qmd-progress icon indicating copy to clipboard operation
qmd-progress copied to clipboard

Kernel

Open linneamk opened this issue 4 years ago • 10 comments

O(N) implementation of pre-conditioned kernel calculation using implicit Fermi-Dirac response calculations.


This change is Reviewable

linneamk avatar May 12 '21 18:05 linneamk

Sorry for the delay @linneamk . Our CI tests are currently broken because we were using Travis-CI who changed their business model and stopped offering their free open source software support. We will have to switch to GitHub Actions before we can run CI again. If anyone else can verify manually that this PR is passing the tests I am happy to merge it.

nicolasbock avatar May 14 '21 17:05 nicolasbock

Could you rebase this PR? We just merged a fix for CI into master.

nicolasbock avatar Jun 04 '21 02:06 nicolasbock

I rebased the change.

nicolasbock avatar Jun 04 '21 12:06 nicolasbock

Hi! Sorry for my delay, I have updated my code and fixed the test.

linneamk avatar Jul 16 '21 16:07 linneamk

Thanks @linneamk . You need to rebase your changes. Right now the merge conflicts prevent us from merging the pull request.

Thanks!

nicolasbock avatar Jul 17 '21 11:07 nicolasbock

Thanks @linneamk .

Something went wrong during the rebase though. What commands did you run to rebase?

nicolasbock avatar Jul 18 '21 11:07 nicolasbock

Hi @linneamk

I spent a good hour just now trying to rebase your commits onto the current master branch. Unfortunately I ran into a wall and was not able to complete the rebase.

You had based the original PR on a rather old commit (4e73df7664d468ed0ba213bf21768ef41a5adb3d). I rebased that commit in https://github.com/lanl/qmd-progress/pull/169#issuecomment-854667688 as 0d6316ba4207841877075fa6dbd8d24e5436f7ba.

However, you ignored that rebase and force pushed 02e93a6c87bbfaa9fdfd775ee5c4b77c10f0f0fd which was still based on this old commit.

After asking you to rebase you merged your branch with master instead and pushed 6304e52fbc1f45de355658e384d72acab31c580a which is quite frankly a bit of a mess. Your PR now contains duplicate commits

* 6304e52 Fixed indentation
*   f5eeefd Rebased
|\  
| *   1fe5601 Merge branch 'master' into kernel
| |\  
| * | 02e93a6 Fixed test
| * | 2ac9f50 Added kernel routine
| * | 7960bcf O(N) pre-conditioned kernel matrix with implicit FD-expansion response
| * | 383736e Added kernel routines for XL-BOMD
* | | 1256e74 Fixed test
* | | da4eb81 Added kernel routine
* | | bf9a368 O(N) pre-conditioned kernel matrix with implicit FD-expansion response
* | | ce0e474 Added kernel routines for XL-BOMD
| |/  
|/|   
* |   eab14aa Merge pull request #201 from nicolasbock/ci_scripts

and it's unclear to me whether your changes are even correctly represented in your branch.

Merging does not always produce the desired code since the merging algorithm is really pretty dumb because it doesn't know what your intention and the intention of the person making changes in the branch you merged with was when the changes were made. It's really a best guess kind of effort.

Could you carefully go through your original changes and apply them to a fresh branch that is based on the current master branch?

Please don't hesitate to ask if any of what I wrote is unclear. Git is a little tricky to use :smile:

Thanks.

nicolasbock avatar Sep 17 '21 22:09 nicolasbock

Hi!

Wow, I'm sorry and thank you for trying to clean up my mess 😅 I will try to create a new branch and commit my changes to, I'll let you know if I need any help. Thanks!

linneamk avatar Sep 18 '21 11:09 linneamk

Thanks @linneamk ! Don't hesitate to ask if you run into any difficulties.

nicolasbock avatar Sep 18 '21 13:09 nicolasbock

@linneamk Are you still working on this one? Or did these changes make it into another PR?

jeanlucf22 avatar Oct 13 '23 17:10 jeanlucf22