pcsx2/plugins/zzogl-pg/opengl/HostMemory.cpp
gregory.hainaut e3c741bb2a zzogl: painfully merge the zzogl-dev branch
* new memory management
* asm was replaced by intrinsic
* new GLSL backend (AMD only) Cmake is probably broken anyway with the 2 plugins...
* and lots of others stuff that I forgot about it ;)


git-svn-id: http://pcsx2.googlecode.com/svn/trunk@5166 96395faa-99c1-11dd-bbfe-3dabce05a288
2012-04-19 21:22:08 +00:00

621 lines
18 KiB
C++

/* ZZ Open GL graphics plugin
* Copyright (c)2009-2010 zeydlitz@gmail.com, arcum42@gmail.com
* Based on Zerofrog's ZeroGS KOSMOS (c)2005-2008
*
* This program 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; either version 2 of the License, or
* (at your option) any later version.
*
* This program 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 this program; if not, write to the Free Software
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA
*/
#include "GS.h"
#include <stdlib.h>
#include "Mem.h"
#include "x86.h"
#include "targets.h"
#include "ZZoglVB.h"
// flush current vertices, call before setting new registers (the main render method)
extern void Flush(int context);
u8* g_pbyGSMemory = NULL; // 4Mb GS system mem
void GSMemory::init()
{
const u32 mem_size = MEMORY_END + 0x10000; // leave some room for out of range accesses (saves on the checks)
// clear
g_pbyGSMemory = (u8*)_aligned_malloc(mem_size, 1024);
memset(g_pbyGSMemory, 0, mem_size);
}
void GSMemory::destroy()
{
_aligned_free(g_pbyGSMemory);
g_pbyGSMemory = NULL;
}
u8* GSMemory::get()
{
return g_pbyGSMemory;
}
u8* GSMemory::get(u32 addr)
{
return &g_pbyGSMemory[addr*8];
}
u8* GSMemory::get_raw(u32 addr)
{
return &g_pbyGSMemory[addr];
}
u8* g_pbyGSClut = NULL; // ZZ
void GSClut::init()
{
g_pbyGSClut = (u8*)_aligned_malloc(256 * 8, 1024); // need 512 alignment!
memset(g_pbyGSClut, 0, 256*8);
}
void GSClut::destroy()
{
_aligned_free(g_pbyGSClut);
g_pbyGSClut = NULL;
}
u8* GSClut::get()
{
return g_pbyGSClut;
}
u8* GSClut::get(u32 addr)
{
return &g_pbyGSClut[addr*8];
}
u8* GSClut::get_raw(u32 addr)
{
return &g_pbyGSClut[addr];
}
extern _getPixelAddress getPixelFun[64];
extern CRangeManager s_RangeMngr; // manages overwritten memory
extern void ResolveInRange(int start, int end);
static vector<u8> s_vTempBuffer, s_vTransferCache;
static int gs_imageEnd = 0;
// From the start of monster labs. In all 3 cases, psm == 0.
// ZZogl-PG: GetRectMemAddress(0x3f4000, 0x404000, 0x0, 0x0, 0x0, 0x100, 0x40, 0x3f40, 0x100);
// ZZogl-PG: GetRectMemAddress(0x3f8000, 0x408000, 0x0, 0x0, 0x0, 0x100, 0x40, 0x3f80, 0x100);
// ZZogl-PG: GetRectMemAddress(0x3fc000, 0x40c000, 0x0, 0x0, 0x0, 0x100, 0x40, 0x3fc0, 0x100);
void GetRectMemAddress(int& start, int& end, int psm, int x, int y, int w, int h, int bp, int bw)
{
FUNCLOG
u32 bits = 0;
if (m_Blocks[psm].bpp == 0)
{
ZZLog::Error_Log("ZeroGS: Bad psm 0x%x.", psm);
start = 0;
end = MEMORY_END;
return;
}
if (PSMT_ISZTEX(psm))
{
// This still needs an eye kept on it.
