Thread 106115566

>>106115566 (OP)
Skill issue.

>>106115663
Yeah all the skilled programmers are leaving to other languages that respect the programmer's time.

I honestly don't understand why linkers are still slow outside of backwards compatibility boomer bullshit reasons because how hard can it be? Isn't the ideal linker basically just cat with some offset fix ups?

I don't fucking get it. It's insane how a linker step can kill all parallelism in your codebase's build system, give you static init ordering nasal demons and a desire to kill oneself. Runtime linking is also fucking dumb, but it's also why win32 and M$s obsession with stable ABIs still work in Windows 11. Fuck computers and fuck jannies.

>>106115748
Niggers don't know about my Digital Mars C's linker.

>>106115748
>Runtime linking is also fucking dumb
Not really.

>>106115748
>It's insane how a linker step can kill all parallelism
it's a myth
https://github.com/rui314/mold
and there have been some other parallel linkers after that

>>106115748
0/10 bait

>>106115693
>leaving to other languages
to what languages?

Back in the day, trolls tried to make people angry.
Nowadays, people get angry randomly all by themselves and make threads seething, whining and crying about the most unimportant minor things.

>>106115972
>>106116015
give exactly one (1) reason why it's necessary for linkers to not underflow 64 bit functions to 32 bit on systems without AVX2. it's obsolete technology.

>>106115566 (OP) (Checked)
Hilariously enough, one of the few things in CS that was unironically invented by a woman.

It has many drawbacks and potential system issues for.. literally almost no tangible benefit whatsoever at this point in hardware tech.

filtered

>>106115942
Used to be java, now is the web browser

>>106117080
really? thought linker was invented by John Mauchly

>>106115566 (OP)
Not sure I would go that far but it's pretty clear the original design of linkers doesn't fit how they're used today. Many modern languages avoid the linker almost entirely during build themselves and only really rely on the linker for integrating with libc.

>>106115566 (OP)
Go solved this, and compilation times weren't affected. in fact, compiling Go programs is fast.
I don't get why they couldn't come up with a similar solution in the 70s.

I wish I knew what you guys were talking about sometimes but I'm too busy making 3x as a security analyst.

>>106117994
>security analyst
>doesn't know shit about programming
you should kys

>>106118008
AI. I have no reason to learn anymore.

>>106118030
you will end up destroying something, getting fired and put in a blacklist. and that is a good thing. enjoy your """AI""".

>>106118062
I've been doing this for 12 years lmao but I guess we'll see! I've made enough money over the years it's fine either way :)

>>106115566 (OP)
That's not Docker.

>>106115748
>Isn't the ideal linker basically just cat with some offset fix ups?
Link Time Optimization.

>>106117967
>I don't get why they couldn't come up with a similar solution in the 70s.
You fail to appreciate just how little compute power, memory, and disk there really was back then.
But that's because you're a weenie.

>>106120706
ok, late 80s then.
I'm sure modern C compilers are not much worse in terms of "compute power, memory, and disk" than the Go compiler lmao

>>106121855
>retard misses the point
many such cases

>>106115693
didnt get a real answer lmao

>>106121855
func Contains[T comparable](slice []T, val T) bool {
for _, item := range slice {
if item == val {
return true
}
}
return false
}

Gay retard

>>106122492
that's wrong, though

>>106116015
And most of the time it's anons not wanting to learn something