Skip to content

GH-123894: Simplify ascii decode using unaligned loads #123895

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
wants to merge 1 commit into from
Closed

GH-123894: Simplify ascii decode using unaligned loads #123895

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
62 changes: 18 additions & 44 deletions Objects/unicodeobject.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4973,58 +4973,32 @@ static Py_ssize_t
ascii_decode(const char *start, const char *end, Py_UCS1 *dest)
{
const char *p = start;

#if SIZEOF_SIZE_T <= SIZEOF_VOID_P
if (_Py_IS_ALIGNED(p, ALIGNOF_SIZE_T)
&& _Py_IS_ALIGNED(dest, ALIGNOF_SIZE_T))
{
/* Fast path, see in STRINGLIB(utf8_decode) for
an explanation. */
/* Help allocation */
const char *_p = p;
Py_UCS1 * q = dest;
while (_p + SIZEOF_SIZE_T <= end) {
size_t value = *(const size_t *) _p;
if (value & ASCII_CHAR_MASK)
break;
*((size_t *)q) = value;
_p += SIZEOF_SIZE_T;
q += SIZEOF_SIZE_T;
}
p = _p;
while (p < end) {
if ((unsigned char)*p & 0x80)
break;
*q++ = *p++;
Py_UCS1 *q = dest;
const char *size_t_end = end - (SIZEOF_SIZE_T - 1);
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Not sure if we already discussed it but would _Py_ALIGN_DOWN(end, SIZEOF_SIZE_T) work?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Probably. Is using a macro here important?


while (p < size_t_end) {
/* Loading and storing a size_t using memcpy leads to unaligned loads
for platforms that can handle it. */
const size_t *restrict _p = (const size_t *)p;
size_t *restrict _q = (size_t *)q;
size_t value;
memcpy(&value, _p, SIZEOF_SIZE_T);
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

How are performances affected by those multiple memcpy() calls?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Memcpy isn't called. Memcpy with a given size gets optimized to a single load.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Q.E.D: https://godbolt.org/z/a9xdjfoq4

Copy link
Member

@picnixz picnixz Sep 10, 2024 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Ah I see the ASM yes. I'd suggest reformulating the comment by saying that "When supported, compilers optimize memcpy() into a single LOAD instruction if the number of bytes to copy is a literal value".

I confirmed it on GCC, but is it the case for MSVC & clang by the way? (NVM, I posted before seeing your reply)

if (value & ASCII_CHAR_MASK) {
break;
}
return p - start;
memcpy(_q, &value, SIZEOF_SIZE_T);
p += SIZEOF_SIZE_T;
q += SIZEOF_SIZE_T;
}
#endif
while (p < end) {
/* Fast path, see in STRINGLIB(utf8_decode) in stringlib/codecs.h
for an explanation. */
if (_Py_IS_ALIGNED(p, ALIGNOF_SIZE_T)) {
/* Help allocation */
const char *_p = p;
while (_p + SIZEOF_SIZE_T <= end) {
size_t value = *(const size_t *) _p;
if (value & ASCII_CHAR_MASK)
break;
_p += SIZEOF_SIZE_T;
}
p = _p;
if (_p == end)
break;
}
if ((unsigned char)*p & 0x80)
if ((unsigned char)*p & 0x80) {
break;
++p;
}
*q++ = *p++;
}
memcpy(dest, start, p - start);
return p - start;
}


static int
unicode_decode_utf8_impl(_PyUnicodeWriter *writer,
const char *starts, const char *s, const char *end,
Expand Down
Loading