1 /* $NetBSD: subr_disk.c,v 1.25 1999/02/22 16:00:01 drochner Exp $ */ 2 3 /*- 4 * Copyright (c) 1996, 1997 The NetBSD Foundation, Inc. 5 * All rights reserved. 6 * 7 * This code is derived from software contributed to The NetBSD Foundation 8 * by Jason R. Thorpe of the Numerical Aerospace Simulation Facility, 9 * NASA Ames Research Center. 10 * 11 * Redistribution and use in source and binary forms, with or without 12 * modification, are permitted provided that the following conditions 13 * are met: 14 * 1. Redistributions of source code must retain the above copyright 15 * notice, this list of conditions and the following disclaimer. 16 * 2. Redistributions in binary form must reproduce the above copyright 17 * notice, this list of conditions and the following disclaimer in the 18 * documentation and/or other materials provided with the distribution. 19 * 3. All advertising materials mentioning features or use of this software 20 * must display the following acknowledgement: 21 * This product includes software developed by the NetBSD 22 * Foundation, Inc. and its contributors. 23 * 4. Neither the name of The NetBSD Foundation nor the names of its 24 * contributors may be used to endorse or promote products derived 25 * from this software without specific prior written permission. 26 * 27 * THIS SOFTWARE IS PROVIDED BY THE NETBSD FOUNDATION, INC. AND CONTRIBUTORS 28 * ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED 29 * TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR 30 * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE FOUNDATION OR CONTRIBUTORS 31 * BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR 32 * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF 33 * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS 34 * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN 35 * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) 36 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE 37 * POSSIBILITY OF SUCH DAMAGE. 38 */ 39 40 /* 41 * Copyright (c) 1982, 1986, 1988, 1993 42 * The Regents of the University of California. All rights reserved. 43 * (c) UNIX System Laboratories, Inc. 44 * All or some portions of this file are derived from material licensed 45 * to the University of California by American Telephone and Telegraph 46 * Co. or Unix System Laboratories, Inc. and are reproduced herein with 47 * the permission of UNIX System Laboratories, Inc. 48 * 49 * Redistribution and use in source and binary forms, with or without 50 * modification, are permitted provided that the following conditions 51 * are met: 52 * 1. Redistributions of source code must retain the above copyright 53 * notice, this list of conditions and the following disclaimer. 54 * 2. Redistributions in binary form must reproduce the above copyright 55 * notice, this list of conditions and the following disclaimer in the 56 * documentation and/or other materials provided with the distribution. 57 * 3. All advertising materials mentioning features or use of this software 58 * must display the following acknowledgement: 59 * This product includes software developed by the University of 60 * California, Berkeley and its contributors. 61 * 4. Neither the name of the University nor the names of its contributors 62 * may be used to endorse or promote products derived from this software 63 * without specific prior written permission. 64 * 65 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND 66 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 67 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 68 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE 69 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 70 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 71 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 72 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 73 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 74 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 75 * SUCH DAMAGE. 76 * 77 * @(#)ufs_disksubr.c 8.5 (Berkeley) 1/21/94 78 */ 79 80 #include <sys/param.h> 81 #include <sys/systm.h> 82 #include <sys/kernel.h> 83 #include <sys/malloc.h> 84 #include <sys/buf.h> 85 #include <sys/syslog.h> 86 #include <sys/time.h> 87 #include <sys/disklabel.h> 88 #include <sys/disk.h> 89 90 /* 91 * A global list of all disks attached to the system. May grow or 92 * shrink over time. 93 */ 94 struct disklist_head disklist; /* TAILQ_HEAD */ 95 int disk_count; /* number of drives in global disklist */ 96 97 /* 98 * Seek sort for disks. We depend on the driver which calls us using b_resid 99 * as the current cylinder number. 