home.social

#lcrq — Public Fediverse posts

Live and recent posts from across the Fediverse tagged #lcrq, aggregated by home.social.

fetched live
  1. Less than a week and we've got a minor point release for LCRQ.

    Version 0.3.1 is now available from @Codeberg It's also already in @debian thanks to @vagrantc

    lists.sr.ht/~librecast/libreca

    This was funded by #NGICore thanks to @NGIZero for funding these optimizations of our RaptorQ FEC Library!

    #Librecast #lcrq

  2. It's a new lcrq release with MOAR SPEED !

    Version 0.3.0 is now available from @Codeberg

    We have optimized lcrq for some of the AMD64 SIMD instruction sets which has achieved some large performance gains on some CPUs, especially those with
    AVX512.

    lists.sr.ht/~librecast/libreca

    This was funded by #NGICore thanks to @NGIZero for funding these optimizations of our RaptorQ FEC Library!

    #Librecast #lcrq

  3. You can read more on the Wiki at @Codeberg

    lists.sr.ht/~librecast/libreca

    Changes:

    [0.2.0] - 2024-07-11

    Added

    Runtime CPU Dispatcher
    configure: --enable-native combine compilation units into a single build with all supported isets

    Removed

    configure: --enable-simd - removed obsolete option

    Fixed

    Makefile: ensure header changes trigger rebuild
    configure: check for false
    configure: check if pthreads required for sem_post()
    pass configure-detected LIBS to LDLIBS

    #lcrq

  4. In previous versions, it was necessary to build from source on the target machine to achieve optimal performance. However, most users of linux and *BSDs install their software from binary packages provided by their distribution. These packages need to run on the lowest common denominator in their CPU family e.g for x86_64 binary packages target a CPU released in 2004 (20 Years ago)!

    #lcrq #NGIAssure

  5. It's a new release of lcrq!

    lcrq now makes use of a CPU dispatcher to detect the available SIMD instruction sets at runtime, ensuring that the code runs as fast as possible on the target machine.

    Thanks to @nlnet and #NGIAssure for funding this work.

    codeberg.org/librecast/lcrq/re

    #simd #RaptorQ #lcrq