Qorivva MPC5xxx/SPC5xx Debugger and NEXUS Trace
55
©1989-2021 Lauterbach GmbH
Tracing of Data (read/write) Transactions
General data tracing is enabled using the command
. This command enables the data trace for
the full address space. The amount of generated trace messages is usually too high to be sent through the
trace port and the on-chip message FIFO will overflow.
The amount of generated trace messages can be reduced by defining address ranges for which data trace
is generated. Up to four address ranges are possible.
Example: Data Trace with Address Range
Use TraceData to limit the data trace to an address range. Depending on the core, 2 or 4 address ranges are
possible. TraceData has no impact on program trace messaging setting.
Another method of reducing trace data is
event-triggered trace filtering
.
Tracing of Context Switches
NEXUS supports two methods of tracing context (= process, task, thread etc) changes. One method is using
the data trace, the other method uses Ownership Trace Messaging (OTM).
In order to have the debugger process and display task information properly, it is required to set up OS
Awareness. If the operating system in use implements the OSEK Runtime Interface (ORTI), which is often
true for this processor series, see
(rtos_orti.pdf) for instructions. For other operating
systems, please check the instructions in the appropriate
Trace Context Switches using Data Trace Messaging (DTM)
If the core implements NEXUS Class 3, the preferred method to trace context switches is to enable data
trace for the memory location that holds the ID of the currently active process. If the OS Awareness is set up,
the address of the memory location can be retrieved using the function
;Enable data trace for read/write accesses to all peripherals
Break.Set 0xC0000000--0xFFFFFFFF /ReadWrite /TraceData
;In addition to full program trace, enable data trace for read accesses
;to the array
flags
NEXUS.BTM ON
Var.Break.Set flags /Read /TraceData
