mmu.c 47.2 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
/*
 * Copyright (C) 2012 - Virtual Open Systems and Columbia University
 * Author: Christoffer Dall <c.dall@virtualopensystems.com>
 *
 * This program is free software; you can redistribute it and/or modify
 * it under the terms of the GNU General Public License, version 2, as
 * published by the Free Software Foundation.
 *
 * This program is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 * GNU General Public License for more details.
 *
 * You should have received a copy of the GNU General Public License
 * along with this program; if not, write to the Free Software
 * Foundation, 51 Franklin Street, Fifth Floor, Boston, MA  02110-1301, USA.
 */
18 19 20 21

#include <linux/mman.h>
#include <linux/kvm_host.h>
#include <linux/io.h>
22
#include <linux/hugetlb.h>
C
Christoffer Dall 已提交
23
#include <trace/events/kvm.h>
24
#include <asm/pgalloc.h>
25
#include <asm/cacheflush.h>
26 27
#include <asm/kvm_arm.h>
#include <asm/kvm_mmu.h>
C
Christoffer Dall 已提交
28
#include <asm/kvm_mmio.h>
29
#include <asm/kvm_asm.h>
30
#include <asm/kvm_emulate.h>
31 32

#include "trace.h"
33 34 35

extern char  __hyp_idmap_text_start[], __hyp_idmap_text_end[];

36
static pgd_t *boot_hyp_pgd;
37
static pgd_t *hyp_pgd;
38 39
static DEFINE_MUTEX(kvm_hyp_pgd_mutex);

40 41 42 43 44
static void *init_bounce_page;
static unsigned long hyp_idmap_start;
static unsigned long hyp_idmap_end;
static phys_addr_t hyp_idmap_vector;

45
#define hyp_pgd_order get_order(PTRS_PER_PGD * sizeof(pgd_t))
46

47
#define kvm_pmd_huge(_x)	(pmd_huge(_x) || pmd_trans_huge(_x))
48
#define kvm_pud_huge(_x)	pud_huge(_x)
49

50 51 52 53 54 55
#define KVM_S2PTE_FLAG_IS_IOMAP		(1UL << 0)
#define KVM_S2_FLAG_LOGGING_ACTIVE	(1UL << 1)

static bool memslot_is_logging(struct kvm_memory_slot *memslot)
{
	return memslot->dirty_bitmap && !(memslot->flags & KVM_MEM_READONLY);
56 57 58 59 60 61 62 63 64 65 66
}

/**
 * kvm_flush_remote_tlbs() - flush all VM TLB entries for v7/8
 * @kvm:	pointer to kvm structure.
 *
 * Interface to HYP function to flush all VM TLB entries
 */
void kvm_flush_remote_tlbs(struct kvm *kvm)
{
	kvm_call_hyp(__kvm_tlb_flush_vmid, kvm);
67
}
68

69
static void kvm_tlb_flush_vmid_ipa(struct kvm *kvm, phys_addr_t ipa)
70
{
71 72 73 74 75 76 77 78
	/*
	 * This function also gets called when dealing with HYP page
	 * tables. As HYP doesn't have an associated struct kvm (and
	 * the HYP page tables are fairly static), we don't do
	 * anything there.
	 */
	if (kvm)
		kvm_call_hyp(__kvm_tlb_flush_vmid_ipa, kvm, ipa);
79 80
}

81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100
/*
 * D-Cache management functions. They take the page table entries by
 * value, as they are flushing the cache using the kernel mapping (or
 * kmap on 32bit).
 */
static void kvm_flush_dcache_pte(pte_t pte)
{
	__kvm_flush_dcache_pte(pte);
}

static void kvm_flush_dcache_pmd(pmd_t pmd)
{
	__kvm_flush_dcache_pmd(pmd);
}

static void kvm_flush_dcache_pud(pud_t pud)
{
	__kvm_flush_dcache_pud(pud);
}

101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119
/**
 * stage2_dissolve_pmd() - clear and flush huge PMD entry
 * @kvm:	pointer to kvm structure.
 * @addr:	IPA
 * @pmd:	pmd pointer for IPA
 *
 * Function clears a PMD entry, flushes addr 1st and 2nd stage TLBs. Marks all
 * pages in the range dirty.
 */
static void stage2_dissolve_pmd(struct kvm *kvm, phys_addr_t addr, pmd_t *pmd)
{
	if (!kvm_pmd_huge(*pmd))
		return;

	pmd_clear(pmd);
	kvm_tlb_flush_vmid_ipa(kvm, addr);
	put_page(virt_to_page(pmd));
}

120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151
static int mmu_topup_memory_cache(struct kvm_mmu_memory_cache *cache,
				  int min, int max)
{
	void *page;

	BUG_ON(max > KVM_NR_MEM_OBJS);
	if (cache->nobjs >= min)
		return 0;
	while (cache->nobjs < max) {
		page = (void *)__get_free_page(PGALLOC_GFP);
		if (!page)
			return -ENOMEM;
		cache->objects[cache->nobjs++] = page;
	}
	return 0;
}

static void mmu_free_memory_cache(struct kvm_mmu_memory_cache *mc)
{
	while (mc->nobjs)
		free_page((unsigned long)mc->objects[--mc->nobjs]);
}

static void *mmu_memory_cache_alloc(struct kvm_mmu_memory_cache *mc)
{
	void *p;

	BUG_ON(!mc || !mc->nobjs);
	p = mc->objects[--mc->nobjs];
	return p;
}

152
static void clear_pgd_entry(struct kvm *kvm, pgd_t *pgd, phys_addr_t addr)
153
{
154 155 156 157 158
	pud_t *pud_table __maybe_unused = pud_offset(pgd, 0);
	pgd_clear(pgd);
	kvm_tlb_flush_vmid_ipa(kvm, addr);
	pud_free(NULL, pud_table);
	put_page(virt_to_page(pgd));
159 160
}

161
static void clear_pud_entry(struct kvm *kvm, pud_t *pud, phys_addr_t addr)
162
{
163 164 165 166 167
	pmd_t *pmd_table = pmd_offset(pud, 0);
	VM_BUG_ON(pud_huge(*pud));
	pud_clear(pud);
	kvm_tlb_flush_vmid_ipa(kvm, addr);
	pmd_free(NULL, pmd_table);
168 169
	put_page(virt_to_page(pud));
}
170

171
static void clear_pmd_entry(struct kvm *kvm, pmd_t *pmd, phys_addr_t addr)
172
{
173 174 175 176 177
	pte_t *pte_table = pte_offset_kernel(pmd, 0);
	VM_BUG_ON(kvm_pmd_huge(*pmd));
	pmd_clear(pmd);
	kvm_tlb_flush_vmid_ipa(kvm, addr);
	pte_free_kernel(NULL, pte_table);
178 179 180
	put_page(virt_to_page(pmd));
}

181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200
/*
 * Unmapping vs dcache management:
 *
 * If a guest maps certain memory pages as uncached, all writes will
 * bypass the data cache and go directly to RAM.  However, the CPUs
 * can still speculate reads (not writes) and fill cache lines with
 * data.
 *
 * Those cache lines will be *clean* cache lines though, so a
 * clean+invalidate operation is equivalent to an invalidate
 * operation, because no cache lines are marked dirty.
 *
 * Those clean cache lines could be filled prior to an uncached write
 * by the guest, and the cache coherent IO subsystem would therefore
 * end up writing old data to disk.
 *
 * This is why right after unmapping a page/section and invalidating
 * the corresponding TLBs, we call kvm_flush_dcache_p*() to make sure
 * the IO subsystem will never hit in the cache.
 */
201 202
static void unmap_ptes(struct kvm *kvm, pmd_t *pmd,
		       phys_addr_t addr, phys_addr_t end)
203
{
204 205 206 207 208 209
	phys_addr_t start_addr = addr;
	pte_t *pte, *start_pte;

	start_pte = pte = pte_offset_kernel(pmd, addr);
	do {
		if (!pte_none(*pte)) {
210 211
			pte_t old_pte = *pte;

212 213
			kvm_set_pte(pte, __pte(0));
			kvm_tlb_flush_vmid_ipa(kvm, addr);
214 215 216 217 218 219

			/* No need to invalidate the cache for device mappings */
			if ((pte_val(old_pte) & PAGE_S2_DEVICE) != PAGE_S2_DEVICE)
				kvm_flush_dcache_pte(old_pte);

			put_page(virt_to_page(pte));
220 221 222
		}
	} while (pte++, addr += PAGE_SIZE, addr != end);

223
	if (kvm_pte_table_empty(kvm, start_pte))
224
		clear_pmd_entry(kvm, pmd, start_addr);
225 226
}

