diff_asm_f256ece8.asm

diff_asm_f256ece8.asm

Samples: 96K of event 'cpu-clock', 11999 Hz, Event count (approx.): 8028475560, [percent: local period]
void [System.Private.CoreLib] System.Buffers.Binary.BinaryPrimitives::ReverseEndianness(valuetype System.ReadOnlySpan`1<!!0>,valuetype System.Span`1<!!0>)[OptimizedTier1]() /tmp/jitted-60783-7253.so
Percent        
               
               
             Disassembly of section .text:
               
             0000000000000080 <void [System.Private.CoreLib] System.Buffers.Binary.BinaryPrimitives::ReverseEndianness(valuetype System.ReadOnlySpan`1<!!0>,valuetype System.Span`1<!!0>)[OptimizedTier1]>:
             void [System.Private.CoreLib] System.Buffers.Binary.BinaryPrimitives::ReverseEndianness(valuetype System.ReadOnlySpan`1<!!0>,valuetype System.Span`1<!!0>)[OptimizedTier1]():
  0.58         push       rbp 
               push       rbx 
               push       rax 
               lea        rbp,[rsp+0x10]
  0.01         cmp        ecx,esi
             ↓ jl         19a 
               cmp        rdi,rdx
             ↓ je         3c  
               test       esi,esi
             ↓ je         3c  
  0.26         test       ecx,ecx
             ↓ je         3c  
  0.00         mov        rbx,rdx
               sub        rbx,rdi
               mov        eax,esi
  0.02         shl        rax,0x2
               cmp        rbx,rax
             ↓ jb         90  
  0.00         mov        eax,ecx
               shl        rax,0x2
  0.23         neg        rax 
  0.01         cmp        rax,rbx
             ↓ jb         90  
  0.03   3c:   xor        eax,eax
               lea        ecx,[rsi-0x10]
  0.03         test       ecx,ecx
             ↓ jl         68  
  0.01         vmovups    zmm0,ZMMWORD PTR [rip+0x191]        # 260 <void [System.Private.CoreLib] System.Buffers.Binary.BinaryPrimitives::ReverseEndianness(valuetype System.ReadOnlySpan`1<!!0>,valuetype System.Span`1<!!0>)[OptimizedTier1]+0x1e0>
  0.39         nop            
  4.94   50:   mov        r8d,eax
  0.35         vpermb     zmm1,zmm0,ZMMWORD PTR [rdi+r8*4]
 27.17         vmovups    ZMMWORD PTR [rdx+r8*4],zmm1
 65.29         add        eax,0x10
  0.37         cmp        eax,ecx
  0.00       ↑ jle        50  
         68:   lea        ecx,[rsi-0x8]
         6b:   cmp        eax,ecx
             ↓ jle        177 
  0.01         lea        ecx,[rsi-0x4]
         76:   cmp        eax,ecx
             ↓ jle        154 
         7e:   cmp        eax,esi
             ↓ jl         13e 
         86:   vzeroupper     
  0.27         add        rsp,0x8
  0.01         pop        rbx 
  0.00         pop        rbp 
  0.01       ← ret            
         90:   test       bl,0x3
             ↓ je         a2  
               movabs     rax,0x75b4e5bacc78
             → call       QWORD PTR [rax]
               int3           
         a2:   mov        rax,rbx
               sar        rax,0x3f
               and        rax,0x3
               add        rax,rbx
               sar        rax,0x2
               test       eax,eax
             ↑ jl         3c  
               cmp        esi,0x10
             ↓ jl         e1  
         bd:   add        esi,0xfffffff0
               mov        eax,esi
               vmovups    zmm0,ZMMWORD PTR [rip+0x114]        # 260 <void [System.Private.CoreLib] System.Buffers.Binary.BinaryPrimitives::ReverseEndianness(valuetype System.ReadOnlySpan`1<!!0>,valuetype System.Span`1<!!0>)[OptimizedTier1]+0x1e0>
               vpermb     zmm0,zmm0,ZMMWORD PTR [rdi+rax*4]
               mov        eax,esi
               vmovups    ZMMWORD PTR [rdx+rax*4],zmm0
               cmp        esi,0x10
             ↑ jge        bd  
         e1:   cmp        esi,0x8
             ↓ jl         103 
               add        esi,0xfffffff8
               mov        eax,esi
               vmovups    ymm0,YMMWORD PTR [rip+0xed]        # 260 <void [System.Private.CoreLib] System.Buffers.Binary.BinaryPrimitives::ReverseEndianness(valuetype System.ReadOnlySpan`1<!!0>,valuetype System.Span`1<!!0>)[OptimizedTier1]+0x1e0>
               vpermb     ymm0,ymm0,YMMWORD PTR [rdi+rax*4]
               mov        eax,esi
               vmovups    YMMWORD PTR [rdx+rax*4],ymm0
             ↑ jmp        e1  
        103:   cmp        esi,0x4
             ↓ jl         124 
               add        esi,0xfffffffc
               mov        eax,esi
               vmovups    xmm0,XMMWORD PTR [rdi+rax*4]
               vpshufb    xmm0,xmm0,XMMWORD PTR [rip+0xc5]        # 260 <void [System.Private.CoreLib] System.Buffers.Binary.BinaryPrimitives::ReverseEndianness(valuetype System.ReadOnlySpan`1<!!0>,valuetype System.Span`1<!!0>)[OptimizedTier1]+0x1e0>
               mov        eax,esi
               vmovups    XMMWORD PTR [rdx+rax*4],xmm0
             ↑ jmp        103 
        124:   test       esi,esi
             ↑ jle        86  
               dec        esi 
               movsxd     rax,esi
               mov        eax,DWORD PTR [rdi+rax*4]
               movsxd     rcx,esi
               movbe      DWORD PTR [rdx+rcx*4],eax
             ↑ jmp        124 
        13e:   movsxd     rcx,eax
               mov        ecx,DWORD PTR [rdi+rcx*4]
               movsxd     r8,eax
               movbe      DWORD PTR [rdx+r8*4],ecx
               inc        eax 
             ↑ jmp        7e  
        154:   mov        r8d,eax
               vmovups    xmm0,XMMWORD PTR [rdi+r8*4]
               vpshufb    xmm0,xmm0,XMMWORD PTR [rip+0x7a]        # 260 <void [System.Private.CoreLib] System.Buffers.Binary.BinaryPrimitives::ReverseEndianness(valuetype System.ReadOnlySpan`1<!!0>,valuetype System.Span`1<!!0>)[OptimizedTier1]+0x1e0>
               mov        r8d,eax
               vmovups    XMMWORD PTR [rdx+r8*4],xmm0
               add        eax,0x4
             ↑ jmp        76  
        177:   mov        r8d,eax
               vmovups    ymm0,YMMWORD PTR [rip+0x5e]        # 260 <void [System.Private.CoreLib] System.Buffers.Binary.BinaryPrimitives::ReverseEndianness(valuetype System.ReadOnlySpan`1<!!0>,valuetype System.Span`1<!!0>)[OptimizedTier1]+0x1e0>
               vpermb     ymm0,ymm0,YMMWORD PTR [rdi+r8*4]
               mov        r8d,eax
               vmovups    YMMWORD PTR [rdx+r8*4],ymm0
               add        eax,0x8
             ↑ jmp        6b  
        19a:   movabs     rax,0x75b4e5bacbe8
             → call       QWORD PTR [rax]
               int3           
Samples: 96K of event 'cpu-clock', 11999 Hz, Event count (approx.): 8028475560, [percent: local period]
instance void [benchapp] Bench::Reverse()[OptimizedTier1]() /tmp/jitted-60783-7255.so
Percent        
               
               
            Disassembly of section .text:
               
            0000000000000080 <instance void [benchapp] Bench::Reverse()[OptimizedTier1]>:
            instance void [benchapp] Bench::Reverse()[OptimizedTier1]():
  0.40        push   rbp    
              mov    rbp,rsp
  0.50        mov    rsi,QWORD PTR [rdi+0x8]
  7.78        test   rsi,rsi
            ↓ jne    38     
              xor    edx,edx
              xor    esi,esi
  0.40  11:   mov    rdi,QWORD PTR [rdi+0x10]
              test   rdi,rdi
            ↓ je     41     
  1.09        lea    rcx,[rdi+0x10]
  1.14        mov    eax,DWORD PTR [rdi+0x8]
 22.46  21:   mov    rdi,rdx
              mov    rdx,rcx
 11.60        mov    ecx,eax
  2.23        movabs rax,0x75b4e5b0e310
  0.25      → call   QWORD PTR [rax]
  0.55        nop           
 11.50        pop    rbp    
  0.10      ← ret           
  1.04  38:   lea    rdx,[rsi+0x10]
 17.20        mov    esi,DWORD PTR [rsi+0x8]
 21.76      ↑ jmp    11     
        41:   xor    ecx,ecx
              xor    eax,eax
            ↑ jmp    21