[llvm] [AArch64] Update zero latency instructions in Neoverse scheduling tables (PR #165690)
Simon Wallis via llvm-commits
llvm-commits at lists.llvm.org
Wed Nov 19 03:21:46 PST 2025
https://github.com/simonwallis2 updated https://github.com/llvm/llvm-project/pull/165690
>From 90fdc605033f2142f647d4aa2273210f52393d2c Mon Sep 17 00:00:00 2001
From: Simon Wallis <simon.wallis2 at arm.com>
Date: Thu, 30 Oct 2025 10:43:12 +0000
Subject: [PATCH 1/7] [AArch64] Update zero latency instructions in scheduling
tables for Neoverse cores
NeoverseZeroMove was introduced for Neoverse-V2 and was added to V3 and V3AE.
Use NeoverseZeroMove for Neoverse-V1, N2, N3 in the same way, including these instructions:
MOV Xd|Wd, #0|XZR|WZR
For all Neoverse targets, the following instructions are also decoded as not utilizing the scheduling and execution resources of the machine:
MOV Wd,Wn
MOV Xd,Xn
For Neoverse-N3 only, these instructions also have zero latency
FMOV Dd, Dn
FMOV Sd, Sn
Change-Id: I1a5f86e049798582d33d96ba99389e4b2ffb210e
---
.../Target/AArch64/AArch64SchedNeoverseN2.td | 30 ++++++++++++-
.../Target/AArch64/AArch64SchedNeoverseN3.td | 26 ++++++++++--
.../Target/AArch64/AArch64SchedNeoverseV1.td | 22 +++++++++-
.../Target/AArch64/AArch64SchedNeoverseV2.td | 5 ++-
.../Target/AArch64/AArch64SchedNeoverseV3.td | 5 ++-
.../AArch64/AArch64SchedNeoverseV3AE.td | 5 ++-
.../AArch64/Neoverse/N2-basic-instructions.s | 22 +++++-----
.../AArch64/Neoverse/N3-basic-instructions.s | 30 ++++++-------
.../AArch64/Neoverse/V1-basic-instructions.s | 22 +++++-----
.../AArch64/Neoverse/V1-zero-dependency.s | 42 +++++++++----------
.../AArch64/Neoverse/V2-basic-instructions.s | 10 ++---
.../AArch64/Neoverse/V2-zero-lat-movs.s | 24 +++++------
.../AArch64/Neoverse/V3-basic-instructions.s | 10 ++---
.../AArch64/Neoverse/V3-zero-lat-movs.s | 24 +++++------
.../Neoverse/V3AE-basic-instructions.s | 10 ++---
.../AArch64/Neoverse/V3AE-zero-lat-movs.s | 24 +++++------
16 files changed, 192 insertions(+), 119 deletions(-)
diff --git a/llvm/lib/Target/AArch64/AArch64SchedNeoverseN2.td b/llvm/lib/Target/AArch64/AArch64SchedNeoverseN2.td
index 50f10114989d0..d1ce5a13d0510 100644
--- a/llvm/lib/Target/AArch64/AArch64SchedNeoverseN2.td
+++ b/llvm/lib/Target/AArch64/AArch64SchedNeoverseN2.td
@@ -72,6 +72,13 @@ def : WriteRes<WriteLDHi, []> { let Latency = 4; }
// Define customized scheduler read/write types specific to the Neoverse N2.
//===----------------------------------------------------------------------===//
+
+// Define generic 0 micro-op types
+def N2Write_0c : SchedWriteRes<[]> {
+ let Latency = 0;
+ let NumMicroOps = 0;
+}
+
// Define generic 1 micro-op types
def N2Write_1c_1B : SchedWriteRes<[N2UnitB]> { let Latency = 1; }
@@ -645,6 +652,21 @@ def N2Write_11c_9L01_9S_9V : SchedWriteRes<[N2UnitL01, N2UnitL01, N2UnitL01,
let NumMicroOps = 27;
}
+//===----------------------------------------------------------------------===//
+// Define predicate-controlled types
+
+def N2Write_0or1c_1I : SchedWriteVariant<[
+ SchedVar<NeoverseZeroMove, [N2Write_0c]>,
+ SchedVar<NoSchedPred, [N2Write_1c_1I]>]>;
+
+def N2Write_0or2c_1V : SchedWriteVariant<[
+ SchedVar<NeoverseZeroMove, [N2Write_0c]>,
+ SchedVar<NoSchedPred, [N2Write_2c_1V]>]>;
+
+def N2Write_0or3c_1M0 : SchedWriteVariant<[
+ SchedVar<NeoverseZeroMove, [N2Write_0c]>,
+ SchedVar<NoSchedPred, [N2Write_3c_1M0]>]>;
+
//===----------------------------------------------------------------------===//
// Define types for arithmetic and logical ops with short shifts
def N2Write_Arith : SchedWriteVariant<[
@@ -680,6 +702,7 @@ def : InstRW<[N2Write_1c_1B_1S], (instrs BL, BLR)>;
// ALU, basic
// ALU, basic, flagset
def : SchedAlias<WriteI, N2Write_1c_1I>;
+def : InstRW<[N2Write_0or1c_1I], (instregex "^MOVZ[WX]i$")>;
// ALU, extend and shift
def : SchedAlias<WriteIEReg, N2Write_2c_1M>;
@@ -691,7 +714,8 @@ def : SchedAlias<WriteISReg, N2Write_Arith>;
// Logical, shift, no flagset
def : InstRW<[N2Write_1c_1I],
- (instregex "^(AND|BIC|EON|EOR|ORN|ORR)[WX]rs$")>;
+ (instregex "^(AND|BIC|EON|EOR|ORN)[WX]rs$")>;
+def : InstRW<[N2Write_0or1c_1I], (instregex "^ORR[WX]rs$")>;
// Logical, shift, flagset
def : InstRW<[N2Write_Logical], (instregex "^(AND|BIC)S[WX]rs$")>;
@@ -882,7 +906,7 @@ def : SchedAlias<WriteFImm, N2Write_2c_1V>;
def : InstRW<[N2Write_2c_1V], (instrs FMOVHr, FMOVSr, FMOVDr)>;
// FP transfer, from gen to low half of vec reg
-def : InstRW<[N2Write_3c_1M0], (instrs FMOVWHr, FMOVXHr, FMOVWSr, FMOVXDr,
+def : InstRW<[N2Write_0or3c_1M0], (instrs FMOVWHr, FMOVXHr, FMOVWSr, FMOVXDr,
FMOVHWr, FMOVHXr, FMOVSWr, FMOVDXr)>;
// FP transfer, from gen to high half of vec reg
@@ -1225,6 +1249,8 @@ def : InstRW<[N2Write_3c_1V0], (instrs BFCVT)>;
// ASIMD unzip/zip
// Handled by SchedAlias<WriteV[dq], ...>
+def : InstRW<[N2Write_0or2c_1V], (instrs MOVID, MOVIv2d_ns)>;
+
// ASIMD duplicate, gen reg
def : InstRW<[N2Write_3c_1M0], (instregex "^DUPv.+gpr")>;
diff --git a/llvm/lib/Target/AArch64/AArch64SchedNeoverseN3.td b/llvm/lib/Target/AArch64/AArch64SchedNeoverseN3.td
index 411b372a3f533..32d48ca66ee2d 100644
--- a/llvm/lib/Target/AArch64/AArch64SchedNeoverseN3.td
+++ b/llvm/lib/Target/AArch64/AArch64SchedNeoverseN3.td
@@ -553,6 +553,22 @@ def N3Write_16c_16V0 : SchedWriteRes<[N3UnitV0, N3UnitV0, N3UnitV0, N3UnitV0,
let NumMicroOps = 16;
}
+
+//===----------------------------------------------------------------------===//
+// Define predicate-controlled types
+
+def N3Write_0or1c_1I : SchedWriteVariant<[
+ SchedVar<NeoverseZeroMove, [N3Write_0c]>,
+ SchedVar<NoSchedPred, [N3Write_1c_1I]>]>;
+
+def N3Write_0or2c_1V : SchedWriteVariant<[
+ SchedVar<NeoverseZeroMove, [N3Write_0c]>,
+ SchedVar<NoSchedPred, [N3Write_2c_1V]>]>;
+
+def N3Write_0or3c_1M0 : SchedWriteVariant<[
+ SchedVar<NeoverseZeroMove, [N3Write_0c]>,
+ SchedVar<NoSchedPred, [N3Write_3c_1M0]>]>;
+
// Miscellaneous
// -----------------------------------------------------------------------------
@@ -581,6 +597,7 @@ def : InstRW<[N3Write_1c_1B_1S], (instrs BL, BLR)>;
// Conditional compare
// Conditional select
def : SchedAlias<WriteI, N3Write_1c_1I>;
+def : InstRW<[N3Write_0or1c_1I], (instregex "^MOVZ[WX]i$")>;
// ALU, extend and shift
def : SchedAlias<WriteIEReg, N3Write_2c_1M>;
@@ -610,7 +627,8 @@ def : InstRW<[N3Write_1c_1I], (instrs GMI, SUBP, SUBPS)>;
// Logical, shift, no flagset
def : InstRW<[N3Write_1c_1I],
- (instregex "^(AND|BIC|EON|EOR|ORN|ORR)[WX]rs$")>;
+ (instregex "^(AND|BIC|EON|EOR|ORN)[WX]rs$")>;
+def : InstRW<[N3Write_0or1c_1I], (instregex "^ORR[WX]rs$")>;
// Logical, shift, flagset
def : InstRW<[N3Write_2c_1M], (instregex "^(AND|BIC)S[WX]rs$")>;
@@ -855,10 +873,11 @@ def : SchedAlias<WriteFCvt, N3Write_3c_1V0>;
def : SchedAlias<WriteFImm, N3Write_2c_1V>;
// FP move, register
-def : InstRW<[N3Write_2c_1V], (instrs FMOVHr, FMOVSr, FMOVDr)>;
+def : InstRW<[N3Write_2c_1V], (instrs FMOVHr)>;
+def : InstRW<[N3Write_0c], (instrs FMOVSr, FMOVDr)>;
// FP transfer, from gen to low half of vec reg
-def : InstRW<[N3Write_3c_1M0], (instrs FMOVWHr, FMOVXHr, FMOVWSr, FMOVXDr)>;
+def : InstRW<[N3Write_0or3c_1M0], (instrs FMOVWHr, FMOVXHr, FMOVWSr, FMOVXDr)>;
// FP transfer, from gen to high half of vec reg
def : InstRW<[N3Write_5c_1M0_1V], (instrs FMOVXDHighr)>;
@@ -1186,6 +1205,7 @@ def : InstRW<[N3Write_3c_1V0], (instrs BFCVT)>;
// ASIMD transpose
// ASIMD unzip/zip
// Covered by WriteV[dq]
+def : InstRW<[N3Write_0or2c_1V], (instrs MOVID, MOVIv2d_ns)>;
// ASIMD duplicate, gen reg
def : InstRW<[N3Write_3c_1M0], (instregex "^DUPv.+gpr")>;
diff --git a/llvm/lib/Target/AArch64/AArch64SchedNeoverseV1.td b/llvm/lib/Target/AArch64/AArch64SchedNeoverseV1.td
index 3cbfc59423c9a..8d33ca22616c2 100644
--- a/llvm/lib/Target/AArch64/AArch64SchedNeoverseV1.td
+++ b/llvm/lib/Target/AArch64/AArch64SchedNeoverseV1.td
@@ -472,6 +472,21 @@ def V1Write_11c_9L01_9S_9V : SchedWriteRes<[V1UnitL01, V1UnitL01, V1UnitL01,
V1UnitV, V1UnitV, V1UnitV,
V1UnitV, V1UnitV, V1UnitV]>;
+//===----------------------------------------------------------------------===//
+// Define predicate-controlled types
+
+def V1Write_0or1c_1I : SchedWriteVariant<[
+ SchedVar<NeoverseZeroMove, [V1Write_0c_0Z]>,
+ SchedVar<NoSchedPred, [V1Write_1c_1I]>]>;
+
+def V1Write_0or2c_1V : SchedWriteVariant<[
+ SchedVar<NeoverseZeroMove, [V1Write_0c_0Z]>,
+ SchedVar<NoSchedPred, [V1Write_2c_1V]>]>;
+
+def V1Write_0or3c_1M0 : SchedWriteVariant<[
+ SchedVar<NeoverseZeroMove, [V1Write_0c_0Z]>,
+ SchedVar<NoSchedPred, [V1Write_3c_1M0]>]>;
+
//===----------------------------------------------------------------------===//
// Define forwarded types
@@ -603,6 +618,7 @@ def : InstRW<[V1Write_1c_1I_1Flg],
"^(ADC|SBC)S[WX]r$",
"^ANDS[WX]ri$",
"^(AND|BIC)S[WX]rr$")>;
+def : InstRW<[V1Write_0or1c_1I], (instregex "^MOVZ[WX]i$")>;
// ALU, extend and shift
def : SchedAlias<WriteIEReg, V1Write_2c_1M>;
@@ -623,7 +639,8 @@ def : InstRW<[V1WriteISRegS],
(instregex "^(ADD|SUB)S(([WX]r[sx])|Xrx64)$")>;
// Logical, shift, no flagset
-def : InstRW<[V1Write_1c_1I], (instregex "^(AND|BIC|EON|EOR|ORN|ORR)[WX]rs$")>;
+def : InstRW<[V1Write_1c_1I], (instregex "^(AND|BIC|EON|EOR|ORN)[WX]rs$")>;
+def : InstRW<[V1Write_0or1c_1I], (instregex "^ORR[WX]rs$")>;
// Logical, shift, flagset
def : InstRW<[V1Write_2c_1M_1Flg], (instregex "^(AND|BIC)S[WX]rs$")>;
@@ -805,7 +822,7 @@ def : SchedAlias<WriteFImm, V1Write_2c_1V>;
def : InstRW<[V1Write_2c_1V], (instrs FMOVHr, FMOVSr, FMOVDr)>;
// FP transfer, from gen to low half of vec reg
-def : InstRW<[V1Write_3c_1M0], (instrs FMOVWHr, FMOVXHr, FMOVWSr, FMOVXDr)>;
+def : InstRW<[V1Write_0or3c_1M0], (instrs FMOVWHr, FMOVXHr, FMOVWSr, FMOVXDr)>;
// FP transfer, from gen to high half of vec reg
def : InstRW<[V1Write_5c_1M0_1V], (instrs FMOVXDHighr)>;
@@ -1122,6 +1139,7 @@ def : InstRW<[V1Write_3c_1V02], (instrs BFCVT)>;
// ASIMD transpose
// ASIMD unzip/zip
// Covered by "SchedAlias (WriteV[dq]...)" above
+def : InstRW<[V1Write_0or2c_1V], (instrs MOVID, MOVIv2d_ns)>;
// ASIMD duplicate, gen reg
def : InstRW<[V1Write_3c_1M0],
diff --git a/llvm/lib/Target/AArch64/AArch64SchedNeoverseV2.td b/llvm/lib/Target/AArch64/AArch64SchedNeoverseV2.td
index 2387f176f3051..1ef087f07022d 100644
--- a/llvm/lib/Target/AArch64/AArch64SchedNeoverseV2.td
+++ b/llvm/lib/Target/AArch64/AArch64SchedNeoverseV2.td
@@ -94,7 +94,10 @@ def : WriteRes<WriteLDHi, []> { let Latency = 4; }
//===----------------------------------------------------------------------===//
// Define generic 0 micro-op types
-def V2Write_0c : SchedWriteRes<[]> { let Latency = 0; }
+def V2Write_0c : SchedWriteRes<[]> {
+ let Latency = 0;
+ let NumMicroOps = 0;
+}
// Define generic 1 micro-op types
diff --git a/llvm/lib/Target/AArch64/AArch64SchedNeoverseV3.td b/llvm/lib/Target/AArch64/AArch64SchedNeoverseV3.td
index e23576a20d277..3dd2988088f0b 100644
--- a/llvm/lib/Target/AArch64/AArch64SchedNeoverseV3.td
+++ b/llvm/lib/Target/AArch64/AArch64SchedNeoverseV3.td
@@ -94,7 +94,10 @@ def : WriteRes<WriteLDHi, []> { let Latency = 4; }
//===----------------------------------------------------------------------===//
// Define generic 0 micro-op types
-def V3Write_0c : SchedWriteRes<[]> { let Latency = 0; }
+def V3Write_0c : SchedWriteRes<[]> {
+ let Latency = 0;
+ let NumMicroOps = 0;
+}
// Define generic 1 micro-op types
diff --git a/llvm/lib/Target/AArch64/AArch64SchedNeoverseV3AE.td b/llvm/lib/Target/AArch64/AArch64SchedNeoverseV3AE.td
index 0f1ec669a4e5e..19b56260387e1 100644
--- a/llvm/lib/Target/AArch64/AArch64SchedNeoverseV3AE.td
+++ b/llvm/lib/Target/AArch64/AArch64SchedNeoverseV3AE.td
@@ -89,7 +89,10 @@ def : WriteRes<WriteLDHi, []> { let Latency = 4; }
//===----------------------------------------------------------------------===//
// Define generic 0 micro-op types
-def V3AEWrite_0c : SchedWriteRes<[]> { let Latency = 0; }
+def V3AEWrite_0c : SchedWriteRes<[]> {
+ let Latency = 0;
+ let NumMicroOps = 0;
+}
// Define generic 1 micro-op types
diff --git a/llvm/test/tools/llvm-mca/AArch64/Neoverse/N2-basic-instructions.s b/llvm/test/tools/llvm-mca/AArch64/Neoverse/N2-basic-instructions.s
index cf1cf0e98c801..d3343ab055887 100644
--- a/llvm/test/tools/llvm-mca/AArch64/Neoverse/N2-basic-instructions.s
+++ b/llvm/test/tools/llvm-mca/AArch64/Neoverse/N2-basic-instructions.s
@@ -2508,14 +2508,14 @@ drps
# CHECK-NEXT: 1 2 0.50 bics x3, xzr, x3, lsl #1
# CHECK-NEXT: 1 2 0.50 tst w3, w7, lsl #31
# CHECK-NEXT: 1 2 0.50 tst x2, x20, asr #2
-# CHECK-NEXT: 1 1 0.25 mov x3, x6
-# CHECK-NEXT: 1 1 0.25 mov x3, xzr
-# CHECK-NEXT: 1 1 0.25 mov wzr, w2
-# CHECK-NEXT: 1 1 0.25 mov w3, w5
+# CHECK-NEXT: 0 0 0.00 mov x3, x6
+# CHECK-NEXT: 0 0 0.00 mov x3, xzr
+# CHECK-NEXT: 0 0 0.00 mov wzr, w2
+# CHECK-NEXT: 0 0 0.00 mov w3, w5
# CHECK-NEXT: 1 1 0.25 movz w2, #0, lsl #16
# CHECK-NEXT: 1 1 0.25 mov w2, #-1235
# CHECK-NEXT: 1 1 0.25 mov x2, #5299989643264
-# CHECK-NEXT: 1 1 0.25 mov x2, #0
+# CHECK-NEXT: 0 0 0.00 mov x2, #0
# CHECK-NEXT: 1 1 0.25 movk w3, #0
# CHECK-NEXT: 1 1 0.25 movz x4, #0, lsl #16
# CHECK-NEXT: 1 1 0.25 movk w5, #0, lsl #16
@@ -2557,7 +2557,7 @@ drps