227 228
static void unmap_pmds(struct kvm *kvm, pud_t *pud,
		       phys_addr_t addr, phys_addr_t end)
229
{
230 231
	phys_addr_t next, start_addr = addr;
	pmd_t *pmd, *start_pmd;
232

233 234 235 236 237
	start_pmd = pmd = pmd_offset(pud, addr);
	do {
		next = kvm_pmd_addr_end(addr, end);
		if (!pmd_none(*pmd)) {
			if (kvm_pmd_huge(*pmd)) {
238 239
				pmd_t old_pmd = *pmd;

240 241
				pmd_clear(pmd);
				kvm_tlb_flush_vmid_ipa(kvm, addr);
242 243 244

				kvm_flush_dcache_pmd(old_pmd);

245 246 247 248
				put_page(virt_to_page(pmd));
			} else {
				unmap_ptes(kvm, pmd, addr, next);
			}
249
		}
250
	} while (pmd++, addr = next, addr != end);
251

252
	if (kvm_pmd_table_empty(kvm, start_pmd))
253 254
		clear_pud_entry(kvm, pud, start_addr);
}
255

256 257 258 259 260
static void unmap_puds(struct kvm *kvm, pgd_t *pgd,
		       phys_addr_t addr, phys_addr_t end)
{
	phys_addr_t next, start_addr = addr;
	pud_t *pud, *start_pud;
261

262 263 264 265 266
	start_pud = pud = pud_offset(pgd, addr);
	do {
		next = kvm_pud_addr_end(addr, end);
		if (!pud_none(*pud)) {
			if (pud_huge(*pud)) {
267 268
				pud_t old_pud = *pud;

269 270
				pud_clear(pud);
				kvm_tlb_flush_vmid_ipa(kvm, addr);
271 272 273

				kvm_flush_dcache_pud(old_pud);

274 275 276
				put_page(virt_to_page(pud));
			} else {
				unmap_pmds(kvm, pud, addr, next);
277 278
			}
		}
279
	} while (pud++, addr = next, addr != end);
280

281
	if (kvm_pud_table_empty(kvm, start_pud))
282 283 284 285 286 287 288 289 290 291 292 293 294 295
		clear_pgd_entry(kvm, pgd, start_addr);
}


static void unmap_range(struct kvm *kvm, pgd_t *pgdp,
			phys_addr_t start, u64 size)
{
	pgd_t *pgd;
	phys_addr_t addr = start, end = start + size;
	phys_addr_t next;

	pgd = pgdp + pgd_index(addr);
	do {
		next = kvm_pgd_addr_end(addr, end);
296 297
		if (!pgd_none(*pgd))
			unmap_puds(kvm, pgd, addr, next);
298
	} while (pgd++, addr = next, addr != end);
299 300
}

301 302 303 304 305 306 307
static void stage2_flush_ptes(struct kvm *kvm, pmd_t *pmd,
			      phys_addr_t addr, phys_addr_t end)
{
	pte_t *pte;

	pte = pte_offset_kernel(pmd, addr);
	do {
308 309 310
		if (!pte_none(*pte) &&
		    (pte_val(*pte) & PAGE_S2_DEVICE) != PAGE_S2_DEVICE)
			kvm_flush_dcache_pte(*pte);
311 312 313 314 315 316 317 318 319 320 321 322 323
	} while (pte++, addr += PAGE_SIZE, addr != end);
}

static void stage2_flush_pmds(struct kvm *kvm, pud_t *pud,
			      phys_addr_t addr, phys_addr_t end)
{
	pmd_t *pmd;
	phys_addr_t next;

	pmd = pmd_offset(pud, addr);
	do {
		next = kvm_pmd_addr_end(addr, end);
		if (!pmd_none(*pmd)) {
324 325 326
			if (kvm_pmd_huge(*pmd))
				kvm_flush_dcache_pmd(*pmd);
			else
327 328 329 330 331 332 333 334 335 336 337 338 339 340 341
				stage2_flush_ptes(kvm, pmd, addr, next);
		}
	} while (pmd++, addr = next, addr != end);
}

static void stage2_flush_puds(struct kvm *kvm, pgd_t *pgd,
			      phys_addr_t addr, phys_addr_t end)
{
	pud_t *pud;
	phys_addr_t next;

	pud = pud_offset(pgd, addr);
	do {
		next = kvm_pud_addr_end(addr, end);
		if (!pud_none(*pud)) {
342 343 344
			if (pud_huge(*pud))
				kvm_flush_dcache_pud(*pud);
			else
345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371
				stage2_flush_pmds(kvm, pud, addr, next);
		}
	} while (pud++, addr = next, addr != end);
}

static void stage2_flush_memslot(struct kvm *kvm,
				 struct kvm_memory_slot *memslot)
{
	phys_addr_t addr = memslot->base_gfn << PAGE_SHIFT;
	phys_addr_t end = addr + PAGE_SIZE * memslot->npages;
	phys_addr_t next;
	pgd_t *pgd;

	pgd = kvm->arch.pgd + pgd_index(addr);
	do {
		next = kvm_pgd_addr_end(addr, end);
		stage2_flush_puds(kvm, pgd, addr, next);
	} while (pgd++, addr = next, addr != end);
}

/**
 * stage2_flush_vm - Invalidate cache for pages mapped in stage 2
 * @kvm: The struct kvm pointer
 *
 * Go through the stage 2 page tables and invalidate any cache lines
 * backing memory already mapped to the VM.
 */
372
static void stage2_flush_vm(struct kvm *kvm)
373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388
{
	struct kvm_memslots *slots;
	struct kvm_memory_slot *memslot;
	int idx;

	idx = srcu_read_lock(&kvm->srcu);
	spin_lock(&kvm->mmu_lock);

	slots = kvm_memslots(kvm);
	kvm_for_each_memslot(memslot, slots)
		stage2_flush_memslot(kvm, memslot);

	spin_unlock(&kvm->mmu_lock);
	srcu_read_unlock(&kvm->srcu, idx);
}

389 390 391 392 393 394 395 396 397 398
/**
 * free_boot_hyp_pgd - free HYP boot page tables
 *
 * Free the HYP boot page tables. The bounce page is also freed.
 */
void free_boot_hyp_pgd(void)
{
	mutex_lock(&kvm_hyp_pgd_mutex);

	if (boot_hyp_pgd) {
399 400
		unmap_range(NULL, boot_hyp_pgd, hyp_idmap_start, PAGE_SIZE);
		unmap_range(NULL, boot_hyp_pgd, TRAMPOLINE_VA, PAGE_SIZE);
401
		free_pages((unsigned long)boot_hyp_pgd, hyp_pgd_order);
402 403 404 405
		boot_hyp_pgd = NULL;
	}

	if (hyp_pgd)
406
		unmap_range(NULL, hyp_pgd, TRAMPOLINE_VA, PAGE_SIZE);
407

408
	free_page((unsigned long)init_bounce_page);
409 410 411 412 413
	init_bounce_page = NULL;

	mutex_unlock(&kvm_hyp_pgd_mutex);
}

414
/**
415
 * free_hyp_pgds - free Hyp-mode page tables
416
 *
417 418 419 420 421 422
 * Assumes hyp_pgd is a page table used strictly in Hyp-mode and
 * therefore contains either mappings in the kernel memory area (above
 * PAGE_OFFSET), or device mappings in the vmalloc range (from
 * VMALLOC_START to VMALLOC_END).
 *
 * boot_hyp_pgd should only map two pages for the init code.
423
 */
424
void free_hyp_pgds(void)
425 426 427
{
	unsigned long addr;

428
	free_boot_hyp_pgd();
429

430
	mutex_lock(&kvm_hyp_pgd_mutex);
431

432 433
	if (hyp_pgd) {
		for (addr = PAGE_OFFSET; virt_addr_valid(addr); addr += PGDIR_SIZE)
434
			unmap_range(NULL, hyp_pgd, KERN_TO_HYP(addr), PGDIR_SIZE);
435
		for (addr = VMALLOC_START; is_vmalloc_addr((void*)addr); addr += PGDIR_SIZE)
436 437
			unmap_range(NULL, hyp_pgd, KERN_TO_HYP(addr), PGDIR_SIZE);

438
		free_pages((unsigned long)hyp_pgd, hyp_pgd_order);
439
		hyp_pgd = NULL;
440 441
	}

442 443 444 445
	mutex_unlock(&kvm_hyp_pgd_mutex);
}

static void create_hyp_pte_mappings(pmd_t *pmd, unsigned long start,
446 447
				    unsigned long end, unsigned long pfn,
				    pgprot_t prot)
448 449 450 451
{
	pte_t *pte;
	unsigned long addr;

452 453
	addr = start;
	do {
454 455
		pte = pte_offset_kernel(pmd, addr);
		kvm_set_pte(pte, pfn_pte(pfn, prot));
456
		get_page(virt_to_page(pte));
457
		kvm_flush_dcache_to_poc(pte, sizeof(*pte));
458
		pfn++;
459
	} while (addr += PAGE_SIZE, addr != end);
460 461 462
}

static int create_hyp_pmd_mappings(pud_t *pud, unsigned long start,
463 464
				   unsigned long end, unsigned long pfn,
				   pgprot_t prot)
465 466 467 468 469
{
	pmd_t *pmd;
	pte_t *pte;
	unsigned long addr, next;

470 471
	addr = start;
	do {
472
		pmd = pmd_offset(pud, addr);
473 474 475 476

		BUG_ON(pmd_sect(*pmd));

		if (pmd_none(*pmd)) {
477
			pte = pte_alloc_one_kernel(NULL, addr);
478 479 480 481 482
			if (!pte) {
				kvm_err("Cannot allocate Hyp pte\n");
				return -ENOMEM;
			}
			pmd_populate_kernel(NULL, pmd, pte);
483
			get_page(virt_to_page(pmd));
484
			kvm_flush_dcache_to_poc(pmd, sizeof(*pmd));
485 486 487 488
		}

		next = pmd_addr_end(addr, end);

489 490
		create_hyp_pte_mappings(pmd, addr, next, pfn, prot);
		pfn += (next - addr) >> PAGE_SHIFT;
491
	} while (addr = next, addr != end);
492 493 494 495

	return 0;
}

496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529
static int create_hyp_pud_mappings(pgd_t *pgd, unsigned long start,
				   unsigned long end, unsigned long pfn,
				   pgprot_t prot)
{
	pud_t *pud;
	pmd_t *pmd;
	unsigned long addr, next;
	int ret;

	addr = start;
	do {
		pud = pud_offset(pgd, addr);

		if (pud_none_or_clear_bad(pud)) {
			pmd = pmd_alloc_one(NULL, addr);
			if (!pmd) {
				kvm_err("Cannot allocate Hyp pmd\n");
				return -ENOMEM;
			}
			pud_populate(NULL, pud, pmd);
			get_page(virt_to_page(pud));
			kvm_flush_dcache_to_poc(pud, sizeof(*pud));
		}

		next = pud_addr_end(addr, end);
		ret = create_hyp_pmd_mappings(pud, addr, next, pfn, prot);
		if (ret)
			return ret;
		pfn += (next - addr) >> PAGE_SHIFT;
	} while (addr = next, addr != end);

	return 0;
}

530 531 532
static int __create_hyp_mappings(pgd_t *pgdp,
				 unsigned long start, unsigned long end,
				 unsigned long pfn, pgprot_t prot)
533 534 535 536 537 538 539
{
	pgd_t *pgd;
	pud_t *pud;
	unsigned long addr, next;
	int err = 0;

	mutex_lock(&kvm_hyp_pgd_mutex);
540 541 542
	addr = start & PAGE_MASK;
	end = PAGE_ALIGN(end);
	do {
543
		pgd = pgdp + pgd_index(addr);
544

545 546 547 548
		if (pgd_none(*pgd)) {
			pud = pud_alloc_one(NULL, addr);
			if (!pud) {
				kvm_err("Cannot allocate Hyp pud\n");
549 550 551
				err = -ENOMEM;
				goto out;
			}
552 553 554
			pgd_populate(NULL, pgd, pud);
			get_page(virt_to_page(pgd));
			kvm_flush_dcache_to_poc(pgd, sizeof(*pgd));
555 556 557
		}

		next = pgd_addr_end(addr, end);
558
		err = create_hyp_pud_mappings(pgd, addr, next, pfn, prot);
559 560
		if (err)
			goto out;
561
		pfn += (next - addr) >> PAGE_SHIFT;
562
	} while (addr = next, addr != end);
563 564 565 566 567
out:
	mutex_unlock(&kvm_hyp_pgd_mutex);
	return err;
}

568 569 570 571 572 573 574 575 576 577 578
static phys_addr_t kvm_kaddr_to_phys(void *kaddr)
{
	if (!is_vmalloc_addr(kaddr)) {
		BUG_ON(!virt_addr_valid(kaddr));
		return __pa(kaddr);
	} else {
		return page_to_phys(vmalloc_to_page(kaddr)) +
		       offset_in_page(kaddr);
	}
}

579
/**
580
 * create_hyp_mappings - duplicate a kernel virtual address range in Hyp mode
581 582 583
 * @from:	The virtual kernel start address of the range
 * @to:		The virtual kernel end address of the range (exclusive)
 *
584 585 586
 * The same virtual address as the kernel virtual address is also used
 * in Hyp-mode mapping (modulo HYP_PAGE_OFFSET) to the same underlying
 * physical pages.
587 588 589
 */
int create_hyp_mappings(void *from, void *to)
{
590 591
	phys_addr_t phys_addr;
	unsigned long virt_addr;
592 593 594
	unsigned long start = KERN_TO_HYP((unsigned long)from);
	unsigned long end = KERN_TO_HYP((unsigned long)to);

595 596
	start = start & PAGE_MASK;
	end = PAGE_ALIGN(end);
597

598 599
	for (virt_addr = start; virt_addr < end; virt_addr += PAGE_SIZE) {
		int err;
600

601 602 603 604 605 606 607 608 609 610
		phys_addr = kvm_kaddr_to_phys(from + virt_addr - start);
		err = __create_hyp_mappings(hyp_pgd, virt_addr,
					    virt_addr + PAGE_SIZE,
					    __phys_to_pfn(phys_addr),
					    PAGE_HYP);
		if (err)
			return err;
	}

	return 0;
611 612 613
}

/**
614 615 616
 * create_hyp_io_mappings - duplicate a kernel IO mapping into Hyp mode
 * @from:	The kernel start VA of the range
 * @to:		The kernel end VA of the range (exclusive)
617
 * @phys_addr:	The physical start address which gets mapped
618 619 620
 *
 * The resulting HYP VA is the same as the kernel VA, modulo
 * HYP_PAGE_OFFSET.
621
 */
622
int create_hyp_io_mappings(void *from, void *to, phys_addr_t phys_addr)
623
{
624 625 626 627 628 629 630 631 632
	unsigned long start = KERN_TO_HYP((unsigned long)from);
	unsigned long end = KERN_TO_HYP((unsigned long)to);

	/* Check for a valid kernel IO mapping */
	if (!is_vmalloc_addr(from) || !is_vmalloc_addr(to - 1))
		return -EINVAL;

	return __create_hyp_mappings(hyp_pgd, start, end,
				     __phys_to_pfn(phys_addr), PAGE_HYP_DEVICE);
633 634
}

635 636 637 638 639 640 641 642 643 644 645 646 647
/**
 * kvm_alloc_stage2_pgd - allocate level-1 table for stage-2 translation.
 * @kvm:	The KVM struct pointer for the VM.
 *
 * Allocates the 1st level table only of size defined by S2_PGD_ORDER (can
 * support either full 40-bit input addresses or limited to 32-bit input
 * addresses). Clears the allocated pages.
 *
 * Note we don't need locking here as this is only called when the VM is
 * created, which can only be done once.
 */
int kvm_alloc_stage2_pgd(struct kvm *kvm)
{
648
	int ret;
649 650 651 652 653 654 655
	pgd_t *pgd;

	if (kvm->arch.pgd != NULL) {
		kvm_err("kvm_arch already initialized?\n");
		return -EINVAL;
	}

656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671
	if (KVM_PREALLOC_LEVEL > 0) {
		/*
		 * Allocate fake pgd for the page table manipulation macros to
		 * work.  This is not used by the hardware and we have no
		 * alignment requirement for this allocation.
		 */
		pgd = (pgd_t *)kmalloc(PTRS_PER_S2_PGD * sizeof(pgd_t),
				       GFP_KERNEL | __GFP_ZERO);
	} else {
		/*
		 * Allocate actual first-level Stage-2 page table used by the
		 * hardware for Stage-2 page table walks.
		 */
		pgd = (pgd_t *)__get_free_pages(GFP_KERNEL | __GFP_ZERO, S2_PGD_ORDER);
	}

672 673 674
	if (!pgd)
		return -ENOMEM;

675 676 677 678
	ret = kvm_prealloc_hwpgd(kvm, pgd);
	if (ret)
		goto out_err;

679
	kvm_clean_pgd(pgd);
680 681
	kvm->arch.pgd = pgd;
	return 0;
682 683 684 685 686 687
out_err:
	if (KVM_PREALLOC_LEVEL > 0)
		kfree(pgd);
	else
		free_pages((unsigned long)pgd, S2_PGD_ORDER);
	return ret;
688 689 690 691 692 693 694 695 696 697 698 699 700 701 702
}

