Re: [PATCH v6 2/6] iommufd: Add IOMMU_HWPT_INVALIDATE

[Nicolin Chen]

On Tue, Nov 21, 2023 at 02:50:05AM +0000, Tian, Kevin wrote:
> > From: Nicolin Chen <nicolinc@xxxxxxxxxx>
> > Sent: Tuesday, November 21, 2023 1:37 AM
> >
> > On Mon, Nov 20, 2023 at 08:34:58AM +0000, Tian, Kevin wrote:
> > > > From: Liu, Yi L <yi.l.liu@xxxxxxxxx>
> > > > Sent: Monday, November 20, 2023 4:30 PM
> > > >
> > > > On 2023/11/20 16:09, Tian, Kevin wrote:
> > > > >> From: Liu, Yi L <yi.l.liu@xxxxxxxxx>
> > > > >> Sent: Friday, November 17, 2023 9:07 PM
> > > > >> + * @req_len: Length (in bytes) of a request entry in the request array
> > > > >> + * @req_num: Input the number of cache invalidation requests in the
> > > > array.
> > > > >> + *           Output the number of requests successfully handled by
> > kernel.
> > > > >> + * @out_driver_error_code: Report a driver speicifc error code upon
> > > > failure.
> > > > >> + *                         It's optional, driver has a choice to fill it or
> > > > >> + *                         not.
> > > > >
> > > > > Being optional how does the user tell whether the code is filled or not?
> >
> > Well, naming it "error_code" indicates zero means no error while
> > non-zero means something? An error return from this ioctl could
> > also tell the user space to look up for this driver error code,
> > if it ever cares.
> 
> probably over-thinking but I'm not sure whether zero is guaranteed to
> mean no error in all implementations...

Well, you are right. Usually HW conveniently raises a flag in a
register to indicate something wrong, yet it is probably unsafe
to say it definitely.

> > > > seems like we need a flag for it. otherwise, a reserved special value is
> > > > required. or is it enough to just document it that this field is available
> > > > or not per the iommu_hw_info_type?
> > > >
> > >
> > > No guarantee that a reserved special value applies to all vendors.
> > >
> > > I'll just remove the optional part. the viommu driver knows what a valid
> > > code is, if the driver fills it.
> >
> > Hmm, remove out_driver_error_code? Do you mean by reusing ioctl
> > error code to tell the user space what driver error code it gets?
> >
> 
> No. I just meant to remove the last sentence "It's optional ..."

OK. Would it force all IOMMU drivers to report something upon
an error/failure?

Thanks
Nic