Merge tag 'stable/for-linus-3.6-rc0-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/konrad/xen

Pull Xen update from Konrad Rzeszutek Wilk:
 "Features:
   * Performance improvement to lower the amount of traps the hypervisor
     has to do 32-bit guests.  Mainly for setting PTE entries and
     updating TLS descriptors.
   * MCE polling driver to collect hypervisor MCE buffer and present
     them to /dev/mcelog.
   * Physical CPU online/offline support.  When an privileged guest is
     booted it is present with virtual CPUs, which might have an 1:1 to
     physical CPUs but usually don't.  This provides mechanism to
     offline/online physical CPUs.
  Bug-fixes for:
   * Coverity found fixes in the console and ACPI processor driver.
   * PVonHVM kexec fixes along with some cleanups.
   * Pages that fall within E820 gaps and non-RAM regions (and had been
     released to hypervisor) would be populated back, but potentially in
     non-RAM regions."

* tag 'stable/for-linus-3.6-rc0-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/konrad/xen:
  xen: populate correct number of pages when across mem boundary (v2)
  xen PVonHVM: move shared_info to MMIO before kexec
  xen: simplify init_hvm_pv_info
  xen: remove cast from HYPERVISOR_shared_info assignment
  xen: enable platform-pci only in a Xen guest
  xen/pv-on-hvm kexec: shutdown watches from old kernel
  xen/x86: avoid updating TLS descriptors if they haven't changed
  xen/x86: add desc_equal() to compare GDT descriptors
  xen/mm: zero PTEs for non-present MFNs in the initial page table
  xen/mm: do direct hypercall in xen_set_pte() if batching is unavailable
  xen/hvc: Fix up checks when the info is allocated.
  xen/acpi: Fix potential memory leak.
  xen/mce: add .poll method for mcelog device driver
  xen/mce: schedule a workqueue to avoid sleep in atomic context
  xen/pcpu: Xen physical cpus online/offline sys interface
  xen/mce: Register native mce handler as vMCE bounce back point
  x86, MCE, AMD: Adjust initcall sequence for xen
  xen/mce: Add mcelog support for Xen platform

5 files changed, 217 insertions, 73 deletions

diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c
index ed7d54985d0c..bf4bda6d3e9a 100644
--- a/arch/x86/xen/enlighten.c
+++ b/arch/x86/xen/enlighten.c
@@ -31,6 +31,7 @@
 #include <linux/pci.h>
 #include <linux/gfp.h>
 #include <linux/memblock.h>
+#include <linux/syscore_ops.h>
 
 #include <xen/xen.h>
 #include <xen/interface/xen.h>
@@ -38,6 +39,7 @@
 #include <xen/interface/physdev.h>
 #include <xen/interface/vcpu.h>
 #include <xen/interface/memory.h>
+#include <xen/interface/xen-mca.h>
 #include <xen/features.h>
 #include <xen/page.h>
 #include <xen/hvm.h>
@@ -107,7 +109,7 @@ EXPORT_SYMBOL_GPL(xen_have_vector_callback);
  * Point at some empty memory to start with. We map the real shared_info
  * page as soon as fixmap is up and running.
  */
-struct shared_info *HYPERVISOR_shared_info = (void *)&xen_dummy_shared_info;
+struct shared_info *HYPERVISOR_shared_info = &xen_dummy_shared_info;
 
 /*
  * Flag to determine whether vcpu info placement is available on all
@@ -124,6 +126,19 @@ struct shared_info *HYPERVISOR_shared_info = (void *)&xen_dummy_shared_info;
  */
 static int have_vcpu_info_placement = 1;
 
+struct tls_descs {
+	struct desc_struct desc[3];
+};
+
+/*
+ * Updating the 3 TLS descriptors in the GDT on every task switch is
+ * surprisingly expensive so we avoid updating them if they haven't
+ * changed.  Since Xen writes different descriptors than the one
+ * passed in the update_descriptor hypercall we keep shadow copies to
+ * compare against.
+ */
+static DEFINE_PER_CPU(struct tls_descs, shadow_tls_desc);
+
 static void clamp_max_cpus(void)
 {
 #ifdef CONFIG_SMP
@@ -341,9 +356,7 @@ static void __init xen_init_cpuid_mask(void)
 	unsigned int xsave_mask;
 
 	cpuid_leaf1_edx_mask =
-		~((1 << X86_FEATURE_MCE)  |  /* disable MCE */
-		  (1 << X86_FEATURE_MCA)  |  /* disable MCA */
-		  (1 << X86_FEATURE_MTRR) |  /* disable MTRR */
+		~((1 << X86_FEATURE_MTRR) |  /* disable MTRR */
 		  (1 << X86_FEATURE_ACC));   /* thermal monitoring */
 
 	if (!xen_initial_domain())
@@ -540,12 +553,28 @@ static void __init xen_load_gdt_boot(const struct desc_ptr *dtr)
 		BUG();
 }
 
