On Tue, 2 Aug 2022 11:04:52 -0300 Jason Gunthorpe <jgg@xxxxxxxxxx> wrote: > On Mon, Aug 01, 2022 at 12:42:53PM -0600, Alex Williamson wrote: > > On Wed, 27 Jul 2022 11:37:02 +0530 > > Abhishek Sahu <abhsahu@xxxxxxxxxx> wrote: > > > > > On 7/26/2022 10:53 PM, Jason Gunthorpe wrote: > > > > On Tue, Jul 26, 2022 at 06:17:18PM +0530, Abhishek Sahu wrote: > > > >> Thanks Alex for your thorough review of uAPI. > > > >> I have incorporated all the suggestions. > > > >> Following is the updated uAPI. > > > >> > > > >> /* > > > >> * Upon VFIO_DEVICE_FEATURE_SET, allow the device to be moved into a low power > > > >> * state with the platform-based power management. Device use of lower power > > > >> * states depends on factors managed by the runtime power management core, > > > >> * including system level support and coordinating support among dependent > > > >> * devices. Enabling device low power entry does not guarantee lower power > > > >> * usage by the device, nor is a mechanism provided through this feature to > > > >> * know the current power state of the device. If any device access happens > > > >> * (either from the host or through the vfio uAPI) when the device is in the > > > >> * low power state, then the host will move the device out of the low power > > > >> * state as necessary prior to the access. Once the access is completed, the > > > >> * device may re-enter the low power state. For single shot low power support > > > >> * with wake-up notification, see > > > >> * VFIO_DEVICE_FEATURE_LOW_POWER_ENTRY_WITH_WAKEUP below. Access to mmap'd > > > >> * device regions is disabled on LOW_POWER_ENTRY and may only be resumed after > > > >> * calling LOW_POWER_EXIT. > > > >> */ > > > >> #define VFIO_DEVICE_FEATURE_LOW_POWER_ENTRY 3 > > > >> > > > >> /* > > > >> * This device feature has the same behavior as > > > >> * VFIO_DEVICE_FEATURE_LOW_POWER_ENTRY with the exception that the user > > > >> * provides an eventfd for wake-up notification. > > > > > > > > It feels like this should be one entry point instead of two. > > > > > > > > A flag "automatic re-sleep" and an optional eventfd (-1 means not > > > > provided) seems to capture both of these behaviors in a bit clearer > > > > and extendable way. > > > > I think the mutual exclusion between re-entrant mode and one-shot is > > quite a bit more subtle in the version below, so I don't particularly > > find this cleaner. Potentially we could have variant drivers support > > one w/o the other in the previously proposed model as well. It's > > interesting to see this suggestion since since we seem to have a theme > > of making features single purpose elsewhere. Thanks, > > It is still quite single purpose, just > VFIO_DEVICE_LOW_POWER_REENTERY_DISABLE is some minor customization of > that single purpose. > > Either the flag is set or not, it isn't subtle.. The subtlety is that there's a flag and a field and the flag can only be set if the field is set, the flag can only be clear if the field is clear, so we return -EINVAL for the other cases? Why do we have both a flag and a field? This isn't like we're adding a feature later and the flag needs to indicate that the field is present and valid. It's just not a very clean interface, imo. Thanks, Alex
