Date: Wed, 15 Aug 2018 10:45:03 -0600 From: Warner Losh <imp@bsdimp.com> To: Kyle Evans <kevans@freebsd.org> Cc: David Wolfskill <david@catwhisker.org>, current <current@freebsd.org> Subject: Re: r337738 -> r337834: Forth loader OK; Lua loader says "BTX halted" Message-ID: <CANCZdfqJDAWzHCgGctPv-Ao=-2P4Uwz95DxLQKp=7nkoxZ7MSw@mail.gmail.com> In-Reply-To: <CACNAnaEQLDucX0=arGoLv51WWbqpfXnth5LVh_8oNOC9U5k4cQ@mail.gmail.com> References: <20180815124000.GB1190@albert.catwhisker.org> <CACNAnaEewf%2BXztU1pk=um-vkQFN0rvbHo1ettBhiTkSyvHQuhg@mail.gmail.com> <CACNAnaEQLDucX0=arGoLv51WWbqpfXnth5LVh_8oNOC9U5k4cQ@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Wed, Aug 15, 2018 at 7:00 AM, Kyle Evans <kevans@freebsd.org> wrote:
> On Wed, Aug 15, 2018 at 7:53 AM, Kyle Evans <kevans@freebsd.org> wrote:
> > On Wed, Aug 15, 2018 at 7:40 AM, David Wolfskill <david@catwhisker.org>
> wrote:
> >> I'm tracking head/amd64 daily twice on each of two machines: my "build
> >> machine" ("freebeast") and my laptop, each first using the traditional
> >> Forth loader, then (on a different slice), the Lua loader.
> >>
> >> Each is using BIOS and MBR (not UEFI; not GPT).
> >>
> >> Yesterday's update was to r337738, and was uneventful for each of
> >> the four environments.
> >>
> >> Today's was to r337834, and was uneventful for the build machine (both
> >> Forth and Lua loaders -- though there remains no means of interacting
> >> with the loader on the build machine if it's booted using the Lua
> >> loader).
> >>
> >> The laptop had no issues using the Forth loader, but the boot process
> >> was ... severely abbreviated ... using the Lua loader.
> >>
> >> I don't have a serial consoloe on the laptop, so I snapped a photo; it
> >> may be found at
> >> <http://www.catwhisker.org/~david/FreeBSD/head/r337834/BYX_halted.jpg>.
> >>
> >> As the name suggests, the final message was:
> >>
> >> BTX halted
> >>
> >> at which point, I found no way to get the laptop to respond to anything
> >> but a power cycle.
> >
> > Interesting. It seems to have died in the middle of loading a
> > module... any insight into what that module may have been?
> >
> > Any way to hack beastie_disable="YES" into loader.conf(5) and see if
> > that changes things?
> >
>
> To be clear- the test I'm interested in will not be productive unless
> beastie_disable="YES" is present in loader.conf or the environment
> before lualoader engages at all.
You may need to build a smaller loader_lua. If so please see
https://reviews.freebsd.org/D16724 and apply and rebuild. The good news is
that you only need to rebuild src/stand, which is fast.
Warner
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CANCZdfqJDAWzHCgGctPv-Ao=-2P4Uwz95DxLQKp=7nkoxZ7MSw>