1 /****************************************************************************** 2 * balloon.c 3 * 4 * Xen balloon driver - enables returning/claiming memory to/from Xen. 5 * 6 * Copyright (c) 2003, B Dragovic 7 * Copyright (c) 2003-2004, M Williamson, K Fraser 8 * Copyright (c) 2005 Dan M. Smith, IBM Corporation 9 * 10 * This file may be distributed separately from the Linux kernel, or 11 * incorporated into other software packages, subject to the following license: 12 * 13 * Permission is hereby granted, free of charge, to any person obtaining a copy 14 * of this source file (the "Software"), to deal in the Software without 15 * restriction, including without limitation the rights to use, copy, modify, 16 * merge, publish, distribute, sublicense, and/or sell copies of the Software, 17 * and to permit persons to whom the Software is furnished to do so, subject to 18 * the following conditions: 19 * 20 * The above copyright notice and this permission notice shall be included in 21 * all copies or substantial portions of the Software. 22 * 23 * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR 24 * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, 25 * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE 26 * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER 27 * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING 28 * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS 29 * IN THE SOFTWARE. 30 */ 31 32 #include <sys/cdefs.h> 33 __FBSDID("$FreeBSD$"); 34 35 #include <sys/param.h> 36 #include <sys/lock.h> 37 #include <sys/kernel.h> 38 #include <sys/kthread.h> 39 #include <sys/malloc.h> 40 #include <sys/mutex.h> 41 #include <sys/sysctl.h> 42 43 #include <machine/xen/xen-os.h> 44 #include <machine/xen/xenfunc.h> 45 #include <machine/xen/xenvar.h> 46 #include <xen/hypervisor.h> 47 #include <xen/xenbus/xenbusvar.h> 48 49 #include <vm/vm.h> 50 #include <vm/vm_page.h> 51 52 MALLOC_DEFINE(M_BALLOON, "Balloon", "Xen Balloon Driver"); 53 54 struct mtx balloon_mutex; 55 56 /* 57 * Protects atomic reservation decrease/increase against concurrent increases. 58 * Also protects non-atomic updates of current_pages and driver_pages, and 59 * balloon lists. 60 */ 61 struct mtx balloon_lock; 62 63 /* We increase/decrease in batches which fit in a page */ 64 static unsigned long frame_list[PAGE_SIZE / sizeof(unsigned long)]; 65 #define ARRAY_SIZE(A) (sizeof(A) / sizeof(A[0])) 66 67 struct balloon_stats { 68 /* We aim for 'current allocation' == 'target allocation'. */ 69 unsigned long current_pages; 70 unsigned long target_pages; 71 /* We may hit the hard limit in Xen. If we do then we remember it. */ 72 unsigned long hard_limit; 73 /* 74 * Drivers may alter the memory reservation independently, but they 75 * must inform the balloon driver so we avoid hitting the hard limit. 