ShuriZma
/
dolphin
mirror of https://github.com/dolphin-emu/dolphin.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Clean up more space/tab mismatches in AudioCommon, Common, and VideoCommon. 

Not planning to touch Core since it's the most actively changed part of the project.
This commit is contained in:
lioncash 2013-03-19 21:51:12 -04:00
parent 0e3d8e2e9f
commit edd9d0e0ef
91 changed files with 2151 additions and 2169 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
Source/Core/AudioCommon/Src/AOSoundStream.cpp
View File

@ -34,7 +34,7 @@ void AOSound::SoundLoop()
format.channels = 2; format.channels = 2;
format.rate = m_mixer->GetSampleRate(); format.rate = m_mixer->GetSampleRate();
format.byte_format = AO_FMT_LITTLE; format.byte_format = AO_FMT_LITTLE;
device = ao_open_live(default_driver, &format, NULL /* no options */); device = ao_open_live(default_driver, &format, NULL /* no options */);
if (!device) if (!device)
{ {
@ -49,7 +49,7 @@ void AOSound::SoundLoop()
while (!threadData) while (!threadData)
{ {
m_mixer->Mix(realtimeBuffer, numBytesToRender >> 2); m_mixer->Mix(realtimeBuffer, numBytesToRender >> 2);
{ {
std::lock_guard<std::mutex> lk(soundCriticalSection); std::lock_guard<std::mutex> lk(soundCriticalSection);
ao_play(device, (char*)realtimeBuffer, numBytesToRender); ao_play(device, (char*)realtimeBuffer, numBytesToRender);
@ -62,7 +62,7 @@ void AOSound::SoundLoop()
bool AOSound::Start() bool AOSound::Start()
{ {
memset(realtimeBuffer, 0, sizeof(realtimeBuffer)); memset(realtimeBuffer, 0, sizeof(realtimeBuffer));
thread = std::thread(std::mem_fun(&AOSound::SoundLoop), this); thread = std::thread(std::mem_fun(&AOSound::SoundLoop), this);
return true; return true;
} }

12
Source/Core/AudioCommon/Src/AOSoundStream.h
View File

@ -45,21 +45,21 @@ public:
AOSound(CMixer *mixer) : SoundStream(mixer) {} AOSound(CMixer *mixer) : SoundStream(mixer) {}
virtual ~AOSound(); virtual ~AOSound();
virtual bool Start(); virtual bool Start();
virtual void SoundLoop(); virtual void SoundLoop();
virtual void Stop(); virtual void Stop();
static bool isValid() { static bool isValid() {
return true; return true;
} }
virtual bool usesMixer() const { virtual bool usesMixer() const {
return true; return true;
} }
virtual void Update(); virtual void Update();
#else #else

32
Source/Core/AudioCommon/Src/AlsaSoundStream.cpp
View File

@ -88,7 +88,7 @@ bool AlsaSound::AlsaInit()
snd_pcm_hw_params_t *hwparams; snd_pcm_hw_params_t *hwparams;
snd_pcm_uframes_t buffer_size,buffer_size_max; snd_pcm_uframes_t buffer_size,buffer_size_max;
unsigned int periods; unsigned int periods;
err = snd_pcm_open(&handle, "default", SND_PCM_STREAM_PLAYBACK, 0); err = snd_pcm_open(&handle, "default", SND_PCM_STREAM_PLAYBACK, 0);
if (err < 0) if (err < 0)
{ {
@ -97,7 +97,7 @@ bool AlsaSound::AlsaInit()
} }
snd_pcm_hw_params_alloca(&hwparams); snd_pcm_hw_params_alloca(&hwparams);
err = snd_pcm_hw_params_any(handle, hwparams); err = snd_pcm_hw_params_any(handle, hwparams);
if (err < 0) if (err < 0)
{ {
@ -111,8 +111,8 @@ bool AlsaSound::AlsaInit()
ERROR_LOG(AUDIO, "Access type not available: %s\n", snd_strerror(err)); ERROR_LOG(AUDIO, "Access type not available: %s\n", snd_strerror(err));
return false; return false;
} }
err = snd_pcm_hw_params_set_format(handle, hwparams, SND_PCM_FORMAT_S16_LE); err = snd_pcm_hw_params_set_format(handle, hwparams, SND_PCM_FORMAT_S16_LE);
if (err < 0) if (err < 0)
{ {
ERROR_LOG(AUDIO, "Sample format not available: %s\n", snd_strerror(err)); ERROR_LOG(AUDIO, "Sample format not available: %s\n", snd_strerror(err));
@ -126,14 +126,14 @@ bool AlsaSound::AlsaInit()
ERROR_LOG(AUDIO, "Rate not available: %s\n", snd_strerror(err)); ERROR_LOG(AUDIO, "Rate not available: %s\n", snd_strerror(err));
return false; return false;
} }
err = snd_pcm_hw_params_set_channels(handle, hwparams, 2); err = snd_pcm_hw_params_set_channels(handle, hwparams, 2);
if (err < 0) if (err < 0)
{ {
ERROR_LOG(AUDIO, "Channels count not available: %s\n", snd_strerror(err)); ERROR_LOG(AUDIO, "Channels count not available: %s\n", snd_strerror(err));
return false; return false;
} }
periods = BUFFER_SIZE_MAX / FRAME_COUNT_MIN; periods = BUFFER_SIZE_MAX / FRAME_COUNT_MIN;
err = snd_pcm_hw_params_set_periods_max(handle, hwparams, &periods, &dir); err = snd_pcm_hw_params_set_periods_max(handle, hwparams, &periods, &dir);
if (err < 0) if (err < 0)
@ -153,10 +153,10 @@ bool AlsaSound::AlsaInit()
err = snd_pcm_hw_params(handle, hwparams); err = snd_pcm_hw_params(handle, hwparams);
if (err < 0) if (err < 0)
{ {
ERROR_LOG(AUDIO, "Unable to install hw params: %s\n", snd_strerror(err)); ERROR_LOG(AUDIO, "Unable to install hw params: %s\n", snd_strerror(err));
return false; return false;
} }
err = snd_pcm_hw_params_get_buffer_size(hwparams, &buffer_size); err = snd_pcm_hw_params_get_buffer_size(hwparams, &buffer_size);
if (err < 0) if (err < 0)
{ {
@ -176,10 +176,10 @@ bool AlsaSound::AlsaInit()
frames_to_deliver = buffer_size / periods; frames_to_deliver = buffer_size / periods;
//limit the minimum size. pulseaudio advertises a minimum of 32 samples. //limit the minimum size. pulseaudio advertises a minimum of 32 samples.
if (frames_to_deliver < FRAME_COUNT_MIN) if (frames_to_deliver < FRAME_COUNT_MIN)
frames_to_deliver = FRAME_COUNT_MIN; frames_to_deliver = FRAME_COUNT_MIN;
//it is probably a bad idea to try to send more than one buffer of data //it is probably a bad idea to try to send more than one buffer of data
if ((unsigned int)frames_to_deliver > buffer_size) if ((unsigned int)frames_to_deliver > buffer_size)
frames_to_deliver = buffer_size; frames_to_deliver = buffer_size;
NOTICE_LOG(AUDIO, "ALSA gave us a %ld sample \"hardware\" buffer with %d periods. Will send %d samples per fragments.\n", buffer_size, periods, frames_to_deliver); NOTICE_LOG(AUDIO, "ALSA gave us a %ld sample \"hardware\" buffer with %d periods. Will send %d samples per fragments.\n", buffer_size, periods, frames_to_deliver);
snd_pcm_sw_params_alloca(&swparams); snd_pcm_sw_params_alloca(&swparams);
@ -187,21 +187,21 @@ bool AlsaSound::AlsaInit()
err = snd_pcm_sw_params_current(handle, swparams); err = snd_pcm_sw_params_current(handle, swparams);
if (err < 0) if (err < 0)
{ {
ERROR_LOG(AUDIO, "cannot init sw params: %s\n", snd_strerror(err)); ERROR_LOG(AUDIO, "cannot init sw params: %s\n", snd_strerror(err));
return false; return false;
} }
err = snd_pcm_sw_params_set_start_threshold(handle, swparams, 0U); err = snd_pcm_sw_params_set_start_threshold(handle, swparams, 0U);
if (err < 0) if (err < 0)
{ {
ERROR_LOG(AUDIO, "cannot set start thresh: %s\n", snd_strerror(err)); ERROR_LOG(AUDIO, "cannot set start thresh: %s\n", snd_strerror(err));
return false; return false;
} }
err = snd_pcm_sw_params(handle, swparams); err = snd_pcm_sw_params(handle, swparams);
if (err < 0) if (err < 0)
{ {
ERROR_LOG(AUDIO, "cannot set sw params: %s\n", snd_strerror(err)); ERROR_LOG(AUDIO, "cannot set sw params: %s\n", snd_strerror(err));
return false; return false;
} }

6
Source/Core/AudioCommon/Src/CoreAudioSoundStream.h
View File

@ -30,19 +30,19 @@ class CoreAudioSound : public SoundStream
public: public:
CoreAudioSound(CMixer *mixer); CoreAudioSound(CMixer *mixer);
virtual ~CoreAudioSound(); virtual ~CoreAudioSound();
virtual bool Start(); virtual bool Start();
virtual void SetVolume(int volume); virtual void SetVolume(int volume);
virtual void SoundLoop(); virtual void SoundLoop();
virtual void Stop(); virtual void Stop();
static bool isValid() { static bool isValid() {
return true; return true;
} }
virtual bool usesMixer() const { virtual bool usesMixer() const {
return true; return true;
} }
virtual void Update(); virtual void Update();
private: private:

52
Source/Core/AudioCommon/Src/DPL2Decoder.cpp
View File

@ -130,15 +130,15 @@ returns 0 if OK, -1 if fail
*/ */
float* design_fir(unsigned int *n, float* fc, float opt) float* design_fir(unsigned int *n, float* fc, float opt)
{ {
unsigned int o = *n & 1; // Indicator for odd filter length unsigned int o = *n & 1; // Indicator for odd filter length
unsigned int end = ((*n + 1) >> 1) - o; // Loop end unsigned int end = ((*n + 1) >> 1) - o; // Loop end
unsigned int i; // Loop index unsigned int i; // Loop index
float k1 = 2 * float(M_PI); // 2*pi*fc1 float k1 = 2 * float(M_PI); // 2*pi*fc1
float k2 = 0.5f * (float)(1 - o);// Constant used if the filter has even length float k2 = 0.5f * (float)(1 - o); // Constant used if the filter has even length
float g = 0.0f; // Gain float g = 0.0f; // Gain
float t1; // Temporary variables float t1; // Temporary variables
float fc1; // Cutoff frequencies float fc1; // Cutoff frequencies
// Sanity check // Sanity check
if(*n==0) return NULL; if(*n==0) return NULL;
@ -241,18 +241,18 @@ void matrix_decode(const float *in, const int k, const int il,
float *_rr, float *_cf) float *_rr, float *_cf)
{ {
static const float M9_03DB = 0.3535533906f; static const float M9_03DB = 0.3535533906f;
static const float MATAGCTRIG = 8.0f; /* (Fuzzy) AGC trigger */ static const float MATAGCTRIG = 8.0f; /* (Fuzzy) AGC trigger */
static const float MATAGCDECAY = 1.0f; /* AGC baseline decay rate (1/samp.) */ static const float MATAGCDECAY = 1.0f; /* AGC baseline decay rate (1/samp.) */
static const float MATCOMPGAIN = 0.37f; /* Cross talk compensation gain, 0.50 - 0.55 is full cancellation. */ static const float MATCOMPGAIN = 0.37f; /* Cross talk compensation gain, 0.50 - 0.55 is full cancellation. */
const int kr = (k + olddelay) % _dlbuflen; const int kr = (k + olddelay) % _dlbuflen;
float l_gain = (_l_fwr + _r_fwr) / (1 + _l_fwr + _l_fwr); float l_gain = (_l_fwr + _r_fwr) / (1 + _l_fwr + _l_fwr);
float r_gain = (_l_fwr + _r_fwr) / (1 + _r_fwr + _r_fwr); float r_gain = (_l_fwr + _r_fwr) / (1 + _r_fwr + _r_fwr);
/* The 2nd axis has strong gain fluctuations, and therefore require // The 2nd axis has strong gain fluctuations, and therefore require
limits. The factor corresponds to the 1 / amplification of (Lt // limits. The factor corresponds to the 1 / amplification of (Lt
- Rt) when (Lt, Rt) is strongly correlated. (e.g. during // - Rt) when (Lt, Rt) is strongly correlated. (e.g. during
dialogues). It should be bigger than -12 dB to prevent // dialogues). It should be bigger than -12 dB to prevent
distortion. */ // distortion.
float lmr_lim_fwr = _lmr_fwr > M9_03DB * _lpr_fwr ? _lmr_fwr : M9_03DB * _lpr_fwr; float lmr_lim_fwr = _lmr_fwr > M9_03DB * _lpr_fwr ? _lmr_fwr : M9_03DB * _lpr_fwr;
float lpr_gain = (_lpr_fwr + lmr_lim_fwr) / (1 + _lpr_fwr + _lpr_fwr); float lpr_gain = (_lpr_fwr + lmr_lim_fwr) / (1 + _lpr_fwr + _lpr_fwr);
float lmr_gain = (_lpr_fwr + lmr_lim_fwr) / (1 + lmr_lim_fwr + lmr_lim_fwr); float lmr_gain = (_lpr_fwr + lmr_lim_fwr) / (1 + lmr_lim_fwr + lmr_lim_fwr);
@ -275,9 +275,9 @@ void matrix_decode(const float *in, const int k, const int il,
if (decode_rear) if (decode_rear)
{ {
_lr[kr] = _rr[kr] = (l_agc - r_agc) * (float)M_SQRT1_2; _lr[kr] = _rr[kr] = (l_agc - r_agc) * (float)M_SQRT1_2;
/* Stereo rear channel is steered with the same AGC steering as // Stereo rear channel is steered with the same AGC steering as
the decoding matrix. Note this requires a fast updating AGC // the decoding matrix. Note this requires a fast updating AGC
at the order of 20 ms (which is the case here). */ // at the order of 20 ms (which is the case here).
_lr[kr] *= (_l_fwr + _l_fwr) / (1 + _l_fwr + _r_fwr); _lr[kr] *= (_l_fwr + _l_fwr) / (1 + _l_fwr + _r_fwr);
_rr[kr] *= (_r_fwr + _r_fwr) / (1 + _l_fwr + _r_fwr); _rr[kr] *= (_r_fwr + _r_fwr) / (1 + _l_fwr + _r_fwr);
} }
@ -298,16 +298,16 @@ void matrix_decode(const float *in, const int k, const int il,
_rf[k] = (lpr_agc - lmr_agc) * (float)M_SQRT1_2; _rf[k] = (lpr_agc - lmr_agc) * (float)M_SQRT1_2;
/*** CENTER FRONT CANCELLATION ***/ /*** CENTER FRONT CANCELLATION ***/
/* A heuristic approach exploits that Lt + Rt gain contains the // A heuristic approach exploits that Lt + Rt gain contains the
information about Lt, Rt correlation. This effectively reshapes // information about Lt, Rt correlation. This effectively reshapes
the front and rear "cones" to concentrate Lt + Rt to C and // the front and rear "cones" to concentrate Lt + Rt to C and
introduce Lt - Rt in L, R. */ // introduce Lt - Rt in L, R.
/* 0.67677 is the empirical lower bound for lpr_gain. */ /* 0.67677 is the empirical lower bound for lpr_gain. */
c_gain = 8 * (*_adapt_lpr_gain - 0.67677f); c_gain = 8 * (*_adapt_lpr_gain - 0.67677f);
c_gain = c_gain > 0 ? c_gain : 0; c_gain = c_gain > 0 ? c_gain : 0;
/* c_gain should not be too high, not even reaching full // c_gain should not be too high, not even reaching full
cancellation (~ 0.50 - 0.55 at current AGC implementation), or // cancellation (~ 0.50 - 0.55 at current AGC implementation), or
the center will sound too narrow. */ // the center will sound too narrow. */
c_gain = MATCOMPGAIN / (1 + c_gain * c_gain); c_gain = MATCOMPGAIN / (1 + c_gain * c_gain);
c_agc_cfk = c_gain * _cf[k]; c_agc_cfk = c_gain * _cf[k];
_lf[k] -= c_agc_cfk; _lf[k] -= c_agc_cfk;
@ -317,7 +317,7 @@ void matrix_decode(const float *in, const int k, const int il,
void dpl2decode(float *samples, int numsamples, float *out) void dpl2decode(float *samples, int numsamples, float *out)
{ {
static const unsigned int FWRDURATION = 240; /* FWR average duration (samples) */ static const unsigned int FWRDURATION = 240; // FWR average duration (samples)
static const int cfg_delay = 0; static const int cfg_delay = 0;
static const unsigned int fmt_freq = 48000; static const unsigned int fmt_freq = 48000;
static const unsigned int fmt_nchannels = 2; // input channels static const unsigned int fmt_nchannels = 2; // input channels

2
Source/Core/AudioCommon/Src/DSoundStream.cpp
View File

@ -122,7 +122,7 @@ void DSound::SoundLoop()
bool DSound::Start() bool DSound::Start()
{ {
if (FAILED(DirectSoundCreate8(0, &ds, 0))) if (FAILED(DirectSoundCreate8(0, &ds, 0)))
return false; return false;
if (hWnd) if (hWnd)
{ {
HRESULT hr = ds->SetCooperativeLevel((HWND)hWnd, DSSCL_PRIORITY); HRESULT hr = ds->SetCooperativeLevel((HWND)hWnd, DSSCL_PRIORITY);

50
Source/Core/AudioCommon/Src/DSoundStream.h
View File

@ -31,33 +31,33 @@
class DSound : public SoundStream class DSound : public SoundStream
{ {
#ifdef _WIN32 #ifdef _WIN32
std::thread thread; std::thread thread;
Common::Event soundSyncEvent; Common::Event soundSyncEvent;
void *hWnd; void *hWnd;
IDirectSound8* ds; IDirectSound8* ds;
IDirectSoundBuffer* dsBuffer; IDirectSoundBuffer* dsBuffer;
int bufferSize; //i bytes int bufferSize; //i bytes
int m_volume; int m_volume;
// playback position // playback position
int currentPos; int currentPos;
int lastPos; int lastPos;
short realtimeBuffer[BUFSIZE / sizeof(short)]; short realtimeBuffer[BUFSIZE / sizeof(short)];
inline int FIX128(int x) inline int FIX128(int x)
{ {
return x & (~127); return x & (~127);
} }
inline int ModBufferSize(int x) inline int ModBufferSize(int x)
{ {
return (x + bufferSize) % bufferSize; return (x + bufferSize) % bufferSize;
} }
bool CreateBuffer(); bool CreateBuffer();
bool WriteDataToBuffer(DWORD dwOffset, char* soundData, DWORD dwSoundBytes); bool WriteDataToBuffer(DWORD dwOffset, char* soundData, DWORD dwSoundBytes);
public: public:
DSound(CMixer *mixer, void *_hWnd = NULL) DSound(CMixer *mixer, void *_hWnd = NULL)
@ -70,16 +70,16 @@ public:
, hWnd(_hWnd) , hWnd(_hWnd)
{} {}
virtual ~DSound() {} virtual ~DSound() {}
virtual bool Start(); virtual bool Start();
virtual void SoundLoop(); virtual void SoundLoop();
virtual void SetVolume(int volume); virtual void SetVolume(int volume);
virtual void Stop(); virtual void Stop();
virtual void Clear(bool mute); virtual void Clear(bool mute);
static bool isValid() { return true; } static bool isValid() { return true; }
virtual bool usesMixer() const { return true; } virtual bool usesMixer() const { return true; }
virtual void Update(); virtual void Update();
#else #else
public: public:

11
Source/Core/AudioCommon/Src/Mixer.cpp
View File

@ -67,7 +67,7 @@ unsigned int CMixer::Mix(short* samples, unsigned int numSamples)
{ {
static const __m128i sr_mask = static const __m128i sr_mask =
_mm_set_epi32(0x0C0D0E0FL, 0x08090A0BL, _mm_set_epi32(0x0C0D0E0FL, 0x08090A0BL,
0x04050607L, 0x00010203L); 0x04050607L, 0x00010203L);
for (unsigned int i = 0; i < numLeft * 2; i += 8) for (unsigned int i = 0; i < numLeft * 2; i += 8)
{ {
@ -99,12 +99,11 @@ unsigned int CMixer::Mix(short* samples, unsigned int numSamples)
if ((m_indexR2 & INDEX_MASK) == (m_indexW & INDEX_MASK)) //..if it exists if ((m_indexR2 & INDEX_MASK) == (m_indexW & INDEX_MASK)) //..if it exists
m_indexR2 = m_indexR; m_indexR2 = m_indexR;
s16 l1 = Common::swap16(m_buffer[m_indexR & INDEX_MASK]); //current s16 l1 = Common::swap16(m_buffer[m_indexR & INDEX_MASK]); //current
s16 l2 = Common::swap16(m_buffer[m_indexR2 & INDEX_MASK]); //next s16 l2 = Common::swap16(m_buffer[m_indexR2 & INDEX_MASK]); //next
int sampleL = ((l1 << 16) + (l2 - l1) * (u16)frac) >> 16; int sampleL = ((l1 << 16) + (l2 - l1) * (u16)frac) >> 16;
samples[i+1] = sampleL; samples[i+1] = sampleL;
s16 r1 = Common::swap16(m_buffer[(m_indexR + 1) & INDEX_MASK]); //current s16 r1 = Common::swap16(m_buffer[(m_indexR + 1) & INDEX_MASK]); //current
s16 r2 = Common::swap16(m_buffer[(m_indexR2 + 1) & INDEX_MASK]); //next s16 r2 = Common::swap16(m_buffer[(m_indexR2 + 1) & INDEX_MASK]); //next
int sampleR = ((r1 << 16) + (r2 - r1) * (u16)frac) >> 16; int sampleR = ((r1 << 16) + (r2 - r1) * (u16)frac) >> 16;
@ -116,8 +115,6 @@ unsigned int CMixer::Mix(short* samples, unsigned int numSamples)
} }
} }
} else { } else {
numLeft = 0; numLeft = 0;
} }

18
Source/Core/AudioCommon/Src/NullSoundStream.h
View File

@ -25,24 +25,24 @@
class NullSound : public SoundStream class NullSound : public SoundStream
{ {
// playback position // playback position
short realtimeBuffer[BUF_SIZE / sizeof(short)]; short realtimeBuffer[BUF_SIZE / sizeof(short)];
public: public:
NullSound(CMixer *mixer, void *hWnd = NULL) NullSound(CMixer *mixer, void *hWnd = NULL)
: SoundStream(mixer) : SoundStream(mixer)
{} {}
virtual ~NullSound() {} virtual ~NullSound() {}
virtual bool Start(); virtual bool Start();
virtual void SoundLoop(); virtual void SoundLoop();
virtual void SetVolume(int volume); virtual void SetVolume(int volume);
virtual void Stop(); virtual void Stop();
virtual void Clear(bool mute); virtual void Clear(bool mute);
static bool isValid() { return true; } static bool isValid() { return true; }
virtual bool usesMixer() const { return true; } virtual bool usesMixer() const { return true; }
virtual void Update(); virtual void Update();
}; };
#endif //_NULLSOUNDSTREAM_H_ #endif //_NULLSOUNDSTREAM_H_

2
Source/Core/AudioCommon/Src/OpenALStream.cpp
View File

@ -295,7 +295,6 @@ void OpenALStream::SoundLoop()
{ {
ERROR_LOG(AUDIO, "Error occurred while buffering float32 data: %08x", err); ERROR_LOG(AUDIO, "Error occurred while buffering float32 data: %08x", err);
} }
} }
#endif #endif
if (!float32_capable) if (!float32_capable)
@ -308,7 +307,6 @@ void OpenALStream::SoundLoop()
stereo[i * 2 + 1] = (short)((float)sampleBuffer[i * 2 + 1] * (1 << 16)); stereo[i * 2 + 1] = (short)((float)sampleBuffer[i * 2 + 1] * (1 << 16));
} }
alBufferData(uiBufferTemp[iBuffersFilled], AL_FORMAT_STEREO16, stereo, nSamples * 2 * 2, ulFrequency); alBufferData(uiBufferTemp[iBuffersFilled], AL_FORMAT_STEREO16, stereo, nSamples * 2 * 2, ulFrequency);
} }
} }

10
Source/Core/AudioCommon/Src/SoundStream.h
View File

@ -27,17 +27,17 @@ class SoundStream
protected: protected:
CMixer *m_mixer; CMixer *m_mixer;
// We set this to shut down the sound thread. // We set this to shut down the sound thread.
// 0=keep playing, 1=stop playing NOW. // 0=keep playing, 1=stop playing NOW.
volatile int threadData; volatile int threadData;
bool m_logAudio; bool m_logAudio;
WaveFileWriter g_wave_writer; WaveFileWriter g_wave_writer;
bool m_muted; bool m_muted;
public: public:
SoundStream(CMixer *mixer) : m_mixer(mixer), threadData(0), m_logAudio(false), m_muted(false) {} SoundStream(CMixer *mixer) : m_mixer(mixer), threadData(0), m_logAudio(false), m_muted(false) {}
virtual ~SoundStream() { delete m_mixer;} virtual ~SoundStream() { delete m_mixer;}
static bool isValid() { return false; } static bool isValid() { return false; }
virtual CMixer *GetMixer() const { return m_mixer; } virtual CMixer *GetMixer() const { return m_mixer; }
virtual bool Start() { return false; } virtual bool Start() { return false; }

10
Source/Core/AudioCommon/Src/WaveFile.cpp
View File

@ -104,21 +104,21 @@ void WaveFileWriter::AddStereoSamples(const short *sample_data, u32 count)
{ {
if (!file) if (!file)
PanicAlertT("WaveFileWriter - file not open."); PanicAlertT("WaveFileWriter - file not open.");
if (skip_silence) if (skip_silence)
{ {
bool all_zero = true; bool all_zero = true;
for (u32 i = 0; i < count * 2; i++) for (u32 i = 0; i < count * 2; i++)
{ {
if (sample_data[i]) if (sample_data[i])
all_zero = false; all_zero = false;
} }
if (all_zero) if (all_zero)
return; return;
} }
file.WriteBytes(sample_data, count * 4); file.WriteBytes(sample_data, count * 4);
audio_size += count * 4; audio_size += count * 4;
} }
@ -134,7 +134,7 @@ void WaveFileWriter::AddStereoSamplesBE(const short *sample_data, u32 count)
if (skip_silence) if (skip_silence)
{ {
bool all_zero = true; bool all_zero = true;
for (u32 i = 0; i < count * 2; i++) for (u32 i = 0; i < count * 2; i++)
{ {
if (sample_data[i]) if (sample_data[i])

6
Source/Core/AudioCommon/Src/XAudio2Stream.cpp
View File

@ -95,10 +95,10 @@ void StreamingVoiceContext::OnBufferEnd(void* context)
if (!m_source_voice || !context) if (!m_source_voice || !context)
return; return;
//m_sound_sync_event->Wait(); // sync //m_sound_sync_event->Wait(); // sync
//m_sound_sync_event->Spin(); // or tight sync //m_sound_sync_event->Spin(); // or tight sync
m_mixer->Mix(static_cast<short*>(context), SAMPLES_PER_BUFFER); m_mixer->Mix(static_cast<short*>(context), SAMPLES_PER_BUFFER);
SubmitBuffer(static_cast<BYTE*>(context)); SubmitBuffer(static_cast<BYTE*>(context));
} }
@ -183,6 +183,6 @@ void XAudio2::Stop()
m_mastering_voice->DestroyVoice(); m_mastering_voice->DestroyVoice();
m_mastering_voice = nullptr; m_mastering_voice = nullptr;
} }
m_xaudio2.reset(); // release interface m_xaudio2.reset(); // release interface
} }

14
Source/Core/Common/Src/ArmCPUDetect.cpp
View File

@ -33,7 +33,7 @@ char *GetCPUString()
auto const fp = file.GetHandle(); auto const fp = file.GetHandle();
if (!fp) if (!fp)
return 0; return 0;
while (fgets(buf, sizeof(buf), fp)) while (fgets(buf, sizeof(buf), fp))
{ {
if (strncmp(buf, marker, sizeof(marker) - 1)) if (strncmp(buf, marker, sizeof(marker) - 1))
@ -74,12 +74,12 @@ int GetCoreCount()
const char marker[] = "processor\t: "; const char marker[] = "processor\t: ";
int cores = 0; int cores = 0;
char buf[1024]; char buf[1024];
File::IOFile file(procfile, "r"); File::IOFile file(procfile, "r");
auto const fp = file.GetHandle(); auto const fp = file.GetHandle();
if (!fp) if (!fp)
return 0; return 0;
while (fgets(buf, sizeof(buf), fp)) while (fgets(buf, sizeof(buf), fp))
{ {
if (strncmp(buf, marker, sizeof(marker) - 1)) if (strncmp(buf, marker, sizeof(marker) - 1))
@ -103,12 +103,12 @@ void CPUInfo::Detect()
HTT = false; HTT = false;
OS64bit = false; OS64bit = false;
CPU64bit = false; CPU64bit = false;
Mode64bit = false; Mode64bit = false;
vendor = VENDOR_ARM; vendor = VENDOR_ARM;
// Get the information about the CPU // Get the information about the CPU
strncpy(cpu_string, GetCPUString(), sizeof(cpu_string)); strncpy(cpu_string, GetCPUString(), sizeof(cpu_string));
num_cores = GetCoreCount(); num_cores = GetCoreCount();
bSwp = CheckCPUFeature("swp"); bSwp = CheckCPUFeature("swp");
bHalf = CheckCPUFeature("half"); bHalf = CheckCPUFeature("half");
bThumb = CheckCPUFeature("thumb"); bThumb = CheckCPUFeature("thumb");
@ -122,7 +122,7 @@ void CPUInfo::Detect()
bVFPv4 = CheckCPUFeature("vfpv4"); bVFPv4 = CheckCPUFeature("vfpv4");
bIDIVa = CheckCPUFeature("idiva"); bIDIVa = CheckCPUFeature("idiva");
bIDIVt = CheckCPUFeature("idivt"); bIDIVt = CheckCPUFeature("idivt");
// On some buggy kernels(Qualcomm) they show that they support VFPv4 but not IDIVa // On some buggy kernels(Qualcomm) they show that they support VFPv4 but not IDIVa
// All VFPv4 CPUs will support IDIVa // All VFPv4 CPUs will support IDIVa
if (bVFPv4) if (bVFPv4)

2
Source/Core/Common/Src/ArmEmitter.cpp
View File

@ -157,7 +157,7 @@ void ARMXEmitter::MOVI2R(ARMReg reg, u32 val, bool optimize)
{ {
Operand2 op2; Operand2 op2;
bool inverse; bool inverse;
if (cpu_info.bArmV7 && !optimize) if (cpu_info.bArmV7 && !optimize)
{ {
// For backpatching on ARMv7 // For backpatching on ARMv7

28
Source/Core/Common/Src/ArmEmitter.h
View File

@ -136,7 +136,7 @@ protected:
u32 Value; u32 Value;
private: private:
OpType Type; OpType Type;
// IMM types // IMM types
u8 Rotation; // Only for u8 values u8 Rotation; // Only for u8 values
@ -154,7 +154,7 @@ public:
{ {
Type = type; Type = type;
Value = imm; Value = imm;
Rotation = 0; Rotation = 0;
} }
Operand2(ARMReg Reg) Operand2(ARMReg Reg)
@ -297,7 +297,7 @@ public:
u32 Imm24() u32 Imm24()
{ {
_assert_msg_(DYNA_REC, (Type == TYPE_IMM), "Imm16 not IMM"); _assert_msg_(DYNA_REC, (Type == TYPE_IMM), "Imm16 not IMM");
return (Value & 0x0FFFFFFF); return (Value & 0x0FFFFFFF);
} }
// NEON and ASIMD specific // NEON and ASIMD specific
u32 Imm8ASIMD() u32 Imm8ASIMD()
@ -336,9 +336,9 @@ struct FixupBranch
struct LiteralPool struct LiteralPool
{ {
s32 loc; s32 loc;
u8* ldr_address; u8* ldr_address;
u32 val; u32 val;
}; };
typedef const u8* JumpTarget; typedef const u8* JumpTarget;
@ -357,7 +357,7 @@ private:
void WriteShiftedDataOp(u32 op, bool SetFlags, ARMReg dest, ARMReg src, ARMReg op2); void WriteShiftedDataOp(u32 op, bool SetFlags, ARMReg dest, ARMReg src, ARMReg op2);
void WriteShiftedDataOp(u32 op, bool SetFlags, ARMReg dest, ARMReg src, Operand2 op2); void WriteShiftedDataOp(u32 op, bool SetFlags, ARMReg dest, ARMReg src, Operand2 op2);
void WriteSignedMultiply(u32 Op, u32 Op2, u32 Op3, ARMReg dest, ARMReg r1, ARMReg r2); void WriteSignedMultiply(u32 Op, u32 Op2, u32 Op3, ARMReg dest, ARMReg r1, ARMReg r2);
u32 EncodeVd(ARMReg Vd); u32 EncodeVd(ARMReg Vd);
u32 EncodeVn(ARMReg Vn); u32 EncodeVn(ARMReg Vn);
u32 EncodeVm(ARMReg Vm); u32 EncodeVm(ARMReg Vm);
@ -407,10 +407,10 @@ public:
// Hint instruction // Hint instruction
void YIELD(); void YIELD();
// Do nothing // Do nothing
void NOP(int count = 1); //nop padding - TODO: fast nop slides, for amd and intel (check their manuals) void NOP(int count = 1); //nop padding - TODO: fast nop slides, for amd and intel (check their manuals)
#ifdef CALL #ifdef CALL
#undef CALL #undef CALL
#endif #endif
@ -422,7 +422,7 @@ public:
FixupBranch BL(); FixupBranch BL();
FixupBranch BL_CC(CCFlags Cond); FixupBranch BL_CC(CCFlags Cond);
void SetJumpTarget(FixupBranch const &branch); void SetJumpTarget(FixupBranch const &branch);
void B (const void *fnptr); void B (const void *fnptr);
void B (ARMReg src); void B (ARMReg src);
void BL(const void *fnptr); void BL(const void *fnptr);
@ -468,7 +468,7 @@ public:
void BICS(ARMReg dest, ARMReg src, Operand2 op2); void BICS(ARMReg dest, ARMReg src, Operand2 op2);
void MVN (ARMReg dest, Operand2 op2); void MVN (ARMReg dest, Operand2 op2);
void MVNS(ARMReg dest, Operand2 op2); void MVNS(ARMReg dest, Operand2 op2);
void MOVW(ARMReg dest, Operand2 op2); void MOVW(ARMReg dest, Operand2 op2);
void MOVT(ARMReg dest, Operand2 op2, bool TopBits = false); void MOVT(ARMReg dest, Operand2 op2, bool TopBits = false);
// UDIV and SDIV are only available on CPUs that have // UDIV and SDIV are only available on CPUs that have
@ -510,7 +510,7 @@ public:
void STMFD(ARMReg dest, bool WriteBack, const int Regnum, ...); void STMFD(ARMReg dest, bool WriteBack, const int Regnum, ...);
void LDMFD(ARMReg dest, bool WriteBack, const int Regnum, ...); void LDMFD(ARMReg dest, bool WriteBack, const int Regnum, ...);
// Exclusive Access operations // Exclusive Access operations
void LDREX(ARMReg dest, ARMReg base); void LDREX(ARMReg dest, ARMReg base);
// result contains the result if the instruction managed to store the value // result contains the result if the instruction managed to store the value
@ -528,7 +528,7 @@ public:
// NEON Only // NEON Only
void VADD(IntegerSize Size, ARMReg Vd, ARMReg Vn, ARMReg Vm); void VADD(IntegerSize Size, ARMReg Vd, ARMReg Vn, ARMReg Vm);
void VSUB(IntegerSize Size, ARMReg Vd, ARMReg Vn, ARMReg Vm); void VSUB(IntegerSize Size, ARMReg Vd, ARMReg Vn, ARMReg Vm);
// VFP Only // VFP Only
void VLDR(ARMReg Dest, ARMReg Base, s16 offset); void VLDR(ARMReg Dest, ARMReg Base, s16 offset);
void VSTR(ARMReg Src, ARMReg Base, s16 offset); void VSTR(ARMReg Src, ARMReg Base, s16 offset);
@ -539,7 +539,7 @@ public:
void VCMPE(ARMReg Vd); void VCMPE(ARMReg Vd);
void VDIV(ARMReg Vd, ARMReg Vn, ARMReg Vm); void VDIV(ARMReg Vd, ARMReg Vn, ARMReg Vm);
void VSQRT(ARMReg Vd, ARMReg Vm); void VSQRT(ARMReg Vd, ARMReg Vm);
// NEON and VFP // NEON and VFP
void VADD(ARMReg Vd, ARMReg Vn, ARMReg Vm); void VADD(ARMReg Vd, ARMReg Vn, ARMReg Vm);
void VSUB(ARMReg Vd, ARMReg Vn, ARMReg Vm); void VSUB(ARMReg Vd, ARMReg Vn, ARMReg Vm);

8
Source/Core/Common/Src/BreakPoints.h
View File

@ -27,9 +27,9 @@ class DebugInterface;
struct TBreakPoint struct TBreakPoint
{ {
u32 iAddress; u32 iAddress;
bool bOn; bool bOn;
bool bTemporary; bool bTemporary;
}; };
struct TMemCheck struct TMemCheck
@ -96,9 +96,9 @@ public:
typedef std::vector<std::string> TMemChecksStr; typedef std::vector<std::string> TMemChecksStr;
TMemChecks m_MemChecks; TMemChecks m_MemChecks;
const TMemChecks& GetMemChecks() { return m_MemChecks; } const TMemChecks& GetMemChecks() { return m_MemChecks; }
TMemChecksStr GetStrings() const; TMemChecksStr GetStrings() const;
void AddFromStrings(const TMemChecksStr& mcs); void AddFromStrings(const TMemChecksStr& mcs);

10
Source/Core/Common/Src/CPUDetect.h
View File

@ -32,13 +32,13 @@ enum CPUVendor
struct CPUInfo struct CPUInfo
{ {
CPUVendor vendor; CPUVendor vendor;
char cpu_string[0x21]; char cpu_string[0x21];
char brand_string[0x41]; char brand_string[0x41];
bool OS64bit; bool OS64bit;
bool CPU64bit; bool CPU64bit;
bool Mode64bit; bool Mode64bit;
bool HTT; bool HTT;
int num_cores; int num_cores;
int logical_cpu_count; int logical_cpu_count;
@ -56,7 +56,7 @@ struct CPUInfo
bool bAES; bool bAES;
bool bLAHFSAHF64; bool bLAHFSAHF64;
bool bLongMode; bool bLongMode;
// ARM specific CPUInfo // ARM specific CPUInfo
bool bSwp; bool bSwp;
bool bHalf; bool bHalf;
@ -72,14 +72,14 @@ struct CPUInfo
bool bIDIVa; bool bIDIVa;
bool bIDIVt; bool bIDIVt;
bool bArmV7; // enable MOVT, MOVW etc bool bArmV7; // enable MOVT, MOVW etc
// ARMv8 specific // ARMv8 specific
bool bFP; bool bFP;
bool bASIMD; bool bASIMD;
// Call Detect() // Call Detect()
explicit CPUInfo(); explicit CPUInfo();
// Turn the cpu info into a string we can show // Turn the cpu info into a string we can show
std::string Summarize(); std::string Summarize();

52
Source/Core/Common/Src/ChunkFile.h
View File

@ -69,7 +69,7 @@ public:
{ {
u32 count = (u32)x.size(); u32 count = (u32)x.size();
Do(count); Do(count);
switch (mode) switch (mode)
{ {
case MODE_READ: case MODE_READ:
@ -81,7 +81,7 @@ public:
x.insert(pair); x.insert(pair);
} }
break; break;
case MODE_WRITE: case MODE_WRITE:
case MODE_MEASURE: case MODE_MEASURE:
case MODE_VERIFY: case MODE_VERIFY:
@ -93,14 +93,14 @@ public:
break; break;
} }
} }
template <typename T> template <typename T>
void DoContainer(T& x) void DoContainer(T& x)
{ {
u32 size = (u32)x.size(); u32 size = (u32)x.size();
Do(size); Do(size);
x.resize(size); x.resize(size);
for (auto itr = x.begin(); itr != x.end(); ++itr) for (auto itr = x.begin(); itr != x.end(); ++itr)
Do(*itr); Do(*itr);
} }
@ -110,32 +110,32 @@ public:
{ {
DoContainer(x); DoContainer(x);
} }
template <typename T> template <typename T>
void Do(std::list<T>& x) void Do(std::list<T>& x)
{ {
DoContainer(x); DoContainer(x);
} }
template <typename T> template <typename T>
void Do(std::deque<T>& x) void Do(std::deque<T>& x)
{ {
DoContainer(x); DoContainer(x);
} }
template <typename T> template <typename T>
void Do(std::basic_string<T>& x) void Do(std::basic_string<T>& x)
{ {
DoContainer(x); DoContainer(x);
} }
template <typename T> template <typename T>
void DoArray(T* x, u32 count) void DoArray(T* x, u32 count)
{ {
for (u32 i = 0; i != count; ++i) for (u32 i = 0; i != count; ++i)
Do(x[i]); Do(x[i]);
} }
template <typename T> template <typename T>
void Do(T& x) void Do(T& x)
{ {
@ -220,7 +220,7 @@ public:
{ {
u32 cookie = arbitraryNumber; u32 cookie = arbitraryNumber;
Do(cookie); Do(cookie);
if (mode == PointerWrap::MODE_READ && cookie != arbitraryNumber) if (mode == PointerWrap::MODE_READ && cookie != arbitraryNumber)
{ {
PanicAlertT("Error: After \"%s\", found %d (0x%X) instead of save marker %d (0x%X). Aborting savestate load...", PanicAlertT("Error: After \"%s\", found %d (0x%X) instead of save marker %d (0x%X). Aborting savestate load...",
@ -228,7 +228,7 @@ public:
mode = PointerWrap::MODE_MEASURE; mode = PointerWrap::MODE_MEASURE;
} }
} }
private: private:
__forceinline void DoByte(u8& x) __forceinline void DoByte(u8& x)
{ {
@ -237,27 +237,27 @@ private:
case MODE_READ: case MODE_READ:
x = **ptr; x = **ptr;
break; break;
case MODE_WRITE: case MODE_WRITE:
**ptr = x; **ptr = x;
break; break;
case MODE_MEASURE: case MODE_MEASURE:
break; break;
case MODE_VERIFY: case MODE_VERIFY:
_dbg_assert_msg_(COMMON, (x == **ptr), _dbg_assert_msg_(COMMON, (x == **ptr),
"Savestate verification failure: %d (0x%X) (at %p) != %d (0x%X) (at %p).\n", "Savestate verification failure: %d (0x%X) (at %p) != %d (0x%X) (at %p).\n",
x, x, &x, **ptr, **ptr, *ptr); x, x, &x, **ptr, **ptr, *ptr);
break; break;
default: default:
break; break;
} }
++(*ptr); ++(*ptr);
} }
void DoVoid(void *data, u32 size) void DoVoid(void *data, u32 size)
{ {
for(u32 i = 0; i != size; ++i) for(u32 i = 0; i != size; ++i)
@ -300,7 +300,7 @@ public:
ERROR_LOG(COMMON,"ChunkReader: Bad header size"); ERROR_LOG(COMMON,"ChunkReader: Bad header size");
return false; return false;
} }
// Check revision // Check revision
if (header.Revision != _Revision) if (header.Revision != _Revision)
{ {
@ -308,7 +308,7 @@ public:
header.Revision, _Revision); header.Revision, _Revision);
return false; return false;
} }
// get size // get size
const u32 sz = (u32)(fileSize - headerSize); const u32 sz = (u32)(fileSize - headerSize);
if (header.ExpectedSize != sz) if (header.ExpectedSize != sz)
@ -317,7 +317,7 @@ public:
sz, header.ExpectedSize); sz, header.ExpectedSize);
return false; return false;
} }
// read the state // read the state
std::vector<u8> buffer(sz); std::vector<u8> buffer(sz);
if (!pFile.ReadArray(&buffer[0], sz)) if (!pFile.ReadArray(&buffer[0], sz))
@ -333,7 +333,7 @@ public:
INFO_LOG(COMMON, "ChunkReader: Done loading %s" , _rFilename.c_str()); INFO_LOG(COMMON, "ChunkReader: Done loading %s" , _rFilename.c_str());
return true; return true;
} }
// Save file template // Save file template
template<class T> template<class T>
static bool Save(const std::string& _rFilename, u32 _Revision, T& _class) static bool Save(const std::string& _rFilename, u32 _Revision, T& _class)
@ -355,12 +355,12 @@ public:
ptr = &buffer[0]; ptr = &buffer[0];
p.SetMode(PointerWrap::MODE_WRITE); p.SetMode(PointerWrap::MODE_WRITE);
_class.DoState(p); _class.DoState(p);
// Create header // Create header
SChunkHeader header; SChunkHeader header;
header.Revision = _Revision; header.Revision = _Revision;
header.ExpectedSize = (u32)sz; header.ExpectedSize = (u32)sz;
// Write to file // Write to file
if (!pFile.WriteArray(&header, 1)) if (!pFile.WriteArray(&header, 1))
{ {
@ -373,11 +373,11 @@ public:
ERROR_LOG(COMMON,"ChunkReader: Failed writing data"); ERROR_LOG(COMMON,"ChunkReader: Failed writing data");
return false; return false;
} }
INFO_LOG(COMMON,"ChunkReader: Done writing %s", _rFilename.c_str()); INFO_LOG(COMMON,"ChunkReader: Done writing %s", _rFilename.c_str());
return true; return true;
} }
private: private:
struct SChunkHeader struct SChunkHeader
{ {

