Received-SPF: pass (google.com: domain of linux-kernel+bounces-91466-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:45e3:2400::1 as permitted sender) client-ip=2604:1380:45e3:2400::1;
Date: Mon, 4 Mar 2024 19:49:08 -0500
From: Steven Rostedt <rostedt@goodmis.org>
To: Randy Dunlap <rdunlap@infradead.org>
Cc: LKML <linux-kernel@vger.kernel.org>, Linux Trace Kernel
 <linux-trace-kernel@vger.kernel.org>, Masami Hiramatsu
 <mhiramat@kernel.org>, Mathieu Desnoyers <mathieu.desnoyers@efficios.com>,
 Linus Torvalds <torvalds@linux-foundation.org>, Sachin Sant
 <sachinp@linux.ibm.com>
Subject: Re: [PATCH] tracing: Have trace_marker writes be just half of
 TRACE_SEQ_SIZE
Message-ID: <20240304194908.6d8d4fba@gandalf.local.home>
In-Reply-To: <fa113c5c-07c7-417b-8287-c9bf5f9b42a1@infradead.org>
References: <20240304192710.4c99677c@gandalf.local.home>
	<fa113c5c-07c7-417b-8287-c9bf5f9b42a1@infradead.org>
Precedence: bulk
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit

On Mon, 4 Mar 2024 16:43:46 -0800
Randy Dunlap <rdunlap@infradead.org> wrote:

> > diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c
> > index 8198bfc54b58..d68544aef65f 100644
> > --- a/kernel/trace/trace.c
> > +++ b/kernel/trace/trace.c
> > @@ -7320,6 +7320,17 @@ tracing_mark_write(struct file *filp, const char __user *ubuf,
> >  	if ((ssize_t)cnt < 0)
> >  		return -EINVAL;
> >  
> > +	/*
> > +	 * TRACE_SEQ_SIZE is the total size of trace_seq buffer used
> > +	 * for output. As the print event outputs more than just
> > +	 * the string written, keep it smaller than the trace_seq
> > +	 * as it could drop the event if the extra data makes it bigger
> > +	 * than what the trace_seq can hold. Half he TRACE_SEQ_SIZE  
> 
> 	                                          the

I honestly think my 't' key isn't triggering as much. At least when before
hitting 'h', as I noticed I've been writing "he", "hey" and "here" a lot,
and spell check isn't (obviously) catching it ;-)

-- Steve


> 
> > +	 * is more than enough.
> > +	 */
> > +	if (cnt > TRACE_SEQ_SIZE / 2)
> > +		cnt = TRACE_SEQ_SIZE / 2;
> > +
> >  	meta_size = sizeof(*entry) + 2;  /* add '\0' and possible '\n' */
> >   again:
> >  	size = cnt + meta_size;  
> 
>