root/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/llama.cpp.git synced 2024-12-28 12:24:35 +00:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

rebase to master (except ggml-cuda)

Browse Source
This commit is contained in:
slaren 2023-07-16 14:36:32 +02:00
parent 33ab185dd1
commit 9c72e7e916
5 changed files with 21 additions and 30 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
ggml-backend.c
View File

@ -289,7 +289,7 @@ void ggml_graph_splits_add_n_va(struct ggml_graph_splits * splits, struct ggml_t
if ((*inputs[0])->backend == ggml_get_ctx_backend(ctx)) { if ((*inputs[0])->backend == ggml_get_ctx_backend(ctx)) {
if (splits->n_splits > 0) { if (splits->n_splits > 0) {
char name[GGML_MAX_NAME - 1]; // silence -Wformat-truncation char name[GGML_MAX_NAME];
vsnprintf(name, sizeof(name), fmt, args); vsnprintf(name, sizeof(name), fmt, args);
char new_name[GGML_MAX_NAME]; char new_name[GGML_MAX_NAME];
snprintf(new_name, sizeof(new_name), "%s,%s", splits->splits[splits->n_splits - 1].name, name); snprintf(new_name, sizeof(new_name), "%s,%s", splits->splits[splits->n_splits - 1].name, name);

30
ggml-cuda.cu
View File

@ -1475,8 +1475,8 @@ static void ggml_cuda_mul_mat(ggml_cuda_context * ctx, ggml_tensor * src0, ggml_
} }
static void ggml_cuda_exec_node(ggml_cuda_context * ctx, ggml_tensor * node, cudaStream_t stream) { static void ggml_cuda_exec_node(ggml_cuda_context * ctx, ggml_tensor * node, cudaStream_t stream) {
ggml_tensor * src0 = node->src0; ggml_tensor * src0 = node->src[0];
ggml_tensor * src1 = node->src1; ggml_tensor * src1 = node->src[1];
ggml_tensor * dst = node; ggml_tensor * dst = node;
#if 0 #if 0
@ -1551,8 +1551,6 @@ static void ggml_cuda_exec_node(ggml_cuda_context * ctx, ggml_tensor * node, cud
} }
} }
static const int GGML_MAX_PARENTS = 2 + GGML_MAX_OPT;
static bool ggml_is_noop(ggml_tensor * t) { static bool ggml_is_noop(ggml_tensor * t) {
return t->op == GGML_OP_RESHAPE || t->op == GGML_OP_VIEW || t->op == GGML_OP_TRANSPOSE || return t->op == GGML_OP_RESHAPE || t->op == GGML_OP_VIEW || t->op == GGML_OP_TRANSPOSE ||
t->op == GGML_OP_PERMUTE || t->op == GGML_OP_NONE; t->op == GGML_OP_PERMUTE || t->op == GGML_OP_NONE;
@ -1581,26 +1579,20 @@ static void ggml_cuda_graph_exec_parallel(ggml_cuda_context * ctx, ggml_cgraph *
ggml_tensor * node = gf->nodes[i]; ggml_tensor * node = gf->nodes[i];
const bool is_noop = ggml_is_noop(node); const bool is_noop = ggml_is_noop(node);
// build a list of parents
ggml_tensor * parents[GGML_MAX_PARENTS] = { node->src0, node->src1 };
for (int j = 0; j < GGML_MAX_OPT; j++) {
parents[j + 2] = node->opt[j];
}
// assign an stream for the node // assign an stream for the node
cudaStream_t stream = nullptr; cudaStream_t stream = nullptr;
// take a stream from a parent // take a stream from a parent
for (int j = 0; j < GGML_MAX_PARENTS; j++) { for (int j = 0; j < GGML_MAX_SRC; j++) {
if (parents[j] && stream_map.count(parents[j]) && stream_map[parents[j]] != nullptr) { if (node->src[j] && stream_map.count(node->src[j]) && stream_map[node->src[j]] != nullptr) {
stream = stream_map[parents[j]]; stream = stream_map[node->src[j]];
stream_map.erase(parents[j]); stream_map.erase(node->src[j]);
if (is_noop) { if (is_noop) {
// if this is a noop, we can use the parent's event // if this is a noop, we can use the parent's event
stream_map[node] = stream; stream_map[node] = stream;
if (event_map.count(parents[j]) > 0) { if (event_map.count(node->src[j]) > 0) {
event_map[node] = event_map[parents[j]]; event_map[node] = event_map[node->src[j]];
} }
} }
break; break;
@ -1624,9 +1616,9 @@ static void ggml_cuda_graph_exec_parallel(ggml_cuda_context * ctx, ggml_cgraph *
// wait on parent streams // wait on parent streams
bool waited = false; bool waited = false;
for (int j = 0; j < GGML_MAX_PARENTS; j++) { for (int j = 0; j < GGML_MAX_SRC; j++) {
if (parents[j] && event_map.count(parents[j]) > 0) { if (node->src[j] && event_map.count(node->src[j]) > 0) {
CUDA_CHECK(cudaStreamWaitEvent(stream, event_map[parents[j]], 0)); CUDA_CHECK(cudaStreamWaitEvent(stream, event_map[node->src[j]], 0));
waited = true; waited = true;
} }
} }

