Reviewed-by: Samuel Pitoiset <[email protected]>
On 2/18/19 2:23 PM, Rhys Perry wrote:
I don't see a 16-bit version of tbuffer.store in IntrinsicsAMDGPU.td
and simply changing "llvm.amdgcn.tbuffer.store.i32" to
"llvm.amdgcn.tbuffer.store.i16" and removing the zext doesn't seem to
work.
On Mon, 18 Feb 2019 at 08:55, Samuel Pitoiset <[email protected]> wrote:
Does this fix anything know? There is a 16-bit version of tbuffer.store,
maybe we should use it?
On 2/16/19 1:21 AM, Rhys Perry wrote:
Signed-off-by: Rhys Perry <[email protected]>
---
src/amd/common/ac_nir_to_llvm.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/src/amd/common/ac_nir_to_llvm.c b/src/amd/common/ac_nir_to_llvm.c
index 89a78b43c6f..b260142c177 100644
--- a/src/amd/common/ac_nir_to_llvm.c
+++ b/src/amd/common/ac_nir_to_llvm.c
@@ -1586,6 +1586,8 @@ static void visit_store_ssbo(struct ac_nir_context *ctx,
} else if (num_bytes == 2) {
store_name = "llvm.amdgcn.tbuffer.store.i32";
data_type = ctx->ac.i32;
+ data = LLVMBuildBitCast(ctx->ac.builder, data, ctx->ac.i16,
"");
+ data = LLVMBuildZExt(ctx->ac.builder, data, data_type,
"");
LLVMValueRef tbuffer_params[] = {
data,
rsrc,
_______________________________________________
mesa-dev mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/mesa-dev