const BLOCK& b = m_Blocks[psm];
const int x2 = x + w + b.width - 1;
const int y2 = y + h - 1;
bw = bw / b.width;
start = (bp + ((y / b.height) * bw + (x / b.width)) * 0x20) * 0x100;
end = (bp + ((y2 / b.height) * bw + (x2 / b.width)) * 0x20) * 0x100;
return;
}
bits = PSMT_BITS_NUM(psm);
start = getPixelFun[psm](x, y, bp, bw);
end = getPixelFun[psm](x + w - 1, y + h - 1, bp, bw) + 1;
if (bits > 0)
{
start *= bits;
end *= bits;
}
else
{
start /= 2;
end /= 2;
}
}
// Same as GetRectMemAddress, except that we know x & y are zero, so it's simplified a bit.
void GetRectMemAddressZero(int& start, int& end, int psm, int w, int h, int bp, int bw)
{
FUNCLOG
u32 bits = 0;
if (m_Blocks[psm].bpp == 0)
{
ZZLog::Error_Log("ZeroGS: Bad psm 0x%x.", psm);
start = 0;
end = MEMORY_END;
return;
}
if (PSMT_ISZTEX(psm))
{
// This still needs an eye kept on it.
const BLOCK& b = m_Blocks[psm];
const int x2 = w + b.width - 1;
const int y2 = h - 1;
bw = bw / b.width;
start = bp * 0x100;
end = (bp + ((y2 / b.height) * bw + (x2 / b.width)) * 0x20) * 0x100;
return;
}
bits = PSMT_BITS_NUM(psm);
start = getPixelFun[psm](0, 0, bp, bw);
end = getPixelFun[psm](w - 1, h - 1, bp, bw) + 1;
if (bits > 0)
{
start *= bits;
end *= bits;
}
else
{
start /= 2;
end /= 2;
}
}
void GetRectMemAddress(int& start, int& end, int psm, Point p, Size s, int bp, int bw)
{
GetRectMemAddress(start, end, psm, p.x, p.y, s.w, s.h, bp, bw);
}
void GetRectMemAddress(int& start, int& end, int psm, int x, int y, Size s, int bp, int bw)
{
GetRectMemAddress(start, end, psm, x, y, s.w, s.h, bp, bw);
}
void GetRectMemAddressZero(int& start, int& end, int psm, Size s, int bp, int bw)
{
GetRectMemAddressZero(start, end, psm, s.w, s.h, bp, bw);
}
void InitTransferHostLocal()
{
FUNCLOG
#if defined(_DEBUG)
// Xenosaga 1.
if (gs.trxpos.dx + gs.imageNew.w > gs.dstbuf.bw)
ZZLog::Debug_Log("Transfer error, width exceeded. (0x%x > 0X%x)", gs.trxpos.dx + gs.imageNew.w, gs.dstbuf.bw);
#endif
//bool bHasFlushed = false;
gs.image.x = gs.trxpos.dx;
gs.image.y = gs.trxpos.dy;
gs.imageEnd.x = gs.image.x + gs.imageNew.w;
gs.imageEnd.y = gs.image.y + gs.imageNew.h;
assert(gs.imageEnd.x < 2048 && gs.imageEnd.y < 2048);
// This needs to be looked in to, since psm should *not* be 63.
// hack! viewful joe
if (gs.dstbuf.psm == 63)
{
ZZLog::WriteLn("gs.dstbuf.psm set to 0!");
gs.dstbuf.psm = 0;
}
int start, end;
GetRectMemAddress(start, end, gs.dstbuf.psm, gs.trxpos.dx, gs.trxpos.dy, gs.imageNew, gs.dstbuf.bp, gs.dstbuf.bw);
if (end > MEMORY_END)
{
// Monster Lab - the screwed up title screen
// Init host local out of bounds! (end == 0x404000)
// Init host local out of bounds! (end == 0x408000)
// Init host local out of bounds! (end == 0x40c000)
// MEMORY_END is 0x400000...
ZZLog::Warn_Log("Init host local out of bounds! (end == 0x%x)", end);
//gs.transferring = false;
end = MEMORY_END;
}
gs_imageEnd = end;
if (vb[0].nCount > 0) Flush(0);
if (vb[1].nCount > 0) Flush(1);
//ZZLog::Prim_Log("trans: bp:%x x:%x y:%x w:%x h:%x\n", gs.dstbuf.bp, gs.trxpos.dx, gs.trxpos.dy, gs.imageNew.w, gs.imageNew.h);
}
void TransferHostLocal(const void* pbyMem, u32 nQWordSize)
{
FUNCLOG
int start = -1, end = -1;
GetRectMemAddress(start, end, gs.dstbuf.psm, gs.image, gs.imageNew, gs.dstbuf.bp, gs.dstbuf.bw);
if ((start == -1) || (end == -1)) ZZLog::WriteLn("start == %d, end == %d", start, end);
assert(start < gs_imageEnd);
end = gs_imageEnd;
// sometimes games can decompress to alpha channel of render target only, in this case
// do a resolve right away. wolverine x2
if (((gs.dstbuf.psm == PSMT8H) || (gs.dstbuf.psm == PSMT4HL) || (gs.dstbuf.psm == PSMT4HH)) && !(conf.settings().gust))
{
list<CRenderTarget*> listTransmissionUpdateTargs;
s_RTs.GetTargs(start, end, listTransmissionUpdateTargs);
for (list<CRenderTarget*>::iterator it = listTransmissionUpdateTargs.begin(); it != listTransmissionUpdateTargs.end(); ++it)
{
CRenderTarget* ptarg = *it;
if ((ptarg->status & CRenderTarget::TS_Virtual)) continue;
//ZZLog::Error_Log("Resolving to alpha channel.");
ptarg->Resolve();
}
}
s_RangeMngr.Insert(start, min(end, start + (int)nQWordSize*16));
const u8* porgend = (const u8*)pbyMem + 4 * nQWordSize;
if (s_vTransferCache.size() > 0)
{
int imagecache = s_vTransferCache.size();
s_vTempBuffer.resize(imagecache + nQWordSize*4);
memcpy(&s_vTempBuffer[0], &s_vTransferCache[0], imagecache);
memcpy(&s_vTempBuffer[imagecache], pbyMem, nQWordSize*4);
pbyMem = (const void*) & s_vTempBuffer[0];
porgend = &s_vTempBuffer[0] + s_vTempBuffer.size();
int wordinc = imagecache / 4;
if ((nQWordSize * 4 + imagecache) / 3 == ((nQWordSize + wordinc) * 4) / 3)
{
// can use the data
nQWordSize += wordinc;
}
}
int leftover = m_Blocks[gs.dstbuf.psm].TransferHostLocal(pbyMem, nQWordSize);
if (leftover > 0)
{
// copy the last gs.image24bitOffset to the cache
s_vTransferCache.resize(leftover);
memcpy(&s_vTransferCache[0], porgend - leftover, leftover);
}
else
{
s_vTransferCache.resize(0);
}
#if defined(_DEBUG)
if (g_bSaveTrans)
{
tex0Info t;
t.tbp0 = gs.dstbuf.bp;
t.tw = gs.imageNew.w;
t.th = gs.imageNew.h;
t.tbw = gs.dstbuf.bw;
t.psm = gs.dstbuf.psm;
SaveTex(&t, 0);
}
#endif
}
void InitTransferLocalHost()
{
FUNCLOG
assert(gs.trxpos.sx + gs.imageNew.w <= 2048 && gs.trxpos.sy + gs.imageNew.h <= 2048);
#if defined(_DEBUG)
if (gs.trxpos.sx + gs.imageNew.w > gs.srcbuf.bw)
ZZLog::Debug_Log("Transfer error, width exceeded. (0x%x > 0x%x)", gs.trxpos.sx + gs.imageNew.w, gs.srcbuf.bw);
#endif
gs.image.x = gs.trxpos.sx;
gs.image.y = gs.trxpos.sy;
gs.imageEnd.x = gs.image.x + gs.imageNew.w;
gs.imageEnd.y = gs.image.y + gs.imageNew.h;
s_vTransferCache.resize(0);
int start, end;
GetRectMemAddress(start, end, gs.srcbuf.psm, gs.trxpos.sx, gs.trxpos.sy, gs.imageNew, gs.srcbuf.bp, gs.srcbuf.bw);
ResolveInRange(start, end);
}
template <class T>
void TransferLocalHost(void* pbyMem, u32 nQWordSize, int& x, int& y, u8 *pstart)
{
_readPixel_0 rp = readPixelFun_0[gs.srcbuf.psm];
int i = x, j = y;
T* pbuf = (T*)pbyMem;
u32 nSize = nQWordSize * 16 / sizeof(T);
for (; i < gs.imageEnd.y; ++i)
{
for (; j < gs.imageEnd.x && nSize > 0; ++j, --nSize)
{
*pbuf++ = rp(pstart, j % 2048, i % 2048, gs.srcbuf.bw);
}
if (j >= gs.imageEnd.x)
{
assert(j == gs.imageEnd.x);
j = gs.trxpos.sx;
}
else
{
assert(nSize == 0);
break;
}
}
}
void TransferLocalHost_24(void* pbyMem, u32 nQWordSize, int& x, int& y, u8 *pstart)
{
_readPixel_0 rp = readPixelFun_0[gs.srcbuf.psm];
int i = x, j = y;
u8* pbuf = (u8*)pbyMem;
u32 nSize = nQWordSize * 16 / 3;
for (; i < gs.imageEnd.y; ++i)
{
for (; j < gs.imageEnd.x && nSize > 0; ++j, --nSize)
{
u32 p = rp(pstart, j % 2048, i % 2048, gs.srcbuf.bw);
pbuf[0] = (u8)p;
pbuf[1] = (u8)(p >> 8);
pbuf[2] = (u8)(p >> 16);
pbuf += 3;
}
if (j >= gs.imageEnd.x)
{
assert(j == gs.imageEnd.x);
j = gs.trxpos.sx;
}
else
{
assert(nSize == 0);
break;
}
}
}
// left/right, top/down
void TransferLocalHost(void* pbyMem, u32 nQWordSize)
{
FUNCLOG
assert(gs.imageTransfer == XFER_LOCAL_TO_HOST);
u8* pstart = g_pbyGSMemory + 256 * gs.srcbuf.bp;
switch(PSMT_BITMODE(gs.srcbuf.psm))
{
case 0:
TransferLocalHost<u32>(pbyMem, nQWordSize, gs.image.y, gs.image.x, pstart);
break;
case 1:
TransferLocalHost_24(pbyMem, nQWordSize, gs.image.y, gs.image.x, pstart);
break;
case 2:
TransferLocalHost<u16>(pbyMem, nQWordSize, gs.image.y, gs.image.x, pstart);
break;
case 3:
TransferLocalHost<u8>(pbyMem, nQWordSize, gs.image.y, gs.image.x, pstart);
break;
default:
assert(0);
break;
}
if (gs.image.y >= gs.imageEnd.y)
{
ZZLog::Error_Log("gs.image.y >= gs.imageEnd.y!");
assert(gs.image.y == gs.imageEnd.y);
gs.transferring = false;
}
}
__forceinline void _TransferLocalLocal()
{
//ZZLog::Error_Log("TransferLocalLocal(0x%x, 0x%x)", gs.srcbuf.psm, gs.dstbuf.psm);
_writePixel_0 wp = writePixelFun_0[gs.srcbuf.psm];
_readPixel_0 rp = readPixelFun_0[gs.dstbuf.psm];
u8* pSrcBuf = g_pbyGSMemory + gs.srcbuf.bp * 256;
u8* pDstBuf = g_pbyGSMemory + gs.dstbuf.bp * 256;
u32 widthlimit = 4;
u32 maxX = gs.trxpos.sx + gs.imageNew.w;
u32 maxY = gs.trxpos.sy + gs.imageNew.h;
if (PSMT_BITMODE(gs.srcbuf.psm) == 0) widthlimit = 2;
if ((gs.imageNew.w & widthlimit) != 0) return;
for(u32 i = gs.trxpos.sy, i2 = gs.trxpos.dy; i < maxY; i++, i2++)
{
for(u32 j = gs.trxpos.sx, j2 = gs.trxpos.dx; j < maxX; j += widthlimit, j2 += widthlimit)
{
wp(pDstBuf, j2%2048, i2%2048,
rp(pSrcBuf, j%2048, i%2048, gs.srcbuf.bw), gs.dstbuf.bw);
wp(pDstBuf, (j2+1)%2048, i2%2048,
rp(pSrcBuf, (j+1)%2048, i%2048, gs.srcbuf.bw), gs.dstbuf.bw);
if (widthlimit > 2)
{
// Then widthlimit == 4.
