perf tools: Support insn and insnlen in perf script

index f545ec1e758a..8bc7fec817d7 100644
@@ -319,15 +319,12 @@ static int intel_bts_get_next_insn(struct intel_bts_queue *btsq, u64 ip)
 	struct machine *machine = btsq->bts->machine;
 	struct thread *thread;
 	struct addr_location al;
-	unsigned char buf[1024];
-	size_t bufsz;
+	unsigned char buf[INTEL_PT_INSN_BUF_SZ];
 	ssize_t len;
 	int x86_64;
 	uint8_t cpumode;
 	int err = -1;
 
-	bufsz = intel_pt_insn_max_size();
-
 	if (machine__kernel_ip(machine, ip))
 		cpumode = PERF_RECORD_MISC_KERNEL;
 	else

@@ -341,7 +338,8 @@ static int intel_bts_get_next_insn(struct intel_bts_queue *btsq, u64 ip)
 	if (!al.map || !al.map->dso)
 		goto out_put;
 
-	len = dso__data_read_addr(al.map->dso, al.map, machine, ip, buf, bufsz);
+	len = dso__data_read_addr(al.map->dso, al.map, machine, ip, buf,
+				  INTEL_PT_INSN_BUF_SZ);
 	if (len <= 0)
 		goto out_put;
 

index d23138c06665..5f95cd442075 100644
@@ -27,6 +27,10 @@
 
 #include "intel-pt-insn-decoder.h"
 
+#if INTEL_PT_INSN_BUF_SZ < MAX_INSN_SIZE
+#error Instruction buffer size too small
+#endif
+
 /* Based on branch_type() from perf_event_intel_lbr.c */
 static void intel_pt_insn_decoder(struct insn *insn,
 				  struct intel_pt_insn *intel_pt_insn)

@@ -166,10 +170,10 @@ int intel_pt_get_insn(const unsigned char *buf, size_t len, int x86_64,
 	if (!insn_complete(&insn) || insn.length > len)
 		return -1;
 	intel_pt_insn_decoder(&insn, intel_pt_insn);
-	if (insn.length < INTEL_PT_INSN_DBG_BUF_SZ)
+	if (insn.length < INTEL_PT_INSN_BUF_SZ)
 		memcpy(intel_pt_insn->buf, buf, insn.length);
 	else
-		memcpy(intel_pt_insn->buf, buf, INTEL_PT_INSN_DBG_BUF_SZ);
+		memcpy(intel_pt_insn->buf, buf, INTEL_PT_INSN_BUF_SZ);
 	return 0;
 }
 

@@ -211,11 +215,6 @@ int intel_pt_insn_desc(const struct intel_pt_insn *intel_pt_insn, char *buf,
 	return 0;
 }
 
-size_t intel_pt_insn_max_size(void)
-{
-	return MAX_INSN_SIZE;
-}
-
 int intel_pt_insn_type(enum intel_pt_insn_op op)
 {
 	switch (op) {

index b0adbf37323e..37ec5627ae9b 100644
@@ -20,7 +20,7 @@
 #include <stdint.h>
 
 #define INTEL_PT_INSN_DESC_MAX		32
-#define INTEL_PT_INSN_DBG_BUF_SZ	16
+#define INTEL_PT_INSN_BUF_SZ		16
 
 enum intel_pt_insn_op {
 	INTEL_PT_OP_OTHER,

@@ -47,7 +47,7 @@ struct intel_pt_insn {
 	enum intel_pt_insn_branch	branch;
 	int				length;
 	int32_t				rel;
-	unsigned char			buf[INTEL_PT_INSN_DBG_BUF_SZ];
+	unsigned char			buf[INTEL_PT_INSN_BUF_SZ];
 };
 
 int intel_pt_get_insn(const unsigned char *buf, size_t len, int x86_64,

@@ -58,8 +58,6 @@ const char *intel_pt_insn_name(enum intel_pt_insn_op op);
 int intel_pt_insn_desc(const struct intel_pt_insn *intel_pt_insn, char *buf,
 		       size_t buf_len);
 
-size_t intel_pt_insn_max_size(void);
-
 int intel_pt_insn_type(enum intel_pt_insn_op op);
 
 #endif

index 319bef33a64b..e02bc7b166a0 100644
@@ -119,8 +119,8 @@ void __intel_pt_log_insn(struct intel_pt_insn *intel_pt_insn, uint64_t ip)
 	if (intel_pt_log_open())
 		return;
 
-	if (len > INTEL_PT_INSN_DBG_BUF_SZ)
-		len = INTEL_PT_INSN_DBG_BUF_SZ;
+	if (len > INTEL_PT_INSN_BUF_SZ)
+		len = INTEL_PT_INSN_BUF_SZ;
 	intel_pt_print_data(intel_pt_insn->buf, len, ip, 8);
 	if (intel_pt_insn_desc(intel_pt_insn, desc, INTEL_PT_INSN_DESC_MAX) > 0)
 		fprintf(f, "%s\n", desc);

index dc041d4368c8..815a14d8904b 100644
@@ -428,8 +428,7 @@ static int intel_pt_walk_next_insn(struct intel_pt_insn *intel_pt_insn,
 	struct machine *machine = ptq->pt->machine;
 	struct thread *thread;
 	struct addr_location al;
-	unsigned char buf[1024];
-	size_t bufsz;
+	unsigned char buf[INTEL_PT_INSN_BUF_SZ];
 	ssize_t len;
 	int x86_64;
 	u8 cpumode;

@@ -440,8 +439,6 @@ static int intel_pt_walk_next_insn(struct intel_pt_insn *intel_pt_insn,
 	if (to_ip && *ip == to_ip)
 		goto out_no_cache;
 
-	bufsz = intel_pt_insn_max_size();
-
 	if (*ip >= ptq->pt->kernel_start)
 		cpumode = PERF_RECORD_MISC_KERNEL;
 	else

@@ -493,7 +490,8 @@ static int intel_pt_walk_next_insn(struct intel_pt_insn *intel_pt_insn,
 
 		while (1) {
 			len = dso__data_read_offset(al.map->dso, machine,
-						    offset, buf, bufsz);
+						    offset, buf,
+						    INTEL_PT_INSN_BUF_SZ);
 			if (len <= 0)
 				return -EINVAL;
 

index 8d363d5e65a2..c735c53a26f8 100644
@@ -177,6 +177,8 @@ enum {
 	PERF_IP_FLAG_TRACE_BEGIN	|\
 	PERF_IP_FLAG_TRACE_END)
 
+#define MAX_INSN 16
+
 struct perf_sample {
 	u64 ip;
 	u32 pid, tid;

@@ -193,6 +195,7 @@ struct perf_sample {
 	u32 flags;
 	u16 insn_len;
 	u8  cpumode;
+	char insn[MAX_INSN];
 	void *raw_data;
 	struct ip_callchain *callchain;
 	struct branch_stack *branch_stack;

index 8bc7fec817d7..6c2eb5da4afc 100644
@@ -295,6 +295,7 @@ static int intel_bts_synth_branch_sample(struct intel_bts_queue *btsq,
 	sample.cpu = btsq->cpu;
 	sample.flags = btsq->sample_flags;
 	sample.insn_len = btsq->intel_pt_insn.length;
+	memcpy(sample.insn, btsq->intel_pt_insn.buf, INTEL_PT_INSN_BUF_SZ);
 
 	if (bts->synth_opts.inject) {
 		event.sample.header.size = bts->branches_event_size;

index 16c06d3ae577..e4e7dc781d21 100644
@@ -980,6 +980,8 @@ out:
 out_no_progress:
 	decoder->state.insn_op = intel_pt_insn->op;
 	decoder->state.insn_len = intel_pt_insn->length;
+	memcpy(decoder->state.insn, intel_pt_insn->buf,
+	       INTEL_PT_INSN_BUF_SZ);
 
