LinuxQuestions.org

LinuxQuestions.org (/questions/)
-   Slackware (https://www.linuxquestions.org/questions/slackware-14/)
-   -   elilo-c(line 77): Kernel file not found umlinuz-huge-5.15.154 (https://www.linuxquestions.org/questions/slackware-14/elilo-c-line-77-kernel-file-not-found-umlinuz-huge-5-15-154-a-4175736275/)

babydr 04-20-2024 08:30 PM

elilo-c(line 77): Kernel file not found umlinuz-huge-5.15.154
 
1 Attachment(s)
Hello All ,
A Very frustrating issue here , the item that is being complained about IS there and IS a bzImage file ...
the one hint the 'elilo.c(line 77):Kernel file' could either be a line in the original elilo.c file or it thinks there is a line 77 in lilo.conf and the rest of the file name is being stripped off ...

More info comeing in a bit . Enabling the serial console .

I'll also look into the elilo.c file , to hopefully garner some insight .

The vmlinuz-huge-5.15.154 is a locally complied kernel .

Please see what I have below & the attched elilo.conf file .

Below is hand typed in from a photo ...

Code:

read_config=Success
selected chooser simple
command list (must be first character):
=:print device list, %:print variable list, &:print paths, ?:help
TAB:print label infrmation
default file path: scsi0:\EFI\Slackware\
config file: \EFI\Slackware\elilo.conf

ELILO boot: ..........
kernel is 'vmlinuz-huge-5.15.154'
arguments are ' root=PARTUUID=4856b1c5-xxxx-yyyy-zzzz-xyzxyzxyzxyz vg...
sbcore-autosuspend=-1 ro ipu6.disable=1 ro'
initrd is 'vmlinuz-huge-5.15.154-initrd-gz'

elilo.c(line 77):Kernel file not  found vmlinuz-huge-5.15.154

First one is visible to the one time boot menu in elilo ...
Next one says Not Found from above ...


Code:


# file /boot/efi/EFI/Slackware/vmlinuz-huge-5.15.145
/boot/efi/EFI/Slackware/vmlinuz-huge-5.15.145: Linux kernel x86 boot executable bzImage, version 5.15.145 (root@z-mp.slackware.lan) #1 SMP PREEMPT Sun Dec 24 00:07:06 CST 2023, RO-rootFS, swap_dev 0XB, Normal VGA

# file /boot/efi/EFI/Slackware/vmlinuz-huge-5.15.154
/boot/efi/EFI/Slackware/vmlinuz-huge-5.15.154: Linux kernel x86 boot executable bzImage, version 5.15.154 (root@replaceme.baby-dragons.com) #1 SMP PREEMPT Fri Apr 12 18:02:35 AKDT 2024, RO-rootFS, swap_dev 0XB, Normal VGA

# file /boot/efi/EFI/Slackware/vmlinuz-huge-5.15.145-initrd.gz
/boot/efi/EFI/Slackware/vmlinuz-huge-5.15.145-initrd.gz: gzip compressed data, max compression, from Unix, original size modulo 2^32 33037824

# file /boot/efi/EFI/Slackware/vmlinuz-huge-5.15.154-initrd.gz
/boot/efi/EFI/Slackware/vmlinuz-huge-5.15.154-initrd.gz: gzip compressed data, max compression, from Unix, original size modulo 2^32 29868544


Aeterna 04-21-2024 07:08 AM

Shorten vmlinuz name.

babydr 04-21-2024 04:41 PM

@Aeterna , Thank You for the idea .

Although , please then tell me why ?

vmlinuz-huge-5.15.154

vmlinuz-huge-5.15.145

The first doesn't work the second one boots fine .

If the first one which doesn't work is requested to shorten the filename .
The filenames are the same length .

Now if it's that the file content that may need to be reduced , that would be odd too ...

Code:

--- Not booting ---
11378 -rwxr-xr-x 1 root root 11651040 Apr 13 12:49 vmlinuz-huge-5.15.154*
 9959 -rwxr-xr-x 1 root root 10197826 Apr 13 12:50 vmlinuz-huge-5.15.154-initrd.gz*

 --- Does boot ---
11350 -rwxr-xr-x 1 root root 11622048 Apr  4 19:41 vmlinuz-huge-5.15.145*
10712 -rwxr-xr-x 1 root root 10968290 Apr  4 19:41 vmlinuz-huge-5.15.145-initrd.gz*

This is the reason for my confusion .
That there is nothing substantially odd between the two .

.

kingbeowulf 04-21-2024 05:34 PM

If you are booting huge kernels, you do not need an initrd.
Since you are booting x86, both kernels should be huge-smp, or generic-smp with appropriate initrd. "kernel not found" can also mean "no suitable kernel found" and not that the file is missing. There may have been changes to 5.15.154 that now require you to use smp kernel - especially if you are on a x86_64 CPU.

