[ClangIR][Lowering][NFC] Split LowerToLLVM.h from LowerToLLVM.cpp #1102
Conversation
|
I think this makes sense overall: regardless of how we setup pipelines, the functionality should be available in a more composable way, which this PR allows for. This is purely NFC, right? I wonder if as part of this we should also rename the classes to be more descriptive with what they implement (because they might as well later be intermixed), for example, should we follow a pattern like |
bcardosolopes
changed the title
|
I'd be in favor of the more explicit names, especially if we'll potentially be mixing lowering paths in the future. |
ChuanqiXu9
force-pushed
the
SplitLowerToLLVM
branch
from
f049f59 to
fcdd282
Compare
|
Done renaming.
Absolutely if I didn't make some oversight during the rebase. But even if there is any oversight, I believe it can be caught easily by the tests. |
|
✅ With the latest revision this PR passed the C/C++ code formatter. |
ChuanqiXu9
force-pushed
the
SplitLowerToLLVM
branch
from
fcdd282 to
79f7326
Compare
ChuanqiXu9
force-pushed
the
SplitLowerToLLVM
branch
from
79f7326 to
de10d03
Compare
Done! |
) This is a NFC patch that moves declaration from LowerToLLVM.cpp. The motivation of the patch is, we hope we can use the abilities from MLIR's standard dialects without lowering **ALL** clangir operation to MLIR's standard dialects. For example, currently we have 86 operations in LowerToLLVM.cpp but only 45 operations under though MLIR. It won't be easy to add proper lowering for all operation to **different** dialects. I think the solution may be to allow **mixed** IR. So that we can lowering CIR to MLIR's standard dialects partially and we can use some existing analysis and optimizations in MLIR and then we can lower all of them (the MLIR dialects and unlowered clangir) to LLVM IR. The hybrid IR is one of the goals of MLIR as far as I know. NOTE: I completely understand that the DirectlyLLVM pipeline is the tier-1 pipeline that we want to support. The idea above won't change this. I just want to offer some oppotunities for the downstream projects and finally some chances to improve the overall ecosystem.
This is a NFC patch that moves declaration from LowerToLLVM.cpp.
The motivation of the patch is, we hope we can use the abilities from MLIR's standard dialects without lowering ALL clangir operation to MLIR's standard dialects. For example, currently we have 86 operations in LowerToLLVM.cpp but only 45 operations under though MLIR. It won't be easy to add proper lowering for all operation to different dialects.
I think the solution may be to allow mixed IR. So that we can lowering CIR to MLIR's standard dialects partially and we can use some existing analysis and optimizations in MLIR and then we can lower all of them (the MLIR dialects and unlowered clangir) to LLVM IR. The hybrid IR is one of the goals of MLIR as far as I know.
NOTE: I completely understand that the DirectlyLLVM pipeline is the tier-1 pipeline that we want to support. The idea above won't change this. I just want to offer some oppotunities for the downstream projects and finally some chances to improve the overall ecosystem.