All of lore.kernel.org
 help / color / mirror / Atom feed
  • * Re: [REGRESSION] linux 6.6.18 and later fails to boot with "initramfs unpacking failed: invalid magic at start of compressed archive"
           [not found] <172096653882.8.16813493378478071178.379356753@simplelogin.com>
           [not found] ` <32bd7226-5d26-46fd-a3bb-5b14344bc0b5@pm.me>
    @ 2024-07-17 15:55 ` Ard Biesheuvel
      2024-07-17 17:16   ` Ard Biesheuvel
      2024-07-17 21:59   ` mavrix#kernel
      1 sibling, 2 replies; 32+ messages in thread
    From: Ard Biesheuvel @ 2024-07-17 15:55 UTC (permalink / raw)
      To: mavrix#kernel; +Cc: Radek Podgorny, regressions, regressions, christian
    
    Hi,
    
    On Sun, 14 Jul 2024 at 07:15, <mavrix#kernel@simplelogin.com> wrote:
    >
    > Hi
    >
    > My system appears to be affected by a regression originating with commit 8117961d98fb2d335ab6de2cad7afb8b6171f5fe as well, with the main difference being that the current fix does not resolve the issue for me. What I don't know is whether it's a different regression resulting from the same commit or if the core issue is the same and the patch only fixes it is some cases. I decided that replying here would be the most efficient way to bring this to attention, but if I should start a new report altogether, please let me know.
    >
    > Here's the story so far: After updating from 6.6.9-arch1-1 to the next regular kernel version, I had an unbootable system where I could still select the systemd-boot loader entry but then the computer became unresponsive with a black screen, where not even sysrq worked anymore. For a while the Arch LTS kernels still worked as a fallback after that, but eventually they resulted in the same behaviour. I recently posted on the Arch forum about the problem - https://bbs.archlinux.org/viewtopic.php?id=297453 - where Christian Heusel, who I CC'd in this reply as well - helped me bisect the first bad Arch LTS kernel, i.e. 6.6.18. This lead to the aforementioned commit. Mainline unfortunately does not fix it for me (as of linux-mainline-6.10rc6-1). I also have other machines with a similar setup as far as the operating system goes (but on different hardware) that work fine.
    >
    > I have an otherwise up to date Arch system that I've originally set up well over 5 years ago on a Dell XPS 13 and that has been regularly maintained and updated, mostly without issues. The information requested so far in this thread in my case is as follows:
    >
    > bootctl - https://drive.proton.me/urls/P1TKZPA98R#YHmvPFBCRy7v
    > dmidecode - https://drive.proton.me/urls/GXTZVVSTK0#43CCvNy1m6dh
    > dmesg - https://drive.proton.me/urls/6DSXK4R2D8#8TFGkw6yvFfx
    >
    > I hope this all is of some help; if you need further information, please let me know.
    >
    
    Does it make any difference if you boot with 'nokaslr' on the command line?
    
    Another thing you might try is applying the hunk below, to try and see
    whether the placement of boot_params in the binary is causing the
    issue.
    
    I can also provide a partial backport of the offending patch, to
    figure out whether it is the allocation that triggers the issue, or
    the way the boot_params data structure is populated.
    
    
    
    --- a/arch/x86/boot/compressed/vmlinux.lds.S
    +++ b/arch/x86/boot/compressed/vmlinux.lds.S
    @@ -55,8 +55,7 @@ SECTIONS
            . = ALIGN(L1_CACHE_BYTES);
            .bss : {
                    _bss = . ;
    -               *(.bss)
    -               *(.bss.*)
    +               *(SORT_BY_ALIGNMENT(.bss*))
                    *(COMMON)
                    . = ALIGN(8);   /* For convenience during zeroing */
                    _ebss = .;
    
    ^ permalink raw reply	[flat|nested] 32+ messages in thread
  • * [REGRESSION] linux 6.6.18 and later fails to boot with "initramfs unpacking failed: invalid magic at start of compressed archive"
    @ 2024-03-14 18:31 Radek Podgorny
      2024-03-14 19:35 ` Ard Biesheuvel
      0 siblings, 1 reply; 32+ messages in thread
    From: Radek Podgorny @ 2024-03-14 18:31 UTC (permalink / raw)
      To: Ard Biesheuvel, stable, regressions; +Cc: regressions
    
    
    [-- Attachment #1.1: Type: text/plain, Size: 2269 bytes --]
    
    hi,
    
    i seem to be the only one in the world to have this problem. :-(
    
    on one of my machines, updating to 6.6.18 and later (including mainline 
    branch) leads to unbootable system. all other computers are unaffected.
    
    bisecting the history leads to:
    
    commit 8117961d98fb2d335ab6de2cad7afb8b6171f5fe
    Author: Ard Biesheuvel <ardb@kernel.org> 
     
    
    Date:   Tue Sep 12 09:00:53 2023 +0000 
     
    
     
     
    
         x86/efi: Disregard setup header of loaded image 
     
    
     
     
    
         commit 7e50262229faad0c7b8c54477cd1c883f31cc4a7 upstream. 
     
