slimsag

If existing already, delete the default k3d cluster:

k3d cluster delete

Start a k3d cluster and port forward the loadbalancer:

k3d cluster create -p "30080:80@loadbalancer"

slimsag

If existing already, delete the default k3d cluster:

k3d cluster delete

Start a k3d cluster and port forward the loadbalancer:

k3d cluster create -p "30080:80@loadbalancer"

slimsag

OS	Arch	Model	CPU	RAM	Storage
macOS	arm64	2020 Mac Mini	Apple Silicon M1 8 core 3.2ghz	8GB	512G SSD
macOS	x86	Late 2014 Mac Mini	Intel i7 8 core 3ghz	16GB	256G SSD
linux	x86	Beelink GTR5 5900HX	Ryzen 9 5900HX 8 core 4.6ghz	32GB	512G SSD
windows	x86	Beelink GTR5 5900HX	Ryzen 9 5900HX 8 core 4.6ghz	32GB	512G SSD
windows	arm64	Windows Dev Kit 2023	Snapdragon 8cx Gen 3, 8 core ~3ghz	32GB	512G SSD
linux	arm64	Radxa ROCK pi 5B	RK3588 8 core ~2.4ghz	16G	512G SSD

• Case: KENUCO SOHO Mini 10" Rack

slimsag

Proprietary mesh triangulation in open source software?

Recently I was looking for a solid, reliable way to do path triangulation. Looking at state-of-the-art research, I came across the UC Berkley researcher Jonathan Shewchuk's outstanding Triangle library.

Jonathan is very clear on the Triangle website, and in source code, about the project not being open-source:

This program may be freely redistributed under the condition that the copyright notices (including this entire header and the copyright notice printed when the `-h' switch is selected) are not removed, and no compensation is received. Private, research, and institutional use is free.

slimsag

const std = @import("std");
const builtin = @import("builtin");
const testing = std.testing;
const process = std.process;
const fs = std.fs;
const ChildProcess = std.ChildProcess;

var a: *std.mem.Allocator = undefined;

pub fn main() !void {

slimsag

https://twitter.com/DiedeApers

https://twitter.com/GPUOpen

https://twitter.com/kineticpoet

https://twitter.com/luxeengine

https://twitter.com/ajmmertens

slimsag

// avxSupport uses CPUID to check if the CPU supports AVX
// func avxSupport() bool
// Return: +0(FP)
TEXT ·avxSupport(SB),$0
	MOVL $0x1, AX
	CPUID

	// AND ECX, 0x18000000
	BYTE $0x81; BYTE $0xE1; BYTE $0x00; BYTE $0x00; BYTE $0x00; BYTE $0x18;

slimsag

slimsag@hexopsblade15:~/Desktop/beyang/symf$ CGO_ENABLED=1 GOOS=darwin GOARCH=amd64 CC='zcc' CXX='zxx' go build -ldflags='-extldflags=-F/home/slimsag/Desktop/beyang/symf/sdk-macos-12.0/root/System/Library/Frameworks' -o test ./cmd/symf
# github.com/sourcegraph/symf/cmd/symf
warning: unsupported linker arg: -headerpad
warning: unsupported linker arg: 1144
warning: unsupported linker arg: -no_pie
warning: unsupported linker arg: -pagezero_size
warning: unsupported linker arg: 4000000
warning: unsupported linker arg: --compress-debug-sections
warning: unsupported linker arg: zlib-gnu

slimsag

C:\Users\slimsag\Desktop\zig\zig-out\bin\zig.exe build-lib C:\Users\slimsag\Desktop\mach2\gpu-dawn\src\dawn\dummy.zig -lc++ C:\Users\slimsag\Desktop\mach2\gpu-dawn\zig-cache\o\6bc980b450b4b05fab4cd621a1ba86cd\glfw.lib -lc -lgdi32 -cflags -g1 -std=c++17 -DDAWN_ENABLE_BACKEND_NULL -DDAWN_ENABLE_BACKEND_D3D12 -IC:\Users\slimsag\Desktop\mach2\gpu-dawn/libs/mach-glfw/upstream/glfw/include -IC:\Users\slimsag\Desktop\mach2\gpu-dawn/libs/dawn/out/Debug/gen/include -IC:\Users\slimsag\Desktop\mach2\gpu-dawn/libs/dawn/out/Debug/gen/src -IC:\Users\slimsag\Desktop\mach2\gpu-dawn/libs/dawn/include -IC:\Users\slimsag\Desktop\mach2\gpu-dawn/libs/dawn/src -D_DEBUG -D_MT -D_DLL -- C:\Users\slimsag\Desktop\mach2\gpu-dawn\src\dawn\dawn_native_mach.cpp -cflags -IC:\Users\slimsag\Desktop\mach2\gpu-dawn/libs/dawn/src -g1 -std=c++17 -- C:\Users\slimsag\Desktop\mach2\gpu-dawn\libs\dawn\src\dawn\common\Assert.cpp C:\Users\slimsag\Desktop\mach2\gpu-dawn\libs\dawn\src\dawn\common\DynamicLib.cpp C:\Users\slimsag\Desktop\mach2\gpu-dawn\li

slimsag

WIDL compiler incompatibilities with Windows SDK

I attempted to compile the IDL files distributed in https://github.com/microsoft/win32metadata/tree/master/generation/WinSDK/RecompiledIdlHeaders using the latest (7.0.1-rc.1) version of the Wine IDL compiler, called WIDL. Unlike Microsoft's MIDL compiler which ships with Visual Studio, WIDL is open source.

More importantly, WIDL contains patches to generated header files that make e.g. COM calling conventions compatible with mingw-w64 and clang when targetting GNU ABI.

• Bugs
  • Imports are incorrectly case sensitive
  • WIDL does not allow double slashes in string literals
• Segmentation faults