76 */ 77 unsigned long driver_pages; 78 /* Number of pages in high- and low-memory balloons. */ 79 unsigned long balloon_low; 80 unsigned long balloon_high; 81 }; 82 83 static struct balloon_stats balloon_stats; 84 #define bs balloon_stats 85 86 SYSCTL_DECL(_dev_xen); 87 SYSCTL_NODE(_dev_xen, OID_AUTO, balloon, CTLFLAG_RD, NULL, "Balloon"); 88 SYSCTL_ULONG(_dev_xen_balloon, OID_AUTO, current, CTLFLAG_RD, 89 &bs.current_pages, 0, "Current allocation"); 90 SYSCTL_ULONG(_dev_xen_balloon, OID_AUTO, target, CTLFLAG_RD, 91 &bs.target_pages, 0, "Target allocation"); 92 SYSCTL_ULONG(_dev_xen_balloon, OID_AUTO, driver_pages, CTLFLAG_RD, 93 &bs.driver_pages, 0, "Driver pages"); 94 SYSCTL_ULONG(_dev_xen_balloon, OID_AUTO, hard_limit, CTLFLAG_RD, 95 &bs.hard_limit, 0, "Xen hard limit"); 96 SYSCTL_ULONG(_dev_xen_balloon, OID_AUTO, low_mem, CTLFLAG_RD, 97 &bs.balloon_low, 0, "Low-mem balloon"); 98 SYSCTL_ULONG(_dev_xen_balloon, OID_AUTO, high_mem, CTLFLAG_RD, 99 &bs.balloon_high, 0, "High-mem balloon"); 100 101 struct balloon_entry { 102 vm_page_t page; 103 STAILQ_ENTRY(balloon_entry) list; 104 }; 105 106 /* List of ballooned pages, threaded through the mem_map array. */ 107 static STAILQ_HEAD(,balloon_entry) ballooned_pages; 108 109 /* Main work function, always executed in process context. */ 110 static void balloon_process(void *unused); 111 112 #define IPRINTK(fmt, args...) \ 113 printk(KERN_INFO "xen_mem: " fmt, ##args) 114 #define WPRINTK(fmt, args...) \ 115 printk(KERN_WARNING "xen_mem: " fmt, ##args) 116 117 /* balloon_append: add the given page to the balloon. */ 118 static void 119 balloon_append(vm_page_t page) 120 { 121 struct balloon_entry *entry; 122 123 entry = malloc(sizeof(struct balloon_entry), M_BALLOON, M_WAITOK); 124 entry->page = page; 125 STAILQ_INSERT_HEAD(&ballooned_pages, entry, list); 126 bs.balloon_low++; 127 } 128 129 /* balloon_retrieve: rescue a page from the balloon, if it is not empty. */ 130 static vm_page_t 131 balloon_retrieve(void) 132 { 133 vm_page_t page; 134 struct balloon_entry *entry; 135 136 if (STAILQ_EMPTY(&ballooned_pages)) 137 return NULL; 138 139 entry = STAILQ_FIRST(&ballooned_pages); 140 STAILQ_REMOVE_HEAD(&ballooned_pages, list); 141 142 page = entry->page; 143 free(entry, M_DEVBUF); 144 145 bs.balloon_low--; 146 147 return page; 148 } 149 150 static void 151 balloon_alarm(void *unused) 152 { 153 wakeup(balloon_process); 154 } 155 156 static unsigned long 157 current_target(void) 158 { 159 unsigned long target = min(bs.target_pages, bs.hard_limit); 160 if (target > (bs.current_pages + bs.balloon_low + bs.balloon_high)) 161 target = bs.current_pages + bs.balloon_low + bs.balloon_high; 162 return target; 163 } 164 165 static unsigned long 166 minimum_target(void) 167 { 168 #ifdef XENHVM 169 #define max_pfn physmem 170 #endif 171 unsigned long min_pages, curr_pages = current_target(); 172 173 #define MB2PAGES(mb) ((mb) << (20 - PAGE_SHIFT)) 174 /* Simple continuous piecewiese linear function: 175 * max MiB -> min MiB gradient 176 * 0 0 177 * 16 16 178 * 32 24 179 * 128 72 (1/2) 180 * 512 168 (1/4) 181 * 2048 360 (1/8) 182 * 8192 552 (1/32) 183 * 32768 1320 184 * 131072 4392 185 */ 186 if (max_pfn < MB2PAGES(128)) 187 min_pages = MB2PAGES(8) + (max_pfn >> 1); 188 else if (max_pfn < MB2PAGES(512)) 189 min_pages = MB2PAGES(40) + (max_pfn >> 2); 190 else if (max_pfn < MB2PAGES(2048)) 191 min_pages = MB2PAGES(104) + (max_pfn >> 3); 192 else 193 min_pages = MB2PAGES(296) + (max_pfn >> 5); 194 #undef MB2PAGES 195 196 /* Don't enforce growth */ 197 return min(min_pages, curr_pages); 