Not dumb enough to merge BOTH repos into NVIDIA and force push

EEyeCantCU Not dumb enough to merge BOTH repos into NVIDIA and force push

B

bsherman•12/8/23, 4:39 PM

i'm not quite sure how it happened either, that's why I'm bringing it up

E

EyeCantCUOP•12/8/23, 4:39 PM

I did this via GitHub last night so maybe I overlooked something there?

E

EyeCantCUOP•12/8/23, 4:39 PM

Very confused

Bbsherman it is protected... but... i think we allow bypassing of rules by those with perm...

A

akdev•12/8/23, 4:39 PM

Yeah but you still need to force push I believe

Aakdev Yeah but you still need to force push I believe

B

bsherman•12/8/23, 4:39 PM

i agree, i think that's required

E

EyeCantCUOP•12/8/23, 4:40 PM

The PRs never even existed for NVIDIA which makes zero sense

EEyeCantCU The PRs never even existed for NVIDIA which makes zero sense

B

bsherman•12/8/23, 4:40 PM

yep, exactly why i was extra confused

E

EyeCantCUOP•12/8/23, 4:41 PM

Like how would both repos get changed from one PR?

B

bsherman•12/8/23, 4:41 PM

honestly, i'm a little nervous about the pull app

E

EyeCantCUOP•12/8/23, 4:42 PM

Shouldn't the pull app only work one way?

B

bsherman•12/8/23, 4:42 PM

yes, but i had weird behavior and ended up almost pushing stuff to the wrong repo after adding it

B

bsherman•12/8/23, 4:43 PM

i would like to resolve this by force pushing

ublue-os/nvidia

ublue-os/nvidia

ublue-os/nvidia

ublue-os/nvidia such that those last 3 commits never existed

E

EyeCantCUOP•12/8/23, 4:44 PM

I agree

E

EyeCantCUOP•12/8/23, 4:44 PM

Apologies for everyone. Really screwed up there

E

EyeCantCUOP•12/8/23, 4:44 PM

I hadn't even noticed

B

bsherman•12/8/23, 4:44 PM

I only noticed due to build failures... and then i was surprised because it was obvious we didn't have PRs for those commits

E

EyeCantCUOP•12/8/23, 4:45 PM

Maybe GitHub tried to merge both repositories when the PR was opened?

Bbsherman i would like to resolve this by force pushing `ublue-os/nvidia` such that those ...

B

bsherman•12/8/23, 4:45 PM

@EyeCantCU @akdev

I'll do this now

E

EyeCantCUOP•12/8/23, 4:45 PM

Thank you so much

A

akdev•12/8/23, 4:45 PM

Sounds good to me, thanks for fixing it

J

jstone•12/8/23, 4:46 PM

This sounds related to when I requested somebody with the right permissions to sync the asus-nvidia repo, which looks like a fork of the nvidia repo. I wonder what happened there. Did github try to be "smart" by automatically applying merges to the fork and the upstream?

E

EyeCantCUOP•12/8/23, 4:47 PM

I honestly have no idea but that's definitely not smart behavior lol

B

bsherman•12/8/23, 4:47 PM

is that request above?

J

jstone•12/8/23, 4:48 PM

💾ublue-dev

B

bsherman•12/8/23, 4:48 PM

ok, i read it

B

bsherman•12/8/23, 4:48 PM

SEEMS simple...

Jjstone This sounds related to when I requested somebody with the right permissions to s...

B

bsherman•12/8/23, 4:50 PM

thanks for calling out the correlation.

J

jstone•12/8/23, 4:50 PM

The changes seem reasonable in asus-nvidia

J

jstone•12/8/23, 4:50 PM

I only detected the drivers being updated

B

bsherman•12/8/23, 4:52 PM

hmm... it looks like
1) changes synced from nvidia > asus-nvidia
2) changes synced from asus-nvidia > nvidia
3) changes synced from nvidia > surface-nvidia
4) changes synced from surface-nvidia > nvidia

E

EyeCantCUOP•12/8/23, 4:53 PM

With 2 PRs though?

