/* Copyright (C) 1994-1995 Apogee Software, Ltd. Copyright (C) 2015 EDuke32 developers Copyright (C) 2015 Voidpoint, LLC This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2 of the License, or (at your option) any later version. This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details. You should have received a copy of the GNU General Public License along with this program; if not, write to the Free Software Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA. */ /********************************************************************** module: MULTIVOC.C author: James R. Dose date: December 20, 1993 Routines to provide multichannel digitized sound playback for Sound Blaster compatible sound cards. (c) Copyright 1993 James R. Dose. All Rights Reserved. **********************************************************************/ #include "compat.h" #include "pragmas.h" #include #include #include #include #include #include "linklist.h" #include "drivers.h" #include "pitch.h" #include "multivoc.h" #include "_multivc.h" #include "fx_man.h" static void MV_StopVoice(VoiceNode *voice); static void MV_ServiceVoc(void); static VoiceNode *MV_GetVoice(int32_t handle); static const int16_t *MV_GetVolumeTable(int32_t vol); #define IS_QUIET(ptr) ((void const *)(ptr) == (void *)&MV_VolumeTable[0]) static int32_t MV_ReverbLevel; static int32_t MV_ReverbDelay; static int16_t *MV_ReverbTable = NULL; static int16_t MV_VolumeTable[MV_MAXVOLUME + 1][256]; Pan MV_PanTable[MV_NUMPANPOSITIONS][MV_MAXVOLUME + 1]; int32_t MV_Installed = FALSE; static int32_t MV_TotalVolume = MV_MAXTOTALVOLUME; static int32_t MV_MaxVoices = 1; static int32_t MV_BufferSize = MV_MIXBUFFERSIZE; static int32_t MV_BufferLength; static int32_t MV_NumberOfBuffers = MV_NUMBEROFBUFFERS; static int32_t MV_Channels = 1; static int32_t MV_ReverseStereo = FALSE; int32_t MV_MixRate; static int32_t MV_BufferEmpty[MV_NUMBEROFBUFFERS]; char *MV_MixBuffer[MV_NUMBEROFBUFFERS + 1]; static VoiceNode *MV_Voices = NULL; static VoiceNode VoiceList; static VoiceNode VoicePool; static int32_t MV_MixPage = 0; void (*MV_Printf)(const char *fmt, ...) = NULL; static void (*MV_CallBackFunc)(uint32_t) = NULL; char *MV_MixDestination; const int16_t *MV_LeftVolume; const int16_t *MV_RightVolume; int32_t MV_SampleSize = 1; int32_t MV_RightChannelOffset; uint32_t MV_MixPosition; int32_t MV_ErrorCode = MV_NotInstalled; static int32_t lockdepth = 0; FORCE_INLINE void DisableInterrupts(void) { if (lockdepth++ <= 0) SoundDriver_Lock(); } FORCE_INLINE void RestoreInterrupts(void) { if (--lockdepth <= 0) SoundDriver_Unlock(); } const char *MV_ErrorString(int32_t ErrorNumber) { switch (ErrorNumber) { case MV_Error: return MV_ErrorString(MV_ErrorCode); case MV_Ok: return "Multivoc ok."; case MV_NotInstalled: return "Multivoc not installed."; case MV_DriverError: return SoundDriver_ErrorString(SoundDriver_GetError()); case MV_NoVoices: return "No free voices available to Multivoc."; case MV_NoMem: return "Out of memory in Multivoc."; case MV_VoiceNotFound: return "No voice with matching handle found."; case MV_InvalidFile: return "Invalid file passed in to Multivoc."; default: return "Unknown Multivoc error code."; } } static void MV_Mix(VoiceNode *voice, int const buffer) { /* cheap fix for a crash under 64-bit linux */ /* v v v v */ if (voice->length == 0 && (voice->GetSound == NULL || voice->GetSound(voice) != KeepPlaying)) return; int32_t length = MV_MIXBUFFERSIZE; uint32_t FixedPointBufferSize = voice->FixedPointBufferSize; MV_MixDestination = MV_MixBuffer[buffer]; MV_LeftVolume = voice->LeftVolume; MV_RightVolume = voice->RightVolume; if ((MV_Channels == 2) && (IS_QUIET(MV_LeftVolume))) { MV_LeftVolume = MV_RightVolume; MV_MixDestination += MV_RightChannelOffset; } // Add this voice to the mix do { const char *start = voice->sound; uint32_t const rate = voice->RateScale; uint32_t const position = voice->position; int32_t voclength; // Check if the last sample in this buffer would be // beyond the length of the sample block if ((position + FixedPointBufferSize) >= voice->length) { if (position >= voice->length) { voice->GetSound(voice); return; } voclength = (voice->length - position + rate - voice->channels) / rate; } else voclength = length; if (voice->mix) voice->mix(position, rate, start, voclength); voice->position = MV_MixPosition; length -= voclength; if (voice->position >= voice->length) { // Get the next block of sound if (voice->GetSound(voice) != KeepPlaying) return; if (length > (voice->channels - 1)) { // Get the position of the last sample in the buffer FixedPointBufferSize = voice->RateScale * (length - voice->channels); } } } while (length > 0); } void MV_PlayVoice(VoiceNode *voice) { DisableInterrupts(); LL_SortedInsertion(&VoiceList, voice, prev, next, VoiceNode, priority); RestoreInterrupts(); } static void MV_StopVoice(VoiceNode *voice) { DisableInterrupts(); // move the voice from the play list to the free list LL_Remove(voice, next, prev); LL_Add((VoiceNode*) &VoicePool, voice, next, prev); RestoreInterrupts(); switch (voice->wavetype) { #ifdef HAVE_VORBIS case FMT_VORBIS: MV_ReleaseVorbisVoice(voice); break; #endif #ifdef HAVE_FLAC case FMT_FLAC: MV_ReleaseFLACVoice(voice); break; #endif case FMT_XA: MV_ReleaseXAVoice(voice); break; default: break; } voice->handle = 0; } /*--------------------------------------------------------------------- JBF: no synchronisation happens inside MV_ServiceVoc nor the supporting functions it calls. This would cause a deadlock between the mixer thread in the driver vs the nested locking in the user-space functions of MultiVoc. The call to MV_ServiceVoc is synchronised in the driver. ---------------------------------------------------------------------*/ static void MV_ServiceVoc(void) { // Toggle which buffer we'll mix next if (++MV_MixPage >= MV_NumberOfBuffers) MV_MixPage -= MV_NumberOfBuffers; if (MV_ReverbLevel == 0) { // Initialize buffer //Commented out so that the buffer is always cleared. //This is so the guys at Echo Speech can mix into the //buffer even when no sounds are playing. if (!MV_BufferEmpty[MV_MixPage]) { Bmemset(MV_MixBuffer[MV_MixPage], 0, MV_BufferSize); MV_BufferEmpty[ MV_MixPage ] = TRUE; } } else { char const *const end = MV_MixBuffer[0] + MV_BufferLength; char *dest = MV_MixBuffer[MV_MixPage]; char const *source = MV_MixBuffer[MV_MixPage] - MV_ReverbDelay; if (source < MV_MixBuffer[ 0 ]) source += MV_BufferLength; int32_t length = MV_BufferSize; while (length > 0) { int const count = (source + length > end) ? (end - source) : length; MV_16BitReverb(source, dest, MV_ReverbTable, count / 2); // if we go through the loop again, it means that we've wrapped around the buffer source = MV_MixBuffer[ 0 ]; dest += count; length -= count; } } // Play any waiting voices //DisableInterrupts(); if (!VoiceList.next || VoiceList.next == &VoiceList) return; VoiceNode *voice = VoiceList.next; int iter = 0; VoiceNode *next; do { next = voice->next; if (++iter > MV_MaxVoices && MV_Printf) MV_Printf("more iterations than voices! iter: %d\n",iter); if (voice->Paused) continue; MV_BufferEmpty[ MV_MixPage ] = FALSE; MV_Mix(voice, MV_MixPage); // Is this voice done? if (!voice->Playing) { //JBF: prevent a deadlock caused by MV_StopVoice grabbing the mutex again //MV_StopVoice( voice ); LL_Remove(voice, next, prev); LL_Add((VoiceNode*) &VoicePool, voice, next, prev); switch (voice->wavetype) { #ifdef HAVE_VORBIS case FMT_VORBIS: MV_ReleaseVorbisVoice(voice); break; #endif #ifdef HAVE_FLAC case FMT_FLAC: MV_ReleaseFLACVoice(voice); break; #endif case FMT_XA: MV_ReleaseXAVoice(voice); break; default: break; } voice->handle = 0; if (MV_CallBackFunc) MV_CallBackFunc(voice->callbackval); } } while ((voice = next) != &VoiceList); //RestoreInterrupts(); } static VoiceNode *MV_GetVoice(int32_t handle) { if (handle < MV_MINVOICEHANDLE || handle > MV_MaxVoices) { if (MV_Printf) MV_Printf("MV_GetVoice(): bad handle (%d)!\n", handle); return NULL; } DisableInterrupts(); for (VoiceNode *voice = VoiceList.next; voice != &VoiceList; voice = voice->next) { if (handle == voice->handle) { RestoreInterrupts(); return voice; } } RestoreInterrupts(); MV_SetErrorCode(MV_VoiceNotFound); return NULL; } VoiceNode *MV_BeginService(int32_t handle) { if (!MV_Installed) return NULL; DisableInterrupts(); VoiceNode *voice; if ((voice = MV_GetVoice(handle)) == NULL) { RestoreInterrupts(); MV_SetErrorCode(MV_VoiceNotFound); return NULL; } return voice; } static inline void MV_EndService(void) { RestoreInterrupts(); } int32_t MV_VoicePlaying(int32_t handle) { return (MV_Installed && MV_GetVoice(handle)) ? TRUE : FALSE; } int32_t MV_KillAllVoices(void) { if (!MV_Installed) return MV_Error; DisableInterrupts(); if (&VoiceList == VoiceList.next) { RestoreInterrupts(); return MV_Ok; } VoiceNode * voice = VoiceList.prev; // Remove all the voices from the list while (voice != &VoiceList) { if (voice->priority == MV_MUSIC_PRIORITY) { voice = voice->prev; continue; } MV_Kill(voice->handle); voice = VoiceList.prev; } RestoreInterrupts(); return MV_Ok; } int32_t MV_Kill(int32_t handle) { VoiceNode *voice = MV_BeginService(handle); if (voice == NULL) return MV_Error; uint32_t const callbackval = voice->callbackval; MV_StopVoice(voice); MV_EndService(); if (MV_CallBackFunc) MV_CallBackFunc(callbackval); return MV_Ok; } int32_t MV_VoicesPlaying(void) { if (!MV_Installed) return 0; DisableInterrupts(); int NumVoices = 0; for (VoiceNode *voice = VoiceList.next; voice != &VoiceList; voice = voice->next) NumVoices++; RestoreInterrupts(); return NumVoices; } VoiceNode *MV_AllocVoice(int32_t priority) { VoiceNode *voice, *node; DisableInterrupts(); // Check if we have any free voices if (LL_Empty(&VoicePool, next, prev)) { // check if we have a higher priority than a voice that is playing. for (voice = node = VoiceList.next; node != &VoiceList; node = node->next) { if (node->priority < voice->priority) voice = node; } if (priority >= voice->priority && voice != &VoiceList && voice->handle >= MV_MINVOICEHANDLE) MV_Kill(voice->handle); if (LL_Empty(&VoicePool, next, prev)) { // No free voices RestoreInterrupts(); return NULL; } } voice = VoicePool.next; LL_Remove(voice, next, prev); RestoreInterrupts(); int32_t vhan = MV_MINVOICEHANDLE; // Find a free voice handle do { if (++vhan < MV_MINVOICEHANDLE || vhan > MV_MaxVoices) vhan = MV_MINVOICEHANDLE; } while (MV_VoicePlaying(vhan)); voice->handle = vhan; return voice; } int32_t MV_VoiceAvailable(int32_t priority) { // Check if we have any free voices if (!LL_Empty(&VoicePool, next, prev)) return TRUE; DisableInterrupts(); VoiceNode *voice, *node; // check if we have a higher priority than a voice that is playing. for (voice = node = VoiceList.next; node != &VoiceList; node = node->next) { if (node->priority < voice->priority) voice = node; } if ((voice == &VoiceList) || (priority < voice->priority)) { RestoreInterrupts(); return FALSE; } RestoreInterrupts(); return TRUE; } void MV_SetVoicePitch(VoiceNode *voice, uint32_t rate, int32_t pitchoffset) { voice->SamplingRate = rate; voice->PitchScale = PITCH_GetScale(pitchoffset); voice->RateScale = (rate * voice->PitchScale) / MV_MixRate; // Multiply by MV_MIXBUFFERSIZE - 1 voice->FixedPointBufferSize = (voice->RateScale * MV_MIXBUFFERSIZE) - voice->RateScale; } int32_t MV_SetPitch(int32_t handle, int32_t pitchoffset) { VoiceNode *voice = MV_BeginService(handle); if (voice == NULL) return MV_Error; MV_SetVoicePitch(voice, voice->SamplingRate, pitchoffset); MV_EndService(); return MV_Ok; } int32_t MV_SetFrequency(int32_t handle, int32_t frequency) { VoiceNode *voice = MV_BeginService(handle); if (voice == NULL) return MV_Error; MV_SetVoicePitch(voice, frequency, 0); MV_EndService(); return MV_Ok; } static inline const int16_t *MV_GetVolumeTable(int32_t vol) { return MV_VolumeTable[MIX_VOLUME(vol)]; } /*--------------------------------------------------------------------- Function: MV_SetVoiceMixMode Selects which method should be used to mix the voice. 16Bit 16Bit | 8Bit 16Bit 8Bit 16Bit | Mono Ster | Mono Mono Ster Ster | Mixer Out Out | In In In In | ----------------------+---------------------------+------------- X | X | Mix16BitMono16 X | X | Mix16BitMono X | X | Mix16BitStereo16 X | X | Mix16BitStereo ----------------------+---------------------------+------------- X | X | Mix16BitStereo16Stereo X | X | Mix16BitStereo8Stereo X | X | Mix16BitMono16Stereo X | X | Mix16BitMono8Stereo ---------------------------------------------------------------------*/ void MV_SetVoiceMixMode(VoiceNode *voice) { int32_t type = T_DEFAULT; if (MV_Channels == 1) type |= T_MONO; else { if (IS_QUIET(voice->RightVolume)) type |= T_RIGHTQUIET; else if (IS_QUIET(voice->LeftVolume)) type |= T_LEFTQUIET; } if (voice->bits == 16) type |= T_16BITSOURCE; if (voice->channels == 2) { type |= T_STEREOSOURCE; type &= ~(T_RIGHTQUIET | T_LEFTQUIET); } switch (type) { case T_16BITSOURCE | T_LEFTQUIET: MV_LeftVolume = MV_RightVolume; case T_16BITSOURCE | T_MONO: case T_16BITSOURCE | T_RIGHTQUIET: voice->mix = MV_Mix16BitMono16; break; case T_LEFTQUIET: MV_LeftVolume = MV_RightVolume; case T_MONO: case T_RIGHTQUIET: voice->mix = MV_Mix16BitMono; break; case T_16BITSOURCE: voice->mix = MV_Mix16BitStereo16; break; case T_SIXTEENBIT_STEREO: voice->mix = MV_Mix16BitStereo; break; case T_16BITSOURCE | T_STEREOSOURCE: voice->mix = MV_Mix16BitStereo16Stereo; break; case T_16BITSOURCE | T_STEREOSOURCE | T_MONO: voice->mix = MV_Mix16BitMono16Stereo; break; case T_STEREOSOURCE: voice->mix = MV_Mix16BitStereo8Stereo; break; case T_STEREOSOURCE | T_MONO: voice->mix = MV_Mix16BitMono8Stereo; break; default: voice->mix = NULL; break; } } void MV_SetVoiceVolume(VoiceNode *voice, int32_t vol, int32_t left, int32_t right) { if (MV_Channels == 1) left = right = vol; voice->LeftVolume = MV_GetVolumeTable(left); if (left == right) voice->RightVolume = voice->LeftVolume; else { voice->RightVolume = MV_GetVolumeTable(right); if (MV_ReverseStereo) swapptr(&voice->LeftVolume, &voice->RightVolume); } MV_SetVoiceMixMode(voice); } int32_t MV_PauseVoice(int32_t handle, int32_t pause) { VoiceNode *voice = MV_BeginService(handle); if (voice == NULL) return MV_Error; voice->Paused = pause; MV_EndService(); return MV_Ok; } int32_t MV_GetPosition(int32_t handle, int32_t *position) { VoiceNode *voice = MV_BeginService(handle); if (voice == NULL) return MV_Error; switch (voice->wavetype) { #ifdef HAVE_VORBIS case