> > My feeling is that an "end user" of a library expects > fast and reliable execution and low overhead. He would not > care if the lines that generated the hex file were human > readable or not.
> I guess the end user also wants reliable code. I mentioned "reliable" :-) > Research > proves there > is a distinct correlation between easy-to-read code and > error rate I did not search for proof, as it seemed obvious to me that readable code is easier to debug. But, for those that use assembler, documented programs can be readable. I do not speak for unreadable code, I've probably used the wrong words. All libs should be readable for those who are familiar with the used language. Just if someone doesn't speak asm at all, he can still use the libs, and leave debugging and modifications to those who do. I wouldn't recommend to translate the whole USB or FAT lib to assembler. They would be converted from great pieces of work to probably unusable piles of bugs. But, for the format lib, one input must bring one output. So, it is easy to check if it works right. Output all possible numbers, and check the results using proven-good programs. That test may run a month... Ok, but after that, the lib can be considered reliable. I've written programs in pure asm, and if the compiler handles inline as it claims to do, there should not arise new problems by using asm. Greets, Kiste -- You received this message because you are subscribed to the Google Groups "jallib" group. To post to this group, send email to [email protected]. To unsubscribe from this group, send email to [email protected]. For more options, visit this group at http://groups.google.com/group/jallib?hl=en.
