perf tools: Introduce an abstraction for Instruction Tracing

index 7e23dae..cfbe2b9 100644
@@ -816,7 +816,15 @@ static int perf_evlist__mmap_per_evsel(struct perf_evlist *evlist, int idx,
 			perf_evlist__mmap_get(evlist, idx);
 		}
 
-		if (__perf_evlist__add_pollfd(evlist, fd, idx) < 0) {
+		/*
+		 * The system_wide flag causes a selected event to be opened
+		 * always without a pid.  Consequently it will never get a
+		 * POLLHUP, but it is used for tracking in combination with
+		 * other events, so it should not need to be polled anyway.
+		 * Therefore don't add it for polling.
+		 */
+		if (!evsel->system_wide &&
+		    __perf_evlist__add_pollfd(evlist, fd, idx) < 0) {
 			perf_evlist__mmap_put(evlist, idx);
 			return -1;
 		}

index a2157f0..9ebc8b1 100644
@@ -103,15 +103,14 @@ struct comm *thread__exec_comm(const struct thread *thread)
 	return last;
 }
 
-/* CHECKME: time should always be 0 if event aren't ordered */
 int __thread__set_comm(struct thread *thread, const char *str, u64 timestamp,
 		       bool exec)
 {
 	struct comm *new, *curr = thread__comm(thread);
 	int err;
 
-	/* Override latest entry if it had no specific time coverage */
-	if (!curr->start && !curr->exec) {
+	/* Override the default :tid entry */
+	if (!thread->comm_set) {
 		err = comm__override(curr, str, timestamp, exec);
 		if (err)
 			return err;

index e86fffd..27f8fb5 100644
@@ -869,6 +869,13 @@ static int write_branch_stack(int fd __maybe_unused,
 	return 0;
 }
 
+static int write_itrace(int fd __maybe_unused,
+			struct perf_header *h __maybe_unused,
+			struct perf_evlist *evlist __maybe_unused)
+{
+	return 0;
+}
+
 static void print_hostname(struct perf_header *ph, int fd __maybe_unused,
 			   FILE *fp)
 {

@@ -1163,6 +1170,12 @@ static void print_branch_stack(struct perf_header *ph __maybe_unused,
 	fprintf(fp, "# contains samples with branch stack\n");
 }
 
+static void print_itrace(struct perf_header *ph __maybe_unused,
+			 int fd __maybe_unused, FILE *fp)
+{
+	fprintf(fp, "# contains Instruction Traces\n");
+}
+
 static void print_pmu_mappings(struct perf_header *ph, int fd __maybe_unused,
 			       FILE *fp)
 {

@@ -1873,6 +1886,7 @@ static const struct feature_ops feat_ops[HEADER_LAST_FEATURE] = {
 	FEAT_OPA(HEADER_BRANCH_STACK,	branch_stack),
 	FEAT_OPP(HEADER_PMU_MAPPINGS,	pmu_mappings),
 	FEAT_OPP(HEADER_GROUP_DESC,	group_desc),
+	FEAT_OPA(HEADER_ITRACE,		itrace),
 };
 
 struct header_print_data {

index 3bb90ac..990edcf 100644
@@ -30,6 +30,7 @@ enum {
 	HEADER_BRANCH_STACK,
 	HEADER_PMU_MAPPINGS,
 	HEADER_GROUP_DESC,
+	HEADER_ITRACE,
 	HEADER_LAST_FEATURE,
 	HEADER_FEAT_BITS	= 256,
 };

124 lines not on this page · raw message

index 649b0c5..f19d89c 100644
@@ -8,6 +8,7 @@
 #include "event.h"
 #include "evsel.h"
 #include "util.h"
+#include "itrace.h"
 #include <unistd.h>
 
 struct pollfd;

@@ -28,6 +29,7 @@ struct perf_mmap {
 	int		 mask;
 	int		 refcnt;
 	unsigned int	 prev;
+	struct itrace_mmap itrace_mmap;
 	char		 event_copy[PERF_SAMPLE_MAX_SIZE];
 };
 

@@ -123,6 +125,9 @@ int perf_evlist__parse_mmap_pages(const struct option *opt,
 				  const char *str,
 				  int unset);
 
+int perf_evlist__mmap_ex(struct perf_evlist *evlist, unsigned int pages,
+			 bool overwrite, unsigned int itrace_pages,
+			 bool itrace_overwrite);
 int perf_evlist__mmap(struct perf_evlist *evlist, unsigned int pages,
 		      bool overwrite);
 void perf_evlist__munmap(struct perf_evlist *evlist);

99 lines not on this page · raw message

96 lines not on this page · raw message

index 6c6d044..efe6475 100644
@@ -29,6 +29,8 @@ static const char *perf_event__names[] = {
 	[PERF_RECORD_HEADER_BUILD_ID]		= "BUILD_ID",
 	[PERF_RECORD_FINISHED_ROUND]		= "FINISHED_ROUND",
 	[PERF_RECORD_ID_INDEX]			= "ID_INDEX",
+	[PERF_RECORD_ITRACE_INFO]		= "ITRACE_INFO",
+	[PERF_RECORD_ITRACE]			= "ITRACE",
 };
 
 const char *perf_event__name(unsigned int id)

index 7be3897..21d341e 100644
@@ -214,6 +214,8 @@ enum perf_user_event_type { /* above any possible kernel type */
 	PERF_RECORD_HEADER_BUILD_ID		= 67,
 	PERF_RECORD_FINISHED_ROUND		= 68,
 	PERF_RECORD_ID_INDEX			= 69,
+	PERF_RECORD_ITRACE_INFO			= 70,
+	PERF_RECORD_ITRACE			= 71,
 	PERF_RECORD_HEADER_MAX
 };
 

@@ -279,6 +281,24 @@ struct id_index_event {
 	struct id_index_entry entries[0];
 };
 
+struct itrace_info_event {
+	struct perf_event_header header;
+	u32 type;
+	u32 reserved__; /* For alignment */
+	u64 priv[];
+};
+
+struct itrace_event {
+	struct perf_event_header header;
+	u64 size;
+	u64 offset;
+	u64 reference;
+	u32 idx;
+	u32 tid;
+	u32 cpu;
+	u32 reserved__; /* For alignment */
+};
+
 union perf_event {
 	struct perf_event_header	header;
 	struct mmap_event		mmap;

@@ -294,6 +314,8 @@ union perf_event {
 	struct tracing_data_event	tracing_data;
 	struct build_id_event		build_id;
 	struct id_index_event		id_index;
+	struct itrace_info_event	itrace_info;
+	struct itrace_event		itrace;
 };
 
 void perf_event__print_totals(void);

99 lines not on this page · raw message

index bb2708b..b734c1a 100644
@@ -3,6 +3,8 @@
 
 #include <stdbool.h>
 
+#include <linux/types.h>
+
 struct perf_session;
 union perf_event;
 struct perf_evlist;

@@ -25,6 +27,9 @@ typedef int (*event_attr_op)(struct perf_tool *tool,
 typedef int (*event_op2)(struct perf_tool *tool, union perf_event *event,
 			 struct perf_session *session);
 
+typedef s64 (*event_op3)(struct perf_tool *tool, union perf_event *event,
+			 struct perf_session *session);
+
 struct perf_tool {
 	event_sample	sample,
 			read;

@@ -40,7 +45,9 @@ struct perf_tool {
 	event_op2	tracing_data;
 	event_op2	finished_round,
 			build_id,
-			id_index;
+			id_index,
+			itrace_info;
+	event_op3	itrace;
 	bool		ordered_events;
 	bool		ordering_requires_timestamps;
 };

index 511c2831..24c97bc 100644
@@ -52,8 +52,10 @@ struct record_opts {
 	bool	     sample_weight;
 	bool	     sample_time;
 	bool	     period;
+	bool	     full_itrace;
 	unsigned int freq;
 	unsigned int mmap_pages;
+	unsigned int itrace_mmap_pages;
 	unsigned int user_freq;
 	u64          branch_stack;
 	u64	     default_interval;

210 lines not on this page · raw message

81 lines not on this page · raw message

index cf69325..656eaa3 100644
@@ -119,7 +119,16 @@ void perf_evlist__config(struct perf_evlist *evlist, struct record_opts *opts)
 			evsel->attr.comm_exec = 1;
 	}
 
-	if (evlist->nr_entries > 1) {
+	if (opts->full_itrace) {
+		/*
+		 * Need to be able to synthesize and parse selected events with
+		 * arbitrary sample types, which requires always being able to
+		 * match the id.
+		 */
+		use_sample_identifier = true;
+		evlist__for_each(evlist, evsel)
+			perf_evsel__set_sample_id(evsel, use_sample_identifier);
+	} else if (evlist->nr_entries > 1) {
 		struct perf_evsel *first = perf_evlist__first(evlist);
 
 		evlist__for_each(evlist, evsel) {

index 582c4da..783b5ba 100644
@@ -27,6 +27,7 @@
 #include "util/cpumap.h"
 #include "util/thread_map.h"
 #include "util/data.h"
+#include "util/itrace.h"
 
 #include <unistd.h>
 #include <sched.h>

@@ -38,6 +39,7 @@ struct record {
 	struct record_opts	opts;
 	u64			bytes_written;
 	struct perf_data_file	file;
+	struct itrace_record	*itr;
 	struct perf_evlist	*evlist;
 	struct perf_session	*session;
 	const char		*progname;

@@ -110,6 +112,43 @@ out:
 	return rc;
 }
 
+static int record__process_itrace(struct perf_tool *tool,
+				  union perf_event *event, void *data1,
+				  size_t len1, void *data2, size_t len2)
+{
+	struct record *rec = container_of(tool, struct record, tool);
+	size_t padding;
+	u8 pad[8] = {0};
+
+	/* event.itrace.size includes padding, see __itrace_mmap__read() */
+	padding = (len1 + len2) & 7;
+	if (padding)
+		padding = 8 - padding;
+
+	record__write(rec, event, event->header.size);
+	record__write(rec, data1, len1);
+	record__write(rec, data2, len2);
+	record__write(rec, &pad, padding);
+
+	return 0;
+}
+
+static int record__itrace_mmap_read(struct record *rec,
+				    struct itrace_mmap *mm)
+{
+	int ret;
+
+	ret = itrace_mmap__read(mm, rec->itr, &rec->tool,
+				record__process_itrace);
+	if (ret < 0)
+		return ret;
+
+	if (ret)
+		rec->samples++;
+
+	return 0;
+}
+
 static volatile int done = 0;
 static volatile int signr = -1;
 static volatile int child_finished = 0;