FMT_VORBIS: *position = MV_GetVorbisPosition(voice); break; #endif #ifdef HAVE_FLAC case FMT_FLAC: *position = MV_GetFLACPosition(voice); break; #endif case FMT_XA: *position = MV_GetXAPosition(voice); break; default: break; } MV_EndService(); return MV_Ok; } int32_t MV_SetPosition(int32_t handle, int32_t position) { VoiceNode *voice = MV_BeginService(handle); if (voice == NULL) return MV_Error; switch (voice->wavetype) { #ifdef HAVE_VORBIS case FMT_VORBIS: MV_SetVorbisPosition(voice, position); break; #endif #ifdef HAVE_FLAC case FMT_FLAC: MV_SetFLACPosition(voice, position); break; #endif case FMT_XA: MV_SetXAPosition(voice, position); break; default: break; } MV_EndService(); return MV_Ok; } int32_t MV_EndLooping(int32_t handle) { VoiceNode *voice = MV_BeginService(handle); if (voice == NULL) return MV_Error; voice->LoopCount = 0; voice->LoopStart = NULL; voice->LoopEnd = NULL; MV_EndService(); return MV_Ok; } int32_t MV_SetPan(int32_t handle, int32_t vol, int32_t left, int32_t right) { VoiceNode *voice = MV_BeginService(handle); if (voice == NULL) return MV_Error; MV_SetVoiceVolume(voice, vol, left, right); MV_EndService(); return MV_Ok; } int32_t MV_Pan3D(int32_t handle, int32_t angle, int32_t distance) { if (distance < 0) { distance = -distance; angle += MV_NUMPANPOSITIONS / 2; } int const volume = MIX_VOLUME(distance); angle &= MV_MAXPANPOSITION; return MV_SetPan(handle, max(0, 255 - distance), MV_PanTable[ angle ][ volume ].left, MV_PanTable[ angle ][ volume ].right); } void MV_SetReverb(int32_t reverb) { MV_ReverbLevel = MIX_VOLUME(reverb); MV_ReverbTable = &MV_VolumeTable[MV_ReverbLevel][0]; } int32_t MV_GetMaxReverbDelay(void) { return MV_MIXBUFFERSIZE * MV_NumberOfBuffers; } int32_t MV_GetReverbDelay(void) { return MV_ReverbDelay / MV_SampleSize; } void MV_SetReverbDelay(int32_t delay) { MV_ReverbDelay = max(MV_MIXBUFFERSIZE, min(delay, MV_GetMaxReverbDelay())) * MV_SampleSize; } static int32_t MV_SetMixMode(int32_t numchannels) { if (!MV_Installed) return MV_Error; MV_Channels = 1 + (numchannels == 2); MV_SampleSize = sizeof(int8_t) * MV_Channels * 2; MV_BufferSize = MV_MIXBUFFERSIZE * MV_SampleSize; MV_NumberOfBuffers = MV_TOTALBUFFERSIZE / MV_BufferSize; MV_BufferLength = MV_TOTALBUFFERSIZE; MV_RightChannelOffset = MV_SampleSize >> 1; return MV_Ok; } static int32_t MV_StartPlayback(void) { // Initialize the buffers Bmemset(MV_MixBuffer[0], 0, MV_TOTALBUFFERSIZE); for (int buffer = 0; buffer < MV_NumberOfBuffers; buffer++) MV_BufferEmpty[buffer] = TRUE; MV_MixPage = 1; if (SoundDriver_BeginPlayback(MV_MixBuffer[0], MV_BufferSize, MV_NumberOfBuffers, MV_ServiceVoc) != MV_Ok) { MV_SetErrorCode(MV_DriverError); return MV_Error; } return MV_Ok; } static void MV_StopPlayback(void) { SoundDriver_StopPlayback(); // Make sure all callbacks are done. DisableInterrupts(); for (VoiceNode *voice = VoiceList.next, *next; voice != &VoiceList; voice = next) { next = voice->next; MV_StopVoice(voice); if (MV_CallBackFunc) MV_CallBackFunc(voice->callbackval); } RestoreInterrupts(); } static void MV_CalcVolume(int32_t MaxVolume) { // For each volume level, create a translation table with the // appropriate volume calculated. for (int volume = 0; volume <= MV_MAXVOLUME; volume++) { int const level = (volume * MaxVolume) / MV_MAXTOTALVOLUME; for (int i = 0; i < 65536; i += 256) MV_VolumeTable[volume][i / 256] = ((i - 0x8000) * level) / MV_MAXVOLUME; } } static void MV_CalcPanTable(void) { const int32_t HalfAngle = MV_NUMPANPOSITIONS / 2; const int32_t QuarterAngle = HalfAngle / 2; for (int distance = 0; distance <= MV_MAXVOLUME; distance++) { const int32_t