[bloaf]

As long as rowhammer is still out there, aint none of your memory safe. Fixing rowhammer is the memory safety absolutism I want to hear more about.

[inigyou]

ECC memory fixes rowhammer. And random bitflips. Everyone should have ECC memory, but Intel disagrees because they are greedy.

My system sometimes detects a few bitflips per day.

[bloaf]

There have been demonstrated rowhammer-based ECC bypass attacks. ECC mitigates but does not fix rowhammer.

[userbinator]

My system sometimes detects a few bitflips per day.

Perhaps you should replace your RAM, or it's a sign that you have a massive source of radiation lurking somewhere?

[inigyou]

I assumed this is just normal in DDR5 and the reason I only had motherboards with ECC to choose from. I do have quite a lot of it.

It wouldn't be the first time engineers pushed something close to the unreliability limit and compensated with a mechanism to make it more reliable.

[mikewarot]

Amen, any RAM that is subject to bitflips, regardless of how it's arranged, should be rejected as a defective design. It's insane that we permit this crap to happen in the first place.

[e2le]

Defective memory is another issue to be aware of and somewhat common. Around 10% of Firefox crash reports are due to bit flips.

https://news.ycombinator.com/item?id=47252971