LLVM Clang 13 Bencmarking Intel Xeon Ice Lake

LLVM Clang benchmarks for a future article.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2108236-TJ-LLVMCLANG38
Jump To Table - Results

View

Do Not Show Noisy Results
Do Not Show Results With Incomplete Data
Do Not Show Results With Little Change/Spread
List Notable Results

Limit displaying results to tests within:

AV1 3 Tests
C/C++ Compiler Tests 18 Tests
Compression Tests 2 Tests
CPU Massive 17 Tests
Creator Workloads 15 Tests
Cryptography 4 Tests
Database Test Suite 3 Tests
Encoding 7 Tests
Finance 2 Tests
Go Language Tests 2 Tests
HPC - High Performance Computing 4 Tests
Imaging 2 Tests
Common Kernel Benchmarks 3 Tests
Machine Learning 3 Tests
Multi-Core 17 Tests
NVIDIA GPU Compute 2 Tests
Programmer / Developer System Benchmarks 3 Tests
Raytracing 2 Tests
Renderers 3 Tests
Server 5 Tests
Server CPU Tests 9 Tests
Single-Threaded 2 Tests
Telephony 2 Tests
Video Encoding 6 Tests

Statistics

Show Overall Harmonic Mean(s)
Show Overall Geometric Mean
Show Geometric Means Per-Suite/Category
Show Wins / Losses Counts (Pie Chart)
Normalize Results
Remove Outliers Before Calculating Averages

Graph Settings

Force Line Graphs Where Applicable
Convert To Scalar Where Applicable
Prefer Vertical Bar Graphs

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Hide
Result
Result
Identifier
View Logs
Performance Per
Dollar
Date
Run
  Test
  Duration
Clang 11
August 23 2021
  11 Hours, 2 Minutes
Clang 12
August 22 2021
  6 Hours, 55 Minutes
Clang 13
August 22 2021
  5 Hours, 53 Minutes
Invert Hiding All Results Option
  7 Hours, 56 Minutes

Only show results where is faster than
Only show results matching title/arguments (delimit multiple options with a comma):
Do not show results matching title/arguments (delimit multiple options with a comma):

