Re: [PATCH] Move kfree_call_rcu() to slab_common.c

[Rao Shoaib <rao.shoaib@oracle.com>]

[-- Attachment #1: Type: text/plain, Size: 1729 bytes --]



On 12/21/2017 05:39 PM, Paul E. McKenney wrote:
>> I left it out on purpose because the call in tiny is a little different
>>
>> rcutiny.h:
>>
>> static inline void kfree_call_rcu(struct rcu_head *head,
>>  A A A  A A A  A A A  A A A  A  void (*func)(struct rcu_head *rcu))
>> {
>>  A A A  call_rcu(head, func);
>> }
>>
>> tree.c:
>>
>> void kfree_call_rcu(struct rcu_head *head,
>>  A A A  A A A  A A A  void (*func)(struct rcu_head *rcu))
>> {
>>  A A A  __call_rcu(head, func, rcu_state_p, -1, 1);
>> }
>> EXPORT_SYMBOL_GPL(kfree_call_rcu);
>>
>> If we want the code to be exactly same I can create a lazy version
>> for tiny as well. However,A  I don not know where to move
>> kfree_call_rcu() from it's current home in rcutiny.h though. Any
>> thoughts ?
> I might be missing something subtle here, but in case I am not, my
> suggestion is to simply rename rcutiny.h's kfree_call_rcu() and otherwise
> leave it as is.  If you want to update the type of the second argument,
> which got missed back in the day, there is always this:
>
> static inline void call_rcu_lazy(struct rcu_head *head, rcu_callback_t func)
> {
> 	call_rcu(head, func);
> }
>
> The reason that Tiny RCU doesn't handle laziness specially is because
> Tree RCU's handling of laziness is a big no-op on the single CPU systems
> on which Tiny RCU runs.  So Tiny RCU need do nothing special to support
> laziness.
>
> 							Thanx, Paul
>
Hi Paul,

I can not just change the name as __kfree_call_rcu macro calls 
kfree_call_rcu(). I have made tiny version of kfree_call_rcu() call 
rcu_call_lazy() which calls call_rcu(). As far as the type is concerned, 
my bad, I cut and posted from an older release. Latest code is already 
using the typedef.

Shoaib

[-- Attachment #2: Type: text/html, Size: 2157 bytes --]