File: cortex-a57-misched-ldm.ll

package info (click to toggle)
llvm-toolchain-16 1%3A16.0.6-15~deb12u1
  • links: PTS, VCS
  • area: main
  • in suites: bookworm
  • size: 1,634,792 kB
  • sloc: cpp: 6,179,261; ansic: 1,216,205; asm: 741,319; python: 196,614; objc: 75,325; f90: 49,640; lisp: 32,396; pascal: 12,286; sh: 9,394; perl: 7,442; ml: 5,494; awk: 3,523; makefile: 2,723; javascript: 1,206; xml: 886; fortran: 581; cs: 573
file content (28 lines) | stat: -rw-r--r-- 873 bytes parent folder | download | duplicates (12) 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
 
; REQUIRES: asserts
; RUN: llc < %s -mtriple=armv8r-eabi -mcpu=cortex-a57 -mattr=use-misched -verify-misched -debug-only=machine-scheduler -o - 2>&1 > /dev/null | FileCheck %s

; CHECK:       ********** MI Scheduling **********
; We need second, post-ra scheduling to have LDM instruction combined from single-loads
; CHECK:       ********** MI Scheduling **********
; CHECK:       LDMIA
; CHECK:       rdefs left
; CHECK-NEXT:  Latency            : 3
; CHECK:       Successors:
; CHECK:       Data
; CHECK-SAME:  Latency=3
; CHECK-NEXT:  Data
; CHECK-SAME:  Latency=0

define i32 @foo(ptr %a) nounwind optsize {
entry:
  %b = getelementptr i32, ptr %a, i32 1
  %c = getelementptr i32, ptr %a, i32 2 
  %0 = load i32, ptr %a, align 4
  %1 = load i32, ptr %b, align 4
  %2 = load i32, ptr %c, align 4

  %mul1 = mul i32 %0, %1
  %mul2 = mul i32 %mul1, %2
  ret i32 %mul2
}