LLVM Clang 13 Bencmarking Intel Xeon Ice Lake LLVM Clang benchmarks for a future article. Clang 11: Processor: 2 x Intel Xeon Platinum 8380 @ 3.40GHz (80 Cores / 160 Threads), Motherboard: Intel M50CYP2SB2U (SE5C6200.86B.0022.D08.2103221623 BIOS), Chipset: Intel Device 0998, Memory: 504GB, Disk: 7682GB INTEL SSDPF2KX076TZ, Graphics: ASPEED, Monitor: VE228, Network: 2 x Intel X710 for 10GBASE-T + 2 x Intel E810-C for QSFP OS: Ubuntu 21.04, Kernel: 5.14.0-rc1-folio (x86_64) 20210715, Desktop: GNOME Shell 3.38.4, Display Server: X Server 1.20.11, Compiler: Clang 11.0.1-2ubuntu4, File-System: ext4, Screen Resolution: 1920x1080 Clang 12: Processor: 2 x Intel Xeon Platinum 8380 @ 3.40GHz (80 Cores / 160 Threads), Motherboard: Intel M50CYP2SB2U (SE5C6200.86B.0022.D08.2103221623 BIOS), Chipset: Intel Device 0998, Memory: 504GB, Disk: 7682GB INTEL SSDPF2KX076TZ, Graphics: ASPEED, Monitor: VE228, Network: 2 x Intel X710 for 10GBASE-T + 2 x Intel E810-C for QSFP OS: Ubuntu 21.04, Kernel: 5.14.0-rc1-folio (x86_64) 20210715, Desktop: GNOME Shell 3.38.4, Display Server: X Server 1.20.11, Compiler: Clang 12.0.0-3ubuntu1~21.04.1, File-System: ext4, Screen Resolution: 1920x1080 Clang 13: Processor: 2 x Intel Xeon Platinum 8380 @ 3.40GHz (80 Cores / 160 Threads), Motherboard: Intel M50CYP2SB2U (SE5C6200.86B.0022.D08.2103221623 BIOS), Chipset: Intel Device 0998, Memory: 504GB, Disk: 7682GB INTEL SSDPF2KX076TZ, Graphics: ASPEED, Monitor: VE228, Network: 2 x Intel X710 for 10GBASE-T + 2 x Intel E810-C for QSFP OS: Ubuntu 21.04, Kernel: 5.14.0-rc1-folio (x86_64) 20210715, Desktop: GNOME Shell 3.38.4, Display Server: X Server 1.20.11, Compiler: Clang 13.0.0-++20210820072921+23ba3732246a-1~exp1~20210820174536.53, File-System: ext4, Screen Resolution: 1920x1080 TNN 0.3 Target: CPU - Model: SqueezeNet v1.1 ms < Lower Is Better Clang 11 . 657.10 |============================================================ Clang 12 . 402.26 |===================================== Clang 13 . 400.26 |===================================== TNN 0.3 Target: CPU - Model: DenseNet ms < Lower Is Better Clang 11 . 6912.12 |=========================================================== Clang 12 . 4353.39 |===================================== Clang 13 . 4370.10 |===================================== TNN 0.3 Target: CPU - Model: MobileNet v2 ms < Lower Is Better Clang 11 . 773.47 |============================================================ Clang 12 . 541.93 |========================================== Clang 13 . 539.65 |========================================== TNN 0.3 Target: CPU - Model: SqueezeNet v2 ms < Lower Is Better Clang 11 . 105.40 |============================================================ Clang 12 . 83.86 |================================================ Clang 13 . 84.61 |================================================ C-Ray 1.1 Total Time - 4K, 16 Rays Per Pixel Seconds < Lower Is Better Clang 11 . 14.80 |=========================================================== Clang 12 . 15.22 |============================================================= Clang 13 . 12.59 |================================================== Apache HTTP Server 2.4.48 Concurrent Requests: 500 Requests Per Second > Higher Is Better Clang 11 . 122576.13 |================================================ Clang 12 . 121643.80 |================================================ Clang 13 . 145589.94 |========================================================= GraphicsMagick 1.3.33 Operation: Sharpen Iterations Per Minute > Higher Is Better Clang 11 . 745 |====================================================== Clang 12 . 758 |====================================================== Clang 13 . 877 |=============================================================== John The Ripper 1.9.0-jumbo-1 Test: MD5 Real C/S > Higher Is Better Clang 11 . 8880000 |================================================= Clang 12 . 10275333 |========================================================= Clang 13 . 10430000 |========================================================== LibRaw 0.20 Post-Processing Benchmark Mpix/sec > Higher Is Better Clang 11 . 34.55 |===================================================== Clang 12 . 36.65 |======================================================== Clang 13 . 39.99 |============================================================= VP9 libvpx Encoding 1.10.0 Speed: Speed 5 - Input: Bosphorus 4K Frames Per Second > Higher Is Better Clang 11 . 14.32 |========================================================= Clang 12 . 14.11 |======================================================== Clang 13 . 15.35 |============================================================= Zstd Compression 1.5.0 Compression Level: 8 - Compression Speed MB/s > Higher Is Better Clang 11 . 2729.0 |=========================================================== Clang 12 . 2560.6 |======================================================= Clang 13 . 2775.9 |============================================================ Apache HTTP Server 2.4.48 Concurrent Requests: 1000 Requests Per Second > Higher Is Better Clang 11 . 120827.45 |========================================================= Clang 12 . 112287.17 |===================================================== Clang 13 . 112600.02 |===================================================== oneDNN 2.1.2 Harness: Deconvolution Batch shapes_1d - Data Type: bf16bf16bf16 - Engine: CPU ms < Lower Is Better Clang 11 . 3.05315 |======================================================= Clang 12 . 3.05821 |======================================================= Clang 13 . 3.28114 |=========================================================== GraphicsMagick 1.3.33 Operation: Enhanced Iterations Per Minute > Higher Is Better Clang 11 . 1104 |========================================================== Clang 12 . 1104 |========================================================== Clang 13 . 1174 |============================================================== Zstd Compression 1.5.0 Compression Level: 3 - Compression Speed MB/s > Higher Is Better Clang 11 . 6433.3 |========================================================== Clang 12 . 6321.0 |========================================================= Clang 13 . 6697.9 |============================================================ NCNN 20210720 Target: CPU - Model: yolov4-tiny ms < Lower Is Better Clang 11 . 26.79 |========================================================== Clang 12 . 28.35 |============================================================= Clang 13 . 27.94 |============================================================ Tachyon 0.99b6 Total Time Seconds < Lower Is Better Clang 11 . 14.25 |============================================================= Clang 12 . 13.95 |============================================================ Clang 13 . 13.47 |========================================================== Zstd Compression 1.5.0 Compression Level: 19 - Compression Speed MB/s > Higher Is Better Clang 11 . 84.8 |============================================================== Clang 12 . 80.4 |=========================================================== Clang 13 . 83.6 |============================================================= Opus Codec Encoding 1.3.1 WAV To Opus Encode Seconds < Lower Is Better Clang 11 . 9.423 |========================================================== Clang 12 . 9.908 |============================================================= Clang 13 . 9.632 |=========================================================== SecureMark 1.0.4 Benchmark: SecureMark-TLS marks > Higher Is Better Clang 11 . 252089 |============================================================ Clang 12 . 250849 |============================================================ Clang 13 . 240447 |========================================================= Zstd Compression 1.5.0 Compression Level: 3, Long Mode - Compression Speed MB/s > Higher Is Better Clang 11 . 826.7 |========================================================== Clang 12 . 865.6 |============================================================= Clang 13 . 863.1 |============================================================= SVT-VP9 0.3 Tuning: PSNR/SSIM Optimized - Input: Bosphorus 1080p Frames Per Second > Higher Is Better Clang 11 . 460.20 |========================================================== Clang 12 . 456.18 |========================================================= Clang 13 . 476.72 |============================================================ Google SynthMark 20201109 Test: VoiceMark_100 Voices > Higher Is Better Clang 11 . 491.03 |========================================================= Clang 12 . 503.93 |=========================================================== Clang 13 . 513.01 |============================================================ Zstd Compression 1.5.0 Compression Level: 8, Long Mode - Compression Speed MB/s > Higher Is Better Clang 11 . 798.8 |========================================================== Clang 12 . 833.6 |============================================================= Clang 13 . 828.0 |============================================================= AOM AV1 3.1 Encoder Mode: Speed 9 Realtime - Input: Bosphorus 4K Frames Per Second > Higher Is Better Clang 11 . 54.44 |=========================================================== Clang 12 . 54.47 |=========================================================== Clang 13 . 56.65 |============================================================= SVT-HEVC 1.5.0 Tuning: 7 - Input: Bosphorus 1080p Frames Per Second > Higher Is Better Clang 11 . 344.59 |========================================================== Clang 12 . 340.21 |========================================================== Clang 13 . 353.72 |============================================================ C-Blosc 2.0 Compressor: blosclz MB/s > Higher Is Better Clang 11 . 23751.9 |========================================================= Clang 12 . 24249.3 |========================================================== Clang 13 . 24671.5 |=========================================================== Coremark 1.0 CoreMark Size 666 - Iterations Per Second Iterations/Sec > Higher Is Better Clang 11 . 2103884.89 |====================================================== Clang 12 . 2146447.38 |======================================================= Clang 13 . 2185330.04 |======================================================== GraphicsMagick 1.3.33 Operation: Noise-Gaussian Iterations Per Minute > Higher Is Better Clang 11 . 623 |=============================================================== Clang 12 . 600 |============================================================= Clang 13 . 600 |============================================================= Botan 2.17.3 Test: Twofish - Decrypt MiB/s > Higher Is Better Clang 11 . 279.96 |========================================================== Clang 12 . 288.61 |============================================================ Clang 13 . 