level = (255 * (MV_MAXVOLUME - distance)) / MV_MAXVOLUME; for (int angle = 0; angle <= QuarterAngle; angle++) { const int32_t ramp = level - (level * angle) / QuarterAngle; MV_PanTable[angle][distance].left = ramp; MV_PanTable[angle][distance].right = level; MV_PanTable[HalfAngle - angle][distance].left = ramp; MV_PanTable[HalfAngle - angle][distance].right = level; MV_PanTable[HalfAngle + angle][distance].left = level; MV_PanTable[HalfAngle + angle][distance].right = ramp; MV_PanTable[MV_MAXPANPOSITION - angle][distance].left = level; MV_PanTable[MV_MAXPANPOSITION - angle][distance].right = ramp; } } } void MV_SetVolume(int32_t volume) { MV_TotalVolume = min(max(0, volume), MV_MAXTOTALVOLUME); MV_CalcVolume(MV_TotalVolume); } int32_t MV_GetVolume(void) { return MV_TotalVolume; } void MV_SetCallBack(void (*function)(uint32_t)) { MV_CallBackFunc = function; } void MV_SetReverseStereo(int32_t setting) { MV_ReverseStereo = setting; } int32_t MV_GetReverseStereo(void) { return MV_ReverseStereo; } int32_t MV_Init(int32_t soundcard, int32_t MixRate, int32_t Voices, int32_t numchannels, void *initdata) { if (MV_Installed) MV_Shutdown(); MV_SetErrorCode(MV_Ok); // MV_TotalMemory + 2: FIXME, see valgrind_errors.log int const totalmem = Voices * sizeof(VoiceNode) + MV_TOTALBUFFERSIZE + 2; char *ptr = (char *) Xaligned_alloc(16, totalmem); if (!ptr) { MV_SetErrorCode(MV_NoMem); return MV_Error; } Bmemset(ptr, 0, totalmem); MV_Voices = (VoiceNode *)ptr; ptr += Voices * sizeof(VoiceNode); MV_MaxVoices = Voices; LL_Reset((VoiceNode*) &VoiceList, next, prev); LL_Reset((VoiceNode*) &VoicePool, next, prev); for (int index = 0; index < Voices; index++) LL_Add((VoiceNode*) &VoicePool, &MV_Voices[ index ], next, prev); MV_SetReverseStereo(FALSE); ASS_SoundDriver = soundcard; // Initialize the sound card if (SoundDriver_Init(&MixRate, &numchannels, initdata) != MV_Ok) MV_SetErrorCode(MV_DriverError); if (MV_ErrorCode != MV_Ok) { ALIGNED_FREE_AND_NULL(MV_Voices); return MV_Error; } MV_Installed = TRUE; MV_CallBackFunc = NULL; MV_ReverbLevel = 0; MV_ReverbTable = NULL; // Set the sampling rate MV_MixRate = MixRate; // Set Mixer to play stereo digitized sound MV_SetMixMode(numchannels); MV_ReverbDelay = MV_BufferSize * 3; // Make sure we don't cross a physical page MV_MixBuffer[ MV_NumberOfBuffers ] = ptr; for (int buffer = 0; buffer < MV_NumberOfBuffers; buffer++) { MV_MixBuffer[ buffer ] = ptr; ptr += MV_BufferSize; } // Calculate pan table MV_CalcPanTable(); MV_SetVolume(MV_MAXTOTALVOLUME); // Start the playback engine if (MV_StartPlayback() != MV_Ok) { // Preserve error code while we shutdown. int status = MV_ErrorCode; MV_Shutdown(); MV_SetErrorCode(status); return MV_Error; } return MV_Ok; } int32_t MV_Shutdown(void) { if (!MV_Installed) return MV_Ok; MV_KillAllVoices(); MV_Installed = FALSE; // Stop the sound playback engine MV_StopPlayback(); // Shutdown the sound card SoundDriver_Shutdown(); // Free any voices we allocated ALIGNED_FREE_AND_NULL(MV_Voices); LL_Reset((VoiceNode*) &VoiceList, next, prev); LL_Reset((VoiceNode*) &VoicePool, next, prev); MV_MaxVoices = 1; // Release the descriptor from our mix buffer for (int buffer = 0; buffer < MV_NUMBEROFBUFFERS; buffer++) MV_MixBuffer[ buffer ] = NULL; MV_SetErrorCode(MV_NotInstalled); return MV_Ok; } void MV_SetPrintf(void (*function)(const char *, ...)) { MV_Printf = function; } const char *loopStartTags[loopStartTagCount] = { "LOOP_START", "LOOPSTART", "LOOP" }; const char *loopEndTags[loopEndTagCount] = { "LOOP_END", "LOOPEND" }; const char *loopLengthTags[loopLengthTagCount] = { "LOOP_LENGTH", "LOOPLENGTH" };