Merge tag 'stable/for-linus-3.7-x86-tag' of...

Merge tag 'stable/for-linus-3.7-x86-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/konrad/xen

Pull Xen update from Konrad Rzeszutek Wilk:
 "Features:
   - When hotplugging PCI devices in a PV guest we can allocate
     Xen-SWIOTLB later.
   - Cleanup Xen SWIOTLB.
   - Support pages out grants from HVM domains in the backends.
   - Support wild cards in xen-pciback.hide=(BDF) arguments.
   - Update grant status updates with upstream hypervisor.
   - Boot PV guests with more than 128GB.
   - Cleanup Xen MMU code/add comments.
   - Obtain XENVERS using a preferred method.
   - Lay out generic changes to support Xen ARM.
   - Allow privcmd ioctl for HVM (used to do only PV).
   - Do v2 of mmap_batch for privcmd ioctls.
   - If hypervisor saves the LED keyboard light - we will now instruct
     the kernel about its state.
  Fixes:
   - More fixes to Xen PCI backend for various calls/FLR/etc.
   - With more than 4GB in a 64-bit PV guest disable native SWIOTLB.
   - Fix up smatch warnings.
   - Fix up various return values in privmcmd and mm."

* tag 'stable/for-linus-3.7-x86-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/konrad/xen: (48 commits)
  xen/pciback: Restore the PCI config space after an FLR.
  xen-pciback: properly clean up after calling pcistub_device_find()
  xen/vga: add the xen EFI video mode support
  xen/x86: retrieve keyboard shift status flags from hypervisor.
  xen/gndev: Xen backend support for paged out grant targets V4.
  xen-pciback: support wild cards in slot specifications
  xen/swiotlb: Fix compile warnings when using plain integer instead of NULL pointer.
  xen/swiotlb: Remove functions not needed anymore.
  xen/pcifront: Use Xen-SWIOTLB when initting if required.
  xen/swiotlb: For early initialization, return zero on success.
  xen/swiotlb: Use the swiotlb_late_init_with_tbl to init Xen-SWIOTLB late when PV PCI is used.
  xen/swiotlb: Move the error strings to its own function.
  xen/swiotlb: Move the nr_tbl determination in its own function.
  xen/arm: compile and run xenbus
  xen: resynchronise grant table status codes with upstream
  xen/privcmd: return -EFAULT on error
  xen/privcmd: Fix mmap batch ioctl error status copy back.
  xen/privcmd: add PRIVCMD_MMAPBATCH_V2 ioctl
  xen/mm: return more precise error from xen_remap_domain_range()
  xen/mmu: If the revector fails, don't attempt to revector anything else.
  ...

arch/ia64/include/asm/xen/interface.h

@@ -67,6 +67,10 @@
 #define set_xen_guest_handle(hnd, val)	do { (hnd).p = val; } while (0)
 
 #ifndef __ASSEMBLY__
+/* Explicitly size integers that represent pfns in the public interface
+ * with Xen so that we could have one ABI that works for 32 and 64 bit
+ * guests. */
+typedef unsigned long xen_pfn_t;
 /* Guest handles for primitive C types. */
 __DEFINE_GUEST_HANDLE(uchar, unsigned char);
 __DEFINE_GUEST_HANDLE(uint, unsigned int);
@@ -79,7 +83,6 @@ DEFINE_GUEST_HANDLE(void);
 DEFINE_GUEST_HANDLE(uint64_t);
 DEFINE_GUEST_HANDLE(uint32_t);
 
-typedef unsigned long xen_pfn_t;
 DEFINE_GUEST_HANDLE(xen_pfn_t);
 #define PRI_xen_pfn	"lx"
 #endif
@@ -265,6 +268,8 @@ typedef struct xen_callback xen_callback_t;
 
 #endif /* !__ASSEMBLY__ */
 
+#include <asm/pvclock-abi.h>
+
 /* Size of the shared_info area (this is not related to page size).  */
 #define XSI_SHIFT			14
 #define XSI_SIZE			(1 << XSI_SHIFT)

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

@@ -47,6 +47,10 @@
 #endif
 
 #ifndef __ASSEMBLY__
+/* Explicitly size integers that represent pfns in the public interface
+ * with Xen so that on ARM we can have one ABI that works for 32 and 64
+ * bit guests. */
+typedef unsigned long xen_pfn_t;
 /* Guest handles for primitive C types. */
 __DEFINE_GUEST_HANDLE(uchar, unsigned char);
 __DEFINE_GUEST_HANDLE(uint,  unsigned int);
@@ -57,6 +61,7 @@ DEFINE_GUEST_HANDLE(long);
 DEFINE_GUEST_HANDLE(void);
 DEFINE_GUEST_HANDLE(uint64_t);
 DEFINE_GUEST_HANDLE(uint32_t);
+DEFINE_GUEST_HANDLE(xen_pfn_t);
 #endif
 
 #ifndef HYPERVISOR_VIRT_START
@@ -121,6 +126,8 @@ struct arch_shared_info {
 #include "interface_64.h"
 #endif
 
+#include <asm/pvclock-abi.h>
+
 #ifndef __ASSEMBLY__
 /*
  * The following is all CPU context. Note that the fpu_ctxt block is filled

arch/x86/include/asm/xen/swiotlb-xen.h

@@ -5,10 +5,12 @@
 extern int xen_swiotlb;
 extern int __init pci_xen_swiotlb_detect(void);
 extern void __init pci_xen_swiotlb_init(void);
+extern int pci_xen_swiotlb_init_late(void);
 #else
 #define xen_swiotlb (0)
 static inline int __init pci_xen_swiotlb_detect(void) { return 0; }
 static inline void __init pci_xen_swiotlb_init(void) { }
+static inline int pci_xen_swiotlb_init_late(void) { return -ENXIO; }
 #endif
 
 #endif /* _ASM_X86_SWIOTLB_XEN_H */

arch/x86/xen/apic.c

@@ -6,8 +6,9 @@
 
 #include <xen/xen.h>
 #include <xen/interface/physdev.h>
+#include "xen-ops.h"
 
-unsigned int xen_io_apic_read(unsigned apic, unsigned reg)
+static unsigned int xen_io_apic_read(unsigned apic, unsigned reg)
 {
 	struct physdev_apic apic_op;
 	int ret;

arch/x86/xen/enlighten.c

@@ -80,6 +80,8 @@
 #include "smp.h"
 #include "multicalls.h"
 
+#include <xen/events.h>
+
 EXPORT_SYMBOL_GPL(hypercall_page);
 
 DEFINE_PER_CPU(struct vcpu_info *, xen_vcpu);
@@ -1288,7 +1290,6 @@ asmlinkage void __init xen_start_kernel(void)
 {
 	struct physdev_set_iopl set_iopl;
 	int rc;
-	pgd_t *pgd;
 
 	if (!xen_start_info)
 		return;
@@ -1380,8 +1381,6 @@ asmlinkage void __init xen_start_kernel(void)
 	acpi_numa = -1;
 #endif
 
-	pgd = (pgd_t *)xen_start_info->pt_base;
-
 	/* Don't do the full vcpu_info placement stuff until we have a
 	   possible map and a non-dummy shared_info. */
 	per_cpu(xen_vcpu, 0) = &HYPERVISOR_shared_info->vcpu_info[0];
@@ -1390,7 +1389,7 @@ asmlinkage void __init xen_start_kernel(void)
 	early_boot_irqs_disabled = true;
 
 	xen_raw_console_write("mapping kernel into physical memory\n");
-	pgd = xen_setup_kernel_pagetable(pgd, xen_start_info->nr_pages);
+	xen_setup_kernel_pagetable((pgd_t *)xen_start_info->pt_base, xen_start_info->nr_pages);
 
 	/* Allocate and initialize top and mid mfn levels for p2m structure */
 	xen_build_mfn_list_list();
@@ -1441,11 +1440,19 @@ asmlinkage void __init xen_start_kernel(void)
 		const struct dom0_vga_console_info *info =
 			(void *)((char *)xen_start_info +
 				 xen_start_info->console.dom0.info_off);
+		struct xen_platform_op op = {
+			.cmd = XENPF_firmware_info,
+			.interface_version = XENPF_INTERFACE_VERSION,
+			.u.firmware_info.type = XEN_FW_KBD_SHIFT_FLAGS,
+		};
 
 		xen_init_vga(info, xen_start_info->console.dom0.info_size);
 		xen_start_info->console.domU.mfn = 0;
 		xen_start_info->console.domU.evtchn = 0;
 
+		if (HYPERVISOR_dom0_op(&op) == 0)
+			boot_params.kbd_status = op.u.firmware_info.u.kbd_shift_flags;
+
 		xen_init_apic();
 
 		/* Make sure ACS will be enabled */

arch/x86/xen/mmu.c

@@ -84,6 +84,7 @@
  */
 DEFINE_SPINLOCK(xen_reservation_lock);
 
+#ifdef CONFIG_X86_32
 /*
  * Identity map, in addition to plain kernel map.  This needs to be
  * large enough to allocate page table pages to allocate the rest.
@@ -91,7 +92,7 @@ DEFINE_SPINLOCK(xen_reservation_lock);
  */
 #define LEVEL1_IDENT_ENTRIES	(PTRS_PER_PTE * 4)
 static RESERVE_BRK_ARRAY(pte_t, level1_ident_pgt, LEVEL1_IDENT_ENTRIES);
-
+#endif
 #ifdef CONFIG_X86_64
 /* l3 pud for userspace vsyscall mapping */
 static pud_t level3_user_vsyscall[PTRS_PER_PUD] __page_aligned_bss;
@@ -1176,13 +1177,6 @@ static void xen_exit_mmap(struct mm_struct *mm)
 
 static void xen_post_allocator_init(void);
 
-static void __init xen_pagetable_init(void)
-{
-	paging_init();
-	xen_setup_shared_info();
-	xen_post_allocator_init();
-}
-
 static __init void xen_mapping_pagetable_reserve(u64 start, u64 end)
 {
 	/* reserve the range used */
@@ -1197,6 +1191,87 @@ static __init void xen_mapping_pagetable_reserve(u64 start, u64 end)
 	}
 }
 
+#ifdef CONFIG_X86_64
+static void __init xen_cleanhighmap(unsigned long vaddr,
+				    unsigned long vaddr_end)
+{
+	unsigned long kernel_end = roundup((unsigned long)_brk_end, PMD_SIZE) - 1;
+	pmd_t *pmd = level2_kernel_pgt + pmd_index(vaddr);
+
+	/* NOTE: The loop is more greedy than the cleanup_highmap variant.
+	 * We include the PMD passed in on _both_ boundaries. */
+	for (; vaddr <= vaddr_end && (pmd < (level2_kernel_pgt + PAGE_SIZE));
+			pmd++, vaddr += PMD_SIZE) {
+		if (pmd_none(*pmd))
+			continue;
+		if (vaddr < (unsigned long) _text || vaddr > kernel_end)
+			set_pmd(pmd, __pmd(0));
+	}
+	/* In case we did something silly, we should crash in this function
+	 * instead of somewhere later and be confusing. */
+	xen_mc_flush();
+}
+#endif
+static void __init xen_pagetable_init(void)
+{
+#ifdef CONFIG_X86_64
+	unsigned long size;
+	unsigned long addr;
+#endif
+	paging_init();
+	xen_setup_shared_info();
+#ifdef CONFIG_X86_64
+	if (!xen_feature(XENFEAT_auto_translated_physmap)) {
+		unsigned long new_mfn_list;
+
+		size = PAGE_ALIGN(xen_start_info->nr_pages * sizeof(unsigned long));
+
+		/* On 32-bit, we get zero so this never gets executed. */
+		new_mfn_list = xen_revector_p2m_tree();
+		if (new_mfn_list && new_mfn_list != xen_start_info->mfn_list) {
+			/* using __ka address and sticking INVALID_P2M_ENTRY! */
+			memset((void *)xen_start_info->mfn_list, 0xff, size);
+
+			/* We should be in __ka space. */
+			BUG_ON(xen_start_info->mfn_list < __START_KERNEL_map);
+			addr = xen_start_info->mfn_list;
+			/* We roundup to the PMD, which means that if anybody at this stage is
+			 * using the __ka address of xen_start_info or xen_start_info->shared_info
+			 * they are in going to crash. Fortunatly we have already revectored
+			 * in xen_setup_kernel_pagetable and in xen_setup_shared_info. */
+			size = roundup(size, PMD_SIZE);
+			xen_cleanhighmap(addr, addr + size);
+
+			size = PAGE_ALIGN(xen_start_info->nr_pages * sizeof(unsigned long));
+			memblock_free(__pa(xen_start_info->mfn_list), size);
+			/* And revector! Bye bye old array */
+			xen_start_info->mfn_list = new_mfn_list;
+		} else
+			goto skip;
+	}
+	/* At this stage, cleanup_highmap has already cleaned __ka space
+	 * from _brk_limit way up to the max_pfn_mapped (which is the end of
+	 * the ramdisk). We continue on, erasing PMD entries that point to page
+	 * tables - do note that they are accessible at this stage via __va.
+	 * For good measure we also round up to the PMD - which means that if
+	 * anybody is using __ka address to the initial boot-stack - and try
+	 * to use it - they are going to crash. The xen_start_info has been
+	 * taken care of already in xen_setup_kernel_pagetable. */
+	addr = xen_start_info->pt_base;
+	size = roundup(xen_start_info->nr_pt_frames * PAGE_SIZE, PMD_SIZE);
+
+	xen_cleanhighmap(addr, addr + size);
+	xen_start_info->pt_base = (unsigned long)__va(__pa(xen_start_info->pt_base));
+#ifdef DEBUG
+	/* This is superflous and is not neccessary, but you know what
+	 * lets do it. The MODULES_VADDR -> MODULES_END should be clear of
+	 * anything at this stage. */
+	xen_cleanhighmap(MODULES_VADDR, roundup(MODULES_VADDR, PUD_SIZE) - 1);
+#endif
+skip:
+#endif
+	xen_post_allocator_init();
+}
 static void xen_write_cr2(unsigned long cr2)
 {
 	this_cpu_read(xen_vcpu)->arch.cr2 = cr2;
@@ -1652,7 +1727,7 @@ static void set_page_prot(void *addr, pgprot_t prot)
 	if (HYPERVISOR_update_va_mapping((unsigned long)addr, pte, 0))
 		BUG();
 }
-
+#ifdef CONFIG_X86_32
 static void __init xen_map_identity_early(pmd_t *pmd, unsigned long max_pfn)
 {
 	unsigned pmdidx, pteidx;
@@ -1703,7 +1778,7 @@ static void __init xen_map_identity_early(pmd_t *pmd, unsigned long max_pfn)
 
 	set_page_prot(pmd, PAGE_KERNEL_RO);
 }
-
+#endif
 void __init xen_setup_machphys_mapping(void)
 {
 	struct xen_machphys_mapping mapping;
@@ -1731,7 +1806,20 @@ static void convert_pfn_mfn(void *v)
 	for (i = 0; i < PTRS_PER_PTE; i++)
 		pte[i] = xen_make_pte(pte[i].pte);
 }
-
+static void __init check_pt_base(unsigned long *pt_base, unsigned long *pt_end,
+				 unsigned long addr)
+{
+	if (*pt_base == PFN_DOWN(__pa(addr))) {
+		set_page_prot((void *)addr, PAGE_KERNEL);
+		clear_page((void *)addr);
+		(*pt_base)++;
+	}
+	if (*pt_end == PFN_DOWN(__pa(addr))) {
+		set_page_prot((void *)addr, PAGE_KERNEL);
+		clear_page((void *)addr);
+		(*pt_end)--;
+	}
+}
 /*
  * Set up the initial kernel pagetable.
  *
@@ -1743,11 +1831,13 @@ static void convert_pfn_mfn(void *v)
  * of the physical mapping once some sort of allocator has been set
  * up.
  */
-pgd_t * __init xen_setup_kernel_pagetable(pgd_t *pgd,
-					 unsigned long max_pfn)
+void __init xen_setup_kernel_pagetable(pgd_t *pgd, unsigned long max_pfn)
 {
 	pud_t *l3;
 	pmd_t *l2;
+	unsigned long addr[3];
+	unsigned long pt_base, pt_end;
+	unsigned i;
 
 	/* max_pfn_mapped is the last pfn mapped in the initial memory
 	 * mappings. Considering that on Xen after the kernel mappings we
@@ -1755,32 +1845,53 @@ pgd_t * __init xen_setup_kernel_pagetable(pgd_t *pgd,
 	 * set max_pfn_mapped to the last real pfn mapped. */
 	max_pfn_mapped = PFN_DOWN(__pa(xen_start_info->mfn_list));
 
+	pt_base = PFN_DOWN(__pa(xen_start_info->pt_base));
+	pt_end = pt_base + xen_start_info->nr_pt_frames;
+
 	/* Zap identity mapping */
 	init_level4_pgt[0] = __pgd(0);
 
 	/* Pre-constructed entries are in pfn, so convert to mfn */
+	/* L4[272] -> level3_ident_pgt
+	 * L4[511] -> level3_kernel_pgt */
 	convert_pfn_mfn(init_level4_pgt);
+
+	/* L3_i[0] -> level2_ident_pgt */
 	convert_pfn_mfn(level3_ident_pgt);
+	/* L3_k[510] -> level2_kernel_pgt
+	 * L3_i[511] -> level2_fixmap_pgt */
 	convert_pfn_mfn(level3_kernel_pgt);
 
+	/* We get [511][511] and have Xen's version of level2_kernel_pgt */
 	l3 = m2v(pgd[pgd_index(__START_KERNEL_map)].pgd);
 	l2 = m2v(l3[pud_index(__START_KERNEL_map)].pud);
 
-	memcpy(level2_ident_pgt, l2, sizeof(pmd_t) * PTRS_PER_PMD);
-	memcpy(level2_kernel_pgt, l2, sizeof(pmd_t) * PTRS_PER_PMD);
-
+	addr[0] = (unsigned long)pgd;
+	addr[1] = (unsigned long)l3;
+	addr[2] = (unsigned long)l2;
+	/* Graft it onto L4[272][0]. Note that we creating an aliasing problem:
+	 * Both L4[272][0] and L4[511][511] have entries that point to the same
+	 * L2 (PMD) tables. Meaning that if you modify it in __va space
+	 * it will be also modified in the __ka space! (But if you just
+	 * modify the PMD table to point to other PTE's or none, then you
+	 * are OK - which is what cleanup_highmap does) */
+	copy_page(level2_ident_pgt, l2);
+	/* Graft it onto L4[511][511] */
+	copy_page(level2_kernel_pgt, l2);
+
+	/* Get [511][510] and graft that in level2_fixmap_pgt */
 	l3 = m2v(pgd[pgd_index(__START_KERNEL_map + PMD_SIZE)].pgd);
 	l2 = m2v(l3[pud_index(__START_KERNEL_map + PMD_SIZE)].pud);
-	memcpy(level2_fixmap_pgt, l2, sizeof(pmd_t) * PTRS_PER_PMD);
-
-	/* Set up identity map */
-	xen_map_identity_early(level2_ident_pgt, max_pfn);
+	copy_page(level2_fixmap_pgt, l2);
+	/* Note that we don't do anything with level1_fixmap_pgt which
+	 * we don't need. */
 
 	/* Make pagetable pieces RO */
 	set_page_prot(init_level4_pgt, PAGE_KERNEL_RO);
 	set_page_prot(level3_ident_pgt, PAGE_KERNEL_RO);
 	set_page_prot(level3_kernel_pgt, PAGE_KERNEL_RO);
 	set_page_prot(level3_user_vsyscall, PAGE_KERNEL_RO);
+	set_page_prot(level2_ident_pgt, PAGE_KERNEL_RO);
 	set_page_prot(level2_kernel_pgt, PAGE_KERNEL_RO);
 	set_page_prot(level2_fixmap_pgt, PAGE_KERNEL_RO);
 
@@ -1791,22 +1902,28 @@ pgd_t * __init xen_setup_kernel_pagetable(pgd_t *pgd,
 	/* Unpin Xen-provided one */
 	pin_pagetable_pfn(MMUEXT_UNPIN_TABLE, PFN_DOWN(__pa(pgd)));
 
-	/* Switch over */
-	pgd = init_level4_pgt;
-
 	/*
 	 * At this stage there can be no user pgd, and no page
 	 * structure to attach it to, so make sure we just set kernel
 	 * pgd.
 	 */
 	xen_mc_batch();
-	__xen_write_cr3(true, __pa(pgd));
+	__xen_write_cr3(true, __pa(init_level4_pgt));
 	xen_mc_issue(PARAVIRT_LAZY_CPU);
 
-	memblock_reserve(__pa(xen_start_info->pt_base),
-			 xen_start_info->nr_pt_frames * PAGE_SIZE);
+	/* We can't that easily rip out L3 and L2, as the Xen pagetables are
+	 * set out this way: [L4], [L1], [L2], [L3], [L1], [L1] ...  for
+	 * the initial domain. For guests using the toolstack, they are in:
+	 * [L4], [L3], [L2], [L1], [L1], order .. So for dom0 we can only
+	 * rip out the [L4] (pgd), but for guests we shave off three pages.
+	 */
+	for (i = 0; i < ARRAY_SIZE(addr); i++)
+		check_pt_base(&pt_base, &pt_end, addr[i]);
 
-	return pgd;
+	/* Our (by three pages) smaller Xen pagetable that we are using */
+	memblock_reserve(PFN_PHYS(pt_base), (pt_end - pt_base) * PAGE_SIZE);
+	/* Revector the xen_start_info */
+	xen_start_info = (struct start_info *)__va(__pa(xen_start_info));
 }
 #else	/* !CONFIG_X86_64 */
 static RESERVE_BRK_ARRAY(pmd_t, initial_kernel_pmd, PTRS_PER_PMD);
@@ -1831,8 +1948,7 @@ static void __init xen_write_cr3_init(unsigned long cr3)
 	 */
 	swapper_kernel_pmd =
 		extend_brk(sizeof(pmd_t) * PTRS_PER_PMD, PAGE_SIZE);
-	memcpy(swapper_kernel_pmd, initial_kernel_pmd,
-	       sizeof(pmd_t) * PTRS_PER_PMD);
+	copy_page(swapper_kernel_pmd, initial_kernel_pmd);
 	swapper_pg_dir[KERNEL_PGD_BOUNDARY] =
 		__pgd(__pa(swapper_kernel_pmd) | _PAGE_PRESENT);
 	set_page_prot(swapper_kernel_pmd, PAGE_KERNEL_RO);
@@ -1849,8 +1965,7 @@ static void __init xen_write_cr3_init(unsigned long cr3)
 	pv_mmu_ops.write_cr3 = &xen_write_cr3;
 }
 
-pgd_t * __init xen_setup_kernel_pagetable(pgd_t *pgd,
-					 unsigned long max_pfn)
+void __init xen_setup_kernel_pagetable(pgd_t *pgd, unsigned long max_pfn)
 {
 	pmd_t *kernel_pmd;
 
@@ -1862,11 +1977,11 @@ pgd_t * __init xen_setup_kernel_pagetable(pgd_t *pgd,
 				  512*1024);
 
 	kernel_pmd = m2v(pgd[KERNEL_PGD_BOUNDARY].pgd);
-	memcpy(initial_kernel_pmd, kernel_pmd, sizeof(pmd_t) * PTRS_PER_PMD);
+	copy_page(initial_kernel_pmd, kernel_pmd);
 
 	xen_map_identity_early(initial_kernel_pmd, max_pfn);
 
-	memcpy(initial_page_table, pgd, sizeof(pgd_t) * PTRS_PER_PGD);
+	copy_page(initial_page_table, pgd);
 	initial_page_table[KERNEL_PGD_BOUNDARY] =
 		__pgd(__pa(initial_kernel_pmd) | _PAGE_PRESENT);
 
@@ -1882,8 +1997,6 @@ pgd_t * __init xen_setup_kernel_pagetable(pgd_t *pgd,
 
 	memblock_reserve(__pa(xen_start_info->pt_base),
 			 xen_start_info->nr_pt_frames * PAGE_SIZE);
-
-	return initial_page_table;
 }
 #endif	/* CONFIG_X86_64 */
 
@@ -2333,6 +2446,9 @@ int xen_remap_domain_mfn_range(struct vm_area_struct *vma,
 	unsigned long range;
 	int err = 0;
 
+	if (xen_feature(XENFEAT_auto_translated_physmap))
+		return -EINVAL;
+
 	prot = __pgprot(pgprot_val(prot) | _PAGE_IOMAP);
 
 	BUG_ON(!((vma->vm_flags & (VM_PFNMAP | VM_RESERVED | VM_IO)) ==
@@ -2351,8 +2467,8 @@ int xen_remap_domain_mfn_range(struct vm_area_struct *vma,
 		if (err)
 			goto out;
 