 	if (decoder->tx_flags & INTEL_PT_IN_TX)
 		decoder->state.flags |= INTEL_PT_IN_TX;

index 89399985fa4d..e90619a43c0c 100644
@@ -66,6 +66,7 @@ struct intel_pt_state {
 	uint32_t flags;
 	enum intel_pt_insn_op insn_op;
 	int insn_len;
+	char insn[INTEL_PT_INSN_BUF_SZ];
 };
 
 struct intel_pt_insn;

index 5f95cd442075..7913363bde5c 100644
@@ -27,7 +27,7 @@
 
 #include "intel-pt-insn-decoder.h"
 
-#if INTEL_PT_INSN_BUF_SZ < MAX_INSN_SIZE
+#if INTEL_PT_INSN_BUF_SZ < MAX_INSN_SIZE || INTEL_PT_INSN_BUF_SZ > MAX_INSN
 #error Instruction buffer size too small
 #endif
 

index 815a14d8904b..85d5eeb66c75 100644
@@ -143,6 +143,7 @@ struct intel_pt_queue {
 	u32 flags;
 	u16 insn_len;
 	u64 last_insn_cnt;
+	char insn[INTEL_PT_INSN_BUF_SZ];
 };
 
 static void intel_pt_dump(struct intel_pt *pt __maybe_unused,

@@ -315,6 +316,7 @@ struct intel_pt_cache_entry {
 	enum intel_pt_insn_branch	branch;
 	int				length;
 	int32_t				rel;
+	char				insn[INTEL_PT_INSN_BUF_SZ];
 };
 
 static int intel_pt_config_div(const char *var, const char *value, void *data)

@@ -400,6 +402,7 @@ static int intel_pt_cache_add(struct dso *dso, struct machine *machine,
 	e->branch = intel_pt_insn->branch;
 	e->length = intel_pt_insn->length;
 	e->rel = intel_pt_insn->rel;
+	memcpy(e->insn, intel_pt_insn->buf, INTEL_PT_INSN_BUF_SZ);
 
 	err = auxtrace_cache__add(c, offset, &e->entry);
 	if (err)

@@ -436,6 +439,8 @@ static int intel_pt_walk_next_insn(struct intel_pt_insn *intel_pt_insn,
 	u64 insn_cnt = 0;
 	bool one_map = true;
 
+	intel_pt_insn->length = 0;
+
 	if (to_ip && *ip == to_ip)
 		goto out_no_cache;
 

@@ -475,6 +480,8 @@ static int intel_pt_walk_next_insn(struct intel_pt_insn *intel_pt_insn,
 				intel_pt_insn->branch = e->branch;
 				intel_pt_insn->length = e->length;
 				intel_pt_insn->rel = e->rel;
+				memcpy(intel_pt_insn->buf, e->insn,
+				       INTEL_PT_INSN_BUF_SZ);
 				intel_pt_log_insn_no_data(intel_pt_insn, *ip);
 				return 0;
 			}

@@ -898,6 +905,7 @@ static void intel_pt_sample_flags(struct intel_pt_queue *ptq)
 		if (ptq->state->flags & INTEL_PT_IN_TX)
 			ptq->flags |= PERF_IP_FLAG_IN_TX;
 		ptq->insn_len = ptq->state->insn_len;
+		memcpy(ptq->insn, ptq->state->insn, INTEL_PT_INSN_BUF_SZ);
 	}
 }
 

@@ -1078,6 +1086,7 @@ static int intel_pt_synth_branch_sample(struct intel_pt_queue *ptq)
 	sample.cpu = ptq->cpu;
 	sample.flags = ptq->flags;
 	sample.insn_len = ptq->insn_len;
+	memcpy(sample.insn, ptq->insn, INTEL_PT_INSN_BUF_SZ);
 
