On Fri Feb 6, 2026 at 5:13 PM GMT, Danilo Krummrich wrote: > On Fri Feb 6, 2026 at 5:13 PM CET, Gary Guo wrote: >> On Fri Feb 6, 2026 at 4:05 PM GMT, Joel Fernandes wrote: >>> >>> >>> On 2/6/2026 10:53 AM, Danilo Krummrich wrote: >>>> On Fri Feb 6, 2026 at 4:25 PM CET, Gary Guo wrote: >>>>> On Fri Feb 6, 2026 at 12:41 AM GMT, Joel Fernandes wrote: >>>>>> diff --git a/drivers/gpu/Kconfig b/drivers/gpu/Kconfig >>>>>> index 22dd29cd50b5..2c3dec070645 100644 >>>>>> --- a/drivers/gpu/Kconfig >>>>>> +++ b/drivers/gpu/Kconfig >>>>>> @@ -1,7 +1,14 @@ >>>>>> # SPDX-License-Identifier: GPL-2.0 >>>>>> >>>>>> +config RUST_CLIST >>>>>> + bool >>>>>> + depends on RUST >>>>>> + help >>>>>> + Rust abstraction for interfacing with C linked lists. >>>>> >>>>> I am not sure if we need extra config entry. This is fully generic so >>>>> shouldn't >>>>> generate any code unless there is an user. >>>> >>>> I also don't think we need a Kconfig for this. >>>> >>>> In any case, it shouln't be in drivers/gpu/Kconfig. >>> >>> Fair point, I believe I was having trouble compiling this into the kernel >>> crate >>> without warnings (I believe if !GPU_BUDDY). I'll try to drop it and see if >>> we >>> can get rid of it. >> >> If you run into dead code warnings, I think it is fine to just >> >> #[allow(dead_code, reason = "all users might be cfg-ed out")] >> >> the overhead of just let rustc type-checking this module isn't worth the >> extra >> Kconfig plumbing, I think. > > You mean because there are pub(crate) in clist.rs? I don't think the Kconfig > would help with that, nothing prevents people from enabling RUST_CLIST, but > none > of the users. > > Besides that, once we have the new build system, the users of CList are likely > in other crates anyways, so I think we should just change things to pub.
I asked for this to be changed to `pub(crate)` because I think this isn't something that should be used by drivers. As you said, tt might be tricky to enforce that with new build system when subsystems are inside different crates. But until then I think it's better to limit visibility. Best, Gary
