[Mesa-dev] [PATCH 2/2] ac/nir: Fix ordering of parameters for image atomic cmpswap intrinsics

[Bas Nieuwenhuizen]

Thanks! Pushed and cc'd it to stable.

Not pushing the first patch as I assume that is superseded by Connors patches.

On Fri, Jun 30, 2017 at 12:15 PM, Alex Smith
<asmith at feralinteractive.com> wrote:
> The NIR parameters are ordered "compare, data", matching GLSL, but both
> the image and buffer LLVM intrinsics take them the other way around.
> This is already handled correctly for SSBO atomics.
>
> Signed-off-by: Alex Smith <asmith at feralinteractive.com>
> ---
>  src/amd/common/ac_nir_to_llvm.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/src/amd/common/ac_nir_to_llvm.c b/src/amd/common/ac_nir_to_llvm.c
> index 6845df8..89b7103 100644
> --- a/src/amd/common/ac_nir_to_llvm.c
> +++ b/src/amd/common/ac_nir_to_llvm.c
> @@ -3442,9 +3442,9 @@ static LLVMValueRef visit_image_atomic(struct nir_to_llvm_context *ctx,
>                 abort();
>         }
>
> -       params[param_count++] = get_src(ctx, instr->src[2]);
>         if (instr->intrinsic == nir_intrinsic_image_atomic_comp_swap)
>                 params[param_count++] = get_src(ctx, instr->src[3]);
> +       params[param_count++] = get_src(ctx, instr->src[2]);
>
>         if (glsl_get_sampler_dim(type) == GLSL_SAMPLER_DIM_BUF) {
>                 params[param_count++] = get_sampler_desc(ctx, instr->variables[0], DESC_BUFFER);
> --
> 2.9.4
>