Skip site navigation (1)Skip section navigation (2)
Date:      10 Sep 2001 07:14:33 +0200
From:      Kent Boortz <kent@erix.ericsson.se>
To:        freebsd-hackers@freebsd.org
Subject:   Re: How to force small TCP packets?
Message-ID:  <d2heubk4p2.fsf@erix.ericsson.se>
References:  <d2y9nogetm.fsf@erix.ericsson.se>, <20010909182745.O2965@elvis.mu.org>

next in thread | previous in thread | raw e-mail | index | archive | help

bright@mu.org (Alfred Perlstein) writes:
> there's a setsockopt for this called TCP_NODELAY in netinet/tcp.h.
> 
> you should read further into stevens before posting such questions.

I seem to have expressed myself really bad :-( I want to debug an
application. We have had code that broke only occasionally on a
specific Linux version under heavy load. I want to detect bad code
like that and had hoped that some FreeBSD hackers had some tricks how
to use kernel configuration to do that.

If application A do a write like

  write(socket, buf, 100);

and application B read this like

  read(socket, buf, 100);

without checking the result from the read operation, then this code
will probably work 99.9999% of the time. But if I could force the
kernel to pass data from a socket on to the application say one byte
at the time then this code will break 100% of the time. This makes 
finding these problems more easy.

This is highly simplified example, else I could just grep the sources
for all calls to read() and manually inspect the code. The data is
passed on to other parts of the system using message queues.

But using the kernel to debug user applications like this was probably
a stupid idea. I'm sorry for posting a question to this group without
the proper knowledge wasting your time,

kent

To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-hackers" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?d2heubk4p2.fsf>