New topic to pick up where the topic began in another thread topic.

uboot is no longer at sourceforge, it's now hosted by Denx at this git repository

The AmigaOne specifics can be found under board/MAI/AmigaOneG3SE path. Do the XE and Micro inherit everything identically fromt eh G3SE stuff, or should there be differences? The history shows activity in there up to Oct 2009!

Do we need to go back to the March 2005 Hyperion release?

If we tinker with uboot, we may be testing in Linux only land, due to lack of being able to add the OS4 special sauce to the new image.

I'm thinking up a uboot switch card, comparable to a classic kickstart ROM switcher. Put the original flash chip in the primary, readonly socket, so meddling with uboot won't leave you with a bricked machine. Secondary flash chip is programmable with your test image, reboot and see what happens, though limited to linux testing I guess. If bad things happen, switch back to primary flash and try again.

Last edited by billt on 02-Feb-2010 at 12:07 AM.

_________________
All glory to the Hypnotoad!

@billt

I would like update uboot for A1

@Kicko

Quote:

I would like update uboot for A1

Just nag on the betatesters a bit or send it for a repair


@billt

There is some old threads on the subject with som good info as well, I'll see if I can find some of those (Or number6 has links saved somewhere).

Initially a good starting point could be to make a list of what is lacking with the current one from 2005 as well as bugs that needs to be ironed out.

_________________
Site admins are people too..pooff!

@billt

U-Boot is GPL. If the special sauce is a compiled part of U-Boot, then the special sauce is also GPL.

_________________
Sam440ep-flex 733 MHz/1 GB RAM/Radeon 9250/AmigaOS4.1 Update 2
borked A1200/Blizzard1260+SCSI-IV/Z4+MediatorZIV/Deneb/Voodoo3/CatweaselMk3
more borked A1200/MBX1200z/Indivision
A500/clockport/RRNet
A600/A603

@Trev

Quote:

U-Boot is GPL. If the special sauce is a compiled part of U-Boot, then the special sauce is also GPL.

Yes, good luck with that, BTW.

_________________
Common sense - So rare it's almost like a super power

@zerohero

We'll see how my request goes. Hopefully, Hyperion respects the copyrights of others and doesn't condone copyright infringement or violations of software licensing terms.

_________________
Sam440ep-flex 733 MHz/1 GB RAM/Radeon 9250/AmigaOS4.1 Update 2
borked A1200/Blizzard1260+SCSI-IV/Z4+MediatorZIV/Deneb/Voodoo3/CatweaselMk3
more borked A1200/MBX1200z/Indivision
A500/clockport/RRNet
A600/A603

@Trev

The "dongle code" that is added to the ROM image is not part of U-Boot, not referenced by U-Boot and not GPL. It just happens to be stored in spare space within the ROM and is not GPL or likely to be released as GPL.

Since we don't have it, we can't produce working binary contents that you could load into a ROM.

_________________
cheers
tony

Hyperion Support Forum: http://forum.hyperion-entertainment.biz/index.php

@tonyw

If it's part of the ROM image, we already have it. Regardless, anyone that asks is entitled to their modified U-Boot sources.

_________________
Sam440ep-flex 733 MHz/1 GB RAM/Radeon 9250/AmigaOS4.1 Update 2
borked A1200/Blizzard1260+SCSI-IV/Z4+MediatorZIV/Deneb/Voodoo3/CatweaselMk3
more borked A1200/MBX1200z/Indivision
A500/clockport/RRNet
A600/A603

@tonyw

Quote:

tonyw wrote: Since we don't have it, we can't produce working binary contents that you could load into a ROM.

Surely, the public code should at least allow Linux to boot? That would be enough to test things out.

Hans

_________________
http://hdrlab.org.nz/ - Amiga OS 4 projects, programming articles and more. Home of the RadeonHD driver for Amiga OS 4.x project.
https://keasigmadelta.com/ - More of my work.

@billt

Quote:

billt wrote: The AmigaOne specifics can be found under board/MAI/AmigaOneG3SE path. Do the XE and Micro inherit everything identically fromt eh G3SE stuff, or should there be differences?

AFAIK the A1G3, A1XE and uA1 have different PCI interrupt routing. Naturally u-boot needs to know about this, when it sets up the PCI devices.

Quote:

The history shows activity in there up to Oct 2009!

As you can see in the GIT shortlog, these are mostly cleanups and no bugfixes. The last entry from an OS4 developers seems to be from 2002.

Quote:

I'm thinking up a uboot switch card, comparable to a classic kickstart ROM switcher. Put the original flash chip in the primary, readonly socket, so meddling with uboot won't leave you with a bricked machine. Secondary flash chip is programmable with your test image, reboot and see what happens, though limited to linux testing I guess. If bad things happen, switch back to primary flash and try again.

I already designed such a card. I just didn't build it up yet. See here.

_________________
A1SE: G3@600MHz, 2GB, 1GBit network card
A1XE: G4@933MHz, 2GB, refitted AC'97 codec
microA1: G3@800MHz, 1GB

- A1 Linux support -

@tomazkid

My main complaint is that it cannot scan for PCI devices across bridges. At least not very far. I have some things I'd like to tinker with that involve a bridge and a switch (logically a bridge or set of them) before getting to any devices. I also need to add a software workaround for some pcie1 devices that don't properly honor reserved bits and thus are not scanned properly by the pcie2 switch.

_________________
All glory to the Hypnotoad!

@Geri

Quote:

I already designed such a card. I just didn't build it up yet. See here.

Neat! Why not for Micros as well?

