Restores autocomplete and makes bash more usable as an interactive
shell. Added 106 KB to compressed initrd (checked librem_14).
Signed-off-by: Jonathon Hall <jonathon.hall@puri.sm>
Support LVM2 thin-provisioned volumes. LVM2 wants the thin_check
utility by default, but it has multiple dependencies we do not
currently ship (boost, libexpat, others), so disable it.
Signed-off-by: Jonathon Hall <jonathon.hall@puri.sm>
Add TRACE_FUNC to trace the file, line, and name of the calling
function. File and function names don't have to be duplicated in a
TRACE statement with this (they tend to become inaccurate as functions
are renamed and the TRACE statement is forgotten).
Add DEBUG_STACK to dump the bash stack to debug output.
Configure bash with --enable-debugger. Bash doesn't actually include
the entire debugger, this is just some supporting variables for it.
Evidently, BASH_SOURCE[n] is only set within a function if this is
enabled. I couldn't find this indicated in any documentation, but it
happened in practice.
Compressed initrd size only increased by 2560 bytes for librem_mini_v2,
I think that is fine. This also gives us BASH_ARGC/BASH_ARGV which
might be useful for diagnostics.
Signed-off-by: Jonathon Hall <jonathon.hall@puri.sm>
popt: too old to have a working config.guess
libusb-compat: not needed for gpg2
gpg2: depend on libusb not libusb-compat
Signed-off-by: Thierry Laurion <insurgo@riseup.net>
Laptops can include optional USB keyboard support (default off unless
the board also sets the default to 'y'). The setting is in the
configuration GUI.
CONFIG_USER_USB_KEYBOARD is now the user-controlled setting on those
boards. 'CONFIG_USB_KEYBOARD' is no longer used to avoid any conflict
with prior releases that expect this to be a compile-time setting only
(conflicts risk total lock out requiring hardware flash, so some
caution is justified IMO).
Boards previously exporting CONFIG_USB_KEYBOARD now export
CONFIG_USB_KEYBOARD_REQUIRED. Those boards don't have built-in
keyboards, USB keyboard is always enabled. (librem_mini,
librem_mini_v2, librem_11, librem_l1um, librem_l1um_v2, talos-2,
kgpe-d16_workstation-usb_keyboard, x230-hotp-maximized_usb-kb).
Librem laptops now export CONFIG_SUPPORT_USB_KEYBOARD to enable
optional support. The default is still 'off'.
Signed-off-by: Jonathon Hall <jonathon.hall@puri.sm>
Disable the Ada compiler, as it no longer compiles on Debian 12 and is
not needed.
The Ada compiler is only used for libgfxinit - Intel native graphics
initialization. Neither of the boards on coreboot 4.11 uses this;
Aspeed graphics initialization is written in C (but is not used yet as
it only supports text mode in 4.11).
Signed-off-by: Jonathon Hall <jonathon.hall@puri.sm>
Download coreboot toolchain archives into packages/<arch> before
coreboot tries to download them. This allows us to use mirrors to get
the archives. We could also update the primary source this way if it
goes down instead of patching coreboot itself (has happened for IASL).
The archive versions and digests are retrieved from the coreboot
module, so there isn't another copy of that info to maintain. That is
done in bin/fetch_coreboot_crossgcc_archive.sh, which uses the
existing fetch script to do the actual download, leveraging mirrors.
bin/fetch_source_archive.sh supports using a SHA-1 digest instead of
SHA-256, since coreboot has SHA-1 digests. It also checks if the file
already exists (deleting the coreboot directory will cause it to be
re-run, but the packages are already there and can be used from cache).
The coreboot-4.11 IASL patch is updated to delete the outdated acpica
archive digest (it already added the new one, but the old one was still
there). bin/fetch_coreboot_crossgcc_archive.sh finds the archive
version and digest from the digest files, so only one acpica file must
be present.
