News for hackers. Stuff that matters.
Re:Hertzbleed Attack by ancientsofmumu on Tuesday June 14, 2022 @03:52PM (#31745119)
This paper relies on Turbo P-states, where they measure the oscillation when that is active; it is not measuring general SpeedStep (OS software controlled) as some seem to have taken away from it. Turbo state is the HWP (hardware P-state controlled) layer above SpeedStep; turning off Turbo in the BIOS still fully allows OS controlled SpeedStep P-states to function, it just disables the hardware level bursting P-states above that max listed CPU level for short periods of time. As others have noted, Turbo state can really kill a laptop battery and/or drive up the thermals pretty quick, a lot of folks disable it anyways if they've tinkered around before.

The abstract writes it as "When frequency boost is disabled, the frequency stays fixed at the base frequency during workload execution, preventing leakage via Hertzbleed. This is not a recommended mitigation strategy as it will very significantly impact performance." This is a confusing grammatical way to state it, as SpeedStep will still work at the OS layer, you'll scale min to max "as usual" and just lose temporary hardware boost max+ capability when under stress (full load at P0 state) - not really "fixed" as it were in layperson's terms. That would be more akin to saying SpeedStep had to be disabled, IMHO.

https://www.kernel.org/doc/html/v4.19/admin-guide/pm/intel_p...

Re:Hertzbleed Attack by fpoling on Tuesday June 14, 2022 @04:48PM (#31745857)
On MacOS Low-Power mode in the Power section of system settings disables turbo-boost. On Linux

    echo 1 > /sys/devices/system/cpu/intel_pstate/no_turbo 
does the rick. Note that this is not the same as power-saving mode in Gnome settings.

I have found that for heavy C++ compilation that lasts for many minutes the slowdown was about 20% on my ThinkPad X1 laptop. The big plus is that it made the laptop almost silent.

Re:Hertzbleed Attack by ancientsofmumu on Tuesday June 14, 2022 @04:59PM (#31746008)
I think you're running into changing the governor mode here, which is a related but different part of the same ballpark. Modern Intel even has a "bias hint" allowed in addition to just a governor, where the user can help tell the power saving features what tradeoffs they prefer; power-saving mode is an additional limitation in conjunction with SpeedStep (or Turbo) P-state use; if the laptop is almost silent (no fans) you're surely clocking it down to avoid heat/thermal buildup (no fans) - this is usually used to conserve/extend battery to the max possible, at the expense of CPU clock speed.

The Arch Wiki has a nice page outlining all of it, there's a lot of knobs to twiddle. https://wiki.archlinux.org/title/CPU_frequency_scaling

Re:Hertzbleed Attack by fpoling on Tuesday June 14, 2022 @05:09PM (#31746138)
It is Gnome Power setting dialog changes the governor. The above command just disables Turbo boost while allowing CPU to spend 100% of its time at the base frequency.
Re:Hertzbleed Attack by lostmsu on Tuesday June 14, 2022 @08:21PM (#31747721)
Any way to do it on Windows without getting into BIOS?
Re:Hertzbleed Attack by qayxc on Wednesday June 15, 2022 @06:47AM (#31751211)
You can use tools like ThrottleStop [0]

I use it on my laptop and run it to disable turbo boost most of the time -interestingly for performance reasons. Turbo boost leads to very erratic behaviour on laptops when you have long running CPU-intensive tasks (e.g. the cores run hot and it has to throttle down hard to cool them down again).

[0] https://www.techpowerup.com/download/techpowerup-throttlesto...

[dead]
Re:Hertzbleed Attack by omginternets on Tuesday June 14, 2022 @05:43PM (#31746441)
Could you elaborate?
[dead]
Re:Hertzbleed Attack by NamTaf on Tuesday June 14, 2022 @07:02PM (#31747070)
This has almost no similarities to stuxnet. A more analogous hypothetical attack to stuxnet would be if they repeatedly cycled spinning rust drive heads in a certain way to cause the motors to fail and corrupt data, all the while faking the SMART data of the drive to not report drive head parking cycles.

Your understanding of either this attack and/or stuxnet is flawed.

Re:Hertzbleed Attack by samstave on Tuesday June 14, 2022 @08:12PM (#31747649)
Rspndplz.

Defute

Re:Hertzbleed Attack by Dylan16807 on Tuesday June 14, 2022 @10:46PM (#31748580)
They did.

Also if this counts as "Cause some crazy busllshit at a very small level, targetting specific people, such that I get the outcome I want!" then so does every vulnerability.

Re:Hertzbleed Attack by vore on Tuesday June 14, 2022 @06:15PM (#31746709)
This is about measuring CPU frequency, where doing so definitely does not cause your computer to explode.
Re:Hertzbleed Attack by dan000892 on Tuesday June 14, 2022 @06:59PM (#31747044)
OR DOES IT???

Oh, it doesn’t? Carry on.

I’m quite fatigued by the recent (?) increase in comparisons of current vulnerabilities, attacks, and adversarial capabilities to Stuxnet and can’t help but tune out when it’s invoked. Yes, the ‘96 Bulls were the best team of all time. That has no bearing on how good the Bulls are now and sure as hell shouldn’t blind you to how good other teams have gotten since…

[dead]
Re:Hertzbleed Attack by saagarjha on Tuesday June 14, 2022 @06:54PM (#31747010)
Being unnecessarily cryptic and sounding like a crackpot while calling everyone else out for being uninformed is generally unlikely to get you support on Hacker News.
Re:Hertzbleed Attack by B1FF_PSUVM on Tuesday June 14, 2022 @08:08PM (#31747625)
Also advised: "Be kind. Don't be snarky. Have curious conversation; don't cross-examine. Please don't fulminate. Please don't sneer, including at the rest of the community."
[dead]
Re:Hertzbleed Attack by dang on Wednesday June 15, 2022 @12:20AM (#31749069)
Sam: (1) you can't do this; (2) you know you can't do this; (3) do you want us to ban you? because I don't want to ban you but your recent posts are way over the line; (4) I've put the rate limit back on your account; (5) please stop.
Re:Hertzbleed Attack by samstave on Wednesday June 15, 2022 @12:55AM (#31749267)
Apologies Dang.