EngineX-Ascend/enginex-ascend-910-llama.cpp
10
0
Fork 0
Code Issues Pull Requests Actions 4 Projects Releases Wiki Activity

metal : refactor + optimize v2 (#15995)

Browse Source 
* metal : improve naming

* metal : refactor device

ggml-ci

* cont : props

ggml-ci

* metal : apply ggml_mem_ranges_t

ggml-ci

* metal : remove GGML_METAL_USE_BF16

ggml-ci

* metal : refactor device buffer

ggml-ci

* cont : fix naming

* metal : sync before destroying the backend

ggml-ci

* metal : refactor context

ggml-ci

* metal : migrate ggml-metal.m to ggml-metal.cpp

ggml-ci

* metal : adjust ops API

ggml-ci

* metal : use C++ to store piplienes

ggml-ci

* metal : migrate ops to separate functions

ggml-ci

* metal : add ggml_metal_library_t

ggml-ci

* metal : improve naming

ggml-ci

* metal : cleanp

ggml-ci

* metal : add support for GGML_OP_LOG

ggml-ci

* metal : fix error handling

ggml-ci
This commit is contained in:
Georgi Gerganov
2025-09-17 20:38:12 +03:00
committed by GitHub
parent a7a98e0fff
commit 0320ac5264
19 changed files with 7873 additions and 7184 deletions
Download Patch File Download Diff File Expand all files Collapse all files

35
ggml/src/ggml-metal/ggml-metal-impl.h
View File

@@ -165,6 +165,16 @@ typedef struct {
uint64_t nb3;
} ggml_metal_kargs_repeat;
typedef struct {
float scale;
float bias;
} ggml_metal_kargs_scale;
typedef struct {
float min;
float max;
} ggml_metal_kargs_clamp;
typedef struct {
int64_t ne00;
int64_t ne01;
@@ -453,7 +463,7 @@ typedef struct {
uint64_t nb00;
uint64_t nb01;
uint64_t nb02;
int32_t n_groups;
int32_t ngrp;
float eps;
} ggml_metal_kargs_group_norm;
@@ -506,14 +516,6 @@ typedef struct {
uint64_t nb01;
uint64_t nb02;
uint64_t nb03;
int64_t ne10;
int64_t ne11;
int64_t ne12;
int64_t ne13;
uint64_t nb10;
uint64_t nb11;
uint64_t nb12;
uint64_t nb13;
int64_t ne0;
int64_t ne1;
int64_t ne2;
@@ -547,12 +549,6 @@ typedef struct {
int32_t n_head_log2;
} ggml_metal_kargs_soft_max;
typedef struct {
int64_t ne00;
int64_t ne01;
int n_past;
} ggml_metal_kargs_diag_mask_inf;
typedef struct {
int64_t ne00;
int64_t ne01;
@@ -579,7 +575,7 @@ typedef struct {
int64_t n_group;
int64_t n_seq_tokens;
int64_t n_seqs;
int64_t s_off;
uint64_t s_off;
uint64_t nb01;
uint64_t nb02;
uint64_t nb03;
@@ -719,7 +715,12 @@ typedef struct {
int64_t IW;
int64_t OH;
int64_t OW;
int64_t parallel_elements;
int64_t np;
} ggml_metal_kargs_pool_2d;
typedef struct {
int64_t ne00;
uint64_t nb01;
} ggml_metal_kargs_argmax;
#endif // GGML_METAL_IMPL