1 |
$Id: BUGS,v 1.36 2005/08/15 07:01:55 debug Exp $ |
2 |
|
3 |
x) An unknown math coprocessor bug. (Ultrix' dxclock sometimes looks |
4 |
weird.) |
5 |
|
6 |
x) Enabling cache emulation (./configure --enable-caches) triggers bugs. |
7 |
|
8 |
x) Linux/DECstation (Debian) oopses extremely often unless -U is used |
9 |
at run-time. I'm not sure yet why it bugs out. With -U, the risk is |
10 |
lower, but not completely gone. _Maybe_ this is a bug in Linux. Why? |
11 |
Because the oops message contains things like ANSI escape codes and |
12 |
characters in registers (including the pc and return address register); |
13 |
this looks like a buffer overflow in the serial driver. (Another thing |
14 |
that gives weight to this theory is that the serial driver in Linux is |
15 |
still being developed.) But this is just a guess. |
16 |
|
17 |
o) Hardware device ticks are done at cycle specific intervals, not |
18 |
instruction intervals, so sometimes a fraction of a cycle can be |
19 |
"lost". |
20 |
|
21 |
o) Running Linux/DECstation 2.4.26 with no scsi disks attached causes |
22 |
a warning message to be printed by Linux. |
23 |
|
24 |
o) UDP packets that are too large are not handled well by the Lance device. |
25 |
|
26 |
o) Colors in X11 framebuffers on MacOS X hosts are wrong. (I'm not sure |
27 |
how to solve this; the code works on both little-endian (Alpha) and |
28 |
big-endian (UltraSPARC) X-servers...) |
29 |
|
30 |
o) Binary-translated 64-bit stuff checks to see if the top 32 bits are |
31 |
all zeroes or all ones, and then uses 32-bit tables and such. This |
32 |
is a bug. It should check the top 33 bits, not 32. (Alpha only, already |
33 |
fixed for i386?) |
34 |
|
35 |
o) NetBSD/arc 2.0 uses the ASC controller in a way which GXemul cannot yet |
36 |
handle. (NetBSD 1.6.2 works ok.) |
37 |
|