perf, tools, script: Add brstackinsn for branch stacks

Commit-ID:  48d02a1d5c137d362defd11a5d57d0af4a75a983
Gitweb:     http://git.kernel.org/tip/48d02a1d5c137d362defd11a5d57d0af4a75a983
Author:     Andi Kleen <[email protected]>
AuthorDate: Thu, 23 Feb 2017 15:46:34 -0800
Committer:  Arnaldo Carvalho de Melo <[email protected]>
CommitDate: Thu, 16 Mar 2017 09:24:35 -0300

perf script: Add 'brstackinsn' for branch stacks

Implement printing instruction sequences as hex dump for branch stacks.

This relies on the x86 instruction decoder used by the PT decoder to find the lengths of instructions to dump them individually.

This is good enough for pattern matching.

This allows to study hot paths for individual samples, together with branch misprediction and cycle count / IPC information if available (on Skylake systems).

% perf record -b ...
% perf script -F brstackinsn
...

    read_hpet+67:
          ffffffff9905b843        insn: 74 ea                     # PRED
          ffffffff9905b82f        insn: 85 c9
          ffffffff9905b831        insn: 74 12
          ffffffff9905b833        insn: f3 90
          ffffffff9905b835        insn: 48 8b 0f
          ffffffff9905b838        insn: 48 89 ca
          ffffffff9905b83b        insn: 48 c1 ea 20
          ffffffff9905b83f        insn: 39 f2
          ffffffff9905b841        insn: 89 d0
          ffffffff9905b843        insn: 74 ea                     # PRED

Only works when no special branch filters are specified.

Occasionally the path does not reach up to the sample IP, as the LBRs may be frozen before executing a final jump. In this case we print a special message.

The instruction dumper piggy backs on the existing infrastructure from the IP PT decoder.

An earlier iteration of this patch relied on a disassembler, but this version only uses the existing instruction decoder.

Committer note:

Added hint about how to get suitable perf.data files for use with '-F brstackinsm':

  $ perf record usleep 1
  [ perf record: Woken up 1 times to write data ]
  [ perf record: Captured and wrote 0.018 MB perf.data (8 samples) ]
  $
  $ perf script -F brstackinsn
  Display of branch stack assembler requested, but non all-branch filter set
  Hint: run 'perf record -b ...'
  $

---
 tools/perf/Documentation/perf-script.txt           |  13 +-
 tools/perf/builtin-script.c                        | 264 ++++++++++++++++++++-
 tools/perf/util/Build                              |   1 +
 tools/perf/util/dump-insn.c                        |  14 ++
 tools/perf/util/dump-insn.h                        |  22 ++
 .../util/intel-pt-decoder/intel-pt-insn-decoder.c  |  24 ++
 6 files changed, 327 insertions(+), 11 deletions(-)

index 62c9b0c..cb0eda3 100644
@@ -116,7 +116,7 @@ OPTIONS
 --fields::
         Comma separated list of fields to print. Options are:
         comm, tid, pid, time, cpu, event, trace, ip, sym, dso, addr, symoff,
-        srcline, period, iregs, brstack, brstacksym, flags, bpf-output,
+        srcline, period, iregs, brstack, brstacksym, flags, bpf-output, brstackinsn,
         callindent, insn, insnlen. Field list can be prepended with the type, trace, sw or hw,
         to indicate to which event type the field list applies.
         e.g., -F sw:comm,tid,time,ip,sym  and -F trace:time,cpu,trace

@@ -189,15 +189,20 @@ OPTIONS
 	i.e., -F "" is not allowed.
 
 	The brstack output includes branch related information with raw addresses using the
-	/v/v/v/v/ syntax in the following order:
+	/v/v/v/v/cycles syntax in the following order:
 	FROM: branch source instruction
 	TO  : branch target instruction
         M/P/-: M=branch target mispredicted or branch direction was mispredicted, P=target predicted or direction predicted, -=not supported
 	X/- : X=branch inside a transactional region, -=not in transaction region or not supported
 	A/- : A=TSX abort entry, -=not aborted region or not supported
+	cycles
 
 	The brstacksym is identical to brstack, except that the FROM and TO addresses are printed in a symbolic form if possible.
 
+	When brstackinsn is specified the full assembler sequences of branch sequences for each sample
+	is printed. This is the full execution path leading to the sample. This is only supported when the
+	sample was recorded with perf record -b or -j any.
+
 -k::
 --vmlinux=<file>::
         vmlinux pathname

@@ -302,6 +307,10 @@ include::itrace.txt[]
 	stop time is not given (i.e, time string is 'x.y,') then analysis goes
 	to end of file.
 
+--max-blocks::
+	Set the maximum number of program blocks to print with brstackasm for
+	each sample.
+
 SEE ALSO
 --------
 linkperf:perf-record[1], linkperf:perf-script-perl[1],

index 2ea5ee1..fb4f42f 100644
@@ -82,6 +82,7 @@ libperf-$(CONFIG_AUXTRACE) += intel-pt-decoder/
 libperf-$(CONFIG_AUXTRACE) += intel-pt.o
 libperf-$(CONFIG_AUXTRACE) += intel-bts.o
 libperf-y += parse-branch-options.o
+libperf-y += dump-insn.o
 libperf-y += parse-regs-options.o
 libperf-y += term.o
 libperf-y += help-unknown-cmd.o

new file mode 100644
index 0000000..ffbdb19
@@ -0,0 +1,14 @@
+#include <linux/compiler.h>
+#include "dump-insn.h"
+
+/* Fallback code */
+
+__weak
+const char *dump_insn(struct perf_insn *x __maybe_unused,
+		      u64 ip __maybe_unused, u8 *inbuf __maybe_unused,
+		      int inlen __maybe_unused, int *lenp)
+{
+	if (lenp)
+		*lenp = 0;
+	return "?";
+}

new file mode 100644
index 0000000..90fb115
@@ -0,0 +1,22 @@
+#ifndef __PERF_DUMP_INSN_H
+#define __PERF_DUMP_INSN_H 1
+
+#define MAXINSN 15
+
+#include <linux/types.h>
+
+struct thread;
+
+struct perf_insn {
+	/* Initialized by callers: */
+	struct thread *thread;
+	u8	      cpumode;
+	bool	      is64bit;
+	int	      cpu;
+	/* Temporary */
+	char	      out[256];
+};
+
+const char *dump_insn(struct perf_insn *x, u64 ip,
+		      u8 *inbuf, int inlen, int *lenp);
+#endif

index 55b6250..a5f35b2 100644
@@ -26,6 +26,7 @@
 #include "insn.c"
 
 #include "intel-pt-insn-decoder.h"
+#include "dump-insn.h"
 
 #if INTEL_PT_INSN_BUF_SZ < MAX_INSN_SIZE || INTEL_PT_INSN_BUF_SZ > MAX_INSN
 #error Instruction buffer size too small

@@ -179,6 +180,29 @@ int intel_pt_get_insn(const unsigned char *buf, size_t len, int x86_64,
 	return 0;
 }
 
+const char *dump_insn(struct perf_insn *x, uint64_t ip __maybe_unused,
+		      u8 *inbuf, int inlen, int *lenp)
+{
+	struct insn insn;
+	int n, i;
+	int left;
+
+	insn_init(&insn, inbuf, inlen, x->is64bit);
+	insn_get_length(&insn);
+	if (!insn_complete(&insn) || insn.length > inlen)
+		return "<bad>";
+	if (lenp)
+		*lenp = insn.length;
+	left = sizeof(x->out);
+	n = snprintf(x->out, left, "insn: ");
+	left -= n;
+	for (i = 0; i < insn.length; i++) {
+		n += snprintf(x->out + n, left, "%02x ", inbuf[i]);
+		left -= n;
+	}
+	return x->out;
+}
+
 const char *branch_name[] = {
 	[INTEL_PT_OP_OTHER]	= "Other",
 	[INTEL_PT_OP_CALL]	= "Call",