[pulseaudio-discuss] [ANNOUNCE] PulseAudio 0.9.23

Fri Jun 24 14:15:09 PDT 2011

On Fri, Jun 24, 2011 at 12:35 AM, Colin Guthrie <gmane at colin.guthr.ie> wrote:
> Hi,
>
> http://freedesktop.org/software/pulseaudio/releases/pulseaudio-0.9.23.tar.gz
>
> MD5:  7391205a337d1e04a9ff38025f684034
> SHA1: 361f7d89606e8423aad19691bb27f6552e0b679e
>
>
> Following on from the previous release a little over six months ago, I'm
> very happy to announce the next (and likely last) instalment in the
> 0.9.x series.
>
>
> As some of you know, Lennart is currently primarily focused on systemd
> stuff (which is pretty awesome, but it's not PulseAudio!) so I've been
> delegated the job of dong the donkey work of the release, even if
> Lennart is still ACKing the changes overall.
>
> After our v1.0 is out, I shall push out releases much more quickly (at
> least that is the plan!), often without Lennart's direct involvement
> (even tho' I would still like him to give his approval for larger
> changes, I'm more than capable of making appropriate judgement calls).
>
> As before, the 0.9.23 release is primarily a bugfix release. It also
> includes a preview of module-echo-cancel. Originally this was intended
> to gather feedback on it's progress rather than a recommendation for
> general use. As time has slipped somewhat on the release, numerous
> improvements have subsequently been made in the master branch, so we do
> not recommend using it in 0.9.23 to any great extent - feel free to have
> a play tho' :)
>
>
> Anyway without further ado:

The shortlog seems incomplete - there were a lot more changes between
v0.9.22 and v0.9.23 :)

Daniel


> Alexander Kurtz (3):
>      vala: Some bugfixes for the vala bindings
>      vala: move GLibMainLoop class into separate file to fix linker errors
>      vala: More vala fixes
>
> Arnaud Fontaine (1):
>      x11: More XCB fixes.
>
> Arun Raghavan (16):
>      volume: Add explicit checks for ARMv6 instructions
>      rescue-streams: Fix segfault in some conditions
>      client: Don't update smoother while corked
>      echo-cancel: Move the module into it's own directory
>      echo-cancel: Split out speex code from the core module
>      echo-cancel: Pass arguments to the specific canceller module
>      echo-cancel: Let AEC module determine source/sink spec
>      echo-cancel: Add alternative echo-cancellation implementation
>      echo-cancel: Allow selection of AEC method using modargs
>      echo-cancel: Make blocksize a module-wide parameter
>      echo-cancel: Mark immutable parameters as const in vfunc
>      echo-cancel: Fix out-of-tree build
>      echo-cancel: Ensure correct handling of endianness
>      echo-cancel: Use S16NE for adrian module
>      echo-cancel: Fix source may_move_to function
>      i18n: Update POTFILES.*
>
> Chen Rui (1):
>      fix bug about get source-output status
>
> Colin Guthrie (16):
>      build-sys: Replace dummy Makefiles with proper Makefile.am's
>      build-sys: Fix make distcheck failure due to udev rules fixed
> install path
>      build-sys: Include the .version file in the distribution tarball
>      console-kit: Console Kit support is dependent on DBUS and is thus
> optional.
>      build-sys: Put in specific warnings when there is no udev or DBUS
> support
>      build-sys: Mention dbus support in the summary
>      build-sys: Make --disable-dbus actually work.
>      tunnel: Fix automatic names when source/sink_name argument is missing.
>      conf: Make system.pa use udev-detect and not hal-detect.
>      volume: Add a PA_VOLUME_UI_MAX define for the recommended max
> volume to show in UIs
>      vala: Fix path error in the last commit.
>      streams: Fix the actual resampler method shown in debug messages.
>      pulsecore: Add a couple pa_asserts() on pa_tagstruct* calls.
>      device-manager: Free a hook slot to prevent segv on unload
>      build-sys: Add echo-cancel dummy Makefile
>      build-sys: bump soname
>
> Daniel T Chen (2):
>      trivial: Mention speex as a resampler reference in pulse-daemon.conf
>      Further clarify src- and speex- references in pulse-daemon.conf
>
> David Henningsson (5):
>      jack: Add module-jackdbus-detect
>      Fighting rewinds: Seek and write data in the same message
>      Fighting rewinds: Reduce calls to handle_seek
>      module-jack-sink/source: protect against null return in jack_get_ports
>      SSE/MMX: Fix problem with highpitched noise on i386
>
> Forest Bond (3):
>      module-combine-sink: Initialize smoother with offset pa_rtclock_now()
>      module-combine-sink: Initialize smoother in paused state
>      module-combine-sink: Check running flag before rendering in null mode
>
> Jyri Sarha (3):
>      build-sys: Link binaries linking libpulse to libpulsecommon too
>      build-sys: Link binaries linking libpulsecommon to libpulse too
>      build-sys: Link lirc libraries in right order
>
> Lennart Poettering (3):
>      ratelimit: fix log levels of log suppression messages
>      memblockq: implement new call pa_memblockq_peek_fixed_size()
>      rtkit: use private bus connection in order to avoid threading
> issues when invoking pa_make_realtime()
>
> Maarten Bosmans (5):
>      module-loopback: Add adjust_time to valid args
>      Limit rate adjustments to small, inaudible jumps
>      module-rtp-recv: Use new algorithm for adjusting sample rate
>      module-rtp-recv: Average the estimated real sample rate
>      module-rtp-recv: Remove smoother from write index
>
> Maarten Lankhorst (1):
>      module-loopback: Prevent an infinite loop when rate adjusting is
> disabled
>
> Maciej Grela (1):
>      x11: Fix build errors with newest xcb-util.
>
> Matthijs Kooijman (1):
>      http: support HTTP HEAD
>
> Michal Schmidt (1):
>      padsp: wrap __open_2 and __open64_2
>
> Rico Tzschichholz (1):
>      echo-cancel: Fix make distcheck
>
> Scott Reeves (1):
>      Core: Fix incorrect check of return value
>
> Sean McNamara (2):
>      vala: delegate FreeCb does not have a target.
>      vala: ChannelMap has no destroy function.
>
> Vincent Becker (1):
>      Fix return value of pa_sndfile_format_from_string
>
> Wim Taymans (9):
>      make echo-cancel module that exposes a new sink and source
>      echo-cancel: take into account snapshot delay
>      echo-cancel: use the phone media role
>      echo-cancel: tweak the resync code a little
>      echo-cancel: improve debug
>      echo-cancel: keep frame_size a power of 2
>      echo-cancel: rework alignment code
>      echo-cancel: improve accuracy
>      echo-cancel: pause timer when echo canceling is off
>
> Yang Xichuan (1):
>      core: Fix variable "has_whined" value bug
>
> Zhang Wanming (1):
>      Fix typos
>
> --
>
> Colin Guthrie
> gmane(at)colin.guthr.ie
> http://colin.guthr.ie/
>
> Day Job:
>  Tribalogic Limited [http://www.tribalogic.net/]
> Open Source:
>  Mageia Contributor [http://www.mageia.org/]
>  PulseAudio Hacker [http://www.pulseaudio.org/]
>  Trac Hacker [http://trac.edgewall.org/]
>
> _______________________________________________
> pulseaudio-discuss mailing list
> pulseaudio-discuss at lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/pulseaudio-discuss
>