290.47 |============================================================ PJSIP 2.11 Method: OPTIONS, Stateless Responses Per Second > Higher Is Better Clang 11 . 41305 |============================================================ Clang 12 . 40364 |=========================================================== Clang 13 . 41871 |============================================================= Botan 2.17.3 Test: KASUMI - Decrypt MiB/s > Higher Is Better Clang 11 . 73.56 |=========================================================== Clang 12 . 76.24 |============================================================= Clang 13 . 76.25 |============================================================= Botan 2.17.3 Test: KASUMI MiB/s > Higher Is Better Clang 11 . 74.40 |=========================================================== Clang 12 . 76.92 |============================================================= Clang 13 . 77.09 |============================================================= SVT-HEVC 1.5.0 Tuning: 10 - Input: Bosphorus 1080p Frames Per Second > Higher Is Better Clang 11 . 616.45 |=========================================================== Clang 12 . 604.45 |========================================================== Clang 13 . 626.33 |============================================================ SVT-VP9 0.3 Tuning: Visual Quality Optimized - Input: Bosphorus 1080p Frames Per Second > Higher Is Better Clang 11 . 362.67 |========================================================== Clang 12 . 365.79 |=========================================================== Clang 13 . 374.91 |============================================================ Zstd Compression 1.5.0 Compression Level: 19, Long Mode - Compression Speed MB/s > Higher Is Better Clang 11 . 45.5 |============================================================ Clang 12 . 45.7 |============================================================ Clang 13 . 47.0 |============================================================== Botan 2.17.3 Test: ChaCha20Poly1305 MiB/s > Higher Is Better Clang 11 . 869.35 |=========================================================== Clang 12 . 882.17 |============================================================ Clang 13 . 855.76 |========================================================== Botan 2.17.3 Test: ChaCha20Poly1305 - Decrypt MiB/s > Higher Is Better Clang 11 . 866.10 |=========================================================== Clang 12 . 874.91 |============================================================ Clang 13 . 849.96 |========================================================== Google Draco 1.4.1 Model: Church Facade ms < Lower Is Better Clang 11 . 7000 |============================================================ Clang 12 . 7057 |============================================================= Clang 13 . 7202 |============================================================== AOM AV1 3.1 Encoder Mode: Speed 8 Realtime - Input: Bosphorus 4K Frames Per Second > Higher Is Better Clang 11 . 44.06 |=========================================================== Clang 12 . 44.84 |============================================================ Clang 13 . 45.30 |============================================================= Botan 2.17.3 Test: Twofish MiB/s > Higher Is Better Clang 11 . 286.03 |========================================================== Clang 12 . 293.73 |============================================================ Clang 13 . 288.81 |=========================================================== QuantLib 1.21 MFLOPS > Higher Is Better Clang 11 . 2589.5 |========================================================== Clang 12 . 2606.3 |=========================================================== Clang 13 . 2657.3 |============================================================ FinanceBench 2016-07-25 Benchmark: Repo OpenMP ms < Lower Is Better Clang 11 . 37221.00 |========================================================= Clang 12 . 37760.89 |========================================================= Clang 13 . 38153.40 |========================================================== SVT-AV1 0.8.7 Encoder Mode: Preset 4 - Input: Bosphorus 4K Frames Per Second > Higher Is Better Clang 11 . 4.895 |============================================================ Clang 12 . 4.849 |============================================================ Clang 13 . 4.967 |============================================================= Google Draco 1.4.1 Model: Lion ms < Lower Is Better Clang 11 . 5827 |============================================================= Clang 12 . 5870 |============================================================= Clang 13 . 5967 |============================================================== Botan 2.17.3 Test: Blowfish - Decrypt MiB/s > Higher Is Better Clang 11 . 325.70 |=========================================================== Clang 12 . 326.80 |=========================================================== Clang 13 . 333.28 |============================================================ oneDNN 2.1.2 Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU ms < Lower Is Better Clang 11 . 608.60 |============================================================ Clang 12 . 611.93 |============================================================ Clang 13 . 598.02 |=========================================================== NCNN 20210720 Target: CPU - Model: mobilenet ms < Lower Is Better Clang 11 . 17.30 |============================================================ Clang 12 . 