mirror of
				https://github.com/ggml-org/llama.cpp.git
				synced 2025-11-04 09:32:00 +00:00 
			
		
		
		
	Revert "ggml-cpu: fix duplicate func names during compile"
This reverts commit fbb733451f.
Signed-off-by: Aaron Teo <aaron.teo1@ibm.com>
			
			
This commit is contained in:
		@@ -265,18 +265,18 @@
 | 
			
		||||
#undef GGML_COMPUTE_FP32_TO_FP16
 | 
			
		||||
#endif
 | 
			
		||||
 | 
			
		||||
#define GGML_COMPUTE_FP16_TO_FP32(x) neon_compute_fp16_to_fp32(x)
 | 
			
		||||
#define GGML_COMPUTE_FP32_TO_FP16(x) neon_compute_fp32_to_fp16(x)
 | 
			
		||||
#define GGML_COMPUTE_FP16_TO_FP32(x) ggml_compute_fp16_to_fp32(x)
 | 
			
		||||
#define GGML_COMPUTE_FP32_TO_FP16(x) ggml_compute_fp32_to_fp16(x)
 | 
			
		||||
 | 
			
		||||
#define GGML_FP16_TO_FP32(x) neon_compute_fp16_to_fp32(x)
 | 
			
		||||
#define GGML_FP16_TO_FP32(x) ggml_compute_fp16_to_fp32(x)
 | 
			
		||||
 | 
			
		||||
static inline float neon_compute_fp16_to_fp32(ggml_fp16_t h) {
 | 
			
		||||
static inline float ggml_compute_fp16_to_fp32(ggml_fp16_t h) {
 | 
			
		||||
    __fp16 tmp;
 | 
			
		||||
    memcpy(&tmp, &h, sizeof(ggml_fp16_t));
 | 
			
		||||
    return (float)tmp;
 | 
			
		||||
}
 | 
			
		||||
 | 
			
		||||
static inline ggml_fp16_t neon_compute_fp32_to_fp16(float f) {
 | 
			
		||||
static inline ggml_fp16_t ggml_compute_fp32_to_fp16(float f) {
 | 
			
		||||
    ggml_fp16_t res;
 | 
			
		||||
    __fp16 tmp = f;
 | 
			
		||||
    memcpy(&res, &tmp, sizeof(ggml_fp16_t));
 | 
			
		||||
@@ -589,13 +589,13 @@ static inline unsigned char ggml_endian_byte(int i) {
 | 
			
		||||
#undef GGML_COMPUTE_FP32_TO_FP16
 | 
			
		||||
#endif
 | 
			
		||||
 | 
			
		||||
#define GGML_COMPUTE_FP16_TO_FP32(x) power_compute_fp16_to_fp32(x)
 | 
			
		||||
#define GGML_COMPUTE_FP32_TO_FP16(x) power_compute_fp32_to_fp16(x)
 | 
			
		||||
#define GGML_COMPUTE_FP16_TO_FP32(x) ggml_compute_fp16_to_fp32(x)
 | 
			
		||||
#define GGML_COMPUTE_FP32_TO_FP16(x) ggml_compute_fp32_to_fp16(x)
 | 
			
		||||
/* the inline asm below is about 12% faster than the lookup method */
 | 
			
		||||
#define GGML_FP16_TO_FP32(x) GGML_COMPUTE_FP16_TO_FP32(x)
 | 
			
		||||
#define GGML_FP32_TO_FP16(x) GGML_COMPUTE_FP32_TO_FP16(x)
 | 
			
		||||
 | 
			
		||||
static inline float power_compute_fp16_to_fp32(ggml_fp16_t h) {
 | 
			
		||||
static inline float ggml_compute_fp16_to_fp32(ggml_fp16_t h) {
 | 
			
		||||
    float f;
 | 
			
		||||
    double d;
 | 
			
		||||
    __asm__(
 | 
			
		||||
@@ -608,7 +608,7 @@ static inline float power_compute_fp16_to_fp32(ggml_fp16_t h) {
 | 
			
		||||
    return f;
 | 
			
		||||
}
 | 
			
		||||
 | 
			
		||||
static inline ggml_fp16_t power_compute_fp32_to_fp16(float f) {
 | 
			
		||||
static inline ggml_fp16_t ggml_compute_fp32_to_fp16(float f) {
 | 
			
		||||
    double d;
 | 
			
		||||
    ggml_fp16_t r;
 | 
			
		||||
    __asm__( /* xscvdphp can work on double or single precision */
 | 
			
		||||
@@ -1155,7 +1155,7 @@ static inline ggml_fp16_t nnpa_compute_fp32_to_fp16(float f) {
 | 
			
		||||
 | 
			
		||||
#elif defined(__riscv) && defined(__riscv_zfhmin)
 | 
			
		||||
 | 
			
		||||
static inline float riscv_compute_fp16_to_fp32(ggml_fp16_t h) {
 | 
			
		||||
static inline float ggml_compute_fp16_to_fp32(ggml_fp16_t h) {
 | 
			
		||||
    float f;
 | 
			
		||||
    __asm__(
 | 
			
		||||
        "fmv.h.x %[f], %[h]\n\t"
 | 
			
		||||
@@ -1166,7 +1166,7 @@ static inline float riscv_compute_fp16_to_fp32(ggml_fp16_t h) {
 | 
			
		||||
    return f;
 | 
			
		||||
}
 | 
			
		||||
 | 
			
		||||
static inline ggml_fp16_t riscv_compute_fp32_to_fp16(float f) {
 | 
			
		||||
static inline ggml_fp16_t ggml_compute_fp32_to_fp16(float f) {
 | 
			
		||||
    ggml_fp16_t res;
 | 
			
		||||
    __asm__(
 | 
			
		||||
        "fcvt.h.s %[f], %[f]\n\t"
 | 
			
		||||
@@ -1177,8 +1177,8 @@ static inline ggml_fp16_t riscv_compute_fp32_to_fp16(float f) {
 | 
			
		||||
    return res;
 | 
			
		||||
}
 | 
			
		||||
 | 
			
		||||
#define GGML_COMPUTE_FP16_TO_FP32(x) riscv_compute_fp16_to_fp32(x)
 | 
			
		||||
#define GGML_COMPUTE_FP32_TO_FP16(x) riscv_compute_fp32_to_fp16(x)
 | 
			
		||||
#define GGML_COMPUTE_FP16_TO_FP32(x) ggml_compute_fp16_to_fp32(x)
 | 
			
		||||
#define GGML_COMPUTE_FP32_TO_FP16(x) ggml_compute_fp32_to_fp16(x)
 | 
			
		||||
#define GGML_FP16_TO_FP32(x) GGML_COMPUTE_FP16_TO_FP32(x)
 | 
			
		||||
#define GGML_FP32_TO_FP16(x) GGML_COMPUTE_FP32_TO_FP16(x)
 | 
			
		||||
 | 
			
		||||
 
 | 
			
		||||
		Reference in New Issue
	
	Block a user