# CHECK: Resource pressure per iteration:
# CHECK-NEXT: [0.0] [0.1] [1.0] [1.1] [2] [3.0] [3.1] [4] [5] [6.0] [6.1] [7] [8]
-# CHECK-NEXT: 11.00 11.00 33.00 33.00 87.33 151.33 151.33 517.00 251.00 162.50 162.50 215.50 85.50
+# CHECK-NEXT: 11.00 11.00 33.00 33.00 87.33 151.33 151.33 515.75 249.75 161.25 161.25 215.50 85.50
# CHECK: Resource pressure by instruction:
# CHECK-NEXT: [0.0] [0.1] [1.0] [1.1] [2] [3.0] [3.1] [4] [5] [6.0] [6.1] [7] [8] Instructions:
@@ -3692,14 +3692,14 @@ drps
# CHECK-NEXT: - - - - - - - 0.50 0.50 - - - - bics x3, xzr, x3, lsl #1
# CHECK-NEXT: - - - - - - - 0.50 0.50 - - - - tst w3, w7, lsl #31
# CHECK-NEXT: - - - - - - - 0.50 0.50 - - - - tst x2, x20, asr #2
-# CHECK-NEXT: - - - - - - - 0.25 0.25 0.25 0.25 - - mov x3, x6
-# CHECK-NEXT: - - - - - - - 0.25 0.25 0.25 0.25 - - mov x3, xzr
-# CHECK-NEXT: - - - - - - - 0.25 0.25 0.25 0.25 - - mov wzr, w2
-# CHECK-NEXT: - - - - - - - 0.25 0.25 0.25 0.25 - - mov w3, w5
+# CHECK-NEXT: - - - - - - - - - - - - - mov x3, x6
+# CHECK-NEXT: - - - - - - - - - - - - - mov x3, xzr
+# CHECK-NEXT: - - - - - - - - - - - - - mov wzr, w2
+# CHECK-NEXT: - - - - - - - - - - - - - mov w3, w5
# CHECK-NEXT: - - - - - - - 0.25 0.25 0.25 0.25 - - movz w2, #0, lsl #16
# CHECK-NEXT: - - - - - - - 0.25 0.25 0.25 0.25 - - mov w2, #-1235
# CHECK-NEXT: - - - - - - - 0.25 0.25 0.25 0.25 - - mov x2, #5299989643264
-# CHECK-NEXT: - - - - - - - 0.25 0.25 0.25 0.25 - - mov x2, #0
+# CHECK-NEXT: - - - - - - - - - - - - - mov x2, #0
# CHECK-NEXT: - - - - - - - 0.25 0.25 0.25 0.25 - - movk w3, #0
# CHECK-NEXT: - - - - - - - 0.25 0.25 0.25 0.25 - - movz x4, #0, lsl #16
# CHECK-NEXT: - - - - - - - 0.25 0.25 0.25 0.25 - - movk w5, #0, lsl #16
diff --git a/llvm/test/tools/llvm-mca/AArch64/Neoverse/N3-basic-instructions.s b/llvm/test/tools/llvm-mca/AArch64/Neoverse/N3-basic-instructions.s
index b9758280e2491..f7311b5e41b2e 100644
--- a/llvm/test/tools/llvm-mca/AArch64/Neoverse/N3-basic-instructions.s
+++ b/llvm/test/tools/llvm-mca/AArch64/Neoverse/N3-basic-instructions.s
@@ -1888,7 +1888,7 @@ drps
# CHECK-NEXT: 1 2 0.50 fccmpe d31, d5, #7, ne
# CHECK-NEXT: 1 2 0.50 fcsel s3, s20, s9, pl
# CHECK-NEXT: 1 2 0.50 fcsel d9, d10, d11, mi
-# CHECK-NEXT: 1 2 0.50 fmov s0, s1
+# CHECK-NEXT: 0 0 0.00 fmov s0, s1
# CHECK-NEXT: 1 2 0.50 fabs s2, s3
# CHECK-NEXT: 1 2 0.50 fneg s4, s5
# CHECK-NEXT: 1 7 1.00 fsqrt s6, s7
@@ -1901,7 +1901,7 @@ drps
# CHECK-NEXT: 1 3 1.00 frinta s20, s21
# CHECK-NEXT: 1 3 1.00 frintx s22, s23
# CHECK-NEXT: 1 3 1.00 frinti s24, s25
-# CHECK-NEXT: 1 2 0.50 fmov d0, d1
+# CHECK-NEXT: 0 0 0.00 fmov d0, d1
# CHECK-NEXT: 1 2 0.50 fabs d2, d3
# CHECK-NEXT: 1 2 0.50 fneg d4, d5
# CHECK-NEXT: 1 12 1.00 fsqrt d6, d7
@@ -2508,14 +2508,14 @@ drps
# CHECK-NEXT: 1 2 0.50 bics x3, xzr, x3, lsl #1
# CHECK-NEXT: 1 2 0.50 tst w3, w7, lsl #31
# CHECK-NEXT: 1 2 0.50 tst x2, x20, asr #2
-# CHECK-NEXT: 1 1 0.25 mov x3, x6
-# CHECK-NEXT: 1 1 0.25 mov x3, xzr
-# CHECK-NEXT: 1 1 0.25 mov wzr, w2
-# CHECK-NEXT: 1 1 0.25 mov w3, w5
+# CHECK-NEXT: 0 0 0.00 mov x3, x6
+# CHECK-NEXT: 0 0 0.00 mov x3, xzr
+# CHECK-NEXT: 0 0 0.00 mov wzr, w2
+# CHECK-NEXT: 0 0 0.00 mov w3, w5
# CHECK-NEXT: 1 1 0.25 movz w2, #0, lsl #16
# CHECK-NEXT: 1 1 0.25 mov w2, #-1235
# CHECK-NEXT: 1 1 0.25 mov x2, #5299989643264
-# CHECK-NEXT: 1 1 0.25 mov x2, #0
+# CHECK-NEXT: 0 0 0.00 mov x2, #0
# CHECK-NEXT: 1 1 0.25 movk w3, #0
# CHECK-NEXT: 1 1 0.25 movz x4, #0, lsl #16
# CHECK-NEXT: 1 1 0.25 movk w5, #0, lsl #16
@@ -2557,7 +2557,7 @@ drps
# CHECK: Resource pressure per iteration:
# CHECK-NEXT: [0.0] [0.1] [1.0] [1.1] [2] [3.0] [3.1] [4] [5] [6.0] [6.1] [7] [8]
-# CHECK-NEXT: 11.00 11.00 33.00 33.00 99.33 163.33 163.33 357.75 212.75 156.25 156.25 184.50 64.50
+# CHECK-NEXT: 11.00 11.00 33.00 33.00 99.33 163.33 163.33 356.50 211.50 155.00 155.00 183.50 63.50
# CHECK: Resource pressure by instruction:
# CHECK-NEXT: [0.0] [0.1] [1.0] [1.1] [2] [3.0] [3.1] [4] [5] [6.0] [6.1] [7] [8] Instructions:
@@ -3072,7 +3072,7 @@ drps
# CHECK-NEXT: - - - - - - - - - - - 0.50 0.50 fccmpe d31, d5, #7, ne
# CHECK-NEXT: - - - - - - - - - - - 0.50 0.50 fcsel s3, s20, s9, pl
# CHECK-NEXT: - - - - - - - - - - - 0.50 0.50 fcsel d9, d10, d11, mi
-# CHECK-NEXT: - - - - - - - - - - - 0.50 0.50 fmov s0, s1
+# CHECK-NEXT: - - - - - - - - - - - - - fmov s0, s1
# CHECK-NEXT: - - - - - - - - - - - 0.50 0.50 fabs s2, s3
# CHECK-NEXT: - - - - - - - - - - - 0.50 0.50 fneg s4, s5
# CHECK-NEXT: - - - - - - - - - - - 1.00 - fsqrt s6, s7
@@ -3085,7 +3085,7 @@ drps
# CHECK-NEXT: - - - - - - - - - - - 1.00 - frinta s20, s21
# CHECK-NEXT: - - - - - - - - - - - 1.00 - frintx s22, s23
# CHECK-NEXT: - - - - - - - - - - - 1.00 - frinti s24, s25
-# CHECK-NEXT: - - - - - - - - - - - 0.50 0.50 fmov d0, d1
+# CHECK-NEXT: - - - - - - - - - - - - - fmov d0, d1
# CHECK-NEXT: - - - - - - - - - - - 0.50 0.50 fabs d2, d3
# CHECK-NEXT: - - - - - - - - - - - 0.50 0.50 fneg d4, d5
# CHECK-NEXT: - - - - - - - - - - - 1.00 - fsqrt d6, d7
@@ -3692,14 +3692,14 @@ drps
# CHECK-NEXT: - - - - - - - 0.50 0.50 - - - - bics x3, xzr, x3, lsl #1
# CHECK-NEXT: - - - - - - - 0.50 0.50 - - - - tst w3, w7, lsl #31
# CHECK-NEXT: - - - - - - - 0.50 0.50 - - - - tst x2, x20, asr #2
-# CHECK-NEXT: - - - - - - - 0.25 0.25 0.25 0.25 - - mov x3, x6
-# CHECK-NEXT: - - - - - - - 0.25 0.25 0.25 0.25 - - mov x3, xzr
-# CHECK-NEXT: - - - - - - - 0.25 0.25 0.25 0.25 - - mov wzr, w2
-# CHECK-NEXT: - - - - - - - 0.25 0.25 0.25 0.25 - - mov w3, w5
+# CHECK-NEXT: - - - - - - - - - - - - - mov x3, x6
+# CHECK-NEXT: - - - - - - - - - - - - - mov x3, xzr
+# CHECK-NEXT: - - - - - - - - - - - - - mov wzr, w2
+# CHECK-NEXT: - - - - - - - - - - - - - mov w3, w5
# CHECK-NEXT: - - - - - - - 0.25 0.25 0.25 0.25 - - movz w2, #0, lsl #16
# CHECK-NEXT: - - - - - - - 0.25 0.25 0.25 0.25 - - mov w2, #-1235
# CHECK-NEXT: - - - - - - - 0.25 0.25 0.25 0.25 - - mov x2, #5299989643264
-# CHECK-NEXT: - - - - - - - 0.25 0.25 0.25 0.25 - - mov x2, #0
+# CHECK-NEXT: - - - - - - - - - - - - - mov x2, #0
# CHECK-NEXT: - - - - - - - 0.25 0.25 0.25 0.25 - - movk w3, #0
# CHECK-NEXT: - - - - - - - 0.25 0.25 0.25 0.25 - - movz x4, #0, lsl #16
# CHECK-NEXT: - - - - - - - 0.25 0.25 0.25 0.25 - - movk w5, #0, lsl #16
diff --git a/llvm/test/tools/llvm-mca/AArch64/Neoverse/V1-basic-instructions.s b/llvm/test/tools/llvm-mca/AArch64/Neoverse/V1-basic-instructions.s
index eddc3e565c353..f75222f27a94a 100644
--- a/llvm/test/tools/llvm-mca/AArch64/Neoverse/V1-basic-instructions.s
+++ b/llvm/test/tools/llvm-mca/AArch64/Neoverse/V1-basic-instructions.s
@@ -2673,14 +2673,14 @@ drps
# CHECK-NEXT: 1 2 0.50 bics x3, xzr, x3, lsl #1
# CHECK-NEXT: 1 2 0.50 tst w3, w7, lsl #31
# CHECK-NEXT: 1 2 0.50 tst x2, x20, asr #2
-# CHECK-NEXT: 1 1 0.25 mov x3, x6
-# CHECK-NEXT: 1 1 0.25 mov x3, xzr
-# CHECK-NEXT: 1 1 0.25 mov wzr, w2
-# CHECK-NEXT: 1 1 0.25 mov w3, w5
+# CHECK-NEXT: 0 0 0.00 mov x3, x6
+# CHECK-NEXT: 0 0 0.00 mov x3, xzr
+# CHECK-NEXT: 0 0 0.00 mov wzr, w2
+# CHECK-NEXT: 0 0 0.00 mov w3, w5
# CHECK-NEXT: 1 1 0.25 movz w2, #0, lsl #16
# CHECK-NEXT: 1 1 0.25 mov w2, #-1235
# CHECK-NEXT: 1 1 0.25 mov x2, #5299989643264
-# CHECK-NEXT: 1 1 0.25 mov x2, #0
+# CHECK-NEXT: 0 0 0.00 mov x2, #0
# CHECK-NEXT: 1 1 0.25 movk w3, #0
# CHECK-NEXT: 1 1 0.25 movz x4, #0, lsl #16
# CHECK-NEXT: 1 1 0.25 movk w5, #0, lsl #16
@@ -2731,7 +2731,7 @@ drps
# CHECK: Resource pressure per iteration:
# CHECK-NEXT: [0.0] [0.1] [1.0] [1.1] [2.0] [2.1] [2.2] [3] [4.0] [4.1] [5] [6] [7.0] [7.1] [8] [9] [10] [11]
-# CHECK-NEXT: 13.00 13.00 40.50 40.50 48.00 48.00 48.00 96.67 175.17 175.17 322.50 209.50 142.00 142.00 189.00 55.50 65.50 13.00
+# CHECK-NEXT: 13.00 13.00 40.50 40.50 48.00 48.00 48.00 96.67 175.17 175.17 321.25 208.25 140.75 140.75 189.00 55.50 65.50 13.00
# CHECK: Resource pressure by instruction:
# CHECK-NEXT: [0.0] [0.1] [1.0] [1.1] [2.0] [2.1] [2.2] [3] [4.0] [4.1] [5] [6] [7.0] [7.1] [8] [9] [10] [11] Instructions:
@@ -3944,14 +3944,14 @@ drps
# CHECK-NEXT: - - - - 0.33 0.33 0.33 - - - 0.50 0.50 - - - - - - bics x3, xzr, x3, lsl #1
# CHECK-NEXT: - - - - 0.33 0.33 0.33 - - - 0.50 0.50 - - - - - - tst w3, w7, lsl #31
# CHECK-NEXT: - - - - 0.33 0.33 0.33 - - - 0.50 0.50 - - - - - - tst x2, x20, asr #2
-# CHECK-NEXT: - - - - - - - - - - 0.25 0.25 0.25 0.25 - - - - mov x3, x6
-# CHECK-NEXT: - - - - - - - - - - 0.25 0.25 0.25 0.25 - - - - mov x3, xzr
-# CHECK-NEXT: - - - - - - - - - - 0.25 0.25 0.25 0.25 - - - - mov wzr, w2
-# CHECK-NEXT: - - - - - - - - - - 0.25 0.25 0.25 0.25 - - - - mov w3, w5
+# CHECK-NEXT: - - - - - - - - - - - - - - - - - - mov x3, x6
+# CHECK-NEXT: - - - - - - - - - - - - - - - - - - mov x3, xzr
+# CHECK-NEXT: - - - - - - - - - - - - - - - - - - mov wzr, w2
+# CHECK-NEXT: - - - - - - - - - - - - - - - - - - mov w3, w5
# CHECK-NEXT: - - - - - - - - - - 0.25 0.25 0.25 0.25 - - - - movz w2, #0, lsl #16
# CHECK-NEXT: - - - - - - - - - - 0.25 0.25 0.25 0.25 - - - - mov w2, #-1235
# CHECK-NEXT: - - - - - - - - - - 0.25 0.25 0.25 0.25 - - - - mov x2, #5299989643264
-# CHECK-NEXT: - - - - - - - - - - 0.25 0.25 0.25 0.25 - - - - mov x2, #0
+# CHECK-NEXT: - - - - - - - - - - - - - - - - - - mov x2, #0
# CHECK-NEXT: - - - - - - - - - - 0.25 0.25 0.25 0.25 - - - - movk w3, #0
# CHECK-NEXT: - - - - - - - - - - 0.25 0.25 0.25 0.25 - - - - movz x4, #0, lsl #16
# CHECK-NEXT: - - - - - - - - - - 0.25 0.25 0.25 0.25 - - - - movk w5, #0, lsl #16
diff --git a/llvm/test/tools/llvm-mca/AArch64/Neoverse/V1-zero-dependency.s b/llvm/test/tools/llvm-mca/AArch64/Neoverse/V1-zero-dependency.s
index 3954cbd8c5490..7767b95ff98ea 100644
--- a/llvm/test/tools/llvm-mca/AArch64/Neoverse/V1-zero-dependency.s
+++ b/llvm/test/tools/llvm-mca/AArch64/Neoverse/V1-zero-dependency.s
@@ -6,13 +6,13 @@ cmp x0, #4
# CHECK: Iterations: 100
# CHECK-NEXT: Instructions: 200
-# CHECK-NEXT: Total Cycles: 54
-# CHECK-NEXT: Total uOps: 200
+# CHECK-NEXT: Total Cycles: 37
+# CHECK-NEXT: Total uOps: 100
# CHECK: Dispatch Width: 8
-# CHECK-NEXT: uOps Per Cycle: 3.70
-# CHECK-NEXT: IPC: 3.70
-# CHECK-NEXT: Block RThroughput: 0.5
+# CHECK-NEXT: uOps Per Cycle: 2.70
+# CHECK-NEXT: IPC: 5.41
+# CHECK-NEXT: Block RThroughput: 0.3
# CHECK: Instruction Info:
# CHECK-NEXT: [1]: #uOps
@@ -23,7 +23,7 @@ cmp x0, #4
# CHECK-NEXT: [6]: HasSideEffects (U)
# CHECK: [1] [2] [3] [4] [5] [6] Instructions:
-# CHECK-NEXT: 1 1 0.25 mov x0, x1
+# CHECK-NEXT: 0 0 0.00 mov x0, x1
# CHECK-NEXT: 1 1 0.33 cmp x0, #4
# CHECK: Resources:
@@ -48,24 +48,24 @@ cmp x0, #4
# CHECK: Resource pressure per iteration:
# CHECK-NEXT: [0.0] [0.1] [1.0] [1.1] [2.0] [2.1] [2.2] [3] [4.0] [4.1] [5] [6] [7.0] [7.1] [8] [9] [10] [11]
-# CHECK-NEXT: - - - - 0.33 0.33 0.34 - - - 0.50 0.50 0.50 0.50 - - - -
+# CHECK-NEXT: - - - - 0.33 0.33 0.34 - - - 0.22 0.22 0.28 0.28 - - - -
# CHECK: Resource pressure by instruction:
# CHECK-NEXT: [0.0] [0.1] [1.0] [1.1] [2.0] [2.1] [2.2] [3] [4.0] [4.1] [5] [6] [7.0] [7.1] [8] [9] [10] [11] Instructions:
-# CHECK-NEXT: - - - - - - - - - - 0.49 0.49 0.01 0.01 - - - - mov x0, x1
-# CHECK-NEXT: - - - - 0.33 0.33 0.34 - - - 0.01 0.01 0.49 0.49 - - - - cmp x0, #4
+# CHECK-NEXT: - - - - - - - - - - - - - - - - - - mov x0, x1
+# CHECK-NEXT: - - - - 0.33 0.33 0.34 - - - 0.22 0.22 0.28 0.28 - - - - cmp x0, #4
# CHECK: Timeline view:
-# CHECK-NEXT: Index 012345
+# CHECK-NEXT: Index 01234
-# CHECK: [0,0] DeER . mov x0, x1
-# CHECK-NEXT: [0,1] D=eER. cmp x0, #4
-# CHECK-NEXT: [1,0] DeE-R. mov x0, x1
-# CHECK-NEXT: [1,1] D=eER. cmp x0, #4
-# CHECK-NEXT: [2,0] DeE-R. mov x0, x1
-# CHECK-NEXT: [2,1] D=eER. cmp x0, #4
-# CHECK-NEXT: [3,0] DeE-R. mov x0, x1
-# CHECK-NEXT: [3,1] D==eER cmp x0, #4
+# CHECK: [0,0] DR . mov x0, x1
+# CHECK-NEXT: [0,1] DeER. cmp x0, #4
+# CHECK-NEXT: [1,0] D--R. mov x0, x1
+# CHECK-NEXT: [1,1] DeER. cmp x0, #4
+# CHECK-NEXT: [2,0] D--R. mov x0, x1
+# CHECK-NEXT: [2,1] DeER. cmp x0, #4
+# CHECK-NEXT: [3,0] D--R. mov x0, x1
+# CHECK-NEXT: [3,1] D=eER cmp x0, #4
# CHECK: Average Wait times (based on the timeline view):
# CHECK-NEXT: [0]: Executions
@@ -74,6 +74,6 @@ cmp x0, #4
# CHECK-NEXT: [3]: Average time elapsed from WB until retire stage
# CHECK: [0] [1] [2] [3]
-# CHECK-NEXT: 0. 4 1.0 1.0 0.8 mov x0, x1
-# CHECK-NEXT: 1. 4 2.3 0.3 0.0 cmp x0, #4
-# CHECK-NEXT: 4 1.6 0.6 0.4 <total>
+# CHECK-NEXT: 0. 4 0.0 0.0 1.5 mov x0, x1
+# CHECK-NEXT: 1. 4 1.3 1.3 0.0 cmp x0, #4
+# CHECK-NEXT: 4 0.6 0.6 0.8 <total>
diff --git a/llvm/test/tools/llvm-mca/AArch64/Neoverse/V2-basic-instructions.s b/llvm/test/tools/llvm-mca/AArch64/Neoverse/V2-basic-instructions.s
index 54b5f1644be48..9c987d54d2350 100644
--- a/llvm/test/tools/llvm-mca/AArch64/Neoverse/V2-basic-instructions.s
+++ b/llvm/test/tools/llvm-mca/AArch64/Neoverse/V2-basic-instructions.s
@@ -2536,14 +2536,14 @@ drps
# CHECK-NEXT: 1 2 0.50 bics x3, xzr, x3, lsl #1
# CHECK-NEXT: 1 2 0.50 tst w3, w7, lsl #31
# CHECK-NEXT: 1 2 0.50 tst x2, x20, asr #2
-# CHECK-NEXT: 1 0 0.17 mov x3, x6
-# CHECK-NEXT: 1 0 0.17 mov x3, xzr
-# CHECK-NEXT: 1 0 0.17 mov wzr, w2
-# CHECK-NEXT: 1 0 0.17 mov w3, w5
+# CHECK-NEXT: 0 0 0.00 mov x3, x6
+# CHECK-NEXT: 0 0 0.00 mov x3, xzr
+# CHECK-NEXT: 0 0 0.00 mov wzr, w2
+# CHECK-NEXT: 0 0 0.00 mov w3, w5
# CHECK-NEXT: 1 1 0.17 movz w2, #0, lsl #16
# CHECK-NEXT: 1 1 0.17 mov w2, #-1235
# CHECK-NEXT: 1 1 0.17 mov x2, #5299989643264
-# CHECK-NEXT: 1 0 0.17 mov x2, #0
