CSE 522 Real-Time Scheduling (3)

Presentation on theme: "CSE 522 Real-Time Scheduling (3)"— Presentation transcript:

CSE 522 Real-Time Scheduling (3)
Computer Science & Engineering Department Arizona State University Tempe, AZ 85287 Dr. Yann-Hang Lee (480)

Priority Inversion in Synchronization
Blocked 2(M) 3(L) Time 1 :{…P(S1)…V(S1)…} 3 :{…P(S1)…V(S1)…} S1 unlocked Attempt to lock S1 (blocked) S1 locked

Priority Inversion Delay to a task’s execution caused by interference from lower priority tasks is known as priority inversion Priority inversion is modeled by blocking time Identifying and evaluating the effect of sources of priority inversion is important in schedulability analysis Sources of priority Inversion Synchronization and mutual exclusion Non-preemtable regions of code FIFO (first-in-first-out) queues

Accounting for Priority Inversion
Recall that task schedulability is affected by preemption: two types of preemption can occur several times per period can occur once per period execution: once per period blocking: at most once per period for each request to a source The schedulability formulas are modified to add a “blocking” or “priority inversion” term to account for inversion effects

UB Test with Blocking Include blocking while calculating effective utilization for each tasks: Hn Preemption(can hit n times) Execution Blocking H1 Preemption(can hit once)

RT Test with Blocking Blocking is also included in the RT test
Perform test as before, including blocking effect

Example: Considering Blocking
Consider the following example Periodic tasks What is the worst case blocking effect (priority inversion) experienced by each task ? 25 msec 100 msec 50 msec 200 msec 300 msec 10 msec 30 msec Data Structure 1 2 3

Task 2 does not use the data structure. Task 2 does experiences no priority inversion Task 1 shares the data structure with 3 . Task 1 could have to wait for 3 to complete its critical section. But worse, if 2 preempts while 1 is waiting for the data structure, 1 could have to wait for 2’s entire computation. This is the resulting table task Period Execution Time Priority Blocking delay Deadline 1 100 25 High 30+50 2 200 50 Medium 3 300 Low

UB Test for Example UB test with blocking: Not schedulable
with additional RT test, 3 is shown to be schedulable

Synchronization Protocols
No preemption Basic priority inheritance Highest locker’s priority Priority ceiling Each protocol prevents unbounded priority inversion.

Nonpreemption Protocol

Simplicity Use with fixed-priority and dynamic-priority systems No priori knowledge about resource requirement by each task Good when all critical sections are short Disadvantages: Every task can be blocked by every lower priority task, even when there is no resource sharing between the tasks. Blocking time: max(csi)

Basic Inheritance Protocol (BIP)
2:{…P(S1)…V(S1)…} 4:{…P(S1)…V(S1)…} ready 1(H) attempts to lock S1 S1 locked S1 unlocked ready 2 blocked ready 3 S1 locked S1 unlocked 4(L) inherits the priority of 2 after 2 is blocked

Some Notations Ji is the i-th job in Task T.
i = Assigned priority of Job Ji i(t) = current priority of Ji If the decision to change the priority of Job Ji is made at t = t1 then i(t1-) = priority at and immediately before, i(t1+) = priority immediately after the priority change  = nonexistent priority, lower than the lowest priority

Terminology and Assumptions
At time t1, job Ji requests resource Rk. Rk  Jl: Resource Rk is held by Job Jl Ji  Rk: Job Ji is blocked waiting for resource Rk to be released (Ji  Rk  Jl) Scheduling Rules: Ready Jobs scheduled on processors preemptively in a priority driven manner according to their current priorities, i(t). At job release time the priority is equal to its assigned priority if Ji is release at t = t’, then i(t’) = i Resource allocation: If a resource is free then it is allocated when requested if not free then the request is denied and the requesting job is blocked

Priority Inheritance Rules
Scheduling Rule: same as the assumptions Allocation Rule: same as the assumptions Priority-Inheritance Rule: if Ji  Rk  Jl and l(t1-) = priority of Jl at t = t1 then l(t1+) = i(t1) until Jl releases Rk at t2 when l(t2+) = l(t1-) L. Sha, R. Rajkumar, J. Lehoczky, “Priority Inheritance Protocols: An Approach to Real-Time Synchronization”, IEEE Transactions on Computers, Vol. 39, No. 9, pp , 1990

Properties of Priority Inheritance
For each resource (semaphore), a list of blocked tasks must be stored in a priority queue. A task (job) ti uses its assigned priority, and uses (inherits) the highest dynamic priority of all the tasks it blocks when it is in its critical section and blocks some higher priority tasks. Priority inheritance is transitive; that is, if taskti blocks tj and tj blocks tk , then ti can inherit the priority of tk. When taskti releases a resource, which priority it should use? Chained blocking if requesting multiple resources (nested mutex requests) Direct blocking and indirect (inheritance) blocking (when the lower priority task inherits the higher priority task’s priority).

