Bug#664998: Too much CPU used

[Miguel Colon]

I cannot reproduce that on my personal box. I have a Duron at 800 MHz with 
debian squeeze on it. Running mpg123 -v needs around 1% of CPU! There must be 
something else going on. Also please note that nowadays, the audio output 
happily gobbles up way more CPU than mpg123 needs. An example is the ALSA 
library resampling 44100 Hz material to 48000 Hz.
Can you produce a profile of your CPU-consuming mpg123 (using the perf tool, 
for example)?

Hello Santiago:

One of the upstream authors replied to your bug report but forgot to CC you.

This is email is just to relay the message.

Cheers,
Miguel



--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org

Bug#664998: Too much CPU used

[Thomas Orgis]

I cannot reproduce that on my personal box. I have a Duron at 800 MHz with 
debian squeeze on it. Running mpg123 -v needs around 1% of CPU! There must be 
something else going on. Also please note that nowadays, the audio output 
happily gobbles up way more CPU than mpg123 needs. An example is the ALSA 
library resampling 44100 Hz material to 48000 Hz.

Can you produce a profile of your CPU-consuming mpg123 (using the perf tool, 
for example)?


Alrighty then,

Thomas


signature.asc
Description: PGP signature

Bug#664998: Too much CPU used

[Santiago Garcia Mantinan]

Package: mpg123
Version: 1.12.1-3.2
Severity: wishlist

It is long time since I last used mpg123, but it used to be the lowest
consuming of all players, today I played a song (typical 128 kb/s joint
stereo encoding) and saw it take 23% CPU on my  AMD Athlon(tm) 64 Processor
3000+, I have just played the very same file under audacious and it was
always under 3%, so I guess something is wrong here.

I tried to investigate, as I saw it was using x86-64 (SSE) for decoding, to
see if I was able to select a more standard decoder, but it seems that the
only ones compiled on amd64 are: x86-64 generic generic_dither
and choosing generic was even worse, 31% of CPU eaten.

Wow, I have done more tests right now and saw that this was due to using -v,
I mean, if I play it without -v it takes under 3% and if I run it with -v it
goes to 30-40%.

I was reporting this bug thinking it was a problem in our building of the
package, but it seems that there is a problem in how -v is implemented, thus
I'd say it is an upstream thing.

I don't know what else to add here, if you are not able to reproduce it or
need more info just add for it.

Regards.

-- System Information:
Debian Release: wheezy/sid
  APT prefers unstable
  APT policy: (500, 'unstable'), (500, 'testing'), (500, 'stable'), (1, 
'experimental')
Architecture: amd64 (x86_64)

Kernel: Linux 3.2.0-2-amd64 (SMP w/1 CPU core)
Locale: LANG=gl_ES.UTF-8, LC_CTYPE=gl_ES.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash

Versions of packages mpg123 depends on:
ii  libc62.13-27
ii  libltdl7 2.4.2-1
ii  libmpg123-0  1.12.1-3.2

Versions of packages mpg123 recommends:
ii  libasound2  1.0.25-2
ii  oss-compat  2

Versions of packages mpg123 suggests:
ii  libasound21.0.25-2
ii  libaudio2 1.9.3-4
ii  libesd0   0.2.41-10+b1
ii  libjack0 [libjack-0.116]  1:0.121.3+20120222git81e8f031-1
ii  libpulse0 1.1-3

-- no debconf information



-- 
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org