gallivm: some bits of seamless cube filtering implementation
authorRoland Scheidegger <sroland@vmware.com>
Fri, 13 Sep 2013 17:52:09 +0000 (19:52 +0200)
committerRoland Scheidegger <sroland@vmware.com>
Tue, 17 Sep 2013 22:00:37 +0000 (00:00 +0200)
commit93b5f7117962ce6f30f4970eb7177b6755f90181
tree821a64003b01797e80315c5c4c5edd7be5140dad
parentb8244b00569828f62137693f251297bc83dee67c
gallivm: some bits of seamless cube filtering implementation

Simply adjust wrap mode to clamp_to_edge. This is all that's needed for a
correct implementation for nearest filtering, and it's way better than
using repeat wrap for instance for linear filtering (though obviously this
doesn't actually do seamless filtering).

v2: fix s/t wrap not r/s...

Reviewed-by: Brian Paul <brianp@vmware.com>
Reviewed-by: Jose Fonseca <jfonseca@vmware.com>
src/gallium/auxiliary/gallivm/lp_bld_sample.c
src/gallium/auxiliary/gallivm/lp_bld_sample.h
src/gallium/auxiliary/gallivm/lp_bld_sample_soa.c