1/****************************************************************************** 2 * gntdev.c 3 * 4 * Device for accessing (in user-space) pages that have been granted by other 5 * domains. 6 * 7 * Copyright (c) 2006-2007, D G Murray. 8 * (c) 2009 Gerd Hoffmann <kraxel@redhat.com> 9 * (c) 2018 Oleksandr Andrushchenko, EPAM Systems Inc. 10 * 11 * This program is distributed in the hope that it will be useful, 12 * but WITHOUT ANY WARRANTY; without even the implied warranty of 13 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the 14 * GNU General Public License for more details. 15 * 16 * You should have received a copy of the GNU General Public License 17 * along with this program; if not, write to the Free Software 18 * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA 19 */ 20 21#undef DEBUG 22 23#define pr_fmt(fmt) "xen:" KBUILD_MODNAME ": " fmt 24 25#include <linux/dma-mapping.h> 26#include <linux/module.h> 27#include <linux/kernel.h> 28#include <linux/init.h> 29#include <linux/miscdevice.h> 30#include <linux/fs.h> 31#include <linux/uaccess.h> 32#include <linux/sched.h> 33#include <linux/sched/mm.h> 34#include <linux/spinlock.h> 35#include <linux/slab.h> 36#include <linux/highmem.h> 37#include <linux/refcount.h> 38#include <linux/workqueue.h> 39 40#include <xen/xen.h> 41#include <xen/grant_table.h> 42#include <xen/balloon.h> 43#include <xen/gntdev.h> 44#include <xen/events.h> 45#include <xen/page.h> 46#include <asm/xen/hypervisor.h> 47#include <asm/xen/hypercall.h> 48 49#include "gntdev-common.h" 50#ifdef CONFIG_XEN_GNTDEV_DMABUF 51#include "gntdev-dmabuf.h" 52#endif 53 54MODULE_LICENSE("GPL"); 55MODULE_AUTHOR("Derek G. Murray <Derek.Murray@cl.cam.ac.uk>, " 56 "Gerd Hoffmann <kraxel@redhat.com>"); 57MODULE_DESCRIPTION("User-space granted page access driver"); 58 59static unsigned int limit = 64*1024; 60module_param(limit, uint, 0644); 61MODULE_PARM_DESC(limit, 62 "Maximum number of grants that may be mapped by one mapping request"); 63 64/* True in PV mode, false otherwise */ 65static int use_ptemod; 66 67static void unmap_grant_pages(struct gntdev_grant_map *map, 68 int offset, int pages); 69 70static struct miscdevice gntdev_miscdev; 71 72/* ------------------------------------------------------------------ */ 73 74bool gntdev_test_page_count(unsigned int count) 75{ 76 return !count || count > limit; 77} 78 79static void gntdev_print_maps(struct gntdev_priv *priv, 80 char *text, int text_index) 81{ 82#ifdef DEBUG 83 struct gntdev_grant_map *map; 84 85 pr_debug("%s: maps list (priv %p)\n", __func__, priv); 86 list_for_each_entry(map, &priv->maps, next) 87 pr_debug(" index %2d, count %2d %s\n", 88 map->index, map->count, 89 map->index == text_index && text ? text : ""); 90#endif 91} 92 93static void gntdev_free_map(struct gntdev_grant_map *map) 94{ 95 if (map == NULL) 96 return; 97 98#ifdef CONFIG_XEN_GRANT_DMA_ALLOC 99 if (map->dma_vaddr) { 100 struct gnttab_dma_alloc_args args; 101 102 args.dev = map->dma_dev; 103 args.coherent = !!(map->dma_flags & GNTDEV_DMA_FLAG_COHERENT); 104 args.nr_pages = map->count; 105 args.pages = map->pages; 106 args.frames = map->frames; 107 args.vaddr = map->dma_vaddr; 108 args.dev_bus_addr = map->dma_bus_addr; 109 110 gnttab_dma_free_pages(&args); 111 } else 112#endif 113 if (map->pages) 114 gnttab_free_pages(map->count, map->pages); 115 116#ifdef CONFIG_XEN_GRANT_DMA_ALLOC 117 kvfree(map->frames); 118#endif 119 kvfree(map->pages); 120 kvfree(map->grants); 121 kvfree(map->map_ops); 122 kvfree(map->unmap_ops); 123 kvfree(map->kmap_ops); 124 kvfree(map->kunmap_ops); 125 kvfree(map->being_removed); 126 kfree(map); 127} 128 129struct gntdev_grant_map *gntdev_alloc_map(struct gntdev_priv *priv, int count, 130 int dma_flags) 131{ 132 struct gntdev_grant_map *add; 133 int i; 134 135 add = kzalloc(sizeof(*add), GFP_KERNEL); 136 if (NULL == add) 137 return NULL; 138 139 add->grants = kvcalloc(count, sizeof(add->grants[0]), GFP_KERNEL); 140 add->map_ops = kvcalloc(count, sizeof(add->map_ops[0]), GFP_KERNEL); 141 add->unmap_ops = kvcalloc(count, sizeof(add->unmap_ops[0]), GFP_KERNEL); 142 