Parallel Programming using OpenMP
Mike Bailey
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License
Computer Graphics
Copyright By PowCoder代写 加微信 powcoder
openmp.pptx
mjb – March 22, 2021
OpenMP Multithreaded Programming
• OpenMP stands for “Open Multi-Processing”
• OpenMP is a multi-vendor (see next page) standard to perform shared-memory multithreading
• OpenMP uses the fork-join model
• OpenMP is both directive- and library-based
• OpenMP threads share a single executable, global memory, and heap (malloc, new)
• Each OpenMP thread has its own stack (function arguments, function return address, local variables)
• Using OpenMP requires no dramatic code changes
• OpenMP probably gives you the biggest multithread benefit per amount of work you have to put in to using it
Much of your use of OpenMP will be accomplished by issuing C/C++ “pragmas” to tell the compiler how to build the threads into the executable
Computer Graphics
#pragma omp directive [clause]
mjb – March 22, 2021
Who is in the OpenMP Consortium?
Computer Graphics
mjb – March 22, 2021
What OpenMP Isn’t:
• OpenMP doesn’t check for data dependencies, data conflicts, deadlocks, or race conditions. You are responsible for avoiding those yourself
• OpenMP doesn’t check for non-conforming code sequences
• OpenMP doesn’t guarantee identical behavior across vendors or hardware, or even between multiple runs on the same vendor’s hardware
• OpenMP doesn’t guarantee the order in which threads execute, just that they do execute • OpenMP is not overhead-free
• OpenMP does not prevent you from writing code that triggers cache performance problems (such as in false-sharing), in fact, it makes it really easy
Computer Graphics
We will get to “false sharing” in the cache notes
mjb – March 22, 2021
Memory Allocation in a Multithreaded Program
Program Executable
One-thread
Multiple-threads
Don’t take this completely literally. The exact arrangement depends on the operating system and the compiler. For example, sometimes the stack and heap are arranged so that they grow towards each other.
Common Program Executable
Common Globals
Common Heap
Computer Graphics
mjb – March 22, 2021
Using OpenMP on Linux
g++ -o proj proj.cpp -lm -fopenmp
icpc -o proj proj.cpp -lm -openmp -align -qopt-report=3 -qopt-report-phase=vec
Using OpenMP in Microsoft Visual Studio
1. Go to the Project menu → Project Properties
2. Change the setting Configuration Properties → C/C++ → Language →
OpenMP Support to “Yes (/openmp)”
Seeing if OpenMP is Supported on Your System
#ifndef _OPENMP
fprintf( stderr, “OpenMP is not supported – sorry!\n” ); exit( 0 );
Computer Graphics
mjb – March 22, 2021
A Potential OpenMP/Visual Studio Problem
If you are using Visual Studio 2019 and get a compile message that looks like this:
1>c1xx: error C2338: two-phase name lookup is not supported for C++/CLI, C++/CX, or OpenMP; use /Zc:twoPhase-
then do this:
1. Go to “Project Properties“→ “C/C++” → “Command Line“
2. Add /Zc:twoPhase- in “Additional Options” in the bottom section
3. Press OK
Computer Graphics
No, I don’t know what this means either …
mjb – March 22, 2021
Numbers of OpenMP threads
How to specify how many OpenMP threads you want to have available:
omp_set_num_threads( num );
Asking how many cores this program has access to:
num = omp_get_num_procs( );
Actually returns the number of hyperthreads, not the number of physical cores
Setting the number of available threads to the exact number of cores available:
omp_set_num_threads( omp_get_num_procs( ) );
Asking how many OpenMP threads this program is using right now:
num = omp_get_num_threads( );
Asking which thread number this one is:
me = omp_get_thread_num( );
Computer Graphics
mjb – March 22, 2021
Creating an OpenMP Team of Threads
#pragma omp parallel default(none) {
This creates a team of threads
Each thread then executes all lines of code in this block.
Computer Graphics
Think of it this way:
#pragma omp parallel default(none)
mjb – March 22, 2021
Creating an OpenMP Team of Threads
#include
return 0; }
omp_set_num_threads( 8 ); #pragma omp parallel default(none) {
printf( “Hello, World, from thread #%d ! \n” , omp_get_thread_num( ) ); }
Hint: run it several times in a row. What do you see? Why?
Computer Graphics
mjb – March 22, 2021
Second Run
Hello, World, from thread #6 ! Hello, World, from thread #1 ! Hello, World, from thread #7 ! Hello, World, from thread #5 ! Hello, World, from thread #4 ! Hello, World, from thread #3 ! Hello, World, from thread #2 ! Hello, World, from thread #0 !
Hello, World, from thread #0 ! Hello, World, from thread #7 ! Hello, World, from thread #4 ! Hello, World, from thread #6 ! Hello, World, from thread #1 ! Hello, World, from thread #3 ! Hello, World, from thread #5 ! Hello, World, from thread #2 !
