1 files changed, 138 insertions, 84 deletions

diff --git a/arch/x86/xen/mmu.c b/arch/x86/xen/mmu.c
index a0aed70..fe00be6 100644
--- a/arch/x86/xen/mmu.c
+++ b/arch/x86/xen/mmu.c
@@ -48,6 +48,8 @@
 #include <linux/memblock.h>
 #include <linux/seq_file.h>
 
+#include <trace/events/xen.h>
+
 #include <asm/pgtable.h>
 #include <asm/tlbflush.h>
 #include <asm/fixmap.h>
@@ -194,6 +196,8 @@ void xen_set_domain_pte(pte_t *ptep, pte_t pteval, unsigned domid)
 	struct multicall_space mcs;
 	struct mmu_update *u;
 
+	trace_xen_mmu_set_domain_pte(ptep, pteval, domid);
+
 	mcs = xen_mc_entry(sizeof(*u));
 	u = mcs.args;
 
@@ -225,6 +229,24 @@ static void xen_extend_mmu_update(const struct mmu_update *update)
 	*u = *update;
 }
 
+static void xen_extend_mmuext_op(const struct mmuext_op *op)
+{
+	struct multicall_space mcs;
+	struct mmuext_op *u;
+
+	mcs = xen_mc_extend_args(__HYPERVISOR_mmuext_op, sizeof(*u));
+
+	if (mcs.mc != NULL) {
+		mcs.mc->args[1]++;
+	} else {
+		mcs = __xen_mc_entry(sizeof(*u));
+		MULTI_mmuext_op(mcs.mc, mcs.args, 1, NULL, DOMID_SELF);
+	}
+
+	u = mcs.args;
+	*u = *op;
+}
+
 static void xen_set_pmd_hyper(pmd_t *ptr, pmd_t val)
 {
 	struct mmu_update u;
@@ -245,6 +267,8 @@ static void xen_set_pmd_hyper(pmd_t *ptr, pmd_t val)
 
 static void xen_set_pmd(pmd_t *ptr, pmd_t val)
 {
+	trace_xen_mmu_set_pmd(ptr, val);
+
 	/* If page is not pinned, we can just update the entry
 	   directly */
 	if (!xen_page_pinned(ptr)) {
@@ -282,22 +306,30 @@ static bool xen_batched_set_pte(pte_t *ptep, pte_t pteval)
 	return true;
 }
 
-static void xen_set_pte(pte_t *ptep, pte_t pteval)
+static inline void __xen_set_pte(pte_t *ptep, pte_t pteval)
 {
 	if (!xen_batched_set_pte(ptep, pteval))
 		native_set_pte(ptep, pteval);
 }
 
+static void xen_set_pte(pte_t *ptep, pte_t pteval)
+{
+	trace_xen_mmu_set_pte(ptep, pteval);
+	__xen_set_pte(ptep, pteval);
+}
+
 static void xen_set_pte_at(struct mm_struct *mm, unsigned long addr,
 		    pte_t *ptep, pte_t pteval)
 {
-	xen_set_pte(ptep, pteval);
+	trace_xen_mmu_set_pte_at(mm, addr, ptep, pteval);
+	__xen_set_pte(ptep, pteval);
 }
 
 pte_t xen_ptep_modify_prot_start(struct mm_struct *mm,
 				 unsigned long addr, pte_t *ptep)
 {
 	/* Just return the pte as-is.  We preserve the bits on commit */
+	trace_xen_mmu_ptep_modify_prot_start(mm, addr, ptep, *ptep);
 	return *ptep;
 }
 
@@ -306,6 +338,7 @@ void xen_ptep_modify_prot_commit(struct mm_struct *mm, unsigned long addr,
 {
 	struct mmu_update u;
 
+	trace_xen_mmu_ptep_modify_prot_commit(mm, addr, ptep, pte);
 	xen_mc_batch();
 
 	u.ptr = virt_to_machine(ptep).maddr | MMU_PT_UPDATE_PRESERVE_AD;
@@ -467,41 +500,6 @@ static pte_t xen_make_pte(pteval_t pte)
 }
 PV_CALLEE_SAVE_REGS_THUNK(xen_make_pte);
 
