Skip to content

Instantly share code, notes, and snippets.

@EgorBot
Created October 26, 2024 15:59
Show Gist options
  • Save EgorBot/9bad032058afeb9f28207dbd7148248d to your computer and use it in GitHub Desktop.
Save EgorBot/9bad032058afeb9f28207dbd7148248d to your computer and use it in GitHub Desktop.
diff_asm_b84ab9f0.asm
Samples: 10K of event 'cpu-clock', 1999 Hz, Event count (approx.): 5006502000, [percent: local period]
instance int32 [benchapp] Bench::TestInner(class [System.Runtime]System.Collections.Generic.ICollection`1<string>)[OptimizedTier1]() /tmp/jitted-53947-7271.so
Percent
Disassembly of section .text:
0000000000000080 <instance int32 [benchapp] Bench::TestInner(class [System.Runtime]System.Collections.Generic.ICollection`1<string>)[OptimizedTier1]>:
30.62 stp x29, x30, [sp, #-16]!
0.11 mov x29, sp
3.20 ldr x0, [x1]
1.72 mov x11, #0x9070 // #36976
0.17 movk x11, #0x51e, lsl #16
1.64 movk x11, #0xe4aa, lsl #32
1.73 cmp x0, x11
↓ b.ne 2c
1.79 ldr w0, [x1, #8]
1.72 24: ldp x29, x30, [sp], #16
57.30 ← ret
2c: mov x0, x1
mov x11, #0x2458 // #9304
movk x11, #0x404, lsl #16
movk x11, #0xe4aa, lsl #32
ldr x16, [x11]
→ blr x16
↑ b 24
Samples: 10K of event 'cpu-clock', 1999 Hz, Event count (approx.): 5006502000, [percent: local period]
instance void [d8b03265-5d61-4709-b1d8-b442eacaeb14Emitted] BenchmarkDotNet.Autogenerated.Runnable_0::WorkloadActionUnroll(int64)[Optimized]() /tmp/jitted-53947-7234.so
Percent
Disassembly of section .text:
0000000000000080 <instance void [d8b03265-5d61-4709-b1d8-b442eacaeb14Emitted] BenchmarkDotNet.Autogenerated.Runnable_0::WorkloadActionUnroll(int64)[Optimized]>:
stp x29, x30, [sp, #-48]!
stp x19, x20, [sp, #24]
str x21, [sp, #40]
mov x29, sp
mov x19, x0
cmp x1, #0x0
↓ b.le 168
mov x20, x1
0.72 20: ldp x1, x21, [x19, #48]
0.65 ldr x0, [x1, #8]
0.51 ldr x1, [x1, #24]
0.65 → blr x1
3.75 stlur w0, [x21, #56]
0.65 ldp x1, x21, [x19, #48]
0.36 ldr x0, [x1, #8]
0.58 ldr x1, [x1, #24]
0.87 → blr x1
3.97 stlur w0, [x21, #56]
0.07 ldp x1, x21, [x19, #48]
0.87 ldr x0, [x1, #8]
0.43 ldr x1, [x1, #24]
0.94 → blr x1
4.62 stlur w0, [x21, #56]
0.22 ldp x1, x21, [x19, #48]
0.29 ldr x0, [x1, #8]
0.79 ldr x1, [x1, #24]
0.43 → blr x1
4.40 stlur w0, [x21, #56]
0.36 ldp x1, x21, [x19, #48]
0.36 ldr x0, [x1, #8]
0.72 ldr x1, [x1, #24]
0.65 → blr x1
4.83 stlur w0, [x21, #56]
0.58 ldp x1, x21, [x19, #48]
0.29 ldr x0, [x1, #8]
0.14 ldr x1, [x1, #24]
0.65 → blr x1
4.40 stlur w0, [x21, #56]
0.43 ldp x1, x21, [x19, #48]
0.65 ldr x0, [x1, #8]
0.51 ldr x1, [x1, #24]
0.51 → blr x1
4.11 stlur w0, [x21, #56]
0.51 ldp x1, x21, [x19, #48]
0.51 ldr x0, [x1, #8]
0.51 ldr x1, [x1, #24]
0.79 → blr x1
4.47 stlur w0, [x21, #56]
0.36 ldp x1, x21, [x19, #48]
0.51 ldr x0, [x1, #8]
0.36 ldr x1, [x1, #24]
0.14 → blr x1
3.97 stlur w0, [x21, #56]
0.58 ldp x1, x21, [x19, #48]
0.29 ldr x0, [x1, #8]
0.58 ldr x1, [x1, #24]
0.58 → blr x1
3.39 stlur w0, [x21, #56]
0.36 ldp x1, x21, [x19, #48]
0.79 ldr x0, [x1, #8]
0.58 ldr x1, [x1, #24]
0.29 → blr x1
5.12 stlur w0, [x21, #56]
0.72 ldp x1, x21, [x19, #48]
0.72 ldr x0, [x1, #8]
0.51 ldr x1, [x1, #24]
0.43 → blr x1
3.39 stlur w0, [x21, #56]
0.72 ldp x1, x21, [x19, #48]
0.43 ldr x0, [x1, #8]
0.51 ldr x1, [x1, #24]
0.65 → blr x1
3.75 stlur w0, [x21, #56]
0.65 ldp x1, x21, [x19, #48]
0.51 ldr x0, [x1, #8]
0.51 ldr x1, [x1, #24]
0.79 → blr x1
4.04 stlur w0, [x21, #56]
0.72 ldp x1, x21, [x19, #48]
0.51 ldr x0, [x1, #8]
0.72 ldr x1, [x1, #24]
0.72 → blr x1
3.39 stlur w0, [x21, #56]
0.51 ldp x1, x21, [x19, #48]
0.36 ldr x0, [x1, #8]
0.36 ldr x1, [x1, #24]
0.65 → blr x1
3.75 stlur w0, [x21, #56]
0.36 sub x20, x20, #0x1
↑ cbnz x20, 20
168: ldr x21, [sp, #40]
ldp x19, x20, [sp, #24]
ldp x29, x30, [sp], #48
← ret
Samples: 10K of event 'cpu-clock', 1999 Hz, Event count (approx.): 5006502000, [percent: local period]
instance int32 [benchapp] Bench::Test()[OptimizedTier1]() /tmp/jitted-53947-7270.so
Percent
Disassembly of section .text:
0000000000000080 <instance int32 [benchapp] Bench::Test()[OptimizedTier1]>:
10.74 stp x29, x30, [sp, #-16]!
0.92 mov x29, sp
23.72 mov x1, #0x2d00 // #11520
1.23 movk x1, #0xf800, lsl #16
11.76 movk x1, #0xe469, lsl #32
13.50 ldr x1, [x1]
0.10 mov x2, #0xd260 // #53856
13.39 movk x2, #0x5c1, lsl #16
0.31 movk x2, #0xe4aa, lsl #32
10.94 ldr x2, [x2]
13.39 ldp x29, x30, [sp], #16
→ br x2
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment