Add Imath support for fp16-fp32 conversions

Makefile

@@ -130,6 +130,10 @@ ifdef WHISPER_GPROF
 	CFLAGS  += -pg
 	CXXFLAGS  += -pg
 endif
+ifdef WHISPER_IMATH
+	CFLAGS += -DGGML_USE_IMATH
+	LDFLAGS += -lImath
+endif
 ifneq ($(filter aarch64%,$(UNAME_M)),)
 endif
 ifneq ($(filter armv6%,$(UNAME_M)),)

ggml.c

@@ -158,6 +158,13 @@ ggml_fp16_t ggml_fp32_to_fp16(float f) {
 #define GGML_FP16_TO_FP32(x) _cvtsh_ss(x)
 #define GGML_FP32_TO_FP16(x) _cvtss_sh(x, 0)
 
+#elif GGML_USE_IMATH
+
+#include <Imath/half.h>
+
+#define GGML_FP16_TO_FP32(x) imath_half_to_float(x)
+#define GGML_FP32_TO_FP16(x) imath_float_to_half(x)
+
 #else
 
 // FP16 <-> FP32