Date: Thu, 12 Feb 1998 19:51:54 -0500 From: Brian McGovern <bmcgover@cisco.com> To: Mike Smith <mike@smith.net.au> Cc: dg@root.com, hackers@FreeBSD.ORG Subject: Re: Mapping phyical memory in to the PCI address range... Message-ID: <199802130051.TAA10455@bmcgover-pc.cisco.com> In-Reply-To: Your message of "Thu, 12 Feb 1998 15:16:00 PST." <199802122316.PAA04707@dingo.cdrom.com>
next in thread | previous in thread | raw e-mail | index | archive | help
I guess I've mis-represented some terminology, so I think we're off on a bit of a tangent. Let me start over, and hopefully, we'll get on track. The Cyclades card has (onboard) 1MB+ of RAM, some used for the board's firmware, and some used as circular queues for sending and receiving data (about 4K on the transmit side, and 8K on the receive). These buffers can be moved by reprogramming the PLX9060 to use a memory address on the HOST so long as its visable on the PCI bus. Once this is done, the BOARD's CPU will handle transferring the data to these queues, thereby eliminating the need for the HOST CPU to copy the data from the board. The question at hand was whether it was 'faster' to have the board transfer the data to the HOST's memory, so that a bcopy or b_to_q() call would be moving data from HOST memory to HOST memory, or is the gain insufficient, and its more prudent to have the HOST CPU move the data from the board itself. As far as programming goes, you merely have to tell the PLX9060 the address of the buffer, and the size, and the rest is handled by the board. The goal is to reduce the HOST CPU time spent moving data around without affecting throughput to the board. There, now that that is said, is the picture any clearer? -Brian To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe hackers" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199802130051.TAA10455>