/**
 * unmap_stage2_range -- Clear stage2 page table entries to unmap a range
 * @kvm:   The VM pointer
 * @start: The intermediate physical base address of the range to unmap
 * @size:  The size of the area to unmap
 *
 * Clear a range of stage-2 mappings, lowering the various ref-counts.  Must
 * be called while holding mmu_lock (unless for freeing the stage2 pgd before
 * destroying the VM), otherwise another faulting VCPU may come in and mess
 * with things behind our backs.
 */
static void unmap_stage2_range(struct kvm *kvm, phys_addr_t start, u64 size)
{
703
	unmap_range(kvm, kvm->arch.pgd, start, size);
704 705
}

706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770
static void stage2_unmap_memslot(struct kvm *kvm,
				 struct kvm_memory_slot *memslot)
{
	hva_t hva = memslot->userspace_addr;
	phys_addr_t addr = memslot->base_gfn << PAGE_SHIFT;
	phys_addr_t size = PAGE_SIZE * memslot->npages;
	hva_t reg_end = hva + size;

	/*
	 * A memory region could potentially cover multiple VMAs, and any holes
	 * between them, so iterate over all of them to find out if we should
	 * unmap any of them.
	 *
	 *     +--------------------------------------------+
	 * +---------------+----------------+   +----------------+
	 * |   : VMA 1     |      VMA 2     |   |    VMA 3  :    |
	 * +---------------+----------------+   +----------------+
	 *     |               memory region                |
	 *     +--------------------------------------------+
	 */
	do {
		struct vm_area_struct *vma = find_vma(current->mm, hva);
		hva_t vm_start, vm_end;

		if (!vma || vma->vm_start >= reg_end)
			break;

		/*
		 * Take the intersection of this VMA with the memory region
		 */
		vm_start = max(hva, vma->vm_start);
		vm_end = min(reg_end, vma->vm_end);

		if (!(vma->vm_flags & VM_PFNMAP)) {
			gpa_t gpa = addr + (vm_start - memslot->userspace_addr);
			unmap_stage2_range(kvm, gpa, vm_end - vm_start);
		}
		hva = vm_end;
	} while (hva < reg_end);
}

/**
 * stage2_unmap_vm - Unmap Stage-2 RAM mappings
 * @kvm: The struct kvm pointer
 *
 * Go through the memregions and unmap any reguler RAM
 * backing memory already mapped to the VM.
 */
void stage2_unmap_vm(struct kvm *kvm)
{
	struct kvm_memslots *slots;
	struct kvm_memory_slot *memslot;
	int idx;

	idx = srcu_read_lock(&kvm->srcu);
	spin_lock(&kvm->mmu_lock);

	slots = kvm_memslots(kvm);
	kvm_for_each_memslot(memslot, slots)
		stage2_unmap_memslot(kvm, memslot);

	spin_unlock(&kvm->mmu_lock);
	srcu_read_unlock(&kvm->srcu, idx);
}

771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787
/**
 * kvm_free_stage2_pgd - free all stage-2 tables
 * @kvm:	The KVM struct pointer for the VM.
 *
 * Walks the level-1 page table pointed to by kvm->arch.pgd and frees all
 * underlying level-2 and level-3 tables before freeing the actual level-1 table
 * and setting the struct pointer to NULL.
 *
 * Note we don't need locking here as this is only called when the VM is
 * destroyed, which can only be done once.
 */
void kvm_free_stage2_pgd(struct kvm *kvm)
{
	if (kvm->arch.pgd == NULL)
		return;

	unmap_stage2_range(kvm, 0, KVM_PHYS_SIZE);
788 789 790 791 792
	kvm_free_hwpgd(kvm);
	if (KVM_PREALLOC_LEVEL > 0)
		kfree(kvm->arch.pgd);
	else
		free_pages((unsigned long)kvm->arch.pgd, S2_PGD_ORDER);
793 794 795
	kvm->arch.pgd = NULL;
}

796
static pud_t *stage2_get_pud(struct kvm *kvm, struct kvm_mmu_memory_cache *cache,
797
			     phys_addr_t addr)
798 799 800 801 802
{
	pgd_t *pgd;
	pud_t *pud;

	pgd = kvm->arch.pgd + pgd_index(addr);
803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820
	if (WARN_ON(pgd_none(*pgd))) {
		if (!cache)
			return NULL;
		pud = mmu_memory_cache_alloc(cache);
		pgd_populate(NULL, pgd, pud);
		get_page(virt_to_page(pgd));
	}

	return pud_offset(pgd, addr);
}

static pmd_t *stage2_get_pmd(struct kvm *kvm, struct kvm_mmu_memory_cache *cache,
			     phys_addr_t addr)
{
	pud_t *pud;
	pmd_t *pmd;

	pud = stage2_get_pud(kvm, cache, addr);
821 822
	if (pud_none(*pud)) {
		if (!cache)
823
			return NULL;
824 825 826
		pmd = mmu_memory_cache_alloc(cache);
		pud_populate(NULL, pud, pmd);
		get_page(virt_to_page(pud));
827 828
	}

829 830 831 832 833 834 835 836 837 838
	return pmd_offset(pud, addr);
}

static int stage2_set_pmd_huge(struct kvm *kvm, struct kvm_mmu_memory_cache
			       *cache, phys_addr_t addr, const pmd_t *new_pmd)
{
	pmd_t *pmd, old_pmd;

	pmd = stage2_get_pmd(kvm, cache, addr);
	VM_BUG_ON(!pmd);
839

840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860
	/*
	 * Mapping in huge pages should only happen through a fault.  If a
	 * page is merged into a transparent huge page, the individual
	 * subpages of that huge page should be unmapped through MMU
	 * notifiers before we get here.
	 *
	 * Merging of CompoundPages is not supported; they should become
	 * splitting first, unmapped, merged, and mapped back in on-demand.
	 */
	VM_BUG_ON(pmd_present(*pmd) && pmd_pfn(*pmd) != pmd_pfn(*new_pmd));

	old_pmd = *pmd;
	kvm_set_pmd(pmd, *new_pmd);
	if (pmd_present(old_pmd))
		kvm_tlb_flush_vmid_ipa(kvm, addr);
	else
		get_page(virt_to_page(pmd));
	return 0;
}

static int stage2_set_pte(struct kvm *kvm, struct kvm_mmu_memory_cache *cache,
861 862
			  phys_addr_t addr, const pte_t *new_pte,
			  unsigned long flags)
863 864 865
{
	pmd_t *pmd;
	pte_t *pte, old_pte;
866 867 868 869
	bool iomap = flags & KVM_S2PTE_FLAG_IS_IOMAP;
	bool logging_active = flags & KVM_S2_FLAG_LOGGING_ACTIVE;

	VM_BUG_ON(logging_active && !cache);
870

871
	/* Create stage-2 page table mapping - Levels 0 and 1 */
872 873 874 875 876 877 878 879 880
	pmd = stage2_get_pmd(kvm, cache, addr);
	if (!pmd) {
		/*
		 * Ignore calls from kvm_set_spte_hva for unallocated
		 * address ranges.
		 */
		return 0;
	}

881 882 883 884 885 886 887
	/*
	 * While dirty page logging - dissolve huge PMD, then continue on to
	 * allocate page.
	 */
	if (logging_active)
		stage2_dissolve_pmd(kvm, addr, pmd);

888
	/* Create stage-2 page mappings - Level 2 */
889 890 891 892
	if (pmd_none(*pmd)) {
		if (!cache)
			return 0; /* ignore calls from kvm_set_spte_hva */
		pte = mmu_memory_cache_alloc(cache);
893
		kvm_clean_pte(pte);
894 895
		pmd_populate_kernel(NULL, pmd, pte);
		get_page(virt_to_page(pmd));
896 897 898
	}

	pte = pte_offset_kernel(pmd, addr);
899 900 901 902 903 904 905 906

	if (iomap && pte_present(*pte))
		return -EFAULT;

	/* Create 2nd stage page table mapping - Level 3 */
	old_pte = *pte;
	kvm_set_pte(pte, *new_pte);
	if (pte_present(old_pte))
907
		kvm_tlb_flush_vmid_ipa(kvm, addr);
908 909 910 911 912 913 914 915 916 917 918 919 920 921 922
	else
		get_page(virt_to_page(pte));

	return 0;
}

/**
 * kvm_phys_addr_ioremap - map a device range to guest IPA
 *
 * @kvm:	The KVM pointer
 * @guest_ipa:	The IPA at which to insert the mapping
 * @pa:		The physical address of the device
 * @size:	The size of the mapping
 */
int kvm_phys_addr_ioremap(struct kvm *kvm, phys_addr_t guest_ipa,
923
			  phys_addr_t pa, unsigned long size, bool writable)
924 925 926 927 928 929 930 931 932 933
{
	phys_addr_t addr, end;
	int ret = 0;
	unsigned long pfn;
	struct kvm_mmu_memory_cache cache = { 0, };

	end = (guest_ipa + size + PAGE_SIZE - 1) & PAGE_MASK;
	pfn = __phys_to_pfn(pa);

	for (addr = guest_ipa; addr < end; addr += PAGE_SIZE) {
934
		pte_t pte = pfn_pte(pfn, PAGE_S2_DEVICE);
935

