This message is from the T13 list server.
On Wed, 30 Jan 2002 12:53:44 -0700, Pat LaVarre wrote: >This message is from the T13 list server. >Yes, AtapiPio sustains more special purposes than does AtapiDma: >in particular, AtapiPio lets people count bytes precisely. And we could have this in DMA is the host side DMA engines we use would count the bytes received and stored into system memory. >In AtapiPio, whether the data has a pad (aka residual) byte is >locally visible: the data clocked across the bus ends with >precisely one residual byte if and only if the sum of x1F5:1F4 >Cylinder (aka byte count) was odd. And it is locally visible in DMA too because how else would the host side software know to add one to the byte count (for the pad byte) when programming the host side DMA engine? >AtapiDma makes the count of residual bytes locally invisible. >Atapi UDma by design raises the max possible count of residual >bytes with increasing burst rate. Absolutely not. Only a "broken host" would fail to understand how DMA works, especially DMA on an x86. >In Intel asm AtapiPio terms we see repeat in string quad word, >then repeat in string word, then repeat in string byte. We don't >see just repeat in string word. Not in AtapiPio. Now this is what I consider to be a broken host. Mixing Different sized I/O reads/write within one PIO data burst is just plain bad design. BAD HOST - BROKEN HOST! I know of x86 host adapters that will not understand this at all and they will corrupt the data! *** Hale Landis *** www.ata-atapi.com *** Subscribe/Unsubscribe instructions can be found at www.t13.org.
