Forum > Discussion Forum
AVX Optimized App Development
arkayn:
--- Code: ---=========================================================
Ftst_v7_J34 started.
Optimal function choices:
--------------------------------------------------------
name timing error
--------------------------------------------------------
v_BaseLineSmooth (no other)
v_GetPowerSpectrum 0.000480 0.00000 test
v_vGetPowerSpectrum 0.000301 0.00000 test
v_vGetPowerSpectrum2 0.000327 0.00000 test
v_vGetPowerSpectrumUnrolled 0.000314 0.00000 test
v_vGetPowerSpectrumUnrolled2 0.000294 0.00000 test
v_vGetPowerSpectrumUnrolled2 0.000294 0.00000 choice
v_ChirpData 0.019478 0.00000 test
fpu_ChirpData 0.025356 0.00000 test
fpu_opt_ChirpData 0.015757 0.00000 test
v_vChirpData_x86_64 0.079464 0.00000 test
sse1_ChirpData_ak 0.011689 0.00000 test
sse2_ChirpData_ak 0.011893 0.00000 test
sse2_ChirpData_ak8 0.008098 0.00000 test
sse3_ChirpData_ak 0.011029 0.00000 test
sse2_ChirpData_ak8 0.008098 0.00000 choice
v_Transpose 0.041660 0.00000 test
v_Transpose2 0.025839 0.00000 test
v_Transpose4 0.012987 0.00000 test
v_Transpose8 0.020351 0.00000 test
v_pfTranspose2 0.025092 0.00000 test
v_pfTranspose4 0.012726 0.00000 test
v_pfTranspose8 0.019991 0.00000 test
v_vTranspose4 0.012808 0.00000 test
v_vTranspose4np 0.013273 0.00000 test
v_vTranspose4ntw 0.008225 0.00000 test
v_vTranspose4x8ntw 0.008911 0.00000 test
v_vTranspose4x16ntw 0.007548 0.00000 test
v_vpfTranspose8x4ntw 0.008831 0.00000 test
v_vTranspose4x16ntw 0.007548 0.00000 choice
FPU opt folding 0.003467 0.00000 test
AK SSE folding 0.001317 0.00000 test
BH SSE folding 0.001285 0.00000 test
BH SSE folding 0.001285 0.00000 choice
Test duration 6.44 seconds
Ftst_v7 completed successfully.
--- End code ---
Miep:
1st boinc running 2nd snoozed
Claggy:
@arkayn your posted stderr.txt says Ftst_v7_J34 and not J37
Claggy
Claggy:
Here'a run on my C2D E8500 @4.14GHz with J37 (5 times with Boinc and apps running and 5 times with Boinc shut down)
Claggy
Fredericx51:
Did some reading about AVX and checked its output with this Test-file.
Whithout BOINC running: sterr.txt :
=========================================================
Ftst_v7_J34 started.
Optimal function choices:
--------------------------------------------------------
name timing error
--------------------------------------------------------
v_BaseLineSmooth (no other)
v_GetPowerSpectrum 0.000105 0.00000 test
v_vGetPowerSpectrum 0.000052 0.00000 test
v_vGetPowerSpectrum2 0.000063 0.00000 test
v_vGetPowerSpectrumUnrolled 0.000049 0.00000 test
v_vGetPowerSpectrumUnrolled2 0.000066 0.00000 test
v_avxGetPowerSpectrum 0.000043 0.00000 test
v_avxGetPowerSpectrum 0.000043 0.00000 choice
v_ChirpData 0.005899 0.00000 test
fpu_ChirpData 0.010711 0.00000 test
fpu_opt_ChirpData 0.005305 0.00000 test
v_vChirpData_x86_64 0.051195 0.00000 test
sse1_ChirpData_ak 0.006250 0.00000 test
sse2_ChirpData_ak 0.005789 0.00000 test
sse2_ChirpData_ak8 0.003679 0.00000 test
sse3_ChirpData_ak 0.005621 0.00000 test
avx_ChirpData_a 0.001884 0.00000 test
avx_ChirpData_b 0.002139 0.00000 test
avx_ChirpData_a 0.001884 0.00000 choice
v_Transpose 0.002753 0.00000 test
v_Transpose2 0.002947 0.00000 test
v_Transpose4 0.001516 0.00000 test
v_Transpose8 0.002775 0.00000 test
v_pfTranspose2 0.001659 0.00000 test
v_pfTranspose4 0.001586 0.00000 test
v_pfTranspose8 0.002802 0.00000 test
v_vTranspose4 0.000915 0.00000 test
v_vTranspose4np 0.001169 0.00000 test
v_vTranspose4ntw 0.007690 0.00000 test
v_vTranspose4x8ntw 0.003222 0.00000 test
v_vTranspose4x16ntw 0.000900 0.00000 test
v_vpfTranspose8x4ntw 0.007704 0.00000 test
v_avxTranspose4x8ntw 0.003195 0.00000 test
v_avxTranspose4x16ntw 0.000817 0.00000 test
v_avxTranspose8x4ntw 0.007712 0.00000 test
v_avxTranspose8x8ntw_a 0.002666 0.00000 test
v_avxTranspose8x8ntw_b 0.003011 0.00000 test
v_avxTranspose4x16ntw 0.000817 0.00000 choice
FPU opt folding 0.002047 0.00000 test
AK SSE folding 0.000464 0.00000 test
BH SSE folding 0.000451 0.00000 test
JS AVX folding 0.000405 0.00000 test
JS AVX folding 0.000405 0.00000 choice
Test duration 2.90 seconds
Ftst_v7 completed successfully.
With BOINC (6.10.60 X64)(i7-2600 + 2x HD5870) 8x MB (CPU)+ 4 ATi MB rev.177 or AP rev.524.