198 #ifndef CONFIG_XEN 199 #undef max_pfn 200 #endif 201 } 202 203 static int 204 increase_reservation(unsigned long nr_pages) 205 { 206 unsigned long pfn, i; 207 struct balloon_entry *entry; 208 vm_page_t page; 209 long rc; 210 struct xen_memory_reservation reservation = { 211 .address_bits = 0, 212 .extent_order = 0, 213 .domid = DOMID_SELF 214 }; 215 216 if (nr_pages > ARRAY_SIZE(frame_list)) 217 nr_pages = ARRAY_SIZE(frame_list); 218 219 mtx_lock(&balloon_lock); 220 221 for (entry = STAILQ_FIRST(&ballooned_pages), i = 0; 222 i < nr_pages; i++, entry = STAILQ_NEXT(entry, list)) { 223 KASSERT(entry, ("ballooned_pages list corrupt")); 224 page = entry->page; 225 frame_list[i] = (VM_PAGE_TO_PHYS(page) >> PAGE_SHIFT); 226 } 227 228 set_xen_guest_handle(reservation.extent_start, frame_list); 229 reservation.nr_extents = nr_pages; 230 rc = HYPERVISOR_memory_op( 231 XENMEM_populate_physmap, &reservation); 232 if (rc < nr_pages) { 233 if (rc > 0) { 234 int ret; 235 236 /* We hit the Xen hard limit: reprobe. */ 237 reservation.nr_extents = rc; 238 ret = HYPERVISOR_memory_op(XENMEM_decrease_reservation, 239 &reservation); 240 KASSERT(ret == rc, ("HYPERVISOR_memory_op failed")); 241 } 242 if (rc >= 0) 243 bs.hard_limit = (bs.current_pages + rc - 244 bs.driver_pages); 245 goto out; 246 } 247 248 for (i = 0; i < nr_pages; i++) { 249 page = balloon_retrieve(); 250 KASSERT(page, ("balloon_retrieve failed")); 251 252 pfn = (VM_PAGE_TO_PHYS(page) >> PAGE_SHIFT); 253 KASSERT((xen_feature(XENFEAT_auto_translated_physmap) || 254 !phys_to_machine_mapping_valid(pfn)), 255 ("auto translated physmap but mapping is valid")); 256 257 set_phys_to_machine(pfn, frame_list[i]); 258 259 #ifndef XENHVM 260 /* Link back into the page tables if not highmem. */ 261 if (pfn < max_low_pfn) { 262 int ret; 263 ret = HYPERVISOR_update_va_mapping( 264 (unsigned long)__va(pfn << PAGE_SHIFT), 265 pfn_pte_ma(frame_list[i], PAGE_KERNEL), 266 0); 267 PASSING(ret == 0, 268 ("HYPERVISOR_update_va_mapping failed")); 269 } 270 #endif 271 272 /* Relinquish the page back to the allocator. */ 273 vm_page_unwire(page, 0); 274 vm_page_free(page); 275 } 276 277 bs.current_pages += nr_pages; 278 //totalram_pages = bs.current_pages; 279 280 out: 281 mtx_unlock(&balloon_lock); 282 283 return 0; 284 } 285 286 static int 287 decrease_reservation(unsigned long nr_pages) 288 { 289 unsigned long pfn, i; 290 vm_page_t page; 291 int need_sleep = 0; 292 int ret; 293 struct xen_memory_reservation reservation = { 294 .address_bits = 0, 295 .extent_order = 0, 296 .domid = DOMID_SELF 297 }; 298 299 if (nr_pages > ARRAY_SIZE(frame_list)) 300 nr_pages = ARRAY_SIZE(frame_list); 301 302 for (i = 0; i < nr_pages; i++) { 303 int color = 0; 304 if ((page = vm_page_alloc(NULL, color++, 305 VM_ALLOC_NORMAL | VM_ALLOC_NOOBJ | 306 VM_ALLOC_WIRED | VM_ALLOC_ZERO)) == NULL) { 307 nr_pages = i; 308 need_sleep = 1; 309 break; 310 } 311 312 pfn = (VM_PAGE_TO_PHYS(page) >> PAGE_SHIFT); 313 frame_list[i] = PFNTOMFN(pfn); 314 315 #if 0 316 if (!PageHighMem(page)) { 317 v = phys_to_virt(pfn << PAGE_SHIFT); 318 scrub_pages(v, 1); 319 #ifdef CONFIG_XEN 320 ret = HYPERVISOR_update_va_mapping( 321 (unsigned long)v, __pte_ma(0), 0); 322 BUG_ON(ret); 323 #endif 324 } 325 #endif 326 #ifdef CONFIG_XEN_SCRUB_PAGES 327 else { 328 v = kmap(page); 329 scrub_pages(v, 1); 330 kunmap(page); 331 } 332 #endif 333 } 334 335 #ifdef CONFIG_XEN 336 /* Ensure that ballooned highmem pages don't have kmaps. */ 337 kmap_flush_unused(); 338 flush_tlb_all(); 339 #endif 340 341 mtx_lock(&balloon_lock); 342 343 /* No more mappings: invalidate P2M and add to balloon. */ 344 for (i = 0; i < nr_pages; i++) { 345 pfn = MFNTOPFN(frame_list[i]); 346 set_phys_to_machine(pfn, INVALID_P2M_ENTRY); 347 balloon_append(PHYS_TO_VM_PAGE(pfn << PAGE_SHIFT)); 348 } 349 350 set_xen_guest_handle(reservation.extent_start, frame_list); 351 reservation.nr_extents = nr_pages; 352 ret = HYPERVISOR_memory_op(XENMEM_decrease_reservation, &reservation); 353 KASSERT(ret == nr_pages, ("HYPERVISOR_memory_op failed")); 354 355 bs.current_pages -= nr_pages; 356 //totalram_pages = bs.current_pages; 357 358 mtx_unlock(&balloon_lock); 359 360 return (need_sleep); 361 } 362 363 /* 364 * We avoid multiple worker processes conflicting via the balloon mutex. 365 * We may of course race updates of the target counts (which are protected 366 * by the balloon lock), or with changes to the Xen hard limit, but we will 367 * recover from these in time. 368 */ 369 static void 370 balloon_process(void *unused) 371 { 372 int need_sleep = 0; 373 long credit; 374 375 mtx_lock(&balloon_mutex); 376 for (;;) { 377 do { 378 credit = current_target() - bs.current_pages; 379 if (credit > 0) 380 need_sleep = (increase_reservation(credit) != 0); 381 if (credit < 0) 382 need_sleep = (decrease_reservation(-credit) != 0); 383 384 } while ((credit != 0) && !need_sleep); 385 386 /* Schedule more work if there is some still to be done. */ 387 if (current_target() != bs.current_pages) 388 timeout(balloon_alarm, NULL, ticks + hz); 389 390 msleep(balloon_process, &balloon_mutex, 0, "balloon", -1); 391 } 392 mtx_unlock(&balloon_mutex); 393 } 394 395 /* Resets the Xen limit, sets new target, and kicks off processing. */ 396 static void 397 set_new_target(unsigned long target) 398 { 399 /* No need for lock. Not read-modify-write updates. */ 400 bs.hard_limit = ~0UL; 401 bs.target_pages = max(target, minimum_target()); 402 wakeup(balloon_process); 403 } 404 405 static struct xenbus_watch target_watch = 406 { 407 .node = "memory/target" 408 }; 409 410 /* React to a change in the target key */ 411 static void 412 watch_target(struct xenbus_watch *watch, 413 const char **vec, unsigned int len) 414 { 415 unsigned long long new_target; 416 int err; 417 418 err = xenbus_scanf(XBT_NIL, "memory", "target", NULL, 419 "%llu", &new_target); 420 if (err) { 421 /* This is ok (for domain0 at least) - so just return */ 422 return; 423 } 424 425 /* The given memory/target value is in KiB, so it needs converting to 426 pages. PAGE_SHIFT converts bytes to pages, hence PAGE_SHIFT - 10. 