-		err = -EFAULT;
-		if (HYPERVISOR_mmu_update(mmu_update, batch, NULL, domid) < 0)
+		err = HYPERVISOR_mmu_update(mmu_update, batch, NULL, domid);
+		if (err < 0)
 			goto out;
 
 		nr -= batch;

arch/x86/xen/p2m.c

@@ -22,7 +22,7 @@
  *
  * P2M_PER_PAGE depends on the architecture, as a mfn is always
  * unsigned long (8 bytes on 64-bit, 4 bytes on 32), leading to
- * 512 and 1024 entries respectively. 
+ * 512 and 1024 entries respectively.
  *
  * In short, these structures contain the Machine Frame Number (MFN) of the PFN.
  *
@@ -139,11 +139,11 @@
  *      /    | ~0, ~0, ....  |
  *     |     \---------------/
  *     |
- *     p2m_missing             p2m_missing
- * /------------------\     /------------\
- * | [p2m_mid_missing]+---->| ~0, ~0, ~0 |
- * | [p2m_mid_missing]+---->| ..., ~0    |
- * \------------------/     \------------/
+ *   p2m_mid_missing           p2m_missing
+ * /-----------------\     /------------\
+ * | [p2m_missing]   +---->| ~0, ~0, ~0 |
+ * | [p2m_missing]   +---->| ..., ~0    |
+ * \-----------------/     \------------/
  *
  * where ~0 is INVALID_P2M_ENTRY. IDENTITY is (PFN | IDENTITY_BIT)
  */
@@ -396,7 +396,85 @@ void __init xen_build_dynamic_phys_to_machine(void)
 
 	m2p_override_init();
 }
+#ifdef CONFIG_X86_64
+#include <linux/bootmem.h>
+unsigned long __init xen_revector_p2m_tree(void)
+{
+	unsigned long va_start;
+	unsigned long va_end;
+	unsigned long pfn;
+	unsigned long pfn_free = 0;
+	unsigned long *mfn_list = NULL;
+	unsigned long size;
+
+	va_start = xen_start_info->mfn_list;
+	/*We copy in increments of P2M_PER_PAGE * sizeof(unsigned long),
+	 * so make sure it is rounded up to that */
+	size = PAGE_ALIGN(xen_start_info->nr_pages * sizeof(unsigned long));
+	va_end = va_start + size;
+
+	/* If we were revectored already, don't do it again. */
+	if (va_start <= __START_KERNEL_map && va_start >= __PAGE_OFFSET)
+		return 0;
+
+	mfn_list = alloc_bootmem_align(size, PAGE_SIZE);
+	if (!mfn_list) {
+		pr_warn("Could not allocate space for a new P2M tree!\n");
+		return xen_start_info->mfn_list;
+	}
+	/* Fill it out with INVALID_P2M_ENTRY value */
+	memset(mfn_list, 0xFF, size);
+
+	for (pfn = 0; pfn < ALIGN(MAX_DOMAIN_PAGES, P2M_PER_PAGE); pfn += P2M_PER_PAGE) {
+		unsigned topidx = p2m_top_index(pfn);
+		unsigned mididx;
+		unsigned long *mid_p;
+
+		if (!p2m_top[topidx])
+			continue;
+
+		if (p2m_top[topidx] == p2m_mid_missing)
+			continue;
+
+		mididx = p2m_mid_index(pfn);
+		mid_p = p2m_top[topidx][mididx];
+		if (!mid_p)
+			continue;
+		if ((mid_p == p2m_missing) || (mid_p == p2m_identity))
+			continue;
+
+		if ((unsigned long)mid_p == INVALID_P2M_ENTRY)
+			continue;
+
+		/* The old va. Rebase it on mfn_list */
+		if (mid_p >= (unsigned long *)va_start && mid_p <= (unsigned long *)va_end) {
+			unsigned long *new;
+
+			if (pfn_free  > (size / sizeof(unsigned long))) {
+				WARN(1, "Only allocated for %ld pages, but we want %ld!\n",
+				     size / sizeof(unsigned long), pfn_free);
+				return 0;
+			}
+			new = &mfn_list[pfn_free];
+
+			copy_page(new, mid_p);
+			p2m_top[topidx][mididx] = &mfn_list[pfn_free];
+			p2m_top_mfn_p[topidx][mididx] = virt_to_mfn(&mfn_list[pfn_free]);
+
+			pfn_free += P2M_PER_PAGE;
 
+		}
+		/* This should be the leafs allocated for identity from _brk. */
+	}
+	return (unsigned long)mfn_list;
+
+}
+#else
+unsigned long __init xen_revector_p2m_tree(void)
+{
+	return 0;
+}
+#endif
 unsigned long get_phys_to_machine(unsigned long pfn)
 {
 	unsigned topidx, mididx, idx;
@@ -430,7 +508,7 @@ static void free_p2m_page(void *p)
 	free_page((unsigned long)p);
 }
 
-/* 
+/*
  * Fully allocate the p2m structure for a given pfn.  We need to check
  * that both the top and mid levels are allocated, and make sure the
  * parallel mfn tree is kept in sync.  We may race with other cpus, so

arch/x86/xen/pci-swiotlb-xen.c

@@ -8,6 +8,14 @@
 #include <xen/xen.h>
 #include <asm/iommu_table.h>
 
+
+#include <asm/xen/swiotlb-xen.h>
+#ifdef CONFIG_X86_64
+#include <asm/iommu.h>
+#include <asm/dma.h>
+#endif
+#include <linux/export.h>
+
 int xen_swiotlb __read_mostly;
 
 static struct dma_map_ops xen_swiotlb_dma_ops = {
@@ -34,34 +42,64 @@ static struct dma_map_ops xen_swiotlb_dma_ops = {
 int __init pci_xen_swiotlb_detect(void)
 {
 
+	if (!xen_pv_domain())
+		return 0;
+
 	/* If running as PV guest, either iommu=soft, or swiotlb=force will
 	 * activate this IOMMU. If running as PV privileged, activate it
 	 * irregardless.
 	 */
-	if ((xen_initial_domain() || swiotlb || swiotlb_force) &&
-	    (xen_pv_domain()))
+	if ((xen_initial_domain() || swiotlb || swiotlb_force))
 		xen_swiotlb = 1;
 
 	/* If we are running under Xen, we MUST disable the native SWIOTLB.
 	 * Don't worry about swiotlb_force flag activating the native, as
 	 * the 'swiotlb' flag is the only one turning it on. */
-	if (xen_pv_domain())
-		swiotlb = 0;
+	swiotlb = 0;
 
+#ifdef CONFIG_X86_64
+	/* pci_swiotlb_detect_4gb turns on native SWIOTLB if no_iommu == 0
+	 * (so no iommu=X command line over-writes).
+	 * Considering that PV guests do not want the *native SWIOTLB* but
+	 * only Xen SWIOTLB it is not useful to us so set no_iommu=1 here.
+	 */
+	if (max_pfn > MAX_DMA32_PFN)
+		no_iommu = 1;
+#endif
 	return xen_swiotlb;
 }
 
 void __init pci_xen_swiotlb_init(void)
 {
 	if (xen_swiotlb) {
-		xen_swiotlb_init(1);
+		xen_swiotlb_init(1, true /* early */);
 		dma_ops = &xen_swiotlb_dma_ops;
 
 		/* Make sure ACS will be enabled */
 		pci_request_acs();
 	}
 }
+
+int pci_xen_swiotlb_init_late(void)
+{
+	int rc;
+
+	if (xen_swiotlb)
+		return 0;
+
+	rc = xen_swiotlb_init(1, false /* late */);
+	if (rc)
+		return rc;
+
+	dma_ops = &xen_swiotlb_dma_ops;
+	/* Make sure ACS will be enabled */
+	pci_request_acs();
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(pci_xen_swiotlb_init_late);
+
 IOMMU_INIT_FINISH(pci_xen_swiotlb_detect,
-		  0,
+		  NULL,
 		  pci_xen_swiotlb_init,
-		  0);
+		  NULL);

arch/x86/xen/platform-pci-unplug.c

@@ -24,6 +24,7 @@
 #include <linux/module.h>
 
 #include <xen/platform_pci.h>
+#include "xen-ops.h"
 
 #define XEN_PLATFORM_ERR_MAGIC -1
 #define XEN_PLATFORM_ERR_PROTOCOL -2

arch/x86/xen/setup.c

@@ -432,6 +432,24 @@ char * __init xen_memory_setup(void)
 	 *  - mfn_list
 	 *  - xen_start_info
 	 * See comment above "struct start_info" in <xen/interface/xen.h>
+	 * We tried to make the the memblock_reserve more selective so
+	 * that it would be clear what region is reserved. Sadly we ran
+	 * in the problem wherein on a 64-bit hypervisor with a 32-bit
+	 * initial domain, the pt_base has the cr3 value which is not
+	 * neccessarily where the pagetable starts! As Jan put it: "
+	 * Actually, the adjustment turns out to be correct: The page
+	 * tables for a 32-on-64 dom0 get allocated in the order "first L1",
+	 * "first L2", "first L3", so the offset to the page table base is
+	 * indeed 2. When reading xen/include/public/xen.h's comment
+	 * very strictly, this is not a violation (since there nothing is said
+	 * that the first thing in the page table space is pointed to by
+	 * pt_base; I admit that this seems to be implied though, namely
+	 * do I think that it is implied that the page table space is the
+	 * range [pt_base, pt_base + nt_pt_frames), whereas that
+	 * range here indeed is [pt_base - 2, pt_base - 2 + nt_pt_frames),
+	 * which - without a priori knowledge - the kernel would have
+	 * difficulty to figure out)." - so lets just fall back to the
+	 * easy way and reserve the whole region.
 	 */
 	memblock_reserve(__pa(xen_start_info->mfn_list),
 			 xen_start_info->pt_base - xen_start_info->mfn_list);

arch/x86/xen/vga.c

@@ -35,6 +35,7 @@ void __init xen_init_vga(const struct dom0_vga_console_info *info, size_t size)
 			info->u.text_mode_3.font_height;
 		break;
 
+	case XEN_VGATYPE_EFI_LFB:
 	case XEN_VGATYPE_VESA_LFB:
 		if (size < offsetof(struct dom0_vga_console_info,
 				    u.vesa_lfb.gbl_caps))
@@ -54,6 +55,12 @@ void __init xen_init_vga(const struct dom0_vga_console_info *info, size_t size)
 		screen_info->blue_pos = info->u.vesa_lfb.blue_pos;
 		screen_info->rsvd_size = info->u.vesa_lfb.rsvd_size;
 		screen_info->rsvd_pos = info->u.vesa_lfb.rsvd_pos;
+
+		if (info->video_type == XEN_VGATYPE_EFI_LFB) {
+			screen_info->orig_video_isVGA = VIDEO_TYPE_EFI;
+			break;
+		}
+
 		if (size >= offsetof(struct dom0_vga_console_info,
 				     u.vesa_lfb.gbl_caps)
 		    + sizeof(info->u.vesa_lfb.gbl_caps))

arch/x86/xen/xen-head.S

@@ -28,9 +28,61 @@ ENTRY(startup_xen)
 	__FINIT
 
 .pushsection .text
-	.align PAGE_SIZE
+	.balign PAGE_SIZE
 ENTRY(hypercall_page)
-	.skip PAGE_SIZE
+#define NEXT_HYPERCALL(x) \
+	ENTRY(xen_hypercall_##x) \
+	.skip 32
+
+NEXT_HYPERCALL(set_trap_table)
+NEXT_HYPERCALL(mmu_update)
+NEXT_HYPERCALL(set_gdt)
+NEXT_HYPERCALL(stack_switch)
+NEXT_HYPERCALL(set_callbacks)
+NEXT_HYPERCALL(fpu_taskswitch)
+NEXT_HYPERCALL(sched_op_compat)
+NEXT_HYPERCALL(platform_op)
+NEXT_HYPERCALL(set_debugreg)
+NEXT_HYPERCALL(get_debugreg)
+NEXT_HYPERCALL(update_descriptor)
+NEXT_HYPERCALL(ni)
+NEXT_HYPERCALL(memory_op)
+NEXT_HYPERCALL(multicall)
+NEXT_HYPERCALL(update_va_mapping)
+NEXT_HYPERCALL(set_timer_op)
+NEXT_HYPERCALL(event_channel_op_compat)
+NEXT_HYPERCALL(xen_version)
+NEXT_HYPERCALL(console_io)
+NEXT_HYPERCALL(physdev_op_compat)
+NEXT_HYPERCALL(grant_table_op)
+NEXT_HYPERCALL(vm_assist)
+NEXT_HYPERCALL(update_va_mapping_otherdomain)
+NEXT_HYPERCALL(iret)
+NEXT_HYPERCALL(vcpu_op)
+NEXT_HYPERCALL(set_segment_base)
+NEXT_HYPERCALL(mmuext_op)
+NEXT_HYPERCALL(xsm_op)
+NEXT_HYPERCALL(nmi_op)
+NEXT_HYPERCALL(sched_op)
+NEXT_HYPERCALL(callback_op)
+NEXT_HYPERCALL(xenoprof_op)
+NEXT_HYPERCALL(event_channel_op)
+NEXT_HYPERCALL(physdev_op)
+NEXT_HYPERCALL(hvm_op)
+NEXT_HYPERCALL(sysctl)
+NEXT_HYPERCALL(domctl)
+NEXT_HYPERCALL(kexec_op)
+NEXT_HYPERCALL(tmem_op) /* 38 */
+ENTRY(xen_hypercall_rsvr)
+	.skip 320
+NEXT_HYPERCALL(mca) /* 48 */
+NEXT_HYPERCALL(arch_1)
+NEXT_HYPERCALL(arch_2)
+NEXT_HYPERCALL(arch_3)
+NEXT_HYPERCALL(arch_4)
+NEXT_HYPERCALL(arch_5)
+NEXT_HYPERCALL(arch_6)
+	.balign PAGE_SIZE
 .popsection
 
 	ELFNOTE(Xen, XEN_ELFNOTE_GUEST_OS,       .asciz "linux")

arch/x86/xen/xen-ops.h

@@ -27,7 +27,7 @@ void xen_setup_mfn_list_list(void);
 void xen_setup_shared_info(void);
 void xen_build_mfn_list_list(void);
 void xen_setup_machphys_mapping(void);
-pgd_t *xen_setup_kernel_pagetable(pgd_t *pgd, unsigned long max_pfn);
+void xen_setup_kernel_pagetable(pgd_t *pgd, unsigned long max_pfn);
 void xen_reserve_top(void);
 extern unsigned long xen_max_p2m_pfn;
 
@@ -45,6 +45,7 @@ void xen_hvm_init_shared_info(void);
 void xen_unplug_emulated_devices(void);
 
 void __init xen_build_dynamic_phys_to_machine(void);
+unsigned long __init xen_revector_p2m_tree(void);
 
 void xen_init_irq_ops(void);
 void xen_setup_timer(int cpu);

drivers/net/xen-netback/netback.c

@@ -635,9 +635,7 @@ static void xen_netbk_rx_action(struct xen_netbk *netbk)
 		return;
 
 	BUG_ON(npo.copy_prod > ARRAY_SIZE(netbk->grant_copy_op));
-	ret = HYPERVISOR_grant_table_op(GNTTABOP_copy, &netbk->grant_copy_op,
-					npo.copy_prod);
-	BUG_ON(ret != 0);
+	gnttab_batch_copy(netbk->grant_copy_op, npo.copy_prod);
 
 	while ((skb = __skb_dequeue(&rxq)) != NULL) {
 		sco = (struct skb_cb_overlay *)skb->cb;
@@ -1460,18 +1458,15 @@ static void xen_netbk_tx_submit(struct xen_netbk *netbk)
 static void xen_netbk_tx_action(struct xen_netbk *netbk)
 {
 	unsigned nr_gops;
-	int ret;
 
 	nr_gops = xen_netbk_tx_build_gops(netbk);
 
 	if (nr_gops == 0)
 		return;
-	ret = HYPERVISOR_grant_table_op(GNTTABOP_copy,
-					netbk->tx_copy_ops, nr_gops);
-	BUG_ON(ret);
 
-	xen_netbk_tx_submit(netbk);
+	gnttab_batch_copy(netbk->tx_copy_ops, nr_gops);
 
+	xen_netbk_tx_submit(netbk);
 }
 
 static void xen_netbk_idx_release(struct xen_netbk *netbk, u16 pending_idx)

drivers/pci/xen-pcifront.c

@@ -21,6 +21,7 @@
 #include <linux/bitops.h>
 #include <linux/time.h>
 
+#include <asm/xen/swiotlb-xen.h>
 #define INVALID_GRANT_REF (0)
 #define INVALID_EVTCHN    (-1)
 
@@ -236,7 +237,7 @@ static int pcifront_bus_write(struct pci_bus *bus, unsigned int devfn,
 	return errno_to_pcibios_err(do_pci_op(pdev, &op));
 }
 
-struct pci_ops pcifront_bus_ops = {
+static struct pci_ops pcifront_bus_ops = {
 	.read = pcifront_bus_read,
 	.write = pcifront_bus_write,
 };
@@ -668,7 +669,7 @@ static irqreturn_t pcifront_handler_aer(int irq, void *dev)
 	schedule_pcifront_aer_op(pdev);
 	return IRQ_HANDLED;
 }
-static int pcifront_connect(struct pcifront_device *pdev)
+static int pcifront_connect_and_init_dma(struct pcifront_device *pdev)
 {
 	int err = 0;
 
@@ -681,9 +682,13 @@ static int pcifront_connect(struct pcifront_device *pdev)
 		dev_err(&pdev->xdev->dev, "PCI frontend already installed!\n");
 		err = -EEXIST;
 	}
-
 	spin_unlock(&pcifront_dev_lock);
 
+	if (!err && !swiotlb_nr_tbl()) {
+		err = pci_xen_swiotlb_init_late();
+		if (err)
+			dev_err(&pdev->xdev->dev, "Could not setup SWIOTLB!\n");
+	}
 	return err;
 }
 
@@ -842,10 +847,10 @@ static int __devinit pcifront_try_connect(struct pcifront_device *pdev)
 	    XenbusStateInitialised)
 		goto out;
 
-	err = pcifront_connect(pdev);
+	err = pcifront_connect_and_init_dma(pdev);
 	if (err) {
 		xenbus_dev_fatal(pdev->xdev, err,
-				 "Error connecting PCI Frontend");
+				 "Error setting up PCI Frontend");
 		goto out;
 	}
 

drivers/tty/hvc/hvc_xen.c

@@ -21,6 +21,7 @@
 #include <linux/console.h>
 #include <linux/delay.h>
 #include <linux/err.h>
+#include <linux/irq.h>
 #include <linux/init.h>
 #include <linux/types.h>
 #include <linux/list.h>
@@ -35,6 +36,7 @@
 #include <xen/page.h>
 #include <xen/events.h>
 #include <xen/interface/io/console.h>
+#include <xen/interface/sched.h>
 #include <xen/hvc-console.h>
 #include <xen/xenbus.h>
 

drivers/xen/events.c

@@ -373,11 +373,22 @@ static void unmask_evtchn(int port)
 {
 	struct shared_info *s = HYPERVISOR_shared_info;
 	unsigned int cpu = get_cpu();
+	int do_hypercall = 0, evtchn_pending = 0;
 
 	BUG_ON(!irqs_disabled());
 
-	/* Slow path (hypercall) if this is a non-local port. */
-	if (unlikely(cpu != cpu_from_evtchn(port))) {
+	if (unlikely((cpu != cpu_from_evtchn(port))))
+		do_hypercall = 1;
+	else
+		evtchn_pending = sync_test_bit(port, &s->evtchn_pending[0]);
+
+	if (unlikely(evtchn_pending && xen_hvm_domain()))
+		do_hypercall = 1;
+
+	/* Slow path (hypercall) if this is a non-local port or if this is
+	 * an hvm domain and an event is pending (hvm domains don't have
+	 * their own implementation of irq_enable). */
+	if (do_hypercall) {
 		struct evtchn_unmask unmask = { .port = port };
 		(void)HYPERVISOR_event_channel_op(EVTCHNOP_unmask, &unmask);
 	} else {
@@ -390,7 +401,7 @@ static void unmask_evtchn(int port)
 		 * 'hw_resend_irq'. Just like a real IO-APIC we 'lose
 		 * the interrupt edge' if the channel is masked.
 		 */
-		if (sync_test_bit(port, &s->evtchn_pending[0]) &&
+		if (evtchn_pending &&
 		    !sync_test_and_set_bit(port / BITS_PER_LONG,
 					   &vcpu_info->evtchn_pending_sel))
 			vcpu_info->evtchn_upcall_pending = 1;
@@ -831,6 +842,7 @@ int bind_evtchn_to_irq(unsigned int evtchn)
 		struct irq_info *info = info_for_irq(irq);
 		WARN_ON(info == NULL || info->type != IRQT_EVTCHN);
 	}
+	irq_clear_status_flags(irq, IRQ_NOREQUEST|IRQ_NOAUTOEN);
 
 out:
 	mutex_unlock(&irq_mapping_update_lock);

drivers/xen/gntdev.c

@@ -446,7 +446,7 @@ static void mn_release(struct mmu_notifier *mn,
 	spin_unlock(&priv->lock);
 }
 
-struct mmu_notifier_ops gntdev_mmu_ops = {
+static struct mmu_notifier_ops gntdev_mmu_ops = {
 	.release                = mn_release,
 	.invalidate_page        = mn_invl_page,
 	.invalidate_range_start = mn_invl_range_start,

drivers/xen/grant-table.c

@@ -38,6 +38,7 @@
 #include <linux/vmalloc.h>
 #include <linux/uaccess.h>
 #include <linux/io.h>
+#include <linux/delay.h>
 #include <linux/hardirq.h>
 
 #include <xen/xen.h>
@@ -47,6 +48,7 @@
 #include <xen/interface/memory.h>
 #include <xen/hvc-console.h>
 #include <asm/xen/hypercall.h>
+#include <asm/xen/interface.h>
 
 #include <asm/pgtable.h>
 #include <asm/sync_bitops.h>
@@ -285,10 +287,9 @@ int gnttab_grant_foreign_access(domid_t domid, unsigned long frame,
 }
 EXPORT_SYMBOL_GPL(gnttab_grant_foreign_access);
 
-void gnttab_update_subpage_entry_v2(grant_ref_t ref, domid_t domid,
-				    unsigned long frame, int flags,
-				    unsigned page_off,
-				    unsigned length)
+static void gnttab_update_subpage_entry_v2(grant_ref_t ref, domid_t domid,
+					   unsigned long frame, int flags,
+					   unsigned page_off, unsigned length)
 {
 	gnttab_shared.v2[ref].sub_page.frame = frame;
 	gnttab_shared.v2[ref].sub_page.page_off = page_off;
@@ -345,9 +346,9 @@ bool gnttab_subpage_grants_available(void)
 }
 EXPORT_SYMBOL_GPL(gnttab_subpage_grants_available);
 
-void gnttab_update_trans_entry_v2(grant_ref_t ref, domid_t domid,
-				  int flags, domid_t trans_domid,
-				  grant_ref_t trans_gref)
+static void gnttab_update_trans_entry_v2(grant_ref_t ref, domid_t domid,
+					 int flags, domid_t trans_domid,
+					 grant_ref_t trans_gref)
 {
 	gnttab_shared.v2[ref].transitive.trans_domid = trans_domid;
 	gnttab_shared.v2[ref].transitive.gref = trans_gref;
@@ -823,6 +824,52 @@ unsigned int gnttab_max_grant_frames(void)
 }
 EXPORT_SYMBOL_GPL(gnttab_max_grant_frames);
 
+/* Handling of paged out grant targets (GNTST_eagain) */
+#define MAX_DELAY 256
+static inline void
+gnttab_retry_eagain_gop(unsigned int cmd, void *gop, int16_t *status,
+						const char *func)
+{
+	unsigned delay = 1;
+
+	do {
+		BUG_ON(HYPERVISOR_grant_table_op(cmd, gop, 1));
+		if (*status == GNTST_eagain)
+			msleep(delay++);
+	} while ((*status == GNTST_eagain) && (delay < MAX_DELAY));
+
+	if (delay >= MAX_DELAY) {
+		printk(KERN_ERR "%s: %s eagain grant\n", func, current->comm);
+		*status = GNTST_bad_page;
+	}
+}
+
+void gnttab_batch_map(struct gnttab_map_grant_ref *batch, unsigned count)
+{
+	struct gnttab_map_grant_ref *op;
+
+	if (HYPERVISOR_grant_table_op(GNTTABOP_map_grant_ref, batch, count))
+		BUG();
+	for (op = batch; op < batch + count; op++)
+		if (op->status == GNTST_eagain)
+			gnttab_retry_eagain_gop(GNTTABOP_map_grant_ref, op,
+						&op->status, __func__);
+}
+EXPORT_SYMBOL_GPL(gnttab_batch_map);
+
+void gnttab_batch_copy(struct gnttab_copy *batch, unsigned count)
+{
+	struct gnttab_copy *op;
+
+	if (HYPERVISOR_grant_table_op(GNTTABOP_copy, batch, count))
+		BUG();
+	for (op = batch; op < batch + count; op++)
+		if (op->status == GNTST_eagain)
+			gnttab_retry_eagain_gop(GNTTABOP_copy, op,
+						&op->status, __func__);
+}
+EXPORT_SYMBOL_GPL(gnttab_batch_copy);
+
 int gnttab_map_refs(struct gnttab_map_grant_ref *map_ops,
 		    struct gnttab_map_grant_ref *kmap_ops,
 		    struct page **pages, unsigned int count)
@@ -836,6 +883,12 @@ int gnttab_map_refs(struct gnttab_map_grant_ref *map_ops,
 	if (ret)
 		return ret;
 
