fix last commit, add missing files; git ignore mulmat-tune.txt

Author: mqy

.gitignore

@@ -49,3 +49,4 @@ qnt-*.txt
 perf-*.txt
 
 examples/jeopardy/results.txt
+mulmat-tune.txt

Makefile

@@ -228,7 +228,7 @@ OBJS += ggml-tune.o
 # Build library
 #
 
-ggml.o: ggml.c ggml.h ggml-cuda.h
+ggml.o: ggml.c ggml.h ggml-cuda.h ggml-tune.h
 	$(CC)  $(CFLAGS)   -c $< -o $@
 
 llama.o: llama.cpp ggml.h ggml-cuda.h llama.h llama-util.h
@@ -290,7 +290,7 @@ benchmark-matmult: examples/benchmark/benchmark-matmult.cpp build-info.h ggml.o
 vdot: pocs/vdot/vdot.cpp ggml.o $(OBJS)
 	$(CXX) $(CXXFLAGS) $^ -o $@ $(LDFLAGS)
 
-ggml-tune.o: ggml-tune.c ggml-tune.h
+ggml-tune.o: ggml-tune.c ggml.h ggml-tune.h
 	$(CC)  $(CFLAGS) -c $< -o $@
 
 mulmat-tune: examples/mulmat-tune/mulmat-tune.c ggml.o $(OBJS)

examples/mulmat-tune/README.md

@@ -116,7 +116,7 @@ In conclusion, let me list the key point how does it work.
    estimate total computing time for any M (even if out of bench range), thus
    could select the fastest profile.
 5. On llama start, it loads the bench data from file (if exists). Before computing
-   node, we select the fastest profile. When compute, both `dst->task_conf` and 
+   node, we select the fastest profile. When compute, both `dst->task_conf` and
    `params` control which part of the codes to run.
 
 

examples/mulmat-tune/mulmat-tune.c

@@ -1,14 +1,5 @@
-#include "ggml-tune.h"
 #include "ggml.h"
 
-// #if defined GGML_USE_CLBLAST
-// #include "ggml-cuda.h"
-// #endif
-
-// #if defined GGML_USE_CLBLAST
-// #include "ggml-opencl.h"
-// #endif
-
 #include <inttypes.h>
 #include <stdio.h>
 #include <stdlib.h>