121 lines
12 KiB
Plaintext
121 lines
12 KiB
Plaintext
|
Intel(R) Architecture Code Analyzer Version - v3.0-28-g1ba2cbb build date: 2017-10-23;16:42:45
|
||
|
|
Analyzed File - gromacs-avx512-dp.o
|
||
|
|
Binary Format - 64Bit
|
||
|
|
Architecture - SKX
|
||
|
|
Analysis Type - Throughput
|
||
|
|
|
||
|
|
Throughput Analysis Report
|
||
|
|
--------------------------
|
||
|
|
Block Throughput: 45.95 Cycles Throughput Bottleneck: Backend
|
||
|
|
Loop Count: 22
|
||
|
|
Port Binding In Cycles Per Iteration:
|
||
|
|
--------------------------------------------------------------------------------------------------
|
||
|
|
| Port | 0 - DV | 1 | 2 - D | 3 - D | 4 | 5 | 6 | 7 |
|
||
|
|
--------------------------------------------------------------------------------------------------
|
||
|
|
| Cycles | 40.0 0.0 | 2.0 | 5.0 5.0 | 5.0 5.0 | 0.0 | 40.0 | 2.0 | 0.0 |
|
||
|
|
--------------------------------------------------------------------------------------------------
|
||
|
|
|
||
|
|
DV - Divider pipe (on port 0)
|
||
|
|
D - Data fetch pipe (on ports 2 and 3)
|
||
|
|
F - Macro Fusion with the previous instruction occurred
|
||
|
|
* - instruction micro-ops not bound to a port
|
||
|
|
^ - Micro Fusion occurred
|
||
|
|
# - ESP Tracking sync uop was issued
|
||
|
|
@ - SSE instruction followed an AVX256/AVX512 instruction, dozens of cycles penalty is expected
|
||
|
|
X - instruction not supported, was not accounted in Analysis
|
||
|
|
|
||
|
|
| Num Of | Ports pressure in cycles | |
|
||
|
|
| Uops | 0 - DV | 1 | 2 - D | 3 - D | 4 | 5 | 6 | 7 |
|
||
|
|
-----------------------------------------------------------------------------------------
|
||
|
|
| 1 | | | 1.0 1.0 | | | | | | movsxd rax, dword ptr [rdx+rcx*4]
|
||
|
|
| 1 | | 1.0 | | | | | | | lea rax, ptr [rax+rax*2]
|
||
|
|
| 1 | | | | | | | 1.0 | | shl rax, 0x6
|
||
|
|
| 1 | | | | 1.0 1.0 | | | | | vmovapd zmm28, zmmword ptr [rdi+rax*1]
|
||
|
|
| 1 | | | 1.0 1.0 | | | | | | vmovapd zmm29, zmmword ptr [rdi+rax*1+0x40]
|
||
|
|
| 1 | | | | 1.0 1.0 | | | | | vmovapd zmm30, zmmword ptr [rdi+rax*1+0x80]
|
||
|
|
| 1 | | | 1.0 1.0 | | | | | | vmovupd zmm3, zmmword ptr [rsp+0x40]
|
||
|
|
| 1 | | | | | | 1.0 | | | vsubpd zmm14, zmm3, zmm28
|
||
|
|
| 1 | 1.0 | | | | | | | | vsubpd zmm12, zmm26, zmm29
|
||
|
|
| 1 | | | | | | 1.0 | | | vsubpd zmm31, zmm25, zmm30
|
||
|
|
| 1 | 1.0 | | | | | | | | vsubpd zmm3, zmm23, zmm28
|
||
|
|
| 1 | | | | | | 1.0 | | | vmulpd zmm15, zmm31, zmm31
|
||
|
|
