Date: Wed, 27 Nov 2019 09:48:37 -0500 From: Tom Pusateri <pusateri@bangj.com> To: Tom Pusateri <pusateri@bangj.com> Cc: Ian Lepore <ian@freebsd.org>, freebsd-arm@freebsd.org Subject: Re: cubox-i armv7 12.1-RELEASE image fails to boot Message-ID: <15ABD54F-4193-4413-8045-061A6AEA98A7@bangj.com> In-Reply-To: <02866BFF-03B3-4335-B42B-EB71453CB516@bangj.com> References: <E5E65FF1-C2ED-45B1-AD76-2C9E3CB60468@bangj.com> <26c21aefb0fbc4f97046f8d966d3cec5d9311885.camel@freebsd.org> <02866BFF-03B3-4335-B42B-EB71453CB516@bangj.com>
next in thread | previous in thread | raw e-mail | index | archive | help
> On Nov 27, 2019, at 12:05 AM, Tom Pusateri via freebsd-arm = <freebsd-arm@freebsd.org> wrote: >=20 >=20 >> On Nov 26, 2019, at 10:24 PM, Ian Lepore <ian@freebsd.org> wrote: >>=20 >> On Mon, 2019-11-25 at 21:37 -0500, Tom Pusateri via freebsd-arm = wrote: >>> I tried booting the 12.1-RELEASE image on my cubox-i and it goes = into >>> an endless loop before it boots the kernel but it scrolls so fast >>> it=E2=80=99s hard to make out or figure out how to stop it. >>>=20 >>> The last line I can read before it hits the endless loop is: >>>=20 >>> Loading /boot/loader.conf.local >>> / >>>=20 >>> If I freeze frame a slow mo video of the endless loop it repeats: >>>=20 >>> Code: b1116931 f0074806 4806fea4 fea1f007 (e7cf682d) >>> data abort >>> pc : [<8ffafd74>] lr : [<0ffafd2f>] >>> reloc pc : [<1782bd74>] lr : [<1782bd2f>] >>> sp : 8f57ae38 ip : 8f5a fp : 8ffc2f1c >>> r10 : 8ffbb240 r9 : 8f58 r8 : 8ffafd74 >>> r7 : 8ffd06c4 r6 : 8e55 r5 : ffffffff r4 : 8f8a6318 >>> r3 : 9ff01ce5 r2 : 9ff0 r1 : 8ffbb240 r0 : 0000005e >>> Flags : nzCv IRQs on FIQs on Mode SVC_32 >>>=20 >>>=20 >>> This unit is a couple of years old now but still works fine with >>> Linux. >>>=20 >>> Thanks, >>> Tom >>>=20 >>=20 >> Hmmm, I just downloaded the image, burned it to a fresh sdcard, and = my >> cubox boots from it fine. I'm using a serial console, not video, = maybe >> that's related somehow. >>=20 >> Yep, that's what it is... I dug out a monitor and hooked it up, and >> that gives me the same symptoms you're seeing. >>=20 >> The problem is, I have no idea what to do about that. Those register >> dumps are printed by uboot's exception handlers, and this happens = while >> the efi loader is loading the kernel. I don't know much about efi or >> about running arm systems with video monitors connected. >>=20 >> =E2=80=94 Ian >>=20 >=20 > Thanks! >=20 > unplugging the keyboard/display worked for me too. This is fine for me = but if I can help test any changes, I=E2=80=99ll be glad to. I had to force install an updated version of pkg because the = 12.1-RELEASE version depended on an old version of libssl.so.9. I used: pkg-static bootstrap -f And there=E2=80=99s still some packages that want the old version of = openssl in the =E2=80=9Clatest=E2=80=9D pkg repository: $ curl ld-elf.so.1: Shared object "libssl.so.9" not found, required by = "libcurl.so.4=E2=80=9D $ ls /usr/lib/libssl* /usr/lib/libssl.a /usr/lib/libssl.so.111 /usr/lib/libssl.so /usr/lib/libssl_p.a I=E2=80=99m rebuilding curl from ports now. Thanks again, Tom
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?15ABD54F-4193-4413-8045-061A6AEA98A7>