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: <20241002.135832.841519218420629933.fujita.tomonori@gmail.com>
Date: Wed, 02 Oct 2024 13:58:32 +0000 (UTC)
From: FUJITA Tomonori <fujita.tomonori@...il.com>
To: aliceryhl@...gle.com
Cc: andrew@...n.ch, 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
Subject: Re: [PATCH net-next v1 1/2] rust: add delay abstraction

On Wed, 2 Oct 2024 14:37:55 +0200
Alice Ryhl <aliceryhl@...gle.com> wrote:

> On Wed, Oct 2, 2024 at 2:19 PM Andrew Lunn <andrew@...n.ch> wrote:
>>
>> > > I would also document the units for the parameter. Is it picoseconds
>> > > or centuries?
>> >
>> > Rust's Duration is created from seconds and nanoseconds.
>>
>> 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?
>>
>> Anyway, i would still document the parameter is a Duration, since it
>> is different to how C fsleep() works.
> 
> I'm not necessarily convinced we want to use the Rust Duration type.
> Similar questions came up when I added the Ktime type. The Rust
> Duration type is rather large.

core::mem::size_of::<core::time::Duration>() says 16 bytes.

You prefer to add a simpler Duration structure to kernel/time.rs?
Something like:

struct Duration {
    nanos: u64,
}

u64 in nanoseconds is enough for delay in the kernel, I think.


btw, core::time::Duration::new() could panic if a driver does
something stupid; for example,

let d = core::time::Duration::new(u64::MAX, 1_000_000_000);

Is this acceptable?