the stack trace isn't even that suspicious!pic.twitter.com/jhr3RrCXB6
U tweetove putem weba ili aplikacija drugih proizvođača možete dodati podatke o lokaciji, kao što su grad ili točna lokacija. Povijest lokacija tweetova uvijek možete izbrisati. Saznajte više
it's trying to free a bit of memory that belongs to a region mmapped from a file whatpic.twitter.com/Tpfszoc8m5
ooh, LLDB shows more deets, looks like it crashes when dropping `delf::ProgramHeader`pic.twitter.com/UeSoWrYsSG
my current working theory is that one of my `mmap` calls is accidentally unmapping a previously-mapped region used as backing memory for a `Vec`, and when dropping the `Vec`, well, all hell breaks loose This only happens with large enough Vecs, so it didn't happen beforepic.twitter.com/QGyNt7TpiY
lessons learned:
- don't write articles while extremely tired
- mmap with MAP_FIXED is *way dangerous*
- using LLDB for day-to-day stuff sounds worth it 
Okay, after some additional bug-hunting, https://fasterthanli.me/blog/2020/dynamic-symbol-resolution/ … is all fixed up and our codebase should *not* blow up the minute we run it on real-world executables
There was another subtle bug btw: malloc/free misuse is SO EASY to identify as compared to mmap misuse. The memory manager will happily unmap crucial stuff from under you and it'll blow up way later.pic.twitter.com/Z4UANRnq6g
The worst part is that you can't really write a sanitizer for mmap mis-use because remapping regions is 100% legitimate usage and something the dynamic linker does a bunch
tl;dr we need a "unfuckingsafe" keyword for the https://crates.io/crates/mmap crate
Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.