drivers/block/xen-blkfront.c

   1 /*
   2  * blkfront.c
   3  *
   4  * XenLinux virtual block device driver.
   5  *
   6  * Copyright (c) 2003-2004, Keir Fraser & Steve Hand
   7  * Modifications by Mark A. Williamson are (c) Intel Research Cambridge
   8  * Copyright (c) 2004, Christian Limpach
   9  * Copyright (c) 2004, Andrew Warfield
  10  * Copyright (c) 2005, Christopher Clark
  11  * Copyright (c) 2005, XenSource Ltd
  12  *
  13  * This program is free software; you can redistribute it and/or
  14  * modify it under the terms of the GNU General Public License version 2
  15  * as published by the Free Software Foundation; or, when distributed
  16  * separately from the Linux kernel or incorporated into other
  17  * software packages, subject to the following license:
  18  *
  19  * Permission is hereby granted, free of charge, to any person obtaining a copy
  20  * of this source file (the "Software"), to deal in the Software without
  21  * restriction, including without limitation the rights to use, copy, modify,
  22  * merge, publish, distribute, sublicense, and/or sell copies of the Software,
  23  * and to permit persons to whom the Software is furnished to do so, subject to
  24  * the following conditions:
  25  *
  26  * The above copyright notice and this permission notice shall be included in
  27  * all copies or substantial portions of the Software.
  28  *
  29  * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
  30  * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
  31  * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
  32  * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
  33  * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
  34  * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
  35  * IN THE SOFTWARE.
  36  */
  37
  38 #include <linux/interrupt.h>
  39 #include <linux/blkdev.h>
  40 #include <linux/blk-mq.h>
  41 #include <linux/hdreg.h>
  42 #include <linux/cdrom.h>
  43 #include <linux/module.h>
  44 #include <linux/slab.h>
  45 #include <linux/mutex.h>
  46 #include <linux/scatterlist.h>
  47 #include <linux/bitmap.h>
  48 #include <linux/list.h>
  49 #include <linux/workqueue.h>
  50 #include <linux/sched/mm.h>
  51
  52 #include <xen/xen.h>
  53 #include <xen/xenbus.h>
  54 #include <xen/grant_table.h>
  55 #include <xen/events.h>
  56 #include <xen/page.h>
  57 #include <xen/platform_pci.h>
  58
  59 #include <xen/interface/grant_table.h>
  60 #include <xen/interface/io/blkif.h>
  61 #include <xen/interface/io/protocols.h>
  62
  63 #include <asm/xen/hypervisor.h>
  64
  65 /*
  66  * The minimal size of segment supported by the block framework is PAGE_SIZE.
  67  * When Linux is using a different page size than Xen, it may not be possible
  68  * to put all the data in a single segment.
  69  * This can happen when the backend doesn't support indirect descriptor and
  70  * therefore the maximum amount of data that a request can carry is
  71  * BLKIF_MAX_SEGMENTS_PER_REQUEST * XEN_PAGE_SIZE = 44KB
  72  *
  73  * Note that we only support one extra request. So the Linux page size
  74  * should be <= ( 2 * BLKIF_MAX_SEGMENTS_PER_REQUEST * XEN_PAGE_SIZE) =
  75  * 88KB.
  76  */
  77 #define HAS_EXTRA_REQ (BLKIF_MAX_SEGMENTS_PER_REQUEST < XEN_PFN_PER_PAGE)
  78
  79 enum blkif_state {
  80         BLKIF_STATE_DISCONNECTED,
  81         BLKIF_STATE_CONNECTED,
  82         BLKIF_STATE_SUSPENDED,
  83 };
  84
  85 struct grant {
  86         grant_ref_t gref;
  87         struct page *page;
  88         struct list_head node;
  89 };
  90
  91 enum blk_req_status {
  92         REQ_WAITING,
  93         REQ_DONE,
  94         REQ_ERROR,
  95         REQ_EOPNOTSUPP,
  96 };
  97
  98 struct blk_shadow {
  99         struct blkif_request req;
 100         struct request *request;
 101         struct grant **grants_used;
 102         struct grant **indirect_grants;
 103         struct scatterlist *sg;
 104         unsigned int num_sg;
 105         enum blk_req_status status;
 106
 107         #define NO_ASSOCIATED_ID ~0UL
 108         /*
 109          * Id of the sibling if we ever need 2 requests when handling a
 110          * block I/O request
 111          */
 112         unsigned long associated_id;
 113 };
 114
 115 struct blkif_req {
 116         blk_status_t    error;
 117 };
 118
 119 static inline struct blkif_req *blkif_req(struct request *rq)
 120 {
 121         return blk_mq_rq_to_pdu(rq);
 122 }
 123
 124 static DEFINE_MUTEX(blkfront_mutex);
 125 static const struct block_device_operations xlvbd_block_fops;
 126 static struct delayed_work blkfront_work;
 127 static LIST_HEAD(info_list);
 128
 129 /*
 130  * Maximum number of segments in indirect requests, the actual value used by
 131  * the frontend driver is the minimum of this value and the value provided
 132  * by the backend driver.
 133  */
 134
 135 static unsigned int xen_blkif_max_segments = 32;
 136 module_param_named(max_indirect_segments, xen_blkif_max_segments, uint, 0444);
 137 MODULE_PARM_DESC(max_indirect_segments,
 138                  "Maximum amount of segments in indirect requests (default is 32)");
 139
 140 static unsigned int xen_blkif_max_queues = 4;
 141 module_param_named(max_queues, xen_blkif_max_queues, uint, 0444);
 142 MODULE_PARM_DESC(max_queues, "Maximum number of hardware queues/rings used per virtual disk");
 143
 144 /*
 145  * Maximum order of pages to be used for the shared ring between front and
 146  * backend, 4KB page granularity is used.
 147  */
 148 static unsigned int xen_blkif_max_ring_order;
 149 module_param_named(max_ring_page_order, xen_blkif_max_ring_order, int, 0444);
 150 MODULE_PARM_DESC(max_ring_page_order, "Maximum order of pages to be used for the shared ring");
 151
 152 #define BLK_RING_SIZE(info)     \
 153         __CONST_RING_SIZE(blkif, XEN_PAGE_SIZE * (info)->nr_ring_pages)
 154
 155 /*
 156  * ring-ref%u i=(-1UL) would take 11 characters + 'ring-ref' is 8, so 19
 157  * characters are enough. Define to 20 to keep consistent with backend.
 158  */
 159 #define RINGREF_NAME_LEN (20)
 160 /*
 161  * queue-%u would take 7 + 10(UINT_MAX) = 17 characters.
 162  */
 163 #define QUEUE_NAME_LEN (17)
 164
 165 /*
 166  *  Per-ring info.
 167  *  Every blkfront device can associate with one or more blkfront_ring_info,
 168  *  depending on how many hardware queues/rings to be used.
 169  */
 170 struct blkfront_ring_info {
 171         /* Lock to protect data in every ring buffer. */
 172         spinlock_t ring_lock;
 173         struct blkif_front_ring ring;
 174         unsigned int ring_ref[XENBUS_MAX_RING_GRANTS];
 175         unsigned int evtchn, irq;
 176         struct work_struct work;
 177         struct gnttab_free_callback callback;
 178         struct list_head indirect_pages;
 179         struct list_head grants;
 180         unsigned int persistent_gnts_c;
 181         unsigned long shadow_free;
 182         struct blkfront_info *dev_info;
 183         struct blk_shadow shadow[];
 184 };
 185
 186 /*
 187  * We have one of these per vbd, whether ide, scsi or 'other'.  They
 188  * hang in private_data off the gendisk structure. We may end up
 189  * putting all kinds of interesting stuff here :-)
 190  */
 191 struct blkfront_info
 192 {
 193         struct mutex mutex;
 194         struct xenbus_device *xbdev;
 195         struct gendisk *gd;
 196         u16 sector_size;
 197         unsigned int physical_sector_size;
 198         int vdevice;
 199         blkif_vdev_t handle;
 200         enum blkif_state connected;
 201         /* Number of pages per ring buffer. */
 202         unsigned int nr_ring_pages;
 203         struct request_queue *rq;
 204         unsigned int feature_flush:1;
 205         unsigned int feature_fua:1;
 206         unsigned int feature_discard:1;
 207         unsigned int feature_secdiscard:1;
 208         unsigned int feature_persistent:1;
 209         unsigned int discard_granularity;
 210         unsigned int discard_alignment;
 211         /* Number of 4KB segments handled */
 212         unsigned int max_indirect_segments;
 213         int is_ready;
 214         struct blk_mq_tag_set tag_set;
 215         struct blkfront_ring_info *rinfo;
 216         unsigned int nr_rings;
 217         unsigned int rinfo_size;
 218         /* Save uncomplete reqs and bios for migration. */
 219         struct list_head requests;
 220         struct bio_list bio_list;
 221         struct list_head info_list;
 222 };
 223
 224 static unsigned int nr_minors;
 225 static unsigned long *minors;
 226 static DEFINE_SPINLOCK(minor_lock);
 227
 228 #define GRANT_INVALID_REF       0
 229
 230 #define PARTS_PER_DISK          16
 231 #define PARTS_PER_EXT_DISK      256
 232
 233 #define BLKIF_MAJOR(dev) ((dev)>>8)
 234 #define BLKIF_MINOR(dev) ((dev) & 0xff)
 235
 236 #define EXT_SHIFT 28
 237 #define EXTENDED (1<<EXT_SHIFT)
 238 #define VDEV_IS_EXTENDED(dev) ((dev)&(EXTENDED))
 239 #define BLKIF_MINOR_EXT(dev) ((dev)&(~EXTENDED))
 240 #define EMULATED_HD_DISK_MINOR_OFFSET (0)
 241 #define EMULATED_HD_DISK_NAME_OFFSET (EMULATED_HD_DISK_MINOR_OFFSET / 256)
 242 #define EMULATED_SD_DISK_MINOR_OFFSET (0)
 243 #define EMULATED_SD_DISK_NAME_OFFSET (EMULATED_SD_DISK_MINOR_OFFSET / 256)
 244
 245 #define DEV_NAME        "xvd"   /* name in /dev */
 246
 247 /*
 248  * Grants are always the same size as a Xen page (i.e 4KB).
 249  * A physical segment is always the same size as a Linux page.
