Dan Klishch

I've reevaluated my life choices and decided to go with Tim's simpler version.

crti.S should be deleted. It is not used by Clang since (my) 19.1.0 update, and we should just do the same thing for GCC. (Ref: 869d71fe19b4b05065a34e5f5c56301428230aa9)

Hmm, @ADKaster and @spholz liked new pointer-to-member formatting and assembly formatting, respectively. Personally, I don't really care about pointers-to-member or empty blocks (but I like that that strange `{};` construct...

~~The empty block changes are due to a bug fix that made `SpaceInEmptyBlock: true` from WebKit preset actually work. New rule in clang-format and empty blocks revert incoming.~~ See discussion...

> Maybe it'd be nice to omit LibWeb here. CI won't let me do this, and then it will yell at you every time you change a file with wrong...

The test case passes out of the box with AsyncDeflateDecompressor from #24567.

@MarekKnapek, it would be great if you "resolved" review conversations (clicked on the "resolve conversation" button) that you've fixed—this greatly reduces cognitive load when one tries to figure out what...

Now with #24668 finally merged, can you make this use the same `AK/CPUFeatures.h` / static member function pointer machinery? I think this will cover most of my very nonconstructive "feels...

> getting http/3 support implemented in Ladybird Not exactly. Ladybird can just yoink some third-party library and call it a day. Supporting (asynchronous version of, as opposed to callback hell...

Hmm, this "Can't keep up" spam might be a separate issue. Can you figure out which timer causes it? Since I don't think there is a time-travel debugger for MacOS,...