6 Passes with no errors, but then errors on exit - Bug or memory error

[SuperSeven]

Hi

Running MemTest86 on a MacBook Pro 13 (Mid-2010) machine with new 8Gb of ram fitted.

Pass 1 ran fine and displayed press esc to exit message, but I left it running overnight.
In the morning it had completed 6 passes with 0 errors.

I then pressed esc to exit. After a couple of minutes MemTest86 displayed a screen showing that it had now found 2 errors with a low/high address of 0.0Mb.

It's a bit odd this was only displayed after I pressed exit.

Does this mean I have a memory issue or is it a bug in MemTest ?

[David (PassMark)]

I agree is it a bit strange.

The Lowest memory address of 0.0MB is just because it was rounded down. So this is potentially normal.

The highest memory address being 0.0MB is a known bug. We have fixed it so it will be in the next release.

We'll have a look into what test #9 does when exiting.

[David (PassMark)]

Looks like the errors after exit is a side effect of a different known bug as well.

As best we can tell the USB keyboard functionality is memory mapped into a portion of low memory on some (maybe many) machines.

So typing on a USB keyboard changes some values in RAM as the key presses are stored in memory as you type. When pressing the exit key, the tests don't stop immediately as they in a tight loop. So there is a small window of opportunity for the test to see the changed values in RAM as memory corruption and report the error.

We are going to do a V4.2 release either today or tomorrow to fix and other bugs up. We'll be excluding this area of memory from the ranges of memory to test.

[ndimmick]

I'll chime in here. I had the same ESC problem with my 2009 Macbook Pro. However, when running the new 4.2.0 version, the keyboard doesn't respond at all. The only way to stop the test is a hard power off.