ok, sounds good. thanks. I think I had a version that does that. I'll
revert back to that version.
I'll let the maintainer there know so that we can proceed with that direction.
Willy Tu
On Tue, Nov 8, 2022 at 11:37 AM Alexander Kanavin
wrote:
>
> Ah, I thought this was in a separate layer. I
Ah, I thought this was in a separate layer. I think you're best off
getting a private copy of openssh sources then.
Alex
On Tue, 8 Nov 2022 at 20:30, Willy Tu wrote:
>
> So do you suggest adding bbappend in meta-openembedded where
> https://github.com/openembedded/meta-openembedded/pull/597 wil
So do you suggest adding bbappend in meta-openembedded where
https://github.com/openembedded/meta-openembedded/pull/597 will live?
I'll check with the maintainer there. And Google/hiba users will maintain
that.
Best,
Willy Tu
On Tue, Nov 8, 2022 at 11:25 AM Alexander Kanavin
wrote:
> Actuall
Actually, you don't need to do this in HIBA - a bbappend to openssh will do.
Alex
On Tue, 8 Nov 2022 at 20:21, Alexander Kanavin via
lists.openembedded.org
wrote:
>
> I'd rather have this hidden in your recipes, yes. It's clearly a
> special needs issue, and something you shouldn't ask us to mai
I'd rather have this hidden in your recipes, yes. It's clearly a
special needs issue, and something you shouldn't ask us to maintain
for you (what I mean is that you're not going to rebase the patch when
a new openssh version is out, or adjust do_install - this falls on
usually me, and my time to d
I think they are fine with the hack here. oops. Ok, I'll work with
them to get the fix without this change.
Thanks for the feedback.
Willy Tu
On Tue, Nov 8, 2022 at 9:12 AM Alexander Kanavin wrote:
>
> On Tue, 8 Nov 2022 at 18:08, Willy Tu wrote:
> >
> > Thanks Alexander for the feedback,
> >
On Tue, 8 Nov 2022 at 18:08, Willy Tu wrote:
>
> Thanks Alexander for the feedback,
>
> We had some discussion internally with one of the OpenSSH maintainers
> and they are fine with it. I'll try to see if there is a public forum
> where we can share the discussion.
If they are fine with it, then
Thanks Alexander for the feedback,
We had some discussion internally with one of the OpenSSH maintainers
and they are fine with it. I'll try to see if there is a public forum
where we can share the discussion.
Willy Tu
On Tue, Nov 8, 2022 at 8:27 AM Alexander Kanavin wrote:
>
> The correct way
The correct way to do this is to discuss the issue with openssh
upstream, and provide a link to any such discussion if it has already
happened. If they don't want to install these things, they have
reasons, and you need to at least try to convince them otherwise.
We generally do not go against ups
Is there anything else that I need to do to help move this forward?
Willy Tu
On Mon, Oct 31, 2022 at 8:42 AM Willy Tu via lists.openembedded.org wrote:
> Friendly ping. Is there any more concern on this?
>
> Willy Tu
>
>
> On Mon, Oct 24, 2022 at 10:07 AM Willy Tu via lists.openembedded.org
>
Friendly ping. Is there any more concern on this?
Willy Tu
On Mon, Oct 24, 2022 at 10:07 AM Willy Tu via lists.openembedded.org
wrote:
>
> Testing it out more and it seems like the current change is fine in that
> regard.
>
> I was building it against an OpenBMC image which include the change
Testing it out more and it seems like the current change is fine in that
regard.
I was building it against an OpenBMC image which include the change and the
header and libssh.a files doesn't show up in the image itself
```
./tmp/work/izumi-openbmc-linux-gnueabi/obmc-phosphor-image
```
doesn't hav
I am testing it out again and it seems like it does install to the target .
```
$ find -name libssh.a
./tmp/sysroots-components/armv7a/openssh/usr/lib/libssh.a
./tmp/work/armv7a-openbmc-linux-gnueabi/openssh/9.0p1-r0/image/usr/lib/libssh.a
./tmp/work/armv7a-openbmc-linux-gnueabi/openssh/9.0p1-r0/s
Adding +Tristan Lelong for visibility and comment on
any concern.
On Mon, Oct 24, 2022 at 7:27 AM Willy Tu via lists.openembedded.org wrote:
> Hi Michael,
>
> Thanks for the response. Sorry for the delay. I must have missed the email
> when I was setting up the filters. I have sent the patch di
Hi Michael,
Thanks for the response. Sorry for the delay. I must have missed the email
when I was setting up the filters. I have sent the patch directly to you
privately to see if that works.
As for
```
git config --global sendemail.from "w...@google.com"
```
,
I think I already have to setup pr
Hi Willy
Thanks for the patch!
On 10/21/22 15:17, Willy Tu via lists.openembedded.org wrote:
Install API only if opensshinternals is in DISTRO_FEATURES.
Install the OpenSSH header + lib files for other repo to build using it.
For example, https://github.com/google/hiba is dependent on those fil
Hi Ross,
Thanks for the feedback.
Can you point to the right repo to update? Just to make sure that I
understand the concern, you are not referring to
https://github.com/openembedded/meta-openembedded/pull/597, right? I am
making changes to
http://cgit.openembedded.org/openembedded-core/tree/meta
Hi,
oe-core doesn’t accept merge requests on the github mirrors, and you pushed to
meta-oe, which is a different repository.
>> + if ${@bb.utils.contains('DISTRO_FEATURES', 'opensshinternals', 'true',
>> 'false', d)}; then
This should be a PACKAGECONFIG, as it’s not a distro-wide setting.
>>
just for visibility.
On Fri, Oct 21, 2022 at 6:17 AM Willy Tu via lists.openembedded.org
wrote:
>
> Install API only if opensshinternals is in DISTRO_FEATURES.
>
> Install the OpenSSH header + lib files for other repo to build using it.
> For example, https://github.com/google/hiba is dependent
Install API only if opensshinternals is in DISTRO_FEATURES.
Install the OpenSSH header + lib files for other repo to build using it.
For example, https://github.com/google/hiba is dependent on those files
to build and this change is needed to meet that requirement.
https://github.com/openembedded
20 matches
Mail list logo
