---
- branch: netbsd-8
  date: Tue Jan  9 19:35:03 UTC 2018
  files:
  - new: 1.26.2.4
    old: 1.26.2.3
    path: src/sys/dev/auconv.c
    pathrev: src/sys/dev/auconv.c@1.26.2.4
    type: modified
  - new: 1.16.42.3
    old: 1.16.42.2
    path: src/sys/dev/auconv.h
    pathrev: src/sys/dev/auconv.h@1.16.42.3
    type: modified
  - new: 1.28.42.3
    old: 1.28.42.2
    path: src/sys/dev/mulaw.c
    pathrev: src/sys/dev/mulaw.c@1.28.42.3
    type: modified
  - new: 1.20.80.2
    old: 1.20.80.1
    path: src/sys/dev/mulaw.h
    pathrev: src/sys/dev/mulaw.h@1.20.80.2
    type: modified
  id: 20180109T193503Z.3a11f97432b9460345919cee3757adf024b70667
  log: "Pull up following revision(s) (requested by nat in ticket #485):\n\tsys/dev/auconv.c:
    1.32-1.35\n\tsys/dev/auconv.h: 1.20-1.21\n\tsys/dev/mulaw.c: 1.33\n\tsys/dev/mulaw.h:
    1.25\nImprove LINEAR_LINEARN macro - inparticular BE/LE conversions.\nAddresses
    part of PR kern/52586 analyzed and reported by isaki@.\n--\nAllow maxused to be
    an odd number of bytes.  This makes single byte\nconversions work for the test
    described in the PR.\nAddresses another part of PR kern/52586.\n--\nRemove 24
    bit in a 32 bit container filters.\nAt present this filter did not work well and
    has caused problems with\nthe audio mixer disabled for hdaudio(4).\nThese filters
    may return when they are fixed.\n--\nReplace LINEARN_TO_LINEARN macro with individual
    optimized functions for\n8, 16, 24 and 32 bits linear to linear conversions.\nAddresses
    PR kern/52586.\n--\nmulaw/alaw_to_linearN macros replaced with individual faster
    filters.\nNFC.\n"
  module: src
  subject: 'CVS commit: [netbsd-8] src/sys/dev'
  unixtime: '1515526503'
  user: snj