public inbox for gcc@gcc.gnu.org
 help / color / mirror / Atom feed
From: Mohamed Atef <mohamedatef1698@gmail.com>
To: Martin Jambor <mjambor@suse.cz>
Cc: gcc@gcc.gnu.org
Subject: Re: Beginning OpenMP Threads
Date: Mon, 7 Mar 2022 19:57:19 +0200	[thread overview]
Message-ID: <CAPFh8NJb-FXvLHF2BvNXk3WX3mMSm9h4iRB4TR3bCVbv8+O4bg@mail.gmail.com> (raw)
In-Reply-To: <ri6ee3dpso3.fsf@suse.cz>

Hi,
  Thanks a lot.
The good news is that by the end of this month I think we will finish most
of OMPD.
The remaining parts need OMPT support which is missing in GNU but all the
other functions work perfectly fine.

Finished work,
  Initialization
  Thread handles
  ICVs

remaining work,
  task handles
  parallel region handles

Thanks

Mohamed

On Mon, Mar 7, 2022 at 7:40 PM Martin Jambor <mjambor@suse.cz> wrote:

> Hi,
>
> it's been a while since I had a serious look into libgomp...
>
> On Fri, Feb 25 2022, Mohamed Atef via Gcc wrote:
> > Hello everyone,
> >   From OpenMP specs.
> > "The OpenMP implementation must execute ompd_bp_thread_begin at every
> > native-thread-begin and initial-thread-begin event. This execution occurs
> > before the thread starts
> > the execution of any OpenMP region"
> >
> > "When terminating an OpenMP thread, the implementation executes
> > ompd_bp_thread_end."
> >
> > are these team_start and team_end in team.c file?
>
> ...but - based on a rather quick look - while I think you correctly
> tracked down that gomp_team_start is the place actually creating the
> threads, I think that the OMPD implementation wants to execute the
> ompd_bp_thread_begin function for each thread and by the native
> (pthread) thread.  And, as far as I can tell, that seems to be the
> function gomp_thread_start.  At the moment I cannot quite tell if the
> end of the function would be a good place for the thread termination
> event - or rather if it is sufficient and there may be others, I would
> need to dig deeper.  But perhaps this already help?
>
> Martin
>
>

      reply	other threads:[~2022-03-07 17:57 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-02-25  3:00 Mohamed Atef
2022-03-07 17:40 ` Martin Jambor
2022-03-07 17:57   ` Mohamed Atef [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=CAPFh8NJb-FXvLHF2BvNXk3WX3mMSm9h4iRB4TR3bCVbv8+O4bg@mail.gmail.com \
    --to=mohamedatef1698@gmail.com \
    --cc=gcc@gcc.gnu.org \
    --cc=mjambor@suse.cz \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).