view mercurial/mpatch_module.c @ 30451:94ca0e13d1fc

perf: add command for measuring revlog chunk operations Upcoming commits will teach revlogs to leverage the new compression engine API so that new compression formats can more easily be leveraged in revlogs. We want to be sure this refactoring doesn't regress performance. So this commit introduces "perfrevchunks" to explicitly test performance of reading, decompressing, and recompressing revlog chunks. Here is output when run on the mozilla-unified repo: $ hg perfrevlogchunks -c ! read ! wall 0.346603 comb 0.350000 user 0.340000 sys 0.010000 (best of 28) ! read w/ reused fd ! wall 0.337707 comb 0.340000 user 0.320000 sys 0.020000 (best of 30) ! read batch ! wall 0.013206 comb 0.020000 user 0.000000 sys 0.020000 (best of 221) ! read batch w/ reused fd ! wall 0.013259 comb 0.030000 user 0.010000 sys 0.020000 (best of 222) ! chunk ! wall 1.909939 comb 1.910000 user 1.900000 sys 0.010000 (best of 6) ! chunk batch ! wall 1.750677 comb 1.760000 user 1.740000 sys 0.020000 (best of 6) ! compress ! wall 5.668004 comb 5.670000 user 5.670000 sys 0.000000 (best of 3) $ hg perfrevlogchunks -m ! read ! wall 0.365834 comb 0.370000 user 0.350000 sys 0.020000 (best of 26) ! read w/ reused fd ! wall 0.350160 comb 0.350000 user 0.320000 sys 0.030000 (best of 28) ! read batch ! wall 0.024777 comb 0.020000 user 0.000000 sys 0.020000 (best of 119) ! read batch w/ reused fd ! wall 0.024895 comb 0.030000 user 0.000000 sys 0.030000 (best of 118) ! chunk ! wall 2.514061 comb 2.520000 user 2.480000 sys 0.040000 (best of 4) ! chunk batch ! wall 2.380788 comb 2.380000 user 2.360000 sys 0.020000 (best of 5) ! compress ! wall 9.815297 comb 9.820000 user 9.820000 sys 0.000000 (best of 3) We already see some interesting data, such as how much slower non-batched chunk reading is and that zlib compression appears to be >2x slower than decompression. I didn't have the data when I wrote this commit message, but I ran this on Mozilla's NFS-based Mercurial server and the time for reading with a reused file descriptor was faster. So I think it is worth testing both with and without file descriptor reuse so we can make informed decisions about recycling file descriptors.
author Gregory Szorc <gregory.szorc@gmail.com>
date Thu, 17 Nov 2016 20:17:51 -0800
parents 155f0cc3f813
children 5fc3459d0493
line wrap: on
line source

/*
 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}
};

#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.mpatch.mpatchError",
					  NULL, NULL);
	Py_INCREF(mpatch_Error);
	PyModule_AddObject(m, "mpatchError", mpatch_Error);

	return m;
}
#else
PyMODINIT_FUNC
initmpatch(void)
{
	Py_InitModule3("mpatch", methods, mpatch_doc);
	mpatch_Error = PyErr_NewException("mercurial.mpatch.mpatchError",
					  NULL, NULL);
}
#endif