Merge branch 'core/percpu' into x86/paravirt

9 files changed, 36 insertions, 22 deletions

diff --git a/arch/x86/include/asm/processor.h b/arch/x86/include/asm/processor.h
index befa20b4a68c..656d02ea509b 100644
--- a/arch/x86/include/asm/processor.h
+++ b/arch/x86/include/asm/processor.h
@@ -768,7 +768,8 @@ extern int sysenter_setup(void);
 extern struct desc_ptr		early_gdt_descr;
 
 extern void cpu_set_gdt(int);
-extern void switch_to_new_gdt(void);
+extern void switch_to_new_gdt(int);
+extern void load_percpu_segment(int);
 extern void cpu_init(void);
 
 static inline unsigned long get_debugctlmsr(void)
diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c
index 652fdc9a757a..0f73ea423089 100644
--- a/arch/x86/kernel/cpu/common.c
+++ b/arch/x86/kernel/cpu/common.c
@@ -253,23 +253,28 @@ static char __cpuinit *table_lookup_model(struct cpuinfo_x86 *c)
 
 __u32 cleared_cpu_caps[NCAPINTS] __cpuinitdata;
 
+void load_percpu_segment(int cpu)
+{
+#ifdef CONFIG_X86_32
+	loadsegment(fs, __KERNEL_PERCPU);
+#else
+	loadsegment(gs, 0);
+	wrmsrl(MSR_GS_BASE, (unsigned long)per_cpu(irq_stack_union.gs_base, cpu));
+#endif
+}
+
 /* Current gdt points %fs at the "master" per-cpu area: after this,
  * it's on the real one. */
-void switch_to_new_gdt(void)
+void switch_to_new_gdt(int cpu)
 {
 	struct desc_ptr gdt_descr;
-	int cpu = smp_processor_id();
 
 	gdt_descr.address = (long)get_cpu_gdt_table(cpu);
 	gdt_descr.size = GDT_SIZE - 1;
 	load_gdt(&gdt_descr);
 	/* Reload the per-cpu base */
-#ifdef CONFIG_X86_32
-	loadsegment(fs, __KERNEL_PERCPU);
-#else
-	loadsegment(gs, 0);
-	wrmsrl(MSR_GS_BASE, (unsigned long)per_cpu(irq_stack_union.gs_base, cpu));
-#endif
+
+	load_percpu_segment(cpu);
 }
 
 static struct cpu_dev *cpu_devs[X86_VENDOR_NUM] = {};
@@ -993,7 +998,7 @@ void __cpuinit cpu_init(void)
 	 * and set up the GDT descriptor:
 	 */
 
-	switch_to_new_gdt();
+	switch_to_new_gdt(cpu);
 	loadsegment(fs, 0);
 
 	load_idt((const struct desc_ptr *)&idt_descr);
@@ -1098,7 +1103,7 @@ void __cpuinit cpu_init(void)
 		clear_in_cr4(X86_CR4_VME|X86_CR4_PVI|X86_CR4_TSD|X86_CR4_DE);
 
 	load_idt(&idt_descr);
-	switch_to_new_gdt();
+	switch_to_new_gdt(cpu);
 
 	/*
 	 * Set up and load the per-CPU TSS and LDT
diff --git a/arch/x86/kernel/setup_percpu.c b/arch/x86/kernel/setup_percpu.c
index 0d1e7ac439f4..ef91747bbed5 100644
--- a/arch/x86/kernel/setup_percpu.c
+++ b/arch/x86/kernel/setup_percpu.c
@@ -122,7 +122,7 @@ void __init setup_per_cpu_areas(void)
 		 * area.  Reload any changed state for the boot CPU.
 		 */
 		if (cpu == boot_cpu_id)
-			switch_to_new_gdt();
+			switch_to_new_gdt(cpu);
 
 		DBG("PERCPU: cpu %4d %p\n", cpu, ptr);
 	}
diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c
index f9dbcff43546..612d3c74f6a3 100644
--- a/arch/x86/kernel/smpboot.c
+++ b/arch/x86/kernel/smpboot.c
@@ -1185,7 +1185,7 @@ out:
 void __init native_smp_prepare_boot_cpu(void)
 {
 	int me = smp_processor_id();
-	switch_to_new_gdt();
+	switch_to_new_gdt(me);
 	/* already set me in cpu_online_mask in boot_cpu_init() */
 	cpumask_set_cpu(me, cpu_callout_mask);
 	per_cpu(cpu_state, me) = CPU_ONLINE;
diff --git a/arch/x86/kernel/tlb_uv.c b/arch/x86/kernel/tlb_uv.c
index 89fce1b6d01f..f4b2f27d19b9 100644
--- a/arch/x86/kernel/tlb_uv.c
+++ b/arch/x86/kernel/tlb_uv.c
@@ -259,7 +259,7 @@ const struct cpumask *uv_flush_send_and_wait(int cpu, int this_blade,
 		 * the cpu's, all of which are still in the mask.
 		 */
 		__get_cpu_var(ptcstats).ptc_i++;
-		return 0;
+		return flush_mask;
 	}
 
 	/*
diff --git a/arch/x86/kernel/vmlinux_64.lds.S b/arch/x86/kernel/vmlinux_64.lds.S
index c9740996430a..07f62d287ff0 100644
--- a/arch/x86/kernel/vmlinux_64.lds.S
+++ b/arch/x86/kernel/vmlinux_64.lds.S
@@ -22,6 +22,7 @@ PHDRS {
 #ifdef CONFIG_SMP
 	percpu PT_LOAD FLAGS(7);	/* RWE */
 #endif
+	data.init2 PT_LOAD FLAGS(7);	/* RWE */
 	note PT_NOTE FLAGS(0);	/* ___ */
 }
 SECTIONS
@@ -215,7 +216,7 @@ SECTIONS
   /*
    * percpu offsets are zero-based on SMP.  PERCPU_VADDR() changes the
    * output PHDR, so the next output section - __data_nosave - should
-   * switch it back to data.init.  Also, pda should be at the head of
+   * start another section data.init2.  Also, pda should be at the head of
    * percpu area.  Preallocate it and define the percpu offset symbol
    * so that it can be accessed as a percpu variable.
    */
@@ -232,7 +233,7 @@ SECTIONS
   __nosave_begin = .;
   .data_nosave : AT(ADDR(.data_nosave) - LOAD_OFFSET) {
       *(.data.nosave)
-  } :data.init	/* switch back to data.init, see PERCPU_VADDR() above */
+  } :data.init2 /* use another section data.init2, see PERCPU_VADDR() above */
   . = ALIGN(PAGE_SIZE);
   __nosave_end = .;
 
diff --git a/arch/x86/mach-voyager/voyager_smp.c b/arch/x86/mach-voyager/voyager_smp.c
index 331cd6d56483..58c7cac3440d 100644
--- a/arch/x86/mach-voyager/voyager_smp.c
+++ b/arch/x86/mach-voyager/voyager_smp.c
@@ -1746,12 +1746,13 @@ static void __init voyager_smp_prepare_cpus(unsigned int max_cpus)
 
 static void __cpuinit voyager_smp_prepare_boot_cpu(void)
 {
-	switch_to_new_gdt();
+	int cpu = smp_processor_id();
+	switch_to_new_gdt(cpu);
 
-	cpu_set(smp_processor_id(), cpu_online_map);
-	cpu_set(smp_processor_id(), cpu_callout_map);
-	cpu_set(smp_processor_id(), cpu_possible_map);
-	cpu_set(smp_processor_id(), cpu_present_map);
+	cpu_set(cpu, cpu_online_map);
+	cpu_set(cpu, cpu_callout_map);
+	cpu_set(cpu, cpu_possible_map);
+	cpu_set(cpu, cpu_present_map);
 }
 
 static int __cpuinit voyager_cpu_up(unsigned int cpu)
diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c
index ff6d530ccc77..cd022c43dfbc 100644
--- a/arch/x86/xen/enlighten.c
+++ b/arch/x86/xen/enlighten.c
@@ -917,6 +917,9 @@ asmlinkage void __init xen_start_kernel(void)
 	have_vcpu_info_placement = 0;
 #endif
 
+	/* setup percpu state */
+	load_percpu_segment(0);
+
 	xen_smp_init();
 
 	/* Get mfn list */
diff --git a/arch/x86/xen/smp.c b/arch/x86/xen/smp.c
index 7735e3dd359c..88d5d5ec6beb 100644
--- a/arch/x86/xen/smp.c
+++ b/arch/x86/xen/smp.c
@@ -170,7 +170,8 @@ static void __init xen_smp_prepare_boot_cpu(void)
 
 	/* We've switched to the "real" per-cpu gdt, so make sure the
 	   old memory can be recycled */
-	make_lowmem_page_readwrite(&per_cpu_var(gdt_page));
+	make_lowmem_page_readwrite(__per_cpu_load +
+				   (unsigned long)&per_cpu_var(gdt_page));
 
 	xen_setup_vcpu_info_placement();
 }
@@ -235,6 +236,8 @@ cpu_initialize_context(unsigned int cpu, struct task_struct *idle)
 	ctxt->user_regs.ss = __KERNEL_DS;
 #ifdef CONFIG_X86_32
 	ctxt->user_regs.fs = __KERNEL_PERCPU;
+#else
+	ctxt->gs_base_kernel = per_cpu_offset(cpu);
 #endif
 	ctxt->user_regs.eip = (unsigned long)cpu_bringup_and_idle;
 	ctxt->user_regs.eflags = 0x1000; /* IOPL_RING1 */