upstream-archive/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2026-09-30 18:07:38 -05:00
Code Issues Packages Projects Releases Wiki Activity

ggml : speed up model loading (#29598)

Browse Source 
* ggml : speed up model loading

A crafted model could hang the server for a very long time, try with:

    llama-cli -hf angt/test-gguf-1Mkv -hff model.gguf

Signed-off-by: Adrien Gallouët <angt@huggingface.co>

* Avoid empty keys

Signed-off-by: Adrien Gallouët <angt@huggingface.co>

* Fix

Co-authored-by: Johannes Gäßler <johannesg@5d6.de>

---------

Signed-off-by: Adrien Gallouët <angt@huggingface.co>
Co-authored-by: Johannes Gäßler <johannesg@5d6.de>
This commit is contained in:
Adrien Gallouët
2026-09-29 10:35:08 +02:00
committed by GitHub
co-authored by Johannes Gäßler
parent c8cda8b4fe
commit c13e04e1dd
1 changed files with 12 additions and 11 deletions
Download Patch File Download Diff File Expand all files Collapse all files
ggml/src/gguf.cpp
+12 -11
View File
@@ -14,6 +14,7 @@
#include <new>
#include <stdexcept>
#include <string>
#include <unordered_set>
#include <vector>
#define GGUF_MAX_STRING_LENGTH (1024*1024*1024)
@@ -550,6 +551,8 @@ static struct gguf_context * gguf_init_from_reader(const struct gguf_reader & gr
// KV pairs
{
std::unordered_set<std::string> seen_keys;
for (int64_t i = 0; ok && i < n_kv; ++i) {
std::string key;
gguf_type type = gguf_type(-1);
@@ -569,11 +572,9 @@ static struct gguf_context * gguf_init_from_reader(const struct gguf_reader & gr
GGML_LOG_ERROR("%s: key %" PRIi64 " is empty\n", __func__, i);
ok = false;
}
for (size_t j = 0; ok && j < ctx->kv.size(); ++j) {
if (key == ctx->kv[j].key) {
GGML_LOG_ERROR("%s: duplicate key '%s' for tensors %zu and %" PRIi64 " \n", __func__, key.c_str(), j, i);
ok = false;
}
if (ok && !seen_keys.insert(key).second) {
GGML_LOG_ERROR("%s: duplicate key '%s' for KV pair %" PRIi64 "\n", __func__, key.c_str(), i);
ok = false;
}
if (!ok) {
break;
@@ -636,6 +637,8 @@ static struct gguf_context * gguf_init_from_reader(const struct gguf_reader & gr
}
// read the tensor info
std::unordered_set<std::string> seen_tensor_names;
for (int64_t i = 0; ok && i < n_tensors; ++i) {
struct gguf_tensor_info info;
@@ -659,12 +662,10 @@ static struct gguf_context * gguf_init_from_reader(const struct gguf_reader & gr
ggml_set_name(&info.t, name.c_str());
// make sure there are no duplicate tensor names
for (int64_t j = 0; ok && j < i; ++j) {
if (strcmp(info.t.name, ctx->info[j].t.name) == 0) {
GGML_LOG_ERROR("%s: duplicate tensor name '%s' for tensors %" PRIi64 " and %" PRIi64 "\n", __func__, info.t.name, j, i);
ok = false;
break;
}
if (ok && !seen_tensor_names.insert(name).second) {
GGML_LOG_ERROR("%s: duplicate tensor name '%s' for tensor %" PRIi64 "\n", __func__, info.t.name, i);
ok = false;
break;
}
}
if (!ok) {