Skip to contentSkip to conversation

How stable is the v2 API?

libgpiod Ben Hutchings 3 participants 7 messages Jul 12, 2021 – Jul 15, 2021 GPIO

lore
Share & tools
lore
#0 lore raw

I'm working on a project that requires GPIO debouncing and is planned to use libgpiod. But debouncing is only available on the unreleased next/libgpiod-2.0 branch.

What we'd like to know is whether the current C API on that branch is expected to change much before release. I don't expect any commitment to API or ABI stability, but it would be helpful to have some indication of how much change is likely to be needed in a client that is written for the current API.

Thanks,

Ben.

#1 lore raw
Quoted 10 lines · Ben Hutchings

I'm working on a project that requires GPIO debouncing and is planned to use libgpiod. But debouncing is only available on the unreleased next/libgpiod-2.0 branch.

What we'd like to know is whether the current C API on that branch is expected to change much before release. I don't expect any commitment to API or ABI stability, but it would be helpful to have some indication of how much change is likely to be needed in a client that is written for the current API.

I believe this is the right thread to discuss and influence if needed.

https://lore.kernel.org/linux-gpio/[email protected]/T/#ma70b1ab80804c09644f4a018ff5eff10464b195a

Signature · 2 lines
With Best Regards,
Andy Shevchenko
#2 lore raw
Quoted 14 lines · Andy Shevchenko
Quoted 10 lines · Ben Hutchings

I'm working on a project that requires GPIO debouncing and is planned to use libgpiod. But debouncing is only available on the unreleased next/libgpiod-2.0 branch.

What we'd like to know is whether the current C API on that branch is expected to change much before release. I don't expect any commitment to API or ABI stability, but it would be helpful to have some indication of how much change is likely to be needed in a client that is written for the current API.

Quoted 3 lines

I believe this is the right thread to discuss and influence if needed.

https://lore.kernel.org/linux-gpio/[email protected]/T/#ma70b1ab80804c09644f4a018ff5eff10464b195a

Oops, it was a follow up, but the core bindings were here:
https://lore.kernel.org/linux-gpio/20210607045950.GA18318@sol/T/#u

Signature · 2 lines
With Best Regards,
Andy Shevchenko
#3 lore raw
In reply to Andy Shevchenko
Quoted 13 lines · Andy Shevchenko
Quoted 10 lines · Ben Hutchings

I'm working on a project that requires GPIO debouncing and is planned to use libgpiod. But debouncing is only available on the unreleased next/libgpiod-2.0 branch.

What we'd like to know is whether the current C API on that branch is expected to change much before release. I don't expect any commitment to API or ABI stability, but it would be helpful to have some indication of how much change is likely to be needed in a client that is written for the current API.

Quoted 3 lines

I believe this is the right thread to discuss and influence if needed.

https://lore.kernel.org/linux-gpio/[email protected]/T/#ma70b1ab80804c09644f4a018ff5eff10464b195a

Thanks. I've actually seen that already. I don't think I need to exert any "influence" as the current v2 API seems to do cover our needs.

Ben.

#4 lore raw
In reply to Ben Hutchings
Quoted 11 lines · Ben Hutchings

I'm working on a project that requires GPIO debouncing and is planned to use libgpiod. But debouncing is only available on the unreleased next/libgpiod-2.0 branch.

What we'd like to know is whether the current C API on that branch is expected to change much before release. I don't expect any commitment to API or ABI stability, but it would be helpful to have some indication of how much change is likely to be needed in a client that is written for the current API.

Quoted 4 lines

Thanks,

Ben.

Hi Ben!

While - as you already assumed - there are no commitments to any API & ABI stability yet, the interface should not change very much. I have a single set of API changes in my queue that we discussed while reviewing the C++ bindings and I don't expect there to be many more coming after that.

Again: this is an estimation, not a promise.

Best Regards
Bartosz

#5 lore raw

On Tue, Jul 13, 2021 at 10:10:12PM +0200, Bartosz Golaszewski wrote:
[...]

Quoted 5 lines

While - as you already assumed - there are no commitments to any API & ABI stability yet, the interface should not change very much. I have a single set of API changes in my queue that we discussed while reviewing the C++ bindings and I don't expect there to be many more coming after that.

Is that single set of API changes the same as
<https://patchwork.ozlabs.org/project/linux-gpio/list/?series=250005>
or do you have more beyond that?

Quoted 1 line

Again: this is an estimation, not a promise.

Understood.

Thanks,

Ben.

#6 lore raw
Quoted 9 lines · Ben Hutchings

On Tue, Jul 13, 2021 at 10:10:12PM +0200, Bartosz Golaszewski wrote:
[...]

Quoted 5 lines

While - as you already assumed - there are no commitments to any API & ABI stability yet, the interface should not change very much. I have a single set of API changes in my queue that we discussed while reviewing the C++ bindings and I don't expect there to be many more coming after that.

Quoted 4 lines

Is that single set of API changes the same as
<https://patchwork.ozlabs.org/project/linux-gpio/list/?series=250005>
or do you have more beyond that?

These have already been squashed with the top commit in the next/libgpiod-2.0 branch (except for the last one). I have two more that I will try to send out today. I'll Cc you.

Bart

Quoted 7 lines
Quoted 1 line

Again: this is an estimation, not a promise.

Understood.

Thanks,

Ben.