Fourth Run
Hello, World, from thread #2 ! Hello, World, from thread #5 ! Hello, World, from thread #0 ! Hello, World, from thread #7 ! Hello, World, from thread #1 ! Hello, World, from thread #3 ! Hello, World, from thread #4 ! Hello, World, from thread #6 !
Hello, World, from thread #1 ! Hello, World, from thread #3 ! Hello, World, from thread #5 ! Hello, World, from thread #2 ! Hello, World, from thread #4 ! Hello, World, from thread #7 ! Hello, World, from thread #6 ! Hello, World, from thread #0 !
Computer Graphics
There is no guarantee of thread execution order!
mjb – March 22, 2021
#include
omp_set_num_threads( NUMT ); …
#pragma omp parallel for default(none)
Creating OpenMP threads in Loops
for( int i = 0; i < arraySize; i++ ) {
The code starts out executing in a single thread
This sets how many threads will be in the thread pool. It doesn’t create them yet, it just says how many will be used the next time you ask for them.
This creates a team of threads from the thread pool and divides the for-loop passes up among those threads
There is an “implied barrier” at the end where each thread waits until all threads are done, then the code continues in a single thread
This tells the compiler to parallelize the for-loop into multiple threads. Each thread automatically gets its own personal copy of the variable i because it is defined within the for-loop body.
The default(none) directive forces you to explicitly declare all variables declared outside the
parallel region to be either private or shared while they are in the parallel region. Variables
declared within the for-loop are automatically private.
Computer Graphics
mjb – March 22, 2021
OpenMP for-Loop Rules
#pragma omp parallel for default(none), shared(...), private(...) for( int index = start ; index terminate condition; index changed )
The index must be an int or a pointer
The start and terminate conditions must have compatible types
Neither the start nor the terminate conditions can be changed during the execution of the loop
The index can only be modified by the changed expression (i.e., not modified inside the loop itself)
You cannot use a break or a goto to get out of the loop There can be no inter-loop data dependencies such as:
a[ i ] = a[ i-1 ] + 1.;
a[101] = a[100] + 1.; // what if this is the last line of thread #0’s work? a[102] = a[101] + 1.; // what if this is the first line of thread #1’s work?
Computer Graphics
mjb – March 22, 2021
for( index = start ;
index < end index <= end ; index > end index >= end
–index ) index += incr
index = index + incr index = incr + index index -= decr
index = index – decr
OpenMP For-Loop Rules
Computer Graphics
mjb – March 22, 2021
What to do about Variables Declared Before the for-loop Starts?
float x = 0.;
#pragma omp parallel for …
for( int i = 0; i < N; i++ ) {
x = (float) i;
float y = x*x;
<< more code... >
private(x)
Means that each thread will get its own version of the variable
Means that all threads will share a common version of the variable
default(none)
I recommend that you include this in your OpenMP for-loop directive. This will force you to explicitly flag all of your externally-declared variables as shared or private. Don’t make a mistake by leaving it up to the default!
Computer Graphics
i and y are automatically private because they are defined within the loop.
Good practice demands that x be explicitly declared to be shared or private!
#pragma omp parallel for default(none), private(x)
mjb – March 22, 2021
For-loop “Fission” 16 Because of the loop dependency, this whole thing is not parallelizable:
x[ 0 ] = 0.;
y[ 0 ] *= 2.;
for( int i = 1; i < N; i++ ) {
x[ i ] = x[ i-1 ] + 1.;
y[ i ] *= 2.; }
But, it can be broken into one loop that is not parallelizable, plus one that is:
x[ 0 ] = 0.;
for( int i = 1; i < N; i++ ) {
x[ i ] = x[ i-1 ] + 1.;
#pragma omp parallel for shared(y) for( int i = 0; i < N; i++ )
y[ i ] *= 2.;
Computer Graphics
mjb – March 22, 2021
Uh-oh, which for-loop do you put the #pragma on?
For-loop “Collapsing”
for( int i = 1; i < N; i++ ) {
for( int j = 0; j < M; j++ ) {
Ah-ha – trick question. You put it on both!
