Re: [RFC PATCH v10 3/4] blk-iocost: add iocost_ioc_tick tracepoint for per-period device summary

[Tao Cui <cui.tao@linux.dev>]

Hello Tejun,

在 2026/10/17 01:49, Tejun Heo 写道:
> Hello, Tao.
> 
> On Sat, 03 Oct 2026 20:40:44 +0800, Tao Cui wrote:
>> + * Periodic per-device summary, emitted once per period from the tail of
>> + * ioc_timer_fn().  Unlike the state-change events above, this fires every
> 
> The event is emitted before the vrate adjustment and the period
> transition, not from the tail.
> 
Updated both the commit message and the comment to say that the event
is emitted before the vrate adjustment and the period transition.

>> +		/* whether the controller still has active iocgs */
>> +		__field(int, running)
> 
> nr_active counts every entry on ioc->active_iocgs, and ioc->lock is held
> from the count to the emit point, so running always equals active != 0.
> Maybe drop the field?
> 
Dropped, along with the printk column and the corresponding commit
message paragraph. With the lock held across the count and the
tracepoint emission, it always equals active != 0, so it carries no
additional information.

Thanks,
Tao

> Thanks.
>