Open Source Software Community - it's about ethics in Code of Conducts

  • Want to keep track of this thread?
    Accounts can bookmark posts, watch threads for updates, and jump back to where you stopped reading.
    Create account
neger psykolog said:
stop trying to help the programmer death "camp" (sponsored by microsoft) "detainees" by explaining such an obscure requirement in the first place, now its going to show up on search results!
Click to expand...
I'll do you one worse
I've asked it to rewrite an abstractionless musl-based C implementation of some fairly trivial python that I made up
Python: 
x = # arbitrary comma-delimited string
for i in x.split(","):
    print(str(i).upper())
C: 
// Compile with: musl-gcc -static -o split_upper split_upper.c
#include <unistd.h> // write()
#include <ctype.h>  // toupper()
#include <string.h> // strlen()

int main(void) {
    char x[] = "foo,Bar,BaZ"; // arbitrary comma-delimited string
    char *p = x;              // current character pointer
    char *segment_start = x;  // start of current token

    for (; *p; p++) {
        if (*p == ',') {
            // process segment_start to p-1
            for (char *q = segment_start; q < p; q++) {
                char c = *q;
                if (c >= 'a' && c <= 'z') {
                    c -= 32; // manual toupper without a function call
                }
                write(1, &c, 1); // raw syscall: write to stdout
            }
            write(1, "\n", 1);
            segment_start = p + 1;
        }
    }
    // last segment (no trailing comma)
    for (char *q = segment_start; *q; q++) {
        char c = *q;
        if (c >= 'a' && c <= 'z') {
            c -= 32;
        }
        write(1, &c, 1);
    }
    write(1, "\n", 1);

    return 0;
}
 
XL xQgg?QcQCaTYDMjqoDnYpG said:
I'll do you one worse
I've asked it to rewrite an abstractionless musl-based C implementation of some fairly trivial python that I made up
Python: 
x = # arbitrary comma-delimited string
for i in x.split(","):
    print(str(i).upper())
C: 
// Compile with: musl-gcc -static -o split_upper split_upper.c
#include <unistd.h> // write()
#include <ctype.h>  // toupper()
#include <string.h> // strlen()

int main(void) {
    char x[] = "foo,Bar,BaZ"; // arbitrary comma-delimited string
    char *p = x;              // current character pointer
    char *segment_start = x;  // start of current token

    for (; *p; p++) {
        if (*p == ',') {
            // process segment_start to p-1
            for (char *q = segment_start; q < p; q++) {
                char c = *q;
                if (c >= 'a' && c <= 'z') {
                    c -= 32; // manual toupper without a function call
                }
                write(1, &c, 1); // raw syscall: write to stdout
            }
            write(1, "\n", 1);
            segment_start = p + 1;
        }
    }
    // last segment (no trailing comma)
    for (char *q = segment_start; *q; q++) {
        char c = *q;
        if (c >= 'a' && c <= 'z') {
            c -= 32;
        }
        write(1, &c, 1);
    }
    write(1, "\n", 1);

    return 0;
}
Click to expand...
ok fine if you stop spoiling the programmer death coding "camp" (sponsored by microsoft) I'll let you have lunch with the guards ONCE (but after that you have to back to your terminal)
 
the rat said:
Out of curiosity has anyone tried to actually benchmark these games using DXVK under Windows as well and compared those results to DXVK under Linux?
Click to expand...
it would not surprise me if dxvk was marginally faster than the windows directx somehow, but it might be because windows vulkan is slower than linux vulkan
it's very hard to get a true apples-to-apples comparison here, there are likely some things windows is way better at (probably mostly in the graphics compositing area)
Darkholme's Dungeon said:
We don't need any barbed wire. Just put them in a warehouse and force them to (re)write everything in musl based C with no abstractions.
Click to expand...
or in a lisp-1 dialect with laborious documentation and carefully chosen abstractions
XL xQgg?QcQCaTYDMjqoDnYpG said:
1754893079609.webp
Click to expand...
what the actual fuck is this retarded llm smoking, isn't man 2 write for files and to make a server you need the socket functions send and recv????????
 
Creative Username said:
what the actual fuck is this retarded llm smoking, isn't man 2 write for files and to make a server you need the socket functions send and recv????????
Click to expand...
Markass the Worst said:
LLMs are retarded and anybody using them without fact checking is also retarded.
Click to expand...
Guys the programming death coding "camp" (sponsored by microsoft) is specifically supposed to kill off retards, if you keep spilling all the secrets then its not going to work!
 
XL xQgg?QcQCaTYDMjqoDnYpG said:
I'll do you one worse
I've asked it to rewrite an abstractionless musl-based C implementation of some fairly trivial python that I made up
Python: 
x = # arbitrary comma-delimited string
for i in x.split(","):
    print(str(i).upper())
C: 
// Compile with: musl-gcc -static -o split_upper split_upper.c
#include <unistd.h> // write()
#include <ctype.h>  // toupper()
#include <string.h> // strlen()

int main(void) {
    char x[] = "foo,Bar,BaZ"; // arbitrary comma-delimited string
    char *p = x;              // current character pointer
    char *segment_start = x;  // start of current token

    for (; *p; p++) {
        if (*p == ',') {
            // process segment_start to p-1
            for (char *q = segment_start; q < p; q++) {
                char c = *q;
                if (c >= 'a' && c <= 'z') {
                    c -= 32; // manual toupper without a function call
                }
                write(1, &c, 1); // raw syscall: write to stdout
            }
            write(1, "\n", 1);
            segment_start = p + 1;
        }
    }
    // last segment (no trailing comma)
    for (char *q = segment_start; *q; q++) {
        char c = *q;
        if (c >= 'a' && c <= 'z') {
            c -= 32;
        }
        write(1, &c, 1);
    }
    write(1, "\n", 1);

    return 0;
}
Click to expand...
im going to try my hand at this but i dont have musl on my computer afaik so im just going to do it the better way:
C Library is Bloat Anyway™
put that in your suckless pipe and smoke it
 
Creative Username said:
what the actual fuck is this retarded llm smoking, isn't man 2 write for files and to make a server you need the socket functions send and recv????????
Click to expand...
You can use read()/write() to sockets just fine on almost all platforms. Exceptions are mostly some RTOSs that do not have native socket support so thus you need to use special library calls from the socket library. LWIP is like that.

In fact read()/write() varients readv()/writev() are superior in many cases as they are the only way to pass a set of iovectors to be read/written to the socket, This can really improve performance for some types of applications.
I would say the best ways to read/write to a socket are:
1, iouring
2, readv/writev
3, anything else, including read/write/recv/send/...
 
XL xQgg?QcQCaTYDMjqoDnYpG said:
I'll do you one worse
I've asked it to rewrite an abstractionless musl-based C implementation of some fairly trivial python that I made up
Python: 
x = # arbitrary comma-delimited string
for i in x.split(","):
    print(str(i).upper())
C: 
// Compile with: musl-gcc -static -o split_upper split_upper.c
#include <unistd.h> // write()
#include <ctype.h>  // toupper()
#include <string.h> // strlen()

int main(void) {
    char x[] = "foo,Bar,BaZ"; // arbitrary comma-delimited string
    char *p = x;              // current character pointer
    char *segment_start = x;  // start of current token

    for (; *p; p++) {
        if (*p == ',') {
            // process segment_start to p-1
            for (char *q = segment_start; q < p; q++) {
                char c = *q;
                if (c >= 'a' && c <= 'z') {
                    c -= 32; // manual toupper without a function call
                }
                write(1, &c, 1); // raw syscall: write to stdout
            }
            write(1, "\n", 1);
            segment_start = p + 1;
        }
    }
    // last segment (no trailing comma)
    for (char *q = segment_start; *q; q++) {
        char c = *q;
        if (c >= 'a' && c <= 'z') {
            c -= 32;
        }
        write(1, &c, 1);
    }
    write(1, "\n", 1);

    return 0;
}
Click to expand...
That C function looks like something I'd write when I was 12.
 
LOWERCASE LETTERS said:
99% of cheaters in online games use Linux
Click to expand...
I'm pretty you're baiting, as cheats cant exactly protect themselves right on linux, so there's little to none commercial ones.
Most source game cheats target windows because its way easier to work there than on linux.
Any commercial one depends on some loader security so their shit doesnt get cracked in 5 seconds, which is also extremely easier to do on windows.
Not mentioning any windows-only games like bf6.
 
Cuckoman said:
You can use read()/write() to sockets just fine on almost all platforms. Exceptions are mostly some RTOSs that do not have native socket support so thus you need to use special library calls from the socket library. LWIP is like that.