17.65 |============================================================= Clang 13 . 17.38 |============================================================ MariaDB 10.6.4 Clients: 2048 Queries Per Second > Higher Is Better Clang 11 . 700 |============================================================== Clang 12 . 698 |============================================================== Clang 13 . 712 |=============================================================== SVT-AV1 0.8.7 Encoder Mode: Preset 8 - Input: Bosphorus 4K Frames Per Second > Higher Is Better Clang 11 . 57.84 |============================================================ Clang 12 . 57.92 |============================================================ Clang 13 . 58.98 |============================================================= VP9 libvpx Encoding 1.10.0 Speed: Speed 0 - Input: Bosphorus 4K Frames Per Second > Higher Is Better Clang 11 . 5.60 |============================================================= Clang 12 . 5.62 |============================================================= Clang 13 . 5.71 |============================================================== FFTW 3.3.6 Build: Float + SSE - Size: 2D FFT Size 4096 Mflops > Higher Is Better Clang 11 . 18135 |============================================================ Clang 12 . 18443 |============================================================= Clang 13 . 18489 |============================================================= John The Ripper 1.9.0-jumbo-1 Test: Blowfish Real C/S > Higher Is Better Clang 11 . 113200 |============================================================ Clang 12 . 111335 |=========================================================== Clang 13 . 112407 |============================================================ GraphicsMagick 1.3.33 Operation: HWB Color Space Iterations Per Minute > Higher Is Better Clang 11 . 852 |============================================================== Clang 12 . 852 |============================================================== Clang 13 . 866 |=============================================================== AOM AV1 3.1 Encoder Mode: Speed 6 Realtime - Input: Bosphorus 4K Frames Per Second > Higher Is Better Clang 11 . 19.32 |============================================================ Clang 12 . 19.39 |============================================================ Clang 13 . 19.60 |============================================================= nginx 1.21.1 Concurrent Requests: 100 Requests Per Second > Higher Is Better Clang 11 . 205660.13 |======================================================== Clang 12 . 206297.92 |======================================================== Clang 13 . 208590.36 |========================================================= SQLite Speedtest 3.30 Timed Time - Size 1,000 Seconds < Lower Is Better Clang 11 . 64.42 |============================================================ Clang 12 . 65.29 |============================================================= Clang 13 . 64.64 |============================================================ Aircrack-ng 1.5.2 k/s > Higher Is Better Clang 11 . 211799.58 |========================================================= Clang 12 . 210157.65 |======================================================== Clang 13 . 212825.68 |========================================================= PostgreSQL pgbench 13.0 Scaling Factor: 100 - Clients: 250 - Mode: Read Write TPS > Higher Is Better Clang 11 . 85162 |============================================================= Clang 12 . 84132 |============================================================ Clang 13 . 84854 |============================================================= PostgreSQL pgbench 13.0 Scaling Factor: 100 - Clients: 250 - Mode: Read Write - Average Latency ms < Lower Is Better Clang 11 . 2.938 |============================================================ Clang 12 . 2.973 |============================================================= Clang 13 . 2.948 |============================================================ AOBench Size: 2048 x 2048 - Total Time Seconds < Lower Is Better Clang 11 . 36.17 |============================================================= Clang 12 . 36.43 |============================================================= Clang 13 . 36.00 |============================================================ nginx 1.21.1 Concurrent Requests: 20 Requests Per Second > Higher Is Better Clang 11 . 254323.92 |========================================================= Clang 12 . 251352.56 |======================================================== Clang 13 . 254133.55 |========================================================= Botan 2.17.3 Test: CAST-256 MiB/s > Higher Is Better Clang 11 . 114.75 |=========================================================== Clang 12 . 115.33 |============================================================ Clang 13 . 116.10 |============================================================ Botan 2.17.3 Test: CAST-256 - Decrypt MiB/s > Higher Is Better Clang 11 . 115.65 |============================================================ Clang 12 . 114.83 |============================================================ Clang 13 . 114.50 |=========================================================== oneDNN 2.1.2 Harness: Deconvolution Batch shapes_3d - Data Type: bf16bf16bf16 - Engine: CPU ms < Lower Is Better Clang 11 . 3.59463 |=========================================================== Clang 12 . 3.57952 |========================================================== Clang 13 . 