936 937 938
		if (writable)
			kvm_set_s2pte_writable(&pte);

939 940
		ret = mmu_topup_memory_cache(&cache, KVM_MMU_CACHE_MIN_PAGES,
						KVM_NR_MEM_OBJS);
941 942 943
		if (ret)
			goto out;
		spin_lock(&kvm->mmu_lock);
944 945
		ret = stage2_set_pte(kvm, &cache, addr, &pte,
						KVM_S2PTE_FLAG_IS_IOMAP);
946 947 948 949 950 951 952 953 954 955 956 957
		spin_unlock(&kvm->mmu_lock);
		if (ret)
			goto out;

		pfn++;
	}

out:
	mmu_free_memory_cache(&cache);
	return ret;
}

958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998
static bool transparent_hugepage_adjust(pfn_t *pfnp, phys_addr_t *ipap)
{
	pfn_t pfn = *pfnp;
	gfn_t gfn = *ipap >> PAGE_SHIFT;

	if (PageTransCompound(pfn_to_page(pfn))) {
		unsigned long mask;
		/*
		 * The address we faulted on is backed by a transparent huge
		 * page.  However, because we map the compound huge page and
		 * not the individual tail page, we need to transfer the
		 * refcount to the head page.  We have to be careful that the
		 * THP doesn't start to split while we are adjusting the
		 * refcounts.
		 *
		 * We are sure this doesn't happen, because mmu_notifier_retry
		 * was successful and we are holding the mmu_lock, so if this
		 * THP is trying to split, it will be blocked in the mmu
		 * notifier before touching any of the pages, specifically
		 * before being able to call __split_huge_page_refcount().
		 *
		 * We can therefore safely transfer the refcount from PG_tail
		 * to PG_head and switch the pfn from a tail page to the head
		 * page accordingly.
		 */
		mask = PTRS_PER_PMD - 1;
		VM_BUG_ON((gfn & mask) != (pfn & mask));
		if (pfn & mask) {
			*ipap &= PMD_MASK;
			kvm_release_pfn_clean(pfn);
			pfn &= ~mask;
			kvm_get_pfn(pfn);
			*pfnp = pfn;
		}

		return true;
	}

	return false;
}

999 1000 1001 1002 1003 1004 1005 1006
static bool kvm_is_write_fault(struct kvm_vcpu *vcpu)
{
	if (kvm_vcpu_trap_is_iabt(vcpu))
		return false;

	return kvm_vcpu_dabt_iswrite(vcpu);
}

1007 1008 1009 1010 1011
static bool kvm_is_device_pfn(unsigned long pfn)
{
	return !pfn_valid(pfn);
}

1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096
/**
 * stage2_wp_ptes - write protect PMD range
 * @pmd:	pointer to pmd entry
 * @addr:	range start address
 * @end:	range end address
 */
static void stage2_wp_ptes(pmd_t *pmd, phys_addr_t addr, phys_addr_t end)
{
	pte_t *pte;

	pte = pte_offset_kernel(pmd, addr);
	do {
		if (!pte_none(*pte)) {
			if (!kvm_s2pte_readonly(pte))
				kvm_set_s2pte_readonly(pte);
		}
	} while (pte++, addr += PAGE_SIZE, addr != end);
}

/**
 * stage2_wp_pmds - write protect PUD range
 * @pud:	pointer to pud entry
 * @addr:	range start address
 * @end:	range end address
 */
static void stage2_wp_pmds(pud_t *pud, phys_addr_t addr, phys_addr_t end)
{
	pmd_t *pmd;
	phys_addr_t next;

	pmd = pmd_offset(pud, addr);

	do {
		next = kvm_pmd_addr_end(addr, end);
		if (!pmd_none(*pmd)) {
			if (kvm_pmd_huge(*pmd)) {
				if (!kvm_s2pmd_readonly(pmd))
					kvm_set_s2pmd_readonly(pmd);
			} else {
				stage2_wp_ptes(pmd, addr, next);
			}
		}
	} while (pmd++, addr = next, addr != end);
}

/**
  * stage2_wp_puds - write protect PGD range
  * @pgd:	pointer to pgd entry
  * @addr:	range start address
  * @end:	range end address
  *
  * Process PUD entries, for a huge PUD we cause a panic.
  */
static void  stage2_wp_puds(pgd_t *pgd, phys_addr_t addr, phys_addr_t end)
{
	pud_t *pud;
	phys_addr_t next;

	pud = pud_offset(pgd, addr);
	do {
		next = kvm_pud_addr_end(addr, end);
		if (!pud_none(*pud)) {
			/* TODO:PUD not supported, revisit later if supported */
			BUG_ON(kvm_pud_huge(*pud));
			stage2_wp_pmds(pud, addr, next);
		}
	} while (pud++, addr = next, addr != end);
}

/**
 * stage2_wp_range() - write protect stage2 memory region range
 * @kvm:	The KVM pointer
 * @addr:	Start address of range
 * @end:	End address of range
 */
static void stage2_wp_range(struct kvm *kvm, phys_addr_t addr, phys_addr_t end)
{
	pgd_t *pgd;
	phys_addr_t next;

	pgd = kvm->arch.pgd + pgd_index(addr);
	do {
		/*
		 * Release kvm_mmu_lock periodically if the memory region is
		 * large. Otherwise, we may see kernel panics with
1097 1098
		 * CONFIG_DETECT_HUNG_TASK, CONFIG_LOCKUP_DETECTOR,
		 * CONFIG_LOCKDEP. Additionally, holding the lock too long
1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133
		 * will also starve other vCPUs.
		 */
		if (need_resched() || spin_needbreak(&kvm->mmu_lock))
			cond_resched_lock(&kvm->mmu_lock);

		next = kvm_pgd_addr_end(addr, end);
		if (pgd_present(*pgd))
			stage2_wp_puds(pgd, addr, next);
	} while (pgd++, addr = next, addr != end);
}

/**
 * kvm_mmu_wp_memory_region() - write protect stage 2 entries for memory slot
 * @kvm:	The KVM pointer
 * @slot:	The memory slot to write protect
 *
 * Called to start logging dirty pages after memory region
 * KVM_MEM_LOG_DIRTY_PAGES operation is called. After this function returns
 * all present PMD and PTEs are write protected in the memory region.
 * Afterwards read of dirty page log can be called.
 *
 * Acquires kvm_mmu_lock. Called with kvm->slots_lock mutex acquired,
 * serializing operations for VM memory regions.
 */
void kvm_mmu_wp_memory_region(struct kvm *kvm, int slot)
{
	struct kvm_memory_slot *memslot = id_to_memslot(kvm->memslots, slot);
	phys_addr_t start = memslot->base_gfn << PAGE_SHIFT;
	phys_addr_t end = (memslot->base_gfn + memslot->npages) << PAGE_SHIFT;

	spin_lock(&kvm->mmu_lock);
	stage2_wp_range(kvm, start, end);
	spin_unlock(&kvm->mmu_lock);
	kvm_flush_remote_tlbs(kvm);
}
1134 1135

/**
1136
 * kvm_mmu_write_protect_pt_masked() - write protect dirty pages
1137 1138 1139 1140 1141 1142 1143 1144 1145
 * @kvm:	The KVM pointer
 * @slot:	The memory slot associated with mask
 * @gfn_offset:	The gfn offset in memory slot
 * @mask:	The mask of dirty pages at offset 'gfn_offset' in this memory
 *		slot to be write protected
 *
 * Walks bits set in mask write protects the associated pte's. Caller must
 * acquire kvm_mmu_lock.
 */
1146
static void kvm_mmu_write_protect_pt_masked(struct kvm *kvm,
1147 1148 1149 1150 1151 1152 1153 1154 1155
		struct kvm_memory_slot *slot,
		gfn_t gfn_offset, unsigned long mask)
{
	phys_addr_t base_gfn = slot->base_gfn + gfn_offset;
	phys_addr_t start = (base_gfn +  __ffs(mask)) << PAGE_SHIFT;
	phys_addr_t end = (base_gfn + __fls(mask) + 1) << PAGE_SHIFT;

	stage2_wp_range(kvm, start, end);
}
1156

1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170
/*
 * kvm_arch_mmu_enable_log_dirty_pt_masked - enable dirty logging for selected
 * dirty pages.
 *
 * It calls kvm_mmu_write_protect_pt_masked to write protect selected pages to
 * enable dirty logging for them.
 */
void kvm_arch_mmu_enable_log_dirty_pt_masked(struct kvm *kvm,
		struct kvm_memory_slot *slot,
		gfn_t gfn_offset, unsigned long mask)
{
	kvm_mmu_write_protect_pt_masked(kvm, slot, gfn_offset, mask);
}

1171 1172 1173 1174 1175 1176
static void coherent_cache_guest_page(struct kvm_vcpu *vcpu, pfn_t pfn,
				      unsigned long size, bool uncached)
{
	__coherent_cache_guest_page(vcpu, pfn, size, uncached);
}

1177
static int user_mem_abort(struct kvm_vcpu *vcpu, phys_addr_t fault_ipa,
1178
			  struct kvm_memory_slot *memslot, unsigned long hva,
1179 1180 1181
			  unsigned long fault_status)
{
	int ret;
1182
	bool write_fault, writable, hugetlb = false, force_pte = false;
1183
	unsigned long mmu_seq;
1184 1185
	gfn_t gfn = fault_ipa >> PAGE_SHIFT;
	struct kvm *kvm = vcpu->kvm;
1186
	struct kvm_mmu_memory_cache *memcache = &vcpu->arch.mmu_page_cache;
1187 1188
	struct vm_area_struct *vma;
	pfn_t pfn;
1189
	pgprot_t mem_type = PAGE_S2;
1190
	bool fault_ipa_uncached;
1191 1192
	bool logging_active = memslot_is_logging(memslot);
	unsigned long flags = 0;
1193

1194
	write_fault = kvm_is_write_fault(vcpu);
1195 1196 1197 1198 1199
	if (fault_status == FSC_PERM && !write_fault) {
		kvm_err("Unexpected L2 read permission error\n");
		return -EFAULT;
	}

1200 1201 1202
	/* Let's check if we will get back a huge page backed by hugetlbfs */
	down_read(&current->mm->mmap_sem);
	vma = find_vma_intersection(current->mm, hva, hva + 1);
1203 1204 1205 1206 1207 1208
	if (unlikely(!vma)) {
		kvm_err("Failed to find VMA for hva 0x%lx\n", hva);
		up_read(&current->mm->mmap_sem);
		return -EFAULT;
	}

1209
	if (is_vm_hugetlb_page(vma) && !logging_active) {
1210 1211
		hugetlb = true;
		gfn = (fault_ipa & PMD_MASK) >> PAGE_SHIFT;
1212 1213
	} else {
		/*
1214 1215 1216 1217 1218 1219 1220
		 * Pages belonging to memslots that don't have the same
		 * alignment for userspace and IPA cannot be mapped using
		 * block descriptors even if the pages belong to a THP for
		 * the process, because the stage-2 block descriptor will
		 * cover more than a single THP and we loose atomicity for
		 * unmapping, updates, and splits of the THP or other pages
		 * in the stage-2 block range.
1221
		 */
1222 1223
		if ((memslot->userspace_addr & ~PMD_MASK) !=
		    ((memslot->base_gfn << PAGE_SHIFT) & ~PMD_MASK))
1224
			force_pte = true;
1225 1226 1227
	}
	up_read(&current->mm->mmap_sem);

1228
	/* We need minimum second+third level pages */
1229 1230
	ret = mmu_topup_memory_cache(memcache, KVM_MMU_CACHE_MIN_PAGES,
				     KVM_NR_MEM_OBJS);
1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245
	if (ret)
		return ret;

	mmu_seq = vcpu->kvm->mmu_notifier_seq;
	/*
	 * Ensure the read of mmu_notifier_seq happens before we call
	 * gfn_to_pfn_prot (which calls get_user_pages), so that we don't risk
	 * the page we just got a reference to gets unmapped before we have a
	 * chance to grab the mmu_lock, which ensure that if the page gets
	 * unmapped afterwards, the call to kvm_unmap_hva will take it away
	 * from us again properly. This smp_rmb() interacts with the smp_wmb()
	 * in kvm_mmu_notifier_invalidate_<page|range_end>.
	 */
	smp_rmb();

1246
	pfn = gfn_to_pfn_prot(kvm, gfn, write_fault, &writable);
1247 1248 1249
	if (is_error_pfn(pfn))
		return -EFAULT;

1250
	if (kvm_is_device_pfn(pfn)) {
1251
		mem_type = PAGE_S2_DEVICE;
1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268
		flags |= KVM_S2PTE_FLAG_IS_IOMAP;
	} else if (logging_active) {
		/*
		 * Faults on pages in a memslot with logging enabled
		 * should not be mapped with huge pages (it introduces churn
		 * and performance degradation), so force a pte mapping.
		 */
		force_pte = true;
		flags |= KVM_S2_FLAG_LOGGING_ACTIVE;

		/*
		 * Only actually map the page as writable if this was a write
		 * fault.
		 */
		if (!write_fault)
			writable = false;
	}
1269

1270 1271
	spin_lock(&kvm->mmu_lock);
	if (mmu_notifier_retry(kvm, mmu_seq))
1272
		goto out_unlock;
1273

1274 1275
	if (!hugetlb && !force_pte)
		hugetlb = transparent_hugepage_adjust(&pfn, &fault_ipa);
1276

1277
	fault_ipa_uncached = memslot->flags & KVM_MEMSLOT_INCOHERENT;
1278

1279
	if (hugetlb) {
1280
		pmd_t new_pmd = pfn_pmd(pfn, mem_type);
1281 1282 1283 1284 1285
		new_pmd = pmd_mkhuge(new_pmd);
		if (writable) {
			kvm_set_s2pmd_writable(&new_pmd);
			kvm_set_pfn_dirty(pfn);
		}
1286
		coherent_cache_guest_page(vcpu, pfn, PMD_SIZE, fault_ipa_uncached);
1287 1288
		ret = stage2_set_pmd_huge(kvm, memcache, fault_ipa, &new_pmd);
	} else {
1289
		pte_t new_pte = pfn_pte(pfn, mem_type);
1290

1291 1292 1293
		if (writable) {
			kvm_set_s2pte_writable(&new_pte);
			kvm_set_pfn_dirty(pfn);
1294
			mark_page_dirty(kvm, gfn);
1295
		}
1296
		coherent_cache_guest_page(vcpu, pfn, PAGE_SIZE, fault_ipa_uncached);
1297
		ret = stage2_set_pte(kvm, memcache, fault_ipa, &new_pte, flags);
1298
	}
1299

1300
out_unlock:
1301
	spin_unlock(&kvm->mmu_lock);
1302
	kvm_release_pfn_clean(pfn);
1303
	return ret;
1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317
}

/**
 * kvm_handle_guest_abort - handles all 2nd stage aborts
 * @vcpu:	the VCPU pointer
 * @run:	the kvm_run structure
 *
 * Any abort that gets to the host is almost guaranteed to be caused by a
 * missing second stage translation table entry, which can mean that either the
 * guest simply needs more memory and we must allocate an appropriate page or it
 * can mean that the guest tried to access I/O memory, which is emulated by user
 * space. The distinction is based on the IPA causing the fault and whether this
 * memory region has been registered as standard RAM by user space.
 */
1318 1319
int kvm_handle_guest_abort(struct kvm_vcpu *vcpu, struct kvm_run *run)
{
1320 1321 1322
	unsigned long fault_status;
	phys_addr_t fault_ipa;
	struct kvm_memory_slot *memslot;
1323 1324
	unsigned long hva;
	bool is_iabt, write_fault, writable;
1325 1326 1327
	gfn_t gfn;
	int ret, idx;

1328
	is_iabt = kvm_vcpu_trap_is_iabt(vcpu);
1329
	fault_ipa = kvm_vcpu_get_fault_ipa(vcpu);
1330

1331 1332
	trace_kvm_guest_fault(*vcpu_pc(vcpu), kvm_vcpu_get_hsr(vcpu),
			      kvm_vcpu_get_hfar(vcpu), fault_ipa);
1333 1334

	/* Check the stage-2 fault is trans. fault or write fault */
1335
	fault_status = kvm_vcpu_trap_get_fault_type(vcpu);
1336
	if (fault_status != FSC_FAULT && fault_status != FSC_PERM) {
1337 1338 1339 1340
		kvm_err("Unsupported FSC: EC=%#x xFSC=%#lx ESR_EL2=%#lx\n",
			kvm_vcpu_trap_get_class(vcpu),
			(unsigned long)kvm_vcpu_trap_get_fault(vcpu),
			(unsigned long)kvm_vcpu_get_hsr(vcpu));
1341 1342 1343 1344 1345 1346
		return -EFAULT;
	}

	idx = srcu_read_lock(&vcpu->kvm->srcu);

	gfn = fault_ipa >> PAGE_SHIFT;
1347 1348
	memslot = gfn_to_memslot(vcpu->kvm, gfn);
	hva = gfn_to_hva_memslot_prot(memslot, gfn, &writable);
1349
	write_fault = kvm_is_write_fault(vcpu);
1350
	if (kvm_is_error_hva(hva) || (write_fault && !writable)) {
1351 1352
		if (is_iabt) {
			/* Prefetch Abort on I/O address */
1353
			kvm_inject_pabt(vcpu, kvm_vcpu_get_hfar(vcpu));
1354 1355 1356 1357
			ret = 1;
			goto out_unlock;
		}

