The Performance Impact Of AMD Changing Their Retpoline Method For Spectre V2

[onlyLinuxLuvUBack]

On the 8-core AMD Ryzen 9 5900HX the test with the biggest negative impact was the generic Context Switching test of the Stress-NG suite with the prior default being 22% faster.
On the 16-core AMD Ryzen 9 5950X the test with the biggest negative impact was the generic Context Switching test of the Stress-NG suite with the prior default being 54% faster. That's more than double the impact on twice the amount of cores. I theorize that context switching on higher cores are heavily impacted by the patch.

I would love to see the result for this test on the 8-core EPYC (actually, I really want to see the impact on the 64-core EPYC, but that config wasn't tested), but the result is not listed.

What could phoronix test though with 64 cores?

128 podman containers running nim(c -r) running sqlite3 memdb sql reads ?
256 containers?
what is the load for such a beast?

[Steffo]

I just disabled the mitigations and it seems, that browsing is significant faster!

[Teggs]

You're not alone birdie and @Danny3!!

What the hell is up with those graphs? I mean, what are they tracking? Worse is the further away it is from the center line? Because I see +X% at both sides of it. I'm a simple gal, box and whisker plots are the most I can understand at first glance haha

Not 'worse'. If both perform the same on a test, there is no bar at all. If one is 'better', the bar is on that side (look at the names on the top of the chart, left and right). The bar length and attached number show how much faster/better it was in %.

In the first graph, AMD / LFENCE Prior Default is 54% faster at the Stress-NG Context Switching benchmark. New Default - Generic Retpolines is 3.2% faster at the W.I test for Selenium in Google Chrome.

[SofS]

Michael They made an important consideration, "probably on systems with SMT enabled", have you noticed something on this regard? If so an interesting set of tests would be to compare between the current implementation with SMT enabled and the older implementation with SMT disabled.

[middy]

If this new patch hurts context switching that much, I can only assume gaming performance is going to take a hit. I'll probably run it either with the old patch set or completely off if gaming takes a hit.