Merge branch 'upstream/xen-tracing2' of git://git.kernel.org/pub/scm/linux/kernel/git/jeremy/xen

* 'upstream/xen-tracing2' of git://git.kernel.org/pub/scm/linux/kernel/git/jeremy/xen:
  xen/trace: use class for multicall trace
  xen/trace: convert mmu events to use DECLARE_EVENT_CLASS()/DEFINE_EVENT()
  xen/multicall: move *idx fields to start of mc_buffer
  xen/multicall: special-case singleton hypercalls
  xen/multicalls: add unlikely around slowpath in __xen_mc_entry()
  xen/multicalls: disable MC_DEBUG
  xen/mmu: tune pgtable alloc/release
  xen/mmu: use extend_args for more mmuext updates
  xen/trace: add tlb flush tracepoints
  xen/trace: add segment desc tracing
  xen/trace: add xen_pgd_(un)pin tracepoints
  xen/trace: add ptpage alloc/release tracepoints
  xen/trace: add mmu tracepoints
  xen/trace: add multicall tracing
  xen/trace: set up tracepoint skeleton
  xen/multicalls: remove debugfs stats
  trace/xen: add skeleton for Xen trace events

6 files changed, 243 insertions, 150 deletions

diff --git a/arch/x86/xen/Makefile b/arch/x86/xen/Makefile
index a6575b949b1..ccf73b2f3e6 100644
--- a/arch/x86/xen/Makefile
+++ b/arch/x86/xen/Makefile
@@ -13,7 +13,7 @@ CFLAGS_mmu.o			:= $(nostackp)
 obj-y           := enlighten.o setup.o multicalls.o mmu.o irq.o \
                         time.o xen-asm.o xen-asm_$(BITS).o \
                         grant-table.o suspend.o platform-pci-unplug.o \
-                        p2m.o
+                        p2m.o trace.o
 
 obj-$(CONFIG_SMP)               += smp.o
 obj-$(CONFIG_PARAVIRT_SPINLOCKS)+= spinlock.o
diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c
index 53257421082..974a528458a 100644
--- a/arch/x86/xen/enlighten.c
+++ b/arch/x86/xen/enlighten.c
@@ -341,6 +341,8 @@ static void xen_set_ldt(const void *addr, unsigned entries)
         struct mmuext_op *op;
         struct multicall_space mcs = xen_mc_entry(sizeof(*op));
 
+        trace_xen_cpu_set_ldt(addr, entries);
+
         op = mcs.args;
         op->cmd = MMUEXT_SET_LDT;
         op->arg1.linear_addr = (unsigned long)addr;
@@ -496,6 +498,8 @@ static void xen_write_ldt_entry(struct desc_struct *dt, int entrynum,
         xmaddr_t mach_lp = arbitrary_virt_to_machine(&dt[entrynum]);
         u64 entry = *(u64 *)ptr;
 
+        trace_xen_cpu_write_ldt_entry(dt, entrynum, entry);
+
         preempt_disable();
 
         xen_mc_flush();
@@ -565,6 +569,8 @@ static void xen_write_idt_entry(gate_desc *dt, int entrynum, const gate_desc *g)
         unsigned long p = (unsigned long)&dt[entrynum];
         unsigned long start, end;
 
+        trace_xen_cpu_write_idt_entry(dt, entrynum, g);
+
         preempt_disable();
 
         start = __this_cpu_read(idt_desc.address);
@@ -619,6 +625,8 @@ static void xen_load_idt(const struct desc_ptr *desc)
         static DEFINE_SPINLOCK(lock);
         static struct trap_info traps[257];
 
+        trace_xen_cpu_load_idt(desc);
+
         spin_lock(&lock);
 
         __get_cpu_var(idt_desc) = *desc;
@@ -637,6 +645,8 @@ static void xen_load_idt(const struct desc_ptr *desc)
 static void xen_write_gdt_entry(struct desc_struct *dt, int entry,
                                 const void *desc, int type)
 {
+        trace_xen_cpu_write_gdt_entry(dt, entry, desc, type);
+
         preempt_disable();
 
