These results are interesting, especially aver the Cascalake test https://www.phoronix.com/scan.php?pa...mds-xeon&num=5 , where you conclude that the H/W mitigations on Cascadelake make a pretty big impact.
And now the results in this test are showing no difference at all. Which seems after seeing the above linked Cascade Lake test a little surprising.
Then i looked into it and in this document by Intel ( https://www.intel.com/content/www/us...-hardware.html )
It shows that the 9400F can come in 2 flavours, a Intel64 Family 6 Model 158 Stepping 12 and a Intel64 Family 6 Model 158 Stepping 13, where the "Model 12" doesn't have the hardware mitigations in the same fashion as Cascade Lake has, but the "Model 13" does have full H/W mitigations.
In your review i can't see you mentioning if you tested a Intel64 Family 6 Model 158 Stepping 12 9400F Sku or a Intel64 Family 6 Model 158 Stepping 13 9400F Sku, however this could potentionally make quite the difference.
If looking at the geometric mean of all the benchmarks carried out, the EPYC 7601 averages out to about a 1% hit with its Spectre mitigations. The dual Xeon Platinum 8280 Cascadelake setup with its mostly hardware-based mitigations was slower by 4% with the relevant mitigations enabled . Meanwhile the dual Xeon Gold 6138 server that unfortunately doesn't have the hardware mitigations saw a 11% hit from the benchmarks run with these Spectre/Meltdown/L1TF/MDS mitigations.
Then i looked into it and in this document by Intel ( https://www.intel.com/content/www/us...-hardware.html )
It shows that the 9400F can come in 2 flavours, a Intel64 Family 6 Model 158 Stepping 12 and a Intel64 Family 6 Model 158 Stepping 13, where the "Model 12" doesn't have the hardware mitigations in the same fashion as Cascade Lake has, but the "Model 13" does have full H/W mitigations.
In your review i can't see you mentioning if you tested a Intel64 Family 6 Model 158 Stepping 12 9400F Sku or a Intel64 Family 6 Model 158 Stepping 13 9400F Sku, however this could potentionally make quite the difference.
Comment