Question Google Finds a Bug in Intel CPUs

[nicalandia]

Google Uncovers CPU Bug For Geminilake

Google Uncovers CPU Bug For Geminilake, Affecting At Least Firefox & Chrome - Phoronix Forums

Phoronix: Google Uncovers CPU Bug For Geminilake, Affecting At Least Firefox & Chrome We were alerted this morning to a CPU bug resulting in crashes for Intel Geminilake processors. At least Chrome and Firefox are affected but sounds like other software may be affected too, just that Google has...

www.phoronix.com

I can't wait for those class action Lawsuits for the 40% performance decrease with Intel CPU, now buggy CPUs?

 

[dmens]

Heh, I wonder if Tremont has the same decoder bug.

 

[Ichinisan]

Deja vu.

Pentium FDIV bug - Wikipedia

en.m.wikipedia.org

 

[ondma]

Google Uncovers CPU Bug For Geminilake

Google Uncovers CPU Bug For Geminilake, Affecting At Least Firefox & Chrome - Phoronix Forums

Phoronix: Google Uncovers CPU Bug For Geminilake, Affecting At Least Firefox & Chrome We were alerted this morning to a CPU bug resulting in crashes for Intel Geminilake processors. At least Chrome and Firefox are affected but sounds like other software may be affected too, just that Google has...

www.phoronix.com

I can't wait for those class action Lawsuits for the 40% performance decrease with Intel CPU, now buggy CPUs?

Seriously?? Where did you get that 40% number?? And apparently there are already workarounds and it has been fixed in Firefox. Seems like a minor, hard to reproduce issue in a low end processer which is not even used that widely.

 

[LikeLinus]

Are you waiting for the class action lawsuits for the AMD bug?

AMD acknowledges 3rd-gen Ryzen boost bug, says fix is coming September 10

AMD says that it will provide a BIOS fix to address reports that its 3rd-gen Ryzen chips can't hit advertised turbo boost speeds.

www.pcworld.com

Probably not, because you seem to be an AMD shill and only post negative responses about Intel.

Calling a member a shill is not allowed.
Markfw
Anandtech Moderator

 

[dmens]

Seriously?? Where did you get that 40% number?? And apparently there are already workarounds and it has been fixed in Firefox. Seems like a minor, hard to reproduce issue in a low end processer which is not even used that widely.

Incorrectly decoding lengthy x86 instructions that cross 16 byte boundaries, even if it is a corner case, is a very serious issue.

 

[Thunder 57]

Seriously?? Where did you get that 40% number?? And apparently there are already workarounds and it has been fixed in Firefox. Seems like a minor, hard to reproduce issue in a low end processer which is not even used that widely.

Nice, I actually agree with you. At least about the 40% part. Don't just make up numbers. This is how you should make an argument, instead of just crapping on AMD when you feel like it.

 

[Zucker2k]

And the 40% troll post, not to mention the false "Intel CPUs," gets a reprieve - as usual.

 

[Topweasel]

Are you waiting for the class action lawsuits for the AMD bug?

AMD acknowledges 3rd-gen Ryzen boost bug, says fix is coming September 10

AMD says that it will provide a BIOS fix to address reports that its 3rd-gen Ryzen chips can't hit advertised turbo boost speeds.

www.pcworld.com

Not that this isn't probably being way overblown but there is a huge difference between this and bugs like a boosting algorithm issue in the bios (easily fixable) or a bug that limits clock speed (in essence manufacturing issue that limits their retail spec performance) neither of those are really actionable because one is easy to fix and the other is their problem not ours.

Not properly handling code is pretty big. Though still possibly fixable with a microcode update. Not like back in the day where it basically forced a full on recall.

 

[nicalandia]

Seriously?? Where did you get that 40% number?

From Apple?

Mac Performance Drops Up to 40% With Intel Vulnerability Mitigations

Apple told its customers to disable Hyper-Threading on their systems to defend against Microarchitectural Data Sampling (MDS) issues affecting Intel processors.

www.tomshardware.com

Sorry if I was not clear enough, posting a thread requires that I leave a comment, I was trying to say that with Intel Recent Security Issues and now this Bug?

 

[DrMrLordX]

And apparently there are already workarounds and it has been fixed in Firefox. Seems like a minor, hard to reproduce issue in a low end processer which is not even used that widely.

Firefox had to fix something that never should have required a fix in the first place. See comment by @dmens .

Also, Gemini Lake is used pretty widely. It's present in numerous office/low-end PCs and some notebooks. Intel will be refreshing Gemini Lake as well (instead of introducing a consumer-level Atom based on Tremont, which will come later). One wonders if the same bug will be present in Gemini Lake R.

Are you waiting for the class action lawsuits for the AMD bug?

You joke, but AMD has been sued for less. At least the "boost bug" has been fixed, and caused exactly nobody's system to crash.

 

[VirtualLarry]

Also, Gemini Lake is used pretty widely.

I'm going to have to check my NAS units. Some of the Asustor NAS units, I believe, advertise that they use Gemini Lake CPUs.

 

[DrMrLordX]

I'm going to have to check my NAS units. Some of the Asustor NAS units, I believe, advertise that they use Gemini Lake CPUs.

Interesting. I doubt the bug would affect them much though.

 

[VirtualLarry]

Interesting. I doubt the bug would affect them much though.

Maybe not the purpose-built firmware, they can implement the workarounds in the base firmware image, but these NAS units have the ability to download and run "apps" on them, from 3rd-parties, and whatnot, that's what I'm concerned more about.

 

[DrMrLordX]

but these NAS units have the ability to download and run "apps" on them, from 3rd-parties, and whatnot, that's what I'm concerned more about.

. . . oh. That's potentially bad then.