File: swp-vsum.ll

package info (click to toggle)
llvm-toolchain-14 1%3A14.0.6-12
  • links: PTS, VCS
  • area: main
  • in suites: bookworm
  • size: 1,496,180 kB
  • sloc: cpp: 5,593,972; ansic: 986,872; asm: 585,869; python: 184,223; objc: 72,530; lisp: 31,119; f90: 27,793; javascript: 9,780; pascal: 9,762; sh: 9,482; perl: 7,468; ml: 5,432; awk: 3,523; makefile: 2,538; xml: 953; cs: 573; fortran: 567
file content (32 lines) | stat: -rw-r--r-- 1,079 bytes parent folder | download | duplicates (9) 
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
29
30
31
32
 
; RUN: llc -march=hexagon -mcpu=hexagonv5 -enable-pipeliner < %s -pipeliner-experimental-cg=true | FileCheck %s
; RUN: llc -march=hexagon -mcpu=hexagonv60 -enable-pipeliner < %s -pipeliner-experimental-cg=true | FileCheck %s --check-prefix=CHECKV60

; Simple vector total.
; CHECK: loop0(.LBB0_[[LOOP:.]],
; CHECK: .LBB0_[[LOOP]]:
; CHECK: add(r{{[0-9]+}},r{{[0-9]+}})
; CHECK-NEXT: memw(r{{[0-9]+}}++#4)
; CHECK-NEXT: endloop0

; V60 does not pipeline due to latencies.
; CHECKV60: memw(r{{[0-9]+}}++#4)
; CHECKV60: add(r{{[0-9]+}},r{{[0-9]+}})

define i32 @f0(i32* %a0, i32 %a1) {
b0:
  br label %b1

b1:                                               ; preds = %b1, %b0
  %v0 = phi i32 [ 0, %b0 ], [ %v4, %b1 ]
  %v1 = phi i32* [ %a0, %b0 ], [ %v7, %b1 ]
  %v2 = phi i32 [ 0, %b0 ], [ %v5, %b1 ]
  %v3 = load i32, i32* %v1, align 4
  %v4 = add nsw i32 %v3, %v0
  %v5 = add nsw i32 %v2, 1
  %v6 = icmp eq i32 %v5, 10000
  %v7 = getelementptr i32, i32* %v1, i32 1
  br i1 %v6, label %b2, label %b1

b2:                                               ; preds = %b1
  ret i32 %v4
}