Benchmarking The Updated Intel CPU Microcode For SRBDS / CrossTalk Mitigation
Yesterday afternoon I did note that the actual RdRand performance is certainly slaughtered with this updated microcode. When simply measuring the RdRand performance, the mitigated performance is indeed down to ~3% the original performance.
On the Xeon E3-1275 v6 system, indeed when steressing RdRand the new microcode lowered the performance to just 3% the original performance.
The "good" news is that RdRand isn't used prolificily by real-world workloads. So when stressing a wide range of server-ish workloads, the overall impact from the SRBDS mitigation was not impacted to within a few percent the original performance as a worst case. It was just in a few cases the SRBDS mitigated microcode slightly lowered performance while for most tests the performance was fortunately stable. The dozens of flat benchmark results omitted.