mirror of
https://github.com/ollama/ollama.git
synced 2026-06-05 21:05:00 +08:00
4b5cf3420a
Previous addition for glm-ocr added 4 lines to upstream-edits.patch
(1 in constructor for set_loader_path + 3 in load_all_data for the
maybe_load_text_tensor if/continue/closing-brace block). Refactor to
+1 line:
* Fold path capture into translate_metadata: signature now takes
`const char * fname` and stashes it on the per-loader registry
internally. The constructor still has just one llama_compat call
(we just changed the argument list).
* Make maybe_load_text_tensor self-contained: it derives the buffer
type from `cur->buffer` rather than the caller passing it. The
hook line in load_all_data collapses to a single
`if (llama_ollama_compat::maybe_load_text_tensor(this, cur, weight->offs)) continue;`.
* Drop the public `set_loader_path` symbol — it's now an internal
detail of translate_metadata.
Net patch growth from glm-ocr support: +1 line. Total upstream patch
is back to its original 17-line surface.
Functional: glm-ocr text generation still works ("Paris" via raw
turn-template completion), concat ops still fire (28MB per block).
75 lines
3.3 KiB
C++
Vendored
75 lines
3.3 KiB
C++
Vendored
#pragma once
|
|
|
|
// Ollama-format GGUF compatibility shim.
|
|
//
|
|
// Older Ollama builds ship GGUFs that differ from upstream in a handful of
|
|
// ways per-architecture (arch names, KV keys, tensor names, file layout).
|
|
// This shim detects those files during load and translates them in-memory
|
|
// so the rest of llama.cpp can load them unmodified.
|
|
//
|
|
// Three upstream hook points call into this namespace — one per insertion:
|
|
//
|
|
// 1. llama-model-loader.cpp (main model load):
|
|
// translate_metadata() — mutate KVs / tensor metadata
|
|
// should_skip_tensor() — filter weights_map population
|
|
//
|
|
// 2. tools/mtmd/clip.cpp (mmproj load):
|
|
// translate_clip_metadata() — rewrite KVs + tensor names for clip
|
|
// maybe_load_tensor() — override file read (e.g. F16->F32)
|
|
//
|
|
// Detection is per-arch; for any non-Ollama file every entry point is a
|
|
// no-op. Per-arch logic lives in anonymous-namespace handle_<arch>()
|
|
// functions in the .cpp; adding a new arch is a new handler plus one
|
|
// dispatch line in each translate_* entry point.
|
|
|
|
#include <cstddef>
|
|
#include <string>
|
|
|
|
#include "ggml-backend.h" // for ggml_backend_buffer_type_t
|
|
|
|
struct gguf_context;
|
|
struct ggml_context;
|
|
struct ggml_tensor;
|
|
struct llama_model_loader;
|
|
|
|
namespace llama_ollama_compat {
|
|
|
|
// Called from llama_model_loader's constructor, right after the arch is read.
|
|
// `fname` is the model file path, captured here so later load-time hooks
|
|
// (maybe_load_text_tensor) can read raw bytes from it.
|
|
void translate_metadata(const llama_model_loader * ml,
|
|
gguf_context * meta,
|
|
ggml_context * ctx,
|
|
std::string & arch_name,
|
|
const char * fname);
|
|
|
|
// Called from llama_model_loader's weights_map population loop. Returns
|
|
// true to drop a tensor from the loader — used to hide embedded vision
|
|
// tensors from the text model's view without modifying the gguf_context.
|
|
bool should_skip_tensor(const llama_model_loader * ml, const char * tensor_name);
|
|
|
|
// Called from clip_model_loader's constructor. Rewrites the clip-facing
|
|
// view of the metadata (arch=clip, clip.vision.* KVs, renamed tensors)
|
|
// so the rest of clip.cpp can load an Ollama monolithic GGUF unchanged.
|
|
void translate_clip_metadata(gguf_context * meta, ggml_context * ctx);
|
|
|
|
// Called from clip.cpp's tensor-loading loop, before the normal file read.
|
|
// If this tensor was marked for type promotion by translate_clip_metadata
|
|
// (e.g. F16->F32), performs the conversion and writes the result into
|
|
// `cur` (host memcpy or backend_tensor_set based on `buft`). Returns true
|
|
// when the tensor was handled — caller should skip its normal read path.
|
|
bool maybe_load_tensor(ggml_tensor * cur,
|
|
const char * source_file,
|
|
size_t file_offset,
|
|
ggml_backend_buffer_type_t buft);
|
|
|
|
// Text-side counterpart to maybe_load_tensor. Self-contained: looks up
|
|
// the model file path from the per-loader registry populated by
|
|
// translate_metadata, and derives the buffer type from cur->buffer
|
|
// internally — keeps the call site (and the upstream patch) to one line.
|
|
bool maybe_load_text_tensor(const llama_model_loader * ml,
|
|
ggml_tensor * cur,
|
|
size_t file_offset);
|
|
|
|
} // namespace llama_ollama_compat
|