aboutsummaryrefslogtreecommitdiff
path: root/llvm/test/CodeGen/AArch64/sme-streaming-interface.ll
diff options
context:
space:
mode:
Diffstat (limited to 'llvm/test/CodeGen/AArch64/sme-streaming-interface.ll')
-rw-r--r--llvm/test/CodeGen/AArch64/sme-streaming-interface.ll49
1 files changed, 31 insertions, 18 deletions
diff --git a/llvm/test/CodeGen/AArch64/sme-streaming-interface.ll b/llvm/test/CodeGen/AArch64/sme-streaming-interface.ll
index 465fb466..4321493 100644
--- a/llvm/test/CodeGen/AArch64/sme-streaming-interface.ll
+++ b/llvm/test/CodeGen/AArch64/sme-streaming-interface.ll
@@ -1,5 +1,5 @@
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
-; RUN: llc -mtriple=aarch64-linux-gnu -mattr=+sme -verify-machineinstrs < %s | FileCheck %s
+; RUN: llc -mtriple=aarch64-linux-gnu -mattr=+sve -mattr=+sme -verify-machineinstrs < %s | FileCheck %s
; This file tests the following combinations related to streaming-enabled functions:
; [ ] N -> S (Normal -> Streaming)
@@ -22,10 +22,11 @@ define void @normal_caller_streaming_callee() nounwind {
; CHECK-LABEL: normal_caller_streaming_callee:
; CHECK: // %bb.0:
; CHECK-NEXT: stp d15, d14, [sp, #-80]! // 16-byte Folded Spill
+; CHECK-NEXT: cntd x9
; CHECK-NEXT: stp d13, d12, [sp, #16] // 16-byte Folded Spill
; CHECK-NEXT: stp d11, d10, [sp, #32] // 16-byte Folded Spill
; CHECK-NEXT: stp d9, d8, [sp, #48] // 16-byte Folded Spill
-; CHECK-NEXT: str x30, [sp, #64] // 8-byte Folded Spill
+; CHECK-NEXT: stp x30, x9, [sp, #64] // 16-byte Folded Spill
; CHECK-NEXT: smstart sm
; CHECK-NEXT: bl streaming_callee
; CHECK-NEXT: smstop sm
@@ -47,10 +48,11 @@ define void @streaming_caller_normal_callee() nounwind "aarch64_pstate_sm_enable
; CHECK-LABEL: streaming_caller_normal_callee:
; CHECK: // %bb.0:
; CHECK-NEXT: stp d15, d14, [sp, #-80]! // 16-byte Folded Spill
+; CHECK-NEXT: cntd x9
; CHECK-NEXT: stp d13, d12, [sp, #16] // 16-byte Folded Spill
; CHECK-NEXT: stp d11, d10, [sp, #32] // 16-byte Folded Spill
; CHECK-NEXT: stp d9, d8, [sp, #48] // 16-byte Folded Spill
-; CHECK-NEXT: str x30, [sp, #64] // 8-byte Folded Spill
+; CHECK-NEXT: stp x30, x9, [sp, #64] // 16-byte Folded Spill
; CHECK-NEXT: smstop sm
; CHECK-NEXT: bl normal_callee
; CHECK-NEXT: smstart sm
@@ -103,10 +105,11 @@ define void @call_to_function_pointer_streaming_enabled(ptr %p) nounwind {
; CHECK-LABEL: call_to_function_pointer_streaming_enabled:
; CHECK: // %bb.0:
; CHECK-NEXT: stp d15, d14, [sp, #-80]! // 16-byte Folded Spill
+; CHECK-NEXT: cntd x9
; CHECK-NEXT: stp d13, d12, [sp, #16] // 16-byte Folded Spill
; CHECK-NEXT: stp d11, d10, [sp, #32] // 16-byte Folded Spill
; CHECK-NEXT: stp d9, d8, [sp, #48] // 16-byte Folded Spill
-; CHECK-NEXT: str x30, [sp, #64] // 8-byte Folded Spill
+; CHECK-NEXT: stp x30, x9, [sp, #64] // 16-byte Folded Spill
; CHECK-NEXT: smstart sm
; CHECK-NEXT: blr x0
; CHECK-NEXT: smstop sm
@@ -125,19 +128,20 @@ define <4 x i32> @smstart_clobber_simdfp(<4 x i32> %x) nounwind {
; CHECK-LABEL: smstart_clobber_simdfp:
; CHECK: // %bb.0:
; CHECK-NEXT: sub sp, sp, #96
+; CHECK-NEXT: cntd x9
; CHECK-NEXT: stp d15, d14, [sp, #16] // 16-byte Folded Spill
; CHECK-NEXT: stp d13, d12, [sp, #32] // 16-byte Folded Spill
; CHECK-NEXT: stp d11, d10, [sp, #48] // 16-byte Folded Spill
; CHECK-NEXT: stp d9, d8, [sp, #64] // 16-byte Folded Spill
-; CHECK-NEXT: str x30, [sp, #80] // 8-byte Folded Spill
+; CHECK-NEXT: stp x30, x9, [sp, #80] // 16-byte Folded Spill
; CHECK-NEXT: str q0, [sp] // 16-byte Folded Spill
; CHECK-NEXT: smstart sm
; CHECK-NEXT: bl streaming_callee
; CHECK-NEXT: smstop sm
-; CHECK-NEXT: ldp d9, d8, [sp, #64] // 16-byte Folded Reload
; CHECK-NEXT: ldr q0, [sp] // 16-byte Folded Reload
-; CHECK-NEXT: ldp d11, d10, [sp, #48] // 16-byte Folded Reload
+; CHECK-NEXT: ldp d9, d8, [sp, #64] // 16-byte Folded Reload
; CHECK-NEXT: ldr x30, [sp, #80] // 8-byte Folded Reload
+; CHECK-NEXT: ldp d11, d10, [sp, #48] // 16-byte Folded Reload
; CHECK-NEXT: ldp d13, d12, [sp, #32] // 16-byte Folded Reload
; CHECK-NEXT: ldp d15, d14, [sp, #16] // 16-byte Folded Reload
; CHECK-NEXT: add sp, sp, #96
@@ -150,7 +154,9 @@ define <4 x i32> @smstart_clobber_simdfp(<4 x i32> %x) nounwind {
define <vscale x 4 x i32> @smstart_clobber_sve(<vscale x 4 x i32> %x) nounwind {
; CHECK-LABEL: smstart_clobber_sve:
; CHECK: // %bb.0:
-; CHECK-NEXT: stp x29, x30, [sp, #-16]! // 16-byte Folded Spill
+; CHECK-NEXT: stp x29, x30, [sp, #-32]! // 16-byte Folded Spill
+; CHECK-NEXT: cntd x9
+; CHECK-NEXT: str x9, [sp, #16] // 8-byte Folded Spill
; CHECK-NEXT: addvl sp, sp, #-18
; CHECK-NEXT: str p15, [sp, #4, mul vl] // 2-byte Folded Spill
; CHECK-NEXT: str p14, [sp, #5, mul vl] // 2-byte Folded Spill
@@ -216,7 +222,7 @@ define <vscale x 4 x i32> @smstart_clobber_sve(<vscale x 4 x i32> %x) nounwind {
; CHECK-NEXT: ldr p5, [sp, #14, mul vl] // 2-byte Folded Reload
; CHECK-NEXT: ldr p4, [sp, #15, mul vl] // 2-byte Folded Reload
; CHECK-NEXT: addvl sp, sp, #18
-; CHECK-NEXT: ldp x29, x30, [sp], #16 // 16-byte Folded Reload
+; CHECK-NEXT: ldp x29, x30, [sp], #32 // 16-byte Folded Reload
; CHECK-NEXT: ret
call void @streaming_callee()
ret <vscale x 4 x i32> %x;
@@ -227,7 +233,9 @@ define <vscale x 4 x i32> @smstart_clobber_sve(<vscale x 4 x i32> %x) nounwind {
define <vscale x 4 x i32> @smstart_clobber_sve_duplicate(<vscale x 4 x i32> %x) nounwind {
; CHECK-LABEL: smstart_clobber_sve_duplicate:
; CHECK: // %bb.0:
-; CHECK-NEXT: stp x29, x30, [sp, #-16]! // 16-byte Folded Spill
+; CHECK-NEXT: stp x29, x30, [sp, #-32]! // 16-byte Folded Spill
+; CHECK-NEXT: cntd x9
+; CHECK-NEXT: str x9, [sp, #16] // 8-byte Folded Spill
; CHECK-NEXT: addvl sp, sp, #-18
; CHECK-NEXT: str p15, [sp, #4, mul vl] // 2-byte Folded Spill
; CHECK-NEXT: str p14, [sp, #5, mul vl] // 2-byte Folded Spill
@@ -296,7 +304,7 @@ define <vscale x 4 x i32> @smstart_clobber_sve_duplicate(<vscale x 4 x i32> %x)
; CHECK-NEXT: ldr p5, [sp, #14, mul vl] // 2-byte Folded Reload
; CHECK-NEXT: ldr p4, [sp, #15, mul vl] // 2-byte Folded Reload
; CHECK-NEXT: addvl sp, sp, #18
-; CHECK-NEXT: ldp x29, x30, [sp], #16 // 16-byte Folded Reload
+; CHECK-NEXT: ldp x29, x30, [sp], #32 // 16-byte Folded Reload
; CHECK-NEXT: ret
call void @streaming_callee()
call void @streaming_callee()
@@ -308,11 +316,12 @@ define double @call_to_intrinsic_without_chain(double %x) nounwind "aarch64_psta
; CHECK-LABEL: call_to_intrinsic_without_chain:
; CHECK: // %bb.0: // %entry
; CHECK-NEXT: sub sp, sp, #96
+; CHECK-NEXT: cntd x9
; CHECK-NEXT: stp d15, d14, [sp, #16] // 16-byte Folded Spill
; CHECK-NEXT: stp d13, d12, [sp, #32] // 16-byte Folded Spill
; CHECK-NEXT: stp d11, d10, [sp, #48] // 16-byte Folded Spill
; CHECK-NEXT: stp d9, d8, [sp, #64] // 16-byte Folded Spill
-; CHECK-NEXT: str x30, [sp, #80] // 8-byte Folded Spill
+; CHECK-NEXT: stp x30, x9, [sp, #80] // 16-byte Folded Spill
; CHECK-NEXT: stp d0, d0, [sp] // 16-byte Folded Spill
; CHECK-NEXT: smstop sm
; CHECK-NEXT: ldr d0, [sp] // 8-byte Folded Reload
@@ -320,11 +329,11 @@ define double @call_to_intrinsic_without_chain(double %x) nounwind "aarch64_psta
; CHECK-NEXT: str d0, [sp] // 8-byte Folded Spill
; CHECK-NEXT: smstart sm
; CHECK-NEXT: ldp d1, d0, [sp] // 16-byte Folded Reload
-; CHECK-NEXT: ldr x30, [sp, #80] // 8-byte Folded Reload
+; CHECK-NEXT: fadd d0, d1, d0
; CHECK-NEXT: ldp d9, d8, [sp, #64] // 16-byte Folded Reload
+; CHECK-NEXT: ldr x30, [sp, #80] // 8-byte Folded Reload
; CHECK-NEXT: ldp d11, d10, [sp, #48] // 16-byte Folded Reload
; CHECK-NEXT: ldp d13, d12, [sp, #32] // 16-byte Folded Reload
-; CHECK-NEXT: fadd d0, d1, d0
; CHECK-NEXT: ldp d15, d14, [sp, #16] // 16-byte Folded Reload
; CHECK-NEXT: add sp, sp, #96
; CHECK-NEXT: ret
@@ -342,10 +351,11 @@ define void @disable_tailcallopt() nounwind {
