Max-Planck-Institut für Informatik
max planck institut
informatik
mpii logo Minerva of the Max Planck Society
 

MPI-INF or MPI-SWS or Local Campus Event Calendar

<< Previous Entry Next Entry >> New Event Entry Edit this Entry Login to DB (to update, delete)
What and Who
Title:Blocking Analysis of Spin Locks under Partitioned Fixed-Priority Scheduling
Speaker:Alexander Wieder
coming from:Max Planck Institute for Software Systems
Speakers Bio:
Event Type:SWS Student Defense Talks - Thesis Defense
Visibility:SWS
We use this to send out email in the morning.
Level:Public Audience
Language:English
Date, Time and Location
Date:Wednesday, 29 November 2017
Time:16:00
Duration:-- Not specified --
Location:Saarbr├╝cken
Building:E1 5
Room:029
Abstract
Partitioned fixed-priority scheduling is widely used in embedded
multicore real-time systems. In multicore systems, spin locks are one
well-known technique used to synchronize conflicting accesses from
different processor cores to shared resources (e.g., data structures).
The use of spin locks can cause blocking. Accounting for blocking is a
crucial part of static analysis techniques to establish correct temporal
behavior.

In this thesis, we consider two aspects inherent to the partitioned
fixed-priority scheduling of tasks sharing resources protected by spin
locks: (1) the assignment of tasks to processor cores to ensure correct
timing, and (2) the blocking analysis required to derive bounds on the
blocking.

Heuristics commonly used for task assignment fail to produce assignments
that ensure correct timing when shared resources protected by spin locks
are used. We present an optimal approach that is guaranteed to find such
an assignment if it exists (under the original MSRP analysis). Further,
we present a well-performing and inexpensive heuristic.

For most spin lock types, no blocking analysis is available in prior
work, which renders them unusable in real-time systems. We present a
blocking analysis approach that supports eight different types and is
less pessimistic than prior analyses, where available. Further, we show
that allowing nested requests for FIFO- and priority-ordered locks
renders the blocking analysis problem NP-hard.
Contact
Name(s):
Video Broadcast
Video Broadcast:YesTo Location:Kaiserslautern
To Building:G26To Room:111
Tags, Category, Keywords and additional notes
Note:
Attachments, File(s):
Created by:Maria-Louise Albrecht/MPI-KLSB, 11/21/2017 10:41 AMLast modified by:Maria-Louise Albrecht/MPI-KLSB, 11/21/2017 10:43 AM
  • Maria-Louise Albrecht, 11/21/2017 10:43 AM -- Created document.