Re: [pulseaudio-discuss] [PATCH v2 1/6] gccmacro: Disable printf-like format checking on mingw32 compilers.
On Tue, 2012-08-21 at 13:32 +0200, Thomas Martitz wrote: Am 21.08.2012 08:51, schrieb Rémi Denis-Courmont: Le mardi 21 août 2012 00:50:34 Thomas Martitz, vous avez écrit : There are tons of warnings, most of them because the function is not recognized as printf-like. Removing checks looks very fishy. To use C99 and/or GNU format specifiers on MingW, you need to use the gnuprintf attribute instead of printf. With printf, the format string is validated according to the antiquated MSVC rules. Interesting, I didn't know about gnuprintf. FWIW, what are those antiquated MSVC rules? I assumed the return value which isn't int for some affected functions? Is this one going to be respun? -- Arun ___ pulseaudio-discuss mailing list pulseaudio-discuss@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/pulseaudio-discuss
Re: [pulseaudio-discuss] [PATCH v2 3/6] Proper poll() emulation to fix pacat and friends on windows.
On Mon, 2012-08-20 at 23:50 +0200, Thomas Martitz wrote: Currently, Windows versions of pacat and friends fail because the current poll emulation is not sufficient (it only works for socket fds). Luckily Gnulib has a much better emulation that seems to work perfect all over. The implementation has been largely copied (except a bug fix regarding timeout handling, to be pushed upstream) and works on pipes and files as well. This fixes the Assertion (!e-dead) failed error coming and lets pacat and friends stream happily to/from a server (I didn't actually test parec). --- The changes look okay to me, but the gnulib code is GPL -- any chance that the author would agree to a dual-license for this? -- Arun ___ pulseaudio-discuss mailing list pulseaudio-discuss@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/pulseaudio-discuss
Re: [pulseaudio-discuss] [PATCH v2 0/6] Fixes for pulseaudio on Windows.
On Mon, 2012-08-20 at 23:50 +0200, Thomas Martitz wrote: I've been tryng to run pulseaudio (including utils) on windows. This series is the first to contribute fixes to achieve this. One issue I came across is the assertion failue (!e-dead) in pacat. This is fixed by the third patch. Uncovered by this and fixed by patch 2 is that stdio needs to be switched to binary mode in order to pipe data through it. Patch 1 works around massive build warnings on mingw32, but it's not mission critical (note: I don't know why the patch order of 1-3 is different to the first set). Another issue was pa_read() failure in flush(), which wasn't actually failure, but rather not handling non-blocking sockets. Actually, I'm not sure if the fix is correct (perhaps because pa_read() should not wait for non-block sockets or the fix is windows-only) but it functions as a work- around. The last 2 patches are more or less minor and speak for themselves. Please review and comment on my patches, especually patch#4. This is my first contribution to PA, but surely not the last one so I would like to get things right. Sorry about an awful wait for your first patches. I've commented on a few of them, and I'm pushing out the rest now. There are minor cosmetic changes to some of them (and I've not bothered rewrapping them at 80 columns, but that would be nice). Thanks for the work! Regards, Arun ___ pulseaudio-discuss mailing list pulseaudio-discuss@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/pulseaudio-discuss
Re: [pulseaudio-discuss] [PATCH] modules: Fix false warning in webrtc AEC
On Mon, 2012-10-22 at 23:48 +0200, Peter Meerwald wrote: From: Peter Meerwald p.meerw...@bct-electronic.com CXX libwebrtc_util_la-webrtc.lo modules/echo-cancel/webrtc.cc: In function 'pa_bool_t pa_webrtc_ec_init(pa_core*, pa_echo_canceller*, pa_sample_spec*, pa_channel_map*, pa_sample_spec*, pa_channel_map*, uint32_t*, const char*)': modules/echo-cancel/webrtc.cc:196:9: warning: 'rm' may be used uninitialized in this function [-Wuninitialized] Signed-off-by: Peter Meerwald p.meerw...@bct-electronic.com Thanks, applied. (Changed the modules label to echo-cancel.) -- Tanu ___ pulseaudio-discuss mailing list pulseaudio-discuss@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/pulseaudio-discuss
Re: [pulseaudio-discuss] [PATCH v0 01/11] card: Support adding profiles dynamically
On 10/23/2012 08:01 AM, Mikel Astiz wrote: Hi David, On Mon, Oct 22, 2012 at 6:10 PM, David Henningsson david.hennings...@canonical.com wrote: On 10/22/2012 10:46 AM, Mikel Astiz wrote: From: Mikel Astiz mikel.as...@bmw-carit.de Some cards might need to add profiles during their lifetime, that is, after the card has been created. How are volume control UIs, other modules, etc made aware that the profiles or ports of a card has changed? I believe the D-Bus API as well as the subscription events would have to be extended for this purpose, but I haven't addressed either of these here. Any thoughts? Just that being able to change a card in ways that was not possible before, is something that require careful thought. Removal even more so; what if a volume control UI starts (that was built before we made this change), and suddenly a profile is removed, and then the user tries to select this profile? I'm also not sure if there are other parts of the PulseAudio core that needs to know if things like these change, do you know? I'm not opposed to the change, maybe it is inevitable - e g, the capabilities of HDMI sinks can also change during PulseAudio's life time, so that's another example of dynamic profiles. What I'm trying to say here is that: 1) This is a quite fundamental change of behaviour, have we thought this through thoroughly, with all the consequences not only for us but for our client software as well? and 2) I don't want to release with this being half-done; e g, if we add profiles without a subscription mechanism and then release, we might end up with volume control UIs polling us all the time to work around our own shortcomings. -- David Henningsson, Canonical Ltd. https://launchpad.net/~diwic ___ pulseaudio-discuss mailing list pulseaudio-discuss@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/pulseaudio-discuss
Re: [pulseaudio-discuss] directory to place the 90-pulseaudio.rule and other profile configure files
On Mon, 2012-10-22 at 09:46 -0700, Dongjiu Geng wrote: Hi David: The target device is a embedded device mobile phone. not computer. Thanks (Please don't top-post.) I don't see how the form factor is relevant here. You asked which directory to place the files, and David answered that. Were you somehow unhappy with the answer? -- Tanu ___ pulseaudio-discuss mailing list pulseaudio-discuss@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/pulseaudio-discuss
Re: [pulseaudio-discuss] [PATCH v0 09/11] bluetooth: Refactor code to create card profiles
On Mon, 2012-10-22 at 10:46 +0200, Mikel Astiz wrote: From: Mikel Astiz mikel.as...@bmw-carit.de Use a helper function to create the card profiles, given an UUID. Thanks, applied. -- Tanu ___ pulseaudio-discuss mailing list pulseaudio-discuss@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/pulseaudio-discuss
Re: [pulseaudio-discuss] [PATCH v0 10/11] bluetooth: Refactor code to create profile ports
On Mon, 2012-10-22 at 10:46 +0200, Mikel Astiz wrote: From: Mikel Astiz mikel.as...@bmw-carit.de Put the created ports in a hashmap, but not necessarily inside a struct pa_card_new_data. Thanks, applied. -- Tanu ___ pulseaudio-discuss mailing list pulseaudio-discuss@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/pulseaudio-discuss
Re: [pulseaudio-discuss] [pulseaudio] error: Sufficiently new version of libasound not found.
