Hello everyone,
I'm an EE student at Sogang University Korea.
I have recently submitted a paper on parallel loop scheduling algorithm and had
to modify libgomp a bit in the process.
It is known that the...
/* For now map to schedule(static), later on we could play with feedback driven
choic
e some work.
Also, out of curiosity, is there any plan to add work stealing (affinity
schedules) to gomp?
The clang implementation seem have work stealing.
Ray Kim
-Original Message-
From: "Jakub Jelinek"
To: "김규래";
Cc: "gcc Mailing List";
Sent: 2019-03
ects we are discussing (Additional scheduling
algorithms, doing something about auto, WS etc..)
could make the cut for a GSoC 2019 project?
Ray Kim
-Original Message-
From: "Jakub Jelinek"
To: "김규래";
Cc: "gcc Mailing List";
Sent: 2019-03-02 (토) 02:46:14 (GM
> > > It is on the wish list, but I'm afraid I won't have cycles for it in the
> > > next year or two at least (once GCC 9 is released, I need to work on the
> > > remaining OpenMP 5.0 features). Of course if somebody implements it and
> > > submits
> > > and it is of reasonable quality/performan
Hello,
I was trying to write my application for GSoC 2019 but couldn't find an
application form for GCC.
Some other organizations seem to have GSoC application forms/templates.
The gcc GSoC wiki page doesn't mention any.
Is the application format free? or did I miss it somewhere?
If it is free, is
ent/d/1UwrBOsiAXsZSClsq1r6tAEw6QdqUzJvgW74SQpcWlIU/edit?usp=sharing
Ray Kim.
-Original Message-
From: "Martin Jambor"
To: "김규래"; ;
Cc: "Jakub Jelinek";
Sent: 2019-03-26 (화) 08:32:51 (GMT+09:00)
Subject: Re: [GSoC 2019] No application template?
Hi,
On Sun, Mar 24 2019, 김규래 wrote:
Hi everyone,
Thanks for accepting my proposal Implementing OpenMP Work Stealing Scheduling
for GSoC 2019.
I'll do my best to accomplish the proposed goals.
Right now the semester hasn't ended so progress would be limited.
However, I'll start working on the project as soon as the semester ends.
Dur
Hi everyone,
Just wanted to share some quick baseline benchmark results [3].
I ran LU decomposition on a AMD Ryzen Threadripper 1950x 16C/32T system.
LAPACK is currently plain loop parallel BLAS as far I believe.
And the upstream version of PLASMA uses OpenMP tasks [1].
The colored region is the 9
Hi,
My bad.
I broke the link capitalization by mistake.
This is the correct link:
http://imgur.com/YsxS5Ol
Ray Kim
-Original Message-
From: "Martin Jambor"
To: "김규래"; ;
Cc: ;
Sent: 2019-05-30 (목) 18:11:31 (GMT+09:00)
Subject: Re: [GSoC'19, libgomp work ste
Hi,
I've been studying the libgomp task parallelism system.
I have a few questions.
First, Tracing the events shows that only the main thread calls GOMP_task.
How do the other worker threads enter the libgomp runtime?
I can't find the entry point of the worker threads from the event tracing and
th
correct, I guess the task priority should be given
some special treatment?
Ray Kim
-Original Message-
From: "Jakub Jelinek"
To: "김규래";
Cc: ;
Sent: 2019-06-04 (화) 03:21:01 (GMT+09:00)
Subject: Re: [GSoC'19, libgomp work-stealing] Task parallelism runtime
On T
> On Wed, Jun 5, 2019 at 9:25 PM 김규래 wrote:
>
> > Hi, thanks for the detailed explanation.
> > I think I now get the picture.
> > Judging from my current understanding, the task-parallelism currently works
> > as follows:
> > 1. Tasks are placed in a global
> Another option, which I guess starts to go out of scope of your gsoc, is
> parallel depth first (PDF) search (Blelloch 1999) as an alternative to work
> stealing. Here's a presentation about some recent work in this area,
> although for Julia and not OpenMP (no idea if PDF would fit with OpenMP a
Hi,
Currently, gcc (specifically libgomp) contains some amount of lock-free code.
Does gcc test for the correctness of the lock-free sections?
Or is it tested just as any other code?
Ray Kim
Hi,
I'm not very familiar with the gomp plugin system.
However, looking at 'GOMP_PLUGIN_target_task_completion' seem like tasks have
to go in and out of the runtime.
In that case, is it right that the tasks have to know from which queue they
came from?
I think I'll have to add the id of the corre
Hi everyone,
I'll share my status for GSoC first evaluation.
Current status of libgomp task system:
I'll first summarize my understanding of libgomp.
Please correct me if I'm wrong.
Currently libgomp has 3 different queues: children_queue, taskloop_queue and
team_queue.
These three queues are pr
> > Implementation of competing systems:
> > The intel OpenMP implementation [1] is simpler.
> > It uses a single queue for each thread and a single subroutine for
> > dequeuing and executing the tasks [2, 3].
> > The taskgroup tasks and childen tasks are only counted (not queued) [4, 5,
> > 6].
Hi,
This is an update about my status.
I've been working on unifying the three queues into a single queue.
I'm almost finished and passed all the tests except for the dependency handling
part.
Ray Kim
Hi Jakub,
About the snippet below,
if (gomp_barrier_last_thread (state))
{
if (team->task_count == 0)
{
gomp_team_barrier_done (&team->barrier, state);
gomp_mutex_unlock (&team->task_lock);
gomp_team_barrier_wake (&team->barrier, 0);
return;
}
gomp_team_barrier_set_wa
Hi,
Just submitted a WIP patch for my current status.
I've finished unifying the three queues and reducing the execution paths.
From now on, I will reduce the locked region so that in the end, only the queue
accesses are locked.
Once this is done splitting the queues and implementing work-stealing
wanted to be sure that's the general case.
Thanks.
Ray Kim
-Original Message-
From: "Jakub Jelinek"
To: "김규래";
Cc: ;
Sent: 2019-07-23 (화) 03:54:13 (GMT+09:00)
Subject: Re: Re: [GSoC'19, libgomp work-stealing] Task parallelism runtime
On Sun, Jul 21, 2019 a
Hi,
I'm currently having trouble implementing the thread sleeping mechanism when
the queue is out of tasks.
Problem is, it's hard to maintain consistency between the thread sleeping
routine and the queues.
See the pseudocode below,
1. check queue is empty
2. go to sleep
if we go lock-free, th
> I thought we don't want to go lock-free, the queue operations > aren't easily
> implementable lock-free, but instead with a lock for each of > the queues,
Hi,
By lock-free I meant to use locks only for the queues,
But my terminology was indeed confusing sorry about that.
> mean we don't in so
Hi Jakub,
I think the current semaphore sleep system ought to be improved.
I'm not sure how but since the GSoC deadline is approaching I'll just post the
results without the semaphores.
Instead of sleeping on a per-task basis (for example there are depend waits,
task waits, taskgroup waits etc..)
Hi, my name is Ray Kim.
I am a Junior EE student in Sogang Univ. Korea.
I wish applying to gcc for GSoC 2018.
I'm have a little experience on C, C++, Parallelism/Concurrency and CUDA,
however mostly very new to compiler technology.
Though I'm personally interested in the projects below,
- Par
25 matches
Mail list logo
