Re: Strange USB boot problem, Lenovo T450S

Marius Gedminas <[email protected]> Tue, 30 Mar 2021 22:08:28 +0300
Newsgroups gmane.linux.hardware.thinkpad
Message-ID <20210330190828.bxhjqygbpn5qldbm@blynas>
On Tue, Mar 30, 2021 at 03:19:17PM +0100, Paul Marwick wrote:
> I've (quite accidentally) stumbled on a strange problem with USB booting on
> this machine. It seems that there are some flash drives that the machine
> will not boot from. For instance, I have a 4GB Verbatim Store 'n Go which
> has MX 19.3 on it. That works fine, boots, allows me to install. I have
> another which allowed me to install Windows 10. But there are some disks
> which don't work at all.
> 
> I have several Integral flash drives, one quite old, some recently bought.
> The T450 won't boot from any of them. If I try, it accesses the flash drive,
> the screen goes blank and I might just as well switch it off. Can't use
> CTRL-ALT-DEL no response. Power switch is the only option. Applies even if I
> write the flash drive on the T450. But, the same flash drive will boot any
> other machine I've tried it on.

Hmm, my X220 does the same thing if an external 4 TB Intenso HDD is
plugged in.  It doesn't matter that it's not included in the boot list
in the BIOS, merely having it plugged in makes the BIOS freeze during
boot.  Black screen, no reaction to any keys, fans spinning up to full
speed after a couple of minutes -- smells like an infinite busy loop
somewhere in the firmware.  A long power button press is the only way
out.

If I get a boot menu (F12), then whenever I select any device, the same
thing happens, no matter which boot device I choose.

I haven't seen this with any other external USB HDD, and I've had a few.

I'd love to upgrade the X220's firmware, but the hardware is not
supported by fwupd, I don't have a Windows install, and extracting
bootable floppy images from firmware upgrade ISOs is too much of a
hassle.  I did check the firmware changelog and it didn't mention
anything about boot problem fixes.

Marius Gedminas
-- 
I would suggest re-naming "rmbdd()". I _assume_ that "dd" stands for "data
dependent", but quite frankly, "rmbdd" looks like the standard IBM "we
lost every vowel ever invented" kind of assembly lanaguage to me.
		-- Linus Torvalds

-- 
The linux-thinkpad mailing list home page is at:
http://mailman.linux-thinkpad.org/mailman/listinfo/linux-thinkpad
signature.asc (application/pgp-signature, 195 B)
-----BEGIN PGP SIGNATURE-----

iF0EABECAB0WIQSBIa0y8AqAlHSKbNCRV0Rd56bXjwUCYGN3JwAKCRCRV0Rd56bX
j2gJAJoC5T5Yuod+9E51SnKZLbWTbkZ2kgCeInrs7OSb06QAOnYHWL/va5W3dG4=
=LwRM
-----END PGP SIGNATURE-----