early stage kdump support
a project by mbrugger
Updated 8 months ago. 17 hacker ♥️. 12 followers.

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/

  • TODOs

    • fix elfcorehdr so that we actually can make use of all this...
    • test where in the boot __init() chain we can/should call kexec_early_dump()
    • do we really need memdup() or can we used the complied kernel for creating the segments?
    • refactor and rename everything (Kconfig menu shows in wrong place, Kconfig entry needs to go somewhere else, ekdump vs early_dump vs early_kdump

Resources

Join this project

Looking for hackers with the skills:

kernel kdump

This project is part of:

Hack Week 21 Hack Week 22 Hack Week 23 Hack Week 24 Hack Week 25

Activity

  • 7 months ago: djais liked this project.
  • 8 months ago: livdywan liked this project.
  • 8 months ago: huanxie liked this project.
  • 8 months ago: djais joined this project.
  • 9 months ago: jmodak liked this project.
  • 10 months ago: eomelette joined this project.
  • 10 months ago: m.crivellari liked this project.
  • over 1 year ago: vbabka liked this project.
  • over 1 year ago: pvorel liked this project.
  • over 2 years ago: jzerebecki liked this project.
  • over 2 years ago: tdz liked this project.
  • over 2 years ago: Zildj1an liked this project.
  • over 2 years ago: Zildj1an joined this project.
  • over 2 years ago: tschmitz liked this project.
  • over 2 years ago: paolodepa liked this project.
  • over 2 years ago: afeijoo liked this project.
  • about 3 years ago: clin liked this project.
  • about 3 years ago: ptesarik liked this project.
  • about 3 years ago: akumar joined this project.
  • about 3 years ago: akumar liked this project.
  • about 3 years ago: coolgw liked this project.
  • about 3 years ago: mbrugger started this project.
  • about 3 years ago: mbrugger added keyword "kdump" to this project.
  • about 3 years ago: mbrugger added keyword "kernel" to this project.
  • about 3 years ago: mbrugger originated this project.


    • Comments

    • mbrugger
      over 1 year ago by mbrugger | Reply

      I got a tip to look into grub2 multiboot. Maybe we can load the crash kernel to the reserved memory region from grub. We then would only need to pass a kernel argument so that the kernel takes this into account.

    • ptesarik
      11 months ago by ptesarik | Reply

      FWIW I was contemplating a similar scheme back in 2016. My idea was to load: 1. kdump kernel 2. kdump initrd 3. production kernel 4. production initrd Then boot into the kdump kernel, update memory maps and kexec to the production kernel. When the production kernel crashes, pass control back to the kdump kernel. For the return to the kdump kernel, I was looking at the KEXEC_PRESERVE_CONTEXT flag, but in the end I doubt it's really helpful without further modifications to the production kernel. At this point, it's probably easier to boot the production kernel first and set up an initial crash kernel at early boot.

      Good luck!

    • mbrugger
      7 months ago by mbrugger | Reply

      [ 0.221029] Unable to handle kernel level 3 address size fault at virtual address ffff800080aa0000
      [ 0.222848] Mem abort info:
      [ 0.223419] ESR = 0x0000000096000003
      [ 0.224187] EC = 0x25: DABT (current EL), IL = 32 bits
      [ 0.225300] SET = 0, FnV = 0
      [ 0.225933] EA = 0, S1PTW = 0
      [ 0.226587] FSC = 0x03: level 3 address size fault
      [ 0.227600] Data abort info:
      [ 0.228198] ISV = 0, ISS = 0x00000003, ISS2 = 0x00000000
      [ 0.229351] CM = 0, WnR = 0, TnD = 0, TagAccess = 0
      [ 0.230385] GCS = 0, Overlay = 0, DirtyBit = 0, Xs = 0
      [ 0.231466] swapper pgtable: 64k pages, 48-bit VAs, pgdp=000000005d6b0000
      [ 0.232850] [ffff800080aa0000] pgd=100000005ef80003, p4d=100000005ef80003, pud=100000005ef80003, pmd=100000005ef90003, pte=00681591c0000f03
      [ 0.235548] Internal error: Oops: 0000000096000003 [#1] PREEMPT SMP
      [ 0.236828] Modules linked in:
      [ 0.237504] CPU: 0 UID: 0 PID: 1 Comm: swapper/0 Not tainted 6.12.0-dirty #9
      [ 0.239037] Hardware name: linux,dummy-virt (DT)
      [ 0.240047] pstate: a0400005 (NzCv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--)
      [ 0.241563] pc : _memcpy+0x110/0x230
      [ 0.242374] lr : _copy      toiter+0x374/0x670
      [ 0.243276] sp : ffff8000800efa20
      [ 0.244009] x29: ffff8000800efa70 x28: 0000000000000000 x27: ffff800080aa0000
      [ 0.245577] x26: ffff8000800efba0 x25: 00000000000001a8 x24: ffffd7d750365000
      [ 0.247124] x23: ffff8000800efbb0 x22: 0000000000000000 x21: ffff8000800efba0
      [ 0.248672] x20: 00000000000001a8 x19: 0000000000000000 x18: ffffffffffffffff
      [ 0.250234] x17: 0000000087130253 x16: 00000000e547dfaa x15: 0720072007200720
      [ 0.251792] x14: ffffa2fe3f221a00 x13: ffffd7d750365fb8 x12: ffffd7d75162c9c8
      [ 0.253349] x11: ffffd7d75169ca30 x10: ffffd7d7516849f0 x9 : ffffd7d751684a48
      [ 0.254900] x8 : 0000000000017fe8 x7 : c0000000ffffefff x6 : 0000000000000001
      [ 0.256457] x5 : ffff22febfcc1ba8 x4 : ffff800080aa01a8 x3 : 00000000ffffefff
      [ 0.258013] x2 : 00000000000001a8 x1 : ffff800080aa0000 x0 : ffff22febfcc1a00
      [ 0.259564] Call trace:
      [ 0.260109] _      memcpy+0x110/0x230
      [ 0.260842] copyoldmempage+0xc8/0x110
      [ 0.261713] readfromoldmem+0x1bc/0x268
      [ 0.262595] elfcorehdrreadnotes+0x9c/0xd0
      [ 0.263536] mergenoteheaderself64.constprop.15+0x110/0x3b0
      [ 0.264813] vmcore      init+0x298/0x794
      [ 0.265612] dooneinitcall+0x64/0x1e8
      [ 0.266455] kernelinitfreeable+0x238/0x288

    Similar Projects

    This project is one of its kind!