When assigning numbers to registers, skip any with neither uses nor defs. This is will not have any impact at all on the final SASS but it makes for slightly more readable PTX. This change should also ensure that future minor changes are less likely to cause noisy diffs in register numbering.
25 lines
837 B
LLVM
25 lines
837 B
LLVM
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --version 5
|
|
; RUN: llc < %s -mtriple=nvptx64 -mcpu=sm_20 | FileCheck %s
|
|
; RUN: %if ptxas %{ llc < %s -mtriple=nvptx64 -mcpu=sm_20 | %ptxas-verify %}
|
|
|
|
declare <4 x float> @bar()
|
|
|
|
define void @foo(ptr %ptr) {
|
|
; CHECK-LABEL: foo(
|
|
; CHECK: {
|
|
; CHECK-NEXT: .reg .b64 %rd<4>;
|
|
; CHECK-EMPTY:
|
|
; CHECK-NEXT: // %bb.0:
|
|
; CHECK-NEXT: ld.param.b64 %rd1, [foo_param_0];
|
|
; CHECK-NEXT: { // callseq 0, 0
|
|
; CHECK-NEXT: .param .align 16 .b8 retval0[16];
|
|
; CHECK-NEXT: call.uni (retval0), bar, ();
|
|
; CHECK-NEXT: ld.param.v2.b64 {%rd2, %rd3}, [retval0];
|
|
; CHECK-NEXT: } // callseq 0
|
|
; CHECK-NEXT: st.v2.b64 [%rd1], {%rd2, %rd3};
|
|
; CHECK-NEXT: ret;
|
|
%val = tail call <4 x float> @bar()
|
|
store <4 x float> %val, ptr %ptr
|
|
ret void
|
|
}
|