Hello Smitha,
I would advise you to read carefully the documentation provided with the library - particularly the section "OpenCSD Library - Programmers Guide". This explains how the decoder works - the three stage decode process - and the outputs it provides.
The input requirements for decoding are the raw trace stream, the memory image of the program being decoded and the programmed settings of the ETM. The library output for full decode is a list of executed instruction address ranges.
ETM trace is a highly compressed format - a single bit in the output stream can represent a range of many executed instructions. Thus it requires much more time / computing power to decode trace than to generate it. Bear this in mind if you are considering the realtime decode of a trace stream.
I cannot comment on the feasibility of converting the C++ decoder classes into a format suitable for implementing on an FPGA. However, I would say that for a specific application, it may be possible to ignore the usual library API (DeviceTree) and use the C++ classes directly to build a custom decoder using only the most essential components.
Regards
Mike
> -----Original Message-----
> From: CoreSight <
coresight-bounces@lists.linaro.org> On Behalf Of Smitha
> Gautham
> Sent: 04 June 2019 16:56
> To:
coresight@lists.linaro.org> Subject: ETM trace decode using OpenCSD
>
> Hello,
>
> I am a graduate student from Virginia Commonwealth University working on
> execution and data monitoring of my application running on a STM32F4 board.
> I have a runtime monitor on an FPGA that would analyse the trace
> information.
>
> But, to transfer the instruction traces from ETM and data traces to the FPGA,
> I would need a trace decoder on the FPGA. This trace decoder would be help
> decode the ETM traces from STM32 board to the FPGA monitor. I want to do
> monitoring at runtime. I came across the OpenCSD github repository. Do you
> think this could fit well for our application? Can the OpenCSD be
> implemented on an FPGA to decode ETM traces?
>
> Thank you for any inputs on this.
>
> regards,
> Smitha
IMPORTANT NOTICE: The contents of this email and any attachments are confidential and may also be privileged. If you are not the intended recipient, please notify the sender immediately and do not disclose the contents to any other person, use it for any purpose, or store or copy the information in any medium. Thank you.