4 files changed, 58 insertions, 5 deletions

diff --git a/arch/x86/include/asm/apic.h b/arch/x86/include/asm/apic.h
index 7b3ca8324b6..9b7273cb219 100644
--- a/arch/x86/include/asm/apic.h
+++ b/arch/x86/include/asm/apic.h
@@ -495,7 +495,7 @@ static inline void default_wait_for_init_deassert(atomic_t *deassert)
 	return;
 }
 
-extern struct apic *generic_bigsmp_probe(void);
+extern void generic_bigsmp_probe(void);
 
 
 #ifdef CONFIG_X86_LOCAL_APIC
diff --git a/arch/x86/include/asm/timer.h b/arch/x86/include/asm/timer.h
index fa7b9176b76..431793e5d48 100644
--- a/arch/x86/include/asm/timer.h
+++ b/arch/x86/include/asm/timer.h
@@ -32,6 +32,22 @@ extern int no_timer_check;
  *  (mathieu.desnoyers@polymtl.ca)
  *
  *			-johnstul@us.ibm.com "math is hard, lets go shopping!"
+ *
+ * In:
+ *
+ * ns = cycles * cyc2ns_scale / SC
+ *
+ * Although we may still have enough bits to store the value of ns,
+ * in some cases, we may not have enough bits to store cycles * cyc2ns_scale,
+ * leading to an incorrect result.
+ *
+ * To avoid this, we can decompose 'cycles' into quotient and remainder
+ * of division by SC.  Then,
+ *
+ * ns = (quot * SC + rem) * cyc2ns_scale / SC
+ *    = quot * cyc2ns_scale + (rem * cyc2ns_scale) / SC
+ *
+ *			- sqazi@google.com
  */
 
 DECLARE_PER_CPU(unsigned long, cyc2ns);
@@ -41,9 +57,14 @@ DECLARE_PER_CPU(unsigned long long, cyc2ns_offset);
 
 static inline unsigned long long __cycles_2_ns(unsigned long long cyc)
 {
+	unsigned long long quot;
+	unsigned long long rem;
 	int cpu = smp_processor_id();
 	unsigned long long ns = per_cpu(cyc2ns_offset, cpu);
-	ns += cyc * per_cpu(cyc2ns, cpu) >> CYC2NS_SCALE_FACTOR;
+	quot = (cyc >> CYC2NS_SCALE_FACTOR);
+	rem = cyc & ((1ULL << CYC2NS_SCALE_FACTOR) - 1);
+	ns += quot * per_cpu(cyc2ns, cpu) +
+		((rem * per_cpu(cyc2ns, cpu)) >> CYC2NS_SCALE_FACTOR);
 	return ns;
 }
 
diff --git a/arch/x86/include/asm/uv/uv_bau.h b/arch/x86/include/asm/uv/uv_bau.h
index 37d369859c8..0c767a8e000 100644
--- a/arch/x86/include/asm/uv/uv_bau.h
+++ b/arch/x86/include/asm/uv/uv_bau.h
@@ -55,6 +55,7 @@
 #define UV_BAU_TUNABLES_DIR		"sgi_uv"
 #define UV_BAU_TUNABLES_FILE		"bau_tunables"
 #define WHITESPACE			" \t\n"
+#define uv_mmask			((1UL << uv_hub_info->m_val) - 1)
 #define uv_physnodeaddr(x)		((__pa((unsigned long)(x)) & uv_mmask))
 #define cpubit_isset(cpu, bau_local_cpumask) \
 	test_bit((cpu), (bau_local_cpumask).bits)
diff --git a/arch/x86/include/asm/uv/uv_hub.h b/arch/x86/include/asm/uv/uv_hub.h
index f26544a1521..54a13aaebc4 100644
--- a/arch/x86/include/asm/uv/uv_hub.h
+++ b/arch/x86/include/asm/uv/uv_hub.h
@@ -46,6 +46,13 @@
  *	PNODE   - the low N bits of the GNODE. The PNODE is the most useful variant
  *		  of the nasid for socket usage.
  *
+ *	GPA	- (global physical address) a socket physical address converted
+ *		  so that it can be used by the GRU as a global address. Socket
+ *		  physical addresses 1) need additional NASID (node) bits added
+ *		  to the high end of the address, and 2) unaliased if the
+ *		  partition does not have a physical address 0. In addition, on
+ *		  UV2 rev 1, GPAs need the gnode left shifted to bits 39 or 40.
+ *
  *
  *  NumaLink Global Physical Address Format:
  *  +--------------------------------+---------------------+
@@ -141,6 +148,8 @@ struct uv_hub_info_s {
 	unsigned int		gnode_extra;
 	unsigned char		hub_revision;
 	unsigned char		apic_pnode_shift;
+	unsigned char		m_shift;
+	unsigned char		n_lshift;
 	unsigned long		gnode_upper;
 	unsigned long		lowmem_remap_top;
 	unsigned long		lowmem_remap_base;
@@ -177,6 +186,16 @@ static inline int is_uv2_hub(void)
 	return uv_hub_info->hub_revision >= UV2_HUB_REVISION_BASE;
 }
 
+static inline int is_uv2_1_hub(void)
+{
+	return uv_hub_info->hub_revision == UV2_HUB_REVISION_BASE;
+}
+
+static inline int is_uv2_2_hub(void)
+{
+	return uv_hub_info->hub_revision == UV2_HUB_REVISION_BASE + 1;
+}
+
 union uvh_apicid {
     unsigned long       v;
     struct uvh_apicid_s {
@@ -276,7 +295,10 @@ static inline unsigned long uv_soc_phys_ram_to_gpa(unsigned long paddr)
 {
 	if (paddr < uv_hub_info->lowmem_remap_top)
 		paddr |= uv_hub_info->lowmem_remap_base;
-	return paddr | uv_hub_info->gnode_upper;
+	paddr |= uv_hub_info->gnode_upper;
+	paddr = ((paddr << uv_hub_info->m_shift) >> uv_hub_info->m_shift) |
+		((paddr >> uv_hub_info->m_val) << uv_hub_info->n_lshift);
+	return paddr;
 }
 
 
@@ -300,16 +322,19 @@ static inline unsigned long uv_gpa_to_soc_phys_ram(unsigned long gpa)
 	unsigned long remap_base = uv_hub_info->lowmem_remap_base;
 	unsigned long remap_top =  uv_hub_info->lowmem_remap_top;
 
+	gpa = ((gpa << uv_hub_info->m_shift) >> uv_hub_info->m_shift) |
+		((gpa >> uv_hub_info->n_lshift) << uv_hub_info->m_val);
+	gpa = gpa & uv_hub_info->gpa_mask;
 	if (paddr >= remap_base && paddr < remap_base + remap_top)
 		paddr -= remap_base;
 	return paddr;
 }
 
 
-/* gnode -> pnode */
+/* gpa -> pnode */
 static inline unsigned long uv_gpa_to_gnode(unsigned long gpa)
 {
-	return gpa >> uv_hub_info->m_val;
+	return gpa >> uv_hub_info->n_lshift;
 }
 
 /* gpa -> pnode */
@@ -320,6 +345,12 @@ static inline int uv_gpa_to_pnode(unsigned long gpa)
 	return uv_gpa_to_gnode(gpa) & n_mask;
 }
 
+/* gpa -> node offset*/
+static inline unsigned long uv_gpa_to_offset(unsigned long gpa)
+{
+	return (gpa << uv_hub_info->m_shift) >> uv_hub_info->m_shift;
+}
+
 /* pnode, offset --> socket virtual */
 static inline void *uv_pnode_offset_to_vaddr(int pnode, unsigned long offset)
 {