mirror of
https://github.com/dogkeeper886/ollama37.git
synced 2025-12-09 23:37:06 +00:00
ef14fb5b26
This commit represents a complete rework after pulling the latest changes from official ollama/ollama repository and re-applying Tesla K80 compatibility patches. ## Key Changes ### CUDA Compute Capability 3.7 Support (Tesla K80) - Added sm_37 (compute 3.7) to CMAKE_CUDA_ARCHITECTURES in CMakeLists.txt - Updated CMakePresets.json to include compute 3.7 in "CUDA 11" preset - Using 37-virtual (PTX with JIT compilation) for maximum compatibility ### Legacy Toolchain Compatibility - **NVIDIA Driver**: 470.256.02 (last version supporting Kepler/K80) - **CUDA Version**: 11.4.4 (last CUDA 11.x supporting compute 3.7) - **GCC Version**: 10.5.0 (required by CUDA 11.4 host_config.h) ### CPU Architecture Trade-offs Due to GCC 10.5 limitation, sacrificed newer CPU optimizations: - Alderlake CPU variant enabled WITHOUT AVX_VNNI (requires GCC 11+) - Still supports: SSE4.2, AVX, F16C, AVX2, BMI2, FMA - Performance impact: ~3-7% on newer CPUs (acceptable for K80 compatibility) ### Build System Updates - Modified ml/backend/ggml/ggml/src/ggml-cuda/CMakeLists.txt for compute 3.7 - Added -Wno-deprecated-gpu-targets flag to suppress warnings - Updated ml/backend/ggml/ggml/src/CMakeLists.txt for Alderlake without AVX_VNNI ### Upstream Sync Merged latest llama.cpp changes including: - Enhanced KV cache management with ISWA and hybrid memory support - Improved multi-modal support (mtmd framework) - New model architectures (Gemma3, Llama4, Qwen3, etc.) - GPU backend improvements for CUDA, Metal, and ROCm - Updated quantization support and GGUF format handling ### Documentation - Updated CLAUDE.md with comprehensive build instructions - Documented toolchain constraints and CPU architecture trade-offs - Removed outdated CI/CD workflows (tesla-k80-*.yml) - Cleaned up temporary development artifacts ## Rationale This fork maintains Tesla K80 GPU support (compute 3.7) which was dropped in official Ollama due to legacy driver/CUDA requirements. The toolchain constraint creates a deadlock: - K80 → Driver 470 → CUDA 11.4 → GCC 10 → No AVX_VNNI We accept the loss of cutting-edge CPU optimizations to enable running modern LLMs on legacy but still capable Tesla K80 hardware (12GB VRAM per GPU). 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>
73 lines
3.3 KiB
Diff
73 lines
3.3 KiB
Diff
From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
|
||
From: jmorganca <jmorganca@gmail.com>
|
||
Date: Tue, 8 Apr 2025 19:43:06 -0700
|
||
Subject: [PATCH] fix deepseek deseret regex
|
||
|
||
on some systems, deepseek's regex would throw an error
|
||
on windows due to the deseret characters in the matching
|
||
regex
|
||
---
|
||
src/llama-vocab.cpp | 2 +-
|
||
src/unicode.cpp | 21 +++++++++++++++++++++
|
||
2 files changed, 22 insertions(+), 1 deletion(-)
|
||
|
||
diff --git a/src/llama-vocab.cpp b/src/llama-vocab.cpp
|
||
index 0b6edaf4..3de95c67 100644
|
||
--- a/src/llama-vocab.cpp
|
||
+++ b/src/llama-vocab.cpp
|
||
@@ -299,7 +299,7 @@ struct llm_tokenizer_bpe : llm_tokenizer {
|
||
case LLAMA_VOCAB_PRE_TYPE_DEEPSEEK_LLM:
|
||
regex_exprs = {
|
||
"[\r\n]",
|
||
- "\\s?[A-Za-zµÀ-ÖØ-öø-ƺƼ-ƿDŽ-ʓʕ-ʯͰ-ͳͶͷͻ-ͽͿΆΈ-ΊΌΎ-ΡΣ-ϵϷ-ҁҊ-ԯԱ-ՖႠ-ჅᎠ-Ᏽᏸ-ᏽᲐ-ᲺᲽ-Ჿᴀ-ᴫᵫ-ᵷᵹ-ᶚḀ-ἕἘ-Ἕἠ-ὅὈ-Ὅὐ-ὗὙὛὝὟ-ώᾀ-ᾴᾶ-ᾼιῂ-ῄῆ-ῌῐ-ΐῖ-Ίῠ-Ῥῲ-ῴῶ-ῼℂℇℊ-ℓℕℙ-ℝℤΩℨK-ℭℯ-ℴℹℼ-ℿⅅ-ⅉⅎↃↄⰀ-ⱻⱾ-ⳤⳫ-ⳮⳲⳳꙀ-ꙭꚀ-ꚛꜢ-ꝯꝱ-ꞇꞋ-ꞎꭰ-ꮿff-stﬓ-ﬗＡ-Ｚａ-ｚ𐐀-𐑏𐒰-𐓓𐓘-𐓻𐲀-𐲲𐳀-𐳲𑢠-𑣟𞤀-𞥃]+",
|
||
+ "\\s?[A-Za-zµÀ-ÖØ-öø-ƺƼ-ƿDŽ-ʓʕ-ʯͰ-ͳͶͷͻ-ͽͿΆΈ-ΊΌΎ-ΡΣ-ϵϷ-ҁҊ-ԯԱ-ՖႠ-ჅᎠ-Ᏽᏸ-ᏽᲐ-ᲺᲽ-Ჿᴀ-ᴫᵫ-ᵷᵹ-ᶚḀ-ἕἘ-Ἕἠ-ὅὈ-Ὅὐ-ὗὙὛὝὟ-ώᾀ-ᾴᾶ-ᾼιῂ-ῄῆ-ῌῐ-ΐῖ-Ίῠ-Ῥῲ-ῴῶ-ῼℂℇℊ-ℓℕℙ-ℝℤΩℨK-ℭℯ-ℴℹℼ-ℿⅅ-ⅉⅎↃↄⰀ-ⱻⱾ-ⳤⳫ-ⳮⳲⳳꙀ-ꙭꚀ-ꚛꜢ-ꝯꝱ-ꞇꞋ-ꞎꭰ-ꮿff-stﬓ-ﬗＡ-Ｚａ-ｚ\U00010400-\U0001044f𐒰-𐓓𐓘-𐓻𐲀-𐲲𐳀-𐳲𑢠-𑣟𞤀-𞥃]+",
|
||
"\\s?[!-/:-~！-／：-～‘-‟ -。]+",
|
||
"\\s+$",
|
||
"[一-龥ࠀ-一가-]+",
|
||
diff --git a/src/unicode.cpp b/src/unicode.cpp
|
||
index 65f36651..ce336a22 100644
|
||
--- a/src/unicode.cpp
|
||
+++ b/src/unicode.cpp
|
||
@@ -2,6 +2,11 @@
|
||
#define _SILENCE_CXX17_CODECVT_HEADER_DEPRECATION_WARNING
|
||
#endif
|
||
|
||
+#if defined(_WIN32)
|
||
+#define WIN32_LEAN_AND_MEAN
|
||
+#include <windows.h>
|
||
+#endif
|
||
+
|
||
#include "unicode.h"
|
||
#include "unicode-data.h"
|
||
|
||
@@ -200,6 +205,21 @@ 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) {
|
||
+#ifdef _WIN32
|
||
+ int wlen = MultiByteToWideChar(CP_UTF8, 0, s.c_str(), -1, NULL, 0);
|
||
+ if (!wlen) {
|
||
+ throw std::invalid_argument("failed to convert regex");
|
||
+ }
|
||
+ wchar_t * wbuf = (wchar_t *) malloc(wlen * sizeof(wchar_t));
|
||
+ wlen = MultiByteToWideChar(CP_UTF8, 0, s.c_str(), -1, wbuf, wlen);
|
||
+ if (!wlen) {
|
||
+ free(wbuf);
|
||
+ throw std::invalid_argument("failed to convert regex");
|
||
+ }
|
||
+ std::wstring ret = std::wstring(wbuf);
|
||
+ free(wbuf);
|
||
+ return ret;
|
||
+#else
|
||
#if defined(__clang__)
|
||
// disable C++17 deprecation warning for std::codecvt_utf8
|
||
# pragma clang diagnostic push
|
||
@@ -218,6 +238,7 @@ static inline std::wstring unicode_wstring_from_utf8(const std::string & s) {
|
||
#endif
|
||
|
||
return conv.from_bytes(s);
|
||
+#endif
|
||
}
|
||
|
||
static std::vector<std::string> unicode_byte_encoding_process(const std::vector<std::string> & bpe_words) {
|