Can it "autoswitch" like some kickstart switchers, or does it require a physical toggle?

One thing I was thinking of, to work around possibly not getting the OS4 special part in any test images, is to figure out where it lives in the original chip, and have a pld detect addressing and switch over to the other chip to execute whatever the special OS4 stuff is and boot the OS. We all have a copy of that code in there somewhere, just need to jump and switch to it...

_________________
All glory to the Hypnotoad!

@billt

Quote:

billt wrote: @Geri Quote: I already designed such a card. I just didn't build it up yet. See here. Neat! Why not for Micros as well?

Because I don't know the dimensions the PCB should/can have. Otherwise it shouldn't be a problem.

Quote:

Can it "autoswitch" like some kickstart switchers, or does it require a physical toggle?

It requires a physical toggle. It was very hard to route all the traces from the first flash ROM to the second one, as is is just a double layer PCB (cheaper production). Additional components would have made the layout much harder. But there is always room for improvement!

Quote:

One thing I was thinking of, to work around possibly not getting the OS4 special part in any test images, is to figure out where it lives in the original chip, and have a pld detect addressing and switch over to the other chip to execute whatever the special OS4 stuff is and boot the OS. We all have a copy of that code in there somewhere, just need to jump and switch to it...

The question is, if it is executable code, or if it is some sort of signature of the whole flash ROM. In that case it may be impossible to simply switch to the original ROM. I guess you would have to use a CPU on chip debugger (like the BDI2000) to find that code.

IMHO it would be much simpler, if the A1 U-boot could be developed further in a publicly available GIT repository. Hyperion could automatically build U-boot images every day/week/whatever from this reqository (e.g. Debian does that with their packages) and could put them on their OS4 download homepage (which requires a log in anyway). Well, it could be sooo easy....

_________________
A1SE: G3@600MHz, 2GB, 1GBit network card
A1XE: G4@933MHz, 2GB, refitted AC'97 codec
microA1: G3@800MHz, 1GB

- A1 Linux support -

@billt etal

Wouldn't it help also to talk to @olegil and learn the details about :

Quote:

I have a riser design here, but it's for a riser with 2xPCI and 1xPCIe, so it has two bridges on it. Since this is primarily for the µA1 I kinda put it on hold until the µA1 can handle bridge chips (which it doesn't at the moment).

Source: Post #27

iirc, there was a demo of something working in early 2006. @shoe could also provide some details.

#6

_________________
This posting, in its entirety, represents solely the perspective of the author.
*Secrecy has served us so well*

@Geri

Quote:

I already designed such a card. I just didn't build it up yet.

I can't beleive no one replied in that thread! That was a great idea! Everyone else has a dual BIOS.

BTW what point is there for a RADIish, dual core like, dual BIOS in a PC?

@billt

Quote:

figure out where it lives in the original chip

Is it really that hard? We have the firmware code in UBoot updates. And I don't need to tell anyone the magic is in the boota command.

@Trev

Quote:

U-Boot is GPL. If the special sauce is a compiled part of U-Boot, then the special sauce is also GPL.

Things that are distributed on a Linux CD are not automatically GPL'ed just because they are on the same CD.

Things that are distributed in an U-Boot ROM are not automatically GPL'ed.

_________________
Seriously, if you want to contact me do not bother sending me a PM here. Write me a mail

@Rogue

Quote:

Rogue wrote: @Trev Things that are distributed on a Linux CD are not automatically GPL'ed just because they are on the same CD. Things that are distributed in an U-Boot ROM are not automatically GPL'ed.

That's true. Nonetheless the A1 u-boot needs an update.

It looks like there are some users that would like to work on u-boot. So would it be possible for Hyperion to make it's U-boot repository publicly available (as I proposed above)?

@Hypex

Quote:

BTW what point is there for a RADIish, dual core like, dual BIOS in a PC?

Probably to prevent a broken system, if a BIOS upgrade fails or to switch back to an older BIOS version, if the new one causes some incompatibilities.

Last edited by Geri on 30-Jan-2010 at 05:03 PM.

_________________
A1SE: G3@600MHz, 2GB, 1GBit network card
A1XE: G4@933MHz, 2GB, refitted AC'97 codec
microA1: G3@800MHz, 1GB

- A1 Linux support -

@Rogue

I didn't say in the ROM image; I said a compiled part of U-Boot. I should have said linked with U-Boot and not granted an exception or separate license by all U-Boot copyright holders/contributers. "boota" and friends aren't necessarily GPL according to the U-Boot license, and that's fine.

In any case, where's the latest source? Messages to u-boot@hyperion-entertainment.biz bounce. I guess we need to ask Thomas directly, although he's not the one personally distributing the binaries.

_________________
Sam440ep-flex 733 MHz/1 GB RAM/Radeon 9250/AmigaOS4.1 Update 2
borked A1200/Blizzard1260+SCSI-IV/Z4+MediatorZIV/Deneb/Voodoo3/CatweaselMk3
more borked A1200/MBX1200z/Indivision
A500/clockport/RRNet
A600/A603

@Trev

Being unnecessary sarcastic here, but I did tell you good luck with that, right?

_________________
Common sense - So rare it's almost like a super power

@zerohero

We're not pulling teeth yet.

_________________
Sam440ep-flex 733 MHz/1 GB RAM/Radeon 9250/AmigaOS4.1 Update 2
borked A1200/Blizzard1260+SCSI-IV/Z4+MediatorZIV/Deneb/Voodoo3/CatweaselMk3
more borked A1200/MBX1200z/Indivision
A500/clockport/RRNet
A600/A603