[RFC PATCH 5/5] perf cs-etm: split decode by aux records.

[James Clark]

The trace data between aux records is not continuous, so the decoder
must be reset between each record to ensure that parsing happens
correctly and without any early exits.

Signed-off-by: James Clark <james.clark@xxxxxxx>
---
 tools/perf/util/cs-etm.c | 108 ++++++++++++++++++++++++---------------
 1 file changed, 66 insertions(+), 42 deletions(-)

diff --git a/tools/perf/util/cs-etm.c b/tools/perf/util/cs-etm.c
index 0aaa1f6d2822..b0f464a50e2f 100644
--- a/tools/perf/util/cs-etm.c
+++ b/tools/perf/util/cs-etm.c
@@ -95,6 +95,7 @@ struct cs_etm_queue {
 	int aux_record_list_len;
 	int aux_record_list_idx;
 	struct perf_record_aux *aux_record_list;
+	bool timestamp_found;
 };
 
 /* RB tree for quick conversion between traceID and metadata pointers */
@@ -788,6 +789,9 @@ static int cs_etm__seach_first_timestamp(struct cs_etm_queue *etmq,
 
 	etmq->aux_record_list[etmq->aux_record_list_len++] = *aux_record;
 
+	if (etmq->timestamp_found)
+		return 0;
+
 	/*
 	 * We are under a CPU-wide trace scenario.  As such we need to know
 	 * when the code that generated the traces started to execute so that
@@ -796,56 +800,60 @@ static int cs_etm__seach_first_timestamp(struct cs_etm_queue *etmq,
 	 * timestamp.  The timestamp is then added to the auxtrace min heap
 	 * in order to know what nibble (of all the etmqs) to decode first.
 	 */
-	while (1) {
-		/*
-		 * Fetch an aux_buffer from this etmq.  Bail if no more
-		 * blocks or an error has been encountered.
-		 */
-		ret = cs_etm__get_data_block(etmq);
-		if (ret <= 0)
-			return ret;
-
-		/*
-		 * Run decoder on the trace block.  The decoder will stop when
-		 * encountering a timestamp, a full packet queue or the end of
-		 * trace for that block.
-		 */
-		ret = cs_etm__decode_data_block(etmq);
+	/*
+	 * Fetch an aux_buffer from this etmq.  Bail if no more
+	 * blocks or an error has been encountered.
+	 */
+	if (etmq->aux_record_list[etmq->aux_record_list_idx].aux_size <= 0) {
+		etmq->aux_record_list_idx++;
+		ret = cs_etm_decoder__reset(etmq->decoder);
 		if (ret)
 			return ret;
+	}
+	ret = cs_etm__get_data_block(etmq);
+	if (ret <= 0)
+		return ret;
 
-		/*
-		 * Function cs_etm_decoder__do_{hard|soft}_timestamp() does all
-		 * the timestamp calculation for us.
-		 */
-		timestamp = cs_etm__etmq_get_timestamp(etmq, &trace_chan_id);
+	/*
+	 * Run decoder on the trace block.  The decoder will stop when
+	 * encountering a timestamp, a full packet queue or the end of
+	 * trace for that block.
+	 */
+	ret = cs_etm__decode_data_block(etmq);
+	if (ret)
+		return ret;
 
-		/* We found a timestamp, no need to continue. */
-		if (timestamp)
-			break;
+	/*
+	 * Function cs_etm_decoder__do_{hard|soft}_timestamp() does all
+	 * the timestamp calculation for us.
+	 */
+	timestamp = cs_etm__etmq_get_timestamp(etmq, &trace_chan_id);
 
