1 /* 2 * XenBSD block device driver 3 * 4 * Copyright (c) 2010-2013 Spectra Logic Corporation 5 * Copyright (c) 2009 Scott Long, Yahoo! 6 * Copyright (c) 2009 Frank Suchomel, Citrix 7 * Copyright (c) 2009 Doug F. Rabson, Citrix 8 * Copyright (c) 2005 Kip Macy 9 * Copyright (c) 2003-2004, Keir Fraser & Steve Hand 10 * Modifications by Mark A. Williamson are (c) Intel Research Cambridge 11 * 12 * 13 * Permission is hereby granted, free of charge, to any person obtaining a copy 14 * of this software and associated documentation files (the "Software"), to 15 * deal in the Software without restriction, including without limitation the 16 * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or 17 * sell copies of the Software, and to permit persons to whom the Software is 18 * furnished to do so, subject to the following conditions: 19 * 20 * The above copyright notice and this permission notice shall be included in 21 * all copies or substantial portions of the Software. 22 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR 23 * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, 24 * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE 25 * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER 26 * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING 27 * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER 28 * DEALINGS IN THE SOFTWARE. 29 */ 30 31 #include <sys/cdefs.h> 32 __FBSDID("$FreeBSD$"); 33 34 #include <sys/param.h> 35 #include <sys/systm.h> 36 #include <sys/malloc.h> 37 #include <sys/kernel.h> 38 #include <vm/vm.h> 39 #include <vm/pmap.h> 40 41 #include <sys/bio.h> 42 #include <sys/bus.h> 43 #include <sys/conf.h> 44 #include <sys/module.h> 45 #include <sys/sysctl.h> 46 47 #include <machine/bus.h> 48 #include <sys/rman.h> 49 #include <machine/resource.h> 50 #include <machine/intr_machdep.h> 51 #include <machine/vmparam.h> 52 53 #include <xen/xen-os.h> 54 #include <xen/hypervisor.h> 55 #include <xen/xen_intr.h> 56 #include <xen/gnttab.h> 57 #include <xen/interface/grant_table.h> 58 #include <xen/interface/io/protocols.h> 59 #include <xen/xenbus/xenbusvar.h> 60 61 #include <machine/_inttypes.h> 62 63 #include <geom/geom_disk.h> 64 65 #include <dev/xen/blkfront/block.h> 66 67 #include "xenbus_if.h" 68 69 /*--------------------------- Forward Declarations ---------------------------*/ 70 static void xbd_closing(device_t); 71 static void xbd_startio(struct xbd_softc *sc); 72 73 /*---------------------------------- Macros ----------------------------------*/ 74 #if 0 75 #define DPRINTK(fmt, args...) printf("[XEN] %s:%d: " fmt ".\n", __func__, __LINE__, ##args) 76 #else 77 #define DPRINTK(fmt, args...) 78 #endif 79 80 #define XBD_SECTOR_SHFT 9 81 82 /*---------------------------- Global Static Data ----------------------------*/ 83 static MALLOC_DEFINE(M_XENBLOCKFRONT, "xbd", "Xen Block Front driver data"); 84 85 static int xbd_enable_indirect = 1; 86 SYSCTL_NODE(_hw, OID_AUTO, xbd, CTLFLAG_RD | CTLFLAG_MPSAFE, 0, 87 "xbd driver parameters"); 88 SYSCTL_INT(_hw_xbd, OID_AUTO, xbd_enable_indirect, CTLFLAG_RDTUN, 89 &xbd_enable_indirect, 0, "Enable xbd indirect segments"); 90 91 /*---------------------------- Command Processing ----------------------------*/ 92 static void 93 xbd_freeze(struct xbd_softc *sc, xbd_flag_t xbd_flag) 94 { 95 if (xbd_flag != XBDF_NONE && (sc->xbd_flags & xbd_flag) != 0) 96 return; 97 98 sc->xbd_flags |= xbd_flag; 99 sc->xbd_qfrozen_cnt++; 100 } 101 102 static void 103 xbd_thaw(struct xbd_softc *sc, xbd_flag_t xbd_flag) 104 { 105 if (xbd_flag != XBDF_NONE && (sc->xbd_flags & xbd_flag) == 0) 106 return; 107 108 if (sc->xbd_qfrozen_cnt == 0) 109 panic("%s: Thaw with flag 0x%x while not frozen.", 110 __func__, xbd_flag); 111 112 sc->xbd_flags &= ~xbd_flag; 113 sc->xbd_qfrozen_cnt--; 114 } 115 116 static void 117 xbd_cm_freeze(struct xbd_softc *sc, struct xbd_command *cm, xbdc_flag_t cm_flag) 118 { 119 if ((cm->cm_flags & XBDCF_FROZEN) != 0) 120 return; 121 122 cm->cm_flags |= XBDCF_FROZEN|cm_flag; 123 xbd_freeze(sc, XBDF_NONE); 124 } 125 126 static void 127 xbd_cm_thaw(struct xbd_softc *sc, struct xbd_command *cm) 128 { 129 if ((cm->cm_flags & XBDCF_FROZEN) == 0) 130 return; 131 132 cm->cm_flags &= ~XBDCF_FROZEN; 133 xbd_thaw(sc, XBDF_NONE); 134 } 135 136 static inline void 137 xbd_flush_requests(struct xbd_softc *sc) 138 { 139 int notify; 140 141 RING_PUSH_REQUESTS_AND_CHECK_NOTIFY(&sc->xbd_ring, notify); 142 143 if (notify) 144 xen_intr_signal(sc->xen_intr_handle); 145 } 146 147 static void 148 xbd_free_command(struct xbd_command *cm) 149 { 150 151 KASSERT((cm->cm_flags & XBDCF_Q_MASK) == XBD_Q_NONE, 152 ("Freeing command that is still on queue %d.", 153 cm->cm_flags & XBDCF_Q_MASK)); 154 155 cm->cm_flags = XBDCF_INITIALIZER; 156 cm->cm_bp = NULL; 157 cm->cm_complete = NULL; 158 xbd_enqueue_cm(cm, XBD_Q_FREE); 159 xbd_thaw(cm->cm_sc, XBDF_CM_SHORTAGE); 160 } 161 162 static void 163 xbd_mksegarray(bus_dma_segment_t *segs, int nsegs, 164 grant_ref_t * gref_head, int otherend_id, int readonly, 165 grant_ref_t * sg_ref, struct blkif_request_segment *sg) 166 { 167 struct blkif_request_segment *last_block_sg = sg + nsegs; 168 vm_paddr_t buffer_ma; 169 uint64_t fsect, lsect; 170 int ref; 171 172 while (sg < last_block_sg) { 173 KASSERT(segs->ds_addr % (1 << XBD_SECTOR_SHFT) == 0, 174 ("XEN disk driver I/O must be sector aligned")); 175 KASSERT(segs->ds_len % (1 << XBD_SECTOR_SHFT) == 0, 176 ("XEN disk driver I/Os must be a multiple of " 177 "the sector length")); 178 buffer_ma = segs->ds_addr; 179 fsect = (buffer_ma & PAGE_MASK) >> XBD_SECTOR_SHFT; 180 lsect = fsect + (segs->ds_len >> XBD_SECTOR_SHFT) - 1; 181 182 KASSERT(lsect <= 7, ("XEN disk driver data cannot " 183 "cross a page boundary")); 184 185 /* install a grant reference. */ 186 ref = gnttab_claim_grant_reference(gref_head); 