=========================================================
Ftst_v7_J34 started.
Optimal function choices:
--------------------------------------------------------
name timing error
--------------------------------------------------------
v_BaseLineSmooth (no other)
v_GetPowerSpectrum 0.000105 0.00000 test
v_vGetPowerSpectrum 0.000052 0.00000 test
v_vGetPowerSpectrum2 0.000063 0.00000 test
v_vGetPowerSpectrumUnrolled 0.000049 0.00000 test
v_vGetPowerSpectrumUnrolled2 0.000066 0.00000 test
v_avxGetPowerSpectrum 0.000043 0.00000 test
v_avxGetPowerSpectrum 0.000043 0.00000 choice
v_ChirpData 0.005899 0.00000 test
fpu_ChirpData 0.010711 0.00000 test
fpu_opt_ChirpData 0.005305 0.00000 test
v_vChirpData_x86_64 0.051195 0.00000 test
sse1_ChirpData_ak 0.006250 0.00000 test
sse2_ChirpData_ak 0.005789 0.00000 test
sse2_ChirpData_ak8 0.003679 0.00000 test
sse3_ChirpData_ak 0.005621 0.00000 test
avx_ChirpData_a 0.001884 0.00000 test
avx_ChirpData_b 0.002139 0.00000 test
avx_ChirpData_a 0.001884 0.00000 choice
v_Transpose 0.002753 0.00000 test
v_Transpose2 0.002947 0.00000 test
v_Transpose4 0.001516 0.00000 test
v_Transpose8 0.002775 0.00000 test
v_pfTranspose2 0.001659 0.00000 test
v_pfTranspose4 0.001586 0.00000 test
v_pfTranspose8 0.002802 0.00000 test
v_vTranspose4 0.000915 0.00000 test
v_vTranspose4np 0.001169 0.00000 test
v_vTranspose4ntw 0.007690 0.00000 test
v_vTranspose4x8ntw 0.003222 0.00000 test
v_vTranspose4x16ntw 0.000900 0.00000 test
v_vpfTranspose8x4ntw 0.007704 0.00000 test
v_avxTranspose4x8ntw 0.003195 0.00000 test
v_avxTranspose4x16ntw 0.000817 0.00000 test
v_avxTranspose8x4ntw 0.007712 0.00000 test
v_avxTranspose8x8ntw_a 0.002666 0.00000 test
v_avxTranspose8x8ntw_b 0.003011 0.00000 test
v_avxTranspose4x16ntw 0.000817 0.00000 choice
FPU opt folding 0.002047 0.00000 test
AK SSE folding 0.000464 0.00000 test
BH SSE folding 0.000451 0.00000 test
JS AVX folding 0.000405 0.00000 test
JS AVX folding 0.000405 0.00000 choice
Test duration 2.90 seconds
Ftst_v7 completed successfully.
=========================================================
Ftst_v7_J34 started.
Optimal function choices:
--------------------------------------------------------
name timing error
--------------------------------------------------------
v_BaseLineSmooth (no other)
v_GetPowerSpectrum 0.000234 0.00000 test
v_vGetPowerSpectrum 0.000105 0.00000 test
v_vGetPowerSpectrum2 0.000100 0.00000 test
v_vGetPowerSpectrumUnrolled 0.000082 0.00000 test
v_vGetPowerSpectrumUnrolled2 0.000098 0.00000 test
v_avxGetPowerSpectrum 0.000061 0.00000 test
v_avxGetPowerSpectrum 0.000061 0.00000 choice
v_ChirpData 0.011899 0.00000 test
fpu_ChirpData 0.019045 0.00000 test
fpu_opt_ChirpData 0.012640 0.00000 test
v_vChirpData_x86_64 0.063979 0.00000 test
sse1_ChirpData_ak 0.010132 0.00000 test
sse2_ChirpData_ak 0.009260 0.00000 test
sse2_ChirpData_ak8 0.006961 0.00000 test
sse3_ChirpData_ak 0.008636 0.00000 test
avx_ChirpData_a 0.003490 0.00000 test
avx_ChirpData_b 0.003833 0.00000 test
avx_ChirpData_a 0.003490 0.00000 choice
v_Transpose 0.007700 0.00000 test
v_Transpose2 0.004792 0.00000 test
v_Transpose4 0.008537 0.00000 test
v_Transpose8 0.014129 0.00000 test
v_pfTranspose2 0.015112 0.00000 test
v_pfTranspose4 0.011302 0.00000 test
v_pfTranspose8 0.012998 0.00000 test
v_vTranspose4 0.002625 0.00000 test
v_vTranspose4np 0.005798 0.00000 test
v_vTranspose4ntw 0.008330 0.00000 test
v_vTranspose4x8ntw 0.004689 0.00000 test
v_vTranspose4x16ntw 0.002755 0.00000 test
v_vpfTranspose8x4ntw 0.008488 0.00000 test
v_avxTranspose4x8ntw 0.003759 0.00000 test
v_avxTranspose4x16ntw 0.002249 0.00000 test
v_avxTranspose8x4ntw 0.008294 0.00000 test
v_avxTranspose8x8ntw_a 0.003551 0.00000 test
v_avxTranspose8x8ntw_b 0.004706 0.00000 test
v_avxTranspose4x16ntw 0.002249 0.00000 choice
FPU opt folding 0.003407 0.00000 test
AK SSE folding 0.000878 0.00000 test
BH SSE folding 0.000816 0.00000 test
JS AVX folding 0.000656 0.00000 test
JS AVX folding 0.000656 0.00000 choice
Test duration 5.03 seconds
Ftst_v7 completed successfully.
Navigation
[0] Message Index
[#] Next page
[*] Previous page
Go to full version