Merge tag 'perf-tools-for-v5.18-2022-03-26' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux

Pull perf tools updates from Arnaldo Carvalho de Melo:
 "New features:

  perf ftrace:

   - Add -n/--use-nsec option to the 'latency' subcommand.

     Default: usecs:

     $ sudo perf ftrace latency -T dput -a sleep 1
     #   DURATION     |      COUNT | GRAPH                          |
          0 - 1    us |    2098375 | #############################  |
          1 - 2    us |         61 |                                |
          2 - 4    us |         33 |                                |
          4 - 8    us |         13 |                                |
          8 - 16   us |        124 |                                |
         16 - 32   us |        123 |                                |
         32 - 64   us |          1 |                                |
         64 - 128  us |          0 |                                |
        128 - 256  us |          1 |                                |
        256 - 512  us |          0 |                                |

     Better granularity with nsec:

     $ sudo perf ftrace latency -T dput -a -n sleep 1
     #   DURATION     |      COUNT | GRAPH                          |
          0 - 1    us |          0 |                                |
          1 - 2    ns |          0 |                                |
          2 - 4    ns |          0 |                                |
          4 - 8    ns |          0 |                                |
          8 - 16   ns |          0 |                                |
         16 - 32   ns |          0 |                                |
         32 - 64   ns |          0 |                                |
         64 - 128  ns |    1163434 | ##############                 |
        128 - 256  ns |     914102 | #############                  |
        256 - 512  ns |        884 |                                |
        512 - 1024 ns |        613 |                                |
          1 - 2    us |         31 |                                |
          2 - 4    us |         17 |                                |
          4 - 8    us |          7 |                                |
          8 - 16   us |        123 |                                |
         16 - 32   us |         83 |                                |

  perf lock:

   - Add -c/--combine-locks option to merge lock instances in the same
     class into a single entry.

     # perf lock report -c
                    Name acquired contended avg wait(ns) total wait(ns) max wait(ns) min wait(ns)

           rcu_read_lock   251225         0            0              0            0            0
      hrtimer_bases.lock    39450         0            0              0            0            0
     &sb->s_type->i_l...    10301         1          662            662          662          662
        ptlock_ptr(page)    10173         2          701           1402          760          642
     &(ei->i_block_re...     8732         0            0              0            0            0
            &xa->xa_lock     8088         0            0              0            0            0
             &base->lock     6705         0            0              0            0            0
             &p->pi_lock     5549         0            0              0            0            0
     &dentry->d_lockr...     5010         4         1274           5097         1844          789
               &ep->lock     3958         0            0              0            0            0

      - Add -F/--field option to customize the list of fields to output:

     $ perf lock report -F contended,wait_max -k avg_wait
                     Name contended max wait(ns) avg wait(ns)

           slock-AF_INET6         1        23543        23543
        &lruvec->lru_lock         5        18317        11254
           slock-AF_INET6         1        10379        10379
               rcu_node_1         1         2104         2104
      &dentry->d_lockr...         1         1844         1844
      &dentry->d_lockr...         1         1672         1672
         &newf->file_lock        15         2279         1025
      &dentry->d_lockr...         1          792          792

   - Add --synth=no option for record, as there is no need to symbolize,
     lock names comes from the tracepoints.

  perf record:

   - Threaded recording, opt-in, via the new --threads command line
     option.

   - Improve AMD IBS (Instruction-Based Sampling) error handling
     messages.

  perf script:

   - Add 'brstackinsnlen' field (use it with -F) for branch stacks.

   - Output branch sample type in 'perf script'.

  perf report:

   - Add "addr_from" and "addr_to" sort dimensions.

   - Print branch stack entry type in 'perf report --dump-raw-trace'

   - Fix symbolization for chrooted workloads.

  Hardware tracing:

  Intel PT:

   - Add CFE (Control Flow Event) and EVD (Event Data) packets support.

   - Add MODE.Exec IFLAG bit support.

     Explanation about these features from the "Intel® 64 and IA-32
     architectures software developer’s manual combined volumes: 1, 2A,
     2B, 2C, 2D, 3A, 3B, 3C, 3D, and 4" PDF at:

        https://cdrdv2.intel.com/v1/dl/getContent/671200

     At page 3951:
      "32.2.4

       Event Trace is a capability that exposes details about the
       asynchronous events, when they are generated, and when their
       corresponding software event handler completes execution. These
       include:

        o Interrupts, including NMI and SMI, including the interrupt
          vector when defined.

        o Faults, exceptions including the fault vector.

           - Page faults additionally include the page fault address,
             when in context.

        o Event handler returns, including IRET and RSM.

        o VM exits and VM entries.¹

           - VM exits include the values written to the “exit reason”
             and “exit qualification” VMCS fields. INIT and SIPI events.

        o TSX aborts, including the abort status returned for the RTM
          instructions.

        o Shutdown.

       Additionally, it provides indication of the status of the
       Interrupt Flag (IF), to indicate when interrupts are masked"

  ARM CoreSight:

   - Use advertised caps/min_interval as default sample_period on ARM
     spe.

   - Update deduction of TRCCONFIGR register for branch broadcast on
     ARM's CoreSight ETM.

  Vendor Events (JSON):

  Intel:

   - Update events and metrics for: Alderlake, Broadwell, Broadwell DE,
     BroadwellX, CascadelakeX, Elkhartlake, Bonnell, Goldmont,
     GoldmontPlus, Westmere EP-DP, Haswell, HaswellX, Icelake, IcelakeX,
     Ivybridge, Ivytown, Jaketown, Knights Landing, Nehalem EP,
     Sandybridge, Silvermont, Skylake, Skylake Server, SkylakeX,
     Tigerlake, TremontX, Westmere EP-SP, and Westmere EX.

  ARM:

   - Add support for HiSilicon CPA PMU aliasing.

  perf stat:

   - Fix forked applications enablement of counters.

   - The 'slots' should only be printed on a different order than the
     one specified on the command line when 'topdown' events are
     present, fix it.

  Miscellaneous:

   - Sync msr-index, cpufeatures header files with the kernel sources.

   - Stop using some deprecated libbpf APIs in 'perf trace'.

   - Fix some spelling mistakes.

   - Refactor the maps pointers usage to pave the way for using refcount
     debugging.

   - Only offer the --tui option on perf top, report and annotate when
     perf was built with libslang.

   - Don't mention --to-ctf in 'perf data --help' when not linking with
     the required library, libbabeltrace.

   - Use ARRAY_SIZE() instead of ad hoc equivalent, spotted by
     array_size.cocci.

   - Enhance the matching of sub-commands abbreviations:
	'perf c2c rec' -> 'perf c2c record'
	'perf c2c recport -> error

   - Set build-id using build-id header on new mmap records.

   - Fix generation of 'perf --version' string.

  perf test:

   - Add test for the arm_spe event.

   - Add test to check unwinding using fame-pointer (fp) mode on arm64.

   - Make metric testing more robust in 'perf test'.

   - Add error message for unsupported branch stack cases.

  libperf:

   - Add API for allocating new thread map array.

   - Fix typo in perf_evlist__open() failure error messages in libperf
     tests.

  perf c2c:

   - Replace bitmap_weight() with bitmap_empty() where appropriate"

* tag 'perf-tools-for-v5.18-2022-03-26' of git://git.kernel.org/pub/scm/linux/kernel/git/acme/linux: (143 commits)
  perf evsel: Improve AMD IBS (Instruction-Based Sampling) error handling messages
  perf python: Add perf_env stubs that will be needed in evsel__open_strerror()
  perf tools: Enhance the matching of sub-commands abbreviations
  libperf tests: Fix typo in perf_evlist__open() failure error messages
  tools arm64: Import cputype.h
  perf lock: Add -F/--field option to control output
  perf lock: Extend struct lock_key to have print function
  perf lock: Add --synth=no option for record
  tools headers cpufeatures: Sync with the kernel sources
  tools headers cpufeatures: Sync with the kernel sources
  perf stat: Fix forked applications enablement of counters
  tools arch x86: Sync the msr-index.h copy with the kernel sources
  perf evsel: Make evsel__env() always return a valid env
  perf build-id: Fix spelling mistake "Cant" -> "Can't"
  perf header: Fix spelling mistake "could't" -> "couldn't"
  perf script: Add 'brstackinsnlen' for branch stacks
  perf parse-events: Move slots only with topdown
  perf ftrace latency: Update documentation
  perf ftrace latency: Add -n/--use-nsec option
  perf tools: Fix version kernel tag
  ...

4 files changed, 290 insertions, 26 deletions

diff --git a/tools/perf/util/intel-pt-decoder/intel-pt-decoder.c b/tools/perf/util/intel-pt-decoder/intel-pt-decoder.c
index 0e013c2d9eb4..e1d8f7504cbe 100644
--- a/tools/perf/util/intel-pt-decoder/intel-pt-decoder.c
+++ b/tools/perf/util/intel-pt-decoder/intel-pt-decoder.c
@@ -144,6 +144,8 @@ struct intel_pt_decoder {
 	bool vm_tm_corr_continuous;
 	bool nr;
 	bool next_nr;
+	bool iflag;
+	bool next_iflag;
 	enum intel_pt_param_flags flags;
 	uint64_t pos;
 	uint64_t last_ip;
@@ -213,6 +215,9 @@ 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 set_fup_mode_exec;
 	bool sample_cyc;
 	unsigned int fup_tx_flags;
 	unsigned int tx_flags;
@@ -223,6 +228,7 @@ 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 +237,8 @@ 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)
@@ -820,6 +828,9 @@ static int intel_pt_calc_cyc_cb(struct intel_pt_pkt_info *pkt_info)
 	case INTEL_PT_BIP:
 	case INTEL_PT_BEP:
 	case INTEL_PT_BEP_IP:
+	case INTEL_PT_CFE:
+	case INTEL_PT_CFE_IP:
+	case INTEL_PT_EVD:
 		return 0;
 
 	case INTEL_PT_MTC:
@@ -1094,6 +1105,52 @@ static void intel_pt_sample_insn(struct intel_pt_decoder *decoder)
 	decoder->state.type |= INTEL_PT_INSTRUCTION;
 }
 
+/*
+ * Sample FUP instruction at the same time as reporting the FUP event, so the
+ * instruction sample gets the same flags as the FUP event.
+ */
+static void intel_pt_sample_fup_insn(struct intel_pt_decoder *decoder)
+{
+	struct intel_pt_insn intel_pt_insn;
+	uint64_t max_insn_cnt, insn_cnt = 0;
+	int err;
+
+	decoder->state.insn_op = INTEL_PT_OP_OTHER;
+	decoder->state.insn_len = 0;
+
+	if (!decoder->branch_enable || !decoder->pge || decoder->hop ||
+	    decoder->ip != decoder->last_ip)
+		return;
+
+	if (!decoder->mtc_insn)
+		decoder->mtc_insn = true;
+
+	max_insn_cnt = intel_pt_next_sample(decoder);
+	if (max_insn_cnt != 1)
+		return;
+
+	err = decoder->walk_insn(&intel_pt_insn, &insn_cnt, &decoder->ip,
+				 0, max_insn_cnt, decoder->data);
+	/* Ignore error, it will be reported next walk anyway */
+	if (err)
+		return;
+
+	if (intel_pt_insn.branch != INTEL_PT_BR_NO_BRANCH) {
+		intel_pt_log_at("ERROR: Unexpected branch at FUP instruction", decoder->ip);
+		return;
+	}
+
+	decoder->tot_insn_cnt += insn_cnt;
+	decoder->timestamp_insn_cnt += insn_cnt;
+	decoder->sample_insn_cnt += insn_cnt;
+	decoder->period_insn_cnt += insn_cnt;
+
+	intel_pt_sample_insn(decoder);
+
+	decoder->state.type |= INTEL_PT_INSTRUCTION;
+	decoder->ip += intel_pt_insn.length;
+}
+
 static int intel_pt_walk_insn(struct intel_pt_decoder *decoder,
 			      struct intel_pt_insn *intel_pt_insn, uint64_t ip)
 {
@@ -1203,13 +1260,85 @@ out_no_progress:
 	return err;
 }
 
-static bool intel_pt_fup_event(struct intel_pt_decoder *decoder)
+static void intel_pt_mode_exec_status(struct intel_pt_decoder *decoder)
+{
+	bool iflag = decoder->packet.count & INTEL_PT_IFLAG;
+
+	decoder->exec_mode = decoder->packet.payload;
+	decoder->iflag = iflag;
+	decoder->next_iflag = iflag;
+	decoder->state.from_iflag = iflag;
+	decoder->state.to_iflag = iflag;
+}
+
+static void intel_pt_mode_exec(struct intel_pt_decoder *decoder)
+{
+	bool iflag = decoder->packet.count & INTEL_PT_IFLAG;
+
+	decoder->exec_mode = decoder->packet.payload;
+	decoder->next_iflag = iflag;
+}
+
+static void intel_pt_sample_iflag(struct intel_pt_decoder *decoder)
+{
+	decoder->state.type |= INTEL_PT_IFLAG_CHG;
+	decoder->state.from_iflag = decoder->iflag;
+	decoder->state.to_iflag = decoder->next_iflag;
+	decoder->iflag = decoder->next_iflag;
+}
+
+static void intel_pt_sample_iflag_chg(struct intel_pt_decoder *decoder)
+{
+	if (decoder->iflag != decoder->next_iflag)
+		intel_pt_sample_iflag(decoder);
+}
+
+static void intel_pt_clear_fup_event(struct intel_pt_decoder *decoder)
+{
+	decoder->set_fup_tx_flags = false;
+	decoder->set_fup_ptw = false;
+	decoder->set_fup_mwait = false;
+	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;
+	decoder->set_fup_mode_exec = false;
+	decoder->iflag = decoder->next_iflag;
+}
+
+static bool intel_pt_fup_event(struct intel_pt_decoder *decoder, bool no_tip)
 {
 	enum intel_pt_sample_type type = decoder->state.type;
+	bool sample_fup_insn = false;
 	bool ret = false;
 
 	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_mode_exec) {
+		decoder->set_fup_mode_exec = false;
+		intel_pt_sample_iflag(decoder);
+		sample_fup_insn = no_tip;
+		ret = true;
+	}
 	if (decoder->set_fup_tx_flags) {
 		decoder->set_fup_tx_flags = false;
 		decoder->tx_flags = decoder->fup_tx_flags;
@@ -1266,6 +1395,8 @@ static bool intel_pt_fup_event(struct intel_pt_decoder *decoder)
 	if (ret) {
 		decoder->state.from_ip = decoder->ip;
 		decoder->state.to_ip = 0;
+		if (sample_fup_insn)
+			intel_pt_sample_fup_insn(decoder);
 	} else {
 		decoder->state.type = type;
 	}
@@ -1298,7 +1429,7 @@ static int intel_pt_walk_fup(struct intel_pt_decoder *decoder)
 			bool no_tip = decoder->pkt_state != INTEL_PT_STATE_FUP;
 
 			decoder->pkt_state = INTEL_PT_STATE_IN_SYNC;
-			if (intel_pt_fup_event(decoder) && no_tip)
+			if (intel_pt_fup_event(decoder, no_tip) && no_tip)
 				return 0;
 			return -EAGAIN;
 		}
@@ -1350,6 +1481,7 @@ static int intel_pt_walk_tip(struct intel_pt_decoder *decoder)
 		return err;
 
 	intel_pt_update_nr(decoder);
+	intel_pt_sample_iflag_chg(decoder);
 
 	if (intel_pt_insn.branch == INTEL_PT_BR_INDIRECT) {
 		if (decoder->pkt_state == INTEL_PT_STATE_TIP_PGD) {
@@ -1464,6 +1596,7 @@ static int intel_pt_walk_tnt(struct intel_pt_decoder *decoder)
 			decoder->state.to_ip = decoder->last_ip;
 			decoder->ip = decoder->last_ip;
 			intel_pt_update_nr(decoder);
+			intel_pt_sample_iflag_chg(decoder);
 			return 0;
 		}
 
@@ -1527,6 +1660,19 @@ 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));
@@ -1620,12 +1766,7 @@ static int intel_pt_overflow(struct intel_pt_decoder *decoder)
 	decoder->state.from_ip = decoder->ip;
 	decoder->ip = 0;
 	decoder->pge = false;
-	decoder->set_fup_tx_flags = false;
-	decoder->set_fup_ptw = false;
-	decoder->set_fup_mwait = false;
-	decoder->set_fup_pwre = false;
-	decoder->set_fup_exstop = false;
-	decoder->set_fup_bep = false;
+	intel_pt_clear_fup_event(decoder);
 	decoder->overflow = true;
 	return -EOVERFLOW;
 }
@@ -1873,6 +2014,9 @@ static int intel_pt_walk_psbend(struct intel_pt_decoder *decoder)
 		case INTEL_PT_BIP:
 		case INTEL_PT_BEP:
 		case INTEL_PT_BEP_IP:
+		case INTEL_PT_CFE:
+		case INTEL_PT_CFE_IP:
+		case INTEL_PT_EVD:
 			decoder->have_tma = false;
 			intel_pt_log("ERROR: Unexpected packet\n");
 			err = -EAGAIN;
@@ -1895,7 +2039,7 @@ static int intel_pt_walk_psbend(struct intel_pt_decoder *decoder)
 			break;
 
 		case INTEL_PT_MODE_EXEC:
-			decoder->exec_mode = decoder->packet.payload;
+			intel_pt_mode_exec_status(decoder);
 			break;
 
 		case INTEL_PT_PIP:
@@ -1975,6 +2119,9 @@ static int intel_pt_walk_fup_tip(struct intel_pt_decoder *decoder)
 		case INTEL_PT_BIP:
 		case INTEL_PT_BEP:
 		case INTEL_PT_BEP_IP:
+		case INTEL_PT_CFE:
+		case INTEL_PT_CFE_IP:
+		case INTEL_PT_EVD:
 			intel_pt_log("ERROR: Missing TIP after FUP\n");
 			decoder->pkt_state = INTEL_PT_STATE_ERR3;
 			decoder->pkt_step = 0;
@@ -2026,6 +2173,7 @@ static int intel_pt_walk_fup_tip(struct intel_pt_decoder *decoder)
 				decoder->state.to_ip = decoder->ip;
 			}
 			intel_pt_update_nr(decoder);
+			intel_pt_sample_iflag_chg(decoder);
 			return 0;
 
 		case INTEL_PT_PIP:
@@ -2043,7 +2191,7 @@ static int intel_pt_walk_fup_tip(struct intel_pt_decoder *decoder)
 			break;
 
 		case INTEL_PT_MODE_EXEC:
-			decoder->exec_mode = decoder->packet.payload;
+			intel_pt_mode_exec(decoder);
 			break;
 
 		case INTEL_PT_VMCS:
@@ -2134,6 +2282,9 @@ static int intel_pt_vm_psb_lookahead_cb(struct intel_pt_pkt_info *pkt_info)
 	case INTEL_PT_TIP:
 	case INTEL_PT_PSB:
 	case INTEL_PT_TRACESTOP:
+	case INTEL_PT_CFE:
+	case INTEL_PT_CFE_IP:
+	case INTEL_PT_EVD:
 	default:
 		return 1;
 	}
@@ -2653,6 +2804,9 @@ static int intel_pt_vm_time_correlation(struct intel_pt_decoder *decoder)
 			decoder->blk_type = 0;
 			break;
 
+		case INTEL_PT_CFE:
+		case INTEL_PT_CFE_IP:
+		case INTEL_PT_EVD:
 		case INTEL_PT_MODE_EXEC:
 		case INTEL_PT_MODE_TSX:
 		case INTEL_PT_MNT:
@@ -2719,6 +2873,7 @@ static int intel_pt_hop_trace(struct intel_pt_decoder *decoder, bool *no_tip, in
 		decoder->state.from_ip = decoder->ip;
 		decoder->state.to_ip = 0;
 		intel_pt_update_nr(decoder);
+		intel_pt_sample_iflag_chg(decoder);
 		return HOP_RETURN;
 
