Re: [PATCH] workqueue: format pr_warn exceeds line length in wq_numa_init

[张元瀚 Tio Zhang]

Hi Joe, 

Thanks for your advice.

IMHO, in workqueue, some prints start with "workqueue %s:" which means the workqueue_struct, not the workqueue module. Since we could not remove those, starting with "workqueue: workqueue %s" might be confusing sometimes I guess.

So maybe
#define pr_fmt(fmt) "WORKQUEUE: " fmt 
might work?



在 2023/6/19 10:41，“Joe Perches”<joe@xxxxxxxxxxx <mailto:joe@xxxxxxxxxxx> <mailto:joe@xxxxxxxxxxx <mailto:joe@xxxxxxxxxxx>>> 写入:




On Sun, 2023-06-18 at 09:19 +0800, Lai Jiangshan wrote:
> On Fri, Jun 16, 2023 at 3:00 PM tiozhang <tiozhang@xxxxxxxxxxxxxx <mailto:tiozhang@xxxxxxxxxxxxxx> <mailto:tiozhang@xxxxxxxxxxxxxx <mailto:tiozhang@xxxxxxxxxxxxxx>>> wrote:
> > 
> > Format this long line which would potentially let checkpatch complain
> > "WARNING: line length of 103 exceeds 100 columns".
[]
> This patch hurts the readability actually. A few extra characters exceeding
> is Okay for me.
[]
> > diff --git a/kernel/workqueue.c b/kernel/workqueue.c
[]
> > @@ -5959,7 +5959,8 @@ static void __init wq_numa_init(void)
> > 
> > for_each_possible_cpu(cpu) {
> > if (WARN_ON(cpu_to_node(cpu) == NUMA_NO_NODE)) {
> > - pr_warn("workqueue: NUMA node mapping not available for cpu%d, disabling NUMA support\n", cpu);
> > + pr_warn("workqueue: NUMA node mapping not available for cpu%d, disabling NUMA support\n",
> > + cpu);




What _might_ work reasonably well is to add




#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt




before any include and remove "workqueue: " from
all pr_<foo> uses.