 889 if (cs == NULL || (unit >= 0 && unit != cs->sc_unit)) 890 continue; 891 sbuf_printf(sb, "ccd%d\t\t%d\t%d\t", 892 cs->sc_unit, cs->sc_ileave, cs->sc_flags & CCDF_USERMASK); 893 894 for (i = 0; i < cs->sc_ndisks; ++i) { 895 sbuf_printf(sb, "%s/dev/%s", i == 0 ? "" : " ", 896 cs->sc_cinfo[i].ci_provider->name); 897 } 898 sbuf_printf(sb, "\n"); 899 } 900 sbuf_finish(sb); 901 gctl_set_param_err(req, "output", sbuf_data(sb), sbuf_len(sb) + 1); 902 sbuf_delete(sb); 903 } 904 905 static void 906 g_ccd_config(struct gctl_req *req, struct g_class *mp, char const *verb) 907 { 908 struct g_geom *gp; 909 910 g_topology_assert(); 911 if (!strcmp(verb, "create geom")) { 912 g_ccd_create(req, mp); 913 } else if (!strcmp(verb, "destroy geom")) { 914 gp = gctl_get_geom(req, mp, "geom"); 915 if (gp != NULL) 916 g_ccd_destroy_geom(req, mp, gp); 917 } else if (!strcmp(verb, "list")) { 918 g_ccd_list(req, mp); 919 } else { 920 gctl_error(req, "unknown verb"); 921 } 922 } 923 924 static struct g_class g_ccd_class = { 925 .name = "CCD", 926 .version = G_VERSION, 927 .ctlreq = g_ccd_config, 928 .destroy_geom = g_ccd_destroy_geom, 929 .start = g_ccd_start, 930 .orphan = g_ccd_orphan, 931 .access = g_ccd_access, 932 }; 933 934 DECLARE_GEOM_CLASS(g_ccd_class, g_ccd); 935 MODULE_VERSION(geom_ccd, 0); 936