- remove the __sha3 function symbol: this is provided by the pallet
- remove the storage address spaces: they are not mapped into memory
---------
Signed-off-by: Cyrill Leutwiler <bigcyrill@hotmail.com>
Co-authored-by: Alexander Theißen <alex.theissen@me.com>
This PR changes the CI build scripts to also build LLVM for windows.
**It doesn't build `revive` itself for windows**. This will come in a
follow up. But once we have a LLVM binary release the turn around time
will be much quicker for experimenting with the revive windows build.
I manually uploaded the release those changes produce
[here](https://github.com/paritytech/revive-alex-workflowtest/releases/tag/llvm-18.1.8-revive.22f3ceb).
This enables this PR's CI to find the proper release. This is necessary
because I am also making changes to the folder structure and artifact
naming that the other CI jobs are depending on.
Releases generated from this branch can be inspected here:
https://github.com/paritytech/revive-alex-workflowtest/releases/tag/v0.1.0-dev.12
Summary of changes:
- Change `llvm-builder` to use MSVC toolchain on windows
- Fix `llvm-builder` to work with `.exe` files
- Unify the llvm release jobs into a single one. This removed a lot of
copy pasted code and also speeds up the build by giving each their own
runner.
- Use the LLVM target triple to name the binary releases instead of an
ad-hoc naming convention
- Remove the nested folder hierarchy inside the llvm release. Its just
now a single folder `llvm-<target>` that contains the toolchain.
- Give jobs and workflows consistent names
- Replace all runners bei their `*-latest` counterpart
- Only use `parity-large` to build llvm now. All other jobs use github
runners
- Add the revive runtime function interface to minimize boiler plate
code.
- Outline heavily repeated code into dedicated functions to bring down
code size.
- The code size tests builds optimized for size.
- Function attributes are passed as slices.
This significantly brings down the code size for all OpenZeppelin wizard
contracts (using all possible features) compiled against OpenZeppelin
`v5.0.0` with size optimizations.
|contract|| `-Oz` main | `-Oz` PR || `-O3` main | `-O3` PR |
|-|-|-|-|-|-|-|
|erc1155.sol||100K|67K||114K|147K|
|erc20.sol||120K|90K||160K|191K|
|erc721.sol||128K|101K||178K|214K|
|governor.sol||226K|165K||293K|349K|
|rwa.sol||116K|85K||154K|185K|
|stable.sol||116K|86K||155K|192K|
On the flip side this introduces a heavy penalty for cycle optimized
builds. Setting the no-inline attributes for cycle optimized builds
helps a lot but heavily penalizes runtime speed (LLVM does not yet
inline everything properly - to be investigated later on).
Next steps:
- Modularize more functions
- Refactor the YUL function arguments to use pointers instead of values
- Afterwards check if LLVM still has trouble inline-ing properly on O3
or set the no-inline attribute if it does not penalize runtime
performance too bad.
- llvm artifacts search, download and extract now in `get-llvm` action
- `get-emsdk` action - clone, install, activate emsdk, so we don't need
to build `revive-builder` for this
- switch workflows to new actions
- `concurrency` for remaining workflows (cancel run if new run is
triggered)
- `target_commitish` for main release, fixes release commit
- Run release workflow in PR without creating a release if PR is labeled
with `release-test`
---------
Co-authored-by: cornholio <0@mcornholio.ru>
Co-authored-by: Alexander Theißen <alex.theissen@me.com>
Co-authored-by: xermicus <cyrill@parity.io>
Co-authored-by: xermicus <bigcyrill@hotmail.com>
xermicus