We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
There was an error while loading. Please reload this page.
1 parent f28af0d commit 05c51f9Copy full SHA for 05c51f9
ggml-cuda.cu
@@ -7117,6 +7117,9 @@ static void ggml_cuda_op_mul_mat(
7117
7118
CUDA_CHECK(ggml_cuda_set_device(g_main_device));
7119
for (int64_t id = 0; id < g_device_count; ++id) {
7120
+ if (g_tensor_split[id] >= (id + 1 < g_device_count ? g_tensor_split[id + 1] : 1.0f)) {
7121
+ continue;
7122
+ }
7123
for (int64_t is = 0; is < is_max; ++is) {
7124
CUDA_CHECK(cudaStreamWaitEvent(g_cudaStreams[g_main_device][0], src0_extra->events[id][is], 0));
7125
}
0 commit comments