Hacking Luma3DS - Noob-proof 3DS Custom Firmware

[Supster131]

all that needs to be done is get homebrew apps to work on 10.5 sysnand and fix the 9.0 firm launching when trying to launch emunand 9.2 but get black screen instead.

Well, thanks to a9lh, we can enter sysNAND with sigpatches on 10.5 now, so I would definitely expect some things to be worked on soon.

 

[Svaethier]

Well, thanks to a9lh, we can enter sysNAND with sigpatches on 10.5 now, so I would definitely expect some things to be worked on soon.

yea,would the L button not launching the 9.0 firm file with emunand 9.2 be caused by a semi downgrade if someone had one? don't think my downgrade is semi as more than 47 cia files were installed before firm file got installed.

 

[Supster131]

yea,would the L button not launching the 9.0 firm file with emunand 9.2 be caused by a semi downgrade if someone had one? don't think my downgrade is semi as more than 47 cia files were installed before firm file got installed.

I wouldn't know, and I wouldn't try that out right now

The thing is, at the moment, there's no extra benefit to running your sysNAND on 10.5. It'll probably shave off a few seconds from your boot time, but other than that, it's the same thing.

 

[JacksonS]

yea,would the L button not launching the 9.0 firm file with emunand 9.2 be caused by a semi downgrade if someone had one? don't think my downgrade is semi as more than 47 cia files were installed before firm file got installed.

No, it's due to a fault in the arm9loaderhax.bin payload.

Edit: If you can't load emuNAND even with the 9.6-10.4 FIRM, then you may have a semi-downgrade. But there is definitely something wrong with the officially released payload.

 

[Svaethier]

No, it's due to a fault in the arm9loaderhax.bin payload.

Edit: If you can't load emuNAND even with the 9.6-10.4 FIRM, then you may have a semi-downgrade. But there is definitely something wrong with the officially released payload.

yea you got a point, loving my sysnand 10.5. could't even use homebrew apps with it on 9.2 anyways for some odd reason as decrypt9 didn't work xD atleast I already used the homebrew fbi app using the 3.4 guide to install fbi.cia onto my sysnand.

 

[Just Passing By]

Thanks for this. I got it running now!

 

[Svaethier]

well a decrypt9 cia has been made in the decrypt9 thread. but it needs the 9.0 firm patch that the 3.4 build has. is it safe to use sysupdater to downgrade back to sysnand 9.2 or does it not work with 10.5 sysnand running a9lh?

 

[Trill Spector]

i recently switched to aurienand set to autoboot to emunand with arm9loaderhax and for some reason any game that requires resets will always give me an error, is there a known workaround for this? i made sure to copy the correct /rei/reboot

 

[Supster131]

well a decrypt9 cia has been made in the decrypt9 thread. but it needs the 9.0 firm patch that the 3.4 build has. is it safe to use sysupdater to downgrade back to sysnand 9.2 or does it not work with 10.5 sysnand running a9lh?

I think you should be fine, @JacksonS downgraded to 9.2 earlier. Pretty sure he used the sysUpdater cia.

 

[Syphurith]

Thanks for your contribution aiming the A9LH. However i could not get it built from git repo.

