If an installed OS is not detected, then skip setting the
default boot device or generating /boot checksums.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
Check for presence of CONFIG_BOOT_DEV, and if missing or
unable to be mounted, present the user with a menu offering the
option to select another disk, boot from USB, continue to
main menu, or drop to a recovery shell.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
Set and export currently-used defaults in gui-init, but still
allow for inidividual boards to override via config if desired.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
Strip the path prefix from the ROM filename, and place on own
line to prevent truncation with long filenames / narrow screens.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
This will allow it to be used elsewhere within the UI.
Rename CONFIG_BOOT_GUI_MENU_NAME to better indicate use/function.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
* modules/coreboot: add option to use coreboot 4.11
Port patches from coreboot 4.8.1 to 4.11:
* 0000-measure-boot -> 0001
* 0010-cross-compiler-support
All other patches for coreboot 4.8.1 have either already been
integrated, or are for platforms which do not need to be migrated
to coreboot 4.11 (they will move to 4.12 or newer).
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
* patches/coreboot-4.11: Add Broadwell-DE platform patch
Add a patch for FSP Broadwell-DE to make use of Heads' measured boot.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
* patches/coreboot-4.11: Add patch to read serial # from CBFS
Will be used by multiple Librem boards.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
* patches/coreboot-4.11: add board support for Librem Server L1UM
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
* Librem Server L1UM: add new board
Add board config, coreboot config, kernel config files.
Add conditional purism-blobs dependency to coreboot-4.11 module.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
* flash.sh: add special handling for librem_l1um board
Add support for persisting PCIe config via PCHSTRP9 in flash descriptor.
This is needed to support multiple variants of the L1UM server which
use the same firmware but differ in PCIe lane configuration via the
PCH straps configuration in the flash descriptor.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
* patches/coreboot-4.11: Add 'Use PRIxPTR to print uintptr_t' patch
Cherry-picked from upstream coreboot (post-4.11), fixes compilation issue.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
* CircleCI: add target to build board librem_l1um
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
Update_checksum was already changing files in /boot, befor checking for
gpg card. If no card is present, the user will end up in the recovery
next time instead of getting the same dialog again. Therefore, the
confirm_gpg_card should be checked before altering files.
The dead -u flag/$update_counter is used to mark the necessisty to
update the hash files now.
Currently, exiting the recovery shell results in a kernel panic,
necessitating a hard reset / power cycle. As this is less than ideal,
drop the exec and add a loop to restart the shell.
Addresses issue #833
Tested under qemu-coreboot-fbwhiptail
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
Changing the default boot was failing because remove the old entries did
not work as `rm "/some/path/*.txt"` does not work as intended, e.g. the
asterisk is no catch-all.
Some grub configs/bls entries contain the full paths to the
kernel/initrd files, which the parsers currently fail to handle,
causing a failed boot without any useful error being presented to the user.
To fix this, strip the bootdir prefix from the menu entries when parsing,
should it exist.
Test: build/boot Librem 13v2 w/F32 and bls entries containing absolute paths.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
When a new /boot device is selected, wait until after
successfully mounting the newly-selected device before
updating CONFIG_BOOT_DEV.
Also, don't assume /boot already mounted, as this can cause
a false failure and prevent mounting of the newly-selected device.
Lastly, tidy up the error output in case mounting /boot fails.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
Using 'let' in these scripts fails when evaluating to zero
for some reason, so replace with '$(())' which works as intended.
Test: Boot device selection menu shown properly when
new/unpartitioned drive installed.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
Attempting to reset the TPM when once isn't present causes a kernel
panic, so let's not allow users to do that.
Test: verify 'No TPM Detected' shown on Librem Mini when Reset TPM
option selected from menu.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
-r will always succeed since the file will be generated regardless
of number of boot entries found. Use -s instead to check for zero
file size.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
Since we sort the boot options prior to selecting the new default entry,
we need to use the index of the entry in the list prior to being sorted,
vs always setting it as 1. This fixes setting/booting of the default
OS target where the list entries are changed when calling sort.
Test: perform OEM factory reset with Fedora 32 installed, verify
default boot succeeds followng reset.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
The same grub parsing logic used in kexec-select-boot should
be used here as well, so copy it over.
Test: oem-factory-reset succeeds with Fedora 32 installed.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
With current implementation, Librem Keys with VID 0x316d are
not identified properly; correct the if/else logic to resolve.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
Add a linebreak before showing list of files with changed
checksums. Fix text truncation on checksum update prompt.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
Adjust text on GUI dialogs to prevent filenames from being truncated
and to improve clarity/readability.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
Currently, /media is mounted once per boot, which causes issues
if a user need to change USB sticks, or unknowning performs an
operation that mounts /media and then needs to access a different
USB stick later (eg, updating the firmware).
To mitigate this, always unmount /media if mounted before scanning
for USB devices, so the user can choose the correct device at the
time of its use.
Additionally, add a unique exit code for user abort so we're not
treating it the same as a failure, and use it to prevent unnecessary
GUI prompts when cancelling selection of a USB device.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
For the handful of operations which need to be done with /boot
as the pwd, encapsulate them in subshells to ensure the pwd
doesn't unexpectedly change for other operations, as functions
which need to mount/unmount /boot may fail if the pwd isn't root.
Also, set the pwd to root at the start of detect_boot_device as an
added safety measure.
Test: run oem-factory-reset function, ensure it doesn't fail to
detect boot device due to incorrect working directory.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
Move code duplicated across several GUI scripts into a common
gui_functions file and include/use that.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
Show RAM in GB, since the calculation in MB is imprecise as
it excludes RAM allocated for GPU (eg).
Fix display of firmware version strings which contain spaces by
adjusting cut and simply chopping off the date at the end, which
is a fixed 10-char length.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
Since #758 is merged, users have a option to export GnuPG pubkey if
necessary. Thus, we they do not need to insert a USB drive during
factory reset. Until now the whole process failed just because a user
did not provide a USB drive instead.
This shall be fixed by this commit
If smartcard Nitrokey Storage was factory-reset, we delete AES keys on
it as well.
Explaination: After oem-factory-reset was started the AES on the Nitrokey Storage that is used for the encrypted volume and the password safe is is not usable anymore because the smart card was factory-reset. To make it usable, a user needs to delete it via Nitrokey App. By doing so, the HOTP secret is deleted as well, resulting in a bad warning in Heads. Therefore, we are resetting AES key right after factory-reset with hotp_verification
Add CONFIG_TPM_NO_LUKS_DISK_UNLOCK to allow Librem boards to opt
out of using TPM to store LUKS key, and use it to guard the user
option to add the disk encryption key to the TPM.
Select this option for all Librem boards; all other boards which
select CONFIG_TPM=y will have no change in functionality.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
Rather than wait a fixed 5s for the usb storage kernel modules
to load, and the user to insert a drive, check for new USB drives
inserted every 1s with a 5s timeout.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
mount-usb switched to dynamic USB device detection a while back,
so eliminate instances of CONFIG_BOOT_USB_DEV, and derive the
mounted USB device from /etc/mtab in the one place where it's
actually needed (usb-scan). Clean up areas around calls to
mount-usb for clarity/readability.
Addresses issue #673
Test: Build Librem 13v4, boot ISO file on USB
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
Commit 6b5adcca moved the call to enable_usb from gui-init
to init and guarded it with CONFIG_USB_KEYBOARD, but it was
missed that this is needed for the clean boot check logic
when a librem key is used. Add the call back to gui-init
and guard it properly
Test: clean_boot_detect works properly on a librem 13v4
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
Some (out of tree) servers require use of a USB keyboard, and need
the USB kernel modules loaded prior to checking for keypress to enter
a recovery console. Since loading the modules affects the value in PRC5
and can cause issues putting a LUKS key in TPM, guard the loading of the
USB modules with CONFIG_USB_KEYBOARD and remove the unguarded call from
gui-init.
