News for hackers. Stuff that matters.

A comment on The mysterious case of the Linux Page Table Isolation patches. Parent comment

Re:The mysterious case of the Linux Page Table Isolation pat... by hujji on Monday January 01, 2018 @12:47PM (#16047047)
The current miniaturization of DRAM circuitry doesn't really allow for a hardware fix for the RowHammer attack. During DRAM manufacturing a test similar to the RowHammer attack exists. This test has certain bounds for passing. If the bounds were tightened up to the level of perfection to prevent the attack it would drop the yield a considerable amount.
[ Parent | on HN ]
Re:The mysterious case of the Linux Page Table Isolation pat... by userbinator on Monday January 01, 2018 @01:01PM (#16047112)
If the bounds were tightened up to the level of perfection to prevent the attack it would drop the yield a considerable amount.

The fact that DRAM older than a few years is effectively immune to RH suggests it is possible to manufacture such. Yes, it will cost more, but I think many would be willing to pay for it like they used to, for none other than the assurance of having more reliable memory.

[ Parent | on HN ]
Re:The mysterious case of the Linux Page Table Isolation pat... by fiddlerwoaroof on Monday January 01, 2018 @01:21PM (#16047225)
Isn’t the problem related to the capacity of current memory modules? So you can have, say, a 32GB module that’s vulnerable or a 8GB module that isn’t? (Assuming the 8GB module uses lower density DRAM chips)
[ Parent | on HN ]
Re:The mysterious case of the Linux Page Table Isolation pat... by HelloNurse on Monday January 01, 2018 @02:12PM (#16047491)
A new 8GB DIMM module has about 1/4 the silicon area of a new 32GB module and the same density (often 1/4 the number of identical chips); only an old 8GB module, made with an entirely different process, would have larger and less dense safer DRAM cells.
[ Parent | on HN ]
Re:The mysterious case of the Linux Page Table Isolation pat... by fiddlerwoaroof on Monday January 01, 2018 @02:38PM (#16047629)
Yeah, that's why I included the qualification
[ Parent | on HN ]
Re:The mysterious case of the Linux Page Table Isolation pat... by joshtopher on Tuesday January 02, 2018 @12:05AM (#16049991)
I only buy DRAM that's aged at least 18 months
[ Parent | on HN ]
Re:The mysterious case of the Linux Page Table Isolation pat... by bloaf on Monday January 01, 2018 @01:09PM (#16047160)
Aren't some manufacturers better than others, though?

https://arstechnica.com/information-technology/2016/03/once-...

[ Parent | on HN ]
Re:The mysterious case of the Linux Page Table Isolation pat... by fulafel on Monday January 01, 2018 @01:45PM (#16047346)
There are many examples of hardware using unreliable underlying layers, countered with whitening/scrambling, redudancy and other data encoding tricks. What would make RowHammer imprevious to these?
[ Parent | on HN ]
Re:The mysterious case of the Linux Page Table Isolation pat... by JoshTriplett on Monday January 01, 2018 @01:55PM (#16047399)
Those tricks cost latency. You can get away with hiding some of that latency in access time to persistent storage. It's much harder to do so with RAM.

ECC helps, and can be done at full rate, but isn't a complete solution for all possible problems. And anything you do in hardware at full RAM speed is expensive.

[ Parent | on HN ]
Re:The mysterious case of the Linux Page Table Isolation pat... by fulafel on Monday January 01, 2018 @02:28PM (#16047569)
This is far away from "hardware fix is impossible" though. In reputable hardware, hardware vendors are expected to maintain correctness in spite of performance advances.

Also, DDR3/4 DRAM is glacially slow in latency terms, it's far from clear that there would be appreciable slowdown. There are already big latency compromises in the standardized JEDEC protocols that are not inherent in DRAM - it would be very two-faced for DRAM vendors to say they'll only trade off latency over backward compatibility or tiny cost savings, but not over correctness.

[ Parent | on HN ]
Re:The mysterious case of the Linux Page Table Isolation pat... by olliej on Monday January 01, 2018 @04:33PM (#16048168)
The trick I heard was to literally throttle the writes if you see repeated parallel writes or similar -- as row hammer depends on rapid writes that should be sufficient. But I also have no real idea how hardware actually works :) (magnets? how do they work?)
[ Parent | on HN ]
Re:The mysterious case of the Linux Page Table Isolation pat... by Dylan16807 on Tuesday January 02, 2018 @06:03AM (#16051179)
You don't need to go that far at all. Just refresh neighbors occasionally when there are tons of accesses to the same place.
[ Parent | on HN ]
Re:The mysterious case of the Linux Page Table Isolation pat... by rasz on Tuesday January 02, 2018 @03:50PM (#16055490)
You dont have to overbuild silicon with huge margins, you can build rowhammer detection mechanism and force refresh when it triggers, thats what target row refresh is doing. Problem is Jedec didnt even bother enforcing it, instead making it optional :/
[ Parent | on HN ]
Re:The mysterious case of the Linux Page Table Isolation pat... by brokensegue on Monday January 01, 2018 @01:20PM (#16047217)
maybe add more bits of ECC?
[ Parent | on HN ]
Re:The mysterious case of the Linux Page Table Isolation pat... by kps on Monday January 01, 2018 @01:26PM (#16047242)
Or indeed any bits of ECC. The fact that Intel omits or disables it on most CPUs is shameful.
[ Parent | on HN ]
Re:The mysterious case of the Linux Page Table Isolation pat... by brokensegue on Monday January 01, 2018 @02:02PM (#16047438)
one bit of ECC isn't sufficient...
[ Parent | on HN ]
Re:The mysterious case of the Linux Page Table Isolation pat... by kps on Tuesday January 02, 2018 @11:40AM (#16053002)
Best kind of correct… but modern wide memory controllers turned (9,8), which only detects one error, into (72,64), which is SECDED, “for free”.
[ Parent | on HN ]
Re:The mysterious case of the Linux Page Table Isolation pat... by brokensegue on Tuesday January 02, 2018 @01:39PM (#16054129)
that is still not sufficient
[ Parent | on HN ]
Re:The mysterious case of the Linux Page Table Isolation pat... by dboreham on Monday January 01, 2018 @01:31PM (#16047269)
There's ECC inside the chips.
[ Parent | on HN ]
Re:The mysterious case of the Linux Page Table Isolation pat... by Symmetry on Monday January 01, 2018 @02:50PM (#16047679)
The internal caches have ECC but rowhammer targets the DIMMs. ECC DIMMs would solve the issue but Intel, for market segmentation reasons, fuses off the ability to handle ECC memory in consumer chips.
[ Parent | on HN ]
Re:The mysterious case of the Linux Page Table Isolation pat... by baq on Tuesday January 02, 2018 @02:51AM (#16050531)
ECC doesn't protect against rowhammer though. It makes it more difficult but not impossible.
[ Parent | on HN ]
Re:The mysterious case of the Linux Page Table Isolation pat... by Dylan16807 on Tuesday January 02, 2018 @06:07AM (#16051191)
If you shut down whatever is triggering errors instead of completely ignoring them, the difficulty level is pretty close to "impossible".
[ Parent | on HN ]