amos 16M Nov 25 13:03 /tmp/pdftocairo.compressed-debug $ objcopy --strip-all ./target/debug/pdftocairo /tmp/pdftocairo.stripped $ ls -lhA /tmp/pdftocairo.stripped -rwxr-xr-x. 1 amos amos 5.3M Nov 25 13:04 /tmp/pdftocairo.stripped During my tenure as “chief fucking around and finding out officer”, I’ve built software from sources “manually

B801000000 mov eax,0x1 00401019 0F05 syscall 0040101B 4831FF xor rdi,rdi 0040101E B83C000000 mov eax,0x3c 00401023 0F05 syscall Executing "samples/hello" in memory... code @ 0x561828e69000 entry offset @ 00000000 entry point @ 0x561828e69000 Fuck. I’m pretty sure hello was supposed

If only someone… contributed that feature upstream… Oh no no no not ag- Looking into gtk-rs/gir …fuck’s sake. So the first wrinkle is that, at the time of this writing, there’s gir 0.14, and gir’s default branch, adn they generate incompatible code. So we’ll need to regenerate all crates. Weren’t we planning on regenerating all of

proprietary codebases solo just so your blog slash video platform is juuuuust the way you like it. So instead of buying a fuck-you CPU (like a Threadripper, or something more consumery like the latest Ryzens), maybe you rent a big cloud machine that you can turn on and off as needed. Just for the big stuff. If you need a bunch of CPU to work

joy. What the fuck is happening? …is a fair question, considering. See, the thing with match is, it lets you do pattern matching. As opposed to if, which just does equality comparison. That means with match, we can do something like this: enum Node<T> { Left(T), Right(T), } impl<T> Node<T> { fn get(&self) -> &T { match self

hello-pie zsh: no such file or directory: ./hello-pie WHAT? That’s it. I give up. I quit. Thanks for the support everyone, but I can’t take this anymore. Nothing makes sense, up is down, files exist and are executables but they won’t execute, fuck all this, I’m out. Hey what’s that “interpreter” part of file’s output? I don’t know

assertion failing is: // (in tokio code, in `src/time/driver/entry.rs`) debug_assert!(cur_state < STATE_MIN_VALUE); Huh, a debug_assert!. That means that in production… $ RUST_BACKTRACE=1 cargo run --quiet --release (program never outputs anything, never exits either) Hah. I guess they’re not fucking around about the “undefined

a write method so you can just write anywhere in memory is very much in the spirit of Rust. Sure. Well, no, not that part, but for example.. Oh yeah, sure, in debug builds Rust will check that when you add up two random numbers you’ve read from fuck-knows-where in a poor ELF file, they don’t underflow an u64. Oh yeah. Well done. Bravo. Super

need to reorganize all my manifests carefully. But I was already down, and it’s been a shit month, so, I decided to treat myself. The deploy-manifests script, even in its latest incarnation, still blows. A better version would merge all resources in a single YAML file, and compare against what the control node already has, requiring --fuck-me-up

is still a warning - just something we should look at when we get a chance, but not a showstopper. (Unless we slap #![deny(unreachable_code)] at the start of our main.rs, the equivalent of passing -Werror=something to gcc/clang). Fuck around now, find out… when? Let’s change our sample a little bit. Say we remove the definition of bar entirely