The 3rd Birthday

[[Unknown]]

Well, if you can look at the disassembly near the crash and post a screenshot, that would help a lot. But sometimes you won't get a great trace. It still might give good info.

Otherwise, there are code changes you can make to test it. In Core/MIPS/ARM/ there are the following files:

ArmCompALU.cpp
ArmCompFPU.cpp
ArmCompLoadStore.cpp
ArmCompVFPU.cpp

Near the top of each of these files is two lines that look like this:

Code:

//#define CONDITIONAL_DISABLE { Comp_Generic(op); return; }
#define CONDITIONAL_DISABLE ;

If you switch them (delete the // from the first one, and put a // before the second one), it will disable that section of the jit. You'll need to compile it after doing so, you can find instructions here:

https://github.com/hrydgard/ppsspp/wiki/...-ios-cmake

You may need an iOS dev membership or an open source toolchain or something. I don't know much about that side of things, but I think it says what's needed on that wiki page at the top.

Anyway, if you switch the CONDITIONAL_DISABLEs in each file, you'll end up with something that is *basically* like jit disabled, but faster. It will probably not crash. If it doesn't, we know that the cause of the crash is in one of these sections of the jit.

If it still crashes, it means we've excluded the sections of the jit, which means it's the "base" of the jit, the mechanics. It might be another issue like the one we had on iOS with R9.

Hopefully, it won't be mechanics, and the change I described above will make it not crash (but slow.) In this case, there's more we can do. But, I'll explain that if you're up for it and after looking to see if maybe we can solve it just from the disassembly.

-[Unknown]