Limine won't install

[don570]

Tested on Easy OS 4.4 ( and previous)
With Dell optiplex 990 desktop

When I try the limine installer I get this message

uefi-warning-message.png (14.54 KiB) Viewed 2098 times

So I tried removing negate symbol in the script...
From

Code: Select all

if [ ! -e /sys/firmware/efi ];then

to

Code: Select all

if [  -e /sys/firmware/efi ];then

and the result was this message...

screenshot-alternate.png (12.92 KiB) Viewed 2098 times

_____________________________________________________
My first partition on this windows 10 machine is standard for non uefi machine..

screenshot-first-partition.png (57.55 KiB) Viewed 2098 times

______________________________________________

To help track down problem I investigated at the terminal
There is no efi

Code: Select all

# cat /sys/firmware/efi
cat: /sys/firmware/efi: No such file or directory

______________________________________
Here is dmidecode output

Code: Select all

# dmidecode -t bios
# dmidecode 3.2
Getting SMBIOS data from sysfs.
SMBIOS 2.6 present.

Handle 0x0000, DMI type 0, 24 bytes
BIOS Information
	Vendor: Dell Inc.
	Version: A23
	Release Date: 02/11/2018
	Address: 0xF0000
	Runtime Size: 64 kB
	ROM Size: 10240 kB
	Characteristics:
		PCI is supported
		BIOS is upgradeable
		BIOS shadowing is allowed
		Boot from CD is supported
		Selectable boot is supported
		EDD is supported
		5.25"/1.2 MB floppy services are supported (int 13h)
		3.5"/720 kB floppy services are supported (int 13h)
		3.5"/2.88 MB floppy services are supported (int 13h)
		Print screen service is supported (int 5h)
		8042 keyboard services are supported (int 9h)
		Serial services are supported (int 14h)
		Printer services are supported (int 17h)
		ACPI is supported
		USB legacy is supported
		BIOS boot specification is supported
		Function key-initiated network boot is supported
		Targeted content distribution is supported
	BIOS Revision: 4.6

Handle 0x002B, DMI type 13, 22 bytes
BIOS Language Information
	Language Description Format: Long
	Installable Languages: 1
		en|US|iso8859-1
	Currently Installed Language: en|US|iso8859-1

_________________________________________________________
Here is cpuinfo...

Code: Select all

# cat /proc/cpuinfo
processor	: 0
vendor_id	: GenuineIntel
cpu family	: 6
model		: 42
model name	: Intel(R) Core(TM) i5-2400 CPU @ 3.10GHz
stepping	: 7
microcode	: 0x2d
cpu MHz		: 1853.357
cache size	: 6144 KB
physical id	: 0
siblings	: 4
core id		: 0
cpu cores	: 4
apicid		: 0
initial apicid	: 0
fpu		: yes
fpu_exception	: yes
cpuid level	: 13
wp		: yes
flags		: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ht tm pbe syscall nx rdtscp lm constant_tsc arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 cx16 xtpr pdcm pcid sse4_1 sse4_2 x2apic popcnt tsc_deadline_timer aes xsave avx lahf_lm epb pti ibrs ibpb stibp tpr_shadow vnmi flexpriority ept vpid xsaveopt dtherm ida arat pln pts
vmx flags	: vnmi preemption_timer invvpid ept_x_only flexpriority tsc_offset vtpr mtf vapic ept vpid unrestricted_guest
bugs		: cpu_meltdown spectre_v1 spectre_v2 spec_store_bypass l1tf mds swapgs itlb_multihit mmio_unknown
bogomips	: 6188.36
clflush size	: 64
cache_alignment	: 64
address sizes	: 36 bits physical, 48 bits virtual
power management:

processor	: 1
vendor_id	: GenuineIntel
cpu family	: 6
model		: 42
model name	: Intel(R) Core(TM) i5-2400 CPU @ 3.10GHz
stepping	: 7
microcode	: 0x2d
cpu MHz		: 1610.784
cache size	: 6144 KB
physical id	: 0
siblings	: 4
core id		: 1
cpu cores	: 4
apicid		: 2
initial apicid	: 2
fpu		: yes
fpu_exception	: yes
cpuid level	: 13
wp		: yes
flags		: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ht tm pbe syscall nx rdtscp lm constant_tsc arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 cx16 xtpr pdcm pcid sse4_1 sse4_2 x2apic popcnt tsc_deadline_timer aes xsave avx lahf_lm epb pti ibrs ibpb stibp tpr_shadow vnmi flexpriority ept vpid xsaveopt dtherm ida arat pln pts
vmx flags	: vnmi preemption_timer invvpid ept_x_only flexpriority tsc_offset vtpr mtf vapic ept vpid unrestricted_guest
bugs		: cpu_meltdown spectre_v1 spectre_v2 spec_store_bypass l1tf mds swapgs itlb_multihit mmio_unknown
bogomips	: 6188.36
clflush size	: 64
cache_alignment	: 64
address sizes	: 36 bits physical, 48 bits virtual
power management:

processor	: 2
vendor_id	: GenuineIntel
cpu family	: 6
model		: 42
model name	: Intel(R) Core(TM) i5-2400 CPU @ 3.10GHz
stepping	: 7
microcode	: 0x2d
cpu MHz		: 1600.000
cache size	: 6144 KB
physical id	: 0
siblings	: 4
core id		: 2
cpu cores	: 4
apicid		: 4
initial apicid	: 4
fpu		: yes
fpu_exception	: yes
cpuid level	: 13
wp		: yes
flags		: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ht tm pbe syscall nx rdtscp lm constant_tsc arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 cx16 xtpr pdcm pcid sse4_1 sse4_2 x2apic popcnt tsc_deadline_timer aes xsave avx lahf_lm epb pti ibrs ibpb stibp tpr_shadow vnmi flexpriority ept vpid xsaveopt dtherm ida arat pln pts
vmx flags	: vnmi preemption_timer invvpid ept_x_only flexpriority tsc_offset vtpr mtf vapic ept vpid unrestricted_guest
bugs		: cpu_meltdown spectre_v1 spectre_v2 spec_store_bypass l1tf mds swapgs itlb_multihit mmio_unknown
bogomips	: 6188.36
clflush size	: 64
cache_alignment	: 64
address sizes	: 36 bits physical, 48 bits virtual
power management:

processor	: 3
vendor_id	: GenuineIntel
cpu family	: 6
model		: 42
model name	: Intel(R) Core(TM) i5-2400 CPU @ 3.10GHz
stepping	: 7
microcode	: 0x2d
cpu MHz		: 1600.000
cache size	: 6144 KB
physical id	: 0
siblings	: 4
core id		: 3
cpu cores	: 4
apicid		: 6
initial apicid	: 6
fpu		: yes
fpu_exception	: yes
cpuid level	: 13
wp		: yes
flags		: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ht tm pbe syscall nx rdtscp lm constant_tsc arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 cx16 xtpr pdcm pcid sse4_1 sse4_2 x2apic popcnt tsc_deadline_timer aes xsave avx lahf_lm epb pti ibrs ibpb stibp tpr_shadow vnmi flexpriority ept vpid xsaveopt dtherm ida arat pln pts
vmx flags	: vnmi preemption_timer invvpid ept_x_only flexpriority tsc_offset vtpr mtf vapic ept vpid unrestricted_guest
bugs		: cpu_meltdown spectre_v1 spectre_v2 spec_store_bypass l1tf mds swapgs itlb_multihit mmio_unknown
bogomips	: 6188.36
clflush size	: 64
cache_alignment	: 64
address sizes	: 36 bits physical, 48 bits virtual
power management:

[don570]

I tracked down the reason why limine can't be installed in Dell desktop...

Code: Select all

# dmidecode -s bios-release-date
02/11/2018

The bios release date is too late. Dell was the last manufacturer to switch over to UEFI.

_______________________________________________________

puppy install list

screenshot-limine-list.png (138.75 KiB) Viewed 2066 times

I modified the script to run.....
When I ran the limine install script it gave a huge list of puppies that I had installed.
The instructions are at the bottom of the list.
It would be better to have the instructions at the top of the list.

[BarryK]

Anyone reading this who has a Dell computer that does have UEFI firmware?

If so, would you mind posting the result of running this:

Code: Select all

# dmidecode -t bios

...I would like to know if it does explicitly state that it supports UEFI.

[BarryK]

don570,
I think that I see what the problem is. The Dell optiplex 990 was manufactured in 2011, and came with Windows 7 or Vista.
And, it seems, has legacy-BIOS.
In your case, it looks like the BIOS was upgraded, but it is still legacy-BIOS not UEFI.

So, for Limine-installer to check the firmware date is wrong in your case.
I introduced that date check because there was someone that had a computer with UEFI firmware, but 'dmidecode' did not say anything about UEFI.

So, leaves me in a quandary how to handle it. Hmmm, but if the computer is UEFI then they should have booted up in UEFI mode, not legacy-mode, so /sys/firmware/efi will exist. So...

[BarryK]

Ok, in /usr/local/limine-installer/limine-installer, I have commented-out the date detection:

Code: Select all

###bios type###
UEFIflg=0
BU="$(dmidecode -t bios | grep -o 'UEFI is supported')"
if [ "$BU" ];then
 UEFIflg=1
elif [ -e /sys/firmware/efi ];then #20220725 has booted up in efi mode.
 UEFIflg=1
#20220929 bios release date not reliable to determine if bios or uefi...
#else
# BRD="$(dmidecode -s bios-release-date)"
# BRD=${BRD##*/}
# [ $BRD -gt 2013 ] && UEFIflg=1 #20220721 changed from 2012
fi

[don570]

'dmidecode -s' allows for searching for keywords....
so I checked some of the words...

The best seems to be

Code: Select all

dmidecode -s system-product-name | grep -iq  "OptiPlex 9"

since it would include a lot of old Dell desktops.

Code: Select all

# dmidecode -s system-product-name
OptiPlex 990

# dmidecode -s chassis-manufacturer
Dell Inc.
# dmidecode -s baseboard-version
A00
# dmidecode -s baseboard-manufacturer
Dell Inc.
# dmidecode -s baseboard-product-name
0VNP2H
# dmidecode -s baseboard-version
A00
# dmidecode -s baseboard-serial-number
/45ZVTR1/CN7360417F015I/
# dmidecode -s bios-version
A23
# dmidecode -s bios-release-date
02/11/2018

[don570]

In your case, it looks like the BIOS was upgraded, but it is still legacy-BIOS not UEFI.

Yes , when I bought the computer used, I went to the Dell site and downloaded the latest firmware.
-_________________________________________________________

[BarryK]

don570,
Thanks for the info. I have inserted this in the script:

Code: Select all

SPN="$(dmidecode -s system-product-name)"
case "$SPN" in
 "OptiPlex 990") UEFIflg=0 ;;
esac

[Berto]

@ BarryK, don570 and mitsuki (and everyone else)

Welllllll, hm, I appreciate your work on limemine much, hope it will be a real good slim bootloader soon. It's looking promising, very nice and easy.

Now here comes the but: I'm using grub2, not as passion, just, coz there's no alternative at the moment.

When I want to boot BSD, grub2 does, when I want to boot androidx86, grub2 does, when I want to boot an illumus kernel (Openindiana, based on Solaris 10), grub2 does, aaaand: if I want to boot a Xen kernel, it does also!

Well, the last point is only half true: It boots the Xen kernel shipped with the distro, but no others.

Experimental I've Installed Xen on Debian Strech, OpenSuse and CentOS5 on my Asus Laptop (i7 core, Nvidia 540M, 4 GB RAM). Grub boots every Xen kernel and automatically gives the alternative to boot the standard kernel of the distro, but it doesn't recognize the Xen kernel of the other distros .....

So .... I think there's still a long way to go for Limemine... I always searched an alternative to grub2, but I don't see any at the moment....

This post shouldn't sound negative, see it as inspiration please, Sirs.

Btw. Easy 4.4 booted without any probs with grub2.

Regards,
Berto

[don570]

don570,
Thanks for the info. I have inserted this in the script:

The problem is that Dell manufactured a huge number of computers in the optiplex 900 series.
They were named 960 970 980 990 . Numbers like that.

There was some difference in the motherboards,
but they all used the old-fashion MBR to boot up so Grub works well with them.

By the way Dell is now using 4 numbers rather than 3 in the optiplex series. like 6000 7000
_________________________________________________________________

[BarryK]

don570,
Thanks for the info. I have inserted this in the script:

The problem is that Dell manufactured a huge number of computers in the optiplex 900 series.
They were named 960 970 980 990 . Numbers like that.

There was some difference in the motherboards,
but they all used the old-fashion MBR to boot up so Grub works well with them.

By the way Dell is now using 4 numbers rather than 3 in the optiplex series. like 6000 7000
_________________________________________________________________

Yes, I discovered that and modified the code:

Code: Select all

# ref: https://engineering.purdue.edu/ECN/Support/KB/Docs/DellModelYears
SPN="$(dmidecode -s system-product-name)"
case "${SPN/ */}" in
 OptiPlex)
  case "${SPN/* /}" in
   990|790|980|780|960|760|755|745) UEFIflg=0 ;;
   GX620|GX280|GX270|170L|GX260|GX240) UEFIflg=0 ;;
  esac
 ;;
esac

...hopefully they all return "OptiPlex" followed by space then the number.

https://engineering.purdue.edu/ECN/Supp ... ModelYears

I also read that "dmidecode -s system-product-name" returns nothing on modern Dell computers.

[don570]

Dell computers are like cockroaches.
A local store is selling refurbished Dell 7040 Sff I5 processor $179 Canadian dollars
This is equivalent to what I have but was manufactured in 2015 and probably uses UEFI.

https://www.factorydirect.ca/dell-5040- ... 0gbdvdw10h
_______________________________

[BarryK]

@Caramel
Thanks for the fixes.
latest /usr/local/limine-installer folder attached.

[BarryK]

Limine installer is maintained as part of woofQ, which is now on github.

This is where to find Limine installer:

https://github.com/bkauler/woofq/tree/m ... -installer

...anyone who is a git expert, knows how to download just that folder and its contents?

For ongoing discussion about Limine, note that there have been other threads:

"Limine installer PET version 1.3"
viewtopic.php?t=6406

"Limine installer, request for help"
viewtopic.php?t=6284

EasyOS 4.5.1 has Limine version 3.20221014.1, but that is not the latest.

Anyone having issues with Limine?

If so, maybe we need to bump to the latest version of Limine.

And probably the installer may need fixing; it is still a work-in-progress.

[bugnaw333]

Limine bootloader does not work on my system. I dunno why?

[BarryK]

Limine bootloader does not work on my system. I dunno why?

You will have to post what there is installed, what distros, and where.
Also, is it UEFI or legacy-BIOS
Is the drive GPT or traditional MBR partition table
Make and model of computer

[Xyz2022]

Limine fails in ACER machine. The Output of "dmidecode -t bios" is attached. The PC boots from a USB and OS works fine. You just have to select the boot device every time. Also, WINDOWs was booting only in Legacy mode but Not in UEFI mode from a SSD. After Limine, WINDOWS refuses to boot with a blank screen.

# dmidecode 3.2
Getting SMBIOS data from sysfs.
SMBIOS 3.0.0 present.

Handle 0x0000, DMI type 0, 24 bytes
BIOS Information
Vendor: Acer
Version: R01-A3
Release Date: 12/28/2015
Address: 0xF0000
Runtime Size: 64 kB
ROM Size: 8192 kB
Characteristics:
PCI is supported
BIOS is upgradeable
BIOS shadowing is allowed
Boot from CD is supported
Selectable boot is supported
BIOS ROM is socketed
EDD is supported
5.25"/1.2 MB floppy services are supported (int 13h)
3.5"/720 kB floppy services are supported (int 13h)
3.5"/2.88 MB floppy services are supported (int 13h)
Print screen service is supported (int 5h)
Serial services are supported (int 14h)
Printer services are supported (int 17h)
ACPI is supported
USB legacy is supported
BIOS boot specification is supported
Targeted content distribution is supported
UEFI is supported
BIOS Revision: 5.11
Firmware Revision: 1.2

Handle 0x0073, DMI type 13, 22 bytes
BIOS Language Information
Language Description Format: Long
Installable Languages: 10
en|US|iso8859-1
zh|CN|unicode
<BAD INDEX>
<BAD INDEX>
<BAD INDEX>
<BAD INDEX>
<BAD INDEX>
<BAD INDEX>
<BAD INDEX>
<BAD INDEX>
Currently Installed Language: en|US|iso8859-1

[Misaligned]

Same problems with "limine installer".

My computer is no name brand with Biostar Motherboard, Bios date is 2017.
I can choose between
Legacy only
UEFI only - or
Legacy and UEFI

By default I have it set to Legacy, which gives me the same message
as shown in the first post. If I switch to UEFI only, I get the message
"Nothing Found, Aborting".

What I did -
Booted Easy-4.4.1 from USB (no HD available)
plugged in second USB with 2 partitions
Partition 1, FAT16, boot, esp
Partition 2, ext, with vmlinuz and compressed system in a directory /boot

I still get "Nothing Found" even if I mount those 2 partitions -
are they supposed to be mounted by default?

[bugnaw333]

Limine bootloader does not work on my system. I dunno why?

You will have to post what there is installed, what distros, and where.
Also, is it UEFI or legacy-BIOS
Is the drive GPT or traditional MBR partition table
Make and model of computer

sda1--Windows7
sda2--Windows10
MBR
Fujitsu Futro MA574
Secured BIOS

Best regards,
Reynaldo

[Xyz2022]

Please do a favor to the present and future users of EASYOS.

Just add an example for /EFI/BOOT/refind.conf at either web page:
https://easyos.org/install/how-to-insta ... drive.html
Or
https://easyos.org/install/install-to-p ... nager.html

Thanks

[Misaligned]

@Xyz2022, Thanks for listing the links.

Tried it the manual way as described at the 1st link (Limine)
No luck with that one either, see attached screenshot...

limine-copy-error.png (34.38 KiB) Viewed 1867 times

Same error message, if I boot in Legacy Mode or
in EFI Mode

[BarryK]

@Xyz2022, Thanks for listing the links.

Tried it the manual way as described at the 1st link (Limine)
No luck with that one either, see attached screenshot...

limine-copy-error.png

Same error message, if I boot in Legacy Mode or
in EFI Mode

You were trying to drag-and-drop, as described in that link?

The error that you are reporting has nothing to do with Limine, it is a simple copy operation.
Which should work.

The error message about memory is puzzling. The target is a vfat partition, you can test drag-and-drop by just copying any file.
If it reports the same error, it would suggest something wrong with the target partition.

Thinking about it some more, really puzzled, have never seen that error message before.
You can perform the same copy operation in a terminal:

Code: Select all

# cp -a /tmp/mount-img-1-vfat-loop1/limine.cfg /mnt/sdb1/

It does seem to be some kind of problem with how /mnt/sdb1 is mounted.
I presume it is a vfat filesystem, with sufficient free space. You can verify that it is mounted OK:

Code: Select all

# mount | grep sdb1

You could even try creating a file in /mnt/sdb1, to verify that it is mounted read-write.

[Misaligned]

@BarryK,
Yes it was a drag and drop operation using Rox.
The partition I tried to copy to is vfat (Fat12), 7M
with boot and esp flag set. Just like the Easy Boot
Partition.

Meanwhile I just "dd"-ed the Easy Boot partition
to this particular partition (same usb drive), which
helped , somewhat . The boot process is not
finishing.

The Limine Boot Screen shows up, and after the preset
timeout it continues, but does not finish ... I can see
the vmlinuz message showing up and then it stops,
screen stays black.

It could be that I need a waitdev=x since is only a usb2
drive, but I am not sure if Limine supports this option.
I briefly looked at documentation at github, but didn't
find anything. It could be just a misconfigured cfg file?

I will try this again with a different drive.

[Xyz2022]

The best thing would be that the next EASY OS update 4.5.3 contains the Latest version of Limine.

It is a bit too much to learn linux coding from www.google.com when you just using PC for fairly routine tasks.

[BarryK]

@BarryK,
Yes it was a drag and drop operation using Rox.
The partition I tried to copy to is vfat (Fat12), 7M
with boot and esp flag set. Just like the Easy Boot
Partition.

Meanwhile I just "dd"-ed the Easy Boot partition
to this particular partition (same usb drive), which
helped , somewhat . The boot process is not
finishing.

The Limine Boot Screen shows up, and after the preset
timeout it continues, but does not finish ... I can see
the vmlinuz message showing up and then it stops,
screen stays black.

It could be that I need a waitdev=x since is only a usb2
drive, but I am not sure if Limine supports this option.
I briefly looked at documentation at github, but didn't
find anything. It could be just a misconfigured cfg file?

I will try this again with a different drive.

dd should not be used to copy partitions, only entire drive images.

You won't need "waitdev" on the kernel commandline.

