biondizzle/DeepGEMM
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

[Public release 26/04] Introducing Mega MoE, FP4 Indexer and other features/fixes (#304)

Browse Source 
* Merge with private repo

* Update README

* Update README

* Update README

* Add PyTorch requirements

* Fix sync scopes for MQA logits (#256)

* Update README
This commit is contained in:
Chenggang Zhao
2026-04-17 09:45:14 +08:00
committed by GitHub
parent d30fc36c8f
commit 7f2a703ed5
109 changed files with 12101 additions and 3219 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
csrc/utils/exception.hpp
View File

@@ -42,7 +42,7 @@ do { \
#ifndef DG_NVRTC_CHECK
#define DG_NVRTC_CHECK(cmd) \
do { \
const auto& e = (cmd); \
const auto e = (cmd); \
if (e != NVRTC_SUCCESS) { \
throw DGException("NVRTC", __FILE__, __LINE__, nvrtcGetErrorString(e)); \
} \
@@ -52,7 +52,7 @@ do { \
#ifndef DG_CUDA_DRIVER_CHECK
#define DG_CUDA_DRIVER_CHECK(cmd) \
do { \
const auto& e = (cmd); \
const auto e = (cmd); \
if (e != CUDA_SUCCESS) { \
std::stringstream ss; \
const char *name, *info; \
@@ -66,7 +66,7 @@ do { \
#ifndef DG_CUDA_RUNTIME_CHECK
#define DG_CUDA_RUNTIME_CHECK(cmd) \
do { \
const auto& e = (cmd); \
const auto e = (cmd); \
if (e != cudaSuccess) { \
std::stringstream ss; \
ss << static_cast<int>(e) << " (" << cudaGetErrorName(e) << ", " << cudaGetErrorString(e) << ")"; \
@@ -97,7 +97,7 @@ inline const char* cublasGetStatusString(cublasStatus_t status) {
#define DG_CUBLASLT_CHECK(cmd) \
do { \
const auto& e = (cmd); \
const auto e = (cmd); \
if (e != CUBLAS_STATUS_SUCCESS) { \
std::ostringstream ss; \
ss << static_cast<int>(e) << " (" << cublasGetStatusString(e) << ")"; \