AMD Znver2 (Rome) Scheduler enablement

[GGanesh]

Summary

The patch gives out the details of the znver2 scheduler model.
Rome is second generation zen family processor.
There are few improvements with respect to execution units, latencies and throughput when compared with znver1.
The tests that were present for znver1 for llvm-mca tool were replicated. The latencies, execution units, timeline and throughput information are updated for znver2.
I am working on another patch focusing on the information as mandated by exegesis.

[GGanesh]

Ping!

[hfinkel]

llvm/lib/Target/X86/X86ScheduleZnver2.td

107 ↗

(On Diff #214628)

Indent.

350 ↗

(On Diff #214628)

Commented out?

605 ↗

(On Diff #214628)

Can this "definitely wrong" thing be fixed?

628 ↗

(On Diff #214628)

Here, and below, some of these are commented out. Why? Can/should these just be removed?

1194 ↗

(On Diff #214628)

Commented out?

[GGanesh]

The changes for review comments are incorporated.
The latency information in CLZERO, VZEROUPPER, MUL instructions are updated.

[GGanesh]

Updated for review comments and latency modifications for MUL, vzeroupper, CLZERO instructions.

[jdoerfert]

llvm/lib/Target/X86/X86ScheduleZnver2.td

49 ↗

(On Diff #225324)

Nit: Three?

[lebedev.ri]

llvm/lib/Target/X86/X86ScheduleZnver2.td

90–91 ↗

(On Diff #225324)

But let LoadLatency = 4;.
Shouldn't this also be 4?

155 ↗

(On Diff #225324)

But ReadAdvance<ReadAfterVecLd is at 7?

[GGanesh]

Updated the patch for review comments.

[GGanesh]

Ping!

[GGanesh]

Ping!

[lebedev.ri]

Not sure how this (any) sched model should be reviewed.
Perhaps it would help to post the exegesis reports.

FWIW i'd just merge this.

llvm/lib/Target/X86/X86ScheduleZnver2.td

179 ↗

(On Diff #226091)

This is correct?
Shouldn't it be 4+1 ?

[craig.topper]

In D66088#1740344, @lebedev.ri wrote:

Not sure how this (any) sched model should be reviewed.
Perhaps it would help to post the exegesis reports.

FWIW i'd just merge this.

I agree with Roman that there's not a good way to review this other than comparing with exegesis data. Sounds like from the initial description there is planned follow up with more data from exegesis. So I think this is fine to start with.

[GGanesh]

I agree on having a patch to enable exegesis. Will post that in couple of days. As mentioned, this is part of the initial plan as well.

[hfinkel]

In D66088#1750813, @GGanesh wrote:

I agree on having a patch to enable exegesis. Will post that in couple of days. As mentioned, this is part of the initial plan as well.

It looks like everyone is in agreement: We should move forward with this, exegesis next as follow-up, and then updates from there as appropriate. Thus, LGTM.

[xbolva00]

Do you have the commit rights?

llvm/lib/Target/X86/X86.td

 //===----------------------------------------------------------------------===//
 
 include "X86ScheduleAtom.td"
 include "X86SchedSandyBridge.td"
 include "X86SchedHaswell.td"
 include "X86SchedBroadwell.td"
 include "X86ScheduleSLM.td"
 include "X86ScheduleZnver1.td"
+include "X86ScheduleZnver2.td"
 include "X86ScheduleBdVer2.td"
 include "X86ScheduleBtVer2.td"
 include "X86SchedSkylakeClient.td"
 include "X86SchedSkylakeServer.td"
 
 //===----------------------------------------------------------------------===//
 // X86 Processor Feature Lists
 //===----------------------------------------------------------------------===//

 // Piledriver
 def : ProcessorModel<"bdver2", BdVer2Model, ProcessorFeatures.BdVer2Features>;
 // Steamroller
 def : Proc<"bdver3", ProcessorFeatures.BdVer3Features>;
 // Excavator
 def : Proc<"bdver4", ProcessorFeatures.BdVer4Features>;
 
 def : ProcessorModel<"znver1", Znver1Model, ProcessorFeatures.ZNFeatures>;
-def : ProcessorModel<"znver2", Znver1Model, ProcessorFeatures.ZN2Features>;
+def : ProcessorModel<"znver2", Znver2Model, ProcessorFeatures.ZN2Features>;
 
 def : Proc<"geode",           [FeatureX87, FeatureSlowUAMem16, FeatureCMPXCHG8B,
                                Feature3DNowA]>;
 
 def : Proc<"winchip-c6",      [FeatureX87, FeatureSlowUAMem16, FeatureMMX]>;
 def : Proc<"winchip2",        [FeatureX87, FeatureSlowUAMem16, Feature3DNow]>;
 def : Proc<"c3",              [FeatureX87, FeatureSlowUAMem16, Feature3DNow]>;
 def : Proc<"c3-2",            [FeatureX87, FeatureSlowUAMem16, FeatureCMPXCHG8B,

llvm/lib/Target/X86/X86InstrInfo.td

def ENQCMDS64 : I<0xF8, MRMSrcMem, (outs), (ins GR64:$dst, i512mem:$src),
                  "enqcmds\t{$src, $dst|$dst, $src}",
                  [(set EFLAGS, (X86enqcmds GR64:$dst, addr:$src))]>,
                  T8XS, AdSize64, Requires<[HasENQCMD, In64BitMode]>;
 }
 
 //===----------------------------------------------------------------------===//
 // CLZERO Instruction
 //
-let SchedRW = [WriteSystem] in {
+let SchedRW = [WriteLoad] in {
   let Uses = [EAX] in
   def CLZERO32r : I<0x01, MRM_FC, (outs), (ins), "clzero", []>,
                   TB, Requires<[HasCLZERO, Not64BitMode]>;
   let Uses = [RAX] in
   def CLZERO64r : I<0x01, MRM_FC, (outs), (ins), "clzero", []>,
                   TB, Requires<[HasCLZERO, In64BitMode]>;
 } // SchedRW
 

llvm/test/MC/X86/x86_long_nop.s

 # RUN: llvm-mc -filetype=obj -arch=x86 -triple=x86_64-pc-linux-gnu -mcpu=pentiumpro %s | llvm-objdump -d -no-show-raw-insn - | FileCheck %s --check-prefix=LNOP10
 # RUN: llvm-mc -filetype=obj -arch=x86 -triple=i686-pc-linux-gnu %s -mcpu=pentiumpro | llvm-objdump -d -no-show-raw-insn - | FileCheck %s --check-prefix=LNOP10
 # RUN: llvm-mc -filetype=obj -arch=x86 -triple=x86_64-apple-darwin10.0 -mcpu=pentiumpro %s | llvm-objdump -d -no-show-raw-insn - | FileCheck %s --check-prefix=LNOP10
 # RUN: llvm-mc -filetype=obj -arch=x86 -triple=i686-apple-darwin8 -mcpu=pentiumpro %s | llvm-objdump -d -no-show-raw-insn - | FileCheck %s --check-prefix=LNOP10
 # RUN: llvm-mc -filetype=obj -arch=x86 -triple=i686-pc-linux-gnu -mcpu=slm %s | llvm-objdump -d -no-show-raw-insn - | FileCheck --check-prefix=LNOP7 %s
 # RUN: llvm-mc -filetype=obj -arch=x86 -triple=i686-pc-linux-gnu -mcpu=silvermont %s | llvm-objdump -d -no-show-raw-insn - | FileCheck --check-prefix=LNOP7 %s
 # RUN: llvm-mc -filetype=obj -arch=x86 -triple=i686-pc-linux-gnu -mcpu=lakemont %s | llvm-objdump -d -no-show-raw-insn - | FileCheck --check-prefix=NOP1 %s
 # RUN: llvm-mc -filetype=obj -arch=x86 -triple=x86_64-pc-linux-gnu -mcpu=bdver1 %s | llvm-objdump -d -no-show-raw-insn - | FileCheck %s --check-prefix=LNOP11
 # RUN: llvm-mc -filetype=obj -arch=x86 -triple=i686-pc-linux-gnu %s -mcpu=bdver1 | llvm-objdump -d -no-show-raw-insn - | FileCheck %s --check-prefix=LNOP11
 # RUN: llvm-mc -filetype=obj -arch=x86 -triple=x86_64-pc-linux-gnu -mcpu=btver1 %s | llvm-objdump -d -no-show-raw-insn - | FileCheck %s --check-prefix=LNOP15
 # RUN: llvm-mc -filetype=obj -arch=x86 -triple=i686-pc-linux-gnu %s -mcpu=btver1 | llvm-objdump -d -no-show-raw-insn - | FileCheck %s --check-prefix=LNOP15
 # RUN: llvm-mc -filetype=obj -arch=x86 -triple=x86_64-pc-linux-gnu -mcpu=btver2 %s | llvm-objdump -d -no-show-raw-insn - | FileCheck %s --check-prefix=LNOP15
 # RUN: llvm-mc -filetype=obj -arch=x86 -triple=i686-pc-linux-gnu %s -mcpu=btver2 | llvm-objdump -d -no-show-raw-insn - | FileCheck %s --check-prefix=LNOP15
 # RUN: llvm-mc -filetype=obj -arch=x86 -triple=x86_64-pc-linux-gnu -mcpu=znver1 %s | llvm-objdump -d -no-show-raw-insn - | FileCheck %s --check-prefix=LNOP15
 # RUN: llvm-mc -filetype=obj -arch=x86 -triple=i686-pc-linux-gnu %s -mcpu=znver1 | llvm-objdump -d -no-show-raw-insn - | FileCheck %s --check-prefix=LNOP15
+# RUN: llvm-mc -filetype=obj -arch=x86 -triple=x86_64-pc-linux-gnu -mcpu=znver2 %s | llvm-objdump -d -no-show-raw-insn - | FileCheck %s --check-prefix=LNOP15
+# RUN: llvm-mc -filetype=obj -arch=x86 -triple=i686-pc-linux-gnu %s -mcpu=znver2 | llvm-objdump -d -no-show-raw-insn - | FileCheck %s --check-prefix=LNOP15
 
 # Ensure alignment directives also emit sequences of 10, 11 and 15-byte NOPs on processors
 # capable of using long NOPs.
 inc %eax
 .p2align 5
 inc %eax
 # LNOP15: 0:  inc
 # LNOP15-NEXT: 1:  nop

llvm/test/tools/llvm-mca/X86/Generic/resources-clzero.s

 # NOTE: Assertions have been autogenerated by utils/update_mca_test_checks.py
 # RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=x86-64 -instruction-tables < %s | FileCheck %s
 
 clzero
 
 # CHECK:      Instruction Info:
 # CHECK-NEXT: [1]: #uOps
 # CHECK-NEXT: [2]: Latency
 # CHECK-NEXT: [3]: RThroughput
 # CHECK-NEXT: [4]: MayLoad
 # CHECK-NEXT: [5]: MayStore
 # CHECK-NEXT: [6]: HasSideEffects (U)
 
