TrevorBradley
Active Member
- Joined
- Nov 6, 2007
- Messages
- 732
Hi there... things going well on my C++ SDL project. I'm still targeting release for the Pandora, but it's possible I might be able to release a version for the GP2X.
I've managed to compile my game statically using the open2x toolchain, but there are still consistent crash bugs I believe might be related to memory. My program at present needs about 12MB of memory.
I notice I get a crash at runtime if I compile on Windows using Cygwin. I've managed to make these problems going away by passing the command line option "-Wl,--stack,10000000" to g++. I could also make them go away by reducing my memory usage to about 8 MB. I don't need any such parameters to compile and run on x86 linux.
A bit of research indicates these parameters are passed to the linker ld, and that the parameters needed are OS specific. I was hoping I could pass these parameters, or something similar to the open2x toolchain, but looking at open2x' ld --help command doesn't reveal anything obvious.
If I want to compile a C++ program using the GP2X toolchain that wont crash with a segmentation fault when allocating a larger amounts of memory, what command line parameter should I pass to it?
It's possible my problems have to do with something else .other than memory.. If so it's going to be painful to debug on an F200..
Thanks in advance for your replies! The GP32x community has been a great help in getting this project off the ground.
I've managed to compile my game statically using the open2x toolchain, but there are still consistent crash bugs I believe might be related to memory. My program at present needs about 12MB of memory.
I notice I get a crash at runtime if I compile on Windows using Cygwin. I've managed to make these problems going away by passing the command line option "-Wl,--stack,10000000" to g++. I could also make them go away by reducing my memory usage to about 8 MB. I don't need any such parameters to compile and run on x86 linux.
A bit of research indicates these parameters are passed to the linker ld, and that the parameters needed are OS specific. I was hoping I could pass these parameters, or something similar to the open2x toolchain, but looking at open2x' ld --help command doesn't reveal anything obvious.
If I want to compile a C++ program using the GP2X toolchain that wont crash with a segmentation fault when allocating a larger amounts of memory, what command line parameter should I pass to it?
It's possible my problems have to do with something else .other than memory.. If so it's going to be painful to debug on an F200..
Thanks in advance for your replies! The GP32x community has been a great help in getting this project off the ground.