From owner-freebsd-hackers  Sat Aug 16 17:54:04 1997
Return-Path: <owner-freebsd-hackers>
Received: (from root@localhost)
          by hub.freebsd.org (8.8.5/8.8.5) id RAA21502
          for hackers-outgoing; Sat, 16 Aug 1997 17:54:04 -0700 (PDT)
Received: from agni.nuko.com (dummy.nuko.com [206.79.130.80])
          by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id RAA21497
          for <freebsd-hackers@freebsd.org>; Sat, 16 Aug 1997 17:54:02 -0700 (PDT)
Received: (from vinay@localhost) by agni.nuko.com (8.8.5/8.7.3) id RAA25354 for freebsd-hackers@freebsd.org; Sat, 16 Aug 1997 17:52:23 -0700 (PDT)
From: Vinay Bannai <vinay@agni.nuko.com>
Message-Id: <199708170052.RAA25354@agni.nuko.com>
Subject: Device drivers and DMA details??
To: freebsd-hackers@freebsd.org
Date: Sat, 16 Aug 1997 17:52:23 -0700 (PDT)
X-Mailer: ELM [version 2.4ME+ PL22 (25)]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Sender: owner-freebsd-hackers@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

Hi folks,

I had a few kernel and DMA related questions about a PCI driver I am about
to finish. The DMA controller is built into the card. I have been able to
successfully DMA back and forth several hundreds of bytes of data. My
question being if the transfer size is greated than the physical page
size, do we need to initiate separate DMA (or descriptors) for each page?

Also, while examining the fxp driver, the fxp_start() seems to be
invoked at the time of each incoming packet (or interrupt from the
device). Wouldn't this not create a delay in sending packets? Or is it
that the fxp_start() also gets invoked whenever a packet is shoved on
the network IF queues? 

Thanks
Vinay