I changed my test program from using memset to write a constant value to memory
to using my pid as a seed to rand and then filling memory with random numbers.
Now when I run multiple copies, the first one gets 50% of physical memory but
the additional copies stop with less.

I tried the modified test program with my simulation app running and things got
nasty : First my sim app reported out of memory, then update (bdflush) reported
out of memory, then a killed message appeared. Linux killed my simulation app
and gdb couldn't catch it so I couldn't find out what part of the simm app was
trying to allocate more memory.


--- [rtl] ---
To unsubscribe:
echo "unsubscribe rtl" | mail [EMAIL PROTECTED] OR
echo "unsubscribe rtl <Your_email>" | mail [EMAIL PROTECTED]
----
For more information on Real-Time Linux see:
http://www.rtlinux.org/~rtlinux/

Reply via email to