 250  * Number of grants per physical segment
 251  */
 252 #define GRANTS_PER_PSEG (PAGE_SIZE / XEN_PAGE_SIZE)
 253
 254 #define GRANTS_PER_INDIRECT_FRAME \
 255         (XEN_PAGE_SIZE / sizeof(struct blkif_request_segment))
 256
 257 #define INDIRECT_GREFS(_grants)         \
 258         DIV_ROUND_UP(_grants, GRANTS_PER_INDIRECT_FRAME)
 259
 260 static int blkfront_setup_indirect(struct blkfront_ring_info *rinfo);
 261 static void blkfront_gather_backend_features(struct blkfront_info *info);
 262 static int negotiate_mq(struct blkfront_info *info);
 263
 264 #define for_each_rinfo(info, ptr, idx)                          \
 265         for ((ptr) = (info)->rinfo, (idx) = 0;                  \
 266              (idx) < (info)->nr_rings;                          \
 267              (idx)++, (ptr) = (void *)(ptr) + (info)->rinfo_size)
 268
 269 static inline struct blkfront_ring_info *
 270 get_rinfo(const struct blkfront_info *info, unsigned int i)
 271 {
 272         BUG_ON(i >= info->nr_rings);
 273         return (void *)info->rinfo + i * info->rinfo_size;
 274 }
 275
 276 static int get_id_from_freelist(struct blkfront_ring_info *rinfo)
 277 {
 278         unsigned long free = rinfo->shadow_free;
 279
 280         BUG_ON(free >= BLK_RING_SIZE(rinfo->dev_info));
 281         rinfo->shadow_free = rinfo->shadow[free].req.u.rw.id;
 282         rinfo->shadow[free].req.u.rw.id = 0x0fffffee; /* debug */
 283         return free;
 284 }
 285
 286 static int add_id_to_freelist(struct blkfront_ring_info *rinfo,
 287                               unsigned long id)
 288 {
 289         if (rinfo->shadow[id].req.u.rw.id != id)
 290                 return -EINVAL;
 291         if (rinfo->shadow[id].request == NULL)
 292                 return -EINVAL;
 293         rinfo->shadow[id].req.u.rw.id  = rinfo->shadow_free;
 294         rinfo->shadow[id].request = NULL;
 295         rinfo->shadow_free = id;
 296         return 0;
 297 }
 298
 299 static int fill_grant_buffer(struct blkfront_ring_info *rinfo, int num)
 300 {
 301         struct blkfront_info *info = rinfo->dev_info;
 302         struct page *granted_page;
 303         struct grant *gnt_list_entry, *n;
 304         int i = 0;
 305
 306         while (i < num) {
 307                 gnt_list_entry = kzalloc(sizeof(struct grant), GFP_NOIO);
 308                 if (!gnt_list_entry)
 309                         goto out_of_memory;
 310
 311                 if (info->feature_persistent) {
 312                         granted_page = alloc_page(GFP_NOIO);
 313                         if (!granted_page) {
 314                                 kfree(gnt_list_entry);
 315                                 goto out_of_memory;
 316                         }
 317                         gnt_list_entry->page = granted_page;
 318                 }
 319
 320                 gnt_list_entry->gref = GRANT_INVALID_REF;
 321                 list_add(&gnt_list_entry->node, &rinfo->grants);
 322                 i++;
 323         }
 324
 325         return 0;
 326
 327 out_of_memory:
 328         list_for_each_entry_safe(gnt_list_entry, n,
 329                                  &rinfo->grants, node) {
 330                 list_del(&gnt_list_entry->node);
 331                 if (info->feature_persistent)
 332                         __free_page(gnt_list_entry->page);
 333                 kfree(gnt_list_entry);
 334                 i--;
 335         }
 336         BUG_ON(i != 0);
 337         return -ENOMEM;
 338 }
 339
 340 static struct grant *get_free_grant(struct blkfront_ring_info *rinfo)
 341 {
 342         struct grant *gnt_list_entry;
 343
 344         BUG_ON(list_empty(&rinfo->grants));
 345         gnt_list_entry = list_first_entry(&rinfo->grants, struct grant,
 346                                           node);
 347         list_del(&gnt_list_entry->node);
 348
 349         if (gnt_list_entry->gref != GRANT_INVALID_REF)
 350                 rinfo->persistent_gnts_c--;
 351
 352         return gnt_list_entry;
 353 }
 354
 355 static inline void grant_foreign_access(const struct grant *gnt_list_entry,
 356                                         const struct blkfront_info *info)
 357 {
 358         gnttab_page_grant_foreign_access_ref_one(gnt_list_entry->gref,
 359                                                  info->xbdev->otherend_id,
 360                                                  gnt_list_entry->page,
 361                                                  0);
 362 }
 363
 364 static struct grant *get_grant(grant_ref_t *gref_head,
 365                                unsigned long gfn,
 366                                struct blkfront_ring_info *rinfo)
 367 {
 368         struct grant *gnt_list_entry = get_free_grant(rinfo);
 369         struct blkfront_info *info = rinfo->dev_info;
 370
 371         if (gnt_list_entry->gref != GRANT_INVALID_REF)
 372                 return gnt_list_entry;
 373
 374         /* Assign a gref to this page */
 375         gnt_list_entry->gref = gnttab_claim_grant_reference(gref_head);
 376         BUG_ON(gnt_list_entry->gref == -ENOSPC);
 377         if (info->feature_persistent)
 378                 grant_foreign_access(gnt_list_entry, info);
 379         else {
 380                 /* Grant access to the GFN passed by the caller */
 381                 gnttab_grant_foreign_access_ref(gnt_list_entry->gref,
 382                                                 info->xbdev->otherend_id,
 383                                                 gfn, 0);
 384         }
 385
 386         return gnt_list_entry;
 387 }
 388
 389 static struct grant *get_indirect_grant(grant_ref_t *gref_head,
 390                                         struct blkfront_ring_info *rinfo)
 391 {
 392         struct grant *gnt_list_entry = get_free_grant(rinfo);
 393         struct blkfront_info *info = rinfo->dev_info;
 394
 395         if (gnt_list_entry->gref != GRANT_INVALID_REF)
 396                 return gnt_list_entry;
 397
 398         /* Assign a gref to this page */
 399         gnt_list_entry->gref = gnttab_claim_grant_reference(gref_head);
 400         BUG_ON(gnt_list_entry->gref == -ENOSPC);
 401         if (!info->feature_persistent) {
 402                 struct page *indirect_page;
 403
 404                 /* Fetch a pre-allocated page to use for indirect grefs */
 405                 BUG_ON(list_empty(&rinfo->indirect_pages));
 406                 indirect_page = list_first_entry(&rinfo->indirect_pages,
 407                                                  struct page, lru);
 408                 list_del(&indirect_page->lru);
 409                 gnt_list_entry->page = indirect_page;
 410         }
 411         grant_foreign_access(gnt_list_entry, info);
 412
 413         return gnt_list_entry;
 414 }
 415
 416 static const char *op_name(int op)
 417 {
 418         static const char *const names[] = {
 419                 [BLKIF_OP_READ] = "read",
 420                 [BLKIF_OP_WRITE] = "write",
 421                 [BLKIF_OP_WRITE_BARRIER] = "barrier",
 422                 [BLKIF_OP_FLUSH_DISKCACHE] = "flush",
 423                 [BLKIF_OP_DISCARD] = "discard" };
 424
 425         if (op < 0 || op >= ARRAY_SIZE(names))
 426                 return "unknown";
 427
 428         if (!names[op])
 429                 return "reserved";
 430
 431         return names[op];
 432 }
 433 static int xlbd_reserve_minors(unsigned int minor, unsigned int nr)
 434 {
 435         unsigned int end = minor + nr;
 436         int rc;
 437
 438         if (end > nr_minors) {
 439                 unsigned long *bitmap, *old;
 440
 441                 bitmap = kcalloc(BITS_TO_LONGS(end), sizeof(*bitmap),
 442                                  GFP_KERNEL);
 443                 if (bitmap == NULL)
 444                         return -ENOMEM;
 445
 446                 spin_lock(&minor_lock);
 447                 if (end > nr_minors) {
 448                         old = minors;
 449                         memcpy(bitmap, minors,
 450                                BITS_TO_LONGS(nr_minors) * sizeof(*bitmap));
 451                         minors = bitmap;
 452                         nr_minors = BITS_TO_LONGS(end) * BITS_PER_LONG;
 453                 } else
 454                         old = bitmap;
 455                 spin_unlock(&minor_lock);
 456                 kfree(old);
 457         }
 458
 459         spin_lock(&minor_lock);
 460         if (find_next_bit(minors, end, minor) >= end) {
 461                 bitmap_set(minors, minor, nr);
 462                 rc = 0;
 463         } else
 464                 rc = -EBUSY;
 465         spin_unlock(&minor_lock);
 466
 467         return rc;
 468 }
 469
 470 static void xlbd_release_minors(unsigned int minor, unsigned int nr)
 471 {
 472         unsigned int end = minor + nr;
 473
 474         BUG_ON(end > nr_minors);
 475         spin_lock(&minor_lock);
 476         bitmap_clear(minors,  minor, nr);
 477         spin_unlock(&minor_lock);
 478 }
 479
 480 static void blkif_restart_queue_callback(void *arg)
 481 {
 482         struct blkfront_ring_info *rinfo = (struct blkfront_ring_info *)arg;
 483         schedule_work(&rinfo->work);
 484 }
 485
 486 static int blkif_getgeo(struct block_device *bd, struct hd_geometry *hg)
 487 {
 488         /* We don't have real geometry info, but let's at least return
 489            values consistent with the size of the device */
 490         sector_t nsect = get_capacity(bd->bd_disk);
 491         sector_t cylinders = nsect;
 492
 493         hg->heads = 0xff;
 494         hg->sectors = 0x3f;
 495         sector_div(cylinders, hg->heads * hg->sectors);
 496         hg->cylinders = cylinders;
 497         if ((sector_t)(hg->cylinders + 1) * hg->heads * hg->sectors < nsect)
 498                 hg->cylinders = 0xffff;
 499         return 0;
 500 }
 501
 502 static int blkif_ioctl(struct block_device *bdev, fmode_t mode,
 503                        unsigned command, unsigned long argument)
 504 {
 505         struct blkfront_info *info = bdev->bd_disk->private_data;
 506         int i;
 507
 508         dev_dbg(&info->xbdev->dev, "command: 0x%x, argument: 0x%lx\n",
 509                 command, (long)argument);
 510
 511         switch (command) {
 512         case CDROMMULTISESSION:
 513                 dev_dbg(&info->xbdev->dev, "FIXME: support multisession CDs later\n");
 514                 for (i = 0; i < sizeof(struct cdrom_multisession); i++)
 515                         if (put_user(0, (char __user *)(argument + i)))
 516                                 return -EFAULT;
 517                 return 0;
 518
 519         case CDROM_GET_CAPABILITY: {
 520                 struct gendisk *gd = info->gd;
 521                 if (gd->flags & GENHD_FL_CD)
 522                         return 0;
 523                 return -EINVAL;
 524         }
 525
 526         default:
 527                 /*printk(KERN_ALERT "ioctl %08x not supported by Xen blkdev\n",
 528                   command);*/
 529                 return -EINVAL; /* same return as native Linux */
 530         }
 531
 532         return 0;
 533 }
 534
 535 static unsigned long blkif_ring_get_request(struct blkfront_ring_info *rinfo,
 536                                             struct request *req,
 537                                             struct blkif_request **ring_req)
 538 {
 539         unsigned long id;
 540
 541         *ring_req = RING_GET_REQUEST(&rinfo->ring, rinfo->ring.req_prod_pvt);
 542         rinfo->ring.req_prod_pvt++;
 543
 544         id = get_id_from_freelist(rinfo);
 545         rinfo->shadow[id].request = req;
 546         rinfo->shadow[id].status = REQ_WAITING;
 547         rinfo->shadow[id].associated_id = NO_ASSOCIATED_ID;
 548
 549         (*ring_req)->u.rw.id = id;
 550
 551         return id;
 552 }
 553
 554 static int blkif_queue_discard_req(struct request *req, struct blkfront_ring_info *rinfo)
 555 {
 556         struct blkfront_info *info = rinfo->dev_info;
 557         struct blkif_request *ring_req;
 558         unsigned long id;
 559
 560         /* Fill out a communications ring structure. */
 561         id = blkif_ring_get_request(rinfo, req, &ring_req);
 562
 563         ring_req->operation = BLKIF_OP_DISCARD;
 564         ring_req->u.discard.nr_sectors = blk_rq_sectors(req);
 565         ring_req->u.discard.id = id;
 566         ring_req->u.discard.sector_number = (blkif_sector_t)blk_rq_pos(req);
 567         if (req_op(req) == REQ_OP_SECURE_ERASE && info->feature_secdiscard)
 568                 ring_req->u.discard.flag = BLKIF_DISCARD_SECURE;
 569         else
 570                 ring_req->u.discard.flag = 0;
 571
 572         /* Keep a private copy so we can reissue requests when recovering. */
 573         rinfo->shadow[id].req = *ring_req;
 574
 575         return 0;
 576 }
 577
 578 struct setup_rw_req {
 579         unsigned int grant_idx;
 580         struct blkif_request_segment *segments;
 581         struct blkfront_ring_info *rinfo;
 582         struct blkif_request *ring_req;
 583         grant_ref_t gref_head;
 584         unsigned int id;
 585         /* Only used when persistent grant is used and it's a read request */
 586         bool need_copy;
 587         unsigned int bvec_off;
 588         char *bvec_data;
 589
 590         bool require_extra_req;
 591         struct blkif_request *extra_ring_req;
 592 };
 593
 594 static void blkif_setup_rw_req_grant(unsigned long gfn, unsigned int offset,
 595                                      unsigned int len, void *data)
 596 {
 597         struct setup_rw_req *setup = data;
 598         int n, ref;
 599         struct grant *gnt_list_entry;
 600         unsigned int fsect, lsect;
 601         /* Convenient aliases */
 602         unsigned int grant_idx = setup->grant_idx;
 603         struct blkif_request *ring_req = setup->ring_req;
 604         struct blkfront_ring_info *rinfo = setup->rinfo;
 605         /*
 606          * We always use the shadow of the first request to store the list
 607          * of grant associated to the block I/O request. This made the
 608          * completion more easy to handle even if the block I/O request is
 609          * split.
 610          */
 611         struct blk_shadow *shadow = &rinfo->shadow[setup->id];
 612
 613         if (unlikely(setup->require_extra_req &&
 614                      grant_idx >= BLKIF_MAX_SEGMENTS_PER_REQUEST)) {
 615                 /*
 616                  * We are using the second request, setup grant_idx
 617                  * to be the index of the segment array.
 618                  */
 619                 grant_idx -= BLKIF_MAX_SEGMENTS_PER_REQUEST;
 620                 ring_req = setup->extra_ring_req;
 621         }
 622
 623         if ((ring_req->operation == BLKIF_OP_INDIRECT) &&
 624             (grant_idx % GRANTS_PER_INDIRECT_FRAME == 0)) {
 625                 if (setup->segments)
 626                         kunmap_atomic(setup->segments);
 627
 628                 n = grant_idx / GRANTS_PER_INDIRECT_FRAME;
 629                 gnt_list_entry = get_indirect_grant(&setup->gref_head, rinfo);
 630                 shadow->indirect_grants[n] = gnt_list_entry;
 631                 setup->segments = kmap_atomic(gnt_list_entry->page);
 632                 ring_req->u.indirect.indirect_grefs[n] = gnt_list_entry->gref;
 633         }
 634
 635         gnt_list_entry = get_grant(&setup->gref_head, gfn, rinfo);
 636         ref = gnt_list_entry->gref;
 637         /*
 638          * All the grants are stored in the shadow of the first
 639          * request. Therefore we have to use the global index.
 640          */
 641         shadow->grants_used[setup->grant_idx] = gnt_list_entry;
 642
 643         if (setup->need_copy) {
 644                 void *shared_data;
 645
 646                 shared_data = kmap_atomic(gnt_list_entry->page);
 647                 /*
 648                  * this does not wipe data stored outside the
 649                  * range sg->offset..sg->offset+sg->length.
 650                  * Therefore, blkback *could* see data from
 651                  * previous requests. This is OK as long as
 652                  * persistent grants are shared with just one
 653                  * domain. It may need refactoring if this
 654                  * changes
 655                  */
 656                 memcpy(shared_data + offset,
 657                        setup->bvec_data + setup->bvec_off,
 658                        len);
 659
 660                 kunmap_atomic(shared_data);
 661                 setup->bvec_off += len;
 662         }
 663
 664         fsect = offset >> 9;
 665         lsect = fsect + (len >> 9) - 1;
 666         if (ring_req->operation != BLKIF_OP_INDIRECT) {
 667                 ring_req->u.rw.seg[grant_idx] =
 668                         (struct blkif_request_segment) {
 669                                 .gref       = ref,
 670                                 .first_sect = fsect,
 671                                 .last_sect  = lsect };
 672         } else {
 673                 setup->segments[grant_idx % GRANTS_PER_INDIRECT_FRAME] =
 674                         (struct blkif_request_segment) {
 675                                 .gref       = ref,
 676                                 .first_sect = fsect,
 677                                 .last_sect  = lsect };
 678         }
 679
 680         (setup->grant_idx)++;
 681 }
 682
 683 static void blkif_setup_extra_req(struct blkif_request *first,
 684                                   struct blkif_request *second)
 685 {
 686         uint16_t nr_segments = first->u.rw.nr_segments;
 687
 688         /*
 689          * The second request is only present when the first request uses
 690          * all its segments. It's always the continuity of the first one.
