Trace messaging device and methods thereof
A method of generating timestamped trace messages includes generating a trace message in response to an event at an instruction pipeline of a data processing device. If timestamping is enabled, timestamps are only included in the trace message only if a programmable condition is detected. For example, a timestamp can be included in the trace message if the amount of space used to store messages at a trace message buffer exceeds a watermark value. The condition that results in a timestamped trace message is programmable, and can be selected via a debug interface. Because timestamps are only included in trace messages when the programmable condition is satisfied, some trace messages will not include a timestamp, thereby reducing the amount of buffer space needed to store the trace messages.
1. A method, comprising:
generating a first trace message at a data processing device in response to a first trace event, comprising:
in response to determining timestamping for a first condition is enabled, generating the first trace message including a first timestamp based on detecting the first condition; and
in response to determining timestamping for a second condition is enabled, generating the first trace message including a second timestamp based on detecting the second condition.
2. The method of claim 1 , wherein the first condition is based on an amount of buffer space used to store trace messages exceeding a threshold.
3. The method of claim 2 , wherein the threshold is a programmable value.
4. The method of claim 1 , wherein the first condition is based on a number of trace messages generated at the data processing device.
5. The method of claim 1 , wherein the first condition is based on occurrence of a first interrupt.
6. The method of claim 1 , wherein the first condition is based on occurrence of a first programmable watchpoint and the second condition is based on occurrence of a second programmable watchpoint different than the first.
7. The method of claim 1 , wherein the first condition is based on determination of a first instruction type at an instruction pipeline of the data processing device.
8. The method of claim 1 , wherein the first condition is based on determination of a first data address at an instruction pipeline of the data processing device.
9. The method of claim 1 , wherein the first condition is based on a change in frequency of a clock signal associated with the data processing device.
10. The method of claim 1 , wherein the first condition is based on a change in a power mode associated with the data processing device.
11. The method of claim 1 , wherein the first condition is based on a condition associated with a first thread executing at the data processing device and the second condition is based on a condition associated with a second thread executing at the data processing device.
12. The method of claim 1 , wherein the first condition comprises a request received at the data processing device from a second device.
13. The method of claim 1 , wherein the first condition is associated with a first transport message stream, and the second condition is associated with a second transport message stream.
14. The method of claim 1 , further comprising:
adjusting the second condition based detecting the first condition.
15. A device, comprising:
an instruction pipeline;
a trace monitoring module connected to the instruction pipeline, the trace monitoring module configured to:
generate a first trace message at a data processing device in response to a first trace event;
in response to determining timestamping for a first condition is enabled, include a first timestamp in the first trace message based on detecting the first condition; and
in response to determining timestamping for a second condition is enabled, include the first timestamp in the first trace message based on detecting the second condition.
16. The device of claim 15 , wherein the trace monitoring module is further configured to:
in response to determining timestamping for the first condition is enabled, not include the first timestamp in the first trace message based on not detecting the first condition.
17. The method of claim 15 , wherein the device further comprises a buffer, and wherein first condition is based on an amount of space available at the buffer to store trace messages exceeding a threshold.