M
Marc Zyngier 已提交
1358 1359 1360 1361 1362 1363 1364
		/*
		 * The IPA is reported as [MAX:12], so we need to
		 * complement it with the bottom 12 bits from the
		 * faulting VA. This is always 12 bits, irrespective
		 * of the page size.
		 */
		fault_ipa |= kvm_vcpu_get_hfar(vcpu) & ((1 << 12) - 1);
C
Christoffer Dall 已提交
1365
		ret = io_mem_abort(vcpu, run, fault_ipa);
1366 1367 1368
		goto out_unlock;
	}

1369 1370 1371
	/* Userspace should not be able to register out-of-bounds IPAs */
	VM_BUG_ON(fault_ipa >= KVM_PHYS_SIZE);

1372
	ret = user_mem_abort(vcpu, fault_ipa, memslot, hva, fault_status);
1373 1374 1375 1376 1377
	if (ret == 0)
		ret = 1;
out_unlock:
	srcu_read_unlock(&vcpu->kvm->srcu, idx);
	return ret;
1378 1379
}

1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448
static void handle_hva_to_gpa(struct kvm *kvm,
			      unsigned long start,
			      unsigned long end,
			      void (*handler)(struct kvm *kvm,
					      gpa_t gpa, void *data),
			      void *data)
{
	struct kvm_memslots *slots;
	struct kvm_memory_slot *memslot;

	slots = kvm_memslots(kvm);

	/* we only care about the pages that the guest sees */
	kvm_for_each_memslot(memslot, slots) {
		unsigned long hva_start, hva_end;
		gfn_t gfn, gfn_end;

		hva_start = max(start, memslot->userspace_addr);
		hva_end = min(end, memslot->userspace_addr +
					(memslot->npages << PAGE_SHIFT));
		if (hva_start >= hva_end)
			continue;

		/*
		 * {gfn(page) | page intersects with [hva_start, hva_end)} =
		 * {gfn_start, gfn_start+1, ..., gfn_end-1}.
		 */
		gfn = hva_to_gfn_memslot(hva_start, memslot);
		gfn_end = hva_to_gfn_memslot(hva_end + PAGE_SIZE - 1, memslot);

		for (; gfn < gfn_end; ++gfn) {
			gpa_t gpa = gfn << PAGE_SHIFT;
			handler(kvm, gpa, data);
		}
	}
}

static void kvm_unmap_hva_handler(struct kvm *kvm, gpa_t gpa, void *data)
{
	unmap_stage2_range(kvm, gpa, PAGE_SIZE);
}

int kvm_unmap_hva(struct kvm *kvm, unsigned long hva)
{
	unsigned long end = hva + PAGE_SIZE;

	if (!kvm->arch.pgd)
		return 0;

	trace_kvm_unmap_hva(hva);
	handle_hva_to_gpa(kvm, hva, end, &kvm_unmap_hva_handler, NULL);
	return 0;
}

int kvm_unmap_hva_range(struct kvm *kvm,
			unsigned long start, unsigned long end)
{
	if (!kvm->arch.pgd)
		return 0;

	trace_kvm_unmap_hva_range(start, end);
	handle_hva_to_gpa(kvm, start, end, &kvm_unmap_hva_handler, NULL);
	return 0;
}

static void kvm_set_spte_handler(struct kvm *kvm, gpa_t gpa, void *data)
{
	pte_t *pte = (pte_t *)data;

1449 1450 1451 1452 1453 1454 1455 1456
	/*
	 * We can always call stage2_set_pte with KVM_S2PTE_FLAG_LOGGING_ACTIVE
	 * flag clear because MMU notifiers will have unmapped a huge PMD before
	 * calling ->change_pte() (which in turn calls kvm_set_spte_hva()) and
	 * therefore stage2_set_pte() never needs to clear out a huge PMD
	 * through this calling path.
	 */
	stage2_set_pte(kvm, NULL, gpa, pte, 0);
1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477
}


void kvm_set_spte_hva(struct kvm *kvm, unsigned long hva, pte_t pte)
{
	unsigned long end = hva + PAGE_SIZE;
	pte_t stage2_pte;

	if (!kvm->arch.pgd)
		return;

	trace_kvm_set_spte_hva(hva);
	stage2_pte = pfn_pte(pte_pfn(pte), PAGE_S2);
	handle_hva_to_gpa(kvm, hva, end, &kvm_set_spte_handler, &stage2_pte);
}

void kvm_mmu_free_memory_caches(struct kvm_vcpu *vcpu)
{
	mmu_free_memory_cache(&vcpu->arch.mmu_page_cache);
}

1478 1479 1480 1481 1482
phys_addr_t kvm_mmu_get_httbr(void)
{
	return virt_to_phys(hyp_pgd);
}

1483 1484 1485 1486 1487 1488 1489 1490 1491 1492
phys_addr_t kvm_mmu_get_boot_httbr(void)
{
	return virt_to_phys(boot_hyp_pgd);
}

phys_addr_t kvm_get_idmap_vector(void)
{
	return hyp_idmap_vector;
}

1493 1494
int kvm_mmu_init(void)
{
1495 1496
	int err;

1497 1498 1499
	hyp_idmap_start = kvm_virt_to_phys(__hyp_idmap_text_start);
	hyp_idmap_end = kvm_virt_to_phys(__hyp_idmap_text_end);
	hyp_idmap_vector = kvm_virt_to_phys(__kvm_hyp_init);
1500 1501 1502 1503 1504 1505 1506 1507 1508

	if ((hyp_idmap_start ^ hyp_idmap_end) & PAGE_MASK) {
		/*
		 * Our init code is crossing a page boundary. Allocate
		 * a bounce page, copy the code over and use that.
		 */
		size_t len = __hyp_idmap_text_end - __hyp_idmap_text_start;
		phys_addr_t phys_base;

1509
		init_bounce_page = (void *)__get_free_page(GFP_KERNEL);
1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525
		if (!init_bounce_page) {
			kvm_err("Couldn't allocate HYP init bounce page\n");
			err = -ENOMEM;
			goto out;
		}

		memcpy(init_bounce_page, __hyp_idmap_text_start, len);
		/*
		 * Warning: the code we just copied to the bounce page
		 * must be flushed to the point of coherency.
		 * Otherwise, the data may be sitting in L2, and HYP
		 * mode won't be able to observe it as it runs with
		 * caches off at that point.
		 */
		kvm_flush_dcache_to_poc(init_bounce_page, len);

1526
		phys_base = kvm_virt_to_phys(init_bounce_page);
1527 1528 1529 1530 1531 1532 1533 1534
		hyp_idmap_vector += phys_base - hyp_idmap_start;
		hyp_idmap_start = phys_base;
		hyp_idmap_end = phys_base + len;

		kvm_info("Using HYP init bounce page @%lx\n",
			 (unsigned long)phys_base);
	}

1535 1536
	hyp_pgd = (pgd_t *)__get_free_pages(GFP_KERNEL | __GFP_ZERO, hyp_pgd_order);
	boot_hyp_pgd = (pgd_t *)__get_free_pages(GFP_KERNEL | __GFP_ZERO, hyp_pgd_order);
1537

1538
	if (!hyp_pgd || !boot_hyp_pgd) {
1539
		kvm_err("Hyp mode PGD not allocated\n");
1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553
		err = -ENOMEM;
		goto out;
	}

	/* Create the idmap in the boot page tables */
	err = 	__create_hyp_mappings(boot_hyp_pgd,
				      hyp_idmap_start, hyp_idmap_end,
				      __phys_to_pfn(hyp_idmap_start),
				      PAGE_HYP);

	if (err) {
		kvm_err("Failed to idmap %lx-%lx\n",
			hyp_idmap_start, hyp_idmap_end);
		goto out;
1554 1555
	}

1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577
	/* Map the very same page at the trampoline VA */
	err = 	__create_hyp_mappings(boot_hyp_pgd,
				      TRAMPOLINE_VA, TRAMPOLINE_VA + PAGE_SIZE,
				      __phys_to_pfn(hyp_idmap_start),
				      PAGE_HYP);
	if (err) {
		kvm_err("Failed to map trampoline @%lx into boot HYP pgd\n",
			TRAMPOLINE_VA);
		goto out;
	}

