AST2400 x86 bringup LPC problems

[Tao Ren]

Looks like you are facing the same/similar issue mentioned in below page?

https://www.anandtech.com/show/11110/semi-critical-intel-atom-c2000-flaw-discovered


- Tao

From: openbmc <openbmc-bounces+taoren=fb.com at lists.ozlabs.org> on behalf of Aaron <aaron_ppus at fmad.com>
Date: Thursday, July 12, 2018 at 7:35 AM
To: OpenBMC Maillist <openbmc at lists.ozlabs.org>
Subject: Re: AST2400 x86 bringup LPC problems

Per usual found the issue after spamming the list. The LCLK was disabled because lpc_ctrl was not setup correctly and error`ed out which cleans up and disables the lclk entirely... doh.

Aaron

On 12 July 2018 at 16:47, Aaron <aaron_ppus at fmad.com<mailto:aaron_ppus at fmad.com>> wrote:
Hi,

Currently stuck bringup an x86 server board with the AST2400 and looking for advice.

Have based the code on the Quanta machine, theres a bunch of dst changes and different gpio / i2c etc mappings. Most things work now, power control, can control, gpios are good etc however the LPC is broken.

What I cant working out is why the LPC interface is completely dead.

Symptoms are:
- LPC post code sniffer caches nothing
- Host UART is completely missing
- Host IO port range reading returns 0xff`s

Its as if the LPC block is not being reset or theres no clock or being muxed out or something. Been a few days now and still cant work it out.

Originally thought it was because Strap[14] = 1  (LPC reset not connected to PCI reset) but on the board they physically routed PERST to the to the pin, so it should be ok.

Any suggestions or advice ?

Thanks
Aaron



-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ozlabs.org/pipermail/openbmc/attachments/20180712/394265c6/attachment.html>