+# CHECK-NEXT: 0 0 0.00 mov x2, #0
# CHECK-NEXT: 1 1 0.17 movk w3, #0
# CHECK-NEXT: 1 1 0.17 movz x4, #0, lsl #16
# CHECK-NEXT: 1 1 0.17 movk w5, #0, lsl #16
diff --git a/llvm/test/tools/llvm-mca/AArch64/Neoverse/V2-zero-lat-movs.s b/llvm/test/tools/llvm-mca/AArch64/Neoverse/V2-zero-lat-movs.s
index 3ddb525327015..1cec5897db425 100644
--- a/llvm/test/tools/llvm-mca/AArch64/Neoverse/V2-zero-lat-movs.s
+++ b/llvm/test/tools/llvm-mca/AArch64/Neoverse/V2-zero-lat-movs.s
@@ -23,18 +23,18 @@ mov x1, x2
# CHECK-NEXT: [6]: HasSideEffects (U)
# CHECK: [1] [2] [3] [4] [5] [6] Instructions:
-# CHECK-NEXT: 1 0 0.17 mov x1, #0
-# CHECK-NEXT: 1 0 0.17 mov x1, xzr
-# CHECK-NEXT: 1 0 0.17 mov w1, #0
-# CHECK-NEXT: 1 0 0.17 mov w1, wzr
-# CHECK-NEXT: 1 0 0.17 fmov h1, wzr
-# CHECK-NEXT: 1 0 0.17 fmov h1, xzr
-# CHECK-NEXT: 1 0 0.17 fmov s1, wzr
-# CHECK-NEXT: 1 0 0.17 fmov d1, xzr
-# CHECK-NEXT: 1 0 0.17 movi d1, #0000000000000000
-# CHECK-NEXT: 1 0 0.17 movi v1.2d, #0000000000000000
-# CHECK-NEXT: 1 0 0.17 mov w1, w2
-# CHECK-NEXT: 1 0 0.17 mov x1, x2
+# CHECK-NEXT: 0 0 0.00 mov x1, #0
+# CHECK-NEXT: 0 0 0.00 mov x1, xzr
+# CHECK-NEXT: 0 0 0.00 mov w1, #0
+# CHECK-NEXT: 0 0 0.00 mov w1, wzr
+# CHECK-NEXT: 0 0 0.00 fmov h1, wzr
+# CHECK-NEXT: 0 0 0.00 fmov h1, xzr
+# CHECK-NEXT: 0 0 0.00 fmov s1, wzr
+# CHECK-NEXT: 0 0 0.00 fmov d1, xzr
+# CHECK-NEXT: 0 0 0.00 movi d1, #0000000000000000
+# CHECK-NEXT: 0 0 0.00 movi v1.2d, #0000000000000000
+# CHECK-NEXT: 0 0 0.00 mov w1, w2
+# CHECK-NEXT: 0 0 0.00 mov x1, x2
# CHECK: Resources:
# CHECK-NEXT: [0.0] - V2UnitB
diff --git a/llvm/test/tools/llvm-mca/AArch64/Neoverse/V3-basic-instructions.s b/llvm/test/tools/llvm-mca/AArch64/Neoverse/V3-basic-instructions.s
index 73fd95d6e4a5b..67af391e52863 100644
--- a/llvm/test/tools/llvm-mca/AArch64/Neoverse/V3-basic-instructions.s
+++ b/llvm/test/tools/llvm-mca/AArch64/Neoverse/V3-basic-instructions.s
@@ -2536,14 +2536,14 @@ drps
# CHECK-NEXT: 1 2 0.50 bics x3, xzr, x3, lsl #1
# CHECK-NEXT: 1 2 0.50 tst w3, w7, lsl #31
# CHECK-NEXT: 1 2 0.50 tst x2, x20, asr #2
-# CHECK-NEXT: 1 0 0.10 mov x3, x6
-# CHECK-NEXT: 1 0 0.10 mov x3, xzr
-# CHECK-NEXT: 1 0 0.10 mov wzr, w2
-# CHECK-NEXT: 1 0 0.10 mov w3, w5
+# CHECK-NEXT: 0 0 0.00 mov x3, x6
+# CHECK-NEXT: 0 0 0.00 mov x3, xzr
+# CHECK-NEXT: 0 0 0.00 mov wzr, w2
+# CHECK-NEXT: 0 0 0.00 mov w3, w5
# CHECK-NEXT: 1 1 0.13 movz w2, #0, lsl #16
# CHECK-NEXT: 1 1 0.13 mov w2, #-1235
# CHECK-NEXT: 1 1 0.13 mov x2, #5299989643264
-# CHECK-NEXT: 1 0 0.10 mov x2, #0
+# CHECK-NEXT: 0 0 0.00 mov x2, #0
# CHECK-NEXT: 1 1 0.13 movk w3, #0
# CHECK-NEXT: 1 1 0.13 movz x4, #0, lsl #16
# CHECK-NEXT: 1 1 0.13 movk w5, #0, lsl #16
diff --git a/llvm/test/tools/llvm-mca/AArch64/Neoverse/V3-zero-lat-movs.s b/llvm/test/tools/llvm-mca/AArch64/Neoverse/V3-zero-lat-movs.s
index 1eef230b8174e..9b4834b12a79b 100644
--- a/llvm/test/tools/llvm-mca/AArch64/Neoverse/V3-zero-lat-movs.s
+++ b/llvm/test/tools/llvm-mca/AArch64/Neoverse/V3-zero-lat-movs.s
@@ -23,18 +23,18 @@ mov x1, x2
# CHECK-NEXT: [6]: HasSideEffects (U)
# CHECK: [1] [2] [3] [4] [5] [6] Instructions:
-# CHECK-NEXT: 1 0 0.10 mov x1, #0
-# CHECK-NEXT: 1 0 0.10 mov x1, xzr
-# CHECK-NEXT: 1 0 0.10 mov w1, #0
-# CHECK-NEXT: 1 0 0.10 mov w1, wzr
-# CHECK-NEXT: 1 0 0.10 fmov h1, wzr
-# CHECK-NEXT: 1 0 0.10 fmov h1, xzr
-# CHECK-NEXT: 1 0 0.10 fmov s1, wzr
-# CHECK-NEXT: 1 0 0.10 fmov d1, xzr
-# CHECK-NEXT: 1 0 0.10 movi d1, #0000000000000000
-# CHECK-NEXT: 1 0 0.10 movi v1.2d, #0000000000000000
-# CHECK-NEXT: 1 0 0.10 mov w1, w2
-# CHECK-NEXT: 1 0 0.10 mov x1, x2
+# CHECK-NEXT: 0 0 0.00 mov x1, #0
+# CHECK-NEXT: 0 0 0.00 mov x1, xzr
+# CHECK-NEXT: 0 0 0.00 mov w1, #0
+# CHECK-NEXT: 0 0 0.00 mov w1, wzr
+# CHECK-NEXT: 0 0 0.00 fmov h1, wzr
+# CHECK-NEXT: 0 0 0.00 fmov h1, xzr
+# CHECK-NEXT: 0 0 0.00 fmov s1, wzr
+# CHECK-NEXT: 0 0 0.00 fmov d1, xzr
+# CHECK-NEXT: 0 0 0.00 movi d1, #0000000000000000
+# CHECK-NEXT: 0 0 0.00 movi v1.2d, #0000000000000000
+# CHECK-NEXT: 0 0 0.00 mov w1, w2
+# CHECK-NEXT: 0 0 0.00 mov x1, x2
# CHECK: Resources:
# CHECK-NEXT: [0.0] - V3UnitB
diff --git a/llvm/test/tools/llvm-mca/AArch64/Neoverse/V3AE-basic-instructions.s b/llvm/test/tools/llvm-mca/AArch64/Neoverse/V3AE-basic-instructions.s
index 7ab2be5eaa365..5009ce1d54a86 100644
--- a/llvm/test/tools/llvm-mca/AArch64/Neoverse/V3AE-basic-instructions.s
+++ b/llvm/test/tools/llvm-mca/AArch64/Neoverse/V3AE-basic-instructions.s
@@ -2536,14 +2536,14 @@ drps
# CHECK-NEXT: 1 2 0.50 bics x3, xzr, x3, lsl #1
# CHECK-NEXT: 1 2 0.50 tst w3, w7, lsl #31
# CHECK-NEXT: 1 2 0.50 tst x2, x20, asr #2
-# CHECK-NEXT: 1 0 0.10 mov x3, x6
-# CHECK-NEXT: 1 0 0.10 mov x3, xzr
-# CHECK-NEXT: 1 0 0.10 mov wzr, w2
-# CHECK-NEXT: 1 0 0.10 mov w3, w5
+# CHECK-NEXT: 0 0 0.00 mov x3, x6
+# CHECK-NEXT: 0 0 0.00 mov x3, xzr
+# CHECK-NEXT: 0 0 0.00 mov wzr, w2
+# CHECK-NEXT: 0 0 0.00 mov w3, w5
# CHECK-NEXT: 1 1 0.13 movz w2, #0, lsl #16
# CHECK-NEXT: 1 1 0.13 mov w2, #-1235
# CHECK-NEXT: 1 1 0.13 mov x2, #5299989643264
-# CHECK-NEXT: 1 0 0.10 mov x2, #0
+# CHECK-NEXT: 0 0 0.00 mov x2, #0
# CHECK-NEXT: 1 1 0.13 movk w3, #0
# CHECK-NEXT: 1 1 0.13 movz x4, #0, lsl #16
# CHECK-NEXT: 1 1 0.13 movk w5, #0, lsl #16
diff --git a/llvm/test/tools/llvm-mca/AArch64/Neoverse/V3AE-zero-lat-movs.s b/llvm/test/tools/llvm-mca/AArch64/Neoverse/V3AE-zero-lat-movs.s
index a0840dcddcbab..783bea288b121 100644
--- a/llvm/test/tools/llvm-mca/AArch64/Neoverse/V3AE-zero-lat-movs.s
+++ b/llvm/test/tools/llvm-mca/AArch64/Neoverse/V3AE-zero-lat-movs.s
@@ -23,18 +23,18 @@ mov x1, x2
# CHECK-NEXT: [6]: HasSideEffects (U)
# CHECK: [1] [2] [3] [4] [5] [6] Instructions:
-# CHECK-NEXT: 1 0 0.10 mov x1, #0
-# CHECK-NEXT: 1 0 0.10 mov x1, xzr
-# CHECK-NEXT: 1 0 0.10 mov w1, #0
-# CHECK-NEXT: 1 0 0.10 mov w1, wzr
-# CHECK-NEXT: 1 0 0.10 fmov h1, wzr
-# CHECK-NEXT: 1 0 0.10 fmov h1, xzr
-# CHECK-NEXT: 1 0 0.10 fmov s1, wzr
-# CHECK-NEXT: 1 0 0.10 fmov d1, xzr
-# CHECK-NEXT: 1 0 0.10 movi d1, #0000000000000000
-# CHECK-NEXT: 1 0 0.10 movi v1.2d, #0000000000000000
-# CHECK-NEXT: 1 0 0.10 mov w1, w2
-# CHECK-NEXT: 1 0 0.10 mov x1, x2
+# CHECK-NEXT: 0 0 0.00 mov x1, #0
+# CHECK-NEXT: 0 0 0.00 mov x1, xzr
+# CHECK-NEXT: 0 0 0.00 mov w1, #0
+# CHECK-NEXT: 0 0 0.00 mov w1, wzr
+# CHECK-NEXT: 0 0 0.00 fmov h1, wzr
+# CHECK-NEXT: 0 0 0.00 fmov h1, xzr
+# CHECK-NEXT: 0 0 0.00 fmov s1, wzr
+# CHECK-NEXT: 0 0 0.00 fmov d1, xzr
+# CHECK-NEXT: 0 0 0.00 movi d1, #0000000000000000
+# CHECK-NEXT: 0 0 0.00 movi v1.2d, #0000000000000000
+# CHECK-NEXT: 0 0 0.00 mov w1, w2
+# CHECK-NEXT: 0 0 0.00 mov x1, x2
# CHECK: Resources:
# CHECK-NEXT: [0.0] - V3AEUnitB
>From 695b27d6540de00f9feb943744715758607380ff Mon Sep 17 00:00:00 2001
From: Simon Wallis <simon.wallis2 at arm.com>
Date: Thu, 30 Oct 2025 11:29:34 +0000
Subject: [PATCH 2/7] [AArch64] Update zero latency instructions in scheduling
tables for Neoverse cores
NeoverseZeroMove was introduced for Neoverse-V2 and was added to V3 and V3AE.
Use NeoverseZeroMove for Neoverse-V1, N2, N3 in the same way, including these instructions:
MOV Xd|Wd, #0|XZR|WZR
For all Neoverse targets, the following instructions are also decoded as not utilizing the scheduling and execution resources of the machine:
MOV Wd,Wn
MOV Xd,Xn
For Neoverse-N3 only, these instructions also have zero latency
FMOV Dd, Dn
FMOV Sd, Sn
Change-Id: I955cfe3efc689bea305a708eb6d7259dced6fe04
---
llvm/test/CodeGen/AArch64/pr164181.ll | 8 ++++----
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/llvm/test/CodeGen/AArch64/pr164181.ll b/llvm/test/CodeGen/AArch64/pr164181.ll
index 4ec63ecb2eeb4..987c92b084001 100644
--- a/llvm/test/CodeGen/AArch64/pr164181.ll
+++ b/llvm/test/CodeGen/AArch64/pr164181.ll
@@ -52,11 +52,11 @@ define void @f(i1 %var_0, i16 %var_1, i64 %var_2, i8 %var_3, i16 %var_4, i1 %var
; CHECK-NEXT: mov x10, xzr
; CHECK-NEXT: mov w23, wzr
; CHECK-NEXT: mov w30, wzr
-; CHECK-NEXT: ldrb w19, [sp, #240]
; CHECK-NEXT: mov w25, wzr
; CHECK-NEXT: mov x24, xzr
-; CHECK-NEXT: str w8, [sp, #108] // 4-byte Folded Spill
; CHECK-NEXT: mov x3, x26
+; CHECK-NEXT: str w8, [sp, #108] // 4-byte Folded Spill
+; CHECK-NEXT: ldrb w19, [sp, #240]
; CHECK-NEXT: ldp x9, x8, [sp, #344]
; CHECK-NEXT: str w12, [sp, #92] // 4-byte Folded Spill
; CHECK-NEXT: mov w12, #1 // =0x1
@@ -123,8 +123,8 @@ define void @f(i1 %var_0, i16 %var_1, i64 %var_2, i8 %var_3, i16 %var_4, i1 %var
; CHECK-NEXT: mov x12, #-30 // =0xffffffffffffffe2
; CHECK-NEXT: add x19, x4, w8, sxtw #2
; CHECK-NEXT: mov x9, xzr
-; CHECK-NEXT: csel x12, x24, x12, lo
; CHECK-NEXT: mov w4, w30
+; CHECK-NEXT: csel x12, x24, x12, lo
; CHECK-NEXT: str x12, [sp, #56] // 8-byte Folded Spill
; CHECK-NEXT: b .LBB0_8
; CHECK-NEXT: .p2align 5, , 16
@@ -341,8 +341,8 @@ define void @f(i1 %var_0, i16 %var_1, i64 %var_2, i8 %var_3, i16 %var_4, i1 %var
; CHECK-NEXT: mov x24, x27
; CHECK-NEXT: lsl x23, x14, #1
; CHECK-NEXT: mov x27, #-1 // =0xffffffffffffffff
-; CHECK-NEXT: madd x14, x14, x3, x11
; CHECK-NEXT: mov w28, w30
+; CHECK-NEXT: madd x14, x14, x3, x11
; CHECK-NEXT: mov w3, #-7680 // =0xffffe200
; CHECK-NEXT: b .LBB0_39
; CHECK-NEXT: .p2align 5, , 16
>From 9cebca2344bd9e41b51a1b6ad8ed44c0682d1fe4 Mon Sep 17 00:00:00 2001
From: Simon Wallis <simon.wallis2 at arm.com>
Date: Fri, 7 Nov 2025 09:30:55 +0000
Subject: [PATCH 3/7] [AArch64] Update zero latency instructions in Neoverse
scheduling tables
NeoverseZeroMove was introduced for Neoverse-V2 and was added to V3 and V3AE.
Use NeoverseZeroMove for Neoverse-V1, N2, N3 in the same way, including these instructions:
MOV Xd|Wd, #0|XZR|WZR
For all Neoverse targets, the following instructions are also decoded as not utilizing the scheduling and execution resources of the machine:
MOV Wd,Wn
MOV Xd,Xn
For Neoverse-N3 only, these instructions also have zero latency
FMOV Dd, Dn
FMOV Sd, Sn
Change-Id: Ie6b1f5c3f4d74f26bdd4c67c5e6c5acf6a8e00cc
---
llvm/lib/Target/AArch64/AArch64SchedNeoverseN2.td | 3 +--
.../llvm-mca/AArch64/Neoverse/N2-basic-instructions.s | 10 +++++-----
2 files changed, 6 insertions(+), 7 deletions(-)
diff --git a/llvm/lib/Target/AArch64/AArch64SchedNeoverseN2.td b/llvm/lib/Target/AArch64/AArch64SchedNeoverseN2.td
index d1ce5a13d0510..b4690e6045f85 100644
--- a/llvm/lib/Target/AArch64/AArch64SchedNeoverseN2.td
+++ b/llvm/lib/Target/AArch64/AArch64SchedNeoverseN2.td
@@ -906,8 +906,7 @@ def : SchedAlias<WriteFImm, N2Write_2c_1V>;
def : InstRW<[N2Write_2c_1V], (instrs FMOVHr, FMOVSr, FMOVDr)>;
// FP transfer, from gen to low half of vec reg
-def : InstRW<[N2Write_0or3c_1M0], (instrs FMOVWHr, FMOVXHr, FMOVWSr, FMOVXDr,
- FMOVHWr, FMOVHXr, FMOVSWr, FMOVDXr)>;
+def : InstRW<[N2Write_0or3c_1M0], (instrs FMOVWHr, FMOVXHr, FMOVWSr, FMOVXDr)>;
// FP transfer, from gen to high half of vec reg
def : InstRW<[N2Write_5c_1M0_1V], (instrs FMOVXDHighr)>;
diff --git a/llvm/test/tools/llvm-mca/AArch64/Neoverse/N2-basic-instructions.s b/llvm/test/tools/llvm-mca/AArch64/Neoverse/N2-basic-instructions.s
index d3343ab055887..b140e9a9927ff 100644
--- a/llvm/test/tools/llvm-mca/AArch64/Neoverse/N2-basic-instructions.s
+++ b/llvm/test/tools/llvm-mca/AArch64/Neoverse/N2-basic-instructions.s
@@ -2086,9 +2086,9 @@ drps
# CHECK-NEXT: 1 3 0.50 fcvtas x27, d28
# CHECK-NEXT: 1 3 0.50 fcvtau w29, d30
# CHECK-NEXT: 1 3 0.50 fcvtau xzr, d0
-# CHECK-NEXT: 1 3 3.00 fmov w3, s9
+# CHECK-NEXT: 1 2 0.50 fmov w3, s9
# CHECK-NEXT: 1 3 3.00 fmov s9, w3
-# CHECK-NEXT: 1 3 3.00 fmov x20, d31
+# CHECK-NEXT: 1 2 0.50 fmov x20, d31
# CHECK-NEXT: 1 3 3.00 fmov d1, x15
# CHECK-NEXT: 1 2 0.50 fmov x3, v12.d[1]
# CHECK-NEXT: 2 5 1.00 fmov v1.d[1], x19
@@ -2557,7 +2557,7 @@ drps
# CHECK: Resource pressure per iteration:
# CHECK-NEXT: [0.0] [0.1] [1.0] [1.1] [2] [3.0] [3.1] [4] [5] [6.0] [6.1] [7] [8]
-# CHECK-NEXT: 11.00 11.00 33.00 33.00 87.33 151.33 151.33 515.75 249.75 161.25 161.25 215.50 85.50
+# CHECK-NEXT: 11.00 11.00 33.00 33.00 87.33 151.33 151.33 509.75 249.75 161.25 161.25 216.50 86.50
# CHECK: Resource pressure by instruction:
# CHECK-NEXT: [0.0] [0.1] [1.0] [1.1] [2] [3.0] [3.1] [4] [5] [6.0] [6.1] [7] [8] Instructions:
@@ -3270,9 +3270,9 @@ drps
# CHECK-NEXT: - - - - - - - - - - - 0.50 0.50 fcvtas x27, d28
# CHECK-NEXT: - - - - - - - - - - - 0.50 0.50 fcvtau w29, d30
# CHECK-NEXT: - - - - - - - - - - - 0.50 0.50 fcvtau xzr, d0
-# CHECK-NEXT: - - - - - - - 3.00 - - - - - fmov w3, s9
+# CHECK-NEXT: - - - - - - - - - - - 0.50 0.50 fmov w3, s9
# CHECK-NEXT: - - - - - - - 3.00 - - - - - fmov s9, w3
-# CHECK-NEXT: - - - - - - - 3.00 - - - - - fmov x20, d31
+# CHECK-NEXT: - - - - - - - - - - - 0.50 0.50 fmov x20, d31
# CHECK-NEXT: - - - - - - - 3.00 - - - - - fmov d1, x15
# CHECK-NEXT: - - - - - - - - - - - 0.50 0.50 fmov x3, v12.d[1]
# CHECK-NEXT: - - - - - - - 1.00 - - - 0.50 0.50 fmov v1.d[1], x19
>From 824b64079174a30813ef86bd672171994d80f17e Mon Sep 17 00:00:00 2001
From: Simon Wallis <simon.wallis2 at arm.com>
Date: Thu, 13 Nov 2025 10:45:23 +0000
Subject: [PATCH 4/7] [AArch64] Update zero latency instructions in Neoverse
scheduling tables
NeoverseZeroMove was introduced for Neoverse-V2 and was added to V3 and V3AE.
