Intel Is Unable Deliver a Microcode Patch Some Older CPU Models

Discussion in 'HardForum Tech News' started by FrgMstr, Apr 4, 2018.

  1. FrgMstr

    FrgMstr Just Plain Mean Staff Member

    Messages:
    47,984
    Joined:
    May 18, 1997
    Intel has released a document that explains which CPUs will not receive a Spectre / Meltdown patch. Chipzilla believes that these CPUs are typically implemented in closed systems and are expected to have a lower likelihood of exposure to vulnerabilities. Also it was deemed not practical to patch some CPUs.

    Stopped – After a comprehensive investigation of the microarchitectures and microcode capabilities for these products, Intel has determined to not release
    microcode updates for these products for one or more reasons including, but not limited to the following:
    • Micro-architectural characteristics that preclude a practical implementation of features mitigating Variant 2 (CVE-2017-5715)
    • Limited Commercially Available System Software support
    • Based on customer inputs, most of these products are implemented as “closed systems” and therefore are expected to have a lower likelihood of
    exposure to these vulnerabilities


    A lot of our readership purchases old servers through the resale market for projects. So on one hand I would like to see everything properly patched, but if something is truly ancient then what's the point?
     
  2. bigdogchris

    bigdogchris [H]ard as it Gets

    Messages:
    17,785
    Joined:
    Feb 19, 2008
    Ehh, wasn't really expecting them to release an update to 10 year old CPU's anyways. I doubt board makers or Microsoft would of even published the update.
     
    PantherBlitz likes this.
  3. PantherBlitz

    PantherBlitz Limp Gawd

    Messages:
    421
    Joined:
    Apr 14, 2011
    The fix hurts performance of Server 2008 and Win 7 a good bit, so I'm suspecting that may have had something to do with it as well.
     
  4. DeathFromBelow

    DeathFromBelow [H]ardness Supreme

    Messages:
    7,105
    Joined:
    Jul 15, 2005
    That's all you really need to know. Everything else sounds like PR BS.
     
  5. Armenius

    Armenius I Drive Myself to the [H]ospital

    Messages:
    16,791
    Joined:
    Jan 28, 2014
    More notable archs that will affect users here are Bloomfield, Gulftown, Wolfdale and Yorkfield.
     
    TrailRunner likes this.
  6. Absalom

    Absalom Gawd

    Messages:
    564
    Joined:
    Oct 3, 2007
    This is basically everything, relatively speaking, right up until (but not including) Sandy Bridge.

    Well, at least we now have the proverbial line-in-the-sand from Intel.
     
    Last edited: Apr 4, 2018
  7. ir0nw0lf

    ir0nw0lf [H]ardness Supreme

    Messages:
    6,257
    Joined:
    Feb 7, 2003
    line-in-the-sand[y bridge] :p
     
  8. Rauelius

    Rauelius 2[H]4U

    Messages:
    2,240
    Joined:
    Jan 2, 2008
    Phew...my Q6600 @ 3.6Ghz is safe.
     
  9. viper1152012

    viper1152012 [H]ard|Gawd

    Messages:
    1,025
    Joined:
    Jun 20, 2012
    Iwas hoping they woud own up and fix everything.

    For me they have persuaded me to stay away from the used market.

    So going forward I'll just get new cheap amd procs for my budget builds.

    Sigh.
     
  10. nutzo

    nutzo [H]ardness Supreme

    Messages:
    7,370
    Joined:
    Feb 15, 2004
    I have a lot of older equipment at home & the office, yet it appears every CPU I have in use will have the fix available except for my 2 oldest servers.
    Considering the servers are already 10 years old and should have been retired by now, it's not a big deal.
    I working on getting rid of one of them, guess I just add the other to the list.
     
  11. Absalom

    Absalom Gawd

    Messages:
    564
    Joined:
    Oct 3, 2007
    I wouldn't get my hopes up on that. Just because you're not (explicitly) on the list, doesn't mean you're getting a microcode update.

    My powers of deduction tell me that if neither an E8500 nor a Q9500 are getting an MCU, your Q6600 ain't getting one either. Someone at Intel just probably forgot to add that family to the list.
     
    Rauelius likes this.
  12. /dev/null

    /dev/null [H]ardForum Junkie

    Messages:
    13,845
    Joined:
    Mar 31, 2001
    I don't think any C2Qs are covered with fixes...
     
  13. /dev/null

    /dev/null [H]ardForum Junkie

    Messages:
    13,845
    Joined:
    Mar 31, 2001
    Actually, it looks like nehalem-ep and westmere-ep are covered...

    Those are pre-sandy.
     
  14. Jagger100

    Jagger100 [H]ardness Supreme

    Messages:
    7,419
    Joined:
    Oct 31, 2004
    Bloomfield was dangled for Desktop gaming use. I'm still rocking one. Guess I'm screwed.
     
  15. YeuEmMaiMai

    YeuEmMaiMai Death Incarnate

    Messages:
    13,841
    Joined:
    Jun 11, 2004
    have nothing on the list that cannot be patched outside of hardware that is literally in pieces...all currently running machines are good to go microcode update wise...
     
  16. FlawleZ

    FlawleZ Gawd

    Messages:
    752
    Joined:
    Oct 20, 2010
    I wouldn't sweat it. The vulnerability has always been there and your still marching on without issue. Especially if it's just a dedicated gaming system.
     
  17. Jagger100

    Jagger100 [H]ardness Supreme

    Messages:
    7,419
    Joined:
    Oct 31, 2004
    May be some truth to that, but the vulnerability was only widely known recently.
     
  18. Absalom

    Absalom Gawd

    Messages:
    564
    Joined:
    Oct 3, 2007
    Nice catch. Upon second glance of the updated list, it seems that Intel started work on the i7-9xx series but decided it wasn't worth hence the "Stopped" state. Interestingly enough it seems they completed the MCU work for the Xeon & Mobile family equivalents of Nehalem and Westmere archs.

    So to more accurately reflect my previous statement, pre-Sandy Bridge mainstream cpus are getting the MCU shaft or lack thereof.
     
  19. Nuxius

    Nuxius Limp Gawd

    Messages:
    249
    Joined:
    Dec 16, 2006
    Nope, because the Lynnfield (i7 8xx, i5 7xx) and Clarkdale (i5 6xx, i3 5xx, Pentium G69xx) are still slated for the update.
     
    IdiotInCharge likes this.
  20. Johan Steyn

    Johan Steyn [H]Lite

    Messages:
    84
    Joined:
    Dec 1, 2016
    One of my PC's has one of these CPU's and it is working just fine with Windows 10. Why should I be forced to buy a new system just because Intel has a flaw that I did not create? Now this machine cannot be trusted on my network, which it is on and has rights on the network. In which universe is this OK?

    Intel is rubbish and I will never ever support such a useless company again, never ever - did I say never? At least AMD supported their old CPU's and yet Intel's CEO stated that Intel is the most secure CPU company - bull twang!!!!
     
  21. cjcox

    cjcox [H]ard|Gawd

    Messages:
    1,076
    Joined:
    Jun 7, 2004
    With regards to these recent issues your subject headline could be abbreviated: "Intel is Unable"
     
    Johan Steyn likes this.
  22. Johan Steyn

    Johan Steyn [H]Lite

    Messages:
    84
    Joined:
    Dec 1, 2016

    Sorry, I cannot improve on you in this regard. Haha
     
  23. Absalom

    Absalom Gawd

    Messages:
    564
    Joined:
    Oct 3, 2007
    So add LGA 1156 to the exception list then.
     
  24. IdiotInCharge

    IdiotInCharge [H]ardForum Junkie

    Messages:
    8,759
    Joined:
    Jun 13, 2003
    I love this asshattery.

    You realize that the equivalent AMD products went backward in performance?
     
  25. Johan Steyn

    Johan Steyn [H]Lite

    Messages:
    84
    Joined:
    Dec 1, 2016
    How is your comment even relevant? The question is not about performance, but security. Maybe you did not get it?

    The fact that AMD made lower performing CPU's is in no way relevant to the topic. But it is OK, support Intel so that we can have another decade of no innovation, just go ahead...
     
  26. Zepher

    Zepher [H]ipster Replacement

    Messages:
    16,652
    Joined:
    Sep 29, 2001
    Not sure if the machine being on for so long without a reboot made the scores lower, or if the Meltdown patch is the cause, but went from 1942 to 1817.

    Intel probably won't do a microcode update for this old machine.

    cinebench15-dual-xeon.jpg

    hal-9000-cinebench-1817.jpg
     
  27. IdiotInCharge

    IdiotInCharge [H]ardForum Junkie

    Messages:
    8,759
    Joined:
    Jun 13, 2003
    Because the AMD hardware in question wasn't relevant upon release.