[llvm] [MCA][AArch64] Fixes Neoverse N2 pair-load RThroughput (PR #171853)
via llvm-commits
llvm-commits at lists.llvm.org
Thu Dec 11 08:57:32 PST 2025
llvmbot wrote:
<!--LLVM PR SUMMARY COMMENT-->
@llvm/pr-subscribers-backend-aarch64
Author: Amina Chabane (Amichaxx)
<details>
<summary>Changes</summary>
As per the SWOG for [Neoverse N2](https://developer.arm.com/documentation/109914/latest/), throughput is 1.5.
- Treat LDPXi/LDNPXi/LDPXpost/LDPXpre as dual loads so they now reserve two N2Write_4c_1L
resources
- Update relevant tests to reflect new RThroughput
---
Patch is 29.21 KiB, truncated to 20.00 KiB below, full version: https://github.com/llvm/llvm-project/pull/171853.diff
3 Files Affected:
- (modified) llvm/lib/Target/AArch64/AArch64SchedNeoverseN2.td (+9)
- (modified) llvm/test/tools/llvm-mca/AArch64/Neoverse/N2-basic-instructions.s (+25-25)
- (modified) llvm/test/tools/llvm-mca/AArch64/Neoverse/N2-writeback.s (+54-54)
``````````diff
diff --git a/llvm/lib/Target/AArch64/AArch64SchedNeoverseN2.td b/llvm/lib/Target/AArch64/AArch64SchedNeoverseN2.td
index a02130f8390a7..1ab314c779e79 100644
--- a/llvm/lib/Target/AArch64/AArch64SchedNeoverseN2.td
+++ b/llvm/lib/Target/AArch64/AArch64SchedNeoverseN2.td
@@ -796,6 +796,15 @@ def : InstRW<[N2Write_2c_1M], (instregex "^BFM[WX]ri$")>;
def : SchedAlias<WriteLD, N2Write_4c_1L>;
def : SchedAlias<WriteLDIdx, N2Write_4c_1I_1L>;
+// Load pair, signed immed offset, normal, X-form
+def : InstRW<[N2Write_4c_1L, N2Write_4c_1L, WriteLDHi],
+ (instrs LDPXi, LDNPXi)>;
+
+// Load pair, immed post-index, normal, X-form
+// Load pair, immed pre-index, normal, X-form
+def : InstRW<[WriteAdr, N2Write_4c_1L, N2Write_4c_1L, WriteLDHi],
+ (instregex "^LDPX(post|pre)$")>;
+
// Load pair, signed immed offset, signed words
def : InstRW<[N2Write_5c_1M0, WriteLDHi], (instrs LDPSWi)>;
// Load pair, immed post-index or immed pre-index, signed words
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 18c853c2427a6..92cc68940f6b5 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
@@ -1110,9 +1110,9 @@
# CHECK-NEXT: 2 5 5.00 * ldpsw x9, x10, [sp, #4]
# CHECK-NEXT: 2 5 5.00 * ldpsw x9, x10, [x2, #-256]
# CHECK-NEXT: 2 5 5.00 * ldpsw x20, x30, [sp, #252]
-# CHECK-NEXT: 2 4 0.33 * ldp x21, x29, [x2, #504]
-# CHECK-NEXT: 2 4 0.33 * ldp x22, x23, [x3, #-512]
-# CHECK-NEXT: 2 4 0.33 * ldp x24, x25, [x4, #8]
+# 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: 2 2 0.50 * stp s27, s26, [sp, #-256]
# CHECK-NEXT: 2 6 0.33 * ldp s1, s2, [x3, #44]
@@ -1131,9 +1131,9 @@
# CHECK-NEXT: 3 5 5.00 * ldpsw x9, x10, [sp], #4
# CHECK-NEXT: 3 5 5.00 * ldpsw x9, x10, [x2], #-256
# CHECK-NEXT: 3 5 5.00 * ldpsw x20, x30, [sp], #252
-# CHECK-NEXT: 3 4 0.33 * ldp x21, x29, [x2], #504
-# CHECK-NEXT: 3 4 0.33 * ldp x22, x23, [x3], #-512
-# CHECK-NEXT: 3 4 0.33 * ldp x24, x25, [x4], #8
+# CHECK-NEXT: 4 4 0.67 * ldp x21, x29, [x2], #504
+# CHECK-NEXT: 4 4 0.67 * ldp x22, x23, [x3], #-512
+# CHECK-NEXT: 4 4 0.67 * ldp x24, x25, [x4], #8
# CHECK-NEXT: 4 6 0.50 * ldp s29, s28, [sp], #252
# CHECK-NEXT: 4 2 0.50 * stp s27, s26, [sp], #-256
# CHECK-NEXT: 4 6 0.50 * ldp s1, s2, [x3], #44
@@ -1150,9 +1150,9 @@
# CHECK-NEXT: 3 5 5.00 * ldpsw x9, x10, [sp, #4]!
# CHECK-NEXT: 3 5 5.00 * ldpsw x9, x10, [x2, #-256]!
# CHECK-NEXT: 3 5 5.00 * ldpsw x20, x30, [sp, #252]!
-# CHECK-NEXT: 3 4 0.33 * ldp x21, x29, [x2, #504]!
-# CHECK-NEXT: 3 4 0.33 * ldp x22, x23, [x3, #-512]!
-# CHECK-NEXT: 3 4 0.33 * ldp x24, x25, [x4, #8]!
+# CHECK-NEXT: 4 4 0.67 * ldp x21, x29, [x2, #504]!
+# CHECK-NEXT: 4 4 0.67 * ldp x22, x23, [x3, #-512]!
+# CHECK-NEXT: 4 4 0.67 * ldp x24, x25, [x4, #8]!
# CHECK-NEXT: 4 6 0.50 * ldp s29, s28, [sp, #252]!
# CHECK-NEXT: 4 2 0.50 * stp s27, s26, [sp, #-256]!
# CHECK-NEXT: 4 6 0.50 * ldp s1, s2, [x3, #44]!
@@ -1166,9 +1166,9 @@
# 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: 2 4 0.33 * ldnp x21, x29, [x2, #504]
-# CHECK-NEXT: 2 4 0.33 * ldnp x22, x23, [x3, #-512]
-# CHECK-NEXT: 2 4 0.33 * ldnp x24, x25, [x4, #8]
+# CHECK-NEXT: 3 4 0.67 * ldnp x21, x29, [x2, #504]
+# CHECK-NEXT: 3 4 0.67 * ldnp x22, x23, [x3, #-512]
+# CHECK-NEXT: 3 4 0.67 * ldnp x24, x25, [x4, #8]
# CHECK-NEXT: 2 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]
@@ -1267,7 +1267,7 @@
# 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: 13.00 13.00 34.00 34.00 90.33 161.33 161.33 539.50 253.50 169.00 169.00 254.00 101.00
+# CHECK-NEXT: 13.00 13.00 34.00 34.00 94.33 165.33 165.33 539.50 253.50 169.00 169.00 254.00 101.00
# 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:
@@ -2371,9 +2371,9 @@
# CHECK-NEXT: - - - - - - - 5.00 - - - - - ldpsw x9, x10, [sp, #4]
# CHECK-NEXT: - - - - - - - 5.00 - - - - - ldpsw x9, x10, [x2, #-256]
# CHECK-NEXT: - - - - - - - 5.00 - - - - - ldpsw x20, x30, [sp, #252]
-# CHECK-NEXT: - - - - 0.33 0.33 0.33 - - - - - - ldp x21, x29, [x2, #504]
