mirror of
https://github.com/borgbackup/borg.git
synced 2026-02-25 10:59:33 -05:00
9afebead84
CRC slice by 8 for generic CPUs outperforms zlib CRC32 on ppc and x86 (ARM untested but expected to as well). PCLMULQDQ derived from Intel's zlib patches outperforms every other CRC implementation by a huge margin.
41 lines
1.1 KiB
Cython
41 lines
1.1 KiB
Cython
|
|
from libc.stdint cimport uint32_t
|
|
from cpython.buffer cimport PyBUF_SIMPLE, PyObject_GetBuffer, PyBuffer_Release
|
|
|
|
|
|
cdef extern from "_crc32/crc32.c":
|
|
uint32_t _crc32_slice_by_8 "crc32_slice_by_8"(const void* data, size_t length, uint32_t initial_crc)
|
|
uint32_t _crc32_clmul "crc32_clmul"(const void* data, size_t length, uint32_t initial_crc)
|
|
|
|
int _have_clmul "have_clmul"()
|
|
|
|
|
|
cdef Py_buffer ro_buffer(object data) except *:
|
|
cdef Py_buffer view
|
|
PyObject_GetBuffer(data, &view, PyBUF_SIMPLE)
|
|
return view
|
|
|
|
|
|
def crc32_slice_by_8(data, value=0):
|
|
cdef Py_buffer data_buf = ro_buffer(data)
|
|
cdef uint32_t val = value
|
|
try:
|
|
return _crc32_slice_by_8(data_buf.buf, data_buf.len, val)
|
|
finally:
|
|
PyBuffer_Release(&data_buf)
|
|
|
|
|
|
def crc32_clmul(data, value=0):
|
|
cdef Py_buffer data_buf = ro_buffer(data)
|
|
cdef uint32_t val = value
|
|
try:
|
|
return _crc32_clmul(data_buf.buf, data_buf.len, val)
|
|
finally:
|
|
PyBuffer_Release(&data_buf)
|
|
|
|
|
|
have_clmul = _have_clmul()
|
|
if have_clmul:
|
|
crc32 = crc32_clmul
|
|
else:
|
|
crc32 = crc32_slice_by_8
|