rf_netbsdkintf.c revision 1.202.6.3
1 /* $NetBSD: rf_netbsdkintf.c,v 1.202.6.3 2006/05/24 15:50:29 tron Exp $ */ 2 /*- 3 * Copyright (c) 1996, 1997, 1998 The NetBSD Foundation, Inc. 4 * All rights reserved. 5 * 6 * This code is derived from software contributed to The NetBSD Foundation 7 * by Greg Oster; Jason R. Thorpe. 8 * 9 * Redistribution and use in source and binary forms, with or without 10 * modification, are permitted provided that the following conditions 11 * are met: 12 * 1. Redistributions of source code must retain the above copyright 13 * notice, this list of conditions and the following disclaimer. 14 * 2. Redistributions in binary form must reproduce the above copyright 15 * notice, this list of conditions and the following disclaimer in the 16 * documentation and/or other materials provided with the distribution. 17 * 3. All advertising materials mentioning features or use of this software 18 * must display the following acknowledgement: 19 * This product includes software developed by the NetBSD 20 * Foundation, Inc. and its contributors. 21 * 4. Neither the name of The NetBSD Foundation nor the names of its 22 * contributors may be used to endorse or promote products derived 23 * from this software without specific prior written permission. 24 * 25 * THIS SOFTWARE IS PROVIDED BY THE NETBSD FOUNDATION, INC. AND CONTRIBUTORS 26 * ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED 27 * TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR 28 * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE FOUNDATION OR CONTRIBUTORS 29 * BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR 30 * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF 31 * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS 32 * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN 33 * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) 34 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE 35 * POSSIBILITY OF SUCH DAMAGE. 36 */ 37 38 /* 39 * Copyright (c) 1990, 1993 40 * The Regents of the University of California. All rights reserved. 41 * 42 * This code is derived from software contributed to Berkeley by 43 * the Systems Programming Group of the University of Utah Computer 44 * Science Department. 45 * 46 * Redistribution and use in source and binary forms, with or without 47 * modification, are permitted provided that the following conditions 48 * are met: 49 * 1. Redistributions of source code must retain the above copyright 50 * notice, this list of conditions and the following disclaimer. 51 * 2. Redistributions in binary form must reproduce the above copyright 52 * notice, this list of conditions and the following disclaimer in the 53 * documentation and/or other materials provided with the distribution. 54 * 3. Neither the name of the University nor the names of its contributors 55 * may be used to endorse or promote products derived from this software 56 * without specific prior written permission. 57 * 58 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND 59 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 60 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 61 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE 62 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 63 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 64 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 65 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 66 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 67 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 68 * SUCH DAMAGE. 69 * 70 * from: Utah $Hdr: cd.c 1.6 90/11/28$ 71 * 72 * @(#)cd.c 8.2 (Berkeley) 11/16/93 73 */ 74 75 /* 76 * Copyright (c) 1988 University of Utah. 77 * 78 * This code is derived from software contributed to Berkeley by 79 * the Systems Programming Group of the University of Utah Computer 80 * Science Department. 81 * 82 * Redistribution and use in source and binary forms, with or without 83 * modification, are permitted provided that the following conditions 84 * are met: 85 * 1. Redistributions of source code must retain the above copyright 86 * notice, this list of conditions and the following disclaimer. 87 * 2. Redistributions in binary form must reproduce the above copyright 88 * notice, this list of conditions and the following disclaimer in the 89 * documentation and/or other materials provided with the distribution. 90 * 3. All advertising materials mentioning features or use of this software 91 * must display the following acknowledgement: 92 * This product includes software developed by the University of 93 * California, Berkeley and its contributors. 94 * 4. Neither the name of the University nor the names of its contributors 95 * may be used to endorse or promote products derived from this software 96 * without specific prior written permission. 97 * 98 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND 99 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 100 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 101 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE 102 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 103 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 104 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 105 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 106 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 107 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 108 * SUCH DAMAGE. 109 * 110 * from: Utah $Hdr: cd.c 1.6 90/11/28$ 111 * 112 * @(#)cd.c 8.2 (Berkeley) 11/16/93 113 */ 114 115 /* 116 * Copyright (c) 1995 Carnegie-Mellon University. 117 * All rights reserved. 118 * 119 * Authors: Mark Holland, Jim Zelenka 120 * 121 * Permission to use, copy, modify and distribute this software and 122 * its documentation is hereby granted, provided that both the copyright 123 * notice and this permission notice appear in all copies of the 124 * software, derivative works or modified versions, and any portions 125 * thereof, and that both notices appear in supporting documentation. 126 * 127 * CARNEGIE MELLON ALLOWS FREE USE OF THIS SOFTWARE IN ITS "AS IS" 128 * CONDITION. CARNEGIE MELLON DISCLAIMS ANY LIABILITY OF ANY KIND 129 * FOR ANY DAMAGES WHATSOEVER RESULTING FROM THE USE OF THIS SOFTWARE. 130 * 131 * Carnegie Mellon requests users of this software to return to 132 * 133 * Software Distribution Coordinator or Software.Distribution (at) CS.CMU.EDU 134 * School of Computer Science 135 * Carnegie Mellon University 136 * Pittsburgh PA 15213-3890 137 * 138 * any improvements or extensions that they make and grant Carnegie the 139 * rights to redistribute these changes. 140 */ 141 142 /*********************************************************** 143 * 144 * rf_kintf.c -- the kernel interface routines for RAIDframe 145 * 146 ***********************************************************/ 147 148 #include <sys/cdefs.h> 149 __KERNEL_RCSID(0, "$NetBSD: rf_netbsdkintf.c,v 1.202.6.3 2006/05/24 15:50:29 tron Exp $"); 150 151 #include <sys/param.h> 152 #include <sys/errno.h> 153 #include <sys/pool.h> 154 #include <sys/proc.h> 155 #include <sys/queue.h> 156 #include <sys/disk.h> 157 #include <sys/device.h> 158 #include <sys/stat.h> 159 #include <sys/ioctl.h> 160 #include <sys/fcntl.h> 161 #include <sys/systm.h> 162 #include <sys/namei.h> 163 #include <sys/vnode.h> 164 #include <sys/disklabel.h> 165 #include <sys/conf.h> 166 #include <sys/lock.h> 167 #include <sys/buf.h> 168 #include <sys/bufq.h> 169 #include <sys/user.h> 170 #include <sys/reboot.h> 171 #include <sys/kauth.h> 172 173 #include <dev/raidframe/raidframevar.h> 174 #include <dev/raidframe/raidframeio.h> 175 #include "raid.h" 176 #include "opt_raid_autoconfig.h" 177 #include "rf_raid.h" 178 #include "rf_copyback.h" 179 #include "rf_dag.h" 180 #include "rf_dagflags.h" 181 #include "rf_desc.h" 182 #include "rf_diskqueue.h" 183 #include "rf_etimer.h" 184 #include "rf_general.h" 185 #include "rf_kintf.h" 186 #include "rf_options.h" 187 #include "rf_driver.h" 188 #include "rf_parityscan.h" 189 #include "rf_threadstuff.h" 190 191 #ifdef DEBUG 192 int rf_kdebug_level = 0; 193 #define db1_printf(a) if (rf_kdebug_level > 0) printf a 194 #else /* DEBUG */ 195 #define db1_printf(a) { } 196 #endif /* DEBUG */ 197 198 static RF_Raid_t **raidPtrs; /* global raid device descriptors */ 199 200 RF_DECLARE_STATIC_MUTEX(rf_sparet_wait_mutex) 201 202 static RF_SparetWait_t *rf_sparet_wait_queue; /* requests to install a 203 * spare table */ 204 static RF_SparetWait_t *rf_sparet_resp_queue; /* responses from 205 * installation process */ 206 207 MALLOC_DEFINE(M_RAIDFRAME, "RAIDframe", "RAIDframe structures"); 208 209 /* prototypes */ 210 static void KernelWakeupFunc(struct buf *); 211 static void InitBP(struct buf *, struct vnode *, unsigned, 212 dev_t, RF_SectorNum_t, RF_SectorCount_t, caddr_t, void (*) (struct buf *), 213 void *, int, struct proc *); 214 static void raidinit(RF_Raid_t *); 215 216 void raidattach(int); 217 218 dev_type_open(raidopen); 219 dev_type_close(raidclose); 220 dev_type_read(raidread); 221 dev_type_write(raidwrite); 222 dev_type_ioctl(raidioctl); 223 dev_type_strategy(raidstrategy); 224 dev_type_dump(raiddump); 225 dev_type_size(raidsize); 226 227 const struct bdevsw raid_bdevsw = { 228 raidopen, raidclose, raidstrategy, raidioctl, 229 raiddump, raidsize, D_DISK 230 }; 231 232 const struct cdevsw raid_cdevsw = { 233 raidopen, raidclose, raidread, raidwrite, raidioctl, 234 nostop, notty, nopoll, nommap, nokqfilter, D_DISK 235 }; 236 237 /* XXX Not sure if the following should be replacing the raidPtrs above, 238 or if it should be used in conjunction with that... 239 */ 240 241 struct raid_softc { 242 int sc_flags; /* flags */ 243 int sc_cflags; /* configuration flags */ 244 size_t sc_size; /* size of the raid device */ 245 char sc_xname[20]; /* XXX external name */ 246 struct disk sc_dkdev; /* generic disk device info */ 247 struct bufq_state *buf_queue; /* used for the device queue */ 248 }; 249 /* sc_flags */ 250 #define RAIDF_INITED 0x01 /* unit has been initialized */ 251 #define RAIDF_WLABEL 0x02 /* label area is writable */ 252 #define RAIDF_LABELLING 0x04 /* unit is currently being labelled */ 253 #define RAIDF_WANTED 0x40 /* someone is waiting to obtain a lock */ 254 #define RAIDF_LOCKED 0x80 /* unit is locked */ 255 256 #define raidunit(x) DISKUNIT(x) 257 int numraid = 0; 258 259 extern struct cfdriver raid_cd; 260 261 /* 262 * Allow RAIDOUTSTANDING number of simultaneous IO's to this RAID device. 263 * Be aware that large numbers can allow the driver to consume a lot of 264 * kernel memory, especially on writes, and in degraded mode reads. 265 * 266 * For example: with a stripe width of 64 blocks (32k) and 5 disks, 267 * a single 64K write will typically require 64K for the old data, 268 * 64K for the old parity, and 64K for the new parity, for a total 269 * of 192K (if the parity buffer is not re-used immediately). 270 * Even it if is used immediately, that's still 128K, which when multiplied 271 * by say 10 requests, is 1280K, *on top* of the 640K of incoming data. 272 * 273 * Now in degraded mode, for example, a 64K read on the above setup may 274 * require data reconstruction, which will require *all* of the 4 remaining 275 * disks to participate -- 4 * 32K/disk == 128K again. 276 */ 277 278 #ifndef RAIDOUTSTANDING 279 #define RAIDOUTSTANDING 6 280 #endif 281 282 #define RAIDLABELDEV(dev) \ 283 (MAKEDISKDEV(major((dev)), raidunit((dev)), RAW_PART)) 284 285 /* declared here, and made public, for the benefit of KVM stuff.. */ 286 struct raid_softc *raid_softc; 287 288 static void raidgetdefaultlabel(RF_Raid_t *, struct raid_softc *, 289 struct disklabel *); 290 static void raidgetdisklabel(dev_t); 291 static void raidmakedisklabel(struct raid_softc *); 292 293 static int raidlock(struct raid_softc *); 294 static void raidunlock(struct raid_softc *); 295 296 static void rf_markalldirty(RF_Raid_t *); 297 298 struct device *raidrootdev; 299 300 void rf_ReconThread(struct rf_recon_req *); 301 void rf_RewriteParityThread(RF_Raid_t *raidPtr); 302 void rf_CopybackThread(RF_Raid_t *raidPtr); 303 void rf_ReconstructInPlaceThread(struct rf_recon_req *); 304 int rf_autoconfig(struct device *self); 305 void rf_buildroothack(RF_ConfigSet_t *); 306 307 RF_AutoConfig_t *rf_find_raid_components(void); 308 RF_ConfigSet_t *rf_create_auto_sets(RF_AutoConfig_t *); 309 static int rf_does_it_fit(RF_ConfigSet_t *,RF_AutoConfig_t *); 310 static int rf_reasonable_label(RF_ComponentLabel_t *); 311 void rf_create_configuration(RF_AutoConfig_t *,RF_Config_t *, RF_Raid_t *); 312 int rf_set_autoconfig(RF_Raid_t *, int); 313 int rf_set_rootpartition(RF_Raid_t *, int); 314 void rf_release_all_vps(RF_ConfigSet_t *); 315 void rf_cleanup_config_set(RF_ConfigSet_t *); 316 int rf_have_enough_components(RF_ConfigSet_t *); 317 int rf_auto_config_set(RF_ConfigSet_t *, int *); 318 319 static int raidautoconfig = 0; /* Debugging, mostly. Set to 0 to not 320 allow autoconfig to take place. 321 Note that this is overridden by having 322 RAID_AUTOCONFIG as an option in the 323 kernel config file. */ 324 325 struct RF_Pools_s rf_pools; 326 327 void 328 raidattach(int num) 329 { 330 int raidID; 331 int i, rc; 332 333 #ifdef DEBUG 334 printf("raidattach: Asked for %d units\n", num); 335 #endif 336 337 if (num <= 0) { 338 #ifdef DIAGNOSTIC 339 panic("raidattach: count <= 0"); 340 #endif 341 return; 342 } 343 /* This is where all the initialization stuff gets done. */ 344 345 numraid = num; 346 347 /* Make some space for requested number of units... */ 348 349 RF_Malloc(raidPtrs, num * sizeof(RF_Raid_t *), (RF_Raid_t **)); 350 if (raidPtrs == NULL) { 351 panic("raidPtrs is NULL!!"); 352 } 353 354 rf_mutex_init(&rf_sparet_wait_mutex); 355 356 rf_sparet_wait_queue = rf_sparet_resp_queue = NULL; 357 358 for (i = 0; i < num; i++) 359 raidPtrs[i] = NULL; 360 rc = rf_BootRaidframe(); 361 if (rc == 0) 362 printf("Kernelized RAIDframe activated\n"); 363 else 364 panic("Serious error booting RAID!!"); 365 366 /* put together some datastructures like the CCD device does.. This 367 * lets us lock the device and what-not when it gets opened. */ 368 369 raid_softc = (struct raid_softc *) 370 malloc(num * sizeof(struct raid_softc), 371 M_RAIDFRAME, M_NOWAIT); 372 if (raid_softc == NULL) { 373 printf("WARNING: no memory for RAIDframe driver\n"); 374 return; 375 } 376 377 memset(raid_softc, 0, num * sizeof(struct raid_softc)); 378 379 raidrootdev = (struct device *)malloc(num * sizeof(struct device), 380 M_RAIDFRAME, M_NOWAIT); 381 if (raidrootdev == NULL) { 382 panic("No memory for RAIDframe driver!!?!?!"); 383 } 384 385 for (raidID = 0; raidID < num; raidID++) { 386 bufq_alloc(&raid_softc[raidID].buf_queue, "fcfs", 0); 387 pseudo_disk_init(&raid_softc[raidID].sc_dkdev); 388 389 /* XXXJRT Should use config_attach_pseudo() */ 390 391 raidrootdev[raidID].dv_class = DV_DISK; 392 raidrootdev[raidID].dv_cfdata = NULL; 393 raidrootdev[raidID].dv_unit = raidID; 394 raidrootdev[raidID].dv_parent = NULL; 395 raidrootdev[raidID].dv_flags = 0; 396 raidrootdev[raidID].dv_cfdriver = &raid_cd; 397 snprintf(raidrootdev[raidID].dv_xname, 398 sizeof(raidrootdev[raidID].dv_xname), "raid%d", raidID); 399 400 RF_Malloc(raidPtrs[raidID], sizeof(RF_Raid_t), 401 (RF_Raid_t *)); 402 if (raidPtrs[raidID] == NULL) { 403 printf("WARNING: raidPtrs[%d] is NULL\n", raidID); 404 numraid = raidID; 405 return; 406 } 407 } 408 409 #ifdef RAID_AUTOCONFIG 410 raidautoconfig = 1; 411 #endif 412 413 /* 414 * Register a finalizer which will be used to auto-config RAID 415 * sets once all real hardware devices have been found. 416 */ 417 if (config_finalize_register(NULL, rf_autoconfig) != 0) 418 printf("WARNING: unable to register RAIDframe finalizer\n"); 419 } 420 421 int 422 rf_autoconfig(struct device *self) 423 { 424 RF_AutoConfig_t *ac_list; 425 RF_ConfigSet_t *config_sets; 426 427 if (raidautoconfig == 0) 428 return (0); 429 430 /* XXX This code can only be run once. */ 431 raidautoconfig = 0; 432 433 /* 1. locate all RAID components on the system */ 434 #ifdef DEBUG 435 printf("Searching for RAID components...\n"); 436 #endif 437 ac_list = rf_find_raid_components(); 438 439 /* 2. Sort them into their respective sets. */ 440 config_sets = rf_create_auto_sets(ac_list); 441 442 /* 443 * 3. Evaluate each set andconfigure the valid ones. 444 * This gets done in rf_buildroothack(). 445 */ 446 rf_buildroothack(config_sets); 447 448 return (1); 449 } 450 451 void 452 rf_buildroothack(RF_ConfigSet_t *config_sets) 453 { 454 RF_ConfigSet_t *cset; 455 RF_ConfigSet_t *next_cset; 456 int retcode; 457 int raidID; 458 int rootID; 459 int num_root; 460 461 rootID = 0; 462 num_root = 0; 463 cset = config_sets; 464 while(cset != NULL ) { 465 next_cset = cset->next; 466 if (rf_have_enough_components(cset) && 467 cset->ac->clabel->autoconfigure==1) { 468 retcode = rf_auto_config_set(cset,&raidID); 469 if (!retcode) { 470 if (cset->rootable) { 471 rootID = raidID; 472 num_root++; 473 } 474 } else { 475 /* The autoconfig didn't work :( */ 476 #if DEBUG 477 printf("Autoconfig failed with code %d for raid%d\n", retcode, raidID); 478 #endif 479 rf_release_all_vps(cset); 480 } 481 } else { 482 /* we're not autoconfiguring this set... 483 release the associated resources */ 484 rf_release_all_vps(cset); 485 } 486 /* cleanup */ 487 rf_cleanup_config_set(cset); 488 cset = next_cset; 489 } 490 491 /* we found something bootable... */ 492 493 if (num_root == 1) { 494 booted_device = &raidrootdev[rootID]; 495 } else if (num_root > 1) { 496 /* we can't guess.. require the user to answer... */ 497 boothowto |= RB_ASKNAME; 498 } 499 } 500 501 502 int 503 raidsize(dev_t dev) 504 { 505 struct raid_softc *rs; 506 struct disklabel *lp; 507 int part, unit, omask, size; 508 509 unit = raidunit(dev); 510 if (unit >= numraid) 511 return (-1); 512 rs = &raid_softc[unit]; 513 514 if ((rs->sc_flags & RAIDF_INITED) == 0) 515 return (-1); 516 517 part = DISKPART(dev); 518 omask = rs->sc_dkdev.dk_openmask & (1 << part); 519 lp = rs->sc_dkdev.dk_label; 520 521 if (omask == 0 && raidopen(dev, 0, S_IFBLK, curlwp)) 522 return (-1); 523 524 if (lp->d_partitions[part].p_fstype != FS_SWAP) 525 size = -1; 526 else 527 size = lp->d_partitions[part].p_size * 528 (lp->d_secsize / DEV_BSIZE); 529 530 if (omask == 0 && raidclose(dev, 0, S_IFBLK, curlwp)) 531 return (-1); 532 533 return (size); 534 535 } 536 537 int 538 raiddump(dev_t dev, daddr_t blkno, caddr_t va, size_t size) 539 { 540 /* Not implemented. */ 541 return ENXIO; 542 } 543 /* ARGSUSED */ 544 int 545 raidopen(dev_t dev, int flags, int fmt, struct lwp *l) 546 { 547 int unit = raidunit(dev); 548 struct raid_softc *rs; 549 struct disklabel *lp; 550 int part, pmask; 551 int error = 0; 552 553 if (unit >= numraid) 554 return (ENXIO); 555 rs = &raid_softc[unit]; 556 557 if ((error = raidlock(rs)) != 0) 558 return (error); 559 lp = rs->sc_dkdev.dk_label; 560 561 part = DISKPART(dev); 562 pmask = (1 << part); 563 564 if ((rs->sc_flags & RAIDF_INITED) && 565 (rs->sc_dkdev.dk_openmask == 0)) 566 raidgetdisklabel(dev); 567 568 /* make sure that this partition exists */ 569 570 if (part != RAW_PART) { 571 if (((rs->sc_flags & RAIDF_INITED) == 0) || 572 ((part >= lp->d_npartitions) || 573 (lp->d_partitions[part].p_fstype == FS_UNUSED))) { 574 error = ENXIO; 575 raidunlock(rs); 576 return (error); 577 } 578 } 579 /* Prevent this unit from being unconfigured while open. */ 580 switch (fmt) { 581 case S_IFCHR: 582 rs->sc_dkdev.dk_copenmask |= pmask; 583 break; 584 585 case S_IFBLK: 586 rs->sc_dkdev.dk_bopenmask |= pmask; 587 break; 588 } 589 590 if ((rs->sc_dkdev.dk_openmask == 0) && 591 ((rs->sc_flags & RAIDF_INITED) != 0)) { 592 /* First one... mark things as dirty... Note that we *MUST* 593 have done a configure before this. I DO NOT WANT TO BE 594 SCRIBBLING TO RANDOM COMPONENTS UNTIL IT'S BEEN DETERMINED 595 THAT THEY BELONG TOGETHER!!!!! */ 596 /* XXX should check to see if we're only open for reading 597 here... If so, we needn't do this, but then need some 598 other way of keeping track of what's happened.. */ 599 600 rf_markalldirty( raidPtrs[unit] ); 601 } 602 603 604 rs->sc_dkdev.dk_openmask = 605 rs->sc_dkdev.dk_copenmask | rs->sc_dkdev.dk_bopenmask; 606 607 raidunlock(rs); 608 609 return (error); 610 611 612 } 613 /* ARGSUSED */ 614 int 615 raidclose(dev_t dev, int flags, int fmt, struct lwp *l) 616 { 617 int unit = raidunit(dev); 618 struct raid_softc *rs; 619 int error = 0; 620 int part; 621 622 if (unit >= numraid) 623 return (ENXIO); 624 rs = &raid_softc[unit]; 625 626 if ((error = raidlock(rs)) != 0) 627 return (error); 628 629 part = DISKPART(dev); 630 631 /* ...that much closer to allowing unconfiguration... */ 632 switch (fmt) { 633 case S_IFCHR: 634 rs->sc_dkdev.dk_copenmask &= ~(1 << part); 635 break; 636 637 case S_IFBLK: 638 rs->sc_dkdev.dk_bopenmask &= ~(1 << part); 639 break; 640 } 641 rs->sc_dkdev.dk_openmask = 642 rs->sc_dkdev.dk_copenmask | rs->sc_dkdev.dk_bopenmask; 643 644 if ((rs->sc_dkdev.dk_openmask == 0) && 645 ((rs->sc_flags & RAIDF_INITED) != 0)) { 646 /* Last one... device is not unconfigured yet. 647 Device shutdown has taken care of setting the 648 clean bits if RAIDF_INITED is not set 649 mark things as clean... */ 650 651 rf_update_component_labels(raidPtrs[unit], 652 RF_FINAL_COMPONENT_UPDATE); 653 if (doing_shutdown) { 654 /* last one, and we're going down, so 655 lights out for this RAID set too. */ 656 error = rf_Shutdown(raidPtrs[unit]); 657 658 /* It's no longer initialized... */ 659 rs->sc_flags &= ~RAIDF_INITED; 660 661 /* Detach the disk. */ 662 pseudo_disk_detach(&rs->sc_dkdev); 663 } 664 } 665 666 raidunlock(rs); 667 return (0); 668 669 } 670 671 void 672 raidstrategy(struct buf *bp) 673 { 674 int s; 675 676 unsigned int raidID = raidunit(bp->b_dev); 677 RF_Raid_t *raidPtr; 678 struct raid_softc *rs = &raid_softc[raidID]; 679 int wlabel; 680 681 if ((rs->sc_flags & RAIDF_INITED) ==0) { 682 bp->b_error = ENXIO; 683 bp->b_flags |= B_ERROR; 684 goto done; 685 } 686 if (raidID >= numraid || !raidPtrs[raidID]) { 687 bp->b_error = ENODEV; 688 bp->b_flags |= B_ERROR; 689 goto done; 690 } 691 raidPtr = raidPtrs[raidID]; 692 if (!raidPtr->valid) { 693 bp->b_error = ENODEV; 694 bp->b_flags |= B_ERROR; 695 goto done; 696 } 697 if (bp->b_bcount == 0) { 698 db1_printf(("b_bcount is zero..\n")); 699 goto done; 700 } 701 702 /* 703 * Do bounds checking and adjust transfer. If there's an 704 * error, the bounds check will flag that for us. 705 */ 706 707 wlabel = rs->sc_flags & (RAIDF_WLABEL | RAIDF_LABELLING); 708 if (DISKPART(bp->b_dev) == RAW_PART) { 709 uint64_t size; /* device size in DEV_BSIZE unit */ 710 711 if (raidPtr->logBytesPerSector > DEV_BSHIFT) { 712 size = raidPtr->totalSectors << 713 (raidPtr->logBytesPerSector - DEV_BSHIFT); 714 } else { 715 size = raidPtr->totalSectors >> 716 (DEV_BSHIFT - raidPtr->logBytesPerSector); 717 } 718 if (bounds_check_with_mediasize(bp, DEV_BSIZE, size) <= 0) { 719 goto done; 720 } 721 } else { 722 if (bounds_check_with_label(&rs->sc_dkdev, bp, wlabel) <= 0) { 723 db1_printf(("Bounds check failed!!:%d %d\n", 724 (int) bp->b_blkno, (int) wlabel)); 725 goto done; 726 } 727 } 728 s = splbio(); 729 730 bp->b_resid = 0; 731 732 /* stuff it onto our queue */ 733 BUFQ_PUT(rs->buf_queue, bp); 734 735 /* scheduled the IO to happen at the next convenient time */ 736 wakeup(&(raidPtrs[raidID]->iodone)); 737 738 splx(s); 739 return; 740 741 done: 742 bp->b_resid = bp->b_bcount; 743 biodone(bp); 744 } 745 /* ARGSUSED */ 746 int 747 raidread(dev_t dev, struct uio *uio, int flags) 748 { 749 int unit = raidunit(dev); 750 struct raid_softc *rs; 751 752 if (unit >= numraid) 753 return (ENXIO); 754 rs = &raid_softc[unit]; 755 756 if ((rs->sc_flags & RAIDF_INITED) == 0) 757 return (ENXIO); 758 759 return (physio(raidstrategy, NULL, dev, B_READ, minphys, uio)); 760 761 } 762 /* ARGSUSED */ 763 int 764 raidwrite(dev_t dev, struct uio *uio, int flags) 765 { 766 int unit = raidunit(dev); 767 struct raid_softc *rs; 768 769 if (unit >= numraid) 770 return (ENXIO); 771 rs = &raid_softc[unit]; 772 773 if ((rs->sc_flags & RAIDF_INITED) == 0) 774 return (ENXIO); 775 776 return (physio(raidstrategy, NULL, dev, B_WRITE, minphys, uio)); 777 778 } 779 780 int 781 raidioctl(dev_t dev, u_long cmd, caddr_t data, int flag, struct lwp *l) 782 { 783 int unit = raidunit(dev); 784 int error = 0; 785 int part, pmask; 786 struct raid_softc *rs; 787 RF_Config_t *k_cfg, *u_cfg; 788 RF_Raid_t *raidPtr; 789 RF_RaidDisk_t *diskPtr; 790 RF_AccTotals_t *totals; 791 RF_DeviceConfig_t *d_cfg, **ucfgp; 792 u_char *specific_buf; 793 int retcode = 0; 794 int column; 795 int raidid; 796 struct rf_recon_req *rrcopy, *rr; 797 RF_ComponentLabel_t *clabel; 798 RF_ComponentLabel_t ci_label; 799 RF_ComponentLabel_t **clabel_ptr; 800 RF_SingleComponent_t *sparePtr,*componentPtr; 801 RF_SingleComponent_t hot_spare; 802 RF_SingleComponent_t component; 803 RF_ProgressInfo_t progressInfo, **progressInfoPtr; 804 int i, j, d; 805 #ifdef __HAVE_OLD_DISKLABEL 806 struct disklabel newlabel; 807 #endif 808 809 if (unit >= numraid) 810 return (ENXIO); 811 rs = &raid_softc[unit]; 812 raidPtr = raidPtrs[unit]; 813 814 db1_printf(("raidioctl: %d %d %d %d\n", (int) dev, 815 (int) DISKPART(dev), (int) unit, (int) cmd)); 816 817 /* Must be open for writes for these commands... */ 818 switch (cmd) { 819 case DIOCSDINFO: 820 case DIOCWDINFO: 821 #ifdef __HAVE_OLD_DISKLABEL 822 case ODIOCWDINFO: 823 case ODIOCSDINFO: 824 #endif 825 case DIOCWLABEL: 826 if ((flag & FWRITE) == 0) 827 return (EBADF); 828 } 829 830 /* Must be initialized for these... */ 831 switch (cmd) { 832 case DIOCGDINFO: 833 case DIOCSDINFO: 834 case DIOCWDINFO: 835 #ifdef __HAVE_OLD_DISKLABEL 836 case ODIOCGDINFO: 837 case ODIOCWDINFO: 838 case ODIOCSDINFO: 839 case ODIOCGDEFLABEL: 840 #endif 841 case DIOCGPART: 842 case DIOCWLABEL: 843 case DIOCGDEFLABEL: 844 case RAIDFRAME_SHUTDOWN: 845 case RAIDFRAME_REWRITEPARITY: 846 case RAIDFRAME_GET_INFO: 847 case RAIDFRAME_RESET_ACCTOTALS: 848 case RAIDFRAME_GET_ACCTOTALS: 849 case RAIDFRAME_KEEP_ACCTOTALS: 850 case RAIDFRAME_GET_SIZE: 851 case RAIDFRAME_FAIL_DISK: 852 case RAIDFRAME_COPYBACK: 853 case RAIDFRAME_CHECK_RECON_STATUS: 854 case RAIDFRAME_CHECK_RECON_STATUS_EXT: 855 case RAIDFRAME_GET_COMPONENT_LABEL: 856 case RAIDFRAME_SET_COMPONENT_LABEL: 857 case RAIDFRAME_ADD_HOT_SPARE: 858 case RAIDFRAME_REMOVE_HOT_SPARE: 859 case RAIDFRAME_INIT_LABELS: 860 case RAIDFRAME_REBUILD_IN_PLACE: 861 case RAIDFRAME_CHECK_PARITY: 862 case RAIDFRAME_CHECK_PARITYREWRITE_STATUS: 863 case RAIDFRAME_CHECK_PARITYREWRITE_STATUS_EXT: 864 case RAIDFRAME_CHECK_COPYBACK_STATUS: 865 case RAIDFRAME_CHECK_COPYBACK_STATUS_EXT: 866 case RAIDFRAME_SET_AUTOCONFIG: 867 case RAIDFRAME_SET_ROOT: 868 case RAIDFRAME_DELETE_COMPONENT: 869 case RAIDFRAME_INCORPORATE_HOT_SPARE: 870 if ((rs->sc_flags & RAIDF_INITED) == 0) 871 return (ENXIO); 872 } 873 874 switch (cmd) { 875 876 /* configure the system */ 877 case RAIDFRAME_CONFIGURE: 878 879 if (raidPtr->valid) { 880 /* There is a valid RAID set running on this unit! */ 881 printf("raid%d: Device already configured!\n",unit); 882 return(EINVAL); 883 } 884 885 /* copy-in the configuration information */ 886 /* data points to a pointer to the configuration structure */ 887 888 u_cfg = *((RF_Config_t **) data); 889 RF_Malloc(k_cfg, sizeof(RF_Config_t), (RF_Config_t *)); 890 if (k_cfg == NULL) { 891 return (ENOMEM); 892 } 893 retcode = copyin(u_cfg, k_cfg, sizeof(RF_Config_t)); 894 if (retcode) { 895 RF_Free(k_cfg, sizeof(RF_Config_t)); 896 db1_printf(("rf_ioctl: retcode=%d copyin.1\n", 897 retcode)); 898 return (retcode); 899 } 900 /* allocate a buffer for the layout-specific data, and copy it 901 * in */ 902 if (k_cfg->layoutSpecificSize) { 903 if (k_cfg->layoutSpecificSize > 10000) { 904 /* sanity check */ 905 RF_Free(k_cfg, sizeof(RF_Config_t)); 906 return (EINVAL); 907 } 908 RF_Malloc(specific_buf, k_cfg->layoutSpecificSize, 909 (u_char *)); 910 if (specific_buf == NULL) { 911 RF_Free(k_cfg, sizeof(RF_Config_t)); 912 return (ENOMEM); 913 } 914 retcode = copyin(k_cfg->layoutSpecific, specific_buf, 915 k_cfg->layoutSpecificSize); 916 if (retcode) { 917 RF_Free(k_cfg, sizeof(RF_Config_t)); 918 RF_Free(specific_buf, 919 k_cfg->layoutSpecificSize); 920 db1_printf(("rf_ioctl: retcode=%d copyin.2\n", 921 retcode)); 922 return (retcode); 923 } 924 } else 925 specific_buf = NULL; 926 k_cfg->layoutSpecific = specific_buf; 927 928 /* should do some kind of sanity check on the configuration. 929 * Store the sum of all the bytes in the last byte? */ 930 931 /* configure the system */ 932 933 /* 934 * Clear the entire RAID descriptor, just to make sure 935 * there is no stale data left in the case of a 936 * reconfiguration 937 */ 938 memset((char *) raidPtr, 0, sizeof(RF_Raid_t)); 939 raidPtr->raidid = unit; 940 941 retcode = rf_Configure(raidPtr, k_cfg, NULL); 942 943 if (retcode == 0) { 944 945 /* allow this many simultaneous IO's to 946 this RAID device */ 947 raidPtr->openings = RAIDOUTSTANDING; 948 949 raidinit(raidPtr); 950 rf_markalldirty(raidPtr); 951 } 952 /* free the buffers. No return code here. */ 953 if (k_cfg->layoutSpecificSize) { 954 RF_Free(specific_buf, k_cfg->layoutSpecificSize); 955 } 956 RF_Free(k_cfg, sizeof(RF_Config_t)); 957 958 return (retcode); 959 960 /* shutdown the system */ 961 case RAIDFRAME_SHUTDOWN: 962 963 if ((error = raidlock(rs)) != 0) 964 return (error); 965 966 /* 967 * If somebody has a partition mounted, we shouldn't 968 * shutdown. 969 */ 970 971 part = DISKPART(dev); 972 pmask = (1 << part); 973 if ((rs->sc_dkdev.dk_openmask & ~pmask) || 974 ((rs->sc_dkdev.dk_bopenmask & pmask) && 975 (rs->sc_dkdev.dk_copenmask & pmask))) { 976 raidunlock(rs); 977 return (EBUSY); 978 } 979 980 retcode = rf_Shutdown(raidPtr); 981 982 /* It's no longer initialized... */ 983 rs->sc_flags &= ~RAIDF_INITED; 984 985 /* Detach the disk. */ 986 pseudo_disk_detach(&rs->sc_dkdev); 987 988 raidunlock(rs); 989 990 return (retcode); 991 case RAIDFRAME_GET_COMPONENT_LABEL: 992 clabel_ptr = (RF_ComponentLabel_t **) data; 993 /* need to read the component label for the disk indicated 994 by row,column in clabel */ 995 996 /* For practice, let's get it directly fromdisk, rather 997 than from the in-core copy */ 998 RF_Malloc( clabel, sizeof( RF_ComponentLabel_t ), 999 (RF_ComponentLabel_t *)); 1000 if (clabel == NULL) 1001 return (ENOMEM); 1002 1003 retcode = copyin( *clabel_ptr, clabel, 1004 sizeof(RF_ComponentLabel_t)); 1005 1006 if (retcode) { 1007 RF_Free( clabel, sizeof(RF_ComponentLabel_t)); 1008 return(retcode); 1009 } 1010 1011 clabel->row = 0; /* Don't allow looking at anything else.*/ 1012 1013 column = clabel->column; 1014 1015 if ((column < 0) || (column >= raidPtr->numCol + 1016 raidPtr->numSpare)) { 1017 RF_Free( clabel, sizeof(RF_ComponentLabel_t)); 1018 return(EINVAL); 1019 } 1020 1021 retcode = raidread_component_label(raidPtr->Disks[column].dev, 1022 raidPtr->raid_cinfo[column].ci_vp, 1023 clabel ); 1024 1025 if (retcode == 0) { 1026 retcode = copyout(clabel, *clabel_ptr, 1027 sizeof(RF_ComponentLabel_t)); 1028 } 1029 RF_Free(clabel, sizeof(RF_ComponentLabel_t)); 1030 return (retcode); 1031 1032 case RAIDFRAME_SET_COMPONENT_LABEL: 1033 clabel = (RF_ComponentLabel_t *) data; 1034 1035 /* XXX check the label for valid stuff... */ 1036 /* Note that some things *should not* get modified -- 1037 the user should be re-initing the labels instead of 1038 trying to patch things. 1039 */ 1040 1041 raidid = raidPtr->raidid; 1042 #if DEBUG 1043 printf("raid%d: Got component label:\n", raidid); 1044 printf("raid%d: Version: %d\n", raidid, clabel->version); 1045 printf("raid%d: Serial Number: %d\n", raidid, clabel->serial_number); 1046 printf("raid%d: Mod counter: %d\n", raidid, clabel->mod_counter); 1047 printf("raid%d: Column: %d\n", raidid, clabel->column); 1048 printf("raid%d: Num Columns: %d\n", raidid, clabel->num_columns); 1049 printf("raid%d: Clean: %d\n", raidid, clabel->clean); 1050 printf("raid%d: Status: %d\n", raidid, clabel->status); 1051 #endif 1052 clabel->row = 0; 1053 column = clabel->column; 1054 1055 if ((column < 0) || (column >= raidPtr->numCol)) { 1056 return(EINVAL); 1057 } 1058 1059 /* XXX this isn't allowed to do anything for now :-) */ 1060 1061 /* XXX and before it is, we need to fill in the rest 1062 of the fields!?!?!?! */ 1063 #if 0 1064 raidwrite_component_label( 1065 raidPtr->Disks[column].dev, 1066 raidPtr->raid_cinfo[column].ci_vp, 1067 clabel ); 1068 #endif 1069 return (0); 1070 1071 case RAIDFRAME_INIT_LABELS: 1072 clabel = (RF_ComponentLabel_t *) data; 1073 /* 1074 we only want the serial number from 1075 the above. We get all the rest of the information 1076 from the config that was used to create this RAID 1077 set. 1078 */ 1079 1080 raidPtr->serial_number = clabel->serial_number; 1081 1082 raid_init_component_label(raidPtr, &ci_label); 1083 ci_label.serial_number = clabel->serial_number; 1084 ci_label.row = 0; /* we dont' pretend to support more */ 1085 1086 for(column=0;column<raidPtr->numCol;column++) { 1087 diskPtr = &raidPtr->Disks[column]; 1088 if (!RF_DEAD_DISK(diskPtr->status)) { 1089 ci_label.partitionSize = diskPtr->partitionSize; 1090 ci_label.column = column; 1091 raidwrite_component_label( 1092 raidPtr->Disks[column].dev, 1093 raidPtr->raid_cinfo[column].ci_vp, 1094 &ci_label ); 1095 } 1096 } 1097 1098 return (retcode); 1099 case RAIDFRAME_SET_AUTOCONFIG: 1100 d = rf_set_autoconfig(raidPtr, *(int *) data); 1101 printf("raid%d: New autoconfig value is: %d\n", 1102 raidPtr->raidid, d); 1103 *(int *) data = d; 1104 return (retcode); 1105 1106 case RAIDFRAME_SET_ROOT: 1107 d = rf_set_rootpartition(raidPtr, *(int *) data); 1108 printf("raid%d: New rootpartition value is: %d\n", 1109 raidPtr->raidid, d); 1110 *(int *) data = d; 1111 return (retcode); 1112 1113 /* initialize all parity */ 1114 case RAIDFRAME_REWRITEPARITY: 1115 1116 if (raidPtr->Layout.map->faultsTolerated == 0) { 1117 /* Parity for RAID 0 is trivially correct */ 1118 raidPtr->parity_good = RF_RAID_CLEAN; 1119 return(0); 1120 } 1121 1122 if (raidPtr->parity_rewrite_in_progress == 1) { 1123 /* Re-write is already in progress! */ 1124 return(EINVAL); 1125 } 1126 1127 retcode = RF_CREATE_THREAD(raidPtr->parity_rewrite_thread, 1128 rf_RewriteParityThread, 1129 raidPtr,"raid_parity"); 1130 return (retcode); 1131 1132 1133 case RAIDFRAME_ADD_HOT_SPARE: 1134 sparePtr = (RF_SingleComponent_t *) data; 1135 memcpy( &hot_spare, sparePtr, sizeof(RF_SingleComponent_t)); 1136 retcode = rf_add_hot_spare(raidPtr, &hot_spare); 1137 return(retcode); 1138 1139 case RAIDFRAME_REMOVE_HOT_SPARE: 1140 return(retcode); 1141 1142 case RAIDFRAME_DELETE_COMPONENT: 1143 componentPtr = (RF_SingleComponent_t *)data; 1144 memcpy( &component, componentPtr, 1145 sizeof(RF_SingleComponent_t)); 1146 retcode = rf_delete_component(raidPtr, &component); 1147 return(retcode); 1148 1149 case RAIDFRAME_INCORPORATE_HOT_SPARE: 1150 componentPtr = (RF_SingleComponent_t *)data; 1151 memcpy( &component, componentPtr, 1152 sizeof(RF_SingleComponent_t)); 1153 retcode = rf_incorporate_hot_spare(raidPtr, &component); 1154 return(retcode); 1155 1156 case RAIDFRAME_REBUILD_IN_PLACE: 1157 1158 if (raidPtr->Layout.map->faultsTolerated == 0) { 1159 /* Can't do this on a RAID 0!! */ 1160 return(EINVAL); 1161 } 1162 1163 if (raidPtr->recon_in_progress == 1) { 1164 /* a reconstruct is already in progress! */ 1165 return(EINVAL); 1166 } 1167 1168 componentPtr = (RF_SingleComponent_t *) data; 1169 memcpy( &component, componentPtr, 1170 sizeof(RF_SingleComponent_t)); 1171 component.row = 0; /* we don't support any more */ 1172 column = component.column; 1173 1174 if ((column < 0) || (column >= raidPtr->numCol)) { 1175 return(EINVAL); 1176 } 1177 1178 RF_LOCK_MUTEX(raidPtr->mutex); 1179 if ((raidPtr->Disks[column].status == rf_ds_optimal) && 1180 (raidPtr->numFailures > 0)) { 1181 /* XXX 0 above shouldn't be constant!!! */ 1182 /* some component other than this has failed. 1183 Let's not make things worse than they already 1184 are... */ 1185 printf("raid%d: Unable to reconstruct to disk at:\n", 1186 raidPtr->raidid); 1187 printf("raid%d: Col: %d Too many failures.\n", 1188 raidPtr->raidid, column); 1189 RF_UNLOCK_MUTEX(raidPtr->mutex); 1190 return (EINVAL); 1191 } 1192 if (raidPtr->Disks[column].status == 1193 rf_ds_reconstructing) { 1194 printf("raid%d: Unable to reconstruct to disk at:\n", 1195 raidPtr->raidid); 1196 printf("raid%d: Col: %d Reconstruction already occuring!