On 12.12.2017 06:13, Timothy Arceri wrote:
Without this we get the error "FPExt only operates on FP" when
converting the following:
vec1 32 ssa_5 = b2f ssa_4
vec1 64 ssa_6 = f2f64 ssa_5
Which results in:
%44 = and i32 %43, 1065353216
%45 = fpext i32 %44 to double
With this patch we now get:
%44 = and i32 %43, 1065353216
%45 = bitcast i32 %44 to float
%46 = fpext float %45 to double
Yay for informative commit messages :)
Reviewed-by: Nicolai Hähnle <[email protected]>
---
src/amd/common/ac_nir_to_llvm.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/src/amd/common/ac_nir_to_llvm.c b/src/amd/common/ac_nir_to_llvm.c
index f5cf9d271d3..a1958b4332d 100644
--- a/src/amd/common/ac_nir_to_llvm.c
+++ b/src/amd/common/ac_nir_to_llvm.c
@@ -1770,20 +1770,21 @@ static void visit_alu(struct ac_nir_context *ctx, const
nir_alu_instr *instr)
case nir_op_i2f64:
src[0] = ac_to_integer(&ctx->ac, src[0]);
result = LLVMBuildSIToFP(ctx->ac.builder, src[0],
ac_to_float_type(&ctx->ac, def_type), "");
break;
case nir_op_u2f32:
case nir_op_u2f64:
src[0] = ac_to_integer(&ctx->ac, src[0]);
result = LLVMBuildUIToFP(ctx->ac.builder, src[0],
ac_to_float_type(&ctx->ac, def_type), "");
break;
case nir_op_f2f64:
+ src[0] = ac_to_float(&ctx->ac, src[0]);
result = LLVMBuildFPExt(ctx->ac.builder, src[0],
ac_to_float_type(&ctx->ac, def_type), "");
break;
case nir_op_f2f32:
result = LLVMBuildFPTrunc(ctx->ac.builder, src[0],
ac_to_float_type(&ctx->ac, def_type), "");
break;
case nir_op_u2u32:
case nir_op_u2u64:
src[0] = ac_to_integer(&ctx->ac, src[0]);
if (get_elem_bits(&ctx->ac, LLVMTypeOf(src[0])) <
get_elem_bits(&ctx->ac, def_type))
result = LLVMBuildZExt(ctx->ac.builder, src[0], def_type,
"");
--
Lerne, wie die Welt wirklich ist,
Aber vergiss niemals, wie sie sein sollte.
_______________________________________________
mesa-dev mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/mesa-dev