12
Source/Core/Common/Src/CommonFuncs.h
View File

@ -68,9 +68,9 @@ _mm_shuffle_epi8(__m128i a, __m128i mask)
#define ARRAYSIZE(A) (sizeof(A)/sizeof((A)[0])) #define ARRAYSIZE(A) (sizeof(A)/sizeof((A)[0]))
inline u32 _rotl(u32 x, int shift) { inline u32 _rotl(u32 x, int shift) {
shift &= 31; shift &= 31;
if (!shift) return x; if (!shift) return x;
return (x << shift) | (x >> (32 - shift)); return (x << shift) | (x >> (32 - shift));
} }
inline u64 _rotl64(u64 x, unsigned int shift){ inline u64 _rotl64(u64 x, unsigned int shift){
@ -79,9 +79,9 @@ inline u64 _rotl64(u64 x, unsigned int shift){
} }
inline u32 _rotr(u32 x, int shift) { inline u32 _rotr(u32 x, int shift) {
shift &= 31; shift &= 31;
if (!shift) return x; if (!shift) return x;
return (x >> shift) | (x << (32 - shift)); return (x >> shift) | (x << (32 - shift));
} }
inline u64 _rotr64(u64 x, unsigned int shift){ inline u64 _rotr64(u64 x, unsigned int shift){

6
Source/Core/Common/Src/CommonPaths.h
View File

@ -56,7 +56,7 @@
#define SHARED_USER_DIR File::GetBundleDirectory() + \ #define SHARED_USER_DIR File::GetBundleDirectory() + \
DIR_SEP USERDATA_DIR DIR_SEP DIR_SEP USERDATA_DIR DIR_SEP
#elif defined ANDROID #elif defined ANDROID
#define SYSDATA_DIR "/sdcard/dolphin-emu" #define SYSDATA_DIR "/sdcard/dolphin-emu"
#else #else
#ifdef DATA_DIR #ifdef DATA_DIR
#define SYSDATA_DIR DATA_DIR "sys" #define SYSDATA_DIR DATA_DIR "sys"
@ -127,9 +127,9 @@
#define GC_MEMCARDA "MemoryCardA" #define GC_MEMCARDA "MemoryCardA"
#define GC_MEMCARDB "MemoryCardB" #define GC_MEMCARDB "MemoryCardB"
#define WII_STATE "state.dat" #define WII_STATE "state.dat"
#define WII_SETTING "setting.txt" #define WII_SETTING "setting.txt"
#define WII_EUR_SETTING "setting-eur.txt" #define WII_EUR_SETTING "setting-eur.txt"
#define WII_USA_SETTING "setting-usa.txt" #define WII_USA_SETTING "setting-usa.txt"
#define WII_JAP_SETTING "setting-jpn.txt" #define WII_JAP_SETTING "setting-jpn.txt"

8
Source/Core/Common/Src/ConsoleListener.cpp
View File

@ -227,7 +227,7 @@ void ConsoleListener::PixelSpace(int Left, int Top, int Width, int Height, bool
LetterSpace(LBufWidth, LBufHeight); LetterSpace(LBufWidth, LBufHeight);
ClearScreen(true); ClearScreen(true);
coordScreen.Y = 0; coordScreen.Y = 0;
coordScreen.X = 0; coordScreen.X = 0;
DWORD cCharsWritten = 0; DWORD cCharsWritten = 0;
@ -270,7 +270,7 @@ void ConsoleListener::Log(LogTypes::LOG_LEVELS Level, const char *Text)
*/ */
DWORD cCharsWritten; DWORD cCharsWritten;
WORD Color; WORD Color;
switch (Level) switch (Level)
{ {
case NOTICE_LEVEL: // light green case NOTICE_LEVEL: // light green
@ -334,9 +334,9 @@ void ConsoleListener::ClearScreen(bool Cursor)
DWORD cCharsWritten; DWORD cCharsWritten;
CONSOLE_SCREEN_BUFFER_INFO csbi; CONSOLE_SCREEN_BUFFER_INFO csbi;
DWORD dwConSize; DWORD dwConSize;
HANDLE hConsole = GetStdHandle(STD_OUTPUT_HANDLE); HANDLE hConsole = GetStdHandle(STD_OUTPUT_HANDLE);
GetConsoleScreenBufferInfo(hConsole, &csbi); GetConsoleScreenBufferInfo(hConsole, &csbi);
dwConSize = csbi.dwSize.X * csbi.dwSize.Y; dwConSize = csbi.dwSize.X * csbi.dwSize.Y;
// Write space to the entire console // Write space to the entire console

10
Source/Core/Common/Src/ExtendedTrace.h
View File

@ -25,7 +25,7 @@
#pragma comment( lib, "imagehlp.lib" ) #pragma comment( lib, "imagehlp.lib" )
#define EXTENDEDTRACEINITIALIZE( IniSymbolPath ) InitSymInfo( IniSymbolPath ) #define EXTENDEDTRACEINITIALIZE( IniSymbolPath ) InitSymInfo( IniSymbolPath )
#define EXTENDEDTRACEUNINITIALIZE() UninitSymInfo() #define EXTENDEDTRACEUNINITIALIZE() UninitSymInfo()
#define STACKTRACE(file) StackTrace( GetCurrentThread(), "", file) #define STACKTRACE(file) StackTrace( GetCurrentThread(), "", file)
#define STACKTRACE2(file, eip, esp, ebp) StackTrace(GetCurrentThread(), "", file, eip, esp, ebp) #define STACKTRACE2(file, eip, esp, ebp) StackTrace(GetCurrentThread(), "", file, eip, esp, ebp)
// class File; // class File;
@ -43,10 +43,10 @@ extern char g_uefbuf[UEFBUFSIZE];
#else // not WIN32 #else // not WIN32
#define EXTENDEDTRACEINITIALIZE( IniSymbolPath ) ((void)0) #define EXTENDEDTRACEINITIALIZE( IniSymbolPath ) ((void)0)
#define EXTENDEDTRACEUNINITIALIZE() ((void)0) #define EXTENDEDTRACEUNINITIALIZE() ((void)0)
#define STACKTRACE(file) ((void)0) #define STACKTRACE(file) ((void)0)
#define STACKTRACE2(file, eip, esp, ebp) ((void)0) #define STACKTRACE2(file, eip, esp, ebp) ((void)0)
#endif // WIN32 #endif // WIN32

11
Source/Core/Common/Src/FPURoundMode.h
View File

@ -38,11 +38,12 @@ namespace FPURoundMode
void SetSIMDMode(u32 mode); void SetSIMDMode(u32 mode);
/* /*
There are two different flavors of float to int conversion: * There are two different flavors of float to int conversion:
_mm_cvtps_epi32() and _mm_cvttps_epi32(). The first rounds * _mm_cvtps_epi32() and _mm_cvttps_epi32().
according to the MXCSR rounding bits. The second one always *
uses round towards zero. * The first rounds according to the MXCSR rounding bits.
* The second one always uses round towards zero.
*/ */
void SaveSIMDState(); void SaveSIMDState();
void LoadSIMDState(); void LoadSIMDState();

16
Source/Core/Common/Src/FileUtil.cpp
View File

@ -190,7 +190,7 @@ bool CreateFullPath(const std::string &fullPath)
{ {
int panicCounter = 100; int panicCounter = 100;
INFO_LOG(COMMON, "CreateFullPath: path %s", fullPath.c_str()); INFO_LOG(COMMON, "CreateFullPath: path %s", fullPath.c_str());
if (File::Exists(fullPath)) if (File::Exists(fullPath))
{ {
INFO_LOG(COMMON, "CreateFullPath: path exists %s", fullPath.c_str()); INFO_LOG(COMMON, "CreateFullPath: path exists %s", fullPath.c_str());
@ -206,7 +206,7 @@ bool CreateFullPath(const std::string &fullPath)
// we're done, yay! // we're done, yay!
if (position == fullPath.npos) if (position == fullPath.npos)
return true; return true;
// Include the '/' so the first call is CreateDir("/") rather than CreateDir("") // Include the '/' so the first call is CreateDir("/") rather than CreateDir("")
std::string const subPath(fullPath.substr(0, position + 1)); std::string const subPath(fullPath.substr(0, position + 1));
if (!File::IsDirectory(subPath)) if (!File::IsDirectory(subPath))
@ -486,7 +486,7 @@ u32 ScanDirectoryTree(const std::string &directory, FSTEntry& parentEntry)
return foundEntries; return foundEntries;
} }
// Deletes the given directory and anything under it. Returns true on success. // Deletes the given directory and anything under it. Returns true on success.
bool DeleteDirRecursively(const std::string &directory) bool DeleteDirRecursively(const std::string &directory)
{ {
@ -501,7 +501,7 @@ bool DeleteDirRecursively(const std::string &directory)
FindClose(hFind); FindClose(hFind);
return false; return false;
} }
// windows loop // windows loop
do do
{ {
@ -532,7 +532,7 @@ bool DeleteDirRecursively(const std::string &directory)
#ifndef _WIN32 #ifndef _WIN32
closedir(dirp); closedir(dirp);
#endif #endif
return false; return false;
} }
} }
@ -543,7 +543,7 @@ bool DeleteDirRecursively(const std::string &directory)
#ifndef _WIN32 #ifndef _WIN32
closedir(dirp); closedir(dirp);
#endif #endif
return false; return false;
} }
} }
@ -605,7 +605,7 @@ std::string GetCurrentDir()
if (!(dir = __getcwd(NULL, 0))) { if (!(dir = __getcwd(NULL, 0))) {
ERROR_LOG(COMMON, "GetCurrentDirectory failed: %s", ERROR_LOG(COMMON, "GetCurrentDirectory failed: %s",
GetLastErrorMsg()); GetLastErrorMsg());
return NULL; return NULL;
} }
std::string strDir = dir; std::string strDir = dir;
@ -849,7 +849,7 @@ bool IOFile::Seek(s64 off, int origin)
} }
u64 IOFile::Tell() u64 IOFile::Tell()
{ {
if (IsOpen()) if (IsOpen())
return ftello(m_file); return ftello(m_file);
else else

170
Source/Core/Common/Src/Hash.cpp
View File

@ -119,7 +119,7 @@ u32 HashEctor(const u8* ptr, int length)
inline u64 getblock(const u64 * p, int i) inline u64 getblock(const u64 * p, int i)
{ {
return p[i]; return p[i];
} }
//---------- //----------
@ -127,25 +127,25 @@ inline u64 getblock(const u64 * p, int i)
inline void bmix64(u64 & h1, u64 & h2, u64 & k1, u64 & k2, u64 & c1, u64 & c2) inline void bmix64(u64 & h1, u64 & h2, u64 & k1, u64 & k2, u64 & c1, u64 & c2)
{ {
k1 *= c1; k1 *= c1;
k1 = _rotl64(k1,23); k1 = _rotl64(k1,23);
k1 *= c2; k1 *= c2;
h1 ^= k1; h1 ^= k1;
h1 += h2; h1 += h2;
h2 = _rotl64(h2,41); h2 = _rotl64(h2,41);
k2 *= c2; k2 *= c2;
k2 = _rotl64(k2,23); k2 = _rotl64(k2,23);
k2 *= c1; k2 *= c1;
h2 ^= k2; h2 ^= k2;
h2 += h1; h2 += h1;
h1 = h1*3+0x52dce729; h1 = h1*3+0x52dce729;
h2 = h2*3+0x38495ab5; h2 = h2*3+0x38495ab5;
c1 = c1*5+0x7b7d159c; c1 = c1*5+0x7b7d159c;
c2 = c2*5+0x6bce6396; c2 = c2*5+0x6bce6396;
} }
//---------- //----------
@ -153,87 +153,87 @@ inline void bmix64(u64 & h1, u64 & h2, u64 & k1, u64 & k2, u64 & c1, u64 & c2)
inline u64 fmix64(u64 k) inline u64 fmix64(u64 k)
{ {
k ^= k >> 33; k ^= k >> 33;
k *= 0xff51afd7ed558ccd; k *= 0xff51afd7ed558ccd;
k ^= k >> 33; k ^= k >> 33;
k *= 0xc4ceb9fe1a85ec53; k *= 0xc4ceb9fe1a85ec53;
k ^= k >> 33; k ^= k >> 33;
return k; return k;
} }
u64 GetMurmurHash3(const u8 *src, int len, u32 samples) u64 GetMurmurHash3(const u8 *src, int len, u32 samples)
{ {
const u8 * data = (const u8*)src; const u8 * data = (const u8*)src;
const int nblocks = len / 16; const int nblocks = len / 16;
u32 Step = (len / 8); u32 Step = (len / 8);
if(samples == 0) samples = max(Step, 1u); if(samples == 0) samples = max(Step, 1u);
Step = Step / samples; Step = Step / samples;
if(Step < 1) Step = 1; if(Step < 1) Step = 1;
u64 h1 = 0x9368e53c2f6af274; u64 h1 = 0x9368e53c2f6af274;
u64 h2 = 0x586dcd208f7cd3fd; u64 h2 = 0x586dcd208f7cd3fd;
u64 c1 = 0x87c37b91114253d5; u64 c1 = 0x87c37b91114253d5;
u64 c2 = 0x4cf5ad432745937f; u64 c2 = 0x4cf5ad432745937f;
//---------- //----------
// body // body
const u64 * blocks = (const u64 *)(data); const u64 * blocks = (const u64 *)(data);
for(int i = 0; i < nblocks; i+=Step) for(int i = 0; i < nblocks; i+=Step)
{ {
u64 k1 = getblock(blocks,i*2+0); u64 k1 = getblock(blocks,i*2+0);
u64 k2 = getblock(blocks,i*2+1); u64 k2 = getblock(blocks,i*2+1);
bmix64(h1,h2,k1,k2,c1,c2); bmix64(h1,h2,k1,k2,c1,c2);
} }
//---------- //----------
// tail // tail
const u8 * tail = (const u8*)(data + nblocks*16); const u8 * tail = (const u8*)(data + nblocks*16);
u64 k1 = 0; u64 k1 = 0;
u64 k2 = 0; u64 k2 = 0;
switch(len & 15) switch(len & 15)
{ {
case 15: k2 ^= u64(tail[14]) << 48; case 15: k2 ^= u64(tail[14]) << 48;
case 14: k2 ^= u64(tail[13]) << 40; case 14: k2 ^= u64(tail[13]) << 40;
case 13: k2 ^= u64(tail[12]) << 32; case 13: k2 ^= u64(tail[12]) << 32;
case 12: k2 ^= u64(tail[11]) << 24; case 12: k2 ^= u64(tail[11]) << 24;
case 11: k2 ^= u64(tail[10]) << 16; case 11: k2 ^= u64(tail[10]) << 16;
case 10: k2 ^= u64(tail[ 9]) << 8; case 10: k2 ^= u64(tail[ 9]) << 8;
case 9: k2 ^= u64(tail[ 8]) << 0; case 9: k2 ^= u64(tail[ 8]) << 0;
case 8: k1 ^= u64(tail[ 7]) << 56; case 8: k1 ^= u64(tail[ 7]) << 56;
case 7: k1 ^= u64(tail[ 6]) << 48; case 7: k1 ^= u64(tail[ 6]) << 48;
case 6: k1 ^= u64(tail[ 5]) << 40; case 6: k1 ^= u64(tail[ 5]) << 40;
case 5: k1 ^= u64(tail[ 4]) << 32; case 5: k1 ^= u64(tail[ 4]) << 32;
case 4: k1 ^= u64(tail[ 3]) << 24; case 4: k1 ^= u64(tail[ 3]) << 24;
case 3: k1 ^= u64(tail[ 2]) << 16; case 3: k1 ^= u64(tail[ 2]) << 16;
case 2: k1 ^= u64(tail[ 1]) << 8; case 2: k1 ^= u64(tail[ 1]) << 8;
case 1: k1 ^= u64(tail[ 0]) << 0; case 1: k1 ^= u64(tail[ 0]) << 0;
bmix64(h1,h2,k1,k2,c1,c2); bmix64(h1,h2,k1,k2,c1,c2);
}; };
//---------- //----------
// finalization // finalization
h2 ^= len; h2 ^= len;
h1 += h2; h1 += h2;
h2 += h1; h2 += h1;
h1 = fmix64(h1); h1 = fmix64(h1);
h2 = fmix64(h2); h2 = fmix64(h2);
h1 += h2; h1 += h2;
return h1; return h1;
} }
@ -262,11 +262,13 @@ u64 GetCRC32(const u8 *src, int len, u32 samples)
} }
/* NOTE: This hash function is used for custom texture loading/dumping, so /*
it should not be changed, which would require all custom textures to be * NOTE: This hash function is used for custom texture loading/dumping, so
recalculated for their new hash values. If the hashing function is * it should not be changed, which would require all custom textures to be
changed, make sure this one is still used when the legacy parameter is * recalculated for their new hash values. If the hashing function is
true. */ * changed, make sure this one is still used when the legacy parameter is
* true.
*/
u64 GetHashHiresTexture(const u8 *src, int len, u32 samples) u64 GetHashHiresTexture(const u8 *src, int len, u32 samples)
{ {
const u64 m = 0xc6a4a7935bd1e995; const u64 m = 0xc6a4a7935bd1e995;
@ -282,11 +284,11 @@ u64 GetHashHiresTexture(const u8 *src, int len, u32 samples)
{ {
u64 k = data[0]; u64 k = data[0];
data+=Step; data+=Step;
k *= m; k *= m;
k ^= k >> r; k ^= k >> r;
k *= m; k *= m;
h ^= k; h ^= k;
h *= m; h *= m;
} }
const u8 * data2 = (const u8*)end; const u8 * data2 = (const u8*)end;
@ -453,9 +455,11 @@ u64 GetMurmurHash3(const u8* src, int len, u32 samples)
return *((u64 *)&out); return *((u64 *)&out);
} }
/* FIXME: The old 32-bit version of this hash made different hashes than the /*
64-bit version. Until someone can make a new version of the 32-bit one that * FIXME: The old 32-bit version of this hash made different hashes than the
makes identical hashes, this is just a c/p of the 64-bit one. */ * 64-bit version. Until someone can make a new version of the 32-bit one that
* makes identical hashes, this is just a c/p of the 64-bit one.
*/
u64 GetHashHiresTexture(const u8 *src, int len, u32 samples) u64 GetHashHiresTexture(const u8 *src, int len, u32 samples)
{ {
const u64 m = 0xc6a4a7935bd1e995ULL; const u64 m = 0xc6a4a7935bd1e995ULL;
@ -473,7 +477,7 @@ u64 GetHashHiresTexture(const u8 *src, int len, u32 samples)
data+=Step; data+=Step;
k *= m; k *= m;
k ^= k >> r; k ^= k >> r;
k *= m; k *= m;
h ^= k; h ^= k;
h *= m; h *= m;
} }
@ -502,7 +506,7 @@ u64 GetHashHiresTexture(const u8 *src, int len, u32 samples)
u64 GetHash64(const u8 *src, int len, u32 samples) u64 GetHash64(const u8 *src, int len, u32 samples)
{ {
return ptrHashFunction(src, len, samples); return ptrHashFunction(src, len, samples);
} }
// sets the hash function used for the texture cache // sets the hash function used for the texture cache

20
Source/Core/Common/Src/LogManager.cpp
View File

@ -45,9 +45,9 @@ LogManager::LogManager()
m_Log[LogTypes::MASTER_LOG] = new LogContainer("*", "Master Log"); m_Log[LogTypes::MASTER_LOG] = new LogContainer("*", "Master Log");
m_Log[LogTypes::BOOT] = new LogContainer("BOOT", "Boot"); m_Log[LogTypes::BOOT] = new LogContainer("BOOT", "Boot");
m_Log[LogTypes::COMMON] = new LogContainer("COMMON", "Common"); m_Log[LogTypes::COMMON] = new LogContainer("COMMON", "Common");
m_Log[LogTypes::DISCIO] = new LogContainer("DIO", "Disc IO"); m_Log[LogTypes::DISCIO] = new LogContainer("DIO", "Disc IO");
m_Log[LogTypes::FILEMON] = new LogContainer("FileMon", "File Monitor"); m_Log[LogTypes::FILEMON] = new LogContainer("FileMon", "File Monitor");
m_Log[LogTypes::PAD] = new LogContainer("PAD", "Pad"); m_Log[LogTypes::PAD] = new LogContainer("PAD", "Pad");
m_Log[LogTypes::PIXELENGINE] = new LogContainer("PE", "PixelEngine"); m_Log[LogTypes::PIXELENGINE] = new LogContainer("PE", "PixelEngine");
m_Log[LogTypes::COMMANDPROCESSOR] = new LogContainer("CP", "CommandProc"); m_Log[LogTypes::COMMANDPROCESSOR] = new LogContainer("CP", "CommandProc");
m_Log[LogTypes::VIDEOINTERFACE] = new LogContainer("VI", "VideoInt"); m_Log[LogTypes::VIDEOINTERFACE] = new LogContainer("VI", "VideoInt");
@ -63,15 +63,15 @@ LogManager::LogManager()
m_Log[LogTypes::AUDIO_INTERFACE] = new LogContainer("AI", "AudioInt"); m_Log[LogTypes::AUDIO_INTERFACE] = new LogContainer("AI", "AudioInt");
m_Log[LogTypes::POWERPC] = new LogContainer("PowerPC", "IBM CPU"); m_Log[LogTypes::POWERPC] = new LogContainer("PowerPC", "IBM CPU");
m_Log[LogTypes::OSHLE] = new LogContainer("HLE", "HLE"); m_Log[LogTypes::OSHLE] = new LogContainer("HLE", "HLE");
m_Log[LogTypes::DSPHLE] = new LogContainer("DSPHLE", "DSP HLE"); m_Log[LogTypes::DSPHLE] = new LogContainer("DSPHLE", "DSP HLE");
m_Log[LogTypes::DSPLLE] = new LogContainer("DSPLLE", "DSP LLE"); m_Log[LogTypes::DSPLLE] = new LogContainer("DSPLLE", "DSP LLE");
m_Log[LogTypes::DSP_MAIL] = new LogContainer("DSPMails", "DSP Mails"); m_Log[LogTypes::DSP_MAIL] = new LogContainer("DSPMails", "DSP Mails");
m_Log[LogTypes::VIDEO] = new LogContainer("Video", "Video Backend"); m_Log[LogTypes::VIDEO] = new LogContainer("Video", "Video Backend");
m_Log[LogTypes::AUDIO] = new LogContainer("Audio", "Audio Emulator"); m_Log[LogTypes::AUDIO] = new LogContainer("Audio", "Audio Emulator");
m_Log[LogTypes::DYNA_REC] = new LogContainer("JIT", "Dynamic Recompiler"); m_Log[LogTypes::DYNA_REC] = new LogContainer("JIT", "Dynamic Recompiler");
m_Log[LogTypes::CONSOLE] = new LogContainer("CONSOLE", "Dolphin Console"); m_Log[LogTypes::CONSOLE] = new LogContainer("CONSOLE", "Dolphin Console");
m_Log[LogTypes::OSREPORT] = new LogContainer("OSREPORT", "OSReport"); m_Log[LogTypes::OSREPORT] = new LogContainer("OSREPORT", "OSReport");
m_Log[LogTypes::WIIMOTE] = new LogContainer("Wiimote", "Wiimote"); m_Log[LogTypes::WIIMOTE] = new LogContainer("Wiimote", "Wiimote");
m_Log[LogTypes::WII_IOB] = new LogContainer("WII_IOB", "WII IO Bridge"); m_Log[LogTypes::WII_IOB] = new LogContainer("WII_IOB", "WII IO Bridge");
m_Log[LogTypes::WII_IPC] = new LogContainer("WII_IPC", "WII IPC"); m_Log[LogTypes::WII_IPC] = new LogContainer("WII_IPC", "WII IPC");
m_Log[LogTypes::WII_IPC_HLE] = new LogContainer("WII_IPC_HLE", "WII IPC HLE"); m_Log[LogTypes::WII_IPC_HLE] = new LogContainer("WII_IPC_HLE", "WII IPC HLE");
@ -82,7 +82,7 @@ LogManager::LogManager()
m_Log[LogTypes::WII_IPC_STM] = new LogContainer("WII_IPC_STM", "WII IPC STM"); m_Log[LogTypes::WII_IPC_STM] = new LogContainer("WII_IPC_STM", "WII IPC STM");
m_Log[LogTypes::WII_IPC_NET] = new LogContainer("WII_IPC_NET", "WII IPC NET"); m_Log[LogTypes::WII_IPC_NET] = new LogContainer("WII_IPC_NET", "WII IPC NET");
m_Log[LogTypes::WII_IPC_WIIMOTE] = new LogContainer("WII_IPC_WIIMOTE","WII IPC WIIMOTE"); m_Log[LogTypes::WII_IPC_WIIMOTE] = new LogContainer("WII_IPC_WIIMOTE","WII IPC WIIMOTE");
m_Log[LogTypes::ACTIONREPLAY] = new LogContainer("ActionReplay", "ActionReplay"); m_Log[LogTypes::ACTIONREPLAY] = new LogContainer("ActionReplay", "ActionReplay");
m_Log[LogTypes::MEMCARD_MANAGER] = new LogContainer("MemCard Manager", "MemCard Manager"); m_Log[LogTypes::MEMCARD_MANAGER] = new LogContainer("MemCard Manager", "MemCard Manager");
m_Log[LogTypes::NETPLAY] = new LogContainer("NETPLAY", "Netplay"); m_Log[LogTypes::NETPLAY] = new LogContainer("NETPLAY", "Netplay");

2
Source/Core/Common/Src/LogManager.h
View File

@ -26,7 +26,7 @@
#include <set> #include <set>
#include <string.h> #include <string.h>
#define MAX_MESSAGES 8000 #define MAX_MESSAGES 8000
#define MAX_MSGLEN 1024 #define MAX_MSGLEN 1024

124
Source/Core/Common/Src/MathUtil.cpp
View File

@ -88,7 +88,7 @@ u32 ClassifyFloat(float fvalue)
else else
{ {
// Denormalized number. // Denormalized number.
return sign ? PPC_FPCLASS_ND : PPC_FPCLASS_PD; return sign ? PPC_FPCLASS_ND : PPC_FPCLASS_PD;
} }
} }
else if (exp) else if (exp)
@ -101,26 +101,26 @@ u32 ClassifyFloat(float fvalue)
//Zero //Zero
return sign ? PPC_FPCLASS_NZ : PPC_FPCLASS_PZ; return sign ? PPC_FPCLASS_NZ : PPC_FPCLASS_PZ;
} }
} }
} }
} // namespace } // namespace
inline void MatrixMul(int n, const float *a, const float *b, float *result) inline void MatrixMul(int n, const float *a, const float *b, float *result)
{ {
for (int i = 0; i < n; ++i) for (int i = 0; i < n; ++i)
{ {
for (int j = 0; j < n; ++j) for (int j = 0; j < n; ++j)
{ {
float temp = 0; float temp = 0;
for (int k = 0; k < n; ++k) for (int k = 0; k < n; ++k)
{ {
temp += a[i * n + k] * b[k * n + j]; temp += a[i * n + k] * b[k * n + j];
} }
result[i * n + j] = temp; result[i * n + j] = temp;
} }
} }
} }
// Calculate sum of a float list // Calculate sum of a float list
@ -131,94 +131,94 @@ float MathFloatVectorSum(const std::vector<float>& Vec)
void Matrix33::LoadIdentity(Matrix33 &mtx) void Matrix33::LoadIdentity(Matrix33 &mtx)
{ {
memset(mtx.data, 0, sizeof(mtx.data)); memset(mtx.data, 0, sizeof(mtx.data));
mtx.data[0] = 1.0f; mtx.data[0] = 1.0f;
mtx.data[4] = 1.0f; mtx.data[4] = 1.0f;
mtx.data[8] = 1.0f; mtx.data[8] = 1.0f;
} }
void Matrix33::RotateX(Matrix33 &mtx, float rad) void Matrix33::RotateX(Matrix33 &mtx, float rad)
{ {
float s = sin(rad); float s = sin(rad);
float c = cos(rad); float c = cos(rad);
memset(mtx.data, 0, sizeof(mtx.data)); memset(mtx.data, 0, sizeof(mtx.data));
mtx.data[0] = 1; mtx.data[0] = 1;
mtx.data[4] = c; mtx.data[4] = c;
mtx.data[5] = -s; mtx.data[5] = -s;
mtx.data[7] = s; mtx.data[7] = s;
mtx.data[8] = c; mtx.data[8] = c;
} }
void Matrix33::RotateY(Matrix33 &mtx, float rad) void Matrix33::RotateY(Matrix33 &mtx, float rad)
{ {
float s = sin(rad); float s = sin(rad);
float c = cos(rad); float c = cos(rad);
memset(mtx.data, 0, sizeof(mtx.data)); memset(mtx.data, 0, sizeof(mtx.data));
mtx.data[0] = c; mtx.data[0] = c;
mtx.data[2] = s; mtx.data[2] = s;
mtx.data[4] = 1; mtx.data[4] = 1;
mtx.data[6] = -s; mtx.data[6] = -s;
mtx.data[8] = c; mtx.data[8] = c;
} }
void Matrix33::Multiply(const Matrix33 &a, const Matrix33 &b, Matrix33 &result) void Matrix33::Multiply(const Matrix33 &a, const Matrix33 &b, Matrix33 &result)
{ {
MatrixMul(3, a.data, b.data, result.data); MatrixMul(3, a.data, b.data, result.data);
} }
void Matrix33::Multiply(const Matrix33 &a, const float vec[3], float result[3]) void Matrix33::Multiply(const Matrix33 &a, const float vec[3], float result[3])
{ {
for (int i = 0; i < 3; ++i) { for (int i = 0; i < 3; ++i) {
result[i] = 0; result[i] = 0;
for (int k = 0; k < 3; ++k) { for (int k = 0; k < 3; ++k) {
result[i] += a.data[i * 3 + k] * vec[k]; result[i] += a.data[i * 3 + k] * vec[k];
} }
} }
} }
void Matrix44::LoadIdentity(Matrix44 &mtx) void Matrix44::LoadIdentity(Matrix44 &mtx)
{ {
memset(mtx.data, 0, sizeof(mtx.data)); memset(mtx.data, 0, sizeof(mtx.data));
mtx.data[0] = 1.0f; mtx.data[0] = 1.0f;
mtx.data[5] = 1.0f; mtx.data[5] = 1.0f;
mtx.data[10] = 1.0f; mtx.data[10] = 1.0f;
mtx.data[15] = 1.0f; mtx.data[15] = 1.0f;
} }
void Matrix44::LoadMatrix33(Matrix44 &mtx, const Matrix33 &m33) void Matrix44::LoadMatrix33(Matrix44 &mtx, const Matrix33 &m33)
{ {
for (int i = 0; i < 3; ++i) for (int i = 0; i < 3; ++i)
{ {
for (int j = 0; j < 3; ++j) for (int j = 0; j < 3; ++j)
{ {
mtx.data[i * 4 + j] = m33.data[i * 3 + j]; mtx.data[i * 4 + j] = m33.data[i * 3 + j];
} }
} }
for (int i = 0; i < 3; ++i) for (int i = 0; i < 3; ++i)
{ {
mtx.data[i * 4 + 3] = 0; mtx.data[i * 4 + 3] = 0;
mtx.data[i + 12] = 0; mtx.data[i + 12] = 0;
} }
mtx.data[15] = 1.0f; mtx.data[15] = 1.0f;
} }
void Matrix44::Set(Matrix44 &mtx, const float mtxArray[16]) void Matrix44::Set(Matrix44 &mtx, const float mtxArray[16])
{ {
for(int i = 0; i < 16; ++i) { for(int i = 0; i < 16; ++i) {
mtx.data[i] = mtxArray[i]; mtx.data[i] = mtxArray[i];
} }
} }
void Matrix44::Translate(Matrix44 &mtx, const float vec[3]) void Matrix44::Translate(Matrix44 &mtx, const float vec[3])
{ {
LoadIdentity(mtx); LoadIdentity(mtx);
mtx.data[3] = vec[0]; mtx.data[3] = vec[0];
mtx.data[7] = vec[1]; mtx.data[7] = vec[1];
mtx.data[11] = vec[2]; mtx.data[11] = vec[2];
} }
void Matrix44::Multiply(const Matrix44 &a, const Matrix44 &b, Matrix44 &result) void Matrix44::Multiply(const Matrix44 &a, const Matrix44 &b, Matrix44 &result)
{ {
MatrixMul(4, a.data, b.data, result.data); MatrixMul(4, a.data, b.data, result.data);
} }

34
Source/Core/Common/Src/MathUtil.h
View File

@ -159,12 +159,12 @@ inline u64 Log2(u64 val)
{ {
#if defined(__GNUC__) #if defined(__GNUC__)
return 63 - __builtin_clzll(val); return 63 - __builtin_clzll(val);
#elif defined(_MSC_VER) && defined(_M_X64) #elif defined(_MSC_VER) && defined(_M_X64)
unsigned long result = -1; unsigned long result = -1;
_BitScanReverse64(&result, val); _BitScanReverse64(&result, val);
return result; return result;
#else #else
u64 result = -1; u64 result = -1;
while (val != 0) while (val != 0)
@ -182,32 +182,32 @@ inline u64 Log2(u64 val)
class Matrix33 class Matrix33
{ {
public: public:
static void LoadIdentity(Matrix33 &mtx); static void LoadIdentity(Matrix33 &mtx);
// set mtx to be a rotation matrix around the x axis // set mtx to be a rotation matrix around the x axis
static void RotateX(Matrix33 &mtx, float rad); static void RotateX(Matrix33 &mtx, float rad);
// set mtx to be a rotation matrix around the y axis // set mtx to be a rotation matrix around the y axis
static void RotateY(Matrix33 &mtx, float rad); static void RotateY(Matrix33 &mtx, float rad);
// set result = a x b // set result = a x b
static void Multiply(const Matrix33 &a, const Matrix33 &b, Matrix33 &result); static void Multiply(const Matrix33 &a, const Matrix33 &b, Matrix33 &result);
static void Multiply(const Matrix33 &a, const float vec[3], float result[3]); static void Multiply(const Matrix33 &a, const float vec[3], float result[3]);
float data[9]; float data[9];
}; };
class Matrix44 class Matrix44
{ {
public: public:
static void LoadIdentity(Matrix44 &mtx); static void LoadIdentity(Matrix44 &mtx);
static void LoadMatrix33(Matrix44 &mtx, const Matrix33 &m33); static void LoadMatrix33(Matrix44 &mtx, const Matrix33 &m33);
static void Set(Matrix44 &mtx, const float mtxArray[16]); static void Set(Matrix44 &mtx, const float mtxArray[16]);
static void Translate(Matrix44 &mtx, const float vec[3]); static void Translate(Matrix44 &mtx, const float vec[3]);
static void Multiply(const Matrix44 &a, const Matrix44 &b, Matrix44 &result); static void Multiply(const Matrix44 &a, const Matrix44 &b, Matrix44 &result);
float data[16]; float data[16];
}; };
#endif // _MATH_UTIL_H_ #endif // _MATH_UTIL_H_

4
Source/Core/Common/Src/MemArena.cpp
View File

@ -47,10 +47,10 @@ int AshmemCreateFileMapping(const char *name, size_t size)
fd = open(ASHMEM_DEVICE, O_RDWR); fd = open(ASHMEM_DEVICE, O_RDWR);
if (fd < 0) if (fd < 0)
return fd; return fd;
// We don't really care if we can't set the name, it is optional // We don't really care if we can't set the name, it is optional
ret = ioctl(fd, ASHMEM_SET_NAME, name); ret = ioctl(fd, ASHMEM_SET_NAME, name);
ret = ioctl(fd, ASHMEM_SET_SIZE, size); ret = ioctl(fd, ASHMEM_SET_SIZE, size);
if (ret < 0) if (ret < 0)
{ {

4
Source/Core/Common/Src/MemoryUtil.cpp
View File

@ -155,9 +155,7 @@ void FreeAlignedMemory(void* ptr)
if (ptr) if (ptr)
{ {
#ifdef _WIN32 #ifdef _WIN32
_aligned_free(ptr);
_aligned_free(ptr);
#else #else
free(ptr); free(ptr);
#endif #endif

15
Source/Core/Common/Src/MsgHandler.cpp
View File

@ -102,15 +102,14 @@ bool MsgAlert(bool yes_no, int Style, const char* format, ...)
bool DefaultMsgHandler(const char* caption, const char* text, bool yes_no, int Style) bool DefaultMsgHandler(const char* caption, const char* text, bool yes_no, int Style)
{ {
#ifdef _WIN32 #ifdef _WIN32
int STYLE = MB_ICONINFORMATION; int STYLE = MB_ICONINFORMATION;
if (Style == QUESTION) STYLE = MB_ICONQUESTION; if (Style == QUESTION) STYLE = MB_ICONQUESTION;
if (Style == WARNING) STYLE = MB_ICONWARNING; if (Style == WARNING) STYLE = MB_ICONWARNING;
return IDYES == MessageBox(0, UTF8ToTStr(text).c_str(), UTF8ToTStr(caption).c_str(), STYLE | (yes_no ? MB_YESNO : MB_OK)); return IDYES == MessageBox(0, UTF8ToTStr(text).c_str(), UTF8ToTStr(caption).c_str(), STYLE | (yes_no ? MB_YESNO : MB_OK));
#else #else
printf("%s\n", text); printf("%s\n", text);
return true; return true;
#endif #endif
} }

2
Source/Core/Common/Src/StdConditionVariable.h
View File

@ -133,7 +133,7 @@ public:
// bool wait_for(unique_lock<mutex>& lock, // bool wait_for(unique_lock<mutex>& lock,
// const chrono::duration<Rep, Period>& rel_time, // const chrono::duration<Rep, Period>& rel_time,
// Predicate pred); // Predicate pred);
native_handle_type native_handle() native_handle_type native_handle()
{ {
#ifdef USE_EVENTS #ifdef USE_EVENTS

2
Source/Core/Common/Src/StdThread.h
View File

@ -88,7 +88,7 @@ public:
{ {
return !(*this == rhs); return !(*this == rhs);
} }
bool operator<(const id& rhs) const bool operator<(const id& rhs) const
{ {
return m_thread < rhs.m_thread; return m_thread < rhs.m_thread;

12
Source/Core/Common/Src/StringUtil.cpp
View File

@ -435,10 +435,10 @@ template <typename T>
std::string CodeToUTF8(const char* fromcode, const std::basic_string<T>& input) std::string CodeToUTF8(const char* fromcode, const std::basic_string<T>& input)
{ {
std::string result; std::string result;
#if defined(ANDROID) #if defined(ANDROID)
result = "Not implemented on Android!"; result = "Not implemented on Android!";
#else #else
iconv_t const conv_desc = iconv_open("UTF-8", fromcode); iconv_t const conv_desc = iconv_open("UTF-8", fromcode);
if ((iconv_t)-1 == conv_desc) if ((iconv_t)-1 == conv_desc)
@ -449,7 +449,7 @@ std::string CodeToUTF8(const char* fromcode, const std::basic_string<T>& input)
{ {
size_t const in_bytes = sizeof(T) * input.size(); size_t const in_bytes = sizeof(T) * input.size();
size_t const out_buffer_size = 4 * in_bytes; size_t const out_buffer_size = 4 * in_bytes;
std::string out_buffer; std::string out_buffer;
out_buffer.resize(out_buffer_size); out_buffer.resize(out_buffer_size);
@ -457,12 +457,12 @@ std::string CodeToUTF8(const char* fromcode, const std::basic_string<T>& input)
size_t src_bytes = in_bytes; size_t src_bytes = in_bytes;
auto dst_buffer = &out_buffer[0]; auto dst_buffer = &out_buffer[0];
size_t dst_bytes = out_buffer.size(); size_t dst_bytes = out_buffer.size();
while (src_bytes != 0) while (src_bytes != 0)
{ {
size_t const iconv_result = iconv(conv_desc, (char**)(&src_buffer), &src_bytes, size_t const iconv_result = iconv(conv_desc, (char**)(&src_buffer), &src_bytes,
&dst_buffer, &dst_bytes); &dst_buffer, &dst_bytes);
if ((size_t)-1 == iconv_result) if ((size_t)-1 == iconv_result)
{ {
if (EILSEQ == errno || EINVAL == errno) if (EILSEQ == errno || EINVAL == errno)
@ -512,7 +512,7 @@ std::string UTF16ToUTF8(const std::wstring& input)
// CodeToUTF8("UCS-2LE", input); // CodeToUTF8("UCS-2LE", input);
// CodeToUTF8("UTF-16", input); // CodeToUTF8("UTF-16", input);
CodeToUTF8("UTF-16LE", input); CodeToUTF8("UTF-16LE", input);
// TODO: why is this needed? // TODO: why is this needed?
result.erase(std::remove(result.begin(), result.end(), 0x00), result.end()); result.erase(std::remove(result.begin(), result.end(), 0x00), result.end());
return result; return result;

