##// END OF EJS Templates
perf: benchmark command for revlog indexes...
perf: benchmark command for revlog indexes We didn't have explicit microbenchmark coverage for loading revlog indexes. That seems like a useful thing to have, so let's add it. We currently measure the low-level nodemap APIs. There is room to hook in at the actual revlog layer. This could be done as a follow-up. The hackiest thing about this patch is specifying revlog paths. Other commands have arguments that allow resolution of changelog, manifest, and filelog. I needed to hook in at a lower level of the revlog API than what the existing helper functions to resolve revlogs allowed. I was too lazy to write some new APIs. This could be done as a follow-up easily enough. Example output for `hg perfrevlogindex 00changelog.i` on my Firefox repo (404418 revisions): ! revlog constructor ! wall 0.003106 comb 0.000000 user 0.000000 sys 0.000000 (best of 912) ! read ! wall 0.003077 comb 0.000000 user 0.000000 sys 0.000000 (best of 924) ! create index object ! wall 0.000000 comb 0.000000 user 0.000000 sys 0.000000 (best of 1803994) ! retrieve index entry for rev 0 ! wall 0.000193 comb 0.000000 user 0.000000 sys 0.000000 (best of 14037) ! look up missing node ! wall 0.003313 comb 0.000000 user 0.000000 sys 0.000000 (best of 865) ! look up node at rev 0 ! wall 0.003295 comb 0.010000 user 0.010000 sys 0.000000 (best of 858) ! look up node at 1/4 len ! wall 0.002598 comb 0.010000 user 0.010000 sys 0.000000 (best of 1103) ! look up node at 1/2 len ! wall 0.001909 comb 0.000000 user 0.000000 sys 0.000000 (best of 1507) ! look up node at 3/4 len ! wall 0.001213 comb 0.000000 user 0.000000 sys 0.000000 (best of 2275) ! look up node at tip ! wall 0.000453 comb 0.000000 user 0.000000 sys 0.000000 (best of 5697) ! look up all nodes (forward) ! wall 0.094615 comb 0.100000 user 0.100000 sys 0.000000 (best of 100) ! look up all nodes (reverse) ! wall 0.045889 comb 0.050000 user 0.050000 sys 0.000000 (best of 100) ! retrieve all index entries (forward) ! wall 0.078398 comb 0.080000 user 0.060000 sys 0.020000 (best of 100) ! retrieve all index entries (reverse) ! wall 0.079376 comb 0.080000 user 0.070000 sys 0.010000 (best of 100)

File last commit:

r32371:151cc3b3 default
r32532:e4f51462 default
Show More
mpatch.c
200 lines | 4.4 KiB | text/x-c | CLexer
/*
mpatch.c - efficient binary patching for Mercurial
This implements a patch algorithm that's O(m + nlog n) where m is the
size of the output and n is the number of patches.
Given a list of binary patches, it unpacks each into a hunk list,
then combines the hunk lists with a treewise recursion to form a
single hunk list. This hunk list is then applied to the original
text.
The text (or binary) fragments are copied directly from their source
Python objects into a preallocated output string to avoid the
allocation of intermediate Python objects. Working memory is about 2x
the total number of hunks.
Copyright 2005, 2006 Matt Mackall <mpm@selenic.com>
This software may be used and distributed according to the terms
of the GNU General Public License, incorporated herein by reference.
