RdRand Performance As Bad As ~3% Original Speed With CrossTalk/SRBDS Mitigation

[phoronix]

Phoronix: RdRand Performance As Bad As ~3% Original Speed With CrossTalk/SRBDS Mitigation

Following today's disclosure by Intel of the CrossTalk/SRBDS vulnerability that is MDS-based and vulnerable across physical cores with affected instructions, Intel released new CPU microcode to mitigate the most prone/significant instructions. I've been benchmarking the impact of this new microcode on multiple systems and will have a full report tonight or tomorrow morning... But here is a look specifically at the look at the impact on the RdRand performance...

http://www.phoronix.com/scan.php?page=news_item&px=RdRand-3-Percent

[Adarion]

Ouch. Just ouch. And it is microcode so it will affect everything that makes use of RdRand.
Either that slipped through testing or it is inevitable.
Either way: Ouch.

Good for uncovering that, Michael. (I don't have these processors but others with be puzzled why some things drop in performance without obvious explanation.)

[Aryma]

so we can't disable this patch ?

[discordian]

Meh, stick a fork in it and bring on ARM and RISC-V.

[Michael]

so we can't disable this patch ?

It's mentioned in the article.

[tildearrow]

After all of this, my Skylake Core i7 processor goes head-to-head against the Nintendo GameCube, therefore taking me 19 years back.

[ms178]

AMD should consider thanking Intel at this point for continuously providing more performance regressions every couple of months to their customers! That is one argument more to go with EPYC every time a new one drops.

[Anarchy]

Classic Intel.

I'm looking forward to the shitstorm. This is getting out of hand.

[CochainComplex]

.....

[Pepec9124]

oof

What more can be said to be honest. Intel better has a new uarch planned to replace this mess.