gallivm: add fp64 support. (v2.1)

This adds support for ARB_gpu_shader_fp64 and ARB_vertex_attrib_64bit to
llvmpipe.

Two things that don't mix well are SoA and doubles, see
emit_fetch_double, and emit_store_double_chan in this.

I've also had to split emit_data.chan, to add src_chan,
which can be different for doubles.

It handles indirect double fetches from temps, inputs, constants
and immediates. It doesn't handle double stores to indirects,
however it appears the mesa/st doesn't currently emit these,
it always does UARL/MOV combos, which will work fine.

tested with piglit, no regressions, all the fp64 tests seem to pass.

v2:
switch to using shuffles for fetch/store (Roland)
assert on indirect double stores - mesa/st never emits these (it uses MOV)
fix indirect temp/input/constant/immediates (Roland)
typos/formatting fixes (Roland)

v2.1:
cleanup some long lines, emit_store_double_chan cleanups.

Reviewed-by: Roland Scheidegger <sroland@vmware.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>

1 files changed, 12 insertions, 0 deletions

diff --git a/src/gallium/auxiliary/gallivm/lp_bld_arit.c b/src/gallium/auxiliary/gallivm/lp_bld_arit.c
index 9daa93e..8fba43f 100644
--- a/src/gallium/auxiliary/gallivm/lp_bld_arit.c
+++ b/src/gallium/auxiliary/gallivm/lp_bld_arit.c
@@ -1997,6 +1997,12 @@ lp_build_floor(struct lp_build_context *bld,
       LLVMTypeRef int_vec_type = bld->int_vec_type;
       LLVMTypeRef vec_type = bld->vec_type;
 
+      if (type.width != 32) {
+         char intrinsic[32];
+         util_snprintf(intrinsic, sizeof intrinsic, "llvm.floor.v%uf%u", type.length, type.width);
+         return lp_build_intrinsic_unary(builder, intrinsic, vec_type, a);
+      }
+
       assert(type.width == 32); /* might want to handle doubles at some point */
 
       inttype = type;
@@ -2066,6 +2072,12 @@ lp_build_ceil(struct lp_build_context *bld,
       LLVMTypeRef int_vec_type = bld->int_vec_type;
       LLVMTypeRef vec_type = bld->vec_type;
 
+      if (type.width != 32) {
+         char intrinsic[32];
+         util_snprintf(intrinsic, sizeof intrinsic, "llvm.ceil.v%uf%u", type.length, type.width);
+         return lp_build_intrinsic_unary(builder, intrinsic, vec_type, a);
+      }
+
       assert(type.width == 32); /* might want to handle doubles at some point */
 
       inttype = type;