 691          */
 692         first->u.rw.nr_segments = BLKIF_MAX_SEGMENTS_PER_REQUEST;
 693
 694         second->u.rw.nr_segments = nr_segments - BLKIF_MAX_SEGMENTS_PER_REQUEST;
 695         second->u.rw.sector_number = first->u.rw.sector_number +
 696                 (BLKIF_MAX_SEGMENTS_PER_REQUEST * XEN_PAGE_SIZE) / 512;
 697
 698         second->u.rw.handle = first->u.rw.handle;
 699         second->operation = first->operation;
 700 }
 701
 702 static int blkif_queue_rw_req(struct request *req, struct blkfront_ring_info *rinfo)
 703 {
 704         struct blkfront_info *info = rinfo->dev_info;
 705         struct blkif_request *ring_req, *extra_ring_req = NULL;
 706         unsigned long id, extra_id = NO_ASSOCIATED_ID;
 707         bool require_extra_req = false;
 708         int i;
 709         struct setup_rw_req setup = {
 710                 .grant_idx = 0,
 711                 .segments = NULL,
 712                 .rinfo = rinfo,
 713                 .need_copy = rq_data_dir(req) && info->feature_persistent,
 714         };
 715
 716         /*
 717          * Used to store if we are able to queue the request by just using
 718          * existing persistent grants, or if we have to get new grants,
 719          * as there are not sufficiently many free.
 720          */
 721         bool new_persistent_gnts = false;
 722         struct scatterlist *sg;
 723         int num_sg, max_grefs, num_grant;
 724
 725         max_grefs = req->nr_phys_segments * GRANTS_PER_PSEG;
 726         if (max_grefs > BLKIF_MAX_SEGMENTS_PER_REQUEST)
 727                 /*
 728                  * If we are using indirect segments we need to account
 729                  * for the indirect grefs used in the request.
 730                  */
 731                 max_grefs += INDIRECT_GREFS(max_grefs);
 732
 733         /* Check if we have enough persistent grants to allocate a requests */
 734         if (rinfo->persistent_gnts_c < max_grefs) {
 735                 new_persistent_gnts = true;
 736
 737                 if (gnttab_alloc_grant_references(
 738                     max_grefs - rinfo->persistent_gnts_c,
 739                     &setup.gref_head) < 0) {
 740                         gnttab_request_free_callback(
 741                                 &rinfo->callback,
 742                                 blkif_restart_queue_callback,
 743                                 rinfo,
 744                                 max_grefs - rinfo->persistent_gnts_c);
 745                         return 1;
 746                 }
 747         }
 748
 749         /* Fill out a communications ring structure. */
 750         id = blkif_ring_get_request(rinfo, req, &ring_req);
 751
 752         num_sg = blk_rq_map_sg(req->q, req, rinfo->shadow[id].sg);
 753         num_grant = 0;
 754         /* Calculate the number of grant used */
 755         for_each_sg(rinfo->shadow[id].sg, sg, num_sg, i)
 756                num_grant += gnttab_count_grant(sg->offset, sg->length);
 757
 758         require_extra_req = info->max_indirect_segments == 0 &&
 759                 num_grant > BLKIF_MAX_SEGMENTS_PER_REQUEST;
 760         BUG_ON(!HAS_EXTRA_REQ && require_extra_req);
 761
 762         rinfo->shadow[id].num_sg = num_sg;
 763         if (num_grant > BLKIF_MAX_SEGMENTS_PER_REQUEST &&
 764             likely(!require_extra_req)) {
 765                 /*
 766                  * The indirect operation can only be a BLKIF_OP_READ or
 767                  * BLKIF_OP_WRITE
 768                  */
 769                 BUG_ON(req_op(req) == REQ_OP_FLUSH || req->cmd_flags & REQ_FUA);
 770                 ring_req->operation = BLKIF_OP_INDIRECT;
 771                 ring_req->u.indirect.indirect_op = rq_data_dir(req) ?
 772                         BLKIF_OP_WRITE : BLKIF_OP_READ;
 773                 ring_req->u.indirect.sector_number = (blkif_sector_t)blk_rq_pos(req);
 774                 ring_req->u.indirect.handle = info->handle;
 775                 ring_req->u.indirect.nr_segments = num_grant;
 776         } else {
 777                 ring_req->u.rw.sector_number = (blkif_sector_t)blk_rq_pos(req);
 778                 ring_req->u.rw.handle = info->handle;
 779                 ring_req->operation = rq_data_dir(req) ?
 780                         BLKIF_OP_WRITE : BLKIF_OP_READ;
 781                 if (req_op(req) == REQ_OP_FLUSH || req->cmd_flags & REQ_FUA) {
 782                         /*
 783                          * Ideally we can do an unordered flush-to-disk.
 784                          * In case the backend onlysupports barriers, use that.
 785                          * A barrier request a superset of FUA, so we can
 786                          * implement it the same way.  (It's also a FLUSH+FUA,
 787                          * since it is guaranteed ordered WRT previous writes.)
 788                          */
 789                         if (info->feature_flush && info->feature_fua)
 790                                 ring_req->operation =
 791                                         BLKIF_OP_WRITE_BARRIER;
 792                         else if (info->feature_flush)
 793                                 ring_req->operation =
 794                                         BLKIF_OP_FLUSH_DISKCACHE;
 795                         else
 796                                 ring_req->operation = 0;
 797                 }
 798                 ring_req->u.rw.nr_segments = num_grant;
 799                 if (unlikely(require_extra_req)) {
 800                         extra_id = blkif_ring_get_request(rinfo, req,
 801                                                           &extra_ring_req);
 802                         /*
 803                          * Only the first request contains the scatter-gather
 804                          * list.
 805                          */
 806                         rinfo->shadow[extra_id].num_sg = 0;
 807
 808                         blkif_setup_extra_req(ring_req, extra_ring_req);
 809
 810                         /* Link the 2 requests together */
 811                         rinfo->shadow[extra_id].associated_id = id;
 812                         rinfo->shadow[id].associated_id = extra_id;
 813                 }
 814         }
 815
 816         setup.ring_req = ring_req;
 817         setup.id = id;
 818
 819         setup.require_extra_req = require_extra_req;
 820         if (unlikely(require_extra_req))
 821                 setup.extra_ring_req = extra_ring_req;
 822
 823         for_each_sg(rinfo->shadow[id].sg, sg, num_sg, i) {
 824                 BUG_ON(sg->offset + sg->length > PAGE_SIZE);
 825
 826                 if (setup.need_copy) {
 827                         setup.bvec_off = sg->offset;
 828                         setup.bvec_data = kmap_atomic(sg_page(sg));
 829                 }
 830
 831                 gnttab_foreach_grant_in_range(sg_page(sg),
 832                                               sg->offset,
 833                                               sg->length,
 834                                               blkif_setup_rw_req_grant,
 835                                               &setup);
 836
 837                 if (setup.need_copy)
 838                         kunmap_atomic(setup.bvec_data);
 839         }
 840         if (setup.segments)
 841                 kunmap_atomic(setup.segments);
 842
 843         /* Keep a private copy so we can reissue requests when recovering. */
 844         rinfo->shadow[id].req = *ring_req;
 845         if (unlikely(require_extra_req))
 846                 rinfo->shadow[extra_id].req = *extra_ring_req;
 847
 848         if (new_persistent_gnts)
 849                 gnttab_free_grant_references(setup.gref_head);
 850
 851         return 0;
 852 }
 853
 854 /*
 855  * Generate a Xen blkfront IO request from a blk layer request.  Reads
 856  * and writes are handled as expected.
 857  *
 858  * @req: a request struct
 859  */
 860 static int blkif_queue_request(struct request *req, struct blkfront_ring_info *rinfo)
 861 {
 862         if (unlikely(rinfo->dev_info->connected != BLKIF_STATE_CONNECTED))
 863                 return 1;
 864
 865         if (unlikely(req_op(req) == REQ_OP_DISCARD ||
 866                      req_op(req) == REQ_OP_SECURE_ERASE))
 867                 return blkif_queue_discard_req(req, rinfo);
 868         else
 869                 return blkif_queue_rw_req(req, rinfo);
 870 }
 871
 872 static inline void flush_requests(struct blkfront_ring_info *rinfo)
 873 {
 874         int notify;
 875
 876         RING_PUSH_REQUESTS_AND_CHECK_NOTIFY(&rinfo->ring, notify);
 877
 878         if (notify)
 879                 notify_remote_via_irq(rinfo->irq);
 880 }
 881
 882 static inline bool blkif_request_flush_invalid(struct request *req,
 883                                                struct blkfront_info *info)
 884 {
 885         return (blk_rq_is_passthrough(req) ||
 886                 ((req_op(req) == REQ_OP_FLUSH) &&
 887                  !info->feature_flush) ||
 888                 ((req->cmd_flags & REQ_FUA) &&
 889                  !info->feature_fua));
 890 }
 891
 892 static blk_status_t blkif_queue_rq(struct blk_mq_hw_ctx *hctx,
 893                           const struct blk_mq_queue_data *qd)
 894 {
 895         unsigned long flags;
 896         int qid = hctx->queue_num;
 897         struct blkfront_info *info = hctx->queue->queuedata;
 898         struct blkfront_ring_info *rinfo = NULL;
 899
 900         rinfo = get_rinfo(info, qid);
 901         blk_mq_start_request(qd->rq);
 902         spin_lock_irqsave(&rinfo->ring_lock, flags);
 903         if (RING_FULL(&rinfo->ring))
 904                 goto out_busy;
 905
 906         if (blkif_request_flush_invalid(qd->rq, rinfo->dev_info))
 907                 goto out_err;
 908
 909         if (blkif_queue_request(qd->rq, rinfo))
 910                 goto out_busy;
 911
 912         flush_requests(rinfo);
 913         spin_unlock_irqrestore(&rinfo->ring_lock, flags);
 914         return BLK_STS_OK;
 915
 916 out_err:
 917         spin_unlock_irqrestore(&rinfo->ring_lock, flags);
 918         return BLK_STS_IOERR;
 919
 920 out_busy:
 921         blk_mq_stop_hw_queue(hctx);
 922         spin_unlock_irqrestore(&rinfo->ring_lock, flags);
 923         return BLK_STS_DEV_RESOURCE;
 924 }
 925
 926 static void blkif_complete_rq(struct request *rq)
 927 {
 928         blk_mq_end_request(rq, blkif_req(rq)->error);
 929 }
 930
 931 static const struct blk_mq_ops blkfront_mq_ops = {
 932         .queue_rq = blkif_queue_rq,
 933         .complete = blkif_complete_rq,
 934 };
 935
 936 static void blkif_set_queue_limits(struct blkfront_info *info)
 937 {
 938         struct request_queue *rq = info->rq;
 939         struct gendisk *gd = info->gd;
 940         unsigned int segments = info->max_indirect_segments ? :
 941                                 BLKIF_MAX_SEGMENTS_PER_REQUEST;
 942
 943         blk_queue_flag_set(QUEUE_FLAG_VIRT, rq);
 944
 945         if (info->feature_discard) {
 946                 blk_queue_flag_set(QUEUE_FLAG_DISCARD, rq);
 947                 blk_queue_max_discard_sectors(rq, get_capacity(gd));
 948                 rq->limits.discard_granularity = info->discard_granularity;
 949                 rq->limits.discard_alignment = info->discard_alignment;
 950                 if (info->feature_secdiscard)
 951                         blk_queue_flag_set(QUEUE_FLAG_SECERASE, rq);
 952         }
 953
 954         /* Hard sector size and max sectors impersonate the equiv. hardware. */
 955         blk_queue_logical_block_size(rq, info->sector_size);
 956         blk_queue_physical_block_size(rq, info->physical_sector_size);
 957         blk_queue_max_hw_sectors(rq, (segments * XEN_PAGE_SIZE) / 512);
 958
 959         /* Each segment in a request is up to an aligned page in size. */
 960         blk_queue_segment_boundary(rq, PAGE_SIZE - 1);
 961         blk_queue_max_segment_size(rq, PAGE_SIZE);
 962
 963         /* Ensure a merged request will fit in a single I/O ring slot. */
 964         blk_queue_max_segments(rq, segments / GRANTS_PER_PSEG);
 965
 966         /* Make sure buffer addresses are sector-aligned. */
 967         blk_queue_dma_alignment(rq, 511);
 968 }
 969
 970 static int xlvbd_init_blk_queue(struct gendisk *gd, u16 sector_size,
 971                                 unsigned int physical_sector_size)
 972 {
 973         struct request_queue *rq;
 974         struct blkfront_info *info = gd->private_data;
 975
 976         memset(&info->tag_set, 0, sizeof(info->tag_set));
 977         info->tag_set.ops = &blkfront_mq_ops;
 978         info->tag_set.nr_hw_queues = info->nr_rings;
 979         if (HAS_EXTRA_REQ && info->max_indirect_segments == 0) {
 980                 /*
 981                  * When indirect descriptior is not supported, the I/O request
 982                  * will be split between multiple request in the ring.
 983                  * To avoid problems when sending the request, divide by
 984                  * 2 the depth of the queue.
 985                  */
 986                 info->tag_set.queue_depth =  BLK_RING_SIZE(info) / 2;
 987         } else
 988                 info->tag_set.queue_depth = BLK_RING_SIZE(info);
 989         info->tag_set.numa_node = NUMA_NO_NODE;
 990         info->tag_set.flags = BLK_MQ_F_SHOULD_MERGE;
 991         info->tag_set.cmd_size = sizeof(struct blkif_req);
 992         info->tag_set.driver_data = info;
 993
 994         if (blk_mq_alloc_tag_set(&info->tag_set))
 995                 return -EINVAL;
 996         rq = blk_mq_init_queue(&info->tag_set);
 997         if (IS_ERR(rq)) {
 998                 blk_mq_free_tag_set(&info->tag_set);
 999                 return PTR_ERR(rq);
1000         }
1001
1002         rq->queuedata = info;
1003         info->rq = gd->queue = rq;
1004         info->gd = gd;
1005         info->sector_size = sector_size;
1006         info->physical_sector_size = physical_sector_size;
1007         blkif_set_queue_limits(info);
1008
1009         return 0;
1010 }
1011
1012 static const char *flush_info(struct blkfront_info *info)
1013 {
1014         if (info->feature_flush && info->feature_fua)
1015                 return "barrier: enabled;";
1016         else if (info->feature_flush)
1017                 return "flush diskcache: enabled;";
1018         else
1019                 return "barrier or flush: disabled;";
1020 }
1021
1022 static void xlvbd_flush(struct blkfront_info *info)
1023 {
1024         blk_queue_write_cache(info->rq, info->feature_flush ? true : false,
1025                               info->feature_fua ? true : false);
1026         pr_info("blkfront: %s: %s %s %s %s %s\n",
1027                 info->gd->disk_name, flush_info(info),
1028                 "persistent grants:", info->feature_persistent ?
