|Re: [eigen] Help on solving a race condition|
[ Thread Index |
| More lists.tuxfamily.org/eigen Archives
- To: eigen@xxxxxxxxxxxxxxxxxxx
- Subject: Re: [eigen] Help on solving a race condition
- From: Hauke Heibel <hauke.heibel@xxxxxxxxxxxxxx>
- Date: Fri, 8 Jun 2012 17:19:16 +0200
- Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=googlemail.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :content-type; bh=015wiWHqGgFIc7VUaXCzd7vPhQzLSdhROyFSUbnYbxI=; b=niA6vqZWT00bich1zfgoOjj5uu+Jtjh49nX7TrTFObkoq4VyFAtc4PnpEjN39GZ1uY P/brigsyAUP3n+vo1r86gvCK9hjA/+1Msv2eJtD+fRqZHNdnDaIbPj+Q4n4X4z31rah0 1BlPG+s7FKAFY/7lcoZ/ytvv8rdtUTUcQG9h1S4YS0wAtEU5H4P43qn5i3ZSwSNizCBd U5cUc8xEeaiTIbqk6ClTEAQt0IyJqYu7dF1llmhDaPfSMh0Y5YCIYyO6ep/TmBQwCKPh FQ11hHPfsGgSH7hSpGxjnxyXFBOWxfru/UOVqEOVQC/0PuNghhFBpiP7YYdvbIkAUWDY raWA==
On Fri, Jun 8, 2012 at 4:57 PM, Gael Guennebaud
> yes that would work, but mutex introduces a too large overhead. This
> function is called for every matrix product!
True. It seems to be worse than the thread local storage solution.
> It seems OpenMP's support for atomics is not very good.
Also true, its pretty tricky. In particular since one has to keep in
mind that e.g. MSVC supports "only" OpenMP 2.0.