On Mon, 2012-10-22 at 00:12 -0700, Dongjiu Geng wrote: HI all: when I build the pulseaudio with alsa-lib, the error log is checking for libasound headers version = 1.0.24... not present. configure: error: Sufficiently new version of libasound not found. could you help me about this bug? Thank you very much. Either your alsa-lib version is too old in the build environment (you're doing cross-compilation, right?) or the alsa-lib pkg-config file doesn't exist where pkg-config is looking for it. The pkg-config file for alsa-lib has name alsa.pc. Check that the file exists in the right directory. On my system it's at /usr/lib/x86_64-linux-gnu/pkgconfig/alsa.pc. If I would have compiled and installed alsa-lib myself from source, it would probably be at /usr/local/lib/pkgconfig/alsa.pc. If you're cross-compiling, the right directory may be something else. -- Tanu ___ pulseaudio-discuss mailing list pulseaudio-discuss@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/pulseaudio-discuss
Re: [pulseaudio-discuss] pulseaudio stops to work suddenly
2012/10/23 Paul Menzel paulepan...@users.sourceforge.net Dear Antonio, please just send plain text messages and no HTML to lists [1]. OpenSUSE mailing list ? I did not know of sending mail to this list. Am Montag, den 22.10.2012, 19:25 +0200 schrieb Antonio Trande: 2012/10/19 Antonio Trande anto.tra...@gmail.com […] 2012/10/18 Tanu Kaskinen ta...@iki.fi On Thu, 2012-10-18 at 20:07 +0200, Antonio Trande wrote: No sound after reboot. There's some other problem then too. The pulseaudio log would be useful. $ pulseaudio - snip E: [pulseaudio] pid.c: Daemon already running. E: [pulseaudio] main.c: pa_pid_file_create() failed. That's not very useful, though. You need to ensure that there's no daemon running before you try to start pulseaudio in a terminal. Killing the old daemon instance doesn't necessary help, due to autospawning. Therefore, do this first to disable autospawning: echo autospawn = no ~/.pulse/client.conf Ok. If I start Fedora with 'cat /etc/modprobe.d/dist-alsa.conf options snd-hda-intel model=auto', sound is present for few seconds. Then, it is no more udible. In attached you can see the 'pulseaudio -' log (this log doesn't stop, I must do a Ctrl-C) using the gnome-shell audio settings menu (choose of the sound hardware, sound tests, ...). […] I forgot ... If I plug an headphone, audio can be heared always. Resolved with 'options snd-hda-intel model=acer-aspire' on '/etc/modprobe.d/dist-alsa.conf' file. Thank you for the follow up. I still do not understand how PulseAudio can work for some though and then stops. Anyway, could you please run the Skript `alsa-info.sh` [2] on your system and report your issue and solution to the alsa-devel mailing list [3] or create a ticket in the Linux Kernel bug tracker [4] or in the bug tracker of your distribution. I wil do it. Thank you. -- *Antonio Trande Fedora Ambassador* *Fedora italian translation group* *Blogger **mail*: mailto:sagit...@fedoraproject.org sagit...@fedoraproject.org *Homepage*: http://www.fedora-os.org *Sip Address* : sip:sagitter AT ekiga.net *Jabber http://jabber.org/* :sagitter AT jabber.org *GPG Key: **D400D6C4* ___ pulseaudio-discuss mailing list pulseaudio-discuss@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/pulseaudio-discuss
Re: [pulseaudio-discuss] [PATCH v2 3/6] Proper poll() emulation to fix pacat and friends on windows.
Am 23.10.2012 08:40, schrieb Arun Raghavan: On Mon, 2012-08-20 at 23:50 +0200, Thomas Martitz wrote: Currently, Windows versions of pacat and friends fail because the current poll emulation is not sufficient (it only works for socket fds). Luckily Gnulib has a much better emulation that seems to work perfect all over. The implementation has been largely copied (except a bug fix regarding timeout handling, to be pushed upstream) and works on pipes and files as well. This fixes the Assertion (!e-dead) failed error coming and lets pacat and friends stream happily to/from a server (I didn't actually test parec). --- The changes look okay to me, but the gnulib code is GPL -- any chance that the author would agree to a dual-license for this? Actually, I took the copy from git[1], which has it under GPLv2 or later. [1] https://github.com/git/git/blob/master/compat/poll/poll.c Best regards. ___ pulseaudio-discuss mailing list pulseaudio-discuss@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/pulseaudio-discuss
Re: [pulseaudio-discuss] Brrr ... freeze imminent
On Fri, 2012-10-19 at 19:26 +0300, Tanu Kaskinen wrote:
On Fri, 2012-10-19 at 19:24 +0530, Arun Raghavan wrote:
Hello,
I'm hoping to freeze master for 3.0 early next week, with a release in
2-3 weeks after that if everything is quiet. Mostly, it's just pending
on me finishing with Peter's NEON patches.
Thanks, I've been meaning to bring up the release discussion again...
If there's anything that
needs to go into this release that is in your sekrit branches or on the
list and not been given attention, now's the time to holler!
Here's a list of fixes from me that have not been reviewed yet,
according to my bookkeeping. I'd like to get them in 3.0. There
shouldn't be anything particularly complicated. The earliest patches are
from March, so I'm not sure if they all apply cleanly anymore... If
there are patches that don't apply cleanly, complain about it and I'll
resend rebased versions.
bluetooth: Don't free read_smoother in pa__done().
bluetooth: Fix uninitialized variable.
rtp: Get rid of an unitialized value warning from Coverity.
pasuspender: Check pa_context_connect() return value.
Fix pa_parse_boolean() return value checking.
rtsp-client: Fix rtp_port reading.
smoother-test: Fix array overflow.
mainloop-signal: Explicitly ignore pa_write() return value.
mainloop: Change wakeup_requested type from pa_bool_t to pa_atomic_t.
mainloop: Remove useless pa_mainloop_wakeup() calls.
mainloop: Remove redundant wakeup_pipe validity checks.
mainloop: Write to the wakeup pipe unconditionally when waking up the
mainloop.
mainloop: Check pa_write() return value.
oss: Check pa_read() return value.
cpulimit: Explicitly ignore pa_read() return value.
daemon: Don't rely on prctl(PR_SET_KEEPCAPS, 0) for dropping caps.
device-restore: Fix empty argument list declaration.
memblockq: Use pa_xnew0() to avoid manual zeroing.
memblockq: Fix the order of setting minreq and prebuf.
modargs: Don't fail needlessly in
pa_modargs_get_sample_spec_and_channel_map().
sink, source: New functions: pa_{sink,source}_new_data_add_port().
core-util: Don't accept random words in pa_parse_boolean().
pasuspender: Resume before exiting in case of SIGINT or fork() failure.
object: Initialize allocated memory to zero.
card: Store a pa_card pointer in pa_card_profile.
switch-on-port-available: Prepare for port-is_{out, in}put going away.
alsa-mixer: Introduce description-key option for paths.
alsa-mixer: Remove the name option from the General section of path
configuration files.
device-port: Make it impossible to have a dual-direction port.
resampler: Make sure that there are no overflows when multiplying potentially
big numbers.
loopback: Use the real sample spec once it's known.
loopback: Don't fix the source output format/rate/channels.
virtual-surround-sink: Fix setting max_request and max_rewind.
sink: Process rewind requests also when suspended.
combine: Keep the timer active in the null mode only when running.
Propagate the PA_{SINK, SOURCE}_LATENCY flag changes through a chain of
filter sinks/sources.
Propagate the PA_{SINK, SOURCE}_DYNAMIC_LATENCY flag changes through a chain
of filter sinks/sources.
sink, source: Really set the fixed latency in
set_fixed_latency_within_thread(), always.
build: Don't enable BlueZ if libbluetooth is not found.
match: Use the SINK_INPUT_FIXATE hook instead of NEW.
build: Add PROTOCOL to EXTRA_DIST.
build: Add a2dp-codecs.h to libbluetooth-util sources.
build: Add .gitignore files to EXTRA_DIST.
:) Let's try to work through the list and get as much done as possible.
Are we going to make a pavucontrol release too? Here's one patch for
pavucontrol:
Enable Automake's silent-rules option.
If that's the only one, I'm not to keen on a new release.
-- Arun
___
pulseaudio-discuss mailing list
pulseaudio-discuss@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/pulseaudio-discuss
Re: [pulseaudio-discuss] [PATCH v2 1/6] gccmacro: Disable printf-like format checking on mingw32 compilers.