+	/* 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 (xen_feature(XENFEAT_auto_translated_physmap))
 		return ret;
 

drivers/xen/privcmd.c

@@ -76,7 +76,7 @@ static void free_page_list(struct list_head *pages)
  */
 static int gather_array(struct list_head *pagelist,
 			unsigned nelem, size_t size,
-			void __user *data)
+			const void __user *data)
 {
 	unsigned pageidx;
 	void *pagedata;
@@ -246,61 +246,117 @@ struct mmap_batch_state {
 	domid_t domain;
 	unsigned long va;
 	struct vm_area_struct *vma;
-	int err;
-
-	xen_pfn_t __user *user;
+	/* A tristate:
+	 *      0 for no errors
+	 *      1 if at least one error has happened (and no
+	 *          -ENOENT errors have happened)
+	 *      -ENOENT if at least 1 -ENOENT has happened.
+	 */
+	int global_error;
+	/* An array for individual errors */
+	int *err;
+
+	/* User-space mfn array to store errors in the second pass for V1. */
+	xen_pfn_t __user *user_mfn;
 };
 
 static int mmap_batch_fn(void *data, void *state)
 {
 	xen_pfn_t *mfnp = data;
 	struct mmap_batch_state *st = state;
+	int ret;
+
+	ret = xen_remap_domain_mfn_range(st->vma, st->va & PAGE_MASK, *mfnp, 1,
+					 st->vma->vm_page_prot, st->domain);
 
-	if (xen_remap_domain_mfn_range(st->vma, st->va & PAGE_MASK, *mfnp, 1,
-				       st->vma->vm_page_prot, st->domain) < 0) {
-		*mfnp |= 0xf0000000U;
-		st->err++;
+	/* Store error code for second pass. */
+	*(st->err++) = ret;
+
+	/* And see if it affects the global_error. */
+	if (ret < 0) {
+		if (ret == -ENOENT)
+			st->global_error = -ENOENT;
+		else {
+			/* Record that at least one error has happened. */
+			if (st->global_error == 0)
+				st->global_error = 1;
+		}
 	}
 	st->va += PAGE_SIZE;
 
 	return 0;
 }
 
-static int mmap_return_errors(void *data, void *state)
+static int mmap_return_errors_v1(void *data, void *state)
 {
 	xen_pfn_t *mfnp = data;
 	struct mmap_batch_state *st = state;
-
-	return put_user(*mfnp, st->user++);
+	int err = *(st->err++);
+
+	/*
+	 * V1 encodes the error codes in the 32bit top nibble of the
+	 * mfn (with its known limitations vis-a-vis 64 bit callers).
+	 */
+	*mfnp |= (err == -ENOENT) ?
+				PRIVCMD_MMAPBATCH_PAGED_ERROR :
+				PRIVCMD_MMAPBATCH_MFN_ERROR;
+	return __put_user(*mfnp, st->user_mfn++);
 }
 
 static struct vm_operations_struct privcmd_vm_ops;
 
-static long privcmd_ioctl_mmap_batch(void __user *udata)
+static long privcmd_ioctl_mmap_batch(void __user *udata, int version)
 {
 	int ret;
-	struct privcmd_mmapbatch m;
+	struct privcmd_mmapbatch_v2 m;
 	struct mm_struct *mm = current->mm;
 	struct vm_area_struct *vma;
 	unsigned long nr_pages;
 	LIST_HEAD(pagelist);
+	int *err_array = NULL;
 	struct mmap_batch_state state;
 
 	if (!xen_initial_domain())
 		return -EPERM;
 
-	if (copy_from_user(&m, udata, sizeof(m)))
-		return -EFAULT;
+	switch (version) {
+	case 1:
+		if (copy_from_user(&m, udata, sizeof(struct privcmd_mmapbatch)))
+			return -EFAULT;
+		/* Returns per-frame error in m.arr. */
+		m.err = NULL;
+		if (!access_ok(VERIFY_WRITE, m.arr, m.num * sizeof(*m.arr)))
+			return -EFAULT;
+		break;
+	case 2:
+		if (copy_from_user(&m, udata, sizeof(struct privcmd_mmapbatch_v2)))
+			return -EFAULT;
+		/* Returns per-frame error code in m.err. */
+		if (!access_ok(VERIFY_WRITE, m.err, m.num * (sizeof(*m.err))))
+			return -EFAULT;
+		break;
+	default:
+		return -EINVAL;
+	}
 
 	nr_pages = m.num;
 	if ((m.num <= 0) || (nr_pages > (LONG_MAX >> PAGE_SHIFT)))
 		return -EINVAL;
 
-	ret = gather_array(&pagelist, m.num, sizeof(xen_pfn_t),
-			   m.arr);
+	ret = gather_array(&pagelist, m.num, sizeof(xen_pfn_t), m.arr);
 
-	if (ret || list_empty(&pagelist))
+	if (ret)
 		goto out;
+	if (list_empty(&pagelist)) {
+		ret = -EINVAL;
+		goto out;
+	}
+
+	err_array = kcalloc(m.num, sizeof(int), GFP_KERNEL);
+	if (err_array == NULL) {
+		ret = -ENOMEM;
+		goto out;
+	}
 
 	down_write(&mm->mmap_sem);
 
@@ -315,24 +371,37 @@ static long privcmd_ioctl_mmap_batch(void __user *udata)
 		goto out;
 	}
 
-	state.domain = m.dom;
-	state.vma = vma;
-	state.va = m.addr;
-	state.err = 0;
+	state.domain        = m.dom;
+	state.vma           = vma;
+	state.va            = m.addr;
+	state.global_error  = 0;
+	state.err           = err_array;
 
-	ret = traverse_pages(m.num, sizeof(xen_pfn_t),
-			     &pagelist, mmap_batch_fn, &state);
+	/* mmap_batch_fn guarantees ret == 0 */
+	BUG_ON(traverse_pages(m.num, sizeof(xen_pfn_t),
+			     &pagelist, mmap_batch_fn, &state));
 
 	up_write(&mm->mmap_sem);
 
-	if (state.err > 0) {
-		state.user = m.arr;
+	if (state.global_error && (version == 1)) {
+		/* Write back errors in second pass. */
+		state.user_mfn = (xen_pfn_t *)m.arr;
+		state.err      = err_array;
 		ret = traverse_pages(m.num, sizeof(xen_pfn_t),
-			       &pagelist,
-			       mmap_return_errors, &state);
+				     &pagelist, mmap_return_errors_v1, &state);
+	} else if (version == 2) {
+		ret = __copy_to_user(m.err, err_array, m.num * sizeof(int));
+		if (ret)
+			ret = -EFAULT;
 	}
 
+	/* If we have not had any EFAULT-like global errors then set the global
+	 * error to -ENOENT if necessary. */
+	if ((ret == 0) && (state.global_error == -ENOENT))
+		ret = -ENOENT;
+
 out:
+	kfree(err_array);
 	free_page_list(&pagelist);
 
 	return ret;
@@ -354,7 +423,11 @@ static long privcmd_ioctl(struct file *file,
 		break;
 
 	case IOCTL_PRIVCMD_MMAPBATCH:
-		ret = privcmd_ioctl_mmap_batch(udata);
+		ret = privcmd_ioctl_mmap_batch(udata, 1);
+		break;
+
+	case IOCTL_PRIVCMD_MMAPBATCH_V2:
+		ret = privcmd_ioctl_mmap_batch(udata, 2);
 		break;
 
 	default:
@@ -380,10 +453,6 @@ static struct vm_operations_struct privcmd_vm_ops = {
 
 static int privcmd_mmap(struct file *file, struct vm_area_struct *vma)
 {
-	/* Unsupported for auto-translate guests. */
-	if (xen_feature(XENFEAT_auto_translated_physmap))
-		return -ENOSYS;
-
 	/* DONTCOPY is essential for Xen because copy_page_range doesn't know
 	 * how to recreate these mappings */
 	vma->vm_flags |= VM_RESERVED | VM_IO | VM_DONTCOPY | VM_PFNMAP;

drivers/xen/swiotlb-xen.c

@@ -52,7 +52,7 @@ static unsigned long xen_io_tlb_nslabs;
  * Quick lookup value of the bus address of the IOTLB.
  */
 
-u64 start_dma_addr;
+static u64 start_dma_addr;
 
 static dma_addr_t xen_phys_to_bus(phys_addr_t paddr)
 {
@@ -144,31 +144,72 @@ xen_swiotlb_fixup(void *buf, size_t size, unsigned long nslabs)
 	} while (i < nslabs);
 	return 0;
 }
+static unsigned long xen_set_nslabs(unsigned long nr_tbl)
+{
+	if (!nr_tbl) {
+		xen_io_tlb_nslabs = (64 * 1024 * 1024 >> IO_TLB_SHIFT);
+		xen_io_tlb_nslabs = ALIGN(xen_io_tlb_nslabs, IO_TLB_SEGSIZE);
+	} else
+		xen_io_tlb_nslabs = nr_tbl;
 
-void __init xen_swiotlb_init(int verbose)
+	return xen_io_tlb_nslabs << IO_TLB_SHIFT;
+}
+
+enum xen_swiotlb_err {
+	XEN_SWIOTLB_UNKNOWN = 0,
+	XEN_SWIOTLB_ENOMEM,
+	XEN_SWIOTLB_EFIXUP
+};
+
+static const char *xen_swiotlb_error(enum xen_swiotlb_err err)
+{
+	switch (err) {
+	case XEN_SWIOTLB_ENOMEM:
+		return "Cannot allocate Xen-SWIOTLB buffer\n";
+	case XEN_SWIOTLB_EFIXUP:
+		return "Failed to get contiguous memory for DMA from Xen!\n"\
+		    "You either: don't have the permissions, do not have"\
+		    " enough free memory under 4GB, or the hypervisor memory"\
+		    " is too fragmented!";
+	default:
+		break;
+	}
+	return "";
+}
+int __ref xen_swiotlb_init(int verbose, bool early)
 {
-	unsigned long bytes;
+	unsigned long bytes, order;
 	int rc = -ENOMEM;
-	unsigned long nr_tbl;
-	char *m = NULL;
+	enum xen_swiotlb_err m_ret = XEN_SWIOTLB_UNKNOWN;
 	unsigned int repeat = 3;
 