 	/*
 	 * perf report cannot handle events without a branch stack when using

@@ -1139,6 +1148,7 @@ static int intel_pt_synth_instruction_sample(struct intel_pt_queue *ptq)
 	sample.cpu = ptq->cpu;
 	sample.flags = ptq->flags;
 	sample.insn_len = ptq->insn_len;
+	memcpy(sample.insn, ptq->insn, INTEL_PT_INSN_BUF_SZ);
 
 	ptq->last_insn_cnt = ptq->state->tot_insn_cnt;
 

@@ -1201,6 +1211,7 @@ static int intel_pt_synth_transaction_sample(struct intel_pt_queue *ptq)
 	sample.cpu = ptq->cpu;
 	sample.flags = ptq->flags;
 	sample.insn_len = ptq->insn_len;
+	memcpy(sample.insn, ptq->insn, INTEL_PT_INSN_BUF_SZ);
 
 	if (pt->synth_opts.callchain) {
 		thread_stack__sample(ptq->thread, ptq->chain,

index 7228d141a789..412fb6e65ac0 100644
@@ -66,6 +66,8 @@ enum perf_output_field {
 	PERF_OUTPUT_WEIGHT	    = 1U << 18,
 	PERF_OUTPUT_BPF_OUTPUT	    = 1U << 19,
 	PERF_OUTPUT_CALLINDENT	    = 1U << 20,
+	PERF_OUTPUT_INSN	    = 1U << 21,
+	PERF_OUTPUT_INSNLEN	    = 1U << 22,
 };
 
 struct output_option {

@@ -93,6 +95,8 @@ struct output_option {
 	{.str = "weight",   .field = PERF_OUTPUT_WEIGHT},
 	{.str = "bpf-output",   .field = PERF_OUTPUT_BPF_OUTPUT},
 	{.str = "callindent", .field = PERF_OUTPUT_CALLINDENT},
+	{.str = "insn", .field = PERF_OUTPUT_INSN},
+	{.str = "insnlen", .field = PERF_OUTPUT_INSNLEN},
 };
 
 /* default set to maintain compatibility with current format */

@@ -624,6 +628,20 @@ static void print_sample_callindent(struct perf_sample *sample,
 		printf("%*s", spacing - len, "");
 }
 
+static void print_insn(struct perf_sample *sample,
+		       struct perf_event_attr *attr)
+{
+	if (PRINT_FIELD(INSNLEN))
+		printf(" ilen: %d", sample->insn_len);
+	if (PRINT_FIELD(INSN)) {
+		int i;
+
+		printf(" insn:");
+		for (i = 0; i < sample->insn_len; i++)
+			printf(" %02x", (unsigned char)sample->insn[i]);
+	}
+}
+
 static void print_sample_bts(struct perf_sample *sample,
 			     struct perf_evsel *evsel,
 			     struct thread *thread,

@@ -668,6 +686,8 @@ static void print_sample_bts(struct perf_sample *sample,
 	if (print_srcline_last)
 		map__fprintf_srcline(al->map, al->addr, "\n  ", stdout);
 
+	print_insn(sample, attr);
+
 	printf("\n");
 }
 

@@ -911,7 +931,7 @@ static void process_event(struct perf_script *script,
 
 	if (perf_evsel__is_bpf_output(evsel) && PRINT_FIELD(BPF_OUTPUT))
 		print_sample_bpf_output(sample);
-
+	print_insn(sample, attr);
 	printf("\n");
 }
 

@@ -2124,7 +2144,7 @@ int cmd_script(int argc, const char **argv, const char *prefix __maybe_unused)
 		     "Valid types: hw,sw,trace,raw. "
 		     "Fields: comm,tid,pid,time,cpu,event,trace,ip,sym,dso,"
 		     "addr,symoff,period,iregs,brstack,brstacksym,flags,"
-		     "bpf-output,callindent", parse_output_fields),
+		     "bpf-output,callindent,insn,insnlen", parse_output_fields),
 	OPT_BOOLEAN('a', "all-cpus", &system_wide,
 		    "system-wide collection from all CPUs"),
 	OPT_STRING('S', "symbols", &symbol_conf.sym_list_str, "symbol[,symbol...]",