##// END OF EJS Templates
perf: split obtaining chunks from decompression...
perf: split obtaining chunks from decompression Previously, the code was similar to what revlog._chunks() was doing, which took a raw data segment and delta chain, obtained buffers for the raw revlog chunks within, and decompressed them. This commit splits the "get raw chunks" action from "decompress." The goal of this change is to more accurately measurely decompression performance. On a ~50k deltachain for a manifest in mozilla-central: ! full ! wall 0.430548 comb 0.440000 user 0.410000 sys 0.030000 (best of 24) ! deltachain ! wall 0.016053 comb 0.010000 user 0.010000 sys 0.000000 (best of 181) ! read ! wall 0.008078 comb 0.010000 user 0.000000 sys 0.010000 (best of 362) ! rawchunks ! wall 0.033785 comb 0.040000 user 0.040000 sys 0.000000 (best of 100) ! decompress ! wall 0.327126 comb 0.320000 user 0.320000 sys 0.000000 (best of 31) ! patch ! wall 0.032391 comb 0.030000 user 0.030000 sys 0.000000 (best of 100) ! hash ! wall 0.012587 comb 0.010000 user 0.010000 sys 0.000000 (best of 233)

File last commit:

r30822:b54a2984 default
r30882:74cfc435 default
Show More
zstd.c
136 lines | 4.3 KiB | text/x-c | CLexer
/**
* Copyright (c) 2016-present, Gregory Szorc
* All rights reserved.
*
* This software may be modified and distributed under the terms
* of the BSD license. See the LICENSE file for details.
*/
/* A Python C extension for Zstandard. */
#include "python-zstandard.h"
PyObject *ZstdError;
PyDoc_STRVAR(estimate_compression_context_size__doc__,
"estimate_compression_context_size(compression_parameters)\n"
"\n"
"Give the amount of memory allocated for a compression context given a\n"
"CompressionParameters instance");
PyDoc_STRVAR(estimate_decompression_context_size__doc__,
"estimate_decompression_context_size()\n"
"\n"
"Estimate the amount of memory allocated to a decompression context.\n"
);
static PyObject* estimate_decompression_context_size(PyObject* self) {
return PyLong_FromSize_t(ZSTD_estimateDCtxSize());
}
PyDoc_STRVAR(get_compression_parameters__doc__,
"get_compression_parameters(compression_level[, source_size[, dict_size]])\n"
"\n"
"Obtains a ``CompressionParameters`` instance from a compression level and\n"
"optional input size and dictionary size");
PyDoc_STRVAR(train_dictionary__doc__,
"train_dictionary(dict_size, samples)\n"
"\n"
"Train a dictionary from sample data.\n"
"\n"
"A compression dictionary of size ``dict_size`` will be created from the\n"
"iterable of samples provided by ``samples``.\n"
"\n"
"The raw dictionary content will be returned\n");
static char zstd_doc[] = "Interface to zstandard";
static PyMethodDef zstd_methods[] = {
{ "estimate_compression_context_size", (PyCFunction)estimate_compression_context_size,
METH_VARARGS, estimate_compression_context_size__doc__ },
{ "estimate_decompression_context_size", (PyCFunction)estimate_decompression_context_size,
METH_NOARGS, estimate_decompression_context_size__doc__ },
{ "get_compression_parameters", (PyCFunction)get_compression_parameters,
METH_VARARGS, get_compression_parameters__doc__ },
{ "train_dictionary", (PyCFunction)train_dictionary,
METH_VARARGS | METH_KEYWORDS, train_dictionary__doc__ },
{ NULL, NULL }
};
void compressobj_module_init(PyObject* mod);
void compressor_module_init(PyObject* mod);
void compressionparams_module_init(PyObject* mod);
void constants_module_init(PyObject* mod);
void dictparams_module_init(PyObject* mod);
void compressiondict_module_init(PyObject* mod);
void compressionwriter_module_init(PyObject* mod);
void compressoriterator_module_init(PyObject* mod);
void decompressor_module_init(PyObject* mod);
void decompressobj_module_init(PyObject* mod);
void decompressionwriter_module_init(PyObject* mod);
void decompressoriterator_module_init(PyObject* mod);
void zstd_module_init(PyObject* m) {
/* python-zstandard relies on unstable zstd C API features. This means
that changes in zstd may break expectations in python-zstandard.
python-zstandard is distributed with a copy of the zstd sources.
python-zstandard is only guaranteed to work with the bundled version
of zstd.
However, downstream redistributors or packagers may unbundle zstd
from python-zstandard. This can result in a mismatch between zstd
versions and API semantics. This essentially "voids the warranty"
of python-zstandard and may cause undefined behavior.
We detect this mismatch here and refuse to load the module if this
scenario is detected.
*/
if (ZSTD_VERSION_NUMBER != 10102 || ZSTD_versionNumber() != 10102) {
PyErr_SetString(PyExc_ImportError, "zstd C API mismatch; Python bindings not compiled against expected zstd version");
return;
}
compressionparams_module_init(m);
dictparams_module_init(m);
compressiondict_module_init(m);
compressobj_module_init(m);
compressor_module_init(m);
compressionwriter_module_init(m);
compressoriterator_module_init(m);
constants_module_init(m);
decompressor_module_init(m);
decompressobj_module_init(m);
decompressionwriter_module_init(m);
decompressoriterator_module_init(m);
}
#if PY_MAJOR_VERSION >= 3
static struct PyModuleDef zstd_module = {
PyModuleDef_HEAD_INIT,
"zstd",
zstd_doc,
-1,
zstd_methods
};
PyMODINIT_FUNC PyInit_zstd(void) {
PyObject *m = PyModule_Create(&zstd_module);
if (m) {
zstd_module_init(m);
if (PyErr_Occurred()) {
Py_DECREF(m);
m = NULL;
}
}
return m;
}
#else
PyMODINIT_FUNC initzstd(void) {
PyObject *m = Py_InitModule3("zstd", zstd_methods, zstd_doc);
if (m) {
zstd_module_init(m);
}
}
#endif