pavelmachek (pavelmachek) wrote,
pavelmachek
pavelmachek

Hey Intel, what about an apology?

Hey, Intel. You were selling faulty CPUs for 15+ years, you are still selling faulty CPUs, and there are no signs you even intend to fix them. You sold faulty CPUs for half a year, knowing they are faulty, without telling you customers. You helped develop band-aids for subset of problems, and subset of configurations. Yeah, so there's work around for Meltdown on 64-bit Linux. Where's work around for Meltdown on 32-bit? What about BSDs? MINIX? L4? Where are work arounds for Spectre? And more importantly -- where are real fixes? You know, your CPUs fail to do security checks in time. Somehow I think that maybe you should fix your CPUs? I hearing you want to achieve “quantum supremacy". But maybe I'd like to hear how you intend to fix the mess you created, first? I actually started creating a workaround for x86-32, but I somehow feel like I should not be the one fixing this. I'm willing to test the patches...

(And yes, Spectre is industry-wide problem. Meltdown is -- you screwed it up.)
Subscribe
  • Post a new comment

    Error

    default userpic

    Your reply will be screened

    Your IP address will be recorded 

    When you submit the form an invisible reCAPTCHA check will be performed.
    You must follow the Privacy Policy and Google Terms of use.
  • 6 comments

Anonymous

January 25 2018, 09:56:36 UTC 8 months ago

Hi Pavel,

Do you have an idea in which kernel version is it going to be released? It seems that so far that in each branch 4.15, 4.14, 3.16,3.18,3.2 the requirement of page isolation option in the kernel is a 64-bit.
You may want to google the patches. I don't think they'll make it into v4.15, more like v4.16. Not sure about backports... you may want to help :-).