# FatFs requires libgcc for __aeabi_uidiv
arm-none-eabi-gcc -nostartfiles  -T linker.ld -o build/main.elf build/fs.o build/patches.o build/start.o build/memory.o build/main.o build/emunand.o build/draw.o build/crypto.o build/firm.o build/fatf
s/diskio.o build/fatfs/ff.o build/fatfs/option/syscall.o build/fatfs/sdmmc/delay.o build/fatfs/sdmmc/sdmmc.o
d:/station/devkitpro/devkitarm/bin/../lib/gcc/arm-none-eabi/5.3.0/../../../../arm-none-eabi/bin/ld.exe: warning: d:/station/devkitpro/devkitarm/bin/../lib/gcc/arm-none-eabi/5.3.0/../../../../arm-none-
eabi/lib\libc.a(lib_a-atoi.o) uses 4-byte wchar_t yet the output is to use 2-byte wchar_t; use of wchar_t values across objects may fail
d:/station/devkitpro/devkitarm/bin/../lib/gcc/arm-none-eabi/5.3.0/../../../../arm-none-eabi/bin/ld.exe: warning: d:/station/devkitpro/devkitarm/bin/../lib/gcc/arm-none-eabi/5.3.0/../../../../arm-none-
eabi/lib\libc.a(lib_a-strtol.o) uses 4-byte wchar_t yet the output is to use 2-byte wchar_t; use of wchar_t values across objects may fail
d:/station/devkitpro/devkitarm/bin/../lib/gcc/arm-none-eabi/5.3.0/../../../../arm-none-eabi/bin/ld.exe: warning: d:/station/devkitpro/devkitarm/bin/../lib/gcc/arm-none-eabi/5.3.0/../../../../arm-none-
eabi/lib\libc.a(lib_a-ctype_.o) uses 4-byte wchar_t yet the output is to use 2-byte wchar_t; use of wchar_t values across objects may fail
d:/station/devkitpro/devkitarm/bin/../lib/gcc/arm-none-eabi/5.3.0/../../../../arm-none-eabi/bin/ld.exe: warning: d:/station/devkitpro/devkitarm/bin/../lib/gcc/arm-none-eabi/5.3.0/../../../../arm-none-
eabi/lib\libsysbase.a(getreent.o) uses 4-byte wchar_t yet the output is to use 2-byte wchar_t; use of wchar_t values across objects may fail
d:/station/devkitpro/devkitarm/bin/../lib/gcc/arm-none-eabi/5.3.0/../../../../arm-none-eabi/bin/ld.exe: warning: d:/station/devkitpro/devkitarm/bin/../lib/gcc/arm-none-eabi/5.3.0/../../../../arm-none-
eabi/lib\libsysbase.a(syscall_support.o) uses 4-byte wchar_t yet the output is to use 2-byte wchar_t; use of wchar_t values across objects may fail
d:/station/devkitpro/devkitarm/bin/../lib/gcc/arm-none-eabi/5.3.0/../../../../arm-none-eabi/bin/ld.exe: warning: d:/station/devkitpro/devkitarm/bin/../lib/gcc/arm-none-eabi/5.3.0/../../../../arm-none-
eabi/lib\libc.a(lib_a-impure.o) uses 4-byte wchar_t yet the output is to use 2-byte wchar_t; use of wchar_t values across objects may fail
arm-none-eabi-objcopy -S -O binary build/main.elf build/main.bin
make: *** No rule to make target 'data/firmware.bin', needed by 'out/rei-n3ds/'.  Stop.

I've already have the FIRM.7z for this. (Win8.1x64, MSYS2+DevkitPror45, Python2, armips.. I can build cakeFW and other easily, but not this one. orz)
Just to note: Checking the Makefile in the repo, there is really no actions building the target 'data/firmware.bin'. Could you check it?

 

[JacksonS]

I think you should be fine, @JacksonS downgraded to 9.2 earlier. Pretty sure he used the sysUpdater cia.

I did not, actually. I used the word downgrade but I was really restoring a 9.2 backup via hardmod. I used the wrong terminology, and that was my mistake.

 

[Svaethier]

I did not, actually. I used the word downgrade but I was really restoring a 9.2 backup via hardmod. I used the wrong terminology, and that was my mistake.

well I'm back to 9.2 sysnand after installing the sysupdater.cia so you can infact downgrade sysnand with a9lh

edit: I suggest going back to 3.4 build if you have 9.2 sysnand still and install this decrypt cia: http://www.battlefieldsingleplayer.com/apachethunder/3ds/Decrypt9WIPCIA_NAND.rar
and install it to your NAND using fbi installer as it only works if installed to NAND

credit goes to @Apache Thunder for making this cia build that works.

 

[subcon959]

Thanks for your contribution aiming the A9LH. However i could not get it built from git repo.

# FatFs requires libgcc for __aeabi_uidiv
arm-none-eabi-gcc -nostartfiles  -T linker.ld -o build/main.elf build/fs.o build/patches.o build/start.o build/memory.o build/main.o build/emunand.o build/draw.o build/crypto.o build/firm.o build/fatf
s/diskio.o build/fatfs/ff.o build/fatfs/option/syscall.o build/fatfs/sdmmc/delay.o build/fatfs/sdmmc/sdmmc.o
d:/station/devkitpro/devkitarm/bin/../lib/gcc/arm-none-eabi/5.3.0/../../../../arm-none-eabi/bin/ld.exe: warning: d:/station/devkitpro/devkitarm/bin/../lib/gcc/arm-none-eabi/5.3.0/../../../../arm-none-
eabi/lib\libc.a(lib_a-atoi.o) uses 4-byte wchar_t yet the output is to use 2-byte wchar_t; use of wchar_t values across objects may fail
d:/station/devkitpro/devkitarm/bin/../lib/gcc/arm-none-eabi/5.3.0/../../../../arm-none-eabi/bin/ld.exe: warning: d:/station/devkitpro/devkitarm/bin/../lib/gcc/arm-none-eabi/5.3.0/../../../../arm-none-
eabi/lib\libc.a(lib_a-strtol.o) uses 4-byte wchar_t yet the output is to use 2-byte wchar_t; use of wchar_t values across objects may fail
d:/station/devkitpro/devkitarm/bin/../lib/gcc/arm-none-eabi/5.3.0/../../../../arm-none-eabi/bin/ld.exe: warning: d:/station/devkitpro/devkitarm/bin/../lib/gcc/arm-none-eabi/5.3.0/../../../../arm-none-
eabi/lib\libc.a(lib_a-ctype_.o) uses 4-byte wchar_t yet the output is to use 2-byte wchar_t; use of wchar_t values across objects may fail
d:/station/devkitpro/devkitarm/bin/../lib/gcc/arm-none-eabi/5.3.0/../../../../arm-none-eabi/bin/ld.exe: warning: d:/station/devkitpro/devkitarm/bin/../lib/gcc/arm-none-eabi/5.3.0/../../../../arm-none-
eabi/lib\libsysbase.a(getreent.o) uses 4-byte wchar_t yet the output is to use 2-byte wchar_t; use of wchar_t values across objects may fail
d:/station/devkitpro/devkitarm/bin/../lib/gcc/arm-none-eabi/5.3.0/../../../../arm-none-eabi/bin/ld.exe: warning: d:/station/devkitpro/devkitarm/bin/../lib/gcc/arm-none-eabi/5.3.0/../../../../arm-none-
eabi/lib\libsysbase.a(syscall_support.o) uses 4-byte wchar_t yet the output is to use 2-byte wchar_t; use of wchar_t values across objects may fail
d:/station/devkitpro/devkitarm/bin/../lib/gcc/arm-none-eabi/5.3.0/../../../../arm-none-eabi/bin/ld.exe: warning: d:/station/devkitpro/devkitarm/bin/../lib/gcc/arm-none-eabi/5.3.0/../../../../arm-none-
eabi/lib\libc.a(lib_a-impure.o) uses 4-byte wchar_t yet the output is to use 2-byte wchar_t; use of wchar_t values across objects may fail
arm-none-eabi-objcopy -S -O binary build/main.elf build/main.bin
make: *** No rule to make target 'data/firmware.bin', needed by 'out/rei-n3ds/'.  Stop.

I've already have the FIRM.7z for this. (Win8.1x64, MSYS2+DevkitPror45, Python2, armips.. I can build cakeFW and other easily, but not this one. orz)
Just to note: Checking the Makefile in the repo, there is really no actions building the target 'data/firmware.bin'. Could you check it?

The repo doesn't include the data folder you have to create it yourself and populate with correctly named firmware files.

 

[Supster131]

The repo doesn't include the data folder you have to create it yourself and populate with correctly named firmware files.

For some reason it told me it couldn't create the reboot patch files. So I couldn't compile the whole thing.

 

[JacksonS]

I suggest going back to 3.4 build if you have 9.2 sysnand still and install this decrypt cia: http://www.battlefieldsingleplayer.com/apachethunder/3ds/Decrypt9WIPCIA_NAND.rar
and install it to your NAND using fbi installer as it only works if installed to NAND

This will not boot if you're on 10.5 with 10.5 FIRM, right?

 

[Svaethier]

This will not boot if you're on 10.5 with 10.5 FIRM, right?

the decrypt9 cia only works with the 9.0 firm patch with sysnand that 3.4 build only does by holding L+R at boot.

 

[JacksonS]

the decrypt9 cia only works with the 9.0 firm patch with sysnand that 3.4 build only does by holding L+R at boot.

I'm using sysNAND as my primary NAND. That's why I wanted 3.5, so it would boot right into sysNAND. It doesn't provide a way to easily boot into sysNAND with 9.0 FIRM.

 

[Svaethier]

I'm using sysNAND as my primary NAND. That's why I wanted 3.5, so it would boot right into sysNAND. It doesn't provide a way to easily boot into sysNAND with 9.0 FIRM.

well maybe 3.6 will fix the L button to launch 9.2 emunand and the 9.0 firm issues it has, those problems are what made me go back to 3.4 as it's more stable than 3.5.

 

[Ninoh-FOX]

I try compiler AuReiNand for autoboot in sysnand10.5 (I not have emunand), but amr9loaderhax.bin not work, I use ubuntu 12.04 and devkitpro (I cant install this in windows... make commander said me that gcc no found)

 

[NeoSlyde]

I cant get ssb to work on my o3ds XL with reinand mod
I have A9lh..