thevar1able/ClickHouse
1
0
Fork 0
mirror of https://github.com/ClickHouse/ClickHouse.git synced 2024-11-25 09:02:00 +00:00
Code Issues Packages Projects Releases Wiki Activity

More conversions

Browse Source
This commit is contained in:
Robert Schulze 2022-07-31 18:25:50 +00:00
parent a7734672b9
commit d39259a4c0
No known key found for this signature in database
GPG Key ID: 26703B55FB13728A
2 changed files with 9 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
src/Common/HashTable/StringHashTable.h
View File

@ -3,6 +3,7 @@
#include <Common/HashTable/HashMap.h>
#include <Common/HashTable/HashTable.h>
#include <bit>
#include <new>
#include <variant>
@ -21,17 +22,17 @@ struct StringKey24
inline StringRef ALWAYS_INLINE toStringRef(const StringKey8 & n)
{
assert(n != 0);
return {reinterpret_cast<const char *>(&n), 8ul - (__builtin_clzll(n) >> 3)};
return {reinterpret_cast<const char *>(&n), 8ul - (std::countl_zero(n) >> 3)};
}
inline StringRef ALWAYS_INLINE toStringRef(const StringKey16 & n)
{
assert(n.items[1] != 0);
return {reinterpret_cast<const char *>(&n), 16ul - (__builtin_clzll(n.items[1]) >> 3)};
return {reinterpret_cast<const char *>(&n), 16ul - (std::countl_zero(n.items[1]) >> 3)};
}
inline StringRef ALWAYS_INLINE toStringRef(const StringKey24 & n)
{
assert(n.c != 0);
return {reinterpret_cast<const char *>(&n), 24ul - (__builtin_clzll(n.c) >> 3)};
return {reinterpret_cast<const char *>(&n), 24ul - (std::countl_zero(n.c) >> 3)};
}
struct StringHashTableHash

9
src/Interpreters/ITokenExtractor.cpp
View File

@ -4,6 +4,7 @@
#include <Common/StringUtils/StringUtils.h>
#include <Common/UTF8Helpers.h>
#include <bit>
#if defined(__SSE2__)
#include <emmintrin.h>
@ -122,7 +123,7 @@ bool SplitTokenExtractor::nextInStringPadded(const char * data, size_t length, s
const auto alnum_chars_ranges = _mm_set_epi8(0, 0, 0, 0, 0, 0, 0, 0,
'\xFF', '\x80', 'z', 'a', 'Z', 'A', '9', '0');
// Every bit represents if `haystack` character is in the ranges (1) or not (0)
const int result_bitmask = _mm_cvtsi128_si32(_mm_cmpestrm(alnum_chars_ranges, 8, haystack, haystack_length, _SIDD_CMP_RANGES));
const unsigned result_bitmask = _mm_cvtsi128_si32(_mm_cmpestrm(alnum_chars_ranges, 8, haystack, haystack_length, _SIDD_CMP_RANGES));
#else
// NOTE: -1 and +1 required since SSE2 has no `>=` and `<=` instructions on packed 8-bit integers (epi8).
const auto number_begin = _mm_set1_epi8('0' - 1);
@ -136,7 +137,7 @@ bool SplitTokenExtractor::nextInStringPadded(const char * data, size_t length, s
// every bit represents if `haystack` character `c` satisfies condition:
// (c < 0) || (c > '0' - 1 && c < '9' + 1) || (c > 'a' - 1 && c < 'z' + 1) || (c > 'A' - 1 && c < 'Z' + 1)
// < 0 since _mm_cmplt_epi8 threats chars as SIGNED, and so all chars > 0x80 are negative.
const int result_bitmask = _mm_movemask_epi8(_mm_or_si128(_mm_or_si128(_mm_or_si128(
const unsigned result_bitmask = _mm_movemask_epi8(_mm_or_si128(_mm_or_si128(_mm_or_si128(
_mm_cmplt_epi8(haystack, zero),
_mm_and_si128(_mm_cmpgt_epi8(haystack, number_begin), _mm_cmplt_epi8(haystack, number_end))),
_mm_and_si128(_mm_cmpgt_epi8(haystack, alpha_lower_begin), _mm_cmplt_epi8(haystack, alpha_lower_end))),
@ -152,7 +153,7 @@ bool SplitTokenExtractor::nextInStringPadded(const char * data, size_t length, s
continue;
}
const auto token_start_pos_in_current_haystack = getTrailingZeroBitsUnsafe(result_bitmask);
const auto token_start_pos_in_current_haystack = std::countr_zero(result_bitmask);
if (*token_length == 0)
// new token
*token_start = *pos + token_start_pos_in_current_haystack;
@ -160,7 +161,7 @@ bool SplitTokenExtractor::nextInStringPadded(const char * data, size_t length, s
// end of token starting in one of previous haystacks
return true;
const auto token_bytes_in_current_haystack = getTrailingZeroBitsUnsafe(~(result_bitmask >> token_start_pos_in_current_haystack));
const auto token_bytes_in_current_haystack = std::countr_zero(~(result_bitmask >> token_start_pos_in_current_haystack));
*token_length += token_bytes_in_current_haystack;
*pos += token_start_pos_in_current_haystack + token_bytes_in_current_haystack;