perf intel-pt/bts: Report instruction bytes and length in sample

Change Intel PT and BTS to pass up the length and the instruction
bytes of the decoded or sampled instruction in the perf sample.

The decoder already knows this information, we just need to pass it
up. Since it is only a couple of movs it is not very expensive.

Handle instruction cache too. Make sure ilen is always initialized.

Used in the next patch.

[Adrian: re-base on top (and adjust for) instruction buffer size tidy-up]
[Adrian: add BTS support and adjust commit message accordingly]

Signed-off-by: Adrian Hunter <adrian.hunter@intel.com>
Link: http://lkml.kernel.org/r/1475847747-30994-3-git-send-email-adrian.hunter@intel.com
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>

tools/perf/util/event.h

@@ -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;

tools/perf/util/intel-bts.c

@@ -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;

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

@@ -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;

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

@@ -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;

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

@@ -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
 

tools/perf/util/intel-pt.c

@@ -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,