From 0ea5fc66924303d1bf73ba283a383e2aadee02f2 Mon Sep 17 00:00:00 2001 From: neodarz Date: Sat, 11 Aug 2018 20:21:34 +0200 Subject: Initial commit --- pipermail/nel/2001-February/000301.html | 74 +++++++++++++++++++++++++++++++++ 1 file changed, 74 insertions(+) create mode 100644 pipermail/nel/2001-February/000301.html (limited to 'pipermail/nel/2001-February/000301.html') diff --git a/pipermail/nel/2001-February/000301.html b/pipermail/nel/2001-February/000301.html new file mode 100644 index 00000000..57cabbb7 --- /dev/null +++ b/pipermail/nel/2001-February/000301.html @@ -0,0 +1,74 @@ + + + + [Nel] NeL Network Engine + + + + + + +

[Nel] NeL Network Engine

+ Nicolas Hognon + cblt@cblt.org
+ Wed, 28 Feb 2001 16:11:22 +0100 +

+
+ +
i'm aggry with you
+at this time i'll use select for windows and linux (or poll for linux)
+and design my server so that i can implement another better solution.
+
+has i said before better solution under windows are
+ - are windows message and event and i don't want use it
+ - overlapped I/O but i don't really understand how it works
+   so it's not easy to design a server which is able to use select and this
+   technic. perhaps you already know it but you can take a look at
+http://www.cyberport.com/~tangent/programming/winsock/articles/io-strategies.html
+
+
+> I was more or less suggesting than the solution is not a matter of adopting
+> select() or not, but rather implies a deeper reflexion on the pb of service
+> scheduling... And being cross-platform doesn't mean being uniform, it would
+> be a shame to ignore each OS's specificity. An approach could be to get the
+> best implementation, say, for Linux and Windows, find out an abstract model
+> that unifies both and brings this into NeL. Here again, just a hint ...
+
+
+
+-- 
+Nicolas Hognon
+nicolash@virtools.com
+
+Virtools - The Behavior Company
+Tel. (+33) 1 42 71 46 86 / Fax. (+33) 1 42 71 86 53
+http://www.virtools.com/
+
+
+
+ + + +
+

+ -- cgit v1.2.1