I'm trying to convince Intel that I have a hardware problem, but they say it is a problem with the Linux OS, and that I should use Windows instead.

The hardware has the same problem when I boot from a memtest86 USB, but I don't know for sure that memtest86 doesn't also rely on a Linux kernel.

Please confirm or deny that memtest86 uses Linux.

  • If that's a simple open source program, why can't just send source code to Intel tell them this code can reproduce the problem and slap on their face? Commented Sep 12, 2019 at 4:58
  • @炸鱼薯条德里克, 最好和羔羊说话而不是与老虎搏斗。 Commented Sep 13, 2019 at 13:59

1 Answer 1


memtest86 is self-contained, it doesn’t rely on an underlying OS. (I’m using memtest86+ here but the answer is the same for memtest86+ and memtest86.) Its boot process is described in the README.background file; it goes from boot to running memory tests without any operating system being involved.

There are however parts of the source code which are copied from the Linux kernel; for example, head.S originally came from Linux. This copied code is a very small part of memtest86+, and memtest86+ is much simpler than Linux, so I wouldn’t let that be used as the basis for any argument over the validity of the memory tests’ results.

  • They finally said they'd honour the warranty. The link to how memtest86 loads probably helped a lot. Thanks. Why they assume everyone runs Windows and they don't have their own stand-alone diagnostic program, I've no idea. It's not like Intel is a small company. Commented Sep 23, 2019 at 19:57

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .