1 |
-----BEGIN PGP SIGNED MESSAGE-----
|
2 |
Hash: SHA1
|
3 |
|
4 |
|
5 |
On Thursday July 17, Adam Stylinski wrote:
|
6 |
|
7 |
> Pro's: |
8 |
> |
9 |
> 1.) Bloody fast machine code. Intel obfuscates their architecture |
10 |
> but they give back to the community as much as possible to make their |
11 |
> hardware marketable toward the open source sysadmin, developer, etc |
12 |
> etc. Their drivers are open and they develop for the kernel |
13 |
> constantly. This cooperation leads me to believe that they would |
14 |
> assist a team of developers in making 100% icc compatible code. |
15 |
|
16 |
Gentoo is not supported from Intel, and they had not plans doing so. As
|
17 |
of October 2007, I asked their Software channel whether Gentoo
|
18 |
users have similar support as RedHat or SUSE users and the answer was:
|
19 |
|
20 |
"No, we have no current plans to support Gentoo. Also, Gentoo is NOT a
|
21 |
derivative of a Linux we do support. My understanding is that it is
|
22 |
independently derived from Kernel.org. Thus it is less likely to work
|
23 |
than a distro which is a derivative of a supported distribution.
|
24 |
|
25 |
Meanwhile, Debian/Ubuntu got support, so things might change if
|
26 |
Gentoo re-becomes/remains popular. Any Intel dev reading this list,
|
27 |
please contact us.
|
28 |
|
29 |
And as Luca mentioned, having sunstudio, xlc (is this one free?) or llvm
|
30 |
would not make Intel a privileged case for Gentoo.
|
31 |
|
32 |
|
33 |
> 2.) Bloody fast compilation time. In my experience the compiler |
34 |
> works much faster even with heavy optimization. |
35 |
|
36 |
I don't experience this that much, but I really don't use it much
|
37 |
either. Would be nice to have benchmarks here.
|
38 |
|
39 |
|
40 |
> 4.) will project gentoo toward the power user more, helps the gentoo |
41 |
> image, and overall will make linux a more professional operating |
42 |
> system (and a quite competitive alternative to something like a |
43 |
> SPARC+Solaris configuration). This would also make cluster farms and |
44 |
> science application more respectful toward the gentoo community. The |
45 |
> academic and research world already uses ICC to compile their apps |
46 |
> for the sake of speed. The interprocedural optimizations for both |
47 |
> the fortran and c/c++ compilers make it a must. |
48 |
|
49 |
I would be careful about this, and this needs benchmarks, especially
|
50 |
with gcc > 4.3. By default icc flags are fairly agressive. For example,
|
51 |
for many scientific applications, you don't want a simple -O2 where you
|
52 |
loose floating point precision. Add -mp or -mp1 to your icc flags, add
|
53 |
some decent gcc flags, and improvement over gcc is much smaller.
|
54 |
|
55 |
> 5.) It's free, albeit a commercial product. As gentoo is entirely |
56 |
> non-profit, there is no restriction when it comes to licensing. The |
57 |
> binaries won't be sold for the intel-compiled livecd, and the |
58 |
> compiler itself with a fetch restriction allows the user to legally |
59 |
> register for their free non-commercial license. |
60 |
|
61 |
Again, as long as you're not being compensated for doing it (for Gentoo
|
62 |
I'm not).
|
63 |
|
64 |
In summary, I'm completely in favor of trying projects like this, but
|
65 |
first, this needs a few benchmarks before going further.
|
66 |
|
67 |
|
68 |
- --
|
69 |
Sébastien
|
70 |
-----BEGIN PGP SIGNATURE-----
|
71 |
Version: GnuPG v2.0.9 (GNU/Linux)
|
72 |
|
73 |
iEYEARECAAYFAkiApcoACgkQ1ycZbhPLE2C0TQCfYLD2+mazHKjosK7wiHFU6FGb
|
74 |
d3EAnR1FWZ92O2B9xBJerCj4dj2GRUZ4
|
75 |
=YhT2
|
76 |
-----END PGP SIGNATURE-----
|
77 |
éí¢‡^¾X¬¶ÈžÚ(¢¸&j)bž b² |