[CIR] Implement CGM.getIntrinsic for LLVMCallIntrinsicOp
#1710
+1,109
−277
Conversation
This file contains hidden or 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
|
✅ With the latest revision this PR passed the C/C++ code formatter. |
RiverDave
force-pushed
the
dave/get-intrinsic
branch
3 times, most recently
from
eac2942 to
bba1ab7
Compare
…ntroduce `IntrinsicDescriptor` struct for better intrinsic management
RiverDave
force-pushed
the
dave/get-intrinsic
branch
from
bba1ab7 to
71f3dc1
Compare
RiverDave
commented
Jun 23, 2025
| return true; | ||
| case llvm::Intrinsic::x86_rdtsc: | ||
| case llvm::Intrinsic::x86_rdtscp: | ||
| case llvm::Intrinsic::ctlz: |
There was a problem hiding this comment.
This was one of my points regarding integer signedness, as much as I think It's hard to track these Id's, It seems to be the only way so far...
|
cc. @bcardosolopes |
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Outdated suggestions cannot be applied.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.
Related: #1692
The current implementation supports non overloaded intrinsics for now...
Couple of nits so far and limitations I've found as compared to CodeGen. Would like to get some guidance based on what I've done so far:
How should we handle intrinsic-specific module operations in CIR/MLIR that LLVM normally provides (naming, uniquing, lookups)? specifically since
LLVM::Moduleimplements:getOrInsertFunction. Given that codegen represents these as functions and we have a dedicated op for intrinsics, how would the global -> intrinsic conversion work?Given the fact that we need to give signedness to integer types in CIR, the facilities designated to hold intrinsic metadata(
IITDescriptor) dont contain any signedness. One approach I've taken is hardcoding and determining signedness based on the Intrinsic ID, however that could potentially be a burden to mantain.There's also some code duplication, specially in sections that handle with
IITDescriptors