x86: pmu: Improve LLC misses event verification

When running pmu test on SPR, sometimes the following failure is
reported.

  1 <= 0 <= 1000000
  FAIL: Intel: llc misses-4

Currently The LLC misses occurring only depends on probability. It's
possible that there is no LLC misses happened in the whole loop(),
especially along with processors have larger and larger cache size just
like what we observed on SPR.

Thus, add clflush instruction into the loop() asm blob and ensure once
LLC miss is triggered at least.

Suggested-by: Jim Mattson <jmattson@google.com>
Signed-off-by: Dapeng Mi <dapeng1.mi@linux.intel.com>
Link: https://lore.kernel.org/r/20250215013636.1214612-15-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>

Author: Dapeng Mi

x86/pmu.c

@@ -19,19 +19,30 @@
 #define EXPECTED_INSTR 17
 #define EXPECTED_BRNCH 5
 
-/* Enable GLOBAL_CTRL + disable GLOBAL_CTRL instructions */
-#define EXTRA_INSNS  (3 + 3)
+/* Enable GLOBAL_CTRL + disable GLOBAL_CTRL + clflush/mfence instructions */
+#define EXTRA_INSNS  (3 + 3 +2)
 #define LOOP_INSNS   (N * 10 + EXTRA_INSNS)
 #define LOOP_BRANCHES  (N)
-#define LOOP_ASM(_wrmsr)						\
+#define LOOP_ASM(_wrmsr, _clflush)					\
 	_wrmsr "\n\t"							\
 	"mov %%ecx, %%edi; mov %%ebx, %%ecx;\n\t"			\
+	_clflush "\n\t"                                 		\
+	"mfence;\n\t"                                   		\
 	"1: mov (%1), %2; add $64, %1;\n\t"				\
 	"nop; nop; nop; nop; nop; nop; nop;\n\t"			\
 	"loop 1b;\n\t"							\
 	"mov %%edi, %%ecx; xor %%eax, %%eax; xor %%edx, %%edx;\n\t"	\
 	_wrmsr "\n\t"
 
+#define _loop_asm(_wrmsr, _clflush)				\
+do {								\
+	asm volatile(LOOP_ASM(_wrmsr, _clflush)			\
+		     : "=b"(tmp), "=r"(tmp2), "=r"(tmp3)	\
+		     : "a"(eax), "d"(edx), "c"(global_ctl),	\
+		       "0"(N), "1"(buf)				\
+		     : "edi");					\
+} while (0)
+
 typedef struct {
 	uint32_t ctr;
 	uint32_t idx;
@@ -88,14 +99,17 @@ static struct pmu_event *gp_events;
 static unsigned int gp_events_size;
 static unsigned int fixed_counters_num;
 
-
 static inline void __loop(void)
 {
 	unsigned long tmp, tmp2, tmp3;
+	u32 global_ctl = 0;
+	u32 eax = 0;
+	u32 edx = 0;
 
-	asm volatile(LOOP_ASM("nop")
-		     : "=c"(tmp), "=r"(tmp2), "=r"(tmp3)
-		     : "0"(N), "1"(buf));
+	if (this_cpu_has(X86_FEATURE_CLFLUSH))
+		_loop_asm("nop", "clflush (%1)");
+	else
+		_loop_asm("nop", "nop");
 }
 
 /*
@@ -108,15 +122,14 @@ static inline void __loop(void)
 static inline void __precise_loop(u64 cntrs)
 {
 	unsigned long tmp, tmp2, tmp3;
-	unsigned int global_ctl = pmu.msr_global_ctl;
+	u32 global_ctl = pmu.msr_global_ctl;
 	u32 eax = cntrs & (BIT_ULL(32) - 1);
 	u32 edx = cntrs >> 32;
 
-	asm volatile(LOOP_ASM("wrmsr")
-		     : "=b"(tmp), "=r"(tmp2), "=r"(tmp3)
-		     : "a"(eax), "d"(edx), "c"(global_ctl),
-		       "0"(N), "1"(buf)
-		     : "edi");
+	if (this_cpu_has(X86_FEATURE_CLFLUSH))
+		_loop_asm("wrmsr", "clflush (%1)");
+	else
+		_loop_asm("wrmsr", "nop");
 }
 
 static inline void loop(u64 cntrs)