1029                 "enabled;" : "disabled;", "indirect descriptors:",
1030                 info->max_indirect_segments ? "enabled;" : "disabled;");
1031 }
1032
1033 static int xen_translate_vdev(int vdevice, int *minor, unsigned int *offset)
1034 {
1035         int major;
1036         major = BLKIF_MAJOR(vdevice);
1037         *minor = BLKIF_MINOR(vdevice);
1038         switch (major) {
1039                 case XEN_IDE0_MAJOR:
1040                         *offset = (*minor / 64) + EMULATED_HD_DISK_NAME_OFFSET;
1041                         *minor = ((*minor / 64) * PARTS_PER_DISK) +
1042                                 EMULATED_HD_DISK_MINOR_OFFSET;
1043                         break;
1044                 case XEN_IDE1_MAJOR:
1045                         *offset = (*minor / 64) + 2 + EMULATED_HD_DISK_NAME_OFFSET;
1046                         *minor = (((*minor / 64) + 2) * PARTS_PER_DISK) +
1047                                 EMULATED_HD_DISK_MINOR_OFFSET;
1048                         break;
1049                 case XEN_SCSI_DISK0_MAJOR:
1050                         *offset = (*minor / PARTS_PER_DISK) + EMULATED_SD_DISK_NAME_OFFSET;
1051                         *minor = *minor + EMULATED_SD_DISK_MINOR_OFFSET;
1052                         break;
1053                 case XEN_SCSI_DISK1_MAJOR:
1054                 case XEN_SCSI_DISK2_MAJOR:
1055                 case XEN_SCSI_DISK3_MAJOR:
1056                 case XEN_SCSI_DISK4_MAJOR:
1057                 case XEN_SCSI_DISK5_MAJOR:
1058                 case XEN_SCSI_DISK6_MAJOR:
1059                 case XEN_SCSI_DISK7_MAJOR:
1060                         *offset = (*minor / PARTS_PER_DISK) +
1061                                 ((major - XEN_SCSI_DISK1_MAJOR + 1) * 16) +
1062                                 EMULATED_SD_DISK_NAME_OFFSET;
1063                         *minor = *minor +
1064                                 ((major - XEN_SCSI_DISK1_MAJOR + 1) * 16 * PARTS_PER_DISK) +
1065                                 EMULATED_SD_DISK_MINOR_OFFSET;
1066                         break;
1067                 case XEN_SCSI_DISK8_MAJOR:
1068                 case XEN_SCSI_DISK9_MAJOR:
1069                 case XEN_SCSI_DISK10_MAJOR:
1070                 case XEN_SCSI_DISK11_MAJOR:
1071                 case XEN_SCSI_DISK12_MAJOR:
1072                 case XEN_SCSI_DISK13_MAJOR:
1073                 case XEN_SCSI_DISK14_MAJOR:
1074                 case XEN_SCSI_DISK15_MAJOR:
1075                         *offset = (*minor / PARTS_PER_DISK) +
1076                                 ((major - XEN_SCSI_DISK8_MAJOR + 8) * 16) +
1077                                 EMULATED_SD_DISK_NAME_OFFSET;
1078                         *minor = *minor +
1079                                 ((major - XEN_SCSI_DISK8_MAJOR + 8) * 16 * PARTS_PER_DISK) +
1080                                 EMULATED_SD_DISK_MINOR_OFFSET;
1081                         break;
1082                 case XENVBD_MAJOR:
1083                         *offset = *minor / PARTS_PER_DISK;
1084                         break;
1085                 default:
1086                         printk(KERN_WARNING "blkfront: your disk configuration is "
1087                                         "incorrect, please use an xvd device instead\n");
1088                         return -ENODEV;
1089         }
1090         return 0;
1091 }
1092
1093 static char *encode_disk_name(char *ptr, unsigned int n)
1094 {
1095         if (n >= 26)
1096                 ptr = encode_disk_name(ptr, n / 26 - 1);
1097         *ptr = 'a' + n % 26;
1098         return ptr + 1;
1099 }
1100
1101 static int xlvbd_alloc_gendisk(blkif_sector_t capacity,
1102                                struct blkfront_info *info,
1103                                u16 vdisk_info, u16 sector_size,
1104                                unsigned int physical_sector_size)
1105 {
1106         struct gendisk *gd;
1107         int nr_minors = 1;
1108         int err;
1109         unsigned int offset;
1110         int minor;
1111         int nr_parts;
1112         char *ptr;
1113
1114         BUG_ON(info->gd != NULL);
1115         BUG_ON(info->rq != NULL);
1116
1117         if ((info->vdevice>>EXT_SHIFT) > 1) {
1118                 /* this is above the extended range; something is wrong */
1119                 printk(KERN_WARNING "blkfront: vdevice 0x%x is above the extended range; ignoring\n", info->vdevice);
1120                 return -ENODEV;
1121         }
1122
1123         if (!VDEV_IS_EXTENDED(info->vdevice)) {
1124                 err = xen_translate_vdev(info->vdevice, &minor, &offset);
1125                 if (err)
1126                         return err;
1127                 nr_parts = PARTS_PER_DISK;
1128         } else {
1129                 minor = BLKIF_MINOR_EXT(info->vdevice);
1130                 nr_parts = PARTS_PER_EXT_DISK;
1131                 offset = minor / nr_parts;
1132                 if (xen_hvm_domain() && offset < EMULATED_HD_DISK_NAME_OFFSET + 4)
1133                         printk(KERN_WARNING "blkfront: vdevice 0x%x might conflict with "
1134                                         "emulated IDE disks,\n\t choose an xvd device name"
1135                                         "from xvde on\n", info->vdevice);
1136         }
1137         if (minor >> MINORBITS) {
1138                 pr_warn("blkfront: %#x's minor (%#x) out of range; ignoring\n",
1139                         info->vdevice, minor);
1140                 return -ENODEV;
1141         }
1142
1143         if ((minor % nr_parts) == 0)
1144                 nr_minors = nr_parts;
1145
1146         err = xlbd_reserve_minors(minor, nr_minors);
1147         if (err)
1148                 goto out;
1149         err = -ENODEV;
1150
1151         gd = alloc_disk(nr_minors);
1152         if (gd == NULL)
1153                 goto release;
1154
1155         strcpy(gd->disk_name, DEV_NAME);
1156         ptr = encode_disk_name(gd->disk_name + sizeof(DEV_NAME) - 1, offset);
1157         BUG_ON(ptr >= gd->disk_name + DISK_NAME_LEN);
1158         if (nr_minors > 1)
1159                 *ptr = 0;
1160         else
1161                 snprintf(ptr, gd->disk_name + DISK_NAME_LEN - ptr,
1162                          "%d", minor & (nr_parts - 1));
1163
1164         gd->major = XENVBD_MAJOR;
1165         gd->first_minor = minor;
1166         gd->fops = &xlvbd_block_fops;
1167         gd->private_data = info;
1168         set_capacity(gd, capacity);
1169
1170         if (xlvbd_init_blk_queue(gd, sector_size, physical_sector_size)) {
1171                 del_gendisk(gd);
1172                 goto release;
1173         }
1174
1175         xlvbd_flush(info);
1176
1177         if (vdisk_info & VDISK_READONLY)
1178                 set_disk_ro(gd, 1);
1179
1180         if (vdisk_info & VDISK_REMOVABLE)
1181                 gd->flags |= GENHD_FL_REMOVABLE;
1182
1183         if (vdisk_info & VDISK_CDROM)
1184                 gd->flags |= GENHD_FL_CD;
1185
1186         return 0;
1187
1188  release:
1189         xlbd_release_minors(minor, nr_minors);
1190  out:
1191         return err;
1192 }
1193
1194 static void xlvbd_release_gendisk(struct blkfront_info *info)
1195 {
1196         unsigned int minor, nr_minors, i;
1197         struct blkfront_ring_info *rinfo;
1198
1199         if (info->rq == NULL)
1200                 return;
1201
1202         /* No more blkif_request(). */
1203         blk_mq_stop_hw_queues(info->rq);
1204
1205         for_each_rinfo(info, rinfo, i) {
1206                 /* No more gnttab callback work. */
1207                 gnttab_cancel_free_callback(&rinfo->callback);
1208
1209                 /* Flush gnttab callback work. Must be done with no locks held. */
1210                 flush_work(&rinfo->work);
1211         }
1212
1213         del_gendisk(info->gd);
1214
1215         minor = info->gd->first_minor;
1216         nr_minors = info->gd->minors;
1217         xlbd_release_minors(minor, nr_minors);
1218
1219         blk_cleanup_queue(info->rq);
1220         blk_mq_free_tag_set(&info->tag_set);
1221         info->rq = NULL;
1222
1223         put_disk(info->gd);
1224         info->gd = NULL;
1225 }
1226
1227 /* Already hold rinfo->ring_lock. */
1228 static inline void kick_pending_request_queues_locked(struct blkfront_ring_info *rinfo)
1229 {
1230         if (!RING_FULL(&rinfo->ring))
1231                 blk_mq_start_stopped_hw_queues(rinfo->dev_info->rq, true);
1232 }
1233
1234 static void kick_pending_request_queues(struct blkfront_ring_info *rinfo)
1235 {
1236         unsigned long flags;
1237
1238         spin_lock_irqsave(&rinfo->ring_lock, flags);
1239         kick_pending_request_queues_locked(rinfo);
1240         spin_unlock_irqrestore(&rinfo->ring_lock, flags);
1241 }
1242
1243 static void blkif_restart_queue(struct work_struct *work)
1244 {
1245         struct blkfront_ring_info *rinfo = container_of(work, struct blkfront_ring_info, work);
1246
1247         if (rinfo->dev_info->connected == BLKIF_STATE_CONNECTED)
1248                 kick_pending_request_queues(rinfo);
1249 }
1250
1251 static void blkif_free_ring(struct blkfront_ring_info *rinfo)
1252 {
1253         struct grant *persistent_gnt, *n;
1254         struct blkfront_info *info = rinfo->dev_info;
1255         int i, j, segs;
1256
1257         /*
1258          * Remove indirect pages, this only happens when using indirect
1259          * descriptors but not persistent grants
1260          */
1261         if (!list_empty(&rinfo->indirect_pages)) {
1262                 struct page *indirect_page, *n;
1263
1264                 BUG_ON(info->feature_persistent);
1265                 list_for_each_entry_safe(indirect_page, n, &rinfo->indirect_pages, lru) {
1266                         list_del(&indirect_page->lru);
1267                         __free_page(indirect_page);
1268                 }
1269         }
1270
1271         /* Remove all persistent grants. */
1272         if (!list_empty(&rinfo->grants)) {
1273                 list_for_each_entry_safe(persistent_gnt, n,
1274                                          &rinfo->grants, node) {
1275                         list_del(&persistent_gnt->node);
1276                         if (persistent_gnt->gref != GRANT_INVALID_REF) {
1277                                 gnttab_end_foreign_access(persistent_gnt->gref,
1278                                                           0, 0UL);
1279                                 rinfo->persistent_gnts_c--;
1280                         }
1281                         if (info->feature_persistent)
1282                                 __free_page(persistent_gnt->page);
1283                         kfree(persistent_gnt);
1284                 }
1285         }
1286         BUG_ON(rinfo->persistent_gnts_c != 0);
1287
1288         for (i = 0; i < BLK_RING_SIZE(info); i++) {
1289                 /*
1290                  * Clear persistent grants present in requests already
1291                  * on the shared ring
1292                  */
1293                 if (!rinfo->shadow[i].request)
1294                         goto free_shadow;
1295
1296                 segs = rinfo->shadow[i].req.operation == BLKIF_OP_INDIRECT ?
1297                        rinfo->shadow[i].req.u.indirect.nr_segments :
1298                        rinfo->shadow[i].req.u.rw.nr_segments;
1299                 for (j = 0; j < segs; j++) {
1300                         persistent_gnt = rinfo->shadow[i].grants_used[j];
1301                         gnttab_end_foreign_access(persistent_gnt->gref, 0, 0UL);
1302                         if (info->feature_persistent)
1303                                 __free_page(persistent_gnt->page);
1304                         kfree(persistent_gnt);
1305                 }
1306
1307                 if (rinfo->shadow[i].req.operation != BLKIF_OP_INDIRECT)
1308                         /*
1309                          * If this is not an indirect operation don't try to
1310                          * free indirect segments
1311                          */
1312                         goto free_shadow;
1313
1314                 for (j = 0; j < INDIRECT_GREFS(segs); j++) {
1315                         persistent_gnt = rinfo->shadow[i].indirect_grants[j];
1316                         gnttab_end_foreign_access(persistent_gnt->gref, 0, 0UL);
1317                         __free_page(persistent_gnt->page);
1318                         kfree(persistent_gnt);
1319                 }
1320
1321 free_shadow:
1322                 kvfree(rinfo->shadow[i].grants_used);
1323                 rinfo->shadow[i].grants_used = NULL;
1324                 kvfree(rinfo->shadow[i].indirect_grants);
1325                 rinfo->shadow[i].indirect_grants = NULL;
1326                 kvfree(rinfo->shadow[i].sg);
1327                 rinfo->shadow[i].sg = NULL;
1328         }
1329
1330         /* No more gnttab callback work. */
1331         gnttab_cancel_free_callback(&rinfo->callback);
1332
1333         /* Flush gnttab callback work. Must be done with no locks held. */
1334         flush_work(&rinfo->work);
1335
1336         /* Free resources associated with old device channel. */
1337         for (i = 0; i < info->nr_ring_pages; i++) {
1338                 if (rinfo->ring_ref[i] != GRANT_INVALID_REF) {
1339                         gnttab_end_foreign_access(rinfo->ring_ref[i], 0, 0);
1340                         rinfo->ring_ref[i] = GRANT_INVALID_REF;
1341                 }
1342         }
1343         free_pages((unsigned long)rinfo->ring.sring, get_order(info->nr_ring_pages * XEN_PAGE_SIZE));
1344         rinfo->ring.sring = NULL;
1345
1346         if (rinfo->irq)
1347                 unbind_from_irqhandler(rinfo->irq, rinfo);
1348         rinfo->evtchn = rinfo->irq = 0;
1349 }
1350
1351 static void blkif_free(struct blkfront_info *info, int suspend)
1352 {
1353         unsigned int i;
1354         struct blkfront_ring_info *rinfo;
1355
1356         /* Prevent new requests being issued until we fix things up. */
1357         info->connected = suspend ?
