Use wdata for storing dequantized weights

ggml.c

@@ -6843,16 +6843,15 @@ static void ggml_compute_forward_mul_mat_q_f32(
     const int ir0 = dr*ith;
     const int ir1 = MIN(ir0 + dr, nr);
 
-    // TODO Alloc somewhere else, or maybe use wdata
-    float * dequantized = calloc(ne00, sizeof(float));
-
     for (int ir = ir0; ir < ir1; ++ir) {
         // src0 indices
         const int i03 = ir/(ne02*ne01);
         const int i02 = (ir - i03*ne02*ne01)/ne01;
         const int i01 = (ir - i03*ne02*ne01 - i02*ne01);
 
-        dequantize_row_q((char *) src0->data + (i01 * nb01 + i02 * nb02 + i03 * nb03), dequantized, ne00);
+        float * const wdata = (float *) ((char *) params->wdata + (i01 * nb01 + i02 * nb02 + i03 * nb03));
+
+        dequantize_row_q((char *) src0->data + (i01 * nb01 + i02 * nb02 + i03 * nb03), wdata, ne00);
 
         for (int ic = 0; ic < ne11; ++ic) {
             // src1 indices
@@ -6869,13 +6868,11 @@ static void ggml_compute_forward_mul_mat_q_f32(
             ggml_vec_dot_f32(
                     ne00,
                     (float *) ((char *) dst->data + (i0 * nb0 + i1 * nb1 + i2 * nb2 + i3 * nb3)),
-                    dequantized,
+                    wdata,
                     (float *) ((char *) src1->data + (i11 * nb11 + i12 * nb12 + i13 * nb13)));
         }
     }
 
-    free(dequantized);
-
     //int64_t t1 = ggml_time_us();
     //static int64_t acc = 0;
     //acc += t1 - t0;
@@ -9546,7 +9543,7 @@ void ggml_graph_compute(struct ggml_context * ctx, struct ggml_cgraph * cgraph)
                             } else
 #endif
                             {
-                                cur = GGML_TYPE_SIZE[node->src0->type]*ggml_nelements(node->src1)/GGML_BLCK_SIZE[node->src0->type];
+                                cur = GGML_TYPE_SIZE[GGML_TYPE_F32]*(node->src0->ne[0]*node->src0->ne[1]);
                             }
                         } else {
                             GGML_ASSERT(false);