2
Source/Core/Common/Src/Thread.h
View File

@ -62,7 +62,7 @@ public:
m_condvar.notify_one(); m_condvar.notify_one();
} }
} }
void Wait() void Wait()
{ {
std::unique_lock<std::mutex> lk(m_mutex); std::unique_lock<std::mutex> lk(m_mutex);

96
Source/Core/Common/Src/x64ABI.cpp
View File

@ -226,11 +226,11 @@ void XEmitter::ABI_CallFunction(void *func) {
u64 distance = u64(func) - (u64(code) + 5); u64 distance = u64(func) - (u64(code) + 5);
if (distance >= 0x0000000080000000ULL if (distance >= 0x0000000080000000ULL
&& distance < 0xFFFFFFFF80000000ULL) { && distance < 0xFFFFFFFF80000000ULL) {
// Far call // Far call
MOV(64, R(RAX), Imm64((u64)func)); MOV(64, R(RAX), Imm64((u64)func));
CALLptr(R(RAX)); CALLptr(R(RAX));
} else { } else {
CALL(func); CALL(func);
} }
} }
@ -239,11 +239,11 @@ void XEmitter::ABI_CallFunctionC16(void *func, u16 param1) {
u64 distance = u64(func) - (u64(code) + 5); u64 distance = u64(func) - (u64(code) + 5);
if (distance >= 0x0000000080000000ULL if (distance >= 0x0000000080000000ULL
&& distance < 0xFFFFFFFF80000000ULL) { && distance < 0xFFFFFFFF80000000ULL) {
// Far call // Far call
MOV(64, R(RAX), Imm64((u64)func)); MOV(64, R(RAX), Imm64((u64)func));
CALLptr(R(RAX)); CALLptr(R(RAX));
} else { } else {
CALL(func); CALL(func);
} }
} }
@ -266,11 +266,11 @@ void XEmitter::ABI_CallFunctionC(void *func, u32 param1) {
u64 distance = u64(func) - (u64(code) + 5); u64 distance = u64(func) - (u64(code) + 5);
if (distance >= 0x0000000080000000ULL if (distance >= 0x0000000080000000ULL
&& distance < 0xFFFFFFFF80000000ULL) { && distance < 0xFFFFFFFF80000000ULL) {
// Far call // Far call
MOV(64, R(RAX), Imm64((u64)func)); MOV(64, R(RAX), Imm64((u64)func));
CALLptr(R(RAX)); CALLptr(R(RAX));
} else { } else {
CALL(func); CALL(func);
} }
} }
@ -280,11 +280,11 @@ void XEmitter::ABI_CallFunctionCC(void *func, u32 param1, u32 param2) {
u64 distance = u64(func) - (u64(code) + 5); u64 distance = u64(func) - (u64(code) + 5);
if (distance >= 0x0000000080000000ULL if (distance >= 0x0000000080000000ULL
&& distance < 0xFFFFFFFF80000000ULL) { && distance < 0xFFFFFFFF80000000ULL) {
// Far call // Far call
MOV(64, R(RAX), Imm64((u64)func)); MOV(64, R(RAX), Imm64((u64)func));
CALLptr(R(RAX)); CALLptr(R(RAX));
} else { } else {
CALL(func); CALL(func);
} }
} }
@ -295,11 +295,11 @@ void XEmitter::ABI_CallFunctionCCC(void *func, u32 param1, u32 param2, u32 param
u64 distance = u64(func) - (u64(code) + 5); u64 distance = u64(func) - (u64(code) + 5);
if (distance >= 0x0000000080000000ULL if (distance >= 0x0000000080000000ULL
&& distance < 0xFFFFFFFF80000000ULL) { && distance < 0xFFFFFFFF80000000ULL) {
// Far call // Far call
MOV(64, R(RAX), Imm64((u64)func)); MOV(64, R(RAX), Imm64((u64)func));
CALLptr(R(RAX)); CALLptr(R(RAX));
} else { } else {
CALL(func); CALL(func);
} }
} }
@ -310,11 +310,11 @@ void XEmitter::ABI_CallFunctionCCP(void *func, u32 param1, u32 param2, void *par
u64 distance = u64(func) - (u64(code) + 5); u64 distance = u64(func) - (u64(code) + 5);
if (distance >= 0x0000000080000000ULL if (distance >= 0x0000000080000000ULL
&& distance < 0xFFFFFFFF80000000ULL) { && distance < 0xFFFFFFFF80000000ULL) {
// Far call // Far call
MOV(64, R(RAX), Imm64((u64)func)); MOV(64, R(RAX), Imm64((u64)func));
CALLptr(R(RAX)); CALLptr(R(RAX));
} else { } else {
CALL(func); CALL(func);
} }
} }
@ -326,11 +326,11 @@ void XEmitter::ABI_CallFunctionCCCP(void *func, u32 param1, u32 param2, u32 para
u64 distance = u64(func) - (u64(code) + 5); u64 distance = u64(func) - (u64(code) + 5);
if (distance >= 0x0000000080000000ULL if (distance >= 0x0000000080000000ULL
&& distance < 0xFFFFFFFF80000000ULL) { && distance < 0xFFFFFFFF80000000ULL) {
// Far call // Far call
MOV(64, R(RAX), Imm64((u64)func)); MOV(64, R(RAX), Imm64((u64)func));
CALLptr(R(RAX)); CALLptr(R(RAX));
} else { } else {
CALL(func); CALL(func);
} }
} }
@ -341,11 +341,11 @@ void XEmitter::ABI_CallFunctionPPC(void *func, void *param1, void *param2, u32 p
u64 distance = u64(func) - (u64(code) + 5); u64 distance = u64(func) - (u64(code) + 5);
if (distance >= 0x0000000080000000ULL if (distance >= 0x0000000080000000ULL
&& distance < 0xFFFFFFFF80000000ULL) { && distance < 0xFFFFFFFF80000000ULL) {
// Far call // Far call
MOV(64, R(RAX), Imm64((u64)func)); MOV(64, R(RAX), Imm64((u64)func));
CALLptr(R(RAX)); CALLptr(R(RAX));
} else { } else {
CALL(func); CALL(func);
} }
} }
@ -356,11 +356,11 @@ void XEmitter::ABI_CallFunctionR(void *func, X64Reg reg1) {
u64 distance = u64(func) - (u64(code) + 5); u64 distance = u64(func) - (u64(code) + 5);
if (distance >= 0x0000000080000000ULL if (distance >= 0x0000000080000000ULL
&& distance < 0xFFFFFFFF80000000ULL) { && distance < 0xFFFFFFFF80000000ULL) {
// Far call // Far call
MOV(64, R(RAX), Imm64((u64)func)); MOV(64, R(RAX), Imm64((u64)func));
CALLptr(R(RAX)); CALLptr(R(RAX));
} else { } else {
CALL(func); CALL(func);
} }
} }
@ -380,11 +380,11 @@ void XEmitter::ABI_CallFunctionRR(void *func, X64Reg reg1, X64Reg reg2) {
u64 distance = u64(func) - (u64(code) + 5); u64 distance = u64(func) - (u64(code) + 5);
if (distance >= 0x0000000080000000ULL if (distance >= 0x0000000080000000ULL
&& distance < 0xFFFFFFFF80000000ULL) { && distance < 0xFFFFFFFF80000000ULL) {
// Far call // Far call
MOV(64, R(RAX), Imm64((u64)func)); MOV(64, R(RAX), Imm64((u64)func));
CALLptr(R(RAX)); CALLptr(R(RAX));
} else { } else {
CALL(func); CALL(func);
} }
} }
@ -396,11 +396,11 @@ void XEmitter::ABI_CallFunctionAC(void *func, const Gen::OpArg &arg1, u32 param2
u64 distance = u64(func) - (u64(code) + 5); u64 distance = u64(func) - (u64(code) + 5);
if (distance >= 0x0000000080000000ULL if (distance >= 0x0000000080000000ULL
&& distance < 0xFFFFFFFF80000000ULL) { && distance < 0xFFFFFFFF80000000ULL) {
// Far call // Far call
MOV(64, R(RAX), Imm64((u64)func)); MOV(64, R(RAX), Imm64((u64)func));
CALLptr(R(RAX)); CALLptr(R(RAX));
} else { } else {
CALL(func); CALL(func);
} }
} }
@ -411,11 +411,11 @@ void XEmitter::ABI_CallFunctionA(void *func, const Gen::OpArg &arg1)
u64 distance = u64(func) - (u64(code) + 5); u64 distance = u64(func) - (u64(code) + 5);
if (distance >= 0x0000000080000000ULL if (distance >= 0x0000000080000000ULL
&& distance < 0xFFFFFFFF80000000ULL) { && distance < 0xFFFFFFFF80000000ULL) {
// Far call // Far call
MOV(64, R(RAX), Imm64((u64)func)); MOV(64, R(RAX), Imm64((u64)func));
CALLptr(R(RAX)); CALLptr(R(RAX));
} else { } else {
CALL(func); CALL(func);
} }
} }

42
Source/Core/Common/Src/x64Analyzer.cpp
View File

@ -60,18 +60,18 @@ bool DisassembleMov(const unsigned char *codePtr, InstructionInfo &info, int acc
codeByte = *codePtr++; codeByte = *codePtr++;
// Skip two-byte opcode byte // Skip two-byte opcode byte
bool twoByte = false; bool twoByte = false;
if(codeByte == 0x0F) if(codeByte == 0x0F)
{ {
twoByte = true; twoByte = true;
codeByte2 = *codePtr++; codeByte2 = *codePtr++;
} }
if (!twoByte) if (!twoByte)
{ {
if ((codeByte & 0xF0) == 0x80 || if ((codeByte & 0xF0) == 0x80 ||
((codeByte & 0xF8) == 0xC0 && (codeByte & 0x0E) != 0x02)) ((codeByte & 0xF8) == 0xC0 && (codeByte & 0x0E) != 0x02))
{ {
modRMbyte = *codePtr++; modRMbyte = *codePtr++;
hasModRM = true; hasModRM = true;
@ -79,20 +79,20 @@ bool DisassembleMov(const unsigned char *codePtr, InstructionInfo &info, int acc
} }
else else
{ {
if (((codeByte2 & 0xF0) == 0x00 && (codeByte2 & 0x0F) >= 0x04 && (codeByte2 & 0x0D) != 0x0D) || if (((codeByte2 & 0xF0) == 0x00 && (codeByte2 & 0x0F) >= 0x04 && (codeByte2 & 0x0D) != 0x0D) ||
(codeByte2 & 0xF0) == 0x30 || (codeByte2 & 0xF0) == 0x30 ||
codeByte2 == 0x77 || codeByte2 == 0x77 ||
(codeByte2 & 0xF0) == 0x80 || (codeByte2 & 0xF0) == 0x80 ||
((codeByte2 & 0xF0) == 0xA0 && (codeByte2 & 0x07) <= 0x02) || ((codeByte2 & 0xF0) == 0xA0 && (codeByte2 & 0x07) <= 0x02) ||
(codeByte2 & 0xF8) == 0xC8) (codeByte2 & 0xF8) == 0xC8)
{ {
// No mod R/M byte // No mod R/M byte
} }
else else
{ {
modRMbyte = *codePtr++; modRMbyte = *codePtr++;
hasModRM = true; hasModRM = true;
} }
} }
if (hasModRM) if (hasModRM)
@ -130,7 +130,7 @@ bool DisassembleMov(const unsigned char *codePtr, InstructionInfo &info, int acc
info.displacement = *((s32 *)codePtr); info.displacement = *((s32 *)codePtr);
codePtr += displacementSize; codePtr += displacementSize;
if (accessType == 1) if (accessType == 1)
{ {
info.isMemoryWrite = true; info.isMemoryWrite = true;

2
Source/Core/Common/Src/x64Analyzer.h
View File

@ -51,7 +51,7 @@ enum{
MOVZX_SHORT = 0xB7, //movzx on short MOVZX_SHORT = 0xB7, //movzx on short
MOVSX_BYTE = 0xBE, //movsx on byte MOVSX_BYTE = 0xBE, //movsx on byte
MOVSX_SHORT = 0xBF, //movsx on short MOVSX_SHORT = 0xBF, //movsx on short
MOVE_8BIT = 0xC6, //move 8-bit immediate MOVE_8BIT = 0xC6, //move 8-bit immediate
MOVE_16_32BIT = 0xC7, //move 16 or 32-bit immediate MOVE_16_32BIT = 0xC7, //move 16 or 32-bit immediate
MOVE_REG_TO_MEM = 0x89, //move reg to memory MOVE_REG_TO_MEM = 0x89, //move reg to memory
}; };

2
Source/Core/Common/Src/x64CPUDetect.cpp
View File

@ -39,7 +39,7 @@
#include <machine/cpufunc.h> #include <machine/cpufunc.h>
#else #else
static inline void do_cpuid(unsigned int *eax, unsigned int *ebx, static inline void do_cpuid(unsigned int *eax, unsigned int *ebx,
unsigned int *ecx, unsigned int *edx) unsigned int *ecx, unsigned int *edx)
{ {
#if defined _M_GENERIC #if defined _M_GENERIC
(*eax) = (*ebx) = (*ecx) = (*edx) = 0; (*eax) = (*ebx) = (*ecx) = (*edx) = 0;

216
Source/Core/Common/Src/x64Emitter.cpp
View File

@ -51,13 +51,13 @@ static const NormalOpDef nops[11] =
enum NormalSSEOps enum NormalSSEOps
{ {
sseCMP = 0xC2, sseCMP = 0xC2,
sseADD = 0x58, //ADD sseADD = 0x58, //ADD
sseSUB = 0x5C, //SUB sseSUB = 0x5C, //SUB
sseAND = 0x54, //AND sseAND = 0x54, //AND
sseANDN = 0x55, //ANDN sseANDN = 0x55, //ANDN
sseOR = 0x56, sseOR = 0x56,
sseXOR = 0x57, sseXOR = 0x57,
sseMUL = 0x59, //MUL, sseMUL = 0x59, //MUL,
sseDIV = 0x5E, //DIV sseDIV = 0x5E, //DIV
sseMIN = 0x5D, //MIN sseMIN = 0x5D, //MIN
@ -143,18 +143,18 @@ void OpArg::WriteRex(XEmitter *emit, int opBits, int bits, int customOp) const
if (indexReg & 8) op |= 2; if (indexReg & 8) op |= 2;
if (offsetOrBaseReg & 8) op |= 1; //TODO investigate if this is dangerous if (offsetOrBaseReg & 8) op |= 1; //TODO investigate if this is dangerous
if (op != 0x40 || if (op != 0x40 ||
(bits == 8 && (offsetOrBaseReg & 0x10c) == 4) || (bits == 8 && (offsetOrBaseReg & 0x10c) == 4) ||
(opBits == 8 && (customOp & 0x10c) == 4)) { (opBits == 8 && (customOp & 0x10c) == 4)) {
emit->Write8(op); emit->Write8(op);
_dbg_assert_(DYNA_REC, (offsetOrBaseReg & 0x100) == 0 || bits != 8); _dbg_assert_(DYNA_REC, (offsetOrBaseReg & 0x100) == 0 || bits != 8);
_dbg_assert_(DYNA_REC, (customOp & 0x100) == 0 || opBits != 8); _dbg_assert_(DYNA_REC, (customOp & 0x100) == 0 || opBits != 8);
} else { } else {
_dbg_assert_(DYNA_REC, (offsetOrBaseReg & 0x10c) == 0 || _dbg_assert_(DYNA_REC, (offsetOrBaseReg & 0x10c) == 0 ||
(offsetOrBaseReg & 0x10c) == 0x104 || (offsetOrBaseReg & 0x10c) == 0x104 ||
bits != 8); bits != 8);
_dbg_assert_(DYNA_REC, (customOp & 0x10c) == 0 || _dbg_assert_(DYNA_REC, (customOp & 0x10c) == 0 ||
(customOp & 0x10c) == 0x104 || (customOp & 0x10c) == 0x104 ||
opBits != 8); opBits != 8);
} }
#else #else
@ -969,15 +969,15 @@ void OpArg::WriteNormalOp(XEmitter *emit, bool toRM, NormalOp op, const OpArg &o
immToWrite = 8; immToWrite = 8;
} }
else if ((operand.scale == SCALE_IMM16 && bits == 16) || else if ((operand.scale == SCALE_IMM16 && bits == 16) ||
(operand.scale == SCALE_IMM32 && bits == 32) || (operand.scale == SCALE_IMM32 && bits == 32) ||
(operand.scale == SCALE_IMM32 && bits == 64)) (operand.scale == SCALE_IMM32 && bits == 64))
{ {
emit->Write8(nops[op].imm32); emit->Write8(nops[op].imm32);
immToWrite = bits == 16 ? 16 : 32; immToWrite = bits == 16 ? 16 : 32;
} }
else if ((operand.scale == SCALE_IMM8 && bits == 16) || else if ((operand.scale == SCALE_IMM8 && bits == 16) ||
(operand.scale == SCALE_IMM8 && bits == 32) || (operand.scale == SCALE_IMM8 && bits == 32) ||
(operand.scale == SCALE_IMM8 && bits == 64)) (operand.scale == SCALE_IMM8 && bits == 64))
{ {
emit->Write8(nops[op].simm8); emit->Write8(nops[op].simm8);
immToWrite = 8; immToWrite = 8;
@ -1471,142 +1471,142 @@ void XEmitter::RTDSC() { Write8(0x0F); Write8(0x31); }
// helper routines for setting pointers // helper routines for setting pointers
void XEmitter::CallCdeclFunction3(void* fnptr, u32 arg0, u32 arg1, u32 arg2) void XEmitter::CallCdeclFunction3(void* fnptr, u32 arg0, u32 arg1, u32 arg2)
{ {
using namespace Gen; using namespace Gen;
#ifdef _M_X64 #ifdef _M_X64
#ifdef _MSC_VER #ifdef _MSC_VER
MOV(32, R(RCX), Imm32(arg0)); MOV(32, R(RCX), Imm32(arg0));
MOV(32, R(RDX), Imm32(arg1)); MOV(32, R(RDX), Imm32(arg1));
MOV(32, R(R8), Imm32(arg2)); MOV(32, R(R8), Imm32(arg2));
CALL(fnptr); CALL(fnptr);
#else #else
MOV(32, R(RDI), Imm32(arg0)); MOV(32, R(RDI), Imm32(arg0));
MOV(32, R(RSI), Imm32(arg1)); MOV(32, R(RSI), Imm32(arg1));
MOV(32, R(RDX), Imm32(arg2)); MOV(32, R(RDX), Imm32(arg2));
CALL(fnptr); CALL(fnptr);
#endif #endif
#else #else
ABI_AlignStack(3 * 4); ABI_AlignStack(3 * 4);
PUSH(32, Imm32(arg2)); PUSH(32, Imm32(arg2));
PUSH(32, Imm32(arg1)); PUSH(32, Imm32(arg1));
PUSH(32, Imm32(arg0)); PUSH(32, Imm32(arg0));
CALL(fnptr); CALL(fnptr);
#ifdef _WIN32 #ifdef _WIN32
// don't inc stack // don't inc stack
#else #else
ABI_RestoreStack(3 * 4); ABI_RestoreStack(3 * 4);
#endif #endif
#endif #endif
} }
void XEmitter::CallCdeclFunction4(void* fnptr, u32 arg0, u32 arg1, u32 arg2, u32 arg3) void XEmitter::CallCdeclFunction4(void* fnptr, u32 arg0, u32 arg1, u32 arg2, u32 arg3)
{ {
using namespace Gen; using namespace Gen;
#ifdef _M_X64 #ifdef _M_X64
#ifdef _MSC_VER #ifdef _MSC_VER
MOV(32, R(RCX), Imm32(arg0)); MOV(32, R(RCX), Imm32(arg0));
MOV(32, R(RDX), Imm32(arg1)); MOV(32, R(RDX), Imm32(arg1));
MOV(32, R(R8), Imm32(arg2)); MOV(32, R(R8), Imm32(arg2));
MOV(32, R(R9), Imm32(arg3)); MOV(32, R(R9), Imm32(arg3));
CALL(fnptr); CALL(fnptr);
#else #else
MOV(32, R(RDI), Imm32(arg0)); MOV(32, R(RDI), Imm32(arg0));
MOV(32, R(RSI), Imm32(arg1)); MOV(32, R(RSI), Imm32(arg1));
MOV(32, R(RDX), Imm32(arg2)); MOV(32, R(RDX), Imm32(arg2));
MOV(32, R(RCX), Imm32(arg3)); MOV(32, R(RCX), Imm32(arg3));
CALL(fnptr); CALL(fnptr);
#endif #endif
#else #else
ABI_AlignStack(4 * 4); ABI_AlignStack(4 * 4);
PUSH(32, Imm32(arg3)); PUSH(32, Imm32(arg3));
PUSH(32, Imm32(arg2)); PUSH(32, Imm32(arg2));
PUSH(32, Imm32(arg1)); PUSH(32, Imm32(arg1));
PUSH(32, Imm32(arg0)); PUSH(32, Imm32(arg0));
CALL(fnptr); CALL(fnptr);
#ifdef _WIN32 #ifdef _WIN32
// don't inc stack // don't inc stack
#else #else
ABI_RestoreStack(4 * 4); ABI_RestoreStack(4 * 4);
#endif #endif
#endif #endif
} }
void XEmitter::CallCdeclFunction5(void* fnptr, u32 arg0, u32 arg1, u32 arg2, u32 arg3, u32 arg4) void XEmitter::CallCdeclFunction5(void* fnptr, u32 arg0, u32 arg1, u32 arg2, u32 arg3, u32 arg4)
{ {
using namespace Gen; using namespace Gen;
#ifdef _M_X64 #ifdef _M_X64
#ifdef _MSC_VER #ifdef _MSC_VER
MOV(32, R(RCX), Imm32(arg0)); MOV(32, R(RCX), Imm32(arg0));
MOV(32, R(RDX), Imm32(arg1)); MOV(32, R(RDX), Imm32(arg1));
MOV(32, R(R8), Imm32(arg2)); MOV(32, R(R8), Imm32(arg2));
MOV(32, R(R9), Imm32(arg3)); MOV(32, R(R9), Imm32(arg3));
MOV(32, MDisp(RSP, 0x20), Imm32(arg4)); MOV(32, MDisp(RSP, 0x20), Imm32(arg4));
CALL(fnptr); CALL(fnptr);
#else #else
MOV(32, R(RDI), Imm32(arg0)); MOV(32, R(RDI), Imm32(arg0));
MOV(32, R(RSI), Imm32(arg1)); MOV(32, R(RSI), Imm32(arg1));
MOV(32, R(RDX), Imm32(arg2)); MOV(32, R(RDX), Imm32(arg2));
MOV(32, R(RCX), Imm32(arg3)); MOV(32, R(RCX), Imm32(arg3));
MOV(32, R(R8), Imm32(arg4)); MOV(32, R(R8), Imm32(arg4));
CALL(fnptr); CALL(fnptr);
#endif #endif
#else #else
ABI_AlignStack(5 * 4); ABI_AlignStack(5 * 4);
PUSH(32, Imm32(arg4)); PUSH(32, Imm32(arg4));
PUSH(32, Imm32(arg3)); PUSH(32, Imm32(arg3));
PUSH(32, Imm32(arg2)); PUSH(32, Imm32(arg2));
PUSH(32, Imm32(arg1)); PUSH(32, Imm32(arg1));
PUSH(32, Imm32(arg0)); PUSH(32, Imm32(arg0));
CALL(fnptr); CALL(fnptr);
#ifdef _WIN32 #ifdef _WIN32
// don't inc stack // don't inc stack
#else #else
ABI_RestoreStack(5 * 4); ABI_RestoreStack(5 * 4);
#endif #endif
#endif #endif
} }
void XEmitter::CallCdeclFunction6(void* fnptr, u32 arg0, u32 arg1, u32 arg2, u32 arg3, u32 arg4, u32 arg5) void XEmitter::CallCdeclFunction6(void* fnptr, u32 arg0, u32 arg1, u32 arg2, u32 arg3, u32 arg4, u32 arg5)
{ {
using namespace Gen; using namespace Gen;
#ifdef _M_X64 #ifdef _M_X64
#ifdef _MSC_VER #ifdef _MSC_VER
MOV(32, R(RCX), Imm32(arg0)); MOV(32, R(RCX), Imm32(arg0));
MOV(32, R(RDX), Imm32(arg1)); MOV(32, R(RDX), Imm32(arg1));
MOV(32, R(R8), Imm32(arg2)); MOV(32, R(R8), Imm32(arg2));
MOV(32, R(R9), Imm32(arg3)); MOV(32, R(R9), Imm32(arg3));
MOV(32, MDisp(RSP, 0x20), Imm32(arg4)); MOV(32, MDisp(RSP, 0x20), Imm32(arg4));
MOV(32, MDisp(RSP, 0x28), Imm32(arg5)); MOV(32, MDisp(RSP, 0x28), Imm32(arg5));
CALL(fnptr); CALL(fnptr);
#else #else
MOV(32, R(RDI), Imm32(arg0)); MOV(32, R(RDI), Imm32(arg0));
MOV(32, R(RSI), Imm32(arg1)); MOV(32, R(RSI), Imm32(arg1));
MOV(32, R(RDX), Imm32(arg2)); MOV(32, R(RDX), Imm32(arg2));
MOV(32, R(RCX), Imm32(arg3)); MOV(32, R(RCX), Imm32(arg3));
MOV(32, R(R8), Imm32(arg4)); MOV(32, R(R8), Imm32(arg4));
MOV(32, R(R9), Imm32(arg5)); MOV(32, R(R9), Imm32(arg5));
CALL(fnptr); CALL(fnptr);
#endif #endif
#else #else
ABI_AlignStack(6 * 4); ABI_AlignStack(6 * 4);
PUSH(32, Imm32(arg5)); PUSH(32, Imm32(arg5));
PUSH(32, Imm32(arg4)); PUSH(32, Imm32(arg4));
PUSH(32, Imm32(arg3)); PUSH(32, Imm32(arg3));
PUSH(32, Imm32(arg2)); PUSH(32, Imm32(arg2));
PUSH(32, Imm32(arg1)); PUSH(32, Imm32(arg1));
PUSH(32, Imm32(arg0)); PUSH(32, Imm32(arg0));
CALL(fnptr); CALL(fnptr);
#ifdef _WIN32 #ifdef _WIN32
// don't inc stack // don't inc stack
#else #else
ABI_RestoreStack(6 * 4); ABI_RestoreStack(6 * 4);
#endif #endif
#endif #endif
} }
@ -1615,31 +1615,31 @@ void XEmitter::CallCdeclFunction6(void* fnptr, u32 arg0, u32 arg1, u32 arg2, u32
// See header // See header
void XEmitter::___CallCdeclImport3(void* impptr, u32 arg0, u32 arg1, u32 arg2) { void XEmitter::___CallCdeclImport3(void* impptr, u32 arg0, u32 arg1, u32 arg2) {
MOV(32, R(RCX), Imm32(arg0)); MOV(32, R(RCX), Imm32(arg0));
MOV(32, R(RDX), Imm32(arg1)); MOV(32, R(RDX), Imm32(arg1));
MOV(32, R(R8), Imm32(arg2)); MOV(32, R(R8), Imm32(arg2));
CALLptr(M(impptr)); CALLptr(M(impptr));
} }
void XEmitter::___CallCdeclImport4(void* impptr, u32 arg0, u32 arg1, u32 arg2, u32 arg3) { void XEmitter::___CallCdeclImport4(void* impptr, u32 arg0, u32 arg1, u32 arg2, u32 arg3) {
MOV(32, R(RCX), Imm32(arg0)); MOV(32, R(RCX), Imm32(arg0));
MOV(32, R(RDX), Imm32(arg1)); MOV(32, R(RDX), Imm32(arg1));
MOV(32, R(R8), Imm32(arg2)); MOV(32, R(R8), Imm32(arg2));
MOV(32, R(R9), Imm32(arg3)); MOV(32, R(R9), Imm32(arg3));
CALLptr(M(impptr)); CALLptr(M(impptr));
} }
void XEmitter::___CallCdeclImport5(void* impptr, u32 arg0, u32 arg1, u32 arg2, u32 arg3, u32 arg4) { void XEmitter::___CallCdeclImport5(void* impptr, u32 arg0, u32 arg1, u32 arg2, u32 arg3, u32 arg4) {
MOV(32, R(RCX), Imm32(arg0)); MOV(32, R(RCX), Imm32(arg0));
MOV(32, R(RDX), Imm32(arg1)); MOV(32, R(RDX), Imm32(arg1));
MOV(32, R(R8), Imm32(arg2)); MOV(32, R(R8), Imm32(arg2));
MOV(32, R(R9), Imm32(arg3)); MOV(32, R(R9), Imm32(arg3));
MOV(32, MDisp(RSP, 0x20), Imm32(arg4)); MOV(32, MDisp(RSP, 0x20), Imm32(arg4));
CALLptr(M(impptr)); CALLptr(M(impptr));
} }
void XEmitter::___CallCdeclImport6(void* impptr, u32 arg0, u32 arg1, u32 arg2, u32 arg3, u32 arg4, u32 arg5) { void XEmitter::___CallCdeclImport6(void* impptr, u32 arg0, u32 arg1, u32 arg2, u32 arg3, u32 arg4, u32 arg5) {
MOV(32, R(RCX), Imm32(arg0)); MOV(32, R(RCX), Imm32(arg0));
MOV(32, R(RDX), Imm32(arg1)); MOV(32, R(RDX), Imm32(arg1));
MOV(32, R(R8), Imm32(arg2)); MOV(32, R(R8), Imm32(arg2));
MOV(32, R(R9), Imm32(arg3)); MOV(32, R(R9), Imm32(arg3));
MOV(32, MDisp(RSP, 0x20), Imm32(arg4)); MOV(32, MDisp(RSP, 0x20), Imm32(arg4));
MOV(32, MDisp(RSP, 0x28), Imm32(arg5)); MOV(32, MDisp(RSP, 0x28), Imm32(arg5));
CALLptr(M(impptr)); CALLptr(M(impptr));

2
Source/Core/Common/Src/x64Emitter.h
View File

@ -766,7 +766,7 @@ public:
// Start over if you need to change the code (call FreeCodeSpace(), AllocCodeSpace()). // Start over if you need to change the code (call FreeCodeSpace(), AllocCodeSpace()).
void WriteProtect() void WriteProtect()
{ {
WriteProtectMemory(region, region_size, true); WriteProtectMemory(region, region_size, true);
} }
void ResetCodePtr() void ResetCodePtr()

4
Source/Core/Common/Src/x64FPURoundMode.cpp
View File

@ -68,7 +68,7 @@ namespace FPURoundMode
} }
void SetPrecisionMode(u32 mode) void SetPrecisionMode(u32 mode)
{ {
#ifdef _M_IX86 #ifdef _M_IX86
// sets the floating-point lib to 53-bit // sets the floating-point lib to 53-bit
// PowerPC has a 53bit floating pipeline only // PowerPC has a 53bit floating pipeline only
@ -104,7 +104,7 @@ namespace FPURoundMode
u32 csr = ssetable[mode]; u32 csr = ssetable[mode];
_mm_setcsr(csr); _mm_setcsr(csr);
} }
void SaveSIMDState() void SaveSIMDState()
{ {
saved_sse_state = _mm_getcsr(); saved_sse_state = _mm_getcsr();

860
Source/Core/VideoCommon/Src/BPMemory.h
View File

File diff suppressed because it is too large Load Diff

260
Source/Core/VideoCommon/Src/CPMemory.h
View File

@ -25,18 +25,18 @@ enum
{ {
ARRAY_POSITION = 0, ARRAY_POSITION = 0,
ARRAY_NORMAL = 1, ARRAY_NORMAL = 1,
ARRAY_COLOR = 2, ARRAY_COLOR = 2,
ARRAY_COLOR2 = 3, ARRAY_COLOR2 = 3,
ARRAY_TEXCOORD0 = 4, ARRAY_TEXCOORD0 = 4,
}; };
// Vertex components // Vertex components
enum enum
{ {
NOT_PRESENT = 0, NOT_PRESENT = 0,
DIRECT = 1, DIRECT = 1,
INDEX8 = 2, INDEX8 = 2,
INDEX16 = 3, INDEX16 = 3,
}; };
enum enum
@ -74,14 +74,14 @@ union TVtxDesc
// 0: not present // 0: not present
// 1: present // 1: present
u32 PosMatIdx : 1; u32 PosMatIdx : 1;
u32 Tex0MatIdx : 1; u32 Tex0MatIdx : 1;
u32 Tex1MatIdx : 1; u32 Tex1MatIdx : 1;
u32 Tex2MatIdx : 1; u32 Tex2MatIdx : 1;
u32 Tex3MatIdx : 1; u32 Tex3MatIdx : 1;
u32 Tex4MatIdx : 1; u32 Tex4MatIdx : 1;
u32 Tex5MatIdx : 1; u32 Tex5MatIdx : 1;
u32 Tex6MatIdx : 1; u32 Tex6MatIdx : 1;
u32 Tex7MatIdx : 1; u32 Tex7MatIdx : 1;
// 00: not present // 00: not present
// 01: direct // 01: direct
@ -89,156 +89,156 @@ union TVtxDesc
// 11: 16 bit index // 11: 16 bit index
u32 Position : 2; u32 Position : 2;
u32 Normal : 2; u32 Normal : 2;
u32 Color0 : 2; u32 Color0 : 2;
u32 Color1 : 2; u32 Color1 : 2;
u32 Tex0Coord : 2; u32 Tex0Coord : 2;
u32 Tex1Coord : 2; u32 Tex1Coord : 2;
u32 Tex2Coord : 2; u32 Tex2Coord : 2;
u32 Tex3Coord : 2; u32 Tex3Coord : 2;
u32 Tex4Coord : 2; u32 Tex4Coord : 2;
u32 Tex5Coord : 2; u32 Tex5Coord : 2;
u32 Tex6Coord : 2; u32 Tex6Coord : 2;
u32 Tex7Coord : 2; u32 Tex7Coord : 2;
u32 :31; u32 :31;
}; };
struct { struct {
u32 Hex0, Hex1; u32 Hex0, Hex1;
}; };
}; };
union UVAT_group0 union UVAT_group0
{ {
u32 Hex; u32 Hex;
struct struct
{ {
// 0:8 // 0:8
u32 PosElements : 1; u32 PosElements : 1;
u32 PosFormat : 3; u32 PosFormat : 3;
u32 PosFrac : 5; u32 PosFrac : 5;
// 9:12 // 9:12
u32 NormalElements : 1; u32 NormalElements : 1;
u32 NormalFormat : 3; u32 NormalFormat : 3;
// 13:16 // 13:16
u32 Color0Elements : 1; u32 Color0Elements : 1;
u32 Color0Comp : 3; u32 Color0Comp : 3;
// 17:20 // 17:20
u32 Color1Elements : 1; u32 Color1Elements : 1;
u32 Color1Comp : 3; u32 Color1Comp : 3;
// 21:29 // 21:29
u32 Tex0CoordElements : 1; u32 Tex0CoordElements : 1;
u32 Tex0CoordFormat : 3; u32 Tex0CoordFormat : 3;
u32 Tex0Frac : 5; u32 Tex0Frac : 5;
// 30:31 // 30:31
u32 ByteDequant : 1; u32 ByteDequant : 1;
u32 NormalIndex3 : 1; u32 NormalIndex3 : 1;
}; };
}; };
union UVAT_group1 union UVAT_group1
{ {
u32 Hex; u32 Hex;
struct struct
{ {
// 0:8 // 0:8
u32 Tex1CoordElements : 1; u32 Tex1CoordElements : 1;
u32 Tex1CoordFormat : 3; u32 Tex1CoordFormat : 3;
u32 Tex1Frac : 5; u32 Tex1Frac : 5;
// 9:17 // 9:17
u32 Tex2CoordElements : 1; u32 Tex2CoordElements : 1;
u32 Tex2CoordFormat : 3; u32 Tex2CoordFormat : 3;
u32 Tex2Frac : 5; u32 Tex2Frac : 5;
// 18:26 // 18:26
u32 Tex3CoordElements : 1; u32 Tex3CoordElements : 1;
u32 Tex3CoordFormat : 3; u32 Tex3CoordFormat : 3;
u32 Tex3Frac : 5; u32 Tex3Frac : 5;
// 27:30 // 27:30
u32 Tex4CoordElements : 1; u32 Tex4CoordElements : 1;
u32 Tex4CoordFormat : 3; u32 Tex4CoordFormat : 3;
// //
u32 : 1; u32 : 1;
}; };
}; };
union UVAT_group2 union UVAT_group2
{ {
u32 Hex; u32 Hex;
struct struct
{ {
// 0:4 // 0:4
u32 Tex4Frac : 5; u32 Tex4Frac : 5;
// 5:13 // 5:13
u32 Tex5CoordElements : 1; u32 Tex5CoordElements : 1;
u32 Tex5CoordFormat : 3; u32 Tex5CoordFormat : 3;
u32 Tex5Frac : 5; u32 Tex5Frac : 5;
// 14:22 // 14:22
u32 Tex6CoordElements : 1; u32 Tex6CoordElements : 1;
u32 Tex6CoordFormat : 3; u32 Tex6CoordFormat : 3;
u32 Tex6Frac : 5; u32 Tex6Frac : 5;
// 23:31 // 23:31
u32 Tex7CoordElements : 1; u32 Tex7CoordElements : 1;
u32 Tex7CoordFormat : 3; u32 Tex7CoordFormat : 3;
u32 Tex7Frac : 5; u32 Tex7Frac : 5;
}; };
}; };
struct ColorAttr struct ColorAttr
{ {
u8 Elements; u8 Elements;
u8 Comp; u8 Comp;
}; };
struct TexAttr struct TexAttr
{ {
u8 Elements; u8 Elements;
u8 Format; u8 Format;
u8 Frac; u8 Frac;
}; };
struct TVtxAttr struct TVtxAttr
{ {
u8 PosElements; u8 PosElements;
u8 PosFormat; u8 PosFormat;
u8 PosFrac; u8 PosFrac;
u8 NormalElements; u8 NormalElements;
u8 NormalFormat; u8 NormalFormat;
ColorAttr color[2]; ColorAttr color[2];
TexAttr texCoord[8]; TexAttr texCoord[8];
u8 ByteDequant; u8 ByteDequant;
u8 NormalIndex3; u8 NormalIndex3;
}; };
// Matrix indices // Matrix indices
union TMatrixIndexA union TMatrixIndexA
{ {
struct struct
{ {
u32 PosNormalMtxIdx : 6; u32 PosNormalMtxIdx : 6;
u32 Tex0MtxIdx : 6; u32 Tex0MtxIdx : 6;
u32 Tex1MtxIdx : 6; u32 Tex1MtxIdx : 6;
u32 Tex2MtxIdx : 6; u32 Tex2MtxIdx : 6;
u32 Tex3MtxIdx : 6; u32 Tex3MtxIdx : 6;
}; };
struct struct
{ {
u32 Hex : 30; u32 Hex : 30;
u32 unused : 2; u32 unused : 2;
}; };
}; };
union TMatrixIndexB union TMatrixIndexB
{ {
struct struct
{ {
u32 Tex4MtxIdx : 6; u32 Tex4MtxIdx : 6;
u32 Tex5MtxIdx : 6; u32 Tex5MtxIdx : 6;
u32 Tex6MtxIdx : 6; u32 Tex6MtxIdx : 6;
u32 Tex7MtxIdx : 6; u32 Tex7MtxIdx : 6;
}; };
struct struct
{ {
u32 Hex : 24; u32 Hex : 24;
u32 unused : 8; u32 unused : 8;
}; };
}; };
#pragma pack() #pragma pack()

43
Source/Core/VideoCommon/Src/CommandProcessor.cpp
View File

@ -325,7 +325,7 @@ void Write16(const u16 _Value, const u32 _Address)
case CLEAR_REGISTER: case CLEAR_REGISTER:
{ {
UCPClearReg tmpCtrl(_Value); UCPClearReg tmpCtrl(_Value);
m_CPClearReg.Hex = tmpCtrl.Hex; m_CPClearReg.Hex = tmpCtrl.Hex;
DEBUG_LOG(COMMANDPROCESSOR,"\t write to CLEAR_REGISTER : %04x", _Value); DEBUG_LOG(COMMANDPROCESSOR,"\t write to CLEAR_REGISTER : %04x", _Value);
SetCpClearRegister(); SetCpClearRegister();
} }
@ -414,7 +414,7 @@ void Write16(const u16 _Value, const u32 _Address)
} }
else else
{ {
ResetVideoBuffer(); ResetVideoBuffer();
} }
IncrementCheckContextId(); IncrementCheckContextId();
DEBUG_LOG(COMMANDPROCESSOR,"try to write to FIFO_RW_DISTANCE_HI : %04x", _Value); DEBUG_LOG(COMMANDPROCESSOR,"try to write to FIFO_RW_DISTANCE_HI : %04x", _Value);
@ -462,7 +462,6 @@ void STACKALIGN GatherPipeBursted()
ProcessFifoAllDistance(); ProcessFifoAllDistance();
waitingForPEInterruptDisable = false; waitingForPEInterruptDisable = false;
} }
} }
return; return;
} }
@ -492,11 +491,11 @@ void STACKALIGN GatherPipeBursted()
void UpdateInterrupts(u64 userdata) void UpdateInterrupts(u64 userdata)
{ {
if (userdata) if (userdata)
{ {
interruptSet = true; interruptSet = true;
INFO_LOG(COMMANDPROCESSOR,"Interrupt set"); INFO_LOG(COMMANDPROCESSOR,"Interrupt set");
ProcessorInterface::SetInterrupt(INT_CAUSE_CP, true); ProcessorInterface::SetInterrupt(INT_CAUSE_CP, true);
} }
else else
{ {
@ -504,7 +503,7 @@ void UpdateInterrupts(u64 userdata)
INFO_LOG(COMMANDPROCESSOR,"Interrupt cleared"); INFO_LOG(COMMANDPROCESSOR,"Interrupt cleared");
ProcessorInterface::SetInterrupt(INT_CAUSE_CP, false); ProcessorInterface::SetInterrupt(INT_CAUSE_CP, false);
} }
interruptWaiting = false; interruptWaiting = false;
} }
void UpdateInterruptsFromVideoBackend(u64 userdata) void UpdateInterruptsFromVideoBackend(u64 userdata)
@ -520,11 +519,11 @@ void AbortFrame()
void SetCpStatus(bool isCPUThread) void SetCpStatus(bool isCPUThread)
{ {
// overflow & underflow check // overflow & underflow check
fifo.bFF_HiWatermark = (fifo.CPReadWriteDistance > fifo.CPHiWatermark); fifo.bFF_HiWatermark = (fifo.CPReadWriteDistance > fifo.CPHiWatermark);
fifo.bFF_LoWatermark = (fifo.CPReadWriteDistance < fifo.CPLoWatermark); fifo.bFF_LoWatermark = (fifo.CPReadWriteDistance < fifo.CPLoWatermark);
// breakpoint // breakpoint
if (!isCPUThread) if (!isCPUThread)
{ {
if (fifo.bFF_BPEnable) if (fifo.bFF_BPEnable)
@ -562,12 +561,12 @@ void SetCpStatus(bool isCPUThread)
isHiWatermarkActive = ovfInt && m_CPCtrlReg.GPReadEnable; isHiWatermarkActive = ovfInt && m_CPCtrlReg.GPReadEnable;
isLoWatermarkActive = undfInt && m_CPCtrlReg.GPReadEnable; isLoWatermarkActive = undfInt && m_CPCtrlReg.GPReadEnable;
if (interrupt != interruptSet && !interruptWaiting) if (interrupt != interruptSet && !interruptWaiting)
{ {
u64 userdata = interrupt?1:0; u64 userdata = interrupt?1:0;
if (IsOnThread()) if (IsOnThread())
{ {
if (!interrupt || bpInt || undfInt || ovfInt) if (!interrupt || bpInt || undfInt || ovfInt)
{ {
if (!isCPUThread) if (!isCPUThread)
{ {
@ -579,14 +578,14 @@ void SetCpStatus(bool isCPUThread)
{ {
// CPU thread: // CPU thread:
interruptSet = interrupt; interruptSet = interrupt;
INFO_LOG(COMMANDPROCESSOR,"Interrupt set"); INFO_LOG(COMMANDPROCESSOR,"Interrupt set");
ProcessorInterface::SetInterrupt(INT_CAUSE_CP, interrupt); ProcessorInterface::SetInterrupt(INT_CAUSE_CP, interrupt);
} }
} }
} }
else else
CommandProcessor::UpdateInterrupts(userdata); CommandProcessor::UpdateInterrupts(userdata);
} }
} }
void ProcessFifoToLoWatermark() void ProcessFifoToLoWatermark()

