book3s_64_vio.c 18.4 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
/*
 * 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.
 *
 * Copyright 2010 Paul Mackerras, IBM Corp. <paulus@au1.ibm.com>
 * Copyright 2011 David Gibson, IBM Corporation <dwg@au1.ibm.com>
17
 * Copyright 2016 Alexey Kardashevskiy, IBM Corporation <aik@au1.ibm.com>
18 19 20 21 22 23 24 25 26
 */

#include <linux/types.h>
#include <linux/string.h>
#include <linux/kvm.h>
#include <linux/kvm_host.h>
#include <linux/highmem.h>
#include <linux/gfp.h>
#include <linux/slab.h>
27
#include <linux/sched/signal.h>
28 29 30
#include <linux/hugetlb.h>
#include <linux/list.h>
#include <linux/anon_inodes.h>
31 32
#include <linux/iommu.h>
#include <linux/file.h>
33 34 35

#include <asm/kvm_ppc.h>
#include <asm/kvm_book3s.h>
36
#include <asm/book3s/64/mmu-hash.h>
37 38 39 40 41
#include <asm/hvcall.h>
#include <asm/synch.h>
#include <asm/ppc-opcode.h>
#include <asm/kvm_host.h>
#include <asm/udbg.h>
42
#include <asm/iommu.h>
43
#include <asm/tce.h>
44
#include <asm/mmu_context.h>
45

46
static unsigned long kvmppc_tce_pages(unsigned long iommu_pages)
47
{
48
	return ALIGN(iommu_pages * sizeof(u64), PAGE_SIZE) / PAGE_SIZE;
49 50
}

51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95
static unsigned long kvmppc_stt_pages(unsigned long tce_pages)
{
	unsigned long stt_bytes = sizeof(struct kvmppc_spapr_tce_table) +
			(tce_pages * sizeof(struct page *));

	return tce_pages + ALIGN(stt_bytes, PAGE_SIZE) / PAGE_SIZE;
}

static long kvmppc_account_memlimit(unsigned long stt_pages, bool inc)
{
	long ret = 0;

	if (!current || !current->mm)
		return ret; /* process exited */

	down_write(&current->mm->mmap_sem);

	if (inc) {
		unsigned long locked, lock_limit;

		locked = current->mm->locked_vm + stt_pages;
		lock_limit = rlimit(RLIMIT_MEMLOCK) >> PAGE_SHIFT;
		if (locked > lock_limit && !capable(CAP_IPC_LOCK))
			ret = -ENOMEM;
		else
			current->mm->locked_vm += stt_pages;
	} else {
		if (WARN_ON_ONCE(stt_pages > current->mm->locked_vm))
			stt_pages = current->mm->locked_vm;

		current->mm->locked_vm -= stt_pages;
	}

	pr_debug("[%d] RLIMIT_MEMLOCK KVM %c%ld %ld/%ld%s\n", current->pid,
			inc ? '+' : '-',
			stt_pages << PAGE_SHIFT,
			current->mm->locked_vm << PAGE_SHIFT,
			rlimit(RLIMIT_MEMLOCK),
			ret ? " - exceeded" : "");

	up_write(&current->mm->mmap_sem);

	return ret;
}

96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176
static void kvm_spapr_tce_iommu_table_free(struct rcu_head *head)
{
	struct kvmppc_spapr_tce_iommu_table *stit = container_of(head,
			struct kvmppc_spapr_tce_iommu_table, rcu);

	iommu_tce_table_put(stit->tbl);

	kfree(stit);
}

static void kvm_spapr_tce_liobn_put(struct kref *kref)
{
	struct kvmppc_spapr_tce_iommu_table *stit = container_of(kref,
			struct kvmppc_spapr_tce_iommu_table, kref);

	list_del_rcu(&stit->next);

	call_rcu(&stit->rcu, kvm_spapr_tce_iommu_table_free);
}

extern void kvm_spapr_tce_release_iommu_group(struct kvm *kvm,
		struct iommu_group *grp)
{
	int i;
	struct kvmppc_spapr_tce_table *stt;
	struct kvmppc_spapr_tce_iommu_table *stit, *tmp;
	struct iommu_table_group *table_group = NULL;

