Fix for EvictManagedOnUnlock & SWTOR performance #2126
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
Did a quick test with the build artifacts and it looks correct. Thanks! |
K0bin
changed the title
- Fix interleaved locks with evictManagedOnUnlock We need to make sure there are no other subresources of a texture locked before clearing the dirty box. Otherwise the data for those other subresources won't get copied into VRAM. - Clear dirty box regardless of texture pool Otherwise we keep repacking and copying the whole texture for every single lock. This causes performance problems in Star Wars: The Old Republic.
1 task
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Fixes #2125
Borderlands 2 locks textures like this:
which leads to us clearing the dirty box too soon and not copying the smaller mip maps over.
Fixes #2130
SWTOR locks a 1024x1024 D3DPOOL_DEFAULT, D3DFORMAT_A8, USAGE_DYNAMIC texture very often. (Probably cached font glyphs?). It properly sets the dirty rect whenever it makes a change.
On current master we never clear the dirty so it stays at the initial 1024x1024 rect. That means we keep repacking (from 4 aligned pitch managed buffer into tightly packed staging buffer) the entire image a bunch of times per frame. That's really slow.
If we just clear the dirty box like we do for managed textures, all we have to repack and copy are <30x30 parts of the texture.