Implementation Issues of BIP
PI semaphore and basic semaphore Priority is changed when acquiring and releasing a lock When release a lock, can the priority be restore to the one before acquiring? Need to maintain a list of PI semaphores locked by a task when holding multiple locks and release one when a PI semaphore is deleted when a task waiting for a PI semaphore quits the waiting (due to timeout) when the priority of a task waiting for a PI semaphore is changed Note that for each semaphore, a queue for all waiting tasks

Example Of Chained Blocking (BIP)
1:{…P(S1)…P(S2)…V(S2)…V(S1)…} 2:{…P(S1)…V(S1)…} 3:{…P(S2)…V(S2)…} S2 locked S2 unlocked 3(L) 1(H) 2(M) B Attempts to lock S1(blocked) Attempts to lock S2(blocked) S1 locked S1 unlocked

Attempts to lock S1 (blocked)
Deadlock: Using BIP 1 :{…P(S1)…P(S2)…V(S2)…V(S1)..} 2 :{…P(S2)…P(S1)…V(S1)…V(S2)..} Attempts to lock S2 (blocked) Locks S1 1 (H) B Attempts to lock S1 (blocked) S2 locked 2 (M)

Blocking Time Under BIP
Example T1 = {.. P(A) .3. P(B) .2. V(B) .1. V(A) ..} T2 = {.. P(C) .2. V(C) ..} T3 = {.. P(A) .1. P(B) .2. V(B) .2. V(A) .. } T4 = {.. P(A) .1. P(C) .1. P(B) .3. V(B) .1. V(C) .1. V(A).. } direct blocking by indirect blocking by blocking time T2 T3 T4 T1 N Y 5 7

Priority Ceiling Protocol (PCP)
2:{…P(S1)…V(S1)…}  3:{…P(S2)…V(S2)…} 4:{…P(S1)…V(S1)…} 1(H) attempts to lock S1 ready S1 locked S1 unlocked B 2 attempts to lock S2 ready S2 locked 3 blocked by ceiling S1 locked S1 unlocked 4(L)

Basic Priority Ceiling Rules (1)
(R) = priority ceiling of resource R – the highest priority of the tasks that request R S(t) = system priority ceiling -- the highest priority ceiling of the resources that are in use at time t Scheduling Rule: same as the assumptions Allocation Rule: if Ji  Rk  Jl at t = t1 then block Ji (no change) Rk free at t1, if i(t1) > S(t1), then Rk  Ji else (i.e. i(t1)  S(t1) ) if for some Rx  Ji and (Rx) = S(t1), then Rk  Ji [ Ji holds a resource Rx whose priority ceiling is S(t1) ] else deny and block (Ji  Rk)

Basic Priority Ceiling Rules (2)
Priority-Inheritance Rule: if Ji  Rk at t = t1 and is blocked by Jl (and l(t1-) = priority of Jl) either Rk  Jl, (Jl holds the resource Rk) or Jl  Rx and (Rx) = S(t1)  i(t1) then l(t1+) = i(t1) (inherited priority) until Jl releases all Rx with (Rx)  i(t1), l(t2+) = l(t1-) at t = t2.

Blocking in PCP A task TH can be blocked by a lower-priority task TL in three ways: directly, i.e. when TL inherits a priority higher than the priority pH of TH. When TH requests a resource the priority ceiling of resources held by TL is equal to or higher than pH: TH X TL request for allocated to T  TH X TL (p > pH) TH Y X TL (pH  PX)

Blocked At Most Once (PCP)
1:{…P(S1)…P(S2)…V(S2)…V(S1)…} 2:{…P(S1)…V(S1)…} 3:{…P(S2)…V(S2)…} S2 locked S2 unlocked 3(L) 1(H) 2(M) attempts to lock S1(blocked) B S1 locked S1 unlocked

1 :{…P(S1)…P(S2)…V(S2)…V(S1)..} 2 :{…P(S2)…P(S1)…V(S1)…V(S2)..} Locks S2 Unlocks S2 Attempts to lock S1 (blocked) Locks S1 Unlocks S1 1 (H) C Locks S2 Locks S1 Unlocks S1 Unlocks S2 2 (M)

Stack Sharing T1 T1 Ti Ti Tn Tn
Sharing of the stack among tasks eliminates stack space fragmentation and so allows for memory savings: However: Once job is preempted, it can only resume when it returns to be on top of stack. Otherwise, it may cause a deadlock. Stack becomes a resource that allows for “one-way preemption”. T1 T1 Ti Ti Tn Tn no stack sharing stack sharing

Stack-Based Priority Ceiling Protocol
To avoid deadlocks: Once execution begins, make sure that job is not blocked due to resource access allow preemption only if the priority is higher than the ceiling of the resources in use Update Current Ceiling in the usual manner If no resource allocated, S (t) =  Scheduling Rule: Ji released and blocked until i(t) > S(t) When not blocked jobs are scheduled in the usual manner. Allocation Rule: Allocate when requested

Stack-Based PCP (cont)
The Stack-Based Priority-Ceiling Protocol is deadlock-free: When a job begins to execute, all the resources it will ever need are free. Otherwise, S(t) would be higher or equal to the priority of the job. Whenever a job is preempted, all the resources needed by the preempting job are free. The preempting job can complete, and the preempted job can resume. Worst-case blocking time of Stack-Based Protocol is the same as for Basic Priority Ceiling Protocol. Stack-Based Protocol smaller context-switch overhead 2 context switches since once execution starts, job cannot be blocked (may be preempted) 4 context switches for PCP since a job may be blocked at most once

SRP + EDF Let 1, 2, , n denote the tasks, and R1,R2, , Rm denote the non-preemptable resources. Each resource Rj is statically assigned a ceiling (Rj), which is set equal to the minimum relative deadline parameter of any task that may access it:  (Rj) = min{Di | i accesses Rj} A system ceiling is computed each instant during runtime. This is set equal to the minimum ceiling of any resource that is currently being held by some job. At any instant in time, a job generated by i may begin execution only if it is the earliest-deadline active job, and Di is strictly less than the system ceiling. (From “Resource sharing in EDF-scheduled systems: a closer look,” by Sanjoy K. Baruah, RTSS’06)

Ceiling-Priority Protocol
Re-formulation of stack-based priority ceiling protocol for multiple stacks (no stack-sharing) Update Current Ceiling in the usual manner Scheduling Rule: No resources held by Ji, i(t) = i Resource held by Ji,i(t) = max((Rx)) for all resource Rx held by Ji FIFO scheduling among jobs with equal priority Allocation Rule: Allocate when requested Also called: Priority ceiling emulation, highest locker protocol, Immediate priority ceiling protocol (ICPP)

Comparison Worst case performance of stack-based and basic ceiling protocols are the same Stack-based version supported by the Real-Time systems Annex of Ada95 Jobs must not self-suspend When jobs do not self-suspend, stack-based and ceiling-priority protocols yield the same schedules. Stack-based and ceiling-priority have the same worst-case blocking time.

Highest Locker’s Priority Protocol
2:{…P(S1)…V(S1)…} 4:{…P(S1)…V(S1)…} ready S1 locked S1 unlocked 1(H) 4(L) 2 3 B run with the priority of 2

Summary of Synchronization Protocols
Bounded Priority Inversion Blocked at Most Once Deadlock Avoidance Nonpreemptible critical sections Yes Yes1 Highest locker's priority Basic inheritance No Priority Ceiling Yes2 1 Only if tasks do not suspend within critical sections 2 PCP is not affected if tasks suspend within critical sections.

BIP and PCP BIP CPP (highest locker protocol)
no penalty when the locks are not contended, which covers the vast majority of time-constrained systems. many extra context switches are avoided, and medium priority tasks are not preempted unnecessarily, excellent average performance CPP (highest locker protocol) change a thread's priority twice regardless of whether there is contention for the lock or not higher overhead and many unnecessary context switches and blocking in unrelated tasks worst-case priority inversion control, it handles nested locks well, and can avoid deadlock in some cases static analysis of the system to find the priority ceiling