+static inline bool desc_equal(const struct desc_struct *d1,
+			      const struct desc_struct *d2)
+{
+	return d1->a == d2->a && d1->b == d2->b;
+}
+
 static void load_TLS_descriptor(struct thread_struct *t,
 				unsigned int cpu, unsigned int i)
 {
-	struct desc_struct *gdt = get_cpu_gdt_table(cpu);
-	xmaddr_t maddr = arbitrary_virt_to_machine(&gdt[GDT_ENTRY_TLS_MIN+i]);
-	struct multicall_space mc = __xen_mc_entry(0);
+	struct desc_struct *shadow = &per_cpu(shadow_tls_desc, cpu).desc[i];
+	struct desc_struct *gdt;
+	xmaddr_t maddr;
+	struct multicall_space mc;
+
+	if (desc_equal(shadow, &t->tls_array[i]))
+		return;
+
+	*shadow = t->tls_array[i];
+
+	gdt = get_cpu_gdt_table(cpu);
+	maddr = arbitrary_virt_to_machine(&gdt[GDT_ENTRY_TLS_MIN+i]);
+	mc = __xen_mc_entry(0);
 
 	MULTI_update_descriptor(mc.mc, maddr.maddr, t->tls_array[i]);
 }
@@ -627,8 +656,8 @@ static int cvt_gate_to_trap(int vector, const gate_desc *val,
 	/*
 	 * Look for known traps using IST, and substitute them
 	 * appropriately.  The debugger ones are the only ones we care
-	 * about.  Xen will handle faults like double_fault and
-	 * machine_check, so we should never see them.  Warn if
+	 * about.  Xen will handle faults like double_fault,
+	 * so we should never see them.  Warn if
 	 * there's an unexpected IST-using fault handler.
 	 */
 	if (addr == (unsigned long)debug)
@@ -643,7 +672,11 @@ static int cvt_gate_to_trap(int vector, const gate_desc *val,
 		return 0;
 #ifdef CONFIG_X86_MCE
 	} else if (addr == (unsigned long)machine_check) {
-		return 0;
+		/*
+		 * when xen hypervisor inject vMCE to guest,
+		 * use native mce handler to handle it
+		 */
+		;
 #endif
 	} else {
 		/* Some other trap using IST? */
@@ -1437,64 +1470,155 @@ asmlinkage void __init xen_start_kernel(void)
 #endif
 }
 
-static int init_hvm_pv_info(int *major, int *minor)
-{
-	uint32_t eax, ebx, ecx, edx, pages, msr, base;
-	u64 pfn;
-
-	base = xen_cpuid_base();
-	cpuid(base + 1, &eax, &ebx, &ecx, &edx);
-
-	*major = eax >> 16;
-	*minor = eax & 0xffff;
-	printk(KERN_INFO "Xen version %d.%d.\n", *major, *minor);
-
-	cpuid(base + 2, &pages, &msr, &ecx, &edx);
-
-	pfn = __pa(hypercall_page);
-	wrmsr_safe(msr, (u32)pfn, (u32)(pfn >> 32));
-
-	xen_setup_features();
-
-	pv_info.name = "Xen HVM";
-
-	xen_domain_type = XEN_HVM_DOMAIN;
+#ifdef CONFIG_XEN_PVHVM
+/*
+ * The pfn containing the shared_info is located somewhere in RAM. This
+ * will cause trouble if the current kernel is doing a kexec boot into a
+ * new kernel. The new kernel (and its startup code) can not know where
+ * the pfn is, so it can not reserve the page. The hypervisor will
+ * continue to update the pfn, and as a result memory corruption occours
+ * in the new kernel.
+ *
+ * One way to work around this issue is to allocate a page in the
+ * xen-platform pci device's BAR memory range. But pci init is done very
+ * late and the shared_info page is already in use very early to read
+ * the pvclock. So moving the pfn from RAM to MMIO is racy because some
+ * code paths on other vcpus could access the pfn during the small
+ * window when the old pfn is moved to the new pfn. There is even a
+ * small window were the old pfn is not backed by a mfn, and during that
+ * time all reads return -1.
+ *
+ * Because it is not known upfront where the MMIO region is located it
+ * can not be used right from the start in xen_hvm_init_shared_info.
+ *
+ * To minimise trouble the move of the pfn is done shortly before kexec.
+ * This does not eliminate the race because all vcpus are still online
+ * when the syscore_ops will be called. But hopefully there is no work
+ * pending at this point in time. Also the syscore_op is run last which
+ * reduces the risk further.
+ */
 
-	return 0;
-}
+static struct shared_info *xen_hvm_shared_info;
 
-void __ref xen_hvm_init_shared_info(void)
+static void xen_hvm_connect_shared_info(unsigned long pfn)
 {
-	int cpu;
 	struct xen_add_to_physmap xatp;
-	static struct shared_info *shared_info_page = 0;
 
-	if (!shared_info_page)
-		shared_info_page = (struct shared_info *)
-			extend_brk(PAGE_SIZE, PAGE_SIZE);
 	xatp.domid = DOMID_SELF;
 	xatp.idx = 0;
 	xatp.space = XENMAPSPACE_shared_info;
-	xatp.gpfn = __pa(shared_info_page) >> PAGE_SHIFT;
+	xatp.gpfn = pfn;
 	if (HYPERVISOR_memory_op(XENMEM_add_to_physmap, &xatp))
 		BUG();
 
-	HYPERVISOR_shared_info = (struct shared_info *)shared_info_page;
+}
+static void xen_hvm_set_shared_info(struct shared_info *sip)
+{
+	int cpu;
+
+	HYPERVISOR_shared_info = sip;
 
 	/* xen_vcpu is a pointer to the vcpu_info struct in the shared_info
 	 * page, we use it in the event channel upcall and in some pvclock
 	 * related functions. We don't need the vcpu_info placement
 	 * optimizations because we don't use any pv_mmu or pv_irq op on
 	 * HVM.
-	 * When xen_hvm_init_shared_info is run at boot time only vcpu 0 is
-	 * online but xen_hvm_init_shared_info is run at resume time too and
+	 * When xen_hvm_set_shared_info is run at boot time only vcpu 0 is
+	 * online but xen_hvm_set_shared_info is run at resume time too and
 	 * in that case multiple vcpus might be online. */
 	for_each_online_cpu(cpu) {
 		per_cpu(xen_vcpu, cpu) = &HYPERVISOR_shared_info->vcpu_info[cpu];
 	}
 }
 
-#ifdef CONFIG_XEN_PVHVM
+/* Reconnect the shared_info pfn to a mfn */
+void xen_hvm_resume_shared_info(void)
+{
+	xen_hvm_connect_shared_info(__pa(xen_hvm_shared_info) >> PAGE_SHIFT);
+}
+
+#ifdef CONFIG_KEXEC
+static struct shared_info *xen_hvm_shared_info_kexec;
+static unsigned long xen_hvm_shared_info_pfn_kexec;
+
+/* Remember a pfn in MMIO space for kexec reboot */
+void __devinit xen_hvm_prepare_kexec(struct shared_info *sip, unsigned long pfn)
+{
+	xen_hvm_shared_info_kexec = sip;
+	xen_hvm_shared_info_pfn_kexec = pfn;
+}
+
+static void xen_hvm_syscore_shutdown(void)
+{
+	struct xen_memory_reservation reservation = {
+		.domid = DOMID_SELF,
+		.nr_extents = 1,
+	};
+	unsigned long prev_pfn;
+	int rc;
+
+	if (!xen_hvm_shared_info_kexec)
+		return;
+
+	prev_pfn = __pa(xen_hvm_shared_info) >> PAGE_SHIFT;
+	set_xen_guest_handle(reservation.extent_start, &prev_pfn);
+
+	/* Move pfn to MMIO, disconnects previous pfn from mfn */
+	xen_hvm_connect_shared_info(xen_hvm_shared_info_pfn_kexec);
+
+	/* Update pointers, following hypercall is also a memory barrier */
+	xen_hvm_set_shared_info(xen_hvm_shared_info_kexec);
+
+	/* Allocate new mfn for previous pfn */
+	do {
+		rc = HYPERVISOR_memory_op(XENMEM_populate_physmap, &reservation);
+		if (rc == 0)
+			msleep(123);
+	} while (rc == 0);
+
+	/* Make sure the previous pfn is really connected to a (new) mfn */
+	BUG_ON(rc != 1);
+}
+
+static struct syscore_ops xen_hvm_syscore_ops = {
+	.shutdown = xen_hvm_syscore_shutdown,
+};
+#endif
+
+/* Use a pfn in RAM, may move to MMIO before kexec. */
+static void __init xen_hvm_init_shared_info(void)
+{
+	/* Remember pointer for resume */
+	xen_hvm_shared_info = extend_brk(PAGE_SIZE, PAGE_SIZE);
+	xen_hvm_connect_shared_info(__pa(xen_hvm_shared_info) >> PAGE_SHIFT);
+	xen_hvm_set_shared_info(xen_hvm_shared_info);
+}
+
+static void __init init_hvm_pv_info(void)
+{
+	int major, minor;
+	uint32_t eax, ebx, ecx, edx, pages, msr, base;
+	u64 pfn;
+
+	base = xen_cpuid_base();
+	cpuid(base + 1, &eax, &ebx, &ecx, &edx);
+
+	major = eax >> 16;
+	minor = eax & 0xffff;
+	printk(KERN_INFO "Xen version %d.%d.\n", major, minor);
+
+	cpuid(base + 2, &pages, &msr, &ecx, &edx);
+
+	pfn = __pa(hypercall_page);
+	wrmsr_safe(msr, (u32)pfn, (u32)(pfn >> 32));
+
+	xen_setup_features();
+
+	pv_info.name = "Xen HVM";
+
+	xen_domain_type = XEN_HVM_DOMAIN;
+}
+
 static int __cpuinit xen_hvm_cpu_notify(struct notifier_block *self,
 				    unsigned long action, void *hcpu)
 {
@@ -1517,14 +1641,12 @@ static struct notifier_block xen_hvm_cpu_notifier __cpuinitdata = {
 
 static void __init xen_hvm_guest_init(void)
 {
-	int r;
-	int major, minor;
-
-	r = init_hvm_pv_info(&major, &minor);
-	if (r < 0)
-		return;
+	init_hvm_pv_info();
 
 	xen_hvm_init_shared_info();
+#ifdef CONFIG_KEXEC
+	register_syscore_ops(&xen_hvm_syscore_ops);
+#endif
 
 	if (xen_feature(XENFEAT_hvm_callback_vector))
 		xen_have_vector_callback = 1;
diff --git a/arch/x86/xen/mmu.c b/arch/x86/xen/mmu.c
index 3a73785631ce..27336dfcda8e 100644
--- a/arch/x86/xen/mmu.c
+++ b/arch/x86/xen/mmu.c
@@ -308,8 +308,20 @@ static bool xen_batched_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);
+	if (!xen_batched_set_pte(ptep, pteval)) {
+		/*
+		 * Could call native_set_pte() here and trap and
+		 * emulate the PTE write but with 32-bit guests this
+		 * needs two traps (one for each of the two 32-bit
+		 * words in the PTE) so do one hypercall directly
+		 * instead.
+		 */
+		struct mmu_update u;
+
+		u.ptr = virt_to_machine(ptep).maddr | MMU_NORMAL_PT_UPDATE;
+		u.val = pte_val_ma(pteval);
+		HYPERVISOR_mmu_update(&u, 1, NULL, DOMID_SELF);
+	}
 }
 
 static void xen_set_pte(pte_t *ptep, pte_t pteval)
@@ -1416,13 +1428,28 @@ static pte_t __init mask_rw_pte(pte_t *ptep, pte_t pte)
 }
 #endif /* CONFIG_X86_64 */
 
-/* Init-time set_pte while constructing initial pagetables, which
-   doesn't allow RO pagetable pages to be remapped RW */
+/*
+ * Init-time set_pte while constructing initial pagetables, which
+ * doesn't allow RO page table pages to be remapped RW.
+ *
+ * If there is no MFN for this PFN then this page is initially
+ * ballooned out so clear the PTE (as in decrease_reservation() in
+ * drivers/xen/balloon.c).
+ *
+ * Many of these PTE updates are done on unpinned and writable pages
+ * and doing a hypercall for these is unnecessary and expensive.  At
+ * this point it is not possible to tell if a page is pinned or not,
+ * so always write the PTE directly and rely on Xen trapping and
+ * emulating any updates as necessary.
+ */
 static void __init xen_set_pte_init(pte_t *ptep, pte_t pte)
 {
-	pte = mask_rw_pte(ptep, pte);
+	if (pte_mfn(pte) != INVALID_P2M_ENTRY)
+		pte = mask_rw_pte(ptep, pte);
+	else
+		pte = __pte_ma(0);
 
-	xen_set_pte(ptep, pte);
+	native_set_pte(ptep, pte);
 }
 
 static void pin_pagetable_pfn(unsigned cmd, unsigned long pfn)
diff --git a/arch/x86/xen/setup.c b/arch/x86/xen/setup.c
index a4790bf22c59..ead85576d54a 100644
--- a/arch/x86/xen/setup.c
+++ b/arch/x86/xen/setup.c
@@ -157,25 +157,24 @@ static unsigned long __init xen_populate_chunk(
 	unsigned long dest_pfn;
 
 	for (i = 0, entry = list; i < map_size; i++, entry++) {
-		unsigned long credits = credits_left;
 		unsigned long s_pfn;
 		unsigned long e_pfn;
 		unsigned long pfns;
 		long capacity;
 
-		if (credits <= 0)
+		if (credits_left <= 0)
 			break;
 
 		if (entry->type != E820_RAM)
 			continue;
 
-		e_pfn = PFN_UP(entry->addr + entry->size);
+		e_pfn = PFN_DOWN(entry->addr + entry->size);
 
 		/* We only care about E820 after the xen_start_info->nr_pages */
 		if (e_pfn <= max_pfn)
 			continue;
 
-		s_pfn = PFN_DOWN(entry->addr);
+		s_pfn = PFN_UP(entry->addr);
 		/* If the E820 falls within the nr_pages, we want to start
 		 * at the nr_pages PFN.
 		 * If that would mean going past the E820 entry, skip it
@@ -184,23 +183,19 @@ static unsigned long __init xen_populate_chunk(
 			capacity = e_pfn - max_pfn;
 			dest_pfn = max_pfn;
 		} else {
-			/* last_pfn MUST be within E820_RAM regions */
-			if (*last_pfn && e_pfn >= *last_pfn)
-				s_pfn = *last_pfn;
 			capacity = e_pfn - s_pfn;
 			dest_pfn = s_pfn;
 		}
-		/* If we had filled this E820_RAM entry, go to the next one. */
-		if (capacity <= 0)
-			continue;
 
-		if (credits > capacity)
-			credits = capacity;
+		if (credits_left < capacity)
+			capacity = credits_left;
 
-		pfns = xen_do_chunk(dest_pfn, dest_pfn + credits, false);
+		pfns = xen_do_chunk(dest_pfn, dest_pfn + capacity, false);
 		done += pfns;
-		credits_left -= pfns;
 		*last_pfn = (dest_pfn + pfns);
+		if (pfns < capacity)
+			break;
+		credits_left -= pfns;
 	}
 	return done;
 }
diff --git a/arch/x86/xen/suspend.c b/arch/x86/xen/suspend.c
index 45329c8c226e..ae8a00c39de4 100644
--- a/arch/x86/xen/suspend.c
+++ b/arch/x86/xen/suspend.c
@@ -30,7 +30,7 @@ void xen_arch_hvm_post_suspend(int suspend_cancelled)
 {
 #ifdef CONFIG_XEN_PVHVM
 	int cpu;
-	xen_hvm_init_shared_info();
+	xen_hvm_resume_shared_info();
 	xen_callback_vector();
 	xen_unplug_emulated_devices();
 	if (xen_feature(XENFEAT_hvm_safe_pvclock)) {
diff --git a/arch/x86/xen/xen-ops.h b/arch/x86/xen/xen-ops.h
index 202d4c150154..1e4329e04e0f 100644
--- a/arch/x86/xen/xen-ops.h
+++ b/arch/x86/xen/xen-ops.h
@@ -41,7 +41,7 @@ void xen_enable_syscall(void);
 void xen_vcpu_restore(void);
 
 void xen_callback_vector(void);
-void xen_hvm_init_shared_info(void);
+void xen_hvm_resume_shared_info(void);
 void xen_unplug_emulated_devices(void);
 
 void __init xen_build_dynamic_phys_to_machine(void);