mov rax, [SIZEOF_SOCKADDR] segfaults while lea rax, SIZEOF_SOCKADDR does not 
make[2]: *** No rule to make target 'release/libddnet_engine_shared.a', needed by 'DDNet.exe'. Stop.
make[1]: *** [CMakeFiles/Makefile2:500: CMakeFiles/game-client.dir/all] Error 2
make: *** [Makefile:156: all] Error
when building using cmake .. -G"Unix Makefile" && make j(30) on windows
rust toolchain is stable-x86_64-pc-windows-msvc
python --version: 3.12.3
why is it always libddnet_engine_shared bullying me? - 
db here https://github.com/ChillerDragon/teeworlds-asmr/blob/2050b2bb80e70de58ad7a86e62db399345b564e6/src/teeworlds_asmr.asm#L98 but recvfrom takes pointer to an int, so it was overwriting ADDR_LOCALHOST
your fix works perfectly for me! thank you so much @MilkeeyCat
git worktree it's quite nice. I have some branches permanently checed out in their own folders
-i along with the onto, then I drop the commit I think is duplicated
git checkout -b mybranch
git commit
git checkout -b tmp origin/main
git cherry-pick mybranch
git branch -D mybranch
git branch -M tmp mybranch git rebase -i?git rebase -i? 
rustup target add x86_64-pc-windows-gnu
installed? rustup target add x86_64-pc-windows-gnu
installed? 
rustc --version --verbose please
1
rustup default x86_64-pc-windows-gnu or rustup default x86_64-pc-windows-msvc, I think 
rustup default stable-x86_64-pc-windows-gnu 1
1
async).
Do you think it's actually worth to have a third database?
Why wouldn't you simply always write to the sqlite database (similar to what is now done in the backup table).
I can only imagine the risk of having duplicate ranks, but is that really a huge problem? shouldnt it be easy to detect it with a timestamp or smth.
Or was it mostly bcs it was easier to deal with like that?async).
Do you think it's actually worth to have a third database?
Why wouldn't you simply always write to the sqlite database (similar to what is now done in the backup table).
I can only imagine the risk of having duplicate ranks, but is that really a huge problem? shouldnt it be easy to detect it with a timestamp or smth.
Or was it mostly bcs it was easier to deal with like that? src/game/server/. Why is such a high level of the code even aware of threading? game/server just needs a huuuge refactor in general xdsaves/ranks service. But that only works with writing, with a read idk how the router that's supposed to be message agnostic figure out where to route the read message CJobs that can run somewhere else
src/game/server/. Why is such a high level of the code even aware of threading? 
2#[repr(C)]
1
One important step will be that our devs can compile the ddnet-libs themself (e.g. for updates)
One important step will be that our devs can compile the ddnet-libs themself (e.g. for updates) 
