From: Ilya Verbin <iverbin@gmail.com>
To: Julian Brown <julian@codesourcery.com>
Cc: Thomas Schwinge <thomas@codesourcery.com>,
gcc-patches@gcc.gnu.org, Jakub Jelinek <jakub@redhat.com>,
Kirill Yukhin <kirill.yukhin@gmail.com>
Subject: Re: Merge current set of OpenACC changes from gomp-4_0-branch
Date: Tue, 03 Feb 2015 20:01:00 -0000 [thread overview]
Message-ID: <20150203200104.GB54416@msticlxl57.ims.intel.com> (raw)
In-Reply-To: <20150203130000.7be35f5e@octopus>
On 03 Feb 13:00, Julian Brown wrote:
> On Tue, 3 Feb 2015 14:28:44 +0300
> Ilya Verbin <iverbin@gmail.com> wrote:
> > On 27 Jan 14:07, Julian Brown wrote:
> > > On Mon, 26 Jan 2015 17:34:26 +0300
> > > Ilya Verbin <iverbin@gmail.com> wrote:
> > > > Here is my current patch, it works for OpenMP->MIC, but obviously
> > > > will not work for PTX, since it requires symmetrical changes in
> > > > the plugin. Could you please take a look, whether it is possible
> > > > to support this new interface in PTX plugin?
> > >
> > > I think it can probably be made to work. I'll have a look in more
> > > detail.
> >
> > Do you have any progress on this?
>
> I'm still working on a patch to update OpenACC support and the PTX
> backend to use load/unload_image and to unify initialisation/"opening".
> So far I think the answer is basically "yes, the new interface can be
> supported", though I might request a minor tweak -- e.g. that
> load_image takes an extra "void **" argument so that a libgomp backend
> can allocate a block of generic metadata relating to the image, then
> that same block would be passed (void *) to the unload hook so the
> backend can use it there and deallocate it when it's finished with.
>
> Would that be possible? (It'd mostly be for a "CUmodule" handle: this
> could be stashed away somewhere within the nvptx backend, but it might
> be neater to put it in generic code since it'll probably be useful for
> other backends anyway.)
An extra argument is not a problem, however I don't quite get the idea.
PTX plugin allocates some data while loading, and needs this data while
unloading? Then why not to create a hash table with image_ptr -> metadata
mapping inside the plugin? In this case, to the unload hook can deallocate
metadata using the image_ptr key. Since this metadata is target-specific,
I believe it would be better to keep it inside the plugin.
-- Ilya
next prev parent reply other threads:[~2015-02-03 20:01 UTC|newest]
Thread overview: 95+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-01-15 20:44 Thomas Schwinge
2015-01-15 20:47 ` Jeff Law
2015-01-15 22:47 ` Tobias Burnus
2015-01-16 12:34 ` Gerald Pfeifer
2015-01-16 20:37 ` Thomas Schwinge
2015-01-16 15:04 ` Gerald Pfeifer
2015-01-16 15:06 ` Jakub Jelinek
2015-01-16 15:37 ` David Malcolm
2015-01-16 21:13 ` [PR libgomp/64625] Remove __OFFLOAD_TABLE__ variable/formal parameter (was: Merge current set of OpenACC changes from gomp-4_0-branch) Thomas Schwinge
2015-01-16 23:19 ` Ilya Verbin
2015-01-16 23:38 ` Jack Howarth
2015-01-16 23:48 ` Ilya Verbin
2015-01-17 0:37 ` Jack Howarth
2015-01-17 1:23 ` Ilya Verbin
2015-01-17 3:09 ` Jack Howarth
2015-02-24 17:23 ` [PR libgomp/64625] Remove __OFFLOAD_TABLE__ variable/formal parameter Thomas Schwinge
2015-01-16 22:41 ` Merge current set of OpenACC changes from gomp-4_0-branch Andreas Schwab
2015-02-04 9:41 ` [RFC testsuite] Fix PR64850, tweak acc_on_device* tests Thomas Schwinge
2015-02-10 12:02 ` Thomas Schwinge
2015-02-12 0:23 ` Kaz Kojima
2015-01-16 23:22 ` Merge current set of OpenACC changes from gomp-4_0-branch Ilya Verbin
2015-01-23 18:28 ` Ilya Verbin
2015-01-23 19:11 ` Jakub Jelinek
2015-01-26 14:01 ` Thomas Schwinge
2015-01-26 15:23 ` Ilya Verbin
2015-01-27 14:41 ` Julian Brown
2015-02-03 11:28 ` Ilya Verbin
2015-02-03 13:00 ` Julian Brown
2015-02-03 20:01 ` Ilya Verbin [this message]
2015-02-04 15:06 ` Julian Brown
2015-02-18 12:25 ` Ilya Verbin
2015-02-24 12:49 ` Julian Brown
2015-02-25 9:54 ` libgomp nvptx plugin: rework initialisation and support the proposed load/unload hooks (was: Merge current set of OpenACC changes from gomp-4_0-branch) Thomas Schwinge
2015-02-25 12:17 ` Julian Brown
2015-02-25 12:23 ` Ilya Verbin
2015-02-26 17:31 ` Ilya Verbin
2015-03-06 14:01 ` Ilya Verbin
2015-03-09 14:46 ` Julian Brown
2015-03-23 19:44 ` Ilya Verbin
2015-03-26 10:07 ` libgomp nvptx plugin: rework initialisation and support the proposed load/unload hooks Thomas Schwinge
2015-03-26 12:09 ` libgomp nvptx plugin: rework initialisation and support the proposed load/unload hooks (was: Merge current set of OpenACC changes from gomp-4_0-branch) Jakub Jelinek
2015-03-26 20:41 ` Ilya Verbin
2015-03-30 16:42 ` Jakub Jelinek
2015-03-30 21:43 ` Julian Brown
2015-03-31 12:52 ` Ilya Verbin
2015-03-31 13:08 ` Jakub Jelinek
2015-03-31 16:10 ` Ilya Verbin
2015-03-31 23:53 ` Ilya Verbin
2015-04-01 5:21 ` Jakub Jelinek
2015-04-01 13:14 ` Ilya Verbin
2015-04-01 13:20 ` Jakub Jelinek
2015-04-01 17:26 ` Ilya Verbin
2015-04-06 12:46 ` Ilya Verbin
2015-04-07 15:26 ` Jakub Jelinek
2015-04-08 14:32 ` Julian Brown
2015-04-08 14:34 ` Jakub Jelinek
2015-04-08 14:59 ` Ilya Verbin
2015-04-08 16:14 ` Julian Brown
2015-04-14 14:15 ` Julian Brown
2015-04-14 15:35 ` Using -foffload=[...] to cycle through accelerators (was: libgomp nvptx plugin: rework initialisation and support the proposed load/unload hooks) Thomas Schwinge
2015-04-14 15:43 ` acc_on_device for device_type_host_nonshm " Thomas Schwinge
2015-04-17 13:16 ` Jakub Jelinek
2015-05-07 18:32 ` acc_on_device for device_type_host_nonshm (was: libgomp nvptx plugin: rework initialisation and support the proposed load/unload hooks) (PR65742) Julian Brown
2015-05-21 11:32 ` acc_on_device for device_type_host_nonshm Thomas Schwinge
2015-05-21 11:42 ` Jakub Jelinek
2015-05-28 11:56 ` H.J. Lu
2015-05-28 13:29 ` Julian Brown
2015-06-04 7:25 ` [gomp4] " Tom de Vries
2015-06-02 12:08 ` [PR libgomp/65742, PR middle-end/66332] XFAIL acc_on_device compile-time evaluation (was: acc_on_device for device_type_host_nonshm) Thomas Schwinge
2015-07-14 20:26 ` PR65742: OpenACC acc_on_device fixes Thomas Schwinge
2015-07-15 7:27 ` Richard Biener
2015-04-17 9:54 ` libgomp nvptx plugin: rework initialisation and support the proposed load/unload hooks (was: Merge current set of OpenACC changes from gomp-4_0-branch) (PR65742) Julian Brown
2015-03-31 18:25 ` libgomp nvptx plugin: rework initialisation and support the proposed load/unload hooks (was: Merge current set of OpenACC changes from gomp-4_0-branch) Ilya Verbin
2015-03-31 19:06 ` Jakub Jelinek
2015-09-25 15:10 ` libgomp: Compile-time error for non-portable gomp_mutex_t initialization (was: libgomp nvptx plugin: rework initialisation and support the proposed load/unload hooks) Thomas Schwinge
2015-09-25 15:59 ` Jakub Jelinek
2015-11-18 15:20 ` libgomp: Compile-time error for non-portable gomp_mutex_t initialization Ilya Verbin
2015-11-19 12:31 ` Jakub Jelinek
2015-09-25 16:56 ` libgomp: Guard all offload_images/num_offload_images access by register_lock (was: libgomp nvptx plugin: rework initialisation and support the proposed load/unload hooks) Thomas Schwinge
2015-09-25 16:58 ` Ilya Verbin
2015-09-28 10:03 ` libgomp: Guard all devices/num_devices/num_devices_openmp access by register_lock (was: libgomp: Guard all offload_images/num_offload_images access by register_lock) Thomas Schwinge
2015-10-06 11:49 ` Thomas Schwinge
2015-10-09 11:58 ` libgomp: Guard all devices/num_devices/num_devices_openmp access by register_lock Bernd Schmidt
2015-10-09 14:39 ` Ilya Verbin
2015-03-27 15:21 ` libgomp nvptx plugin: rework initialisation and support the proposed load/unload hooks (was: Merge current set of OpenACC changes from gomp-4_0-branch) Julian Brown
2015-01-27 13:43 ` Merge current set of OpenACC changes from gomp-4_0-branch Julian Brown
2015-01-27 19:50 ` Jack Howarth
2015-02-17 18:06 ` Thomas Schwinge
2015-02-23 10:31 ` Fix number of arguments parameter in Ada DEF_FUNCTION_TYPE_* (was: Merge current set of OpenACC changes from gomp-4_0-branch) Thomas Schwinge
2015-04-20 14:24 ` Merge current set of OpenACC changes from gomp-4_0-branch Thomas Schwinge
2015-04-20 20:14 ` Gerald Pfeifer
2015-01-16 10:46 Uros Bizjak
2015-01-16 10:54 ` Uros Bizjak
2015-01-16 11:16 ` Jakub Jelinek
2015-02-03 23:41 [RFC testsuite] Fix PR64850, tweak acc_on_device* tests Kaz Kojima
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=20150203200104.GB54416@msticlxl57.ims.intel.com \
--to=iverbin@gmail.com \
--cc=gcc-patches@gcc.gnu.org \
--cc=jakub@redhat.com \
--cc=julian@codesourcery.com \
--cc=kirill.yukhin@gmail.com \
--cc=thomas@codesourcery.com \
/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).