dmd -run speed trends

[Siarhei Siamashka]

On Sunday, 17 December 2023 at 06:40:33 UTC, Witold Baryluk wrote:
> On Saturday, 16 December 2023 at 03:31:05 UTC, Walter Bright 
> wrote:
>> It would be illuminating to compare using printf rather than 
>> writefln.
>
> Ok, added `extern(C) int printf(const char *format, ...);`, and 
> measured various variants.

I don't think that there's much practical value in testing 
`printf` because it's not compatible with `@safe` and can't be 
recommended for developing normal D applications. This scenario 
gets out of touch with reality and becomes way too artificial.

> * mold 2.3.3 and mold 2.4.0, segfault, when using with dmd or 
> ldc2.

https://github.com/dlang/dmd/pull/15915 improves dmd's 
compatibility with mold 2.4.0 or maybe even fixes all problems if 
we are optimistic.

> [...]
> ```
> stat("/usr/include/dmd/druntime/import/core/internal/array/comparison.di", 0x7ffecfc2a0f0) = -1 ENOENT (No such file or directory)
> stat("/usr/include/dmd/druntime/import/core/internal/array/comparison.d", {st_mode=S_IFREG|0644, st_size=7333, ...}) = 0
> stat("/usr/include/dmd/druntime/import/core/internal/array/comparison.d", {st_mode=S_IFREG|0644, st_size=7333, ...}) = 0
> openat(AT_FDCWD, 
> "/usr/include/dmd/druntime/import/core/internal/array/comparison.d", O_RDONLY) = 3
> fstat(3, {st_mode=S_IFREG|0644, st_size=7333, ...}) = 0
> read(3, "/**\n * This module contains comp"..., 7333) = 7333
> close(3)
> ```
>
> Each of these syscalls is about 15μs on my system (when 
> stracing, probably little less in real run without `strace` 
> overheads)
>
> There should be a way to reduce this in half with smarter 
> sequencing (i.e. do `open` first instead of `stat + open + 
> fstat`).

That's an interesting discovery. Now it's necessary to implement 
a proof of concept patch for dmd to check how much this can 
actually help in reality. Can you try this?