Re: [PATCH 13/13] rcu/kvfree: Eliminate k[v]free_rcu() single argument macro

[Joel Fernandes]

> On Mar 5, 2023, at 6:41 AM, Uladzislau Rezki <urezki@xxxxxxxxx> wrote:
> 
> 
>> 
>>>> On Mar 5, 2023, at 5:29 AM, Joel Fernandes <joel@xxxxxxxxxxxxxxxxx> wrote:
>>> 
>>> Hi, All,
>>> 
>>>> On Wed, Feb 1, 2023 at 10:11 AM Uladzislau Rezki (Sony)
>>>> <urezki@xxxxxxxxx> wrote:
>>>> 
>>>> For a single argument invocations a new kfree_rcu_mightsleep()
>>>> and kvfree_rcu_mightsleep() macroses are used. This is done in
>>>> order to prevent users from calling a single argument from
>>>> atomic contexts as "_mightsleep" prefix signals that it can
>>>> schedule().
>>>> 
>>> 
>>> Since this commit in -dev branch [1] suggests more users still need
>>> conversion, let us drop this single patch for 6.4 and move the rest of
>>> the series forward? Let me know if you disagree.
>>> https://git.kernel.org/pub/scm/linux/kernel/git/paulmck/linux-rcu.git/commit/?h=dev&id=9bf5e3a2626ed474d080f695007541b6ecd6e60b
>>> 
>>> All -- please supply Ack/Review tags for patches 1-12.
>> 
>> Or put another way, what is the transition plan for these remaining users?
>> 
>> I am getting on a plane right now but I can research which users are remaining later.
>> 
> I am not sure. I think we can cover it on the meeting.

Cool, thanks.

> My feeling is
> that, we introduced "_mightsleep" macros first and after that try to
> convert users.

One stopgap could be to add a checkpatch error if anyone tries to use old API,
and then in the meanwhile convert all users.
Though, that requires people listening to checkpatch complaints.

Thanks,

 - Joel


> 
> @Paul what is your view?
> 
> Thanks!
> 
> --
> Uladzislau Rezki