[PATCH 4/9] x86, perf: Explicitly optimize vm86 handling in code_segment_base

[Andy Lutomirski]

There's no point in checking the VM bit on 64-bit, and, since we're
explicitly checking it, we can use user_mode_ignore_vm86 after the
check.

While we're at it, rearrange the ifdef slightly to make the code
flow a bit clearer.

Cc: Peter Zijlstra <a.p.zijlstra@xxxxxxxxx>
Signed-off-by: Andy Lutomirski <luto@xxxxxxxxxx>
---
 arch/x86/kernel/cpu/perf_event.c | 13 +++++++------
 1 file changed, 7 insertions(+), 6 deletions(-)

diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c
index 979963bb3977..56f7e60ad732 100644
--- a/arch/x86/kernel/cpu/perf_event.c
+++ b/arch/x86/kernel/cpu/perf_event.c
@@ -2147,18 +2147,19 @@ perf_callchain_user(struct perf_callchain_entry *entry, struct pt_regs *regs)
 static unsigned long code_segment_base(struct pt_regs *regs)
 {
 	/*
+	 * For IA32 we look at the GDT/LDT segment base to convert the
+	 * effective IP to a linear address.
+	 */
+
+#ifdef CONFIG_X86_32
+	/*
 	 * If we are in VM86 mode, add the segment offset to convert to a
 	 * linear address.
 	 */
 	if (regs->flags & X86_VM_MASK)
 		return 0x10 * regs->cs;
 
-	/*
-	 * For IA32 we look at the GDT/LDT segment base to convert the
-	 * effective IP to a linear address.
-	 */
-#ifdef CONFIG_X86_32
-	if (user_mode(regs) && regs->cs != __USER_CS)
+	if (user_mode_ignore_vm86(regs) && regs->cs != __USER_CS)
 		return get_segment_base(regs->cs);
 #else
 	if (user_mode(regs) && !user_64bit_mode(regs) &&
-- 
2.3.0

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/