Streaming Trace - Segger J-Link User Manual

14.5 Streaming trace

With introducing streaming trace, some additional concepts needed to be introduced
in order to make real time analysis of the trace data possible. In the following, some
considerations and specifics, that need to be kept in mind when using streaming
trace, are explained.
14.5.1 Download and execution address differ
Analysis of trace data requires that J-Trace needs know which instruction is present
at what address on the target device. As reading from the target memory everytime
is not feasible during live analysis (would lead to a too big performance drop), a copy
of the application contents is cached in the J-Link software at the time, the applica-
tion download is performed. This implies that streaming trace is only possible with
prior download of the application in the same debug session. This also implies that
the execution address needs to be the same as the download address.
In case both addresses differ from each other, the J-Link software needs to be told
that the unknown addresses hold the same data as the cached ones. This is done via
the ReadIntoTraceCache command string. For more information about the syntax
this command string, please refer to Command strings on page 223. For more infor-
mation about how to use command strings in different environments, please refer to
Using command strings on page 241.
14.5.2 Do streaming trace without prior download
Same specifics as for "load and execution address differ" applies. Please refer to
Download and execution address differ on page 381.
J-Link / J-Trace (UM08001) © 2004-2017 SEGGER Microcontroller GmbH & Co. KG
381