4
Source/Core/VideoCommon/Src/EmuWindow.cpp
View File

@ -80,7 +80,7 @@ void FreeLookInput( UINT iMsg, WPARAM wParam )
static bool mouseLookEnabled = false; static bool mouseLookEnabled = false;
static bool mouseMoveEnabled = false; static bool mouseMoveEnabled = false;
static float lastMouse[2]; static float lastMouse[2];
POINT point; POINT point;
switch(iMsg) switch(iMsg)
{ {
case WM_USER_KEYDOWN: case WM_USER_KEYDOWN:
@ -133,7 +133,7 @@ void FreeLookInput( UINT iMsg, WPARAM wParam )
lastMouse[1] = (float)point.y; lastMouse[1] = (float)point.y;
mouseLookEnabled= true; mouseLookEnabled= true;
break; break;
case WM_MBUTTONDOWN: case WM_MBUTTONDOWN:
GetCursorPos(&point); GetCursorPos(&point);
lastMouse[0] = (float)point.x; lastMouse[0] = (float)point.x;
lastMouse[1] = (float)point.y; lastMouse[1] = (float)point.y;

40
Source/Core/VideoCommon/Src/Fifo.cpp
View File

@ -43,10 +43,10 @@ static int size = 0;
void Fifo_DoState(PointerWrap &p) void Fifo_DoState(PointerWrap &p)
{ {
p.DoArray(videoBuffer, FIFO_SIZE); p.DoArray(videoBuffer, FIFO_SIZE);
p.Do(size); p.Do(size);
p.DoPointer(g_pVideoData, videoBuffer); p.DoPointer(g_pVideoData, videoBuffer);
p.Do(g_bSkipCurrentFrame); p.Do(g_bSkipCurrentFrame);
} }
void Fifo_PauseAndLock(bool doLock, bool unpauseOnUnlock) void Fifo_PauseAndLock(bool doLock, bool unpauseOnUnlock)
@ -69,8 +69,8 @@ void Fifo_PauseAndLock(bool doLock, bool unpauseOnUnlock)
void Fifo_Init() void Fifo_Init()
{ {
videoBuffer = (u8*)AllocateMemoryPages(FIFO_SIZE); videoBuffer = (u8*)AllocateMemoryPages(FIFO_SIZE);
size = 0; size = 0;
GpuRunningState = false; GpuRunningState = false;
Common::AtomicStore(CommandProcessor::VITicks, CommandProcessor::m_cpClockOrigin); Common::AtomicStore(CommandProcessor::VITicks, CommandProcessor::m_cpClockOrigin);
@ -79,12 +79,12 @@ void Fifo_Init()
void Fifo_Shutdown() void Fifo_Shutdown()
{ {
if (GpuRunningState) PanicAlert("Fifo shutting down while active"); if (GpuRunningState) PanicAlert("Fifo shutting down while active");
FreeMemoryPages(videoBuffer, FIFO_SIZE); FreeMemoryPages(videoBuffer, FIFO_SIZE);
} }
u8* GetVideoBufferStartPtr() u8* GetVideoBufferStartPtr()
{ {
return videoBuffer; return videoBuffer;
} }
u8* GetVideoBufferEndPtr() u8* GetVideoBufferEndPtr()
@ -119,20 +119,20 @@ void EmulatorState(bool running)
// Description: RunGpuLoop() sends data through this function. // Description: RunGpuLoop() sends data through this function.
void ReadDataFromFifo(u8* _uData, u32 len) void ReadDataFromFifo(u8* _uData, u32 len)
{ {
if (size + len >= FIFO_SIZE) if (size + len >= FIFO_SIZE)
{ {
int pos = (int)(g_pVideoData - videoBuffer); int pos = (int)(g_pVideoData - videoBuffer);
size -= pos; size -= pos;
if (size + len > FIFO_SIZE) if (size + len > FIFO_SIZE)
{ {
PanicAlert("FIFO out of bounds (sz = %i, len = %i at %08x)", size, len, pos); PanicAlert("FIFO out of bounds (sz = %i, len = %i at %08x)", size, len, pos);
} }
memmove(&videoBuffer[0], &videoBuffer[pos], size); memmove(&videoBuffer[0], &videoBuffer[pos], size);
g_pVideoData = videoBuffer; g_pVideoData = videoBuffer;
} }
// Copy new video instructions to videoBuffer for future use in rendering the new picture // Copy new video instructions to videoBuffer for future use in rendering the new picture
memcpy(videoBuffer + size, _uData, len); memcpy(videoBuffer + size, _uData, len);
size += len; size += len;
} }
void ResetVideoBuffer() void ResetVideoBuffer()
@ -153,7 +153,7 @@ void RunGpuLoop()
while (GpuRunningState) while (GpuRunningState)
{ {
g_video_backend->PeekMessages(); g_video_backend->PeekMessages();
VideoFifo_CheckAsyncRequest(); VideoFifo_CheckAsyncRequest();
@ -240,7 +240,7 @@ void RunGpu()
while (fifo.bFF_GPReadEnable && fifo.CPReadWriteDistance && !AtBreakpoint() ) while (fifo.bFF_GPReadEnable && fifo.CPReadWriteDistance && !AtBreakpoint() )
{ {
u8 *uData = Memory::GetPointer(fifo.CPReadPointer); u8 *uData = Memory::GetPointer(fifo.CPReadPointer);
FPURoundMode::SaveSIMDState(); FPURoundMode::SaveSIMDState();
FPURoundMode::LoadDefaultSIMDState(); FPURoundMode::LoadDefaultSIMDState();
ReadDataFromFifo(uData, 32); ReadDataFromFifo(uData, 32);

12
Source/Core/VideoCommon/Src/FramebufferManagerBase.cpp
View File

@ -60,12 +60,12 @@ const XFBSourceBase* const* FramebufferManagerBase::GetRealXFBSource(u32 xfbAddr
m_realXFBSource->texHeight = fbHeight; m_realXFBSource->texHeight = fbHeight;
// TODO: stuff only used by OGL... :/ // TODO: stuff only used by OGL... :/
// OpenGL texture coordinates originate at the lower left, which is why // OpenGL texture coordinates originate at the lower left, which is why
// sourceRc.top = fbHeight and sourceRc.bottom = 0. // sourceRc.top = fbHeight and sourceRc.bottom = 0.
m_realXFBSource->sourceRc.left = 0; m_realXFBSource->sourceRc.left = 0;
m_realXFBSource->sourceRc.top = fbHeight; m_realXFBSource->sourceRc.top = fbHeight;
m_realXFBSource->sourceRc.right = fbWidth; m_realXFBSource->sourceRc.right = fbWidth;
m_realXFBSource->sourceRc.bottom = 0; m_realXFBSource->sourceRc.bottom = 0;
// Decode YUYV data from GameCube RAM // Decode YUYV data from GameCube RAM
m_realXFBSource->DecodeToTexture(xfbAddr, fbWidth, fbHeight); m_realXFBSource->DecodeToTexture(xfbAddr, fbWidth, fbHeight);

4
Source/Core/VideoCommon/Src/GenericDLCache.cpp
View File

@ -23,7 +23,7 @@
namespace DLCache namespace DLCache
{ {
void Init() void Init()
{ {
} }
@ -33,7 +33,7 @@ void Shutdown()
} }
void Clear() void Clear()
{ {
} }
void ProgressiveCleanup() void ProgressiveCleanup()

128
Source/Core/VideoCommon/Src/GenericTextureDecoder.cpp
View File

@ -480,7 +480,7 @@ inline void decodebytesARGB8_4ToRgba(u32 *dst, const u16 *src, const u16 * src2)
{ {
#if 0 #if 0
for (int x = 0; x < 4; x++) { for (int x = 0; x < 4; x++) {
dst[x] = ((src[x] & 0xFF) << 24) | ((src[x] & 0xFF00)>>8) | (src2[x] << 8); dst[x] = ((src[x] & 0xFF) << 24) | ((src[x] & 0xFF00)>>8) | (src2[x] << 8);
} }
#else #else
dst[0] = ((src[0] & 0xFF) << 24) | ((src[0] & 0xFF00)>>8) | (src2[0] << 8); dst[0] = ((src[0] & 0xFF) << 24) | ((src[0] & 0xFF00)>>8) | (src2[0] << 8);
@ -519,7 +519,7 @@ void decodeDXTBlock(u32 *dst, const DXTBlock *src, int pitch)
{ {
int blue3 = ((blue2 - blue1) >> 1) - ((blue2 - blue1) >> 3); int blue3 = ((blue2 - blue1) >> 1) - ((blue2 - blue1) >> 3);
int green3 = ((green2 - green1) >> 1) - ((green2 - green1) >> 3); int green3 = ((green2 - green1) >> 1) - ((green2 - green1) >> 3);
int red3 = ((red2 - red1) >> 1) - ((red2 - red1) >> 3); int red3 = ((red2 - red1) >> 1) - ((red2 - red1) >> 3);
colors[2] = makecol(red1 + red3, green1 + green3, blue1 + blue3, 255); colors[2] = makecol(red1 + red3, green1 + green3, blue1 + blue3, 255);
colors[3] = makecol(red2 - red3, green2 - green3, blue2 - blue3, 255); colors[3] = makecol(red2 - red3, green2 - green3, blue2 - blue3, 255);
} }
@ -562,7 +562,7 @@ void decodeDXTBlockRGBA(u32 *dst, const DXTBlock *src, int pitch)
{ {
int blue3 = ((blue2 - blue1) >> 1) - ((blue2 - blue1) >> 3); int blue3 = ((blue2 - blue1) >> 1) - ((blue2 - blue1) >> 3);
int green3 = ((green2 - green1) >> 1) - ((green2 - green1) >> 3); int green3 = ((green2 - green1) >> 1) - ((green2 - green1) >> 3);
int red3 = ((red2 - red1) >> 1) - ((red2 - red1) >> 3); int red3 = ((red2 - red1) >> 1) - ((red2 - red1) >> 3);
colors[2] = makeRGBA(red1 + red3, green1 + green3, blue1 + blue3, 255); colors[2] = makeRGBA(red1 + red3, green1 + green3, blue1 + blue3, 255);
colors[3] = makeRGBA(red2 - red3, green2 - green3, blue2 - blue3, 255); colors[3] = makeRGBA(red2 - red3, green2 - green3, blue2 - blue3, 255);
} }
@ -664,22 +664,22 @@ PC_TexFormat TexDecoder_Decode_real(u8 *dst, const u8 *src, int width, int heigh
if (tlutfmt == 2) if (tlutfmt == 2)
{ {
// Special decoding is required for TLUT format 5A3 // Special decoding is required for TLUT format 5A3
for (int y = 0; y < height; y += 8) for (int y = 0; y < height; y += 8)
for (int x = 0, yStep = (y / 8) * Wsteps8; x < width; x += 8, yStep++) for (int x = 0, yStep = (y / 8) * Wsteps8; x < width; x += 8, yStep++)
for (int iy = 0, xStep = yStep * 8; iy < 8; iy++, xStep++) for (int iy = 0, xStep = yStep * 8; iy < 8; iy++, xStep++)
decodebytesC4_5A3_To_BGRA32((u32*)dst + (y + iy) * width + x, src + 4 * xStep, tlutaddr); decodebytesC4_5A3_To_BGRA32((u32*)dst + (y + iy) * width + x, src + 4 * xStep, tlutaddr);
} }
else else
{ {
for (int y = 0; y < height; y += 8) for (int y = 0; y < height; y += 8)
for (int x = 0, yStep = (y / 8) * Wsteps8; x < width; x += 8, yStep++) for (int x = 0, yStep = (y / 8) * Wsteps8; x < width; x += 8, yStep++)
for (int iy = 0, xStep = yStep * 8; iy < 8; iy++, xStep++) for (int iy = 0, xStep = yStep * 8; iy < 8; iy++, xStep++)
decodebytesC4_To_Raw16((u16*)dst + (y + iy) * width + x, src + 4 * xStep, tlutaddr); decodebytesC4_To_Raw16((u16*)dst + (y + iy) * width + x, src + 4 * xStep, tlutaddr);
} }
return GetPCFormatFromTLUTFormat(tlutfmt); return GetPCFormatFromTLUTFormat(tlutfmt);
case GX_TF_I4: case GX_TF_I4:
{ {
for (int y = 0; y < height; y += 8) for (int y = 0; y < height; y += 8)
for (int x = 0, yStep = (y / 8) * Wsteps8; x < width; x += 8, yStep++) for (int x = 0, yStep = (y / 8) * Wsteps8; x < width; x += 8, yStep++)
for (int iy = 0, xStep = yStep * 8 ; iy < 8; iy++,xStep++) for (int iy = 0, xStep = yStep * 8 ; iy < 8; iy++,xStep++)
for (int ix = 0; ix < 4; ix++) for (int ix = 0; ix < 4; ix++)
@ -692,19 +692,17 @@ PC_TexFormat TexDecoder_Decode_real(u8 *dst, const u8 *src, int width, int heigh
return PC_TEX_FMT_I4_AS_I8; return PC_TEX_FMT_I4_AS_I8;
case GX_TF_I8: // speed critical case GX_TF_I8: // speed critical
{ {
for (int y = 0; y < height; y += 4) for (int y = 0; y < height; y += 4)
for (int x = 0, yStep = (y / 4) * Wsteps8; x < width; x += 8, yStep++) for (int x = 0, yStep = (y / 4) * Wsteps8; x < width; x += 8, yStep++)
for (int iy = 0, xStep = 4 * yStep; iy < 4; iy++, xStep++) for (int iy = 0, xStep = 4 * yStep; iy < 4; iy++, xStep++)
{
((u64*)(dst + (y + iy) * width + x))[0] = ((u64*)(src + 8 * xStep))[0]; ((u64*)(dst + (y + iy) * width + x))[0] = ((u64*)(src + 8 * xStep))[0];
}
} }
return PC_TEX_FMT_I8; return PC_TEX_FMT_I8;
case GX_TF_C8: case GX_TF_C8:
if (tlutfmt == 2) if (tlutfmt == 2)
{ {
// Special decoding is required for TLUT format 5A3 // Special decoding is required for TLUT format 5A3
for (int y = 0; y < height; y += 4) for (int y = 0; y < height; y += 4)
for (int x = 0, yStep = (y / 4) * Wsteps8; x < width; x += 8, yStep++) for (int x = 0, yStep = (y / 4) * Wsteps8; x < width; x += 8, yStep++)
for (int iy = 0, xStep = 4 * yStep; iy < 4; iy++, xStep++) for (int iy = 0, xStep = 4 * yStep; iy < 4; iy++, xStep++)
decodebytesC8_5A3_To_BGRA32((u32*)dst + (y + iy) * width + x, src + 8 * xStep, tlutaddr); decodebytesC8_5A3_To_BGRA32((u32*)dst + (y + iy) * width + x, src + 8 * xStep, tlutaddr);
@ -713,7 +711,7 @@ PC_TexFormat TexDecoder_Decode_real(u8 *dst, const u8 *src, int width, int heigh
{ {
{ {
for (int y = 0; y < height; y += 4) for (int y = 0; y < height; y += 4)
for (int x = 0, yStep = (y / 4) * Wsteps8; x < width; x += 8, yStep++) for (int x = 0, yStep = (y / 4) * Wsteps8; x < width; x += 8, yStep++)
for (int iy = 0, xStep = 4 * yStep; iy < 4; iy++, xStep++) for (int iy = 0, xStep = 4 * yStep; iy < 4; iy++, xStep++)
decodebytesC8_To_Raw16((u16*)dst + (y + iy) * width + x, src + 8 * xStep, tlutaddr); decodebytesC8_To_Raw16((u16*)dst + (y + iy) * width + x, src + 8 * xStep, tlutaddr);
@ -722,7 +720,7 @@ PC_TexFormat TexDecoder_Decode_real(u8 *dst, const u8 *src, int width, int heigh
return GetPCFormatFromTLUTFormat(tlutfmt); return GetPCFormatFromTLUTFormat(tlutfmt);
case GX_TF_IA4: case GX_TF_IA4:
{ {
for (int y = 0; y < height; y += 4) for (int y = 0; y < height; y += 4)
for (int x = 0, yStep = (y / 4) * Wsteps8; x < width; x += 8, yStep++) for (int x = 0, yStep = (y / 4) * Wsteps8; x < width; x += 8, yStep++)
for (int iy = 0, xStep = 4 * yStep; iy < 4; iy++, xStep++) for (int iy = 0, xStep = 4 * yStep; iy < 4; iy++, xStep++)
decodebytesIA4((u16*)dst + (y + iy) * width + x, src + 8 * xStep); decodebytesIA4((u16*)dst + (y + iy) * width + x, src + 8 * xStep);
@ -730,7 +728,7 @@ PC_TexFormat TexDecoder_Decode_real(u8 *dst, const u8 *src, int width, int heigh
return PC_TEX_FMT_IA4_AS_IA8; return PC_TEX_FMT_IA4_AS_IA8;
case GX_TF_IA8: case GX_TF_IA8:
{ {
for (int y = 0; y < height; y += 4) for (int y = 0; y < height; y += 4)
for (int x = 0, yStep = (y / 4) * Wsteps4; x < width; x += 4, yStep++) for (int x = 0, yStep = (y / 4) * Wsteps4; x < width; x += 4, yStep++)
for (int iy = 0, xStep = yStep * 4; iy < 4; iy++, xStep++) for (int iy = 0, xStep = yStep * 4; iy < 4; iy++, xStep++)
{ {
@ -739,21 +737,20 @@ PC_TexFormat TexDecoder_Decode_real(u8 *dst, const u8 *src, int width, int heigh
for(int j = 0; j < 4; j++) for(int j = 0; j < 4; j++)
*ptr++ = Common::swap16(*s++); *ptr++ = Common::swap16(*s++);
} }
} }
return PC_TEX_FMT_IA8; return PC_TEX_FMT_IA8;
case GX_TF_C14X2: case GX_TF_C14X2:
if (tlutfmt == 2) if (tlutfmt == 2)
{ {
// Special decoding is required for TLUT format 5A3 // Special decoding is required for TLUT format 5A3
for (int y = 0; y < height; y += 4) for (int y = 0; y < height; y += 4)
for (int x = 0, yStep = (y / 4) * Wsteps4; x < width; x += 4, yStep++) for (int x = 0, yStep = (y / 4) * Wsteps4; x < width; x += 4, yStep++)
for (int iy = 0, xStep = 4 * yStep; iy < 4; iy++, xStep++) for (int iy = 0, xStep = 4 * yStep; iy < 4; iy++, xStep++)
decodebytesC14X2_5A3_To_BGRA32((u32*)dst + (y + iy) * width + x, (u16*)(src + 8 * xStep), tlutaddr); decodebytesC14X2_5A3_To_BGRA32((u32*)dst + (y + iy) * width + x, (u16*)(src + 8 * xStep), tlutaddr);
} }
else else
{ {
for (int y = 0; y < height; y += 4) for (int y = 0; y < height; y += 4)
for (int x = 0, yStep = (y / 4) * Wsteps4; x < width; x += 4, yStep++) for (int x = 0, yStep = (y / 4) * Wsteps4; x < width; x += 4, yStep++)
for (int iy = 0, xStep = 4 * yStep; iy < 4; iy++, xStep++) for (int iy = 0, xStep = 4 * yStep; iy < 4; iy++, xStep++)
decodebytesC14X2_To_Raw16((u16*)dst + (y + iy) * width + x,(u16*)(src + 8 * xStep), tlutaddr); decodebytesC14X2_To_Raw16((u16*)dst + (y + iy) * width + x,(u16*)(src + 8 * xStep), tlutaddr);
@ -761,7 +758,7 @@ PC_TexFormat TexDecoder_Decode_real(u8 *dst, const u8 *src, int width, int heigh
return GetPCFormatFromTLUTFormat(tlutfmt); return GetPCFormatFromTLUTFormat(tlutfmt);
case GX_TF_RGB565: case GX_TF_RGB565:
{ {
for (int y = 0; y < height; y += 4) for (int y = 0; y < height; y += 4)
for (int x = 0, yStep = (y / 4) * Wsteps4; x < width; x += 4, yStep++) for (int x = 0, yStep = (y / 4) * Wsteps4; x < width; x += 4, yStep++)
for (int iy = 0, xStep = 4 * yStep; iy < 4; iy++, xStep++) for (int iy = 0, xStep = 4 * yStep; iy < 4; iy++, xStep++)
{ {
@ -774,7 +771,7 @@ PC_TexFormat TexDecoder_Decode_real(u8 *dst, const u8 *src, int width, int heigh
return PC_TEX_FMT_RGB565; return PC_TEX_FMT_RGB565;
case GX_TF_RGB5A3: case GX_TF_RGB5A3:
{ {
for (int y = 0; y < height; y += 4) for (int y = 0; y < height; y += 4)
for (int x = 0, yStep = (y / 4) * Wsteps4; x < width; x += 4, yStep++) for (int x = 0, yStep = (y / 4) * Wsteps4; x < width; x += 4, yStep++)
for (int iy = 0, xStep = 4 * yStep; iy < 4; iy++, xStep++) for (int iy = 0, xStep = 4 * yStep; iy < 4; iy++, xStep++)
//decodebytesRGB5A3((u32*)dst+(y+iy)*width+x, (u16*)src, 4); //decodebytesRGB5A3((u32*)dst+(y+iy)*width+x, (u16*)src, 4);
@ -783,15 +780,14 @@ PC_TexFormat TexDecoder_Decode_real(u8 *dst, const u8 *src, int width, int heigh
return PC_TEX_FMT_BGRA32; return PC_TEX_FMT_BGRA32;
case GX_TF_RGBA8: // speed critical case GX_TF_RGBA8: // speed critical
{ {
for (int y = 0; y < height; y += 4)
{ {
for (int y = 0; y < height; y += 4) for (int x = 0, yStep = (y / 4) * Wsteps4; x < width; x += 4, yStep++)
for (int x = 0, yStep = (y / 4) * Wsteps4; x < width; x += 4, yStep++) {
{ const u8* src2 = src + 64 * yStep;
const u8* src2 = src + 64 * yStep; for (int iy = 0; iy < 4; iy++)
for (int iy = 0; iy < 4; iy++) decodebytesARGB8_4((u32*)dst + (y+iy)*width + x, (u16*)src2 + 4 * iy, (u16*)src2 + 4 * iy + 16);
decodebytesARGB8_4((u32*)dst + (y+iy)*width + x, (u16*)src2 + 4 * iy, (u16*)src2 + 4 * iy + 16); }
}
} }
} }
return PC_TEX_FMT_BGRA32; return PC_TEX_FMT_BGRA32;
@ -817,7 +813,7 @@ PC_TexFormat TexDecoder_Decode_real(u8 *dst, const u8 *src, int width, int heigh
} }
return PC_TEX_FMT_DXT1; return PC_TEX_FMT_DXT1;
#else #else
for (int y = 0; y < height; y += 8) for (int y = 0; y < height; y += 8)
{ {
for (int x = 0, yStep = (y / 8) * Wsteps8; x < width; x += 8, yStep++) for (int x = 0, yStep = (y / 8) * Wsteps8; x < width; x += 8, yStep++)
{ {
@ -862,22 +858,21 @@ PC_TexFormat TexDecoder_Decode_RGBA(u32 * dst, const u8 * src, int width, int he
if (tlutfmt == 2) if (tlutfmt == 2)
{ {
// Special decoding is required for TLUT format 5A3 // Special decoding is required for TLUT format 5A3
for (int y = 0; y < height; y += 8) for (int y = 0; y < height; y += 8)
for (int x = 0, yStep = (y / 8) * Wsteps8; x < width; x += 8,yStep++) for (int x = 0, yStep = (y / 8) * Wsteps8; x < width; x += 8,yStep++)
for (int iy = 0, xStep = 8 * yStep; iy < 8; iy++,xStep++) for (int iy = 0, xStep = 8 * yStep; iy < 8; iy++,xStep++)
decodebytesC4_5A3_To_rgba32(dst + (y + iy) * width + x, src + 4 * xStep, tlutaddr); decodebytesC4_5A3_To_rgba32(dst + (y + iy) * width + x, src + 4 * xStep, tlutaddr);
} }
else if(tlutfmt == 0) else if(tlutfmt == 0)
{ {
for (int y = 0; y < height; y += 8) for (int y = 0; y < height; y += 8)
for (int x = 0, yStep = (y / 8) * Wsteps8; x < width; x += 8,yStep++) for (int x = 0, yStep = (y / 8) * Wsteps8; x < width; x += 8,yStep++)
for (int iy = 0, xStep = 8 * yStep; iy < 8; iy++,xStep++) for (int iy = 0, xStep = 8 * yStep; iy < 8; iy++,xStep++)
decodebytesC4IA8_To_RGBA(dst + (y + iy) * width + x, src + 4 * xStep, tlutaddr); decodebytesC4IA8_To_RGBA(dst + (y + iy) * width + x, src + 4 * xStep, tlutaddr);
} }
else else
{ {
for (int y = 0; y < height; y += 8) for (int y = 0; y < height; y += 8)
for (int x = 0, yStep = (y / 8) * Wsteps8; x < width; x += 8,yStep++) for (int x = 0, yStep = (y / 8) * Wsteps8; x < width; x += 8,yStep++)
for (int iy = 0, xStep = 8 * yStep; iy < 8; iy++,xStep++) for (int iy = 0, xStep = 8 * yStep; iy < 8; iy++,xStep++)
decodebytesC4RGB565_To_RGBA(dst + (y + iy) * width + x, src + 4 * xStep, tlutaddr); decodebytesC4RGB565_To_RGBA(dst + (y + iy) * width + x, src + 4 * xStep, tlutaddr);
@ -925,34 +920,33 @@ PC_TexFormat TexDecoder_Decode_RGBA(u32 * dst, const u8 * src, int width, int he
if (tlutfmt == 2) if (tlutfmt == 2)
{ {
// Special decoding is required for TLUT format 5A3 // Special decoding is required for TLUT format 5A3
for (int y = 0; y < height; y += 4) for (int y = 0; y < height; y += 4)
for (int x = 0, yStep = (y / 4) * Wsteps8; x < width; x += 8, yStep++) for (int x = 0, yStep = (y / 4) * Wsteps8; x < width; x += 8, yStep++)
for (int iy = 0, xStep = 4 * yStep; iy < 4; iy++, xStep++) for (int iy = 0, xStep = 4 * yStep; iy < 4; iy++, xStep++)
decodebytesC8_5A3_To_RGBA32((u32*)dst + (y + iy) * width + x, src + 8 * xStep, tlutaddr); decodebytesC8_5A3_To_RGBA32((u32*)dst + (y + iy) * width + x, src + 8 * xStep, tlutaddr);
} }
else if(tlutfmt == 0) else if(tlutfmt == 0)
{ {
for (int y = 0; y < height; y += 4) for (int y = 0; y < height; y += 4)
for (int x = 0, yStep = (y / 4) * Wsteps8; x < width; x += 8, yStep++) for (int x = 0, yStep = (y / 4) * Wsteps8; x < width; x += 8, yStep++)
for (int iy = 0, xStep = 4 * yStep; iy < 4; iy++, xStep++) for (int iy = 0, xStep = 4 * yStep; iy < 4; iy++, xStep++)
decodebytesC8IA8_To_RGBA(dst + (y + iy) * width + x, src + 8 * xStep, tlutaddr); decodebytesC8IA8_To_RGBA(dst + (y + iy) * width + x, src + 8 * xStep, tlutaddr);
} }
else else
{ {
for (int y = 0; y < height; y += 4) for (int y = 0; y < height; y += 4)
for (int x = 0, yStep = (y / 4) * Wsteps8; x < width; x += 8, yStep++) for (int x = 0, yStep = (y / 4) * Wsteps8; x < width; x += 8, yStep++)
for (int iy = 0, xStep = 4 * yStep; iy < 4; iy++, xStep++) for (int iy = 0, xStep = 4 * yStep; iy < 4; iy++, xStep++)
decodebytesC8RGB565_To_RGBA(dst + (y + iy) * width + x, src + 8 * xStep, tlutaddr); decodebytesC8RGB565_To_RGBA(dst + (y + iy) * width + x, src + 8 * xStep, tlutaddr);
} }
break; break;
case GX_TF_IA4: case GX_TF_IA4:
{ {
for (int y = 0; y < height; y += 4) for (int y = 0; y < height; y += 4)
for (int x = 0, yStep = (y / 4) * Wsteps8; x < width; x += 8, yStep++) for (int x = 0, yStep = (y / 4) * Wsteps8; x < width; x += 8, yStep++)
for (int iy = 0, xStep = 4 * yStep; iy < 4; iy++, xStep++) for (int iy = 0, xStep = 4 * yStep; iy < 4; iy++, xStep++)
decodebytesIA4RGBA(dst + (y + iy) * width + x, src + 8 * xStep); decodebytesIA4RGBA(dst + (y + iy) * width + x, src + 8 * xStep);
} }
break; break;
case GX_TF_IA8: case GX_TF_IA8:
@ -975,21 +969,21 @@ PC_TexFormat TexDecoder_Decode_RGBA(u32 * dst, const u8 * src, int width, int he
if (tlutfmt == 2) if (tlutfmt == 2)
{ {
// Special decoding is required for TLUT format 5A3 // Special decoding is required for TLUT format 5A3
for (int y = 0; y < height; y += 4) for (int y = 0; y < height; y += 4)
for (int x = 0, yStep = (y / 4) * Wsteps4; x < width; x += 4, yStep++) for (int x = 0, yStep = (y / 4) * Wsteps4; x < width; x += 4, yStep++)
for (int iy = 0, xStep = 4 * yStep; iy < 4; iy++, xStep++) for (int iy = 0, xStep = 4 * yStep; iy < 4; iy++, xStep++)
decodebytesC14X2_5A3_To_BGRA32(dst + (y + iy) * width + x, (u16*)(src + 8 * xStep), tlutaddr); decodebytesC14X2_5A3_To_BGRA32(dst + (y + iy) * width + x, (u16*)(src + 8 * xStep), tlutaddr);
} }
else if (tlutfmt == 0) else if (tlutfmt == 0)
{ {
for (int y = 0; y < height; y += 4) for (int y = 0; y < height; y += 4)
for (int x = 0, yStep = (y / 4) * Wsteps4; x < width; x += 4, yStep++) for (int x = 0, yStep = (y / 4) * Wsteps4; x < width; x += 4, yStep++)
for (int iy = 0, xStep = 4 * yStep; iy < 4; iy++, xStep++) for (int iy = 0, xStep = 4 * yStep; iy < 4; iy++, xStep++)
decodebytesC14X2IA8_To_RGBA(dst + (y + iy) * width + x, (u16*)(src + 8 * xStep), tlutaddr); decodebytesC14X2IA8_To_RGBA(dst + (y + iy) * width + x, (u16*)(src + 8 * xStep), tlutaddr);
} }
else else
{ {
for (int y = 0; y < height; y += 4) for (int y = 0; y < height; y += 4)
for (int x = 0, yStep = (y / 4) * Wsteps4; x < width; x += 4, yStep++) for (int x = 0, yStep = (y / 4) * Wsteps4; x < width; x += 4, yStep++)
for (int iy = 0, xStep = 4 * yStep; iy < 4; iy++, xStep++) for (int iy = 0, xStep = 4 * yStep; iy < 4; iy++, xStep++)
decodebytesC14X2rgb565_To_RGBA(dst + (y + iy) * width + x, (u16*)(src + 8 * xStep), tlutaddr); decodebytesC14X2rgb565_To_RGBA(dst + (y + iy) * width + x, (u16*)(src + 8 * xStep), tlutaddr);
@ -1130,7 +1124,7 @@ PC_TexFormat TexDecoder_Decode(u8 *dst, const u8 *src, int width, int height, in
dtp[(y + yoff)*width + x + xoff] = ptr[x] ? 0xFFFF : 0x0000; dtp[(y + yoff)*width + x + xoff] = ptr[x] ? 0xFFFF : 0x0000;
break; break;
} }
default: default:
case PC_TEX_FMT_BGRA32: case PC_TEX_FMT_BGRA32:
{ {
int *dtp = (int*)dst; int *dtp = (int*)dst;
@ -1160,7 +1154,7 @@ void TexDecoder_DecodeTexel(u8 *dst, const u8 *src, int s, int t, int imageWidth
u32 base = (tBlk * widthBlks + sBlk) * blockWidth * blockHeight; u32 base = (tBlk * widthBlks + sBlk) * blockWidth * blockHeight;
u16 blkS = s & (blockWidth - 1); u16 blkS = s & (blockWidth - 1);
u16 blkT = t & (blockHeight - 1); u16 blkT = t & (blockHeight - 1);
u32 blkOff = blkT * blockWidth + blkS; u32 blkOff = blkT * blockWidth + blkS;
*/ */
switch (texformat) switch (texformat)
@ -1174,9 +1168,9 @@ void TexDecoder_DecodeTexel(u8 *dst, const u8 *src, int s, int t, int imageWidth
u16 blkS = s & 7; u16 blkS = s & 7;
u16 blkT = t & 7; u16 blkT = t & 7;
u32 blkOff = (blkT << 3) + blkS; u32 blkOff = (blkT << 3) + blkS;
int rs = (blkOff & 1)?0:4; int rs = (blkOff & 1)?0:4;
u32 offset = base + (blkOff >> 1); u32 offset = base + (blkOff >> 1);
u8 val = (*(src + offset) >> rs) & 0xF; u8 val = (*(src + offset) >> rs) & 0xF;
u16 *tlut = (u16*)(texMem + tlutaddr); u16 *tlut = (u16*)(texMem + tlutaddr);
@ -1186,7 +1180,7 @@ void TexDecoder_DecodeTexel(u8 *dst, const u8 *src, int s, int t, int imageWidth
case 0: case 0:
*((u32*)dst) = decodeIA8Swapped(tlut[val]); *((u32*)dst) = decodeIA8Swapped(tlut[val]);
break; break;
case 1: case 1:
*((u32*)dst) = decode565RGBA(Common::swap16(tlut[val])); *((u32*)dst) = decode565RGBA(Common::swap16(tlut[val]));
break; break;
case 2: case 2:
@ -1204,9 +1198,9 @@ void TexDecoder_DecodeTexel(u8 *dst, const u8 *src, int s, int t, int imageWidth
u16 blkS = s & 7; u16 blkS = s & 7;
u16 blkT = t & 7; u16 blkT = t & 7;
u32 blkOff = (blkT << 3) + blkS; u32 blkOff = (blkT << 3) + blkS;
int rs = (blkOff & 1)?0:4; int rs = (blkOff & 1)?0:4;
u32 offset = base + (blkOff >> 1); u32 offset = base + (blkOff >> 1);
u8 val = (*(src + offset) >> rs) & 0xF; u8 val = (*(src + offset) >> rs) & 0xF;
val = Convert4To8(val); val = Convert4To8(val);
@ -1225,7 +1219,7 @@ void TexDecoder_DecodeTexel(u8 *dst, const u8 *src, int s, int t, int imageWidth
u16 blkS = s & 7; u16 blkS = s & 7;
u16 blkT = t & 3; u16 blkT = t & 3;
u32 blkOff = (blkT << 3) + blkS; u32 blkOff = (blkT << 3) + blkS;
u8 val = *(src + base + blkOff); u8 val = *(src + base + blkOff);
dst[0] = val; dst[0] = val;
dst[1] = val; dst[1] = val;
@ -1242,7 +1236,7 @@ void TexDecoder_DecodeTexel(u8 *dst, const u8 *src, int s, int t, int imageWidth
u16 blkS = s & 7; u16 blkS = s & 7;
u16 blkT = t & 3; u16 blkT = t & 3;
u32 blkOff = (blkT << 3) + blkS; u32 blkOff = (blkT << 3) + blkS;
u8 val = *(src + base + blkOff); u8 val = *(src + base + blkOff);
u16 *tlut = (u16*)(texMem + tlutaddr); u16 *tlut = (u16*)(texMem + tlutaddr);
@ -1251,7 +1245,7 @@ void TexDecoder_DecodeTexel(u8 *dst, const u8 *src, int s, int t, int imageWidth
case 0: case 0:
*((u32*)dst) = decodeIA8Swapped(tlut[val]); *((u32*)dst) = decodeIA8Swapped(tlut[val]);
break; break;
case 1: case 1:
*((u32*)dst) = decode565RGBA(Common::swap16(tlut[val])); *((u32*)dst) = decode565RGBA(Common::swap16(tlut[val]));
break; break;
case 2: case 2:
@ -1269,7 +1263,7 @@ void TexDecoder_DecodeTexel(u8 *dst, const u8 *src, int s, int t, int imageWidth
u16 blkS = s & 7; u16 blkS = s & 7;
u16 blkT = t & 3; u16 blkT = t & 3;
u32 blkOff = (blkT << 3) + blkS; u32 blkOff = (blkT << 3) + blkS;
u8 val = *(src + base + blkOff); u8 val = *(src + base + blkOff);
const u8 a = Convert4To8(val>>4); const u8 a = Convert4To8(val>>4);
const u8 l = Convert4To8(val&0xF); const u8 l = Convert4To8(val&0xF);
@ -1288,7 +1282,7 @@ void TexDecoder_DecodeTexel(u8 *dst, const u8 *src, int s, int t, int imageWidth
u16 blkS = s & 3; u16 blkS = s & 3;
u16 blkT = t & 3; u16 blkT = t & 3;
u32 blkOff = (blkT << 2) + blkS; u32 blkOff = (blkT << 2) + blkS;
u32 offset = (base + blkOff) << 1; u32 offset = (base + blkOff) << 1;
const u16* valAddr = (u16*)(src + offset); const u16* valAddr = (u16*)(src + offset);
@ -1304,7 +1298,7 @@ void TexDecoder_DecodeTexel(u8 *dst, const u8 *src, int s, int t, int imageWidth
u16 blkS = s & 3; u16 blkS = s & 3;
u16 blkT = t & 3; u16 blkT = t & 3;
u32 blkOff = (blkT << 2) + blkS; u32 blkOff = (blkT << 2) + blkS;
u32 offset = (base + blkOff) << 1; u32 offset = (base + blkOff) << 1;
const u16* valAddr = (u16*)(src + offset); const u16* valAddr = (u16*)(src + offset);
@ -1316,7 +1310,7 @@ void TexDecoder_DecodeTexel(u8 *dst, const u8 *src, int s, int t, int imageWidth
case 0: case 0:
*((u32*)dst) = decodeIA8Swapped(tlut[val]); *((u32*)dst) = decodeIA8Swapped(tlut[val]);
break; break;
case 1: case 1:
*((u32*)dst) = decode565RGBA(Common::swap16(tlut[val])); *((u32*)dst) = decode565RGBA(Common::swap16(tlut[val]));
break; break;
case 2: case 2:
@ -1334,7 +1328,7 @@ void TexDecoder_DecodeTexel(u8 *dst, const u8 *src, int s, int t, int imageWidth
u16 blkS = s & 3; u16 blkS = s & 3;
u16 blkT = t & 3; u16 blkT = t & 3;
u32 blkOff = (blkT << 2) + blkS; u32 blkOff = (blkT << 2) + blkS;
u32 offset = (base + blkOff) << 1; u32 offset = (base + blkOff) << 1;
const u16* valAddr = (u16*)(src + offset); const u16* valAddr = (u16*)(src + offset);
@ -1392,7 +1386,7 @@ void TexDecoder_DecodeTexel(u8 *dst, const u8 *src, int s, int t, int imageWidth
u32 offset = (base + blkOff) << 3; u32 offset = (base + blkOff) << 3;
const DXTBlock* dxtBlock = (const DXTBlock*)(src + offset); const DXTBlock* dxtBlock = (const DXTBlock*)(src + offset);
u16 c1 = Common::swap16(dxtBlock->color1); u16 c1 = Common::swap16(dxtBlock->color1);
u16 c2 = Common::swap16(dxtBlock->color2); u16 c2 = Common::swap16(dxtBlock->color2);
int blue1 = Convert5To8(c1 & 0x1F); int blue1 = Convert5To8(c1 & 0x1F);
@ -1411,7 +1405,7 @@ void TexDecoder_DecodeTexel(u8 *dst, const u8 *src, int s, int t, int imageWidth
colorSel |= c1 > c2?0:4; colorSel |= c1 > c2?0:4;
u32 color = 0; u32 color = 0;
switch (colorSel) switch (colorSel)
{ {
case 0: case 0:

30
Source/Core/VideoCommon/Src/ImageWrite.cpp
View File

@ -25,22 +25,22 @@
struct TGA_HEADER struct TGA_HEADER
{ {
u8 identsize; // size of ID field that follows 18 u8 header (0 usually) u8 identsize; // size of ID field that follows 18 u8 header (0 usually)
u8 colourmaptype; // type of colour map 0=none, 1=has palette u8 colourmaptype; // type of colour map 0=none, 1=has palette
u8 imagetype; // type of image 0=none,1=indexed,2=rgb,3=grey,+8=rle packed u8 imagetype; // type of image 0=none,1=indexed,2=rgb,3=grey,+8=rle packed
s16 colourmapstart; // first colour map entry in palette s16 colourmapstart; // first colour map entry in palette
s16 colourmaplength; // number of colours in palette s16 colourmaplength; // number of colours in palette
u8 colourmapbits; // number of bits per palette entry 15,16,24,32 u8 colourmapbits; // number of bits per palette entry 15,16,24,32
s16 xstart; // image x origin s16 xstart; // image x origin
s16 ystart; // image y origin s16 ystart; // image y origin
s16 width; // image width in pixels s16 width; // image width in pixels
s16 height; // image height in pixels s16 height; // image height in pixels
u8 bits; // image bits per pixel 8,16,24,32 u8 bits; // image bits per pixel 8,16,24,32
u8 descriptor; // image descriptor bits (vh flip bits) u8 descriptor; // image descriptor bits (vh flip bits)
// pixel data follows header // pixel data follows header
}; };
#pragma pack(pop) #pragma pack(pop)
@ -53,7 +53,7 @@ bool SaveTGA(const char* filename, int width, int height, void* pdata)
return false; return false;
_assert_(sizeof(TGA_HEADER) == 18 && sizeof(hdr) == 18); _assert_(sizeof(TGA_HEADER) == 18 && sizeof(hdr) == 18);
memset(&hdr, 0, sizeof(hdr)); memset(&hdr, 0, sizeof(hdr));
hdr.imagetype = 2; hdr.imagetype = 2;
hdr.bits = 32; hdr.bits = 32;

