On Thu, 6 Apr 2017 15:36:43 +0100, Bruce Richardson <bruce.richard...@intel.com> wrote: > On Thu, Apr 06, 2017 at 03:31:23PM +0100, Zhang, Helin wrote: > > > > > > -----Original Message----- > > From: Olivier Matz [mailto:olivier.m...@6wind.com] > > Sent: Thursday, April 6, 2017 10:17 PM > > To: dev@dpdk.org > > Cc: Zhang, Helin <helin.zh...@intel.com>; Wu, Jingjing > > <jingjing...@intel.com>; Richardson, Bruce <bruce.richard...@intel.com>; > > Yigit, Ferruh <ferruh.yi...@intel.com> > > Subject: [PATCH] net/i40e: disable init and driver logs by default > > > > Since "net/i40e: use dynamic log type for control logs", the i40e driver is > > more verbose by default, which could result in testpmd being flooded by log > > messages in some conditions: > > > > Checking link statuses... > > i40e_dev_handle_aq_msg(): Request 2561 is not supported yet > > i40e_dev_handle_aq_msg(): Request 2561 is not supported yet > > i40e_dev_handle_aq_msg(): Request 2561 is not supported yet > > i40e_dev_handle_aq_msg(): Request 2561 is not supported yet > > Port 0 Link Up - speed 40000 Mbps - full-duplex > > Port 1 Link Up - speed 40000 Mbps - full-duplex > > Done > > testpmd> i40e_dev_handle_aq_msg(): Request 4097 is not supported yet > > i40e_dev_handle_aq_msg(): Request 4097 is not supported yet > > i40e_dev_handle_aq_msg(): Request 4097 is not supported yet > > > > Fix this by disabling the dynamic logs by default. It is still possible to > > enable them at runtime. > > > > Fixes: c143e5a3d9e1 ("net/i40e: use dynamic log type for control logs") > > > > Signed-off-by: Olivier Matz <olivier.m...@6wind.com> > > Acked-by: Helin Zhang <helin.zh...@intel.com> > > --- > > drivers/net/i40e/i40e_ethdev.c | 4 ++-- > > 1 file changed, 2 insertions(+), 2 deletions(-) > > > > diff --git a/drivers/net/i40e/i40e_ethdev.c > > b/drivers/net/i40e/i40e_ethdev.c index cb8bf2b49..86f5669f0 100644 > > --- a/drivers/net/i40e/i40e_ethdev.c > > +++ b/drivers/net/i40e/i40e_ethdev.c > > @@ -12456,8 +12456,8 @@ i40e_init_log(void) { > > i40e_logtype_init = rte_log_register("pmd.i40e.init"); > > if (i40e_logtype_init >= 0) > > - rte_log_set_level(i40e_logtype_init, RTE_LOG_NOTICE); > > + rte_log_set_level(i40e_logtype_init, RTE_LOG_EMERG); > > i40e_logtype_driver = rte_log_register("pmd.i40e.driver"); > > if (i40e_logtype_driver >= 0) > > - rte_log_set_level(i40e_logtype_driver, RTE_LOG_NOTICE); > > + rte_log_set_level(i40e_logtype_driver, RTE_LOG_EMERG); > > } > > -- > > Hi Helin, > > Is this the correct fix? IMHO, if this is a problem, then we should > surely not be hiding and ignoring the error. If it's not a problem, then > the log level should be reduced to a lower level, e.g. INFO or NOTICE. > > Alternatively, this code could be modified to only print an error once > for each unsupported request type. > > Ideally both solutions should be used, I think. I'm not sure I like > setting the default log level to just show EMERG messages.
Yes, now that these logs are not stripped at compilation, we should do an effort to have the proper default log value (notice, warning, error?), and have the proper log level on each log, depending on its criticality and maybe its frequency (maybe we'll want to rate limit the logs in the future). This is a more global discussion, which is not especially related to i40e (the only part of code that currently use dynamic logs). Olivier