 # CHECK:      [1]    [2]    [3]    [4]    [5]    [6]    Instructions:
-# CHECK-NEXT:  1      100   0.33                  U     clzero
+# CHECK-NEXT:  1      5   0.50                  U     clzero
 
 # CHECK:      Resources:
 # CHECK-NEXT: [0]   - SBDivider
 # CHECK-NEXT: [1]   - SBFPDivider
 # CHECK-NEXT: [2]   - SBPort0
 # CHECK-NEXT: [3]   - SBPort1
 # CHECK-NEXT: [4]   - SBPort4
 # CHECK-NEXT: [5]   - SBPort5
 # CHECK-NEXT: [6.0] - SBPort23
 # CHECK-NEXT: [6.1] - SBPort23
 
 # CHECK:      Resource pressure per iteration:
 # CHECK-NEXT: [0]    [1]    [2]    [3]    [4]    [5]    [6.0]  [6.1]
-# CHECK-NEXT:  -      -     0.33   0.33    -     0.33    -      -
+# CHECK-NEXT:  -      -      -      -      -      -     0.50   0.50
 
 # CHECK:      Resource pressure by instruction:
 # CHECK-NEXT: [0]    [1]    [2]    [3]    [4]    [5]    [6.0]  [6.1]  Instructions:
-# CHECK-NEXT:  -      -     0.33   0.33    -     0.33    -      -     clzero
+# CHECK-NEXT: -      -      -      -      -      -     0.50   0.50   clzero

llvm/test/tools/llvm-mca/X86/Znver1/resources-clzero.s

 # NOTE: Assertions have been autogenerated by utils/update_mca_test_checks.py
 # RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=znver1 -instruction-tables < %s | FileCheck %s
 
 clzero
 
 # CHECK:      Instruction Info:
 # CHECK-NEXT: [1]: #uOps
 # CHECK-NEXT: [2]: Latency
 # CHECK-NEXT: [3]: RThroughput
 # CHECK-NEXT: [4]: MayLoad
 # CHECK-NEXT: [5]: MayStore
 # CHECK-NEXT: [6]: HasSideEffects (U)
 
 # CHECK:      [1]    [2]    [3]    [4]    [5]    [6]    Instructions:
-# CHECK-NEXT:  1      100   0.25                  U     clzero
+# CHECK-NEXT:  1      8    0.50                   U       clzero
 
 # CHECK:      Resources:
 # CHECK-NEXT: [0]   - ZnAGU0
 # CHECK-NEXT: [1]   - ZnAGU1
 # CHECK-NEXT: [2]   - ZnALU0
 # CHECK-NEXT: [3]   - ZnALU1
 # CHECK-NEXT: [4]   - ZnALU2
 # CHECK-NEXT: [5]   - ZnALU3
 # CHECK-NEXT: [6]   - ZnDivider
 # CHECK-NEXT: [7]   - ZnFPU0
 # CHECK-NEXT: [8]   - ZnFPU1
 # CHECK-NEXT: [9]   - ZnFPU2
 # CHECK-NEXT: [10]  - ZnFPU3
 # CHECK-NEXT: [11]  - ZnMultiplier
 
 # CHECK:      Resource pressure per iteration:
 # CHECK-NEXT: [0]    [1]    [2]    [3]    [4]    [5]    [6]    [7]    [8]    [9]    [10]   [11]
-# CHECK-NEXT:  -      -      -      -      -      -      -      -      -      -      -      -
+# CHECK-NEXT:  0.50   0.50    -      -      -      -      -      -      -      -      -      -
 
 # CHECK:      Resource pressure by instruction:
 # CHECK-NEXT: [0]    [1]    [2]    [3]    [4]    [5]    [6]    [7]    [8]    [9]    [10]   [11]   Instructions:
-# CHECK-NEXT:  -      -      -      -      -      -      -      -      -      -      -      -     clzero
+# CHECK-NEXT: 0.50   0.50    -      -      -      -      -      -      -      -      -      -     clzero

llvm/test/tools/llvm-mca/X86/bextr-read-after-ld.s

 # NOTE: Assertions have been autogenerated by utils/update_mca_test_checks.py
 # RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=haswell -iterations=1 -timeline -resource-pressure=false < %s | FileCheck %s -check-prefix=ALL -check-prefix=HASWELL
 # RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=broadwell -iterations=1 -timeline -resource-pressure=false < %s | FileCheck %s -check-prefix=ALL -check-prefix=BDWELL
 # RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=skylake -iterations=1 -timeline -resource-pressure=false < %s | FileCheck %s -check-prefix=ALL -check-prefix=SKYLAKE
 # RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=bdver2 -iterations=1 -timeline -resource-pressure=false < %s | FileCheck %s -check-prefix=ALL -check-prefix=BDVER2
 # RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=btver2 -iterations=1 -timeline -resource-pressure=false < %s | FileCheck %s -check-prefix=ALL -check-prefix=BTVER2
 # RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=znver1 -iterations=1 -timeline -resource-pressure=false < %s | FileCheck %s -check-prefix=ALL -check-prefix=ZNVER1
+# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=znver2 -iterations=1 -timeline -resource-pressure=false < %s | FileCheck %s -check-prefix=ALL -check-prefix=ZNVER2
 
 add     %edi, %esi
 bextrl  %esi, (%rdi), %eax
 
 # ALL:          Iterations:        1
 # ALL-NEXT:     Instructions:      2
 
 # BDVER2-NEXT:  Total Cycles:      9

 # HASWELL-NEXT: Total uOps:        4
 
 # SKYLAKE-NEXT: Total Cycles:      10
 # SKYLAKE-NEXT: Total uOps:        4
 
 # ZNVER1-NEXT:  Total Cycles:      8
 # ZNVER1-NEXT:  Total uOps:        3
 
+# ZNVER2-NEXT:  Total Cycles:      8
+# ZNVER2-NEXT:  Total uOps:        3
+
 # BDVER2:       Dispatch Width:    4
 # BDVER2-NEXT:  uOps Per Cycle:    0.33
 # BDVER2-NEXT:  IPC:               0.22
 # BDVER2-NEXT:  Block RThroughput: 2.0
 
 # BDWELL:       Dispatch Width:    4
 # BDWELL-NEXT:  uOps Per Cycle:    0.40
 # BDWELL-NEXT:  IPC:               0.20

llvm/test/tools/llvm-mca/X86/bzhi-read-after-ld.s

 # NOTE: Assertions have been autogenerated by utils/update_mca_test_checks.py
 # RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=haswell -iterations=1 -timeline -resource-pressure=false < %s | FileCheck %s -check-prefix=ALL -check-prefix=HASWELL
 # RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=broadwell -iterations=1 -timeline -resource-pressure=false < %s | FileCheck %s -check-prefix=ALL -check-prefix=BDWELL
 # RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=skylake -iterations=1 -timeline -resource-pressure=false < %s | FileCheck %s -check-prefix=ALL -check-prefix=SKYLAKE
 # RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=znver1 -iterations=1 -timeline -resource-pressure=false < %s | FileCheck %s -check-prefix=ALL -check-prefix=ZNVER1
+# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=znver2 -iterations=1 -timeline -resource-pressure=false < %s | FileCheck %s -check-prefix=ALL -check-prefix=ZNVER2
 
 add     %edi, %esi
 bzhil %esi, (%rdi), %eax
 
 # ALL:          Iterations:        1
 # ALL-NEXT:     Instructions:      2
 
 # BDWELL-NEXT:  Total Cycles:      9
 # HASWELL-NEXT: Total Cycles:      9
 # SKYLAKE-NEXT: Total Cycles:      9
 # ZNVER1-NEXT:  Total Cycles:      8
+# ZNVER2-NEXT:  Total Cycles:      8
 
 # ALL-NEXT:     Total uOps:        3
 
 # BDWELL:       Dispatch Width:    4
 # BDWELL-NEXT:  uOps Per Cycle:    0.33
 # BDWELL-NEXT:  IPC:               0.22
 # BDWELL-NEXT:  Block RThroughput: 0.8
 

llvm/test/tools/llvm-mca/X86/cpus.s

 # NOTE: Assertions have been autogenerated by utils/update_mca_test_checks.py
 # RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=x86-64 -resource-pressure=false -instruction-info=false < %s | FileCheck --check-prefix=ALL --check-prefix=BARCELONA %s
 # RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=bdver2 -resource-pressure=false -instruction-info=false < %s | FileCheck --check-prefix=ALL --check-prefix=BDVER2 %s
 # RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=btver2 -resource-pressure=false -instruction-info=false < %s | FileCheck --check-prefix=ALL --check-prefix=BTVER2 %s
 # RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=znver1 -resource-pressure=false -instruction-info=false < %s | FileCheck --check-prefix=ALL --check-prefix=ZNVER1 %s
+# RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=znver2 -resource-pressure=false -instruction-info=false < %s | FileCheck --check-prefix=ALL --check-prefix=ZNVER2 %s
 # RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=sandybridge -resource-pressure=false -instruction-info=false < %s | FileCheck --check-prefix=ALL --check-prefix=SANDYBRIDGE %s
 # RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=ivybridge -resource-pressure=false -instruction-info=false < %s | FileCheck --check-prefix=ALL --check-prefix=IVYBRIDGE %s
 # RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=haswell -resource-pressure=false -instruction-info=false < %s | FileCheck --check-prefix=ALL --check-prefix=HASWELL %s
 # RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=broadwell -resource-pressure=false -instruction-info=false < %s | FileCheck --check-prefix=ALL --check-prefix=BROADWELL %s
 # RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=knl -resource-pressure=false -instruction-info=false < %s | FileCheck --check-prefix=ALL --check-prefix=KNL %s
 # RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=skylake -resource-pressure=false -instruction-info=false < %s | FileCheck --check-prefix=ALL --check-prefix=SKX %s
 # RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=skylake-avx512 -resource-pressure=false -instruction-info=false < %s | FileCheck --check-prefix=ALL --check-prefix=SKX-AVX512 %s
 # RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=slm -resource-pressure=false -instruction-info=false < %s | FileCheck --check-prefix=ALL --check-prefix=SLM %s

 # SLM-NEXT:         uOps Per Cycle:    0.97
 # SLM-NEXT:         IPC:               0.97
 # SLM-NEXT:         Block RThroughput: 0.5
 
 # ZNVER1:           Dispatch Width:    4
 # ZNVER1-NEXT:      uOps Per Cycle:    0.97
 # ZNVER1-NEXT:      IPC:               0.97
 # ZNVER1-NEXT:      Block RThroughput: 0.3
+
+# ZNVER2:           Dispatch Width:    4
+# ZNVER2-NEXT:      uOps Per Cycle:    0.97
+# ZNVER2-NEXT:      IPC:               0.97
+# ZNVER2-NEXT:      Block RThroughput: 0.3

llvm/test/tools/llvm-mca/X86/fma3-read-after-ld-1.s

 # NOTE: Assertions have been autogenerated by utils/update_mca_test_checks.py
 # RUN: llvm-mca -mtriple=x86_64-unknown-unknown  -mcpu=haswell -iterations=1 -resource-pressure=false -instruction-info=false -timeline < %s | FileCheck %s -check-prefix=ALL -check-prefix=HASWELL
 
 # RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=broadwell -iterations=1 -resource-pressure=false -instruction-info=false -timeline < %s | FileCheck %s -check-prefix=ALL -check-prefix=BDWELL
 
 # RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=skylake -iterations=1 -resource-pressure=false -instruction-info=false -timeline < %s | FileCheck %s -check-prefix=ALL -check-prefix=SKYLAKE
 
 # RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=znver1 -iterations=1 -resource-pressure=false -instruction-info=false -timeline < %s | FileCheck %s -check-prefix=ALL -check-prefix=ZNVER1
 
+# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=znver2 -iterations=1 -resource-pressure=false -instruction-info=false -timeline < %s | FileCheck %s -check-prefix=ALL -check-prefix=ZNVER1
+
 vaddps %xmm0, %xmm0, %xmm1
 vfmadd213ps (%rdi), %xmm1, %xmm2
 
 # ALL:          Iterations:        1
 # ALL-NEXT:     Instructions:      2
 
 # BDWELL-NEXT:  Total Cycles:      13
 # BDWELL-NEXT:  Total uOps:        3

llvm/test/tools/llvm-mca/X86/fma3-read-after-ld-2.s

 # NOTE: Assertions have been autogenerated by utils/update_mca_test_checks.py
 # RUN: llvm-mca -mtriple=x86_64-unknown-unknown  -mcpu=haswell -iterations=1 -resource-pressure=false -instruction-info=false -timeline < %s | FileCheck %s -check-prefix=ALL -check-prefix=HASWELL
 
 # RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=broadwell -iterations=1 -resource-pressure=false -instruction-info=false -timeline < %s | FileCheck %s -check-prefix=ALL -check-prefix=BDWELL
 
 # RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=skylake -iterations=1 -resource-pressure=false -instruction-info=false -timeline < %s | FileCheck %s -check-prefix=ALL -check-prefix=SKYLAKE
 
 # RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=znver1 -iterations=1 -resource-pressure=false -instruction-info=false -timeline < %s | FileCheck %s -check-prefix=ALL -check-prefix=ZNVER1
 
+# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=znver2 -iterations=1 -resource-pressure=false -instruction-info=false -timeline < %s | FileCheck %s -check-prefix=ALL -check-prefix=ZNVER1
+
 vaddps %xmm0, %xmm0, %xmm2
 vfmadd213ps (%rdi), %xmm1, %xmm2
 
 # ALL:          Iterations:        1
 # ALL-NEXT:     Instructions:      2
 
 # BDWELL-NEXT:  Total Cycles:      13
 # BDWELL-NEXT:  Total uOps:        3

llvm/test/tools/llvm-mca/X86/read-after-ld-1.s

 # NOTE: Assertions have been autogenerated by utils/update_mca_test_checks.py
 # RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=sandybridge -iterations=1 -resource-pressure=false -instruction-info=false -timeline < %s | FileCheck %s -check-prefix=ALL -check-prefix=SANDY
 # RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=haswell -iterations=1 -resource-pressure=false -instruction-info=false -timeline < %s | FileCheck %s -check-prefix=ALL -check-prefix=HASWELL
 # RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=broadwell -iterations=1 -resource-pressure=false -instruction-info=false -timeline < %s | FileCheck %s -check-prefix=ALL -check-prefix=BDWELL
 # RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=skylake -iterations=1 -resource-pressure=false -instruction-info=false -timeline < %s | FileCheck %s -check-prefix=ALL -check-prefix=SKYLAKE
 # RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=x86-64 -iterations=1 -resource-pressure=false -instruction-info=false -timeline < %s | FileCheck %s -check-prefix=ALL -check-prefix=BARCELONA
 # RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=bdver2 -iterations=1 -resource-pressure=false -instruction-info=false -timeline < %s | FileCheck %s -check-prefix=ALL -check-prefix=BDVER2
 # RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=btver2 -iterations=1 -resource-pressure=false -instruction-info=false -timeline < %s | FileCheck %s -check-prefix=ALL -check-prefix=BTVER2
 # RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=znver1 -iterations=1 -resource-pressure=false -instruction-info=false -timeline < %s | FileCheck %s -check-prefix=ALL -check-prefix=ZNVER1
+# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=znver2 -iterations=1 -resource-pressure=false -instruction-info=false -timeline < %s | FileCheck %s -check-prefix=ALL -check-prefix=ZNVER2
 
 vdivps  %xmm0, %xmm1, %xmm1
 vaddps  (%rax), %xmm1, %xmm1
 
 # ALL:            Iterations:        1
 # ALL-NEXT:       Instructions:      2
 
 # BARCELONA-NEXT: Total Cycles:      20

 # SANDY-NEXT:     Total uOps:        3
 
 # SKYLAKE-NEXT:   Total Cycles:      18
 # SKYLAKE-NEXT:   Total uOps:        3
 
 # ZNVER1-NEXT:    Total Cycles:      20
 # ZNVER1-NEXT:    Total uOps:        2
 
+# ZNVER2-NEXT:    Total Cycles:      20
+# ZNVER2-NEXT:    Total uOps:        2
+
 # BARCELONA:      Dispatch Width:    4
 # BARCELONA-NEXT: uOps Per Cycle:    0.15
 # BARCELONA-NEXT: IPC:               0.10
 # BARCELONA-NEXT: Block RThroughput: 14.0
 
 # BDVER2:         Dispatch Width:    4
 # BDVER2-NEXT:    uOps Per Cycle:    0.12
 # BDVER2-NEXT:    IPC:               0.12

 # SKYLAKE-NEXT:   IPC:               0.11
 # SKYLAKE-NEXT:   Block RThroughput: 3.0
 
 # ZNVER1:         Dispatch Width:    4
 # ZNVER1-NEXT:    uOps Per Cycle:    0.10
 # ZNVER1-NEXT:    IPC:               0.10
 # ZNVER1-NEXT:    Block RThroughput: 1.0
 
+# ZNVER2:         Dispatch Width:    4
+# ZNVER2-NEXT:    uOps Per Cycle:    0.10
+# ZNVER2-NEXT:    IPC:               0.10
+# ZNVER2-NEXT:    Block RThroughput: 1.0
+
 # ALL:            Timeline view:
 
 # BARCELONA-NEXT:                     0123456789
 # BARCELONA-NEXT: Index     0123456789
 
 # BDVER2-NEXT:                        0123456
 # BDVER2-NEXT:    Index     0123456789
 

 # SANDY-NEXT:     Index     0123456789
 
 # SKYLAKE-NEXT:                       01234567
 # SKYLAKE-NEXT:   Index     0123456789
 
 # ZNVER1-NEXT:                        0123456789
 # ZNVER1-NEXT:    Index     0123456789
 
+# ZNVER2-NEXT:                        0123456789
+# ZNVER2-NEXT:    Index     0123456789
+
 # BARCELONA:      [0,0]     DeeeeeeeeeeeeeeER  .   vdivps %xmm0, %xmm1, %xmm1
 # BARCELONA-NEXT: [0,1]     D========eeeeeeeeeER   vaddps (%rax), %xmm1, %xmm1
 
 # BDVER2:         [0,0]     DeeeeeeeeeER   ..   vdivps  %xmm0, %xmm1, %xmm1
 # BDVER2-NEXT:    [0,1]     D====eeeeeeeeeeER   vaddps  (%rax), %xmm1, %xmm1
 
 # BDWELL:         [0,0]     DeeeeeeeeeeeER ..   vdivps  %xmm0, %xmm1, %xmm1
 # BDWELL-NEXT:    [0,1]     D======eeeeeeeeER   vaddps  (%rax), %xmm1, %xmm1
 
 # BTVER2:         [0,0]     DeeeeeeeeeeeeeeeeeeeER  .   vdivps  %xmm0, %xmm1, %xmm1
 # BTVER2-NEXT:    [0,1]     D==============eeeeeeeeER   vaddps  (%rax), %xmm1, %xmm1
 
 # HASWELL:        [0,0]     DeeeeeeeeeeeeeER  .   vdivps  %xmm0, %xmm1, %xmm1
 # HASWELL-NEXT:   [0,1]     D=======eeeeeeeeeER   vaddps  (%rax), %xmm1, %xmm1
 
 # SANDY:          [0,0]     DeeeeeeeeeeeeeeER  .   vdivps %xmm0, %xmm1, %xmm1
 # SANDY-NEXT:     [0,1]     D========eeeeeeeeeER   vaddps (%rax), %xmm1, %xmm1
 
 # SKYLAKE:        [0,0]     DeeeeeeeeeeeER . .   vdivps %xmm0, %xmm1, %xmm1
 # SKYLAKE-NEXT:   [0,1]     D=====eeeeeeeeeeER   vaddps (%rax), %xmm1, %xmm1
 
 # ZNVER1:         [0,0]     DeeeeeeeeeeeeeeeER .   vdivps %xmm0, %xmm1, %xmm1
 # ZNVER1-NEXT:    [0,1]     D=======eeeeeeeeeeER   vaddps (%rax), %xmm1, %xmm1
 
+# ZNVER2:         [0,0]     DeeeeeeeeeeeeeeeER .   vdivps %xmm0, %xmm1, %xmm1
+# ZNVER2-NEXT:    [0,1]     D=======eeeeeeeeeeER   vaddps (%rax), %xmm1, %xmm1
+
 # ALL:            Average Wait times (based on the timeline view):
 # ALL-NEXT:       [0]: Executions
 # ALL-NEXT:       [1]: Average time spent waiting in a scheduler's queue
 # ALL-NEXT:       [2]: Average time spent waiting in a scheduler's queue while ready
 # ALL-NEXT:       [3]: Average time elapsed from WB until retire stage
 