1358                 BLKIF_STATE_SUSPENDED : BLKIF_STATE_DISCONNECTED;
1359         /* No more blkif_request(). */
1360         if (info->rq)
1361                 blk_mq_stop_hw_queues(info->rq);
1362
1363         for_each_rinfo(info, rinfo, i)
1364                 blkif_free_ring(rinfo);
1365
1366         kvfree(info->rinfo);
1367         info->rinfo = NULL;
1368         info->nr_rings = 0;
1369 }
1370
1371 struct copy_from_grant {
1372         const struct blk_shadow *s;
1373         unsigned int grant_idx;
1374         unsigned int bvec_offset;
1375         char *bvec_data;
1376 };
1377
1378 static void blkif_copy_from_grant(unsigned long gfn, unsigned int offset,
1379                                   unsigned int len, void *data)
1380 {
1381         struct copy_from_grant *info = data;
1382         char *shared_data;
1383         /* Convenient aliases */
1384         const struct blk_shadow *s = info->s;
1385
1386         shared_data = kmap_atomic(s->grants_used[info->grant_idx]->page);
1387
1388         memcpy(info->bvec_data + info->bvec_offset,
1389                shared_data + offset, len);
1390
1391         info->bvec_offset += len;
1392         info->grant_idx++;
1393
1394         kunmap_atomic(shared_data);
1395 }
1396
1397 static enum blk_req_status blkif_rsp_to_req_status(int rsp)
1398 {
1399         switch (rsp)
1400         {
1401         case BLKIF_RSP_OKAY:
1402                 return REQ_DONE;
1403         case BLKIF_RSP_EOPNOTSUPP:
1404                 return REQ_EOPNOTSUPP;
1405         case BLKIF_RSP_ERROR:
1406         default:
1407                 return REQ_ERROR;
1408         }
1409 }
1410
1411 /*
1412  * Get the final status of the block request based on two ring response
1413  */
1414 static int blkif_get_final_status(enum blk_req_status s1,
1415                                   enum blk_req_status s2)
1416 {
1417         BUG_ON(s1 == REQ_WAITING);
1418         BUG_ON(s2 == REQ_WAITING);
1419
1420         if (s1 == REQ_ERROR || s2 == REQ_ERROR)
1421                 return BLKIF_RSP_ERROR;
1422         else if (s1 == REQ_EOPNOTSUPP || s2 == REQ_EOPNOTSUPP)
1423                 return BLKIF_RSP_EOPNOTSUPP;
1424         return BLKIF_RSP_OKAY;
1425 }
1426
1427 static bool blkif_completion(unsigned long *id,
1428                              struct blkfront_ring_info *rinfo,
1429                              struct blkif_response *bret)
1430 {
1431         int i = 0;
1432         struct scatterlist *sg;
1433         int num_sg, num_grant;
1434         struct blkfront_info *info = rinfo->dev_info;
1435         struct blk_shadow *s = &rinfo->shadow[*id];
1436         struct copy_from_grant data = {
1437                 .grant_idx = 0,
1438         };
1439
1440         num_grant = s->req.operation == BLKIF_OP_INDIRECT ?
1441                 s->req.u.indirect.nr_segments : s->req.u.rw.nr_segments;
1442
1443         /* The I/O request may be split in two. */
1444         if (unlikely(s->associated_id != NO_ASSOCIATED_ID)) {
1445                 struct blk_shadow *s2 = &rinfo->shadow[s->associated_id];
1446
1447                 /* Keep the status of the current response in shadow. */
1448                 s->status = blkif_rsp_to_req_status(bret->status);
1449
1450                 /* Wait the second response if not yet here. */
1451                 if (s2->status == REQ_WAITING)
1452                         return false;
1453
1454                 bret->status = blkif_get_final_status(s->status,
1455                                                       s2->status);
1456
1457                 /*
1458                  * All the grants is stored in the first shadow in order
1459                  * to make the completion code simpler.
1460                  */
1461                 num_grant += s2->req.u.rw.nr_segments;
1462
1463                 /*
1464                  * The two responses may not come in order. Only the
1465                  * first request will store the scatter-gather list.
1466                  */
1467                 if (s2->num_sg != 0) {
1468                         /* Update "id" with the ID of the first response. */
1469                         *id = s->associated_id;
1470                         s = s2;
1471                 }
1472
1473                 /*
1474                  * We don't need anymore the second request, so recycling
1475                  * it now.
1476                  */
1477                 if (add_id_to_freelist(rinfo, s->associated_id))
1478                         WARN(1, "%s: can't recycle the second part (id = %ld) of the request\n",
1479                              info->gd->disk_name, s->associated_id);
1480         }
1481
1482         data.s = s;
1483         num_sg = s->num_sg;
1484
1485         if (bret->operation == BLKIF_OP_READ && info->feature_persistent) {
1486                 for_each_sg(s->sg, sg, num_sg, i) {
1487                         BUG_ON(sg->offset + sg->length > PAGE_SIZE);
1488
1489                         data.bvec_offset = sg->offset;
1490                         data.bvec_data = kmap_atomic(sg_page(sg));
1491
1492                         gnttab_foreach_grant_in_range(sg_page(sg),
1493                                                       sg->offset,
1494                                                       sg->length,
1495                                                       blkif_copy_from_grant,
1496                                                       &data);
1497
1498                         kunmap_atomic(data.bvec_data);
1499                 }
1500         }
1501         /* Add the persistent grant into the list of free grants */
1502         for (i = 0; i < num_grant; i++) {
1503                 if (gnttab_query_foreign_access(s->grants_used[i]->gref)) {
1504                         /*
1505                          * If the grant is still mapped by the backend (the
1506                          * backend has chosen to make this grant persistent)
1507                          * we add it at the head of the list, so it will be
1508                          * reused first.
1509                          */
1510                         if (!info->feature_persistent)
1511                                 pr_alert_ratelimited("backed has not unmapped grant: %u\n",
1512                                                      s->grants_used[i]->gref);
1513                         list_add(&s->grants_used[i]->node, &rinfo->grants);
1514                         rinfo->persistent_gnts_c++;
1515                 } else {
1516                         /*
1517                          * If the grant is not mapped by the backend we end the
1518                          * foreign access and add it to the tail of the list,
1519                          * so it will not be picked again unless we run out of
1520                          * persistent grants.
1521                          */
1522                         gnttab_end_foreign_access(s->grants_used[i]->gref, 0, 0UL);
1523                         s->grants_used[i]->gref = GRANT_INVALID_REF;
1524                         list_add_tail(&s->grants_used[i]->node, &rinfo->grants);
1525                 }
1526         }
1527         if (s->req.operation == BLKIF_OP_INDIRECT) {
1528                 for (i = 0; i < INDIRECT_GREFS(num_grant); i++) {
1529                         if (gnttab_query_foreign_access(s->indirect_grants[i]->gref)) {
1530                                 if (!info->feature_persistent)
1531                                         pr_alert_ratelimited("backed has not unmapped grant: %u\n",
1532                                                              s->indirect_grants[i]->gref);
1533                                 list_add(&s->indirect_grants[i]->node, &rinfo->grants);
1534                                 rinfo->persistent_gnts_c++;
1535                         } else {
1536                                 struct page *indirect_page;
1537
1538                                 gnttab_end_foreign_access(s->indirect_grants[i]->gref, 0, 0UL);
1539                                 /*
1540                                  * Add the used indirect page back to the list of
1541                                  * available pages for indirect grefs.
1542                                  */
1543                                 if (!info->feature_persistent) {
1544                                         indirect_page = s->indirect_grants[i]->page;
1545                                         list_add(&indirect_page->lru, &rinfo->indirect_pages);
1546                                 }
1547                                 s->indirect_grants[i]->gref = GRANT_INVALID_REF;
1548                                 list_add_tail(&s->indirect_grants[i]->node, &rinfo->grants);
1549                         }
1550                 }
1551         }
1552
1553         return true;
1554 }
1555
1556 static irqreturn_t blkif_interrupt(int irq, void *dev_id)
1557 {
1558         struct request *req;
1559         struct blkif_response *bret;
1560         RING_IDX i, rp;
1561         unsigned long flags;
1562         struct blkfront_ring_info *rinfo = (struct blkfront_ring_info *)dev_id;
1563         struct blkfront_info *info = rinfo->dev_info;
1564
1565         if (unlikely(info->connected != BLKIF_STATE_CONNECTED))
1566                 return IRQ_HANDLED;
1567
1568         spin_lock_irqsave(&rinfo->ring_lock, flags);
1569  again:
1570         rp = rinfo->ring.sring->rsp_prod;
1571         rmb(); /* Ensure we see queued responses up to 'rp'. */
1572
1573         for (i = rinfo->ring.rsp_cons; i != rp; i++) {
1574                 unsigned long id;
1575
1576                 bret = RING_GET_RESPONSE(&rinfo->ring, i);
1577                 id   = bret->id;
1578                 /*
1579                  * The backend has messed up and given us an id that we would
1580                  * never have given to it (we stamp it up to BLK_RING_SIZE -
1581                  * look in get_id_from_freelist.
1582                  */
1583                 if (id >= BLK_RING_SIZE(info)) {
1584                         WARN(1, "%s: response to %s has incorrect id (%ld)\n",
1585                              info->gd->disk_name, op_name(bret->operation), id);
1586                         /* We can't safely get the 'struct request' as
1587                          * the id is busted. */
1588                         continue;
1589                 }
1590                 req  = rinfo->shadow[id].request;
1591
1592                 if (bret->operation != BLKIF_OP_DISCARD) {
1593                         /*
1594                          * We may need to wait for an extra response if the
1595                          * I/O request is split in 2
1596                          */
1597                         if (!blkif_completion(&id, rinfo, bret))
1598                                 continue;
1599                 }
1600
1601                 if (add_id_to_freelist(rinfo, id)) {
1602                         WARN(1, "%s: response to %s (id %ld) couldn't be recycled!\n",
1603                              info->gd->disk_name, op_name(bret->operation), id);
1604                         continue;
1605                 }
1606
1607                 if (bret->status == BLKIF_RSP_OKAY)
1608                         blkif_req(req)->error = BLK_STS_OK;
1609                 else
1610                         blkif_req(req)->error = BLK_STS_IOERR;
1611
1612                 switch (bret->operation) {
1613                 case BLKIF_OP_DISCARD:
1614                         if (unlikely(bret->status == BLKIF_RSP_EOPNOTSUPP)) {
1615                                 struct request_queue *rq = info->rq;
1616                                 printk(KERN_WARNING "blkfront: %s: %s op failed\n",
1617                                            info->gd->disk_name, op_name(bret->operation));
1618                                 blkif_req(req)->error = BLK_STS_NOTSUPP;
1619                                 info->feature_discard = 0;
1620                                 info->feature_secdiscard = 0;
1621                                 blk_queue_flag_clear(QUEUE_FLAG_DISCARD, rq);
1622                                 blk_queue_flag_clear(QUEUE_FLAG_SECERASE, rq);
1623                         }
1624                         break;
1625                 case BLKIF_OP_FLUSH_DISKCACHE:
1626                 case BLKIF_OP_WRITE_BARRIER:
1627                         if (unlikely(bret->status == BLKIF_RSP_EOPNOTSUPP)) {
1628                                 printk(KERN_WARNING "blkfront: %s: %s op failed\n",
1629                                        info->gd->disk_name, op_name(bret->operation));
1630                                 blkif_req(req)->error = BLK_STS_NOTSUPP;
1631                         }
1632                         if (unlikely(bret->status == BLKIF_RSP_ERROR &&
1633                                      rinfo->shadow[id].req.u.rw.nr_segments == 0)) {
1634                                 printk(KERN_WARNING "blkfront: %s: empty %s op failed\n",
1635                                        info->gd->disk_name, op_name(bret->operation));
1636                                 blkif_req(req)->error = BLK_STS_NOTSUPP;
1637                         }
1638                         if (unlikely(blkif_req(req)->error)) {
1639                                 if (blkif_req(req)->error == BLK_STS_NOTSUPP)
1640                                         blkif_req(req)->error = BLK_STS_OK;
1641                                 info->feature_fua = 0;
1642                                 info->feature_flush = 0;
1643                                 xlvbd_flush(info);
1644                         }
1645                         fallthrough;
1646                 case BLKIF_OP_READ:
1647                 case BLKIF_OP_WRITE:
1648                         if (unlikely(bret->status != BLKIF_RSP_OKAY))
1649                                 dev_dbg(&info->xbdev->dev, "Bad return from blkdev data "
1650                                         "request: %x\n", bret->status);
1651
1652                         break;
1653                 default:
1654                         BUG();
1655                 }
1656
1657                 if (likely(!blk_should_fake_timeout(req->q)))
1658                         blk_mq_complete_request(req);
1659         }
1660
1661         rinfo->ring.rsp_cons = i;
1662
1663         if (i != rinfo->ring.req_prod_pvt) {
1664                 int more_to_do;
1665                 RING_FINAL_CHECK_FOR_RESPONSES(&rinfo->ring, more_to_do);
1666                 if (more_to_do)
1667                         goto again;
1668         } else
1669                 rinfo->ring.sring->rsp_event = i + 1;
1670
1671         kick_pending_request_queues_locked(rinfo);
1672
1673         spin_unlock_irqrestore(&rinfo->ring_lock, flags);
1674
1675         return IRQ_HANDLED;
1676 }
1677
1678
1679 static int setup_blkring(struct xenbus_device *dev,
1680                          struct blkfront_ring_info *rinfo)
1681 {
1682         struct blkif_sring *sring;
1683         int err, i;
1684         struct blkfront_info *info = rinfo->dev_info;
1685         unsigned long ring_size = info->nr_ring_pages * XEN_PAGE_SIZE;
1686         grant_ref_t gref[XENBUS_MAX_RING_GRANTS];
1687
1688         for (i = 0; i < info->nr_ring_pages; i++)
1689                 rinfo->ring_ref[i] = GRANT_INVALID_REF;
1690
1691         sring = (struct blkif_sring *)__get_free_pages(GFP_NOIO | __GFP_HIGH,
1692                                                        get_order(ring_size));
1693         if (!sring) {
1694                 xenbus_dev_fatal(dev, -ENOMEM, "allocating shared ring");
1695                 return -ENOMEM;
1696         }
1697         SHARED_RING_INIT(sring);
1698         FRONT_RING_INIT(&rinfo->ring, sring, ring_size);
1699
1700         err = xenbus_grant_ring(dev, rinfo->ring.sring, info->nr_ring_pages, gref);
1701         if (err < 0) {
1702                 free_pages((unsigned long)sring, get_order(ring_size));
1703                 rinfo->ring.sring = NULL;
1704                 goto fail;
1705         }
1706         for (i = 0; i < info->nr_ring_pages; i++)
1707                 rinfo->ring_ref[i] = gref[i];
1708
1709         err = xenbus_alloc_evtchn(dev, &rinfo->evtchn);
1710         if (err)
1711                 goto fail;
1712
1713         err = bind_evtchn_to_irqhandler(rinfo->evtchn, blkif_interrupt, 0,
1714                                         "blkif", rinfo);
1715         if (err <= 0) {
1716                 xenbus_dev_fatal(dev, err,
1717                                  "bind_evtchn_to_irqhandler failed");
1718                 goto fail;
1719         }
1720         rinfo->irq = err;
1721
1722         return 0;
1723 fail:
1724         blkif_free(info, 0);
1725         return err;
1726 }
1727
1728 /*
1729  * Write out per-ring/queue nodes including ring-ref and event-channel, and each
1730  * ring buffer may have multi pages depending on ->nr_ring_pages.
