>>109900275
kek you positively DO hate the maidniggers, dont you?

cant blame you though
theyve been quite cancerous lately

>>109900283
I am a man in his 30s.
I don't have any interest in other men pretending to be little girls just computer programming.

File: screenshot_2026-09-25_06:52:15.png (25 KB, 502x483)

25 KB PNG

I have incorporated my password manager into serpent my desktop automation tool.
It get's my password and then opens up the login web page of a specified account.

>>109900312
You are a redditor, and you belong on reddit.

>>109900312
same here
i dont mind their antics but theyre really a hindrance to discussion when they become overbearing
even if thats not your intent i salute the fact that you create new threads a wee bit earlier than usual

dpt should be neutral grounds. and a neutral picrel promotes that

>>109900334
baste

>>109900349
You are a fat redditor.

File: Screenshot from 2026-09-2(...).png (26 KB, 280x196)

26 KB PNG

>>109900381
how bout we keep shitposting to the previous thread?

>>109900391
Silence redditor. You are a redditor and you belong on reddit, not here.

>>109900391
Don't say brown or the jeet will use his verbal IQ against you.

>>109900429
Go collect more Funko Pop dolls, redditor.

File: Screenshot from 2026-09-2(...).png (473 KB, 1419x1000)

473 KB PNG

>>109900397
post code, then we decide about who belongs where
how bout?

>>109900275
R stands for "retarded" as all of its users are exactly that

>>109900447
Post your code on reddit. That is where you belong.

Wrote a do while loop in java and set up a cron job using the task scheduler. I'm surprised my java jar and python script survived a new install. I can actually just keep working where I left off.

>>109900468
>Wrote a do while loop in java
nice jobbing sirs

>>109900447
>Giant commented out section that's larger than the uncommented code
Learn to use VCS you fucking retard

>>109900464
R stands for "Reddit", because this is the reddit thread.

>>109900467
>not code
oh, so you mean youre just straight up trolling because you feel like annoying everyone. ok.

im curious- why?
are you a maidshitter?

>>109900474
fuck vcs w/a stick
wtf is it, even?

>>109900480
Nobody cares about your fizzbuzz-tier toy C programming 101 garbage. You are a redditor.

File: snailcat-laughing.png (1.52 MB, 1254x1254)

1.52 MB PNG

>>109900517
>t. nocode
dw, i know you dont care about code
you made it abundantly clear, maidshitter

what are you even doing in a programming thread, then?

File: IMG_2361.jpg (42 KB, 474x766)

42 KB JPG

main(\= puts("The table language is progressing"))

>>109900536
>autistic screeching
Silence, redditor.

File: thumbs-up.jpg (14 KB, 250x203)

14 KB JPG

>>109900539

File: manhattan cafe.jpg (279 KB, 850x1134)

279 KB JPG

>>109900539
If you get it self-hosting, are you going to try to build an OS about it? <3

>>109900562
Yes, I'm working on getting it to interface with more foreign functions not just from the C library.

>>109900275
god
what's up with that book cover? why are R faggots the most boring turbo autistic fucks on the face of the earth?

This is a general about COMPUTER PROGRAMMING not your damn paedophile fetish.

>>109900596
>>109900542
i intuit these two posters are the same person

>>109900596
You are a redditor.

R language for statistical math
sample range too small out of funds

>>109900618
R is a toy language that got completely deprecated by Python.

>>109900624
>completely deprecated
python doesn't have 10% of the dependencies needed for statistical bullshit that exist in R bro

>>109900283
Reddit spacing.

>>109900474
I didn't even notice that.
That's fucking based. I do that, too.
Have some working code, comment it out and rewrite it.

>>109900684
>waaa notice me, senpai
i did
but i chose to ignore you

you sound mentally handicapped

>>109900762
>reddit opinions

File: screenshot_2026-09-25_07:50:47.png (341 KB, 2544x1399)

341 KB PNG

>>109900334
This is the whole password generation routine. There is a part that gets a keyword from you that's erased with escape sequences like the final output is when you continue.

File: Screenshot from 2026-09-2(...).png (289 KB, 1419x1000)

289 KB PNG

>>109900728
im actually working on a lib
and instead of messing around with versioning tools and whatnot i just comment the thing out

you never know when its gonna come in handy, and here it did

i was mucking about with various ways to abstract opencl's inner workings but keeping most of the functionality of the framework nonetheless

i tried various formulas but im going back to an extremely thin abstraction layer
which the code i posted actually is

so i go back to the second version of my opencl lib and i have coding artifacts that allow me to remember how things went in 15 seconds (thats the live code. the commented out part is from the 1st version)

>>109900800
>toy project
>reddit poster

>"im working on something!!"
>open the image
>it's in C
>close the image
zero products people actually use are being built in C

>>109900800
Incredibly based.

>>109900728
>>109900800
its still gonna be more verbose than that

in the end i decided that buffer and kernel argument things are gonna be straight up opencl-like, and the only thing ill change is that i use wrapper types for the bookkeeping metadata

the idea was to have abstract "channels" for each kernel but it ended up devolving into a fucking mess
and all that because it pissed me off that i have to use setKernelArg when the kernel argument is a scalar
but i have to do setKernelArg and enqueueReadBuffer when the arg is a buffer

but then i went into the intricacies of buffers and yeah
you cannot change much to that part without either crippling opencl or making a c++ out of it

what broke the model was buffer mapping. i could pretend host side buffers dont exist (buffer thats host-side, but opencl takes care of everything) but not mapped buffers. theyre just too powerful (adresses host-side that directly map to an abstract opencl buffer. this activiates all the funny integration technologies that exist bw a cpu and a gpu like amds infinity fabric)

anyhoo
2 weeks lost but at least i learned shit

>>109900872
>>109900828
sry for the delay btw
amds v-cache vs infinity cache vs infinity fabric took some googling to sort out

so yeah
no progress recently on my lib but thats because ive been experimenting with stuff

i still intend to post it for anons to make opencl more accessible to all of you
im sure now and again youre in a situations where youre like
>fkn... i need to compute a large chunk of computation, but i dont wanna deal with the opencl autism...
>aw, fuck it, lets go with google compute and pay the 20 bucks its gonna take to run xyz on 50 cpus for 5 minutes

well
ill expose the workings of opencl and abstract the tedious shit
i think you might like the result

also shit's moderately tidy
i have a query struct you can hook into if you need finer control, at all stages
im not certain why the fuck i did that, but also i have provision for hooks where you can hook into any point of failure during the execution of an opencl program if you want to do your own error handling
even though rn its premature af
but yeah, cschizo type of development i guess
i have an idea then i implement it
then i wonder why the fuck i even did that lmao

File: Screenshot from 2026-09-2(...).png (189 KB, 874x914)

189 KB PNG

-forgot picrel again-
also the struct members are named just like the args in opencl's docs
i didnt do renaming autism. the goal is for the lib to be as comfy and discrete as possible

Erlang is the best programming language.
Learn Erlang!

tri(\n =
    result = 0
    | n > 0 : result = n + tri(n - 1) _
    result
)

File: nijisanji.jpg (392 KB, 850x1133)

392 KB JPG

>>109901032
Can you show some outputs from your table language?

>>109900983
Hey man can you kill yourself today? I keep seeing you post your retarded code and I'm a little bored, so if you can wrap your life up I'd appreciate that.

>>109901045
>t. nocoder

>>109900800
You talk like the biggest faggot on the entire planet. If I could just put my hands around your little throat and squeeze, I'd be so fucking euphoric.

File: pepe.jpg (28 KB, 460x320)

28 KB JPG

>>109901045
>t jealous
if you werent such a sad faggot i could explain everything to you

when will we be free from the tyranny of memory access time

>>109901065
when you think about it
youre saying that tiered memory is bad...

youre also saying that it sucks that all the needed memory doesnt exist in the same proximity to the cpu
bc the hard limit is the speed it takes for a signal to go down a wire

>>109901072
what if the memory was more compressed e.g. not binary, more 3d, or quantum mumbo jumbo

>>109901097
i immediately thought about it thats why i mentioned cable length
your cells still have an xyz coordiante somewhere
unless you deliberately arrange your cells in a sphere, one layer thick, we are gonna have slower or faster memory

unless your whole cpu exists in quantum mumbo jumbo, but then all bets are off, init?

even if in practice, surely there will be a point where were gonna have enough cache to run everything a normal person needs
kinda like
ssd->cache directly, then you go from there bc you have 10 Tb cache bc youre on 0.00000001 nm or something
physicsfags feel free to chime in, theres surely physical limits to how small we can go...

