Let's Create a Compiler (Pt.1)

When he searches for the size of rdi the only thing I could think of is how easy chatGPT made those kinda small searches easier

This is a great video and a nice footage, just the solving how C++ works is too much, better to focus on the ASM, since C++ is not interesting here.

You used to be Edward Snowden right? Jk; you think similarly to me; subscribed

i dont know who is learing me or he

This dude really just woke up and decided to make tutorial with out prepearing and using language he don't fully remember.

Rust will die off like every other language that went up against
C++.

I has a question, as Assembler aren't the same depending on the os,
how would you make the compiler compile the program on all os (without making a different compiler for each os)

I thought of compiling the code to a higher language like C and then letting the C compiler compiler it to assembler but it doesn't feel really right

IOS stands for input-output string...

I’ve been missing this kind of content for a while. This is great, keep it up!!

Don’t know anything about C++, assembly or compilers but watched this from start to finish. Great video

"If your ide isnt using 10g of ram" the jetbrain fanboy in me in feeling called out 😂😂

I may have felt in love
finally, someone to talk about compiler design

good

it is happening again, someone ask how to open files in C++ and ended up creating a new programming language

cool

Bro you did mute you mic, but you forgot to mute your arm’s movement 😂

Omg you have no idea how happy it made me to see you forgot the argc argv stuff. I literally always forget that, LOL

"Ladies and gentlemen we have out own programming language, which we can use to return any exit code that we want" 😂😂😂😂😂😂

😂

bro says vscode is for soys and then opens up even more soy PAID CLOSED-SOURCE IDE

Has anyone ever told you that your vocal patterns are a mixture of Tsoding and General Sam?

This was very entertaining, now I want to try writing my own compiler for memes. Keep up the good work, I’m subbing for sure.

Way cool 😎

I don't know if this is unique to the STM microprocessors we use at my job, but the primary function of our linker script is organizing memory. It was explained to me (maybe not explained well, or maybe i misunderstood) that the linker takes all the variable names and translates those into RAM addresses and makes sure the rest of the code doesn't accidentally overwrite them.

The term libraries never even came up during that explanation.

The C++ destroyed quality of this video

Vscode is for soys. Enough said

truly one of the programming moments of all time

bro you're a brilliant

C++ eveything

Sup, I know this is kinda newbie but how can I install nasm and gcc on ubuntu wsl?

Great series btw. Just submitted a compiler for class using llvm-ir code and was pretty fun. Its a horrible compiler that barely works but it works noneless!

All great coders know that it still counts as first try if the previous tries all errored out

ChatGPT
The size of the rdi register depends on the architecture of the processor. In x86-64 architecture, which is commonly used in modern desktop and server systems, including most PCs, the rdi register is a 64-bit register. This means it can hold 64 bits or 8 bytes of data.

In general, registers in computer architecture can vary in size based on the specific architecture. Common register sizes include 8-bit, 16-bit, 32-bit, and 64-bit, with the size determining the amount of data they can hold at once. The rdi register, as part of the x86-64 architecture, is a 64-bit register.

I love the chaos of all this...fumbling to read a file with the end goal of writing a compiler is majestic haha

Does it translate it to assembly language or machine language?

Im not even a programmer but i watched through entire stuff great video man!

As I do not know C++ I think C++ is hardly 10 levels higher than assembly language whereas python is 10000 levels higher than assembly language.

someone able to run ld test.o - o test on mac? please help me i am getting error like this
ld: warning: platform not specified

ld: warning: -arch not specified
ld: warning: No platform min-version specified on command line
ld: warning: ignoring file test.o, building for -unknown but attempting to link with file built for unknown-unsupported file format ( 0x7F 0x45 0x4C 0x46 0x02 0x01 0x01 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 0x00 )
Undefined symbols for architecture unknown:
"_main", referenced from:
implicit entry/start for main executable
ld: symbol(s) not found for architecture unknown

Thank you very much. ❤❤❤ THANKS GOD AND THE VIRGIN MARY AND SAINT JOSEPH. ❤❤❤❤❤ Happy thanksgiving to Pixeled and your family and Everyone. ❤ 25.11.2023 ❤ 6h23 ❤ 14.12.2023

Why is this so entertaining

doesn't ios stand for input output stream?

and this is why high level languages exist :D

I have many respect from that guy, he writes asm for fun

this is great stuff, carry on

One underated aspect of this kind of videos is that making mistakes is actually good to keep in the video instead of editing it out. Really makes it engaging and relatable as if we're doing it together. Great content !

the never nester in me is screaming at those functions

CMake is crap in my opinion. Linux folk talk about it a lot. I guess they use it a lot. On Windows, I've only ever needed it for open source stuff and the nashing of teeth that follows is painful. It usually fails big time. It's dog awful. The Linux way of doing stuff akways seems to include unnecessary complexity. I guess this is designed to turn one's hair white prematurely? And since mine is mostly white already, Linux is now in my no go zone.

My approach to tokenize or separate out a stream was to move from beginning to end, setting the type to whatever it was I was currently "eating," like "now an number" or "now a string" or "now a whatever." That type had expectations as to what should come next and any new character that did not fit that expectation would immediately terminate the token. Since the token and its type were already being formed as part of the eating process, they would be "closed off" and I would then prepare for the next one. The next one would begin with any character that satisfied the rules for any known type.

To test against the current key, I did not do a series of nested if-thens, but more a dictionary type of things with the key being the type and the valid characters comprising the value, like a "semicolon" entry would be just ";" for example. Or a number key would be have the value "01234567890.," or whatever you considered to be valid for a number. Then you just do a check to see if the value contains the current character. If it doesn't, you close out and then the more complicated process of locating the next key begins via comparison to value contents. There are surely ways to optimize it.

I like what you’re doing in this video but fu$& dude… watching you struggle to know how to even put rags in your c main function. Groan. You really need to know the basics and if you can’t get that part right do some prep in advance.

Umm, we all know what’s on his mind eh. The exit code is 60 but he keeps saying 69.

Where are you bro please come back and make our day better