94057c3d3e
PR Test (NPU) / check-changes (push) Has been cancelled
PR Test (NPU) / pr-gate (push) Has been cancelled
PR Test (NPU) / set-image-config (push) Has been cancelled
PR Test (NPU) / stage-b-test-1-npu-a2 (0) (push) Has been cancelled
PR Test (NPU) / stage-b-test-1-npu-a2 (1) (push) Has been cancelled
PR Test (NPU) / stage-b-test-2-npu-a2 (0) (push) Has been cancelled
PR Test (NPU) / stage-b-test-2-npu-a2 (1) (push) Has been cancelled
PR Test (NPU) / stage-b-test-4-npu-a3 (push) Has been cancelled
PR Test (NPU) / stage-b-test-16-npu-a3 (push) Has been cancelled
PR Test (NPU) / multimodal-gen-test-1-npu-a3 (push) Has been cancelled
PR Test (NPU) / multimodal-gen-test-2-npu-a3 (push) Has been cancelled
PR Test (Arm64) / pr-gate (push) Has been cancelled
PR Test (Arm64) / check-changes (push) Has been cancelled
PR Test (Arm64) / build-test (push) Has been cancelled
PR Test (sgl-router) / gate (push) Has been cancelled
PR Test (sgl-router) / tier-1 — lint (push) Has been cancelled
PR Test (sgl-router) / tier-2 — build + test (push) Has been cancelled
PR Test (sgl-router) / tier-3 — docker (placeholder) (push) Has been cancelled
PR Test (sgl-router) / tier-3 — k8s integration (push) Has been cancelled
PR Test (sgl-router) / tier-3 — e2e (push) Has been cancelled
PR Test (sgl-router) / finish (push) Has been cancelled
PR Test (NPU) / single-node-poc (map[name:qwen3_6_27b_w8a8_1p_in64k_out1k_50ms runner:linux-aarch64-a3-2 test_case:test/registered/ascend/performance/qwen3_6_27b/test_npu_qwen3_6_27b_w8a8_1p_in64k_out1k_50ms.py test_type:perf]) (push) Has been cancelled
PR Test (NPU) / pr-test-npu-finish (push) Has been cancelled
PR Test (Xeon) / pr-gate (push) Has been cancelled
PR Test (Xeon) / check-changes (push) Has been cancelled
PR Test (Xeon) / build-test (, xeon-gnr, base-b-test-cpu) (push) Has been cancelled
PR Test (XPU) / check-changes (push) Has been cancelled
PR Test (XPU) / pr-gate (push) Has been cancelled
PR Test (XPU) / stage-a-test-1-gpu-xpu (push) Has been cancelled
PR Test (XPU) / wait-for-stage-a (push) Has been cancelled
PR Test (XPU) / stage-b-test-1-gpu-xpu (push) Has been cancelled
PR Test (XPU) / finish (push) Has been cancelled
CI Model Inventory / build-inventory (push) Has been cancelled
Lint / lint (push) Has been cancelled
PR Benchmark (SMG Components) / Benchmark Compilation Check (push) Has been cancelled
PR Benchmark (SMG Components) / Benchmark - Manual Policy (push) Has been cancelled
PR Benchmark (SMG Components) / Benchmark - Request Processing (push) Has been cancelled
PR Benchmark (SMG Components) / Benchmark Summary (push) Has been cancelled
PR Test (SMG) / build-wheel (push) Has been cancelled
Release SGLang Model Gateway to PyPI / build on windows (x86_64 - auto) (push) Has been cancelled
Release SGLang Model Gateway to PyPI / build on macos (x86_64 - auto) (push) Has been cancelled
PR Test (SMG) / python-unit-tests (push) Has been cancelled
PR Test (SMG) / unit-tests (push) Has been cancelled
PR Test (SMG) / benchmarks (push) Has been cancelled
PR Test (SMG) / chat-completions (push) Has been cancelled
PR Test (SMG) / chat-completions-4gpu (push) Has been cancelled
PR Test (SMG) / e2e (push) Has been cancelled
PR Test (SMG) / docker-build-test (push) Has been cancelled
PR Test (SMG) / k8s-integration (push) Has been cancelled
PR Test (SMG) / finish (push) Has been cancelled
PR Test (SMG) / summarize-benchmarks (push) Has been cancelled
Release SGLang Model Gateway Docker Image / publish (push) Has been cancelled
Release SGLang Model Gateway to PyPI / build on macos (aarch64 - auto) (push) Has been cancelled
Release SGLang Model Gateway to PyPI / build on linux (aarch64 - auto) (push) Has been cancelled
Release SGLang Model Gateway to PyPI / build on linux (x86_64 - auto) (push) Has been cancelled
Release SGLang Model Gateway to PyPI / build on linux (aarch64 - musllinux_1_1) (push) Has been cancelled
Release SGLang Model Gateway to PyPI / build on linux (x86_64 - musllinux_1_1) (push) Has been cancelled
Release SGLang Model Gateway to PyPI / Build SDist (push) Has been cancelled
Release SGLang Model Gateway to PyPI / Upload to PyPI (push) Has been cancelled
Release SGLang Kernels / build-cu129-matrix (aarch64, 12.9, 3.10, arm-kernel-build-node) (push) Has been cancelled
Release SGLang Kernels / build-cu129-matrix (x86_64, 12.9, 3.10, x64-kernel-build-node) (push) Has been cancelled
Release SGLang Kernels / release-cu129 (push) Has been cancelled
Release SGLang Kernels / build-cu130-matrix (aarch64, 13.0, 3.10, arm-kernel-build-node) (push) Has been cancelled
Release SGLang Kernels / build-cu130-matrix (x86_64, 13.0, 3.10, x64-kernel-build-node) (push) Has been cancelled
Release SGLang Kernels / release-cu130 (push) Has been cancelled
Release SGLang Kernels / build-rocm-matrix (3.10, 700) (push) Has been cancelled
Release SGLang Kernels / build-rocm-matrix (3.10, 720) (push) Has been cancelled
Release SGLang Kernels / release-rocm700 (push) Has been cancelled
Release SGLang Kernels / release-rocm720 (push) Has been cancelled
Release SGLang Kernels / build-musa43 (43, 3.10) (push) Has been cancelled
Release SGLang Kernels / release-musa43 (push) Has been cancelled
76 lines
2.0 KiB
Plaintext
76 lines
2.0 KiB
Plaintext
/// \file math.cuh
|
|
/// \brief Device-side math helper functions and constants.
|
|
///
|
|
/// Provides type-generic wrappers around CUDA math intrinsics by
|
|
/// dispatching through `dtype_trait<T>`. All functions are forced-inline
|
|
/// device functions.
|
|
|
|
#pragma once
|
|
#include <sgl_kernel/type.cuh>
|
|
|
|
#include <cmath>
|
|
|
|
namespace device::math {
|
|
|
|
/// \brief Constant: log2(e)
|
|
inline constexpr float log2e = 1.44269504088896340736f;
|
|
/// \brief Constant: ln(2)
|
|
inline constexpr float loge2 = 0.693147180559945309417f;
|
|
/// \brief Maximum representable value for FP8 E4M3 format.
|
|
/// Arch-aware: 448 on CUDA / AMD OCP e4m3fn (gfx950), 224 on AMD e4m3fnuz
|
|
/// (gfx942). Mirrors kFP8E4M3Max so fp8 quant scale divisors and clamps in
|
|
/// the dsv4 compute path (indexer Q-quant, MoE silu+mul / dispatch quant,
|
|
/// GEMM per-tensor quant) do not over-saturate fnuz hardware.
|
|
inline constexpr float FP8_E4M3_MAX = ::kFP8E4M3Max;
|
|
static_assert(log2e * loge2 == 1.0f, "log2e * loge2 must be 1");
|
|
|
|
/// \brief Returns the larger of `a` and `b`.
|
|
template <typename T>
|
|
SGL_DEVICE T max(T a, T b) {
|
|
return dtype_trait<T>::max(a, b);
|
|
}
|
|
|
|
/// \brief Returns the smaller of `a` and `b`.
|
|
template <typename T>
|
|
SGL_DEVICE T min(T a, T b) {
|
|
return dtype_trait<T>::min(a, b);
|
|
}
|
|
|
|
/// \brief Returns the absolute value of `a`.
|
|
template <typename T>
|
|
SGL_DEVICE T abs(T a) {
|
|
return dtype_trait<T>::abs(a);
|
|
}
|
|
|
|
/// \brief Returns the square root of `a`.
|
|
template <typename T>
|
|
SGL_DEVICE T sqrt(T a) {
|
|
return dtype_trait<T>::sqrt(a);
|
|
}
|
|
|
|
/// \brief Returns the reciprocal square root of `a` (i.e. 1 / sqrt(a)).
|
|
template <typename T>
|
|
SGL_DEVICE T rsqrt(T a) {
|
|
return dtype_trait<T>::rsqrt(a);
|
|
}
|
|
|
|
/// \brief Returns e^a.
|
|
template <typename T>
|
|
SGL_DEVICE T exp(T a) {
|
|
return dtype_trait<T>::exp(a);
|
|
}
|
|
|
|
/// \brief Returns sin(a).
|
|
template <typename T>
|
|
SGL_DEVICE T sin(T a) {
|
|
return dtype_trait<T>::sin(a);
|
|
}
|
|
|
|
/// \brief Returns cos(a).
|
|
template <typename T>
|
|
SGL_DEVICE T cos(T a) {
|
|
return dtype_trait<T>::cos(a);
|
|
}
|
|
|
|
} // namespace device::math
|