 # ALL:                  [0]    [1]    [2]    [3]
 # ALL-NEXT:       0.     1     1.0    1.0    0.0       vdivps %xmm0, %xmm1, %xmm1

 # SANDY-NEXT:     1.     1     9.0    0.0    0.0       vaddps (%rax), %xmm1, %xmm1
 # SANDY-NEXT:            1     5.0    0.5    0.0       <total>
 
 # SKYLAKE-NEXT:   1.     1     6.0    0.0    0.0       vaddps (%rax), %xmm1, %xmm1
 # SKYLAKE-NEXT:          1     3.5    0.5    0.0       <total>
 
 # ZNVER1-NEXT:    1.     1     8.0    0.0    0.0       vaddps (%rax), %xmm1, %xmm1
 # ZNVER1-NEXT:           1     4.5    0.5    0.0       <total>
+
+# ZNVER2-NEXT:    1.     1     8.0    0.0    0.0       vaddps (%rax), %xmm1, %xmm1
+# ZNVER2-NEXT:           1     4.5    0.5    0.0       <total>

llvm/test/tools/llvm-mca/X86/read-after-ld-2.s

 # NOTE: Assertions have been autogenerated by utils/update_mca_test_checks.py
 # RUN: llvm-mca -mtriple=x86_64-unknown-unknown  -mcpu=haswell -iterations=100 -resource-pressure=false -instruction-info=false -timeline < %s | FileCheck %s -check-prefix=ALL -check-prefix=HASWELL
 
 # RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=broadwell -iterations=100 -resource-pressure=false -instruction-info=false -timeline < %s | FileCheck %s -check-prefix=ALL -check-prefix=BDWELL
 
 # RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=skylake -iterations=100 -resource-pressure=false -instruction-info=false -timeline < %s | FileCheck %s -check-prefix=ALL -check-prefix=SKYLAKE
 
 # RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=znver1 -iterations=100 -resource-pressure=false -instruction-info=false -timeline < %s | FileCheck %s -check-prefix=ALL -check-prefix=ZNVER1
 
+# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=znver2 -iterations=100 -resource-pressure=false -instruction-info=false -timeline < %s | FileCheck %s -check-prefix=ALL -check-prefix=ZNVER2
+
 # Code Snippet from "Ithemal: Accurate, Portable and Fast Basic Block Throughput Estimation using Deep Neural Networks"
 # Charith Mendis, Saman Amarasinghe, Michael Carbin
 add     $1, %edx
 vpaddd (%r8), %ymm0, %ymm0
 add     $32, %r8
 cmp     %edi, %edx
 
 # ALL:          Iterations:        100

 # SKYLAKE-NEXT: IPC:               3.64
 # SKYLAKE-NEXT: Block RThroughput: 0.8
 
 # ZNVER1:       Dispatch Width:    4
 # ZNVER1-NEXT:  uOps Per Cycle:    3.64
 # ZNVER1-NEXT:  IPC:               3.64
 # ZNVER1-NEXT:  Block RThroughput: 1.0
 
+# ZNVER2:       Dispatch Width:    4
+# ZNVER2-NEXT:  uOps Per Cycle:    3.64
+# ZNVER2-NEXT:  IPC:               3.64
+# ZNVER2-NEXT:  Block RThroughput: 1.0
+
 # ALL:          Timeline view:
 
 # BDWELL-NEXT:                      0123456789
 # BDWELL-NEXT:  Index     0123456789          01
 
 # HASWELL-NEXT:                     0123456789
 # HASWELL-NEXT: Index     0123456789          012
 
 # SKYLAKE-NEXT:                     0123456789
 # SKYLAKE-NEXT: Index     0123456789
 
 # ZNVER1-NEXT:                      0123456789
 # ZNVER1-NEXT:  Index     0123456789
 
+# ZNVER2-NEXT:                      0123456789
+# ZNVER2-NEXT:  Index     0123456789
+
 # BDWELL:       [0,0]     DeER .    .    .    ..   addl $1, %edx
 # BDWELL-NEXT:  [0,1]     DeeeeeeeER.    .    ..   vpaddd (%r8), %ymm0, %ymm0
 # BDWELL-NEXT:  [0,2]     DeE------R.    .    ..   addq $32, %r8
 # BDWELL-NEXT:  [0,3]     .DeE-----R.    .    ..   cmpl %edi, %edx
 # BDWELL-NEXT:  [1,0]     .DeE-----R.    .    ..   addl $1, %edx
 # BDWELL-NEXT:  [1,1]     .DeeeeeeeER    .    ..   vpaddd (%r8), %ymm0, %ymm0
 # BDWELL-NEXT:  [1,2]     . DeE-----R    .    ..   addq $32, %r8
 # BDWELL-NEXT:  [1,3]     . DeE-----R    .    ..   cmpl %edi, %edx

 # ZNVER1-NEXT:  [8,1]     .    .  DeeeeeeeeER.   vpaddd (%r8), %ymm0, %ymm0
 # ZNVER1-NEXT:  [8,2]     .    .  DeE-------R.   addq $32, %r8
 # ZNVER1-NEXT:  [8,3]     .    .  D=eE------R.   cmpl %edi, %edx
 # ZNVER1-NEXT:  [9,0]     .    .   DeE------R.   addl $1, %edx
 # ZNVER1-NEXT:  [9,1]     .    .   DeeeeeeeeER   vpaddd (%r8), %ymm0, %ymm0
 # ZNVER1-NEXT:  [9,2]     .    .   DeE-------R   addq $32, %r8
 # ZNVER1-NEXT:  [9,3]     .    .   D=eE------R   cmpl %edi, %edx
 
+# ZNVER2:       [0,0]     DeER .    .    .   .   addl   $1, %edx
+# ZNVER2-NEXT:  [0,1]     DeeeeeeeeER    .   .   vpaddd (%r8), %ymm0, %ymm0
+# ZNVER2-NEXT:  [0,2]     DeE-------R    .   .   addq   $32, %r8
+# ZNVER2-NEXT:  [0,3]     D=eE------R    .   .   cmpl   %edi, %edx
+# ZNVER2-NEXT:  [1,0]     .DeE------R    .   .   addl   $1, %edx
+# ZNVER2-NEXT:  [1,1]     .DeeeeeeeeER   .   .   vpaddd (%r8), %ymm0, %ymm0
+# ZNVER2-NEXT:  [1,2]     .DeE-------R   .   .   addq   $32, %r8
+# ZNVER2-NEXT:  [1,3]     .D=eE------R   .   .   cmpl   %edi, %edx
+# ZNVER2-NEXT:  [2,0]     . DeE------R   .   .   addl   $1, %edx
+# ZNVER2-NEXT:  [2,1]     . DeeeeeeeeER  .   .   vpaddd (%r8), %ymm0, %ymm0
+# ZNVER2-NEXT:  [2,2]     . DeE-------R  .   .   addq   $32, %r8
+# ZNVER2-NEXT:  [2,3]     . D=eE------R  .   .   cmpl   %edi, %edx
+# ZNVER2-NEXT:  [3,0]     .  DeE------R  .   .   addl   $1, %edx
+# ZNVER2-NEXT:  [3,1]     .  DeeeeeeeeER .   .   vpaddd (%r8), %ymm0, %ymm0
+# ZNVER2-NEXT:  [3,2]     .  DeE-------R .   .   addq   $32, %r8
+# ZNVER2-NEXT:  [3,3]     .  D=eE------R .   .   cmpl   %edi, %edx
+# ZNVER2-NEXT:  [4,0]     .   DeE------R .   .   addl   $1, %edx
+# ZNVER2-NEXT:  [4,1]     .   DeeeeeeeeER.   .   vpaddd (%r8), %ymm0, %ymm0
+# ZNVER2-NEXT:  [4,2]     .   DeE-------R.   .   addq   $32, %r8
+# ZNVER2-NEXT:  [4,3]     .   D=eE------R.   .   cmpl   %edi, %edx
+# ZNVER2-NEXT:  [5,0]     .    DeE------R.   .   addl   $1, %edx
+# ZNVER2-NEXT:  [5,1]     .    DeeeeeeeeER   .   vpaddd (%r8), %ymm0, %ymm0
+# ZNVER2-NEXT:  [5,2]     .    DeE-------R   .   addq   $32, %r8
+# ZNVER2-NEXT:  [5,3]     .    D=eE------R   .   cmpl   %edi, %edx
+# ZNVER2-NEXT:  [6,0]     .    .DeE------R   .   addl   $1, %edx
+# ZNVER2-NEXT:  [6,1]     .    .DeeeeeeeeER  .   vpaddd (%r8), %ymm0, %ymm0
+# ZNVER2-NEXT:  [6,2]     .    .DeE-------R  .   addq   $32, %r8
+# ZNVER2-NEXT:  [6,3]     .    .D=eE------R  .   cmpl   %edi, %edx
+# ZNVER2-NEXT:  [7,0]     .    . DeE------R  .   addl   $1, %edx
+# ZNVER2-NEXT:  [7,1]     .    . DeeeeeeeeER .   vpaddd (%r8), %ymm0, %ymm0
+# ZNVER2-NEXT:  [7,2]     .    . DeE-------R .   addq   $32, %r8
+# ZNVER2-NEXT:  [7,3]     .    . D=eE------R .   cmpl   %edi, %edx
+# ZNVER2-NEXT:  [8,0]     .    .  DeE------R .   addl   $1, %edx
+# ZNVER2-NEXT:  [8,1]     .    .  DeeeeeeeeER.   vpaddd (%r8), %ymm0, %ymm0
+# ZNVER2-NEXT:  [8,2]     .    .  DeE-------R.   addq   $32, %r8
+# ZNVER2-NEXT:  [8,3]     .    .  D=eE------R.   cmpl   %edi, %edx
+# ZNVER2-NEXT:  [9,0]     .    .   DeE------R.   addl   $1, %edx
+# ZNVER2-NEXT:  [9,1]     .    .   DeeeeeeeeER   vpaddd (%r8), %ymm0, %ymm0
+# ZNVER2-NEXT:  [9,2]     .    .   DeE-------R   addq   $32, %r8
+# ZNVER2-NEXT:  [9,3]     .    .   D=eE------R   cmpl   %edi, %edx
+
 # ALL:          Average Wait times (based on the timeline view):
 # ALL-NEXT:     [0]: Executions
 # ALL-NEXT:     [1]: Average time spent waiting in a scheduler's queue
 # ALL-NEXT:     [2]: Average time spent waiting in a scheduler's queue while ready
 # ALL-NEXT:     [3]: Average time elapsed from WB until retire stage
 