@@ -168,13 +207,15 @@ try_again:
 		goto out;
 	}
 
-	if (perf_evlist__mmap(evlist, opts->mmap_pages, false) < 0) {
+	if (perf_evlist__mmap_ex(evlist, opts->mmap_pages, false,
+				 opts->itrace_mmap_pages, false) < 0) {
 		if (errno == EPERM) {
 			pr_err("Permission error mapping pages.\n"
 			       "Consider increasing "
 			       "/proc/sys/kernel/perf_event_mlock_kb,\n"
 			       "or try again with a smaller value of -m/--mmap_pages.\n"
-			       "(current value: %u)\n", opts->mmap_pages);
+			       "(current value: %u,%u)\n",
+			       opts->mmap_pages, opts->itrace_mmap_pages);
 			rc = -errno;
 		} else {
 			pr_err("failed to mmap with %d (%s)\n", errno,

@@ -257,12 +298,20 @@ static int record__mmap_read_all(struct record *rec)
 	int rc = 0;
 
 	for (i = 0; i < rec->evlist->nr_mmaps; i++) {
+		struct itrace_mmap *mm = &rec->evlist->mmap[i].itrace_mmap;
+
 		if (rec->evlist->mmap[i].base) {
 			if (record__mmap_read(rec, i) != 0) {
 				rc = -1;
 				goto out;
 			}
 		}
+
+		if (mm->base &&
+		    record__itrace_mmap_read(rec, mm) != 0) {
+			rc = -1;
+			goto out;
+		}
 	}
 
 	/*

@@ -292,6 +341,9 @@ static void record__init_features(struct record *rec)
 
 	if (!rec->opts.branch_stack)
 		perf_header__clear_feat(&session->header, HEADER_BRANCH_STACK);
+
+	if (!rec->opts.full_itrace)
+		perf_header__clear_feat(&session->header, HEADER_ITRACE);
 }
 
 static volatile int workload_exec_errno;

@@ -407,6 +459,13 @@ static int __cmd_record(struct record *rec, int argc, const char **argv)
 		}
 	}
 
+	if (rec->opts.full_itrace) {
+		err = perf_event__synthesize_itrace_info(rec->itr, tool,
+					session, process_synthesized_event);
+		if (err)
+			goto out_delete_session;
+	}
+
 	err = perf_event__synthesize_kernel_mmap(tool, process_synthesized_event,
 						 machine);
 	if (err < 0)

@@ -505,16 +564,17 @@ static int __cmd_record(struct record *rec, int argc, const char **argv)
 	}
 
 	if (!quiet) {
-		fprintf(stderr, "[ perf record: Woken up %ld times to write data ]\n", waking);
-
-		/*
-		 * Approximate RIP event size: 24 bytes.
-		 */
-		fprintf(stderr,
-			"[ perf record: Captured and wrote %.3f MB %s (~%" PRIu64 " samples) ]\n",
-			(double)rec->bytes_written / 1024.0 / 1024.0,
-			file->path,
-			rec->bytes_written / 24);
+		fprintf(stderr, "[ perf record: Woken up %ld times to write data ]\n",
+			waking);
+		fprintf(stderr, "[ perf record: Captured and wrote %.3f MB %s",
+			(double)rec->bytes_written / 1024.0 / 1024.0, file->path);
+		if (rec->opts.full_itrace) {
+			fprintf(stderr, " ]\n");
+		} else {
+			/* Approximate RIP event size: 24 bytes */
+			fprintf(stderr, " (~%" PRIu64 " samples) ]\n",
+				rec->bytes_written / 24);
+		}
 	}
 
 out_child:

@@ -818,13 +878,18 @@ struct option *record_options = __record_options;
 
 int cmd_record(int argc, const char **argv, const char *prefix __maybe_unused)
 {
-	int err = -ENOMEM;
+	int err;
 	struct record *rec = &record;
 	char errbuf[BUFSIZ];
 
+	rec->itr = itrace_record__init(&err);
+	if (err)
+		return err;
+
+	err = -ENOMEM;
 	rec->evlist = perf_evlist__new();
 	if (rec->evlist == NULL)
-		return -ENOMEM;
+		goto out_itrace_free;
 
 	perf_config(perf_record_config, rec);
 

@@ -884,6 +949,10 @@ int cmd_record(int argc, const char **argv, const char *prefix __maybe_unused)
 	if (perf_evlist__create_maps(rec->evlist, &rec->opts.target) < 0)
 		usage_with_options(record_usage, record_options);
 
+	err = itrace_record__options(rec->itr, rec->evlist, &rec->opts);
+	if (err)
+		goto out_symbol_exit;
+
 	if (record_opts__config(&rec->opts)) {
 		err = -EINVAL;
 		goto out_symbol_exit;

@@ -893,5 +962,7 @@ int cmd_record(int argc, const char **argv, const char *prefix __maybe_unused)
 out_symbol_exit:
 	perf_evlist__delete(rec->evlist);
 	symbol__exit();
+out_itrace_free:
+	itrace_record__free(rec->itr);
 	return err;
 }

61 lines not on this page · raw message

index 2ee7f46..27d4de3 100644
@@ -984,10 +984,8 @@ static long parse_pages_arg(const char *str, unsigned long min,
 	return pages;
 }
 
-int perf_evlist__parse_mmap_pages(const struct option *opt, const char *str,
-				  int unset __maybe_unused)
+int __perf_evlist__parse_mmap_pages(unsigned int *mmap_pages, const char *str)
 {
-	unsigned int *mmap_pages = opt->value;
 	unsigned long max = UINT_MAX;
 	long pages;
 

@@ -1004,6 +1002,12 @@ int perf_evlist__parse_mmap_pages(const struct option *opt, const char *str,
 	return 0;
 }
 
+int perf_evlist__parse_mmap_pages(const struct option *opt, const char *str,
+				  int unset __maybe_unused)
+{
+	return __perf_evlist__parse_mmap_pages(opt->value, str);
+}
+
 /**
  * perf_evlist__mmap_ex - Create mmaps to receive events.
  * @evlist: list of events

index f19d89c..fcf2975c 100644
@@ -121,6 +121,7 @@ int perf_evlist__start_workload(struct perf_evlist *evlist);
 
 struct option;
 
+int __perf_evlist__parse_mmap_pages(unsigned int *mmap_pages, const char *str);
 int perf_evlist__parse_mmap_pages(const struct option *opt,
 				  const char *str,
 				  int unset);

index efe6475..573fed2 100644
@@ -31,6 +31,7 @@ static const char *perf_event__names[] = {
 	[PERF_RECORD_ID_INDEX]			= "ID_INDEX",
 	[PERF_RECORD_ITRACE_INFO]		= "ITRACE_INFO",
 	[PERF_RECORD_ITRACE]			= "ITRACE",
+	[PERF_RECORD_ITRACE_ERROR]		= "ITRACE_ERROR",
 };
 
 const char *perf_event__name(unsigned int id)

index 21d341e..2a9b99f 100644
@@ -216,6 +216,7 @@ enum perf_user_event_type { /* above any possible kernel type */
 	PERF_RECORD_ID_INDEX			= 69,
 	PERF_RECORD_ITRACE_INFO			= 70,
 	PERF_RECORD_ITRACE			= 71,
+	PERF_RECORD_ITRACE_ERROR		= 72,
 	PERF_RECORD_HEADER_MAX
 };
 

@@ -299,6 +300,20 @@ struct itrace_event {
 	u32 reserved__; /* For alignment */
 };
 
+#define MAX_ITRACE_ERROR_MSG 64
+
+struct itrace_error_event {
+	struct perf_event_header header;
+	u32 type;
+	u32 code;
+	u32 cpu;
+	u32 pid;
+	u32 tid;
+	u32 reserved__; /* For alignment */
+	u64 ip;
+	char msg[MAX_ITRACE_ERROR_MSG];
+};
+
 union perf_event {
 	struct perf_event_header	header;
 	struct mmap_event		mmap;

@@ -316,6 +331,7 @@ union perf_event {
 	struct id_index_event		id_index;
 	struct itrace_info_event	itrace_info;
 	struct itrace_event		itrace;
+	struct itrace_error_event	itrace_error;
 };
 
 void perf_event__print_totals(void);

index c1639c7..89ca176 100644
@@ -271,6 +271,15 @@ static s64 process_event_itrace_stub(struct perf_tool *tool __maybe_unused,
 	return event->itrace.size;
 }
 
+static
+int process_event_itrace_error_stub(struct perf_tool *tool __maybe_unused,
+				    union perf_event *event __maybe_unused,
+				    struct perf_session *session __maybe_unused)
+{
+	dump_printf(": unhandled!\n");
+	return 0;
+}
+
 void perf_tool__fill_defaults(struct perf_tool *tool)
 {
 	if (tool->sample == NULL)

@@ -311,6 +320,8 @@ void perf_tool__fill_defaults(struct perf_tool *tool)
 		tool->itrace_info = process_event_itrace_info_stub;
 	if (tool->itrace == NULL)
 		tool->itrace = process_event_itrace_stub;
+	if (tool->itrace_error == NULL)
+		tool->itrace_error = process_event_itrace_error_stub;
 }
  
 static void swap_sample_id_all(union perf_event *event, void *data)

@@ -514,6 +525,17 @@ static void perf_event__itrace_swap(union perf_event *event,
 	event->itrace.cpu       = bswap_32(event->itrace.cpu);
 }
 