Daily future manual laborers thread

>>109901038

tri(\n =
    result = 0
    | n > 0 : result = n + tri(n - 1) _
    result
)

_printtris(\n, i =
    | i < n : (
        printi(i)
        prints(": ")
        printi(tri(i))
        ln()
        _printtris(n, i + 1)
    ) _
)

printtris(\n = _printtris(n, 0))

main(\= printtris(20))

Output:

0: 0
1: 1
2: 3
3: 6
4: 10
5: 15
6: 21
7: 28
8: 36
9: 45
10: 55
11: 66
12: 78
13: 91
14: 105
15: 120
16: 136
17: 153
18: 171
19: 190

>>109901140
imagine being so uninformed as to think manual labor is a saving grace...

>>109901145
>print i
>print s
beware of c++ and duplication of possibly overlapping interfaces

why not print + a format string?
or type cognizance?

>>109901165
>why not print + a format string?
>or type cognizance?
Because I haven't implemented a printf clone or function overloading yet. Right now I'm trying to add tablecopy() and use references as much as possible anyway to avoid becoming C++ and making a bajillion copies everywhere.

File: clin-d-oeil-2.jpg (60 KB, 800x534)

60 KB JPG

>>109901178
100% fair
after i posted i remembered youre building as you go

File: clin-d-oeil-2-thumbs.jpg (60 KB, 800x534)

60 KB JPG

>>109901178
sory wrong picrel
i misremembered that in that one it the guy did a thumbs up

Why does vscode look so fucking good. Damnit I wish it wasn't a huge piping pile of slow shit

If VSKode is so good why is there no VSKarlie?

I use Zed because it's written in Rust and 4chan is wrong about everything.

>>109901294
g is right about everything and i use... vscodium... bc i do, okay?

File: screenshot_2026-09-25_09:32:04.png (75 KB, 1175x603)

75 KB PNG

I stole some ascii art.
fuck the police.

>>109901311
snek. i wanna pet.

File: this thread.jpg (232 KB, 850x555)

232 KB JPG

>>109901145
This looks nice. Thank you for showing me <3

im too drunk to code btw
how do i deal with that?

all i have to do is to implement 4 functions
set_arg
create_buffer
read_buffer
write_buffer

but i cant be mentally afk while doing that

wat do?

File: IMG_2367.jpg (131 KB, 1200x900)

131 KB JPG

>>109901309
Based. Disobeying 4chan's obsession with normal vscode.

File: THE-face-when.png (115 KB, 245x251)

115 KB PNG

>>109901393
wtf
4keks is into vscode?
why would someone use vscode if they can use vscodium?
i dont get it, mang

File: oai-influencers.png (352 KB, 1012x595)

352 KB PNG

(its the exact effing same but opensourced so youre sure youre not feeding m$ft your code through whatever proxy that could exist. bc its opensores'd. and dozens of people look at the code. thats more than one person, when was the last time you checked for vulns in source you dled, faggot?)

its actually GlassIt, isnt it?
it doesnt work on vscodium, but you can achieve the same using devilspie
its kinda comfy when you have 10.000 windows open, but its optional to be completely honest

File: Screenshot from 2026-09-2(...).png (1.23 MB, 3127x998)

1.23 MB PNG

>optional
i press the windows key and i see... windows
glassit improves my throughput by 0.00001%
not even worth it to make things work

i could, and did use devilspie
but the improvement isnt worth even that

>but the improvement isnt worth even that
imho that is
you do you
im just opinionated, i can also be wrong

File: 1761151233364327.png (84 KB, 739x982)

84 KB PNG

https://leetcode.com/problems/brace-expansion-ii

Why doesn't /g/ ever talk about C++? It's always just the one brown jeet screeching C is god because he learned it first in India.

File: IMG_2370.png (148 KB, 400x225)

148 KB PNG

This is now a C++ thread.

>>109901836
u sound angry DOE
also somewhat mentally handicapped to be quite famalam with you
who would wanna engage with you?

>>109901860
>ad hominem
>ad hominem
>ad hominem
Jeet stfu. You're obsessed with me.

>>109901871
no, its just that you sharteens are obvious as fuck and low iq
>shit up g
>but this is still the place you come for genuine discussion
commit sepulchre using a broken bottle you unironical homofaggot

Every time I try to write C/C++ I'm astounded by how poorly conceived the #include/header file system is. Truly astonishing that most of the world's software was written this way.

File: IMG_2373.png (171 KB, 421x314)

171 KB PNG

>>109901881
>weird schizo flurry shitstorm of verbiage in the name of c

>>109901898
>t.hates having a table of contents coming with the book
...

>>109901898
At least C++ considered adding modules.

>>109901836
Use case for C++?

>>109901911
go back to your python, you

>>109901912
Its hand warmers cover up the footguns. I only had one footgun happen to me and it was using a reference to a vector element.

>>109901915
Jeet stfu

>>109901911
And no one uses them.

File: Screenshot from 2026-09-2(...).png (37 KB, 394x369)

37 KB PNG

>>109901925
>>109901928
>t. vector is a self resizing heap array
nigga, please...
youre embarassing the family w/y shit

>>109901944
I'm glad to be an embarrassment to your home nation of India.

>>109901963
>t. seethink bc filtered

>>109901979
Filtered by what? Not using references to vector elements anymore?

File: oldman-shrug.jpg (9 KB, 250x203)

9 KB JPG

>>109901990
no, filtered by making interfaces
you just cant do that. 404.
so you want prefabs

and this is the essence of c++.
c, but with normalized interfaces, enforced by the standard

think about it

are debuggers are a meme? programmers in the '80s were so constrained by the specs of their computers that there was no spare memory to run a debugger or even print out memory locations of variables/functions and they still managed to write software. carmack swears debuggers are the most important tools available and relies on them heavily but i know he remembers when they didn't exist.

>>109902039
I don't use them for C/++. In C I just suck it up. In C++ I use .at() because I'm not a masochist.

>>109902039
>carmack swears debuggers are the most important tools available and relies on them heavily but i know he remembers when they didn't exist.
this should tell you everything you need to know

valgrind is cool. it tells you where your program segs and where the nearest memory comes from

>>109902047
i only ask because i want to write k&r c on a c64 and there's no debugger at least with a native compiler. some languages have a REPL like forth but i'm afraid the weird syntax will slow me down more than not having access to a debugger.

>>109902047
>In C I just suck it up
that equates to move an fprintf(stderr, ...) statement until it stops showing

why bother. use valgrind. you gain nothing from doing it by hand

>>109902039
I can't write assembly with a debugger because I need to watch the registers.
I used gdb to step through C code to reduce the mental load sometimes.

>>109900998
but i haven't reproduced yet
would learning erlang lower or increase those chances?

C++ and Rust exist so you don't have to install training wheels on C. Just use a real language instead of autism.

>>109902049
my options are programming in c with no debugger, programming in a language with arcane syntax but a feature rich development environment including a REPL (Forth) or programming in a language that is really unforgiving but i can inspect the program execution/memory in a machine language monitor (assembly).

which path would choose? each is arduous in its own way.

>>109902072
honestly
if you use gdb for c
youre either debugging compiledshit like drivers
or youre a fucking autist and should immediately start getting bullied

how would fukken c code become more legible when you look at the asm, nigger?
how would reading fucking hex be more legible than reading the equivalent chars?

i think gdb is for when you deal with inline asm and shit
as far as c goes, more strictly, gdb is a waste of time

>>109901836
hobbyists like C for the joy of programming. there is no joy to be had in C++, it is universally hated, including by C++ developers and pOOP is for pea brains that can't see the forest for the trees

>>109902093
if i can do c w/o inline asm, i go with c w/o inline asm
just because then 99% of whatever the exotic fuck youre working with is abstracted away

if theres inline asm involved then i probably would go full asm because i would need to check how the c abi interfaces with what im writing.
unless its like, < 15% of the code
limited entrypoints means limited places where the two things interface

>unless its like, < 15% of the code
unless *asm* is like, < 15% of the code

also
>c, forth, or asm
did you have boughtered the weird cpu?
i memeber some anon mentioned it and i wanted to try it out
dont remember the name though

>>109902094
>how would fukken c code become more legible when you look at the asm, nigger?
95% of the time you're not doing that with a debugger. Do you seriously not understand you can step through code with debug information and verify its doing what it's supposed to and that has nothing to do with looking at assembly output? It sounds like you're a complete novice screaming from the peak of mount stupid.

>>109902094
You have never used a debugger in your life.
Do you even know what step through means?