Use NeoverseZeroMove for Neoverse-V1, N2, N3 in the same way, including these instructions:
MOV Xd|Wd, #0|XZR|WZR
For all Neoverse targets, the following instructions are also decoded as not utilizing the scheduling and execution resources of the machine:
MOV Wd,Wn
MOV Xd,Xn
For Neoverse-N3 only, these instructions also have zero latency
FMOV Dd, Dn
FMOV Sd, Sn
Change-Id: I7a6a971cf75c60d8f75b210f0529c4ad813775a3
---
.../Target/AArch64/AArch64SchedNeoverseN2.td | 5 +-
.../Target/AArch64/AArch64SchedNeoverseN3.td | 2 +-
.../Target/AArch64/AArch64SchedNeoverseV1.td | 2 +-
.../Target/AArch64/AArch64SchedNeoverseV2.td | 5 +-
.../Target/AArch64/AArch64SchedNeoverseV3.td | 5 +-
.../AArch64/AArch64SchedNeoverseV3AE.td | 5 +-
.../AArch64/Neoverse/N2-basic-instructions.s | 10 +--
.../AArch64/Neoverse/N3-basic-instructions.s | 14 ++--
.../AArch64/Neoverse/N3-sve-instructions.s | 2 +-
.../AArch64/Neoverse/V1-basic-instructions.s | 76 +++++++++----------
.../llvm-mca/AArch64/Neoverse/V1-writeback.s | 31 ++++----
.../AArch64/Neoverse/V1-zero-dependency.s | 6 +-
.../AArch64/Neoverse/V2-basic-instructions.s | 10 +--
.../AArch64/Neoverse/V2-zero-lat-movs.s | 24 +++---
.../AArch64/Neoverse/V3-basic-instructions.s | 10 +--
.../AArch64/Neoverse/V3-zero-lat-movs.s | 24 +++---
.../Neoverse/V3AE-basic-instructions.s | 10 +--
.../AArch64/Neoverse/V3AE-zero-lat-movs.s | 24 +++---
18 files changed, 126 insertions(+), 139 deletions(-)
diff --git a/llvm/lib/Target/AArch64/AArch64SchedNeoverseN2.td b/llvm/lib/Target/AArch64/AArch64SchedNeoverseN2.td
index b4690e6045f85..a02130f8390a7 100644
--- a/llvm/lib/Target/AArch64/AArch64SchedNeoverseN2.td
+++ b/llvm/lib/Target/AArch64/AArch64SchedNeoverseN2.td
@@ -74,10 +74,7 @@ def : WriteRes<WriteLDHi, []> { let Latency = 4; }
//===----------------------------------------------------------------------===//
// Define generic 0 micro-op types
-def N2Write_0c : SchedWriteRes<[]> {
- let Latency = 0;
- let NumMicroOps = 0;
-}
+def N2Write_0c : SchedWriteRes<[]> { let Latency = 0; }
// Define generic 1 micro-op types
diff --git a/llvm/lib/Target/AArch64/AArch64SchedNeoverseN3.td b/llvm/lib/Target/AArch64/AArch64SchedNeoverseN3.td
index 32d48ca66ee2d..ae24fc1e35a89 100644
--- a/llvm/lib/Target/AArch64/AArch64SchedNeoverseN3.td
+++ b/llvm/lib/Target/AArch64/AArch64SchedNeoverseN3.td
@@ -75,7 +75,7 @@ def : WriteRes<WriteHint, []> { let Latency = 1; }
def N3Write_0c : SchedWriteRes<[]> {
let Latency = 0;
- let NumMicroOps = 0;
+ let NumMicroOps = 1;
}
def N3Write_4c : SchedWriteRes<[]> {
diff --git a/llvm/lib/Target/AArch64/AArch64SchedNeoverseV1.td b/llvm/lib/Target/AArch64/AArch64SchedNeoverseV1.td
index 8d33ca22616c2..ac3b9a3d40192 100644
--- a/llvm/lib/Target/AArch64/AArch64SchedNeoverseV1.td
+++ b/llvm/lib/Target/AArch64/AArch64SchedNeoverseV1.td
@@ -91,7 +91,7 @@ def : WriteRes<WriteHint, []> { let Latency = 1; }
//===----------------------------------------------------------------------===//
// Define generic 0 micro-op types
-let Latency = 0, NumMicroOps = 0 in
+let Latency = 0, NumMicroOps = 1 in
def V1Write_0c_0Z : SchedWriteRes<[]>;
diff --git a/llvm/lib/Target/AArch64/AArch64SchedNeoverseV2.td b/llvm/lib/Target/AArch64/AArch64SchedNeoverseV2.td
index 1ef087f07022d..2387f176f3051 100644
--- a/llvm/lib/Target/AArch64/AArch64SchedNeoverseV2.td
+++ b/llvm/lib/Target/AArch64/AArch64SchedNeoverseV2.td
@@ -94,10 +94,7 @@ def : WriteRes<WriteLDHi, []> { let Latency = 4; }
//===----------------------------------------------------------------------===//
// Define generic 0 micro-op types
-def V2Write_0c : SchedWriteRes<[]> {
- let Latency = 0;
- let NumMicroOps = 0;
-}
+def V2Write_0c : SchedWriteRes<[]> { let Latency = 0; }
// Define generic 1 micro-op types
diff --git a/llvm/lib/Target/AArch64/AArch64SchedNeoverseV3.td b/llvm/lib/Target/AArch64/AArch64SchedNeoverseV3.td
index 3dd2988088f0b..e23576a20d277 100644
--- a/llvm/lib/Target/AArch64/AArch64SchedNeoverseV3.td
+++ b/llvm/lib/Target/AArch64/AArch64SchedNeoverseV3.td
@@ -94,10 +94,7 @@ def : WriteRes<WriteLDHi, []> { let Latency = 4; }
//===----------------------------------------------------------------------===//
// Define generic 0 micro-op types
-def V3Write_0c : SchedWriteRes<[]> {
- let Latency = 0;
- let NumMicroOps = 0;
-}
+def V3Write_0c : SchedWriteRes<[]> { let Latency = 0; }
// Define generic 1 micro-op types
diff --git a/llvm/lib/Target/AArch64/AArch64SchedNeoverseV3AE.td b/llvm/lib/Target/AArch64/AArch64SchedNeoverseV3AE.td
index 19b56260387e1..0f1ec669a4e5e 100644
--- a/llvm/lib/Target/AArch64/AArch64SchedNeoverseV3AE.td
+++ b/llvm/lib/Target/AArch64/AArch64SchedNeoverseV3AE.td
@@ -89,10 +89,7 @@ def : WriteRes<WriteLDHi, []> { let Latency = 4; }
//===----------------------------------------------------------------------===//
// Define generic 0 micro-op types
-def V3AEWrite_0c : SchedWriteRes<[]> {
- let Latency = 0;
- let NumMicroOps = 0;
-}
+def V3AEWrite_0c : SchedWriteRes<[]> { let Latency = 0; }
// Define generic 1 micro-op types
diff --git a/llvm/test/tools/llvm-mca/AArch64/Neoverse/N2-basic-instructions.s b/llvm/test/tools/llvm-mca/AArch64/Neoverse/N2-basic-instructions.s
index b140e9a9927ff..f6ea4c4769c0a 100644
--- a/llvm/test/tools/llvm-mca/AArch64/Neoverse/N2-basic-instructions.s
+++ b/llvm/test/tools/llvm-mca/AArch64/Neoverse/N2-basic-instructions.s
@@ -2508,14 +2508,14 @@ drps
# CHECK-NEXT: 1 2 0.50 bics x3, xzr, x3, lsl #1
# CHECK-NEXT: 1 2 0.50 tst w3, w7, lsl #31
# CHECK-NEXT: 1 2 0.50 tst x2, x20, asr #2
-# CHECK-NEXT: 0 0 0.00 mov x3, x6
-# CHECK-NEXT: 0 0 0.00 mov x3, xzr
-# CHECK-NEXT: 0 0 0.00 mov wzr, w2
-# CHECK-NEXT: 0 0 0.00 mov w3, w5
+# CHECK-NEXT: 1 0 0.20 mov x3, x6
+# CHECK-NEXT: 1 0 0.20 mov x3, xzr
+# CHECK-NEXT: 1 0 0.20 mov wzr, w2
+# CHECK-NEXT: 1 0 0.20 mov w3, w5
# CHECK-NEXT: 1 1 0.25 movz w2, #0, lsl #16
# CHECK-NEXT: 1 1 0.25 mov w2, #-1235
# CHECK-NEXT: 1 1 0.25 mov x2, #5299989643264
-# CHECK-NEXT: 0 0 0.00 mov x2, #0
+# CHECK-NEXT: 1 0 0.20 mov x2, #0
# CHECK-NEXT: 1 1 0.25 movk w3, #0
# CHECK-NEXT: 1 1 0.25 movz x4, #0, lsl #16
# CHECK-NEXT: 1 1 0.25 movk w5, #0, lsl #16
diff --git a/llvm/test/tools/llvm-mca/AArch64/Neoverse/N3-basic-instructions.s b/llvm/test/tools/llvm-mca/AArch64/Neoverse/N3-basic-instructions.s
index f7311b5e41b2e..5f48217f8fab9 100644
--- a/llvm/test/tools/llvm-mca/AArch64/Neoverse/N3-basic-instructions.s
+++ b/llvm/test/tools/llvm-mca/AArch64/Neoverse/N3-basic-instructions.s
@@ -1888,7 +1888,7 @@ drps
# CHECK-NEXT: 1 2 0.50 fccmpe d31, d5, #7, ne
# CHECK-NEXT: 1 2 0.50 fcsel s3, s20, s9, pl
# CHECK-NEXT: 1 2 0.50 fcsel d9, d10, d11, mi
-# CHECK-NEXT: 0 0 0.00 fmov s0, s1
+# CHECK-NEXT: 1 0 0.20 fmov s0, s1
# CHECK-NEXT: 1 2 0.50 fabs s2, s3
# CHECK-NEXT: 1 2 0.50 fneg s4, s5
# CHECK-NEXT: 1 7 1.00 fsqrt s6, s7
@@ -1901,7 +1901,7 @@ drps
# CHECK-NEXT: 1 3 1.00 frinta s20, s21
# CHECK-NEXT: 1 3 1.00 frintx s22, s23
# CHECK-NEXT: 1 3 1.00 frinti s24, s25
-# CHECK-NEXT: 0 0 0.00 fmov d0, d1
+# CHECK-NEXT: 1 0 0.20 fmov d0, d1
# CHECK-NEXT: 1 2 0.50 fabs d2, d3
# CHECK-NEXT: 1 2 0.50 fneg d4, d5
# CHECK-NEXT: 1 12 1.00 fsqrt d6, d7
@@ -2508,14 +2508,14 @@ drps
# CHECK-NEXT: 1 2 0.50 bics x3, xzr, x3, lsl #1
# CHECK-NEXT: 1 2 0.50 tst w3, w7, lsl #31
# CHECK-NEXT: 1 2 0.50 tst x2, x20, asr #2
-# CHECK-NEXT: 0 0 0.00 mov x3, x6
-# CHECK-NEXT: 0 0 0.00 mov x3, xzr
-# CHECK-NEXT: 0 0 0.00 mov wzr, w2
-# CHECK-NEXT: 0 0 0.00 mov w3, w5
+# CHECK-NEXT: 1 0 0.20 mov x3, x6
+# CHECK-NEXT: 1 0 0.20 mov x3, xzr
+# CHECK-NEXT: 1 0 0.20 mov wzr, w2
+# CHECK-NEXT: 1 0 0.20 mov w3, w5
# CHECK-NEXT: 1 1 0.25 movz w2, #0, lsl #16
# CHECK-NEXT: 1 1 0.25 mov w2, #-1235
# CHECK-NEXT: 1 1 0.25 mov x2, #5299989643264
-# CHECK-NEXT: 0 0 0.00 mov x2, #0
+# CHECK-NEXT: 1 0 0.20 mov x2, #0
# CHECK-NEXT: 1 1 0.25 movk w3, #0
# CHECK-NEXT: 1 1 0.25 movz x4, #0, lsl #16
# CHECK-NEXT: 1 1 0.25 movk w5, #0, lsl #16
diff --git a/llvm/test/tools/llvm-mca/AArch64/Neoverse/N3-sve-instructions.s b/llvm/test/tools/llvm-mca/AArch64/Neoverse/N3-sve-instructions.s
index 395aa1141abb5..8977802be6bfc 100644
--- a/llvm/test/tools/llvm-mca/AArch64/Neoverse/N3-sve-instructions.s
+++ b/llvm/test/tools/llvm-mca/AArch64/Neoverse/N3-sve-instructions.s
@@ -5471,7 +5471,7 @@ zip2 z31.s, z31.s, z31.s
# CHECK-NEXT: 1 2 0.50 sel z23.d, p11, z13.d, z8.d
# CHECK-NEXT: 1 2 0.50 sel z23.h, p11, z13.h, z8.h
# CHECK-NEXT: 1 2 0.50 sel z23.s, p11, z13.s, z8.s
-# CHECK-NEXT: 0 0 0.00 * U setffr
+# CHECK-NEXT: 1 0 0.20 * U setffr
# CHECK-NEXT: 1 2 0.50 shadd z0.b, p0/m, z0.b, z1.b
# CHECK-NEXT: 1 2 0.50 shadd z0.h, p0/m, z0.h, z1.h
# CHECK-NEXT: 1 2 0.50 shadd z29.s, p7/m, z29.s, z30.s
diff --git a/llvm/test/tools/llvm-mca/AArch64/Neoverse/V1-basic-instructions.s b/llvm/test/tools/llvm-mca/AArch64/Neoverse/V1-basic-instructions.s
index f75222f27a94a..eaf128acc0195 100644
--- a/llvm/test/tools/llvm-mca/AArch64/Neoverse/V1-basic-instructions.s
+++ b/llvm/test/tools/llvm-mca/AArch64/Neoverse/V1-basic-instructions.s
@@ -2562,78 +2562,78 @@ drps
# CHECK-NEXT: 2 7 0.33 * ldr q17, [x23, w9, sxtw]
# CHECK-NEXT: 3 2 0.50 * str q18, [x22, w10, sxtw]
# CHECK-NEXT: 2 7 0.33 * ldr q19, [x21, wzr, sxtw #4]
-# CHECK-NEXT: 1 4 0.33 * ldp w3, w5, [sp]
+# CHECK-NEXT: 2 4 0.33 * ldp w3, w5, [sp]
# CHECK-NEXT: 2 1 0.50 * stp wzr, w9, [sp, #252]
-# CHECK-NEXT: 1 4 0.33 * ldp w2, wzr, [sp, #-256]
-# CHECK-NEXT: 1 4 0.33 * ldp w9, w10, [sp, #4]
-# CHECK-NEXT: 2 5 0.33 * ldpsw x9, x10, [sp, #4]
-# CHECK-NEXT: 2 5 0.33 * ldpsw x9, x10, [x2, #-256]
-# CHECK-NEXT: 2 5 0.33 * ldpsw x20, x30, [sp, #252]
+# CHECK-NEXT: 2 4 0.33 * ldp w2, wzr, [sp, #-256]
+# CHECK-NEXT: 2 4 0.33 * ldp w9, w10, [sp, #4]
+# CHECK-NEXT: 3 5 0.33 * ldpsw x9, x10, [sp, #4]
+# CHECK-NEXT: 3 5 0.33 * ldpsw x9, x10, [x2, #-256]
+# CHECK-NEXT: 3 5 0.33 * ldpsw x20, x30, [sp, #252]
# CHECK-NEXT: 2 4 0.67 * ldp x21, x29, [x2, #504]
# CHECK-NEXT: 2 4 0.67 * ldp x22, x23, [x3, #-512]
# CHECK-NEXT: 2 4 0.67 * ldp x24, x25, [x4, #8]
-# CHECK-NEXT: 1 6 0.33 * ldp s29, s28, [sp, #252]
+# CHECK-NEXT: 2 6 0.33 * ldp s29, s28, [sp, #252]
# CHECK-NEXT: 2 2 0.50 * stp s27, s26, [sp, #-256]
-# CHECK-NEXT: 1 6 0.33 * ldp s1, s2, [x3, #44]
+# CHECK-NEXT: 2 6 0.33 * ldp s1, s2, [x3, #44]
# CHECK-NEXT: 2 2 0.50 * stp d3, d5, [x9, #504]
# CHECK-NEXT: 2 2 0.50 * stp d7, d11, [x10, #-512]
# CHECK-NEXT: 2 1 0.50 * stnp x20, x16, [x8]
# CHECK-NEXT: 2 1 0.50 * stp x3, x6, [x16]
-# CHECK-NEXT: 1 6 0.33 * ldp d2, d3, [x30, #-8]
+# CHECK-NEXT: 2 6 0.33 * ldp d2, d3, [x30, #-8]
# CHECK-NEXT: 2 2 0.50 * stp q3, q5, [sp]
# CHECK-NEXT: 2 2 0.50 * stp q17, q19, [sp, #1008]
# CHECK-NEXT: 2 6 0.67 * ldp q23, q29, [x1, #-1024]
-# CHECK-NEXT: 2 4 0.33 * ldp w3, w5, [sp], #0
+# CHECK-NEXT: 3 4 0.33 * ldp w3, w5, [sp], #0
# CHECK-NEXT: 3 1 0.50 * stp wzr, w9, [sp], #252
-# CHECK-NEXT: 2 4 0.33 * ldp w2, wzr, [sp], #-256
-# CHECK-NEXT: 2 4 0.33 * ldp w9, w10, [sp], #4
-# CHECK-NEXT: 3 5 0.50 * ldpsw x9, x10, [sp], #4
-# CHECK-NEXT: 3 5 0.50 * ldpsw x9, x10, [x2], #-256
-# CHECK-NEXT: 3 5 0.50 * ldpsw x20, x30, [sp], #252
+# CHECK-NEXT: 3 4 0.33 * ldp w2, wzr, [sp], #-256
+# CHECK-NEXT: 3 4 0.33 * ldp w9, w10, [sp], #4
+# CHECK-NEXT: 4 5 0.50 * ldpsw x9, x10, [sp], #4
+# CHECK-NEXT: 4 5 0.50 * ldpsw x9, x10, [x2], #-256
+# CHECK-NEXT: 4 5 0.50 * ldpsw x20, x30, [sp], #252
# CHECK-NEXT: 3 4 0.67 * ldp x21, x29, [x2], #504
# CHECK-NEXT: 3 4 0.67 * ldp x22, x23, [x3], #-512
# CHECK-NEXT: 3 4 0.67 * ldp x24, x25, [x4], #8
-# CHECK-NEXT: 2 6 0.33 * ldp s29, s28, [sp], #252
+# CHECK-NEXT: 3 6 0.33 * ldp s29, s28, [sp], #252
# CHECK-NEXT: 3 2 0.50 * stp s27, s26, [sp], #-256
-# CHECK-NEXT: 2 6 0.33 * ldp s1, s2, [x3], #44
+# CHECK-NEXT: 3 6 0.33 * ldp s1, s2, [x3], #44
# CHECK-NEXT: 3 2 0.50 * stp d3, d5, [x9], #504
# CHECK-NEXT: 3 2 0.50 * stp d7, d11, [x10], #-512
-# CHECK-NEXT: 2 6 0.33 * ldp d2, d3, [x30], #-8
+# CHECK-NEXT: 3 6 0.33 * ldp d2, d3, [x30], #-8
# CHECK-NEXT: 4 2 1.00 * stp q3, q5, [sp], #0
# CHECK-NEXT: 4 2 1.00 * stp q17, q19, [sp], #1008
# CHECK-NEXT: 3 6 0.67 * ldp q23, q29, [x1], #-1024
-# CHECK-NEXT: 2 4 0.33 * ldp w3, w5, [sp, #0]!