\n", raidPtr->raidid, column); 1197 1198 RF_UNLOCK_MUTEX(raidPtr->mutex); 1199 return (EINVAL); 1200 } 1201 if (raidPtr->Disks[column].status == rf_ds_spared) { 1202 RF_UNLOCK_MUTEX(raidPtr->mutex); 1203 return (EINVAL); 1204 } 1205 RF_UNLOCK_MUTEX(raidPtr->mutex); 1206 1207 RF_Malloc(rrcopy, sizeof(*rrcopy), (struct rf_recon_req *)); 1208 if (rrcopy == NULL) 1209 return(ENOMEM); 1210 1211 rrcopy->raidPtr = (void *) raidPtr; 1212 rrcopy->col = column; 1213 1214 retcode = RF_CREATE_THREAD(raidPtr->recon_thread, 1215 rf_ReconstructInPlaceThread, 1216 rrcopy,"raid_reconip"); 1217 return(retcode); 1218 1219 case RAIDFRAME_GET_INFO: 1220 if (!raidPtr->valid) 1221 return (ENODEV); 1222 ucfgp = (RF_DeviceConfig_t **) data; 1223 RF_Malloc(d_cfg, sizeof(RF_DeviceConfig_t), 1224 (RF_DeviceConfig_t *)); 1225 if (d_cfg == NULL) 1226 return (ENOMEM); 1227 d_cfg->rows = 1; /* there is only 1 row now */ 1228 d_cfg->cols = raidPtr->numCol; 1229 d_cfg->ndevs = raidPtr->numCol; 1230 if (d_cfg->ndevs >= RF_MAX_DISKS) { 1231 RF_Free(d_cfg, sizeof(RF_DeviceConfig_t)); 1232 return (ENOMEM); 1233 } 1234 d_cfg->nspares = raidPtr->numSpare; 1235 if (d_cfg->nspares >= RF_MAX_DISKS) { 1236 RF_Free(d_cfg, sizeof(RF_DeviceConfig_t)); 1237 return (ENOMEM); 1238 } 1239 d_cfg->maxqdepth = raidPtr->maxQueueDepth; 1240 d = 0; 1241 for (j = 0; j < d_cfg->cols; j++) { 1242 d_cfg->devs[d] = raidPtr->Disks[j]; 1243 d++; 1244 } 1245 for (j = d_cfg->cols, i = 0; i < d_cfg->nspares; i++, j++) { 1246 d_cfg->spares[i] = raidPtr->Disks[j]; 1247 } 1248 retcode = copyout(d_cfg, *ucfgp, sizeof(RF_DeviceConfig_t)); 1249 RF_Free(d_cfg, sizeof(RF_DeviceConfig_t)); 1250 1251 return (retcode); 1252 1253 case RAIDFRAME_CHECK_PARITY: 1254 *(int *) data = raidPtr->parity_good; 1255 return (0); 1256 1257 case RAIDFRAME_RESET_ACCTOTALS: 1258 memset(&raidPtr->acc_totals, 0, sizeof(raidPtr->acc_totals)); 1259 return (0); 1260 1261 case RAIDFRAME_GET_ACCTOTALS: 1262 totals = (RF_AccTotals_t *) data; 1263 *totals = raidPtr->acc_totals; 1264 return (0); 1265 1266 case RAIDFRAME_KEEP_ACCTOTALS: 1267 raidPtr->keep_acc_totals = *(int *)data; 1268 return (0); 1269 1270 case RAIDFRAME_GET_SIZE: 1271 *(int *) data = raidPtr->totalSectors; 1272 return (0); 1273 1274 /* fail a disk & optionally start reconstruction */ 1275 case RAIDFRAME_FAIL_DISK: 1276 1277 if (raidPtr->Layout.map->faultsTolerated == 0) { 1278 /* Can't do this on a RAID 0!! */ 1279 return(EINVAL); 1280 } 1281 1282 rr = (struct rf_recon_req *) data; 1283 rr->row = 0; 1284 if (rr->col < 0 || rr->col >= raidPtr->numCol) 1285 return (EINVAL); 1286 1287 1288 RF_LOCK_MUTEX(raidPtr->mutex); 1289 if (raidPtr->status == rf_rs_reconstructing) { 1290 /* you can't fail a disk while we're reconstructing! */ 1291 /* XXX wrong for RAID6 */ 1292 RF_UNLOCK_MUTEX(raidPtr->mutex); 1293 return (EINVAL); 1294 } 1295 if ((raidPtr->Disks[rr->col].status == 1296 rf_ds_optimal) && (raidPtr->numFailures > 0)) { 1297 /* some other component has failed. Let's not make 1298 things worse. XXX wrong for RAID6 */ 1299 RF_UNLOCK_MUTEX(raidPtr->mutex); 1300 return (EINVAL); 1301 } 1302 if (raidPtr->Disks[rr->col].status == rf_ds_spared) { 1303 /* Can't fail a spared disk! */ 1304 RF_UNLOCK_MUTEX(raidPtr->mutex); 1305 return (EINVAL); 1306 } 1307 RF_UNLOCK_MUTEX(raidPtr->mutex); 1308 1309 /* make a copy of the recon request so that we don't rely on 1310 * the user's buffer */ 1311 RF_Malloc(rrcopy, sizeof(*rrcopy), (struct rf_recon_req *)); 1312 if (rrcopy == NULL) 1313 return(ENOMEM); 1314 memcpy(rrcopy, rr, sizeof(*rr)); 1315 rrcopy->raidPtr = (void *) raidPtr; 1316 1317 retcode = RF_CREATE_THREAD(raidPtr->recon_thread, 1318 rf_ReconThread, 1319 rrcopy,"raid_recon"); 1320 return (0); 1321 1322 /* invoke a copyback operation after recon on whatever disk 1323 * needs it, if any */ 1324 case RAIDFRAME_COPYBACK: 1325 1326 if (raidPtr->Layout.map->faultsTolerated == 0) { 1327 /* This makes no sense on a RAID 0!! */ 1328 return(EINVAL); 1329 } 1330 1331 if (raidPtr->copyback_in_progress == 1) { 1332 /* Copyback is already in progress! */ 1333 return(EINVAL); 1334 } 1335 1336 retcode = RF_CREATE_THREAD(raidPtr->copyback_thread, 1337 rf_CopybackThread, 1338 raidPtr,"raid_copyback"); 1339 return (retcode); 1340 1341 /* return the percentage completion of reconstruction */ 1342 case RAIDFRAME_CHECK_RECON_STATUS: 1343 if (raidPtr->Layout.map->faultsTolerated == 0) { 1344 /* This makes no sense on a RAID 0, so tell the 1345 user it's done. */ 1346 *(int *) data = 100; 1347 return(0); 1348 } 1349 if (raidPtr->status != rf_rs_reconstructing) 1350 *(int *) data = 100; 1351 else { 1352 if (raidPtr->reconControl->numRUsTotal > 0) { 1353 *(int *) data = (raidPtr->reconControl->numRUsComplete * 100 / raidPtr->reconControl->numRUsTotal); 1354 } else { 1355 *(int *) data = 0; 1356 } 1357 } 1358 return (0); 1359 case RAIDFRAME_CHECK_RECON_STATUS_EXT: 1360 progressInfoPtr = (RF_ProgressInfo_t **) data; 1361 if (raidPtr->status != rf_rs_reconstructing) { 1362 progressInfo.remaining = 0; 1363 progressInfo.completed = 100; 1364 progressInfo.total = 100; 1365 } else { 1366 progressInfo.total = 1367 raidPtr->reconControl->numRUsTotal; 1368 progressInfo.completed = 1369 raidPtr->reconControl->numRUsComplete; 1370 progressInfo.remaining = progressInfo.total - 1371 progressInfo.completed; 1372 } 1373 retcode = copyout(&progressInfo, *progressInfoPtr, 1374 sizeof(RF_ProgressInfo_t)); 1375 return (retcode); 1376 1377 case RAIDFRAME_CHECK_PARITYREWRITE_STATUS: 1378 if (raidPtr->Layout.map->faultsTolerated == 0) { 1379 /* This makes no sense on a RAID 0, so tell the 1380 user it's done. */ 1381 *(int *) data = 100; 1382 return(0); 1383 } 1384 if (raidPtr->parity_rewrite_in_progress == 1) { 1385 *(int *) data = 100 * 1386 raidPtr->parity_rewrite_stripes_done / 1387 raidPtr->Layout.numStripe; 1388 } else { 1389 *(int *) data = 100; 1390 } 1391 return (0); 1392 1393 case RAIDFRAME_CHECK_PARITYREWRITE_STATUS_EXT: 1394 progressInfoPtr = (RF_ProgressInfo_t **) data; 1395 if (raidPtr->parity_rewrite_in_progress == 1) { 1396 progressInfo.total = raidPtr->Layout.numStripe; 1397 progressInfo.completed = 1398 raidPtr->parity_rewrite_stripes_done; 1399 progressInfo.remaining = progressInfo.total - 1400 progressInfo.completed; 1401 } else { 1402 progressInfo.remaining = 0; 1403 progressInfo.completed = 100; 1404 progressInfo.total = 100; 1405 } 1406 retcode = copyout(&progressInfo, *progressInfoPtr, 1407 sizeof(RF_ProgressInfo_t)); 1408 return (retcode); 1409 1410 case RAIDFRAME_CHECK_COPYBACK_STATUS: 1411 if (raidPtr->Layout.map->faultsTolerated == 0) { 1412 /* This makes no sense on a RAID 0 */ 1413 *(int *) data = 100; 1414 return(0); 1415 } 1416 if (raidPtr->copyback_in_progress == 1) { 1417 *(int *) data = 100 * raidPtr->copyback_stripes_done / 1418 raidPtr->Layout.numStripe; 1419 } else { 1420 *(int *) data = 100; 1421 } 1422 return (0); 1423 1424 case RAIDFRAME_CHECK_COPYBACK_STATUS_EXT: 1425 progressInfoPtr = (RF_ProgressInfo_t **) data; 1426 if (raidPtr->copyback_in_progress == 1) { 1427 progressInfo.total = raidPtr->Layout.numStripe; 1428 progressInfo.completed = 1429 raidPtr->copyback_stripes_done; 1430 progressInfo.remaining = progressInfo.total - 1431 progressInfo.completed; 1432 } else { 1433 progressInfo.remaining = 0; 1434 progressInfo.completed = 100; 1435 progressInfo.total = 100; 1436 } 1437 retcode = copyout(&progressInfo, *progressInfoPtr, 1438 sizeof(RF_ProgressInfo_t)); 1439 return (retcode); 1440 1441 /* the sparetable daemon calls this to wait for the kernel to 1442 * need a spare table. this ioctl does not return until a 1443 * spare table is needed. XXX -- calling mpsleep here in the 1444 * ioctl code is almost certainly wrong and evil. -- XXX XXX 1445 * -- I should either compute the spare table in the kernel, 1446 * or have a different -- XXX XXX -- interface (a different 1447 * character device) for delivering the table -- XXX */ 1448 #if 0 1449 case RAIDFRAME_SPARET_WAIT: 1450 RF_LOCK_MUTEX(rf_sparet_wait_mutex); 1451 while (!rf_sparet_wait_queue) 1452 mpsleep(&rf_sparet_wait_queue, (PZERO + 1) | PCATCH, "sparet wait", 0, (void *) simple_lock_addr(rf_sparet_wait_mutex), MS_LOCK_SIMPLE); 1453 waitreq = rf_sparet_wait_queue; 1454 rf_sparet_wait_queue = rf_sparet_wait_queue->next; 1455 RF_UNLOCK_MUTEX(rf_sparet_wait_mutex); 1456 1457 /* structure assignment */ 1458 *((RF_SparetWait_t *) data) = *waitreq; 1459 1460 RF_Free(waitreq, sizeof(*waitreq)); 1461 return (0); 1462 1463 /* wakes up a process waiting on SPARET_WAIT and puts an error 1464 * code in it that will cause the dameon to exit */ 1465 case RAIDFRAME_ABORT_SPARET_WAIT: 1466 RF_Malloc(waitreq, sizeof(*waitreq), (RF_SparetWait_t *)); 1467 waitreq->fcol = -1; 1468 RF_LOCK_MUTEX(rf_sparet_wait_mutex); 1469 waitreq->next = rf_sparet_wait_queue; 1470 rf_sparet_wait_queue = waitreq; 1471 RF_UNLOCK_MUTEX(rf_sparet_wait_mutex); 1472 wakeup(&rf_sparet_wait_queue); 1473 return (0); 1474 1475 /* used by the spare table daemon to deliver a spare table 1476 * into the kernel */ 1477 case RAIDFRAME_SEND_SPARET: 1478 1479 /* install the spare table */ 1480 retcode = rf_SetSpareTable(raidPtr, *(void **) data); 1481 1482 /* respond to the requestor. the return status of the spare 1483 * table installation is passed in the "fcol" field */ 1484 RF_Malloc(waitreq, sizeof(*waitreq), (RF_SparetWait_t *)); 1485 waitreq->fcol = retcode; 1486 RF_LOCK_MUTEX(rf_sparet_wait_mutex); 1487 waitreq->next = rf_sparet_resp_queue; 1488 rf_sparet_resp_queue = waitreq; 1489 wakeup(&rf_sparet_resp_queue); 1490 RF_UNLOCK_MUTEX(rf_sparet_wait_mutex); 1491 1492 return (retcode); 1493 #endif 1494 1495 default: 1496 break; /* fall through to the os-specific code below */ 1497 1498 } 1499 1500 if (!raidPtr->valid) 1501 return (EINVAL); 1502 1503 /* 1504 * Add support for "regular" device ioctls here. 1505 */ 1506 1507 switch (cmd) { 1508 case DIOCGDINFO: 1509 *(struct disklabel *) data = *(rs->sc_dkdev.dk_label); 1510 break; 1511 #ifdef __HAVE_OLD_DISKLABEL 1512 case ODIOCGDINFO: 1513 newlabel = *(rs->sc_dkdev.dk_label); 1514 if (newlabel.d_npartitions > OLDMAXPARTITIONS) 1515 return ENOTTY; 1516 memcpy(data, &newlabel, sizeof (struct olddisklabel)); 1517 break; 1518 #endif 1519 1520 case DIOCGPART: 1521 ((struct partinfo *) data)->disklab = rs->sc_dkdev.dk_label; 1522 ((struct partinfo *) data)->part = 1523 &rs->sc_dkdev.dk_label->d_partitions[DISKPART(dev)]; 1524 break; 1525 1526 case DIOCWDINFO: 1527 case DIOCSDINFO: 1528 #ifdef __HAVE_OLD_DISKLABEL 1529 case ODIOCWDINFO: 1530 case ODIOCSDINFO: 1531 #endif 1532 { 1533 struct disklabel *lp; 1534 #ifdef __HAVE_OLD_DISKLABEL 1535 if (cmd == ODIOCSDINFO || cmd == ODIOCWDINFO) { 1536 memset(&newlabel, 0, sizeof newlabel); 1537 memcpy(&newlabel, data, sizeof (struct olddisklabel)); 1538 lp = &newlabel; 1539 } else 1540 #endif 1541 lp = (struct disklabel *)data; 1542 1543 if ((error = raidlock(rs)) != 0) 1544 return (error); 1545 1546 rs->sc_flags |= RAIDF_LABELLING; 1547 1548 error = setdisklabel(rs->sc_dkdev.dk_label, 1549 lp, 0, rs->sc_dkdev.dk_cpulabel); 1550 if (error == 0) { 1551 if (cmd == DIOCWDINFO 1552 #ifdef __HAVE_OLD_DISKLABEL 1553 || cmd == ODIOCWDINFO 1554 #endif 1555 ) 1556 error = writedisklabel(RAIDLABELDEV(dev), 1557 raidstrategy, rs->sc_dkdev.dk_label, 1558 rs->sc_dkdev.dk_cpulabel); 1559 } 1560 rs->sc_flags &= ~RAIDF_LABELLING; 1561 1562 raidunlock(rs); 1563 1564 if (error) 1565 return (error); 1566 break; 1567 } 1568 1569 case DIOCWLABEL: 1570 if (*(int *) data != 0) 1571 rs->sc_flags |= RAIDF_WLABEL; 1572 else 1573 rs->sc_flags &= ~RAIDF_WLABEL; 1574 break; 1575 1576 case DIOCGDEFLABEL: 1577 raidgetdefaultlabel(raidPtr, rs, (struct disklabel *) data); 1578 break; 1579 1580 #ifdef __HAVE_OLD_DISKLABEL 1581 case ODIOCGDEFLABEL: 1582 raidgetdefaultlabel(raidPtr, rs, &newlabel); 1583 if (newlabel.d_npartitions > OLDMAXPARTITIONS) 1584 return ENOTTY; 1585 memcpy(data, &newlabel, sizeof (struct olddisklabel)); 1586 break; 1587 #endif 1588 1589 default: 1590 retcode = ENOTTY; 1591 } 1592 return (retcode); 1593 1594 } 1595 1596 1597 /* raidinit -- complete the rest of the initialization for the 1598 RAIDframe device. */ 1599 1600 1601 static void 1602 raidinit(RF_Raid_t *raidPtr) 1603 { 1604 struct raid_softc *rs; 1605 int unit; 1606 1607 unit = raidPtr->raidid; 1608 1609 rs = &raid_softc[unit]; 1610 1611 /* XXX should check return code first... */ 1612 rs->sc_flags |= RAIDF_INITED; 1613 1614 /* XXX doesn't check bounds. */ 1615 snprintf(rs->sc_xname, sizeof(rs->sc_xname), "raid%d", unit); 1616 1617 rs->sc_dkdev.dk_name = rs->sc_xname; 1618 1619 /* disk_attach actually creates space for the CPU disklabel, among 1620 * other things, so it's critical to call this *BEFORE* we try putzing 1621 * with disklabels. */ 1622 1623 pseudo_disk_attach(&rs->sc_dkdev); 1624 1625 /* XXX There may be a weird interaction here between this, and 1626 * protectedSectors, as used in RAIDframe. */ 1627 1628 rs->sc_size = raidPtr->totalSectors; 1629 } 1630 #if (RF_INCLUDE_PARITY_DECLUSTERING_DS > 0) 1631 /* wake up the daemon & tell it to get us a spare table 1632 * XXX 1633 * the entries in the queues should be tagged with the raidPtr 1634 * so that in the extremely rare case that two recons happen at once, 1635 * we know for which device were requesting a spare table 1636 * XXX 1637 * 1638 * XXX This code is not currently used. GO 1639 */ 1640 int 1641 rf_GetSpareTableFromDaemon(RF_SparetWait_t *req) 1642 { 1643 int retcode; 1644 1645 RF_LOCK_MUTEX(rf_sparet_wait_mutex); 1646 req->next = rf_sparet_wait_queue; 1647 rf_sparet_wait_queue = req; 1648 wakeup(&rf_sparet_wait_queue); 1649 1650 /* mpsleep unlocks the mutex */ 1651 while (!rf_sparet_resp_queue) { 1652 tsleep(&rf_sparet_resp_queue, PRIBIO, 1653 "raidframe getsparetable", 0); 1654 } 1655 req = rf_sparet_resp_queue; 1656 rf_sparet_resp_queue = req->next; 1657 RF_UNLOCK_MUTEX(rf_sparet_wait_mutex); 1658 1659 retcode = req->fcol; 1660 RF_Free(req, sizeof(*req)); /* this is not the same req as we 1661 * alloc'd */ 1662 return (retcode); 1663 } 1664 #endif 1665 1666 /* a wrapper around rf_DoAccess that extracts appropriate info from the 1667 * bp & passes it down. 1668 * any calls originating in the kernel must use non-blocking I/O 1669 * do some extra sanity checking to return "appropriate" error values for 1670 * certain conditions (to make some standard utilities work) 1671 * 1672 * Formerly known as: rf_DoAccessKernel 1673 */ 1674 void 1675 raidstart(RF_Raid_t *raidPtr) 1676 { 1677 RF_SectorCount_t num_blocks, pb, sum; 1678 RF_RaidAddr_t raid_addr; 1679 struct partition *pp; 1680 daddr_t blocknum; 1681 int unit; 1682 struct raid_softc *rs; 1683 int do_async; 1684 struct buf *bp; 1685 int rc; 1686 1687 unit = raidPtr->raidid; 1688 rs = &raid_softc[unit]; 1689 1690 /* quick check to see if anything has died recently */ 1691 RF_LOCK_MUTEX(raidPtr->mutex); 1692 if (raidPtr->numNewFailures > 0) { 1693 RF_UNLOCK_MUTEX(raidPtr->mutex); 1694 rf_update_component_labels(raidPtr, 1695 RF_NORMAL_COMPONENT_UPDATE); 1696 RF_LOCK_MUTEX(raidPtr->mutex); 1697 raidPtr->numNewFailures--; 1698 } 1699 1700 /* Check to see if we're at the limit... */ 1701 while (raidPtr->openings > 0) { 1702 RF_UNLOCK_MUTEX(raidPtr->mutex); 1703 1704 /* get the next item, if any, from the queue */ 1705 if ((bp = BUFQ_GET(rs->buf_queue)) == NULL) { 1706 /* nothing more to do */ 1707 return; 1708 } 1709 1710 /* Ok, for the bp we have here, bp->b_blkno is relative to the 1711 * partition.. Need to make it absolute to the underlying 1712 * device.. */ 1713 1714 blocknum = bp->b_blkno; 1715 if (DISKPART(bp->b_dev) != RAW_PART) { 1716 pp = &rs->sc_dkdev.dk_label->d_partitions[DISKPART(bp->b_dev)]; 1717 blocknum += pp->p_offset; 1718 } 1719 1720 db1_printf(("Blocks: %d, %d\n", (int) bp->b_blkno, 1721 (int) blocknum)); 1722 1723 db1_printf(("bp->b_bcount = %d\n", (int) bp->b_bcount)); 1724 db1_printf(("bp->b_resid = %d\n", (int) bp->b_resid)); 1725 1726 /* *THIS* is where we adjust what block we're going to... 1727 * but DO NOT TOUCH bp->b_blkno!!! */ 1728 raid_addr = blocknum; 1729 1730 num_blocks = bp->b_bcount >> raidPtr->logBytesPerSector; 1731 pb = (bp->b_bcount & raidPtr->sectorMask) ? 1 : 0; 1732 sum = raid_addr + num_blocks + pb; 1733 if (1 || rf_debugKernelAccess) { 1734 db1_printf(("raid_addr=%d sum=%d num_blocks=%d(+%d) (%d)\n", 1735 (int) raid_addr, (int) sum, (int) num_blocks, 1736 (int) pb, (int) bp->b_resid)); 1737 } 1738 if ((sum > raidPtr->totalSectors) || (sum < raid_addr) 1739 || (sum < num_blocks) || (sum < pb)) { 1740 bp->b_error = ENOSPC; 1741 bp->b_flags |= B_ERROR; 1742 bp->b_resid = bp->b_bcount; 1743 biodone(bp); 1744 RF_LOCK_MUTEX(raidPtr->mutex); 1745 continue; 1746 } 1747 /* 1748 * XXX rf_DoAccess() should do this, not just DoAccessKernel() 1749 */ 1750 1751 if (bp->b_bcount & raidPtr->sectorMask) { 1752 bp->b_error = EINVAL; 1753 bp->b_flags |= B_ERROR; 1754 bp->b_resid = bp->b_bcount; 1755 biodone(bp); 1756 RF_LOCK_MUTEX(raidPtr->mutex); 1757 continue; 1758 1759 } 1760 db1_printf(("Calling DoAccess..\n")); 1761 1762 1763 RF_LOCK_MUTEX(raidPtr->mutex); 1764 raidPtr->openings--; 1765 RF_UNLOCK_MUTEX(raidPtr->mutex); 1766 1767 /* 1768 * Everything is async. 1769 */ 1770 do_async = 1; 1771 1772 disk_busy(&rs->sc_dkdev); 1773 1774 /* XXX we're still at splbio() here... do we *really* 1775 need to be? */ 1776 1777 /* don't ever condition on bp->b_flags & B_WRITE. 1778 * always condition on B_READ instead */ 1779 1780 rc = rf_DoAccess(raidPtr, (bp->b_flags & B_READ) ? 1781 RF_IO_TYPE_READ : RF_IO_TYPE_WRITE, 1782 do_async, raid_addr, num_blocks, 1783 bp->b_data, bp, RF_DAG_NONBLOCKING_IO); 1784 1785 if (rc) { 1786 bp->b_error = rc; 1787 bp->b_flags |= B_ERROR; 1788 bp->b_resid = bp->b_bcount; 1789 biodone(bp); 1790 /* continue loop */ 1791 } 1792 1793 RF_LOCK_MUTEX(raidPtr->mutex); 1794 } 1795 RF_UNLOCK_MUTEX(raidPtr->mutex); 1796 } 1797 1798 1799 1800 1801 /* invoke an I/O from kernel mode. Disk queue should be locked upon entry */ 1802 1803 int 1804 rf_DispatchKernelIO(RF_DiskQueue_t *queue, RF_DiskQueueData_t *req) 1805 { 1806 int op = (req->type == RF_IO_TYPE_READ) ? B_READ : B_WRITE; 1807 struct buf *bp; 1808 1809 req->queue = queue; 1810 1811 #if DIAGNOSTIC 1812 if (queue->raidPtr->raidid >= numraid) { 1813 printf("Invalid unit number: %d %d\n", queue->raidPtr->raidid, 1814 numraid); 1815 panic("Invalid Unit number in rf_DispatchKernelIO"); 1816 } 1817 #endif 1818 1819 bp = req->bp; 1820 1821 switch (req->type) { 1822 case RF_IO_TYPE_NOP: /* used primarily to unlock a locked queue */ 1823 /* XXX need to do something extra here.. */ 1824 /* I'm leaving this in, as I've never actually seen it used, 1825 * and I'd like folks to report it... GO */ 1826 printf(("WAKEUP CALLED\n")); 1827 queue->numOutstanding++; 1828 1829 bp->b_flags = 0; 1830 bp->b_private = req; 1831 1832 KernelWakeupFunc(bp); 1833 break; 1834 1835 case RF_IO_TYPE_READ: 1836 case RF_IO_TYPE_WRITE: 1837 #if RF_ACC_TRACE > 0 1838 if (req->tracerec) { 1839 RF_ETIMER_START(req->tracerec->timer); 1840 } 1841 #endif 1842 InitBP(bp, queue->rf_cinfo->ci_vp, 1843 op, queue->rf_cinfo->ci_dev, 1844 req->sectorOffset, req->numSector, 1845 req->buf, KernelWakeupFunc, (void *) req, 1846 queue->raidPtr->logBytesPerSector, req->b_proc); 1847 1848 if (rf_debugKernelAccess) { 1849 db1_printf(("dispatch: bp->b_blkno = %ld\n", 1850 (long) bp->b_blkno)); 1851 } 1852 queue->numOutstanding++; 1853 queue->last_deq_sector = req->sectorOffset; 1854 /* acc wouldn't have been let in if there were any pending 1855 * reqs at any other priority */ 1856 queue->curPriority = req->priority; 1857 1858 db1_printf(("Going for %c to unit %d col %d\n", 1859 req->type, queue->raidPtr->raidid, 1860 queue->col)); 1861 db1_printf(("sector %d count %d (%d bytes) %d\n", 1862 (int) req->sectorOffset, (int) req->numSector, 1863 (int) (req->numSector << 1864 queue->raidPtr->logBytesPerSector), 1865 (int) queue->raidPtr->logBytesPerSector)); 1866 VOP_STRATEGY(bp->b_vp, bp); 1867 1868 break; 1869 1870 default: 1871 panic("bad req->type in rf_DispatchKernelIO"); 1872 } 1873 db1_printf(("Exiting from DispatchKernelIO\n")); 1874 1875 return (0); 1876 } 1877 /* this is the callback function associated with a I/O invoked from 1878 kernel code. 1879 */ 1880 static void 1881 KernelWakeupFunc(struct buf *bp) 1882 { 1883 RF_DiskQueueData_t *req = NULL; 1884 RF_DiskQueue_t *queue; 1885 int s; 1886 1887 s = splbio(); 1888 db1_printf(("recovering the request queue:\n")); 1889 req = bp->b_private; 1890 1891 queue = (RF_DiskQueue_t *) req->queue; 1892 1893 #if RF_ACC_TRACE > 0 1894 if (req->tracerec) { 1895 RF_ETIMER_STOP(req->tracerec->timer); 1896 RF_ETIMER_EVAL(req->tracerec->timer); 1897 RF_LOCK_MUTEX(rf_tracing_mutex); 1898 req->tracerec->diskwait_us += RF_ETIMER_VAL_US(req->tracerec->timer); 1899 req->tracerec->phys_io_us += RF_ETIMER_VAL_US(req->tracerec->timer); 1900 req->tracerec->num_phys_ios++; 1901 RF_UNLOCK_MUTEX(rf_tracing_mutex); 1902 } 1903 #endif 1904 1905 /* XXX Ok, let's get aggressive... If B_ERROR is set, let's go 1906 * ballistic, and mark the component as hosed... */ 1907 1908 if (bp->b_flags & B_ERROR) { 1909 /* Mark the disk as dead */ 1910 /* but only mark it once... */ 1911 /* and only if it wouldn't leave this RAID set 1912 completely broken */ 1913 if (((queue->raidPtr->Disks[queue->col].status == 1914 rf_ds_optimal) || 1915 (queue->raidPtr->Disks[queue->col].status == 1916 rf_ds_used_spare)) && 1917 (queue->raidPtr->numFailures < 1918 queue->raidPtr->Layout.map->faultsTolerated)) { 1919 printf("raid%d: IO Error. Marking %s as failed.\n", 1920 queue->raidPtr->raidid, 1921 queue->raidPtr->Disks[queue->col].devname); 1922 queue->raidPtr->Disks[queue->col].status = 1923 rf_ds_failed; 1924 queue->raidPtr->status = rf_rs_degraded; 1925 queue->raidPtr->numFailures++; 1926 queue->raidPtr->numNewFailures++; 1927 } else { /* Disk is already dead... */ 1928 /* printf("Disk already marked as dead!\n"); */ 1929 } 1930 1931 } 1932 1933 /* Fill in the error value */ 1934 1935 req->error = (bp->b_flags & B_ERROR) ? bp->b_error : 0; 1936 1937 simple_lock(&queue->raidPtr->iodone_lock); 1938 1939 /* Drop this one on the "finished" queue... */ 1940 TAILQ_INSERT_TAIL(&(queue->raidPtr->iodone), req, iodone_entries); 1941 1942 /* Let the raidio thread know there is work to be done. */ 1943 wakeup(&(queue->raidPtr->iodone)); 1944 1945 simple_unlock(&queue->raidPtr->iodone_lock); 1946 1947 splx(s); 1948 } 1949 1950 1951 1952 /* 1953 * initialize a buf structure for doing an I/O in the kernel. 1954 */ 1955 static void 1956 InitBP(struct buf *bp, struct vnode *b_vp, unsigned rw_flag, dev_t dev, 1957 RF_SectorNum_t startSect, RF_SectorCount_t numSect, caddr_t bf, 1958 void (*cbFunc) (struct buf *), void *cbArg, int logBytesPerSector, 1959 struct proc *b_proc) 1960 { 1961 /* bp->b_flags = B_PHYS | rw_flag; */ 1962 bp->b_flags = B_CALL | rw_flag; /* XXX need B_PHYS here too??? */ 1963 bp->b_bcount = numSect << logBytesPerSector; 1964 bp->b_bufsize = bp->b_bcount; 1965 bp->b_error = 0; 1966 bp->b_dev = dev; 1967 bp->b_data = bf; 1968 bp->b_blkno = startSect; 1969 bp->b_resid = bp->b_bcount; /* XXX is this right!??!?!! */ 1970 if (bp->b_bcount == 0) { 1971 panic("bp->b_bcount is zero in InitBP!!"); 1972 } 1973 bp->b_proc = b_proc; 1974 bp->b_iodone = cbFunc; 1975 bp->b_private = cbArg; 1976 bp->b_vp = b_vp; 1977 if ((bp->b_flags & B_READ) == 0) { 1978 bp->b_vp->v_numoutput++; 1979 } 1980 1981 } 1982 1983 static void 1984 raidgetdefaultlabel(RF_Raid_t *raidPtr, struct raid_softc *rs, 1985 struct disklabel *lp) 1986 { 1987 memset(lp, 0, sizeof(*lp)); 1988 1989 /* fabricate a label... */ 1990 lp->d_secperunit = raidPtr->totalSectors; 1991 lp->d_secsize = raidPtr->bytesPerSector; 1992 lp->d_nsectors = raidPtr->Layout.dataSectorsPerStripe; 1993 lp->d_ntracks = 4 * raidPtr->numCol; 1994 lp->d_ncylinders = raidPtr->totalSectors / 1995 (lp->d_nsectors * lp->d_ntracks); 1996 lp->d_secpercyl = lp->d_ntracks * lp->d_nsectors; 1997 1998 strncpy(lp->d_typename, "raid", sizeof(lp->d_typename)); 1999 lp->d_type = DTYPE_RAID; 2000 strncpy(lp->d_packname, "fictitious", sizeof(lp->d_packname)); 2001 lp->d_rpm = 3600; 2002 lp->d_interleave = 1; 2003 lp->d_flags = 0; 2004 2005 lp->d_partitions[RAW_PART].p_offset = 0; 2006 lp->d_partitions[RAW_PART].p_size = raidPtr->totalSectors; 2007 lp->d_partitions[RAW_PART].p_fstype = FS_UNUSED; 2008 lp->d_npartitions = RAW_PART + 1; 2009 2010 lp->d_magic = DISKMAGIC; 2011 lp->d_magic2 = DISKMAGIC; 2012 lp->d_checksum = dkcksum(rs->sc_dkdev.dk_label); 2013 2014 } 2015 /* 2016 * Read the disklabel from the raid device. If one is not present, fake one 2017 * up. 2018 */ 2019 static void 2020 raidgetdisklabel(dev_t dev) 2021 { 2022 int unit = raidunit(dev); 2023 struct raid_softc *rs = &raid_softc[unit]; 2024 const char *errstring; 2025 struct disklabel *lp = rs->sc_dkdev.dk_label; 2026 struct cpu_disklabel *clp = rs->sc_dkdev.dk_cpulabel; 2027 RF_Raid_t *raidPtr; 2028 2029 db1_printf(("Getting the disklabel...\n")); 2030 2031 memset(clp, 0, sizeof(*clp)); 2032 2033 raidPtr = raidPtrs[unit]; 2034 2035 raidgetdefaultlabel(raidPtr, rs, lp); 2036 2037 /* 2038 * Call the generic disklabel extraction routine. 2039 */ 2040 errstring = readdisklabel(RAIDLABELDEV(dev), raidstrategy, 2041 rs->sc_dkdev.dk_label, rs->sc_dkdev.dk_cpulabel); 2042 if (errstring) 2043 raidmakedisklabel(rs); 2044 else { 2045 int i; 2046 struct partition *pp; 2047 2048 /* 2049 * Sanity check whether the found disklabel is valid. 2050 * 2051 * This is necessary since total size of the raid device 2052 * may vary when an interleave is changed even though exactly 2053 * same componets are used, and old disklabel may used 2054 * if that is found. 2055 */ 2056 if (lp->d_secperunit != rs->sc_size) 2057 printf("raid%d: WARNING: %s: " 2058 "total sector size in disklabel (%d) != " 2059 "the size of raid (%ld)\n", unit, rs->sc_xname, 2060 lp->d_secperunit, (long) rs->sc_size); 2061 for (i = 0; i < lp->d_npartitions; i++) { 2062 pp = &lp->d_partitions[i]; 2063 if (pp->p_offset + pp->p_size > rs->sc_size) 2064 printf("raid%d: WARNING: %s: end of partition `%c' " 2065 "exceeds the size of raid (%ld)\n", 2066 unit, rs->sc_xname, 'a' + i, (long) rs->sc_size); 2067 } 2068 } 2069 2070 } 2071 /* 2072 * Take care of things one might want to take care of in the event 2073 * that a disklabel isn't present. 2074 */ 2075 static void 2076 raidmakedisklabel(struct raid_softc *rs) 2077 { 2078 struct disklabel *lp = rs->sc_dkdev.dk_label; 2079 db1_printf(("Making a label..\n")); 2080 2081 /* 2082 * For historical reasons, if there's no disklabel present 2083 * the raw partition must be marked FS_BSDFFS. 2084 */ 2085 2086 lp->d_partitions[RAW_PART].p_fstype = FS_BSDFFS; 2087 2088 strncpy(lp->d_packname, "default label", sizeof(lp->d_packname)); 2089 2090 lp->d_checksum = dkcksum(lp); 2091 } 2092 /* 2093 * Lookup the provided name in the filesystem. If the file exists, 2094 * is a valid block device, and isn't being used by anyone else, 2095 * set *vpp to the file's vnode. 2096 * You'll find the original of this in ccd.c 2097 */ 2098 int 2099 raidlookup(char *path, struct lwp *l, struct vnode **vpp) 2100 { 2101 struct nameidata nd; 2102 struct vnode *vp; 2103 struct proc *p; 2104 struct vattr va; 2105 int error; 2106 2107 if (l == NULL) 2108 return(ESRCH); /* Is ESRCH the best choice? */ 2109 p = l->l_proc; 2110 2111 NDINIT(&nd, LOOKUP, FOLLOW, UIO_SYSSPACE, path, l); 2112 if ((error = vn_open(&nd, FREAD | FWRITE, 0)) != 0) { 2113 return (error); 2114 } 2115 vp = nd.ni_vp; 2116 if (vp->v_usecount > 1) { 2117 VOP_UNLOCK(vp, 0); 2118 (void) vn_close(vp, FREAD | FWRITE, p->p_cred, l); 2119 return (EBUSY); 2120 } 2121 if ((error = VOP_GETATTR(vp, &va, p->p_cred, l)) != 0) { 2122 VOP_UNLOCK(vp, 0); 2123 (void) vn_close(vp, FREAD | FWRITE, p->p_cred, l); 2124 return (error); 2125 } 2126 /* XXX: eventually we should handle VREG, too. */ 2127 if (va.va_type != VBLK) { 2128 VOP_UNLOCK(vp, 0); 2129 (void) vn_close(vp, FREAD | FWRITE, p->p_cred, l); 2130 return (ENOTBLK); 2131 } 2132 VOP_UNLOCK(vp, 0); 2133 *vpp = vp; 2134 return (0); 2135 } 2136 /* 2137 * Wait interruptibly for an exclusive lock. 2138 * 2139 * XXX 2140 * Several drivers do this; it should be abstracted and made MP-safe. 2141 * (Hmm... where have we seen this warning before :-> GO ) 2142 */ 2143 static int 2144 raidlock(struct raid_softc *rs) 2145 { 2146 int error; 2147 2148 while ((rs->sc_flags & RAIDF_LOCKED) != 0) { 2149 rs->sc_flags |= RAIDF_WANTED; 2150 if ((error = 2151 tsleep(rs, PRIBIO | PCATCH, "raidlck", 0)) != 0) 2152 return (error); 2153 } 2154 rs->sc_flags |= RAIDF_LOCKED; 2155 return (0); 2156 } 2157 /* 2158 * Unlock and wake up any waiters. 2159 */ 2160 static void 2161 raidunlock(struct raid_softc *rs) 2162 { 2163 2164 rs->sc_flags &= ~RAIDF_LOCKED; 2165 if ((rs->sc_flags & RAIDF_WANTED) != 0) { 2166 rs->sc_flags &= ~RAIDF_WANTED; 2167 wakeup(rs); 2168 } 2169 } 2170 2171 2172 #define RF_COMPONENT_INFO_OFFSET 16384 /* bytes */ 2173 #define RF_COMPONENT_INFO_SIZE 1024 /* bytes */ 2174 2175 int 2176 raidmarkclean(dev_t dev, struct vnode *b_vp, int mod_counter) 2177 { 2178 RF_ComponentLabel_t clabel; 2179 raidread_component_label(dev, b_vp, &clabel); 2180 clabel.mod_counter = mod_counter; 2181 clabel.clean = RF_RAID_CLEAN; 2182 raidwrite_component_label(dev, b_vp, &clabel); 2183 return(0); 2184 } 2185 2186 2187 int 2188 raidmarkdirty(dev_t dev, struct vnode *b_vp, int mod_counter) 2189 { 2190 RF_ComponentLabel_t clabel; 2191 raidread_component_label(dev, b_vp, &clabel); 2192 clabel.mod_counter = mod_counter; 2193 clabel.clean = RF_RAID_DIRTY; 2194 raidwrite_component_label(dev, b_vp, &clabel); 2195 return(0); 2196 } 2197 2198 /* ARGSUSED */ 2199 int 2200 raidread_component_label(dev_t dev, struct vnode *b_vp, 2201 RF_ComponentLabel_t *clabel) 2202 { 2203 struct buf *bp; 2204 const struct bdevsw *bdev; 2205 int error; 2206 2207 /* XXX should probably ensure that we don't try to do this if 2208 someone has changed rf_protected_sectors. */ 2209 2210 if (b_vp == NULL) { 2211 /* For whatever reason, this component is not valid. 2212 Don't try to read a component label from it. */ 2213 return(EINVAL); 2214 } 2215 2216 /* get a block of the appropriate size... */ 2217 bp = geteblk((int)RF_COMPONENT_INFO_SIZE); 2218 bp->b_dev = dev; 2219 2220 /* get our ducks in a row for the read */ 2221 bp->b_blkno = RF_COMPONENT_INFO_OFFSET / DEV_BSIZE; 2222 bp->b_bcount = RF_COMPONENT_INFO_SIZE; 2223 bp->b_flags |= B_READ; 2224 bp->b_resid = RF_COMPONENT_INFO_SIZE / DEV_BSIZE; 2225 2226 bdev = bdevsw_lookup(bp->b_dev); 2227 if (bdev == NULL) 2228 return (ENXIO); 2229 (*bdev->d_strategy)(bp); 2230 2231 error = biowait(bp); 2232 2233 if (!error) { 2234 memcpy(clabel, bp->b_data, 2235 sizeof(RF_ComponentLabel_t)); 2236 } 2237 2238 brelse(bp); 2239 return(error); 2240 } 2241 /* ARGSUSED */ 2242 int 2243 raidwrite_component_label(dev_t dev, struct vnode *b_vp, 2244 RF_ComponentLabel_t *clabel) 2245 { 2246 struct buf *bp; 2247 const struct bdevsw *bdev; 2248 int error; 2249 2250 /* get a block of the appropriate size... */ 2251 bp = geteblk((int)RF_COMPONENT_INFO_SIZE); 2252 bp->b_dev = dev; 2253 2254 /* get our ducks in a row for the write */ 2255 bp->b_blkno = RF_COMPONENT_INFO_OFFSET / DEV_BSIZE; 2256 bp->b_bcount = RF_COMPONENT_INFO_SIZE; 2257 bp->b_flags |= B_WRITE; 2258 bp->b_resid = RF_COMPONENT_INFO_SIZE / DEV_BSIZE; 2259 2260 memset(bp->b_data, 0, RF_COMPONENT_INFO_SIZE ); 2261 2262 memcpy(bp->b_data, clabel, sizeof(RF_ComponentLabel_t)); 2263 2264 bdev = bdevsw_lookup(bp->b_dev); 2265 if (bdev == NULL) 2266 return (ENXIO); 2267 (*bdev->d_strategy)(bp); 2268 error = biowait(bp); 2269 brelse(bp); 2270 if (error) { 2271 #if 1 2272 printf("Failed to write RAID component info!\n"); 2273 #endif 2274 } 2275 2276 return(error); 2277 } 2278 2279 void 2280 rf_markalldirty(RF_Raid_t *raidPtr) 2281 { 2282 RF_ComponentLabel_t clabel; 2283 int sparecol; 2284 int c; 2285 int j; 2286 int scol = -1; 2287 2288 raidPtr->mod_counter++; 2289 for (c = 0; c < raidPtr->numCol; c++) { 2290 /* we don't want to touch (at all) a disk that has 2291 failed */ 2292 if (!RF_DEAD_DISK(raidPtr->Disks[c].status)) { 2293 raidread_component_label( 2294 raidPtr->Disks[c].dev, 2295 raidPtr->raid_cinfo[c].ci_vp, 2296 &clabel); 2297 if (clabel.status == rf_ds_spared) { 2298 /* XXX do something special... 2299 but whatever you do, don't 2300 try to access it!! */ 2301 } else { 2302 raidmarkdirty( 2303 raidPtr->Disks[c].dev, 2304 raidPtr->raid_cinfo[c].ci_vp, 2305 raidPtr->mod_counter); 2306 } 2307 } 2308 } 2309 2310 for( c = 0; c < raidPtr->numSpare ; c++) { 2311 sparecol = raidPtr->numCol + c; 2312 if (raidPtr->Disks[sparecol].status == rf_ds_used_spare) { 2313 /* 2314 2315 we claim this disk is "optimal" if it's 2316 rf_ds_used_spare, as that means it should be 2317 directly substitutable for the disk it replaced. 2318 We note that too... 2319 2320 */ 2321 2322 for(j=0;j<raidPtr->numCol;j++) { 2323 if (raidPtr->Disks[j].spareCol == sparecol) { 2324 scol = j; 2325 break; 2326 } 2327 } 2328 2329 raidread_component_label( 2330 raidPtr->Disks[sparecol].dev, 2331 raidPtr->raid_cinfo[sparecol].ci_vp, 2332 &clabel); 2333 /* make sure status is noted */ 2334 2335 raid_init_component_label(raidPtr, &clabel); 2336 2337 clabel.row = 0; 2338 clabel.column = scol; 2339 /* Note: we *don't* change status from rf_ds_used_spare 2340 to rf_ds_optimal */ 2341 /* clabel.status = rf_ds_optimal; */ 2342 2343 raidmarkdirty(raidPtr->Disks[sparecol].dev, 2344 raidPtr->raid_cinfo[sparecol].ci_vp, 2345 raidPtr->mod_counter); 2346 } 2347 } 2348 } 2349 2350 2351 void 2352 rf_update_component_labels(RF_Raid_t *raidPtr, int final) 2353 { 2354 RF_ComponentLabel_t clabel; 2355 int sparecol; 2356 int c; 2357 int j; 2358 int scol; 2359 2360 scol = -1; 2361 2362 /* XXX should do extra checks to make sure things really are clean, 2363 rather than blindly setting the clean bit... */ 2364 2365 raidPtr->mod_counter++; 2366 2367 for (c = 0; c < raidPtr->numCol; c++) { 2368 if (raidPtr->Disks[c].status == rf_ds_optimal) { 2369 raidread_component_label( 2370 raidPtr->Disks[c].dev, 2371 raidPtr->raid_cinfo[c].ci_vp, 2372 &clabel); 2373 /* make sure status is noted */ 2374 clabel.status = rf_ds_optimal; 2375 2376 /* bump the counter */ 2377 clabel.mod_counter = raidPtr->mod_counter; 2378 2379 raidwrite_component_label( 2380 raidPtr->Disks[c].dev, 2381 raidPtr->raid_cinfo[c].ci_vp, 2382 &clabel); 2383 if (final == RF_FINAL_COMPONENT_UPDATE) { 2384 if (raidPtr->parity_good == RF_RAID_CLEAN) { 2385 raidmarkclean( 2386 raidPtr->Disks[c].dev, 2387 raidPtr->raid_cinfo[c].ci_vp, 2388 raidPtr->mod_counter); 2389 } 2390 } 2391 } 2392 /* else we don't touch it.. */ 2393 } 2394 2395 for( c = 0; c < raidPtr->numSpare ; c++) { 2396 sparecol = raidPtr->numCol + c; 2397 /* Need to ensure that the reconstruct actually completed! */ 2398 if (raidPtr->Disks[sparecol].status == rf_ds_used_spare) { 2399 /* 2400 2401 we claim this disk is "optimal" if it's 2402 rf_ds_used_spare, as that means it should be 2403 directly substitutable for the disk it replaced. 2404 We note that too... 2405 2406 */ 2407 2408 for(j=0;j<raidPtr->numCol;j++) { 2409 if (raidPtr->Disks[j].spareCol == sparecol) { 2410 scol = j; 2411 break; 2412 } 2413 } 2414 2415 /* XXX shouldn't *really* need this... */ 2416 raidread_component_label( 2417 raidPtr->Disks[sparecol].dev, 2418 raidPtr->raid_cinfo[sparecol].ci_vp, 2419 &clabel); 2420 /* make sure status is noted */ 2421 2422 raid_init_component_label(raidPtr, &clabel); 2423 2424 clabel.mod_counter = raidPtr->mod_counter; 2425 clabel.column = scol; 2426 clabel.status = rf_ds_optimal; 2427 2428 raidwrite_component_label( 2429 raidPtr->Disks[sparecol].dev, 2430 raidPtr->raid_cinfo[sparecol].ci_vp, 2431 &clabel); 2432 if (final == RF_FINAL_COMPONENT_UPDATE) { 2433 if (raidPtr->parity_good == RF_RAID_CLEAN) { 2434 raidmarkclean( raidPtr->Disks[sparecol].dev, 2435 raidPtr->raid_cinfo[sparecol].ci_vp, 2436 raidPtr->mod_counter); 2437 } 2438 } 2439 } 2440 } 2441 } 2442 2443 void 2444 rf_close_component(RF_Raid_t *raidPtr, struct vnode *vp, int auto_configured) 2445 { 2446 struct proc *p; 2447 struct lwp *l; 2448 2449 p = raidPtr->engine_thread; 2450 l = LIST_FIRST(&p->p_lwps); 2451 2452 if (vp != NULL) { 2453 if (auto_configured == 1) { 2454 vn_lock(vp, LK_EXCLUSIVE | LK_RETRY); 2455 VOP_CLOSE(vp, FREAD | FWRITE, NOCRED, 0); 2456 vput(vp); 2457 2458 } else { 2459 (void) vn_close(vp, FREAD | FWRITE, p->p_cred, l); 2460 } 2461 } 2462 } 2463 2464 2465 void 2466 rf_UnconfigureVnodes(RF_Raid_t *raidPtr) 2467 { 2468 int r,c; 2469 struct vnode *vp; 2470 int acd; 2471 2472 2473 /* We take this opportunity to close the vnodes like we should.. */ 2474 2475 for (c = 0; c < raidPtr->numCol; c++) { 2476 vp = raidPtr->raid_cinfo[c].ci_vp; 2477 acd = raidPtr->Disks[c].auto_configured; 2478 rf_close_component(raidPtr, vp, acd); 2479 raidPtr->raid_cinfo[c].ci_vp = NULL; 2480 raidPtr->Disks[c].auto_configured = 0; 2481 } 2482 2483 for (r = 0; r < raidPtr->numSpare; r++) { 2484 vp = raidPtr->raid_cinfo[raidPtr->numCol + r].ci_vp; 2485 acd = raidPtr->Disks[raidPtr->numCol + r].auto_configured; 2486 rf_close_component(raidPtr, vp, acd); 2487 raidPtr->raid_cinfo[raidPtr->numCol + r].ci_vp = NULL; 2488 raidPtr->Disks[raidPtr->numCol + r].auto_configured = 0; 2489 } 2490 } 2491 2492 2493 void 2494 rf_ReconThread(struct rf_recon_req *req) 2495 { 2496 int s; 2497 RF_Raid_t *raidPtr; 2498 2499 s = splbio(); 2500 raidPtr = (RF_Raid_t *) req->raidPtr; 2501 raidPtr->recon_in_progress = 1; 2502 2503 rf_FailDisk((RF_Raid_t *) req->raidPtr, req->col, 2504 ((req->flags & RF_FDFLAGS_RECON) ? 1 : 0)); 2505 2506 RF_Free(req, sizeof(*req)); 2507 2508 raidPtr->recon_in_progress = 0; 2509 splx(s); 2510 2511 /* That's all... */ 2512 kthread_exit(0); /* does not return */ 2513 } 2514 2515 void 2516 rf_RewriteParityThread(RF_Raid_t *raidPtr) 2517 { 2518 int retcode; 2519 int s; 2520 2521 raidPtr->parity_rewrite_stripes_done = 0; 2522 raidPtr->parity_rewrite_in_progress = 1; 2523 s = splbio(); 2524 retcode = rf_RewriteParity(raidPtr); 2525 splx(s); 2526 if (retcode) { 2527 printf("raid%d: Error re-writing parity!\n",raidPtr->raidid); 2528 } else { 2529 /* set the clean bit! If we shutdown correctly, 2530 the clean bit on each component label will get 2531 set */ 2532 raidPtr->parity_good = RF_RAID_CLEAN; 2533 } 2534 raidPtr->parity_rewrite_in_progress = 0; 2535 2536 /* Anyone waiting for us to stop? If so, inform them... */ 2537 if (raidPtr->waitShutdown) { 2538 wakeup(&raidPtr->parity_rewrite_in_progress); 2539 } 2540 2541 /* That's all... */ 2542 kthread_exit(0); /* does not return */ 2543 } 2544 2545 2546 void 2547 rf_CopybackThread(RF_Raid_t *raidPtr) 2548 { 2549 int s; 2550 2551 raidPtr->copyback_in_progress = 1; 2552 s = splbio(); 2553 rf_CopybackReconstructedData(raidPtr); 2554 splx(s); 2555 raidPtr->copyback_in_progress = 0; 2556 2557 /* That's all... */ 2558 kthread_exit(0); /* does not return */ 2559 } 2560 2561 2562 void 2563 rf_ReconstructInPlaceThread(struct rf_recon_req *req) 2564 { 2565 int s; 2566 RF_Raid_t *raidPtr; 2567 2568 s = splbio(); 2569 raidPtr = req->raidPtr; 2570 raidPtr->recon_in_progress = 1; 2571 rf_ReconstructInPlace(raidPtr, req->col); 2572 RF_Free(req, sizeof(*req)); 2573 raidPtr->recon_in_progress = 0; 2574 splx(s); 2575 2576 /* That's all... */ 2577 kthread_exit(0); /* does not return */ 2578 } 2579 2580 RF_AutoConfig_t * 2581 rf_find_raid_components() 2582 { 2583 struct vnode *vp; 2584 struct disklabel label; 2585 struct device *dv; 2586 dev_t dev; 2587 int bmajor; 2588 int error; 2589 int i; 2590 int good_one; 2591 RF_ComponentLabel_t *clabel; 2592 RF_AutoConfig_t *ac_list; 2593 RF_AutoConfig_t *ac; 2594 2595 2596 /* initialize the AutoConfig list */ 2597 ac_list = NULL; 2598 2599 /* we begin by trolling through *all* the devices on the system */ 2600 2601 for (dv = alldevs.tqh_first; dv != NULL; 2602 dv = dv->dv_list.tqe_next) { 2603 2604 /* we are only interested in disks... */ 2605 if (device_class(dv) != DV_DISK) 2606 continue; 2607 2608 /* we don't care about floppies... */ 2609 if (device_is_a(dv, "fd")) { 2610 continue; 2611 } 2612 2613 /* we don't care about CD's... */ 2614 if (device_is_a(dv, "cd")) { 2615 continue; 2616 } 2617 2618 /* hdfd is the Atari/Hades floppy driver */ 2619 if (device_is_a(dv, "hdfd")) { 2620 continue; 2621 } 2622 2623 /* fdisa is the Atari/Milan floppy driver */ 2624 if (device_is_a(dv, "fdisa")) { 2625 continue; 2626 } 2627 2628 /* need to find the device_name_to_block_device_major stuff */ 2629 bmajor = devsw_name2blk(dv->dv_xname, NULL, 0); 2630 2631 /* get a vnode for the raw partition of this disk */ 2632 2633 dev = MAKEDISKDEV(bmajor, device_unit(dv), RAW_PART); 2634 if (bdevvp(dev, &vp)) 2635 panic("RAID can't alloc vnode"); 2636 2637 error = VOP_OPEN(vp, FREAD, NOCRED, 0); 2638 2639 if (error) { 2640 /* "Who cares." Continue looking 2641 for something that exists*/ 2642 vput(vp); 2643 continue; 2644 } 2645 2646 /* Ok, the disk exists. Go get the disklabel. */ 2647 error = VOP_IOCTL(vp, DIOCGDINFO, &label, FREAD, NOCRED, 0); 2648 if (error) { 2649 /* 2650 * XXX can't happen - open() would 2651 * have errored out (or faked up one) 2652 */ 2653 if (error != ENOTTY) 2654 printf("RAIDframe: can't get label for dev " 2655 "%s (%d)\n", dv->dv_xname, error); 2656 } 2657 2658 /* don't need this any more. We'll allocate it again 2659 a little later if we really do... */ 2660 vn_lock(vp, LK_EXCLUSIVE | LK_RETRY); 2661 VOP_CLOSE(vp, FREAD | FWRITE, NOCRED, 0); 2662 vput(vp); 2663 2664 if (error) 2665 continue; 2666 2667 for (i=0; i < label.d_npartitions; i++) { 2668 /* We only support partitions marked as RAID */ 2669 if (label.d_partitions[i].p_fstype != FS_RAID) 2670 continue; 2671 2672 dev = MAKEDISKDEV(bmajor, device_unit(dv), i); 2673 if (bdevvp(dev, &vp)) 2674 panic("RAID can't alloc vnode"); 2675 2676 error = VOP_OPEN(vp, FREAD, NOCRED, 0); 2677 if (error) { 2678 /* Whatever... */ 2679 vput(vp); 2680 continue; 2681 } 2682 2683 good_one = 0; 2684 2685 clabel = (RF_ComponentLabel_t *) 2686 malloc(sizeof(RF_ComponentLabel_t), 2687 M_RAIDFRAME, M_NOWAIT); 2688 if (clabel == NULL) { 2689 while(ac_list) { 2690 ac = ac_list; 2691 if (ac->clabel) 2692 free(ac->clabel, M_RAIDFRAME); 2693 ac_list = ac_list->next; 2694 free(ac, M_RAIDFRAME); 2695 }; 2696 printf("RAID auto config: out of memory!\n"); 2697 return(NULL); /* XXX probably should panic? */ 2698 } 2699 2700 if (!raidread_component_label(dev, vp, clabel)) { 2701 /* Got the label. Does it look reasonable? */ 2702 if (rf_reasonable_label(clabel) && 2703 (clabel->partitionSize <= 2704 label.d_partitions[i].p_size)) { 2705 #if DEBUG 2706 printf("Component on: %s%c: %d\n", 2707 dv->dv_xname, 'a'+i, 2708 label.d_partitions[i].p_size); 2709 rf_print_component_label(clabel); 2710 #endif 2711 /* if it's reasonable, add it, 2712 else ignore it. */ 2713 ac = (RF_AutoConfig_t *) 2714 malloc(sizeof(RF_AutoConfig_t), 2715 M_RAIDFRAME, 2716 M_NOWAIT); 2717 if (ac == NULL) { 2718 /* XXX should panic?? */ 2719 while(ac_list) { 2720 ac = ac_list; 2721 if (ac->clabel) 2722 free(ac->clabel, 2723 M_RAIDFRAME); 2724 ac_list = ac_list->next; 2725 free(ac, M_RAIDFRAME); 2726 } 2727 free(clabel, M_RAIDFRAME); 2728 return(NULL); 2729 } 2730 2731 snprintf(ac->devname, 2732 sizeof(ac->devname), "%s%c", 2733 dv->dv_xname, 'a'+i); 2734 ac->dev = dev; 2735 ac->vp = vp; 2736 ac->clabel = clabel; 2737 ac->next = ac_list; 2738 ac_list = ac; 2739 good_one = 1; 2740 } 2741 } 2742 if (!good_one) { 2743 /* cleanup */ 2744 free(clabel, M_RAIDFRAME); 2745 vn_lock(vp, LK_EXCLUSIVE | LK_RETRY); 2746 VOP_CLOSE(vp, FREAD | FWRITE, NOCRED, 0); 2747 vput(vp); 2748 } 2749 } 2750 } 2751 return(ac_list); 2752 } 2753 2754 static int 2755 rf_reasonable_label(RF_ComponentLabel_t *clabel) 2756 { 2757 2758 if (((clabel->version==RF_COMPONENT_LABEL_VERSION_1) || 2759 (clabel->version==RF_COMPONENT_LABEL_VERSION)) && 2760 ((clabel->clean == RF_RAID_CLEAN) || 2761 (clabel->clean == RF_RAID_DIRTY)) && 2762 clabel->row >=0 && 2763 clabel->column >= 0 && 2764 clabel->num_rows > 0 && 2765 clabel->num_columns > 0 && 2766 clabel->row < clabel->num_rows && 2767 clabel->column < clabel->num_columns && 2768 clabel->blockSize > 0 && 2769 clabel->numBlocks > 0) { 2770 /* label looks reasonable enough... */ 2771 return(1); 2772 } 2773 return(0); 2774 } 2775 2776 2777 #if DEBUG 2778 void 2779 rf_print_component_label(RF_ComponentLabel_t *clabel) 2780 { 2781 printf(" Row: %d Column: %d Num Rows: %d Num Columns: %d\n", 2782 clabel->row, clabel->column, 2783 clabel->num_rows, clabel->num_columns); 2784 printf(" Version: %d Serial Number: %d Mod Counter: %d\n", 2785 clabel->version, clabel->serial_number, 2786 clabel->mod_counter); 2787 printf(" Clean: %s Status: %d\n", 2788 clabel->clean ? "Yes" : "No", clabel->status ); 2789 printf(" sectPerSU: %d SUsPerPU: %d SUsPerRU: %d\n", 2790 clabel->sectPerSU, clabel->SUsPerPU, clabel->SUsPerRU); 2791 printf(" RAID Level: %c blocksize: %d numBlocks: %d\n", 2792 (char) clabel->parityConfig, clabel->blockSize, 2793 clabel->numBlocks); 2794 printf(" Autoconfig: %s\n", clabel->autoconfigure ? "Yes" : "No" ); 2795 printf(" Contains root partition: %s\n", 2796 clabel->root_partition ? "Yes" : "No" ); 2797 printf(" Last configured as: raid%d\n", clabel->last_unit ); 2798 #if 0 2799 printf(" Config order: %d\n", clabel->config_order); 2800 #endif 2801 2802 } 2803 #endif 2804 2805 RF_ConfigSet_t * 2806 rf_create_auto_sets(RF_AutoConfig_t *ac_list) 2807 { 2808 RF_AutoConfig_t *ac; 2809 RF_ConfigSet_t *config_sets; 2810 RF_ConfigSet_t *cset; 2811 RF_AutoConfig_t *ac_next; 2812 2813 2814 config_sets = NULL; 2815 2816 /* Go through the AutoConfig list, and figure out which components 2817 belong to what sets. */ 2818 ac = ac_list; 2819 while(ac!=NULL) { 2820 /* we're going to putz with ac->next, so save it here 2821 for use at the end of the loop */ 2822 ac_next = ac->next; 2823 2824 if (config_sets == NULL) { 2825 /* will need at least this one... */ 2826 config_sets = (RF_ConfigSet_t *) 2827 malloc(sizeof(RF_ConfigSet_t), 2828 M_RAIDFRAME, M_NOWAIT); 2829 if (config_sets == NULL) { 2830 panic("rf_create_auto_sets: No memory!"); 2831 } 2832 /* this one is easy :) */ 2833 config_sets->ac = ac; 2834 config_sets->next = NULL; 2835 config_sets->rootable = 0; 2836 ac->next = NULL; 2837 } else { 2838 /* which set does this component fit into? */ 2839 cset = config_sets; 2840 while(cset!=NULL) { 2841 if (rf_does_it_fit(cset, ac)) { 2842 /* looks like it matches... */ 2843 ac->next = cset->ac; 2844 cset->ac = ac; 2845 break; 2846 } 2847 cset = cset->next; 2848 } 2849 if (cset==NULL) { 2850 /* didn't find a match above... new set..*/ 2851 cset = (RF_ConfigSet_t *) 2852 malloc(sizeof(RF_ConfigSet_t), 2853 M_RAIDFRAME, M_NOWAIT); 2854 if (cset == NULL) { 2855 panic("rf_create_auto_sets: No memory!"); 2856 } 2857 cset->ac = ac; 2858 ac->next = NULL; 2859 cset->next = config_sets; 2860 cset->rootable = 0; 2861 config_sets = cset; 2862 } 2863 } 2864 ac = ac_next; 2865 } 2866 2867 2868 return(config_sets); 2869 } 2870 2871 static int 2872 rf_does_it_fit(RF_ConfigSet_t *cset, RF_AutoConfig_t *ac) 2873 { 2874 RF_ComponentLabel_t *clabel1, *clabel2; 2875 2876 /* If this one matches the *first* one in the set, that's good 2877 enough, since the other members of the set would have been 2878 through here too... */ 2879 /* note that we are not checking partitionSize here.. 2880 2881 Note that we are also not checking the mod_counters here. 2882 If everything else matches execpt the mod_counter, that's 2883 good enough for this test. We will deal with the mod_counters 2884 a little later in the autoconfiguration process. 2885 2886 (clabel1->mod_counter == clabel2->mod_counter) && 2887 2888 The reason we don't check for this is that failed disks 2889 will have lower modification counts. If those disks are 2890 not added to the set they used to belong to, then they will 2891 form their own set, which may result in 2 different sets, 2892 for example, competing to be configured at raid0, and 2893 perhaps competing to be the root filesystem set. If the 2894 wrong ones get configured, or both attempt to become /, 2895 weird behaviour and or serious lossage will occur. Thus we 2896 need to bring them into the fold here, and kick them out at 2897 a later point. 