187 188 /* 189 * GNTTAB_LIST_END == 0xffffffff, but it is private 190 * to gnttab.c. 191 */ 192 KASSERT(ref != ~0, ("grant_reference failed")); 193 194 gnttab_grant_foreign_access_ref( 195 ref, 196 otherend_id, 197 buffer_ma >> PAGE_SHIFT, 198 readonly); 199 200 *sg_ref = ref; 201 *sg = (struct blkif_request_segment) { 202 .gref = ref, 203 .first_sect = fsect, 204 .last_sect = lsect 205 }; 206 sg++; 207 sg_ref++; 208 segs++; 209 } 210 } 211 212 static void 213 xbd_queue_cb(void *arg, bus_dma_segment_t *segs, int nsegs, int error) 214 { 215 struct xbd_softc *sc; 216 struct xbd_command *cm; 217 int op; 218 219 cm = arg; 220 sc = cm->cm_sc; 221 222 if (error) { 223 cm->cm_bp->bio_error = EIO; 224 biodone(cm->cm_bp); 225 xbd_free_command(cm); 226 return; 227 } 228 229 KASSERT(nsegs <= sc->xbd_max_request_segments, 230 ("Too many segments in a blkfront I/O")); 231 232 if (nsegs <= BLKIF_MAX_SEGMENTS_PER_REQUEST) { 233 blkif_request_t *ring_req; 234 235 /* Fill out a blkif_request_t structure. */ 236 ring_req = (blkif_request_t *) 237 RING_GET_REQUEST(&sc->xbd_ring, sc->xbd_ring.req_prod_pvt); 238 sc->xbd_ring.req_prod_pvt++; 239 ring_req->id = cm->cm_id; 240 ring_req->operation = cm->cm_operation; 241 ring_req->sector_number = cm->cm_sector_number; 242 ring_req->handle = (blkif_vdev_t)(uintptr_t)sc->xbd_disk; 243 ring_req->nr_segments = nsegs; 244 cm->cm_nseg = nsegs; 245 xbd_mksegarray(segs, nsegs, &cm->cm_gref_head, 246 xenbus_get_otherend_id(sc->xbd_dev), 247 cm->cm_operation == BLKIF_OP_WRITE, 248 cm->cm_sg_refs, ring_req->seg); 249 } else { 250 blkif_request_indirect_t *ring_req; 251 252 /* Fill out a blkif_request_indirect_t structure. */ 253 ring_req = (blkif_request_indirect_t *) 254 RING_GET_REQUEST(&sc->xbd_ring, sc->xbd_ring.req_prod_pvt); 255 sc->xbd_ring.req_prod_pvt++; 256 ring_req->id = cm->cm_id; 257 ring_req->operation = BLKIF_OP_INDIRECT; 258 ring_req->indirect_op = cm->cm_operation; 259 ring_req->sector_number = cm->cm_sector_number; 260 ring_req->handle = (blkif_vdev_t)(uintptr_t)sc->xbd_disk; 261 ring_req->nr_segments = nsegs; 262 cm->cm_nseg = nsegs; 263 xbd_mksegarray(segs, nsegs, &cm->cm_gref_head, 264 xenbus_get_otherend_id(sc->xbd_dev), 265 cm->cm_operation == BLKIF_OP_WRITE, 266 cm->cm_sg_refs, cm->cm_indirectionpages); 267 memcpy(ring_req->indirect_grefs, &cm->cm_indirectionrefs, 268 sizeof(grant_ref_t) * sc->xbd_max_request_indirectpages); 269 } 270 271 if (cm->cm_operation == BLKIF_OP_READ) 272 op = BUS_DMASYNC_PREREAD; 273 else if (cm->cm_operation == BLKIF_OP_WRITE) 274 op = BUS_DMASYNC_PREWRITE; 275 else 276 op = 0; 277 bus_dmamap_sync(sc->xbd_io_dmat, cm->cm_map, op); 278 279 gnttab_free_grant_references(cm->cm_gref_head); 280 281 xbd_enqueue_cm(cm, XBD_Q_BUSY); 282 283 /* 284 * If bus dma had to asynchronously call us back to dispatch 285 * this command, we are no longer executing in the context of 286 * xbd_startio(). Thus we cannot rely on xbd_startio()'s call to 287 * xbd_flush_requests() to publish this command to the backend 288 * along with any other commands that it could batch. 289 */ 290 if ((cm->cm_flags & XBDCF_ASYNC_MAPPING) != 0) 291 xbd_flush_requests(sc); 292 293 return; 294 } 295 296 static int 297 xbd_queue_request(struct xbd_softc *sc, struct xbd_command *cm) 298 { 299 int error; 300 301 if (cm->cm_bp != NULL) 302 error = bus_dmamap_load_bio(sc->xbd_io_dmat, cm->cm_map, 303 cm->cm_bp, xbd_queue_cb, cm, 0); 304 else 305 error = bus_dmamap_load(sc->xbd_io_dmat, cm->cm_map, 306 cm->cm_data, cm->cm_datalen, xbd_queue_cb, cm, 0); 307 if (error == EINPROGRESS) { 308 /* 309 * Maintain queuing order by freezing the queue. The next 310 * command may not require as many resources as the command 311 * we just attempted to map, so we can't rely on bus dma 312 * blocking for it too. 313 */ 314 xbd_cm_freeze(sc, cm, XBDCF_ASYNC_MAPPING); 315 return (0); 316 } 317 318 return (error); 319 } 320 321 static void 322 xbd_restart_queue_callback(void *arg) 323 { 324 struct xbd_softc *sc = arg; 325 326 mtx_lock(&sc->xbd_io_lock); 327 328 xbd_thaw(sc, XBDF_GNT_SHORTAGE); 329 330 xbd_startio(sc); 331 332 mtx_unlock(&sc->xbd_io_lock); 333 } 334 335 static struct xbd_command * 336 xbd_bio_command(struct xbd_softc *sc) 337 { 338 struct xbd_command *cm; 339 struct bio *bp; 340 341 if (__predict_false(sc->xbd_state != XBD_STATE_CONNECTED)) 342 return (NULL); 343 344 bp = xbd_dequeue_bio(sc); 345 if (bp == NULL) 346 return (NULL); 347 348 if ((cm = xbd_dequeue_cm(sc, XBD_Q_FREE)) == NULL) { 349 xbd_freeze(sc, XBDF_CM_SHORTAGE); 350 xbd_requeue_bio(sc, bp); 351 return (NULL); 352 } 353 354 if (gnttab_alloc_grant_references(sc->xbd_max_request_segments, 355 &cm->cm_gref_head) != 0) { 356 gnttab_request_free_callback(&sc->xbd_callback, 357 xbd_restart_queue_callback, sc, 358 sc->xbd_max_request_segments); 359 xbd_freeze(sc, XBDF_GNT_SHORTAGE); 360 xbd_requeue_bio(sc, bp); 361 xbd_enqueue_cm(cm, XBD_Q_FREE); 362 return (NULL); 363 } 364 365 cm->cm_bp = bp; 366 cm->cm_sector_number = (blkif_sector_t)bp->bio_pblkno; 367 368 switch (bp->bio_cmd) { 369 case BIO_READ: 370 cm->cm_operation = BLKIF_OP_READ; 371 break; 372 case BIO_WRITE: 373 cm->cm_operation = BLKIF_OP_WRITE; 374 if ((bp->bio_flags & BIO_ORDERED) != 0) { 375 if ((sc->xbd_flags & XBDF_BARRIER) != 0) { 376 cm->cm_operation = BLKIF_OP_WRITE_BARRIER; 377 } else { 378 /* 379 * Single step this command. 380 */ 381 cm->cm_flags |= XBDCF_Q_FREEZE; 382 if (xbd_queue_length(sc, XBD_Q_BUSY) != 0) { 383 /* 384 * Wait for in-flight requests to 385 * finish. 