This should resolve issues #603 and #674.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
Not setting USB_FAILED when call to mount-usb succeeds results
in a spurious 'sh: 0 unknown operand' error printed to console.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
Since the custom password is used to set the GPG admin
password as well as the TPM and GPG user passwords, an
8-character minimum is required. Inform the user of this,
and validate custom password length upon entry.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
If /boot isn't mounted, we can't read the HOTP counter, so no
point in reading from the TPM. This speeds up getting to the
main menu in the case of an inaccessible or non-existant /boot.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
Update text on TOTP error prompt to provide better
guidance for users following the use of the OEM
factory reset function
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
If kexec-sign-config fails due to GPG key not present,
the double die() results in a kernel panic (and if it didn't,
/boot would be left mounted RW). Fix this by removing call to
die() and ensuring /boot remounted RO regardless checksum
update success or failure.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
Normally we resort to default passwords for OEM reset, however we have a
use case where it would be convenient to set a custom password instead.
This patch adds a simple prompt (that defaults to the defaults if you
hit Enter) that enables someone using the OEM reset to enter a single
password that will replace the defaults (TPM, GPG Admin, GPG User).
If the user chooses to flash a "cleaned" ROM (not persisting settings
or GPG keys) then the signatures on /boot are no longer valid, so clear
them out. This allows for the OEM factory reset prompt to be shown on
the next boot.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
Users may wish to temporarily boot an OS from a drive other than
their primary boot drive, without changing the default and saving
to ROM. Mounting /boot after changing the device selection
facilitates this by allowing the user to then choose an unsafe boot
from the newly-selected boot drive.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
when commit [928f003] config-gui: add 'Full Reset' option
was added, the bottom end of the save config option was
accidentally truncated; restore it to fix save config option
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
Add a check to determine if first boot after flashing a cleaned
ROM, and prompt user to run the OEM Factory Reset if so
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
Add an OEM Factory Reset menu option, which performs an
unattended reset and configuration of the device's TPM,
GPG security token, and boot device / boot selection.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
Add oem-factory-reset script which performs an unattended
reset and configuration of the device's TPM, GPG security token,
and boot device / boot selection.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
Add function to detect boot device. Start by checking
CONFIG_BOOT_DEV, then iterate thru all bootable partitions.
Check if partition is mountable, contains grub directory.
Update CONFIG_BOOT_DEV and mount on /boot if successful.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
Initial card detection can sometimes fail, so prompt the user
to remove/reinsert their GPG card before retrying. Since
errexit is likely set, disable it prior to calling --card-status so
we can handle the error ourself, then re-set if necessary when done.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
Next prompt will be to ensure GPG key is attached, which defaults
to Y, so default here as well for consistency
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
Check that any files added to cbfs exist before attempting to
add them, so flashing doesn't fail after a reset.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
Add Full Reset option to clear all GPG keys and user settings,
both from the local filesystem and running firmware, and
clear/reset the TPM
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
If the boot device is invalid or unabled to be mounted
and the user opts to select a new boot device, jump
directly to that menu option and avoid the additional
step of showing the config main menu.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
Updating TOTP/HOTP is mainly necessary after a user updates
their firmware, and no need to force another reboot.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
use similar filtering logic as with USB drives to provide
the user a more sane list of boot device options. Show user
only valid bootable partitions, not block devices.
There's no point in showing /dev/nvme0 and /dev/nvme0n1 (eg)
when /dev/nvme0n1p[1..n] (eg) exist, as the former are not
valid boot devices.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
Add optional parameter to bypass menu selection and
immediately select a menu option. This allows us to call
the 'Set Boot Device' option directly, saving the user
an unnecessary step.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
Introduce option to remove any existing key(s) from firmware
and add a user-suppled key, before reflashing.
Move existing code for adding a new key to a separate function
so it can be reused for new feature without duplication.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
If CONFIG_BOOT_DEV isn't set or otherwise fails to mount,
we prompt the user to change the /boot device, but never
attempt to mount it to ensure it's sane, leading to
potential failures later in the config/boot process.
Ensure that CONFIG_BOOT_DEV is updated after the /boot device
is changed, and attempt to re-mount /boot after the change.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
If the user has changed the /boot config but not yet flashed,
it's reasonable for them to expect that flashing to add/update
a GPG key won't revert those changes.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
In order to streamline the initial setup, prompt user to
update checksums/sign /boot files after adding a key and
updating the firmware, in order to avoid an extra reboot.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
Move the non-GUI component of update_checksums() to
/etc/functions so it can be reused outside of gui-init.
Add check that /boot/kexec_default_hashes.txt exists before parsing
it, since doesn't exist if there's no default boot target set yet.
Eliminates spurious error text and/or premature exit depending on
state of errexit.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
'Options' makes more sense than 'Settings' given the submenu's contents.
Tidy up Boot Options as well
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
Before anything else, ensure that a default boot device
is set. If not, prompt the user to set one. If set, ensure
that /boot can be mounted successfully; else prompt the
user to select a new boot device.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
If /boot isn't mounted, we can't read the HOTP secret, so no
point in reading from the TPM. This speeds up getting to the
main menu in the case of an inaccessible or non-existant /boot,
and maintains the warning condition from not being able to
validate the HOTP.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
Since a USB boot target can't be the default (at least currently,
/boot must be on internal media), skip the extraneous prompt to
set it as such when booting from USB.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
Exclude dot folders from ROM search path, so that files in
.Trash (eg) aren't shown. Sort the remaining options.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
Using sort on USB boot options produces a reverse-ordered list,
leading users to often select the wrong option. Add the -r
parameter to sort to correct the list order and make the default
option the first in the list.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
Because of the way unseal-hotp is called from gui-init,
dropping to a recovery shell when failing to mount /boot
causes it to hang, leaving the user stranded until they
kill it with CTRL+C. Instead, simply return and continue
to the main GUI menu where the user can address the problem.
Rename the function to clarify difference from other versions
of mount_boot() which do drop to the recovery shell.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
Fix HOTP verfication failure if LK admin pin/passphrase contains
spaces by quoting the variables when passed to functions.
Test: set LK admin pin to passphrase with spaces, generate
new TOTP/HOTP, verification passes.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
Reading the file into a variable and then redirecting to stdin
via echo() can cause the binary data to be truncated, leading
to an invalid base32 value and failure to properly generate
and validate the HOTP code.
To resolve this, pass the file directly to hotp(), and ensure
it is removed properly regardless of success or failure to
prevent leakage.
Fixes "Invalid base32 string" error seen when attempting to
generate a new TOTP secret.
Signed-off-by: Matt DeVillier <matt.devillier@puri.sm>
Add a main boot menu entry to power off. This enables users to
only verify the firmware integrity using OTP, and do nothing more.
After having left the device out of sight, one might want to do
a quick sanity check only.
Since we already have a script to safely power down, we make use of
it now.
Instead of relying on a hard-coded USB disk, it would be better if the
mount script attempted to dynamically detect available USB disks. This
modification to the USB mount script attempts to handle the common case
of a single USB disk but can also handle the case of multiple disks
where it will present the user with all available USB disks
This mimics tlauion's OEM work in the sense that a user (or OEM) could
choose this option and it will reset an OpenPGP smart card and
automatically generate a random key on it. The idea is to allow an OEM
to set up a Librem Key and Heads on a machine before shipping with a
random key, so the user can test for tampering when they receive the
machine, and then the user can choose to reset all of the keys with
their chosen keys after that fact.