    
    
         The native EFI entrypoint does not take a struct boot_params from the
         loader, but instead, it constructs one from scratch, using the setup
         header data placed at the start of the image.
    
         This setup header is placed in a way that permits legacy loaders to
         manipulate the contents (i.e., to pass the kernel command line or the
         address and size of an initial ramdisk), but EFI boot does not use 
    it in
         that way - it only copies the contents that were placed there at build
         time, but EFI loaders will not (and should not) manipulate the setup
         header to configure the boot. (Commit 63bf28ceb3ebbe76 "efi: x86: Wipe
         setup_data on pure EFI boot" deals with some of the fallout of using
         setup_data in a way that breaks EFI boot.)
    
         Given that none of the non-zero values that are copied from the setup
         header into the EFI stub's struct boot_params are relevant to the boot
         now that the EFI stub no longer enters via the legacy decompressor, the
         copy can be omitted altogether.
    
         Signed-off-by: Ard Biesheuvel <ardb@kernel.org>
         Signed-off-by: Ingo Molnar <mingo@kernel.org>
         Link: 
    https://lore.kernel.org/r/20230912090051.4014114-19-ardb@google.com
         Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
    
    this seems to be the commit to introduce the regression.
    
    i have no idea where to look and what information to provide to explain 
    what the problem might be or why this single machine is different. :-(
    
    please don't hesitate to ask me further questions - i can do any 
    debugging you may need on your behalf.
    
    sincerely,
    R.
    
    #regzbot introduced: 8117961d98fb2d335ab6de2cad7afb8b6171f5fe
    
    [-- Attachment #2: OpenPGP digital signature --]
    [-- Type: application/pgp-signature, Size: 840 bytes --]
    
    ^ permalink raw reply	[flat|nested] 32+ messages in thread

    end of thread, other threads:[~2024-07-22  8:35 UTC | newest]
    
    Thread overview: 32+ messages (download: mbox.gz follow: Atom feed
    -- links below jump to the message on this page --
         [not found] <172096653882.8.16813493378478071178.379356753@simplelogin.com>
         [not found] ` <32bd7226-5d26-46fd-a3bb-5b14344bc0b5@pm.me>
    2024-07-14 14:40   ` [REGRESSION] linux 6.6.18 and later fails to boot with "initramfs unpacking failed: invalid magic at start of compressed archive" mavrix#kernel
    2024-07-17 15:55 ` Ard Biesheuvel
    2024-07-17 17:16   ` Ard Biesheuvel
    2024-07-17 21:44     ` Christian Heusel
    2024-07-17 21:59   ` mavrix#kernel
    2024-07-17 22:15     ` Ard Biesheuvel
    2024-07-18 20:28       ` mavrix#kernel
    2024-07-22  8:35         ` Ard Biesheuvel
    2024-03-14 18:31 Radek Podgorny
    2024-03-14 19:35 ` Ard Biesheuvel
    2024-03-14 21:53   ` Ard Biesheuvel
    2024-03-15  7:42     ` Ard Biesheuvel
    2024-03-15 14:11       ` Radek Podgorny
    2024-03-15 14:25         ` Ard Biesheuvel
    2024-03-15 14:51           ` Radek Podgorny
    2024-03-15 15:33         ` Ard Biesheuvel
    2024-03-15 16:08           ` Ard Biesheuvel
    2024-03-15 16:24             ` Radek Podgorny
    2024-03-15 17:48               ` Ard Biesheuvel
    2024-03-15 18:11                 ` Radek Podgorny
    2024-03-15 18:25                   ` Ard Biesheuvel
    2024-03-18 13:14                     ` Radek Podgorny
    2024-03-18 13:19                       ` Ard Biesheuvel
    2024-03-22 16:28                         ` Radek Podgorny
    2024-03-22 17:23                           ` Ard Biesheuvel
    2024-03-25 14:56                             ` Radek Podgorny
    2024-03-25 16:24                               ` Ard Biesheuvel
    2024-03-28  9:17                                 ` Radek Podgorny
    2024-03-28  9:41                                   ` Ard Biesheuvel
    2024-03-28 14:49                                     ` Ard Biesheuvel
    2024-03-29 10:43                                       ` Radek Podgorny
    2024-03-29 14:36                                         ` Ard Biesheuvel
    

    This is an external index of several public inboxes,
    see mirroring instructions on how to clone and mirror
    all data and code used by this external index.