2898 2899 */ 2900 2901 clabel1 = cset->ac->clabel; 2902 clabel2 = ac->clabel; 2903 if ((clabel1->version == clabel2->version) && 2904 (clabel1->serial_number == clabel2->serial_number) && 2905 (clabel1->num_rows == clabel2->num_rows) && 2906 (clabel1->num_columns == clabel2->num_columns) && 2907 (clabel1->sectPerSU == clabel2->sectPerSU) && 2908 (clabel1->SUsPerPU == clabel2->SUsPerPU) && 2909 (clabel1->SUsPerRU == clabel2->SUsPerRU) && 2910 (clabel1->parityConfig == clabel2->parityConfig) && 2911 (clabel1->maxOutstanding == clabel2->maxOutstanding) && 2912 (clabel1->blockSize == clabel2->blockSize) && 2913 (clabel1->numBlocks == clabel2->numBlocks) && 2914 (clabel1->autoconfigure == clabel2->autoconfigure) && 2915 (clabel1->root_partition == clabel2->root_partition) && 2916 (clabel1->last_unit == clabel2->last_unit) && 2917 (clabel1->config_order == clabel2->config_order)) { 2918 /* if it get's here, it almost *has* to be a match */ 2919 } else { 2920 /* it's not consistent with somebody in the set.. 2921 punt */ 2922 return(0); 2923 } 2924 /* all was fine.. it must fit... */ 2925 return(1); 2926 } 2927 2928 int 2929 rf_have_enough_components(RF_ConfigSet_t *cset) 2930 { 2931 RF_AutoConfig_t *ac; 2932 RF_AutoConfig_t *auto_config; 2933 RF_ComponentLabel_t *clabel; 2934 int c; 2935 int num_cols; 2936 int num_missing; 2937 int mod_counter; 2938 int mod_counter_found; 2939 int even_pair_failed; 2940 char parity_type; 2941 2942 2943 /* check to see that we have enough 'live' components 2944 of this set. If so, we can configure it if necessary */ 2945 2946 num_cols = cset->ac->clabel->num_columns; 2947 parity_type = cset->ac->clabel->parityConfig; 2948 2949 /* XXX Check for duplicate components!?!?!? */ 2950 2951 /* Determine what the mod_counter is supposed to be for this set. */ 2952 2953 mod_counter_found = 0; 2954 mod_counter = 0; 2955 ac = cset->ac; 2956 while(ac!=NULL) { 2957 if (mod_counter_found==0) { 2958 mod_counter = ac->clabel->mod_counter; 2959 mod_counter_found = 1; 2960 } else { 2961 if (ac->clabel->mod_counter > mod_counter) { 2962 mod_counter = ac->clabel->mod_counter; 2963 } 2964 } 2965 ac = ac->next; 2966 } 2967 2968 num_missing = 0; 2969 auto_config = cset->ac; 2970 2971 even_pair_failed = 0; 2972 for(c=0; c<num_cols; c++) { 2973 ac = auto_config; 2974 while(ac!=NULL) { 2975 if ((ac->clabel->column == c) && 2976 (ac->clabel->mod_counter == mod_counter)) { 2977 /* it's this one... */ 2978 #if DEBUG 2979 printf("Found: %s at %d\n", 2980 ac->devname,c); 2981 #endif 2982 break; 2983 } 2984 ac=ac->next; 2985 } 2986 if (ac==NULL) { 2987 /* Didn't find one here! */ 2988 /* special case for RAID 1, especially 2989 where there are more than 2 2990 components (where RAIDframe treats 2991 things a little differently :( ) */ 2992 if (parity_type == '1') { 2993 if (c%2 == 0) { /* even component */ 2994 even_pair_failed = 1; 2995 } else { /* odd component. If 2996 we're failed, and 2997 so is the even 2998 component, it's 2999 "Good Night, Charlie" */ 3000 if (even_pair_failed == 1) { 3001 return(0); 3002 } 3003 } 3004 } else { 3005 /* normal accounting */ 3006 num_missing++; 3007 } 3008 } 3009 if ((parity_type == '1') && (c%2 == 1)) { 3010 /* Just did an even component, and we didn't 3011 bail.. reset the even_pair_failed flag, 3012 and go on to the next component.... */ 3013 even_pair_failed = 0; 3014 } 3015 } 3016 3017 clabel = cset->ac->clabel; 3018 3019 if (((clabel->parityConfig == '0') && (num_missing > 0)) || 3020 ((clabel->parityConfig == '4') && (num_missing > 1)) || 3021 ((clabel->parityConfig == '5') && (num_missing > 1))) { 3022 /* XXX this needs to be made *much* more general */ 3023 /* Too many failures */ 3024 return(0); 3025 } 3026 /* otherwise, all is well, and we've got enough to take a kick 3027 at autoconfiguring this set */ 3028 return(1); 3029 } 3030 3031 void 3032 rf_create_configuration(RF_AutoConfig_t *ac, RF_Config_t *config, 3033 RF_Raid_t *raidPtr) 3034 { 3035 RF_ComponentLabel_t *clabel; 3036 int i; 3037 3038 clabel = ac->clabel; 3039 3040 /* 1. Fill in the common stuff */ 3041 config->numRow = clabel->num_rows = 1; 3042 config->numCol = clabel->num_columns; 3043 config->numSpare = 0; /* XXX should this be set here? */ 3044 config->sectPerSU = clabel->sectPerSU; 3045 config->SUsPerPU = clabel->SUsPerPU; 3046 config->SUsPerRU = clabel->SUsPerRU; 3047 config->parityConfig = clabel->parityConfig; 3048 /* XXX... */ 3049 strcpy(config->diskQueueType,"fifo"); 3050 config->maxOutstandingDiskReqs = clabel->maxOutstanding; 3051 config->layoutSpecificSize = 0; /* XXX ?? */ 3052 3053 while(ac!=NULL) { 3054 /* row/col values will be in range due to the checks 3055 in reasonable_label() */ 3056 strcpy(config->devnames[0][ac->clabel->column], 3057 ac->devname); 3058 ac = ac->next; 3059 } 3060 3061 for(i=0;i<RF_MAXDBGV;i++) { 3062 config->debugVars[i][0] = 0; 3063 } 3064 } 3065 3066 int 3067 rf_set_autoconfig(RF_Raid_t *raidPtr, int new_value) 3068 { 3069 RF_ComponentLabel_t clabel; 3070 struct vnode *vp; 3071 dev_t dev; 3072 int column; 3073 int sparecol; 3074 3075 raidPtr->autoconfigure = new_value; 3076 3077 for(column=0; column<raidPtr->numCol; column++) { 3078 if (raidPtr->Disks[column].status == rf_ds_optimal) { 3079 dev = raidPtr->Disks[column].dev; 3080 vp = raidPtr->raid_cinfo[column].ci_vp; 3081 raidread_component_label(dev, vp, &clabel); 3082 clabel.autoconfigure = new_value; 3083 raidwrite_component_label(dev, vp, &clabel); 3084 } 3085 } 3086 for(column = 0; column < raidPtr->numSpare ; column++) { 3087 sparecol = raidPtr->numCol + column; 3088 if (raidPtr->Disks[sparecol].status == rf_ds_used_spare) { 3089 dev = raidPtr->Disks[sparecol].dev; 3090 vp = raidPtr->raid_cinfo[sparecol].ci_vp; 3091 raidread_component_label(dev, vp, &clabel); 3092 clabel.autoconfigure = new_value; 3093 raidwrite_component_label(dev, vp, &clabel); 3094 } 3095 } 3096 return(new_value); 3097 } 3098 3099 int 3100 rf_set_rootpartition(RF_Raid_t *raidPtr, int new_value) 3101 { 3102 RF_ComponentLabel_t clabel; 3103 struct vnode *vp; 3104 dev_t dev; 3105 int column; 3106 int sparecol; 3107 3108 raidPtr->root_partition = new_value; 3109 for(column=0; column<raidPtr->numCol; column++) { 3110 if (raidPtr->Disks[column].status == rf_ds_optimal) { 3111 dev = raidPtr->Disks[column].dev; 3112 vp = raidPtr->raid_cinfo[column].ci_vp; 3113 raidread_component_label(dev, vp, &clabel); 3114 clabel.root_partition = new_value; 3115 raidwrite_component_label(dev, vp, &clabel); 3116 } 3117 } 3118 for(column = 0; column < raidPtr->numSpare ; column++) { 3119 sparecol = raidPtr->numCol + column; 3120 if (raidPtr->Disks[sparecol].status == rf_ds_used_spare) { 3121 dev = raidPtr->Disks[sparecol].dev; 3122 vp = raidPtr->raid_cinfo[sparecol].ci_vp; 3123 raidread_component_label(dev, vp, &clabel); 3124 clabel.root_partition = new_value; 3125 raidwrite_component_label(dev, vp, &clabel); 3126 } 3127 } 3128 return(new_value); 3129 } 3130 3131 void 3132 rf_release_all_vps(RF_ConfigSet_t *cset) 3133 { 3134 RF_AutoConfig_t *ac; 3135 3136 ac = cset->ac; 3137 while(ac!=NULL) { 3138 /* Close the vp, and give it back */ 3139 if (ac->vp) { 3140 vn_lock(ac->vp, LK_EXCLUSIVE | LK_RETRY); 3141 VOP_CLOSE(ac->vp, FREAD, NOCRED, 0); 3142 vput(ac->vp); 3143 ac->vp = NULL; 3144 } 3145 ac = ac->next; 3146 } 3147 } 3148 3149 3150 void 3151 rf_cleanup_config_set(RF_ConfigSet_t *cset) 3152 { 3153 RF_AutoConfig_t *ac; 3154 RF_AutoConfig_t *next_ac; 3155 3156 ac = cset->ac; 3157 while(ac!=NULL) { 3158 next_ac = ac->next; 3159 /* nuke the label */ 3160 free(ac->clabel, M_RAIDFRAME); 3161 /* cleanup the config structure */ 3162 free(ac, M_RAIDFRAME); 3163 /* "next.." */ 3164 ac = next_ac; 3165 } 3166 /* and, finally, nuke the config set */ 3167 free(cset, M_RAIDFRAME); 3168 } 3169 3170 3171 void 3172 raid_init_component_label(RF_Raid_t *raidPtr, RF_ComponentLabel_t *clabel) 3173 { 3174 /* current version number */ 3175 clabel->version = RF_COMPONENT_LABEL_VERSION; 3176 clabel->serial_number = raidPtr->serial_number; 3177 clabel->mod_counter = raidPtr->mod_counter; 3178 clabel->num_rows = 1; 3179 clabel->num_columns = raidPtr->numCol; 3180 clabel->clean = RF_RAID_DIRTY; /* not clean */ 3181 clabel->status = rf_ds_optimal; /* "It's good!" */ 3182 3183 clabel->sectPerSU = raidPtr->Layout.sectorsPerStripeUnit; 3184 clabel->SUsPerPU = raidPtr->Layout.SUsPerPU; 3185 clabel->SUsPerRU = raidPtr->Layout.SUsPerRU; 3186 3187 clabel->blockSize = raidPtr->bytesPerSector; 3188 clabel->numBlocks = raidPtr->sectorsPerDisk; 3189 3190 /* XXX not portable */ 3191 clabel->parityConfig = raidPtr->Layout.map->parityConfig; 3192 clabel->maxOutstanding = raidPtr->maxOutstanding; 3193 clabel->autoconfigure = raidPtr->autoconfigure; 3194 clabel->root_partition = raidPtr->root_partition; 3195 clabel->last_unit = raidPtr->raidid; 3196 clabel->config_order = raidPtr->config_order; 3197 } 3198 3199 int 3200 rf_auto_config_set(RF_ConfigSet_t *cset, int *unit) 3201 { 3202 RF_Raid_t *raidPtr; 3203 RF_Config_t *config; 3204 int raidID; 3205 int retcode; 3206 3207 #if DEBUG 3208 printf("RAID autoconfigure\n"); 3209 #endif 3210 3211 retcode = 0; 3212 *unit = -1; 3213 3214 /* 1. Create a config structure */ 3215 3216 config = (RF_Config_t *)malloc(sizeof(RF_Config_t), 3217 M_RAIDFRAME, 3218 M_NOWAIT); 3219 if (config==NULL) { 3220 printf("Out of mem!?!?\n"); 3221 /* XXX do something more intelligent here. */ 3222 return(1); 3223 } 3224 3225 memset(config, 0, sizeof(RF_Config_t)); 3226 3227 /* 3228 2. Figure out what RAID ID this one is supposed to live at 3229 See if we can get the same RAID dev that it was configured 3230 on last time.. 3231 */ 3232 3233 raidID = cset->ac->clabel->last_unit; 3234 if ((raidID < 0) || (raidID >= numraid)) { 3235 /* let's not wander off into lala land. */ 3236 raidID = numraid - 1; 3237 } 3238 if (raidPtrs[raidID]->valid != 0) { 3239 3240 /* 3241 Nope... Go looking for an alternative... 3242 Start high so we don't immediately use raid0 if that's 3243 not taken. 3244 */ 3245 3246 for(raidID = numraid - 1; raidID >= 0; raidID--) { 3247 if (raidPtrs[raidID]->valid == 0) { 3248 /* can use this one! */ 3249 break; 3250 } 3251 } 3252 } 3253 3254 if (raidID < 0) { 3255 /* punt... */ 3256 printf("Unable to auto configure this set!\n"); 3257 printf("(Out of RAID devs!)\n"); 3258 free(config, M_RAIDFRAME); 3259 return(1); 3260 } 3261 3262 #if DEBUG 3263 printf("Configuring raid%d:\n",raidID); 3264 #endif 3265 3266 raidPtr = raidPtrs[raidID]; 3267 3268 /* XXX all this stuff should be done SOMEWHERE ELSE! */ 3269 raidPtr->raidid = raidID; 3270 raidPtr->openings = RAIDOUTSTANDING; 3271 3272 /* 3. Build the configuration structure */ 3273 rf_create_configuration(cset->ac, config, raidPtr); 3274 3275 /* 4. Do the configuration */ 3276 retcode = rf_Configure(raidPtr, config, cset->ac); 3277 3278 if (retcode == 0) { 3279 3280 raidinit(raidPtrs[raidID]); 3281 3282 rf_markalldirty(raidPtrs[raidID]); 3283 raidPtrs[raidID]->autoconfigure = 1; /* XXX do this here? */ 3284 if (cset->ac->clabel->root_partition==1) { 3285 /* everything configured just fine. Make a note 3286 that this set is eligible to be root. */ 3287 cset->rootable = 1; 3288 /* XXX do this here? */ 3289 raidPtrs[raidID]->root_partition = 1; 3290 } 3291 } 3292 3293 /* 5. Cleanup */ 3294 free(config, M_RAIDFRAME); 3295 3296 *unit = raidID; 3297 return(retcode); 3298 } 3299 3300 void 3301 rf_disk_unbusy(RF_RaidAccessDesc_t *desc) 3302 { 3303 struct buf *bp; 3304 3305 bp = (struct buf *)desc->bp; 3306 disk_unbusy(&raid_softc[desc->raidPtr->raidid].sc_dkdev, 3307 (bp->b_bcount - bp->b_resid), (bp->b_flags & B_READ)); 3308 } 3309 3310 void 3311 rf_pool_init(struct pool *p, size_t size, const char *w_chan, 3312 size_t xmin, size_t xmax) 3313 { 3314 pool_init(p, size, 0, 0, 0, w_chan, NULL); 3315 pool_sethiwat(p, xmax); 3316 pool_prime(p, xmin); 3317 pool_setlowat(p, xmin); 3318 } 3319 3320 /* 3321 * rf_buf_queue_check(int raidid) -- looks into the buf_queue to see 3322 * if there is IO pending and if that IO could possibly be done for a 3323 * given RAID set. Returns 0 if IO is waiting and can be done, 1 3324 * otherwise. 3325 * 3326 */ 3327 3328 int 3329 rf_buf_queue_check(int raidid) 3330 { 3331 if ((BUFQ_PEEK(raid_softc[raidid].buf_queue) != NULL) && 3332 raidPtrs[raidid]->openings > 0) { 3333 /* there is work to do */ 3334 return 0; 3335 } 3336 /* default is nothing to do */ 3337 return 1; 3338 } 3339
Indexes created Tue Sep 30 20:09:53 GMT 2025