1 |
Hi! |
2 |
|
3 |
On Thu, 29 Jun 2017, James Le Cuirot wrote: |
4 |
> > No. Alpha is little endian. |
5 |
> |
6 |
> Wikipedia says it is bi. tc-native() reports alpha* as big so I guess |
7 |
> that's the only variant we support? Then again, this page says it is |
8 |
> usually little. Is tc-native() wrong? |
9 |
> |
10 |
> https://kernelnewbies.org/EndianIssues |
11 |
|
12 |
For the purposes of explanation, let's distinguish |
13 |
Alpha-the-processor from Alpha-the-systems here. |
14 |
|
15 |
Yes, the AtP can be switched between big- and little-endianess. |
16 |
|
17 |
AtS can't. That is, once built, hardware-wise, it has to be |
18 |
either, but can never switch. Even the firmware has to be |
19 |
different between LE and BE systems. |
20 |
|
21 |
There are a lot of LE Alpha systems, in essence, everything that |
22 |
was ever sold as an Alpha system by DEC, Compaq, HP and sundry |
23 |
others (like Samsung, who made the UP1500 and related systems). |
24 |
As a guideline: if it ever ran True64, OSF/1 or digital alpha |
25 |
UNIX, it is little-endian. |
26 |
|
27 |
The machines that run Alpha CPUs in big-endian mode are |
28 |
exclusively Cray supercomputers like the Cray T3D and T3E series. |
29 |
|
30 |
Linux only ever supported parts of the former group (e.g. the |
31 |
high-end Alpha server GS1280 series from Compaq is definitely not, |
32 |
despite running in little-endian mode). The big endian Crays were |
33 |
never even close to be supported. |
34 |
|
35 |
tl;dr: Alpha is little-endian only for (our) practical purposes. |
36 |
|
37 |
Regards, |
38 |
Tobias |
39 |
|
40 |
PS: There may be obscure one-off or developer boards for Alphas |
41 |
which can switch, but the tl;dr still stands. |
42 |
|
43 |
|
44 |
-- |
45 |
Sent from aboard the Culture ship |
46 |
GSV (Range Class) Ethics Gradient |