AMD Ryzen 7 3800X review
PowerColor 5700 RED DEVIL review
MSI Radeon RX 5700 XT Evoke review
ASUS Radeon RX 5700 XT ROG STRIX review
ViewSonic Elite XG240R Monitor Review
Promo: URCDKey Summer Sale Windows 10 Pro for $11
X570 Aorus master review
Cooler Master Silencio S600 review
AMD Ryzen 5 3400G review
ASUS RoG Swift PG35VQ Monitor review
Researchers Find Seven New Meltdown and Spectre Attacks
And they keep on coming, no less than Seven new Meltdown and Spectre attacks affecting AMD, ARM, and Intel CPUs were revealed by researchers this week. These include Meltdown-BR, which exploits an x86 bound instruction on Intel and AMD processors, and Meltdown-PK, which bypasses memory protection keys on Intel CPUs. ARM and Intel have already acknowledged the researchers' findings.
Researchers say they've discovered the seven new CPU attacks while performing "a sound and extensible systematization of transient execution attacks" -- a catch-all term the research team used to describe attacks on the various internal mechanisms that a CPU uses to process data, such as the speculative execution process, the CPU's internal caches, and other internal execution stages. The team, which partly consists of people who were already involved in the 'original' Specter / Meltdown research, has published a paper (pdf) in which it explains the new vulnerabilities.
There are two new Meltdown bugs, the other five are Specter variants. The research team says they've successfully demonstrated all seven attacks with proof-of-concept code. Experiments to confirm six other Meltdown-attacks did not succeed, according to a graph published by researchers.
Researchers successfully bypass hardware encryption in Samsung and Crucial SSDs - 11/06/2018 10:52 AM
Researchers from the Netherlands discovered several vulnerabilities in various SSDs from Crucial and Samsung. The embedded hardware encryption can be bypassed. ...
Researchers find uefi rootkit that is actively being abused - 09/28/2018 08:36 AM
ESET researchers have shown that the Sednit operators used different components of the LoJax malware to target a few government organizations in the Balkans as well as in Central and Eastern Europe. T...
NVIDIA CEO Hands out Special CEO Edition Titan V GPUs to AI Researchers - 06/21/2018 04:21 PM
At the Computer Vision and Pattern Recognition conference in Salt Lake City Wednesday, NVIDIA’s Jensen Huang told top AI researchers he wanted to help them achieve a very different kind of p...
Researchers reveal Variant 4 of Spectre vulnerability - 05/22/2018 05:38 PM
As discussed a few weeks ago, a new Spectre vulnerability has been shared and made public today. Earlier on it was reported that there are eight new vulnerabilities, grouped and named as Spectre-ng...
Researcher Demonstrates USB Stick That Can BSOD Any Windows 10 Device Even If Locked - 05/03/2018 07:28 AM
Microsoft has a security issue that affects both Windows 7 and Windows 10 operating systems. The code exploits a vulnerability in Microsoft's handling of NTFS filesystem images and was discovered by...
fantaskarsef
Senior Member
Posts: 10573
Joined: 2014-07-21
Senior Member
Posts: 10573
Joined: 2014-07-21
#5608673 Posted on: 11/19/2018 02:06 PM
These issues won't disappear in the years to come....
These issues won't disappear in the years to come....
alanm
Senior Member
Posts: 8816
Joined: 2004-05-10
Senior Member
Posts: 8816
Joined: 2004-05-10
#5608676 Posted on: 11/19/2018 02:11 PM
Yep, will always be something new. Even if spectre/meltdown resolved, something else will come along.
These issues won't disappear in the years to come....
Yep, will always be something new. Even if spectre/meltdown resolved, something else will come along.
emperorsfist
Senior Member
Posts: 597
Joined: 2018-08-29
Senior Member
Posts: 597
Joined: 2018-08-29
#5608678 Posted on: 11/19/2018 02:18 PM
Well, at least they're brand-agnostic
Well, at least they're brand-agnostic
fantaskarsef
Senior Member
Posts: 10573
Joined: 2014-07-21
Senior Member
Posts: 10573
Joined: 2014-07-21
#5608681 Posted on: 11/19/2018 02:20 PM
Exactly. And Meltdown / Spetre are also here to stay as long as speculative execution / prefetching of certain things is a thing.
Yep, will always be something new. Even if spectre/meltdown resolved, something else will come along.
