You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on Nov 1, 2021. It is now read-only.
Right now each time the DRM backend wants to scan-out a client buffer it imports it via GBM. On some drivers I believe this degrades performance. For instance with #2240 and AMD, I end up with the compositor using near 100% CPU, spending a lot of time in Addr::V2::CoordEq::solve in radeonsi.so.
Instead we should import each buffer once, and then re-use the gbm_bo.
The text was updated successfully, but these errors were encountered:
Right now each time the DRM backend wants to scan-out a client buffer it imports it via GBM. On some drivers I believe this degrades performance. For instance with #2240 and AMD, I end up with the compositor using near 100% CPU, spending a lot of time in
Addr::V2::CoordEq::solve
inradeonsi.so
.Instead we should import each buffer once, and then re-use the
gbm_bo
.The text was updated successfully, but these errors were encountered: