Files
@ r13257:4c5b8120be59
Branch filter:
Location: cpp/openttd-patchpack/source/src/core/sort_func.hpp - annotation
r13257:4c5b8120be59
2.5 KiB
text/x-c++hdr
(svn r17776) -Codechange: [SDL] make "update the video card"-process asynchronious. Profiling with gprof etc. hasn't shown us that DrawSurfaceToScreen takes a significant amount of CPU; only using TIC/TOC it became apparant that it was a heavy CPU-cycle user or that it was waiting for something.
The benefit of making this function asynchronious ranges from 2%-25% (real time) during fast forward on dual core/hyperthreading-enabled CPUs; 8bpp improvements are, in my test cases, significantly smaller than 32bpp improvements.
On single core non-hyperthreading-enabled CPUs the extra locking/scheduling costs up to 1% extra realtime in fast forward. You can use -v sdl:no_threads to disable threading and undo this loss.
During normal non-fast-forwarded games the benefit/costs are negligable except when the gameloop takes more than about 90% of the time of a tick.
Note that allegro's performance does not improve with this system, likely due to their way of getting data to the video card. It is not implemented for the OS X/Windows video backends, unless (ofcourse) SDL is used there.
Funny is that the performance of the 32bpp(-anim) blitter is, at least in some test cases, significantly faster (more than 10%) than the 8bpp(-optimized) blitter when looking at real time in fast forward on a dual core CPU; it was slower.
The idea comes from a paper/report by Idar Borlaug and Knut Imar Hagen.
The benefit of making this function asynchronious ranges from 2%-25% (real time) during fast forward on dual core/hyperthreading-enabled CPUs; 8bpp improvements are, in my test cases, significantly smaller than 32bpp improvements.
On single core non-hyperthreading-enabled CPUs the extra locking/scheduling costs up to 1% extra realtime in fast forward. You can use -v sdl:no_threads to disable threading and undo this loss.
During normal non-fast-forwarded games the benefit/costs are negligable except when the gameloop takes more than about 90% of the time of a tick.
Note that allegro's performance does not improve with this system, likely due to their way of getting data to the video card. It is not implemented for the OS X/Windows video backends, unless (ofcourse) SDL is used there.
Funny is that the performance of the 32bpp(-anim) blitter is, at least in some test cases, significantly faster (more than 10%) than the 8bpp(-optimized) blitter when looking at real time in fast forward on a dual core CPU; it was slower.
The idea comes from a paper/report by Idar Borlaug and Knut Imar Hagen.
r9520:7cbc98578045 r9520:7cbc98578045 r12768:980ae0491352 r12768:980ae0491352 r12768:980ae0491352 r12768:980ae0491352 r12768:980ae0491352 r12768:980ae0491352 r12768:980ae0491352 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9575:665324c39b56 r9575:665324c39b56 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9575:665324c39b56 r9575:665324c39b56 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 r9520:7cbc98578045 | /* $Id$ */
/*
* This file is part of OpenTTD.
* OpenTTD is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, version 2.
* OpenTTD is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
* See the GNU General Public License for more details. You should have received a copy of the GNU General Public License along with OpenTTD. If not, see <http://www.gnu.org/licenses/>.
*/
/** @file sort_func.hpp Functions related to sorting operations. */
#ifndef SORT_FUNC_HPP
#define SORT_FUNC_HPP
#include <stdlib.h>
#include "math_func.hpp"
#include "mem_func.hpp"
/**
* Type safe qsort()
*
* @note Use this sort for irregular sorted data.
*
* @param base Pointer to the first element of the array to be sorted.
* @param num Number of elements in the array pointed by base.
* @param comparator Function that compares two elements.
* @param desc Sort descending.
*/
template <typename T>
static FORCEINLINE void QSortT(T *base, uint num, int (CDECL *comparator)(const T*, const T*), bool desc = false)
{
if (num < 2) return;
qsort(base, num, sizeof(T), (int (CDECL *)(const void *, const void *))comparator);
if (desc) MemReverseT(base, num);
}
/**
* Type safe Gnome Sort.
*
* This is a slightly modifyied Gnome search. The basic
* Gnome search trys to sort already sorted list parts.
* The modification skips these.
*
* @note Use this sort for presorted / regular sorted data.
*
* @param base Pointer to the first element of the array to be sorted.
* @param num Number of elements in the array pointed by base.
* @param comparator Function that compares two elements.
* @param desc Sort descending.
*/
template <typename T>
static inline void GSortT(T *base, uint num, int (CDECL *comparator)(const T*, const T*), bool desc = false)
{
if (num < 2) return;
assert(base != NULL);
assert(comparator != NULL);
T *a = base;
T *b = base + 1;
uint offset = 0;
while (num > 1) {
const int diff = comparator(a, b);
if ((!desc && diff <= 0) || (desc && diff >= 0)) {
if (offset != 0) {
/* Jump back to the last direction switch point */
a += offset;
b += offset;
offset = 0;
continue;
}
a++;
b++;
num--;
} else {
Swap(*a, *b);
if (a == base) continue;
a--;
b--;
offset++;
}
}
}
#endif /* SORT_FUNC_HPP */
|