100 * 101 * The argument ap structure holds a b_actf activity chain pointer on which we 102 * keep two queues, sorted in ascending cylinder order. The first queue holds 103 * those requests which are positioned after the current cylinder (in the first 104 * request); the second holds requests which came in after their cylinder number 105 * was passed. Thus we implement a one way scan, retracting after reaching the 106 * end of the drive to the first request on the second queue, at which time it 107 * becomes the first queue. 108 * 109 * A one-way scan is natural because of the way UNIX read-ahead blocks are 110 * allocated. 111 */ 112 113 void 114 disksort(ap, bp) 115 register struct buf *ap, *bp; 116 { 117 register struct buf *bq; 118 119 /* If the queue is empty, then it's easy. */ 120 if (ap->b_actf == NULL) { 121 bp->b_actf = NULL; 122 ap->b_actf = bp; 123 return; 124 } 125 126 /* 127 * If we lie after the first (currently active) request, then we 128 * must locate the second request list and add ourselves to it. 129 */ 130 bq = ap->b_actf; 131 if (bp->b_cylinder < bq->b_cylinder) { 132 while (bq->b_actf) { 133 /* 134 * Check for an ``inversion'' in the normally ascending 135 * cylinder numbers, indicating the start of the second 136 * request list. 137 */ 138 if (bq->b_actf->b_cylinder < bq->b_cylinder) { 139 /* 140 * Search the second request list for the first 141 * request at a larger cylinder number. We go 142 * before that; if there is no such request, we 143 * go at end. 144 */ 145 do { 146 if (bp->b_cylinder < 147 bq->b_actf->b_cylinder) 148 goto insert; 149 if (bp->b_cylinder == 150 bq->b_actf->b_cylinder && 151 bp->b_blkno < bq->b_actf->b_blkno) 152 goto insert; 153 bq = bq->b_actf; 154 } while (bq->b_actf); 155 goto insert; /* after last */ 156 } 157 bq = bq->b_actf; 158 } 159 /* 160 * No inversions... we will go after the last, and 161 * be the first request in the second request list. 162 */ 163 goto insert; 164 } 165 /* 166 * Request is at/after the current request... 167 * sort in the first request list. 168 */ 169 while (bq->b_actf) { 170 /* 171 * We want to go after the current request if there is an 172 * inversion after it (i.e. it is the end of the first 173 * request list), or if the next request is a larger cylinder 174 * than our request. 175 */ 176 if (bq->b_actf->b_cylinder < bq->b_cylinder || 177 bp->b_cylinder < bq->b_actf->b_cylinder || 178 (bp->b_cylinder == bq->b_actf->b_cylinder && 179 bp->b_blkno < bq->b_actf->b_blkno)) 180 goto insert; 181 bq = bq->b_actf; 182 } 183 /* 184 * Neither a second list nor a larger request... we go at the end of 185 * the first list, which is the same as the end of the whole schebang. 186 */ 187 insert: bp->b_actf = bq->b_actf; 188 bq->b_actf = bp; 189 } 190 191 /* 192 * Compute checksum for disk label. 193 */ 194 u_int 195 dkcksum(lp) 196 register struct disklabel *lp; 197 { 198 register u_short *start, *end; 199 register u_short sum = 0; 200 201 start = (u_short *)lp; 202 end = (u_short *)&lp->d_partitions[lp->d_npartitions]; 203 while (start < end) 204 sum ^= *start++; 205 return (sum); 206 } 207 208 /* 209 * Disk error is the preface to plaintive error messages 210 * about failing disk transfers. It prints messages of the form 211 212 hp0g: hard error reading fsbn 12345 of 12344-12347 (hp0 bn %d cn %d tn %d sn %d) 213 214 * if the offset of the error in the transfer and a disk label 215 * are both available. blkdone should be -1 if the position of the error 216 * is unknown; the disklabel pointer may be null from drivers that have not 217 * been converted to use them. The message is printed with printf 218 * if pri is LOG_PRINTF, otherwise it uses log at the specified priority. 219 * The message should be completed (with at least a newline) with printf 220 * or addlog, respectively. There is no trailing space. 221 */ 222 void 223 diskerr(bp, dname, what, pri, blkdone, lp) 224 register struct buf *bp; 225 char *dname, *what; 226 int pri, blkdone; 227 register struct disklabel *lp; 228 { 229 int unit = DISKUNIT(bp->b_dev), part = DISKPART(bp->b_dev); 230 register void (*pr) __P((const char *, ...)); 231 char partname = 'a' + part; 232 int sn; 233 234 if (pri != LOG_PRINTF) { 235 static const char fmt[] = ""; 236 log(pri, fmt); 237 pr = addlog; 238 } else 239 pr = printf; 240 (*pr)("%s%d%c: %s %sing fsbn ", dname, unit, partname, what, 241 bp->b_flags & B_READ ? "read" : "writ"); 242 sn = bp->b_blkno; 243 if (bp->b_bcount <= DEV_BSIZE) 244 (*pr)("%d", sn); 245 else { 246 if (blkdone >= 0) { 247 sn += blkdone; 248 (*pr)("%d of ", sn); 249 } 250 (*pr)("%d-%d", bp->b_blkno, 251 bp->b_blkno + (bp->b_bcount - 1) / DEV_BSIZE); 252 } 253 if (lp && (blkdone >= 0 || bp->b_bcount <= lp->d_secsize)) { 254 sn += lp->d_partitions[part].p_offset; 255 (*pr)(" (%s%d bn %d; cn %d", dname, unit, sn, 256 sn / lp->d_secpercyl); 257 sn %= lp->d_secpercyl; 258 (*pr)(" tn %d sn %d)", sn / lp->d_nsectors, sn % lp->d_nsectors); 259 } 260 } 261 262 /* 263 * Initialize the disklist. Called by main() before autoconfiguration. 264 */ 265 void 266 disk_init() 267 { 268 269 TAILQ_INIT(&disklist); 270 disk_count = 0; 271 } 272 273 /* 274 * Searches the disklist for the disk corresponding to the 275 * name provided. 276 */ 277 struct disk * 278 disk_find(name) 279 char *name; 280 { 281 struct disk *diskp; 282 283 if ((name == NULL) || (disk_count <= 0)) 284 return (NULL); 285 286 for (diskp = disklist.tqh_first; diskp != NULL; 287 diskp = diskp->dk_link.tqe_next) 288 if (strcmp(diskp->dk_name, name) == 0) 289 return (diskp); 290 291 return (NULL); 292 } 293 294 /* 295 * Attach a disk. 296 */ 297 void 298 disk_attach(diskp) 299 struct disk *diskp; 300 { 301 int s; 302 303 /* 304 * Allocate and initialize the disklabel structures. Note that 305 * it's not safe to sleep here, since we're probably going to be 306 * called during autoconfiguration. 307 */ 308 diskp->dk_label = malloc(sizeof(struct disklabel), M_DEVBUF, M_NOWAIT); 309 diskp->dk_cpulabel = malloc(sizeof(struct cpu_disklabel), M_DEVBUF, 310 M_NOWAIT); 311 if ((diskp->dk_label == NULL) || (diskp->dk_cpulabel == NULL)) 312 panic("disk_attach: can't allocate storage for disklabel"); 313 314 memset(diskp->dk_label, 0, sizeof(struct disklabel)); 315 memset(diskp->dk_cpulabel, 0, sizeof(struct cpu_disklabel)); 316 317 /* 318 * Set the attached timestamp. 319 */ 320 s = splclock(); 321 diskp->dk_attachtime = mono_time; 322 splx(s); 323 324 /* 325 * Link into the disklist. 326 */ 327 TAILQ_INSERT_TAIL(&disklist, diskp, dk_link); 328 ++disk_count; 329 } 330 331 /* 332 * Detach a disk. 333 */ 334 void 335 disk_detach(diskp) 336 struct disk *diskp; 337 { 338 339 /* 340 * Remove from the disklist. 341 */ 342 if (--disk_count < 0) 343 panic("disk_detach: disk_count < 0"); 344 TAILQ_REMOVE(&disklist, diskp, dk_link); 345 346 /* 347 * Free the space used by the disklabel structures. 348 */ 349 free(diskp->dk_label, M_DEVBUF); 350 free(diskp->dk_cpulabel, M_DEVBUF); 351 } 352 353 /* 354 * Increment a disk's busy counter. If the counter is going from 355 * 0 to 1, set the timestamp. 356 */ 357 void 358 disk_busy(diskp) 359 struct disk *diskp; 360 { 361 int s; 362 363 /* 364 * XXX We'd like to use something as accurate as microtime(), 365 * but that doesn't depend on the system TOD clock. 366 */ 367 if (diskp->dk_busy++ == 0) { 368 s = splclock(); 369 diskp->dk_timestamp = mono_time; 370 splx(s); 371 } 372 } 373 374 /* 375 * Decrement a disk's busy counter, increment the byte count, total busy 376 * time, and reset the timestamp. 377 */ 378 void 379 disk_unbusy(diskp, bcount) 380 struct disk *diskp; 381 long bcount; 382 { 383 int s; 384 struct timeval dv_time, diff_time; 385 386 if (diskp->dk_busy-- == 0) { 387 printf("%s: dk_busy < 0\n", diskp->dk_name); 388 panic("disk_unbusy"); 389 } 390 391 s = splclock(); 392 dv_time = mono_time; 393 splx(s); 394 395 timersub(&dv_time, &diskp->dk_timestamp, &diff_time); 396 timeradd(&diskp->dk_time, &diff_time, &diskp->dk_time); 397 398 diskp->dk_timestamp = dv_time; 399 if (bcount > 0) { 400 diskp->dk_bytes += bcount; 401 diskp->dk_xfer++; 402 } 403 } 404 405 /* 406 * Reset the metrics counters on the given disk. Note that we cannot 407 * reset the busy counter, as it may case a panic in disk_unbusy(). 408 * We also must avoid playing with the timestamp information, as it 409 * may skew any pending transfer results. 410 */ 411 void 412 disk_resetstat(diskp) 413 struct disk *diskp; 414 { 415 int s = splbio(), t; 416 417 diskp->dk_xfer = 0; 418 diskp->dk_bytes = 0; 419 420 t = splclock(); 421 diskp->dk_attachtime = mono_time; 422 splx(t); 423 424 timerclear(&diskp->dk_time); 425 426 splx(s); 427 } 428