perf intel-pt: decoder: Add CFE (Control Flow Event) and EVD (Event Data) processing

As of Intel SDM (https://www.intel.com/sdm) version 076, there is a new
Intel PT feature called Event Trace which requires 2 new packets CFE
(Control Flow Event) and EVD (Event Data).

Each Event Trace event is represented by a CFE packet that is preceded
by zero or more EVD packets. It may be bound to a following FUP (Flow
Update) packet that provides the IP.

Event Trace exposes details about asynchronous events. The CFE packet
contains a type field to identify one of the following:

	 1	INTR		interrupt, fault, exception, NMI
	 2	IRET		interrupt return
	 3	SMI		system management interrupt
	 4	RSM		resume from system management mode
	 5	SIPI		startup interprocessor interrupt
	 6	INIT		INIT signal
	 7	VMENTRY		VM-Entry
	 8	VMEXIT		VM-Entry
	 9	VMEXIT_INTR	VM-Exit due to interrupt
	10	SHUTDOWN	Shutdown

For more details, refer to the Intel SDM, Intel Processor Trace chapter.

Add processing to the decoder for the new packets.
Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Andi Kleen <ak@linux.intel.com>
Cc: Jiri Olsa <jolsa@redhat.com>
Link: https://lore.kernel.org/r/20220124084201.2699795-8-adrian.hunter@intel.comSigned-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>

tools/perf/util/intel-pt-decoder/intel-pt-decoder.c

@@ -213,6 +213,8 @@ struct intel_pt_decoder {
 	bool set_fup_pwre;
 	bool set_fup_exstop;
 	bool set_fup_bep;
+	bool set_fup_cfe_ip;
+	bool set_fup_cfe;
 	bool sample_cyc;
 	unsigned int fup_tx_flags;
 	unsigned int tx_flags;
@@ -223,6 +225,7 @@ struct intel_pt_decoder {
 	uint64_t timestamp_insn_cnt;
 	uint64_t sample_insn_cnt;
 	uint64_t stuck_ip;
+	struct intel_pt_pkt fup_cfe_pkt;
 	int max_loops;
 	int no_progress;
 	int stuck_ip_prd;
@@ -231,6 +234,8 @@ struct intel_pt_decoder {
 	const unsigned char *next_buf;
 	size_t next_len;
 	unsigned char temp_buf[INTEL_PT_PKT_MAX_SZ];
+	int evd_cnt;
+	struct intel_pt_evd evd[INTEL_PT_MAX_EVDS];
 };
 
 static uint64_t intel_pt_lower_power_of_2(uint64_t x)
@@ -1214,6 +1219,9 @@ static void intel_pt_clear_fup_event(struct intel_pt_decoder *decoder)
 	decoder->set_fup_pwre = false;
 	decoder->set_fup_exstop = false;
 	decoder->set_fup_bep = false;
+	decoder->set_fup_cfe_ip = false;
+	decoder->set_fup_cfe = false;
+	decoder->evd_cnt = 0;
 }
 
 static bool intel_pt_fup_event(struct intel_pt_decoder *decoder)
@@ -1223,6 +1231,23 @@ static bool intel_pt_fup_event(struct intel_pt_decoder *decoder)
 
 	decoder->state.type &= ~INTEL_PT_BRANCH;
 
+	if (decoder->set_fup_cfe_ip || decoder->set_fup_cfe) {
+		bool ip = decoder->set_fup_cfe_ip;
+
+		decoder->set_fup_cfe_ip = false;
+		decoder->set_fup_cfe = false;
+		decoder->state.type |= INTEL_PT_EVT;
+		if (!ip && decoder->pge)
+			decoder->state.type |= INTEL_PT_BRANCH;
+		decoder->state.cfe_type = decoder->fup_cfe_pkt.count;
+		decoder->state.cfe_vector = decoder->fup_cfe_pkt.payload;
+		decoder->state.evd_cnt = decoder->evd_cnt;
+		decoder->state.evd = decoder->evd;
+		decoder->evd_cnt = 0;
+		if (ip || decoder->pge)
+			decoder->state.flags |= INTEL_PT_FUP_IP;
+		ret = true;
+	}
 	if (decoder->set_fup_tx_flags) {
 		decoder->set_fup_tx_flags = false;
 		decoder->tx_flags = decoder->fup_tx_flags;
@@ -1540,6 +1565,19 @@ static int intel_pt_mode_tsx(struct intel_pt_decoder *decoder, bool *no_tip)
 	return 0;
 }
 
+static int intel_pt_evd(struct intel_pt_decoder *decoder)
+{
+	if (decoder->evd_cnt >= INTEL_PT_MAX_EVDS) {
+		intel_pt_log_at("ERROR: Too many EVD packets", decoder->pos);
+		return -ENOSYS;
+	}
+	decoder->evd[decoder->evd_cnt++] = (struct intel_pt_evd){
+		.type = decoder->packet.count,
+		.payload = decoder->packet.payload,
+	};
+	return 0;
+}
+
 static uint64_t intel_pt_8b_tsc(uint64_t timestamp, uint64_t ref_timestamp)
 {
 	timestamp |= (ref_timestamp & (0xffULL << 56));
@@ -3250,8 +3288,32 @@ static int intel_pt_walk_trace(struct intel_pt_decoder *decoder)
 			goto next;
 
 		case INTEL_PT_CFE:
+			decoder->fup_cfe_pkt = decoder->packet;
+			decoder->set_fup_cfe = true;
+			if (!decoder->pge) {
+				intel_pt_fup_event(decoder);
+				return 0;
+			}
+			break;
+
 		case INTEL_PT_CFE_IP:
+			decoder->fup_cfe_pkt = decoder->packet;
+			err = intel_pt_get_next_packet(decoder);
+			if (err)
+				return err;
+			if (decoder->packet.type == INTEL_PT_FUP) {
+				decoder->set_fup_cfe_ip = true;
+				no_tip = true;
+			} else {
+				intel_pt_log_at("ERROR: Missing FUP after CFE",
+						decoder->pos);
+			}
+			goto next;
+
 		case INTEL_PT_EVD:
+			err = intel_pt_evd(decoder);
+			if (err)
+				return err;
 			break;
 
 		default:

tools/perf/util/intel-pt-decoder/intel-pt-decoder.h

@@ -35,6 +35,7 @@ enum intel_pt_sample_type {
 	INTEL_PT_TRACE_END	= 1 << 10,
 	INTEL_PT_BLK_ITEMS	= 1 << 11,
 	INTEL_PT_PSB_EVT	= 1 << 12,
+	INTEL_PT_EVT		= 1 << 13,
 };
 
 enum intel_pt_period_type {
@@ -209,6 +210,18 @@ struct intel_pt_vmcs_info {
 	bool error_printed;
 };
 
+/*
+ * Maximum number of event trace data in one go, assuming at most 1 per type
+ * and 6-bits of type in the EVD packet.
+ */
+#define INTEL_PT_MAX_EVDS 64
+
+/* Event trace data from EVD packet */
+struct intel_pt_evd {
+	int type;
+	uint64_t payload;
+};
+
 struct intel_pt_state {
 	enum intel_pt_sample_type type;
 	bool from_nr;
@@ -234,6 +247,10 @@ struct intel_pt_state {
 	int insn_len;
 	char insn[INTEL_PT_INSN_BUF_SZ];
 	struct intel_pt_blk_items items;
+	int cfe_type;
+	int cfe_vector;
+	int evd_cnt;
+	struct intel_pt_evd *evd;
 };
 
 struct intel_pt_insn;