Am 23.10.2012 08:25, schrieb Arun Raghavan: On Tue, 2012-08-21 at 13:32 +0200, Thomas Martitz wrote: Am 21.08.2012 08:51, schrieb Rémi Denis-Courmont: Le mardi 21 août 2012 00:50:34 Thomas Martitz, vous avez écrit : There are tons of warnings, most of them because the function is not recognized as printf-like. Removing checks looks very fishy. To use C99 and/or GNU format specifiers on MingW, you need to use the gnuprintf attribute instead of printf. With printf, the format string is validated according to the antiquated MSVC rules. Interesting, I didn't know about gnuprintf. FWIW, what are those antiquated MSVC rules? I assumed the return value which isn't int for some affected functions? Is this one going to be respun? Yes, here you go. From c5f15eec69bf95c9a1261e0d82abbd039156e75e Mon Sep 17 00:00:00 2001 From: Thomas Martitz k...@rockbox.org Date: Wed, 8 Aug 2012 17:38:04 +0200 Subject: [PATCH 1/3] gccmacro: Work around warnings due to printf redirection by libintl. Libintl defines printf as libintl_printf, which breaks the format attribue. Unfortunately the workaround around provided by libintl is only enabled for cygwin, but not for mingw builds. Therefore install the workaround manually. --- src/pulse/gccmacro.h |9 - 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/src/pulse/gccmacro.h b/src/pulse/gccmacro.h index 1e818ff..e906f59 100644 --- a/src/pulse/gccmacro.h +++ b/src/pulse/gccmacro.h @@ -25,8 +25,15 @@ /** \file * GCC attribute macros */ -#ifdef __GNUC__ +#if defined(__GNUC__) +#ifdef __MINGW32__ +/* libintl overrides printf with a #define. As this breaks this attribute, + * it has a workaround. However the workaround isn't enabled for MINGW + * builds (only cygwin) */ +#define PA_GCC_PRINTF_ATTR(a,b) __attribute__ ((format (__printf__, a, b))) +#else #define PA_GCC_PRINTF_ATTR(a,b) __attribute__ ((format (printf, a, b))) +#endif #else /** If we're in GNU C, use some magic for detecting invalid format strings */ #define PA_GCC_PRINTF_ATTR(a,b) -- 1.7.10.4 ___ pulseaudio-discuss mailing list pulseaudio-discuss@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/pulseaudio-discuss
Re: [pulseaudio-discuss] Brrr ... freeze imminent
On Mon, 2012-10-22 at 15:21 +0200, Peter Meerwald wrote: Hello Arun, I'm hoping to freeze master for 3.0 early next week, with a release in 2-3 weeks after that if everything is quiet. Mostly, it's just pending on me finishing with Peter's NEON patches. If there's anything that needs to go into this release that is in your sekrit branches or on the list and not been given attention, now's the time to holler! what approach are we taking w.r.t ARM NEON patches? compile-time check vs. runtime check + separate build of NEON code? I'm tending towards building the code in a .S separately as Rémi suggests and selecting the right function at run-time. We probably will need a configure-time option to enable compiling the code in at all, but it'll allow us to have a single binary for multiple architectures, which should be useful to binary ARM distros. -- Arun ___ pulseaudio-discuss mailing list pulseaudio-discuss@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/pulseaudio-discuss
[pulseaudio-discuss] svolume_mmx fails to build on OS X
Hi, A bug[1] was filed about a build failure on OS X. Both Clang and GCC give similar errors for the assembly code in svolume_mmx.c. Since bugzilla probably isn't followed actively by many people, and since I'm pretty clueless about this assembly stuff, I decided to raise the issue here on the mailing list, where there may be more people with more clue. Here are the errors from Clang: CC libpulsecore_2.98_la-svolume_mmx.lo pulsecore/svolume_mmx.c:108:43: error: register %rcx is only available in 64-bit mode je 2f \n\t ^ inline asm:5:9: note: instantiated into assembly here movd (%rcx, %edi, 4), %mm0 ^~~~ pulsecore/svolume_mmx.c:124:43: error: register %rcx is only available in 64-bit mode 3: \n\t /* do samples in groups of 2 */ ^ inline asm:33:9: note: instantiated into assembly here movq (%rcx, %edi, 4), %mm0 ^~~~ pulsecore/svolume_mmx.c:137:43: error: register %rcx is only available in 64-bit mode 5: \n\t /* do samples in groups of 4 */ ^ inline asm:59:9: note: instantiated into assembly here movq (%rcx, %edi, 4), %mm0 ^~~~ pulsecore/svolume_mmx.c:138:43: error: register %rcx is only available in 64-bit mode movq (%q1, %3, 4), %%mm0 \n\t /* | v1h | v1l | v0h | v0l | */ ^ inline asm:60:10: note: instantiated into assembly here movq 8(%rcx, %edi, 4), %mm2 ^~~~ pulsecore/svolume_mmx.c:181:43: error: register %rcx is only available in 64-bit mode je 2f \n\t ^ inline asm:9:9: note: instantiated into assembly here movd (%rcx, %edi, 4), %mm0 ^~~~ pulsecore/svolume_mmx.c:199:43: error: register %rcx is only available in 64-bit mode 3: \n\t /* do samples in groups of 2 */ ^ inline asm:39:9: note: instantiated into assembly here movq (%rcx, %edi, 4), %mm0 ^~~~ pulsecore/svolume_mmx.c:214:43: error: register %rcx is only available in 64-bit mode 5: \n\t /* do samples in groups of 4 */ ^ inline asm:73:9: note: instantiated into assembly here movq (%rcx, %edi, 4), %mm0 ^~~~ pulsecore/svolume_mmx.c:215:43: error: register %rcx is only available in 64-bit mode movq (%q1, %3, 4), %%mm0 \n\t /* | v1h | v1l | v0h | v0l | */ ^ inline asm:74:10: note: instantiated into assembly here movq 8(%rcx, %edi, 4), %mm2 ^~~~ 8 errors generated. make[3]: *** [libpulsecore_2.98_la-svolume_mmx.lo] Error 1 make[2]: *** [install] Error 2 make[1]: *** [install-recursive] Error 1 make: *** [install] Error 2 [1] https://bugs.freedesktop.org/show_bug.cgi?id=55733 -- Tanu ___ pulseaudio-discuss mailing list pulseaudio-discuss@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/pulseaudio-discuss
Re: [pulseaudio-discuss] Brrr ... freeze imminent
Am 23.10.2012 12:10, schrieb Arun Raghavan: Are we going to make a pavucontrol release too? Here's one patch for pavucontrol: Enable Automake's silent-rules option. If that's the only one, I'm not to keen on a new release. Speaking of it, I had it working on Windows with some (more or less) nasty workarounds in it and related libs. I hope to clean it up and publish the patches sooner or later, but I doubt it will be in time for 3.0. Are pavucontrol releases generally aligned to PA? Best regards. ___ pulseaudio-discuss mailing list pulseaudio-discuss@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/pulseaudio-discuss
Re: [pulseaudio-discuss] Brrr ... freeze imminent
On Tue, 2012-10-23 at 15:40 +0530, Arun Raghavan wrote:
On Fri, 2012-10-19 at 19:26 +0300, Tanu Kaskinen wrote:
On Fri, 2012-10-19 at 19:24 +0530, Arun Raghavan wrote:
Hello,
I'm hoping to freeze master for 3.0 early next week, with a release in
2-3 weeks after that if everything is quiet. Mostly, it's just pending
on me finishing with Peter's NEON patches.
Thanks, I've been meaning to bring up the release discussion again...
If there's anything that
needs to go into this release that is in your sekrit branches or on the
list and not been given attention, now's the time to holler!
Here's a list of fixes from me that have not been reviewed yet,
according to my bookkeeping. I'd like to get them in 3.0. There
shouldn't be anything particularly complicated. The earliest patches are
from March, so I'm not sure if they all apply cleanly anymore... If
there are patches that don't apply cleanly, complain about it and I'll
resend rebased versions.
bluetooth: Don't free read_smoother in pa__done().
bluetooth: Fix uninitialized variable.
rtp: Get rid of an unitialized value warning from Coverity.
pasuspender: Check pa_context_connect() return value.
Fix pa_parse_boolean() return value checking.
rtsp-client: Fix rtp_port reading.
