MD-Bench/arch_analysis/avx512/osaca_force_soa_lt8_iln1000.txt
Rafael Ravedutti 9c28ff1e9e Add arch_analysis directory and first AVX2 results
Signed-off-by: Rafael Ravedutti <rafaelravedutti@gmail.com>
2021-05-06 23:18:28 +02:00

71 lines
8.2 KiB
Plaintext

iwia021h@testfront1:~/MD-Bench$ /home/hpc/iwia/iwia021h/.local/bin/osaca --ignore-unknown --arch=CSX ICC/force.s
Open Source Architecture Code Analyzer (OSACA) - 0.3.14
Analyzed file: ICC/force.s
Architecture: CSX
Timestamp: 2021-04-30 16:08:44
P - Throughput of LOAD operation can be hidden behind a past or future STORE instruction
* - Instruction micro-ops not bound to a port
X - No throughput/latency information for this instruction in data file
Combined Analysis Report
------------------------
Port pressure in cycles
| 0 - 0DV | 1 | 2 - 2D | 3 - 3D | 4 | 5 | 6 | 7 || CP | LCD |
-------------------------------------------------------------------------------------------------
306 | | | | | | | | || | | # LOE rbp rdi r8 r9 r10 edx ecx r11d r12d r13d r14d r15d ymm13 ymm14 zmm2 zmm3 zmm4 zmm5 zmm6 zmm7 zmm8 zmm9 zmm10 zmm11 zmm12 zmm15
307 | | | | | | | | || | | ..B1.29: # Preds ..B1.28
308 | | | | | | | | || | | # Execution count [2.50e+04]
309 | 0.00 | 0.50 | | | | 0.00 | 0.50 | || | | movl %r14d, %eax #64.13
310 | | | | | | | | || | | X subl %ecx, %eax #64.13
311 | | | | | | | | || | | X vpbroadcastd %eax, %ymm0 #64.13
312 | | | | | | 1.00 | | || | | vpcmpgtd %ymm14, %ymm0, %k5 #64.13
313 | 0.00 | 0.16 | | | | 0.00 | 0.84 | || | | movslq %ecx, %rcx #64.13
314 | | | | | | | | || | | * vmovaps %zmm15, %zmm17 #67.40
315 | 1.00 | | | | | | | || | | kmovw %k5, %k2 #67.40
316 | | | | | | | | || | | * vmovaps %zmm15, %zmm16 #66.40
317 | 0.00 | 1.00 | 0.50 0.50 | 0.50 0.50 | | 0.00 | | || 4.0 | | vmovdqu32 (%rdi,%rcx,4), %ymm1{%k5}{z} #65.25
318 | 1.00 | | | | | | | || | | kmovw %k5, %k1 #66.40
319 | | | | | | | | || | | * vmovaps %zmm15, %zmm18 #68.40
320 | 1.00 | | | | | | | || | | kmovw %k5, %k3 #68.40
321 | 1.50 | 0.50 | 4.00 0.50 | 4.00 0.50 | | 0.50 | 0.50 | || | | vgatherdpd (%r8,%ymm1,8), %zmm18{%k3} #68.40
322 | 1.50 | 0.50 | 4.00 0.50 | 4.00 0.50 | | 0.50 | 0.50 | || 4.0 | | vgatherdpd (%r9,%ymm1,8), %zmm17{%k2} #67.40
323 | 1.50 | 0.34 | 4.00 0.50 | 4.00 0.50 | | 0.50 | 0.66 | || | | vgatherdpd (%r10,%ymm1,8), %zmm16{%k1} #66.40
324 | 0.50 | | | | | 0.50 | | || | | vsubpd %zmm18, %zmm3, %zmm31 #68.40
325 | 0.50 | | | | | 0.50 | | || 4.0 | | vsubpd %zmm17, %zmm2, %zmm29 #67.40
326 | 0.50 | | | | | 0.50 | | || | | vsubpd %zmm16, %zmm4, %zmm28 #66.40
327 | 0.50 | | | | | 0.50 | | || 4.0 | | vmulpd %zmm29, %zmm29, %zmm27 #69.53
328 | 0.50 | | | | | 0.50 | | || 4.0 | | vfmadd231pd %zmm28, %zmm28, %zmm27 #69.53
329 | 0.50 | | | | | 0.50 | | || 4.0 | | vfmadd231pd %zmm31, %zmm31, %zmm27 #69.67
330 | 2.50 | | | | | 0.50 | | || 8.0 | | vrcp14pd %zmm27, %zmm26 #72.42
331 | | | | | | 1.00 | | || | | vcmppd $1, %zmm12, %zmm27, %k6{%k5} #71.26
332 | | | | | | 1.00 | | || | | vfpclasspd $30, %zmm26, %k0 #72.42
333 | | | | | | | | || | | * vmovaps %zmm27, %zmm19 #72.42
334 | 0.50 | | 0.50 0.50 | 0.50 0.50 | | 0.50 | | || 4.0 | | vfnmadd213pd .L_2il0floatpacket.5(%rip){1to8}, %zmm26, %zmm19 #72.42
335 | 1.00 | | | | | | | || | | knotw %k0, %k4 #72.42
336 | 0.50 | | | | | 0.50 | | || 4.0 | | vmulpd %zmm19, %zmm19, %zmm20 #72.42
337 | 0.50 | | | | | 0.50 | | || | | vfmadd213pd %zmm26, %zmm19, %zmm26{%k4} #72.42
338 | 0.50 | | | | | 0.50 | | || 4.0 | | vfmadd213pd %zmm26, %zmm20, %zmm26{%k4} #72.42
339 | 0.50 | | | | | 0.50 | | || 4.0 | | vmulpd %zmm11, %zmm26, %zmm21 #73.42
340 | 0.50 | | | | | 0.50 | | || | | vmulpd %zmm9, %zmm26, %zmm23 #74.58
341 | 0.50 | | | | | 0.50 | | || 4.0 | | vmulpd %zmm21, %zmm26, %zmm24 #73.48
342 | 0.50 | | | | | 0.50 | | || 4.0 | | vmulpd %zmm24, %zmm26, %zmm22 #73.54
343 | 0.00 | | | | | 1.00 | | || | | vfmsub213pd %zmm10, %zmm24, %zmm26 #74.58
344 | 0.00 | | | | | 1.00 | | || 4.0 | | vmulpd %zmm23, %zmm22, %zmm25 #74.65
345 | 0.00 | | | | | 1.00 | | || 4.0 | | vmulpd %zmm26, %zmm25, %zmm30 #74.71
346 | 0.00 | | | | | 1.00 | | || | | vfmadd231pd %zmm28, %zmm30, %zmm7{%k6} #75.21
347 | 0.00 | | | | | 1.00 | | || | | vfmadd231pd %zmm29, %zmm30, %zmm6{%k6} #76.21
348 | 0.00 | | | | | 1.00 | | || 4.0 | 4.0 | vfmadd231pd %zmm31, %zmm30, %zmm5{%k6} #77.21
18.0 3.00 13.0 2.50 13.0 2.50 18.0 3.00 68.0 4
Loop-Carried Dependencies Analysis Report
-----------------------------------------
313 | 1.0 | movslq %ecx, %rcx #64.13| [313]
348 | 4.0 | vfmadd231pd %zmm31, %zmm30, %zmm5{%k6} #77.21| [348]
347 | 4.0 | vfmadd231pd %zmm29, %zmm30, %zmm6{%k6} #76.21| [347]
346 | 4.0 | vfmadd231pd %zmm28, %zmm30, %zmm7{%k6} #75.21| [346]