Re: [PATCH] tracing/function-graph-tracer: strip ending newlineson comments

[Ingo Molnar]

* Frédéric Weisbecker <fweisbec@xxxxxxxxx> wrote:

> 2008/12/23 Ingo Molnar <mingo@xxxxxxx>:
> >
> > * Frederic Weisbecker <fweisbec@xxxxxxxxx> wrote:
> >
> >> Impact: trivial output improvement
> >>
> >> Ending newlines are appended automatically on comments by the function graph tracer
> >> because the newlines needs to be placed after the "*/" characters.
> >> So if the user puts an ending whitespace, we want to strip it.
> >>
> >> Signed-off-by: Frederic Weisbecker <fweisbec@xxxxxxxxx>
> >
> > applied to tip/tracing/function-graph-tracer, thanks Frederic!
> >
> >> @@ -584,6 +585,11 @@ print_graph_comment(struct print_entry *trace, struct trace_seq *s,
> >>                               return TRACE_TYPE_PARTIAL_LINE;
> >>               }
> >>
> >> +     /* Strip ending newline on the comment */
> >> +     len = strlen(trace->buf);
> >> +     if (trace->buf[len - 1] == '\n')
> >> +             trace->buf[len - 1] = '\0';
> >> +
> >
> > Hm, dont we know the length of the buffer already, so that we could avoid
> > the strlen() overhead?
> >
> >        Ingo
> >
> 
> We know it during the insertion. But we need a new field struct 
> print_entry. That makes me remind that I didn't considered the 
> TRACE_CONT rest of the buffer in this patch. Actually, the most easy 
> solution would be to print entirely the buffer to the seq, and then 
> delete the newline directly to the seq. That would be a bit crappy but 
> actually the only solution.
> 
> So my patch is wrong in cases of long messages.

ok, i zapped it for now and will wait for v2.

	Ingo
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/