From ae1ee11be77f51cedb6c569887dddc70c163ab6d Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Wed, 2 May 2007 19:27:10 +0200
Subject: [PATCH] i386: Use per-cpu variables for GDT, PDA

Allocating PDA and GDT at boot is a pain.  Using simple per-cpu variables adds
happiness (although we need the GDT page-aligned for Xen, which we do in a
followup patch).

[akpm@linux-foundation.org: build fix]
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---
 arch/i386/kernel/cpu/common.c | 94 ++++++-------------------------------------
 1 file changed, 13 insertions(+), 81 deletions(-)

(limited to 'arch/i386/kernel/cpu/common.c')

diff --git a/arch/i386/kernel/cpu/common.c b/arch/i386/kernel/cpu/common.c
index dcbbd0a8bfc2..2335f4464ead 100644
--- a/arch/i386/kernel/cpu/common.c
+++ b/arch/i386/kernel/cpu/common.c
@@ -25,8 +25,10 @@
 DEFINE_PER_CPU(struct Xgt_desc_struct, cpu_gdt_descr);
 EXPORT_PER_CPU_SYMBOL(cpu_gdt_descr);
 
-struct i386_pda *_cpu_pda[NR_CPUS] __read_mostly;
-EXPORT_SYMBOL(_cpu_pda);
+DEFINE_PER_CPU(struct desc_struct, cpu_gdt[GDT_ENTRIES]);
+
+DEFINE_PER_CPU(struct i386_pda, _cpu_pda);
+EXPORT_PER_CPU_SYMBOL(_cpu_pda);
 
 static int cachesize_override __cpuinitdata = -1;
 static int disable_x86_fxsr __cpuinitdata;
@@ -609,52 +611,6 @@ struct pt_regs * __devinit idle_regs(struct pt_regs *regs)
 	return regs;
 }
 
-static __cpuinit int alloc_gdt(int cpu)
-{
-	struct Xgt_desc_struct *cpu_gdt_descr = &per_cpu(cpu_gdt_descr, cpu);
-	struct desc_struct *gdt;
-	struct i386_pda *pda;
-
-	gdt = (struct desc_struct *)cpu_gdt_descr->address;
-	pda = cpu_pda(cpu);
-
-	/*
-	 * This is a horrible hack to allocate the GDT.  The problem
-	 * is that cpu_init() is called really early for the boot CPU
-	 * (and hence needs bootmem) but much later for the secondary
-	 * CPUs, when bootmem will have gone away
-	 */
-	if (NODE_DATA(0)->bdata->node_bootmem_map) {
-		BUG_ON(gdt != NULL || pda != NULL);
-
-		gdt = alloc_bootmem_pages(PAGE_SIZE);
-		pda = alloc_bootmem(sizeof(*pda));
-		/* alloc_bootmem(_pages) panics on failure, so no check */
-
-		memset(gdt, 0, PAGE_SIZE);
-		memset(pda, 0, sizeof(*pda));
-	} else {
-		/* GDT and PDA might already have been allocated if
-		   this is a CPU hotplug re-insertion. */
-		if (gdt == NULL)
-			gdt = (struct desc_struct *)get_zeroed_page(GFP_KERNEL);
-
-		if (pda == NULL)
-			pda = kmalloc_node(sizeof(*pda), GFP_KERNEL, cpu_to_node(cpu));
-
-		if (unlikely(!gdt || !pda)) {
-			free_pages((unsigned long)gdt, 0);
-			kfree(pda);
-			return 0;
-		}
-	}
-
- 	cpu_gdt_descr->address = (unsigned long)gdt;
-	cpu_pda(cpu) = pda;
-
-	return 1;
-}
-
 /* Initial PDA used by boot CPU */
 struct i386_pda boot_pda = {
 	._pda = &boot_pda,
@@ -670,31 +626,17 @@ static inline void set_kernel_fs(void)
 	asm volatile ("mov %0, %%fs" : : "r" (__KERNEL_PDA) : "memory");
 }
 
-/* Initialize the CPU's GDT and PDA.  The boot CPU does this for
-   itself, but secondaries find this done for them. */
-__cpuinit int init_gdt(int cpu, struct task_struct *idle)
+/* Initialize the CPU's GDT and PDA.  This is either the boot CPU doing itself
+   (still using cpu_gdt_table), or a CPU doing it for a secondary which
+   will soon come up. */
+__cpuinit void init_gdt(int cpu, struct task_struct *idle)
 {
 	struct Xgt_desc_struct *cpu_gdt_descr = &per_cpu(cpu_gdt_descr, cpu);
-	struct desc_struct *gdt;
-	struct i386_pda *pda;
-
-	/* For non-boot CPUs, the GDT and PDA should already have been
-	   allocated. */
-	if (!alloc_gdt(cpu)) {
-		printk(KERN_CRIT "CPU%d failed to allocate GDT or PDA\n", cpu);
-		return 0;
-	}
+	struct desc_struct *gdt = per_cpu(cpu_gdt, cpu);
+	struct i386_pda *pda = &per_cpu(_cpu_pda, cpu);
 
-	gdt = (struct desc_struct *)cpu_gdt_descr->address;
-	pda = cpu_pda(cpu);
-
-	BUG_ON(gdt == NULL || pda == NULL);
-
-	/*
-	 * Initialize the per-CPU GDT with the boot GDT,
-	 * and set up the GDT descriptor:
-	 */
  	memcpy(gdt, cpu_gdt_table, GDT_SIZE);
+ 	cpu_gdt_descr->address = (unsigned long)gdt;
 	cpu_gdt_descr->size = GDT_SIZE - 1;
 
 	pack_descriptor((u32 *)&gdt[GDT_ENTRY_PDA].a,
@@ -706,17 +648,12 @@ __cpuinit int init_gdt(int cpu, struct task_struct *idle)
 	pda->_pda = pda;
 	pda->cpu_number = cpu;
 	pda->pcurrent = idle;
-
-	return 1;
 }
 
 void __cpuinit cpu_set_gdt(int cpu)
 {
 	struct Xgt_desc_struct *cpu_gdt_descr = &per_cpu(cpu_gdt_descr, cpu);
 
-	/* Reinit these anyway, even if they've already been done (on
-	   the boot CPU, this will transition from the boot gdt+pda to
-	   the real ones). */
 	load_gdt(cpu_gdt_descr);
 	set_kernel_fs();
 }
@@ -804,13 +741,8 @@ void __cpuinit cpu_init(void)
 	struct task_struct *curr = current;
 
 	/* Set up the real GDT and PDA, so we can transition from the
-	   boot versions. */
-	if (!init_gdt(cpu, curr)) {
-		/* failed to allocate something; not much we can do... */
-		for (;;)
-			local_irq_enable();
-	}
-
+	   boot_gdt_table & boot_pda. */
+	init_gdt(cpu, curr);
 	cpu_set_gdt(cpu);
 	_cpu_init(cpu, curr);
 }
-- 
cgit v1.2.3-59-g8ed1b


From bf50467204b435421d8de33ad080fa46c6f3d50b Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Wed, 2 May 2007 19:27:10 +0200
Subject: [PATCH] i386: Use per-cpu GDT immediately upon boot

Now we are no longer dynamically allocating the GDT, we don't need the
"cpu_gdt_table" at all: we can switch straight from "boot_gdt_table" to the
per-cpu GDT.  This means initializing the cpu_gdt array in C.

The boot CPU uses the per-cpu var directly, then in smp_prepare_cpus() it
switches to the per-cpu copy just allocated.  For secondary CPUs, the
early_gdt_descr is set to point directly to their per-cpu copy.

For UP the code is very simple: it keeps using the "per-cpu" GDT as per SMP,
but we never have to move.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---
 arch/i386/kernel/cpu/common.c        | 72 ++++++++++++++----------------------
 arch/i386/kernel/head.S              | 55 +--------------------------
 arch/i386/kernel/smpboot.c           | 59 +++++++++++++++++++++++------
 arch/i386/mach-voyager/voyager_smp.c |  6 ---
 include/asm-i386/desc.h              |  2 -
 include/asm-i386/processor.h         |  1 -
 6 files changed, 75 insertions(+), 120 deletions(-)

(limited to 'arch/i386/kernel/cpu/common.c')

diff --git a/arch/i386/kernel/cpu/common.c b/arch/i386/kernel/cpu/common.c
index 2335f4464ead..fd6b079f7609 100644
--- a/arch/i386/kernel/cpu/common.c
+++ b/arch/i386/kernel/cpu/common.c
@@ -25,7 +25,33 @@
 DEFINE_PER_CPU(struct Xgt_desc_struct, cpu_gdt_descr);
 EXPORT_PER_CPU_SYMBOL(cpu_gdt_descr);
 
-DEFINE_PER_CPU(struct desc_struct, cpu_gdt[GDT_ENTRIES]);
+DEFINE_PER_CPU(struct desc_struct, cpu_gdt[GDT_ENTRIES]) = {
+	[GDT_ENTRY_KERNEL_CS] = { 0x0000ffff, 0x00cf9a00 },
+	[GDT_ENTRY_KERNEL_DS] = { 0x0000ffff, 0x00cf9200 },
+	[GDT_ENTRY_DEFAULT_USER_CS] = { 0x0000ffff, 0x00cffa00 },
+	[GDT_ENTRY_DEFAULT_USER_DS] = { 0x0000ffff, 0x00cff200 },
+	/*
+	 * Segments used for calling PnP BIOS have byte granularity.
+	 * They code segments and data segments have fixed 64k limits,
+	 * the transfer segment sizes are set at run time.
+	 */
+	[GDT_ENTRY_PNPBIOS_CS32] = { 0x0000ffff, 0x00409a00 },/* 32-bit code */
+	[GDT_ENTRY_PNPBIOS_CS16] = { 0x0000ffff, 0x00009a00 },/* 16-bit code */
+	[GDT_ENTRY_PNPBIOS_DS] = { 0x0000ffff, 0x00009200 }, /* 16-bit data */
+	[GDT_ENTRY_PNPBIOS_TS1] = { 0x00000000, 0x00009200 },/* 16-bit data */
+	[GDT_ENTRY_PNPBIOS_TS2] = { 0x00000000, 0x00009200 },/* 16-bit data */
+	/*
+	 * The APM segments have byte granularity and their bases
+	 * are set at run time.  All have 64k limits.
+	 */
+	[GDT_ENTRY_APMBIOS_BASE] = { 0x0000ffff, 0x00409a00 },/* 32-bit code */
+	/* 16-bit code */
+	[GDT_ENTRY_APMBIOS_BASE+1] = { 0x0000ffff, 0x00009a00 },
+	[GDT_ENTRY_APMBIOS_BASE+2] = { 0x0000ffff, 0x00409200 }, /* data */
+
+	[GDT_ENTRY_ESPFIX_SS] = { 0x00000000, 0x00c09200 },
+	[GDT_ENTRY_PDA] = { 0x00000000, 0x00c09200 }, /* set in setup_pda */
+};
 
