[qt/clang/llvm-project]: Summary of bulk changes made
KDE Git Services - Bulk Change <[email protected]>
| Newsgroups | gmane.comp.kde.cvs |
|---|---|
| Message-ID | <[email protected]> |
Git repository change summary for qt/clang/llvm-project
Pushed by mirror-service into branch 'upstream/users/michalpaszkowski/pisa_backend_upstreaming_6'.
Changed from 0000000000000000000000000000000000000000 to bd8a2d16f618a3daa9287c79458ccbc541151e19
Acknowledgement was received that this change introduces only existing code that has been pushed to another public open source repository.
This change contains the following new commits:
Git commit bd8a2d16f618a3daa9287c79458ccbc541151e19 by Michal Paszkowski on 06/08/2026 at 03:02..
Add PISA codegen pipeline and lowering passes
Add the IR- and MIR-level passes that make up the PISA code generation
pipeline and wire them into PISAPassConfig in PISATargetMachine:
- Intrinsic handling: PISAEmitIntrinsics, PISAExpandIntrinsics,
PISAReplaceIntrinsics
- Call/ABI lowering: PISALegalizeCalls, PISAKernelByValArgsLowering
- IR preparation: PISAPropagateNullPointers, PISALayout, PISAConstProp
- Legalization/verification: PISALegalizePredicates, PISAVerifyTypes,
PISAVerifier
- Post-selection: PISAScopeSelector, PISACacheHintSelector,
PISALegalizeSubregAccess, PISAOptimizeSubregAccess,
PISAOptimizeRedundantCopies, PISAMarkConvergentNoMerge,
PISAInsertLifetimeStart
PISAPassConfig now overrides the GlobalISel and pre/post-regalloc hooks
to run these passes, disables the machine passes that assume physical
registers, and runs the load/store vectorizer and atomic expansion in
the IR stage.
AsmPrinter is added in a follow-up change, so this commit is tested with
MIR-level -run-pass/-stop-after tests.
https://invent.kde.org/qt/clang/llvm-project/-/commit/bd8a2d16f618a3daa9287c79458ccbc541151e19