* [Qemu-devel] [PATCH v2 0/6] -audiodev option
@ 2015-06-16 12:49 Kővágó, Zoltán
2015-06-16 12:49 ` [Qemu-devel] [PATCH v2 1/6] qapi: qapi for audio backends Kővágó, Zoltán
` (5 more replies)
0 siblings, 6 replies; 42+ messages in thread
From: Kővágó, Zoltán @ 2015-06-16 12:49 UTC (permalink / raw)
To: qemu-devel; +Cc: Gerd Hoffmann
This series of patches adds a new -audiodev command line option to specify audio
subsytem parameters instead of environment variables. This will later allow us
to specify multiple audio backends. The syntax is something like this:
-audiodev driver_name,property=value,...
like:
-audiodev alsa,frequency=8000,channels=1
The first commit adds a qapi Audiodev struct that describes the audio backend
options. The next 4 commits are some miscellaneous additions that are needed by
the last commit which finally adds the -audiodev option.
For users with esoteric platforms or needs please check I did not break anything
accidentally. For easier testing, pull https://github.com/DirtYiCE/qemu.git tag
audio-cmdline-v2.
Please review.
---
Changes from v1:
* updated to master, removed commits already merged
* now every time-related option takes usecs
* some small fixes
Kővágó, Zoltán (6):
qapi: qapi for audio backends
qapi: support nested structs in OptsVisitor
opts: do not print separator before first item in qemu_opts_print
qapi: AllocVisitor
audio: use qapi AudioFormat instead of audfmt_e
audio: -audiodev command line option
Makefile | 4 +-
audio/Makefile.objs | 2 +-
audio/alsaaudio.c | 337 +++++---------
audio/audio.c | 798 ++++++++++----------------------
audio/audio.h | 34 +-
audio/audio_int.h | 7 +-
audio/audio_legacy.c | 328 +++++++++++++
audio/audio_template.h | 13 +-
audio/audio_win_int.c | 18 +-
audio/coreaudio.c | 49 +-
audio/dsound_template.h | 6 +-
audio/dsoundaudio.c | 60 +--
audio/noaudio.c | 3 +-
audio/ossaudio.c | 183 +++-----
audio/paaudio.c | 109 ++---
audio/sdlaudio.c | 50 +-
audio/spiceaudio.c | 11 +-
audio/wavaudio.c | 76 +--
audio/wavcapture.c | 2 +-
block.c | 2 +-
hw/arm/omap2.c | 2 +-
hw/audio/ac97.c | 2 +-
hw/audio/adlib.c | 2 +-
hw/audio/cs4231a.c | 6 +-
hw/audio/es1370.c | 4 +-
hw/audio/gus.c | 2 +-
hw/audio/hda-codec.c | 18 +-
hw/audio/lm4549.c | 6 +-
hw/audio/milkymist-ac97.c | 2 +-
hw/audio/pcspk.c | 2 +-
hw/audio/sb16.c | 14 +-
hw/audio/wm8750.c | 4 +-
hw/input/tsc210x.c | 2 +-
hw/usb/dev-audio.c | 2 +-
include/qapi/alloc-visitor.h | 18 +
qapi-schema.json | 3 +
qapi/Makefile.objs | 1 +
qapi/alloc-visitor.c | 62 +++
qapi/audio.json | 223 +++++++++
qapi/opts-visitor.c | 144 ++++--
qemu-options.hx | 218 ++++++++-
tests/qapi-schema/qapi-schema-test.json | 9 +-
tests/test-opts-visitor.c | 34 ++
ui/vnc.c | 14 +-
util/qemu-option.c | 5 +-
vl.c | 9 +-
46 files changed, 1627 insertions(+), 1273 deletions(-)
create mode 100644 audio/audio_legacy.c
create mode 100644 include/qapi/alloc-visitor.h
create mode 100644 qapi/alloc-visitor.c
create mode 100644 qapi/audio.json
--
2.4.3
^ permalink raw reply [flat|nested] 42+ messages in thread* [Qemu-devel] [PATCH v2 1/6] qapi: qapi for audio backends 2015-06-16 12:49 [Qemu-devel] [PATCH v2 0/6] -audiodev option Kővágó, Zoltán @ 2015-06-16 12:49 ` Kővágó, Zoltán 2015-06-17 7:46 ` Markus Armbruster 2015-06-16 12:49 ` [Qemu-devel] [PATCH v2 2/6] qapi: support nested structs in OptsVisitor Kővágó, Zoltán ` (4 subsequent siblings) 5 siblings, 1 reply; 42+ messages in thread From: Kővágó, Zoltán @ 2015-06-16 12:49 UTC (permalink / raw) To: qemu-devel; +Cc: Michael Roth, Gerd Hoffmann, Markus Armbruster This patch adds structures into qapi to replace the existing configuration structures used by audio backends currently. This qapi will be the base of the -audiodev command line parameter (that replaces the old environment variables based config). This is not a 1:1 translation of the old options, I've tried to make them much more consistent (e.g. almost every backend had an option to specify buffer size, but the name was different for every backend, and some backends required usecs, while some other required frames, samples or bytes). Also tried to reduce the number of abbreviations used by the config keys. Some of the more important changes: * use `in` and `out` instead of `ADC` and `DAC`, as the former is more user friendly imho * moved buffer settings into the global setting area (so it's the same for all backends that support it. Backends that can't change buffer size will simply ignore them). Also using usecs, as it's probably more user friendly than samples or bytes. * try-poll is now an alsa and oss backend specific option (as all other backends currently ignore it) Signed-off-by: Kővágó, Zoltán <DirtY.iCE.hu@gmail.com> --- Changes from v1 patch: * every time-related field now take usecs (and removed -usecs, -millis suffixes) * fixed inconsisten optional marking, language issues Changes from v2 RFC patch: * in, out are no longer optional * try-poll: moved to alsa and oss (as no other backend used them) * voices: added (env variables had this option) * dsound: removed primary buffer related fields Changes from v1 RFC patch: * fixed style issues * moved definitions into a separate file * documented undocumented options (hopefully) * removed plive option. It was useless even years ago so it can probably safely go away: https://lists.nongnu.org/archive/html/qemu-devel/2012-03/msg02427.html * removed verbose, debug options. Backends should use trace events instead. * removed *_retries options from dsound. It's a kludge. * moved buffer_usecs and buffer_count to the global config options. Some driver might ignore it (as they do not expose API to change them). * wav backend: removed frequecy, format, channels as AudiodevPerDirectionOptions already have them. Makefile | 4 +- qapi-schema.json | 3 + qapi/audio.json | 223 +++++++++++++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 228 insertions(+), 2 deletions(-) create mode 100644 qapi/audio.json diff --git a/Makefile b/Makefile index 3f97904..ac566fa 100644 --- a/Makefile +++ b/Makefile @@ -257,8 +257,8 @@ $(SRC_PATH)/qga/qapi-schema.json $(SRC_PATH)/scripts/qapi-commands.py $(qapi-py) " GEN $@") qapi-modules = $(SRC_PATH)/qapi-schema.json $(SRC_PATH)/qapi/common.json \ - $(SRC_PATH)/qapi/block.json $(SRC_PATH)/qapi/block-core.json \ - $(SRC_PATH)/qapi/event.json + $(SRC_PATH)/qapi/audio.json $(SRC_PATH)/qapi/block.json \ + $(SRC_PATH)/qapi/block-core.json $(SRC_PATH)/qapi/event.json qapi-types.c qapi-types.h :\ $(qapi-modules) $(SRC_PATH)/scripts/qapi-types.py $(qapi-py) diff --git a/qapi-schema.json b/qapi-schema.json index 106008c..e751ea3 100644 --- a/qapi-schema.json +++ b/qapi-schema.json @@ -5,6 +5,9 @@ # QAPI common definitions { 'include': 'qapi/common.json' } +# QAPI audio definitions +{ 'include': 'qapi/audio.json' } + # QAPI block definitions { 'include': 'qapi/block.json' } diff --git a/qapi/audio.json b/qapi/audio.json new file mode 100644 index 0000000..2851689 --- /dev/null +++ b/qapi/audio.json @@ -0,0 +1,223 @@ +# -*- mode: python -*- +# +# Copyright (C) 2015 Zoltán Kővágó <DirtY.iCE.hu@gmail.com> +# +# This work is licensed under the terms of the GNU GPL, version 2 or later. +# See the COPYING file in the top-level directory. + +## +# @AudiodevNoOptions +# +# The none, coreaudio, sdl and spice audio backend have no options. +# +# Since: 2.4 +## +{ 'struct': 'AudiodevNoOptions', + 'data': { } } + +## +# @AudiodevAlsaPerDirectionOptions +# +# Options of the alsa backend that are used for both playback and recording. +# +# @dev: #optional the name of the alsa device to use +# +# @try-poll: #optional attempt to use poll mode +# +# Since: 2.4 +## +{ 'struct': 'AudiodevAlsaPerDirectionOptions', + 'data': { + '*dev': 'str', + '*try-poll': 'bool' } } + +## +# @AudiodevAlsaOptions +# +# Options of the alsa audio backend. +# +# @in: options of the capture stream +# +# @out: options of the playback stream +# +# @threshold: #optional set the threshold (in frames) when playback starts +# +# Since: 2.4 +## +{ 'struct': 'AudiodevAlsaOptions', + 'data': { + 'in': 'AudiodevAlsaPerDirectionOptions', + 'out': 'AudiodevAlsaPerDirectionOptions', + '*threshold': 'int' } } + +## +# @AudiodevDsoundOptions +# +# Options of the dsound audio backend. +# +# @latency: #optional add extra latency to playback (in microseconds) +# +# Since: 2.4 +## +{ 'struct': 'AudiodevDsoundOptions', + 'data': { + '*latency': 'int' } } + +## +# @AudiodevOssPerDirectionOptions +# +# Options of the oss backend that are used for both playback and recording. +# +# @dev: #optional path of the oss device +# +# @try-poll: #optional attempt to use poll mode +# +# Since: 2.4 +## +{ 'struct': 'AudiodevOssPerDirectionOptions', + 'data': { + '*dev': 'str', + '*try-poll': 'bool' } } + +## +# @AudiodevOssOptions +# +# Options of the oss audio backend. +# +# @in: options of the capture stream +# +# @out: options of the playback stream +# +# @mmap: #optional try using memory mapped access +# +# @exclusive: #optional open device in exclusive mode (vmix won't work) +# +# @dsp-policy: #optional set the timing policy of the device, -1 to use fragment +# mode (option ignored on some platforms) +# +# Since: 2.4 +## +{ 'struct': 'AudiodevOssOptions', + 'data': { + 'in': 'AudiodevOssPerDirectionOptions', + 'out': 'AudiodevOssPerDirectionOptions', + '*mmap': 'bool', + '*exclusive': 'bool', + '*dsp-policy': 'int' } } + +## +# @AudiodevPaOptions +# +# Options of the pa (PulseAudio) audio backend. +# +# @server: #optional PulseAudio server address +# +# @sink: #optional sink device name +# +# @source: #optional source device name +# +# Since: 2.4 +## +{ 'struct': 'AudiodevPaOptions', + 'data': { + '*server': 'str', + '*sink': 'str', + '*source': 'str' } } + +## +# @AudiodevWavOptions +# +# Options of the wav audio backend. +# +# @path: #optional path of the wav file to record +# +# Since: 2.4 +## +{ 'struct': 'AudiodevWavOptions', + 'data': { + '*path': 'str' } } + + +## +# @AudiodevBackendOptions +# +# A discriminated record of audio backends. +# +# Since: 2.4 +## +{ 'union': 'AudiodevBackendOptions', + 'data': { + 'none': 'AudiodevNoOptions', + 'alsa': 'AudiodevAlsaOptions', + 'coreaudio': 'AudiodevNoOptions', + 'dsound': 'AudiodevDsoundOptions', + 'oss': 'AudiodevOssOptions', + 'pa': 'AudiodevPaOptions', + 'sdl': 'AudiodevNoOptions', + 'spice': 'AudiodevNoOptions', + 'wav': 'AudiodevWavOptions' } } + +## +# @AudioFormat +# +# An enumeration of possible audio formats. +# +# Since: 2.4 +## +{ 'enum': 'AudioFormat', + 'data': [ 'u8', 's8', 'u16', 's16', 'u32', 's32' ] } + +## +# @AudiodevPerDirectionOptions +# +# General audio backend options that are used for both playback and recording. +# +# @fixed-settings: #optional use fixed settings for host DAC/ADC +# +# @frequency: #optional frequency to use when using fixed settings +# +# @channels: #optional number of channels when using fixed settings +# +# @format: #optional sample format to use when using fixed settings +# +# @buffer: #optional the buffer size (in microseconds) +# +# @buffer-count: #optional number of buffers +# +# Since: 2.4 +## +{ 'struct': 'AudiodevPerDirectionOptions', + 'data': { + '*fixed-settings': 'bool', + '*frequency': 'int', + '*channels': 'int', + '*voices': 'int', + '*format': 'AudioFormat', + '*buffer': 'int', + '*buffer-count': 'int' } } + +## +# @Audiodev +# +# Captures the configuration of an audio backend. +# +# @id: identifier of the backend +# +# @in: options of the capture stream +# +# @out: options of the playback stream +# +# @timer-period: #optional timer period (in microseconds, 0: use lowest +# possible) +# +# @opts: audio backend specific options +# +# Since: 2.4 +## +{ 'struct': 'Audiodev', + 'data': { + '*id': 'str', + 'in': 'AudiodevPerDirectionOptions', + 'out': 'AudiodevPerDirectionOptions', + '*timer-period': 'int', + 'opts': 'AudiodevBackendOptions' } } -- 2.4.3 ^ permalink raw reply related [flat|nested] 42+ messages in thread
* Re: [Qemu-devel] [PATCH v2 1/6] qapi: qapi for audio backends 2015-06-16 12:49 ` [Qemu-devel] [PATCH v2 1/6] qapi: qapi for audio backends Kővágó, Zoltán @ 2015-06-17 7:46 ` Markus Armbruster 2015-06-17 10:54 ` Kővágó Zoltán 0 siblings, 1 reply; 42+ messages in thread From: Markus Armbruster @ 2015-06-17 7:46 UTC (permalink / raw) To: Kővágó, Zoltán Cc: Gerd Hoffmann, qemu-devel, Michael Roth Copying Eric for additional QAPI schema expertise. My questions inline, pretty sure they show my ignorance. "Kővágó, Zoltán" <dirty.ice.hu@gmail.com> writes: > This patch adds structures into qapi to replace the existing configuration > structures used by audio backends currently. This qapi will be the base of the > -audiodev command line parameter (that replaces the old environment variables > based config). > > This is not a 1:1 translation of the old options, I've tried to make them much > more consistent (e.g. almost every backend had an option to specify buffer size, > but the name was different for every backend, and some backends required usecs, > while some other required frames, samples or bytes). Also tried to reduce the > number of abbreviations used by the config keys. > > Some of the more important changes: > * use `in` and `out` instead of `ADC` and `DAC`, as the former is more user > friendly imho > * moved buffer settings into the global setting area (so it's the same for all > backends that support it. Backends that can't change buffer size will simply > ignore them). Also using usecs, as it's probably more user friendly than > samples or bytes. > * try-poll is now an alsa and oss backend specific option (as all other backends > currently ignore it) > > Signed-off-by: Kővágó, Zoltán <DirtY.iCE.hu@gmail.com> > > --- > > Changes from v1 patch: > * every time-related field now take usecs (and removed -usecs, -millis suffixes) > * fixed inconsisten optional marking, language issues > > Changes from v2 RFC patch: > * in, out are no longer optional > * try-poll: moved to alsa and oss (as no other backend used them) > * voices: added (env variables had this option) > * dsound: removed primary buffer related fields > > Changes from v1 RFC patch: > * fixed style issues > * moved definitions into a separate file > * documented undocumented options (hopefully) > * removed plive option. It was useless even years ago so it can probably safely > go away: https://lists.nongnu.org/archive/html/qemu-devel/2012-03/msg02427.html > * removed verbose, debug options. Backends should use trace events instead. > * removed *_retries options from dsound. It's a kludge. > * moved buffer_usecs and buffer_count to the global config options. Some driver > might ignore it (as they do not expose API to change them). > * wav backend: removed frequecy, format, channels as AudiodevPerDirectionOptions > already have them. > > Makefile | 4 +- > qapi-schema.json | 3 + > qapi/audio.json | 223 +++++++++++++++++++++++++++++++++++++++++++++++++++++++ > 3 files changed, 228 insertions(+), 2 deletions(-) > create mode 100644 qapi/audio.json > > diff --git a/Makefile b/Makefile > index 3f97904..ac566fa 100644 > --- a/Makefile > +++ b/Makefile > @@ -257,8 +257,8 @@ $(SRC_PATH)/qga/qapi-schema.json $(SRC_PATH)/scripts/qapi-commands.py $(qapi-py) > " GEN $@") > > qapi-modules = $(SRC_PATH)/qapi-schema.json $(SRC_PATH)/qapi/common.json \ > - $(SRC_PATH)/qapi/block.json $(SRC_PATH)/qapi/block-core.json \ > - $(SRC_PATH)/qapi/event.json > + $(SRC_PATH)/qapi/audio.json $(SRC_PATH)/qapi/block.json \ > + $(SRC_PATH)/qapi/block-core.json $(SRC_PATH)/qapi/event.json > > qapi-types.c qapi-types.h :\ > $(qapi-modules) $(SRC_PATH)/scripts/qapi-types.py $(qapi-py) > diff --git a/qapi-schema.json b/qapi-schema.json > index 106008c..e751ea3 100644 > --- a/qapi-schema.json > +++ b/qapi-schema.json > @@ -5,6 +5,9 @@ > # QAPI common definitions > { 'include': 'qapi/common.json' } > > +# QAPI audio definitions > +{ 'include': 'qapi/audio.json' } > + > # QAPI block definitions > { 'include': 'qapi/block.json' } > > diff --git a/qapi/audio.json b/qapi/audio.json > new file mode 100644 > index 0000000..2851689 > --- /dev/null > +++ b/qapi/audio.json > @@ -0,0 +1,223 @@ > +# -*- mode: python -*- > +# > +# Copyright (C) 2015 Zoltán Kővágó <DirtY.iCE.hu@gmail.com> > +# > +# This work is licensed under the terms of the GNU GPL, version 2 or later. > +# See the COPYING file in the top-level directory. > + > +## > +# @AudiodevNoOptions > +# > +# The none, coreaudio, sdl and spice audio backend have no options. > +# > +# Since: 2.4 > +## > +{ 'struct': 'AudiodevNoOptions', > + 'data': { } } > + > +## > +# @AudiodevAlsaPerDirectionOptions > +# > +# Options of the alsa backend that are used for both playback and recording. > +# > +# @dev: #optional the name of the alsa device to use Who picks the default, QEMU or ALSA? > +# > +# @try-poll: #optional attempt to use poll mode What happens when the attempt fails? > +# > +# Since: 2.4 > +## > +{ 'struct': 'AudiodevAlsaPerDirectionOptions', > + 'data': { > + '*dev': 'str', > + '*try-poll': 'bool' } } > + > +## > +# @AudiodevAlsaOptions > +# > +# Options of the alsa audio backend. > +# > +# @in: options of the capture stream > +# > +# @out: options of the playback stream > +# > +# @threshold: #optional set the threshold (in frames) when playback starts > +# > +# Since: 2.4 > +## > +{ 'struct': 'AudiodevAlsaOptions', > + 'data': { > + 'in': 'AudiodevAlsaPerDirectionOptions', > + 'out': 'AudiodevAlsaPerDirectionOptions', > + '*threshold': 'int' } } Do we have an established term for a "direction"? If not: the use with 'in' and 'out' tempts me to name the type AudiodevAlsaIOOptions. Just rambling, use what you think is best. > + > +## > +# @AudiodevDsoundOptions > +# > +# Options of the dsound audio backend. > +# > +# @latency: #optional add extra latency to playback (in microseconds) We already use seconds, milliseconds and nanoseconds. You make the zoo complete. > +# > +# Since: 2.4 > +## > +{ 'struct': 'AudiodevDsoundOptions', > + 'data': { > + '*latency': 'int' } } > + > +## > +# @AudiodevOssPerDirectionOptions > +# > +# Options of the oss backend that are used for both playback and recording. > +# > +# @dev: #optional path of the oss device Who picks the default, QEMU or OSS? For ALSA, you documented @dev to be "the name", here it's "path". Intentional difference? > +# > +# @try-poll: #optional attempt to use poll mode > +# > +# Since: 2.4 > +## > +{ 'struct': 'AudiodevOssPerDirectionOptions', > + 'data': { > + '*dev': 'str', > + '*try-poll': 'bool' } } Same as for ALSA. Keeping them separate is fine with me. > + > +## > +# @AudiodevOssOptions > +# > +# Options of the oss audio backend. > +# > +# @in: options of the capture stream > +# > +# @out: options of the playback stream > +# > +# @mmap: #optional try using memory mapped access What happens when the attempt fails? Should this be called try-mmap? > +# > +# @exclusive: #optional open device in exclusive mode (vmix won't work) > +# > +# @dsp-policy: #optional set the timing policy of the device, -1 to use fragment > +# mode (option ignored on some platforms) What are the possible values besides -1? > +# > +# Since: 2.4 > +## > +{ 'struct': 'AudiodevOssOptions', > + 'data': { > + 'in': 'AudiodevOssPerDirectionOptions', > + 'out': 'AudiodevOssPerDirectionOptions', > + '*mmap': 'bool', > + '*exclusive': 'bool', > + '*dsp-policy': 'int' } } > + > +## > +# @AudiodevPaOptions > +# > +# Options of the pa (PulseAudio) audio backend. > +# > +# @server: #optional PulseAudio server address > +# > +# @sink: #optional sink device name > +# > +# @source: #optional source device name Who picks the defaults, QEMU or PA? > +# > +# Since: 2.4 > +## > +{ 'struct': 'AudiodevPaOptions', > + 'data': { > + '*server': 'str', > + '*sink': 'str', > + '*source': 'str' } } > + > +## > +# @AudiodevWavOptions > +# > +# Options of the wav audio backend. > +# > +# @path: #optional path of the wav file to record > +# > +# Since: 2.4 > +## > +{ 'struct': 'AudiodevWavOptions', > + 'data': { > + '*path': 'str' } } Who picks the default? > + > + > +## > +# @AudiodevBackendOptions > +# > +# A discriminated record of audio backends. > +# > +# Since: 2.4 > +## > +{ 'union': 'AudiodevBackendOptions', > + 'data': { > + 'none': 'AudiodevNoOptions', > + 'alsa': 'AudiodevAlsaOptions', > + 'coreaudio': 'AudiodevNoOptions', > + 'dsound': 'AudiodevDsoundOptions', > + 'oss': 'AudiodevOssOptions', > + 'pa': 'AudiodevPaOptions', > + 'sdl': 'AudiodevNoOptions', > + 'spice': 'AudiodevNoOptions', > + 'wav': 'AudiodevWavOptions' } } > + > +## > +# @AudioFormat > +# > +# An enumeration of possible audio formats. > +# > +# Since: 2.4 > +## > +{ 'enum': 'AudioFormat', > + 'data': [ 'u8', 's8', 'u16', 's16', 'u32', 's32' ] } > + > +## > +# @AudiodevPerDirectionOptions > +# > +# General audio backend options that are used for both playback and recording. > +# > +# @fixed-settings: #optional use fixed settings for host DAC/ADC > +# > +# @frequency: #optional frequency to use when using fixed settings > +# > +# @channels: #optional number of channels when using fixed settings > +# > +# @format: #optional sample format to use when using fixed settings Are these guys used when @fixed-settings is off? > +# > +# @buffer: #optional the buffer size (in microseconds) @buffer suggests this is a buffer, not a buffer length given as time span. @buffer-len? > +# > +# @buffer-count: #optional number of buffers > +# > +# Since: 2.4 > +## > +{ 'struct': 'AudiodevPerDirectionOptions', > + 'data': { > + '*fixed-settings': 'bool', > + '*frequency': 'int', > + '*channels': 'int', > + '*voices': 'int', > + '*format': 'AudioFormat', > + '*buffer': 'int', > + '*buffer-count': 'int' } } > + > +## > +# @Audiodev > +# > +# Captures the configuration of an audio backend. > +# > +# @id: identifier of the backend > +# > +# @in: options of the capture stream > +# > +# @out: options of the playback stream > +# > +# @timer-period: #optional timer period (in microseconds, 0: use lowest > +# possible) > +# > +# @opts: audio backend specific options > +# > +# Since: 2.4 > +## > +{ 'struct': 'Audiodev', > + 'data': { > + '*id': 'str', > + 'in': 'AudiodevPerDirectionOptions', > + 'out': 'AudiodevPerDirectionOptions', > + '*timer-period': 'int', > + 'opts': 'AudiodevBackendOptions' } } Have you considered making this a flat union, similar ro BlockdevOptions? Don't get deceived by the number of my questions, this is solid work. ^ permalink raw reply [flat|nested] 42+ messages in thread
* Re: [Qemu-devel] [PATCH v2 1/6] qapi: qapi for audio backends 2015-06-17 7:46 ` Markus Armbruster @ 2015-06-17 10:54 ` Kővágó Zoltán 2015-06-17 11:48 ` Markus Armbruster 0 siblings, 1 reply; 42+ messages in thread From: Kővágó Zoltán @ 2015-06-17 10:54 UTC (permalink / raw) To: Markus Armbruster; +Cc: Gerd Hoffmann, qemu-devel, Michael Roth 2015-06-17 09:46 keltezéssel, Markus Armbruster írta: > Copying Eric for additional QAPI schema expertise. > > My questions inline, pretty sure they show my ignorance. > > "Kővágó, Zoltán" <dirty.ice.hu@gmail.com> writes: > >> This patch adds structures into qapi to replace the existing configuration >> structures used by audio backends currently. This qapi will be the base of the >> -audiodev command line parameter (that replaces the old environment variables >> based config). >> >> This is not a 1:1 translation of the old options, I've tried to make them much >> more consistent (e.g. almost every backend had an option to specify buffer size, >> but the name was different for every backend, and some backends required usecs, >> while some other required frames, samples or bytes). Also tried to reduce the >> number of abbreviations used by the config keys. >> >> Some of the more important changes: >> * use `in` and `out` instead of `ADC` and `DAC`, as the former is more user >> friendly imho >> * moved buffer settings into the global setting area (so it's the same for all >> backends that support it. Backends that can't change buffer size will simply >> ignore them). Also using usecs, as it's probably more user friendly than >> samples or bytes. >> * try-poll is now an alsa and oss backend specific option (as all other backends >> currently ignore it) >> >> Signed-off-by: Kővágó, Zoltán <DirtY.iCE.hu@gmail.com> >> >> --- >> >> Changes from v1 patch: >> * every time-related field now take usecs (and removed -usecs, -millis suffixes) >> * fixed inconsisten optional marking, language issues >> >> Changes from v2 RFC patch: >> * in, out are no longer optional >> * try-poll: moved to alsa and oss (as no other backend used them) >> * voices: added (env variables had this option) >> * dsound: removed primary buffer related fields >> >> Changes from v1 RFC patch: >> * fixed style issues >> * moved definitions into a separate file >> * documented undocumented options (hopefully) >> * removed plive option. It was useless even years ago so it can probably safely >> go away: https://lists.nongnu.org/archive/html/qemu-devel/2012-03/msg02427.html >> * removed verbose, debug options. Backends should use trace events instead. >> * removed *_retries options from dsound. It's a kludge. >> * moved buffer_usecs and buffer_count to the global config options. Some driver >> might ignore it (as they do not expose API to change them). >> * wav backend: removed frequecy, format, channels as AudiodevPerDirectionOptions >> already have them. >> >> Makefile | 4 +- >> qapi-schema.json | 3 + >> qapi/audio.json | 223 +++++++++++++++++++++++++++++++++++++++++++++++++++++++ >> 3 files changed, 228 insertions(+), 2 deletions(-) >> create mode 100644 qapi/audio.json >> >> diff --git a/Makefile b/Makefile >> index 3f97904..ac566fa 100644 >> --- a/Makefile >> +++ b/Makefile >> @@ -257,8 +257,8 @@ $(SRC_PATH)/qga/qapi-schema.json $(SRC_PATH)/scripts/qapi-commands.py $(qapi-py) >> " GEN $@") >> >> qapi-modules = $(SRC_PATH)/qapi-schema.json $(SRC_PATH)/qapi/common.json \ >> - $(SRC_PATH)/qapi/block.json $(SRC_PATH)/qapi/block-core.json \ >> - $(SRC_PATH)/qapi/event.json >> + $(SRC_PATH)/qapi/audio.json $(SRC_PATH)/qapi/block.json \ >> + $(SRC_PATH)/qapi/block-core.json $(SRC_PATH)/qapi/event.json >> >> qapi-types.c qapi-types.h :\ >> $(qapi-modules) $(SRC_PATH)/scripts/qapi-types.py $(qapi-py) >> diff --git a/qapi-schema.json b/qapi-schema.json >> index 106008c..e751ea3 100644 >> --- a/qapi-schema.json >> +++ b/qapi-schema.json >> @@ -5,6 +5,9 @@ >> # QAPI common definitions >> { 'include': 'qapi/common.json' } >> >> +# QAPI audio definitions >> +{ 'include': 'qapi/audio.json' } >> + >> # QAPI block definitions >> { 'include': 'qapi/block.json' } >> >> diff --git a/qapi/audio.json b/qapi/audio.json >> new file mode 100644 >> index 0000000..2851689 >> --- /dev/null >> +++ b/qapi/audio.json >> @@ -0,0 +1,223 @@ >> +# -*- mode: python -*- >> +# >> +# Copyright (C) 2015 Zoltán Kővágó <DirtY.iCE.hu@gmail.com> >> +# >> +# This work is licensed under the terms of the GNU GPL, version 2 or later. >> +# See the COPYING file in the top-level directory. >> + >> +## >> +# @AudiodevNoOptions >> +# >> +# The none, coreaudio, sdl and spice audio backend have no options. >> +# >> +# Since: 2.4 >> +## >> +{ 'struct': 'AudiodevNoOptions', >> + 'data': { } } >> + >> +## >> +# @AudiodevAlsaPerDirectionOptions >> +# >> +# Options of the alsa backend that are used for both playback and recording. >> +# >> +# @dev: #optional the name of the alsa device to use > > Who picks the default, QEMU or ALSA? It defaults to "default", which tells alsa to use the default device... > >> +# >> +# @try-poll: #optional attempt to use poll mode > > What happens when the attempt fails? It falls back to non polling (timer based) mode. >> +# >> +# Since: 2.4 >> +## >> +{ 'struct': 'AudiodevAlsaPerDirectionOptions', >> + 'data': { >> + '*dev': 'str', >> + '*try-poll': 'bool' } } >> + >> +## >> +# @AudiodevAlsaOptions >> +# >> +# Options of the alsa audio backend. >> +# >> +# @in: options of the capture stream >> +# >> +# @out: options of the playback stream >> +# >> +# @threshold: #optional set the threshold (in frames) when playback starts >> +# >> +# Since: 2.4 >> +## >> +{ 'struct': 'AudiodevAlsaOptions', >> + 'data': { >> + 'in': 'AudiodevAlsaPerDirectionOptions', >> + 'out': 'AudiodevAlsaPerDirectionOptions', >> + '*threshold': 'int' } } > > Do we have an established term for a "direction"? > > If not: the use with 'in' and 'out' tempts me to name the type > AudiodevAlsaIOOptions. > > Just rambling, use what you think is best. I don't know, so far nobody rambled about my naming... > >> + >> +## >> +# @AudiodevDsoundOptions >> +# >> +# Options of the dsound audio backend. >> +# >> +# @latency: #optional add extra latency to playback (in microseconds) > > We already use seconds, milliseconds and nanoseconds. You make the zoo > complete. Hmm. The audio backend previously used microseconds, milliseconds, Hertz, frames, samples and bytes as time measurement, now at least everything use microseconds. Milliseconds precision is not enough, while nanosecond precision doesn't really make sense imho. >> +# >> +# Since: 2.4 >> +## >> +{ 'struct': 'AudiodevDsoundOptions', >> + 'data': { >> + '*latency': 'int' } } >> + >> +## >> +# @AudiodevOssPerDirectionOptions >> +# >> +# Options of the oss backend that are used for both playback and recording. >> +# >> +# @dev: #optional path of the oss device > > Who picks the default, QEMU or OSS? It defaults to "/dev/dsp". > For ALSA, you documented @dev to be "the name", here it's "path". > Intentional difference? Yes. For oss you have to provide a filesystem path of the audio device (like /dev/dsp), for alsa you provide an alsa specific name, like "default", "hw:1,0", "hdmi:CARD=NVidia,DEV=1", whatever. >> +# >> +# @try-poll: #optional attempt to use poll mode >> +# >> +# Since: 2.4 >> +## >> +{ 'struct': 'AudiodevOssPerDirectionOptions', >> + 'data': { >> + '*dev': 'str', >> + '*try-poll': 'bool' } } > > Same as for ALSA. Keeping them separate is fine with me. This is the same as alsa's try-poll. They're separate because only these two backend use this setting. > >> + >> +## >> +# @AudiodevOssOptions >> +# >> +# Options of the oss audio backend. >> +# >> +# @in: options of the capture stream >> +# >> +# @out: options of the playback stream >> +# >> +# @mmap: #optional try using memory mapped access > > What happens when the attempt fails? It should fall back to non-mmapped access (should because when I checked, it was buggy, at least on linux with alsa emulated oss on pulseaudio alsa device...) > Should this be called try-mmap? Agree. >> +# >> +# @exclusive: #optional open device in exclusive mode (vmix won't work) >> +# >> +# @dsp-policy: #optional set the timing policy of the device, -1 to use fragment >> +# mode (option ignored on some platforms) > > What are the possible values besides -1? It should be a number between 0 and 10 (according to this page: http://manuals.opensound.com/developer/SNDCTL_DSP_POLICY.html) > >> +# >> +# Since: 2.4 >> +## >> +{ 'struct': 'AudiodevOssOptions', >> + 'data': { >> + 'in': 'AudiodevOssPerDirectionOptions', >> + 'out': 'AudiodevOssPerDirectionOptions', >> + '*mmap': 'bool', >> + '*exclusive': 'bool', >> + '*dsp-policy': 'int' } } >> + >> +## >> +# @AudiodevPaOptions >> +# >> +# Options of the pa (PulseAudio) audio backend. >> +# >> +# @server: #optional PulseAudio server address >> +# >> +# @sink: #optional sink device name >> +# >> +# @source: #optional source device name > > Who picks the defaults, QEMU or PA? PA >> +# >> +# Since: 2.4 >> +## >> +{ 'struct': 'AudiodevPaOptions', >> + 'data': { >> + '*server': 'str', >> + '*sink': 'str', >> + '*source': 'str' } } >> + >> +## >> +# @AudiodevWavOptions >> +# >> +# Options of the wav audio backend. >> +# >> +# @path: #optional path of the wav file to record >> +# >> +# Since: 2.4 >> +## >> +{ 'struct': 'AudiodevWavOptions', >> + 'data': { >> + '*path': 'str' } } > > Who picks the default? It defaults to "qemu.wav" >> + >> + >> +## >> +# @AudiodevBackendOptions >> +# >> +# A discriminated record of audio backends. >> +# >> +# Since: 2.4 >> +## >> +{ 'union': 'AudiodevBackendOptions', >> + 'data': { >> + 'none': 'AudiodevNoOptions', >> + 'alsa': 'AudiodevAlsaOptions', >> + 'coreaudio': 'AudiodevNoOptions', >> + 'dsound': 'AudiodevDsoundOptions', >> + 'oss': 'AudiodevOssOptions', >> + 'pa': 'AudiodevPaOptions', >> + 'sdl': 'AudiodevNoOptions', >> + 'spice': 'AudiodevNoOptions', >> + 'wav': 'AudiodevWavOptions' } } >> + >> +## >> +# @AudioFormat >> +# >> +# An enumeration of possible audio formats. >> +# >> +# Since: 2.4 >> +## >> +{ 'enum': 'AudioFormat', >> + 'data': [ 'u8', 's8', 'u16', 's16', 'u32', 's32' ] } >> + >> +## >> +# @AudiodevPerDirectionOptions >> +# >> +# General audio backend options that are used for both playback and recording. >> +# >> +# @fixed-settings: #optional use fixed settings for host DAC/ADC >> +# >> +# @frequency: #optional frequency to use when using fixed settings >> +# >> +# @channels: #optional number of channels when using fixed settings >> +# >> +# @format: #optional sample format to use when using fixed settings > > Are these guys used when @fixed-settings is off? No. >> +# >> +# @buffer: #optional the buffer size (in microseconds) > > @buffer suggests this is a buffer, not a buffer length given as time > span. @buffer-len? Ok. (It used to be called buffer-usecs before I changed everything to microseconds.) > >> +# >> +# @buffer-count: #optional number of buffers >> +# >> +# Since: 2.4 >> +## >> +{ 'struct': 'AudiodevPerDirectionOptions', >> + 'data': { >> + '*fixed-settings': 'bool', >> + '*frequency': 'int', >> + '*channels': 'int', >> + '*voices': 'int', >> + '*format': 'AudioFormat', >> + '*buffer': 'int', >> + '*buffer-count': 'int' } } >> + >> +## >> +# @Audiodev >> +# >> +# Captures the configuration of an audio backend. >> +# >> +# @id: identifier of the backend >> +# >> +# @in: options of the capture stream >> +# >> +# @out: options of the playback stream >> +# >> +# @timer-period: #optional timer period (in microseconds, 0: use lowest >> +# possible) >> +# >> +# @opts: audio backend specific options >> +# >> +# Since: 2.4 >> +## >> +{ 'struct': 'Audiodev', >> + 'data': { >> + '*id': 'str', >> + 'in': 'AudiodevPerDirectionOptions', >> + 'out': 'AudiodevPerDirectionOptions', >> + '*timer-period': 'int', >> + 'opts': 'AudiodevBackendOptions' } } > > Have you considered making this a flat union, similar ro > BlockdevOptions? Not really. If you qapi masters out there think it's better, then I will convert it. > Don't get deceived by the number of my questions, this is solid work. > ^ permalink raw reply [flat|nested] 42+ messages in thread
