| ??? 12/29/04 00:31 Read: times |
#84009 - The human optimization Responding to: ???'s previous message |
Is there any way to test the code effeciency or performance to tell which one is faster an Assembler in this case MetalLink? I don't think so. While the "efficiency" of generated code might be a factor with code written in a high level language like C, where you can choose from various optimization levels, in an assembly coding, the efficiency of generated code depends on you - the programmer. The assembler just translates your mnemonics to machine language and has a minimal or no role to play in optimizing it or otherwise "making" it efficient. Of course there is this aspect of handling generic JMP and CALL instructions. The smart assembler will decide which variety to use - L, A, S. Apart from this, I am not sure if the assembler can contribute anything towards efficiency. With some time consuming calculations (manually )of the instruction cycles in the two assembler codes, one can actually see which is fast ? It was an interesting question that you asked - let me also follow this thread for more enlightnment on this effic`iency stuff. Raghu |
| Topic | Author | Date |
| Code efficiency/performance | 01/01/70 00:00 | |
| The human optimization | 01/01/70 00:00 | |
| No Magic Wand | 01/01/70 00:00 | |
| Optimizing | 01/01/70 00:00 | |
| A non-generalizable task... | 01/01/70 00:00 | |
| Code efficiency | 01/01/70 00:00 | |
| Single step with index finger | 01/01/70 00:00 | |
| Efficiency | 01/01/70 00:00 | |
| No Magic Wand | 01/01/70 00:00 | |
| No Magic Wand | 01/01/70 00:00 | |
Outstanding Effort ! | 01/01/70 00:00 |