| 1 | 1.0 | | | | | | | | vfmadd231pd zmm15, zmm12, zmm12
|
||
|
|
| 1 | | | | | | 1.0 | | | vfmadd231pd zmm15, zmm14, zmm14
|
||
|
|
| 3 | 2.0 | | | | | 1.0 | | | vrcp14pd zmm16, zmm15
|
||
|
|
| 1 | | | | | | 1.0 | | | vcmppd k1, zmm15, zmm0, 0x1
|
||
|
|
| 1 | 1.0 | | | | | | | | vmulpd zmm15, zmm16, zmm20
|
||
|
|
| 1 | | | | | | 1.0 | | | vmulpd zmm18, zmm16, zmm16
|
||
|
|
| 1 | 1.0 | | | | | | | | vmulpd zmm15, zmm18, zmm15
|
||
|
|
| 1 | | | | 1.0 1.0 | | | | | vmovupd zmm18, zmmword ptr [rsp+0x200]
|
||
|
|
| 1 | | | | | | 1.0 | | | vsubpd zmm18, zmm18, zmm29
|
||
|
|
| 1 | 1.0 | | | | | | | | vmulpd zmm16, zmm1, zmm16
|
||
|
|
| 1 | | | | | | 1.0 | | | vmulpd zmm16, zmm16, zmm15
|
||
|
|
| 1 | 1.0 | | | | | | | | vaddpd zmm15, zmm15, zmm2
|
||
|
|
| 1 | | | | | | 1.0 | | | vmulpd zmm15, zmm16, zmm15
|
||
|
|
| 1 | | | 1.0 1.0 | | | | | | vmovupd zmm16, zmmword ptr [rsp+0x1c0]
|
||
|
|
| 1 | 1.0 | | | | | | | | vsubpd zmm16, zmm16, zmm30
|
||
|
|
| 1 | | | | | | 1.0 | | | vfmadd231pd zmm19{k1}, zmm15, zmm14
|
||
|
|
| 1 | 1.0 | | | | | | | | vmulpd zmm14, zmm16, zmm16
|
||
|
|
| 1 | | | | | | 1.0 | | | vfmadd231pd zmm14, zmm18, zmm18
|
||
|
|
| 1 | 1.0 | | | | | | | | vfmadd231pd zmm14, zmm3, zmm3
|
||
|
|
| 1 | | | | | | 1.0 | | | vcmppd k2, zmm14, zmm0, 0x1
|
||
|
|
| 3 | 2.0 | | | | | 1.0 | | | vrcp14pd zmm14, zmm14
|
||
|
|
| 1 | | | | | | 1.0 | | | vfmadd231pd zmm11{k1}, zmm15, zmm12
|
||
|
|
| 1 | 1.0 | | | | | | | | vfmadd231pd zmm7{k1}, zmm15, zmm31
|
||
|
|
| 1 | | | | | | 1.0 | | | vmulpd zmm12, zmm14, zmm20
|
||
|
|
| 1 | 1.0 | | | | | | | | vmulpd zmm15, zmm14, zmm14
|
||
|
|
| 1 | | | | | | 1.0 | | | vmulpd zmm12, zmm15, zmm12
|
||
|
|
| 1 | 1.0 | | | | | | | | vsubpd zmm15, zmm24, zmm28
|
||
|
|
| 1 | | | | | | 1.0 | | | vmulpd zmm14, zmm1, zmm14
|
||
|
|
| 1 | 1.0 | | | | | | | | vmulpd zmm14, zmm14, zmm12
|
||
|
|
| 1 | | | | | | 1.0 | | | vaddpd zmm12, zmm12, zmm2
|
||
|
|
| 1 | 1.0 | | | | | | | | vmulpd zmm12, zmm14, zmm12
|
||
|
|
| 1 | | | | | | 1.0 | | | vsubpd zmm14, zmm22, zmm29
|
||
|
|
| 1 | 1.0 | | | | | | | | vfmadd231pd zmm17{k2}, zmm12, zmm3
|
||
|
|
| 1 | | | | | | 1.0 | | | vsubpd zmm3, zmm27, zmm30
|
||
|
|
| 1 | 1.0 | | | | | | | | vfmadd231pd zmm9{k2}, zmm12, zmm18
|
||
|
|
| 1 | | | | | | 1.0 | | | vmulpd zmm18, zmm3, zmm3
|
||
|
|
| 1 | 1.0 | | | | | | | | vfmadd231pd zmm18, zmm14, zmm14
|
||
|
|
| 1 | | | | | | 1.0 | | | vfmadd231pd zmm18, zmm15, zmm15
|
||
|
|
| 1 | | | | | | 1.0 | | | vcmppd k1, zmm18, zmm0, 0x1
|
||
|
|
| 3 | 2.0 | | | | | 1.0 | | | vrcp14pd zmm18, zmm18
|
||
|
|
| 1 | 1.0 | | | | | | | | vfmadd231pd zmm5{k2}, zmm12, zmm16
|
||
|
|
| 1 | | | | | | 1.0 | | | vmulpd zmm12, zmm18, zmm20
|
||
|
|
| 1 | 1.0 | | | | | | | | vmulpd zmm16, zmm18, zmm18
|
||
|
|
| 1 | | | | | | 1.0 | | | vmulpd zmm12, zmm16, zmm12
|
||
|
|
| 1 | 1.0 | | | | | | | | vmulpd zmm16, zmm1, zmm18
|
||
|
|
| 1 | | | | | | 1.0 | | | vmulpd zmm16, zmm16, zmm12
|
||
|
|
| 1 | 1.0 | | | | | | | | vaddpd zmm12, zmm12, zmm2
|
||
|
|
| 1 | | | | | | 1.0 | | | vmulpd zmm12, zmm16, zmm12
|
||
|
|
| 1 | 1.0 | | | | | | | | vfmadd231pd zmm21{k1}, zmm12, zmm15
|
||
|
|
| 1 | | | | 1.0 1.0 | | | | | vmovupd zmm15, zmmword ptr [rsp+0x240]
|
||
|
|
| 1 | | | | | | 1.0 | | | vsubpd zmm15, zmm15, zmm28
|
||
|
|
| 1 | | | 1.0 1.0 | | | | | | vmovupd zmm16, zmmword ptr [rsp+0x80]
|
||
|
|
| 1 | 1.0 | | | | | | | | vsubpd zmm16, zmm16, zmm29
|
||
|
|
| 1 | | | | 1.0 1.0 | | | | | vmovupd zmm18, zmmword ptr [rsp+0x180]
|
||
|
|
| 1 | | | | | | 1.0 | | | vsubpd zmm18, zmm18, zmm30
|
||
|
|
| 1 | 1.0 | | | | | | | | vfmadd231pd zmm10{k1}, zmm12, zmm14
|
||
|
|
| 1 | | | | | | 1.0 | | | vmulpd zmm14, zmm18, zmm18
|
||
|
|
| 1 | 1.0 | | | | | | | | vfmadd231pd zmm14, zmm16, zmm16
|
||
|
|
| 1 | | | | | | 1.0 | | | vfmadd231pd zmm14, zmm15, zmm15
|
||
|
|
| 1 | | | | | | 1.0 | | | vcmppd k2, zmm14, zmm0, 0x1
|
||
|
|
| 3 | 2.0 | | | | | 1.0 | | | vrcp14pd zmm14, zmm14
|
||
|
|
| 1 | 1.0 | | | | | | | | vfmadd231pd zmm6{k1}, zmm12, zmm3
|
||
|
|
| 1 | | | | | | 1.0 | | | vmulpd zmm3, zmm14, zmm20
|
||
|
|
| 1 | 1.0 | | | | | | | | vmulpd zmm12, zmm14, zmm14
|
||
|
|
| 1 | | | | | | 1.0 | | | vmulpd zmm3, zmm12, zmm3
|
||
|
|
| 1 | 1.0 | | | | | | | | vmulpd zmm12, zmm1, zmm14
|
||
|
|
| 1 | | | | | | 1.0 | | | vmulpd zmm12, zmm12, zmm3
|
||
|
|
| 1 | 1.0 | | | | | | | | vaddpd zmm3, zmm3, zmm2
|
||
|
|
| 1 | | | | | | 1.0 | | | vmulpd zmm3, zmm12, zmm3
|
||
|
|
| 1 | 1.0 | | | | | | | | vfmadd231pd zmm13{k2}, zmm3, zmm15
|
||
|
|
| 1 | | | | | | 1.0 | | | vfmadd231pd zmm8{k2}, zmm3, zmm16
|
||
|
|
| 1 | 1.0 | | | | | | | | vfmadd231pd zmm4{k2}, zmm3, zmm18
|
||
|
|
| 1 | | 1.0 | | | | | | | inc rcx
|
||
|
|
| 1* | | | | | | | | | cmp r11, rcx
|
||
|
|
| 0*F | | | | | | | | | jnz 0xfffffffffffffdf3
|
||
|
|
| 1 | | | | | | | 1.0 | | jmp 0xfffffffffffff7a5
|
||
|
|
Total Num Of Uops: 95
|
||
|
|
Analysis Notes:
|
||
|
|
Backend allocation was stalled due to unavailable allocation resources.
|