Change: Do not disallow persistent buffer mapping on AMD GPUs, as it is actually faster. #8876
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Outdated suggestions cannot be applied.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.
Closes #8790
Motivation / Problem
At beginning, persistent buffer mapping was horribly slow on AMD GPUs (initally discovered on Linux Mesa drivers, but I now checked that same thing happened on Windows). As such, that workaround was added to disable persistent mapping when AMD is detected. This made performance on Mesa driver fine, but on propertiary driver on Windows it was better but still meh.
However, this was fixing the wrong problem. Meanwhile it was discovered that mapping was missing GL_CLIENT_STORAGE_BIT. With that bit set, performance is fine both on Mesa driver and Windows propertiary driver.
Description
Do not disallow persistent buffer mapping on AMD GPUs.
Limitations
Probably needs testing on Windows driver with dedicated GPU. I tested on Linux with dedicated GPU, and on Windows but on APU.
Checklist for review
Some things are not automated, and forgotten often. This list is a reminder for the reviewers.