Re: [PATCH v4 0/7] rust: use Delta instead of raw jiffies for timeouts and delays

On Wed Aug 5, 2026 at 1:48 PM BST, Andreas Hindborg wrote:
> "FUJITA Tomonori" <[email protected]> writes:
>
>> On Thu, 30 Jul 2026 14:09:59 +0100
>> "Gary Guo" <[email protected]> wrote:
>>
>>> On Wed Jul 22, 2026 at 10:49 PM BST, FUJITA Tomonori wrote:
>>>> From: FUJITA Tomonori <[email protected]>
>>>>
>>>> CondVar::wait_interruptible_timeout() and Queue::enqueue_delayed() use
>>>> a raw jiffies count (a plain c_ulong alias with no type
>>>> safety). Callers have to know on their own that the value meant
>>>> jiffies and convert to/from it themselves, which is easy to get wrong
>>>> (e.g. passing a millisecond value where a jiffies value is expected).
>>>>
>>>> Both APIs just take a span of time so, they can use the Delta type
>>>> instead.
>>>>
>>>> This series makes Delta generic over its time unit with Nsec and Jiffy
>>>> types, switches CondVar and Queue to use Delta instead of raw jiffies
>>>> (updating binder's ioctl_freeze(), the only caller of
>>>> wait_interruptible_timeout()), and then removes the now-unused Jiffies
>>>> and Msecs type aliases.
>>>
>>> There are more jiffy users being added so we should take this sooner than later:
>>> https://lore.kernel.org/rust-for-linux/[email protected]/#t
>>>
>>> Andreas, any plan on taking this via timekeeping-next? I suppose Boqun
>>> could also be taken via rust-sync tree if you give an ack?
>>
>> Agreed. Getting this into the next merge window would be great.
>
> I was on vacation, back now.
>
> Getting this in the merge window for v7.3 is probably a little tight. We
> should aim for v7.4.
>
> I will take closer look this week.

Ideally we can still take them this cycle so we can add users in the next one.
I suppose otherwise we might want a topic branch early next cycle.

I think if you can pick timekeeping patches today or tomorrow then they'll still
have a round or two in linux-next before rc7. Or perhaps Miguel can pick these
instead?

Best,
Gary