+# CHECK-NEXT: 3 4 0.33 * ldp w3, w5, [sp, #0]!
# CHECK-NEXT: 3 1 0.50 * stp wzr, w9, [sp, #252]!
-# CHECK-NEXT: 2 4 0.33 * ldp w2, wzr, [sp, #-256]!
-# CHECK-NEXT: 2 4 0.33 * ldp w9, w10, [sp, #4]!
-# CHECK-NEXT: 3 5 0.50 * ldpsw x9, x10, [sp, #4]!
-# CHECK-NEXT: 3 5 0.50 * ldpsw x9, x10, [x2, #-256]!
-# CHECK-NEXT: 3 5 0.50 * ldpsw x20, x30, [sp, #252]!
+# CHECK-NEXT: 3 4 0.33 * ldp w2, wzr, [sp, #-256]!
+# CHECK-NEXT: 3 4 0.33 * ldp w9, w10, [sp, #4]!
+# CHECK-NEXT: 4 5 0.50 * ldpsw x9, x10, [sp, #4]!
+# CHECK-NEXT: 4 5 0.50 * ldpsw x9, x10, [x2, #-256]!
+# CHECK-NEXT: 4 5 0.50 * ldpsw x20, x30, [sp, #252]!
# CHECK-NEXT: 3 4 0.67 * ldp x21, x29, [x2, #504]!
# CHECK-NEXT: 3 4 0.67 * ldp x22, x23, [x3, #-512]!
# CHECK-NEXT: 3 4 0.67 * ldp x24, x25, [x4, #8]!
-# CHECK-NEXT: 2 6 0.33 * ldp s29, s28, [sp, #252]!
+# CHECK-NEXT: 3 6 0.33 * ldp s29, s28, [sp, #252]!
# CHECK-NEXT: 3 2 0.50 * stp s27, s26, [sp, #-256]!
-# CHECK-NEXT: 2 6 0.33 * ldp s1, s2, [x3, #44]!
+# CHECK-NEXT: 3 6 0.33 * ldp s1, s2, [x3, #44]!
# CHECK-NEXT: 3 2 0.50 * stp d3, d5, [x9, #504]!
# CHECK-NEXT: 3 2 0.50 * stp d7, d11, [x10, #-512]!
-# CHECK-NEXT: 2 6 0.33 * ldp d2, d3, [x30, #-8]!
+# CHECK-NEXT: 3 6 0.33 * ldp d2, d3, [x30, #-8]!
# CHECK-NEXT: 4 2 1.00 * stp q3, q5, [sp, #0]!
# CHECK-NEXT: 4 2 1.00 * stp q17, q19, [sp, #1008]!
# CHECK-NEXT: 3 6 0.67 * ldp q23, q29, [x1, #-1024]!
-# CHECK-NEXT: 1 4 0.33 * ldnp w3, w5, [sp]
+# CHECK-NEXT: 2 4 0.33 * ldnp w3, w5, [sp]
# CHECK-NEXT: 2 1 0.50 * stnp wzr, w9, [sp, #252]
-# CHECK-NEXT: 1 4 0.33 * ldnp w2, wzr, [sp, #-256]
-# CHECK-NEXT: 1 4 0.33 * ldnp w9, w10, [sp, #4]
+# CHECK-NEXT: 2 4 0.33 * ldnp w2, wzr, [sp, #-256]
+# CHECK-NEXT: 2 4 0.33 * ldnp w9, w10, [sp, #4]
# CHECK-NEXT: 2 4 0.67 * ldnp x21, x29, [x2, #504]
# CHECK-NEXT: 2 4 0.67 * ldnp x22, x23, [x3, #-512]
# CHECK-NEXT: 2 4 0.67 * ldnp x24, x25, [x4, #8]
-# CHECK-NEXT: 1 6 0.33 * ldnp s29, s28, [sp, #252]
+# CHECK-NEXT: 2 6 0.33 * ldnp s29, s28, [sp, #252]
# CHECK-NEXT: 2 2 0.50 * stnp s27, s26, [sp, #-256]
-# CHECK-NEXT: 1 6 0.33 * ldnp s1, s2, [x3, #44]
+# CHECK-NEXT: 2 6 0.33 * ldnp s1, s2, [x3, #44]
# CHECK-NEXT: 2 2 0.50 * stnp d3, d5, [x9, #504]
# CHECK-NEXT: 2 2 0.50 * stnp d7, d11, [x10, #-512]
-# CHECK-NEXT: 1 6 0.33 * ldnp d2, d3, [x30, #-8]
+# CHECK-NEXT: 2 6 0.33 * ldnp d2, d3, [x30, #-8]
# CHECK-NEXT: 2 2 0.50 * stnp q3, q5, [sp]
# CHECK-NEXT: 2 2 0.50 * stnp q17, q19, [sp, #1008]
# CHECK-NEXT: 2 6 0.67 * ldnp q23, q29, [x1, #-1024]
@@ -2673,14 +2673,14 @@ drps
# CHECK-NEXT: 1 2 0.50 bics x3, xzr, x3, lsl #1
# CHECK-NEXT: 1 2 0.50 tst w3, w7, lsl #31
# CHECK-NEXT: 1 2 0.50 tst x2, x20, asr #2
-# CHECK-NEXT: 0 0 0.00 mov x3, x6
-# CHECK-NEXT: 0 0 0.00 mov x3, xzr
-# CHECK-NEXT: 0 0 0.00 mov wzr, w2
-# CHECK-NEXT: 0 0 0.00 mov w3, w5
+# CHECK-NEXT: 1 0 0.13 mov x3, x6
+# CHECK-NEXT: 1 0 0.13 mov x3, xzr
+# CHECK-NEXT: 1 0 0.13 mov wzr, w2
+# CHECK-NEXT: 1 0 0.13 mov w3, w5
# CHECK-NEXT: 1 1 0.25 movz w2, #0, lsl #16
# CHECK-NEXT: 1 1 0.25 mov w2, #-1235
# CHECK-NEXT: 1 1 0.25 mov x2, #5299989643264
-# CHECK-NEXT: 0 0 0.00 mov x2, #0
+# CHECK-NEXT: 1 0 0.13 mov x2, #0
# CHECK-NEXT: 1 1 0.25 movk w3, #0
# CHECK-NEXT: 1 1 0.25 movz x4, #0, lsl #16
# CHECK-NEXT: 1 1 0.25 movk w5, #0, lsl #16
diff --git a/llvm/test/tools/llvm-mca/AArch64/Neoverse/V1-writeback.s b/llvm/test/tools/llvm-mca/AArch64/Neoverse/V1-writeback.s
index 1961b24ae6aac..fee62ce565b43 100644
--- a/llvm/test/tools/llvm-mca/AArch64/Neoverse/V1-writeback.s
+++ b/llvm/test/tools/llvm-mca/AArch64/Neoverse/V1-writeback.s
@@ -3140,13 +3140,13 @@ add x0, x27, 1
# CHECK: Iterations: 100
# CHECK-NEXT: Instructions: 1000
-# CHECK-NEXT: Total Cycles: 609
-# CHECK-NEXT: Total uOps: 3300
+# CHECK-NEXT: Total Cycles: 708
+# CHECK-NEXT: Total uOps: 3500
# CHECK: Dispatch Width: 8
-# CHECK-NEXT: uOps Per Cycle: 5.42
-# CHECK-NEXT: IPC: 1.64
-# CHECK-NEXT: Block RThroughput: 4.1
+# CHECK-NEXT: uOps Per Cycle: 4.94
+# CHECK-NEXT: IPC: 1.41
+# CHECK-NEXT: Block RThroughput: 4.4
# CHECK: Timeline view:
# CHECK-NEXT: 01234
@@ -3161,7 +3161,7 @@ add x0, x27, 1
# CHECK-NEXT: [0,6] . DeeeeeeE-R ldp s1, s2, [x27], #248
# CHECK-NEXT: [0,7] . D=eE-----R add x0, x27, #1
# CHECK-NEXT: [0,8] . D=eeeeeeER ldp d1, d2, [x27], #496
-# CHECK-NEXT: [0,9] . D==eE----R add x0, x27, #1
+# CHECK-NEXT: [0,9] . .D=eE----R add x0, x27, #1
# CHECK: Average Wait times (based on the timeline view):
# CHECK-NEXT: [0]: Executions
@@ -3179,18 +3179,18 @@ add x0, x27, 1
# CHECK-NEXT: 6. 1 1.0 0.0 1.0 ldp s1, s2, [x27], #248
# CHECK-NEXT: 7. 1 2.0 0.0 5.0 add x0, x27, #1
# CHECK-NEXT: 8. 1 2.0 0.0 0.0 ldp d1, d2, [x27], #496
-# CHECK-NEXT: 9. 1 3.0 0.0 4.0 add x0, x27, #1
-# CHECK-NEXT: 1 1.4 0.3 2.8 <total>
+# CHECK-NEXT: 9. 1 2.0 0.0 4.0 add x0, x27, #1
+# CHECK-NEXT: 1 1.3 0.3 2.8 <total>
# CHECK: [44] Code Region - G45
# CHECK: Iterations: 100
# CHECK-NEXT: Instructions: 1000
# CHECK-NEXT: Total Cycles: 507
-# CHECK-NEXT: Total uOps: 1700
+# CHECK-NEXT: Total uOps: 2000
# CHECK: Dispatch Width: 8
-# CHECK-NEXT: uOps Per Cycle: 3.35
+# CHECK-NEXT: uOps Per Cycle: 3.94
# CHECK-NEXT: IPC: 1.97
# CHECK-NEXT: Block RThroughput: 2.5
@@ -3233,10 +3233,10 @@ add x0, x27, 1
# CHECK: Iterations: 100
# CHECK-NEXT: Instructions: 1000
# CHECK-NEXT: Total Cycles: 507
-# CHECK-NEXT: Total uOps: 1900
+# CHECK-NEXT: Total uOps: 2200
# CHECK: Dispatch Width: 8
-# CHECK-NEXT: uOps Per Cycle: 3.75
+# CHECK-NEXT: uOps Per Cycle: 4.34
# CHECK-NEXT: IPC: 1.97
# CHECK-NEXT: Block RThroughput: 3.0
@@ -3251,7 +3251,7 @@ add x0, x27, 1
# CHECK-NEXT: [0,4] .D=eeeeER .. ldp x1, x2, [x27, #496]!
# CHECK-NEXT: [0,5] .D==eE--R .. add x0, x27, #1
# CHECK-NEXT: [0,6] .D==eeeeeER. ldpsw x1, x2, [x27], #248
-# CHECK-NEXT: [0,7] .D===eE---R. add x0, x27, #1
+# CHECK-NEXT: [0,7] . D==eE---R. add x0, x27, #1
# CHECK-NEXT: [0,8] . D==eeeeeER ldpsw x1, x2, [x27, #248]!
# CHECK-NEXT: [0,9] . D===eE---R add x0, x27, #1
@@ -3269,10 +3269,10 @@ add x0, x27, 1
# CHECK-NEXT: 4. 1 2.0 0.0 0.0 ldp x1, x2, [x27, #496]!
# CHECK-NEXT: 5. 1 3.0 0.0 2.0 add x0, x27, #1
# CHECK-NEXT: 6. 1 3.0 0.0 0.0 ldpsw x1, x2, [x27], #248
-# CHECK-NEXT: 7. 1 4.0 0.0 3.0 add x0, x27, #1
+# CHECK-NEXT: 7. 1 3.0 0.0 3.0 add x0, x27, #1
# CHECK-NEXT: 8. 1 3.0 0.0 0.0 ldpsw x1, x2, [x27, #248]!
# CHECK-NEXT: 9. 1 4.0 0.0 3.0 add x0, x27, #1
-# CHECK-NEXT: 1 2.7 0.1 1.2 <total>
+# CHECK-NEXT: 1 2.6 0.1 1.2 <total>
# CHECK: [46] Code Region - G47
@@ -5272,4 +5272,3 @@ add x0, x27, 1
# CHECK-NEXT: 2. 1 5.0 0.0 0.0 ldr x2, [x1], #254
# CHECK-NEXT: 3. 1 2.0 0.0 6.0 add x0, x27, #1
# CHECK-NEXT: 1 2.5 0.3 2.0 <total>
-
diff --git a/llvm/test/tools/llvm-mca/AArch64/Neoverse/V1-zero-dependency.s b/llvm/test/tools/llvm-mca/AArch64/Neoverse/V1-zero-dependency.s
index 7767b95ff98ea..47c5b7cd513b3 100644
--- a/llvm/test/tools/llvm-mca/AArch64/Neoverse/V1-zero-dependency.s
+++ b/llvm/test/tools/llvm-mca/AArch64/Neoverse/V1-zero-dependency.s
@@ -7,10 +7,10 @@ cmp x0, #4
# CHECK: Iterations: 100
# CHECK-NEXT: Instructions: 200
# CHECK-NEXT: Total Cycles: 37
-# CHECK-NEXT: Total uOps: 100
+# CHECK-NEXT: Total uOps: 200
# CHECK: Dispatch Width: 8
-# CHECK-NEXT: uOps Per Cycle: 2.70
+# CHECK-NEXT: uOps Per Cycle: 5.41
# CHECK-NEXT: IPC: 5.41
# CHECK-NEXT: Block RThroughput: 0.3
@@ -23,7 +23,7 @@ cmp x0, #4
# CHECK-NEXT: [6]: HasSideEffects (U)
# CHECK: [1] [2] [3] [4] [5] [6] Instructions:
-# CHECK-NEXT: 0 0 0.00 mov x0, x1
+# CHECK-NEXT: 1 0 0.13 mov x0, x1
# CHECK-NEXT: 1 1 0.33 cmp x0, #4
# CHECK: Resources:
diff --git a/llvm/test/tools/llvm-mca/AArch64/Neoverse/V2-basic-instructions.s b/llvm/test/tools/llvm-mca/AArch64/Neoverse/V2-basic-instructions.s
index 9c987d54d2350..54b5f1644be48 100644
--- a/llvm/test/tools/llvm-mca/AArch64/Neoverse/V2-basic-instructions.s
+++ b/llvm/test/tools/llvm-mca/AArch64/Neoverse/V2-basic-instructions.s
@@ -2536,14 +2536,14 @@ drps
# CHECK-NEXT: 1 2 0.50 bics x3, xzr, x3, lsl #1
# CHECK-NEXT: 1 2 0.50 tst w3, w7, lsl #31
# CHECK-NEXT: 1 2 0.50 tst x2, x20, asr #2
-# CHECK-NEXT: 0 0 0.00 mov x3, x6
-# CHECK-NEXT: 0 0 0.00 mov x3, xzr
-# CHECK-NEXT: 0 0 0.00 mov wzr, w2
-# CHECK-NEXT: 0 0 0.00 mov w3, w5
+# CHECK-NEXT: 1 0 0.17 mov x3, x6
+# CHECK-NEXT: 1 0 0.17 mov x3, xzr
+# CHECK-NEXT: 1 0 0.17 mov wzr, w2
+# CHECK-NEXT: 1 0 0.17 mov w3, w5
# CHECK-NEXT: 1 1 0.17 movz w2, #0, lsl #16
# CHECK-NEXT: 1 1 0.17 mov w2, #-1235
# CHECK-NEXT: 1 1 0.17 mov x2, #5299989643264
-# CHECK-NEXT: 0 0 0.00 mov x2, #0
+# CHECK-NEXT: 1 0 0.17 mov x2, #0
# CHECK-NEXT: 1 1 0.17 movk w3, #0
# CHECK-NEXT: 1 1 0.17 movz x4, #0, lsl #16
# CHECK-NEXT: 1 1 0.17 movk w5, #0, lsl #16
diff --git a/llvm/test/tools/llvm-mca/AArch64/Neoverse/V2-zero-lat-movs.s b/llvm/test/tools/llvm-mca/AArch64/Neoverse/V2-zero-lat-movs.s
index 1cec5897db425..3ddb525327015 100644
--- a/llvm/test/tools/llvm-mca/AArch64/Neoverse/V2-zero-lat-movs.s
+++ b/llvm/test/tools/llvm-mca/AArch64/Neoverse/V2-zero-lat-movs.s
@@ -23,18 +23,18 @@ mov x1, x2
# CHECK-NEXT: [6]: HasSideEffects (U)
# CHECK: [1] [2] [3] [4] [5] [6] Instructions:
-# CHECK-NEXT: 0 0 0.00 mov x1, #0
-# CHECK-NEXT: 0 0 0.00 mov x1, xzr
-# CHECK-NEXT: 0 0 0.00 mov w1, #0
-# CHECK-NEXT: 0 0 0.00 mov w1, wzr
-# CHECK-NEXT: 0 0 0.00 fmov h1, wzr
-# CHECK-NEXT: 0 0 0.00 fmov h1, xzr
-# CHECK-NEXT: 0 0 0.00 fmov s1, wzr
-# CHECK-NEXT: 0 0 0.00 fmov d1, xzr
-# CHECK-NEXT: 0 0 0.00 movi d1, #0000000000000000
-# CHECK-NEXT: 0 0 0.00 movi v1.2d, #0000000000000000
-# CHECK-NEXT: 0 0 0.00 mov w1, w2
-# CHECK-NEXT: 0 0 0.00 mov x1, x2
+# CHECK-NEXT: 1 0 0.17 mov x1, #0
+# CHECK-NEXT: 1 0 0.17 mov x1, xzr
+# CHECK-NEXT: 1 0 0.17 mov w1, #0
+# CHECK-NEXT: 1 0 0.17 mov w1, wzr
+# CHECK-NEXT: 1 0 0.17 fmov h1, wzr
+# CHECK-NEXT: 1 0 0.17 fmov h1, xzr
+# CHECK-NEXT: 1 0 0.17 fmov s1, wzr
+# CHECK-NEXT: 1 0 0.17 fmov d1, xzr
+# CHECK-NEXT: 1 0 0.17 movi d1, #0000000000000000
+# CHECK-NEXT: 1 0 0.17 movi v1.2d, #0000000000000000
+# CHECK-NEXT: 1 0 0.17 mov w1, w2
+# CHECK-NEXT: 1 0 0.17 mov x1, x2
# CHECK: Resources:
# CHECK-NEXT: [0.0] - V2UnitB
diff --git a/llvm/test/tools/llvm-mca/AArch64/Neoverse/V3-basic-instructions.s b/llvm/test/tools/llvm-mca/AArch64/Neoverse/V3-basic-instructions.s
index 67af391e52863..73fd95d6e4a5b 100644
--- a/llvm/test/tools/llvm-mca/AArch64/Neoverse/V3-basic-instructions.s
+++ b/llvm/test/tools/llvm-mca/AArch64/Neoverse/V3-basic-instructions.s
@@ -2536,14 +2536,14 @@ drps
# CHECK-NEXT: 1 2 0.50 bics x3, xzr, x3, lsl #1
# CHECK-NEXT: 1 2 0.50 tst w3, w7, lsl #31
# CHECK-NEXT: 1 2 0.50 tst x2, x20, asr #2
-# CHECK-NEXT: 0 0 0.00 mov x3, x6
-# CHECK-NEXT: 0 0 0.00 mov x3, xzr
-# CHECK-NEXT: 0 0 0.00 mov wzr, w2
-# CHECK-NEXT: 0 0 0.00 mov w3, w5
+# CHECK-NEXT: 1 0 0.10 mov x3, x6
+# CHECK-NEXT: 1 0 0.10 mov x3, xzr
+# CHECK-NEXT: 1 0 0.10 mov wzr, w2
+# CHECK-NEXT: 1 0 0.10 mov w3, w5
# CHECK-NEXT: 1 1 0.13 movz w2, #0, lsl #16
# CHECK-NEXT: 1 1 0.13 mov w2, #-1235
# CHECK-NEXT: 1 1 0.13 mov x2, #5299989643264
-# CHECK-NEXT: 0 0 0.00 mov x2, #0
+# CHECK-NEXT: 1 0 0.10 mov x2, #0
# CHECK-NEXT: 1 1 0.13 movk w3, #0
# CHECK-NEXT: 1 1 0.13 movz x4, #0, lsl #16
# CHECK-NEXT: 1 1 0.13 movk w5, #0, lsl #16
diff --git a/llvm/test/tools/llvm-mca/AArch64/Neoverse/V3-zero-lat-movs.s b/llvm/test/tools/llvm-mca/AArch64/Neoverse/V3-zero-lat-movs.s
index 9b4834b12a79b..1eef230b8174e 100644
--- a/llvm/test/tools/llvm-mca/AArch64/Neoverse/V3-zero-lat-movs.s
+++ b/llvm/test/tools/llvm-mca/AArch64/Neoverse/V3-zero-lat-movs.s
@@ -23,18 +23,18 @@ mov x1, x2
# CHECK-NEXT: [6]: HasSideEffects (U)
# CHECK: [1] [2] [3] [4] [5] [6] Instructions:
-# CHECK-NEXT: 0 0 0.00 mov x1, #0
-# CHECK-NEXT: 0 0 0.00 mov x1, xzr
-# CHECK-NEXT: 0 0 0.00 mov w1, #0
-# CHECK-NEXT: 0 0 0.00 mov w1, wzr
-# CHECK-NEXT: 0 0 0.00 fmov h1, wzr
-# CHECK-NEXT: 0 0 0.00 fmov h1, xzr
-# CHECK-NEXT: 0 0 0.00 fmov s1, wzr
-# CHECK-NEXT: 0 0 0.00 fmov d1, xzr
-# CHECK-NEXT: 0 0 0.00 movi d1, #0000000000000000
-# CHECK-NEXT: 0 0 0.00 movi v1.2d, #0000000000000000
-# CHECK-NEXT: 0 0 0.00 mov w1, w2
-# CHECK-NEXT: 0 0 0.00 mov x1, x2
+# CHECK-NEXT: 1 0 0.10 mov x1, #0
+# CHECK-NEXT: 1 0 0.10 mov x1, xzr
+# CHECK-NEXT: 1 0 0.10 mov w1, #0
+# CHECK-NEXT: 1 0 0.10 mov w1, wzr
+# CHECK-NEXT: 1 0 0.10 fmov h1, wzr
+# CHECK-NEXT: 1 0 0.10 fmov h1, xzr
+# CHECK-NEXT: 1 0 0.10 fmov s1, wzr
+# CHECK-NEXT: 1 0 0.10 fmov d1, xzr
+# CHECK-NEXT: 1 0 0.10 movi d1, #0000000000000000
+# CHECK-NEXT: 1 0 0.10 movi v1.2d, #0000000000000000
+# CHECK-NEXT: 1 0 0.10 mov w1, w2
+# CHECK-NEXT: 1 0 0.10 mov x1, x2
# CHECK: Resources:
# CHECK-NEXT: [0.0] - V3UnitB
diff --git a/llvm/test/tools/llvm-mca/AArch64/Neoverse/V3AE-basic-instructions.s b/llvm/test/tools/llvm-mca/AArch64/Neoverse/V3AE-basic-instructions.s
index 5009ce1d54a86..7ab2be5eaa365 100644
--- a/llvm/test/tools/llvm-mca/AArch64/Neoverse/V3AE-basic-instructions.s
+++ b/llvm/test/tools/llvm-mca/AArch64/Neoverse/V3AE-basic-instructions.s
@@ -2536,14 +2536,14 @@ drps
# CHECK-NEXT: 1 2 0.50 bics x3, xzr, x3, lsl #1
# CHECK-NEXT: 1 2 0.50 tst w3, w7, lsl #31
# CHECK-NEXT: 1 2 0.50 tst x2, x20, asr #2
-# CHECK-NEXT: 0 0 0.00 mov x3, x6
-# CHECK-NEXT: 0 0 0.00 mov x3, xzr
-# CHECK-NEXT: 0 0 0.00 mov wzr, w2
-# CHECK-NEXT: 0 0 0.00 mov w3, w5
+# CHECK-NEXT: 1 0 0.10 mov x3, x6
+# CHECK-NEXT: 1 0 0.10 mov x3, xzr
+# CHECK-NEXT: 1 0 0.10 mov wzr, w2
+# CHECK-NEXT: 1 0 0.10 mov w3, w5
# CHECK-NEXT: 1 1 0.13 movz w2, #0, lsl #16
# CHECK-NEXT: 1 1 0.13 mov w2, #-1235
# CHECK-NEXT: 1 1 0.13 mov x2, #5299989643264
-# CHECK-NEXT: 0 0 0.00 mov x2, #0
+# CHECK-NEXT: 1 0 0.10 mov x2, #0
# CHECK-NEXT: 1 1 0.13 movk w3, #0
# CHECK-NEXT: 1 1 0.13 movz x4, #0, lsl #16
# CHECK-NEXT: 1 1 0.13 movk w5, #0, lsl #16
diff --git a/llvm/test/tools/llvm-mca/AArch64/Neoverse/V3AE-zero-lat-movs.s b/llvm/test/tools/llvm-mca/AArch64/Neoverse/V3AE-zero-lat-movs.s
index 783bea288b121..a0840dcddcbab 100644
--- a/llvm/test/tools/llvm-mca/AArch64/Neoverse/V3AE-zero-lat-movs.s
+++ b/llvm/test/tools/llvm-mca/AArch64/Neoverse/V3AE-zero-lat-movs.s
@@ -23,18 +23,18 @@ mov x1, x2
# CHECK-NEXT: [6]: HasSideEffects (U)
# CHECK: [1] [2] [3] [4] [5] [6] Instructions:
-# CHECK-NEXT: 0 0 0.00 mov x1, #0
-# CHECK-NEXT: 0 0 0.00 mov x1, xzr
-# CHECK-NEXT: 0 0 0.00 mov w1, #0
-# CHECK-NEXT: 0 0 0.00 mov w1, wzr
-# CHECK-NEXT: 0 0 0.00 fmov h1, wzr
-# CHECK-NEXT: 0 0 0.00 fmov h1, xzr
-# CHECK-NEXT: 0 0 0.00 fmov s1, wzr
-# CHECK-NEXT: 0 0 0.00 fmov d1, xzr
-# CHECK-NEXT: 0 0 0.00 movi d1, #0000000000000000
-# CHECK-NEXT: 0 0 0.00 movi v1.2d, #0000000000000000
-# CHECK-NEXT: 0 0 0.00 mov w1, w2
-# CHECK-NEXT: 0 0 0.00 mov x1, x2
+# CHECK-NEXT: 1 0 0.10 mov x1, #0
+# CHECK-NEXT: 1 0 0.10 mov x1, xzr
+# CHECK-NEXT: 1 0 0.10 mov w1, #0
+# CHECK-NEXT: 1 0 0.10 mov w1, wzr
+# CHECK-NEXT: 1 0 0.10 fmov h1, wzr
+# CHECK-NEXT: 1 0 0.10 fmov h1, xzr
+# CHECK-NEXT: 1 0 0.10 fmov s1, wzr
+# CHECK-NEXT: 1 0 0.10 fmov d1, xzr
+# CHECK-NEXT: 1 0 0.10 movi d1, #0000000000000000
+# CHECK-NEXT: 1 0 0.10 movi v1.2d, #0000000000000000
+# CHECK-NEXT: 1 0 0.10 mov w1, w2
+# CHECK-NEXT: 1 0 0.10 mov x1, x2
# CHECK: Resources:
# CHECK-NEXT: [0.0] - V3AEUnitB
>From f2d0a292068262aef005694980443f6528765f67 Mon Sep 17 00:00:00 2001
From: Simon Wallis <simon.wallis2 at arm.com>
Date: Thu, 13 Nov 2025 10:50:20 +0000
Subject: [PATCH 5/7] [AArch64] Update zero latency instructions in Neoverse
scheduling tables
NeoverseZeroMove was introduced for Neoverse-V2 and was added to V3 and V3AE.
Use NeoverseZeroMove for Neoverse-V1, N2, N3 in the same way, including these instructions:
MOV Xd|Wd, #0|XZR|WZR
For all Neoverse targets, the following instructions are also decoded as not utilizing the scheduling and execution resources of the machine:
MOV Wd,Wn
MOV Xd,Xn
For Neoverse-N3 only, these instructions also have zero latency
FMOV Dd, Dn
FMOV Sd, Sn
Change-Id: I95c53d373f35bb0bea5174a16c7ab3ac25acf684
---
llvm/test/CodeGen/AArch64/pr164181.ll | 8 ++++----
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/llvm/test/CodeGen/AArch64/pr164181.ll b/llvm/test/CodeGen/AArch64/pr164181.ll
index 987c92b084001..4ec63ecb2eeb4 100644
--- a/llvm/test/CodeGen/AArch64/pr164181.ll
+++ b/llvm/test/CodeGen/AArch64/pr164181.ll
@@ -52,11 +52,11 @@ define void @f(i1 %var_0, i16 %var_1, i64 %var_2, i8 %var_3, i16 %var_4, i1 %var
; CHECK-NEXT: mov x10, xzr
; CHECK-NEXT: mov w23, wzr
; CHECK-NEXT: mov w30, wzr
+; CHECK-NEXT: ldrb w19, [sp, #240]
; CHECK-NEXT: mov w25, wzr
; CHECK-NEXT: mov x24, xzr
-; CHECK-NEXT: mov x3, x26
; CHECK-NEXT: str w8, [sp, #108] // 4-byte Folded Spill
-; CHECK-NEXT: ldrb w19, [sp, #240]
+; CHECK-NEXT: mov x3, x26
; CHECK-NEXT: ldp x9, x8, [sp, #344]
; CHECK-NEXT: str w12, [sp, #92] // 4-byte Folded Spill
; CHECK-NEXT: mov w12, #1 // =0x1
@@ -123,8 +123,8 @@ define void @f(i1 %var_0, i16 %var_1, i64 %var_2, i8 %var_3, i16 %var_4, i1 %var
; CHECK-NEXT: mov x12, #-30 // =0xffffffffffffffe2
; CHECK-NEXT: add x19, x4, w8, sxtw #2
; CHECK-NEXT: mov x9, xzr
-; CHECK-NEXT: mov w4, w30
; CHECK-NEXT: csel x12, x24, x12, lo
+; CHECK-NEXT: mov w4, w30
; CHECK-NEXT: str x12, [sp, #56] // 8-byte Folded Spill
; CHECK-NEXT: b .LBB0_8
; CHECK-NEXT: .p2align 5, , 16
@@ -341,8 +341,8 @@ define void @f(i1 %var_0, i16 %var_1, i64 %var_2, i8 %var_3, i16 %var_4, i1 %var
; CHECK-NEXT: mov x24, x27
; CHECK-NEXT: lsl x23, x14, #1
; CHECK-NEXT: mov x27, #-1 // =0xffffffffffffffff
-; CHECK-NEXT: mov w28, w30
; CHECK-NEXT: madd x14, x14, x3, x11
+; CHECK-NEXT: mov w28, w30
; CHECK-NEXT: mov w3, #-7680 // =0xffffe200
; CHECK-NEXT: b .LBB0_39
; CHECK-NEXT: .p2align 5, , 16
>From 684ad3e5e2724ec3de3071cfb30fe26273a4cc91 Mon Sep 17 00:00:00 2001
From: Simon Wallis <simon.wallis2 at arm.com>
Date: Fri, 14 Nov 2025 13:21:45 +0000
Subject: [PATCH 6/7] [AArch64] Update zero latency instructions in Neoverse
scheduling tables
NeoverseZeroMove was introduced for Neoverse-V2 and was added to V3 and V3AE.
Use NeoverseZeroMove for Neoverse-V1, N2, N3 in the same way, including these instructions:
MOV Xd|Wd, #0|XZR|WZR
For all Neoverse targets, the following instructions are also decoded as not utilizing the scheduling and execution resources of the machine:
MOV Wd,Wn
MOV Xd,Xn
For Neoverse-N3 only, these instructions also have zero latency
FMOV Dd, Dn
FMOV Sd, Sn
Change-Id: Ibbc0ba1da02dd4bf5ca28b33164d8fa4e93958d6
---
.../Target/AArch64/AArch64SchedNeoverseN3.td | 14 +++--
.../Target/AArch64/AArch64SchedNeoverseV1.td | 7 +--
.../AArch64/AArch64SchedPredNeoverse.td | 19 ++++++-
.../AArch64/Neoverse/N3-neon-instructions.s | 14 ++---
.../AArch64/Neoverse/N3-sve-instructions.s | 54 +++++++++----------
5 files changed, 64 insertions(+), 44 deletions(-)
diff --git a/llvm/lib/Target/AArch64/AArch64SchedNeoverseN3.td b/llvm/lib/Target/AArch64/AArch64SchedNeoverseN3.td
index ae24fc1e35a89..1f78ebb57da5e 100644
--- a/llvm/lib/Target/AArch64/AArch64SchedNeoverseN3.td
+++ b/llvm/lib/Target/AArch64/AArch64SchedNeoverseN3.td
@@ -565,6 +565,10 @@ def N3Write_0or2c_1V : SchedWriteVariant<[
SchedVar<NeoverseZeroMove, [N3Write_0c]>,
SchedVar<NoSchedPred, [N3Write_2c_1V]>]>;
+def N3Write_0or2c_1M : SchedWriteVariant<[
+ SchedVar<NeoverseAllElement, [N3Write_0c]>,
+ SchedVar<NoSchedPred, [N3Write_2c_1M]>]>;
+
def N3Write_0or3c_1M0 : SchedWriteVariant<[
SchedVar<NeoverseZeroMove, [N3Write_0c]>,
SchedVar<NoSchedPred, [N3Write_3c_1M0]>]>;
@@ -981,6 +985,8 @@ def : InstRW<[WriteAdr, N3Write_2c_1L01_1V_1I], (instregex "^STP[SDQ](post|pre)$
// ASIMD compare
// ASIMD logical
// ASIMD max/min, basic and pair-wise
+def : InstRW<[N3Write_0or2c_1V], (instregex "^ORRv16i8", "^ORRv8i8")>;
+
def : SchedAlias<WriteVd, N3Write_2c_1V>;
def : SchedAlias<WriteVq, N3Write_2c_1V>;
@@ -1584,10 +1590,11 @@ def : InstRW<[N3Write_2c_1M], (instregex "^REV_PP_[BHSD]")>;
def : InstRW<[N3Write_1c_1M], (instrs SEL_PPPP)>;
// Predicate set
-def : InstRW<[N3Write_2c_1M], (instregex "^PFALSE", "^PTRUE_[BHSD]")>;
+def : InstRW<[N3Write_0c], (instregex "^PFALSE")>;
+def : InstRW<[N3Write_0or2c_1M], (instregex "^PTRUE_[BHSD]")>;
// Predicate set/initialize, set flags
-def : InstRW<[N3Write_2c_1M], (instregex "^PTRUES_[BHSD]")>;
+def : InstRW<[N3Write_0or2c_1M], (instregex "^PTRUES_[BHSD]")>;
// Predicate find first/next
def : InstRW<[N3Write_2c_1M], (instregex "^PFIRST_B$", "^PNEXT_[BHSD]$")>;
@@ -1810,10 +1817,11 @@ def : InstRW<[N3Write_5c_1M0_1V], (instregex "^INDEX_(IR|RI|RR)_D$")>;
// Logical
def : InstRW<[N3Write_2c_1V],
(instregex "^(AND|EOR|ORR)_ZI",
- "^(AND|BIC|EOR|ORR)_ZZZ",
+ "^(AND|BIC|EOR)_ZZZ",
"^EOR(BT|TB)_ZZZ_[BHSD]",
"^(AND|BIC|EOR|NOT|ORR)_(ZPmZ|ZPZZ)_[BHSD]",
"^NOT_ZPmZ_[BHSD]")>;
+def : InstRW<[N3Write_0or2c_1V], (instregex "^ORR_ZZZ")>;
// Max/min, basic and pairwise
def : InstRW<[N3Write_2c_1V], (instregex "^[SU](MAX|MIN)_ZI_[BHSD]",
diff --git a/llvm/lib/Target/AArch64/AArch64SchedNeoverseV1.td b/llvm/lib/Target/AArch64/AArch64SchedNeoverseV1.td
index ac3b9a3d40192..a880c1a034598 100644
--- a/llvm/lib/Target/AArch64/AArch64SchedNeoverseV1.td
+++ b/llvm/lib/Target/AArch64/AArch64SchedNeoverseV1.td
@@ -479,10 +479,6 @@ def V1Write_0or1c_1I : SchedWriteVariant<[
SchedVar<NeoverseZeroMove, [V1Write_0c_0Z]>,
SchedVar<NoSchedPred, [V1Write_1c_1I]>]>;
-def V1Write_0or2c_1V : SchedWriteVariant<[
- SchedVar<NeoverseZeroMove, [V1Write_0c_0Z]>,
- SchedVar<NoSchedPred, [V1Write_2c_1V]>]>;
-
def V1Write_0or3c_1M0 : SchedWriteVariant<[
SchedVar<NeoverseZeroMove, [V1Write_0c_0Z]>,
SchedVar<NoSchedPred, [V1Write_3c_1M0]>]>;
@@ -822,7 +818,7 @@ def : SchedAlias<WriteFImm, V1Write_2c_1V>;
def : InstRW<[V1Write_2c_1V], (instrs FMOVHr, FMOVSr, FMOVDr)>;
// FP transfer, from gen to low half of vec reg
-def : InstRW<[V1Write_0or3c_1M0], (instrs FMOVWHr, FMOVXHr, FMOVWSr, FMOVXDr)>;
+def : InstRW<[V1Write_3c_1M0], (instrs FMOVWHr, FMOVXHr, FMOVWSr, FMOVXDr)>;
// FP transfer, from gen to high half of vec reg
def : InstRW<[V1Write_5c_1M0_1V], (instrs FMOVXDHighr)>;
@@ -1139,7 +1135,6 @@ def : InstRW<[V1Write_3c_1V02], (instrs BFCVT)>;
// ASIMD transpose
// ASIMD unzip/zip
// Covered by "SchedAlias (WriteV[dq]...)" above
-def : InstRW<[V1Write_0or2c_1V], (instrs MOVID, MOVIv2d_ns)>;
// ASIMD duplicate, gen reg
def : InstRW<[V1Write_3c_1M0],
