[dpdk-dev] [PATCH v3 0/8] eal: dynamic logs

[Thomas Monjalon]

2017-04-12 13:11, De Lara Guarch, Pablo:
> From: Thomas Monjalon [mailto:thomas.monjalon at 6wind.com]
> > Sent: Wednesday, April 12, 2017 11:38 AM
> > 2017-04-12 09:26, De Lara Guarch, Pablo:
> > > From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Olivier Matz
> > > >
> > > > The objective of this patchset is to introduce a framework to
> > > > support dynamic log types in EAL. It also provides one example of use
> > > > (in i40e).
> > > >
> > > > Features:
> > > > - log types are identified by a string
> > > > - at registration, a uniq identifier is associated to a log type
> > > > - each log type can have its level changed dynamically
> > > > - extend command line parameters to set the log level of a specific
> > > >   type, or logs matching a regular expression
> > > > - keep compat with other legacy types (eal, malloc, ring, user*,
> > > >   etc... keep their hardcoded log type value)
> > > >
> > > > Next step is to adapt drivers, libs and apps to use this new API. At the
> > > > end, we can expect that all non-dataplane logs are moved to be
> > dynamic,
> > > > so we can enable/disable them at runtime, without recompiling. Many
> > > > debug options can probably be removed from configuration:
> > > >   $ git grep DEBUG config/common_base | wc -l
> > > >   89
> > [...]
> > > With this patch, all logs that use logtype "USERX" (e.g.
> > RTE_LOGTYPE_USER1) are not shown anymore.
> > > Should these macro be removed?
> > >
> > > Right now, all applications using this won't show these, so I assume that
> > all of them
> > > should be fixed before the release is out.
> > > Is that correct?
> > 
> > Is it a bug in the commit http://dpdk.org/commit/c1b5fa9 ?
> > Note this line:
> > 	__rte_log_register("user1", 24);
> 
> This also happens with libraries, like HASH. I think the problem might be here:
> 
> @@ -139,7 +266,11 @@ rte_vlog(uint32_t level, uint32_t logtype, const char *format, va_list ap)
>  		}
>  	}
>  
> -	if ((level > rte_logs.level) || !(logtype & rte_logs.type))
> +	if (level > rte_logs.level)
> +		return 0;
> +	if (logtype >= rte_logs.dynamic_types_len)
> +		return -1;
> +	if (level > rte_logs.dynamic_types[logtype].loglevel)
>  		return 0;
> 
> Type used to be a mask, so for instance, RTE_LOGTYPE_HASH 0x0...40 = 64 in decimal.
> Therefore, logtype = 64 >= 34 = rte_logs.dynamic_types_len), so this won't be shown.
> However, the PMDs or EAL will show it, because their logtype is less than 34.
> 
> Am I missing something?

I think you are right.
We are mixing old logtype and new ones.
Should we redefine RTE_LOGTYPE_* values to the new values registered
in rte_log_init?