piwalker/dolphin
1
0
Fork 0
mirror of https://github.com/dolphin-emu/dolphin.git synced 2025-07-23 22:29:39 -06:00
Code Issues Packages Projects Releases Wiki Activity

implement some code to reduce the amounts of calls to setup vertex format, in d3d9 it gives no noticeable speedup, in opengl it still does not work right.

Browse Source 
thanks to neobrain for the idea
This commit is contained in:
rodolfoosvaldobogado
2012-10-26 23:18:09 -03:00
parent eaa1ea71c1
commit ee72852491
9 changed files with 305 additions and 110 deletions
Download Patch File Download Diff File Expand all files Collapse all files

96
Source/Plugins/Plugin_VideoOGL/Src/VertexManager.cpp
View File

@ -114,8 +114,9 @@ void VertexManager::CreateDeviceObjects()
m_buffers_count = MAX_VBUFFER_COUNT;
m_current_index_buffer = 0;
m_current_vertex_buffer = 0;
m_index_buffer_cursor = 0;
m_vertex_buffer_cursor = 0;
m_index_buffer_cursor = m_index_buffer_size;
m_vertex_buffer_cursor = m_vertex_buffer_size;
m_CurrentVertexFmt = NULL;
}
void VertexManager::DestroyDeviceObjects()
{
@ -153,18 +154,20 @@ void VertexManager::PrepareDrawBuffers(u32 stride)
int line_index_size = IndexGenerator::GetLineindexLen() * sizeof(u16);
int point_index_size = IndexGenerator::GetPointindexLen() * sizeof(u16);
int index_data_size = triangle_index_size + line_index_size + point_index_size;
GLbitfield LockMode = GL_MAP_WRITE_BIT;
GLbitfield LockMode = GL_MAP_WRITE_BIT;
m_vertex_buffer_cursor--;
m_vertex_buffer_cursor = m_vertex_buffer_cursor - (m_vertex_buffer_cursor % stride) + stride;
if (m_vertex_buffer_cursor > m_vertex_buffer_size - vertex_data_size)
{
LockMode |= GL_MAP_INVALIDATE_BUFFER_BIT;
m_vertex_buffer_cursor = 0;
m_current_vertex_buffer = (m_current_vertex_buffer + 1) % m_buffers_count;
glBindBuffer(GL_ARRAY_BUFFER, m_vertex_buffers[m_current_vertex_buffer]);
}
else
{
LockMode |= GL_MAP_INVALIDATE_RANGE_BIT | GL_MAP_UNSYNCHRONIZED_BIT;
}
glBindBuffer(GL_ARRAY_BUFFER, m_vertex_buffers[m_current_vertex_buffer]);
}
if(GLEW_ARB_map_buffer_range)
{
pVertices = (u8*)glMapBufferRange(GL_ARRAY_BUFFER, m_vertex_buffer_cursor, vertex_data_size, LockMode);
@ -189,13 +192,13 @@ void VertexManager::PrepareDrawBuffers(u32 stride)
{
LockMode |= GL_MAP_INVALIDATE_BUFFER_BIT;
m_index_buffer_cursor = 0;
m_current_index_buffer = (m_current_index_buffer + 1) % m_buffers_count;
m_current_index_buffer = (m_current_index_buffer + 1) % m_buffers_count;
glBindBuffer(GL_ELEMENT_ARRAY_BUFFER, m_index_buffers[m_current_index_buffer]);
}
else
{
LockMode |= GL_MAP_INVALIDATE_RANGE_BIT | GL_MAP_UNSYNCHRONIZED_BIT;
}
glBindBuffer(GL_ELEMENT_ARRAY_BUFFER, m_index_buffers[m_current_index_buffer]);
}
if(GLEW_ARB_map_buffer_range)
{
pIndices = (u16*)glMapBufferRange(GL_ELEMENT_ARRAY_BUFFER, m_index_buffer_cursor , index_data_size, LockMode);
@ -279,11 +282,33 @@ void VertexManager::DrawVertexBufferObject()
}
}
void VertexManager::DrawVertexBufferObjectBase(u32 stride)