* Re: [Qemu-devel] [PATCH v2 1/6] qapi: qapi for audio backends 2015-06-17 10:54 ` Kővágó Zoltán @ 2015-06-17 11:48 ` Markus Armbruster 2015-06-17 12:07 ` Kővágó Zoltán 2015-06-17 15:50 ` Eric Blake 0 siblings, 2 replies; 42+ messages in thread From: Markus Armbruster @ 2015-06-17 11:48 UTC (permalink / raw) To: Kővágó Zoltán; +Cc: Michael Roth, Gerd Hoffmann, qemu-devel "Kővágó Zoltán" <dirty.ice.hu@gmail.com> writes: > 2015-06-17 09:46 keltezéssel, Markus Armbruster írta: >> Copying Eric for additional QAPI schema expertise. >> >> My questions inline, pretty sure they show my ignorance. >> >> "Kővágó, Zoltán" <dirty.ice.hu@gmail.com> writes: >> >>> This patch adds structures into qapi to replace the existing configuration >>> structures used by audio backends currently. This qapi will be the base of the >>> -audiodev command line parameter (that replaces the old environment variables >>> based config). >>> >>> This is not a 1:1 translation of the old options, I've tried to make them much >>> more consistent (e.g. almost every backend had an option to specify buffer size, >>> but the name was different for every backend, and some backends required usecs, >>> while some other required frames, samples or bytes). Also tried to reduce the >>> number of abbreviations used by the config keys. >>> >>> Some of the more important changes: >>> * use `in` and `out` instead of `ADC` and `DAC`, as the former is more user >>> friendly imho >>> * moved buffer settings into the global setting area (so it's the same for all >>> backends that support it. Backends that can't change buffer size will simply >>> ignore them). Also using usecs, as it's probably more user friendly than >>> samples or bytes. >>> * try-poll is now an alsa and oss backend specific option (as all other backends >>> currently ignore it) >>> >>> Signed-off-by: Kővágó, Zoltán <DirtY.iCE.hu@gmail.com> >>> >>> --- >>> >>> Changes from v1 patch: >>> * every time-related field now take usecs (and removed -usecs, -millis suffixes) >>> * fixed inconsisten optional marking, language issues >>> >>> Changes from v2 RFC patch: >>> * in, out are no longer optional >>> * try-poll: moved to alsa and oss (as no other backend used them) >>> * voices: added (env variables had this option) >>> * dsound: removed primary buffer related fields >>> >>> Changes from v1 RFC patch: >>> * fixed style issues >>> * moved definitions into a separate file >>> * documented undocumented options (hopefully) >>> * removed plive option. It was useless even years ago so it can probably safely >>> go away: https://lists.nongnu.org/archive/html/qemu-devel/2012-03/msg02427.html >>> * removed verbose, debug options. Backends should use trace events instead. >>> * removed *_retries options from dsound. It's a kludge. >>> * moved buffer_usecs and buffer_count to the global config options. Some driver >>> might ignore it (as they do not expose API to change them). >>> * wav backend: removed frequecy, format, channels as AudiodevPerDirectionOptions >>> already have them. >>> >>> Makefile | 4 +- >>> qapi-schema.json | 3 + >>> qapi/audio.json | 223 +++++++++++++++++++++++++++++++++++++++++++++++++++++++ >>> 3 files changed, 228 insertions(+), 2 deletions(-) >>> create mode 100644 qapi/audio.json >>> >>> diff --git a/Makefile b/Makefile >>> index 3f97904..ac566fa 100644 >>> --- a/Makefile >>> +++ b/Makefile >>> @@ -257,8 +257,8 @@ $(SRC_PATH)/qga/qapi-schema.json $(SRC_PATH)/scripts/qapi-commands.py $(qapi-py) >>> " GEN $@") >>> >>> qapi-modules = $(SRC_PATH)/qapi-schema.json $(SRC_PATH)/qapi/common.json \ >>> - $(SRC_PATH)/qapi/block.json $(SRC_PATH)/qapi/block-core.json \ >>> - $(SRC_PATH)/qapi/event.json >>> + $(SRC_PATH)/qapi/audio.json $(SRC_PATH)/qapi/block.json \ >>> + $(SRC_PATH)/qapi/block-core.json $(SRC_PATH)/qapi/event.json >>> >>> qapi-types.c qapi-types.h :\ >>> $(qapi-modules) $(SRC_PATH)/scripts/qapi-types.py $(qapi-py) >>> diff --git a/qapi-schema.json b/qapi-schema.json >>> index 106008c..e751ea3 100644 >>> --- a/qapi-schema.json >>> +++ b/qapi-schema.json >>> @@ -5,6 +5,9 @@ >>> # QAPI common definitions >>> { 'include': 'qapi/common.json' } >>> >>> +# QAPI audio definitions >>> +{ 'include': 'qapi/audio.json' } >>> + >>> # QAPI block definitions >>> { 'include': 'qapi/block.json' } >>> >>> diff --git a/qapi/audio.json b/qapi/audio.json >>> new file mode 100644 >>> index 0000000..2851689 >>> --- /dev/null >>> +++ b/qapi/audio.json >>> @@ -0,0 +1,223 @@ >>> +# -*- mode: python -*- >>> +# >>> +# Copyright (C) 2015 Zoltán Kővágó <DirtY.iCE.hu@gmail.com> >>> +# >>> +# This work is licensed under the terms of the GNU GPL, version 2 or later. >>> +# See the COPYING file in the top-level directory. >>> + >>> +## >>> +# @AudiodevNoOptions >>> +# >>> +# The none, coreaudio, sdl and spice audio backend have no options. >>> +# >>> +# Since: 2.4 >>> +## >>> +{ 'struct': 'AudiodevNoOptions', >>> + 'data': { } } >>> + >>> +## >>> +# @AudiodevAlsaPerDirectionOptions >>> +# >>> +# Options of the alsa backend that are used for both playback and recording. >>> +# >>> +# @dev: #optional the name of the alsa device to use >> >> Who picks the default, QEMU or ALSA? > > It defaults to "default", which tells alsa to use the default device... Then make it # @dev: #optional the name of the alsa device to use (default 'default') >> >>> +# >>> +# @try-poll: #optional attempt to use poll mode >> >> What happens when the attempt fails? > > It falls back to non polling (timer based) mode. Okay, assuming that's the user interface we want. >>> +# >>> +# Since: 2.4 >>> +## >>> +{ 'struct': 'AudiodevAlsaPerDirectionOptions', >>> + 'data': { >>> + '*dev': 'str', >>> + '*try-poll': 'bool' } } >>> + >>> +## >>> +# @AudiodevAlsaOptions >>> +# >>> +# Options of the alsa audio backend. >>> +# >>> +# @in: options of the capture stream >>> +# >>> +# @out: options of the playback stream >>> +# >>> +# @threshold: #optional set the threshold (in frames) when playback starts >>> +# >>> +# Since: 2.4 >>> +## >>> +{ 'struct': 'AudiodevAlsaOptions', >>> + 'data': { >>> + 'in': 'AudiodevAlsaPerDirectionOptions', >>> + 'out': 'AudiodevAlsaPerDirectionOptions', >>> + '*threshold': 'int' } } >> >> Do we have an established term for a "direction"? >> >> If not: the use with 'in' and 'out' tempts me to name the type >> AudiodevAlsaIOOptions. >> >> Just rambling, use what you think is best. > > I don't know, so far nobody rambled about my naming... > >> >>> + >>> +## >>> +# @AudiodevDsoundOptions >>> +# >>> +# Options of the dsound audio backend. >>> +# >>> +# @latency: #optional add extra latency to playback (in microseconds) >> >> We already use seconds, milliseconds and nanoseconds. You make the zoo >> complete. > > Hmm. The audio backend previously used microseconds, milliseconds, > Hertz, frames, samples and bytes as time measurement, now at least > everything use microseconds. Milliseconds precision is not enough, > while nanosecond precision doesn't really make sense imho. I'm just poking fun at our inability to pick a time unit for QAPI/QMP. No objection to your use of microseconds. >>> +# >>> +# Since: 2.4 >>> +## >>> +{ 'struct': 'AudiodevDsoundOptions', >>> + 'data': { >>> + '*latency': 'int' } } >>> + >>> +## >>> +# @AudiodevOssPerDirectionOptions >>> +# >>> +# Options of the oss backend that are used for both playback and recording. >>> +# >>> +# @dev: #optional path of the oss device >> >> Who picks the default, QEMU or OSS? > > It defaults to "/dev/dsp". > >> For ALSA, you documented @dev to be "the name", here it's "path". >> Intentional difference? > > Yes. For oss you have to provide a filesystem path of the audio device > (like /dev/dsp), for alsa you provide an alsa specific name, like > "default", "hw:1,0", "hdmi:CARD=NVidia,DEV=1", whatever. I don't like calling a filename a path, even though we already do in many places. Let's do: # @dev: #optional file name of the oss device (default '/dev/dsp') >>> +# >>> +# @try-poll: #optional attempt to use poll mode >>> +# >>> +# Since: 2.4 >>> +## >>> +{ 'struct': 'AudiodevOssPerDirectionOptions', >>> + 'data': { >>> + '*dev': 'str', >>> + '*try-poll': 'bool' } } >> >> Same as for ALSA. Keeping them separate is fine with me. > > This is the same as alsa's try-poll. They're separate because only > these two backend use this setting. >> >>> + >>> +## >>> +# @AudiodevOssOptions >>> +# >>> +# Options of the oss audio backend. >>> +# >>> +# @in: options of the capture stream >>> +# >>> +# @out: options of the playback stream >>> +# >>> +# @mmap: #optional try using memory mapped access >> >> What happens when the attempt fails? > > It should fall back to non-mmapped access (should because when I > checked, it was buggy, at least on linux with alsa emulated oss on > pulseaudio alsa device...) Okay, assuming that's the user interface we want. >> Should this be called try-mmap? > Agree. > >>> +# >>> +# @exclusive: #optional open device in exclusive mode (vmix won't work) >>> +# >>> +# @dsp-policy: #optional set the timing policy of the device, -1 >>> to use fragment >>> +# mode (option ignored on some platforms) >> >> What are the possible values besides -1? > > It should be a number between 0 and 10 (according to this page: > http://manuals.opensound.com/developer/SNDCTL_DSP_POLICY.html) Please cover that in your comment. >>> +# >>> +# Since: 2.4 >>> +## >>> +{ 'struct': 'AudiodevOssOptions', >>> + 'data': { >>> + 'in': 'AudiodevOssPerDirectionOptions', >>> + 'out': 'AudiodevOssPerDirectionOptions', >>> + '*mmap': 'bool', >>> + '*exclusive': 'bool', >>> + '*dsp-policy': 'int' } } >>> + >>> +## >>> +# @AudiodevPaOptions >>> +# >>> +# Options of the pa (PulseAudio) audio backend. >>> +# >>> +# @server: #optional PulseAudio server address >>> +# >>> +# @sink: #optional sink device name >>> +# >>> +# @source: #optional source device name >> >> Who picks the defaults, QEMU or PA? > > PA Is there a way to explicitly ask for the PA default? Something like source=default? >>> +# >>> +# Since: 2.4 >>> +## >>> +{ 'struct': 'AudiodevPaOptions', >>> + 'data': { >>> + '*server': 'str', >>> + '*sink': 'str', >>> + '*source': 'str' } } >>> + >>> +## >>> +# @AudiodevWavOptions >>> +# >>> +# Options of the wav audio backend. >>> +# >>> +# @path: #optional path of the wav file to record >>> +# >>> +# Since: 2.4 >>> +## >>> +{ 'struct': 'AudiodevWavOptions', >>> + 'data': { >>> + '*path': 'str' } } >> >> Who picks the default? > > It defaults to "qemu.wav" Make it # @path: #optional path of the wav file to record (default 'qemu.wav') >>> + >>> + >>> +## >>> +# @AudiodevBackendOptions >>> +# >>> +# A discriminated record of audio backends. >>> +# >>> +# Since: 2.4 >>> +## >>> +{ 'union': 'AudiodevBackendOptions', >>> + 'data': { >>> + 'none': 'AudiodevNoOptions', >>> + 'alsa': 'AudiodevAlsaOptions', >>> + 'coreaudio': 'AudiodevNoOptions', >>> + 'dsound': 'AudiodevDsoundOptions', >>> + 'oss': 'AudiodevOssOptions', >>> + 'pa': 'AudiodevPaOptions', >>> + 'sdl': 'AudiodevNoOptions', >>> + 'spice': 'AudiodevNoOptions', >>> + 'wav': 'AudiodevWavOptions' } } >>> + >>> +## >>> +# @AudioFormat >>> +# >>> +# An enumeration of possible audio formats. >>> +# >>> +# Since: 2.4 >>> +## >>> +{ 'enum': 'AudioFormat', >>> + 'data': [ 'u8', 's8', 'u16', 's16', 'u32', 's32' ] } >>> + >>> +## >>> +# @AudiodevPerDirectionOptions >>> +# >>> +# General audio backend options that are used for both playback >>> and recording. >>> +# >>> +# @fixed-settings: #optional use fixed settings for host DAC/ADC >>> +# >>> +# @frequency: #optional frequency to use when using fixed settings >>> +# >>> +# @channels: #optional number of channels when using fixed settings >>> +# >>> +# @format: #optional sample format to use when using fixed settings >> >> Are these guys used when @fixed-settings is off? > > No. If @fixed-settings, are the other three all required? If not, what are their defaults? >>> +# >>> +# @buffer: #optional the buffer size (in microseconds) >> >> @buffer suggests this is a buffer, not a buffer length given as time >> span. @buffer-len? > > Ok. (It used to be called buffer-usecs before I changed everything to > microseconds.) > >> >>> +# >>> +# @buffer-count: #optional number of buffers >>> +# >>> +# Since: 2.4 >>> +## >>> +{ 'struct': 'AudiodevPerDirectionOptions', >>> + 'data': { >>> + '*fixed-settings': 'bool', >>> + '*frequency': 'int', >>> + '*channels': 'int', >>> + '*voices': 'int', >>> + '*format': 'AudioFormat', >>> + '*buffer': 'int', >>> + '*buffer-count': 'int' } } >>> + >>> +## >>> +# @Audiodev >>> +# >>> +# Captures the configuration of an audio backend. >>> +# >>> +# @id: identifier of the backend >>> +# >>> +# @in: options of the capture stream >>> +# >>> +# @out: options of the playback stream >>> +# >>> +# @timer-period: #optional timer period (in microseconds, 0: use lowest >>> +# possible) >>> +# >>> +# @opts: audio backend specific options >>> +# >>> +# Since: 2.4 >>> +## >>> +{ 'struct': 'Audiodev', >>> + 'data': { >>> + '*id': 'str', >>> + 'in': 'AudiodevPerDirectionOptions', >>> + 'out': 'AudiodevPerDirectionOptions', >>> + '*timer-period': 'int', >>> + 'opts': 'AudiodevBackendOptions' } } >> >> Have you considered making this a flat union, similar ro >> BlockdevOptions? > > Not really. If you qapi masters out there think it's better, then I > will convert it. Related: discussion about flattening in review of PATCH 2. >> Don't get deceived by the number of my questions, this is solid work. >> ^ permalink raw reply [flat|nested] 42+ messages in thread
* Re: [Qemu-devel] [PATCH v2 1/6] qapi: qapi for audio backends 2015-06-17 11:48 ` Markus Armbruster @ 2015-06-17 12:07 ` Kővágó Zoltán 2015-06-17 13:37 ` Markus Armbruster 2015-06-17 15:50 ` Eric Blake 1 sibling, 1 reply; 42+ messages in thread From: Kővágó Zoltán @ 2015-06-17 12:07 UTC (permalink / raw) To: Markus Armbruster; +Cc: Michael Roth, Gerd Hoffmann, qemu-devel 2015-06-17 13:48 keltezéssel, Markus Armbruster írta: > "Kővágó Zoltán" <dirty.ice.hu@gmail.com> writes: > >> 2015-06-17 09:46 keltezéssel, Markus Armbruster írta: >>> Copying Eric for additional QAPI schema expertise. >>> >>> My questions inline, pretty sure they show my ignorance. >>> >>> "Kővágó, Zoltán" <dirty.ice.hu@gmail.com> writes: >>> >>>> This patch adds structures into qapi to replace the existing configuration >>>> structures used by audio backends currently. This qapi will be the base of the >>>> -audiodev command line parameter (that replaces the old environment variables >>>> based config). >>>> >>>> This is not a 1:1 translation of the old options, I've tried to make them much >>>> more consistent (e.g. almost every backend had an option to specify buffer size, >>>> but the name was different for every backend, and some backends required usecs, >>>> while some other required frames, samples or bytes). Also tried to reduce the >>>> number of abbreviations used by the config keys. >>>> >>>> Some of the more important changes: >>>> * use `in` and `out` instead of `ADC` and `DAC`, as the former is more user >>>> friendly imho >>>> * moved buffer settings into the global setting area (so it's the same for all >>>> backends that support it. Backends that can't change buffer size will simply >>>> ignore them). Also using usecs, as it's probably more user friendly than >>>> samples or bytes. >>>> * try-poll is now an alsa and oss backend specific option (as all other backends >>>> currently ignore it) >>>> >>>> Signed-off-by: Kővágó, Zoltán <DirtY.iCE.hu@gmail.com> >>>> >>>> --- >>>> >>>> Changes from v1 patch: >>>> * every time-related field now take usecs (and removed -usecs, -millis suffixes) >>>> * fixed inconsisten optional marking, language issues >>>> >>>> Changes from v2 RFC patch: >>>> * in, out are no longer optional >>>> * try-poll: moved to alsa and oss (as no other backend used them) >>>> * voices: added (env variables had this option) >>>> * dsound: removed primary buffer related fields >>>> >>>> Changes from v1 RFC patch: >>>> * fixed style issues >>>> * moved definitions into a separate file >>>> * documented undocumented options (hopefully) >>>> * removed plive option. It was useless even years ago so it can probably safely >>>> go away: https://lists.nongnu.org/archive/html/qemu-devel/2012-03/msg02427.html >>>> * removed verbose, debug options. Backends should use trace events instead. >>>> * removed *_retries options from dsound. It's a kludge. >>>> * moved buffer_usecs and buffer_count to the global config options. Some driver >>>> might ignore it (as they do not expose API to change them). >>>> * wav backend: removed frequecy, format, channels as AudiodevPerDirectionOptions >>>> already have them. >>>> >>>> Makefile | 4 +- >>>> qapi-schema.json | 3 + >>>> qapi/audio.json | 223 +++++++++++++++++++++++++++++++++++++++++++++++++++++++ >>>> 3 files changed, 228 insertions(+), 2 deletions(-) >>>> create mode 100644 qapi/audio.json >>>> >>>> diff --git a/Makefile b/Makefile >>>> index 3f97904..ac566fa 100644 >>>> --- a/Makefile >>>> +++ b/Makefile >>>> @@ -257,8 +257,8 @@ $(SRC_PATH)/qga/qapi-schema.json $(SRC_PATH)/scripts/qapi-commands.py $(qapi-py) >>>> " GEN $@") >>>> >>>> qapi-modules = $(SRC_PATH)/qapi-schema.json $(SRC_PATH)/qapi/common.json \ >>>> - $(SRC_PATH)/qapi/block.json $(SRC_PATH)/qapi/block-core.json \ >>>> - $(SRC_PATH)/qapi/event.json >>>> + $(SRC_PATH)/qapi/audio.json $(SRC_PATH)/qapi/block.json \ >>>> + $(SRC_PATH)/qapi/block-core.json $(SRC_PATH)/qapi/event.json >>>> >>>> qapi-types.c qapi-types.h :\ >>>> $(qapi-modules) $(SRC_PATH)/scripts/qapi-types.py $(qapi-py) >>>> diff --git a/qapi-schema.json b/qapi-schema.json >>>> index 106008c..e751ea3 100644 >>>> --- a/qapi-schema.json >>>> +++ b/qapi-schema.json >>>> @@ -5,6 +5,9 @@ >>>> # QAPI common definitions >>>> { 'include': 'qapi/common.json' } >>>> >>>> +# QAPI audio definitions >>>> +{ 'include': 'qapi/audio.json' } >>>> + >>>> # QAPI block definitions >>>> { 'include': 'qapi/block.json' } >>>> >>>> diff --git a/qapi/audio.json b/qapi/audio.json >>>> new file mode 100644 >>>> index 0000000..2851689 >>>> --- /dev/null >>>> +++ b/qapi/audio.json >>>> @@ -0,0 +1,223 @@ >>>> +# -*- mode: python -*- >>>> +# >>>> +# Copyright (C) 2015 Zoltán Kővágó <DirtY.iCE.hu@gmail.com> >>>> +# >>>> +# This work is licensed under the terms of the GNU GPL, version 2 or later. >>>> +# See the COPYING file in the top-level directory. >>>> + >>>> +## >>>> +# @AudiodevNoOptions >>>> +# >>>> +# The none, coreaudio, sdl and spice audio backend have no options. >>>> +# >>>> +# Since: 2.4 >>>> +## >>>> +{ 'struct': 'AudiodevNoOptions', >>>> + 'data': { } } >>>> + >>>> +## >>>> +# @AudiodevAlsaPerDirectionOptions >>>> +# >>>> +# Options of the alsa backend that are used for both playback and recording. >>>> +# >>>> +# @dev: #optional the name of the alsa device to use >>> >>> Who picks the default, QEMU or ALSA? >> >> It defaults to "default", which tells alsa to use the default device... > > Then make it > > # @dev: #optional the name of the alsa device to use (default 'default') > >>> >>>> +# >>>> +# @try-poll: #optional attempt to use poll mode >>> >>> What happens when the attempt fails? >> >> It falls back to non polling (timer based) mode. > > Okay, assuming that's the user interface we want. > >>>> +# >>>> +# Since: 2.4 >>>> +## >>>> +{ 'struct': 'AudiodevAlsaPerDirectionOptions', >>>> + 'data': { >>>> + '*dev': 'str', >>>> + '*try-poll': 'bool' } } >>>> + >>>> +## >>>> +# @AudiodevAlsaOptions >>>> +# >>>> +# Options of the alsa audio backend. >>>> +# >>>> +# @in: options of the capture stream >>>> +# >>>> +# @out: options of the playback stream >>>> +# >>>> +# @threshold: #optional set the threshold (in frames) when playback starts >>>> +# >>>> +# Since: 2.4 >>>> +## >>>> +{ 'struct': 'AudiodevAlsaOptions', >>>> + 'data': { >>>> + 'in': 'AudiodevAlsaPerDirectionOptions', >>>> + 'out': 'AudiodevAlsaPerDirectionOptions', >>>> + '*threshold': 'int' } } >>> >>> Do we have an established term for a "direction"? >>> >>> If not: the use with 'in' and 'out' tempts me to name the type >>> AudiodevAlsaIOOptions. >>> >>> Just rambling, use what you think is best. >> >> I don't know, so far nobody rambled about my naming... >> >>> >>>> + >>>> +## >>>> +# @AudiodevDsoundOptions >>>> +# >>>> +# Options of the dsound audio backend. >>>> +# >>>> +# @latency: #optional add extra latency to playback (in microseconds) >>> >>> We already use seconds, milliseconds and nanoseconds. You make the zoo >>> complete. >> >> Hmm. The audio backend previously used microseconds, milliseconds, >> Hertz, frames, samples and bytes as time measurement, now at least >> everything use microseconds. Milliseconds precision is not enough, >> while nanosecond precision doesn't really make sense imho. > > I'm just poking fun at our inability to pick a time unit for QAPI/QMP. > No objection to your use of microseconds. > >>>> +# >>>> +# Since: 2.4 >>>> +## >>>> +{ 'struct': 'AudiodevDsoundOptions', >>>> + 'data': { >>>> + '*latency': 'int' } } >>>> + >>>> +## >>>> +# @AudiodevOssPerDirectionOptions >>>> +# >>>> +# Options of the oss backend that are used for both playback and recording. >>>> +# >>>> +# @dev: #optional path of the oss device >>> >>> Who picks the default, QEMU or OSS? >> >> It defaults to "/dev/dsp". >> >>> For ALSA, you documented @dev to be "the name", here it's "path". >>> Intentional difference? >> >> Yes. For oss you have to provide a filesystem path of the audio device >> (like /dev/dsp), for alsa you provide an alsa specific name, like >> "default", "hw:1,0", "hdmi:CARD=NVidia,DEV=1", whatever. > > I don't like calling a filename a path, even though we already do in > many places. Let's do: > > # @dev: #optional file name of the oss device (default '/dev/dsp') > >>>> +# >>>> +# @try-poll: #optional attempt to use poll mode >>>> +# >>>> +# Since: 2.4 >>>> +## >>>> +{ 'struct': 'AudiodevOssPerDirectionOptions', >>>> + 'data': { >>>> + '*dev': 'str', >>>> + '*try-poll': 'bool' } } >>> >>> Same as for ALSA. Keeping them separate is fine with me. >> >> This is the same as alsa's try-poll. They're separate because only >> these two backend use this setting. >>> >>>> + >>>> +## >>>> +# @AudiodevOssOptions >>>> +# >>>> +# Options of the oss audio backend. >>>> +# >>>> +# @in: options of the capture stream >>>> +# >>>> +# @out: options of the playback stream >>>> +# >>>> +# @mmap: #optional try using memory mapped access >>> >>> What happens when the attempt fails? >> >> It should fall back to non-mmapped access (should because when I >> checked, it was buggy, at least on linux with alsa emulated oss on >> pulseaudio alsa device...) > > Okay, assuming that's the user interface we want. > >>> Should this be called try-mmap? >> Agree. >> >>>> +# >>>> +# @exclusive: #optional open device in exclusive mode (vmix won't work) >>>> +# >>>> +# @dsp-policy: #optional set the timing policy of the device, -1 >>>> to use fragment >>>> +# mode (option ignored on some platforms) >>> >>> What are the possible values besides -1? >> >> It should be a number between 0 and 10 (according to this page: >> http://manuals.opensound.com/developer/SNDCTL_DSP_POLICY.html) > > Please cover that in your comment. > >>>> +# >>>> +# Since: 2.4 >>>> +## >>>> +{ 'struct': 'AudiodevOssOptions', >>>> + 'data': { >>>> + 'in': 'AudiodevOssPerDirectionOptions', >>>> + 'out': 'AudiodevOssPerDirectionOptions', >>>> + '*mmap': 'bool', >>>> + '*exclusive': 'bool', >>>> + '*dsp-policy': 'int' } } >>>> + >>>> +## >>>> +# @AudiodevPaOptions >>>> +# >>>> +# Options of the pa (PulseAudio) audio backend. >>>> +# >>>> +# @server: #optional PulseAudio server address >>>> +# >>>> +# @sink: #optional sink device name >>>> +# >>>> +# @source: #optional source device name >>> >>> Who picks the defaults, QEMU or PA? >> >> PA > > Is there a way to explicitly ask for the PA default? Something like > source=default? Not really right now. The default is a NULL pointer (pulseaudio api wise), so unless we add an arbitrary keyword (like default), it's not possible to ask explicitly for the default. (But omitting them will choose the default, of course.) >>>> +# >>>> +# Since: 2.4 >>>> +## >>>> +{ 'struct': 'AudiodevPaOptions', >>>> + 'data': { >>>> + '*server': 'str', >>>> + '*sink': 'str', >>>> + '*source': 'str' } } >>>> + >>>> +## >>>> +# @AudiodevWavOptions >>>> +# >>>> +# Options of the wav audio backend. >>>> +# >>>> +# @path: #optional path of the wav file to record >>>> +# >>>> +# Since: 2.4 >>>> +## >>>> +{ 'struct': 'AudiodevWavOptions', >>>> + 'data': { >>>> + '*path': 'str' } } >>> >>> Who picks the default? >> >> It defaults to "qemu.wav" > > Make it > > # @path: #optional path of the wav file to record (default 'qemu.wav') > >>>> + >>>> + >>>> +## >>>> +# @AudiodevBackendOptions >>>> +# >>>> +# A discriminated record of audio backends. >>>> +# >>>> +# Since: 2.4 >>>> +## >>>> +{ 'union': 'AudiodevBackendOptions', >>>> + 'data': { >>>> + 'none': 'AudiodevNoOptions', >>>> + 'alsa': 'AudiodevAlsaOptions', >>>> + 'coreaudio': 'AudiodevNoOptions', >>>> + 'dsound': 'AudiodevDsoundOptions', >>>> + 'oss': 'AudiodevOssOptions', >>>> + 'pa': 'AudiodevPaOptions', >>>> + 'sdl': 'AudiodevNoOptions', >>>> + 'spice': 'AudiodevNoOptions', >>>> + 'wav': 'AudiodevWavOptions' } } >>>> + >>>> +## >>>> +# @AudioFormat >>>> +# >>>> +# An enumeration of possible audio formats. >>>> +# >>>> +# Since: 2.4 >>>> +## >>>> +{ 'enum': 'AudioFormat', >>>> + 'data': [ 'u8', 's8', 'u16', 's16', 'u32', 's32' ] } >>>> + >>>> +## >>>> +# @AudiodevPerDirectionOptions >>>> +# >>>> +# General audio backend options that are used for both playback >>>> and recording. >>>> +# >>>> +# @fixed-settings: #optional use fixed settings for host DAC/ADC >>>> +# >>>> +# @frequency: #optional frequency to use when using fixed settings >>>> +# >>>> +# @channels: #optional number of channels when using fixed settings >>>> +# >>>> +# @format: #optional sample format to use when using fixed settings >>> >>> Are these guys used when @fixed-settings is off? >> >> No. > > If @fixed-settings, are the other three all required? If not, what are > their defaults? No, they all have defaults: 44100 Hz, 2 channels and s16 format. I guess I should also document it... >>>> +# >>>> +# @buffer: #optional the buffer size (in microseconds) >>> >>> @buffer suggests this is a buffer, not a buffer length given as time >>> span. @buffer-len? >> >> Ok. (It used to be called buffer-usecs before I changed everything to >> microseconds.) >> >>> >>>> +# >>>> +# @buffer-count: #optional number of buffers >>>> +# >>>> +# Since: 2.4 >>>> +## >>>> +{ 'struct': 'AudiodevPerDirectionOptions', >>>> + 'data': { >>>> + '*fixed-settings': 'bool', >>>> + '*frequency': 'int', >>>> + '*channels': 'int', >>>> + '*voices': 'int', >>>> + '*format': 'AudioFormat', >>>> + '*buffer': 'int', >>>> + '*buffer-count': 'int' } } >>>> + >>>> +## >>>> +# @Audiodev >>>> +# >>>> +# Captures the configuration of an audio backend. >>>> +# >>>> +# @id: identifier of the backend >>>> +# >>>> +# @in: options of the capture stream >>>> +# >>>> +# @out: options of the playback stream >>>> +# >>>> +# @timer-period: #optional timer period (in microseconds, 0: use lowest >>>> +# possible) >>>> +# >>>> +# @opts: audio backend specific options >>>> +# >>>> +# Since: 2.4 >>>> +## >>>> +{ 'struct': 'Audiodev', >>>> + 'data': { >>>> + '*id': 'str', >>>> + 'in': 'AudiodevPerDirectionOptions', >>>> + 'out': 'AudiodevPerDirectionOptions', >>>> + '*timer-period': 'int', >>>> + 'opts': 'AudiodevBackendOptions' } } >>> >>> Have you considered making this a flat union, similar ro >>> BlockdevOptions? >> >> Not really. If you qapi masters out there think it's better, then I >> will convert it. > > Related: discussion about flattening in review of PATCH 2. > >>> Don't get deceived by the number of my questions, this is solid work. >>> ^ permalink raw reply [flat|nested] 42+ messages in thread
* Re: [Qemu-devel] [PATCH v2 1/6] qapi: qapi for audio backends 2015-06-17 12:07 ` Kővágó Zoltán @ 2015-06-17 13:37 ` Markus Armbruster 2015-06-17 13:53 ` Kővágó Zoltán 0 siblings, 1 reply; 42+ messages in thread From: Markus Armbruster @ 2015-06-17 13:37 UTC (permalink / raw) To: Kővágó Zoltán; +Cc: qemu-devel, Michael Roth, Gerd Hoffmann "Kővágó Zoltán" <dirty.ice.hu@gmail.com> writes: > 2015-06-17 13:48 keltezéssel, Markus Armbruster írta: >> "Kővágó Zoltán" <dirty.ice.hu@gmail.com> writes: >> >>> 2015-06-17 09:46 keltezéssel, Markus Armbruster írta: >>>> Copying Eric for additional QAPI schema expertise. >>>> >>>> My questions inline, pretty sure they show my ignorance. >>>> >>>> "Kővágó, Zoltán" <dirty.ice.hu@gmail.com> writes: [...] >>>>> +## >>>>> +# @AudiodevPaOptions >>>>> +# >>>>> +# Options of the pa (PulseAudio) audio backend. >>>>> +# >>>>> +# @server: #optional PulseAudio server address >>>>> +# >>>>> +# @sink: #optional sink device name >>>>> +# >>>>> +# @source: #optional source device name >>>> >>>> Who picks the defaults, QEMU or PA? >>> >>> PA >> >> Is there a way to explicitly ask for the PA default? Something like >> source=default? > > Not really right now. The default is a NULL pointer (pulseaudio api > wise), so unless we add an arbitrary keyword (like default), it's not > possible to ask explicitly for the default. (But omitting them will > choose the default, of course.) Treating an empty string like NULL should get us a way to ask for the default, and a way to document the default concisely, like (default '') plus a suitable explanation what '' means. I'm not saying you should do that. I'm saying whatever you do, document what happens when an optional parameter is absent :) >>>>> +# >>>>> +# Since: 2.4 >>>>> +## >>>>> +{ 'struct': 'AudiodevPaOptions', >>>>> + 'data': { >>>>> + '*server': 'str', >>>>> + '*sink': 'str', >>>>> + '*source': 'str' } } >>>>> + >>>>> +## >>>>> +# @AudiodevWavOptions >>>>> +# >>>>> +# Options of the wav audio backend. >>>>> +# >>>>> +# @path: #optional path of the wav file to record >>>>> +# >>>>> +# Since: 2.4 >>>>> +## >>>>> +{ 'struct': 'AudiodevWavOptions', >>>>> + 'data': { >>>>> + '*path': 'str' } } >>>> >>>> Who picks the default? >>> >>> It defaults to "qemu.wav" >> >> Make it >> >> # @path: #optional path of the wav file to record (default 'qemu.wav') >> >>>>> + >>>>> + >>>>> +## >>>>> +# @AudiodevBackendOptions >>>>> +# >>>>> +# A discriminated record of audio backends. >>>>> +# >>>>> +# Since: 2.4 >>>>> +## >>>>> +{ 'union': 'AudiodevBackendOptions', >>>>> + 'data': { >>>>> + 'none': 'AudiodevNoOptions', >>>>> + 'alsa': 'AudiodevAlsaOptions', >>>>> + 'coreaudio': 'AudiodevNoOptions', >>>>> + 'dsound': 'AudiodevDsoundOptions', >>>>> + 'oss': 'AudiodevOssOptions', >>>>> + 'pa': 'AudiodevPaOptions', >>>>> + 'sdl': 'AudiodevNoOptions', >>>>> + 'spice': 'AudiodevNoOptions', >>>>> + 'wav': 'AudiodevWavOptions' } } >>>>> + >>>>> +## >>>>> +# @AudioFormat >>>>> +# >>>>> +# An enumeration of possible audio formats. >>>>> +# >>>>> +# Since: 2.4 >>>>> +## >>>>> +{ 'enum': 'AudioFormat', >>>>> + 'data': [ 'u8', 's8', 'u16', 's16', 'u32', 's32' ] } >>>>> + >>>>> +## >>>>> +# @AudiodevPerDirectionOptions >>>>> +# >>>>> +# General audio backend options that are used for both playback >>>>> and recording. >>>>> +# >>>>> +# @fixed-settings: #optional use fixed settings for host DAC/ADC >>>>> +# >>>>> +# @frequency: #optional frequency to use when using fixed settings >>>>> +# >>>>> +# @channels: #optional number of channels when using fixed settings >>>>> +# >>>>> +# @format: #optional sample format to use when using fixed settings >>>> >>>> Are these guys used when @fixed-settings is off? >>> >>> No. >> >> If @fixed-settings, are the other three all required? If not, what are >> their defaults? > > No, they all have defaults: 44100 Hz, 2 channels and s16 format. Okay, this sort of explains why you have @fixed-settings. My first thought was that @fixed-settings is redundant, because we can have any of @frequency, @channels, @format imply fixed settings. Except that doesn't let you ask for the *default* fixed settings, as you have to specify at least one. What's the default for @fixed-settings? What if I specify frequency, channels or format together with explicit fixed-settings: false? > I > guess I should also document it... Yes, please. >>>>> +# >>>>> +# @buffer: #optional the buffer size (in microseconds) >>>> >>>> @buffer suggests this is a buffer, not a buffer length given as time >>>> span. @buffer-len? >>> >>> Ok. (It used to be called buffer-usecs before I changed everything to >>> microseconds.) >>> >>>> >>>>> +# >>>>> +# @buffer-count: #optional number of buffers >>>>> +# >>>>> +# Since: 2.4 >>>>> +## >>>>> +{ 'struct': 'AudiodevPerDirectionOptions', >>>>> + 'data': { >>>>> + '*fixed-settings': 'bool', >>>>> + '*frequency': 'int', >>>>> + '*channels': 'int', >>>>> + '*voices': 'int', >>>>> + '*format': 'AudioFormat', >>>>> + '*buffer': 'int', >>>>> + '*buffer-count': 'int' } } [...] ^ permalink raw reply [flat|nested] 42+ messages in thread
* Re: [Qemu-devel] [PATCH v2 1/6] qapi: qapi for audio backends 2015-06-17 13:37 ` Markus Armbruster @ 2015-06-17 13:53 ` Kővágó Zoltán 2015-06-17 16:06 ` Markus Armbruster 0 siblings, 1 reply; 42+ messages in thread From: Kővágó Zoltán @ 2015-06-17 13:53 UTC (permalink / raw) To: Markus Armbruster; +Cc: qemu-devel, Michael Roth, Gerd Hoffmann 2015-06-17 15:37 keltezéssel, Markus Armbruster írta: > "Kővágó Zoltán" <dirty.ice.hu@gmail.com> writes: > >> 2015-06-17 13:48 keltezéssel, Markus Armbruster írta: >>> "Kővágó Zoltán" <dirty.ice.hu@gmail.com> writes: >>> >>>> 2015-06-17 09:46 keltezéssel, Markus Armbruster írta: >>>>> Copying Eric for additional QAPI schema expertise. >>>>> >>>>> My questions inline, pretty sure they show my ignorance. >>>>> >>>>> "Kővágó, Zoltán" <dirty.ice.hu@gmail.com> writes: > [...] >>>>>> +## >>>>>> +# @AudiodevPaOptions >>>>>> +# >>>>>> +# Options of the pa (PulseAudio) audio backend. >>>>>> +# >>>>>> +# @server: #optional PulseAudio server address >>>>>> +# >>>>>> +# @sink: #optional sink device name >>>>>> +# >>>>>> +# @source: #optional source device name >>>>> >>>>> Who picks the defaults, QEMU or PA? >>>> >>>> PA >>> >>> Is there a way to explicitly ask for the PA default? Something like >>> source=default? >> >> Not really right now. The default is a NULL pointer (pulseaudio api >> wise), so unless we add an arbitrary keyword (like default), it's not >> possible to ask explicitly for the default. (But omitting them will >> choose the default, of course.) > > Treating an empty string like NULL should get us a way to ask for the > default, and a way to document the default concisely, like (default '') > plus a suitable explanation what '' means. > > I'm not saying you should do that. I'm saying whatever you do, document > what happens when an optional parameter is absent :) > >>>>>> +# >>>>>> +# Since: 2.4 >>>>>> +## >>>>>> +{ 'struct': 'AudiodevPaOptions', >>>>>> + 'data': { >>>>>> + '*server': 'str', >>>>>> + '*sink': 'str', >>>>>> + '*source': 'str' } } >>>>>> + >>>>>> +## >>>>>> +# @AudiodevWavOptions >>>>>> +# >>>>>> +# Options of the wav audio backend. >>>>>> +# >>>>>> +# @path: #optional path of the wav file to record >>>>>> +# >>>>>> +# Since: 2.4 >>>>>> +## >>>>>> +{ 'struct': 'AudiodevWavOptions', >>>>>> + 'data': { >>>>>> + '*path': 'str' } } >>>>> >>>>> Who picks the default? >>>> >>>> It defaults to "qemu.wav" >>> >>> Make it >>> >>> # @path: #optional path of the wav file to record (default 'qemu.wav') >>> >>>>>> + >>>>>> + >>>>>> +## >>>>>> +# @AudiodevBackendOptions >>>>>> +# >>>>>> +# A discriminated record of audio backends. >>>>>> +# >>>>>> +# Since: 2.4 >>>>>> +## >>>>>> +{ 'union': 'AudiodevBackendOptions', >>>>>> + 'data': { >>>>>> + 'none': 'AudiodevNoOptions', >>>>>> + 'alsa': 'AudiodevAlsaOptions', >>>>>> + 'coreaudio': 'AudiodevNoOptions', >>>>>> + 'dsound': 'AudiodevDsoundOptions', >>>>>> + 'oss': 'AudiodevOssOptions', >>>>>> + 'pa': 'AudiodevPaOptions', >>>>>> + 'sdl': 'AudiodevNoOptions', >>>>>> + 'spice': 'AudiodevNoOptions', >>>>>> + 'wav': 'AudiodevWavOptions' } } >>>>>> + >>>>>> +## >>>>>> +# @AudioFormat >>>>>> +# >>>>>> +# An enumeration of possible audio formats. >>>>>> +# >>>>>> +# Since: 2.4 >>>>>> +## >>>>>> +{ 'enum': 'AudioFormat', >>>>>> + 'data': [ 'u8', 's8', 'u16', 's16', 'u32', 's32' ] } >>>>>> + >>>>>> +## >>>>>> +# @AudiodevPerDirectionOptions >>>>>> +# >>>>>> +# General audio backend options that are used for both playback >>>>>> and recording. >>>>>> +# >>>>>> +# @fixed-settings: #optional use fixed settings for host DAC/ADC >>>>>> +# >>>>>> +# @frequency: #optional frequency to use when using fixed settings >>>>>> +# >>>>>> +# @channels: #optional number of channels when using fixed settings >>>>>> +# >>>>>> +# @format: #optional sample format to use when using fixed settings >>>>> >>>>> Are these guys used when @fixed-settings is off? >>>> >>>> No. >>> >>> If @fixed-settings, are the other three all required? If not, what are >>> their defaults? >> >> No, they all have defaults: 44100 Hz, 2 channels and s16 format. > > Okay, this sort of explains why you have @fixed-settings. > > My first thought was that @fixed-settings is redundant, because we can > have any of @frequency, @channels, @format imply fixed settings. Except > that doesn't let you ask for the *default* fixed settings, as you have > to specify at least one. > > What's the default for @fixed-settings? It's on by default. > What if I specify frequency, channels or format together with explicit > fixed-settings: false? They will be ignored. The audio system currently work like this: when an audio frontend wants to open an output with some format (frequency, channels, format) it checks fixed-settings. If it's false, it will just open the stream with the frontend specified settings. If it's true, it'll convert it into the format specified by @frequency, @channels, @format, then pass this converted/recoded stream to the backend. > >> I >> guess I should also document it... > > Yes, please. > >>>>>> +# >>>>>> +# @buffer: #optional the buffer size (in microseconds) >>>>> >>>>> @buffer suggests this is a buffer, not a buffer length given as time >>>>> span. @buffer-len? >>>> >>>> Ok. (It used to be called buffer-usecs before I changed everything to >>>> microseconds.) >>>> >>>>> >>>>>> +# >>>>>> +# @buffer-count: #optional number of buffers >>>>>> +# >>>>>> +# Since: 2.4 >>>>>> +## >>>>>> +{ 'struct': 'AudiodevPerDirectionOptions', >>>>>> + 'data': { >>>>>> + '*fixed-settings': 'bool', >>>>>> + '*frequency': 'int', >>>>>> + '*channels': 'int', >>>>>> + '*voices': 'int', >>>>>> + '*format': 'AudioFormat', >>>>>> + '*buffer': 'int', >>>>>> + '*buffer-count': 'int' } } > [...] > ^ permalink raw reply [flat|nested] 42+ messages in thread
* Re: [Qemu-devel] [PATCH v2 1/6] qapi: qapi for audio backends 2015-06-17 13:53 ` Kővágó Zoltán @ 2015-06-17 16:06 ` Markus Armbruster 2015-06-18 0:21 ` Kővágó Zoltán 0 siblings, 1 reply; 42+ messages in thread From: Markus Armbruster @ 2015-06-17 16:06 UTC (permalink / raw) To: Kővágó Zoltán; +Cc: Gerd Hoffmann, qemu-devel, Michael Roth "Kővágó Zoltán" <dirty.ice.hu@gmail.com> writes: > 2015-06-17 15:37 keltezéssel, Markus Armbruster írta: >> "Kővágó Zoltán" <dirty.ice.hu@gmail.com> writes: >> >>> 2015-06-17 13:48 keltezéssel, Markus Armbruster írta: >>>> "Kővágó Zoltán" <dirty.ice.hu@gmail.com> writes: >>>> >>>>> 2015-06-17 09:46 keltezéssel, Markus Armbruster írta: [...] >>>>>>> +## >>>>>>> +# @AudiodevBackendOptions >>>>>>> +# >>>>>>> +# A discriminated record of audio backends. >>>>>>> +# >>>>>>> +# Since: 2.4 >>>>>>> +## >>>>>>> +{ 'union': 'AudiodevBackendOptions', >>>>>>> + 'data': { >>>>>>> + 'none': 'AudiodevNoOptions', >>>>>>> + 'alsa': 'AudiodevAlsaOptions', >>>>>>> + 'coreaudio': 'AudiodevNoOptions', >>>>>>> + 'dsound': 'AudiodevDsoundOptions', >>>>>>> + 'oss': 'AudiodevOssOptions', >>>>>>> + 'pa': 'AudiodevPaOptions', >>>>>>> + 'sdl': 'AudiodevNoOptions', >>>>>>> + 'spice': 'AudiodevNoOptions', >>>>>>> + 'wav': 'AudiodevWavOptions' } } >>>>>>> + >>>>>>> +## >>>>>>> +# @AudioFormat >>>>>>> +# >>>>>>> +# An enumeration of possible audio formats. >>>>>>> +# >>>>>>> +# Since: 2.4 >>>>>>> +## >>>>>>> +{ 'enum': 'AudioFormat', >>>>>>> + 'data': [ 'u8', 's8', 'u16', 's16', 'u32', 's32' ] } >>>>>>> + >>>>>>> +## >>>>>>> +# @AudiodevPerDirectionOptions >>>>>>> +# >>>>>>> +# General audio backend options that are used for both playback >>>>>>> and recording. >>>>>>> +# >>>>>>> +# @fixed-settings: #optional use fixed settings for host DAC/ADC >>>>>>> +# >>>>>>> +# @frequency: #optional frequency to use when using fixed settings >>>>>>> +# >>>>>>> +# @channels: #optional number of channels when using fixed settings >>>>>>> +# >>>>>>> +# @format: #optional sample format to use when using fixed settings >>>>>> >>>>>> Are these guys used when @fixed-settings is off? >>>>> >>>>> No. >>>> >>>> If @fixed-settings, are the other three all required? If not, what are >>>> their defaults? >>> >>> No, they all have defaults: 44100 Hz, 2 channels and s16 format. >> >> Okay, this sort of explains why you have @fixed-settings. >> >> My first thought was that @fixed-settings is redundant, because we can >> have any of @frequency, @channels, @format imply fixed settings. Except >> that doesn't let you ask for the *default* fixed settings, as you have >> to specify at least one. >> >> What's the default for @fixed-settings? > > It's on by default. > >> What if I specify frequency, channels or format together with explicit >> fixed-settings: false? > > They will be ignored. > > The audio system currently work like this: when an audio frontend > wants to open an output with some format (frequency, channels, format) > it checks fixed-settings. If it's false, it will just open the stream > with the frontend specified settings. If it's true, it'll convert it > into the format specified by @frequency, @channels, @format, then pass > this converted/recoded stream to the backend. So user typically specifies either fixed-settings=off, or any combination of the other three (including none of them). Correct? We could reject the non-sensical combination of fixed-settings=off plus any of the other three instead of silently ignoring their values. Matter of taste, your choice. Whatever you do, make sure to document how these four work together. Thank you for educating me so patiently. ^ permalink raw reply [flat|nested] 42+ messages in thread
* Re: [Qemu-devel] [PATCH v2 1/6] qapi: qapi for audio backends 2015-06-17 16:06 ` Markus Armbruster @ 2015-06-18 0:21 ` Kővágó Zoltán 2015-06-18 8:51 ` Markus Armbruster 0 siblings, 1 reply; 42+ messages in thread From: Kővágó Zoltán @ 2015-06-18 0:21 UTC (permalink / raw) To: Markus Armbruster; +Cc: Gerd Hoffmann, qemu-devel, Michael Roth 2015-06-17 18:06 keltezéssel, Markus Armbruster írta: > "Kővágó Zoltán" <dirty.ice.hu@gmail.com> writes: > >> 2015-06-17 15:37 keltezéssel, Markus Armbruster írta: >>> "Kővágó Zoltán" <dirty.ice.hu@gmail.com> writes: >>> >>>> 2015-06-17 13:48 keltezéssel, Markus Armbruster írta: >>>>> "Kővágó Zoltán" <dirty.ice.hu@gmail.com> writes: >>>>> >>>>>> 2015-06-17 09:46 keltezéssel, Markus Armbruster írta: > [...] >>>>>>>> +## >>>>>>>> +# @AudiodevBackendOptions >>>>>>>> +# >>>>>>>> +# A discriminated record of audio backends. >>>>>>>> +# >>>>>>>> +# Since: 2.4 >>>>>>>> +## >>>>>>>> +{ 'union': 'AudiodevBackendOptions', >>>>>>>> + 'data': { >>>>>>>> + 'none': 'AudiodevNoOptions', >>>>>>>> + 'alsa': 'AudiodevAlsaOptions', >>>>>>>> + 'coreaudio': 'AudiodevNoOptions', >>>>>>>> + 'dsound': 'AudiodevDsoundOptions', >>>>>>>> + 'oss': 'AudiodevOssOptions', >>>>>>>> + 'pa': 'AudiodevPaOptions', >>>>>>>> + 'sdl': 'AudiodevNoOptions', >>>>>>>> + 'spice': 'AudiodevNoOptions', >>>>>>>> + 'wav': 'AudiodevWavOptions' } } >>>>>>>> + >>>>>>>> +## >>>>>>>> +# @AudioFormat >>>>>>>> +# >>>>>>>> +# An enumeration of possible audio formats. >>>>>>>> +# >>>>>>>> +# Since: 2.4 >>>>>>>> +## >>>>>>>> +{ 'enum': 'AudioFormat', >>>>>>>> + 'data': [ 'u8', 's8', 'u16', 's16', 'u32', 's32' ] } >>>>>>>> + >>>>>>>> +## >>>>>>>> +# @AudiodevPerDirectionOptions >>>>>>>> +# >>>>>>>> +# General audio backend options that are used for both playback >>>>>>>> and recording. >>>>>>>> +# >>>>>>>> +# @fixed-settings: #optional use fixed settings for host DAC/ADC >>>>>>>> +# >>>>>>>> +# @frequency: #optional frequency to use when using fixed settings >>>>>>>> +# >>>>>>>> +# @channels: #optional number of channels when using fixed settings >>>>>>>> +# >>>>>>>> +# @format: #optional sample format to use when using fixed settings >>>>>>> >>>>>>> Are these guys used when @fixed-settings is off? >>>>>> >>>>>> No. >>>>> >>>>> If @fixed-settings, are the other three all required? If not, what are >>>>> their defaults? >>>> >>>> No, they all have defaults: 44100 Hz, 2 channels and s16 format. >>> >>> Okay, this sort of explains why you have @fixed-settings. >>> >>> My first thought was that @fixed-settings is redundant, because we can >>> have any of @frequency, @channels, @format imply fixed settings. Except >>> that doesn't let you ask for the *default* fixed settings, as you have >>> to specify at least one. >>> >>> What's the default for @fixed-settings? >> >> It's on by default. >> >>> What if I specify frequency, channels or format together with explicit >>> fixed-settings: false? >> >> They will be ignored. >> >> The audio system currently work like this: when an audio frontend >> wants to open an output with some format (frequency, channels, format) >> it checks fixed-settings. If it's false, it will just open the stream >> with the frontend specified settings. If it's true, it'll convert it >> into the format specified by @frequency, @channels, @format, then pass >> this converted/recoded stream to the backend. > > So user typically specifies either fixed-settings=off, or any > combination of the other three (including none of them). Correct? > > We could reject the non-sensical combination of fixed-settings=off plus > any of the other three instead of silently ignoring their values. > Matter of taste, your choice. > > Whatever you do, make sure to document how these four work together. > > Thank you for educating me so patiently. The audio backend currently works like that you can pass any non-sensical values to it, like negative frequency, or 'kdp' count of channels, it will silently fallback to some default value, or just fail, but qemu will continue to run. We can make the new config more strict (and we should, I think), so if you have any idea where should we be more strict (without creating a backward compatibility headache), don't hesitate to point it out. ^ permalink raw reply [flat|nested] 42+ messages in thread
* Re: [Qemu-devel] [PATCH v2 1/6] qapi: qapi for audio backends 2015-06-18 0:21 ` Kővágó Zoltán @ 2015-06-18 8:51 ` Markus Armbruster 0 siblings, 0 replies; 42+ messages in thread From: Markus Armbruster @ 2015-06-18 8:51 UTC (permalink / raw) To: Kővágó Zoltán; +Cc: Michael Roth, Gerd Hoffmann, qemu-devel "Kővágó Zoltán" <dirty.ice.hu@gmail.com> writes: > 2015-06-17 18:06 keltezéssel, Markus Armbruster írta: >> "Kővágó Zoltán" <dirty.ice.hu@gmail.com> writes: >> >>> 2015-06-17 15:37 keltezéssel, Markus Armbruster írta: >>>> "Kővágó Zoltán" <dirty.ice.hu@gmail.com> writes: >>>> >>>>> 2015-06-17 13:48 keltezéssel, Markus Armbruster írta: >>>>>> "Kővágó Zoltán" <dirty.ice.hu@gmail.com> writes: >>>>>> >>>>>>> 2015-06-17 09:46 keltezéssel, Markus Armbruster írta: >> [...] >>>>>>>>> +## >>>>>>>>> +# @AudiodevBackendOptions >>>>>>>>> +# >>>>>>>>> +# A discriminated record of audio backends. >>>>>>>>> +# >>>>>>>>> +# Since: 2.4 >>>>>>>>> +## >>>>>>>>> +{ 'union': 'AudiodevBackendOptions', >>>>>>>>> + 'data': { >>>>>>>>> + 'none': 'AudiodevNoOptions', >>>>>>>>> + 'alsa': 'AudiodevAlsaOptions', >>>>>>>>> + 'coreaudio': 'AudiodevNoOptions', >>>>>>>>> + 'dsound': 'AudiodevDsoundOptions', >>>>>>>>> + 'oss': 'AudiodevOssOptions', >>>>>>>>> + 'pa': 'AudiodevPaOptions', >>>>>>>>> + 'sdl': 'AudiodevNoOptions', >>>>>>>>> + 'spice': 'AudiodevNoOptions', >>>>>>>>> + 'wav': 'AudiodevWavOptions' } } >>>>>>>>> + >>>>>>>>> +## >>>>>>>>> +# @AudioFormat >>>>>>>>> +# >>>>>>>>> +# An enumeration of possible audio formats. >>>>>>>>> +# >>>>>>>>> +# Since: 2.4 >>>>>>>>> +## >>>>>>>>> +{ 'enum': 'AudioFormat', >>>>>>>>> + 'data': [ 'u8', 's8', 'u16', 's16', 'u32', 's32' ] } >>>>>>>>> + >>>>>>>>> +## >>>>>>>>> +# @AudiodevPerDirectionOptions >>>>>>>>> +# >>>>>>>>> +# General audio backend options that are used for both playback >>>>>>>>> and recording. >>>>>>>>> +# >>>>>>>>> +# @fixed-settings: #optional use fixed settings for host DAC/ADC >>>>>>>>> +# >>>>>>>>> +# @frequency: #optional frequency to use when using fixed settings >>>>>>>>> +# >>>>>>>>> +# @channels: #optional number of channels when using fixed settings >>>>>>>>> +# >>>>>>>>> +# @format: #optional sample format to use when using fixed settings >>>>>>>> >>>>>>>> Are these guys used when @fixed-settings is off? >>>>>>> >>>>>>> No. >>>>>> >>>>>> If @fixed-settings, are the other three all required? If not, what are >>>>>> their defaults? >>>>> >>>>> No, they all have defaults: 44100 Hz, 2 channels and s16 format. >>>> >>>> Okay, this sort of explains why you have @fixed-settings. >>>> >>>> My first thought was that @fixed-settings is redundant, because we can >>>> have any of @frequency, @channels, @format imply fixed settings. Except >>>> that doesn't let you ask for the *default* fixed settings, as you have >>>> to specify at least one. >>>> >>>> What's the default for @fixed-settings? >>> >>> It's on by default. >>> >>>> What if I specify frequency, channels or format together with explicit >>>> fixed-settings: false? >>> >>> They will be ignored. >>> >>> The audio system currently work like this: when an audio frontend >>> wants to open an output with some format (frequency, channels, format) >>> it checks fixed-settings. If it's false, it will just open the stream >>> with the frontend specified settings. If it's true, it'll convert it >>> into the format specified by @frequency, @channels, @format, then pass >>> this converted/recoded stream to the backend. >> >> So user typically specifies either fixed-settings=off, or any >> combination of the other three (including none of them). Correct? >> >> We could reject the non-sensical combination of fixed-settings=off plus >> any of the other three instead of silently ignoring their values. >> Matter of taste, your choice. >> >> Whatever you do, make sure to document how these four work together. >> >> Thank you for educating me so patiently. > > The audio backend currently works like that you can pass any > non-sensical values to it, like negative frequency, or 'kdp' count of > channels, it will silently fallback to some default value, or just > fail, but qemu will continue to run. We can make the new config more > strict (and we should, I think), so if you have any idea where should > we be more strict (without creating a backward compatibility > headache), don't hesitate to point it out. When a sensible default value exists, making the parameter optional is usually a good idea. We should refuse to start on non-sensical configuration, not silently substitute defaults or disable the affected component (here: audio backend). I can't give you more specific guidance, because I'm an audio ignoramus :) ^ permalink raw reply [flat|nested] 42+ messages in thread
* Re: [Qemu-devel] [PATCH v2 1/6] qapi: qapi for audio backends 2015-06-17 11:48 ` Markus Armbruster 2015-06-17 12:07 ` Kővágó Zoltán @ 2015-06-17 15:50 ` Eric Blake 1 sibling, 0 replies; 42+ messages in thread From: Eric Blake @ 2015-06-17 15:50 UTC (permalink / raw) To: Markus Armbruster, Kővágó Zoltán Cc: qemu-devel, Michael Roth, Gerd Hoffmann [-- Attachment #1: Type: text/plain, Size: 1105 bytes --] On 06/17/2015 05:48 AM, Markus Armbruster wrote: > "Kővágó Zoltán" <dirty.ice.hu@gmail.com> writes: > >> 2015-06-17 09:46 keltezéssel, Markus Armbruster írta: >>> Copying Eric for additional QAPI schema expertise. >>> >>>> +{ 'struct': 'Audiodev', >>>> + 'data': { >>>> + '*id': 'str', >>>> + 'in': 'AudiodevPerDirectionOptions', >>>> + 'out': 'AudiodevPerDirectionOptions', >>>> + '*timer-period': 'int', >>>> + 'opts': 'AudiodevBackendOptions' } } >>> >>> Have you considered making this a flat union, similar ro >>> BlockdevOptions? >> >> Not really. If you qapi masters out there think it's better, then I >> will convert it. > > Related: discussion about flattening in review of PATCH 2. Indeed - I think a flat union makes for nicer command line structure, and fewer {} nesting in QMP structure. I still need to spend time on the overall thread, but wanted to chime in on this comment up front. -- Eric Blake eblake redhat com +1-919-301-3266 Libvirt virtualization library http://libvirt.org [-- Attachment #2: OpenPGP digital signature --] [-- Type: application/pgp-signature, Size: 604 bytes --] ^ permalink raw reply [flat|nested] 42+ messages in thread
* [Qemu-devel] [PATCH v2 2/6] qapi: support nested structs in OptsVisitor 2015-06-16 12:49 [Qemu-devel] [PATCH v2 0/6] -audiodev option Kővágó, Zoltán 2015-06-16 12:49 ` [Qemu-devel] [PATCH v2 1/6] qapi: qapi for audio backends Kővágó, Zoltán @ 2015-06-16 12:49 ` Kővágó, Zoltán 2015-06-17 7:50 ` Markus Armbruster 2015-06-16 12:49 ` [Qemu-devel] [PATCH v2 3/6] opts: do not print separator before first item in qemu_opts_print Kővágó, Zoltán ` (3 subsequent siblings) 5 siblings, 1 reply; 42+ messages in thread From: Kővágó, Zoltán @ 2015-06-16 12:49 UTC (permalink / raw) To: qemu-devel; +Cc: Michael Roth, Gerd Hoffmann, Markus Armbruster The current OptsVisitor flattens the whole structure, if there are same named fields under different paths (like `in' and `out' in `Audiodev'), the current visitor can't cope with them (for example setting `frequency=44100' will set the in's frequency to 44100 and leave out's frequency unspecified). This patch fixes it, by the following changes: 1) Specifying just the field name will apply to all fields that has the specified name (this means it would set both in's and out's frequency to 44100 in the above example). 2) Optionally user can specify the path in the hierarchy. Names are separated by a dot (e.g. `in.frequency', `foo.bar.something', etc). The user need not specify the whole path, only the last few components (i.e. `bar.something' is equivalent to `foo.bar.something' if only `foo' has a `bar' field). This way 1) is just a special case of this when only the last component is specified. 3) In case of an ambiguity (e.g `frequency=44100,in.frequency=8000') the longest matching (the most specific) path wins (so in this example, in's frequency would become 8000, because `in.frequency' is more specific that `frequency', and out's frequency would become 44100, because only `frequency' matches it). Signed-off-by: Kővágó, Zoltán <DirtY.iCE.hu@gmail.com> --- qapi/opts-visitor.c | 144 +++++++++++++++++++++++++------- tests/qapi-schema/qapi-schema-test.json | 9 +- tests/test-opts-visitor.c | 34 ++++++++ 3 files changed, 157 insertions(+), 30 deletions(-) diff --git a/qapi/opts-visitor.c b/qapi/opts-visitor.c index f2ad6d7..409d8b7 100644 --- a/qapi/opts-visitor.c +++ b/qapi/opts-visitor.c @@ -64,13 +64,14 @@ struct OptsVisitor /* Non-null iff depth is positive. Each key is a QemuOpt name. Each value * is a non-empty GQueue, enumerating all QemuOpt occurrences with that * name. */ - GHashTable *unprocessed_opts; + GHashTable *unprocessed_opts, *opts; /* The list currently being traversed with opts_start_list() / * opts_next_list(). The list must have a struct element type in the * schema, with a single mandatory scalar member. */ ListMode list_mode; GQueue *repeated_opts; + char *repeated_name; /* When parsing a list of repeating options as integers, values of the form * "a-b", representing a closed interval, are allowed. Elements in the @@ -86,6 +87,9 @@ struct OptsVisitor * not survive or escape the OptsVisitor object. */ QemuOpt *fake_id_opt; + + /* List of field names leading to the current structure. */ + GQueue *nested_names; }; @@ -97,11 +101,12 @@ destroy_list(gpointer list) static void -opts_visitor_insert(GHashTable *unprocessed_opts, const QemuOpt *opt) +opts_visitor_insert(OptsVisitor *ov, const QemuOpt *opt) { GQueue *list; + assert(opt); - list = g_hash_table_lookup(unprocessed_opts, opt->name); + list = g_hash_table_lookup(ov->opts, opt->name); if (list == NULL) { list = g_queue_new(); @@ -109,7 +114,8 @@ opts_visitor_insert(GHashTable *unprocessed_opts, const QemuOpt *opt) * "key_destroy_func" in opts_start_struct(). Thus cast away key * const-ness in order to suppress gcc's warning. */ - g_hash_table_insert(unprocessed_opts, (gpointer)opt->name, list); + g_hash_table_insert(ov->opts, (gpointer)opt->name, list); + g_hash_table_insert(ov->unprocessed_opts, (gpointer)opt->name, list); } /* Similarly, destroy_list() doesn't call g_queue_free_full(). */ @@ -127,17 +133,27 @@ opts_start_struct(Visitor *v, void **obj, const char *kind, if (obj) { *obj = g_malloc0(size > 0 ? size : 1); } + + /* assuming name is a statically allocated string (or at least it's lifetime + * is longer than the visitor's) */ + if (!name) { + name = ""; + } + g_queue_push_tail(ov->nested_names, (gpointer) name); + if (ov->depth++ > 0) { return; } - ov->unprocessed_opts = g_hash_table_new_full(&g_str_hash, &g_str_equal, - NULL, &destroy_list); + ov->opts = g_hash_table_new_full(&g_str_hash, &g_str_equal, + NULL, &destroy_list); + ov->unprocessed_opts = g_hash_table_new(&g_str_hash, &g_str_equal); + QTAILQ_FOREACH(opt, &ov->opts_root->head, next) { /* ensured by qemu-option.c::opts_do_parse() */ assert(strcmp(opt->name, "id") != 0); - opts_visitor_insert(ov->unprocessed_opts, opt); + opts_visitor_insert(ov, opt); } if (ov->opts_root->id != NULL) { @@ -145,7 +161,7 @@ opts_start_struct(Visitor *v, void **obj, const char *kind, ov->fake_id_opt->name = g_strdup("id"); ov->fake_id_opt->str = g_strdup(ov->opts_root->id); - opts_visitor_insert(ov->unprocessed_opts, ov->fake_id_opt); + opts_visitor_insert(ov, ov->fake_id_opt); } } @@ -163,6 +179,8 @@ opts_end_struct(Visitor *v, Error **errp) OptsVisitor *ov = DO_UPCAST(OptsVisitor, visitor, v); GQueue *any; + g_queue_pop_tail(ov->nested_names); + if (--ov->depth > 0) { return; } @@ -177,6 +195,8 @@ opts_end_struct(Visitor *v, Error **errp) } g_hash_table_destroy(ov->unprocessed_opts); ov->unprocessed_opts = NULL; + g_hash_table_destroy(ov->opts); + ov->opts = NULL; if (ov->fake_id_opt) { g_free(ov->fake_id_opt->name); g_free(ov->fake_id_opt->str); @@ -185,16 +205,56 @@ opts_end_struct(Visitor *v, Error **errp) ov->fake_id_opt = NULL; } +static void +sum_strlen(gpointer data, gpointer user_data) +{ + const char *str = data; + size_t *sum_len = user_data; + *sum_len += strlen(str) + 1; +} + +static void +append_str(gpointer data, gpointer user_data) +{ + strcat(user_data, data); + strcat(user_data, "."); +} + +/* lookup a name, trying from the most qualified version (e.g. foo.bar.asd) to + * least qualified ones (i.e. foo.bar.asd overrides bar.asd or asd) */ static GQueue * -lookup_distinct(const OptsVisitor *ov, const char *name, Error **errp) +lookup_distinct(const OptsVisitor *ov, const char *name, char **out_key, + Error **errp) { - GQueue *list; + GQueue *list = NULL; + char *key, *key2; + size_t sum_len = strlen(name); - list = g_hash_table_lookup(ov->unprocessed_opts, name); + g_queue_foreach(ov->nested_names, sum_strlen, &sum_len); + key = g_malloc(sum_len+1); + key[0] = 0; + g_queue_foreach(ov->nested_names, append_str, key); + strcat(key, name); + + key2 = key; + while (*key2) { + list = g_hash_table_lookup(ov->opts, key2); + if (list) { + if (out_key) { + *out_key = g_strdup(key2); + } + break; + } + + while (*key2 && *key2++ != '.') { + } + } if (!list) { error_set(errp, QERR_MISSING_PARAMETER, name); } + + g_free(key); return list; } @@ -206,7 +266,7 @@ opts_start_list(Visitor *v, const char *name, Error **errp) /* we can't traverse a list in a list */ assert(ov->list_mode == LM_NONE); - ov->repeated_opts = lookup_distinct(ov, name, errp); + ov->repeated_opts = lookup_distinct(ov, name, &ov->repeated_name, errp); if (ov->repeated_opts != NULL) { ov->list_mode = LM_STARTED; } @@ -242,11 +302,9 @@ opts_next_list(Visitor *v, GenericList **list, Error **errp) /* range has been completed, fall through in order to pop option */ case LM_IN_PROGRESS: { - const QemuOpt *opt; - - opt = g_queue_pop_head(ov->repeated_opts); + g_queue_pop_head(ov->repeated_opts); if (g_queue_is_empty(ov->repeated_opts)) { - g_hash_table_remove(ov->unprocessed_opts, opt->name); + g_hash_table_remove(ov->unprocessed_opts, ov->repeated_name); return NULL; } link = &(*list)->next; @@ -272,22 +330,28 @@ opts_end_list(Visitor *v, Error **errp) ov->list_mode == LM_SIGNED_INTERVAL || ov->list_mode == LM_UNSIGNED_INTERVAL); ov->repeated_opts = NULL; + + g_free(ov->repeated_name); + ov->repeated_name = NULL; + ov->list_mode = LM_NONE; } static const QemuOpt * -lookup_scalar(const OptsVisitor *ov, const char *name, Error **errp) +lookup_scalar(const OptsVisitor *ov, const char *name, char** out_key, + Error **errp) { if (ov->list_mode == LM_NONE) { GQueue *list; /* the last occurrence of any QemuOpt takes effect when queried by name */ - list = lookup_distinct(ov, name, errp); + list = lookup_distinct(ov, name, out_key, errp); return list ? g_queue_peek_tail(list) : NULL; } assert(ov->list_mode == LM_IN_PROGRESS); + assert(out_key == NULL || *out_key == NULL); return g_queue_peek_head(ov->repeated_opts); } @@ -309,13 +373,15 @@ opts_type_str(Visitor *v, char **obj, const char *name, Error **errp) { OptsVisitor *ov = DO_UPCAST(OptsVisitor, visitor, v); const QemuOpt *opt; + char *key = NULL; - opt = lookup_scalar(ov, name, errp); + opt = lookup_scalar(ov, name, &key, errp); if (!opt) { return; } *obj = g_strdup(opt->str ? opt->str : ""); - processed(ov, name); + processed(ov, key); + g_free(key); } @@ -325,8 +391,9 @@ opts_type_bool(Visitor *v, bool *obj, const char *name, Error **errp) { OptsVisitor *ov = DO_UPCAST(OptsVisitor, visitor, v); const QemuOpt *opt; + char *key = NULL; - opt = lookup_scalar(ov, name, errp); + opt = lookup_scalar(ov, name, &key, errp); if (!opt) { return; } @@ -343,13 +410,15 @@ opts_type_bool(Visitor *v, bool *obj, const char *name, Error **errp) } else { error_set(errp, QERR_INVALID_PARAMETER_VALUE, opt->name, "on|yes|y|off|no|n"); + g_free(key); return; } } else { *obj = true; } - processed(ov, name); + processed(ov, key); + g_free(key); } @@ -361,13 +430,14 @@ opts_type_int(Visitor *v, int64_t *obj, const char *name, Error **errp) const char *str; long long val; char *endptr; + char *key = NULL; if (ov->list_mode == LM_SIGNED_INTERVAL) { *obj = ov->range_next.s; return; } - opt = lookup_scalar(ov, name, errp); + opt = lookup_scalar(ov, name, &key, errp); if (!opt) { return; } @@ -381,11 +451,13 @@ opts_type_int(Visitor *v, int64_t *obj, const char *name, Error **errp) if (errno == 0 && endptr > str && INT64_MIN <= val && val <= INT64_MAX) { if (*endptr == '\0') { *obj = val; - processed(ov, name); + processed(ov, key); + g_free(key); return; } if (*endptr == '-' && ov->list_mode == LM_IN_PROGRESS) { long long val2; + assert(key == NULL); str = endptr + 1; val2 = strtoll(str, &endptr, 0); @@ -406,6 +478,7 @@ opts_type_int(Visitor *v, int64_t *obj, const char *name, Error **errp) error_set(errp, QERR_INVALID_PARAMETER_VALUE, opt->name, (ov->list_mode == LM_NONE) ? "an int64 value" : "an int64 value or range"); + g_free(key); } @@ -417,13 +490,14 @@ opts_type_uint64(Visitor *v, uint64_t *obj, const char *name, Error **errp) const char *str; unsigned long long val; char *endptr; + char *key = NULL; if (ov->list_mode == LM_UNSIGNED_INTERVAL) { *obj = ov->range_next.u; return; } - opt = lookup_scalar(ov, name, errp); + opt = lookup_scalar(ov, name, &key, errp); if (!opt) { return; } @@ -435,11 +509,13 @@ opts_type_uint64(Visitor *v, uint64_t *obj, const char *name, Error **errp) if (parse_uint(str, &val, &endptr, 0) == 0 && val <= UINT64_MAX) { if (*endptr == '\0') { *obj = val; - processed(ov, name); + processed(ov, key); + g_free(key); return; } if (*endptr == '-' && ov->list_mode == LM_IN_PROGRESS) { unsigned long long val2; + assert(key == NULL); str = endptr + 1; if (parse_uint_full(str, &val2, 0) == 0 && @@ -458,6 +534,7 @@ opts_type_uint64(Visitor *v, uint64_t *obj, const char *name, Error **errp) error_set(errp, QERR_INVALID_PARAMETER_VALUE, opt->name, (ov->list_mode == LM_NONE) ? "a uint64 value" : "a uint64 value or range"); + g_free(key); } @@ -468,8 +545,9 @@ opts_type_size(Visitor *v, uint64_t *obj, const char *name, Error **errp) const QemuOpt *opt; int64_t val; char *endptr; + char *key = NULL; - opt = lookup_scalar(ov, name, errp); + opt = lookup_scalar(ov, name, &key, errp); if (!opt) { return; } @@ -479,11 +557,13 @@ opts_type_size(Visitor *v, uint64_t *obj, const char *name, Error **errp) if (val < 0 || *endptr) { error_set(errp, QERR_INVALID_PARAMETER_VALUE, opt->name, "a size value representible as a non-negative int64"); + g_free(key); return; } *obj = val; - processed(ov, name); + processed(ov, key); + g_free(key); } @@ -494,7 +574,7 @@ opts_optional(Visitor *v, bool *present, const char *name, Error **errp) /* we only support a single mandatory scalar field in a list node */ assert(ov->list_mode == LM_NONE); - *present = (lookup_distinct(ov, name, NULL) != NULL); + *present = (lookup_distinct(ov, name, NULL, NULL) != NULL); } @@ -505,6 +585,8 @@ opts_visitor_new(const QemuOpts *opts) ov = g_malloc0(sizeof *ov); + ov->nested_names = g_queue_new(); + ov->visitor.start_struct = &opts_start_struct; ov->visitor.end_struct = &opts_end_struct; @@ -545,6 +627,10 @@ opts_visitor_cleanup(OptsVisitor *ov) if (ov->unprocessed_opts != NULL) { g_hash_table_destroy(ov->unprocessed_opts); } + if (ov->opts != NULL) { + g_hash_table_destroy(ov->opts); + } + g_queue_free(ov->nested_names); g_free(ov->fake_id_opt); g_free(ov); } diff --git a/tests/qapi-schema/qapi-schema-test.json b/tests/qapi-schema/qapi-schema-test.json index c7eaa86..a818eff 100644 --- a/tests/qapi-schema/qapi-schema-test.json +++ b/tests/qapi-schema/qapi-schema-test.json @@ -81,6 +81,11 @@ { 'command': 'user_def_cmd3', 'data': {'a': 'int', '*b': 'int' }, 'returns': 'int' } +# For testing hierarchy support in opts-visitor +{ 'struct': 'UserDefOptionsSub', + 'data': { + '*nint': 'int' } } + # For testing integer range flattening in opts-visitor. The following schema # corresponds to the option format: # @@ -94,7 +99,9 @@ '*u64' : [ 'uint64' ], '*u16' : [ 'uint16' ], '*i64x': 'int' , - '*u64x': 'uint64' } } + '*u64x': 'uint64' , + 'sub0': 'UserDefOptionsSub', + 'sub1': 'UserDefOptionsSub' } } # testing event { 'struct': 'EventStructOne', diff --git a/tests/test-opts-visitor.c b/tests/test-opts-visitor.c index ebeee5d..5862c7c 100644 --- a/tests/test-opts-visitor.c +++ b/tests/test-opts-visitor.c @@ -177,6 +177,34 @@ expect_u64_max(OptsVisitorFixture *f, gconstpointer test_data) g_assert(f->userdef->u64->value == UINT64_MAX); } +static void +expect_both(OptsVisitorFixture *f, gconstpointer test_data) +{ + expect_ok(f, test_data); + g_assert(f->userdef->sub0->has_nint); + g_assert(f->userdef->sub0->nint == 13); + g_assert(f->userdef->sub1->has_nint); + g_assert(f->userdef->sub1->nint == 13); +} + +static void +expect_sub0(OptsVisitorFixture *f, gconstpointer test_data) +{ + expect_ok(f, test_data); + g_assert(f->userdef->sub0->has_nint); + g_assert(f->userdef->sub0->nint == 13); + g_assert(!f->userdef->sub1->has_nint); +} + +static void +expect_sub1(OptsVisitorFixture *f, gconstpointer test_data) +{ + expect_ok(f, test_data); + g_assert(!f->userdef->sub0->has_nint); + g_assert(f->userdef->sub1->has_nint); + g_assert(f->userdef->sub1->nint == 13); +} + /* test cases */ int @@ -270,6 +298,12 @@ main(int argc, char **argv) add_test("/visitor/opts/i64/range/2big/full", &expect_fail, "i64=-0x8000000000000000-0x7fffffffffffffff"); + /* Test nested structs support */ + add_test("/visitor/opts/nested/unqualified", &expect_both, "nint=13"); + add_test("/visitor/opts/nested/both", &expect_both, + "sub0.nint=13,sub1.nint=13"); + add_test("/visitor/opts/nested/sub0", &expect_sub0, "sub0.nint=13"); + add_test("/visitor/opts/nested/sub1", &expect_sub1, "sub1.nint=13"); g_test_run(); return 0; } -- 2.4.3 ^ permalink raw reply related [flat|nested] 42+ messages in thread
* Re: [Qemu-devel] [PATCH v2 2/6] qapi: support nested structs in OptsVisitor 2015-06-16 12:49 ` [Qemu-devel] [PATCH v2 2/6] qapi: support nested structs in OptsVisitor Kővágó, Zoltán @ 2015-06-17 7:50 ` Markus Armbruster 2015-06-17 8:41 ` Gerd Hoffmann 0 siblings, 1 reply; 42+ messages in thread From: Markus Armbruster @ 2015-06-17 7:50 UTC (permalink / raw) To: Kővágó, Zoltán Cc: László Ersek, Gerd Hoffmann, qemu-devel, Michael Roth Copying László because his fingerprints are on OptsVisitor. "Kővágó, Zoltán" <dirty.ice.hu@gmail.com> writes: > The current OptsVisitor flattens the whole structure, if there are same named > fields under different paths (like `in' and `out' in `Audiodev'), the current > visitor can't cope with them (for example setting `frequency=44100' will set the > in's frequency to 44100 and leave out's frequency unspecified). > > This patch fixes it, by the following changes: > 1) Specifying just the field name will apply to all fields that has the > specified name (this means it would set both in's and out's frequency to > 44100 in the above example). > 2) Optionally user can specify the path in the hierarchy. Names are separated by > a dot (e.g. `in.frequency', `foo.bar.something', etc). The user need not > specify the whole path, only the last few components (i.e. `bar.something' is > equivalent to `foo.bar.something' if only `foo' has a `bar' field). This way > 1) is just a special case of this when only the last component is specified. > 3) In case of an ambiguity (e.g `frequency=44100,in.frequency=8000') the longest > matching (the most specific) path wins (so in this example, in's frequency > would become 8000, because `in.frequency' is more specific that `frequency', > and out's frequency would become 44100, because only `frequency' matches it). Can you explain why the complexity is needed, i.e. why we can't just require full paths always? ^ permalink raw reply [flat|nested] 42+ messages in thread
* Re: [Qemu-devel] [PATCH v2 2/6] qapi: support nested structs in OptsVisitor 2015-06-17 7:50 ` Markus Armbruster @ 2015-06-17 8:41 ` Gerd Hoffmann 2015-06-17 11:01 ` Kővágó Zoltán 2015-06-17 11:18 ` Markus Armbruster 0 siblings, 2 replies; 42+ messages in thread From: Gerd Hoffmann @ 2015-06-17 8:41 UTC (permalink / raw) To: Markus Armbruster Cc: Michael Roth, László Ersek, qemu-devel, Kővágó, Zoltán On Mi, 2015-06-17 at 09:50 +0200, Markus Armbruster wrote: > Copying László because his fingerprints are on OptsVisitor. > > "Kővágó, Zoltán" <dirty.ice.hu@gmail.com> writes: > > > The current OptsVisitor flattens the whole structure, if there are same named > > fields under different paths (like `in' and `out' in `Audiodev'), the current > > visitor can't cope with them (for example setting `frequency=44100' will set the > > in's frequency to 44100 and leave out's frequency unspecified). > > > > This patch fixes it, by the following changes: > > 1) Specifying just the field name will apply to all fields that has the > > specified name (this means it would set both in's and out's frequency to > > 44100 in the above example). > > 2) Optionally user can specify the path in the hierarchy. Names are separated by > > a dot (e.g. `in.frequency', `foo.bar.something', etc). The user need not > > specify the whole path, only the last few components (i.e. `bar.something' is > > equivalent to `foo.bar.something' if only `foo' has a `bar' field). This way > > 1) is just a special case of this when only the last component is specified. > > 3) In case of an ambiguity (e.g `frequency=44100,in.frequency=8000') the longest > > matching (the most specific) path wins (so in this example, in's frequency > > would become 8000, because `in.frequency' is more specific that `frequency', > > and out's frequency would become 44100, because only `frequency' matches it). > > Can you explain why the complexity is needed, i.e. why we can't just > require full paths always? Keeping the short names is required for -netdev backward compatibility. Restricting to short or full (i.e. something= or foo.bar.something=, but disallow bar.something=) should not be a problem. I'm not sure this simplifies things much though. We have to build the full path anyway, and I think bar.something= is just a convenient thing we get almost for free ... cheers, Gerd ^ permalink raw reply [flat|nested] 42+ messages in thread
* Re: [Qemu-devel] [PATCH v2 2/6] qapi: support nested structs in OptsVisitor 2015-06-17 8:41 ` Gerd Hoffmann @ 2015-06-17 11:01 ` Kővágó Zoltán 2015-06-17 11:50 ` Markus Armbruster 2015-06-17 15:47 ` Eric Blake 2015-06-17 11:18 ` Markus Armbruster 1 sibling, 2 replies; 42+ messages in thread From: Kővágó Zoltán @ 2015-06-17 11:01 UTC (permalink / raw) To: Gerd Hoffmann, Markus Armbruster Cc: László Ersek, qemu-devel, Michael Roth 2015-06-17 10:41 keltezéssel, Gerd Hoffmann írta: > On Mi, 2015-06-17 at 09:50 +0200, Markus Armbruster wrote: >> Copying László because his fingerprints are on OptsVisitor. >> >> "Kővágó, Zoltán" <dirty.ice.hu@gmail.com> writes: >> >>> The current OptsVisitor flattens the whole structure, if there are same named >>> fields under different paths (like `in' and `out' in `Audiodev'), the current >>> visitor can't cope with them (for example setting `frequency=44100' will set the >>> in's frequency to 44100 and leave out's frequency unspecified). >>> >>> This patch fixes it, by the following changes: >>> 1) Specifying just the field name will apply to all fields that has the >>> specified name (this means it would set both in's and out's frequency to >>> 44100 in the above example). >>> 2) Optionally user can specify the path in the hierarchy. Names are separated by >>> a dot (e.g. `in.frequency', `foo.bar.something', etc). The user need not >>> specify the whole path, only the last few components (i.e. `bar.something' is >>> equivalent to `foo.bar.something' if only `foo' has a `bar' field). This way >>> 1) is just a special case of this when only the last component is specified. >>> 3) In case of an ambiguity (e.g `frequency=44100,in.frequency=8000') the longest >>> matching (the most specific) path wins (so in this example, in's frequency >>> would become 8000, because `in.frequency' is more specific that `frequency', >>> and out's frequency would become 44100, because only `frequency' matches it). >> >> Can you explain why the complexity is needed, i.e. why we can't just >> require full paths always? > > Keeping the short names is required for -netdev backward compatibility. > > Restricting to short or full (i.e. something= or foo.bar.something=, but > disallow bar.something=) should not be a problem. I'm not sure this > simplifies things much though. We have to build the full path anyway, > and I think bar.something= is just a convenient thing we get almost for > free ... With the current implementation you can specify (see my previous patch) in.try-poll=off in case of alsa. If we would need full paths, it would look like opts.data.in.try-poll=off, which is probably not something we want. ^ permalink raw reply [flat|nested] 42+ messages in thread
* Re: [Qemu-devel] [PATCH v2 2/6] qapi: support nested structs in OptsVisitor 2015-06-17 11:01 ` Kővágó Zoltán @ 2015-06-17 11:50 ` Markus Armbruster 2015-06-17 15:47 ` Eric Blake 1 sibling, 0 replies; 42+ messages in thread From: Markus Armbruster @ 2015-06-17 11:50 UTC (permalink / raw) To: Kővágó Zoltán Cc: Michael Roth, László Ersek, Gerd Hoffmann, qemu-devel "Kővágó Zoltán" <dirty.ice.hu@gmail.com> writes: > 2015-06-17 10:41 keltezéssel, Gerd Hoffmann írta: >> On Mi, 2015-06-17 at 09:50 +0200, Markus Armbruster wrote: >>> Copying László because his fingerprints are on OptsVisitor. >>> >>> "Kővágó, Zoltán" <dirty.ice.hu@gmail.com> writes: >>> >>>> The current OptsVisitor flattens the whole structure, if there are >>>> same named >>>> fields under different paths (like `in' and `out' in `Audiodev'), >>>> the current >>>> visitor can't cope with them (for example setting >>>> frequency=44100' will set the >>>> in's frequency to 44100 and leave out's frequency unspecified). >>>> >>>> This patch fixes it, by the following changes: >>>> 1) Specifying just the field name will apply to all fields that has the >>>> specified name (this means it would set both in's and out's frequency to >>>> 44100 in the above example). >>>> 2) Optionally user can specify the path in the hierarchy. Names >>>> are separated by >>>> a dot (e.g. `in.frequency', `foo.bar.something', etc). The user need not >>>> specify the whole path, only the last few components >>>> (i.e. `bar.something' is >>>> equivalent to `foo.bar.something' if only `foo' has a `bar' >>>> field). This way >>>> 1) is just a special case of this when only the last component >>>> is specified. >>>> 3) In case of an ambiguity (e.g >>>> frequency=44100,in.frequency=8000') the longest >>>> matching (the most specific) path wins (so in this example, >>>> in's frequency >>>> would become 8000, because `in.frequency' is more specific >>>> that `frequency', >>>> and out's frequency would become 44100, because only >>>> frequency' matches it). >>> >>> Can you explain why the complexity is needed, i.e. why we can't just >>> require full paths always? >> >> Keeping the short names is required for -netdev backward compatibility. >> >> Restricting to short or full (i.e. something= or foo.bar.something=, but >> disallow bar.something=) should not be a problem. I'm not sure this >> simplifies things much though. We have to build the full path anyway, >> and I think bar.something= is just a convenient thing we get almost for >> free ... > > With the current implementation you can specify (see my previous > patch) in.try-poll=off in case of alsa. If we would need full paths, > it would look like opts.data.in.try-poll=off, which is probably not > something we want. I agree opts.data.in.try-poll would be an ugly user interface. But instead of hiding opts.data. with a convenience feature, I'd like us to investigate avoiding it altogether. ^ permalink raw reply [flat|nested] 42+ messages in thread
* Re: [Qemu-devel] [PATCH v2 2/6] qapi: support nested structs in OptsVisitor 2015-06-17 11:01 ` Kővágó Zoltán 2015-06-17 11:50 ` Markus Armbruster @ 2015-06-17 15:47 ` Eric Blake 1 sibling, 0 replies; 42+ messages in thread From: Eric Blake @ 2015-06-17 15:47 UTC (permalink / raw) To: Kővágó Zoltán, Gerd Hoffmann, Markus Armbruster Cc: László Ersek, qemu-devel, Michael Roth [-- Attachment #1: Type: text/plain, Size: 1068 bytes --] On 06/17/2015 05:01 AM, Kővágó Zoltán wrote: >>> Can you explain why the complexity is needed, i.e. why we can't just >>> require full paths always? >> >> Keeping the short names is required for -netdev backward compatibility. >> >> Restricting to short or full (i.e. something= or foo.bar.something=, but >> disallow bar.something=) should not be a problem. I'm not sure this >> simplifies things much though. We have to build the full path anyway, >> and I think bar.something= is just a convenient thing we get almost for >> free ... > > With the current implementation you can specify (see my previous patch) > in.try-poll=off in case of alsa. If we would need full paths, it would > look like opts.data.in.try-poll=off, which is probably not something we > want. Elsewhere in the thread, it was suggested that you use a flat union. That would simplify the full path to opts.in.try-poll, rather than opts.data.in.try-poll. -- Eric Blake eblake redhat com +1-919-301-3266 Libvirt virtualization library http://libvirt.org [-- Attachment #2: OpenPGP digital signature --] [-- Type: application/pgp-signature, Size: 604 bytes --] ^ permalink raw reply [flat|nested] 42+ messages in thread
* Re: [Qemu-devel] [PATCH v2 2/6] qapi: support nested structs in OptsVisitor 2015-06-17 8:41 ` Gerd Hoffmann 2015-06-17 11:01 ` Kővágó Zoltán @ 2015-06-17 11:18 ` Markus Armbruster 2015-06-17 12:11 ` Kővágó Zoltán 1 sibling, 1 reply; 42+ messages in thread From: Markus Armbruster @ 2015-06-17 11:18 UTC (permalink / raw) To: Gerd Hoffmann Cc: Kevin Wolf, dirty.ice.hu, László Ersek, Michael Roth, qemu-devel Copying Kevin because similar issues exist in the block layer. Gerd Hoffmann <kraxel@redhat.com> writes: > On Mi, 2015-06-17 at 09:50 +0200, Markus Armbruster wrote: >> Copying László because his fingerprints are on OptsVisitor. >> >> "Kővágó, Zoltán" <dirty.ice.hu@gmail.com> writes: >> >> > The current OptsVisitor flattens the whole structure, if there are >> > same named >> > fields under different paths (like `in' and `out' in `Audiodev'), >> > the current >> > visitor can't cope with them (for example setting >> > frequency=44100' will set the >> > in's frequency to 44100 and leave out's frequency unspecified). >> > >> > This patch fixes it, by the following changes: >> > 1) Specifying just the field name will apply to all fields that has the >> > specified name (this means it would set both in's and out's frequency to >> > 44100 in the above example). What if they have different types? What if one of them can't take the value? >> > 2) Optionally user can specify the path in the hierarchy. Names >> > are separated by >> > a dot (e.g. `in.frequency', `foo.bar.something', etc). The user need not >> > specify the whole path, only the last few components >> > (i.e. `bar.something' is >> > equivalent to `foo.bar.something' if only `foo' has a `bar' >> > field). This way >> > 1) is just a special case of this when only the last component >> > is specified. >> > 3) In case of an ambiguity (e.g >> > frequency=44100,in.frequency=8000') the longest >> > matching (the most specific) path wins (so in this example, >> > in's frequency >> > would become 8000, because `in.frequency' is more specific that >> > frequency', >> > and out's frequency would become 44100, because only >> > frequency' matches it). The current rule for multiple assignments is "last one wins". E.g. in -drive if=none,file=tmp.img,file=tmp.qcow2 file=tmp.qcow2 wins. If I understand correctly, this patch amends the rule to "last most specific one wins". Correct? >> Can you explain why the complexity is needed, i.e. why we can't just >> require full paths always? > > Keeping the short names is required for -netdev backward compatibility. I suspect mostly because NetLegacy and Netdev aren't flat unions. Could be self-inflicted pain. What about flattening them instead? Assuming that's possible; I'd have to try. > Restricting to short or full (i.e. something= or foo.bar.something=, but > disallow bar.something=) should not be a problem. I'm not sure this > simplifies things much though. We have to build the full path anyway, > and I think bar.something= is just a convenient thing we get almost for > free ... We've been bitten by convenience features before. Adding them tends to be cheap, but maintaining compatibility can become a terrible headache. ^ permalink raw reply [flat|nested] 42+ messages in thread
* Re: [Qemu-devel] [PATCH v2 2/6] qapi: support nested structs in OptsVisitor 2015-06-17 11:18 ` Markus Armbruster @ 2015-06-17 12:11 ` Kővágó Zoltán 2015-06-17 13:41 ` Markus Armbruster 0 siblings, 1 reply; 42+ messages in thread From: Kővágó Zoltán @ 2015-06-17 12:11 UTC (permalink / raw) To: Markus Armbruster, Gerd Hoffmann Cc: Kevin Wolf, László Ersek, Michael Roth, qemu-devel 2015-06-17 13:18 keltezéssel, Markus Armbruster írta: > Copying Kevin because similar issues exist in the block layer. > > Gerd Hoffmann <kraxel@redhat.com> writes: > >> On Mi, 2015-06-17 at 09:50 +0200, Markus Armbruster wrote: >>> Copying László because his fingerprints are on OptsVisitor. >>> >>> "Kővágó, Zoltán" <dirty.ice.hu@gmail.com> writes: >>> >>>> The current OptsVisitor flattens the whole structure, if there are >>>> same named >>>> fields under different paths (like `in' and `out' in `Audiodev'), >>>> the current >>>> visitor can't cope with them (for example setting >>>> frequency=44100' will set the >>>> in's frequency to 44100 and leave out's frequency unspecified). >>>> >>>> This patch fixes it, by the following changes: >>>> 1) Specifying just the field name will apply to all fields that has the >>>> specified name (this means it would set both in's and out's frequency to >>>> 44100 in the above example). > > What if they have different types? > > What if one of them can't take the value? Currently it will error out, requiring the user to be more explicit. Probably not the best solution, but I can't really think of a better solution. (If we would ignore invalid values that would be very confusing imho.) >>>> 2) Optionally user can specify the path in the hierarchy. Names >>>> are separated by >>>> a dot (e.g. `in.frequency', `foo.bar.something', etc). The user need not >>>> specify the whole path, only the last few components >>>> (i.e. `bar.something' is >>>> equivalent to `foo.bar.something' if only `foo' has a `bar' >>>> field). This way >>>> 1) is just a special case of this when only the last component >>>> is specified. >>>> 3) In case of an ambiguity (e.g >>>> frequency=44100,in.frequency=8000') the longest >>>> matching (the most specific) path wins (so in this example, >>>> in's frequency >>>> would become 8000, because `in.frequency' is more specific that >>>> frequency', >>>> and out's frequency would become 44100, because only >>>> frequency' matches it). > > The current rule for multiple assignments is "last one wins". E.g. in > > -drive if=none,file=tmp.img,file=tmp.qcow2 > > file=tmp.qcow2 wins. > > If I understand correctly, this patch amends the rule to "last most > specific one wins". Correct? Yes. (But I didn't really checked that as I didn't know about the "last one win", and just thought it's an artifact of the current implementation.) >>> Can you explain why the complexity is needed, i.e. why we can't just >>> require full paths always? >> >> Keeping the short names is required for -netdev backward compatibility. > > I suspect mostly because NetLegacy and Netdev aren't flat unions. > Could be self-inflicted pain. > > What about flattening them instead? Assuming that's possible; I'd have > to try. > >> Restricting to short or full (i.e. something= or foo.bar.something=, but >> disallow bar.something=) should not be a problem. I'm not sure this >> simplifies things much though. We have to build the full path anyway, >> and I think bar.something= is just a convenient thing we get almost for >> free ... > > We've been bitten by convenience features before. Adding them tends to > be cheap, but maintaining compatibility can become a terrible headache. > ^ permalink raw reply [flat|nested] 42+ messages in thread
* Re: [Qemu-devel] [PATCH v2 2/6] qapi: support nested structs in OptsVisitor 2015-06-17 12:11 ` Kővágó Zoltán @ 2015-06-17 13:41 ` Markus Armbruster 2015-06-17 14:02 ` Kővágó Zoltán 0 siblings, 1 reply; 42+ messages in thread From: Markus Armbruster @ 2015-06-17 13:41 UTC (permalink / raw) To: Kővágó Zoltán Cc: Kevin Wolf, qemu-devel, László Ersek, Gerd Hoffmann, Michael Roth "Kővágó Zoltán" <dirty.ice.hu@gmail.com> writes: > 2015-06-17 13:18 keltezéssel, Markus Armbruster írta: >> Copying Kevin because similar issues exist in the block layer. >> >> Gerd Hoffmann <kraxel@redhat.com> writes: >> >>> On Mi, 2015-06-17 at 09:50 +0200, Markus Armbruster wrote: >>>> Copying László because his fingerprints are on OptsVisitor. >>>> >>>> "Kővágó, Zoltán" <dirty.ice.hu@gmail.com> writes: >>>> >>>>> The current OptsVisitor flattens the whole structure, if there are >>>>> same named >>>>> fields under different paths (like `in' and `out' in `Audiodev'), >>>>> the current >>>>> visitor can't cope with them (for example setting >>>>> frequency=44100' will set the >>>>> in's frequency to 44100 and leave out's frequency unspecified). >>>>> >>>>> This patch fixes it, by the following changes: >>>>> 1) Specifying just the field name will apply to all fields that has the >>>>> specified name (this means it would set both in's and out's >>>>> frequency to >>>>> 44100 in the above example). >> >> What if they have different types? >> >> What if one of them can't take the value? > > Currently it will error out, requiring the user to be more > explicit. Probably not the best solution, but I can't really think of > a better solution. (If we would ignore invalid values that would be > very confusing imho.) Yes, we clearly don't want foo=0 to set a.foo and b.foo, but foo=x set only a.foo, because the former can take any string, but the latter only a number. Can we require the LHS to be unambiguous? [...] ^ permalink raw reply [flat|nested] 42+ messages in thread
* Re: [Qemu-devel] [PATCH v2 2/6] qapi: support nested structs in OptsVisitor 2015-06-17 13:41 ` Markus Armbruster @ 2015-06-17 14:02 ` Kővágó Zoltán 2015-06-17 16:10 ` Markus Armbruster 0 siblings, 1 reply; 42+ messages in thread From: Kővágó Zoltán @ 2015-06-17 14:02 UTC (permalink / raw) To: Markus Armbruster Cc: Kevin Wolf, qemu-devel, László Ersek, Gerd Hoffmann, Michael Roth 2015-06-17 15:41 keltezéssel, Markus Armbruster írta: > "Kővágó Zoltán" <dirty.ice.hu@gmail.com> writes: > >> 2015-06-17 13:18 keltezéssel, Markus Armbruster írta: >>> Copying Kevin because similar issues exist in the block layer. >>> >>> Gerd Hoffmann <kraxel@redhat.com> writes: >>> >>>> On Mi, 2015-06-17 at 09:50 +0200, Markus Armbruster wrote: >>>>> Copying László because his fingerprints are on OptsVisitor. >>>>> >>>>> "Kővágó, Zoltán" <dirty.ice.hu@gmail.com> writes: >>>>> >>>>>> The current OptsVisitor flattens the whole structure, if there are >>>>>> same named >>>>>> fields under different paths (like `in' and `out' in `Audiodev'), >>>>>> the current >>>>>> visitor can't cope with them (for example setting >>>>>> frequency=44100' will set the >>>>>> in's frequency to 44100 and leave out's frequency unspecified). >>>>>> >>>>>> This patch fixes it, by the following changes: >>>>>> 1) Specifying just the field name will apply to all fields that has the >>>>>> specified name (this means it would set both in's and out's >>>>>> frequency to >>>>>> 44100 in the above example). >>> >>> What if they have different types? >>> >>> What if one of them can't take the value? >> >> Currently it will error out, requiring the user to be more >> explicit. Probably not the best solution, but I can't really think of >> a better solution. (If we would ignore invalid values that would be >> very confusing imho.) > > Yes, we clearly don't want foo=0 to set a.foo and b.foo, but foo=x set > only a.foo, because the former can take any string, but the latter only > a number. > > Can we require the LHS to be unambiguous? Originally I designed it that way because it allows you to specify frequency=44100 and set both in.frequency and out.frequency. But this could also be the convenience feature that we not really need. I don't see any other downside of making it unambigous. ^ permalink raw reply [flat|nested] 42+ messages in thread
* Re: [Qemu-devel] [PATCH v2 2/6] qapi: support nested structs in OptsVisitor 2015-06-17 14:02 ` Kővágó Zoltán @ 2015-06-17 16:10 ` Markus Armbruster 0 siblings, 0 replies; 42+ messages in thread From: Markus Armbruster @ 2015-06-17 16:10 UTC (permalink / raw) To: Kővágó Zoltán Cc: Kevin Wolf, Michael Roth, László Ersek, qemu-devel, Gerd Hoffmann "Kővágó Zoltán" <dirty.ice.hu@gmail.com> writes: > 2015-06-17 15:41 keltezéssel, Markus Armbruster írta: >> "Kővágó Zoltán" <dirty.ice.hu@gmail.com> writes: >> >>> 2015-06-17 13:18 keltezéssel, Markus Armbruster írta: >>>> Copying Kevin because similar issues exist in the block layer. >>>> >>>> Gerd Hoffmann <kraxel@redhat.com> writes: >>>> >>>>> On Mi, 2015-06-17 at 09:50 +0200, Markus Armbruster wrote: >>>>>> Copying László because his fingerprints are on OptsVisitor. >>>>>> >>>>>> "Kővágó, Zoltán" <dirty.ice.hu@gmail.com> writes: >>>>>> >>>>>>> The current OptsVisitor flattens the whole structure, if there are >>>>>>> same named >>>>>>> fields under different paths (like `in' and `out' in `Audiodev'), >>>>>>> the current >>>>>>> visitor can't cope with them (for example setting >>>>>>> frequency=44100' will set the >>>>>>> in's frequency to 44100 and leave out's frequency unspecified). >>>>>>> >>>>>>> This patch fixes it, by the following changes: >>>>>>> 1) Specifying just the field name will apply to all fields that has the >>>>>>> specified name (this means it would set both in's and out's >>>>>>> frequency to >>>>>>> 44100 in the above example). >>>> >>>> What if they have different types? >>>> >>>> What if one of them can't take the value? >>> >>> Currently it will error out, requiring the user to be more >>> explicit. Probably not the best solution, but I can't really think of >>> a better solution. (If we would ignore invalid values that would be >>> very confusing imho.) >> >> Yes, we clearly don't want foo=0 to set a.foo and b.foo, but foo=x set >> only a.foo, because the former can take any string, but the latter only >> a number. >> >> Can we require the LHS to be unambiguous? > > Originally I designed it that way because it allows you to specify > frequency=44100 and set both in.frequency and out.frequency. Is it common to set both to the same value? We could also default one to the other. Not sure that's actually a good idea. > But this > could also be the convenience feature that we not really need. I don't > see any other downside of making it unambigous. Good to know. We've struggled with unforeseen aftereffects of cute convenience features too much, and by now I'm quite reluctant to support them without a really compelling reason. ^ permalink raw reply [flat|nested] 42+ messages in thread
* [Qemu-devel] [PATCH v2 3/6] opts: do not print separator before first item in qemu_opts_print 2015-06-16 12:49 [Qemu-devel] [PATCH v2 0/6] -audiodev option Kővágó, Zoltán 2015-06-16 12:49 ` [Qemu-devel] [PATCH v2 1/6] qapi: qapi for audio backends Kővágó, Zoltán 2015-06-16 12:49 ` [Qemu-devel] [PATCH v2 2/6] qapi: support nested structs in OptsVisitor Kővágó, Zoltán @ 2015-06-16 12:49 ` Kővágó, Zoltán 2015-06-17 7:53 ` Markus Armbruster 2015-06-17 9:02 ` Kevin Wolf 2015-06-16 12:49 ` [Qemu-devel] [PATCH v2 4/6] qapi: AllocVisitor Kővágó, Zoltán ` (2 subsequent siblings) 5 siblings, 2 replies; 42+ messages in thread From: Kővágó, Zoltán @ 2015-06-16 12:49 UTC (permalink / raw) To: qemu-devel; +Cc: Kevin Wolf, Gerd Hoffmann, open list:Block layer core This allows to print options in a format that the user would actually write it on the command line (foo=bar,baz=asd,etc=def), without prepending a spurious comma at the beginning of the list. Only block.c depended on the old behavior, but it was also updated. Signed-off-by: Kővágó, Zoltán <DirtY.iCE.hu@gmail.com> --- block.c | 2 +- util/qemu-option.c | 5 ++++- 2 files changed, 5 insertions(+), 2 deletions(-) diff --git a/block.c b/block.c index dd4f58d..c5d456c 100644 --- a/block.c +++ b/block.c @@ -3823,7 +3823,7 @@ void bdrv_img_create(const char *filename, const char *fmt, } if (!quiet) { - printf("Formatting '%s', fmt=%s", filename, fmt); + printf("Formatting '%s', fmt=%s ", filename, fmt); qemu_opts_print(opts, " "); puts(""); } diff --git a/util/qemu-option.c b/util/qemu-option.c index 840f5f7..b347d92 100644 --- a/util/qemu-option.c +++ b/util/qemu-option.c @@ -728,14 +728,16 @@ void qemu_opts_del(QemuOpts *opts) g_free(opts); } -void qemu_opts_print(QemuOpts *opts, const char *sep) +void qemu_opts_print(QemuOpts *opts, const char *d_sep) { QemuOpt *opt; QemuOptDesc *desc = opts->list->desc; + const char *sep = ""; if (desc[0].name == NULL) { QTAILQ_FOREACH(opt, &opts->head, next) { printf("%s%s=\"%s\"", sep, opt->name, opt->str); + sep = d_sep; } return; } @@ -755,6 +757,7 @@ void qemu_opts_print(QemuOpts *opts, const char *sep) } else { printf("%s%s=%s", sep, desc->name, value); } + sep = d_sep; } } -- 2.4.3 ^ permalink raw reply related [flat|nested] 42+ messages in thread
* Re: [Qemu-devel] [PATCH v2 3/6] opts: do not print separator before first item in qemu_opts_print 2015-06-16 12:49 ` [Qemu-devel] [PATCH v2 3/6] opts: do not print separator before first item in qemu_opts_print Kővágó, Zoltán @ 2015-06-17 7:53 ` Markus Armbruster 2015-06-17 9:02 ` Kevin Wolf 1 sibling, 0 replies; 42+ messages in thread From: Markus Armbruster @ 2015-06-17 7:53 UTC (permalink / raw) To: Kővágó, Zoltán Cc: Kevin Wolf, qemu-devel, open list:Block layer core, Gerd Hoffmann "Kővágó, Zoltán" <dirty.ice.hu@gmail.com> writes: > This allows to print options in a format that the user would actually write it > on the command line (foo=bar,baz=asd,etc=def), without prepending a spurious > comma at the beginning of the list. Suggest: This will let us print... > Only block.c depended on the old behavior, but it was also updated. I'd drop this sentence. > > Signed-off-by: Kővágó, Zoltán <DirtY.iCE.hu@gmail.com> > --- > block.c | 2 +- > util/qemu-option.c | 5 ++++- > 2 files changed, 5 insertions(+), 2 deletions(-) > > diff --git a/block.c b/block.c > index dd4f58d..c5d456c 100644 > --- a/block.c > +++ b/block.c > @@ -3823,7 +3823,7 @@ void bdrv_img_create(const char *filename, const char *fmt, > } > > if (!quiet) { > - printf("Formatting '%s', fmt=%s", filename, fmt); > + printf("Formatting '%s', fmt=%s ", filename, fmt); > qemu_opts_print(opts, " "); > puts(""); > } > diff --git a/util/qemu-option.c b/util/qemu-option.c > index 840f5f7..b347d92 100644 > --- a/util/qemu-option.c > +++ b/util/qemu-option.c > @@ -728,14 +728,16 @@ void qemu_opts_del(QemuOpts *opts) > g_free(opts); > } > > -void qemu_opts_print(QemuOpts *opts, const char *sep) > +void qemu_opts_print(QemuOpts *opts, const char *d_sep) > { > QemuOpt *opt; > QemuOptDesc *desc = opts->list->desc; > + const char *sep = ""; > > if (desc[0].name == NULL) { > QTAILQ_FOREACH(opt, &opts->head, next) { > printf("%s%s=\"%s\"", sep, opt->name, opt->str); > + sep = d_sep; > } > return; > } > @@ -755,6 +757,7 @@ void qemu_opts_print(QemuOpts *opts, const char *sep) > } else { > printf("%s%s=%s", sep, desc->name, value); > } > + sep = d_sep; > } > } No idea what d_ means. Let's rename d_sep to separator. With that rename: Reviewed-by: Markus Armbruster <armbru@redhat.com> ^ permalink raw reply [flat|nested] 42+ messages in thread
* Re: [Qemu-devel] [PATCH v2 3/6] opts: do not print separator before first item in qemu_opts_print 2015-06-16 12:49 ` [Qemu-devel] [PATCH v2 3/6] opts: do not print separator before first item in qemu_opts_print Kővágó, Zoltán 2015-06-17 7:53 ` Markus Armbruster @ 2015-06-17 9:02 ` Kevin Wolf 1 sibling, 0 replies; 42+ messages in thread From: Kevin Wolf @ 2015-06-17 9:02 UTC (permalink / raw) To: Kővágó, Zoltán Cc: qemu-devel, open list:Block layer core, Gerd Hoffmann Am 16.06.2015 um 14:49 hat Kővágó, Zoltán geschrieben: > This allows to print options in a format that the user would actually write it > on the command line (foo=bar,baz=asd,etc=def), without prepending a spurious > comma at the beginning of the list. Note that if you really want to produce a valid command line, you would have to escape any commas in option values. The quotes arounds strings might not be wanted any more then either. > Only block.c depended on the old behavior, but it was also updated. > > Signed-off-by: Kővágó, Zoltán <DirtY.iCE.hu@gmail.com> Seems that no behaviour changes from a block layer perspective, so: Acked-by: Kevin Wolf <kwolf@redhat.com> ^ permalink raw reply [flat|nested] 42+ messages in thread
* [Qemu-devel] [PATCH v2 4/6] qapi: AllocVisitor 2015-06-16 12:49 [Qemu-devel] [PATCH v2 0/6] -audiodev option Kővágó, Zoltán ` (2 preceding siblings ...) 2015-06-16 12:49 ` [Qemu-devel] [PATCH v2 3/6] opts: do not print separator before first item in qemu_opts_print Kővágó, Zoltán @ 2015-06-16 12:49 ` Kővágó, Zoltán 2015-06-17 7:56 ` Markus Armbruster 2015-06-16 12:49 ` [Qemu-devel] [PATCH v2 5/6] audio: use qapi AudioFormat instead of audfmt_e Kővágó, Zoltán 2015-06-16 12:49 ` [Qemu-devel] [PATCH v2 6/6] audio: -audiodev command line option Kővágó, Zoltán 5 siblings, 1 reply; 42+ messages in thread From: Kővágó, Zoltán @ 2015-06-16 12:49 UTC (permalink / raw) To: qemu-devel; +Cc: Michael Roth, Gerd Hoffmann, Markus Armbruster Simple visitor that recursively allocates structures with only optional variables. Unions are initialized to the first type specified. Other non optional types are not supported. Signed-off-by: Kővágó, Zoltán <DirtY.iCE.hu@gmail.com> --- include/qapi/alloc-visitor.h | 18 +++++++++++++ qapi/Makefile.objs | 1 + qapi/alloc-visitor.c | 62 ++++++++++++++++++++++++++++++++++++++++++++ 3 files changed, 81 insertions(+) create mode 100644 include/qapi/alloc-visitor.h create mode 100644 qapi/alloc-visitor.c diff --git a/include/qapi/alloc-visitor.h b/include/qapi/alloc-visitor.h new file mode 100644 index 0000000..3d54295 --- /dev/null +++ b/include/qapi/alloc-visitor.h @@ -0,0 +1,18 @@ +/* + * Alloc Visitor. + * Recursively allocates structs, leaving all optional fields unset. In case of + * a non-optional field it fails. + */ + +#ifndef ALLOC_VISITOR_H +#define ALLOC_VISITOR_H + +#include "qapi/visitor.h" + +typedef struct AllocVisitor AllocVisitor; + +AllocVisitor *alloc_visitor_new(void); +void alloc_visitor_cleanup(AllocVisitor *v); +Visitor *alloc_visitor_get_visitor(AllocVisitor *v); + +#endif diff --git a/qapi/Makefile.objs b/qapi/Makefile.objs index 2278970..7bc26a3 100644 --- a/qapi/Makefile.objs +++ b/qapi/Makefile.objs @@ -4,3 +4,4 @@ util-obj-y += string-input-visitor.o string-output-visitor.o util-obj-y += opts-visitor.o util-obj-y += qmp-event.o util-obj-y += qapi-util.o +util-obj-y += alloc-visitor.o diff --git a/qapi/alloc-visitor.c b/qapi/alloc-visitor.c new file mode 100644 index 0000000..dbb83af --- /dev/null +++ b/qapi/alloc-visitor.c @@ -0,0 +1,62 @@ +#include "qapi/alloc-visitor.h" +#include "qemu-common.h" +#include "qapi/visitor-impl.h" + +struct AllocVisitor { + Visitor visitor; +}; + +static void alloc_start_struct(Visitor *v, void **obj, const char* kind, + const char *name, size_t size, Error **errp) +{ + if (obj) { + *obj = g_malloc0(size); + } +} + +static void alloc_end_struct(Visitor *v, Error **errp) +{ +} + +static void alloc_start_implicit_struct(Visitor *v, void **obj, size_t size, + Error **errp) +{ + if (obj) { + *obj = g_malloc0(size); + } +} + +static void alloc_end_implicit_struct(Visitor *v, Error **errp) +{ +} + +static void alloc_type_enum(Visitor *v, int *obj, const char *strings[], + const char *kind, const char *name, Error **errp) +{ + assert(*strings); /* there is at least one valid enum value... */ + *obj = 0; +} + +AllocVisitor *alloc_visitor_new(void) +{ + AllocVisitor *v = g_malloc0(sizeof(AllocVisitor)); + + v->visitor.start_struct = alloc_start_struct; + v->visitor.end_struct = alloc_end_struct; + v->visitor.start_implicit_struct = alloc_start_implicit_struct; + v->visitor.end_implicit_struct = alloc_end_implicit_struct; + + v->visitor.type_enum = alloc_type_enum; + + return v; +} + +void alloc_visitor_cleanup(AllocVisitor *v) +{ + g_free(v); +} + +Visitor *alloc_visitor_get_visitor(AllocVisitor *v) +{ + return &v->visitor; +} -- 2.4.3 ^ permalink raw reply related [flat|nested] 42+ messages in thread
* Re: [Qemu-devel] [PATCH v2 4/6] qapi: AllocVisitor 2015-06-16 12:49 ` [Qemu-devel] [PATCH v2 4/6] qapi: AllocVisitor Kővágó, Zoltán @ 2015-06-17 7:56 ` Markus Armbruster 2015-06-17 12:01 ` Kővágó Zoltán 0 siblings, 1 reply; 42+ messages in thread From: Markus Armbruster @ 2015-06-17 7:56 UTC (permalink / raw) To: Kővágó, Zoltán Cc: Gerd Hoffmann, qemu-devel, Michael Roth "Kővágó, Zoltán" <dirty.ice.hu@gmail.com> writes: > Simple visitor that recursively allocates structures with only optional > variables. Unions are initialized to the first type specified. Other non > optional types are not supported. Sounds dubious :) Can you explain why it's useful? I guess later patches provide an example. ^ permalink raw reply [flat|nested] 42+ messages in thread
* Re: [Qemu-devel] [PATCH v2 4/6] qapi: AllocVisitor 2015-06-17 7:56 ` Markus Armbruster @ 2015-06-17 12:01 ` Kővágó Zoltán 2015-06-17 13:42 ` Markus Armbruster 0 siblings, 1 reply; 42+ messages in thread From: Kővágó Zoltán @ 2015-06-17 12:01 UTC (permalink / raw) To: Markus Armbruster; +Cc: Gerd Hoffmann, qemu-devel, Michael Roth 2015-06-17 09:56 keltezéssel, Markus Armbruster írta: > "Kővágó, Zoltán" <dirty.ice.hu@gmail.com> writes: > >> Simple visitor that recursively allocates structures with only optional >> variables. Unions are initialized to the first type specified. Other non >> optional types are not supported. > > Sounds dubious :) > > Can you explain why it's useful? I guess later patches provide an > example. Oh, crap, ignore this commit. Just realized that in the refactorings I did I removed all references to this visitor... so it's not needed. ^ permalink raw reply [flat|nested] 42+ messages in thread
* Re: [Qemu-devel] [PATCH v2 4/6] qapi: AllocVisitor 2015-06-17 12:01 ` Kővágó Zoltán @ 2015-06-17 13:42 ` Markus Armbruster 0 siblings, 0 replies; 42+ messages in thread From: Markus Armbruster @ 2015-06-17 13:42 UTC (permalink / raw) To: Kővágó Zoltán; +Cc: Michael Roth, Gerd Hoffmann, qemu-devel "Kővágó Zoltán" <dirty.ice.hu@gmail.com> writes: > 2015-06-17 09:56 keltezéssel, Markus Armbruster írta: >> "Kővágó, Zoltán" <dirty.ice.hu@gmail.com> writes: >> >>> Simple visitor that recursively allocates structures with only optional >>> variables. Unions are initialized to the first type specified. Other non >>> optional types are not supported. >> >> Sounds dubious :) >> >> Can you explain why it's useful? I guess later patches provide an >> example. > > Oh, crap, ignore this commit. Just realized that in the refactorings I > did I removed all references to this visitor... so it's not needed. Glad I asked before reviewing the code :) ^ permalink raw reply [flat|nested] 42+ messages in thread
* [Qemu-devel] [PATCH v2 5/6] audio: use qapi AudioFormat instead of audfmt_e 2015-06-16 12:49 [Qemu-devel] [PATCH v2 0/6] -audiodev option Kővágó, Zoltán ` (3 preceding siblings ...) 2015-06-16 12:49 ` [Qemu-devel] [PATCH v2 4/6] qapi: AllocVisitor Kővágó, Zoltán @ 2015-06-16 12:49 ` Kővágó, Zoltán 2015-06-17 8:01 ` Markus Armbruster 2015-06-16 12:49 ` [Qemu-devel] [PATCH v2 6/6] audio: -audiodev command line option Kővágó, Zoltán 5 siblings, 1 reply; 42+ messages in thread From: Kővágó, Zoltán @ 2015-06-16 12:49 UTC (permalink / raw) To: qemu-devel; +Cc: Peter Maydell, Michael Walle, Gerd Hoffmann I had to include an enum for audio sampling formats into qapi, but that meant duplicating the audfmt_e enum. This patch replaces audfmt_e and associated values with the qapi generated AudioFormat enum. This patch is mostly a search-and-replace, except for switches where the qapi generated AUDIO_FORMAT_MAX caused problems. Signed-off-by: Kővágó, Zoltán <DirtY.iCE.hu@gmail.com> --- audio/alsaaudio.c | 53 ++++++++++++++------------ audio/audio.c | 97 ++++++++++++++++++++++++++--------------------- audio/audio.h | 11 +----- audio/audio_win_int.c | 18 ++++----- audio/ossaudio.c | 30 +++++++-------- audio/paaudio.c | 28 +++++++------- audio/sdlaudio.c | 26 ++++++------- audio/spiceaudio.c | 4 +- audio/wavaudio.c | 17 +++++---- audio/wavcapture.c | 2 +- hw/arm/omap2.c | 2 +- hw/audio/ac97.c | 2 +- hw/audio/adlib.c | 2 +- hw/audio/cs4231a.c | 6 +-- hw/audio/es1370.c | 4 +- hw/audio/gus.c | 2 +- hw/audio/hda-codec.c | 18 ++++----- hw/audio/lm4549.c | 6 +-- hw/audio/milkymist-ac97.c | 2 +- hw/audio/pcspk.c | 2 +- hw/audio/sb16.c | 14 +++---- hw/audio/wm8750.c | 4 +- hw/input/tsc210x.c | 2 +- hw/usb/dev-audio.c | 2 +- ui/vnc.c | 14 +++---- 25 files changed, 187 insertions(+), 181 deletions(-) diff --git a/audio/alsaaudio.c b/audio/alsaaudio.c index 6315b2d..4d38f5d 100644 --- a/audio/alsaaudio.c +++ b/audio/alsaaudio.c @@ -88,7 +88,7 @@ struct alsa_params_req { struct alsa_params_obt { int freq; - audfmt_e fmt; + AudioFormat fmt; int endianness; int nchannels; snd_pcm_uframes_t samples; @@ -295,16 +295,16 @@ static int alsa_write (SWVoiceOut *sw, void *buf, int len) return audio_pcm_sw_write (sw, buf, len); } -static snd_pcm_format_t aud_to_alsafmt (audfmt_e fmt, int endianness) +static snd_pcm_format_t aud_to_alsafmt (AudioFormat fmt, int endianness) { switch (fmt) { - case AUD_FMT_S8: + case AUDIO_FORMAT_S8: return SND_PCM_FORMAT_S8; - case AUD_FMT_U8: + case AUDIO_FORMAT_U8: return SND_PCM_FORMAT_U8; - case AUD_FMT_S16: + case AUDIO_FORMAT_S16: if (endianness) { return SND_PCM_FORMAT_S16_BE; } @@ -312,7 +312,7 @@ static snd_pcm_format_t aud_to_alsafmt (audfmt_e fmt, int endianness) return SND_PCM_FORMAT_S16_LE; } - case AUD_FMT_U16: + case AUDIO_FORMAT_U16: if (endianness) { return SND_PCM_FORMAT_U16_BE; } @@ -320,7 +320,7 @@ static snd_pcm_format_t aud_to_alsafmt (audfmt_e fmt, int endianness) return SND_PCM_FORMAT_U16_LE; } - case AUD_FMT_S32: + case AUDIO_FORMAT_S32: if (endianness) { return SND_PCM_FORMAT_S32_BE; } @@ -328,7 +328,7 @@ static snd_pcm_format_t aud_to_alsafmt (audfmt_e fmt, int endianness) return SND_PCM_FORMAT_S32_LE; } - case AUD_FMT_U32: + case AUDIO_FORMAT_U32: if (endianness) { return SND_PCM_FORMAT_U32_BE; } @@ -345,58 +345,58 @@ static snd_pcm_format_t aud_to_alsafmt (audfmt_e fmt, int endianness) } } -static int alsa_to_audfmt (snd_pcm_format_t alsafmt, audfmt_e *fmt, +static int alsa_to_audfmt (snd_pcm_format_t alsafmt, AudioFormat *fmt, int *endianness) { switch (alsafmt) { case SND_PCM_FORMAT_S8: *endianness = 0; - *fmt = AUD_FMT_S8; + *fmt = AUDIO_FORMAT_S8; break; case SND_PCM_FORMAT_U8: *endianness = 0; - *fmt = AUD_FMT_U8; + *fmt = AUDIO_FORMAT_U8; break; case SND_PCM_FORMAT_S16_LE: *endianness = 0; - *fmt = AUD_FMT_S16; + *fmt = AUDIO_FORMAT_S16; break; case SND_PCM_FORMAT_U16_LE: *endianness = 0; - *fmt = AUD_FMT_U16; + *fmt = AUDIO_FORMAT_U16; break; case SND_PCM_FORMAT_S16_BE: *endianness = 1; - *fmt = AUD_FMT_S16; + *fmt = AUDIO_FORMAT_S16; break; case SND_PCM_FORMAT_U16_BE: *endianness = 1; - *fmt = AUD_FMT_U16; + *fmt = AUDIO_FORMAT_U16; break; case SND_PCM_FORMAT_S32_LE: *endianness = 0; - *fmt = AUD_FMT_S32; + *fmt = AUDIO_FORMAT_S32; break; case SND_PCM_FORMAT_U32_LE: *endianness = 0; - *fmt = AUD_FMT_U32; + *fmt = AUDIO_FORMAT_U32; break; case SND_PCM_FORMAT_S32_BE: *endianness = 1; - *fmt = AUD_FMT_S32; + *fmt = AUDIO_FORMAT_S32; break; case SND_PCM_FORMAT_U32_BE: *endianness = 1; - *fmt = AUD_FMT_U32; + *fmt = AUDIO_FORMAT_U32; break; default: @@ -639,19 +639,22 @@ static int alsa_open (int in, struct alsa_params_req *req, bytes_per_sec = freq << (nchannels == 2); switch (obt->fmt) { - case AUD_FMT_S8: - case AUD_FMT_U8: + case AUDIO_FORMAT_S8: + case AUDIO_FORMAT_U8: break; - case AUD_FMT_S16: - case AUD_FMT_U16: + case AUDIO_FORMAT_S16: + case AUDIO_FORMAT_U16: bytes_per_sec <<= 1; break; - case AUD_FMT_S32: - case AUD_FMT_U32: + case AUDIO_FORMAT_S32: + case AUDIO_FORMAT_U32: bytes_per_sec <<= 2; break; + + case AUDIO_FORMAT_MAX: + break; } threshold = (conf->threshold * bytes_per_sec) / 1000; diff --git a/audio/audio.c b/audio/audio.c index 5be4b15..112b57b 100644 --- a/audio/audio.c +++ b/audio/audio.c @@ -75,7 +75,7 @@ static struct { .settings = { .freq = 44100, .nchannels = 2, - .fmt = AUD_FMT_S16, + .fmt = AUDIO_FORMAT_S16, .endianness = AUDIO_HOST_ENDIANNESS, } }, @@ -87,7 +87,7 @@ static struct { .settings = { .freq = 44100, .nchannels = 2, - .fmt = AUD_FMT_S16, + .fmt = AUDIO_FORMAT_S16, .endianness = AUDIO_HOST_ENDIANNESS, } }, @@ -219,58 +219,61 @@ static char *audio_alloc_prefix (const char *s) return r; } -static const char *audio_audfmt_to_string (audfmt_e fmt) +static const char *audio_audfmt_to_string (AudioFormat fmt) { switch (fmt) { - case AUD_FMT_U8: + case AUDIO_FORMAT_U8: return "U8"; - case AUD_FMT_U16: + case AUDIO_FORMAT_U16: return "U16"; - case AUD_FMT_S8: + case AUDIO_FORMAT_S8: return "S8"; - case AUD_FMT_S16: + case AUDIO_FORMAT_S16: return "S16"; - case AUD_FMT_U32: + case AUDIO_FORMAT_U32: return "U32"; - case AUD_FMT_S32: + case AUDIO_FORMAT_S32: return "S32"; + + case AUDIO_FORMAT_MAX: + abort(); } dolog ("Bogus audfmt %d returning S16\n", fmt); return "S16"; } -static audfmt_e audio_string_to_audfmt (const char *s, audfmt_e defval, +static AudioFormat audio_string_to_audfmt (const char *s, AudioFormat defval, int *defaultp) { if (!strcasecmp (s, "u8")) { *defaultp = 0; - return AUD_FMT_U8; + return AUDIO_FORMAT_U8; } else if (!strcasecmp (s, "u16")) { *defaultp = 0; - return AUD_FMT_U16; + return AUDIO_FORMAT_U16; } else if (!strcasecmp (s, "u32")) { *defaultp = 0; - return AUD_FMT_U32; + return AUDIO_FORMAT_U32; } else if (!strcasecmp (s, "s8")) { *defaultp = 0; - return AUD_FMT_S8; + return AUDIO_FORMAT_S8; } else if (!strcasecmp (s, "s16")) { *defaultp = 0; - return AUD_FMT_S16; + return AUDIO_FORMAT_S16; } else if (!strcasecmp (s, "s32")) { *defaultp = 0; - return AUD_FMT_S32; + return AUDIO_FORMAT_S32; } else { dolog ("Bogus audio format `%s' using %s\n", @@ -280,8 +283,8 @@ static audfmt_e audio_string_to_audfmt (const char *s, audfmt_e defval, } } -static audfmt_e audio_get_conf_fmt (const char *envname, - audfmt_e defval, +static AudioFormat audio_get_conf_fmt (const char *envname, + AudioFormat defval, int *defaultp) { const char *var = getenv (envname); @@ -384,7 +387,7 @@ static void audio_print_options (const char *prefix, case AUD_OPT_FMT: { - audfmt_e *fmtp = opt->valp; + AudioFormat *fmtp = opt->valp; printf ( "format, %s = %s, (one of: U8 S8 U16 S16 U32 S32)\n", state, @@ -471,7 +474,7 @@ static void audio_process_options (const char *prefix, case AUD_OPT_FMT: { - audfmt_e *fmtp = opt->valp; + AudioFormat *fmtp = opt->valp; *fmtp = audio_get_conf_fmt (optname, *fmtp, &def); } break; @@ -502,22 +505,22 @@ static void audio_print_settings (struct audsettings *as) dolog ("frequency=%d nchannels=%d fmt=", as->freq, as->nchannels); switch (as->fmt) { - case AUD_FMT_S8: + case AUDIO_FORMAT_S8: AUD_log (NULL, "S8"); break; - case AUD_FMT_U8: + case AUDIO_FORMAT_U8: AUD_log (NULL, "U8"); break; - case AUD_FMT_S16: + case AUDIO_FORMAT_S16: AUD_log (NULL, "S16"); break; - case AUD_FMT_U16: + case AUDIO_FORMAT_U16: AUD_log (NULL, "U16"); break; - case AUD_FMT_S32: + case AUDIO_FORMAT_S32: AUD_log (NULL, "S32"); break; - case AUD_FMT_U32: + case AUDIO_FORMAT_U32: AUD_log (NULL, "U32"); break; default: @@ -548,12 +551,12 @@ static int audio_validate_settings (struct audsettings *as) invalid |= as->endianness != 0 && as->endianness != 1; switch (as->fmt) { - case AUD_FMT_S8: - case AUD_FMT_U8: - case AUD_FMT_S16: - case AUD_FMT_U16: - case AUD_FMT_S32: - case AUD_FMT_U32: + case AUDIO_FORMAT_S8: + case AUDIO_FORMAT_U8: + case AUDIO_FORMAT_S16: + case AUDIO_FORMAT_U16: + case AUDIO_FORMAT_S32: + case AUDIO_FORMAT_U32: break; default: invalid = 1; @@ -569,25 +572,28 @@ static int audio_pcm_info_eq (struct audio_pcm_info *info, struct audsettings *a int bits = 8, sign = 0; switch (as->fmt) { - case AUD_FMT_S8: + case AUDIO_FORMAT_S8: sign = 1; /* fall through */ - case AUD_FMT_U8: + case AUDIO_FORMAT_U8: break; - case AUD_FMT_S16: + case AUDIO_FORMAT_S16: sign = 1; /* fall through */ - case AUD_FMT_U16: + case AUDIO_FORMAT_U16: bits = 16; break; - case AUD_FMT_S32: + case AUDIO_FORMAT_S32: sign = 1; /* fall through */ - case AUD_FMT_U32: + case AUDIO_FORMAT_U32: bits = 32; break; + + case AUDIO_FORMAT_MAX: + abort(); } return info->freq == as->freq && info->nchannels == as->nchannels @@ -601,24 +607,27 @@ void audio_pcm_init_info (struct audio_pcm_info *info, struct audsettings *as) int bits = 8, sign = 0, shift = 0; switch (as->fmt) { - case AUD_FMT_S8: + case AUDIO_FORMAT_S8: sign = 1; - case AUD_FMT_U8: + case AUDIO_FORMAT_U8: break; - case AUD_FMT_S16: + case AUDIO_FORMAT_S16: sign = 1; - case AUD_FMT_U16: + case AUDIO_FORMAT_U16: bits = 16; shift = 1; break; - case AUD_FMT_S32: + case AUDIO_FORMAT_S32: sign = 1; - case AUD_FMT_U32: + case AUDIO_FORMAT_U32: bits = 32; shift = 2; break; + + case AUDIO_FORMAT_MAX: + abort(); } info->freq = as->freq; diff --git a/audio/audio.h b/audio/audio.h index e7ea397..e300511 100644 --- a/audio/audio.h +++ b/audio/audio.h @@ -29,15 +29,6 @@ typedef void (*audio_callback_fn) (void *opaque, int avail); -typedef enum { - AUD_FMT_U8, - AUD_FMT_S8, - AUD_FMT_U16, - AUD_FMT_S16, - AUD_FMT_U32, - AUD_FMT_S32 -} audfmt_e; - #ifdef HOST_WORDS_BIGENDIAN #define AUDIO_HOST_ENDIANNESS 1 #else @@ -47,7 +38,7 @@ typedef enum { struct audsettings { int freq; int nchannels; - audfmt_e fmt; + AudioFormat fmt; int endianness; }; diff --git a/audio/audio_win_int.c b/audio/audio_win_int.c index e132405..a8cfa77 100644 --- a/audio/audio_win_int.c +++ b/audio/audio_win_int.c @@ -23,20 +23,20 @@ int waveformat_from_audio_settings (WAVEFORMATEX *wfx, wfx->cbSize = 0; switch (as->fmt) { - case AUD_FMT_S8: - case AUD_FMT_U8: + case AUDIO_FORMAT_S8: + case AUDIO_FORMAT_U8: wfx->wBitsPerSample = 8; break; - case AUD_FMT_S16: - case AUD_FMT_U16: + case AUDIO_FORMAT_S16: + case AUDIO_FORMAT_U16: wfx->wBitsPerSample = 16; wfx->nAvgBytesPerSec <<= 1; wfx->nBlockAlign <<= 1; break; - case AUD_FMT_S32: - case AUD_FMT_U32: + case AUDIO_FORMAT_S32: + case AUDIO_FORMAT_U32: wfx->wBitsPerSample = 32; wfx->nAvgBytesPerSec <<= 2; wfx->nBlockAlign <<= 2; @@ -84,15 +84,15 @@ int waveformat_to_audio_settings (WAVEFORMATEX *wfx, switch (wfx->wBitsPerSample) { case 8: - as->fmt = AUD_FMT_U8; + as->fmt = AUDIO_FORMAT_U8; break; case 16: - as->fmt = AUD_FMT_S16; + as->fmt = AUDIO_FORMAT_S16; break; case 32: - as->fmt = AUD_FMT_S32; + as->fmt = AUDIO_FORMAT_S32; break; default: diff --git a/audio/ossaudio.c b/audio/ossaudio.c index 11e76a1..02a3a95 100644 --- a/audio/ossaudio.c +++ b/audio/ossaudio.c @@ -72,7 +72,7 @@ typedef struct OSSVoiceIn { struct oss_params { int freq; - audfmt_e fmt; + AudioFormat fmt; int nchannels; int nfrags; int fragsize; @@ -150,16 +150,16 @@ static int oss_write (SWVoiceOut *sw, void *buf, int len) return audio_pcm_sw_write (sw, buf, len); } -static int aud_to_ossfmt (audfmt_e fmt, int endianness) +static int aud_to_ossfmt (AudioFormat fmt, int endianness) { switch (fmt) { - case AUD_FMT_S8: + case AUDIO_FORMAT_S8: return AFMT_S8; - case AUD_FMT_U8: + case AUDIO_FORMAT_U8: return AFMT_U8; - case AUD_FMT_S16: + case AUDIO_FORMAT_S16: if (endianness) { return AFMT_S16_BE; } @@ -167,7 +167,7 @@ static int aud_to_ossfmt (audfmt_e fmt, int endianness) return AFMT_S16_LE; } - case AUD_FMT_U16: + case AUDIO_FORMAT_U16: if (endianness) { return AFMT_U16_BE; } @@ -184,37 +184,37 @@ static int aud_to_ossfmt (audfmt_e fmt, int endianness) } } -static int oss_to_audfmt (int ossfmt, audfmt_e *fmt, int *endianness) +static int oss_to_audfmt (int ossfmt, AudioFormat *fmt, int *endianness) { switch (ossfmt) { case AFMT_S8: *endianness = 0; - *fmt = AUD_FMT_S8; + *fmt = AUDIO_FORMAT_S8; break; case AFMT_U8: *endianness = 0; - *fmt = AUD_FMT_U8; + *fmt = AUDIO_FORMAT_U8; break; case AFMT_S16_LE: *endianness = 0; - *fmt = AUD_FMT_S16; + *fmt = AUDIO_FORMAT_S16; break; case AFMT_U16_LE: *endianness = 0; - *fmt = AUD_FMT_U16; + *fmt = AUDIO_FORMAT_U16; break; case AFMT_S16_BE: *endianness = 1; - *fmt = AUD_FMT_S16; + *fmt = AUDIO_FORMAT_S16; break; case AFMT_U16_BE: *endianness = 1; - *fmt = AUD_FMT_U16; + *fmt = AUDIO_FORMAT_U16; break; default: @@ -502,7 +502,7 @@ static int oss_init_out(HWVoiceOut *hw, struct audsettings *as, int endianness; int err; int fd; - audfmt_e effective_fmt; + AudioFormat effective_fmt; struct audsettings obt_as; OSSConf *conf = drv_opaque; @@ -671,7 +671,7 @@ static int oss_init_in(HWVoiceIn *hw, struct audsettings *as, void *drv_opaque) int endianness; int err; int fd; - audfmt_e effective_fmt; + AudioFormat effective_fmt; struct audsettings obt_as; OSSConf *conf = drv_opaque; diff --git a/audio/paaudio.c b/audio/paaudio.c index fea6071..cfdbdc6 100644 --- a/audio/paaudio.c +++ b/audio/paaudio.c @@ -384,21 +384,21 @@ static int qpa_read (SWVoiceIn *sw, void *buf, int len) return audio_pcm_sw_read (sw, buf, len); } -static pa_sample_format_t audfmt_to_pa (audfmt_e afmt, int endianness) +static pa_sample_format_t audfmt_to_pa (AudioFormat afmt, int endianness) { int format; switch (afmt) { - case AUD_FMT_S8: - case AUD_FMT_U8: + case AUDIO_FORMAT_S8: + case AUDIO_FORMAT_U8: format = PA_SAMPLE_U8; break; - case AUD_FMT_S16: - case AUD_FMT_U16: + case AUDIO_FORMAT_S16: + case AUDIO_FORMAT_U16: format = endianness ? PA_SAMPLE_S16BE : PA_SAMPLE_S16LE; break; - case AUD_FMT_S32: - case AUD_FMT_U32: + case AUDIO_FORMAT_S32: + case AUDIO_FORMAT_U32: format = endianness ? PA_SAMPLE_S32BE : PA_SAMPLE_S32LE; break; default: @@ -409,26 +409,26 @@ static pa_sample_format_t audfmt_to_pa (audfmt_e afmt, int endianness) return format; } -static audfmt_e pa_to_audfmt (pa_sample_format_t fmt, int *endianness) +static AudioFormat pa_to_audfmt (pa_sample_format_t fmt, int *endianness) { switch (fmt) { case PA_SAMPLE_U8: - return AUD_FMT_U8; + return AUDIO_FORMAT_U8; case PA_SAMPLE_S16BE: *endianness = 1; - return AUD_FMT_S16; + return AUDIO_FORMAT_S16; case PA_SAMPLE_S16LE: *endianness = 0; - return AUD_FMT_S16; + return AUDIO_FORMAT_S16; case PA_SAMPLE_S32BE: *endianness = 1; - return AUD_FMT_S32; + return AUDIO_FORMAT_S32; case PA_SAMPLE_S32LE: *endianness = 0; - return AUD_FMT_S32; + return AUDIO_FORMAT_S32; default: dolog ("Internal logic error: Bad pa_sample_format %d\n", fmt); - return AUD_FMT_U8; + return AUDIO_FORMAT_U8; } } diff --git a/audio/sdlaudio.c b/audio/sdlaudio.c index 1140f2e..db0f95a 100644 --- a/audio/sdlaudio.c +++ b/audio/sdlaudio.c @@ -115,19 +115,19 @@ static int sdl_unlock_and_post (SDLAudioState *s, const char *forfn) return sdl_post (s, forfn); } -static int aud_to_sdlfmt (audfmt_e fmt) +static int aud_to_sdlfmt (AudioFormat fmt) { switch (fmt) { - case AUD_FMT_S8: + case AUDIO_FORMAT_S8: return AUDIO_S8; - case AUD_FMT_U8: + case AUDIO_FORMAT_U8: return AUDIO_U8; - case AUD_FMT_S16: + case AUDIO_FORMAT_S16: return AUDIO_S16LSB; - case AUD_FMT_U16: + case AUDIO_FORMAT_U16: return AUDIO_U16LSB; default: @@ -139,37 +139,37 @@ static int aud_to_sdlfmt (audfmt_e fmt) } } -static int sdl_to_audfmt(int sdlfmt, audfmt_e *fmt, int *endianness) +static int sdl_to_audfmt(int sdlfmt, AudioFormat *fmt, int *endianness) { switch (sdlfmt) { case AUDIO_S8: *endianness = 0; - *fmt = AUD_FMT_S8; + *fmt = AUDIO_FORMAT_S8; break; case AUDIO_U8: *endianness = 0; - *fmt = AUD_FMT_U8; + *fmt = AUDIO_FORMAT_U8; break; case AUDIO_S16LSB: *endianness = 0; - *fmt = AUD_FMT_S16; + *fmt = AUDIO_FORMAT_S16; break; case AUDIO_U16LSB: *endianness = 0; - *fmt = AUD_FMT_U16; + *fmt = AUDIO_FORMAT_U16; break; case AUDIO_S16MSB: *endianness = 1; - *fmt = AUD_FMT_S16; + *fmt = AUDIO_FORMAT_S16; break; case AUDIO_U16MSB: *endianness = 1; - *fmt = AUD_FMT_U16; + *fmt = AUDIO_FORMAT_U16; break; default: @@ -341,7 +341,7 @@ static int sdl_init_out(HWVoiceOut *hw, struct audsettings *as, SDL_AudioSpec req, obt; int endianness; int err; - audfmt_e effective_fmt; + AudioFormat effective_fmt; struct audsettings obt_as; req.freq = as->freq; diff --git a/audio/spiceaudio.c b/audio/spiceaudio.c index 5c6f726..f556b3b 100644 --- a/audio/spiceaudio.c +++ b/audio/spiceaudio.c @@ -127,7 +127,7 @@ static int line_out_init(HWVoiceOut *hw, struct audsettings *as, settings.freq = SPICE_INTERFACE_PLAYBACK_FREQ; #endif settings.nchannels = SPICE_INTERFACE_PLAYBACK_CHAN; - settings.fmt = AUD_FMT_S16; + settings.fmt = AUDIO_FORMAT_S16; settings.endianness = AUDIO_HOST_ENDIANNESS; audio_pcm_init_info (&hw->info, &settings); @@ -255,7 +255,7 @@ static int line_in_init(HWVoiceIn *hw, struct audsettings *as, void *drv_opaque) settings.freq = SPICE_INTERFACE_RECORD_FREQ; #endif settings.nchannels = SPICE_INTERFACE_RECORD_CHAN; - settings.fmt = AUD_FMT_S16; + settings.fmt = AUDIO_FORMAT_S16; settings.endianness = AUDIO_HOST_ENDIANNESS; audio_pcm_init_info (&hw->info, &settings); diff --git a/audio/wavaudio.c b/audio/wavaudio.c index c586020..62017de 100644 --- a/audio/wavaudio.c +++ b/audio/wavaudio.c @@ -116,20 +116,23 @@ static int wav_init_out(HWVoiceOut *hw, struct audsettings *as, stereo = wav_as.nchannels == 2; switch (wav_as.fmt) { - case AUD_FMT_S8: - case AUD_FMT_U8: + case AUDIO_FORMAT_S8: + case AUDIO_FORMAT_U8: bits16 = 0; break; - case AUD_FMT_S16: - case AUD_FMT_U16: + case AUDIO_FORMAT_S16: + case AUDIO_FORMAT_U16: bits16 = 1; break; - case AUD_FMT_S32: - case AUD_FMT_U32: + case AUDIO_FORMAT_S32: + case AUDIO_FORMAT_U32: dolog ("WAVE files can not handle 32bit formats\n"); return -1; + + case AUDIO_FORMAT_MAX: + abort(); } hdr[34] = bits16 ? 0x10 : 0x08; @@ -224,7 +227,7 @@ static int wav_ctl_out (HWVoiceOut *hw, int cmd, ...) static WAVConf glob_conf = { .settings.freq = 44100, .settings.nchannels = 2, - .settings.fmt = AUD_FMT_S16, + .settings.fmt = AUDIO_FORMAT_S16, .wav_path = "qemu.wav" }; diff --git a/audio/wavcapture.c b/audio/wavcapture.c index 6f6d792..b03c244 100644 --- a/audio/wavcapture.c +++ b/audio/wavcapture.c @@ -135,7 +135,7 @@ int wav_start_capture (CaptureState *s, const char *path, int freq, as.freq = freq; as.nchannels = 1 << stereo; - as.fmt = bits16 ? AUD_FMT_S16 : AUD_FMT_U8; + as.fmt = bits16 ? AUDIO_FORMAT_S16 : AUDIO_FORMAT_U8; as.endianness = 0; ops.notify = wav_notify; diff --git a/hw/arm/omap2.c b/hw/arm/omap2.c index e39b317..3b14a5d 100644 --- a/hw/arm/omap2.c +++ b/hw/arm/omap2.c @@ -269,7 +269,7 @@ static void omap_eac_format_update(struct omap_eac_s *s) * does I2S specify it? */ /* All register writes are 16 bits so we we store 16-bit samples * in the buffers regardless of AGCFR[B8_16] value. */ - fmt.fmt = AUD_FMT_U16; + fmt.fmt = AUDIO_FORMAT_U16; s->codec.in_voice = AUD_open_in(&s->codec.card, s->codec.in_voice, "eac.codec.in", s, omap_eac_in_cb, &fmt); diff --git a/hw/audio/ac97.c b/hw/audio/ac97.c index b173835..fa75f33 100644 --- a/hw/audio/ac97.c +++ b/hw/audio/ac97.c @@ -360,7 +360,7 @@ static void open_voice (AC97LinkState *s, int index, int freq) as.freq = freq; as.nchannels = 2; - as.fmt = AUD_FMT_S16; + as.fmt = AUDIO_FORMAT_S16; as.endianness = 0; if (freq > 0) { diff --git a/hw/audio/adlib.c b/hw/audio/adlib.c index 656eb37..f8f0f55 100644 --- a/hw/audio/adlib.c +++ b/hw/audio/adlib.c @@ -323,7 +323,7 @@ static void adlib_realizefn (DeviceState *dev, Error **errp) as.freq = s->freq; as.nchannels = SHIFT; - as.fmt = AUD_FMT_S16; + as.fmt = AUDIO_FORMAT_S16; as.endianness = AUDIO_HOST_ENDIANNESS; AUD_register_card ("adlib", &s->card); diff --git a/hw/audio/cs4231a.c b/hw/audio/cs4231a.c index f96f561..626a173 100644 --- a/hw/audio/cs4231a.c +++ b/hw/audio/cs4231a.c @@ -284,7 +284,7 @@ static void cs_reset_voices (CSState *s, uint32_t val) switch ((val >> 5) & ((s->dregs[MODE_And_ID] & MODE2) ? 7 : 3)) { case 0: - as.fmt = AUD_FMT_U8; + as.fmt = AUDIO_FORMAT_U8; s->shift = as.nchannels == 2; break; @@ -294,7 +294,7 @@ static void cs_reset_voices (CSState *s, uint32_t val) case 3: s->tab = ALawDecompressTable; x_law: - as.fmt = AUD_FMT_S16; + as.fmt = AUDIO_FORMAT_S16; as.endianness = AUDIO_HOST_ENDIANNESS; s->shift = as.nchannels == 2; break; @@ -302,7 +302,7 @@ static void cs_reset_voices (CSState *s, uint32_t val) case 6: as.endianness = 1; case 2: - as.fmt = AUD_FMT_S16; + as.fmt = AUDIO_FORMAT_S16; s->shift = as.nchannels; break; diff --git a/hw/audio/es1370.c b/hw/audio/es1370.c index 8e7bcf5..f6e74cb 100644 --- a/hw/audio/es1370.c +++ b/hw/audio/es1370.c @@ -414,14 +414,14 @@ static void es1370_update_voices (ES1370State *s, uint32_t ctl, uint32_t sctl) i, new_freq, 1 << (new_fmt & 1), - (new_fmt & 2) ? AUD_FMT_S16 : AUD_FMT_U8, + (new_fmt & 2) ? AUDIO_FORMAT_S16 : AUDIO_FORMAT_U8, d->shift); if (new_freq) { struct audsettings as; as.freq = new_freq; as.nchannels = 1 << (new_fmt & 1); - as.fmt = (new_fmt & 2) ? AUD_FMT_S16 : AUD_FMT_U8; + as.fmt = (new_fmt & 2) ? AUDIO_FORMAT_S16 : AUDIO_FORMAT_U8; as.endianness = 0; if (i == ADC_CHANNEL) { diff --git a/hw/audio/gus.c b/hw/audio/gus.c index 86223a9..6107824 100644 --- a/hw/audio/gus.c +++ b/hw/audio/gus.c @@ -242,7 +242,7 @@ static void gus_realizefn (DeviceState *dev, Error **errp) as.freq = s->freq; as.nchannels = 2; - as.fmt = AUD_FMT_S16; + as.fmt = AUDIO_FORMAT_S16; as.endianness = GUS_ENDIANNESS; s->voice = AUD_open_out ( diff --git a/hw/audio/hda-codec.c b/hw/audio/hda-codec.c index 3c03ff5..8693b7a 100644 --- a/hw/audio/hda-codec.c +++ b/hw/audio/hda-codec.c @@ -97,9 +97,9 @@ static void hda_codec_parse_fmt(uint32_t format, struct audsettings *as) } switch (format & AC_FMT_BITS_MASK) { - case AC_FMT_BITS_8: as->fmt = AUD_FMT_S8; break; - case AC_FMT_BITS_16: as->fmt = AUD_FMT_S16; break; - case AC_FMT_BITS_32: as->fmt = AUD_FMT_S32; break; + case AC_FMT_BITS_8: as->fmt = AUDIO_FORMAT_S8; break; + case AC_FMT_BITS_16: as->fmt = AUDIO_FORMAT_S16; break; + case AC_FMT_BITS_32: as->fmt = AUDIO_FORMAT_S32; break; } as->nchannels = ((format & AC_FMT_CHAN_MASK) >> AC_FMT_CHAN_SHIFT) + 1; @@ -128,12 +128,12 @@ static void hda_codec_parse_fmt(uint32_t format, struct audsettings *as) /* -------------------------------------------------------------------------- */ static const char *fmt2name[] = { - [ AUD_FMT_U8 ] = "PCM-U8", - [ AUD_FMT_S8 ] = "PCM-S8", - [ AUD_FMT_U16 ] = "PCM-U16", - [ AUD_FMT_S16 ] = "PCM-S16", - [ AUD_FMT_U32 ] = "PCM-U32", - [ AUD_FMT_S32 ] = "PCM-S32", + [ AUDIO_FORMAT_U8 ] = "PCM-U8", + [ AUDIO_FORMAT_S8 ] = "PCM-S8", + [ AUDIO_FORMAT_U16 ] = "PCM-U16", + [ AUDIO_FORMAT_S16 ] = "PCM-S16", + [ AUDIO_FORMAT_U32 ] = "PCM-U32", + [ AUDIO_FORMAT_S32 ] = "PCM-S32", }; typedef struct HDAAudioState HDAAudioState; diff --git a/hw/audio/lm4549.c b/hw/audio/lm4549.c index 380ef60..9d4f4b5 100644 --- a/hw/audio/lm4549.c +++ b/hw/audio/lm4549.c @@ -185,7 +185,7 @@ void lm4549_write(lm4549_state *s, struct audsettings as; as.freq = value; as.nchannels = 2; - as.fmt = AUD_FMT_S16; + as.fmt = AUDIO_FORMAT_S16; as.endianness = 0; s->voice = AUD_open_out( @@ -255,7 +255,7 @@ static int lm4549_post_load(void *opaque, int version_id) struct audsettings as; as.freq = freq; as.nchannels = 2; - as.fmt = AUD_FMT_S16; + as.fmt = AUDIO_FORMAT_S16; as.endianness = 0; s->voice = AUD_open_out( @@ -292,7 +292,7 @@ void lm4549_init(lm4549_state *s, lm4549_callback data_req_cb, void* opaque) /* Open a default voice */ as.freq = 48000; as.nchannels = 2; - as.fmt = AUD_FMT_S16; + as.fmt = AUDIO_FORMAT_S16; as.endianness = 0; s->voice = AUD_open_out( diff --git a/hw/audio/milkymist-ac97.c b/hw/audio/milkymist-ac97.c index 28f55e8..15169e2 100644 --- a/hw/audio/milkymist-ac97.c +++ b/hw/audio/milkymist-ac97.c @@ -297,7 +297,7 @@ static int milkymist_ac97_init(SysBusDevice *dev) as.freq = 48000; as.nchannels = 2; - as.fmt = AUD_FMT_S16; + as.fmt = AUDIO_FORMAT_S16; as.endianness = 1; s->voice_in = AUD_open_in(&s->card, s->voice_in, diff --git a/hw/audio/pcspk.c b/hw/audio/pcspk.c index 5266fb5..302debf 100644 --- a/hw/audio/pcspk.c +++ b/hw/audio/pcspk.c @@ -112,7 +112,7 @@ static void pcspk_callback(void *opaque, int free) static int pcspk_audio_init(ISABus *bus) { PCSpkState *s = pcspk_state; - struct audsettings as = {PCSPK_SAMPLE_RATE, 1, AUD_FMT_U8, 0}; + struct audsettings as = {PCSPK_SAMPLE_RATE, 1, AUDIO_FORMAT_U8, 0}; AUD_register_card(s_spk, &s->card); diff --git a/hw/audio/sb16.c b/hw/audio/sb16.c index b052de5..a159dcc 100644 --- a/hw/audio/sb16.c +++ b/hw/audio/sb16.c @@ -66,7 +66,7 @@ typedef struct SB16State { int fmt_stereo; int fmt_signed; int fmt_bits; - audfmt_e fmt; + AudioFormat fmt; int dma_auto; int block_size; int fifo; @@ -221,7 +221,7 @@ static void continue_dma8 (SB16State *s) static void dma_cmd8 (SB16State *s, int mask, int dma_len) { - s->fmt = AUD_FMT_U8; + s->fmt = AUDIO_FORMAT_U8; s->use_hdma = 0; s->fmt_bits = 8; s->fmt_signed = 0; @@ -316,18 +316,18 @@ static void dma_cmd (SB16State *s, uint8_t cmd, uint8_t d0, int dma_len) if (16 == s->fmt_bits) { if (s->fmt_signed) { - s->fmt = AUD_FMT_S16; + s->fmt = AUDIO_FORMAT_S16; } else { - s->fmt = AUD_FMT_U16; + s->fmt = AUDIO_FORMAT_U16; } } else { if (s->fmt_signed) { - s->fmt = AUD_FMT_S8; + s->fmt = AUDIO_FORMAT_S8; } else { - s->fmt = AUD_FMT_U8; + s->fmt = AUDIO_FORMAT_U8; } } @@ -839,7 +839,7 @@ static void legacy_reset (SB16State *s) as.freq = s->freq; as.nchannels = 1; - as.fmt = AUD_FMT_U8; + as.fmt = AUDIO_FORMAT_U8; as.endianness = 0; s->voice = AUD_open_out ( diff --git a/hw/audio/wm8750.c b/hw/audio/wm8750.c index b50b331..4c4333c 100644 --- a/hw/audio/wm8750.c +++ b/hw/audio/wm8750.c @@ -201,7 +201,7 @@ static void wm8750_set_format(WM8750State *s) in_fmt.endianness = 0; in_fmt.nchannels = 2; in_fmt.freq = s->adc_hz; - in_fmt.fmt = AUD_FMT_S16; + in_fmt.fmt = AUDIO_FORMAT_S16; s->adc_voice[0] = AUD_open_in(&s->card, s->adc_voice[0], CODEC ".input1", s, wm8750_audio_in_cb, &in_fmt); @@ -214,7 +214,7 @@ static void wm8750_set_format(WM8750State *s) out_fmt.endianness = 0; out_fmt.nchannels = 2; out_fmt.freq = s->dac_hz; - out_fmt.fmt = AUD_FMT_S16; + out_fmt.fmt = AUDIO_FORMAT_S16; s->dac_voice[0] = AUD_open_out(&s->card, s->dac_voice[0], CODEC ".speaker", s, wm8750_audio_out_cb, &out_fmt); diff --git a/hw/input/tsc210x.c b/hw/input/tsc210x.c index fae3385..3cf938b 100644 --- a/hw/input/tsc210x.c +++ b/hw/input/tsc210x.c @@ -315,7 +315,7 @@ static void tsc2102_audio_output_update(TSC210xState *s) fmt.endianness = 0; fmt.nchannels = 2; fmt.freq = s->codec.tx_rate; - fmt.fmt = AUD_FMT_S16; + fmt.fmt = AUDIO_FORMAT_S16; s->dac_voice[0] = AUD_open_out(&s->card, s->dac_voice[0], "tsc2102.sink", s, (void *) tsc210x_audio_out_cb, &fmt); diff --git a/hw/usb/dev-audio.c b/hw/usb/dev-audio.c index f092bb8..0171579 100644 --- a/hw/usb/dev-audio.c +++ b/hw/usb/dev-audio.c @@ -646,7 +646,7 @@ static void usb_audio_realize(USBDevice *dev, Error **errp) s->out.vol[1] = 240; /* 0 dB */ s->out.as.freq = USBAUDIO_SAMPLE_RATE; s->out.as.nchannels = 2; - s->out.as.fmt = AUD_FMT_S16; + s->out.as.fmt = AUDIO_FORMAT_S16; s->out.as.endianness = 0; streambuf_init(&s->out.buf, s->buffer); diff --git a/ui/vnc.c b/ui/vnc.c index 69b605c..42cefb7 100644 --- a/ui/vnc.c +++ b/ui/vnc.c @@ -2379,12 +2379,12 @@ static int protocol_client_msg(VncState *vs, uint8_t *data, size_t len) if (len == 4) return 10; switch (read_u8(data, 4)) { - case 0: vs->as.fmt = AUD_FMT_U8; break; - case 1: vs->as.fmt = AUD_FMT_S8; break; - case 2: vs->as.fmt = AUD_FMT_U16; break; - case 3: vs->as.fmt = AUD_FMT_S16; break; - case 4: vs->as.fmt = AUD_FMT_U32; break; - case 5: vs->as.fmt = AUD_FMT_S32; break; + case 0: vs->as.fmt = AUDIO_FORMAT_U8; break; + case 1: vs->as.fmt = AUDIO_FORMAT_S8; break; + case 2: vs->as.fmt = AUDIO_FORMAT_U16; break; + case 3: vs->as.fmt = AUDIO_FORMAT_S16; break; + case 4: vs->as.fmt = AUDIO_FORMAT_U32; break; + case 5: vs->as.fmt = AUDIO_FORMAT_S32; break; default: VNC_DEBUG("Invalid audio format %d\n", read_u8(data, 4)); vnc_client_error(vs); @@ -3065,7 +3065,7 @@ void vnc_init_state(VncState *vs) vs->as.freq = 44100; vs->as.nchannels = 2; - vs->as.fmt = AUD_FMT_S16; + vs->as.fmt = AUDIO_FORMAT_S16; vs->as.endianness = 0; qemu_mutex_init(&vs->output_mutex); -- 2.4.3 ^ permalink raw reply related [flat|nested] 42+ messages in thread
* Re: [Qemu-devel] [PATCH v2 5/6] audio: use qapi AudioFormat instead of audfmt_e 2015-06-16 12:49 ` [Qemu-devel] [PATCH v2 5/6] audio: use qapi AudioFormat instead of audfmt_e Kővágó, Zoltán @ 2015-06-17 8:01 ` Markus Armbruster 2015-06-17 11:05 ` Kővágó Zoltán 0 siblings, 1 reply; 42+ messages in thread From: Markus Armbruster @ 2015-06-17 8:01 UTC (permalink / raw) To: Kővágó, Zoltán Cc: Peter Maydell, Michael Walle, qemu-devel, Gerd Hoffmann "Kővágó, Zoltán" <dirty.ice.hu@gmail.com> writes: > I had to include an enum for audio sampling formats into qapi, but that meant > duplicating the audfmt_e enum. This patch replaces audfmt_e and associated > values with the qapi generated AudioFormat enum. > > This patch is mostly a search-and-replace, except for switches where the qapi > generated AUDIO_FORMAT_MAX caused problems. [...] > diff --git a/audio/alsaaudio.c b/audio/alsaaudio.c > index 6315b2d..4d38f5d 100644 > --- a/audio/alsaaudio.c > +++ b/audio/alsaaudio.c [...] > @@ -639,19 +639,22 @@ static int alsa_open (int in, struct alsa_params_req *req, > bytes_per_sec = freq << (nchannels == 2); > > switch (obt->fmt) { > - case AUD_FMT_S8: > - case AUD_FMT_U8: > + case AUDIO_FORMAT_S8: > + case AUDIO_FORMAT_U8: > break; > > - case AUD_FMT_S16: > - case AUD_FMT_U16: > + case AUDIO_FORMAT_S16: > + case AUDIO_FORMAT_U16: > bytes_per_sec <<= 1; > break; > > - case AUD_FMT_S32: > - case AUD_FMT_U32: > + case AUDIO_FORMAT_S32: > + case AUDIO_FORMAT_U32: > bytes_per_sec <<= 2; > break; > + > + case AUDIO_FORMAT_MAX: > + break; Can this happen? > } > > threshold = (conf->threshold * bytes_per_sec) / 1000; > diff --git a/audio/audio.c b/audio/audio.c > index 5be4b15..112b57b 100644 > --- a/audio/audio.c > +++ b/audio/audio.c > @@ -75,7 +75,7 @@ static struct { > .settings = { > .freq = 44100, > .nchannels = 2, > - .fmt = AUD_FMT_S16, > + .fmt = AUDIO_FORMAT_S16, > .endianness = AUDIO_HOST_ENDIANNESS, > } > }, > @@ -87,7 +87,7 @@ static struct { > .settings = { > .freq = 44100, > .nchannels = 2, > - .fmt = AUD_FMT_S16, > + .fmt = AUDIO_FORMAT_S16, > .endianness = AUDIO_HOST_ENDIANNESS, > } > }, > @@ -219,58 +219,61 @@ static char *audio_alloc_prefix (const char *s) > return r; > } > > -static const char *audio_audfmt_to_string (audfmt_e fmt) > +static const char *audio_audfmt_to_string (AudioFormat fmt) > { > switch (fmt) { > - case AUD_FMT_U8: > + case AUDIO_FORMAT_U8: > return "U8"; > > - case AUD_FMT_U16: > + case AUDIO_FORMAT_U16: > return "U16"; > > - case AUD_FMT_S8: > + case AUDIO_FORMAT_S8: > return "S8"; > > - case AUD_FMT_S16: > + case AUDIO_FORMAT_S16: > return "S16"; > > - case AUD_FMT_U32: > + case AUDIO_FORMAT_U32: > return "U32"; > > - case AUD_FMT_S32: > + case AUDIO_FORMAT_S32: > return "S32"; > + > + case AUDIO_FORMAT_MAX: default: would be more defensive. Same elsewhere. > + abort(); > } > > dolog ("Bogus audfmt %d returning S16\n", fmt); > return "S16"; > } [...] ^ permalink raw reply [flat|nested] 42+ messages in thread
* Re: [Qemu-devel] [PATCH v2 5/6] audio: use qapi AudioFormat instead of audfmt_e 2015-06-17 8:01 ` Markus Armbruster @ 2015-06-17 11:05 ` Kővágó Zoltán 2015-06-17 11:51 ` Markus Armbruster 0 siblings, 1 reply; 42+ messages in thread From: Kővágó Zoltán @ 2015-06-17 11:05 UTC (permalink / raw) To: Markus Armbruster; +Cc: Peter Maydell, Michael Walle, qemu-devel, Gerd Hoffmann 2015-06-17 10:01 keltezéssel, Markus Armbruster írta: > "Kővágó, Zoltán" <dirty.ice.hu@gmail.com> writes: > >> I had to include an enum for audio sampling formats into qapi, but that meant >> duplicating the audfmt_e enum. This patch replaces audfmt_e and associated >> values with the qapi generated AudioFormat enum. >> >> This patch is mostly a search-and-replace, except for switches where the qapi >> generated AUDIO_FORMAT_MAX caused problems. > [...] >> diff --git a/audio/alsaaudio.c b/audio/alsaaudio.c >> index 6315b2d..4d38f5d 100644 >> --- a/audio/alsaaudio.c >> +++ b/audio/alsaaudio.c > [...] >> @@ -639,19 +639,22 @@ static int alsa_open (int in, struct alsa_params_req *req, >> bytes_per_sec = freq << (nchannels == 2); >> >> switch (obt->fmt) { >> - case AUD_FMT_S8: >> - case AUD_FMT_U8: >> + case AUDIO_FORMAT_S8: >> + case AUDIO_FORMAT_U8: >> break; >> >> - case AUD_FMT_S16: >> - case AUD_FMT_U16: >> + case AUDIO_FORMAT_S16: >> + case AUDIO_FORMAT_U16: >> bytes_per_sec <<= 1; >> break; >> >> - case AUD_FMT_S32: >> - case AUD_FMT_U32: >> + case AUDIO_FORMAT_S32: >> + case AUDIO_FORMAT_U32: >> bytes_per_sec <<= 2; >> break; >> + >> + case AUDIO_FORMAT_MAX: >> + break; > > Can this happen? Not under normal circumstances, but gcc warns otherwise. >> } >> >> threshold = (conf->threshold * bytes_per_sec) / 1000; >> diff --git a/audio/audio.c b/audio/audio.c >> index 5be4b15..112b57b 100644 >> --- a/audio/audio.c >> +++ b/audio/audio.c >> @@ -75,7 +75,7 @@ static struct { >> .settings = { >> .freq = 44100, >> .nchannels = 2, >> - .fmt = AUD_FMT_S16, >> + .fmt = AUDIO_FORMAT_S16, >> .endianness = AUDIO_HOST_ENDIANNESS, >> } >> }, >> @@ -87,7 +87,7 @@ static struct { >> .settings = { >> .freq = 44100, >> .nchannels = 2, >> - .fmt = AUD_FMT_S16, >> + .fmt = AUDIO_FORMAT_S16, >> .endianness = AUDIO_HOST_ENDIANNESS, >> } >> }, >> @@ -219,58 +219,61 @@ static char *audio_alloc_prefix (const char *s) >> return r; >> } >> >> -static const char *audio_audfmt_to_string (audfmt_e fmt) >> +static const char *audio_audfmt_to_string (AudioFormat fmt) >> { >> switch (fmt) { >> - case AUD_FMT_U8: >> + case AUDIO_FORMAT_U8: >> return "U8"; >> >> - case AUD_FMT_U16: >> + case AUDIO_FORMAT_U16: >> return "U16"; >> >> - case AUD_FMT_S8: >> + case AUDIO_FORMAT_S8: >> return "S8"; >> >> - case AUD_FMT_S16: >> + case AUDIO_FORMAT_S16: >> return "S16"; >> >> - case AUD_FMT_U32: >> + case AUDIO_FORMAT_U32: >> return "U32"; >> >> - case AUD_FMT_S32: >> + case AUDIO_FORMAT_S32: >> return "S32"; >> + >> + case AUDIO_FORMAT_MAX: > > default: would be more defensive. Same elsewhere. Ok, I'll change them to default. >> + abort(); >> } >> >> dolog ("Bogus audfmt %d returning S16\n", fmt); >> return "S16"; >> } > [...] > ^ permalink raw reply [flat|nested] 42+ messages in thread
* Re: [Qemu-devel] [PATCH v2 5/6] audio: use qapi AudioFormat instead of audfmt_e 2015-06-17 11:05 ` Kővágó Zoltán @ 2015-06-17 11:51 ` Markus Armbruster 2015-06-17 16:01 ` Eric Blake 0 siblings, 1 reply; 42+ messages in thread From: Markus Armbruster @ 2015-06-17 11:51 UTC (permalink / raw) To: Kővágó Zoltán Cc: Peter Maydell, Michael Walle, qemu-devel, Gerd Hoffmann "Kővágó Zoltán" <dirty.ice.hu@gmail.com> writes: > 2015-06-17 10:01 keltezéssel, Markus Armbruster írta: >> "Kővágó, Zoltán" <dirty.ice.hu@gmail.com> writes: >> >>> I had to include an enum for audio sampling formats into qapi, but that meant >>> duplicating the audfmt_e enum. This patch replaces audfmt_e and associated >>> values with the qapi generated AudioFormat enum. >>> >>> This patch is mostly a search-and-replace, except for switches where the qapi >>> generated AUDIO_FORMAT_MAX caused problems. >> [...] >>> diff --git a/audio/alsaaudio.c b/audio/alsaaudio.c >>> index 6315b2d..4d38f5d 100644 >>> --- a/audio/alsaaudio.c >>> +++ b/audio/alsaaudio.c >> [...] >>> @@ -639,19 +639,22 @@ static int alsa_open (int in, struct alsa_params_req *req, >>> bytes_per_sec = freq << (nchannels == 2); >>> >>> switch (obt->fmt) { >>> - case AUD_FMT_S8: >>> - case AUD_FMT_U8: >>> + case AUDIO_FORMAT_S8: >>> + case AUDIO_FORMAT_U8: >>> break; >>> >>> - case AUD_FMT_S16: >>> - case AUD_FMT_U16: >>> + case AUDIO_FORMAT_S16: >>> + case AUDIO_FORMAT_U16: >>> bytes_per_sec <<= 1; >>> break; >>> >>> - case AUD_FMT_S32: >>> - case AUD_FMT_U32: >>> + case AUDIO_FORMAT_S32: >>> + case AUDIO_FORMAT_U32: >>> bytes_per_sec <<= 2; >>> break; >>> + >>> + case AUDIO_FORMAT_MAX: >>> + break; >> >> Can this happen? > > Not under normal circumstances, but gcc warns otherwise. Okay, sounds like another case of "default: abort();" to me :) [...] ^ permalink raw reply [flat|nested] 42+ messages in thread
* Re: [Qemu-devel] [PATCH v2 5/6] audio: use qapi AudioFormat instead of audfmt_e 2015-06-17 11:51 ` Markus Armbruster @ 2015-06-17 16:01 ` Eric Blake 0 siblings, 0 replies; 42+ messages in thread From: Eric Blake @ 2015-06-17 16:01 UTC (permalink / raw) To: Markus Armbruster, Kővágó Zoltán Cc: Peter Maydell, Michael Walle, qemu-devel, Gerd Hoffmann [-- Attachment #1: Type: text/plain, Size: 1374 bytes --] On 06/17/2015 05:51 AM, Markus Armbruster wrote: >>>> @@ -639,19 +639,22 @@ static int alsa_open (int in, struct alsa_params_req *req, >>>> bytes_per_sec = freq << (nchannels == 2); >>>> >>>> switch (obt->fmt) { >>>> - case AUD_FMT_S8: >>>> - case AUD_FMT_U8: >>>> + case AUDIO_FORMAT_S8: >>>> + case AUDIO_FORMAT_U8: >>>> + >>>> + case AUDIO_FORMAT_MAX: >>>> + break; >>> >>> Can this happen? >> >> Not under normal circumstances, but gcc warns otherwise. > > Okay, sounds like another case of "default: abort();" to me :) gcc has the annoying habit that it can warn you about missing enum labels in a switch statement, but ONLY if you do not use a 'default:' label. Warning about missing labels is nice if you plan to add more enum values down the road (then the compiler helpfully points out all the code spots that need to deal with the new value) but with the drawback that non-enum values bypass the switch completely. So if you anticipate this set growing, then s/break/abort()/ is all the more you need. On the other hand, if the set is pretty much fixed, then switching to default: covers more cases including non-enum values assigned into obt->fmt. -- Eric Blake eblake redhat com +1-919-301-3266 Libvirt virtualization library http://libvirt.org [-- Attachment #2: OpenPGP digital signature --] [-- Type: application/pgp-signature, Size: 604 bytes --] ^ permalink raw reply [flat|nested] 42+ messages in thread
* [Qemu-devel] [PATCH v2 6/6] audio: -audiodev command line option 2015-06-16 12:49 [Qemu-devel] [PATCH v2 0/6] -audiodev option Kővágó, Zoltán ` (4 preceding siblings ...) 2015-06-16 12:49 ` [Qemu-devel] [PATCH v2 5/6] audio: use qapi AudioFormat instead of audfmt_e Kővágó, Zoltán @ 2015-06-16 12:49 ` Kővágó, Zoltán 2015-06-17 8:13 ` Markus Armbruster 5 siblings, 1 reply; 42+ messages in thread From: Kővágó, Zoltán @ 2015-06-16 12:49 UTC (permalink / raw) To: qemu-devel; +Cc: Paolo Bonzini, Gerd Hoffmann This patch adds an -audiodev command line option, and deprecates the QEMU_* environment variables for audio backend configuration. It's syntax is similar to existing options (-netdev, -device, etc): -audiodev driver_name,property=value,... Audio drivers now get an Audiodev * as config paramters, instead of the global audio_option structs. There is some code in audio/audio_legacy.c that converts the old environment variables to audiodev options (this way backends do not have to worry about legacy options, also print out them with -audio-help, to ease migrating to -audiodev). Although now it's possible to specify multiple -audiodev options on command line, multiple audio backends are not supported yet. Signed-off-by: Kővágó, Zoltán <DirtY.iCE.hu@gmail.com> --- Changes from v1: * updated to everything usecs without suffix * better rounding when converting between usecs and frames/samples/bytes * bugfixes in audio_legacy.c * fixed code broken by qemu_opts changes by Markus audio/Makefile.objs | 2 +- audio/alsaaudio.c | 284 ++++++------------ audio/audio.c | 747 +++++++++++++----------------------------------- audio/audio.h | 23 +- audio/audio_int.h | 7 +- audio/audio_legacy.c | 328 +++++++++++++++++++++ audio/audio_template.h | 13 +- audio/coreaudio.c | 49 +--- audio/dsound_template.h | 6 +- audio/dsoundaudio.c | 60 ++-- audio/noaudio.c | 3 +- audio/ossaudio.c | 155 +++------- audio/paaudio.c | 81 ++---- audio/sdlaudio.c | 24 +- audio/spiceaudio.c | 7 +- audio/wavaudio.c | 61 +--- qemu-options.hx | 218 +++++++++++++- vl.c | 9 +- 18 files changed, 994 insertions(+), 1083 deletions(-) create mode 100644 audio/audio_legacy.c diff --git a/audio/Makefile.objs b/audio/Makefile.objs index 481d1aa..9d8f579 100644 --- a/audio/Makefile.objs +++ b/audio/Makefile.objs @@ -1,4 +1,4 @@ -common-obj-y = audio.o noaudio.o wavaudio.o mixeng.o +common-obj-y = audio.o audio_legacy.o noaudio.o wavaudio.o mixeng.o common-obj-$(CONFIG_SDL) += sdlaudio.o common-obj-$(CONFIG_OSS) += ossaudio.o common-obj-$(CONFIG_SPICE) += spiceaudio.o diff --git a/audio/alsaaudio.c b/audio/alsaaudio.c index 4d38f5d..a660ae1 100644 --- a/audio/alsaaudio.c +++ b/audio/alsaaudio.c @@ -22,6 +22,8 @@ * THE SOFTWARE. */ #include <alsa/asoundlib.h> +#include "qapi/alloc-visitor.h" +#include "qapi-visit.h" #include "qemu-common.h" #include "qemu/main-loop.h" #include "audio.h" @@ -34,28 +36,9 @@ #define AUDIO_CAP "alsa" #include "audio_int.h" -typedef struct ALSAConf { - int size_in_usec_in; - int size_in_usec_out; - const char *pcm_name_in; - const char *pcm_name_out; - unsigned int buffer_size_in; - unsigned int period_size_in; - unsigned int buffer_size_out; - unsigned int period_size_out; - unsigned int threshold; - - int buffer_size_in_overridden; - int period_size_in_overridden; - - int buffer_size_out_overridden; - int period_size_out_overridden; -} ALSAConf; - struct pollhlp { snd_pcm_t *handle; struct pollfd *pfds; - ALSAConf *conf; int count; int mask; }; @@ -67,6 +50,7 @@ typedef struct ALSAVoiceOut { void *pcm_buf; snd_pcm_t *handle; struct pollhlp pollhlp; + Audiodev *dev; } ALSAVoiceOut; typedef struct ALSAVoiceIn { @@ -74,16 +58,13 @@ typedef struct ALSAVoiceIn { snd_pcm_t *handle; void *pcm_buf; struct pollhlp pollhlp; + Audiodev *dev; } ALSAVoiceIn; struct alsa_params_req { int freq; snd_pcm_format_t fmt; int nchannels; - int size_in_usec; - int override_mask; - unsigned int buffer_size; - unsigned int period_size; }; struct alsa_params_obt { @@ -409,7 +390,8 @@ static int alsa_to_audfmt (snd_pcm_format_t alsafmt, AudioFormat *fmt, static void alsa_dump_info (struct alsa_params_req *req, struct alsa_params_obt *obt, - snd_pcm_format_t obtfmt) + snd_pcm_format_t obtfmt, + AudiodevPerDirectionOptions *pdo) { dolog ("parameter | requested value | obtained value\n"); dolog ("format | %10d | %10d\n", req->fmt, obtfmt); @@ -417,8 +399,9 @@ static void alsa_dump_info (struct alsa_params_req *req, req->nchannels, obt->nchannels); dolog ("frequency | %10d | %10d\n", req->freq, obt->freq); dolog ("============================================\n"); - dolog ("requested: buffer size %d period size %d\n", - req->buffer_size, req->period_size); + dolog ("requested: buffer size %" PRId64 " buffer count %" PRId64 "\n", + pdo->has_buffer ? pdo->buffer : 0, + pdo->has_buffer ? pdo->buffer : 0); dolog ("obtained: samples %ld\n", obt->samples); } @@ -452,23 +435,24 @@ static void alsa_set_threshold (snd_pcm_t *handle, snd_pcm_uframes_t threshold) } } -static int alsa_open (int in, struct alsa_params_req *req, - struct alsa_params_obt *obt, snd_pcm_t **handlep, - ALSAConf *conf) +static int alsa_open(bool in, struct alsa_params_req *req, + struct alsa_params_obt *obt, snd_pcm_t **handlep, + Audiodev *dev) { + AudiodevPerDirectionOptions *pdo = in ? dev->in : dev->out; + AudiodevAlsaOptions *aopts = dev->opts->alsa; + AudiodevAlsaPerDirectionOptions *apdo = in ? aopts->in : aopts->out; snd_pcm_t *handle; snd_pcm_hw_params_t *hw_params; int err; - int size_in_usec; unsigned int freq, nchannels; - const char *pcm_name = in ? conf->pcm_name_in : conf->pcm_name_out; + const char *pcm_name = apdo->has_dev ? apdo->dev : "default"; snd_pcm_uframes_t obt_buffer_size; const char *typ = in ? "ADC" : "DAC"; snd_pcm_format_t obtfmt; freq = req->freq; nchannels = req->nchannels; - size_in_usec = req->size_in_usec; snd_pcm_hw_params_alloca (&hw_params); @@ -528,79 +512,49 @@ static int alsa_open (int in, struct alsa_params_req *req, goto err; } - if (req->buffer_size) { - unsigned long obt; + if (pdo->buffer_count) { + if (pdo->buffer) { + int64_t req = pdo->buffer * pdo->buffer_count; - if (size_in_usec) { int dir = 0; - unsigned int btime = req->buffer_size; + unsigned int btime = req; - err = snd_pcm_hw_params_set_buffer_time_near ( - handle, - hw_params, - &btime, - &dir - ); - obt = btime; - } - else { - snd_pcm_uframes_t bsize = req->buffer_size; + err = snd_pcm_hw_params_set_buffer_time_near( + handle, hw_params, &btime, &dir); - err = snd_pcm_hw_params_set_buffer_size_near ( - handle, - hw_params, - &bsize - ); - obt = bsize; - } - if (err < 0) { - alsa_logerr2 (err, typ, "Failed to set buffer %s to %d\n", - size_in_usec ? "time" : "size", req->buffer_size); - goto err; - } + if (err < 0) { + alsa_logerr2(err, typ, + "Failed to set buffer time to %" PRId64 "\n", + req); + goto err; + } - if ((req->override_mask & 2) && (obt - req->buffer_size)) - dolog ("Requested buffer %s %u was rejected, using %lu\n", - size_in_usec ? "time" : "size", req->buffer_size, obt); + if (pdo->has_buffer_count && btime != req) { + dolog("Requested buffer time %" PRId64 + " was rejected, using %u\n", req, btime); + } + } else { + dolog("Can't set buffer_count without buffer_size!\n"); + } } - if (req->period_size) { - unsigned long obt; + if (pdo->buffer) { + int dir = 0; + unsigned int ptime = pdo->buffer; - if (size_in_usec) { - int dir = 0; - unsigned int ptime = req->period_size; - - err = snd_pcm_hw_params_set_period_time_near ( - handle, - hw_params, - &ptime, - &dir - ); - obt = ptime; - } - else { - int dir = 0; - snd_pcm_uframes_t psize = req->period_size; - - err = snd_pcm_hw_params_set_period_size_near ( - handle, - hw_params, - &psize, - &dir - ); - obt = psize; - } + err = snd_pcm_hw_params_set_period_time_near(handle, hw_params, &ptime, + &dir); if (err < 0) { - alsa_logerr2 (err, typ, "Failed to set period %s to %d\n", - size_in_usec ? "time" : "size", req->period_size); + alsa_logerr2(err, typ, "Failed to set period time to %" PRId64 "\n", + pdo->buffer); goto err; } - if (((req->override_mask & 1) && (obt - req->period_size))) - dolog ("Requested period %s %u was rejected, using %lu\n", - size_in_usec ? "time" : "size", req->period_size, obt); + if (pdo->has_buffer && ptime != pdo->buffer) { + dolog("Requested period time %" PRId64 " was rejected, using %d\n", + pdo->buffer, ptime); + } } err = snd_pcm_hw_params (handle, hw_params); @@ -632,7 +586,7 @@ static int alsa_open (int in, struct alsa_params_req *req, goto err; } - if (!in && conf->threshold) { + if (!in && aopts->has_threshold && aopts->threshold) { snd_pcm_uframes_t threshold; int bytes_per_sec; @@ -657,7 +611,7 @@ static int alsa_open (int in, struct alsa_params_req *req, break; } - threshold = (conf->threshold * bytes_per_sec) / 1000; + threshold = (aopts->threshold * bytes_per_sec) / 1000; alsa_set_threshold (handle, threshold); } @@ -671,11 +625,11 @@ static int alsa_open (int in, struct alsa_params_req *req, obt->nchannels != req->nchannels || obt->freq != req->freq) { dolog ("Audio parameters for %s\n", typ); - alsa_dump_info (req, obt, obtfmt); + alsa_dump_info (req, obt, obtfmt, pdo); } #ifdef DEBUG - alsa_dump_info (req, obt, obtfmt); + alsa_dump_info (req, obt, obtfmt, pdo); #endif return 0; @@ -801,19 +755,13 @@ static int alsa_init_out(HWVoiceOut *hw, struct audsettings *as, struct alsa_params_obt obt; snd_pcm_t *handle; struct audsettings obt_as; - ALSAConf *conf = drv_opaque; + Audiodev *dev = drv_opaque; req.fmt = aud_to_alsafmt (as->fmt, as->endianness); req.freq = as->freq; req.nchannels = as->nchannels; - req.period_size = conf->period_size_out; - req.buffer_size = conf->buffer_size_out; - req.size_in_usec = conf->size_in_usec_out; - req.override_mask = - (conf->period_size_out_overridden ? 1 : 0) | - (conf->buffer_size_out_overridden ? 2 : 0); - if (alsa_open (0, &req, &obt, &handle, conf)) { + if (alsa_open (0, &req, &obt, &handle, dev)) { return -1; } @@ -834,7 +782,7 @@ static int alsa_init_out(HWVoiceOut *hw, struct audsettings *as, } alsa->handle = handle; - alsa->pollhlp.conf = conf; + alsa->dev = dev; return 0; } @@ -874,16 +822,12 @@ static int alsa_voice_ctl (snd_pcm_t *handle, const char *typ, int ctl) static int alsa_ctl_out (HWVoiceOut *hw, int cmd, ...) { ALSAVoiceOut *alsa = (ALSAVoiceOut *) hw; + AudiodevAlsaPerDirectionOptions *apdo = alsa->dev->opts->alsa->out; switch (cmd) { case VOICE_ENABLE: { - va_list ap; - int poll_mode; - - va_start (ap, cmd); - poll_mode = va_arg (ap, int); - va_end (ap); + bool poll_mode = !apdo->has_try_poll || apdo->try_poll; ldebug ("enabling voice\n"); if (poll_mode && alsa_poll_out (hw)) { @@ -912,19 +856,13 @@ static int alsa_init_in(HWVoiceIn *hw, struct audsettings *as, void *drv_opaque) struct alsa_params_obt obt; snd_pcm_t *handle; struct audsettings obt_as; - ALSAConf *conf = drv_opaque; + Audiodev *dev = drv_opaque; req.fmt = aud_to_alsafmt (as->fmt, as->endianness); req.freq = as->freq; req.nchannels = as->nchannels; - req.period_size = conf->period_size_in; - req.buffer_size = conf->buffer_size_in; - req.size_in_usec = conf->size_in_usec_in; - req.override_mask = - (conf->period_size_in_overridden ? 1 : 0) | - (conf->buffer_size_in_overridden ? 2 : 0); - if (alsa_open (1, &req, &obt, &handle, conf)) { + if (alsa_open (1, &req, &obt, &handle, dev)) { return -1; } @@ -945,7 +883,7 @@ static int alsa_init_in(HWVoiceIn *hw, struct audsettings *as, void *drv_opaque) } alsa->handle = handle; - alsa->pollhlp.conf = conf; + alsa->dev = dev; return 0; } @@ -1087,16 +1025,12 @@ static int alsa_read (SWVoiceIn *sw, void *buf, int size) static int alsa_ctl_in (HWVoiceIn *hw, int cmd, ...) { ALSAVoiceIn *alsa = (ALSAVoiceIn *) hw; + AudiodevAlsaPerDirectionOptions *apdo = alsa->dev->opts->alsa->in; switch (cmd) { case VOICE_ENABLE: { - va_list ap; - int poll_mode; - - va_start (ap, cmd); - poll_mode = va_arg (ap, int); - va_end (ap); + bool poll_mode = !apdo->has_try_poll || apdo->try_poll; ldebug ("enabling voice\n"); if (poll_mode && alsa_poll_in (hw)) { @@ -1119,88 +1053,35 @@ static int alsa_ctl_in (HWVoiceIn *hw, int cmd, ...) return -1; } -static ALSAConf glob_conf = { - .buffer_size_out = 4096, - .period_size_out = 1024, - .pcm_name_out = "default", - .pcm_name_in = "default", -}; - -static void *alsa_audio_init (void) +static void *alsa_audio_init(Audiodev *dev) { - ALSAConf *conf = g_malloc(sizeof(ALSAConf)); - *conf = glob_conf; - return conf; + assert(dev->opts->kind == AUDIODEV_BACKEND_OPTIONS_KIND_ALSA); + + /* need to define them, as otherwise alsa produces no sound + * doesn't set has_* so alsa_open can identify it wasn't set by the user */ + if (!dev->out->has_buffer_count) { + dev->out->buffer_count = 4; + } + if (!dev->out->has_buffer) { + dev->out->buffer = 23219; /* 1024 frames assuming 44100Hz */ + } + + /* OptsVisitor sets unspecified optional fields to zero, but do not depend + * on it... */ + if (!dev->in->has_buffer_count) { + dev->in->buffer_count = 0; + } + if (!dev->in->has_buffer) { + dev->in->buffer = 0; + } + + return dev; } static void alsa_audio_fini (void *opaque) { - g_free(opaque); } -static struct audio_option alsa_options[] = { - { - .name = "DAC_SIZE_IN_USEC", - .tag = AUD_OPT_BOOL, - .valp = &glob_conf.size_in_usec_out, - .descr = "DAC period/buffer size in microseconds (otherwise in frames)" - }, - { - .name = "DAC_PERIOD_SIZE", - .tag = AUD_OPT_INT, - .valp = &glob_conf.period_size_out, - .descr = "DAC period size (0 to go with system default)", - .overriddenp = &glob_conf.period_size_out_overridden - }, - { - .name = "DAC_BUFFER_SIZE", - .tag = AUD_OPT_INT, - .valp = &glob_conf.buffer_size_out, - .descr = "DAC buffer size (0 to go with system default)", - .overriddenp = &glob_conf.buffer_size_out_overridden - }, - { - .name = "ADC_SIZE_IN_USEC", - .tag = AUD_OPT_BOOL, - .valp = &glob_conf.size_in_usec_in, - .descr = - "ADC period/buffer size in microseconds (otherwise in frames)" - }, - { - .name = "ADC_PERIOD_SIZE", - .tag = AUD_OPT_INT, - .valp = &glob_conf.period_size_in, - .descr = "ADC period size (0 to go with system default)", - .overriddenp = &glob_conf.period_size_in_overridden - }, - { - .name = "ADC_BUFFER_SIZE", - .tag = AUD_OPT_INT, - .valp = &glob_conf.buffer_size_in, - .descr = "ADC buffer size (0 to go with system default)", - .overriddenp = &glob_conf.buffer_size_in_overridden - }, - { - .name = "THRESHOLD", - .tag = AUD_OPT_INT, - .valp = &glob_conf.threshold, - .descr = "(undocumented)" - }, - { - .name = "DAC_DEV", - .tag = AUD_OPT_STR, - .valp = &glob_conf.pcm_name_out, - .descr = "DAC device name (for instance dmix)" - }, - { - .name = "ADC_DEV", - .tag = AUD_OPT_STR, - .valp = &glob_conf.pcm_name_in, - .descr = "ADC device name" - }, - { /* End of list */ } -}; - static struct audio_pcm_ops alsa_pcm_ops = { .init_out = alsa_init_out, .fini_out = alsa_fini_out, @@ -1218,7 +1099,6 @@ static struct audio_pcm_ops alsa_pcm_ops = { struct audio_driver alsa_audio_driver = { .name = "alsa", .descr = "ALSA http://www.alsa-project.org", - .options = alsa_options, .init = alsa_audio_init, .fini = alsa_audio_fini, .pcm_ops = &alsa_pcm_ops, diff --git a/audio/audio.c b/audio/audio.c index 112b57b..daee117 100644 --- a/audio/audio.c +++ b/audio/audio.c @@ -24,7 +24,11 @@ #include "hw/hw.h" #include "audio.h" #include "monitor/monitor.h" +#include "qapi-visit.h" +#include "qapi/alloc-visitor.h" +#include "qapi/opts-visitor.h" #include "qemu/timer.h" +#include "qemu/config-file.h" #include "sysemu/sysemu.h" #define AUDIO_CAP "audio" @@ -42,59 +46,14 @@ The 1st one is the one used by default, that is the reason that we generate the list. */ -static struct audio_driver *drvtab[] = { +struct audio_driver *drvtab[] = { #ifdef CONFIG_SPICE &spice_audio_driver, #endif CONFIG_AUDIO_DRIVERS &no_audio_driver, - &wav_audio_driver -}; - -struct fixed_settings { - int enabled; - int nb_voices; - int greedy; - struct audsettings settings; -}; - -static struct { - struct fixed_settings fixed_out; - struct fixed_settings fixed_in; - union { - int hertz; - int64_t ticks; - } period; - int try_poll_in; - int try_poll_out; -} conf = { - .fixed_out = { /* DAC fixed settings */ - .enabled = 1, - .nb_voices = 1, - .greedy = 1, - .settings = { - .freq = 44100, - .nchannels = 2, - .fmt = AUDIO_FORMAT_S16, - .endianness = AUDIO_HOST_ENDIANNESS, - } - }, - - .fixed_in = { /* ADC fixed settings */ - .enabled = 1, - .nb_voices = 1, - .greedy = 1, - .settings = { - .freq = 44100, - .nchannels = 2, - .fmt = AUDIO_FORMAT_S16, - .endianness = AUDIO_HOST_ENDIANNESS, - } - }, - - .period = { .hertz = 100 }, - .try_poll_in = 1, - .try_poll_out = 1, + &wav_audio_driver, + NULL }; static AudioState glob_audio_state; @@ -113,9 +72,6 @@ const struct mixeng_volume nominal_volume = { #ifdef AUDIO_IS_FLAWLESS_AND_NO_CHECKS_ARE_REQURIED #error No its not #else -static void audio_print_options (const char *prefix, - struct audio_option *opt); - int audio_bug (const char *funcname, int cond) { if (cond) { @@ -123,16 +79,9 @@ int audio_bug (const char *funcname, int cond) AUD_log (NULL, "A bug was just triggered in %s\n", funcname); if (!shown) { - struct audio_driver *d; - shown = 1; AUD_log (NULL, "Save all your work and restart without audio\n"); - AUD_log (NULL, "Please send bug report to av1474@comtv.ru\n"); AUD_log (NULL, "I am sorry\n"); - d = glob_audio_state.drv; - if (d) { - audio_print_options (d->name, d->options); - } } AUD_log (NULL, "Context:\n"); @@ -194,139 +143,6 @@ void *audio_calloc (const char *funcname, int nmemb, size_t size) return g_malloc0 (len); } -static char *audio_alloc_prefix (const char *s) -{ - const char qemu_prefix[] = "QEMU_"; - size_t len, i; - char *r, *u; - - if (!s) { - return NULL; - } - - len = strlen (s); - r = g_malloc (len + sizeof (qemu_prefix)); - - u = r + sizeof (qemu_prefix) - 1; - - pstrcpy (r, len + sizeof (qemu_prefix), qemu_prefix); - pstrcat (r, len + sizeof (qemu_prefix), s); - - for (i = 0; i < len; ++i) { - u[i] = qemu_toupper(u[i]); - } - - return r; -} - -static const char *audio_audfmt_to_string (AudioFormat fmt) -{ - switch (fmt) { - case AUDIO_FORMAT_U8: - return "U8"; - - case AUDIO_FORMAT_U16: - return "U16"; - - case AUDIO_FORMAT_S8: - return "S8"; - - case AUDIO_FORMAT_S16: - return "S16"; - - case AUDIO_FORMAT_U32: - return "U32"; - - case AUDIO_FORMAT_S32: - return "S32"; - - case AUDIO_FORMAT_MAX: - abort(); - } - - dolog ("Bogus audfmt %d returning S16\n", fmt); - return "S16"; -} - -static AudioFormat audio_string_to_audfmt (const char *s, AudioFormat defval, - int *defaultp) -{ - if (!strcasecmp (s, "u8")) { - *defaultp = 0; - return AUDIO_FORMAT_U8; - } - else if (!strcasecmp (s, "u16")) { - *defaultp = 0; - return AUDIO_FORMAT_U16; - } - else if (!strcasecmp (s, "u32")) { - *defaultp = 0; - return AUDIO_FORMAT_U32; - } - else if (!strcasecmp (s, "s8")) { - *defaultp = 0; - return AUDIO_FORMAT_S8; - } - else if (!strcasecmp (s, "s16")) { - *defaultp = 0; - return AUDIO_FORMAT_S16; - } - else if (!strcasecmp (s, "s32")) { - *defaultp = 0; - return AUDIO_FORMAT_S32; - } - else { - dolog ("Bogus audio format `%s' using %s\n", - s, audio_audfmt_to_string (defval)); - *defaultp = 1; - return defval; - } -} - -static AudioFormat audio_get_conf_fmt (const char *envname, - AudioFormat defval, - int *defaultp) -{ - const char *var = getenv (envname); - if (!var) { - *defaultp = 1; - return defval; - } - return audio_string_to_audfmt (var, defval, defaultp); -} - -static int audio_get_conf_int (const char *key, int defval, int *defaultp) -{ - int val; - char *strval; - - strval = getenv (key); - if (strval) { - *defaultp = 0; - val = atoi (strval); - return val; - } - else { - *defaultp = 1; - return defval; - } -} - -static const char *audio_get_conf_str (const char *key, - const char *defval, - int *defaultp) -{ - const char *val = getenv (key); - if (!val) { - *defaultp = 1; - return defval; - } - else { - *defaultp = 0; - return val; - } -} - void AUD_vlog (const char *cap, const char *fmt, va_list ap) { if (cap) { @@ -345,161 +161,6 @@ void AUD_log (const char *cap, const char *fmt, ...) va_end (ap); } -static void audio_print_options (const char *prefix, - struct audio_option *opt) -{ - char *uprefix; - - if (!prefix) { - dolog ("No prefix specified\n"); - return; - } - - if (!opt) { - dolog ("No options\n"); - return; - } - - uprefix = audio_alloc_prefix (prefix); - - for (; opt->name; opt++) { - const char *state = "default"; - printf (" %s_%s: ", uprefix, opt->name); - - if (opt->overriddenp && *opt->overriddenp) { - state = "current"; - } - - switch (opt->tag) { - case AUD_OPT_BOOL: - { - int *intp = opt->valp; - printf ("boolean, %s = %d\n", state, *intp ? 1 : 0); - } - break; - - case AUD_OPT_INT: - { - int *intp = opt->valp; - printf ("integer, %s = %d\n", state, *intp); - } - break; - - case AUD_OPT_FMT: - { - AudioFormat *fmtp = opt->valp; - printf ( - "format, %s = %s, (one of: U8 S8 U16 S16 U32 S32)\n", - state, - audio_audfmt_to_string (*fmtp) - ); - } - break; - - case AUD_OPT_STR: - { - const char **strp = opt->valp; - printf ("string, %s = %s\n", - state, - *strp ? *strp : "(not set)"); - } - break; - - default: - printf ("???\n"); - dolog ("Bad value tag for option %s_%s %d\n", - uprefix, opt->name, opt->tag); - break; - } - printf (" %s\n", opt->descr); - } - - g_free (uprefix); -} - -static void audio_process_options (const char *prefix, - struct audio_option *opt) -{ - char *optname; - const char qemu_prefix[] = "QEMU_"; - size_t preflen, optlen; - - if (audio_bug (AUDIO_FUNC, !prefix)) { - dolog ("prefix = NULL\n"); - return; - } - - if (audio_bug (AUDIO_FUNC, !opt)) { - dolog ("opt = NULL\n"); - return; - } - - preflen = strlen (prefix); - - for (; opt->name; opt++) { - size_t len, i; - int def; - - if (!opt->valp) { - dolog ("Option value pointer for `%s' is not set\n", - opt->name); - continue; - } - - len = strlen (opt->name); - /* len of opt->name + len of prefix + size of qemu_prefix - * (includes trailing zero) + zero + underscore (on behalf of - * sizeof) */ - optlen = len + preflen + sizeof (qemu_prefix) + 1; - optname = g_malloc (optlen); - - pstrcpy (optname, optlen, qemu_prefix); - - /* copy while upper-casing, including trailing zero */ - for (i = 0; i <= preflen; ++i) { - optname[i + sizeof (qemu_prefix) - 1] = qemu_toupper(prefix[i]); - } - pstrcat (optname, optlen, "_"); - pstrcat (optname, optlen, opt->name); - - def = 1; - switch (opt->tag) { - case AUD_OPT_BOOL: - case AUD_OPT_INT: - { - int *intp = opt->valp; - *intp = audio_get_conf_int (optname, *intp, &def); - } - break; - - case AUD_OPT_FMT: - { - AudioFormat *fmtp = opt->valp; - *fmtp = audio_get_conf_fmt (optname, *fmtp, &def); - } - break; - - case AUD_OPT_STR: - { - const char **strp = opt->valp; - *strp = audio_get_conf_str (optname, *strp, &def); - } - break; - - default: - dolog ("Bad value tag for option `%s' - %d\n", - optname, opt->tag); - break; - } - - if (!opt->overriddenp) { - opt->overriddenp = &opt->overridden; - } - *opt->overriddenp = !def; - g_free (optname); - } -} - static void audio_print_settings (struct audsettings *as) { dolog ("frequency=%d nchannels=%d fmt=", as->freq, as->nchannels); @@ -1120,7 +781,7 @@ static void audio_reset_timer (AudioState *s) { if (audio_is_timer_needed ()) { timer_mod (s->ts, - qemu_clock_get_ns(QEMU_CLOCK_VIRTUAL) + conf.period.ticks); + qemu_clock_get_ns(QEMU_CLOCK_VIRTUAL) + s->period_ticks); } else { timer_del (s->ts); @@ -1196,7 +857,7 @@ void AUD_set_active_out (SWVoiceOut *sw, int on) if (!hw->enabled) { hw->enabled = 1; if (s->vm_running) { - hw->pcm_ops->ctl_out (hw, VOICE_ENABLE, conf.try_poll_out); + hw->pcm_ops->ctl_out (hw, VOICE_ENABLE); audio_reset_timer (s); } } @@ -1241,7 +902,7 @@ void AUD_set_active_in (SWVoiceIn *sw, int on) if (!hw->enabled) { hw->enabled = 1; if (s->vm_running) { - hw->pcm_ops->ctl_in (hw, VOICE_ENABLE, conf.try_poll_in); + hw->pcm_ops->ctl_in (hw, VOICE_ENABLE); audio_reset_timer (s); } } @@ -1558,168 +1219,10 @@ void audio_run (const char *msg) #endif } -static struct audio_option audio_options[] = { - /* DAC */ - { - .name = "DAC_FIXED_SETTINGS", - .tag = AUD_OPT_BOOL, - .valp = &conf.fixed_out.enabled, - .descr = "Use fixed settings for host DAC" - }, - { - .name = "DAC_FIXED_FREQ", - .tag = AUD_OPT_INT, - .valp = &conf.fixed_out.settings.freq, - .descr = "Frequency for fixed host DAC" - }, - { - .name = "DAC_FIXED_FMT", - .tag = AUD_OPT_FMT, - .valp = &conf.fixed_out.settings.fmt, - .descr = "Format for fixed host DAC" - }, - { - .name = "DAC_FIXED_CHANNELS", - .tag = AUD_OPT_INT, - .valp = &conf.fixed_out.settings.nchannels, - .descr = "Number of channels for fixed DAC (1 - mono, 2 - stereo)" - }, - { - .name = "DAC_VOICES", - .tag = AUD_OPT_INT, - .valp = &conf.fixed_out.nb_voices, - .descr = "Number of voices for DAC" - }, - { - .name = "DAC_TRY_POLL", - .tag = AUD_OPT_BOOL, - .valp = &conf.try_poll_out, - .descr = "Attempt using poll mode for DAC" - }, - /* ADC */ - { - .name = "ADC_FIXED_SETTINGS", - .tag = AUD_OPT_BOOL, - .valp = &conf.fixed_in.enabled, - .descr = "Use fixed settings for host ADC" - }, - { - .name = "ADC_FIXED_FREQ", - .tag = AUD_OPT_INT, - .valp = &conf.fixed_in.settings.freq, - .descr = "Frequency for fixed host ADC" - }, - { - .name = "ADC_FIXED_FMT", - .tag = AUD_OPT_FMT, - .valp = &conf.fixed_in.settings.fmt, - .descr = "Format for fixed host ADC" - }, - { - .name = "ADC_FIXED_CHANNELS", - .tag = AUD_OPT_INT, - .valp = &conf.fixed_in.settings.nchannels, - .descr = "Number of channels for fixed ADC (1 - mono, 2 - stereo)" - }, - { - .name = "ADC_VOICES", - .tag = AUD_OPT_INT, - .valp = &conf.fixed_in.nb_voices, - .descr = "Number of voices for ADC" - }, - { - .name = "ADC_TRY_POLL", - .tag = AUD_OPT_BOOL, - .valp = &conf.try_poll_in, - .descr = "Attempt using poll mode for ADC" - }, - /* Misc */ - { - .name = "TIMER_PERIOD", - .tag = AUD_OPT_INT, - .valp = &conf.period.hertz, - .descr = "Timer period in HZ (0 - use lowest possible)" - }, - { /* End of list */ } -}; - -static void audio_pp_nb_voices (const char *typ, int nb) +static int audio_driver_init(AudioState *s, struct audio_driver *drv, + Audiodev *dev) { - switch (nb) { - case 0: - printf ("Does not support %s\n", typ); - break; - case 1: - printf ("One %s voice\n", typ); - break; - case INT_MAX: - printf ("Theoretically supports many %s voices\n", typ); - break; - default: - printf ("Theoretically supports up to %d %s voices\n", nb, typ); - break; - } - -} - -void AUD_help (void) -{ - size_t i; - - audio_process_options ("AUDIO", audio_options); - for (i = 0; i < ARRAY_SIZE (drvtab); i++) { - struct audio_driver *d = drvtab[i]; - if (d->options) { - audio_process_options (d->name, d->options); - } - } - - printf ("Audio options:\n"); - audio_print_options ("AUDIO", audio_options); - printf ("\n"); - - printf ("Available drivers:\n"); - - for (i = 0; i < ARRAY_SIZE (drvtab); i++) { - struct audio_driver *d = drvtab[i]; - - printf ("Name: %s\n", d->name); - printf ("Description: %s\n", d->descr); - - audio_pp_nb_voices ("playback", d->max_voices_out); - audio_pp_nb_voices ("capture", d->max_voices_in); - - if (d->options) { - printf ("Options:\n"); - audio_print_options (d->name, d->options); - } - else { - printf ("No options\n"); - } - printf ("\n"); - } - - printf ( - "Options are settable through environment variables.\n" - "Example:\n" -#ifdef _WIN32 - " set QEMU_AUDIO_DRV=wav\n" - " set QEMU_WAV_PATH=c:\\tune.wav\n" -#else - " export QEMU_AUDIO_DRV=wav\n" - " export QEMU_WAV_PATH=$HOME/tune.wav\n" - "(for csh replace export with setenv in the above)\n" -#endif - " qemu ...\n\n" - ); -} - -static int audio_driver_init (AudioState *s, struct audio_driver *drv) -{ - if (drv->options) { - audio_process_options (drv->name, drv->options); - } - s->drv_opaque = drv->init (); + s->drv_opaque = drv->init(dev); if (s->drv_opaque) { audio_init_nb_voices_out (drv); @@ -1743,11 +1246,11 @@ static void audio_vm_change_state_handler (void *opaque, int running, s->vm_running = running; while ((hwo = audio_pcm_hw_find_any_enabled_out (hwo))) { - hwo->pcm_ops->ctl_out (hwo, op, conf.try_poll_out); + hwo->pcm_ops->ctl_out (hwo, op); } while ((hwi = audio_pcm_hw_find_any_enabled_in (hwi))) { - hwi->pcm_ops->ctl_in (hwi, op, conf.try_poll_in); + hwi->pcm_ops->ctl_in (hwi, op); } audio_reset_timer (s); } @@ -1786,6 +1289,8 @@ static void audio_atexit (void) if (s->drv) { s->drv->fini (s->drv_opaque); } + + qapi_free_Audiodev(s->dev); } static const VMStateDescription vmstate_audio = { @@ -1797,18 +1302,37 @@ static const VMStateDescription vmstate_audio = { } }; -static void audio_init (void) +static Audiodev *parse_option(QemuOpts *opts, Error **errp); +static int audio_init(Audiodev *dev) { size_t i; int done = 0; - const char *drvname; + const char *drvname = NULL; VMChangeStateEntry *e; AudioState *s = &glob_audio_state; + QemuOptsList *list = NULL; /* silence gcc warning about uninitialized + * variable */ if (s->drv) { - return; + if (dev) { + dolog("Cannot create more than one audio backend, sorry\n"); + qapi_free_Audiodev(dev); + } + return -1; } + if (dev) { + drvname = AudiodevBackendOptionsKind_lookup[dev->opts->kind]; + } else { + audio_handle_legacy_opts(); + list = qemu_find_opts("audiodev"); + dev = parse_option(QTAILQ_FIRST(&list->head), &error_abort); + if (!dev) { + exit(1); + } + } + s->dev = dev; + QLIST_INIT (&s->hw_head_out); QLIST_INIT (&s->hw_head_in); QLIST_INIT (&s->cap_head); @@ -1819,10 +1343,8 @@ static void audio_init (void) hw_error("Could not create audio timer\n"); } - audio_process_options ("AUDIO", audio_options); - - s->nb_hw_voices_out = conf.fixed_out.nb_voices; - s->nb_hw_voices_in = conf.fixed_in.nb_voices; + s->nb_hw_voices_out = dev->out->voices; + s->nb_hw_voices_in = dev->in->voices; if (s->nb_hw_voices_out <= 0) { dolog ("Bogus number of playback voices %d, setting to 1\n", @@ -1836,17 +1358,12 @@ static void audio_init (void) s->nb_hw_voices_in = 0; } - { - int def; - drvname = audio_get_conf_str ("QEMU_AUDIO_DRV", NULL, &def); - } - if (drvname) { int found = 0; - for (i = 0; i < ARRAY_SIZE (drvtab); i++) { + for (i = 0; drvtab[i]; i++) { if (!strcmp (drvname, drvtab[i]->name)) { - done = !audio_driver_init (s, drvtab[i]); + done = !audio_driver_init (s, drvtab[i], dev); found = 1; break; } @@ -1854,20 +1371,24 @@ static void audio_init (void) if (!found) { dolog ("Unknown audio driver `%s'\n", drvname); - dolog ("Run with -audio-help to list available drivers\n"); } - } - - if (!done) { - for (i = 0; !done && i < ARRAY_SIZE (drvtab); i++) { - if (drvtab[i]->can_be_default) { - done = !audio_driver_init (s, drvtab[i]); + } else { + for (i = 0; !done && drvtab[i]; i++) { + QemuOpts *opts = qemu_opts_find(list, drvtab[i]->name); + if (opts) { + qapi_free_Audiodev(dev); + dev = parse_option(opts, &error_abort); + if (!dev) { + exit(1); + } + s->dev = dev; + done = !audio_driver_init(s, drvtab[i], dev); } } } if (!done) { - done = !audio_driver_init (s, &no_audio_driver); + done = !audio_driver_init (s, &no_audio_driver, dev); if (!done) { hw_error("Could not initialize audio subsystem\n"); } @@ -1876,16 +1397,16 @@ static void audio_init (void) } } - if (conf.period.hertz <= 0) { - if (conf.period.hertz < 0) { - dolog ("warning: Timer period is negative - %d " - "treating as zero\n", - conf.period.hertz); + if (dev->timer_period <= 0) { + if (dev->timer_period < 0) { + dolog ("warning: Timer period is negative - %" PRId64 + " treating as zero\n", + dev->timer_period); } - conf.period.ticks = 1; + s->period_ticks = 1; } else { - conf.period.ticks = - muldiv64 (1, get_ticks_per_sec (), conf.period.hertz); + s->period_ticks = + muldiv64(dev->timer_period, get_ticks_per_sec(), 1000000); } e = qemu_add_vm_change_state_handler (audio_vm_change_state_handler, s); @@ -1896,11 +1417,12 @@ static void audio_init (void) QLIST_INIT (&s->card_head); vmstate_register (NULL, 0, &vmstate_audio, s); + return 0; } void AUD_register_card (const char *name, QEMUSoundCard *card) { - audio_init (); + audio_init(NULL); card->name = g_strdup (name); memset (&card->entries, 0, sizeof (card->entries)); QLIST_INSERT_HEAD (&glob_audio_state.card_head, card, entries); @@ -2070,3 +1592,142 @@ void AUD_set_volume_in (SWVoiceIn *sw, int mute, uint8_t lvol, uint8_t rvol) } } } + +QemuOptsList qemu_audiodev_opts = { + .name = "audiodev", + .head = QTAILQ_HEAD_INITIALIZER(qemu_audiodev_opts.head), + .implied_opt_name = "type", + .desc = { + /* + * no elements => accept any params + * sanity checking will happen later + */ + { /* end of list */ } + }, +}; + +static void set_per_direction_defaults(AudiodevPerDirectionOptions *pdo) +{ + if (!pdo->has_fixed_settings) { + pdo->has_fixed_settings = true; + pdo->fixed_settings = true; + } + if (!pdo->has_frequency) { + pdo->has_frequency = true; + pdo->frequency = 44100; + } + if (!pdo->has_channels) { + pdo->has_channels = true; + pdo->channels = 2; + } + if (!pdo->has_voices) { + pdo->has_voices = true; + pdo->voices = 1; + } + if (!pdo->has_format) { + pdo->has_format = true; + pdo->format = AUDIO_FORMAT_S16; + } +} + +static Audiodev *parse_option(QemuOpts *opts, Error **errp) +{ + Error *local_err = NULL; + OptsVisitor *ov = opts_visitor_new(opts); + Audiodev *dev = NULL; + visit_type_Audiodev(opts_get_visitor(ov), &dev, NULL, &local_err); + opts_visitor_cleanup(ov); + + if (local_err) { + error_propagate(errp, local_err); + return NULL; + } + + if (!dev->has_id) { + dev->has_id = true; + dev->id = g_strdup("default"); + } + + set_per_direction_defaults(dev->in); + set_per_direction_defaults(dev->out); + + if (!dev->has_timer_period) { + dev->has_timer_period = true; + dev->timer_period = 10000; /* 100Hz -> 10ms */ + } + + return dev; +} + +static int each_option(void *opaque, QemuOpts *opts, Error **errp) +{ + Audiodev *dev = parse_option(opts, errp); + if (!dev) { + return -1; + } + return audio_init(dev); +} + +void audio_set_options(void) +{ + if (qemu_opts_foreach(qemu_find_opts("audiodev"), each_option, NULL, + &error_abort)) { + exit(1); + } +} + +audsettings audiodev_to_audsettings(AudiodevPerDirectionOptions *pdo) +{ + return (audsettings) { + .freq = pdo->frequency, + .nchannels = pdo->channels, + .fmt = pdo->format, + .endianness = AUDIO_HOST_ENDIANNESS, + }; +} + +int audioformat_bytes_per_sample(AudioFormat fmt) +{ + switch (fmt) { + case AUDIO_FORMAT_U8: + case AUDIO_FORMAT_S8: + return 1; + + case AUDIO_FORMAT_U16: + case AUDIO_FORMAT_S16: + return 2; + + case AUDIO_FORMAT_U32: + case AUDIO_FORMAT_S32: + return 4; + + case AUDIO_FORMAT_MAX: + ; + } + abort(); +} + + +/* frames = freq * usec / 1e6 */ +int audio_buffer_frames(AudiodevPerDirectionOptions *pdo, + audsettings *as, int def_usecs) +{ + uint64_t usecs = pdo->has_buffer ? pdo->buffer : def_usecs; + return (as->freq * usecs + 500000) / 1000000; +} + +/* samples = channels * frames = channels * freq * usec / 1e6 */ +int audio_buffer_samples(AudiodevPerDirectionOptions *pdo, + audsettings *as, int def_usecs) +{ + return as->nchannels * audio_buffer_frames(pdo, as, def_usecs); +} + +/* bytes = bytes_per_sample * samples = + * bytes_per_sample * channels * freq * usec / 1e6 */ +int audio_buffer_bytes(AudiodevPerDirectionOptions *pdo, + audsettings *as, int def_usecs) +{ + return audio_buffer_samples(pdo, as, def_usecs) * + audioformat_bytes_per_sample(as->fmt); +} diff --git a/audio/audio.h b/audio/audio.h index e300511..177a673 100644 --- a/audio/audio.h +++ b/audio/audio.h @@ -24,7 +24,10 @@ #ifndef QEMU_AUDIO_H #define QEMU_AUDIO_H +#include <stdarg.h> #include "config-host.h" +#include "qapi-types.h" +#include "qemu/option.h" #include "qemu/queue.h" typedef void (*audio_callback_fn) (void *opaque, int avail); @@ -35,12 +38,21 @@ typedef void (*audio_callback_fn) (void *opaque, int avail); #define AUDIO_HOST_ENDIANNESS 0 #endif -struct audsettings { +typedef struct audsettings { int freq; int nchannels; AudioFormat fmt; int endianness; -}; +} audsettings; + +audsettings audiodev_to_audsettings(AudiodevPerDirectionOptions *pdo); +int audioformat_bytes_per_sample(AudioFormat fmt); +int audio_buffer_frames(AudiodevPerDirectionOptions *pdo, + audsettings *as, int def_usecs); +int audio_buffer_samples(AudiodevPerDirectionOptions *pdo, + audsettings *as, int def_usecs); +int audio_buffer_bytes(AudiodevPerDirectionOptions *pdo, + audsettings *as, int def_usecs); typedef enum { AUD_CNOTIFY_ENABLE, @@ -77,10 +89,11 @@ typedef struct QEMUAudioTimeStamp { uint64_t old_ts; } QEMUAudioTimeStamp; +extern QemuOptsList qemu_audiodev_opts; + void AUD_vlog (const char *cap, const char *fmt, va_list