RE: [PATCH v2] cxl/pci: Set the device timestamp

[Dan Williams <dan.j.williams@intel.com>]

Jonathan Cameron wrote:
> CXL r3.0 section 8.2.9.4.2 "Set Timestamp" recommends that the host sets
> the timestamp after every Conventional or CXL Reset to ensure accurate
> timestamps. This should include on initial boot up. The time base that
> is being set is used by a device for the poison list overflow timestamp
> and all event timestamps.  Note that the command is optional and if
> not supported and the device cannot return accurate timestamps it will
> fill the fields in with an appropriate marker (see the specification
> description of each timestamp).
> 
> Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
> 
> ---
> v2: Thanks to DavidLohr, Alison and Dan for quick reviews.
>  - Drop exposure to userspace.
>  - Drop prevention of raw command based access.
>  - Check for a Not Supported return code. If that happens eat the error.
>  - Fix missing endian conversion.
>  - Switch to ktime.h include rather than timekeeping.h as per comments in
>    the headers.
> 
> Based on cxl/pending as of today.
> 
> Open question: Should we only do this if Linux has control of the
> error handling?  In theory it should be safe anyway given the
> specification is clear that the timestamp base should always be the
> same - so subject to small errors we shouldn't cause any firmware first
> handling to get confused.

I think it's safe for the reasons you give, but I will ask our platform
firmware team what they think / give them a heads up that Linux is
planning to do this unconditionally.