386 */ 387 xbd_freeze(sc, XBDF_WAIT_IDLE); 388 xbd_requeue_cm(cm, XBD_Q_READY); 389 return (NULL); 390 } 391 } 392 } 393 break; 394 case BIO_FLUSH: 395 if ((sc->xbd_flags & XBDF_FLUSH) != 0) 396 cm->cm_operation = BLKIF_OP_FLUSH_DISKCACHE; 397 else if ((sc->xbd_flags & XBDF_BARRIER) != 0) 398 cm->cm_operation = BLKIF_OP_WRITE_BARRIER; 399 else 400 panic("flush request, but no flush support available"); 401 break; 402 default: 403 biofinish(bp, NULL, EOPNOTSUPP); 404 xbd_enqueue_cm(cm, XBD_Q_FREE); 405 return (NULL); 406 } 407 408 return (cm); 409 } 410 411 /* 412 * Dequeue buffers and place them in the shared communication ring. 413 * Return when no more requests can be accepted or all buffers have 414 * been queued. 415 * 416 * Signal XEN once the ring has been filled out. 417 */ 418 static void 419 xbd_startio(struct xbd_softc *sc) 420 { 421 struct xbd_command *cm; 422 int error, queued = 0; 423 424 mtx_assert(&sc->xbd_io_lock, MA_OWNED); 425 426 if (sc->xbd_state != XBD_STATE_CONNECTED) 427 return; 428 429 while (!RING_FULL(&sc->xbd_ring)) { 430 431 if (sc->xbd_qfrozen_cnt != 0) 432 break; 433 434 cm = xbd_dequeue_cm(sc, XBD_Q_READY); 435 436 if (cm == NULL) 437 cm = xbd_bio_command(sc); 438 439 if (cm == NULL) 440 break; 441 442 if ((cm->cm_flags & XBDCF_Q_FREEZE) != 0) { 443 /* 444 * Single step command. Future work is 445 * held off until this command completes. 446 */ 447 xbd_cm_freeze(sc, cm, XBDCF_Q_FREEZE); 448 } 449 450 if ((error = xbd_queue_request(sc, cm)) != 0) { 451 printf("xbd_queue_request returned %d\n", error); 452 break; 453 } 454 queued++; 455 } 456 457 if (queued != 0) 458 xbd_flush_requests(sc); 459 } 460 461 static void 462 xbd_bio_complete(struct xbd_softc *sc, struct xbd_command *cm) 463 { 464 struct bio *bp; 465 466 bp = cm->cm_bp; 467 468 if (__predict_false(cm->cm_status != BLKIF_RSP_OKAY)) { 469 disk_err(bp, "disk error" , -1, 0); 470 printf(" status: %x\n", cm->cm_status); 471 bp->bio_flags |= BIO_ERROR; 472 } 473 474 if (bp->bio_flags & BIO_ERROR) 475 bp->bio_error = EIO; 476 else 477 bp->bio_resid = 0; 478 479 xbd_free_command(cm); 480 biodone(bp); 481 } 482 483 static void 484 xbd_int(void *xsc) 485 { 486 struct xbd_softc *sc = xsc; 487 struct xbd_command *cm; 488 blkif_response_t *bret; 489 RING_IDX i, rp; 490 int op; 491 492 mtx_lock(&sc->xbd_io_lock); 493 494 if (__predict_false(sc->xbd_state == XBD_STATE_DISCONNECTED)) { 495 mtx_unlock(&sc->xbd_io_lock); 496 return; 497 } 498 499 again: 500 rp = sc->xbd_ring.sring->rsp_prod; 501 rmb(); /* Ensure we see queued responses up to 'rp'. */ 502 503 for (i = sc->xbd_ring.rsp_cons; i != rp;) { 504 bret = RING_GET_RESPONSE(&sc->xbd_ring, i); 505 cm = &sc->xbd_shadow[bret->id]; 506 507 xbd_remove_cm(cm, XBD_Q_BUSY); 508 gnttab_end_foreign_access_references(cm->cm_nseg, 509 cm->cm_sg_refs); 510 i++; 511 512 if (cm->cm_operation == BLKIF_OP_READ) 513 op = BUS_DMASYNC_POSTREAD; 514 else if (cm->cm_operation == BLKIF_OP_WRITE || 515 cm->cm_operation == BLKIF_OP_WRITE_BARRIER) 516 op = BUS_DMASYNC_POSTWRITE; 517 else 518 op = 0; 519 bus_dmamap_sync(sc->xbd_io_dmat, cm->cm_map, op); 520 bus_dmamap_unload(sc->xbd_io_dmat, cm->cm_map); 521 522 /* 523 * Release any hold this command has on future command 524 * dispatch. 525 */ 526 xbd_cm_thaw(sc, cm); 527 528 /* 529 * Directly call the i/o complete routine to save an 530 * an indirection in the common case. 531 */ 532 cm->cm_status = bret->status; 533 if (cm->cm_bp) 534 xbd_bio_complete(sc, cm); 535 else if (cm->cm_complete != NULL) 536 cm->cm_complete(cm); 537 else 538 xbd_free_command(cm); 539 } 540 541 sc->xbd_ring.rsp_cons = i; 542 543 if (i != sc->xbd_ring.req_prod_pvt) { 544 int more_to_do; 545 RING_FINAL_CHECK_FOR_RESPONSES(&sc->xbd_ring, more_to_do); 546 if (more_to_do) 547 goto again; 548 } else { 549 sc->xbd_ring.sring->rsp_event = i + 1; 550 } 551 552 if (xbd_queue_length(sc, XBD_Q_BUSY) == 0) 553 xbd_thaw(sc, XBDF_WAIT_IDLE); 554 555 xbd_startio(sc); 556 557 if (__predict_false(sc->xbd_state == XBD_STATE_SUSPENDED)) 558 wakeup(&sc->xbd_cm_q[XBD_Q_BUSY]); 559 560 mtx_unlock(&sc->xbd_io_lock); 561 } 562 563 /*------------------------------- Dump Support -------------------------------*/ 564 /** 565 * Quiesce the disk writes for a dump file before allowing the next buffer. 566 */ 567 static void 568 xbd_quiesce(struct xbd_softc *sc) 569 { 570 int mtd; 571 572 // While there are outstanding requests 573 while (xbd_queue_length(sc, XBD_Q_BUSY) != 0) { 574 RING_FINAL_CHECK_FOR_RESPONSES(&sc->xbd_ring, mtd); 575 if (mtd) { 576 /* Received request completions, update queue. */ 577 xbd_int(sc); 578 } 579 if (xbd_queue_length(sc, XBD_Q_BUSY) != 0) { 580 /* 581 * Still pending requests, wait for the disk i/o 582 * to complete. 583 */ 584 HYPERVISOR_yield(); 585 } 586 } 587 } 588 589 /* Kernel dump function for a paravirtualized disk device */ 590 static void 591 xbd_dump_complete(struct xbd_command *cm) 592 { 593 594 xbd_enqueue_cm(cm, XBD_Q_COMPLETE); 595 } 596 597 static int 598 xbd_dump(void *arg, void *virtual, vm_offset_t physical, off_t offset, 599 size_t length) 600 { 601 struct disk *dp = arg; 602 struct xbd_softc *sc = dp->d_drv1; 603 struct xbd_command *cm; 604 size_t chunk; 605 int sbp; 606 int rc = 0; 607 608 if (length == 0) 609 return (0); 610 611 xbd_quiesce(sc); /* All quiet on the western front. */ 612 613 /* 614 * If this lock is held, then this module is failing, and a 615 * successful kernel dump is highly unlikely anyway. 616 */ 617 mtx_lock(&sc->xbd_io_lock); 618 619 /* Split the 64KB block as needed */ 620 for (sbp=0; length > 0; sbp++) { 621 cm = xbd_dequeue_cm(sc, XBD_Q_FREE); 622 if (cm == NULL) { 623 mtx_unlock(&sc->xbd_io_lock); 624 device_printf(sc->xbd_dev, "dump: no more commands?\n"); 625 return (EBUSY); 626 } 627 628 if (gnttab_alloc_grant_references(sc->xbd_max_request_segments, 629 &cm->cm_gref_head) != 0) { 630 xbd_free_command(cm); 631 mtx_unlock(&sc->xbd_io_lock); 632 device_printf(sc->xbd_dev, "no more grant allocs?\n"); 633 return (EBUSY); 634 } 635 636 chunk = length > sc->xbd_max_request_size ? 