On 21 10月 21 09:58:58, Steven Rostedt wrote:

On Thu, 21 Oct 2021 21:53:12 +0800
Cai Huoqing [off-list ref] wrote:

quoted

quoted

quoted

+/**
+ * kthread_run_on_cpu - create and wake a cpu bound thread.
+ * @threadfn: the function to run until signal_pending(current).
+ * @data: data ptr for @threadfn.
+ * @cpu: The cpu on which the thread should be bound,
+ * @namefmt: printf-style name for the thread. Format is restricted
+ *	     to "name.*%u". Code fills in cpu number.
+ *
+ * Description: Convenient wrapper for kthread_create_on_node()
+ * followed by wake_up_process().  Returns the kthread or
+ * ERR_PTR(-ENOMEM).
+ */
+#define kthread_run_on_cpu(threadfn, data, cpu, namefmt)		  \  

Why is this a macro and not a static inline function?

-- Steve  

Hi,Thanks for your feedback,

I think using static inline function is nice, but here try to keep
consistent with the other macros,
sush as kthread_create/kthread_init_work...

Which they did because they didn't want to use va_list to have variable
arguments, which you don't have.

Which begs the question, should you?

-- Steve

Hi, Thanks your reply.

I get it, V3 here:
https://lore.kernel.org/lkml/20211022025711.3673-2-caihuoqing@baidu.com/ (local)

Thanks,
Cai