-#ifdef CONFIG_XEN_DEBUG
-pte_t xen_make_pte_debug(pteval_t pte)
-{
-	phys_addr_t addr = (pte & PTE_PFN_MASK);
-	phys_addr_t other_addr;
-	bool io_page = false;
-	pte_t _pte;
-
-	if (pte & _PAGE_IOMAP)
-		io_page = true;
-
-	_pte = xen_make_pte(pte);
-
-	if (!addr)
-		return _pte;
-
-	if (io_page &&
-	    (xen_initial_domain() || addr >= ISA_END_ADDRESS)) {
-		other_addr = pfn_to_mfn(addr >> PAGE_SHIFT) << PAGE_SHIFT;
-		WARN_ONCE(addr != other_addr,
-			"0x%lx is using VM_IO, but it is 0x%lx!\n",
-			(unsigned long)addr, (unsigned long)other_addr);
-	} else {
-		pteval_t iomap_set = (_pte.pte & PTE_FLAGS_MASK) & _PAGE_IOMAP;
-		other_addr = (_pte.pte & PTE_PFN_MASK);
-		WARN_ONCE((addr == other_addr) && (!io_page) && (!iomap_set),
-			"0x%lx is missing VM_IO (and wasn't fixed)!\n",
-			(unsigned long)addr);
-	}
-
-	return _pte;
-}
-PV_CALLEE_SAVE_REGS_THUNK(xen_make_pte_debug);
-#endif
-
 static pgd_t xen_make_pgd(pgdval_t pgd)
 {
 	pgd = pte_pfn_to_mfn(pgd);
@@ -535,6 +533,8 @@ static void xen_set_pud_hyper(pud_t *ptr, pud_t val)
 
 static void xen_set_pud(pud_t *ptr, pud_t val)
 {
+	trace_xen_mmu_set_pud(ptr, val);
+
 	/* If page is not pinned, we can just update the entry
 	   directly */
 	if (!xen_page_pinned(ptr)) {
@@ -548,17 +548,20 @@ static void xen_set_pud(pud_t *ptr, pud_t val)
 #ifdef CONFIG_X86_PAE
 static void xen_set_pte_atomic(pte_t *ptep, pte_t pte)
 {
+	trace_xen_mmu_set_pte_atomic(ptep, pte);
 	set_64bit((u64 *)ptep, native_pte_val(pte));
 }
 
 static void xen_pte_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
 {
+	trace_xen_mmu_pte_clear(mm, addr, ptep);
 	if (!xen_batched_set_pte(ptep, native_make_pte(0)))
 		native_pte_clear(mm, addr, ptep);
 }
 
 static void xen_pmd_clear(pmd_t *pmdp)
 {
+	trace_xen_mmu_pmd_clear(pmdp);
 	set_pmd(pmdp, __pmd(0));
 }
 #endif	/* CONFIG_X86_PAE */
@@ -634,6 +637,8 @@ static void xen_set_pgd(pgd_t *ptr, pgd_t val)
 {
 	pgd_t *user_ptr = xen_get_user_pgd(ptr);
 
+	trace_xen_mmu_set_pgd(ptr, user_ptr, val);
+
 	/* If page is not pinned, we can just update the entry
 	   directly */
 	if (!xen_page_pinned(ptr)) {
@@ -793,14 +798,12 @@ static void xen_pte_unlock(void *v)
 
 static void xen_do_pin(unsigned level, unsigned long pfn)
 {
-	struct mmuext_op *op;
-	struct multicall_space mcs;
+	struct mmuext_op op;
 
-	mcs = __xen_mc_entry(sizeof(*op));
-	op = mcs.args;
-	op->cmd = level;
-	op->arg1.mfn = pfn_to_mfn(pfn);
-	MULTI_mmuext_op(mcs.mc, op, 1, NULL, DOMID_SELF);
+	op.cmd = level;
+	op.arg1.mfn = pfn_to_mfn(pfn);
+
+	xen_extend_mmuext_op(&op);
 }
 
 static int xen_pin_page(struct mm_struct *mm, struct page *page,
@@ -868,6 +871,8 @@ static int xen_pin_page(struct mm_struct *mm, struct page *page,
    read-only, and can be pinned. */
 static void __xen_pgd_pin(struct mm_struct *mm, pgd_t *pgd)
 {
+	trace_xen_mmu_pgd_pin(mm, pgd);
+
 	xen_mc_batch();
 
 	if (__xen_pgd_walk(mm, pgd, xen_pin_page, USER_LIMIT)) {
@@ -993,6 +998,8 @@ static int xen_unpin_page(struct mm_struct *mm, struct page *page,
 /* Release a pagetables pages back as normal RW */
 static void __xen_pgd_unpin(struct mm_struct *mm, pgd_t *pgd)
 {
+	trace_xen_mmu_pgd_unpin(mm, pgd);
+
 	xen_mc_batch();
 
 	xen_do_pin(MMUEXT_UNPIN_TABLE, PFN_DOWN(__pa(pgd)));
@@ -1196,11 +1203,32 @@ unsigned long xen_read_cr2_direct(void)
 	return percpu_read(xen_vcpu_info.arch.cr2);
 }
 