This one is actually already reviewed, so I pushed it now.
smoother-test: Fix array overflow.
mainloop-signal: Explicitly ignore pa_write() return value.
mainloop: Change wakeup_requested type from pa_bool_t to pa_atomic_t.
mainloop: Remove useless pa_mainloop_wakeup() calls.
mainloop: Remove redundant wakeup_pipe validity checks.
mainloop: Write to the wakeup pipe unconditionally when waking up the
mainloop.
mainloop: Check pa_write() return value.
oss: Check pa_read() return value.
cpulimit: Explicitly ignore pa_read() return value.
daemon: Don't rely on prctl(PR_SET_KEEPCAPS, 0) for dropping caps.
device-restore: Fix empty argument list declaration.
memblockq: Use pa_xnew0() to avoid manual zeroing.
memblockq: Fix the order of setting minreq and prebuf.
modargs: Don't fail needlessly in
pa_modargs_get_sample_spec_and_channel_map().
sink, source: New functions: pa_{sink,source}_new_data_add_port().
core-util: Don't accept random words in pa_parse_boolean().
pasuspender: Resume before exiting in case of SIGINT or fork() failure.
object: Initialize allocated memory to zero.
card: Store a pa_card pointer in pa_card_profile.
switch-on-port-available: Prepare for port-is_{out, in}put going away.
alsa-mixer: Introduce description-key option for paths.
alsa-mixer: Remove the name option from the General section of path
configuration files.
device-port: Make it impossible to have a dual-direction port.
resampler: Make sure that there are no overflows when multiplying
potentially big numbers.
loopback: Use the real sample spec once it's known.
loopback: Don't fix the source output format/rate/channels.
virtual-surround-sink: Fix setting max_request and max_rewind.
sink: Process rewind requests also when suspended.
combine: Keep the timer active in the null mode only when running.
Propagate the PA_{SINK, SOURCE}_LATENCY flag changes through a chain of
filter sinks/sources.
Propagate the PA_{SINK, SOURCE}_DYNAMIC_LATENCY flag changes through a
chain of filter sinks/sources.
sink, source: Really set the fixed latency in
set_fixed_latency_within_thread(), always.
build: Don't enable BlueZ if libbluetooth is not found.
match: Use the SINK_INPUT_FIXATE hook instead of NEW.
build: Add PROTOCOL to EXTRA_DIST.
build: Add a2dp-codecs.h to libbluetooth-util sources.
build: Add .gitignore files to EXTRA_DIST.
:) Let's try to work through the list and get as much done as possible.
The list is much longer than what I anticipated when I wrote I'd like
to get them in 3.0. Here's a list of serious bug fixes, they could be
prioritized for 3.0:
memblockq: Fix the order of setting minreq and prebuf.
pasuspender: Resume before exiting in case of SIGINT or fork() failure.
resampler: Make sure that there are no overflows when multiplying potentially
big numbers.
loopback: Use the real sample spec once it's known.
loopback: Don't fix the source output format/rate/channels.
virtual-surround-sink: Fix setting max_request and max_rewind.
sink: Process rewind requests also when suspended.
combine: Keep the timer active in the null mode only when running.
match: Use the SINK_INPUT_FIXATE hook instead of NEW.
build: Add a2dp-codecs.h to libbluetooth-util sources.
I think make distcheck doesn't work without that last patch...
These would be nice too, since we need them in Mer and Nemo (they are
currently shipped as patches on top of v2.1, but I'd really like to
minimize the amount of patches):
Propagate the PA_{SINK, SOURCE}_LATENCY flag changes through a chain of filter
sinks/sources.
Propagate the PA_{SINK, SOURCE}_DYNAMIC_LATENCY flag changes through a chain
Re: [pulseaudio-discuss] [PATCH] core: Fix a litte-endian bug in ARM svolume code
Hello Arun, checking NEON volume_float32ne NEON: 10223 usec. ref: 46480 usec. checking NEON volume_s16ne NEON: 8484 usec. ARM: 339272 usec. ref: 20203 usec. I was testing with SAMPLES 1019; while you are likely testing with SAMPLES 1022 Checking ARMv6 svolume (with 1019 samples) func: 33923743 usec (min = 338868, max = 341919, stddev = 365.753). orig: 2430664 usec (min = 24261, max = 24445, stddev = 42.2141). Checking ARMv6 svolume (with 1022 samples) func: 915036 usec (min = 9094, max = 9338, stddev = 50.2385). orig: 2437988 usec (min = 24322, max = 24536, stddev = 48.1282). That's odd indeed. I have this on a Freescale i.mx53 (also Cortex A8) Checking ARM svolume func: 905150 usec (min = 9006, max = 9562, stddev = 76.1938). orig: 2278824 usec (min = 22760, max = 23252, stddev = 65.5575). I get similar numbers with SAMPLES 1022 on a beagle-xm; I think you'll see catastrophic runtime with SAMPLES 1019 comparing ARM vs. NEON code, the svolume s16 NEON code uses two MULs, while ARM can do with one -- the ARM instructions (smulwb, ssat) look ideal for the svolume_s16 code three observations: (1) when the number of samples is odd, the ARM code processes the first sample before switching to the unrolled 4-samples-at-a-time loop; this causes the samples pointer to become misaligned (2-byte align) (assuming it was 4-byte aligned initially) I am not sure what guarantees PulseAudio gives on buffer alignment (2) the NEON code generally fails when input data length 4; can be easily fixed (3) neither ARM nor NEON code cares about alignment; just the strategy is different ARM handles cases where length % 3 != 0 first (before entering the unrolled loop); which is bad when the sample buffer is aligned NEON takes care of length % 3 != 0 for the last samples; which is good when the smaple buffer is aligned # ./cpu-test Running suite(s): CPU CPU flags: V6 V7 VFP EDSP NEON VFPV3 Cortex-A8 Initialising ARM optimized volume functions. Checking ARM svolume 0: 1ac8 != 390e (43e9 * d716) Orc not supported. Skipping 50%: Checks: 2, Failures: 1, Errors: 0 tests/cpu-test.c:52:F:svolume:svolume_arm_test:0: Failed Does this include the little-endianness fix? my fault; took the latest source, but failed to make sure that the proper .so was linked - it works with little-endian fixes actually deployed My current testing shows NEON svolume code with int16 samples consistently slower than the ARM code (tried on the Pandaboard, i.mx51, i,mx53, imx.6) by ~10% in most cases. I agree; I think the ARM code is pretty good for s16 handling plus I think enabling NEON is expensive power-wise -- so svolume_s16_neon should be dropped and svolume_s16_arm should be improved to handle odd buffers nicely I just ignored the ARM code before since I was always testing with SAMPLES 1019 this hitting the worst case runtime wise regards, p. -- Peter Meerwald +43-664-218 (mobile) ___ pulseaudio-discuss mailing list pulseaudio-discuss@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/pulseaudio-discuss
Re: [pulseaudio-discuss] [PATCH] core: Fix a litte-endian bug in ARM svolume code
On Tue, 2012-10-23 at 15:48 +0200, Peter Meerwald wrote:
Hello myself,
comparing ARM vs. NEON code, the svolume s16 NEON code uses two MULs,
while ARM can do with one -- the ARM instructions (smulwb, ssat) look
ideal for the svolume_s16 code
for the records, NEON can also do it with one MUL:
static inline void vol_s16_neon(const uint32x4_t *vol4, int16_t *samples,
unsigned length) {
asm volatile (
mov%[length], %[length], lsr #2\n\t
vld1.s32 {q1}, [%[vol]]\n\t
1:\n\t
vld1.16{d0}, [%[samples]]\n\t
vshll.s16 q0, d0, #15\n\t
vqdmulhq.s32 q0, q0, q1\n\t
vmovn.s32 d0, q0\n\t
subs %[length], %[length], #1\n\t
vst1.16{d0}, [%[samples]]!\n\t
bgt1b\n\t
/* output operands (or input operands that get modified) */
: [samples] +r (samples), [length] +r (length)
: [vol] r (vol4) /* input operands */
: memory, cc, q0, q1 /* clobber list */
);
}
Checking ARM NEON svolume
func: 1291289 usec (min = 12817, max = 13184, stddev = 65.9113).
orig: 2438875 usec (min = 24322, max = 25605, stddev = 130.359).
Orc not supported. Skipping
100%: Checks: 3, Failures: 0, Errors: 0
this is a bit better than the previous NEON code (~130 vs. ~151),
but still slower than ARM (~92)
Nice catch on the alignment. I'm trying to extend our tests to catch
these cases. A couple of notes: Rémi Denis-Courmont mentions that you
will likely see performance benefits in the NEON code by sprinkling in
some preloads (PLD). I've also factored out the sconv code and that does
provide a win on all the boards I tried.