2
Source/Core/VideoCommon/Src/IndexGenerator.cpp
View File

@ -63,7 +63,7 @@ void IndexGenerator::AddIndices(int primitive, u32 numVerts)
//case GX_DRAW_TRIANGLES: IndexGenerator::AddList(numVerts); break; //case GX_DRAW_TRIANGLES: IndexGenerator::AddList(numVerts); break;
//case GX_DRAW_TRIANGLE_STRIP: IndexGenerator::AddStrip(numVerts); break; //case GX_DRAW_TRIANGLE_STRIP: IndexGenerator::AddStrip(numVerts); break;
//case GX_DRAW_TRIANGLE_FAN: IndexGenerator::AddFan(numVerts); break; //case GX_DRAW_TRIANGLE_FAN: IndexGenerator::AddFan(numVerts); break;
//case GX_DRAW_LINES: IndexGenerator::AddLineList(numVerts); break; //case GX_DRAW_LINES: IndexGenerator::AddLineList(numVerts); break;
//case GX_DRAW_LINE_STRIP: IndexGenerator::AddLineStrip(numVerts); break; //case GX_DRAW_LINE_STRIP: IndexGenerator::AddLineStrip(numVerts); break;
//case GX_DRAW_POINTS: IndexGenerator::AddPoints(numVerts); break; //case GX_DRAW_POINTS: IndexGenerator::AddPoints(numVerts); break;
//} //}

16
Source/Core/VideoCommon/Src/IndexGenerator.h
View File

@ -27,17 +27,17 @@ class IndexGenerator
public: public:
// Init // Init
static void Start(u16 *Triangleptr,u16 *Lineptr,u16 *Pointptr); static void Start(u16 *Triangleptr,u16 *Lineptr,u16 *Pointptr);
static void AddIndices(int primitive, u32 numVertices); static void AddIndices(int primitive, u32 numVertices);
// Interface // Interface
static u32 GetNumTriangles() {return numT;} static u32 GetNumTriangles() {return numT;}
static u32 GetNumLines() {return numL;} static u32 GetNumLines() {return numL;}
static u32 GetNumPoints() {return numP;} static u32 GetNumPoints() {return numP;}
// returns numprimitives // returns numprimitives
static u32 GetNumVerts() {return index;} static u32 GetNumVerts() {return index;}
static u32 GetTriangleindexLen() {return (u32)(Tptr - BASETptr);} static u32 GetTriangleindexLen() {return (u32)(Tptr - BASETptr);}
static u32 GetLineindexLen() {return (u32)(Lptr - BASELptr);} static u32 GetLineindexLen() {return (u32)(Lptr - BASELptr);}
static u32 GetPointindexLen() {return (u32)(Pptr - BASEPptr);} static u32 GetPointindexLen() {return (u32)(Pptr - BASEPptr);}
@ -56,16 +56,16 @@ private:
static void AddStrip(u32 numVerts); static void AddStrip(u32 numVerts);
static void AddFan(u32 numVerts); static void AddFan(u32 numVerts);
static void AddQuads(u32 numVerts); static void AddQuads(u32 numVerts);
// Lines // Lines
static void AddLineList(u32 numVerts); static void AddLineList(u32 numVerts);
static void AddLineStrip(u32 numVerts); static void AddLineStrip(u32 numVerts);
// Points // Points
static void AddPoints(u32 numVerts); static void AddPoints(u32 numVerts);
static void WriteTriangle(u32 index1, u32 index2, u32 index3); static void WriteTriangle(u32 index1, u32 index2, u32 index3);
static u16 *Tptr; static u16 *Tptr;
static u16 *BASETptr; static u16 *BASETptr;
static u16 *Lptr; static u16 *Lptr;

362
Source/Core/VideoCommon/Src/OpcodeDecoding.cpp
View File

@ -123,7 +123,7 @@ void InterpretDisplayList(u32 address, u32 size)
Statistics::SwapDL(); Statistics::SwapDL();
} }
// reset to the old pointer // reset to the old pointer
g_pVideoData = old_pVideoData; g_pVideoData = old_pVideoData;
} }
@ -140,14 +140,14 @@ u32 FifoCommandRunnable(u32 &command_size)
{ {
u32 cycleTime = 0; u32 cycleTime = 0;
u32 buffer_size = (u32)(GetVideoBufferEndPtr() - g_pVideoData); u32 buffer_size = (u32)(GetVideoBufferEndPtr() - g_pVideoData);
if (buffer_size == 0) if (buffer_size == 0)
return 0; // can't peek return 0; // can't peek
u8 cmd_byte = DataPeek8(0); u8 cmd_byte = DataPeek8(0);
switch (cmd_byte) switch (cmd_byte)
{ {
case GX_NOP: // Hm, this means that we scan over nop streams pretty slowly... case GX_NOP: // Hm, this means that we scan over nop streams pretty slowly...
command_size = 1; command_size = 1;
cycleTime = 6; cycleTime = 6;
break; break;
@ -155,30 +155,30 @@ u32 FifoCommandRunnable(u32 &command_size)
command_size = 1; command_size = 1;
cycleTime = 6; cycleTime = 6;
break; break;
case GX_CMD_UNKNOWN_METRICS: // zelda 4 swords calls it and checks the metrics registers after that case GX_CMD_UNKNOWN_METRICS: // zelda 4 swords calls it and checks the metrics registers after that
command_size = 1; command_size = 1;
cycleTime = 6; cycleTime = 6;
break; break;
case GX_LOAD_BP_REG: case GX_LOAD_BP_REG:
command_size = 5; command_size = 5;
cycleTime = 12; cycleTime = 12;
break; break;
case GX_LOAD_CP_REG: case GX_LOAD_CP_REG:
command_size = 6; command_size = 6;
cycleTime = 12; cycleTime = 12;
break; break;
case GX_LOAD_INDX_A: case GX_LOAD_INDX_A:
case GX_LOAD_INDX_B: case GX_LOAD_INDX_B:
case GX_LOAD_INDX_C: case GX_LOAD_INDX_C:
case GX_LOAD_INDX_D: case GX_LOAD_INDX_D:
command_size = 5; command_size = 5;
cycleTime = 6; // TODO cycleTime = 6; // TODO
break; break;
case GX_CMD_CALL_DL: case GX_CMD_CALL_DL:
{ {
// FIXME: Calculate the cycle time of the display list. // FIXME: Calculate the cycle time of the display list.
//u32 address = DataPeek32(1); //u32 address = DataPeek32(1);
@ -208,42 +208,42 @@ u32 FifoCommandRunnable(u32 &command_size)
command_size = 9; command_size = 9;
cycleTime = 45; // This is unverified cycleTime = 45; // This is unverified
} }
break; break;
case GX_LOAD_XF_REG: case GX_LOAD_XF_REG:
{ {
// check if we can read the header // check if we can read the header
if (buffer_size >= 5) if (buffer_size >= 5)
{ {
command_size = 1 + 4; command_size = 1 + 4;
u32 Cmd2 = DataPeek32(1); u32 Cmd2 = DataPeek32(1);
int transfer_size = ((Cmd2 >> 16) & 15) + 1; int transfer_size = ((Cmd2 >> 16) & 15) + 1;
command_size += transfer_size * 4; command_size += transfer_size * 4;
cycleTime = 18 + 6 * transfer_size; cycleTime = 18 + 6 * transfer_size;
} }
else else
{ {
return 0; return 0;
} }
} }
break; break;
default: default:
if (cmd_byte & 0x80) if (cmd_byte & 0x80)
{ {
// check if we can read the header // check if we can read the header
if (buffer_size >= 3) if (buffer_size >= 3)
{ {
command_size = 1 + 2; command_size = 1 + 2;
u16 numVertices = DataPeek16(1); u16 numVertices = DataPeek16(1);
command_size += numVertices * VertexLoaderManager::GetVertexSize(cmd_byte & GX_VAT_MASK); command_size += numVertices * VertexLoaderManager::GetVertexSize(cmd_byte & GX_VAT_MASK);
cycleTime = 1600; // This depends on the number of pixels rendered cycleTime = 1600; // This depends on the number of pixels rendered
} }
else else
{ {
return 0; return 0;
} }
} }
else else
{ {
// TODO(Omega): Maybe dump FIFO to file on this error // TODO(Omega): Maybe dump FIFO to file on this error
@ -258,7 +258,7 @@ u32 FifoCommandRunnable(u32 &command_size)
Host_SysMessage(szTemp); Host_SysMessage(szTemp);
INFO_LOG(VIDEO, "%s", szTemp); INFO_LOG(VIDEO, "%s", szTemp);
{ {
SCPFifoStruct &fifo = CommandProcessor::fifo; SCPFifoStruct &fifo = CommandProcessor::fifo;
char szTmp[512]; char szTmp[512];
// sprintf(szTmp, "Illegal command %02x (at %08x)",cmd_byte,g_pDataReader->GetPtr()); // sprintf(szTmp, "Illegal command %02x (at %08x)",cmd_byte,g_pDataReader->GetPtr());
@ -283,18 +283,18 @@ u32 FifoCommandRunnable(u32 &command_size)
Host_SysMessage(szTmp); Host_SysMessage(szTmp);
INFO_LOG(VIDEO, "%s", szTmp); INFO_LOG(VIDEO, "%s", szTmp);
} }
} }
break; break;
} }
if (command_size > buffer_size)
return 0;
// INFO_LOG("OP detected: cmd_byte 0x%x size %i buffer %i",cmd_byte, command_size, buffer_size); if (command_size > buffer_size)
return 0;
// INFO_LOG("OP detected: cmd_byte 0x%x size %i buffer %i",cmd_byte, command_size, buffer_size);
if (cycleTime == 0) if (cycleTime == 0)
cycleTime = 6; cycleTime = 6;
return cycleTime; return cycleTime;
} }
u32 FifoCommandRunnable() u32 FifoCommandRunnable()
@ -305,92 +305,92 @@ u32 FifoCommandRunnable()
static void Decode() static void Decode()
{ {
u8 *opcodeStart = g_pVideoData; u8 *opcodeStart = g_pVideoData;
int cmd_byte = DataReadU8(); int cmd_byte = DataReadU8();
switch (cmd_byte) switch (cmd_byte)
{ {
case GX_NOP: case GX_NOP:
break; break;
case GX_LOAD_CP_REG: //0x08 case GX_LOAD_CP_REG: //0x08
{ {
u8 sub_cmd = DataReadU8(); u8 sub_cmd = DataReadU8();
u32 value = DataReadU32(); u32 value = DataReadU32();
LoadCPReg(sub_cmd, value); LoadCPReg(sub_cmd, value);
INCSTAT(stats.thisFrame.numCPLoads); INCSTAT(stats.thisFrame.numCPLoads);
} }
break; break;
case GX_LOAD_XF_REG: case GX_LOAD_XF_REG:
{ {
u32 Cmd2 = DataReadU32(); u32 Cmd2 = DataReadU32();
int transfer_size = ((Cmd2 >> 16) & 15) + 1; int transfer_size = ((Cmd2 >> 16) & 15) + 1;
u32 xf_address = Cmd2 & 0xFFFF; u32 xf_address = Cmd2 & 0xFFFF;
GC_ALIGNED128(u32 data_buffer[16]); GC_ALIGNED128(u32 data_buffer[16]);
DataReadU32xFuncs[transfer_size-1](data_buffer); DataReadU32xFuncs[transfer_size-1](data_buffer);
LoadXFReg(transfer_size, xf_address, data_buffer); LoadXFReg(transfer_size, xf_address, data_buffer);
INCSTAT(stats.thisFrame.numXFLoads); INCSTAT(stats.thisFrame.numXFLoads);
} }
break; break;
case GX_LOAD_INDX_A: //used for position matrices case GX_LOAD_INDX_A: //used for position matrices
LoadIndexedXF(DataReadU32(), 0xC); LoadIndexedXF(DataReadU32(), 0xC);
break; break;
case GX_LOAD_INDX_B: //used for normal matrices case GX_LOAD_INDX_B: //used for normal matrices
LoadIndexedXF(DataReadU32(), 0xD); LoadIndexedXF(DataReadU32(), 0xD);
break; break;
case GX_LOAD_INDX_C: //used for postmatrices case GX_LOAD_INDX_C: //used for postmatrices
LoadIndexedXF(DataReadU32(), 0xE); LoadIndexedXF(DataReadU32(), 0xE);
break; break;
case GX_LOAD_INDX_D: //used for lights case GX_LOAD_INDX_D: //used for lights
LoadIndexedXF(DataReadU32(), 0xF); LoadIndexedXF(DataReadU32(), 0xF);
break; break;
case GX_CMD_CALL_DL: case GX_CMD_CALL_DL:
{ {
u32 address = DataReadU32(); u32 address = DataReadU32();
u32 count = DataReadU32(); u32 count = DataReadU32();
ExecuteDisplayList(address, count); ExecuteDisplayList(address, count);
} }
break; break;
case GX_CMD_UNKNOWN_METRICS: // zelda 4 swords calls it and checks the metrics registers after that case GX_CMD_UNKNOWN_METRICS: // zelda 4 swords calls it and checks the metrics registers after that
DEBUG_LOG(VIDEO, "GX 0x44: %08x", cmd_byte); DEBUG_LOG(VIDEO, "GX 0x44: %08x", cmd_byte);
break; break;
case GX_CMD_INVL_VC: // Invalidate Vertex Cache case GX_CMD_INVL_VC: // Invalidate Vertex Cache
DEBUG_LOG(VIDEO, "Invalidate (vertex cache?)"); DEBUG_LOG(VIDEO, "Invalidate (vertex cache?)");
break; break;
case GX_LOAD_BP_REG: //0x61 case GX_LOAD_BP_REG: //0x61
{ {
u32 bp_cmd = DataReadU32(); u32 bp_cmd = DataReadU32();
LoadBPReg(bp_cmd); LoadBPReg(bp_cmd);
INCSTAT(stats.thisFrame.numBPLoads); INCSTAT(stats.thisFrame.numBPLoads);
} }
break; break;
// draw primitives // draw primitives
default: default:
if (cmd_byte & 0x80) if (cmd_byte & 0x80)
{ {
// load vertices (use computed vertex size from FifoCommandRunnable above) // load vertices (use computed vertex size from FifoCommandRunnable above)
u16 numVertices = DataReadU16(); u16 numVertices = DataReadU16();
VertexLoaderManager::RunVertices( VertexLoaderManager::RunVertices(
cmd_byte & GX_VAT_MASK, // Vertex loader index (0 - 7) cmd_byte & GX_VAT_MASK, // Vertex loader index (0 - 7)
(cmd_byte & GX_PRIMITIVE_MASK) >> GX_PRIMITIVE_SHIFT, (cmd_byte & GX_PRIMITIVE_MASK) >> GX_PRIMITIVE_SHIFT,
numVertices); numVertices);
} }
else else
{ {
ERROR_LOG(VIDEO, "OpcodeDecoding::Decode: Illegal command %02x", cmd_byte); ERROR_LOG(VIDEO, "OpcodeDecoding::Decode: Illegal command %02x", cmd_byte);
break; break;
} }
break; break;
} }
// Display lists get added directly into the FIFO stream // Display lists get added directly into the FIFO stream
if (g_bRecordFifoData && cmd_byte != GX_CMD_CALL_DL) if (g_bRecordFifoData && cmd_byte != GX_CMD_CALL_DL)
@ -399,89 +399,89 @@ static void Decode()
static void DecodeSemiNop() static void DecodeSemiNop()
{ {
u8 *opcodeStart = g_pVideoData; u8 *opcodeStart = g_pVideoData;
int cmd_byte = DataReadU8(); int cmd_byte = DataReadU8();
switch (cmd_byte) switch (cmd_byte)
{ {
case GX_CMD_UNKNOWN_METRICS: // zelda 4 swords calls it and checks the metrics registers after that case GX_CMD_UNKNOWN_METRICS: // zelda 4 swords calls it and checks the metrics registers after that
case GX_CMD_INVL_VC: // Invalidate Vertex Cache case GX_CMD_INVL_VC: // Invalidate Vertex Cache
case GX_NOP: case GX_NOP:
break; break;
case GX_LOAD_CP_REG: //0x08 case GX_LOAD_CP_REG: //0x08
// We have to let CP writes through because they determine the size of vertices. // We have to let CP writes through because they determine the size of vertices.
{ {
u8 sub_cmd = DataReadU8(); u8 sub_cmd = DataReadU8();
u32 value = DataReadU32(); u32 value = DataReadU32();
LoadCPReg(sub_cmd, value); LoadCPReg(sub_cmd, value);
INCSTAT(stats.thisFrame.numCPLoads); INCSTAT(stats.thisFrame.numCPLoads);
} }
break; break;
case GX_LOAD_XF_REG: case GX_LOAD_XF_REG:
{ {
u32 Cmd2 = DataReadU32(); u32 Cmd2 = DataReadU32();
int transfer_size = ((Cmd2 >> 16) & 15) + 1; int transfer_size = ((Cmd2 >> 16) & 15) + 1;
u32 address = Cmd2 & 0xFFFF; u32 address = Cmd2 & 0xFFFF;
GC_ALIGNED128(u32 data_buffer[16]); GC_ALIGNED128(u32 data_buffer[16]);
DataReadU32xFuncs[transfer_size-1](data_buffer); DataReadU32xFuncs[transfer_size-1](data_buffer);
LoadXFReg(transfer_size, address, data_buffer); LoadXFReg(transfer_size, address, data_buffer);
INCSTAT(stats.thisFrame.numXFLoads); INCSTAT(stats.thisFrame.numXFLoads);
} }
break; break;
case GX_LOAD_INDX_A: //used for position matrices case GX_LOAD_INDX_A: //used for position matrices
LoadIndexedXF(DataReadU32(), 0xC); LoadIndexedXF(DataReadU32(), 0xC);
break; break;
case GX_LOAD_INDX_B: //used for normal matrices case GX_LOAD_INDX_B: //used for normal matrices
LoadIndexedXF(DataReadU32(), 0xD); LoadIndexedXF(DataReadU32(), 0xD);
break; break;
case GX_LOAD_INDX_C: //used for postmatrices case GX_LOAD_INDX_C: //used for postmatrices
LoadIndexedXF(DataReadU32(), 0xE); LoadIndexedXF(DataReadU32(), 0xE);
break; break;
case GX_LOAD_INDX_D: //used for lights case GX_LOAD_INDX_D: //used for lights
LoadIndexedXF(DataReadU32(), 0xF); LoadIndexedXF(DataReadU32(), 0xF);
break; break;
case GX_CMD_CALL_DL: case GX_CMD_CALL_DL:
// Hm, wonder if any games put tokens in display lists - in that case, // Hm, wonder if any games put tokens in display lists - in that case,
// we'll have to parse them too. // we'll have to parse them too.
DataSkip(8); DataSkip(8);
break; break;
case GX_LOAD_BP_REG: //0x61 case GX_LOAD_BP_REG: //0x61
// We have to let BP writes through because they set tokens and stuff. // We have to let BP writes through because they set tokens and stuff.
// TODO: Call a much simplified LoadBPReg instead. // TODO: Call a much simplified LoadBPReg instead.
{ {
u32 bp_cmd = DataReadU32(); u32 bp_cmd = DataReadU32();
LoadBPReg(bp_cmd); LoadBPReg(bp_cmd);
INCSTAT(stats.thisFrame.numBPLoads); INCSTAT(stats.thisFrame.numBPLoads);
} }
break; break;
// draw primitives // draw primitives
default: default:
if (cmd_byte & 0x80) if (cmd_byte & 0x80)
{ {
// load vertices (use computed vertex size from FifoCommandRunnable above) // load vertices (use computed vertex size from FifoCommandRunnable above)
u16 numVertices = DataReadU16(); u16 numVertices = DataReadU16();
DataSkip(numVertices * VertexLoaderManager::GetVertexSize(cmd_byte & GX_VAT_MASK)); DataSkip(numVertices * VertexLoaderManager::GetVertexSize(cmd_byte & GX_VAT_MASK));
} }
else else
{ {
ERROR_LOG(VIDEO, "OpcodeDecoding::Decode: Illegal command %02x", cmd_byte); ERROR_LOG(VIDEO, "OpcodeDecoding::Decode: Illegal command %02x", cmd_byte);
break; break;
} }
break; break;
} }
if (g_bRecordFifoData && cmd_byte != GX_CMD_CALL_DL) if (g_bRecordFifoData && cmd_byte != GX_CMD_CALL_DL)
FifoRecorder::GetInstance().WriteGPCommand(opcodeStart, u32(g_pVideoData - opcodeStart)); FifoRecorder::GetInstance().WriteGPCommand(opcodeStart, u32(g_pVideoData - opcodeStart));
} }
void OpcodeDecoder_Init() void OpcodeDecoder_Init()
{ {
g_pVideoData = GetVideoBufferStartPtr(); g_pVideoData = GetVideoBufferStartPtr();
#if _M_SSE >= 0x301 #if _M_SSE >= 0x301
@ -494,8 +494,8 @@ void OpcodeDecoder_Init()
if (g_Config.bEnableOpenCL) if (g_Config.bEnableOpenCL)
{ {
OpenCL::Initialize(); OpenCL::Initialize();
TexDecoder_OpenCL_Initialize(); TexDecoder_OpenCL_Initialize();
} }
} }
@ -504,8 +504,8 @@ void OpcodeDecoder_Shutdown()
{ {
if (g_Config.bEnableOpenCL) if (g_Config.bEnableOpenCL)
{ {
TexDecoder_OpenCL_Shutdown(); TexDecoder_OpenCL_Shutdown();
OpenCL::Destroy(); OpenCL::Destroy();
} }
} }

2
Source/Core/VideoCommon/Src/OpcodeDecoding.h
View File

@ -44,7 +44,7 @@
#define GX_DRAW_LINES 0x5 // 0xA8 #define GX_DRAW_LINES 0x5 // 0xA8
#define GX_DRAW_LINE_STRIP 0x6 // 0xB0 #define GX_DRAW_LINE_STRIP 0x6 // 0xB0
#define GX_DRAW_POINTS 0x7 // 0xB8 #define GX_DRAW_POINTS 0x7 // 0xB8
#define GX_DRAW_NONE 0x1; //Tis is a fake value to used in the backends #define GX_DRAW_NONE 0x1; //Tis is a fake value to used in the backends
extern bool g_bRecordFifoData; extern bool g_bRecordFifoData;

28
Source/Core/VideoCommon/Src/OpenCL.cpp
View File

@ -148,25 +148,25 @@ cl_program CompileProgram(const char *Kernel)
// Build the program executable // Build the program executable
err = clBuildProgram(program , 0, NULL, NULL, NULL, NULL); err = clBuildProgram(program , 0, NULL, NULL, NULL, NULL);
if(err != CL_SUCCESS) { if(err != CL_SUCCESS) {
HandleCLError(err, "Error: failed to build program"); HandleCLError(err, "Error: failed to build program");
char *buildlog = NULL; char *buildlog = NULL;
size_t buildlog_size = 0; size_t buildlog_size = 0;
clGetProgramBuildInfo(program, OpenCL::device_id, CL_PROGRAM_BUILD_LOG, 0, NULL, &buildlog_size); clGetProgramBuildInfo(program, OpenCL::device_id, CL_PROGRAM_BUILD_LOG, 0, NULL, &buildlog_size);
buildlog = new char[buildlog_size + 1]; buildlog = new char[buildlog_size + 1];
err = clGetProgramBuildInfo(program, OpenCL::device_id, CL_PROGRAM_BUILD_LOG, buildlog_size, buildlog, NULL); err = clGetProgramBuildInfo(program, OpenCL::device_id, CL_PROGRAM_BUILD_LOG, buildlog_size, buildlog, NULL);
buildlog[buildlog_size] = 0; buildlog[buildlog_size] = 0;
if(err != CL_SUCCESS) if(err != CL_SUCCESS)
{ {
HandleCLError(err, "Error: can't get build log"); HandleCLError(err, "Error: can't get build log");
} else } else
{ {
ERROR_LOG(COMMON, "Error log:\n%s\n", buildlog); ERROR_LOG(COMMON, "Error log:\n%s\n", buildlog);
} }
delete[] buildlog; delete[] buildlog;
return NULL; return NULL;
} }

4
Source/Core/VideoCommon/Src/OpenCL/OCLTextureDecoder.cpp
View File

@ -98,7 +98,7 @@ void TexDecoder_OpenCL_Initialize()
size_t nDevices = 0; size_t nDevices = 0;
cl_device_id *devices = NULL; cl_device_id *devices = NULL;
size_t *binary_sizes = NULL; size_t *binary_sizes = NULL;
char **binaries = NULL; char **binaries = NULL;
std::string filename; std::string filename;
char dolphin_rev[HEADER_SIZE]; char dolphin_rev[HEADER_SIZE];
@ -132,7 +132,7 @@ void TexDecoder_OpenCL_Initialize()
{ {
OpenCL::HandleCLError(err, "clCreateProgramWithBinary"); OpenCL::HandleCLError(err, "clCreateProgramWithBinary");
} }
if (!err) if (!err)
{ {
err = clBuildProgram(g_program, 1, &OpenCL::device_id, NULL, NULL, NULL); err = clBuildProgram(g_program, 1, &OpenCL::device_id, NULL, NULL, NULL);

4
Source/Core/VideoCommon/Src/PixelEngine.cpp
View File

@ -126,8 +126,8 @@ bool bbox_active;
enum enum
{ {
INT_CAUSE_PE_TOKEN = 0x200, // GP Token INT_CAUSE_PE_TOKEN = 0x200, // GP Token
INT_CAUSE_PE_FINISH = 0x400, // GP Finished INT_CAUSE_PE_FINISH = 0x400, // GP Finished
}; };
void DoState(PointerWrap &p) void DoState(PointerWrap &p)

12
Source/Core/VideoCommon/Src/PixelEngine.h
View File

@ -24,13 +24,13 @@ class PointerWrap;
// internal hardware addresses // internal hardware addresses
enum enum
{ {
PE_ZCONF = 0x00, // Z Config PE_ZCONF = 0x00, // Z Config
PE_ALPHACONF = 0x02, // Alpha Config PE_ALPHACONF = 0x02, // Alpha Config
PE_DSTALPHACONF = 0x04, // Destination Alpha Config PE_DSTALPHACONF = 0x04, // Destination Alpha Config
PE_ALPHAMODE = 0x06, // Alpha Mode Config PE_ALPHAMODE = 0x06, // Alpha Mode Config
PE_ALPHAREAD = 0x08, // Alpha Read PE_ALPHAREAD = 0x08, // Alpha Read
PE_CTRL_REGISTER = 0x0a, // Control PE_CTRL_REGISTER = 0x0a, // Control
PE_TOKEN_REG = 0x0e, // Token PE_TOKEN_REG = 0x0e, // Token
PE_BBOX_LEFT = 0x10, // Flip Left PE_BBOX_LEFT = 0x10, // Flip Left
PE_BBOX_RIGHT = 0x12, // Flip Right PE_BBOX_RIGHT = 0x12, // Flip Right
PE_BBOX_TOP = 0x14, // Flip Top PE_BBOX_TOP = 0x14, // Flip Top

44
Source/Core/VideoCommon/Src/PixelShaderGen.cpp
View File

@ -374,38 +374,38 @@ static const char *tevOpTable[] = { // TEV
static const char *tevCInputTable[] = // CC static const char *tevCInputTable[] = // CC
{ {
"(prev.rgb)", // CPREV, "(prev.rgb)", // CPREV,
"(prev.aaa)", // APREV, "(prev.aaa)", // APREV,
"(c0.rgb)", // C0, "(c0.rgb)", // C0,
"(c0.aaa)", // A0, "(c0.aaa)", // A0,
"(c1.rgb)", // C1, "(c1.rgb)", // C1,
"(c1.aaa)", // A1, "(c1.aaa)", // A1,
"(c2.rgb)", // C2, "(c2.rgb)", // C2,
"(c2.aaa)", // A2, "(c2.aaa)", // A2,
"(textemp.rgb)", // TEXC, "(textemp.rgb)", // TEXC,
"(textemp.aaa)", // TEXA, "(textemp.aaa)", // TEXA,
"(rastemp.rgb)", // RASC, "(rastemp.rgb)", // RASC,
"(rastemp.aaa)", // RASA, "(rastemp.aaa)", // RASA,
"float3(1.0f, 1.0f, 1.0f)", // ONE "float3(1.0f, 1.0f, 1.0f)", // ONE
"float3(0.5f, 0.5f, 0.5f)", // HALF "float3(0.5f, 0.5f, 0.5f)", // HALF
"(konsttemp.rgb)", //"konsttemp.rgb", // KONST "(konsttemp.rgb)", //"konsttemp.rgb", // KONST
"float3(0.0f, 0.0f, 0.0f)", // ZERO "float3(0.0f, 0.0f, 0.0f)", // ZERO
///aded extra values to map clamped values ///aded extra values to map clamped values
"(cprev.rgb)", // CPREV, "(cprev.rgb)", // CPREV,
"(cprev.aaa)", // APREV, "(cprev.aaa)", // APREV,
"(cc0.rgb)", // C0, "(cc0.rgb)", // C0,
"(cc0.aaa)", // A0, "(cc0.aaa)", // A0,
"(cc1.rgb)", // C1, "(cc1.rgb)", // C1,
"(cc1.aaa)", // A1, "(cc1.aaa)", // A1,
"(cc2.rgb)", // C2, "(cc2.rgb)", // C2,
"(cc2.aaa)", // A2, "(cc2.aaa)", // A2,
"(textemp.rgb)", // TEXC, "(textemp.rgb)", // TEXC,
"(textemp.aaa)", // TEXA, "(textemp.aaa)", // TEXA,
"(crastemp.rgb)", // RASC, "(crastemp.rgb)", // RASC,
"(crastemp.aaa)", // RASA, "(crastemp.aaa)", // RASA,
"float3(1.0f, 1.0f, 1.0f)", // ONE "float3(1.0f, 1.0f, 1.0f)", // ONE
"float3(0.5f, 0.5f, 0.5f)", // HALF "float3(0.5f, 0.5f, 0.5f)", // HALF
"(ckonsttemp.rgb)", //"konsttemp.rgb", // KONST "(ckonsttemp.rgb)", //"konsttemp.rgb", // KONST
"float3(0.0f, 0.0f, 0.0f)", // ZERO "float3(0.0f, 0.0f, 0.0f)", // ZERO
"PADERROR1", "PADERROR2", "PADERROR3", "PADERROR4" "PADERROR1", "PADERROR2", "PADERROR3", "PADERROR4"
}; };
@ -425,7 +425,7 @@ static const char *tevAInputTable[] = // CA
"cc0", // A0, "cc0", // A0,
"cc1", // A1, "cc1", // A1,
"cc2", // A2, "cc2", // A2,
"textemp", // TEXA, "textemp", // TEXA,
"crastemp", // RASA, "crastemp", // RASA,
"ckonsttemp", // KONST, (hw1 had quarter) "ckonsttemp", // KONST, (hw1 had quarter)
"float4(0.0f, 0.0f, 0.0f, 0.0f)", // ZERO "float4(0.0f, 0.0f, 0.0f, 0.0f)", // ZERO

4
Source/Core/VideoCommon/Src/PixelShaderGen.h
View File

@ -28,8 +28,8 @@
#define I_INDTEXSCALE "cindscale" #define I_INDTEXSCALE "cindscale"
#define I_INDTEXMTX "cindmtx" #define I_INDTEXMTX "cindmtx"
#define I_FOG "cfog" #define I_FOG "cfog"
#define I_PLIGHTS "cPLights" #define I_PLIGHTS "cPLights"
#define I_PMATERIALS "cPmtrl" #define I_PMATERIALS "cPmtrl"
#define C_COLORMATRIX 0 // 0 #define C_COLORMATRIX 0 // 0
#define C_COLORS 0 // 0 #define C_COLORS 0 // 0

216
Source/Core/VideoCommon/Src/PixelShaderManager.cpp
View File

@ -87,61 +87,61 @@ void PixelShaderManager::SetConstants()
{ {
if (g_ActiveConfig.backend_info.APIType == API_OPENGL && !g_ActiveConfig.backend_info.bSupportsGLSLUBO) if (g_ActiveConfig.backend_info.APIType == API_OPENGL && !g_ActiveConfig.backend_info.bSupportsGLSLUBO)
Dirty(); Dirty();
for (int i = 0; i < 2; ++i) for (int i = 0; i < 2; ++i)
{ {
if (s_nColorsChanged[i]) if (s_nColorsChanged[i])
{ {
int baseind = i ? C_KCOLORS : C_COLORS; int baseind = i ? C_KCOLORS : C_COLORS;
for (int j = 0; j < 4; ++j) for (int j = 0; j < 4; ++j)
{ {
if (s_nColorsChanged[i] & (1 << j)) if (s_nColorsChanged[i] & (1 << j))
SetPSConstant4fv(baseind+j, &lastRGBAfull[i][j][0]); SetPSConstant4fv(baseind+j, &lastRGBAfull[i][j][0]);
} }
s_nColorsChanged[i] = 0; s_nColorsChanged[i] = 0;
} }
} }
if (s_nTexDimsChanged) if (s_nTexDimsChanged)
{ {
for (int i = 0; i < 8; ++i) for (int i = 0; i < 8; ++i)
{ {
if (s_nTexDimsChanged & (1<<i)) if (s_nTexDimsChanged & (1<<i))
SetPSTextureDims(i); SetPSTextureDims(i);
} }
s_nTexDimsChanged = 0; s_nTexDimsChanged = 0;
} }
if (s_bAlphaChanged) if (s_bAlphaChanged)
{ {
SetPSConstant4f(C_ALPHA, (lastAlpha&0xff)/255.0f, ((lastAlpha>>8)&0xff)/255.0f, 0, ((lastAlpha>>16)&0xff)/255.0f); SetPSConstant4f(C_ALPHA, (lastAlpha&0xff)/255.0f, ((lastAlpha>>8)&0xff)/255.0f, 0, ((lastAlpha>>16)&0xff)/255.0f);
s_bAlphaChanged = false; s_bAlphaChanged = false;
} }
if (s_bZTextureTypeChanged) if (s_bZTextureTypeChanged)
{ {
float ftemp[4]; float ftemp[4];
switch (bpmem.ztex2.type) switch (bpmem.ztex2.type)
{ {
case 0: case 0:
// 8 bits // 8 bits
ftemp[0] = 0; ftemp[1] = 0; ftemp[2] = 0; ftemp[3] = 255.0f/16777215.0f; ftemp[0] = 0; ftemp[1] = 0; ftemp[2] = 0; ftemp[3] = 255.0f/16777215.0f;
break; break;
case 1: case 1:
// 16 bits // 16 bits
ftemp[0] = 255.0f/16777215.0f; ftemp[1] = 0; ftemp[2] = 0; ftemp[3] = 65280.0f/16777215.0f; ftemp[0] = 255.0f/16777215.0f; ftemp[1] = 0; ftemp[2] = 0; ftemp[3] = 65280.0f/16777215.0f;
break; break;
case 2: case 2:
// 24 bits // 24 bits
ftemp[0] = 16711680.0f/16777215.0f; ftemp[1] = 65280.0f/16777215.0f; ftemp[2] = 255.0f/16777215.0f; ftemp[3] = 0; ftemp[0] = 16711680.0f/16777215.0f; ftemp[1] = 65280.0f/16777215.0f; ftemp[2] = 255.0f/16777215.0f; ftemp[3] = 0;
break; break;
} }
SetPSConstant4fv(C_ZBIAS, ftemp); SetPSConstant4fv(C_ZBIAS, ftemp);
s_bZTextureTypeChanged = false; s_bZTextureTypeChanged = false;
} }
if (s_bZBiasChanged || s_bDepthRangeChanged) if (s_bZBiasChanged || s_bDepthRangeChanged)
{ {
// reversed gxsetviewport(xorig, yorig, width, height, nearz, farz) // reversed gxsetviewport(xorig, yorig, width, height, nearz, farz)
// [0] = width/2 // [0] = width/2
// [1] = height/2 // [1] = height/2
// [2] = 16777215 * (farz - nearz) // [2] = 16777215 * (farz - nearz)
@ -152,78 +152,78 @@ void PixelShaderManager::SetConstants()
//ERROR_LOG("pixel=%x,%x, bias=%x\n", bpmem.zcontrol.pixel_format, bpmem.ztex2.type, lastZBias); //ERROR_LOG("pixel=%x,%x, bias=%x\n", bpmem.zcontrol.pixel_format, bpmem.ztex2.type, lastZBias);
SetPSConstant4f(C_ZBIAS+1, xfregs.viewport.farZ / 16777216.0f, xfregs.viewport.zRange / 16777216.0f, 0, (float)(lastZBias)/16777215.0f); SetPSConstant4f(C_ZBIAS+1, xfregs.viewport.farZ / 16777216.0f, xfregs.viewport.zRange / 16777216.0f, 0, (float)(lastZBias)/16777215.0f);
s_bZBiasChanged = s_bDepthRangeChanged = false; s_bZBiasChanged = s_bDepthRangeChanged = false;
} }
// indirect incoming texture scales // indirect incoming texture scales
if (s_nIndTexScaleChanged) if (s_nIndTexScaleChanged)
{ {
// set as two sets of vec4s, each containing S and T of two ind stages. // set as two sets of vec4s, each containing S and T of two ind stages.
float f[8]; float f[8];
if (s_nIndTexScaleChanged & 0x03) if (s_nIndTexScaleChanged & 0x03)
{ {
for (u32 i = 0; i < 2; ++i) for (u32 i = 0; i < 2; ++i)
{ {
f[2 * i] = bpmem.texscale[0].getScaleS(i & 1); f[2 * i] = bpmem.texscale[0].getScaleS(i & 1);
f[2 * i + 1] = bpmem.texscale[0].getScaleT(i & 1); f[2 * i + 1] = bpmem.texscale[0].getScaleT(i & 1);
PRIM_LOG("tex indscale%d: %f %f\n", i, f[2 * i], f[2 * i + 1]); PRIM_LOG("tex indscale%d: %f %f\n", i, f[2 * i], f[2 * i + 1]);
} }
SetPSConstant4fv(C_INDTEXSCALE, f); SetPSConstant4fv(C_INDTEXSCALE, f);
} }
if (s_nIndTexScaleChanged & 0x0c) { if (s_nIndTexScaleChanged & 0x0c) {
for (u32 i = 2; i < 4; ++i) { for (u32 i = 2; i < 4; ++i) {
f[2 * i] = bpmem.texscale[1].getScaleS(i & 1); f[2 * i] = bpmem.texscale[1].getScaleS(i & 1);
f[2 * i + 1] = bpmem.texscale[1].getScaleT(i & 1); f[2 * i + 1] = bpmem.texscale[1].getScaleT(i & 1);
PRIM_LOG("tex indscale%d: %f %f\n", i, f[2 * i], f[2 * i + 1]); PRIM_LOG("tex indscale%d: %f %f\n", i, f[2 * i], f[2 * i + 1]);
} }
SetPSConstant4fv(C_INDTEXSCALE+1, &f[4]); SetPSConstant4fv(C_INDTEXSCALE+1, &f[4]);
} }
s_nIndTexScaleChanged = 0; s_nIndTexScaleChanged = 0;
} }
if (s_nIndTexMtxChanged) if (s_nIndTexMtxChanged)
{ {
for (int i = 0; i < 3; ++i) for (int i = 0; i < 3; ++i)
{ {
if (s_nIndTexMtxChanged & (1 << i)) if (s_nIndTexMtxChanged & (1 << i))
{ {
int scale = ((u32)bpmem.indmtx[i].col0.s0 << 0) | int scale = ((u32)bpmem.indmtx[i].col0.s0 << 0) |
((u32)bpmem.indmtx[i].col1.s1 << 2) | ((u32)bpmem.indmtx[i].col1.s1 << 2) |
((u32)bpmem.indmtx[i].col2.s2 << 4); ((u32)bpmem.indmtx[i].col2.s2 << 4);
float fscale = powf(2.0f, (float)(scale - 17)) / 1024.0f; float fscale = powf(2.0f, (float)(scale - 17)) / 1024.0f;
// xyz - static matrix // xyz - static matrix
// TODO w - dynamic matrix scale / 256...... somehow / 4 works better // TODO w - dynamic matrix scale / 256...... somehow / 4 works better
// rev 2972 - now using / 256.... verify that this works // rev 2972 - now using / 256.... verify that this works
SetPSConstant4f(C_INDTEXMTX + 2 * i, SetPSConstant4f(C_INDTEXMTX + 2 * i,
bpmem.indmtx[i].col0.ma * fscale, bpmem.indmtx[i].col0.ma * fscale,
bpmem.indmtx[i].col1.mc * fscale, bpmem.indmtx[i].col1.mc * fscale,
bpmem.indmtx[i].col2.me * fscale, bpmem.indmtx[i].col2.me * fscale,
fscale * 4.0f); fscale * 4.0f);
SetPSConstant4f(C_INDTEXMTX + 2 * i + 1, SetPSConstant4f(C_INDTEXMTX + 2 * i + 1,
bpmem.indmtx[i].col0.mb * fscale, bpmem.indmtx[i].col0.mb * fscale,
bpmem.indmtx[i].col1.md * fscale, bpmem.indmtx[i].col1.md * fscale,
bpmem.indmtx[i].col2.mf * fscale, bpmem.indmtx[i].col2.mf * fscale,
fscale * 4.0f); fscale * 4.0f);
PRIM_LOG("indmtx%d: scale=%f, mat=(%f %f %f; %f %f %f)\n", PRIM_LOG("indmtx%d: scale=%f, mat=(%f %f %f; %f %f %f)\n",
i, 1024.0f*fscale, i, 1024.0f*fscale,
bpmem.indmtx[i].col0.ma * fscale, bpmem.indmtx[i].col1.mc * fscale, bpmem.indmtx[i].col2.me * fscale, bpmem.indmtx[i].col0.ma * fscale, bpmem.indmtx[i].col1.mc * fscale, bpmem.indmtx[i].col2.me * fscale,
bpmem.indmtx[i].col0.mb * fscale, bpmem.indmtx[i].col1.md * fscale, bpmem.indmtx[i].col2.mf * fscale); bpmem.indmtx[i].col0.mb * fscale, bpmem.indmtx[i].col1.md * fscale, bpmem.indmtx[i].col2.mf * fscale);
} }
} }
s_nIndTexMtxChanged = 0; s_nIndTexMtxChanged = 0;
} }
if (s_bFogColorChanged) if (s_bFogColorChanged)
{ {
SetPSConstant4f(C_FOG, bpmem.fog.color.r / 255.0f, bpmem.fog.color.g / 255.0f, bpmem.fog.color.b / 255.0f, 0); SetPSConstant4f(C_FOG, bpmem.fog.color.r / 255.0f, bpmem.fog.color.g / 255.0f, bpmem.fog.color.b / 255.0f, 0);
s_bFogColorChanged = false; s_bFogColorChanged = false;
} }
if (s_bFogParamChanged) if (s_bFogParamChanged)
{ {
if(!g_ActiveConfig.bDisableFog) if(!g_ActiveConfig.bDisableFog)
{ {
@ -236,8 +236,8 @@ void PixelShaderManager::SetConstants()
else else
SetPSConstant4f(C_FOG + 1, 0.0, 1.0, 0.0, 1.0); SetPSConstant4f(C_FOG + 1, 0.0, 1.0, 0.0, 1.0);
s_bFogParamChanged = false; s_bFogParamChanged = false;
} }
if (s_bFogRangeAdjustChanged) if (s_bFogRangeAdjustChanged)
{ {
@ -341,11 +341,11 @@ void PixelShaderManager::SetConstants()
void PixelShaderManager::SetPSTextureDims(int texid) void PixelShaderManager::SetPSTextureDims(int texid)
{ {
// texdims.xy are reciprocals of the real texture dimensions // texdims.xy are reciprocals of the real texture dimensions
// texdims.zw are the scaled dimensions // texdims.zw are the scaled dimensions
float fdims[4]; float fdims[4];
TCoordInfo& tc = bpmem.texcoords[texid]; TCoordInfo& tc = bpmem.texcoords[texid];
fdims[0] = 1.0f / (float)(lastTexDims[texid] & 0xffff); fdims[0] = 1.0f / (float)(lastTexDims[texid] & 0xffff);
fdims[1] = 1.0f / (float)((lastTexDims[texid] >> 16) & 0xfff); fdims[1] = 1.0f / (float)((lastTexDims[texid] >> 16) & 0xfff);
fdims[2] = (float)(tc.s.scale_minus_1 + 1); fdims[2] = (float)(tc.s.scale_minus_1 + 1);
@ -359,7 +359,7 @@ void PixelShaderManager::SetPSTextureDims(int texid)
// and update it when the shader constant is set, only. // and update it when the shader constant is set, only.
void PixelShaderManager::SetColorChanged(int type, int num, bool high) void PixelShaderManager::SetColorChanged(int type, int num, bool high)
{ {
float *pf = &lastRGBAfull[type][num][0]; float *pf = &lastRGBAfull[type][num][0];
if (!high) { if (!high) {
int r = bpmem.tevregs[num].low.a; int r = bpmem.tevregs[num].low.a;
int a = bpmem.tevregs[num].low.b; int a = bpmem.tevregs[num].low.b;
@ -371,45 +371,45 @@ void PixelShaderManager::SetColorChanged(int type, int num, bool high)
pf[1] = (float)g * (1.0f / 255.0f); pf[1] = (float)g * (1.0f / 255.0f);
pf[2] = (float)b * (1.0f / 255.0f); pf[2] = (float)b * (1.0f / 255.0f);
} }
s_nColorsChanged[type] |= 1 << num; s_nColorsChanged[type] |= 1 << num;
PRIM_LOG("pixel %scolor%d: %f %f %f %f\n", type?"k":"", num, pf[0], pf[1], pf[2], pf[3]); PRIM_LOG("pixel %scolor%d: %f %f %f %f\n", type?"k":"", num, pf[0], pf[1], pf[2], pf[3]);
} }
void PixelShaderManager::SetAlpha(const AlphaTest& alpha) void PixelShaderManager::SetAlpha(const AlphaTest& alpha)
{ {
if ((alpha.hex & 0xffff) != lastAlpha) if ((alpha.hex & 0xffff) != lastAlpha)
{ {
lastAlpha = (lastAlpha & ~0xffff) | (alpha.hex & 0xffff); lastAlpha = (lastAlpha & ~0xffff) | (alpha.hex & 0xffff);
s_bAlphaChanged = true; s_bAlphaChanged = true;
} }
} }
void PixelShaderManager::SetDestAlpha(const ConstantAlpha& alpha) void PixelShaderManager::SetDestAlpha(const ConstantAlpha& alpha)
{ {
if (alpha.alpha != (lastAlpha >> 16)) if (alpha.alpha != (lastAlpha >> 16))
{ {
lastAlpha = (lastAlpha & ~0xff0000) | ((alpha.hex & 0xff) << 16); lastAlpha = (lastAlpha & ~0xff0000) | ((alpha.hex & 0xff) << 16);
s_bAlphaChanged = true; s_bAlphaChanged = true;
} }
} }
void PixelShaderManager::SetTexDims(int texmapid, u32 width, u32 height, u32 wraps, u32 wrapt) void PixelShaderManager::SetTexDims(int texmapid, u32 width, u32 height, u32 wraps, u32 wrapt)
{ {
u32 wh = width | (height << 16) | (wraps << 28) | (wrapt << 30); u32 wh = width | (height << 16) | (wraps << 28) | (wrapt << 30);
if (lastTexDims[texmapid] != wh) if (lastTexDims[texmapid] != wh)
{ {
lastTexDims[texmapid] = wh; lastTexDims[texmapid] = wh;
s_nTexDimsChanged |= 1 << texmapid; s_nTexDimsChanged |= 1 << texmapid;
} }
} }
void PixelShaderManager::SetZTextureBias(u32 bias) void PixelShaderManager::SetZTextureBias(u32 bias)
{ {
if (lastZBias != bias) if (lastZBias != bias)
{ {
s_bZBiasChanged = true; s_bZBiasChanged = true;
lastZBias = bias; lastZBias = bias;
} }
} }
void PixelShaderManager::SetViewportChanged() void PixelShaderManager::SetViewportChanged()
@ -420,12 +420,12 @@ void PixelShaderManager::SetViewportChanged()
void PixelShaderManager::SetIndTexScaleChanged(u8 stagemask) void PixelShaderManager::SetIndTexScaleChanged(u8 stagemask)
{ {
s_nIndTexScaleChanged |= stagemask; s_nIndTexScaleChanged |= stagemask;
} }
void PixelShaderManager::SetIndMatrixChanged(int matrixidx) void PixelShaderManager::SetIndMatrixChanged(int matrixidx)
{ {
s_nIndTexMtxChanged |= 1 << matrixidx; s_nIndTexMtxChanged |= 1 << matrixidx;
} }
void PixelShaderManager::SetZTextureTypeChanged() void PixelShaderManager::SetZTextureTypeChanged()
@ -435,22 +435,22 @@ void PixelShaderManager::SetZTextureTypeChanged()
void PixelShaderManager::SetTexCoordChanged(u8 texmapid) void PixelShaderManager::SetTexCoordChanged(u8 texmapid)
{ {
s_nTexDimsChanged |= 1 << texmapid; s_nTexDimsChanged |= 1 << texmapid;
} }
void PixelShaderManager::SetFogColorChanged() void PixelShaderManager::SetFogColorChanged()
{ {
s_bFogColorChanged = true; s_bFogColorChanged = true;
} }
void PixelShaderManager::SetFogParamChanged() void PixelShaderManager::SetFogParamChanged()
{ {
s_bFogParamChanged = true; s_bFogParamChanged = true;
} }
void PixelShaderManager::SetFogRangeAdjustChanged() void PixelShaderManager::SetFogRangeAdjustChanged()
{ {
s_bFogRangeAdjustChanged = true; s_bFogRangeAdjustChanged = true;
} }
void PixelShaderManager::SetColorMatrix(const float* pmatrix) void PixelShaderManager::SetColorMatrix(const float* pmatrix)

1
Source/Core/VideoCommon/Src/PixelShaderManager.h
View File

@ -54,7 +54,6 @@ public:
static void SetColorMatrix(const float* pmatrix); static void SetColorMatrix(const float* pmatrix);
static void InvalidateXFRange(int start, int end); static void InvalidateXFRange(int start, int end);
static void SetMaterialColorChanged(int index); static void SetMaterialColorChanged(int index);
}; };

4
Source/Core/VideoCommon/Src/Statistics.cpp
View File

@ -53,8 +53,8 @@ char *Statistics::ToString(char *ptr)
ptr+=sprintf(ptr,"pshaders (unique, delete cache first): %i\n",stats.numUniquePixelShaders); ptr+=sprintf(ptr,"pshaders (unique, delete cache first): %i\n",stats.numUniquePixelShaders);
ptr+=sprintf(ptr,"vshaders created: %i\n",stats.numVertexShadersCreated); ptr+=sprintf(ptr,"vshaders created: %i\n",stats.numVertexShadersCreated);
ptr+=sprintf(ptr,"vshaders alive: %i\n",stats.numVertexShadersAlive); ptr+=sprintf(ptr,"vshaders alive: %i\n",stats.numVertexShadersAlive);
ptr+=sprintf(ptr,"dlists called: %i\n",stats.numDListsCalled); ptr+=sprintf(ptr,"dlists called: %i\n",stats.numDListsCalled);
ptr+=sprintf(ptr,"dlists called(f): %i\n",stats.thisFrame.numDListsCalled); ptr+=sprintf(ptr,"dlists called(f): %i\n",stats.thisFrame.numDListsCalled);
ptr+=sprintf(ptr,"dlists alive: %i\n",stats.numDListsAlive); ptr+=sprintf(ptr,"dlists alive: %i\n",stats.numDListsAlive);
ptr+=sprintf(ptr,"primitive joins: %i\n",stats.thisFrame.numPrimitiveJoins); ptr+=sprintf(ptr,"primitive joins: %i\n",stats.thisFrame.numPrimitiveJoins);
ptr+=sprintf(ptr,"draw calls: %i\n",stats.thisFrame.numDrawCalls); ptr+=sprintf(ptr,"draw calls: %i\n",stats.thisFrame.numDrawCalls);

66
Source/Core/VideoCommon/Src/Statistics.h
View File

@ -24,20 +24,20 @@
struct Statistics struct Statistics
{ {
int numPixelShadersCreated; int numPixelShadersCreated;
int numPixelShadersAlive; int numPixelShadersAlive;
int numVertexShadersCreated; int numVertexShadersCreated;
int numVertexShadersAlive; int numVertexShadersAlive;
int numTexturesCreated; int numTexturesCreated;
int numTexturesAlive; int numTexturesAlive;
int numRenderTargetsCreated; int numRenderTargetsCreated;
int numRenderTargetsAlive; int numRenderTargetsAlive;
int numDListsCalled; int numDListsCalled;
int numDListsCreated; int numDListsCreated;
int numDListsAlive; int numDListsAlive;
int numVertexLoaders; int numVertexLoaders;
@ -52,30 +52,30 @@ struct Statistics
std::vector<EFBRectangle> efb_regions; std::vector<EFBRectangle> efb_regions;
struct ThisFrame struct ThisFrame
{ {
int numBPLoads; int numBPLoads;
int numCPLoads; int numCPLoads;
int numXFLoads; int numXFLoads;
int numBPLoadsInDL; int numBPLoadsInDL;
int numCPLoadsInDL; int numCPLoadsInDL;
int numXFLoadsInDL; int numXFLoadsInDL;
int numDLs; int numDLs;
int numPrims; int numPrims;
int numDLPrims; int numDLPrims;
int numShaderChanges; int numShaderChanges;
int numPrimitiveJoins; int numPrimitiveJoins;
int numDrawCalls; int numDrawCalls;
int numIndexedDrawCalls; int numIndexedDrawCalls;
int numBufferSplits; int numBufferSplits;
int numDListsCalled; int numDListsCalled;
}; };
ThisFrame thisFrame; ThisFrame thisFrame;
void ResetFrame(); void ResetFrame();
static void SwapDL(); static void SwapDL();
// Yeah, this is unsafe, but we really don't wanna faff around allocating // Yeah, this is unsafe, but we really don't wanna faff around allocating

10
Source/Core/VideoCommon/Src/TextureCacheBase.cpp
View File

@ -56,7 +56,7 @@ TextureCache::TextureCache()
if (!temp) if (!temp)
temp = (u8*)AllocateAlignedMemory(temp_size, 16); temp = (u8*)AllocateAlignedMemory(temp_size, 16);
TexDecoder_SetTexFmtOverlayOptions(g_ActiveConfig.bTexFmtOverlayEnable, g_ActiveConfig.bTexFmtOverlayCenter); TexDecoder_SetTexFmtOverlayOptions(g_ActiveConfig.bTexFmtOverlayEnable, g_ActiveConfig.bTexFmtOverlayCenter);
if(g_ActiveConfig.bHiresTextures && !g_ActiveConfig.bDumpTextures) if(g_ActiveConfig.bHiresTextures && !g_ActiveConfig.bDumpTextures)
HiresTextures::Init(SConfig::GetInstance().m_LocalCoreStartupParameter.m_strUniqueID.c_str()); HiresTextures::Init(SConfig::GetInstance().m_LocalCoreStartupParameter.m_strUniqueID.c_str());
SetHash64Function(g_ActiveConfig.bHiresTextures || g_ActiveConfig.bDumpTextures); SetHash64Function(g_ActiveConfig.bHiresTextures || g_ActiveConfig.bDumpTextures);
} }
@ -725,7 +725,7 @@ void TextureCache::CopyRenderTargetToTexture(u32 dstAddr, unsigned int dstFormat
case 1: // R8 case 1: // R8
case 8: // R8 case 8: // R8
colmat[0] = colmat[4] = colmat[8] = colmat[12] = 1; colmat[0] = colmat[4] = colmat[8] = colmat[12] = 1;
cbufid = 13; cbufid = 13;
break; break;
case 2: // RA4 case 2: // RA4
@ -746,11 +746,11 @@ void TextureCache::CopyRenderTargetToTexture(u32 dstAddr, unsigned int dstFormat
case 9: // G8 case 9: // G8
colmat[1] = colmat[5] = colmat[9] = colmat[13] = 1.0f; colmat[1] = colmat[5] = colmat[9] = colmat[13] = 1.0f;
cbufid = 17; cbufid = 17;
break; break;
case 10: // B8 case 10: // B8
colmat[2] = colmat[6] = colmat[10] = colmat[14] = 1.0f; colmat[2] = colmat[6] = colmat[10] = colmat[14] = 1.0f;
cbufid = 18; cbufid = 18;
break; break;
case 11: // RG8 case 11: // RG8
@ -759,7 +759,7 @@ void TextureCache::CopyRenderTargetToTexture(u32 dstAddr, unsigned int dstFormat
break; break;
case 12: // GB8 case 12: // GB8
colmat[1] = colmat[5] = colmat[9] = colmat[14] = 1.0f; colmat[1] = colmat[5] = colmat[9] = colmat[14] = 1.0f;
cbufid = 20; cbufid = 20;
break; break;

245
Source/Core/VideoCommon/Src/TextureConversionShader.cpp
View File

@ -39,23 +39,23 @@ namespace TextureConversionShader
u16 GetEncodedSampleCount(u32 format) u16 GetEncodedSampleCount(u32 format)
{ {
switch (format) { switch (format) {
case GX_TF_I4: return 8; case GX_TF_I4: return 8;
case GX_TF_I8: return 4; case GX_TF_I8: return 4;
case GX_TF_IA4: return 4; case GX_TF_IA4: return 4;
case GX_TF_IA8: return 2; case GX_TF_IA8: return 2;
case GX_TF_RGB565: return 2; case GX_TF_RGB565: return 2;
case GX_TF_RGB5A3: return 2; case GX_TF_RGB5A3: return 2;
case GX_TF_RGBA8: return 1; case GX_TF_RGBA8: return 1;
case GX_CTF_R4: return 8; case GX_CTF_R4: return 8;
case GX_CTF_RA4: return 4; case GX_CTF_RA4: return 4;
case GX_CTF_RA8: return 2; case GX_CTF_RA8: return 2;
case GX_CTF_A8: return 4; case GX_CTF_A8: return 4;
case GX_CTF_R8: return 4; case GX_CTF_R8: return 4;
case GX_CTF_G8: return 4; case GX_CTF_G8: return 4;
case GX_CTF_B8: return 4; case GX_CTF_B8: return 4;
case GX_CTF_RG8: return 2; case GX_CTF_RG8: return 2;
case GX_CTF_GB8: return 2; case GX_CTF_GB8: return 2;
case GX_TF_Z8: return 4; case GX_TF_Z8: return 4;
case GX_TF_Z16: return 2; case GX_TF_Z16: return 2;
case GX_TF_Z24X8: return 1; case GX_TF_Z24X8: return 1;
@ -63,8 +63,8 @@ u16 GetEncodedSampleCount(u32 format)
case GX_CTF_Z8M: return 4; case GX_CTF_Z8M: return 4;
case GX_CTF_Z8L: return 4; case GX_CTF_Z8L: return 4;
case GX_CTF_Z16L: return 2; case GX_CTF_Z16L: return 2;
default: return 1; default: return 1;
} }
} }
const char* WriteRegister(API_TYPE ApiType, const char *prefix, const u32 num) const char* WriteRegister(API_TYPE ApiType, const char *prefix, const u32 num)
@ -100,10 +100,9 @@ void WriteSwizzler(char*& p, u32 format, API_TYPE ApiType)
else else
{ {
WRITE(p,"sampler samp0 : register(s0);\n"); WRITE(p,"sampler samp0 : register(s0);\n");
WRITE(p, "Texture2D Tex0 : register(t0);\n"); WRITE(p, "Texture2D Tex0 : register(t0);\n");
} }
if (ApiType == API_OPENGL) if (ApiType == API_OPENGL)
{ {
WRITE(p, " out float4 ocol0;\n"); WRITE(p, " out float4 ocol0;\n");
@ -123,9 +122,9 @@ void WriteSwizzler(char*& p, u32 format, API_TYPE ApiType)
} }
WRITE(p," in float2 uv0 : TEXCOORD0)\n"); WRITE(p," in float2 uv0 : TEXCOORD0)\n");
} }
WRITE(p, "{\n" WRITE(p, "{\n"
" float2 sampleUv;\n" " float2 sampleUv;\n"
" float2 uv1 = floor(uv0);\n"); " float2 uv1 = floor(uv0);\n");
WRITE(p, " uv1.x = uv1.x * %f;\n", samples); WRITE(p, " uv1.x = uv1.x * %f;\n", samples);
@ -147,14 +146,14 @@ void WriteSwizzler(char*& p, u32 format, API_TYPE ApiType)
if (ApiType == API_OPENGL) if (ApiType == API_OPENGL)
WRITE(p," sampleUv.y = " I_COLORS"[1].y - sampleUv.y;\n"); WRITE(p," sampleUv.y = " I_COLORS"[1].y - sampleUv.y;\n");
WRITE(p, " sampleUv = sampleUv + " I_COLORS"[1].zw;\n"); WRITE(p, " sampleUv = sampleUv + " I_COLORS"[1].zw;\n");
if (ApiType != API_OPENGL) if (ApiType != API_OPENGL)
{ {
WRITE(p, " sampleUv = sampleUv + float2(0.0f,1.0f);\n");// still to determine the reason for this WRITE(p, " sampleUv = sampleUv + float2(0.0f,1.0f);\n");// still to determine the reason for this
WRITE(p, " sampleUv = sampleUv / " I_COLORS"[0].zw;\n"); WRITE(p, " sampleUv = sampleUv / " I_COLORS"[0].zw;\n");
} }
} }
// block dimensions : widthStride, heightStride // block dimensions : widthStride, heightStride
@ -182,7 +181,7 @@ void Write32BitSwizzler(char*& p, u32 format, API_TYPE ApiType)
else else
{ {
WRITE(p,"sampler samp0 : register(s0);\n"); WRITE(p,"sampler samp0 : register(s0);\n");
WRITE(p, "Texture2D Tex0 : register(t0);\n"); WRITE(p, "Texture2D Tex0 : register(t0);\n");
} }
if (ApiType == API_OPENGL) if (ApiType == API_OPENGL)
@ -204,12 +203,12 @@ void Write32BitSwizzler(char*& p, u32 format, API_TYPE ApiType)
} }
WRITE(p," in float2 uv0 : TEXCOORD0)\n"); WRITE(p," in float2 uv0 : TEXCOORD0)\n");
} }
WRITE(p, "{\n" WRITE(p, "{\n"
" float2 sampleUv;\n" " float2 sampleUv;\n"
" float2 uv1 = floor(uv0);\n"); " float2 uv1 = floor(uv0);\n");
WRITE(p, " float yl = floor(uv1.y / %f);\n", blkH); WRITE(p, " float yl = floor(uv1.y / %f);\n", blkH);
WRITE(p, " float yb = yl * %f;\n", blkH); WRITE(p, " float yb = yl * %f;\n", blkH);
WRITE(p, " float yoff = uv1.y - yb;\n"); WRITE(p, " float yoff = uv1.y - yb;\n");
@ -217,26 +216,26 @@ void Write32BitSwizzler(char*& p, u32 format, API_TYPE ApiType)
WRITE(p, " float xel = floor(xp / 2);\n"); WRITE(p, " float xel = floor(xp / 2);\n");
WRITE(p, " float xb = floor(xel / %f);\n", blkH); WRITE(p, " float xb = floor(xel / %f);\n", blkH);
WRITE(p, " float xoff = xel - (xb * %f);\n", blkH); WRITE(p, " float xoff = xel - (xb * %f);\n", blkH);
WRITE(p, " float x2 = uv1.x * 2;\n"); WRITE(p, " float x2 = uv1.x * 2;\n");
WRITE(p, " float xl = floor(x2 / %f);\n", blkW); WRITE(p, " float xl = floor(x2 / %f);\n", blkW);
WRITE(p, " float xib = x2 - (xl * %f);\n", blkW); WRITE(p, " float xib = x2 - (xl * %f);\n", blkW);
WRITE(p, " float halfxb = floor(xb / 2);\n"); WRITE(p, " float halfxb = floor(xb / 2);\n");
WRITE(p, " sampleUv.x = xib + (halfxb * %f);\n", blkW); WRITE(p, " sampleUv.x = xib + (halfxb * %f);\n", blkW);
WRITE(p, " sampleUv.y = yb + xoff;\n"); WRITE(p, " sampleUv.y = yb + xoff;\n");
WRITE(p, " sampleUv = sampleUv * " I_COLORS"[0].xy;\n"); WRITE(p, " sampleUv = sampleUv * " I_COLORS"[0].xy;\n");
if (ApiType == API_OPENGL) if (ApiType == API_OPENGL)
WRITE(p," sampleUv.y = " I_COLORS"[1].y - sampleUv.y;\n"); WRITE(p," sampleUv.y = " I_COLORS"[1].y - sampleUv.y;\n");
WRITE(p, " sampleUv = sampleUv + " I_COLORS"[1].zw;\n"); WRITE(p, " sampleUv = sampleUv + " I_COLORS"[1].zw;\n");
if (ApiType != API_OPENGL) if (ApiType != API_OPENGL)
{ {
WRITE(p, " sampleUv = sampleUv + float2(0.0f,1.0f);\n");// still to determine the reason for this WRITE(p, " sampleUv = sampleUv + float2(0.0f,1.0f);\n");// still to determine the reason for this
WRITE(p, " sampleUv = sampleUv / " I_COLORS"[0].zw;\n"); WRITE(p, " sampleUv = sampleUv / " I_COLORS"[0].zw;\n");
} }
} }
void WriteSampleColor(char*& p, const char* colorComp, const char* dest, API_TYPE ApiType) void WriteSampleColor(char*& p, const char* colorComp, const char* dest, API_TYPE ApiType)
@ -465,11 +464,11 @@ void WriteRGB5A3Encoder(char* p,API_TYPE ApiType)
WriteSampleColor(p, "rgba", "texSample", ApiType); WriteSampleColor(p, "rgba", "texSample", ApiType);
// 0.8784 = 224 / 255 which is the maximum alpha value that can be represented in 3 bits // 0.8784 = 224 / 255 which is the maximum alpha value that can be represented in 3 bits
WRITE(p, "if(texSample.a > 0.878f) {\n"); WRITE(p, "if(texSample.a > 0.878f) {\n");
WriteToBitDepth(p, 5, "texSample.g", "color0"); WriteToBitDepth(p, 5, "texSample.g", "color0");
WRITE(p, " gUpper = floor(color0 / 8.0f);\n"); WRITE(p, " gUpper = floor(color0 / 8.0f);\n");
WRITE(p, " gLower = color0 - gUpper * 8.0f;\n"); WRITE(p, " gLower = color0 - gUpper * 8.0f;\n");
WriteToBitDepth(p, 5, "texSample.r", "ocol0.b"); WriteToBitDepth(p, 5, "texSample.r", "ocol0.b");
@ -477,27 +476,27 @@ void WriteRGB5A3Encoder(char* p,API_TYPE ApiType)
WriteToBitDepth(p, 5, "texSample.b", "ocol0.g"); WriteToBitDepth(p, 5, "texSample.b", "ocol0.g");
WRITE(p, " ocol0.g = ocol0.g + gLower * 32.0f;\n"); WRITE(p, " ocol0.g = ocol0.g + gLower * 32.0f;\n");
WRITE(p, "} else {\n"); WRITE(p, "} else {\n");
WriteToBitDepth(p, 4, "texSample.r", "ocol0.b"); WriteToBitDepth(p, 4, "texSample.r", "ocol0.b");
WriteToBitDepth(p, 4, "texSample.b", "ocol0.g"); WriteToBitDepth(p, 4, "texSample.b", "ocol0.g");
WriteToBitDepth(p, 3, "texSample.a", "color0"); WriteToBitDepth(p, 3, "texSample.a", "color0");
WRITE(p, "ocol0.b = ocol0.b + color0 * 16.0f;\n"); WRITE(p, "ocol0.b = ocol0.b + color0 * 16.0f;\n");
WriteToBitDepth(p, 4, "texSample.g", "color0"); WriteToBitDepth(p, 4, "texSample.g", "color0");
WRITE(p, "ocol0.g = ocol0.g + color0 * 16.0f;\n"); WRITE(p, "ocol0.g = ocol0.g + color0 * 16.0f;\n");
WRITE(p, "}\n"); WRITE(p, "}\n");
WriteIncrementSampleX(p, ApiType); WriteIncrementSampleX(p, ApiType);
WriteSampleColor(p, "rgba", "texSample", ApiType); WriteSampleColor(p, "rgba", "texSample", ApiType);
WRITE(p, "if(texSample.a > 0.878f) {\n"); WRITE(p, "if(texSample.a > 0.878f) {\n");
WriteToBitDepth(p, 5, "texSample.g", "color0"); WriteToBitDepth(p, 5, "texSample.g", "color0");
WRITE(p, " gUpper = floor(color0 / 8.0f);\n"); WRITE(p, " gUpper = floor(color0 / 8.0f);\n");
WRITE(p, " gLower = color0 - gUpper * 8.0f;\n"); WRITE(p, " gLower = color0 - gUpper * 8.0f;\n");
WriteToBitDepth(p, 5, "texSample.r", "ocol0.r"); WriteToBitDepth(p, 5, "texSample.r", "ocol0.r");
@ -505,17 +504,17 @@ void WriteRGB5A3Encoder(char* p,API_TYPE ApiType)
WriteToBitDepth(p, 5, "texSample.b", "ocol0.a"); WriteToBitDepth(p, 5, "texSample.b", "ocol0.a");
WRITE(p, " ocol0.a = ocol0.a + gLower * 32.0f;\n"); WRITE(p, " ocol0.a = ocol0.a + gLower * 32.0f;\n");
WRITE(p, "} else {\n"); WRITE(p, "} else {\n");
WriteToBitDepth(p, 4, "texSample.r", "ocol0.r"); WriteToBitDepth(p, 4, "texSample.r", "ocol0.r");
WriteToBitDepth(p, 4, "texSample.b", "ocol0.a"); WriteToBitDepth(p, 4, "texSample.b", "ocol0.a");
WriteToBitDepth(p, 3, "texSample.a", "color0"); WriteToBitDepth(p, 3, "texSample.a", "color0");
WRITE(p, "ocol0.r = ocol0.r + color0 * 16.0f;\n"); WRITE(p, "ocol0.r = ocol0.r + color0 * 16.0f;\n");
WriteToBitDepth(p, 4, "texSample.g", "color0"); WriteToBitDepth(p, 4, "texSample.g", "color0");
WRITE(p, "ocol0.a = ocol0.a + color0 * 16.0f;\n"); WRITE(p, "ocol0.a = ocol0.a + color0 * 16.0f;\n");
WRITE(p, "}\n"); WRITE(p, "}\n");
WRITE(p, " ocol0 = ocol0 / 255.0f;\n"); WRITE(p, " ocol0 = ocol0 / 255.0f;\n");
WriteEncoderEnd(p, ApiType); WriteEncoderEnd(p, ApiType);
@ -680,96 +679,96 @@ void WriteZ8Encoder(char* p, const char* multiplier,API_TYPE ApiType)
{ {
WriteSwizzler(p, GX_CTF_Z8M, ApiType); WriteSwizzler(p, GX_CTF_Z8M, ApiType);
WRITE(p, " float depth;\n"); WRITE(p, " float depth;\n");
WriteSampleColor(p, "b", "depth", ApiType); WriteSampleColor(p, "b", "depth", ApiType);
WRITE(p, "ocol0.b = frac(depth * %s);\n", multiplier); WRITE(p, "ocol0.b = frac(depth * %s);\n", multiplier);
WriteIncrementSampleX(p, ApiType); WriteIncrementSampleX(p, ApiType);
WriteSampleColor(p, "b", "depth", ApiType); WriteSampleColor(p, "b", "depth", ApiType);
WRITE(p, "ocol0.g = frac(depth * %s);\n", multiplier); WRITE(p, "ocol0.g = frac(depth * %s);\n", multiplier);
WriteIncrementSampleX(p, ApiType); WriteIncrementSampleX(p, ApiType);
WriteSampleColor(p, "b", "depth", ApiType); WriteSampleColor(p, "b", "depth", ApiType);
WRITE(p, "ocol0.r = frac(depth * %s);\n", multiplier); WRITE(p, "ocol0.r = frac(depth * %s);\n", multiplier);
WriteIncrementSampleX(p, ApiType); WriteIncrementSampleX(p, ApiType);
WriteSampleColor(p, "b", "depth", ApiType); WriteSampleColor(p, "b", "depth", ApiType);
WRITE(p, "ocol0.a = frac(depth * %s);\n", multiplier); WRITE(p, "ocol0.a = frac(depth * %s);\n", multiplier);
WriteEncoderEnd(p, ApiType); WriteEncoderEnd(p, ApiType);
} }
void WriteZ16Encoder(char* p,API_TYPE ApiType) void WriteZ16Encoder(char* p,API_TYPE ApiType)
{ {
WriteSwizzler(p, GX_TF_Z16, ApiType); WriteSwizzler(p, GX_TF_Z16, ApiType);
WRITE(p, " float depth;\n"); WRITE(p, " float depth;\n");
WRITE(p, " float3 expanded;\n"); WRITE(p, " float3 expanded;\n");
// byte order is reversed // byte order is reversed
WriteSampleColor(p, "b", "depth", ApiType); WriteSampleColor(p, "b", "depth", ApiType);
WRITE(p, " depth *= 16777215.0f;\n"); WRITE(p, " depth *= 16777215.0f;\n");
WRITE(p, " expanded.r = floor(depth / (256 * 256));\n"); WRITE(p, " expanded.r = floor(depth / (256 * 256));\n");
WRITE(p, " depth -= expanded.r * 256 * 256;\n"); WRITE(p, " depth -= expanded.r * 256 * 256;\n");
WRITE(p, " expanded.g = floor(depth / 256);\n"); WRITE(p, " expanded.g = floor(depth / 256);\n");
WRITE(p, " ocol0.b = expanded.g / 255;\n"); WRITE(p, " ocol0.b = expanded.g / 255;\n");
WRITE(p, " ocol0.g = expanded.r / 255;\n"); WRITE(p, " ocol0.g = expanded.r / 255;\n");
WriteIncrementSampleX(p, ApiType); WriteIncrementSampleX(p, ApiType);
WriteSampleColor(p, "b", "depth", ApiType); WriteSampleColor(p, "b", "depth", ApiType);
WRITE(p, " depth *= 16777215.0f;\n"); WRITE(p, " depth *= 16777215.0f;\n");
WRITE(p, " expanded.r = floor(depth / (256 * 256));\n"); WRITE(p, " expanded.r = floor(depth / (256 * 256));\n");
WRITE(p, " depth -= expanded.r * 256 * 256;\n"); WRITE(p, " depth -= expanded.r * 256 * 256;\n");
WRITE(p, " expanded.g = floor(depth / 256);\n"); WRITE(p, " expanded.g = floor(depth / 256);\n");
WRITE(p, " ocol0.r = expanded.g / 255;\n"); WRITE(p, " ocol0.r = expanded.g / 255;\n");
WRITE(p, " ocol0.a = expanded.r / 255;\n"); WRITE(p, " ocol0.a = expanded.r / 255;\n");
WriteEncoderEnd(p, ApiType); WriteEncoderEnd(p, ApiType);
} }
void WriteZ16LEncoder(char* p,API_TYPE ApiType) void WriteZ16LEncoder(char* p,API_TYPE ApiType)
{ {
WriteSwizzler(p, GX_CTF_Z16L, ApiType); WriteSwizzler(p, GX_CTF_Z16L, ApiType);
WRITE(p, " float depth;\n"); WRITE(p, " float depth;\n");
WRITE(p, " float3 expanded;\n"); WRITE(p, " float3 expanded;\n");
// byte order is reversed // byte order is reversed
WriteSampleColor(p, "b", "depth", ApiType); WriteSampleColor(p, "b", "depth", ApiType);
WRITE(p, " depth *= 16777215.0f;\n"); WRITE(p, " depth *= 16777215.0f;\n");
WRITE(p, " expanded.r = floor(depth / (256 * 256));\n"); WRITE(p, " expanded.r = floor(depth / (256 * 256));\n");
WRITE(p, " depth -= expanded.r * 256 * 256;\n"); WRITE(p, " depth -= expanded.r * 256 * 256;\n");
WRITE(p, " expanded.g = floor(depth / 256);\n"); WRITE(p, " expanded.g = floor(depth / 256);\n");
WRITE(p, " depth -= expanded.g * 256;\n"); WRITE(p, " depth -= expanded.g * 256;\n");
WRITE(p, " expanded.b = depth;\n"); WRITE(p, " expanded.b = depth;\n");
WRITE(p, " ocol0.b = expanded.b / 255;\n"); WRITE(p, " ocol0.b = expanded.b / 255;\n");
WRITE(p, " ocol0.g = expanded.g / 255;\n"); WRITE(p, " ocol0.g = expanded.g / 255;\n");
WriteIncrementSampleX(p, ApiType); WriteIncrementSampleX(p, ApiType);
WriteSampleColor(p, "b", "depth", ApiType); WriteSampleColor(p, "b", "depth", ApiType);
WRITE(p, " depth *= 16777215.0f;\n"); WRITE(p, " depth *= 16777215.0f;\n");
WRITE(p, " expanded.r = floor(depth / (256 * 256));\n"); WRITE(p, " expanded.r = floor(depth / (256 * 256));\n");
WRITE(p, " depth -= expanded.r * 256 * 256;\n"); WRITE(p, " depth -= expanded.r * 256 * 256;\n");
WRITE(p, " expanded.g = floor(depth / 256);\n"); WRITE(p, " expanded.g = floor(depth / 256);\n");
WRITE(p, " depth -= expanded.g * 256;\n"); WRITE(p, " depth -= expanded.g * 256;\n");
WRITE(p, " expanded.b = depth;\n"); WRITE(p, " expanded.b = depth;\n");
WRITE(p, " ocol0.r = expanded.b;\n"); WRITE(p, " ocol0.r = expanded.b;\n");
WRITE(p, " ocol0.a = expanded.g;\n"); WRITE(p, " ocol0.a = expanded.g;\n");
WriteEncoderEnd(p, ApiType); WriteEncoderEnd(p, ApiType);
} }
void WriteZ24Encoder(char* p, API_TYPE ApiType) void WriteZ24Encoder(char* p, API_TYPE ApiType)
@ -779,16 +778,16 @@ void WriteZ24Encoder(char* p, API_TYPE ApiType)
WRITE(p, " float cl = xb - (halfxb * 2);\n"); WRITE(p, " float cl = xb - (halfxb * 2);\n");
WRITE(p, " float depth0;\n"); WRITE(p, " float depth0;\n");
WRITE(p, " float depth1;\n"); WRITE(p, " float depth1;\n");
WRITE(p, " float3 expanded0;\n"); WRITE(p, " float3 expanded0;\n");
WRITE(p, " float3 expanded1;\n"); WRITE(p, " float3 expanded1;\n");
WriteSampleColor(p, "b", "depth0", ApiType); WriteSampleColor(p, "b", "depth0", ApiType);
WriteIncrementSampleX(p, ApiType); WriteIncrementSampleX(p, ApiType);
WriteSampleColor(p, "b", "depth1", ApiType); WriteSampleColor(p, "b", "depth1", ApiType);
for (int i = 0; i < 2; i++) for (int i = 0; i < 2; i++)
{ {
WRITE(p, " depth%i *= 16777215.0f;\n", i); WRITE(p, " depth%i *= 16777215.0f;\n", i);
WRITE(p, " expanded%i.r = floor(depth%i / (256 * 256));\n", i, i); WRITE(p, " expanded%i.r = floor(depth%i / (256 * 256));\n", i, i);
@ -796,23 +795,23 @@ void WriteZ24Encoder(char* p, API_TYPE ApiType)
WRITE(p, " expanded%i.g = floor(depth%i / 256);\n", i, i); WRITE(p, " expanded%i.g = floor(depth%i / 256);\n", i, i);
WRITE(p, " depth%i -= expanded%i.g * 256;\n", i, i); WRITE(p, " depth%i -= expanded%i.g * 256;\n", i, i);
WRITE(p, " expanded%i.b = depth%i;\n", i, i); WRITE(p, " expanded%i.b = depth%i;\n", i, i);
} }
WRITE(p, " if(cl > 0.5f) {\n"); WRITE(p, " if(cl > 0.5f) {\n");
// upper 16 // upper 16
WRITE(p, " ocol0.b = expanded0.g / 255;\n"); WRITE(p, " ocol0.b = expanded0.g / 255;\n");
WRITE(p, " ocol0.g = expanded0.b / 255;\n"); WRITE(p, " ocol0.g = expanded0.b / 255;\n");
WRITE(p, " ocol0.r = expanded1.g / 255;\n"); WRITE(p, " ocol0.r = expanded1.g / 255;\n");
WRITE(p, " ocol0.a = expanded1.b / 255;\n"); WRITE(p, " ocol0.a = expanded1.b / 255;\n");
WRITE(p, " } else {\n"); WRITE(p, " } else {\n");
// lower 8 // lower 8
WRITE(p, " ocol0.b = 1.0f;\n"); WRITE(p, " ocol0.b = 1.0f;\n");
WRITE(p, " ocol0.g = expanded0.r / 255;\n"); WRITE(p, " ocol0.g = expanded0.r / 255;\n");
WRITE(p, " ocol0.r = 1.0f;\n"); WRITE(p, " ocol0.r = 1.0f;\n");
WRITE(p, " ocol0.a = expanded1.r / 255;\n"); WRITE(p, " ocol0.a = expanded1.r / 255;\n");
WRITE(p, " }\n"); WRITE(p, " }\n");
WriteEncoderEnd(p, ApiType); WriteEncoderEnd(p, ApiType);
} }
const char *GenerateEncodingShader(u32 format,API_TYPE ApiType) const char *GenerateEncodingShader(u32 format,API_TYPE ApiType)
@ -902,7 +901,7 @@ const char *GenerateEncodingShader(u32 format,API_TYPE ApiType)
PanicAlert("TextureConversionShader generator - buffer too small, canary has been eaten!"); PanicAlert("TextureConversionShader generator - buffer too small, canary has been eaten!");
setlocale(LC_NUMERIC, ""); // restore locale setlocale(LC_NUMERIC, ""); // restore locale
return text; return text;
} }
void SetShaderParameters(float width, float height, float offsetX, float offsetY, float widthStride, float heightStride,float buffW,float buffH) void SetShaderParameters(float width, float height, float offsetX, float offsetY, float widthStride, float heightStride,float buffW,float buffH)

62
Source/Core/VideoCommon/Src/TextureDecoder.h
View File

@ -27,41 +27,41 @@ extern GC_ALIGNED16(u8 texMem[TMEM_SIZE]);
enum TextureFormat enum TextureFormat
{ {
GX_TF_I4 = 0x0, GX_TF_I4 = 0x0,
GX_TF_I8 = 0x1, GX_TF_I8 = 0x1,
GX_TF_IA4 = 0x2, GX_TF_IA4 = 0x2,
GX_TF_IA8 = 0x3, GX_TF_IA8 = 0x3,
GX_TF_RGB565 = 0x4, GX_TF_RGB565 = 0x4,
GX_TF_RGB5A3 = 0x5, GX_TF_RGB5A3 = 0x5,
GX_TF_RGBA8 = 0x6, GX_TF_RGBA8 = 0x6,
GX_TF_C4 = 0x8, GX_TF_C4 = 0x8,
GX_TF_C8 = 0x9, GX_TF_C8 = 0x9,
GX_TF_C14X2 = 0xA, GX_TF_C14X2 = 0xA,
GX_TF_CMPR = 0xE, GX_TF_CMPR = 0xE,
_GX_TF_CTF = 0x20, // copy-texture-format only (simply means linear?) _GX_TF_CTF = 0x20, // copy-texture-format only (simply means linear?)
_GX_TF_ZTF = 0x10, // Z-texture-format _GX_TF_ZTF = 0x10, // Z-texture-format
// these formats are also valid when copying targets // these formats are also valid when copying targets
GX_CTF_R4 = 0x0 | _GX_TF_CTF, GX_CTF_R4 = 0x0 | _GX_TF_CTF,
GX_CTF_RA4 = 0x2 | _GX_TF_CTF, GX_CTF_RA4 = 0x2 | _GX_TF_CTF,
GX_CTF_RA8 = 0x3 | _GX_TF_CTF, GX_CTF_RA8 = 0x3 | _GX_TF_CTF,
GX_CTF_YUVA8 = 0x6 | _GX_TF_CTF, GX_CTF_YUVA8 = 0x6 | _GX_TF_CTF,
GX_CTF_A8 = 0x7 | _GX_TF_CTF, GX_CTF_A8 = 0x7 | _GX_TF_CTF,
GX_CTF_R8 = 0x8 | _GX_TF_CTF, GX_CTF_R8 = 0x8 | _GX_TF_CTF,
GX_CTF_G8 = 0x9 | _GX_TF_CTF, GX_CTF_G8 = 0x9 | _GX_TF_CTF,
GX_CTF_B8 = 0xA | _GX_TF_CTF, GX_CTF_B8 = 0xA | _GX_TF_CTF,
GX_CTF_RG8 = 0xB | _GX_TF_CTF, GX_CTF_RG8 = 0xB | _GX_TF_CTF,
GX_CTF_GB8 = 0xC | _GX_TF_CTF, GX_CTF_GB8 = 0xC | _GX_TF_CTF,
GX_TF_Z8 = 0x1 | _GX_TF_ZTF, GX_TF_Z8 = 0x1 | _GX_TF_ZTF,
GX_TF_Z16 = 0x3 | _GX_TF_ZTF, GX_TF_Z16 = 0x3 | _GX_TF_ZTF,
GX_TF_Z24X8 = 0x6 | _GX_TF_ZTF, GX_TF_Z24X8 = 0x6 | _GX_TF_ZTF,
GX_CTF_Z4 = 0x0 | _GX_TF_ZTF | _GX_TF_CTF, GX_CTF_Z4 = 0x0 | _GX_TF_ZTF | _GX_TF_CTF,
GX_CTF_Z8M = 0x9 | _GX_TF_ZTF | _GX_TF_CTF, GX_CTF_Z8M = 0x9 | _GX_TF_ZTF | _GX_TF_CTF,
GX_CTF_Z8L = 0xA | _GX_TF_ZTF | _GX_TF_CTF, GX_CTF_Z8L = 0xA | _GX_TF_ZTF | _GX_TF_CTF,
GX_CTF_Z16L = 0xC | _GX_TF_ZTF | _GX_TF_CTF, GX_CTF_Z16L = 0xC | _GX_TF_ZTF | _GX_TF_CTF,
}; };
int TexDecoder_GetTexelSizeInNibbles(int format); int TexDecoder_GetTexelSizeInNibbles(int format);

