Skip to content

Commit

Permalink
ggml: avoid rebuild of GGML graph for each token (ggml-org#7456)
Browse files Browse the repository at this point in the history 
Introduces caching of GGML graph to avoid unnecessary full rebuild
between each token. KV cache parameters, which change with each token,
are updated directly in cached GGML graph. Can be disabled with
GGML_DISABLE_GRAPH_CACHING environment variable.

fix seg fault

restrict to nsplit=2

Improve identification of K and V nodes for param updates

Up LCPP Graph PR by Agray3
  • Loading branch information
@agray3 @Nexesenex
agray3 authored and Nexesenex committed Jul 27, 2024
1 parent e8a4d58 commit 8409f2f
Show file tree
Hide file tree
Showing 5 changed files with 168 additions and 13 deletions.
5 changes: 5 additions & 0 deletions ggml/include/ggml-backend.h
View file
Original file line number Diff line number Diff line change
Expand Up @@ -232,6 +232,11 @@ extern "C" {
GGML_API void ggml_backend_tensor_alloc(ggml_backend_buffer_t buffer, struct ggml_tensor * tensor, void * addr);
GGML_API void ggml_backend_view_init(struct ggml_tensor * tensor);

// Utility to query whether cached GGML graph is in use
GGML_API bool ggml_use_cached_graph(ggml_backend_sched_t sched);

// Set whether or not to use GGML graph caching
GGML_API void ggml_set_cached_graph(ggml_backend_sched_t sched, bool set_value);

#ifdef __cplusplus
}
Expand Down
11 changes: 10 additions & 1 deletion ggml/include/ggml.h
View file
Original file line number Diff line number Diff line change
Expand Up @@ -568,6 +568,13 @@ extern "C" {
GGML_TENSOR_FLAG_PARAM = 4,
};

// Flag (used on GGML_OP_CPY nodes) on whether node is associated with K or V cache
enum ggml_kv_cache_flag {
GGML_KV_CACHE_FLAG_NONE = 0,
GGML_KV_CACHE_FLAG_K = 1,
GGML_KV_CACHE_FLAG_V = 2
};

// ggml object
struct ggml_object {
size_t offs;
Expand Down Expand Up @@ -602,6 +609,8 @@ extern "C" {
// op params - allocated as int32_t for alignment
int32_t op_params[GGML_MAX_OP_PARAMS / sizeof(int32_t)];

enum ggml_kv_cache_flag kv_cache_flag;

int32_t flags;

struct ggml_tensor * grad;
Expand All @@ -617,7 +626,7 @@ extern "C" {

void * extra; // extra things e.g. for ggml-cuda.cu

// char padding[4];
char padding[1];
};

static const size_t GGML_TENSOR_SIZE = sizeof(struct ggml_tensor);
Expand Down
31 changes: 30 additions & 1 deletion ggml/src/ggml-backend.c
View file
Original file line number Diff line number Diff line change
Expand Up @@ -1045,6 +1045,13 @@ struct ggml_backend_sched_split {
struct ggml_cgraph graph;
};

// Object to facilitate GML graph caching
struct ggml_cached_graph {
bool is_active;
ggml_backend_t input_backend;
struct ggml_tensor * input_cpy[GGML_SCHED_MAX_SPLIT_INPUTS];
};

struct ggml_backend_sched {
bool is_reset; // true if the scheduler has been reset since the last graph split
bool is_alloc;
Expand Down Expand Up @@ -1767,6 +1774,14 @@ static enum ggml_status ggml_backend_sched_compute_splits(ggml_backend_sched_t s
struct ggml_tensor * input = split->inputs[j];
struct ggml_tensor * input_cpy = tensor_copy(input, split_backend_id, sched->cur_copy);