add->kmap_ops = kvcalloc(count, sizeof(add->kmap_ops[0]), GFP_KERNEL); 143 add->kunmap_ops = kvcalloc(count, 144 sizeof(add->kunmap_ops[0]), GFP_KERNEL); 145 add->pages = kvcalloc(count, sizeof(add->pages[0]), GFP_KERNEL); 146 add->being_removed = 147 kvcalloc(count, sizeof(add->being_removed[0]), GFP_KERNEL); 148 if (NULL == add->grants || 149 NULL == add->map_ops || 150 NULL == add->unmap_ops || 151 NULL == add->kmap_ops || 152 NULL == add->kunmap_ops || 153 NULL == add->pages || 154 NULL == add->being_removed) 155 goto err; 156 157#ifdef CONFIG_XEN_GRANT_DMA_ALLOC 158 add->dma_flags = dma_flags; 159 160 /* 161 * Check if this mapping is requested to be backed 162 * by a DMA buffer. 163 */ 164 if (dma_flags & (GNTDEV_DMA_FLAG_WC | GNTDEV_DMA_FLAG_COHERENT)) { 165 struct gnttab_dma_alloc_args args; 166 167 add->frames = kvcalloc(count, sizeof(add->frames[0]), 168 GFP_KERNEL); 169 if (!add->frames) 170 goto err; 171 172 /* Remember the device, so we can free DMA memory. */ 173 add->dma_dev = priv->dma_dev; 174 175 args.dev = priv->dma_dev; 176 args.coherent = !!(dma_flags & GNTDEV_DMA_FLAG_COHERENT); 177 args.nr_pages = count; 178 args.pages = add->pages; 179 args.frames = add->frames; 180 181 if (gnttab_dma_alloc_pages(&args)) 182 goto err; 183 184 add->dma_vaddr = args.vaddr; 185 add->dma_bus_addr = args.dev_bus_addr; 186 } else 187#endif 188 if (gnttab_alloc_pages(count, add->pages)) 189 goto err; 190 191 for (i = 0; i < count; i++) { 192 add->map_ops[i].handle = -1; 193 add->unmap_ops[i].handle = -1; 194 add->kmap_ops[i].handle = -1; 195 add->kunmap_ops[i].handle = -1; 196 } 197 198 add->index = 0; 199 add->count = count; 200 refcount_set(&add->users, 1); 201 202 return add; 203 204err: 205 gntdev_free_map(add); 206 return NULL; 207} 208 209void gntdev_add_map(struct gntdev_priv *priv, struct gntdev_grant_map *add) 210{ 211 struct gntdev_grant_map *map; 212 213 list_for_each_entry(map, &priv->maps, next) { 214 if (add->index + add->count < map->index) { 215 list_add_tail(&add->next, &map->next); 216 goto done; 217 } 218 add->index = map->index + map->count; 219 } 220 list_add_tail(&add->next, &priv->maps); 221 222done: 223 gntdev_print_maps(priv, "[new]", add->index); 224} 225 226static struct gntdev_grant_map *gntdev_find_map_index(struct gntdev_priv *priv, 227 int index, int count) 228{ 229 struct gntdev_grant_map *map; 230 231 list_for_each_entry(map, &priv->maps, next) { 232 if (map->index != index) 233 continue; 234 if (count && map->count != count) 235 continue; 236 return map; 237 } 238 return NULL; 239} 240 241void gntdev_put_map(struct gntdev_priv *priv, struct gntdev_grant_map *map) 242{ 243 if (!map) 244 return; 245 246 if (!refcount_dec_and_test(&map->users)) 247 return; 248 249 if (map->pages && !use_ptemod) { 250 /* 251 * Increment the reference count. This ensures that the 252 * subsequent call to unmap_grant_pages() will not wind up 253 * re-entering itself. It *can* wind up calling 254 * gntdev_put_map() recursively, but such calls will be with a 255 * reference count greater than 1, so they will return before 256 * this code is reached. The recursion depth is thus limited to 257 * 1. Do NOT use refcount_inc() here, as it will detect that 258 * the reference count is zero and WARN(). 259 */ 260 refcount_set(&map->users, 1); 261 262 /* 263 * Unmap the grants. This may or may not be asynchronous, so it 264 * is possible that the reference count is 1 on return, but it 265 * could also be greater than 1. 