Merge tag 'stable/for-linus-3.20-rc0-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/xen/tip

Pull xen features and fixes from David Vrabel:

 - Reworked handling for foreign (grant mapped) pages to simplify the
   code, enable a number of additional use cases and fix a number of
   long-standing bugs.

 - Prefer the TSC over the Xen PV clock when dom0 (and the TSC is
   stable).

 - Assorted other cleanup and minor bug fixes.

* tag 'stable/for-linus-3.20-rc0-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/xen/tip: (25 commits)
  xen/manage: Fix USB interaction issues when resuming
  xenbus: Add proper handling of XS_ERROR from Xenbus for transactions.
  xen/gntdev: provide find_special_page VMA operation
  xen/gntdev: mark userspace PTEs as special on x86 PV guests
  xen-blkback: safely unmap grants in case they are still in use
  xen/gntdev: safely unmap grants in case they are still in use
  xen/gntdev: convert priv->lock to a mutex
  xen/grant-table: add a mechanism to safely unmap pages that are in use
  xen-netback: use foreign page information from the pages themselves
  xen: mark grant mapped pages as foreign
  xen/grant-table: add helpers for allocating pages
  x86/xen: require ballooned pages for grant maps
  xen: remove scratch frames for ballooned pages and m2p override
  xen/grant-table: pre-populate kernel unmap ops for xen_gnttab_unmap_refs()
  mm: add 'foreign' alias for the 'pinned' page flag
  mm: provide a find_special_page vma operation
  x86/xen: cleanup arch/x86/xen/mmu.c
  x86/xen: add some __init annotations in arch/x86/xen/mmu.c
  x86/xen: add some __init and static annotations in arch/x86/xen/setup.c
  x86/xen: use correct types for addresses in arch/x86/xen/setup.c
  ...

arch/arm/include/asm/xen/page.h

@@ -92,7 +92,7 @@ extern int set_foreign_p2m_mapping(struct gnttab_map_grant_ref *map_ops,
 				   struct page **pages, unsigned int count);
 
 extern int clear_foreign_p2m_mapping(struct gnttab_unmap_grant_ref *unmap_ops,
-				     struct gnttab_map_grant_ref *kmap_ops,
+				     struct gnttab_unmap_grant_ref *kunmap_ops,
 				     struct page **pages, unsigned int count);
 
 bool __set_phys_to_machine(unsigned long pfn, unsigned long mfn);

arch/arm/xen/enlighten.c

@@ -29,10 +29,10 @@
 
 struct start_info _xen_start_info;
 struct start_info *xen_start_info = &_xen_start_info;
-EXPORT_SYMBOL_GPL(xen_start_info);
+EXPORT_SYMBOL(xen_start_info);
 
 enum xen_domain_type xen_domain_type = XEN_NATIVE;
-EXPORT_SYMBOL_GPL(xen_domain_type);
+EXPORT_SYMBOL(xen_domain_type);
 
 struct shared_info xen_dummy_shared_info;
 struct shared_info *HYPERVISOR_shared_info = (void *)&xen_dummy_shared_info;

arch/arm/xen/mm.c

@@ -149,7 +149,7 @@ void xen_destroy_contiguous_region(phys_addr_t pstart, unsigned int order)
 EXPORT_SYMBOL_GPL(xen_destroy_contiguous_region);
 
 struct dma_map_ops *xen_dma_ops;
-EXPORT_SYMBOL_GPL(xen_dma_ops);
+EXPORT_SYMBOL(xen_dma_ops);
 
 static struct dma_map_ops xen_swiotlb_dma_ops = {
 	.mapping_error = xen_swiotlb_dma_mapping_error,

arch/arm/xen/p2m.c

@@ -102,7 +102,7 @@ int set_foreign_p2m_mapping(struct gnttab_map_grant_ref *map_ops,
 EXPORT_SYMBOL_GPL(set_foreign_p2m_mapping);
 
 int clear_foreign_p2m_mapping(struct gnttab_unmap_grant_ref *unmap_ops,
-			      struct gnttab_map_grant_ref *kmap_ops,
+			      struct gnttab_unmap_grant_ref *kunmap_ops,
 			      struct page **pages, unsigned int count)
 {
 	int i;

arch/x86/include/asm/xen/page.h

@@ -55,9 +55,8 @@ extern int set_foreign_p2m_mapping(struct gnttab_map_grant_ref *map_ops,
 				   struct gnttab_map_grant_ref *kmap_ops,
 				   struct page **pages, unsigned int count);
 extern int clear_foreign_p2m_mapping(struct gnttab_unmap_grant_ref *unmap_ops,
-				     struct gnttab_map_grant_ref *kmap_ops,
+				     struct gnttab_unmap_grant_ref *kunmap_ops,
 				     struct page **pages, unsigned int count);
-extern unsigned long m2p_find_override_pfn(unsigned long mfn, unsigned long pfn);
 
 /*
  * Helper functions to write or read unsigned long values to/from
@@ -154,21 +153,12 @@ static inline unsigned long mfn_to_pfn(unsigned long mfn)
 		return mfn;
 
 	pfn = mfn_to_pfn_no_overrides(mfn);
-	if (__pfn_to_mfn(pfn) != mfn) {
-		/*
-		 * If this appears to be a foreign mfn (because the pfn
-		 * doesn't map back to the mfn), then check the local override
-		 * table to see if there's a better pfn to use.
-		 *
-		 * m2p_find_override_pfn returns ~0 if it doesn't find anything.
-		 */
-		pfn = m2p_find_override_pfn(mfn, ~0);
-	}
+	if (__pfn_to_mfn(pfn) != mfn)
+		pfn = ~0;
 
 	/*
-	 * pfn is ~0 if there are no entries in the m2p for mfn or if the
-	 * entry doesn't map back to the mfn and m2p_override doesn't have a
-	 * valid entry for it.
+	 * pfn is ~0 if there are no entries in the m2p for mfn or the
+	 * entry doesn't map back to the mfn.
 	 */
 	if (pfn == ~0 && __pfn_to_mfn(mfn) == IDENTITY_FRAME(mfn))
 		pfn = mfn;

arch/x86/xen/mmu.c

@@ -1489,7 +1489,7 @@ static void __init xen_set_pte_init(pte_t *ptep, pte_t pte)
 	native_set_pte(ptep, pte);
 }
 
-static void pin_pagetable_pfn(unsigned cmd, unsigned long pfn)
+static void __init pin_pagetable_pfn(unsigned cmd, unsigned long pfn)
 {
 	struct mmuext_op op;
 	op.cmd = cmd;
@@ -1657,7 +1657,7 @@ void __init xen_reserve_top(void)
  * Like __va(), but returns address in the kernel mapping (which is
  * all we have until the physical memory mapping has been set up.
  */
-static void *__ka(phys_addr_t paddr)
+static void * __init __ka(phys_addr_t paddr)
 {
 #ifdef CONFIG_X86_64
 	return (void *)(paddr + __START_KERNEL_map);
@@ -1667,7 +1667,7 @@ static void *__ka(phys_addr_t paddr)
 }
 
 /* Convert a machine address to physical address */
-static unsigned long m2p(phys_addr_t maddr)
+static unsigned long __init m2p(phys_addr_t maddr)
 {
 	phys_addr_t paddr;
 
@@ -1678,13 +1678,14 @@ static unsigned long m2p(phys_addr_t maddr)
 }
 
 /* Convert a machine address to kernel virtual */
-static void *m2v(phys_addr_t maddr)
+static void * __init m2v(phys_addr_t maddr)
 {
 	return __ka(m2p(maddr));
 }
 
 /* Set the page permissions on an identity-mapped pages */
-static void set_page_prot_flags(void *addr, pgprot_t prot, unsigned long flags)
+static void __init set_page_prot_flags(void *addr, pgprot_t prot,
+				       unsigned long flags)
 {
 	unsigned long pfn = __pa(addr) >> PAGE_SHIFT;
 	pte_t pte = pfn_pte(pfn, prot);
@@ -1696,7 +1697,7 @@ static void set_page_prot_flags(void *addr, pgprot_t prot, unsigned long flags)
 	if (HYPERVISOR_update_va_mapping((unsigned long)addr, pte, flags))
 		BUG();
 }
-static void set_page_prot(void *addr, pgprot_t prot)
+static void __init set_page_prot(void *addr, pgprot_t prot)
 {
 	return set_page_prot_flags(addr, prot, UVMF_NONE);
 }
@@ -1733,10 +1734,8 @@ static void __init xen_map_identity_early(pmd_t *pmd, unsigned long max_pfn)
 		for (pteidx = 0; pteidx < PTRS_PER_PTE; pteidx++, pfn++) {
 			pte_t pte;
 
-#ifdef CONFIG_X86_32
 			if (pfn > max_pfn_mapped)
 				max_pfn_mapped = pfn;
-#endif
 
 			if (!pte_none(pte_page[pteidx]))
 				continue;
@@ -1769,7 +1768,7 @@ void __init xen_setup_machphys_mapping(void)
 }
 
 #ifdef CONFIG_X86_64
-static void convert_pfn_mfn(void *v)
+static void __init convert_pfn_mfn(void *v)
 {
 	pte_t *pte = v;
 	int i;

arch/x86/xen/p2m.c

@@ -84,8 +84,6 @@
 
 #define PMDS_PER_MID_PAGE	(P2M_MID_PER_PAGE / PTRS_PER_PTE)
 
-static void __init m2p_override_init(void);
-
 unsigned long *xen_p2m_addr __read_mostly;
 EXPORT_SYMBOL_GPL(xen_p2m_addr);
 unsigned long xen_p2m_size __read_mostly;
@@ -402,8 +400,6 @@ void __init xen_vmalloc_p2m_tree(void)
 	xen_p2m_size = xen_max_p2m_pfn;
 
 	xen_inv_extra_mem();
-
-	m2p_override_init();
 }
 
 unsigned long get_phys_to_machine(unsigned long pfn)
@@ -652,100 +648,21 @@ bool set_phys_to_machine(unsigned long pfn, unsigned long mfn)
 	return true;
 }
 
-#define M2P_OVERRIDE_HASH_SHIFT	10
-#define M2P_OVERRIDE_HASH	(1 << M2P_OVERRIDE_HASH_SHIFT)
-
-static struct list_head *m2p_overrides;
-static DEFINE_SPINLOCK(m2p_override_lock);
-
-static void __init m2p_override_init(void)
-{
-	unsigned i;
-
-	m2p_overrides = alloc_bootmem_align(
-				sizeof(*m2p_overrides) * M2P_OVERRIDE_HASH,
-				sizeof(unsigned long));
-
-	for (i = 0; i < M2P_OVERRIDE_HASH; i++)
-		INIT_LIST_HEAD(&m2p_overrides[i]);
-}
-
-static unsigned long mfn_hash(unsigned long mfn)
-{
-	return hash_long(mfn, M2P_OVERRIDE_HASH_SHIFT);
-}
-
-/* Add an MFN override for a particular page */
-static int m2p_add_override(unsigned long mfn, struct page *page,
-			    struct gnttab_map_grant_ref *kmap_op)
-{
-	unsigned long flags;
-	unsigned long pfn;
-	unsigned long uninitialized_var(address);
-	unsigned level;
-	pte_t *ptep = NULL;
-
-	pfn = page_to_pfn(page);
-	if (!PageHighMem(page)) {
-		address = (unsigned long)__va(pfn << PAGE_SHIFT);
-		ptep = lookup_address(address, &level);
-		if (WARN(ptep == NULL || level != PG_LEVEL_4K,
-			 "m2p_add_override: pfn %lx not mapped", pfn))
-			return -EINVAL;
-	}
-
-	if (kmap_op != NULL) {
-		if (!PageHighMem(page)) {
-			struct multicall_space mcs =
-				xen_mc_entry(sizeof(*kmap_op));
-
-			MULTI_grant_table_op(mcs.mc,
-					GNTTABOP_map_grant_ref, kmap_op, 1);
-
-			xen_mc_issue(PARAVIRT_LAZY_MMU);
-		}
-	}
-	spin_lock_irqsave(&m2p_override_lock, flags);
-	list_add(&page->lru,  &m2p_overrides[mfn_hash(mfn)]);
-	spin_unlock_irqrestore(&m2p_override_lock, flags);
-
-	/* p2m(m2p(mfn)) == mfn: the mfn is already present somewhere in
-	 * this domain. Set the FOREIGN_FRAME_BIT in the p2m for the other
-	 * pfn so that the following mfn_to_pfn(mfn) calls will return the
-	 * pfn from the m2p_override (the backend pfn) instead.
-	 * We need to do this because the pages shared by the frontend
-	 * (xen-blkfront) can be already locked (lock_page, called by
-	 * do_read_cache_page); when the userspace backend tries to use them
-	 * with direct_IO, mfn_to_pfn returns the pfn of the frontend, so
-	 * do_blockdev_direct_IO is going to try to lock the same pages
-	 * again resulting in a deadlock.
-	 * As a side effect get_user_pages_fast might not be safe on the
-	 * frontend pages while they are being shared with the backend,
-	 * because mfn_to_pfn (that ends up being called by GUPF) will
-	 * return the backend pfn rather than the frontend pfn. */
-	pfn = mfn_to_pfn_no_overrides(mfn);
-	if (__pfn_to_mfn(pfn) == mfn)
-		set_phys_to_machine(pfn, FOREIGN_FRAME(mfn));
-
-	return 0;
-}
-
 int set_foreign_p2m_mapping(struct gnttab_map_grant_ref *map_ops,
 			    struct gnttab_map_grant_ref *kmap_ops,
 			    struct page **pages, unsigned int count)
 {
 	int i, ret = 0;
-	bool lazy = false;
 	pte_t *pte;
 
 	if (xen_feature(XENFEAT_auto_translated_physmap))
 		return 0;
 
-	if (kmap_ops &&
-	    !in_interrupt() &&
-	    paravirt_get_lazy_mode() == PARAVIRT_LAZY_NONE) {
-		arch_enter_lazy_mmu_mode();
-		lazy = true;
+	if (kmap_ops) {
+		ret = HYPERVISOR_grant_table_op(GNTTABOP_map_grant_ref,
+						kmap_ops, count);
+		if (ret)
+			goto out;
 	}
 
 	for (i = 0; i < count; i++) {
@@ -764,170 +681,28 @@ int set_foreign_p2m_mapping(struct gnttab_map_grant_ref *map_ops,
 		}
 		pfn = page_to_pfn(pages[i]);
 
-		WARN_ON(PagePrivate(pages[i]));
-		SetPagePrivate(pages[i]);
-		set_page_private(pages[i], mfn);
-		pages[i]->index = pfn_to_mfn(pfn);
+		WARN(pfn_to_mfn(pfn) != INVALID_P2M_ENTRY, "page must be ballooned");
 
 		if (unlikely(!set_phys_to_machine(pfn, FOREIGN_FRAME(mfn)))) {
 			ret = -ENOMEM;
 			goto out;
 		}
-
-		if (kmap_ops) {
-			ret = m2p_add_override(mfn, pages[i], &kmap_ops[i]);
-			if (ret)
-				goto out;
-		}
 	}
 
 out:
-	if (lazy)
-		arch_leave_lazy_mmu_mode();
-
 	return ret;
 }
 EXPORT_SYMBOL_GPL(set_foreign_p2m_mapping);
 
