highperfocused/ollama
1
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2025-11-11 11:47:29 +01:00
Code Issues Packages Projects Releases Wiki Activity

Update GGML to b6646 (#12245)

Browse Source 
Notable EOLs with this change:
- MacOS v12 and v13 are no longer supported (v14+ required)
- AMD gfx900 and gfx906 are no longer supported
This commit is contained in:
Daniel Hiltgen
2025-10-02 14:47:10 -07:00
committed by GitHub
parent fdb109469f
commit c68f367ef6
326 changed files with 30615 additions and 20624 deletions
Download Patch File Download Diff File Expand all files Collapse all files

199
llama/llama.cpp/include/llama.h vendored
View File

@@ -64,8 +64,6 @@ extern "C" {
typedef struct llama_memory_i * llama_memory_t;
struct llama_kv_cache; // DEPRECATED (use llama_memory instead)
typedef int32_t llama_pos;
typedef int32_t llama_token;
typedef int32_t llama_seq_id;
@@ -181,6 +179,14 @@ extern "C" {
LLAMA_ATTENTION_TYPE_NON_CAUSAL = 1,
};
enum llama_flash_attn_type {
LLAMA_FLASH_ATTN_TYPE_AUTO = -1,
LLAMA_FLASH_ATTN_TYPE_DISABLED = 0,
LLAMA_FLASH_ATTN_TYPE_ENABLED = 1,
};
LLAMA_API const char * llama_flash_attn_type_name(enum llama_flash_attn_type flash_attn_type);
enum llama_split_mode {
LLAMA_SPLIT_MODE_NONE = 0, // single GPU
LLAMA_SPLIT_MODE_LAYER = 1, // split layers and KV across GPUs
@@ -200,7 +206,7 @@ extern "C" {
llama_token_data * data;
size_t size;
int64_t selected; // this is the index in the data array (i.e. not the token id)
bool sorted;
bool sorted; // note: do not assume the data is sorted - always check this flag
} llama_token_data_array;
typedef bool (*llama_progress_callback)(float progress, void * user_data);
@@ -305,6 +311,7 @@ extern "C" {
enum llama_rope_scaling_type rope_scaling_type; // RoPE scaling type, from `enum llama_rope_scaling_type`
enum llama_pooling_type pooling_type; // whether to pool (sum) embedding results by sequence id
enum llama_attention_type attention_type; // attention type to use for embeddings
enum llama_flash_attn_type flash_attn_type; // when to enable Flash Attention
// ref: https://github.com/ggml-org/llama.cpp/pull/2054
float rope_freq_base; // RoPE base frequency, 0 = from model
@@ -314,7 +321,7 @@ extern "C" {
float yarn_beta_fast; // YaRN low correction dim
float yarn_beta_slow; // YaRN high correction dim
uint32_t yarn_orig_ctx; // YaRN original context size
float defrag_thold; // defragment the KV cache if holes/size > thold, <= 0 disabled (default)
float defrag_thold; // [DEPRECATED] defragment the KV cache if holes/size > thold, <= 0 disabled (default)
ggml_backend_sched_eval_callback cb_eval;
void * cb_eval_user_data;
@@ -331,7 +338,6 @@ extern "C" {
// Keep the booleans together and at the end of the struct to avoid misalignment during copy-by-value.
bool embeddings; // if true, extract embeddings (together with logits)
bool offload_kqv; // offload the KQV ops (including the KV cache) to GPU
bool flash_attn; // use flash attention [EXPERIMENTAL]
bool no_perf; // measure performance timings
bool op_offload; // offload host tensor operations to device
bool swa_full; // use full-size SWA cache (https://github.com/ggml-org/llama.cpp/pull/13194#issuecomment-2868343055)
@@ -469,8 +475,6 @@ extern "C" {
LLAMA_API llama_memory_t llama_get_memory (const struct llama_context * ctx);
LLAMA_API enum llama_pooling_type llama_pooling_type(const struct llama_context * ctx); // TODO: rename to llama_get_pooling_type
DEPRECATED(LLAMA_API struct llama_kv_cache * llama_get_kv_self(struct llama_context * ctx), "use llama_get_memory instead");
LLAMA_API const struct llama_vocab * llama_model_get_vocab(const struct llama_model * model);
LLAMA_API enum llama_rope_type llama_model_rope_type(const struct llama_model * model);
@@ -557,10 +561,32 @@ extern "C" {
struct llama_model * model,
const char * path_lora);
// Functions to access the adapter's GGUF metadata scalar values
// - The functions return the length of the string on success, or -1 on failure
// - The output string is always null-terminated and cleared on failure
// - When retrieving a string, an extra byte must be allocated to account for the null terminator
// - GGUF array values are not supported by these functions
// Get metadata value as a string by key name
LLAMA_API int32_t llama_adapter_meta_val_str(const struct llama_adapter_lora * adapter, const char * key, char * buf, size_t buf_size);
// Get the number of metadata key/value pairs
LLAMA_API int32_t llama_adapter_meta_count(const struct llama_adapter_lora * adapter);
// Get metadata key name by index
LLAMA_API int32_t llama_adapter_meta_key_by_index(const struct llama_adapter_lora * adapter, int32_t i, char * buf, size_t buf_size);
// Get metadata value as a string by index
LLAMA_API int32_t llama_adapter_meta_val_str_by_index(const struct llama_adapter_lora * adapter, int32_t i, char * buf, size_t buf_size);
// Manually free a LoRA adapter
// Note: loaded adapters will be free when the associated model is deleted
LLAMA_API void llama_adapter_lora_free(struct llama_adapter_lora * adapter);
// Get the invocation tokens if the current lora is an alora
LLAMA_API uint64_t llama_adapter_get_alora_n_invocation_tokens(const struct llama_adapter_lora * adapter);
LLAMA_API const llama_token * llama_adapter_get_alora_invocation_tokens (const struct llama_adapter_lora * adapter);
// The following functions operate on a llama_context, hence the naming: llama_verb_...
// Add a loaded LoRA adapter to given context
@@ -667,111 +693,6 @@ extern "C" {
// Check if the memory supports shifting
LLAMA_API bool llama_memory_can_shift(llama_memory_t mem);
//
// KV cache for self-attention (TODO: deprecate in favor of llama_memory)
//
// Returns the number of tokens in the KV cache (slow, use only for debug)
// If a KV cell has multiple sequences assigned to it, it will be counted multiple times
DEPRECATED(LLAMA_API int32_t llama_kv_self_n_tokens(const struct llama_context * ctx),
"Use llama_kv_self_seq_pos_max() and llama_kv_self_seq_pos_min() instead (https://github.com/ggml-org/llama.cpp/issues/13793)");
// Returns the number of used KV cells (i.e. have at least one sequence assigned to them)
DEPRECATED(LLAMA_API int32_t llama_kv_self_used_cells(const struct llama_context * ctx),
"Use llama_kv_self_seq_pos_max() and llama_kv_self_seq_pos_min() instead (https://github.com/ggml-org/llama.cpp/issues/13793)");
// Clear the KV cache - both cell info is erased and KV data is zeroed
DEPRECATED(LLAMA_API void llama_kv_self_clear(
struct llama_context * ctx),
"Use llama_memory_clear() instead");
// Removes all tokens that belong to the specified sequence and have positions in [p0, p1)
// Returns false if a partial sequence cannot be removed. Removing a whole sequence never fails
// seq_id < 0 : match any sequence
// p0 < 0 : [0, p1]
// p1 < 0 : [p0, inf)
DEPRECATED(LLAMA_API bool llama_kv_self_seq_rm(
struct llama_context * ctx,
llama_seq_id seq_id,
llama_pos p0,
llama_pos p1),
"Use llama_memory_seq_rm() instead");
// Copy all tokens that belong to the specified sequence to another sequence
// Note that this does not allocate extra KV cache memory - it simply assigns the tokens to the new sequence
// p0 < 0 : [0, p1]
// p1 < 0 : [p0, inf)
DEPRECATED(LLAMA_API void llama_kv_self_seq_cp(
struct llama_context * ctx,
llama_seq_id seq_id_src,
llama_seq_id seq_id_dst,
llama_pos p0,
llama_pos p1),
"Use llama_memory_seq_cp() instead");
// Removes all tokens that do not belong to the specified sequence
DEPRECATED(LLAMA_API void llama_kv_self_seq_keep(
struct llama_context * ctx,
llama_seq_id seq_id),
"Use llama_memory_seq_keep() instead");
// Adds relative position "delta" to all tokens that belong to the specified sequence and have positions in [p0, p1)
// If the KV cache is RoPEd, the KV data is updated accordingly:
// - lazily on next llama_decode()
// p0 < 0 : [0, p1]
// p1 < 0 : [p0, inf)
DEPRECATED(LLAMA_API void llama_kv_self_seq_add(
struct llama_context * ctx,
llama_seq_id seq_id,
llama_pos p0,
llama_pos p1,
llama_pos delta),
"Use llama_memory_seq_add() instead");
// Integer division of the positions by factor of `d > 1`
// If the KV cache is RoPEd, the KV data is updated accordingly:
// - lazily on next llama_decode()
// p0 < 0 : [0, p1]
// p1 < 0 : [p0, inf)
DEPRECATED(LLAMA_API void llama_kv_self_seq_div(
struct llama_context * ctx,
llama_seq_id seq_id,
llama_pos p0,
llama_pos p1,
int d),
"Use llama_memory_seq_div() instead");
// Returns the smallest position present in the KV cache for the specified sequence
// This is typically non-zero only for SWA caches
// Note that all positions in the range [pos_min, pos_max] are guaranteed to be present in the KV cache
// Return -1 if the sequence is empty
DEPRECATED(LLAMA_API llama_pos llama_kv_self_seq_pos_min(
struct llama_context * ctx,
llama_seq_id seq_id),
"Use llama_memory_seq_pos_min() instead");
// Returns the largest position present in the KV cache for the specified sequence
// Note that all positions in the range [pos_min, pos_max] are guaranteed to be present in the KV cache
// Return -1 if the sequence is empty
DEPRECATED(LLAMA_API llama_pos llama_kv_self_seq_pos_max(
struct llama_context * ctx,
llama_seq_id seq_id),
"Use llama_memory_seq_pos_max() instead");
// Defragment the KV cache
// This will be applied:
// - lazily on next llama_decode()
DEPRECATED(LLAMA_API void llama_kv_self_defrag(struct llama_context * ctx),
"simply remove this call, the context will automatically decide when to do a defragmentation based on 'defrag_thold'");
// Check if the context supports KV cache shifting
DEPRECATED(LLAMA_API bool llama_kv_self_can_shift(const struct llama_context * ctx),
"use llama_memory_can_shift() instead");
// Apply the KV cache updates (such as K-shifts, defragmentation, etc.)
DEPRECATED(LLAMA_API void llama_kv_self_update(struct llama_context * ctx),
"simply remove this call, updates are applied lazily on the next llama_decode()");
//
// State / sessions
//
@@ -870,6 +791,29 @@ extern "C" {
size_t n_token_capacity,
size_t * n_token_count_out);
#define LLAMA_STATE_SEQ_FLAGS_SWA_ONLY 1
typedef uint32_t llama_state_seq_flags;
LLAMA_API size_t llama_state_seq_get_size_ext(
struct llama_context * ctx,
llama_seq_id seq_id,
llama_state_seq_flags flags);
LLAMA_API size_t llama_state_seq_get_data_ext(
struct llama_context * ctx,
uint8_t * dst,
size_t size,
llama_seq_id seq_id,
llama_state_seq_flags flags);
LLAMA_API size_t llama_state_seq_set_data_ext(
struct llama_context * ctx,
const uint8_t * src,
size_t size,
llama_seq_id dest_seq_id,
llama_state_seq_flags flags);
//
// Decoding
//
@@ -1216,11 +1160,6 @@ extern "C" {
LLAMA_API struct llama_sampler * llama_sampler_init_greedy(void);
LLAMA_API struct llama_sampler * llama_sampler_init_dist (uint32_t seed);
/// @details Sorts candidate tokens by their logits in descending order and calculate probabilities based on logits.
/// NOTE: Avoid using on the full vocabulary as the sorting can become slow. For example, apply top-k or top-p sampling first.
DEPRECATED(LLAMA_API struct llama_sampler * llama_sampler_init_softmax (void),
"will be removed in the future (see https://github.com/ggml-org/llama.cpp/pull/9896#discussion_r1800920915)");
/// @details Top-K sampling described in academic paper "The Curious Case of Neural Text Degeneration" https://arxiv.org/abs/1904.09751
/// Setting k <= 0 makes this a noop
LLAMA_API struct llama_sampler * llama_sampler_init_top_k (int32_t k);
@@ -1390,24 +1329,25 @@ extern "C" {
//
// Performance utils
//
// NOTE: Used by llama.cpp examples, avoid using in third-party apps. Instead, do your own performance measurements.
// NOTE: Used by llama.cpp examples/tools, avoid using in third-party apps. Instead, do your own performance measurements.
//
struct llama_perf_context_data {
double t_start_ms;
double t_load_ms;
double t_p_eval_ms;
double t_eval_ms;
// ms == milliseconds
double t_start_ms; // absolute start time
double t_load_ms; // time needed for loading the model
double t_p_eval_ms; // time needed for processing the prompt
double t_eval_ms; // time needed for generating tokens
int32_t n_p_eval;
int32_t n_eval;
int32_t n_reused; // number of times a ggml compute graph had been reused
int32_t n_p_eval; // number of prompt tokens
int32_t n_eval; // number of generated tokens
int32_t n_reused; // number of times a ggml compute graph had been reused
};
struct llama_perf_sampler_data {
double t_sample_ms;
double t_sample_ms; // time needed for sampling in ms
int32_t n_sample;
int32_t n_sample; // number of sampled tokens
};
LLAMA_API struct llama_perf_context_data llama_perf_context (const struct llama_context * ctx);
@@ -1419,6 +1359,9 @@ extern "C" {
LLAMA_API void llama_perf_sampler_print(const struct llama_sampler * chain);
LLAMA_API void llama_perf_sampler_reset( struct llama_sampler * chain);
// print a breakdown of per-device memory use via LLAMA_LOG:
LLAMA_API void llama_memory_breakdown_print(const struct llama_context * ctx);
//
// training
//
@@ -1437,6 +1380,8 @@ extern "C" {
ggml_opt_get_optimizer_params get_opt_pars; // callback for calculating optimizer parameters
void * get_opt_pars_ud; // userdata for calculating optimizer parameters
enum ggml_opt_optimizer_type optimizer_type;
};
LLAMA_API void llama_opt_init(struct llama_context * lctx, struct llama_model * model, struct llama_opt_params lopt_params);