>>109902106
C has literally no advantages over C++

>>109902152
>It sounds like you're a complete novice screaming from the peak of mount stupid.
i might be

i legit never needed anything else than using unbuffered printfs to be completely honest with you
i saw gdb
i played with it for 5 mins then wondered wtf would i even need that

and legit, i never missed it
i see no usecase for it unless im dismantling what im linking with

but then yeah
i can code wtfever i want
but im not a turbo wizard who can fix everything there is

thats why i say, maybe im screaming from mount stupid
but in the context of c i really dont see the need for gdb

>>109902184
so you don't know what you're talking about and you think disassembly is the only form of debugging a debugger can offer, of course you don't like it. you are doing it wrong.

compile with -ggdb next time and learn how to use a watch window. and use a gdb frontend like gf2. gdb is just a shit debugger in general and linux has always sucked in this regard but anyone who comes from a windows environment is going to tell you how incredibly important debuggers are.

>>109902214
ok, i work on loonix, but if i can get any of the names you mentioned to work, ill try them
but legit
i think this is gonna be a whole detour to gain percentile amounts of devtime because i dont need to printout my buffers that often

>>109902182
it has simplicity and compile speed. it keeps you in the mindset of just doing the simple procedural thing. C++ has nothing half measures that don't solve anything and footguns.

>>109902230
>because i dont need to printout my buffers that often
You genuinely don't understand the purpose of a debugger. It's about verifying the state of your program at any point. If you write new code you should be stepping through it to ensure it's actually doing what you want. Right now you are programming like a blind rat and you can only peek in to the state when you printf a value.

>>109902254
>stepping through a program, verifying states
i dont need a debugger for that
that happens at the design stage

its exceedingly rare that i need to slow down and go through the program step by step

>nuh persoonal attack
i guess im superior to you, then.
i always programmed like people did in the 80's and im better for that.
kneel, faggot

>>109902254
I make a mental simulation of the entire sequential run of my program by checking one printf thoughbeit
debuggers are obsolete-by-design

>>109902274
>>109902282
no one on the planet is capable of making a complete mental model of their program, including carmack. you are just guessing and assuming it will behave as you want until you actually verify it. if this weren't the case, code wouldn't have bugs, people wouldn't write unit tests. they wouldn't need printf for debugging in the first place. you wouldn't need compiler error messages because you'd just get it right the first time. that's fantasy world.

>>109902298
carmack didnt invent fisr.
and last time i had a serious error was when dealing with a lib that takes a void * and for some reason it was the only place in the api (opencl) where i was supposed to pass a reference, not an object
and since i was interacting with a separate device the error made no sense bc there was no shared state

algo was flawless. as fukken usual
it was a type error

when you do your job properly
you work shit out on paper before even comitting to one line of code

ofc then you iterate, but keeping your ducks in a row is not a superhuman ask

>>109902298
>no one on the planet is capable of making a complete mental model of their program, including carmack. you are just guessing and assuming it will behave as you want until you actually verify it.
Of course not. Hence the printf. See Death Note: L, Anonymity & Eluding Entropy, for example, or Quantum Mechanics. A single printf decreases the possible program states dramatically by several bits and collapses the superposition. The bug cause can then easily be deduced from that much.

OMFG >< I HATE YOU

>>109902184
not anon, but I agree with you.
debuggers don't debug, they only inspect.
address sanitizer (and the rest of the sanitizers/tools) are the only real debuggers that do real work.
But after you start enabling runtime checks, that's when you start running your code inside of a debugger, because if an error is found I want to see where it happened (the alternative is to open a core dump, which takes too much time on linux since you can't just double click on the .dmp coredump, unlike windows).
But on windows I only do mini-dumps which only contain a stacktrace, maybe registers (because address sanitizer creates +500mb uncompressed dumps, which takes a few seconds to finish writing on my ssd, On linux asan disables core dumps).
To enable crash dumps on windows, I need to set the "LocalDumps" and "CrashDumps" in the registry. You can set it to full dumps and I think there is a mode that only captures references of memory only on the stack (but I remember being annoyed by it). Also if you were using msvc, the "Release" build has no function info (unlike linux unless you set the default visibility to hidden/strip) so you won't even get a stacktrace. So enabling debug info with /DEBUG or cmake's RelWithDebInfo or CMAKE_WINDOWS_EXPORT_ALL_SYMBOLS can fix that.
And it's tempting to have a in-process SEGV/crash handler that just prints backtrace_symbols, which is ok on linux I think? but on windows the most important crashes tend to turn into "fastfail" errors which can't be handled without a out-of-process debugger, for example double-free/mem corruption from malloc/free is a fastfail, and /GS stack-guard, and stack overflow's are hard to handle without causing another stack overflow (I use a fiber).

