• serenissi@lemmy.world
    link
    fedilink
    English
    arrow-up
    9
    arrow-down
    1
    ·
    2 days ago

    I don’t understand. writing large amount of data at once breaks nand controller? and how that’s an os issue?

    • Kay Ohtie@pawb.social
      link
      fedilink
      English
      arrow-up
      3
      ·
      1 day ago

      My assumption was it had something to do with the drivers for the controller, and the update flooding it faster than it could take data and not caching everything it couldn’t cram into controller DRAM, causing parts to just get dropped wholesale.

      • serenissi@lemmy.world
        link
        fedilink
        English
        arrow-up
        2
        ·
        1 day ago

        modern ssds have on board firmware that executes actual nvme (or sata) commands. so it’s possible a bad dma driver can coincide with a firmware bug to result in unexpected behaviour.

    • darksiderbun@lemmy.ca
      link
      fedilink
      English
      arrow-up
      4
      ·
      2 days ago

      Maybe has something to do with making assumptions about how the nand controller is going to allocate things across superpages but that is a stupid random guess and also I have no idea why it’s an OS issue but boy do they keep finding ways

      • serenissi@lemmy.world
        link
        fedilink
        English
        arrow-up
        4
        ·
        2 days ago

        os updates write large amount of data to disk at once. especially a highly bloated os like modern windows. whatever the trigger, it’s still faulty nand controller. you can very well cooy a 20G file and see disk is corrupted.

        • darksiderbun@lemmy.ca
          link
          fedilink
          English
          arrow-up
          2
          ·
          2 days ago

          That’s a good point on how the OS could be “causing” it. I don’t think my take was very likely after I thought about it for a while.