64-bit doesn't use the entry for per CPU data, but for CPU
(and node) numbers. The change will clarify the real usage
of this entry in GDT.
Suggested-by: H. Peter Anvin <h...@zytor.com>
Signed-off-by: Chang S. Bae <chang.seok....@intel.com>
Acked-by: Andy Lutomirski <l...@kernel.org>
Reviewed-by: Thomas Gleixner <t...@linutronix.de>
Cc: Ingo Molnar <mi...@kernel.org>
Cc: Andi Kleen <a...@linux.intel.com>
Cc: Dave Hansen <dave.han...@linux.intel.com>
---
arch/x86/entry/vdso/vma.c | 2 +-
arch/x86/include/asm/segment.h | 5 ++---
arch/x86/include/asm/vgtod.h | 8 ++++----
3 files changed, 7 insertions(+), 8 deletions(-)
diff --git a/arch/x86/entry/vdso/vma.c b/arch/x86/entry/vdso/vma.c
index 5b8b556..0b114aa 100644
--- a/arch/x86/entry/vdso/vma.c
+++ b/arch/x86/entry/vdso/vma.c
@@ -359,7 +359,7 @@ static void vgetcpu_cpu_init(void *arg)
d.p = 1; /* Present */
d.d = 1; /* 32-bit */
- write_gdt_entry(get_cpu_gdt_rw(cpu), GDT_ENTRY_PER_CPU, &d, DESCTYPE_S);
+ write_gdt_entry(get_cpu_gdt_rw(cpu), GDT_ENTRY_CPU_NUMBER, &d,
DESCTYPE_S);
}
static int vgetcpu_online(unsigned int cpu)
diff --git a/arch/x86/include/asm/segment.h b/arch/x86/include/asm/segment.h
index 0ffbe95..3cb2aa5 100644
--- a/arch/x86/include/asm/segment.h
+++ b/arch/x86/include/asm/segment.h
@@ -186,8 +186,7 @@
#define GDT_ENTRY_TLS_MIN 12
#define GDT_ENTRY_TLS_MAX 14
-/* Abused to load per CPU data from limit */
-#define GDT_ENTRY_PER_CPU 15
+#define GDT_ENTRY_CPU_NUMBER 15
/*
* Number of entries in the GDT table:
@@ -207,7 +206,7 @@
#define __USER_DS (GDT_ENTRY_DEFAULT_USER_DS*8 + 3)
#define __USER32_DS __USER_DS
#define __USER_CS (GDT_ENTRY_DEFAULT_USER_CS*8 + 3)
-#define __PER_CPU_SEG (GDT_ENTRY_PER_CPU*8 + 3)
+#define __CPU_NUMBER_SEG (GDT_ENTRY_CPU_NUMBER*8 + 3)
#endif
diff --git a/arch/x86/include/asm/vgtod.h b/arch/x86/include/asm/vgtod.h
index 5374854..4e81ea9 100644
--- a/arch/x86/include/asm/vgtod.h
+++ b/arch/x86/include/asm/vgtod.h
@@ -86,9 +86,9 @@ static inline unsigned int __getcpu(void)
unsigned int p;
/*
- * Load per CPU data from GDT. LSL is faster than RDTSCP and
- * works on all CPUs. This is volatile so that it orders
- * correctly wrt barrier() and to keep gcc from cleverly
+ * Load CPU (and node) number from GDT. LSL is faster than RDTSCP
+ * and works on all CPUs. This is volatile so that it orders
+ * correctly with respect to barrier() and to keep GCC from cleverly
* hoisting it out of the calling function.
*
* If RDPID is available, use it.
@@ -96,7 +96,7 @@ static inline unsigned int __getcpu(void)
alternative_io ("lsl %[seg],%[p]",
".byte 0xf3,0x0f,0xc7,0xf8", /* RDPID %eax/rax */
X86_FEATURE_RDPID,
- [p] "=a" (p), [seg] "r" (__PER_CPU_SEG));
+ [p] "=a" (p), [seg] "r" (__CPU_NUMBER_SEG));
return p;
}
--
2.7.4