         switch (type) {
@@ -665,6 +675,8 @@ static void xen_write_gdt_entry(struct desc_struct *dt, int entry,
 static void __init xen_write_gdt_entry_boot(struct desc_struct *dt, int entry,
                                             const void *desc, int type)
 {
+        trace_xen_cpu_write_gdt_entry(dt, entry, desc, type);
+
         switch (type) {
         case DESC_LDT:
         case DESC_TSS:
@@ -684,7 +696,9 @@ static void __init xen_write_gdt_entry_boot(struct desc_struct *dt, int entry,
 static void xen_load_sp0(struct tss_struct *tss,
                          struct thread_struct *thread)
 {
-        struct multicall_space mcs = xen_mc_entry(0);
+        struct multicall_space mcs;
+
+        mcs = xen_mc_entry(0);
         MULTI_stack_switch(mcs.mc, __KERNEL_DS, thread->sp0);
         xen_mc_issue(PARAVIRT_LAZY_CPU);
 }
diff --git a/arch/x86/xen/mmu.c b/arch/x86/xen/mmu.c
index 0ccccb67a99..f987bde77c4 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;
@@ -530,6 +563,8 @@ 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)) {
@@ -543,17 +578,20 @@ 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 */
@@ -629,6 +667,8 @@ 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)) {
@@ -788,14 +828,12 @@ 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,
@@ -863,6 +901,8 @@ 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)) {
@@ -988,6 +1028,8 @@ 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)));
@@ -1196,6 +1238,8 @@ 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));
@@ -1214,6 +1258,8 @@ 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));
@@ -1240,6 +1286,8 @@ 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 */
 
@@ -1275,10 +1323,11 @@ 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
@@ -1286,13 +1335,10 @@ static void __xen_write_cr3(bool kernel, unsigned long cr3)
 
         WARN_ON(mfn == 0 && kernel);
 
-        mcs = __xen_mc_entry(sizeof(*op));
-
-        op = mcs.args;
-        op->cmd = kernel ? MMUEXT_NEW_BASEPTR : MMUEXT_NEW_USER_BASEPTR;
-        op->arg1.mfn = mfn;
+        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);
@@ -1451,19 +1497,52 @@ 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 */
@@ -1483,15 +1562,23 @@ 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);
 
-        if (PagePinned(page)) {
+        trace_xen_mmu_release_ptpage(pfn, level, pinned);
+
+        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);
         }
diff --git a/arch/x86/xen/multicalls.c b/arch/x86/xen/multicalls.c
index 1b2b73ff0a6..0d82003e76a 100644
--- a/arch/x86/xen/multicalls.c
+++ b/arch/x86/xen/multicalls.c
@@ -30,12 +30,13 @@
 
 #define MC_BATCH        32
 
-#define MC_DEBUG        1
+#define MC_DEBUG        0
 
 #define MC_ARGS         (MC_BATCH * 16)
 
 
 struct mc_buffer {
+        unsigned mcidx, argidx, cbidx;
         struct multicall_entry entries[MC_BATCH];
 #if MC_DEBUG
         struct multicall_entry debug[MC_BATCH];
@@ -46,85 +47,15 @@ struct mc_buffer {
                 void (*fn)(void *);
                 void *data;
         } callbacks[MC_BATCH];
-        unsigned mcidx, argidx, cbidx;
 };
 
 static DEFINE_PER_CPU(struct mc_buffer, mc_buffer);
 DEFINE_PER_CPU(unsigned long, xen_mc_irq_flags);
 
-/* flush reasons 0- slots, 1- args, 2- callbacks */
-enum flush_reasons
-{
-        FL_SLOTS,
-        FL_ARGS,
-        FL_CALLBACKS,
-
-        FL_N_REASONS
-};
-
-#ifdef CONFIG_XEN_DEBUG_FS
-#define NHYPERCALLS     40              /* not really */
-
-static struct {
-        unsigned histo[MC_BATCH+1];
-
-        unsigned issued;
-        unsigned arg_total;
-        unsigned hypercalls;
-        unsigned histo_hypercalls[NHYPERCALLS];
-
-        unsigned flush[FL_N_REASONS];
-} mc_stats;
-
-static u8 zero_stats;
-
-static inline void check_zero(void)
-{
-        if (unlikely(zero_stats)) {
-                memset(&mc_stats, 0, sizeof(mc_stats));
-                zero_stats = 0;
-        }
-}
-
-static void mc_add_stats(const struct mc_buffer *mc)
-{
-        int i;
-
-        check_zero();
-
-        mc_stats.issued++;
-        mc_stats.hypercalls += mc->mcidx;
-        mc_stats.arg_total += mc->argidx;
-
-        mc_stats.histo[mc->mcidx]++;
-        for(i = 0; i < mc->mcidx; i++) {
-                unsigned op = mc->entries[i].op;
-                if (op < NHYPERCALLS)
-                        mc_stats.histo_hypercalls[op]++;
-        }
-}
-
-static void mc_stats_flush(enum flush_reasons idx)
-{
-        check_zero();
-
-        mc_stats.flush[idx]++;
-}
-
-#else  /* !CONFIG_XEN_DEBUG_FS */
-
-static inline void mc_add_stats(const struct mc_buffer *mc)
-{
-}
-
-static inline void mc_stats_flush(enum flush_reasons idx)
-{
-}
-#endif  /* CONFIG_XEN_DEBUG_FS */
-
 void xen_mc_flush(void)
 {
         struct mc_buffer *b = &__get_cpu_var(mc_buffer);
+        struct multicall_entry *mc;
         int ret = 0;
         unsigned long flags;
         int i;
@@ -135,9 +66,26 @@ void xen_mc_flush(void)
            something in the middle */
         local_irq_save(flags);
 
-        mc_add_stats(b);
+        trace_xen_mc_flush(b->mcidx, b->argidx, b->cbidx);
+
+        switch (b->mcidx) {
+        case 0:
+                /* no-op */
+                BUG_ON(b->argidx != 0);
+                break;
+
+        case 1:
+                /* Singleton multicall - bypass multicall machinery
+                   and just do the call directly. */
+                mc = &b->entries[0];
+
+                mc->result = privcmd_call(mc->op,
+                                          mc->args[0], mc->args[1], mc->args[2], 
+                                          mc->args[3], mc->args[4]);
+                ret = mc->result < 0;
+                break;
 
-        if (b->mcidx) {
+        default:
 #if MC_DEBUG
                 memcpy(b->debug, b->entries,
                        b->mcidx * sizeof(struct multicall_entry));
@@ -164,11 +112,10 @@ void xen_mc_flush(void)
                         }
                 }
 #endif
+        }
 
-                b->mcidx = 0;
-                b->argidx = 0;
-        } else
-                BUG_ON(b->argidx != 0);
+        b->mcidx = 0;
+        b->argidx = 0;
 
         for (i = 0; i < b->cbidx; i++) {
                 struct callback *cb = &b->callbacks[i];
@@ -188,18 +135,21 @@ struct multicall_space __xen_mc_entry(size_t args)
         struct multicall_space ret;
         unsigned argidx = roundup(b->argidx, sizeof(u64));
 
+        trace_xen_mc_entry_alloc(args);
+
         BUG_ON(preemptible());
         BUG_ON(b->argidx >= MC_ARGS);
 
-        if (b->mcidx == MC_BATCH ||
-            (argidx + args) >= MC_ARGS) {
-                mc_stats_flush(b->mcidx == MC_BATCH ? FL_SLOTS : FL_ARGS);
+        if (unlikely(b->mcidx == MC_BATCH ||
+                     (argidx + args) >= MC_ARGS)) {
+                trace_xen_mc_flush_reason((b->mcidx == MC_BATCH) ?
+                                          XEN_MC_FL_BATCH : XEN_MC_FL_ARGS);
                 xen_mc_flush();
                 argidx = roundup(b->argidx, sizeof(u64));
         }
 
         ret.mc = &b->entries[b->mcidx];
-#ifdef MC_DEBUG
+#if MC_DEBUG
         b->caller[b->mcidx] = __builtin_return_address(0);
 #endif
         b->mcidx++;
@@ -218,20 +168,25 @@ struct multicall_space xen_mc_extend_args(unsigned long op, size_t size)
         BUG_ON(preemptible());
         BUG_ON(b->argidx >= MC_ARGS);
 
-        if (b->mcidx == 0)
-                return ret;
-
-        if (b->entries[b->mcidx - 1].op != op)
-                return ret;
+        if (unlikely(b->mcidx == 0 ||
+                     b->entries[b->mcidx - 1].op != op)) {
+                trace_xen_mc_extend_args(op, size, XEN_MC_XE_BAD_OP);
+                goto out;
+        }
 
-        if ((b->argidx + size) >= MC_ARGS)
-                return ret;
+        if (unlikely((b->argidx + size) >= MC_ARGS)) {
+                trace_xen_mc_extend_args(op, size, XEN_MC_XE_NO_SPACE);
+                goto out;
+        }
 
         ret.mc = &b->entries[b->mcidx - 1];
         ret.args = &b->args[b->argidx];
         b->argidx += size;
 
         BUG_ON(b->argidx >= MC_ARGS);
+
+        trace_xen_mc_extend_args(op, size, XEN_MC_XE_OK);
+out:
         return ret;
 }
 
@@ -241,43 +196,13 @@ void xen_mc_callback(void (*fn)(void *), void *data)
         struct callback *cb;
 
         if (b->cbidx == MC_BATCH) {
-                mc_stats_flush(FL_CALLBACKS);
+                trace_xen_mc_flush_reason(XEN_MC_FL_CALLBACK);
                 xen_mc_flush();
         }
 
+        trace_xen_mc_callback(fn, data);
+
         cb = &b->callbacks[b->cbidx++];
         cb->fn = fn;
         cb->data = data;
 }
-
-#ifdef CONFIG_XEN_DEBUG_FS
-
-static struct dentry *d_mc_debug;
-
-static int __init xen_mc_debugfs(void)
-{
-        struct dentry *d_xen = xen_init_debugfs();
-
-        if (d_xen == NULL)
-                return -ENOMEM;
-
-        d_mc_debug = debugfs_create_dir("multicalls", d_xen);
-
-        debugfs_create_u8("zero_stats", 0644, d_mc_debug, &zero_stats);
-
-        debugfs_create_u32("batches", 0444, d_mc_debug, &mc_stats.issued);
-        debugfs_create_u32("hypercalls", 0444, d_mc_debug, &mc_stats.hypercalls);
-        debugfs_create_u32("arg_total", 0444, d_mc_debug, &mc_stats.arg_total);
-
-        xen_debugfs_create_u32_array("batch_histo", 0444, d_mc_debug,
-                                     mc_stats.histo, MC_BATCH);
-        xen_debugfs_create_u32_array("hypercall_histo", 0444, d_mc_debug,
-                                     mc_stats.histo_hypercalls, NHYPERCALLS);
-        xen_debugfs_create_u32_array("flush_reasons", 0444, d_mc_debug,
-                                     mc_stats.flush, FL_N_REASONS);
-
-        return 0;
-}
-fs_initcall(xen_mc_debugfs);
-
-#endif  /* CONFIG_XEN_DEBUG_FS */
diff --git a/arch/x86/xen/multicalls.h b/arch/x86/xen/multicalls.h
index 4ec8035e321..dee79b78a90 100644
--- a/arch/x86/xen/multicalls.h
+++ b/arch/x86/xen/multicalls.h
@@ -1,6 +1,8 @@
 #ifndef _XEN_MULTICALLS_H
 #define _XEN_MULTICALLS_H
 
+#include <trace/events/xen.h>
+
 #include "xen-ops.h"
 
 /* Multicalls */
@@ -20,8 +22,10 @@ DECLARE_PER_CPU(unsigned long, xen_mc_irq_flags);
 static inline void xen_mc_batch(void)
 {
         unsigned long flags;
+
         /* need to disable interrupts until this entry is complete */
         local_irq_save(flags);
+        trace_xen_mc_batch(paravirt_get_lazy_mode());
         __this_cpu_write(xen_mc_irq_flags, flags);
 }
 
@@ -37,6 +41,8 @@ void xen_mc_flush(void);
 /* Issue a multicall if we're not in a lazy mode */
 static inline void xen_mc_issue(unsigned mode)
 {
+        trace_xen_mc_issue(mode);
+
         if ((paravirt_get_lazy_mode() & mode) == 0)
                 xen_mc_flush();
 
diff --git a/arch/x86/xen/trace.c b/arch/x86/xen/trace.c
new file mode 100644
index 00000000000..734beba2a08
--- /dev/null
+++ b/arch/x86/xen/trace.c
@@ -0,0 +1,61 @@
+#include <linux/ftrace.h>
+
+#define N(x)    [__HYPERVISOR_##x] = "("#x")"
+static const char *xen_hypercall_names[] = {
+        N(set_trap_table),
+        N(mmu_update),
+        N(set_gdt),
+        N(stack_switch),
+        N(set_callbacks),
+        N(fpu_taskswitch),
+        N(sched_op_compat),
+        N(dom0_op),
+        N(set_debugreg),
+        N(get_debugreg),
+        N(update_descriptor),
+        N(memory_op),
+        N(multicall),
+        N(update_va_mapping),
+        N(set_timer_op),
+        N(event_channel_op_compat),
+        N(xen_version),
+        N(console_io),
+        N(physdev_op_compat),
+        N(grant_table_op),
+        N(vm_assist),
+        N(update_va_mapping_otherdomain),
+        N(iret),
+        N(vcpu_op),
+        N(set_segment_base),
+        N(mmuext_op),
+        N(acm_op),
+        N(nmi_op),
+        N(sched_op),
+        N(callback_op),
+        N(xenoprof_op),
+        N(event_channel_op),
+        N(physdev_op),
+        N(hvm_op),
+
+/* Architecture-specific hypercall definitions. */
+        N(arch_0),
+        N(arch_1),
+        N(arch_2),
+        N(arch_3),
+        N(arch_4),
+        N(arch_5),
+        N(arch_6),
+        N(arch_7),
+};
+#undef N
+
+static const char *xen_hypercall_name(unsigned op)
+{
+        if (op < ARRAY_SIZE(xen_hypercall_names) && xen_hypercall_names[op] != NULL)
+                return xen_hypercall_names[op];
+
+        return "";
+}
+
+#define CREATE_TRACE_POINTS
+#include <trace/events/xen.h>