ARM: perf: add support for stalled cycle ABI events

Commit 8f622422 ("perf events: Add generic front-end and back-end
stalled cycle event definitions") added two new ABI events for counting
stalled cycles.

This patch adds support for these new events to the ARM perf
implementation.

Cc: Jamie Iles <jamie@jamieiles.com>
Cc: Jean Pihet <j-pihet@ti.com>
Signed-off-by: Will Deacon <will.deacon@arm.com>

3 files changed, 66 insertions, 49 deletions

diff --git a/arch/arm/kernel/perf_event_v6.c b/arch/arm/kernel/perf_event_v6.c
index e63d811..533be99 100644
--- a/arch/arm/kernel/perf_event_v6.c
+++ b/arch/arm/kernel/perf_event_v6.c
@@ -65,13 +65,15 @@ enum armv6_counters {
  * accesses/misses in hardware.
  */
 static const unsigned armv6_perf_map[PERF_COUNT_HW_MAX] = {
-	[PERF_COUNT_HW_CPU_CYCLES]	    = ARMV6_PERFCTR_CPU_CYCLES,
-	[PERF_COUNT_HW_INSTRUCTIONS]	    = ARMV6_PERFCTR_INSTR_EXEC,
-	[PERF_COUNT_HW_CACHE_REFERENCES]    = HW_OP_UNSUPPORTED,
-	[PERF_COUNT_HW_CACHE_MISSES]	    = HW_OP_UNSUPPORTED,
-	[PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = ARMV6_PERFCTR_BR_EXEC,
-	[PERF_COUNT_HW_BRANCH_MISSES]	    = ARMV6_PERFCTR_BR_MISPREDICT,
-	[PERF_COUNT_HW_BUS_CYCLES]	    = HW_OP_UNSUPPORTED,
+	[PERF_COUNT_HW_CPU_CYCLES]		= ARMV6_PERFCTR_CPU_CYCLES,
+	[PERF_COUNT_HW_INSTRUCTIONS]		= ARMV6_PERFCTR_INSTR_EXEC,
+	[PERF_COUNT_HW_CACHE_REFERENCES]	= HW_OP_UNSUPPORTED,
+	[PERF_COUNT_HW_CACHE_MISSES]		= HW_OP_UNSUPPORTED,
+	[PERF_COUNT_HW_BRANCH_INSTRUCTIONS]	= ARMV6_PERFCTR_BR_EXEC,
+	[PERF_COUNT_HW_BRANCH_MISSES]		= ARMV6_PERFCTR_BR_MISPREDICT,
+	[PERF_COUNT_HW_BUS_CYCLES]		= HW_OP_UNSUPPORTED,
+	[PERF_COUNT_HW_STALLED_CYCLES_FRONTEND]	= ARMV6_PERFCTR_IBUF_STALL,
+	[PERF_COUNT_HW_STALLED_CYCLES_BACKEND]	= ARMV6_PERFCTR_LSU_FULL_STALL,
 };
 
