DSP/BIOS execution graph question

Hmm... I'm not too familiar with the execution graph in CCS3, but 'missing data' is definitely an issue for execution graphs in general. 

In both scenarios, if you look at the BIOS events in the event log, do you see any gaps in the sequence numbers? A gap in sequence numbers would indicate that some data has been lost, which could cause some different behavior in the graph...

Chris

i don't see any gaps in the message log.

i tried to activate the option realtime mode under the debug menu in hope to have some exact drawings, but i got many ccs errors.

anyway i'll migrate to ccs 4 soon.

does anyone know if this issue occurs also on ccs 4 ??

Note that BIOS 5 RTA in CCSv4 does not support the execution graph. In CCSv4, the execution graph is only for BIOS 6.

http://processors.wiki.ti.com/index.php/BIOS_5_Real-Time_Analysis_(RTA)_in_CCSv4

Is it possible that halting the target is affecting the target execution? Maybe a timer is still running while the target is halted or something?

Can you tell whether the records in the log (the TSK switch events, etc., not the printf log) match exactly between the two use cases? We should look at the events and see whether the execution graph in the 'halt' case looks correct or not based on the events.

Thanks,

Chris

thanx chris for your replies,

sorry for the late response,i couldn't respond the question since i was constrained in work to remove the CCS 3.3 and BIOS 5.33, and to migrate to CCS 4.2.0 and BIOS 6.3.

i encountered some difficulties doing the realtime analysis. I can't use execution graph anymore, also i can't locate the STS module.

if you have some documents or pointers, please  provide me with them.

thanks again.