 DEFINE_PER_CPU(struct i386_pda, _cpu_pda);
 EXPORT_PER_CPU_SYMBOL(_cpu_pda);
@@ -618,46 +644,6 @@ struct i386_pda boot_pda = {
 	.pcurrent = &init_task,
 };
 
-static inline void set_kernel_fs(void)
-{
-	/* Set %fs for this CPU's PDA.  Memory clobber is to create a
-	   barrier with respect to any PDA operations, so the compiler
-	   doesn't move any before here. */
-	asm volatile ("mov %0, %%fs" : : "r" (__KERNEL_PDA) : "memory");
-}
-
-/* Initialize the CPU's GDT and PDA.  This is either the boot CPU doing itself
-   (still using cpu_gdt_table), or a CPU doing it for a secondary which
-   will soon come up. */
-__cpuinit void init_gdt(int cpu, struct task_struct *idle)
-{
-	struct Xgt_desc_struct *cpu_gdt_descr = &per_cpu(cpu_gdt_descr, cpu);
-	struct desc_struct *gdt = per_cpu(cpu_gdt, cpu);
-	struct i386_pda *pda = &per_cpu(_cpu_pda, cpu);
-
- 	memcpy(gdt, cpu_gdt_table, GDT_SIZE);
- 	cpu_gdt_descr->address = (unsigned long)gdt;
-	cpu_gdt_descr->size = GDT_SIZE - 1;
-
-	pack_descriptor((u32 *)&gdt[GDT_ENTRY_PDA].a,
-			(u32 *)&gdt[GDT_ENTRY_PDA].b,
-			(unsigned long)pda, sizeof(*pda) - 1,
-			0x80 | DESCTYPE_S | 0x2, 0); /* present read-write data segment */
-
-	memset(pda, 0, sizeof(*pda));
-	pda->_pda = pda;
-	pda->cpu_number = cpu;
-	pda->pcurrent = idle;
-}
-
-void __cpuinit cpu_set_gdt(int cpu)
-{
-	struct Xgt_desc_struct *cpu_gdt_descr = &per_cpu(cpu_gdt_descr, cpu);
-
-	load_gdt(cpu_gdt_descr);
-	set_kernel_fs();
-}
-
 /* Common CPU init for both boot and secondary CPUs */
 static void __cpuinit _cpu_init(int cpu, struct task_struct *curr)
 {
@@ -740,10 +726,6 @@ void __cpuinit cpu_init(void)
 	int cpu = smp_processor_id();
 	struct task_struct *curr = current;
 
-	/* Set up the real GDT and PDA, so we can transition from the
-	   boot_gdt_table & boot_pda. */
-	init_gdt(cpu, curr);
-	cpu_set_gdt(cpu);
 	_cpu_init(cpu, curr);
 }
 
diff --git a/arch/i386/kernel/head.S b/arch/i386/kernel/head.S
index cb185f40c282..633fd2f47429 100644
--- a/arch/i386/kernel/head.S
+++ b/arch/i386/kernel/head.S
@@ -599,7 +599,7 @@ idt_descr:
 	.word 0				# 32 bit align gdt_desc.address
 ENTRY(early_gdt_descr)
 	.word GDT_ENTRIES*8-1
-	.long cpu_gdt_table
+	.long per_cpu__cpu_gdt		/* Overwritten for secondary CPUs */
 
 /*
  * The boot_gdt_table must mirror the equivalent in setup.S and is
@@ -610,56 +610,3 @@ ENTRY(boot_gdt_table)
 	.fill GDT_ENTRY_BOOT_CS,8,0
 	.quad 0x00cf9a000000ffff	/* kernel 4GB code at 0x00000000 */
 	.quad 0x00cf92000000ffff	/* kernel 4GB data at 0x00000000 */
-
-/*
- * The Global Descriptor Table contains 32 quadwords, per-CPU.
- */
-	.align L1_CACHE_BYTES
-ENTRY(cpu_gdt_table)
-	.quad 0x0000000000000000	/* NULL descriptor */
-	.quad 0x0000000000000000	/* 0x0b reserved */
-	.quad 0x0000000000000000	/* 0x13 reserved */
-	.quad 0x0000000000000000	/* 0x1b reserved */
-	.quad 0x0000000000000000	/* 0x20 unused */
-	.quad 0x0000000000000000	/* 0x28 unused */
-	.quad 0x0000000000000000	/* 0x33 TLS entry 1 */
-	.quad 0x0000000000000000	/* 0x3b TLS entry 2 */
-	.quad 0x0000000000000000	/* 0x43 TLS entry 3 */
-	.quad 0x0000000000000000	/* 0x4b reserved */
-	.quad 0x0000000000000000	/* 0x53 reserved */
-	.quad 0x0000000000000000	/* 0x5b reserved */
-
-	.quad 0x00cf9a000000ffff	/* 0x60 kernel 4GB code at 0x00000000 */
-	.quad 0x00cf92000000ffff	/* 0x68 kernel 4GB data at 0x00000000 */
-	.quad 0x00cffa000000ffff	/* 0x73 user 4GB code at 0x00000000 */
-	.quad 0x00cff2000000ffff	/* 0x7b user 4GB data at 0x00000000 */
-
-	.quad 0x0000000000000000	/* 0x80 TSS descriptor */
-	.quad 0x0000000000000000	/* 0x88 LDT descriptor */
-
-	/*
-	 * Segments used for calling PnP BIOS have byte granularity.
-	 * The code segments and data segments have fixed 64k limits,
-	 * the transfer segment sizes are set at run time.
-	 */
-	.quad 0x00409a000000ffff	/* 0x90 32-bit code */
-	.quad 0x00009a000000ffff	/* 0x98 16-bit code */
-	.quad 0x000092000000ffff	/* 0xa0 16-bit data */
-	.quad 0x0000920000000000	/* 0xa8 16-bit data */
-	.quad 0x0000920000000000	/* 0xb0 16-bit data */
-
-	/*
-	 * The APM segments have byte granularity and their bases
-	 * are set at run time.  All have 64k limits.
-	 */
-	.quad 0x00409a000000ffff	/* 0xb8 APM CS    code */
-	.quad 0x00009a000000ffff	/* 0xc0 APM CS 16 code (16 bit) */
-	.quad 0x004092000000ffff	/* 0xc8 APM DS    data */
-
-	.quad 0x00c0920000000000	/* 0xd0 - ESPFIX SS */
-	.quad 0x00cf92000000ffff	/* 0xd8 - PDA */
-	.quad 0x0000000000000000	/* 0xe0 - unused */
-	.quad 0x0000000000000000	/* 0xe8 - unused */
-	.quad 0x0000000000000000	/* 0xf0 - unused */
-	.quad 0x0000000000000000	/* 0xf8 - GDT entry 31: double-fault TSS */
-
diff --git a/arch/i386/kernel/smpboot.c b/arch/i386/kernel/smpboot.c
index b36a5f174cc9..a9447c3e86dd 100644
--- a/arch/i386/kernel/smpboot.c
+++ b/arch/i386/kernel/smpboot.c
@@ -439,12 +439,6 @@ static void __cpuinit start_secondary(void *unused)
  */
 void __devinit initialize_secondary(void)
 {
-	/*
-	 * switch to the per CPU GDT we already set up
-	 * in do_boot_cpu()
-	 */
-	cpu_set_gdt(current_thread_info()->cpu);
-
 	/*
 	 * We don't actually need to load the full TSS,
 	 * basically just the stack pointer and the eip.
@@ -787,6 +781,32 @@ static inline struct task_struct * alloc_idle_task(int cpu)
 #define alloc_idle_task(cpu) fork_idle(cpu)
 #endif
 
+/* Initialize the CPU's GDT.  This is either the boot CPU doing itself
+   (still using the master per-cpu area), or a CPU doing it for a
+   secondary which will soon come up. */
+static __cpuinit void init_gdt(int cpu, struct task_struct *idle)
+{
+	struct Xgt_desc_struct *cpu_gdt_descr = &per_cpu(cpu_gdt_descr, cpu);
+	struct desc_struct *gdt = per_cpu(cpu_gdt, cpu);
+	struct i386_pda *pda = &per_cpu(_cpu_pda, cpu);
+
+ 	cpu_gdt_descr->address = (unsigned long)gdt;
+	cpu_gdt_descr->size = GDT_SIZE - 1;
+
+	pack_descriptor((u32 *)&gdt[GDT_ENTRY_PDA].a,
+			(u32 *)&gdt[GDT_ENTRY_PDA].b,
+			(unsigned long)pda, sizeof(*pda) - 1,
+			0x80 | DESCTYPE_S | 0x2, 0); /* present read-write data segment */
+
+	memset(pda, 0, sizeof(*pda));
+	pda->_pda = pda;
+	pda->cpu_number = cpu;
+	pda->pcurrent = idle;
+}
+
+/* Defined in head.S */
+extern struct Xgt_desc_struct early_gdt_descr;
+
 static int __cpuinit do_boot_cpu(int apicid, int cpu)
 /*
  * NOTE - on most systems this is a PHYSICAL apic ID, but on multiquad
@@ -809,6 +829,8 @@ static int __cpuinit do_boot_cpu(int apicid, int cpu)
 		panic("failed fork for CPU %d", cpu);
 
 	init_gdt(cpu, idle);
+	early_gdt_descr.address = (unsigned long)get_cpu_gdt_table(cpu);
+	start_pda = cpu_pda(cpu);
 
 	idle->thread.eip = (unsigned long) start_secondary;
 	/* start_eip had better be page-aligned! */
@@ -1161,13 +1183,26 @@ void __init smp_prepare_cpus(unsigned int max_cpus)
 	smp_boot_cpus(max_cpus);
 }
 
-void __devinit smp_prepare_boot_cpu(void)
+/* Current gdt points %fs at the "master" per-cpu area: after this,
+ * it's on the real one. */
+static inline void switch_to_new_gdt(void)
 {
-	cpu_set(smp_processor_id(), cpu_online_map);
-	cpu_set(smp_processor_id(), cpu_callout_map);
-	cpu_set(smp_processor_id(), cpu_present_map);
-	cpu_set(smp_processor_id(), cpu_possible_map);
-	per_cpu(cpu_state, smp_processor_id()) = CPU_ONLINE;
+	load_gdt(&per_cpu(cpu_gdt_descr, smp_processor_id()));
+	asm volatile ("mov %0, %%fs" : : "r" (__KERNEL_PDA) : "memory");
+}
+
+void __init smp_prepare_boot_cpu(void)
+{
+	unsigned int cpu = smp_processor_id();
+
+	init_gdt(cpu, current);
+	switch_to_new_gdt();
+
+	cpu_set(cpu, cpu_online_map);
+	cpu_set(cpu, cpu_callout_map);
+	cpu_set(cpu, cpu_present_map);
+	cpu_set(cpu, cpu_possible_map);
+	__get_cpu_var(cpu_state) = CPU_ONLINE;
 }
 
