> On 2020-06-19 23:53, Kiwoong Kim wrote: > > Some SoC specific might need command history for various reasons, such > > as stacking command contexts in system memory to check for debugging > > in the future or scaling some DVFS knobs to boost IO throughput. > > > > What you would do with the information could be variant per SoC > > vendor. > > Isn't this something that should be done in an I/O scheduler instead of in > a SCSI LLD? I don't like the idea behind this patch series at all. > > Bart. If you could get the information, Many would exploit it for their respective purposes. But, it's important for the information to contain accurate timestamps when the driver hooks it, if you're trying to figure out something wrong. As for scaling DVFS knobs to boost UFS throughput, locating in the ufs driver is more Beneficial because SoC vendors have their own power domains and thus lead to make different way of what to scale up to boost. If it's populated in block layer, there is no way to introduce boosting per SoC.
