Development discussion of WireGuard
 help / color / mirror / Atom feed
From: "Jason A. Donenfeld" <Jason@zx2c4.com>
To: Egbert Verhage <egbert@eggiecode.org>
Cc: WireGuard mailing list <wireguard@lists.zx2c4.com>
Subject: [WireGuard] Hosting Companies with Dishonest CPUID [Was: Re: Seeking Ubuntu PPA Maintainer]
Date: Wed, 16 Nov 2016 15:14:50 +0100	[thread overview]
Message-ID: <CAHmME9pWgcg3L9p1m8SPvy1NurLrPeG2E4km-9oEZAgPktZSRg@mail.gmail.com> (raw)

Hey Egbert,

I'm glad that my avx2disabler module worked for you. I believe it
would be in your interest to pressure the hosting provider into either
reenabling AVX, or fixing their CPUID to report correct information.
CPUID is the kernel's only means of judging capabilities, and if the
[virtual] hardware lies, it is impossible to make a reliable operating
system.

Jason

On Wed, Nov 16, 2016 at 11:03 AM, Egbert Verhage <egbert@eggiecode.org> wrote:
> Hey Jason,
>
> The kernel module worked for my. ([1020751.674357] AVX2 disabler loaded:
> 1 -> 0)
> Going to update the PPA to remove my disable of AVX2.
>
> The problem is that the hosting company hides the cpuuid, but from the
> call with the sysadmin it is a XEON E5-2??? v3.
>
> Greetz,
> Egbert
>
>
> On 2016-11-16 02:31, Jason A. Donenfeld wrote:
>> Hey Egbert,
>>
>> On Wed, Nov 16, 2016 at 1:34 AM, Egbert Verhage <egbert@eggiecode.org> wrote:
>>> I mailed/called with one of the sys admins and discussed with him about the
>>> problem in the AVX2 extention.
>>> I think I'm going to mail/call the hosting company again.
>> It seems like the crux of the issue is that if they disable AVX2
>> instructions, they need to make this known in the CPUID. Otherwise the
>> operating system will assume it's there when it isn't.
>>
>>> Can give you a vps if you want to. Just 3 euro the month.
>> See the attached tarball. Compile the kernel module in there and load
>> it BEFORE WireGuard. You might have to manually unload wireguard for
>> this to work: "rmmod wireguard && insmod ./avx2disabler.ko && modprobe
>> wireguard", for example. It should disable avx2 inside the kernel,
>> solving the issue.
>>
>> Jason
>

             reply	other threads:[~2016-11-16 14:12 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-11-16 14:14 Jason A. Donenfeld [this message]
2016-11-16 14:44 ` Egbert Verhage
2016-11-16 15:05   ` Jason A. Donenfeld
2016-11-16 15:08     ` Egbert Verhage

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=CAHmME9pWgcg3L9p1m8SPvy1NurLrPeG2E4km-9oEZAgPktZSRg@mail.gmail.com \
    --to=jason@zx2c4.com \
    --cc=egbert@eggiecode.org \
    --cc=wireguard@lists.zx2c4.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).