Merge branch 'x86-amd-nb-for-linus' of git://git.kernel.org/pub/scm/l…

…inux/kernel/git/tip/linux-2.6-tip

* 'x86-amd-nb-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
  x86, amd_nb: Enable GART support for AMD family 0x15 CPUs
  x86, amd: Use compute unit information to determine thread siblings
  x86, amd: Extract compute unit information for AMD CPUs
  x86, amd: Add support for CPUID topology extension of AMD CPUs
  x86, nmi: Support NMI watchdog on newer AMD CPU families
  x86, mtrr: Assume SYS_CFG[Tom2ForceMemTypeWB] exists on all future AMD CPUs
  x86, k8: Rename k8.[ch] to amd_nb.[ch] and CONFIG_K8_NB to CONFIG_AMD_NB
  x86, k8-gart: Decouple handling of garts and northbridges
  x86, cacheinfo: Fix dependency of AMD L3 CID
  x86, kvm: add new AMD SVM feature bits
  x86, cpu: Fix allowed CPUID bits for KVM guests
  x86, cpu: Update AMD CPUID feature bits
  x86, cpu: Fix renamed, not-yet-shipping AMD CPUID feature bit
  x86, AMD: Remove needless CPU family check (for L3 cache info)
  x86, tsc: Remove CPU frequency calibration on AMD

arch/x86/Kconfig

@@ -674,7 +674,7 @@ config GART_IOMMU
 	bool "GART IOMMU support" if EMBEDDED
 	default y
 	select SWIOTLB
-	depends on X86_64 && PCI && K8_NB
+	depends on X86_64 && PCI && AMD_NB
 	---help---
 	  Support for full DMA access of devices with 32bit memory access only
 	  on systems with more than 3GB. This is usually needed for USB,
@@ -2091,7 +2091,7 @@ config OLPC_OPENFIRMWARE
 
 endif # X86_32
 
-config K8_NB
+config AMD_NB
 	def_bool y
 	depends on CPU_SUP_AMD && PCI
 

arch/x86/include/asm/k8.h → arch/x86/include/asm/amd_nb.h

@@ -1,30 +1,33 @@
-#ifndef _ASM_X86_K8_H
-#define _ASM_X86_K8_H
+#ifndef _ASM_X86_AMD_NB_H
+#define _ASM_X86_AMD_NB_H
 
 #include <linux/pci.h>
 
 extern struct pci_device_id k8_nb_ids[];
 struct bootnode;
 
 extern int early_is_k8_nb(u32 value);
-extern struct pci_dev **k8_northbridges;
-extern int num_k8_northbridges;
 extern int cache_k8_northbridges(void);
 extern void k8_flush_garts(void);
 extern int k8_get_nodes(struct bootnode *nodes);
 extern int k8_numa_init(unsigned long start_pfn, unsigned long end_pfn);
 extern int k8_scan_nodes(void);
 
-#ifdef CONFIG_K8_NB
-extern int num_k8_northbridges;
+struct k8_northbridge_info {
+	u16 num;
+	u8 gart_supported;
+	struct pci_dev **nb_misc;
+};
+extern struct k8_northbridge_info k8_northbridges;
+
+#ifdef CONFIG_AMD_NB
 
 static inline struct pci_dev *node_to_k8_nb_misc(int node)
 {
-	return (node < num_k8_northbridges) ? k8_northbridges[node] : NULL;
+	return (node < k8_northbridges.num) ? k8_northbridges.nb_misc[node] : NULL;
 }
 
 #else
-#define num_k8_northbridges 0
 
 static inline struct pci_dev *node_to_k8_nb_misc(int node)
 {
@@ -33,4 +36,4 @@ static inline struct pci_dev *node_to_k8_nb_misc(int node)
 #endif
 
