[Bug 1188592] New: Tumbleweed cannot boot with 32-bit EFI (and 64bit kernel)
http://bugzilla.opensuse.org/show_bug.cgi?id=1188592 Bug ID: 1188592 Summary: Tumbleweed cannot boot with 32-bit EFI (and 64bit kernel) Classification: openSUSE Product: openSUSE Tumbleweed Version: Current Hardware: x86-64 OS: Other Status: NEW Severity: Normal Priority: P5 - None Component: Bootloader Assignee: screening-team-bugs@suse.de Reporter: nwr10cst-oslnx@yahoo.com QA Contact: qa-bugs@suse.de Found By: --- Blocker: --- User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:78.0) Gecko/20100101 Firefox/78.0 Build Identifier: Note that this is being tested in a KVM virtual machine with the ovmf-ia32 firmware. Using grub2-i386-efi version 2.06.2.1 I see: Kernel 5.13.2-1-default always fails. The machine resets after attempting to boot. Kernel 5.13.1-1-default sometimes fails and sometimes succeeds. Kernel 5.12.13-1-default boots successfully. However, if I instead use an older version of grub2-i386-efi (version 2.04-16.1 from last summer), then all three kernels boot successfully. I have listed the component (for the bug report) as bootloader, because that's what seems to me to be broken. But perhaps it should instead be considered a kernel issue. Reproducible: Always -- You are receiving this mail because: You are on the CC list for the bug.
http://bugzilla.opensuse.org/show_bug.cgi?id=1188592 http://bugzilla.opensuse.org/show_bug.cgi?id=1188592#c2 --- Comment #2 from Neil Rickert <nwr10cst-oslnx@yahoo.com> ---
"Last summer" sounds a bit early before this change was introduced, probably you didn't use efi handover but 32bit boot protocol to boot the kernel ?
That's probably about right. I've kept a copy of the old "rpm", which has a file date of Aug 20, 2020. I think it's the last grub2 version from Tumbleweed before you made those changes. As for whether it should be considered a kernel bug or a bootloader bug -- that's really not up to me to judge. I just report the problem, and do what I can to work around it. -- You are receiving this mail because: You are on the CC list for the bug.
http://bugzilla.opensuse.org/show_bug.cgi?id=1188592 http://bugzilla.opensuse.org/show_bug.cgi?id=1188592#c3 Michael Chang <mchang@suse.com> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |mchang@suse.com Assignee|mchang@suse.com |kernel-bugs@suse.de --- Comment #3 from Michael Chang <mchang@suse.com> --- (In reply to Neil Rickert from comment #2)
"Last summer" sounds a bit early before this change was introduced, probably you didn't use efi handover but 32bit boot protocol to boot the kernel ?
That's probably about right. I've kept a copy of the old "rpm", which has a file date of Aug 20, 2020. I think it's the last grub2 version from Tumbleweed before you made those changes.
Thanks you very much for the confirmation and that is more clear to me a kernel regression.
As for whether it should be considered a kernel bug or a bootloader bug -- that's really not up to me to judge. I just report the problem, and do what I can to work around it.
(In reply to Michael Chang from comment #1)
To me I would say this much likely kernel regression from 5.12.13-1 to 5.13.0-1.
Reassign to kernel component as this happens on my test since 5.13.0-1. TIA. -- You are receiving this mail because: You are on the CC list for the bug.
http://bugzilla.opensuse.org/show_bug.cgi?id=1188592 http://bugzilla.opensuse.org/show_bug.cgi?id=1188592#c4 --- Comment #4 from Neil Rickert <nwr10cst-oslnx@yahoo.com> --- This seems to be fixed with the 5.14 kernels. Both 5.14.0 and 5.14.1 are booting fine with grub2-i386-efi-2.06-6.1 -- You are receiving this mail because: You are on the CC list for the bug.
http://bugzilla.opensuse.org/show_bug.cgi?id=1188592 http://bugzilla.opensuse.org/show_bug.cgi?id=1188592#c5 Michael Chang <mchang@suse.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution|--- |FIXED --- Comment #5 from Michael Chang <mchang@suse.com> --- (In reply to Neil Rickert from comment #4)
This seems to be fixed with the 5.14 kernels.
Both 5.14.0 and 5.14.1 are booting fine with grub2-i386-efi-2.06-6.1
Hi Neil, Yes, it is : https://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git/commit/?id=22aa4... The fix was merged into 5.14
git describe --contains 22aa45cb465be v5.14~5^2~1
... while the regression was introduced v5.13
git describe --contains 79419e13e808 v5.13-rc1~195^2~5
Let's close and feel free to reopen if you still have problem. Thanks. -- You are receiving this mail because: You are on the CC list for the bug.
participants (1)
-
bugzilla_noreply＠suse.com