Re: Need help with New Build -- Skylake

On Thu, 2015-12-24 at 12:48 +0100, Oliver Pinter wrote:
> On 12/24/15, Vijay Rajah <me_at_rvijay.me> wrote:
> > 
> > 
> > On 12/24/15 12:22 PM, Ian Smith wrote:
> > >   ~2 minutes delay there.  sendmail (mta and msp both) at least
> > > are unhappy about your hostname, and sleep on it.  I don't know
> > > whether
> > > that's significant or related to the longer delay you report.  I
> > > just
> > > skimmed through your dmesg, but didn't spot anything glaringly
> > > obvious.
> > > 
> > > FWIW, cheers, Ian
> > Ian,
> > 
> > Thanks for taking the time to help me.
> > 
> > The delay is (mostly) before the boot loader menu is presented.
> > Once the
> > system starts to boot.. It is pretty fast...
> > 
> > When the system boots, after the BIOS hands over the control to the
> > OS..
> > there are 6-7 lines, which give boot loader version etc.. (the last
> > of
> > this gives the details of the build host etc..)
> > 
> > Then there is a spinning wheel, an the system just sits there for
> > some
> > time.. (even the wheel spins slowly). after 3-4 mins I see it loads
> > the
> > /boot/default/loader.conf . After this the menu is loaded..
> 
> Confirmed this issue, seems like the forth parser or something forth
> related code in slow.
> Loading the modules takes ~30 sec, and the kernel takes once more 30
> sec.
> 
> This is a Gigabyte GA-H170N-WIFI board. I can test some change after
> Christmas.
> 
> > 
> > so, the issue is long before this dmesg even begins...
> > Unfortunately, I
> > do not know how to force a verbose boot before the boot menu...
> > 
> > -Thanks
> > Vijay

We had exactly this symptom -- long delay with spincursor before
loading the kernel -- on arm systems when we first enabled forth in
loader.  The problem turned out to be the fact that loader was running
with instruction and data caches disabled, and it took about 90-100
seconds to parse the 547 lines of text (almost all useless) in
/boot/defaults/loader.conf.  We stripped that file down to the dozen or
so lines that actually needed to be there and booting became much
faster.  Eventually we got the caches enabled in the prior-stage
bootloader and it became really fast.

-- Ian