 # ALL:                [0]    [1]    [2]    [3]
 
 # BDWELL-NEXT:  0.     10    1.0    0.4    4.5       addl $1, %edx
 # HASWELL-NEXT: 0.     10    1.0    0.4    5.4       addl $1, %edx
 # SKYLAKE-NEXT: 0.     10    1.9    0.1    5.4       addl $1, %edx
 # ZNVER1-NEXT:  0.     10    1.0    0.1    5.4       addl $1, %edx
+# ZNVER2-NEXT:  0.     10    1.0    0.1    5.4       addl $1, %edx
 
 # ALL-NEXT:     1.     10    1.0    0.1    0.0       vpaddd (%r8), %ymm0, %ymm0
 
 # BDWELL-NEXT:  2.     10    1.0    0.4    5.7       addq $32, %r8
 # BDWELL-NEXT:  3.     10    1.0    0.0    5.3       cmpl %edi, %edx
 # BDWELL-NEXT:         10    1.0    0.2    3.9       <total>
 
 # HASWELL-NEXT: 2.     10    1.0    0.4    6.7       addq $32, %r8
 # HASWELL-NEXT: 3.     10    1.0    0.0    6.3       cmpl %edi, %edx
 # HASWELL-NEXT:        10    1.0    0.2    4.6       <total>
 
 # SKYLAKE-NEXT: 2.     10    1.0    0.1    7.0       addq $32, %r8
 # SKYLAKE-NEXT: 3.     10    2.0    0.0    6.0       cmpl %edi, %edx
 # SKYLAKE-NEXT:        10    1.5    0.1    4.6       <total>
 
 # ZNVER1-NEXT:  2.     10    1.0    0.1    7.0       addq $32, %r8
 # ZNVER1-NEXT:  3.     10    2.0    0.0    6.0       cmpl %edi, %edx
 # ZNVER1-NEXT:         10    1.3    0.1    4.6       <total>
+
+# ZNVER2-NEXT:  2.     10    1.0    0.1    7.0       addq $32, %r8
+# ZNVER2-NEXT:  3.     10    2.0    0.0    6.0       cmpl %edi, %edx
+# ZNVER2-NEXT:         10    1.3    0.1    4.6       <total>

llvm/test/tools/llvm-mca/X86/register-file-statistics.s

 # NOTE: Assertions have been autogenerated by utils/update_mca_test_checks.py
 # RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=x86-64 -iterations=1 -all-stats=false -all-views=false -register-file-stats < %s | FileCheck --check-prefixes=ALL %s
 # RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=x86-64 -iterations=1 -all-stats=false -all-views=false -register-file-stats < %s | FileCheck --check-prefixes=BARCELONA %s
 # RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=bdver2 -iterations=1 -all-stats=false -all-views=false -register-file-stats < %s | FileCheck --check-prefixes=ALL,BDVER2 %s
 # RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=btver2 -iterations=1 -all-stats=false -all-views=false -register-file-stats < %s | FileCheck --check-prefixes=ALL,BTVER2 %s
 # RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=znver1 -iterations=1 -all-stats=false -all-views=false -register-file-stats < %s | FileCheck --check-prefixes=ALL,ZNVER1 %s
+# RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=znver2 -iterations=1 -all-stats=false -all-views=false -register-file-stats < %s | FileCheck --check-prefixes=ALL,ZNVER2 %s
 # RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=sandybridge -iterations=1 -all-stats=false -all-views=false -register-file-stats < %s | FileCheck --check-prefixes=ALL %s
 # RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=ivybridge -iterations=1 -all-stats=false -all-views=false -register-file-stats < %s | FileCheck --check-prefixes=ALL %s
 # RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=haswell -iterations=1 -all-stats=false -all-views=false -register-file-stats < %s | FileCheck --check-prefixes=ALL %s
 # RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=broadwell -iterations=1 -all-stats=false -all-views=false -register-file-stats < %s | FileCheck --check-prefixes=ALL %s
 # RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=knl -iterations=1 -all-stats=false -all-views=false -register-file-stats < %s | FileCheck --check-prefixes=ALL %s
 # RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=skylake -iterations=1 -all-stats=false -all-views=false -register-file-stats < %s | FileCheck --check-prefixes=ALL %s
 # RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=skylake-avx512 -iterations=1 -all-stats=false -all-views=false -register-file-stats < %s | FileCheck --check-prefixes=ALL %s
 # RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=slm -iterations=1 -all-stats=false -all-views=false -register-file-stats < %s | FileCheck --check-prefixes=ALL %s

 # BTVER2-NEXT:       Total number of mappings created: 0
 # BTVER2-NEXT:       Max number of mappings used:      0
 
 # ZNVER1:         *  Register File #1 -- ZnFpuPRF:
 # ZNVER1-NEXT:       Number of physical registers:     160
 # ZNVER1-NEXT:       Total number of mappings created: 0
 # ZNVER1-NEXT:       Max number of mappings used:      0
 
+# ZNVER2:         *  Register File #1 -- Zn2FpuPRF:
+# ZNVER2-NEXT:       Number of physical registers:     160
+# ZNVER2-NEXT:       Total number of mappings created: 0
+# ZNVER2-NEXT:       Max number of mappings used:      0
+
 # BDVER2:         *  Register File #2 -- PdIntegerPRF:
 # BDVER2-NEXT:       Number of physical registers:     96
 # BDVER2-NEXT:       Total number of mappings created: 2
 # BDVER2-NEXT:       Max number of mappings used:      2
 
 # BTVER2:         *  Register File #2 -- JIntegerPRF:
 # BTVER2-NEXT:       Number of physical registers:     64
 # BTVER2-NEXT:       Total number of mappings created: 2
 # BTVER2-NEXT:       Max number of mappings used:      2
 
 # ZNVER1:         *  Register File #2 -- ZnIntegerPRF:
 # ZNVER1-NEXT:       Number of physical registers:     168
 # ZNVER1-NEXT:       Total number of mappings created: 2
 # ZNVER1-NEXT:       Max number of mappings used:      2
+
+# ZNVER2:         *  Register File #2 -- Zn2IntegerPRF:
+# ZNVER2-NEXT:       Number of physical registers:     168
+# ZNVER2-NEXT:       Total number of mappings created: 2
+# ZNVER2-NEXT:       Max number of mappings used:      2

llvm/test/tools/llvm-mca/X86/scheduler-queue-usage.s

 # NOTE: Assertions have been autogenerated by utils/update_mca_test_checks.py
 # RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=x86-64 -iterations=1 -all-stats=false -all-views=false -scheduler-stats < %s | FileCheck --check-prefixes=ALL,BARCELONA %s
 # RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=bdver2 -iterations=1 -all-stats=false -all-views=false -scheduler-stats < %s | FileCheck --check-prefixes=ALL,BDVER2 %s
 # RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=btver2 -iterations=1 -all-stats=false -all-views=false -scheduler-stats < %s | FileCheck --check-prefixes=ALL,BTVER2 %s
 # RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=znver1 -iterations=1 -all-stats=false -all-views=false -scheduler-stats < %s | FileCheck --check-prefixes=ALL,ZNVER1 %s
+# RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=znver2 -iterations=1 -all-stats=false -all-views=false -scheduler-stats < %s | FileCheck --check-prefixes=ALL,ZNVER2 %s
 # RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=sandybridge -iterations=1 -all-stats=false -all-views=false -scheduler-stats < %s | FileCheck --check-prefixes=ALL,SNB %s
 # RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=ivybridge -iterations=1 -all-stats=false -all-views=false -scheduler-stats < %s | FileCheck --check-prefixes=ALL,IVB %s
 # RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=haswell -iterations=1 -all-stats=false -all-views=false -scheduler-stats < %s | FileCheck --check-prefixes=ALL,HSW %s
 # RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=broadwell -iterations=1 -all-stats=false -all-views=false -scheduler-stats < %s | FileCheck --check-prefixes=ALL,BDW %s
 # RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=knl -iterations=1 -all-stats=false -all-views=false -scheduler-stats < %s | FileCheck --check-prefixes=ALL,KNL %s
 # RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=skylake -iterations=1 -all-stats=false -all-views=false -scheduler-stats < %s | FileCheck --check-prefixes=ALL,SKX %s
 # RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=skylake-avx512 -iterations=1 -all-stats=false -all-views=false -scheduler-stats < %s | FileCheck --check-prefixes=ALL,SKX-AVX512 %s
 # RUN: llvm-mca %s -mtriple=x86_64-unknown-unknown -mcpu=slm -iterations=1 -all-stats=false -all-views=false -scheduler-stats < %s | FileCheck --check-prefixes=ALL,SLM %s

 # SNB-NEXT:        [4] Total number of buffer entries.
 
 # ZNVER1:          Scheduler's queue usage:
 # ZNVER1-NEXT:     [1] Resource name.
 # ZNVER1-NEXT:     [2] Average number of used buffer entries.
 # ZNVER1-NEXT:     [3] Maximum number of used buffer entries.
 # ZNVER1-NEXT:     [4] Total number of buffer entries.
 