ap) GCC_FMT_ATTR(2, 0); void AUD_log (const char *cap, const char *fmt, ...) GCC_FMT_ATTR(2, 3); -void AUD_help (void); void AUD_register_card (const char *name, QEMUSoundCard *card); void AUD_remove_card (QEMUSoundCard *card); CaptureVoiceOut *AUD_add_capture ( @@ -154,4 +167,8 @@ static inline void *advance (void *p, int incr) int wav_start_capture (CaptureState *s, const char *path, int freq, int bits, int nchannels); +void audio_set_options(void); +void audio_handle_legacy_opts(void); +void audio_legacy_help(void); + #endif /* audio.h */ diff --git a/audio/audio_int.h b/audio/audio_int.h index 566df5e..c4539e7 100644 --- a/audio/audio_int.h +++ b/audio/audio_int.h @@ -143,8 +143,7 @@ struct SWVoiceIn { struct audio_driver { const char *name; const char *descr; - struct audio_option *options; - void *(*init) (void); + void *(*init) (Audiodev *); void (*fini) (void *); struct audio_pcm_ops *pcm_ops; int can_be_default; @@ -190,6 +189,7 @@ struct SWVoiceCap { struct AudioState { struct audio_driver *drv; + Audiodev *dev; void *drv_opaque; QEMUTimer *ts; @@ -200,6 +200,7 @@ struct AudioState { int nb_hw_voices_out; int nb_hw_voices_in; int vm_running; + int64_t period_ticks; }; extern struct audio_driver no_audio_driver; @@ -213,6 +214,8 @@ extern struct audio_driver pa_audio_driver; extern struct audio_driver spice_audio_driver; extern const struct mixeng_volume nominal_volume; +extern struct audio_driver *drvtab[]; + void audio_pcm_init_info (struct audio_pcm_info *info, struct audsettings *as); void audio_pcm_info_clear_buf (struct audio_pcm_info *info, void *buf, int len); diff --git a/audio/audio_legacy.c b/audio/audio_legacy.c new file mode 100644 index 0000000..4987427 --- /dev/null +++ b/audio/audio_legacy.c @@ -0,0 +1,328 @@ +#include "audio.h" +#include "qemu-common.h" +#include "qemu/config-file.h" + +#define AUDIO_CAP "audio-legacy" +#include "audio_int.h" + +typedef enum EnvTransform { + ENV_TRANSFORM_NONE, + ENV_TRANSFORM_BOOL, + ENV_TRANSFORM_FMT, + ENV_TRANSFORM_FRAMES_TO_USECS_IN, + ENV_TRANSFORM_FRAMES_TO_USECS_OUT, + ENV_TRANSFORM_SAMPLES_TO_USECS_IN, + ENV_TRANSFORM_SAMPLES_TO_USECS_OUT, + ENV_TRANSFORM_BYTES_TO_USECS_IN, + ENV_TRANSFORM_BYTES_TO_USECS_OUT, + ENV_TRANSFORM_MILLIS_TO_USECS, + ENV_TRANSFORM_HZ_TO_USECS, +} EnvTransform; + +typedef struct SimpleEnvMap { + const char *name; + const char *option; + EnvTransform transform; +} SimpleEnvMap; + +SimpleEnvMap global_map[] = { + /* DAC/out settings */ + { "QEMU_AUDIO_DAC_FIXED_SETTINGS", "out.fixed-settings", + ENV_TRANSFORM_BOOL }, + { "QEMU_AUDIO_DAC_FIXED_FREQ", "out.frequency" }, + { "QEMU_AUDIO_DAC_FIXED_FMT", "out.format", ENV_TRANSFORM_FMT }, + { "QEMU_AUDIO_DAC_FIXED_CHANNELS", "out.channels" }, + { "QEMU_AUDIO_DAC_VOICES", "out.voices" }, + + /* ADC/in settings */ + { "QEMU_AUDIO_ADC_FIXED_SETTINGS", "in.fixed-settings", + ENV_TRANSFORM_BOOL }, + { "QEMU_AUDIO_ADC_FIXED_FREQ", "in.frequency" }, + { "QEMU_AUDIO_ADC_FIXED_FMT", "in.format", ENV_TRANSFORM_FMT }, + { "QEMU_AUDIO_ADC_FIXED_CHANNELS", "in.channels" }, + { "QEMU_AUDIO_ADC_VOICES", "in.voices" }, + + /* general */ + { "QEMU_AUDIO_TIMER_PERIOD", "timer-period", ENV_TRANSFORM_HZ_TO_USECS }, + { /* End of list */ } +}; + +SimpleEnvMap alsa_map[] = { + { "QEMU_AUDIO_DAC_TRY_POLL", "out.try-poll", ENV_TRANSFORM_BOOL }, + { "QEMU_AUDIO_ADC_TRY_POLL", "in.try-poll", ENV_TRANSFORM_BOOL }, + + { "QEMU_ALSA_THRESHOLD", "threshold" }, + { "QEMU_ALSA_DAC_DEV", "out.dev" }, + { "QEMU_ALSA_ADC_DEV", "in.dev" }, + + { /* End of list */ } +}; + +SimpleEnvMap coreaudio_map[] = { + { "QEMU_COREAUDIO_BUFFER_SIZE", "buffer", + ENV_TRANSFORM_FRAMES_TO_USECS_OUT }, + { "QEMU_COREAUDIO_BUFFER_COUNT", "buffer-count" }, + + { /* End of list */ } +}; + +SimpleEnvMap dsound_map[] = { + { "QEMU_DSOUND_LATENCY_MILLIS", "latency", ENV_TRANSFORM_MILLIS_TO_USECS }, + { "QEMU_DSOUND_BUFSIZE_OUT", "out.buffer", + ENV_TRANSFORM_BYTES_TO_USECS_OUT }, + { "QEMU_DSOUND_BUFSIZE_IN", "in.buffer", + ENV_TRANSFORM_BYTES_TO_USECS_IN }, + + { /* End of list */ } +}; + +SimpleEnvMap oss_map[] = { + { "QEMU_AUDIO_DAC_TRY_POLL", "out.try-poll", ENV_TRANSFORM_BOOL }, + { "QEMU_AUDIO_ADC_TRY_POLL", "in.try-poll", ENV_TRANSFORM_BOOL }, + + { "QEMU_OSS_FRAGSIZE", "buffer", ENV_TRANSFORM_BYTES_TO_USECS_OUT }, + { "QEMU_OSS_NFRAGS", "buffer-count" }, + { "QEMU_OSS_MMAP", "mmap", ENV_TRANSFORM_BOOL }, + { "QEMU_OSS_DAC_DEV", "out.dev" }, + { "QEMU_OSS_ADC_DEV", "in.dev" }, + { "QEMU_OSS_EXCLUSIVE", "exclusive", ENV_TRANSFORM_BOOL }, + { "QEMU_OSS_POLICY", "dsp-policy" }, + + { /* End of list */ } +}; + +SimpleEnvMap pa_map[] = { + { "QEMU_PA_SAMPLES", "buffer", ENV_TRANSFORM_SAMPLES_TO_USECS_OUT }, + { "QEMU_PA_SERVER", "server" }, + { "QEMU_PA_SINK", "sink" }, + { "QEMU_PA_SOURCE", "source" }, + + { /* End of list */ } +}; + +SimpleEnvMap sdl_map[] = { + { "QEMU_SDL_SAMPLES", "buffer", ENV_TRANSFORM_SAMPLES_TO_USECS_OUT }, + { /* End of list */ } +}; + +SimpleEnvMap wav_map[] = { + { "QEMU_WAV_FREQUENCY", "out.frequency" }, + { "QEMU_WAV_FORMAT", "out.format", ENV_TRANSFORM_FMT }, + { "QEMU_WAV_DAC_FIXED_CHANNELS", "out.channels" }, + { "QEMU_WAV_PATH", "path" }, + { /* End of list */ } +}; + +static unsigned long long toull(const char *str) +{ + unsigned long long ret; + if (parse_uint_full(str, &ret, 10)) { + dolog("Invalid boolean value `%s'\n", str); + exit(1); + } + return ret; +} + +/* non reentrant typesafe or anything, but enough in this small c file */ +static const char *tostr(unsigned long long val) +{ + #define LEN ((CHAR_BIT * sizeof(int) - 1) / 3 + 2) + static char ret[LEN]; + snprintf(ret, LEN, "%llu", val); + return ret; +} + +static uint64_t frames_to_usecs(QemuOpts *opts, uint64_t frames, bool in) +{ + const char *opt = in ? "in.frequency" : "out.frequency"; + uint64_t freq = qemu_opt_get_number(opts, opt, 44100); + return (frames * 1000000 + freq/2) / freq; +} + +static uint64_t samples_to_usecs(QemuOpts *opts, uint64_t samples, bool in) +{ + const char *opt = in ? "in.channels" : "out.channels"; + uint64_t channels = qemu_opt_get_number(opts, opt, 2); + return frames_to_usecs(opts, samples/channels, in); +} + +static uint64_t bytes_to_usecs(QemuOpts *opts, uint64_t bytes, bool in) +{ + const char *opt = in ? "in.format" : "out.format"; + const char *val = qemu_opt_get(opts, opt); + uint64_t bytes_per_sample = (val ? toull(val) : 16) / 8; + return samples_to_usecs(opts, bytes * bytes_per_sample, in); +} + +static const char *transform_val(QemuOpts *opts, const char *val, + EnvTransform transform) +{ + switch (transform) { + case ENV_TRANSFORM_NONE: + return val; + + case ENV_TRANSFORM_BOOL: + return toull(val) ? "on" : "off"; + + case ENV_TRANSFORM_FMT: + if (strcasecmp(val, "u8") == 0) { + return "u8"; + } else if (strcasecmp(val, "u16") == 0) { + return "u16"; + } else if (strcasecmp(val, "u32") == 0) { + return "u32"; + } else if (strcasecmp(val, "s8") == 0) { + return "s8"; + } else if (strcasecmp(val, "s16") == 0) { + return "s16"; + } else if (strcasecmp(val, "s32") == 0) { + return "s32"; + } else { + dolog("Invalid audio format `%s'\n", val); + exit(1); + } + + case ENV_TRANSFORM_FRAMES_TO_USECS_IN: + return tostr(frames_to_usecs(opts, toull(val), true)); + case ENV_TRANSFORM_FRAMES_TO_USECS_OUT: + return tostr(frames_to_usecs(opts, toull(val), false)); + + case ENV_TRANSFORM_SAMPLES_TO_USECS_IN: + return tostr(samples_to_usecs(opts, toull(val), true)); + case ENV_TRANSFORM_SAMPLES_TO_USECS_OUT: + return tostr(samples_to_usecs(opts, toull(val), false)); + + case ENV_TRANSFORM_BYTES_TO_USECS_IN: + return tostr(bytes_to_usecs(opts, toull(val), true)); + case ENV_TRANSFORM_BYTES_TO_USECS_OUT: + return tostr(bytes_to_usecs(opts, toull(val), false)); + + case ENV_TRANSFORM_MILLIS_TO_USECS: + return tostr(toull(val) * 1000); + + case ENV_TRANSFORM_HZ_TO_USECS: + return tostr(1000000 / toull(val)); + } + + abort(); /* it's unreachable, gcc */ +} + +static void handle_env_opts(QemuOpts *opts, SimpleEnvMap *map) +{ + while (map->name) { + const char *val = getenv(map->name); + + if (val) { + qemu_opt_set(opts, map->option, + transform_val(opts, val, map->transform), + &error_abort); + } + + ++map; + } +} + +static void handle_alsa_side(QemuOpts *opts, int period, int buffer, + const char *usec_env, const char *period_env, + const char *buffer_env, const char *usec_opt, + const char *count_opt, bool in) +{ + char *usec_s, *period_s, *buffer_s; + bool usec = false; + + usec_s = getenv("QEMU_ALSA_DAC_SIZE_IN_USEC"); + if (usec_s) { + usec = toull(usec_s); + } + + period_s = getenv(period_env); + if (period_s) { + period = toull(period_s); + } + if (!usec) { + period = frames_to_usecs(opts, period, in); + } + if (period_s) { + qemu_opt_set(opts, usec_opt, tostr(period), &error_abort); + } + + buffer_s = getenv(buffer_env); + if (buffer_s) { + buffer = toull(buffer_s); + if (!usec) { + buffer = frames_to_usecs(opts, buffer, in); + } + printf("buffer %d period %d\n", buffer, period); + qemu_opt_set(opts, count_opt, tostr((buffer+period/2)/period), + &error_abort); + } +} + +static void handle_alsa(QemuOpts *opts) +{ + handle_alsa_side(opts, 1024, 4096, + "QEMU_ALSA_DAC_SIZE_IN_USEC", "QEMU_ALSA_DAC_PERIOD_SIZE", + "QEMU_ALSA_DAC_BUFFER_SIZE", + "out.buffer", "out.buffer-count", false); + handle_alsa_side(opts, 0, 0, + "QEMU_ALSA_ADC_SIZE_IN_USEC", "QEMU_ALSA_ADC_PERIOD_SIZE", + "QEMU_ALSA_ADC_BUFFER_SIZE", + "in.buffer", "in.buffer-count", true); +} + +static void legacy_opt(const char *drv) +{ + QemuOpts *opts; + opts = qemu_opts_create(qemu_find_opts("audiodev"), drv, true, + &error_abort); + qemu_opt_set(opts, "type", drv, &error_abort); + + handle_env_opts(opts, global_map); + + if (strcmp(drv, "alsa") == 0) { + handle_env_opts(opts, alsa_map); + handle_alsa(opts); + } else if (strcmp(drv, "oss") == 0) { + handle_env_opts(opts, oss_map); + } else if (strcmp(drv, "pa") == 0) { + handle_env_opts(opts, pa_map); + } else if (strcmp(drv, "sdl") == 0) { + handle_env_opts(opts, sdl_map); + } else if (strcmp(drv, "wav") == 0) { + handle_env_opts(opts, wav_map); + } +} + +void audio_handle_legacy_opts(void) +{ + const char *drv = getenv("QEMU_AUDIO_DRV"); + + if (drv) { + legacy_opt(drv); + } else { + struct audio_driver **drv; + for (drv = drvtab; *drv; ++drv) { + if ((*drv)->can_be_default) { + legacy_opt((*drv)->name); + } + } + } +} + +static int legacy_help_each(void *opaque, QemuOpts *opts, Error **errp) +{ + printf("-audiodev "); + qemu_opts_print(opts, ","); + printf("\n"); + return 0; +} + +void audio_legacy_help(void) +{ + printf("Environment variable based configuration deprecated.\n"); + printf("Please use the new -audiodev option.\n"); + + audio_handle_legacy_opts(); + printf("\nEquivalent -audiodev to your current environment variables:\n"); + qemu_opts_foreach(qemu_find_opts("audiodev"), legacy_help_each, NULL, NULL); +} diff --git a/audio/audio_template.h b/audio/audio_template.h index 99b27b2..096b2b3 100644 --- a/audio/audio_template.h +++ b/audio/audio_template.h @@ -302,8 +302,10 @@ static HW *glue (audio_pcm_hw_add_new_, TYPE) (struct audsettings *as) static HW *glue (audio_pcm_hw_add_, TYPE) (struct audsettings *as) { HW *hw; + AudioState *s = &glob_audio_state; + AudiodevPerDirectionOptions *pdo = s->dev->TYPE; - if (glue (conf.fixed_, TYPE).enabled && glue (conf.fixed_, TYPE).greedy) { + if (pdo->fixed_settings) { hw = glue (audio_pcm_hw_add_new_, TYPE) (as); if (hw) { return hw; @@ -331,9 +333,11 @@ static SW *glue (audio_pcm_create_voice_pair_, TYPE) ( SW *sw; HW *hw; struct audsettings hw_as; + AudioState *s = &glob_audio_state; + AudiodevPerDirectionOptions *pdo = s->dev->TYPE; - if (glue (conf.fixed_, TYPE).enabled) { - hw_as = glue (conf.fixed_, TYPE).settings; + if (pdo->fixed_settings) { + hw_as = audiodev_to_audsettings(pdo); } else { hw_as = *as; @@ -398,6 +402,7 @@ SW *glue (AUD_open_, TYPE) ( ) { AudioState *s = &glob_audio_state; + AudiodevPerDirectionOptions *pdo = s->dev->TYPE; if (audio_bug (AUDIO_FUNC, !card || !name || !callback_fn || !as)) { dolog ("card=%p name=%p callback_fn=%p as=%p\n", @@ -422,7 +427,7 @@ SW *glue (AUD_open_, TYPE) ( return sw; } - if (!glue (conf.fixed_, TYPE).enabled && sw) { + if (!pdo->fixed_settings && sw) { glue (AUD_close_, TYPE) (card, sw); sw = NULL; } diff --git a/audio/coreaudio.c b/audio/coreaudio.c index 6dfd63e..dfa5e79 100644 --- a/audio/coreaudio.c +++ b/audio/coreaudio.c @@ -34,11 +34,6 @@ static int isAtexit; -typedef struct { - int buffer_frames; - int nbuffers; -} CoreaudioConf; - typedef struct coreaudioVoiceOut { HWVoiceOut hw; pthread_mutex_t mutex; @@ -292,7 +287,9 @@ static int coreaudio_init_out(HWVoiceOut *hw, struct audsettings *as, int err; const char *typ = "playback"; AudioValueRange frameRange; - CoreaudioConf *conf = drv_opaque; + Audiodev *dev = drv_opaque; + AudiodevPerDirectionOptions *pdo = dev->out; + int frames; /* create mutex */ err = pthread_mutex_init(&core->mutex, NULL); @@ -334,16 +331,17 @@ static int coreaudio_init_out(HWVoiceOut *hw, struct audsettings *as, return -1; } - if (frameRange.mMinimum > conf->buffer_frames) { + frames = audio_buffer_frames(pdo, as, 11610); + if (frameRange.mMinimum > frames) { core->audioDevicePropertyBufferFrameSize = (UInt32) frameRange.mMinimum; dolog ("warning: Upsizing Buffer Frames to %f\n", frameRange.mMinimum); } - else if (frameRange.mMaximum < conf->buffer_frames) { + else if (frameRange.mMaximum < frames) { core->audioDevicePropertyBufferFrameSize = (UInt32) frameRange.mMaximum; dolog ("warning: Downsizing Buffer Frames to %f\n", frameRange.mMaximum); } else { - core->audioDevicePropertyBufferFrameSize = conf->buffer_frames; + core->audioDevicePropertyBufferFrameSize = frames; } /* set Buffer Frame Size */ @@ -377,7 +375,8 @@ static int coreaudio_init_out(HWVoiceOut *hw, struct audsettings *as, "Could not get device buffer frame size\n"); return -1; } - hw->samples = conf->nbuffers * core->audioDevicePropertyBufferFrameSize; + hw->samples = (pdo->has_buffer_count ? pdo->buffer_count : 4) * + core->audioDevicePropertyBufferFrameSize; /* get StreamFormat */ propertySize = sizeof(core->outputStreamBasicDescription); @@ -497,41 +496,16 @@ static int coreaudio_ctl_out (HWVoiceOut *hw, int cmd, ...) return 0; } -static CoreaudioConf glob_conf = { - .buffer_frames = 512, - .nbuffers = 4, -}; - -static void *coreaudio_audio_init (void) +static void *coreaudio_audio_init(Audiodev *dev) { - CoreaudioConf *conf = g_malloc(sizeof(CoreaudioConf)); - *conf = glob_conf; - atexit(coreaudio_atexit); - return conf; + return dev; } static void coreaudio_audio_fini (void *opaque) { - g_free(opaque); } -static struct audio_option coreaudio_options[] = { - { - .name = "BUFFER_SIZE", - .tag = AUD_OPT_INT, - .valp = &glob_conf.buffer_frames, - .descr = "Size of the buffer in frames" - }, - { - .name = "BUFFER_COUNT", - .tag = AUD_OPT_INT, - .valp = &glob_conf.nbuffers, - .descr = "Number of buffers" - }, - { /* End of list */ } -}; - static struct audio_pcm_ops coreaudio_pcm_ops = { .init_out = coreaudio_init_out, .fini_out = coreaudio_fini_out, @@ -543,7 +517,6 @@ static struct audio_pcm_ops coreaudio_pcm_ops = { struct audio_driver coreaudio_audio_driver = { .name = "coreaudio", .descr = "CoreAudio http://developer.apple.com/audio/coreaudio.html", - .options = coreaudio_options, .init = coreaudio_audio_init, .fini = coreaudio_audio_fini, .pcm_ops = &coreaudio_pcm_ops, diff --git a/audio/dsound_template.h b/audio/dsound_template.h index b439f33..96181ef 100644 --- a/audio/dsound_template.h +++ b/audio/dsound_template.h @@ -167,17 +167,18 @@ static int dsound_init_out(HWVoiceOut *hw, struct audsettings *as, dsound *s = drv_opaque; WAVEFORMATEX wfx; struct audsettings obt_as; - DSoundConf *conf = &s->conf; #ifdef DSBTYPE_IN const char *typ = "ADC"; DSoundVoiceIn *ds = (DSoundVoiceIn *) hw; DSCBUFFERDESC bd; DSCBCAPS bc; + AudiodevPerDirectionOptions *pdo = s->dev->in; #else const char *typ = "DAC"; DSoundVoiceOut *ds = (DSoundVoiceOut *) hw; DSBUFFERDESC bd; DSBCAPS bc; + AudiodevPerDirectionOptions *pdo = s->dev->out; #endif if (!s->FIELD2) { @@ -193,8 +194,8 @@ static int dsound_init_out(HWVoiceOut *hw, struct audsettings *as, memset (&bd, 0, sizeof (bd)); bd.dwSize = sizeof (bd); bd.lpwfxFormat = &wfx; + bd.dwBufferBytes = audio_buffer_bytes(pdo, as, 92880); #ifdef DSBTYPE_IN - bd.dwBufferBytes = conf->bufsize_in; hr = IDirectSoundCapture_CreateCaptureBuffer ( s->dsound_capture, &bd, @@ -203,7 +204,6 @@ static int dsound_init_out(HWVoiceOut *hw, struct audsettings *as, ); #else bd.dwFlags = DSBCAPS_STICKYFOCUS | DSBCAPS_GETCURRENTPOSITION2; - bd.dwBufferBytes = conf->bufsize_out; hr = IDirectSound_CreateSoundBuffer ( s->dsound, &bd, diff --git a/audio/dsoundaudio.c b/audio/dsoundaudio.c index e9472c1..fdbf2e6 100644 --- a/audio/dsoundaudio.c +++ b/audio/dsoundaudio.c @@ -42,16 +42,10 @@ /* #define DEBUG_DSOUND */ typedef struct { - int bufsize_in; - int bufsize_out; - int latency_millis; -} DSoundConf; - -typedef struct { LPDIRECTSOUND dsound; LPDIRECTSOUNDCAPTURE dsound_capture; struct audsettings settings; - DSoundConf conf; + Audiodev *dev; } dsound; typedef struct { @@ -247,9 +241,9 @@ static void GCC_FMT_ATTR (3, 4) dsound_logerr2 ( dsound_log_hresult (hr); } -static DWORD millis_to_bytes (struct audio_pcm_info *info, DWORD millis) +static uint64_t usecs_to_bytes(struct audio_pcm_info *info, uint32_t usecs) { - return (millis * info->bytes_per_second) / 1000; + return muldiv64(usecs, info->bytes_per_second, 1000000); } #ifdef DEBUG_DSOUND @@ -477,7 +471,7 @@ static int dsound_run_out (HWVoiceOut *hw, int live) LPVOID p1, p2; int bufsize; dsound *s = ds->s; - DSoundConf *conf = &s->conf; + AudiodevDsoundOptions *dso = s->dev->opts->dsound; if (!dsb) { dolog ("Attempt to run empty with playback buffer\n"); @@ -500,14 +494,14 @@ static int dsound_run_out (HWVoiceOut *hw, int live) len = live << hwshift; if (ds->first_time) { - if (conf->latency_millis) { + if (dso->latency) { DWORD cur_blat; cur_blat = audio_ring_dist (wpos, ppos, bufsize); ds->first_time = 0; old_pos = wpos; old_pos += - millis_to_bytes (&hw->info, conf->latency_millis) - cur_blat; + usecs_to_bytes(&hw->info, dso->latency) - cur_blat; old_pos %= bufsize; old_pos &= ~hw->info.align; } @@ -746,12 +740,6 @@ static int dsound_run_in (HWVoiceIn *hw) return decr; } -static DSoundConf glob_conf = { - .bufsize_in = 16384, - .bufsize_out = 16384, - .latency_millis = 10 -}; - static void dsound_audio_fini (void *opaque) { HRESULT hr; @@ -782,13 +770,22 @@ static void dsound_audio_fini (void *opaque) g_free(s); } -static void *dsound_audio_init (void) +static void *dsound_audio_init(Audiodev *dev) { int err; HRESULT hr; dsound *s = g_malloc0(sizeof(dsound)); + AudiodevDsoundOptions *dso; + + assert(dev->opts->kind == AUDIODEV_BACKEND_OPTIONS_KIND_DSOUND); + s->dev = dev; + dso = dev->opts->dsound; + + if (!dso->has_latency) { + dso->has_latency = true; + dso->latency = 10000; /* 10 ms */ + } - s->conf = glob_conf; hr = CoInitialize (NULL); if (FAILED (hr)) { dsound_logerr (hr, "Could not initialize COM\n"); @@ -853,28 +850,6 @@ static void *dsound_audio_init (void) return s; } -static struct audio_option dsound_options[] = { - { - .name = "LATENCY_MILLIS", - .tag = AUD_OPT_INT, - .valp = &glob_conf.latency_millis, - .descr = "(undocumented)" - }, - { - .name = "BUFSIZE_OUT", - .tag = AUD_OPT_INT, - .valp = &glob_conf.bufsize_out, - .descr = "(undocumented)" - }, - { - .name = "BUFSIZE_IN", - .tag = AUD_OPT_INT, - .valp = &glob_conf.bufsize_in, - .descr = "(undocumented)" - }, - { /* End of list */ } -}; - static struct audio_pcm_ops dsound_pcm_ops = { .init_out = dsound_init_out, .fini_out = dsound_fini_out, @@ -892,7 +867,6 @@ static struct audio_pcm_ops dsound_pcm_ops = { struct audio_driver dsound_audio_driver = { .name = "dsound", .descr = "DirectSound http://wikipedia.org/wiki/DirectSound", - .options = dsound_options, .init = dsound_audio_init, .fini = dsound_audio_fini, .pcm_ops = &dsound_pcm_ops, diff --git a/audio/noaudio.c b/audio/noaudio.c index 50db1f3..4c94a26 100644 --- a/audio/noaudio.c +++ b/audio/noaudio.c @@ -134,7 +134,7 @@ static int no_ctl_in (HWVoiceIn *hw, int cmd, ...) return 0; } -static void *no_audio_init (void) +static void *no_audio_init (Audiodev *dev) { return &no_audio_init; } @@ -161,7 +161,6 @@ static struct audio_pcm_ops no_pcm_ops = { struct audio_driver no_audio_driver = { .name = "none", .descr = "Timer based audio emulation", - .options = NULL, .init = no_audio_init, .fini = no_audio_fini, .pcm_ops = &no_pcm_ops, diff --git a/audio/ossaudio.c b/audio/ossaudio.c index 02a3a95..3cc04a7 100644 --- a/audio/ossaudio.c +++ b/audio/ossaudio.c @@ -29,6 +29,8 @@ #include "qemu-common.h" #include "qemu/main-loop.h" #include "qemu/host-utils.h" +#include "qapi/alloc-visitor.h" +#include "qapi-visit.h" #include "audio.h" #include "trace.h" @@ -39,16 +41,6 @@ #define USE_DSP_POLICY #endif -typedef struct OSSConf { - int try_mmap; - int nfrags; - int fragsize; - const char *devpath_out; - const char *devpath_in; - int exclusive; - int policy; -} OSSConf; - typedef struct OSSVoiceOut { HWVoiceOut hw; void *pcm_buf; @@ -58,7 +50,7 @@ typedef struct OSSVoiceOut { int fragsize; int mmapped; int pending; - OSSConf *conf; + Audiodev *dev; } OSSVoiceOut; typedef struct OSSVoiceIn { @@ -67,12 +59,12 @@ typedef struct OSSVoiceIn { int fd; int nfrags; int fragsize; - OSSConf *conf; + Audiodev *dev; } OSSVoiceIn; struct oss_params { int freq; - AudioFormat fmt; + int fmt; int nchannels; int nfrags; int fragsize; @@ -264,19 +256,26 @@ static int oss_get_version (int fd, int *version, const char *typ) } #endif -static int oss_open (int in, struct oss_params *req, - struct oss_params *obt, int *pfd, OSSConf* conf) +static int oss_open(int in, struct oss_params *req, audsettings *as, + struct oss_params *obt, int *pfd, Audiodev *dev) { + AudiodevOssOptions *oopts = dev->opts->oss; + AudiodevOssPerDirectionOptions *opdo = in ? oopts->in : oopts->out; + AudiodevPerDirectionOptions *pdo = in ? dev->in : dev->out; int fd; - int oflags = conf->exclusive ? O_EXCL : 0; + int oflags = (oopts->has_exclusive && oopts->exclusive) ? O_EXCL : 0; audio_buf_info abinfo; int fmt, freq, nchannels; int setfragment = 1; - const char *dspname = in ? conf->devpath_in : conf->devpath_out; + const char *dspname = opdo->has_dev ? opdo->dev : "/dev/dsp"; const char *typ = in ? "ADC" : "DAC"; +#ifdef USE_DSP_POLICY + int policy = oopts->has_dsp_policy ? oopts->dsp_policy : 5; +#endif /* Kludge needed to have working mmap on Linux */ - oflags |= conf->try_mmap ? O_RDWR : (in ? O_RDONLY : O_WRONLY); + oflags |= (oopts->has_mmap && oopts->mmap) ? + O_RDWR : (in ? O_RDONLY : O_WRONLY); fd = open (dspname, oflags | O_NONBLOCK); if (-1 == fd) { @@ -287,6 +286,8 @@ static int oss_open (int in, struct oss_params *req, freq = req->freq; nchannels = req->nchannels; fmt = req->fmt; + req->nfrags = pdo->has_buffer_count ? pdo->buffer_count : 4; + req->fragsize = audio_buffer_bytes(pdo, as, 23220); if (ioctl (fd, SNDCTL_DSP_SAMPLESIZE, &fmt)) { oss_logerr2 (errno, typ, "Failed to set sample size %d\n", req->fmt); @@ -310,18 +311,18 @@ static int oss_open (int in, struct oss_params *req, } #ifdef USE_DSP_POLICY - if (conf->policy >= 0) { + if (policy >= 0) { int version; if (!oss_get_version (fd, &version, typ)) { trace_oss_version(version); if (version >= 0x040000) { - int policy = conf->policy; - if (ioctl (fd, SNDCTL_DSP_POLICY, &policy)) { + int policy2 = policy; + if (ioctl (fd, SNDCTL_DSP_POLICY, &policy2)) { oss_logerr2 (errno, typ, "Failed to set timing policy to %d\n", - conf->policy); + policy); goto err; } setfragment = 0; @@ -504,17 +505,16 @@ static int oss_init_out(HWVoiceOut *hw, struct audsettings *as, int fd; AudioFormat effective_fmt; struct audsettings obt_as; - OSSConf *conf = drv_opaque; + Audiodev *dev = drv_opaque; + AudiodevOssOptions *oopts = dev->opts->oss; oss->fd = -1; req.fmt = aud_to_ossfmt (as->fmt, as->endianness); req.freq = as->freq; req.nchannels = as->nchannels; - req.fragsize = conf->fragsize; - req.nfrags = conf->nfrags; - if (oss_open (0, &req, &obt, &fd, conf)) { + if (oss_open(0, &req, as, &obt, &fd, dev)) { return -1; } @@ -541,7 +541,7 @@ static int oss_init_out(HWVoiceOut *hw, struct audsettings *as, hw->samples = (obt.nfrags * obt.fragsize) >> hw->info.shift; oss->mmapped = 0; - if (conf->try_mmap) { + if (oopts->has_mmap && oopts->mmap) { oss->pcm_buf = mmap ( NULL, hw->samples << hw->info.shift, @@ -601,7 +601,7 @@ static int oss_init_out(HWVoiceOut *hw, struct audsettings *as, } oss->fd = fd; - oss->conf = conf; + oss->dev = dev; return 0; } @@ -609,16 +609,12 @@ static int oss_ctl_out (HWVoiceOut *hw, int cmd, ...) { int trig; OSSVoiceOut *oss = (OSSVoiceOut *) hw; + AudiodevOssPerDirectionOptions *opdo = oss->dev->opts->oss->out; switch (cmd) { case VOICE_ENABLE: { - va_list ap; - int poll_mode; - - va_start (ap, cmd); - poll_mode = va_arg (ap, int); - va_end (ap); + bool poll_mode = !opdo->has_try_poll || opdo->try_poll; ldebug ("enabling voice\n"); if (poll_mode) { @@ -673,16 +669,14 @@ static int oss_init_in(HWVoiceIn *hw, struct audsettings *as, void *drv_opaque) int fd; AudioFormat effective_fmt; struct audsettings obt_as; - OSSConf *conf = drv_opaque; + Audiodev *dev = drv_opaque; oss->fd = -1; req.fmt = aud_to_ossfmt (as->fmt, as->endianness); req.freq = as->freq; req.nchannels = as->nchannels; - req.fragsize = conf->fragsize; - req.nfrags = conf->nfrags; - if (oss_open (1, &req, &obt, &fd, conf)) { + if (oss_open(1, &req, as, &obt, &fd, dev)) { return -1; } @@ -716,7 +710,7 @@ static int oss_init_in(HWVoiceIn *hw, struct audsettings *as, void *drv_opaque) } oss->fd = fd; - oss->conf = conf; + oss->dev = dev; return 0; } @@ -807,16 +801,12 @@ static int oss_read (SWVoiceIn *sw, void *buf, int size) static int oss_ctl_in (HWVoiceIn *hw, int cmd, ...) { OSSVoiceIn *oss = (OSSVoiceIn *) hw; + AudiodevOssPerDirectionOptions *opdo = oss->dev->opts->oss->out; switch (cmd) { case VOICE_ENABLE: { - va_list ap; - int poll_mode; - - va_start (ap, cmd); - poll_mode = va_arg (ap, int); - va_end (ap); + bool poll_mode = !opdo->has_try_poll || opdo->try_poll; if (poll_mode) { oss_poll_in (hw); @@ -836,81 +826,25 @@ static int oss_ctl_in (HWVoiceIn *hw, int cmd, ...) return 0; } -static OSSConf glob_conf = { - .try_mmap = 0, - .nfrags = 4, - .fragsize = 4096, - .devpath_out = "/dev/dsp", - .devpath_in = "/dev/dsp", - .exclusive = 0, - .policy = 5 -}; - -static void *oss_audio_init (void) +static void *oss_audio_init(Audiodev *dev) { - OSSConf *conf = g_malloc(sizeof(OSSConf)); - *conf = glob_conf; + AudiodevOssOptions *oopts; + assert(dev->opts->kind == AUDIODEV_BACKEND_OPTIONS_KIND_OSS); - if (access(conf->devpath_in, R_OK | W_OK) < 0 || - access(conf->devpath_out, R_OK | W_OK) < 0) { + oopts = dev->opts->oss; + if (access(oopts->in->has_dev ? oopts->in->dev : "/dev/dsp", + R_OK | W_OK) < 0 || + access(oopts->out->has_dev ? oopts->out->dev : "/dev/dsp", + R_OK | W_OK) < 0) { return NULL; } - return conf; + return dev; } static void oss_audio_fini (void *opaque) { - g_free(opaque); } -static struct audio_option oss_options[] = { - { - .name = "FRAGSIZE", - .tag = AUD_OPT_INT, - .valp = &glob_conf.fragsize, - .descr = "Fragment size in bytes" - }, - { - .name = "NFRAGS", - .tag = AUD_OPT_INT, - .valp = &glob_conf.nfrags, - .descr = "Number of fragments" - }, - { - .name = "MMAP", - .tag = AUD_OPT_BOOL, - .valp = &glob_conf.try_mmap, - .descr = "Try using memory mapped access" - }, - { - .name = "DAC_DEV", - .tag = AUD_OPT_STR, - .valp = &glob_conf.devpath_out, - .descr = "Path to DAC device" - }, - { - .name = "ADC_DEV", - .tag = AUD_OPT_STR, - .valp = &glob_conf.devpath_in, - .descr = "Path to ADC device" - }, - { - .name = "EXCLUSIVE", - .tag = AUD_OPT_BOOL, - .valp = &glob_conf.exclusive, - .descr = "Open device in exclusive mode (vmix wont work)" - }, -#ifdef USE_DSP_POLICY - { - .name = "POLICY", - .tag = AUD_OPT_INT, - .valp = &glob_conf.policy, - .descr = "Set the timing policy of the device, -1 to use fragment mode", - }, -#endif - { /* End of list */ } -}; - static struct audio_pcm_ops oss_pcm_ops = { .init_out = oss_init_out, .fini_out = oss_fini_out, @@ -928,7 +862,6 @@ static struct audio_pcm_ops oss_pcm_ops = { struct audio_driver oss_audio_driver = { .name = "oss", .descr = "OSS http://www.opensound.com", - .options = oss_options, .init = oss_audio_init, .fini = oss_audio_fini, .pcm_ops = &oss_pcm_ops, diff --git a/audio/paaudio.c b/audio/paaudio.c index cfdbdc6..60afb04 100644 --- a/audio/paaudio.c +++ b/audio/paaudio.c @@ -1,6 +1,8 @@ /* public domain */ #include "qemu-common.h" #include "audio.h" +#include "qapi/alloc-visitor.h" +#include "qapi-visit.h" #include <pulse/pulseaudio.h> @@ -9,14 +11,7 @@ #include "audio_pt_int.h" typedef struct { - int samples; - char *server; - char *sink; - char *source; -} PAConf; - -typedef struct { - PAConf conf; + Audiodev *dev; pa_threaded_mainloop *mainloop; pa_context *context; } paaudio; @@ -31,6 +26,7 @@ typedef struct { void *pcm_buf; struct audio_pt pt; paaudio *g; + int samples; } PAVoiceOut; typedef struct { @@ -45,6 +41,7 @@ typedef struct { const void *read_data; size_t read_index, read_length; paaudio *g; + int samples; } PAVoiceIn; static void qpa_audio_fini(void *opaque); @@ -226,7 +223,7 @@ static void *qpa_thread_out (void *arg) } } - decr = to_mix = audio_MIN (pa->live, pa->g->conf.samples >> 2); + decr = to_mix = audio_MIN (pa->live, pa->samples >> 2); rpos = pa->rpos; if (audio_pt_unlock (&pa->pt, AUDIO_FUNC)) { @@ -318,7 +315,7 @@ static void *qpa_thread_in (void *arg) } } - incr = to_grab = audio_MIN (pa->dead, pa->g->conf.samples >> 2); + incr = to_grab = audio_MIN (pa->dead, pa->samples >> 2); wpos = pa->wpos; if (audio_pt_unlock (&pa->pt, AUDIO_FUNC)) { @@ -545,6 +542,7 @@ static int qpa_init_out(HWVoiceOut *hw, struct audsettings *as, struct audsettings obt_as = *as; PAVoiceOut *pa = (PAVoiceOut *) hw; paaudio *g = pa->g = drv_opaque; + AudiodevPaOptions *popts = g->dev->opts->pa; ss.format = audfmt_to_pa (as->fmt, as->endianness); ss.channels = as->nchannels; @@ -565,7 +563,7 @@ static int qpa_init_out(HWVoiceOut *hw, struct audsettings *as, g, "qemu", PA_STREAM_PLAYBACK, - g->conf.sink, + popts->has_sink ? popts->sink : NULL, &ss, NULL, /* channel map */ &ba, /* buffering attributes */ @@ -577,7 +575,8 @@ static int qpa_init_out(HWVoiceOut *hw, struct audsettings *as, } audio_pcm_init_info (&hw->info, &obt_as); - hw->samples = g->conf.samples; + hw->samples = pa->samples = audio_buffer_samples(g->dev->out, &obt_as, + 46440); pa->pcm_buf = audio_calloc (AUDIO_FUNC, hw->samples, 1 << hw->info.shift); pa->rpos = hw->rpos; if (!pa->pcm_buf) { @@ -611,6 +610,7 @@ static int qpa_init_in(HWVoiceIn *hw, struct audsettings *as, void *drv_opaque) struct audsettings obt_as = *as; PAVoiceIn *pa = (PAVoiceIn *) hw; paaudio *g = pa->g = drv_opaque; + AudiodevPaOptions *popts = g->dev->opts->pa; ss.format = audfmt_to_pa (as->fmt, as->endianness); ss.channels = as->nchannels; @@ -622,7 +622,7 @@ static int qpa_init_in(HWVoiceIn *hw, struct audsettings *as, void *drv_opaque) g, "qemu", PA_STREAM_RECORD, - g->conf.source, + popts->has_source ? popts->source : NULL, &ss, NULL, /* channel map */ NULL, /* buffering attributes */ @@ -634,7 +634,8 @@ static int qpa_init_in(HWVoiceIn *hw, struct audsettings *as, void *drv_opaque) } audio_pcm_init_info (&hw->info, &obt_as); - hw->samples = g->conf.samples; + hw->samples = pa->samples = audio_buffer_samples(g->dev->in, &obt_as, + 46440); pa->pcm_buf = audio_calloc (AUDIO_FUNC, hw->samples, 1 << hw->info.shift); pa->wpos = hw->wpos; if (!pa->pcm_buf) { @@ -808,14 +809,19 @@ static int qpa_ctl_in (HWVoiceIn *hw, int cmd, ...) } /* common */ -static PAConf glob_conf = { - .samples = 4096, -}; - -static void *qpa_audio_init (void) +static void *qpa_audio_init(Audiodev *dev) { - paaudio *g = g_malloc(sizeof(paaudio)); - g->conf = glob_conf; + paaudio *g; + AudiodevPaOptions *popts; + const char *server; + + assert(dev->opts->kind == AUDIODEV_BACKEND_OPTIONS_KIND_PA); + + g = g_malloc(sizeof(paaudio)); + popts = dev->opts->pa; + server = popts->has_server ? popts->server : NULL; + + g->dev = dev; g->mainloop = NULL; g->context = NULL; @@ -825,14 +831,14 @@ static void *qpa_audio_init (void) } g->context = pa_context_new (pa_threaded_mainloop_get_api (g->mainloop), - g->conf.server); + server); if (!g->context) { goto fail; } pa_context_set_state_callback (g->context, context_state_cb, g); - if (pa_context_connect (g->context, g->conf.server, 0, NULL) < 0) { + if (pa_context_connect (g->context, server, 0, NULL) < 0) { qpa_logerr (pa_context_errno (g->context), "pa_context_connect() failed\n"); goto fail; @@ -895,34 +901,6 @@ static void qpa_audio_fini (void *opaque) g_free(g); } -struct audio_option qpa_options[] = { - { - .name = "SAMPLES", - .tag = AUD_OPT_INT, - .valp = &glob_conf.samples, - .descr = "buffer size in samples" - }, - { - .name = "SERVER", - .tag = AUD_OPT_STR, - .valp = &glob_conf.server, - .descr = "server address" - }, - { - .name = "SINK", - .tag = AUD_OPT_STR, - .valp = &glob_conf.sink, - .descr = "sink device name" - }, - { - .name = "SOURCE", - .tag = AUD_OPT_STR, - .valp = &glob_conf.source, - .descr = "source device name" - }, - { /* End of list */ } -}; - static struct audio_pcm_ops qpa_pcm_ops = { .init_out = qpa_init_out, .fini_out = qpa_fini_out, @@ -940,7 +918,6 @@ static struct audio_pcm_ops qpa_pcm_ops = { struct audio_driver pa_audio_driver = { .name = "pa", .descr = "http://www.pulseaudio.org/", - .options = qpa_options, .init = qpa_audio_init, .fini = qpa_audio_fini, .pcm_ops = &qpa_pcm_ops, diff --git a/audio/sdlaudio.c b/audio/sdlaudio.c index db0f95a..796238a 100644 --- a/audio/sdlaudio.c +++ b/audio/sdlaudio.c @@ -44,18 +44,13 @@ typedef struct SDLVoiceOut { int decr; } SDLVoiceOut; -static struct { - int nb_samples; -} conf = { - .nb_samples = 1024 -}; - static struct SDLAudioState { int exit; SDL_mutex *mutex; SDL_sem *sem; int initialized; bool driver_created; + Audiodev *dev; } glob_sdl; typedef struct SDLAudioState SDLAudioState; @@ -347,7 +342,7 @@ static int sdl_init_out(HWVoiceOut *hw, struct audsettings *as, req.freq = as->freq; req.format = aud_to_sdlfmt (as->fmt); req.channels = as->nchannels; - req.samples = conf.nb_samples; + req.samples = audio_buffer_samples(s->dev->out, as, 11610); req.callback = sdl_callback; req.userdata = sdl; @@ -391,7 +386,7 @@ static int sdl_ctl_out (HWVoiceOut *hw, int cmd, ...) return 0; } -static void *sdl_audio_init (void) +static void *sdl_audio_init(Audiodev *dev) { SDLAudioState *s = &glob_sdl; if (s->driver_created) { @@ -420,6 +415,7 @@ static void *sdl_audio_init (void) } s->driver_created = true; + s->dev = dev; return s; } @@ -431,18 +427,9 @@ static void sdl_audio_fini (void *opaque) SDL_DestroyMutex (s->mutex); SDL_QuitSubSystem (SDL_INIT_AUDIO); s->driver_created = false; + s->dev = NULL; } -static struct audio_option sdl_options[] = { - { - .name = "SAMPLES", - .tag = AUD_OPT_INT, - .valp = &conf.nb_samples, - .descr = "Size of SDL buffer in samples" - }, - { /* End of list */ } -}; - static struct audio_pcm_ops sdl_pcm_ops = { .init_out = sdl_init_out, .fini_out = sdl_fini_out, @@ -454,7 +441,6 @@ static struct audio_pcm_ops sdl_pcm_ops = { struct audio_driver sdl_audio_driver = { .name = "sdl", .descr = "SDL http://www.libsdl.org", - .options = sdl_options, .init = sdl_audio_init, .fini = sdl_audio_fini, .pcm_ops = &sdl_pcm_ops, diff --git a/audio/spiceaudio.c b/audio/spiceaudio.c index f556b3b..441fbcb 100644 --- a/audio/spiceaudio.c +++ b/audio/spiceaudio.c @@ -74,7 +74,7 @@ static const SpiceRecordInterface record_sif = { .base.minor_version = SPICE_INTERFACE_RECORD_MINOR, }; -static void *spice_audio_init (void) +static void *spice_audio_init(Audiodev *dev) { if (!using_spice) { return NULL; @@ -370,10 +370,6 @@ static int line_in_ctl (HWVoiceIn *hw, int cmd, ...) return 0; } -static struct audio_option audio_options[] = { - { /* end of list */ }, -}; - static struct audio_pcm_ops audio_callbacks = { .init_out = line_out_init, .fini_out = line_out_fini, @@ -391,7 +387,6 @@ static struct audio_pcm_ops audio_callbacks = { struct audio_driver spice_audio_driver = { .name = "spice", .descr = "spice audio driver", - .options = audio_options, .init = spice_audio_init, .fini = spice_audio_fini, .pcm_ops = &audio_callbacks, diff --git a/audio/wavaudio.c b/audio/wavaudio.c index 62017de..fa807b9 100644 --- a/audio/wavaudio.c +++ b/audio/wavaudio.c @@ -23,6 +23,8 @@ */ #include "hw/hw.h" #include "qemu/timer.h" +#include "qapi/alloc-visitor.h" +#include "qapi-visit.h" #include "audio.h" #define AUDIO_CAP "wav" @@ -36,11 +38,6 @@ typedef struct WAVVoiceOut { int total_samples; } WAVVoiceOut; -typedef struct { - struct audsettings settings; - const char *wav_path; -} WAVConf; - static int wav_run_out (HWVoiceOut *hw, int live) { WAVVoiceOut *wav = (WAVVoiceOut *) hw; @@ -111,8 +108,10 @@ static int wav_init_out(HWVoiceOut *hw, struct audsettings *as, 0x02, 0x00, 0x44, 0xac, 0x00, 0x00, 0x10, 0xb1, 0x02, 0x00, 0x04, 0x00, 0x10, 0x00, 0x64, 0x61, 0x74, 0x61, 0x00, 0x00, 0x00, 0x00 }; - WAVConf *conf = drv_opaque; - struct audsettings wav_as = conf->settings; + Audiodev *dev = drv_opaque; + AudiodevWavOptions *wopts = dev->opts->wav; + struct audsettings wav_as = audiodev_to_audsettings(dev->out); + const char *wav_path = wopts->has_path ? wopts->path : "qemu.wav"; stereo = wav_as.nchannels == 2; switch (wav_as.fmt) { @@ -153,10 +152,10 @@ static int wav_init_out(HWVoiceOut *hw, struct audsettings *as, le_store (hdr + 28, hw->info.freq << (bits16 + stereo), 4); le_store (hdr + 32, 1 << (bits16 + stereo), 2); - wav->f = fopen (conf->wav_path, "wb"); + wav->f = fopen(wav_path, "wb"); if (!wav->f) { dolog ("Failed to open wave file `%s'\nReason: %s\n", - conf->wav_path, strerror (errno)); + wav_path, strerror(errno)); g_free (wav->pcm_buf); wav->pcm_buf = NULL; return -1; @@ -224,54 +223,17 @@ static int wav_ctl_out (HWVoiceOut *hw, int cmd, ...) return 0; } -static WAVConf glob_conf = { - .settings.freq = 44100, - .settings.nchannels = 2, - .settings.fmt = AUDIO_FORMAT_S16, - .wav_path = "qemu.wav" -}; - -static void *wav_audio_init (void) +static void *wav_audio_init(Audiodev *dev) { - WAVConf *conf = g_malloc(sizeof(WAVConf)); - *conf = glob_conf; - return conf; + assert(dev->opts->kind == AUDIODEV_BACKEND_OPTIONS_KIND_WAV); + return dev; } static void wav_audio_fini (void *opaque) { ldebug ("wav_fini"); - g_free(opaque); } -static struct audio_option wav_options[] = { - { - .name = "FREQUENCY", - .tag = AUD_OPT_INT, - .valp = &glob_conf.settings.freq, - .descr = "Frequency" - }, - { - .name = "FORMAT", - .tag = AUD_OPT_FMT, - .valp = &glob_conf.settings.fmt, - .descr = "Format" - }, - { - .name = "DAC_FIXED_CHANNELS", - .tag = AUD_OPT_INT, - .valp = &glob_conf.settings.nchannels, - .descr = "Number of channels (1 - mono, 2 - stereo)" - }, - { - .name = "PATH", - .tag = AUD_OPT_STR, - .valp = &glob_conf.wav_path, - .descr = "Path to wave file" - }, - { /* End of list */ } -}; - static struct audio_pcm_ops wav_pcm_ops = { .init_out = wav_init_out, .fini_out = wav_fini_out, @@ -283,7 +245,6 @@ static struct audio_pcm_ops wav_pcm_ops = { struct audio_driver wav_audio_driver = { .name = "wav", .descr = "WAV renderer http://wikipedia.org/wiki/WAV", - .options = wav_options, .init = wav_audio_init, .fini = wav_audio_fini, .pcm_ops = &wav_pcm_ops, diff --git a/qemu-options.hx b/qemu-options.hx index 5438f98..2889ea4 100644 --- a/qemu-options.hx +++ b/qemu-options.hx @@ -312,14 +312,226 @@ The default is @code{en-us}. ETEXI +HXCOMM Deprecated by -audiodev DEF("audio-help", 0, QEMU_OPTION_audio_help, - "-audio-help print list of audio drivers and their options\n", + "-audio-help show -audiodev equivalent of the current audio settings\n", QEMU_ARCH_ALL) STEXI @item -audio-help @findex -audio-help -Will show the audio subsystem help: list of drivers, tunable -parameters. +Will show the -audiodev equivalent of the currently specified +(deprecated) environment variables. +ETEXI + +DEF("audiodev", HAS_ARG, QEMU_OPTION_audiodev, + "-audiodev driver[,prop[=value][,...]]