djb@wjh12.harvard.edu (David J. Birnbaum) (09/19/89)
In article <398@wjh12.harvard.edu> djb@wjh12.UUCP (David J. Birnbaum) I wrote: >While snooping around in high RAM I noticed a peculiarity. My CompuAdd >286 has a Phoenix ROM BIOS, version 3.07, dated 5/27/87. Doing a debug >dump of f000:8000 shows that the BIOS begins at this address (at least >the display begins with PPhhooeenniixx, etc.). But dumping e000:8000 >(note: different segment) shows the identical information. In fact, >dumping several random addresses between e000:8000 and e000:ffff produced >exactly the same display as dumping the same offsets in the f000 segment. A couple of people reported noticing the same phenomenon: >I have a 10 Mhz DTK AT motherboard with Phoenix version 3.07 >I see the same thing. >I see the same thing with an Award BIOS 3.01. As for the explanation: >... the sockets on the motherboards will take a 2 64kx8 chips >while the BIOS is just 2 32kx8 chips (there is room for 4 roms on my >motherboard) Of course enough address lines had to be run to the socket >to account for the bigger chips, and A15 is unused on a 32kx8 and goes >to the Vpp (programing voltage) pin. So A15 is just ignored because the >Vpp pin isn't functional unless you have 12.5 volts running to it. Or is it A16?: >I suspect the most likely cause is that Address Bit A16 is just not >tied to the chip. In any case, the moral of the story seems to be that I'm not entitled to the e000 segment (which I knew), because IBM reserved it for ROM BIOS. And although my BIOS doesn't need the addresses, the motherboard uses (or, rather, takes but doesn't use) them anyway. Thanks to all who responded. --David ================================================================== David J. Birnbaum djb@wjh12.harvard.edu [Internet] djb@harvunxw.bitnet [Bitnet] ==================================================================
phil@diablo.amd.com (Phil Ngai) (09/20/89)
In article <403@wjh12.harvard.edu> djb@wjh12.UUCP (David J. Birnbaum) writes: |>... the sockets on the motherboards will take a 2 64kx8 chips |>while the BIOS is just 2 32kx8 chips (there is room for 4 roms on my |>motherboard) Of course enough address lines had to be run to the socket |>to account for the bigger chips, and A15 is unused on a 32kx8 and goes |>to the Vpp (programing voltage) pin. So A15 is just ignored because the |>Vpp pin isn't functional unless you have 12.5 volts running to it. | |Or is it A16?: You have to "read between the lines" here. The A15 mentioned is the EPROM's A15. Since there are two of them, that's equivalent to the 286's A16. Of course, this is all a little off the point. What counts is what the control logic does, but I suppose talking about memory chips is an easy, if slightly misleading, way to explain why the address map is the way it is. -- Phil Ngai, phil@diablo.amd.com {uunet,decwrl,ucbvax}!amdcad!phil "Should the US send assault rifles to Colombia? How about small arms?"