Hello!
>> I wrote a write-buffering stream. Anyone interested? Adding read and seek
>> support and extending it so it can write to other streams shouldn't be
>> difficult.
>>
> I thought windows buffered writes anyway? Is this so you can have a
> larger
> buffer or something completely different
>> IMO Linux lack both true async socket, and messaging system.
>
> I wouldn't say so. By using fnctl() one can do the following: set a
> socket descriptor F_SETFL - O_ASYNC and possibly F_SETOWN if the process
> receiving SIGIO has to be something else than this one and you have
> truly asynchrono
>> it is safe to define NO_ADV_MT when a TWSocket instance is use from a
>> single thread only
>
> It seems I missed that one. I have checked the code and it seems to me
> it is only there for if you want to call TWSocket methods from another
> thread context as where it is created in. Is that cor
Sorry for the silly question,
Which platform you will support with ICS linux, both kylix and FPC2 ?
Thanks :)
-Messaggio originale-
Da: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Per
conto di Ari Sundholm
Inviato: domenica 4 settembre 2005 16.59
A: ICS support mailing
Oggetto: Re: [tws
Hello Roberto, hello everyone!
On Sun, 2005-09-04 at 13:22 +0200, Roberto Della Pasqua wrote:
> Please permit me to suggest you what I consider the best model for highend
> scalability under linux:
Yes, epoll() seems like a good idea. It seems libevent supports it, too.
That makes me wonder if li
Please permit me to suggest you what I consider the best model for highend
scalability under linux:
http://www.xmailserver.org/linux-patches/nio-improve.html use epoll(), a
sort of linear select that check the status of all sockets dispatched in
kernel (also 100,000 sockets concurrently without do
On Sat, 2005-09-03 at 18:54 +0200, Francois PIETTE wrote:
> IMO Linux lack both true async socket, and messaging system.
I wouldn't say so. By using fnctl() one can do the following: set a
socket descriptor F_SETFL - O_ASYNC and possibly F_SETOWN if the process
receiving SIGIO has to be something
Hello Francois,
> it is safe to define NO_ADV_MT when a TWSocket instance is use from a
> single thread only
It seems I missed that one. I have checked the code and it seems to me
it is only there for if you want to call TWSocket methods from another
thread context as where it is created in. Is t
OK for me.
--
Contribute to the SSL Effort. Visit http://www.overbyte.be/eng/ssl.html
--
[EMAIL PROTECTED]
http://www.overbyte.be
- Original Message -
From: "Erich Kuba" <[EMAIL PROTECTED]>
To: "ICS support mailing"
Sent: Sunday, September 04, 2005 2:16 AM
Subject: Re: [twsocket] ICS f