To get this moving for 3.0, could you respin just the sconv patches on
top of master (I'll push out my testing code soon) so that we can push
that bit out first while we work on the others?
Cheers,
Arun
___
pulseaudio-discuss mailing list
pulseaudio-discuss@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/pulseaudio-discuss
[pulseaudio-discuss] [PATCH v1 0/6] Bluetooth: Late UUID support
From: Mikel Astiz mikel.as...@bmw-carit.de v1 proposal with the changes suggested by Tanu, most significantly: - Assertions used to detect duplicated card profiles instead of errors. - Hashmap memory freed in callers-side after pa_card_add_ports(). - Added core hooks to report added profiles and ports. - Subscription mechanism used to report card changes. - New UUID hook renamed to _HOOK_UUID_ADDED. - D-Bus API extended with signal NewProfile (patch v1 4/6). From original cover-letter: PulseAudio does not currently support creating card profiles dynamically. This patchset proposes such feature in order to be used by the Bluetooth modules. Mikel Astiz (6): card: Support adding profiles dynamically card: Support adding ports dynamically dbus: Support dynamically created card profiles dbus: Add signal org.PulseAudio.Core1.Card.NewProfile bluetooth: Add hook to announce late UUIDs bluetooth: Handle UUIDs announced later src/modules/bluetooth/bluetooth-util.c | 8 src/modules/bluetooth/bluetooth-util.h | 1 + src/modules/bluetooth/module-bluetooth-device.c | 39 +++ src/modules/dbus/iface-card.c | 52 + src/pulsecore/card.c| 29 ++ src/pulsecore/card.h| 3 ++ src/pulsecore/core.h| 2 + 7 files changed, 134 insertions(+) -- 1.7.11.7 ___ pulseaudio-discuss mailing list pulseaudio-discuss@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/pulseaudio-discuss
[pulseaudio-discuss] [PATCH v1 1/6] card: Support adding profiles dynamically
From: Mikel Astiz mikel.as...@bmw-carit.de
Some cards might need to add profiles during their lifetime, that is,
after the card has been created.
---
src/pulsecore/card.c | 12
src/pulsecore/card.h | 2 ++
src/pulsecore/core.h | 1 +
3 files changed, 15 insertions(+)
diff --git a/src/pulsecore/card.c b/src/pulsecore/card.c
index 492c051..eb1fcc0 100644
--- a/src/pulsecore/card.c
+++ b/src/pulsecore/card.c
@@ -79,6 +79,18 @@ void pa_card_new_data_set_name(pa_card_new_data *data, const
char *name) {
data-name = pa_xstrdup(name);
}
+void pa_card_add_profile(pa_card *c, struct pa_card_profile *profile) {
+pa_assert(c);
+pa_assert(profile);
+
+/* take ownership of the profile */
+pa_assert_se(pa_hashmap_put(c-profiles, profile-name, profile) = 0);
+
+pa_subscription_post(c-core,
PA_SUBSCRIPTION_EVENT_CARD|PA_SUBSCRIPTION_EVENT_CHANGE, c-index);
+
+pa_hook_fire(c-core-hooks[PA_CORE_HOOK_CARD_PROFILE_ADDED], profile);
+}
+
void pa_card_new_data_set_profile(pa_card_new_data *data, const char *profile)
{
pa_assert(data);
diff --git a/src/pulsecore/card.h b/src/pulsecore/card.h
index 6f942e8..4d0a9a2 100644
--- a/src/pulsecore/card.h
+++ b/src/pulsecore/card.h
@@ -100,6 +100,8 @@ void pa_card_new_data_done(pa_card_new_data *data);
pa_card *pa_card_new(pa_core *c, pa_card_new_data *data);
void pa_card_free(pa_card *c);
+void pa_card_add_profile(pa_card *c, struct pa_card_profile *profile);
+
int pa_card_set_profile(pa_card *c, const char *name, pa_bool_t save);
int pa_card_suspend(pa_card *c, pa_bool_t suspend, pa_suspend_cause_t cause);
diff --git a/src/pulsecore/core.h b/src/pulsecore/core.h
index ba21fa9..aec7f80 100644
--- a/src/pulsecore/core.h
+++ b/src/pulsecore/core.h
@@ -113,6 +113,7 @@ typedef enum pa_core_hook {
PA_CORE_HOOK_CARD_PUT,
PA_CORE_HOOK_CARD_UNLINK,
PA_CORE_HOOK_CARD_PROFILE_CHANGED,
+PA_CORE_HOOK_CARD_PROFILE_ADDED,
PA_CORE_HOOK_PORT_AVAILABLE_CHANGED,
PA_CORE_HOOK_MAX
} pa_core_hook_t;
--
1.7.11.7
___
pulseaudio-discuss mailing list
pulseaudio-discuss@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/pulseaudio-discuss
[pulseaudio-discuss] [PATCH v1 2/6] card: Support adding ports dynamically
From: Mikel Astiz mikel.as...@bmw-carit.de
Card profiles -specially the ones registered with pa_card_add_profile()-
might need to create new ports during the lifetime of the card.
---
src/pulsecore/card.c | 17 +
src/pulsecore/card.h | 1 +
src/pulsecore/core.h | 1 +
3 files changed, 19 insertions(+)
diff --git a/src/pulsecore/card.c b/src/pulsecore/card.c
index eb1fcc0..7c18c26 100644
--- a/src/pulsecore/card.c
+++ b/src/pulsecore/card.c
@@ -91,6 +91,23 @@ void pa_card_add_profile(pa_card *c, struct pa_card_profile
*profile) {
pa_hook_fire(c-core-hooks[PA_CORE_HOOK_CARD_PROFILE_ADDED], profile);
}
+void pa_card_add_ports(pa_card *c, pa_hashmap *ports) {
+pa_device_port *p;
+void *state;
+
+pa_assert(c);
+pa_assert(ports);
+
+/* take ownership of the ports */
+PA_HASHMAP_FOREACH(p, ports, state)
+pa_assert_se(pa_hashmap_put(c-ports, p-name, p) = 0);
+
+pa_subscription_post(c-core,
PA_SUBSCRIPTION_EVENT_CARD|PA_SUBSCRIPTION_EVENT_CHANGE, c-index);
+
+while ((p = pa_hashmap_steal_first(ports)) != NULL)
+pa_hook_fire(c-core-hooks[PA_CORE_HOOK_PORT_ADDED], p);
+}
+
void pa_card_new_data_set_profile(pa_card_new_data *data, const char *profile)
{
pa_assert(data);
diff --git a/src/pulsecore/card.h b/src/pulsecore/card.h
index 4d0a9a2..c82a369 100644
--- a/src/pulsecore/card.h
+++ b/src/pulsecore/card.h
@@ -101,6 +101,7 @@ pa_card *pa_card_new(pa_core *c, pa_card_new_data *data);
void pa_card_free(pa_card *c);
void pa_card_add_profile(pa_card *c, struct pa_card_profile *profile);
+void pa_card_add_ports(pa_card *c, pa_hashmap *ports);
int pa_card_set_profile(pa_card *c, const char *name, pa_bool_t save);
diff --git a/src/pulsecore/core.h b/src/pulsecore/core.h
index aec7f80..cf7cc11 100644
--- a/src/pulsecore/core.h
+++ b/src/pulsecore/core.h
@@ -115,6 +115,7 @@ typedef enum pa_core_hook {
PA_CORE_HOOK_CARD_PROFILE_CHANGED,
PA_CORE_HOOK_CARD_PROFILE_ADDED,
PA_CORE_HOOK_PORT_AVAILABLE_CHANGED,
+PA_CORE_HOOK_PORT_ADDED,
PA_CORE_HOOK_MAX
} pa_core_hook_t;
--
1.7.11.7
___
pulseaudio-discuss mailing list
pulseaudio-discuss@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/pulseaudio-discuss
[pulseaudio-discuss] [PATCH v1 4/6] dbus: Add signal org.PulseAudio.Core1.Card.NewProfile
From: Mikel Astiz mikel.as...@bmw-carit.de
Add a new D-Bus signal to report profiles that have been dynamically
created for already existing cards.
---
src/modules/dbus/iface-card.c | 16
1 file changed, 16 insertions(+)
diff --git a/src/modules/dbus/iface-card.c b/src/modules/dbus/iface-card.c
index 902b21c..57fd248 100644
--- a/src/modules/dbus/iface-card.c
+++ b/src/modules/dbus/iface-card.c
@@ -108,15 +108,18 @@ static pa_dbus_method_handler
method_handlers[METHOD_HANDLER_MAX] = {
enum signal_index {
SIGNAL_ACTIVE_PROFILE_UPDATED,
+SIGNAL_NEW_PROFILE,
SIGNAL_PROPERTY_LIST_UPDATED,
SIGNAL_MAX
};
static pa_dbus_arg_info active_profile_updated_args[] = { { profile,
o, NULL } };
+static pa_dbus_arg_info new_profile_args[] ={ { profile,
o, NULL } };
static pa_dbus_arg_info property_list_updated_args[] = { { property_list,
a{say}, NULL } };
static pa_dbus_signal_info signals[SIGNAL_MAX] = {
[SIGNAL_ACTIVE_PROFILE_UPDATED] = { .name = ActiveProfileUpdated,
.arguments = active_profile_updated_args, .n_arguments = 1 },
+[SIGNAL_NEW_PROFILE]= { .name = NewProfile,
.arguments = new_profile_args,.n_arguments = 1 },
[SIGNAL_PROPERTY_LIST_UPDATED] = { .name = PropertyListUpdated,
.arguments = property_list_updated_args, .n_arguments = 1 }
};
@@ -496,6 +499,8 @@ static pa_hook_result_t card_profile_added_cb(void
*hook_data, void *call_data,
pa_dbusiface_card *c = slot_data;
pa_card_profile *profile = call_data;
pa_dbusiface_card_profile *p;
+const char *object_path;
+DBusMessage *signal_msg;
if (!card_has_profile(c-card, profile))
return PA_HOOK_OK;
@@ -506,6 +511,17 @@ static pa_hook_result_t card_profile_added_cb(void
*hook_data, void *call_data,
p = pa_dbusiface_card_profile_new(c, core, profile,
c-next_profile_index++);
pa_hashmap_put(c-profiles, pa_dbusiface_card_profile_get_name(p), p);
+/* Send D-Bus signal */
+object_path = pa_dbusiface_card_profile_get_path(p);
+
+pa_assert_se(signal_msg = dbus_message_new_signal(c-path,
+
PA_DBUSIFACE_CARD_INTERFACE,
+
signals[SIGNAL_NEW_PROFILE].name));
+pa_assert_se(dbus_message_append_args(signal_msg, DBUS_TYPE_OBJECT_PATH,
object_path, DBUS_TYPE_INVALID));
+
+pa_dbus_protocol_send_signal(c-dbus_protocol, signal_msg);
+dbus_message_unref(signal_msg);
+
return PA_HOOK_OK;
}
--
1.7.11.7
___
pulseaudio-discuss mailing list
pulseaudio-discuss@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/pulseaudio-discuss
[pulseaudio-discuss] [PATCH v1 5/6] bluetooth: Add hook to announce late UUIDs
From: Mikel Astiz mikel.as...@bmw-carit.de
UUIDs might be announced at any time, so a hook is needed to notify any
interested module. In practice, the UUIDs are quite stable with the
exception of the pairing procedure, where the UUIDs are reported by
BlueZ as soon as they are discovered.
---
src/modules/bluetooth/bluetooth-util.c | 8
src/modules/bluetooth/bluetooth-util.h | 1 +
2 files changed, 9 insertions(+)
diff --git a/src/modules/bluetooth/bluetooth-util.c
b/src/modules/bluetooth/bluetooth-util.c
index 1e81864..272b6ce 100644
--- a/src/modules/bluetooth/bluetooth-util.c
+++ b/src/modules/bluetooth/bluetooth-util.c
@@ -383,9 +383,17 @@ static int parse_device_property(pa_bluetooth_discovery
*y, pa_bluetooth_device
const char *value;
dbus_message_iter_get_basic(ai, value);
+
+if (pa_bluetooth_uuid_has(d-uuids, value)) {
+dbus_message_iter_next(ai);
+continue;
+}
+
node = uuid_new(value);
PA_LLIST_PREPEND(pa_bluetooth_uuid, d-uuids, node);
+
pa_hook_fire(d-hooks[PA_BLUETOOTH_DEVICE_HOOK_UUID_ADDED], (char *) value);
+
/* Vudentz said the interfaces are here when the UUIDs are
announced */
if (strcasecmp(HSP_AG_UUID, value) == 0 ||
strcasecmp(HFP_AG_UUID, value) == 0) {
pa_assert_se(m =
dbus_message_new_method_call(org.bluez, d-path,
org.bluez.HandsfreeGateway, GetProperties));
diff --git a/src/modules/bluetooth/bluetooth-util.h
b/src/modules/bluetooth/bluetooth-util.h
index 4e3b5f0..8a3f2ad 100644
--- a/src/modules/bluetooth/bluetooth-util.h
+++ b/src/modules/bluetooth/bluetooth-util.h
@@ -94,6 +94,7 @@ typedef enum pa_bt_audio_state {
/* Hook data: pa_bluetooth_device pointer. */
typedef enum pa_bluetooth_device_hook {
PA_BLUETOOTH_DEVICE_HOOK_REMOVED, /* Call data: NULL. */
+PA_BLUETOOTH_DEVICE_HOOK_UUID_ADDED, /* Call data: const char *uuid. */
PA_BLUETOOTH_DEVICE_HOOK_MAX
} pa_bluetooth_device_hook_t;
--
1.7.11.7
___
pulseaudio-discuss mailing list
pulseaudio-discuss@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/pulseaudio-discuss
[pulseaudio-discuss] [PATCH v1 6/6] bluetooth: Handle UUIDs announced later
From: Mikel Astiz mikel.as...@bmw-carit.de
In some cases (typically during pairing) UUIDs might be reported by
BlueZ incrementally, that is, as soon as they have been discovered. At
this point module-bluetooth-device might already be loaded, so the late
UUID announcements need to be handled and additional card profiles
might need to be created accordingly.
---
src/modules/bluetooth/module-bluetooth-device.c | 39 +
1 file changed, 39 insertions(+)
diff --git a/src/modules/bluetooth/module-bluetooth-device.c
b/src/modules/bluetooth/module-bluetooth-device.c
index 8c2efa4..0442035 100644
--- a/src/modules/bluetooth/module-bluetooth-device.c
+++ b/src/modules/bluetooth/module-bluetooth-device.c
@@ -140,6 +140,7 @@ struct userdata {
pa_module *module;
pa_bluetooth_device *device;
+pa_hook_slot *uuid_added_slot;
char *address;
char *path;
pa_bluetooth_transport *transport;
@@ -2495,6 +2496,38 @@ static pa_bluetooth_device* find_device(struct userdata
*u, const char *address,
}
/* Run from main thread */
+static pa_hook_result_t uuid_added_cb(pa_bluetooth_device *d, const char
*uuid, struct userdata *u) {
+pa_card_profile *p;
+pa_hashmap *new_ports;
+
+pa_assert(d);
+pa_assert(uuid);
+pa_assert(u);
+
+p = create_card_profile(u, uuid);
+
+if (!p)
+return PA_HOOK_OK;
+
+if (pa_hashmap_get(u-card-profiles, p-name)) {
+pa_card_profile_free(p);
+return PA_HOOK_OK;
+}
+
+pa_card_add_profile(u-card, p);
+
+new_ports = pa_hashmap_new(pa_idxset_string_hash_func,
pa_idxset_string_compare_func);
+
+create_ports_for_profile(u, new_ports, p);
+
+pa_card_add_ports(u-card, new_ports);
+
+pa_device_port_hashmap_free(new_ports);
+
+return PA_HOOK_OK;
+}
+
+/* Run from main thread */
static int setup_dbus(struct userdata *u) {
DBusError err;
@@ -2597,6 +2630,9 @@ int pa__init(pa_module* m) {
u-device = device;
+u-uuid_added_slot =
pa_hook_connect(device-hooks[PA_BLUETOOTH_DEVICE_HOOK_UUID_ADDED],
PA_HOOK_NORMAL,
+ (pa_hook_cb_t) uuid_added_cb, u);
+
/* Add the card structure. This will also initialize the default profile */
if (add_card(u) 0)
goto fail;
@@ -2714,6 +2750,9 @@ void pa__done(pa_module *m) {
pa_dbus_connection_unref(u-connection);
}
+if (u-uuid_added_slot)
+pa_hook_slot_free(u-uuid_added_slot);
+
if (u-msg)
pa_xfree(u-msg);
--
1.7.11.7
___
pulseaudio-discuss mailing list
pulseaudio-discuss@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/pulseaudio-discuss
[pulseaudio-discuss] [PATCH] module rtp-recv: Fix warning using pa_assert_not_reached()
From: Peter Meerwald p.meerw...@bct-electronic.com
modules/rtp/module-rtp-recv.c:462:8: warning: 'r' may be used uninitialized
in this function [-Wuninitialized]
Signed-off-by: Peter Meerwald p.meerw...@bct-electronic.com
---
src/modules/rtp/module-rtp-recv.c |5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/src/modules/rtp/module-rtp-recv.c
b/src/modules/rtp/module-rtp-recv.c
index 412f4c3..3e9250c 100644
--- a/src/modules/rtp/module-rtp-recv.c
+++ b/src/modules/rtp/module-rtp-recv.c
@@ -451,13 +451,14 @@ static int mcast_socket(const struct sockaddr* sa,
socklen_t salen) {
mr4.imr_multiaddr = ((const struct sockaddr_in*) sa)-sin_addr;
r = setsockopt(fd, IPPROTO_IP, IP_ADD_MEMBERSHIP, mr4, sizeof(mr4));
#ifdef HAVE_IPV6
-} else {
+} else if (af == AF_INET6) {
struct ipv6_mreq mr6;
memset(mr6, 0, sizeof(mr6));
mr6.ipv6mr_multiaddr = ((const struct sockaddr_in6*) sa)-sin6_addr;
r = setsockopt(fd, IPPROTO_IPV6, IPV6_JOIN_GROUP, mr6, sizeof(mr6));
#endif
-}
+} else
+pa_assert_not_reached();
if (r 0) {
pa_log_info(Joining mcast group failed: %s, pa_cstrerror(errno));
--
1.7.9.5
___
pulseaudio-discuss mailing list
pulseaudio-discuss@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/pulseaudio-discuss
[pulseaudio-discuss] [PATCH 1/2] core: Initialize ARM NEON code if available
From: Peter Meerwald p.meerw...@bct-electronic.com
Signed-off-by: Peter Meerwald p.meerw...@bct-electronic.com
---
src/pulsecore/cpu-arm.c |4 +++-
src/pulsecore/cpu-arm.h |2 ++
2 files changed, 5 insertions(+), 1 deletion(-)
diff --git a/src/pulsecore/cpu-arm.c b/src/pulsecore/cpu-arm.c
index 934c7fe..60e047f 100644
--- a/src/pulsecore/cpu-arm.c
+++ b/src/pulsecore/cpu-arm.c
@@ -140,7 +140,9 @@ pa_bool_t pa_cpu_init_arm(pa_cpu_arm_flag_t *flags) {
#if defined (__linux__)
pa_cpu_get_arm_flags(flags);
-if (*flags PA_CPU_ARM_V6)
+if (*flags PA_CPU_ARM_NEON)
+pa_convert_func_init_neon(*flags);
+else if (*flags PA_CPU_ARM_V6)
pa_volume_func_init_arm(*flags);
return TRUE;
diff --git a/src/pulsecore/cpu-arm.h b/src/pulsecore/cpu-arm.h
index 91aa659..88b2aec 100644
--- a/src/pulsecore/cpu-arm.h
+++ b/src/pulsecore/cpu-arm.h
@@ -41,4 +41,6 @@ pa_bool_t pa_cpu_init_arm(pa_cpu_arm_flag_t *flags);
/* some optimized functions */
void pa_volume_func_init_arm(pa_cpu_arm_flag_t flags);
+void pa_convert_func_init_neon(pa_cpu_arm_flag_t flags);
+
#endif /* foocpuarmhfoo */
--
1.7.9.5
___
pulseaudio-discuss mailing list
pulseaudio-discuss@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/pulseaudio-discuss
[pulseaudio-discuss] [PATCH 2/2] core: Add ARM NEON optimized sample conversion code
From: Peter Meerwald p.meerw...@bct-electronic.com
v4:
* fix for sample length 4
v3:
* convert from intrinsics to inline assembly
v2:
* load and store data with vld1/vld1q and vst1/vst1q, resp., to work
around alignment issues of compiler-generated vldmia instruction
* remove redundant check for NEON flags
Ubuntu/Linaro gcc 4.6.3
arm-linux-gnueabi-gcc -O2 -mcpu=cortex-a8 -mfloat-abi=softfp -mfpu=neon
runtime on beagle-xm:
D: [pulseaudio] sconv_neon.c: checking NEON sconv_s16le_from_float
I: [pulseaudio] sconv_neon.c: NEON: 3754 usec.
I: [pulseaudio] sconv_neon.c: ref: 58594 usec.
D: [pulseaudio] sconv_neon.c: checking NEON sconv_s16le_to_float
I: [pulseaudio] sconv_neon.c: NEON: 1831 usec.
I: [pulseaudio] sconv_neon.c: ref: 10528 usec.
I: [pulseaudio] sconv_neon.c: Initialising ARM NEON optimized conversions.
conversion may be off by one for some samples due to rounding issues
Signed-off-by: Peter Meerwald p.meerw...@bct-electronic.com
---
src/Makefile.am|2 +-
src/pulsecore/sconv_neon.c | 213
2 files changed, 214 insertions(+), 1 deletion(-)
create mode 100644 src/pulsecore/sconv_neon.c
diff --git a/src/Makefile.am b/src/Makefile.am
index 6212c74..0ca1ec2 100644
--- a/src/Makefile.am
+++ b/src/Makefile.am
@@ -850,7 +850,7 @@ libpulsecore_@PA_MAJORMINOR@_la_SOURCES = \
pulsecore/svolume_mmx.c pulsecore/svolume_sse.c \
pulsecore/sconv-s16be.c pulsecore/sconv-s16be.h \
pulsecore/sconv-s16le.c pulsecore/sconv-s16le.h \
- pulsecore/sconv_sse.c \
+ pulsecore/sconv_sse.c pulsecore/sconv_neon.c \
pulsecore/sconv.c pulsecore/sconv.h \
pulsecore/shared.c pulsecore/shared.h \
pulsecore/sink-input.c pulsecore/sink-input.h \
diff --git a/src/pulsecore/sconv_neon.c b/src/pulsecore/sconv_neon.c
new file mode 100644
index 000..96e91ca
--- /dev/null
+++ b/src/pulsecore/sconv_neon.c
@@ -0,0 +1,213 @@
+/***
+ This file is part of PulseAudio.
+
+ Copyright 2012 Peter Meerwald p.meerw...@bct-electronic.com
+
+ PulseAudio is free software; you can redistribute it and/or modify
+ it under the terms of the GNU Lesser General Public License as published
+ by the Free Software Foundation; either version 2.1 of the License,
+ or (at your option) any later version.
+
+ PulseAudio is distributed in the hope that it will be useful, but
+ WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ General Public License for more details.
