x86, perf: Fix static_key bug in load_mm_cr4()

Mikulas reported his K6-3 not booting. This is because the
static_key API confusion struck and bit Andy, this wants to be
static_key_false().

Reported-by: Mikulas Patocka <[e-mail]>
Tested-by: Mikulas Patocka <[e-mail]>
Signed-off-by: Peter Zijlstra (Intel) <[e-mail]>
Cc: <[e-mail]>
Cc: Andrea Arcangeli <[e-mail]>
Cc: Andy Lutomirski <[e-mail]>
Cc: Arnaldo Carvalho de Melo <[e-mail]>
Cc: Borislav Petkov <[e-mail]>
Cc: Brian Gerst <[e-mail]>
Cc: Denys Vlasenko <[e-mail]>
Cc: H. Peter Anvin <[e-mail]>
Cc: Kees Cook <[e-mail]>
Cc: Linus Torvalds <[e-mail]>
Cc: Paul Mackerras <[e-mail]>
Cc: Peter Zijlstra <[e-mail]>
Cc: Thomas Gleixner <[e-mail]>
Cc: Valdis Kletnieks <[e-mail]>
Cc: Vince Weaver <[e-mail]>
Cc: hillf.zj <[e-mail]>
Fixes: a66734297f78 ("perf/x86: Add /sys/devices/cpu/rdpmc=2 to allow rdpmc for all tasks")
Link: http://lkml.kernel.org/r/[e-mail]
Signed-off-by: Ingo Molnar <[e-mail]>

arch/x86/include/asm/mmu_context.h

···
 
 static inline void load_mm_cr4(struct mm_struct *mm)
 {
-	if (static_key_true(&rdpmc_always_available) ||
+	if (static_key_false(&rdpmc_always_available) ||
 	    atomic_read(&mm->context.perf_rdpmc_allowed))
 		cr4_set_bits(X86_CR4_PCE);
 	else