 	case INTEL_PT_FUP:
@@ -2733,10 +2888,10 @@ static int intel_pt_hop_trace(struct intel_pt_decoder *decoder, bool *no_tip, in
 			decoder->state.type = INTEL_PT_INSTRUCTION;
 			decoder->state.from_ip = decoder->ip;
 			decoder->state.to_ip = 0;
-			intel_pt_fup_event(decoder);
+			intel_pt_fup_event(decoder, *no_tip);
 			return HOP_RETURN;
 		}
-		intel_pt_fup_event(decoder);
+		intel_pt_fup_event(decoder, *no_tip);
 		decoder->state.type |= INTEL_PT_INSTRUCTION | INTEL_PT_BRANCH;
 		*err = intel_pt_walk_fup_tip(decoder);
 		if (!*err && decoder->state.to_ip)
@@ -2789,6 +2944,9 @@ static int intel_pt_hop_trace(struct intel_pt_decoder *decoder, bool *no_tip, in
 	case INTEL_PT_BIP:
 	case INTEL_PT_BEP:
 	case INTEL_PT_BEP_IP:
+	case INTEL_PT_CFE:
+	case INTEL_PT_CFE_IP:
+	case INTEL_PT_EVD:
 	default:
 		return HOP_PROCESS;
 	}
@@ -2857,6 +3015,9 @@ static int intel_pt_psb_lookahead_cb(struct intel_pt_pkt_info *pkt_info)
 	case INTEL_PT_BIP:
 	case INTEL_PT_BEP:
 	case INTEL_PT_BEP_IP:
+	case INTEL_PT_CFE:
+	case INTEL_PT_CFE_IP:
+	case INTEL_PT_EVD:
 		if (data->after_psbend) {
 			data->after_psbend -= 1;
 			if (!data->after_psbend)
@@ -2994,6 +3155,7 @@ next:
 						decoder->pos);
 				break;
 			}
+			intel_pt_sample_iflag_chg(decoder);
 			intel_pt_set_ip(decoder);
 			decoder->state.from_ip = 0;
 			decoder->state.to_ip = decoder->ip;
@@ -3026,7 +3188,7 @@ next:
 			intel_pt_set_last_ip(decoder);
 			if (!decoder->branch_enable || !decoder->pge) {
 				decoder->ip = decoder->last_ip;
-				if (intel_pt_fup_event(decoder))
+				if (intel_pt_fup_event(decoder, no_tip))
 					return 0;
 				no_tip = false;
 				break;
@@ -3108,8 +3270,15 @@ next:
 			break;
 
 		case INTEL_PT_MODE_EXEC:
-			decoder->exec_mode = decoder->packet.payload;
-			break;
+			intel_pt_mode_exec(decoder);
+			err = intel_pt_get_next_packet(decoder);
+			if (err)
+				return err;
+			if (decoder->packet.type == INTEL_PT_FUP) {
+				decoder->set_fup_mode_exec = true;
+				no_tip = true;
+			}
+			goto next;
 
 		case INTEL_PT_MODE_TSX:
 			/* MODE_TSX need not be followed by FUP */
@@ -3223,6 +3392,35 @@ next:
 			}
 			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, true);
+				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:
 			return intel_pt_bug(decoder);
 		}
@@ -3265,6 +3463,9 @@ static int intel_pt_walk_psb(struct intel_pt_decoder *decoder)
 		case INTEL_PT_BIP:
 		case INTEL_PT_BEP:
 		case INTEL_PT_BEP_IP:
+		case INTEL_PT_CFE:
+		case INTEL_PT_CFE_IP:
+		case INTEL_PT_EVD:
 			intel_pt_log("ERROR: Unexpected packet\n");
 			err = -ENOENT;
 			goto out;
@@ -3307,7 +3508,7 @@ static int intel_pt_walk_psb(struct intel_pt_decoder *decoder)
 			break;
 
 		case INTEL_PT_MODE_EXEC:
-			decoder->exec_mode = decoder->packet.payload;
+			intel_pt_mode_exec_status(decoder);
 			break;
 
 		case INTEL_PT_MODE_TSX:
@@ -3426,7 +3627,7 @@ static int intel_pt_walk_to_ip(struct intel_pt_decoder *decoder)
 			break;
 
 		case INTEL_PT_MODE_EXEC:
-			decoder->exec_mode = decoder->packet.payload;
+			intel_pt_mode_exec_status(decoder);
 			break;
 
 		case INTEL_PT_MODE_TSX:
@@ -3476,6 +3677,9 @@ static int intel_pt_walk_to_ip(struct intel_pt_decoder *decoder)
 		case INTEL_PT_BIP:
 		case INTEL_PT_BEP:
 		case INTEL_PT_BEP_IP:
+		case INTEL_PT_CFE:
+		case INTEL_PT_CFE_IP:
+		case INTEL_PT_EVD:
 		default:
 			break;
 		}
@@ -3486,12 +3690,7 @@ static int intel_pt_sync_ip(struct intel_pt_decoder *decoder)
 {
 	int err;
 