11
ggml.c
View File

@ -6855,7 +6855,9 @@ struct ggml_tensor * ggml_rope_impl(
struct ggml_tensor * result = inplace ? ggml_view_tensor(ctx, a) : ggml_dup_tensor(ctx, a); struct ggml_tensor * result = inplace ? ggml_view_tensor(ctx, a) : ggml_dup_tensor(ctx, a);
// TODO: just use a struct // TODO: just use a struct
int32_t params[] = { n_past, n_dims, mode, n_ctx, *(int32_t*)&freq_base, *(int32_t*)&freq_scale}; int32_t params[6] = { n_past, n_dims, mode, n_ctx };
memcpy(params + 4, &freq_base, sizeof(float));
memcpy(params + 5, &freq_scale, sizeof(float));
assert(GGML_MAX_OP_PARAMS >= sizeof(params)); assert(GGML_MAX_OP_PARAMS >= sizeof(params));
memcpy(result->params, &params, sizeof(params)); memcpy(result->params, &params, sizeof(params));
@ -7127,13 +7129,11 @@ struct ggml_tensor* ggml_pool_1d(
}; };
struct ggml_tensor* result = ggml_new_tensor(ctx, GGML_TYPE_F32, 2, ne); struct ggml_tensor* result = ggml_new_tensor(ctx, GGML_TYPE_F32, 2, ne);
ggml_scratch_save(ctx);
struct ggml_tensor* c = ggml_new_tensor_1d(ctx, GGML_TYPE_I32, 4); struct ggml_tensor* c = ggml_new_tensor_1d(ctx, GGML_TYPE_I32, 4);
((int32_t*)c->data)[0] = op; ((int32_t*)c->data)[0] = op;
((int32_t*)c->data)[1] = k0; ((int32_t*)c->data)[1] = k0;
((int32_t*)c->data)[2] = s0; ((int32_t*)c->data)[2] = s0;
((int32_t*)c->data)[3] = p0; ((int32_t*)c->data)[3] = p0;
ggml_scratch_load(ctx);
result->op = GGML_OP_POOL_1D; result->op = GGML_OP_POOL_1D;
result->grad = is_node ? ggml_dup_tensor(ctx, result) : NULL; result->grad = is_node ? ggml_dup_tensor(ctx, result) : NULL;
@ -7170,7 +7170,6 @@ struct ggml_tensor* ggml_pool_2d(
}; };
struct ggml_tensor* result = ggml_new_tensor(ctx, GGML_TYPE_F32, 3, ne); struct ggml_tensor* result = ggml_new_tensor(ctx, GGML_TYPE_F32, 3, ne);
ggml_scratch_save(ctx);
struct ggml_tensor* c = ggml_new_tensor_1d(ctx, GGML_TYPE_I32, 7); struct ggml_tensor* c = ggml_new_tensor_1d(ctx, GGML_TYPE_I32, 7);
((int32_t*)c->data)[0] = op; ((int32_t*)c->data)[0] = op;
((int32_t*)c->data)[1] = k0; ((int32_t*)c->data)[1] = k0;
@ -7179,7 +7178,6 @@ struct ggml_tensor* ggml_pool_2d(
((int32_t*)c->data)[4] = s1; ((int32_t*)c->data)[4] = s1;
((int32_t*)c->data)[5] = p0; ((int32_t*)c->data)[5] = p0;
((int32_t*)c->data)[6] = p1; ((int32_t*)c->data)[6] = p1;
ggml_scratch_load(ctx);
result->op = GGML_OP_POOL_2D; result->op = GGML_OP_POOL_2D;
result->grad = is_node ? ggml_dup_tensor(ctx, result) : NULL; result->grad = is_node ? ggml_dup_tensor(ctx, result) : NULL;
@ -15823,7 +15821,8 @@ static void ggml_visit_parents(struct ggml_cgraph * cgraph, struct ggml_tensor *
} }
} }
if (node->op == GGML_OP_NONE && node->src0 == NULL && node->src1 == NULL && node->grad == NULL) { // TODO: add ggml_dependency instead of checking for NULL
if (node->op == GGML_OP_NONE && node->src[0] == NULL && node->src[1] == NULL && node->grad == NULL) {
// reached a leaf node, not part of the gradient graph (e.g. a constant) // reached a leaf node, not part of the gradient graph (e.g. a constant)
GGML_ASSERT(cgraph->n_leafs < GGML_MAX_NODES); GGML_ASSERT(cgraph->n_leafs < GGML_MAX_NODES);

2
ggml.h
View File

@ -199,7 +199,7 @@
#define GGML_MAX_CONTEXTS 64 #define GGML_MAX_CONTEXTS 64
#define GGML_MAX_SRC 6 #define GGML_MAX_SRC 6
#define GGML_MAX_NAME 48 #define GGML_MAX_NAME 48
#define GGML_MAX_OP_PARAMS 16 #define GGML_MAX_OP_PARAMS 32
#define GGML_DEFAULT_N_THREADS 4 #define GGML_DEFAULT_N_THREADS 4

6
llama.cpp
View File

@ -1168,7 +1168,7 @@ static ggml_graph_splits llama_build_graph(
struct ggml_graph_splits splits = ggml_graph_split_init(); struct ggml_graph_splits splits = ggml_graph_split_init();
// initalize contexts for every backend // initialize contexts for every backend
struct ggml_context * ctx_cpu = nullptr; struct ggml_context * ctx_cpu = nullptr;
// TODO: don't create context if there are no CPU layers // TODO: don't create context if there are no CPU layers
@ -1295,8 +1295,8 @@ static ggml_graph_splits llama_build_graph(
// TODO: replace with ggml_dependency / ggml_depends_on // TODO: replace with ggml_dependency / ggml_depends_on
k = ggml_view_tensor(ctx_kv, kv_self.k); k = ggml_view_tensor(ctx_kv, kv_self.k);
v = ggml_view_tensor(ctx_kv, kv_self.v); v = ggml_view_tensor(ctx_kv, kv_self.v);
k->src0 = k_cpy; k->src[0] = k_cpy;
v->src0 = v_cpy; v->src[0] = v_cpy;
} }
struct ggml_tensor * Q = struct ggml_tensor * Q =