This site is devoted to the x86 code optimizer/auto-parallelizer ( dco
) developed by Dalsoft. Optimizers, developed by Dalsoft, for other architectures described here.
Work on the LLVM version of the auto-prarallelizer is in progress.
Visit this for the description of
Dalsoft Parallel Library - and all other products offered by Dalsoft.
is used to perform optimization and automatic parallelization of the code generated by a compiler.
The programmer uses a compiler (C/C++, FORTRAN etc.) to translate his code
into x86 assembly code. This code is used as an input to dco.
The output, generated by dco, is a highly optimized/parallelized x86
assembly code that is logically identical to the original; dco
rearranges the existing code, utilizing options and features that are supported by the x86 architecture. To create a final object file the
generated code should be assembled.
Note that dco
does not require preprocessing or any other involvement by the user. It is fully automated and it is possible to
incorporate dco into makefiles or other product generation tools.
The x86 code optimizer currently supports code for the IA-32 and x86-64 architectures which include SSE* extensions and is available on Linux and Windows OS's.