266 */ 267 unmap_grant_pages(map, 0, map->count); 268 269 /* Check if the memory now needs to be freed */ 270 if (!refcount_dec_and_test(&map->users)) 271 return; 272 273 /* 274 * All pages have been returned to the hypervisor, so free the 275 * map. 276 */ 277 } 278 279 if (use_ptemod && map->notifier_init) 280 mmu_interval_notifier_remove(&map->notifier); 281 282 if (map->notify.flags & UNMAP_NOTIFY_SEND_EVENT) { 283 notify_remote_via_evtchn(map->notify.event); 284 evtchn_put(map->notify.event); 285 } 286 gntdev_free_map(map); 287} 288 289/* ------------------------------------------------------------------ */ 290 291static int find_grant_ptes(pte_t *pte, unsigned long addr, void *data) 292{ 293 struct gntdev_grant_map *map = data; 294 unsigned int pgnr = (addr - map->pages_vm_start) >> PAGE_SHIFT; 295 int flags = map->flags | GNTMAP_application_map | GNTMAP_contains_pte | 296 (1 << _GNTMAP_guest_avail0); 297 u64 pte_maddr; 298 299 BUG_ON(pgnr >= map->count); 300 pte_maddr = arbitrary_virt_to_machine(pte).maddr; 301 302 gnttab_set_map_op(&map->map_ops[pgnr], pte_maddr, flags, 303 map->grants[pgnr].ref, 304 map->grants[pgnr].domid); 305 gnttab_set_unmap_op(&map->unmap_ops[pgnr], pte_maddr, flags, 306 -1 /* handle */); 307 return 0; 308} 309 310int gntdev_map_grant_pages(struct gntdev_grant_map *map) 311{ 312 size_t alloced = 0; 313 int i, err = 0; 314 315 if (!use_ptemod) { 316 /* Note: it could already be mapped */ 317 if (map->map_ops[0].handle != -1) 318 return 0; 319 for (i = 0; i < map->count; i++) { 320 unsigned long addr = (unsigned long) 321 pfn_to_kaddr(page_to_pfn(map->pages[i])); 322 gnttab_set_map_op(&map->map_ops[i], addr, map->flags, 323 map->grants[i].ref, 324 map->grants[i].domid); 325 gnttab_set_unmap_op(&map->unmap_ops[i], addr, 326 map->flags, -1 /* handle */); 327 } 328 } else { 329 /* 330 * Setup the map_ops corresponding to the pte entries pointing 331 * to the kernel linear addresses of the struct pages. 332 * These ptes are completely different from the user ptes dealt 333 * with find_grant_ptes. 334 * Note that GNTMAP_device_map isn't needed here: The 335 * dev_bus_addr output field gets consumed only from ->map_ops, 336 * and by not requesting it when mapping we also avoid needing 337 * to mirror dev_bus_addr into ->unmap_ops (and holding an extra 338 * reference to the page in the hypervisor). 339 */ 340 unsigned int flags = (map->flags & ~GNTMAP_device_map) | 341 GNTMAP_host_map; 342 343 for (i = 0; i < map->count; i++) { 344 unsigned long address = (unsigned long) 345 pfn_to_kaddr(page_to_pfn(map->pages[i])); 346 BUG_ON(PageHighMem(map->pages[i])); 347 348 gnttab_set_map_op(&map->kmap_ops[i], address, flags, 349 map->grants[i].ref, 350 map->grants[i].domid); 351 gnttab_set_unmap_op(&map->kunmap_ops[i], address, 352 flags, -1); 353 } 354 } 355 356 pr_debug("map %d+%d\n", map->index, map->count); 357 err = gnttab_map_refs(map->map_ops, use_ptemod ? map->kmap_ops : NULL, 358 map->pages, map->count); 359 360 for (i = 0; i < map->count; i++) { 361 if (map->map_ops[i].status == GNTST_okay) { 362 map->unmap_ops[i].handle = map->map_ops[i].handle; 363 alloced++; 364 } else if (!err) 365 err = -EINVAL; 366 367 if (map->flags & GNTMAP_device_map) 368 map->unmap_ops[i].dev_bus_addr = map->map_ops[i].dev_bus_addr; 369 370 if (use_ptemod) { 371 if (map->kmap_ops[i].status == GNTST_okay) { 372 alloced++; 373 map->kunmap_ops[i].handle = map->kmap_ops[i].handle; 374 } else if (!err) 375 err = -EINVAL; 376 } 377 } 378 atomic_add(alloced, &map->live_grants); 379 return err; 380} 381 382static void __unmap_grant_pages_done(int result, 383 struct gntab_unmap_queue_data *data) 384{ 385 unsigned int i; 386 struct gntdev_grant_map *map = data->data; 387 unsigned int offset = data->unmap_ops - map->unmap_ops; 388 int successful_unmaps = 0; 389 int live_grants; 390 391 for (i = 0; i < data->count; i++) { 392 if (map->unmap_ops[offset + i].status == GNTST_okay && 393 map->unmap_ops[offset + i].handle != -1) 394 successful_unmaps++; 395 396 WARN_ON(map->unmap_ops[offset+i].status && 397 map->unmap_ops[offset+i].handle != -1); 398 pr_debug("unmap handle=%d st=%d\n", 399 map->unmap_ops[offset+i].handle, 400 map->unmap_ops[offset+i].status); 401 map->unmap_ops[offset+i].handle = -1; 402 if (use_ptemod) { 403 if (map->kunmap_ops[offset + i].status == GNTST_okay && 404 map->kunmap_ops[offset + i].handle != -1) 405 successful_unmaps++; 406 407 WARN_ON(map->kunmap_ops[offset+i].status && 408 map->kunmap_ops[offset+i].handle != -1); 409 pr_debug("kunmap handle=%u st=%d\n", 410 map->kunmap_ops[offset+i].handle, 411 map->kunmap_ops[offset+i].status); 412 map->kunmap_ops[offset+i].handle = -1; 413 } 414 } 415 416 /* 417 * Decrease the live-grant counter. This must happen after the loop to 418 * prevent premature reuse of the grants by gnttab_mmap(). 419 */ 420 live_grants = atomic_sub_return(successful_unmaps, &map->live_grants); 421 if (WARN_ON(live_grants < 0)) 422 pr_err("%s: live_grants became negative (%d) after unmapping %d pages!\n", 423 __func__, live_grants, successful_unmaps); 424 425 /* Release reference taken by __unmap_grant_pages */ 426 gntdev_put_map(NULL, map); 427} 428 429static void __unmap_grant_pages(struct gntdev_grant_map *map, int offset, 430 int pages) 431{ 432 if (map->notify.flags & UNMAP_NOTIFY_CLEAR_BYTE) { 433 int pgno = (map->notify.addr >> PAGE_SHIFT); 434 435 if (pgno >= offset && pgno < offset + pages) { 436 /* No need for kmap, pages are in lowmem */ 437 uint8_t *tmp = pfn_to_kaddr(page_to_pfn(map->pages[pgno])); 438 439 tmp[map->notify.addr & (PAGE_SIZE-1)] = 0; 440 map->notify.flags &= ~UNMAP_NOTIFY_CLEAR_BYTE; 441 } 442 } 443 444 map->unmap_data.unmap_ops = map->unmap_ops + offset; 445 map->unmap_data.kunmap_ops = use_ptemod ? map->kunmap_ops + offset : NULL; 446 map->unmap_data.pages = map->pages + offset; 447 map->unmap_data.count = pages; 448 map->unmap_data.done = __unmap_grant_pages_done; 449 map->unmap_data.data = map; 450 refcount_inc(&map->users); /* to keep map alive during async call below */ 451 452 gnttab_unmap_refs_async(&map->unmap_data); 453} 454 455static void unmap_grant_pages(struct gntdev_grant_map *map, int offset, 456 int pages) 457{ 458 int range; 459 460 if (atomic_read(&map->live_grants) == 0) 461 return; /* Nothing to do */ 462 463 pr_debug("unmap %d+%d [%d+%d]\n", map->index, map->count, offset, pages); 464 465 /* It is possible the requested range will have a "hole" where we 466 * already unmapped some of the grants. Only unmap valid ranges. 467 */ 468 while (pages) { 469 while (pages && map->being_removed[offset]) { 470 offset++; 471 pages--; 472 } 473 range = 0; 474 while (range < pages) { 475 if (map->being_removed[offset + range]) 476 break; 477 map->being_removed[offset + range] = true; 478 range++; 479 } 480 if (range) 481 __unmap_grant_pages(map, offset, range); 482 offset += range; 483 pages -= range; 484 } 485} 486 487/* ------------------------------------------------------------------ */ 488 489static void gntdev_vma_open(struct vm_area_struct *vma) 490{ 491 struct gntdev_grant_map *map = vma->vm_private_data; 492 493 pr_debug("gntdev_vma_open %p\n", vma); 494 refcount_inc(&map->users); 495} 496 497static void gntdev_vma_close(struct vm_area_struct *vma) 498{ 499 struct gntdev_grant_map *map = vma->vm_private_data; 500 struct file *file = vma->vm_file; 501 struct gntdev_priv *priv = file->private_data; 502 503 pr_debug("gntdev_vma_close %p\n", vma); 504 505 vma->vm_private_data = NULL; 506 gntdev_put_map(priv, map); 507} 508 509static struct page *gntdev_vma_find_special_page(struct vm_area_struct *vma, 510 unsigned long addr) 511{ 512 struct gntdev_grant_map *map = vma->vm_private_data; 513 514 return map->pages[(addr - map->pages_vm_start) >> PAGE_SHIFT]; 515} 516 517static const struct vm_operations_struct gntdev_vmops = { 518 .open = gntdev_vma_open, 519 .close = gntdev_vma_close, 520 .find_special_page = gntdev_vma_find_special_page, 521}; 522 523/* ------------------------------------------------------------------ */ 524 525static bool gntdev_invalidate(struct mmu_interval_notifier *mn, 526 const struct mmu_notifier_range *range, 527 unsigned long cur_seq) 528{ 529 struct gntdev_grant_map *map = 530 container_of(mn, struct gntdev_grant_map, notifier); 531 unsigned long mstart, mend; 532 unsigned long map_start, map_end; 533 534 if (!mmu_notifier_range_blockable(range)) 535 return false; 536 537 map_start = map->pages_vm_start; 538 map_end = map->pages_vm_start + (map->count << PAGE_SHIFT); 539 540 /* 541 * If the VMA is split or otherwise changed the notifier is not 542 * updated, but we don't want to process VA's outside the modified 543 * VMA. FIXME: It would be much more understandable to just prevent 544 * modifying the VMA in the first place. 545 */ 546 if (map_start >= range->end || map_end <= range->start) 547 return true; 548 549 mstart = max(range->start, map_start); 550 mend = min(range->end, map_end); 551 pr_debug("map %d+%d (%lx %lx), range %lx %lx, mrange %lx %lx\n", 552 map->index, map->count, map_start, map_end, 553 range->start, range->end, mstart, mend); 554 unmap_grant_pages(map, (mstart - map_start) >> PAGE_SHIFT, 555 (mend - mstart) >> PAGE_SHIFT); 556 557 return true; 558} 559 560static const struct mmu_interval_notifier_ops gntdev_mmu_ops = { 561 .invalidate = gntdev_invalidate, 562}; 563 564/* ------------------------------------------------------------------ */ 565 566static int gntdev_open(struct inode *inode, struct file *flip) 567{ 568 struct gntdev_priv *priv; 569 570 priv = kzalloc(sizeof(*priv), GFP_KERNEL); 571 if (!priv) 572 return -ENOMEM; 573 574 INIT_LIST_HEAD(&priv->maps); 575 mutex_init(&priv->lock); 576 577#ifdef CONFIG_XEN_GNTDEV_DMABUF 578 priv->dmabuf_priv = gntdev_dmabuf_init(flip); 579 if (IS_ERR(priv->dmabuf_priv)) { 580 int ret = PTR_ERR(priv->dmabuf_priv); 581 582 kfree(priv); 583 return ret; 584 } 585#endif 586 587 flip->private_data = priv; 588#ifdef CONFIG_XEN_GRANT_DMA_ALLOC 589 priv->dma_dev = gntdev_miscdev.this_device; 590 dma_coerce_mask_and_coherent(priv->dma_dev, DMA_BIT_MASK(64)); 591#endif 592 pr_debug("priv %p\n", priv); 593 594 return 0; 595} 596 597static int gntdev_release(struct inode *inode, struct file *flip) 598{ 599 struct gntdev_priv *priv = flip->private_data; 600 struct gntdev_grant_map *map; 601 602 pr_debug("priv %p\n", priv); 603 604 mutex_lock(&priv->lock); 605 while (!list_empty(&priv->maps)) { 606 map = list_entry(priv->maps.next, 607 struct gntdev_grant_map, next); 608 list_del(&map->next); 609 gntdev_put_map(NULL /* already removed */, map); 610 } 611 mutex_unlock(&priv->lock); 612 613#ifdef CONFIG_XEN_GNTDEV_DMABUF 614 gntdev_dmabuf_fini(priv->dmabuf_priv); 615#endif 616 617 kfree(priv); 618 return 0; 619} 620 621static long gntdev_ioctl_map_grant_ref(struct gntdev_priv *priv, 622 struct ioctl_gntdev_map_grant_ref __user *u) 623{ 624 struct ioctl_gntdev_map_grant_ref op; 625 struct gntdev_grant_map *map; 626 int err; 627 628 if (copy_from_user(&op, u, sizeof(op)) != 0) 629 return -EFAULT; 630 pr_debug("priv %p, add %d\n", priv, op.count); 631 if (unlikely(gntdev_test_page_count(op.count))) 632 return -EINVAL; 633 634 err = -ENOMEM; 635 map = gntdev_alloc_map(priv, op.count, 0 /* This is not a dma-buf. */); 636 if (!map) 637 return err; 638 639 if (copy_from_user(map->grants, &u->refs, 640 sizeof(map->grants[0]) * op.count) != 0) { 641 gntdev_put_map(NULL, map); 642 return -EFAULT; 643 } 644 645 mutex_lock(&priv->lock); 646 gntdev_add_map(priv, map); 647 op.index = map->index << PAGE_SHIFT; 648 mutex_unlock(&priv->lock); 649 650 if (copy_to_user(u, &op, sizeof(op)) != 0) 651 return -EFAULT; 652 653 return 0; 654} 655 656static long gntdev_ioctl_unmap_grant_ref(struct gntdev_priv *priv, 657 struct ioctl_gntdev_unmap_grant_ref __user *u) 658{ 659 struct ioctl_gntdev_unmap_grant_ref op; 660 struct gntdev_grant_map *map; 661 int err = -ENOENT; 662 663 if (copy_from_user(&op, u, sizeof(op)) != 0) 664 return -EFAULT; 665 pr_debug("priv %p, del %d+%d\n", priv, (int)op.index, (int)op.count); 666 667 mutex_lock(&priv->lock); 668 map = gntdev_find_map_index(priv, op.index >> PAGE_SHIFT, op.count); 669 if (map) { 670 list_del(&map->next); 671 err = 0; 672 } 673 mutex_unlock(&priv->lock); 674 if (map) 675 gntdev_put_map(priv, map); 676 return err; 677} 678 679static long gntdev_ioctl_get_offset_for_vaddr(struct gntdev_priv *priv, 680 struct ioctl_gntdev_get_offset_for_vaddr __user *u) 681{ 682 struct ioctl_gntdev_get_offset_for_vaddr op; 683 struct vm_area_struct *vma; 684 struct gntdev_grant_map *map; 685 int rv = -EINVAL; 686 687 if (copy_from_user(&op, u, sizeof(op)) != 0) 688 return -EFAULT; 689 pr_debug("priv %p, offset for vaddr %lx\n", priv, (unsigned long)op.vaddr); 690 691 mmap_read_lock(current->mm); 692 vma = find_vma(current->mm, op.vaddr); 693 if (!vma || vma->vm_ops != &gntdev_vmops) 694 goto out_unlock; 