1 files changed, 137 insertions, 0 deletions
diff --git a/arch/x86/include/asm/pda.h b/arch/x86/include/asm/pda.h
new file mode 100644
index 000000000000..80860afffbdb
--- /dev/null
+++ b/arch/x86/include/asm/pda.h
@@ -0,0 +1,137 @@
+#ifndef ASM_X86__PDA_H
+#define ASM_X86__PDA_H
+#ifndef __ASSEMBLY__
+#include <linux/stddef.h>
+#include <linux/types.h>
+#include <linux/cache.h>
+#include <asm/page.h>
+/* Per processor datastructure. %gs points to it while the kernel runs */
+struct x8664_pda {
+        struct task_struct *pcurrent;   /* 0  Current process */
+        unsigned long data_offset;      /* 8 Per cpu data offset from linker
+                                           address */
+        unsigned long kernelstack;      /* 16 top of kernel stack for current */
+        unsigned long oldrsp;           /* 24 user rsp for system call */
+        int irqcount;                   /* 32 Irq nesting counter. Starts -1 */
+        unsigned int cpunumber;         /* 36 Logical CPU number */
+#ifdef CONFIG_CC_STACKPROTECTOR
+        unsigned long stack_canary;     /* 40 stack canary value */
+                                        /* gcc-ABI: this canary MUST be at
+                                           offset 40!!! */
+#endif
+        char *irqstackptr;
+        short nodenumber;               /* number of current node (32k max) */
+        short in_bootmem;               /* pda lives in bootmem */
+        unsigned int __softirq_pending;
+        unsigned int __nmi_count;       /* number of NMI on this CPUs */
+        short mmu_state;
+        short isidle;
+        struct mm_struct *active_mm;
+        unsigned apic_timer_irqs;
+        unsigned irq0_irqs;
+        unsigned irq_resched_count;
+        unsigned irq_call_count;
+        unsigned irq_tlb_count;
+        unsigned irq_thermal_count;
+        unsigned irq_threshold_count;
+        unsigned irq_spurious_count;
+} ____cacheline_aligned_in_smp;
+extern struct x8664_pda **_cpu_pda;
+extern void pda_init(int);
+#define cpu_pda(i) (_cpu_pda[i])
+/*
+ * There is no fast way to get the base address of the PDA, all the accesses
+ * have to mention %fs/%gs.  So it needs to be done this Torvaldian way.
+ */
+extern void __bad_pda_field(void) __attribute__((noreturn));
+/*
+ * proxy_pda doesn't actually exist, but tell gcc it is accessed for
+ * all PDA accesses so it gets read/write dependencies right.
+ */
+extern struct x8664_pda _proxy_pda;
+#define pda_offset(field) offsetof(struct x8664_pda, field)
+#define pda_to_op(op, field, val)                                       \
+do {                                                                    \
+        typedef typeof(_proxy_pda.field) T__;                           \
+        if (0) { T__ tmp__; tmp__ = (val); }    /* type checking */     \
+        switch (sizeof(_proxy_pda.field)) {                             \
+        case 2:                                                         \
+                asm(op "w %1,%%gs:%c2" :                                \
+                    "+m" (_proxy_pda.field) :                           \
+                    "ri" ((T__)val),                                    \
+                    "i"(pda_offset(field)));                            \
+                break;                                                  \
+        case 4:                                                         \
+                asm(op "l %1,%%gs:%c2" :                                \
+                    "+m" (_proxy_pda.field) :                           \
+                    "ri" ((T__)val),                                    \
+                    "i" (pda_offset(field)));                           \
+                break;                                                  \
+        case 8:                                                         \
+                asm(op "q %1,%%gs:%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 2:                                 \
+                asm(op "w %%gs:%c1,%0" :        \
+                    "=r" (ret__) :              \
+                    "i" (pda_offset(field)),    \
+                    "m" (_proxy_pda.field));    \
+                break;                          \
+        case 4:                                 \
+                asm(op "l %%gs:%c1,%0":         \
+                    "=r" (ret__):               \
+                    "i" (pda_offset(field)),    \
+                    "m" (_proxy_pda.field));    \
+                break;                          \
+        case 8:                                 \
+                asm(op "q %%gs:%c1,%0":         \
+                    "=r" (ret__) :              \
+                    "i" (pda_offset(field)),    \
+                    "m" (_proxy_pda.field));    \
+                break;                          \
+        default:                                \
+                __bad_pda_field();              \
+        }                                       \
+        ret__;                                  \
