Martijn van Oosterhout wrote: > The first is just clamps x2/y2 as previously discussed. It just hasn't > been committed yet. > > The second is more complicated. I've noticed that occasionally the > output view is offset in the Y direction from where the actual window > is. With some poking around I've come to the conclusion it's due to > the hardware viewport (as calculated by r128CalcViewport) not being > updated after resizing the window. What I think is happening is: > > 1. I resize window > 2. Resize event propegates down to the actual drawable containing the > OpenGL output > 3. User program does a new glViewport() > 4. This calls r128CalcViewport() but produces the wrong output because > drawable->h is not yet updated > 5. Sometime later the client program does some drawing which grabs the > lock and revalidates the context, updating the width/height values. > > All this patch does is call r128CalcViewport() again after the new > window coordinates have been detected. > > However, I've noticed other people with other drivers complaining > about a similar issue, so maybe there should be a general call by the > DRI code to ctx->Driver.Viewport().
I've checked in your patches. Perhaps r128CalcViewport() should grab the lock and get the current window pos/size at that point. -Brian Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 _______________________________________________ Mesa3d-dev mailing list Mesa3d-dev@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/mesa3d-dev
