If the setup of socket 0 to 3 with 4+2+1+1KB is detected then the W5100 is _not_ initialized, otherwise it does set up socket 0 and 1 with 4KB each. Either way socket 0 is used - now with 4KB instead of 8KB as before.
There are scenarios in which it is beneficial to search for an Etherne chip at several i/o locations. To do so the chip initialization is performed at several i/o locations until it succeeds. In order to allow for that operation model the i/o location fixup needs to be repeatable.
Note: This won't work with the RR-Net because the fixup bits overlap with the chip i/o bits.
* CS8900A
The Contiki driver allows to adjust the chip base addr at runtime (which allows to support different slots in the Apple II) and removes received frames from the chip if there's no room to send frames.
* LAN91C96
The Contiki driver was used by IP65 more or less unchanged in the first place.
* W5100
The Contiki driver allows to adjust the chip base addr at runtime (which allows to support different slots in the Apple II) and stays clear from the W5100 hybrid mode. It presumes a fully functional W5100 register auto-increment and pre-calculates necessary W5100 frame buffer wrap-arounds and thus achieves the maximal 6502 <-> W5100 transfer speed.
It's the very purpose of the DATA segment to consolidate all items that potentially need to be copied from ROM to RAM. So there's no need to / point in doing that on a per-item-base.
- Tab chars were partially used with the unusual width of two blanks. I removed them altogether.
- Line endings were inconsistent even iniside individual files. I normalized them.
- I introduced a consistent coding style regarding comment indenting and blank line usage.
- I removed trailing spaces.
- I removed newlines following unnamed labels.
- ...
