diff options
| author | Daniel Dunbar <daniel@zuster.org> | 2010-09-23 01:54:28 +0000 |
|---|---|---|
| committer | Daniel Dunbar <daniel@zuster.org> | 2010-09-23 01:54:28 +0000 |
| commit | 8aa87c71d9bfec14e135c683b0d7b9de999dbcb0 (patch) | |
| tree | 770576d0ca9fb91432b7c463921b7277dfb8adca /lib/CodeGen | |
| parent | 3df2f31c4e0eec3af3a010bc1b24963d3437ee42 (diff) | |
IRgen/ABI/ARM: Trust the backend to pass vectors correctly for the given ABI.
- Therefore, we can lower out the NEON wrapper structs and pass the vectors
directly. This makes a huge difference in the cleanliness of the IR after
optimization.
- I will trust, but verify, via future ABITest testing (for APCS-GNU, at
least).
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@114618 91177308-0d34-0410-b5e6-96231b3b80d8
Diffstat (limited to 'lib/CodeGen')
| -rw-r--r-- | lib/CodeGen/TargetInfo.cpp | 11 |
1 files changed, 11 insertions, 0 deletions
diff --git a/lib/CodeGen/TargetInfo.cpp b/lib/CodeGen/TargetInfo.cpp index 34a6d37726..f1da3c3903 100644 --- a/lib/CodeGen/TargetInfo.cpp +++ b/lib/CodeGen/TargetInfo.cpp @@ -2272,6 +2272,17 @@ ABIArgInfo ARMABIInfo::classifyArgumentType(QualType Ty) const { if (isRecordWithNonTrivialDestructorOrCopyConstructor(Ty)) return ABIArgInfo::getIndirect(0, /*ByVal=*/false); + // NEON vectors are implemented as (theoretically) opaque structures wrapping + // the underlying vector type. We trust the backend to pass the underlying + // vectors appropriately, so we can unwrap the structs which generally will + // lead to much cleaner IR. + if (const Type *SeltTy = isSingleElementStruct(Ty, getContext())) { + if (SeltTy->isVectorType()) + return ABIArgInfo::getDirect(CGT.ConvertType(QualType(SeltTy, 0))); + } + + // Otherwise, pass by coercing to a structure of the appropriate size. + // // FIXME: This is kind of nasty... but there isn't much choice because the ARM // backend doesn't support byval. // FIXME: This doesn't handle alignment > 64 bits. |