On Thursday 17 April 2003 08:07, Randolph Bentson wrote:
>
> That's true if documentation serves only to describe _what_ the
> code does. You've ignored the need to describe _why_ the code
> was written in this way. For example, documentation can note some
> feature of the hardware which requires special handling, or it can
> describe some emergent property which isn't obvious even if all
> the code is understood.
>
> That's why local comments should explain non-obvious trickery used,
> perhaps the exploitation of a poorly documented side-effect of some
> instruction, and block comments or external documentation should
> help the reader understand why things are done some particular way.
> For instance, if the code implements some specific, well documented
> algorithm, it should reference the algorithm by name.
You're right; this is very true. I suppose I overlooked that due to
my background in other kinds of code bases... In an "ideal" code
base, written from scratch, you'd probably need very very few
comments. If you're interfacing with buggy legacy code or
problematic hardware, I can see there'd be a much greater need.
Eric
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Wed Apr 23 2003 - 22:00:22 EST