Amirkabir University of

Technology
(Tehran Polytechnic)
Department of Computer Engineering and Information
Technology

Threads (‫ نخ‬،‫)ریسمان‬

Hamid R. Zarandi
h_zarandi@[Link]

Hamid R. Zarandi Amirkabir Univ. of Tech. (Tehran Polytechnic) 2018/10/16

 Operating Systems

Definition
A basic unit of CPU utilization
o Private: Thread ID, program counter, register set, stack
o Shared: code section, data section, OS resources (IO & file)

Examples:
o Web browsers
o Word processors
o Database engines
o RPC!

Versus Process:
o Time consuming
o Resource intensive

Hamid R. Zarandi Amirkabir Univ. of Tech. (Tehran Polytechnic) 2015/10/03 2

 Operating Systems

Web server application

Hamid R. Zarandi Amirkabir Univ. of Tech. (Tehran Polytechnic) 2015/10/03 3

 Operating Systems

Advantages of using threads

Responsiveness
o Allowing a program to continue running even part of it is blocked or
lengthy

Resource sharing
o Memory, resources

Economy
o Fast

Scalability
o Threads may be running in parallel on processing cores
Hamid R. Zarandi Amirkabir Univ. of Tech. (Tehran Polytechnic) 2015/10/03 4
 Operating Systems

Multicore programming
Multicore or multiprocessor systems putting pressure on programmers,
challenges include:
o Dividing activities
o Balance
o Data splitting
o Data dependency
o Testing and debugging

Parallelism implies a system can perform more than one task simultaneously

Concurrency supports more than one task making progress

o Single processor / core, scheduler providing concurrency

Hamid R. Zarandi Amirkabir Univ. of Tech. (Tehran Polytechnic) 2015/10/03 5

 Operating Systems

Multicore programming
Concurrent execution on single-core system:

Parallelism on a multi-core system:

Hamid R. Zarandi Amirkabir Univ. of Tech. (Tehran Polytechnic) 2015/10/03 6

 Operating Systems

Types of parallelism
Types of parallelism
o Data parallelism
o Task parallelism

As # of threads grows, so does architectural support for threading

o CPUs have cores as well as hardware threads
o Consider Oracle SPARC T4 with 8 cores, and 8 hardware threads per core

Hamid R. Zarandi Amirkabir Univ. of Tech. (Tehran Polytechnic) 2015/10/03 7

 Operating Systems

RPC using threads

Hamid R. Zarandi Amirkabir Univ. of Tech. (Tehran Polytechnic) 2015/10/03 8

 Operating Systems

User threads and kernel threads

User threads - management done by user-level threads library
Three primary thread libraries:
o POSIX Pthreads (kernel-level lib, user-level lib)
o Windows threads (kernel-level lib)
o Java threads (kernel-level lib)

Kernel threads - Supported by the Kernel

Asynchronous vs. synchronous threading

o Parent & child threads

Hamid R. Zarandi Amirkabir Univ. of Tech. (Tehran Polytechnic) 2015/10/03 9

 Operating Systems

User level vs. kernel level threads

Hamid R. Zarandi Amirkabir Univ. of Tech. (Tehran Polytechnic) 2015/10/03 10

 Operating Systems

Multithreading models
Many-to-One

One-to-One

Many-to-Many

Hamid R. Zarandi Amirkabir Univ. of Tech. (Tehran Polytechnic) 2015/10/03 11

 Operating Systems

Many-to-one
 Many user-level threads mapped to single kernel thread

 One thread blocking causes all to block

 Multiple threads may not run in parallel on multicore

system because only one may be in kernel at a time

 Few systems currently use this model

 Examples:
o Solaris Green Threads
o GNU Portable Threads

 Used in very few systems.

