Well Known Linux Kernel Developer Recommends Against Buying Skylake Systems

[phoronix]

Phoronix: Well Known Linux Kernel Developer Recommends Against Buying Skylake Systems

Well known Linux kernel developer Matthew Garrett who has led the charge for a number of years about UEFI/SecureBoot issues, poorly secured devices, and more, has taken aim now at Intel's latest-generation "Skylake" systems...

http://www.phoronix.com/scan.php?page=news_item&px=Skylake-Concerns-MJG

[Danny3]

I wasn't planning anyway since these days they keep throwing at us binary blobs so they can put their crap in there.
I think I will go with a previous version or change sides.

[imirkin]

Works for me... in powertop:

          Package   |
Powered On  0.0%    |
C1E-SKL     1.2%    |
                    |
C3-SKL      0.4%    |
RC6pp       0.0%    |
C7s-SKL     0.1%    |
C8-SKL     89.4%    |

Looks like I get a decent amount of time in C8. Desktop Core i7-6700 here.

[Sloth]

Works here too:

          Package   |             Core    |            CPU 0       CPU 4
                    |                     | C0 active   1.5%        0.8%
                    |                     | POLL        0.0%    0.0 ms  0.0%    0.0 ms
                    |                     | C1E-SKL     2.7%    0.9 ms  0.2%    0.2 ms
C2 (pc2)   27.2%    |                     |
C3 (pc3)    0.0%    | C3 (cc3)    0.7%    | C3-SKL      0.7%    0.4 ms  0.0%    0.1 ms
C6 (pc6)    2.5%    | C6 (cc6)    4.2%    | C6-SKL      5.5%    0.8 ms  0.2%    1.3 ms
C7 (pc7)    0.0%    | C7 (cc7)   80.2%    | C7s-SKL     0.1%    1.1 ms  0.0%    0.0 ms
C8 (pc8)   38.5%    |                     | C8-SKL     33.3%    2.4 ms  8.0%    4.7 ms
C9 (pc9)    0.0%    |                     | C9-SKL      0.0%    0.0 ms  0.0%    0.0 ms
C10 (pc10)  0.0%    |                     | C10-SKL    49.6%    9.1 ms 85.7%   19.6 ms

[skriticos]

Yup, works for me too (NUC6i5):

          Package   |             Core    |            CPU 0       CPU 2
Powered On  0,0%    | POLL        0,1%    | POLL        0,2%    0,8 ms  0,0%
C1E-SKL     1,1%    | C1E-SKL     0,5%    | C1E-SKL     0,6%    0,3 ms  0,5%
                    |                     |
C3-SKL      0,1%    | C3-SKL      0,2%    | C3-SKL      0,2%    0,1 ms  0,1%
RC6pp       0,0%    | C6-SKL      2,2%    | C6-SKL      2,7%    0,6 ms  1,6%
C7s-SKL     0,0%    | C7s-SKL     0,0%    | C7s-SKL     0,0%    0,0 ms  0,0%
C8-SKL     33,5%    | C8-SKL     41,1%    | C8-SKL     49,4%    3,5 ms 32,6%
C10-SKL    60,2%    | C10-SKL    51,2%    | C10-SKL    40,3%    6,2 ms 61,8%

[mjg59]

Yup, works for me too (NUC6i5):

Yes, it's not an issue on desktop parts. I've updated the post to say that.

[BillBroadley]

Yes, it's not an issue on desktop parts. I've updated the post to say that.

Er the NUC6i5 has a i5-6260U, proof at http://www.intel.com/content/www/us/...nuc6i5syk.html.

The i5-6260U is a mobile CPU, proof at http://ark.intel.com/products/series...Series#@Mobile.

So what exactly doesn't work?

[Sloth]

Yeah, mine is mobile as well: model name : Intel(R) Xeon(R) CPU E3-1505M v5 @ 2.80GHz.

[jacob]

I wasn't planning anyway since these days they keep throwing at us binary blobs so they can put their crap in there.
I think I will go with a previous version or change sides.

Which side would you change to? AMD? It's no better blob-wise. ARM? Welcome to the land of obfuscated proprietariness par excellence.
AFAIK currently the only option for a genuinely blobless platform is POWER8, but it has its own downsides.

[Delgarde]

AFAIK currently the only option for a genuinely blobless platform is POWER8, but it has its own downsides.

Most notably, the fact that it's practically non-existent for anything but heavy-grade server hardware? I know there are one or two companies selling Power-based workstations, but it's not exactly something you can get your hands on easily... or cheaply.