; CHECK-LABEL: disable_tailcallopt:
; CHECK: // %bb.0:
; CHECK-NEXT: stp d15, d14, [sp, #-80]! // 16-byte Folded Spill
+; CHECK-NEXT: cntd x9
; CHECK-NEXT: stp d13, d12, [sp, #16] // 16-byte Folded Spill
; CHECK-NEXT: stp d11, d10, [sp, #32] // 16-byte Folded Spill
; CHECK-NEXT: stp d9, d8, [sp, #48] // 16-byte Folded Spill
-; CHECK-NEXT: str x30, [sp, #64] // 8-byte Folded Spill
+; CHECK-NEXT: stp x30, x9, [sp, #64] // 16-byte Folded Spill
; CHECK-NEXT: smstart sm
; CHECK-NEXT: bl streaming_callee
; CHECK-NEXT: smstop sm
@@ -362,11 +372,13 @@ define void @disable_tailcallopt() nounwind {
define i8 @call_to_non_streaming_pass_sve_objects(ptr nocapture noundef readnone %ptr) #0 {
; CHECK-LABEL: call_to_non_streaming_pass_sve_objects:
; CHECK: // %bb.0: // %entry
-; CHECK-NEXT: stp d15, d14, [sp, #-80]! // 16-byte Folded Spill
+; CHECK-NEXT: stp d15, d14, [sp, #-96]! // 16-byte Folded Spill
+; CHECK-NEXT: cntd x9
; CHECK-NEXT: stp d13, d12, [sp, #16] // 16-byte Folded Spill
; CHECK-NEXT: stp d11, d10, [sp, #32] // 16-byte Folded Spill
; CHECK-NEXT: stp d9, d8, [sp, #48] // 16-byte Folded Spill
; CHECK-NEXT: stp x29, x30, [sp, #64] // 16-byte Folded Spill
+; CHECK-NEXT: str x9, [sp, #80] // 8-byte Folded Spill
; CHECK-NEXT: addvl sp, sp, #-3
; CHECK-NEXT: rdsvl x3, #1
; CHECK-NEXT: addvl x0, sp, #2
@@ -383,7 +395,7 @@ define i8 @call_to_non_streaming_pass_sve_objects(ptr nocapture noundef readnone
; CHECK-NEXT: ldp d9, d8, [sp, #48] // 16-byte Folded Reload
; CHECK-NEXT: ldp d11, d10, [sp, #32] // 16-byte Folded Reload
; CHECK-NEXT: ldp d13, d12, [sp, #16] // 16-byte Folded Reload
-; CHECK-NEXT: ldp d15, d14, [sp], #80 // 16-byte Folded Reload
+; CHECK-NEXT: ldp d15, d14, [sp], #96 // 16-byte Folded Reload
; CHECK-NEXT: ret
entry:
%Data1 = alloca <vscale x 16 x i8>, align 16
@@ -400,11 +412,12 @@ define void @call_to_non_streaming_pass_args(ptr nocapture noundef readnone %ptr
; CHECK-LABEL: call_to_non_streaming_pass_args:
; CHECK: // %bb.0: // %entry
; CHECK-NEXT: sub sp, sp, #112
+; CHECK-NEXT: cntd x9
; CHECK-NEXT: stp d15, d14, [sp, #32] // 16-byte Folded Spill
; CHECK-NEXT: stp d13, d12, [sp, #48] // 16-byte Folded Spill
; CHECK-NEXT: stp d11, d10, [sp, #64] // 16-byte Folded Spill
; CHECK-NEXT: stp d9, d8, [sp, #80] // 16-byte Folded Spill
-; CHECK-NEXT: str x30, [sp, #96] // 8-byte Folded Spill
+; CHECK-NEXT: stp x30, x9, [sp, #96] // 16-byte Folded Spill
; CHECK-NEXT: stp d2, d3, [sp, #16] // 16-byte Folded Spill
; CHECK-NEXT: stp s0, s1, [sp, #8] // 8-byte Folded Spill
; CHECK-NEXT: smstop sm
generated by cgit v1.1 at 2024-08-01 10:28:26 +0000