Hello,
I dare to ask special attention for this group of small fixes.
They are currently holding N900 audio for hostage.
Bug 15950 - "No audio or audio routing on N900 with Pulseaudio 0.9.22"
https://bugs.meego.com/show_bug.cgi?id=15950
It's been classified as 'high' 'critical'.
We made the fixes as small and simple as possible.
- The two fixes in SR#16652 have both been in pulseaudio before.
- SR#16717 will only affect n900 (tested and working)
- SR#16718 removing a few asserts will not break mfld audio. (resource
policy which brings this issue up is not even in use. AFAIK)
Note that the 2 SR's that this one (16652) depends on have changed due
to cleanup. They are: SR#16717 and SR#16718.
They change the handling of null pointer to be more allowing.
The fix is likely to be temporary until better fix is found.
thank you,
SSirkia
On 04/21/2011 11:50 AM, Sami Sirkia wrote:
Hi,
I have made the following changes to pulseaudio in project devel:multimedia.
Please review and accept ASAP.
Thank You,
Sami Sirkia
[This message was auto-generated]
---
Request #16652:
submit: home:sami.sirkia:branches:devel:multimedia/pulseaudio(r5)(cleanup)
-> devel:multimedia/pulseaudio
Message:
This should be merged after SR#16642 and SR16651. They change null pointer
handling in mainvolume module
for packages pulseaudio-modules-n900 and pulseaudio-modules-mfld.
It this fix is taken alone mainvolume will assert on N900.
Mfld does not use resource policy, so this problem does not occur.
* Tue Apr 19 2011 Sami Sirkia<[email protected]> - 0.9.22
- Fix for BMC#15950 - No audio or audio routing on N900 with Pulseaudio 0.9.22
A strict checking of audio driver properties caused a failure to load
alsamixer paths.
The patch is temporary but old, so this has to be handled in driver soon.
- Another fix for the same BMC#15950 corrects a input/output typo. We have had
this fix before in .19.
State: new 2011-04-21T01:50:27 sami.sirkia
Comment: None
changes files:
--------------
--- pulseaudio.changes
+++ pulseaudio.changes
@@ -0,0 +1,6 @@
+* Tue Apr 19 2011 Sami Sirkia<[email protected]> - 0.9.22
+- Fix for BMC#15950 - No audio or audio routing on N900 with Pulseaudio 0.9.22
+ A strict checking of audio driver properties caused a failure to load
alsamixer paths.
+ The patch is temporary but old, so this has to be handled in driver soon.
+- Another fix for the same BMC#15950 corrects a input/output typo. We have had
this fix before in .19.
+
new:
----
tmp-alsa-mixer-Adapt-to-conflicting-dB-range-information.patch
spec files:
-----------
--- pulseaudio.spec
+++ pulseaudio.spec
@@ -279,6 +279,7 @@
Patch255: use-a2dp-profile-for-non-mfld.patch
Patch256: fix-9972-auto-route-audio-to-bt-sink.patch
Patch257: pa-0.9.22-add-old-sink-and-source.patch
+Patch258: tmp-alsa-mixer-Adapt-to-conflicting-dB-range-information.patch
Requires: udev>= 145-3
Requires: xcb-util
Requires(post): /sbin/ldconfig
@@ -861,6 +862,8 @@
%patch256 -p1
# pa-0.9.22-add-old-sink-and-source.patch
%patch257 -p1
+# tmp-alsa-mixer-Adapt-to-conflicting-dB-range-information.patch
+%patch258 -p1
#>> setup
#<< setup
other changes:
--------------
++++++ pa-0.9.22-add-old-sink-and-source.patch
--- pa-0.9.22-add-old-sink-and-source.patch
+++ pa-0.9.22-add-old-sink-and-source.patch
@@ -2732,7 +2732,7 @@
+ if (!(u->mixer_handle = pa_alsa_old_open_mixer(mixer_device)))
+ return;
+
-+ if (!(u->mixer_path_set = pa_alsa_path_set_new(mapping,
PA_ALSA_DIRECTION_OUTPUT))) {
++ if (!(u->mixer_path_set = pa_alsa_path_set_new(mapping,
PA_ALSA_DIRECTION_INPUT))) {
+ pa_log_debug("pa_alsa_path_set_new() failed.");
+ goto fail;
+ }
++++++ pulseaudio.yaml
--- pulseaudio.yaml
+++ pulseaudio.yaml
@@ -275,6 +275,7 @@
- use-a2dp-profile-for-non-mfld.patch
- fix-9972-auto-route-audio-to-bt-sink.patch
- pa-0.9.22-add-old-sink-and-source.patch
+ - tmp-alsa-mixer-Adapt-to-conflicting-dB-range-information.patch
Description: |
PulseAudio is a sound server for Linux and other Unix like operating
++++++ tmp-alsa-mixer-Adapt-to-conflicting-dB-range-information.patch (new)
--- tmp-alsa-mixer-Adapt-to-conflicting-dB-range-information.patch
+++ tmp-alsa-mixer-Adapt-to-conflicting-dB-range-information.patch
+From d17cefcf855128c2e1375666e637937af0400b14 Mon Sep 17 00:00:00 2001
+From: Tanu Kaskinen<[email protected]>
+Date: Wed, 19 May 2010 14:38:07 +0300
+Subject: [PATCH 093/172] alsa-mixer: Adapt to conflicting dB range information
from alsa.
+
+I think this can be classified as a "temporary hack", because in the general
+case we don't know what's the reason for the conflicting dB range information,
+and therefore the solution that's used here for adapting to the situation can't
+be used in all cases.
+---
+ src/modules/alsa/alsa-mixer.c | 13 ++++++++++++-
+ 1 files changed, 12 insertions(+), 1 deletions(-)
+
+diff --git a/src/modules/alsa/alsa-mixer.c b/src/modules/alsa/alsa-mixer.c
+index 3f6aa79..7f16965 100644
+--- a/src/modules/alsa/alsa-mixer.c
++++ b/src/modules/alsa/alsa-mixer.c
+@@ -1279,7 +1279,18 @@ static int element_probe(pa_alsa_element *e,
snd_mixer_t *m) {
+ e->alsa_name,
+ min_dB / 100.0, max_dB / 100.0,
+ min_dB_checked / 100.0, e->min_volume,
max_dB_checked / 100.0, e->max_volume);
+- return -1;
++ /*return -1;*/
++
++ /* FIXME: Returning -1 should be an appropriate
general
++ * way to react to a broken driver. The following code
++ * is a workaround for a known alsa bug, and should be
++ * removed as soon as the bug gets fixed. This comment
++ * was written on 2010-05-19. If you're reading this
++ * many weeks after that date, it's time to remove the
++ * following three lines of code and return -1 here.
*/
++ if (min_dB_checked != SND_CTL_TLV_DB_GAIN_MUTE)
++ min_dB = min_dB_checked;
++ max_dB = max_dB_checked;
+ }
+ }
+
+--
+1.7.3.2
+
_______________________________________________
MeeGo-commits mailing list
[email protected]
http://lists.meego.com/listinfo/meego-commits
_______________________________________________
MeeGo-packaging mailing list
[email protected]
http://lists.meego.com/listinfo/meego-packaging
