Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/x86/linux-2.6-x86

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/x86/linux-2.6-x86:
  x86: revert assign IRQs to hpet timer
  x86: tsc prevent time going backwards
  xen: Clear PG_pinned in release_{pt,pd}()
  xen: Do not pin/unpin PMD pages
  xen: refactor xen_{alloc,release}_{pt,pd}()
  x86, agpgart: scary messages are fortunately obsolete
  xen: fix grant table bug
  x86: fix breakage of vSMP irq operations
  x86: print message if nmi_watchdog=2 cannot be enabled
  x86: fix nmi_watchdog=2 on Pentium-D CPUs

8 files changed, 76 insertions, 31 deletions

diff --git a/arch/x86/kernel/cpu/perfctr-watchdog.c b/arch/x86/kernel/cpu/perfctr-watchdog.c
index 9b83832..b943e10 100644
--- a/arch/x86/kernel/cpu/perfctr-watchdog.c
+++ b/arch/x86/kernel/cpu/perfctr-watchdog.c
@@ -652,9 +652,6 @@ static void probe_nmi_watchdog(void)
 			wd_ops = &p6_wd_ops;
 			break;
 		case 15:
-			if (boot_cpu_data.x86_model > 0x4)
-				return;
-
 			wd_ops = &p4_wd_ops;
 			break;
 		default:
@@ -670,8 +667,10 @@ int lapic_watchdog_init(unsigned nmi_hz)
 {
 	if (!wd_ops) {
 		probe_nmi_watchdog();
-		if (!wd_ops)
+		if (!wd_ops) {
+			printk(KERN_INFO "NMI watchdog: CPU not supported\n");
 			return -1;
+		}
 
 		if (!wd_ops->reserve()) {
 			printk(KERN_ERR
diff --git a/arch/x86/kernel/hpet.c b/arch/x86/kernel/hpet.c
index 235fd6c..36652ea 100644
--- a/arch/x86/kernel/hpet.c
+++ b/arch/x86/kernel/hpet.c
@@ -133,13 +133,16 @@ static void hpet_reserve_platform_timers(unsigned long id)
 #ifdef CONFIG_HPET_EMULATE_RTC
 	hpet_reserve_timer(&hd, 1);
 #endif
+
 	hd.hd_irq[0] = HPET_LEGACY_8254;
 	hd.hd_irq[1] = HPET_LEGACY_RTC;
 
-       for (i = 2; i < nrtimers; timer++, i++)
-	       hd.hd_irq[i] = (timer->hpet_config & Tn_INT_ROUTE_CNF_MASK) >>
-		       Tn_INT_ROUTE_CNF_SHIFT;
+	for (i = 2; i < nrtimers; timer++, i++)
+		hd.hd_irq[i] = (timer->hpet_config & Tn_INT_ROUTE_CNF_MASK) >>
+			Tn_INT_ROUTE_CNF_SHIFT;
+
 	hpet_alloc(&hd);
+
 }
 #else
 static void hpet_reserve_platform_timers(unsigned long id) { }
diff --git a/arch/x86/kernel/pci-gart_64.c b/arch/x86/kernel/pci-gart_64.c
index faf3229..700e464 100644
--- a/arch/x86/kernel/pci-gart_64.c
+++ b/arch/x86/kernel/pci-gart_64.c
@@ -615,8 +615,8 @@ static __init int init_k8_gatt(struct agp_kern_info *info)
 
  nommu:
 	/* Should not happen anymore */
-	printk(KERN_ERR "PCI-DMA: More than 4GB of RAM and no IOMMU\n"
-	       KERN_ERR "PCI-DMA: 32bit PCI IO may malfunction.\n");
+	printk(KERN_WARNING "PCI-DMA: More than 4GB of RAM and no IOMMU\n"
+	       KERN_WARNING "falling back to iommu=soft.\n");
 	return -1;
 }
 
