forked from OSchip/llvm-project
Fix MSVC warning about extending a uint32_t shift result to uint64_t. NFCI.
llvm-svn: 366808
This commit is contained in:
parent
5d4bb8628c
commit
c60c12fb10
|
@ -4502,9 +4502,9 @@ void AMDGPUTargetLowering::computeKnownBitsForTargetNode(
|
|||
Known.One |= ((LHSKnown.One.getZExtValue() >> SelBits) & 0xff) << I;
|
||||
Known.Zero |= ((LHSKnown.Zero.getZExtValue() >> SelBits) & 0xff) << I;
|
||||
} else if (SelBits == 0x0c) {
|
||||
Known.Zero |= 0xff << I;
|
||||
Known.Zero |= 0xFFull << I;
|
||||
} else if (SelBits > 0x0c) {
|
||||
Known.One |= 0xff << I;
|
||||
Known.One |= 0xFFull << I;
|
||||
}
|
||||
Sel >>= 8;
|
||||
}
|
||||
|
|
Loading…
Reference in New Issue