gallivm: consolidate some half-to-float and r11g11b10-to-float code
authorRoland Scheidegger <sroland@vmware.com>
Fri, 29 Mar 2013 05:16:33 +0000 (06:16 +0100)
committerRoland Scheidegger <sroland@vmware.com>
Fri, 29 Mar 2013 15:39:40 +0000 (16:39 +0100)
commit5f41e08cf39d585d600aa506cdcd2f5380c60ddd
tree21d5737a38f0f87fad86c95aa0207e024ba260e0
parent4412f3bc13886751f91f265babd78bac28c7ba41
gallivm: consolidate some half-to-float and r11g11b10-to-float code

Similar enough that we can try to use shared code.
v2: fix a stupid bug using wrong variable causing mayhem with Inf and NaNs.

Reviewed-by: Jose Fonseca <jfonseca@vmware.com
src/gallium/auxiliary/gallivm/lp_bld_conv.c
src/gallium/auxiliary/gallivm/lp_bld_format.h
src/gallium/auxiliary/gallivm/lp_bld_format_float.c