[PATCH] Add the vgetcpu vsyscall

This patch adds a vgetcpu vsyscall, which depending on the CPU RDTSCP
capability uses either the RDTSCP or CPUID to obtain a CPU and node
numbers and pass them to the program.

AK: Lots of changes over Vojtech's original code:
Better prototype for vgetcpu()
It's better to pass the cpu / node numbers as separate arguments
to avoid mistakes when going from SMP to NUMA.
Also add a fast time stamp based cache using a user supplied
argument to speed things more up.
Use fast method from Chuck Ebbert to retrieve node/cpu from
GDT limit instead of CPUID
Made sure RDTSCP init is always executed after node is known.
Drop printk

Signed-off-by: Vojtech Pavlik <vojtech@suse.cz>
Signed-off-by: Andi Kleen <ak@suse.de>

4 files changed, 37 insertions, 5 deletions

diff --git a/include/asm-x86_64/segment.h b/include/asm-x86_64/segment.h
index d4bed33fb32c..334ddcdd8f92 100644
--- a/include/asm-x86_64/segment.h
+++ b/include/asm-x86_64/segment.h
@@ -20,15 +20,16 @@
 #define __USER_CS     0x33   /* 6*8+3 */ 
 #define __USER32_DS     __USER_DS 
 
-#define GDT_ENTRY_TLS 1
 #define GDT_ENTRY_TSS 8 /* needs two entries */
 #define GDT_ENTRY_LDT 10 /* needs two entries */
 #define GDT_ENTRY_TLS_MIN 12
 #define GDT_ENTRY_TLS_MAX 14
-/* 15 free */
 
 #define GDT_ENTRY_TLS_ENTRIES 3
 
+#define GDT_ENTRY_PER_CPU 15    /* Abused to load per CPU data from limit */
+#define __PER_CPU_SEG   (GDT_ENTRY_PER_CPU * 8 + 3)
+
 /* TLS indexes for 64bit - hardcoded in arch_prctl */
 #define FS_TLS 0        
 #define GS_TLS 1        
diff --git a/include/asm-x86_64/smp.h b/include/asm-x86_64/smp.h
index 6805e1feb300..d61547fd833b 100644
--- a/include/asm-x86_64/smp.h
+++ b/include/asm-x86_64/smp.h
@@ -133,13 +133,19 @@ static __inline int logical_smp_processor_id(void)
         /* we don't want to mark this access volatile - bad code generation */
         return GET_APIC_LOGICAL_ID(*(unsigned long *)(APIC_BASE+APIC_LDR));
 }
-#endif
 
 #ifdef CONFIG_SMP
 #define cpu_physical_id(cpu)            x86_cpu_to_apicid[cpu]
 #else
 #define cpu_physical_id(cpu)            boot_cpu_id
-#endif
-
+static inline int smp_call_function_single(int cpuid, void (*func) (void *info),
+                                void *info, int retry, int wait)
+{
+        /* Disable interrupts here? */
+        func(info);
+        return 0;
+}
+#endif /* !CONFIG_SMP */
+#endif /* !__ASSEMBLY */
 #endif
 
diff --git a/include/asm-x86_64/vsyscall.h b/include/asm-x86_64/vsyscall.h
index 146b24402a5f..2281e9399b96 100644
--- a/include/asm-x86_64/vsyscall.h
+++ b/include/asm-x86_64/vsyscall.h
@@ -4,6 +4,7 @@
 enum vsyscall_num {
         __NR_vgettimeofday,
         __NR_vtime,
+        __NR_vgetcpu,
 };
 
 #define VSYSCALL_START (-10UL << 20)
@@ -15,6 +16,7 @@ enum vsyscall_num {
 #include <linux/seqlock.h>
 
 #define __section_vxtime __attribute__ ((unused, __section__ (".vxtime"), aligned(16)))
+#define __section_vgetcpu_mode __attribute__ ((unused, __section__ (".vgetcpu_mode"), aligned(16)))
 #define __section_wall_jiffies __attribute__ ((unused, __section__ (".wall_jiffies"), aligned(16)))
 #define __section_jiffies __attribute__ ((unused, __section__ (".jiffies"), aligned(16)))
 #define __section_sys_tz __attribute__ ((unused, __section__ (".sys_tz"), aligned(16)))
@@ -26,6 +28,9 @@ enum vsyscall_num {
 #define VXTIME_HPET     2
 #define VXTIME_PMTMR    3
 
+#define VGETCPU_RDTSCP  1
+#define VGETCPU_LSL     2
+
 struct vxtime_data {
         long hpet_address;      /* HPET base address */
         int last;
@@ -40,6 +45,7 @@ struct vxtime_data {
 
 /* vsyscall space (readonly) */
 extern struct vxtime_data __vxtime;
+extern int __vgetcpu_mode;
 extern struct timespec __xtime;
 extern volatile unsigned long __jiffies;
 extern unsigned long __wall_jiffies;
@@ -48,6 +54,7 @@ extern seqlock_t __xtime_lock;
 
 /* kernel space (writeable) */
 extern struct vxtime_data vxtime;
+extern int vgetcpu_mode;
 extern unsigned long wall_jiffies;
 extern struct timezone sys_tz;
 extern int sysctl_vsyscall;
@@ -55,6 +62,8 @@ extern seqlock_t xtime_lock;
 
 extern int sysctl_vsyscall;
 
+extern void vsyscall_set_cpu(int cpu);
+
 #define ARCH_HAVE_XTIME_LOCK 1
 
 #endif /* __KERNEL__ */
diff --git a/include/linux/getcpu.h b/include/linux/getcpu.h
new file mode 100644
index 000000000000..031ed3780e45
--- /dev/null
+++ b/include/linux/getcpu.h
@@ -0,0 +1,16 @@
+#ifndef _LINUX_GETCPU_H
+#define _LINUX_GETCPU_H 1
+
+/* Cache for getcpu() to speed it up. Results might be upto a jiffie
+   out of date, but will be faster.
+   User programs should not refer to the contents of this structure.
+   It is only a cache for vgetcpu(). It might change in future kernels.
+   The user program must store this information per thread (__thread)
+   If you want 100% accurate information pass NULL instead. */
+struct getcpu_cache {
+        unsigned long t0;
+        unsigned long t1;
+        unsigned long res[4];
+};
+
+#endif