+# ZNVER2:         Scheduler's queue usage:
+# ZNVER2-NEXT:     [1] Resource name.
+# ZNVER2-NEXT:     [2] Average number of used buffer entries.
+# ZNVER2-NEXT:     [3] Maximum number of used buffer entries.
+# ZNVER2-NEXT:     [4] Total number of buffer entries.
+
 # BARCELONA:        [1]            [2]        [3]        [4]
 # BARCELONA-NEXT:  SBPortAny        0          1          54
 
 # BDVER2:           [1]            [2]        [3]        [4]
 # BDVER2-NEXT:     PdEX             0          1          40
 # BDVER2-NEXT:     PdFPU            0          0          64
 # BDVER2-NEXT:     PdLoad           0          0          40
 # BDVER2-NEXT:     PdStore          0          0          24

llvm/test/tools/llvm-mca/X86/sqrt-rsqrt-rcp-memop.s

 # NOTE: Assertions have been autogenerated by utils/update_mca_test_checks.py
 # RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=x86-64 -iterations=1 -all-views=false -timeline < %s | FileCheck %s -check-prefix=ALL -check-prefix=BARCELONA
 # RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=bdver2 -iterations=1 -all-views=false -timeline < %s | FileCheck %s -check-prefix=ALL -check-prefix=BDVER2
 # RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=btver2 -iterations=1 -all-views=false -timeline < %s | FileCheck %s -check-prefix=ALL -check-prefix=BTVER2
 # RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=znver1 -iterations=1 -all-views=false -timeline < %s | FileCheck %s -check-prefix=ALL -check-prefix=ZNVER1
+# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=znver2 -iterations=1 -all-views=false -timeline < %s | FileCheck %s -check-prefix=ALL -check-prefix=ZNVER2
 # RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=haswell -iterations=1 -all-views=false -timeline < %s | FileCheck %s -check-prefix=ALL -check-prefix=HASWELL
 # RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=broadwell -iterations=1 -all-views=false -timeline < %s | FileCheck %s -check-prefix=ALL -check-prefix=BROADWELL
 # RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=skylake -iterations=1 -all-views=false -timeline < %s | FileCheck %s -check-prefix=ALL -check-prefix=SKYLAKE
 
 #LLVM-MCA-BEGIN  test_sqrtss
 leaq 8(%rsp, %rdi, 2), %rax
 sqrtss (%rax), %xmm1
 #LLVM-MCA-END

 # HASWELL-NEXT:   Index     0123456789
 
 # SKYLAKE-NEXT:                       0123456789
 # SKYLAKE-NEXT:   Index     0123456789          0
 
 # ZNVER1-NEXT:                        0123456789          0
 # ZNVER1-NEXT:    Index     0123456789          0123456789
 
+# ZNVER2-NEXT:                        0123456789          0
+# ZNVER2-NEXT:    Index     0123456789          0123456789
+
 # BARCELONA:      [0,0]     DeER .    .    .    .  .   leaq 8(%rsp,%rdi,2), %rax
 # BARCELONA-NEXT: [0,1]     D=eeeeeeeeeeeeeeeeeeeeER   sqrtss (%rax), %xmm1
 
 # BDVER2:         [0,0]     DeER .    .    . .   leaq 8(%rsp,%rdi,2), %rax
 # BDVER2-NEXT:    [0,1]     D=eeeeeeeeeeeeeeER   sqrtss (%rax), %xmm1
 
 # BROADWELL:      [0,0]     DeER .    .    .   .   leaq 8(%rsp,%rdi,2), %rax
 # BROADWELL-NEXT: [0,1]     D=eeeeeeeeeeeeeeeeER   sqrtss (%rax), %xmm1
 
 # BTVER2:         [0,0]     DeeER.    .    .    .    .    .   leaq  8(%rsp,%rdi,2), %rax
 # BTVER2-NEXT:    [0,1]     D==eeeeeeeeeeeeeeeeeeeeeeeeeeER   sqrtss  (%rax), %xmm1
 
 # HASWELL:        [0,0]     DeER .    .    .   .   leaq 8(%rsp,%rdi,2), %rax
 # HASWELL-NEXT:   [0,1]     D=eeeeeeeeeeeeeeeeER   sqrtss (%rax), %xmm1
 
 # SKYLAKE:        [0,0]     DeER .    .    .    .   leaq  8(%rsp,%rdi,2), %rax
 # SKYLAKE-NEXT:   [0,1]     D=eeeeeeeeeeeeeeeeeER   sqrtss  (%rax), %xmm1
 
 # ZNVER1:         [0,0]     DeER .    .    .    .    .    .   leaq  8(%rsp,%rdi,2), %rax
 # ZNVER1-NEXT:    [0,1]     D=eeeeeeeeeeeeeeeeeeeeeeeeeeeER   sqrtss  (%rax), %xmm1
 
+# ZNVER2:         [0,0]     DeER .    .    .    .    .    .   leaq  8(%rsp,%rdi,2), %rax
+# ZNVER2-NEXT:    [0,1]     D=eeeeeeeeeeeeeeeeeeeeeeeeeeeER   sqrtss  (%rax), %xmm1
+
 # ALL:            Average Wait times (based on the timeline view):
 # ALL-NEXT:       [0]: Executions
 # ALL-NEXT:       [1]: Average time spent waiting in a scheduler's queue
 # ALL-NEXT:       [2]: Average time spent waiting in a scheduler's queue while ready
 # ALL-NEXT:       [3]: Average time elapsed from WB until retire stage
 
 # ALL:                  [0]    [1]    [2]    [3]
 # ALL-NEXT:       0.     1     1.0    1.0    0.0       leaq 8(%rsp,%rdi,2), %rax

 # HASWELL-NEXT:          1     1.5    0.5    0.0       <total>
 
 # SKYLAKE-NEXT:   1.     1     2.0    0.0    0.0       sqrtss (%rax), %xmm1
 # SKYLAKE-NEXT:          1     1.5    0.5    0.0       <total>
 
 # ZNVER1-NEXT:    1.     1     2.0    0.0    0.0       sqrtss (%rax), %xmm1
 # ZNVER1-NEXT:           1     1.5    0.5    0.0       <total>
 
+# ZNVER2-NEXT:    1.     1     2.0    0.0    0.0       sqrtss (%rax), %xmm1
+# ZNVER2-NEXT:           1     1.5    0.5    0.0       <total>
+
 # ALL:            [1] Code Region - test_sqrtsd
 
 # ALL:            Timeline view:
 
 # BARCELONA-NEXT:                     0123456789          0
 # BARCELONA-NEXT: Index     0123456789          0123456789
 
 # BDVER2-NEXT:                        01234567

 # HASWELL-NEXT:   Index     0123456789          01234
 
 # SKYLAKE-NEXT:                       0123456789
 # SKYLAKE-NEXT:   Index     0123456789          0123456
 
 # ZNVER1-NEXT:                        0123456789          0
 # ZNVER1-NEXT:    Index     0123456789          0123456789
 
+# ZNVER2-NEXT:                        0123456789          0
+# ZNVER2-NEXT:    Index     0123456789          0123456789
+
 # BARCELONA:      [0,0]     DeER .    .    .    .    .    .   leaq  8(%rsp,%rdi,2), %rax
 # BARCELONA-NEXT: [0,1]     D=eeeeeeeeeeeeeeeeeeeeeeeeeeeER   sqrtsd  (%rax), %xmm1
 
 # BDVER2:         [0,0]     DeER .    .    . .   leaq 8(%rsp,%rdi,2), %rax
 # BDVER2-NEXT:    [0,1]     D=eeeeeeeeeeeeeeER   sqrtsd (%rax), %xmm1
 
 # BROADWELL:      [0,0]     DeER .    .    .    .   .   leaq  8(%rsp,%rdi,2), %rax
 # BROADWELL-NEXT: [0,1]     D=eeeeeeeeeeeeeeeeeeeeeER   sqrtsd  (%rax), %xmm1
 
 # BTVER2:         [0,0]     DeeER.    .    .    .    .    .    ..   leaq  8(%rsp,%rdi,2), %rax
 # BTVER2-NEXT:    [0,1]     D==eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeER   sqrtsd  (%rax), %xmm1
 
 # HASWELL:        [0,0]     DeER .    .    .    .   .   leaq  8(%rsp,%rdi,2), %rax
 # HASWELL-NEXT:   [0,1]     D=eeeeeeeeeeeeeeeeeeeeeER   sqrtsd  (%rax), %xmm1
 
 # SKYLAKE:        [0,0]     DeER .    .    .    .    ..   leaq  8(%rsp,%rdi,2), %rax
 # SKYLAKE-NEXT:   [0,1]     D=eeeeeeeeeeeeeeeeeeeeeeeER   sqrtsd  (%rax), %xmm1
 
 # ZNVER1:         [0,0]     DeER .    .    .    .    .    .   leaq  8(%rsp,%rdi,2), %rax
 # ZNVER1-NEXT:    [0,1]     D=eeeeeeeeeeeeeeeeeeeeeeeeeeeER   sqrtsd  (%rax), %xmm1
 
+# ZNVER2:         [0,0]     DeER .    .    .    .    .    .   leaq  8(%rsp,%rdi,2), %rax
+# ZNVER2-NEXT:    [0,1]     D=eeeeeeeeeeeeeeeeeeeeeeeeeeeER   sqrtsd  (%rax), %xmm1
+
 # ALL:            Average Wait times (based on the timeline view):
 # ALL-NEXT:       [0]: Executions
 # ALL-NEXT:       [1]: Average time spent waiting in a scheduler's queue
 # ALL-NEXT:       [2]: Average time spent waiting in a scheduler's queue while ready
 # ALL-NEXT:       [3]: Average time elapsed from WB until retire stage
 
 # ALL:                  [0]    [1]    [2]    [3]
 # ALL-NEXT:       0.     1     1.0    1.0    0.0       leaq 8(%rsp,%rdi,2), %rax

 # HASWELL-NEXT:          1     1.5    0.5    0.0       <total>
 
 # SKYLAKE-NEXT:   1.     1     2.0    0.0    0.0       sqrtsd (%rax), %xmm1
 # SKYLAKE-NEXT:          1     1.5    0.5    0.0       <total>
 
 # ZNVER1-NEXT:    1.     1     2.0    0.0    0.0       sqrtsd (%rax), %xmm1
 # ZNVER1-NEXT:           1     1.5    0.5    0.0       <total>
 
+# ZNVER2-NEXT:    1.     1     2.0    0.0    0.0       sqrtsd (%rax), %xmm1
+# ZNVER2-NEXT:           1     1.5    0.5    0.0       <total>
+
 # ALL:            [2] Code Region - test_rsqrtss
 
 # ALL:            Timeline view:
 
 # BARCELONA-NEXT:                     01234
 # BDVER2-NEXT:                        0123
 # BROADWELL-NEXT:                     0123
 # BTVER2-NEXT:                        01
 # HASWELL-NEXT:                       0123
 # SKYLAKE-NEXT:                       012
 # ZNVER1-NEXT:                        012345
+# ZNVER2-NEXT:                        012345
 
 # ALL-NEXT:       Index     0123456789
 
 # BARCELONA:      [0,0]     DeER .    .   .   leaq  8(%rsp,%rdi,2), %rax
 # BARCELONA-NEXT: [0,1]     D=eeeeeeeeeeeER   rsqrtss (%rax), %xmm1
 
 # BDVER2:         [0,0]     DeER .    .  .   leaq 8(%rsp,%rdi,2), %rax
 # BDVER2-NEXT:    [0,1]     D=eeeeeeeeeeER   rsqrtss  (%rax), %xmm1
 
 # BROADWELL:      [0,0]     DeER .    .  .   leaq 8(%rsp,%rdi,2), %rax
 # BROADWELL-NEXT: [0,1]     D=eeeeeeeeeeER   rsqrtss  (%rax), %xmm1
 
 # BTVER2:         [0,0]     DeeER.    ..   leaq 8(%rsp,%rdi,2), %rax
 # BTVER2-NEXT:    [0,1]     D==eeeeeeeER   rsqrtss  (%rax), %xmm1
 