1731  */
1732 static int write_per_ring_nodes(struct xenbus_transaction xbt,
1733                                 struct blkfront_ring_info *rinfo, const char *dir)
1734 {
1735         int err;
1736         unsigned int i;
1737         const char *message = NULL;
1738         struct blkfront_info *info = rinfo->dev_info;
1739
1740         if (info->nr_ring_pages == 1) {
1741                 err = xenbus_printf(xbt, dir, "ring-ref", "%u", rinfo->ring_ref[0]);
1742                 if (err) {
1743                         message = "writing ring-ref";
1744                         goto abort_transaction;
1745                 }
1746         } else {
1747                 for (i = 0; i < info->nr_ring_pages; i++) {
1748                         char ring_ref_name[RINGREF_NAME_LEN];
1749
1750                         snprintf(ring_ref_name, RINGREF_NAME_LEN, "ring-ref%u", i);
1751                         err = xenbus_printf(xbt, dir, ring_ref_name,
1752                                             "%u", rinfo->ring_ref[i]);
1753                         if (err) {
1754                                 message = "writing ring-ref";
1755                                 goto abort_transaction;
1756                         }
1757                 }
1758         }
1759
1760         err = xenbus_printf(xbt, dir, "event-channel", "%u", rinfo->evtchn);
1761         if (err) {
1762                 message = "writing event-channel";
1763                 goto abort_transaction;
1764         }
1765
1766         return 0;
1767
1768 abort_transaction:
1769         xenbus_transaction_end(xbt, 1);
1770         if (message)
1771                 xenbus_dev_fatal(info->xbdev, err, "%s", message);
1772
1773         return err;
1774 }
1775
1776 static void free_info(struct blkfront_info *info)
1777 {
1778         list_del(&info->info_list);
1779         kfree(info);
1780 }
1781
1782 /* Common code used when first setting up, and when resuming. */
1783 static int talk_to_blkback(struct xenbus_device *dev,
1784                            struct blkfront_info *info)
1785 {
1786         const char *message = NULL;
1787         struct xenbus_transaction xbt;
1788         int err;
1789         unsigned int i, max_page_order;
1790         unsigned int ring_page_order;
1791         struct blkfront_ring_info *rinfo;
1792
1793         if (!info)
1794                 return -ENODEV;
1795
1796         max_page_order = xenbus_read_unsigned(info->xbdev->otherend,
1797                                               "max-ring-page-order", 0);
1798         ring_page_order = min(xen_blkif_max_ring_order, max_page_order);
1799         info->nr_ring_pages = 1 << ring_page_order;
1800
1801         err = negotiate_mq(info);
1802         if (err)
1803                 goto destroy_blkring;
1804
1805         for_each_rinfo(info, rinfo, i) {
1806                 /* Create shared ring, alloc event channel. */
1807                 err = setup_blkring(dev, rinfo);
1808                 if (err)
1809                         goto destroy_blkring;
1810         }
1811
1812 again:
1813         err = xenbus_transaction_start(&xbt);
1814         if (err) {
1815                 xenbus_dev_fatal(dev, err, "starting transaction");
1816                 goto destroy_blkring;
1817         }
1818
1819         if (info->nr_ring_pages > 1) {
1820                 err = xenbus_printf(xbt, dev->nodename, "ring-page-order", "%u",
1821                                     ring_page_order);
1822                 if (err) {
1823                         message = "writing ring-page-order";
1824                         goto abort_transaction;
1825                 }
1826         }
1827
1828         /* We already got the number of queues/rings in _probe */
1829         if (info->nr_rings == 1) {
1830                 err = write_per_ring_nodes(xbt, info->rinfo, dev->nodename);
1831                 if (err)
1832                         goto destroy_blkring;
1833         } else {
1834                 char *path;
1835                 size_t pathsize;
1836
1837                 err = xenbus_printf(xbt, dev->nodename, "multi-queue-num-queues", "%u",
1838                                     info->nr_rings);
1839                 if (err) {
1840                         message = "writing multi-queue-num-queues";
1841                         goto abort_transaction;
1842                 }
1843
1844                 pathsize = strlen(dev->nodename) + QUEUE_NAME_LEN;
1845                 path = kmalloc(pathsize, GFP_KERNEL);
1846                 if (!path) {
1847                         err = -ENOMEM;
1848                         message = "ENOMEM while writing ring references";
1849                         goto abort_transaction;
1850                 }
1851
1852                 for_each_rinfo(info, rinfo, i) {
1853                         memset(path, 0, pathsize);
1854                         snprintf(path, pathsize, "%s/queue-%u", dev->nodename, i);
1855                         err = write_per_ring_nodes(xbt, rinfo, path);
1856                         if (err) {
1857                                 kfree(path);
1858                                 goto destroy_blkring;
1859                         }
1860                 }
1861                 kfree(path);
1862         }
1863         err = xenbus_printf(xbt, dev->nodename, "protocol", "%s",
1864                             XEN_IO_PROTO_ABI_NATIVE);
1865         if (err) {
1866                 message = "writing protocol";
1867                 goto abort_transaction;
1868         }
1869         err = xenbus_printf(xbt, dev->nodename, "feature-persistent", "%u",
1870                         info->feature_persistent);
1871         if (err)
1872                 dev_warn(&dev->dev,
1873                          "writing persistent grants feature to xenbus");
1874
1875         err = xenbus_transaction_end(xbt, 0);
1876         if (err) {
1877                 if (err == -EAGAIN)
1878                         goto again;
1879                 xenbus_dev_fatal(dev, err, "completing transaction");
1880                 goto destroy_blkring;
1881         }
1882
1883         for_each_rinfo(info, rinfo, i) {
1884                 unsigned int j;
1885
1886                 for (j = 0; j < BLK_RING_SIZE(info); j++)
1887                         rinfo->shadow[j].req.u.rw.id = j + 1;
1888                 rinfo->shadow[BLK_RING_SIZE(info)-1].req.u.rw.id = 0x0fffffff;
1889         }
1890         xenbus_switch_state(dev, XenbusStateInitialised);
1891
1892         return 0;
1893
1894  abort_transaction:
1895         xenbus_transaction_end(xbt, 1);
1896         if (message)
1897                 xenbus_dev_fatal(dev, err, "%s", message);
1898  destroy_blkring:
1899         blkif_free(info, 0);
1900
1901         mutex_lock(&blkfront_mutex);
1902         free_info(info);
1903         mutex_unlock(&blkfront_mutex);
1904
1905         dev_set_drvdata(&dev->dev, NULL);
1906
1907         return err;
1908 }
1909
1910 static int negotiate_mq(struct blkfront_info *info)
1911 {
1912         unsigned int backend_max_queues;
1913         unsigned int i;
1914         struct blkfront_ring_info *rinfo;
1915
1916         BUG_ON(info->nr_rings);
1917
1918         /* Check if backend supports multiple queues. */
1919         backend_max_queues = xenbus_read_unsigned(info->xbdev->otherend,
1920                                                   "multi-queue-max-queues", 1);
1921         info->nr_rings = min(backend_max_queues, xen_blkif_max_queues);
1922         /* We need at least one ring. */
1923         if (!info->nr_rings)
1924                 info->nr_rings = 1;
1925
1926         info->rinfo_size = struct_size(info->rinfo, shadow,
1927                                        BLK_RING_SIZE(info));
1928         info->rinfo = kvcalloc(info->nr_rings, info->rinfo_size, GFP_KERNEL);
1929         if (!info->rinfo) {
1930                 xenbus_dev_fatal(info->xbdev, -ENOMEM, "allocating ring_info structure");
1931                 info->nr_rings = 0;
1932                 return -ENOMEM;
1933         }
1934
1935         for_each_rinfo(info, rinfo, i) {
1936                 INIT_LIST_HEAD(&rinfo->indirect_pages);
1937                 INIT_LIST_HEAD(&rinfo->grants);
1938                 rinfo->dev_info = info;
1939                 INIT_WORK(&rinfo->work, blkif_restart_queue);
1940                 spin_lock_init(&rinfo->ring_lock);
1941         }
1942         return 0;
1943 }
1944
1945 /* Enable the persistent grants feature. */
1946 static bool feature_persistent = true;
1947 module_param(feature_persistent, bool, 0644);
1948 MODULE_PARM_DESC(feature_persistent,
1949                 "Enables the persistent grants feature");
1950
1951 /**
1952  * Entry point to this code when a new device is created.  Allocate the basic
1953  * structures and the ring buffer for communication with the backend, and
1954  * inform the backend of the appropriate details for those.  Switch to
1955  * Initialised state.
1956  */
1957 static int blkfront_probe(struct xenbus_device *dev,
1958                           const struct xenbus_device_id *id)
1959 {
1960         int err, vdevice;
1961         struct blkfront_info *info;
1962
1963         /* FIXME: Use dynamic device id if this is not set. */
1964         err = xenbus_scanf(XBT_NIL, dev->nodename,
1965                            "virtual-device", "%i", &vdevice);
1966         if (err != 1) {
1967                 /* go looking in the extended area instead */
1968                 err = xenbus_scanf(XBT_NIL, dev->nodename, "virtual-device-ext",
1969                                    "%i", &vdevice);
1970                 if (err != 1) {
1971                         xenbus_dev_fatal(dev, err, "reading virtual-device");
1972                         return err;
1973                 }
1974         }
1975
1976         if (xen_hvm_domain()) {
1977                 char *type;
1978                 int len;
1979                 /* no unplug has been done: do not hook devices != xen vbds */
1980                 if (xen_has_pv_and_legacy_disk_devices()) {
1981                         int major;
1982
1983                         if (!VDEV_IS_EXTENDED(vdevice))
1984                                 major = BLKIF_MAJOR(vdevice);
1985                         else
1986                                 major = XENVBD_MAJOR;
1987
1988                         if (major != XENVBD_MAJOR) {
1989                                 printk(KERN_INFO
1990                                                 "%s: HVM does not support vbd %d as xen block device\n",
1991                                                 __func__, vdevice);
1992                                 return -ENODEV;
1993                         }
1994                 }
1995                 /* do not create a PV cdrom device if we are an HVM guest */
1996                 type = xenbus_read(XBT_NIL, dev->nodename, "device-type", &len);
1997                 if (IS_ERR(type))
1998                         return -ENODEV;
1999                 if (strncmp(type, "cdrom", 5) == 0) {
2000                         kfree(type);
2001                         return -ENODEV;
2002                 }
2003                 kfree(type);
2004         }
2005         info = kzalloc(sizeof(*info), GFP_KERNEL);
2006         if (!info) {
2007                 xenbus_dev_fatal(dev, -ENOMEM, "allocating info structure");
2008                 return -ENOMEM;
2009         }
2010
2011         info->xbdev = dev;
2012
2013         mutex_init(&info->mutex);
2014         info->vdevice = vdevice;
2015         info->connected = BLKIF_STATE_DISCONNECTED;
2016
2017         info->feature_persistent = feature_persistent;
2018
2019         /* Front end dir is a number, which is used as the id. */
2020         info->handle = simple_strtoul(strrchr(dev->nodename, '/')+1, NULL, 0);
2021         dev_set_drvdata(&dev->dev, info);
2022
2023         mutex_lock(&blkfront_mutex);
2024         list_add(&info->info_list, &info_list);
2025         mutex_unlock(&blkfront_mutex);
2026
2027         return 0;
2028 }
2029
2030 static int blkif_recover(struct blkfront_info *info)
2031 {
2032         unsigned int r_index;
2033         struct request *req, *n;
2034         int rc;
2035         struct bio *bio;
2036         unsigned int segs;
2037         struct blkfront_ring_info *rinfo;
2038
2039         blkfront_gather_backend_features(info);
2040         /* Reset limits changed by blk_mq_update_nr_hw_queues(). */
2041         blkif_set_queue_limits(info);
2042         segs = info->max_indirect_segments ? : BLKIF_MAX_SEGMENTS_PER_REQUEST;
2043         blk_queue_max_segments(info->rq, segs / GRANTS_PER_PSEG);
2044
2045         for_each_rinfo(info, rinfo, r_index) {
2046                 rc = blkfront_setup_indirect(rinfo);
2047                 if (rc)
2048                         return rc;
2049         }
2050         xenbus_switch_state(info->xbdev, XenbusStateConnected);
2051
2052         /* Now safe for us to use the shared ring */
2053         info->connected = BLKIF_STATE_CONNECTED;
2054
2055         for_each_rinfo(info, rinfo, r_index) {
2056                 /* Kick any other new requests queued since we resumed */
2057                 kick_pending_request_queues(rinfo);
2058         }
2059
2060         list_for_each_entry_safe(req, n, &info->requests, queuelist) {
2061                 /* Requeue pending requests (flush or discard) */
2062                 list_del_init(&req->queuelist);
2063                 BUG_ON(req->nr_phys_segments > segs);
2064                 blk_mq_requeue_request(req, false);
2065         }
2066         blk_mq_start_stopped_hw_queues(info->rq, true);
2067         blk_mq_kick_requeue_list(info->rq);
2068
2069         while ((bio = bio_list_pop(&info->bio_list)) != NULL) {
2070                 /* Traverse the list of pending bios and re-queue them */
2071                 submit_bio(bio);
2072         }
2073
2074         return 0;
2075 }
2076
2077 /**
2078  * We are reconnecting to the backend, due to a suspend/resume, or a backend
2079  * driver restart.  We tear down our blkif structure and recreate it, but
2080  * leave the device-layer structures intact so that this is transparent to the
2081  * rest of the kernel.
