Analysing PPU programs for stalls

[cashew]

Has anyone had any experience using any simulators to analyse their PPU code for stalls, pipeline or otherwise?

I see there's tools on IBM's alphaworks site which look potentially useful. I'm particularly interesting in analysing the performance of some altivec code that executes on the PPU. A lot of stuff seems geared towards the SPUs.

Failing that, has anyone got a good cheat sheet or tips guide that I could use to inspect my PPU vectorised code for stalls?

Thanks.

[Vitaly Vidmirov]

I think CELL simulator is the only tool that can do this.

CBE programming handbook: Appendix A. PPE Instruction Set and Intrinsics
http://www-01.ibm.com/chips/techlib/techlib.nsf/techdocs/9F820A5FFA3ECE8C8725716A0062585F?Open&S_TACT=105AGX16&S_CMP=LP

some notes about profiling:
http://www-01.ibm.com/chips/techlib/techlib.nsf/techdocs/1DAAA0A3B6404763002573530066008C?Open&S_TACT=105AGX16&S_CMP=LP