Enabling No Overhead Profiling - 2021.2 English

When profiling is enabled in xrt.ini with opencl_summary or native_xrt_trace, there is operational overhead added and events in the timeline can show longer delay in between events. However, XRT provides an option to dump just the OpenCL events on the timeline trace without any overhead from profiling. It is a simple method of displaying events on the timeline without any overhead.

Because the goal is to provide visibility into events with absolutely zero overhead, there are limitations to the number of events that can be logged. Additionally, there is no command queue information in this view, so this view is not a replacement of the Timeline Trace; just an alternative view.

You can use the "no overhead" view to confirm OpenCL command dependencies and to observe actual event overhead for the command execution from the host application.

[Debug]
xocl_debug=true
#xocl_event_begin= 0 (default)
#xocl_event_end=1000 (default)

By default only 1000 events can be visualized. OpenCL commands can be visualized on the same timeline. This feature should not be used with opencl_trace enabled, but can be used with device profile enabled.

vp_analyze xocl -i xocl.log // generates debug_log.csv
vp_analyze trace -i debug_log.csv // generates debug_log.wdb
vitis_analyzer debug_log.wdb // loads the wdb file in Vitis analyzer

Figure 1. No Overhead Timeline

Generally "no overhead" profiling should be disabled with regular or low-overhead profiling, as the more detailed profiling will add overhead in any case.