[RFH] Please test current code on your devices before coreboot 4.6 release

[Paul Menzel]

Dear coreboot folks,

coreboot 4.6 is planned to be released on Monday, so please take ten
minutes, and build the current master branch for your board, flash it,
boot it, and upload the status to the board status repository.

And if you notice a regression, please send a message to the list or
create an issue in the issue tracker [1].

Currently it’s unknown if on Lenovo laptops [2], the TPM is still
works, when TPM support is selected in Kconfig. Please note, that there
were two regressions in the Linux Kernel, so that you should not use
Linux 4.9 or the Linux 4.11 release candidates for testing.

If somebody tested the different QEMU targets that’d be great too.

Thanks,

Paul

[1] https://ticket.coreboot.org/
[2] https://review.coreboot.org/10411

[persmule]

在 2017年04月14日 07:48, Paul Menzel via coreboot 写道:
> And if you notice a regression, please send a message to the list or
> create an issue in the issue tracker [1].

On a0729d9a56, if the first boot after flash is done on dock 433615W
(http://www.thinkwiki.org/wiki/ThinkPad_Dock_Series_3#ThinkPad_Series_3_Docki...),
it will highly possibly fail to light the screen when trying to resume
from suspension. Besides, if usb 3.0 ports are used before suspension,
the screen nearly always fails to resume after suspension.

The ME region has been cleansed and shrinked to minimal size with
me_cleaner (0ac4b4bfd).

My board status has been uploaded via commit 201b7eeb0 of its own
repository.

The tpm on my x230 seems fine.

[Paul Menzel]

Dear persmule,

Am Freitag, den 14.04.2017, 21:59 +0800 schrieb persmule:
> 在 2017年04月14日 07:48, Paul Menzel via coreboot 写道:
> > And if you notice a regression, please send a message to the list or
> > create an issue in the issue tracker [1].

Thank you for testing coreboot on the Lenovo X230.
> On a0729d9a56, if the first boot after flash is done on dock 433615W
> (http://www.thinkwiki.org/wiki/ThinkPad_Dock_Series_3#ThinkPad_Series_3_Docki...),
> it will highly possibly fail to light the screen when trying to resume
> from suspension.

Interesting. Please report that to the issue tracker. It’d be great if
you could get the Linux messages during resume with `drm.debug=0xe`.
> Besides, if usb 3.0 ports are used before suspension,
> the screen nearly always fails to resume after suspension.

I never heard of that. An issue with the corresponding logs would be
helpful too.
> The ME region has been cleansed and shrinked to minimal size with
> me_cleaner (0ac4b4bfd).
>
> My board status has been uploaded via commit 201b7eeb0 of its own
> repository.

Thank you!
> The tpm on my x230 seems fine.

Interesting. Linux 4.9.13 and 4.9.18 seem to be able to detect the TPM
on the Lenovo X230.

```
+[    5.155247] tpm_tis 00:06: 1.2 TPM (device-id 0x0, rev-id 78)
+[    5.259246] tpm tpm0: A TPM error (6) occurred attempting to read a pcr value
+[    5.259291] tpm tpm0: TPM is disabled/deactivated (0x6)
```

`device-id 0x0` looks strange though, and the Linux driver says it’s
deactivated.

I wouldn’t think that’s expected, but I am not sure about that.

Thanks,

Paul

[Marek Behun]

> The ME region has been cleansed and shrinked to minimal size with
> me_cleaner (0ac4b4bfd).

You have also shrinked the ME region? What is it's size now? I would
like to have more space for the payload, so if I could shrink the ME
region from current 5 MB, that would be great. How can this be done?

Marek

[persmule]

Marek,

Your could check the newist me_cleaner
(https://github.com/corna/me_cleaner/). I will attach my flash layout
file to ease your hacking.

Persmule

在 2017年04月14日 22:59, Marek Behun 写道:
> > The ME region has been cleansed and shrinked to minimal size with
> > me_cleaner (0ac4b4bfd).
>
> You have also shrinked the ME region? What is it's size now? I would
> like to have more space for the payload, so if I could shrink the ME
> region from current 5 MB, that would be great. How can this be done?
>
> Marek

[persmule]

Hi Dima:

Currently config for T400 could mostly be used for T400s, except
UDELAY_IO should be used instead of the default UDELAY_LAPIC, otherwise
EC will be unable to access during power cycle, but become accessible
after a reboot.

Have you observed any similar phenomina?

Besides, T400s has sata 5 and 6 as Esata on its back and dock.

Persmule

在 2017年04月14日 07:48, Paul Menzel via coreboot 写道:
> Dear coreboot folks,
>
> coreboot 4.6 is planned to be released on Monday, so please take ten
> minutes, and build the current master branch for your board, flash it,
> boot it, and upload the status to the board status repository.
>
> And if you notice a regression, please send a message to the list or
> create an issue in the issue tracker [1].
>
> Currently it’s unknown if on Lenovo laptops [2], the TPM is still
> works, when TPM support is selected in Kconfig. Please note, that there
> were two regressions in the Linux Kernel, so that you should not use
> Linux 4.9 or the Linux 4.11 release candidates for testing.
>
> If somebody tested the different QEMU targets that’d be great too.
>
> Thanks,
>
> Paul
>
> [1] https://ticket.coreboot.org/
> [2] https://review.coreboot.org/10411

[persmule]

Hi Paul,

Today I found that on my thinkpad x230
(https://review.coreboot.org/cgit/board-status.git/tree/lenovo/x230/4.5-1640-...),
TPM remains in deactivated state after boot, and there is no config menu
item in seabios' menu, while my t420's
(https://review.coreboot.org/cgit/board-status.git/tree/lenovo/t420/4.5-1640-...)
TPM works fine. Their TPM Kconfig seem same.

Besides, I have also tried an OEM bios with ME cleansed just like my
coreboot build on my x230, and TPM works fine on it.

It seems TPM init and activate code breaks a little on x230.

Persmule

在 2017年04月14日 07:48, Paul Menzel via coreboot 写道:
> Dear coreboot folks,
>
> coreboot 4.6 is planned to be released on Monday, so please take ten
> minutes, and build the current master branch for your board, flash it,
> boot it, and upload the status to the board status repository.
>
> And if you notice a regression, please send a message to the list or
> create an issue in the issue tracker [1].
>
> Currently it’s unknown if on Lenovo laptops [2], the TPM is still
> works, when TPM support is selected in Kconfig. Please note, that there
> were two regressions in the Linux Kernel, so that you should not use
> Linux 4.9 or the Linux 4.11 release candidates for testing.
>
> If somebody tested the different QEMU targets that’d be great too.
>
> Thanks,
>
> Paul
>
> [1] https://ticket.coreboot.org/
> [2] https://review.coreboot.org/10411