Re: [Regression] Commit "nvme/pci: Use host managed power state for suspend" has problems

[Keith Busch]

On Thu, Aug 01, 2019 at 02:05:54AM -0700, Kai-Heng Feng wrote:
> at 06:33, Rafael J. Wysocki <rafael@xxxxxxxxxx> wrote:
> > On Thu, Aug 1, 2019 at 12:22 AM Keith Busch <kbusch@xxxxxxxxxx> wrote:
> >
> >> In which case we do need to reintroduce the HMB handling.
> >
> > Right.
> 
> The patch alone doesnât break HMB Toshiba NVMe I tested. But I think itâs  
> still safer to do proper HMB handling.

Spec requires host request controller release HMB for D3cold. I suspect
you're only getting to D3hot.