| 1 |
36 |
wfjm |
$Id: README_known_issues.txt 779 2016-06-26 15:37:16Z mueller $
|
| 2 |
30 |
wfjm |
|
| 3 |
|
|
Known issues for this release.
|
| 4 |
|
|
The case id indicates the release when the issue was first recognized.
|
| 5 |
|
|
|
| 6 |
36 |
wfjm |
- V0.73-1: as of vivado 2016.2 xelab shows sometimes extremely long build times,
|
| 7 |
|
|
especially for generated post-synthesis vhdl models. But also building a
|
| 8 |
|
|
behavioral simulation for a w11a design can take 25 min. Even though
|
| 9 |
|
|
post-synthesis or post-routing models are now generated in verilog working
|
| 10 |
|
|
with xsim is cumbersome and time consuming.
|
| 11 |
|
|
- V0.73-2: Many post-synthesis functional and especially post-routing timing
|
| 12 |
|
|
simulations currently fail due to startup and initialization problems.
|
| 13 |
|
|
Cause is MMCM/PLL startup, which is not properly reflected in the test
|
| 14 |
|
|
bench. Will be resolved in an upcoming release.
|
| 15 |
|
|
- V0.73-3: The 'state number generator' code in pdp11_sequencer causes in vivado
|
| 16 |
|
|
2016.1 (and .2) that the main FSM isn't re-coded anymore, which has high
|
| 17 |
|
|
impact on achievable clock rate. The two optional debug units depending on
|
| 18 |
|
|
the state number, dmscnt and dmcmon, are therefore currently deactivated in
|
| 19 |
|
|
all Artix based systems (but are available on all Spartan based systems).
|
| 20 |
35 |
wfjm |
|
| 21 |
31 |
wfjm |
- V0.66-1: the TM11 controller transfers data byte wise (all disk do it 16bit
|
| 22 |
|
|
word wise) and allows for odd byte length transfers. Odd length transfers
|
| 23 |
|
|
are currently not supported and rejected as invalid command. Odd byte
|
| 24 |
|
|
length records aren't used by OS, if at all, so in practice this limitation
|
| 25 |
|
|
isn't relevant.
|
| 26 |
|
|
|
| 27 |
30 |
wfjm |
- V0.65-1: ti_rri sometimes crashes in normal rundown (exit or ^D) when
|
| 28 |
|
|
a cuff: type rlink is active. One gets
|
| 29 |
|
|
terminate called after throwing an instance of 'Retro::Rexception'
|
| 30 |
|
|
what(): RlinkPortCuff::Cleanup(): driver thread failed to stop
|
| 31 |
|
|
doesn't affect normal operation, will be fixed in upcoming release.
|
| 32 |
|
|
- V0.65-2: some exotic RH70/RP/RM features and conditions not implemented yet
|
| 33 |
|
|
- last block transfered flag (in DS)
|
| 34 |
|
|
- CS2.BAI currently ignored and not handled
|
| 35 |
|
|
- read or write 'with header' gives currently ILF
|
| 36 |
|
|
All this isn't used by any OS, so in practice not relevant.
|
| 37 |
|
|
|
| 38 |
|
|
- V0.64-6: IO delays still unconstraint in vivado. All critical IOs use
|
| 39 |
|
|
explicitly IOB flops, thus timing well defined.
|
| 40 |
|
|
- V0.64-2: rlink throughput on basys3/nexys4 limited by serial port stack
|
| 41 |
|
|
round trip times. Will be overcome by libusb based custom driver.
|
| 42 |
|
|
- V0.64-1: The large default transfer size for disk accesses leads to bad
|
| 43 |
|
|
throughput in the DL11 emulation for low speed links, like the
|
| 44 |
|
|
460kBaud the S3board is limited too. Will be overcome by a DL11
|
| 45 |
|
|
controller with more buffering.
|
| 46 |
|
|
|
| 47 |
|
|
- V0.62-2: rlink v4 error recovery not yet implemented, will crash on error
|
| 48 |
|
|
- V0.62-1: Command lists aren't split to fit in retransmit buffer size
|
| 49 |
|
|
{last two issues not relevant for w11 backend over USB usage because
|
| 50 |
|
|
the backend produces proper command lists and the USB channel is
|
| 51 |
|
|
usually error free}
|
