Struct memchr::arch::x86_64::sse2::memchr::One

Note that it is a common misconception that if one compiles for an x86_64 target, then they therefore automatically have access to SSE2 instructions. While this is almost always the case, it isn’t true in 100% of cases.

source

pub fn is_available() -> bool

Returns true when this implementation is available in the current environment.

When this is true, it is guaranteed that One::new will return a Some value. Similarly, when it is false, it is guaranteed that One::new will return a None value.

Note also that for the lifetime of a single program, if this returns true then it will always return true.

source

pub fn find(&self, haystack: &[u8]) -> Option<usize>

Return the first occurrence of one of the needle bytes in the given haystack. If no such occurrence exists, then None is returned.

The occurrence is reported as an offset into haystack. Its maximum value is haystack.len() - 1.

source

pub fn rfind(&self, haystack: &[u8]) -> Option<usize>

Return the last occurrence of one of the needle bytes in the given haystack. If no such occurrence exists, then None is returned.

The occurrence is reported as an offset into haystack. Its maximum value is haystack.len() - 1.

source

pub fn count(&self, haystack: &[u8]) -> usize

Counts all occurrences of this byte in the given haystack.

source

pub unsafe fn find_raw( &self, start: const u8, end: const u8, ) -> Option<*const u8>

Like find, but accepts and returns raw pointers.

When a match is found, the pointer returned is guaranteed to be >= start and < end.

This routine is useful if you’re already using raw pointers and would like to avoid converting back to a slice before executing a search.

§Safety

Both start and end must be valid for reads.
Both start and end must point to an initialized value.
Both start and end must point to the same allocated object and must either be in bounds or at most one byte past the end of the allocated object.
Both start and end must be derived from a pointer to the same object.
The distance between start and end must not overflow isize.
The distance being in bounds must not rely on “wrapping around” the address space.

Note that callers may pass a pair of pointers such that start >= end. In that case, None will always be returned.

source

pub unsafe fn rfind_raw( &self, start: const u8, end: const u8, ) -> Option<*const u8>

Like rfind, but accepts and returns raw pointers.

When a match is found, the pointer returned is guaranteed to be >= start and < end.

This routine is useful if you’re already using raw pointers and would like to avoid converting back to a slice before executing a search.

§Safety

Both start and end must be valid for reads.
Both start and end must point to an initialized value.
Both start and end must point to the same allocated object and must either be in bounds or at most one byte past the end of the allocated object.
Both start and end must be derived from a pointer to the same object.
The distance between start and end must not overflow isize.
The distance being in bounds must not rely on “wrapping around” the address space.

Note that callers may pass a pair of pointers such that start >= end. In that case, None will always be returned.

source

pub unsafe fn count_raw(&self, start: const u8, end: const u8) -> usize

Counts all occurrences of this byte in the given haystack represented by raw pointers.

This routine is useful if you’re already using raw pointers and would like to avoid converting back to a slice before executing a search.

§Safety

Both start and end must be valid for reads.
Both start and end must point to an initialized value.
Both start and end must point to the same allocated object and must either be in bounds or at most one byte past the end of the allocated object.
Both start and end must be derived from a pointer to the same object.
The distance between start and end must not overflow isize.
The distance being in bounds must not rely on “wrapping around” the address space.

Note that callers may pass a pair of pointers such that start >= end. In that case, 0 will always be returned.

source

unsafe fn find_raw_impl( &self, start: const u8, end: const u8, ) -> Option<*const u8>

Available with target feature sse2 only.

Execute a search using SSE2 vectors and routines.

§Safety

Same as One::find_raw, except the distance between start and end must be at least the size of an SSE2 vector (in bytes).

(The target feature safety obligation is automatically fulfilled by virtue of being a method on One, which can only be constructed when it is safe to call sse2 routines.)

source

unsafe fn rfind_raw_impl( &self, start: const u8, end: const u8, ) -> Option<*const u8>

Available with target feature sse2 only.

Execute a search using SSE2 vectors and routines.

§Safety

Same as One::rfind_raw, except the distance between start and end must be at least the size of an SSE2 vector (in bytes).

(The target feature safety obligation is automatically fulfilled by virtue of being a method on One, which can only be constructed when it is safe to call sse2 routines.)

source

unsafe fn count_raw_impl(&self, start: const u8, end: const u8) -> usize

Available with target feature sse2 only.

Execute a count using SSE2 vectors and routines.

§Safety

Same as One::count_raw, except the distance between start and end must be at least the size of an SSE2 vector (in bytes).

(The target feature safety obligation is automatically fulfilled by virtue of being a method on One, which can only be constructed when it is safe to call sse2 routines.)

source