16
Source/Core/VideoCommon/Src/VertexLoader.cpp
View File

@ -121,10 +121,10 @@ void LOADERDECL UpdateBoundingBox()
{ {
if (!PixelEngine::bbox_active) if (!PixelEngine::bbox_active)
return; return;
// reset videodata pointer // reset videodata pointer
VertexManager::s_pCurBufferPointer = s_bbox_pCurBufferPointer_orig; VertexManager::s_pCurBufferPointer = s_bbox_pCurBufferPointer_orig;
// copy vertex pointers // copy vertex pointers
memcpy(VertexManager::s_pCurBufferPointer, s_bbox_vertex_buffer, 12); memcpy(VertexManager::s_pCurBufferPointer, s_bbox_vertex_buffer, 12);
VertexManager::s_pCurBufferPointer += 12; VertexManager::s_pCurBufferPointer += 12;
@ -335,7 +335,7 @@ void VertexLoader::CompileVertexTranslator()
nat_offset += 12; nat_offset += 12;
vtx_decl.normal_offset[2] = nat_offset; vtx_decl.normal_offset[2] = nat_offset;
nat_offset += 12; nat_offset += 12;
} }
int numNormals = (m_VtxAttr.NormalElements == 1) ? NRM_THREE : NRM_ONE; int numNormals = (m_VtxAttr.NormalElements == 1) ? NRM_THREE : NRM_ONE;
m_NativeFmt->m_components |= VB_HAS_NRM0; m_NativeFmt->m_components |= VB_HAS_NRM0;
@ -367,7 +367,7 @@ void VertexLoader::CompileVertexTranslator()
default: _assert_(0); break; default: _assert_(0); break;
} }
break; break;
case INDEX8: case INDEX8:
m_VertexSize += 1; m_VertexSize += 1;
switch (m_VtxAttr.color[i].Comp) switch (m_VtxAttr.color[i].Comp)
{ {
@ -558,7 +558,7 @@ int VertexLoader::SetupRunVertices(int vtx_attr_group, int primitive, int const
m_VtxAttr.texCoord[0].Frac = g_VtxAttr[vtx_attr_group].g0.Tex0Frac; m_VtxAttr.texCoord[0].Frac = g_VtxAttr[vtx_attr_group].g0.Tex0Frac;
m_VtxAttr.texCoord[1].Frac = g_VtxAttr[vtx_attr_group].g1.Tex1Frac; m_VtxAttr.texCoord[1].Frac = g_VtxAttr[vtx_attr_group].g1.Tex1Frac;
m_VtxAttr.texCoord[2].Frac = g_VtxAttr[vtx_attr_group].g1.Tex2Frac; m_VtxAttr.texCoord[2].Frac = g_VtxAttr[vtx_attr_group].g1.Tex2Frac;
m_VtxAttr.texCoord[3].Frac = g_VtxAttr[vtx_attr_group].g1.Tex3Frac; m_VtxAttr.texCoord[3].Frac = g_VtxAttr[vtx_attr_group].g1.Tex3Frac;
m_VtxAttr.texCoord[4].Frac = g_VtxAttr[vtx_attr_group].g2.Tex4Frac; m_VtxAttr.texCoord[4].Frac = g_VtxAttr[vtx_attr_group].g2.Tex4Frac;
m_VtxAttr.texCoord[5].Frac = g_VtxAttr[vtx_attr_group].g2.Tex5Frac; m_VtxAttr.texCoord[5].Frac = g_VtxAttr[vtx_attr_group].g2.Tex5Frac;
m_VtxAttr.texCoord[6].Frac = g_VtxAttr[vtx_attr_group].g2.Tex6Frac; m_VtxAttr.texCoord[6].Frac = g_VtxAttr[vtx_attr_group].g2.Tex6Frac;
@ -607,11 +607,11 @@ void VertexLoader::ConvertVertices ( int count )
void VertexLoader::RunCompiledVertices(int vtx_attr_group, int primitive, int const count, u8* Data) void VertexLoader::RunCompiledVertices(int vtx_attr_group, int primitive, int const count, u8* Data)
{ {
auto const new_count = SetupRunVertices(vtx_attr_group, primitive, count); auto const new_count = SetupRunVertices(vtx_attr_group, primitive, count);
memcpy_gc(VertexManager::s_pCurBufferPointer, Data, native_stride * new_count); memcpy_gc(VertexManager::s_pCurBufferPointer, Data, native_stride * new_count);
VertexManager::s_pCurBufferPointer += native_stride * new_count; VertexManager::s_pCurBufferPointer += native_stride * new_count;
DataSkip(new_count * m_VertexSize); DataSkip(new_count * m_VertexSize);
VertexManager::AddVertices(primitive, new_count); VertexManager::AddVertices(primitive, new_count);
} }
@ -645,7 +645,7 @@ void VertexLoader::SetVAT(u32 _group0, u32 _group1, u32 _group2)
m_VtxAttr.texCoord[2].Frac = vat.g1.Tex2Frac; m_VtxAttr.texCoord[2].Frac = vat.g1.Tex2Frac;
m_VtxAttr.texCoord[3].Elements = vat.g1.Tex3CoordElements; m_VtxAttr.texCoord[3].Elements = vat.g1.Tex3CoordElements;
m_VtxAttr.texCoord[3].Format = vat.g1.Tex3CoordFormat; m_VtxAttr.texCoord[3].Format = vat.g1.Tex3CoordFormat;
m_VtxAttr.texCoord[3].Frac = vat.g1.Tex3Frac; m_VtxAttr.texCoord[3].Frac = vat.g1.Tex3Frac;
m_VtxAttr.texCoord[4].Elements = vat.g1.Tex4CoordElements; m_VtxAttr.texCoord[4].Elements = vat.g1.Tex4CoordElements;
m_VtxAttr.texCoord[4].Format = vat.g1.Tex4CoordFormat; m_VtxAttr.texCoord[4].Format = vat.g1.Tex4CoordFormat;

2
Source/Core/VideoCommon/Src/VertexLoader.h
View File

@ -134,6 +134,6 @@ private:
void WriteGetVariable(int bits, Gen::OpArg dest, void *address); void WriteGetVariable(int bits, Gen::OpArg dest, void *address);
void WriteSetVariable(int bits, void *address, Gen::OpArg dest); void WriteSetVariable(int bits, void *address, Gen::OpArg dest);
#endif #endif
}; };
#endif #endif

10
Source/Core/VideoCommon/Src/VertexLoader_Normal.cpp
View File

@ -46,7 +46,7 @@ __forceinline float FracAdjust(T val)
//auto const U8FRAC = 1.f / (1u << 7); //auto const U8FRAC = 1.f / (1u << 7);
//auto const S16FRAC = 1.f / (1u << 14); //auto const S16FRAC = 1.f / (1u << 14);
//auto const U16FRAC = 1.f / (1u << 15); //auto const U16FRAC = 1.f / (1u << 15);
// TODO: is this right? // TODO: is this right?
return val / float(1u << (sizeof(T) * 8 - std::numeric_limits<T>::is_signed - 1)); return val / float(1u << (sizeof(T) * 8 - std::numeric_limits<T>::is_signed - 1));
} }
@ -64,7 +64,7 @@ __forceinline void ReadIndirect(const T* data)
{ {
DataWrite(FracAdjust(Common::FromBigEndian(data[i]))); DataWrite(FracAdjust(Common::FromBigEndian(data[i])));
} }
LOG_NORM(); LOG_NORM();
} }
@ -77,7 +77,7 @@ struct Normal_Direct
ReadIndirect<T, N * 3>(source); ReadIndirect<T, N * 3>(source);
DataSkip<N * 3 * sizeof(T)>(); DataSkip<N * 3 * sizeof(T)>();
} }
static const int size = sizeof(T) * N * 3; static const int size = sizeof(T) * N * 3;
}; };
@ -99,7 +99,7 @@ struct Normal_Index
{ {
Normal_Index_Offset<I, T, N, 0>(); Normal_Index_Offset<I, T, N, 0>();
} }
static const int size = sizeof(I); static const int size = sizeof(I);
}; };
@ -112,7 +112,7 @@ struct Normal_Index_Indices3
Normal_Index_Offset<I, T, 1, 1>(); Normal_Index_Offset<I, T, 1, 1>();
Normal_Index_Offset<I, T, 1, 2>(); Normal_Index_Offset<I, T, 1, 2>();
} }
static const int size = sizeof(I) * 3; static const int size = sizeof(I) * 3;
}; };

58
Source/Core/VideoCommon/Src/VertexLoader_Normal.h
View File

@ -25,43 +25,43 @@ class VertexLoader_Normal
{ {
public: public:
// Init // Init
static void Init(void); static void Init(void);
// GetSize // GetSize
static unsigned int GetSize(unsigned int _type, unsigned int _format, static unsigned int GetSize(unsigned int _type, unsigned int _format,
unsigned int _elements, unsigned int _index3); unsigned int _elements, unsigned int _index3);
// GetFunction // GetFunction
static TPipelineFunction GetFunction(unsigned int _type, static TPipelineFunction GetFunction(unsigned int _type,
unsigned int _format, unsigned int _elements, unsigned int _index3); unsigned int _format, unsigned int _elements, unsigned int _index3);
private: private:
enum ENormalType enum ENormalType
{ {
NRM_NOT_PRESENT = 0, NRM_NOT_PRESENT = 0,
NRM_DIRECT = 1, NRM_DIRECT = 1,
NRM_INDEX8 = 2, NRM_INDEX8 = 2,
NRM_INDEX16 = 3, NRM_INDEX16 = 3,
NUM_NRM_TYPE NUM_NRM_TYPE
}; };
enum ENormalFormat enum ENormalFormat
{ {
FORMAT_UBYTE = 0, FORMAT_UBYTE = 0,
FORMAT_BYTE = 1, FORMAT_BYTE = 1,
FORMAT_USHORT = 2, FORMAT_USHORT = 2,
FORMAT_SHORT = 3, FORMAT_SHORT = 3,
FORMAT_FLOAT = 4, FORMAT_FLOAT = 4,
NUM_NRM_FORMAT NUM_NRM_FORMAT
}; };
enum ENormalElements enum ENormalElements
{ {
NRM_NBT = 0, NRM_NBT = 0,
NRM_NBT3 = 1, NRM_NBT3 = 1,
NUM_NRM_ELEMENTS NUM_NRM_ELEMENTS
}; };
enum ENormalIndices enum ENormalIndices
{ {

10
Source/Core/VideoCommon/Src/VertexLoader_Position.cpp
View File

@ -87,10 +87,10 @@ template <typename T, int N>
void LOADERDECL Pos_ReadDirect() void LOADERDECL Pos_ReadDirect()
{ {
static_assert(N <= 3, "N > 3 is not sane!"); static_assert(N <= 3, "N > 3 is not sane!");
for (int i = 0; i < 3; ++i) for (int i = 0; i < 3; ++i)
DataWrite(i<N ? PosScale(DataRead<T>()) : 0.f); DataWrite(i<N ? PosScale(DataRead<T>()) : 0.f);
LOG_VTX(); LOG_VTX();
} }
@ -99,15 +99,15 @@ void LOADERDECL Pos_ReadIndex()
{ {
static_assert(!std::numeric_limits<I>::is_signed, "Only unsigned I is sane!"); static_assert(!std::numeric_limits<I>::is_signed, "Only unsigned I is sane!");
static_assert(N <= 3, "N > 3 is not sane!"); static_assert(N <= 3, "N > 3 is not sane!");
auto const index = DataRead<I>(); auto const index = DataRead<I>();
if (index < std::numeric_limits<I>::max()) if (index < std::numeric_limits<I>::max())
{ {
auto const data = reinterpret_cast<const T*>(cached_arraybases[ARRAY_POSITION] + (index * arraystrides[ARRAY_POSITION])); auto const data = reinterpret_cast<const T*>(cached_arraybases[ARRAY_POSITION] + (index * arraystrides[ARRAY_POSITION]));
for (int i = 0; i < 3; ++i) for (int i = 0; i < 3; ++i)
DataWrite(i<N ? PosScale(Common::FromBigEndian(data[i])) : 0.f); DataWrite(i<N ? PosScale(Common::FromBigEndian(data[i])) : 0.f);
LOG_VTX(); LOG_VTX();
} }
} }

12
Source/Core/VideoCommon/Src/VertexLoader_Position.h
View File

@ -21,14 +21,14 @@
class VertexLoader_Position { class VertexLoader_Position {
public: public:
// Init // Init
static void Init(void); static void Init(void);
// GetSize // GetSize
static unsigned int GetSize(unsigned int _type, unsigned int _format, unsigned int _elements); static unsigned int GetSize(unsigned int _type, unsigned int _format, unsigned int _elements);
// GetFunction // GetFunction
static TPipelineFunction GetFunction(unsigned int _type, unsigned int _format, unsigned int _elements); static TPipelineFunction GetFunction(unsigned int _type, unsigned int _format, unsigned int _elements);
}; };
#endif #endif

10
Source/Core/VideoCommon/Src/VertexLoader_TextCoord.cpp
View File

@ -70,7 +70,7 @@ void LOADERDECL TexCoord_ReadDirect()
DataWrite(TCScale(DataRead<T>())); DataWrite(TCScale(DataRead<T>()));
LOG_TEX<N>(); LOG_TEX<N>();
++tcIndex; ++tcIndex;
} }
@ -82,10 +82,10 @@ void LOADERDECL TexCoord_ReadIndex()
auto const index = DataRead<I>(); auto const index = DataRead<I>();
auto const data = reinterpret_cast<const T*>(cached_arraybases[ARRAY_TEXCOORD0 + tcIndex] auto const data = reinterpret_cast<const T*>(cached_arraybases[ARRAY_TEXCOORD0 + tcIndex]
+ (index * arraystrides[ARRAY_TEXCOORD0 + tcIndex])); + (index * arraystrides[ARRAY_TEXCOORD0 + tcIndex]));
for (int i = 0; i != N; ++i) for (int i = 0; i != N; ++i)
DataWrite(TCScale(Common::FromBigEndian(data[i]))); DataWrite(TCScale(Common::FromBigEndian(data[i])));
LOG_TEX<N>(); LOG_TEX<N>();
++tcIndex; ++tcIndex;
} }
@ -97,7 +97,7 @@ template <typename I>
void LOADERDECL TexCoord_ReadIndex_Short2_SSE4() void LOADERDECL TexCoord_ReadIndex_Short2_SSE4()
{ {
static_assert(!std::numeric_limits<I>::is_signed, "Only unsigned I is sane!"); static_assert(!std::numeric_limits<I>::is_signed, "Only unsigned I is sane!");
// Heavy in ZWW // Heavy in ZWW
auto const index = DataRead<I>(); auto const index = DataRead<I>();
const s32 *pData = (const s32*)(cached_arraybases[ARRAY_TEXCOORD0+tcIndex] + (index * arraystrides[ARRAY_TEXCOORD0+tcIndex])); const s32 *pData = (const s32*)(cached_arraybases[ARRAY_TEXCOORD0+tcIndex] + (index * arraystrides[ARRAY_TEXCOORD0+tcIndex]));
@ -121,7 +121,7 @@ template <typename I>
void LOADERDECL TexCoord_ReadIndex_Float2_SSSE3() void LOADERDECL TexCoord_ReadIndex_Float2_SSSE3()
{ {
static_assert(!std::numeric_limits<I>::is_signed, "Only unsigned I is sane!"); static_assert(!std::numeric_limits<I>::is_signed, "Only unsigned I is sane!");
auto const index = DataRead<I>(); auto const index = DataRead<I>();
const u32 *pData = (const u32 *)(cached_arraybases[ARRAY_TEXCOORD0+tcIndex] + (index * arraystrides[ARRAY_TEXCOORD0+tcIndex])); const u32 *pData = (const u32 *)(cached_arraybases[ARRAY_TEXCOORD0+tcIndex] + (index * arraystrides[ARRAY_TEXCOORD0+tcIndex]));
GC_ALIGNED128(const __m128i a = _mm_loadl_epi64((__m128i*)pData)); GC_ALIGNED128(const __m128i a = _mm_loadl_epi64((__m128i*)pData));

16
Source/Core/VideoCommon/Src/VertexLoader_TextCoord.h
View File

@ -24,18 +24,18 @@ class VertexLoader_TextCoord
{ {
public: public:
// Init // Init
static void Init(void); static void Init(void);
// GetSize // GetSize
static unsigned int GetSize(unsigned int _type, unsigned int _format, unsigned int _elements); static unsigned int GetSize(unsigned int _type, unsigned int _format, unsigned int _elements);
// GetFunction // GetFunction
static TPipelineFunction GetFunction(unsigned int _type, unsigned int _format, unsigned int _elements); static TPipelineFunction GetFunction(unsigned int _type, unsigned int _format, unsigned int _elements);
// GetDummyFunction // GetDummyFunction
// It is important to synchronize tcIndex. // It is important to synchronize tcIndex.
static TPipelineFunction GetDummyFunction(); static TPipelineFunction GetDummyFunction();
}; };
#endif #endif

10
Source/Core/VideoCommon/Src/VertexManagerBase.cpp
View File

@ -103,7 +103,7 @@ void VertexManager::AddVertices(int primitive, u32 numVertices)
ADDSTAT(stats.thisFrame.numPrims, numVertices); ADDSTAT(stats.thisFrame.numPrims, numVertices);
INCSTAT(stats.thisFrame.numPrimitiveJoins); INCSTAT(stats.thisFrame.numPrimitiveJoins);
IndexGenerator::AddIndices(primitive, numVertices); IndexGenerator::AddIndices(primitive, numVertices);
} }
@ -111,14 +111,14 @@ void VertexManager::Flush()
{ {
if (g_vertex_manager->IsFlushed()) if (g_vertex_manager->IsFlushed())
return; return;
// loading a state will invalidate BP, so check for it // loading a state will invalidate BP, so check for it
g_video_backend->CheckInvalidState(); g_video_backend->CheckInvalidState();
VideoFifo_CheckEFBAccess(); VideoFifo_CheckEFBAccess();
g_vertex_manager->vFlush(); g_vertex_manager->vFlush();
g_vertex_manager->ResetBuffer(); g_vertex_manager->ResetBuffer();
} }

8
Source/Core/VideoCommon/Src/VertexManagerBase.h
View File

@ -45,10 +45,10 @@ public:
virtual void DestroyDeviceObjects(){}; virtual void DestroyDeviceObjects(){};
protected: protected:
u16* GetTriangleIndexBuffer() { return &TIBuffer[0]; } u16* GetTriangleIndexBuffer() { return &TIBuffer[0]; }
u16* GetLineIndexBuffer() { return &LIBuffer[0]; } u16* GetLineIndexBuffer() { return &LIBuffer[0]; }
u16* GetPointIndexBuffer() { return &PIBuffer[0]; } u16* GetPointIndexBuffer() { return &PIBuffer[0]; }
u8* GetVertexBuffer() { return &s_pBaseBufferPointer[0]; } u8* GetVertexBuffer() { return &s_pBaseBufferPointer[0]; }
virtual void vDoState(PointerWrap& p) { DoStateShared(p); } virtual void vDoState(PointerWrap& p) { DoStateShared(p); }
void DoStateShared(PointerWrap& p); void DoStateShared(PointerWrap& p);

59
Source/Core/VideoCommon/Src/VertexShaderGen.cpp
View File

@ -174,7 +174,7 @@ const char *GenerateVertexShaderCode(u32 components, API_TYPE ApiType)
_assert_(bpmem.genMode.numtexgens == xfregs.numTexGen.numTexGens); _assert_(bpmem.genMode.numtexgens == xfregs.numTexGen.numTexGens);
_assert_(bpmem.genMode.numcolchans == xfregs.numChan.numColorChans); _assert_(bpmem.genMode.numcolchans == xfregs.numChan.numColorChans);
bool is_d3d = (ApiType & API_D3D9 || ApiType == API_D3D11); bool is_d3d = (ApiType & API_D3D9 || ApiType == API_D3D11);
u32 lightMask = 0; u32 lightMask = 0;
if (xfregs.numChan.numColorChans > 0) if (xfregs.numChan.numColorChans > 0)
@ -184,11 +184,11 @@ const char *GenerateVertexShaderCode(u32 components, API_TYPE ApiType)
char *p = text; char *p = text;
WRITE(p, "//Vertex Shader: comp:%x, \n", components); WRITE(p, "//Vertex Shader: comp:%x, \n", components);
// uniforms // uniforms
if (g_ActiveConfig.backend_info.bSupportsGLSLUBO) if (g_ActiveConfig.backend_info.bSupportsGLSLUBO)
WRITE(p, "layout(std140) uniform VSBlock {\n"); WRITE(p, "layout(std140) uniform VSBlock {\n");
WRITE(p, "%sfloat4 " I_POSNORMALMATRIX"[6] %s;\n", WriteLocation(ApiType), WriteRegister(ApiType, "c", C_POSNORMALMATRIX)); WRITE(p, "%sfloat4 " I_POSNORMALMATRIX"[6] %s;\n", WriteLocation(ApiType), WriteRegister(ApiType, "c", C_POSNORMALMATRIX));
WRITE(p, "%sfloat4 " I_PROJECTION"[4] %s;\n", WriteLocation(ApiType), WriteRegister(ApiType, "c", C_PROJECTION)); WRITE(p, "%sfloat4 " I_PROJECTION"[4] %s;\n", WriteLocation(ApiType), WriteRegister(ApiType, "c", C_PROJECTION));
WRITE(p, "%sfloat4 " I_MATERIALS"[4] %s;\n", WriteLocation(ApiType), WriteRegister(ApiType, "c", C_MATERIALS)); WRITE(p, "%sfloat4 " I_MATERIALS"[4] %s;\n", WriteLocation(ApiType), WriteRegister(ApiType, "c", C_MATERIALS));
@ -198,10 +198,9 @@ const char *GenerateVertexShaderCode(u32 components, API_TYPE ApiType)
WRITE(p, "%sfloat4 " I_NORMALMATRICES"[32] %s;\n", WriteLocation(ApiType), WriteRegister(ApiType, "c", C_NORMALMATRICES)); WRITE(p, "%sfloat4 " I_NORMALMATRICES"[32] %s;\n", WriteLocation(ApiType), WriteRegister(ApiType, "c", C_NORMALMATRICES));
WRITE(p, "%sfloat4 " I_POSTTRANSFORMMATRICES"[64] %s;\n", WriteLocation(ApiType), WriteRegister(ApiType, "c", C_POSTTRANSFORMMATRICES)); WRITE(p, "%sfloat4 " I_POSTTRANSFORMMATRICES"[64] %s;\n", WriteLocation(ApiType), WriteRegister(ApiType, "c", C_POSTTRANSFORMMATRICES));
WRITE(p, "%sfloat4 " I_DEPTHPARAMS" %s;\n", WriteLocation(ApiType), WriteRegister(ApiType, "c", C_DEPTHPARAMS)); WRITE(p, "%sfloat4 " I_DEPTHPARAMS" %s;\n", WriteLocation(ApiType), WriteRegister(ApiType, "c", C_DEPTHPARAMS));
if (g_ActiveConfig.backend_info.bSupportsGLSLUBO) if (g_ActiveConfig.backend_info.bSupportsGLSLUBO)
WRITE(p, "};\n"); WRITE(p, "};\n");
p = GenerateVSOutputStruct(p, components, ApiType); p = GenerateVSOutputStruct(p, components, ApiType);
@ -221,7 +220,7 @@ const char *GenerateVertexShaderCode(u32 components, API_TYPE ApiType)
WRITE(p, "ATTRIN float4 color0; // ATTR%d,\n", SHADER_COLOR0_ATTRIB); WRITE(p, "ATTRIN float4 color0; // ATTR%d,\n", SHADER_COLOR0_ATTRIB);
if (components & VB_HAS_COL1) if (components & VB_HAS_COL1)
WRITE(p, "ATTRIN float4 color1; // ATTR%d,\n", SHADER_COLOR1_ATTRIB); WRITE(p, "ATTRIN float4 color1; // ATTR%d,\n", SHADER_COLOR1_ATTRIB);
for (int i = 0; i < 8; ++i) { for (int i = 0; i < 8; ++i) {
u32 hastexmtx = (components & (VB_HAS_TEXMTXIDX0<<i)); u32 hastexmtx = (components & (VB_HAS_TEXMTXIDX0<<i));
if ((components & (VB_HAS_UV0<<i)) || hastexmtx) if ((components & (VB_HAS_UV0<<i)) || hastexmtx)
@ -258,8 +257,8 @@ const char *GenerateVertexShaderCode(u32 components, API_TYPE ApiType)
} }
else else
{ {
WRITE(p, "VS_OUTPUT main(\n"); WRITE(p, "VS_OUTPUT main(\n");
// inputs // inputs
if (components & VB_HAS_NRM0) if (components & VB_HAS_NRM0)
WRITE(p, " float3 rawnorm0 : NORMAL0,\n"); WRITE(p, " float3 rawnorm0 : NORMAL0,\n");
@ -335,8 +334,6 @@ const char *GenerateVertexShaderCode(u32 components, API_TYPE ApiType)
if (!(components & VB_HAS_NRM0)) if (!(components & VB_HAS_NRM0))
WRITE(p, "float3 _norm0 = float3(0.0f, 0.0f, 0.0f);\n"); WRITE(p, "float3 _norm0 = float3(0.0f, 0.0f, 0.0f);\n");
WRITE(p, "o.pos = float4(dot(" I_PROJECTION"[0], pos), dot(" I_PROJECTION"[1], pos), dot(" I_PROJECTION"[2], pos), dot(" I_PROJECTION"[3], pos));\n"); WRITE(p, "o.pos = float4(dot(" I_PROJECTION"[0], pos), dot(" I_PROJECTION"[1], pos), dot(" I_PROJECTION"[2], pos), dot(" I_PROJECTION"[3], pos));\n");
WRITE(p, "float4 mat, lacc;\n" WRITE(p, "float4 mat, lacc;\n"
@ -506,9 +503,9 @@ const char *GenerateVertexShaderCode(u32 components, API_TYPE ApiType)
WRITE(p, "o.tex7 = pos.xyzz;\n"); WRITE(p, "o.tex7 = pos.xyzz;\n");
else else
WRITE(p, "o.tex7.w = pos.z;\n"); WRITE(p, "o.tex7.w = pos.z;\n");
} }
if (components & VB_HAS_COL0) if (components & VB_HAS_COL0)
WRITE(p, "o.colors_0 = color0;\n"); WRITE(p, "o.colors_0 = color0;\n");
if (components & VB_HAS_COL1) if (components & VB_HAS_COL1)
WRITE(p, "o.colors_1 = color1;\n"); WRITE(p, "o.colors_1 = color1;\n");
@ -522,26 +519,26 @@ const char *GenerateVertexShaderCode(u32 components, API_TYPE ApiType)
} }
else else
{ {
// this results in a scale from -1..0 to -1..1 after perspective // this results in a scale from -1..0 to -1..1 after perspective
// divide // divide
WRITE(p, "o.pos.z = o.pos.w + o.pos.z * 2.0f;\n"); WRITE(p, "o.pos.z = o.pos.w + o.pos.z * 2.0f;\n");
// Sonic Unleashed puts its final rendering at the near or // Sonic Unleashed puts its final rendering at the near or
// far plane of the viewing frustrum(actually box, they use // far plane of the viewing frustrum(actually box, they use
// orthogonal projection for that), and we end up putting it // orthogonal projection for that), and we end up putting it
// just beyond, and the rendering gets clipped away. (The // just beyond, and the rendering gets clipped away. (The
// primitive gets dropped) // primitive gets dropped)
WRITE(p, "o.pos.z = o.pos.z * 1048575.0f/1048576.0f;\n"); WRITE(p, "o.pos.z = o.pos.z * 1048575.0f/1048576.0f;\n");
// the next steps of the OGL pipeline are: // the next steps of the OGL pipeline are:
// (x_c,y_c,z_c,w_c) = o.pos //switch to OGL spec terminology // (x_c,y_c,z_c,w_c) = o.pos //switch to OGL spec terminology
// clipping to -w_c <= (x_c,y_c,z_c) <= w_c // clipping to -w_c <= (x_c,y_c,z_c) <= w_c
// (x_d,y_d,z_d) = (x_c,y_c,z_c)/w_c//perspective divide // (x_d,y_d,z_d) = (x_c,y_c,z_c)/w_c//perspective divide
// z_w = (f-n)/2*z_d + (n+f)/2 // z_w = (f-n)/2*z_d + (n+f)/2
// z_w now contains the value to go to the 0..1 depth buffer // z_w now contains the value to go to the 0..1 depth buffer
//trying to get the correct semantic while not using glDepthRange //trying to get the correct semantic while not using glDepthRange
//seems to get rather complicated //seems to get rather complicated
} }
if (ApiType & API_D3D9) if (ApiType & API_D3D9)
@ -579,7 +576,7 @@ const char *GenerateVertexShaderCode(u32 components, API_TYPE ApiType)
for (unsigned int i = 0; i < xfregs.numTexGen.numTexGens; ++i) for (unsigned int i = 0; i < xfregs.numTexGen.numTexGens; ++i)
WRITE(p, " uv%d_2%s = o.tex%d;\n", i, i < 4 ? ".xyzw" : ".xyz" , i); WRITE(p, " uv%d_2%s = o.tex%d;\n", i, i < 4 ? ".xyzw" : ".xyz" , i);
} }
} }
WRITE(p, "colors_02 = o.colors_0;\n"); WRITE(p, "colors_02 = o.colors_0;\n");
WRITE(p, "colors_12 = o.colors_1;\n"); WRITE(p, "colors_12 = o.colors_1;\n");
WRITE(p, "gl_Position = o.pos;\n"); WRITE(p, "gl_Position = o.pos;\n");

2
Source/Core/VideoCommon/Src/VertexShaderGen.h
View File

@ -61,7 +61,7 @@
#define C_NORMALMATRICES (C_TRANSFORMMATRICES + 64) #define C_NORMALMATRICES (C_TRANSFORMMATRICES + 64)
#define C_POSTTRANSFORMMATRICES (C_NORMALMATRICES + 32) #define C_POSTTRANSFORMMATRICES (C_NORMALMATRICES + 32)
#define C_DEPTHPARAMS (C_POSTTRANSFORMMATRICES + 64) #define C_DEPTHPARAMS (C_POSTTRANSFORMMATRICES + 64)
#define C_VENVCONST_END (C_DEPTHPARAMS + 1) #define C_VENVCONST_END (C_DEPTHPARAMS + 1)
const s_svar VSVar_Loc[] = { {I_POSNORMALMATRIX, C_POSNORMALMATRIX, 6 }, const s_svar VSVar_Loc[] = { {I_POSNORMALMATRIX, C_POSNORMALMATRIX, 6 },
{I_PROJECTION , C_PROJECTION, 4 }, {I_PROJECTION , C_PROJECTION, 4 },
{I_MATERIALS, C_MATERIALS, 4 }, {I_MATERIALS, C_MATERIALS, 4 },

36
Source/Core/VideoCommon/Src/VertexShaderManager.cpp
View File

@ -102,7 +102,7 @@ float PHackValue(std::string sValue)
c[i] = (cStr[i] == ',') ? '.' : *(cStr+i); c[i] = (cStr[i] == ',') ? '.' : *(cStr+i);
if (c[i] == '.') if (c[i] == '.')
fp = true; fp = true;
} }
cStr = c; cStr = c;
sTof.str(cStr); sTof.str(cStr);
sTof >> f; sTof >> f;
@ -119,22 +119,22 @@ void UpdateProjectionHack(int iPhackvalue[], std::string sPhackvalue[])
float fhacksign1 = 1.0, fhacksign2 = 1.0; float fhacksign1 = 1.0, fhacksign2 = 1.0;
bool bProjHack3 = false; bool bProjHack3 = false;
const char *sTemp[2]; const char *sTemp[2];
if (iPhackvalue[0] == 1) if (iPhackvalue[0] == 1)
{ {
NOTICE_LOG(VIDEO, "\t\t--- Ortographic Projection Hack ON ---"); NOTICE_LOG(VIDEO, "\t\t--- Ortographic Projection Hack ON ---");
fhacksign1 *= (iPhackvalue[1] == 1) ? -1.0f : fhacksign1; fhacksign1 *= (iPhackvalue[1] == 1) ? -1.0f : fhacksign1;
sTemp[0] = (iPhackvalue[1] == 1) ? " * (-1)" : ""; sTemp[0] = (iPhackvalue[1] == 1) ? " * (-1)" : "";
fhacksign2 *= (iPhackvalue[2] == 1) ? -1.0f : fhacksign2; fhacksign2 *= (iPhackvalue[2] == 1) ? -1.0f : fhacksign2;
sTemp[1] = (iPhackvalue[2] == 1) ? " * (-1)" : ""; sTemp[1] = (iPhackvalue[2] == 1) ? " * (-1)" : "";
fhackvalue1 = PHackValue(sPhackvalue[0]); fhackvalue1 = PHackValue(sPhackvalue[0]);
NOTICE_LOG(VIDEO, "- zNear Correction = (%f + zNear)%s", fhackvalue1, sTemp[0]); NOTICE_LOG(VIDEO, "- zNear Correction = (%f + zNear)%s", fhackvalue1, sTemp[0]);
fhackvalue2 = PHackValue(sPhackvalue[1]); fhackvalue2 = PHackValue(sPhackvalue[1]);
NOTICE_LOG(VIDEO, "- zFar Correction = (%f + zFar)%s", fhackvalue2, sTemp[1]); NOTICE_LOG(VIDEO, "- zFar Correction = (%f + zFar)%s", fhackvalue2, sTemp[1]);
sTemp[0] = "DISABLED"; sTemp[0] = "DISABLED";
bProjHack3 = (iPhackvalue[3] == 1) ? true : bProjHack3; bProjHack3 = (iPhackvalue[3] == 1) ? true : bProjHack3;
if (bProjHack3) if (bProjHack3)
@ -171,22 +171,22 @@ void VertexShaderManager::Dirty()
{ {
nTransformMatricesChanged[0] = 0; nTransformMatricesChanged[0] = 0;
nTransformMatricesChanged[1] = 256; nTransformMatricesChanged[1] = 256;
nNormalMatricesChanged[0] = 0; nNormalMatricesChanged[0] = 0;
nNormalMatricesChanged[1] = 96; nNormalMatricesChanged[1] = 96;
nPostTransformMatricesChanged[0] = 0; nPostTransformMatricesChanged[0] = 0;
nPostTransformMatricesChanged[1] = 256; nPostTransformMatricesChanged[1] = 256;
nLightsChanged[0] = 0; nLightsChanged[0] = 0;
nLightsChanged[1] = 0x80; nLightsChanged[1] = 0x80;
bPosNormalMatrixChanged = true; bPosNormalMatrixChanged = true;
bTexMatricesChanged[0] = true; bTexMatricesChanged[0] = true;
bTexMatricesChanged[1] = true; bTexMatricesChanged[1] = true;
bProjectionChanged = true; bProjectionChanged = true;
nMaterialsChanged = 15; nMaterialsChanged = 15;
} }
@ -358,7 +358,7 @@ void VertexShaderManager::SetConstants()
switch(xfregs.projection.type) switch(xfregs.projection.type)
{ {
case GX_PERSPECTIVE: case GX_PERSPECTIVE:
g_fProjectionMatrix[0] = rawProjection[0] * g_ActiveConfig.fAspectRatioHackW; g_fProjectionMatrix[0] = rawProjection[0] * g_ActiveConfig.fAspectRatioHackW;
g_fProjectionMatrix[1] = 0.0f; g_fProjectionMatrix[1] = 0.0f;
g_fProjectionMatrix[2] = rawProjection[1]; g_fProjectionMatrix[2] = rawProjection[1];
@ -374,7 +374,7 @@ void VertexShaderManager::SetConstants()
g_fProjectionMatrix[10] = rawProjection[4]; g_fProjectionMatrix[10] = rawProjection[4];
g_fProjectionMatrix[11] = rawProjection[5]; g_fProjectionMatrix[11] = rawProjection[5];
g_fProjectionMatrix[12] = 0.0f; g_fProjectionMatrix[12] = 0.0f;
g_fProjectionMatrix[13] = 0.0f; g_fProjectionMatrix[13] = 0.0f;
// donkopunchstania: GC GPU rounds differently? // donkopunchstania: GC GPU rounds differently?
@ -399,9 +399,9 @@ void VertexShaderManager::SetConstants()
SETSTAT_FT(stats.gproj_14, g_fProjectionMatrix[14]); SETSTAT_FT(stats.gproj_14, g_fProjectionMatrix[14]);
SETSTAT_FT(stats.gproj_15, g_fProjectionMatrix[15]); SETSTAT_FT(stats.gproj_15, g_fProjectionMatrix[15]);
break; break;
case GX_ORTHOGRAPHIC: case GX_ORTHOGRAPHIC:
g_fProjectionMatrix[0] = rawProjection[0]; g_fProjectionMatrix[0] = rawProjection[0];
g_fProjectionMatrix[1] = 0.0f; g_fProjectionMatrix[1] = 0.0f;
g_fProjectionMatrix[2] = 0.0f; g_fProjectionMatrix[2] = 0.0f;
@ -426,10 +426,10 @@ void VertexShaderManager::SetConstants()
this hack was added...setting g_fProjectionMatrix[14] to -1 might make the hack more stable, needs more testing. this hack was added...setting g_fProjectionMatrix[14] to -1 might make the hack more stable, needs more testing.
Only works for OGL and DX9...this is not helping DX11 Only works for OGL and DX9...this is not helping DX11
*/ */
g_fProjectionMatrix[14] = 0.0f; g_fProjectionMatrix[14] = 0.0f;
g_fProjectionMatrix[15] = (g_ProjHack3 && rawProjection[0] == 2.0f ? 0.0f : 1.0f); //causes either the efb copy or bloom layer not to show if proj hack enabled g_fProjectionMatrix[15] = (g_ProjHack3 && rawProjection[0] == 2.0f ? 0.0f : 1.0f); //causes either the efb copy or bloom layer not to show if proj hack enabled
SETSTAT_FT(stats.g2proj_0, g_fProjectionMatrix[0]); SETSTAT_FT(stats.g2proj_0, g_fProjectionMatrix[0]);
SETSTAT_FT(stats.g2proj_1, g_fProjectionMatrix[1]); SETSTAT_FT(stats.g2proj_1, g_fProjectionMatrix[1]);
SETSTAT_FT(stats.g2proj_2, g_fProjectionMatrix[2]); SETSTAT_FT(stats.g2proj_2, g_fProjectionMatrix[2]);
@ -453,7 +453,7 @@ void VertexShaderManager::SetConstants()
SETSTAT_FT(stats.proj_4, rawProjection[4]); SETSTAT_FT(stats.proj_4, rawProjection[4]);
SETSTAT_FT(stats.proj_5, rawProjection[5]); SETSTAT_FT(stats.proj_5, rawProjection[5]);
break; break;
default: default:
ERROR_LOG(VIDEO, "unknown projection type: %d", xfregs.projection.type); ERROR_LOG(VIDEO, "unknown projection type: %d", xfregs.projection.type);
} }

2
Source/Core/VideoCommon/Src/VertexShaderManager.h
View File

@ -43,7 +43,7 @@ public:
static void Dirty(); static void Dirty();
static void Shutdown(); static void Shutdown();
static void DoState(PointerWrap &p); static void DoState(PointerWrap &p);
;
// constant management // constant management
static void SetConstants(); static void SetConstants();