>>109902039
>A year or two after I'd joined the Labs, I was pair programming with Ken Thompson on an on-the-fly compiler for a little interactive graphics language designed by Gerard Holzmann. I was the faster typist, so I was at the keyboard and Ken was standing behind me as we programmed. We were working fast, and things broke, often visibly—it was a graphics language, after all. When something went wrong, I'd reflexively start to dig in to the problem, examining stack traces, sticking in print statements, invoking a debugger, and so on. But Ken would just stand and think, ignoring me and the code we'd just written. After a while I noticed a pattern: Ken would often understand the problem before I would, and would suddenly announce, "I know what's wrong." He was usually correct. I realized that Ken was building a mental model of the code and when something broke it was an error in the model. By thinking about *how* that problem could happen, he'd intuit where the model was wrong or where our code must not be satisfying the model.
>Ken taught me that thinking before debugging is extremely important. If you dive into the bug, you tend to fix the local issue in the code, but if you think about the bug first, how the bug came to be, you often find and correct a higher-level problem in the code that will improve the design and prevent further bugs.
>I recognize this is largely a matter of style. Some people insist on line-by-line tool-driven debugging for everything. But I now believe that thinking—without looking at the code—is the best debugging tool of all, because it leads to better software.
I can't use debuggers on distributed concurrent systems.

>>109902363
>you work shit out on paper before even comitting to one line of code
too slow, just keep writing code and testing it until it works

>>109902363
>when you do your job properly
AI can do your job properly

File: Screenshot_20260925_071525.png (409 KB, 2690x1524)

409 KB PNG

>>109900275
>What are you working on /g/?
Working more on my RISC-V assembler.
I get a feeling of this being pointless as hell, but I am so close to make it work. I should get back to Rust OS dev maybe

File: Screenshot_20260925_081159.png (485 KB, 2690x1524)

485 KB PNG

>>109902952
Made the label lookup with a simple loop now. By rewriting this I think I found my error with the hashmap handling, but whatever.
This should work now. Lets generate some machine kode, get that blinky blinky, then improve

File: file.png (246 KB, 2514x2268)

246 KB PNG

Dealt with Section and Theme Flags creation. Kind of a pain to set the media pipeline, even after it was theoretically done, but all set up and proven working now.

>>109902079
100% lower them. Erlang comes with a lifetime guarantee of unemployment, childlessness and celibacy. Learn erlang today!

>>109903189
ok made the code generation.
It also compiles, but i gotta prep for interview now. REEEEE I DONT WANT TO HAVE AN INTERVIEW

If I vibe code a software, or pieces of it, who owns the copyright?
Especially since you know all the AI models were trained on pirated data, so who knows whos code it might spit out at me?

>>109902184
>>109902214
Reddit spacing.

>>109903849
>If I vibe code a software, or pieces of it, who owns the copyright?
No one. It's not covered by copyright.

I'm pretty good at C++, I think.
I also want to learn Python, C, Lua, maybe Rust and Javascript, and an even bigger maybe for re-learning C# and Java since it's been so long since I used them.

What kinds of projects would you start with?

>>109903911
You should start with a reddit clone because you are a redditor.

>>109903935
I'm not too into web development. I know it's a marketable skill, but I've never really resonated with it the same way I have with the underlying logic of C++

>>109903911
Rewrite all your projects in each language.
Or more realistically, just pick a language and write your next project in that.
You do have some idea of what kinds of projects you want to be making, right? Right, Anon?

>>109902072
>I can't write assembly with a debugger because I need to watch the registers.
I mean, with gdb you can do either something like:
display/4i $pc
display/16xg $sp
display/x $rax
...
display/x $rdx

or:
define step_asm
stepi
printf "eax=%d ... edx=%d\n",$eax,...,$edx
x /4i $pc
end

or:
define hookpost-stepi
printf "rax=%016lx ...\n",$rax,...
end

and then just:
break *0x400100 # or whatever the address of your entry point or point of interest is
run
stepi # or step_asm, if you used the second option

there's also a "layout asm" specifically for assembly, but I haven't really used it (although I haven't used hookpost myself, either)

and of course there is "info registers" which displays all the registers, but only does it once unless you combine it with the second or third option.

>>109900275
Manually setting up NeoVIM from scratch without using plugins or kickstarters. It's also my first time working with NeoVIM.

I decided to hook it up to an existing tool chain bundle and found out that's actually excruciatingly painful to do vs. just hooking it up to a clean/proper install. It completely spiraled out of control and ended with compiler diving to get clangd to work. Took me 4 hours to figure out while also learning LUA.

Which btw, coming from C++, is an absolute slut of a lang lol.

>>109903997
all of my what now?

really though, I should probably try just writing some games in each of these and see what happens. I've been trying to come up with portfolio filler that I can put on a github to show to prospective employers, and the only thing I can think of that might be actually usable is an image viewer that's a bit more well suited to my porn viewing needs, and my own take at a music player gui since Foobar2000 isn't on Linux natively, and DeadBeef breaks more often than it opens successfully.

>>109904056
>all of my what now?
Oh. In that case write hello world in each language then shut down your computer and go outside.

>Portfolio is an image viewer and a music player
This is a meme, right?

>>109904085
It's more so that I just haven't put any effort into writing any presentable code for anyone else. I've been only focused on learning and trying to create half-formed projects for myself.
Those ideas are just the only ones I could remember thinking of that don't already exist in a better form. Part of me knows that I just need to make anything even if it already exists, but another part actually wants to make useful software that people would willfully download.

>>109904140
>that don't already exist in a better form
This doesn't matter if you need a project to practice a language or other technology. I keeps saying there's a satisfaction in using tools you've made yourself.

>actually wants to make useful software that people would willfully download
This conflicts with your desire to learn new languages.

>>109904216
yep, quite the mental paradox, ain't it?
No wonder I haven't got a portfolio yet.

>>109904293
Have you ever heard of Buridan's ass?

>>109904018
Do you know what gdb stands for anon?

my hyperfocus broke

>>109901898
it's pretty good yeah

>>109904355
sure. don't know what that has to do with anything tho. (well, unless you're the kind of person who will go "hur hur, he said 'WRITE assembly WITH a DEBUGGER', obviously you can't do that", blatantly ignoring the "because I need to watch the registers" which quite clearly shows that it was just a figure of speech - he clearly wasn't *literally* trying to write assembly with a debugger, just trying to use a debugger *with* code written in assembly, but not knowing you *can* in fact inspect the registers with it -- but hopefully that wasn't what you were trying to say; I'm just including this here as a pre-emptive response if that was the case, since I don't really want to have a long back-and-forth, since replying in 4chan is a bit of a chore nowadays - it doesn't like my ip range, so I have to move to another device with a mobile connection to comment at all, and then deal with the annoying captcha on top of that)

I currently work on shitty old desktop apps and I want to break out of this dead end niche. What books/subjects do your recommend?
I am planing on doing the following:
>In depth angular course
>Some stuff on spring boot or asp .net + in depth API design
>Reading designing data intensive applications.

>>109902432
if it works for you, then it works
to me, going through registers and asm would be another layer of abstraction from the code im writing + i barely can read asm
to be efficient i would need to know which register maps to what variable

printing out the values of my variables with fprintf(stderr, ...) [its unbuffered then] works just fine for me
also in knowing where the error actually happens part (as opposed to where its caught)

my sop is that i identify the offending variable, and then i follow the changes to its value throughout its lifetime. most often starting where the error is caught, going backwards
once i find where the value starts to deviate from whats expected in the algo, thats where the bug is
fprintf is enough for that

>>109902906
might work with higher level languages
but in c thats the best path to code yourself into a corner time and again, and cut down your total efficiency by 90%, esp once human factors kick in, like burnout

i dont do things "properly" myself anymore, because im a lazy sack of shit with an exceedingly high self esteem
but im still close. i still figure out the general layout and interfacing of my project first, and then i run the part im currently working on, on paper (if justified)
only then i start coding

in c you have to define the memory layout and storage methods
and if a refactor goes deep enough, its oftentimes more productive to just rewrite the thing in another version, salvaging all the code that remains unchanged
spaghettified c is extremely hard to manage, and it becomes a time drain that compounds for as long as its not fixed