 # HASWELL:        [0,0]     DeER .    .  .   leaq 8(%rsp,%rdi,2), %rax
 # HASWELL-NEXT:   [0,1]     D=eeeeeeeeeeER   rsqrtss  (%rax), %xmm1
 
 # SKYLAKE:        [0,0]     DeER .    . .   leaq  8(%rsp,%rdi,2), %rax
 # SKYLAKE-NEXT:   [0,1]     D=eeeeeeeeeER   rsqrtss (%rax), %xmm1
 
 # ZNVER1:         [0,0]     DeER .    .    .   leaq 8(%rsp,%rdi,2), %rax
 # ZNVER1-NEXT:    [0,1]     D=eeeeeeeeeeeeER   rsqrtss  (%rax), %xmm1
 
+# ZNVER2:         [0,0]     DeER .    .    .   leaq 8(%rsp,%rdi,2), %rax
+# ZNVER2-NEXT:    [0,1]     D=eeeeeeeeeeeeER   rsqrtss  (%rax), %xmm1
+
 # ALL:            Average Wait times (based on the timeline view):
 # ALL-NEXT:       [0]: Executions
 # ALL-NEXT:       [1]: Average time spent waiting in a scheduler's queue
 # ALL-NEXT:       [2]: Average time spent waiting in a scheduler's queue while ready
 # ALL-NEXT:       [3]: Average time elapsed from WB until retire stage
 
 # ALL:                  [0]    [1]    [2]    [3]
 # ALL-NEXT:       0.     1     1.0    1.0    0.0       leaq 8(%rsp,%rdi,2), %rax

 # HASWELL-NEXT:          1     1.5    0.5    0.0       <total>
 
 # SKYLAKE-NEXT:   1.     1     2.0    0.0    0.0       rsqrtss  (%rax), %xmm1
 # SKYLAKE-NEXT:          1     1.5    0.5    0.0       <total>
 
 # ZNVER1-NEXT:    1.     1     2.0    0.0    0.0       rsqrtss  (%rax), %xmm1
 # ZNVER1-NEXT:           1     1.5    0.5    0.0       <total>
 
+# ZNVER2-NEXT:    1.     1     2.0    0.0    0.0       rsqrtss  (%rax), %xmm1
+# ZNVER2-NEXT:           1     1.5    0.5    0.0       <total>
+
 # ALL:            [3] Code Region - test_rcp
 
 # ALL:            Timeline view:
 
 # BARCELONA-NEXT:                     01234
 # BDVER2-NEXT:                        0123
 # BROADWELL-NEXT:                     0123
 # BTVER2-NEXT:                        01
 # HASWELL-NEXT:                       0123
 # SKYLAKE-NEXT:                       012
 # ZNVER1-NEXT:                        012345
+# ZNVER2-NEXT:                        012345
 
 # ALL-NEXT:       Index     0123456789
 
 # BARCELONA:      [0,0]     DeER .    .   .   leaq  8(%rsp,%rdi,2), %rax
 # BARCELONA-NEXT: [0,1]     D=eeeeeeeeeeeER   rcpss (%rax), %xmm1
 
 # BDVER2:         [0,0]     DeER .    .  .   leaq 8(%rsp,%rdi,2), %rax
 # BDVER2-NEXT:    [0,1]     D=eeeeeeeeeeER   rcpss  (%rax), %xmm1
 
 # BROADWELL:      [0,0]     DeER .    .  .   leaq 8(%rsp,%rdi,2), %rax
 # BROADWELL-NEXT: [0,1]     D=eeeeeeeeeeER   rcpss  (%rax), %xmm1
 
 # BTVER2:         [0,0]     DeeER.    ..   leaq 8(%rsp,%rdi,2), %rax
 # BTVER2-NEXT:    [0,1]     D==eeeeeeeER   rcpss  (%rax), %xmm1
 
 # HASWELL:        [0,0]     DeER .    .  .   leaq 8(%rsp,%rdi,2), %rax
 # HASWELL-NEXT:   [0,1]     D=eeeeeeeeeeER   rcpss  (%rax), %xmm1
 
 # SKYLAKE:        [0,0]     DeER .    . .   leaq  8(%rsp,%rdi,2), %rax
 # SKYLAKE-NEXT:   [0,1]     D=eeeeeeeeeER   rcpss (%rax), %xmm1
 
 # ZNVER1:         [0,0]     DeER .    .    .   leaq 8(%rsp,%rdi,2), %rax
 # ZNVER1-NEXT:    [0,1]     D=eeeeeeeeeeeeER   rcpss  (%rax), %xmm1
 
+# ZNVER2:         [0,0]     DeER .    .    .   leaq 8(%rsp,%rdi,2), %rax
+# ZNVER2-NEXT:    [0,1]     D=eeeeeeeeeeeeER   rcpss  (%rax), %xmm1
+
 # ALL:            Average Wait times (based on the timeline view):
 # ALL-NEXT:       [0]: Executions
 # ALL-NEXT:       [1]: Average time spent waiting in a scheduler's queue
 # ALL-NEXT:       [2]: Average time spent waiting in a scheduler's queue while ready
 # ALL-NEXT:       [3]: Average time elapsed from WB until retire stage
 
 # ALL:                  [0]    [1]    [2]    [3]
 # ALL-NEXT:       0.     1     1.0    1.0    0.0       leaq 8(%rsp,%rdi,2), %rax

 # HASWELL-NEXT:   1.     1     2.0    0.0    0.0       rcpss  (%rax), %xmm1
 # HASWELL-NEXT:          1     1.5    0.5    0.0       <total>
 
 # SKYLAKE-NEXT:   1.     1     2.0    0.0    0.0       rcpss  (%rax), %xmm1
 # SKYLAKE-NEXT:          1     1.5    0.5    0.0       <total>
 
 # ZNVER1-NEXT:    1.     1     2.0    0.0    0.0       rcpss  (%rax), %xmm1
 # ZNVER1-NEXT:           1     1.5    0.5    0.0       <total>
+
+# ZNVER2-NEXT:    1.     1     2.0    0.0    0.0       rcpss  (%rax), %xmm1
+# ZNVER2-NEXT:           1     1.5    0.5    0.0       <total>

llvm/test/tools/llvm-mca/X86/variable-blend-read-after-ld-1.s

 # RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=skylake -iterations=1 -timeline -instruction-info=false -resource-pressure=false < %s | FileCheck %s -check-prefix=ALL -check-prefix=SKYLAKE
 
 # RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=bdver2 -iterations=1 -timeline -instruction-info=false -resource-pressure=false < %s | FileCheck %s -check-prefix=ALL -check-prefix=BDVER2
 
 # RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=btver2 -iterations=1 -timeline -instruction-info=false -resource-pressure=false < %s | FileCheck %s -check-prefix=ALL -check-prefix=BTVER2
 
 # RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=znver1 -iterations=1 -timeline -instruction-info=false -resource-pressure=false < %s | FileCheck %s -check-prefix=ALL -check-prefix=ZNVER1
 
+# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=znver2 -iterations=1 -timeline -instruction-info=false -resource-pressure=false < %s | FileCheck %s -check-prefix=ALL -check-prefix=ZNVER2
+
 vaddps %xmm0, %xmm0, %xmm1
 vblendvps %xmm1, (%rdi), %xmm2, %xmm3
 
 # ALL:          Iterations:        1
 # ALL-NEXT:     Instructions:      2
 
 # BDVER2-NEXT:  Total Cycles:      10
 # BDVER2-NEXT:  Total uOps:        2

 # SANDY-NEXT:   Total uOps:        4
 
 # SKYLAKE-NEXT: Total Cycles:      11
 # SKYLAKE-NEXT: Total uOps:        4
 
 # ZNVER1-NEXT:  Total Cycles:      11
 # ZNVER1-NEXT:  Total uOps:        2
 
+# ZNVER2-NEXT:  Total Cycles:      11
+# ZNVER2-NEXT:  Total uOps:        2
+
 # BDVER2:       Dispatch Width:    4
 # BDVER2-NEXT:  uOps Per Cycle:    0.20
 # BDVER2-NEXT:  IPC:               0.20
 # BDVER2-NEXT:  Block RThroughput: 2.0
 
 # BDWELL:       Dispatch Width:    4
 # BDWELL-NEXT:  uOps Per Cycle:    0.40
 # BDWELL-NEXT:  IPC:               0.20

 # SKYLAKE-NEXT: IPC:               0.18
 # SKYLAKE-NEXT: Block RThroughput: 0.7
 
 # ZNVER1:       Dispatch Width:    4
 # ZNVER1-NEXT:  uOps Per Cycle:    0.18
 # ZNVER1-NEXT:  IPC:               0.18
 # ZNVER1-NEXT:  Block RThroughput: 1.0
 
+# ZNVER2:       Dispatch Width:    4
+# ZNVER2-NEXT:  uOps Per Cycle:    0.18
+# ZNVER2-NEXT:  IPC:               0.18
+# ZNVER2-NEXT:  Block RThroughput: 1.0
+
 # BDVER2:       Timeline view:
 # BDVER2-NEXT:  Index     0123456789
 
 # BDWELL:       Timeline view:
 # BDWELL-NEXT:  Index     0123456789
 
 # BTVER2:       Timeline view:
 # BTVER2-NEXT:                      0

 # SKYLAKE:      Timeline view:
 # SKYLAKE-NEXT:                     0
 # SKYLAKE-NEXT: Index     0123456789
 
 # ZNVER1:       Timeline view:
 # ZNVER1-NEXT:                      0
 # ZNVER1-NEXT:  Index     0123456789
 
+# ZNVER2:       Timeline view:
+# ZNVER2-NEXT:                      0
+# ZNVER2-NEXT:  Index     0123456789
+
 # BDVER2:       [0,0]     DeeeeeER .   vaddps %xmm0, %xmm0, %xmm1
 # BDVER2-NEXT:  [0,1]     DeeeeeeeER   vblendvps  %xmm1, (%rdi), %xmm2, %xmm3
 
 # BDWELL:       [0,0]     DeeeER   .   vaddps %xmm0, %xmm0, %xmm1
 # BDWELL-NEXT:  [0,1]     DeeeeeeeER   vblendvps  %xmm1, (%rdi), %xmm2, %xmm3
 
 # BTVER2:       [0,0]     DeeeER    .   vaddps  %xmm0, %xmm0, %xmm1
 # BTVER2-NEXT:  [0,1]     .DeeeeeeeER   vblendvps %xmm1, (%rdi), %xmm2, %xmm3
 
