vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

ggml : move AMX to the CPU backend (#10570)

Browse source 
* ggml : move AMX to the CPU backend

---------

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
This commit is contained in:
Diego Devesa 2024-11-29 21:54:58 +01:00 • committed by GitHub
parent b782e5c7d4
commit 7cc2d2c889
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
64 changed files with 514 additions and 801 deletions
Download patch file Download diff file Expand all files Collapse all files

3
ggml/src/ggml-backend.cpp
View file

@ -742,7 +742,8 @@ static int ggml_backend_sched_backend_id_from_cur(ggml_backend_sched_t sched, st
if (tensor->buffer || (tensor->view_src && tensor->view_src->buffer)) {
// since the tensor is pre-allocated, it cannot be moved to another backend
GGML_ABORT("pre-allocated tensor (%s) in a backend that cannot run the operation", tensor->name);
ggml_backend_buffer_t buffer = tensor->view_src ? tensor->view_src->buffer : tensor->buffer;
GGML_ABORT("pre-allocated tensor (%s) in a buffer (%s) that cannot run the operation (%s)", tensor->name, ggml_backend_buffer_name(buffer), ggml_op_name(tensor->op));
}
// graph input