Re: Automatic "code-pattern" recognition, in DSP codes

The usual instruction selection is itself a pattern matching like you
are talking abt. I am not sure if any compiler does it, but I would
imagine that it can be trivially accomodated in the present algorithm
itself. It is just a matter of adding the pattern for call fir() to the
library of opcodes. Now the only hindrance might be if the instruction
selection is on a data dependence graph based on a basic block.
        Yet another related work could be those which synthesize some loops
into FPGA. There again you have to detect the loop, find its
characteristics, estimate the hardware and then synthesis. A possible
paper on this is the following -
www.comp.nus.edu.sg/~tulika/fpl03.pdf








Miguel Casás-Sánchez wrote:
> I am researching in compilers and source to source code transformations,
> for Digital Signal Processors, in particular with the Texas Instruments
> Code Composer Studio 2.2 for the 55x family.
>
> I was wondering if somebody has ever heard about something like
> "automatic typical DSP-like kernel recognition", for example, as the
> compiler reads the program and sets up the data and control graphs,
> recognise that
>
> for(i=0, acc = 0; i < N; i++) {
> acc = h[n] * X[N-n];
> }
>
> can be a FIR filter, that may be substituted as a whole by a fir()
> library function call, most optimised.
>
> Can be also understood as pattern-search in the code control/data graphs?

Return to the comp.compilers page.
Search the comp.compilers archives again.