Gireesh, At a guess, I'd say those are probably thread stacks. You're measuring the wrong thing - the virtual size of the application. You don't really need to care about the virtual size - these thread stacks don't get faulted into actual physical ram. Mike On Mon, Apr 14, 2008 at 1:13 AM, Gireesh Kumar M <gireesh.mnair at gmail.com> wrote: > Hi, > > I'm running GStreamer, built with CodeSourcery arm toolchain [version: > (CodeSourcery Sourcery G++ Lite 2007q3-51) 4.2.1], on TI's OMAP board. > The runtime memory consumption (as reported by 'top' command) is 38% on a > 128MB RAM (~48MBytes), which is very high. Verifying the memory map log > (using pmap), I found the presence of some un-named memory blocks of size > 8188 KBytes which contributes major share to the 38% RAM usage (PMAP log > attached). Is this a GStreamer bug or a memory leak? Can anyone help me to > fix this? > > Currently I'm using the mp3 playback with the following pipeline > $gst-launch filesrc location=test.mp3 ! queue ! mad ! alsasink > > Any help on this is highly appreciated. > > Thank you, > Gireesh > > > ------------------------------------------------------------------------- > This SF.net email is sponsored by the 2008 JavaOne(SM) Conference > Don't miss this year's exciting event. There's still time to save $100. > Use priority code J8TL2D2. > > http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone > _______________________________________________ > Gstreamer-embedded mailing list > Gstreamer-embedded at lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/gstreamer-embedded > >
