LOVE/love
1
0
Fork 0
mirror of https://github.com/love2d/love.git synced 2026-08-16 16:20:42 +02:00
Code Issues Packages Projects Releases Wiki Activity

Don't use persistently mapped buffers on Intel Ivy Bridge GPUs on Windows. glClientWaitSync can hang with those drivers, apparently.

Browse Source
This commit is contained in:
Alex Szpakowski
2018-01-06 00:42:16 -04:00
parent fd514d8176
commit 3857d09132
3 changed files with 27 additions and 8 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/modules/graphics/opengl/StreamBuffer.cpp
+11 -8
View File
@@ -538,14 +538,17 @@ love::graphics::StreamBuffer *CreateStreamBuffer(BufferType mode, size_t size)
{
if (gl.isCoreProfile())
{
// AMD's pinned memory seems to be faster than persistent mapping, on
// AMD GPUs.
if (GLAD_AMD_pinned_memory)
return new StreamBufferPinnedMemory(mode, size);
else if (GLAD_VERSION_4_4 || GLAD_ARB_buffer_storage)
return new StreamBufferPersistentMapSync(mode, size);
else
return new StreamBufferSubDataOrphan(mode, size);
if (!gl.bugs.clientWaitSyncStalls)
{
// AMD's pinned memory seems to be faster than persistent mapping,
// on AMD GPUs.
if (GLAD_AMD_pinned_memory)
return new StreamBufferPinnedMemory(mode, size);
else if (GLAD_VERSION_4_4 || GLAD_ARB_buffer_storage)
return new StreamBufferPersistentMapSync(mode, size);
}
return new StreamBufferSubDataOrphan(mode, size);
}
else
return new StreamBufferClientMemory(mode, size);