diff options
| author | Craig Topper <craig.topper@sifive.com> | 2024-05-29 11:47:21 -0700 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2024-05-29 11:47:21 -0700 |
| commit | 8aceb7a53d82f9566a7cf4775effa4089b22a75b (patch) | |
| tree | a2e1d821c7737ef4e95ce6f04cd0bbd50ced608f /llvm/lib/CodeGen/TargetLoweringBase.cpp | |
| parent | 472184db24f2f5f7ddca76d7aadd13a3dca05991 (diff) | |
| download | llvm-8aceb7a53d82f9566a7cf4775effa4089b22a75b.zip llvm-8aceb7a53d82f9566a7cf4775effa4089b22a75b.tar.gz llvm-8aceb7a53d82f9566a7cf4775effa4089b22a75b.tar.bz2 | |
[ValueTypes] Remove MVT::MAX_ALLOWED_VALUETYPE. NFC (#93654)
Despite the comment, this isn't used to size bit vectors or tables.
That's done by VALUETYPE_SIZE. MAX_ALLOWED_VALUETYPE is only used by
some static_asserts that compare it to VALUETYPE_SIZE.
This patch removes it and most of the static_asserts. I left one where I
compared VALUETYPE_SIZE to token which is the first type that isn't part
of the VALUETYPE range. This isn't strictly needed, we'd probably catch
duplication error from VTEmitter.cpp first.
Diffstat (limited to 'llvm/lib/CodeGen/TargetLoweringBase.cpp')
| -rw-r--r-- | llvm/lib/CodeGen/TargetLoweringBase.cpp | 3 |
1 files changed, 0 insertions, 3 deletions
diff --git a/llvm/lib/CodeGen/TargetLoweringBase.cpp b/llvm/lib/CodeGen/TargetLoweringBase.cpp index 82a5991..f2e4632 100644 --- a/llvm/lib/CodeGen/TargetLoweringBase.cpp +++ b/llvm/lib/CodeGen/TargetLoweringBase.cpp @@ -1430,9 +1430,6 @@ TargetLoweringBase::findRepresentativeClass(const TargetRegisterInfo *TRI, /// this allows us to compute derived properties we expose. void TargetLoweringBase::computeRegisterProperties( const TargetRegisterInfo *TRI) { - static_assert(MVT::VALUETYPE_SIZE <= MVT::MAX_ALLOWED_VALUETYPE, - "Too many value types for ValueTypeActions to hold!"); - // Everything defaults to needing one register. for (unsigned i = 0; i != MVT::VALUETYPE_SIZE; ++i) { NumRegistersForVT[i] = 1; |