[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: move constructors likely a requirement
From: |
kingcrimson |
Subject: |
Re: move constructors likely a requirement |
Date: |
Fri, 6 Sep 2019 08:49:28 +0200 |
> Il giorno 6 set 2019, alle ore 08:36, address@hidden ha scritto:
>
> My suggestion would be to just remove NoAlias then ...
> changeset attached.
I tested the following code with and without the patch :
A = randn (10000); tic, B = triu (A); toc
time without the patch averages to 0.68528, while with the patch it is 0.48319.
I think the copy is effectively being avoided.
Is it OK if I push the changeset to default?
c.
- Re: performance improvements, (continued)
- Re: move constructors likely a requirement, Carlo De Falco, 2019/09/04
- Re: move constructors likely a requirement, John W. Eaton, 2019/09/04
- Re: move constructors likely a requirement, Carlo De Falco, 2019/09/05
- Re: move constructors likely a requirement, John W. Eaton, 2019/09/05
- Re: move constructors likely a requirement, Carlo De Falco, 2019/09/05
- Re: move constructors likely a requirement, John W. Eaton, 2019/09/05
- Re: move constructors likely a requirement, Carlo de Falco, 2019/09/05
- Re: move constructors likely a requirement, Carlo de Falco, 2019/09/05
- Re: move constructors likely a requirement, kingcrimson, 2019/09/06
- Re: move constructors likely a requirement,
kingcrimson <=
- Re: move constructors likely a requirement, Carlo De Falco, 2019/09/06
- Re: move constructors likely a requirement, Carlo De Falco, 2019/09/05
- Re: move constructors likely a requirement, John W. Eaton, 2019/09/05
- Re: move constructors likely a requirement, Carlo De Falco, 2019/09/05
- Re: move constructors likely a requirement, Carlo De Falco, 2019/09/05
- Re: move constructors likely a requirement, Carlo De Falco, 2019/09/05