Hello everyone!
Does anyone know how to create shared ELF libraries in DJGPP that actually works?
I remember before that i downloaded the elf-binutils for DJGPP, but they didn't work for shared libraries. First there was som error saying __GLOBAL_OFFSET_TABLE couldn't be founded. I founded the solution to this on some page so at least it seemed to work but when I checked in the .so files I created they where all fucked up (at least the relocation entries).
What to do? I would be very great grateful if someone could help me here...
Working shared ELF libraries in DJGPP
Re:Working shared ELF libraries in DJGPP
It would be easier if you just use Cygwinand a proper cross compiler.
Re:Working shared ELF libraries in DJGPP
Yes, but I want to use dos to test my os on real hardware (dos is very fast to boot as you might know...)It would be easier if you just use Cygwin and a proper cross compiler.
Re:Working shared ELF libraries in DJGPP
You can try to make a 32-bit crosscompiler with DJGPP, although I'm not at all sure how/whether that will work. You'll need unmodified sources to GCC, and then compile that with DJGPP. If I'm thinking straight you can then use the resulting binary as a normal compiler under DOS, except that it can then no longer create DOS binaries but only 32-bit files.Sam the man wrote:Yes, but I want to use dos to test my os on real hardware (dos is very fast to boot as you might know...)It would be easier if you just use Cygwin and a proper cross compiler.
- Pype.Clicker
- Member
- Posts: 5964
- Joined: Wed Oct 18, 2006 2:31 am
- Location: In a galaxy, far, far away
- Contact:
Re:Working shared ELF libraries in DJGPP
well, the only problem i could foresee with using DJGPP to build a cross-compiler will be the time it'll require. It won't be fast under DOS (probably take the night long) and it will even get worse under windows (due to the double emulation required to run 32 bits program in an emulated environment that's there for 16 bits compatibility).
Seriously, you should consider using cygwin instead, as we all did the day we discovered cygwin. All the code you gather from DJGPP tutorials will run almost unmodified under cygwin and you'll be set up for the future.
Seriously, you should consider using cygwin instead, as we all did the day we discovered cygwin. All the code you gather from DJGPP tutorials will run almost unmodified under cygwin and you'll be set up for the future.