 # HASWELL:      [0,0]     DeeeER    .   vaddps  %xmm0, %xmm0, %xmm1
 # HASWELL-NEXT: [0,1]     DeeeeeeeeER   vblendvps %xmm1, (%rdi), %xmm2, %xmm3
 
 # IVY:          [0,0]     DeeeER    .   vaddps  %xmm0, %xmm0, %xmm1
 # IVY-NEXT:     [0,1]     DeeeeeeeeER   vblendvps %xmm1, (%rdi), %xmm2, %xmm3
 
 # SANDY:        [0,0]     DeeeER    .   vaddps  %xmm0, %xmm0, %xmm1
 # SANDY-NEXT:   [0,1]     DeeeeeeeeER   vblendvps %xmm1, (%rdi), %xmm2, %xmm3
 
 # SKYLAKE:      [0,0]     DeeeeER   .   vaddps  %xmm0, %xmm0, %xmm1
 # SKYLAKE-NEXT: [0,1]     DeeeeeeeeER   vblendvps %xmm1, (%rdi), %xmm2, %xmm3
 
 # ZNVER1:       [0,0]     DeeeER    .   vaddps  %xmm0, %xmm0, %xmm1
 # ZNVER1-NEXT:  [0,1]     DeeeeeeeeER   vblendvps %xmm1, (%rdi), %xmm2, %xmm3
 
+# ZNVER2:       [0,0]     DeeeER    .   vaddps  %xmm0, %xmm0, %xmm1
+# ZNVER2-NEXT:  [0,1]     DeeeeeeeeER   vblendvps %xmm1, (%rdi), %xmm2, %xmm3
+
 # ALL:          Average Wait times (based on the timeline view):
 # ALL-NEXT:     [0]: Executions
 # ALL-NEXT:     [1]: Average time spent waiting in a scheduler's queue
 # ALL-NEXT:     [2]: Average time spent waiting in a scheduler's queue while ready
 # ALL-NEXT:     [3]: Average time elapsed from WB until retire stage
 
 # ALL:                [0]    [1]    [2]    [3]
 # ALL-NEXT:     0.     1     1.0    1.0    0.0       vaddps %xmm0, %xmm0, %xmm1

 # SANDY-NEXT:   1.     1     1.0    0.0    0.0       vblendvps  %xmm1, (%rdi), %xmm2, %xmm3
 # SANDY-NEXT:          1     1.0    0.5    0.0       <total>
 
 # SKYLAKE-NEXT: 1.     1     1.0    0.0    0.0       vblendvps  %xmm1, (%rdi), %xmm2, %xmm3
 # SKYLAKE-NEXT:        1     1.0    0.5    0.0       <total>
 
 # ZNVER1-NEXT:  1.     1     1.0    0.0    0.0       vblendvps  %xmm1, (%rdi), %xmm2, %xmm3
 # ZNVER1-NEXT:         1     1.0    0.5    0.0       <total>
+
+# ZNVER2-NEXT:  1.     1     1.0    0.0    0.0       vblendvps  %xmm1, (%rdi), %xmm2, %xmm3
+# ZNVER2-NEXT:         1     1.0    0.5    0.0       <total>

llvm/test/tools/llvm-mca/X86/variable-blend-read-after-ld-2.s

 # RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=skylake -iterations=1 -timeline -instruction-info=false -resource-pressure=false < %s | FileCheck %s -check-prefix=ALL -check-prefix=SKYLAKE
 
 # RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=bdver2 -iterations=1 -timeline -instruction-info=false -resource-pressure=false < %s | FileCheck %s -check-prefix=ALL -check-prefix=BDVER2
 
 # RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=btver2 -iterations=1 -timeline -instruction-info=false -resource-pressure=false < %s | FileCheck %s -check-prefix=ALL -check-prefix=BTVER2
 
 # RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=znver1 -iterations=1 -timeline -instruction-info=false -resource-pressure=false < %s | FileCheck %s -check-prefix=ALL -check-prefix=ZNVER1
 
+# RUN: llvm-mca -mtriple=x86_64-unknown-unknown -mcpu=znver2 -iterations=1 -timeline -instruction-info=false -resource-pressure=false < %s | FileCheck %s -check-prefix=ALL -check-prefix=ZNVER2
+
 vaddps %xmm0, %xmm0, %xmm2
 vblendvps %xmm1, (%rdi), %xmm2, %xmm3
 
 # ALL:          Iterations:        1
 # ALL-NEXT:     Instructions:      2
 
 # BDVER2-NEXT:  Total Cycles:      10
 # BDVER2-NEXT:  Total uOps:        2

 # SANDY-NEXT:   Total uOps:        4
 
 # SKYLAKE-NEXT: Total Cycles:      11
 # SKYLAKE-NEXT: Total uOps:        4
 
 # ZNVER1-NEXT:  Total Cycles:      11
 # ZNVER1-NEXT:  Total uOps:        2
 
+# ZNVER2-NEXT:  Total Cycles:      11
+# ZNVER2-NEXT:  Total uOps:        2
+
 # BDVER2:       Dispatch Width:    4
 # BDVER2-NEXT:  uOps Per Cycle:    0.20
 # BDVER2-NEXT:  IPC:               0.20
 # BDVER2-NEXT:  Block RThroughput: 2.0
 
 # BDWELL:       Dispatch Width:    4
 # BDWELL-NEXT:  uOps Per Cycle:    0.40
 # BDWELL-NEXT:  IPC:               0.20

 # SKYLAKE-NEXT: IPC:               0.18
 # SKYLAKE-NEXT: Block RThroughput: 0.7
 
 # ZNVER1:       Dispatch Width:    4
 # ZNVER1-NEXT:  uOps Per Cycle:    0.18
 # ZNVER1-NEXT:  IPC:               0.18
 # ZNVER1-NEXT:  Block RThroughput: 1.0
 
+# ZNVER2:       Dispatch Width:    4
+# ZNVER2-NEXT:  uOps Per Cycle:    0.18
+# ZNVER2-NEXT:  IPC:               0.18
+# ZNVER2-NEXT:  Block RThroughput: 1.0
+
 # BDVER2:       Timeline view:
 # BDVER2-NEXT:  Index     0123456789
 
 # BDWELL:       Timeline view:
 # BDWELL-NEXT:  Index     0123456789
 
 # BTVER2:       Timeline view:
 # BTVER2-NEXT:                      0

 # SKYLAKE:      Timeline view:
 # SKYLAKE-NEXT:                     0
 # SKYLAKE-NEXT: Index     0123456789
 
 # ZNVER1:       Timeline view:
 # ZNVER1-NEXT:                      0
 # ZNVER1-NEXT:  Index     0123456789
 
+# ZNVER2:       Timeline view:
+# ZNVER2-NEXT:                      0
+# ZNVER2-NEXT:  Index     0123456789
+
 # BDVER2:       [0,0]     DeeeeeER .   vaddps %xmm0, %xmm0, %xmm2
 # BDVER2-NEXT:  [0,1]     DeeeeeeeER   vblendvps  %xmm1, (%rdi), %xmm2, %xmm3
 
 # BDWELL:       [0,0]     DeeeER   .   vaddps %xmm0, %xmm0, %xmm2
 # BDWELL-NEXT:  [0,1]     DeeeeeeeER   vblendvps  %xmm1, (%rdi), %xmm2, %xmm3
 
 # BTVER2:       [0,0]     DeeeER    .   vaddps  %xmm0, %xmm0, %xmm2
 # BTVER2-NEXT:  [0,1]     .DeeeeeeeER   vblendvps %xmm1, (%rdi), %xmm2, %xmm3
 
 # HASWELL:      [0,0]     DeeeER    .   vaddps  %xmm0, %xmm0, %xmm2
 # HASWELL-NEXT: [0,1]     DeeeeeeeeER   vblendvps %xmm1, (%rdi), %xmm2, %xmm3
 
 # IVY:          [0,0]     DeeeER    .   vaddps  %xmm0, %xmm0, %xmm2
 # IVY-NEXT:     [0,1]     DeeeeeeeeER   vblendvps %xmm1, (%rdi), %xmm2, %xmm3
 
 # SANDY:        [0,0]     DeeeER    .   vaddps  %xmm0, %xmm0, %xmm2
 # SANDY-NEXT:   [0,1]     DeeeeeeeeER   vblendvps %xmm1, (%rdi), %xmm2, %xmm3
 
 # SKYLAKE:      [0,0]     DeeeeER   .   vaddps  %xmm0, %xmm0, %xmm2
 # SKYLAKE-NEXT: [0,1]     DeeeeeeeeER   vblendvps %xmm1, (%rdi), %xmm2, %xmm3
 
 # ZNVER1:       [0,0]     DeeeER    .   vaddps  %xmm0, %xmm0, %xmm2
 # ZNVER1-NEXT:  [0,1]     DeeeeeeeeER   vblendvps %xmm1, (%rdi), %xmm2, %xmm3
 
+# ZNVER2:       [0,0]     DeeeER    .   vaddps  %xmm0, %xmm0, %xmm2
+# ZNVER2-NEXT:  [0,1]     DeeeeeeeeER   vblendvps %xmm1, (%rdi), %xmm2, %xmm3
+
 # ALL:          Average Wait times (based on the timeline view):
 # ALL-NEXT:     [0]: Executions
 # ALL-NEXT:     [1]: Average time spent waiting in a scheduler's queue
 # ALL-NEXT:     [2]: Average time spent waiting in a scheduler's queue while ready
 # ALL-NEXT:     [3]: Average time elapsed from WB until retire stage
 
 # ALL:                [0]    [1]    [2]    [3]
 # ALL-NEXT:     0.     1     1.0    1.0    0.0       vaddps %xmm0, %xmm0, %xmm2

 # SANDY-NEXT:   1.     1     1.0    0.0    0.0       vblendvps  %xmm1, (%rdi), %xmm2, %xmm3
 # SANDY-NEXT:          1     1.0    0.5    0.0       <total>
 
 # SKYLAKE-NEXT: 1.     1     1.0    0.0    0.0       vblendvps  %xmm1, (%rdi), %xmm2, %xmm3
 # SKYLAKE-NEXT:        1     1.0    0.5    0.0       <total>
 
 # ZNVER1-NEXT:  1.     1     1.0    0.0    0.0       vblendvps  %xmm1, (%rdi), %xmm2, %xmm3
 # ZNVER1-NEXT:         1     1.0    0.5    0.0       <total>
+
+# ZNVER2-NEXT:  1.     1     1.0    0.0    0.0       vblendvps  %xmm1, (%rdi), %xmm2, %xmm3
+# ZNVER2-NEXT:         1     1.0    0.5    0.0       <total>
+