I'm wondering if your problem has something to do with this:

https://bkhome.org/news/202212/boot-par ... -vfat.html

...I don't know why, copy should work regardless if mounted as msdos or vfat.

Though, you could try a bigger boot-partition in the internal drive and use fat16 instead of fat12. Let's see minimum size:

https://recoverit.wondershare.com/file- ... ystem.html

Quoting:

Criteria

Code: Select all

FAT16 File System
First Introduced 	1984
Minimum Partition Size:
8 MB (with 128-byte sectors)
32 MB (with 512-byte sectors)
256 MB (with 4KB sectors)

512-byte sectors is what we normally expect, so create a 32MiB boot-partition with gparted and format it as fat16.

[Misaligned]

@Xyz2022, yo man, it's painful, but I am on a mission . I am not making fun of you.

@BarryK, thanks for giving me all these pointers. Well, I got Limine to work, by copying
the bootloader on to my boot partition, folowing the steps you describe in your blog.

I just recreated all the partitions again (same usb drive), this time I used fdisk. The first
time I used cfdisk.

cfdisk seems to have some issues? The reason I am saying this, after I deleted the partitions
and recreated them (in cfdisk), the previously copied files showed up again? I tried this several
times, I even exited cfdisk after deleting everything, checked with pmount and blkid, no partitions,
restarted cfdisk, recreated the partitions and ... and the files where back, on both partitions.

I finally deleted the partitions with gparted and then I used fdisk to create my partitions. Did not
want to dedicate 16M for the boot partition and gparted does not go below this.

I do wonder though about the 8M min for FAT12 .... Easy Os has a 7M vfat Boot Partition? But then
this could be because blkid does not report this correctly, I will look into this some more.

The reason I was mentioning the waitdev option, FatDog needs it, well at least in my case with
the grub4dos loader.

[Misaligned]

Just wanted to mention, blkid gives me SEC_TYPE=msdos for Easy Boot and
the Boot Partition I created, just checked it.

[Misaligned]

Finally solved Limine install ... after sorting out some hardware issues.

I still doesn't work using the Limine-Setup, have to do it manually.
Arch-Wiki was helpful .

All manually done
First I created my two partitions
-boot partition, 7M, Fat12, EFI (reflects Easy Boot Partition)
-2nd partition to hold vmlinuz and initrd
-created /EFI/BOOT on boot-partition
-copied BOOTIA32.EFI and BOOTX64.EFI to /EFI/BOOT
-copied limine.sys to /limine.sys on boot-partition
-ran "limine-deploy /dev/sdx" (sdx - your name here/there, no partition #)
the device was unmounted at this point, not sure if it makes a difference,
if it is mounted? Not tested.

Seems like you don't need the last step if you only boot in EFI mode, limine.sys is
also only neccessary if you plan on being able to boot legacy bios.

Anyways ... the above mentioned steps, let's me boot in legacy or Efi mode.
And, just wanted to mention I copied all the Limine files from /usr/share/limine,
which is present in EasyOs. Not sure if it makes a difference, especially the limine.sys.
Haven't tried all these options.
Ohh ... and don't forget, you need the limine.cfg as well. Adjust to your needs.

One oddity I noticed creating the /EFI directory ... although I used all capital letters,
it only shows up with lower case letters (rox and cli). In Fat Dog it shows up correctly???
The BOOT sub-directory shows up correct (all capitals).

Hope this helps?

[Misaligned]

Just saw release notes on 4.5.3.

Maybe the msdos Boot-Partition is responsible for capital letters showing up as lowercase
letters. Does not really sound plausible? Just a thought.

[Misaligned]

@BarryK,
Just a thought/suggestion -

Maybe the limine-setup could be modified to just
create the directories /EFI/BOOT copy the
required files *.EFI into it, and copy the limine.sys
in the root of that partition and maybe a skeleton
limine.cfg. Than run limine-deploy. This would of
course require the user to edit the cfg file to fit
the needs.

This would elimate to check what Bios is in that
particular machine. It would boot either/or. I asume.

Or make it a multi step process with user interaction
Step1 - automate the creation of the partitions (boot and ext fs)
Step2 - copy the neccessary boot loader files
Step3 - ask for location of vmlinuz/initrd
Step4 - copy those to ext fs partition
Step5 - run limine-deploy.
Step6 - create limine.cfg

Like I said, just a thought.