gallivm/format: convert unsigned values to float properly.
authorDave Airlie <airlied@redhat.com>
Tue, 9 Jun 2020 03:58:59 +0000 (13:58 +1000)
committerDave Airlie <airlied@redhat.com>
Thu, 11 Jun 2020 04:41:14 +0000 (14:41 +1000)
commit45606ee80497be34f080e784831504c8642416a8
treea7535e54d66bc5baa19e23a330136107a0bce3ba
parenta2c16ecb2ee7f00dff17dcf658935f5b3cb03626
gallivm/format: convert unsigned values to float properly.

This fixes:
dEQP-GLES31.functional.draw_indirect.random.2

which ends up with 3x32-bit USCALED values going down this path
some of which have the top bit set, and end up converted to signed
float instead of unsigned float values.

Reviewed-by: Roland Scheidegger <sroland@vmware.com>
Part-of: <https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/5379>
.gitlab-ci/deqp-virgl-fails.txt
src/gallium/auxiliary/gallivm/lp_bld_format_soa.c