2082  */
2083 static int blkfront_resume(struct xenbus_device *dev)
2084 {
2085         struct blkfront_info *info = dev_get_drvdata(&dev->dev);
2086         int err = 0;
2087         unsigned int i, j;
2088         struct blkfront_ring_info *rinfo;
2089
2090         dev_dbg(&dev->dev, "blkfront_resume: %s\n", dev->nodename);
2091
2092         bio_list_init(&info->bio_list);
2093         INIT_LIST_HEAD(&info->requests);
2094         for_each_rinfo(info, rinfo, i) {
2095                 struct bio_list merge_bio;
2096                 struct blk_shadow *shadow = rinfo->shadow;
2097
2098                 for (j = 0; j < BLK_RING_SIZE(info); j++) {
2099                         /* Not in use? */
2100                         if (!shadow[j].request)
2101                                 continue;
2102
2103                         /*
2104                          * Get the bios in the request so we can re-queue them.
2105                          */
2106                         if (req_op(shadow[j].request) == REQ_OP_FLUSH ||
2107                             req_op(shadow[j].request) == REQ_OP_DISCARD ||
2108                             req_op(shadow[j].request) == REQ_OP_SECURE_ERASE ||
2109                             shadow[j].request->cmd_flags & REQ_FUA) {
2110                                 /*
2111                                  * Flush operations don't contain bios, so
2112                                  * we need to requeue the whole request
2113                                  *
2114                                  * XXX: but this doesn't make any sense for a
2115                                  * write with the FUA flag set..
2116                                  */
2117                                 list_add(&shadow[j].request->queuelist, &info->requests);
2118                                 continue;
2119                         }
2120                         merge_bio.head = shadow[j].request->bio;
2121                         merge_bio.tail = shadow[j].request->biotail;
2122                         bio_list_merge(&info->bio_list, &merge_bio);
2123                         shadow[j].request->bio = NULL;
2124                         blk_mq_end_request(shadow[j].request, BLK_STS_OK);
2125                 }
2126         }
2127
2128         blkif_free(info, info->connected == BLKIF_STATE_CONNECTED);
2129
2130         err = talk_to_blkback(dev, info);
2131         if (!err)
2132                 blk_mq_update_nr_hw_queues(&info->tag_set, info->nr_rings);
2133
2134         /*
2135          * We have to wait for the backend to switch to
2136          * connected state, since we want to read which
2137          * features it supports.
2138          */
2139
2140         return err;
2141 }
2142
2143 static void blkfront_closing(struct blkfront_info *info)
2144 {
2145         struct xenbus_device *xbdev = info->xbdev;
2146         struct block_device *bdev = NULL;
2147
2148         mutex_lock(&info->mutex);
2149
2150         if (xbdev->state == XenbusStateClosing) {
2151                 mutex_unlock(&info->mutex);
2152                 return;
2153         }
2154
2155         if (info->gd)
2156                 bdev = bdget_disk(info->gd, 0);
2157
2158         mutex_unlock(&info->mutex);
2159
2160         if (!bdev) {
2161                 xenbus_frontend_closed(xbdev);
2162                 return;
2163         }
2164
2165         mutex_lock(&bdev->bd_mutex);
2166
2167         if (bdev->bd_openers) {
2168                 xenbus_dev_error(xbdev, -EBUSY,
2169                                  "Device in use; refusing to close");
2170                 xenbus_switch_state(xbdev, XenbusStateClosing);
2171         } else {
2172                 xlvbd_release_gendisk(info);
2173                 xenbus_frontend_closed(xbdev);
2174         }
2175
2176         mutex_unlock(&bdev->bd_mutex);
2177         bdput(bdev);
2178 }
2179
2180 static void blkfront_setup_discard(struct blkfront_info *info)
2181 {
2182         int err;
2183         unsigned int discard_granularity;
2184         unsigned int discard_alignment;
2185
2186         info->feature_discard = 1;
2187         err = xenbus_gather(XBT_NIL, info->xbdev->otherend,
2188                 "discard-granularity", "%u", &discard_granularity,
2189                 "discard-alignment", "%u", &discard_alignment,
2190                 NULL);
2191         if (!err) {
2192                 info->discard_granularity = discard_granularity;
2193                 info->discard_alignment = discard_alignment;
2194         }
2195         info->feature_secdiscard =
2196                 !!xenbus_read_unsigned(info->xbdev->otherend, "discard-secure",
2197                                        0);
2198 }
2199
2200 static int blkfront_setup_indirect(struct blkfront_ring_info *rinfo)
2201 {
2202         unsigned int psegs, grants, memflags;
2203         int err, i;
2204         struct blkfront_info *info = rinfo->dev_info;
2205
2206         memflags = memalloc_noio_save();
2207
2208         if (info->max_indirect_segments == 0) {
2209                 if (!HAS_EXTRA_REQ)
2210                         grants = BLKIF_MAX_SEGMENTS_PER_REQUEST;
2211                 else {
2212                         /*
2213                          * When an extra req is required, the maximum
2214                          * grants supported is related to the size of the
2215                          * Linux block segment.
2216                          */
2217                         grants = GRANTS_PER_PSEG;
2218                 }
2219         }
2220         else
2221                 grants = info->max_indirect_segments;
2222         psegs = DIV_ROUND_UP(grants, GRANTS_PER_PSEG);
2223
2224         err = fill_grant_buffer(rinfo,
2225                                 (grants + INDIRECT_GREFS(grants)) * BLK_RING_SIZE(info));
2226         if (err)
2227                 goto out_of_memory;
2228
2229         if (!info->feature_persistent && info->max_indirect_segments) {
2230                 /*
2231                  * We are using indirect descriptors but not persistent
2232                  * grants, we need to allocate a set of pages that can be
2233                  * used for mapping indirect grefs
2234                  */
2235                 int num = INDIRECT_GREFS(grants) * BLK_RING_SIZE(info);
2236
2237                 BUG_ON(!list_empty(&rinfo->indirect_pages));
2238                 for (i = 0; i < num; i++) {
2239                         struct page *indirect_page = alloc_page(GFP_KERNEL);
2240                         if (!indirect_page)
2241                                 goto out_of_memory;
2242                         list_add(&indirect_page->lru, &rinfo->indirect_pages);
2243                 }
2244         }
2245
2246         for (i = 0; i < BLK_RING_SIZE(info); i++) {
2247                 rinfo->shadow[i].grants_used =
2248                         kvcalloc(grants,
2249                                  sizeof(rinfo->shadow[i].grants_used[0]),
2250                                  GFP_KERNEL);
2251                 rinfo->shadow[i].sg = kvcalloc(psegs,
2252                                                sizeof(rinfo->shadow[i].sg[0]),
2253                                                GFP_KERNEL);
2254                 if (info->max_indirect_segments)
2255                         rinfo->shadow[i].indirect_grants =
2256                                 kvcalloc(INDIRECT_GREFS(grants),
2257                                          sizeof(rinfo->shadow[i].indirect_grants[0]),
2258                                          GFP_KERNEL);
2259                 if ((rinfo->shadow[i].grants_used == NULL) ||
2260                         (rinfo->shadow[i].sg == NULL) ||
2261                      (info->max_indirect_segments &&
2262                      (rinfo->shadow[i].indirect_grants == NULL)))
2263                         goto out_of_memory;
2264                 sg_init_table(rinfo->shadow[i].sg, psegs);
2265         }
2266
2267         memalloc_noio_restore(memflags);
2268
2269         return 0;
2270
2271 out_of_memory:
2272         for (i = 0; i < BLK_RING_SIZE(info); i++) {
2273                 kvfree(rinfo->shadow[i].grants_used);
2274                 rinfo->shadow[i].grants_used = NULL;
2275                 kvfree(rinfo->shadow[i].sg);
2276                 rinfo->shadow[i].sg = NULL;
2277                 kvfree(rinfo->shadow[i].indirect_grants);
2278                 rinfo->shadow[i].indirect_grants = NULL;
2279         }
2280         if (!list_empty(&rinfo->indirect_pages)) {
2281                 struct page *indirect_page, *n;
2282                 list_for_each_entry_safe(indirect_page, n, &rinfo->indirect_pages, lru) {
2283                         list_del(&indirect_page->lru);
2284                         __free_page(indirect_page);
2285                 }
2286         }
2287
2288         memalloc_noio_restore(memflags);
2289
2290         return -ENOMEM;
2291 }
2292
2293 /*
2294  * Gather all backend feature-*
2295  */
2296 static void blkfront_gather_backend_features(struct blkfront_info *info)
2297 {
2298         unsigned int indirect_segments;
2299
2300         info->feature_flush = 0;
2301         info->feature_fua = 0;
2302
2303         /*
2304          * If there's no "feature-barrier" defined, then it means
2305          * we're dealing with a very old backend which writes
2306          * synchronously; nothing to do.
2307          *
2308          * If there are barriers, then we use flush.
2309          */
2310         if (xenbus_read_unsigned(info->xbdev->otherend, "feature-barrier", 0)) {
2311                 info->feature_flush = 1;
2312                 info->feature_fua = 1;
2313         }
2314
2315         /*
2316          * And if there is "feature-flush-cache" use that above
2317          * barriers.
2318          */
2319         if (xenbus_read_unsigned(info->xbdev->otherend, "feature-flush-cache",
2320                                  0)) {
2321                 info->feature_flush = 1;
2322                 info->feature_fua = 0;
2323         }
2324
2325         if (xenbus_read_unsigned(info->xbdev->otherend, "feature-discard", 0))
2326                 blkfront_setup_discard(info);
2327
2328         if (info->feature_persistent)
2329                 info->feature_persistent =
2330                         !!xenbus_read_unsigned(info->xbdev->otherend,
2331                                                "feature-persistent", 0);
2332
2333         indirect_segments = xenbus_read_unsigned(info->xbdev->otherend,
2334                                         "feature-max-indirect-segments", 0);
2335         if (indirect_segments > xen_blkif_max_segments)
2336                 indirect_segments = xen_blkif_max_segments;
2337         if (indirect_segments <= BLKIF_MAX_SEGMENTS_PER_REQUEST)
2338                 indirect_segments = 0;
2339         info->max_indirect_segments = indirect_segments;
2340
2341         if (info->feature_persistent) {
2342                 mutex_lock(&blkfront_mutex);
2343                 schedule_delayed_work(&blkfront_work, HZ * 10);
2344                 mutex_unlock(&blkfront_mutex);
2345         }
2346 }
2347
2348 /*
2349  * Invoked when the backend is finally 'ready' (and has told produced
2350  * the details about the physical device - #sectors, size, etc).
2351  */
2352 static void blkfront_connect(struct blkfront_info *info)
2353 {
2354         unsigned long long sectors;
2355         unsigned long sector_size;
2356         unsigned int physical_sector_size;
2357         unsigned int binfo;
2358         int err, i;
2359         struct blkfront_ring_info *rinfo;
2360
2361         switch (info->connected) {
2362         case BLKIF_STATE_CONNECTED:
2363                 /*
2364                  * Potentially, the back-end may be signalling
2365                  * a capacity change; update the capacity.
2366                  */
2367                 err = xenbus_scanf(XBT_NIL, info->xbdev->otherend,
2368                                    "sectors", "%Lu", &sectors);
2369                 if (XENBUS_EXIST_ERR(err))
2370                         return;
2371                 printk(KERN_INFO "Setting capacity to %Lu\n",
2372                        sectors);
2373                 set_capacity_and_notify(info->gd, sectors);
2374
2375                 return;
2376         case BLKIF_STATE_SUSPENDED:
2377                 /*
2378                  * If we are recovering from suspension, we need to wait
2379                  * for the backend to announce it's features before
2380                  * reconnecting, at least we need to know if the backend
2381                  * supports indirect descriptors, and how many.