	list_for_each_entry_rcu(stt, &kvm->arch.spapr_tce_tables, list) {

		table_group = iommu_group_get_iommudata(grp);
		if (WARN_ON(!table_group))
			continue;

		list_for_each_entry_safe(stit, tmp, &stt->iommu_tables, next) {
			for (i = 0; i < IOMMU_TABLE_GROUP_MAX_TABLES; ++i) {
				if (table_group->tables[i] != stit->tbl)
					continue;

				kref_put(&stit->kref, kvm_spapr_tce_liobn_put);
			}
		}
	}
}

extern long kvm_spapr_tce_attach_iommu_group(struct kvm *kvm, int tablefd,
		struct iommu_group *grp)
{
	struct kvmppc_spapr_tce_table *stt = NULL;
	bool found = false;
	struct iommu_table *tbl = NULL;
	struct iommu_table_group *table_group;
	long i;
	struct kvmppc_spapr_tce_iommu_table *stit;
	struct fd f;

	f = fdget(tablefd);
	if (!f.file)
		return -EBADF;

	list_for_each_entry_rcu(stt, &kvm->arch.spapr_tce_tables, list) {
		if (stt == f.file->private_data) {
			found = true;
			break;
		}
	}

	fdput(f);

	if (!found)
		return -EINVAL;

	table_group = iommu_group_get_iommudata(grp);
	if (WARN_ON(!table_group))
		return -EFAULT;

	for (i = 0; i < IOMMU_TABLE_GROUP_MAX_TABLES; ++i) {
		struct iommu_table *tbltmp = table_group->tables[i];

		if (!tbltmp)
			continue;
177 178 179 180
		/* Make sure hardware table parameters are compatible */
		if ((tbltmp->it_page_shift <= stt->page_shift) &&
				(tbltmp->it_offset << tbltmp->it_page_shift ==
				 stt->offset << stt->page_shift) &&
181
				(tbltmp->it_size << tbltmp->it_page_shift >=
182
				 stt->size << stt->page_shift)) {
183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223
			/*
			 * Reference the table to avoid races with
			 * add/remove DMA windows.
			 */
			tbl = iommu_tce_table_get(tbltmp);
			break;
		}
	}
	if (!tbl)
		return -EINVAL;

	list_for_each_entry_rcu(stit, &stt->iommu_tables, next) {
		if (tbl != stit->tbl)
			continue;

		if (!kref_get_unless_zero(&stit->kref)) {
			/* stit is being destroyed */
			iommu_tce_table_put(tbl);
			return -ENOTTY;
		}
		/*
		 * The table is already known to this KVM, we just increased
		 * its KVM reference counter and can return.
		 */
		return 0;
	}

	stit = kzalloc(sizeof(*stit), GFP_KERNEL);
	if (!stit) {
		iommu_tce_table_put(tbl);
		return -ENOMEM;
	}

	stit->tbl = tbl;
	kref_init(&stit->kref);

	list_add_rcu(&stit->next, &stt->iommu_tables);

	return 0;
}

224
static void release_spapr_tce_table(struct rcu_head *head)
225
{
226 227
	struct kvmppc_spapr_tce_table *stt = container_of(head,
			struct kvmppc_spapr_tce_table, rcu);
228
	unsigned long i, npages = kvmppc_tce_pages(stt->size);
229

230
	for (i = 0; i < npages; i++)
231 232
		if (stt->pages[i])
			__free_page(stt->pages[i]);
233

234
	kfree(stt);
235 236
}

237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257
static struct page *kvm_spapr_get_tce_page(struct kvmppc_spapr_tce_table *stt,
		unsigned long sttpage)
{
	struct page *page = stt->pages[sttpage];

	if (page)
		return page;

	mutex_lock(&stt->alloc_lock);
	page = stt->pages[sttpage];
	if (!page) {
		page = alloc_page(GFP_KERNEL | __GFP_ZERO);
		WARN_ON_ONCE(!page);
		if (page)
			stt->pages[sttpage] = page;
	}
	mutex_unlock(&stt->alloc_lock);

	return page;
}

258
static vm_fault_t kvm_spapr_tce_fault(struct vm_fault *vmf)
259
{
260
	struct kvmppc_spapr_tce_table *stt = vmf->vma->vm_file->private_data;
261 262
	struct page *page;

