npm - cui-llama.rn - Versions diffs - 1.7.3 → 1.7.6 - Mend

cui-llama.rn 1.7.3 → 1.7.6

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.

Files changed (276) hide show

package/cpp/ggml-cpu/{ggml-cpu-quants.h → quants.h} RENAMED Viewed

@@ -58,6 +58,32 @@ void lm_ggml_vec_dot_iq4_nl_q8_0 (int n, float * LM_GGML_RESTRICT s, size_t bs,
 void lm_ggml_vec_dot_iq4_xs_q8_K (int n, float * LM_GGML_RESTRICT s, size_t bs, const void * LM_GGML_RESTRICT vx, size_t bx, const void * LM_GGML_RESTRICT vy, size_t by, int nrc);
 void lm_ggml_vec_dot_iq3_s_q8_K  (int n, float * LM_GGML_RESTRICT s, size_t bs, const void * LM_GGML_RESTRICT vx, size_t bx, const void * LM_GGML_RESTRICT vy, size_t by, int nrc);
+// Generic implementation
+void quantize_row_q8_0_generic(const float * LM_GGML_RESTRICT x, void * LM_GGML_RESTRICT vy, int64_t k);
+void quantize_row_q8_1_generic(const float * LM_GGML_RESTRICT x, void * LM_GGML_RESTRICT vy, int64_t k);
+void quantize_row_q8_K_generic(const float * LM_GGML_RESTRICT x, void * LM_GGML_RESTRICT y, int64_t k);
+void lm_ggml_vec_dot_q4_0_q8_0_generic(int n, float * LM_GGML_RESTRICT s, size_t bs, const void * LM_GGML_RESTRICT vx, size_t bx, const void * LM_GGML_RESTRICT vy, size_t by, int nrc);
+void lm_ggml_vec_dot_q4_1_q8_1_generic(int n, float * LM_GGML_RESTRICT s, size_t bs, const void * LM_GGML_RESTRICT vx, size_t bx, const void * LM_GGML_RESTRICT vy, size_t by, int nrc);
+void lm_ggml_vec_dot_q5_0_q8_0_generic(int n, float * LM_GGML_RESTRICT s, size_t bs, const void * LM_GGML_RESTRICT vx, size_t bx, const void * LM_GGML_RESTRICT vy, size_t by, int nrc);
+void lm_ggml_vec_dot_q5_1_q8_1_generic(int n, float * LM_GGML_RESTRICT s, size_t bs, const void * LM_GGML_RESTRICT vx, size_t bx, const void * LM_GGML_RESTRICT vy, size_t by, int nrc);
+void lm_ggml_vec_dot_q8_0_q8_0_generic(int n, float * LM_GGML_RESTRICT s, size_t bs, const void * LM_GGML_RESTRICT vx, size_t bx, const void * LM_GGML_RESTRICT vy, size_t by, int nrc);
+void lm_ggml_vec_dot_tq1_0_q8_K_generic(int n, float * LM_GGML_RESTRICT s, size_t bs, const void * LM_GGML_RESTRICT vx, size_t bx, const void * LM_GGML_RESTRICT vy, size_t by, int nrc);
+void lm_ggml_vec_dot_tq2_0_q8_K_generic(int n, float * LM_GGML_RESTRICT s, size_t bs, const void * LM_GGML_RESTRICT vx, size_t bx, const void * LM_GGML_RESTRICT vy, size_t by, int nrc);
+void lm_ggml_vec_dot_q2_K_q8_K_generic(int n, float * LM_GGML_RESTRICT s, size_t bs, const void * LM_GGML_RESTRICT vx, size_t bx, const void * LM_GGML_RESTRICT vy, size_t by, int nrc);
+void lm_ggml_vec_dot_q3_K_q8_K_generic(int n, float * LM_GGML_RESTRICT s, size_t bs, const void * LM_GGML_RESTRICT vx, size_t bx, const void * LM_GGML_RESTRICT vy, size_t by, int nrc);
+void lm_ggml_vec_dot_q4_K_q8_K_generic(int n, float * LM_GGML_RESTRICT s, size_t bs, const void * LM_GGML_RESTRICT vx, size_t bx, const void * LM_GGML_RESTRICT vy, size_t by, int nrc);
+void lm_ggml_vec_dot_q5_K_q8_K_generic(int n, float * LM_GGML_RESTRICT s, size_t bs, const void * LM_GGML_RESTRICT vx, size_t bx, const void * LM_GGML_RESTRICT vy,  size_t by, int nrc);
+void lm_ggml_vec_dot_q6_K_q8_K_generic(int n, float * LM_GGML_RESTRICT s, size_t bs, const void * LM_GGML_RESTRICT vx, size_t bx, const void * LM_GGML_RESTRICT vy, size_t by, int nrc);
+void lm_ggml_vec_dot_iq2_xxs_q8_K_generic(int n, float * LM_GGML_RESTRICT s, size_t bs, const void * LM_GGML_RESTRICT vx, size_t bx, const void * LM_GGML_RESTRICT vy, size_t by, int nrc);
+void lm_ggml_vec_dot_iq2_xs_q8_K_generic(int n, float * LM_GGML_RESTRICT s, size_t bs, const void * LM_GGML_RESTRICT vx, size_t bx, const void * LM_GGML_RESTRICT vy, size_t by, int nrc);
+void lm_ggml_vec_dot_iq2_s_q8_K_generic(int n, float * LM_GGML_RESTRICT s, size_t bs, const void * LM_GGML_RESTRICT vx, size_t bx, const void * LM_GGML_RESTRICT vy, size_t by, int nrc);
+void lm_ggml_vec_dot_iq3_xxs_q8_K_generic(int n, float * LM_GGML_RESTRICT s, size_t bs, const void * LM_GGML_RESTRICT vx, size_t bx, const void * LM_GGML_RESTRICT vy, size_t by, int nrc);
+void lm_ggml_vec_dot_iq3_s_q8_K_generic(int n, float * LM_GGML_RESTRICT s, size_t bs, const void * LM_GGML_RESTRICT vx, size_t bx, const void * LM_GGML_RESTRICT vy, size_t by, int nrc);
+void lm_ggml_vec_dot_iq1_s_q8_K_generic(int n, float * LM_GGML_RESTRICT s, size_t bs, const void * LM_GGML_RESTRICT vx, size_t bx, const void * LM_GGML_RESTRICT vy, size_t by, int nrc);
+void lm_ggml_vec_dot_iq1_m_q8_K_generic(int n, float * LM_GGML_RESTRICT s, size_t bs, const void * LM_GGML_RESTRICT vx, size_t bx, const void * LM_GGML_RESTRICT vy, size_t by, int nrc);
+void lm_ggml_vec_dot_iq4_nl_q8_0_generic(int n, float * LM_GGML_RESTRICT s, size_t bs, const void * LM_GGML_RESTRICT vx, size_t bx, const void * LM_GGML_RESTRICT vy, size_t by, int nrc);
+void lm_ggml_vec_dot_iq4_xs_q8_K_generic(int n, float * LM_GGML_RESTRICT s, size_t bs, const void * LM_GGML_RESTRICT vx, size_t bx, const void * LM_GGML_RESTRICT vy, size_t by, int nrc);
 #ifdef __cplusplus
 }
 #endif