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 -resource-pressure=false < %s | FileCheck %s
3
4# The vmul can start executing 3cy in advance. That is beause the first use
5# operand (i.e. %xmm1) is a ReadAfterLd. That means, the memory operand is
6# evaluated before %xmm1.
7
8vaddps  %xmm0, %xmm0, %xmm1
9vmulps  (%rdi), %xmm1, %xmm2
10
11# CHECK:      Iterations:        1
12# CHECK-NEXT: Instructions:      2
13# CHECK-NEXT: Total Cycles:      10
14# CHECK-NEXT: Total uOps:        2
15
16# CHECK:      Dispatch Width:    2
17# CHECK-NEXT: uOps Per Cycle:    0.20
18# CHECK-NEXT: IPC:               0.20
19# CHECK-NEXT: Block RThroughput: 1.0
20
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)
28
29# CHECK:      [1]    [2]    [3]    [4]    [5]    [6]    Instructions:
30# CHECK-NEXT:  1      3     1.00                        vaddps	%xmm0, %xmm0, %xmm1
31# CHECK-NEXT:  1      7     1.00    *                   vmulps	(%rdi), %xmm1, %xmm2
32
33# CHECK:      Timeline view:
34# CHECK-NEXT: Index     0123456789
35
36# CHECK:      [0,0]     DeeeER   .   vaddps	%xmm0, %xmm0, %xmm1
37# CHECK-NEXT: [0,1]     DeeeeeeeER   vmulps	(%rdi), %xmm1, %xmm2
38
39# CHECK:      Average Wait times (based on the timeline view):
40# CHECK-NEXT: [0]: Executions
41# CHECK-NEXT: [1]: Average time spent waiting in a scheduler's queue
42# CHECK-NEXT: [2]: Average time spent waiting in a scheduler's queue while ready
43# CHECK-NEXT: [3]: Average time elapsed from WB until retire stage
44
45# CHECK:            [0]    [1]    [2]    [3]
46# CHECK-NEXT: 0.     1     1.0    1.0    0.0       vaddps	%xmm0, %xmm0, %xmm1
47# CHECK-NEXT: 1.     1     1.0    0.0    0.0       vmulps	(%rdi), %xmm1, %xmm2
48# CHECK-NEXT:        1     1.0    0.5    0.0       <total>
49