Run DCE after a LoopFlatten test to reduce spurious output [nfc]
[llvm-project.git] / llvm / test / tools / llvm-mca / X86 / BtVer2 / negative-read-advance.s
blobde409e1c28d04d7bb59ba0215ad13bfb59871059
1 # NOTE: Assertions have been autogenerated by utils/update_mca_test_checks.py
2 # RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=btver2 -iterations=1 -timeline < %s | FileCheck %s
4 add %ebx, %ebx
5 vpinsrd $1, %ebx, %xmm0, %xmm1
6 vpinsrd $1, %ebx, %xmm2, %xmm3
7 vpinsrd $2, %ebx, %xmm4, %xmm5
8 vpinsrd $2, %ebx, %xmm6, %xmm7
9 vpinsrd $3, %ebx, %xmm8, %xmm10
11 # CHECK: Iterations: 1
12 # CHECK-NEXT: Instructions: 6
13 # CHECK-NEXT: Total Cycles: 13
14 # CHECK-NEXT: Total uOps: 11
16 # CHECK: Dispatch Width: 2
17 # CHECK-NEXT: uOps Per Cycle: 0.85
18 # CHECK-NEXT: IPC: 0.46
19 # CHECK-NEXT: Block RThroughput: 5.5
21 # CHECK: Instruction Info:
22 # CHECK-NEXT: [1]: #uOps
23 # CHECK-NEXT: [2]: Latency
24 # CHECK-NEXT: [3]: RThroughput
25 # CHECK-NEXT: [4]: MayLoad
26 # CHECK-NEXT: [5]: MayStore
27 # CHECK-NEXT: [6]: HasSideEffects (U)
29 # CHECK: [1] [2] [3] [4] [5] [6] Instructions:
30 # CHECK-NEXT: 1 1 0.50 addl %ebx, %ebx
31 # CHECK-NEXT: 2 7 0.50 vpinsrd $1, %ebx, %xmm0, %xmm1
32 # CHECK-NEXT: 2 7 0.50 vpinsrd $1, %ebx, %xmm2, %xmm3
33 # CHECK-NEXT: 2 7 0.50 vpinsrd $2, %ebx, %xmm4, %xmm5
34 # CHECK-NEXT: 2 7 0.50 vpinsrd $2, %ebx, %xmm6, %xmm7
35 # CHECK-NEXT: 2 7 0.50 vpinsrd $3, %ebx, %xmm8, %xmm10
37 # CHECK: Resources:
38 # CHECK-NEXT: [0] - JALU0
39 # CHECK-NEXT: [1] - JALU1
40 # CHECK-NEXT: [2] - JDiv
41 # CHECK-NEXT: [3] - JFPA
42 # CHECK-NEXT: [4] - JFPM
43 # CHECK-NEXT: [5] - JFPU0
44 # CHECK-NEXT: [6] - JFPU1
45 # CHECK-NEXT: [7] - JLAGU
46 # CHECK-NEXT: [8] - JMul
47 # CHECK-NEXT: [9] - JSAGU
48 # CHECK-NEXT: [10] - JSTC
49 # CHECK-NEXT: [11] - JVALU0
50 # CHECK-NEXT: [12] - JVALU1
51 # CHECK-NEXT: [13] - JVIMUL
53 # CHECK: Resource pressure per iteration:
54 # CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13]
55 # CHECK-NEXT: - 1.00 - - - 2.00 3.00 - - - - 2.00 3.00 -
57 # CHECK: Resource pressure by instruction:
58 # CHECK-NEXT: [0] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] Instructions:
59 # CHECK-NEXT: - 1.00 - - - - - - - - - - - - addl %ebx, %ebx
60 # CHECK-NEXT: - - - - - - 1.00 - - - - - 1.00 - vpinsrd $1, %ebx, %xmm0, %xmm1
61 # CHECK-NEXT: - - - - - 1.00 - - - - - 1.00 - - vpinsrd $1, %ebx, %xmm2, %xmm3
62 # CHECK-NEXT: - - - - - - 1.00 - - - - - 1.00 - vpinsrd $2, %ebx, %xmm4, %xmm5
63 # CHECK-NEXT: - - - - - 1.00 - - - - - 1.00 - - vpinsrd $2, %ebx, %xmm6, %xmm7
64 # CHECK-NEXT: - - - - - - 1.00 - - - - - 1.00 - vpinsrd $3, %ebx, %xmm8, %xmm10
66 # CHECK: Timeline view:
67 # CHECK-NEXT: 012
68 # CHECK-NEXT: Index 0123456789
70 # CHECK: [0,0] DeER . . . addl %ebx, %ebx
71 # CHECK-NEXT: [0,1] .D======eER . vpinsrd $1, %ebx, %xmm0, %xmm1
72 # CHECK-NEXT: [0,2] . D=====eER . vpinsrd $1, %ebx, %xmm2, %xmm3
73 # CHECK-NEXT: [0,3] . D=====eER. vpinsrd $2, %ebx, %xmm4, %xmm5
74 # CHECK-NEXT: [0,4] . D====eER. vpinsrd $2, %ebx, %xmm6, %xmm7
75 # CHECK-NEXT: [0,5] . D====eER vpinsrd $3, %ebx, %xmm8, %xmm10
77 # CHECK: Average Wait times (based on the timeline view):
78 # CHECK-NEXT: [0]: Executions
79 # CHECK-NEXT: [1]: Average time spent waiting in a scheduler's queue
80 # CHECK-NEXT: [2]: Average time spent waiting in a scheduler's queue while ready
81 # CHECK-NEXT: [3]: Average time elapsed from WB until retire stage
83 # CHECK: [0] [1] [2] [3]
84 # CHECK-NEXT: 0. 1 1.0 1.0 0.0 addl %ebx, %ebx
85 # CHECK-NEXT: 1. 1 7.0 0.0 0.0 vpinsrd $1, %ebx, %xmm0, %xmm1
86 # CHECK-NEXT: 2. 1 6.0 0.0 0.0 vpinsrd $1, %ebx, %xmm2, %xmm3
87 # CHECK-NEXT: 3. 1 6.0 1.0 0.0 vpinsrd $2, %ebx, %xmm4, %xmm5
88 # CHECK-NEXT: 4. 1 5.0 1.0 0.0 vpinsrd $2, %ebx, %xmm6, %xmm7
89 # CHECK-NEXT: 5. 1 5.0 2.0 0.0 vpinsrd $3, %ebx, %xmm8, %xmm10
90 # CHECK-NEXT: 1 5.0 0.8 0.0 <total>