Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

merge consecutive lines when stippling #419

Open
wants to merge 1 commit into
base: main
Choose a base branch
from

Conversation

kusma
Copy link
Contributor

@kusma kusma commented Oct 17, 2023

When we draw the reference images, we split line segments into single-pixel length line segment, and we consider the stipple pattern for each segment individually. This leads to close to every covered pixel being considered an edge pixel, because we draw rectangles that are a single pixel long.

We should merge the consecutive line segments before rasterizing them, so the filled portions are considered as a whole.

Without this, the line-stippling tests are close to pointless; very close to every single covered pixel is allowed to be either filled or not, leading to very very few pixels actually being tested.

When we draw the reference images, we split line segments into
single-pixel length line segment, and we consider the stipple pattern
for each segment individually. This leads to close to every covered
pixel being considered an edge pixel, because we draw rectangles that
are a single pixel long.

We should merge the consecutive line segments before rasterizing them,
so the filled portions are considered as a whole.

Without this, the line-stippling tests are close to pointless; very
close to every single covered pixel is allowed to be either filled or
not, leading to very very few pixels actually being tested.
@kusma
Copy link
Contributor Author

kusma commented Oct 17, 2023

Here's an old MR I wrote when debugging CTS failures on Lavapipe... I guess it should be upstreamed.

I don't expect this to cause any regressions on real drivers, because there are some pixels being tested per segment, and it would be very unlikely that a driver would cover those but not the ones that are completely uncovered by accident. But I haven't tested this on much HW.

@kusma
Copy link
Contributor Author

kusma commented Feb 21, 2024

Ping?

@lordalcol
Copy link
Contributor

Hi @kusma I'm currently reviewing this MR.

@kusma
Copy link
Contributor Author

kusma commented Aug 2, 2024

Ping?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants