Re: question about linker

On 27/11/2024 12:38, Thiago Adams wrote:

On 27/11/2024 09:10, Bart wrote:

On 27/11/2024 11:57, Bart wrote:

On 27/11/2024 01:52, Thiago Adams wrote:

>

I also use ILs for my compilers, but I write my own backends. I've worked on two diifferent kinds. One looks like a HLL, and only exists for my language. So this original source:

>
I forgot to say that I've also tried transpiling to C from my language.
>
That makes some things simpler (I don't need to write the backend! And I get optimisation for free), but C is a poor fit for my language. So programs that need to be transpiled to C can only use a restricted, crippled set of features.
>
My HLL example produces this C:
>
    static void $t$f(void) {
         i64 r;
         i64 a;
         i64 b;
         i64 c;
         r = (a + (b * c));
    }
>
The advantage of my ILs is that they don't have the restrictions of C. For example, they will translate this:
>
     (c | a | b) := 0         # (assign 0 to either a or b)
>
with no problem. The C transpiler will produce this:
>
     (!!(c) ? a : b) = (i64)0;
>
But this is not legal C.

 But you can write it in a different way, can't you?
This raises the question what cannot be done in C?

It's just harder. The transformations become harder, if I want to generated structured C. My syntax is expression-based, for example.
I used also a version called 'linear C' which was flattened C code, but as I said in my last post, the result is usually a travesty of the language. And the result absolutely needs an optimising C compiler, otherwise the code would be even half the speed Tiny C.
Sometimes the C was used for distribution: so people can create binaries locally, avoiding the AV issues with downloading binaries.
Now, for people using Windows, I'd be likely to provide a .ASM file instead. For one thing, ASM would support inline ASM sequences in my programs, which C can't handle.