wp(pDstBuf, (j2+2)%2048, i2%2048,
rp(pSrcBuf, (j+2)%2048, i%2048, gs.srcbuf.bw), gs.dstbuf.bw);
wp(pDstBuf, (j2+3)%2048, i2%2048,
rp(pSrcBuf, (j+3)%2048, i%2048, gs.srcbuf.bw), gs.dstbuf.bw);
}
}
}
}
__forceinline void _TransferLocalLocal_4()
{
//ZZLog::Error_Log("TransferLocalLocal_4(0x%x, 0x%x)", gs.srcbuf.psm, gs.dstbuf.psm);
_getPixelAddress_0 gsp = getPixelFun_0[gs.srcbuf.psm];
_getPixelAddress_0 gdp = getPixelFun_0[gs.dstbuf.psm];
u8* pSrcBuf = g_pbyGSMemory + gs.srcbuf.bp * 256;
u8* pDstBuf = g_pbyGSMemory + gs.dstbuf.bp * 256;
u32 maxX = gs.trxpos.sx + gs.imageNew.w;
u32 maxY = gs.trxpos.sy + gs.imageNew.h;
assert((gs.imageNew.w % 8) == 0);
for(u32 i = gs.trxpos.sy, i2 = gs.trxpos.dy; i < maxY; ++i, ++i2)
{
for(u32 j = gs.trxpos.sx, j2 = gs.trxpos.dx; j < maxX; j += 8, j2 += 8)
{
/* NOTE: the 2 conseq 4bit values are in NOT in the same byte */
u32 read = gsp(j%2048, i%2048, gs.srcbuf.bw);
u32 write = gdp(j2%2048, i2%2048, gs.dstbuf.bw);
pDstBuf[write] = (pDstBuf[write]&0xf0)|(pSrcBuf[read]&0x0f);
read = gsp((j+1)%2048, i%2048, gs.srcbuf.bw);
write = gdp((j2+1)%2048, i2%2048, gs.dstbuf.bw);
pDstBuf[write] = (pDstBuf[write]&0x0f)|(pSrcBuf[read]&0xf0);
read = gsp((j+2)%2048, i%2048, gs.srcbuf.bw);
write = gdp((j2+2)%2048, i2%2048, gs.dstbuf.bw);
pDstBuf[write] = (pDstBuf[write]&0xf0)|(pSrcBuf[read]&0x0f);
read = gsp((j+3)%2048, i%2048, gs.srcbuf.bw);
write = gdp((j2+3)%2048, i2%2048, gs.dstbuf.bw);
pDstBuf[write] = (pDstBuf[write]&0x0f)|(pSrcBuf[read]&0xf0);
read = gsp((j+4)%2048, i%2048, gs.srcbuf.bw);
write = gdp((j2+4)%2048, i2%2048, gs.dstbuf.bw);
pDstBuf[write] = (pDstBuf[write]&0xf0)|(pSrcBuf[read]&0x0f);
read = gsp((j+5)%2048, i%2048, gs.srcbuf.bw);
write = gdp((j2+5)%2048, i2%2048, gs.dstbuf.bw);
pDstBuf[write] = (pDstBuf[write]&0x0f)|(pSrcBuf[read]&0xf0);
read = gsp((j+6)%2048, i%2048, gs.srcbuf.bw);
write = gdp((j2+6)%2048, i2%2048, gs.dstbuf.bw);
pDstBuf[write] = (pDstBuf[write]&0xf0)|(pSrcBuf[read]&0x0f);
read = gsp((j+7)%2048, i%2048, gs.srcbuf.bw);
write = gdp((j2+7)%2048, i2%2048, gs.dstbuf.bw);
pDstBuf[write] = (pDstBuf[write]&0x0f)|(pSrcBuf[read]&0xf0);
}
}
}
// dir depends on trxpos.dirx & trxpos.diry
void TransferLocalLocal()
{
FUNCLOG
//ZZLog::Error_Log("I'z in your code, transferring your memory...");
assert(gs.imageTransfer == XFER_LOCAL_TO_LOCAL);
assert(gs.trxpos.sx + gs.imageNew.w < 2048 && gs.trxpos.sy + gs.imageNew.h < 2048);
assert(gs.trxpos.dx + gs.imageNew.w < 2048 && gs.trxpos.dy + gs.imageNew.h < 2048);
assert((gs.srcbuf.psm&0x7) == (gs.dstbuf.psm&0x7));
if (gs.trxpos.sx + gs.imageNew.w > gs.srcbuf.bw)
ZZLog::Debug_Log("Transfer error, src width exceeded.(0x%x > 0x%x)", gs.trxpos.sx + gs.imageNew.w, gs.srcbuf.bw);
if (gs.trxpos.dx + gs.imageNew.w > gs.dstbuf.bw)
ZZLog::Debug_Log("Transfer error, dst width exceeded.(0x%x > 0x%x)", gs.trxpos.dx + gs.imageNew.w, gs.dstbuf.bw);
int srcstart, srcend, dststart, dstend;
GetRectMemAddress(srcstart, srcend, gs.srcbuf.psm, gs.trxpos.sx, gs.trxpos.sy, gs.imageNew, gs.srcbuf.bp, gs.srcbuf.bw);
GetRectMemAddress(dststart, dstend, gs.dstbuf.psm, gs.trxpos.dx, gs.trxpos.dy, gs.imageNew, gs.dstbuf.bp, gs.dstbuf.bw);
// resolve the targs
ResolveInRange(srcstart, srcend);
list<CRenderTarget*> listTargs;
s_RTs.GetTargs(dststart, dstend, listTargs);
for (list<CRenderTarget*>::iterator it = listTargs.begin(); it != listTargs.end(); ++it)
{
if (!((*it)->status & CRenderTarget::TS_Virtual))
{
(*it)->Resolve();
//(*it)->status |= CRenderTarget::TS_NeedUpdate;
}
}
if (PSMT_BITMODE(gs.srcbuf.psm) != 4)
{
_TransferLocalLocal();
}
else
{
_TransferLocalLocal_4();
}
g_MemTargs.ClearRange(dststart, dstend);
#ifdef ZEROGS_DEVBUILD
if (g_bSaveTrans)
{
tex0Info t;
t.tbp0 = gs.dstbuf.bp;
t.tw = gs.imageNew.w;
t.th = gs.imageNew.h;
t.tbw = gs.dstbuf.bw;
t.psm = gs.dstbuf.psm;
SaveTex(&t, 0);
t.tbp0 = gs.srcbuf.bp;
t.tw = gs.imageNew.w;
t.th = gs.imageNew.h;
t.tbw = gs.srcbuf.bw;
t.psm = gs.srcbuf.psm;
SaveTex(&t, 0);
}
#endif
}