diff --git a/llvm/lib/Target/AArch64/AArch64SchedPredNeoverse.td b/llvm/lib/Target/AArch64/AArch64SchedPredNeoverse.td
index 33b76a4f65f05..2d330de16280b 100644
--- a/llvm/lib/Target/AArch64/AArch64SchedPredNeoverse.td
+++ b/llvm/lib/Target/AArch64/AArch64SchedPredNeoverse.td
@@ -80,5 +80,22 @@ def NeoverseZeroMove : MCSchedPredicate<
// MOVI Dd, #0
// MOVI Vd.2D, #0
CheckAll<[CheckOpcode<[MOVID, MOVIv2d_ns]>,
- CheckImmOperand<1, 0>]>
+ CheckImmOperand<1, 0>]>,
+ // MOV Zd, Zn
+ CheckAll<[CheckOpcode<[ORR_ZZZ]>,
+ CheckSameRegOperand<1, 2>]>,
+ // MOV Vd, Vn
+ CheckAll<[CheckOpcode<[ORRv16i8, ORRv8i8]>,
+ CheckSameRegOperand<1, 2>]>,
+ ]>>;
+
+def NeoverseAllElement : MCSchedPredicate<
+ CheckAny<[
+ // PTRUE Pd, ALL
+ // PTRUES Pd, ALL
+ CheckAll<[CheckOpcode<[
+ PTRUE_B, PTRUE_H, PTRUE_S, PTRUE_D,
+ PTRUES_B, PTRUES_H, PTRUES_S, PTRUES_D]>,
+ CheckIsImmOperand<1>,
+ CheckImmOperand<1, 31>]>,
]>>;
diff --git a/llvm/test/tools/llvm-mca/AArch64/Neoverse/N3-neon-instructions.s b/llvm/test/tools/llvm-mca/AArch64/Neoverse/N3-neon-instructions.s
index dddaca34f68dd..9ef4dd9877a14 100644
--- a/llvm/test/tools/llvm-mca/AArch64/Neoverse/N3-neon-instructions.s
+++ b/llvm/test/tools/llvm-mca/AArch64/Neoverse/N3-neon-instructions.s
@@ -1445,8 +1445,8 @@ zip2 v0.8h, v0.8h, v0.8h
# CHECK-NEXT: 2 5 1.00 mov v0.h[1], w8
# CHECK-NEXT: 2 5 1.00 mov v0.s[2], w8
# CHECK-NEXT: 2 5 1.00 mov v0.d[1], x8
-# CHECK-NEXT: 1 2 0.50 mov v0.16b, v0.16b
-# CHECK-NEXT: 1 2 0.50 mov v0.8b, v0.8b
+# CHECK-NEXT: 1 0 0.20 mov v0.16b, v0.16b
+# CHECK-NEXT: 1 0 0.20 mov v0.8b, v0.8b
# CHECK-NEXT: 1 2 0.50 movi d15, #0xff00ff00ff00ff
# CHECK-NEXT: 1 2 0.50 movi v0.16b, #31
# CHECK-NEXT: 1 2 0.50 movi v0.2d, #0xff0000ff0000ffff
@@ -1467,7 +1467,7 @@ zip2 v0.8h, v0.8h, v0.8h
# CHECK-NEXT: 1 2 0.50 mvn v0.16b, v0.16b
# CHECK-NEXT: 1 2 0.50 mvn v0.8b, v0.8b
# CHECK-NEXT: 1 2 0.50 orn v0.16b, v0.16b, v0.16b
-# CHECK-NEXT: 1 2 0.50 mov v0.16b, v0.16b
+# CHECK-NEXT: 1 0 0.20 mov v0.16b, v0.16b
# CHECK-NEXT: 1 2 0.50 orr v0.8h, #31
# CHECK-NEXT: 1 2 1.00 pmul v0.16b, v0.16b, v0.16b
# CHECK-NEXT: 1 2 1.00 pmul v0.8b, v0.8b, v0.8b
@@ -2163,7 +2163,7 @@ zip2 v0.8h, v0.8h, v0.8h
# CHECK: Resource pressure per iteration:
# CHECK-NEXT: [0.0] [0.1] [1.0] [1.1] [2] [3.0] [3.1] [4] [5] [6.0] [6.1] [7] [8]
-# CHECK-NEXT: - - - - 33.00 51.50 51.50 18.75 7.75 7.75 7.75 649.00 584.00
+# CHECK-NEXT: - - - - 33.00 51.50 51.50 18.75 7.75 7.75 7.75 647.50 582.50
# CHECK: Resource pressure by instruction:
# CHECK-NEXT: [0.0] [0.1] [1.0] [1.1] [2] [3.0] [3.1] [4] [5] [6.0] [6.1] [7] [8] Instructions:
@@ -2534,8 +2534,8 @@ zip2 v0.8h, v0.8h, v0.8h
# CHECK-NEXT: - - - - - - - 1.00 - - - 0.50 0.50 mov v0.h[1], w8
# CHECK-NEXT: - - - - - - - 1.00 - - - 0.50 0.50 mov v0.s[2], w8
# CHECK-NEXT: - - - - - - - 1.00 - - - 0.50 0.50 mov v0.d[1], x8
-# CHECK-NEXT: - - - - - - - - - - - 0.50 0.50 mov v0.16b, v0.16b
-# CHECK-NEXT: - - - - - - - - - - - 0.50 0.50 mov v0.8b, v0.8b
+# CHECK-NEXT: - - - - - - - - - - - - - mov v0.16b, v0.16b
+# CHECK-NEXT: - - - - - - - - - - - - - mov v0.8b, v0.8b
# CHECK-NEXT: - - - - - - - - - - - 0.50 0.50 movi d15, #0xff00ff00ff00ff
# CHECK-NEXT: - - - - - - - - - - - 0.50 0.50 movi v0.16b, #31
# CHECK-NEXT: - - - - - - - - - - - 0.50 0.50 movi v0.2d, #0xff0000ff0000ffff
@@ -2556,7 +2556,7 @@ zip2 v0.8h, v0.8h, v0.8h
# CHECK-NEXT: - - - - - - - - - - - 0.50 0.50 mvn v0.16b, v0.16b
# CHECK-NEXT: - - - - - - - - - - - 0.50 0.50 mvn v0.8b, v0.8b
# CHECK-NEXT: - - - - - - - - - - - 0.50 0.50 orn v0.16b, v0.16b, v0.16b
-# CHECK-NEXT: - - - - - - - - - - - 0.50 0.50 mov v0.16b, v0.16b
+# CHECK-NEXT: - - - - - - - - - - - - - mov v0.16b, v0.16b
# CHECK-NEXT: - - - - - - - - - - - 0.50 0.50 orr v0.8h, #31
# CHECK-NEXT: - - - - - - - - - - - 1.00 - pmul v0.16b, v0.16b, v0.16b
# CHECK-NEXT: - - - - - - - - - - - 1.00 - pmul v0.8b, v0.8b, v0.8b
diff --git a/llvm/test/tools/llvm-mca/AArch64/Neoverse/N3-sve-instructions.s b/llvm/test/tools/llvm-mca/AArch64/Neoverse/N3-sve-instructions.s
index 8977802be6bfc..3bc16e693e33f 100644
--- a/llvm/test/tools/llvm-mca/AArch64/Neoverse/N3-sve-instructions.s
+++ b/llvm/test/tools/llvm-mca/AArch64/Neoverse/N3-sve-instructions.s
@@ -4979,7 +4979,7 @@ zip2 z31.s, z31.s, z31.s
# CHECK-NEXT: 1 2 0.50 mov z0.d, p0/m, d0
# CHECK-NEXT: 2 5 1.00 mov z0.d, p0/m, x0
# CHECK-NEXT: 1 3 1.00 mov z0.d, x0
-# CHECK-NEXT: 1 2 0.50 mov z0.d, z0.d
+# CHECK-NEXT: 1 0 0.20 mov z0.d, z0.d
# CHECK-NEXT: 1 2 0.50 mov z0.h, #-256
# CHECK-NEXT: 1 2 0.50 mov z0.h, #-32768
# CHECK-NEXT: 1 2 0.50 mov z0.h, #0
@@ -5039,7 +5039,7 @@ zip2 z31.s, z31.s, z31.s
# CHECK-NEXT: 1 2 0.50 movprfx z31.d, p7/z, z6.d
# CHECK-NEXT: 2 5 1.00 mov z31.d, p7/m, sp
# CHECK-NEXT: 1 3 1.00 mov z31.d, sp
-# CHECK-NEXT: 1 2 0.50 mov z31.d, z0.d
+# CHECK-NEXT: 1 0 0.20 mov z31.d, z0.d
# CHECK-NEXT: 1 2 0.50 mov z31.d, z31.d[7]
# CHECK-NEXT: 1 2 0.50 mov z31.h, p15/m, z31.h
# CHECK-NEXT: 1 2 0.50 mov z31.h, p7/m, h31
@@ -5150,7 +5150,7 @@ zip2 z31.s, z31.s, z31.s
# CHECK-NEXT: 2 5 1.00 orv d0, p7, z31.d
# CHECK-NEXT: 2 5 1.00 orv h0, p7, z31.h
# CHECK-NEXT: 2 5 1.00 orv s0, p7, z31.s
-# CHECK-NEXT: 1 2 0.50 pfalse p15.b
+# CHECK-NEXT: 1 0 0.20 pfalse p15.b
# CHECK-NEXT: 1 2 0.50 pfirst p0.b, p15, p0.b
# CHECK-NEXT: 1 2 0.50 pfirst p15.b, p15, p15.b
# CHECK-NEXT: 1 2 1.00 pmul z0.b, z1.b, z2.b
@@ -5280,11 +5280,11 @@ zip2 z31.s, z31.s, z31.s
# CHECK-NEXT: 1 2 0.50 ptrue p0.d, pow2
# CHECK-NEXT: 1 2 0.50 ptrue p0.h, pow2
# CHECK-NEXT: 1 2 0.50 ptrue p0.s, pow2
-# CHECK-NEXT: 1 2 0.50 ptrue p15.b
-# CHECK-NEXT: 1 2 0.50 ptrue p15.d
-# CHECK-NEXT: 1 2 0.50 ptrue p15.h
-# CHECK-NEXT: 1 2 0.50 ptrue p15.s
-# CHECK-NEXT: 1 2 0.50 ptrue p7.s
+# CHECK-NEXT: 1 0 0.20 ptrue p15.b
+# CHECK-NEXT: 1 0 0.20 ptrue p15.d
+# CHECK-NEXT: 1 0 0.20 ptrue p15.h
+# CHECK-NEXT: 1 0 0.20 ptrue p15.s
+# CHECK-NEXT: 1 0 0.20 ptrue p7.s
# CHECK-NEXT: 1 2 0.50 ptrue p7.s, #14
# CHECK-NEXT: 1 2 0.50 ptrue p7.s, #15
# CHECK-NEXT: 1 2 0.50 ptrue p7.s, #16
@@ -5319,11 +5319,11 @@ zip2 z31.s, z31.s, z31.s
# CHECK-NEXT: 1 2 0.50 ptrues p0.d, pow2
# CHECK-NEXT: 1 2 0.50 ptrues p0.h, pow2
# CHECK-NEXT: 1 2 0.50 ptrues p0.s, pow2
-# CHECK-NEXT: 1 2 0.50 ptrues p15.b
-# CHECK-NEXT: 1 2 0.50 ptrues p15.d
-# CHECK-NEXT: 1 2 0.50 ptrues p15.h
-# CHECK-NEXT: 1 2 0.50 ptrues p15.s
-# CHECK-NEXT: 1 2 0.50 ptrues p7.s
+# CHECK-NEXT: 1 0 0.20 ptrues p15.b
+# CHECK-NEXT: 1 0 0.20 ptrues p15.d
+# CHECK-NEXT: 1 0 0.20 ptrues p15.h
+# CHECK-NEXT: 1 0 0.20 ptrues p15.s
+# CHECK-NEXT: 1 0 0.20 ptrues p7.s
# CHECK-NEXT: 1 2 0.50 ptrues p7.s, #14
# CHECK-NEXT: 1 2 0.50 ptrues p7.s, #15
# CHECK-NEXT: 1 2 0.50 ptrues p7.s, #16
@@ -6847,7 +6847,7 @@ zip2 z31.s, z31.s, z31.s
# CHECK: Resource pressure per iteration:
# CHECK-NEXT: [0.0] [0.1] [1.0] [1.1] [2] [3.0] [3.1] [4] [5] [6.0] [6.1] [7] [8]
-# CHECK-NEXT: - - - - 332.67 481.67 481.67 298.00 230.00 88.50 88.50 1558.00 1401.00
+# CHECK-NEXT: - - - - 332.67 481.67 481.67 292.50 224.50 88.50 88.50 1557.00 1400.00
# CHECK: Resource pressure by instruction:
# CHECK-NEXT: [0.0] [0.1] [1.0] [1.1] [2] [3.0] [3.1] [4] [5] [6.0] [6.1] [7] [8] Instructions:
@@ -8410,7 +8410,7 @@ zip2 z31.s, z31.s, z31.s
# CHECK-NEXT: - - - - - - - - - - - 0.50 0.50 mov z0.d, p0/m, d0
# CHECK-NEXT: - - - - - - - 1.00 - - - 0.50 0.50 mov z0.d, p0/m, x0
# CHECK-NEXT: - - - - - - - 1.00 - - - - - mov z0.d, x0
-# CHECK-NEXT: - - - - - - - - - - - 0.50 0.50 mov z0.d, z0.d
+# CHECK-NEXT: - - - - - - - - - - - - - mov z0.d, z0.d
# CHECK-NEXT: - - - - - - - - - - - 0.50 0.50 mov z0.h, #-256
# CHECK-NEXT: - - - - - - - - - - - 0.50 0.50 mov z0.h, #-32768
# CHECK-NEXT: - - - - - - - - - - - 0.50 0.50 mov z0.h, #0
@@ -8470,7 +8470,7 @@ zip2 z31.s, z31.s, z31.s
# CHECK-NEXT: - - - - - - - - - - - 0.50 0.50 movprfx z31.d, p7/z, z6.d
# CHECK-NEXT: - - - - - - - 1.00 - - - 0.50 0.50 mov z31.d, p7/m, sp
# CHECK-NEXT: - - - - - - - 1.00 - - - - - mov z31.d, sp
-# CHECK-NEXT: - - - - - - - - - - - 0.50 0.50 mov z31.d, z0.d
+# CHECK-NEXT: - - - - - - - - - - - - - mov z31.d, z0.d
# CHECK-NEXT: - - - - - - - - - - - 0.50 0.50 mov z31.d, z31.d[7]
# CHECK-NEXT: - - - - - - - - - - - 0.50 0.50 mov z31.h, p15/m, z31.h
# CHECK-NEXT: - - - - - - - - - - - 0.50 0.50 mov z31.h, p7/m, h31
@@ -8581,7 +8581,7 @@ zip2 z31.s, z31.s, z31.s
# CHECK-NEXT: - - - - - - - - - - - 0.50 1.50 orv d0, p7, z31.d
# CHECK-NEXT: - - - - - - - - - - - 0.50 1.50 orv h0, p7, z31.h
# CHECK-NEXT: - - - - - - - - - - - 0.50 1.50 orv s0, p7, z31.s
-# CHECK-NEXT: - - - - - - - 0.50 0.50 - - - - pfalse p15.b
+# CHECK-NEXT: - - - - - - - - - - - - - pfalse p15.b
# CHECK-NEXT: - - - - - - - 0.50 0.50 - - - - pfirst p0.b, p15, p0.b
# CHECK-NEXT: - - - - - - - 0.50 0.50 - - - - pfirst p15.b, p15, p15.b
# CHECK-NEXT: - - - - - - - - - - - 1.00 - pmul z0.b, z1.b, z2.b
@@ -8711,11 +8711,11 @@ zip2 z31.s, z31.s, z31.s
# CHECK-NEXT: - - - - - - - 0.50 0.50 - - - - ptrue p0.d, pow2
# CHECK-NEXT: - - - - - - - 0.50 0.50 - - - - ptrue p0.h, pow2
# CHECK-NEXT: - - - - - - - 0.50 0.50 - - - - ptrue p0.s, pow2
-# CHECK-NEXT: - - - - - - - 0.50 0.50 - - - - ptrue p15.b
-# CHECK-NEXT: - - - - - - - 0.50 0.50 - - - - ptrue p15.d
-# CHECK-NEXT: - - - - - - - 0.50 0.50 - - - - ptrue p15.h
-# CHECK-NEXT: - - - - - - - 0.50 0.50 - - - - ptrue p15.s
-# CHECK-NEXT: - - - - - - - 0.50 0.50 - - - - ptrue p7.s
+# CHECK-NEXT: - - - - - - - - - - - - - ptrue p15.b
+# CHECK-NEXT: - - - - - - - - - - - - - ptrue p15.d
+# CHECK-NEXT: - - - - - - - - - - - - - ptrue p15.h
+# CHECK-NEXT: - - - - - - - - - - - - - ptrue p15.s
+# CHECK-NEXT: - - - - - - - - - - - - - ptrue p7.s
# CHECK-NEXT: - - - - - - - 0.50 0.50 - - - - ptrue p7.s, #14
# CHECK-NEXT: - - - - - - - 0.50 0.50 - - - - ptrue p7.s, #15
# CHECK-NEXT: - - - - - - - 0.50 0.50 - - - - ptrue p7.s, #16
@@ -8750,11 +8750,11 @@ zip2 z31.s, z31.s, z31.s
# CHECK-NEXT: - - - - - - - 0.50 0.50 - - - - ptrues p0.d, pow2
# CHECK-NEXT: - - - - - - - 0.50 0.50 - - - - ptrues p0.h, pow2
# CHECK-NEXT: - - - - - - - 0.50 0.50 - - - - ptrues p0.s, pow2
-# CHECK-NEXT: - - - - - - - 0.50 0.50 - - - - ptrues p15.b
-# CHECK-NEXT: - - - - - - - 0.50 0.50 - - - - ptrues p15.d
-# CHECK-NEXT: - - - - - - - 0.50 0.50 - - - - ptrues p15.h
-# CHECK-NEXT: - - - - - - - 0.50 0.50 - - - - ptrues p15.s
-# CHECK-NEXT: - - - - - - - 0.50 0.50 - - - - ptrues p7.s
+# CHECK-NEXT: - - - - - - - - - - - - - ptrues p15.b
+# CHECK-NEXT: - - - - - - - - - - - - - ptrues p15.d
+# CHECK-NEXT: - - - - - - - - - - - - - ptrues p15.h
+# CHECK-NEXT: - - - - - - - - - - - - - ptrues p15.s
+# CHECK-NEXT: - - - - - - - - - - - - - ptrues p7.s
# CHECK-NEXT: - - - - - - - 0.50 0.50 - - - - ptrues p7.s, #14
# CHECK-NEXT: - - - - - - - 0.50 0.50 - - - - ptrues p7.s, #15
# CHECK-NEXT: - - - - - - - 0.50 0.50 - - - - ptrues p7.s, #16
>From 54fc6b557dfe7d1a3bdd77682af0641787dac9ce Mon Sep 17 00:00:00 2001
From: Simon Wallis <simon.wallis2 at arm.com>
Date: Tue, 18 Nov 2025 11:12:32 +0000
Subject: [PATCH 7/7] [AArch64] Update zero latency instructions in Neoverse
scheduling tables
NeoverseZeroMove was introduced for Neoverse-V2 and was added to V3 and V3AE.
