So I'm writing a program in c, and I want to get the firmware (e.g. BIOS, UEFI, Coreboot). Since I'm writing a c program it shouldn't really be running commands, I want to be able to check via files or a library but preferably files. I know /sys/firmware/efi in created for UEFI systems, but then what about BIOS and Coreboot? DMI doesn't even seem to have a nice answer either since it seems to consider UEFI as BIOS I guess for compatibility reasons, meaning it's all just called BIOS when it may or may not be. I couldn't find anything under /sys/class/dmi/id that mentioned what firmware was running, only stuff about that firmware, though somehow the command dmidecode listed 'UEFI is supported' under the characteristics of my 'BIOS Information' (I'm using UEFI, not BIOS), I couldn't find where it got that information from though. In the first place DMI instead a great solution since I need this to be able to work for ARM and RISC-V, not just x86 which I think DMI is for.

I ever dislike the keyboards with lights, today after firmware update on my NS5x_NS7xPU the keyboard..become "with backlight". How to turn off? I have tried this command as root sh -c "echo 0 > /sys/class/leds/system76_acpi::kbd_backlight/max_brightness" sh: line 1: /sys/class/leds/system76_acpi::kbd_backlight/max_brightness: Permission denied When I press fn+f7 it appear a "monitor configuration" of Xfce4 When I run this command echo 0 > /sys/class/leds/system76_acpi\:\:kbd_backlight/brightness the keyboard backlight is still on,

I am using Linux Mint Linux version 5.15.0-76-generic on Lenovo Thinkpad x220 with flashed Coreboot BIOS. System boots but with ACPI errors, namely:

[    0.194996] ACPI Error: AE_NO_ACPI_TABLES, While loading namespace from ACPI tables (20210730/tbxfload-68)

which comes from linux tbsfload file. The Lenovo x220 model has 1 USB 3.0 port (works fine) and 2 USB 2.0 ports. USB 2.0 ports do not work which I believe is linked to the ACPI issue. The boot time is around 1 minute because of USB failures:

[    3.431059] i915 0000:00:02.0: [drm] fb0: i915drmfb frame buffer device
[   16.714718] usb 2-1: device not accepting address 2, error -110
[   16.714855] usb 1-1: device not accepting address 2, error -110
[   16.842692] usb 1-1: new high-speed USB device number 3 using ehci-pci
[   16.842732] usb 2-1: new high-speed USB device number 3 using ehci-pci
[   32.842692] usb 2-1: device not accepting address 3, error -110
[   32.842837] usb 1-1: device not accepting address 3, error -110
[   32.842953] usb usb1-port1: attempt power cycle
[   32.842981] usb usb2-port1: attempt power cycle
[   33.286712] usb 1-1: new high-speed USB device number 4 using ehci-pci
[   33.286759] usb 2-1: new high-speed USB device number 4 using ehci-pci
[   44.106690] usb 2-1: device not accepting address 4, error -110
[   44.106832] usb 1-1: device not accepting address 4, error -110
[   44.234721] usb 2-1: new high-speed USB device number 5 using ehci-pci
[   44.234760] usb 1-1: new high-speed USB device number 5 using ehci-pci
[   54.858694] usb 1-1: device not accepting address 5, error -110
[   54.858846] usb 2-1: device not accepting address 5, error -110
[   54.858962] usb usb2-port1: unable to enumerate USB device
[   54.859067] usb usb1-port1: unable to enumerate USB device

I also do not have any visibility on battery state on Linux. My question: How do I provide ACPI tables to the OS so it does see it properly during boot? Is it a matter of installing some missing packages on linux (if so, which ones?) or rather is it coming from Coreboot (how can I fix the Coreboot build so coreboot.rom has necessary ACPI tables?). I have reviewed Coreboot docs on ACPI but there is no meaningful information related to this case. I also tried to boot with acpi=off mode and pci=noacpi in Grub settings but the first fails to boot system (blank screen hangs on boot) and the second makes os unusable (no mouse/trackpad/keyboard working).

I will be trying to flash LibreBoot (a particular coreboot downstream) on to a thinkpad T440p. Even with libreboot - instructions are quite complex, but look entirely doable. There is one thing, however - I cannot be sure if I can make the boot roms using lbmk on a relatively powerful linux computer (running manjaro/arch) and then flash using RaspberryPi, or must I make the roms on the raspberrypi that I will be using for flashing (with a SOIC-8 clip). Intuitively I think there should be no problem, but I just don't know this area. Perhaps someone who understands this better, or who has done libreboot flashing could explain this?

Few days ago, I flashed my x230 with coreboot with this tutorial . Now, while I was installing Funtoo using its GPT-Partitioning guide (not using MBR-Partitioning ), I got following error: EFI variables are not supported on this system Does this mean, Coreboot/SeaBIOS only use MBR partitioning ?

I am interested in investing in hardware that supports libreboot, and I understand that it's a fork of coreboot with firmware blobs removed, however I do not understand what seabios is. I did some research, however could not find information (that I understand) about where exactly seabios stands in all this. It has been explained that seabios is a payload, however an OS kernel (say Linux) can instead be the 'payload'. I've seen mention of seabios when trying to tweak a Chromebook, and also on qemu, and on Purism's products, however from what I understood; seabios is not necessary when using coreboot. Is seabios on a higher level (as the term 'payload' suggests) or is it on a lower level (as the name 'seabios' itself suggests). I'm guessing that its similar to an open source video driver stack in Linux in which you have GPU firmware > kernel driver > userspace driver/API layer (like mesa, SDL,...), then the application. From what I understood, Coreboot/Libreboot is the hardware initialization program, and Seabios would be something that loads the bootloader (GRUB or ISOLINUX), and it goes from there, however the details are still unclear to me. I've been using GNU/Linux full-time for over a decade now, however my knowledge ends beyond systemd and GRUB scripts. Never looked into libre BIOS/UEFI alternatives until now. I'd greatly appreciate clarification on this prior to committing to buying hardware (even though the list of supported hardware is highly limited)!

A of couple days ago I've stumbled upon this speech: https://www.youtube.com/watch?v=3CQUNd3oKBM The guy in the video speaks about Intel Mangement Engine, describes in details how it works and that he exploited it over JTAG. This exploit lets you run your own code on Intel ME. All code on the ME is signed, the keys are in FUSEs (readonly memory), but ME loads manually the keys into register, so you can feed the register with your own keys. So we can trick BG. As I know from the forum page here ( https://groups.google.com/forum/m/#!topic/qubes-users/dlNu2Iv9MCU ) the main coreboot problem is the BG because there is even this video here ( https://www.google.com/url?q=https://twitter.com/_zaolin_/status/1097966252625149952%3Flang%3Den&sa=U&ved=2ahUKEwjGlJar4PznAhXztHEKHcsnD_EQtwIwAXoECAQQAQ&usg=AOvVaw0Ii-ufcV7uluw5LzRYECTY ) where the developer runs coreboot on a ThinkPad T480s!!! BUT the unlocked sample (that's impossible to get), so without BG. I ask for your help professional coders, because I understand not everything of all what I've heard/read (I'm only learning). - Does this mean that "soon" we will have coreboot support for newer hardware and how much time will this "soon" take? (Coreboot isn't dead but at the same time's not active) - Is the T440p coreboot support out/almost out because of this vulnerability research?[ANSWERED] - How the guy from the video (first link) got a T460 to work if there is no coreboot image out there to flash? UPDATE1: T440p is coreboot compatible because of Haswell BG that's incomplete and works only on BGA chips. Coreboot is active in chromebooks support development. Cheers Yugene:)

After flashing coreboot I get the wrong date on my linux install.

# date
Mon 21 Mar 2089 04:03:42 AM UTC

And when I try to read the date with hwclock I get this error

# hwclock -r
hwclock: ioctl(RTC_RD_TIME) to /dev/rtc0 to read the time failed: Invalid argument

Here's the output of dmesg Could this be caused by a dying/dead CMOS battery ? Thanks in advance! PD: If I boot coreinfo I see what it seems to be an invalid date on the bottom right (71/131/1983 - 52:16:58).

I am having incredible trouble with networking. (original post , different forum). I'm upgrading a secondhand Thinkpad x230 (installed: coreboot with SeaBIOS, have run me_cleaner, using Lubuntu Cosmic). Suddenly, internal wifi refuses to work. Stopped inexplicably 3 weeks ago after I reassembled the machine (keyboard, etc). Tried a different motherboard (original config), booted live USBs (Fedora, Manjaro, etc), traced antenna wires and different wifi card (2nd hand). No joy. - Wifi sometimes works with a DWA-131 Dlink "Wireless N Nano USB Adapter", but usually not. - A Huawei 4G modem seems to work perfectly, recognized as a wired ethernet connection. - Not sure, but DWA-131 adapter seems to start working after the 4G modem has been inserted. My last dmsesg dump is here , if that's of any use. (Timestamp is wrong, so clock is out, apparently). The last few hours shows me plugging and unplugging things (including a computer-to-computer ethernet link, below), and the DWA-131 adapter starting to work somehow. Really not sure how to proceed. --- There's another issue I'll post separately: a cable connection from the x230 to a Dell XPS13 prompted a strange network in someone else's name.

I read it's about replacing the firmware on a board, but why do they need to do this, and does it effect the ability to install other OSs on the board?