 #ifdef CONFIG_HOTPLUG_CPU
diff --git a/arch/i386/mach-voyager/voyager_smp.c b/arch/i386/mach-voyager/voyager_smp.c
index 15d8132d4f5e..b9ce33c0c202 100644
--- a/arch/i386/mach-voyager/voyager_smp.c
+++ b/arch/i386/mach-voyager/voyager_smp.c
@@ -764,12 +764,6 @@ initialize_secondary(void)
 	set_current(hard_get_current());
 #endif
 
-	/*
-	 * switch to the per CPU GDT we already set up
-	 * in do_boot_cpu()
-	 */
-	cpu_set_gdt(current_thread_info()->cpu);
-
 	/*
 	 * We don't actually need to load the full TSS,
 	 * basically just the stack pointer and the eip.
diff --git a/include/asm-i386/desc.h b/include/asm-i386/desc.h
index 53c5916687b6..a75ae6b97860 100644
--- a/include/asm-i386/desc.h
+++ b/include/asm-i386/desc.h
@@ -12,8 +12,6 @@
 
 #include <asm/mmu.h>
 
-extern struct desc_struct cpu_gdt_table[GDT_ENTRIES];
-
 struct Xgt_desc_struct {
 	unsigned short size;
 	unsigned long address __attribute__((packed));
diff --git a/include/asm-i386/processor.h b/include/asm-i386/processor.h
index cd940befef53..b25a2f5b5375 100644
--- a/include/asm-i386/processor.h
+++ b/include/asm-i386/processor.h
@@ -743,7 +743,6 @@ extern unsigned long boot_option_idle_override;
 extern void enable_sep_cpu(void);
 extern int sysenter_setup(void);
 
-extern void init_gdt(int cpu, struct task_struct *idle);
 extern void cpu_set_gdt(int);
 extern void secondary_cpu_init(void);
 
-- 
cgit v1.2.3-59-g8ed1b


From d2cbcc49e2bfd6eaa44d7e4e5e5f171aaa5ec80d Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Wed, 2 May 2007 19:27:10 +0200
Subject: [PATCH] i386: clean up cpu_init()

We now have cpu_init() and secondary_cpu_init() doing nothing but calling
_cpu_init() with the same arguments.  Rename _cpu_init() to cpu_init() and use
it as a replcement for secondary_cpu_init().

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---
 arch/i386/kernel/cpu/common.c | 34 +++++++++-------------------------
 arch/i386/kernel/smpboot.c    |  8 ++++----
 include/asm-i386/processor.h  |  2 +-
 3 files changed, 14 insertions(+), 30 deletions(-)

(limited to 'arch/i386/kernel/cpu/common.c')

diff --git a/arch/i386/kernel/cpu/common.c b/arch/i386/kernel/cpu/common.c
index fd6b079f7609..2a26956fce42 100644
--- a/arch/i386/kernel/cpu/common.c
+++ b/arch/i386/kernel/cpu/common.c
@@ -644,9 +644,16 @@ struct i386_pda boot_pda = {
 	.pcurrent = &init_task,
 };
 
-/* Common CPU init for both boot and secondary CPUs */
-static void __cpuinit _cpu_init(int cpu, struct task_struct *curr)
+/*
+ * cpu_init() initializes state that is per-CPU. Some data is already
+ * initialized (naturally) in the bootstrap process, such as the GDT
+ * and IDT. We reload them nevertheless, this function acts as a
+ * 'CPU state barrier', nothing should get across.
+ */
+void __cpuinit cpu_init(void)
 {
+	int cpu = smp_processor_id();
+	struct task_struct *curr = current;
 	struct tss_struct * t = &per_cpu(init_tss, cpu);
 	struct thread_struct *thread = &curr->thread;
 
@@ -706,29 +713,6 @@ static void __cpuinit _cpu_init(int cpu, struct task_struct *curr)
 	mxcsr_feature_mask_init();
 }
 
-/* Entrypoint to initialize secondary CPU */
-void __cpuinit secondary_cpu_init(void)
-{
-	int cpu = smp_processor_id();
-	struct task_struct *curr = current;
-
-	_cpu_init(cpu, curr);
-}
-
-/*
- * cpu_init() initializes state that is per-CPU. Some data is already
- * initialized (naturally) in the bootstrap process, such as the GDT
- * and IDT. We reload them nevertheless, this function acts as a
- * 'CPU state barrier', nothing should get across.
- */
-void __cpuinit cpu_init(void)
-{
-	int cpu = smp_processor_id();
-	struct task_struct *curr = current;
-
-	_cpu_init(cpu, curr);
-}
-
 #ifdef CONFIG_HOTPLUG_CPU
 void __cpuinit cpu_uninit(void)
 {
diff --git a/arch/i386/kernel/smpboot.c b/arch/i386/kernel/smpboot.c
index a9447c3e86dd..954245f6d307 100644
--- a/arch/i386/kernel/smpboot.c
+++ b/arch/i386/kernel/smpboot.c
@@ -378,14 +378,14 @@ set_cpu_sibling_map(int cpu)
 static void __cpuinit start_secondary(void *unused)
 {
 	/*
-	 * Don't put *anything* before secondary_cpu_init(), SMP
-	 * booting is too fragile that we want to limit the
-	 * things done here to the most necessary things.
+	 * Don't put *anything* before cpu_init(), SMP booting is too
+	 * fragile that we want to limit the things done here to the
+	 * most necessary things.
 	 */
 #ifdef CONFIG_VMI
 	vmi_bringup();
 #endif
-	secondary_cpu_init();
+	cpu_init();
 	preempt_disable();
 	smp_callin();
 	while (!cpu_isset(smp_processor_id(), smp_commenced_mask))
diff --git a/include/asm-i386/processor.h b/include/asm-i386/processor.h
index b25a2f5b5375..80f7e8a1e878 100644
--- a/include/asm-i386/processor.h
+++ b/include/asm-i386/processor.h
@@ -744,6 +744,6 @@ extern void enable_sep_cpu(void);
 extern int sysenter_setup(void);
 
 extern void cpu_set_gdt(int);
-extern void secondary_cpu_init(void);
+extern void cpu_init(void);
 
 #endif /* __ASM_I386_PROCESSOR_H */
-- 
cgit v1.2.3-59-g8ed1b


From 4fbb5968810b237e81977f131986b9efd5245368 Mon Sep 17 00:00:00 2001
From: Rusty Russell <rusty@rustcorp.com.au>
Date: Wed, 2 May 2007 19:27:11 +0200
Subject: [PATCH] i386: cleanup GDT Access

Now we have an explicit per-cpu GDT variable, we don't need to keep the
descriptors around to use them to find the GDT: expose cpu_gdt directly.

We could go further and make load_gdt() pack the descriptor for us, or even
assume it means "load the current cpu's GDT" which is what it always does.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---
 arch/i386/kernel/cpu/common.c |  4 +---
 arch/i386/kernel/efi.c        | 16 ++++++++--------
 arch/i386/kernel/entry.S      |  3 +--
 arch/i386/kernel/smpboot.c    | 12 ++++++------
 arch/i386/kernel/traps.c      |  4 +---
 include/asm-i386/desc.h       |  7 ++-----
 6 files changed, 19 insertions(+), 27 deletions(-)

(limited to 'arch/i386/kernel/cpu/common.c')

diff --git a/arch/i386/kernel/cpu/common.c b/arch/i386/kernel/cpu/common.c
index 2a26956fce42..5faf675aab4b 100644
--- a/arch/i386/kernel/cpu/common.c
+++ b/arch/i386/kernel/cpu/common.c
@@ -22,9 +22,6 @@
 
 #include "cpu.h"
 
-DEFINE_PER_CPU(struct Xgt_desc_struct, cpu_gdt_descr);
-EXPORT_PER_CPU_SYMBOL(cpu_gdt_descr);
-
 DEFINE_PER_CPU(struct desc_struct, cpu_gdt[GDT_ENTRIES]) = {
 	[GDT_ENTRY_KERNEL_CS] = { 0x0000ffff, 0x00cf9a00 },
 	[GDT_ENTRY_KERNEL_DS] = { 0x0000ffff, 0x00cf9200 },
@@ -52,6 +49,7 @@ DEFINE_PER_CPU(struct desc_struct, cpu_gdt[GDT_ENTRIES]) = {
 	[GDT_ENTRY_ESPFIX_SS] = { 0x00000000, 0x00c09200 },
 	[GDT_ENTRY_PDA] = { 0x00000000, 0x00c09200 }, /* set in setup_pda */
 };
+EXPORT_PER_CPU_SYMBOL_GPL(cpu_gdt);
 
 DEFINE_PER_CPU(struct i386_pda, _cpu_pda);
 EXPORT_PER_CPU_SYMBOL(_cpu_pda);
diff --git a/arch/i386/kernel/efi.c b/arch/i386/kernel/efi.c
index 8f9c624ace6f..dd9e7faafa7c 100644
--- a/arch/i386/kernel/efi.c
+++ b/arch/i386/kernel/efi.c
@@ -69,13 +69,11 @@ static void efi_call_phys_prelog(void) __acquires(efi_rt_lock)
 {
 	unsigned long cr4;
 	unsigned long temp;
-	struct Xgt_desc_struct *cpu_gdt_descr;
+	struct Xgt_desc_struct gdt_descr;
 
 	spin_lock(&efi_rt_lock);
 	local_irq_save(efi_rt_eflags);
 
-	cpu_gdt_descr = &per_cpu(cpu_gdt_descr, 0);
-
 	/*
 	 * If I don't have PSE, I should just duplicate two entries in page
 	 * directory. If I have PSE, I just need to duplicate one entry in
@@ -105,17 +103,19 @@ static void efi_call_phys_prelog(void) __acquires(efi_rt_lock)
 	 */
 	local_flush_tlb();
 
-	cpu_gdt_descr->address = __pa(cpu_gdt_descr->address);
-	load_gdt(cpu_gdt_descr);
+	gdt_descr.address = __pa(get_cpu_gdt_table(0));
+	gdt_descr.size = GDT_SIZE - 1;
+	load_gdt(&gdt_descr);
 }
 
 static void efi_call_phys_epilog(void) __releases(efi_rt_lock)
 {
 	unsigned long cr4;
-	struct Xgt_desc_struct *cpu_gdt_descr = &per_cpu(cpu_gdt_descr, 0);
+	struct Xgt_desc_struct gdt_descr;
 
