I assume this is not a problem if I disable hyperthreading?
Probably you'll lose more performance.
As always, this depends on your workload.
Bisected: The Unfortunate Reason Linux 4.20 Is Running Slower
11–20 of 114 posts
Re: Bisected: The Unfortunate Reason Linux 4.20 Is Running Slower
#12Re: Bisected: The Unfortunate Reason Linux 4.20 Is Running Slower
#13I get slower during 4.20 moments too...
Re: Bisected: The Unfortunate Reason Linux 4.20 Is Running Slower
#14I get slower during 4.20 moments too...
If only the kernel went more creative..
Re: Bisected: The Unfortunate Reason Linux 4.20 Is Running Slower
#15Earlier quoted context omitted.
As always, this depends on your workload.
As of today still most of the software I use seems to get stuck in one core of my CPU when calculating stuff so I'd argue that by enabling HT I'm losing performance.
That's not really clear from the article, but I suspect that performance drop is from multithreaded software. And sure, if you are observing performance degradation with hyper-threading, there's 0 reason to keep it enabled. OpenBSD guys suggest to turn it off after all.
Re: Bisected: The Unfortunate Reason Linux 4.20 Is Running Slower
#16Does the article not actually mention the reason? (I mean, I know it's STIBP, but it never says that?
Re: Bisected: The Unfortunate Reason Linux 4.20 Is Running Slower
#17Does the article not actually mention the reason? (I mean, I know it's STIBP, but it never says that?
it's on page two. Why a short article like this has pages though...
Re: Bisected: The Unfortunate Reason Linux 4.20 Is Running Slower
#18If you could turn off speculative execution completely I bet that would eliminate all the security concerns, but performance would be abysmal. The Linux kernel really can't be a "one size fits all" type of thing --- an environment where no mutually untrusting code is run has very different security requirements from shared cloud hosts, for example. Personally I think it's all a bit overblown, probably due to owners of the latter environments.
Re: Bisected: The Unfortunate Reason Linux 4.20 Is Running Slower
#19Is Intel already selling CPUs which don't need these patches?
Not in x86 architecture.
Re: Bisected: The Unfortunate Reason Linux 4.20 Is Running Slower
#20Earlier quoted context omitted.
As of today still most of the software I use seems to get stuck in one core of my CPU when calculating stuff so I'd argue that by enabling HT I'm losing performance.
That's not really clear from the article, but I suspect that performance drop is from multithreaded software. And sure, if you are observing performance degradation with hyper-threading, there's 0 reason to keep it enabled. OpenBSD guys suggest to turn it off after all.
The article claims a 30% perf slowdown with PHP. The majority of PHP workloads are single threaded. That causes me to suspect this impacts single core workloads too.