haraldwolff/llama.cpp
1
0
Fork 0
Code Issues Pull requests Packages Projects Releases Wiki Activity

tests : Fix compilation warnings (Linux/GCC) (#2451)

Browse source 
* fix hellaswag print format, cast away warning in test-double-float

* c++11 cannot use designated initializers

* add static to test-grad0.c internal functions

* use memcpy in test-double-float.c

* port c tests to c++

* use initializer list for ggml_init_params
This commit is contained in:
Eve 2023-08-02 04:06:19 -04:00 committed by GitHub
parent a312193e18
commit 81844fbcfd
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
7 changed files with 40 additions and 37 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

6
Makefile
View file

@ -411,13 +411,13 @@ benchmark-matmult: examples/benchmark/benchmark-matmult.cpp build-info.h ggml.o
vdot: pocs/vdot/vdot.cpp ggml.o $(OBJS) vdot: pocs/vdot/vdot.cpp ggml.o $(OBJS)
$(CXX) $(CXXFLAGS) $^ -o $@ $(LDFLAGS) $(CXX) $(CXXFLAGS) $^ -o $@ $(LDFLAGS)
tests/test-double-float: tests/test-double-float.c build-info.h ggml.o llama.o common.o $(OBJS) tests/test-double-float: tests/test-double-float.cpp build-info.h ggml.o llama.o common.o $(OBJS)
$(CXX) $(CXXFLAGS) $(filter-out %.txt,$^) -o $@ $(LDFLAGS) $(CXX) $(CXXFLAGS) $(filter-out %.txt,$^) -o $@ $(LDFLAGS)
tests/test-grad0: tests/test-grad0.c build-info.h ggml.o llama.o common.o $(OBJS) tests/test-grad0: tests/test-grad0.cpp build-info.h ggml.o llama.o common.o $(OBJS)
$(CXX) $(CXXFLAGS) $(filter-out %.txt,$^) -o $@ $(LDFLAGS) $(CXX) $(CXXFLAGS) $(filter-out %.txt,$^) -o $@ $(LDFLAGS)
tests/test-opt: tests/test-opt.c build-info.h ggml.o llama.o common.o $(OBJS) tests/test-opt: tests/test-opt.cpp build-info.h ggml.o llama.o common.o $(OBJS)
$(CXX) $(CXXFLAGS) $(filter-out %.txt,$^) -o $@ $(LDFLAGS) $(CXX) $(CXXFLAGS) $(filter-out %.txt,$^) -o $@ $(LDFLAGS)
tests/test-quantize-fns: tests/test-quantize-fns.cpp build-info.h ggml.o llama.o common.o $(OBJS) tests/test-quantize-fns: tests/test-quantize-fns.cpp build-info.h ggml.o llama.o common.o $(OBJS)

2
examples/common.cpp
View file

@ -572,7 +572,7 @@ void gpt_print_usage(int /*argc*/, char ** argv, const gpt_params & params) {
fprintf(stdout, " --temp N temperature (default: %.1f)\n", (double)params.temp); fprintf(stdout, " --temp N temperature (default: %.1f)\n", (double)params.temp);
fprintf(stdout, " --perplexity compute perplexity over each ctx window of the prompt\n"); fprintf(stdout, " --perplexity compute perplexity over each ctx window of the prompt\n");
fprintf(stdout, " --hellaswag compute HellaSwag score over random tasks from datafile supplied with -f\n"); fprintf(stdout, " --hellaswag compute HellaSwag score over random tasks from datafile supplied with -f\n");
fprintf(stdout, " --hellaswag-tasks N number of tasks to use when computing the HellaSwag score (default: %d)\n", params.hellaswag_tasks); fprintf(stdout, " --hellaswag-tasks N number of tasks to use when computing the HellaSwag score (default: %zu)\n", params.hellaswag_tasks);
fprintf(stdout, " --keep N number of tokens to keep from the initial prompt (default: %d, -1 = all)\n", params.n_keep); fprintf(stdout, " --keep N number of tokens to keep from the initial prompt (default: %d, -1 = all)\n", params.n_keep);
fprintf(stdout, " --chunks N max number of chunks to process (default: %d, -1 = all)\n", params.n_chunks); fprintf(stdout, " --chunks N max number of chunks to process (default: %d, -1 = all)\n", params.n_chunks);
if (llama_mlock_supported()) { if (llama_mlock_supported()) {

