RAM may be vulnerable to high frequency row hammer bit flips

[Foomate]

I bought a new(*edit: new for me, but used condition ) RAM stick for my laptop, identical memory stick (exactly same model number), to get the best dual channel results.
It's a 4GB stick of ram. (currently with 8GB)
I ran memtest86, (not memtest86+), and I got this note:
[Note] RAM may be vulnerable to high frequency row hammer bit flips
But then I ran the memtest on each stick alone and did not get any reports.

Is it a bad thing to be vulenrable to row hammer bit flips while using both sticks together? Maybe they're not really synced well (They're identical, however not the same batch/mfg date)

ty!

 

[jasonkaler]

have you tried swapping them around?
also try playing with timing settings (may not be possible on a laptop).
also see if you can turn off dual channel access mode as you say it does not happen with only one.

to answer your question "Is it a bad thing" - this type of error has a lower chance of happening than other ram errors, so the likelyhood of something going wrong is low...
however if it does happen, what you are working on at the time may become corrupt, or your computer could totally crash, blue screen of death.

If there's anything critical on your laptop you can't afford to lose, then this is a very bad thing.

I would suggest that if you're not working on anything that can't be replaced, then don't stress about it. If you think it's the end of the world to lose a game because of a hardware fault, then it's worth your while changing that ram

 

[Foomate]

I can't change timings on my laptop, and how can I turn off dual channel? (But I bought the same stick for the dual channel performance )
I haven't swapped them yet, however I did test each individually in the slot where the original RAM was placed(when I got no errors), and now they're actually swapped so all I need to do is to run the memtest again and report

Let's hope it will be good now, because sometimes I do stuff that matter (not gaming), so hopefully I won't get the BSOD while doing them!

 

[jasonkaler]

Fingers crossed. just remember to save often.
also, a big tip: Each day, save a copy of the important stuff you're working on instead of just overwriting the old file. e.g. when opening ImportantWork.doc , save it as ImportantWork_monday.doc - just in case you lose the original.

 

[Pinhedd]

I bought a new(*edit: new for me, but used condition ) RAM stick for my laptop, identical memory stick (exactly same model number), to get the best dual channel results.
It's a 4GB stick of ram. (currently with 8GB)
I ran memtest86, (not memtest86+), and I got this note:
[Note] RAM may be vulnerable to high frequency row hammer bit flips
But then I ran the memtest on each stick alone and did not get any reports.

Is it a bad thing to be vulenrable to row hammer bit flips while using both sticks together? Maybe they're not really synced well (They're identical, however not the same batch/mfg date)

ty!

Row hammering is a design vulnerability that an attacker can exploit to write to memory that it would normally not be permitted to write to. Although it has been demonstrated, it's unreliable and virtually impossible to exploit in the wild; all proof-of-concepts have been carefully crafted and are run in laboratory conditions.

So no, i'ts not something to worry about.

 

[Foomate]

Thank you, but as jasonkaler said it might cause crashes and BSODS, that would then be my main concern if exploiting it is virtually impossible.
And it's weird that it appears only when both of the sticks are in, why is that happening? I mean, at first I thought that the used stick I bought is at fault, but then I realized that it is OK when it's plugged alone (the other removed for testing)

 

[Pinhedd]

Thank you, but as jasonkaler said it might cause crashes and BSODS, that would then be my main concern if exploiting it is virtually impossible.
And it's weird that it appears only when both of the sticks are in, why is that happening? I mean, at first I thought that the used stick I bought is at fault, but then I realized that it is OK when it's plugged alone (the other removed for testing)

Jasonkaler doesn't know what row hammering is. Row hammering has nothing to do with crashes and BSoDs unless someone is running a program that is attempting to corrupt memory via row hammering, and even then it's most likely going to crash the program that is doing the row hammering rather than anything else.

 

[Foomate]

Alright, thank you. Hopefully I really won't have BSODS and I still find it weird that Memtest shows this warning about row hammering only when both of the sticks are plugged!

 

[antonios.hadji]

Answer from the source.
https://www.memtest86.com/troubleshooting.htm#hammer

Summary:
1. Replace with ECC RAM if running mission critical applications like banking servers or safety equipment (and supported by motherboard)
2. Replace with another brand that is known to not have issues with hammer bit flips if you want no errors on the test
3. Do nothing. It's unlikely to cause noticeable errors for everyday home or office use.

 

[Lynn Hutson]

I own a computer repair store and out of thousands of memtest runs over the years this is only the 2nd time it has happened, Samsung AIO nothing special. 2x8gb ddr3l. Memtest doesn't treat it as an error. Error field still says 0. this is just a warning pretty much. nothing to worry about, move on

Memtest showing 0 errors while receiving bitflip warning. Forget about it, you're
good!

 

[Layarion]

I bought a new(*edit: new for me, but used condition ) RAM stick for my laptop, identical memory stick (exactly same model number), to get the best dual channel results.
It's a 4GB stick of ram. (currently with 8GB)
I ran memtest86, (not memtest86+), and I got this note:
[Note] RAM may be vulnerable to high frequency row hammer bit flips
But then I ran the memtest on each stick alone and did not get any reports.

Is it a bad thing to be vulenrable to row hammer bit flips while using both sticks together? Maybe they're not really synced well (They're identical, however not the same batch/mfg date)

ty!

Row hammering is a design vulnerability that an attacker can exploit to write to memory that it would normally not be permitted to write to. Although it has been demonstrated, it's unreliable and virtually impossible to exploit in the wild; all proof-of-concepts have been carefully crafted and are run in laboratory conditions.

So no, i'ts not something to worry about.

Good to know, thanks