Skip to content

m900 tower port + TPM1 counter auth and defend lock fixes#2118

Open
tlaurion wants to merge 15 commits into
masterfrom
m900_with_tpm1_fixes
Open

m900 tower port + TPM1 counter auth and defend lock fixes#2118
tlaurion wants to merge 15 commits into
masterfrom
m900_with_tpm1_fixes

Conversation

@tlaurion
Copy link
Copy Markdown
Collaborator

@tlaurion tlaurion commented May 14, 2026

Supersedes #2111, rebased on #2117 (tpm1_fixes).

This PR merges the m900 tower board port with the TPM1 fixes from PR #2117.

Board port (notgivenby's commits):

  • Board configs for m900 tower (maximized and hotp-maximized variants)
  • Coreboot config, linux config, ME blob handling
  • Review comments addressed: typos, blob names, board name includes 'tower'
  • CircleCI integration for both variants (requires seed:coreboot-25.09)
  • CircleCI x86_blobs step added for m900 ME download/deguard
  • Blob download script refactored to use shared blobs/lib.sh

TPM1 fixes from PR #2117 (included via rebase on tpm1_fixes):

@notgivenby so you can test artifacts.

Copilot AI review requested due to automatic review settings May 14, 2026 19:52
Copy link
Copy Markdown

Copilot AI left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Pull request overview

This PR ports the Lenovo M900 Tower (Skylake/Kaby Lake LGA1151 mini-tower) to Heads with two board variants (maximized, hotp-maximized), plus targeted TPM1 reliability fixes in tpmr.sh so that auth-failure detection and tpm1_reset() recover from the TPM_DEFEND_LOCK_RUNNING state after multiple bad passphrases. Documentation in doc/tpm.md is expanded with TPM1 vs TPM2 error-stream conventions, auth grep patterns, and the defend-lock recovery flow.

Changes:

  • New EOL_m900_tower-{maximized,hotp-maximized} boards with shared coreboot/linux configs and an ME blob preparation pipeline (download → me_cleaner → deguard).
  • TPM1 auth grep patterns extended to include defend/0x98e/0x149; tpm1_reset() cycles physical presence on defend lock running after forceclear.
  • CircleCI: two new build jobs (depend on the existing EOL_t480-hotp-maximized 25.09 seed).

Reviewed changes

Copilot reviewed 10 out of 14 changed files in this pull request and generated 5 comments.

Show a summary per file
File Description
boards/EOL_m900_tower-maximized/EOL_m900_tower-maximized.config New maximized board config (no HOTP).
boards/EOL_m900_tower-hotp-maximized/EOL_m900_tower-hotp-maximized.config New hotp-maximized variant.
config/coreboot-m900-maximized.config Coreboot 25.09 config for Lenovo M900.
config/linux-m900.config Linux 6.1.8 kernel config for the board.
targets/m900_me_blobs.mk Make rules tying the ME blob script into the board build.
blobs/m900/m900_download_clean_deguard_me.sh Downloads ASRock BIOS, neuters/deguards ME 11.6.0.1126.
blobs/m900/README.md Blob layout, sources and integrity notes.
blobs/m900/hashes.txt SHA256 of ME/IFD/GBE blobs.
blobs/m900/.gitignore Ignores generated me.bin/m900_me.bin.
initrd/bin/tpmr.sh Adds defend-lock detection in auth-retry grep and tpm1_reset() recovery sequence.
doc/tpm.md New sections on TPM1/TPM2 auth error patterns and defend-lock recovery.
.circleci/config.yml Adds the two M900 board build jobs (depending on the 25.09 seed).

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Comment thread blobs/m900/m900_download_clean_deguard_me.sh Outdated
Comment thread doc/tpm.md Outdated
Comment thread doc/tpm.md
Comment thread blobs/m900/README.md Outdated
Comment thread blobs/m900/README.md Outdated
@tlaurion tlaurion force-pushed the m900_with_tpm1_fixes branch 2 times, most recently from 78366db to a9cd1fc Compare May 14, 2026 20:13
@tlaurion tlaurion requested a review from Copilot May 14, 2026 20:18
Copy link
Copy Markdown

Copilot AI left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Pull request overview

Copilot reviewed 10 out of 14 changed files in this pull request and generated 5 comments.

Comment thread blobs/m900/m900_download_clean_deguard_me.sh Outdated
Comment thread blobs/m900/m900_download_clean_deguard_me.sh
Comment thread blobs/m900/README.md Outdated
Comment thread blobs/m900/hashes.txt
Comment thread blobs/m900/.gitignore Outdated
Copy link
Copy Markdown

Copilot AI left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Pull request overview

Copilot reviewed 10 out of 14 changed files in this pull request and generated no new comments.