-static struct page *m2p_find_override(unsigned long mfn)
-{
-	unsigned long flags;
-	struct list_head *bucket;
-	struct page *p, *ret;
-
-	if (unlikely(!m2p_overrides))
-		return NULL;
-
-	ret = NULL;
-	bucket = &m2p_overrides[mfn_hash(mfn)];
-
-	spin_lock_irqsave(&m2p_override_lock, flags);
-
-	list_for_each_entry(p, bucket, lru) {
-		if (page_private(p) == mfn) {
-			ret = p;
-			break;
-		}
-	}
-
-	spin_unlock_irqrestore(&m2p_override_lock, flags);
-
-	return ret;
-}
-
-static int m2p_remove_override(struct page *page,
-			       struct gnttab_map_grant_ref *kmap_op,
-			       unsigned long mfn)
-{
-	unsigned long flags;
-	unsigned long pfn;
-	unsigned long uninitialized_var(address);
-	unsigned level;
-	pte_t *ptep = NULL;
-
-	pfn = page_to_pfn(page);
-
-	if (!PageHighMem(page)) {
-		address = (unsigned long)__va(pfn << PAGE_SHIFT);
-		ptep = lookup_address(address, &level);
-
-		if (WARN(ptep == NULL || level != PG_LEVEL_4K,
-			 "m2p_remove_override: pfn %lx not mapped", pfn))
-			return -EINVAL;
-	}
-
-	spin_lock_irqsave(&m2p_override_lock, flags);
-	list_del(&page->lru);
-	spin_unlock_irqrestore(&m2p_override_lock, flags);
-
-	if (kmap_op != NULL) {
-		if (!PageHighMem(page)) {
-			struct multicall_space mcs;
-			struct gnttab_unmap_and_replace *unmap_op;
-			struct page *scratch_page = get_balloon_scratch_page();
-			unsigned long scratch_page_address = (unsigned long)
-				__va(page_to_pfn(scratch_page) << PAGE_SHIFT);
-
-			/*
-			 * It might be that we queued all the m2p grant table
-			 * hypercalls in a multicall, then m2p_remove_override
-			 * get called before the multicall has actually been
-			 * issued. In this case handle is going to -1 because
-			 * it hasn't been modified yet.
-			 */
-			if (kmap_op->handle == -1)
-				xen_mc_flush();
-			/*
-			 * Now if kmap_op->handle is negative it means that the
-			 * hypercall actually returned an error.
-			 */
-			if (kmap_op->handle == GNTST_general_error) {
-				pr_warn("m2p_remove_override: pfn %lx mfn %lx, failed to modify kernel mappings",
-					pfn, mfn);
-				put_balloon_scratch_page();
-				return -1;
-			}
-
-			xen_mc_batch();
-
-			mcs = __xen_mc_entry(
-				sizeof(struct gnttab_unmap_and_replace));
-			unmap_op = mcs.args;
-			unmap_op->host_addr = kmap_op->host_addr;
-			unmap_op->new_addr = scratch_page_address;
-			unmap_op->handle = kmap_op->handle;
-
-			MULTI_grant_table_op(mcs.mc,
-				GNTTABOP_unmap_and_replace, unmap_op, 1);
-
-			mcs = __xen_mc_entry(0);
-			MULTI_update_va_mapping(mcs.mc, scratch_page_address,
-					pfn_pte(page_to_pfn(scratch_page),
-					PAGE_KERNEL_RO), 0);
-
-			xen_mc_issue(PARAVIRT_LAZY_MMU);
-
-			kmap_op->host_addr = 0;
-			put_balloon_scratch_page();
-		}
-	}
-
-	/* p2m(m2p(mfn)) == FOREIGN_FRAME(mfn): the mfn is already present
-	 * somewhere in this domain, even before being added to the
-	 * m2p_override (see comment above in m2p_add_override).
-	 * If there are no other entries in the m2p_override corresponding
-	 * to this mfn, then remove the FOREIGN_FRAME_BIT from the p2m for
-	 * the original pfn (the one shared by the frontend): the backend
-	 * cannot do any IO on this page anymore because it has been
-	 * unshared. Removing the FOREIGN_FRAME_BIT from the p2m entry of
-	 * the original pfn causes mfn_to_pfn(mfn) to return the frontend
-	 * pfn again. */
-	mfn &= ~FOREIGN_FRAME_BIT;
-	pfn = mfn_to_pfn_no_overrides(mfn);
-	if (__pfn_to_mfn(pfn) == FOREIGN_FRAME(mfn) &&
-			m2p_find_override(mfn) == NULL)
-		set_phys_to_machine(pfn, mfn);
-
-	return 0;
-}
-
 int clear_foreign_p2m_mapping(struct gnttab_unmap_grant_ref *unmap_ops,
-			      struct gnttab_map_grant_ref *kmap_ops,
+			      struct gnttab_unmap_grant_ref *kunmap_ops,
 			      struct page **pages, unsigned int count)
 {
 	int i, ret = 0;
-	bool lazy = false;
 
 	if (xen_feature(XENFEAT_auto_translated_physmap))
 		return 0;
 
-	if (kmap_ops &&
-	    !in_interrupt() &&
-	    paravirt_get_lazy_mode() == PARAVIRT_LAZY_NONE) {
-		arch_enter_lazy_mmu_mode();
-		lazy = true;
-	}
-
 	for (i = 0; i < count; i++) {
 		unsigned long mfn = __pfn_to_mfn(page_to_pfn(pages[i]));
 		unsigned long pfn = page_to_pfn(pages[i]);
@@ -937,36 +712,16 @@ int clear_foreign_p2m_mapping(struct gnttab_unmap_grant_ref *unmap_ops,
 			goto out;
 		}
 
-		set_page_private(pages[i], INVALID_P2M_ENTRY);
-		WARN_ON(!PagePrivate(pages[i]));
-		ClearPagePrivate(pages[i]);
-		set_phys_to_machine(pfn, pages[i]->index);
-
-		if (kmap_ops)
-			ret = m2p_remove_override(pages[i], &kmap_ops[i], mfn);
-		if (ret)
-			goto out;
+		set_phys_to_machine(pfn, INVALID_P2M_ENTRY);
 	}
-
+	if (kunmap_ops)
+		ret = HYPERVISOR_grant_table_op(GNTTABOP_unmap_grant_ref,
+						kunmap_ops, count);
 out:
-	if (lazy)
-		arch_leave_lazy_mmu_mode();
 	return ret;
 }
 EXPORT_SYMBOL_GPL(clear_foreign_p2m_mapping);
 
-unsigned long m2p_find_override_pfn(unsigned long mfn, unsigned long pfn)
-{
-	struct page *p = m2p_find_override(mfn);
-	unsigned long ret = pfn;
-
-	if (p)
-		ret = page_to_pfn(p);
-
-	return ret;
-}
-EXPORT_SYMBOL_GPL(m2p_find_override_pfn);
-
 #ifdef CONFIG_XEN_DEBUG_FS
 #include <linux/debugfs.h>
 #include "debugfs.h"

arch/x86/xen/setup.c

@@ -32,16 +32,6 @@
 #include "p2m.h"
 #include "mmu.h"
 
-/* These are code, but not functions.  Defined in entry.S */
-extern const char xen_hypervisor_callback[];
-extern const char xen_failsafe_callback[];
-#ifdef CONFIG_X86_64
-extern asmlinkage void nmi(void);
-#endif
-extern void xen_sysenter_target(void);
-extern void xen_syscall_target(void);
-extern void xen_syscall32_target(void);
-
 /* Amount of extra memory space we add to the e820 ranges */
 struct xen_memory_region xen_extra_mem[XEN_EXTRA_MEM_MAX_REGIONS] __initdata;
 
@@ -74,7 +64,7 @@ static unsigned long xen_remap_mfn __initdata = INVALID_P2M_ENTRY;
  */
 #define EXTRA_MEM_RATIO		(10)
 
-static void __init xen_add_extra_mem(u64 start, u64 size)
+static void __init xen_add_extra_mem(phys_addr_t start, phys_addr_t size)
 {
 	int i;
 
@@ -97,10 +87,10 @@ static void __init xen_add_extra_mem(u64 start, u64 size)
 	memblock_reserve(start, size);
 }
 
-static void __init xen_del_extra_mem(u64 start, u64 size)
+static void __init xen_del_extra_mem(phys_addr_t start, phys_addr_t size)
 {
 	int i;
-	u64 start_r, size_r;
+	phys_addr_t start_r, size_r;
 
 	for (i = 0; i < XEN_EXTRA_MEM_MAX_REGIONS; i++) {
 		start_r = xen_extra_mem[i].start;
@@ -267,7 +257,7 @@ static void __init xen_set_identity_and_release_chunk(unsigned long start_pfn,
 static void __init xen_update_mem_tables(unsigned long pfn, unsigned long mfn)
 {
 	struct mmu_update update = {
-		.ptr = ((unsigned long long)mfn << PAGE_SHIFT) | MMU_MACHPHYS_UPDATE,
+		.ptr = ((uint64_t)mfn << PAGE_SHIFT) | MMU_MACHPHYS_UPDATE,
 		.val = pfn
 	};
 
@@ -545,20 +535,21 @@ static unsigned long __init xen_get_max_pages(void)
 	return min(max_pages, MAX_DOMAIN_PAGES);
 }
 
-static void xen_align_and_add_e820_region(u64 start, u64 size, int type)
+static void __init xen_align_and_add_e820_region(phys_addr_t start,
+						 phys_addr_t size, int type)
 {
-	u64 end = start + size;
+	phys_addr_t end = start + size;
 
 	/* Align RAM regions to page boundaries. */
 	if (type == E820_RAM) {
 		start = PAGE_ALIGN(start);
-		end &= ~((u64)PAGE_SIZE - 1);
+		end &= ~((phys_addr_t)PAGE_SIZE - 1);
 	}
 
 	e820_add_region(start, end - start, type);
 }
 
-void xen_ignore_unusable(struct e820entry *list, size_t map_size)
+static void __init xen_ignore_unusable(struct e820entry *list, size_t map_size)
 {
 	struct e820entry *entry;
 	unsigned int i;
@@ -577,7 +568,7 @@ char * __init xen_memory_setup(void)
 	static struct e820entry map[E820MAX] __initdata;
 
 	unsigned long max_pfn = xen_start_info->nr_pages;
-	unsigned long long mem_end;
+	phys_addr_t mem_end;
 	int rc;
 	struct xen_memory_map memmap;
 	unsigned long max_pages;
@@ -652,16 +643,16 @@ char * __init xen_memory_setup(void)
 			  extra_pages);
 	i = 0;
 	while (i < memmap.nr_entries) {
-		u64 addr = map[i].addr;
-		u64 size = map[i].size;
+		phys_addr_t addr = map[i].addr;
+		phys_addr_t size = map[i].size;
 		u32 type = map[i].type;
 
 		if (type == E820_RAM) {
 			if (addr < mem_end) {
 				size = min(size, mem_end - addr);
 			} else if (extra_pages) {
-				size = min(size, (u64)extra_pages * PAGE_SIZE);
-				extra_pages -= size / PAGE_SIZE;
+				size = min(size, PFN_PHYS(extra_pages));
+				extra_pages -= PFN_DOWN(size);
 				xen_add_extra_mem(addr, size);
 				xen_max_p2m_pfn = PFN_DOWN(addr + size);
 			} else

arch/x86/xen/smp.c

@@ -507,7 +507,7 @@ static int xen_cpu_disable(void)
 static void xen_cpu_die(unsigned int cpu)
 {
 	while (xen_pv_domain() && HYPERVISOR_vcpu_op(VCPUOP_is_up, cpu, NULL)) {
-		current->state = TASK_UNINTERRUPTIBLE;
+		__set_current_state(TASK_UNINTERRUPTIBLE);
 		schedule_timeout(HZ/10);
 	}
 

arch/x86/xen/time.c

@@ -479,6 +479,10 @@ static void __init xen_time_init(void)
 	int cpu = smp_processor_id();
 	struct timespec tp;
 
+	/* As Dom0 is never moved, no penalty on using TSC there */
+	if (xen_initial_domain())
+		xen_clocksource.rating = 275;
+
 	clocksource_register_hz(&xen_clocksource, NSEC_PER_SEC);
 
 	if (HYPERVISOR_vcpu_op(VCPUOP_stop_periodic_timer, cpu, NULL) == 0) {

arch/x86/xen/xen-ops.h

@@ -10,6 +10,12 @@
 extern const char xen_hypervisor_callback[];
 extern const char xen_failsafe_callback[];
 
+void xen_sysenter_target(void);
+#ifdef CONFIG_X86_64
+void xen_syscall_target(void);
+void xen_syscall32_target(void);
+#endif
+
 extern void *xen_initial_gdt;
 
 struct trap_info;

drivers/block/xen-blkback/blkback.c

@@ -47,6 +47,7 @@
 #include <asm/xen/hypervisor.h>
 #include <asm/xen/hypercall.h>
 #include <xen/balloon.h>
+#include <xen/grant_table.h>
 #include "common.h"
 
 /*
@@ -100,7 +101,7 @@ module_param(log_stats, int, 0644);
 
 #define BLKBACK_INVALID_HANDLE (~0)
 
-/* Number of free pages to remove on each call to free_xenballooned_pages */
+/* Number of free pages to remove on each call to gnttab_free_pages */
 #define NUM_BATCH_FREE_PAGES 10
 
 static inline int get_free_page(struct xen_blkif *blkif, struct page **page)
@@ -111,7 +112,7 @@ static inline int get_free_page(struct xen_blkif *blkif, struct page **page)
 	if (list_empty(&blkif->free_pages)) {
 		BUG_ON(blkif->free_pages_num != 0);
 		spin_unlock_irqrestore(&blkif->free_pages_lock, flags);
-		return alloc_xenballooned_pages(1, page, false);
+		return gnttab_alloc_pages(1, page);
 	}
 	BUG_ON(blkif->free_pages_num == 0);
 	page[0] = list_first_entry(&blkif->free_pages, struct page, lru);
@@ -151,14 +152,14 @@ static inline void shrink_free_pagepool(struct xen_blkif *blkif, int num)
 		blkif->free_pages_num--;
 		if (++num_pages == NUM_BATCH_FREE_PAGES) {
 			spin_unlock_irqrestore(&blkif->free_pages_lock, flags);
-			free_xenballooned_pages(num_pages, page);
+			gnttab_free_pages(num_pages, page);
 			spin_lock_irqsave(&blkif->free_pages_lock, flags);
 			num_pages = 0;
 		}
 	}
 	spin_unlock_irqrestore(&blkif->free_pages_lock, flags);
 	if (num_pages != 0)
-		free_xenballooned_pages(num_pages, page);
+		gnttab_free_pages(num_pages, page);
 }
 
 #define vaddr(page) ((unsigned long)pfn_to_kaddr(page_to_pfn(page)))
@@ -262,6 +263,17 @@ static void put_persistent_gnt(struct xen_blkif *blkif,
 	atomic_dec(&blkif->persistent_gnt_in_use);
 }
 
+static void free_persistent_gnts_unmap_callback(int result,
+						struct gntab_unmap_queue_data *data)
+{
+	struct completion *c = data->data;
+
+	/* BUG_ON used to reproduce existing behaviour,
+	   but is this the best way to deal with this? */
+	BUG_ON(result);
+	complete(c);
+}
+
 static void free_persistent_gnts(struct xen_blkif *blkif, struct rb_root *root,
                                  unsigned int num)
 {
@@ -269,8 +281,17 @@ static void free_persistent_gnts(struct xen_blkif *blkif, struct rb_root *root,
 	struct page *pages[BLKIF_MAX_SEGMENTS_PER_REQUEST];
 	struct persistent_gnt *persistent_gnt;
 	struct rb_node *n;
-	int ret = 0;
 	int segs_to_unmap = 0;
+	struct gntab_unmap_queue_data unmap_data;
+	struct completion unmap_completion;
+
+	init_completion(&unmap_completion);
+
+	unmap_data.data = &unmap_completion;
+	unmap_data.done = &free_persistent_gnts_unmap_callback;
+	unmap_data.pages = pages;
+	unmap_data.unmap_ops = unmap;
+	unmap_data.kunmap_ops = NULL;
 
 	foreach_grant_safe(persistent_gnt, n, root, node) {
 		BUG_ON(persistent_gnt->handle ==
@@ -285,9 +306,11 @@ static void free_persistent_gnts(struct xen_blkif *blkif, struct rb_root *root,
 
 		if (++segs_to_unmap == BLKIF_MAX_SEGMENTS_PER_REQUEST ||
 			!rb_next(&persistent_gnt->node)) {
-			ret = gnttab_unmap_refs(unmap, NULL, pages,
-				segs_to_unmap);
-			BUG_ON(ret);
+
+			unmap_data.count = segs_to_unmap;
+			gnttab_unmap_refs_async(&unmap_data);
+			wait_for_completion(&unmap_completion);
+
 			put_free_pages(blkif, pages, segs_to_unmap);
 			segs_to_unmap = 0;
 		}
@@ -653,18 +676,14 @@ void xen_blkbk_free_caches(struct xen_blkif *blkif)
 	shrink_free_pagepool(blkif, 0 /* All */);
 }
 
