[RFC v2 09/14] rcu: introduce a logging helper

[David Marchand]

On Fri, Dec 8, 2023 at 10:27 PM Tyler Retzlaff
<roretzla at linux.microsoft.com> wrote:
>
> On Fri, Dec 08, 2023 at 03:59:43PM +0100, David Marchand wrote:
> > Signed-off-by: David Marchand <david.marchand at redhat.com>
> > ---
>
> Reviewed-by: Tyler Retzlaff <roretzla at linux.microsoft.com>
>
> >  lib/rcu/rte_rcu_qsbr.c | 62 ++++++++++++++++--------------------------
> >  lib/rcu/rte_rcu_qsbr.h |  1 +
> >  2 files changed, 24 insertions(+), 39 deletions(-)
> >
> > diff --git a/lib/rcu/rte_rcu_qsbr.c b/lib/rcu/rte_rcu_qsbr.c
> > index 41a44be4b9..5b6530788a 100644
> > --- a/lib/rcu/rte_rcu_qsbr.c
> > +++ b/lib/rcu/rte_rcu_qsbr.c
> > @@ -19,6 +19,9 @@
> >  #include "rte_rcu_qsbr.h"
> >  #include "rcu_qsbr_pvt.h"
> >
> > +#define RCU_LOG(level, fmt, args...) \
> > +     RTE_LOG(level, RCU, "%s(): " fmt "\n", __func__, ## args)
> > +
>
> Since you are looking in the area for all the versions of gcc/clang we
> use able to support the non-standard __VA_ARGS__ that discard the comma?

I suspect there is some typo (maybe s/for all/are all/ ?).
Can you please clarify?


>
> I know that some versions of gcc do and if it does I would like to move
> to using __VA_ARGS__ instead of args so we can use the same thing with
> msvc.

If the request is to translate the ## args stuff to __VA_ARGS__, I
would prefer this is done in a separate series and not to mix with
this already huge series.


-- 
David Marchand