fixator/bk0010

Automatically exported from code.google.com/p/bk0010

bk0010's Issues

SD doesn't work @50MHz

Compile for 50MHz core. Everything is in order, but SD card functions fail.

Original issue reported on code.google.com by [email protected] on 3 Jun 2010 at 1:19

STOP key is not handled properly

Now СТОП simply traps the CPU through vector 4 and sets bit 2 in 177716.
This simulates the end result but is not correct.

Real СТОП allegedly works as follows:
1) IRQ1 (unmaskable) is generated
2) CPU attempts to trap through vector ((hi(177716))+2) effectively
(100002) (same as HALT) 
  These addresses are based on 160000 in canon implementations
3) CPU writes PC and PSW to addresses 177674 and 177676, not implemented in BK
3.1) It is highly likely that during this access, bit 2 in 177716 is set
4) CPU receives a BUS ERROR and traps through 4

Original issue reported on code.google.com by [email protected] on 10 May 2010 at 6:51

Compile tiny-fatfs for BK

To implement bootstrapping and .bin loading from SD card, MMS interface and
FAT support are necessary. Elm's library may be naturally the best fit, but
it could require some portage to become compilable with pcc.

Original issue reported on code.google.com by [email protected] on 15 May 2010 at 12:56

SAVAGE.BIN shows credits, then crashes

load SAVAGE.BIN

Original issue reported on code.google.com by [email protected] on 9 Jun 2010 at 12:55

SPHUNT.BIN doesn't start

Just falls back to the monitor ? prompt.

Original issue reported on code.google.com by [email protected] on 8 Jun 2010 at 10:48

Maintainable synchronous 1801VM1 CPU is needed

A fully synchronous вм1-compatible CPU with maintainable code is absolutely
necessary.

The current POP-11 derivative is almost impossible to maintain. It is
_very_ not cycle-accurate, several VM1 features are not implemented while
some features are done PDP-11 way.

Present POP-11 also requires use of a derived clock which is a boo-boo,
even though everything in the core domain lives in the same derived and
buffered clock.

Original issue reported on code.google.com by [email protected] on 28 Jun 2008 at 11:15

Map all special keys

Some keys are not mapped to PS/2 keyboard yet.

Original issue reported on code.google.com by [email protected] on 10 May 2010 at 10:56

Screen RAM should be available to BIN loader

POPCORN.BIN loads itself into screen RAM and starts execution in screen memory.

1) Extend loader window, at least beginning of screen RAM should be available
2) Screen restore should be off by default

Original issue reported on code.google.com by [email protected] on 1 Jun 2010 at 12:12

Implement the timer

Timer not implemented

Original issue reported on code.google.com by [email protected] on 13 May 2010 at 7:14

Tape output doesn't create loadable waveforms

Tape output doesn't create loadable waveforms

Original issue reported on code.google.com by [email protected] on 28 Jun 2008 at 11:39

Implement .bin file loading

From an SD card

Original issue reported on code.google.com by [email protected] on 11 May 2010 at 12:44

SATURN.BIN can't start

doesn't run

Original issue reported on code.google.com by [email protected] on 9 Jun 2010 at 12:55

BLOOD REVENGE shows garbage screens

L.BIN, L.OVL, L0, L1.. code seems to work fine but the tiles are all garbage

Original issue reported on code.google.com by [email protected] on 8 Jun 2010 at 10:47

Caps status is unclear

Latin characters get entered in lowercase, Cyrillic in uppercase. In real
BK it's uppercase by default and both registers must act the same.

Original issue reported on code.google.com by [email protected] on 13 May 2010 at 7:12

50 Hz timer and 177662

Investigate if it really exists.

BKBTL's reference page says that bit 14 in 177662 enables 50Hz timer
interrupt (vector 0100), but any write to 177662 in the real BK seems to
generate bus error.

Original issue reported on code.google.com by [email protected] on 2 Jun 2010 at 3:06

Implement named file loading

This can be done by hooking a kernel-mode handler to EMT 36 (104036).

Original issue reported on code.google.com by [email protected] on 4 Jun 2010 at 7:37