Re: [PATCH v3] io: add io_pgtable abstraction

[Robin Murphy]

On 2025-12-01 9:58 am, Alice Ryhl wrote:
[...]
> > > We need a different signature if it's possible to have mapped != 0 when
> > > returning an error.
> >
> > Aha, thanks for clarifying - indeed this is not the common "value or error"
> > case, it is two (almost) orthogonal return values. However if we're not
> > permitting callers to try to do anything clever with -EEXIST then it might
> > make sense to just embed the inevitable cleanup-on-failure boilerplate here
> > anyway (even if we still leave retry-on-partial-success to the caller).
>
> Is the only possible error -EEXIST? I could encode that in the API if
> that is the case.

No, I was just calling out -EEXIST as the only error where I imagine a 
caller *might* want to continue without cleaning up a partial mapping, 
if for instance they were playing clever tricks like a background 
mapping of a large buffer while already allowing other threads to 
eagerly demand-page bits of it, so the "main" mapping thread just 
adjusts and restarts to skip over already-present pages. Other errors 
are still possible, but generally represent terminal failure conditions 
at the caller's level too - in practice things like -EINVAL and -ENOMEM 
are likely to happen before any mappings can be made, but io-pgtable 
doesn't guarantee any particular behaviour here, so a well-behaved 
caller should still generally handle cleaning up after an error (at 
least if they intend to keep trying to use the pagetable beyond that point).

Cheers,
Robin.