427 */ 428 set_new_target(new_target >> (PAGE_SHIFT - 10)); 429 430 } 431 432 static void 433 balloon_init_watcher(void *arg) 434 { 435 int err; 436 437 err = register_xenbus_watch(&target_watch); 438 if (err) 439 printf("Failed to set balloon watcher\n"); 440 441 } 442 SYSINIT(balloon_init_watcher, SI_SUB_PSEUDO, SI_ORDER_ANY, 443 balloon_init_watcher, NULL); 444 445 static void 446 balloon_init(void *arg) 447 { 448 #ifndef XENHVM 449 vm_page_t page; 450 #endif 451 452 if (!is_running_on_xen()) 453 return; 454 455 mtx_init(&balloon_lock, "balloon_lock", NULL, MTX_DEF); 456 mtx_init(&balloon_mutex, "balloon_mutex", NULL, MTX_DEF); 457 458 #ifndef XENHVM 459 bs.current_pages = min(xen_start_info->nr_pages, max_pfn); 460 #else 461 bs.current_pages = physmem; 462 #endif 463 bs.target_pages = bs.current_pages; 464 bs.balloon_low = 0; 465 bs.balloon_high = 0; 466 bs.driver_pages = 0UL; 467 bs.hard_limit = ~0UL; 468 469 kproc_create(balloon_process, NULL, NULL, 0, 0, "balloon"); 470 // init_timer(&balloon_timer); 471 // balloon_timer.data = 0; 472 // balloon_timer.function = balloon_alarm; 473 474 #ifndef XENHVM 475 /* Initialise the balloon with excess memory space. */ 476 for (pfn = xen_start_info->nr_pages; pfn < max_pfn; pfn++) { 477 page = PHYS_TO_VM_PAGE(pfn << PAGE_SHIFT); 478 balloon_append(page); 479 } 480 #endif 481 482 target_watch.callback = watch_target; 483 484 return; 485 } 486 SYSINIT(balloon_init, SI_SUB_PSEUDO, SI_ORDER_ANY, balloon_init, NULL); 487 488 void balloon_update_driver_allowance(long delta); 489 490 void 491 balloon_update_driver_allowance(long delta) 492 { 493 mtx_lock(&balloon_lock); 494 bs.driver_pages += delta; 495 mtx_unlock(&balloon_lock); 496 } 497 498 #if 0 499 static int dealloc_pte_fn( 500 pte_t *pte, struct page *pte_page, unsigned long addr, void *data) 501 { 502 unsigned long mfn = pte_mfn(*pte); 503 int ret; 504 struct xen_memory_reservation reservation = { 505 .extent_start = &mfn, 506 .nr_extents = 1, 507 .extent_order = 0, 508 .domid = DOMID_SELF 509 }; 510 set_pte_at(&init_mm, addr, pte, __pte_ma(0)); 511 set_phys_to_machine(__pa(addr) >> PAGE_SHIFT, INVALID_P2M_ENTRY); 512 ret = HYPERVISOR_memory_op(XENMEM_decrease_reservation, &reservation); 513 KASSERT(ret == 1, ("HYPERVISOR_memory_op failed")); 514 return 0; 515 } 516 517 #endif 518 519 #if 0 520 vm_page_t 521 balloon_alloc_empty_page_range(unsigned long nr_pages) 522 { 523 vm_page_t pages; 524 int i, rc; 525 unsigned long *mfn_list; 526 struct xen_memory_reservation reservation = { 527 .address_bits = 0, 528 .extent_order = 0, 529 .domid = DOMID_SELF 530 }; 531 532 pages = vm_page_alloc_contig(nr_pages, 0, -1, 4, 4) 533 if (pages == NULL) 534 return NULL; 535 536 mfn_list = malloc(nr_pages*sizeof(unsigned long), M_DEVBUF, M_WAITOK); 537 538 for (i = 0; i < nr_pages; i++) { 539 mfn_list[i] = PFNTOMFN(VM_PAGE_TO_PHYS(pages[i]) >> PAGE_SHIFT); 540 PFNTOMFN(i) = INVALID_P2M_ENTRY; 541 reservation.extent_start = mfn_list; 542 reservation.nr_extents = nr_pages; 543 rc = HYPERVISOR_memory_op(XENMEM_decrease_reservation, 544 &reservation); 545 KASSERT(rc == nr_pages, ("HYPERVISOR_memory_op failed")); 546 } 547 548 current_pages -= nr_pages; 549 550 wakeup(balloon_process); 551 552 return pages; 553 } 554 555 void 556 balloon_dealloc_empty_page_range(vm_page_t page, unsigned long nr_pages) 557 { 558 unsigned long i; 559 560 for (i = 0; i < nr_pages; i++) 561 balloon_append(page + i); 562 563 wakeup(balloon_process); 564 } 565 #endif 566