Signed-off-by: Jonathon Hall <jonathon.hall@puri.sm>
* overwriting a hotp secret is not possible anymore
* make sure to delete the hotp secret before setting a new one
* requires one additional user presence check during HOTP setup
* bump to v1.5
Signed-off-by: Markus Meissner <coder@safemailbox.de>
* remove all previous coreboot patches (as they are already included)
* to be investigated: linux trampoline patch
* add new patch to hardcode sleep configuration
* activate smmstore as dasharo vendor code requires it
Signed-off-by: Markus Meissner <coder@safemailbox.de>
Taken from : https://github.com/Nitrokey/heads/tree/temp-release-v2.3
- Move branding/Heads/bootsplash-1024x768.jpg -> branding/Heads/bootsplash.jpg (We don't care about the size. Make filename generic)
- Adapt all coreboot configs so bootsplash is adapted by BRAND_NAME CONFIG_BOOTSPLASH_FILE="@BRAND_DIR@/bootsplash.jpg"
- Reminders :
- Makefile changes Heads to defined BRAND_NAME in board config
- Makefile changes -e 's!@BRAND_DIR@!$(pwd)/branding/$(BRAND_NAME)!g'
- nv41/nv50
- coreboot oldefconfigs adapted by:
- make BOARD=nitropad-ns50 coreboot.modify_and_save_oldconfig_in_place
- make BOARD=nitropad-nv41 coreboot.modify_and_save_oldconfig_in_place
- linux oldefconfigs adapted by
- make BOARD=nitropad-nv41 linux.modify_and_save_oldconfig_in_place
- since this is shared config across nv41/ns50: it only needs to be done for a single board
Signed-off-by: Thierry Laurion <insurgo@riseup.net>
- Closes https://github.com/linuxboot/heads/pull/1452
- coreboot: Take Talos II 0.7 release coreboot config file that was inside of cbfs and use it as a base upstream.
- linux: Readd sysctl and proc requirements for cbmem to work.
TODO: fix gpg2 module so that the following doesn't happen (a ppc64 thing. Can't figure out why):
```
Adding generated key to current firmware and re-flashing...
Board talos-2 detected, continuing...
37281653053696daf2e40a8efe9451b557d9d6ab586830dc85f814bf2e03a05f /tmp/talos-2.rom
Initializing Flash Programmer
Reading old flash contents. Please wait...
Flashing: [##################################################\] (100%)
Verifying flash contents. Please wait...
The flash contents were verified and the image was flashed correctly.
Signing boot files and generating checksums...
180726119: 000E452213510000005A
gpg: error running '//bin/dirmngr': probably not installed
gpg: failed to start dirmngr '//bin/dirmngr': Configuration error
gpg: can't connect to the dirmngr: Configuration error
gpg: no default secret key: No dirmngr
gpg: signing failed: No dirmngr
```
dirmngr is deactivated per configure statement --disable-dirmngr, and works as expected on x86
Signed-off-by: Thierry Laurion <insurgo@riseup.net>
As on master otherwise with --disable-asm:
config.status: executing gcrypt-conf commands
Libgcrypt v1.10.1 has been configured as follows:
Platform: GNU/Linux (x86_64-pc-linux-musl)
Hardware detection module: none
Enabled cipher algorithms: arcfour blowfish cast5 des aes twofish
serpent rfc2268 seed camellia idea salsa20
gost28147 chacha20 sm4
Enabled digest algorithms: crc gostr3411-94 md4 md5 rmd160 sha1
sha256 sha512 sha3 tiger whirlpool stribog
blake2 sm3
Enabled kdf algorithms: s2k pkdf2 scrypt
Enabled pubkey algorithms: dsa elgamal rsa ecc
Random number generator: default
Try using jitter entropy: yes
Using linux capabilities: no
FIPS module version:
Try using Padlock crypto: n/a
Try using AES-NI crypto: n/a
Try using Intel SHAEXT: n/a
Try using Intel PCLMUL: n/a
Try using Intel SSE4.1: n/a
Try using DRNG (RDRAND): n/a
Try using Intel AVX: n/a
Try using Intel AVX2: n/a
Try using ARM NEON: n/a
Try using ARMv8 crypto: n/a
Try using PPC crypto: n/a
By disabling --disable-asm in libgcrypt 1.10.1:
config.status: executing gcrypt-conf commands
Libgcrypt v1.10.1 has been configured as follows:
Platform: GNU/Linux (x86_64-pc-linux-musl)
Hardware detection module: libgcrypt_la-hwf-x86
Enabled cipher algorithms: arcfour blowfish cast5 des aes twofish
serpent rfc2268 seed camellia idea salsa20
gost28147 chacha20 sm4
Enabled digest algorithms: crc gostr3411-94 md4 md5 rmd160 sha1
sha256 sha512 sha3 tiger whirlpool stribog
blake2 sm3
Enabled kdf algorithms: s2k pkdf2 scrypt
Enabled pubkey algorithms: dsa elgamal rsa ecc
Random number generator: default
Enabled digest algorithms: crc gostr3411-94 md4 md5 rmd160 sha1
sha256 sha512 sha3 tiger whirlpool stribog
blake2 sm3
Enabled kdf algorithms: s2k pkdf2 scrypt
Enabled pubkey algorithms: dsa elgamal rsa ecc
Random number generator: default
Try using jitter entropy: yes
Using linux capabilities: no
FIPS module version:
Try using Padlock crypto: yes
Try using AES-NI crypto: yes
Try using Intel SHAEXT: yes
Try using Intel PCLMUL: yes
Try using Intel SSE4.1: yes
Try using DRNG (RDRAND): yes
Try using Intel AVX: yes
Try using Intel AVX2: yes
Try using ARM NEON: n/a
Try using ARMv8 crypto: n/a
Try using PPC crypto: n/a
To support PPC crypto, it seems we will need yasm.
To support linux capabilities, libcap would be required as well later on. :/ another point for rng-tools (which also depends on libcap-ng)
Allow boards to optionally include loadkeys to set a custom keymap.
showkey and dumpkeys (normally only needed for development) can also be
optionally included.
Remove *.map from .gitignore; this was probably intended for build
artifacts that are now excluded via the build/ directory.
Add reboot and poweroff to shell history, which is useful for devices
lacking full hardware keyboards to escape the recovery shell with just
"up" and "enter".
Signed-off-by: Jonathon Hall <jonathon.hall@puri.sm>
Update to 1.3. Includes navigation improvements for devices with just
up/down/Enter keys, for Librem 11.
Signed-off-by: Jonathon Hall <jonathon.hall@puri.sm>
This is 4.21-Purism-1 plus a fix for native graphics init on Mini
v1/v2: HDMI1 is enabled so passive DisplayPort to DVI/HDMI adapters
will work.
Signed-off-by: Jonathon Hall <jonathon.hall@puri.sm>
Rationale:
cpio -t alone cannot extract initrd past early cpio (microcode) in most packed initrd.
unpack_initramfs.sh already under master comes to the rescue, but its usage up to today was limited to pass firmware blobs to final OS under boards/librem_mini_v2
Debian OSes (and probably others) need to have cryptroot/crypttab overriden directly, otherwise generic generation of crypttab is not enough.
Extracting crypttab and overriding directly what is desired by final OS and exposed into /boot/initrd is the way to go otherwise hacking on top of hacks.
This brings default packed modules under Heads to 5 modules, which needs to be deactivate in board configs if undesired:
user@heads-tests-deb12:~/heads$ grep -Rn "?= y" modules/ | grep -v MUSL
modules/zlib:1:CONFIG_ZLIB ?= y
modules/zstd:3:CONFIG_ZSTD ?= y
modules/exfatprogs:2:CONFIG_EXFATPROGS ?= y
modules/busybox:2:CONFIG_BUSYBOX ?= y
modules/e2fsprogs:2:CONFIG_E2FSPROGS ?= y
prepare_thumb_drive: default to creating 10% LUKS container on usb drive, prompts for passphrase is not provided and scan drives if no --device specified
NOTE: qemu usb_thumb drive of 128 mb are not big enough so that 10% of it (12mb) can be used to create thumb drive.
Adds:
- e2fsprogs to support ext4 filesystem creation through mke2fs
- add /etc/mke2fs.conf so that mke2fs knows how to handle ext2/ext3/ext4
- removes mke2fs support from busybox
- bump busybox to latest version which adds cpu accelerated hash functions (not needed per se here)
- Adds exfatprogs to have mkfs.exfat and fsck.exfat
- Adds prepare_thumb_drive /etc/luks-functions to be able to prepare a thumb drive with percentage of drive assigned to LUKS, rest to exfat
- Modify most board configs to test space requirements failing
- Talos2 linux config: add staging Exfat support
- Make e2fsprogs and exfatprogs included by default unless explicitely deactivate in board configs
- Change cryptsetup calls : luksOpen to open and luksClose to close to addresss review
- etc/luks_functions: cleanup
GOAL here is to have secure thumb drive creation which Heads will be able to use to backup/restore/use generated GPG key material in the future (next PR)
Nothing else shares the 4.20.1 toolchain yet, and upcoming forks are
based on older releases. We'll share it when other boards update to
4.20.1.
Signed-off-by: Jonathon Hall <jonathon.hall@puri.sm>
fbwhiptail scales its UI based on the display size. FBWHIPTAIL_SCALE
can set a specific scale factor for testing.
fbwhiptail no longer looks for a 1080p mode when the default mode is
2160p.
Signed-off-by: Jonathon Hall <jonathon.hall@puri.sm>
Build kbd and ship setfont if enabled with CONFIG_KBD.
When CONFIG_KBD is enabled, setconsolefont.sh will double the console
font size on large displays (>1600 lines tall as a heuristic).
Signed-off-by: Jonathon Hall <jonathon.hall@puri.sm>
CROSS= is needed for skiboot on PPC64 due to different endianness
relative to coreboot.
The talos_2 fork doesn't share the toolchain because it is the only
_fork_, not board, to be precise. We could add more boards using that
fork without having to create a shared toolchain, it only matters if we
add another fork or start building boards from the upstream release
too.
Signed-off-by: Jonathon Hall <jonathon.hall@puri.sm>
Two := assignments were factored out together, the second overwrote the
first. Fix to +=, and remove the nitrokey assignment since it came
from a branch.
Signed-off-by: Jonathon Hall <jonathon.hall@puri.sm>
This was spelled wrong - it's actually '_depends'. 'initrd' isn't a
module any more so the value doesn't make sense, remove it.
Signed-off-by: Jonathon Hall <jonathon.hall@puri.sm>
The skiboot build fails to find the toolchain when it's not in the
default location. There is only one ppc64 board anyway, so there's no
point trying to share a toolchain for now.
Signed-off-by: Jonathon Hall <jonathon.hall@puri.sm>
Use .heads-toolchain to mark that the toolchain was built rather than
.xcompile. coreboot doesn't generate .xcompile until the build step,
so all modules had to build successfully before we would stop trying to
to rebuild the toolchain. Build steps should generally produce the
indicated outputs too, which was not occurring here.
Signed-off-by: Jonathon Hall <jonathon.hall@puri.sm>
Reuse the toolchain from a coreboot release for fork builds. Either
the fork or the release can be built first, in either case the
release's toolchain is built at the default location and reused for
later builds.
Signed-off-by: Jonathon Hall <jonathon.hall@puri.sm>
Define a separate module for each coreboot version, so the module used
to build the ROM will optionally be able to reference the toolchain
from a different module.
This will allow coreboot fork builds to use the toolchain from the
corresponding release.
Signed-off-by: Jonathon Hall <jonathon.hall@puri.sm>
At one time coreboot was built using Heads' musl toolchain, but this
was later reverted. coreboot builds with its own toolchain again.
CROSS= has no effect on coreboot proper (only exception is PPC64
skiboot payload). It was added to coreboot by a patch that was deleted
in 8e44853. COREBOOT_IASL was set to the default, that was only needed
when the toolchain was being overridden to override iasl back to the
coreboot one.
ppc64 still specifies CROSS= since skiboot is unable to find coreboot's
toolchain from XGCCPATH but checks CROSS. This builds skiboot with the
Heads toolchain as before.
Signed-off-by: Jonathon Hall <jonathon.hall@puri.sm>
Remove coreboot 4.8.1, 4.13, and 4.17, which were all unused.
Remove extra copies of EXTRA_FLAGS which duplicated the common
definition. The only difference was
-Wno-error=address-of-packed-member, the warning is now disabled
entirely everywhere with -Wno-address-of-packed-member.
Use separate coreboot_version values for talos_2, nitrokey, and purism,
which gives each a separate build directory.
Move conditional blob definitions out of each coreboot version.
Fix condition for coreboot-blobs - whether a module is a git clone
actually depends on non-empty <module>_repo, not <module>_version==git.
Fix the test so git versions of coreboot can have arbitrary names.
Signed-off-by: Jonathon Hall <jonathon.hall@puri.sm>
Updated cbmem searches for CBMEM exposed by kernel in sysfs before
trying to read it from memory directly. As such, there is no need for
pointing to that file explicitly.
New coreboot revision also fixes output of 'cbmem -t' caused by wrong
endianness.
Signed-off-by: Krystian Hebel <krystian.hebel@3mdeb.com>
Update flashrom - in particular, this includes support for new chipsets
like Jasper Lake.
