Here is a list of known bugs of version 2.0.0-pre1
--------------------------------------------------

- starting ROM program is only possible in the old-style shell
  and it doesn't fill in the command line. Starting o65 files is ok.

- lib6502 does not use SBRK when memory is exhausted.

- slipd libip routines seem to be buggy still - very weird effects!
  I have no idea where they come from. When I start hunting one, it
  disappears and another pops up. Amongs them were:
  - closing telnet seems to set nthreads in kernel tables to $ff!!!!
  - also opening a telnet connection prints some weird characters on the
    wrong screen (stream gets "$ff $ff $c6 $07 $07 $07..." bytes sent,
    I have no idea where from...)
  - telnet works on telnet port somehow - bug hangs if streams are closed...
    should comply to rfc1143
  - httpd doesn't work although using almost all the same routines as
    rsh which work
  - SC_ESTAT does not seem to recognize E_EOF the same way as SC_STAT?
    (slipd, in rsh.a65, label newpkt)
  I have found and fixed a bug in the lib6502 fread/fwrite that could 
  probably have caused this bug, but have not (yet) checked this.

- using inet apps without registering an inet task blocks those programs
  (kernel problem as not-registered task is not recognized)

- listen does not check if port already in use
  also there is no way to free listenports in use by crashed/killed
  programs - probably when reuse, overwrite old listenport such 
  that table does not fill up?

- telnet doesn't really implement local echo mode.

- slipd uses accesses to $dc** (C64 CIA timers) to generate initial
  pseudo-random sequence numbers.

- when getting tcp signal, set error flag in associated streams

- SLIPD: wwwroot path given to chdir, but not used when opening a file;
  only the first letter is used as drive.

- FSIEC does not recognize chdir paths when reading directories
  from the bus (i.e. C64 reads from IEEE488 bus and CSA drive is the 
  fsibm drive), files are ok.

- more diagnostics ("why the beep...", i.e. why the "?") in lsh

- keyboard layout into docs

- when fsiec on C64 is parallel but doesn't find the interface, it
  dies too ungracefully...

- when path overflows, behaviour is undefined

- C64 sbrk always returns ROMSTART instead of RAMSIZE?

- a semaphore can be released when a process is waiting for it

- console: control-character printed?

- (multiple) errors on non-existant IEC drives are not handled correctly:
  gecko may break, C64 sometimes reads empty stream...

- FSIEC hangs on disk full with C64 parallel IEC bus interface (and probably
  others too)

- FSIEC (on CS/A65) seems not to release the interface semaphore after
  init every time?

- serial line drivers ignore handshake value (i.e. rts/cts vs. xon/xoff)

- serial IEC bus for CS/A65 architecture only works with 1MHz (is this a bug?)

- The CS/A65 console doesn't work when interrupting the device is enabled.
  (could be kernel or device problem...)

- C64 fsiec filesystem (serial/parallel?) reads one byte from file even
  when it's not found (in VICE)

- C64 console doesn't generate the F1 key with shift and shift/ctrl?

- CS/A65: non-ROM-started devices don't work.

- spooler device doesn't forward BRK and PUSH flags

- fsiec, fsibm and fsdev ignore stream brk/push/pull flags, and don't set 
  stream error flags (only fsiec sets SCE_[RW]ERRM flags).

- PET don't do shifted characters on keyboard