Hacker News Reader

Heh, I'm curious if he used the mitigations for all the side channel flaws for the intel processors.

The mitigations don't affect CPU bound benchmarks [1] which don't call into the kernel or use specific user-space mitigations, so it won't matter here.

[1] There are some rare exceptions, such as https://travisdowns.github.io/blog/2019/03/19/random-writes-... , but it is unlikely to matter here.

Smt on/off has a large effect.

It's a single threaded test, so I don't think that matters here.

True. But generally it affects cpu bound benchmarks.

SMT off might mean not enough spare threads to run the OS telemetry.

This is Linux but I don't think that would be true even on Windows.

That may have been true, but it is rather dramatically false with the new JCC erratum workaround.

It’s also false if you’re using a hypervisor that mitigates the iTLB multihit issue.

Good point, I forgot about that one, although here there is only a single hot loop with one jump so a high chance the crossing doesn't occur, and even if it does the IPC is low enough the legacy decoder probably does OK (although it adds a cycle or two to misprediction recovery, which matters here).

So it's something worth checking.

No hypervisor involved.

Reply on news.ycombinator.com