Re: [RFC PATCH v4 1/2] trace,smp: Add tracepoints around remotelly called functions

From: Valentin Schneider <vschneid@redhat.com>
To: Leonardo Bras <leobras@redhat.com>,
	Thomas Gleixner <tglx@linutronix.de>,
	Sebastian Andrzej Siewior <bigeasy@linutronix.de>,
	Nadav Amit <namit@vmware.com>,
	Chen Zhongjin <chenzhongjin@huawei.com>,
	Daniel Bristot de Oliveira <bristot@kernel.org>,
	Marcelo Tosatti <mtosatti@redhat.com>,
	Steven Rostedt <rostedt@goodmis.org>,
	Masami Hiramatsu <mhiramat@kernel.org>,
	Leonardo Bras <leobras@redhat.com>,
	Peter Zijlstra <peterz@infradead.org>,
	"Paul E. McKenney" <paulmck@kernel.org>,
	Juergen Gross <jgross@suse.com>,
	Yury Norov <yury.norov@gmail.com>
Cc: linux-kernel@vger.kernel.org, linux-trace-kernel@vger.kernel.org
Subject: Re: [RFC PATCH v4 1/2] trace,smp: Add tracepoints around remotelly called functions
Date: Tue, 30 May 2023 11:36:16 +0100	[thread overview]
Message-ID: <xhsmh353ep0m7.mognet@vschneid.remote.csb> (raw)
In-Reply-To: <20230515183045.654199-2-leobras@redhat.com>

On 15/05/23 15:30, Leonardo Bras wrote:
> @@ -375,7 +386,7 @@ static int generic_exec_single(int cpu, call_single_data_t *csd)
>               csd_lock_record(csd);
>               csd_unlock(csd);
>               local_irq_save(flags);
> -		func(info);
> +		csd_do_func(func, info, csd);

I'd suggest making this match the local case of
smp_call_function_many_cond(), IOW pass NULL as the csd when executing
locally.

IMO this is required for postprocessing with e.g. synthetic events for CSD
delivery measurement, otherwise we'll try to match this execution with a
previous CSD enqueue.

>               csd_lock_record(NULL);
>               local_irq_restore(flags);
>               return 0;