Marcus Müller
Marcus Müller
Upstream changes from #7555 so that we (OK, as of now, @ryanvolz) aren't left alone with having to remember to fiddle with the workflow file after every re-render
From a discussion with @awalls-cx18 on IRC: * FFT filters and such are nice, but unusable for bursty operation, since there's no way to clear internal state * this leads...
… it includes sqrt, which is an analytic function, not transcendental at all. The real problem, though, is that it's a wrapper for a few functions from a specific library...
After a discussion with @osh this sounds like a very good idea: Blocks that don't actually need any block state shouldn't have a (`general_`)`work` method that's allowed to modify any...
Schneider @schneider42 would like to have an indication of how full the message receiver buffers are. That's a very reasonable request, as it would allow our pdu architecture slightly backpressurable...
I had this lying around and $daytimejob used it, so why not put it out in the open? I use a small python script to compare `volk_config`s from different machines.
- CFLAGS should definitely include `-O3` - There's still the disabled check in there, due to the by-now-fixed s390x problems - yarda is a good maintainer, help them
I understand FFTw is necessary to be bundled for a WASM build (is that correct?), but on any other system, we must use the system FFTw, to be packagable. (Also,...
Can't do FetchContent in any distro builds, nor in openembedded, Conan/artifactory builds etc. Simply a show-stopper for distribution, and thus for a 4.0 release. We need all externalities to either...
… bit surprising, tbh, since modern clang is explicitly listed as supported ```shell mkdir testbuild && cd testbuild cmake -DCMAKE_BUILD_TYPE=RelWithDebInfo -GNinja -DCMAKE_CXX_COMPILER=clang++ -DCMAKE_C_COMPILER=clang ${graph-proto directory} ``` yields ```text -- The...