-# CHECK-NEXT: - - - - 0.33 0.33 0.33 - - - - - - ldp x22, x23, [x3, #-512]
-# CHECK-NEXT: - - - - 0.33 0.33 0.33 - - - - - - ldp x24, x25, [x4, #8]
+# CHECK-NEXT: - - - - 0.67 0.67 0.67 - - - - - - ldp x21, x29, [x2, #504]
+# CHECK-NEXT: - - - - 0.67 0.67 0.67 - - - - - - ldp x22, x23, [x3, #-512]
+# CHECK-NEXT: - - - - 0.67 0.67 0.67 - - - - - - ldp x24, x25, [x4, #8]
# CHECK-NEXT: - - - - 0.33 0.33 0.33 - - - - - - ldp s29, s28, [sp, #252]
# CHECK-NEXT: - - - - - 0.50 0.50 - - - - 0.50 0.50 stp s27, s26, [sp, #-256]
# CHECK-NEXT: - - - - 0.33 0.33 0.33 - - - - - - ldp s1, s2, [x3, #44]
@@ -2392,9 +2392,9 @@
# CHECK-NEXT: - - - - - - - 5.25 0.25 0.25 0.25 - - ldpsw x9, x10, [sp], #4
# CHECK-NEXT: - - - - - - - 5.25 0.25 0.25 0.25 - - ldpsw x9, x10, [x2], #-256
# CHECK-NEXT: - - - - - - - 5.25 0.25 0.25 0.25 - - ldpsw x20, x30, [sp], #252
-# CHECK-NEXT: - - - - 0.33 0.33 0.33 0.25 0.25 0.25 0.25 - - ldp x21, x29, [x2], #504
-# CHECK-NEXT: - - - - 0.33 0.33 0.33 0.25 0.25 0.25 0.25 - - ldp x22, x23, [x3], #-512
-# CHECK-NEXT: - - - - 0.33 0.33 0.33 0.25 0.25 0.25 0.25 - - ldp x24, x25, [x4], #8
+# CHECK-NEXT: - - - - 0.67 0.67 0.67 0.25 0.25 0.25 0.25 - - ldp x21, x29, [x2], #504
+# CHECK-NEXT: - - - - 0.67 0.67 0.67 0.25 0.25 0.25 0.25 - - ldp x22, x23, [x3], #-512
+# CHECK-NEXT: - - - - 0.67 0.67 0.67 0.25 0.25 0.25 0.25 - - ldp x24, x25, [x4], #8
# CHECK-NEXT: - - - - 0.33 0.33 0.33 0.50 0.50 0.50 0.50 - - ldp s29, s28, [sp], #252
# CHECK-NEXT: - - - - - 0.50 0.50 0.50 0.50 0.50 0.50 0.50 0.50 stp s27, s26, [sp], #-256
# CHECK-NEXT: - - - - 0.33 0.33 0.33 0.50 0.50 0.50 0.50 - - ldp s1, s2, [x3], #44
@@ -2411,9 +2411,9 @@
# CHECK-NEXT: - - - - - - - 5.25 0.25 0.25 0.25 - - ldpsw x9, x10, [sp, #4]!
# CHECK-NEXT: - - - - - - - 5.25 0.25 0.25 0.25 - - ldpsw x9, x10, [x2, #-256]!
# CHECK-NEXT: - - - - - - - 5.25 0.25 0.25 0.25 - - ldpsw x20, x30, [sp, #252]!
-# CHECK-NEXT: - - - - 0.33 0.33 0.33 0.25 0.25 0.25 0.25 - - ldp x21, x29, [x2, #504]!
-# CHECK-NEXT: - - - - 0.33 0.33 0.33 0.25 0.25 0.25 0.25 - - ldp x22, x23, [x3, #-512]!
-# CHECK-NEXT: - - - - 0.33 0.33 0.33 0.25 0.25 0.25 0.25 - - ldp x24, x25, [x4, #8]!
+# CHECK-NEXT: - - - - 0.67 0.67 0.67 0.25 0.25 0.25 0.25 - - ldp x21, x29, [x2, #504]!
+# CHECK-NEXT: - - - - 0.67 0.67 0.67 0.25 0.25 0.25 0.25 - - ldp x22, x23, [x3, #-512]!
+# CHECK-NEXT: - - - - 0.67 0.67 0.67 0.25 0.25 0.25 0.25 - - ldp x24, x25, [x4, #8]!
# CHECK-NEXT: - - - - 0.33 0.33 0.33 0.50 0.50 0.50 0.50 - - ldp s29, s28, [sp, #252]!
# CHECK-NEXT: - - - - - 0.50 0.50 0.50 0.50 0.50 0.50 0.50 0.50 stp s27, s26, [sp, #-256]!
# CHECK-NEXT: - - - - 0.33 0.33 0.33 0.50 0.50 0.50 0.50 - - ldp s1, s2, [x3, #44]!
@@ -2427,9 +2427,9 @@
# CHECK-NEXT: - - 0.50 0.50 - 0.50 0.50 - - - - - - stnp wzr, w9, [sp, #252]
# CHECK-NEXT: - - - - 0.33 0.33 0.33 - - - - - - ldnp w2, wzr, [sp, #-256]
# CHECK-NEXT: - - - - 0.33 0.33 0.33 - - - - - - ldnp w9, w10, [sp, #4]
-# CHECK-NEXT: - - - - 0.33 0.33 0.33 - - - - - - ldnp x21, x29, [x2, #504]
-# CHECK-NEXT: - - - - 0.33 0.33 0.33 - - - - - - ldnp x22, x23, [x3, #-512]
-# CHECK-NEXT: - - - - 0.33 0.33 0.33 - - - - - - ldnp x24, x25, [x4, #8]
+# CHECK-NEXT: - - - - 0.67 0.67 0.67 - - - - - - ldnp x21, x29, [x2, #504]
+# CHECK-NEXT: - - - - 0.67 0.67 0.67 - - - - - - ldnp x22, x23, [x3, #-512]
+# CHECK-NEXT: - - - - 0.67 0.67 0.67 - - - - - - ldnp x24, x25, [x4, #8]
# CHECK-NEXT: - - - - 0.33 0.33 0.33 - - - - - - ldnp s29, s28, [sp, #252]
# CHECK-NEXT: - - - - - 0.50 0.50 - - - - 0.50 0.50 stnp s27, s26, [sp, #-256]
# CHECK-NEXT: - - - - 0.33 0.33 0.33 - - - - - - ldnp s1, s2, [x3, #44]
diff --git a/llvm/test/tools/llvm-mca/AArch64/Neoverse/N2-writeback.s b/llvm/test/tools/llvm-mca/AArch64/Neoverse/N2-writeback.s
index dee46a304582b..3644f03d3e309 100644
--- a/llvm/test/tools/llvm-mca/AArch64/Neoverse/N2-writeback.s
+++ b/llvm/test/tools/llvm-mca/AArch64/Neoverse/N2-writeback.s
@@ -1605,11 +1605,11 @@ add x0, x27, 1
# CHECK-NEXT: Block RThroughput: 5.7
# CHECK: Timeline view:
-# CHECK-NEXT: 0123
+# CHECK-NEXT: 01234
# CHECK-NEXT: Index 0123456789
-# CHECK: [0,0] DeeeeeeER . . ld1 { v1.8b, v2.8b, v3.8b }, [x27], x28
-# CHECK-NEXT: [0,1] D=eE----R . . add x0, x27, #1
+# CHECK: [0,0] DeeeeeeER . . ld1 { v1.8b, v2.8b, v3.8b }, [x27], x28
+# CHECK-NEXT: [0,1] D=eE----R . . add x0, x27, #1
# CHECK-NEXT: [0,2] .DeeeeeeER. . ld1 { v1.8h, v2.8h, v3.8h }, [x27], x28
# CHECK-NEXT: [0,3] .D=eE----R. . add x0, x27, #1
# CHECK-NEXT: [0,4] . DeeeeeeER . ld1 { v1.16b, v2.16b, v3.16b }, [x27], x28
@@ -1651,7 +1651,7 @@ add x0, x27, 1
# CHECK-NEXT: Block RThroughput: 6.7
# CHECK: Timeline view:
-# CHECK-NEXT: 01234
+# CHECK-NEXT: 01234567
# CHECK-NEXT: Index 0123456789
# CHECK: [0,0] DeeeeeeeER. . . ld1 { v1.2s, v2.2s, v3.2s, v4.2s }, [x27], #32
@@ -1697,7 +1697,7 @@ add x0, x27, 1
# CHECK-NEXT: Block RThroughput: 6.7
# CHECK: Timeline view:
-# CHECK-NEXT: 01234
+# CHECK-NEXT: 01234567
# CHECK-NEXT: Index 0123456789
# CHECK: [0,0] DeeeeeeeER. . . ld1 { v1.16b, v2.16b, v3.16b, v4.16b }, [x27], #64
@@ -1744,7 +1744,7 @@ add x0, x27, 1
# CHECK: Timeline view:
# CHECK-NEXT: 0123456789
-# CHECK-NEXT: Index 0123456789 01
+# CHECK-NEXT: Index 0123456789 0123
# CHECK: [0,0] DeeeeeeeER. . . . ld1 { v1.4s, v2.4s, v3.4s, v4.4s }, [x27], x28
# CHECK-NEXT: [0,1] .DeE-----R. . . . add x0, x27, #1
@@ -2114,8 +2114,8 @@ add x0, x27, 1
# CHECK-NEXT: 012345
# CHECK-NEXT: Index 0123456789
-# CHECK: [0,0] DeeeeeeeeER . ld2 { v1.2s, v2.2s }, [x27], x28
-# CHECK-NEXT: [0,1] D=eE------R . add x0, x27, #1
+# CHECK: [0,0] DeeeeeeeeER . ld2 { v1.2s, v2.2s }, [x27], x28
+# CHECK-NEXT: [0,1] D=eE------R . add x0, x27, #1
# CHECK-NEXT: [0,2] .DeeeeeeeeER . ld2 { v1.4h, v2.4h }, [x27], x28
# CHECK-NEXT: [0,3] .D=eE------R . add x0, x27, #1
# CHECK-NEXT: [0,4] . DeeeeeeeeER . ld2 { v1.4s, v2.4s }, [x27], x28
@@ -2480,7 +2480,7 @@ add x0, x27, 1
# CHECK: Timeline view:
# CHECK-NEXT: 012345678
-# CHECK-NEXT: Index 0123456789
+# CHECK-NEXT: Index 0123456789
# CHECK: [0,0] DeeeeeeeeER . . ld3 { v1.16b, v2.16b, v3.16b }, [x27], #48
# CHECK-NEXT: [0,1] .DeE------R . . add x0, x27, #1
@@ -2526,7 +2526,7 @@ add x0, x27, 1
# CHECK: Timeline view:
# CHECK-NEXT: 0123456789 0
-# CHECK-NEXT: Index 0123456789 0123456789
+# CHECK-NEXT: Index 0123456789 0123456789
# CHECK: [0,0] DeeeeeeeeER . . . . ld3 { v1.8b, v2.8b, v3.8b }, [x27], x28
# CHECK-NEXT: [0,1] .DeE------R . . . . add x0, x27, #1
@@ -2664,7 +2664,7 @@ add x0, x27, 1
# CHECK: Timeline view:
# CHECK-NEXT: 012345678
-# CHECK-NEXT: Index 0123456789
+# CHECK-NEXT: Index 0123456789
# CHECK: [0,0] DeeeeeeeeER . . ld3r { v1.1d, v2.1d, v3.1d }, [x27], #24
# CHECK-NEXT: [0,1] .DeE------R . . add x0, x27, #1
@@ -2710,7 +2710,7 @@ add x0, x27, 1
# CHECK: Timeline view:
# CHECK-NEXT: 012345678
-# CHECK-NEXT: Index 0123456789
+# CHECK-NEXT: Index 0123456789
# CHECK: [0,0] DeeeeeeeeER . . ld3r { v1.8b, v2.8b, v3.8b }, [x27], #3
# CHECK-NEXT: [0,1] .DeE------R . . add x0, x27, #1
@@ -2756,7 +2756,7 @@ add x0, x27, 1
# CHECK: Timeline view:
# CHECK-NEXT: 012345678
-# CHECK-NEXT: Index 0123456789
+# CHECK-NEXT: Index 0123456789
# CHECK: [0,0] DeeeeeeeeER . . ld3r { v1.2s, v2.2s, v3.2s }, [x27], x28
# CHECK-NEXT: [0,1] .DeE------R . . add x0, x27, #1
@@ -2802,7 +2802,7 @@ add x0, x27, 1
# CHECK: Timeline view:
# CHECK-NEXT: 0123456789
-# CHECK-NEXT: Index 0123456789
+# CHECK-NEXT: Index 0123456789
# CHECK: [0,0] DeeeeeeeeER . . ld3r { v1.16b, v2.16b, v3.16b }, [x27], x28
# CHECK-NEXT: [0,1] .DeE------R . . add x0, x27, #1
@@ -2848,7 +2848,7 @@ add x0, x27, 1
# CHECK: Timeline view:
# CHECK-NEXT: 012345678
-# CHECK-NEXT: Index 0123456789
+# CHECK-NEXT: Index 0123456789
# CHECK: [0,0] DeeeeeeeeER . . ld4 { v1.8b, v2.8b, v3.8b, v4.8b }, [x27], #32
# CHECK-NEXT: [0,1] .DeE------R . . add x0, x27, #1
@@ -2894,7 +2894,7 @@ add x0, x27, 1
# CHECK: Timeline view:
# CHECK-NEXT: 0123456789
-# CHECK-NEXT: Index 0123456789
+# CHECK-NEXT: Index 0123456789
# CHECK: [0,0] DeeeeeeeeER . . ld4 { v1.4h, v2.4h, v3.4h, v4.4h }, [x27], x28
# CHECK-NEXT: [0,1] .DeE------R . . add x0, x27, #1
@@ -3078,7 +3078,7 @@ add x0, x27, 1
# CHECK: Timeline view:
# CHECK-NEXT: 012345678
-# CHECK-NEXT: Index 0123456789
+# CHECK-NEXT: Index 0123456789
# CHECK: [0,0] DeeeeeeeeER . . ld4r { v1.4h, v2.4h, v3.4h, v4.4h }, [x27], #8
# CHECK-NEXT: [0,1] .DeE------R . . add x0, x27, #1
@@ -3124,7 +3124,7 @@ add x0, x27, 1
# CHECK: Timeline view:
# CHECK-NEXT: 012345678
-# CHECK-NEXT: Index 0123456789
+# CHECK-NEXT: Index 0123456789
# CHECK: [0,0] DeeeeeeeeER . . ld4r { v1.1d, v2.1d, v3.1d, v4.1d }, [x27], x28
# CHECK-NEXT: [0,1] .DeE------R . . add x0, x27, #1
@@ -3170,7 +3170,7 @@ add x0, x27, 1
# CHECK: Timeline view:
# CHECK-NEXT: 012345
-# CHECK-NEXT: Index 0123456789
+# CHECK-NEXT: Index 0123456789
# CHECK: [0,0] DeeeeeeeeER . ld4r { v1.8b, v2.8b, v3.8b, v4.8b }, [x27], x28
# CHECK-NEXT: [0,1] .DeE------R . add x0, x27, #1
@@ -3216,7 +3216,7 @@ add x0, x27, 1
# CHECK: Timeline view:
# CHECK-NEXT: 012
-# CHECK-NEXT: Index 0123456789
+# CHECK-NEXT: Index 0123456789
# CHECK: [0,0] DeeeeeeER . . ldp q1, q2, [x27], #992
# CHECK-NEXT: [0,1] .DeE----R . . add x0, x27, #1
@@ -3253,10 +3253,10 @@ add x0, x27, 1
# CHECK: Iterations: 100
# CHECK-NEXT: Instructions: 1000
# CHECK-NEXT: Total Cycles: 1006
-# CHECK-NEXT: Total uOps: 2000
+# CHECK-NEXT: Total uOps: 2200
# CHECK: Dispatch Width: 5
-# CHECK-NEXT: uOps Per Cycle: 1.99
+# CHECK-NEXT: uOps Per Cycle: 2.19
# CHECK-NEXT: IPC: 0.99
# CHECK-NEXT: Block RThroughput: 10.0
@@ -3308,7 +3308,7 @@ add x0, x27, 1
# CHECK: Timeline view:
# CHECK-NEXT: 012
-# CHECK-NEXT: Index 0123456789
+# CHECK-NEXT: Index 0123456789
# CHECK: [0,0] DeeeeeeER . . ldr b1, [x27], #254
# CHECK-NEXT: [0,1] D=eE----R . . add x0, x27, #1
@@ -3354,7 +3354,7 @@ add x0, x27, 1
# CHECK: Timeline view:
# CHECK-NEXT: 012
-# CHECK-NEXT: Index 0123456789
+# CHECK-NEXT: Index 0123456789
# CHECK: [0,0] DeeeeeeER . . ldr b1, [x27, #254]!
# CHECK-NEXT: [0,1] D=eE----R . . add x0, x27, #1
@@ -3400,7 +3400,7 @@ add x0, x27, 1
# CHECK: Timeline view:
# CHECK-NEXT: 0
-# CHECK-NEXT: Index 0123456789
+# CHECK-NEXT: Index 0123456789
# CHECK: [0,0] DeeeeER . ldr w1, [x27], #254
# ...
[truncated]
``````````
</details>
https://github.com/llvm/llvm-project/pull/171853
More information about the llvm-commits
mailing list