[dpdk-dev] [PATCH] eal: add option to put timestamp on console output

[Stephen Hemminger]

On Mon, 19 Oct 2020 16:11:19 +0200
Thomas Monjalon <thomas at monjalon.net> wrote:

> There was no ack or conclusion for this patch.
> 
> 17/08/2020 17:11, Stephen Hemminger:
> > On Mon, 17 Aug 2020 11:37:40 +0100
> > Bruce Richardson <bruce.richardson at intel.com> wrote:
> >   
> > > On Fri, Aug 14, 2020 at 12:20:28PM -0700, Stephen Hemminger wrote:  
> > > > On Fri, 14 Aug 2020 22:09:11 +0300
> > > > Dmitry Kozlyuk <dmitry.kozliuk at gmail.com> wrote:
> > > >     
> > > > > On Fri, 14 Aug 2020 11:45:49 -0700, Stephen Hemminger wrote:    
> > > > > > On Fri, 14 Aug 2020 21:39:29 +0300
> > > > > > Dmitry Kozlyuk <dmitry.kozliuk at gmail.com> wrote:      
> > > > > [...]    
> > > > > > FreeBsd doesn't go through the same log code.      
> > > > > 
> > > > > Then --log-timestamp shouldn't appear on FreeBSD if it's unsupported.    
> > > > 
> > > > Was using same ifdef that is already there. There are several options already
> > > > (like syslog facility) which do nothing on FreeBSD anyway.
> > > >     
> > > While that is correct, I think that we really should support this on
> > > FreeBSD also, since I don't think there is any technical reason we can't do
> > > so, right? Even a sub-optimal solution where the timestamp write and log
> > > write are non-atomic would be better than nothing.
> > > 
> > > /Bruce  
> > 
> > FreeBSD now just uses default log stream (stdout).
> > Linux has log stream to syslog and stdout.
> > 
> > There is nothing Linux specific in the old or new code.
> > The Linux code could just be moved to common, and both could use it.
> > That is a separate patch, and someone with expertise and build setup for FreeBSD needs to do it.
> >   

Lets fix FreeBSD to use common code for logging, then timestamp option can be added?