-	nr_tbl = swiotlb_nr_tbl();
-	if (nr_tbl)
-		xen_io_tlb_nslabs = nr_tbl;
-	else {
-		xen_io_tlb_nslabs = (64 * 1024 * 1024 >> IO_TLB_SHIFT);
-		xen_io_tlb_nslabs = ALIGN(xen_io_tlb_nslabs, IO_TLB_SEGSIZE);
-	}
+	xen_io_tlb_nslabs = swiotlb_nr_tbl();
 retry:
-	bytes = xen_io_tlb_nslabs << IO_TLB_SHIFT;
-
+	bytes = xen_set_nslabs(xen_io_tlb_nslabs);
+	order = get_order(xen_io_tlb_nslabs << IO_TLB_SHIFT);
 	/*
 	 * Get IO TLB memory from any location.
 	 */
-	xen_io_tlb_start = alloc_bootmem_pages(PAGE_ALIGN(bytes));
+	if (early)
+		xen_io_tlb_start = alloc_bootmem_pages(PAGE_ALIGN(bytes));
+	else {
+#define SLABS_PER_PAGE (1 << (PAGE_SHIFT - IO_TLB_SHIFT))
+#define IO_TLB_MIN_SLABS ((1<<20) >> IO_TLB_SHIFT)
+		while ((SLABS_PER_PAGE << order) > IO_TLB_MIN_SLABS) {
+			xen_io_tlb_start = (void *)__get_free_pages(__GFP_NOWARN, order);
+			if (xen_io_tlb_start)
+				break;
+			order--;
+		}
+		if (order != get_order(bytes)) {
+			pr_warn("Warning: only able to allocate %ld MB "
+				"for software IO TLB\n", (PAGE_SIZE << order) >> 20);
+			xen_io_tlb_nslabs = SLABS_PER_PAGE << order;
+			bytes = xen_io_tlb_nslabs << IO_TLB_SHIFT;
+		}
+	}
 	if (!xen_io_tlb_start) {
-		m = "Cannot allocate Xen-SWIOTLB buffer!\n";
+		m_ret = XEN_SWIOTLB_ENOMEM;
 		goto error;
 	}
 	xen_io_tlb_end = xen_io_tlb_start + bytes;
@@ -179,17 +220,22 @@ void __init xen_swiotlb_init(int verbose)
 			       bytes,
 			       xen_io_tlb_nslabs);
 	if (rc) {
-		free_bootmem(__pa(xen_io_tlb_start), PAGE_ALIGN(bytes));
-		m = "Failed to get contiguous memory for DMA from Xen!\n"\
-		    "You either: don't have the permissions, do not have"\
-		    " enough free memory under 4GB, or the hypervisor memory"\
-		    "is too fragmented!";
+		if (early)
+			free_bootmem(__pa(xen_io_tlb_start), PAGE_ALIGN(bytes));
+		else {
+			free_pages((unsigned long)xen_io_tlb_start, order);
+			xen_io_tlb_start = NULL;
+		}
+		m_ret = XEN_SWIOTLB_EFIXUP;
 		goto error;
 	}
 	start_dma_addr = xen_virt_to_bus(xen_io_tlb_start);
-	swiotlb_init_with_tbl(xen_io_tlb_start, xen_io_tlb_nslabs, verbose);
-
-	return;
+	if (early) {
+		swiotlb_init_with_tbl(xen_io_tlb_start, xen_io_tlb_nslabs, verbose);
+		rc = 0;
+	} else
+		rc = swiotlb_late_init_with_tbl(xen_io_tlb_start, xen_io_tlb_nslabs);
+	return rc;
 error:
 	if (repeat--) {
 		xen_io_tlb_nslabs = max(1024UL, /* Min is 2MB */
@@ -198,10 +244,13 @@ void __init xen_swiotlb_init(int verbose)
 		      (xen_io_tlb_nslabs << IO_TLB_SHIFT) >> 20);
 		goto retry;
 	}
-	xen_raw_printk("%s (rc:%d)", m, rc);
-	panic("%s (rc:%d)", m, rc);
+	pr_err("%s (rc:%d)", xen_swiotlb_error(m_ret), rc);
+	if (early)
+		panic("%s (rc:%d)", xen_swiotlb_error(m_ret), rc);
+	else
+		free_pages((unsigned long)xen_io_tlb_start, order);
+	return rc;
 }
-
 void *
 xen_swiotlb_alloc_coherent(struct device *hwdev, size_t size,
 			   dma_addr_t *dma_handle, gfp_t flags,
@@ -466,14 +515,6 @@ xen_swiotlb_map_sg_attrs(struct device *hwdev, struct scatterlist *sgl,
 }
 EXPORT_SYMBOL_GPL(xen_swiotlb_map_sg_attrs);
 
-int
-xen_swiotlb_map_sg(struct device *hwdev, struct scatterlist *sgl, int nelems,
-		   enum dma_data_direction dir)
-{
-	return xen_swiotlb_map_sg_attrs(hwdev, sgl, nelems, dir, NULL);
-}
-EXPORT_SYMBOL_GPL(xen_swiotlb_map_sg);
-
 /*
  * Unmap a set of streaming mode DMA translations.  Again, cpu read rules
  * concerning calls here are the same as for swiotlb_unmap_page() above.
@@ -494,14 +535,6 @@ xen_swiotlb_unmap_sg_attrs(struct device *hwdev, struct scatterlist *sgl,
 }
 EXPORT_SYMBOL_GPL(xen_swiotlb_unmap_sg_attrs);
 
-void
-xen_swiotlb_unmap_sg(struct device *hwdev, struct scatterlist *sgl, int nelems,
-		     enum dma_data_direction dir)
-{
-	return xen_swiotlb_unmap_sg_attrs(hwdev, sgl, nelems, dir, NULL);
-}
-EXPORT_SYMBOL_GPL(xen_swiotlb_unmap_sg);
-
 /*
  * Make physical memory consistent for a set of streaming mode DMA translations
  * after a transfer.

drivers/xen/sys-hypervisor.c

@@ -114,7 +114,7 @@ static void xen_sysfs_version_destroy(void)
 
 /* UUID */
 
-static ssize_t uuid_show(struct hyp_sysfs_attr *attr, char *buffer)
+static ssize_t uuid_show_fallback(struct hyp_sysfs_attr *attr, char *buffer)
 {
 	char *vm, *val;
 	int ret;
@@ -135,6 +135,17 @@ static ssize_t uuid_show(struct hyp_sysfs_attr *attr, char *buffer)
 	return ret;
 }
 
+static ssize_t uuid_show(struct hyp_sysfs_attr *attr, char *buffer)
+{
+	xen_domain_handle_t uuid;
+	int ret;
+	ret = HYPERVISOR_xen_version(XENVER_guest_handle, uuid);
+	if (ret)
+		return uuid_show_fallback(attr, buffer);
+	ret = sprintf(buffer, "%pU\n", uuid);
+	return ret;
+}
+
 HYPERVISOR_ATTR_RO(uuid);
 
 static int __init xen_sysfs_uuid_init(void)

drivers/xen/tmem.c

@@ -21,6 +21,7 @@
 #include <asm/xen/hypercall.h>
 #include <asm/xen/page.h>
 #include <asm/xen/hypervisor.h>
+#include <xen/tmem.h>
 
 #define TMEM_CONTROL               0
 #define TMEM_NEW_POOL              1

drivers/xen/xen-pciback/pci_stub.c

@@ -362,6 +362,7 @@ static int __devinit pcistub_init_device(struct pci_dev *dev)
 	else {
 		dev_dbg(&dev->dev, "reseting (FLR, D3, etc) the device\n");
 		__pci_reset_function_locked(dev);
+		pci_restore_state(dev);
 	}
 	/* Now disable the device (this also ensures some private device
 	 * data is setup before we export)
@@ -681,14 +682,14 @@ static pci_ers_result_t xen_pcibk_slot_reset(struct pci_dev *dev)
 		dev_err(&dev->dev, DRV_NAME " device is not connected or owned"
 			" by HVM, kill it\n");
 		kill_domain_by_device(psdev);
-		goto release;
+		goto end;
 	}
 
 	if (!test_bit(_XEN_PCIB_AERHANDLER,
 		(unsigned long *)&psdev->pdev->sh_info->flags)) {
 		dev_err(&dev->dev,
 			"guest with no AER driver should have been killed\n");
-		goto release;
+		goto end;
 	}
 	result = common_process(psdev, 1, XEN_PCI_OP_aer_slotreset, result);
 
@@ -698,9 +699,9 @@ static pci_ers_result_t xen_pcibk_slot_reset(struct pci_dev *dev)
 			"No AER slot_reset service or disconnected!\n");
 		kill_domain_by_device(psdev);
 	}
-release:
-	pcistub_device_put(psdev);
 end:
+	if (psdev)
+		pcistub_device_put(psdev);
 	up_write(&pcistub_sem);
 	return result;
 
@@ -739,14 +740,14 @@ static pci_ers_result_t xen_pcibk_mmio_enabled(struct pci_dev *dev)
 		dev_err(&dev->dev, DRV_NAME " device is not connected or owned"
 			" by HVM, kill it\n");
 		kill_domain_by_device(psdev);
-		goto release;
+		goto end;
 	}
 
 	if (!test_bit(_XEN_PCIB_AERHANDLER,
 		(unsigned long *)&psdev->pdev->sh_info->flags)) {
 		dev_err(&dev->dev,
 			"guest with no AER driver should have been killed\n");
-		goto release;
+		goto end;
 	}
 	result = common_process(psdev, 1, XEN_PCI_OP_aer_mmio, result);
 
@@ -756,9 +757,9 @@ static pci_ers_result_t xen_pcibk_mmio_enabled(struct pci_dev *dev)
 			"No AER mmio_enabled service or disconnected!\n");
 		kill_domain_by_device(psdev);
 	}
-release:
-	pcistub_device_put(psdev);
 end:
+	if (psdev)
+		pcistub_device_put(psdev);
 	up_write(&pcistub_sem);
 	return result;
 }
@@ -797,7 +798,7 @@ static pci_ers_result_t xen_pcibk_error_detected(struct pci_dev *dev,
 		dev_err(&dev->dev, DRV_NAME " device is not connected or owned"
 			" by HVM, kill it\n");
 		kill_domain_by_device(psdev);
-		goto release;
+		goto end;
 	}
 
 	/*Guest owns the device yet no aer handler regiested, kill guest*/
@@ -805,7 +806,7 @@ static pci_ers_result_t xen_pcibk_error_detected(struct pci_dev *dev,
 		(unsigned long *)&psdev->pdev->sh_info->flags)) {
 		dev_dbg(&dev->dev, "guest may have no aer driver, kill it\n");
 		kill_domain_by_device(psdev);
-		goto release;
+		goto end;
 	}
 	result = common_process(psdev, error, XEN_PCI_OP_aer_detected, result);
 
@@ -815,9 +816,9 @@ static pci_ers_result_t xen_pcibk_error_detected(struct pci_dev *dev,
 			"No AER error_detected service or disconnected!\n");
 		kill_domain_by_device(psdev);
 	}
-release:
-	pcistub_device_put(psdev);
 end:
+	if (psdev)
+		pcistub_device_put(psdev);
 	up_write(&pcistub_sem);
 	return result;
 }
@@ -851,7 +852,7 @@ static void xen_pcibk_error_resume(struct pci_dev *dev)
 		dev_err(&dev->dev, DRV_NAME " device is not connected or owned"
 			" by HVM, kill it\n");
 		kill_domain_by_device(psdev);
-		goto release;
+		goto end;
 	}
 
 	if (!test_bit(_XEN_PCIB_AERHANDLER,
@@ -859,13 +860,13 @@ static void xen_pcibk_error_resume(struct pci_dev *dev)
 		dev_err(&dev->dev,
 			"guest with no AER driver should have been killed\n");
 		kill_domain_by_device(psdev);
-		goto release;
+		goto end;
 	}
 	common_process(psdev, 1, XEN_PCI_OP_aer_resume,
 		       PCI_ERS_RESULT_RECOVERED);
-release:
-	pcistub_device_put(psdev);
 end:
+	if (psdev)
+		pcistub_device_put(psdev);
 	up_write(&pcistub_sem);
 	return;
 }
@@ -897,17 +898,41 @@ static inline int str_to_slot(const char *buf, int *domain, int *bus,
 			      int *slot, int *func)
 {
 	int err;
+	char wc = '*';
 
 	err = sscanf(buf, " %x:%x:%x.%x", domain, bus, slot, func);
-	if (err == 4)
+	switch (err) {
+	case 3:
+		*func = -1;
+		err = sscanf(buf, " %x:%x:%x.%c", domain, bus, slot, &wc);
+		break;
+	case 2:
+		*slot = *func = -1;
+		err = sscanf(buf, " %x:%x:*.%c", domain, bus, &wc);
+		if (err >= 2)
+			++err;
+		break;
+	}
+	if (err == 4 && wc == '*')
 		return 0;
 	else if (err < 0)
 		return -EINVAL;
 
 	/* try again without domain */
 	*domain = 0;
+	wc = '*';
 	err = sscanf(buf, " %x:%x.%x", bus, slot, func);
-	if (err == 3)
+	switch (err) {
+	case 2:
+		*func = -1;
+		err = sscanf(buf, " %x:%x.%c", bus, slot, &wc);
+		break;
+	case 1:
+		*slot = *func = -1;
+		err = sscanf(buf, " %x:*.%c", bus, &wc) + 1;
+		break;
+	}
+	if (err == 3 && wc == '*')
 		return 0;
 
 	return -EINVAL;
@@ -930,6 +955,19 @@ static int pcistub_device_id_add(int domain, int bus, int slot, int func)
 {
 	struct pcistub_device_id *pci_dev_id;
 	unsigned long flags;
+	int rc = 0;
+
+	if (slot < 0) {
+		for (slot = 0; !rc && slot < 32; ++slot)
+			rc = pcistub_device_id_add(domain, bus, slot, func);
+		return rc;
+	}
+
+	if (func < 0) {
+		for (func = 0; !rc && func < 8; ++func)
+			rc = pcistub_device_id_add(domain, bus, slot, func);
+		return rc;
+	}
 
 	pci_dev_id = kmalloc(sizeof(*pci_dev_id), GFP_KERNEL);
 	if (!pci_dev_id)
@@ -952,15 +990,15 @@ static int pcistub_device_id_add(int domain, int bus, int slot, int func)
 static int pcistub_device_id_remove(int domain, int bus, int slot, int func)
 {
 	struct pcistub_device_id *pci_dev_id, *t;
-	int devfn = PCI_DEVFN(slot, func);
 	int err = -ENOENT;
 	unsigned long flags;
 
 	spin_lock_irqsave(&device_ids_lock, flags);
 	list_for_each_entry_safe(pci_dev_id, t, &pcistub_device_ids,
 				 slot_list) {
-		if (pci_dev_id->domain == domain
-		    && pci_dev_id->bus == bus && pci_dev_id->devfn == devfn) {
+		if (pci_dev_id->domain == domain && pci_dev_id->bus == bus
+		    && (slot < 0 || PCI_SLOT(pci_dev_id->devfn) == slot)
+		    && (func < 0 || PCI_FUNC(pci_dev_id->devfn) == func)) {
 			/* Don't break; here because it's possible the same
 			 * slot could be in the list more than once
 			 */
@@ -987,7 +1025,7 @@ static int pcistub_reg_add(int domain, int bus, int slot, int func, int reg,
 	struct config_field *field;
 
 	psdev = pcistub_device_find(domain, bus, slot, func);
-	if (!psdev || !psdev->dev) {
+	if (!psdev) {
 		err = -ENODEV;
 		goto out;
 	}
@@ -1011,6 +1049,8 @@ static int pcistub_reg_add(int domain, int bus, int slot, int func, int reg,
 	if (err)
 		kfree(field);
 out:
+	if (psdev)
+		pcistub_device_put(psdev);
 	return err;
 }
 
@@ -1115,10 +1155,9 @@ static ssize_t pcistub_irq_handler_switch(struct device_driver *drv,
 
 	err = str_to_slot(buf, &domain, &bus, &slot, &func);
 	if (err)
-		goto out;
+		return err;
 
 	psdev = pcistub_device_find(domain, bus, slot, func);
-
 	if (!psdev)
 		goto out;
 
@@ -1134,6 +1173,8 @@ static ssize_t pcistub_irq_handler_switch(struct device_driver *drv,
 	if (dev_data->isr_on)
 		dev_data->ack_intr = 1;
 out:
+	if (psdev)
+		pcistub_device_put(psdev);
 	if (!err)
 		err = count;
 	return err;
@@ -1216,15 +1257,16 @@ static ssize_t permissive_add(struct device_driver *drv, const char *buf,
 	err = str_to_slot(buf, &domain, &bus, &slot, &func);
 	if (err)
 		goto out;
+	if (slot < 0 || func < 0) {
+		err = -EINVAL;
+		goto out;
+	}
 	psdev = pcistub_device_find(domain, bus, slot, func);
 	if (!psdev) {
 		err = -ENODEV;
 		goto out;
 	}
-	if (!psdev->dev) {
-		err = -ENODEV;
-		goto release;
-	}
+
 	dev_data = pci_get_drvdata(psdev->dev);
 	/* the driver data for a device should never be null at this point */
 	if (!dev_data) {
@@ -1297,17 +1339,51 @@ static int __init pcistub_init(void)
 
 	if (pci_devs_to_hide && *pci_devs_to_hide) {
 		do {
+			char wc = '*';
+
 			parsed = 0;
 
 			err = sscanf(pci_devs_to_hide + pos,
 				     " (%x:%x:%x.%x) %n",
 				     &domain, &bus, &slot, &func, &parsed);
-			if (err != 4) {
+			switch (err) {
+			case 3:
+				func = -1;
+				err = sscanf(pci_devs_to_hide + pos,
+					     " (%x:%x:%x.%c) %n",
+					     &domain, &bus, &slot, &wc,
+					     &parsed);
+				break;
+			case 2:
+				slot = func = -1;
+				err = sscanf(pci_devs_to_hide + pos,
+					     " (%x:%x:*.%c) %n",
+					     &domain, &bus, &wc, &parsed) + 1;
+				break;
+			}
+
+			if (err != 4 || wc != '*') {
 				domain = 0;
+				wc = '*';
 				err = sscanf(pci_devs_to_hide + pos,
 					     " (%x:%x.%x) %n",
 					     &bus, &slot, &func, &parsed);
-				if (err != 3)
+				switch (err) {
+				case 2:
+					func = -1;
+					err = sscanf(pci_devs_to_hide + pos,
+						     " (%x:%x.%c) %n",
+						     &bus, &slot, &wc,
+						     &parsed);
+					break;
+				case 1:
+					slot = func = -1;
+					err = sscanf(pci_devs_to_hide + pos,
+						     " (%x:*.%c) %n",
+						     &bus, &wc, &parsed) + 1;
+					break;
+				}
+				if (err != 3 || wc != '*')
 					goto parse_error;
 			}
 

drivers/xen/xenbus/xenbus_client.c

@@ -490,8 +490,7 @@ static int xenbus_map_ring_valloc_pv(struct xenbus_device *dev,
 
 	op.host_addr = arbitrary_virt_to_machine(pte).maddr;
 
-	if (HYPERVISOR_grant_table_op(GNTTABOP_map_grant_ref, &op, 1))
-		BUG();
+	gnttab_batch_map(&op, 1);
 
 	if (op.status != GNTST_okay) {
 		free_vm_area(area);
@@ -572,8 +571,7 @@ int xenbus_map_ring(struct xenbus_device *dev, int gnt_ref,
 	gnttab_set_map_op(&op, (unsigned long)vaddr, GNTMAP_host_map, gnt_ref,
 			  dev->otherend_id);
 
-	if (HYPERVISOR_grant_table_op(GNTTABOP_map_grant_ref, &op, 1))
-		BUG();
+	gnttab_batch_map(&op, 1);
 
 	if (op.status != GNTST_okay) {
 		xenbus_dev_fatal(dev, op.status,

drivers/xen/xenbus/xenbus_comms.c

@@ -224,7 +224,7 @@ int xb_init_comms(void)
 		int err;
 		err = bind_evtchn_to_irqhandler(xen_store_evtchn, wake_waiting,
 						0, "xenbus", &xb_waitq);
-		if (err <= 0) {
+		if (err < 0) {
 			printk(KERN_ERR "XENBUS request irq failed %i\n", err);
 			return err;
 		}

drivers/xen/xenbus/xenbus_dev_backend.c

@@ -107,7 +107,7 @@ static int xenbus_backend_mmap(struct file *file, struct vm_area_struct *vma)
 	return 0;
 }
 
-const struct file_operations xenbus_backend_fops = {
+static const struct file_operations xenbus_backend_fops = {
 	.open = xenbus_backend_open,
 	.mmap = xenbus_backend_mmap,
 	.unlocked_ioctl = xenbus_backend_ioctl,

drivers/xen/xenbus/xenbus_probe.c

@@ -324,8 +324,8 @@ static int cmp_dev(struct device *dev, void *data)
 	return 0;
 }
 
-struct xenbus_device *xenbus_device_find(const char *nodename,
-					 struct bus_type *bus)
+static struct xenbus_device *xenbus_device_find(const char *nodename,
+						struct bus_type *bus)
 {
 	struct xb_find_info info = { .dev = NULL, .nodename = nodename };
 
@@ -719,17 +719,47 @@ static int __init xenstored_local_init(void)
 	return err;
 }
 
+enum xenstore_init {
+	UNKNOWN,
+	PV,
+	HVM,
+	LOCAL,
+};
 static int __init xenbus_init(void)
 {
 	int err = 0;
+	enum xenstore_init usage = UNKNOWN;
+	uint64_t v = 0;
 
 	if (!xen_domain())
 		return -ENODEV;
 
 	xenbus_ring_ops_init();
 
-	if (xen_hvm_domain()) {
-		uint64_t v = 0;
+	if (xen_pv_domain())
+		usage = PV;
+	if (xen_hvm_domain())
+		usage = HVM;
+	if (xen_hvm_domain() && xen_initial_domain())
+		usage = LOCAL;
+	if (xen_pv_domain() && !xen_start_info->store_evtchn)
+		usage = LOCAL;
+	if (xen_pv_domain() && xen_start_info->store_evtchn)
+		xenstored_ready = 1;
+
+	switch (usage) {
+	case LOCAL:
+		err = xenstored_local_init();
+		if (err)
+			goto out_error;
+		xen_store_interface = mfn_to_virt(xen_store_mfn);
+		break;
+	case PV:
+		xen_store_evtchn = xen_start_info->store_evtchn;
+		xen_store_mfn = xen_start_info->store_mfn;
+		xen_store_interface = mfn_to_virt(xen_store_mfn);
+		break;
+	case HVM:
 		err = hvm_get_parameter(HVM_PARAM_STORE_EVTCHN, &v);
 		if (err)
 			goto out_error;
@@ -738,18 +768,12 @@ static int __init xenbus_init(void)
 		if (err)
 			goto out_error;
 		xen_store_mfn = (unsigned long)v;
-		xen_store_interface = ioremap(xen_store_mfn << PAGE_SHIFT, PAGE_SIZE);
-	} else {
-		xen_store_evtchn = xen_start_info->store_evtchn;
-		xen_store_mfn = xen_start_info->store_mfn;
-		if (xen_store_evtchn)
-			xenstored_ready = 1;
-		else {
-			err = xenstored_local_init();
-			if (err)
-				goto out_error;
-		}
-		xen_store_interface = mfn_to_virt(xen_store_mfn);
+		xen_store_interface =
+			ioremap(xen_store_mfn << PAGE_SHIFT, PAGE_SIZE);
+		break;
+	default:
+		pr_warn("Xenstore state unknown\n");
+		break;
 	}
 
 	/* Initialize the interface to xenstore. */

drivers/xen/xenbus/xenbus_probe_frontend.c

@@ -21,6 +21,7 @@
 #include <xen/xenbus.h>
 #include <xen/events.h>
 #include <xen/page.h>
+#include <xen/xen.h>
 
 #include <xen/platform_pci.h>
 

drivers/xen/xenbus/xenbus_xs.c

@@ -44,6 +44,7 @@
 #include <linux/rwsem.h>
 #include <linux/module.h>
 #include <linux/mutex.h>
+#include <asm/xen/hypervisor.h>
 #include <xen/xenbus.h>
 #include <xen/xen.h>
 #include "xenbus_comms.h"
@@ -622,7 +623,7 @@ static void xs_reset_watches(void)
 {
 	int err, supported = 0;
 
-	if (!xen_hvm_domain())
+	if (!xen_hvm_domain() || xen_initial_domain())
 		return;
 
 	err = xenbus_scanf(XBT_NIL, "control",

include/linux/swiotlb.h

@@ -25,6 +25,7 @@ extern int swiotlb_force;
 extern void swiotlb_init(int verbose);
 extern void swiotlb_init_with_tbl(char *tlb, unsigned long nslabs, int verbose);
 extern unsigned long swiotlb_nr_tbl(void);
+extern int swiotlb_late_init_with_tbl(char *tlb, unsigned long nslabs);
 
 /*
  * Enumeration for sync targets

include/xen/grant_table.h

@@ -190,4 +190,16 @@ int gnttab_unmap_refs(struct gnttab_unmap_grant_ref *unmap_ops,
 		      struct gnttab_map_grant_ref *kunmap_ops,
 		      struct page **pages, unsigned int count);
 
+/* Perform a batch of grant map/copy operations. Retry every batch slot
+ * for which the hypervisor returns GNTST_eagain. This is typically due
+ * to paged out target frames.
+ *
+ * Will retry for 1, 2, ... 255 ms, i.e. 256 times during 32 seconds.
+ *
+ * Return value in each iand every status field of the batch guaranteed
+ * to not be GNTST_eagain.
+ */
+void gnttab_batch_map(struct gnttab_map_grant_ref *batch, unsigned count);
+void gnttab_batch_copy(struct gnttab_copy *batch, unsigned count);
+
 #endif /* __ASM_GNTTAB_H__ */

include/xen/interface/grant_table.h

@@ -338,7 +338,7 @@ DEFINE_GUEST_HANDLE_STRUCT(gnttab_dump_table);
 #define GNTTABOP_transfer                4
 struct gnttab_transfer {
     /* IN parameters. */
-    unsigned long mfn;
+    xen_pfn_t mfn;
     domid_t       domid;
     grant_ref_t   ref;
     /* OUT parameters. */
@@ -375,7 +375,7 @@ struct gnttab_copy {
 	struct {
 		union {
 			grant_ref_t ref;
-			unsigned long   gmfn;
+			xen_pfn_t   gmfn;
 		} u;
 		domid_t  domid;
 		uint16_t offset;
@@ -519,7 +519,9 @@ DEFINE_GUEST_HANDLE_STRUCT(gnttab_get_version);
 #define GNTST_no_device_space  (-7) /* Out of space in I/O MMU.              */
 #define GNTST_permission_denied (-8) /* Not enough privilege for operation.  */
 #define GNTST_bad_page         (-9) /* Specified page was invalid for op.    */
-#define GNTST_bad_copy_arg    (-10) /* copy arguments cross page boundary */
+#define GNTST_bad_copy_arg    (-10) /* copy arguments cross page boundary.   */
+#define GNTST_address_too_big (-11) /* transfer page address too large.      */
+#define GNTST_eagain          (-12) /* Operation not done; try again.        */
 
