1 |
I'd prefer Kumba's machine specific profiles. Of course, that's the one |
2 |
that's the biggest PITA to implement/update. |
3 |
|
4 |
Curtis Phillips |
5 |
Winemaker |
6 |
Salmon Leap Consulting |
7 |
|
8 |
On Aug 30, 2005, at 8:18 AM, Stephen P. Becker wrote: |
9 |
|
10 |
> Hi folks, |
11 |
> |
12 |
> With the somewhat recent introduction of support for a wide variety of |
13 |
> SGI machines under gentoo, expanding to include all of Indy/Indigo2, |
14 |
> Origin, Octane, Indigo2 Impact (ip28), and O2, I've noticed more than |
15 |
> just a handful of new users have had problems when getting to the |
16 |
> kernel compile phase of the install. The problem is that on systems |
17 |
> that only run 64-bit kernels, you need a mips64-unknown-linux-gnu |
18 |
> toolchain to build the kernel. Since the userland is all 32-bit, the |
19 |
> native toolchain isn't good enough to compile the kernel. However, we |
20 |
> do provide a proper toolchain via the gcc-mips64 ebuild. Furthermore, |
21 |
> binutils supports mips64 by default, but symlinks must exist such that |
22 |
> we have mips64-unknown-linux-gnu-ld -> ld, etc. Both of these are |
23 |
> automatically provided during emerge system if you use the correct |
24 |
> profile, which is default-linux/mips/mips64/2005.0 currently. |
25 |
> |
26 |
> The problem is that all of our stages ship with |
27 |
> default-linux/mips/2005.0 as the default profile, which does *not* |
28 |
> provide gcc-mips64 and the binutils symlinks. Therefore if a user |
29 |
> didn't know any better and didn't change their profile appropriately, |
30 |
> they would be stuck while trying to build their kernel because the |
31 |
> native 32-bit toolchain in the userland will just spit out errors and |
32 |
> die when compiling the kernel. Of course, this is easily fixed by |
33 |
> emerging gcc-mips64 and running "binutils-config --mips", which will |
34 |
> set up a proper toolchain. However, by that time, the user is |
35 |
> discouraged a bit and inevitably finds our irc channel and whines that |
36 |
> Gentoo is broken. |
37 |
> |
38 |
> Now, I have a few ideas for getting around this. Obviously whatever |
39 |
> is decided should be added to the documentation, but here are some |
40 |
> ideas: |
41 |
> |
42 |
> A) Do nothing...document in the handbook that if your machine is |
43 |
> 64-bit, you *must* select the mips64 sub-profile. (I don't like this |
44 |
> because some folks may be confused as to why everything still works |
45 |
> just fine with the mips profile, and/or they will just skim over that |
46 |
> and keep going) |
47 |
> |
48 |
> B) Similar to A, except ship stages without the profile set. That |
49 |
> way, folks really are stuck until they set the proper profile. (I |
50 |
> don't like this because they could still be confused and set the mips |
51 |
> profile) |
52 |
> |
53 |
> C) Make default-linux/mips/ provide all the 64-bit stuff and get rid |
54 |
> of the mips64 sub-profile, since all of the SGI machines we support |
55 |
> can run 64-bit kernels if you so choose (ip22 is the only system that |
56 |
> supports a 32-bit kernel at this time). |
57 |
> |
58 |
> D) (Kumba's idea here...) Have machine specific profiles, e.g. |
59 |
> default-linux/mips/ip22, default-linux/mips/ip32, etc. (This could be |
60 |
> really useful because it would allow us to do some other machine |
61 |
> specific voodoo in the profile). |
62 |
> |
63 |
> Any thoughts? |
64 |
> |
65 |
> -Steve |
66 |
> -- |
67 |
> gentoo-mips@g.o mailing list |
68 |
> |
69 |
|
70 |
-- |
71 |
gentoo-mips@g.o mailing list |