diff options
Diffstat (limited to 'pipermail/nel/2001-March/000338.html')
| -rw-r--r-- | pipermail/nel/2001-March/000338.html | 164 |
1 files changed, 164 insertions, 0 deletions
diff --git a/pipermail/nel/2001-March/000338.html b/pipermail/nel/2001-March/000338.html new file mode 100644 index 00000000..eff91668 --- /dev/null +++ b/pipermail/nel/2001-March/000338.html @@ -0,0 +1,164 @@ +<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN"> +<HTML> + <HEAD> + <TITLE> [Nel] newbies about packet schema</TITLE> + <LINK REL="Index" HREF="index.html" > + <LINK REL="made" HREF="mailto:jaleco%40svr1.gameone.com.tw"> + <LINK REL="Previous" HREF="000337.html"> + <LINK REL="Next" HREF="000352.html"> + </HEAD> + <BODY BGCOLOR="#ffffff"> + <H1>[Nel] newbies about packet schema</H1> + <B>jaleco</B> + <A HREF="mailto:jaleco%40svr1.gameone.com.tw" + TITLE="[Nel] newbies about packet schema">jaleco@svr1.gameone.com.tw</A><BR> + <I>Tue, 6 Mar 2001 12:05:42 +0800</I> + <P><UL> + <LI> Previous message: <A HREF="000337.html">[Nel] problems compiling nel/net/new_message.h</A></li> + <LI> Next message: <A HREF="000352.html">[Nel] newbies about packet schema</A></li> + <LI> <B>Messages sorted by:</B> + <a href="date.html#338">[ date ]</a> + <a href="thread.html#338">[ thread ]</a> + <a href="subject.html#338">[ subject ]</a> + <a href="author.html#338">[ author ]</a> + </LI> + </UL> + <HR> +<!--beginarticle--> +<PRE>This is a multi-part message in MIME format. + +------=_NextPart_000_0010_01C0A635.C4605920 +Content-Type: text/plain; + charset="big5" +Content-Transfer-Encoding: quoted-printable + +I have read artice for winsock program,=20 +if folling article is true, that snow ball netowrk driver , should = +change it's packet schema. +any one can give me suggect ? + +original from = +<A HREF="http://www.cyberport.com/~tangent/programming/winsock/intermediate.html#d=">http://www.cyberport.com/~tangent/programming/winsock/intermediate.html#d=</A> +isable-nagle +Inexperienced Winsockers usually try disabling the Nagle algorithm when = +they are trying to impose some kind of packet scheme on a TCP data = +stream. That is, they want to be able to send, say, two packets, one 40 = +bytes and the other 60, and have the receiver get a 40-byte packet = +followed by a separate 60-byte packet. (With the Nagle algorithm = +enabled, TCP will often coalesce these two packets into a single 100 = +byte packet.) Unfortunately, this is futile, for the following reasons: + + 1.. Even if the sender manages to send its packets individually, the = +receiving TCP/IP stack may still coalesce the received packets into a = +single packet. This can happen any time the sender can send data faster = +than the receiver can deal with it.=20 + 2.. Winsock Layered Service Providers (LSPs) may coalesce or fragment = +stream data, especially LSPs that modify the data as it passes.=20 + 3.. Turning off the Nagle algorithm in a client program will not = +affect the way that the server sends packets, and vice versa.=20 + 4.. Routers and other intermediaries on the network can fragment = +packets, and there is no guarantee of "proper" reassembly with stream = +protocols.=20 + 5.. If packet arrives that is larger than the available space in the = +stack's buffers, it may fragment a packet, queuing up as many bytes as = +it has buffer space for and discarding the rest. (The remote peer will = +resend the remaining data later.)=20 + 6.. Winsock is not required to give you all the data it has queued on = +a socket even if your recv() call gave Winsock enough buffer space. It = +may require several calls to get all the data queued on a socket.=20 + + +------=_NextPart_000_0010_01C0A635.C4605920 +Content-Type: text/html; + charset="big5" +Content-Transfer-Encoding: quoted-printable + +<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN"> +<HTML><HEAD> +<META http-equiv=3DContent-Type content=3D"text/html; charset=3Dbig5"> +<META content=3D"MSHTML 5.50.4522.1800" name=3DGENERATOR> +<STYLE></STYLE> +</HEAD> +<BODY bgColor=3D#ffffff> +<DIV><FONT size=3D2>I have read artice for winsock program, = +</FONT></DIV> +<DIV><FONT size=3D2>if folling article is true, that snow ball netowrk = +driver ,=20 +should change it's packet schema.</FONT></DIV> +<DIV><FONT size=3D2>any one can give me suggect ?</FONT></DIV> +<DIV><FONT size=3D2></FONT>&nbsp;</DIV> +<DIV><FONT size=3D2>original from <A=20 +href=3D"<A HREF="http://www.cyberport.com/~tangent/programming/winsock/intermediat=">http://www.cyberport.com/~tangent/programming/winsock/intermediat=</A> +e.html#disable-nagle"><A HREF="http://www.cyberport.com/~tangent/programming/winso=">http://www.cyberport.com/~tangent/programming/winso=</A> +ck/intermediate.html#disable-nagle</A></FONT></DIV> +<DIV> +<P>Inexperienced Winsockers usually try disabling the Nagle algorithm = +when they=20 +are trying to impose some kind of <A=20 +href=3D"<A HREF="http://www.cyberport.com/~tangent/programming/winsock/intermediat=">http://www.cyberport.com/~tangent/programming/winsock/intermediat=</A> +e.html#packetscheme">packet=20 +scheme</A> on a TCP data stream. That is, they want to be able to send, = +say, two=20 +packets, one 40 bytes and the other 60, and have the receiver get a = +40-byte=20 +packet followed by a separate 60-byte packet. (With the Nagle algorithm = +enabled,=20 +TCP will often coalesce these two packets into a single 100 byte = +packet.)=20 +Unfortunately, this is futile, for the following reasons:</P> +<OL> + <LI>Even if the sender manages to send its packets individually, the = +receiving=20 + TCP/IP stack may still coalesce the received packets into a single = +packet.=20 + This can happen any time the sender can send data faster than the = +receiver can=20 + deal with it.=20 + <LI>Winsock Layered Service Providers (LSPs) may coalesce or fragment = +stream=20 + data, especially LSPs that modify the data as it passes.=20 + <LI>Turning off the Nagle algorithm in a client program will not = +affect the=20 + way that the server sends packets, and vice versa.=20 + <LI>Routers and other intermediaries on the network can fragment = +packets, and=20 + there is no guarantee of "proper" reassembly with stream protocols.=20 + <LI>If packet arrives that is larger than the available space in the = +stack's=20 + buffers, it may fragment a packet, queuing up as many bytes as it has = +buffer=20 + space for and discarding the rest. (The remote peer will resend the = +remaining=20 + data later.)=20 + <LI>Winsock is not required to give you all the data it has queued on = +a socket=20 + even if your <CODE>recv()</CODE> call gave Winsock enough buffer = +space. It may=20 + require several calls to get all the data queued on a socket.=20 +</LI></OL></DIV></BODY></HTML> + +------=_NextPart_000_0010_01C0A635.C4605920-- + + + +</pre> + + + + + + +<!--endarticle--> + <HR> + <P><UL> + <!--threads--> + <LI> Previous message: <A HREF="000337.html">[Nel] problems compiling nel/net/new_message.h</A></li> + <LI> Next message: <A HREF="000352.html">[Nel] newbies about packet schema</A></li> + <LI> <B>Messages sorted by:</B> + <a href="date.html#338">[ date ]</a> + <a href="thread.html#338">[ thread ]</a> + <a href="subject.html#338">[ subject ]</a> + <a href="author.html#338">[ author ]</a> + </LI> + </UL> +</body></html> |