Re: kernel problem 12.4

On 12/10/25 5:31 AM, Thomas Seeling ([email protected] via 
lfs-support Mailing List) wrote:
> Hello dear list,
>
> I have a strange problem with an old 32 bit LFS system which I'm still 
> using as a pet project ...
>
> I used 12.0 on the machine to compile 12.4-r49 to a pristine partition 
> (kernel 6.17) which went fine. Took overnight with jhalfs ... I then 
> took a precompiled kernel 6.17.9 (cross-compile on my faster i7) and 
> installed it.
>
> When I tried to boot it had some SEGV problems (I think) but booted 
> through all the init scripts to the login prompt. I could not enter a 
> username, and the last errormessage I could read was about not being 
> able to mount /boot/efi (vfat).
>
> The kern.log lets me think it has a general problem loading modules, 
> and vfat is the only module for filesystems, everything else is 
> compiled into the kernel. Modules are compressed .gz files.
>
> Strange thing is: the same kernel with its modules runs fine on the 
> older 12.0 partition without any problems, does load all the modules, 
> mounts /boot/efi and works fine. I do the 32-bit kernel cross-compile 
> for years now and have not had problems with that.
>
> Here's the relevant part from kern.log; maybe someone has an idea 
> what's going wrong?
>
> Dec  9 19:06:41 buero kernel: clocksource: tsc: mask: 
> 0xffffffffffffffff max_cycles: 0x2e02c398820, max_idle_ns: 
> 440795273435 ns
> Dec  9 19:06:41 buero kernel: clocksource: Switched to clocksource tsc
> Dec  9 19:06:42 buero kernel: random: crng init done
> Dec  9 19:06:43 buero kernel: ------------[ cut here ]------------
> Dec  9 19:06:43 buero kernel: WARNING: CPU: 0 PID: 925 at 
> mm/highmem.c:622 kunmap_local_indexed+0xd8/0xf0
> Dec  9 19:06:43 buero kernel: Modules linked in:
> Dec  9 19:06:43 buero kernel: CPU: 0 UID: 0 PID: 925 Comm: 
> (udev-worker) Not tainted 6.12.61 #1
> Dec  9 19:06:43 buero kernel: Hardware name: Hewlett-Packard HP Compaq 
> dc7600 Small Form Factor/09F8h, BIOS 786D1 v01.61 07/01/2009
> Dec  9 19:06:43 buero kernel: EIP: kunmap_local_indexed+0xd8/0xf0
> Dec  9 19:06:43 buero kernel: Code: ec ff 64 a1 84 a1 cd c1 85 c0 74 
> 26 e8 81 ea ec ff 5b 5e 5f 5d c3 8d 74 26 00 0f 0b e9 62 ff ff ff 90 
> 0f 0b 90 2e 8d 74 26 00 <0f> 0b eb 9a 8d 74 26 00 e8 cb 7a 65 00 eb d3 
> 90 2e 8d b4 26 00 00
> Dec  9 19:06:43 buero kernel: EAX: 00000001 EBX: fffba000 ECX: 
> 00046000 EDX: fffb9000
> Dec  9 19:06:43 buero kernel: ESI: c2b58000 EDI: 00000001 EBP: 
> c2ae9e60 ESP: c2ae9e54
> Dec  9 19:06:43 buero kernel: DS: 007b ES: 007b FS: 00d8 GS: 0033 SS: 
> 0068 EFLAGS: 00010206
> Dec  9 19:06:43 buero kernel: CR0: 80050033 CR2: 02082a1c CR3: 
> 02e2d000 CR4: 000006d0
> Dec  9 19:06:43 buero kernel: Call Trace:
> Dec  9 19:06:43 buero kernel:  module_decompress+0x12e/0x230
> Dec  9 19:06:43 buero kernel:  init_module_from_file+0x92/0xb0
