mirror of
https://github.com/ggerganov/llama.cpp.git
synced 2025-01-27 20:43:07 +01:00
19d8762ab6
* rename ggml-cpu-aarch64.c to .cpp * reformat extra cpu backend. - clean Q4_0_N_M and IQ4_0_N_M - remove from "file" tensor type - allow only with dynamic repack - extract cpu extra bufts and convert to C++ - hbm - "aarch64" - more generic use of extra buffer - generalise extra_supports_op - new API for "cpu-accel": - amx - aarch64 * clang-format * Clean Q4_0_N_M ref Enable restrict on C++ * add op GGML_OP_MUL_MAT_ID for Q4_0_N_M with runtime repack * added/corrected control on tensor size for Q4 repacking. * Update ggml/src/ggml-cpu/ggml-cpu-aarch64.cpp Co-authored-by: Georgi Gerganov <ggerganov@gmail.com> * Update ggml/src/ggml-cpu/ggml-cpu-aarch64.cpp Co-authored-by: Georgi Gerganov <ggerganov@gmail.com> * add debug logs on repacks. --------- Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
37 lines
1.2 KiB
C++
37 lines
1.2 KiB
C++
#include "ggml-cpu-traits.h"
|
|
|
|
#include "ggml-backend-impl.h"
|
|
#include "ggml-backend.h"
|
|
|
|
namespace ggml::cpu {
|
|
tensor_traits::~tensor_traits() {}
|
|
|
|
extra_buffer_type::~extra_buffer_type() {}
|
|
} // namespace ggml::cpu
|
|
|
|
bool ggml_cpu_extra_compute_forward(struct ggml_compute_params * params, struct ggml_tensor * op) {
|
|
for (auto extra : ggml_backend_cpu_get_extra_buffers_type()) {
|
|
if (extra && extra->context) {
|
|
auto buf_extra = (ggml::cpu::extra_buffer_type *) extra->context;
|
|
auto tensor_traits = buf_extra->get_tensor_traits(op);
|
|
if (tensor_traits && tensor_traits->compute_forward(params, op)) {
|
|
return true;
|
|
}
|
|
}
|
|
}
|
|
return false;
|
|
}
|
|
|
|
bool ggml_cpu_extra_work_size(int n_threads, const struct ggml_tensor * op, size_t * size) {
|
|
for (auto extra : ggml_backend_cpu_get_extra_buffers_type()) {
|
|
if (extra && extra->context) {
|
|
auto buf_extra = (ggml::cpu::extra_buffer_type *) extra->context;
|
|
auto tensor_traits = buf_extra->get_tensor_traits(op);
|
|
if (tensor_traits && tensor_traits->work_size(n_threads, op, *size)) {
|
|
return true;
|
|
}
|
|
}
|
|
}
|
|
return false;
|
|
}
|