if (!sched->cached_graph.is_active) {
sched->cached_graph.input_backend = input_backend;
sched->cached_graph.input_cpy[j] = input_cpy;
}
else {
input_backend = sched->cached_graph.input_backend;
input_cpy = sched->cached_graph.input_cpy[j];
}
if (input->flags & GGML_TENSOR_FLAG_INPUT) {
// inputs from the user must be copied immediately to prevent the user overwriting the data before the copy is done
if (sched->events[split_backend_id][sched->cur_copy] != NULL) {
Expand Down Expand Up @@ -1888,6 +1903,8 @@ ggml_backend_sched_t ggml_backend_sched_new(

ggml_backend_sched_reset(sched);

sched->cached_graph.is_active = false;

return sched;
}

Expand Down Expand Up @@ -1964,6 +1981,9 @@ enum ggml_status ggml_backend_sched_graph_compute(ggml_backend_sched_t sched, st
}

enum ggml_status ggml_backend_sched_graph_compute_async(ggml_backend_sched_t sched, struct ggml_cgraph * graph) {

if(!sched->cached_graph.is_active)
{
if (!sched->is_reset && !sched->is_alloc) {
ggml_backend_sched_reset(sched);
}
Expand All @@ -1973,7 +1993,7 @@ enum ggml_status ggml_backend_sched_graph_compute_async(ggml_backend_sched_t sch
return GGML_STATUS_ALLOC_FAILED;
}
}

}
return ggml_backend_sched_compute_splits(sched);
}

Expand Down Expand Up @@ -2238,3 +2258,12 @@ bool ggml_backend_compare_graph_backend(ggml_backend_t backend1, ggml_backend_t

return true;
}

bool ggml_use_cached_graph(ggml_backend_sched_t sched) {
return sched->cached_graph.is_active;
}

void ggml_set_cached_graph(ggml_backend_sched_t sched, bool set_value) {
sched->cached_graph.is_active = set_value;
}

3 changes: 2 additions & 1 deletion ggml/src/ggml.c
View file
Original file line number Diff line number Diff line change
Expand Up @@ -3746,6 +3746,7 @@ static struct ggml_tensor * ggml_new_tensor_impl(
/*.nb =*/ { 0, 0, 0, 0 },
/*.op =*/ GGML_OP_NONE,
/*.op_params =*/ { 0 },
/*.kv_cache_flag=*/ GGML_KV_CACHE_FLAG_NONE,
/*.flags =*/ 0,
/*.grad =*/ NULL,
/*.src =*/ { NULL },
Expand All @@ -3754,7 +3755,7 @@ static struct ggml_tensor * ggml_new_tensor_impl(
/*.data =*/ obj_alloc_size > 0 ? (void *)(result + 1) : data,
/*.name =*/ { 0 },
/*.extra =*/ NULL,
///*.padding =*/ { 0 },
/*.padding =*/ { 0 },
};

#ifdef __clang__
Expand Down
131 changes: 121 additions & 10 deletions src/llama.cpp
View file
Original file line number Diff line number Diff line change
Expand Up @@ -2648,6 +2648,17 @@ struct llama_model {
}
};

// Object used to allow caching of GGML graph between tokens where possible.
struct ggml_cached_graph {
bool is_active = false;
ggml_cgraph * gf;
size_t n;
ggml_backend_t backend_res;
ggml_backend_t backend_embd;
struct ggml_tensor * res;
struct ggml_tensor * embd;
};

struct llama_context {
llama_context(const llama_model & model)
: model(model)
Expand Down Expand Up @@ -2748,6 +2759,9 @@ struct llama_context {
struct ggml_tensor * inp_pos_bucket; // I32 [n_batch|n_kv, n_batch]
struct ggml_tensor * inp_embd_enc; // F32 [n_embd, n_outputs_enc]
struct ggml_tensor * inp_KQ_mask_cross; // F32 [n_outputs_enc, n_batch]

// cached Cuda Graphs
struct ggml_cached_graph cached_graph;
};

struct llama_lora_weight {
Expand Down Expand Up @@ -7900,7 +7914,9 @@ static void llm_build_kv_store(
cb(k_cache_view, "k_cache_view", il);

// note: storing RoPE-ed version of K in the KV cache
ggml_build_forward_expand(graph, ggml_cpy(ctx, k_cur, k_cache_view));
ggml_tensor * tmp = ggml_cpy(ctx, k_cur, k_cache_view);
tmp->kv_cache_flag = GGML_KV_CACHE_FLAG_K;
ggml_build_forward_expand(graph, tmp);

assert(v_cur->ne[0] == n_embd_v_gqa && v_cur->ne[1] == n_tokens);

Expand All @@ -7918,8 +7934,9 @@ static void llm_build_kv_store(
v_cur = ggml_transpose(ctx, v_cur);
}
cb(v_cache_view, "v_cache_view", il);

ggml_build_forward_expand(graph, ggml_cpy(ctx, v_cur, v_cache_view));
tmp=ggml_cpy(ctx, v_cur, v_cache_view);
tmp->kv_cache_flag = GGML_KV_CACHE_FLAG_V;
ggml_build_forward_expand(graph, tmp);
}

// do mat_mul, while optionally apply lora
Expand Down Expand Up @@ -14726,12 +14743,44 @@ static int llama_decode_internal(
ggml_backend_sched_reset(lctx.sched);
ggml_backend_sched_set_eval_callback(lctx.sched, lctx.cparams.cb_eval, lctx.cparams.cb_eval_user_data);

ggml_cgraph * gf = llama_build_graph(lctx, u_batch, false);

ggml_cgraph * gf;
// the output is always the last tensor in the graph
struct ggml_tensor * res = gf->nodes[gf->n_nodes - 1];
struct ggml_tensor * embd = gf->nodes[gf->n_nodes - 2];
struct ggml_tensor * res;
struct ggml_tensor * embd;

bool n_has_changed_since_last_token = false;
if(lctx.cached_graph.n != kv_self.n) n_has_changed_since_last_token = true;
lctx.cached_graph.n = kv_self.n;

// Re-build graph only if graph caching is not possible
if(!ggml_use_cached_graph(lctx.sched) || n_has_changed_since_last_token) {

gf = llama_build_graph(lctx, u_batch, false);

// Set whether GGML graph caching is in use within GGML module, based on
// whether caching was activated here during the previous token
ggml_set_cached_graph(lctx.sched,lctx.cached_graph.is_active);

// Disable future graph caching in presence of env var,
// if there are multiple devices, if batch size is greater than 1,
// or if nsplits is not 2.
// TO DO enable graph caching for these cases
bool disable_cached_ggml_graph = (getenv("GGML_DISABLE_GRAPH_CACHING") != nullptr)
|| (llama_get_device_count(model) > 1)
|| (ggml_backend_sched_get_n_splits(lctx.sched) != 2);
for (int i = 0 ; i < gf->n_nodes; i++) {
if (gf->nodes[i]->op == GGML_OP_ADD && gf->nodes[i]->src[1] && gf->nodes[i]->src[1]->ne[1] > 1) {
disable_cached_ggml_graph = true;
break;
}
}

// Set whether graph caching should be used for future tokens
lctx.cached_graph.is_active=!disable_cached_ggml_graph;

// the output is always the last tensor in the graph
res = gf->nodes[gf->n_nodes - 1];
embd = gf->nodes[gf->n_nodes - 2];
if (lctx.n_outputs == 0) {
// no output
res = nullptr;
Expand All @@ -14747,10 +14796,62 @@ static int llama_decode_internal(
embd = nullptr; // do not extract embeddings when not needed
GGML_ASSERT(strcmp(res->name, "result_output") == 0 && "missing result_output tensor");
}
lctx.cached_graph.res = res;
lctx.cached_graph.embd = embd;
// LLAMA_LOG_INFO("graph build time: %.3f ms (%d nodes, %d leafs)\n", (ggml_time_us() - t_start_us)/1000.0, gf->n_nodes, gf->n_leafs);

ggml_backend_sched_alloc_graph(lctx.sched, gf);

}
else {
gf = lctx.cached_graph.gf;
res = lctx.cached_graph.res;
embd = lctx.cached_graph.embd;
}
lctx.cached_graph.gf = gf;

if(ggml_use_cached_graph(lctx.sched)) {

// Temporarily store KV cache parameters that will need updated in cached graph.
const struct llama_hparams & hparams = model.hparams;
const int64_t n_layer = hparams.n_layer;
const int64_t kv_head = kv_self.head;
std::vector<void *> k_cache_ptrs;
std::vector<void *> v_cache_ptrs;
for (int il = 0; il < n_layer; ++il) {
const int64_t n_embd_k_gqa = hparams.n_embd_k_gqa(il);
const int64_t n_embd_v_gqa = hparams.n_embd_v_gqa(il);
ggml_tensor * tmp_tensor = kv_self.k_l[il];
size_t tmp_offset = (ggml_row_size(kv_self.k_l[il]->type, n_embd_k_gqa))*kv_head;
k_cache_ptrs.push_back(static_cast<char*>(tmp_tensor->data) + tmp_offset);
tmp_tensor = kv_self.v_l[il];
if (cparams.flash_attn) {
tmp_offset = (kv_head)*ggml_row_size(kv_self.v_l[il]->type, n_embd_v_gqa);
} else {
tmp_offset = (kv_head)*ggml_element_size(kv_self.v_l[il]);
}
v_cache_ptrs.push_back(static_cast<char*>(tmp_tensor->data) + tmp_offset);
}

// Update KV cache parameters in cached graph.
int k_count = 0;
int v_count = 0;
if(gf != nullptr && gf->nodes != nullptr){
for (int i = 0; i < gf->n_nodes; i++) {
ggml_tensor * node = gf->nodes[i];
if (node->op == GGML_OP_CPY) {
if (node->kv_cache_flag == GGML_KV_CACHE_FLAG_K) {
node->src[1]->data = k_cache_ptrs[k_count++];
}
if (node->kv_cache_flag == GGML_KV_CACHE_FLAG_V) {
node->src[1]->data = v_cache_ptrs[v_count++];
}
}
}
}

}

llama_set_inputs(lctx, u_batch);

llama_graph_compute(lctx, gf, n_threads);
Expand All @@ -14773,11 +14874,15 @@ static int llama_decode_internal(
// extract logits
if (res) {
ggml_backend_t backend_res = ggml_backend_sched_get_tensor_backend(lctx.sched, res);
GGML_ASSERT(backend_res != nullptr);
GGML_ASSERT(lctx.logits != nullptr);

float * logits_out = lctx.logits + n_outputs_prev*n_vocab;
const int32_t n_outputs_new = lctx.n_outputs;
if(!ggml_use_cached_graph(lctx.sched))
lctx.cached_graph.backend_res = backend_res;
else
backend_res = lctx.cached_graph.backend_res;

GGML_ASSERT(backend_res != nullptr);
GGML_ASSERT(lctx.logits != nullptr);

if (n_outputs_new) {
GGML_ASSERT( n_outputs_prev + n_outputs_new <= n_outputs);
Expand All @@ -14789,6 +14894,12 @@ static int llama_decode_internal(
// extract embeddings
if (embd) {
ggml_backend_t backend_embd = ggml_backend_sched_get_tensor_backend(lctx.sched, embd);


if(!ggml_use_cached_graph(lctx.sched))
lctx.cached_graph.backend_embd = backend_embd;
else
backend_embd = lctx.cached_graph.backend_embd;
GGML_ASSERT(backend_embd != nullptr);

switch (cparams.pooling_type) {
Expand Down

0 comments on commit 8409f2f

Please sign in to comment.