[PR #3672] [MERGED] video_core: Better handling of image copies with DmaData #3643

New issue

Closed

opened 2026-02-27 22:04:28 +03:00 by kerem · 0 comments

kerem commented

2026-02-27 22:04:28 +03:00

Owner

📋 Pull Request Information

Original PR: https://github.com/shadps4-emu/shadPS4/pull/3672
Author: @raphaelthegreat
Created: 9/29/2025
Status: ✅ Merged
Merged: 9/30/2025
Merged by: @georgemoralis

Base: main ← Head: cursed-dma

📝 Commits (1)

d5bf4c7 video_core: Better handling of image copies with DmaData

📊 Changes

4 files changed (+55 additions, -93 deletions)

View changed files

📝 src/video_core/buffer_cache/buffer.cpp (+36 -0)
📝 src/video_core/buffer_cache/buffer.h (+8 -11)
📝 src/video_core/buffer_cache/buffer_cache.cpp (+11 -77)
📝 src/video_core/buffer_cache/buffer_cache.h (+0 -5)

📄 Description

The patch essentially does 2 things:

Removes cpu fast path for CopyBuffer. This is not that useful and would make copying render targets harder
Always invalidate texture cache on destination range

_{🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.}

## 📋 Pull Request Information **Original PR:** https://github.com/shadps4-emu/shadPS4/pull/3672 **Author:** [@raphaelthegreat](https://github.com/raphaelthegreat) **Created:** 9/29/2025 **Status:** ✅ Merged **Merged:** 9/30/2025 **Merged by:** [@georgemoralis](https://github.com/georgemoralis) **Base:** `main` ← **Head:** `cursed-dma` --- ### 📝 Commits (1) - [`d5bf4c7`](https://github.com/shadps4-emu/shadPS4/commit/d5bf4c72b0149c98c953402d1187b9145818418d) video_core: Better handling of image copies with DmaData ### 📊 Changes **4 files changed** (+55 additions, -93 deletions) <details> <summary>View changed files</summary> 📝 `src/video_core/buffer_cache/buffer.cpp` (+36 -0) 📝 `src/video_core/buffer_cache/buffer.h` (+8 -11) 📝 `src/video_core/buffer_cache/buffer_cache.cpp` (+11 -77) 📝 `src/video_core/buffer_cache/buffer_cache.h` (+0 -5) </details> ### 📄 Description The patch essentially does 2 things: - Removes cpu fast path for CopyBuffer. This is not that useful and would make copying render targets harder - Always invalidate texture cache on destination range --- <sub>🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.</sub>