Amiga.org

Operating System Specific Discussions => AROS Research Operating System => Topic started by: OlafS3 on April 03, 2020, 02:14:04 PM

Title: Could someone test Aros on genuine Amiga?
Post by: OlafS3 on April 03, 2020, 02:14:04 PM
Hi

I have uploaded a slightly adapted version of aros vision that could run on Amiga hardware with Kick 3.0 or 3.1

minimum requirememt is some fastram

Would be interesting to know if it works

http://www.aros-platform.de/Vampire_Classic.hdf
Title: Re: Could someone test Aros on genuine Amiga?
Post by: alexscotti on April 03, 2020, 07:57:37 PM
I’m glad to try if someone points me at a tool to mount this on the Amiga. I’ve tried ‘ImageMount’ so far with no luck.
Title: Re: Could someone test Aros on genuine Amiga?
Post by: kolla on April 04, 2020, 11:55:22 AM
@OlafS3

Please learn to use lha :)

Code: [Select]
cd sys:
lha -x -a -r a aros.lha ~(aros.lha) #?
and then upload that aros.lha somewhere - this whole disk-image stuff works great on emulators and FPGA systems like MiSTer, MiST and Minimig - but sucks on real Amiga and other systems that cannot directly just use a disk image (and that includes Vampire V4 too, for that matter).
Title: Re: Could someone test Aros on genuine Amiga?
Post by: lionstorm on April 04, 2020, 10:11:22 PM
I have the same problem as Alexscotti : I can not mount your hdf image with the tool provided by AOS4 (DiskImage) in order to transfer it to my A1200...
Title: Re: Could someone test Aros on genuine Amiga?
Post by: OlafS3 on April 04, 2020, 11:24:21 PM
I have made a lha archive (like recommended) and uploaded it

http://www.aros-platform.de/aros.lha
Title: Re: Could someone test Aros on genuine Amiga?
Post by: OlafS3 on April 04, 2020, 11:52:03 PM
something did not work

I have reuploaded it as zip file

http://www.aros-platform.de/aros.zip
Title: Re: Could someone test Aros on genuine Amiga?
Post by: kolla on April 05, 2020, 05:23:15 AM
something did not work

The perfect bug report :)
What did not work?

Btw, the zip archive have a few “gliches”, some files exists twice with filenames in different casings, which indicate that your UAE host OS has case sensitive file system, which Amiga doesn’t have. When unzipping the archive, users are asked if they want to over write these files.
Title: Re: Could someone test Aros on genuine Amiga?
Post by: kolla on April 05, 2020, 06:39:24 AM
I just saw wawa's explaination for why AROSBootStrap fails, will try again on V500 and perhaps also 060 later today.
Title: Re: Could someone test Aros on genuine Amiga?
Post by: kolla on April 05, 2020, 01:40:44 PM
Well, that wasn't easy, since Wawa only posted that aros.elf on a slack channel somewhere :p
Title: Re: Could someone test Aros on genuine Amiga?
Post by: alexscotti on April 05, 2020, 06:17:50 PM
Had issues with the archive, executable bit missing on c: contents and others. Once fixed it runs arosbootsrap when it boots then gets to a grey screen and hangs. 060 machine.
Title: Re: Could someone test Aros on genuine Amiga?
Post by: kolla on April 05, 2020, 06:36:36 PM
Once fixed it runs arosbootsrap when it boots then gets to a grey screen and hangs. 060 machine.
That's what I get too, if I use vcontrol to load recent 1MB AROS kickstarts on the V500v2, I must dig around for older disk images.
Title: Re: Could someone test Aros on genuine Amiga?
Post by: F1Lupo on April 06, 2020, 12:06:22 AM
what ROM do you need to use?
Title: Re: Could someone test Aros on genuine Amiga?
Post by: alexscotti on April 06, 2020, 12:50:43 AM
First line in startup-sequence runs arosbootstrap against aros.elf file. Am I supposed to be configuring something to point at a kickstart for my machine?
Title: Re: Could someone test Aros on genuine Amiga?
Post by: tohtoris on April 11, 2020, 03:49:38 PM
Hey,

I tried the .zip-package on my A3000 (040). I unzipped (on the amiga) the package to empty 1G PFS3 partition that was set as bootable. During the installation I noticed the same problems with the packages as some others: unzipping complained about files already exiting (I used the option overwrite all). Also the first boot did not work as all the executables are missing executable bit. Then I did set the executable bit to the most obvious files (C, system, tools, wbstartup etc.) myself and tried booting again.

This time I got the Aros bootscreen and was able to enter the bootmenu. But it seems that the current setup does not recognize the A3000 scsi controller. The only bootable device seems to be df0: . Any idea how to/if it is possible to boot with A3000 scsi?
Title: Re: Could someone test Aros on genuine Amiga?
Post by: tohtoris on April 11, 2020, 09:46:09 PM
Well, i did some testing and using Chris Tonys patched scsi.device (a3000 version) and setting it as a parameter for the arosbootstrap, I was able to get it booting. I also tried the mntgfx.card P96 card file to get va2000 support. I was not able to get the boot logo displayed with the va2000, but it might work when actually booting the os. But currently the boot ends up after some time to Workbench screen with a small window reading Mount FA... Error (Attachment).
Title: Re: Could someone test Aros on genuine Amiga?
Post by: pVC on April 12, 2020, 09:31:20 AM
Uh, releasing something Amiga related as a Zip archive is never a good idea. Zip archives just mess all file attributes, don't preserve exact time stamps, and lose all file comments, to name few issues with them. It's an absolutely no-go with an OS release. Besides those executable protection bits, I bet it also lost pure and script bits too... and the whole OS doesn't work as intented anymore.
Title: Re: Could someone test Aros on genuine Amiga?
Post by: AndyFC on July 11, 2020, 09:37:32 AM
Has there been any further development of this recently?

Yesterday I downloaded, extracted and tried to run this on my real Amiga (see sig for details). I didn't get any warnings of duplicate files on extraction.

It boots and loads the kickstart alternative but, like some others, it then just resets to a light grey screen and stops - I presume it crashes because the floppy drive clicks once and then stops too.

I tried booting from an SD card and from a real IDE drive with the same results.