[Coco] Fwd: IP packets on my coco
John W. Linville
linville at tuxdriver.com
Thu Jun 9 13:36:15 EDT 2016
From page 77:
4.10.9 Basic I/O Mode Receive
I/O Mode receive operations occur in the following order (In this
example, interrupts are enabled to signal the presence of a valid
1) A frame is received by the CS8900A, triggering an enabled
That looks fairly clear to me. Is the datasheet wrong?
On Thu, 2016-06-09 at 12:29 -0400, RETRO Innovations wrote:
> IRQs are allowed, but only in memory mapped mode, which require 4k
> mapping. Given the requirements of memory mode, most folks map
> cs8900a in IO mode, which means IRQs are not supported (nor is
> On June 9, 2016 at 11:19 AM "John W. Linville"
> <linville at tuxdriver.com> wrote:
> On Wed, Jun 08, 2016 at 11:38:23AM -0500, RETRO Innovations
> On 6/7/2016 4:16 PM, John W. Linville wrote:
> On Tue, Jun 07, 2016 at 12:40:29AM -0500, RETRO Innovations
> On 6/7/2016 12:31 AM, Stephen H. Fischer wrote:
> What the ????
> I see the board image but it quickly is replaced with (Soft
> Hmmm, not sure what is going on there... It didn't do that for
> butpostimage looks to be putting ads on the side...
> Anyway, here is another version, off my web site, no ads:
> Ah...well, that is very tempting. What do I have to do to get
> No big deal to get one prototyped. The main question is:
> rtl8019 or cs8900a
> * cs8900a pros
> o a bit more code available from the 65XX crowd
> o Slightly easier to produce, as I have all parts here.
> o AutoCRC?
> The datasheet I find on the web makes it very clear that the
> can generate CRC on transmit.
> * cons
> o No IRQ capability. You must poll for all packets
> Why is this? The datasheet clearly shows IRQ lines -- not
> since the part was designed for the ISA bus.
> o More expensive to produce (cs8900a is ~$10/pc)
> * rtl8019 pros
> o IRQ capability
> o A bit cheaper to produce ($rtl8019 is ~$5.00/pc)
> * cons
> o A bit less code from the 6XXX crowd
> o A bit tougher to produce, as I need to source the parts (not a
> huge deal)
> o No AutoCRC?
> Bit 0 of the Transmit Configuration Register would seem to offer
> generation on transmit.
> I know the 65XX crowd laments the lack of cs8900a IRQ, but the
> design is
> proven. So, I leave it up to the community.
> I have no idea what design issue might have painted them into
> corner, but the datasheet suggests that this shouldn't be a
> issue with the chip.
> John W. Linville Someday the world will need a hero, and you
> linville at tuxdriver.com might be all we have. Be ready.
> Coco mailing list
> Coco at maltedmedia.com
John W. Linville Someday the world will need a hero, and you
linville at tuxdriver.com might be all we have. Be ready.
More information about the Coco