scl/strato
1
0
Fork 0
mirror of https://github.com/Takiiiiiiii/strato.git synced 2025-07-21 13:37:23 +00:00
Code Issues Packages Projects Releases Wiki Activity

Use a linear allocator for most per-execution GPU allocations

Browse Source 
Currently we heavily thrash the heap each draw, with malloc/free taking up about 10% of GPFIFOs execution time. Using a linear allocator for the main offenders of buffer usage callbacks and index/vertex state helps to reduce this to about 4%
This commit is contained in:
Billy Laws
2022-07-31 13:41:28 +01:00
parent 70eec5a414
commit 683cd594ad
6 changed files with 22 additions and 20 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
app/src/main/cpp/skyline/gpu/buffer_manager.cpp
View File

@ -152,8 +152,9 @@ namespace skyline::gpu {
// Transfer all delegates references from the overlapping buffer to the new buffer
for (auto &delegate : srcBuffer->delegates) {
delegate->buffer = *newBuffer;
if (delegate->usageCallback)
delegate->usageCallback(*delegate->view, *newBuffer);
if (delegate->usageCallbacks)
for (auto &callback : *delegate->usageCallbacks)
callback(*delegate->view, *newBuffer);
}
newBuffer->delegates.splice(newBuffer->delegates.end(), srcBuffer->delegates);