263
	if (vmf->pgoff >= kvmppc_tce_pages(stt->size))
264 265
		return VM_FAULT_SIGBUS;

266 267 268 269
	page = kvm_spapr_get_tce_page(stt, vmf->pgoff);
	if (!page)
		return VM_FAULT_OOM;

270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287
	get_page(page);
	vmf->page = page;
	return 0;
}

static const struct vm_operations_struct kvm_spapr_tce_vm_ops = {
	.fault = kvm_spapr_tce_fault,
};

static int kvm_spapr_tce_mmap(struct file *file, struct vm_area_struct *vma)
{
	vma->vm_ops = &kvm_spapr_tce_vm_ops;
	return 0;
}

static int kvm_spapr_tce_release(struct inode *inode, struct file *filp)
{
	struct kvmppc_spapr_tce_table *stt = filp->private_data;
288
	struct kvmppc_spapr_tce_iommu_table *stit, *tmp;
289
	struct kvm *kvm = stt->kvm;
290

291
	mutex_lock(&kvm->lock);
292
	list_del_rcu(&stt->list);
293
	mutex_unlock(&kvm->lock);
294

295 296 297 298 299 300 301 302
	list_for_each_entry_safe(stit, tmp, &stt->iommu_tables, next) {
		WARN_ON(!kref_read(&stit->kref));
		while (1) {
			if (kref_put(&stit->kref, kvm_spapr_tce_liobn_put))
				break;
		}
	}

303 304
	kvm_put_kvm(stt->kvm);

305
	kvmppc_account_memlimit(
306
		kvmppc_stt_pages(kvmppc_tce_pages(stt->size)), false);
307 308
	call_rcu(&stt->rcu, release_spapr_tce_table);

309 310 311
	return 0;
}

312
static const struct file_operations kvm_spapr_tce_fops = {
313 314 315 316 317
	.mmap           = kvm_spapr_tce_mmap,
	.release	= kvm_spapr_tce_release,
};

long kvm_vm_ioctl_create_spapr_tce(struct kvm *kvm,
318
				   struct kvm_create_spapr_tce_64 *args)
319 320
{
	struct kvmppc_spapr_tce_table *stt = NULL;
321
	struct kvmppc_spapr_tce_table *siter;
322
	unsigned long npages, size = args->size;
323 324
	int ret = -ENOMEM;

325 326
	if (!args->size || args->page_shift < 12 || args->page_shift > 34 ||
		(args->offset + args->size > (ULLONG_MAX >> args->page_shift)))
327 328
		return -EINVAL;

329
	npages = kvmppc_tce_pages(size);
330
	ret = kvmppc_account_memlimit(kvmppc_stt_pages(npages), true);
331 332
	if (ret)
		return ret;
333

334
	ret = -ENOMEM;
335 336 337
	stt = kzalloc(sizeof(*stt) + npages * sizeof(struct page *),
		      GFP_KERNEL);
	if (!stt)
338
		goto fail_acct;
339 340

	stt->liobn = args->liobn;
341 342
	stt->page_shift = args->page_shift;
	stt->offset = args->offset;
343
	stt->size = size;
344
	stt->kvm = kvm;
345
	mutex_init(&stt->alloc_lock);
346
	INIT_LIST_HEAD_RCU(&stt->iommu_tables);
347 348

	mutex_lock(&kvm->lock);
349 350 351 352 353 354 355 356 357 358

	/* Check this LIOBN hasn't been previously allocated */
	ret = 0;
	list_for_each_entry(siter, &kvm->arch.spapr_tce_tables, list) {
		if (siter->liobn == args->liobn) {
			ret = -EBUSY;
			break;
		}
	}

359
	kvm_get_kvm(kvm);
360 361 362 363
	if (!ret)
		ret = anon_inode_getfd("kvm-spapr-tce", &kvm_spapr_tce_fops,
				       stt, O_RDWR | O_CLOEXEC);

364
	if (ret >= 0)
365
		list_add_rcu(&stt->list, &kvm->arch.spapr_tce_tables);
366 367
	else
		kvm_put_kvm(kvm);
368 369 370

	mutex_unlock(&kvm->lock);

371 372
	if (ret >= 0)
		return ret;
373

374 375 376
	kfree(stt);
 fail_acct:
	kvmppc_account_memlimit(kvmppc_stt_pages(npages), false);
377 378
	return ret;
}
379

380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395
static long kvmppc_tce_to_ua(struct kvm *kvm, unsigned long tce,
		unsigned long *ua)
{
	unsigned long gfn = tce >> PAGE_SHIFT;
	struct kvm_memory_slot *memslot;

	memslot = search_memslots(kvm_memslots(kvm), gfn);
	if (!memslot)
		return -EINVAL;

	*ua = __gfn_to_hva_memslot(memslot, gfn) |
		(tce & ~(PAGE_MASK | TCE_PCI_READ | TCE_PCI_WRITE));

	return 0;
}

396 397 398 399 400 401 402 403 404 405 406 407 408 409 410
static long kvmppc_tce_validate(struct kvmppc_spapr_tce_table *stt,
		unsigned long tce)
{
	unsigned long gpa = tce & ~(TCE_PCI_READ | TCE_PCI_WRITE);
	enum dma_data_direction dir = iommu_tce_direction(tce);
	struct kvmppc_spapr_tce_iommu_table *stit;
	unsigned long ua = 0;

	/* Allow userspace to poison TCE table */
	if (dir == DMA_NONE)
		return H_SUCCESS;

	if (iommu_tce_check_gpa(stt->page_shift, gpa))
		return H_TOO_HARD;

411
	if (kvmppc_tce_to_ua(stt->kvm, tce, &ua))
412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429
		return H_TOO_HARD;

	list_for_each_entry_rcu(stit, &stt->iommu_tables, next) {
		unsigned long hpa = 0;
		struct mm_iommu_table_group_mem_t *mem;
		long shift = stit->tbl->it_page_shift;

		mem = mm_iommu_lookup(stt->kvm->mm, ua, 1ULL << shift);
		if (!mem)
			return H_TOO_HARD;

		if (mm_iommu_ua_to_hpa(mem, ua, shift, &hpa))
			return H_TOO_HARD;
	}

	return H_SUCCESS;
}

430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459
/*
 * Handles TCE requests for emulated devices.
 * Puts guest TCE values to the table and expects user space to convert them.
 * Cannot fail so kvmppc_tce_validate must be called before it.
 */
static void kvmppc_tce_put(struct kvmppc_spapr_tce_table *stt,
		unsigned long idx, unsigned long tce)
{
	struct page *page;
	u64 *tbl;
	unsigned long sttpage;

	idx -= stt->offset;
	sttpage = idx / TCES_PER_PAGE;
	page = stt->pages[sttpage];

	if (!page) {
		/* We allow any TCE, not just with read|write permissions */
		if (!tce)
			return;

		page = kvm_spapr_get_tce_page(stt, sttpage);
		if (!page)
			return;
	}
	tbl = page_to_virt(page);

	tbl[idx % TCES_PER_PAGE] = tce;
}

460 461
static void kvmppc_clear_tce(struct mm_struct *mm, struct iommu_table *tbl,
		unsigned long entry)
462 463 464 465
{
	unsigned long hpa = 0;
	enum dma_data_direction dir = DMA_NONE;