+static void perf_event__itrace_error_swap(union perf_event *event,
+					  bool sample_id_all __maybe_unused)
+{
+	event->itrace_error.type = bswap_32(event->itrace_error.type);
+	event->itrace_error.code = bswap_32(event->itrace_error.code);
+	event->itrace_error.cpu  = bswap_32(event->itrace_error.cpu);
+	event->itrace_error.pid  = bswap_32(event->itrace_error.pid);
+	event->itrace_error.tid  = bswap_32(event->itrace_error.tid);
+	event->itrace_error.ip   = bswap_64(event->itrace_error.ip);
+}
+
 typedef void (*perf_event__swap_op)(union perf_event *event,
 				    bool sample_id_all);
 

@@ -535,6 +557,7 @@ static perf_event__swap_op perf_event__swap_ops[] = {
 	[PERF_RECORD_ID_INDEX]		  = perf_event__all64_swap,
 	[PERF_RECORD_ITRACE_INFO]	  = perf_event__itrace_info_swap,
 	[PERF_RECORD_ITRACE]		  = perf_event__itrace_swap,
+	[PERF_RECORD_ITRACE_ERROR]	  = perf_event__itrace_error_swap,
 	[PERF_RECORD_HEADER_MAX]	  = NULL,
 };
 

@@ -971,6 +994,8 @@ static s64 perf_session__process_user_event(struct perf_session *session,
 		/* setup for reading amidst mmap */
 		lseek(fd, file_offset + event->header.size, SEEK_SET);
 		return tool->itrace(tool, event, session);
+	case PERF_RECORD_ITRACE_ERROR:
+		return tool->itrace_error(tool, event, session);
 	default:
 		return -EINVAL;
 	}

index b734c1a..b5c55de 100644
@@ -46,7 +46,8 @@ struct perf_tool {
 	event_op2	finished_round,
 			build_id,
 			id_index,
-			itrace_info;
+			itrace_info,
+			itrace_error;
 	event_op3	itrace;
 	bool		ordered_events;
 	bool		ordering_requires_timestamps;

index 44b6a01..1cf5d16 100644
@@ -23,6 +23,7 @@
 #include <linux/types.h>
 
 #include "../perf.h"
+#include "session.h"
 
 union perf_event;
 struct perf_session;

@@ -32,6 +33,26 @@ struct record_opts;
 struct itrace_info_event;
 
 /**
+ * struct itrace - session callbacks to allow Instruction Trace data decoding.
+ * @process_event: lets the decoder see all session events
+ * @flush_events: process any remaining data
+ * @free_events: free resources associated with event processing
+ * @free: free resources associated with the session
+ * @error_count: number of errors
+ */
+struct itrace {
+	int (*process_event)(struct perf_session *session,
+			     union perf_event *event,
+			     struct perf_sample *sample,
+			     struct perf_tool *tool);
+	int (*flush_events)(struct perf_session *session,
+			    struct perf_tool *tool);
+	void (*free_events)(struct perf_session *session);
+	void (*free)(struct perf_session *session);
+	unsigned long long error_count;
+};
+
+/**
  * struct itrace_mmap - records an mmap of the itrace buffer.
  * @base: address of mapped area
  * @userpg: pointer to buffer's perf_event_mmap_page

@@ -150,4 +171,40 @@ int perf_event__synthesize_itrace(struct perf_tool *tool,
 				  size_t size, u64 offset, u64 ref, int idx,
 				  u32 tid, u32 cpu);
 
+static inline int itrace__process_event(struct perf_session *session,
+					union perf_event *event,
+					struct perf_sample *sample,
+					struct perf_tool *tool)
+{
+	if (!session->itrace)
+		return 0;
+
+	return session->itrace->process_event(session, event, sample, tool);
+}
+
+static inline int itrace__flush_events(struct perf_session *session,
+				       struct perf_tool *tool)
+{
+	if (!session->itrace)
+		return 0;
+
+	return session->itrace->flush_events(session, tool);
+}
+
+static inline void itrace__free_events(struct perf_session *session)
+{
+	if (!session->itrace)
+		return;
+
+	return session->itrace->free_events(session);
+}
+
+static inline void itrace__free(struct perf_session *session)
+{
+	if (!session->itrace)
+		return;
+
+	return session->itrace->free(session);
+}
+
 #endif

114 lines not on this page · raw message

index dc26ebf..6f8ac2f 100644
@@ -15,10 +15,13 @@
 struct ip_callchain;
 struct thread;
 
+struct itrace;
+
 struct perf_session {
 	struct perf_header	header;
 	struct machines		machines;
 	struct perf_evlist	*evlist;
+	struct itrace		*itrace;
 	struct trace_event	tevent;
 	struct events_stats	stats;
 	bool			repipe;

index 08372a9..2a3a7ce 100644
@@ -34,6 +34,7 @@
 
 #include "event.h"
 #include "debug.h"
+#include "parse-options.h"
 
 int itrace_mmap__mmap(struct itrace_mmap *mm, struct itrace_mmap_params *mp,
 		      void *userpg, int fd)

@@ -205,6 +206,136 @@ int perf_event__synthesize_itrace(struct perf_tool *tool,
 	return process(tool, &ev, NULL, NULL);
 }
 
+#define PERF_ITRACE_DEFAULT_PERIOD_TYPE		PERF_ITRACE_PERIOD_NANOSECS
+#define PERF_ITRACE_DEFAULT_PERIOD		100000
+#define PERF_ITRACE_DEFAULT_CALLCHAIN_SZ	16
+#define PERF_ITRACE_MAX_CALLCHAIN_SZ		1024
+
+void itrace_synth_opts__set_default(struct itrace_synth_opts *synth_opts)
+{
+	synth_opts->instructions = true;
+	synth_opts->branches = true;
+	synth_opts->errors = true;
+	synth_opts->period_type = PERF_ITRACE_DEFAULT_PERIOD_TYPE;
+	synth_opts->period = PERF_ITRACE_DEFAULT_PERIOD;
+	synth_opts->callchain_sz = PERF_ITRACE_DEFAULT_CALLCHAIN_SZ;
+}
+
+int itrace_parse_synth_opts(const struct option *opt, const char *str,
+			    int unset)
+{
+	struct itrace_synth_opts *synth_opts = opt->value;
+	const char *p;
+	char *endptr;
+
+	synth_opts->set = true;
+
+	if (unset) {
+		synth_opts->dont_decode = true;
+		return 0;
+	}
+
+	if (!str) {
+		itrace_synth_opts__set_default(synth_opts);
+		return 0;
+	}
+
+	for (p = str; *p;) {
+		switch (*p++) {
+		case 'i':
+			synth_opts->instructions = true;
+			while (*p == ' ' || *p == ',')
+				p += 1;
+			if (isdigit(*p)) {
+				synth_opts->period = strtoull(p, &endptr, 10);
+				p = endptr;
+				while (*p == ' ' || *p == ',')
+					p += 1;
+				switch (*p++) {
+				case 'i':
+					synth_opts->period_type =
+						PERF_ITRACE_PERIOD_INSTRUCTIONS;
+					break;
+				case 't':
+					synth_opts->period_type =
+						PERF_ITRACE_PERIOD_TICKS;
+					break;
+				case 'm':
+					synth_opts->period *= 1000;
+					/* Fall through */
+				case 'u':
+					synth_opts->period *= 1000;
+					/* Fall through */
+				case 'n':
+					if (*p++ != 's')
+						goto out_err;
+					synth_opts->period_type =
+						PERF_ITRACE_PERIOD_NANOSECS;
+					break;
+				case '\0':
+					goto out;
+				default:
+					goto out_err;
+				}
+			}
+			break;
+		case 'b':
+			synth_opts->branches = true;
+			break;
+		case 'e':
+			synth_opts->errors = true;
+			break;
+		case 'd':
+			synth_opts->log = true;
+			break;
+		case 'c':
+			synth_opts->branches = true;
+			synth_opts->calls = true;
+			break;
+		case 'r':
+			synth_opts->branches = true;
+			synth_opts->returns = true;
+			break;
+		case 'g':
+			synth_opts->instructions = true;
+			synth_opts->callchain = true;
+			synth_opts->callchain_sz =
+					PERF_ITRACE_DEFAULT_CALLCHAIN_SZ;
+			while (*p == ' ' || *p == ',')
+				p += 1;
+			if (isdigit(*p)) {
+				unsigned int val;
+
+				val = strtoul(p, &endptr, 10);
+				p = endptr;
+				if (!val || val > PERF_ITRACE_MAX_CALLCHAIN_SZ)
+					goto out_err;
+				synth_opts->callchain_sz = val;
+			}
+			break;
+		case ' ':
+		case ',':
+			break;
+		default:
+			goto out_err;
+		}
+	}
+out:
+	if (synth_opts->instructions) {
+		if (!synth_opts->period_type)
+			synth_opts->period_type =
+					PERF_ITRACE_DEFAULT_PERIOD_TYPE;
+		if (!synth_opts->period)
+			synth_opts->period = PERF_ITRACE_DEFAULT_PERIOD;
+	}
+
+	return 0;
+
+out_err:
+	pr_err("Bad instruction trace options '%s'\n", str);
+	return -EINVAL;
+}
+
 int itrace_mmap__read(struct itrace_mmap *mm, struct itrace_record *itr,
 		      struct perf_tool *tool, process_itrace_t fn)
 {

58 lines not on this page · raw message

index 6f8ac2f..ccf6a75 100644
@@ -16,12 +16,14 @@ struct ip_callchain;
 struct thread;
 
 struct itrace;
+struct itrace_synth_opts;
 
 struct perf_session {
 	struct perf_header	header;
 	struct machines		machines;
 	struct perf_evlist	*evlist;
 	struct itrace		*itrace;
+	struct itrace_synth_opts *itrace_synth_opts;
 	struct trace_event	tevent;
 	struct events_stats	stats;
 	bool			repipe;

index 2a3a7ce..b063c26 100644
@@ -22,6 +22,7 @@
 #include <linux/types.h>
 
 #include <stdlib.h>
+#include <stdio.h>
 #include <string.h>
 #include <errno.h>
 

@@ -33,6 +34,7 @@
 #include "itrace.h"
 
 #include "event.h"
+#include "session.h"
 #include "debug.h"
 #include "parse-options.h"
 

@@ -157,6 +159,28 @@ struct itrace_record * __weak itrace_record__init(int *err)
 	return NULL;
 }
 
+void itrace_synth_error(struct itrace_error_event *itrace_error, int type,
+			int code, int cpu, pid_t pid, pid_t tid, u64 ip,
+			const char *msg)
+{
+	size_t size;
+
+	memset(itrace_error, 0, sizeof(struct itrace_error_event));
+
+	itrace_error->header.type = PERF_RECORD_ITRACE_ERROR;
+	itrace_error->type = type;
+	itrace_error->code = code;
+	itrace_error->cpu = cpu;
+	itrace_error->pid = pid;
+	itrace_error->tid = tid;
+	itrace_error->ip = ip;
+	strlcpy(itrace_error->msg, msg, MAX_ITRACE_ERROR_MSG);
+
+	size = (void *)itrace_error->msg - (void *)itrace_error +
+	       strlen(itrace_error->msg) + 1;
+	itrace_error->header.size = PERF_ALIGN(size, sizeof(u64));
+}
+
 int perf_event__synthesize_itrace_info(struct itrace_record *itr,
 				       struct perf_tool *tool,
 				       struct perf_session *session,

@@ -336,6 +360,37 @@ out_err:
 	return -EINVAL;
 }
 
+size_t perf_event__fprintf_itrace_error(union perf_event *event, FILE *fp)
+{
+	struct itrace_error_event *e = &event->itrace_error;
+	int ret;
+
+	ret = fprintf(fp, " Instruction trace error type %u", e->type);
+	ret += fprintf(fp, " cpu %d pid %d tid %d ip %#"PRIx64" code %u: %s\n",
+		       e->cpu, e->pid, e->tid, e->ip, e->code, e->msg);
+	return ret;
+}
+
+int perf_event__process_itrace_error(struct perf_tool *tool __maybe_unused,
+				     union perf_event *event,
+				     struct perf_session *session)
+{
+	if (session->itrace)
+		session->itrace->error_count += 1;
+
+	perf_event__fprintf_itrace_error(event, stdout);
+	return 0;
+}
+
+int perf_event__count_itrace_error(struct perf_tool *tool __maybe_unused,
+				   union perf_event *event __maybe_unused,
+				   struct perf_session *session)
+{
+	if (session->itrace)
+		session->itrace->error_count += 1;
+	return 0;
+}
+
 int itrace_mmap__read(struct itrace_mmap *mm, struct itrace_record *itr,
 		      struct perf_tool *tool, process_itrace_t fn)
 {

index 76e70ea..8275afd 100644
@@ -33,6 +33,10 @@ struct option;
 struct record_opts;
 struct itrace_info_event;
 
+enum itrace_error_type {
+	PERF_ITRACE_DECODER_ERROR = 1,
+};
+
 enum itrace_period_type {
 	PERF_ITRACE_PERIOD_INSTRUCTIONS,
 	PERF_ITRACE_PERIOD_TICKS,

@@ -202,6 +206,10 @@ int itrace_record__info_fill(struct itrace_record *itr,
 void itrace_record__free(struct itrace_record *itr);
 u64 itrace_record__reference(struct itrace_record *itr);
 
+void itrace_synth_error(struct itrace_error_event *itrace_error, int type,
+			int code, int cpu, pid_t pid, pid_t tid, u64 ip,
+			const char *msg);
+
 int perf_event__synthesize_itrace_info(struct itrace_record *itr,
 				       struct perf_tool *tool,
 				       struct perf_session *session,

@@ -210,10 +218,18 @@ int perf_event__synthesize_itrace(struct perf_tool *tool,
 				  perf_event__handler_t process,
 				  size_t size, u64 offset, u64 ref, int idx,
 				  u32 tid, u32 cpu);
+int perf_event__process_itrace_error(struct perf_tool *tool,
+				     union perf_event *event,
+				     struct perf_session *session);
+int perf_event__count_itrace_error(struct perf_tool *tool __maybe_unused,
+				   union perf_event *event __maybe_unused,
+				   struct perf_session *session);
 int itrace_parse_synth_opts(const struct option *opt, const char *str,
 			    int unset);
 void itrace_synth_opts__set_default(struct itrace_synth_opts *synth_opts);
 
+size_t perf_event__fprintf_itrace_error(union perf_event *event, FILE *fp);
+
 static inline int itrace__process_event(struct perf_session *session,
 					union perf_event *event,
 					struct perf_sample *sample,

index b063c26..c67bf10 100644
@@ -20,11 +20,15 @@
 #include <linux/kernel.h>
 #include <linux/perf_event.h>
 #include <linux/types.h>
+#include <linux/string.h>
 
+#include <sys/param.h>
 #include <stdlib.h>
 #include <stdio.h>
 #include <string.h>
+#include <limits.h>
 #include <errno.h>
+#include <linux/list.h>
 
 #include "../perf.h"
 #include "util.h"

@@ -108,6 +112,241 @@ void itrace_mmap_params__set_idx(struct itrace_mmap_params *mp,
 	}
 }
 
+#define ITRACE_INIT_NR_QUEUES	32
+
+static struct itrace_queue *itrace_alloc_queue_array(unsigned int nr_queues)
+{
+	struct itrace_queue *queue_array;
+	unsigned int max_nr_queues, i;
+
+	max_nr_queues = MIN(UINT_MAX, SIZE_MAX) / sizeof(struct itrace_queue);
+	if (nr_queues > max_nr_queues)
+		return NULL;
+
+	queue_array = calloc(nr_queues, sizeof(struct itrace_queue));
+	if (!queue_array)
+		return NULL;
+
+	for (i = 0; i < nr_queues; i++) {
+		INIT_LIST_HEAD(&queue_array[i].head);
+		queue_array[i].priv = NULL;
+	}
+
+	return queue_array;
+}
+
+int itrace_queues__init(struct itrace_queues *queues)
+{
+	queues->nr_queues = ITRACE_INIT_NR_QUEUES;
+	queues->queue_array = itrace_alloc_queue_array(queues->nr_queues);
+	if (!queues->queue_array)
+		return -ENOMEM;
+	return 0;
+}
+
+static int itrace_queues__grow(struct itrace_queues *queues,
+			       unsigned int new_nr_queues)
+{
+	unsigned int nr_queues = queues->nr_queues;
+	struct itrace_queue *queue_array;
+	unsigned int i;
+
+	if (!nr_queues)
+		nr_queues = ITRACE_INIT_NR_QUEUES;
+
+	while (nr_queues && nr_queues < new_nr_queues)
+		nr_queues <<= 1;
+
+	if (nr_queues < queues->nr_queues || nr_queues < new_nr_queues)
+		return -EINVAL;
+
+	queue_array = itrace_alloc_queue_array(nr_queues);
+	if (!queue_array)
+		return -ENOMEM;
+
+	for (i = 0; i < queues->nr_queues; i++) {
+		list_splice_tail(&queues->queue_array[i].head,
+				 &queue_array[i].head);
+		queue_array[i].priv = queues->queue_array[i].priv;
+	}
+
+	queues->nr_queues = nr_queues;
+	queues->queue_array = queue_array;
+
+	return 0;
+}
+
+static void *itrace_copy_data(u64 size, struct perf_session *session)
+{
+	int fd = perf_data_file__fd(session->file);
+	void *p;
+	ssize_t ret;
+
+	if (size > SSIZE_MAX)
+		return NULL;
+
+	p = malloc(size);
+	if (!p)
+		return NULL;
+
+	ret = readn(fd, p, size);
+	if (ret != (ssize_t)size) {
+		free(p);
+		return NULL;
+	}
+
+	return p;
+}
+
+static int itrace_queues__add_buffer(struct itrace_queues *queues,
+				     unsigned int idx,
+				     struct itrace_buffer *buffer)
+{
+	struct itrace_queue *queue;
+	int err;
+
+	if (idx >= queues->nr_queues) {
+		err = itrace_queues__grow(queues, idx + 1);
+		if (err)
+			return err;
+	}
+
+	queue = &queues->queue_array[idx];
+
+	if (!queue->set) {
+		queue->set = true;
+		queue->tid = buffer->tid;
+		queue->cpu = buffer->cpu;
+	} else if (buffer->cpu != queue->cpu || buffer->tid != queue->tid) {
+		pr_err("itrace queue conflict: cpu %d, tid %d vs cpu %d, tid %d\n",
+		       queue->cpu, queue->tid, buffer->cpu, buffer->tid);
+		return -EINVAL;
+	}
+
+	buffer->buffer_nr = queues->next_buffer_nr++;
+
+	list_add_tail(&buffer->list, &queue->head);
+
+	queues->new_data = true;
+	queues->populated = true;
+
+	return 0;
+}
+
+/* Limit buffers to 32MiB on 32-bit */
+#define BUFFER_LIMIT_FOR_32_BIT (32 * 1024 * 1024)
+
+static int itrace_queues__split_buffer(struct itrace_queues *queues,
+				       unsigned int idx,
+				       struct itrace_buffer *buffer)
+{
+	u64 sz = buffer->size;
+	bool consecutive = false;
+	struct itrace_buffer *b;
+	int err;
+
+	while (sz > BUFFER_LIMIT_FOR_32_BIT) {
+		b = memdup(buffer, sizeof(struct itrace_buffer));
+		if (!b)
+			return -ENOMEM;
+		b->size = BUFFER_LIMIT_FOR_32_BIT;
+		b->consecutive = consecutive;
+		err = itrace_queues__add_buffer(queues, idx, b);
+		if (err) {
+			itrace_buffer__free(b);
+			return err;
+		}
+		buffer->data_offset += BUFFER_LIMIT_FOR_32_BIT;
+		sz -= BUFFER_LIMIT_FOR_32_BIT;
+		consecutive = true;
+	}
+
+	buffer->size = sz;
+	buffer->consecutive = consecutive;
+
+	return 0;
+}
+
+static int itrace_queues__add_event_buffer(struct itrace_queues *queues,
+					   struct perf_session *session,
+					   unsigned int idx,
+					   struct itrace_buffer *buffer)
+{
+	if (session->one_mmap) {
+		buffer->data = buffer->data_offset - session->one_mmap_offset +
+			       session->one_mmap_addr;
+	} else if (perf_data_file__is_pipe(session->file)) {
+		buffer->data = itrace_copy_data(buffer->size, session);
+		if (!buffer->data)
+			return -ENOMEM;
+		buffer->data_needs_freeing = true;
+	} else if (BITS_PER_LONG == 32 &&
+		   buffer->size > BUFFER_LIMIT_FOR_32_BIT) {
+		int err;
+
+		err = itrace_queues__split_buffer(queues, idx, buffer);
+		if (err)
+			return err;
+	}
+
+	return itrace_queues__add_buffer(queues, idx, buffer);
+}
+
+int itrace_queues__add_event(struct itrace_queues *queues,
+			     struct perf_session *session,
+			     union perf_event *event, off_t data_offset,
+			     struct itrace_buffer **buffer_ptr)
+{
+	struct itrace_buffer *buffer;
+	unsigned int idx;
+	int err;
+
+	buffer = zalloc(sizeof(struct itrace_buffer));
+	if (!buffer)
+		return -ENOMEM;
+
+	buffer->pid = -1;
+	buffer->tid = event->itrace.tid;
+	buffer->cpu = event->itrace.cpu;
+	buffer->data_offset = data_offset;
+	buffer->offset = event->itrace.offset;
+	buffer->reference = event->itrace.reference;
+	buffer->size = event->itrace.size;
+	idx = event->itrace.idx;
+
+	err = itrace_queues__add_event_buffer(queues, session, idx, buffer);
+	if (err)
+		goto out_err;
+
+	if (buffer_ptr)
+		*buffer_ptr = buffer;
+
+	return 0;
+
+out_err:
+	itrace_buffer__free(buffer);
+	return err;
+}
+
+void itrace_queues__free(struct itrace_queues *queues)
+{
+	unsigned int i;
+
+	for (i = 0; i < queues->nr_queues; i++) {
+		while (!list_empty(&queues->queue_array[i].head)) {
+			struct itrace_buffer *buffer;
+
+			buffer = list_entry(queues->queue_array[i].head.next,
+					    struct itrace_buffer, list);
+			list_del(&buffer->list);
+			itrace_buffer__free(buffer);
+		}
+	}
+
+	zfree(&queues->queue_array);
+	queues->nr_queues = 0;
+}
+
 size_t itrace_record__info_priv_size(struct itrace_record *itr)
 {
 	if (itr)

@@ -159,6 +398,71 @@ struct itrace_record * __weak itrace_record__init(int *err)
 	return NULL;
 }
 
+struct itrace_buffer *itrace_buffer__next(struct itrace_queue *queue,
+					  struct itrace_buffer *buffer)
+{
+	if (buffer) {
+		if (list_is_last(&buffer->list, &queue->head))
+			return NULL;
+		return list_entry(buffer->list.next, struct itrace_buffer,
+				  list);
+	} else {
+		if (list_empty(&queue->head))
+			return NULL;
+		return list_entry(queue->head.next, struct itrace_buffer, list);
+	}
+}
+
+void *itrace_buffer__get_data(struct itrace_buffer *buffer, int fd)
+{
+	size_t adj = buffer->data_offset & (page_size - 1);
+	size_t size = buffer->size + adj;
+	off_t file_offset = buffer->data_offset - adj;
+	void *addr;
+
+	if (buffer->data)
+		return buffer->data;
+
+	addr = mmap(NULL, size, PROT_READ, MAP_SHARED, fd, file_offset);
+	if (addr == MAP_FAILED)
+		return NULL;
+
+	buffer->mmap_addr = addr;
+	buffer->mmap_size = size;
+
+	buffer->data = addr + adj;
+
+	return buffer->data;
+}
+
+void itrace_buffer__put_data(struct itrace_buffer *buffer)
+{
+	if (!buffer->data || !buffer->mmap_addr)
+		return;
+	munmap(buffer->mmap_addr, buffer->mmap_size);
+	buffer->mmap_addr = NULL;
+	buffer->mmap_size = 0;
+	buffer->data = NULL;
+	buffer->use_data = NULL;
+}
+
+void itrace_buffer__drop_data(struct itrace_buffer *buffer)
+{
+	itrace_buffer__put_data(buffer);
+	if (buffer->data_needs_freeing) {
+		buffer->data_needs_freeing = false;
+		zfree(&buffer->data);
+		buffer->use_data = NULL;
+		buffer->size = 0;
+	}
+}
+
+void itrace_buffer__free(struct itrace_buffer *buffer)
+{
+	itrace_buffer__drop_data(buffer);
+	free(buffer);
+}
+
 void itrace_synth_error(struct itrace_error_event *itrace_error, int type,
 			int code, int cpu, pid_t pid, pid_t tid, u64 ip,
 			const char *msg)

105 lines not on this page · raw message

index c67bf10..b50d959 100644
@@ -347,6 +347,91 @@ void itrace_queues__free(struct itrace_queues *queues)
 	queues->nr_queues = 0;
 }
 
+static void itrace_heapify(struct itrace_heap_item *heap_array,
+			   unsigned int pos, unsigned int queue_nr,
+			   u64 ordinal)
+{
+	unsigned int parent;
+
+	while (pos) {
+		parent = (pos - 1) >> 1;
+		if (heap_array[parent].ordinal <= ordinal)
+			break;
+		heap_array[pos] = heap_array[parent];
+		pos = parent;
+	}
+	heap_array[pos].queue_nr = queue_nr;
+	heap_array[pos].ordinal = ordinal;
+}
+
+int itrace_heap__add(struct itrace_heap *heap, unsigned int queue_nr,
+		     u64 ordinal)
+{
+	struct itrace_heap_item *heap_array;
+
+	if (queue_nr >= heap->heap_sz) {
+		unsigned int heap_sz = ITRACE_INIT_NR_QUEUES;
+
+		while (heap_sz <= queue_nr)
+			heap_sz <<= 1;
+		heap_array = realloc(heap->heap_array,
+				     heap_sz * sizeof(struct itrace_heap_item));
+		if (!heap_array)
+			return -ENOMEM;
+		heap->heap_array = heap_array;
+		heap->heap_sz = heap_sz;
+	}
+
+	itrace_heapify(heap->heap_array, heap->heap_cnt++, queue_nr, ordinal);
+
+	return 0;
+}
+
+void itrace_heap__free(struct itrace_heap *heap)
+{
+	zfree(&heap->heap_array);
+	heap->heap_cnt = 0;
+	heap->heap_sz = 0;
+}
+
+void itrace_heap__pop(struct itrace_heap *heap)
+{
+	unsigned int pos, last, heap_cnt = heap->heap_cnt;
+	struct itrace_heap_item *heap_array;
+
+	if (!heap_cnt)
+		return;
+
+	heap->heap_cnt -= 1;
+
+	heap_array = heap->heap_array;
+
+	pos = 0;
+	while (1) {
+		unsigned int left, right;
+
+		left = (pos << 1) + 1;
+		if (left >= heap_cnt)
+			break;
+		right = left + 1;
+		if (right >= heap_cnt) {
+			heap_array[pos] = heap_array[left];
+			return;
+		}
+		if (heap_array[left].ordinal < heap_array[right].ordinal) {
+			heap_array[pos] = heap_array[left];
+			pos = left;
+		} else {
+			heap_array[pos] = heap_array[right];
+			pos = right;
+		}
+	}
+
+	last = heap_cnt - 1;
+	itrace_heapify(heap_array, pos, heap_array[last].queue_nr,
+		       heap_array[last].ordinal);
+}
+
 size_t itrace_record__info_priv_size(struct itrace_record *itr)
 {
 	if (itr)

index c357893..a7d4ad2 100644
@@ -172,6 +172,29 @@ struct itrace_queues {
 };
 
 /**
+ * struct itrace_heap_item - element of struct itrace_heap.
+ * @queue_nr: queue number
+ * @ordinal: value used for sorting (lowest ordinal is top of the heap) expected
+ *           to be a timestamp
+ */
+struct itrace_heap_item {
+	unsigned int		queue_nr;
+	u64			ordinal;
+};
+
+/**
+ * struct itrace_heap - a heap suitable for sorting Instruction Tracing queues.
+ * @heap_array: the heap
+ * @heap_cnt: the number of elements in the heap
+ * @heap_sz: maximum number of elements (grows as needed)
+ */
+struct itrace_heap {
+	struct itrace_heap_item	*heap_array;
+	unsigned int		heap_cnt;
+	unsigned int		heap_sz;
+};
+
+/**
  * struct itrace_mmap - records an mmap of the itrace buffer.
  * @base: address of mapped area
  * @userpg: pointer to buffer's perf_event_mmap_page

@@ -280,6 +303,12 @@ void *itrace_buffer__get_data(struct itrace_buffer *buffer, int fd);
 void itrace_buffer__put_data(struct itrace_buffer *buffer);
 void itrace_buffer__drop_data(struct itrace_buffer *buffer);
 void itrace_buffer__free(struct itrace_buffer *buffer);
+
+int itrace_heap__add(struct itrace_heap *heap, unsigned int queue_nr,
+		     u64 ordinal);
+void itrace_heap__pop(struct itrace_heap *heap);
+void itrace_heap__free(struct itrace_heap *heap);
+
 struct itrace_record *itrace_record__init(int *err);
 
 int itrace_record__options(struct itrace_record *itr,

index b50d959..eb43a7f 100644
@@ -599,6 +599,28 @@ out_free:
 	return err;
 }
 
+static bool itrace__dont_decode(struct perf_session *session)
+{
+	return !session->itrace_synth_opts ||
+	       session->itrace_synth_opts->dont_decode;
+}
+
+int perf_event__process_itrace_info(struct perf_tool *tool __maybe_unused,
+				    union perf_event *event,
+				    struct perf_session *session __maybe_unused)
+{
+	enum itrace_type type = event->itrace_info.type;
+
+	if (dump_trace)
+		fprintf(stdout, " type: %u\n", type);
+
+	switch (type) {
+	case PERF_ITRACE_UNKNOWN:
+	default:
+		return -EINVAL;
+	}
+}
+
 int perf_event__synthesize_itrace(struct perf_tool *tool,
 				  perf_event__handler_t process,
 				  size_t size, u64 offset, u64 ref, int idx,

@@ -619,6 +641,30 @@ int perf_event__synthesize_itrace(struct perf_tool *tool,
 	return process(tool, &ev, NULL, NULL);
 }
 
+s64 perf_event__process_itrace(struct perf_tool *tool, union perf_event *event,
+			       struct perf_session *session)
+{
+	s64 err;
+
+	if (dump_trace)
+		fprintf(stdout, " size: %#"PRIx64"  offset: %#"PRIx64"  ref: %#"PRIx64"  idx: %u  tid: %d  cpu: %d\n",
+			event->itrace.size, event->itrace.offset,
+			event->itrace.reference, event->itrace.idx,
+			event->itrace.tid, event->itrace.cpu);
+
+	if (itrace__dont_decode(session))
+		return event->itrace.size;
+
+	if (!session->itrace || event->header.type != PERF_RECORD_ITRACE)
+		return -EINVAL;
+
+	err = session->itrace->process_itrace_event(session, event, tool);
+	if (err < 0)
+		return err;
+
+	return event->itrace.size;
+}
+
 #define PERF_ITRACE_DEFAULT_PERIOD_TYPE		PERF_ITRACE_PERIOD_NANOSECS
 #define PERF_ITRACE_DEFAULT_PERIOD		100000
 #define PERF_ITRACE_DEFAULT_CALLCHAIN_SZ	16

@@ -764,6 +810,9 @@ int perf_event__process_itrace_error(struct perf_tool *tool __maybe_unused,
 				     union perf_event *event,
 				     struct perf_session *session)
 {
+	if (itrace__dont_decode(session))
+		return 0;
+
 	if (session->itrace)
 		session->itrace->error_count += 1;
 

index a7d4ad2..575037c 100644
@@ -34,6 +34,10 @@ struct option;
 struct record_opts;
 struct itrace_info_event;
 
+enum itrace_type {
+	PERF_ITRACE_UNKNOWN,
+};
+
 enum itrace_error_type {
 	PERF_ITRACE_DECODER_ERROR = 1,
 };

@@ -90,6 +94,9 @@ struct itrace {
 			     union perf_event *event,
 			     struct perf_sample *sample,
 			     struct perf_tool *tool);
+	int (*process_itrace_event)(struct perf_session *session,
+				    union perf_event *event,
+				    struct perf_tool *tool);
 	int (*flush_events)(struct perf_session *session,
 			    struct perf_tool *tool);
 	void (*free_events)(struct perf_session *session);

@@ -330,10 +337,16 @@ int perf_event__synthesize_itrace_info(struct itrace_record *itr,
 				       struct perf_tool *tool,
 				       struct perf_session *session,
 				       perf_event__handler_t process);
+int perf_event__process_itrace_info(struct perf_tool *tool,
+				    union perf_event *event,
+				    struct perf_session *session);
 int perf_event__synthesize_itrace(struct perf_tool *tool,
 				  perf_event__handler_t process,
 				  size_t size, u64 offset, u64 ref, int idx,
 				  u32 tid, u32 cpu);
+s64 perf_event__process_itrace(struct perf_tool *tool,
+			       union perf_event *event,
+			       struct perf_session *session);
 int perf_event__process_itrace_error(struct perf_tool *tool,
 				     union perf_event *event,
 				     struct perf_session *session);

index eb43a7f..fd67a9a 100644
@@ -37,6 +37,8 @@
 #include "thread_map.h"
 #include "itrace.h"
 
+#include <linux/hash.h>
+
 #include "event.h"
 #include "session.h"
 #include "debug.h"

@@ -916,3 +918,124 @@ int itrace_mmap__read(struct itrace_mmap *mm, struct itrace_record *itr,
 
 	return 1;
 }
+
+/**
+ * struct itrace_cache - hash table to cache decoded instruction blocks
+ * @hashtable: the hashtable
+ * @sz: hashtable size (number of hlists)
+ * @entry_size: size of an entry
+ * @limit: limit the number of entries to this maximum, when reached the cache
+ *         is dropped and caching begins again with an empty cache
+ * @cnt: current number of entries
+ * @bits: hashtable size (@sz = 2^@bits)
+ */
+struct itrace_cache {
+	struct hlist_head *hashtable;
+	size_t sz;
+	size_t entry_size;
+	size_t limit;
+	size_t cnt;
+	unsigned int bits;
+};
+
+struct itrace_cache *itrace_cache__new(unsigned int bits, size_t entry_size,
+				       unsigned int limit_percent)
+{
+	struct itrace_cache *c;
+	struct hlist_head *ht;
+	size_t sz, i;
+
+	c = zalloc(sizeof(struct itrace_cache));
+	if (!c)
+		return NULL;
+
+	sz = 1UL << bits;
+
+	ht = calloc(sz, sizeof(struct hlist_head));
+	if (!ht)
+		goto out_free;
+
+	for (i = 0; i < sz; i++)
+		INIT_HLIST_HEAD(&ht[i]);
+
+	c->hashtable = ht;
+	c->sz = sz;
+	c->entry_size = entry_size;
+	c->limit = (c->sz * limit_percent) / 100;
+	c->bits = bits;
+
+	return c;
+
+out_free:
+	free(c);
+	return NULL;
+}
+
+static void itrace_cache__drop(struct itrace_cache *c)
+{
+	struct itrace_cache_entry *entry;
+	struct hlist_node *tmp;
+	size_t i;
+
+	if (!c)
+		return;
+
+	for (i = 0; i < c->sz; i++) {
+		hlist_for_each_entry_safe(entry, tmp, &c->hashtable[i], hash) {
+			hlist_del(&entry->hash);
+			itrace_cache__free_entry(c, entry);
+		}
+	}
+
+	c->cnt = 0;
+}
+
+void itrace_cache__free(struct itrace_cache *c)
+{
+	if (!c)
+		return;
+
+	itrace_cache__drop(c);
+	free(c->hashtable);
+	free(c);
+}
+
+void *itrace_cache__alloc_entry(struct itrace_cache *c)
+{
+	return malloc(c->entry_size);
+}
+
+void itrace_cache__free_entry(struct itrace_cache *c __maybe_unused,
+			      void *entry)
+{
+	free(entry);
+}
+
+int itrace_cache__add(struct itrace_cache *c, u32 key,
+		      struct itrace_cache_entry *entry)
+{
+	if (c->limit && ++c->cnt > c->limit)
+		itrace_cache__drop(c);
+
+	entry->key = key;
+	hlist_add_head(&entry->hash, &c->hashtable[hash_32(key, c->bits)]);
+
+	return 0;
+}
+
+void *itrace_cache__lookup(struct itrace_cache *c, u32 key)
+{
+	struct itrace_cache_entry *entry;
+	struct hlist_head *hlist;
+
+	if (!c)
+		return NULL;
+
+	hlist = &c->hashtable[hash_32(key, c->bits)];
+	hlist_for_each_entry(entry, hlist, hash) {
+		if (entry->key == key)
+			return entry;
+	}
+
+	return NULL;
+}

index 575037c..e67fa34 100644
@@ -316,6 +316,20 @@ int itrace_heap__add(struct itrace_heap *heap, unsigned int queue_nr,
 void itrace_heap__pop(struct itrace_heap *heap);
 void itrace_heap__free(struct itrace_heap *heap);
 
+struct itrace_cache_entry {
+	struct hlist_node hash;
+	u32 key;
+};
+
+struct itrace_cache *itrace_cache__new(unsigned int bits, size_t entry_size,
+				       unsigned int limit_percent);
+void itrace_cache__free(struct itrace_cache *itrace_cache);
+void *itrace_cache__alloc_entry(struct itrace_cache *c);
+void itrace_cache__free_entry(struct itrace_cache *c, void *entry);
+int itrace_cache__add(struct itrace_cache *c, u32 key,
+		      struct itrace_cache_entry *entry);
+void *itrace_cache__lookup(struct itrace_cache *c, u32 key);
+
 struct itrace_record *itrace_record__init(int *err);
 
 int itrace_record__options(struct itrace_record *itr,

index 45be944..73e28ca 100644
@@ -4,6 +4,7 @@
 #include "symbol.h"
 #include "dso.h"
 #include "machine.h"
+#include "itrace.h"
 #include "util.h"
 #include "debug.h"
 

@@ -914,6 +915,7 @@ void dso__delete(struct dso *dso)
 	}
 
 	dso__data_close(dso);
+	itrace_cache__free(dso->itrace_cache);
 	dso_cache__free(&dso->data.cache);
 	dso__free_a2l(dso);
 	zfree(&dso->symsrc_filename);

index 3782c82..c27c8fc 100644
@@ -101,6 +101,8 @@ struct dsos {
 	struct rb_root	 root;	/* rbtree root sorted by long name */
 };
 
+struct itrace_cache;
+
 struct dso {
 	struct list_head node;
 	struct rb_node	 rb_node;	/* rbtree node sorted by long name */

@@ -130,6 +132,7 @@ struct dso {
 	u16		 long_name_len;
 	u16		 short_name_len;
 	void		*dwfl;			/* DWARF debug info */
+	struct itrace_cache *itrace_cache;
 
 	/* dso data file */
 	struct {

index ce304df..dbffd00 100644
@@ -16,6 +16,7 @@
 #include "util/evsel.h"
 #include "util/sort.h"
 #include "util/data.h"
+#include "util/itrace.h"
 #include <linux/bitmap.h>
 
 static char const		*script_name;

@@ -1505,6 +1506,7 @@ int cmd_script(int argc, const char **argv, const char *prefix __maybe_unused)
 	char *rec_script_path = NULL;
 	char *rep_script_path = NULL;
 	struct perf_session *session;
+	struct itrace_synth_opts itrace_synth_opts = {0};
 	char *script_path = NULL;
 	const char **__argv;
 	int i, j, err = 0;

