netdev - Re: [PATCH net-next v1 1/2] rust: add delay abstraction

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <3fc44d62-586f-4ed0-88ee-a561bef1fdaf@lunn.ch>
Date: Wed, 2 Oct 2024 14:51:41 +0200
From: Andrew Lunn <andrew@...n.ch>
To: Miguel Ojeda <miguel.ojeda.sandonis@...il.com>
Cc: Thomas Gleixner <tglx@...utronix.de>,
	FUJITA Tomonori <fujita.tomonori@...il.com>, netdev@...r.kernel.org,
	rust-for-linux@...r.kernel.org, hkallweit1@...il.com,
	tmgross@...ch.edu, ojeda@...nel.org, alex.gaynor@...il.com,
	gary@...yguo.net, bjorn3_gh@...tonmail.com, benno.lossin@...ton.me,
	a.hindborg@...sung.com, aliceryhl@...gle.com
Subject: Re: [PATCH net-next v1 1/2] rust: add delay abstraction

On Wed, Oct 02, 2024 at 02:35:57PM +0200, Miguel Ojeda wrote:
> On Wed, Oct 2, 2024 at 2:19 PM Andrew Lunn <andrew@...n.ch> wrote:
> >
> > How well know is that? And is there a rust-for-linux wide preference
> > to use Duration for time? Are we going to get into a situation that
> > some abstractions use Duration, others seconds, some milliseconds,
> > etc, just like C code?
> 
> We have `Ktime` that wraps `ktime_t`.
> 
> However, note that something like `Ktime` or `Duration` are types, not
> a typedef, i.e. it is not an integer where you may confuse the unit.
> 
> So, for instance, the implementation here calls `as_micros()` to get
> the actual integer. And whoever constructs e.g. a `Duration` has
> several constructors to do so, not just the one that takes seconds and
> nanoseconds, e.g. there is also `from_millis()`.
> 
> Perhaps we may end up needing different abstractions depending on what
> we need (Cc'ing Thomas), but we will almost certainly want to still
> use types like this, rather than plain integers or typedefs where
> units can be confused.
> 
> > Anyway, i would still document the parameter is a Duration, since it
> > is different to how C fsleep() works.
> 
> That is in the signature itself -- so taking into account what I
> mentioned above, does it make sense that seeing the type in the
> signature would be enough?

Which is better, the Rust type system catching the error, or not
making the error in the first place because you read the documentation
and it pointed you in the right direction?

Maybe this is my background as a C programmer, with its sloppy type
system, but i prefer to have this very clear, maybe redundantly
stating it in words in addition to the signature.

	Andrew