
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.
17 lines
1.0 KiB
LLVM
17 lines
1.0 KiB
LLVM
; RUN: llc -mtriple=hexagon -no-integrated-as < %s | FileCheck %s
|
|
|
|
target triple = "hexagon"
|
|
|
|
;CHECK: [[REGH:r[0-9]]]:[[REGL:[0-9]]] = memd_locked
|
|
;CHECK: HIGH([[REGH]])
|
|
;CHECK: LOW(r[[REGL]])
|
|
define i32 @fred(ptr %free_list_ptr, ptr %item_ptr, ptr %free_item_ptr) nounwind {
|
|
entry:
|
|
%free_list_ptr.addr = alloca ptr, align 4
|
|
store ptr %free_list_ptr, ptr %free_list_ptr.addr, align 4
|
|
%0 = load ptr, ptr %item_ptr, align 4
|
|
%1 = call { i64, i32 } asm sideeffect "1: $0 = memd_locked($5)\0A\09 $1 = HIGH(${0:H}) \0A\09 $1 = add($1,#1) \0A\09 memw($6) = LOW(${0:L}) \0A\09 $0 = combine($7,$1) \0A\09 memd_locked($5,p0) = $0 \0A\09 if !p0 jump 1b\0A\09", "=&r,=&r,=*m,=*m,r,r,r,r,*m,*m,~{p0}"(ptr elementtype(ptr) %free_list_ptr.addr, ptr elementtype(ptr) %free_item_ptr, i64 0, ptr %free_list_ptr, ptr %free_item_ptr, ptr %0, ptr elementtype(ptr) %free_list_ptr.addr, ptr elementtype(ptr) %free_item_ptr) nounwind
|
|
%asmresult1 = extractvalue { i64, i32 } %1, 1
|
|
ret i32 %asmresult1
|
|
}
|