Intel(R) Architecture Code Analyzer Version - v3.0-28-g1ba2cbb build date: 2017-10-23;16:42:45 Analyzed File - force_aos_lt8_markers.o Binary Format - 64Bit Architecture - SKX Analysis Type - Throughput Throughput Analysis Report -------------------------- Block Throughput: 69.79 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 | 21.0 0.0 | 5.5 | 13.0 13.0 | 13.0 13.0 | 0.0 | 21.0 | 5.5 | 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 | | | | | | | imul rcx, r8 | 1 | | | | | | 1.0 | | | vbroadcastsd zmm4, xmm6 | 1 | | | | | | | 1.0 | | sub r11d, r14d | 1 | | 0.5 | | | | | 0.5 | | add rcx, r10 | 1 | | | | | | 1.0 | | | vpbroadcastd ymm0, r11d | 1 | | | | | | 1.0 | | | vpcmpgtd k3, ymm0, ymm15 | 1 | | 0.5 | | | | | 0.5 | | movsxd r14, r14d | 1 | 1.0 | | | | | | | | kmovw ebx, k3 | 2 | | 1.0 | 1.0 1.0 | | | | | | vmovdqu32 ymm1{k3}{z}, ymmword ptr [rcx+r14*4] | 1 | | 1.0 | | | | | | | vpaddd ymm2, ymm1, ymm1 | 1 | | 1.0 | | | | | | | vpaddd ymm0, ymm1, ymm2 | 1 | 1.0 | | | | | | | | kmovw k1, k3 | 1 | 1.0 | | | | | | | | kmovw k2, k3 | 1* | | | | | | | | | vpxord zmm1, zmm1, zmm1 | 1* | | | | | | | | | vpxord zmm2, zmm2, zmm2 | 1* | | | | | | | | | vpxord zmm3, zmm3, zmm3 | 5^ | 1.0 | | 4.0 4.0 | 4.0 4.0 | | 1.0 | 1.0 | | vgatherdpd zmm1, k1, zmmword ptr [rdi+ymm0*8+0x10] | 5^ | 1.0 | | 4.0 4.0 | 4.0 4.0 | | 1.0 | 1.0 | | vgatherdpd zmm2, k2, zmmword ptr [rdi+ymm0*8+0x8] | 5^ | 1.0 | | 4.0 4.0 | 4.0 4.0 | | 1.0 | 1.0 | | vgatherdpd zmm3, k3, zmmword ptr [rdi+ymm0*8] | 1 | | | | | | 1.0 | | | vbroadcastsd zmm7, xmm7 | 1 | | | | | | 1.0 | | | vbroadcastsd zmm12, xmm12 | 1 | 1.0 | | | | | | | | vsubpd zmm23, zmm12, zmm1 | 1 | 1.0 | | | | | | | | vsubpd zmm21, zmm7, zmm2 | 1 | 0.5 | | | | | 0.5 | | | vsubpd zmm20, zmm4, zmm3 | 1 | 0.5 | | | | | 0.5 | | | vmulpd zmm19, zmm21, zmm21 | 1 | 0.5 | | | | | 0.5 | | | vfmadd231pd zmm19, zmm20, zmm20 | 1 | 0.5 | | | | | 0.5 | | | vfmadd231pd zmm19, zmm23, zmm23 | 3 | 2.0 | | | | | 1.0 | | | vrcp14pd zmm18, zmm19 | 1 | | | | | | 1.0 | | | vcmppd k2, zmm19, zmm14, 0x1 | 1 | | | | | | 1.0 | | | vfpclasspd k0, zmm18, 0x1e | 1 | 1.0 | | | | | | | | kmovw ecx, k2 | 1 | 1.0 | | | | | | | | knotw k1, k0 | 1* | | | | | | | | | vmovaps zmm0, zmm19 | 1 | | 0.5 | | | | | 0.5 | | and ebx, ecx | 2^ | | | | 1.0 1.0 | | 1.0 | | | vfnmadd213pd zmm0, zmm18, qword ptr [rip]{1to8} | 1 | | | | | | 1.0 | | | kmovw k3, ebx | 1 | 1.0 | | | | | | | | vmulpd zmm1, zmm0, zmm0 | 1 | 0.5 | | | | | 0.5 | | | vfmadd213pd zmm18{k1}, zmm0, zmm18 | 1 | 0.5 | | | | | 0.5 | | | vfmadd213pd zmm18{k1}, zmm1, zmm18 | 1 | 0.5 | | | | | 0.5 | | | vmulpd zmm2, zmm18, zmm13 | 1 | 0.5 | | | | | 0.5 | | | vmulpd zmm4, zmm18, zmm10 | 1 | 0.5 | | | | | 0.5 | | | vmulpd zmm6, zmm18, zmm2 | 1 | 0.5 | | | | | 0.5 | | | vmulpd zmm3, zmm18, zmm6 | 1 | 0.5 | | | | | 0.5 | | | vfmsub213pd zmm18, zmm6, zmm5 | 1 | 0.5 | | | | | 0.5 | | | vmulpd zmm17, zmm3, zmm4 | 1 | 0.5 | | | | | 0.5 | | | vmulpd zmm22, zmm17, zmm18 | 1 | 0.5 | | | | | 0.5 | | | vfmadd231pd zmm9{k3}, zmm22, zmm20 | 1 | 0.5 | | | | | 0.5 | | | vfmadd231pd zmm8{k3}, zmm22, zmm21 | 1 | 0.5 | | | | | 0.5 | | | vfmadd231pd zmm11{k3}, zmm22, zmm23 Total Num Of Uops: 65 Analysis Notes: Backend allocation was stalled due to unavailable allocation resources. There were bubbles in the frontend.