 
-#endif /* _ASM_X86_K8_H */
+#endif /* _ASM_X86_AMD_NB_H */

arch/x86/include/asm/cpufeature.h

@@ -152,10 +152,14 @@
 #define X86_FEATURE_3DNOWPREFETCH (6*32+ 8) /* 3DNow prefetch instructions */
 #define X86_FEATURE_OSVW	(6*32+ 9) /* OS Visible Workaround */
 #define X86_FEATURE_IBS		(6*32+10) /* Instruction Based Sampling */
-#define X86_FEATURE_SSE5	(6*32+11) /* SSE-5 */
+#define X86_FEATURE_XOP		(6*32+11) /* extended AVX instructions */
 #define X86_FEATURE_SKINIT	(6*32+12) /* SKINIT/STGI instructions */
 #define X86_FEATURE_WDT		(6*32+13) /* Watchdog timer */
+#define X86_FEATURE_LWP		(6*32+15) /* Light Weight Profiling */
+#define X86_FEATURE_FMA4	(6*32+16) /* 4 operands MAC instructions */
 #define X86_FEATURE_NODEID_MSR	(6*32+19) /* NodeId MSR */
+#define X86_FEATURE_TBM		(6*32+21) /* trailing bit manipulations */
+#define X86_FEATURE_TOPOEXT	(6*32+22) /* topology extensions CPUID leafs */
 
 /*
  * Auxiliary flags: Linux defined - For features scattered in various
@@ -180,6 +184,13 @@
 #define X86_FEATURE_LBRV	(8*32+ 6) /* AMD LBR Virtualization support */
 #define X86_FEATURE_SVML	(8*32+ 7) /* "svm_lock" AMD SVM locking MSR */
 #define X86_FEATURE_NRIPS	(8*32+ 8) /* "nrip_save" AMD SVM next_rip save */
+#define X86_FEATURE_TSCRATEMSR  (8*32+ 9) /* "tsc_scale" AMD TSC scaling support */
+#define X86_FEATURE_VMCBCLEAN   (8*32+10) /* "vmcb_clean" AMD VMCB clean bits support */
+#define X86_FEATURE_FLUSHBYASID (8*32+11) /* AMD flush-by-ASID support */
+#define X86_FEATURE_DECODEASSISTS (8*32+12) /* AMD Decode Assists support */
+#define X86_FEATURE_PAUSEFILTER (8*32+13) /* AMD filtered pause intercept */
+#define X86_FEATURE_PFTHRESHOLD (8*32+14) /* AMD pause filter threshold */
+
 
 /* Intel-defined CPU features, CPUID level 0x00000007:0 (ebx), word 9 */
 #define X86_FEATURE_FSGSBASE	(9*32+ 0) /* {RD/WR}{FS/GS}BASE instructions*/

arch/x86/include/asm/processor.h

@@ -110,6 +110,8 @@ struct cpuinfo_x86 {
 	u16			phys_proc_id;
 	/* Core id: */
 	u16			cpu_core_id;
+	/* Compute unit id */
+	u8			compute_unit_id;
 	/* Index into per_cpu list: */
 	u16			cpu_index;
 #endif

arch/x86/kernel/Makefile

@@ -90,7 +90,7 @@ obj-$(CONFIG_EARLY_PRINTK)	+= early_printk.o
 obj-$(CONFIG_HPET_TIMER) 	+= hpet.o
 obj-$(CONFIG_APB_TIMER)		+= apb_timer.o
 
-obj-$(CONFIG_K8_NB)		+= k8.o
+obj-$(CONFIG_AMD_NB)		+= amd_nb.o
 obj-$(CONFIG_DEBUG_RODATA_TEST)	+= test_rodata.o
 obj-$(CONFIG_DEBUG_NX_TEST)	+= test_nx.o
 

arch/x86/kernel/k8.c → arch/x86/kernel/amd_nb.c

@@ -8,21 +8,19 @@
 #include <linux/errno.h>
 #include <linux/module.h>
 #include <linux/spinlock.h>
-#include <asm/k8.h>
-
-int num_k8_northbridges;
-EXPORT_SYMBOL(num_k8_northbridges);
+#include <asm/amd_nb.h>
 
 static u32 *flush_words;
 
 struct pci_device_id k8_nb_ids[] = {
 	{ PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_K8_NB_MISC) },
 	{ PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_10H_NB_MISC) },
+	{ PCI_DEVICE(PCI_VENDOR_ID_AMD, PCI_DEVICE_ID_AMD_15H_NB_MISC) },
 	{}
 };
 EXPORT_SYMBOL(k8_nb_ids);
 
-struct pci_dev **k8_northbridges;
+struct k8_northbridge_info k8_northbridges;
 EXPORT_SYMBOL(k8_northbridges);
 
 static struct pci_dev *next_k8_northbridge(struct pci_dev *dev)
@@ -40,36 +38,45 @@ int cache_k8_northbridges(void)
 	int i;
 	struct pci_dev *dev;
 
-	if (num_k8_northbridges)
+	if (k8_northbridges.num)
 		return 0;
 
 	dev = NULL;
 	while ((dev = next_k8_northbridge(dev)) != NULL)
-		num_k8_northbridges++;
+		k8_northbridges.num++;
+
+	/* some CPU families (e.g. family 0x11) do not support GART */
+	if (boot_cpu_data.x86 == 0xf || boot_cpu_data.x86 == 0x10 ||
+	    boot_cpu_data.x86 == 0x15)
+		k8_northbridges.gart_supported = 1;
 
-	k8_northbridges = kmalloc((num_k8_northbridges + 1) * sizeof(void *),
-				  GFP_KERNEL);
-	if (!k8_northbridges)
+	k8_northbridges.nb_misc = kmalloc((k8_northbridges.num + 1) *
+					  sizeof(void *), GFP_KERNEL);
+	if (!k8_northbridges.nb_misc)
 		return -ENOMEM;
 
-	if (!num_k8_northbridges) {
-		k8_northbridges[0] = NULL;
+	if (!k8_northbridges.num) {
+		k8_northbridges.nb_misc[0] = NULL;
 		return 0;
 	}
 
-	flush_words = kmalloc(num_k8_northbridges * sizeof(u32), GFP_KERNEL);
-	if (!flush_words) {
-		kfree(k8_northbridges);
-		return -ENOMEM;
+	if (k8_northbridges.gart_supported) {
+		flush_words = kmalloc(k8_northbridges.num * sizeof(u32),
+				      GFP_KERNEL);
+		if (!flush_words) {
+			kfree(k8_northbridges.nb_misc);
+			return -ENOMEM;
+		}
 	}
 
 	dev = NULL;
 	i = 0;
 	while ((dev = next_k8_northbridge(dev)) != NULL) {
-		k8_northbridges[i] = dev;
-		pci_read_config_dword(dev, 0x9c, &flush_words[i++]);
+		k8_northbridges.nb_misc[i] = dev;
+		if (k8_northbridges.gart_supported)
+			pci_read_config_dword(dev, 0x9c, &flush_words[i++]);
 	}
-	k8_northbridges[i] = NULL;
+	k8_northbridges.nb_misc[i] = NULL;
 	return 0;
 }
 EXPORT_SYMBOL_GPL(cache_k8_northbridges);
@@ -93,22 +100,25 @@ void k8_flush_garts(void)
 	unsigned long flags;
 	static DEFINE_SPINLOCK(gart_lock);
 
+	if (!k8_northbridges.gart_supported)
+		return;
+
 	/* Avoid races between AGP and IOMMU. In theory it's not needed
 	   but I'm not sure if the hardware won't lose flush requests
 	   when another is pending. This whole thing is so expensive anyways
 	   that it doesn't matter to serialize more. -AK */
 	spin_lock_irqsave(&gart_lock, flags);
 	flushed = 0;
-	for (i = 0; i < num_k8_northbridges; i++) {
-		pci_write_config_dword(k8_northbridges[i], 0x9c,
+	for (i = 0; i < k8_northbridges.num; i++) {
+		pci_write_config_dword(k8_northbridges.nb_misc[i], 0x9c,
 				       flush_words[i]|1);
 		flushed++;
 	}
-	for (i = 0; i < num_k8_northbridges; i++) {
+	for (i = 0; i < k8_northbridges.num; i++) {
 		u32 w;
 		/* Make sure the hardware actually executed the flush*/
 		for (;;) {
-			pci_read_config_dword(k8_northbridges[i],
+			pci_read_config_dword(k8_northbridges.nb_misc[i],
 					      0x9c, &w);
 			if (!(w & 1))
 				break;

arch/x86/kernel/aperture_64.c

@@ -27,7 +27,7 @@
 #include <asm/gart.h>
 #include <asm/pci-direct.h>
 #include <asm/dma.h>
-#include <asm/k8.h>
+#include <asm/amd_nb.h>
 #include <asm/x86_init.h>
 
 int gart_iommu_aperture;

arch/x86/kernel/cpu/amd.c

@@ -253,37 +253,51 @@ static int __cpuinit nearby_node(int apicid)
 #endif
 
 /*
- * Fixup core topology information for AMD multi-node processors.
- * Assumption: Number of cores in each internal node is the same.
+ * Fixup core topology information for
+ * (1) AMD multi-node processors
+ *     Assumption: Number of cores in each internal node is the same.
+ * (2) AMD processors supporting compute units
  */
 #ifdef CONFIG_X86_HT
-static void __cpuinit amd_fixup_dcm(struct cpuinfo_x86 *c)
+static void __cpuinit amd_get_topology(struct cpuinfo_x86 *c)
 {
-	unsigned long long value;
-	u32 nodes, cores_per_node;
+	u32 nodes;
+	u8 node_id;
 	int cpu = smp_processor_id();
 