 #define GNTTABOP_error_msgs {                   \
     "okay",                                     \
@@ -532,7 +534,9 @@ DEFINE_GUEST_HANDLE_STRUCT(gnttab_get_version);
     "no spare translation slot in the I/O MMU", \
     "permission denied",                        \
     "bad page",                                 \
-    "copy arguments cross page boundary"        \
+    "copy arguments cross page boundary",       \
+    "page address size too large",              \
+    "operation not done; try again"             \
 }
 
 #endif /* __XEN_PUBLIC_GRANT_TABLE_H__ */

include/xen/interface/memory.h

@@ -31,7 +31,7 @@ struct xen_memory_reservation {
      *   OUT: GMFN bases of extents that were allocated
      *   (NB. This command also updates the mach_to_phys translation table)
      */
-    GUEST_HANDLE(ulong) extent_start;
+    GUEST_HANDLE(xen_pfn_t) extent_start;
 
     /* Number of extents, and size/alignment of each (2^extent_order pages). */
     unsigned long  nr_extents;
@@ -130,7 +130,7 @@ struct xen_machphys_mfn_list {
      * any large discontiguities in the machine address space, 2MB gaps in
      * the machphys table will be represented by an MFN base of zero.
      */
-    GUEST_HANDLE(ulong) extent_start;
+    GUEST_HANDLE(xen_pfn_t) extent_start;
 
     /*
      * Number of extents written to the above array. This will be smaller
@@ -163,6 +163,9 @@ struct xen_add_to_physmap {
     /* Which domain to change the mapping for. */
     domid_t domid;
 
+    /* Number of pages to go through for gmfn_range */
+    uint16_t    size;
+
     /* Source mapping space. */
 #define XENMAPSPACE_shared_info 0 /* shared info page */
 #define XENMAPSPACE_grant_table 1 /* grant table page */
@@ -172,7 +175,7 @@ struct xen_add_to_physmap {
     unsigned long idx;
 
     /* GPFN where the source mapping page should appear. */
-    unsigned long gpfn;
+    xen_pfn_t gpfn;
 };
 DEFINE_GUEST_HANDLE_STRUCT(xen_add_to_physmap);
 

include/xen/interface/platform.h

@@ -54,7 +54,7 @@ DEFINE_GUEST_HANDLE_STRUCT(xenpf_settime_t);
 #define XENPF_add_memtype         31
 struct xenpf_add_memtype {
 	/* IN variables. */
-	unsigned long mfn;
+	xen_pfn_t mfn;
 	uint64_t nr_mfns;
 	uint32_t type;
 	/* OUT variables. */
@@ -84,7 +84,7 @@ struct xenpf_read_memtype {
 	/* IN variables. */
 	uint32_t reg;
 	/* OUT variables. */
-	unsigned long mfn;
+	xen_pfn_t mfn;
 	uint64_t nr_mfns;
 	uint32_t type;
 };
@@ -112,6 +112,7 @@ DEFINE_GUEST_HANDLE_STRUCT(xenpf_platform_quirk_t);
 #define XEN_FW_DISK_INFO          1 /* from int 13 AH=08/41/48 */
 #define XEN_FW_DISK_MBR_SIGNATURE 2 /* from MBR offset 0x1b8 */
 #define XEN_FW_VBEDDC_INFO        3 /* from int 10 AX=4f15 */
+#define XEN_FW_KBD_SHIFT_FLAGS    5 /* Int16, Fn02: Get keyboard shift flags. */
 struct xenpf_firmware_info {
 	/* IN variables. */
 	uint32_t type;
@@ -142,6 +143,8 @@ struct xenpf_firmware_info {
 			/* must refer to 128-byte buffer */
 			GUEST_HANDLE(uchar) edid;
 		} vbeddc_info; /* XEN_FW_VBEDDC_INFO */
+
+		uint8_t kbd_shift_flags; /* XEN_FW_KBD_SHIFT_FLAGS */
 	} u;
 };
 DEFINE_GUEST_HANDLE_STRUCT(xenpf_firmware_info_t);

include/xen/interface/version.h

@@ -60,4 +60,7 @@ struct xen_feature_info {
 /* arg == NULL; returns host memory page size. */
 #define XENVER_pagesize 7
 
+/* arg == xen_domain_handle_t. */
+#define XENVER_guest_handle 8
+
 #endif /* __XEN_PUBLIC_VERSION_H__ */

include/xen/interface/xen.h

@@ -10,7 +10,6 @@
 #define __XEN_PUBLIC_XEN_H__
 
 #include <asm/xen/interface.h>
-#include <asm/pvclock-abi.h>
 
 /*
  * XEN "SYSTEM CALLS" (a.k.a. HYPERCALLS).
@@ -190,7 +189,7 @@ struct mmuext_op {
 	unsigned int cmd;
 	union {
 		/* [UN]PIN_TABLE, NEW_BASEPTR, NEW_USER_BASEPTR */
-		unsigned long mfn;
+		xen_pfn_t mfn;
 		/* INVLPG_LOCAL, INVLPG_ALL, SET_LDT */
 		unsigned long linear_addr;
 	} arg1;
@@ -430,11 +429,11 @@ struct start_info {
 	unsigned long nr_pages;     /* Total pages allocated to this domain.  */
 	unsigned long shared_info;  /* MACHINE address of shared info struct. */
 	uint32_t flags;             /* SIF_xxx flags.                         */
-	unsigned long store_mfn;    /* MACHINE page number of shared page.    */
+	xen_pfn_t store_mfn;        /* MACHINE page number of shared page.    */
 	uint32_t store_evtchn;      /* Event channel for store communication. */
 	union {
 		struct {
-			unsigned long mfn;  /* MACHINE page number of console page.   */
+			xen_pfn_t mfn;      /* MACHINE page number of console page.   */
 			uint32_t  evtchn;   /* Event channel for console page.        */
 		} domU;
 		struct {
@@ -455,6 +454,7 @@ struct dom0_vga_console_info {
 	uint8_t video_type;
 #define XEN_VGATYPE_TEXT_MODE_3 0x03
 #define XEN_VGATYPE_VESA_LFB    0x23
+#define XEN_VGATYPE_EFI_LFB     0x70
 
 	union {
 		struct {

include/xen/privcmd.h

@@ -35,8 +35,7 @@
 
 #include <linux/types.h>
 #include <linux/compiler.h>
-
-typedef unsigned long xen_pfn_t;
+#include <xen/interface/xen.h>
 
 struct privcmd_hypercall {
 	__u64 op;
@@ -59,13 +58,33 @@ struct privcmd_mmapbatch {
 	int num;     /* number of pages to populate */
 	domid_t dom; /* target domain */
 	__u64 addr;  /* virtual address */
-	xen_pfn_t __user *arr; /* array of mfns - top nibble set on err */
+	xen_pfn_t __user *arr; /* array of mfns - or'd with
+				  PRIVCMD_MMAPBATCH_*_ERROR on err */
+};
+
+#define PRIVCMD_MMAPBATCH_MFN_ERROR     0xf0000000U
+#define PRIVCMD_MMAPBATCH_PAGED_ERROR   0x80000000U
+
+struct privcmd_mmapbatch_v2 {
+	unsigned int num; /* number of pages to populate */
+	domid_t dom;      /* target domain */
+	__u64 addr;       /* virtual address */
+	const xen_pfn_t __user *arr; /* array of mfns */
+	int __user *err;  /* array of error codes */
 };
 
 /*
  * @cmd: IOCTL_PRIVCMD_HYPERCALL
  * @arg: &privcmd_hypercall_t
  * Return: Value returned from execution of the specified hypercall.
+ *
+ * @cmd: IOCTL_PRIVCMD_MMAPBATCH_V2
+ * @arg: &struct privcmd_mmapbatch_v2
+ * Return: 0 on success (i.e., arg->err contains valid error codes for
+ * each frame).  On an error other than a failed frame remap, -1 is
+ * returned and errno is set to EINVAL, EFAULT etc.  As an exception,
+ * if the operation was otherwise successful but any frame failed with
+ * -ENOENT, then -1 is returned and errno is set to ENOENT.
  */
 #define IOCTL_PRIVCMD_HYPERCALL					\
 	_IOC(_IOC_NONE, 'P', 0, sizeof(struct privcmd_hypercall))
@@ -73,5 +92,7 @@ struct privcmd_mmapbatch {
 	_IOC(_IOC_NONE, 'P', 2, sizeof(struct privcmd_mmap))
 #define IOCTL_PRIVCMD_MMAPBATCH					\
 	_IOC(_IOC_NONE, 'P', 3, sizeof(struct privcmd_mmapbatch))
+#define IOCTL_PRIVCMD_MMAPBATCH_V2				\
+	_IOC(_IOC_NONE, 'P', 4, sizeof(struct privcmd_mmapbatch_v2))
 
 #endif /* __LINUX_PUBLIC_PRIVCMD_H__ */

include/xen/swiotlb-xen.h

@@ -3,7 +3,7 @@
 
 #include <linux/swiotlb.h>
 
-extern void xen_swiotlb_init(int verbose);
+extern int xen_swiotlb_init(int verbose, bool early);
 
 extern void
 *xen_swiotlb_alloc_coherent(struct device *hwdev, size_t size,
@@ -23,15 +23,6 @@ extern dma_addr_t xen_swiotlb_map_page(struct device *dev, struct page *page,
 extern void xen_swiotlb_unmap_page(struct device *hwdev, dma_addr_t dev_addr,
 				   size_t size, enum dma_data_direction dir,
 				   struct dma_attrs *attrs);
-/*
-extern int
-xen_swiotlb_map_sg(struct device *hwdev, struct scatterlist *sg, int nents,
-		   enum dma_data_direction dir);
-
-extern void
-xen_swiotlb_unmap_sg(struct device *hwdev, struct scatterlist *sg, int nents,
-		     enum dma_data_direction dir);
-*/
 extern int
 xen_swiotlb_map_sg_attrs(struct device *hwdev, struct scatterlist *sgl,
 			 int nelems, enum dma_data_direction dir,

lib/swiotlb.c

@@ -170,7 +170,7 @@ void __init swiotlb_init_with_tbl(char *tlb, unsigned long nslabs, int verbose)
  * Statically reserve bounce buffer space and initialize bounce buffer data
  * structures for the software IO TLB used to implement the DMA API.
  */
-void __init
+static void __init
 swiotlb_init_with_default_size(size_t default_size, int verbose)
 {
 	unsigned long bytes;
@@ -206,8 +206,9 @@ swiotlb_init(int verbose)
 int
 swiotlb_late_init_with_default_size(size_t default_size)
 {
-	unsigned long i, bytes, req_nslabs = io_tlb_nslabs;
+	unsigned long bytes, req_nslabs = io_tlb_nslabs;
 	unsigned int order;
+	int rc = 0;
 
 	if (!io_tlb_nslabs) {
 		io_tlb_nslabs = (default_size >> IO_TLB_SHIFT);
@@ -229,16 +230,32 @@ swiotlb_late_init_with_default_size(size_t default_size)
 		order--;
 	}
 
-	if (!io_tlb_start)
-		goto cleanup1;
-
+	if (!io_tlb_start) {
+		io_tlb_nslabs = req_nslabs;
+		return -ENOMEM;
+	}
 	if (order != get_order(bytes)) {
 		printk(KERN_WARNING "Warning: only able to allocate %ld MB "
 		       "for software IO TLB\n", (PAGE_SIZE << order) >> 20);
 		io_tlb_nslabs = SLABS_PER_PAGE << order;
-		bytes = io_tlb_nslabs << IO_TLB_SHIFT;
 	}
+	rc = swiotlb_late_init_with_tbl(io_tlb_start, io_tlb_nslabs);
+	if (rc)
+		free_pages((unsigned long)io_tlb_start, order);
+	return rc;
+}
+
+int
+swiotlb_late_init_with_tbl(char *tlb, unsigned long nslabs)
+{
+	unsigned long i, bytes;
+
+	bytes = nslabs << IO_TLB_SHIFT;
+
+	io_tlb_nslabs = nslabs;
+	io_tlb_start = tlb;
 	io_tlb_end = io_tlb_start + bytes;
+
 	memset(io_tlb_start, 0, bytes);
 
 	/*
@@ -288,10 +305,8 @@ swiotlb_late_init_with_default_size(size_t default_size)
 	io_tlb_list = NULL;
 cleanup2:
 	io_tlb_end = NULL;
-	free_pages((unsigned long)io_tlb_start, order);
 	io_tlb_start = NULL;
-cleanup1:
-	io_tlb_nslabs = req_nslabs;
+	io_tlb_nslabs = 0;
 	return -ENOMEM;
 }