Lumpiasty/llama.cpp
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

metal : add count_equal op (#18314)

Browse Source 
* add count equal for metal

* remove trailing whitespace

* updated doc ops table

* changed shmem to i32

* added multi tg and templating

* removed BLAS support from Metal docs

* Apply suggestions from code review

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>

* add memset to set dst to 0

* metal : cleanup

---------

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
This commit is contained in:
gatbontonpc
2025-12-31 10:39:48 +02:00
committed by GitHub
co-authored by Georgi Gerganov
parent ecc343de63
commit 9a6369bb60
9 changed files with 585 additions and 324 deletions
Download Patch File Download Diff File Expand all files Collapse all files
ggml/src/ggml-metal/ggml-metal-impl.h
+20
View File
@@ -78,6 +78,7 @@
#define FC_MUL_MM 700
#define FC_ROPE 800
#define FC_SSM_CONV 900
#define FC_COUNT_EQUAL 1000
// op-specific constants
#define OP_FLASH_ATTN_EXT_NQPTG 8
@@ -894,6 +895,25 @@ typedef struct {
float step;
} ggml_metal_kargs_arange;
typedef struct {
int64_t val;
} ggml_metal_kargs_memset;
typedef struct {
int32_t ne00;
int32_t ne01;
int32_t ne02;
int32_t ne03;
uint64_t nb00;
uint64_t nb01;
uint64_t nb02;
uint64_t nb03;
uint64_t nb10;
uint64_t nb11;
uint64_t nb12;
uint64_t nb13;
} ggml_metal_kargs_count_equal;
typedef struct {
int32_t k0;
int32_t k1;