All of lore.kernel.org
 help / color / mirror / Atom feed
* [PATCH v2 0/6] arm64: perf: Broadcom Vulcan PMU support
@ 2016-03-24 12:52 Ashok Kumar
  2016-03-24 12:52 ` [PATCH v2 1/6] arm64/perf: Changed events naming as per ARM ARM Ashok Kumar
                   ` (5 more replies)
  0 siblings, 6 replies; 11+ messages in thread
From: Ashok Kumar @ 2016-03-24 12:52 UTC (permalink / raw)
  To: linux-arm-kernel

Cleaned up event naming convention as per ARM ARM.

Added macros for complete ARMv8 recommended implementation defined events.

Common architectural and micro-architectural events which are exported to /sys
are now filtered using PMCEIDn_EL0

Added support for Broadcom Vulcan PMU. 

changes since v1 [1]:
 Incorporated the following review comments from Will.
 * cleaned up event naming convention as per ARM ARM
 * Filtered common events based on PMCEIDn_EL0
 * Removed exposing implementation defined events to /sys.

[1] http://www.spinics.net/lists/arm-kernel/msg490954.html

Ashok Kumar (6):
  arm64/perf: Changed events naming as per ARM ARM
  arm64/perf: Define complete ARMv8 recommended implementation defined
    events
  arm64/perf: Filter common events based on PMCEIDn_EL0
  arm64/perf: Add Broadcom Vulcan PMU support
  arm64: dts: Add Broadcom Vulcan PMU in dts
  Documentation: arm64: pmu: Add Broadcom Vulcan PMU binding

 Documentation/devicetree/bindings/arm/pmu.txt |   3 +-
 arch/arm64/boot/dts/broadcom/vulcan.dtsi      |   2 +-
 arch/arm64/kernel/perf_event.c                | 717 +++++++++++++++++---------
 3 files changed, 485 insertions(+), 237 deletions(-)

-- 
2.1.0

^ permalink raw reply	[flat|nested] 11+ messages in thread

* [PATCH v2 1/6] arm64/perf: Changed events naming as per ARM ARM
  2016-03-24 12:52 [PATCH v2 0/6] arm64: perf: Broadcom Vulcan PMU support Ashok Kumar
@ 2016-03-24 12:52 ` Ashok Kumar
  2016-03-24 12:52 ` [PATCH v2 2/6] arm64/perf: Define complete ARMv8 recommended implementation defined events Ashok Kumar
                   ` (4 subsequent siblings)
  5 siblings, 0 replies; 11+ messages in thread
From: Ashok Kumar @ 2016-03-24 12:52 UTC (permalink / raw)
  To: linux-arm-kernel

changed all the events name definition as per ARM ARM
naming convention.

Signed-off-by: Ashok Kumar <ashoks@broadcom.com>
---
 arch/arm64/kernel/perf_event.c | 302 ++++++++++++++++++++---------------------
 1 file changed, 151 insertions(+), 151 deletions(-)

diff --git a/arch/arm64/kernel/perf_event.c b/arch/arm64/kernel/perf_event.c
index 1cc61fc..d1a93cf 100644
--- a/arch/arm64/kernel/perf_event.c
+++ b/arch/arm64/kernel/perf_event.c
@@ -31,43 +31,43 @@
  */
 
 /* Required events. */
-#define ARMV8_PMUV3_PERFCTR_PMNC_SW_INCR			0x00
-#define ARMV8_PMUV3_PERFCTR_L1_DCACHE_REFILL			0x03
-#define ARMV8_PMUV3_PERFCTR_L1_DCACHE_ACCESS			0x04
-#define ARMV8_PMUV3_PERFCTR_PC_BRANCH_MIS_PRED			0x10
-#define ARMV8_PMUV3_PERFCTR_CLOCK_CYCLES			0x11
-#define ARMV8_PMUV3_PERFCTR_PC_BRANCH_PRED			0x12
+#define ARMV8_PMUV3_PERFCTR_SW_INCR				0x00
+#define ARMV8_PMUV3_PERFCTR_L1D_CACHE_REFILL			0x03
+#define ARMV8_PMUV3_PERFCTR_L1D_CACHE				0x04
+#define ARMV8_PMUV3_PERFCTR_BR_MIS_PRED				0x10
+#define ARMV8_PMUV3_PERFCTR_CPU_CYCLES				0x11
+#define ARMV8_PMUV3_PERFCTR_BR_PRED				0x12
 
 /* At least one of the following is required. */
-#define ARMV8_PMUV3_PERFCTR_INSTR_EXECUTED			0x08
-#define ARMV8_PMUV3_PERFCTR_OP_SPEC				0x1B
+#define ARMV8_PMUV3_PERFCTR_INST_RETIRED			0x08
+#define ARMV8_PMUV3_PERFCTR_INST_SPEC				0x1B
 
 /* Common architectural events. */
-#define ARMV8_PMUV3_PERFCTR_MEM_READ				0x06
-#define ARMV8_PMUV3_PERFCTR_MEM_WRITE				0x07
+#define ARMV8_PMUV3_PERFCTR_LD_RETIRED				0x06
+#define ARMV8_PMUV3_PERFCTR_ST_RETIRED				0x07
 #define ARMV8_PMUV3_PERFCTR_EXC_TAKEN				0x09
-#define ARMV8_PMUV3_PERFCTR_EXC_EXECUTED			0x0A
-#define ARMV8_PMUV3_PERFCTR_CID_WRITE				0x0B
-#define ARMV8_PMUV3_PERFCTR_PC_WRITE				0x0C
-#define ARMV8_PMUV3_PERFCTR_PC_IMM_BRANCH			0x0D
-#define ARMV8_PMUV3_PERFCTR_PC_PROC_RETURN			0x0E
-#define ARMV8_PMUV3_PERFCTR_MEM_UNALIGNED_ACCESS		0x0F
-#define ARMV8_PMUV3_PERFCTR_TTBR_WRITE				0x1C
+#define ARMV8_PMUV3_PERFCTR_EXC_RETURN				0x0A
+#define ARMV8_PMUV3_PERFCTR_CID_WRITE_RETIRED			0x0B
+#define ARMV8_PMUV3_PERFCTR_PC_WRITE_RETIRED			0x0C
+#define ARMV8_PMUV3_PERFCTR_BR_IMMED_RETIRED			0x0D
+#define ARMV8_PMUV3_PERFCTR_BR_RETURN_RETIRED			0x0E
+#define ARMV8_PMUV3_PERFCTR_UNALIGNED_LDST_RETIRED		0x0F
+#define ARMV8_PMUV3_PERFCTR_TTBR_WRITE_RETIRED			0x1C
 #define ARMV8_PMUV3_PERFCTR_CHAIN				0x1E
 #define ARMV8_PMUV3_PERFCTR_BR_RETIRED				0x21
 
 /* Common microarchitectural events. */
-#define ARMV8_PMUV3_PERFCTR_L1_ICACHE_REFILL			0x01
-#define ARMV8_PMUV3_PERFCTR_ITLB_REFILL				0x02
-#define ARMV8_PMUV3_PERFCTR_DTLB_REFILL				0x05
+#define ARMV8_PMUV3_PERFCTR_L1I_CACHE_REFILL			0x01
+#define ARMV8_PMUV3_PERFCTR_L1I_TLB_REFILL			0x02
+#define ARMV8_PMUV3_PERFCTR_L1D_TLB_REFILL			0x05
 #define ARMV8_PMUV3_PERFCTR_MEM_ACCESS				0x13
-#define ARMV8_PMUV3_PERFCTR_L1_ICACHE_ACCESS			0x14
-#define ARMV8_PMUV3_PERFCTR_L1_DCACHE_WB			0x15
-#define ARMV8_PMUV3_PERFCTR_L2_CACHE_ACCESS			0x16
-#define ARMV8_PMUV3_PERFCTR_L2_CACHE_REFILL			0x17
-#define ARMV8_PMUV3_PERFCTR_L2_CACHE_WB				0x18
+#define ARMV8_PMUV3_PERFCTR_L1I_CACHE				0x14
+#define ARMV8_PMUV3_PERFCTR_L1D_CACHE_WB			0x15
+#define ARMV8_PMUV3_PERFCTR_L2D_CACHE				0x16
+#define ARMV8_PMUV3_PERFCTR_L2D_CACHE_REFILL			0x17
+#define ARMV8_PMUV3_PERFCTR_L2D_CACHE_WB			0x18
 #define ARMV8_PMUV3_PERFCTR_BUS_ACCESS				0x19
-#define ARMV8_PMUV3_PERFCTR_MEM_ERROR				0x1A
+#define ARMV8_PMUV3_PERFCTR_MEMORY_ERROR			0x1A
 #define ARMV8_PMUV3_PERFCTR_BUS_CYCLES				0x1D
 #define ARMV8_PMUV3_PERFCTR_L1D_CACHE_ALLOCATE			0x1F
 #define ARMV8_PMUV3_PERFCTR_L2D_CACHE_ALLOCATE			0x20
@@ -83,71 +83,71 @@
 #define ARMV8_PMUV3_PERFCTR_L3D_CACHE				0x2B
 #define ARMV8_PMUV3_PERFCTR_L3D_CACHE_WB			0x2C
 #define ARMV8_PMUV3_PERFCTR_L2D_TLB_REFILL			0x2D
-#define ARMV8_PMUV3_PERFCTR_L21_TLB_REFILL			0x2E
+#define ARMV8_PMUV3_PERFCTR_L2I_TLB_REFILL			0x2E
 #define ARMV8_PMUV3_PERFCTR_L2D_TLB				0x2F
-#define ARMV8_PMUV3_PERFCTR_L21_TLB				0x30
-
-/* ARMv8 implementation defined event types. */
-#define ARMV8_IMPDEF_PERFCTR_L1_DCACHE_ACCESS_LD		0x40
-#define ARMV8_IMPDEF_PERFCTR_L1_DCACHE_ACCESS_ST		0x41
-#define ARMV8_IMPDEF_PERFCTR_L1_DCACHE_REFILL_LD		0x42
-#define ARMV8_IMPDEF_PERFCTR_L1_DCACHE_REFILL_ST		0x43
-#define ARMV8_IMPDEF_PERFCTR_DTLB_REFILL_LD			0x4C
-#define ARMV8_IMPDEF_PERFCTR_DTLB_REFILL_ST			0x4D
-#define ARMV8_IMPDEF_PERFCTR_DTLB_ACCESS_LD			0x4E
-#define ARMV8_IMPDEF_PERFCTR_DTLB_ACCESS_ST			0x4F
+#define ARMV8_PMUV3_PERFCTR_L2I_TLB				0x30
+
+/* ARMv8 recommended implementation defined event types */
+#define ARMV8_IMPDEF_PERFCTR_L1D_CACHE_RD			0x40
+#define ARMV8_IMPDEF_PERFCTR_L1D_CACHE_WR			0x41
+#define ARMV8_IMPDEF_PERFCTR_L1D_CACHE_REFILL_RD		0x42
+#define ARMV8_IMPDEF_PERFCTR_L1D_CACHE_REFILL_WR		0x43
+#define ARMV8_IMPDEF_PERFCTR_L1D_TLB_REFILL_RD			0x4C
+#define ARMV8_IMPDEF_PERFCTR_L1D_TLB_REFILL_WR			0x4D
+#define ARMV8_IMPDEF_PERFCTR_L1D_TLB_RD				0x4E
+#define ARMV8_IMPDEF_PERFCTR_L1D_TLB_WR				0x4F
 
 /* ARMv8 Cortex-A53 specific event types. */
 #define ARMV8_A53_PERFCTR_PREFETCH_LINEFILL			0xC2
 
 /* ARMv8 Cavium ThunderX specific event types. */
-#define ARMV8_THUNDER_PERFCTR_L1_DCACHE_MISS_ST			0xE9
-#define ARMV8_THUNDER_PERFCTR_L1_DCACHE_PREF_ACCESS		0xEA
-#define ARMV8_THUNDER_PERFCTR_L1_DCACHE_PREF_MISS		0xEB
-#define ARMV8_THUNDER_PERFCTR_L1_ICACHE_PREF_ACCESS		0xEC
-#define ARMV8_THUNDER_PERFCTR_L1_ICACHE_PREF_MISS		0xED
+#define ARMV8_THUNDER_PERFCTR_L1D_CACHE_MISS_ST			0xE9
+#define ARMV8_THUNDER_PERFCTR_L1D_CACHE_PREF_ACCESS		0xEA
+#define ARMV8_THUNDER_PERFCTR_L1D_CACHE_PREF_MISS		0xEB
+#define ARMV8_THUNDER_PERFCTR_L1I_CACHE_PREF_ACCESS		0xEC
+#define ARMV8_THUNDER_PERFCTR_L1I_CACHE_PREF_MISS		0xED
 
 /* PMUv3 HW events mapping. */
 static const unsigned armv8_pmuv3_perf_map[PERF_COUNT_HW_MAX] = {
 	PERF_MAP_ALL_UNSUPPORTED,
-	[PERF_COUNT_HW_CPU_CYCLES]		= ARMV8_PMUV3_PERFCTR_CLOCK_CYCLES,
-	[PERF_COUNT_HW_INSTRUCTIONS]		= ARMV8_PMUV3_PERFCTR_INSTR_EXECUTED,
-	[PERF_COUNT_HW_CACHE_REFERENCES]	= ARMV8_PMUV3_PERFCTR_L1_DCACHE_ACCESS,
-	[PERF_COUNT_HW_CACHE_MISSES]		= ARMV8_PMUV3_PERFCTR_L1_DCACHE_REFILL,
-	[PERF_COUNT_HW_BRANCH_MISSES]		= ARMV8_PMUV3_PERFCTR_PC_BRANCH_MIS_PRED,
+	[PERF_COUNT_HW_CPU_CYCLES]		= ARMV8_PMUV3_PERFCTR_CPU_CYCLES,
+	[PERF_COUNT_HW_INSTRUCTIONS]		= ARMV8_PMUV3_PERFCTR_INST_RETIRED,
+	[PERF_COUNT_HW_CACHE_REFERENCES]	= ARMV8_PMUV3_PERFCTR_L1D_CACHE,
+	[PERF_COUNT_HW_CACHE_MISSES]		= ARMV8_PMUV3_PERFCTR_L1D_CACHE_REFILL,
+	[PERF_COUNT_HW_BRANCH_MISSES]		= ARMV8_PMUV3_PERFCTR_BR_MIS_PRED,
 };
 
 /* ARM Cortex-A53 HW events mapping. */
 static const unsigned armv8_a53_perf_map[PERF_COUNT_HW_MAX] = {
 	PERF_MAP_ALL_UNSUPPORTED,
-	[PERF_COUNT_HW_CPU_CYCLES]		= ARMV8_PMUV3_PERFCTR_CLOCK_CYCLES,
-	[PERF_COUNT_HW_INSTRUCTIONS]		= ARMV8_PMUV3_PERFCTR_INSTR_EXECUTED,
-	[PERF_COUNT_HW_CACHE_REFERENCES]	= ARMV8_PMUV3_PERFCTR_L1_DCACHE_ACCESS,
-	[PERF_COUNT_HW_CACHE_MISSES]		= ARMV8_PMUV3_PERFCTR_L1_DCACHE_REFILL,
-	[PERF_COUNT_HW_BRANCH_INSTRUCTIONS]	= ARMV8_PMUV3_PERFCTR_PC_WRITE,
-	[PERF_COUNT_HW_BRANCH_MISSES]		= ARMV8_PMUV3_PERFCTR_PC_BRANCH_MIS_PRED,
+	[PERF_COUNT_HW_CPU_CYCLES]		= ARMV8_PMUV3_PERFCTR_CPU_CYCLES,
+	[PERF_COUNT_HW_INSTRUCTIONS]		= ARMV8_PMUV3_PERFCTR_INST_RETIRED,
+	[PERF_COUNT_HW_CACHE_REFERENCES]	= ARMV8_PMUV3_PERFCTR_L1D_CACHE,
+	[PERF_COUNT_HW_CACHE_MISSES]		= ARMV8_PMUV3_PERFCTR_L1D_CACHE_REFILL,
+	[PERF_COUNT_HW_BRANCH_INSTRUCTIONS]	= ARMV8_PMUV3_PERFCTR_PC_WRITE_RETIRED,
+	[PERF_COUNT_HW_BRANCH_MISSES]		= ARMV8_PMUV3_PERFCTR_BR_MIS_PRED,
 	[PERF_COUNT_HW_BUS_CYCLES]		= ARMV8_PMUV3_PERFCTR_BUS_CYCLES,
 };
 
 /* ARM Cortex-A57 and Cortex-A72 events mapping. */
 static const unsigned armv8_a57_perf_map[PERF_COUNT_HW_MAX] = {
 	PERF_MAP_ALL_UNSUPPORTED,
-	[PERF_COUNT_HW_CPU_CYCLES]		= ARMV8_PMUV3_PERFCTR_CLOCK_CYCLES,
-	[PERF_COUNT_HW_INSTRUCTIONS]		= ARMV8_PMUV3_PERFCTR_INSTR_EXECUTED,
-	[PERF_COUNT_HW_CACHE_REFERENCES]	= ARMV8_PMUV3_PERFCTR_L1_DCACHE_ACCESS,
-	[PERF_COUNT_HW_CACHE_MISSES]		= ARMV8_PMUV3_PERFCTR_L1_DCACHE_REFILL,
-	[PERF_COUNT_HW_BRANCH_MISSES]		= ARMV8_PMUV3_PERFCTR_PC_BRANCH_MIS_PRED,
+	[PERF_COUNT_HW_CPU_CYCLES]		= ARMV8_PMUV3_PERFCTR_CPU_CYCLES,
+	[PERF_COUNT_HW_INSTRUCTIONS]		= ARMV8_PMUV3_PERFCTR_INST_RETIRED,
+	[PERF_COUNT_HW_CACHE_REFERENCES]	= ARMV8_PMUV3_PERFCTR_L1D_CACHE,
+	[PERF_COUNT_HW_CACHE_MISSES]		= ARMV8_PMUV3_PERFCTR_L1D_CACHE_REFILL,
+	[PERF_COUNT_HW_BRANCH_MISSES]		= ARMV8_PMUV3_PERFCTR_BR_MIS_PRED,
 	[PERF_COUNT_HW_BUS_CYCLES]		= ARMV8_PMUV3_PERFCTR_BUS_CYCLES,
 };
 
 static const unsigned armv8_thunder_perf_map[PERF_COUNT_HW_MAX] = {
 	PERF_MAP_ALL_UNSUPPORTED,
-	[PERF_COUNT_HW_CPU_CYCLES]		= ARMV8_PMUV3_PERFCTR_CLOCK_CYCLES,
-	[PERF_COUNT_HW_INSTRUCTIONS]		= ARMV8_PMUV3_PERFCTR_INSTR_EXECUTED,
-	[PERF_COUNT_HW_CACHE_REFERENCES]	= ARMV8_PMUV3_PERFCTR_L1_DCACHE_ACCESS,
-	[PERF_COUNT_HW_CACHE_MISSES]		= ARMV8_PMUV3_PERFCTR_L1_DCACHE_REFILL,
-	[PERF_COUNT_HW_BRANCH_INSTRUCTIONS]	= ARMV8_PMUV3_PERFCTR_PC_WRITE,
-	[PERF_COUNT_HW_BRANCH_MISSES]		= ARMV8_PMUV3_PERFCTR_PC_BRANCH_MIS_PRED,
+	[PERF_COUNT_HW_CPU_CYCLES]		= ARMV8_PMUV3_PERFCTR_CPU_CYCLES,
+	[PERF_COUNT_HW_INSTRUCTIONS]		= ARMV8_PMUV3_PERFCTR_INST_RETIRED,
+	[PERF_COUNT_HW_CACHE_REFERENCES]	= ARMV8_PMUV3_PERFCTR_L1D_CACHE,
+	[PERF_COUNT_HW_CACHE_MISSES]		= ARMV8_PMUV3_PERFCTR_L1D_CACHE_REFILL,
+	[PERF_COUNT_HW_BRANCH_INSTRUCTIONS]	= ARMV8_PMUV3_PERFCTR_PC_WRITE_RETIRED,
+	[PERF_COUNT_HW_BRANCH_MISSES]		= ARMV8_PMUV3_PERFCTR_BR_MIS_PRED,
 	[PERF_COUNT_HW_STALLED_CYCLES_FRONTEND] = ARMV8_PMUV3_PERFCTR_STALL_FRONTEND,
 	[PERF_COUNT_HW_STALLED_CYCLES_BACKEND]	= ARMV8_PMUV3_PERFCTR_STALL_BACKEND,
 };
@@ -157,15 +157,15 @@ static const unsigned armv8_pmuv3_perf_cache_map[PERF_COUNT_HW_CACHE_MAX]
 						[PERF_COUNT_HW_CACHE_RESULT_MAX] = {
 	PERF_CACHE_MAP_ALL_UNSUPPORTED,
 
-	[C(L1D)][C(OP_READ)][C(RESULT_ACCESS)]	= ARMV8_PMUV3_PERFCTR_L1_DCACHE_ACCESS,
-	[C(L1D)][C(OP_READ)][C(RESULT_MISS)]	= ARMV8_PMUV3_PERFCTR_L1_DCACHE_REFILL,
-	[C(L1D)][C(OP_WRITE)][C(RESULT_ACCESS)]	= ARMV8_PMUV3_PERFCTR_L1_DCACHE_ACCESS,
-	[C(L1D)][C(OP_WRITE)][C(RESULT_MISS)]	= ARMV8_PMUV3_PERFCTR_L1_DCACHE_REFILL,
+	[C(L1D)][C(OP_READ)][C(RESULT_ACCESS)]	= ARMV8_PMUV3_PERFCTR_L1D_CACHE,
+	[C(L1D)][C(OP_READ)][C(RESULT_MISS)]	= ARMV8_PMUV3_PERFCTR_L1D_CACHE_REFILL,
+	[C(L1D)][C(OP_WRITE)][C(RESULT_ACCESS)]	= ARMV8_PMUV3_PERFCTR_L1D_CACHE,
+	[C(L1D)][C(OP_WRITE)][C(RESULT_MISS)]	= ARMV8_PMUV3_PERFCTR_L1D_CACHE_REFILL,
 
-	[C(BPU)][C(OP_READ)][C(RESULT_ACCESS)]	= ARMV8_PMUV3_PERFCTR_PC_BRANCH_PRED,
-	[C(BPU)][C(OP_READ)][C(RESULT_MISS)]	= ARMV8_PMUV3_PERFCTR_PC_BRANCH_MIS_PRED,
-	[C(BPU)][C(OP_WRITE)][C(RESULT_ACCESS)]	= ARMV8_PMUV3_PERFCTR_PC_BRANCH_PRED,
-	[C(BPU)][C(OP_WRITE)][C(RESULT_MISS)]	= ARMV8_PMUV3_PERFCTR_PC_BRANCH_MIS_PRED,
+	[C(BPU)][C(OP_READ)][C(RESULT_ACCESS)]	= ARMV8_PMUV3_PERFCTR_BR_PRED,
+	[C(BPU)][C(OP_READ)][C(RESULT_MISS)]	= ARMV8_PMUV3_PERFCTR_BR_MIS_PRED,
+	[C(BPU)][C(OP_WRITE)][C(RESULT_ACCESS)]	= ARMV8_PMUV3_PERFCTR_BR_PRED,
+	[C(BPU)][C(OP_WRITE)][C(RESULT_MISS)]	= ARMV8_PMUV3_PERFCTR_BR_MIS_PRED,
 };
 
 static const unsigned armv8_a53_perf_cache_map[PERF_COUNT_HW_CACHE_MAX]
@@ -173,21 +173,21 @@ static const unsigned armv8_a53_perf_cache_map[PERF_COUNT_HW_CACHE_MAX]
 					      [PERF_COUNT_HW_CACHE_RESULT_MAX] = {
 	PERF_CACHE_MAP_ALL_UNSUPPORTED,
 
-	[C(L1D)][C(OP_READ)][C(RESULT_ACCESS)]	= ARMV8_PMUV3_PERFCTR_L1_DCACHE_ACCESS,
-	[C(L1D)][C(OP_READ)][C(RESULT_MISS)]	= ARMV8_PMUV3_PERFCTR_L1_DCACHE_REFILL,
-	[C(L1D)][C(OP_WRITE)][C(RESULT_ACCESS)]	= ARMV8_PMUV3_PERFCTR_L1_DCACHE_ACCESS,
-	[C(L1D)][C(OP_WRITE)][C(RESULT_MISS)]	= ARMV8_PMUV3_PERFCTR_L1_DCACHE_REFILL,
+	[C(L1D)][C(OP_READ)][C(RESULT_ACCESS)]	= ARMV8_PMUV3_PERFCTR_L1D_CACHE,
+	[C(L1D)][C(OP_READ)][C(RESULT_MISS)]	= ARMV8_PMUV3_PERFCTR_L1D_CACHE_REFILL,
+	[C(L1D)][C(OP_WRITE)][C(RESULT_ACCESS)]	= ARMV8_PMUV3_PERFCTR_L1D_CACHE,
+	[C(L1D)][C(OP_WRITE)][C(RESULT_MISS)]	= ARMV8_PMUV3_PERFCTR_L1D_CACHE_REFILL,
 	[C(L1D)][C(OP_PREFETCH)][C(RESULT_MISS)] = ARMV8_A53_PERFCTR_PREFETCH_LINEFILL,
 
-	[C(L1I)][C(OP_READ)][C(RESULT_ACCESS)]	= ARMV8_PMUV3_PERFCTR_L1_ICACHE_ACCESS,
-	[C(L1I)][C(OP_READ)][C(RESULT_MISS)]	= ARMV8_PMUV3_PERFCTR_L1_ICACHE_REFILL,
+	[C(L1I)][C(OP_READ)][C(RESULT_ACCESS)]	= ARMV8_PMUV3_PERFCTR_L1I_CACHE,
+	[C(L1I)][C(OP_READ)][C(RESULT_MISS)]	= ARMV8_PMUV3_PERFCTR_L1I_CACHE_REFILL,
 
-	[C(ITLB)][C(OP_READ)][C(RESULT_MISS)]	= ARMV8_PMUV3_PERFCTR_ITLB_REFILL,
+	[C(ITLB)][C(OP_READ)][C(RESULT_MISS)]	= ARMV8_PMUV3_PERFCTR_L1I_TLB_REFILL,
 
-	[C(BPU)][C(OP_READ)][C(RESULT_ACCESS)]	= ARMV8_PMUV3_PERFCTR_PC_BRANCH_PRED,
-	[C(BPU)][C(OP_READ)][C(RESULT_MISS)]	= ARMV8_PMUV3_PERFCTR_PC_BRANCH_MIS_PRED,
-	[C(BPU)][C(OP_WRITE)][C(RESULT_ACCESS)]	= ARMV8_PMUV3_PERFCTR_PC_BRANCH_PRED,
-	[C(BPU)][C(OP_WRITE)][C(RESULT_MISS)]	= ARMV8_PMUV3_PERFCTR_PC_BRANCH_MIS_PRED,
+	[C(BPU)][C(OP_READ)][C(RESULT_ACCESS)]	= ARMV8_PMUV3_PERFCTR_BR_PRED,
+	[C(BPU)][C(OP_READ)][C(RESULT_MISS)]	= ARMV8_PMUV3_PERFCTR_BR_MIS_PRED,
+	[C(BPU)][C(OP_WRITE)][C(RESULT_ACCESS)]	= ARMV8_PMUV3_PERFCTR_BR_PRED,
+	[C(BPU)][C(OP_WRITE)][C(RESULT_MISS)]	= ARMV8_PMUV3_PERFCTR_BR_MIS_PRED,
 };
 
 static const unsigned armv8_a57_perf_cache_map[PERF_COUNT_HW_CACHE_MAX]
@@ -195,23 +195,23 @@ static const unsigned armv8_a57_perf_cache_map[PERF_COUNT_HW_CACHE_MAX]
 					      [PERF_COUNT_HW_CACHE_RESULT_MAX] = {
 	PERF_CACHE_MAP_ALL_UNSUPPORTED,
 
-	[C(L1D)][C(OP_READ)][C(RESULT_ACCESS)]	= ARMV8_IMPDEF_PERFCTR_L1_DCACHE_ACCESS_LD,
-	[C(L1D)][C(OP_READ)][C(RESULT_MISS)]	= ARMV8_IMPDEF_PERFCTR_L1_DCACHE_REFILL_LD,
-	[C(L1D)][C(OP_WRITE)][C(RESULT_ACCESS)]	= ARMV8_IMPDEF_PERFCTR_L1_DCACHE_ACCESS_ST,
-	[C(L1D)][C(OP_WRITE)][C(RESULT_MISS)]	= ARMV8_IMPDEF_PERFCTR_L1_DCACHE_REFILL_ST,
+	[C(L1D)][C(OP_READ)][C(RESULT_ACCESS)]	= ARMV8_IMPDEF_PERFCTR_L1D_CACHE_RD,
+	[C(L1D)][C(OP_READ)][C(RESULT_MISS)]	= ARMV8_IMPDEF_PERFCTR_L1D_CACHE_REFILL_RD,
+	[C(L1D)][C(OP_WRITE)][C(RESULT_ACCESS)]	= ARMV8_IMPDEF_PERFCTR_L1D_CACHE_WR,
+	[C(L1D)][C(OP_WRITE)][C(RESULT_MISS)]	= ARMV8_IMPDEF_PERFCTR_L1D_CACHE_REFILL_WR,
 
-	[C(L1I)][C(OP_READ)][C(RESULT_ACCESS)]	= ARMV8_PMUV3_PERFCTR_L1_ICACHE_ACCESS,
-	[C(L1I)][C(OP_READ)][C(RESULT_MISS)]	= ARMV8_PMUV3_PERFCTR_L1_ICACHE_REFILL,
+	[C(L1I)][C(OP_READ)][C(RESULT_ACCESS)]	= ARMV8_PMUV3_PERFCTR_L1I_CACHE,
+	[C(L1I)][C(OP_READ)][C(RESULT_MISS)]	= ARMV8_PMUV3_PERFCTR_L1I_CACHE_REFILL,
 
-	[C(DTLB)][C(OP_READ)][C(RESULT_MISS)]	= ARMV8_IMPDEF_PERFCTR_DTLB_REFILL_LD,
-	[C(DTLB)][C(OP_WRITE)][C(RESULT_MISS)]	= ARMV8_IMPDEF_PERFCTR_DTLB_REFILL_ST,
+	[C(DTLB)][C(OP_READ)][C(RESULT_MISS)]	= ARMV8_IMPDEF_PERFCTR_L1D_TLB_REFILL_RD,
+	[C(DTLB)][C(OP_WRITE)][C(RESULT_MISS)]	= ARMV8_IMPDEF_PERFCTR_L1D_TLB_REFILL_WR,
 
-	[C(ITLB)][C(OP_READ)][C(RESULT_MISS)]	= ARMV8_PMUV3_PERFCTR_ITLB_REFILL,
+	[C(ITLB)][C(OP_READ)][C(RESULT_MISS)]	= ARMV8_PMUV3_PERFCTR_L1I_TLB_REFILL,
 
-	[C(BPU)][C(OP_READ)][C(RESULT_ACCESS)]	= ARMV8_PMUV3_PERFCTR_PC_BRANCH_PRED,
-	[C(BPU)][C(OP_READ)][C(RESULT_MISS)]	= ARMV8_PMUV3_PERFCTR_PC_BRANCH_MIS_PRED,
-	[C(BPU)][C(OP_WRITE)][C(RESULT_ACCESS)]	= ARMV8_PMUV3_PERFCTR_PC_BRANCH_PRED,
-	[C(BPU)][C(OP_WRITE)][C(RESULT_MISS)]	= ARMV8_PMUV3_PERFCTR_PC_BRANCH_MIS_PRED,
+	[C(BPU)][C(OP_READ)][C(RESULT_ACCESS)]	= ARMV8_PMUV3_PERFCTR_BR_PRED,
+	[C(BPU)][C(OP_READ)][C(RESULT_MISS)]	= ARMV8_PMUV3_PERFCTR_BR_MIS_PRED,
+	[C(BPU)][C(OP_WRITE)][C(RESULT_ACCESS)]	= ARMV8_PMUV3_PERFCTR_BR_PRED,
+	[C(BPU)][C(OP_WRITE)][C(RESULT_MISS)]	= ARMV8_PMUV3_PERFCTR_BR_MIS_PRED,
 };
 
 static const unsigned armv8_thunder_perf_cache_map[PERF_COUNT_HW_CACHE_MAX]
@@ -219,29 +219,29 @@ static const unsigned armv8_thunder_perf_cache_map[PERF_COUNT_HW_CACHE_MAX]
 						   [PERF_COUNT_HW_CACHE_RESULT_MAX] = {
 	PERF_CACHE_MAP_ALL_UNSUPPORTED,
 
-	[C(L1D)][C(OP_READ)][C(RESULT_ACCESS)]	= ARMV8_IMPDEF_PERFCTR_L1_DCACHE_ACCESS_LD,
-	[C(L1D)][C(OP_READ)][C(RESULT_MISS)]	= ARMV8_IMPDEF_PERFCTR_L1_DCACHE_REFILL_LD,
-	[C(L1D)][C(OP_WRITE)][C(RESULT_ACCESS)]	= ARMV8_IMPDEF_PERFCTR_L1_DCACHE_ACCESS_ST,
-	[C(L1D)][C(OP_WRITE)][C(RESULT_MISS)]	= ARMV8_THUNDER_PERFCTR_L1_DCACHE_MISS_ST,
-	[C(L1D)][C(OP_PREFETCH)][C(RESULT_ACCESS)] = ARMV8_THUNDER_PERFCTR_L1_DCACHE_PREF_ACCESS,
-	[C(L1D)][C(OP_PREFETCH)][C(RESULT_MISS)] = ARMV8_THUNDER_PERFCTR_L1_DCACHE_PREF_MISS,
-
-	[C(L1I)][C(OP_READ)][C(RESULT_ACCESS)]	= ARMV8_PMUV3_PERFCTR_L1_ICACHE_ACCESS,
-	[C(L1I)][C(OP_READ)][C(RESULT_MISS)]	= ARMV8_PMUV3_PERFCTR_L1_ICACHE_REFILL,
-	[C(L1I)][C(OP_PREFETCH)][C(RESULT_ACCESS)] = ARMV8_THUNDER_PERFCTR_L1_ICACHE_PREF_ACCESS,
-	[C(L1I)][C(OP_PREFETCH)][C(RESULT_MISS)] = ARMV8_THUNDER_PERFCTR_L1_ICACHE_PREF_MISS,
-
-	[C(DTLB)][C(OP_READ)][C(RESULT_ACCESS)]	= ARMV8_IMPDEF_PERFCTR_DTLB_ACCESS_LD,
-	[C(DTLB)][C(OP_READ)][C(RESULT_MISS)]	= ARMV8_IMPDEF_PERFCTR_DTLB_REFILL_LD,
-	[C(DTLB)][C(OP_WRITE)][C(RESULT_ACCESS)] = ARMV8_IMPDEF_PERFCTR_DTLB_ACCESS_ST,
-	[C(DTLB)][C(OP_WRITE)][C(RESULT_MISS)]	= ARMV8_IMPDEF_PERFCTR_DTLB_REFILL_ST,
-
-	[C(ITLB)][C(OP_READ)][C(RESULT_MISS)]	= ARMV8_PMUV3_PERFCTR_ITLB_REFILL,
-
-	[C(BPU)][C(OP_READ)][C(RESULT_ACCESS)]	= ARMV8_PMUV3_PERFCTR_PC_BRANCH_PRED,
-	[C(BPU)][C(OP_READ)][C(RESULT_MISS)]	= ARMV8_PMUV3_PERFCTR_PC_BRANCH_MIS_PRED,
-	[C(BPU)][C(OP_WRITE)][C(RESULT_ACCESS)]	= ARMV8_PMUV3_PERFCTR_PC_BRANCH_PRED,
-	[C(BPU)][C(OP_WRITE)][C(RESULT_MISS)]	= ARMV8_PMUV3_PERFCTR_PC_BRANCH_MIS_PRED,
+	[C(L1D)][C(OP_READ)][C(RESULT_ACCESS)]	= ARMV8_IMPDEF_PERFCTR_L1D_CACHE_RD,
+	[C(L1D)][C(OP_READ)][C(RESULT_MISS)]	= ARMV8_IMPDEF_PERFCTR_L1D_CACHE_REFILL_RD,
+	[C(L1D)][C(OP_WRITE)][C(RESULT_ACCESS)]	= ARMV8_IMPDEF_PERFCTR_L1D_CACHE_WR,
+	[C(L1D)][C(OP_WRITE)][C(RESULT_MISS)]	= ARMV8_THUNDER_PERFCTR_L1D_CACHE_MISS_ST,
+	[C(L1D)][C(OP_PREFETCH)][C(RESULT_ACCESS)] = ARMV8_THUNDER_PERFCTR_L1D_CACHE_PREF_ACCESS,
+	[C(L1D)][C(OP_PREFETCH)][C(RESULT_MISS)] = ARMV8_THUNDER_PERFCTR_L1D_CACHE_PREF_MISS,
+
+	[C(L1I)][C(OP_READ)][C(RESULT_ACCESS)]	= ARMV8_PMUV3_PERFCTR_L1I_CACHE,
+	[C(L1I)][C(OP_READ)][C(RESULT_MISS)]	= ARMV8_PMUV3_PERFCTR_L1I_CACHE_REFILL,
+	[C(L1I)][C(OP_PREFETCH)][C(RESULT_ACCESS)] = ARMV8_THUNDER_PERFCTR_L1I_CACHE_PREF_ACCESS,
+	[C(L1I)][C(OP_PREFETCH)][C(RESULT_MISS)] = ARMV8_THUNDER_PERFCTR_L1I_CACHE_PREF_MISS,
+
+	[C(DTLB)][C(OP_READ)][C(RESULT_ACCESS)]	= ARMV8_IMPDEF_PERFCTR_L1D_TLB_RD,
+	[C(DTLB)][C(OP_READ)][C(RESULT_MISS)]	= ARMV8_IMPDEF_PERFCTR_L1D_TLB_REFILL_RD,
+	[C(DTLB)][C(OP_WRITE)][C(RESULT_ACCESS)] = ARMV8_IMPDEF_PERFCTR_L1D_TLB_WR,
+	[C(DTLB)][C(OP_WRITE)][C(RESULT_MISS)]	= ARMV8_IMPDEF_PERFCTR_L1D_TLB_REFILL_WR,
+
+	[C(ITLB)][C(OP_READ)][C(RESULT_MISS)]	= ARMV8_PMUV3_PERFCTR_L1I_TLB_REFILL,
+
+	[C(BPU)][C(OP_READ)][C(RESULT_ACCESS)]	= ARMV8_PMUV3_PERFCTR_BR_PRED,
+	[C(BPU)][C(OP_READ)][C(RESULT_MISS)]	= ARMV8_PMUV3_PERFCTR_BR_MIS_PRED,
+	[C(BPU)][C(OP_WRITE)][C(RESULT_ACCESS)]	= ARMV8_PMUV3_PERFCTR_BR_PRED,
+	[C(BPU)][C(OP_WRITE)][C(RESULT_MISS)]	= ARMV8_PMUV3_PERFCTR_BR_MIS_PRED,
 };
 
 #define ARMV8_EVENT_ATTR_RESOLVE(m) #m
@@ -249,35 +249,35 @@ static const unsigned armv8_thunder_perf_cache_map[PERF_COUNT_HW_CACHE_MAX]
 	PMU_EVENT_ATTR_STRING(name, armv8_event_attr_##name, \
 			      "event=" ARMV8_EVENT_ATTR_RESOLVE(config))
 
-ARMV8_EVENT_ATTR(sw_incr, ARMV8_PMUV3_PERFCTR_PMNC_SW_INCR);
-ARMV8_EVENT_ATTR(l1i_cache_refill, ARMV8_PMUV3_PERFCTR_L1_ICACHE_REFILL);
-ARMV8_EVENT_ATTR(l1i_tlb_refill, ARMV8_PMUV3_PERFCTR_ITLB_REFILL);
-ARMV8_EVENT_ATTR(l1d_cache_refill, ARMV8_PMUV3_PERFCTR_L1_DCACHE_REFILL);
-ARMV8_EVENT_ATTR(l1d_cache, ARMV8_PMUV3_PERFCTR_L1_DCACHE_ACCESS);
-ARMV8_EVENT_ATTR(l1d_tlb_refill, ARMV8_PMUV3_PERFCTR_DTLB_REFILL);
-ARMV8_EVENT_ATTR(ld_retired, ARMV8_PMUV3_PERFCTR_MEM_READ);
-ARMV8_EVENT_ATTR(st_retired, ARMV8_PMUV3_PERFCTR_MEM_WRITE);
-ARMV8_EVENT_ATTR(inst_retired, ARMV8_PMUV3_PERFCTR_INSTR_EXECUTED);
+ARMV8_EVENT_ATTR(sw_incr, ARMV8_PMUV3_PERFCTR_SW_INCR);
+ARMV8_EVENT_ATTR(l1i_cache_refill, ARMV8_PMUV3_PERFCTR_L1I_CACHE_REFILL);
+ARMV8_EVENT_ATTR(l1i_tlb_refill, ARMV8_PMUV3_PERFCTR_L1I_TLB_REFILL);
+ARMV8_EVENT_ATTR(l1d_cache_refill, ARMV8_PMUV3_PERFCTR_L1D_CACHE_REFILL);
+ARMV8_EVENT_ATTR(l1d_cache, ARMV8_PMUV3_PERFCTR_L1D_CACHE);
+ARMV8_EVENT_ATTR(l1d_tlb_refill, ARMV8_PMUV3_PERFCTR_L1D_TLB_REFILL);
+ARMV8_EVENT_ATTR(ld_retired, ARMV8_PMUV3_PERFCTR_LD_RETIRED);
+ARMV8_EVENT_ATTR(st_retired, ARMV8_PMUV3_PERFCTR_ST_RETIRED);
+ARMV8_EVENT_ATTR(inst_retired, ARMV8_PMUV3_PERFCTR_INST_RETIRED);
 ARMV8_EVENT_ATTR(exc_taken, ARMV8_PMUV3_PERFCTR_EXC_TAKEN);
-ARMV8_EVENT_ATTR(exc_return, ARMV8_PMUV3_PERFCTR_EXC_EXECUTED);
-ARMV8_EVENT_ATTR(cid_write_retired, ARMV8_PMUV3_PERFCTR_CID_WRITE);
-ARMV8_EVENT_ATTR(pc_write_retired, ARMV8_PMUV3_PERFCTR_PC_WRITE);
-ARMV8_EVENT_ATTR(br_immed_retired, ARMV8_PMUV3_PERFCTR_PC_IMM_BRANCH);
-ARMV8_EVENT_ATTR(br_return_retired, ARMV8_PMUV3_PERFCTR_PC_PROC_RETURN);
-ARMV8_EVENT_ATTR(unaligned_ldst_retired, ARMV8_PMUV3_PERFCTR_MEM_UNALIGNED_ACCESS);
-ARMV8_EVENT_ATTR(br_mis_pred, ARMV8_PMUV3_PERFCTR_PC_BRANCH_MIS_PRED);
-ARMV8_EVENT_ATTR(cpu_cycles, ARMV8_PMUV3_PERFCTR_CLOCK_CYCLES);
-ARMV8_EVENT_ATTR(br_pred, ARMV8_PMUV3_PERFCTR_PC_BRANCH_PRED);
+ARMV8_EVENT_ATTR(exc_return, ARMV8_PMUV3_PERFCTR_EXC_RETURN);
+ARMV8_EVENT_ATTR(cid_write_retired, ARMV8_PMUV3_PERFCTR_CID_WRITE_RETIRED);
+ARMV8_EVENT_ATTR(pc_write_retired, ARMV8_PMUV3_PERFCTR_PC_WRITE_RETIRED);
+ARMV8_EVENT_ATTR(br_immed_retired, ARMV8_PMUV3_PERFCTR_BR_IMMED_RETIRED);
+ARMV8_EVENT_ATTR(br_return_retired, ARMV8_PMUV3_PERFCTR_BR_RETURN_RETIRED);
+ARMV8_EVENT_ATTR(unaligned_ldst_retired, ARMV8_PMUV3_PERFCTR_UNALIGNED_LDST_RETIRED);
+ARMV8_EVENT_ATTR(br_mis_pred, ARMV8_PMUV3_PERFCTR_BR_MIS_PRED);
+ARMV8_EVENT_ATTR(cpu_cycles, ARMV8_PMUV3_PERFCTR_CPU_CYCLES);
+ARMV8_EVENT_ATTR(br_pred, ARMV8_PMUV3_PERFCTR_BR_PRED);
 ARMV8_EVENT_ATTR(mem_access, ARMV8_PMUV3_PERFCTR_MEM_ACCESS);
-ARMV8_EVENT_ATTR(l1i_cache, ARMV8_PMUV3_PERFCTR_L1_ICACHE_ACCESS);
-ARMV8_EVENT_ATTR(l1d_cache_wb, ARMV8_PMUV3_PERFCTR_L1_DCACHE_WB);
-ARMV8_EVENT_ATTR(l2d_cache, ARMV8_PMUV3_PERFCTR_L2_CACHE_ACCESS);
-ARMV8_EVENT_ATTR(l2d_cache_refill, ARMV8_PMUV3_PERFCTR_L2_CACHE_REFILL);
-ARMV8_EVENT_ATTR(l2d_cache_wb, ARMV8_PMUV3_PERFCTR_L2_CACHE_WB);
+ARMV8_EVENT_ATTR(l1i_cache, ARMV8_PMUV3_PERFCTR_L1I_CACHE);
+ARMV8_EVENT_ATTR(l1d_cache_wb, ARMV8_PMUV3_PERFCTR_L1D_CACHE_WB);
+ARMV8_EVENT_ATTR(l2d_cache, ARMV8_PMUV3_PERFCTR_L2D_CACHE);
+ARMV8_EVENT_ATTR(l2d_cache_refill, ARMV8_PMUV3_PERFCTR_L2D_CACHE_REFILL);
+ARMV8_EVENT_ATTR(l2d_cache_wb, ARMV8_PMUV3_PERFCTR_L2D_CACHE_WB);
 ARMV8_EVENT_ATTR(bus_access, ARMV8_PMUV3_PERFCTR_BUS_ACCESS);
-ARMV8_EVENT_ATTR(memory_error, ARMV8_PMUV3_PERFCTR_MEM_ERROR);
-ARMV8_EVENT_ATTR(inst_spec, ARMV8_PMUV3_PERFCTR_OP_SPEC);
-ARMV8_EVENT_ATTR(ttbr_write_retired, ARMV8_PMUV3_PERFCTR_TTBR_WRITE);
+ARMV8_EVENT_ATTR(memory_error, ARMV8_PMUV3_PERFCTR_MEMORY_ERROR);
+ARMV8_EVENT_ATTR(inst_spec, ARMV8_PMUV3_PERFCTR_INST_SPEC);
+ARMV8_EVENT_ATTR(ttbr_write_retired, ARMV8_PMUV3_PERFCTR_TTBR_WRITE_RETIRED);
 ARMV8_EVENT_ATTR(bus_cycles, ARMV8_PMUV3_PERFCTR_BUS_CYCLES);
 ARMV8_EVENT_ATTR(chain, ARMV8_PMUV3_PERFCTR_CHAIN);
 ARMV8_EVENT_ATTR(l1d_cache_allocate, ARMV8_PMUV3_PERFCTR_L1D_CACHE_ALLOCATE);
@@ -295,9 +295,9 @@ ARMV8_EVENT_ATTR(l3d_cache_refill, ARMV8_PMUV3_PERFCTR_L3D_CACHE_REFILL);
 ARMV8_EVENT_ATTR(l3d_cache, ARMV8_PMUV3_PERFCTR_L3D_CACHE);
 ARMV8_EVENT_ATTR(l3d_cache_wb, ARMV8_PMUV3_PERFCTR_L3D_CACHE_WB);
 ARMV8_EVENT_ATTR(l2d_tlb_refill, ARMV8_PMUV3_PERFCTR_L2D_TLB_REFILL);
-ARMV8_EVENT_ATTR(l21_tlb_refill, ARMV8_PMUV3_PERFCTR_L21_TLB_REFILL);
+ARMV8_EVENT_ATTR(l2i_tlb_refill, ARMV8_PMUV3_PERFCTR_L2I_TLB_REFILL);
 ARMV8_EVENT_ATTR(l2d_tlb, ARMV8_PMUV3_PERFCTR_L2D_TLB);
-ARMV8_EVENT_ATTR(l21_tlb, ARMV8_PMUV3_PERFCTR_L21_TLB);
+ARMV8_EVENT_ATTR(l2i_tlb, ARMV8_PMUV3_PERFCTR_L2I_TLB);
 
 static struct attribute *armv8_pmuv3_event_attrs[] = {
 	&armv8_event_attr_sw_incr.attr.attr,
@@ -346,9 +346,9 @@ static struct attribute *armv8_pmuv3_event_attrs[] = {
 	&armv8_event_attr_l3d_cache.attr.attr,
 	&armv8_event_attr_l3d_cache_wb.attr.attr,
 	&armv8_event_attr_l2d_tlb_refill.attr.attr,
-	&armv8_event_attr_l21_tlb_refill.attr.attr,
+	&armv8_event_attr_l2i_tlb_refill.attr.attr,
 	&armv8_event_attr_l2d_tlb.attr.attr,
-	&armv8_event_attr_l21_tlb.attr.attr,
+	&armv8_event_attr_l2i_tlb.attr.attr,
 	NULL,
 };
 
@@ -719,7 +719,7 @@ static int armv8pmu_get_event_idx(struct pmu_hw_events *cpuc,
 	unsigned long evtype = hwc->config_base & ARMV8_EVTYPE_EVENT;
 
 	/* Always place a cycle counter into the cycle counter. */
-	if (evtype == ARMV8_PMUV3_PERFCTR_CLOCK_CYCLES) {
+	if (evtype == ARMV8_PMUV3_PERFCTR_CPU_CYCLES) {
 		if (test_and_set_bit(ARMV8_IDX_CYCLE_COUNTER, cpuc->used_mask))
 			return -EAGAIN;
 
-- 
2.1.0

^ permalink raw reply related	[flat|nested] 11+ messages in thread

* [PATCH v2 2/6] arm64/perf: Define complete ARMv8 recommended implementation defined events
  2016-03-24 12:52 [PATCH v2 0/6] arm64: perf: Broadcom Vulcan PMU support Ashok Kumar
  2016-03-24 12:52 ` [PATCH v2 1/6] arm64/perf: Changed events naming as per ARM ARM Ashok Kumar
@ 2016-03-24 12:52 ` Ashok Kumar
  2016-03-24 12:52 ` [PATCH v2 3/6] arm64/perf: Filter common events based on PMCEIDn_EL0 Ashok Kumar
                   ` (3 subsequent siblings)
  5 siblings, 0 replies; 11+ messages in thread
From: Ashok Kumar @ 2016-03-24 12:52 UTC (permalink / raw)
  To: linux-arm-kernel

Defined all the ARMv8 recommended implementation defined events
from J3 - "ARM recommendations for IMPLEMENTATION DEFINED event numbers"
in ARMv8 ARM.

Signed-off-by: Ashok Kumar <ashoks@broadcom.com>
---
 arch/arm64/kernel/perf_event.c | 79 ++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 79 insertions(+)

diff --git a/arch/arm64/kernel/perf_event.c b/arch/arm64/kernel/perf_event.c
index d1a93cf..5358587 100644
--- a/arch/arm64/kernel/perf_event.c
+++ b/arch/arm64/kernel/perf_event.c
@@ -92,10 +92,89 @@
 #define ARMV8_IMPDEF_PERFCTR_L1D_CACHE_WR			0x41
 #define ARMV8_IMPDEF_PERFCTR_L1D_CACHE_REFILL_RD		0x42
 #define ARMV8_IMPDEF_PERFCTR_L1D_CACHE_REFILL_WR		0x43
+#define ARMV8_IMPDEF_PERFCTR_L1D_CACHE_REFILL_INNER		0x44
+#define ARMV8_IMPDEF_PERFCTR_L1D_CACHE_REFILL_OUTER		0x45
+#define ARMV8_IMPDEF_PERFCTR_L1D_CACHE_WB_VICTIM		0x46
+#define ARMV8_IMPDEF_PERFCTR_L1D_CACHE_WB_CLEAN			0x47
+#define ARMV8_IMPDEF_PERFCTR_L1D_CACHE_INVAL			0x48
+
 #define ARMV8_IMPDEF_PERFCTR_L1D_TLB_REFILL_RD			0x4C
 #define ARMV8_IMPDEF_PERFCTR_L1D_TLB_REFILL_WR			0x4D
 #define ARMV8_IMPDEF_PERFCTR_L1D_TLB_RD				0x4E
 #define ARMV8_IMPDEF_PERFCTR_L1D_TLB_WR				0x4F
+#define ARMV8_IMPDEF_PERFCTR_L2D_CACHE_RD			0x50
+#define ARMV8_IMPDEF_PERFCTR_L2D_CACHE_WR			0x51
+#define ARMV8_IMPDEF_PERFCTR_L2D_CACHE_REFILL_RD		0x52
+#define ARMV8_IMPDEF_PERFCTR_L2D_CACHE_REFILL_WR		0x53
+
+#define ARMV8_IMPDEF_PERFCTR_L2D_CACHE_WB_VICTIM		0x56
+#define ARMV8_IMPDEF_PERFCTR_L2D_CACHE_WB_CLEAN			0x57
+#define ARMV8_IMPDEF_PERFCTR_L2D_CACHE_INVAL			0x58
+
+#define ARMV8_IMPDEF_PERFCTR_L2D_TLB_REFILL_RD			0x5C
+#define ARMV8_IMPDEF_PERFCTR_L2D_TLB_REFILL_WR			0x5D
+#define ARMV8_IMPDEF_PERFCTR_L2D_TLB_RD				0x5E
+#define ARMV8_IMPDEF_PERFCTR_L2D_TLB_WR				0x5F
+
+#define ARMV8_IMPDEF_PERFCTR_BUS_ACCESS_RD			0x60
+#define ARMV8_IMPDEF_PERFCTR_BUS_ACCESS_WR			0x61
+#define ARMV8_IMPDEF_PERFCTR_BUS_ACCESS_SHARED			0x62
+#define ARMV8_IMPDEF_PERFCTR_BUS_ACCESS_NOT_SHARED		0x63
+#define ARMV8_IMPDEF_PERFCTR_BUS_ACCESS_NORMAL			0x64
+#define ARMV8_IMPDEF_PERFCTR_BUS_ACCESS_PERIPH			0x65
+
+#define ARMV8_IMPDEF_PERFCTR_MEM_ACCESS_RD			0x66
+#define ARMV8_IMPDEF_PERFCTR_MEM_ACCESS_WR			0x67
+#define ARMV8_IMPDEF_PERFCTR_UNALIGNED_LD_SPEC			0x68
+#define ARMV8_IMPDEF_PERFCTR_UNALIGNED_ST_SPEC			0x69
+#define ARMV8_IMPDEF_PERFCTR_UNALIGNED_LDST_SPEC		0x6A
+
+#define ARMV8_IMPDEF_PERFCTR_LDREX_SPEC				0x6C
+#define ARMV8_IMPDEF_PERFCTR_STREX_PASS_SPEC			0x6D
+#define ARMV8_IMPDEF_PERFCTR_STREX_FAIL_SPEC			0x6E
+#define ARMV8_IMPDEF_PERFCTR_STREX_SPEC				0x6F
+#define ARMV8_IMPDEF_PERFCTR_LD_SPEC				0x70
+#define ARMV8_IMPDEF_PERFCTR_ST_SPEC				0x71
+#define ARMV8_IMPDEF_PERFCTR_LDST_SPEC				0x72
+#define ARMV8_IMPDEF_PERFCTR_DP_SPEC				0x73
+#define ARMV8_IMPDEF_PERFCTR_ASE_SPEC				0x74
+#define ARMV8_IMPDEF_PERFCTR_VFP_SPEC				0x75
+#define ARMV8_IMPDEF_PERFCTR_PC_WRITE_SPEC			0x76
+#define ARMV8_IMPDEF_PERFCTR_CRYPTO_SPEC			0x77
+#define ARMV8_IMPDEF_PERFCTR_BR_IMMED_SPEC			0x78
+#define ARMV8_IMPDEF_PERFCTR_BR_RETURN_SPEC			0x79
+#define ARMV8_IMPDEF_PERFCTR_BR_INDIRECT_SPEC			0x7A
+
+#define ARMV8_IMPDEF_PERFCTR_ISB_SPEC				0x7C
+#define ARMV8_IMPDEF_PERFCTR_DSB_SPEC				0x7D
+#define ARMV8_IMPDEF_PERFCTR_DMB_SPEC				0x7E
+
+#define ARMV8_IMPDEF_PERFCTR_EXC_UNDEF				0x81
+#define ARMV8_IMPDEF_PERFCTR_EXC_SVC				0x82
+#define ARMV8_IMPDEF_PERFCTR_EXC_PABORT				0x83
+#define ARMV8_IMPDEF_PERFCTR_EXC_DABORT				0x84
+
+#define ARMV8_IMPDEF_PERFCTR_EXC_IRQ				0x86
+#define ARMV8_IMPDEF_PERFCTR_EXC_FIQ				0x87
+#define ARMV8_IMPDEF_PERFCTR_EXC_SMC				0x88
+
+#define ARMV8_IMPDEF_PERFCTR_EXC_HVC				0x8A
+#define ARMV8_IMPDEF_PERFCTR_EXC_TRAP_PABORT			0x8B
+#define ARMV8_IMPDEF_PERFCTR_EXC_TRAP_DABORT			0x8C
+#define ARMV8_IMPDEF_PERFCTR_EXC_TRAP_OTHER			0x8D
+#define ARMV8_IMPDEF_PERFCTR_EXC_TRAP_IRQ			0x8E
+#define ARMV8_IMPDEF_PERFCTR_EXC_TRAP_FIQ			0x8F
+#define ARMV8_IMPDEF_PERFCTR_RC_LD_SPEC				0x90
+#define ARMV8_IMPDEF_PERFCTR_RC_ST_SPEC				0x91
+
+#define ARMV8_IMPDEF_PERFCTR_L3D_CACHE_RD			0xA0
+#define ARMV8_IMPDEF_PERFCTR_L3D_CACHE_WR			0xA1
+#define ARMV8_IMPDEF_PERFCTR_L3D_CACHE_REFILL_RD		0xA2
+#define ARMV8_IMPDEF_PERFCTR_L3D_CACHE_REFILL_WR		0xA3
+
+#define ARMV8_IMPDEF_PERFCTR_L3D_CACHE_WB_VICTIM		0xA6
+#define ARMV8_IMPDEF_PERFCTR_L3D_CACHE_WB_CLEAN			0xA7
+#define ARMV8_IMPDEF_PERFCTR_L3D_CACHE_INVAL			0xA8
 
 /* ARMv8 Cortex-A53 specific event types. */
 #define ARMV8_A53_PERFCTR_PREFETCH_LINEFILL			0xC2
-- 
2.1.0

^ permalink raw reply related	[flat|nested] 11+ messages in thread

* [PATCH v2 3/6] arm64/perf: Filter common events based on PMCEIDn_EL0
  2016-03-24 12:52 [PATCH v2 0/6] arm64: perf: Broadcom Vulcan PMU support Ashok Kumar
  2016-03-24 12:52 ` [PATCH v2 1/6] arm64/perf: Changed events naming as per ARM ARM Ashok Kumar
  2016-03-24 12:52 ` [PATCH v2 2/6] arm64/perf: Define complete ARMv8 recommended implementation defined events Ashok Kumar
@ 2016-03-24 12:52 ` Ashok Kumar
  2016-03-24 15:28   ` Suzuki K. Poulose
  2016-03-24 16:14   ` Mark Rutland
  2016-03-24 12:52 ` [PATCH v2 4/6] arm64/perf: Add Broadcom Vulcan PMU support Ashok Kumar
                   ` (2 subsequent siblings)
  5 siblings, 2 replies; 11+ messages in thread
From: Ashok Kumar @ 2016-03-24 12:52 UTC (permalink / raw)
  To: linux-arm-kernel

The complete common architectural and micro-architectural
event number structure is filtered based on PMCEIDn_EL0 and
copied to a new structure which is exposed to /sys

The function which derives event bitmap from PMCEIDn_EL0 is
executed in the cpus, which has the pmu being initialized,
for heterogeneous pmu support.

Enforced armv8_pmuv3_event_attrs array for event number
ordering as this is indexed with event number now.

Signed-off-by: Ashok Kumar <ashoks@broadcom.com>
---
 arch/arm64/kernel/perf_event.c | 281 ++++++++++++++++++++++++++++-------------
 1 file changed, 195 insertions(+), 86 deletions(-)

diff --git a/arch/arm64/kernel/perf_event.c b/arch/arm64/kernel/perf_event.c
index 5358587..a025ec2 100644
--- a/arch/arm64/kernel/perf_event.c
+++ b/arch/arm64/kernel/perf_event.c
@@ -378,62 +378,106 @@ ARMV8_EVENT_ATTR(l2i_tlb_refill, ARMV8_PMUV3_PERFCTR_L2I_TLB_REFILL);
 ARMV8_EVENT_ATTR(l2d_tlb, ARMV8_PMUV3_PERFCTR_L2D_TLB);
 ARMV8_EVENT_ATTR(l2i_tlb, ARMV8_PMUV3_PERFCTR_L2I_TLB);
 
-static struct attribute *armv8_pmuv3_event_attrs[] = {
-	&armv8_event_attr_sw_incr.attr.attr,
-	&armv8_event_attr_l1i_cache_refill.attr.attr,
-	&armv8_event_attr_l1i_tlb_refill.attr.attr,
-	&armv8_event_attr_l1d_cache_refill.attr.attr,
-	&armv8_event_attr_l1d_cache.attr.attr,
-	&armv8_event_attr_l1d_tlb_refill.attr.attr,
-	&armv8_event_attr_ld_retired.attr.attr,
-	&armv8_event_attr_st_retired.attr.attr,
-	&armv8_event_attr_inst_retired.attr.attr,
-	&armv8_event_attr_exc_taken.attr.attr,
-	&armv8_event_attr_exc_return.attr.attr,
-	&armv8_event_attr_cid_write_retired.attr.attr,
-	&armv8_event_attr_pc_write_retired.attr.attr,
-	&armv8_event_attr_br_immed_retired.attr.attr,
-	&armv8_event_attr_br_return_retired.attr.attr,
-	&armv8_event_attr_unaligned_ldst_retired.attr.attr,
-	&armv8_event_attr_br_mis_pred.attr.attr,
-	&armv8_event_attr_cpu_cycles.attr.attr,
-	&armv8_event_attr_br_pred.attr.attr,
-	&armv8_event_attr_mem_access.attr.attr,
-	&armv8_event_attr_l1i_cache.attr.attr,
-	&armv8_event_attr_l1d_cache_wb.attr.attr,
-	&armv8_event_attr_l2d_cache.attr.attr,
-	&armv8_event_attr_l2d_cache_refill.attr.attr,
-	&armv8_event_attr_l2d_cache_wb.attr.attr,
-	&armv8_event_attr_bus_access.attr.attr,
-	&armv8_event_attr_memory_error.attr.attr,
-	&armv8_event_attr_inst_spec.attr.attr,
-	&armv8_event_attr_ttbr_write_retired.attr.attr,
-	&armv8_event_attr_bus_cycles.attr.attr,
-	&armv8_event_attr_chain.attr.attr,
-	&armv8_event_attr_l1d_cache_allocate.attr.attr,
-	&armv8_event_attr_l2d_cache_allocate.attr.attr,
-	&armv8_event_attr_br_retired.attr.attr,
-	&armv8_event_attr_br_mis_pred_retired.attr.attr,
-	&armv8_event_attr_stall_frontend.attr.attr,
-	&armv8_event_attr_stall_backend.attr.attr,
-	&armv8_event_attr_l1d_tlb.attr.attr,
-	&armv8_event_attr_l1i_tlb.attr.attr,
-	&armv8_event_attr_l2i_cache.attr.attr,
-	&armv8_event_attr_l2i_cache_refill.attr.attr,
-	&armv8_event_attr_l3d_cache_allocate.attr.attr,
-	&armv8_event_attr_l3d_cache_refill.attr.attr,
-	&armv8_event_attr_l3d_cache.attr.attr,
-	&armv8_event_attr_l3d_cache_wb.attr.attr,
-	&armv8_event_attr_l2d_tlb_refill.attr.attr,
-	&armv8_event_attr_l2i_tlb_refill.attr.attr,
-	&armv8_event_attr_l2d_tlb.attr.attr,
-	&armv8_event_attr_l2i_tlb.attr.attr,
-	NULL,
-};
-
-static struct attribute_group armv8_pmuv3_events_attr_group = {
-	.name = "events",
-	.attrs = armv8_pmuv3_event_attrs,
+#define ARMV8_PMUV3_MAX_COMMON_EVENTS 0x40
+static struct attribute *armv8_pmuv3_event_attrs[ARMV8_PMUV3_MAX_COMMON_EVENTS] = {
+	[ARMV8_PMUV3_PERFCTR_SW_INCR] =
+		&armv8_event_attr_sw_incr.attr.attr,
+	[ARMV8_PMUV3_PERFCTR_L1I_CACHE_REFILL] =
+		&armv8_event_attr_l1i_cache_refill.attr.attr,
+	[ARMV8_PMUV3_PERFCTR_L1I_TLB_REFILL] =
+		&armv8_event_attr_l1i_tlb_refill.attr.attr,
+	[ARMV8_PMUV3_PERFCTR_L1D_CACHE_REFILL] =
+		&armv8_event_attr_l1d_cache_refill.attr.attr,
+	[ARMV8_PMUV3_PERFCTR_L1D_CACHE] =
+		&armv8_event_attr_l1d_cache.attr.attr,
+	[ARMV8_PMUV3_PERFCTR_L1D_TLB_REFILL] =
+		&armv8_event_attr_l1d_tlb_refill.attr.attr,
+	[ARMV8_PMUV3_PERFCTR_LD_RETIRED] =
+		&armv8_event_attr_ld_retired.attr.attr,
+	[ARMV8_PMUV3_PERFCTR_ST_RETIRED] =
+		&armv8_event_attr_st_retired.attr.attr,
+	[ARMV8_PMUV3_PERFCTR_INST_RETIRED] =
+		&armv8_event_attr_inst_retired.attr.attr,
+	[ARMV8_PMUV3_PERFCTR_EXC_TAKEN] =
+		&armv8_event_attr_exc_taken.attr.attr,
+	[ARMV8_PMUV3_PERFCTR_EXC_RETURN] =
+		&armv8_event_attr_exc_return.attr.attr,
+	[ARMV8_PMUV3_PERFCTR_CID_WRITE_RETIRED] =
+		&armv8_event_attr_cid_write_retired.attr.attr,
+	[ARMV8_PMUV3_PERFCTR_PC_WRITE_RETIRED] =
+		&armv8_event_attr_pc_write_retired.attr.attr,
+	[ARMV8_PMUV3_PERFCTR_BR_IMMED_RETIRED] =
+		&armv8_event_attr_br_immed_retired.attr.attr,
+	[ARMV8_PMUV3_PERFCTR_BR_RETURN_RETIRED] =
+		&armv8_event_attr_br_return_retired.attr.attr,
+	[ARMV8_PMUV3_PERFCTR_UNALIGNED_LDST_RETIRED] =
+		&armv8_event_attr_unaligned_ldst_retired.attr.attr,
+	[ARMV8_PMUV3_PERFCTR_BR_MIS_PRED] =
+		&armv8_event_attr_br_mis_pred.attr.attr,
+	[ARMV8_PMUV3_PERFCTR_CPU_CYCLES] =
+		&armv8_event_attr_cpu_cycles.attr.attr,
+	[ARMV8_PMUV3_PERFCTR_BR_PRED] =
+		&armv8_event_attr_br_pred.attr.attr,
+	[ARMV8_PMUV3_PERFCTR_MEM_ACCESS] =
+		&armv8_event_attr_mem_access.attr.attr,
+	[ARMV8_PMUV3_PERFCTR_L1I_CACHE] =
+		&armv8_event_attr_l1i_cache.attr.attr,
+	[ARMV8_PMUV3_PERFCTR_L1D_CACHE_WB] =
+		&armv8_event_attr_l1d_cache_wb.attr.attr,
+	[ARMV8_PMUV3_PERFCTR_L2D_CACHE] =
+		&armv8_event_attr_l2d_cache.attr.attr,
+	[ARMV8_PMUV3_PERFCTR_L2D_CACHE_REFILL] =
+		&armv8_event_attr_l2d_cache_refill.attr.attr,
+	[ARMV8_PMUV3_PERFCTR_L2D_CACHE_WB] =
+		&armv8_event_attr_l2d_cache_wb.attr.attr,
+	[ARMV8_PMUV3_PERFCTR_BUS_ACCESS] =
+		&armv8_event_attr_bus_access.attr.attr,
+	[ARMV8_PMUV3_PERFCTR_MEMORY_ERROR] =
+		&armv8_event_attr_memory_error.attr.attr,
+	[ARMV8_PMUV3_PERFCTR_INST_SPEC] =
+		&armv8_event_attr_inst_spec.attr.attr,
+	[ARMV8_PMUV3_PERFCTR_TTBR_WRITE_RETIRED] =
+		&armv8_event_attr_ttbr_write_retired.attr.attr,
+	[ARMV8_PMUV3_PERFCTR_BUS_CYCLES] =
+		&armv8_event_attr_bus_cycles.attr.attr,
+	[ARMV8_PMUV3_PERFCTR_CHAIN] =
+		&armv8_event_attr_chain.attr.attr,
+	[ARMV8_PMUV3_PERFCTR_L1D_CACHE_ALLOCATE] =
+		&armv8_event_attr_l1d_cache_allocate.attr.attr,
+	[ARMV8_PMUV3_PERFCTR_L2D_CACHE_ALLOCATE] =
+		&armv8_event_attr_l2d_cache_allocate.attr.attr,
+	[ARMV8_PMUV3_PERFCTR_BR_RETIRED] =
+		&armv8_event_attr_br_retired.attr.attr,
+	[ARMV8_PMUV3_PERFCTR_BR_MIS_PRED_RETIRED] =
+		&armv8_event_attr_br_mis_pred_retired.attr.attr,
+	[ARMV8_PMUV3_PERFCTR_STALL_FRONTEND] =
+		&armv8_event_attr_stall_frontend.attr.attr,
+	[ARMV8_PMUV3_PERFCTR_STALL_BACKEND] =
+		&armv8_event_attr_stall_backend.attr.attr,
+	[ARMV8_PMUV3_PERFCTR_L1D_TLB] =
+		&armv8_event_attr_l1d_tlb.attr.attr,
+	[ARMV8_PMUV3_PERFCTR_L1I_TLB] =
+		&armv8_event_attr_l1i_tlb.attr.attr,
+	[ARMV8_PMUV3_PERFCTR_L2I_CACHE] =
+		&armv8_event_attr_l2i_cache.attr.attr,
+	[ARMV8_PMUV3_PERFCTR_L2I_CACHE_REFILL] =
+		&armv8_event_attr_l2i_cache_refill.attr.attr,
+	[ARMV8_PMUV3_PERFCTR_L3D_CACHE_ALLOCATE] =
+		&armv8_event_attr_l3d_cache_allocate.attr.attr,
+	[ARMV8_PMUV3_PERFCTR_L3D_CACHE_REFILL] =
+		&armv8_event_attr_l3d_cache_refill.attr.attr,
+	[ARMV8_PMUV3_PERFCTR_L3D_CACHE] =
+		&armv8_event_attr_l3d_cache.attr.attr,
+	[ARMV8_PMUV3_PERFCTR_L3D_CACHE_WB] =
+		&armv8_event_attr_l3d_cache_wb.attr.attr,
+	[ARMV8_PMUV3_PERFCTR_L2D_TLB_REFILL] =
+		&armv8_event_attr_l2d_tlb_refill.attr.attr,
+	[ARMV8_PMUV3_PERFCTR_L2I_TLB_REFILL] =
+		&armv8_event_attr_l2i_tlb_refill.attr.attr,
+	[ARMV8_PMUV3_PERFCTR_L2D_TLB] =
+		&armv8_event_attr_l2d_tlb.attr.attr,
+	[ARMV8_PMUV3_PERFCTR_L2I_TLB] =
+		&armv8_event_attr_l2i_tlb.attr.attr,
 };
 
 PMU_FORMAT_ATTR(event, "config:0-9");
@@ -448,12 +492,6 @@ static struct attribute_group armv8_pmuv3_format_attr_group = {
 	.attrs = armv8_pmuv3_format_attrs,
 };
 
-static const struct attribute_group *armv8_pmuv3_attr_groups[] = {
-	&armv8_pmuv3_events_attr_group,
-	&armv8_pmuv3_format_attr_group,
-	NULL,
-};
-
 /*
  * Perf Events' indices
  */
@@ -522,6 +560,18 @@ static inline void armv8pmu_pmcr_write(u32 val)
 	asm volatile("msr pmcr_el0, %0" :: "r" (val));
 }
 
+static inline u32 armv8pmu_pmceidn_read(int reg)
+{
+	u32 val = 0;
+
+	if (reg == 0)
+		asm volatile("mrs %0, pmceid0_el0" : "=r" (val));
+	else if (reg == 1)
+		asm volatile("mrs %0, pmceid1_el0" : "=r" (val));
+
+	return val;
+}
+
 static inline int armv8pmu_has_overflowed(u32 pmovsr)
 {
 	return pmovsr & ARMV8_OVERFLOWED_MASK;
@@ -890,26 +940,53 @@ static int armv8_thunder_map_event(struct perf_event *event)
 				ARMV8_EVTYPE_EVENT);
 }
 
-static void armv8pmu_read_num_pmnc_events(void *info)
+static unsigned int armv8pmu_read_num_pmnc_events(void)
 {
-	int *nb_cnt = info;
-
+	unsigned int nb_cnt;
 	/* Read the nb of CNTx counters supported from PMNC */
-	*nb_cnt = (armv8pmu_pmcr_read() >> ARMV8_PMCR_N_SHIFT) & ARMV8_PMCR_N_MASK;
+	nb_cnt = (armv8pmu_pmcr_read() >> ARMV8_PMCR_N_SHIFT) & ARMV8_PMCR_N_MASK;
 
 	/* Add the CPU cycles counter */
-	*nb_cnt += 1;
+	nb_cnt += 1;
+
+	return nb_cnt;
+}
+
+static void armv8pmu_read_common_events_bitmap(unsigned long *bmp)
+{
+	u32 reg[2];
+
+	reg[0] = armv8pmu_pmceidn_read(0);
+	reg[1] = armv8pmu_pmceidn_read(1);
+
+	bitmap_from_u32array(bmp, ARMV8_PMUV3_MAX_COMMON_EVENTS,
+			     reg, ARRAY_SIZE(reg));
+	return;
 }
 
-static int armv8pmu_probe_num_events(struct arm_pmu *arm_pmu)
+struct armv8pmu_probe_pmu_data {
+	int num_evt_cntrs;
+	DECLARE_BITMAP(events_bitmap, ARMV8_PMUV3_MAX_COMMON_EVENTS);
+};
+
+static void armv8pmu_probe_pmu(void *info)
 {
-	return smp_call_function_any(&arm_pmu->supported_cpus,
-				    armv8pmu_read_num_pmnc_events,
-				    &arm_pmu->num_events, 1);
+	struct armv8pmu_probe_pmu_data *data = info;
+
+	data->num_evt_cntrs = armv8pmu_read_num_pmnc_events();
+	armv8pmu_read_common_events_bitmap(data->events_bitmap);
 }
 
-static void armv8_pmu_init(struct arm_pmu *cpu_pmu)
+static int armv8_pmu_init(struct arm_pmu *cpu_pmu)
 {
+	u32 evt, idx = 0;
+	struct attribute **event_attrs;
+	struct attribute_group *events_attr_group;
+	const struct attribute_group **attr_groups;
+	struct device *dev = &cpu_pmu->plat_device->dev;
+	struct armv8pmu_probe_pmu_data pmu_data;
+	int error;
+
 	cpu_pmu->handle_irq		= armv8pmu_handle_irq,
 	cpu_pmu->enable			= armv8pmu_enable_event,
 	cpu_pmu->disable		= armv8pmu_disable_event,
@@ -921,50 +998,82 @@ static void armv8_pmu_init(struct arm_pmu *cpu_pmu)
 	cpu_pmu->reset			= armv8pmu_reset,
 	cpu_pmu->max_period		= (1LLU << 32) - 1,
 	cpu_pmu->set_event_filter	= armv8pmu_set_event_filter;
+
+	error = smp_call_function_any(&cpu_pmu->supported_cpus,
+			      armv8pmu_probe_pmu,
+			      &pmu_data, 1);
+	if (error)
+		goto out;
+
+	cpu_pmu->num_events = pmu_data.num_evt_cntrs;
+
+	event_attrs = devm_kcalloc(dev, bitmap_weight(pmu_data.events_bitmap,
+				   ARMV8_PMUV3_MAX_COMMON_EVENTS) + 1,
+				   sizeof(*event_attrs), GFP_KERNEL);
+	if (!event_attrs)
+		goto mem_out;
+
+	events_attr_group = devm_kzalloc(dev, sizeof(*events_attr_group),
+					 GFP_KERNEL);
+	if (!events_attr_group)
+		goto mem_out;
+
+	attr_groups = devm_kcalloc(dev, 3, sizeof(*attr_groups), GFP_KERNEL);
+	if (!attr_groups)
+		goto mem_out;
+
+	for_each_set_bit(evt, pmu_data.events_bitmap,
+			 ARMV8_PMUV3_MAX_COMMON_EVENTS)
+		event_attrs[idx++] = armv8_pmuv3_event_attrs[evt];
+
+	events_attr_group->name = "events";
+	events_attr_group->attrs = event_attrs;
+
+	attr_groups[0] = events_attr_group;
+	attr_groups[1] = &armv8_pmuv3_format_attr_group;
+
+	cpu_pmu->pmu.attr_groups = attr_groups;
+
+	return 0;
+mem_out:
+	error = -ENOMEM;
+out:
+	return error;
 }
 
 static int armv8_pmuv3_init(struct arm_pmu *cpu_pmu)
 {
-	armv8_pmu_init(cpu_pmu);
 	cpu_pmu->name			= "armv8_pmuv3";
 	cpu_pmu->map_event		= armv8_pmuv3_map_event;
-	return armv8pmu_probe_num_events(cpu_pmu);
+	return armv8_pmu_init(cpu_pmu);
 }
 
 static int armv8_a53_pmu_init(struct arm_pmu *cpu_pmu)
 {
-	armv8_pmu_init(cpu_pmu);
 	cpu_pmu->name			= "armv8_cortex_a53";
 	cpu_pmu->map_event		= armv8_a53_map_event;
-	cpu_pmu->pmu.attr_groups	= armv8_pmuv3_attr_groups;
-	return armv8pmu_probe_num_events(cpu_pmu);
+	return armv8_pmu_init(cpu_pmu);
 }
 
 static int armv8_a57_pmu_init(struct arm_pmu *cpu_pmu)
 {
-	armv8_pmu_init(cpu_pmu);
 	cpu_pmu->name			= "armv8_cortex_a57";
 	cpu_pmu->map_event		= armv8_a57_map_event;
-	cpu_pmu->pmu.attr_groups	= armv8_pmuv3_attr_groups;
-	return armv8pmu_probe_num_events(cpu_pmu);
+	return armv8_pmu_init(cpu_pmu);
 }
 
 static int armv8_a72_pmu_init(struct arm_pmu *cpu_pmu)
 {
-	armv8_pmu_init(cpu_pmu);
 	cpu_pmu->name			= "armv8_cortex_a72";
 	cpu_pmu->map_event		= armv8_a57_map_event;
-	cpu_pmu->pmu.attr_groups	= armv8_pmuv3_attr_groups;
-	return armv8pmu_probe_num_events(cpu_pmu);
+	return armv8_pmu_init(cpu_pmu);
 }
 
 static int armv8_thunder_pmu_init(struct arm_pmu *cpu_pmu)
 {
-	armv8_pmu_init(cpu_pmu);
 	cpu_pmu->name			= "armv8_cavium_thunder";
 	cpu_pmu->map_event		= armv8_thunder_map_event;
-	cpu_pmu->pmu.attr_groups	= armv8_pmuv3_attr_groups;
-	return armv8pmu_probe_num_events(cpu_pmu);
+	return armv8_pmu_init(cpu_pmu);
 }
 
 static const struct of_device_id armv8_pmu_of_device_ids[] = {
-- 
2.1.0

^ permalink raw reply related	[flat|nested] 11+ messages in thread

* [PATCH v2 4/6] arm64/perf: Add Broadcom Vulcan PMU support
  2016-03-24 12:52 [PATCH v2 0/6] arm64: perf: Broadcom Vulcan PMU support Ashok Kumar
                   ` (2 preceding siblings ...)
  2016-03-24 12:52 ` [PATCH v2 3/6] arm64/perf: Filter common events based on PMCEIDn_EL0 Ashok Kumar
@ 2016-03-24 12:52 ` Ashok Kumar
  2016-03-24 12:52 ` [PATCH v2 5/6] arm64: dts: Add Broadcom Vulcan PMU in dts Ashok Kumar
  2016-03-24 12:52 ` [PATCH v2 6/6] Documentation: arm64: pmu: Add Broadcom Vulcan PMU binding Ashok Kumar
  5 siblings, 0 replies; 11+ messages in thread
From: Ashok Kumar @ 2016-03-24 12:52 UTC (permalink / raw)
  To: linux-arm-kernel

Broadcom Vulcan uses ARMv8 PMUv3 and supports most of
the ARMv8 recommended implementation defined events.

Added Vulcan events mapping for perf and perf_cache map.

Signed-off-by: Ashok Kumar <ashoks@broadcom.com>
---
 arch/arm64/kernel/perf_event.c | 59 ++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 59 insertions(+)

diff --git a/arch/arm64/kernel/perf_event.c b/arch/arm64/kernel/perf_event.c
index a025ec2..e8b985c 100644
--- a/arch/arm64/kernel/perf_event.c
+++ b/arch/arm64/kernel/perf_event.c
@@ -231,6 +231,20 @@ static const unsigned armv8_thunder_perf_map[PERF_COUNT_HW_MAX] = {
 	[PERF_COUNT_HW_STALLED_CYCLES_BACKEND]	= ARMV8_PMUV3_PERFCTR_STALL_BACKEND,
 };
 
+/* Broadcom Vulcan events mapping */
+static const unsigned armv8_vulcan_perf_map[PERF_COUNT_HW_MAX] = {
+	PERF_MAP_ALL_UNSUPPORTED,
+	[PERF_COUNT_HW_CPU_CYCLES]		= ARMV8_PMUV3_PERFCTR_CPU_CYCLES,
+	[PERF_COUNT_HW_INSTRUCTIONS]		= ARMV8_PMUV3_PERFCTR_INST_RETIRED,
+	[PERF_COUNT_HW_CACHE_REFERENCES]	= ARMV8_PMUV3_PERFCTR_L1D_CACHE,
+	[PERF_COUNT_HW_CACHE_MISSES]		= ARMV8_PMUV3_PERFCTR_L1D_CACHE_REFILL,
+	[PERF_COUNT_HW_BRANCH_INSTRUCTIONS]	= ARMV8_PMUV3_PERFCTR_BR_RETIRED,
+	[PERF_COUNT_HW_BRANCH_MISSES]		= ARMV8_PMUV3_PERFCTR_BR_MIS_PRED,
+	[PERF_COUNT_HW_BUS_CYCLES]		= ARMV8_PMUV3_PERFCTR_BUS_CYCLES,
+	[PERF_COUNT_HW_STALLED_CYCLES_FRONTEND]	= ARMV8_PMUV3_PERFCTR_STALL_FRONTEND,
+	[PERF_COUNT_HW_STALLED_CYCLES_BACKEND]	= ARMV8_PMUV3_PERFCTR_STALL_BACKEND,
+};
+
 static const unsigned armv8_pmuv3_perf_cache_map[PERF_COUNT_HW_CACHE_MAX]
 						[PERF_COUNT_HW_CACHE_OP_MAX]
 						[PERF_COUNT_HW_CACHE_RESULT_MAX] = {
@@ -323,6 +337,36 @@ static const unsigned armv8_thunder_perf_cache_map[PERF_COUNT_HW_CACHE_MAX]
 	[C(BPU)][C(OP_WRITE)][C(RESULT_MISS)]	= ARMV8_PMUV3_PERFCTR_BR_MIS_PRED,
 };
 
+static const unsigned armv8_vulcan_perf_cache_map[PERF_COUNT_HW_CACHE_MAX]
+					      [PERF_COUNT_HW_CACHE_OP_MAX]
+					      [PERF_COUNT_HW_CACHE_RESULT_MAX] = {
+	PERF_CACHE_MAP_ALL_UNSUPPORTED,
+
+	[C(L1D)][C(OP_READ)][C(RESULT_ACCESS)]	= ARMV8_IMPDEF_PERFCTR_L1D_CACHE_RD,
+	[C(L1D)][C(OP_READ)][C(RESULT_MISS)]	= ARMV8_IMPDEF_PERFCTR_L1D_CACHE_REFILL_RD,
+	[C(L1D)][C(OP_WRITE)][C(RESULT_ACCESS)]	= ARMV8_IMPDEF_PERFCTR_L1D_CACHE_WR,
+	[C(L1D)][C(OP_WRITE)][C(RESULT_MISS)]	= ARMV8_IMPDEF_PERFCTR_L1D_CACHE_REFILL_WR,
+
+	[C(L1I)][C(OP_READ)][C(RESULT_ACCESS)]	= ARMV8_PMUV3_PERFCTR_L1I_CACHE,
+	[C(L1I)][C(OP_READ)][C(RESULT_MISS)]	= ARMV8_PMUV3_PERFCTR_L1I_CACHE_REFILL,
+
+	[C(ITLB)][C(OP_READ)][C(RESULT_MISS)]	= ARMV8_PMUV3_PERFCTR_L1I_TLB_REFILL,
+	[C(ITLB)][C(OP_READ)][C(RESULT_ACCESS)]	= ARMV8_PMUV3_PERFCTR_L1I_TLB,
+
+	[C(DTLB)][C(OP_READ)][C(RESULT_ACCESS)]	= ARMV8_IMPDEF_PERFCTR_L1D_TLB_RD,
+	[C(DTLB)][C(OP_WRITE)][C(RESULT_ACCESS)] = ARMV8_IMPDEF_PERFCTR_L1D_TLB_WR,
+	[C(DTLB)][C(OP_READ)][C(RESULT_MISS)]	= ARMV8_IMPDEF_PERFCTR_L1D_TLB_REFILL_RD,
+	[C(DTLB)][C(OP_WRITE)][C(RESULT_MISS)]	= ARMV8_IMPDEF_PERFCTR_L1D_TLB_REFILL_WR,
+
+	[C(BPU)][C(OP_READ)][C(RESULT_ACCESS)]	= ARMV8_PMUV3_PERFCTR_BR_PRED,
+	[C(BPU)][C(OP_READ)][C(RESULT_MISS)]	= ARMV8_PMUV3_PERFCTR_BR_MIS_PRED,
+	[C(BPU)][C(OP_WRITE)][C(RESULT_ACCESS)]	= ARMV8_PMUV3_PERFCTR_BR_PRED,
+	[C(BPU)][C(OP_WRITE)][C(RESULT_MISS)]	= ARMV8_PMUV3_PERFCTR_BR_MIS_PRED,
+
+	[C(NODE)][C(OP_READ)][C(RESULT_ACCESS)]	= ARMV8_IMPDEF_PERFCTR_BUS_ACCESS_RD,
+	[C(NODE)][C(OP_WRITE)][C(RESULT_ACCESS)] = ARMV8_IMPDEF_PERFCTR_BUS_ACCESS_WR,
+};
+
 #define ARMV8_EVENT_ATTR_RESOLVE(m) #m
 #define ARMV8_EVENT_ATTR(name, config) \
 	PMU_EVENT_ATTR_STRING(name, armv8_event_attr_##name, \
@@ -940,6 +984,13 @@ static int armv8_thunder_map_event(struct perf_event *event)
 				ARMV8_EVTYPE_EVENT);
 }
 
+static int armv8_vulcan_map_event(struct perf_event *event)
+{
+	return armpmu_map_event(event, &armv8_vulcan_perf_map,
+				&armv8_vulcan_perf_cache_map,
+				ARMV8_EVTYPE_EVENT);
+}
+
 static unsigned int armv8pmu_read_num_pmnc_events(void)
 {
 	unsigned int nb_cnt;
@@ -1076,12 +1127,20 @@ static int armv8_thunder_pmu_init(struct arm_pmu *cpu_pmu)
 	return armv8_pmu_init(cpu_pmu);
 }
 
+static int armv8_vulcan_pmu_init(struct arm_pmu *cpu_pmu)
+{
+	cpu_pmu->name			= "armv8_brcm_vulcan";
+	cpu_pmu->map_event		= armv8_vulcan_map_event;
+	return armv8_pmu_init(cpu_pmu);
+}
+
 static const struct of_device_id armv8_pmu_of_device_ids[] = {
 	{.compatible = "arm,armv8-pmuv3",	.data = armv8_pmuv3_init},
 	{.compatible = "arm,cortex-a53-pmu",	.data = armv8_a53_pmu_init},
 	{.compatible = "arm,cortex-a57-pmu",	.data = armv8_a57_pmu_init},
 	{.compatible = "arm,cortex-a72-pmu",	.data = armv8_a72_pmu_init},
 	{.compatible = "cavium,thunder-pmu",	.data = armv8_thunder_pmu_init},
+	{.compatible = "brcm,vulcan-pmu",	.data = armv8_vulcan_pmu_init},
 	{},
 };
 
-- 
2.1.0

^ permalink raw reply related	[flat|nested] 11+ messages in thread

* [PATCH v2 5/6] arm64: dts: Add Broadcom Vulcan PMU in dts
  2016-03-24 12:52 [PATCH v2 0/6] arm64: perf: Broadcom Vulcan PMU support Ashok Kumar
                   ` (3 preceding siblings ...)
  2016-03-24 12:52 ` [PATCH v2 4/6] arm64/perf: Add Broadcom Vulcan PMU support Ashok Kumar
@ 2016-03-24 12:52 ` Ashok Kumar
  2016-03-24 12:52 ` [PATCH v2 6/6] Documentation: arm64: pmu: Add Broadcom Vulcan PMU binding Ashok Kumar
  5 siblings, 0 replies; 11+ messages in thread
From: Ashok Kumar @ 2016-03-24 12:52 UTC (permalink / raw)
  To: linux-arm-kernel

Add "brcm,vulcan-pmu" compatible string for Broadcom Vulcan PMU.

Signed-off-by: Ashok Kumar <ashoks@broadcom.com>
---
 arch/arm64/boot/dts/broadcom/vulcan.dtsi | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/arm64/boot/dts/broadcom/vulcan.dtsi b/arch/arm64/boot/dts/broadcom/vulcan.dtsi
index 85820e2..34e11a9 100644
--- a/arch/arm64/boot/dts/broadcom/vulcan.dtsi
+++ b/arch/arm64/boot/dts/broadcom/vulcan.dtsi
@@ -86,7 +86,7 @@
 	};
 
 	pmu {
-		compatible = "arm,armv8-pmuv3";
+		compatible = "brcm,vulcan-pmu", "arm,armv8-pmuv3";
 		interrupts = <GIC_PPI 7 IRQ_TYPE_LEVEL_HIGH>; /* PMU overflow */
 	};
 
-- 
2.1.0

^ permalink raw reply related	[flat|nested] 11+ messages in thread

* [PATCH v2 6/6] Documentation: arm64: pmu: Add Broadcom Vulcan PMU binding
  2016-03-24 12:52 [PATCH v2 0/6] arm64: perf: Broadcom Vulcan PMU support Ashok Kumar
                   ` (4 preceding siblings ...)
  2016-03-24 12:52 ` [PATCH v2 5/6] arm64: dts: Add Broadcom Vulcan PMU in dts Ashok Kumar
@ 2016-03-24 12:52 ` Ashok Kumar
  5 siblings, 0 replies; 11+ messages in thread
From: Ashok Kumar @ 2016-03-24 12:52 UTC (permalink / raw)
  To: linux-arm-kernel

Document the compatible string for Broadcom Vulcan PMU.
Also arranged the list in alphabetical order.

Signed-off-by: Ashok Kumar <ashoks@broadcom.com>
---
 Documentation/devicetree/bindings/arm/pmu.txt | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/Documentation/devicetree/bindings/arm/pmu.txt b/Documentation/devicetree/bindings/arm/pmu.txt
index d3999a1..b73a7c7 100644
--- a/Documentation/devicetree/bindings/arm/pmu.txt
+++ b/Documentation/devicetree/bindings/arm/pmu.txt
@@ -22,10 +22,11 @@ Required properties:
 	"arm,arm11mpcore-pmu"
 	"arm,arm1176-pmu"
 	"arm,arm1136-pmu"
+	"brcm,vulcan-pmu"
+	"cavium,thunder-pmu"
 	"qcom,scorpion-pmu"
 	"qcom,scorpion-mp-pmu"
 	"qcom,krait-pmu"
-	"cavium,thunder-pmu"
 - interrupts : 1 combined interrupt or 1 per core. If the interrupt is a per-cpu
                interrupt (PPI) then 1 interrupt should be specified.
 
-- 
2.1.0

^ permalink raw reply related	[flat|nested] 11+ messages in thread

* [PATCH v2 3/6] arm64/perf: Filter common events based on PMCEIDn_EL0
  2016-03-24 12:52 ` [PATCH v2 3/6] arm64/perf: Filter common events based on PMCEIDn_EL0 Ashok Kumar
@ 2016-03-24 15:28   ` Suzuki K. Poulose
  2016-03-28 12:31     ` Ashok Sekar
  2016-03-24 16:14   ` Mark Rutland
  1 sibling, 1 reply; 11+ messages in thread
From: Suzuki K. Poulose @ 2016-03-24 15:28 UTC (permalink / raw)
  To: linux-arm-kernel

On 24/03/16 12:52, Ashok Kumar wrote:
> The complete common architectural and micro-architectural
> event number structure is filtered based on PMCEIDn_EL0 and
> copied to a new structure which is exposed to /sys
>
> The function which derives event bitmap from PMCEIDn_EL0 is
> executed in the cpus, which has the pmu being initialized,
> for heterogeneous pmu support.
>
> Enforced armv8_pmuv3_event_attrs array for event number
> ordering as this is indexed with event number now.
>
> Signed-off-by: Ashok Kumar <ashoks@broadcom.com>
> ---
>   arch/arm64/kernel/perf_event.c | 281 ++++++++++++++++++++++++++++-------------
>   1 file changed, 195 insertions(+), 86 deletions(-)
>

> +static inline u32 armv8pmu_pmceidn_read(int reg)
> +{
> +	u32 val = 0;
> +
> +	if (reg == 0)
> +		asm volatile("mrs %0, pmceid0_el0" : "=r" (val));
> +	else if (reg == 1)
> +		asm volatile("mrs %0, pmceid1_el0" : "=r" (val));
> +
> +	return val;
> +}

> +	reg[0] = armv8pmu_pmceidn_read(0);
> +	reg[1] = armv8pmu_pmceidn_read(1);

Could we not add definitions for SYS_PMCEID0_EL0 and SYS_PMCEID1_EL0
in asm/sysreg.h and simply do

reg[0] = read_cpuid(SYS_PMCEID0_EL0);
reg[1] = read_cpuid(SYS_PMCEID1_EL0);

?

Thanks
Suzuki

^ permalink raw reply	[flat|nested] 11+ messages in thread

* [PATCH v2 3/6] arm64/perf: Filter common events based on PMCEIDn_EL0
  2016-03-24 12:52 ` [PATCH v2 3/6] arm64/perf: Filter common events based on PMCEIDn_EL0 Ashok Kumar
  2016-03-24 15:28   ` Suzuki K. Poulose
@ 2016-03-24 16:14   ` Mark Rutland
  2016-03-28 12:25     ` Ashok Sekar
  1 sibling, 1 reply; 11+ messages in thread
From: Mark Rutland @ 2016-03-24 16:14 UTC (permalink / raw)
  To: linux-arm-kernel

HI,

On Thu, Mar 24, 2016 at 05:52:37AM -0700, Ashok Kumar wrote:
> The complete common architectural and micro-architectural
> event number structure is filtered based on PMCEIDn_EL0 and
> copied to a new structure which is exposed to /sys
>
> The function which derives event bitmap from PMCEIDn_EL0 is
> executed in the cpus, which has the pmu being initialized,
> for heterogeneous pmu support.

I would prefer it we could instead filter the list at run time by
implementing attribute_group::is_visible() for the events attribute
group, and share a common set of attributes and attr groups. That would
avoid the filtering, copying, and assocaited memory allocation.

e.g. have an array of:

struct event_attribute {
	struct attribute attr;
	int pmceid_idx;
};

Then cache the pmceid value at probe time, and have and something like:

umode_t event_attr_is_visible(struct kobject *kobj, struct attribute *attr, int)
{
	struct arm_pmu *arm_pmu;
	struct event_attribute *e_attr;

	arm_pmu = pmu_kobj_to_armpmu(kobj);
	e_attr = container_of(attr, struct event_attr, attr);
	
	if (test_bit(e_attr->pmceid_idx, &arm_pmu->cached_pmceid))
		return 0444;

	return 0;
}

Thanks,
Mark.

^ permalink raw reply	[flat|nested] 11+ messages in thread

* [PATCH v2 3/6] arm64/perf: Filter common events based on PMCEIDn_EL0
  2016-03-24 16:14   ` Mark Rutland
@ 2016-03-28 12:25     ` Ashok Sekar
  0 siblings, 0 replies; 11+ messages in thread
From: Ashok Sekar @ 2016-03-28 12:25 UTC (permalink / raw)
  To: linux-arm-kernel

Hi Mark,

On Thu, Mar 24, 2016 at 9:44 PM, Mark Rutland <mark.rutland@arm.com> wrote:
> HI,
>
> On Thu, Mar 24, 2016 at 05:52:37AM -0700, Ashok Kumar wrote:
>> The complete common architectural and micro-architectural
>> event number structure is filtered based on PMCEIDn_EL0 and
>> copied to a new structure which is exposed to /sys
>>
>> The function which derives event bitmap from PMCEIDn_EL0 is
>> executed in the cpus, which has the pmu being initialized,
>> for heterogeneous pmu support.
>
> I would prefer it we could instead filter the list at run time by
> implementing attribute_group::is_visible() for the events attribute
> group, and share a common set of attributes and attr groups. That would
> avoid the filtering, copying, and assocaited memory allocation.
>
> e.g. have an array of:
>
> struct event_attribute {
>         struct attribute attr;
>         int pmceid_idx;
> };
>
> Then cache the pmceid value at probe time, and have and something like:
>
> umode_t event_attr_is_visible(struct kobject *kobj, struct attribute *attr, int)
> {
>         struct arm_pmu *arm_pmu;
>         struct event_attribute *e_attr;
>
>         arm_pmu = pmu_kobj_to_armpmu(kobj);
>         e_attr = container_of(attr, struct event_attr, attr);
>
>         if (test_bit(e_attr->pmceid_idx, &arm_pmu->cached_pmceid))
>                 return 0444;
>
>         return 0;
> }
This looks better - posted a v3 based on your suggestion.

Thanks,
Ashok
>
> Thanks,
> Mark.

^ permalink raw reply	[flat|nested] 11+ messages in thread

* [PATCH v2 3/6] arm64/perf: Filter common events based on PMCEIDn_EL0
  2016-03-24 15:28   ` Suzuki K. Poulose
@ 2016-03-28 12:31     ` Ashok Sekar
  0 siblings, 0 replies; 11+ messages in thread
From: Ashok Sekar @ 2016-03-28 12:31 UTC (permalink / raw)
  To: linux-arm-kernel

Hi Suzuki,

On Thu, Mar 24, 2016 at 8:58 PM, Suzuki K. Poulose
<Suzuki.Poulose@arm.com> wrote:
> On 24/03/16 12:52, Ashok Kumar wrote:
>>
>> The complete common architectural and micro-architectural
>> event number structure is filtered based on PMCEIDn_EL0 and
>> copied to a new structure which is exposed to /sys
>>
>> The function which derives event bitmap from PMCEIDn_EL0 is
>> executed in the cpus, which has the pmu being initialized,
>> for heterogeneous pmu support.
>>
>> Enforced armv8_pmuv3_event_attrs array for event number
>> ordering as this is indexed with event number now.
>>
>> Signed-off-by: Ashok Kumar <ashoks@broadcom.com>
>> ---
>>   arch/arm64/kernel/perf_event.c | 281
>> ++++++++++++++++++++++++++++-------------
>>   1 file changed, 195 insertions(+), 86 deletions(-)
>>
>
>> +static inline u32 armv8pmu_pmceidn_read(int reg)
>> +{
>> +       u32 val = 0;
>> +
>> +       if (reg == 0)
>> +               asm volatile("mrs %0, pmceid0_el0" : "=r" (val));
>> +       else if (reg == 1)
>> +               asm volatile("mrs %0, pmceid1_el0" : "=r" (val));
>> +
>> +       return val;
>> +}
>
>
>> +       reg[0] = armv8pmu_pmceidn_read(0);
>> +       reg[1] = armv8pmu_pmceidn_read(1);
>
>
> Could we not add definitions for SYS_PMCEID0_EL0 and SYS_PMCEID1_EL0
> in asm/sysreg.h and simply do
>
> reg[0] = read_cpuid(SYS_PMCEID0_EL0);
> reg[1] = read_cpuid(SYS_PMCEID1_EL0);
>
> ?
Yes, it is possible but looks like read_cpuid is used only in cpuid
related contexts.
Maybe we need to define separate definition for pmu/general context and use it?
perf_event.c has other instances where the pmu registers are directly read and
needs to be converted to this format if we are doing it. Let me know
your suggestion.
Currently I have posted v3 adhering to the convention of already
existing pmu register reads.

Thanks,
Ashok
>
> Thanks
> Suzuki

^ permalink raw reply	[flat|nested] 11+ messages in thread

end of thread, other threads:[~2016-03-28 12:31 UTC | newest]

Thread overview: 11+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2016-03-24 12:52 [PATCH v2 0/6] arm64: perf: Broadcom Vulcan PMU support Ashok Kumar
2016-03-24 12:52 ` [PATCH v2 1/6] arm64/perf: Changed events naming as per ARM ARM Ashok Kumar
2016-03-24 12:52 ` [PATCH v2 2/6] arm64/perf: Define complete ARMv8 recommended implementation defined events Ashok Kumar
2016-03-24 12:52 ` [PATCH v2 3/6] arm64/perf: Filter common events based on PMCEIDn_EL0 Ashok Kumar
2016-03-24 15:28   ` Suzuki K. Poulose
2016-03-28 12:31     ` Ashok Sekar
2016-03-24 16:14   ` Mark Rutland
2016-03-28 12:25     ` Ashok Sekar
2016-03-24 12:52 ` [PATCH v2 4/6] arm64/perf: Add Broadcom Vulcan PMU support Ashok Kumar
2016-03-24 12:52 ` [PATCH v2 5/6] arm64: dts: Add Broadcom Vulcan PMU in dts Ashok Kumar
2016-03-24 12:52 ` [PATCH v2 6/6] Documentation: arm64: pmu: Add Broadcom Vulcan PMU binding Ashok Kumar

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.