1 |
$Id: BUGS,v 1.32 2005/03/05 11:18:31 debug Exp $ |
2 |
|
3 |
|
4 |
Serious bugs: |
5 |
|
6 |
x) An unknown math coprocessor bug. (Ultrix' dxclock sometimes looks |
7 |
weird.) |
8 |
|
9 |
x) Enabling cache emulation (./configure --enable-caches) triggers bugs. |
10 |
|
11 |
x) Linux/DECstation (Debian) oopses extremely often unless -U is used |
12 |
at run-time. I'm not sure yet why it bugs out. With -U, the risk is |
13 |
lower, but not completely gone. _Maybe_ this is a bug in Linux. Why? |
14 |
Because the oops message contains things like ANSI escape codes and |
15 |
characters in registers (including the pc and return address register); |
16 |
this looks like a buffer overflow in the serial driver. (Another thing |
17 |
that gives weight to this theory is that the serial driver in Linux is |
18 |
still being developed.) But this is just a guess. |
19 |
|
20 |
x) bintrans bugs. (For example, NetBSD/sgimips for IP22/24 hangs during |
21 |
boot with bintrans enabled.) |
22 |
|
23 |
|
24 |
Less serious bugs / issues: |
25 |
|
26 |
o) hardware device ticks are done at cycle specific intervals, not |
27 |
instruction intervals, so sometimes a fraction of a cycle can be |
28 |
"lost". |
29 |
|
30 |
o) running Linux/DECstation 2.4.26 with no scsi disks attached causes |
31 |
a warning message to be printed by Linux. |
32 |
|
33 |
o) UDP packets that are too large are not handled well by the Lance device. |
34 |
|
35 |
o) Colors in X11 framebuffers on MacOS X hosts are wrong. (I'm not sure |
36 |
how to solve this; the code works on both little-endian (Alpha) and |
37 |
big-endian (UltraSPARC) X-servers...) |
38 |
|
39 |
o) bintranslated 64-bit stuff checks to see if the top 32 bits are |
40 |
all zeroes or all ones, and then uses 32-bit tables and such. This |
41 |
is a bug. It should check the top 33 bits, not 32. (Alpha only, already |
42 |
fixed for i386?) |
43 |
|
44 |
o) NetBSD/arc 2.0 uses the ASC controller in a way which GXemul cannot yet |
45 |
handle. (NetBSD 1.6.2 works ok.) |
46 |
|