Convergence LCPP/IKL and cleanup

Author: Nexesenex

ggml/src/ggml-cpu/ggml-cpu-quants.h

@@ -5,6 +5,8 @@
 
 #include "ggml.h"
 
+#include "iqk_croco/iqk_quantize_croco.h"
+
 #include <stdint.h>
 #include <stddef.h>
 
@@ -36,21 +38,21 @@ void quantize_row_tq2_0(const float * GGML_RESTRICT x, void * GGML_RESTRICT y, i
 void quantize_row_iq4_nl (const float * GGML_RESTRICT x, void * GGML_RESTRICT y, int64_t k);
 void quantize_row_iq4_xs (const float * GGML_RESTRICT x, void * GGML_RESTRICT y, int64_t k);
 
-void quantize_row_iq2_ks (const float * GGML_RESTRICT x, void * GGML_RESTRICT y, int64_t k);
-void quantize_row_iq2_k (const float * GGML_RESTRICT x, void * GGML_RESTRICT y, int64_t k);
-void quantize_row_iq3_k (const float * GGML_RESTRICT x, void * GGML_RESTRICT y, int64_t k);
-void quantize_row_iq4_kss (const float * GGML_RESTRICT x, void * GGML_RESTRICT y, int64_t k);
-void quantize_row_iq4_ks (const float * GGML_RESTRICT x, void * GGML_RESTRICT y, int64_t k);
-void quantize_row_iq4_k (const float * GGML_RESTRICT x, void * GGML_RESTRICT y, int64_t k);
-void quantize_row_iq5_k (const float * GGML_RESTRICT x, void * GGML_RESTRICT y, int64_t k);
-void quantize_row_iq5_ks (const float * GGML_RESTRICT x, void * GGML_RESTRICT y, int64_t k);
-void quantize_row_iq3_ks (const float * GGML_RESTRICT x, void * GGML_RESTRICT y, int64_t k);
-void quantize_row_iq6_k (const float * GGML_RESTRICT x, void * GGML_RESTRICT y, int64_t k);
-void quantize_row_iq2_kt (const float * GGML_RESTRICT x, void * GGML_RESTRICT y, int64_t k);
-void quantize_row_iq3_kt (const float * GGML_RESTRICT x, void * GGML_RESTRICT y, int64_t k);
-void quantize_row_iq4_kt (const float * GGML_RESTRICT x, void * GGML_RESTRICT y, int64_t k);
-void quantize_row_iq1_bn (const float * GGML_RESTRICT x, void * GGML_RESTRICT y, int64_t k);
-void quantize_row_iq2_bn (const float * GGML_RESTRICT x, void * GGML_RESTRICT y, int64_t k);
+// void quantize_row_iq2_ks (const float * GGML_RESTRICT x, void * GGML_RESTRICT y, int64_t k);
+// void quantize_row_iq2_k (const float * GGML_RESTRICT x, void * GGML_RESTRICT y, int64_t k);
+// void quantize_row_iq3_k (const float * GGML_RESTRICT x, void * GGML_RESTRICT y, int64_t k);
+// void quantize_row_iq4_kss (const float * GGML_RESTRICT x, void * GGML_RESTRICT y, int64_t k);
+// void quantize_row_iq4_ks (const float * GGML_RESTRICT x, void * GGML_RESTRICT y, int64_t k);
+// void quantize_row_iq4_k (const float * GGML_RESTRICT x, void * GGML_RESTRICT y, int64_t k);
+// void quantize_row_iq5_k (const float * GGML_RESTRICT x, void * GGML_RESTRICT y, int64_t k);
+// void quantize_row_iq5_ks (const float * GGML_RESTRICT x, void * GGML_RESTRICT y, int64_t k);
+// void quantize_row_iq3_ks (const float * GGML_RESTRICT x, void * GGML_RESTRICT y, int64_t k);
+// void quantize_row_iq6_k (const float * GGML_RESTRICT x, void * GGML_RESTRICT y, int64_t k);
+// void quantize_row_iq2_kt (const float * GGML_RESTRICT x, void * GGML_RESTRICT y, int64_t k);
+// void quantize_row_iq3_kt (const float * GGML_RESTRICT x, void * GGML_RESTRICT y, int64_t k);
+// void quantize_row_iq4_kt (const float * GGML_RESTRICT x, void * GGML_RESTRICT y, int64_t k);
+// void quantize_row_iq1_bn (const float * GGML_RESTRICT x, void * GGML_RESTRICT y, int64_t k);
+// void quantize_row_iq2_bn (const float * GGML_RESTRICT x, void * GGML_RESTRICT y, int64_t k);
 
 
 // Dot product
@@ -80,22 +82,22 @@ void ggml_vec_dot_iq4_nl_q8_0 (int n, float * GGML_RESTRICT s, size_t bs, const
 void ggml_vec_dot_iq4_xs_q8_K (int n, float * GGML_RESTRICT s, size_t bs, const void * GGML_RESTRICT vx, size_t bx, const void * GGML_RESTRICT vy, size_t by, int nrc);
 void ggml_vec_dot_iq3_s_q8_K  (int n, float * GGML_RESTRICT s, size_t bs, const void * GGML_RESTRICT vx, size_t bx, const void * GGML_RESTRICT vy, size_t by, int nrc);
 
-void vec_dot_iq2_ks_q8_K (int n, float * GGML_RESTRICT s, size_t bs, const void * GGML_RESTRICT vx, size_t bx, const void * GGML_RESTRICT vy, size_t by, int nrc);
-void vec_dot_iq2_k_q8_K (int n, float * GGML_RESTRICT s, size_t bs, const void * GGML_RESTRICT vx, size_t bx, const void * GGML_RESTRICT vy, size_t by, int nrc);
-void vec_dot_iq3_k_q8_K  (int n, float * GGML_RESTRICT s, size_t bs, const void * GGML_RESTRICT vx, size_t bx, const void * GGML_RESTRICT vy, size_t by, int nrc);
-void vec_dot_iq4_kss_q8_K(int n, float * GGML_RESTRICT s, size_t bs, const void * GGML_RESTRICT vx, size_t bx, const void * GGML_RESTRICT vy, size_t by, int nrc);
-void vec_dot_iq4_ks_q8_K  (int n, float * GGML_RESTRICT s, size_t bs, const void * GGML_RESTRICT vx, size_t bx, const void * GGML_RESTRICT vy, size_t by, int nrc);
-void vec_dot_iq4_k_q8_K  (int n, float * GGML_RESTRICT s, size_t bs, const void * GGML_RESTRICT vx, size_t bx, const void * GGML_RESTRICT vy, size_t by, int nrc);
-void vec_dot_iq5_k_q8_K (int n, float * GGML_RESTRICT s, size_t bs, const void * GGML_RESTRICT vx, size_t bx, const void * GGML_RESTRICT vy, size_t by, int nrc);
-void vec_dot_iq5_ks_q8_K (int n, float * GGML_RESTRICT s, size_t bs, const void * GGML_RESTRICT vx, size_t bx, const void * GGML_RESTRICT vy, size_t by, int nrc);
-void vec_dot_iq3_ks_q8_K (int n, float * GGML_RESTRICT s, size_t bs, const void * GGML_RESTRICT vx, size_t bx, const void * GGML_RESTRICT vy, size_t by, int nrc);
-void vec_dot_iq6_k_q8_K (int n, float * GGML_RESTRICT s, size_t bs, const void * GGML_RESTRICT vx, size_t bx, const void * GGML_RESTRICT vy, size_t by, int nrc);
-void vec_dot_iq2_kt_q8_K  (int n, float * GGML_RESTRICT s, size_t bs, const void * GGML_RESTRICT vx, size_t bx, const void * GGML_RESTRICT vy, size_t by, int nrc);
-void vec_dot_iq3_kt_q8_K (int n, float * GGML_RESTRICT s, size_t bs, const void * GGML_RESTRICT vx, size_t bx, const void * GGML_RESTRICT vy, size_t by, int nrc);
-void vec_dot_iq4_kt_q8_K  (int n, float * GGML_RESTRICT s, size_t bs, const void * GGML_RESTRICT vx, size_t bx, const void * GGML_RESTRICT vy, size_t by, int nrc);
-void vec_dot_iq1_bn_q8_0 (int n, float * GGML_RESTRICT s, size_t bs, const void * GGML_RESTRICT vx, size_t bx, const void * GGML_RESTRICT vy, size_t by, int nrc);
-void vec_dot_iq1_bn_q8_K64(int n, float * GGML_RESTRICT s, size_t bs, const void * GGML_RESTRICT vx, size_t bx, const void * GGML_RESTRICT vy, size_t by, int nrc);
-void vec_dot_iq2_bn_q8_0  (int n, float * GGML_RESTRICT s, size_t bs, const void * GGML_RESTRICT vx, size_t bx, const void * GGML_RESTRICT vy, size_t by, int nrc);
+// void vec_dot_iq2_ks_q8_K (int n, float * GGML_RESTRICT s, size_t bs, const void * GGML_RESTRICT vx, size_t bx, const void * GGML_RESTRICT vy, size_t by, int nrc);
+// void vec_dot_iq2_k_q8_K (int n, float * GGML_RESTRICT s, size_t bs, const void * GGML_RESTRICT vx, size_t bx, const void * GGML_RESTRICT vy, size_t by, int nrc);
+// void vec_dot_iq3_k_q8_K  (int n, float * GGML_RESTRICT s, size_t bs, const void * GGML_RESTRICT vx, size_t bx, const void * GGML_RESTRICT vy, size_t by, int nrc);
+// void vec_dot_iq4_kss_q8_K(int n, float * GGML_RESTRICT s, size_t bs, const void * GGML_RESTRICT vx, size_t bx, const void * GGML_RESTRICT vy, size_t by, int nrc);
+// void vec_dot_iq4_ks_q8_K  (int n, float * GGML_RESTRICT s, size_t bs, const void * GGML_RESTRICT vx, size_t bx, const void * GGML_RESTRICT vy, size_t by, int nrc);
+// void vec_dot_iq4_k_q8_K  (int n, float * GGML_RESTRICT s, size_t bs, const void * GGML_RESTRICT vx, size_t bx, const void * GGML_RESTRICT vy, size_t by, int nrc);
+// void vec_dot_iq5_k_q8_K (int n, float * GGML_RESTRICT s, size_t bs, const void * GGML_RESTRICT vx, size_t bx, const void * GGML_RESTRICT vy, size_t by, int nrc);
+// void vec_dot_iq5_ks_q8_K (int n, float * GGML_RESTRICT s, size_t bs, const void * GGML_RESTRICT vx, size_t bx, const void * GGML_RESTRICT vy, size_t by, int nrc);
+// void vec_dot_iq3_ks_q8_K (int n, float * GGML_RESTRICT s, size_t bs, const void * GGML_RESTRICT vx, size_t bx, const void * GGML_RESTRICT vy, size_t by, int nrc);
+// void vec_dot_iq6_k_q8_K (int n, float * GGML_RESTRICT s, size_t bs, const void * GGML_RESTRICT vx, size_t bx, const void * GGML_RESTRICT vy, size_t by, int nrc);
+// void vec_dot_iq2_kt_q8_K  (int n, float * GGML_RESTRICT s, size_t bs, const void * GGML_RESTRICT vx, size_t bx, const void * GGML_RESTRICT vy, size_t by, int nrc);
+// void vec_dot_iq3_kt_q8_K (int n, float * GGML_RESTRICT s, size_t bs, const void * GGML_RESTRICT vx, size_t bx, const void * GGML_RESTRICT vy, size_t by, int nrc);
+// void vec_dot_iq4_kt_q8_K  (int n, float * GGML_RESTRICT s, size_t bs, const void * GGML_RESTRICT vx, size_t bx, const void * GGML_RESTRICT vy, size_t by, int nrc);
+// void vec_dot_iq1_bn_q8_0 (int n, float * GGML_RESTRICT s, size_t bs, const void * GGML_RESTRICT vx, size_t bx, const void * GGML_RESTRICT vy, size_t by, int nrc);
+// void vec_dot_iq1_bn_q8_K64(int n, float * GGML_RESTRICT s, size_t bs, const void * GGML_RESTRICT vx, size_t bx, const void * GGML_RESTRICT vy, size_t by, int nrc);
+// void vec_dot_iq2_bn_q8_0  (int n, float * GGML_RESTRICT s, size_t bs, const void * GGML_RESTRICT vx, size_t bx, const void * GGML_RESTRICT vy, size_t by, int nrc);
 
 #ifdef __cplusplus
 }

ggml/src/ggml-cpu/ggml-cpu.c

@@ -16,7 +16,6 @@
 #include "ggml.h"
 #include "gguf.h"
 
-#include "iqk_croco/iqk_quantize_croco.h"
 #include "iqk_croco/iqk_quantize_croco.h"
 
 #if defined(_MSC_VER) || defined(__MINGW32__)

ggml/src/ggml-cuda/convert (2).cuh_

@@ -212,10 +212,6 @@ void quantize_row_q5_1_ref(const float * GGML_RESTRICT x, block_q5_1 * GGML_REST
     }
 }
 
-// void quantize_row_q5_1(const float * restrict x, void * restrict y, int64_t k) {
-    // quantize_row_q5_1_ref(x, y, k);
-// }
-
 void quantize_row_q6_0_ref(const float * GGML_RESTRICT x, block_q6_0 * GGML_RESTRICT y, int64_t k) {
     static const int qk = QK6_0;
 
@@ -5730,7 +5726,6 @@ bool ggml_validate_row_data(enum ggml_type type, const void * data, size_t nbyte
         // case GGML_TYPE_Q8_K_R8:  break;
         // case GGML_TYPE_Q8_KV:    break;
         // case GGML_TYPE_BF16_R16: break;
-        // case GGML_TYPE_IQ2_KS: break;
         case GGML_TYPE_IQ2_KT: break;
         case GGML_TYPE_IQ3_KT: break;
         case GGML_TYPE_IQ4_KT: break;