CONFIG_INTERAL_X86 was created so CONFIG_INTERNAL could apply to other
platforms, enable it for x86.
The default build target now requires sphinx, just build flashrom
itself.
Update flashrom_progress - filter out noise in newer flashrom that
chokes the progress bar implementation, make size detection more
robust, improve progress bar implementation slightly.
Signed-off-by: Jonathon Hall <jonathon.hall@puri.sm>
Co-signed by: Thierry Laurion <insurgo@riseup.net.
both linux/coreboot:
- save_in_defconfig_format_in_place : takes whatever coreboot config file for a make BOARD=xyz statement and saves it in defconfig
- save_in_oldefconfig_format_in_place : takes whatever coreboot config file for a make BOARD=xyz statement and saves it in oldefconfig
linux:
- linux.prompt_for_new_config_options_for_kernel_version_bump:
- The most useful helper as of now when doing kernel version bump.
- Requires to save current kernel config in oldconfig (make BOARD=xyz linux.save_in_oldefconfig_format_in_place) first, then bump kernel version in board config and then use that helper to review new options and save in tree.
Use CONFIG_BRAND_NAME to control the brand name displayed in the UI.
Override by setting BRAND_NAME when building, either in the Makefile or
on the command line.
Signed-off-by: Jonathon Hall <jonathon.hall@puri.sm>
Debian 12's initrd by default now consists of an uncompressed cpio
archive containing microcode, followed by a zstd-compressed cpio
archive. inject_firmware.sh only supported gzip-compressed cpio, so it
could not extract /init from this archive.
Add zstd-decompress to decompress zstd streams (uncompressed size is
about 180 KB).
Add unpack_initramfs.sh which is able to decompress uncompressed, gzip,
or zstd archives, with multiple segments, much like the Linux kernel
itself does.
Use unpack_initramfs.sh to extract /init for blob jail.
Don't compress the new archive segment containing firmware and the
updated /init.
Signed-off-by: Jonathon Hall <jonathon.hall@puri.sm>
These changes primarily improve server boards using BMC video. The
correct DRI card is selected even if it isn't the first one, and
performance is greatly improved on non-UMA cards.
Signed-off-by: Jonathon Hall <jonathon.hall@puri.sm>
This is particularly beneficial for servers with Aspeed BMC video,
because it introduces framebuffer console acceleration. The
framebuffer console is much more responsive.
Patches were ported from 5.10.5:
0001-fake-acpi.patch: This may not be needed any more, but it applies
cleanly and I don't think it would harm anything.
0002-nmi-squelch.patch: The comment mentions qemu but I see this
message on physical machines occasionally, so I think this is needed.
0003-fake-trampoline.patch: This patch does not apply cleanly. It
could be ported, but I don't think it's needed, I dropped it. Dates
back to a very old commit where Linux was being embedded into a vendor
UEFI firmware: a4d7654b1e.
0010-winterfell-ahci.patch: Minor change of %x to %lx in context.
Signed-off-by: Jonathon Hall <jonathon.hall@puri.sm>
Add ioport module, enable for librem_mini_v2. Only inb and outb are
included, inw/outw/inl/outl aren't needed.
Signed-off-by: Jonathon Hall <jonathon.hall@puri.sm>
On some newer platforms of intel (confirmed on nehalem, sandy/ivy
bridge), coreboot after commit [2ac149d294af795710eb4bb20f093e9920604abd](https://review.coreboot.org/cgit/coreboot.git/commit/?id=2ac149d294af795710eb4bb20f093e9920604abd)
registers an SMI to lockdown some registers on the chipset, as well
as access to the SPI flash, optionally. The SMI will always be triggered
by coreboot during S3 resume, but can be triggered by either coreboot
or the payload during normal boot path.
Enabling lockdown access to SPI flash will effectly write-protect it,
but there is no runtime option for coreboot to control it, so letting
coreboot to trigger such SMI will leave the owner of the machine lost
any possibility to program the SPI flash with its own OS, and becomes
a nightmare if the machine is uneasy to disassemble, so a scheme could
be implement, in which the SMI to lockdown chipset and SPI flash is left
for a payload to trigger, and temporarily disabling such triggering in
order to program the SPI flash needs authentication.
