[PATCH 2/2] perf intel-pt: Add support for new branch instructions ERETS and ERETU

[Adrian Hunter]

Intel Flexible Return and Event Delivery (FRED) adds instructions ERETS
(return to supervisor) and ERETU (return to user). Intel PT instruction
decoder needs to know about these instructions because they are
branch instructions. Similar to IRET instructions, when the decoder
encounters one of these instructions it will match it to a TIP (target
instruction pointer) packet that informs what the branch destination is.

The existing "x86 instruction decoder - new instructions" test can be
used to test the result e.g.

  $ perf test -v ins |& grep eret
  Decoded ok: f2 0f 01 ca         erets
  Decoded ok: f3 0f 01 ca         eretu

Signed-off-by: Adrian Hunter <adrian.hunter@xxxxxxxxx>
---
 tools/perf/arch/x86/tests/insn-x86.c           |  4 ++++
 .../intel-pt-decoder/intel-pt-insn-decoder.c   | 18 ++++++++++++++++++
 .../intel-pt-decoder/intel-pt-insn-decoder.h   |  2 ++
 3 files changed, 24 insertions(+)

diff --git a/tools/perf/arch/x86/tests/insn-x86.c b/tools/perf/arch/x86/tests/insn-x86.c
index 94b490c434d0..735257d205b5 100644
--- a/tools/perf/arch/x86/tests/insn-x86.c
+++ b/tools/perf/arch/x86/tests/insn-x86.c
@@ -29,6 +29,8 @@ struct test_data test_data_64[] = {
 #include "insn-x86-dat-64.c"
 	{{0x0f, 0x01, 0xee}, 3, 0, NULL, NULL, "0f 01 ee             \trdpkru"},
 	{{0x0f, 0x01, 0xef}, 3, 0, NULL, NULL, "0f 01 ef             \twrpkru"},
+	{{0xf2, 0x0f, 0x01, 0xca}, 4, 0, "erets", "indirect", "f2 0f 01 ca  \terets"},
+	{{0xf3, 0x0f, 0x01, 0xca}, 4, 0, "eretu", "indirect", "f3 0f 01 ca  \teretu"},
 	{{0}, 0, 0, NULL, NULL, NULL},
 };
 
@@ -49,6 +51,8 @@ static int get_op(const char *op_str)
 		{"syscall", INTEL_PT_OP_SYSCALL},
 		{"sysret",  INTEL_PT_OP_SYSRET},
 		{"vmentry",  INTEL_PT_OP_VMENTRY},
+		{"erets",   INTEL_PT_OP_ERETS},
+		{"eretu",   INTEL_PT_OP_ERETU},
 		{NULL, 0},
 	};
 	struct val_data *val;
diff --git a/tools/perf/util/intel-pt-decoder/intel-pt-insn-decoder.c b/tools/perf/util/intel-pt-decoder/intel-pt-insn-decoder.c
index 22308dd93010..c5d57027ec23 100644
--- a/tools/perf/util/intel-pt-decoder/intel-pt-insn-decoder.c
+++ b/tools/perf/util/intel-pt-decoder/intel-pt-insn-decoder.c
@@ -52,6 +52,20 @@ static void intel_pt_insn_decoder(struct insn *insn,
 				op = INTEL_PT_OP_VMENTRY;
 				branch = INTEL_PT_BR_INDIRECT;
 				break;
+			case 0xca:
+				switch (insn->prefixes.bytes[3]) {
+				case 0xf2: /* erets */
+					op = INTEL_PT_OP_ERETS;
+					branch = INTEL_PT_BR_INDIRECT;
+					break;
+				case 0xf3: /* eretu */
+					op = INTEL_PT_OP_ERETU;
+					branch = INTEL_PT_BR_INDIRECT;
+					break;
+				default:
+					break;
+				}
+				break;
 			default:
 				break;
 			}
@@ -230,6 +244,8 @@ const char *branch_name[] = {
 	[INTEL_PT_OP_SYSCALL]	= "Syscall",
 	[INTEL_PT_OP_SYSRET]	= "Sysret",
 	[INTEL_PT_OP_VMENTRY]	= "VMentry",
+	[INTEL_PT_OP_ERETS]	= "Erets",
+	[INTEL_PT_OP_ERETU]	= "Eretu",
 };
 
 const char *intel_pt_insn_name(enum intel_pt_insn_op op)
@@ -273,6 +289,8 @@ int intel_pt_insn_type(enum intel_pt_insn_op op)
 	case INTEL_PT_OP_LOOP:
 		return PERF_IP_FLAG_BRANCH | PERF_IP_FLAG_CONDITIONAL;
 	case INTEL_PT_OP_IRET:
+	case INTEL_PT_OP_ERETS:
+	case INTEL_PT_OP_ERETU:
 		return PERF_IP_FLAG_BRANCH | PERF_IP_FLAG_RETURN |
 		       PERF_IP_FLAG_INTERRUPT;
 	case INTEL_PT_OP_INT:
diff --git a/tools/perf/util/intel-pt-decoder/intel-pt-insn-decoder.h b/tools/perf/util/intel-pt-decoder/intel-pt-insn-decoder.h
index e3338b56a75f..7fb7fe3a1566 100644
--- a/tools/perf/util/intel-pt-decoder/intel-pt-insn-decoder.h
+++ b/tools/perf/util/intel-pt-decoder/intel-pt-insn-decoder.h
@@ -25,6 +25,8 @@ enum intel_pt_insn_op {
 	INTEL_PT_OP_SYSCALL,
 	INTEL_PT_OP_SYSRET,
 	INTEL_PT_OP_VMENTRY,
+	INTEL_PT_OP_ERETS,
+	INTEL_PT_OP_ERETU,
 };
 
 enum intel_pt_insn_branch {
-- 
2.34.1