695 696 map = vma->vm_private_data; 697 if (!map) 698 goto out_unlock; 699 700 op.offset = map->index << PAGE_SHIFT; 701 op.count = map->count; 702 rv = 0; 703 704 out_unlock: 705 mmap_read_unlock(current->mm); 706 707 if (rv == 0 && copy_to_user(u, &op, sizeof(op)) != 0) 708 return -EFAULT; 709 return rv; 710} 711 712static long gntdev_ioctl_notify(struct gntdev_priv *priv, void __user *u) 713{ 714 struct ioctl_gntdev_unmap_notify op; 715 struct gntdev_grant_map *map; 716 int rc; 717 int out_flags; 718 evtchn_port_t out_event; 719 720 if (copy_from_user(&op, u, sizeof(op))) 721 return -EFAULT; 722 723 if (op.action & ~(UNMAP_NOTIFY_CLEAR_BYTE|UNMAP_NOTIFY_SEND_EVENT)) 724 return -EINVAL; 725 726 /* We need to grab a reference to the event channel we are going to use 727 * to send the notify before releasing the reference we may already have 728 * (if someone has called this ioctl twice). This is required so that 729 * it is possible to change the clear_byte part of the notification 730 * without disturbing the event channel part, which may now be the last 731 * reference to that event channel. 732 */ 733 if (op.action & UNMAP_NOTIFY_SEND_EVENT) { 734 if (evtchn_get(op.event_channel_port)) 735 return -EINVAL; 736 } 737 738 out_flags = op.action; 739 out_event = op.event_channel_port; 740 741 mutex_lock(&priv->lock); 742 743 list_for_each_entry(map, &priv->maps, next) { 744 uint64_t begin = map->index << PAGE_SHIFT; 745 uint64_t end = (map->index + map->count) << PAGE_SHIFT; 746 if (op.index >= begin && op.index < end) 747 goto found; 748 } 749 rc = -ENOENT; 750 goto unlock_out; 751 752 found: 753 if ((op.action & UNMAP_NOTIFY_CLEAR_BYTE) && 754 (map->flags & GNTMAP_readonly)) { 755 rc = -EINVAL; 756 goto unlock_out; 757 } 758 759 out_flags = map->notify.flags; 760 out_event = map->notify.event; 761 762 map->notify.flags = op.action; 763 map->notify.addr = op.index - (map->index << PAGE_SHIFT); 764 map->notify.event = op.event_channel_port; 765 766 rc = 0; 767 768 unlock_out: 769 mutex_unlock(&priv->lock); 770 771 /* Drop the reference to the event channel we did not save in the map */ 772 if (out_flags & UNMAP_NOTIFY_SEND_EVENT) 773 evtchn_put(out_event); 774 775 return rc; 776} 777 778#define GNTDEV_COPY_BATCH 16 779 780struct gntdev_copy_batch { 781 struct gnttab_copy ops[GNTDEV_COPY_BATCH]; 782 struct page *pages[GNTDEV_COPY_BATCH]; 783 s16 __user *status[GNTDEV_COPY_BATCH]; 784 unsigned int nr_ops; 785 unsigned int nr_pages; 786 bool writeable; 787}; 788 789static int gntdev_get_page(struct gntdev_copy_batch *batch, void __user *virt, 790 unsigned long *gfn) 791{ 792 unsigned long addr = (unsigned long)virt; 793 struct page *page; 794 unsigned long xen_pfn; 795 int ret; 796 797 ret = pin_user_pages_fast(addr, 1, batch->writeable ? FOLL_WRITE : 0, &page); 798 if (ret < 0) 799 return ret; 800 801 batch->pages[batch->nr_pages++] = page; 802 803 xen_pfn = page_to_xen_pfn(page) + XEN_PFN_DOWN(addr & ~PAGE_MASK); 804 *gfn = pfn_to_gfn(xen_pfn); 805 806 return 0; 807} 808 809static void gntdev_put_pages(struct gntdev_copy_batch *batch) 810{ 811 unpin_user_pages_dirty_lock(batch->pages, batch->nr_pages, batch->writeable); 812 batch->nr_pages = 0; 813 batch->writeable = false; 814} 815 816static int gntdev_copy(struct gntdev_copy_batch *batch) 817{ 818 unsigned int i; 819 820 gnttab_batch_copy(batch->ops, batch->nr_ops); 821 gntdev_put_pages(batch); 822 823 /* 824 * For each completed op, update the status if the op failed 825 * and all previous ops for the segment were successful. 826 */ 827 for (i = 0; i < batch->nr_ops; i++) { 828 s16 status = batch->ops[i].status; 829 s16 old_status; 830 831 if (status == GNTST_okay) 832 continue; 833 834 if (__get_user(old_status, batch->status[i])) 835 return -EFAULT; 836 837 if (old_status != GNTST_okay) 838 continue; 839 840 if (__put_user(status, batch->status[i])) 841 return -EFAULT; 842 } 843 844 batch->nr_ops = 0; 845 return 0; 846} 847 848static int gntdev_grant_copy_seg(struct gntdev_copy_batch *batch, 849 struct gntdev_grant_copy_segment *seg, 850 s16 __user *status) 851{ 852 uint16_t copied = 0; 853 854 /* 855 * Disallow local -> local copies since there is only space in 856 * batch->pages for one page per-op and this would be a very 857 * expensive memcpy(). 