On 3/21/19 7:23 PM, Ken Moffat via blfs-dev wrote:
On Thu, Mar 21, 2019 at 09:35:36AM -0500, Bruce Dubbs via blfs-dev wrote:
On 3/21/19 6:19 AM, Tim Tassonis via blfs-dev wrote:

I just noticed on my development system, where nothing has been used for
about nine hours, that pulseaudio is consuming 100% of one core.  t did have
xfce running with one konsole instance but nothing else that could use
audio.  I don't recall running any browser there at all.  I exited xfce and
pulseaudio terminated.  I am running one other app, gkrellm, on the
development system over ssh and that is where I noticed the cpu usage.

I get that sometimes, particularly when changing to different
systems (/home is shared) on the machine where I (still) have the
youtube-alsa problem.  When I notice, I kill pulse and wipe out its
~/.config files.  Obviously, for testing ff/alsa I want it stopped
anyway :)

NB I start pulse from .xinitrc because I don't have a DE to start
it.  On icewm, I eventually notice the red line at the bottom of the
cpu window on the 'panel'.  In xfce I think I would notice it in the
cpu monitor.

Meanwhile, and possibly related to Tim's later reply about a later
kernel, on my haswell I've been seeing very slow or bursty network
i/o causing terms to hang for several seconds (e.g. downloading a
large package and saving it on nfs, or writing backups to nfs, or
even just untarring firefox from nfs), that seemed to start a month
or so back but was ok with _some_ kernels.  Currently I've just
built LFS-svn and am now running 5.0.2 (it's ok, it's an intel -
I've seen a problem report for that version on amd) building Xorg.
Will have to wait and see if the network sluggishness persists.


Just testing firefox 66.0 with alsa and kernel 4.19.30 on a Dell Latitude E7240 and experiencing absolutely no performance issues, as I type this mail into thunderbird 60.6.0.

I gues my problem was somehow kernel/disk related.


Bye
Tim
--
http://lists.linuxfromscratch.org/listinfo/blfs-dev
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Reply via email to