Hamid R. Zarandi Amirkabir Univ. of Tech. (Tehran Polytechnic) 2015/10/03 12

 Operating Systems

One-to-one
Each user-level thread maps to kernel thread

Creating a user-level thread creates a kernel

thread

More concurrency than many-to-one

Number of threads per process sometimes

restricted due to overhead

Examples
o Windows
o Linux
o Solaris 9 and later

Hamid R. Zarandi Amirkabir Univ. of Tech. (Tehran Polytechnic) 2015/10/03 13

 Operating Systems

Many-to-many model
Allows many user level threads to be
mapped to many kernel threads

Allows the operating system to create a

sufficient number of kernel threads

Two-level Model:
o Similar to M:M, except that it allows a user thread
to be bound to kernel thread

Hamid R. Zarandi Amirkabir Univ. of Tech. (Tehran Polytechnic) 2015/10/03 14

 Operating Systems

Thread operations and states

Spawn
o When a new process is spawned, a thread for that process
is also spawned

Block
o When a thread needs to wait for an event, it will block

Unblock
o When the event for which a thread is blocked occurs, the thread is
moved to the Ready queue

Finish
o When a thread completes, its register context and stacks are deallocate

Hamid R. Zarandi Amirkabir Univ. of Tech. (Tehran Polytechnic) 2015/10/03 15

 Operating Systems

Pthread: POSIX thread

Hamid R. Zarandi Amirkabir Univ. of Tech. (Tehran Polytechnic) 2015/10/03 16

 Operating Systems

Pthreads code for joining

Pthreads 10 threads
Code for Joining 10 Threads

Hamid R. Zarandi Amirkabir Univ. of Tech. (Tehran Polytechnic) 2015/10/03

Silberschatz, Galvin and Gagne ©2013
17
th 4. 21
 Operating Systems

Windows multithread C program

Hamid R. Zarandi Amirkabir Univ. of Tech. (Tehran Polytechnic) 2015/10/03 18

 Operating Systems

Java thread programming

Hamid R. Zarandi Amirkabir Univ. of Tech. (Tehran Polytechnic) 2015/10/03 19

 Operating Systems

Implicit threading
Three methods explored
o Thread Pools (Win)
o OpenMP (C lib)
o Grand Central Dispatch (Mac OS, iOS)

Block is in “^{ }” - ˆ{ printf("I am a block"); } #pragma omp parallel for

for(i=0;i<N;i++) {
c[i] = a[i] + b[i];
}
Hamid R. Zarandi Amirkabir Univ. of Tech. (Tehran Polytechnic) 2015/10/03 20
 Operating Systems

Thread-local storage
Thread-local storage (TLS) allows each thread to have its own copy of data

Useful when you do not have control over the thread creation process (i.e.,
when using a thread pool)

Different from local variables

o Local variables visible only during single function invocation
o TLS visible across function invocations

Similar to static data

o TLS is unique to each thread

Hamid R. Zarandi Amirkabir Univ. of Tech. (Tehran Polytechnic) 2015/10/03 21

 Operating Systems

Thread termination
Thread cancelation
o Asynchronous cancellation
o Deferred cancellation

Who is “target thread”?

Hamid R. Zarandi Amirkabir Univ. of Tech. (Tehran Polytechnic) 2015/10/03 22

 Operating Systems

Windows threads data structures

Implements the one-to-one mapping, kernel-level

Each thread contains

o A thread id
o Register set representing state of processor
o Separate user and kernel stacks for when thread runs in
user mode or kernel mode
o Private data storage area used by run-time libraries and
dynamic link libraries (DLLs)

The register set, stacks, and private storage area are

known as the context of the thread

Data structures:
o Execution thread block, kernel thread block and thread
environment block
Hamid R. Zarandi Amirkabir Univ. of Tech. (Tehran Polytechnic) 2015/10/03 23
 Operating Systems

Linux threads
 Linux refers to them as tasks rather than threads

 Thread creation is done through clone() system call

 clone() allows a child task to share the address space of the parent task (process)
o Flags control behavior

 struct task_struct points to process data structures (shared or unique)

Hamid R. Zarandi Amirkabir Univ. of Tech. (Tehran Polytechnic) 2015/10/03 24

 Operating Systems

Questions?

Hamid R. Zarandi Amirkabir Univ. of Tech. (Tehran Polytechnic) 2015/10/03 25