@@ -1519,6 +1521,10 @@ int cmd_script(int argc, const char **argv, const char *prefix __maybe_unused)
 			.attr		 = process_attr,
 			.tracing_data	 = perf_event__process_tracing_data,
 			.build_id	 = perf_event__process_build_id,
+			.id_index	 = perf_event__process_id_index,
+			.itrace_info	 = perf_event__process_itrace_info,
+			.itrace		 = perf_event__process_itrace,
+			.itrace_error	 = perf_event__process_itrace_error,
 			.ordered_events	 = true,
 			.ordering_requires_timestamps = true,
 		},

@@ -1570,6 +1576,9 @@ int cmd_script(int argc, const char **argv, const char *prefix __maybe_unused)
 		    "Show the fork/comm/exit events"),
 	OPT_BOOLEAN('\0', "show-mmap-events", &script.show_mmap_events,
 		    "Show the mmap events"),
+	OPT_CALLBACK_OPTARG('Z', "itrace", &itrace_synth_opts, NULL, "opts",
+			    "Instruction Tracing options",
+			    itrace_parse_synth_opts),
 	OPT_END()
 	};
 	const char * const script_usage[] = {

@@ -1767,6 +1776,8 @@ int cmd_script(int argc, const char **argv, const char *prefix __maybe_unused)
 
 	script.session = session;
 
+	session->itrace_synth_opts = &itrace_synth_opts;
+
 	if (cpu_list) {
 		err = perf_session__cpu_bitmap(session, cpu_list, cpu_bitmap);
 		if (err < 0)

index dbffd00..8976d9b 100644
@@ -147,9 +147,10 @@ static const char *output_field2str(enum perf_output_field field)
 
 #define PRINT_FIELD(x)  (output[attr->type].fields & PERF_OUTPUT_##x)
 
-static int perf_evsel__check_stype(struct perf_evsel *evsel,
-				   u64 sample_type, const char *sample_msg,
-				   enum perf_output_field field)
+static int perf_evsel__do_check_stype(struct perf_evsel *evsel,
+				      u64 sample_type, const char *sample_msg,
+				      enum perf_output_field field,
+				      bool allow_user_set)
 {
 	struct perf_event_attr *attr = &evsel->attr;
 	int type = attr->type;

@@ -159,6 +160,8 @@ static int perf_evsel__check_stype(struct perf_evsel *evsel,
 		return 0;
 
 	if (output[type].user_set) {
+		if (allow_user_set)
+			return 0;
 		evname = perf_evsel__name(evsel);
 		pr_err("Samples for '%s' event do not have %s attribute set. "
 		       "Cannot print '%s' field.\n",

@@ -176,10 +179,21 @@ static int perf_evsel__check_stype(struct perf_evsel *evsel,
 	return 0;
 }
 
+static int perf_evsel__check_stype(struct perf_evsel *evsel,
+				   u64 sample_type, const char *sample_msg,
+				   enum perf_output_field field)
+{
+	return perf_evsel__do_check_stype(evsel, sample_type, sample_msg, field,
+					  false);
+}
+
 static int perf_evsel__check_attr(struct perf_evsel *evsel,
 				  struct perf_session *session)
 {
 	struct perf_event_attr *attr = &evsel->attr;
+	bool allow_user_set;
+
+	allow_user_set = perf_header__has_feat(&session->header, HEADER_ITRACE);
 
 	if (PRINT_FIELD(TRACE) &&
 		!perf_session__has_traces(session, "record -R"))

@@ -192,8 +206,8 @@ static int perf_evsel__check_attr(struct perf_evsel *evsel,
 	}
 
 	if (PRINT_FIELD(ADDR) &&
-		perf_evsel__check_stype(evsel, PERF_SAMPLE_ADDR, "ADDR",
-					PERF_OUTPUT_ADDR))
+		perf_evsel__do_check_stype(evsel, PERF_SAMPLE_ADDR, "ADDR",
+					   PERF_OUTPUT_ADDR, allow_user_set))
 		return -EINVAL;
 
 	if (PRINT_FIELD(SYM) && !PRINT_FIELD(IP) && !PRINT_FIELD(ADDR)) {

@@ -230,8 +244,8 @@ static int perf_evsel__check_attr(struct perf_evsel *evsel,
 		return -EINVAL;
 
 	if (PRINT_FIELD(CPU) &&
-		perf_evsel__check_stype(evsel, PERF_SAMPLE_CPU, "CPU",
-					PERF_OUTPUT_CPU))
+		perf_evsel__do_check_stype(evsel, PERF_SAMPLE_CPU, "CPU",
+					   PERF_OUTPUT_CPU, allow_user_set))
 		return -EINVAL;
 
 	if (PRINT_FIELD(PERIOD) &&

index 140a6cd..1b92c3d 100644
@@ -36,6 +36,8 @@
 #include "util/data.h"
 #include "arch/common.h"
 
+#include "util/itrace.h"
+
 #include <dlfcn.h>
 #include <linux/bitmap.h>
 

@@ -572,6 +574,7 @@ parse_percent_limit(const struct option *opt, const char *str,
 int cmd_report(int argc, const char **argv, const char *prefix __maybe_unused)
 {
 	struct perf_session *session;
+	struct itrace_synth_opts itrace_synth_opts = {0};
 	struct stat st;
 	bool has_br_stack = false;
 	int branch_mode = -1;

@@ -593,6 +596,10 @@ int cmd_report(int argc, const char **argv, const char *prefix __maybe_unused)
 			.attr		 = perf_event__process_attr,
 			.tracing_data	 = perf_event__process_tracing_data,
 			.build_id	 = perf_event__process_build_id,
+			.id_index	 = perf_event__process_id_index,
+			.itrace_info	 = perf_event__process_itrace_info,
+			.itrace		 = perf_event__process_itrace,
+			.itrace_error	 = perf_event__count_itrace_error,
 			.ordered_events	 = true,
 			.ordering_requires_timestamps = true,
 		},

@@ -696,6 +703,9 @@ int cmd_report(int argc, const char **argv, const char *prefix __maybe_unused)
 		     "Don't show entries under that percent", parse_percent_limit),
 	OPT_CALLBACK(0, "percentage", NULL, "relative|absolute",
 		     "how to display percentage of filtered entries", parse_filter_percentage),
+	OPT_CALLBACK_OPTARG('Z', "itrace", &itrace_synth_opts, NULL, "opts",
+			    "Instruction Tracing options",
+			    itrace_parse_synth_opts),
 	OPT_END()
 	};
 	struct perf_data_file file = {

@@ -740,6 +750,8 @@ repeat:
 					       report.queue_size);
 	}
 
+	session->itrace_synth_opts = &itrace_synth_opts;
+
 	report.session = session;
 
 	has_br_stack = perf_header__has_feat(&session->header,

index 84df2de..29e0840 100644
@@ -38,14 +38,11 @@ struct event_entry {
 	union perf_event event[0];
 };
 
-static int perf_event__repipe_synth(struct perf_tool *tool,
-				    union perf_event *event)
+static int output_bytes(struct perf_inject *inject, void *buf, size_t sz)
 {
-	struct perf_inject *inject = container_of(tool, struct perf_inject, tool);
 	ssize_t size;
 
-	size = perf_data_file__write(&inject->output, event,
-				     event->header.size);
+	size = perf_data_file__write(&inject->output, buf, sz);
 	if (size < 0)
 		return -errno;
 

@@ -53,6 +50,34 @@ static int perf_event__repipe_synth(struct perf_tool *tool,
 	return 0;
 }
 
+static int copy_bytes(struct perf_inject *inject, int fd, off_t size)
+{
+	char buf[4096];
+	ssize_t ssz;
+	int ret;
+
+	while (size > 0) {
+		ssz = read(fd, buf, MIN(size, sizeof(buf)));
+		if (ssz < 0)
+			return -errno;
+		ret = output_bytes(inject, buf, ssz);
+		if (ret)
+			return ret;
+		size -= ssz;
+	}
+
+	return 0;
+}
+
+static int perf_event__repipe_synth(struct perf_tool *tool,
+				    union perf_event *event)
+{
+	struct perf_inject *inject = container_of(tool, struct perf_inject,
+						  tool);
+
+	return output_bytes(inject, event, event->header.size);
+}
+
 static int perf_event__repipe_op2_synth(struct perf_tool *tool,
 					union perf_event *event,
 					struct perf_session *session

@@ -79,6 +104,31 @@ static int perf_event__repipe_attr(struct perf_tool *tool,
 	return perf_event__repipe_synth(tool, event);
 }
 
+static s64 perf_event__repipe_itrace(struct perf_tool *tool,
+				     union perf_event *event,
+				     struct perf_session *session
+				     __maybe_unused)
+{
+	struct perf_inject *inject = container_of(tool, struct perf_inject,
+						  tool);
+	int ret;
+
+	if (perf_data_file__is_pipe(session->file) || !session->one_mmap) {
+		ret = output_bytes(inject, event, event->header.size);
+		if (ret < 0)
+			return ret;
+		ret = copy_bytes(inject, perf_data_file__fd(session->file),
+				 event->itrace.size);
+	} else {
+		ret = output_bytes(inject, event,
+				   event->header.size + event->itrace.size);
+	}
+	if (ret < 0)
+		return ret;
+
+	return event->itrace.size;
+}
+
 static int perf_event__repipe(struct perf_tool *tool,
 			      union perf_event *event,
 			      struct perf_sample *sample __maybe_unused,

@@ -407,6 +457,9 @@ int cmd_inject(int argc, const char **argv, const char *prefix __maybe_unused)
 			.unthrottle	= perf_event__repipe,
 			.attr		= perf_event__repipe_attr,
 			.tracing_data	= perf_event__repipe_op2_synth,
+			.itrace_info	= perf_event__repipe_op2_synth,
+			.itrace		= perf_event__repipe_itrace,
+			.itrace_error	= perf_event__repipe_op2_synth,
 			.finished_round	= perf_event__repipe_op2_synth,
 			.build_id	= perf_event__repipe_op2_synth,
 			.id_index	= perf_event__repipe_op2_synth,

