---
 include/asm-x86/percpu.h |   30 ++++++++++++++++++------------
 1 file changed, 18 insertions(+), 12 deletions(-)

Index: linux-2.6/include/asm-x86/percpu.h
===================================================================
--- linux-2.6.orig/include/asm-x86/percpu.h	2007-11-27 21:07:50.600463718 -0800
+++ linux-2.6/include/asm-x86/percpu.h	2007-11-27 21:20:32.368963117 -0800
@@ -11,9 +11,14 @@
 
 #ifdef CONFIG_SMP
 #include <asm/pda.h>
-
 #define __my_cpu_offset read_pda(data_offset)
+#define __percpu_seg "%%gs:"
+/* Calculate the offset to use with the segment register */
+#define per_cpu_offset(name)	*SHIFT_PTR(&per_cpu_var(name), -__per_cpu_start)
 
+#else
+#define __percpu_seg ""
+#define per_cpu_offset(name)	per_cpu_var(name)
 #endif
 #include <asm-generic/percpu.h>
 
@@ -61,16 +66,11 @@ DECLARE_PER_CPU(struct x8664_pda, pda);
  *    PER_CPU(cpu_gdt_descr, %ebx)
  */
 #ifdef CONFIG_SMP
-
 #define __my_cpu_offset x86_read_percpu(this_cpu_off)
-
 /* fs segment starts at (positive) offset == __per_cpu_offset[cpu] */
 #define __percpu_seg "%%fs:"
-
 #else  /* !SMP */
-
 #define __percpu_seg ""
-
 #endif	/* SMP */
 
 #include <asm-generic/percpu.h>
@@ -78,6 +78,13 @@ DECLARE_PER_CPU(struct x8664_pda, pda);
 /* We can use this directly for local CPU (faster). */
 DECLARE_PER_CPU(unsigned long, this_cpu_off);
 
+#define per_cpu_offset(name)	per_cpu_var(name)
+
+#endif /* __ASSEMBLY__ */
+#endif /* !CONFIG_X86_64 */
+
+#ifndef __ASSEMBLY__
+
 /* For arch-specific code, we can use direct single-insn ops (they
  * don't give an lvalue though). */
 extern void __bad_percpu_size(void);
@@ -129,11 +136,10 @@ extern void __bad_percpu_size(void);
 		}						\
 		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)
+#define x86_read_percpu(var) percpu_from_op("mov", per_cpu_offset(var))
+#define x86_write_percpu(var,val) percpu_to_op("mov", per_cpu_offset(var), val)
+#define x86_add_percpu(var,val) percpu_to_op("add", per_cpu_offset(var), val)
+#define x86_sub_percpu(var,val) percpu_to_op("sub", per_cpu_offset(var), val)
+#define x86_or_percpu(var,val) percpu_to_op("or", per_cpu_offset(var), val)
 #endif /* !__ASSEMBLY__ */
-#endif /* !CONFIG_X86_64 */
 #endif /* _ASM_X86_PERCPU_H_ */