February 19, 2013 Re: Possibility of non stop-the-world GC in the future? | ||||
---|---|---|---|---|
| ||||
Posted in reply to bearophile | On 2/19/2013 7:46 AM, bearophile wrote:
> There is some discussion here:
> http://d.puremagic.com/issues/show_bug.cgi?id=5219
>
> Bye,
> bearophile
Thanks for the heads-up. I added a vote. Is it possible/permissible to vote for an issue more than once?
|
February 19, 2013 Re: Possibility of non stop-the-world GC in the future? | ||||
---|---|---|---|---|
| ||||
Posted in reply to Minas Mina | Am 19.02.2013 17:45, schrieb Minas Mina:
> @nogc sounds nice, but how could someone use classes(OOP) with this?
The same way you do in c++. Manual memory management.
|
February 20, 2013 Re: Possibility of non stop-the-world GC in the future? | ||||
---|---|---|---|---|
| ||||
Posted in reply to Benjamin Thaut | On Tuesday, 19 February 2013 at 18:45:00 UTC, Benjamin Thaut wrote:
> Am 19.02.2013 17:45, schrieb Minas Mina:
>> @nogc sounds nice, but how could someone use classes(OOP) with this?
>
> The same way you do in c++. Manual memory management.
Yeah, it would basically reduce memory management of D down to the level of C++, but it would still be much better than C++ and we can target localized sections of code to be @nogc only where needed, so we can get the best out of both worlds.
For @nogc sections of code, it would be very useful if Phobos had smart pointers, similar to what have in C++11.
--rt
|
February 22, 2013 Re: Possibility of non stop-the-world GC in the future? | ||||
---|---|---|---|---|
| ||||
Posted in reply to Benjamin Thaut | On Monday, 18 February 2013 at 14:47:19 UTC, Benjamin Thaut wrote:
> Agreed. But as long as D does not have a Garbage Collector that is as powerfull as the GC of the .NET 4 runtime you will be better of (performance wise) not using a GC at all when programming performance critical parts of game engines.
.NET's GC is really nice for games, as far as GCs go. Something like that would make a memory-managed D acceptable.
I'm happy for a @nogc D though, and in many ways it's a better option (and it's easier to make happen). This is what I need to consider D viable for me. Notify me when this is integrated into D! I think a lack of support for purely manual memory management in D is what causes a lot of C and C++ programmers to turn their nose up at it. It sure is a beautiful language and I'm excited to see it grow. Pure MMM could widen its appeal.
|
February 22, 2013 Re: Possibility of non stop-the-world GC in the future? | ||||
---|---|---|---|---|
| ||||
Posted in reply to JoeCoder | JoeCoder:
> Is it possible/permissible to vote for an issue more than once?
I don't think so.
Bye,
bearophile
|
February 23, 2013 Re: Possibility of non stop-the-world GC in the future? | ||||
---|---|---|---|---|
| ||||
Posted in reply to Benjamin Thaut | On Tuesday, 19 February 2013 at 18:45:00 UTC, Benjamin Thaut wrote:
> Am 19.02.2013 17:45, schrieb Minas Mina:
>> @nogc sounds nice, but how could someone use classes(OOP) with this?
>
> The same way you do in c++. Manual memory management.
But I guess you would like to have OOP as well. Would that be possible with @nogc?
|
February 23, 2013 Re: Possibility of non stop-the-world GC in the future? | ||||
---|---|---|---|---|
| ||||
Posted in reply to Minas Mina | 23-Feb-2013 14:25, Minas Mina пишет: > On Tuesday, 19 February 2013 at 18:45:00 UTC, Benjamin Thaut wrote: >> Am 19.02.2013 17:45, schrieb Minas Mina: >>> @nogc sounds nice, but how could someone use classes(OOP) with this? >> >> The same way you do in c++. Manual memory management. > > But I guess you would like to have OOP as well. Would that be possible > with @nogc? std.conv.emplace It has some bugs presently though I haven't hit a single one but there is. See e.g. this pull: https://github.com/D-Programming-Language/phobos/pull/1082 Vote up. Or was that only for bugzilla ? :) -- Dmitry Olshansky |
Copyright © 1999-2021 by the D Language Foundation