858 */ 859 if (!(seg->flags & (GNTCOPY_source_gref | GNTCOPY_dest_gref))) 860 return -EINVAL; 861 862 /* Can't cross page if source/dest is a grant ref. */ 863 if (seg->flags & GNTCOPY_source_gref) { 864 if (seg->source.foreign.offset + seg->len > XEN_PAGE_SIZE) 865 return -EINVAL; 866 } 867 if (seg->flags & GNTCOPY_dest_gref) { 868 if (seg->dest.foreign.offset + seg->len > XEN_PAGE_SIZE) 869 return -EINVAL; 870 } 871 872 if (put_user(GNTST_okay, status)) 873 return -EFAULT; 874 875 while (copied < seg->len) { 876 struct gnttab_copy *op; 877 void __user *virt; 878 size_t len, off; 879 unsigned long gfn; 880 int ret; 881 882 if (batch->nr_ops >= GNTDEV_COPY_BATCH) { 883 ret = gntdev_copy(batch); 884 if (ret < 0) 885 return ret; 886 } 887 888 len = seg->len - copied; 889 890 op = &batch->ops[batch->nr_ops]; 891 op->flags = 0; 892 893 if (seg->flags & GNTCOPY_source_gref) { 894 op->source.u.ref = seg->source.foreign.ref; 895 op->source.domid = seg->source.foreign.domid; 896 op->source.offset = seg->source.foreign.offset + copied; 897 op->flags |= GNTCOPY_source_gref; 898 } else { 899 virt = seg->source.virt + copied; 900 off = (unsigned long)virt & ~XEN_PAGE_MASK; 901 len = min(len, (size_t)XEN_PAGE_SIZE - off); 902 batch->writeable = false; 903 904 ret = gntdev_get_page(batch, virt, &gfn); 905 if (ret < 0) 906 return ret; 907 908 op->source.u.gmfn = gfn; 909 op->source.domid = DOMID_SELF; 910 op->source.offset = off; 911 } 912 913 if (seg->flags & GNTCOPY_dest_gref) { 914 op->dest.u.ref = seg->dest.foreign.ref; 915 op->dest.domid = seg->dest.foreign.domid; 916 op->dest.offset = seg->dest.foreign.offset + copied; 917 op->flags |= GNTCOPY_dest_gref; 918 } else { 919 virt = seg->dest.virt + copied; 920 off = (unsigned long)virt & ~XEN_PAGE_MASK; 921 len = min(len, (size_t)XEN_PAGE_SIZE - off); 922 batch->writeable = true; 923 924 ret = gntdev_get_page(batch, virt, &gfn); 925 if (ret < 0) 926 return ret; 927 928 op->dest.u.gmfn = gfn; 929 op->dest.domid = DOMID_SELF; 930 op->dest.offset = off; 931 } 932 933 op->len = len; 934 copied += len; 935 936 batch->status[batch->nr_ops] = status; 937 batch->nr_ops++; 938 } 939 940 return 0; 941} 942 943static long gntdev_ioctl_grant_copy(struct gntdev_priv *priv, void __user *u) 944{ 945 struct ioctl_gntdev_grant_copy copy; 946 struct gntdev_copy_batch batch; 947 unsigned int i; 948 int ret = 0; 949 950 if (copy_from_user(©, u, sizeof(copy))) 951 return -EFAULT; 952 953 batch.nr_ops = 0; 954 batch.nr_pages = 0; 955 956 for (i = 0; i < copy.count; i++) { 957 struct gntdev_grant_copy_segment seg; 958 959 if (copy_from_user(&seg, ©.segments[i], sizeof(seg))) { 960 ret = -EFAULT; 961 goto out; 962 } 963 964 ret = gntdev_grant_copy_seg(&batch, &seg, ©.segments[i].status); 965 if (ret < 0) 966 goto out; 967 968 cond_resched(); 969 } 970 if (batch.nr_ops) 971 ret = gntdev_copy(&batch); 972 return ret; 973 974 out: 975 gntdev_put_pages(&batch); 976 return ret; 977} 978 979static long gntdev_ioctl(struct file *flip, 980 unsigned int cmd, unsigned long arg) 981{ 982 struct gntdev_priv *priv = flip->private_data; 983 void __user *ptr = (void __user *)arg; 984 985 switch (cmd) { 986 case IOCTL_GNTDEV_MAP_GRANT_REF: 987 return gntdev_ioctl_map_grant_ref(priv, ptr); 988 989 case IOCTL_GNTDEV_UNMAP_GRANT_REF: 990 return gntdev_ioctl_unmap_grant_ref(priv, ptr); 991 992 case IOCTL_GNTDEV_GET_OFFSET_FOR_VADDR: 993 return gntdev_ioctl_get_offset_for_vaddr(priv, ptr); 994 995 case IOCTL_GNTDEV_SET_UNMAP_NOTIFY: 996 return gntdev_ioctl_notify(priv, ptr); 997 998 case IOCTL_GNTDEV_GRANT_COPY: 999 return gntdev_ioctl_grant_copy(priv, ptr); 1000 1001#ifdef CONFIG_XEN_GNTDEV_DMABUF 1002 case IOCTL_GNTDEV_DMABUF_EXP_FROM_REFS: 1003 return gntdev_ioctl_dmabuf_exp_from_refs(priv, use_ptemod, ptr); 1004 1005 case IOCTL_GNTDEV_DMABUF_EXP_WAIT_RELEASED: 1006 return gntdev_ioctl_dmabuf_exp_wait_released(priv, ptr); 1007 1008 case IOCTL_GNTDEV_DMABUF_IMP_TO_REFS: 1009 return gntdev_ioctl_dmabuf_imp_to_refs(priv, ptr); 1010 1011 case IOCTL_GNTDEV_DMABUF_IMP_RELEASE: 1012 return gntdev_ioctl_dmabuf_imp_release(priv, ptr); 1013#endif 1014 1015 default: 1016 pr_debug("priv %p, unknown cmd %x\n", priv, cmd); 1017 return -ENOIOCTLCMD; 1018 } 1019 1020 return 0; 1021} 1022 1023static int gntdev_mmap(struct file *flip, struct vm_area_struct *vma) 1024{ 1025 struct gntdev_priv *priv = flip->private_data; 1026 int index = vma->vm_pgoff; 1027 int count = vma_pages(vma); 1028 struct gntdev_grant_map *map; 1029 int err = -EINVAL; 1030 1031 if ((vma->vm_flags & VM_WRITE) && !(vma->vm_flags & VM_SHARED)) 1032 return -EINVAL; 1033 1034 pr_debug("map %d+%d at %lx (pgoff %lx)\n", 1035 index, count, vma->vm_start, vma->vm_pgoff); 1036 1037 mutex_lock(&priv->lock); 1038 map = gntdev_find_map_index(priv, index, count); 1039 if (!map) 1040 goto unlock_out; 1041 if (!atomic_add_unless(&map->in_use, 1, 1)) 1042 goto unlock_out; 1043 1044 refcount_inc(&map->users); 1045 1046 vma->vm_ops = &gntdev_vmops; 1047 1048 vma->vm_flags |= VM_DONTEXPAND | VM_DONTDUMP | VM_MIXEDMAP; 1049 1050 if (use_ptemod) 1051 vma->vm_flags |= VM_DONTCOPY; 1052 1053 vma->vm_private_data = map; 1054 if (map->flags) { 1055 if ((vma->vm_flags & VM_WRITE) && 1056 (map->flags & GNTMAP_readonly)) 1057 goto out_unlock_put; 1058 } else { 1059 map->flags = GNTMAP_host_map; 1060 if (!(vma->vm_flags & VM_WRITE)) 1061 map->flags |= GNTMAP_readonly; 1062 } 1063 1064 map->pages_vm_start = vma->vm_start; 1065 1066 if (use_ptemod) { 1067 err = mmu_interval_notifier_insert_locked( 1068 &map->notifier, vma->vm_mm, vma->vm_start, 1069 vma->vm_end - vma->vm_start, &gntdev_mmu_ops); 1070 if (err) 1071 goto out_unlock_put; 1072 1073 map->notifier_init = true; 1074 } 1075 mutex_unlock(&priv->lock); 1076 1077 if (use_ptemod) { 1078 /* 1079 * gntdev takes the address of the PTE in find_grant_ptes() and 1080 * passes it to the hypervisor in gntdev_map_grant_pages(). The 1081 * purpose of the notifier is to prevent the hypervisor pointer 1082 * to the PTE from going stale. 1083 * 1084 * Since this vma's mappings can't be touched without the 1085 * mmap_lock, and we are holding it now, there is no need for 1086 * the notifier_range locking pattern. 1087 */ 1088 mmu_interval_read_begin(&map->notifier); 1089 1090 err = apply_to_page_range(vma->vm_mm, vma->vm_start, 1091 vma->vm_end - vma->vm_start, 1092 find_grant_ptes, map); 1093 if (err) { 1094 pr_warn("find_grant_ptes() failure.\n"); 1095 goto out_put_map; 1096 } 1097 } 1098 1099 err = gntdev_map_grant_pages(map); 1100 if (err) 1101 goto out_put_map; 1102 1103 if (!use_ptemod) { 1104 err = vm_map_pages_zero(vma, map->pages, map->count); 1105 if (err) 1106 goto out_put_map; 1107 } 1108 1109 return 0; 1110 1111unlock_out: 1112 mutex_unlock(&priv->lock); 1113 return err; 1114 1115out_unlock_put: 1116 mutex_unlock(&priv->lock); 1117out_put_map: 1118 if (use_ptemod) 1119 unmap_grant_pages(map, 0, map->count); 1120 gntdev_put_map(priv, map); 1121 return err; 1122} 1123 1124static const struct file_operations gntdev_fops = { 1125 .owner = THIS_MODULE, 1126 .open = gntdev_open, 1127 .release = gntdev_release, 1128 .mmap = gntdev_mmap, 1129 .unlocked_ioctl = gntdev_ioctl 1130}; 1131 1132static struct miscdevice gntdev_miscdev = { 1133 .minor = MISC_DYNAMIC_MINOR, 1134 .name = "xen/gntdev", 1135 .fops = &gntdev_fops, 1136}; 1137 1138/* ------------------------------------------------------------------ */ 1139 1140static int __init gntdev_init(void) 1141{ 1142 int err; 1143 1144 if (!xen_domain()) 1145 return -ENODEV; 1146 1147 use_ptemod = !xen_feature(XENFEAT_auto_translated_physmap); 1148 1149 err = misc_register(&gntdev_miscdev); 1150 if (err != 0) { 1151 pr_err("Could not register gntdev device\n"); 1152 return err; 1153 } 1154 return 0; 1155} 1156 1157static void __exit gntdev_exit(void) 1158{ 1159 misc_deregister(&gntdev_miscdev); 1160} 1161 1162module_init(gntdev_init); 1163module_exit(gntdev_exit); 1164 1165/* ------------------------------------------------------------------ */ 1166