Re: Multithread safety

Alexander Nickolov wrote:
> BSD Sockets happily support multiple socket I/O on a single
> thread. The trick is to use non-blocking sockets and select
> judiciously.

Sure, but this approach doesn't scale well and is not very Windows friendly
(you cannot wait for I/O and other Windows kernel objects simultaneously and
such).


--
Eugene
http://www.gershnik.com


.

Relevant Pages

  • Re: Multithread safety
    ... portability is not important for a lot of folks anyway... ... >> BSD Sockets happily support multiple socket I/O on a single ... The trick is to use non-blocking sockets and select ... > friendly (you cannot wait for I/O and other Windows kernel objects ...
    (microsoft.public.win32.programmer.networks)
  • Re: Multithread safety
    ... BSD Sockets happily support multiple socket I/O on a single ... Microsoft MVP, MCSD ... >> the socket while another one is reading to it (full-duplex), ...
    (microsoft.public.win32.programmer.networks)
  • Re: Socket switch delay
    ... I am using non-blocking sockets. ... I was using blocking sockets in the server before (even with the Overlapped ... when the client was sending data while the server was also sending ...
    (microsoft.public.win32.programmer.networks)
  • Re: WSAEWOULDBLOCK second time I connect to the server?
    ... WSAEWOULDBLOCK is perfectly normal, at least for non-blocking sockets. ... operation would impermissibly block if carried out at this time. ...
    (microsoft.public.vc.mfc)
  • Re: on the semantics of connect(): EINTR, EALREADY, EINPROGRESS
    ... > The quote in question describes what selectand poll() do, ... > and the whole point of using blocking sockets is precisely to do so: ... Since this isn't what I understand the standard to define, ... but this essentially shows that non-blocking sockets are far ...
    (comp.unix.programmer)
Loading