4
scripts/sync-ggml.sh
View file

@ -10,5 +10,5 @@ cp -rpv ../ggml/src/ggml-metal.m ./ggml-metal.m
cp -rpv ../ggml/src/ggml-metal.metal ./ggml-metal.metal cp -rpv ../ggml/src/ggml-metal.metal ./ggml-metal.metal
cp -rpv ../ggml/include/ggml/ggml.h ./ggml.h cp -rpv ../ggml/include/ggml/ggml.h ./ggml.h
cp -rpv ../ggml/tests/test-opt.c ./tests/test-opt.c cp -rpv ../ggml/tests/test-opt.cpp ./tests/test-opt.cpp
cp -rpv ../ggml/tests/test-grad0.c ./tests/test-grad0.c cp -rpv ../ggml/tests/test-grad0.cpp ./tests/test-grad0.cpp

6
tests/CMakeLists.txt
View file

@ -6,10 +6,10 @@ function(llama_add_test source)
add_test(NAME ${TEST_TARGET} COMMAND $<TARGET_FILE:${TEST_TARGET}> ${ARGN}) add_test(NAME ${TEST_TARGET} COMMAND $<TARGET_FILE:${TEST_TARGET}> ${ARGN})
endfunction() endfunction()
# llama_add_test(test-double-float.c) # SLOW # llama_add_test(test-double-float.cpp) # SLOW
llama_add_test(test-quantize-fns.cpp) llama_add_test(test-quantize-fns.cpp)
llama_add_test(test-quantize-perf.cpp) llama_add_test(test-quantize-perf.cpp)
llama_add_test(test-sampling.cpp) llama_add_test(test-sampling.cpp)
llama_add_test(test-tokenizer-0.cpp ${CMAKE_CURRENT_SOURCE_DIR}/../models/ggml-vocab.bin) llama_add_test(test-tokenizer-0.cpp ${CMAKE_CURRENT_SOURCE_DIR}/../models/ggml-vocab.bin)
llama_add_test(test-grad0.c) # SLOW llama_add_test(test-grad0.cpp) # SLOW
# llama_add_test(test-opt.c) # SLOW # llama_add_test(test-opt.cpp) # SLOW

12
tests/test-double-float.c → tests/test-double-float.cpp
View file

@ -3,10 +3,11 @@
// This is done by checking all finite (non-NaN, non-infinite) floats. // This is done by checking all finite (non-NaN, non-infinite) floats.
#undef NDEBUG #undef NDEBUG
#include <assert.h> #include <cassert>
#include <immintrin.h> #include <immintrin.h>
#include <math.h> #include <cmath>
#include <stdint.h> #include <cstdint>
#include <cstring>
#pragma GCC diagnostic push #pragma GCC diagnostic push
#pragma GCC diagnostic ignored "-Wdouble-promotion" #pragma GCC diagnostic ignored "-Wdouble-promotion"
@ -32,8 +33,9 @@ inline static float silu_float(float x) {
int main(void) { int main(void) {
uint32_t x = UINT32_MAX; uint32_t x = UINT32_MAX;
do { do {
float f = *(float *)&x; float f;
assert(!isfinite(f) || (round_orig(f) == round_float(f))); memcpy(&f, &x, sizeof(x));
assert(!std::isfinite(f) || (round_orig(f) == round_float(f)));
} while (x--); } while (x--);
#ifdef __F16C__ #ifdef __F16C__