2382                  */
2383                 blkif_recover(info);
2384                 return;
2385
2386         default:
2387                 break;
2388         }
2389
2390         dev_dbg(&info->xbdev->dev, "%s:%s.\n",
2391                 __func__, info->xbdev->otherend);
2392
2393         err = xenbus_gather(XBT_NIL, info->xbdev->otherend,
2394                             "sectors", "%llu", &sectors,
2395                             "info", "%u", &binfo,
2396                             "sector-size", "%lu", &sector_size,
2397                             NULL);
2398         if (err) {
2399                 xenbus_dev_fatal(info->xbdev, err,
2400                                  "reading backend fields at %s",
2401                                  info->xbdev->otherend);
2402                 return;
2403         }
2404
2405         /*
2406          * physcial-sector-size is a newer field, so old backends may not
2407          * provide this. Assume physical sector size to be the same as
2408          * sector_size in that case.
2409          */
2410         physical_sector_size = xenbus_read_unsigned(info->xbdev->otherend,
2411                                                     "physical-sector-size",
2412                                                     sector_size);
2413         blkfront_gather_backend_features(info);
2414         for_each_rinfo(info, rinfo, i) {
2415                 err = blkfront_setup_indirect(rinfo);
2416                 if (err) {
2417                         xenbus_dev_fatal(info->xbdev, err, "setup_indirect at %s",
2418                                          info->xbdev->otherend);
2419                         blkif_free(info, 0);
2420                         break;
2421                 }
2422         }
2423
2424         err = xlvbd_alloc_gendisk(sectors, info, binfo, sector_size,
2425                                   physical_sector_size);
2426         if (err) {
2427                 xenbus_dev_fatal(info->xbdev, err, "xlvbd_add at %s",
2428                                  info->xbdev->otherend);
2429                 goto fail;
2430         }
2431
2432         xenbus_switch_state(info->xbdev, XenbusStateConnected);
2433
2434         /* Kick pending requests. */
2435         info->connected = BLKIF_STATE_CONNECTED;
2436         for_each_rinfo(info, rinfo, i)
2437                 kick_pending_request_queues(rinfo);
2438
2439         device_add_disk(&info->xbdev->dev, info->gd, NULL);
2440
2441         info->is_ready = 1;
2442         return;
2443
2444 fail:
2445         blkif_free(info, 0);
2446         return;
2447 }
2448
2449 /**
2450  * Callback received when the backend's state changes.
2451  */
2452 static void blkback_changed(struct xenbus_device *dev,
2453                             enum xenbus_state backend_state)
2454 {
2455         struct blkfront_info *info = dev_get_drvdata(&dev->dev);
2456
2457         dev_dbg(&dev->dev, "blkfront:blkback_changed to state %d.\n", backend_state);
2458
2459         switch (backend_state) {
2460         case XenbusStateInitWait:
2461                 if (dev->state != XenbusStateInitialising)
2462                         break;
2463                 if (talk_to_blkback(dev, info))
2464                         break;
2465         case XenbusStateInitialising:
2466         case XenbusStateInitialised:
2467         case XenbusStateReconfiguring:
2468         case XenbusStateReconfigured:
2469         case XenbusStateUnknown:
2470                 break;
2471
2472         case XenbusStateConnected:
2473                 /*
2474                  * talk_to_blkback sets state to XenbusStateInitialised
2475                  * and blkfront_connect sets it to XenbusStateConnected
2476                  * (if connection went OK).
2477                  *
2478                  * If the backend (or toolstack) decides to poke at backend
2479                  * state (and re-trigger the watch by setting the state repeatedly
2480                  * to XenbusStateConnected (4)) we need to deal with this.
2481                  * This is allowed as this is used to communicate to the guest
2482                  * that the size of disk has changed!
2483                  */
2484                 if ((dev->state != XenbusStateInitialised) &&
2485                     (dev->state != XenbusStateConnected)) {
2486                         if (talk_to_blkback(dev, info))
2487                                 break;
2488                 }
2489
2490                 blkfront_connect(info);
2491                 break;
2492
2493         case XenbusStateClosed:
2494                 if (dev->state == XenbusStateClosed)
2495                         break;
2496                 fallthrough;
2497         case XenbusStateClosing:
2498                 if (info)
2499                         blkfront_closing(info);
2500                 break;
2501         }
2502 }
2503
2504 static int blkfront_remove(struct xenbus_device *xbdev)
2505 {
2506         struct blkfront_info *info = dev_get_drvdata(&xbdev->dev);
2507         struct block_device *bdev = NULL;
2508         struct gendisk *disk;
2509
2510         dev_dbg(&xbdev->dev, "%s removed", xbdev->nodename);
2511
2512         if (!info)
2513                 return 0;
2514
2515         blkif_free(info, 0);
2516
2517         mutex_lock(&info->mutex);
2518
2519         disk = info->gd;
2520         if (disk)
2521                 bdev = bdget_disk(disk, 0);
2522
2523         info->xbdev = NULL;
2524         mutex_unlock(&info->mutex);
2525
2526         if (!bdev) {
2527                 mutex_lock(&blkfront_mutex);
2528                 free_info(info);
2529                 mutex_unlock(&blkfront_mutex);
2530                 return 0;
2531         }
2532
2533         /*
2534          * The xbdev was removed before we reached the Closed
2535          * state. See if it's safe to remove the disk. If the bdev
2536          * isn't closed yet, we let release take care of it.
2537          */
2538
2539         mutex_lock(&bdev->bd_mutex);
2540         info = disk->private_data;
2541
2542         dev_warn(disk_to_dev(disk),
2543                  "%s was hot-unplugged, %d stale handles\n",
2544                  xbdev->nodename, bdev->bd_openers);
2545
2546         if (info && !bdev->bd_openers) {
2547                 xlvbd_release_gendisk(info);
2548                 disk->private_data = NULL;
2549                 mutex_lock(&blkfront_mutex);
2550                 free_info(info);
2551                 mutex_unlock(&blkfront_mutex);
2552         }
2553
2554         mutex_unlock(&bdev->bd_mutex);
2555         bdput(bdev);
2556
2557         return 0;
2558 }
2559
2560 static int blkfront_is_ready(struct xenbus_device *dev)
2561 {
2562         struct blkfront_info *info = dev_get_drvdata(&dev->dev);
2563
2564         return info->is_ready && info->xbdev;
2565 }
2566
2567 static int blkif_open(struct block_device *bdev, fmode_t mode)
2568 {
2569         struct gendisk *disk = bdev->bd_disk;
2570         struct blkfront_info *info;
2571         int err = 0;
2572
2573         mutex_lock(&blkfront_mutex);
2574
2575         info = disk->private_data;
2576         if (!info) {
2577                 /* xbdev gone */
2578                 err = -ERESTARTSYS;
2579                 goto out;
2580         }
2581
2582         mutex_lock(&info->mutex);
2583
2584         if (!info->gd)
2585                 /* xbdev is closed */
2586                 err = -ERESTARTSYS;
2587
2588         mutex_unlock(&info->mutex);
2589
2590 out:
2591         mutex_unlock(&blkfront_mutex);
2592         return err;
2593 }
2594
2595 static void blkif_release(struct gendisk *disk, fmode_t mode)
2596 {
2597         struct blkfront_info *info = disk->private_data;
2598         struct block_device *bdev;
2599         struct xenbus_device *xbdev;
2600
2601         mutex_lock(&blkfront_mutex);
2602
2603         bdev = bdget_disk(disk, 0);
2604
2605         if (!bdev) {
2606                 WARN(1, "Block device %s yanked out from us!\n", disk->disk_name);
2607                 goto out_mutex;
2608         }
2609         if (bdev->bd_openers)
2610                 goto out;
2611
2612         /*
2613          * Check if we have been instructed to close. We will have
2614          * deferred this request, because the bdev was still open.
2615          */
2616
2617         mutex_lock(&info->mutex);
2618         xbdev = info->xbdev;
2619
2620         if (xbdev && xbdev->state == XenbusStateClosing) {
2621                 /* pending switch to state closed */
2622                 dev_info(disk_to_dev(bdev->bd_disk), "releasing disk\n");
2623                 xlvbd_release_gendisk(info);
2624                 xenbus_frontend_closed(info->xbdev);
2625         }
2626
2627         mutex_unlock(&info->mutex);
2628
2629         if (!xbdev) {
2630                 /* sudden device removal */
2631                 dev_info(disk_to_dev(bdev->bd_disk), "releasing disk\n");
2632                 xlvbd_release_gendisk(info);
2633                 disk->private_data = NULL;
2634                 free_info(info);
2635         }
2636
2637 out:
2638         bdput(bdev);
2639 out_mutex:
2640         mutex_unlock(&blkfront_mutex);
2641 }
2642
2643 static const struct block_device_operations xlvbd_block_fops =
2644 {
2645         .owner = THIS_MODULE,
2646         .open = blkif_open,
2647         .release = blkif_release,
2648         .getgeo = blkif_getgeo,
2649         .ioctl = blkif_ioctl,
2650         .compat_ioctl = blkdev_compat_ptr_ioctl,
2651 };
2652
2653
2654 static const struct xenbus_device_id blkfront_ids[] = {
2655         { "vbd" },
2656         { "" }
2657 };
2658
2659 static struct xenbus_driver blkfront_driver = {
2660         .ids  = blkfront_ids,
2661         .probe = blkfront_probe,
2662         .remove = blkfront_remove,
2663         .resume = blkfront_resume,
2664         .otherend_changed = blkback_changed,
2665         .is_ready = blkfront_is_ready,
2666 };
2667
2668 static void purge_persistent_grants(struct blkfront_info *info)
2669 {
2670         unsigned int i;
2671         unsigned long flags;
2672         struct blkfront_ring_info *rinfo;
2673
2674         for_each_rinfo(info, rinfo, i) {
2675                 struct grant *gnt_list_entry, *tmp;
2676
2677                 spin_lock_irqsave(&rinfo->ring_lock, flags);
2678
2679                 if (rinfo->persistent_gnts_c == 0) {
2680                         spin_unlock_irqrestore(&rinfo->ring_lock, flags);
2681                         continue;
2682                 }
2683
2684                 list_for_each_entry_safe(gnt_list_entry, tmp, &rinfo->grants,
2685                                          node) {
2686                         if (gnt_list_entry->gref == GRANT_INVALID_REF ||
2687                             gnttab_query_foreign_access(gnt_list_entry->gref))
2688                                 continue;
2689
2690                         list_del(&gnt_list_entry->node);
2691                         gnttab_end_foreign_access(gnt_list_entry->gref, 0, 0UL);
2692                         rinfo->persistent_gnts_c--;
2693                         gnt_list_entry->gref = GRANT_INVALID_REF;
2694                         list_add_tail(&gnt_list_entry->node, &rinfo->grants);
2695                 }
2696
2697                 spin_unlock_irqrestore(&rinfo->ring_lock, flags);
2698         }
2699 }
2700
2701 static void blkfront_delay_work(struct work_struct *work)
2702 {
2703         struct blkfront_info *info;
2704         bool need_schedule_work = false;
2705
2706         mutex_lock(&blkfront_mutex);
2707
2708         list_for_each_entry(info, &info_list, info_list) {
2709                 if (info->feature_persistent) {
2710                         need_schedule_work = true;
2711                         mutex_lock(&info->mutex);
2712                         purge_persistent_grants(info);
2713                         mutex_unlock(&info->mutex);
2714                 }
2715         }
2716
2717         if (need_schedule_work)
2718                 schedule_delayed_work(&blkfront_work, HZ * 10);
2719
2720         mutex_unlock(&blkfront_mutex);
2721 }
2722
2723 static int __init xlblk_init(void)
2724 {
2725         int ret;
2726         int nr_cpus = num_online_cpus();
2727
2728         if (!xen_domain())
2729                 return -ENODEV;
2730
2731         if (!xen_has_pv_disk_devices())
2732                 return -ENODEV;
2733
2734         if (register_blkdev(XENVBD_MAJOR, DEV_NAME)) {
2735                 pr_warn("xen_blk: can't get major %d with name %s\n",
2736                         XENVBD_MAJOR, DEV_NAME);
2737                 return -ENODEV;
2738         }
2739
2740         if (xen_blkif_max_segments < BLKIF_MAX_SEGMENTS_PER_REQUEST)
2741                 xen_blkif_max_segments = BLKIF_MAX_SEGMENTS_PER_REQUEST;
2742
2743         if (xen_blkif_max_ring_order > XENBUS_MAX_RING_GRANT_ORDER) {
2744                 pr_info("Invalid max_ring_order (%d), will use default max: %d.\n",
2745                         xen_blkif_max_ring_order, XENBUS_MAX_RING_GRANT_ORDER);
2746                 xen_blkif_max_ring_order = XENBUS_MAX_RING_GRANT_ORDER;
2747         }
2748
2749         if (xen_blkif_max_queues > nr_cpus) {
2750                 pr_info("Invalid max_queues (%d), will use default max: %d.\n",
2751                         xen_blkif_max_queues, nr_cpus);
2752                 xen_blkif_max_queues = nr_cpus;
2753         }
2754
2755         INIT_DELAYED_WORK(&blkfront_work, blkfront_delay_work);
2756
2757         ret = xenbus_register_frontend(&blkfront_driver);
2758         if (ret) {
2759                 unregister_blkdev(XENVBD_MAJOR, DEV_NAME);
2760                 return ret;
2761         }
2762
2763         return 0;
2764 }
2765 module_init(xlblk_init);
2766
2767
2768 static void __exit xlblk_exit(void)
2769 {
2770         cancel_delayed_work_sync(&blkfront_work);
2771
2772         xenbus_unregister_driver(&blkfront_driver);
2773         unregister_blkdev(XENVBD_MAJOR, DEV_NAME);
2774         kfree(minors);
2775 }
2776 module_exit(xlblk_exit);
2777
2778 MODULE_DESCRIPTION("Xen virtual block device frontend");
2779 MODULE_LICENSE("GPL");
2780 MODULE_ALIAS_BLOCKDEV_MAJOR(XENVBD_MAJOR);
2781 MODULE_ALIAS("xen:vbd");
2782 MODULE_ALIAS("xenblk");