> Dec  9 19:06:43 buero kernel:  __ia32_sys_finit_module+0x115/0x2c0
> Dec  9 19:06:43 buero kernel:  ia32_sys_call+0x1814/0x1e30
> Dec  9 19:06:43 buero kernel:  __do_fast_syscall_32+0x50/0xd0
> Dec  9 19:06:43 buero kernel:  do_fast_syscall_32+0x29/0x60
> Dec  9 19:06:43 buero kernel:  do_SYSENTER_32+0x15/0x20
> Dec  9 19:06:43 buero kernel:  entry_SYSENTER_32+0x98/0xf9
> Dec  9 19:06:43 buero kernel: EIP: 0xb7f52579
> Dec  9 19:06:43 buero kernel: Code: b8 01 10 06 03 74 b4 01 10 07 03 
> 74 b0 01 10 08 03 74 d8 01 00 00 00 00 00 00 00 00 00 00 00 00 00 51 
> 52 55 89 e5 0f 34 cd 80 <5d> 5a 59 c3 90 90 90 90 8d 76 00 58 b8 77 00 
> 00 00 cd 80 90 8d 76
> Dec  9 19:06:43 buero kernel: EAX: ffffffda EBX: 0000002b ECX: 
> b687944f EDX: 00000004
> Dec  9 19:06:43 buero kernel: ESI: 0190ff50 EDI: 01910510 EBP: 
> 00000000 ESP: bfed78fc
> Dec  9 19:06:43 buero kernel: DS: 007b ES: 007b FS: 0000 GS: 0033 SS: 
> 007b EFLAGS: 00000292
> Dec  9 19:06:43 buero kernel: ---[ end trace 0000000000000000 ]---
> Dec  9 19:06:43 buero kernel: ------------[ cut here ]------------
> Dec  9 19:06:43 buero kernel: DEBUG_LOCKS_WARN_ON(current->kmap_ctrl.idx)
> Dec  9 19:06:43 buero kernel: WARNING: CPU: 0 PID: 925 at 
> include/linux/highmem-internal.h:17 syscall_exit_to_user_mode+0xfa/0x120
> Dec  9 19:06:43 buero kernel: Modules linked in: usb_common
> Dec  9 19:06:43 buero kernel: CPU: 0 UID: 0 PID: 925 Comm: 
> (udev-worker) Tainted: G        W          6.12.61 #1
> Dec  9 19:06:43 buero kernel: Tainted: [W]=WARN
> Dec  9 19:06:43 buero kernel: Hardware name: Hewlett-Packard HP Compaq 
> dc7600 Small Form Factor/09F8h, BIOS 786D1 v01.61 07/01/2009
> Dec  9 19:06:43 buero kernel: EIP: syscall_exit_to_user_mode+0xfa/0x120
> Dec  9 19:06:43 buero kernel: Code: e8 4b 9e b9 ff 85 c0 0f 84 6f ff 
> ff ff 8b 0d e0 58 c0 c1 85 c9 0f 85 61 ff ff ff 68 2e 11 9b c1 68 86 
> 9e 9a c1 e8 36 ef 84 ff <0f> 0b 58 5a e9 49 ff ff ff 2e 8d 74 26 00 e8 
> a3 37 8d ff e9 0b ff
> Dec  9 19:06:43 buero kernel: EAX: 0000002b EBX: 00000001 ECX: 
> 00000002 EDX: 00000000
> Dec  9 19:06:43 buero kernel: ESI: 00000000 EDI: 01910510 EBP: 
> c2ae9f84 ESP: c2ae9f70
> Dec  9 19:06:43 buero kernel: DS: 007b ES: 007b FS: 00d8 GS: 0033 SS: 
> 0068 EFLAGS: 00010086
> Dec  9 19:06:43 buero kernel: CR0: 80050033 CR2: 02082a1c CR3: 
> 02e2d000 CR4: 000006d0
> Dec  9 19:06:43 buero kernel: Call Trace:
> Dec  9 19:06:43 buero kernel:  __do_fast_syscall_32+0x5a/0xd0
> Dec  9 19:06:43 buero kernel:  do_fast_syscall_32+0x29/0x60
> Dec  9 19:06:43 buero kernel:  do_SYSENTER_32+0x15/0x20
> Dec  9 19:06:43 buero kernel:  entry_SYSENTER_32+0x98/0xf9
> Dec  9 19:06:43 buero kernel: EIP: 0xb7f52579
> Dec  9 19:06:43 buero kernel: Code: b8 01 10 06 03 74 b4 01 10 07 03 
> 74 b0 01 10 08 03 74 d8 01 00 00 00 00 00 00 00 00 00 00 00 00 00 51 
> 52 55 89 e5 0f 34 cd 80 <5d> 5a 59 c3 90 90 90 90 8d 76 00 58 b8 77 00 
> 00 00 cd 80 90 8d 76
> Dec  9 19:06:43 buero kernel: EAX: 00000000 EBX: 0000002b ECX: 
> b687944f EDX: 00000004
> Dec  9 19:06:43 buero kernel: ESI: 0190ff50 EDI: 01910510 EBP: 
> 00000000 ESP: bfed78fc
> Dec  9 19:06:43 buero kernel: DS: 007b ES: 007b FS: 0000 GS: 0033 SS: 
> 007b EFLAGS: 00000292
> Dec  9 19:06:43 buero kernel: ---[ end trace 0000000000000000 ]---
> Dec  9 19:06:43 buero kernel: Invalid ELF section hdr(type 3951375263)
> Dec  9 19:06:43 buero last message buffered 1 times
> Dec  9 19:06:43 buero kernel: ELF Spec violation: section name table 
> isn't null terminated
> Dec  9 19:06:43 buero kernel: Invalid ELF header magic: != \x7fELF
> Dec  9 19:06:43 buero kernel: Invalid ELF section hdr(type 818574085)
> Dec  9 19:06:43 buero kernel: ------------[ cut here ]------------
> Dec  9 19:06:43 buero kernel: WARNING: CPU: 0 PID: 925 at 
> mm/highmem.c:713 kmap_local_fork+0x14/0x50
> Dec  9 19:06:43 buero kernel: Modules linked in: wmi usb_common
> Dec  9 19:06:43 buero kernel: CPU: 0 UID: 0 PID: 925 Comm: 
> (udev-worker) Tainted: G        W          6.12.61 #1
> Dec  9 19:06:43 buero kernel: Tainted: [W]=WARN
> Dec  9 19:06:43 buero kernel: Hardware name: Hewlett-Packard HP Compaq 
> dc7600 Small Form Factor/09F8h, BIOS 786D1 v01.61 07/01/2009
> Dec  9 19:06:43 buero kernel: EIP: kmap_local_fork+0x14/0x50
> Dec  9 19:06:43 buero kernel: Code: 5d c3 8d 74 26 00 0f 0b eb e8 8d 
> 74 26 00 2e 8d b4 26 00 00 00 00 8b 90 14 08 00 00 85 d2 75 06 c3 2e 
> 8d 74 26 00 55 89 e5 57 <0f> 0b 8d 90 18 08 00 00 8d 88 14 08 00 00 c7 
> 80 14 08 00 00 00 00
> Dec  9 19:06:43 buero kernel: EAX: c2382400 EBX: c2382400 ECX: 
> ffffffff EDX: 00000001
> Dec  9 19:06:43 buero kernel: ESI: 00000000 EDI: c2382400 EBP: 
> c2ae9e20 ESP: c2ae9e1c
> Dec  9 19:06:43 buero kernel: DS: 007b ES: 007b FS: 00d8 GS: 0033 SS: 
> 0068 EFLAGS: 00010202
> Dec  9 19:06:43 buero kernel: CR0: 80050033 CR2: 0191943c CR3: 
> 02e2d000 CR4: 000006d0
> Dec  9 19:06:43 buero kernel: Call Trace:
> Dec  9 19:06:43 buero kernel:  copy_process+0x2b2/0x1f90
> Dec  9 19:06:43 buero kernel:  ? path_init+0x3ca/0x3f0
> Dec  9 19:06:43 buero kernel:  ? walk_component+0xb3/0x150
> Dec  9 19:06:43 buero kernel:  ? terminate_walk+0x67/0x120
> Dec  9 19:06:43 buero kernel:  ? path_lookupat+0xb2/0x150
> Dec  9 19:06:43 buero kernel:  kernel_clone+0xa2/0x3d0
> Dec  9 19:06:43 buero kernel:  ? kmem_cache_alloc_noprof+0x1a5/0x1f0
> Dec  9 19:06:43 buero kernel:  ? getname_flags.part.0+0x1c/0x180
> Dec  9 19:06:43 buero kernel:  ? putname+0x4f/0x60
> Dec  9 19:06:43 buero kernel:  ? mntput_no_expire+0x35/0x210
> Dec  9 19:06:43 buero kernel:  __ia32_sys_clone+0x65/0x80
> Dec  9 19:06:43 buero kernel:  ia32_sys_call+0x18fb/0x1e30
> Dec  9 19:06:43 buero kernel:  __do_fast_syscall_32+0x50/0xd0
> Dec  9 19:06:43 buero kernel:  do_fast_syscall_32+0x29/0x60
> Dec  9 19:06:43 buero kernel:  do_SYSENTER_32+0x15/0x20
