> Everything has gone well for about 72 hours except for Medm. Medm runs
> fine and then, at some (apparently) random time, begins consuming virtual
> memory at a rate of about 1K bytes per second until all is consumed. There
> is no noticeable change in behavior or performance when Medm gets into this
> state (until the system starts swapping).
>
> I realize that the problem could be in the linux kernel, X libs, or C libs. I
> will try to build the test on another platform. Any comments/suggestions
> are appreciated.
>
>
I remember reading that X programs that continuously call XLib, but do not call XNextEvent()
will continue to allocate memory for each incoming X event until the process runs out of virtual
memory.
Perhaps medm is stuck somewhere where it is calling XLib, but it isn't processing
incoming XEvents. If the CA client lib was sufficiently busy perhaps it is calling
monitor callbacks in medm from ca_pend_event() and some of these callbacks are
making XLib calls, but none of these calls directly or indirectly call XNextEvent()?
Jeff
- Navigate by Date:
- Prev:
request for PMAC records Kevin Tsubota
- Next:
RE: help on epics base 3.13beta11 installation Jeff Hill
- Index:
1994
1995
1996
1997
<1998>
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
- Navigate by Thread:
- Prev:
Re: Portable channel access server on linux Garrett D. Rinehart
- Next:
Marty Kraimers UC Distinguished Performance Award Bill McDowell
- Index:
1994
1995
1996
1997
<1998>
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
|