Create a DRM driver for VGA video cards
a project by tdz
Updated almost 3 years ago. 6 hacker ♥️. 4 followers.

Yes, those VGA video cards. The goal of this project is to implement a DRM graphics driver for such devices. While actual hardware is hard to obtain or even run today, qemu emulates VGA output.

VGA has a number of limitations, which make this project interesting.

  • There are only 640x480 pixels (or less) on the screen. That resolution is also a soft lower limit imposed by DRM. It's mostly a problem for desktop environments though.
  • Desktop environments assume 16 million colors, but there are only 16 colors with VGA. VGA's 256 color palette is not available at 640x480. We can choose those 16 colors freely. The interesting part is how to choose them. We have to build a palette for the displayed frame and map each color to one of the palette's 16 entries. This is called dithering, and VGA's limitations are a good opportunity to learn about dithering algorithms.
  • VGA has an interesting memory layout. Most graphics devices use linear framebuffers, which store the pixels byte by byte. VGA uses 4 bitplanes instead. Plane 0 holds all bits 0 of all pixels. Plane 1 holds all bits 1 of all pixels, and so on.

The driver will probably not be useful to many people. But, if finished, it can serve as test environment for low-level hardware. There's some interest in supporting old Amiga and Atari framebuffers in DRM. Those systems have similar limitations as VGA, but are harder to obtain and test with. With qemu, the VGA driver could fill this gap.

Apart from the Wikipedia entry, good resources on VGA are at osdev.net and FreeVGA

Join this project

Looking for hackers with the skills:

graphics drivers kernel drm

This project is part of:

Hack Week 22 Hack Week 24

Activity

  • about 1 year ago: Pharaoh_Atem liked this project.
  • about 1 year ago: llansky3 liked this project.
  • about 1 year ago: michals liked this project.
  • almost 3 years ago: vliaskovitis liked this project.
  • almost 3 years ago: wfrisch liked this project.
  • almost 3 years ago: achaudhary2 joined this project.
  • almost 3 years ago: tdz added keyword "graphics" to this project.
  • almost 3 years ago: tdz added keyword "drivers" to this project.
  • almost 3 years ago: tdz added keyword "kernel" to this project.
  • almost 3 years ago: tdz added keyword "drm" to this project.
  • almost 3 years ago: tdz started this project.
  • almost 3 years ago: tdz liked this project.
  • almost 3 years ago: tdz originated this project.


    • Comments

    • tdz
      almost 3 years ago by tdz | Reply

      I have pushed a skeleton driver to Gitlab. The does not display anything yet, but binds to a provided VGA device. The rsp kernel config for i586 is here.

      I use 32-bit Tumbleweed on qemu as testing environment. Here are some short setup instructions.

      • Download TW for 32-bit and install in a qemu guest
      • There are various emulated graphics devices available with qemu's 32-bit emulation. I use virtio, but anything with VGA should work
      • On the installed system, open /etc/default/grub and set 'GRUB_TERMINAL=console' to enable textmode. Also append 'console=ttyS0,115200n8' to 'GRUB_CMDLINE_LINUX_DEFAULT' to enable serial-console output.
      • Recreate grub's config with ' sudo grub2-mkconfig -o /boot/grub2/grub.cfg'. These steps will set grub to VGA text mode.
      • Build and install the provided kernel. I use 'make O=build-i586/' and then again with module_install and install to get it installed. This requires kernel development tools.
      • After a reboot, grub should start in text mode. The development kernel should load the vgadrm driver. It does not display anything. If you switch to the serial console and login, 'dmesg | grep drm' should mention vgadrm.

    • tdz
      almost 3 years ago by tdz | Reply

      Day 1: I started with the skeleton driver that binds to the VGA device without doing proper output. I added blitting code to copy the framebuffer data into the video memory. The driver also installs a default palette on each pageflip. The screen now turns dark, but at least it's obvious that something changed. Updating the video memory is somewhat complicated due to VGA's awkward memory layout and a number of register settings that modify the written framebuffer data.

      See https://gitlab.suse.de/tzimmermann/linux/-/commits/vgadrm-day1.

    • tdz
      almost 3 years ago by tdz | Reply

      Amit joined the project and is about to install and try the currently available driver code.

    • tdz
      almost 3 years ago by tdz | Reply

      Day 2: I've been fixing register settings for the whole day. The driver is still not functional, but there's at least noise displayed on the screen. The existing material at FreeVGA and osdev.net has been extremely helpful.

      See https://gitlab.suse.de/tzimmermann/linux/-/tree/vgadrm-day2.

    • tdz
      almost 3 years ago by tdz | Reply

      Day 3: I only did register debugging today, as I still don't see an image on the screen. VGA has plenty of settings for text mode; plus compatibility with the even older EGA and CGA cards. I guess those could interfere or I'm simply missing something trivial. The qemu emulator doesn't seems to tell me about possible problems. So I'm going to set up a regular BIOS-based PC to test the driver. Connecting a monitor should give a signal and a video mode. If the signal is there, the problem is in the screen update; if not it's in the mode setting.

      See https://gitlab.suse.de/tzimmermann/linux/-/tree/vgadrm-day3.

      • tdz
        almost 3 years ago by tdz | Reply

        And just after I typed this comment, I tried the driver on a PC and could not get a signal on the monitor. So the mode-setting code appears to be the problem.

    • tdz
      almost 3 years ago by tdz | Reply

      Day 4: I stepped through my commits and made them into smaller pieces until I found the reason why the screen remained dark. It turns out that it's necessary to activate the palette after filling it with colors. The bit the to that is non-intuitively hidden in the attribute controller's index register, which also serve as a data register. The VGA device's interface is often like this and small errors can quickly backfire.

      So there's still the noise on the screen, but with repeating patterns. It looks like the final problem is in the way the VGA chips reads from video memory.

      See https://gitlab.suse.de/tzimmermann/linux/-/tree/vgadrm-day4.

    • tdz
      almost 3 years ago by tdz | Reply

      Day 5: Today I finally got a working output on the screen! Upon boot, the kernel console is being displayed and it is possible to log in and use the command line. The screen flickers, has a few artifacts and sometimes the colors are off, but it mostly works. I assume that the remaining problems come from many little issue within the kernel's DRM framework. The helpers for color-format conversion and blitting are not prepared to handle 16-color palette modes. So a few fixes might be required. Nevertheless, the display output is there and I count this hackweek project as a success. If I find the time to fix the remaining bugs, I'll send the driver to upstream for inclusion in the official Linux kernel.

      See https://gitlab.suse.de/tzimmermann/linux/-/tree/vgadrm-day5.

    Similar Projects

    Improve UML page fault handler by ptesarik

    Description

    Improve UML handling of segmentation faults in kernel mode. Although such page faults are generally caused by a kernel bug, it is annoying if they cause an infinite loop, or panic the kernel. More importantly, a robust implementation allows to write KUnit tests for various guard pages, preventing potential kernel self-protection regressions.

    Goals

    Convert the UML page fault handler to use oops_* helpers, go through a few review rounds and finally get my patch series merged in 6.14.

    Resources

    Wrong initial attempt: https://lore.kernel.org/lkml/20231215121431.680-1-petrtesarik@huaweicloud.com/T/

    Backporting patches using LLM by jankara

    Description

    Backporting Linux kernel fixes (either for CVE issues or as part of general git-fixes workflow) is boring and mostly mechanical work (dealing with changes in context, renamed variables, new helper functions etc.). The idea of this project is to explore usage of LLM for backporting Linux kernel commits to SUSE kernels using LLM.

    Goals

    • Create safe environment allowing LLM to run and backport patches without exposing the whole filesystem to it (for privacy and security reasons).
    • Write prompt that will guide LLM through the backporting process. Fine tune it based on experimental results.
    • Explore success rate of LLMs when backporting various patches.

    Resources

    • Docker
    • Gemini CLI

    Repository

    Current version of the container with some instructions for use are at: https://gitlab.suse.de/jankara/gemini-cli-backporter

    dynticks-testing: analyse perf / trace-cmd output and aggregate data by m.crivellari

    Description

    dynticks-testing is a project started years ago by Frederic Weisbecker. One of the feature is to check the actual configuration (isolcpus, irqaffinity etc etc) and give feedback on it.

    An important goal of this tool is to parse the output of trace-cmd / perf and provide more readable data, showing the duration of every events grouped by PID (showing also the CPU number, if the tasks has been migrated etc).

    An example of data captured on my laptop (incomplete!!):

              -0     [005] dN.2. 20310.270699: sched_wakeup:         WaylandProxy:46380 [120] CPU:005
          -0     [005] d..2. 20310.270702: sched_switch:         swapper/5:0 [120] R ==> WaylandProxy:46380 [120]
...
    WaylandProxy-46380 [004] d..2. 20310.295397: sched_switch:         WaylandProxy:46380 [120] S ==> swapper/4:0 [120]
          -0     [006] d..2. 20310.295397: sched_switch:         swapper/6:0 [120] R ==> firefox:46373 [120]
         firefox-46373 [006] d..2. 20310.295408: sched_switch:         firefox:46373 [120] S ==> swapper/6:0 [120]
          -0     [004] dN.2. 20310.295466: sched_wakeup:         WaylandProxy:46380 [120] CPU:004

    Output of noise_parse.py:

    Task: WaylandProxy Pid: 46380 cpus: {4, 5} (Migrated!!!)
        Wakeup Latency                                Nr:        24     Duration:          89
        Sched switch: kworker/12:2                    Nr:         1     Duration:           6

    My first contribution is around Nov. 2024!

    Goals

    • add more features (eg cpuset)
    • test / bugfix

    Resources

    Progresses

    isolcpus and cpusets implemented and merged in master: dynticks-testing.git commit

    early stage kdump support by mbrugger

    Project Description

    When we experience a early boot crash, we are not able to analyze the kernel dump, as user-space wasn't able to load the crash system. The idea is to make the crash system compiled into the host kernel (think of initramfs) so that we can create a kernel dump really early in the boot process.

    Goal for the Hackweeks

    1. Investigate if this is possible and the implications it would have (done in HW21)
    2. Hack up a PoC (done in HW22 and HW23)
    3. Prepare RFC series (giving it's only one week, we are entering wishful thinking territory here).

    update HW23

    • I was able to include the crash kernel into the kernel Image.
    • I'll need to find a way to load that from init/main.c:start_kernel() probably after kcsan_init()
    • I workaround for a smoke test was to hack kexec_file_load() systemcall which has two problems:
      1. My initramfs in the porduction kernel does not have a new enough kexec version, that's not a blocker but where the week ended
      2. As the crash kernel is part of init.data it will be already stale once I can call kexec_file_load() from user-space.

    The solution is probably to rewrite the POC so that the invocation can be done from init.text (that's my theory) but I'm not sure if I can reuse the kexec infrastructure in the kernel from there, which I rely on heavily.

    update HW24

    • Day1
      • rebased on v6.12 with no problems others then me breaking the config
      • setting up a new compilation and qemu/virtme env
      • getting desperate as nothing works that used to work
    • Day 2
      • getting to call the invocation of loading the early kernel from __init after kcsan_init()

    • Day 3

      • fix problem of memdup not being able to alloc so much memory... use 64K page sizes for now
      • code refactoring
      • I'm now able to load the crash kernel
      • When using virtme I can boot into the crash kernel, also it doesn't boot completely (major milestone!), crash in elfcorehdr_read_notes()

    • Day 4

      • crash systems crashes (no pun intended) in copy_old_mempage() link; will need to understand elfcorehdr...
      • call path vmcore_init() -> parse_crash_elf_headers() -> elfcorehdr_read() -> read_from_oldmem() -> copy_oldmem_page() -> copy_to_iter()

    • Day 5

      • hacking arch/arm64/kernel/crash_dump.c:copy_old_mempage() to see if crash system really starts. It does.
      • fun fact: retested with more reserved memory and with UEFI FW, host kernel crashes in init but directly starts the crash kernel, so it works (somehow) \o/

    update HW25

    • Day 1
      • rebased crash-kernel on v6.12.59 (for now), still crashing

    Add Qualcomm Snapdragon 765G (SM7250) basic device tree to mainline linux kernel by pvorel

    Qualcomm Snapdragon 765G (SM7250) (smartphone SoC) has no support in the linux kernel, nor in u-boot. Try to add basic device tree support. The hardest part will be to create boot.img which will be accepted by phone.

    UART is available for smartphone :).