-/*
- * Unmap the grant references, and also remove the M2P over-rides
- * used in the 'pending_req'.
- */
-static void xen_blkbk_unmap(struct xen_blkif *blkif,
-                            struct grant_page *pages[],
-                            int num)
+static unsigned int xen_blkbk_unmap_prepare(
+	struct xen_blkif *blkif,
+	struct grant_page **pages,
+	unsigned int num,
+	struct gnttab_unmap_grant_ref *unmap_ops,
+	struct page **unmap_pages)
 {
-	struct gnttab_unmap_grant_ref unmap[BLKIF_MAX_SEGMENTS_PER_REQUEST];
-	struct page *unmap_pages[BLKIF_MAX_SEGMENTS_PER_REQUEST];
 	unsigned int i, invcount = 0;
-	int ret;
 
 	for (i = 0; i < num; i++) {
 		if (pages[i]->persistent_gnt != NULL) {
@@ -674,22 +693,96 @@ static void xen_blkbk_unmap(struct xen_blkif *blkif,
 		if (pages[i]->handle == BLKBACK_INVALID_HANDLE)
 			continue;
 		unmap_pages[invcount] = pages[i]->page;
-		gnttab_set_unmap_op(&unmap[invcount], vaddr(pages[i]->page),
+		gnttab_set_unmap_op(&unmap_ops[invcount], vaddr(pages[i]->page),
 				    GNTMAP_host_map, pages[i]->handle);
 		pages[i]->handle = BLKBACK_INVALID_HANDLE;
-		if (++invcount == BLKIF_MAX_SEGMENTS_PER_REQUEST) {
-			ret = gnttab_unmap_refs(unmap, NULL, unmap_pages,
-			                        invcount);
-			BUG_ON(ret);
-			put_free_pages(blkif, unmap_pages, invcount);
-			invcount = 0;
+		invcount++;
        }
+
+       return invcount;
+}
+
+static void xen_blkbk_unmap_and_respond_callback(int result, struct gntab_unmap_queue_data *data)
+{
+	struct pending_req* pending_req = (struct pending_req*) (data->data);
+	struct xen_blkif *blkif = pending_req->blkif;
+
+	/* BUG_ON used to reproduce existing behaviour,
+	   but is this the best way to deal with this? */
+	BUG_ON(result);
+
+	put_free_pages(blkif, data->pages, data->count);
+	make_response(blkif, pending_req->id,
+		      pending_req->operation, pending_req->status);
+	free_req(blkif, pending_req);
+	/*
+	 * Make sure the request is freed before releasing blkif,
+	 * or there could be a race between free_req and the
+	 * cleanup done in xen_blkif_free during shutdown.
+	 *
+	 * NB: The fact that we might try to wake up pending_free_wq
+	 * before drain_complete (in case there's a drain going on)
+	 * it's not a problem with our current implementation
+	 * because we can assure there's no thread waiting on
+	 * pending_free_wq if there's a drain going on, but it has
+	 * to be taken into account if the current model is changed.
+	 */
+	if (atomic_dec_and_test(&blkif->inflight) && atomic_read(&blkif->drain)) {
+		complete(&blkif->drain_complete);
 	}
+	xen_blkif_put(blkif);
+}
+
+static void xen_blkbk_unmap_and_respond(struct pending_req *req)
+{
+	struct gntab_unmap_queue_data* work = &req->gnttab_unmap_data;
+	struct xen_blkif *blkif = req->blkif;
+	struct grant_page **pages = req->segments;
+	unsigned int invcount;
+
+	invcount = xen_blkbk_unmap_prepare(blkif, pages, req->nr_pages,
+					   req->unmap, req->unmap_pages);
+
+	work->data = req;
+	work->done = xen_blkbk_unmap_and_respond_callback;
+	work->unmap_ops = req->unmap;
+	work->kunmap_ops = NULL;
+	work->pages = req->unmap_pages;
+	work->count = invcount;
+
+	gnttab_unmap_refs_async(&req->gnttab_unmap_data);
+}
+
+
+/*
+ * Unmap the grant references.
+ *
+ * This could accumulate ops up to the batch size to reduce the number
+ * of hypercalls, but since this is only used in error paths there's
+ * no real need.
+ */
+static void xen_blkbk_unmap(struct xen_blkif *blkif,
+                            struct grant_page *pages[],
+                            int num)
+{
+	struct gnttab_unmap_grant_ref unmap[BLKIF_MAX_SEGMENTS_PER_REQUEST];
+	struct page *unmap_pages[BLKIF_MAX_SEGMENTS_PER_REQUEST];
+	unsigned int invcount = 0;
+	int ret;
+
+	while (num) {
+		unsigned int batch = min(num, BLKIF_MAX_SEGMENTS_PER_REQUEST);
+		
+		invcount = xen_blkbk_unmap_prepare(blkif, pages, batch,
+						   unmap, unmap_pages);
 		if (invcount) {
 			ret = gnttab_unmap_refs(unmap, NULL, unmap_pages, invcount);
 			BUG_ON(ret);
 			put_free_pages(blkif, unmap_pages, invcount);
 		}
+		pages += batch;
+		num -= batch;
+	}
 }
 
 static int xen_blkbk_map(struct xen_blkif *blkif,
@@ -982,32 +1075,8 @@ static void __end_block_io_op(struct pending_req *pending_req, int error)
 	 * the grant references associated with 'request' and provide
 	 * the proper response on the ring.
 	 */
-	if (atomic_dec_and_test(&pending_req->pendcnt)) {
-		struct xen_blkif *blkif = pending_req->blkif;
-
-		xen_blkbk_unmap(blkif,
-		                pending_req->segments,
-		                pending_req->nr_pages);
-		make_response(blkif, pending_req->id,
-			      pending_req->operation, pending_req->status);
-		free_req(blkif, pending_req);
-		/*
-		 * Make sure the request is freed before releasing blkif,
-		 * or there could be a race between free_req and the
-		 * cleanup done in xen_blkif_free during shutdown.
-		 *
-		 * NB: The fact that we might try to wake up pending_free_wq
-		 * before drain_complete (in case there's a drain going on)
-		 * it's not a problem with our current implementation
-		 * because we can assure there's no thread waiting on
-		 * pending_free_wq if there's a drain going on, but it has
-		 * to be taken into account if the current model is changed.
-		 */
-		if (atomic_dec_and_test(&blkif->inflight) && atomic_read(&blkif->drain)) {
-			complete(&blkif->drain_complete);
-		}
-		xen_blkif_put(blkif);
-	}
+	if (atomic_dec_and_test(&pending_req->pendcnt))
+		xen_blkbk_unmap_and_respond(pending_req);
 }
 
 /*

drivers/block/xen-blkback/common.h

@@ -350,6 +350,9 @@ struct pending_req {
 	struct grant_page	*indirect_pages[MAX_INDIRECT_PAGES];
 	struct seg_buf		seg[MAX_INDIRECT_SEGMENTS];
 	struct bio		*biolist[MAX_INDIRECT_SEGMENTS];
+	struct gnttab_unmap_grant_ref unmap[MAX_INDIRECT_SEGMENTS];
+	struct page                   *unmap_pages[MAX_INDIRECT_SEGMENTS];
+	struct gntab_unmap_queue_data gnttab_unmap_data;
 };
 
 

drivers/net/xen-netback/interface.c

@@ -483,9 +483,8 @@ int xenvif_init_queue(struct xenvif_queue *queue)
 	 * better enable it. The long term solution would be to use just a
 	 * bunch of valid page descriptors, without dependency on ballooning
 	 */
-	err = alloc_xenballooned_pages(MAX_PENDING_REQS,
-				       queue->mmap_pages,
-				       false);
+	err = gnttab_alloc_pages(MAX_PENDING_REQS,
+				 queue->mmap_pages);
 	if (err) {
 		netdev_err(queue->vif->dev, "Could not reserve mmap_pages\n");
 		return -ENOMEM;
@@ -664,7 +663,7 @@ void xenvif_disconnect(struct xenvif *vif)
  */
 void xenvif_deinit_queue(struct xenvif_queue *queue)
 {
-	free_xenballooned_pages(MAX_PENDING_REQS, queue->mmap_pages);
+	gnttab_free_pages(MAX_PENDING_REQS, queue->mmap_pages);
 }
 
 void xenvif_free(struct xenvif *vif)

drivers/net/xen-netback/netback.c

@@ -314,9 +314,7 @@ static struct xenvif_rx_meta *get_next_rx_buffer(struct xenvif_queue *queue,
 static void xenvif_gop_frag_copy(struct xenvif_queue *queue, struct sk_buff *skb,
 				 struct netrx_pending_operations *npo,
 				 struct page *page, unsigned long size,
-				 unsigned long offset, int *head,
-				 struct xenvif_queue *foreign_queue,
-				 grant_ref_t foreign_gref)
+				 unsigned long offset, int *head)
 {
 	struct gnttab_copy *copy_gop;
 	struct xenvif_rx_meta *meta;
@@ -333,6 +331,8 @@ static void xenvif_gop_frag_copy(struct xenvif_queue *queue, struct sk_buff *skb
 	offset &= ~PAGE_MASK;
 
 	while (size > 0) {
+		struct xen_page_foreign *foreign;
+
 		BUG_ON(offset >= PAGE_SIZE);
 		BUG_ON(npo->copy_off > MAX_BUFFER_OFFSET);
 
@@ -361,9 +361,10 @@ static void xenvif_gop_frag_copy(struct xenvif_queue *queue, struct sk_buff *skb
 		copy_gop->flags = GNTCOPY_dest_gref;
 		copy_gop->len = bytes;
 
-		if (foreign_queue) {
-			copy_gop->source.domid = foreign_queue->vif->domid;
-			copy_gop->source.u.ref = foreign_gref;
+		foreign = xen_page_foreign(page);
+		if (foreign) {
+			copy_gop->source.domid = foreign->domid;
+			copy_gop->source.u.ref = foreign->gref;
 			copy_gop->flags |= GNTCOPY_source_gref;
 		} else {
 			copy_gop->source.domid = DOMID_SELF;
@@ -405,35 +406,6 @@ static void xenvif_gop_frag_copy(struct xenvif_queue *queue, struct sk_buff *skb
 	}
 }
 
-/*
- * Find the grant ref for a given frag in a chain of struct ubuf_info's
- * skb: the skb itself
- * i: the frag's number
- * ubuf: a pointer to an element in the chain. It should not be NULL
- *
- * Returns a pointer to the element in the chain where the page were found. If
- * not found, returns NULL.
- * See the definition of callback_struct in common.h for more details about
- * the chain.
- */
-static const struct ubuf_info *xenvif_find_gref(const struct sk_buff *const skb,
-						const int i,
-						const struct ubuf_info *ubuf)
-{
-	struct xenvif_queue *foreign_queue = ubuf_to_queue(ubuf);
-
-	do {
-		u16 pending_idx = ubuf->desc;
-
-		if (skb_shinfo(skb)->frags[i].page.p ==
-		    foreign_queue->mmap_pages[pending_idx])
-			break;
-		ubuf = (struct ubuf_info *) ubuf->ctx;
-	} while (ubuf);
-
-	return ubuf;
-}
-
 /*
  * Prepare an SKB to be transmitted to the frontend.
  *
@@ -459,8 +431,6 @@ static int xenvif_gop_skb(struct sk_buff *skb,
 	int head = 1;
 	int old_meta_prod;
 	int gso_type;
-	const struct ubuf_info *ubuf = skb_shinfo(skb)->destructor_arg;
-	const struct ubuf_info *const head_ubuf = ubuf;
 
 	old_meta_prod = npo->meta_prod;
 
@@ -507,68 +477,16 @@ static int xenvif_gop_skb(struct sk_buff *skb,
 			len = skb_tail_pointer(skb) - data;
 
 		xenvif_gop_frag_copy(queue, skb, npo,
-				     virt_to_page(data), len, offset, &head,
-				     NULL,
-				     0);
+				     virt_to_page(data), len, offset, &head);
 		data += len;
 	}
 
 	for (i = 0; i < nr_frags; i++) {
-		/* This variable also signals whether foreign_gref has a real
-		 * value or not.
-		 */
-		struct xenvif_queue *foreign_queue = NULL;
-		grant_ref_t foreign_gref;
-
-		if ((skb_shinfo(skb)->tx_flags & SKBTX_DEV_ZEROCOPY) &&
-			(ubuf->callback == &xenvif_zerocopy_callback)) {
-			const struct ubuf_info *const startpoint = ubuf;
-
-			/* Ideally ubuf points to the chain element which
-			 * belongs to this frag. Or if frags were removed from
-			 * the beginning, then shortly before it.
-			 */
-			ubuf = xenvif_find_gref(skb, i, ubuf);
-
-			/* Try again from the beginning of the list, if we
-			 * haven't tried from there. This only makes sense in
-			 * the unlikely event of reordering the original frags.
-			 * For injected local pages it's an unnecessary second
-			 * run.
-			 */
-			if (unlikely(!ubuf) && startpoint != head_ubuf)
-				ubuf = xenvif_find_gref(skb, i, head_ubuf);
-
-			if (likely(ubuf)) {
-				u16 pending_idx = ubuf->desc;
-
-				foreign_queue = ubuf_to_queue(ubuf);
-				foreign_gref =
-					foreign_queue->pending_tx_info[pending_idx].req.gref;
-				/* Just a safety measure. If this was the last
-				 * element on the list, the for loop will
-				 * iterate again if a local page were added to
-				 * the end. Using head_ubuf here prevents the
-				 * second search on the chain. Or the original
-				 * frags changed order, but that's less likely.
-				 * In any way, ubuf shouldn't be NULL.
-				 */
-				ubuf = ubuf->ctx ?
-					(struct ubuf_info *) ubuf->ctx :
-					head_ubuf;
-			} else
-				/* This frag was a local page, added to the
-				 * array after the skb left netback.
-				 */
-				ubuf = head_ubuf;
-		}
 		xenvif_gop_frag_copy(queue, skb, npo,
 				     skb_frag_page(&skb_shinfo(skb)->frags[i]),
 				     skb_frag_size(&skb_shinfo(skb)->frags[i]),
 				     skb_shinfo(skb)->frags[i].page_offset,
-				     &head,
-				     foreign_queue,
-				     foreign_queue ? foreign_gref : UINT_MAX);
+				     &head);
 	}
 
 	return npo->meta_prod - old_meta_prod;
@@ -1241,12 +1159,6 @@ static void xenvif_fill_frags(struct xenvif_queue *queue, struct sk_buff *skb)
 		/* Take an extra reference to offset network stack's put_page */
 		get_page(queue->mmap_pages[pending_idx]);
 	}
-	/* FIXME: __skb_fill_page_desc set this to true because page->pfmemalloc
-	 * overlaps with "index", and "mapping" is not set. I think mapping
-	 * should be set. If delivered to local stack, it would drop this
-	 * skb in sk_filter unless the socket has the right to use it.
-	 */
-	skb->pfmemalloc	= false;
 }
 
 static int xenvif_get_extras(struct xenvif_queue *queue,

drivers/xen/balloon.c

@@ -92,7 +92,6 @@ EXPORT_SYMBOL_GPL(balloon_stats);
 
 /* We increase/decrease in batches which fit in a page */
 static xen_pfn_t frame_list[PAGE_SIZE / sizeof(unsigned long)];
-static DEFINE_PER_CPU(struct page *, balloon_scratch_page);
 
 
 /* List of ballooned pages, threaded through the mem_map array. */
@@ -423,22 +422,12 @@ static enum bp_state decrease_reservation(unsigned long nr_pages, gfp_t gfp)
 		page = pfn_to_page(pfn);
 
 #ifdef CONFIG_XEN_HAVE_PVMMU
-		/*
-		 * Ballooned out frames are effectively replaced with
-		 * a scratch frame.  Ensure direct mappings and the
-		 * p2m are consistent.
-		 */
 		if (!xen_feature(XENFEAT_auto_translated_physmap)) {
 			if (!PageHighMem(page)) {
-				struct page *scratch_page = get_balloon_scratch_page();
-
 				ret = HYPERVISOR_update_va_mapping(
 						(unsigned long)__va(pfn << PAGE_SHIFT),
-						pfn_pte(page_to_pfn(scratch_page),
-							PAGE_KERNEL_RO), 0);
+						__pte_ma(0), 0);
 				BUG_ON(ret);
-
-				put_balloon_scratch_page();
 			}
 			__set_phys_to_machine(pfn, INVALID_P2M_ENTRY);
 		}
@@ -500,18 +489,6 @@ static void balloon_process(struct work_struct *work)
 	mutex_unlock(&balloon_mutex);
 }
 
-struct page *get_balloon_scratch_page(void)
-{
-	struct page *ret = get_cpu_var(balloon_scratch_page);
-	BUG_ON(ret == NULL);
-	return ret;
-}
-
-void put_balloon_scratch_page(void)
-{
-	put_cpu_var(balloon_scratch_page);
-}
-
 /* Resets the Xen limit, sets new target, and kicks off processing. */
 void balloon_set_new_target(unsigned long target)
 {
@@ -605,61 +582,13 @@ static void __init balloon_add_region(unsigned long start_pfn,
 	}
 }
 
-static int alloc_balloon_scratch_page(int cpu)
-{
-	if (per_cpu(balloon_scratch_page, cpu) != NULL)
-		return 0;
-
-	per_cpu(balloon_scratch_page, cpu) = alloc_page(GFP_KERNEL);
-	if (per_cpu(balloon_scratch_page, cpu) == NULL) {
-		pr_warn("Failed to allocate balloon_scratch_page for cpu %d\n", cpu);
-		return -ENOMEM;
-	}
-
-	return 0;
-}
-
-
-static int balloon_cpu_notify(struct notifier_block *self,
-				    unsigned long action, void *hcpu)
-{
-	int cpu = (long)hcpu;
-	switch (action) {
-	case CPU_UP_PREPARE:
-		if (alloc_balloon_scratch_page(cpu))
-			return NOTIFY_BAD;
-		break;
-	default:
-		break;
-	}
-	return NOTIFY_OK;
-}
-
-static struct notifier_block balloon_cpu_notifier = {
-	.notifier_call	= balloon_cpu_notify,
-};
-
 static int __init balloon_init(void)
 {
-	int i, cpu;
+	int i;
 
 	if (!xen_domain())
 		return -ENODEV;
 
-	if (!xen_feature(XENFEAT_auto_translated_physmap)) {
-		register_cpu_notifier(&balloon_cpu_notifier);
-
-		get_online_cpus();
-		for_each_online_cpu(cpu) {
-			if (alloc_balloon_scratch_page(cpu)) {
-				put_online_cpus();
-				unregister_cpu_notifier(&balloon_cpu_notifier);
-				return -ENOMEM;
-			}
-		}
-		put_online_cpus();
-	}
-
 	pr_info("Initialising balloon driver\n");
 
 	balloon_stats.current_pages = xen_pv_domain()
@@ -696,15 +625,4 @@ static int __init balloon_init(void)
 
 subsys_initcall(balloon_init);
 
-static int __init balloon_clear(void)
-{
-	int cpu;
-
-	for_each_possible_cpu(cpu)
-		per_cpu(balloon_scratch_page, cpu) = NULL;
-
-	return 0;
-}
-early_initcall(balloon_clear);
-
 MODULE_LICENSE("GPL");

drivers/xen/grant-table.c

@@ -42,6 +42,7 @@
 #include <linux/io.h>
 #include <linux/delay.h>
 #include <linux/hardirq.h>
+#include <linux/workqueue.h>
 
 #include <xen/xen.h>
 #include <xen/interface/xen.h>
@@ -50,6 +51,7 @@
 #include <xen/interface/memory.h>
 #include <xen/hvc-console.h>
 #include <xen/swiotlb-xen.h>
+#include <xen/balloon.h>
 #include <asm/xen/hypercall.h>
 #include <asm/xen/interface.h>
 
@@ -671,6 +673,59 @@ void gnttab_free_auto_xlat_frames(void)
 }
 EXPORT_SYMBOL_GPL(gnttab_free_auto_xlat_frames);
 
+/**
+ * gnttab_alloc_pages - alloc pages suitable for grant mapping into
+ * @nr_pages: number of pages to alloc
+ * @pages: returns the pages
+ */
+int gnttab_alloc_pages(int nr_pages, struct page **pages)
+{
+	int i;
+	int ret;
+
+	ret = alloc_xenballooned_pages(nr_pages, pages, false);
+	if (ret < 0)
+		return ret;
+
+	for (i = 0; i < nr_pages; i++) {
+#if BITS_PER_LONG < 64
+		struct xen_page_foreign *foreign;
+
+		foreign = kzalloc(sizeof(*foreign), GFP_KERNEL);
+		if (!foreign) {
+			gnttab_free_pages(nr_pages, pages);
+			return -ENOMEM;
+		}
+		set_page_private(pages[i], (unsigned long)foreign);
+#endif
+		SetPagePrivate(pages[i]);
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL(gnttab_alloc_pages);
+
+/**
+ * gnttab_free_pages - free pages allocated by gnttab_alloc_pages()
+ * @nr_pages; number of pages to free
+ * @pages: the pages
+ */
+void gnttab_free_pages(int nr_pages, struct page **pages)
+{
+	int i;
+
+	for (i = 0; i < nr_pages; i++) {
+		if (PagePrivate(pages[i])) {
+#if BITS_PER_LONG < 64
+			kfree((void *)page_private(pages[i]));
+#endif
+			ClearPagePrivate(pages[i]);
+		}
+	}
+	free_xenballooned_pages(nr_pages, pages);
+}
+EXPORT_SYMBOL(gnttab_free_pages);
+
 /* Handling of paged out grant targets (GNTST_eagain) */
 #define MAX_DELAY 256
 static inline void
@@ -727,30 +782,87 @@ int gnttab_map_refs(struct gnttab_map_grant_ref *map_ops,
 	if (ret)
 		return ret;
 
+	for (i = 0; i < count; i++) {
 		/* Retry eagain maps */
-	for (i = 0; i < count; i++)
 		if (map_ops[i].status == GNTST_eagain)
 			gnttab_retry_eagain_gop(GNTTABOP_map_grant_ref, map_ops + i,
 						&map_ops[i].status, __func__);
 
+		if (map_ops[i].status == GNTST_okay) {
+			struct xen_page_foreign *foreign;
+
+			SetPageForeign(pages[i]);
+			foreign = xen_page_foreign(pages[i]);
+			foreign->domid = map_ops[i].dom;
+			foreign->gref = map_ops[i].ref;
+		}
+	}
+
 	return set_foreign_p2m_mapping(map_ops, kmap_ops, pages, count);
 }
 EXPORT_SYMBOL_GPL(gnttab_map_refs);
 
 int gnttab_unmap_refs(struct gnttab_unmap_grant_ref *unmap_ops,
-		      struct gnttab_map_grant_ref *kmap_ops,
+		      struct gnttab_unmap_grant_ref *kunmap_ops,
 		      struct page **pages, unsigned int count)
 {
+	unsigned int i;
 	int ret;
 
 	ret = HYPERVISOR_grant_table_op(GNTTABOP_unmap_grant_ref, unmap_ops, count);
 	if (ret)
 		return ret;
 
-	return clear_foreign_p2m_mapping(unmap_ops, kmap_ops, pages, count);
+	for (i = 0; i < count; i++)
+		ClearPageForeign(pages[i]);
+
+	return clear_foreign_p2m_mapping(unmap_ops, kunmap_ops, pages, count);
 }
 EXPORT_SYMBOL_GPL(gnttab_unmap_refs);
 
+#define GNTTAB_UNMAP_REFS_DELAY 5
+
+static void __gnttab_unmap_refs_async(struct gntab_unmap_queue_data* item);
+
+static void gnttab_unmap_work(struct work_struct *work)
+{
+	struct gntab_unmap_queue_data
+		*unmap_data = container_of(work, 
+					   struct gntab_unmap_queue_data,
+					   gnttab_work.work);
+	if (unmap_data->age != UINT_MAX)
+		unmap_data->age++;
+	__gnttab_unmap_refs_async(unmap_data);
+}
+
+static void __gnttab_unmap_refs_async(struct gntab_unmap_queue_data* item)
+{
+	int ret;
+	int pc;
+
+	for (pc = 0; pc < item->count; pc++) {
+		if (page_count(item->pages[pc]) > 1) {
+			unsigned long delay = GNTTAB_UNMAP_REFS_DELAY * (item->age + 1);
+			schedule_delayed_work(&item->gnttab_work,
+					      msecs_to_jiffies(delay));
+			return;
+		}
+	}
+
+	ret = gnttab_unmap_refs(item->unmap_ops, item->kunmap_ops,
+				item->pages, item->count);
+	item->done(ret, item);
+}
+
+void gnttab_unmap_refs_async(struct gntab_unmap_queue_data* item)
+{
+	INIT_DELAYED_WORK(&item->gnttab_work, gnttab_unmap_work);
+	item->age = 0;
+
+	__gnttab_unmap_refs_async(item);
+}
+EXPORT_SYMBOL_GPL(gnttab_unmap_refs_async);
+
 static int gnttab_map_frames_v1(xen_pfn_t *frames, unsigned int nr_gframes)
 {
 	int rc;

drivers/xen/manage.c

@@ -105,10 +105,16 @@ static void do_suspend(void)
 
 	err = freeze_processes();
 	if (err) {
-		pr_err("%s: freeze failed %d\n", __func__, err);
+		pr_err("%s: freeze processes failed %d\n", __func__, err);
 		goto out;
 	}
 
+	err = freeze_kernel_threads();
+	if (err) {
+		pr_err("%s: freeze kernel threads failed %d\n", __func__, err);
+		goto out_thaw;
+	}
+
 	err = dpm_suspend_start(PMSG_FREEZE);
 	if (err) {
 		pr_err("%s: dpm_suspend_start %d\n", __func__, err);

drivers/xen/tmem.c

@@ -374,7 +374,7 @@ static struct frontswap_ops tmem_frontswap_ops = {
 };
 #endif
 
