[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>

7 files changed, 139 insertions, 121 deletions

diff --git a/include/asm-i386/current.h b/include/asm-i386/current.h
index 5252ee0..d352485 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 a1b3f7f..3368b20 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 aef7f73..0000000
--- 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 a10e7c6..c5f12f0 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 9222604..ced2da8 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 065f10b..07e7062 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 2d083cb..090abc1 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;