clean submission: comp168 base + max_model_len=100000

qwen3_6_scripts/flash_qla_sm70/build/.ninja_log

@@ -0,0 +1,5 @@
+# ninja log v5
+0	61739	1786467036068204659	gdn_forward.cuda.o	4fbd18c8f06e5181
+61739	62033	1786467036388208334	flash_qla_sm70_gdn_strided.so	a5d04d69a8ccfcee
+0	60985	1786469746403441679	gdn_forward.cuda.o	15f5cb32976bd0b3
+60985	61271	1786469746711445255	flash_qla_sm70_gdn_strided.so	a5d04d69a8ccfcee

qwen3_6_scripts/flash_qla_sm70/build/build.ninja

@@ -0,0 +1,31 @@
+ninja_required_version = 1.3
+cxx = c++
+nvcc = /usr/local/corex/bin/clang++
+
+cflags = -DTORCH_EXTENSION_NAME=flash_qla_sm70_gdn_strided -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE=\"_gcc\" -DPYBIND11_STDLIB=\"_libstdcpp\" -DPYBIND11_BUILD_ABI=\"_cxxabi1011\" -isystem /usr/local/corex/lib64/python3/dist-packages/torch/include -isystem /usr/local/corex/lib64/python3/dist-packages/torch/include/torch/csrc/api/include -isystem /usr/local/corex/lib64/python3/dist-packages/torch/include/TH -isystem /usr/local/corex/lib64/python3/dist-packages/torch/include/THC -isystem /usr/local/corex/include -isystem /usr/local/include/python3.10 -D_GLIBCXX_USE_CXX11_ABI=0 -fPIC -std=c++17 -O3
+post_cflags = 
+cuda_cflags = -DTORCH_EXTENSION_NAME=flash_qla_sm70_gdn_strided -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE=\"_gcc\" -DPYBIND11_STDLIB=\"_libstdcpp\" -DPYBIND11_BUILD_ABI=\"_cxxabi1011\" -isystem /usr/local/corex/lib64/python3/dist-packages/torch/include -isystem /usr/local/corex/lib64/python3/dist-packages/torch/include/torch/csrc/api/include -isystem /usr/local/corex/lib64/python3/dist-packages/torch/include/TH -isystem /usr/local/corex/lib64/python3/dist-packages/torch/include/THC -isystem /usr/local/corex/include -isystem /usr/local/include/python3.10 -D_GLIBCXX_USE_CXX11_ABI=0 -D__CUDA_NO_HALF_OPERATORS__ -D__CUDA_NO_HALF_CONVERSIONS__ -D__CUDA_NO_BFLOAT16_CONVERSIONS__ -D__CUDA_NO_HALF2_OPERATORS__ -D__ILUVATAR__ -D__ILUVATAR_WORKAROUND__ -D__ILUVATAR_DIAG__ -cl-single-precision-constant -fPIC -mllvm --bonus-inst-threshold=0 -O3 --cuda-gpu-arch=ivcore10 --cuda-path=/usr/local/corex -std=c++17
+cuda_post_cflags = 
+cuda_dlink_post_cflags = 
+ldflags = -shared -L/usr/local/corex/lib64/python3/dist-packages/torch/lib -lc10 -lc10_cuda -ltorch_cpu -ltorch_cuda -ltorch -ltorch_python -L/usr/local/corex/lib64 -lcudart
+
+rule compile
+  command = $cxx -MMD -MF $out.d $cflags -c $in -o $out $post_cflags
+  depfile = $out.d
+  deps = gcc
+
+rule cuda_compile
+  command = $nvcc  $cuda_cflags -c $in -o $out $cuda_post_cflags
+
+
+
+rule link
+  command = $cxx $in $ldflags -o $out
+
+build gdn_forward.cuda.o: cuda_compile /workspace/qwen3_6_scripts/flash_qla_sm70/csrc/gdn_forward.cu
+
+
+
+build flash_qla_sm70_gdn_strided.so: link gdn_forward.cuda.o
+
+default flash_qla_sm70_gdn_strided.so