root/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/llama.cpp.git synced 2025-01-05 00:04:36 +00:00
Code Issues Actions 9 Packages Projects Releases Wiki Activity

clean up vulkan/cpu switch

Browse Source
This commit is contained in:
Aaron Miller 2023-10-12 11:46:30 -07:00 committed by cebtenzzre
parent cd0257ed0d
commit 8dc79ac380
1 changed files with 1 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
llama.cpp
View File

@ -3855,19 +3855,11 @@ static bool llama_eval_internal(
ggml_graph_compute_helper(lctx.work_buffer, gf, n_threads); ggml_graph_compute_helper(lctx.work_buffer, gf, n_threads);
} }
#elif defined(GGML_USE_KOMPUTE) #elif defined(GGML_USE_KOMPUTE)
if (lctx.ctx_kompute) { // && N == 1) { if (lctx.ctx_kompute) {
ggml_vk_graph_compute(lctx.ctx_kompute, gf); ggml_vk_graph_compute(lctx.ctx_kompute, gf);
ggml_vk_d2h_tensor(lctx.ctx_kompute, res); ggml_vk_d2h_tensor(lctx.ctx_kompute, res);
} else { } else {
if (lctx.ctx_kompute) {
ggml_vk_d2h_tensor(lctx.ctx_kompute, kv_self.k);
ggml_vk_d2h_tensor(lctx.ctx_kompute, kv_self.v);
}
ggml_graph_compute_helper(lctx.work_buffer, gf, n_threads); ggml_graph_compute_helper(lctx.work_buffer, gf, n_threads);
if (lctx.ctx_kompute) {
ggml_vk_h2d_tensor(lctx.ctx_kompute, kv_self.k);
ggml_vk_h2d_tensor(lctx.ctx_kompute, kv_self.v);
}
} }
#else #else
ggml_graph_compute_helper(lctx.work_buffer, gf, n_threads); ggml_graph_compute_helper(lctx.work_buffer, gf, n_threads);