Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Bump coreboot 4.13 based boards to 4.18 #1234

Closed
wants to merge 1 commit into from
Closed

Bump coreboot 4.13 based boards to 4.18 #1234

wants to merge 1 commit into from

Conversation

tlaurion
Copy link
Collaborator

@tlaurion tlaurion commented Nov 3, 2022 •
edited
Loading

No config change on that one. Just a coreboot version bump for all boards that were pinned to coreboot 4.13

Logic on that is that a lot of coreboot defaults are now under defconfig for unknown reasons from me. Copying actual x230-hotp-maximized coreboot config over 4.18/.config, make savedefconfig and comparing with saved in tree defconfig results in:

user@heads-tests:~/heads/build/x86/coreboot-4.18$ diff defconfig ../../../config/coreboot-x230-hotp-maximized.config
8d7
< CONFIG_CONSOLE_CBMEM_BUFFER_SIZE=0x80000
10,12d8
< CONFIG_PCIEXP_HOTPLUG_BUSES=8
< CONFIG_PCIEXP_HOTPLUG_MEM=0x800000
< CONFIG_PCIEXP_HOTPLUG_PREFETCH_MEM=0x10000000
14a11
> CONFIG_UART_PCI_ADDR=0
18,21d14
< CONFIG_PCIEXP_HOTPLUG_IO=0x2000
< CONFIG_SUBSYSTEM_VENDOR_ID=0x0000
< CONFIG_SUBSYSTEM_DEVICE_ID=0x0000
< CONFIG_I2C_TRANSFER_TIMEOUT_US=500000
23a17
> CONFIG_CONSOLE_CBMEM_BUFFER_SIZE=0x80000

Only CONFIG_CONSOLE_CBMEM_BUFFER_SIZE=0x80000 changed position in config file.

Which no other config parameters there changed compared to defaults, so those are simply not written back under config/coreboot* as of now. And saving them into non-savedefconfig should be subject to discussion under another issue, where past discussions were advertising into keeping configs under tree as savedefconfig output.
So I'm following this without agreeing for the moment.

Activating other new features that happened since 4.13, like BOM, CBFS file measurements (not TPM but bootblock based) or on boot memory wipe (needs additional boot time, untested) will need additional discussions with the community prior of being added. Personally, I would addmemory wiping on boot as default for all boards supporting it.

Testing that was done

Other platforms tested:

@tlaurion
Copy link
Collaborator Author

tlaurion commented Nov 3, 2022 •
edited
Loading

Time to shine. Please tick when tested successfully and leave a comment below (please do not test if you do not have an external reprogrammer)

Instructions to download ROMs are under https://osresearch.net/Downloading from the green mark above near commit ID (or choose your board's builds here: https://app.circleci.com/pipelines/github/tlaurion/heads/1245/workflows/e4753fab-b47d-41d7-9d20-41d468b849dd):

@natterangell
Copy link
Contributor

natterangell commented Nov 3, 2022 •
edited
Loading

  • t420 (hotp-maximized) 95389c1 works fine.

EDIT: Suspend/resume work without issues.

@tlaurion
Copy link
Collaborator Author

tlaurion commented Nov 3, 2022 •
edited
Loading

@natterangell and everyone else testing, please report any regression (suspend/resume issues or any other).

@tlaurion
Copy link
Collaborator Author

tlaurion commented Nov 3, 2022 •
edited
Loading

I would also be curious of anyone flashing and testing legacy boards with specific commands, just to have an idea of who is still using them if possible (non-maximized boards THAT CANNOT BE USED INTERCHANGEABLY WITH MAXIMIZED BOARDS, PLEASE READ THE DOCS)

@weyounsix
Copy link

weyounsix commented Nov 5, 2022 •
edited
Loading

both pass. tested totp, tpm reset, basic QubesOS installation, disk unlock sealing, suspend/resume
no issues.

@srgrint
Copy link
Contributor

srgrint commented Nov 7, 2022

x220 tested (heads-x220-maximized-v0.2.0-1268-g95389c1.rom) - no issues (tested totp, tpm reset, suspend/resume)

@tlaurion tlaurion mentioned this pull request Nov 8, 2022
Closed
@weyounsix
Copy link

i think @eganonoa is a dead account-- says it is (old) on matrix too. so unless we can track them down or recruit new testers, this is likely all the testing that we will be able to do.

@tlaurion tlaurion force-pushed the pass_coreboot413_coreboot418 branch from 95389c1 to 6f16b2a Compare November 15, 2022 20:49
@tlaurion
Copy link
Collaborator Author

tlaurion commented Nov 16, 2022 •
edited
Loading

Using x230-hotp-maximized on daily driver from https://app.circleci.com/pipelines/github/tlaurion/heads/1268/workflows/dcf334f0-47b6-40e7-9920-15a30dbcd4ca/jobs/12420/artifacts

(this PR on top of master + #1206)

Not sure how to handle untested platforms. Just asked on Heads slack/matrix for t530/t520 board owners with external reprogrammer to test

@tlaurion tlaurion mentioned this pull request Nov 16, 2022
Merged
@tlaurion
Copy link
Collaborator Author

@weyounsix @srgrint @natterangell : using on daily driver?

@srgrint
Copy link
Contributor

srgrint commented Nov 16, 2022

Yes - I have been using heads-x220-maximized-v0.2.0-1268-g95389c1.rom with Debian on an x220 as a daily driver for the last 9 days. No issues so far.

@natterangell
Copy link
Contributor

natterangell commented Nov 16, 2022 via email

@weyounsix
Copy link

weyounsix commented Nov 17, 2022 •
edited
Loading

i am not using on daily driver-- but my daily driver is now an x230, which i believe you already have covered @tlaurion

@tlaurion
Copy link
Collaborator Author

Edited OP with actual testers. t530 and t520 tests still missing from actual board owners.

@tlaurion
Copy link
Collaborator Author

tlaurion commented Nov 18, 2022 •
edited
Loading

@githubisnonfree would you be willing to test on t530/t520? Unfortunately, we seem to have lost @eganonoa to confirm no regression against coreboot 4.13 (should not be buggy since their sibling boards are tested and had no regression)

That would be top+bottom(external flash)/full roms (not top/bottom rom) produced from this PR for t530, and full ROM for t520 :

Let me know! Otherwise, I will have to rename those boards as untested or something...

@jnscmns
Copy link

jnscmns commented Dec 22, 2022

Time to shine. Please tick when tested successfully and leave a comment below (please do not test if you do not have an external reprogrammer)

Instructions to download ROMs are under https://osresearch.net/Downloading from the green mark above near commit ID (or choose your board's builds here: https://app.circleci.com/pipelines/github/tlaurion/heads/1245/workflows/e4753fab-b47d-41d7-9d20-41d468b849dd):

Apologies for the late reply, but I didnt have the external reprogrammer available at the time.
Upgraded on a w530 k1000m - Despite having to re-define to dual graphics with nvramtool, all tests passed.
Cheers

@weyounsix
Copy link

daily driving from this branch now, and very happy with it. on the master branch i had an issue with clock drift on sleep, but this is working flawlessly.

@bwachter bwachter mentioned this pull request Jan 8, 2023
Closed
@rbreslow
Copy link
Contributor

So I'm following this without agreeing for the moment.

Not sure if you've also picked up on this. Despite these values being written via make savedefconfig, they haven't actually deviated from their defaults. e.g., with SUBSYSTEM_VENDOR_ID:

https://github.com/coreboot/coreboot/blob/f3a672908fd27dcf48e24d939dbe8f072843e1e0/src/device/Kconfig#L816-L825

So this may be a regression w/ Coreboot?

@tlaurion
Copy link
Collaborator Author

Superseded with #1312

@tlaurion tlaurion closed this Feb 18, 2023
@tlaurion tlaurion mentioned this pull request Mar 2, 2023
Closed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
6 participants
@tlaurion @natterangell @weyounsix @srgrint @jnscmns @rbreslow