Re: [Mesa-dev] [PATCH] st/nir: Disable varying packing when doing transform feedback.

2018-07-05 Thread Eric Anholt
Timothy Arceri writes: > On 03/07/18 05:51, Eric Anholt wrote: > >> Eric Anholt writes: >> >>> [ Unknown signature status ] >>> Timothy Arceri writes: >>> nir_compact_varyings() is meant to skip over varyings used by xfb: /* We can't repack xfb varyings. */

Re: [Mesa-dev] [PATCH] st/nir: Disable varying packing when doing transform feedback.

2018-07-03 Thread Timothy Arceri
On 03/07/18 05:51, Eric Anholt wrote: Eric Anholt writes: [ Unknown signature status ] Timothy Arceri writes: nir_compact_varyings() is meant to skip over varyings used by xfb: /* We can't repack xfb varyings. */ if (var->data.always_active_io) continue

Re: [Mesa-dev] [PATCH] st/nir: Disable varying packing when doing transform feedback.

2018-07-02 Thread Eric Anholt
Eric Anholt writes: > [ Unknown signature status ] > Timothy Arceri writes: > >> nir_compact_varyings() is meant to skip over varyings used by xfb: >> >> /* We can't repack xfb varyings. */ >> if (var->data.always_active_io) >> continue; >> >> Any idea why that i

Re: [Mesa-dev] [PATCH] st/nir: Disable varying packing when doing transform feedback.

2018-06-21 Thread Eric Anholt
Timothy Arceri writes: > nir_compact_varyings() is meant to skip over varyings used by xfb: > > /* We can't repack xfb varyings. */ > if (var->data.always_active_io) > continue; > > Any idea why that isn't working in this case? Looks like GLSL IR has that flag wr

Re: [Mesa-dev] [PATCH] st/nir: Disable varying packing when doing transform feedback.

2018-06-20 Thread Timothy Arceri
nir_compact_varyings() is meant to skip over varyings used by xfb: /* We can't repack xfb varyings. */ if (var->data.always_active_io) continue; Any idea why that isn't working in this case? On 21/06/18 06:30, Eric Anholt wrote: If one of the TF variables got vary

[Mesa-dev] [PATCH] st/nir: Disable varying packing when doing transform feedback.

2018-06-20 Thread Eric Anholt
If one of the TF variables got varying-packed, it would end up with a new driver_location, despite the pipe_stream_output already being set up for the old driver location. This left the gallium driver with no way to work back to what varying was referenced by pipe_stream_output. Fixes these tests