131 lines not on this page · raw message

index c3323ea..4d0fe53 100644
@@ -115,6 +115,18 @@ static s64 perf_event__repipe_itrace(struct perf_tool *tool,
 						  tool);
 	int ret;
 
+	if (!inject->output.is_pipe) {
+		off_t offset;
+
+		offset = lseek(inject->output.fd, 0, SEEK_CUR);
+		if (offset == -1)
+			return -errno;
+		ret = itrace_index__itrace_event(&session->itrace_index, event,
+						 offset);
+		if (ret < 0)
+			return ret;
+	}
+
 	if (perf_data_file__is_pipe(session->file) || !session->one_mmap) {
 		ret = output_bytes(inject, event, event->header.size);
 		if (ret < 0)

@@ -481,6 +493,9 @@ static int __cmd_inject(struct perf_inject *inject)
 		output_data_offset = 4096;
 	}
 
+	if (!inject->itrace_synth_opts.set)
+		itrace_index__free(&session->itrace_index);
+
 	if (!file_out->is_pipe)
 		lseek(file_out->fd, output_data_offset, SEEK_SET);
 

index 1e65d38..62211b9 100644
@@ -117,9 +117,24 @@ static int record__process_itrace(struct perf_tool *tool,
 				  size_t len1, void *data2, size_t len2)
 {
 	struct record *rec = container_of(tool, struct record, tool);
+	struct perf_data_file *file = &rec->file;
 	size_t padding;
 	u8 pad[8] = {0};
 
+	if (!perf_data_file__is_pipe(file)) {
+		off_t file_offset;
+		int fd = perf_data_file__fd(file);
+		int err;
+
+		file_offset = lseek(fd, 0, SEEK_CUR);
+		if (file_offset == -1)
+			return -1;
+		err = itrace_index__itrace_event(&rec->session->itrace_index,
+						 event, file_offset);
+		if (err)
+			return err;
+	}
+
 	/* event.itrace.size includes padding, see __itrace_mmap__read() */
 	padding = (len1 + len2) & 7;
 	if (padding)

index 27f8fb5..b64a746 100644
@@ -869,11 +869,18 @@ static int write_branch_stack(int fd __maybe_unused,
 	return 0;
 }
 
-static int write_itrace(int fd __maybe_unused,
-			struct perf_header *h __maybe_unused,
+static int write_itrace(int fd, struct perf_header *h,
 			struct perf_evlist *evlist __maybe_unused)
 {
-	return 0;
+	struct perf_session *session;
+	int err;
+
+	session = container_of(h, struct perf_session, header);
+
+	err = itrace_index__write(fd, &session->itrace_index);
+	if (err < 0)
+		pr_err("Failed to write itrace index\n");
+	return err;
 }
 
 static void print_hostname(struct perf_header *ph, int fd __maybe_unused,

@@ -1846,6 +1853,21 @@ out_free:
 	return ret;
 }
 
+static int process_itrace(struct perf_file_section *section,
+			  struct perf_header *ph, int fd,
+			  void *data __maybe_unused)
+{
+	struct perf_session *session;
+	int err;
+
+	session = container_of(ph, struct perf_session, header);
+
+	err = itrace_index__process(fd, section->size, session, ph->needs_swap);
+	if (err < 0)
+		pr_err("Failed to process itrace index\n");
+	return err;
+}
+
 struct feature_ops {
 	int (*write)(int fd, struct perf_header *h, struct perf_evlist *evlist);
 	void (*print)(struct perf_header *h, int fd, FILE *fp);

@@ -1886,7 +1908,7 @@ static const struct feature_ops feat_ops[HEADER_LAST_FEATURE] = {
 	FEAT_OPA(HEADER_BRANCH_STACK,	branch_stack),
 	FEAT_OPP(HEADER_PMU_MAPPINGS,	pmu_mappings),
 	FEAT_OPP(HEADER_GROUP_DESC,	group_desc),
-	FEAT_OPA(HEADER_ITRACE,		itrace),
+	FEAT_OPP(HEADER_ITRACE,		itrace),
 };
 
 struct header_print_data {

226 lines not on this page · raw message

index e67fa34..02b6bed 100644
@@ -82,6 +82,32 @@ struct itrace_synth_opts {
 };
 
 /**
+ * struct itrace_index_entry - indexes a Instruction Tracing event within a
+ *                             perf.data file.
+ * @file_offset: offset within the perf.data file
+ * @sz: size of the event
+ */
+struct itrace_index_entry {
+	u64			file_offset;
+	u64			sz;
+};
+
+#define PERF_ITRACE_INDEX_ENTRY_COUNT 256
+
+/**
+ * struct itrace_index - index of Instruction Tracing events within a perf.data
+ *                       file.
+ * @list: linking a number of arrays of entries
+ * @nr: number of entries
+ * @entries: array of entries
+ */
+struct itrace_index {
+	struct list_head	list;
+	size_t			nr;
+	struct itrace_index_entry entries[PERF_ITRACE_INDEX_ENTRY_COUNT];
+};
+
+/**
  * struct itrace - session callbacks to allow Instruction Trace data decoding.
  * @process_event: lets the decoder see all session events
  * @flush_events: process any remaining data

@@ -304,6 +330,8 @@ int itrace_queues__add_event(struct itrace_queues *queues,
 			     union perf_event *event, off_t data_offset,
 			     struct itrace_buffer **buffer_ptr);
 void itrace_queues__free(struct itrace_queues *queues);
+int itrace_queues__process_index(struct itrace_queues *queues,
+				 struct perf_session *session);
 struct itrace_buffer *itrace_buffer__next(struct itrace_queue *queue,
 					  struct itrace_buffer *buffer);
 void *itrace_buffer__get_data(struct itrace_buffer *buffer, int fd);

@@ -343,6 +371,13 @@ int itrace_record__info_fill(struct itrace_record *itr,
 void itrace_record__free(struct itrace_record *itr);
 u64 itrace_record__reference(struct itrace_record *itr);
 
+int itrace_index__itrace_event(struct list_head *head, union perf_event *event,
+			       off_t file_offset);
+int itrace_index__write(int fd, struct list_head *head);
+int itrace_index__process(int fd, u64 size, struct perf_session *session,
+			  bool needs_swap);
+void itrace_index__free(struct list_head *head);
+
 void itrace_synth_error(struct itrace_error_event *itrace_error, int type,
 			int code, int cpu, pid_t pid, pid_t tid, u64 ip,
 			const char *msg);

index 1f03418..0af02a4 100644
@@ -97,6 +97,7 @@ struct perf_session *perf_session__new(struct perf_data_file *file,
 
 	session->repipe = repipe;
 	ordered_events__init(&session->ordered_events);
+	INIT_LIST_HEAD(&session->itrace_index);
 	machines__init(&session->machines);
 
 	if (file) {

@@ -168,6 +169,7 @@ static void perf_session_env__delete(struct perf_session_env *env)
 void perf_session__delete(struct perf_session *session)
 {
 	itrace__free(session);
+	itrace_index__free(&session->itrace_index);
 	perf_session__destroy_kernel_maps(session);
 	perf_session__delete_dead_threads(session);
 	perf_session__delete_threads(session);

index ccf6a75..01c4aa4 100644
@@ -24,6 +24,7 @@ struct perf_session {
 	struct perf_evlist	*evlist;
 	struct itrace		*itrace;
 	struct itrace_synth_opts *itrace_synth_opts;
+	struct list_head	itrace_index;
 	struct trace_event	tevent;
 	struct events_stats	stats;
 	bool			repipe;

index ed3873b..d694309 100644
@@ -69,6 +69,15 @@ static int perf_session__list_build_ids(bool force, bool with_hits)
 	session = perf_session__new(&file, false, &build_id__mark_dso_hit_ops);
 	if (session == NULL)
 		return -1;
+
+	/*
+	 * We take all buildids when the file contains Instruction Tracing data
+	 * because we do not decode the trace because it would take too long.
+	 */
+	if (!perf_data_file__is_pipe(&file) &&
+	    perf_header__has_feat(&session->header, HEADER_ITRACE))
+		with_hits = false;
+
 	/*
 	 * in pipe-mode, the only way to get the buildids is to parse
 	 * the record stream. Buildids are stored as RECORD_HEADER_BUILD_ID

index 4d0fe53..285f159 100644
@@ -27,6 +27,7 @@ struct perf_inject {
 	struct perf_session	*session;
 	bool			build_ids;
 	bool			sched_stat;
+	bool			have_itrace;
 	const char		*input_name;
 	struct perf_data_file	output;
 	u64			bytes_written;

@@ -115,6 +116,8 @@ static s64 perf_event__repipe_itrace(struct perf_tool *tool,
 						  tool);
 	int ret;
 
+	inject->have_itrace = true;
+
 	if (!inject->output.is_pipe) {
 		off_t offset;
 

@@ -502,9 +505,12 @@ static int __cmd_inject(struct perf_inject *inject)
 	ret = perf_session__process_events(session, &inject->tool);
 
 	if (!file_out->is_pipe) {
-		if (inject->build_ids)
+		if (inject->build_ids) {
 			perf_header__set_feat(&session->header,
 					      HEADER_BUILD_ID);
+			if (inject->have_itrace)
+				dsos__hit_all(session);
+		}
 		/*
 		 * The instruction traces have been removed and replaced with
 		 * synthesized hardware events, so clear the feature flag.

index 62211b9..475082f 100644
@@ -613,8 +613,16 @@ out_child:
 	if (!err && !file->is_pipe) {
 		rec->session->header.data_size += rec->bytes_written;
 
-		if (!rec->no_buildid)
+		if (!rec->no_buildid) {
 			process_buildids(rec);
+			/*
+			 * We take all buildids when the file contains
+			 * Instruction Tracing data because we do not decode the
+			 * trace because it would take too long.
+			 */
+			if (rec->opts.full_itrace)
+				dsos__hit_all(rec->session);
+		}
 		perf_session__write_header(rec->session, rec->evlist,
 					   file->fd, true);
 	}