637 sc->xbd_max_request_size : length; 638 cm->cm_data = virtual; 639 cm->cm_datalen = chunk; 640 cm->cm_operation = BLKIF_OP_WRITE; 641 cm->cm_sector_number = offset / dp->d_sectorsize; 642 cm->cm_complete = xbd_dump_complete; 643 644 xbd_enqueue_cm(cm, XBD_Q_READY); 645 646 length -= chunk; 647 offset += chunk; 648 virtual = (char *) virtual + chunk; 649 } 650 651 /* Tell DOM0 to do the I/O */ 652 xbd_startio(sc); 653 mtx_unlock(&sc->xbd_io_lock); 654 655 /* Poll for the completion. */ 656 xbd_quiesce(sc); /* All quite on the eastern front */ 657 658 /* If there were any errors, bail out... */ 659 while ((cm = xbd_dequeue_cm(sc, XBD_Q_COMPLETE)) != NULL) { 660 if (cm->cm_status != BLKIF_RSP_OKAY) { 661 device_printf(sc->xbd_dev, 662 "Dump I/O failed at sector %jd\n", 663 cm->cm_sector_number); 664 rc = EIO; 665 } 666 xbd_free_command(cm); 667 } 668 669 return (rc); 670 } 671 672 /*----------------------------- Disk Entrypoints -----------------------------*/ 673 static int 674 xbd_open(struct disk *dp) 675 { 676 struct xbd_softc *sc = dp->d_drv1; 677 678 if (sc == NULL) { 679 printf("xbd%d: not found", dp->d_unit); 680 return (ENXIO); 681 } 682 683 sc->xbd_flags |= XBDF_OPEN; 684 sc->xbd_users++; 685 return (0); 686 } 687 688 static int 689 xbd_close(struct disk *dp) 690 { 691 struct xbd_softc *sc = dp->d_drv1; 692 693 if (sc == NULL) 694 return (ENXIO); 695 sc->xbd_flags &= ~XBDF_OPEN; 696 if (--(sc->xbd_users) == 0) { 697 /* 698 * Check whether we have been instructed to close. We will 699 * have ignored this request initially, as the device was 700 * still mounted. 701 */ 702 if (xenbus_get_otherend_state(sc->xbd_dev) == 703 XenbusStateClosing) 704 xbd_closing(sc->xbd_dev); 705 } 706 return (0); 707 } 708 709 static int 710 xbd_ioctl(struct disk *dp, u_long cmd, void *addr, int flag, struct thread *td) 711 { 712 struct xbd_softc *sc = dp->d_drv1; 713 714 if (sc == NULL) 715 return (ENXIO); 716 717 return (ENOTTY); 718 } 719 720 /* 721 * Read/write routine for a buffer. Finds the proper unit, place it on 722 * the sortq and kick the controller. 723 */ 724 static void 725 xbd_strategy(struct bio *bp) 726 { 727 struct xbd_softc *sc = bp->bio_disk->d_drv1; 728 729 /* bogus disk? */ 730 if (sc == NULL) { 731 bp->bio_error = EINVAL; 732 bp->bio_flags |= BIO_ERROR; 733 bp->bio_resid = bp->bio_bcount; 734 biodone(bp); 735 return; 736 } 737 738 /* 739 * Place it in the queue of disk activities for this disk 740 */ 741 mtx_lock(&sc->xbd_io_lock); 742 743 xbd_enqueue_bio(sc, bp); 744 xbd_startio(sc); 745 746 mtx_unlock(&sc->xbd_io_lock); 747 return; 748 } 749 750 /*------------------------------ Ring Management -----------------------------*/ 751 static int 752 xbd_alloc_ring(struct xbd_softc *sc) 753 { 754 blkif_sring_t *sring; 755 uintptr_t sring_page_addr; 756 int error; 757 int i; 758 759 sring = malloc(sc->xbd_ring_pages * PAGE_SIZE, M_XENBLOCKFRONT, 760 M_NOWAIT|M_ZERO); 761 if (sring == NULL) { 762 xenbus_dev_fatal(sc->xbd_dev, ENOMEM, "allocating shared ring"); 763 return (ENOMEM); 764 } 765 SHARED_RING_INIT(sring); 766 FRONT_RING_INIT(&sc->xbd_ring, sring, sc->xbd_ring_pages * PAGE_SIZE); 767 768 for (i = 0, sring_page_addr = (uintptr_t)sring; 769 i < sc->xbd_ring_pages; 770 i++, sring_page_addr += PAGE_SIZE) { 771 772 error = xenbus_grant_ring(sc->xbd_dev, 773 (vtophys(sring_page_addr) >> PAGE_SHIFT), 774 &sc->xbd_ring_ref[i]); 775 if (error) { 776 xenbus_dev_fatal(sc->xbd_dev, error, 777 "granting ring_ref(%d)", i); 778 return (error); 779 } 780 } 781 if (sc->xbd_ring_pages == 1) { 782 error = xs_printf(XST_NIL, xenbus_get_node(sc->xbd_dev), 783 "ring-ref", "%u", sc->xbd_ring_ref[0]); 784 if (error) { 785 xenbus_dev_fatal(sc->xbd_dev, error, 786 "writing %s/ring-ref", 787 xenbus_get_node(sc->xbd_dev)); 788 return (error); 789 } 790 } else { 791 for (i = 0; i < sc->xbd_ring_pages; i++) { 792 char ring_ref_name[]= "ring_refXX"; 793 794 snprintf(ring_ref_name, sizeof(ring_ref_name), 795 "ring-ref%u", i); 796 error = xs_printf(XST_NIL, xenbus_get_node(sc->xbd_dev), 797 ring_ref_name, "%u", sc->xbd_ring_ref[i]); 798 if (error) { 799 xenbus_dev_fatal(sc->xbd_dev, error, 800 "writing %s/%s", 801 xenbus_get_node(sc->xbd_dev), 802 ring_ref_name); 803 return (error); 804 } 805 } 806 } 807 808 error = xen_intr_alloc_and_bind_local_port(sc->xbd_dev, 809 xenbus_get_otherend_id(sc->xbd_dev), NULL, xbd_int, sc, 810 INTR_TYPE_BIO | INTR_MPSAFE, &sc->xen_intr_handle); 811 if (error) { 812 xenbus_dev_fatal(sc->xbd_dev, error, 813 "xen_intr_alloc_and_bind_local_port failed"); 814 return (error); 815 } 816 817 return (0); 818 } 819 820 static void 821 xbd_free_ring(struct xbd_softc *sc) 822 { 823 int i; 824 825 if (sc->xbd_ring.sring == NULL) 826 return; 827 828 for (i = 0; i < sc->xbd_ring_pages; i++) { 829 if (sc->xbd_ring_ref[i] != GRANT_REF_INVALID) { 830 gnttab_end_foreign_access_ref(sc->xbd_ring_ref[i]); 831 sc->xbd_ring_ref[i] = GRANT_REF_INVALID; 832 } 833 } 834 free(sc->xbd_ring.sring, M_XENBLOCKFRONT); 835 sc->xbd_ring.sring = NULL; 836 } 837 838 /*-------------------------- Initialization/Teardown -------------------------*/ 839 static int 840 xbd_feature_string(struct xbd_softc *sc, char *features, size_t len) 841 { 842 struct sbuf sb; 843 int feature_cnt; 844 845 sbuf_new(&sb, features, len, SBUF_FIXEDLEN); 846 847 feature_cnt = 0; 848 if ((sc->xbd_flags & XBDF_FLUSH) != 0) { 849 sbuf_printf(&sb, "flush"); 850 feature_cnt++; 851 } 852 853 if ((sc->xbd_flags & XBDF_BARRIER) != 0) { 854 if (feature_cnt != 0) 855 sbuf_printf(&sb, ", "); 856 sbuf_printf(&sb, "write_barrier"); 857 feature_cnt++; 858 } 859 860 if ((sc->xbd_flags & XBDF_DISCARD) != 0) { 861 if (feature_cnt != 0) 862 sbuf_printf(&sb, ", "); 863 sbuf_printf(&sb, "discard"); 864 feature_cnt++; 865 } 866 867 if ((sc->xbd_flags & XBDF_PERSISTENT) != 0) { 868 if (feature_cnt != 0) 869 