Comment by torstenvl

I'm not so sure that the right take-away is "hand-written assembler is 6x faster than C." It's more like "jumps are a lot slower than conditional arithmetic." And that can [edit:often] be achieved easily in C by simply not using switch statements when an if statement or two will work fine.

Rewriting the C function as follows got a 5.5x speedup:

    int run_switches(char *input) {
        int r = 0;
        char c; 
        while (1) {
            c = *input++;
            if (c == 's') r++;
            if (c == 'p') r--;
            if (c == '\0') break;
        }
        return r;
    }

Results:

    [16:50:14 user@boxer ~/looptest] $ gcc -O3 bench.c loop1.c -o lone
    [16:50:37 user@boxer ~/looptest] $ gcc -O3 bench.c loop2.c -o ltwo
    [16:50:47 user@boxer ~/looptest] $ time ./lone 1000 1
    449000
    ./lone 1000 1  3.58s user 0.00s system 99% cpu 3.589 total
    [16:50:57 user@boxer ~/looptest] $ time ./ltwo 1000 1
    449000
    ./ltwo 1000 1  0.65s user 0.00s system 99% cpu 0.658 total