I have implemented a passcode-protected runtime-disableable lockdown
with grub, described [here](https://github.com/hardenedlinux/Debian-GNU-Linux-Profiles/blob/master/docs/hardened_boot/grub-for-coreboot.md#update-for-coreboot-after-commit-2ac149d294af795710eb4bb20f093e9920604abd). In order to implement a similar scheme for
Heads, I wrote [io386](https://github.com/hardenedlinux/io386).
With this commit, io386 will be called before entering boot routine
to trigger the SMI to finalize the chipset and write protect the SPI
flash at the same time. Entering recovery shell will leave the flash
writable.
(The authentication routine implemented in previous revisions has been
split as an independent commit.)
Originally proposed under PR#326
* Properly initialize sensor IDs of 2nd CPU to fix fan control.
* Use 2s delay for I2C communications with TPM in OPAL (configured in
device tree).
* Stop building unused parts of skiboot using host GCC.
Signed-off-by: Sergii Dmytruk <sergii.dmytruk@3mdeb.com>
Updated to reproducible version of fbwhiptail.
Added flags to remove debug info.
Updated url to current one instead of going through redirect.
Signed-off-by: Daniel Pineda <daniel.pineda@puri.sm>
Most useful to me are:
coreboot.modify_and_save_defconfig_in_place
coreboot.modify_and_save_oldconfig_in_place
linux.modify_and_save_oldconfig_in_place
linux.modify_and_save_defconfig_in_place
Which permit to take current in tree configs and translate them into other format.
This is useful when trying to version bump and build.
Also add helpers to save in versioned version to facilitate change tracking:
linux.generate_and_save-versioned-oldconfig
linux.regenerate_and_save_versioned_defconfig
Update kexec to 2.0.26. Add tracing to framebuffer initialization. In
particular, the driver name is traced if not recognized, and messages
about kernel config are shown if the kernel doesn't provide the
framebuffer pointer.
Signed-off-by: Jonathon Hall <jonathon.hall@puri.sm>
By default json-c builds as debug instead of release.
Adding CMAKE_BUILD_TYPE=minsizerel ensures it does not
add debug info and also optimizes for file size.
Signed-off-by: Daniel Pineda <daniel.pineda@puri.sm>
This patch adds ARCH="$(LINUX_ARCH)" to Linux targets working on config
files. Without it, the architecture defaults to that of host, which for
cross-compilation isn't right.
Signed-off-by: Krystian Hebel <krystian.hebel@3mdeb.com>
Disable all optional algorithms except SM3. (SHA and AES are not
optional.) tpm2-tss uses SHA, AES, and SM3. Reduces size of libcrypto
by almost 1 MB, saves about 140 KB in ROM.
Signed-off-by: Jonathon Hall <jonathon.hall@puri.sm>
tpm2-tools is able to log pcap files of TPM2 commands, which can be
inspected with wireshark. Add CONFIG_TPM2_CAPTURE_PCAP to capture
these from the tpmr wrapper, and enable for qemu TPM2 boards.
Signed-off-by: Jonathon Hall <jonathon.hall@puri.sm>
Update to branch including tpm pcrread until it is merged upstream. tpm
pcrread allows us to use the same logical flow for TPM1 and TPM2 in
seal operations.
Signed-off-by: Jonathon Hall <jonathon.hall@puri.sm>
Update OpenSSL to 3.0.8. Build with -Os. Install only libcrypto,
libssl is not currently needed. Don't buid tests.
Signed-off-by: Jonathon Hall <jonathon.hall@puri.sm>
- legacy-flash boards have a single purpose: to flash BIOS region through flashrom.
- They do not need bash nor have space for it in their 4mb defined coreboot CBFS region
Test build to have legacy boards builds under osresearch#1292
Include bash in all builds. Remove CONFIG_BASH.
Remove CONFIG_BASH_IS_ASH from busybox configuration and clean up hacks
in modules/bash.
Signed-off-by: Jonathon Hall <jonathon.hall@puri.sm>
Disable readline features for interactive shell. This significantly
reduces the size of bash and doesn't affect scripting features. The
interactive shell still functions, but there is no history or command
line editing (backspace works, but arrows do not move cursor).
Enable -Os on bash for more size reduction.
This saves about 180KiB from the compressed initrd for
qemu-coreboot-fbwhiptail-tpm2-hotp, almost half the cost of adding
bash.
Signed-off-by: Jonathon Hall <jonathon.hall@puri.sm>
