What's the purpose of the author just listing these on their site instead of reporting the bugs to LLVM? They even seem to be actively hostile to their proposed fixes being used, given the copyright notices on pretty trivial functions and the 'terms of service' at the bottom of the page (i.e. it might be a bad idea for an LLVM contributer to read this page).
Years ago, bugs I would file against Clang got attention, these days it seems like a black hole.
I have no data or inside insight, but my feeling is that most contributors only work on what their employer wants, and bugs from the public are not relevant.
Still better that’s its open source than not though!
I think many of these bugs are major enough that employers would want them fixed (e.g. not respecting ABI on Windows). A lot of effort has been put into LLVM’s optimizer and code generators so presumably this is something they value.
They track regressions in compile time and other metrics like instruction count for a bunch of benchmark programs on https://llvm-compile-time-tracker.com
Yeah - I didn't understand the hostility - if these fixes had been refused or not even considered, than maybe, but they all seem like honest bugs that need fixing.
