Yat Ho

@jggimi I spun up an OpenBSD 7.7 VM to do some testing with vanilla `4.1.0-beta.2`. I know the the problem doesn't exist anymore in latest `main`, but I want to...

@rafe-s Does the PR work as expected now? * * * I tried installing `devel/dht` on FreeBSD 14.3, set `-DUSE_SYSTEM_DHT=ON`, and got this build error. Cannot figure out why. ```...

@rafe-s Glad to hear that it works for you. This is the CI run where I got my error: https://github.com/tearfur/transmission/actions/runs/18973961378/job/54188663895. Also here's a copy of the build logs for the...

@rafe-s Oh, yeah. I think you're right. Now my FreeBSD CI works. @mikedld Ready for review.

To get this started, I'll start with a few solutions I've heard of. #### Linux 1. Bcache 2. Ext4 (built-in) #### Windows 1. NTFS (built-in)

Just want to reduce noise by urging everyone giving their opinion to support their argument with real test results. We cannot make a good judgement just from a bunch of...

Right now, my stance is not to touch the cache until we have a good reason to believe we can live without it, and by the sound of things we...

I agree that `std::optional` would be better here. `is_dummy()` was introduced to workaround a bug in `small::vector` that requires the `value_type` to be `DefaultInsertable`. See https://github.com/alandefreitas/small/pull/46. Now that I think...

> In any case, I'm curious about that logic when we're using a compiler that supports exceptions by default / not-MSVC. I assume GCC/Clang users don't want exceptions disabled by...

> I can't confirm your changes wouldn't break things for other people (GCC and Clang). Right now `SMALL_BUILD_WITH_EXCEPTIONS` has no effect on any compilers other than MSVC, there isn't any...