466
	iommu_tce_xchg(mm, tbl, entry, &hpa, &dir);
467 468 469 470 471 472 473
}

static long kvmppc_tce_iommu_mapped_dec(struct kvm *kvm,
		struct iommu_table *tbl, unsigned long entry)
{
	struct mm_iommu_table_group_mem_t *mem = NULL;
	const unsigned long pgsize = 1ULL << tbl->it_page_shift;
474
	__be64 *pua = IOMMU_TABLE_USERSPACE_ENTRY_RO(tbl, entry);
475 476

	if (!pua)
477
		return H_SUCCESS;
478

479
	mem = mm_iommu_lookup(kvm->mm, be64_to_cpu(*pua), pgsize);
480 481 482 483 484
	if (!mem)
		return H_TOO_HARD;

	mm_iommu_mapped_dec(mem);

485
	*pua = cpu_to_be64(0);
486 487 488 489

	return H_SUCCESS;
}

490
static long kvmppc_tce_iommu_do_unmap(struct kvm *kvm,
491 492 493 494 495 496
		struct iommu_table *tbl, unsigned long entry)
{
	enum dma_data_direction dir = DMA_NONE;
	unsigned long hpa = 0;
	long ret;

497
	if (WARN_ON_ONCE(iommu_tce_xchg(kvm->mm, tbl, entry, &hpa, &dir)))
498
		return H_TOO_HARD;
499 500 501 502 503 504

	if (dir == DMA_NONE)
		return H_SUCCESS;

	ret = kvmppc_tce_iommu_mapped_dec(kvm, tbl, entry);
	if (ret != H_SUCCESS)
505
		iommu_tce_xchg(kvm->mm, tbl, entry, &hpa, &dir);
506 507 508 509

	return ret;
}

510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527
static long kvmppc_tce_iommu_unmap(struct kvm *kvm,
		struct kvmppc_spapr_tce_table *stt, struct iommu_table *tbl,
		unsigned long entry)
{
	unsigned long i, ret = H_SUCCESS;
	unsigned long subpages = 1ULL << (stt->page_shift - tbl->it_page_shift);
	unsigned long io_entry = entry * subpages;

	for (i = 0; i < subpages; ++i) {
		ret = kvmppc_tce_iommu_do_unmap(kvm, tbl, io_entry + i);
		if (ret != H_SUCCESS)
			break;
	}

	return ret;
}

long kvmppc_tce_iommu_do_map(struct kvm *kvm, struct iommu_table *tbl,
528 529 530 531
		unsigned long entry, unsigned long ua,
		enum dma_data_direction dir)
{
	long ret;
532 533
	unsigned long hpa;
	__be64 *pua = IOMMU_TABLE_USERSPACE_ENTRY(tbl, entry);
534 535 536 537 538 539 540 541 542 543 544
	struct mm_iommu_table_group_mem_t *mem;

	if (!pua)
		/* it_userspace allocation might be delayed */
		return H_TOO_HARD;

	mem = mm_iommu_lookup(kvm->mm, ua, 1ULL << tbl->it_page_shift);
	if (!mem)
		/* This only handles v2 IOMMU type, v1 is handled via ioctl() */
		return H_TOO_HARD;

545
	if (WARN_ON_ONCE(mm_iommu_ua_to_hpa(mem, ua, tbl->it_page_shift, &hpa)))
546
		return H_TOO_HARD;
547 548

	if (mm_iommu_mapped_inc(mem))
549
		return H_TOO_HARD;
550

551
	ret = iommu_tce_xchg(kvm->mm, tbl, entry, &hpa, &dir);
552 553
	if (WARN_ON_ONCE(ret)) {
		mm_iommu_mapped_dec(mem);
554
		return H_TOO_HARD;
555 556 557 558 559
	}

	if (dir != DMA_NONE)
		kvmppc_tce_iommu_mapped_dec(kvm, tbl, entry);

560
	*pua = cpu_to_be64(ua);
561 562 563 564

	return 0;
}

565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585
static long kvmppc_tce_iommu_map(struct kvm *kvm,
		struct kvmppc_spapr_tce_table *stt, struct iommu_table *tbl,
		unsigned long entry, unsigned long ua,
		enum dma_data_direction dir)
{
	unsigned long i, pgoff, ret = H_SUCCESS;
	unsigned long subpages = 1ULL << (stt->page_shift - tbl->it_page_shift);
	unsigned long io_entry = entry * subpages;

	for (i = 0, pgoff = 0; i < subpages;
			++i, pgoff += IOMMU_PAGE_SIZE(tbl)) {

		ret = kvmppc_tce_iommu_do_map(kvm, tbl,
				io_entry + i, ua + pgoff, dir);
		if (ret != H_SUCCESS)
			break;
	}

	return ret;
}

586 587 588
long kvmppc_h_put_tce(struct kvm_vcpu *vcpu, unsigned long liobn,
		      unsigned long ioba, unsigned long tce)
{
589
	struct kvmppc_spapr_tce_table *stt;
590 591 592 593
	long ret, idx;
	struct kvmppc_spapr_tce_iommu_table *stit;
	unsigned long entry, ua = 0;
	enum dma_data_direction dir;
594 595 596 597

	/* udbg_printf("H_PUT_TCE(): liobn=0x%lx ioba=0x%lx, tce=0x%lx\n", */
	/* 	    liobn, ioba, tce); */

598
	stt = kvmppc_find_table(vcpu->kvm, liobn);
599 600 601 602 603 604 605
	if (!stt)
		return H_TOO_HARD;

	ret = kvmppc_ioba_validate(stt, ioba, 1);
	if (ret != H_SUCCESS)
		return ret;

606 607
	idx = srcu_read_lock(&vcpu->kvm->srcu);

608 609
	ret = kvmppc_tce_validate(stt, tce);
	if (ret != H_SUCCESS)
610
		goto unlock_exit;
611

612
	dir = iommu_tce_direction(tce);
613

614
	if ((dir != DMA_NONE) && kvmppc_tce_to_ua(vcpu->kvm, tce, &ua)) {
615 616 617
		ret = H_PARAMETER;
		goto unlock_exit;
	}
618 619 620 621

	entry = ioba >> stt->page_shift;

	list_for_each_entry_lockless(stit, &stt->iommu_tables, next) {
622
		if (dir == DMA_NONE)
623
			ret = kvmppc_tce_iommu_unmap(vcpu->kvm, stt,
624
					stit->tbl, entry);
625
		else
626
			ret = kvmppc_tce_iommu_map(vcpu->kvm, stt, stit->tbl,
627 628
					entry, ua, dir);

629
		if (ret != H_SUCCESS) {
630
			kvmppc_clear_tce(vcpu->kvm->mm, stit->tbl, entry);
631
			goto unlock_exit;
632
		}
633 634 635
	}

	kvmppc_tce_put(stt, entry, tce);
636

637 638 639 640
unlock_exit:
	srcu_read_unlock(&vcpu->kvm->srcu, idx);

	return ret;
641 642 643
}
EXPORT_SYMBOL_GPL(kvmppc_h_put_tce);

644 645 646 647 648 649 650
long kvmppc_h_put_tce_indirect(struct kvm_vcpu *vcpu,
		unsigned long liobn, unsigned long ioba,
		unsigned long tce_list, unsigned long npages)
{
	struct kvmppc_spapr_tce_table *stt;
	long i, ret = H_SUCCESS, idx;
	unsigned long entry, ua = 0;
651 652
	u64 __user *tces;
	u64 tce;
653
	struct kvmppc_spapr_tce_iommu_table *stit;
654

655
	stt = kvmppc_find_table(vcpu->kvm, liobn);
656 657 658
	if (!stt)
		return H_TOO_HARD;

659
	entry = ioba >> stt->page_shift;
660 661 662 663 664 665 666 667 668 669 670 671 672 673 674
	/*
	 * SPAPR spec says that the maximum size of the list is 512 TCEs
	 * so the whole table fits in 4K page
	 */
	if (npages > 512)
		return H_PARAMETER;

	if (tce_list & (SZ_4K - 1))
		return H_PARAMETER;

	ret = kvmppc_ioba_validate(stt, ioba, npages);
	if (ret != H_SUCCESS)
		return ret;

	idx = srcu_read_lock(&vcpu->kvm->srcu);
675
	if (kvmppc_tce_to_ua(vcpu->kvm, tce_list, &ua)) {
676 677 678 679 680 681 682 683 684 685 686 687 688 689 690
		ret = H_TOO_HARD;
		goto unlock_exit;
	}
	tces = (u64 __user *) ua;

	for (i = 0; i < npages; ++i) {
		if (get_user(tce, tces + i)) {
			ret = H_TOO_HARD;
			goto unlock_exit;
		}
		tce = be64_to_cpu(tce);

		ret = kvmppc_tce_validate(stt, tce);
		if (ret != H_SUCCESS)
			goto unlock_exit;
691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708
	}

	for (i = 0; i < npages; ++i) {
		/*
		 * This looks unsafe, because we validate, then regrab
		 * the TCE from userspace which could have been changed by
		 * another thread.
		 *
		 * But it actually is safe, because the relevant checks will be
		 * re-executed in the following code.  If userspace tries to
		 * change this dodgily it will result in a messier failure mode
		 * but won't threaten the host.
		 */
		if (get_user(tce, tces + i)) {
			ret = H_TOO_HARD;
			goto unlock_exit;
		}
		tce = be64_to_cpu(tce);
709

710
		if (kvmppc_tce_to_ua(vcpu->kvm, tce, &ua))
711 712 713
			return H_PARAMETER;

		list_for_each_entry_lockless(stit, &stt->iommu_tables, next) {
714
			ret = kvmppc_tce_iommu_map(vcpu->kvm, stt,
715 716 717
					stit->tbl, entry + i, ua,
					iommu_tce_direction(tce));

718
			if (ret != H_SUCCESS) {
719 720
				kvmppc_clear_tce(vcpu->kvm->mm, stit->tbl,
						entry);
721
				goto unlock_exit;
722
			}
723 724
		}

725 726 727 728 729 730 731 732 733
		kvmppc_tce_put(stt, entry + i, tce);
	}

unlock_exit:
	srcu_read_unlock(&vcpu->kvm->srcu, idx);

	return ret;
}
EXPORT_SYMBOL_GPL(kvmppc_h_put_tce_indirect);
734 735 736 737 738 739 740

long kvmppc_h_stuff_tce(struct kvm_vcpu *vcpu,
		unsigned long liobn, unsigned long ioba,
		unsigned long tce_value, unsigned long npages)
{
	struct kvmppc_spapr_tce_table *stt;
	long i, ret;
741
	struct kvmppc_spapr_tce_iommu_table *stit;
742

743
	stt = kvmppc_find_table(vcpu->kvm, liobn);
744 745 746 747 748 749 750 751 752 753 754
	if (!stt)
		return H_TOO_HARD;

	ret = kvmppc_ioba_validate(stt, ioba, npages);
	if (ret != H_SUCCESS)
		return ret;

	/* Check permission bits only to allow userspace poison TCE for debug */
	if (tce_value & (TCE_PCI_WRITE | TCE_PCI_READ))
		return H_PARAMETER;

755
	list_for_each_entry_lockless(stit, &stt->iommu_tables, next) {
756
		unsigned long entry = ioba >> stt->page_shift;
757 758

		for (i = 0; i < npages; ++i) {
759
			ret = kvmppc_tce_iommu_unmap(vcpu->kvm, stt,
760 761 762 763 764 765 766 767 768
					stit->tbl, entry + i);

			if (ret == H_SUCCESS)
				continue;

			if (ret == H_TOO_HARD)
				return ret;

			WARN_ON_ONCE(1);
769
			kvmppc_clear_tce(vcpu->kvm->mm, stit->tbl, entry);
770 771 772
		}
	}

773 774 775 776 777 778
	for (i = 0; i < npages; ++i, ioba += (1ULL << stt->page_shift))
		kvmppc_tce_put(stt, ioba >> stt->page_shift, tce_value);

	return H_SUCCESS;
}
EXPORT_SYMBOL_GPL(kvmppc_h_stuff_tce);