Re: [OE-core] [PATCH] go: upgrade to 1.17 release

From: "Richard Purdie" <richard.purdie@linuxfoundation.org>
To: Alexander Kanavin <alex.kanavin@gmail.com>,
	Bruce Ashfield <bruce.ashfield@gmail.com>
Cc: Sakib Sajal <sakib.sajal@windriver.com>,
	Patches and discussions about the oe-core layer
	<openembedded-core@lists.openembedded.org>
Subject: Re: [OE-core] [PATCH] go: upgrade to 1.17 release
Date: Wed, 25 Aug 2021 20:43:11 +0100	[thread overview]
Message-ID: <db691c357f05e5e2e34819ecbec4778223eae0d7.camel@linuxfoundation.org> (raw)
In-Reply-To: <CANNYZj8D+D1zBiT=jdHLOcq0-4W615Qtj--FFhRuDjXxRWX41w@mail.gmail.com>

On Wed, 2021-08-25 at 21:31 +0200, Alexander Kanavin wrote:
> On Wed, 25 Aug 2021 at 21:26, Bruce Ashfield <bruce.ashfield@gmail.com> wrote:
> > I mentioned meta-virtualization in the non-threaded part of this
> > discussion, so that specific information has been covered. Sakib is a
> > contributor there (and isn't new to core), and knows what he's doing,
> > hence why I have the luxury of being less detailed in the reponses.
> > 
> > I've said my bit on this, I'll pick up the pieces if anything ends up
> > breaking .. I'm just trying to be proactive since history shows that
> > go always breaks something.
> > 
> 
> 
> Please try to understand where I am coming from. Yours truly is doing the bulk
> of version updates in oe-core, because no one else bothers to, not even listed
> maintainers (and more and more recipes do not even have a listed maintainer
> anyore). So when someone else is showing initiative in version updates - and
> Sakib has not updated go before - I would much rather show appreciation for
> the effort, rather than immediate (and vague!) criticism for not enough
> testing.
> 

Enough, please :)

Alex does a lot of good work with version upgrades in OE-Core and it is much
appreciated.

Bruce does a lot of good work with the kernels, associated bits and other
things, both in core and as maintainer of meta-virtualization and it is also
much appreciated.

Go is a bit of a problem at the moment as previous upgrades introduced bugs
which the triage team has been unable to get resolved. We know the next version
makes further related changes which *will* cause problems. As such we are
worried about it. I'm close to reverting the last uprev over the lack of
attention to bugs.

OE-Core doesn't use go that much and our test coverage is known to be weak
there. meta-virt does use it much more heavily so asking for that to be tested
with go uprevs is a reasonable request IMO.

We are increasing the test matrix the autobuilder covers and meta-virt is
potentially part of that, as is meta-oe. We do already have some basic check-
layer tests.

I think we all agree that it is too late for 1.17 in this cycle, we have enough
risky problems on the go without that one.

Cheers,

Richard

> 