How do I specify the execution time of x86 and PowerPC instructions?

Question

I need to get close to the runtime of PowerPC and x86 assembler code. I understand that I can’t calculate exactly, it depends on many problems (the current state of the processor - the x86 processor extends internal instructions in microcommands, access time to memory, time to get code from the cache from slower memory, etc.).

I found some information in the Intel Optimization help system (APPENDIX C), but does not contain information about all general-purpose instructions. Is there any complete information about this?

What about PowerPC processors? Where can I find such information?

Answer 1

, .

: - , , , . , .

Answer 2

PowerPC , , . IBM 970 (G5). Intel , .

, , , , . x86, PowerPC - , , , , . , PowerPC 970 215 " " . , .

Answer 3

- .

, .

Answer 4

You will need to do a very rigorous analysis. Consider all caches, alignment, pipelining, timing, etc. Etc. Etc. Does x86 even have a hard beat time for each instruction? It is best to write optimized code for speed according to what the processor manual suggests.