Closed gemarcano closed 8 years ago
@173210
Please create NAND FAT16 Xorpad to decrypt the NAND image. I doubt if there is some garbage left after the downgrading. Try open the corresponding folder and check if multiple tmd or app.
If it's broken even with emuNAND, it should be a problem of FatFs library. So, code page or something? Anyway, at least it works for me.
No, it doesn't. I confirmed the issue.
Then this issue needs to be closed
I mean, it doesn't work for me.
Here is diff of the original FatFs and FatFs in rxTools. There are no significant differences. The problem may not be in FatFs.
Freezing at this point: https://github.com/roxas75/rxTools/commit/3adda013ef9e02e21cb828b12cc89562dfb5689f#diff-02ae78568917b1d89f38461647cbde1cR116
b_read = FileRead(&tmp, &tmd_entry, 0x30, size - (cont * 0x30));
Here is the proof: eb75735456e966722d73ba38af9bbc4120395393
Confirmed 88f16f760a88f5bead1f75c56aa1085c77c4eb69 fixes the issue.
downgradeapp.c
is a minefield. Be careful.
As background, I'm using an o3DS downgraded to fw 9.2 in sysNAND, 10.3 in emuNAND.
I was able to use an older (not sure which) version of rxTools to inject FBI. When compiling 647b341 and installing it, trying to read and just check FBI on both sysNAND and emuNAND leads to rxTools to hang. I did a
git bisect
and traced the issue to commit 77c1f93773113d85421402193d338fede5f1120e, which is when the TMIO driver was introduced. I don't know enough about 3DS development yet to know what to do at this point, though.As an aside, in previous commits, although checking FBI would not lead to a hang, all rxTools would output was:
From reading the source code, it's not finding or it's failing to open the app (see rxTools/rxtools/source/features/downgradeapp.c:525, I think that's the conditional that's failing). If this is expected behavior for sysNAND, then so be it, but if not, I can open another issue to track this thing. (EDIT: Forgot to mention sysNAND has the original Health & Safety app, not FBI, so I would think it would be detected...).