 static const unsigned armv6_perf_cache_map[PERF_COUNT_HW_CACHE_MAX]
@@ -218,13 +220,15 @@ enum armv6mpcore_perf_types {
  * accesses/misses in hardware.
  */
 static const unsigned armv6mpcore_perf_map[PERF_COUNT_HW_MAX] = {
-	[PERF_COUNT_HW_CPU_CYCLES]	    = ARMV6MPCORE_PERFCTR_CPU_CYCLES,
-	[PERF_COUNT_HW_INSTRUCTIONS]	    = ARMV6MPCORE_PERFCTR_INSTR_EXEC,
-	[PERF_COUNT_HW_CACHE_REFERENCES]    = HW_OP_UNSUPPORTED,
-	[PERF_COUNT_HW_CACHE_MISSES]	    = HW_OP_UNSUPPORTED,
-	[PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = ARMV6MPCORE_PERFCTR_BR_EXEC,
-	[PERF_COUNT_HW_BRANCH_MISSES]	    = ARMV6MPCORE_PERFCTR_BR_MISPREDICT,
-	[PERF_COUNT_HW_BUS_CYCLES]	    = HW_OP_UNSUPPORTED,
+	[PERF_COUNT_HW_CPU_CYCLES]		= ARMV6MPCORE_PERFCTR_CPU_CYCLES,
+	[PERF_COUNT_HW_INSTRUCTIONS]		= ARMV6MPCORE_PERFCTR_INSTR_EXEC,
+	[PERF_COUNT_HW_CACHE_REFERENCES]	= HW_OP_UNSUPPORTED,
+	[PERF_COUNT_HW_CACHE_MISSES]		= HW_OP_UNSUPPORTED,
+	[PERF_COUNT_HW_BRANCH_INSTRUCTIONS]	= ARMV6MPCORE_PERFCTR_BR_EXEC,
+	[PERF_COUNT_HW_BRANCH_MISSES]		= ARMV6MPCORE_PERFCTR_BR_MISPREDICT,
+	[PERF_COUNT_HW_BUS_CYCLES]		= HW_OP_UNSUPPORTED,
+	[PERF_COUNT_HW_STALLED_CYCLES_FRONTEND]	= ARMV6MPCORE_PERFCTR_IBUF_STALL,
+	[PERF_COUNT_HW_STALLED_CYCLES_BACKEND]	= ARMV6MPCORE_PERFCTR_LSU_FULL_STALL,
 };
 
 static const unsigned armv6mpcore_perf_cache_map[PERF_COUNT_HW_CACHE_MAX]
diff --git a/arch/arm/kernel/perf_event_v7.c b/arch/arm/kernel/perf_event_v7.c
index d2361e7..460bbbb 100644
--- a/arch/arm/kernel/perf_event_v7.c
+++ b/arch/arm/kernel/perf_event_v7.c
@@ -77,11 +77,14 @@ enum armv7_a8_perf_types {
 	ARMV7_A8_PERFCTR_L2_CACHE_ACCESS		= 0x43,
 	ARMV7_A8_PERFCTR_L2_CACHE_REFILL		= 0x44,
 	ARMV7_A8_PERFCTR_L1_ICACHE_ACCESS		= 0x50,
+	ARMV7_A8_PERFCTR_STALL_ISIDE			= 0x56,
 };
 
 /* ARMv7 Cortex-A9 specific event types */
 enum armv7_a9_perf_types {
 	ARMV7_A9_PERFCTR_INSTR_CORE_RENAME		= 0x68,
+	ARMV7_A9_PERFCTR_STALL_ICACHE			= 0x60,
+	ARMV7_A9_PERFCTR_STALL_DISPATCH			= 0x66,
 };
 