32
tests/test-grad0.c → tests/test-grad0.cpp
View file

@ -1,10 +1,10 @@
#define _CRT_SECURE_NO_DEPRECATE // Disables ridiculous "unsafe" warnigns on Windows #define _CRT_SECURE_NO_DEPRECATE // Disables ridiculous "unsafe" warnigns on Windows
#include "ggml.h" #include "ggml.h"
#include <math.h> #include <cmath>
#include <stdio.h> #include <cstdio>
#include <stdlib.h> #include <cstdlib>
#include <assert.h> #include <cassert>
#if defined(_MSC_VER) #if defined(_MSC_VER)
#pragma warning(disable: 4244 4267) // possible loss of data #pragma warning(disable: 4244 4267) // possible loss of data
@ -47,16 +47,16 @@
#define GGML_PRINT(...) printf(__VA_ARGS__) #define GGML_PRINT(...) printf(__VA_ARGS__)
float frand(void) { static float frand(void) {
return (float)rand()/(float)RAND_MAX; return (float)rand()/(float)RAND_MAX;
} }
int irand(int n) { static int irand(int n) {
if (n == 0) return 0; if (n == 0) return 0;
return rand()%n; return rand()%n;
} }
void get_random_dims(int64_t * dims, int ndims) { static void get_random_dims(int64_t * dims, int ndims) {
dims[0] = dims[1] = dims[2] = dims[3] = 1; dims[0] = dims[1] = dims[2] = dims[3] = 1;
for (int i = 0; i < ndims; i++) { for (int i = 0; i < ndims; i++) {
@ -64,7 +64,7 @@ void get_random_dims(int64_t * dims, int ndims) {
} }
} }
struct ggml_tensor * get_random_tensor_f32( static struct ggml_tensor * get_random_tensor_f32(
struct ggml_context * ctx0, struct ggml_context * ctx0,
int ndims, int ndims,
int64_t ne[], int64_t ne[],
@ -112,7 +112,7 @@ struct ggml_tensor * get_random_tensor_f32(
return result; return result;
} }
struct ggml_tensor * get_random_tensor_f16( static struct ggml_tensor * get_random_tensor_f16(
struct ggml_context * ctx0, struct ggml_context * ctx0,
int ndims, int ndims,
int64_t ne[], int64_t ne[],
@ -160,7 +160,7 @@ struct ggml_tensor * get_random_tensor_f16(
return result; return result;
} }
struct ggml_tensor * get_random_tensor_i32( static struct ggml_tensor * get_random_tensor_i32(
struct ggml_context * ctx0, struct ggml_context * ctx0,
int ndims, int ndims,
int64_t ne[], int64_t ne[],
@ -208,7 +208,7 @@ struct ggml_tensor * get_random_tensor_i32(
return result; return result;
} }
void print_elements(const char* label, const struct ggml_tensor * t) { static void print_elements(const char* label, const struct ggml_tensor * t) {
if (!t) { if (!t) {
printf("%s: %s = null\n", __func__, label); printf("%s: %s = null\n", __func__, label);
return; return;
@ -228,7 +228,7 @@ void print_elements(const char* label, const struct ggml_tensor * t) {
} }
bool check_gradient( static bool check_gradient(
const char * op_name, const char * op_name,
struct ggml_context * ctx0, struct ggml_context * ctx0,
struct ggml_tensor * x[], struct ggml_tensor * x[],
@ -310,7 +310,7 @@ bool check_gradient(
} }
// TODO: clean-up this .. // TODO: clean-up this ..
bool check_mat_mul( static bool check_mat_mul(
const struct ggml_tensor * y, const struct ggml_tensor * y,
const struct ggml_tensor * x0, const struct ggml_tensor * x0,
const struct ggml_tensor * x1) { const struct ggml_tensor * x1) {
@ -373,9 +373,9 @@ bool check_mat_mul(
int main(int argc, const char ** argv) { int main(int argc, const char ** argv) {
struct ggml_init_params params = { struct ggml_init_params params = {
.mem_size = 128*1024*1024, /* .mem_size = */ 128*1024*1024,
.mem_buffer = NULL, /* .mem_buffer = */ NULL,
.no_alloc = false, /* .no_alloc = */ false,
}; };
int64_t ne[4]; int64_t ne[4];

15
tests/test-opt.c → tests/test-opt.cpp
View file

@ -1,9 +1,9 @@
#include "ggml.h" #include "ggml.h"
#include <math.h> #include <cmath>
#include <stdio.h> #include <cstdio>
#include <stdlib.h> #include <cstdlib>
#include <assert.h> #include <cassert>
#define MAX_NARGS 2 #define MAX_NARGS 2
@ -119,10 +119,11 @@ void set_element(struct ggml_tensor * t, int idx, float value) {
int main(void) { int main(void) {
struct ggml_init_params params = { struct ggml_init_params params = {
.mem_size = 1024*1024*1024, /* .mem_size = */ 1024*1024*1024,
.mem_buffer = NULL, /* .mem_buffer = */ NULL,
.no_alloc = false, /* .no_alloc = */ false,
}; };
struct ggml_context * ctx = ggml_init(params); struct ggml_context * ctx = ggml_init(params);
int64_t ne1[4] = {4, 128, 1, 1}; int64_t ne1[4] = {4, 128, 1, 1};