-	cpu_gdt_descr->address = (unsigned long)__va(cpu_gdt_descr->address);
-	load_gdt(cpu_gdt_descr);
+	gdt_descr.address = (unsigned long)get_cpu_gdt_table(0);
+	gdt_descr.size = GDT_SIZE - 1;
+	load_gdt(&gdt_descr);
 
 	cr4 = read_cr4();
 
diff --git a/arch/i386/kernel/entry.S b/arch/i386/kernel/entry.S
index 922cc38dc405..c61c6b67e856 100644
--- a/arch/i386/kernel/entry.S
+++ b/arch/i386/kernel/entry.S
@@ -561,8 +561,7 @@ END(syscall_badsys)
 #define FIXUP_ESPFIX_STACK \
 	/* since we are on a wrong stack, we cant make it a C code :( */ \
 	movl %fs:PDA_cpu, %ebx; \
-	PER_CPU(cpu_gdt_descr, %ebx); \
-	movl GDS_address(%ebx), %ebx; \
+	PER_CPU(cpu_gdt, %ebx); \
 	GET_DESC_BASE(GDT_ENTRY_ESPFIX_SS, %ebx, %eax, %ax, %al, %ah); \
 	addl %esp, %eax; \
 	pushl $__KERNEL_DS; \
diff --git a/arch/i386/kernel/smpboot.c b/arch/i386/kernel/smpboot.c
index 954245f6d307..b0ad04d9ef21 100644
--- a/arch/i386/kernel/smpboot.c
+++ b/arch/i386/kernel/smpboot.c
@@ -786,13 +786,9 @@ static inline struct task_struct * alloc_idle_task(int cpu)
    secondary which will soon come up. */
 static __cpuinit void init_gdt(int cpu, struct task_struct *idle)
 {
-	struct Xgt_desc_struct *cpu_gdt_descr = &per_cpu(cpu_gdt_descr, cpu);
-	struct desc_struct *gdt = per_cpu(cpu_gdt, cpu);
+	struct desc_struct *gdt = get_cpu_gdt_table(cpu);
 	struct i386_pda *pda = &per_cpu(_cpu_pda, cpu);
 
- 	cpu_gdt_descr->address = (unsigned long)gdt;
-	cpu_gdt_descr->size = GDT_SIZE - 1;
-
 	pack_descriptor((u32 *)&gdt[GDT_ENTRY_PDA].a,
 			(u32 *)&gdt[GDT_ENTRY_PDA].b,
 			(unsigned long)pda, sizeof(*pda) - 1,
@@ -1187,7 +1183,11 @@ void __init smp_prepare_cpus(unsigned int max_cpus)
  * it's on the real one. */
 static inline void switch_to_new_gdt(void)
 {
-	load_gdt(&per_cpu(cpu_gdt_descr, smp_processor_id()));
+	struct Xgt_desc_struct gdt_descr;
+
+	gdt_descr.address = (long)get_cpu_gdt_table(smp_processor_id());
+	gdt_descr.size = GDT_SIZE - 1;
+	load_gdt(&gdt_descr);
 	asm volatile ("mov %0, %%fs" : : "r" (__KERNEL_PDA) : "memory");
 }
 
diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c
index 58dfecc8e36c..8722444cacaa 100644
--- a/arch/i386/kernel/traps.c
+++ b/arch/i386/kernel/traps.c
@@ -1030,9 +1030,7 @@ fastcall void do_spurious_interrupt_bug(struct pt_regs * regs,
 fastcall unsigned long patch_espfix_desc(unsigned long uesp,
 					  unsigned long kesp)
 {
-	int cpu = smp_processor_id();
-	struct Xgt_desc_struct *cpu_gdt_descr = &per_cpu(cpu_gdt_descr, cpu);
-	struct desc_struct *gdt = (struct desc_struct *)cpu_gdt_descr->address;
+	struct desc_struct *gdt = __get_cpu_var(cpu_gdt);
 	unsigned long base = (kesp - uesp) & -THREAD_SIZE;
 	unsigned long new_kesp = kesp - base;
 	unsigned long lim_pages = (new_kesp | (THREAD_SIZE - 1)) >> PAGE_SHIFT;
diff --git a/include/asm-i386/desc.h b/include/asm-i386/desc.h
index 13f701ea9a8f..4a974064e928 100644
--- a/include/asm-i386/desc.h
+++ b/include/asm-i386/desc.h
@@ -18,16 +18,13 @@ struct Xgt_desc_struct {
 	unsigned short pad;
 } __attribute__ ((packed));
 
-extern struct Xgt_desc_struct idt_descr;
-DECLARE_PER_CPU(struct Xgt_desc_struct, cpu_gdt_descr);
 DECLARE_PER_CPU(struct desc_struct, cpu_gdt[GDT_ENTRIES]);
-extern struct Xgt_desc_struct early_gdt_descr;
-
 static inline struct desc_struct *get_cpu_gdt_table(unsigned int cpu)
 {
-	return (struct desc_struct *)per_cpu(cpu_gdt_descr, cpu).address;
+	return per_cpu(cpu_gdt, cpu);
 }
 
+extern struct Xgt_desc_struct idt_descr;
 extern struct desc_struct idt_table[];
 extern void set_intr_gate(unsigned int irq, void * addr);
 
-- 
cgit v1.2.3-59-g8ed1b


From a6c4e076ee4c1ea670e4faa55814e63dd08e3f29 Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@goop.org>
Date: Wed, 2 May 2007 19:27:12 +0200
Subject: [PATCH] i386: clean up identify_cpu

identify_cpu() is used to identify both the boot CPU and secondary
CPUs, but it performs some actions which only apply to the boot CPU.
Those functions are therefore really __init functions, but because
they're called by identify_cpu(), they must be marked __cpuinit.

This patch splits identify_cpu() into identify_boot_cpu() and
identify_secondary_cpu(), and calls the appropriate init functions
from each.  Also, identify_boot_cpu() and all the functions it
dominates are marked __init.

Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/cpu/common.c | 21 ++++++++++++++-------
 arch/i386/kernel/smpboot.c    |  2 +-
 arch/i386/kernel/sysenter.c   |  2 +-
 include/asm-i386/processor.h  |  3 ++-
 4 files changed, 18 insertions(+), 10 deletions(-)

(limited to 'arch/i386/kernel/cpu/common.c')

diff --git a/arch/i386/kernel/cpu/common.c b/arch/i386/kernel/cpu/common.c
index 5faf675aab4b..58128585ae60 100644
--- a/arch/i386/kernel/cpu/common.c
+++ b/arch/i386/kernel/cpu/common.c
@@ -394,7 +394,7 @@ __setup("serialnumber", x86_serial_nr_setup);
 /*
  * This does the hard work of actually picking apart the CPU stuff...
  */
-void __cpuinit identify_cpu(struct cpuinfo_x86 *c)
+static void __cpuinit identify_cpu(struct cpuinfo_x86 *c)
 {
 	int i;
 
@@ -505,15 +505,22 @@ void __cpuinit identify_cpu(struct cpuinfo_x86 *c)
 
 	/* Init Machine Check Exception if available. */
 	mcheck_init(c);
+}
 
-	if (c == &boot_cpu_data)
-		sysenter_setup();
+void __init identify_boot_cpu(void)
+{
+	identify_cpu(&boot_cpu_data);
+	sysenter_setup();
 	enable_sep_cpu();
+	mtrr_bp_init();
+}
 
-	if (c == &boot_cpu_data)
-		mtrr_bp_init();
-	else
-		mtrr_ap_init();
+void __cpuinit identify_secondary_cpu(struct cpuinfo_x86 *c)
+{
+	BUG_ON(c == &boot_cpu_data);
+	identify_cpu(c);
+	enable_sep_cpu();
+	mtrr_ap_init();
 }
 
 #ifdef CONFIG_X86_HT
diff --git a/arch/i386/kernel/smpboot.c b/arch/i386/kernel/smpboot.c
index 1c3ad9b406ca..61e2842add36 100644
--- a/arch/i386/kernel/smpboot.c
+++ b/arch/i386/kernel/smpboot.c
@@ -155,7 +155,7 @@ static void __cpuinit smp_store_cpu_info(int id)
 
 	*c = boot_cpu_data;
 	if (id!=0)
-		identify_cpu(c);
+		identify_secondary_cpu(c);
 	/*
 	 * Mask B, Pentium, but not Pentium MMX
 	 */
diff --git a/arch/i386/kernel/sysenter.c b/arch/i386/kernel/sysenter.c
index 168f8147d3b4..13ca54a85a1c 100644
--- a/arch/i386/kernel/sysenter.c
+++ b/arch/i386/kernel/sysenter.c
@@ -72,7 +72,7 @@ extern const char vsyscall_int80_start, vsyscall_int80_end;
 extern const char vsyscall_sysenter_start, vsyscall_sysenter_end;
 static struct page *syscall_pages[1];
 
-int __cpuinit sysenter_setup(void)
+int __init sysenter_setup(void)
 {
 	void *syscall_page = (void *)get_zeroed_page(GFP_ATOMIC);
 	syscall_pages[0] = virt_to_page(syscall_page);
diff --git a/include/asm-i386/processor.h b/include/asm-i386/processor.h
index 11838df88603..9d895cc2f312 100644
--- a/include/asm-i386/processor.h
+++ b/include/asm-i386/processor.h
@@ -116,7 +116,8 @@ extern char ignore_fpu_irq;
 
 void __init cpu_detect(struct cpuinfo_x86 *c);
 
-extern void identify_cpu(struct cpuinfo_x86 *);
+extern void identify_boot_cpu(void);
+extern void identify_secondary_cpu(struct cpuinfo_x86 *);
 extern void print_cpu_info(struct cpuinfo_x86 *);
 extern unsigned int init_intel_cacheinfo(struct cpuinfo_x86 *c);
 extern unsigned short num_cache_leaves;
-- 
cgit v1.2.3-59-g8ed1b


From 7a61d35d4b4056e7711031202da7605e052f4137 Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@goop.org>
Date: Wed, 2 May 2007 19:27:15 +0200
Subject: [PATCH] i386: Page-align the GDT

Xen wants a dedicated page for the GDT.  I believe VMI likes it too.
lguest, KVM and native don't care.

Simple transformation to page-aligned "struct gdt_page".