+void xen_flush_tlb_all(void)
+{
+	struct mmuext_op *op;
+	struct multicall_space mcs;
+
+	trace_xen_mmu_flush_tlb_all(0);
+
+	preempt_disable();
+
+	mcs = xen_mc_entry(sizeof(*op));
+
+	op = mcs.args;
+	op->cmd = MMUEXT_TLB_FLUSH_ALL;
+	MULTI_mmuext_op(mcs.mc, op, 1, NULL, DOMID_SELF);
+
+	xen_mc_issue(PARAVIRT_LAZY_MMU);
+
+	preempt_enable();
+}
 static void xen_flush_tlb(void)
 {
 	struct mmuext_op *op;
 	struct multicall_space mcs;
 
+	trace_xen_mmu_flush_tlb(0);
+
 	preempt_disable();
 
 	mcs = xen_mc_entry(sizeof(*op));
@@ -1219,6 +1247,8 @@ static void xen_flush_tlb_single(unsigned long addr)
 	struct mmuext_op *op;
 	struct multicall_space mcs;
 
+	trace_xen_mmu_flush_tlb_single(addr);
+
 	preempt_disable();
 
 	mcs = xen_mc_entry(sizeof(*op));
@@ -1245,6 +1275,8 @@ static void xen_flush_tlb_others(const struct cpumask *cpus,
 	} *args;
 	struct multicall_space mcs;
 
+	trace_xen_mmu_flush_tlb_others(cpus, mm, va);
+
 	if (cpumask_empty(cpus))
 		return;		/* nothing to do */
 
@@ -1280,10 +1312,11 @@ static void set_current_cr3(void *v)
 
 static void __xen_write_cr3(bool kernel, unsigned long cr3)
 {
-	struct mmuext_op *op;
-	struct multicall_space mcs;
+	struct mmuext_op op;
 	unsigned long mfn;
 
+	trace_xen_mmu_write_cr3(kernel, cr3);
+
 	if (cr3)
 		mfn = pfn_to_mfn(PFN_DOWN(cr3));
 	else
@@ -1291,13 +1324,10 @@ static void __xen_write_cr3(bool kernel, unsigned long cr3)
 
 	WARN_ON(mfn == 0 && kernel);
 
-	mcs = __xen_mc_entry(sizeof(*op));
+	op.cmd = kernel ? MMUEXT_NEW_BASEPTR : MMUEXT_NEW_USER_BASEPTR;
+	op.arg1.mfn = mfn;
 
-	op = mcs.args;
-	op->cmd = kernel ? MMUEXT_NEW_BASEPTR : MMUEXT_NEW_USER_BASEPTR;
-	op->arg1.mfn = mfn;
-
-	MULTI_mmuext_op(mcs.mc, op, 1, NULL, DOMID_SELF);
+	xen_extend_mmuext_op(&op);
 
 	if (kernel) {
 		percpu_write(xen_cr3, cr3);
@@ -1456,19 +1486,52 @@ static void __init xen_release_pmd_init(unsigned long pfn)
 	make_lowmem_page_readwrite(__va(PFN_PHYS(pfn)));
 }
 
+static inline void __pin_pagetable_pfn(unsigned cmd, unsigned long pfn)
+{
+	struct multicall_space mcs;
+	struct mmuext_op *op;
+
+	mcs = __xen_mc_entry(sizeof(*op));
+	op = mcs.args;
+	op->cmd = cmd;
+	op->arg1.mfn = pfn_to_mfn(pfn);
+
+	MULTI_mmuext_op(mcs.mc, mcs.args, 1, NULL, DOMID_SELF);
+}
+
+static inline void __set_pfn_prot(unsigned long pfn, pgprot_t prot)
+{
+	struct multicall_space mcs;
+	unsigned long addr = (unsigned long)__va(pfn << PAGE_SHIFT);
+
+	mcs = __xen_mc_entry(0);
+	MULTI_update_va_mapping(mcs.mc, (unsigned long)addr,
+				pfn_pte(pfn, prot), 0);
+}
+
 /* This needs to make sure the new pte page is pinned iff its being
    attached to a pinned pagetable. */
-static void xen_alloc_ptpage(struct mm_struct *mm, unsigned long pfn, unsigned level)
+static inline void xen_alloc_ptpage(struct mm_struct *mm, unsigned long pfn,
+				    unsigned level)
 {
-	struct page *page = pfn_to_page(pfn);
+	bool pinned = PagePinned(virt_to_page(mm->pgd));
+
+	trace_xen_mmu_alloc_ptpage(mm, pfn, level, pinned);
+
+	if (pinned) {
+		struct page *page = pfn_to_page(pfn);
 