Use NeoverseZeroMove for Neoverse-V1, N2, N3 in the same way, including these instructions:
MOV Xd|Wd, #0|XZR|WZR
For all Neoverse targets, the following instructions are also decoded as not utilizing the scheduling and execution resources of the machine:
MOV Wd,Wn
MOV Xd,Xn
For Neoverse-N3 only, these instructions also have zero latency
FMOV Dd, Dn
FMOV Sd, Sn
Change-Id: I2d51b0ee6736d14f8212583f234431c555cc2574
---
.../Target/AArch64/AArch64SchedNeoverseN3.td | 10 +--
.../Target/AArch64/AArch64SchedNeoverseV1.td | 7 +-
.../AArch64/AArch64SchedPredNeoverse.td | 21 +++---
.../AArch64/Neoverse/V1-basic-instructions.s | 66 +++++++++----------
.../llvm-mca/AArch64/Neoverse/V1-writeback.s | 30 ++++-----
5 files changed, 68 insertions(+), 66 deletions(-)
diff --git a/llvm/lib/Target/AArch64/AArch64SchedNeoverseN3.td b/llvm/lib/Target/AArch64/AArch64SchedNeoverseN3.td
index 1f78ebb57da5e..f195c83344c6a 100644
--- a/llvm/lib/Target/AArch64/AArch64SchedNeoverseN3.td
+++ b/llvm/lib/Target/AArch64/AArch64SchedNeoverseN3.td
@@ -566,8 +566,8 @@ def N3Write_0or2c_1V : SchedWriteVariant<[
SchedVar<NoSchedPred, [N3Write_2c_1V]>]>;
def N3Write_0or2c_1M : SchedWriteVariant<[
- SchedVar<NeoverseAllElement, [N3Write_0c]>,
- SchedVar<NoSchedPred, [N3Write_2c_1M]>]>;
+ SchedVar<NeoverseAllActivePredicate, [N3Write_0c]>,
+ SchedVar<NoSchedPred, [N3Write_2c_1M]>]>;
def N3Write_0or3c_1M0 : SchedWriteVariant<[
SchedVar<NeoverseZeroMove, [N3Write_0c]>,
@@ -985,7 +985,7 @@ def : InstRW<[WriteAdr, N3Write_2c_1L01_1V_1I], (instregex "^STP[SDQ](post|pre)$
// ASIMD compare
// ASIMD logical
// ASIMD max/min, basic and pair-wise
-def : InstRW<[N3Write_0or2c_1V], (instregex "^ORRv16i8", "^ORRv8i8")>;
+def : InstRW<[N3Write_0or2c_1V], (instrs ORRv16i8, ORRv8i8)>;
def : SchedAlias<WriteVd, N3Write_2c_1V>;
def : SchedAlias<WriteVq, N3Write_2c_1V>;
@@ -1590,7 +1590,7 @@ def : InstRW<[N3Write_2c_1M], (instregex "^REV_PP_[BHSD]")>;
def : InstRW<[N3Write_1c_1M], (instrs SEL_PPPP)>;
// Predicate set
-def : InstRW<[N3Write_0c], (instregex "^PFALSE")>;
+def : InstRW<[N3Write_0c], (instrs PFALSE)>;
def : InstRW<[N3Write_0or2c_1M], (instregex "^PTRUE_[BHSD]")>;
// Predicate set/initialize, set flags
@@ -1821,7 +1821,7 @@ def : InstRW<[N3Write_2c_1V],
"^EOR(BT|TB)_ZZZ_[BHSD]",
"^(AND|BIC|EOR|NOT|ORR)_(ZPmZ|ZPZZ)_[BHSD]",
"^NOT_ZPmZ_[BHSD]")>;
-def : InstRW<[N3Write_0or2c_1V], (instregex "^ORR_ZZZ")>;
+def : InstRW<[N3Write_0or2c_1V], (instrs ORR_ZZZ)>;
// Max/min, basic and pairwise
def : InstRW<[N3Write_2c_1V], (instregex "^[SU](MAX|MIN)_ZI_[BHSD]",
diff --git a/llvm/lib/Target/AArch64/AArch64SchedNeoverseV1.td b/llvm/lib/Target/AArch64/AArch64SchedNeoverseV1.td
index a880c1a034598..bf65b31f88037 100644
--- a/llvm/lib/Target/AArch64/AArch64SchedNeoverseV1.td
+++ b/llvm/lib/Target/AArch64/AArch64SchedNeoverseV1.td
@@ -91,9 +91,10 @@ def : WriteRes<WriteHint, []> { let Latency = 1; }
//===----------------------------------------------------------------------===//
// Define generic 0 micro-op types
-let Latency = 0, NumMicroOps = 1 in
+let Latency = 0, NumMicroOps = 0 in
def V1Write_0c_0Z : SchedWriteRes<[]>;
+def V1Write_0c : SchedWriteRes<[]> { let Latency = 0; }
//===----------------------------------------------------------------------===//
// Define generic 1 micro-op types
@@ -476,11 +477,11 @@ def V1Write_11c_9L01_9S_9V : SchedWriteRes<[V1UnitL01, V1UnitL01, V1UnitL01,
// Define predicate-controlled types
def V1Write_0or1c_1I : SchedWriteVariant<[
- SchedVar<NeoverseZeroMove, [V1Write_0c_0Z]>,
+ SchedVar<NeoverseZeroMove, [V1Write_0c]>,
SchedVar<NoSchedPred, [V1Write_1c_1I]>]>;
def V1Write_0or3c_1M0 : SchedWriteVariant<[
- SchedVar<NeoverseZeroMove, [V1Write_0c_0Z]>,
+ SchedVar<NeoverseZeroMove, [V1Write_0c]>,
SchedVar<NoSchedPred, [V1Write_3c_1M0]>]>;
//===----------------------------------------------------------------------===//
diff --git a/llvm/lib/Target/AArch64/AArch64SchedPredNeoverse.td b/llvm/lib/Target/AArch64/AArch64SchedPredNeoverse.td
index 2d330de16280b..f841e6072d2b4 100644
--- a/llvm/lib/Target/AArch64/AArch64SchedPredNeoverse.td
+++ b/llvm/lib/Target/AArch64/AArch64SchedPredNeoverse.td
@@ -89,13 +89,14 @@ def NeoverseZeroMove : MCSchedPredicate<
CheckSameRegOperand<1, 2>]>,
]>>;
-def NeoverseAllElement : MCSchedPredicate<
- CheckAny<[
- // PTRUE Pd, ALL
- // PTRUES Pd, ALL
- CheckAll<[CheckOpcode<[
- PTRUE_B, PTRUE_H, PTRUE_S, PTRUE_D,
- PTRUES_B, PTRUES_H, PTRUES_S, PTRUES_D]>,
- CheckIsImmOperand<1>,
- CheckImmOperand<1, 31>]>,
- ]>>;
+def NeoverseAllActivePredicate : MCSchedPredicate<
+ CheckAny<[
+ // PTRUE Pd, ALL
+ // PTRUES Pd, ALL
+ CheckAll<[
+ CheckOpcode<[
+ PTRUE_B, PTRUE_H, PTRUE_S, PTRUE_D,
+ PTRUES_B, PTRUES_H, PTRUES_S, PTRUES_D]>,
+ CheckIsImmOperand<1>,
+ CheckImmOperand<1, 31>]>,
+ ]>>;
diff --git a/llvm/test/tools/llvm-mca/AArch64/Neoverse/V1-basic-instructions.s b/llvm/test/tools/llvm-mca/AArch64/Neoverse/V1-basic-instructions.s
index eaf128acc0195..787acbe91c057 100644
--- a/llvm/test/tools/llvm-mca/AArch64/Neoverse/V1-basic-instructions.s
+++ b/llvm/test/tools/llvm-mca/AArch64/Neoverse/V1-basic-instructions.s
@@ -2562,78 +2562,78 @@ drps
# CHECK-NEXT: 2 7 0.33 * ldr q17, [x23, w9, sxtw]
# CHECK-NEXT: 3 2 0.50 * str q18, [x22, w10, sxtw]
# CHECK-NEXT: 2 7 0.33 * ldr q19, [x21, wzr, sxtw #4]
-# CHECK-NEXT: 2 4 0.33 * ldp w3, w5, [sp]
+# CHECK-NEXT: 1 4 0.33 * ldp w3, w5, [sp]
# CHECK-NEXT: 2 1 0.50 * stp wzr, w9, [sp, #252]
-# CHECK-NEXT: 2 4 0.33 * ldp w2, wzr, [sp, #-256]
-# CHECK-NEXT: 2 4 0.33 * ldp w9, w10, [sp, #4]
-# CHECK-NEXT: 3 5 0.33 * ldpsw x9, x10, [sp, #4]
-# CHECK-NEXT: 3 5 0.33 * ldpsw x9, x10, [x2, #-256]
-# CHECK-NEXT: 3 5 0.33 * ldpsw x20, x30, [sp, #252]
+# CHECK-NEXT: 1 4 0.33 * ldp w2, wzr, [sp, #-256]
+# CHECK-NEXT: 1 4 0.33 * ldp w9, w10, [sp, #4]
+# CHECK-NEXT: 2 5 0.33 * ldpsw x9, x10, [sp, #4]
+# CHECK-NEXT: 2 5 0.33 * ldpsw x9, x10, [x2, #-256]
+# CHECK-NEXT: 2 5 0.33 * ldpsw x20, x30, [sp, #252]
# CHECK-NEXT: 2 4 0.67 * ldp x21, x29, [x2, #504]
# CHECK-NEXT: 2 4 0.67 * ldp x22, x23, [x3, #-512]
# CHECK-NEXT: 2 4 0.67 * ldp x24, x25, [x4, #8]
-# CHECK-NEXT: 2 6 0.33 * ldp s29, s28, [sp, #252]
+# CHECK-NEXT: 1 6 0.33 * ldp s29, s28, [sp, #252]
# CHECK-NEXT: 2 2 0.50 * stp s27, s26, [sp, #-256]
-# CHECK-NEXT: 2 6 0.33 * ldp s1, s2, [x3, #44]
+# CHECK-NEXT: 1 6 0.33 * ldp s1, s2, [x3, #44]
# CHECK-NEXT: 2 2 0.50 * stp d3, d5, [x9, #504]
# CHECK-NEXT: 2 2 0.50 * stp d7, d11, [x10, #-512]
# CHECK-NEXT: 2 1 0.50 * stnp x20, x16, [x8]
# CHECK-NEXT: 2 1 0.50 * stp x3, x6, [x16]
-# CHECK-NEXT: 2 6 0.33 * ldp d2, d3, [x30, #-8]
+# CHECK-NEXT: 1 6 0.33 * ldp d2, d3, [x30, #-8]
# CHECK-NEXT: 2 2 0.50 * stp q3, q5, [sp]
# CHECK-NEXT: 2 2 0.50 * stp q17, q19, [sp, #1008]
# CHECK-NEXT: 2 6 0.67 * ldp q23, q29, [x1, #-1024]
-# CHECK-NEXT: 3 4 0.33 * ldp w3, w5, [sp], #0
+# CHECK-NEXT: 2 4 0.33 * ldp w3, w5, [sp], #0
# CHECK-NEXT: 3 1 0.50 * stp wzr, w9, [sp], #252
-# CHECK-NEXT: 3 4 0.33 * ldp w2, wzr, [sp], #-256
-# CHECK-NEXT: 3 4 0.33 * ldp w9, w10, [sp], #4
-# CHECK-NEXT: 4 5 0.50 * ldpsw x9, x10, [sp], #4
-# CHECK-NEXT: 4 5 0.50 * ldpsw x9, x10, [x2], #-256
-# CHECK-NEXT: 4 5 0.50 * ldpsw x20, x30, [sp], #252
+# CHECK-NEXT: 2 4 0.33 * ldp w2, wzr, [sp], #-256
+# CHECK-NEXT: 2 4 0.33 * ldp w9, w10, [sp], #4
+# CHECK-NEXT: 3 5 0.50 * ldpsw x9, x10, [sp], #4
+# CHECK-NEXT: 3 5 0.50 * ldpsw x9, x10, [x2], #-256
+# CHECK-NEXT: 3 5 0.50 * ldpsw x20, x30, [sp], #252
# CHECK-NEXT: 3 4 0.67 * ldp x21, x29, [x2], #504
# CHECK-NEXT: 3 4 0.67 * ldp x22, x23, [x3], #-512
# CHECK-NEXT: 3 4 0.67 * ldp x24, x25, [x4], #8
-# CHECK-NEXT: 3 6 0.33 * ldp s29, s28, [sp], #252
+# CHECK-NEXT: 2 6 0.33 * ldp s29, s28, [sp], #252
# CHECK-NEXT: 3 2 0.50 * stp s27, s26, [sp], #-256
-# CHECK-NEXT: 3 6 0.33 * ldp s1, s2, [x3], #44
+# CHECK-NEXT: 2 6 0.33 * ldp s1, s2, [x3], #44
# CHECK-NEXT: 3 2 0.50 * stp d3, d5, [x9], #504
# CHECK-NEXT: 3 2 0.50 * stp d7, d11, [x10], #-512
-# CHECK-NEXT: 3 6 0.33 * ldp d2, d3, [x30], #-8
+# CHECK-NEXT: 2 6 0.33 * ldp d2, d3, [x30], #-8
# CHECK-NEXT: 4 2 1.00 * stp q3, q5, [sp], #0
# CHECK-NEXT: 4 2 1.00 * stp q17, q19, [sp], #1008
# CHECK-NEXT: 3 6 0.67 * ldp q23, q29, [x1], #-1024
-# CHECK-NEXT: 3 4 0.33 * ldp w3, w5, [sp, #0]!
+# CHECK-NEXT: 2 4 0.33 * ldp w3, w5, [sp, #0]!
# CHECK-NEXT: 3 1 0.50 * stp wzr, w9, [sp, #252]!
-# CHECK-NEXT: 3 4 0.33 * ldp w2, wzr, [sp, #-256]!
-# CHECK-NEXT: 3 4 0.33 * ldp w9, w10, [sp, #4]!
-# CHECK-NEXT: 4 5 0.50 * ldpsw x9, x10, [sp, #4]!
-# CHECK-NEXT: 4 5 0.50 * ldpsw x9, x10, [x2, #-256]!
-# CHECK-NEXT: 4 5 0.50 * ldpsw x20, x30, [sp, #252]!
+# CHECK-NEXT: 2 4 0.33 * ldp w2, wzr, [sp, #-256]!
+# CHECK-NEXT: 2 4 0.33 * ldp w9, w10, [sp, #4]!
+# CHECK-NEXT: 3 5 0.50 * ldpsw x9, x10, [sp, #4]!
+# CHECK-NEXT: 3 5 0.50 * ldpsw x9, x10, [x2, #-256]!
+# CHECK-NEXT: 3 5 0.50 * ldpsw x20, x30, [sp, #252]!
# CHECK-NEXT: 3 4 0.67 * ldp x21, x29, [x2, #504]!
# CHECK-NEXT: 3 4 0.67 * ldp x22, x23, [x3, #-512]!
# CHECK-NEXT: 3 4 0.67 * ldp x24, x25, [x4, #8]!
-# CHECK-NEXT: 3 6 0.33 * ldp s29, s28, [sp, #252]!
+# CHECK-NEXT: 2 6 0.33 * ldp s29, s28, [sp, #252]!
# CHECK-NEXT: 3 2 0.50 * stp s27, s26, [sp, #-256]!
-# CHECK-NEXT: 3 6 0.33 * ldp s1, s2, [x3, #44]!
+# CHECK-NEXT: 2 6 0.33 * ldp s1, s2, [x3, #44]!
# CHECK-NEXT: 3 2 0.50 * stp d3, d5, [x9, #504]!
# CHECK-NEXT: 3 2 0.50 * stp d7, d11, [x10, #-512]!
-# CHECK-NEXT: 3 6 0.33 * ldp d2, d3, [x30, #-8]!
+# CHECK-NEXT: 2 6 0.33 * ldp d2, d3, [x30, #-8]!
# CHECK-NEXT: 4 2 1.00 * stp q3, q5, [sp, #0]!
# CHECK-NEXT: 4 2 1.00 * stp q17, q19, [sp, #1008]!
# CHECK-NEXT: 3 6 0.67 * ldp q23, q29, [x1, #-1024]!
-# CHECK-NEXT: 2 4 0.33 * ldnp w3, w5, [sp]
+# CHECK-NEXT: 1 4 0.33 * ldnp w3, w5, [sp]
# CHECK-NEXT: 2 1 0.50 * stnp wzr, w9, [sp, #252]
-# CHECK-NEXT: 2 4 0.33 * ldnp w2, wzr, [sp, #-256]
-# CHECK-NEXT: 2 4 0.33 * ldnp w9, w10, [sp, #4]
+# CHECK-NEXT: 1 4 0.33 * ldnp w2, wzr, [sp, #-256]
+# CHECK-NEXT: 1 4 0.33 * ldnp w9, w10, [sp, #4]
# CHECK-NEXT: 2 4 0.67 * ldnp x21, x29, [x2, #504]
# CHECK-NEXT: 2 4 0.67 * ldnp x22, x23, [x3, #-512]
# CHECK-NEXT: 2 4 0.67 * ldnp x24, x25, [x4, #8]
-# CHECK-NEXT: 2 6 0.33 * ldnp s29, s28, [sp, #252]
+# CHECK-NEXT: 1 6 0.33 * ldnp s29, s28, [sp, #252]
# CHECK-NEXT: 2 2 0.50 * stnp s27, s26, [sp, #-256]
-# CHECK-NEXT: 2 6 0.33 * ldnp s1, s2, [x3, #44]
+# CHECK-NEXT: 1 6 0.33 * ldnp s1, s2, [x3, #44]
# CHECK-NEXT: 2 2 0.50 * stnp d3, d5, [x9, #504]
# CHECK-NEXT: 2 2 0.50 * stnp d7, d11, [x10, #-512]
-# CHECK-NEXT: 2 6 0.33 * ldnp d2, d3, [x30, #-8]
+# CHECK-NEXT: 1 6 0.33 * ldnp d2, d3, [x30, #-8]
# CHECK-NEXT: 2 2 0.50 * stnp q3, q5, [sp]
# CHECK-NEXT: 2 2 0.50 * stnp q17, q19, [sp, #1008]
# CHECK-NEXT: 2 6 0.67 * ldnp q23, q29, [x1, #-1024]
diff --git a/llvm/test/tools/llvm-mca/AArch64/Neoverse/V1-writeback.s b/llvm/test/tools/llvm-mca/AArch64/Neoverse/V1-writeback.s
index fee62ce565b43..5efe3d0bbf14f 100644
--- a/llvm/test/tools/llvm-mca/AArch64/Neoverse/V1-writeback.s
+++ b/llvm/test/tools/llvm-mca/AArch64/Neoverse/V1-writeback.s
@@ -3140,13 +3140,13 @@ add x0, x27, 1
# CHECK: Iterations: 100
# CHECK-NEXT: Instructions: 1000
-# CHECK-NEXT: Total Cycles: 708
-# CHECK-NEXT: Total uOps: 3500
+# CHECK-NEXT: Total Cycles: 609
+# CHECK-NEXT: Total uOps: 3300
# CHECK: Dispatch Width: 8
-# CHECK-NEXT: uOps Per Cycle: 4.94
-# CHECK-NEXT: IPC: 1.41
-# CHECK-NEXT: Block RThroughput: 4.4
+# CHECK-NEXT: uOps Per Cycle: 5.42
+# CHECK-NEXT: IPC: 1.64
+# CHECK-NEXT: Block RThroughput: 4.1
# CHECK: Timeline view:
# CHECK-NEXT: 01234
@@ -3161,7 +3161,7 @@ add x0, x27, 1
# CHECK-NEXT: [0,6] . DeeeeeeE-R ldp s1, s2, [x27], #248
# CHECK-NEXT: [0,7] . D=eE-----R add x0, x27, #1
# CHECK-NEXT: [0,8] . D=eeeeeeER ldp d1, d2, [x27], #496
-# CHECK-NEXT: [0,9] . .D=eE----R add x0, x27, #1
+# CHECK-NEXT: [0,9] . D==eE----R add x0, x27, #1
# CHECK: Average Wait times (based on the timeline view):
# CHECK-NEXT: [0]: Executions
@@ -3179,18 +3179,18 @@ add x0, x27, 1
# CHECK-NEXT: 6. 1 1.0 0.0 1.0 ldp s1, s2, [x27], #248
# CHECK-NEXT: 7. 1 2.0 0.0 5.0 add x0, x27, #1
# CHECK-NEXT: 8. 1 2.0 0.0 0.0 ldp d1, d2, [x27], #496
-# CHECK-NEXT: 9. 1 2.0 0.0 4.0 add x0, x27, #1
-# CHECK-NEXT: 1 1.3 0.3 2.8 <total>
+# CHECK-NEXT: 9. 1 3.0 0.0 4.0 add x0, x27, #1
+# CHECK-NEXT: 1 1.4 0.3 2.8 <total>
# CHECK: [44] Code Region - G45
# CHECK: Iterations: 100
# CHECK-NEXT: Instructions: 1000
# CHECK-NEXT: Total Cycles: 507
-# CHECK-NEXT: Total uOps: 2000
+# CHECK-NEXT: Total uOps: 1700
# CHECK: Dispatch Width: 8
-# CHECK-NEXT: uOps Per Cycle: 3.94
+# CHECK-NEXT: uOps Per Cycle: 3.35
# CHECK-NEXT: IPC: 1.97
# CHECK-NEXT: Block RThroughput: 2.5
@@ -3233,10 +3233,10 @@ add x0, x27, 1
# CHECK: Iterations: 100
# CHECK-NEXT: Instructions: 1000
# CHECK-NEXT: Total Cycles: 507
-# CHECK-NEXT: Total uOps: 2200
+# CHECK-NEXT: Total uOps: 1900
# CHECK: Dispatch Width: 8
-# CHECK-NEXT: uOps Per Cycle: 4.34
+# CHECK-NEXT: uOps Per Cycle: 3.75
# CHECK-NEXT: IPC: 1.97
# CHECK-NEXT: Block RThroughput: 3.0
@@ -3251,7 +3251,7 @@ add x0, x27, 1
# CHECK-NEXT: [0,4] .D=eeeeER .. ldp x1, x2, [x27, #496]!
# CHECK-NEXT: [0,5] .D==eE--R .. add x0, x27, #1
# CHECK-NEXT: [0,6] .D==eeeeeER. ldpsw x1, x2, [x27], #248
-# CHECK-NEXT: [0,7] . D==eE---R. add x0, x27, #1
+# CHECK-NEXT: [0,7] .D===eE---R. add x0, x27, #1
# CHECK-NEXT: [0,8] . D==eeeeeER ldpsw x1, x2, [x27, #248]!
# CHECK-NEXT: [0,9] . D===eE---R add x0, x27, #1
@@ -3269,10 +3269,10 @@ add x0, x27, 1
# CHECK-NEXT: 4. 1 2.0 0.0 0.0 ldp x1, x2, [x27, #496]!
# CHECK-NEXT: 5. 1 3.0 0.0 2.0 add x0, x27, #1
# CHECK-NEXT: 6. 1 3.0 0.0 0.0 ldpsw x1, x2, [x27], #248
-# CHECK-NEXT: 7. 1 3.0 0.0 3.0 add x0, x27, #1
+# CHECK-NEXT: 7. 1 4.0 0.0 3.0 add x0, x27, #1
# CHECK-NEXT: 8. 1 3.0 0.0 0.0 ldpsw x1, x2, [x27, #248]!
# CHECK-NEXT: 9. 1 4.0 0.0 3.0 add x0, x27, #1
-# CHECK-NEXT: 1 2.6 0.1 1.2 <total>
+# CHECK-NEXT: 1 2.7 0.1 1.2 <total>
# CHECK: [46] Code Region - G47
More information about the llvm-commits
mailing list