Re: Skew bug with egl

2012-08-20 Thread Pekka Paalanen
On Sun, 19 Aug 2012 22:49:12 -0700 Bill Spitzak wrote: > Also why are wl_egl calls in mesa? The source code seems to not show any > reason for them to be there rather than in wayland. I think it would make a > lot more sense for them to be in wayland and it certainly looks possible. > wl_egl_

Re: Skew bug with egl

2012-08-20 Thread Bill Spitzak
On 08/14/2012 03:29 PM, Scott Moreau wrote: Possibly a nouveau specific bug? It works fine here on radeon gallium and intel dri2. Scott I finally gave up and went back to the nvidia proprietary drivers. Wayland works fine with EGL_SOFTWARE=1. It's not fast but I am just trying to figure out

Re: Skew bug with egl

2012-08-19 Thread Bill Spitzak
On Aug 14, 2012, at 11:21 PM, Scott Moreau wrote: > That looks like a driver bug to me Any hints where to look? Is this in Mesa? Is it in the "gallium" subdirectory of mesa? I grepped for the word "stride" but it occurs hundreds of times. My best guess as to what is happening is that the strid

Re: Skew bug with egl

2012-08-15 Thread Scott Moreau
That looks like a driver bug to me ___ wayland-devel mailing list wayland-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/wayland-devel

Re: Skew bug with egl

2012-08-14 Thread Scott Moreau
Maybe a screenshot would help ___ wayland-devel mailing list wayland-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/wayland-devel

Re: Skew bug with egl

2012-08-14 Thread Scott Moreau
Possibly a nouveau specific bug? It works fine here on radeon gallium and intel dri2. Scott ___ wayland-devel mailing list wayland-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/wayland-devel

Re: Skew bug with egl

2012-08-14 Thread Bill Spitzak
Oops. I described the shift backwards, it is actually -width mod 16. If the surface is N*16 wide it is correct. If the surface is N*16-1 wide then it is skewed with each scanline 1 pixel to the right of the one above it. The writing code thinks the stride is N*16, while the reading code thinks

Re: Skew bug with egl

2012-08-14 Thread Bill Spitzak
Further information: Experimenting with resizing clickdot, it looks like it always skews such that line N is shifted width%16 pixels to the right of line N-1, where 0 is the top of the window. I also see a width%16 line of garbage between the skewed left/right edges of the surface. This impl

Skew bug with egl

2012-08-14 Thread Bill Spitzak
I have updated to new Ubuntu. This seems to have enabled wayland clients to use egl rather than shm. However now every client comes out skewed (sharply enough that the windows are unreadable but you see the colors and can guess what is going on). This previously only happened with simple-egl.