mirrors/cpython
1
0
Fork 0
mirror of https://github.com/python/cpython.git synced 2025-08-31 14:07:50 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 9

gh-126024: unicodeobject: optimize find_first_nonascii (GH-127790)

Browse source 
Remove 1 branch.
This commit is contained in:
Inada Naoki 2024-12-14 01:21:46 +09:00 • committed by GitHub
parent 8bc18182a7
commit 5dd775bed0
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
1 changed files with 14 additions and 9 deletions
Download patch file Download diff file Expand all files Collapse all files

23
Objects/unicodeobject.c
View file

@ -5077,21 +5077,24 @@ load_unaligned(const unsigned char *p, size_t size)
static Py_ssize_t static Py_ssize_t
find_first_nonascii(const unsigned char *start, const unsigned char *end) find_first_nonascii(const unsigned char *start, const unsigned char *end)
{ {
// The search is done in `size_t` chunks.
// The start and end might not be aligned at `size_t` boundaries,
// so they're handled specially.
const unsigned char *p = start; const unsigned char *p = start;
if (end - start >= SIZEOF_SIZE_T) { if (end - start >= SIZEOF_SIZE_T) {
const unsigned char *p2 = _Py_ALIGN_UP(p, SIZEOF_SIZE_T); // Avoid unaligned read.
#if PY_LITTLE_ENDIAN && HAVE_CTZ #if PY_LITTLE_ENDIAN && HAVE_CTZ
if (p < p2) { size_t u;
size_t u; memcpy(&u, p, sizeof(size_t));
memcpy(&u, p, sizeof(size_t)); u &= ASCII_CHAR_MASK;
u &= ASCII_CHAR_MASK; if (u) {
if (u) { return (ctz(u) - 7) / 8;
return (ctz(u) - 7) / 8;
}
p = p2;
} }
p = _Py_ALIGN_DOWN(p + SIZEOF_SIZE_T, SIZEOF_SIZE_T);
#else /* PY_LITTLE_ENDIAN && HAVE_CTZ */ #else /* PY_LITTLE_ENDIAN && HAVE_CTZ */
const unsigned char *p2 = _Py_ALIGN_UP(p, SIZEOF_SIZE_T);
while (p < p2) { while (p < p2) {
if (*p & 0x80) { if (*p & 0x80) {
return p - start; return p - start;
@ -5099,6 +5102,7 @@ find_first_nonascii(const unsigned char *start, const unsigned char *end)
p++; p++;
} }
#endif #endif
const unsigned char *e = end - SIZEOF_SIZE_T; const unsigned char *e = end - SIZEOF_SIZE_T;
while (p <= e) { while (p <= e) {
size_t u = (*(const size_t *)p) & ASCII_CHAR_MASK; size_t u = (*(const size_t *)p) & ASCII_CHAR_MASK;
@ -5115,6 +5119,7 @@ find_first_nonascii(const unsigned char *start, const unsigned char *end)
} }
} }
#if PY_LITTLE_ENDIAN && HAVE_CTZ #if PY_LITTLE_ENDIAN && HAVE_CTZ
assert((end - p) < SIZEOF_SIZE_T);
// we can not use *(const size_t*)p to avoid buffer overrun. // we can not use *(const size_t*)p to avoid buffer overrun.
size_t u = load_unaligned(p, end - p) & ASCII_CHAR_MASK; size_t u = load_unaligned(p, end - p) & ASCII_CHAR_MASK;
if (u) { if (u) {