@@ -692,9 +692,9 @@ void __init gart_iommu_init(void)
 	    !gart_iommu_aperture ||
 	    (no_agp && init_k8_gatt(&info) < 0)) {
 		if (end_pfn > MAX_DMA32_PFN) {
-			printk(KERN_ERR "WARNING more than 4GB of memory "
-					"but GART IOMMU not available.\n"
-			       KERN_ERR "WARNING 32bit PCI may malfunction.\n");
+			printk(KERN_WARNING "More than 4GB of memory "
+			       	          "but GART IOMMU not available.\n"
+			       KERN_WARNING "falling back to iommu=soft.\n");
 		}
 		return;
 	}
diff --git a/arch/x86/kernel/tsc_32.c b/arch/x86/kernel/tsc_32.c
index f14cfd9..d7498b3 100644
--- a/arch/x86/kernel/tsc_32.c
+++ b/arch/x86/kernel/tsc_32.c
@@ -287,14 +287,27 @@ core_initcall(cpufreq_tsc);
 /* clock source code */
 
 static unsigned long current_tsc_khz = 0;
+static struct clocksource clocksource_tsc;
 
+/*
+ * We compare the TSC to the cycle_last value in the clocksource
+ * structure to avoid a nasty time-warp issue. This can be observed in
+ * a very small window right after one CPU updated cycle_last under
+ * xtime lock and the other CPU reads a TSC value which is smaller
+ * than the cycle_last reference value due to a TSC which is slighty
+ * behind. This delta is nowhere else observable, but in that case it
+ * results in a forward time jump in the range of hours due to the
+ * unsigned delta calculation of the time keeping core code, which is
+ * necessary to support wrapping clocksources like pm timer.
+ */
 static cycle_t read_tsc(void)
 {
 	cycle_t ret;
 
 	rdtscll(ret);
 
-	return ret;
+	return ret >= clocksource_tsc.cycle_last ?
+		ret : clocksource_tsc.cycle_last;
 }
 
 static struct clocksource clocksource_tsc = {
diff --git a/arch/x86/kernel/tsc_64.c b/arch/x86/kernel/tsc_64.c
index 947554d..01fc9f0 100644
--- a/arch/x86/kernel/tsc_64.c
+++ b/arch/x86/kernel/tsc_64.c
@@ -11,6 +11,7 @@
 #include <asm/hpet.h>
 #include <asm/timex.h>
 #include <asm/timer.h>
+#include <asm/vgtod.h>
 
 static int notsc __initdata = 0;
 
@@ -290,18 +291,34 @@ int __init notsc_setup(char *s)
 
 __setup("notsc", notsc_setup);
 
+static struct clocksource clocksource_tsc;
 
-/* clock source code: */
+/*
+ * We compare the TSC to the cycle_last value in the clocksource
+ * structure to avoid a nasty time-warp. This can be observed in a
+ * very small window right after one CPU updated cycle_last under
+ * xtime/vsyscall_gtod lock and the other CPU reads a TSC value which
+ * is smaller than the cycle_last reference value due to a TSC which
+ * is slighty behind. This delta is nowhere else observable, but in
+ * that case it results in a forward time jump in the range of hours
+ * due to the unsigned delta calculation of the time keeping core
+ * code, which is necessary to support wrapping clocksources like pm
+ * timer.
+ */
 static cycle_t read_tsc(void)
 {
 	cycle_t ret = (cycle_t)get_cycles();
-	return ret;
+
+	return ret >= clocksource_tsc.cycle_last ?
+		ret : clocksource_tsc.cycle_last;
 }
 
 static cycle_t __vsyscall_fn vread_tsc(void)
 {
 	cycle_t ret = (cycle_t)vget_cycles();
-	return ret;
+
+	return ret >= __vsyscall_gtod_data.clock.cycle_last ?
+		ret : __vsyscall_gtod_data.clock.cycle_last;
 }
 
 static struct clocksource clocksource_tsc = {
diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c
index de4e6f0..27ee26a 100644
--- a/arch/x86/xen/enlighten.c
+++ b/arch/x86/xen/enlighten.c
@@ -667,10 +667,10 @@ static void xen_release_pt_init(u32 pfn)
 	make_lowmem_page_readwrite(__va(PFN_PHYS(pfn)));
 }
 
-static void pin_pagetable_pfn(unsigned level, unsigned long pfn)
+static void pin_pagetable_pfn(unsigned cmd, unsigned long pfn)
 {
 	struct mmuext_op op;
-	op.cmd = level;
+	op.cmd = cmd;
 	op.arg1.mfn = pfn_to_mfn(pfn);
 	if (HYPERVISOR_mmuext_op(&op, 1, NULL, DOMID_SELF))
 		BUG();
@@ -687,7 +687,8 @@ static void xen_alloc_ptpage(struct mm_struct *mm, u32 pfn, unsigned level)
 
 		if (!PageHighMem(page)) {
 			make_lowmem_page_readonly(__va(PFN_PHYS(pfn)));
-			pin_pagetable_pfn(level, pfn);
+			if (level == PT_PTE)
+				pin_pagetable_pfn(MMUEXT_PIN_L1_TABLE, pfn);
 		} else
 			/* make sure there are no stray mappings of
 			   this page */
@@ -697,27 +698,39 @@ static void xen_alloc_ptpage(struct mm_struct *mm, u32 pfn, unsigned level)
 
 static void xen_alloc_pt(struct mm_struct *mm, u32 pfn)
 {
-	xen_alloc_ptpage(mm, pfn, MMUEXT_PIN_L1_TABLE);
+	xen_alloc_ptpage(mm, pfn, PT_PTE);
 }
 
 static void xen_alloc_pd(struct mm_struct *mm, u32 pfn)
 {
-	xen_alloc_ptpage(mm, pfn, MMUEXT_PIN_L2_TABLE);
+	xen_alloc_ptpage(mm, pfn, PT_PMD);
 }
 
 /* This should never happen until we're OK to use struct page */
-static void xen_release_pt(u32 pfn)
+static void xen_release_ptpage(u32 pfn, unsigned level)
 {
 	struct page *page = pfn_to_page(pfn);
 
 	if (PagePinned(page)) {
 		if (!PageHighMem(page)) {
-			pin_pagetable_pfn(MMUEXT_UNPIN_TABLE, pfn);
+			if (level == PT_PTE)
+				pin_pagetable_pfn(MMUEXT_UNPIN_TABLE, pfn);
 			make_lowmem_page_readwrite(__va(PFN_PHYS(pfn)));
 		}
+		ClearPagePinned(page);
 	}
 }
 
+static void xen_release_pt(u32 pfn)
+{
+	xen_release_ptpage(pfn, PT_PTE);
+}
+
+static void xen_release_pd(u32 pfn)
+{
+	xen_release_ptpage(pfn, PT_PMD);
+}
+
 #ifdef CONFIG_HIGHPTE
 static void *xen_kmap_atomic_pte(struct page *page, enum km_type type)
 {
@@ -838,7 +851,7 @@ static __init void xen_pagetable_setup_done(pgd_t *base)
 	pv_mmu_ops.alloc_pt = xen_alloc_pt;
 	pv_mmu_ops.alloc_pd = xen_alloc_pd;
 	pv_mmu_ops.release_pt = xen_release_pt;
-	pv_mmu_ops.release_pd = xen_release_pt;
+	pv_mmu_ops.release_pd = xen_release_pd;
 	pv_mmu_ops.set_pte = xen_set_pte;
 
 	setup_shared_info();
diff --git a/arch/x86/xen/mmu.c b/arch/x86/xen/mmu.c
index 0144395..2a054ef 100644
--- a/arch/x86/xen/mmu.c
+++ b/arch/x86/xen/mmu.c
@@ -310,13 +310,6 @@ pgd_t xen_make_pgd(unsigned long pgd)
 }
 #endif	/* CONFIG_X86_PAE */
 
-enum pt_level {
-	PT_PGD,
-	PT_PUD,
-	PT_PMD,
-	PT_PTE
-};
-
 /*
   (Yet another) pagetable walker.  This one is intended for pinning a
   pagetable.  This means that it walks a pagetable and calls the
diff --git a/arch/x86/xen/mmu.h b/arch/x86/xen/mmu.h
index c9ff27f..b5e189b 100644
--- a/arch/x86/xen/mmu.h
+++ b/arch/x86/xen/mmu.h
@@ -3,6 +3,13 @@
 #include <linux/linkage.h>
 #include <asm/page.h>
 
+enum pt_level {
+	PT_PGD,
+	PT_PUD,
+	PT_PMD,
+	PT_PTE
+};
+
 /*
  * Page-directory addresses above 4GB do not fit into architectural %cr3.
  * When accessing %cr3, or equivalent field in vcpu_guest_context, guests