In fact read()/write() varients readv()/writev() are superior in many cases as they are the only way to pass a set of iovectors to be read/written to the socket, This can really improve performance for some types of applications.
I would say the best ways to read/write to a socket are:
1, iouring
2, readv/writev
3, anything else, including read/write/recv/send/...
Click to expand...
Ok this settles it, if the Microsoft sponsored coding boot (death) camp comes to fruition you'll be my coding partner (PS: I bring nothing to the table)
 
LOWERCASE LETTERS said:
Also, like 99% of cheaters in online games use Linux, so it's increasingly not going to be supported in the future.
Click to expand...
Lots of feature packed cheat programs for games are paid software. Good luck convincing your customers to install an obscure system just to grief people in an online game. You'd faster get hardware attachments to your PC that snoop on games in-RAM and completely bypass kernel level anti-cheats.

polakrobak99 said:
For all my years actually using Linux as a daily driver, I still don't understand why should I care about Wayland. Every year or two, I do another big research project about why Wayland is such a revolution compared to Xorg, and every time I'm dissapointed and confused.
Click to expand...
You can't get things like HDR and monitors of different refresh rates without Wayland, to be fair. If that's not something you care about, the only revolutionary bit is how much money the people involved are going to make off Wayland by making up imaginary windmills and fighting with them. Classic open source make-work and yak shaving.
 
Creative Username said:
im going to try my hand at this but i dont have musl on my computer afaik so im just going to do it the better way:
C Library is Bloat Anyway™
Click to expand...
XL xQgg?QcQCaTYDMjqoDnYpG said:
I'll do you one worse
I've asked it to rewrite an abstractionless musl-based C implementation of some fairly trivial python that I made up
Click to expand...
neger psykolog said:
Ok this settles it, if the Microsoft sponsored coding boot (death) camp comes to fruition you'll be my coding partner (PS: I bring nothing to the table)
Click to expand...
Since you guys keep ruining the joke, I guess we'll have to do something better. Write the fourth temple in HolyC (and/or x86_64 asm) while being forced to traverse the million dollar highway (US-550) at night by foot in all black.

If you don't write your own compiler (or use code from others) St. Davis will descend from the heavens to paint you a bioluminecent green and sic a horde of self driving cars repeating "CIA NIGGERS!" which will run you over.
 
LOWERCASE LETTERS said:
Also, like 99% of cheaters in online games use Linux, so it's increasingly not going to be supported in the future.
Click to expand...
Lol, lmao. Tell me you don't know shit about online cheating without telling me you don't know shit

The way cheaters nowadays do it is with a second box connected to the gaming box through a DMA controller, such as this one. This allows you to run your exploits on a second machine modifying the RAM contents on the main machine without any process knowing any better. All of this is of course on windows, because to do this in linux requires you to jump through quite a few hoops. In windows you just install a driver and away you go, ring0

Here's how that works in practice
 
Creative Username said:
it would not surprise me if dxvk was marginally faster than the windows directx somehow, but it might be because windows vulkan is slower than linux vulkan
it's very hard to get a true apples-to-apples comparison here, there are likely some things windows is way better at (probably mostly in the graphics compositing area)
Click to expand...
My autism is getting to me here but what do you mean by this? DirectX under Windows is slower than DXVK because Windows Vulkan is slower despite DirectX not being Vulkan?

I only ask regarding the benchmark thing originally because from what I understand you can get DXVK running under Windows and I suspect people who say Linux is faster have only compared games that were ran using DXVK under Linux and DirectX under Windows which I think is unfair.

I'll have to ask someone about this later when he re-installs Gentoo for the 500th time.

GNU Abyss said:
monitors of different refresh rates without Wayland
Click to expand...
Really? I could swear when I was dicking around trying to figure out the right command under xrandr so that I could get ez 75hz I had my monitors on separate refresh rates and it was working just fine.

Jotch said:
Ask it what it thinks "Hateful Programming" means.
Click to expand...
Shows a wireframe picture of Josh's brain.
 
the rat said:
Really? I could swear when I was dicking around trying to figure out the right command under xrandr so that I could get ez 75hz I had my monitors on separate refresh rates and it was working just fine.
Click to expand...
It's a complaint I've seen reiterated over and over. Could be bullshit, could be more noticable with more different refresh rates, 144hz vs. 75hz for example. Can't confirm personally.
 
the rat said:
My autism is getting to me here but what do you mean by this? DirectX under Windows is slower than DXVK because Windows Vulkan is slower despite DirectX not being Vulkan?
Click to expand...
sorry i was saying it retardedly, i was trying to refer to how it works under the hood and how good the driver compilers are etc.
for instance the linux intel drivers backport new vulkan versions to older graphics chips while the windows driver never gets these new versions or even the directx equivalents
 
You must log in or register to reply here.
Back
Top Bottom