New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Compile time launch policies #2276
Conversation
- these are specific for particular launch policies
1eff09e
to
a8eb093
Compare
{ | ||
HPX_ASSERT(this->get_id()); | ||
|
||
typedef typename server::template_accumulator<T>::add_action | ||
action_type; | ||
hpx::apply<action_type>(this->get_id(), arg); | ||
hpx::apply<action_type>(hpx::launch::sync, this->get_id(), arg); |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
What does this mean? apply(sync, ...)?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
That's a typo, thanks - will fix.
- free function API - client side objects - AGAS interface - flyby: removing object_semaphore stubs
} | ||
else | ||
{ | ||
id_ = id; |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
do we really need to keep the id around if it is unmanaged?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
an alternative would be to move it into handle_managed_target
.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
We usually pass id_types by const&
, I don't see a way to move away from it in this case.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@sithhell Better now?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
A little, yes. But you are right ... the fact that we take the id_type by const reference is certainly a bit unfortunate here.
I think this PR is ready to be merged now. Some of the other pending PRs could be adjusted based on this work, thus merging this PR first would be preferable. Any objections? |
One general remark that strikes me a bit is that |
Yes, you're certainly right. I'd rather see it as a way of supporting generic programming, where If you need a 'truly' synchronous-looking syntax, use |
Looks good. Merging soon? |
On 08/08/2016 02:11 PM, John Biddiscombe wrote:
Yes, feel free to merge. |
I'm not able to merge stuff that isn't 100% up-to-date so I'll leave it to one of the admins. |
This changes the implementation of launch policies such that they can be used for tag dispaching.