Firefox, flash, alsa-plugins-pulseaudio - lack of "synergy" with pulseaudio... ?

[Tom London]

For some time, I have observed situations where the pulseaudio daemon
would "suddenly" start consuming much more CPU resources, jumping from
about 1-3% to 15-20%.

I think I have localized the "problem" on my system to firefox, flash,
alsa-plugins-pulseaudio and pulseaudio.

On my rawhide x86_64 system, I listen to mp3's all day long with rhythmbox.

Typically, rhythmbox consumes less than 5% and pulseaudio about 1.3-3%
of the CPU (as displayed by "top").

I've noticed on occasion pulseaudio jumping up to as high as 20%, and
consuming up to about 8-12% even when I pause rhythmbox.

I believe I've narrowed this down to sound originating from firefox:
for example, gmail chat sounds and/or flash clips.

For example, playing the flash clips on http://nytimes.com is almost
guaranteed to cause pulseaudio to spike to 8-10% and remain there even
after flash has "stopped".

A reliable way to get the pulseaudio daemon to "drop down" in cycles
is to kill firefox (or kill/start the pulseaudio daemon).  While
firefox continues to run, pulseaudio continues at >10%, typically.

Documentation from gmail chat indicates is uses flash to generate sounds, too.

I'm using the native x86_64 flash plugin
(libflashplayer-10.0.d21.1.linux-x86_64.so.tar.gz).

Anyone else seeing this?

As a "workaround", I've (temporarily?) disabled sounds in gmail chat,
and gotten in the habit of quitting firefox after playing flash clips.

Can someone else test this?
     1. run "top" in a terminal window.
     2. start rhythmbox to play continuously (for example, by
selecting "repeat")
     3. observer pulseaudio's CPU usage for a while
     4. start firefox
     5. create a tab and brows to http://nytimes.com
     6. start the flash clip displayed about 1/3 of the way down the page
     7. observer pulseaudio's CPU usage while the clip is playing.
Does it shoot up?
     8. stop the flash clip (e.g., use the pause button or close the tab).
     9. observe pulseaudio's CPU usage.  Does it remain higher than in step 3?
    10. exit firefox.  does pulseaudio's CPU usage drop?

Here is my setup: Thinkpad X200, x86_64 rawhide
kernel/pulseaudio/etc., native x86_64 flash
(libflashplayer-10.0.d21.1.linux-x86_64.so.tar.gz), no
nspluginwrapper.

I've already filed a bug at Adobe due to firefox segfaulting on exit
(http://bugs.adobe.com/jira/browse/FP-1431), if I can narrow this
down, I can add another report there.

Thanks in advance.  Any thoughts greatly welcomed!

tom
-- 
Tom London