+***/
+
+#ifdef HAVE_CONFIG_H
+#include config.h
+#endif
+
+#include pulse/rtclock.h
+
+#include pulsecore/macro.h
+#include pulsecore/endianmacros.h
+
+#include cpu-arm.h
+#include sconv.h
+
+#if defined(__ARM_NEON__)
+
+#include math.h
+#include arm_neon.h
+
+#define RUN_TEST
+
+static void pa_sconv_s16le_from_f32ne_neon(unsigned n, const float *src,
int16_t *dst) {
+unsigned i = n 3;
+
+asm volatile (
+movs %[n], %[n], lsr #2\n\t
+beq2f\n\t
+vdup.f32 q2, %[plusone]\n\t
+vneg.f32 q3, q2\n\t
+vdup.f32 q4, %[scale]\n\t
+vdup.u32 q5, %[mask]\n\t
+vdup.f32 q6, %[half]\n\t
+1:\n\t
+vld1.32{q0}, [%[src]]!\n\t
+vmin.f32 q0, q0, q2\n\t /* clamp */
+vmax.f32 q0, q0, q3\n\t
+vmul.f32 q0, q0, q4\n\t /* scale */
+vand.u32 q1, q0, q5\n\t
+vorr.u32 q1, q1, q6\n\t /* round */
+vadd.f32 q0, q0, q1\n\t
+vcvt.s32.f32 q0, q0\n\t /* narrow */
+vmovn.i32 d0, q0\n\t
+subs %[n], %[n], #1\n\t
+vst1.16{d0}, [%[dst]]!\n\t
+bgt1b\n\t
+2:\n\t
+ /* output operands (or input operands that get modified) */
+: [dst] +r (dst), [src] +r (src), [n] +r (n)
+: [plusone] r (1.0f), [scale] r (32767.0f),
+ [half] r (0.5f), [mask] r (0x8000) /* input operands */
+: memory, cc, q0, q1, q2, q3, q4, q5, q6 /* clobber list
*/
+);
+
+// leftovers
+while (i--) {
+*dst++ = (int16_t) lrintf(PA_CLAMP_UNLIKELY(*src, -1.0f, 1.0f) *
0x7FFF);
+src++;
+}
+}
+
+static void pa_sconv_s16le_to_f32ne_neon(unsigned n, const int16_t *src, float
*dst) {
+unsigned i = n 3;
+
+const float invscale = 1.0f / 0x7FFF;
+
+asm volatile (
+movs%[n], %[n], lsr #2\n\t
+beq2f\n\t
+vdup.f32 q1, %[invscale]\n\t
+1:\n\t
+vld1.16{d0}, [%[src]]!\n\t
+vmovl.s16 q0, d0\n\t
+vcvt.f32.s32 q0, q0\n\t
+vmul.f32 q0, q0, q1\n\t
+subs %[n], %[n], #1\n\t
+vst1.32{q0}, [%[dst]]!\n\t
+bgt1b\n\t
+2:\n\t
+ /* output operands (or input operands that get modified) */
+: [dst] +r (dst), [src] +r (src), [n] +r (n)
+: [invscale] r (invscale) /* input operands */
+: memory, cc, q0, q1 /* clobber list */
+);
+
+// leftovers
+while (i--) {
+*dst++ = *src++ * invscale;
+}
Re: [pulseaudio-discuss] [PATCH] core: Fix a litte-endian bug in ARM svolume code
To get this moving for 3.0, could you respin just the sconv patches on top of master (I'll push out my testing code soon) so that we can push that bit out first while we work on the others? just sent the sconv stuff only thank you for your efforts! p. -- Peter Meerwald +43-664-218 (mobile) ___ pulseaudio-discuss mailing list pulseaudio-discuss@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/pulseaudio-discuss
[pulseaudio-discuss] [PATCH] device-restore: When restoring volume, print the restored volume to the log.
---
src/modules/module-device-restore.c | 16 ++--
1 file changed, 10 insertions(+), 6 deletions(-)
diff --git a/src/modules/module-device-restore.c
b/src/modules/module-device-restore.c
index b9bd498..c9ffbb1 100644
--- a/src/modules/module-device-restore.c
+++ b/src/modules/module-device-restore.c
@@ -779,12 +779,13 @@ static pa_hook_result_t sink_fixate_hook_callback(pa_core
*c, pa_sink_new_data *
if (!new_data-volume_is_set) {
pa_cvolume v;
+char buf[PA_CVOLUME_SNPRINT_MAX];
pa_log_info(Restoring volume for sink %s., new_data-name);
-
v = e-volume;
pa_cvolume_remap(v, e-channel_map, new_data-channel_map);
pa_sink_new_data_set_volume(new_data, v);
+pa_log_info(Restored volume: %s, pa_cvolume_snprint(buf,
PA_CVOLUME_SNPRINT_MAX, new_data-volume));
new_data-save_volume = TRUE;
} else
@@ -823,14 +824,15 @@ static pa_hook_result_t sink_port_hook_callback(pa_core
*c, pa_sink *sink, struc
if ((e = perportentry_read(u, name, (sink-active_port ?
sink-active_port-name : NULL {
if (u-restore_volume e-volume_valid) {
-
+char buf[PA_CVOLUME_SNPRINT_MAX];
pa_cvolume v;
pa_log_info(Restoring volume for sink %s., sink-name);
-
v = e-volume;
pa_cvolume_remap(v, e-channel_map, sink-channel_map);
pa_sink_set_volume(sink, v, TRUE, FALSE);
+pa_log_info(Restored volume: %s, pa_cvolume_snprint(buf,
PA_CVOLUME_SNPRINT_MAX, sink-reference_volume));
+
sink-save_volume = TRUE;
}
@@ -919,13 +921,14 @@ static pa_hook_result_t
source_fixate_hook_callback(pa_core *c, pa_source_new_da
if (u-restore_volume e-volume_valid) {
if (!new_data-volume_is_set) {
+char buf[PA_CVOLUME_SNPRINT_MAX];
pa_cvolume v;
pa_log_info(Restoring volume for source %s., new_data-name);
-
v = e-volume;
pa_cvolume_remap(v, e-channel_map, new_data-channel_map);
pa_source_new_data_set_volume(new_data, v);
+pa_log_info(Restored volume: %s, pa_cvolume_snprint(buf,
PA_CVOLUME_SNPRINT_MAX, new_data-volume));
new_data-save_volume = TRUE;
} else
@@ -964,14 +967,15 @@ static pa_hook_result_t source_port_hook_callback(pa_core
*c, pa_source *source,
if ((e = perportentry_read(u, name, (source-active_port ?
source-active_port-name : NULL {
if (u-restore_volume e-volume_valid) {
-
+char buf[PA_CVOLUME_SNPRINT_MAX];
pa_cvolume v;
pa_log_info(Restoring volume for source %s., source-name);
-
v = e-volume;
pa_cvolume_remap(v, e-channel_map, source-channel_map);
pa_source_set_volume(source, v, TRUE, FALSE);
+pa_log_info(Restored volume: %s, pa_cvolume_snprint(buf,
PA_CVOLUME_SNPRINT_MAX, source-reference_volume));
+
source-save_volume = TRUE;
}
--
1.7.10.4
___
pulseaudio-discuss mailing list
pulseaudio-discuss@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/pulseaudio-discuss
Re: [pulseaudio-discuss] Brrr ... freeze imminent
Hello, please consider to also merge version 2 of my patch module-virtual-surround- sink: add silence to hrir if necessary. The problem is, that the libsndfile based resamplers do not guarantee to output any samples for too short input sequences. Thus, we should append silence to the hrir if we do not get enough samples from the resampler. Regards, Ole ___ pulseaudio-discuss mailing list pulseaudio-discuss@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/pulseaudio-discuss
Re: [pulseaudio-discuss] [PATCH] pulse: Fix for volume documentation
On Thu, 2012-08-09 at 15:52 -0300, Flavio Ceolin wrote: Fixing the documentation for the function pa_cvolume_dec(). --- src/pulse/volume.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) Pulled, thanks! -- Arun ___ pulseaudio-discuss mailing list pulseaudio-discuss@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/pulseaudio-discuss
Re: [pulseaudio-discuss] a question about audio synchronization between local sink and tunnel sink
On Thu, 2012-10-18 at 23:17 +, Sun, Xiaodong wrote: Tanu, Yes, I am new to PulseAudio code. All I have done so far with code is, reading a few module source code (i.e. module-combine-sink.c, module-tunnel.c, sink.c, etc.) to understand the nature of this unsynchronized issue I met. I didn't expect I will be involved as a developer at all. Since this feature is very important to me, it seems I have to dive deeper. My time is very tight though. Is there any design document you can share with me. I can only find the document for developer on freedesktop.org. [...] The only resources available are: * Assorted developer docs: http://www.freedesktop.org/wiki/Software/PulseAudio/Documentation/Developer * API: http://freedesktop.org/software/pulseaudio/doxygen/ * The code :) As Tanu mentions, feel free to drop by IRC to talk to us, or post any questions you have about the code / design on list. -- Arun ___ pulseaudio-discuss mailing list pulseaudio-discuss@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/pulseaudio-discuss
