GCC Compiler Support #1791
GCC Compiler Support #1791
Conversation
hensldm
added
Tooling
Needs-second-approval
There was a problem hiding this comment.
Thoughts on bringing mkldscript.c over from OoT (though does require adding the else if for STMT_increment on lines 40-41) to have parity between the two (mainly looks like just formatting is the difference between the 2). If you don't want to that is fine, I would probably do it after this gets merged.
| ENDIAN := -EB | ||
| ifeq ($(COMPILER),gcc) | ||
| # MIPS options | ||
| CFLAGS += -G 0 -march=vr4300 -mtune=vr4300 -mfix4300 -mabi=32 -mno-abicalls -mdivide-breaks |
There was a problem hiding this comment.
This reminds me, why do we use -mdivide-breaks? The n64 supports trap instructions
There was a problem hiding this comment.
It's mostly for compatibility. I think it's a reasonable default that people can change themselves depending on their tolerance for what platforms they want to support
| COMMON_DEFINES := $(GBI_DEFINES) | ||
| AS_DEFINES := $(COMMON_DEFINES) -DMIPSEB -D_LANGUAGE_ASSEMBLY -D_ULTRA64 | ||
| C_DEFINES := $(COMMON_DEFINES) -D_LANGUAGE_C | ||
| ENDIAN := |
There was a problem hiding this comment.
Why omit this?
| ENDIAN := | |
| ENDIAN := -EB |
There was a problem hiding this comment.
It's an architecture default, it doesn't really need to be explicit. We already rely on -mhard-float being implicit from the arch
AngheloAlf
added
Merge-ready
This PR adds support for building the project with a modern MIPS GCC compiler.
Thanks to @fig02 for testing a build on hardware, and everyone who engaged in related discussions.