-	decoder->set_fup_tx_flags = false;
-	decoder->set_fup_ptw = false;
-	decoder->set_fup_mwait = false;
-	decoder->set_fup_pwre = false;
-	decoder->set_fup_exstop = false;
-	decoder->set_fup_bep = false;
+	intel_pt_clear_fup_event(decoder);
 	decoder->overflow = false;
 
 	if (!decoder->branch_enable) {
diff --git a/tools/perf/util/intel-pt-decoder/intel-pt-decoder.h b/tools/perf/util/intel-pt-decoder/intel-pt-decoder.h
index 8fd68f7a0963..efb2cb3ae0ca 100644
--- a/tools/perf/util/intel-pt-decoder/intel-pt-decoder.h
+++ b/tools/perf/util/intel-pt-decoder/intel-pt-decoder.h
@@ -17,6 +17,7 @@
 
 #define INTEL_PT_IN_TX		(1 << 0)
 #define INTEL_PT_ABORT_TX	(1 << 1)
+#define INTEL_PT_IFLAG		(1 << 2)
 #define INTEL_PT_ASYNC		(1 << 2)
 #define INTEL_PT_FUP_IP		(1 << 3)
 #define INTEL_PT_SAMPLE_IPC	(1 << 4)
@@ -35,6 +36,8 @@ 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,
+	INTEL_PT_IFLAG_CHG	= 1 << 14,
 };
 
 enum intel_pt_period_type {
@@ -209,10 +212,24 @@ 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;
 	bool to_nr;
+	bool from_iflag;
+	bool to_iflag;
 	int err;
 	uint64_t from_ip;
 	uint64_t to_ip;
@@ -234,6 +251,10 @@ 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;
diff --git a/tools/perf/util/intel-pt-decoder/intel-pt-pkt-decoder.c b/tools/perf/util/intel-pt-decoder/intel-pt-pkt-decoder.c
index 4bd154848cad..18f97f43e01a 100644
--- a/tools/perf/util/intel-pt-decoder/intel-pt-pkt-decoder.c
+++ b/tools/perf/util/intel-pt-decoder/intel-pt-pkt-decoder.c
@@ -64,6 +64,9 @@ static const char * const packet_name[] = {
 	[INTEL_PT_BIP]		= "BIP",
 	[INTEL_PT_BEP]		= "BEP",
 	[INTEL_PT_BEP_IP]	= "BEP",
+	[INTEL_PT_CFE]		= "CFE",
+	[INTEL_PT_CFE_IP]	= "CFE",
+	[INTEL_PT_EVD]		= "EVD",
 };
 
 const char *intel_pt_pkt_name(enum intel_pt_pkt_type type)
@@ -197,8 +200,7 @@ static int intel_pt_get_mnt(const unsigned char *buf, size_t len,
 		return INTEL_PT_NEED_MORE_BYTES;
 	packet->type = INTEL_PT_MNT;
 	memcpy_le64(&packet->payload, buf + 3, 8);
-	return 11
-;
+	return 11;
 }
 
 static int intel_pt_get_3byte(const unsigned char *buf, size_t len,
@@ -329,6 +331,29 @@ static int intel_pt_get_bep_ip(size_t len, struct intel_pt_pkt *packet)
 	return 2;
 }
 
+static int intel_pt_get_cfe(const unsigned char *buf, size_t len,
+			    struct intel_pt_pkt *packet)
+{
+	if (len < 4)
+		return INTEL_PT_NEED_MORE_BYTES;
+	packet->type = buf[2] & 0x80 ? INTEL_PT_CFE_IP : INTEL_PT_CFE;
+	packet->count = buf[2] & 0x1f;
+	packet->payload = buf[3];
+	return 4;
+}
+
+static int intel_pt_get_evd(const unsigned char *buf, size_t len,
+			    struct intel_pt_pkt *packet)
+{
+	if (len < 11)
+		return INTEL_PT_NEED_MORE_BYTES;
+	packet->type = INTEL_PT_EVD;
+	packet->count = buf[2] & 0x3f;
+	packet->payload = buf[3];
+	memcpy_le64(&packet->payload, buf + 3, 8);
+	return 11;
+}
+
 static int intel_pt_get_ext(const unsigned char *buf, size_t len,
 			    struct intel_pt_pkt *packet)
 {
@@ -375,6 +400,10 @@ static int intel_pt_get_ext(const unsigned char *buf, size_t len,
 		return intel_pt_get_bep(len, packet);
 	case 0xb3: /* BEP with IP */
 		return intel_pt_get_bep_ip(len, packet);
+	case 0x13: /* CFE */
+		return intel_pt_get_cfe(buf, len, packet);
+	case 0x53: /* EVD */
+		return intel_pt_get_evd(buf, len, packet);
 	default:
 		return INTEL_PT_BAD_PACKET;
 	}
@@ -475,6 +504,7 @@ static int intel_pt_get_mode(const unsigned char *buf, size_t len,
 	switch (buf[1] >> 5) {
 	case 0:
 		packet->type = INTEL_PT_MODE_EXEC;
+		packet->count = buf[1];
 		switch (buf[1] & 3) {
 		case 0:
 			packet->payload = 16;
@@ -624,6 +654,9 @@ void intel_pt_upd_pkt_ctx(const struct intel_pt_pkt *packet,
 	case INTEL_PT_MWAIT:
 	case INTEL_PT_BEP:
 	case INTEL_PT_BEP_IP:
+	case INTEL_PT_CFE:
+	case INTEL_PT_CFE_IP:
+	case INTEL_PT_EVD:
 		*ctx = INTEL_PT_NO_CTX;
 		break;
 	case INTEL_PT_BBP:
@@ -709,7 +742,8 @@ int intel_pt_pkt_desc(const struct intel_pt_pkt *packet, char *buf,
 		return snprintf(buf, buf_len, "%s CTC 0x%x FC 0x%x", name,
 				(unsigned)payload, packet->count);
 	case INTEL_PT_MODE_EXEC:
-		return snprintf(buf, buf_len, "%s %lld", name, payload);
+		return snprintf(buf, buf_len, "%s IF:%d %lld",
+				name, !!(packet->count & 4), payload);
 	case INTEL_PT_MODE_TSX:
 		return snprintf(buf, buf_len, "%s TXAbort:%u InTX:%u",
 				name, (unsigned)(payload >> 1) & 1,
@@ -751,6 +785,13 @@ int intel_pt_pkt_desc(const struct intel_pt_pkt *packet, char *buf,
 	case INTEL_PT_BIP:
 		return snprintf(buf, buf_len, "%s ID 0x%02x Value 0x%llx",
 				name, packet->count, payload);
+	case INTEL_PT_CFE:
+	case INTEL_PT_CFE_IP:
+		return snprintf(buf, buf_len, "%s IP:%d Type 0x%02x Vector 0x%llx",
+				name, packet->type == INTEL_PT_CFE_IP, packet->count, payload);
+	case INTEL_PT_EVD:
+		return snprintf(buf, buf_len, "%s Type 0x%02x Payload 0x%llx",
+				name, packet->count, payload);
 	default:
 		break;
 	}
diff --git a/tools/perf/util/intel-pt-decoder/intel-pt-pkt-decoder.h b/tools/perf/util/intel-pt-decoder/intel-pt-pkt-decoder.h
index 996090cb84f6..496ba4be875c 100644
--- a/tools/perf/util/intel-pt-decoder/intel-pt-pkt-decoder.h
+++ b/tools/perf/util/intel-pt-decoder/intel-pt-pkt-decoder.h
@@ -56,6 +56,9 @@ enum intel_pt_pkt_type {
 	INTEL_PT_BIP,
 	INTEL_PT_BEP,
 	INTEL_PT_BEP_IP,
+	INTEL_PT_CFE,
+	INTEL_PT_CFE_IP,
+	INTEL_PT_EVD,
 };
 
 struct intel_pt_pkt {