clean up

Author: Djip007

ggml/src/ggml-quants.c

@@ -15744,7 +15744,7 @@ bool ggml_validate_row_data(enum ggml_type type, const void * data, size_t nbyte
                     if (!validate_float(q[i].d, i)) {
                         return false;
                     }
-                    // NAN?
+                    // NAN
                     for (size_t k = 0; k < QK_K; ++k) {
                         nans += (q[i].qs[k] & 0x7f) == 0x7f;
                     }

ggml/src/ggml.c

@@ -588,7 +588,6 @@ void ggml_bf16_to_fp32_row(const ggml_bf16_t * x, float * y, int64_t n) {
     int64_t i = 0;
 #if defined(__AVX512BF16__)
     for (; i + 16 <= n; i += 16) {
-        // __m512 _mm512_cvtpbh_ps (__m256bh a)
         _mm512_storeu_ps(y + i,
                          _mm512_cvtpbh_ps((__m256bh)
                              _mm256_loadu_si256(

include/llama.h

@@ -175,7 +175,6 @@ extern "C" {
         LLAMA_FTYPE_MOSTLY_Q4_0_8_8      = 35, // except 1d tensors
         LLAMA_FTYPE_MOSTLY_TQ1_0         = 36, // except 1d tensors
         LLAMA_FTYPE_MOSTLY_TQ2_0         = 37, // except 1d tensors
-
         LLAMA_FTYPE_MOSTLY_E5M2          = 38, // except 1d tensors
         LLAMA_FTYPE_MOSTLY_E4M3          = 39, // except 1d tensors
         LLAMA_FTYPE_MOSTLY_E4M3_Q        = 40, // except 1d tensors

scripts/sync-ggml.sh

@@ -15,14 +15,14 @@ cp -rpv ../ggml/src/ggml-cann.cpp       ./ggml/src/ggml-cann.cpp
 cp -rpv ../ggml/src/ggml-common.h       ./ggml/src/ggml-common.h
 cp -rpv ../ggml/src/ggml-cuda/*         ./ggml/src/ggml-cuda/
 cp -rpv ../ggml/src/ggml-cuda.cu        ./ggml/src/ggml-cuda.cu
+cp -rpv ../ggml/src/ggml-fp8.cpp        ./ggml/src/ggml-fp8.cpp
+cp -rpv ../ggml/src/ggml-fp8.h          ./ggml/src/ggml-fp8.h
 cp -rpv ../ggml/src/ggml-impl.h         ./ggml/src/ggml-impl.h
 cp -rpv ../ggml/src/ggml-kompute.cpp    ./ggml/src/ggml-kompute.cpp
 cp -rpv ../ggml/src/ggml-metal.m        ./ggml/src/ggml-metal.m
 cp -rpv ../ggml/src/ggml-metal.metal    ./ggml/src/ggml-metal.metal
 cp -rpv ../ggml/src/ggml-quants.c       ./ggml/src/ggml-quants.c
 cp -rpv ../ggml/src/ggml-quants.h       ./ggml/src/ggml-quants.h
-cp -rpv ../ggml/src/ggml-fp8.h          ./ggml/src/ggml-fp8.h
-cp -rpv ../ggml/src/ggml-fp8.cpp        ./ggml/src/ggml-fp8.cpp
 cp -rpv ../ggml/src/ggml-rpc.cpp        ./ggml/src/ggml-rpc.cpp
 cp -rpv ../ggml/src/ggml-sycl/*         ./ggml/src/ggml-sycl/
 cp -rpv ../ggml/src/ggml-sycl.cpp       ./ggml/src/ggml-sycl.cpp

src/llama.cpp

@@ -5298,7 +5298,6 @@ static std::string llama_model_ftype_name(llama_ftype ftype) {
         case LLAMA_FTYPE_MOSTLY_Q4_0_4_4: return "Q4_0_4_4";
         case LLAMA_FTYPE_MOSTLY_Q4_0_4_8: return "Q4_0_4_8";
         case LLAMA_FTYPE_MOSTLY_Q4_0_8_8: return "Q4_0_8_8";
-
         case LLAMA_FTYPE_MOSTLY_E5M2:     return "E5M2";
         case LLAMA_FTYPE_MOSTLY_E4M3:     return "E4M3";
         case LLAMA_FTYPE_MOSTLY_E4M3_Q:   return "E4M3_Q";