 /* ARMv7 Cortex-A5 specific event types */
@@ -116,13 +119,15 @@ enum armv7_a15_perf_types {
  * accesses/misses in hardware.
  */
 static const unsigned armv7_a8_perf_map[PERF_COUNT_HW_MAX] = {
-	[PERF_COUNT_HW_CPU_CYCLES]	    = ARMV7_PERFCTR_CPU_CYCLES,
-	[PERF_COUNT_HW_INSTRUCTIONS]	    = ARMV7_PERFCTR_INSTR_EXECUTED,
-	[PERF_COUNT_HW_CACHE_REFERENCES]    = ARMV7_PERFCTR_L1_DCACHE_ACCESS,
-	[PERF_COUNT_HW_CACHE_MISSES]	    = ARMV7_PERFCTR_L1_DCACHE_REFILL,
-	[PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = ARMV7_PERFCTR_PC_WRITE,
-	[PERF_COUNT_HW_BRANCH_MISSES]	    = ARMV7_PERFCTR_PC_BRANCH_MIS_PRED,
-	[PERF_COUNT_HW_BUS_CYCLES]	    = HW_OP_UNSUPPORTED,
+	[PERF_COUNT_HW_CPU_CYCLES]		= ARMV7_PERFCTR_CPU_CYCLES,
+	[PERF_COUNT_HW_INSTRUCTIONS]		= ARMV7_PERFCTR_INSTR_EXECUTED,
+	[PERF_COUNT_HW_CACHE_REFERENCES]	= ARMV7_PERFCTR_L1_DCACHE_ACCESS,
+	[PERF_COUNT_HW_CACHE_MISSES]		= ARMV7_PERFCTR_L1_DCACHE_REFILL,
+	[PERF_COUNT_HW_BRANCH_INSTRUCTIONS]	= ARMV7_PERFCTR_PC_WRITE,
+	[PERF_COUNT_HW_BRANCH_MISSES]		= ARMV7_PERFCTR_PC_BRANCH_MIS_PRED,
+	[PERF_COUNT_HW_BUS_CYCLES]		= HW_OP_UNSUPPORTED,
+	[PERF_COUNT_HW_STALLED_CYCLES_FRONTEND]	= ARMV7_A8_PERFCTR_STALL_ISIDE,
+	[PERF_COUNT_HW_STALLED_CYCLES_BACKEND]	= HW_OP_UNSUPPORTED,
 };
 
 static const unsigned armv7_a8_perf_cache_map[PERF_COUNT_HW_CACHE_MAX]
@@ -238,13 +243,15 @@ static const unsigned armv7_a8_perf_cache_map[PERF_COUNT_HW_CACHE_MAX]
  * Cortex-A9 HW events mapping
  */
 static const unsigned armv7_a9_perf_map[PERF_COUNT_HW_MAX] = {
-	[PERF_COUNT_HW_CPU_CYCLES]	    = ARMV7_PERFCTR_CPU_CYCLES,
-	[PERF_COUNT_HW_INSTRUCTIONS]	    = ARMV7_A9_PERFCTR_INSTR_CORE_RENAME,
-	[PERF_COUNT_HW_CACHE_REFERENCES]    = ARMV7_PERFCTR_L1_DCACHE_ACCESS,
-	[PERF_COUNT_HW_CACHE_MISSES]	    = ARMV7_PERFCTR_L1_DCACHE_REFILL,
-	[PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = ARMV7_PERFCTR_PC_WRITE,
-	[PERF_COUNT_HW_BRANCH_MISSES]	    = ARMV7_PERFCTR_PC_BRANCH_MIS_PRED,
-	[PERF_COUNT_HW_BUS_CYCLES]	    = HW_OP_UNSUPPORTED,
+	[PERF_COUNT_HW_CPU_CYCLES]		= ARMV7_PERFCTR_CPU_CYCLES,
+	[PERF_COUNT_HW_INSTRUCTIONS]		= ARMV7_A9_PERFCTR_INSTR_CORE_RENAME,
+	[PERF_COUNT_HW_CACHE_REFERENCES]	= ARMV7_PERFCTR_L1_DCACHE_ACCESS,
+	[PERF_COUNT_HW_CACHE_MISSES]		= ARMV7_PERFCTR_L1_DCACHE_REFILL,
+	[PERF_COUNT_HW_BRANCH_INSTRUCTIONS]	= ARMV7_PERFCTR_PC_WRITE,
+	[PERF_COUNT_HW_BRANCH_MISSES]		= ARMV7_PERFCTR_PC_BRANCH_MIS_PRED,
+	[PERF_COUNT_HW_BUS_CYCLES]		= HW_OP_UNSUPPORTED,
+	[PERF_COUNT_HW_STALLED_CYCLES_FRONTEND]	= ARMV7_A9_PERFCTR_STALL_ICACHE,
+	[PERF_COUNT_HW_STALLED_CYCLES_BACKEND]	= ARMV7_A9_PERFCTR_STALL_DISPATCH,
 };
 
 static const unsigned armv7_a9_perf_cache_map[PERF_COUNT_HW_CACHE_MAX]
@@ -360,13 +367,15 @@ static const unsigned armv7_a9_perf_cache_map[PERF_COUNT_HW_CACHE_MAX]
  * Cortex-A5 HW events mapping
  */
 static const unsigned armv7_a5_perf_map[PERF_COUNT_HW_MAX] = {
-	[PERF_COUNT_HW_CPU_CYCLES]	    = ARMV7_PERFCTR_CPU_CYCLES,
-	[PERF_COUNT_HW_INSTRUCTIONS]	    = ARMV7_PERFCTR_INSTR_EXECUTED,
-	[PERF_COUNT_HW_CACHE_REFERENCES]    = ARMV7_PERFCTR_L1_DCACHE_ACCESS,
-	[PERF_COUNT_HW_CACHE_MISSES]	    = ARMV7_PERFCTR_L1_DCACHE_REFILL,
-	[PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = ARMV7_PERFCTR_PC_WRITE,
-	[PERF_COUNT_HW_BRANCH_MISSES]	    = ARMV7_PERFCTR_PC_BRANCH_MIS_PRED,
-	[PERF_COUNT_HW_BUS_CYCLES]	    = HW_OP_UNSUPPORTED,
+	[PERF_COUNT_HW_CPU_CYCLES]		= ARMV7_PERFCTR_CPU_CYCLES,
+	[PERF_COUNT_HW_INSTRUCTIONS]		= ARMV7_PERFCTR_INSTR_EXECUTED,
+	[PERF_COUNT_HW_CACHE_REFERENCES]	= ARMV7_PERFCTR_L1_DCACHE_ACCESS,
+	[PERF_COUNT_HW_CACHE_MISSES]		= ARMV7_PERFCTR_L1_DCACHE_REFILL,
+	[PERF_COUNT_HW_BRANCH_INSTRUCTIONS]	= ARMV7_PERFCTR_PC_WRITE,
+	[PERF_COUNT_HW_BRANCH_MISSES]		= ARMV7_PERFCTR_PC_BRANCH_MIS_PRED,
+	[PERF_COUNT_HW_BUS_CYCLES]		= HW_OP_UNSUPPORTED,
+	[PERF_COUNT_HW_STALLED_CYCLES_FRONTEND]	= HW_OP_UNSUPPORTED,
+	[PERF_COUNT_HW_STALLED_CYCLES_BACKEND]	= HW_OP_UNSUPPORTED,
 };
 
 static const unsigned armv7_a5_perf_cache_map[PERF_COUNT_HW_CACHE_MAX]
@@ -466,13 +475,15 @@ static const unsigned armv7_a5_perf_cache_map[PERF_COUNT_HW_CACHE_MAX]
  * Cortex-A15 HW events mapping
  */
 static const unsigned armv7_a15_perf_map[PERF_COUNT_HW_MAX] = {
-	[PERF_COUNT_HW_CPU_CYCLES]	    = ARMV7_PERFCTR_CPU_CYCLES,
-	[PERF_COUNT_HW_INSTRUCTIONS]	    = ARMV7_PERFCTR_INSTR_EXECUTED,
-	[PERF_COUNT_HW_CACHE_REFERENCES]    = ARMV7_PERFCTR_L1_DCACHE_ACCESS,
-	[PERF_COUNT_HW_CACHE_MISSES]	    = ARMV7_PERFCTR_L1_DCACHE_REFILL,
-	[PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = ARMV7_A15_PERFCTR_PC_WRITE_SPEC,
-	[PERF_COUNT_HW_BRANCH_MISSES]	    = ARMV7_PERFCTR_PC_BRANCH_MIS_PRED,
-	[PERF_COUNT_HW_BUS_CYCLES]	    = ARMV7_PERFCTR_BUS_CYCLES,
+	[PERF_COUNT_HW_CPU_CYCLES]		= ARMV7_PERFCTR_CPU_CYCLES,
+	[PERF_COUNT_HW_INSTRUCTIONS]		= ARMV7_PERFCTR_INSTR_EXECUTED,
+	[PERF_COUNT_HW_CACHE_REFERENCES]	= ARMV7_PERFCTR_L1_DCACHE_ACCESS,
+	[PERF_COUNT_HW_CACHE_MISSES]		= ARMV7_PERFCTR_L1_DCACHE_REFILL,
+	[PERF_COUNT_HW_BRANCH_INSTRUCTIONS]	= ARMV7_A15_PERFCTR_PC_WRITE_SPEC,
+	[PERF_COUNT_HW_BRANCH_MISSES]		= ARMV7_PERFCTR_PC_BRANCH_MIS_PRED,
+	[PERF_COUNT_HW_BUS_CYCLES]		= ARMV7_PERFCTR_BUS_CYCLES,
+	[PERF_COUNT_HW_STALLED_CYCLES_FRONTEND]	= HW_OP_UNSUPPORTED,
+	[PERF_COUNT_HW_STALLED_CYCLES_BACKEND]	= HW_OP_UNSUPPORTED,
 };
 