sbuf_printf(&sb, ", "); 870 sbuf_printf(&sb, "persistent_grants"); 871 feature_cnt++; 872 } 873 874 (void) sbuf_finish(&sb); 875 return (sbuf_len(&sb)); 876 } 877 878 static int 879 xbd_sysctl_features(SYSCTL_HANDLER_ARGS) 880 { 881 char features[80]; 882 struct xbd_softc *sc = arg1; 883 int error; 884 int len; 885 886 error = sysctl_wire_old_buffer(req, 0); 887 if (error != 0) 888 return (error); 889 890 len = xbd_feature_string(sc, features, sizeof(features)); 891 892 /* len is -1 on error, which will make the SYSCTL_OUT a no-op. */ 893 return (SYSCTL_OUT(req, features, len + 1/*NUL*/)); 894 } 895 896 static void 897 xbd_setup_sysctl(struct xbd_softc *xbd) 898 { 899 struct sysctl_ctx_list *sysctl_ctx = NULL; 900 struct sysctl_oid *sysctl_tree = NULL; 901 struct sysctl_oid_list *children; 902 903 sysctl_ctx = device_get_sysctl_ctx(xbd->xbd_dev); 904 if (sysctl_ctx == NULL) 905 return; 906 907 sysctl_tree = device_get_sysctl_tree(xbd->xbd_dev); 908 if (sysctl_tree == NULL) 909 return; 910 911 children = SYSCTL_CHILDREN(sysctl_tree); 912 SYSCTL_ADD_UINT(sysctl_ctx, children, OID_AUTO, 913 "max_requests", CTLFLAG_RD, &xbd->xbd_max_requests, -1, 914 "maximum outstanding requests (negotiated)"); 915 916 SYSCTL_ADD_UINT(sysctl_ctx, children, OID_AUTO, 917 "max_request_segments", CTLFLAG_RD, 918 &xbd->xbd_max_request_segments, 0, 919 "maximum number of pages per requests (negotiated)"); 920 921 SYSCTL_ADD_UINT(sysctl_ctx, children, OID_AUTO, 922 "max_request_size", CTLFLAG_RD, &xbd->xbd_max_request_size, 0, 923 "maximum size in bytes of a request (negotiated)"); 924 925 SYSCTL_ADD_UINT(sysctl_ctx, children, OID_AUTO, 926 "ring_pages", CTLFLAG_RD, &xbd->xbd_ring_pages, 0, 927 "communication channel pages (negotiated)"); 928 929 SYSCTL_ADD_PROC(sysctl_ctx, children, OID_AUTO, 930 "features", CTLTYPE_STRING | CTLFLAG_RD | CTLFLAG_NEEDGIANT, xbd, 931 0, xbd_sysctl_features, "A", "protocol features (negotiated)"); 932 } 933 934 /* 935 * Translate Linux major/minor to an appropriate name and unit 936 * number. For HVM guests, this allows us to use the same drive names 937 * with blkfront as the emulated drives, easing transition slightly. 938 */ 939 static void 940 xbd_vdevice_to_unit(uint32_t vdevice, int *unit, const char **name) 941 { 942 static struct vdev_info { 943 int major; 944 int shift; 945 int base; 946 const char *name; 947 } info[] = { 948 {3, 6, 0, "ada"}, /* ide0 */ 949 {22, 6, 2, "ada"}, /* ide1 */ 950 {33, 6, 4, "ada"}, /* ide2 */ 951 {34, 6, 6, "ada"}, /* ide3 */ 952 {56, 6, 8, "ada"}, /* ide4 */ 953 {57, 6, 10, "ada"}, /* ide5 */ 954 {88, 6, 12, "ada"}, /* ide6 */ 955 {89, 6, 14, "ada"}, /* ide7 */ 956 {90, 6, 16, "ada"}, /* ide8 */ 957 {91, 6, 18, "ada"}, /* ide9 */ 958 959 {8, 4, 0, "da"}, /* scsi disk0 */ 960 {65, 4, 16, "da"}, /* scsi disk1 */ 961 {66, 4, 32, "da"}, /* scsi disk2 */ 962 {67, 4, 48, "da"}, /* scsi disk3 */ 963 {68, 4, 64, "da"}, /* scsi disk4 */ 964 {69, 4, 80, "da"}, /* scsi disk5 */ 965 {70, 4, 96, "da"}, /* scsi disk6 */ 966 {71, 4, 112, "da"}, /* scsi disk7 */ 967 {128, 4, 128, "da"}, /* scsi disk8 */ 968 {129, 4, 144, "da"}, /* scsi disk9 */ 969 {130, 4, 160, "da"}, /* scsi disk10 */ 970 {131, 4, 176, "da"}, /* scsi disk11 */ 971 {132, 4, 192, "da"}, /* scsi disk12 */ 972 {133, 4, 208, "da"}, /* scsi disk13 */ 973 {134, 4, 224, "da"}, /* scsi disk14 */ 974 {135, 4, 240, "da"}, /* scsi disk15 */ 975 976 {202, 4, 0, "xbd"}, /* xbd */ 977 978 {0, 0, 0, NULL}, 979 }; 980 int major = vdevice >> 8; 981 int minor = vdevice & 0xff; 982 int i; 983 984 if (vdevice & (1 << 28)) { 985 *unit = (vdevice & ((1 << 28) - 1)) >> 8; 986 *name = "xbd"; 987 return; 988 } 989 990 for (i = 0; info[i].major; i++) { 991 if (info[i].major == major) { 992 *unit = info[i].base + (minor >> info[i].shift); 993 *name = info[i].name; 994 return; 995 } 996 } 997 998 *unit = minor >> 4; 999 *name = "xbd"; 1000 } 1001 1002 int 1003 xbd_instance_create(struct xbd_softc *sc, blkif_sector_t sectors, 1004 int vdevice, uint16_t vdisk_info, unsigned long sector_size, 1005 unsigned long phys_sector_size) 1006 { 1007 char features[80]; 1008 int unit, error = 0; 1009 const char *name; 1010 1011 xbd_vdevice_to_unit(vdevice, &unit, &name); 1012 1013 sc->xbd_unit = unit; 1014 1015 if (strcmp(name, "xbd") != 0) 1016 device_printf(sc->xbd_dev, "attaching as %s%d\n", name, unit); 1017 1018 if (xbd_feature_string(sc, features, sizeof(features)) > 0) { 1019 device_printf(sc->xbd_dev, "features: %s\n", 1020 features); 1021 } 1022 1023 sc->xbd_disk = disk_alloc(); 1024 sc->xbd_disk->d_unit = sc->xbd_unit; 1025 sc->xbd_disk->d_open = xbd_open; 1026 sc->xbd_disk->d_close = xbd_close; 1027 sc->xbd_disk->d_ioctl = xbd_ioctl; 1028 sc->xbd_disk->d_strategy = xbd_strategy; 1029 sc->xbd_disk->d_dump = xbd_dump; 1030 sc->xbd_disk->d_name = name; 1031 sc->xbd_disk->d_drv1 = sc; 1032 sc->xbd_disk->d_sectorsize = sector_size; 1033 sc->xbd_disk->d_stripesize = phys_sector_size; 1034 sc->xbd_disk->d_stripeoffset = 0; 1035 1036 sc->xbd_disk->d_mediasize = sectors * sector_size; 1037 sc->xbd_disk->d_maxsize = sc->xbd_max_request_size; 1038 sc->xbd_disk->d_flags = DISKFLAG_UNMAPPED_BIO; 1039 if ((sc->xbd_flags & (XBDF_FLUSH|XBDF_BARRIER)) != 0) { 1040 sc->xbd_disk->d_flags |= DISKFLAG_CANFLUSHCACHE; 1041 device_printf(sc->xbd_dev, 1042 "synchronize cache commands enabled.\n"); 1043 } 1044 disk_create(sc->xbd_disk, DISK_VERSION); 1045 1046 return error; 1047 } 1048 1049 static void 1050 xbd_free(struct xbd_softc *sc) 1051 { 1052 int i; 1053 1054 /* Prevent new requests being issued until we fix things up. */ 1055 mtx_lock(&sc->xbd_io_lock); 1056 sc->xbd_state = XBD_STATE_DISCONNECTED; 1057 mtx_unlock(&sc->xbd_io_lock); 1058 1059 /* Free resources associated with old device channel. */ 1060 xbd_free_ring(sc); 1061 if (sc->xbd_shadow) { 1062 1063 for (i = 0; i < sc->xbd_max_requests; i++) { 1064 struct xbd_command *cm; 1065 1066 cm = &sc->xbd_shadow[i]; 1067 if (cm->cm_sg_refs != NULL) { 1068 free(cm->cm_sg_refs, M_XENBLOCKFRONT); 1069 cm->cm_sg_refs = NULL; 1070 } 1071 1072 if (cm->cm_indirectionpages != NULL) { 1073 gnttab_end_foreign_access_references( 1074 sc->xbd_max_request_indirectpages, 1075 &cm->cm_indirectionrefs[0]); 1076 contigfree(cm->cm_indirectionpages, PAGE_SIZE * 1077 sc->xbd_max_request_indirectpages, 1078 M_XENBLOCKFRONT); 1079 cm->cm_indirectionpages = NULL; 1080 } 1081 1082 bus_dmamap_destroy(sc->xbd_io_dmat, cm->cm_map); 1083 } 1084 free(sc->xbd_shadow, M_XENBLOCKFRONT); 1085 sc->xbd_shadow = NULL; 1086 1087 bus_dma_tag_destroy(sc->xbd_io_dmat); 1088 1089 xbd_initq_cm(sc, XBD_Q_FREE); 1090 xbd_initq_cm(sc, XBD_Q_READY); 1091 xbd_initq_cm(sc, XBD_Q_COMPLETE); 1092 } 1093 1094 xen_intr_unbind(&sc->xen_intr_handle); 1095 1096 } 1097 1098 /*--------------------------- State Change Handlers --------------------------*/ 1099 static void 1100 xbd_initialize(struct xbd_softc *sc) 1101 { 1102 const char *otherend_path; 1103 const char *node_path; 1104 uint32_t max_ring_page_order; 1105 int error; 1106 1107 if (xenbus_get_state(sc->xbd_dev) != XenbusStateInitialising) { 1108 /* Initialization has already been performed. */ 1109 return; 1110 } 1111 1112 /* 1113 * Protocol defaults valid even if negotiation for a 1114 * setting fails. 1115 */ 1116 max_ring_page_order = 0; 1117 sc->xbd_ring_pages = 1; 1118 1119 /* 1120 * Protocol negotiation. 1121 * 1122 * \note xs_gather() returns on the first encountered error, so 1123 * we must use independent calls in order to guarantee 1124 * we don't miss information in a sparsly populated back-end 1125 * tree. 1126 * 1127 * \note xs_scanf() does not update variables for unmatched 1128 * fields. 1129 */ 1130 otherend_path = xenbus_get_otherend_path(sc->xbd_dev); 1131 node_path = xenbus_get_node(sc->xbd_dev); 1132 1133 /* Support both backend schemes for relaying ring page limits. */ 1134 (void)xs_scanf(XST_NIL, otherend_path, 1135 "max-ring-page-order", NULL, "%" PRIu32, 1136 &max_ring_page_order); 1137 sc->xbd_ring_pages = 1 << max_ring_page_order; 1138 (void)xs_scanf(XST_NIL, otherend_path, 1139 "max-ring-pages", NULL, "%" PRIu32, 1140 &sc->xbd_ring_pages); 1141 if (sc->xbd_ring_pages < 1) 1142 sc->xbd_ring_pages = 1; 1143 1144 if (sc->xbd_ring_pages > XBD_MAX_RING_PAGES) { 1145 device_printf(sc->xbd_dev, 1146 "Back-end specified ring-pages of %u " 1147 "limited to front-end limit of %u.\n", 1148 sc->xbd_ring_pages, XBD_MAX_RING_PAGES); 1149 sc->xbd_ring_pages = XBD_MAX_RING_PAGES; 1150 } 1151 1152 if (powerof2(sc->xbd_ring_pages) == 0) { 1153 uint32_t new_page_limit; 1154 1155 new_page_limit = 0x01 << (fls(sc->xbd_ring_pages) - 1); 1156 device_printf(sc->xbd_dev, 1157 "Back-end specified ring-pages of %u " 1158 "is not a power of 2. Limited to %u.\n", 1159 sc->xbd_ring_pages, new_page_limit); 1160 sc->xbd_ring_pages = new_page_limit; 1161 } 1162 1163 sc->xbd_max_requests = 1164 BLKIF_MAX_RING_REQUESTS(sc->xbd_ring_pages * PAGE_SIZE); 1165 if (sc->xbd_max_requests > XBD_MAX_REQUESTS) { 1166 device_printf(sc->xbd_dev, 1167 "Back-end specified max_requests of %u " 1168 "limited to front-end limit of %zu.\n", 1169 sc->xbd_max_requests, XBD_MAX_REQUESTS); 1170 sc->xbd_max_requests = XBD_MAX_REQUESTS; 1171 } 1172 1173 if (xbd_alloc_ring(sc) != 0) 1174 return; 1175 1176 /* Support both backend schemes for relaying ring page limits. */ 1177 if (sc->xbd_ring_pages > 1) { 1178 error = xs_printf(XST_NIL, node_path, 1179 "num-ring-pages","%u", 1180 sc->xbd_ring_pages); 1181 if (error) { 1182 xenbus_dev_fatal(sc->xbd_dev, error, 1183 "writing %s/num-ring-pages", 1184 node_path); 1185 return; 1186 } 1187 1188 error = xs_printf(XST_NIL, node_path, 1189 "ring-page-order", "%u", 1190 fls(sc->xbd_ring_pages) - 1); 1191 if (error) { 1192 xenbus_dev_fatal(sc->xbd_dev, error, 1193 "writing %s/ring-page-order", 1194 node_path); 1195 return; 1196 } 1197 } 1198 1199 error = xs_printf(XST_NIL, node_path, "event-channel", 1200 "%u", xen_intr_port(sc->xen_intr_handle)); 1201 if (error) { 1202 xenbus_dev_fatal(sc->xbd_dev, error, 1203 "writing %s/event-channel", 1204 node_path); 1205 return; 1206 } 1207 1208 error = xs_printf(XST_NIL, node_path, "protocol", 1209 "%s", XEN_IO_PROTO_ABI_NATIVE); 1210 if (error) { 1211 xenbus_dev_fatal(sc->xbd_dev, error, 1212 "writing %s/protocol", 1213 node_path); 1214 return; 1215 } 1216 1217 xenbus_set_state(sc->xbd_dev, XenbusStateInitialised); 1218 } 1219 1220 /* 1221 * Invoked when the backend is finally 'ready' (and has published 1222 * the details about the physical device - #sectors, size, etc). 1223 */ 1224 static void 1225 xbd_connect(struct xbd_softc *sc) 1226 { 1227 device_t dev = sc->xbd_dev; 1228 unsigned long sectors, sector_size, phys_sector_size; 1229 unsigned int binfo; 1230 int err, feature_barrier, feature_flush; 1231 int i, j; 1232 1233 DPRINTK("blkfront.c:connect:%s.