Re: Suggestions for custom application-layer protocol?
From: David Schwartz (davids_at_webmaster.com)
Date: 05/26/05
- Previous message: Walter Roberson: "Re: Suggestions for custom application-layer protocol?"
- In reply to: Walter Roberson: "Re: Suggestions for custom application-layer protocol?"
- Next in thread: Grant Edwards: "Re: Suggestions for custom application-layer protocol?"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Date: Thu, 26 May 2005 11:13:38 -0700
"Walter Roberson" <roberson@ibd.nrc-cnrc.gc.ca> wrote in message
news:d753g2$lim$1@canopus.cc.umanitoba.ca...
> You still need to "mess about" if your data might not fit in a single
> packet.
And even if it does.
> Recall that when you read() from a socket you are not certain
> to get all of the data unless the data length is no greater than
> your system's atomicity limits [PIPE_BUF]. The POSIX minimum
> value for PIPE_BUF is 512 -- so in the general case a write() of
> more than 512 bytes may require multiple read() statements.
These atomicity limits don't apply to TCP connections, which have their
own segmentation rules.
DS
- Previous message: Walter Roberson: "Re: Suggestions for custom application-layer protocol?"
- In reply to: Walter Roberson: "Re: Suggestions for custom application-layer protocol?"
- Next in thread: Grant Edwards: "Re: Suggestions for custom application-layer protocol?"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]