-	if (PagePinned(virt_to_page(mm->pgd))) {
 		SetPagePinned(page);
 
 		if (!PageHighMem(page)) {
-			make_lowmem_page_readonly(__va(PFN_PHYS((unsigned long)pfn)));
+			xen_mc_batch();
+
+			__set_pfn_prot(pfn, PAGE_KERNEL_RO);
+
 			if (level == PT_PTE && USE_SPLIT_PTLOCKS)
-				pin_pagetable_pfn(MMUEXT_PIN_L1_TABLE, pfn);
+				__pin_pagetable_pfn(MMUEXT_PIN_L1_TABLE, pfn);
+
+			xen_mc_issue(PARAVIRT_LAZY_MMU);
 		} else {
 			/* make sure there are no stray mappings of
 			   this page */
@@ -1488,15 +1551,23 @@ static void xen_alloc_pmd(struct mm_struct *mm, unsigned long pfn)
 }
 
 /* This should never happen until we're OK to use struct page */
-static void xen_release_ptpage(unsigned long pfn, unsigned level)
+static inline void xen_release_ptpage(unsigned long pfn, unsigned level)
 {
 	struct page *page = pfn_to_page(pfn);
+	bool pinned = PagePinned(page);
+
+	trace_xen_mmu_release_ptpage(pfn, level, pinned);
 
-	if (PagePinned(page)) {
+	if (pinned) {
 		if (!PageHighMem(page)) {
+			xen_mc_batch();
+
 			if (level == PT_PTE && USE_SPLIT_PTLOCKS)
-				pin_pagetable_pfn(MMUEXT_UNPIN_TABLE, pfn);
-			make_lowmem_page_readwrite(__va(PFN_PHYS(pfn)));
+				__pin_pagetable_pfn(MMUEXT_UNPIN_TABLE, pfn);
+
+			__set_pfn_prot(pfn, PAGE_KERNEL);
+
+			xen_mc_issue(PARAVIRT_LAZY_MMU);
 		}
 		ClearPagePinned(page);
 	}
@@ -1639,10 +1710,8 @@ void __init xen_setup_machphys_mapping(void)
 		machine_to_phys_nr = MACH2PHYS_NR_ENTRIES;
 	}
 #ifdef CONFIG_X86_32
-	if ((machine_to_phys_mapping + machine_to_phys_nr)
-	    < machine_to_phys_mapping)
-		machine_to_phys_nr = (unsigned long *)NULL
-				     - machine_to_phys_mapping;
+	WARN_ON((machine_to_phys_mapping + (machine_to_phys_nr - 1))
+		< machine_to_phys_mapping);
 #endif
 }
 
@@ -1838,6 +1907,7 @@ static void xen_set_fixmap(unsigned idx, phys_addr_t phys, pgprot_t prot)
 # endif
 #else
 	case VSYSCALL_LAST_PAGE ... VSYSCALL_FIRST_PAGE:
+	case VVAR_PAGE:
 #endif
 	case FIX_TEXT_POKE0:
 	case FIX_TEXT_POKE1:
@@ -1878,7 +1948,8 @@ static void xen_set_fixmap(unsigned idx, phys_addr_t phys, pgprot_t prot)
 #ifdef CONFIG_X86_64
 	/* Replicate changes to map the vsyscall page into the user
 	   pagetable vsyscall mapping. */
-	if (idx >= VSYSCALL_LAST_PAGE && idx <= VSYSCALL_FIRST_PAGE) {
+	if ((idx >= VSYSCALL_LAST_PAGE && idx <= VSYSCALL_FIRST_PAGE) ||
+	    idx == VVAR_PAGE) {
 		unsigned long vaddr = __fix_to_virt(idx);
 		set_pte_vaddr_pud(level3_user_vsyscall, vaddr, pte);
 	}
@@ -1910,9 +1981,6 @@ void __init xen_ident_map_ISA(void)
 
 static void __init xen_post_allocator_init(void)
 {
-#ifdef CONFIG_XEN_DEBUG
-	pv_mmu_ops.make_pte = PV_CALLEE_SAVE(xen_make_pte_debug);
-#endif
 	pv_mmu_ops.set_pte = xen_set_pte;
 	pv_mmu_ops.set_pmd = xen_set_pmd;
 	pv_mmu_ops.set_pud = xen_set_pud;
@@ -2318,22 +2386,8 @@ int xen_remap_domain_mfn_range(struct vm_area_struct *vma,
 	err = 0;
 out:
 
-	flush_tlb_all();
+	xen_flush_tlb_all();
 
 	return err;
 }
 EXPORT_SYMBOL_GPL(xen_remap_domain_mfn_range);
-
-#ifdef CONFIG_XEN_DEBUG_FS
-static int p2m_dump_open(struct inode *inode, struct file *filp)
-{
-	return single_open(filp, p2m_dump_show, NULL);
-}
-
-static const struct file_operations p2m_dump_fops = {
-	.open		= p2m_dump_open,
-	.read		= seq_read,
-	.llseek		= seq_lseek,
-	.release	= single_release,
-};
-#endif /* CONFIG_XEN_DEBUG_FS */