*/
#define PY_SSIZE_T_CLEAN
#include <Python.h>
#include <stdlib.h>
#include <string.h>
#include "util.h"
#include "bitmanipulation.h"
#include "compat.h"
#include "mpatch.h"
static char mpatch_doc[] = "Efficient binary patching.";
static PyObject *mpatch_Error;
static void setpyerr(int r)
{
switch (r) {
case MPATCH_ERR_NO_MEM:
PyErr_NoMemory();
break;
case MPATCH_ERR_CANNOT_BE_DECODED:
PyErr_SetString(mpatch_Error, "patch cannot be decoded");
break;
case MPATCH_ERR_INVALID_PATCH:
PyErr_SetString(mpatch_Error, "invalid patch");
break;
}
}
struct mpatch_flist *cpygetitem(void *bins, ssize_t pos)
{
const char *buffer;
struct mpatch_flist *res;
ssize_t blen;
int r;
PyObject *tmp = PyList_GetItem((PyObject*)bins, pos);
if (!tmp)
return NULL;
if (PyObject_AsCharBuffer(tmp, &buffer, (Py_ssize_t*)&blen))
return NULL;
if ((r = mpatch_decode(buffer, blen, &res)) < 0) {
if (!PyErr_Occurred())
setpyerr(r);
return NULL;
}
return res;
}
static PyObject *
patches(PyObject *self, PyObject *args)
{
PyObject *text, *bins, *result;
struct mpatch_flist *patch;
const char *in;
int r = 0;
char *out;
Py_ssize_t len, outlen, inlen;
if (!PyArg_ParseTuple(args, "OO:mpatch", &text, &bins))
return NULL;
len = PyList_Size(bins);
if (!len) {
/* nothing to do */
Py_INCREF(text);
return text;
}
if (PyObject_AsCharBuffer(text, &in, &inlen))
return NULL;
patch = mpatch_fold(bins, cpygetitem, 0, len);
if (!patch) { /* error already set or memory error */
if (!PyErr_Occurred())
PyErr_NoMemory();
return NULL;
}
outlen = mpatch_calcsize(inlen, patch);
if (outlen < 0) {
r = (int)outlen;
result = NULL;
goto cleanup;
}
result = PyBytes_FromStringAndSize(NULL, outlen);
if (!result) {
result = NULL;
goto cleanup;
}
out = PyBytes_AsString(result);
if ((r = mpatch_apply(out, in, inlen, patch)) < 0) {
Py_DECREF(result);
result = NULL;
}
cleanup:
mpatch_lfree(patch);
if (!result && !PyErr_Occurred())
setpyerr(r);
return result;
}
/* calculate size of a patched file directly */
static PyObject *
patchedsize(PyObject *self, PyObject *args)
{
long orig, start, end, len, outlen = 0, last = 0, pos = 0;
Py_ssize_t patchlen;
char *bin;
if (!PyArg_ParseTuple(args, "ls#", &orig, &bin, &patchlen))
return NULL;
while (pos >= 0 && pos < patchlen) {
start = getbe32(bin + pos);
end = getbe32(bin + pos + 4);
len = getbe32(bin + pos + 8);
if (start > end)
break; /* sanity check */
pos += 12 + len;
outlen += start - last;
last = end;
outlen += len;
}
if (pos != patchlen) {
if (!PyErr_Occurred())
PyErr_SetString(mpatch_Error, "patch cannot be decoded");
return NULL;
}
outlen += orig - last;
return Py_BuildValue("l", outlen);
}
static PyMethodDef methods[] = {
{"patches", patches, METH_VARARGS, "apply a series of patches\n"},
{"patchedsize", patchedsize, METH_VARARGS, "calculed patched size\n"},
{NULL, NULL}
};
static const int version = 1;
#ifdef IS_PY3K
static struct PyModuleDef mpatch_module = {
PyModuleDef_HEAD_INIT,
"mpatch",
mpatch_doc,
-1,
methods
};
PyMODINIT_FUNC PyInit_mpatch(void)
{
PyObject *m;
m = PyModule_Create(&mpatch_module);
if (m == NULL)
return NULL;
mpatch_Error = PyErr_NewException("mercurial.cext.mpatch.mpatchError",
NULL, NULL);
Py_INCREF(mpatch_Error);
PyModule_AddObject(m, "mpatchError", mpatch_Error);
PyModule_AddIntConstant(m, "version", version);
return m;
}
#else
PyMODINIT_FUNC
initmpatch(void)
{
PyObject *m;
m = Py_InitModule3("mpatch", methods, mpatch_doc);
mpatch_Error = PyErr_NewException("mercurial.cext.mpatch.mpatchError",
NULL, NULL);
PyModule_AddIntConstant(m, "version", version);
}
#endif