Add SSE3 versions of FIPR and FTRV - the latter is about a 4.5% improvement
src/sh4/sh4.c
src/sh4/sh4.h
src/sh4/sh4x86.in
src/sh4/x86op.h
src/sh4/sh4.c
src/sh4/sh4.h
src/sh4/sh4x86.in
src/sh4/x86op.h
Move the precision/size tests to translation-time rather than execution-time,
and flush/retranslate on a mismatch. Shaves a few percent off the core runtime
src/sh4/ia32abi.h
src/sh4/ia32mac.h
src/sh4/ia64abi.h
src/sh4/sh4trans.c
src/sh4/sh4x86.in
src/sh4/x86op.h
src/sh4/xltcache.c
src/sh4/xltcache.h
and flush/retranslate on a mismatch. Shaves a few percent off the core runtime
src/sh4/ia32abi.h
src/sh4/ia32mac.h
src/sh4/ia64abi.h
src/sh4/sh4trans.c
src/sh4/sh4x86.in
src/sh4/x86op.h
src/sh4/xltcache.c
src/sh4/xltcache.h
Add --enable-optimized flag (on by default).
When optimized, add -fomit-frame-pointer -fexceptions on non-Mac x86 (Unwinding doesn't seem to work correctly on Mac)
config.h.in
configure
configure.in
src/sh4/ia32abi.h
When optimized, add -fomit-frame-pointer -fexceptions on non-Mac x86 (Unwinding doesn't seem to work correctly on Mac)
config.h.in
configure
configure.in
src/sh4/ia32abi.h
.