This message is from the T13 list server.
Fun to see I can say "Pio works better than Dma" and you can say "Dma works better than Pio" and yet maybe we're not really disagreeing about any of the concrete details. > added speed, reliability, and lower CPU utilization, ... UDma UDma offers no added speed if Pio is fast enough to keep up with the device in question. Via the Crc Atapi UDma does offer a theoretically better guarantee of the integrity of Data ... but for Atapi the bytes of the Command cross the bus via Pio without Crc no matter what. But what gives Atapi Pio the edge is that Atapi Pio can copy arbitrary counts of bytes and it can count how many bytes it copied. > better choices, depending on your needs. Agreed of course. > Remember that PIO is needed - some commands > can only use PIO for data transfer. Yes. And all Atapi Command's cross the bus via Pio. > performance limitations of PIO (speed i.e. 17e+6 byte/s max burst rate. > single word and multiple word DMA are not useful > - at best you gain better CPU utilization over PIO, > and nothing over UDMA. Glad to see we agree Atapi SwDma & MwDma offer nothing over Pio except less Cpu load in return for their less reliable plug 'n pray. x4402 Pat LaVarre [EMAIL PROTECTED] http://members.aol.com/plscsi/ >>> McGrath, Jim 04/10/02 08:00PM >>> This message is from the T13 list server. If you can live with the performance limitations of PIO (speed and CPU utilization), then its OK. Only reliability is a real problem (lack of CRC). And the state machine for something like a bridge chip is easier to do if it is just PIO. Remember that PIO is needed - some commands can only use PIO for data transfer. But for added speed, reliability, and lower CPU utilization, you should use UDMA. Note that single word and multiple word DMA are not useful - at best you gain better CPU utilization over PIO, and nothing over UDMA. Either PIO only, or PIO and UDMA, are better choices, depending on your needs. Jim -----Original Message----- From: Pat LaVarre [mailto:[EMAIL PROTECTED]] Sent: Wednesday, April 10, 2002 5:57 PM To: [EMAIL PROTECTED] Subject: [t13] Pio by PRD vs. Dma by PRD This message is from the T13 list server. [ BC [EMAIL PROTECTED] since I updated the Subject line. ] > Ok, I meant how is using a PRD list with "PIO" different from "DMA" > (especially from the software perspective) ? Hmmm. I wonder if I'm misunderstanding you? I'm Not trying to be cute or sarcastic ... The low Cpu load is the same, not different. The difference from the host software perspective is that Pio works better than Dma. We can accordingly imagine a PRD list for Pio would work better than a PRD list for Dma. What's happening now is that people write software to talk Pio because it works better than Dma, which makes up for the cost of wasting Cpu cycles on Pio. But that doesn't mean people are happy to waste Cpu cycles. What the people want is Dma that works, and a PRD list for Dma would be a way to deliver that. The commodity Usb/AtapiPio bridges do exactly this: they deliver the Dma experience with Pio cycles at the target. Works great up to burst rates of 17e+6 byte/s. Would be good to see on The Motherboard too. Am I making any more sense yet? Who knows. Maybe serial Ata will learn to count data bytes. Maybe it will learn to copy arbitrary lengths of them from arbitrary addresses. Who knows. x4402 Pat LaVarre [EMAIL PROTECTED] http://members.aol.com/plscsi/ >>> RE: RE: [t13] UDMA Bursts - Pause versus Termination >>> Ooi, Thien Ern 04/10/02 06:19PM >>> This message is from the T13 list server. Ok, I meant how is using a PRD list with "PIO" different from "DMA" (especially from the software perspective) ? -----Original Message----- From: Pat LaVarre [mailto:[EMAIL PROTECTED]] Sent: Wednesday, April 10, 2002 4:38 PM To: [EMAIL PROTECTED] Subject: RE: RE: [t13] UDMA Bursts - Pause versus Termination ... Pio works better than Dma. We know this because we know people use the checkbox that Microsoft provides to turn Dma off. x4402 Pat LaVarre [EMAIL PROTECTED] http://members.aol.com/plscsi/ >>> Ooi, Thien Ern 04/10/02 04:59PM >>> This message is from the T13 list server. I don't understand. How is that different from MW-DMA and UDMA? T.E. -----Original Message----- From: Hale Landis [mailto:[EMAIL PROTECTED]] Sent: Wednesday, April 10, 2002 2:22 PM To: [EMAIL PROTECTED] Subject: RE: RE: [t13] UDMA Bursts - Pause versus Termination ... Another question: Can you give me one good reason why (5 years ago!) the host side should not be able to use a PRD list for PIO data transfers and have the host adapter preform the equivalent of the x86 REP INSx/OUTSx instructions? Given the popularity of ATA/ATAPI and all the improvements done in device designs there is no excuse for the host adapter side of the interface to be in such a sad state! ... *** Hale Landis *** www.ata-atapi.com ***
