RTEMS / Pre-Qualification / Central¶
Go to Issues or Merge Requests
Merge Requests Summary
Issues¶
2 - Update rtems_waf module location.¶
Id |
2 |
State |
closed |
Type |
ISSUE |
Author |
Amar Takhar |
Assignee(s) |
Sebastian Huber |
Closed by |
Kinsey Moore |
Created |
2024-06-11T21:35:38.662Z |
Closed |
2024-06-12T15:31:27.639Z |
Updated |
2024-06-12T15:31:27.656Z |
Milestone |
6.1 |
Link |
https://gitlab.rtems.org/rtems/prequal/rtems-central/-/issues/2 |
Merges |
0 |
Summary¶
The rtems_waf submodule still points to git.rtems.org
Steps to reproduce¶
Author: Amar Takhar
2024-06-11T21:35:38.732Z
assigned to @sebhub
Author: Kinsey Moore
2024-06-12T15:31:27.383Z
rtems-central has no submodule reference to rtems_waf.
1 - Sync commits from master to main¶
Id |
1 |
State |
closed |
Type |
ISSUE |
Author |
Gedare Bloom |
Assignee(s) |
Gedare Bloom |
Closed by |
Gedare Bloom |
Created |
2024-05-02T04:34:17.394Z |
Closed |
2024-05-02T04:35:44.739Z |
Updated |
2024-05-02T04:35:44.753Z |
Milestone |
6.1 |
Link |
https://gitlab.rtems.org/rtems/prequal/rtems-central/-/issues/1 |
Merges |
0 |
Summary¶
Some commits were lost from master during the transition to gitlab and main.
Steps to reproduce¶
Author: Gedare Bloom
2024-05-02T04:34:17.488Z
assigned to @gedare
Author: Gedare Bloom
2024-05-02T04:35:44.471Z
Nevermind, I was mistaken, these are all sync’d.
Merge Requests¶
2 - Draft: Various updates (opened)¶
Id |
2 |
State |
opened |
Assignee(s) |
Sebastian Huber |
Created |
2024-06-26T06:46:51.613Z |
Updated |
2024-07-18T07:52:04.338Z |
Milestone |
6.1 |
Labels |
qualification |
Link |
https://gitlab.rtems.org/rtems/prequal/rtems-central/-/merge_requests/2 |
Commit |
Summary¶
Integrate recent updates in rtems-central.
Author: Sebastian Huber
2024-06-26T06:46:51.764Z
assigned to @sebhub
Author: Chris Johns
2024-06-26T06:51:08.676Z
https://gitlab.rtems.org/rtems/prequal/rtems-central/-/commit/46ced75f93e459da012f655960d865ce0a8836d9?merge_request_iid=2 The message should be:
Make sure the invalidated data cache lines are contained in the buffer.
Author: Sebastian Huber
2024-06-26T06:53:29.638Z
added 3 commits
85955f01 - spec: Fix cache invalidation test case
3e7025f4 - modules: Update rtems
9961311f - modules: Update rtems-docs
Author: Sebastian Huber
2024-06-26T06:54:54.642Z
changed the description
Author: Chris Johns
2024-06-26T06:55:34.229Z
This is fragile as we do not support external references to moving branches like this. In release or branch this link breaks.
Author: Chris Johns
2024-06-26T06:56:16.671Z
I am sorry this again is something I do not like to see.
Author: Sebastian Huber
2024-06-26T06:57:32.746Z
Yes, this is fragile, but all I can offer right now.
Author: Chris Johns
2024-06-26T07:03:55.021Z
We need a solution. Should an issue be raised with “need funding” or whatever we now have to deal with this? The issue should at least note this repo depends on external linking.
I appreciate this is complex and involved and not a reason to block the merge but we need to be careful as having these links is a liability to the project and potentially confusing to users as generated content with the links ages.
