riastradh

The new completion system is unusable for me in its present state, so in case this is helpful for anyone else who needs to revert ledger-mode to a state where...

The first part of this still (again?) seems to be happening for me. Versions: - GNU Emacs 26.3 - ledger-mode 6a54eb2b762dd2a5a6ceff54019eeb430562948a

Thanks for digging into this! There's a lot of threads tangled up here. Some questions: 1. Do you know which threads appear to have missed a wakeup -- that is,...

`___lwp_park60` is the generic syscall for a thread to sleep waiting until another thread has woken it with `_lwp_unpark`. It's used by pthread_mutex, pthread_cond, rtld's internal locks, and golang's semasleep/semawakeup...

Interesting! I thought golang used statically linked executables exclusively, so it didn't occur to me that the rtld/libpthread bug could apply here. But the stack traces in https://gist.github.com/bcmills/c6a33bf8ce39e7b017891d9b8bc176f9 show rtld...

> > short of computing the transcendental functions in extended precision > > There's not even an easy proof that this suffices--how close can (x,y) be to exp(-x) = cos(y)?...

Would funding help with getting this and #491 finished, tested, and merged soon, and 0.8.0 released?

Is anyone actively working on this or planning to work on it in the near, foreseeable future? The branch currently doesn't merge cleanly because the following changes conflict: - https://github.com/mozilla/authenticator-rs/commit/47d69204c56d04f17f4918ad4802c4fb6193016f...

I took a closer look at this branch than just by blindly merging and testing, and I took a look at the ctap2-2021 branch too in order to estimate what...

> @riastradh: Thanks. However, I think `__HAVE_ATOMIC_AS_MEMBAR` is really ugly, error-prone and should better be removed. > > Can we just use C11 here? I would even use a plain...