+})
+#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)
+/* This is not atomic against other CPUs -- CPU preemption needs to be off */
+#define test_and_clear_bit_pda(bit, field)                              \
+({                                                                      \
+        int old__;                                                      \
+        asm volatile("btr %2,%%gs:%c3\n\tsbbl %0,%0"                    \
+                     : "=r" (old__), "+m" (_proxy_pda.field)            \
+                     : "dIr" (bit), "i" (pda_offset(field)) : "memory");\
+        old__;                                                          \
+})
+#endif
+#define PDA_STACKOFFSET (5*8)
+#endif /* ASM_X86__PDA_H */

diff --git a/arch/x86/include/asm/pda.h b/arch/x86/include/asm/pda.h new file mode 100644 index 000000000000..80860afffbdb --- /dev/null +++ b/arch/x86/include/asm/pda.h
@@ -0,0 +1,137 @@
	1	#ifndef ASM_X86__PDA_H
	2	#define ASM_X86__PDA_H
	3
	4	#ifndef __ASSEMBLY__
	5	#include <linux/stddef.h>
	6	#include <linux/types.h>
	7	#include <linux/cache.h>
	8	#include <asm/page.h>
	9
	10	/* Per processor datastructure. %gs points to it while the kernel runs */
	11	struct x8664_pda {
	12	struct task_struct pcurrent; / 0 Current process */
	13	unsigned long data_offset; /* 8 Per cpu data offset from linker
	14	address */
	15	unsigned long kernelstack; /* 16 top of kernel stack for current */
	16	unsigned long oldrsp; /* 24 user rsp for system call */
	17	int irqcount; /* 32 Irq nesting counter. Starts -1 */
	18	unsigned int cpunumber; /* 36 Logical CPU number */
	19	#ifdef CONFIG_CC_STACKPROTECTOR
	20	unsigned long stack_canary; /* 40 stack canary value */
	21	/* gcc-ABI: this canary MUST be at
	22	offset 40!!! */
	23	#endif
	24	char *irqstackptr;
	25	short nodenumber; /* number of current node (32k max) */
	26	short in_bootmem; /* pda lives in bootmem */
	27	unsigned int __softirq_pending;
	28	unsigned int __nmi_count; /* number of NMI on this CPUs */
	29	short mmu_state;
	30	short isidle;
	31	struct mm_struct *active_mm;
	32	unsigned apic_timer_irqs;
	33	unsigned irq0_irqs;
	34	unsigned irq_resched_count;
	35	unsigned irq_call_count;
	36	unsigned irq_tlb_count;
	37	unsigned irq_thermal_count;
	38	unsigned irq_threshold_count;
	39	unsigned irq_spurious_count;
	40	} ____cacheline_aligned_in_smp;
	41
	42	extern struct x8664_pda **_cpu_pda;
	43	extern void pda_init(int);
	44
	45	#define cpu_pda(i) (_cpu_pda[i])
	46
	47	/*
	48	* There is no fast way to get the base address of the PDA, all the accesses
	49	* have to mention %fs/%gs. So it needs to be done this Torvaldian way.
	50	*/
	51	extern void __bad_pda_field(void) __attribute__((noreturn));
	52
	53	/*
	54	* proxy_pda doesn't actually exist, but tell gcc it is accessed for
	55	* all PDA accesses so it gets read/write dependencies right.
	56	*/
	57	extern struct x8664_pda _proxy_pda;
	58
	59	#define pda_offset(field) offsetof(struct x8664_pda, field)
	60
	61	#define pda_to_op(op, field, val) \
	62	do { \
	63	typedef typeof(_proxy_pda.field) T__; \
	64	if (0) { T__ tmp__; tmp__ = (val); } /* type checking */ \
	65	switch (sizeof(_proxy_pda.field)) { \
	66	case 2: \
	67	asm(op "w %1,%%gs:%c2" : \
	68	"+m" (_proxy_pda.field) : \
	69	"ri" ((T__)val), \
	70	"i"(pda_offset(field))); \
	71	break; \
	72	case 4: \
	73	asm(op "l %1,%%gs:%c2" : \
	74	"+m" (_proxy_pda.field) : \
	75	"ri" ((T__)val), \
	76	"i" (pda_offset(field))); \
	77	break; \
	78	case 8: \
	79	asm(op "q %1,%%gs:%c2": \
	80	"+m" (_proxy_pda.field) : \
	81	"ri" ((T__)val), \
	82	"i"(pda_offset(field))); \
	83	break; \
	84	default: \
	85	__bad_pda_field(); \
	86	} \
	87	} while (0)
	88
	89	#define pda_from_op(op, field) \
	90	({ \
	91	typeof(_proxy_pda.field) ret__; \
	92	switch (sizeof(_proxy_pda.field)) { \
	93	case 2: \
	94	asm(op "w %%gs:%c1,%0" : \
	95	"=r" (ret__) : \
	96	"i" (pda_offset(field)), \
	97	"m" (_proxy_pda.field)); \
	98	break; \
	99	case 4: \
	100	asm(op "l %%gs:%c1,%0": \
	101	"=r" (ret__): \
	102	"i" (pda_offset(field)), \
	103	"m" (_proxy_pda.field)); \
	104	break; \
	105	case 8: \
	106	asm(op "q %%gs:%c1,%0": \
	107	"=r" (ret__) : \
	108	"i" (pda_offset(field)), \
	109	"m" (_proxy_pda.field)); \
	110	break; \
	111	default: \
	112	__bad_pda_field(); \
	113	} \
	114	ret__; \
	115	})
	116
	117	#define read_pda(field) pda_from_op("mov", field)
	118	#define write_pda(field, val) pda_to_op("mov", field, val)
	119	#define add_pda(field, val) pda_to_op("add", field, val)
	120	#define sub_pda(field, val) pda_to_op("sub", field, val)
	121	#define or_pda(field, val) pda_to_op("or", field, val)
	122
	123	/* This is not atomic against other CPUs -- CPU preemption needs to be off */
	124	#define test_and_clear_bit_pda(bit, field) \
	125	({ \
	126	int old__; \
	127	asm volatile("btr %2,%%gs:%c3\n\tsbbl %0,%0" \
	128	: "=r" (old__), "+m" (_proxy_pda.field) \
	129	: "dIr" (bit), "i" (pda_offset(field)) : "memory");\
	130	old__; \
	131	})
	132
	133	#endif
	134
	135	#define PDA_STACKOFFSET (5*8)
	136
	137	#endif /* ASM_X86__PDA_H */