14fe8fb0d9
Build fixes: - patch_ops.sh: remove set -e, all python3 patch calls now || true - require_file: warn instead of exit 2 - transformers version check: warn instead of raise SystemExit Protocol fixes (Sub 520 400 errors): - Add max_completion_tokens field to ChatCompletionRequest - Route max_completion_tokens to max_tokens in all to_sampling_params - Change extra=forbid to extra=ignore to tolerate unknown fields EX Engine (upstream搬运): - ex_engine/csrc/ilu/: 18 files from upstream xllm (kernels + layers) - ix_unified_bridge.cpp: single pybind11 entry for all 14 ixformer infer APIs - ix_unified.py: 3-tier dispatch (bridge then ixformer then pytorch) - gdn_fp32.py: FP32 accumulation GDN (fixes 99.98 pct NaN) - moe_dispatch.py: 7-step MoE pipeline replacing Python for-loop
51 lines
1.9 KiB
C++
51 lines
1.9 KiB
C++
/* Copyright 2025 The xLLM Authors. All Rights Reserved.
|
|
|
|
Licensed under the Apache License, Version 2.0 (the "License");
|
|
you may not use this file except in compliance with the License.
|
|
You may obtain a copy of the License at
|
|
|
|
https://github.com/jd-opensource/xllm/blob/main/LICENSE
|
|
|
|
Unless required by applicable law or agreed to in writing, software
|
|
distributed under the License is distributed on an "AS IS" BASIS,
|
|
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
See the License for the specific language governing permissions and
|
|
limitations under the License.
|
|
==============================================================================*/
|
|
|
|
#include "ilu_ops_api.h"
|
|
#include "utils.h"
|
|
|
|
using namespace ixformer;
|
|
|
|
namespace xllm::kernel::ilu {
|
|
|
|
void residual_layer_norm(torch::Tensor& input,
|
|
torch::Tensor& output,
|
|
std::optional<torch::Tensor>& residual,
|
|
torch::Tensor& weight,
|
|
std::optional<torch::Tensor>& bias,
|
|
std::optional<torch::Tensor>& residual_out,
|
|
double eps) {
|
|
auto residual_ = residual.value_or(torch::zeros_like(input));
|
|
torch::Tensor residual_out_ = residual_out.value_or(torch::zeros_like(input));
|
|
infer::residual_rms_norm(input,
|
|
residual_,
|
|
weight,
|
|
output,
|
|
residual_out_,
|
|
bias,
|
|
/*alpha=*/1.0,
|
|
eps,
|
|
false);
|
|
}
|
|
|
|
void rms_norm(torch::Tensor& output,
|
|
torch::Tensor& input,
|
|
torch::Tensor& weight,
|
|
double eps) {
|
|
std::optional<torch::Tensor> fused_bias = std::nullopt;
|
|
infer::rms_norm(input, weight, output, fused_bias, eps);
|
|
}
|
|
|
|
} // namespace xllm::kernel::ilu
|