> Dec  9 19:06:43 buero kernel:  entry_SYSENTER_32+0x98/0xf9
> Dec  9 19:06:43 buero kernel: EIP: 0xb7f52579
> Dec  9 19:06:43 buero kernel: Code: b8 01 10 06 03 74 b4 01 10 07 03 
> 74 b0 01 10 08 03 74 d8 01 00 00 00 00 00 00 00 00 00 00 00 00 00 51 
> 52 55 89 e5 0f 34 cd 80 <5d> 5a 59 c3 90 90 90 90 8d 76 00 58 b8 77 00 
> 00 00 cd 80 90 8d 76
> Dec  9 19:06:43 buero kernel: EAX: ffffffda EBX: 01200011 ECX: 
> 00000000 EDX: 00000000
> Dec  9 19:06:43 buero kernel: ESI: 00000000 EDI: b7522868 EBP: 
> bfed76b4 ESP: bfed7680
> Dec  9 19:06:43 buero kernel: DS: 007b ES: 007b FS: 0000 GS: 0033 SS: 
> 007b EFLAGS: 00000282
> Dec  9 19:06:43 buero kernel: ---[ end trace 0000000000000000 ]---
> Dec  9 19:06:43 buero kernel: Invalid ELF header magic: != \x7fELF
> Dec  9 19:06:43 buero kernel: Invalid ELF section hdr(type 3153546210)
> Dec  9 19:06:43 buero kernel: Invalid ELF header magic: != \x7fELF
> Dec  9 19:06:43 buero last message buffered 1 times
> Dec  9 19:06:46 buero kernel: Adding 3999740k swap on /dev/sda3. 
> Priority:1 extents:1 across:3999740k
> Dec  9 19:06:46 buero kernel: EXT4-fs (sda6): re-mounted 
> ddcc85dd-6fa9-413c-af1c-6151e0244138.
> Dec  9 19:06:46 buero kernel: EXT4-fs (sda6): re-mounted 
> ddcc85dd-6fa9-413c-af1c-6151e0244138 r/w.
> Dec  9 19:06:46 buero kernel: EXT4-fs (sda8): mounted filesystem 
> cf42ca86-596a-47d7-8bc7-3bea49b8c22c r/w with ordered data mode. Quota 
> mode: none.
> Dec  9 19:06:47 buero kernel: EXT4-fs (sda7): mounted filesystem 
> 37b9c980-3ce9-42b5-8e02-98c239983269 r/w with ordered data mode. Quota 
> mode: none.
> Dec  9 19:06:51 buero kernel: (udev-worker) (913) used greatest stack 
> depth: 5812 bytes left
> Dec  9 19:06:51 buero kernel: (udev-worker) (915) used greatest stack 
> depth: 5676 bytes left
>
> Stay healthy,
> Thomas
> -- Do you wanna be a legend or a passing footprint on the sands of time?
>
Hi Thomas,

I haven't seen this particular class of error in a while. "ELF Spec 
violation: section name table isn't null terminated" is very bizarre, 
especially since it works on your 12.0 partition with the same kernel! 
The strange part to me though is that your logs there show 6.12.61 as 
the kernel in use...

I'll see about reproducing this later today since I am currently working 
with a 32-bit system running the development version of LFS.

However, one thought I had is that newer versions of the kernel (6.16+) 
implemented an option called CONFIG_X86_NATIVE_CPU. Enabling that sets 
the -march=native CFLAG which tunes the kernel to the specific CPU in 
use at build time. You should check to make sure CONFIG_X86_NATIVE_CPU 
is set to N, because even though you are compiling a 32-bit kernel here, 
it will still tune to your host system's CPU anyway. I've seen this 
happen in VMs recently while working on my ISC Kea bugs.

- Doug

-- 
http://lists.linuxfromscratch.org/sympa/info/lfs-support
Unsubscribe: See the above information page