March 31, 2014 Re: Increasing performance with static polymorphism (and other neat tricks) - blog post | ||||
---|---|---|---|---|
| ||||
Posted in reply to Meta | On Monday, 31 March 2014 at 15:24:03 UTC, Meta wrote: > On Monday, 31 March 2014 at 15:22:23 UTC, Meta wrote: >> What's the title on Hackernews? I can't seem to find it. > Eh, nevermind, found it. > Hackernews really needs a search function. Erm thats where I am coming from, the search feature is there at the bottom, take a look at the footer. I have a slightly tangential question, repeating what I asked on HN Can't the destructor issue of scopebuffer be mitigated by using hasElaborateDestructor( ) For instance have two versions one with a destructor for non POD and another without. The compiler dispatches to the appropriate one at compile time. There should not be any loss in performance. |
April 01, 2014 Re: Increasing performance with static polymorphism (and other neat tricks) - blog post | ||||
---|---|---|---|---|
| ||||
Posted in reply to srean | On Monday, 31 March 2014 at 23:57:09 UTC, srean wrote:
> On Monday, 31 March 2014 at 15:24:03 UTC, Meta wrote:
>> On Monday, 31 March 2014 at 15:22:23 UTC, Meta wrote:
>>> What's the title on Hackernews? I can't seem to find it.
>> Eh, nevermind, found it.
>
>> Hackernews really needs a search function.
>
> Erm thats where I am coming from, the search feature is there at
> the bottom, take a look at the footer.
>
> I have a slightly tangential question, repeating what I asked on
> HN
>
> Can't the destructor issue of scopebuffer be mitigated by using
>
> hasElaborateDestructor( )
>
> For instance have two versions one with a destructor for non POD
> and another without. The compiler dispatches to the appropriate
> one at compile time. There should not be any loss in performance.
Ah, you're right. That could definitely be better placed... I generally read the Hackernews RSS feed, not the site itself, so I'm not well-accustomed to the actual site.
|
April 01, 2014 Re: Increasing performance with static polymorphism (and other neat tricks) - blog post | ||||
---|---|---|---|---|
| ||||
Posted in reply to Meta | Not a question to you in particular, but dont know how to add to
a thread without replying.
I am very curious about the following (I think by the time I had
clarified my question Walter had left the building)
>> I have a slightly tangential question, repeating what I asked on HN
>>
>> Can't the destructor issue of scopebuffer be mitigated by using
>>
>> hasElaborateDestructor( )
>>
>> For instance have two versions one with a destructor for non POD
>> and another without. The compiler dispatches to the appropriate
>> one at compile time. There should not be any loss in performance.
|
April 01, 2014 Re: Increasing performance with static polymorphism (and other neat tricks) - blog post | ||||
---|---|---|---|---|
| ||||
Posted in reply to srean | On Monday, 31 March 2014 at 23:57:09 UTC, srean wrote: > Can't the destructor issue of scopebuffer be mitigated by using > > hasElaborateDestructor( ) > > For instance have two versions one with a destructor for non POD > and another without. The compiler dispatches to the appropriate > one at compile time. There should not be any loss in performance. Andrei suggested something similar on github a couple weeks ago: https://github.com/D-Programming-Language/phobos/pull/2014#issuecomment-37947056 |
Copyright © 1999-2021 by the D Language Foundation