Move FP DIV/MUL handling to genCodeForBinary
authorMike Danes <onemihaid@hotmail.com>
Thu, 28 Dec 2017 14:03:22 +0000 (16:03 +0200)
committerMike Danes <onemihaid@hotmail.com>
Fri, 7 Sep 2018 21:21:58 +0000 (00:21 +0300)
commit28305b5d566c3898fc0aa39e48f367fd1dc01cfb
tree3098a0dce02170016747017b797ec1128d28b740
parent1d00a61b95ce280886c252360851945937b273b7
Move FP DIV/MUL handling to genCodeForBinary

Floating point (SSE) MUL/DIV instructions have the same formats as ADD/SUB but they're handled by genCodeForMul and genCodeForDivMod. At least in the case of division this results in duplicated logic being required in genCodeForDivMod.
src/jit/codegen.h
src/jit/codegenarm.cpp
src/jit/codegenarm64.cpp
src/jit/codegenarmarch.cpp
src/jit/codegenxarch.cpp
src/jit/instr.cpp
src/jit/lower.cpp
src/jit/lowerarmarch.cpp
src/jit/lowerxarch.cpp