-static int xen_tmem_init(void)
+static int __init xen_tmem_init(void)
 {
 	if (!xen_domain())
 		return 0;

drivers/xen/xen-scsiback.c

@@ -227,7 +227,7 @@ static void put_free_pages(struct page **page, int num)
 		return;
 	if (i > scsiback_max_buffer_pages) {
 		n = min(num, i - scsiback_max_buffer_pages);
-		free_xenballooned_pages(n, page + num - n);
+		gnttab_free_pages(n, page + num - n);
 		n = num - n;
 	}
 	spin_lock_irqsave(&free_pages_lock, flags);
@@ -244,7 +244,7 @@ static int get_free_page(struct page **page)
 	spin_lock_irqsave(&free_pages_lock, flags);
 	if (list_empty(&scsiback_free_pages)) {
 		spin_unlock_irqrestore(&free_pages_lock, flags);
-		return alloc_xenballooned_pages(1, page, false);
+		return gnttab_alloc_pages(1, page);
 	}
 	page[0] = list_first_entry(&scsiback_free_pages, struct page, lru);
 	list_del(&page[0]->lru);
@@ -2106,7 +2106,7 @@ static void __exit scsiback_exit(void)
 	while (free_pages_num) {
 		if (get_free_page(&page))
 			BUG();
-		free_xenballooned_pages(1, &page);
+		gnttab_free_pages(1, &page);
 	}
 	scsiback_deregister_configfs();
 	xenbus_unregister_driver(&scsiback_driver);

drivers/xen/xenbus/xenbus_dev_frontend.c

@@ -326,10 +326,13 @@ static int xenbus_write_transaction(unsigned msg_type,
 	}
 
 	if (msg_type == XS_TRANSACTION_START) {
+		if (u->u.msg.type == XS_ERROR)
+			kfree(trans);
+		else {
 			trans->handle.id = simple_strtoul(reply, NULL, 0);
-
 			list_add(&trans->list, &u->transactions);
-	} else if (msg_type == XS_TRANSACTION_END) {
+		}
+	} else if (u->u.msg.type == XS_TRANSACTION_END) {
 		list_for_each_entry(trans, &u->transactions, list)
 			if (trans->handle.id == u->u.msg.tx_id)
 				break;

include/linux/mm.h

@@ -290,6 +290,14 @@ struct vm_operations_struct {
 	/* called by sys_remap_file_pages() to populate non-linear mapping */
 	int (*remap_pages)(struct vm_area_struct *vma, unsigned long addr,
 			   unsigned long size, pgoff_t pgoff);
+
+	/*
+	 * Called by vm_normal_page() for special PTEs to find the
+	 * page for @addr.  This is useful if the default behavior
+	 * (using pte_page()) would not find the correct page.
+	 */
+	struct page *(*find_special_page)(struct vm_area_struct *vma,
+					  unsigned long addr);
 };
 
 struct mmu_gather;

include/linux/page-flags.h

@@ -121,8 +121,12 @@ enum pageflags {
 	PG_fscache = PG_private_2,	/* page backed by cache */
 
 	/* XEN */
+	/* Pinned in Xen as a read-only pagetable page. */
 	PG_pinned = PG_owner_priv_1,
+	/* Pinned as part of domain save (see xen_mm_pin_all()). */
 	PG_savepinned = PG_dirty,
+	/* Has a grant mapping of another (foreign) domain's page. */
+	PG_foreign = PG_owner_priv_1,
 
 	/* SLOB */
 	PG_slob_free = PG_private,
@@ -215,6 +219,7 @@ __PAGEFLAG(Slab, slab)
 PAGEFLAG(Checked, checked)		/* Used by some filesystems */
 PAGEFLAG(Pinned, pinned) TESTSCFLAG(Pinned, pinned)	/* Xen */
 PAGEFLAG(SavePinned, savepinned);			/* Xen */
+PAGEFLAG(Foreign, foreign);				/* Xen */
 PAGEFLAG(Reserved, reserved) __CLEARPAGEFLAG(Reserved, reserved)
 PAGEFLAG(SwapBacked, swapbacked) __CLEARPAGEFLAG(SwapBacked, swapbacked)
 	__SETPAGEFLAG(SwapBacked, swapbacked)

include/xen/grant_table.h

@@ -45,6 +45,8 @@
 #include <asm/xen/hypervisor.h>
 
 #include <xen/features.h>
+#include <linux/mm_types.h>
+#include <linux/page-flags.h>
 
 #define GNTTAB_RESERVED_XENSTORE 1
 
@@ -58,6 +60,22 @@ struct gnttab_free_callback {
 	u16 count;
 };
 
+struct gntab_unmap_queue_data;
+
+typedef void (*gnttab_unmap_refs_done)(int result, struct gntab_unmap_queue_data *data);
+
+struct gntab_unmap_queue_data
+{
+	struct delayed_work	gnttab_work;
+	void *data;
+	gnttab_unmap_refs_done	done;
+	struct gnttab_unmap_grant_ref *unmap_ops;
+	struct gnttab_unmap_grant_ref *kunmap_ops;
+	struct page **pages;
+	unsigned int count;
+	unsigned int age;
+};
+
 int gnttab_init(void);
 int gnttab_suspend(void);
 int gnttab_resume(void);
@@ -163,12 +181,17 @@ void gnttab_free_auto_xlat_frames(void);
 
 #define gnttab_map_vaddr(map) ((void *)(map.host_virt_addr))
 
+int gnttab_alloc_pages(int nr_pages, struct page **pages);
+void gnttab_free_pages(int nr_pages, struct page **pages);
+
 int gnttab_map_refs(struct gnttab_map_grant_ref *map_ops,
 		    struct gnttab_map_grant_ref *kmap_ops,
 		    struct page **pages, unsigned int count);
 int gnttab_unmap_refs(struct gnttab_unmap_grant_ref *unmap_ops,
-		      struct gnttab_map_grant_ref *kunmap_ops,
+		      struct gnttab_unmap_grant_ref *kunmap_ops,
 		      struct page **pages, unsigned int count);
+void gnttab_unmap_refs_async(struct gntab_unmap_queue_data* item);
+
 
 /* Perform a batch of grant map/copy operations. Retry every batch slot
  * for which the hypervisor returns GNTST_eagain. This is typically due
@@ -182,4 +205,22 @@ int gnttab_unmap_refs(struct gnttab_unmap_grant_ref *unmap_ops,
 void gnttab_batch_map(struct gnttab_map_grant_ref *batch, unsigned count);
 void gnttab_batch_copy(struct gnttab_copy *batch, unsigned count);
 
+
+struct xen_page_foreign {
+	domid_t domid;
+	grant_ref_t gref;
+};
+
+static inline struct xen_page_foreign *xen_page_foreign(struct page *page)
+{
+	if (!PageForeign(page))
+		return NULL;
+#if BITS_PER_LONG < 64
+	return (struct xen_page_foreign *)page->private;
+#else
+	BUILD_BUG_ON(sizeof(struct xen_page_foreign) > BITS_PER_LONG);
+	return (struct xen_page_foreign *)&page->private;
+#endif
+}
+
 #endif /* __ASM_GNTTAB_H__ */

include/xen/interface/features.h

@@ -41,6 +41,12 @@
 /* x86: Does this Xen host support the MMU_PT_UPDATE_PRESERVE_AD hypercall? */
 #define XENFEAT_mmu_pt_update_preserve_ad  5
 
+/*
+ * If set, GNTTABOP_map_grant_ref honors flags to be placed into guest kernel
+ * available pte bits.
+ */
+#define XENFEAT_gnttab_map_avail_bits      7
+
 /* x86: Does this Xen host support the HVM callback vector type? */
 #define XENFEAT_hvm_callback_vector        8
 

include/xen/interface/grant_table.h

@@ -525,6 +525,13 @@ DEFINE_GUEST_HANDLE_STRUCT(gnttab_cache_flush);
 #define _GNTMAP_contains_pte    (4)
 #define GNTMAP_contains_pte     (1<<_GNTMAP_contains_pte)
 
+/*
+ * Bits to be placed in guest kernel available PTE bits (architecture
+ * dependent; only supported when XENFEAT_gnttab_map_avail_bits is set).
+ */
+#define _GNTMAP_guest_avail0    (16)
+#define GNTMAP_guest_avail_mask ((uint32_t)~0 << _GNTMAP_guest_avail0)
+
 /*
  * Values for error status returns. All errors are -ve.
  */

mm/memory.c

@@ -754,6 +754,8 @@ struct page *vm_normal_page(struct vm_area_struct *vma, unsigned long addr,
 	if (HAVE_PTE_SPECIAL) {
 		if (likely(!pte_special(pte)))
 			goto check_pfn;
+		if (vma->vm_ops && vma->vm_ops->find_special_page)
+			return vma->vm_ops->find_special_page(vma, addr);
 		if (vma->vm_flags & (VM_PFNMAP | VM_MIXEDMAP))
 			return NULL;
 		if (!is_zero_pfn(pfn))