ReactOS on ARM, working on the iPhone

[Richardo Stallmanu]

Hello everyone!

I just got freeldr working properly (sort of) on the iPhone 4!
Currently, I use Apple's iBSS
(first stage DFU bootloader), with the go command patched to always branch to
0x41000000. 0x41000000 contains the actual LLB, which is relocated to
0x42000000.
The LLB will initialize the ReactOS loader console and ARM loader
block, which is sent
to FreeLoader, just like on Omap3/Zoom2MDK and Versatile/PB.

Time to get ramdisks working. :)

http://goput.it/uc5.jpg <- There's your picture.

-- winocm

[Sven Barth]

On 22.06.2012 21:06, Richardo Stallmanu wrote:
>   Hello everyone!
>
>  I just got freeldr working properly (sort of) on the iPhone 4!
>  Currently, I use Apple's iBSS
>  (first stage DFU bootloader), with the go command patched to always branch to
>  0x41000000. 0x41000000 contains the actual LLB, which is relocated to
>  0x42000000.
>  The LLB will initialize the ReactOS loader console and ARM loader
>  block, which is sent
>  to FreeLoader, just like on Omap3/Zoom2MDK and Versatile/PB.
>
>  Time to get ramdisks working. :)
>
>  http://goput.it/uc5.jpg <- There's your picture. 
The bootloader is only the first step... getting ReactOS itself to boot 
and utilize the iPhone's hardware correctly (and I'm mostly talking 
about the CPU and the memory here) is a rather big part that remains 
(not even mentioning that ROS won't boot that far into user mode at all 
on ARM...)

Regards,
Sven

Of course. we need to also reinitialize the arm address table translation register as the
mappings created by iBoot will not work for windows nt. 

for the boot chain, I still need to get ramdisks working. this will most likely involve
hooking the iBoot command list and adding our own ramdisk command to copy the image to
some location in memory. 

current state of the arm port is that the kernel does boot, however arm3 memory init is
missing, so we just hang there. 

Sent from my iPhone

On Jun 22, 2012, at 10:55 PM, Sven Barth <pascaldragon(a)googlemail.com> wrote:

>   On 22.06.2012 21:06, Richardo Stallmanu wrote:
> >   Hello everyone!
> >  
> >  I just got freeldr working properly (sort of) on the iPhone 4!
> >  Currently, I use Apple's iBSS
> >  (first stage DFU bootloader), with the go command patched to always branch to
> >  0x41000000. 0x41000000 contains the actual LLB, which is relocated to
> >  0x42000000.
> >  The LLB will initialize the ReactOS loader console and ARM loader
> >  block, which is sent
> >  to FreeLoader, just like on Omap3/Zoom2MDK and Versatile/PB.
> >  
> >  Time to get ramdisks working. :)
> >  
> >  http://goput.it/uc5.jpg <- There's your picture. 
>  
>  The bootloader is only the first step... getting ReactOS itself to boot and utilize the
> iPhone's hardware correctly (and I'm mostly talking about the CPU and the memory
> here) is a rather big part that remains (not even mentioning that ROS won't boot that
> far into user mode at all on ARM...)
>  
>  Regards,
>  Sven
>  
>  
>  
>  _______________________________________________
>  Ros-dev mailing list
>  Ros-dev(a)reactos.org
>  http://www.reactos.org/mailman/listinfo/ros-dev

[Sven Barth]

On 23.06.2012 06:11, rms(a)velocitylimitless.org wrote:
>   Of course. we need to also reinitialize the arm
> address table
>  translation register as the mappings created by iBoot will not work for
>  windows nt.
>
>  for the boot chain, I still need to get ramdisks working. this will most
>  likely involve hooking the iBoot command list and adding our own ramdisk
>  command to copy the image to some location in memory.
>
>  current state of the arm port is that the kernel does boot, however arm3
>  memory init is missing, so we just hang there. 
I definitely wish you good look. If you get ReactOS somehow working on 
the iPhone (and here I'm already happy with "is able to start and run 
smss.exe") this might be the first (and only) reason for me to purchase 
such a device ;)

Regards,
Sven

slight status update for the iPhone LLB: timer functions are hackplemented, along with
working commands for iboot.  now the user has to send the "go reactos" command
to iboot to start the actual boot loader. the LLB will then check for several nvram
variables (ramdisk-size, ramdisk-base, ramdisk-offset and loader-base) and will set up
freeloader as required. the boot menu now displays and the system reboots when
reinitializing paging. :)

-- winocm 

Sent from my iPhone

On Jun 23, 2012, at 2:39 PM, Sven Barth <pascaldragon(a)googlemail.com> wrote:

>   On 23.06.2012 06:11, rms(a)velocitylimitless.org wrote:
> >   Of course. we need to also reinitialize the arm
> > address table
> >  translation register as the mappings created by iBoot will not work for
> >  windows nt.
> >  
> >  for the boot chain, I still need to get ramdisks working. this will most
> >  likely involve hooking the iBoot command list and adding our own ramdisk
> >  command to copy the image to some location in memory.
> >  
> >  current state of the arm port is that the kernel does boot, however arm3
> >  memory init is missing, so we just hang there. 
>  
>  I definitely wish you good look. If you get ReactOS somehow working on the iPhone (and
> here I'm already happy with "is able to start and run smss.exe") this might
> be the first (and only) reason for me to purchase such a device ;)
>  
>  Regards,
>  Sven
>  
>  _______________________________________________
>  Ros-dev mailing list
>  Ros-dev(a)reactos.org
>  http://www.reactos.org/mailman/listinfo/ros-dev