Where did you get 5.15.154? I do not see it in the Official Slackware ChangeLog. Did you compile it yourself?

babydr 04-21-2024 06:06 PM

@kingbeowulf , Thank you for the response .

Yes this is locally built kernel , I did mention that in the original post .
download of sources from www.kernel.org &
I copied the config-huge-5.15.145.x64 from /boot/ ...

Code:


cd /usr/src/linux
make mrproper
cp /boot/config-huge-5.15.145.x64 .config
make olddefconfig
make V=1 KBUILD_VERBOSE=1 prepare
make -j 3 V=1 KBUILD_VERBOSE=1 all
make -j 3 V=1 KBUILD_VERBOSE=1 modules_install

# since /usr/src/linux/arch/x86_64/boot/bzImage is a link to ../../x86/boot/bzImage
cp /usr/src/linux/arch/x86/boot/bzImage /boot/efi/EFI/Slackware/vmlinuz-huge-5.15.154

mkinitrd ...

Had a brain fart & thought I might have used a differant /dev/disk/by-uuid , in the mkinitrd .
It seems I did not ...

Code:


# gunzip --stdout /boot/efi/EFI/Slackware/vmlinuz-huge-5.15.145-initrd.gz | strings | grep -E '(sdc1|23b060b3-4b21-45ba-94f2-0fd9e7dee41f)'
/sbin/mkinitrd -c -R -k 5.15.145 -f ext4 -r UUID=23b060b3-4b21-45ba-94f2-0fd9e7dee41f -m megaraid_sas:mpt3sas:usb-storage:xhci-pci:ohci-pci:ehci-pci:xhci-hcd:uhci-hcd:ehci-hcd:hid:usbhid:i2c-hid:hid_generic:hid-asus:hid-cherry:hid-logitech:hid-logitech-dj:hid-logitech-hidpp:hid-lenovo:hid-microsoft:hid_multitouch:jbd2:mbcache:crc32c_intel:crc32c_generic:ext4 -u -o /boot/initrd-5.15.145-202404041941.gz
UUID=23b060b3-4b21-45ba-94f2-0fd9e7dee41f
070701012E01E5000061B0000000000000000600000001660E1FDA00000000000000080000002100000008000000210000000900000000dev/sdc1

# gunzip --stdout /boot/efi/EFI/Slackware/vmlinuz-huge-5.15.154-initrd.gz | strings | grep -E '(sdc1|23b060b3-4b21-45ba-94f2-0fd9e7dee41f)'
/sbin/mkinitrd -c -R -k 5.15.154 -f ext4 -r UUID=23b060b3-4b21-45ba-94f2-0fd9e7dee41f -m megaraid_sas:mpt3sas:usb-storage:xhci-pci:ohci-pci:ehci-pci:xhci-hcd:uhci-hcd:ehci-hcd:hid:usbhid:i2c-hid:hid_generic:hid-asus:hid-cherry:hid-logitech:hid-logitech-dj:hid-logitech-hidpp:hid-lenovo:hid-microsoft:hid_multitouch:jbd2:mbcache:crc32c_intel:crc32c_generic:ext4 -u -o /boot/initrd-5.15.154-202404131247.gz
UUID=23b060b3-4b21-45ba-94f2-0fd9e7dee41f
070701012E01E5000061B0000000000000000600000001660E1FDA00000000000000080000002100000008000000210000000900000000dev/sdc1

The rest you know about ...

Aeterna 04-21-2024 07:54 PM

Quote:

Originally Posted by babydr (Post 6497557)
@Aeterna , Thank You for the idea .

Although , please then tell me why ?

vmlinuz-huge-5.15.154

vmlinuz-huge-5.15.145

The first doesn't work the second one boots fine .

If the first one which doesn't work is requested to shorten the filename .
The filenames are the same length .

Now if it's that the file content that may need to be reduced , that would be odd too ...

Code:

--- Not booting ---
11378 -rwxr-xr-x 1 root root 11651040 Apr 13 12:49 vmlinuz-huge-5.15.154*
 9959 -rwxr-xr-x 1 root root 10197826 Apr 13 12:50 vmlinuz-huge-5.15.154-initrd.gz*

 --- Does boot ---
11350 -rwxr-xr-x 1 root root 11622048 Apr  4 19:41 vmlinuz-huge-5.15.145*
10712 -rwxr-xr-x 1 root root 10968290 Apr  4 19:41 vmlinuz-huge-5.15.145-initrd.gz*

This is the reason for my confusion .
That there is nothing substantially odd between the two .

1) I don't use initrd. Problem persists if you have two-three "long" vmlinuz names. I noticed this some time ago as I have only custom kernels (all without initrd by the way) at least three at a time.
You can have one longer name though that is why users following default Slackare kernel don't see the problem. The issue is not related to elilo per se as I never upgraded elilo since 2021. The problem was gone for some time but recently reappeared. I never mentioned it because elilo is dying so I doubt that there is an interest in fixing the issue.


All times are GMT -5. The time now is 12:25 PM.