CI ccache rework #2857
Merged
CI ccache rework #2857
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
No need to run IDF installer at all which opens up more options for pruning crud
Setting is provided for CI use which is handy. Even though the submodules aren't used, IDF still initialises them by default.
Default behaviour is closer to ideal Cache doesn't need to be updated after **every** run. Cache naming is cleaner We're already dealing with installation so ccache action actually makes life more complicated.
- Simplify - Use defined environment paths, very confusing otherwise - Avoid trying to clean inside .git - fails anyway, but not the intent - Found some more stuff to get rid of - Further testing and fixes for Windows
Allows builds to be started from web page
Merged
slaff
pushed a commit
that referenced
this pull request
Jul 11, 2024
Broken by #2857. Also enable ccache. - `CI_BUILD_DIR` does not point to Sming build: must use SMING_HOME. - Don't set CI_BUILD_DIR or SMING_HOME in `setenv.ps1` - it's wrong and already set in `.yml`
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This PR makes a few changes to the way ccache is used for CI.
I did consider adding ccache detection to
export.shso it gets enabled by default.Whilst ccache is great and seems quite robust, if things do go wrong it just adds more complication for users to sort out so requiring manual activation seems best I think.
Simplify CI ccache use
Hendrik Muhs
ccache-actiongot things working easily, but it does installation and some other stuff which we don't need and creating versioned caches on every run is un-necessary.Instead, just use the standard
cacheaction as used for IDF tools, so pull requests use thedevelopbranch cache if available or creates one. This keeps things clean and simple.However, once created caches aren't update automatically, so I've added a workflows to rebuilding the tools which is required if the IDF toolchains get updated.
I've also added a second 'cache clean` workflow which removes any cache entries for pull requests (specifically, not develop).
Tidy action scripts
ifclause doesn't require${{ }}Esp32 tools installation
Installer is still re-installing the tools (gdb, etc.) we've previously pruned out.
So just skip esp32 tools installation for CI if directory already present.
IDF also insists on pulling unwanted submodules back in again, even though they're not used.
Fortunately there is
IDF_SKIP_CHECK_SUBMODULES=1to disable this behaviour.Also some improvements to the
clean-tools.pyscript. Works much better. Have added some notes to theTools/ci/README.rst.This is also a useful step in identifying and eliminating stuff we don't need for Sming.
One thing I have in mind is getting rid of the IDF build step completely and just building the required code separately.
Unfortunately the whole SDK is excessively complex and have found the Rp2040 much easier to work with in practice.
If I ever find a serious application for the extra hardware in an esp32 I might revisit this...