2208c97c1b
Similar to 806761a7629df268c8aed49657aeccffa6bca449 -mtriple= specifies the full target triple while -march= merely sets the architecture part of the default target triple, leaving a target triple which may not make sense. Therefore, -march= is error-prone and not recommended for tests without a target triple. The issue has been benign as we recognize $unknown-apple-darwin as ELF instead of rejecting it outrightly.
31 lines
990 B
LLVM
31 lines
990 B
LLVM
; RUN: llc -mtriple=hexagon < %s | FileCheck %s
|
|
|
|
target datalayout = "e-m:e-p:32:32:32-a:0-n16:32-i64:64:64-i32:32:32-i16:16:16-i1:8:8-f32:32:32-f64:64:64-v32:32:32-v64:64:64-v512:512:512-v1024:1024:1024-v2048:2048:2048"
|
|
target triple = "hexagon"
|
|
|
|
; CHECK-LABEL: danny:
|
|
; CHECK: vunpack
|
|
; CHECK-NOT: vinsert
|
|
define void @danny(ptr %a0, ptr %a1) #0 {
|
|
b2:
|
|
%v16 = load <16 x i16>, ptr %a0, align 128
|
|
%v17 = sext <16 x i16> %v16 to <16 x i32>
|
|
store <16 x i32> %v17, ptr %a1, align 128
|
|
ret void
|
|
}
|
|
|
|
; CHECK-LABEL: sammy:
|
|
; CHECK: vunpack
|
|
; CHECK-NOT: vinsert
|
|
define void @sammy(ptr %a0, ptr %a1) #1 {
|
|
b2:
|
|
%v16 = load <32 x i16>, ptr %a0, align 128
|
|
%v17 = sext <32 x i16> %v16 to <32 x i32>
|
|
store <32 x i32> %v17, ptr %a1, align 128
|
|
ret void
|
|
}
|
|
|
|
|
|
attributes #0 = { noinline norecurse nounwind "target-cpu"="hexagonv60" "target-features"="+hvx-length64b,+hvxv60" }
|
|
attributes #1 = { noinline norecurse nounwind "target-cpu"="hexagonv60" "target-features"="+hvx-length128b,+hvxv60" }
|