Re: running out of performer shared memory
Rémi Arnaud (remi++at++remi.engr.sgi.com)
Thu, 26 Jun 1997 10:21:55 -0700 (PDT)
Anita Kishore wrote:
>
> Hi :
>
> I am having trouble with performer shared memory after I put new code
> to pfMalloc (in its shared arena) some structures in pfList. The message I get
> at that line is :
>
> pfMemory::new() Unable to allocate 48 bytes from arena 0x18040000.
> Try using pfSharedArenaSize() to increase the arena size
> (currently 262144.00 KBytes) and check for adequate setrlimit()
> values and available space on swap (or pfTmpDir()).
>
> and the program dumps core after this. The core file also points to the same
> problem.
>
>
> I increased shared arena size to 454288000 bytes (can't increase beyond this).
> But I still get the same message even though it gives current size as the new
> one. The file that I load is very simple and there should be enough memory
> for everything, because very big scenegraph files (minus the new code) have
> so far been loading allright. I am sure that the new allocation is not taking
> more memory than what would be used when loading big scene graphs (unless a
> different arena is used by performer for its nodes?). I made sure that ipcs
> output is clean.
>
> Is there a call to check the shared memory usage from the program? How can I
> further increase the arena size? Any help on this problem is greatly
> appreciated.
>
Looks like you have a memory corruption problem.
Did you try to dmalloc your application ? In case you do not know this
tool, it's available from ftp://sgigate.sgi.com/pub/Performer/src/misc/dmalloc.tar
_ / _ _
|_) _ ._ _ o /\ |_)|\ | /\ | || \
| \(/_| | || /--\| \| \|/--\|_||_/
=======================================================================
List Archives, FAQ, FTP: http://www.sgi.com/Technology/Performer/
Submissions: info-performer++at++sgi.com
Admin. requests: info-performer-request++at++sgi.com
This archive was generated by hypermail 2.0b2
on Mon Aug 10 1998 - 17:55:30 PDT