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

Remove padding and multiple D2D copies for MTP (#24086)

Browse Source 
* Make ggml_gated_delta_net take only the initial recurrent state (D, 1, n_seqs) and passes the snapshot count K as an op parameter instead of inferring it from state->ne[1].

Remove the padding hack and copy all emitted snapshots into the recurrent cache with a single strided ggml_cpy

* Make GDN changes in all backends. Address review comments.

* Fix CI build errors
This commit is contained in:
Gaurav Garg
2026-06-10 23:21:16 +05:30
committed by GitHub
parent d2462f8f7a
commit e95dae18d6
20 changed files with 118 additions and 106 deletions
Download Patch File Download Diff File Expand all files Collapse all files
ggml/src/ggml-opencl/ggml-opencl.cpp
+1 -1
View File
@@ -17750,7 +17750,7 @@ static void ggml_cl_gated_delta_net(ggml_backend_t backend, ggml_tensor * dst) {
const cl_uint H_v = (cl_uint) src_v->ne[1];
const cl_uint n_tokens = (cl_uint) src_v->ne[2];
const cl_uint n_seqs = (cl_uint) src_v->ne[3];
const cl_uint K = (cl_uint) src_state->ne[1];
const cl_uint K = (cl_uint) ggml_get_op_params_i32(dst, 0);
int si;
switch (S_v) {