Hi Rahul, Thank you for the input. I ill get back after checking.
Thanks and Regards Vikram Upparpalli Telephone: 040-3063 3535 Mobile: 9985152191 [cid:[email protected]] From: rahul p [mailto:[EMAIL PROTECTED] Sent: Wednesday, October 22, 2008 1:04 PM To: Vikram_Upparpalli Cc: Peter Teoh; [email protected] Subject: Re: sched summary On 10/22/08, Vikram_Upparpalli <[EMAIL PROTECTED]<mailto:[EMAIL PROTECTED]>> wrote: Hi Peter, Ps: Pardon me if my understanding is wrong. OS is VRTX (Versatile Real Time Execute) from the Mentor Graphics. Yes it is an RTOS.... where the timeliness is determined already... like I have decided to run one task (say X) periodically once in 250ms. So I would assign a priority P for this task X. Confusion is about the timeliness. Like if I have already determined the priority P for this task X and the system works fine with the priority P to decide running the task X. what if I want to add one new task (there is a system call through which I can create a task from some application code dynamically). Wouldn't the timeliness be disturbed? My view of the running tasks is - The priority P assigned to this task X is such that after the every 250ms, the tasks being executed would definitely have the priority P' where P' < P. so after every 250ms, what ever task is running could be swapped and task X could be run since the priority resolves the issue. Now I am adding one new task with some priority (say also < P). would the system still work well as earlier? Though the new task has priority which is < P, would the timeliness still be maintained? The priority of the new task is _less_than_ the priority of current task means that the new task needs to wait for it's chance to be picked by the scheduler. I guess you know how O(1) scheduler picks the next task to run via priority array. In your case all active(runnable) tasks of higher priority that didn't use up their timesclice has to run first. The tasks of same priority are scheduled in a round-robin manner. I hope this helps. Also a higher priority task that is sleeping on some event to happen preempts the currrent lower priority task when it wakes up. - Rahul Pydimukkala -- Smart data structures and dumb code works a lot better than the other way around. -Eric S Raymond ________________________________ DISCLAIMER: This email (including any attachments) is intended for the sole use of the intended recipient/s and may contain material that is CONFIDENTIAL AND PRIVATE COMPANY INFORMATION. Any review or reliance by others or copying or distribution or forwarding of any or all of the contents in this message is STRICTLY PROHIBITED. If you are not the intended recipient, please contact the sender by email and delete all copies; your cooperation in this regard is appreciated.
<<inline: image001.gif>>
