cehteh

> Fine by me. Thou `LinearPOT` could still benefit from the very fact of compile-tine pot checks with `ilog2`. I am pretty sure this optimizes when used with a constant,...

On Tue, 05 Sep 2023 05:21:41 -0700 tower120 ***@***.***> wrote: > As I said before I did try division on pot number in godbolt. It > replaces divisions with bit...

On Tue, 05 Sep 2023 06:31:11 -0700 Jacob McCollum ***@***.***> wrote: > I think we are possibly mixing two concerns here: > > 1. The concern with performance, where powers...

Meanwhile i have a new smartphone (Fairphone4 with Calyxos/microg, no google). Testing with following DC Clients running: a) DC-Desktop 1.30.1 (yes I should update that) on Debian b) DC-Android 1.34.10...

> @cehteh Please read my comment written mid of 2022 and report back, thanks! :) @r10s @gerryfrancis yes i can confirm that, exactly that is happening here too, when I...

> > @cehteh Please read my comment written mid of 2022 and report back, thanks! :) > > @r10s @gerryfrancis yes i can confirm that, exactly that is happening here...

We cant track users eyes/minds but we should have some decently reliable logic about what constitutes as 'message read'. Just a chat being in focus is not enough. Something like...

Notice from the performance department: I expressed before that I am not very fond of using 'thin-vec', now while playing around with the benchmark and different features (trying to see...

i am playing with what i noted before: ``` pub mod detail { use super::MemConfig; #[cfg(feature = "raw-alloc")] pub struct Segment(*mut T); #[cfg(feature = "thin-segments")] pub struct Segment(thin_vec::ThinVec); #[cfg(not(feature =...

> i am playing with what i noted before: > > ``` > pub mod detail { > use super::MemConfig; > #[cfg(feature = "raw-alloc")] > pub struct Segment(*mut T); >...