18
Source/Core/VideoCommon/Src/VideoConfig.cpp
View File

@ -52,7 +52,7 @@ void VideoConfig::Load(const char *ini_file)
std::string temp; std::string temp;
IniFile iniFile; IniFile iniFile;
iniFile.Load(ini_file); iniFile.Load(ini_file);
iniFile.Get("Hardware", "VSync", &bVSync, 0); // Hardware iniFile.Get("Hardware", "VSync", &bVSync, 0); // Hardware
iniFile.Get("Settings", "wideScreenHack", &bWidescreenHack, false); iniFile.Get("Settings", "wideScreenHack", &bWidescreenHack, false);
iniFile.Get("Settings", "AspectRatio", &iAspectRatio, (int)ASPECT_AUTO); iniFile.Get("Settings", "AspectRatio", &iAspectRatio, (int)ASPECT_AUTO);
@ -78,17 +78,17 @@ void VideoConfig::Load(const char *ini_file)
iniFile.Get("Settings", "AnaglyphFocalAngle", &iAnaglyphFocalAngle, 0); iniFile.Get("Settings", "AnaglyphFocalAngle", &iAnaglyphFocalAngle, 0);
iniFile.Get("Settings", "EnablePixelLighting", &bEnablePixelLighting, 0); iniFile.Get("Settings", "EnablePixelLighting", &bEnablePixelLighting, 0);
iniFile.Get("Settings", "HackedBufferUpload", &bHackedBufferUpload, 0); iniFile.Get("Settings", "HackedBufferUpload", &bHackedBufferUpload, 0);
iniFile.Get("Settings", "MSAA", &iMultisampleMode, 0); iniFile.Get("Settings", "MSAA", &iMultisampleMode, 0);
iniFile.Get("Settings", "EFBScale", &iEFBScale, 2); // native iniFile.Get("Settings", "EFBScale", &iEFBScale, 2); // native
iniFile.Get("Settings", "DstAlphaPass", &bDstAlphaPass, false); iniFile.Get("Settings", "DstAlphaPass", &bDstAlphaPass, false);
iniFile.Get("Settings", "TexFmtOverlayEnable", &bTexFmtOverlayEnable, 0); iniFile.Get("Settings", "TexFmtOverlayEnable", &bTexFmtOverlayEnable, 0);
iniFile.Get("Settings", "TexFmtOverlayCenter", &bTexFmtOverlayCenter, 0); iniFile.Get("Settings", "TexFmtOverlayCenter", &bTexFmtOverlayCenter, 0);
iniFile.Get("Settings", "WireFrame", &bWireFrame, 0); iniFile.Get("Settings", "WireFrame", &bWireFrame, 0);
iniFile.Get("Settings", "DisableFog", &bDisableFog, 0); iniFile.Get("Settings", "DisableFog", &bDisableFog, 0);
iniFile.Get("Settings", "EnableOpenCL", &bEnableOpenCL, false); iniFile.Get("Settings", "EnableOpenCL", &bEnableOpenCL, false);
iniFile.Get("Settings", "OMPDecoder", &bOMPDecoder, false); iniFile.Get("Settings", "OMPDecoder", &bOMPDecoder, false);
@ -98,7 +98,7 @@ void VideoConfig::Load(const char *ini_file)
iniFile.Get("Enhancements", "MaxAnisotropy", &iMaxAnisotropy, 0); // NOTE - this is x in (1 << x) iniFile.Get("Enhancements", "MaxAnisotropy", &iMaxAnisotropy, 0); // NOTE - this is x in (1 << x)
iniFile.Get("Enhancements", "PostProcessingShader", &sPostProcessingShader, ""); iniFile.Get("Enhancements", "PostProcessingShader", &sPostProcessingShader, "");
iniFile.Get("Enhancements", "Enable3dVision", &b3DVision, false); iniFile.Get("Enhancements", "Enable3dVision", &b3DVision, false);
iniFile.Get("Hacks", "EFBAccessEnable", &bEFBAccessEnable, true); iniFile.Get("Hacks", "EFBAccessEnable", &bEFBAccessEnable, true);
iniFile.Get("Hacks", "DlistCachingEnable", &bDlistCachingEnable,false); iniFile.Get("Hacks", "DlistCachingEnable", &bDlistCachingEnable,false);
iniFile.Get("Hacks", "EFBCopyEnable", &bEFBCopyEnable, true); iniFile.Get("Hacks", "EFBCopyEnable", &bEFBCopyEnable, true);
@ -219,14 +219,14 @@ void VideoConfig::Save(const char *ini_file)
iniFile.Set("Settings", "EnableOpenCL", bEnableOpenCL); iniFile.Set("Settings", "EnableOpenCL", bEnableOpenCL);
iniFile.Set("Settings", "OMPDecoder", bOMPDecoder); iniFile.Set("Settings", "OMPDecoder", bOMPDecoder);
iniFile.Set("Settings", "EnableShaderDebugging", bEnableShaderDebugging); iniFile.Set("Settings", "EnableShaderDebugging", bEnableShaderDebugging);
iniFile.Set("Enhancements", "ForceFiltering", bForceFiltering); iniFile.Set("Enhancements", "ForceFiltering", bForceFiltering);
iniFile.Set("Enhancements", "MaxAnisotropy", iMaxAnisotropy); iniFile.Set("Enhancements", "MaxAnisotropy", iMaxAnisotropy);
iniFile.Set("Enhancements", "PostProcessingShader", sPostProcessingShader); iniFile.Set("Enhancements", "PostProcessingShader", sPostProcessingShader);
iniFile.Set("Enhancements", "Enable3dVision", b3DVision); iniFile.Set("Enhancements", "Enable3dVision", b3DVision);
iniFile.Set("Hacks", "EFBAccessEnable", bEFBAccessEnable); iniFile.Set("Hacks", "EFBAccessEnable", bEFBAccessEnable);
iniFile.Set("Hacks", "DlistCachingEnable", bDlistCachingEnable); iniFile.Set("Hacks", "DlistCachingEnable", bDlistCachingEnable);
iniFile.Set("Hacks", "EFBCopyEnable", bEFBCopyEnable); iniFile.Set("Hacks", "EFBCopyEnable", bEFBCopyEnable);
@ -237,7 +237,7 @@ void VideoConfig::Save(const char *ini_file)
iniFile.Set("Hacks", "EFBEmulateFormatChanges", bEFBEmulateFormatChanges); iniFile.Set("Hacks", "EFBEmulateFormatChanges", bEFBEmulateFormatChanges);
iniFile.Set("Hardware", "Adapter", iAdapter); iniFile.Set("Hardware", "Adapter", iAdapter);
iniFile.Save(ini_file); iniFile.Save(ini_file);
} }

16
Source/Core/VideoCommon/Src/VideoConfig.h
View File

@ -99,12 +99,12 @@ struct VideoConfig
bool bTexFmtOverlayCenter; bool bTexFmtOverlayCenter;
bool bShowEFBCopyRegions; bool bShowEFBCopyRegions;
bool bLogFPSToFile; bool bLogFPSToFile;
// Render // Render
bool bWireFrame; bool bWireFrame;
bool bDstAlphaPass; bool bDstAlphaPass;
bool bDisableFog; bool bDisableFog;
// Utility // Utility
bool bDumpTextures; bool bDumpTextures;
bool bHiresTextures; bool bHiresTextures;
@ -164,7 +164,7 @@ struct VideoConfig
bool bSupportsDualSourceBlend; // only supported by D3D11 and OpenGL bool bSupportsDualSourceBlend; // only supported by D3D11 and OpenGL
bool bSupportsFormatReinterpretation; bool bSupportsFormatReinterpretation;
bool bSupportsPixelLighting; bool bSupportsPixelLighting;
bool bSupportsGLSLUBO; bool bSupportsGLSLUBO;
bool bSupportsGLSLCache; bool bSupportsGLSLCache;
bool bSupportsGLPinnedMemory; bool bSupportsGLPinnedMemory;
@ -172,11 +172,11 @@ struct VideoConfig
bool bSupportsGLBaseVertex; bool bSupportsGLBaseVertex;
} backend_info; } backend_info;
// Utility // Utility
bool RealXFBEnabled() const { return bUseXFB && bUseRealXFB; } bool RealXFBEnabled() const { return bUseXFB && bUseRealXFB; }
bool VirtualXFBEnabled() const { return bUseXFB && !bUseRealXFB; } bool VirtualXFBEnabled() const { return bUseXFB && !bUseRealXFB; }
bool EFBCopiesToTextureEnabled() const { return bEFBCopyEnable && bCopyEFBToTexture; } bool EFBCopiesToTextureEnabled() const { return bEFBCopyEnable && bCopyEFBToTexture; }
bool EFBCopiesToRamEnabled() const { return bEFBCopyEnable && !bCopyEFBToTexture; } bool EFBCopiesToRamEnabled() const { return bEFBCopyEnable && !bCopyEFBToTexture; }
}; };
extern VideoConfig g_Config; extern VideoConfig g_Config;

30
Source/Core/VideoCommon/Src/VideoState.cpp
View File

@ -30,33 +30,33 @@
static void DoState(PointerWrap &p) static void DoState(PointerWrap &p)
{ {
// BP Memory // BP Memory
p.Do(bpmem); p.Do(bpmem);
p.DoMarker("BP Memory"); p.DoMarker("BP Memory");
// CP Memory // CP Memory
p.DoArray(arraybases, 16); p.DoArray(arraybases, 16);
p.DoArray(arraystrides, 16); p.DoArray(arraystrides, 16);
p.Do(MatrixIndexA); p.Do(MatrixIndexA);
p.Do(MatrixIndexB); p.Do(MatrixIndexB);
p.Do(g_VtxDesc.Hex); p.Do(g_VtxDesc.Hex);
p.DoArray(g_VtxAttr, 8); p.DoArray(g_VtxAttr, 8);
p.DoMarker("CP Memory"); p.DoMarker("CP Memory");
// XF Memory // XF Memory
p.Do(xfregs); p.Do(xfregs);
p.DoArray(xfmem, XFMEM_SIZE); p.DoArray(xfmem, XFMEM_SIZE);
p.DoMarker("XF Memory"); p.DoMarker("XF Memory");
// Texture decoder // Texture decoder
p.DoArray(texMem, TMEM_SIZE); p.DoArray(texMem, TMEM_SIZE);
p.DoMarker("texMem"); p.DoMarker("texMem");
// FIFO // FIFO
Fifo_DoState(p); Fifo_DoState(p);
p.DoMarker("Fifo"); p.DoMarker("Fifo");
CommandProcessor::DoState(p); CommandProcessor::DoState(p);
p.DoMarker("CommandProcessor"); p.DoMarker("CommandProcessor");
PixelEngine::DoState(p); PixelEngine::DoState(p);
@ -77,7 +77,7 @@ static void DoState(PointerWrap &p)
void VideoCommon_DoState(PointerWrap &p) void VideoCommon_DoState(PointerWrap &p)
{ {
DoState(p); DoState(p);
} }
void VideoCommon_RunLoop(bool enable) void VideoCommon_RunLoop(bool enable)

196
Source/Core/VideoCommon/Src/XFMemory.h
View File

@ -110,70 +110,70 @@
union LitChannel union LitChannel
{ {
struct struct
{ {
u32 matsource : 1; u32 matsource : 1;
u32 enablelighting : 1; u32 enablelighting : 1;
u32 lightMask0_3 : 4; u32 lightMask0_3 : 4;
u32 ambsource : 1; u32 ambsource : 1;
u32 diffusefunc : 2; // LIGHTDIF_X u32 diffusefunc : 2; // LIGHTDIF_X
u32 attnfunc : 2; // LIGHTATTN_X u32 attnfunc : 2; // LIGHTATTN_X
u32 lightMask4_7 : 4; u32 lightMask4_7 : 4;
}; };
struct struct
{ {
u32 hex : 15; u32 hex : 15;
u32 unused : 17; u32 unused : 17;
}; };
struct struct
{ {
u32 dummy0 : 7; u32 dummy0 : 7;
u32 lightparams : 4; u32 lightparams : 4;
u32 dummy1 : 21; u32 dummy1 : 21;
}; };
unsigned int GetFullLightMask() const unsigned int GetFullLightMask() const
{ {
return enablelighting ? (lightMask0_3 | (lightMask4_7 << 4)) : 0; return enablelighting ? (lightMask0_3 | (lightMask4_7 << 4)) : 0;
} }
}; };
union INVTXSPEC union INVTXSPEC
{ {
struct struct
{ {
u32 numcolors : 2; u32 numcolors : 2;
u32 numnormals : 2; // 0 - nothing, 1 - just normal, 2 - normals and binormals u32 numnormals : 2; // 0 - nothing, 1 - just normal, 2 - normals and binormals
u32 numtextures : 4; u32 numtextures : 4;
u32 unused : 24; u32 unused : 24;
}; };
u32 hex; u32 hex;
}; };
union TexMtxInfo union TexMtxInfo
{ {
struct struct
{ {
u32 unknown : 1; u32 unknown : 1;
u32 projection : 1; // XF_TEXPROJ_X u32 projection : 1; // XF_TEXPROJ_X
u32 inputform : 2; // XF_TEXINPUT_X u32 inputform : 2; // XF_TEXINPUT_X
u32 texgentype : 3; // XF_TEXGEN_X u32 texgentype : 3; // XF_TEXGEN_X
u32 sourcerow : 5; // XF_SRCGEOM_X u32 sourcerow : 5; // XF_SRCGEOM_X
u32 embosssourceshift : 3; // what generated texcoord to use u32 embosssourceshift : 3; // what generated texcoord to use
u32 embosslightshift : 3; // light index that is used u32 embosslightshift : 3; // light index that is used
}; };
u32 hex; u32 hex;
}; };
union PostMtxInfo union PostMtxInfo
{ {
struct struct
{ {
u32 index : 6; // base row of dual transform matrix u32 index : 6; // base row of dual transform matrix
u32 unused : 2; u32 unused : 2;
u32 normalize : 1; // normalize before send operation u32 normalize : 1; // normalize before send operation
}; };
u32 hex; u32 hex;
}; };
union NumColorChannel union NumColorChannel
@ -207,25 +207,25 @@ union DualTexInfo
struct Light struct Light
{ {
u32 useless[3]; u32 useless[3];
u32 color; //rgba u32 color; //rgba
float a0; //attenuation float a0; //attenuation
float a1; float a1;
float a2; float a2;
float k0; //k stuff float k0; //k stuff
float k1; float k1;
float k2; float k2;
union union
{ {
struct { struct {
float dpos[3]; float dpos[3];
float ddir[3]; // specular lights only float ddir[3]; // specular lights only
}; };
struct { struct {
float sdir[3]; float sdir[3];
float shalfangle[3]; // specular lights only float shalfangle[3]; // specular lights only
}; };
}; };
}; };
struct Viewport struct Viewport
@ -246,35 +246,35 @@ struct Projection
struct XFRegisters struct XFRegisters
{ {
u32 error; // 0x1000 u32 error; // 0x1000
u32 diag; // 0x1001 u32 diag; // 0x1001
u32 state0; // 0x1002 u32 state0; // 0x1002
u32 state1; // 0x1003 u32 state1; // 0x1003
u32 xfClock; // 0x1004 u32 xfClock; // 0x1004
u32 clipDisable; // 0x1005 u32 clipDisable; // 0x1005
u32 perf0; // 0x1006 u32 perf0; // 0x1006
u32 perf1; // 0x1007 u32 perf1; // 0x1007
INVTXSPEC hostinfo; // 0x1008 number of textures,colors,normals from vertex input INVTXSPEC hostinfo; // 0x1008 number of textures,colors,normals from vertex input
NumColorChannel numChan; // 0x1009 NumColorChannel numChan; // 0x1009
u32 ambColor[2]; // 0x100a, 0x100b u32 ambColor[2]; // 0x100a, 0x100b
u32 matColor[2]; // 0x100c, 0x100d u32 matColor[2]; // 0x100c, 0x100d
LitChannel color[2]; // 0x100e, 0x100f LitChannel color[2]; // 0x100e, 0x100f
LitChannel alpha[2]; // 0x1010, 0x1011 LitChannel alpha[2]; // 0x1010, 0x1011
DualTexInfo dualTexTrans; // 0x1012 DualTexInfo dualTexTrans; // 0x1012
u32 unk3; // 0x1013 u32 unk3; // 0x1013
u32 unk4; // 0x1014 u32 unk4; // 0x1014
u32 unk5; // 0x1015 u32 unk5; // 0x1015
u32 unk6; // 0x1016 u32 unk6; // 0x1016
u32 unk7; // 0x1017 u32 unk7; // 0x1017
u32 MatrixIndexA; // 0x1018 u32 MatrixIndexA; // 0x1018
u32 MatrixIndexB; // 0x1019 u32 MatrixIndexB; // 0x1019
Viewport viewport; // 0x101a - 0x101f Viewport viewport; // 0x101a - 0x101f
Projection projection; // 0x1020 - 0x1026 Projection projection; // 0x1020 - 0x1026
u32 unk8[24]; // 0x1027 - 0x103e u32 unk8[24]; // 0x1027 - 0x103e
NumTexGen numTexGen; // 0x103f NumTexGen numTexGen; // 0x103f
TexMtxInfo texMtxInfo[8]; // 0x1040 - 0x1047 TexMtxInfo texMtxInfo[8]; // 0x1040 - 0x1047
u32 unk9[8]; // 0x1048 - 0x104f u32 unk9[8]; // 0x1048 - 0x104f
PostMtxInfo postMtxInfo[8]; // 0x1050 - 0x1057 PostMtxInfo postMtxInfo[8]; // 0x1050 - 0x1057
}; };

128
Source/Core/VideoCommon/Src/XFStructs.cpp
View File

@ -36,37 +36,37 @@ void XFRegWritten(int transferSize, u32 baseAddress, u32 *pData)
u32 address = baseAddress; u32 address = baseAddress;
u32 dataIndex = 0; u32 dataIndex = 0;
while (transferSize > 0 && address < 0x1058) while (transferSize > 0 && address < 0x1058)
{ {
u32 newValue = pData[dataIndex]; u32 newValue = pData[dataIndex];
u32 nextAddress = address + 1; u32 nextAddress = address + 1;
switch (address) switch (address)
{ {
case XFMEM_ERROR: case XFMEM_ERROR:
case XFMEM_DIAG: case XFMEM_DIAG:
case XFMEM_STATE0: // internal state 0 case XFMEM_STATE0: // internal state 0
case XFMEM_STATE1: // internal state 1 case XFMEM_STATE1: // internal state 1
case XFMEM_CLOCK: case XFMEM_CLOCK:
case XFMEM_SETGPMETRIC: case XFMEM_SETGPMETRIC:
nextAddress = 0x1007; nextAddress = 0x1007;
break; break;
case XFMEM_CLIPDISABLE: case XFMEM_CLIPDISABLE:
//if (data & 1) {} // disable clipping detection //if (data & 1) {} // disable clipping detection
//if (data & 2) {} // disable trivial rejection //if (data & 2) {} // disable trivial rejection
//if (data & 4) {} // disable cpoly clipping acceleration //if (data & 4) {} // disable cpoly clipping acceleration
break; break;
case XFMEM_VTXSPECS: //__GXXfVtxSpecs, wrote 0004 case XFMEM_VTXSPECS: //__GXXfVtxSpecs, wrote 0004
break; break;
case XFMEM_SETNUMCHAN: case XFMEM_SETNUMCHAN:
if (xfregs.numChan.numColorChans != (newValue & 3)) if (xfregs.numChan.numColorChans != (newValue & 3))
VertexManager::Flush(); VertexManager::Flush();
break; break;
case XFMEM_SETCHAN0_AMBCOLOR: // Channel Ambient Color case XFMEM_SETCHAN0_AMBCOLOR: // Channel Ambient Color
case XFMEM_SETCHAN1_AMBCOLOR: case XFMEM_SETCHAN1_AMBCOLOR:
{ {
u8 chan = address - XFMEM_SETCHAN0_AMBCOLOR; u8 chan = address - XFMEM_SETCHAN0_AMBCOLOR;
@ -75,11 +75,11 @@ void XFRegWritten(int transferSize, u32 baseAddress, u32 *pData)
VertexManager::Flush(); VertexManager::Flush();
VertexShaderManager::SetMaterialColorChanged(chan); VertexShaderManager::SetMaterialColorChanged(chan);
PixelShaderManager::SetMaterialColorChanged(chan); PixelShaderManager::SetMaterialColorChanged(chan);
} }
break; break;
} }
case XFMEM_SETCHAN0_MATCOLOR: // Channel Material Color case XFMEM_SETCHAN0_MATCOLOR: // Channel Material Color
case XFMEM_SETCHAN1_MATCOLOR: case XFMEM_SETCHAN1_MATCOLOR:
{ {
u8 chan = address - XFMEM_SETCHAN0_MATCOLOR; u8 chan = address - XFMEM_SETCHAN0_MATCOLOR;
@ -92,28 +92,28 @@ void XFRegWritten(int transferSize, u32 baseAddress, u32 *pData)
break; break;
} }
case XFMEM_SETCHAN0_COLOR: // Channel Color case XFMEM_SETCHAN0_COLOR: // Channel Color
case XFMEM_SETCHAN1_COLOR: case XFMEM_SETCHAN1_COLOR:
case XFMEM_SETCHAN0_ALPHA: // Channel Alpha case XFMEM_SETCHAN0_ALPHA: // Channel Alpha
case XFMEM_SETCHAN1_ALPHA: case XFMEM_SETCHAN1_ALPHA:
if (((u32*)&xfregs)[address - 0x1000] != (newValue & 0x7fff)) if (((u32*)&xfregs)[address - 0x1000] != (newValue & 0x7fff))
VertexManager::Flush(); VertexManager::Flush();
break; break;
case XFMEM_DUALTEX: case XFMEM_DUALTEX:
if (xfregs.dualTexTrans.enabled != (newValue & 1)) if (xfregs.dualTexTrans.enabled != (newValue & 1))
VertexManager::Flush(); VertexManager::Flush();
break; break;
case XFMEM_SETMATRIXINDA: case XFMEM_SETMATRIXINDA:
//_assert_msg_(GX_XF, 0, "XF matrixindex0"); //_assert_msg_(GX_XF, 0, "XF matrixindex0");
VertexShaderManager::SetTexMatrixChangedA(newValue); VertexShaderManager::SetTexMatrixChangedA(newValue);
break; break;
case XFMEM_SETMATRIXINDB: case XFMEM_SETMATRIXINDB:
//_assert_msg_(GX_XF, 0, "XF matrixindex1"); //_assert_msg_(GX_XF, 0, "XF matrixindex1");
VertexShaderManager::SetTexMatrixChangedB(newValue); VertexShaderManager::SetTexMatrixChangedB(newValue);
break; break;
case XFMEM_SETVIEWPORT: case XFMEM_SETVIEWPORT:
case XFMEM_SETVIEWPORT+1: case XFMEM_SETVIEWPORT+1:
@ -135,16 +135,16 @@ void XFRegWritten(int transferSize, u32 baseAddress, u32 *pData)
case XFMEM_SETPROJECTION+4: case XFMEM_SETPROJECTION+4:
case XFMEM_SETPROJECTION+5: case XFMEM_SETPROJECTION+5:
case XFMEM_SETPROJECTION+6: case XFMEM_SETPROJECTION+6:
VertexManager::Flush(); VertexManager::Flush();
VertexShaderManager::SetProjectionChanged(); VertexShaderManager::SetProjectionChanged();
nextAddress = XFMEM_SETPROJECTION + 7; nextAddress = XFMEM_SETPROJECTION + 7;
break; break;
case XFMEM_SETNUMTEXGENS: // GXSetNumTexGens case XFMEM_SETNUMTEXGENS: // GXSetNumTexGens
if (xfregs.numTexGen.numTexGens != (newValue & 15)) if (xfregs.numTexGen.numTexGens != (newValue & 15))
VertexManager::Flush(); VertexManager::Flush();
break; break;
case XFMEM_SETTEXMTXINFO: case XFMEM_SETTEXMTXINFO:
case XFMEM_SETTEXMTXINFO+1: case XFMEM_SETTEXMTXINFO+1:
@ -178,47 +178,47 @@ void XFRegWritten(int transferSize, u32 baseAddress, u32 *pData)
// Maybe these are for Normals? // Maybe these are for Normals?
case 0x1048: //xfregs.texcoords[0].nrmmtxinfo.hex = data; break; ?? case 0x1048: //xfregs.texcoords[0].nrmmtxinfo.hex = data; break; ??
case 0x1049: case 0x1049:
case 0x104a: case 0x104a:
case 0x104b: case 0x104b:
case 0x104c: case 0x104c:
case 0x104d: case 0x104d:
case 0x104e: case 0x104e:
case 0x104f: case 0x104f:
DEBUG_LOG(VIDEO, "Possible Normal Mtx XF reg?: %x=%x\n", address, newValue); DEBUG_LOG(VIDEO, "Possible Normal Mtx XF reg?: %x=%x\n", address, newValue);
break; break;
case 0x1013: case 0x1013:
case 0x1014: case 0x1014:
case 0x1015: case 0x1015:
case 0x1016: case 0x1016:
case 0x1017: case 0x1017:
default: default:
WARN_LOG(VIDEO, "Unknown XF Reg: %x=%x\n", address, newValue); WARN_LOG(VIDEO, "Unknown XF Reg: %x=%x\n", address, newValue);
break; break;
} }
int transferred = nextAddress - address; int transferred = nextAddress - address;
address = nextAddress; address = nextAddress;
transferSize -= transferred; transferSize -= transferred;
dataIndex += transferred; dataIndex += transferred;
} }
} }
void LoadXFReg(u32 transferSize, u32 baseAddress, u32 *pData) void LoadXFReg(u32 transferSize, u32 baseAddress, u32 *pData)
{ {
// do not allow writes past registers // do not allow writes past registers
if (baseAddress + transferSize > 0x1058) if (baseAddress + transferSize > 0x1058)
{ {
INFO_LOG(VIDEO, "xf load exceeds address space: %x %d bytes\n", baseAddress, transferSize); INFO_LOG(VIDEO, "xf load exceeds address space: %x %d bytes\n", baseAddress, transferSize);
if (baseAddress >= 0x1058) if (baseAddress >= 0x1058)
transferSize = 0; transferSize = 0;
else else
transferSize = 0x1058 - baseAddress; transferSize = 0x1058 - baseAddress;
} }
// write to XF mem // write to XF mem
if (baseAddress < 0x1000 && transferSize > 0) if (baseAddress < 0x1000 && transferSize > 0)
@ -247,11 +247,11 @@ void LoadXFReg(u32 transferSize, u32 baseAddress, u32 *pData)
} }
// write to XF regs // write to XF regs
if (transferSize > 0) if (transferSize > 0)
{ {
XFRegWritten(transferSize, baseAddress, pData); XFRegWritten(transferSize, baseAddress, pData);
memcpy_gc((u32*)(&xfregs) + (baseAddress - 0x1000), pData, transferSize * 4); memcpy_gc((u32*)(&xfregs) + (baseAddress - 0x1000), pData, transferSize * 4);
} }
} }
// TODO - verify that it is correct. Seems to work, though. // TODO - verify that it is correct. Seems to work, though.

33
Source/Core/VideoCommon/Src/x64DLCache.cpp
View File

@ -111,17 +111,17 @@ struct CachedDisplayList
u32 num_cp_reg; u32 num_cp_reg;
u32 num_bp_reg; u32 num_bp_reg;
u32 num_index_xf; u32 num_index_xf;
u32 num_draw_call; u32 num_draw_call;
u32 pass; u32 pass;
u32 check; u32 check;
int frame_count; int frame_count;
u32 BufferCount; u32 BufferCount;
void InsertRegion(ReferencedDataRegion* NewRegion) void InsertRegion(ReferencedDataRegion* NewRegion)
{ {
if(LastRegion) if(LastRegion)
{ {
LastRegion->NextRegion = NewRegion; LastRegion->NextRegion = NewRegion;
} }
LastRegion = NewRegion; LastRegion = NewRegion;
if(!Regions) if(!Regions)
@ -174,7 +174,7 @@ struct CachedDisplayList
Current = Current->NextRegion; Current = Current->NextRegion;
} }
return true; return true;
} }
ReferencedDataRegion* FindOverlapingRegion(u8* RegionStart, u32 Regionsize) ReferencedDataRegion* FindOverlapingRegion(u8* RegionStart, u32 Regionsize)
{ {
@ -182,7 +182,7 @@ struct CachedDisplayList
while(Current) while(Current)
{ {
if(!Current->IntersectsMemoryRange(RegionStart, Regionsize)) if(!Current->IntersectsMemoryRange(RegionStart, Regionsize))
return Current; return Current;
Current = Current->NextRegion; Current = Current->NextRegion;
} }
return Current; return Current;
@ -330,7 +330,7 @@ u32 AnalyzeAndRunDisplayList(u32 address, u32 size, CachedDisplayList *dl)
u32 addr = DataReadU32(); u32 addr = DataReadU32();
u32 count = DataReadU32(); u32 count = DataReadU32();
ExecuteDisplayList(addr, count); ExecuteDisplayList(addr, count);
} }
break; break;
case GX_CMD_UNKNOWN_METRICS: // zelda 4 swords calls it and checks the metrics registers after that case GX_CMD_UNKNOWN_METRICS: // zelda 4 swords calls it and checks the metrics registers after that
DEBUG_LOG(VIDEO, "GX 0x44: %08x", cmd_byte); DEBUG_LOG(VIDEO, "GX 0x44: %08x", cmd_byte);
@ -392,7 +392,7 @@ u32 AnalyzeAndRunDisplayList(u32 address, u32 size, CachedDisplayList *dl)
dl->num_draw_call = num_draw_call; dl->num_draw_call = num_draw_call;
dl->num_index_xf = num_index_xf; dl->num_index_xf = num_index_xf;
dl->num_xf_reg = num_xf_reg; dl->num_xf_reg = num_xf_reg;
// reset to the old pointer // reset to the old pointer
g_pVideoData = old_pVideoData; g_pVideoData = old_pVideoData;
return result; return result;
} }
@ -458,7 +458,7 @@ void CompileAndRunDisplayList(u32 address, u32 size, CachedDisplayList *dl)
NewRegion->MustClean = true; NewRegion->MustClean = true;
NewRegion->size = transfer_size * 4; NewRegion->size = transfer_size * 4;
NewRegion->start_address = (u8*) new u8[NewRegion->size+15+12]; // alignment and guaranteed space NewRegion->start_address = (u8*) new u8[NewRegion->size+15+12]; // alignment and guaranteed space
NewRegion->hash = 0; NewRegion->hash = 0;
dl->InsertRegion(NewRegion); dl->InsertRegion(NewRegion);
u32 *data_buffer = (u32*)(u8*)(((size_t)NewRegion->start_address+0xf)&~0xf); u32 *data_buffer = (u32*)(u8*)(((size_t)NewRegion->start_address+0xf)&~0xf);
DataReadU32xFuncs[transfer_size-1](data_buffer); DataReadU32xFuncs[transfer_size-1](data_buffer);
@ -539,7 +539,6 @@ void CompileAndRunDisplayList(u32 address, u32 size, CachedDisplayList *dl)
if (cmd_byte & 0x80) if (cmd_byte & 0x80)
{ {
// load vertices (use computed vertex size from FifoCommandRunnable above) // load vertices (use computed vertex size from FifoCommandRunnable above)
u16 numVertices = DataReadU16(); u16 numVertices = DataReadU16();
if(numVertices > 0) if(numVertices > 0)
{ {
@ -558,7 +557,7 @@ void CompileAndRunDisplayList(u32 address, u32 size, CachedDisplayList *dl)
NewRegion->MustClean = true; NewRegion->MustClean = true;
NewRegion->size = Vdatasize; NewRegion->size = Vdatasize;
NewRegion->start_address = (u8*)new u8[Vdatasize]; NewRegion->start_address = (u8*)new u8[Vdatasize];
NewRegion->hash = 0; NewRegion->hash = 0;
dl->InsertRegion(NewRegion); dl->InsertRegion(NewRegion);
memcpy(NewRegion->start_address, StartAddress, Vdatasize); memcpy(NewRegion->start_address, StartAddress, Vdatasize);
emitter.ABI_CallFunctionCCCP((void *)&VertexLoaderManager::RunCompiledVertices, cmd_byte & GX_VAT_MASK, (cmd_byte & GX_PRIMITIVE_MASK) >> GX_PRIMITIVE_SHIFT, numVertices, NewRegion->start_address); emitter.ABI_CallFunctionCCCP((void *)&VertexLoaderManager::RunCompiledVertices, cmd_byte & GX_VAT_MASK, (cmd_byte & GX_PRIMITIVE_MASK) >> GX_PRIMITIVE_SHIFT, numVertices, NewRegion->start_address);
@ -577,7 +576,6 @@ void CompileAndRunDisplayList(u32 address, u32 size, CachedDisplayList *dl)
} }
} }
} }
} }
else else
{ {
@ -626,7 +624,7 @@ void Clear()
} }
dl_map.clear(); dl_map.clear();
// Reset the cache pointers. // Reset the cache pointers.
emitter.SetCodePtr(dlcode_cache); emitter.SetCodePtr(dlcode_cache);
} }
void ProgressiveCleanup() void ProgressiveCleanup()
@ -684,14 +682,14 @@ bool HandleDisplayList(u32 address, u32 size)
{ {
vhash = DLCache::CreateVMapId(Parentiter->second.VATUsed); vhash = DLCache::CreateVMapId(Parentiter->second.VATUsed);
iter = Parentiter->second.dl_map.find(vhash); iter = Parentiter->second.dl_map.find(vhash);
childexist = iter != Parentiter->second.dl_map.end(); childexist = iter != Parentiter->second.dl_map.end();
} }
if (Parentiter != DLCache::dl_map.end() && childexist) if (Parentiter != DLCache::dl_map.end() && childexist)
{ {
DLCache::CachedDisplayList &dl = iter->second; DLCache::CachedDisplayList &dl = iter->second;
if (dl.uncachable) if (dl.uncachable)
{ {
return false; return false;
} }
switch (dl.pass) switch (dl.pass)
@ -700,7 +698,7 @@ bool HandleDisplayList(u32 address, u32 size)
// First, check that the hash is the same as the last time. // First, check that the hash is the same as the last time.
if (dl.dl_hash != GetHash64(Memory::GetPointer(address), size, 0)) if (dl.dl_hash != GetHash64(Memory::GetPointer(address), size, 0))
{ {
dl.uncachable = true; dl.uncachable = true;
return false; return false;
} }
DLCache::CompileAndRunDisplayList(address, size, &dl); DLCache::CompileAndRunDisplayList(address, size, &dl);
@ -717,7 +715,7 @@ bool HandleDisplayList(u32 address, u32 size)
if (DlistChanged) if (DlistChanged)
{ {
dl.uncachable = true; dl.uncachable = true;
dl.ClearRegions(); dl.ClearRegions();
return false; return false;
} }
dl.frame_count= frameCount; dl.frame_count= frameCount;
@ -764,10 +762,9 @@ bool HandleDisplayList(u32 address, u32 size)
vdl.VATUsed = dlvatused; vdl.VATUsed = dlvatused;
vdl.count = 1; vdl.count = 1;
DLCache::dl_map[dl_id] = vdl; DLCache::dl_map[dl_id] = vdl;
} }
return true; return true;
} }
void IncrementCheckContextId() void IncrementCheckContextId()

36
Source/Core/VideoCommon/Src/x64TextureDecoder.cpp
View File

@ -517,7 +517,7 @@ inline void decodebytesARGB8_4ToRgba(u32 *dst, const u16 *src, const u16 * src2)
{ {
#if 0 #if 0
for (int x = 0; x < 4; x++) { for (int x = 0; x < 4; x++) {
dst[x] = ((src[x] & 0xFF) << 24) | ((src[x] & 0xFF00)>>8) | (src2[x] << 8); dst[x] = ((src[x] & 0xFF) << 24) | ((src[x] & 0xFF00)>>8) | (src2[x] << 8);
} }
#else #else
dst[0] = ((src[0] & 0xFF) << 24) | ((src[0] & 0xFF00)>>8) | (src2[0] << 8); dst[0] = ((src[0] & 0xFF) << 24) | ((src[0] & 0xFF00)>>8) | (src2[0] << 8);
@ -556,7 +556,7 @@ void decodeDXTBlock(u32 *dst, const DXTBlock *src, int pitch)
{ {
int blue3 = ((blue2 - blue1) >> 1) - ((blue2 - blue1) >> 3); int blue3 = ((blue2 - blue1) >> 1) - ((blue2 - blue1) >> 3);
int green3 = ((green2 - green1) >> 1) - ((green2 - green1) >> 3); int green3 = ((green2 - green1) >> 1) - ((green2 - green1) >> 3);
int red3 = ((red2 - red1) >> 1) - ((red2 - red1) >> 3); int red3 = ((red2 - red1) >> 1) - ((red2 - red1) >> 3);
colors[2] = makecol(red1 + red3, green1 + green3, blue1 + blue3, 255); colors[2] = makecol(red1 + red3, green1 + green3, blue1 + blue3, 255);
colors[3] = makecol(red2 - red3, green2 - green3, blue2 - blue3, 255); colors[3] = makecol(red2 - red3, green2 - green3, blue2 - blue3, 255);
} }
@ -2140,7 +2140,7 @@ void TexDecoder_DecodeTexel(u8 *dst, const u8 *src, int s, int t, int imageWidth
u32 base = (tBlk * widthBlks + sBlk) * blockWidth * blockHeight; u32 base = (tBlk * widthBlks + sBlk) * blockWidth * blockHeight;
u16 blkS = s & (blockWidth - 1); u16 blkS = s & (blockWidth - 1);
u16 blkT = t & (blockHeight - 1); u16 blkT = t & (blockHeight - 1);
u32 blkOff = blkT * blockWidth + blkS; u32 blkOff = blkT * blockWidth + blkS;
*/ */
switch (texformat) switch (texformat)
@ -2154,9 +2154,9 @@ void TexDecoder_DecodeTexel(u8 *dst, const u8 *src, int s, int t, int imageWidth
u16 blkS = s & 7; u16 blkS = s & 7;
u16 blkT = t & 7; u16 blkT = t & 7;
u32 blkOff = (blkT << 3) + blkS; u32 blkOff = (blkT << 3) + blkS;
int rs = (blkOff & 1)?0:4; int rs = (blkOff & 1)?0:4;
u32 offset = base + (blkOff >> 1); u32 offset = base + (blkOff >> 1);
u8 val = (*(src + offset) >> rs) & 0xF; u8 val = (*(src + offset) >> rs) & 0xF;
u16 *tlut = (u16*)(texMem + tlutaddr); u16 *tlut = (u16*)(texMem + tlutaddr);
@ -2166,7 +2166,7 @@ void TexDecoder_DecodeTexel(u8 *dst, const u8 *src, int s, int t, int imageWidth
case 0: case 0:
*((u32*)dst) = decodeIA8Swapped(tlut[val]); *((u32*)dst) = decodeIA8Swapped(tlut[val]);
break; break;
case 1: case 1:
*((u32*)dst) = decode565RGBA(Common::swap16(tlut[val])); *((u32*)dst) = decode565RGBA(Common::swap16(tlut[val]));
break; break;
case 2: case 2:
@ -2184,9 +2184,9 @@ void TexDecoder_DecodeTexel(u8 *dst, const u8 *src, int s, int t, int imageWidth
u16 blkS = s & 7; u16 blkS = s & 7;
u16 blkT = t & 7; u16 blkT = t & 7;
u32 blkOff = (blkT << 3) + blkS; u32 blkOff = (blkT << 3) + blkS;
int rs = (blkOff & 1)?0:4; int rs = (blkOff & 1)?0:4;
u32 offset = base + (blkOff >> 1); u32 offset = base + (blkOff >> 1);
u8 val = (*(src + offset) >> rs) & 0xF; u8 val = (*(src + offset) >> rs) & 0xF;
val = Convert4To8(val); val = Convert4To8(val);
@ -2205,7 +2205,7 @@ void TexDecoder_DecodeTexel(u8 *dst, const u8 *src, int s, int t, int imageWidth
u16 blkS = s & 7; u16 blkS = s & 7;
u16 blkT = t & 3; u16 blkT = t & 3;
u32 blkOff = (blkT << 3) + blkS; u32 blkOff = (blkT << 3) + blkS;
u8 val = *(src + base + blkOff); u8 val = *(src + base + blkOff);
dst[0] = val; dst[0] = val;
dst[1] = val; dst[1] = val;
@ -2222,7 +2222,7 @@ void TexDecoder_DecodeTexel(u8 *dst, const u8 *src, int s, int t, int imageWidth
u16 blkS = s & 7; u16 blkS = s & 7;
u16 blkT = t & 3; u16 blkT = t & 3;
u32 blkOff = (blkT << 3) + blkS; u32 blkOff = (blkT << 3) + blkS;
u8 val = *(src + base + blkOff); u8 val = *(src + base + blkOff);
u16 *tlut = (u16*)(texMem + tlutaddr); u16 *tlut = (u16*)(texMem + tlutaddr);
@ -2231,7 +2231,7 @@ void TexDecoder_DecodeTexel(u8 *dst, const u8 *src, int s, int t, int imageWidth
case 0: case 0:
*((u32*)dst) = decodeIA8Swapped(tlut[val]); *((u32*)dst) = decodeIA8Swapped(tlut[val]);
break; break;
case 1: case 1:
*((u32*)dst) = decode565RGBA(Common::swap16(tlut[val])); *((u32*)dst) = decode565RGBA(Common::swap16(tlut[val]));
break; break;
case 2: case 2:
@ -2249,7 +2249,7 @@ void TexDecoder_DecodeTexel(u8 *dst, const u8 *src, int s, int t, int imageWidth
u16 blkS = s & 7; u16 blkS = s & 7;
u16 blkT = t & 3; u16 blkT = t & 3;
u32 blkOff = (blkT << 3) + blkS; u32 blkOff = (blkT << 3) + blkS;
u8 val = *(src + base + blkOff); u8 val = *(src + base + blkOff);
const u8 a = Convert4To8(val>>4); const u8 a = Convert4To8(val>>4);
const u8 l = Convert4To8(val&0xF); const u8 l = Convert4To8(val&0xF);
@ -2268,7 +2268,7 @@ void TexDecoder_DecodeTexel(u8 *dst, const u8 *src, int s, int t, int imageWidth
u16 blkS = s & 3; u16 blkS = s & 3;
u16 blkT = t & 3; u16 blkT = t & 3;
u32 blkOff = (blkT << 2) + blkS; u32 blkOff = (blkT << 2) + blkS;
u32 offset = (base + blkOff) << 1; u32 offset = (base + blkOff) << 1;
const u16* valAddr = (u16*)(src + offset); const u16* valAddr = (u16*)(src + offset);
@ -2284,7 +2284,7 @@ void TexDecoder_DecodeTexel(u8 *dst, const u8 *src, int s, int t, int imageWidth
u16 blkS = s & 3; u16 blkS = s & 3;
u16 blkT = t & 3; u16 blkT = t & 3;
u32 blkOff = (blkT << 2) + blkS; u32 blkOff = (blkT << 2) + blkS;
u32 offset = (base + blkOff) << 1; u32 offset = (base + blkOff) << 1;
const u16* valAddr = (u16*)(src + offset); const u16* valAddr = (u16*)(src + offset);
@ -2296,7 +2296,7 @@ void TexDecoder_DecodeTexel(u8 *dst, const u8 *src, int s, int t, int imageWidth
case 0: case 0:
*((u32*)dst) = decodeIA8Swapped(tlut[val]); *((u32*)dst) = decodeIA8Swapped(tlut[val]);
break; break;
case 1: case 1:
*((u32*)dst) = decode565RGBA(Common::swap16(tlut[val])); *((u32*)dst) = decode565RGBA(Common::swap16(tlut[val]));
break; break;
case 2: case 2:
@ -2314,7 +2314,7 @@ void TexDecoder_DecodeTexel(u8 *dst, const u8 *src, int s, int t, int imageWidth
u16 blkS = s & 3; u16 blkS = s & 3;
u16 blkT = t & 3; u16 blkT = t & 3;
u32 blkOff = (blkT << 2) + blkS; u32 blkOff = (blkT << 2) + blkS;
u32 offset = (base + blkOff) << 1; u32 offset = (base + blkOff) << 1;
const u16* valAddr = (u16*)(src + offset); const u16* valAddr = (u16*)(src + offset);
@ -2372,7 +2372,7 @@ void TexDecoder_DecodeTexel(u8 *dst, const u8 *src, int s, int t, int imageWidth
u32 offset = (base + blkOff) << 3; u32 offset = (base + blkOff) << 3;
const DXTBlock* dxtBlock = (const DXTBlock*)(src + offset); const DXTBlock* dxtBlock = (const DXTBlock*)(src + offset);
u16 c1 = Common::swap16(dxtBlock->color1); u16 c1 = Common::swap16(dxtBlock->color1);
u16 c2 = Common::swap16(dxtBlock->color2); u16 c2 = Common::swap16(dxtBlock->color2);
int blue1 = Convert5To8(c1 & 0x1F); int blue1 = Convert5To8(c1 & 0x1F);
@ -2391,7 +2391,7 @@ void TexDecoder_DecodeTexel(u8 *dst, const u8 *src, int s, int t, int imageWidth
colorSel |= c1 > c2?0:4; colorSel |= c1 > c2?0:4;
u32 color = 0; u32 color = 0;
switch (colorSel) switch (colorSel)
{ {
case 0: case 0: