Everything was working fine until I added another ptx instruction
dummy = 0-1; // = ffff ffff ffff ffff
asm("{\n" \
"add.cc.u32 %0, %13, %0;\n\t"
"addc.cc.u32 %12, %12, 0;\n\t" \
"addc.cc.u32 %11, %11, 0;\n\t" \
"addc.cc.u32 %10, %10, 0;\n\t" \
"addc.cc.u32 %9, %9, 0;\n\t" \
"addc.cc.u32 %8, %8, 0;\n\t" \
"addc.cc.u32 %7, %7, 0;\n\t" \
"addc.cc.u32 %6, %6, 0;\n\t" \
"addc.cc.u32 %5, %5, 0;\n\t" \
"addc.cc.u32 %4, %4, 0;\n\t" \
"addc.cc.u32 %3, %3, 0;\n\t" \
"addc.cc.u32 %2, %2, 0;\n\t" \
"addc.u32 %1, %1, 0;\n\t" \
"}" : "+r"(carry),"+r"(x[66]),"+r"(x[67]),"+r"(x[68]),"+r"(x[69]),"+r"(x[70]),"+r"(x[71]),"+r"(x[72]),"+r"(x[73]),"+r"(x[74]),"+r"(x[75]),"+r"(x[76]),"+r"(x[77]) : "r"(dummy) : );
for(i=0; i<66; i++) cuPrintf("33x33[%d] = %x\n", 66+i, x[66 + i]);
And then I get this error, that seems to not referring to the code…
Signal: caught in Code_Expansion phase.
1><input> : error : Signal caught in phase Code_Expansion -- processing aborted
1>CUDACOMPILE : nvopencc error : C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v4.0\bin/../open64/lib//be.exe returned non-zero status 3
1>C:\Program Files (x86)\MSBuild\Microsoft.Cpp\v4.0\BuildCustomizations\CUDA 4.0.targets(285,3): error MSB3721: The command ""C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v4.0\bin\nvcc.exe" -gencode=arch=compute_20,code=\"sm_20,compute_20\" --use-local-env --cl-version 2010 -ccbin "C:\Program Files (x86)\Microsoft Visual Studio 10.0\VC\bin\x86_amd64" -I"C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v4.0\include" -G0 --keep-dir "x64\Debug" -maxrregcount=0 --ptxas-options=-v --machine 64 --compile -D_NEXUS_DEBUG -g -Xcompiler "/EHsc /nologo /Od /Zi /MDd " -o "x64\Debug\Karatsuba.cu.obj" "C:\Users\Giuseppe\Desktop\giuseppe\src\Karatsuba\Karatsuba.cu"" exited with code 2.
1>
1>Build FAILED.
Any idea?