> The 'reason' was computed only when logging at level DBG. However, if a
> node takes more than 'engine_compute_log_timeout_msec' to recompute, the
> log is generated at level INFO, resulting in:
>
> inc_proc_eng|INFO|node: northd, recompute ((null))
>
> We now unconditionally compute 'reason'. In theory it's a bit
> inefficient but given that the number of I-P nodes is bound and usually
> relatively low, there's no real performance hit while the code remains
> readable.
>
> Fixes: e1041d9c0bad ("inc-proc-eng: use VLOG_INFO_RL for recompute time over
> 500ms")
> Signed-off-by: Dumitru Ceara <[email protected]>
Acked-by: Lorenzo Bianconi <[email protected]>
> ---
> lib/inc-proc-eng.c | 11 ++++-------
> 1 file changed, 4 insertions(+), 7 deletions(-)
>
> diff --git a/lib/inc-proc-eng.c b/lib/inc-proc-eng.c
> index 7b43917006..575b774ae6 100644
> --- a/lib/inc-proc-eng.c
> +++ b/lib/inc-proc-eng.c
> @@ -354,14 +354,11 @@ engine_recompute(struct engine_node *node, bool allowed,
> const char *reason_fmt, ...)
> {
> char *reason = NULL;
> + va_list reason_args;
>
> - if (VLOG_IS_DBG_ENABLED()) {
> - va_list reason_args;
> -
> - va_start(reason_args, reason_fmt);
> - reason = xvasprintf(reason_fmt, reason_args);
> - va_end(reason_args);
> - }
> + va_start(reason_args, reason_fmt);
> + reason = xvasprintf(reason_fmt, reason_args);
> + va_end(reason_args);
>
> if (!allowed) {
> VLOG_DBG("node: %s, recompute (%s) aborted", node->name, reason);
> --
> 2.27.0
>
_______________________________________________
dev mailing list
[email protected]
https://mail.openvswitch.org/mailman/listinfo/ovs-dev