B

bsherman•12/8/23, 4:54 PM

i thinik it's more likely the "sync" operations bypassed something

E

EyeCantCUOP•12/8/23, 4:54 PM

Shoot

B

bsherman•12/8/23, 4:54 PM

clearly there was no PR to merge from asus/surface > nvidia

B

bsherman•12/8/23, 4:57 PM

hah, i can't even force push a revert, so clearly you didn't do that @EyeCantCU it had to be some use ofthe github tooling

B

bsherman•12/8/23, 4:58 PM

Ok, fixed

ublue-os/nvidia

ublue-os/nvidia

ublue-os/nvidia

ublue-os/nvidia

$ git push --force
Total 0 (delta 0), reused 0 (delta 0), pack-reused 0
remote: Bypassed rule violations for refs/heads/main:
remote:
remote: - Changes must be made through the merge queue
remote:
remote: - Changes must be made through a pull request.
remote:
remote: - Cannot force-push to this branch
remote:
To github.com:ublue-os/nvidia.git
 + 83ca789...8119dd2 main -> main (forced update)

$ git push --force
Total 0 (delta 0), reused 0 (delta 0), pack-reused 0
remote: Bypassed rule violations for refs/heads/main:
remote:
remote: - Changes must be made through the merge queue
remote:
remote: - Changes must be made through a pull request.
remote:
remote: - Cannot force-push to this branch
remote:
To github.com:ublue-os/nvidia.git
 + 83ca789...8119dd2 main -> main (forced update)

$ git push --force
Total 0 (delta 0), reused 0 (delta 0), pack-reused 0
remote: Bypassed rule violations for refs/heads/main:
remote:
remote: - Changes must be made through the merge queue
remote:
remote: - Changes must be made through a pull request.
remote:
remote: - Cannot force-push to this branch
remote:
To github.com:ublue-os/nvidia.git
 + 83ca789...8119dd2 main -> main (forced update)

$ git push --force
Total 0 (delta 0), reused 0 (delta 0), pack-reused 0
remote: Bypassed rule violations for refs/heads/main:
remote:
remote: - Changes must be made through the merge queue
remote:
remote: - Changes must be made through a pull request.
remote:
remote: - Cannot force-push to this branch
remote:
To github.com:ublue-os/nvidia.git
 + 83ca789...8119dd2 main -> main (forced update)

E

EyeCantCUOP•12/8/23, 4:59 PM

What a mess

B

bsherman•12/8/23, 4:59 PM

re-running the nvidia build for today

B

bsherman•12/8/23, 5:29 PM

regardless, no nvidia builds today due to upstream fedora/rpmfusion churn

E

EyeCantCUOP•12/8/23, 5:54 PM

@jstone refreshed supergfxctl and the plasmoid. Combatting some problems building the GNOME shell extension but that should be done here soon

J

jstone•12/8/23, 6:18 PM

Ah interesting, what's going on here? https://download.copr.fedorainfracloud.org/results/eyecantcu/supergfxctl/fedora-39-x86_64/06734840-gnome-shell-extension-supergfxctl-gex/builder-live.log.gz

Jjstone Ah interesting, what's going on here? https://download.copr.fedorainfracloud.org...

E

EyeCantCUOP•12/8/23, 6:21 PM

Looks like deps changed a bit or something but the instructions haven't changed so I'm kind of throwing stuff at the wall

EEyeCantCU Looks like deps changed a bit or something but the instructions haven't changed ...

J

jstone•12/8/23, 6:29 PM

Has the package moved? https://www.npmjs.com/package/undici-types

npm

undici-types

A stand-alone types package for Undici. Latest version: 6.0.1, last published: 2 days ago. Start using undici-types in your project by running

npm i undici-types

npm i undici-types

npm i undici-types

npm i undici-types. There are 67 other projects in the npm registry using undici-types.

J

jstone•12/8/23, 6:30 PM

I'm not very familiar with NPM

Jjstone Has the package moved? https://www.npmjs.com/package/undici-types

E

EyeCantCUOP•12/8/23, 6:32 PM

I'm not either. Kind of confused because it says it installs then complains during build that it can't find it

E

EyeCantCUOP•12/8/23, 6:33 PM

And I'm almost certain they didn't change tsconfig.json but not upload it?

EEyeCantCU And I'm almost certain they didn't change tsconfig.json but not upload it?

J

jstone•12/8/23, 6:45 PM

I have an idea -- could we use this as a source? https://extensions.gnome.org/extension/5344/supergfxctl-gex/

J

jstone•12/8/23, 6:45 PM

I can confirm that installing this as a user extension works in Gnome 45

E

EyeCantCUOP•12/8/23, 6:45 PM

Would solve the issue of needing to build it. Let me try

J

jstone•12/8/23, 6:46 PM

It would just have to be for the F39 build, F38 should be fine

Not dumb enough to merge BOTH repos into NVIDIA and force push

Similar Threads

Similar Threads

Similar Threads