 static const unsigned armv7_a15_perf_cache_map[PERF_COUNT_HW_CACHE_MAX]
diff --git a/arch/arm/kernel/perf_event_xscale.c b/arch/arm/kernel/perf_event_xscale.c
index e0cca10..3b99d82 100644
--- a/arch/arm/kernel/perf_event_xscale.c
+++ b/arch/arm/kernel/perf_event_xscale.c
@@ -48,13 +48,15 @@ enum xscale_counters {
 };
 
 static const unsigned xscale_perf_map[PERF_COUNT_HW_MAX] = {
-	[PERF_COUNT_HW_CPU_CYCLES]	    = XSCALE_PERFCTR_CCNT,
-	[PERF_COUNT_HW_INSTRUCTIONS]	    = XSCALE_PERFCTR_INSTRUCTION,
-	[PERF_COUNT_HW_CACHE_REFERENCES]    = HW_OP_UNSUPPORTED,
-	[PERF_COUNT_HW_CACHE_MISSES]	    = HW_OP_UNSUPPORTED,
-	[PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = XSCALE_PERFCTR_BRANCH,
-	[PERF_COUNT_HW_BRANCH_MISSES]	    = XSCALE_PERFCTR_BRANCH_MISS,
-	[PERF_COUNT_HW_BUS_CYCLES]	    = HW_OP_UNSUPPORTED,
+	[PERF_COUNT_HW_CPU_CYCLES]		= XSCALE_PERFCTR_CCNT,
+	[PERF_COUNT_HW_INSTRUCTIONS]		= XSCALE_PERFCTR_INSTRUCTION,
+	[PERF_COUNT_HW_CACHE_REFERENCES]	= HW_OP_UNSUPPORTED,
+	[PERF_COUNT_HW_CACHE_MISSES]		= HW_OP_UNSUPPORTED,
+	[PERF_COUNT_HW_BRANCH_INSTRUCTIONS]	= XSCALE_PERFCTR_BRANCH,
+	[PERF_COUNT_HW_BRANCH_MISSES]		= XSCALE_PERFCTR_BRANCH_MISS,
+	[PERF_COUNT_HW_BUS_CYCLES]		= HW_OP_UNSUPPORTED,
+	[PERF_COUNT_HW_STALLED_CYCLES_FRONTEND]	= XSCALE_PERFCTR_ICACHE_NO_DELIVER,
+	[PERF_COUNT_HW_STALLED_CYCLES_BACKEND]	= HW_OP_UNSUPPORTED,
 };
 
 static const unsigned xscale_perf_cache_map[PERF_COUNT_HW_CACHE_MAX]