Hi all,
I am writing a special scheduler based on RTLinux. It schedule realtime task
based on a task relation graph.
1. It first select runable tasks and link them.
2. Alloc CPU for the tasks. Tasks which is alloced to other CPU the
task is send to the CPU use IPI.
3. Tasks which are selected to run locally will be inserted into
sched->rt_tasks, then rtl_schedule is called.
The program run smoothly on UP, but on MP it will dead sometimes, and
WITHOUT OOPS MESSAGE.
At first time I give all processor the same priority, thus there are a lot
of IPI happen. The program will surely dead sliently. Then I changed to let
local processor get higher priority. Thus by far the system has not dead,
but I think the bug remains in it.
Can you give some advice about which kind of problem it maybe?(It dead after
it back to Linux)
Any response is apreciated.
Thank you.
-- [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/