Author: Chris Johns
2024-06-26T07:15:43.105Z
Should:
If the ${../if/disable-data:/name} directive is called, then the system shall
be:
If the ${../if/disable-data:/name} directive is called, the system shall
Author: Sebastian Huber
2024-06-26T07:15:43.105Z
See the EARS syntax (unwanted behaviour):
https://docs.rtems.org/branches/master/eng/req/req-for-req.html#syntax
Author: Chris Johns
2024-06-26T07:15:43.105Z
Does this end up in the projects documentation?
Author: Sebastian Huber
2024-06-26T07:19:54.679Z
It is a requirement, so currently it will not show up in the documentation.
Author: Chris Johns
2024-06-26T07:15:43.105Z
Thanks. It is not easy to tell this.
Author: Sebastian Huber
2024-06-26T07:09:37.258Z
It should be clear from the context. The change is already in rtems-docs.
Author: Chris Johns
2024-06-26T07:14:48.744Z
Must have missed it, sorry about that.
Author: Chris Johns
2024-06-26T06:59:56.025Z
Why reference your personal repo?
Author: Sebastian Huber
2024-06-26T07:11:46.550Z
There are still not all patches integrated in the main branch and this will likely never change.
Author: Chris Johns
2024-06-26T07:14:08.984Z
Does this mean pre-qual release are based on your personal repo?
Author: Sebastian Huber
2024-06-26T07:16:22.487Z
What do you mean with a pre-qual release? The stuff published by ESA (https://rtems-qual.io.esa.int/) uses ESA repositories.
Author: Chris Johns
2024-06-26T07:21:03.953Z
Just that. Did that data come from rtems-central? And if so did it reference your personal repo?
Author: Sebastian Huber
2024-06-26T07:23:26.846Z
A repository is just a container for branches. The ESA repositories use branches from rtems-central and my personal repository as baselines.
Author: Chris Johns
2024-06-26T07:25:21.911Z
That be acceptable to ESA but here your repo is just that. It is a personal space that has no other value. This promotes it to something else. If this was in your personal repo of rtems-central it would be fine.
Author: Sebastian Huber
2024-06-26T07:28:29.303Z
You can also add an RTEMS repository to https://gitlab.rtems.org/rtems/prequal. For each update, a new branch is created. We are at
qual-101right now.
Author: Chris Johns
2024-06-26T23:00:39.950Z
I am not across the specifics of how this repo is run and I am attempting to avoid being drawn into that conversation.
My observation is based on how we run the repositories in this gitlab instance. The model we are using is top level repos cannnot be directly pushed to. All merges need a MR, be reviewed and approved by the rules we have established. If we allow personal repos to be linked from top level repos this process is subverted. I am not convinced we need an exception here.
The simple reality is you need approvers to get things merged and given the “normal” guide lines I use this does not fit. As I said I am not across the detail and becoming so to start approving is not feasible and your efforts here would stall. I think that would be a bad outcome to have.
Author: Sebastian Huber
2024-06-27T05:26:59.294Z
The current work flow is like this: 1. Something in rtems-central is changed locally. 2. The files in
modules/rtemsandmodules/rtems-docsare regenerated. 3. In the modules, patch sets are created and merge requests are issued. 4. Changes are requested, go to 1, otherwise go to 5. 5. The module branches are rebased to main branches (which now include the generated changes from 1). Quite often you have to fix conflicts during the rebase. After rebase, new branches are pushed. 6. Patch sets for rtems-central are created which include submodule updates. 7. A merge request for rtems-central is issued.In theory, you can review the changes in the submodules in the commits updating the submodules. However, since I have to carry about 28 commits around, you would have to spend a lot of time to do this for each merge request.
Author: Sebastian Huber
2024-06-26T10:00:17.457Z
added 4 commits
572576bb - spec: Add cannot disable data cache fatal error
8cc73358 - spec: Fix cache invalidation test case
98150e46 - modules: Update rtems
6ec43b37 - modules: Update rtems-docs
Author: Sebastian Huber
2024-06-26T10:02:44.817Z
marked this merge request as draft
Author: Sebastian Huber
2024-07-02T07:36:11.683Z
added 2 commits
ea7b83f4 - modules: Update rtems
98a53719 - modules: Update rtems-docs
Author: Sebastian Huber
2024-07-02T07:36:43.744Z
marked this merge request as ready
Author: Sebastian Huber
2024-07-08T05:30:28.640Z
added 3 commits
95b32ea4 - spec: Fix bad thread dispatch validation
268747b8 - modules: Update rtems-docs
6d6bb738 - modules: Update rtems
Author: Gedare Bloom
2024-07-10T06:08:33.537Z
this breaks if using a personal fork.
Author: Gedare Bloom
2024-07-10T06:08:33.537Z
also, it would be helpful to split this MR into these submodule fixes, versus other stuff. atm I can’t use the current version of rtems-central.
Author: Gedare Bloom
2024-07-10T06:08:33.537Z
Actually it may not be about the personal fork, but I’m not sure why this isn’t working for me. I get
fatal: git upload-pack: not our ref 7b3db9d6c1a7ecfa2c9ff9fe1f939081de17e743 fatal: remote error: upload-pack: not our ref 7b3db9d6c1a7ecfa2c9ff9fe1f939081de17e743 fatal: Fetched in submodule path 'modules/rtems-docs', but it did not contain 7b3db9d6c1a7ecfa2c9ff9fe1f939081de17e743. Direct fetching of that commit failed. fatal: `gedare@gedare-
Author: Gedare Bloom
2024-07-10T06:08:33.537Z
it turns out that it was trying to get my personal fork. by updating my fork’s main, it worked:
From ssh://gitlab.rtems.org:2222/gedare/rtems-docs d196c8b..af7030d main -> origin/main Submodule path 'modules/rtems-docs': checked out '7b3db9d6c1a7ecfa2c9ff9fe1f939081de17e743'I guess there is something else that must be done besides
git submodule deinitin order to clean up old submodules.
Author: Sebastian Huber
2024-07-10T06:08:40.665Z
You can use
git submodule syncto fetch the URLs from.gitmodules.
Author: Sebastian Huber
2024-07-10T06:08:33.537Z
The problem with the personal forks is due to the new repository structure. It breaks also the setup of external users. This is nothing I can fix.
Author: Sebastian Huber
2024-07-11T05:34:18.528Z
added 15 commits
e60af25b - 1 commit from branch
rtems/prequal:maine60af25b…fc10a0d2 - 4 earlier commits
3b3de38a - spec: Add get/set interrupt priority
f90ec554 - spec: Improve interrupt controller tests
873252df - spec: Indicate if lock objects are needed
8a54726c - spec: Add system termination recommendation
bbf8b26f - spec: Fix URL
253a6193 - spec: Add cannot disable data cache fatal error
223759aa - spec: Fix cache invalidation test case
e549144b - spec: Fix bad thread dispatch validation
e9f95f63 - modules: Update rtems-docs
e1750f40 - modules: Update rtems
Author: Sebastian Huber
2024-07-12T06:38:44.691Z
added 3 commits
496c933f - spec: Add optional extra-files to build items
238c546a - modules: Update rtems-docs
5e33a933 - modules: Update rtems
Author: Sebastian Huber
2024-07-17T17:46:33.548Z
marked this merge request as draft
Author: Sebastian Huber
2024-07-18T07:37:26.740Z
added 4 commits
e67d8178 - spec: Specify BSP-specific fatal extension
67c86a7c - spec: Simplify system termination procedure
3b2127d8 - modules: Update rtems-docs
280327de - modules: Update rtems
Author: Sebastian Huber
2024-07-18T07:52:10.114Z
added 1 commit
cde817bf - modules: Update rtems