DPC Latency

[genetix]

It's not the HPET. Tested it up a bit and there's no change on latency jumps.

Most be something else.

-located-

Well, it seems this issue is releated to S.M.A.R.T. values scan, if I dump 'SMART Dump' I get 3 bars every time all over 130K(μs).

Even if I disable the at AIDA64 stability the 'Low Level SMART operations' the SMART values are still at AIDA64 and so is harddrive temperatures. This is some Gigabyte own design of reading smart even through Intel RAID. This cannot be disabled by the BIOS.

NOTE: I do suspect there's more than just the S.M.A.R.T., but would need an build where this is not read to verify where the next hanging section could be located. (as said these is most likely more than 1 issue as I said on my original post).

-edit-

* Loading 'PCI Bus Operations' causes jump ~65K(μs)

* Issues could be also caused by OCZ Agility 2 'known issues' at scanning S.M.A.R.T. 'B0' (what ever this is ;P), but this cannot be disabled by the board.

-edit-

It's not the OCZ Hard drive S.M.A.R.T. issue. Removed the OCZ harddrive and cloned the root to Intel RAID-0 with Samsung F1's. Problem seems to be at the Gigabyte unique S.M.A.R.T. for ICH10R as problem occures at 2 places: Storage > SMART and at Sensor pages. There is a smaller jump at 'Motherboard->ACPI', but that's still on green limits and assuming this is normal.

[Fiery]

Your tests seem to prove my statement above:

In case the latency spikes are approx. 20 seconds away from each other -- as it seems --, then I suppose it's caused by the SMART detection layer of AIDA64, which is used to measure HDD/SSD temperature. AIDA64 now uses a special driver call to pass through the Intel Rapid Storage RAID drivers, in order to reach the RAID member drives. I guess Intel drivers will need some more iterations to mature

Let's hope Intel can iron out the issues in their Rapid Storage RAID drivers soon