\n", xenbus_get_otherend_path(dev)); 1234 1235 if (sc->xbd_state == XBD_STATE_SUSPENDED) { 1236 return; 1237 } 1238 1239 if (sc->xbd_state == XBD_STATE_CONNECTED) { 1240 struct disk *disk; 1241 1242 disk = sc->xbd_disk; 1243 if (disk == NULL) { 1244 return; 1245 } 1246 err = xs_gather(XST_NIL, xenbus_get_otherend_path(dev), 1247 "sectors", "%lu", §ors, NULL); 1248 if (err != 0) { 1249 xenbus_dev_error(dev, err, 1250 "reading sectors at %s", 1251 xenbus_get_otherend_path(dev)); 1252 return; 1253 } 1254 disk->d_mediasize = disk->d_sectorsize * sectors; 1255 err = disk_resize(disk, M_NOWAIT); 1256 if (err) { 1257 xenbus_dev_error(dev, err, 1258 "unable to resize disk %s%u", 1259 disk->d_name, disk->d_unit); 1260 return; 1261 } 1262 device_printf(sc->xbd_dev, 1263 "changed capacity to %jd\n", 1264 (intmax_t)disk->d_mediasize); 1265 return; 1266 } 1267 1268 err = xs_gather(XST_NIL, xenbus_get_otherend_path(dev), 1269 "sectors", "%lu", §ors, 1270 "info", "%u", &binfo, 1271 "sector-size", "%lu", §or_size, 1272 NULL); 1273 if (err) { 1274 xenbus_dev_fatal(dev, err, 1275 "reading backend fields at %s", 1276 xenbus_get_otherend_path(dev)); 1277 return; 1278 } 1279 if ((sectors == 0) || (sector_size == 0)) { 1280 xenbus_dev_fatal(dev, 0, 1281 "invalid parameters from %s:" 1282 " sectors = %lu, sector_size = %lu", 1283 xenbus_get_otherend_path(dev), 1284 sectors, sector_size); 1285 return; 1286 } 1287 err = xs_gather(XST_NIL, xenbus_get_otherend_path(dev), 1288 "physical-sector-size", "%lu", &phys_sector_size, 1289 NULL); 1290 if (err || phys_sector_size <= sector_size) 1291 phys_sector_size = 0; 1292 err = xs_gather(XST_NIL, xenbus_get_otherend_path(dev), 1293 "feature-barrier", "%d", &feature_barrier, 1294 NULL); 1295 if (err == 0 && feature_barrier != 0) 1296 sc->xbd_flags |= XBDF_BARRIER; 1297 1298 err = xs_gather(XST_NIL, xenbus_get_otherend_path(dev), 1299 "feature-flush-cache", "%d", &feature_flush, 1300 NULL); 1301 if (err == 0 && feature_flush != 0) 1302 sc->xbd_flags |= XBDF_FLUSH; 1303 1304 err = xs_gather(XST_NIL, xenbus_get_otherend_path(dev), 1305 "feature-max-indirect-segments", "%" PRIu32, 1306 &sc->xbd_max_request_segments, NULL); 1307 if ((err != 0) || (xbd_enable_indirect == 0)) 1308 sc->xbd_max_request_segments = 0; 1309 if (sc->xbd_max_request_segments > XBD_MAX_INDIRECT_SEGMENTS) 1310 sc->xbd_max_request_segments = XBD_MAX_INDIRECT_SEGMENTS; 1311 if (sc->xbd_max_request_segments > XBD_SIZE_TO_SEGS(MAXPHYS)) 1312 sc->xbd_max_request_segments = XBD_SIZE_TO_SEGS(MAXPHYS); 1313 sc->xbd_max_request_indirectpages = 1314 XBD_INDIRECT_SEGS_TO_PAGES(sc->xbd_max_request_segments); 1315 if (sc->xbd_max_request_segments < BLKIF_MAX_SEGMENTS_PER_REQUEST) 1316 sc->xbd_max_request_segments = BLKIF_MAX_SEGMENTS_PER_REQUEST; 1317 sc->xbd_max_request_size = 1318 XBD_SEGS_TO_SIZE(sc->xbd_max_request_segments); 1319 1320 /* Allocate datastructures based on negotiated values. */ 1321 err = bus_dma_tag_create( 1322 bus_get_dma_tag(sc->xbd_dev), /* parent */ 1323 512, PAGE_SIZE, /* algnmnt, boundary */ 1324 BUS_SPACE_MAXADDR, /* lowaddr */ 1325 BUS_SPACE_MAXADDR, /* highaddr */ 1326 NULL, NULL, /* filter, filterarg */ 1327 sc->xbd_max_request_size, 1328 sc->xbd_max_request_segments, 1329 PAGE_SIZE, /* maxsegsize */ 1330 BUS_DMA_ALLOCNOW, /* flags */ 1331 busdma_lock_mutex, /* lockfunc */ 1332 &sc->xbd_io_lock, /* lockarg */ 1333 &sc->xbd_io_dmat); 1334 if (err != 0) { 1335 xenbus_dev_fatal(sc->xbd_dev, err, 1336 "Cannot allocate parent DMA tag\n"); 1337 return; 1338 } 1339 1340 /* Per-transaction data allocation. */ 1341 sc->xbd_shadow = malloc(sizeof(*sc->xbd_shadow) * sc->xbd_max_requests, 1342 M_XENBLOCKFRONT, M_NOWAIT|M_ZERO); 1343 if (sc->xbd_shadow == NULL) { 1344 bus_dma_tag_destroy(sc->xbd_io_dmat); 1345 xenbus_dev_fatal(sc->xbd_dev, ENOMEM, 1346 "Cannot allocate request structures\n"); 1347 return; 1348 } 1349 1350 for (i = 0; i < sc->xbd_max_requests; i++) { 1351 struct xbd_command *cm; 1352 void * indirectpages; 1353 1354 cm = &sc->xbd_shadow[i]; 1355 cm->cm_sg_refs = malloc( 1356 sizeof(grant_ref_t) * sc->xbd_max_request_segments, 1357 M_XENBLOCKFRONT, M_NOWAIT); 1358 if (cm->cm_sg_refs == NULL) 1359 break; 1360 cm->cm_id = i; 1361 cm->cm_flags = XBDCF_INITIALIZER; 1362 cm->cm_sc = sc; 1363 if (bus_dmamap_create(sc->xbd_io_dmat, 0, &cm->cm_map) != 0) 1364 break; 1365 if (sc->xbd_max_request_indirectpages > 0) { 1366 indirectpages = contigmalloc( 1367 PAGE_SIZE * sc->xbd_max_request_indirectpages, 1368 M_XENBLOCKFRONT, M_ZERO | M_NOWAIT, 0, ~0, 1369 PAGE_SIZE, 0); 1370 if (indirectpages == NULL) 1371 sc->xbd_max_request_indirectpages = 0; 1372 } else { 1373 indirectpages = NULL; 1374 } 1375 for (j = 0; j < sc->xbd_max_request_indirectpages; j++) { 1376 if (gnttab_grant_foreign_access( 1377 xenbus_get_otherend_id(sc->xbd_dev), 1378 (vtophys(indirectpages) >> PAGE_SHIFT) + j, 1379 1 /* grant read-only access */, 1380 &cm->cm_indirectionrefs[j])) 1381 break; 1382 } 1383 if (j < sc->xbd_max_request_indirectpages) { 1384 contigfree(indirectpages, 1385 PAGE_SIZE * sc->xbd_max_request_indirectpages, 1386 M_XENBLOCKFRONT); 1387 break; 1388 } 1389 cm->cm_indirectionpages = indirectpages; 1390 xbd_free_command(cm); 1391 } 1392 1393 if (sc->xbd_disk == NULL) { 1394 device_printf(dev, "%juMB <%s> at %s", 1395 (uintmax_t) sectors / (1048576 / sector_size), 1396 device_get_desc(dev), 1397 xenbus_get_node(dev)); 1398 bus_print_child_footer(device_get_parent(dev), dev); 1399 1400 xbd_instance_create(sc, sectors, sc->xbd_vdevice, binfo, 1401 sector_size, phys_sector_size); 1402 } 1403 1404 (void)xenbus_set_state(dev, XenbusStateConnected); 1405 1406 /* Kick pending requests. */ 1407 mtx_lock(&sc->xbd_io_lock); 1408 sc->xbd_state = XBD_STATE_CONNECTED; 1409 xbd_startio(sc); 1410 sc->xbd_flags |= XBDF_READY; 1411 mtx_unlock(&sc->xbd_io_lock); 1412 } 1413 1414 /** 1415 * Handle the change of state of the backend to Closing. We must delete our 1416 * device-layer structures now, to ensure that writes are flushed through to 1417 * the backend. Once this is done, we can switch to Closed in 1418 * acknowledgement. 1419 */ 1420 static void 1421 xbd_closing(device_t dev) 1422 { 1423 struct xbd_softc *sc = device_get_softc(dev); 1424 1425 xenbus_set_state(dev, XenbusStateClosing); 1426 1427 DPRINTK("xbd_closing: %s removed\n", xenbus_get_node(dev)); 1428 1429 if (sc->xbd_disk != NULL) { 1430 disk_destroy(sc->xbd_disk); 1431 sc->xbd_disk = NULL; 1432 } 1433 1434 xenbus_set_state(dev, XenbusStateClosed); 1435 } 1436 1437 /*---------------------------- NewBus Entrypoints ----------------------------*/ 1438 static int 1439 xbd_probe(device_t dev) 1440 { 1441 if (strcmp(xenbus_get_type(dev), "vbd") != 0) 1442 return (ENXIO); 1443 1444 if (xen_hvm_domain() && xen_disable_pv_disks != 0) 1445 return (ENXIO); 1446 1447 if (xen_hvm_domain()) { 1448 int error; 1449 char *type; 1450 1451 /* 1452 * When running in an HVM domain, IDE disk emulation is 1453 * disabled early in boot so that native drivers will 1454 * not see emulated hardware. However, CDROM device 1455 * emulation cannot be disabled. 1456 * 1457 * Through use of FreeBSD's vm_guest and xen_hvm_domain() 1458 * APIs, we could modify the native CDROM driver to fail its 1459 * probe when running under Xen. Unfortunatlely, the PV 1460 * CDROM support in XenServer (up through at least version 1461 * 6.2) isn't functional, so we instead rely on the emulated 1462 * CDROM instance, and fail to attach the PV one here in 1463 * the blkfront driver. 1464 */ 1465 error = xs_read(XST_NIL, xenbus_get_node(dev), 1466 "device-type", NULL, (void **) &type); 1467 if (error) 1468 return (ENXIO); 1469 1470 if (strncmp(type, "cdrom", 5) == 0) { 1471 free(type, M_XENSTORE); 1472 return (ENXIO); 1473 } 1474 free(type, M_XENSTORE); 1475 } 1476 1477 device_set_desc(dev, "Virtual Block Device"); 1478 device_quiet(dev); 1479 return (0); 1480 } 1481 1482 /* 1483 * Setup supplies the backend dir, virtual device. We place an event 1484 * channel and shared frame entries. We watch backend to wait if it's 1485 * ok. 1486 */ 1487 static int 1488 xbd_attach(device_t dev) 1489 { 1490 struct xbd_softc *sc; 1491 const char *name; 1492 uint32_t vdevice; 1493 int error; 1494 int i; 1495 int unit; 1496 1497 /* FIXME: Use dynamic device id if this is not set. */ 1498 error = xs_scanf(XST_NIL, xenbus_get_node(dev), 1499 "virtual-device", NULL, "%" PRIu32, &vdevice); 1500 if (error) 1501 error = xs_scanf(XST_NIL, xenbus_get_node(dev), 1502 "virtual-device-ext", NULL, "%" PRIu32, &vdevice); 1503 if (error) { 1504 xenbus_dev_fatal(dev, error, "reading virtual-device"); 1505 device_printf(dev, "Couldn't determine virtual device.\n"); 1506 return (error); 1507 } 1508 1509 xbd_vdevice_to_unit(vdevice, &unit, &name); 1510 if (!strcmp(name, "xbd")) 1511 device_set_unit(dev, unit); 1512 1513 sc = device_get_softc(dev); 1514 mtx_init(&sc->xbd_io_lock, "blkfront i/o lock", NULL, MTX_DEF); 1515 xbd_initqs(sc); 1516 for (i = 0; i < XBD_MAX_RING_PAGES; i++) 1517 sc->xbd_ring_ref[i] = GRANT_REF_INVALID; 1518 1519 sc->xbd_dev = dev; 1520 sc->xbd_vdevice = vdevice; 1521 sc->xbd_state = XBD_STATE_DISCONNECTED; 1522 1523 xbd_setup_sysctl(sc); 1524 1525 /* Wait for backend device to publish its protocol capabilities. */ 1526 xenbus_set_state(dev, XenbusStateInitialising); 1527 1528 return (0); 1529 } 1530 1531 static int 1532 xbd_detach(device_t dev) 1533 { 1534 struct xbd_softc *sc = device_get_softc(dev); 1535 1536 DPRINTK("%s: %s removed\n", __func__, xenbus_get_node(dev)); 1537 1538 xbd_free(sc); 1539 mtx_destroy(&sc->xbd_io_lock); 1540 1541 return 0; 1542 } 1543 1544 static int 1545 xbd_suspend(device_t dev) 1546 { 1547 struct xbd_softc *sc = device_get_softc(dev); 1548 int retval; 1549 int saved_state; 1550 1551 /* Prevent new requests being issued until we fix things up. */ 1552 mtx_lock(&sc->xbd_io_lock); 1553 saved_state = sc->xbd_state; 1554 sc->xbd_state = XBD_STATE_SUSPENDED; 1555 1556 /* Wait for outstanding I/O to drain. */ 1557 retval = 0; 1558 while (xbd_queue_length(sc, XBD_Q_BUSY) != 0) { 1559 if (msleep(&sc->xbd_cm_q[XBD_Q_BUSY], &sc->xbd_io_lock, 1560 PRIBIO, "blkf_susp", 30 * hz) == EWOULDBLOCK) { 1561 retval = EBUSY; 1562 break; 1563 } 1564 } 1565 mtx_unlock(&sc->xbd_io_lock); 1566 1567 if (retval != 0) 1568 sc->xbd_state = saved_state; 1569 1570 return (retval); 1571 } 1572 1573 static int 1574 xbd_resume(device_t dev) 1575 { 1576 struct xbd_softc *sc = device_get_softc(dev); 1577 1578 if (xen_suspend_cancelled) { 1579 sc->xbd_state = XBD_STATE_CONNECTED; 1580 return (0); 1581 } 1582 1583 DPRINTK("xbd_resume: %s\n", xenbus_get_node(dev)); 1584 1585 xbd_free(sc); 1586 xbd_initialize(sc); 1587 return (0); 1588 } 1589 1590 /** 1591 * Callback received when the backend's state changes. 1592 */ 1593 static void 1594 xbd_backend_changed(device_t dev, XenbusState backend_state) 1595 { 1596 struct xbd_softc *sc = device_get_softc(dev); 1597 1598 DPRINTK("backend_state=%d\n", backend_state); 1599 1600 switch (backend_state) { 1601 case XenbusStateUnknown: 1602 case XenbusStateInitialising: 1603 case XenbusStateReconfigured: 1604 case XenbusStateReconfiguring: 1605 case XenbusStateClosed: 1606 break; 1607 1608 case XenbusStateInitWait: 1609 case XenbusStateInitialised: 1610 xbd_initialize(sc); 1611 break; 1612 1613 case XenbusStateConnected: 1614 xbd_initialize(sc); 1615 xbd_connect(sc); 1616 break; 1617 1618 case XenbusStateClosing: 1619 if (sc->xbd_users > 0) { 1620 device_printf(dev, "detaching with pending users\n"); 1621 KASSERT(sc->xbd_disk != NULL, 1622 ("NULL disk with pending users\n")); 1623 disk_gone(sc->xbd_disk); 1624 } else { 1625 xbd_closing(dev); 1626 } 1627 break; 1628 } 1629 } 1630 1631 /*---------------------------- NewBus Registration ---------------------------*/ 1632 static device_method_t xbd_methods[] = { 1633 /* Device interface */ 1634 DEVMETHOD(device_probe, xbd_probe), 1635 DEVMETHOD(device_attach, xbd_attach), 1636 DEVMETHOD(device_detach, xbd_detach), 1637 DEVMETHOD(device_shutdown, bus_generic_shutdown), 1638 DEVMETHOD(device_suspend, xbd_suspend), 1639 DEVMETHOD(device_resume, xbd_resume), 1640 1641 /* Xenbus interface */ 1642 DEVMETHOD(xenbus_otherend_changed, xbd_backend_changed), 1643 1644 { 0, 0 } 1645 }; 1646 1647 static driver_t xbd_driver = { 1648 "xbd", 1649 xbd_methods, 1650 sizeof(struct xbd_softc), 1651 }; 1652 devclass_t xbd_devclass; 1653 1654 DRIVER_MODULE(xbd, xenbusb_front, xbd_driver, xbd_devclass, 0, 0); 1655