esolangs & os dev
- So this is a Rust channel? 🤔*Thread
- a template for osdev
- after some more researchthe fault is actually caused by
gdt::init();but it manifests after interrupts get enabled that's whyinterrupts::enable();were susbut what exactly is the problem idk Replying tokalucky0https://cdn.discordapp.com/attachments/659572602825605146/905493866784890880/unknown.png
Ooh is that linux- Isn't gnome Linux only
- 🏳️🌈
pub fn flag(&mut self, colors: Vec<RGB>) { let h: usize = self.height() / colors.len(); for x in 0..self.width() { for y in 0..self.height() { let mut color = match self.info.pixel_format { PixelFormat::RGB => [0, 0, 0, 0], PixelFormat::BGR => [0, 0, 0, 0], PixelFormat::U8 => [0, 0, 0, 0], _ => [0, 0, 0, 0], }; for i in 0..colors.len() { if y <= h * (i + 1) { color = match self.info.pixel_format { PixelFormat::RGB => [colors[i].r, colors[i].g, colors[i].b, 0], PixelFormat::BGR => [colors[i].b, colors[i].g, colors[i].r, 0], PixelFormat::U8 => [colors[i].r, colors[i].g, colors[i].b, 0], _ => [colors[i].r, colors[i].g, colors[i].b, 0], }; break; } } let pixel_offset = y * self.info.stride + x; let bytes_per_pixel = self.info.bytes_per_pixel; let byte_offset = pixel_offset * bytes_per_pixel; self.framebuffer[byte_offset..(byte_offset + bytes_per_pixel)] .copy_from_slice(&color[..bytes_per_pixel]); let _ = unsafe { ptr::read_volatile(&self.framebuffer[byte_offset]) }; } } }
This looks right to me.#pragma once #ifndef __riscv #error __riscv was not defined, are you compiling for RISC-V? #endif #ifdef __cplusplus #include <cstdint> #include <cstddef> #else #include <stdint.h> #include <stddef.h> #endif typedef int8_t s8; typedef int16_t s16; typedef int32_t s32; typedef int64_t s64; typedef uint8_t u8; typedef uint16_t u16; typedef uint32_t u32; typedef uint64_t u64; typedef volatile s8 vs8; typedef volatile s16 vs16; typedef volatile s32 vs32; typedef volatile s64 vs64; typedef volatile u8 vu8; typedef volatile u16 vu16; typedef volatile u32 vu32; typedef volatile u64 vu64; #ifdef __riscv_xlen >= 128 typedef __int128 s128; typedef unsigned s128 u128; typedef volatile s128 vs128; typedef volatile u128 vu128; #endif #ifdef __riscv_flen #if __riscv_flen >= 32 typedef float f32; typedef volatile f32 vf32; #endif #if __riscv_flen >= 64 typedef double f64; typedef volatile f64 vf64; #endif #if __riscv_flen >= 128 typedef __float128 f128; typedef volatile f128 vf128; #endif #endif
- Only horizontal bars
Replying tokalucky0pub fn flag(&mut self, colors: Vec<RGB>) { let h: usize = self.height() / colors.len()…
You can modify this code to support different orientationsReplying togudenauI don't remember, does LLVM have an assembler?
No idea, my knowledge is near zeroReplying togudenauhttps://craftinginterpreters.com/
You know how short my attention span is? I am not able to read more than the slogan on this page- Anyone know how to either have make change directories or to have configure emit files in a directory that isn't the current working directory?
Neat.$ file build/* build/test.c.o: ELF 64-bit LSB relocatable, UCB RISC-V, RVC, double-float ABI, version 1 (SYSV), not stripped build/test.cpp.o: ELF 64-bit LSB relocatable, UCB RISC-V, RVC, double-float ABI, version 1 (SYSV), not stripped build/test.o: ELF 64-bit LSB relocatable, UCB RISC-V, double-float ABI, version 1 (SYSV), not stripped $I can't use compressed opcodes it seems. - btwwhat am I looking for? I am after few beers and it's getting hard I'm trying to do something like this:
AssertionErrorKind { InvalidArgument = "invalid argument", InvalidIdentifier = "invalid identifier", InvalidIndex = "invalid index", InvalidToken = "invalid token", UnexpectedToken = "unexpected token", UnknownOperator = "unknown operator", } - yes it is
Replying tonailujthe errors something like this... over and over
You need to specify the size of the moves.And such.Replying tokalucky0@nailuj almost 7 hours? what are you implementing there?
Walked away from the computer for schoolTimer kept running- i have no code to make that
Replying togudenauIf something doesn't look right, it is your fault.
Not if I didn't write the bootloader and its happening before my kernel is loaded- @nailuj that might be useful https://github.com/orgs/rust-osdev
- It's really good
.section .bss # Create a 16KiB stack .align 16 stack_bottom: .skip 16 * 1024 stack_top: .section text .global _start .type _start, @function _start: # Only run on a single HART #csrr t0, mhartid #bnez t0, spin # Clear satp, just in case csrw satp, zero # Setup the stack la sp, stack_top # Invoke int kernelMain(int hart, void* deviceTree) csrr a0, mhartid # a1 already has the device tree pointer call kernelMain # Spin, main returned... spin: wfi j spin
This is all of the code I have right now.#include "sys/types.h" extern "C" int kernelMain(int hart, void* deviceTree){ (void)deviceTree; if(hart){ return 0; } vu8* serialPort = (vu8*)0x10000000; // No interrupts serialPort[1] = 0; // Enable DLAB serialPort[3] = 0x80; // Set divisor to 0x0030 for 28400 buad serialPort[0] = 0x30; serialPort[1] = 0x00; // 8 bit word, 1 stop bit, no parity serialPort[3] = 0x03; // Enable and clear FIFO serialPort[2] = 0xC7; // IRQs enabled, RTS/DTS set serialPort[4] = 0x0B; serialPort[0] = 'A'; return 0; }But when I single step with GDB and qemu PC never changes.I don't get it.- https://stackoverflow.com/questions/70029004/risc-v-64gc-ld-truncated-relocation-r-riscv-pcrel-hi20 I'm getting banned again.
IT LIVES/usr/bin/qemu-system-riscv64 -machine virt -cpu rv64 -smp 1 -m 256M -drive if=none,format=raw,file=harddrive.img,id=foo -device virtio-blk-device,scsi=off,drive=foo -serial mon:stdio -bios none -device virtio-rng-device -device virtio-gpu-device -device virtio-net-device -device virtio-tablet-device -device virtio-keyboard-device -bios none -kernel out/bootloader.elf Hello world! Process finished with exit code 0- I know it is passed in
a1at least.I guess I will look into that later. ; ModuleID = 'length' source_filename = "length" define i32 @main() { entry: %stackptr = call i64 @initStack() call void @push(i64 %stackptr, i64 42) call void @lengthOutn(i64 %stackptr) ret i32 0 } declare i64 @initStack() declare void @freeStack(i64) declare void @push(i64, i64) declare i64 @pop(i64) declare void @lengthInp(i64) declare void @lengthAdd(i64) declare void @lengthSub(i64) declare void @lengthDup(i64) declare void @lengthOutn(i64) declare void @lengthOuta(i64) declare void @lengthRol(i64) declare void @lengthSwap(i64) declare void @lengthMul(i64) declare void @lengthDiv(i64) declare void @lengthRor(i64)I generated this so farIt does what is expected, prints 42Hardcoded for now- This one is my favorite
I would have to pay about $250 to be able to sell those - https://github.com/Lengthlang/lengthlvm behold, the dumbest thing I've ever made
Replying tonailujI have multiple screenshots of nft
You can absolutely screenshot my NFTs I can even DM you the originals- If you use Ethereum network then you pay about $250 gas fee to be able to post NFTs If you use the other network you pay $0 to post but you can't do auctions and you don't get a percent from reselling the NFTs you made After you sell an NFT you pay 2.5% commission regardless which network you use
Replying tonailujCan you explain how you do it
I use several NN and other algorithms to create them and then just... https://opensea.io/asset/createReplying toAzzyypaarasI will help you make them be more palatable if you give me a cut :sus:
You have a dealReplying toJack PapelInb4 kalucky starts a massive pump & dump scheme
Ligma grindset tip #3253257: commit white collar crimesReplying tokalucky0Tbf, It might be a little bit not ethical or smth But I love making money on people…
Oh, it is not ethical at all, but ethics are a lesser man's cope- I should try it
Replying toJack PapelImo if you rag on nfts, you should be ragging on crypto equally as much. They're usually…
I only rag on nft owners because they are fucking idiots- mostly the latter of the three
Replying togudenauHow the heck did this channel get to something as useless and wasteful as NFTs?
Good question, I got randomly pinged- I built an llvm frontend for it
Replying togudenauHow the heck did this channel get to something as useless and wasteful as NFTs?
Scams might be wasteful But they may also be very lucrativeReplying toAzzyypaarasNFT sellers... :balls:
some of them are even more stupid tbh flooding the market with autogenerated shit is just stupid1000 almost identical NFTs in one hour(edited)- Nft => #blockchain
Replying togudenauAlso these NFTs are going to vanish in a bit when servers die. NFTs are centralized.
ExactlyThis is a bubble