How many for-loops to collapse into one loop
#pragma omp parallel for collapse(2) for( int i = 1; i < N; i++ )
for( int j = 0; j < M; j++ ) {
Computer Graphics
mjb – March 22, 2021
Single Program Multiple Data (SPMD) in OpenMP
#define NUM 1000000
float A[NUM], B[NUM], C[NUM];
total = omp_get_num_threads( );
#pragma omp parallel default(none),private(me),shared(total) {
me = omp_get_thread_num( );
DoWork( me, total ); }
void DoWork( int me, int total )
int first = NUM * me / total; int last = NUM * (me+1)/total for( int i = first; i <= last; i++ ) {
C[ i ] = A[ i ] * B[ i ];
Computer Graphics
mjb – March 22, 2021
OpenMP Allocation of Work to Threads
Static Threads
• All work is allocated and assigned at runtime
Dynamic Threads
• The pool is statically assigned some of the work at runtime, but not all of it • When a thread from the pool becomes idle, it gets a new assignment
• “Round-robin assignments”
OpenMP Scheduling
schedule(static [,chunksize]) schedule(dynamic [,chunksize]) Defaults to static
chunksize defaults to 1
Computer Graphics
mjb – March 22, 2021
2 8,9,10,11
OpenMP Allocation of Work to Threads
#pragma omp parallel for default(none),schedule(static,chunksize) for( int index = 0 ; index < 12 ; index++ )
1 1,4,7,10
2 2,5,8,11
2 4,5,10,11
chunksize = 1
Each thread is assigned one iteration, then the assignments start over
chunksize = 2
Each thread is assigned two iterations, then the assignments start over
chunksize = 4
Each thread is assigned four iterations, then the assignments start over
Computer Graphics
mjb – March 22, 2021
Arithmetic Operations Among Threads – A Problem
#pragma omp parallel for private(myPartialSum),shared(sum) for( int i = 0; i < N; i++ )
float myPartialSum = ...
sum = sum + myPartialSum; }
• There is no guarantee when each thread will execute this line
• There is not even a guarantee that each thread will finish this line before some other thread interrupts it. (Remember that each line of code usually generates multiple lines of assembly.)
• This is non-deterministic !
Assembly code:
Add myPartialSum Store sum
Computer Graphics
What if the scheduler decides to switch threads right here?
Conclusion: Don’t do it this way!
mjb – March 22, 2021
Here’s a trapezoid integration example.
The partial sums are added up, as shown on the previous page. The integration was done 30 times.
The answer is supposed to be exactly 2.
None of the 30 answers is even close.
And, not only are the answers bad, they are not even consistently bad!
0.469635 0.517984 0.438868 0.437553 0.398761 0.506564 0.489211 0.584810 0.476670 0.530668 0.500062 0.672593 0.411158 0.408718 0.523448
0.398893 0.446419 0.431204 0.501783 0.334996 0.484124 0.506362 0.448226 0.434737 0.444919 0.442432 0.548837 0.363092 0.544778 0.356299
Don’t do it this way! We’ll talk about how to it correctly in the Trapezoid Integration noteset.
Computer Graphics
mjb – March 22, 2021
Here’s a trapezoid integration example.
The partial sums are added up, as shown on the previous page. The integration was done 30 times.
The answer is supposed to be exactly 2.
None of the 30 answers is even close.
And, not only are the answers bad, they are not even consistently bad!
Don’t do it this way! We’ll talk about how to it correctly in the Trapezoid Integration noteset.
Computer Graphics
mjb – March 22, 2021
Mutual Exclusion Locks (Mutexes)
omp_init_lock( omp_lock_t * ); omp_set_lock( omp_lock_t * ); omp_unset_lock( omp_lock_t * ); omp_test_lock( omp_lock_t * );
Blocks if the lock is not available
Then sets it and returns when it is available
If the lock is not available, returns 0
If the lock is available, sets it and returns !0
Synchronization
( omp_lock_t is really an array of 4 unsigned chars )
Critical sections
#pragma omp critical
Restricts execution to one thread at a time
#pragma omp single
Restricts execution to a single thread ever
#pragma omp barrier
Forces each thread to wait here until all threads arrive
Computer Graphics
(Note: there is an implied barrier after parallel for loops and OpenMP sections, unless the nowait clause is used)
mjb – March 22, 2021
omp_lock_t Sync; ...
omp_init_lock( &Sync );
omp_set_lock( &Sync );
<< code that needs the mutual exclusion >>
omp_unset_lock( &Sync ); …
while( omp_test_lock( &Sync ) == 0 ) {
Computer Graphics
DoSomeUsefulWork( );
Synchronization Examples
mjb – March 22, 2021
Single-thread-execution Synchronization
#pragma omp single
Restricts execution to a single thread ever. This is used when an operation only
makes sense for one thread to do. Reading data from a file is a good example.
Computer Graphics
mjb – March 22, 2021
Creating Sections of OpenMP Code
Sections are independent blocks of code, able to be assigned to separate threads if they are available.
#pragma omp parallel sections
#pragma omp section
#pragma omp section
(Note: there is an implied barrier after parallel for loops and OpenMP sections, unless the nowait clause is used)
Computer Graphics
mjb – March 22, 2021
What do OpenMP Sections do for You? They decrease your overall execution time.
omp_set_num_threads( 1 );
omp_set_num_threads( 2 );
omp_set_num_threads( 3 );
Computer Graphics
mjb – March 22, 2021
A Functional Decomposition Sections Example
omp_set_num_threads( 3 ); #pragma omp parallel sections
#pragma omp section
Watcher( );
#pragma omp section
Animals( );
#pragma omp section
Plants( );
} // implied barrier — all functions must return to get past here
Computer Graphics
mjb – March 22, 2021
程序代写 CS代考 加微信: powcoder QQ: 1823890830 Email: powcoder@163.com