This repository has been archived on 2023-10-09. You can view files and clone it. You cannot open issues or pull requests or push a commit.
Files
blender-archive/intern/cycles/util/string.cpp
Michael Jones (Apple) da4ef05e4d Cycles: Apple Silicon optimization to specialize intersection kernels
The Metal backend now compiles and caches a second set of kernels which are
optimized for scene contents, enabled for Apple Silicon.

The implementation supports doing this both for intersection and shading
kernels. However this is currently only enabled for intersection kernels that
are quick to compile, and already give a good speedup. Enabling this for
shading kernels would be faster still, however this also causes a long wait
times and would need a good user interface to control this.

M1 Max samples per minute (macOS 13.0):

                    PSO_GENERIC  PSO_SPECIALIZED_INTERSECT  PSO_SPECIALIZED_SHADE

barbershop_interior       83.4	            89.5                   93.7
bmw27                   1486.1	          1671.0                 1825.8
classroom                175.2	           196.8                  206.3
fishy_cat                674.2	           704.3                  719.3
junkshop                 205.4	           212.0                  257.7
koro                     310.1	           336.1                  342.8
monster                  376.7	           418.6                  424.1
pabellon                 273.5	           325.4                  339.8
sponza                   830.6	           929.6                 1142.4
victor                    86.7              96.4                   96.3
wdas_cloud               111.8	           112.7                  183.1

Code contributed by Jason Fielder, Morteza Mostajabodaveh and Michael Jones

Differential Revision: https://developer.blender.org/D14645
2022-07-15 13:40:04 +02:00

274 lines
5.9 KiB
C++

/* SPDX-License-Identifier: Apache-2.0
* Copyright 2011-2022 Blender Foundation */
#include <stdarg.h>
#include <stdio.h>
#include <algorithm>
#include <cctype>
#include "util/foreach.h"
#include "util/string.h"
#include "util/windows.h"
#ifdef _WIN32
# ifndef vsnprintf
# define vsnprintf _vsnprintf
# endif
#endif /* _WIN32 */
CCL_NAMESPACE_BEGIN
string string_printf(const char *format, ...)
{
vector<char> str(128, 0);
while (1) {
va_list args;
int result;
va_start(args, format);
result = vsnprintf(&str[0], str.size(), format, args);
va_end(args);
if (result == -1) {
/* not enough space or formatting error */
if (str.size() > 65536) {
assert(0);
return string("");
}
str.resize(str.size() * 2, 0);
continue;
}
else if (result >= (int)str.size()) {
/* not enough space */
str.resize(result + 1, 0);
continue;
}
return string(&str[0]);
}
}
bool string_iequals(const string &a, const string &b)
{
if (a.size() == b.size()) {
for (size_t i = 0; i < a.size(); i++)
if (toupper(a[i]) != toupper(b[i]))
return false;
return true;
}
return false;
}
void string_split(vector<string> &tokens,
const string &str,
const string &separators,
bool skip_empty_tokens)
{
size_t token_start = 0, token_length = 0;
for (size_t i = 0; i < str.size(); ++i) {
const char ch = str[i];
if (separators.find(ch) == string::npos) {
/* Current character is not a separator,
* append it to token by increasing token length.
*/
++token_length;
}
else {
/* Current character is a separator,
* append current token to the list.
*/
if (!skip_empty_tokens || token_length > 0) {
string token = str.substr(token_start, token_length);
tokens.push_back(token);
}
token_start = i + 1;
token_length = 0;
}
}
/* Append token from the tail of the string if exists. */
if (token_length) {
string token = str.substr(token_start, token_length);
tokens.push_back(token);
}
}
bool string_startswith(const string_view s, const string_view start)
{
const size_t len = start.size();
if (len > s.size()) {
return false;
}
return strncmp(s.c_str(), start.data(), len) == 0;
}
bool string_endswith(const string_view s, const string_view end)
{
const size_t len = end.size();
if (len > s.size()) {
return false;
}
return strncmp(s.c_str() + s.size() - len, end.data(), len) == 0;
}
string string_strip(const string &s)
{
string result = s;
result.erase(0, result.find_first_not_of(' '));
result.erase(result.find_last_not_of(' ') + 1);
return result;
}
void string_replace(string &haystack, const string &needle, const string &other)
{
size_t i = 0, index;
while ((index = haystack.find(needle, i)) != string::npos) {
haystack.replace(index, needle.size(), other);
i = index + other.size();
}
}
void string_replace_same_length(string &haystack, const string &needle, const string &other)
{
assert(needle.size() == other.size());
size_t pos = 0;
while (pos != string::npos) {
pos = haystack.find(needle, pos);
if (pos != string::npos) {
memcpy(haystack.data() + pos, other.data(), other.size());
pos += other.size();
}
}
}
string string_remove_trademark(const string &s)
{
string result = s;
/* Special case, so we don't leave sequential spaces behind. */
/* TODO(sergey): Consider using regex perhaps? */
string_replace(result, " (TM)", "");
string_replace(result, " (R)", "");
string_replace(result, "(TM)", "");
string_replace(result, "(R)", "");
return string_strip(result);
}
string string_from_bool(bool var)
{
if (var)
return "True";
else
return "False";
}
string to_string(const char *str)
{
return string(str);
}
string to_string(const float4 &v)
{
return string_printf("%f,%f,%f,%f", v.x, v.y, v.z, v.w);
}
string string_to_lower(const string &s)
{
string r = s;
std::transform(r.begin(), r.end(), r.begin(), [](char c) { return std::tolower(c); });
return r;
}
/* Wide char strings helpers for Windows. */
#ifdef _WIN32
wstring string_to_wstring(const string &str)
{
const int length_wc = MultiByteToWideChar(CP_UTF8, 0, str.c_str(), str.length(), NULL, 0);
wstring str_wc(length_wc, 0);
MultiByteToWideChar(CP_UTF8, 0, str.c_str(), str.length(), &str_wc[0], length_wc);
return str_wc;
}
string string_from_wstring(const wstring &str)
{
int length_mb = WideCharToMultiByte(CP_UTF8, 0, str.c_str(), str.size(), NULL, 0, NULL, NULL);
string str_mb(length_mb, 0);
WideCharToMultiByte(CP_UTF8, 0, str.c_str(), str.size(), &str_mb[0], length_mb, NULL, NULL);
return str_mb;
}
string string_to_ansi(const string &str)
{
const int length_wc = MultiByteToWideChar(CP_UTF8, 0, str.c_str(), str.length(), NULL, 0);
wstring str_wc(length_wc, 0);
MultiByteToWideChar(CP_UTF8, 0, str.c_str(), str.length(), &str_wc[0], length_wc);
int length_mb = WideCharToMultiByte(
CP_ACP, 0, str_wc.c_str(), str_wc.size(), NULL, 0, NULL, NULL);
string str_mb(length_mb, 0);
WideCharToMultiByte(CP_ACP, 0, str_wc.c_str(), str_wc.size(), &str_mb[0], length_mb, NULL, NULL);
return str_mb;
}
#endif /* _WIN32 */
string string_human_readable_size(size_t size)
{
static const char suffixes[] = "BKMGTPEZY";
const char *suffix = suffixes;
size_t r = 0;
while (size >= 1024) {
r = size % 1024;
size /= 1024;
suffix++;
}
if (*suffix != 'B')
return string_printf("%.2f%c", double(size * 1024 + r) / 1024.0, *suffix);
else
return string_printf("%zu", size);
}
string string_human_readable_number(size_t num)
{
if (num == 0) {
return "0";
}
/* Add thousands separators. */
char buf[32];
char *p = buf + 31;
*p = '\0';
int i = -1;
while (num) {
if (++i && i % 3 == 0)
*(--p) = ',';
*(--p) = '0' + (num % 10);
num /= 10;
}
return p;
}
CCL_NAMESPACE_END