Exactly. And Meltdown / Spetre are also here to stay as long as speculative execution / prefetching of certain things is a thing.
schmidtbag
Senior Member
Posts: 4249
Joined: 2012-11-10
Senior Member
Posts: 4249
Joined: 2012-11-10
#5608734 Posted on: 11/19/2018 05:13 PM
Meanwhile, the mitigations revolving around the vulnerabilities HyperThreading are severe enough that you might as well just simply turn off HT in BIOS.
I'm not yet sure if Windows has been patched for that.
Meanwhile, the mitigations revolving around the vulnerabilities HyperThreading are severe enough that you might as well just simply turn off HT in BIOS.
I'm not yet sure if Windows has been patched for that.
waltc3
Senior Member
Posts: 993
Joined: 2014-07-22
Senior Member
Posts: 993
Joined: 2014-07-22
#5608752 Posted on: 11/19/2018 06:32 PM
Another false alarm, apparently, by "researchers" seeking publicity for themselves and their associated colleges. Intel writes:
The vulnerabilities documented in this paper can be fully addressed by applying existing mitigation techniques for Spectre and Meltdown, including those previously documented here, and elsewhere by other chipmakers. Protecting customers continues to be a critical priority for us and we are thankful to the teams at Graz University of Technology, imec-DistriNet, KU Leuven, & the College of William and Mary for their ongoing research
In other words, these holes have already been filled! ZDNet always seems to sensationalize things--I never go there, myself, anymore. Haven't for many years. Ah, the battle for page hits goes on! So funny how they put this response by Intel at the *bottom* of the article... There is nothing new to report here.
Another false alarm, apparently, by "researchers" seeking publicity for themselves and their associated colleges. Intel writes:
The vulnerabilities documented in this paper can be fully addressed by applying existing mitigation techniques for Spectre and Meltdown, including those previously documented here, and elsewhere by other chipmakers. Protecting customers continues to be a critical priority for us and we are thankful to the teams at Graz University of Technology, imec-DistriNet, KU Leuven, & the College of William and Mary for their ongoing research
In other words, these holes have already been filled! ZDNet always seems to sensationalize things--I never go there, myself, anymore. Haven't for many years. Ah, the battle for page hits goes on! So funny how they put this response by Intel at the *bottom* of the article...
There is nothing new to report here.
Yxskaft
Senior Member
Posts: 1273
Joined: 2011-02-17
Senior Member
Posts: 1273
Joined: 2011-02-17
#5608754 Posted on: 11/19/2018 06:40 PM
Meanwhile, the mitigations revolving around the vulnerabilities HyperThreading are severe enough that you might as well just simply turn off HT in BIOS.
I'm not yet sure if Windows has been patched for that.
Is the general consensus that the HT vulnerabilities are critical even for the average joe? Because 4C/8T CPUs would suffer a horrible performance decrease in modern games if losing HT.
Meanwhile, the mitigations revolving around the vulnerabilities HyperThreading are severe enough that you might as well just simply turn off HT in BIOS.
I'm not yet sure if Windows has been patched for that.
Is the general consensus that the HT vulnerabilities are critical even for the average joe? Because 4C/8T CPUs would suffer a horrible performance decrease in modern games if losing HT.
Denial
Senior Member
Posts: 12224
Joined: 2004-05-16
Senior Member
Posts: 12224
Joined: 2004-05-16
#5608766 Posted on: 11/19/2018 07:06 PM
Another false alarm, apparently, by "researchers" seeking publicity for themselves and their associated colleges. Intel writes:
The vulnerabilities documented in this paper can be fully addressed by applying existing mitigation techniques for Spectre and Meltdown, including those previously documented here, and elsewhere by other chipmakers. Protecting customers continues to be a critical priority for us and we are thankful to the teams at Graz University of Technology, imec-DistriNet, KU Leuven, & the College of William and Mary for their ongoing research
In other words, these holes have already been filled! ZDNet always seems to sensationalize things--I never go there, myself, anymore. Haven't for many years. Ah, the battle for page hits goes on! So funny how they put this response by Intel at the *bottom* of the article... There is nothing new to report here.
What are you talking about?
https://arxiv.org/pdf/1811.05441.pdf
Here is the paper - Most of the authors were the ones who originally found spectre/meltdown. Doesn't seem like it's "seeking publicity".. just seems like they are publishing their research as any researcher would.
Also the paper concludes that the mitigations Intel is talking about don't block all the attacks here:
Transient execution attacks leak otherwise inaccessible information via the CPU’s microarchitectural state from instructions which are never committed. We presented a sound and extensible systematization of transient execution attacks. Our systematization uncovered 7 (new) transient execution attacks (Spectre and Meltdown variants) which have been overlooked and have not been investigated so far. We demonstrated all these variants in practical proof-of-concept attacks and evaluated their applicability to Intel, AMD, and ARM processors. We also systematically evaluated all defenses, discovering that some transient execution attacks are not successfully mitigated by the rolled out patches and others are not mitigated because they have been overlooked. Hence, we need to think about future defenses carefully and plan to mitigate attacks and variants that are yet unknown. and even if they did the entire point of publishing the attacks regardless to mitigation is A. Not all machines are getting upgraded with mitigation and B. So other people can build off the research and ideas found here and look for other methods that may not be mitigated already.
And even if they did block all the attacks how is that relevant to the research in the paper? Numerous machines aren't being updated and the attack vectors are new and can potentially lead other researchers into ideas that aren't mitigated. This is how science works, you build off the previous work of others and share it.
Another false alarm, apparently, by "researchers" seeking publicity for themselves and their associated colleges. Intel writes:
The vulnerabilities documented in this paper can be fully addressed by applying existing mitigation techniques for Spectre and Meltdown, including those previously documented here, and elsewhere by other chipmakers. Protecting customers continues to be a critical priority for us and we are thankful to the teams at Graz University of Technology, imec-DistriNet, KU Leuven, & the College of William and Mary for their ongoing research
In other words, these holes have already been filled! ZDNet always seems to sensationalize things--I never go there, myself, anymore. Haven't for many years. Ah, the battle for page hits goes on! So funny how they put this response by Intel at the *bottom* of the article...
There is nothing new to report here.What are you talking about?
https://arxiv.org/pdf/1811.05441.pdf
Here is the paper - Most of the authors were the ones who originally found spectre/meltdown. Doesn't seem like it's "seeking publicity".. just seems like they are publishing their research as any researcher would.
Also the paper concludes that the mitigations Intel is talking about don't block all the attacks here:
Transient execution attacks leak otherwise inaccessible information via the CPU’s microarchitectural state from instructions which are never committed. We presented a sound and extensible systematization of transient execution attacks. Our systematization uncovered 7 (new) transient execution attacks (Spectre and Meltdown variants) which have been overlooked and have not been investigated so far. We demonstrated all these variants in practical proof-of-concept attacks and evaluated their applicability to Intel, AMD, and ARM processors. We also systematically evaluated all defenses, discovering that some transient execution attacks are not successfully mitigated by the rolled out patches and others are not mitigated because they have been overlooked. Hence, we need to think about future defenses carefully and plan to mitigate attacks and variants that are yet unknown. and even if they did the entire point of publishing the attacks regardless to mitigation is A. Not all machines are getting upgraded with mitigation and B. So other people can build off the research and ideas found here and look for other methods that may not be mitigated already.
And even if they did block all the attacks how is that relevant to the research in the paper? Numerous machines aren't being updated and the attack vectors are new and can potentially lead other researchers into ideas that aren't mitigated. This is how science works, you build off the previous work of others and share it.
WareTernal
Senior Member
Posts: 245
Joined: 2013-09-27
Senior Member
Posts: 245
Joined: 2013-09-27
#5608768 Posted on: 11/19/2018 07:14 PM
Should the 'average joe' even care?
To get 'hacked', a professional 'security researcher' would bring his lab to your house, then he would stay there for weeks while you play games as he tries to steal a few bytes from your CPU's cache in hopes of getting your Steam password to steal your trading cards :O
Is the general consensus that the HT vulnerabilities are critical even for the average joe?
Should the 'average joe' even care?
To get 'hacked', a professional 'security researcher' would bring his lab to your house, then he would stay there for weeks while you play games as he tries to steal a few bytes from your CPU's cache in hopes of getting your Steam password to steal your trading cards :O
schmidtbag
Senior Member
Posts: 4249
Joined: 2012-11-10
Senior Member
Posts: 4249
Joined: 2012-11-10
#5608769 Posted on: 11/19/2018 07:19 PM
I'm not really sure. I personally would argue no, but it would be irresponsible of companies like Intel, AMD, IBM, MS, and so on to say that it isn't, because in doing so, that basically just welcomes hackers to find a way to exploit it, where it is basically guaranteed to not be prevented. In other words, as long as malware devs are led to believe that a known vulnerability will go un-mitigated, they can effortlessly take advantage of that.
However, I don't think the average gamer really has anything to worry about at all (in terms of keeping HT on without mitigations). I would imagine that most single-threaded applications aren't really in danger either, so long as they lock down their core so another malicious thread in the pipeline can't sneak in with execution prediction.
Is the general consensus that the HT vulnerabilities are critical even for the average joe? Because 4C/8T CPUs would suffer a horrible performance decrease in modern games if losing HT.
I'm not really sure. I personally would argue no, but it would be irresponsible of companies like Intel, AMD, IBM, MS, and so on to say that it isn't, because in doing so, that basically just welcomes hackers to find a way to exploit it, where it is basically guaranteed to not be prevented. In other words, as long as malware devs are led to believe that a known vulnerability will go un-mitigated, they can effortlessly take advantage of that.
However, I don't think the average gamer really has anything to worry about at all (in terms of keeping HT on without mitigations). I would imagine that most single-threaded applications aren't really in danger either, so long as they lock down their core so another malicious thread in the pipeline can't sneak in with execution prediction.
Astyanax
Senior Member
Posts: 2538
Joined: 2018-03-21
Senior Member
Posts: 2538
Joined: 2018-03-21
#5608857 Posted on: 11/20/2018 03:37 AM
Should the 'average joe' even care?
To get 'hacked', a professional 'security researcher' would bring his lab to your house, then he would stay there for weeks while you play games as he tries to steal a few bytes from your CPU's cache in hopes of getting your Steam password to steal your trading cards :O
This is exactly it, for the end user, protection mechanisms implemented in the browser code are enough for peace of mind.
Should the 'average joe' even care?
To get 'hacked', a professional 'security researcher' would bring his lab to your house, then he would stay there for weeks while you play games as he tries to steal a few bytes from your CPU's cache in hopes of getting your Steam password to steal your trading cards :O
This is exactly it, for the end user, protection mechanisms implemented in the browser code are enough for peace of mind.
schmidtbag
Senior Member
Posts: 4249
Joined: 2012-11-10
Senior Member
Posts: 4249
Joined: 2012-11-10
#5608858 Posted on: 11/20/2018 03:43 AM
Actually no, it isn't. Kind of the point why these vulnerabilities are so scary is current protection mechanism can't detect them. This is why there's new microcode that basically causes unanimous performance losses.
This is exactly it, for the end user, protection mechanisms implemented in the browser code are enough for peace of mind.
Actually no, it isn't. Kind of the point why these vulnerabilities are so scary is current protection mechanism can't detect them. This is why there's new microcode that basically causes unanimous performance losses.
Astyanax
Senior Member
Posts: 2538
Joined: 2018-03-21
Senior Member
Posts: 2538
Joined: 2018-03-21
#5608863 Posted on: 11/20/2018 03:57 AM
Yes, actually it is. The only people who are at risk from this crap are companies at risk frpm long term espionage and spying.
No low profile hackers after a quick data leak will be using them, to claim otherwise is to be completely ignorant of the blackhat landscape.
Actually no, it isn't. Kind of the point why these vulnerabilities are so scary is current protection mechanism can't detect them. This is why there's new microcode that basically causes unanimous performance losses.
Yes, actually it is. The only people who are at risk from this crap are companies at risk frpm long term espionage and spying.
No low profile hackers after a quick data leak will be using them, to claim otherwise is to be completely ignorant of the blackhat landscape.
Venix
Senior Member
Posts: 881
Joined: 2016-08-01
Senior Member
Posts: 881
Joined: 2016-08-01
#5608868 Posted on: 11/20/2018 04:05 AM
Should the 'average joe' even care?
To get 'hacked', a professional 'security researcher' would bring his lab to your house, then he would stay there for weeks while you play games as he tries to steal a few bytes from your CPU's cache in hopes of getting your Steam password to steal your trading cards :O
Oh no no no, they will drink all my beer!
Should the 'average joe' even care?
To get 'hacked', a professional 'security researcher' would bring his lab to your house, then he would stay there for weeks while you play games as he tries to steal a few bytes from your CPU's cache in hopes of getting your Steam password to steal your trading cards :O
Oh no no no, they will drink all my beer!
Click here to post a comment for this news story on the message forum.
Senior Member
Posts: 330
Joined: 2016-09-28
No no no no xD