	/* Map the same page again into the runtime page tables */
	err = 	__create_hyp_mappings(hyp_pgd,
				      TRAMPOLINE_VA, TRAMPOLINE_VA + PAGE_SIZE,
				      __phys_to_pfn(hyp_idmap_start),
				      PAGE_HYP);
	if (err) {
		kvm_err("Failed to map trampoline @%lx into runtime HYP pgd\n",
			TRAMPOLINE_VA);
		goto out;
	}

1578
	return 0;
1579
out:
1580
	free_hyp_pgds();
1581
	return err;
1582
}
1583 1584 1585 1586 1587 1588

void kvm_arch_commit_memory_region(struct kvm *kvm,
				   struct kvm_userspace_memory_region *mem,
				   const struct kvm_memory_slot *old,
				   enum kvm_mr_change change)
{
1589 1590 1591 1592 1593 1594 1595
	/*
	 * At this point memslot has been committed and there is an
	 * allocated dirty_bitmap[], dirty pages will be be tracked while the
	 * memory slot is write protected.
	 */
	if (change != KVM_MR_DELETE && mem->flags & KVM_MEM_LOG_DIRTY_PAGES)
		kvm_mmu_wp_memory_region(kvm, mem->slot);
1596 1597 1598 1599 1600 1601 1602
}

int kvm_arch_prepare_memory_region(struct kvm *kvm,
				   struct kvm_memory_slot *memslot,
				   struct kvm_userspace_memory_region *mem,
				   enum kvm_mr_change change)
{
1603 1604 1605 1606 1607
	hva_t hva = mem->userspace_addr;
	hva_t reg_end = hva + mem->memory_size;
	bool writable = !(mem->flags & KVM_MEM_READONLY);
	int ret = 0;

1608 1609
	if (change != KVM_MR_CREATE && change != KVM_MR_MOVE &&
			change != KVM_MR_FLAGS_ONLY)
1610 1611
		return 0;

1612 1613 1614 1615 1616 1617 1618 1619
	/*
	 * Prevent userspace from creating a memory region outside of the IPA
	 * space addressable by the KVM guest IPA space.
	 */
	if (memslot->base_gfn + memslot->npages >=
	    (KVM_PHYS_SIZE >> PAGE_SHIFT))
		return -EFAULT;

1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659
	/*
	 * A memory region could potentially cover multiple VMAs, and any holes
	 * between them, so iterate over all of them to find out if we can map
	 * any of them right now.
	 *
	 *     +--------------------------------------------+
	 * +---------------+----------------+   +----------------+
	 * |   : VMA 1     |      VMA 2     |   |    VMA 3  :    |
	 * +---------------+----------------+   +----------------+
	 *     |               memory region                |
	 *     +--------------------------------------------+
	 */
	do {
		struct vm_area_struct *vma = find_vma(current->mm, hva);
		hva_t vm_start, vm_end;

		if (!vma || vma->vm_start >= reg_end)
			break;

		/*
		 * Mapping a read-only VMA is only allowed if the
		 * memory region is configured as read-only.
		 */
		if (writable && !(vma->vm_flags & VM_WRITE)) {
			ret = -EPERM;
			break;
		}

		/*
		 * Take the intersection of this VMA with the memory region
		 */
		vm_start = max(hva, vma->vm_start);
		vm_end = min(reg_end, vma->vm_end);

		if (vma->vm_flags & VM_PFNMAP) {
			gpa_t gpa = mem->guest_phys_addr +
				    (vm_start - mem->userspace_addr);
			phys_addr_t pa = (vma->vm_pgoff << PAGE_SHIFT) +
					 vm_start - vma->vm_start;

1660 1661 1662 1663
			/* IO region dirty page logging not allowed */
			if (memslot->flags & KVM_MEM_LOG_DIRTY_PAGES)
				return -EINVAL;

1664 1665 1666 1667 1668 1669 1670 1671 1672
			ret = kvm_phys_addr_ioremap(kvm, gpa, pa,
						    vm_end - vm_start,
						    writable);
			if (ret)
				break;
		}
		hva = vm_end;
	} while (hva < reg_end);

1673 1674 1675
	if (change == KVM_MR_FLAGS_ONLY)
		return ret;

1676 1677
	spin_lock(&kvm->mmu_lock);
	if (ret)
1678
		unmap_stage2_range(kvm, mem->guest_phys_addr, mem->memory_size);
1679 1680 1681
	else
		stage2_flush_memslot(kvm, memslot);
	spin_unlock(&kvm->mmu_lock);
1682
	return ret;
1683 1684 1685 1686 1687 1688 1689 1690 1691 1692
}

void kvm_arch_free_memslot(struct kvm *kvm, struct kvm_memory_slot *free,
			   struct kvm_memory_slot *dont)
{
}

int kvm_arch_create_memslot(struct kvm *kvm, struct kvm_memory_slot *slot,
			    unsigned long npages)
{
1693 1694 1695 1696 1697 1698 1699 1700 1701
	/*
	 * Readonly memslots are not incoherent with the caches by definition,
	 * but in practice, they are used mostly to emulate ROMs or NOR flashes
	 * that the guest may consider devices and hence map as uncached.
	 * To prevent incoherency issues in these cases, tag all readonly
	 * regions as incoherent.
	 */
	if (slot->flags & KVM_MEM_READONLY)
		slot->flags |= KVM_MEMSLOT_INCOHERENT;
1702 1703 1704 1705 1706 1707 1708 1709 1710 1711 1712 1713 1714 1715
	return 0;
}

void kvm_arch_memslots_updated(struct kvm *kvm)
{
}

void kvm_arch_flush_shadow_all(struct kvm *kvm)
{
}

void kvm_arch_flush_shadow_memslot(struct kvm *kvm,
				   struct kvm_memory_slot *slot)
{
1716 1717 1718 1719 1720 1721
	gpa_t gpa = slot->base_gfn << PAGE_SHIFT;
	phys_addr_t size = slot->npages << PAGE_SHIFT;

	spin_lock(&kvm->mmu_lock);
	unmap_stage2_range(kvm, gpa, size);
	spin_unlock(&kvm->mmu_lock);
1722
}
1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790

/*
 * See note at ARMv7 ARM B1.14.4 (TL;DR: S/W ops are not easily virtualized).
 *
 * Main problems:
 * - S/W ops are local to a CPU (not broadcast)
 * - We have line migration behind our back (speculation)
 * - System caches don't support S/W at all (damn!)
 *
 * In the face of the above, the best we can do is to try and convert
 * S/W ops to VA ops. Because the guest is not allowed to infer the
 * S/W to PA mapping, it can only use S/W to nuke the whole cache,
 * which is a rather good thing for us.
 *
 * Also, it is only used when turning caches on/off ("The expected
 * usage of the cache maintenance instructions that operate by set/way
 * is associated with the cache maintenance instructions associated
 * with the powerdown and powerup of caches, if this is required by
 * the implementation.").
 *
 * We use the following policy:
 *
 * - If we trap a S/W operation, we enable VM trapping to detect
 *   caches being turned on/off, and do a full clean.
 *
 * - We flush the caches on both caches being turned on and off.
 *
 * - Once the caches are enabled, we stop trapping VM ops.
 */
void kvm_set_way_flush(struct kvm_vcpu *vcpu)
{
	unsigned long hcr = vcpu_get_hcr(vcpu);

	/*
	 * If this is the first time we do a S/W operation
	 * (i.e. HCR_TVM not set) flush the whole memory, and set the
	 * VM trapping.
	 *
	 * Otherwise, rely on the VM trapping to wait for the MMU +
	 * Caches to be turned off. At that point, we'll be able to
	 * clean the caches again.
	 */
	if (!(hcr & HCR_TVM)) {
		trace_kvm_set_way_flush(*vcpu_pc(vcpu),
					vcpu_has_cache_enabled(vcpu));
		stage2_flush_vm(vcpu->kvm);
		vcpu_set_hcr(vcpu, hcr | HCR_TVM);
	}
}

void kvm_toggle_cache(struct kvm_vcpu *vcpu, bool was_enabled)
{
	bool now_enabled = vcpu_has_cache_enabled(vcpu);

	/*
	 * If switching the MMU+caches on, need to invalidate the caches.
	 * If switching it off, need to clean the caches.
	 * Clean + invalidate does the trick always.
	 */
	if (now_enabled != was_enabled)
		stage2_flush_vm(vcpu->kvm);

	/* Caches are now on, stop trapping VM ops (until a S/W op) */
	if (now_enabled)
		vcpu_set_hcr(vcpu, vcpu_get_hcr(vcpu) & ~HCR_TVM);

	trace_kvm_toggle_cache(*vcpu_pc(vcpu), was_enabled, now_enabled);
}