On 7/7/26 10:53, Rasmus Villemoes wrote:
> On Tue, Jul 07 2026, "Patrice CHOTARD" <[email protected]> wrote:
>
>> On 7/6/26 18:00, Marek Vasut wrote:
>>> On 7/6/26 5:16 PM, Patrice CHOTARD wrote:
>>>
>>> Hello Patrice,
>>>
>>>>>> I did a test, it doesn't bring any noticeable improvements, but
>>>>>> nevertheless i will submit a v3 with this update.
>>>>>
>>>>> Does the list iteration exit when it finds the first entry that is not
>>>>> going to be executed at this time ? If it does not, then there will be no
>>>>> improvement. But for this to work, the list would have to be surely fully
>>>>> ordered in incrementing timestamp order, is it ?
>>>>
>>>> No, the list iteration parse all entries.
>>>>
>>>> I see your point, on original implementation, entries are not sort by
>>>> incrementing timestamp order.
>>>> I did some implementation test, by sorting cyclic entries in cyclic_list
>>>> when :
>>>> _ registering a new cyclic callback, put it directly in the right
>>>> position in cyclic_list.
>>>> _ in cyclic_run(), after running a cyclic, compute its next call and
>>>> insrt it in the right position in cyclic_list.
>>>>
>>>> There is no noticeable gain.
>>>
>>> Did you also break from the iteration when you reached the first entry with
>>> future timestamp ? That might actually help.
>>>
>>
>> In cyclic_run(), as now entries was sorted by increasing timestamp, only the
>> first cyclic in cyclic_list is parsed,
>> we don't need to parse all cyclic_list element.
>>
>
> The point is kind of moot as v3 seems to work just as well, but I do
> want to point out that no, ensuring that the first expiring timer is at the
> front
> does not _sort_ the list, and breaking once we hit the first non-expired
> timer would be wrong. It would of course work for a list with at most
> two elements, but not in general.
Rasmus
I was talking about an implementation that i didn't submit to the mailing list.
In this implementation (just for trial) i ensure that new cyclic entry is
inserted
in the right position in the cylic_list in order to have all entries sorted by
increasing
timestamp.
Doing that i obtain around the same gain that your patch proposal, but yours is
more simple.
Patrice
>
> Rasmus
