Commit b1ff83b
hexagon: further optimization and tuning of matmul and dot kernels (ggml-org#19407)
* ggml-hexagon: implement 2x2 matmul kernel
* hexmm: implement vec_dot_rx2x2 for Q8_0 and MXFP4
* hexagon: fix editor config failures
* hexagon: refactor matmul ops to use context struct and remove wrappers
Also implement vec_dot_f16 2x2
* hexagon: refactor dyn quantizers to use mmctx
* hexagon: remove mm fastdiv from op_ctx
* hexagon: refactor matmul entry point to reduce code duplication
---------
Co-authored-by: Trivikram Reddy <tamarnat@qti.qualcomm.com>1 parent 4ae1b75 commit b1ff83b
2 files changed
Lines changed: 847 additions & 671 deletions
| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
64 | 64 | | |
65 | 65 | | |
66 | 66 | | |
67 | | - | |
68 | | - | |
69 | | - | |
70 | | - | |
71 | | - | |
72 | 67 | | |
73 | 68 | | |
74 | 69 | | |
75 | 70 | | |
76 | 71 | | |
77 | 72 | | |
78 | | - | |
79 | | - | |
80 | | - | |
81 | | - | |
82 | | - | |
83 | | - | |
84 | | - | |
85 | | - | |
86 | 73 | | |
87 | 74 | | |
88 | 75 | | |
| |||
0 commit comments