Modern libc's have an SSE version of memchr which is a lot faster than our
hand-rolled version. In the past I was reluctant to use it because Darwin's
memchr used a naive ridiculously slow implementation, but that has been fixed
some versions ago.
Should have zero functional impact.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@232898
91177308-0d34-0410-b5e6-
96231b3b80d8
/// \returns The index of the first occurrence of \p C, or npos if not
/// found.
size_t find(char C, size_t From = 0) const {
- for (size_t i = std::min(From, Length), e = Length; i != e; ++i)
- if (Data[i] == C)
- return i;
+ size_t FindBegin = std::min(From, Length);
+ if (FindBegin < Length) { // Avoid calling memchr with nullptr.
+ // Just forward to memchr, which is faster than a hand-rolled loop.
+ if (const void *P = ::memchr(Data + FindBegin, C, Length - FindBegin))
+ return static_cast<const char *>(P) - Data;
+ }
return npos;
}