Re: [PATCH] entry: fix compile error in dynamic_irqentry_exit_cond_resched()

[Sven Schnelle <svens@linux.ibm.com>]

Mark Rutland <mark.rutland@arm.com> writes:

> On Wed, Mar 30, 2022 at 10:43:28AM +0200, Sven Schnelle wrote:
>> kernel/entry/common.c: In function ‘dynamic_irqentry_exit_cond_resched’:
>> kernel/entry/common.c:409:14: error: implicit declaration of
>> function ‘static_key_unlikely’; did you mean ‘static_key_enable’?
>> [-Werror=implicit-function-declaration]
>>   409 |         if (!static_key_unlikely(&sk_dynamic_irqentry_exit_cond_resched))
>>       |              ^~~~~~~~~~~~~~~~~~~
>>       |              static_key_enable
>> 
>> static_key_unlikely() should be static_branch_unlikely().
>> 
>> Fixes: 99cf983cc8bca ("sched/preempt: Add PREEMPT_DYNAMIC using static keys")
>> Signed-off-by: Sven Schnelle <svens@linux.ibm.com>
>
> Sorry about this. FWIW:
>
> Reviewed-by: Mark Rutland <mark.rutland@arm.com>
>
> For context for others, this'll only show up on architectures which both use
> the generic entry code and select CONFIG_HAVE_PREEMPT_DYNAMIC_KEY. Today, only
> arm64 selects CONFIG_HAVE_PREEMPT_DYNAMIC_KEY, and it doesn't use the generic
> entry code.
>
> Sven, I assume you're looking at wiring this up on s390 or parisc?

Yes, i'm looking whether we can use the same implementation on s390. :)

I reported it already on 03/18, but looks like that Mail was lost
somehow:

https://www.spinics.net/lists/kernel/msg4283802.html

I was wondering whether we can make dynamic_irqentry_exit_cond_resched()
static, so it gets inlined. On s390 the compiler generates a branch to
that function just to return immediately if the static key isn't enabled.
With static it would get inlined, and therefore save one function call.
What do you think?

Thanks
Sven