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Andi Kleen <ak@suse.de>
Acked-by: Jeremy Fitzhardinge <jeremy@xensource.com>
---
 arch/i386/kernel/cpu/common.c | 6 +++---
 arch/i386/kernel/entry.S      | 2 +-
 arch/i386/kernel/head.S       | 2 +-
 arch/i386/kernel/traps.c      | 2 +-
 include/asm-i386/desc.h       | 9 +++++++--
 5 files changed, 13 insertions(+), 8 deletions(-)

(limited to 'arch/i386/kernel/cpu/common.c')

diff --git a/arch/i386/kernel/cpu/common.c b/arch/i386/kernel/cpu/common.c
index 58128585ae60..7a4c036d93c8 100644
--- a/arch/i386/kernel/cpu/common.c
+++ b/arch/i386/kernel/cpu/common.c
@@ -22,7 +22,7 @@
 
 #include "cpu.h"
 
-DEFINE_PER_CPU(struct desc_struct, cpu_gdt[GDT_ENTRIES]) = {
+DEFINE_PER_CPU(struct gdt_page, gdt_page) = { .gdt = {
 	[GDT_ENTRY_KERNEL_CS] = { 0x0000ffff, 0x00cf9a00 },
 	[GDT_ENTRY_KERNEL_DS] = { 0x0000ffff, 0x00cf9200 },
 	[GDT_ENTRY_DEFAULT_USER_CS] = { 0x0000ffff, 0x00cffa00 },
@@ -48,8 +48,8 @@ DEFINE_PER_CPU(struct desc_struct, cpu_gdt[GDT_ENTRIES]) = {
 
 	[GDT_ENTRY_ESPFIX_SS] = { 0x00000000, 0x00c09200 },
 	[GDT_ENTRY_PDA] = { 0x00000000, 0x00c09200 }, /* set in setup_pda */
-};
-EXPORT_PER_CPU_SYMBOL_GPL(cpu_gdt);
+} };
+EXPORT_PER_CPU_SYMBOL_GPL(gdt_page);
 
 DEFINE_PER_CPU(struct i386_pda, _cpu_pda);
 EXPORT_PER_CPU_SYMBOL(_cpu_pda);
diff --git a/arch/i386/kernel/entry.S b/arch/i386/kernel/entry.S
index e07473c0d3e7..3e4aa1fd33e2 100644
--- a/arch/i386/kernel/entry.S
+++ b/arch/i386/kernel/entry.S
@@ -557,7 +557,7 @@ END(syscall_badsys)
 #define FIXUP_ESPFIX_STACK \
 	/* since we are on a wrong stack, we cant make it a C code :( */ \
 	movl %fs:PDA_cpu, %ebx; \
-	PER_CPU(cpu_gdt, %ebx); \
+	PER_CPU(gdt_page, %ebx); \
 	GET_DESC_BASE(GDT_ENTRY_ESPFIX_SS, %ebx, %eax, %ax, %al, %ah); \
 	addl %esp, %eax; \
 	pushl $__KERNEL_DS; \
diff --git a/arch/i386/kernel/head.S b/arch/i386/kernel/head.S
index cc46494787e8..bb36c24311b4 100644
--- a/arch/i386/kernel/head.S
+++ b/arch/i386/kernel/head.S
@@ -598,7 +598,7 @@ idt_descr:
 	.word 0				# 32 bit align gdt_desc.address
 ENTRY(early_gdt_descr)
 	.word GDT_ENTRIES*8-1
-	.long per_cpu__cpu_gdt		/* Overwritten for secondary CPUs */
+	.long per_cpu__gdt_page		/* Overwritten for secondary CPUs */
 
 /*
  * The boot_gdt must mirror the equivalent in setup.S and is
diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c
index e0a23bee6967..f21b41e7770c 100644
--- a/arch/i386/kernel/traps.c
+++ b/arch/i386/kernel/traps.c
@@ -1030,7 +1030,7 @@ fastcall void do_spurious_interrupt_bug(struct pt_regs * regs,
 fastcall unsigned long patch_espfix_desc(unsigned long uesp,
 					  unsigned long kesp)
 {
-	struct desc_struct *gdt = __get_cpu_var(cpu_gdt);
+	struct desc_struct *gdt = __get_cpu_var(gdt_page).gdt;
 	unsigned long base = (kesp - uesp) & -THREAD_SIZE;
 	unsigned long new_kesp = kesp - base;
 	unsigned long lim_pages = (new_kesp | (THREAD_SIZE - 1)) >> PAGE_SHIFT;
diff --git a/include/asm-i386/desc.h b/include/asm-i386/desc.h
index 4a974064e928..c547403f341d 100644
--- a/include/asm-i386/desc.h
+++ b/include/asm-i386/desc.h
@@ -18,10 +18,15 @@ struct Xgt_desc_struct {
 	unsigned short pad;
 } __attribute__ ((packed));
 
-DECLARE_PER_CPU(struct desc_struct, cpu_gdt[GDT_ENTRIES]);
+struct gdt_page
+{
+	struct desc_struct gdt[GDT_ENTRIES];
+} __attribute__((aligned(PAGE_SIZE)));
+DECLARE_PER_CPU(struct gdt_page, gdt_page);
+
 static inline struct desc_struct *get_cpu_gdt_table(unsigned int cpu)
 {
-	return per_cpu(cpu_gdt, cpu);
+	return per_cpu(gdt_page, cpu).gdt;
 }
 
 extern struct Xgt_desc_struct idt_descr;
-- 
cgit v1.2.3-59-g8ed1b


From 7c3576d261ce046789a7db14f43303f8120910c7 Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@goop.org>
Date: Wed, 2 May 2007 19:27:16 +0200
Subject: [PATCH] i386: Convert PDA into the percpu section

Currently x86 (similar to x84-64) has a special per-cpu structure
called "i386_pda" which can be easily and efficiently referenced via
the %fs register.  An ELF section is more flexible than a structure,
allowing any piece of code to use this area.  Indeed, such a section
already exists: the per-cpu area.

So this patch:
(1) Removes the PDA and uses per-cpu variables for each current member.
(2) Replaces the __KERNEL_PDA segment with __KERNEL_PERCPU.
(3) Creates a per-cpu mirror of __per_cpu_offset called this_cpu_off, which
    can be used to calculate addresses for this CPU's variables.
(4) Simplifies startup, because %fs doesn't need to be loaded with a
    special segment at early boot; it can be deferred until the first
    percpu area is allocated (or never for UP).

The result is less code and one less x86-specific concept.

Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Andi Kleen <ak@suse.de>
---
 arch/i386/kernel/asm-offsets.c |   5 --
 arch/i386/kernel/cpu/common.c  |  17 +-----
 arch/i386/kernel/entry.S       |   5 +-
 arch/i386/kernel/head.S        |  31 ++--------
 arch/i386/kernel/i386_ksyms.c  |   2 -
 arch/i386/kernel/irq.c         |   3 +
 arch/i386/kernel/process.c     |  12 +++-
 arch/i386/kernel/smpboot.c     |  30 +++++-----
 arch/i386/kernel/vmi.c         |   6 +-
 arch/i386/kernel/vmlinux.lds.S |   1 -
 include/asm-i386/current.h     |   5 +-
 include/asm-i386/irq_regs.h    |  12 ++--
 include/asm-i386/pda.h         |  99 -------------------------------
 include/asm-i386/percpu.h      | 132 ++++++++++++++++++++++++++++++++++++++---
 include/asm-i386/processor.h   |   2 +-
 include/asm-i386/segment.h     |   6 +-
 include/asm-i386/smp.h         |   4 +-
 17 files changed, 177 insertions(+), 195 deletions(-)
 delete mode 100644 include/asm-i386/pda.h

(limited to 'arch/i386/kernel/cpu/common.c')

diff --git a/arch/i386/kernel/asm-offsets.c b/arch/i386/kernel/asm-offsets.c
index d558adfc293c..b05e85fd1c1e 100644
--- a/arch/i386/kernel/asm-offsets.c
+++ b/arch/i386/kernel/asm-offsets.c
@@ -15,7 +15,6 @@
 #include <asm/processor.h>
 #include <asm/thread_info.h>
 #include <asm/elf.h>
-#include <asm/pda.h>
 
 #define DEFINE(sym, val) \
         asm volatile("\n->" #sym " %0 " #val : : "i" (val))
@@ -101,10 +100,6 @@ void foo(void)
 
 	OFFSET(crypto_tfm_ctx_offset, crypto_tfm, __crt_ctx);
 
-	BLANK();
- 	OFFSET(PDA_cpu, i386_pda, cpu_number);
-	OFFSET(PDA_pcurrent, i386_pda, pcurrent);
-
 #ifdef CONFIG_PARAVIRT
 	BLANK();
 	OFFSET(PARAVIRT_enabled, paravirt_ops, paravirt_enabled);
diff --git a/arch/i386/kernel/cpu/common.c b/arch/i386/kernel/cpu/common.c
index 7a4c036d93c8..27e00565f5e4 100644
--- a/arch/i386/kernel/cpu/common.c
+++ b/arch/i386/kernel/cpu/common.c
@@ -18,7 +18,6 @@
 #include <asm/apic.h>
 #include <mach_apic.h>
 #endif
-#include <asm/pda.h>
 
 #include "cpu.h"
 
@@ -47,13 +46,10 @@ DEFINE_PER_CPU(struct gdt_page, gdt_page) = { .gdt = {
 	[GDT_ENTRY_APMBIOS_BASE+2] = { 0x0000ffff, 0x00409200 }, /* data */
 
 	[GDT_ENTRY_ESPFIX_SS] = { 0x00000000, 0x00c09200 },
-	[GDT_ENTRY_PDA] = { 0x00000000, 0x00c09200 }, /* set in setup_pda */
+	[GDT_ENTRY_PERCPU] = { 0x00000000, 0x00000000 },
 } };
 EXPORT_PER_CPU_SYMBOL_GPL(gdt_page);
 
-DEFINE_PER_CPU(struct i386_pda, _cpu_pda);
-EXPORT_PER_CPU_SYMBOL(_cpu_pda);
-
 static int cachesize_override __cpuinitdata = -1;
 static int disable_x86_fxsr __cpuinitdata;
 static int disable_x86_serial_nr __cpuinitdata = 1;
@@ -634,21 +630,14 @@ void __init early_cpu_init(void)
 #endif
 }
 
-/* Make sure %gs is initialized properly in idle threads */
+/* Make sure %fs is initialized properly in idle threads */
 struct pt_regs * __devinit idle_regs(struct pt_regs *regs)
 {
 	memset(regs, 0, sizeof(struct pt_regs));
-	regs->xfs = __KERNEL_PDA;
+	regs->xfs = __KERNEL_PERCPU;
 	return regs;
 }
 
-/* Initial PDA used by boot CPU */
-struct i386_pda boot_pda = {
-	._pda = &boot_pda,
-	.cpu_number = 0,
-	.pcurrent = &init_task,
-};
-
 /*
  * cpu_init() initializes state that is per-CPU. Some data is already
  * initialized (naturally) in the bootstrap process, such as the GDT
diff --git a/arch/i386/kernel/entry.S b/arch/i386/kernel/entry.S
index 3e4aa1fd33e2..7f92ceb428ad 100644
--- a/arch/i386/kernel/entry.S
+++ b/arch/i386/kernel/entry.S
@@ -132,7 +132,7 @@ VM_MASK		= 0x00020000
 	movl $(__USER_DS), %edx; \
 	movl %edx, %ds; \
 	movl %edx, %es; \
-	movl $(__KERNEL_PDA), %edx; \
+	movl $(__KERNEL_PERCPU), %edx; \
 	movl %edx, %fs
 
 #define RESTORE_INT_REGS \
@@ -556,7 +556,6 @@ END(syscall_badsys)
 
 #define FIXUP_ESPFIX_STACK \
 	/* since we are on a wrong stack, we cant make it a C code :( */ \
-	movl %fs:PDA_cpu, %ebx; \
 	PER_CPU(gdt_page, %ebx); \
 	GET_DESC_BASE(GDT_ENTRY_ESPFIX_SS, %ebx, %eax, %ax, %al, %ah); \
 	addl %esp, %eax; \
@@ -681,7 +680,7 @@ error_code:
 	pushl %fs
 	CFI_ADJUST_CFA_OFFSET 4
 	/*CFI_REL_OFFSET fs, 0*/
-	movl $(__KERNEL_PDA), %ecx
+	movl $(__KERNEL_PERCPU), %ecx
 	movl %ecx, %fs
 	UNWIND_ESPFIX_STACK
 	popl %ecx
diff --git a/arch/i386/kernel/head.S b/arch/i386/kernel/head.S
index bb36c24311b4..12277d8938df 100644
--- a/arch/i386/kernel/head.S
+++ b/arch/i386/kernel/head.S
@@ -317,12 +317,12 @@ is386:	movl $2,%ecx		# set MP
 	movl %eax,%cr0
 
 	call check_x87
-	call setup_pda
 	lgdt early_gdt_descr
 	lidt idt_descr
 	ljmp $(__KERNEL_CS),$1f
 1:	movl $(__KERNEL_DS),%eax	# reload all the segment registers
 	movl %eax,%ss			# after changing gdt.
+	movl %eax,%fs			# gets reset once there's real percpu
 
 	movl $(__USER_DS),%eax		# DS/ES contains default USER segment
 	movl %eax,%ds
@@ -332,16 +332,17 @@ is386:	movl $2,%ecx		# set MP
 	movl %eax,%gs
 	lldt %ax
 
-	movl $(__KERNEL_PDA),%eax
-	mov  %eax,%fs
-
 	cld			# gcc2 wants the direction flag cleared at all times
 	pushl $0		# fake return address for unwinder
 #ifdef CONFIG_SMP
 	movb ready, %cl
 	movb $1, ready
 	cmpb $0,%cl		# the first CPU calls start_kernel
-	jne initialize_secondary # all other CPUs call initialize_secondary
+	je   1f
+	movl $(__KERNEL_PERCPU), %eax
+	movl %eax,%fs		# set this cpu's percpu
+	jmp initialize_secondary # all other CPUs call initialize_secondary
+1:
 #endif /* CONFIG_SMP */
 	jmp start_kernel
 
@@ -364,23 +365,6 @@ check_x87:
 	.byte 0xDB,0xE4		/* fsetpm for 287, ignored by 387 */
 	ret
 
-/*
- * Point the GDT at this CPU's PDA.  On boot this will be
- * cpu_gdt_table and boot_pda; for secondary CPUs, these will be
- * that CPU's GDT and PDA.
- */
-ENTRY(setup_pda)
-	/* get the PDA pointer */
-	movl start_pda, %eax
-
-	/* slot the PDA address into the GDT */
-	mov early_gdt_descr+2, %ecx
-	mov %ax, (__KERNEL_PDA+0+2)(%ecx)		/* base & 0x0000ffff */
-	shr $16, %eax
-	mov %al, (__KERNEL_PDA+4+0)(%ecx)		/* base & 0x00ff0000 */
-	mov %ah, (__KERNEL_PDA+4+3)(%ecx)		/* base & 0xff000000 */
-	ret
-
 /*
  *  setup_idt
  *
@@ -553,9 +537,6 @@ ENTRY(empty_zero_page)
  * This starts the data section.
  */
 .data
-ENTRY(start_pda)
-	.long boot_pda
-
 ENTRY(stack_start)
 	.long init_thread_union+THREAD_SIZE
 	.long __BOOT_DS
diff --git a/arch/i386/kernel/i386_ksyms.c b/arch/i386/kernel/i386_ksyms.c
index 4afe26e86260..e3d4b73bfdb0 100644
--- a/arch/i386/kernel/i386_ksyms.c
+++ b/arch/i386/kernel/i386_ksyms.c
@@ -28,5 +28,3 @@ EXPORT_SYMBOL(__read_lock_failed);
 #endif
 
 EXPORT_SYMBOL(csum_partial);
-
-EXPORT_SYMBOL(_proxy_pda);
diff --git a/arch/i386/kernel/irq.c b/arch/i386/kernel/irq.c
index 8db8d514c9c0..d2daf672f4a2 100644
--- a/arch/i386/kernel/irq.c
+++ b/arch/i386/kernel/irq.c
@@ -24,6 +24,9 @@
 DEFINE_PER_CPU(irq_cpustat_t, irq_stat) ____cacheline_internodealigned_in_smp;
 EXPORT_PER_CPU_SYMBOL(irq_stat);
 
+DEFINE_PER_CPU(struct pt_regs *, irq_regs);
+EXPORT_PER_CPU_SYMBOL(irq_regs);
+
 /*
  * 'what should we do if we get a hw irq event on an illegal vector'.
  * each architecture has to answer this themselves.
diff --git a/arch/i386/kernel/process.c b/arch/i386/kernel/process.c
index 5fb9524c6f4b..61999479b7a4 100644
--- a/arch/i386/kernel/process.c
+++ b/arch/i386/kernel/process.c
@@ -39,6 +39,7 @@
 #include <linux/random.h>
 #include <linux/personality.h>
 #include <linux/tick.h>
+#include <linux/percpu.h>
 
 #include <asm/uaccess.h>
 #include <asm/pgtable.h>
@@ -57,7 +58,6 @@
 
 #include <asm/tlbflush.h>
 #include <asm/cpu.h>
-#include <asm/pda.h>
 
 asmlinkage void ret_from_fork(void) __asm__("ret_from_fork");
 
@@ -66,6 +66,12 @@ static int hlt_counter;
 unsigned long boot_option_idle_override = 0;
 EXPORT_SYMBOL(boot_option_idle_override);
 
+DEFINE_PER_CPU(struct task_struct *, current_task) = &init_task;
+EXPORT_PER_CPU_SYMBOL(current_task);
+
+DEFINE_PER_CPU(int, cpu_number);
+EXPORT_PER_CPU_SYMBOL(cpu_number);
+
 /*
  * Return saved PC of a blocked thread.
  */
@@ -342,7 +348,7 @@ int kernel_thread(int (*fn)(void *), void * arg, unsigned long flags)
 
 	regs.xds = __USER_DS;
 	regs.xes = __USER_DS;
-	regs.xfs = __KERNEL_PDA;
+	regs.xfs = __KERNEL_PERCPU;
 	regs.orig_eax = -1;
 	regs.eip = (unsigned long) kernel_thread_helper;
 	regs.xcs = __KERNEL_CS | get_kernel_rpl();
@@ -711,7 +717,7 @@ struct task_struct fastcall * __switch_to(struct task_struct *prev_p, struct tas
 	if (prev->gs | next->gs)
 		loadsegment(gs, next->gs);
 
-	write_pda(pcurrent, next_p);
+	x86_write_percpu(current_task, next_p);
 
 	return prev_p;
 }
diff --git a/arch/i386/kernel/smpboot.c b/arch/i386/kernel/smpboot.c
index 61e2842add36..f79b6233db78 100644
--- a/arch/i386/kernel/smpboot.c
+++ b/arch/i386/kernel/smpboot.c
@@ -53,7 +53,6 @@
 #include <asm/desc.h>
 #include <asm/arch_hooks.h>
 #include <asm/nmi.h>
-#include <asm/pda.h>
 
 #include <mach_apic.h>
 #include <mach_wakecpu.h>
@@ -99,6 +98,9 @@ EXPORT_SYMBOL(x86_cpu_to_apicid);
 
 u8 apicid_2_node[MAX_APICID];
 
+DEFINE_PER_CPU(unsigned long, this_cpu_off);
+EXPORT_PER_CPU_SYMBOL(this_cpu_off);
+
 /*
  * Trampoline 80x86 program as an array.
  */
@@ -456,7 +458,6 @@ extern struct {
 	void * esp;
 	unsigned short ss;
 } stack_start;
-extern struct i386_pda *start_pda;
 
 #ifdef CONFIG_NUMA
 
@@ -784,20 +785,17 @@ static inline struct task_struct * alloc_idle_task(int cpu)
 /* Initialize the CPU's GDT.  This is either the boot CPU doing itself
    (still using the master per-cpu area), or a CPU doing it for a
    secondary which will soon come up. */
-static __cpuinit void init_gdt(int cpu, struct task_struct *idle)
+static __cpuinit void init_gdt(int cpu)
 {
 	struct desc_struct *gdt = get_cpu_gdt_table(cpu);
-	struct i386_pda *pda = &per_cpu(_cpu_pda, cpu);
 
-	pack_descriptor((u32 *)&gdt[GDT_ENTRY_PDA].a,
-			(u32 *)&gdt[GDT_ENTRY_PDA].b,
-			(unsigned long)pda, sizeof(*pda) - 1,
-			0x80 | DESCTYPE_S | 0x2, 0); /* present read-write data segment */
+	pack_descriptor((u32 *)&gdt[GDT_ENTRY_PERCPU].a,
+			(u32 *)&gdt[GDT_ENTRY_PERCPU].b,
+			__per_cpu_offset[cpu], 0xFFFFF,
+			0x80 | DESCTYPE_S | 0x2, 0x8);
 
-	memset(pda, 0, sizeof(*pda));
-	pda->_pda = pda;
-	pda->cpu_number = cpu;
-	pda->pcurrent = idle;
+	per_cpu(this_cpu_off, cpu) = __per_cpu_offset[cpu];
+	per_cpu(cpu_number, cpu) = cpu;
 }
 
 /* Defined in head.S */
@@ -824,9 +822,9 @@ static int __cpuinit do_boot_cpu(int apicid, int cpu)
 	if (IS_ERR(idle))
 		panic("failed fork for CPU %d", cpu);
 
-	init_gdt(cpu, idle);
+	init_gdt(cpu);
+ 	per_cpu(current_task, cpu) = idle;
 	early_gdt_descr.address = (unsigned long)get_cpu_gdt_table(cpu);
-	start_pda = cpu_pda(cpu);
 
 	idle->thread.eip = (unsigned long) start_secondary;
 	/* start_eip had better be page-aligned! */
@@ -1188,14 +1186,14 @@ static inline void switch_to_new_gdt(void)
 	gdt_descr.address = (long)get_cpu_gdt_table(smp_processor_id());
 	gdt_descr.size = GDT_SIZE - 1;
 	load_gdt(&gdt_descr);
-	asm volatile ("mov %0, %%fs" : : "r" (__KERNEL_PDA) : "memory");
+	asm("mov %0, %%fs" : : "r" (__KERNEL_PERCPU) : "memory");
 }
 
 void __init native_smp_prepare_boot_cpu(void)
 {
 	unsigned int cpu = smp_processor_id();
 
-	init_gdt(cpu, current);
+	init_gdt(cpu);
 	switch_to_new_gdt();
 
 	cpu_set(cpu, cpu_online_map);
diff --git a/arch/i386/kernel/vmi.c b/arch/i386/kernel/vmi.c
index ccad7ee960aa..12312988c626 100644
--- a/arch/i386/kernel/vmi.c
+++ b/arch/i386/kernel/vmi.c
@@ -504,8 +504,6 @@ static void vmi_pmd_clear(pmd_t *pmd)
 #endif
 
 #ifdef CONFIG_SMP
-extern void setup_pda(void);
-
 static void __devinit
 vmi_startup_ipi_hook(int phys_apicid, unsigned long start_eip,
 		     unsigned long start_esp)
@@ -530,13 +528,11 @@ vmi_startup_ipi_hook(int phys_apicid, unsigned long start_eip,
 
 	ap.ds = __USER_DS;
 	ap.es = __USER_DS;
-	ap.fs = __KERNEL_PDA;
+	ap.fs = __KERNEL_PERCPU;
 	ap.gs = 0;
 
 	ap.eflags = 0;
 
-	setup_pda();
-
 #ifdef CONFIG_X86_PAE
 	/* efer should match BSP efer. */
 	if (cpu_has_nx) {
diff --git a/arch/i386/kernel/vmlinux.lds.S b/arch/i386/kernel/vmlinux.lds.S
index 2ce4aa185fc8..d125784ddf5e 100644
--- a/arch/i386/kernel/vmlinux.lds.S
+++ b/arch/i386/kernel/vmlinux.lds.S
@@ -26,7 +26,6 @@ OUTPUT_FORMAT("elf32-i386", "elf32-i386", "elf32-i386")
 OUTPUT_ARCH(i386)
 ENTRY(phys_startup_32)
 jiffies = jiffies_64;
-_proxy_pda = 1;
 
 PHDRS {
 	text PT_LOAD FLAGS(5);	/* R_E */
diff --git a/include/asm-i386/current.h b/include/asm-i386/current.h
index 5252ee0f6d7a..d35248539912 100644
--- a/include/asm-i386/current.h
+++ b/include/asm-i386/current.h
@@ -1,14 +1,15 @@
 #ifndef _I386_CURRENT_H
 #define _I386_CURRENT_H
 
-#include <asm/pda.h>
 #include <linux/compiler.h>
+#include <asm/percpu.h>
 
 struct task_struct;
 
+DECLARE_PER_CPU(struct task_struct *, current_task);
 static __always_inline struct task_struct *get_current(void)
 {
-	return read_pda(pcurrent);
+	return x86_read_percpu(current_task);
 }
  
 #define current get_current()
diff --git a/include/asm-i386/irq_regs.h b/include/asm-i386/irq_regs.h
index a1b3f7f594a2..3368b20c0b48 100644
--- a/include/asm-i386/irq_regs.h
+++ b/include/asm-i386/irq_regs.h
@@ -1,25 +1,27 @@
 /*
  * Per-cpu current frame pointer - the location of the last exception frame on
- * the stack, stored in the PDA.
+ * the stack, stored in the per-cpu area.
  *
  * Jeremy Fitzhardinge <jeremy@goop.org>
  */
 #ifndef _ASM_I386_IRQ_REGS_H
 #define _ASM_I386_IRQ_REGS_H
 
-#include <asm/pda.h>
+#include <asm/percpu.h>
+
+DECLARE_PER_CPU(struct pt_regs *, irq_regs);
 
 static inline struct pt_regs *get_irq_regs(void)
 {
-	return read_pda(irq_regs);
+	return x86_read_percpu(irq_regs);
 }
 
 static inline struct pt_regs *set_irq_regs(struct pt_regs *new_regs)
 {
 	struct pt_regs *old_regs;
 
-	old_regs = read_pda(irq_regs);
-	write_pda(irq_regs, new_regs);
+	old_regs = get_irq_regs();
+	x86_write_percpu(irq_regs, new_regs);
 
 	return old_regs;
 }
diff --git a/include/asm-i386/pda.h b/include/asm-i386/pda.h
deleted file mode 100644
index aef7f732f77e..000000000000
--- a/include/asm-i386/pda.h
+++ /dev/null
@@ -1,99 +0,0 @@
-/*
-   Per-processor Data Areas
-   Jeremy Fitzhardinge <jeremy@goop.org> 2006
-   Based on asm-x86_64/pda.h by Andi Kleen.
- */
-#ifndef _I386_PDA_H
-#define _I386_PDA_H
-
-#include <linux/stddef.h>
-#include <linux/types.h>
-#include <asm/percpu.h>
-
-struct i386_pda
-{
-	struct i386_pda *_pda;		/* pointer to self */
-
-	int cpu_number;
-	struct task_struct *pcurrent;	/* current process */
-	struct pt_regs *irq_regs;
-};
-
-DECLARE_PER_CPU(struct i386_pda, _cpu_pda);
-#define cpu_pda(i)	(&per_cpu(_cpu_pda, (i)))
-#define pda_offset(field) offsetof(struct i386_pda, field)
-
-extern void __bad_pda_field(void);
-
-/* This variable is never instantiated.  It is only used as a stand-in
-   for the real per-cpu PDA memory, so that gcc can understand what
-   memory operations the inline asms() below are performing.  This
-   eliminates the need to make the asms volatile or have memory
-   clobbers, so gcc can readily analyse them. */
-extern struct i386_pda _proxy_pda;
-
-#define pda_to_op(op,field,val)						\
-	do {								\
-		typedef typeof(_proxy_pda.field) T__;			\
-		if (0) { T__ tmp__; tmp__ = (val); }			\
-		switch (sizeof(_proxy_pda.field)) {			\
-		case 1:							\
-			asm(op "b %1,%%fs:%c2"				\
-			    : "+m" (_proxy_pda.field)			\
-			    :"ri" ((T__)val),				\
-			     "i"(pda_offset(field)));			\
-			break;						\
-		case 2:							\
-			asm(op "w %1,%%fs:%c2"				\
-			    : "+m" (_proxy_pda.field)			\
-			    :"ri" ((T__)val),				\
-			     "i"(pda_offset(field)));			\
-			break;						\
-		case 4:							\
-			asm(op "l %1,%%fs:%c2"				\
-			    : "+m" (_proxy_pda.field)			\
-			    :"ri" ((T__)val),				\
-			     "i"(pda_offset(field)));			\
-			break;						\
-		default: __bad_pda_field();				\
-		}							\
-	} while (0)
-
-#define pda_from_op(op,field)						\
-	({								\
-		typeof(_proxy_pda.field) ret__;				\
-		switch (sizeof(_proxy_pda.field)) {			\
-		case 1:							\
-			asm(op "b %%fs:%c1,%0"				\
-			    : "=r" (ret__)				\
-			    : "i" (pda_offset(field)),			\
-			      "m" (_proxy_pda.field));			\
-			break;						\
-		case 2:							\
-			asm(op "w %%fs:%c1,%0"				\
-			    : "=r" (ret__)				\
-			    : "i" (pda_offset(field)),			\
-			      "m" (_proxy_pda.field));			\
-			break;						\
-		case 4:							\
-			asm(op "l %%fs:%c1,%0"				\
-			    : "=r" (ret__)				\
-			    : "i" (pda_offset(field)),			\
-			      "m" (_proxy_pda.field));			\
-			break;						\
-		default: __bad_pda_field();				\
-		}							\
-		ret__; })
-
-/* Return a pointer to a pda field */
-#define pda_addr(field)							\
-	((typeof(_proxy_pda.field) *)((unsigned char *)read_pda(_pda) + \
-				      pda_offset(field)))
-
-#define read_pda(field) pda_from_op("mov",field)
-#define write_pda(field,val) pda_to_op("mov",field,val)
-#define add_pda(field,val) pda_to_op("add",field,val)
-#define sub_pda(field,val) pda_to_op("sub",field,val)
-#define or_pda(field,val) pda_to_op("or",field,val)
-
-#endif	/* _I386_PDA_H */
diff --git a/include/asm-i386/percpu.h b/include/asm-i386/percpu.h
index a10e7c68ae9d..c5f12f0d9c23 100644
--- a/include/asm-i386/percpu.h
+++ b/include/asm-i386/percpu.h
@@ -1,9 +1,30 @@
 #ifndef __ARCH_I386_PERCPU__
 #define __ARCH_I386_PERCPU__
 
-#ifndef __ASSEMBLY__
-#include <asm-generic/percpu.h>
-#else
+#ifdef __ASSEMBLY__
+
+/*
+ * PER_CPU finds an address of a per-cpu variable.
+ *
+ * Args:
+ *    var - variable name
+ *    reg - 32bit register
+ *
+ * The resulting address is stored in the "reg" argument.
+ *
+ * Example:
+ *    PER_CPU(cpu_gdt_descr, %ebx)
+ */
+#ifdef CONFIG_SMP
+#define PER_CPU(var, reg)			\
+	movl %fs:per_cpu__this_cpu_off, reg;		\
+	addl $per_cpu__##var, reg
+#else /* ! SMP */
+#define PER_CPU(var, reg) \
+	movl $per_cpu__##var, reg;
+#endif	/* SMP */
+
+#else /* ...!ASSEMBLY */
 
 /*
  * PER_CPU finds an address of a per-cpu variable.
@@ -18,14 +39,107 @@
  *    PER_CPU(cpu_gdt_descr, %ebx)
  */
 #ifdef CONFIG_SMP
-#define PER_CPU(var, cpu) \
-	movl __per_cpu_offset(,cpu,4), cpu;	\
-	addl $per_cpu__##var, cpu;
-#else /* ! SMP */
-#define PER_CPU(var, cpu) \
-	movl $per_cpu__##var, cpu;
+/* Same as generic implementation except for optimized local access. */
+#define __GENERIC_PER_CPU
+
+/* This is used for other cpus to find our section. */
+extern unsigned long __per_cpu_offset[];
+
+/* Separate out the type, so (int[3], foo) works. */
+#define DECLARE_PER_CPU(type, name) extern __typeof__(type) per_cpu__##name
+#define DEFINE_PER_CPU(type, name) \
+    __attribute__((__section__(".data.percpu"))) __typeof__(type) per_cpu__##name
+
+/* We can use this directly for local CPU (faster). */
+DECLARE_PER_CPU(unsigned long, this_cpu_off);
+
+/* var is in discarded region: offset to particular copy we want */
+#define per_cpu(var, cpu) (*({				\
+	extern int simple_indentifier_##var(void);	\
+	RELOC_HIDE(&per_cpu__##var, __per_cpu_offset[cpu]); }))
+
+#define __raw_get_cpu_var(var) (*({					\
+	extern int simple_indentifier_##var(void);			\
+	RELOC_HIDE(&per_cpu__##var, x86_read_percpu(this_cpu_off));	\
+}))
+
+#define __get_cpu_var(var) __raw_get_cpu_var(var)
+
+/* A macro to avoid #include hell... */
+#define percpu_modcopy(pcpudst, src, size)			\
+do {								\
+	unsigned int __i;					\
+	for_each_possible_cpu(__i)				\
+		memcpy((pcpudst)+__per_cpu_offset[__i],		\
+		       (src), (size));				\
+} while (0)
+
+#define EXPORT_PER_CPU_SYMBOL(var) EXPORT_SYMBOL(per_cpu__##var)
+#define EXPORT_PER_CPU_SYMBOL_GPL(var) EXPORT_SYMBOL_GPL(per_cpu__##var)
+
+/* fs segment starts at (positive) offset == __per_cpu_offset[cpu] */
+#define __percpu_seg "%%fs:"
+#else  /* !SMP */
+#include <asm-generic/percpu.h>
+#define __percpu_seg ""
 #endif	/* SMP */
 
+/* For arch-specific code, we can use direct single-insn ops (they
+ * don't give an lvalue though). */
+extern void __bad_percpu_size(void);
+
+#define percpu_to_op(op,var,val)				\
+	do {							\
+		typedef typeof(var) T__;			\
+		if (0) { T__ tmp__; tmp__ = (val); }		\
+		switch (sizeof(var)) {				\
+		case 1:						\
+			asm(op "b %1,"__percpu_seg"%0"		\
+			    : "+m" (var)			\
+			    :"ri" ((T__)val));			\
+			break;					\
+		case 2:						\
+			asm(op "w %1,"__percpu_seg"%0"		\
+			    : "+m" (var)			\
+			    :"ri" ((T__)val));			\
+			break;					\
+		case 4:						\
+			asm(op "l %1,"__percpu_seg"%0"		\
+			    : "+m" (var)			\
+			    :"ri" ((T__)val));			\
+			break;					\
+		default: __bad_percpu_size();			\
+		}						\
+	} while (0)
+
+#define percpu_from_op(op,var)					\
+	({							\
+		typeof(var) ret__;				\
+		switch (sizeof(var)) {				\
+		case 1:						\
+			asm(op "b "__percpu_seg"%1,%0"		\
+			    : "=r" (ret__)			\
+			    : "m" (var));			\
+			break;					\
+		case 2:						\
+			asm(op "w "__percpu_seg"%1,%0"		\
+			    : "=r" (ret__)			\
+			    : "m" (var));			\
+			break;					\
+		case 4:						\
+			asm(op "l "__percpu_seg"%1,%0"		\
+			    : "=r" (ret__)			\
+			    : "m" (var));			\
+			break;					\
+		default: __bad_percpu_size();			\
+		}						\
+		ret__; })
+
+#define x86_read_percpu(var) percpu_from_op("mov", per_cpu__##var)
+#define x86_write_percpu(var,val) percpu_to_op("mov", per_cpu__##var, val)
+#define x86_add_percpu(var,val) percpu_to_op("add", per_cpu__##var, val)
+#define x86_sub_percpu(var,val) percpu_to_op("sub", per_cpu__##var, val)
+#define x86_or_percpu(var,val) percpu_to_op("or", per_cpu__##var, val)
 #endif /* !__ASSEMBLY__ */
 
 #endif /* __ARCH_I386_PERCPU__ */
diff --git a/include/asm-i386/processor.h b/include/asm-i386/processor.h
index 922260474646..ced2da8a0d65 100644
--- a/include/asm-i386/processor.h
+++ b/include/asm-i386/processor.h
@@ -377,7 +377,7 @@ struct thread_struct {
 	.vm86_info = NULL,						\
 	.sysenter_cs = __KERNEL_CS,					\
 	.io_bitmap_ptr = NULL,						\
-	.fs = __KERNEL_PDA,						\
+	.fs = __KERNEL_PERCPU,						\
 }
 
 /*
diff --git a/include/asm-i386/segment.h b/include/asm-i386/segment.h
index 065f10bfa487..07e70624d87c 100644
--- a/include/asm-i386/segment.h
+++ b/include/asm-i386/segment.h
@@ -39,7 +39,7 @@
  *  25 - APM BIOS support 
  *
  *  26 - ESPFIX small SS
- *  27 - PDA				[ per-cpu private data area ]
+ *  27 - per-cpu			[ offset to per-cpu data area ]
  *  28 - unused
  *  29 - unused
  *  30 - unused
@@ -74,8 +74,8 @@
 #define GDT_ENTRY_ESPFIX_SS		(GDT_ENTRY_KERNEL_BASE + 14)
 #define __ESPFIX_SS (GDT_ENTRY_ESPFIX_SS * 8)
 
-#define GDT_ENTRY_PDA			(GDT_ENTRY_KERNEL_BASE + 15)
-#define __KERNEL_PDA (GDT_ENTRY_PDA * 8)
+#define GDT_ENTRY_PERCPU			(GDT_ENTRY_KERNEL_BASE + 15)
+#define __KERNEL_PERCPU (GDT_ENTRY_PERCPU * 8)
 
 #define GDT_ENTRY_DOUBLEFAULT_TSS	31
 
diff --git a/include/asm-i386/smp.h b/include/asm-i386/smp.h
index 2d083cb4ca93..090abc1da32a 100644
--- a/include/asm-i386/smp.h
+++ b/include/asm-i386/smp.h
@@ -8,7 +8,6 @@
 #include <linux/kernel.h>
 #include <linux/threads.h>
 #include <linux/cpumask.h>
-#include <asm/pda.h>
 #endif
 
 #if defined(CONFIG_X86_LOCAL_APIC) && !defined(__ASSEMBLY__)
@@ -112,7 +111,8 @@ do { } while (0)
  * from the initial startup. We map APIC_BASE very early in page_setup(),
  * so this is correct in the x86 case.
  */
-#define raw_smp_processor_id() (read_pda(cpu_number))
+DECLARE_PER_CPU(int, cpu_number);
+#define raw_smp_processor_id() (x86_read_percpu(cpu_number))
 
 extern cpumask_t cpu_callout_map;
 extern cpumask_t cpu_callin_map;
-- 
cgit v1.2.3-59-g8ed1b


From c5413fbe894924ddb8aa474a4d4da52e7a6c7e0b Mon Sep 17 00:00:00 2001
From: Jeremy Fitzhardinge <jeremy@goop.org>
Date: Wed, 2 May 2007 19:27:16 +0200
Subject: [PATCH] i386: Fix UP gdt bugs

Fixes two problems with the GDT when compiling for uniprocessor:
 - There's no percpu segment, so trying to load its selector into %fs fails.
   Use a null selector instead.
 - The real gdt needs to be loaded at some point.  Do it in cpu_init().

Signed-off-by: Chris Wright <chrisw@sous-sol.org>
Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com>
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Rusty Russell <rusty@rustcorp.com.au>
---
 arch/i386/kernel/cpu/common.c | 13 +++++++++++++
 arch/i386/kernel/smpboot.c    | 12 ------------
 include/asm-i386/processor.h  |  1 +
 include/asm-i386/segment.h    |  4 ++++
 4 files changed, 18 insertions(+), 12 deletions(-)

(limited to 'arch/i386/kernel/cpu/common.c')

diff --git a/arch/i386/kernel/cpu/common.c b/arch/i386/kernel/cpu/common.c
index 27e00565f5e4..794d593c47eb 100644
--- a/arch/i386/kernel/cpu/common.c
+++ b/arch/i386/kernel/cpu/common.c
@@ -638,6 +638,18 @@ struct pt_regs * __devinit idle_regs(struct pt_regs *regs)
 	return regs;
 }
 
+/* Current gdt points %fs at the "master" per-cpu area: after this,
+ * it's on the real one. */
+void switch_to_new_gdt(void)
+{
+	struct Xgt_desc_struct gdt_descr;
+
+	gdt_descr.address = (long)get_cpu_gdt_table(smp_processor_id());
+	gdt_descr.size = GDT_SIZE - 1;
+	load_gdt(&gdt_descr);
+	asm("mov %0, %%fs" : : "r" (__KERNEL_PERCPU) : "memory");
+}
+
 /*
  * cpu_init() initializes state that is per-CPU. Some data is already
  * initialized (naturally) in the bootstrap process, such as the GDT
@@ -668,6 +680,7 @@ void __cpuinit cpu_init(void)
 	}
 
 	load_idt(&idt_descr);
+	switch_to_new_gdt();
 
 	/*
 	 * Set up and load the per-CPU TSS and LDT
diff --git a/arch/i386/kernel/smpboot.c b/arch/i386/kernel/smpboot.c
index f79b6233db78..7c1dbef399cd 100644
--- a/arch/i386/kernel/smpboot.c
+++ b/arch/i386/kernel/smpboot.c
@@ -1177,18 +1177,6 @@ void __init native_smp_prepare_cpus(unsigned int max_cpus)
 	smp_boot_cpus(max_cpus);
 }
 
-/* Current gdt points %fs at the "master" per-cpu area: after this,
- * it's on the real one. */
-static inline void switch_to_new_gdt(void)
-{
-	struct Xgt_desc_struct gdt_descr;
-
-	gdt_descr.address = (long)get_cpu_gdt_table(smp_processor_id());
-	gdt_descr.size = GDT_SIZE - 1;
-	load_gdt(&gdt_descr);
-	asm("mov %0, %%fs" : : "r" (__KERNEL_PERCPU) : "memory");
-}
-
 void __init native_smp_prepare_boot_cpu(void)
 {
 	unsigned int cpu = smp_processor_id();
diff --git a/include/asm-i386/processor.h b/include/asm-i386/processor.h
index ced2da8a0d65..70f3515c3db0 100644
--- a/include/asm-i386/processor.h
+++ b/include/asm-i386/processor.h
@@ -750,6 +750,7 @@ extern void enable_sep_cpu(void);
 extern int sysenter_setup(void);
 
 extern void cpu_set_gdt(int);
+extern void switch_to_new_gdt(void);
 extern void cpu_init(void);
 
 extern int force_mwait;
diff --git a/include/asm-i386/segment.h b/include/asm-i386/segment.h
index 07e70624d87c..597a47c2515f 100644
--- a/include/asm-i386/segment.h
+++ b/include/asm-i386/segment.h
@@ -75,7 +75,11 @@
 #define __ESPFIX_SS (GDT_ENTRY_ESPFIX_SS * 8)
 
 #define GDT_ENTRY_PERCPU			(GDT_ENTRY_KERNEL_BASE + 15)
+#ifdef CONFIG_SMP
 #define __KERNEL_PERCPU (GDT_ENTRY_PERCPU * 8)
+#else
+#define __KERNEL_PERCPU 0
+#endif
 
 #define GDT_ENTRY_DOUBLEFAULT_TSS	31
 
-- 
cgit v1.2.3-59-g8ed1b