Ushering out strlcpy()

Posted Aug 26, 2022 21:05 UTC (Fri) by wtarreau (subscriber, #51152)
In reply to: Ushering out strlcpy() by NYKevin
Parent article: Ushering out strlcpy()

I'm not speaking about being able or not to emit libc calls. I can understand that it can emit them when the call is obvious. What I'm saying is that just like any compiler it's extremely unlikely to spot situations where this is possible based on a loop like above. Even for a human it's not trivial to spot this. If it had strings mapped to arrays for example, like in C, I would easily imagine that we'd have an equivalent of strchr() applying to that array that would rely on the libc call. But a loop like this with tests in the middle is extremely unlikely to be turned to a series of strchr() on one given value. And by the way in such a parser it should be left to the developer's choice, because short headers would cost more with an strchr() call while long ones would benefit from a fast strchr(). Thus it would be nice to be able to call the equivalent of strchr() on positions in this array. But overall it looks like reimplementing the C string model that many people dislike.

Ushering out strlcpy()

Posted Aug 26, 2022 22:00 UTC (Fri) by NYKevin (subscriber, #129325) [Link] (1 responses)

Well, sure, but you wouldn't write that loop (in Rust) in the first place. You'd call str::find[1] with suitable arguments, and then it's the implementation's problem to emit the most efficient way to do this. Since one possible set of arguments to str::find can be statically proven to be exactly equivalent to strchr (purely in terms of the type system, so this can be done in the monomorphization pass), the implementation should emit strchr if it is optimal to do so. If the implementation does not in fact emit strchr, and you think this is suboptimal, then you should file a bug with them. But don't tell me it's impossible. It's clearly possible.

[1]: https://doc.rust-lang.org/std/primitive.str.html#method.find

Ushering out strlcpy()

Posted Aug 29, 2022 9:37 UTC (Mon) by wtarreau (subscriber, #51152) [Link]

To be honest, I think you lost me, especially since initially I understood that it had to be handled as an array of bytes and not a string, but that sort of comforts me in the idea that it can quickly become very complicated to write simple yet efficient functions for a number of low-level use cases. And my fear with languages that are complicated to use is that either the developer will introduce logic bugs when trying to hack around constraints, or that they will give up and fall back to the default, much less efficient solution (which is often what we see in practice).