+	/* We found a timestamp, no need to continue. */
+	if (timestamp) {
 		/*
-		 * We didn't find a timestamp so empty all the traceid packet
-		 * queues before looking for another timestamp packet, either
-		 * in the current data block or a new one.  Packets that were
-		 * just decoded are useless since no timestamp has been
-		 * associated with them.  As such simply discard them.
+		 * We have a timestamp.  Add it to the min heap to reflect when
+		 * instructions conveyed by the range packets of this traceID queue
+		 * started to execute.  Once the same has been done for all the traceID
+		 * queues of each etmq, redenring and decoding can start in
+		 * chronological order.
+		 *
+		 * Note that packets decoded above are still in the traceID's packet
+		 * queue and will be processed in cs_etm__process_queues().
 		 */
-		cs_etm__clear_all_packet_queues(etmq);
+		etmq->timestamp_found = true;
+		cs_queue_nr = TO_CS_QUEUE_NR(etmq->queue_nr, trace_chan_id);
+		return auxtrace_heap__add(&etmq->etm->heap, cs_queue_nr, timestamp);
 	}
-
 	/*
-	 * We have a timestamp.  Add it to the min heap to reflect when
-	 * instructions conveyed by the range packets of this traceID queue
-	 * started to execute.  Once the same has been done for all the traceID
-	 * queues of each etmq, redenring and decoding can start in
-	 * chronological order.
-	 *
-	 * Note that packets decoded above are still in the traceID's packet
-	 * queue and will be processed in cs_etm__process_queues().
+	 * We didn't find a timestamp so empty all the traceid packet
+	 * queues before looking for another timestamp packet, either
+	 * in the current data block or a new one.  Packets that were
+	 * just decoded are useless since no timestamp has been
+	 * associated with them.  As such simply discard them.
 	 */
-	cs_queue_nr = TO_CS_QUEUE_NR(etmq->queue_nr, trace_chan_id);
-	return auxtrace_heap__add(&etmq->etm->heap, cs_queue_nr, timestamp);
+	cs_etm__clear_all_packet_queues(etmq);
+	return 0;
 }
 
 static int cs_etm__setup_queue(struct cs_etm_auxtrace *etm,
@@ -2012,6 +2020,15 @@ static int cs_etm__decode_data_block(struct cs_etm_queue *etmq)
 {
 	int ret = 0;
 	size_t processed = 0;
+	u64 decode_size;
+
+	if (etmq->aux_record_list_idx >= etmq->aux_record_list_len ||
+		etmq->aux_record_list[etmq->aux_record_list_idx].aux_size > etmq->buf_len) {
+		// Assume that aux records always equally divide up the aux buffer
+		// so aux_size should never exceed the remaining buffer to decode.
+		ret = -1;
+		goto out;
+	}
 
 	/*
 	 * Packets are decoded and added to the decoder's packet queue
@@ -2020,10 +2037,11 @@ static int cs_etm__decode_data_block(struct cs_etm_queue *etmq)
 	 * operations that stop processing are a timestamp packet or a full
 	 * decoder buffer queue.
 	 */
+	decode_size = etmq->aux_record_list[etmq->aux_record_list_idx].aux_size;
 	ret = cs_etm_decoder__process_data_block(etmq->decoder,
 						 etmq->offset,
 						 &etmq->buf[etmq->buf_used],
-						 etmq->buf_len,
+						 decode_size,
 						 &processed);
 	if (ret)
 		goto out;
@@ -2031,7 +2049,7 @@ static int cs_etm__decode_data_block(struct cs_etm_queue *etmq)
 	etmq->offset += processed;
 	etmq->buf_used += processed;
 	etmq->buf_len -= processed;
-
+	etmq->aux_record_list[etmq->aux_record_list_idx].aux_size -= processed;
 out:
 	return ret;
 }
@@ -2247,6 +2265,12 @@ static int cs_etm__process_queues(struct cs_etm_auxtrace *etm)
 		 * if need be.
 		 */
 refetch:
+		if (etmq->aux_record_list[etmq->aux_record_list_idx].aux_size <= 0) {
+			etmq->aux_record_list_idx++;
+			ret = cs_etm_decoder__reset(etmq->decoder);
+			if (ret)
+				return ret;
+		}
 		ret = cs_etm__get_data_block(etmq);
 		if (ret < 0)
 			goto out;
-- 
2.28.0