[Riccardo Bestetti]

Hey,This interests me very much, can you provide more pictures/informations?
Thank you very much and sorry for my bad english :)

From: rms(a)velocitylimitless.org
Date: Sat, 23 Jun 2012 23:44:51 +0200
To: ros-dev(a)reactos.org
Subject: Re: [ros-dev] ReactOS on ARM, working on the iPhone

slight status update for the iPhone LLB: timer functions are hackplemented, along with
working commands for iboot.  now the user has to send the "go reactos" command
to iboot to start the actual boot loader. the LLB will then check for several nvram
variables (ramdisk-size, ramdisk-base, ramdisk-offset and loader-base) and will set up
freeloader as required. the boot menu now displays and the system reboots when
reinitializing paging. :)
-- winocm 

Sent from my iPhone
On Jun 23, 2012, at 2:39 PM, Sven Barth <pascaldragon(a)googlemail.com> wrote:

On 23.06.2012 06:11, rms(a)velocitylimitless.org wrote:
Of course. we need to also reinitialize the arm address table
translation register as the mappings created by iBoot will not work for
windows nt.

for the boot chain, I still need to get ramdisks working. this will most
likely involve hooking the iBoot command list and adding our own ramdisk
command to copy the image to some location in memory.

current state of the arm port is that the kernel does boot, however arm3
memory init is missing, so we just hang there.

I definitely wish you good look. If you get ReactOS somehow working on the iPhone (and
here I'm already happy with "is able to start and run smss.exe") this might
be the first (and only) reason for me to purchase such a device ;)

Regards,
Sven

_______________________________________________
Ros-dev mailing list
Ros-dev(a)reactos.org
http://www.reactos.org/mailman/listinfo/ros-dev

_______________________________________________
Ros-dev mailing list
Ros-dev(a)reactos.org
http://www.reactos.org/mailman/listinfo/ros-dev

[chris]

Richardo Stallmanu? the o and the u are fake? they have to be.lol
ps... nice work by the way/



----- Original Message ----- 
From: "Richardo Stallmanu" <rms(a)velocitylimitless.org>
To: <ros-dev(a)reactos.org>
Sent: Saturday, June 23, 2012 5:06 AM
Subject: [ros-dev] ReactOS on ARM, working on the iPhone


>   Hello everyone!
>
>  I just got freeldr working properly (sort of) on the iPhone 4!
>  Currently, I use Apple's iBSS
>  (first stage DFU bootloader), with the go command patched to always branch 
>  to
>  0x41000000. 0x41000000 contains the actual LLB, which is relocated to
>  0x42000000.
>  The LLB will initialize the ReactOS loader console and ARM loader
>  block, which is sent
>  to FreeLoader, just like on Omap3/Zoom2MDK and Versatile/PB.
>
>  Time to get ramdisks working. :)
>
>  http://goput.it/uc5.jpg <- There's your picture.
>
>  -- winocm
>
>  _______________________________________________
>  Ros-dev mailing list
>  Ros-dev(a)reactos.org
>  http://www.reactos.org/mailman/listinfo/ros-dev

[Conan Kudo (ニール・ゴンパ)]

This sounds all very cool. Even if not much has actually been accomplished
yet, it is a great step forward!

On Thu, Jul 5, 2012 at 1:44 AM, chris <christopher57(a)optusnet.com.au> wrote:

>   Richardo Stallmanu? the o and the u are fake? they
> have to be.lol
>  ps... nice work by the way/
>
>
>
>  ----- Original Message ----- From: "Richardo Stallmanu" <
>  rms(a)velocitylimitless.org&gt;
>  To: &lt;ros-dev(a)reactos.org&gt;
>  Sent: Saturday, June 23, 2012 5:06 AM
>  Subject: [ros-dev] ReactOS on ARM, working on the iPhone
>
>
>
>   Hello everyone!
> >  
> >  I just got freeldr working properly (sort of) on the iPhone 4!
> >  Currently, I use Apple's iBSS
> >  (first stage DFU bootloader), with the go command patched to always
> >  branch to
> >  0x41000000. 0x41000000 contains the actual LLB, which is relocated to
> >  0x42000000.
> >  The LLB will initialize the ReactOS loader console and ARM loader
> >  block, which is sent
> >  to FreeLoader, just like on Omap3/Zoom2MDK and Versatile/PB.
> >
> >  Time to get ramdisks working. :)
> >
> >  http://goput.it/uc5.jpg <- There's your picture.
> >
> >  -- winocm
> >
> >  ______________________________**_________________
> >  Ros-dev mailing list
> >  Ros-dev(a)reactos.org
> >
> > http://www.reactos.org/**mailman/listinfo/ros-dev<http://www.reactos.org…
> >  
>
>  ______________________________**_________________
>  Ros-dev mailing list
>  Ros-dev(a)reactos.org
>
> http://www.reactos.org/**mailman/listinfo/ros-dev<http://www.reactos.org…
>