On 6/14/2012 12:05 PM, JMGross wrote: > Hi! > No idea how to do it on hardware side. > Of course you can do a break on instruction fetch and check it for a call > instruction, or a RET/RETI instruction (and the return address > is found on stack). But that's rather slow (if not even slower). Not sure if I follow that. With breaking allowed I was able to do it with hacked mspdebug: run, halt, read PC, repeat. The idea is to use EEM trace functionality without stopping the target. If I read TI appnotes correctly that should be possible.
> Alternatively you can put a write access to a global 'trace' variable on > every point you want to trace, and trigger a breakpoint when this > address is written to. Would be better than a breakpoint on a timer > interrupt, and allows tracing ISRs too. Timer interrupt was programmed to randomize PC acquisition time so that aliasing would not upset statistical nature of profiling. > But it is possible that not all MSPs support write-access triggers. The EEM > isn't really good documented in the users guides. No, it is not. I think one needs to sign an NDA to get proper documentation. Well, I was hoping that since TI released MSP430.dll src code it would be possible to figure out how to use EEM trace without having full documentation. Regards, Sergei ------------------------------------------------------------------------------ Live Security Virtual Conference Exclusive live event will cover all the ways today's security and threat landscape has changed and how IT managers can respond. Discussions will include endpoint security, mobile security and the latest in malware threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ _______________________________________________ Mspgcc-users mailing list Mspgcc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/mspgcc-users
