llama: update llama.cpp vendor code to commit d7cfe1ff (#9356)

This commit is contained in:
Jeffrey Morgan
2025-02-26 20:34:44 -08:00
committed by GitHub
parent 2db96c18e7
commit d7d7e99662
149 changed files with 18215 additions and 11009 deletions

View File

@@ -11,10 +11,10 @@ the characters
2 files changed, 23 insertions(+), 1 deletion(-)
diff --git a/src/llama-vocab.cpp b/src/llama-vocab.cpp
index 3fcfcaa3..8f44705a 100644
index a4eee9b8..1ca827eb 100644
--- a/src/llama-vocab.cpp
+++ b/src/llama-vocab.cpp
@@ -375,7 +375,7 @@ struct llm_tokenizer_bpe : llm_tokenizer {
@@ -295,7 +295,7 @@ struct llm_tokenizer_bpe : llm_tokenizer {
case LLAMA_VOCAB_PRE_TYPE_DEEPSEEK_LLM:
regex_exprs = {
"[\r\n]",
@@ -24,7 +24,7 @@ index 3fcfcaa3..8f44705a 100644
"\\s+$",
"[一-龥ࠀ-一가-퟿]+",
diff --git a/src/unicode.cpp b/src/unicode.cpp
index 7aca6544..6155da80 100644
index e63bb4ab..9dd53b9a 100644
--- a/src/unicode.cpp
+++ b/src/unicode.cpp
@@ -2,6 +2,11 @@
@@ -39,7 +39,7 @@ index 7aca6544..6155da80 100644
#include "unicode.h"
#include "unicode-data.h"
@@ -201,6 +206,22 @@ static std::unordered_map<std::string, uint8_t> unicode_utf8_to_byte_map() {
@@ -200,6 +205,22 @@ static std::unordered_map<std::string, uint8_t> unicode_utf8_to_byte_map() {
}
static inline std::wstring unicode_wstring_from_utf8(const std::string & s) {
@@ -62,7 +62,7 @@ index 7aca6544..6155da80 100644
#if defined(__clang__)
// disable C++17 deprecation warning for std::codecvt_utf8
# pragma clang diagnostic push
@@ -214,6 +235,7 @@ static inline std::wstring unicode_wstring_from_utf8(const std::string & s) {
@@ -213,6 +234,7 @@ static inline std::wstring unicode_wstring_from_utf8(const std::string & s) {
#endif
return conv.from_bytes(s);