[PATCH v2 2/9] log: don't build the trace buffer when log is not ready

[Simon Glass]

Hi Sean,

On Fri, 27 Nov 2020 at 07:50, Sean Anderson <seanga2 at gmail.com> wrote:
>
> On 11/27/20 5:20 AM, Patrick Delaunay wrote:
> > Update _log function to drop any traces when log is yet initialized:
> > vsnprintf is no more executed in this case.
> >
> > This patch allows to reduce the cost for the dropped early debug trace.
> >
> > Reviewed-by: Simon Glass <sjg at chromium.org>
> > Signed-off-by: Patrick Delaunay <patrick.delaunay at st.com>
> > ---
> >
> > (no changes since v1)
> >
> >   common/log.c | 13 ++++++++-----
> >   1 file changed, 8 insertions(+), 5 deletions(-)
> >
> > diff --git a/common/log.c b/common/log.c
> > index ce39918e04..212789d6b3 100644
> > --- a/common/log.c
> > +++ b/common/log.c
> > @@ -228,6 +228,9 @@ int _log(enum log_category_t cat, enum log_level_t level, const char *file,
> >       struct log_rec rec;
> >       va_list args;
> >
> > +     if (!gd)
> > +             return -ENOSYS;
>
> How early are you expecting this function to get called? AFAIK this will
> only return true before board_init_f_init_reserve. Shouldn't functions
> that early just not call log in the first place?

Heinrich may have some thoughts here.

I think we will end up using log very early, since it is sort-of the
replacement for printf() and debug().

Regards,
Simon