{
int triangle_index_size = IndexGenerator::GetTriangleindexLen();
int line_index_size = IndexGenerator::GetLineindexLen();
int point_index_size = IndexGenerator::GetPointindexLen();
int StartIndex = m_index_buffer_cursor;
if (triangle_index_size > 0)
{
glDrawElementsBaseVertex(GL_TRIANGLES, triangle_index_size, GL_UNSIGNED_SHORT, (GLvoid*)StartIndex, m_vertex_buffer_cursor / stride);
INCSTAT(stats.thisFrame.numIndexedDrawCalls);
}
if (line_index_size > 0)
{
StartIndex += triangle_index_size * sizeof(u16);
glDrawElementsBaseVertex(GL_LINES, line_index_size, GL_UNSIGNED_SHORT, (GLvoid*)StartIndex, m_vertex_buffer_cursor / stride);
INCSTAT(stats.thisFrame.numIndexedDrawCalls);
}
if (point_index_size > 0)
{
StartIndex += line_index_size * sizeof(u16);
glDrawElementsBaseVertex(GL_POINTS, point_index_size, GL_UNSIGNED_SHORT, (GLvoid*)StartIndex, m_vertex_buffer_cursor / stride);
INCSTAT(stats.thisFrame.numIndexedDrawCalls);
}
}
void VertexManager::vFlush()
{
if (LocalVBuffer == s_pCurBufferPointer) return;
if (Flushed) return;
Flushed=true;
VideoFifo_CheckEFBAccess();
#if defined(_DEBUG) || defined(DEBUGFAST)
PRIM_LOG("frame%d:\n texgen=%d, numchan=%d, dualtex=%d, ztex=%d, cole=%d, alpe=%d, ze=%d", g_ActiveConfig.iSaveTargetId, xfregs.numTexGen.numTexGens,
@ -318,13 +343,18 @@ void VertexManager::vFlush()
u32 stride = g_nativeVertexFmt->GetVertexStride();
PrepareDrawBuffers(stride);
if(m_buffers_count)
//still testing if this line is enabled to reduce the amount of vertex setup call everything goes wrong
//if(m_CurrentVertexFmt != g_nativeVertexFmt || !GLEW_ARB_draw_elements_base_vertex )
{
((GLVertexFormat*)g_nativeVertexFmt)->SetupVertexPointersOffset(m_vertex_buffer_cursor);
}
else
{
g_nativeVertexFmt->SetupVertexPointers();
if(m_buffers_count)
{
((GLVertexFormat*)g_nativeVertexFmt)->SetupVertexPointersOffset(GLEW_ARB_draw_elements_base_vertex ? 0 : m_vertex_buffer_cursor);
}
else
{
g_nativeVertexFmt->SetupVertexPointers();
}
m_CurrentVertexFmt = g_nativeVertexFmt;
}
GL_REPORT_ERRORD();
@ -402,7 +432,21 @@ void VertexManager::vFlush()
if (ps) PixelShaderCache::SetCurrentShader(ps->glprogid); // Lego Star Wars crashes here.
if (vs) VertexShaderCache::SetCurrentShader(vs->glprogid);
if(m_buffers_count) { DrawVertexBufferObject(); }else{ DrawVertexArray();};
if(m_buffers_count)
{
if(GLEW_ARB_draw_elements_base_vertex)
{
DrawVertexBufferObjectBase(stride);
}
else
{
DrawVertexBufferObject();
}
}
else
{
DrawVertexArray();
}
// run through vertex groups again to set alpha
if (useDstAlpha && !dualSourcePossible)
@ -415,7 +459,21 @@ void VertexManager::vFlush()
glDisable(GL_BLEND);
if(m_buffers_count) { DrawVertexBufferObject(); }else{ DrawVertexArray();};
if(m_buffers_count)
{
if(GLEW_ARB_draw_elements_base_vertex)
{
DrawVertexBufferObjectBase(stride);
}
else
{
DrawVertexBufferObject();
}
}
else
{
DrawVertexArray();
}
// restore color mask
g_renderer->SetColorMask();