-	if (!cpu_has(c, X86_FEATURE_NODEID_MSR))
-		return;
+	/* get information required for multi-node processors */
+	if (cpu_has(c, X86_FEATURE_TOPOEXT)) {
+		u32 eax, ebx, ecx, edx;
 
-	/* fixup topology information only once for a core */
-	if (cpu_has(c, X86_FEATURE_AMD_DCM))
-		return;
+		cpuid(0x8000001e, &eax, &ebx, &ecx, &edx);
+		nodes = ((ecx >> 8) & 7) + 1;
+		node_id = ecx & 7;
 
-	rdmsrl(MSR_FAM10H_NODE_ID, value);
+		/* get compute unit information */
+		smp_num_siblings = ((ebx >> 8) & 3) + 1;
+		c->compute_unit_id = ebx & 0xff;
+	} else if (cpu_has(c, X86_FEATURE_NODEID_MSR)) {
+		u64 value;
 
-	nodes = ((value >> 3) & 7) + 1;
-	if (nodes == 1)
+		rdmsrl(MSR_FAM10H_NODE_ID, value);
+		nodes = ((value >> 3) & 7) + 1;
+		node_id = value & 7;
+	} else
 		return;
 
-	set_cpu_cap(c, X86_FEATURE_AMD_DCM);
-	cores_per_node = c->x86_max_cores / nodes;
+	/* fixup multi-node processor information */
+	if (nodes > 1) {
+		u32 cores_per_node;
+
+		set_cpu_cap(c, X86_FEATURE_AMD_DCM);
+		cores_per_node = c->x86_max_cores / nodes;
 
-	/* store NodeID, use llc_shared_map to store sibling info */
-	per_cpu(cpu_llc_id, cpu) = value & 7;
+		/* store NodeID, use llc_shared_map to store sibling info */
+		per_cpu(cpu_llc_id, cpu) = node_id;
 
-	/* fixup core id to be in range from 0 to (cores_per_node - 1) */
-	c->cpu_core_id = c->cpu_core_id % cores_per_node;
+		/* core id to be in range from 0 to (cores_per_node - 1) */
+		c->cpu_core_id = c->cpu_core_id % cores_per_node;
+	}
 }
 #endif
 
@@ -304,9 +318,7 @@ static void __cpuinit amd_detect_cmp(struct cpuinfo_x86 *c)
 	c->phys_proc_id = c->initial_apicid >> bits;
 	/* use socket ID also for last level cache */
 	per_cpu(cpu_llc_id, cpu) = c->phys_proc_id;
-	/* fixup topology information on multi-node processors */
-	if ((c->x86 == 0x10) && (c->x86_model == 9))
-		amd_fixup_dcm(c);
+	amd_get_topology(c);
 #endif
 }
 
@@ -412,6 +424,23 @@ static void __cpuinit early_init_amd(struct cpuinfo_x86 *c)
 			set_cpu_cap(c, X86_FEATURE_EXTD_APICID);
 	}
 #endif
+
+	/* We need to do the following only once */
+	if (c != &boot_cpu_data)
+		return;
+
+	if (cpu_has(c, X86_FEATURE_CONSTANT_TSC)) {
+
+		if (c->x86 > 0x10 ||
+		    (c->x86 == 0x10 && c->x86_model >= 0x2)) {
+			u64 val;
+
+			rdmsrl(MSR_K7_HWCR, val);
+			if (!(val & BIT(24)))
+				printk(KERN_WARNING FW_BUG "TSC doesn't count "
+					"with P0 frequency!\n");
+		}
+	}
 }
 
 static void __cpuinit init_amd(struct cpuinfo_x86 *c)
@@ -523,7 +552,7 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c)
 #endif
 
 	if (c->extended_cpuid_level >= 0x80000006) {
-		if ((c->x86 >= 0x0f) && (cpuid_edx(0x80000006) & 0xf000))
+		if (cpuid_edx(0x80000006) & 0xf000)
 			num_cache_leaves = 4;
 		else
 			num_cache_leaves = 3;