\n" + " specifies the audio backend to use\n" + " id= identifier of the backend\n" + " timer-period= timer period in microseconds\n" + " [in.|out.]fixed-settings= use fixed settings for host audio\n" + " [in.|out.]frequency= frequency to use with fixed settings\n" + " [in.|out.]channels= number of channels to use with fixed settings\n" + " [in.|out.]format= sample format to use with fixed settings\n" + " valid values: s8, s16, s32, u8, u16, u32\n" + " [in.|out.]voices= number of voices to use\n" + " [in.|out.]buffer= size of buffer in microseconds\n" + " [in.|out.]buffer-count= number of buffers\n" + "-audiodev none[,prop[=value][,...]]\n" + " dummy driver that discards all output\n" +#ifdef CONFIG_ALSA + "-audiodev alsa[,prop[=value][,...]]\n" + " [in.|out]dev= name of the audio device to use\n" + " [in.|out]try-poll= attempt to use poll mode\n" + " threshold= threshold (in frames) when playback starts\n" +#endif +#ifdef CONFIG_COREAUDIO + "-audiodev coreaudio[,prop[=value][,...]]\n" +#endif +#ifdef CONFIG_DSOUND + "-audiodev dsound[,prop[=value][,...]]\n" + " latency= add extra latency to playback in microseconds\n" +#endif +#ifdef CONFIG_OSS + "-audiodev oss[,prop[=value][,...]]\n" + " [in.|out]dev= path of the audio device to use\n" + " [in.|out]try-poll= attempt to use poll mode\n" + " mmap= try using memory mapped access\n" + " exclusive= open device in exclusive mode\n" + " dsp-policy= set timing policy, -1 to use fragment mode\n" +#endif +#ifdef CONFIG_PA + "-audiodev pa[,prop[=value][,...]]\n" + " server= PulseAudio server address\n" + " sink= sink device name\n" + " source= source device name\n" +#endif +#ifdef CONFIG_SDL + "-audiodev sdl[,prop[=value][,...]]\n" +#endif +#ifdef CONFIG_SPICE + "-audiodev spice[,prop[=value][,...]]\n" +#endif + "-audiodev wav[,prop[=value][,...]]\n" + " path= path of wav file to record\n", + QEMU_ARCH_ALL) +STEXI +@item -audiodev @var{driver}[,@var{prop}[=@var{value}][,...]] +@findex -audiodev +Adds a new audio backend @var{driver}. There are global and driver +specific properties. Some values can be set differently for input and +output, they're marked with @code{[in.|out.]}. To only set the input's +option prefix the property with @code{in.}, to only set the output +prefix it with @code{out.}, or leave them to specify both. Thus the two +examples are equal: +@example +-audiodev alsa,in.frequency=44110,out.frequency=44110 +-audiodev alsa,frequency=44100 +@end example + +Valid global options are: + +@table @option +@item id=@var{identifier} +Identifies the audio backend. + +@item timer-period=@var{period} +Sets the timer @var{period} used by the audio subsystem in microseconds. + +@item [in.|out.]fixed-settings=on|off +Use fixed settings for host audio. When off, it will change based on +how the guest opens the sound card. + +@item [in.|out.]frequency=@var{frequency} +Specify the @var{frequency} to use when using @var{fixed-settings}. + +@item [in.|out.]channels=@var{channels} +Specify the number of @var{channels} to use when using +@var{fixed-settings}. + +@item [in.|out.]format=@var{format} +Specify the sample @var{format} to use when using @var{fixed-settings}. +Valid values are: @code{s8}, @code{s16}, @code{s32}, @code{u8}, +@code{u16}, @code{u32}. + +@item [in.|out.]voices=@var{voices} +Specify the number of @var{voices} to use. + +@item [in.|out.]buffer=@var{usecs} +Sets the size of the buffer in microseconds. + +@item [in.|out.]buffer-count=@var{count} +Sets the @var{count} of the buffers. + +@end table + +@item -audiodev none[,@var{prop}[=@var{value}][,...]] +Creates a dummy backend that discards all outputs. This backend has no +backend specific properties. + +@item -audiodev alsa[,@var{prop}[=@var{value}][,...]] +Creates backend using the ALSA. This backend is only available on +Linux. + +ALSA specific options are: + +@table @option +@item [in.|out.]dev=@var{device} +Specify the ALSA @var{device} to use for input and/or output. + +@item [in.|out.]try-poll=on|of +Attempt to use poll mode with the device. + +@item threshold=@var{threshold} +Threshold (in frames) when playback starts. + +@end table + +@item -audiodev coreaudio[,@var{prop}[=@var{value}][,...]] +Creates a backend using Apple's Core Audio. This backend is only +available on Mac OS and only supports playback. This backend has no +backend specific properties. + +@item -audiodev dsound[,@var{prop}[=@var{value}][,...]] +Creates a backend using Microsoft's DirectSound. This backend is only +available on Windows and only supports playback. + +Backend specific options are: + +@table @option + +@item latency=@var{usecs} +Add extra @var{usecs} microseconds latency to playback. + +@end table + +@item -audiodev oss[,@var{prop}[=@var{value}][,...]] +Creates a backend using OSS. This backend is available on most +Unix-like systems. + +OSS specific options are: + +@table @option + +@item [in.|out.]dev=@var{path} +Specify the @var{path} of the OSS device to use. + +@item [in.|out.]try-poll=on|of +Attempt to use poll mode with the device. + +@item mmap=on|off +Try using memory mapped device access. + +@item exclusive=on|off +Open the device in exclusive mode (vmix won't work in this case). + +@item dsp-policy=@var{policy} +Sets the timing policy (between 0 and 10). Use -1 to use buffer sizes +specified by @code{buffer} and @code{buffer-count}. This option is +ignored if you do not have OSS 4. + +@end table + +@item -audiodev pa[,@var{prop}[=@var{value}][,...]] +Creates a backend using PulseAudio. This backend is available on most +systems. + +PulseAudio specific options are: + +@table @option + +@item server=@var{server} +Sets the PulseAudio @var{server} to connect to. + +@item sink=@var{sink} +Use the specified @var{sink} for playback. + +@item source=@var{source} +Use the specified @var{source} for recording. + +@end table + +@item -audiodev sdl[,@var{prop}[=@var{value}][,...]] +Creates a backend using SDL. This backend is available on most systems, +but you should use your platform's native backend if possible. This +backend has no backend specific properties. + +@item -audiodev spice[,@var{prop}[=@var{value}][,...]] +Creates a backend that sends audio through SPICE. This backend requires +@code{-spice} and automatically selected in that case, so usually you +can ignore this option. This backend has no backend specific +properties. + +@item -audiodev wav[,@var{prop}[=@var{value}][,...]] +Creates a backend that writes audio to a WAV file. + +Backend specific options are: + +@table @option + +@item path=@var{path} +Write recorded audio into the specified file. + +@end table ETEXI DEF("soundhw", HAS_ARG, QEMU_OPTION_soundhw, diff --git a/vl.c b/vl.c index 2201e27..c3d6ef9 100644 --- a/vl.c +++ b/vl.c @@ -2873,6 +2873,7 @@ int main(int argc, char **argv, char **envp) qemu_add_opts(&qemu_trace_opts); qemu_add_opts(&qemu_option_rom_opts); qemu_add_opts(&qemu_machine_opts); + qemu_add_opts(&qemu_audiodev_opts); qemu_add_opts(&qemu_mem_opts); qemu_add_opts(&qemu_smp_opts); qemu_add_opts(&qemu_boot_opts); @@ -3170,9 +3171,14 @@ int main(int argc, char **argv, char **envp) add_device_config(DEV_BT, optarg); break; case QEMU_OPTION_audio_help: - AUD_help (); + audio_legacy_help(); exit (0); break; + case QEMU_OPTION_audiodev: + if (!qemu_opts_parse(qemu_find_opts("audiodev"), optarg, 1)) { + exit(1); + } + break; case QEMU_OPTION_soundhw: select_soundhw (optarg); break; @@ -4339,6 +4345,7 @@ int main(int argc, char **argv, char **envp) realtime_init(); + audio_set_options(); audio_init(); cpu_synchronize_all_post_init(); -- 2.4.3 ^ permalink raw reply related [flat|nested] 42+ messages in thread
* Re: [Qemu-devel] [PATCH v2 6/6] audio: -audiodev command line option 2015-06-16 12:49 ` [Qemu-devel] [PATCH v2 6/6] audio: -audiodev command line option Kővágó, Zoltán @ 2015-06-17 8:13 ` Markus Armbruster 2015-06-17 11:18 ` Kővágó Zoltán 0 siblings, 1 reply; 42+ messages in thread From: Markus Armbruster @ 2015-06-17 8:13 UTC (permalink / raw) To: Kővágó, Zoltán Cc: Paolo Bonzini, qemu-devel, Gerd Hoffmann "Kővágó, Zoltán" <dirty.ice.hu@gmail.com> writes: > This patch adds an -audiodev command line option, and deprecates the QEMU_* > environment variables for audio backend configuration. It's syntax is similar to > existing options (-netdev, -device, etc): > -audiodev driver_name,property=value,... Sounds really good. Please wrap your commit message lines a bit earlier, around column 70. > Audio drivers now get an Audiodev * as config paramters, instead of the global > audio_option structs. There is some code in audio/audio_legacy.c that converts > the old environment variables to audiodev options (this way backends do not have > to worry about legacy options, also print out them with -audio-help, to ease > migrating to -audiodev). The parenthesis isn't as clear as the rest of your message, probably because it deals with two separate things. Suggest to move out the bit about help into its own paragraph. > Although now it's possible to specify multiple -audiodev options on command > line, multiple audio backends are not supported yet. What happens when I specify multiple -audiodev? How should the command line look like when multiple audio backends are supported? Do we have a clear backward-compatible path from here to there? [...] > 18 files changed, 994 insertions(+), 1083 deletions(-) > create mode 100644 audio/audio_legacy.c Very nice delta, but the size is a bit intimidating :) [...] ^ permalink raw reply [flat|nested] 42+ messages in thread
* Re: [Qemu-devel] [PATCH v2 6/6] audio: -audiodev command line option 2015-06-17 8:13 ` Markus Armbruster @ 2015-06-17 11:18 ` Kővágó Zoltán 2015-06-17 12:27 ` Markus Armbruster 0 siblings, 1 reply; 42+ messages in thread From: Kővágó Zoltán @ 2015-06-17 11:18 UTC (permalink / raw) To: Markus Armbruster; +Cc: Paolo Bonzini, qemu-devel, Gerd Hoffmann 2015-06-17 10:13 keltezéssel, Markus Armbruster írta: > "Kővágó, Zoltán" <dirty.ice.hu@gmail.com> writes: > >> This patch adds an -audiodev command line option, and deprecates the QEMU_* >> environment variables for audio backend configuration. It's syntax is similar to >> existing options (-netdev, -device, etc): >> -audiodev driver_name,property=value,... > > Sounds really good. > > Please wrap your commit message lines a bit earlier, around column 70. > >> Audio drivers now get an Audiodev * as config paramters, instead of the global >> audio_option structs. There is some code in audio/audio_legacy.c that converts >> the old environment variables to audiodev options (this way backends do not have >> to worry about legacy options, also print out them with -audio-help, to ease >> migrating to -audiodev). > > The parenthesis isn't as clear as the rest of your message, probably > because it deals with two separate things. Suggest to move out the bit > about help into its own paragraph. > >> Although now it's possible to specify multiple -audiodev options on command >> line, multiple audio backends are not supported yet. > > What happens when I specify multiple -audiodev? You get an error and qemu terminates. > How should the command line look like when multiple audio backends are > supported? There's an id property of audiodev, so you can identify them: -audiodev alsa,id=foo,... -audiodev pa,id=bar,... and audio devices should get an extra parameter, like audiodev or something like that: -device usb-audio,audiodev=foo -device usb-audio,audiodev=bar And you have two cards, one connected to the alsa device and the other connected to pulseaudio. > > Do we have a clear backward-compatible path from here to there? Currently if you specify an -audiodev option, the environment variables are completely ignored, and it will create an audio backend using the specified options. If you do not provide an -audiodev, it will initialize the audio subsystem using the old environment variables when you add the first sound card (so no -audiodev and no sound device means no audio subsystem, just like the old times). About multiple backends: if the user does not specify the id of the backend when creating the sound card, just use the first -audiodev specified on the command line (or the legacy config, if there's no -audiodev). This way we stay backward-compatible (there won't be multiple -audiodevs in legacy configs). > > [...] >> 18 files changed, 994 insertions(+), 1083 deletions(-) >> create mode 100644 audio/audio_legacy.c > > Very nice delta, but the size is a bit intimidating :) > > [...] > ^ permalink raw reply [flat|nested] 42+ messages in thread
* Re: [Qemu-devel] [PATCH v2 6/6] audio: -audiodev command line option 2015-06-17 11:18 ` Kővágó Zoltán @ 2015-06-17 12:27 ` Markus Armbruster 2015-06-17 13:25 ` Kővágó Zoltán 0 siblings, 1 reply; 42+ messages in thread From: Markus Armbruster @ 2015-06-17 12:27 UTC (permalink / raw) To: Kővágó Zoltán Cc: Paolo Bonzini, qemu-devel, Gerd Hoffmann "Kővágó Zoltán" <dirty.ice.hu@gmail.com> writes: > 2015-06-17 10:13 keltezéssel, Markus Armbruster írta: >> "Kővágó, Zoltán" <dirty.ice.hu@gmail.com> writes: >> >>> This patch adds an -audiodev command line option, and deprecates the QEMU_* >>> environment variables for audio backend configuration. It's syntax >>> is similar to >>> existing options (-netdev, -device, etc): >>> -audiodev driver_name,property=value,... >> >> Sounds really good. >> >> Please wrap your commit message lines a bit earlier, around column 70. >> >>> Audio drivers now get an Audiodev * as config paramters, instead of >>> the global >>> audio_option structs. There is some code in audio/audio_legacy.c >>> that converts >>> the old environment variables to audiodev options (this way >>> backends do not have >>> to worry about legacy options, also print out them with -audio-help, to ease >>> migrating to -audiodev). >> >> The parenthesis isn't as clear as the rest of your message, probably >> because it deals with two separate things. Suggest to move out the bit >> about help into its own paragraph. >> >>> Although now it's possible to specify multiple -audiodev options on command >>> line, multiple audio backends are not supported yet. >> >> What happens when I specify multiple -audiodev? > > You get an error and qemu terminates. Unlikely to create backward compatibility trouble. Works for me. >> How should the command line look like when multiple audio backends are >> supported? > > There's an id property of audiodev, so you can identify them: > -audiodev alsa,id=foo,... -audiodev pa,id=bar,... > and audio devices should get an extra parameter, like audiodev or > something like that: > -device usb-audio,audiodev=foo -device usb-audio,audiodev=bar > And you have two cards, one connected to the alsa device and the other > connected to pulseaudio. Good, because it's consistent with how we connect other kinds of frontends/backends. Multiple audio frontends (device models) can connect to the same audio backend, can't they? Is backend property "id" mandatory? Hmm, judging from PATCH 1 it isn't. It generally is for other kinds of backends, e.g. -netdev, -chardev and (in the future) -blockdev. It's optional with -drive only because -drive is crazy. >> Do we have a clear backward-compatible path from here to there? > > Currently if you specify an -audiodev option, the environment > variables are completely ignored, and it will create an audio backend > using the specified options. If you do not provide an -audiodev, it > will initialize the audio subsystem using the old environment > variables when you add the first sound card (so no -audiodev and no > sound device means no audio subsystem, just like the old times). Should we document this? Where? > About multiple backends: if the user does not specify the id of the > backend when creating the sound card, just use the first -audiodev > specified on the command line (or the legacy config, if there's no > -audiodev). This way we stay backward-compatible (there won't be > multiple -audiodevs in legacy configs). Old way (before your patch): there is only one audio backend, and it's configuration comes from a bunch of environment variables. New way (we're not there yet): you can have multiple audio backends, and you connect frontends to backends using backend ID as usual, i.e. id=ID on the backend, audiodev=ID on the frontend. Required backward compatibility: the old way still works, i.e. when there's no new-way backend, and no frontend has an audiodev=..., then we create a backend configured from the environment. Anything beyond required backward compatibility should be carefully judged on its UI merits. If I understand your description correctly, the plan is to default a frontend's audiodev to the first backend, and if there is none, create one configured from the environment. That's definitely beyond required. Is it a good user interface? [...] ^ permalink raw reply [flat|nested] 42+ messages in thread
* Re: [Qemu-devel] [PATCH v2 6/6] audio: -audiodev command line option 2015-06-17 12:27 ` Markus Armbruster @ 2015-06-17 13:25 ` Kővágó Zoltán 2015-06-17 16:13 ` Markus Armbruster 0 siblings, 1 reply; 42+ messages in thread From: Kővágó Zoltán @ 2015-06-17 13:25 UTC (permalink / raw) To: Markus Armbruster; +Cc: Paolo Bonzini, qemu-devel, Gerd Hoffmann 2015-06-17 14:27 keltezéssel, Markus Armbruster írta: > "Kővágó Zoltán" <dirty.ice.hu@gmail.com> writes: > >> 2015-06-17 10:13 keltezéssel, Markus Armbruster írta: >>> "Kővágó, Zoltán" <dirty.ice.hu@gmail.com> writes: >>> >>>> This patch adds an -audiodev command line option, and deprecates the QEMU_* >>>> environment variables for audio backend configuration. It's syntax >>>> is similar to >>>> existing options (-netdev, -device, etc): >>>> -audiodev driver_name,property=value,... >>> >>> Sounds really good. >>> >>> Please wrap your commit message lines a bit earlier, around column 70. >>> >>>> Audio drivers now get an Audiodev * as config paramters, instead of >>>> the global >>>> audio_option structs. There is some code in audio/audio_legacy.c >>>> that converts >>>> the old environment variables to audiodev options (this way >>>> backends do not have >>>> to worry about legacy options, also print out them with -audio-help, to ease >>>> migrating to -audiodev). >>> >>> The parenthesis isn't as clear as the rest of your message, probably >>> because it deals with two separate things. Suggest to move out the bit >>> about help into its own paragraph. >>> >>>> Although now it's possible to specify multiple -audiodev options on command >>>> line, multiple audio backends are not supported yet. >>> >>> What happens when I specify multiple -audiodev? >> >> You get an error and qemu terminates. > > Unlikely to create backward compatibility trouble. Works for me. > >>> How should the command line look like when multiple audio backends are >>> supported? >> >> There's an id property of audiodev, so you can identify them: >> -audiodev alsa,id=foo,... -audiodev pa,id=bar,... >> and audio devices should get an extra parameter, like audiodev or >> something like that: >> -device usb-audio,audiodev=foo -device usb-audio,audiodev=bar >> And you have two cards, one connected to the alsa device and the other >> connected to pulseaudio. > > Good, because it's consistent with how we connect other kinds of > frontends/backends. > > Multiple audio frontends (device models) can connect to the same audio > backend, can't they? Yes. (It's mandatory, since currently all audio frontends connect to a single backend.) > Is backend property "id" mandatory? Hmm, judging from PATCH 1 it isn't. > It generally is for other kinds of backends, e.g. -netdev, -chardev and > (in the future) -blockdev. It's optional with -drive only because > -drive is crazy. It's optional right now (because I thought specifying an id when you have a single backend is pretty pointless), but it's probably better then if I change it to mandatory. > >>> Do we have a clear backward-compatible path from here to there? >> >> Currently if you specify an -audiodev option, the environment >> variables are completely ignored, and it will create an audio backend >> using the specified options. If you do not provide an -audiodev, it >> will initialize the audio subsystem using the old environment >> variables when you add the first sound card (so no -audiodev and no >> sound device means no audio subsystem, just like the old times). > > Should we document this? Where? Maybe adding a phrase to -audiodev documentation like "If you specify this option, the deprecated environment variables will be ignored". >> About multiple backends: if the user does not specify the id of the >> backend when creating the sound card, just use the first -audiodev >> specified on the command line (or the legacy config, if there's no >> -audiodev). This way we stay backward-compatible (there won't be >> multiple -audiodevs in legacy configs). > > Old way (before your patch): there is only one audio backend, and it's > configuration comes from a bunch of environment variables. > > New way (we're not there yet): you can have multiple audio backends, and > you connect frontends to backends using backend ID as usual, i.e. id=ID > on the backend, audiodev=ID on the frontend. > > Required backward compatibility: the old way still works, i.e. when > there's no new-way backend, and no frontend has an audiodev=..., then we > create a backend configured from the environment. > > Anything beyond required backward compatibility should be carefully > judged on its UI merits. > > If I understand your description correctly, the plan is to default a > frontend's audiodev to the first backend, and if there is none, create > one configured from the environment. That's definitely beyond required. > Is it a good user interface? The "create one configured from the environment" is needed for backward compatibility. Specifying an audiodev=... on frontend makes no sense without an -audiodev, so that's not allowed. The only question is what happens with an -audiodev, but no audiodev= on the frontend. We can't make audiodev= required because that would break compatibility. So we either a) make audiodev= required but only when there's an -audiodev, or b) try to find a sensible default when audiodev= is not specified. Imho b is better, because having a parameter that's sometimes required, sometimes forbidden is a bit confusing. ^ permalink raw reply [flat|nested] 42+ messages in thread
* Re: [Qemu-devel] [PATCH v2 6/6] audio: -audiodev command line option 2015-06-17 13:25 ` Kővágó Zoltán @ 2015-06-17 16:13 ` Markus Armbruster 2015-06-18 6:54 ` Gerd Hoffmann 0 siblings, 1 reply; 42+ messages in thread From: Markus Armbruster @ 2015-06-17 16:13 UTC (permalink / raw) To: Kővágó Zoltán Cc: Paolo Bonzini, qemu-devel, Gerd Hoffmann "Kővágó Zoltán" <dirty.ice.hu@gmail.com> writes: > 2015-06-17 14:27 keltezéssel, Markus Armbruster írta: >> "Kővágó Zoltán" <dirty.ice.hu@gmail.com> writes: >> >>> 2015-06-17 10:13 keltezéssel, Markus Armbruster írta: >>>> "Kővágó, Zoltán" <dirty.ice.hu@gmail.com> writes: >>>> >>>>> This patch adds an -audiodev command line option, and deprecates the QEMU_* >>>>> environment variables for audio backend configuration. It's syntax >>>>> is similar to >>>>> existing options (-netdev, -device, etc): >>>>> -audiodev driver_name,property=value,... >>>> >>>> Sounds really good. >>>> >>>> Please wrap your commit message lines a bit earlier, around column 70. >>>> >>>>> Audio drivers now get an Audiodev * as config paramters, instead of >>>>> the global >>>>> audio_option structs. There is some code in audio/audio_legacy.c >>>>> that converts >>>>> the old environment variables to audiodev options (this way >>>>> backends do not have >>>>> to worry about legacy options, also print out them with >>>>> -audio-help, to ease >>>>> migrating to -audiodev). >>>> >>>> The parenthesis isn't as clear as the rest of your message, probably >>>> because it deals with two separate things. Suggest to move out the bit >>>> about help into its own paragraph. >>>> >>>>> Although now it's possible to specify multiple -audiodev options on command >>>>> line, multiple audio backends are not supported yet. >>>> >>>> What happens when I specify multiple -audiodev? >>> >>> You get an error and qemu terminates. >> >> Unlikely to create backward compatibility trouble. Works for me. >> >>>> How should the command line look like when multiple audio backends are >>>> supported? >>> >>> There's an id property of audiodev, so you can identify them: >>> -audiodev alsa,id=foo,... -audiodev pa,id=bar,... >>> and audio devices should get an extra parameter, like audiodev or >>> something like that: >>> -device usb-audio,audiodev=foo -device usb-audio,audiodev=bar >>> And you have two cards, one connected to the alsa device and the other >>> connected to pulseaudio. >> >> Good, because it's consistent with how we connect other kinds of >> frontends/backends. >> >> Multiple audio frontends (device models) can connect to the same audio >> backend, can't they? > > Yes. (It's mandatory, since currently all audio frontends connect to a > single backend.) Understood. >> Is backend property "id" mandatory? Hmm, judging from PATCH 1 it isn't. >> It generally is for other kinds of backends, e.g. -netdev, -chardev and >> (in the future) -blockdev. It's optional with -drive only because >> -drive is crazy. > It's optional right now (because I thought specifying an id when you > have a single backend is pretty pointless), but it's probably better > then if I change it to mandatory. Let's make it mandatory for now. We can always change mandatory to optional, but not the other way round. >>>> Do we have a clear backward-compatible path from here to there? >>> >>> Currently if you specify an -audiodev option, the environment >>> variables are completely ignored, and it will create an audio backend >>> using the specified options. If you do not provide an -audiodev, it >>> will initialize the audio subsystem using the old environment >>> variables when you add the first sound card (so no -audiodev and no >>> sound device means no audio subsystem, just like the old times). >> >> Should we document this? Where? > > Maybe adding a phrase to -audiodev documentation like "If you specify > this option, the deprecated environment variables will be ignored". Works for me. >>> About multiple backends: if the user does not specify the id of the >>> backend when creating the sound card, just use the first -audiodev >>> specified on the command line (or the legacy config, if there's no >>> -audiodev). This way we stay backward-compatible (there won't be >>> multiple -audiodevs in legacy configs). >> >> Old way (before your patch): there is only one audio backend, and it's >> configuration comes from a bunch of environment variables. >> >> New way (we're not there yet): you can have multiple audio backends, and >> you connect frontends to backends using backend ID as usual, i.e. id=ID >> on the backend, audiodev=ID on the frontend. >> >> Required backward compatibility: the old way still works, i.e. when >> there's no new-way backend, and no frontend has an audiodev=..., then we >> create a backend configured from the environment. >> >> Anything beyond required backward compatibility should be carefully >> judged on its UI merits. >> >> If I understand your description correctly, the plan is to default a >> frontend's audiodev to the first backend, and if there is none, create >> one configured from the environment. That's definitely beyond required. >> Is it a good user interface? > > The "create one configured from the environment" is needed for > backward compatibility. Specifying an audiodev=... on frontend makes > no sense without an -audiodev, so that's not allowed. The only > question is what happens with an -audiodev, but no audiodev= on the > frontend. We can't make audiodev= required because that would break > compatibility. So we either a) make audiodev= required but only when > there's an -audiodev, or b) try to find a sensible default when > audiodev= is not specified. Imho b is better, because having a > parameter that's sometimes required, sometimes forbidden is a bit > confusing. If we didn't have to do backwards compatibility, we'd make audiodev mandatory, wouldn't we? Here's how I'd add back-compat to that baseline. audiodev becomes optional, but omitting it is deprecated. If you do, you implicitly use the legacy backend that takes its configuration from the environment. The legacy backend gets created when a frontend is using it. If you think you got a better way to do it, the person to convince is Gerd. I'll try to find time to experiment with flattening your QAPI schema, perhaps we can make things a bit simpler. I'm glad you're taking care of turning our bizarro audio UI into something that actually fits in with the rest. Much appreciated! ^ permalink raw reply [flat|nested] 42+ messages in thread
* Re: [Qemu-devel] [PATCH v2 6/6] audio: -audiodev command line option 2015-06-17 16:13 ` Markus Armbruster @ 2015-06-18 6:54 ` Gerd Hoffmann 0 siblings, 0 replies; 42+ messages in thread From: Gerd Hoffmann @ 2015-06-18 6:54 UTC (permalink / raw) To: Markus Armbruster Cc: Paolo Bonzini, qemu-devel, Kővágó Zoltán Hi, > Here's how I'd add back-compat to that baseline. audiodev becomes > optional, but omitting it is deprecated. If you do, you implicitly use > the legacy backend that takes its configuration from the environment. > The legacy backend gets created when a frontend is using it. > > If you think you got a better way to do it, the person to convince is > Gerd. Approach makes sense to me. Either the old config scheme is active, for compatibility, or the new one. Allowing to mix old+new (use -audiodev, but don't specify the name to -device soundcard,audiodev=) is not needed. cheers, Gerd ^ permalink raw reply [flat|nested] 42+ messages in thread
end of thread, other threads:[~2015-06-18 8:51 UTC | newest] Thread overview: 42+ messages (download: mbox.gz follow: Atom feed -- links below jump to the message on this page -- 2015-06-16 12:49 [Qemu-devel] [PATCH v2 0/6] -audiodev option Kővágó, Zoltán 2015-06-16 12:49 ` [Qemu-devel] [PATCH v2 1/6] qapi: qapi for audio backends Kővágó, Zoltán 2015-06-17 7:46 ` Markus Armbruster 2015-06-17 10:54 ` Kővágó Zoltán 2015-06-17 11:48 ` Markus Armbruster 2015-06-17 12:07 ` Kővágó Zoltán 2015-06-17 13:37 ` Markus Armbruster 2015-06-17 13:53 ` Kővágó Zoltán 2015-06-17 16:06 ` Markus Armbruster 2015-06-18 0:21 ` Kővágó Zoltán 2015-06-18 8:51 ` Markus Armbruster 2015-06-17 15:50 ` Eric Blake 2015-06-16 12:49 ` [Qemu-devel] [PATCH v2 2/6] qapi: support nested structs in OptsVisitor Kővágó, Zoltán 2015-06-17 7:50 ` Markus Armbruster 2015-06-17 8:41 ` Gerd Hoffmann 2015-06-17 11:01 ` Kővágó Zoltán 2015-06-17 11:50 ` Markus Armbruster 2015-06-17 15:47 ` Eric Blake 2015-06-17 11:18 ` Markus Armbruster 2015-06-17 12:11 ` Kővágó Zoltán 2015-06-17 13:41 ` Markus Armbruster 2015-06-17 14:02 ` Kővágó Zoltán 2015-06-17 16:10 ` Markus Armbruster 2015-06-16 12:49 ` [Qemu-devel] [PATCH v2 3/6] opts: do not print separator before first item in qemu_opts_print Kővágó, Zoltán 2015-06-17 7:53 ` Markus Armbruster 2015-06-17 9:02 ` Kevin Wolf 2015-06-16 12:49 ` [Qemu-devel] [PATCH v2 4/6] qapi: AllocVisitor Kővágó, Zoltán 2015-06-17 7:56 ` Markus Armbruster 2015-06-17 12:01 ` Kővágó Zoltán 2015-06-17 13:42 ` Markus Armbruster 2015-06-16 12:49 ` [Qemu-devel] [PATCH v2 5/6] audio: use qapi AudioFormat instead of audfmt_e Kővágó, Zoltán 2015-06-17 8:01 ` Markus Armbruster 2015-06-17 11:05 ` Kővágó Zoltán 2015-06-17 11:51 ` Markus Armbruster 2015-06-17 16:01 ` Eric Blake 2015-06-16 12:49 ` [Qemu-devel] [PATCH v2 6/6] audio: -audiodev command line option Kővágó, Zoltán 2015-06-17 8:13 ` Markus Armbruster 2015-06-17 11:18 ` Kővágó Zoltán 2015-06-17 12:27 ` Markus Armbruster 2015-06-17 13:25 ` Kővágó Zoltán 2015-06-17 16:13 ` Markus Armbruster 2015-06-18 6:54 ` Gerd Hoffmann
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for NNTP newsgroup(s).