@tlaurion tlaurion force-pushed the m900_with_tpm1_fixes branch 3 times, most recently from 863c9b7 to 6919d44 Compare May 14, 2026 20:52
@tlaurion tlaurion requested a review from Copilot May 14, 2026 20:53
Copy link
Copy Markdown

Copilot AI left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Pull request overview

Copilot reviewed 10 out of 14 changed files in this pull request and generated no new comments.

@notgivenby
Copy link
Copy Markdown
Contributor

debug.log
measuring_trace.log

the fix did not help.

PR #2068 (tpm_reseal_ux-integrity_report-detect_disk_and_tpm_swap,
merged at d3d8053) changed increment_tpm_counter from hardcoded
-pwdc '' (empty counter auth) to -pwdc "${tpm_passphrase:-}" (owner
passphrase from cache/prompt), but left check_tpm_counter using empty
-pwdc when called from kexec-sign-config.sh without a $3 passphrase
argument.  This caused every counter increment to compute
SHA1(owner_pass) while the counter was created with SHA1("") -
persistent TPM_AUTH_FAIL.

Per TCG TPM Main Spec Part 3, TPM_CreateCounter uses owner auth
(-pwdo) but TPM_IncrementCounter uses the counter's own authData,
not the owner password.  The correct design for Heads' rollback
counter is empty auth: rollback security comes from the signed
/boot/kexec_rollback.txt and TPM sealing, not counter access control.

The repeated auth failures (3 per boot x ~5 boots via the
_tpm_auth_retry loop) triggered TPM 1.2 dictionary-attack lockout
(TPM_DEFEND_LOCK_RUNNING), which persisted through forceclear on
some implementations, causing tpm takeown to fail and TPM reset to
abort - a cascade failure from the counter auth mismatch.

Changes:
- initrd/bin/tpmr.sh (_tpm_auth_retry, tpm2_counter_inc, tpm2_seal,
  tpm1_seal): add 'defend' and '0x98e|0x149' to auth detection grep
  patterns so defend lock and TPM2 RC codes are treated as retryable
  auth failures rather than fatal errors
- initrd/bin/tpmr.sh (tpm1_reset): detect defend lock after takeown
  failure and cycle physical presence to clear the lock state before
  retrying; full AC power cycle remains the fallback if software
  presence is insufficient
- initrd/bin/tpmr.sh (tpm1_counter_increment): detect -pwdc '' and
  call tpm directly, bypassing _tpm_auth_retry which injected the
  owner passphrase.  Use || return to survive set -e on expected
  auth failure.
- initrd/etc/functions.sh (check_tpm_counter): pass -pwdc '' instead
  of -pwdc "${tpm_passphrase:-}" so counters use SHA1("") per TCG
  spec.  Document that $3 is intentionally ignored.
- initrd/etc/functions.sh (increment_tpm_counter): try -pwdc '' first
  for TPM1.  If that fails on a readable counter (created by PR #2068
  era code), prompt for owner passphrase and retry as migration
  fallback with clear WARN explaining the one-time migration and
  TPM reset option.
- initrd/etc/functions.sh (increment_tpm_counter): remove the
  TPM1-specific owner-passphrase prompt block added by PR #2068
- initrd/etc/functions.sh (increment_tpm_counter): DIE-path fallback
  counter_create: -pwdc '' for consistency
- initrd/bin/oem-factory-reset.sh: counter_create -pwdc '' for
  consistency with the empty-auth design
- doc/tpm.md: document TPM1 boot chain, tpmtotp tool selection,
  auth retry patterns, defend lock recovery, and physical presence

Signed-off-by: Thierry Laurion <insurgo@riseup.net>
@tlaurion tlaurion force-pushed the m900_with_tpm1_fixes branch from 6919d44 to 9d1e115 Compare May 16, 2026 01:39
@tlaurion tlaurion changed the title m900 tower port + TPM1 auth/defend lock fixes m900 tower port + TPM1 counter auth and defend lock fixes May 16, 2026
@tlaurion
Copy link
Copy Markdown
Collaborator Author

debug.log measuring_trace.log

the fix did not help.

went down the rabbit hole and found the regression and created fix pushed onto #2117 and rebased on top of it.

@tlaurion tlaurion force-pushed the m900_with_tpm1_fixes branch 2 times, most recently from 31e8838 to 3e31d39 Compare May 16, 2026 01:50
Signed-off-by: notgivenby <notgivenby@gmail.com>
Signed-off-by: Thierry Laurion <insurgo@riseup.net>
Signed-off-by: notgivenby <notgivenby@gmail.com>
Signed-off-by: Thierry Laurion <insurgo@riseup.net>
Signed-off-by: notgivenby <notgivenby@gmail.com>
Signed-off-by: Thierry Laurion <insurgo@riseup.net>
Signed-off-by: notgivenby <notgivenby@gmail.com>
Signed-off-by: Thierry Laurion <insurgo@riseup.net>
Signed-off-by: notgivenby <notgivenby@gmail.com>
Signed-off-by: Thierry Laurion <insurgo@riseup.net>
notgivenby and others added 8 commits May 15, 2026 21:58
Signed-off-by: notgivenby <notgivenby@gmail.com>
Signed-off-by: Thierry Laurion <insurgo@riseup.net>
Signed-off-by: notgivenby <notgivenby@gmail.com>
Signed-off-by: Thierry Laurion <insurgo@riseup.net>
Signed-off-by: notgivenby <notgivenby@gmail.com>
Signed-off-by: Thierry Laurion <insurgo@riseup.net>
…900_tower board

Signed-off-by: notgivenby <notgivenby@gmail.com>
Signed-off-by: Thierry Laurion <insurgo@riseup.net>
Signed-off-by: notgivenby <notgivenby@gmail.com>
Signed-off-by: Thierry Laurion <insurgo@riseup.net>
Signed-off-by: notgivenby <notgivenby@gmail.com>
Signed-off-by: Thierry Laurion <insurgo@riseup.net>
- blobs/m900/README.md: fix blob filenames, spelling (paritally->partially,
  Unfourtunatly->Unfortunately, layot->layout)
- blobs/m900/m900_download_clean_deguard_me.sh: fix Dell->ASRock comment
- boards/EOL_m900_tower-*: fix m900_tiny->m900_tower, fix ME script path,
  add 'tower' to CONFIG_BOARD_NAME
- targets/m900_me_blobs.mk: rewrite header with accurate instructions

Signed-off-by: Thierry Laurion <insurgo@riseup.net>
Signed-off-by: Thierry Laurion <insurgo@riseup.net>
m900 ME blob download/deguard script was not wired into the
x86_blobs CI job.  Add it after the xx80 steps, following the
same pattern as other board families.

Signed-off-by: Thierry Laurion <insurgo@riseup.net>
@tlaurion tlaurion force-pushed the m900_with_tpm1_fixes branch from 8838b18 to 3ef9c50 Compare May 16, 2026 02:13
@tlaurion
Copy link
Copy Markdown
Collaborator Author

tlaurion commented May 16, 2026

@notgivenby hopefully this works. Added m900 blobs download in circleci so it downloads it once and reuses cache if already there and checksums match, reworked the script to reuse lib added, and fixups for tpm1.

Keep me posted. Hopefully #2117 (we are based on it here) fixes your issue.

@notgivenby
Copy link
Copy Markdown
Contributor

notgivenby commented May 16, 2026

debug.log
measuring_trace.log

You invested really a lot of time into debugging...I do not want to misuse you and your time here. Please let me known if @tlaurion you think we need to stop the attempts to port let us say not most popular board for only few people. The issue still maybe releveant for other ports who knows. For future attemps, perhaps I need to select a desktop board from other vendor perhaps with tpm2?

@tlaurion
Copy link
Copy Markdown
Collaborator Author

debug.log measuring_trace.log

You invested really a lot of time into debugging...I do not want to misuse you and your time here. Please let me known if @tlaurion you think we need to stop the attempts to port let us say not most popular board for only few people. The issue still maybe releveant for other ports who knows. For future attemps, perhaps I need to select a desktop board from other vendor perhaps with tpm2?

debug.log measuring_trace.log

You invested really a lot of time into debugging...I do not want to misuse you and your time here. Please let me known if @tlaurion you think we need to stop the attempts to port let us say not most popular board for only few people. The issue still maybe releveant for other ports who knows. For future attemps, perhaps I need to select a desktop board from other vendor perhaps with tpm2?

@notgivenby this has nothing specific to m900, outside of the fact aht as opposed to tpm2, we do not configure the dictionary attack config (how many attempt per timeframe nor resolution period).

As said under #2117 , master contains a regression for how we do counter create and how we increment the counter.
Your platform has DA lockout, might be 24h before automatic resolution. i've worked a bit on it today but haven't finished ading code to troubleshot this properly. Maybe tomorrow, ost probably monday

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants