Hacking (vWii) - Tri-Core Linux

  • Thread starter Thread starter Ninja_Carver
  • Start date Start date
  • Views Views 197,348
  • Replies Replies 841
  • Likes Likes 30
Ok, I think for the interim we can hold off on further testing. I have a pretty good idea as to what's going on.

Seems the 3.0 kernel by Delta kills some functionality. The 2.6.32 from isobel manages to load the most hardware 'out of the box'.

Next step is figuring out how to successfully recompile the kernel. My previous attempts, while the kernel successfully built, failed to actually boot. In the interim, I'm also working on some filesystem/package enhancements and working on enabling a swap partition on the SD card.
 
Ok, I think for the interim we can hold off on further testing. I have a pretty good idea as to what's going on.

Seems the 3.0 kernel by Delta kills some functionality. The 2.6.32 from isobel manages to load the most hardware 'out of the box'.

Next step is figuring out how to successfully recompile the kernel. My previous attempts, while the kernel successfully built, failed to actually boot. In the interim, I'm also working on some filesystem/package enhancements and working on enabling a swap partition on the SD card.

How are you going about building your kernel, if I might ask? I'm curious if the process is similar to what I have done in the past with libxenon.

I'm almost positive you know how to partition a SD card in Linux so I assume by enable you are referring to initializing the swap partition and dedicating it as a mount. If that is the case I would strongly recommend against implementing this in any practical environment in the instance you are indeed referring to a legacy SD card. These cards tend to reach critical throughput just over 10MB/s. As I'm also positive you know that swap is for physical RAM / virtual memory, it goes without saying how this could lead to complications in applicable situations where needed. I recommend initializing an USB (flash / disk) drive partition as a ./linux-swap mount.
 
  • Like
Reactions: Deletedmember331810
How are you going about building your kernel, if I might ask? I'm curious if the process is similar to what I have done in the past with libxenon.

I'm almost positive you know how to partition a SD card in Linux so I assume by enable you are referring to initializing the swap partition and dedicating it as a mount. If that is the case I would strongly recommend against implementing this in any practical environment in the instance you are indeed referring to a legacy SD card. These cards tend to reach critical throughput just over 10MB/s. As I'm also positive you know that swap is for physical RAM / virtual memory, it goes without saying how this could lead to complications in applicable situations where needed. I recommend initializing an USB (flash / disk) drive partition as a ./linux-swap mount.


SL- I understand the concerns about the SD card. One could even go a step further and say that all SD cards WILL eventually fail, with or without a swap partition :)

at this point, we can pretty much guarantee that everyone at least has an SD card. Once we have greater demands for the OS, we will explore routes of using a USB flash drive, or even USB HDD.

EDIT: FYI this is a change that could be easily made now, but just no need to at this point.
 
  • Like
Reactions: Deletedmember331810
Ok everyone, another test. This test will utilize Maxternal's code with a Linux kernel I compiled.

For this test you will need to extract your nand, and get the 00000003.app (1-512), or find it some other way. I won't cover instructions on how to dump your nand here, as there are dozens of resources on that. Future releases of max's code will load 1-512 from nand, so you wont need this file.

Here is a 4GB SD card image.

http://www.idfbest.com/wii/wiilinux-carver-0.3.7z - compressed down to 400MB
write it to your SD card in the same manner. The .app file goes in the sd:/bootmii/ folder.

Run it the same way.

If you can record a video and upload to youtube and post it here, that would be fantastic. This test will validate whether or not we have a working kernel source code base to begin working off of.

The Linux filesystem was also updated in this release. apt-get now works, and you should be able to plug in most USB ethernet to establish Internet access. I picked up a used USB ethernet from gamestop for $12, works great. Future releases will hopefully enable wifi.

After you log in to the system, please issue a 'uname -a' and let me know what it returns.

root/whiite
 
Ok everyone, another test. This test will utilize Maxternal's code with a Linux kernel I compiled.

For this test you will need to extract your nand, and get the 00000003.app (1-512), or find it some other way. I won't cover instructions on how to dump your nand here, as there are dozens of resources on that. Future releases of max's code will load 1-512 from nand, so you wont need this file.

Here is a 4GB SD card image.

http://www.idfbest.com/wii/wiilinux-carver-0.3.7z - compressed down to 400MB
write it to your SD card in the same manner. The .app file goes in the sd:/bootmii/ folder.

Run it the same way.

If you can record a video and upload to youtube and post it here, that would be fantastic. This test will validate whether or not we have a working kernel source code base to begin working off of.

The Linux filesystem was also updated in this release. apt-get now works, and you should be able to plug in most USB ethernet to establish Internet access. I picked up a used USB ethernet from gamestop for $12, works great. Future releases will hopefully enable wifi.

After you log in to the system, please issue a 'uname -a' and let me know what it returns.

root/whiite

I placed the 00000003.app file in the bootmii folder, ran tri-linux, and got this:


err.PNG
 
make sure you have:

sd:/bootmii/ppcboot.elf
sd:/bootmii/00000003.app
sd:/apps/wiilinux/meta.xml, boot.dol,
 
Ok, I think for the interim we can hold off on further testing. I have a pretty good idea as to what's going on.

Seems the 3.0 kernel by Delta kills some functionality. The 2.6.32 from isobel manages to load the most hardware 'out of the box'.

Next step is figuring out how to successfully recompile the kernel. My previous attempts, while the kernel successfully built, failed to actually boot. In the interim, I'm also working on some filesystem/package enhancements and working on enabling a swap partition on the SD card.

Wow, I didn't realize anyone was actually testing my kernel modifications with Wii U consoles. Does my 3.0 kernel actually boot on it like it does with GameCube and Wii consoles? I had to leave out a few things in my forward port as I didn't understand how they worked enough to do a full rewrite. I would like to help, but I don't have a Wii U (just a Wii) and my programming knowledge is limited to what I what learned back in high school. Mostly all I did to forward port it was bisect issues and merge duplicate/conflicting code from the mikep5 patch into mainline kernels incrementally until reaching 3.0.
 
  • Like
Reactions: Deletedmember331810
Wow, I didn't realize anyone was actually testing my kernel modifications with Wii U consoles. Does my 3.0 kernel actually boot on it like it does with GameCube and Wii consoles? I had to leave out a few things in my forward port as I didn't understand how they worked enough to do a full rewrite. I would like to help, but I don't have a Wii U (just a Wii) and my programming knowledge is limited to what I what learned back in high school. Mostly all I did to forward port it was bisect issues and merge duplicate/conflicting code from the mikep5 patch into mainline kernels incrementally until reaching 3.0.
well, booting ANY mini ppcboot.elf on Wii U was a little tricky, but once we got passed that part, the linux ones were the easiest because of the memory area they use (still haven't made the right tweaks to be able to run bootmii gui itself ... but it's on my to-do list)

Actually, ninja_carver has been using mikep5 a bit more just because of mikep7 was giving him some problems. Maybe with a little help considering you'd be a bit more FAMILIAR with the code than us (programming experienced or not) if you want to lend a hand, we'd love some help getting it running more smoothly

our end goal will to add support for all 3 cores (as the thread title says) so if you have any pointers on anything you've seen in the code that would help us there, too, we're willing to listen :P
 
Wow, I didn't realize anyone was actually testing my kernel modifications with Wii U consoles. Does my 3.0 kernel actually boot on it like it does with GameCube and Wii consoles? I had to leave out a few things in my forward port as I didn't understand how they worked enough to do a full rewrite. I would like to help, but I don't have a Wii U (just a Wii) and my programming knowledge is limited to what I what learned back in high school. Mostly all I did to forward port it was bisect issues and merge duplicate/conflicting code from the mikep5 patch into mainline kernels incrementally until reaching 3.0.

yes, it does boot. but im sure in the effort to minimize the size of the kernel you disabled some key components. you have a great base though. will discuss with you further if we can use it for our project. I'm the guy that has been emailing you, by the way :)
 
Shorter version; interesting to see very familiar ideas (360 guy, USB loading for Linux, the most current version of linux mentioned in Wiibrew forums, etc). Good progress;-)
 
Ok everyone, another test, and let me know what it returns.

This test will utilize Maxternal's code with a Linux kernel I compiled.

For this test you will need to extract your nand, and get the 00000003.app (1-512), or find it some other way. I won't cover instructions on how to dump your nand here, as there are dozens of resources on that. Future releases of max's code will load 1-512 from nand, so you wont need this file.

Here is a 4GB SD card image.

http://www.idfbest.com/wii/wiilinux-carver-0.3.7z - compressed down to 400MB
write it to your SD card in the same manner. The .app file goes in the sd:/bootmii/ folder.

Run it the same way.

If you can record a video and upload to youtube and post it here, that would be fantastic. This test will validate whether or not we have a working kernel source code base to begin working off of.

The Linux filesystem was also updated in this release. apt-get now works, and you should be able to plug in most USB ethernet to establish Internet access. I picked up a used USB ethernet from gamestop for $12, works great. Future releases will hopefully enable wifi.

After you log in to the system, please issue a 'uname -a'

ok with your ''4GB card image,'' got stuck on this text.
xla1.jpg

but using only the .elf from that card image, i got the same result, like the one from the ''PM''
d92n.png
 
ok with your ''4GB card image,'' got stuck on this text.
xla1.jpg

but using only the .elf from that card image, i got the same result, like the one from the ''PM''
d92n.png

is this the same SD card? or different cards. wondering if this is a 4GB+ SD card limitation...

Snailface, grab this file and rename ppcboot.elf, place in /bootmii and see if that gets you past where you got stuck...

http://vwii.idfbest.com/upload/zImage-CARVER-2.6.32.4

also- how did you get the .app file? you sure its good? mine is 254KB
 
is this the same SD card? or different cards. wondering if this is a 4GB+ SD card limitation...


i been testing the ''4GB card image'' with diferents .elf`s(vanilla, carver, NOPREEMP, SMP NOPREEMP,) and evrery, .elf stuck on the same text.

but using a 2GB card with the ''2GB card image'' every elf, is booting to the end. (login, whiite)
 
i been testing the ''4GB card image'' with diferents .elf`s(vanilla, carver, NOPREEMP, SMP NOPREEMP,) and evrery, .elf stuck on the same text.

but using a 2GB card with the ''2GB card image'' every elf, is booting to the end. (login, whiite)

wait SMP NOPREEMP worked?
 

Site & Scene News

Popular threads in this forum