Skip to Content.
Sympa Menu

rare-users - Re: [RARE-users] [freertr] Fwd: RARE/RARE-NIX - Commits pushed

Subject: RARE user and assistance email list

List archive

Re: [RARE-users] [freertr] Fwd: RARE/RARE-NIX - Commits pushed


Chronological Thread 
  • From: Alexander Gall <>
  • To: <>
  • Cc: <>
  • Subject: Re: [RARE-users] [freertr] Fwd: RARE/RARE-NIX - Commits pushed
  • Date: Tue, 3 Jan 2023 10:11:21 +0100

On Tue, 3 Jan 2023 08:49:16 +0100, mc36 <> said:

> hi,
> On 1/3/23 08:45, wrote:
>> Hi csaba
>>
>> On Tue, 3 Jan 2023 06:38:17 +0100, mc36 <> said:
>>
>>> thanks! it works like a charm... :)
>>
>> Good to hear :) As you've seen, I also modified the optimizer to make
>> it easier to generate profiles that are guaranteed to work with the
>> Nix packaging. The compiler has become so sensitive to changes in the
>> p4 code that it's now basically mandatory to re-optimize before
>> commiting the p4 changes to the rare-nix repo, otherwise the Hydra
>> builds will fail and tna-install-* breaks :(
>>

> yeahhh, thanks for doing so... shall i need to rerun it by hand,
> or did i saw correctly that you're automating it to the builds?

Currently you'll have to do it by hand. It can't be integrated into
the rare-nix build because the new profiles have to be commited to the
rare repo first (and then to rare-nix). What we could do is trigger
the optimizer when we push to the rare repo via the Bitbucket CI.
Then we'd still have to commit the new profiles. I wouldn't object to
an automated commit in that case but we'd need to be careful to not
create a loop since pushing that commit would trigger antoher
optimizer run :)

That should not be a problem because the optimizer is supposed to be
idempotent. Still, I think I would prefer to push the automated commit
to a separate branch first and create a merge request that must be
acked manually.

--
Alex



Archive powered by MHonArc 2.6.19.

Top of Page