SWAR: test for special bytes in the reduced form

The usual test for a zero byte, (x - ones) & ~x & high, needs `& ~x` only
to keep bytes from 0x80 out. The tests for quotes, backslashes and control
characters in string_scan.hpp and the serializer also flag every byte from
0x80 with `v & high`, so the `& ~x` terms can be dropped. Clang finds this
reduction itself; GCC does not and emits 24 instead of 16 instructions.

The masks are the same for 1.3 billion random words and for every pair of
adjacent byte values. With GCC 15 on x86-64: accept() -2 to -4 %,
sax_parse() -2 %, dump() -3 %. Clang is unchanged.

Signed-off-by: Niels Lohmann <mail@nlohmann.me>
This commit is contained in:
Niels Lohmann authored and GitHub committed 2026-11-10 07:08:48 +01:00
1 parent 33f15727ce
commit 3836e2e5c6
3 files changed
+16 -28

No files matched your search

@@ -724,10 +724,8 @@ class serializer
constexpr std::uint64_t ones = 0x0101010101010101ull;
const std::uint64_t q = v ^ 0x2222222222222222ull; // '"'
const std::uint64_t b = v ^ 0x5C5C5C5C5C5C5C5Cull; // '\\'
return (((q - ones) & ~q) | ((b - ones) & ~b) // == '"', == '\\'
| ((v - 0x2020202020202020ull) & ~v) // < 0x20
| v | (v + ones)) // >= 0x7F
& 0x8080808080808080ull;
// (see swar_string_special(): the bytes from 0x80 are flagged anyway)
return ((q - ones) | (b - ones) | (v - 0x2020202020202020ull) | v | (v + ones)) & 0x8080808080808080ull;
}
template<typename WordType>