3.61489 |=========================================================== nginx 1.21.1 Concurrent Requests: 200 Requests Per Second > Higher Is Better Clang 11 . 196465.21 |========================================================= Clang 12 . 194567.38 |======================================================== Clang 13 . 196023.59 |========================================================= TSCP 1.81 AI Chess Performance Nodes Per Second > Higher Is Better Clang 11 . 1477410 |========================================================== Clang 12 . 1481388 |=========================================================== Clang 13 . 1491816 |=========================================================== Botan 2.17.3 Test: AES-256 - Decrypt MiB/s > Higher Is Better Clang 11 . 5714.92 |=========================================================== Clang 12 . 5700.33 |========================================================== Clang 13 . 5753.64 |=========================================================== nginx 1.21.1 Concurrent Requests: 1 Requests Per Second > Higher Is Better Clang 11 . 74606.79 |========================================================= Clang 12 . 75015.29 |========================================================== Clang 13 . 75270.65 |========================================================== Botan 2.17.3 Test: Blowfish MiB/s > Higher Is Better Clang 11 . 326.77 |=========================================================== Clang 12 . 329.53 |============================================================ Clang 13 . 327.93 |============================================================ dav1d 0.9.1 Video Input: Summer Nature 4K FPS > Higher Is Better Clang 11 . 528.63 |============================================================ Clang 12 . 532.51 |============================================================ Clang 13 . 533.02 |============================================================ GraphicsMagick 1.3.33 Operation: Swirl Iterations Per Minute > Higher Is Better Clang 11 . 1860 |============================================================== Clang 12 . 1864 |============================================================== Clang 13 . 1874 |============================================================== oneDNN 2.1.2 Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU ms < Lower Is Better Clang 11 . 368.25 |============================================================ Clang 12 . 368.45 |============================================================ Clang 13 . 365.73 |============================================================ GraphicsMagick 1.3.33 Operation: Rotate Iterations Per Minute > Higher Is Better Clang 11 . 728 |=============================================================== Clang 12 . 723 |=============================================================== Clang 13 . 728 |=============================================================== MariaDB 10.6.4 Clients: 4096 Queries Per Second > Higher Is Better Clang 11 . 319 |=============================================================== Clang 12 . 319 |=============================================================== Clang 13 . 321 |=============================================================== oneDNN 2.1.2 Harness: Convolution Batch Shapes Auto - Data Type: bf16bf16bf16 - Engine: CPU ms < Lower Is Better Clang 11 . 2.08089 |=========================================================== Clang 12 . 2.06918 |=========================================================== Clang 13 . 2.07168 |=========================================================== PJSIP 2.11 Method: OPTIONS, Stateful Responses Per Second > Higher Is Better Clang 11 . 10195 |============================================================= Clang 12 . 10185 |============================================================= Clang 13 . 10140 |============================================================= oneDNN 2.1.2 Harness: IP Shapes 3D - Data Type: bf16bf16bf16 - Engine: CPU ms < Lower Is Better Clang 11 . 1.71640 |=========================================================== Clang 12 . 1.72558 |=========================================================== Clang 13 . 1.71770 |=========================================================== FinanceBench 2016-07-25 Benchmark: Bonds OpenMP ms < Lower Is Better Clang 11 . 58558.26 |========================================================== Clang 12 . 58502.85 |========================================================== Clang 13 . 58281.79 |========================================================== Botan 2.17.3 Test: AES-256 MiB/s > Higher Is Better Clang 11 . 5743.35 |=========================================================== Clang 12 . 5735.25 |=========================================================== Clang 13 . 5761.30 |=========================================================== nginx 1.21.1 Concurrent Requests: 500 Requests Per Second > Higher Is Better Clang 11 . 204519.00 |========================================================= Clang 12 . 204058.28 |========================================================= Clang 13 . 204880.02 |========================================================= PJSIP 2.11 Method: INVITE Responses Per Second > Higher Is Better Clang 11 . 5289 |============================================================== Clang 12 . 5291 |============================================================== Clang 13 . 5271 |============================================================== nginx 1.21.1 Concurrent Requests: 1000 Requests Per Second > Higher Is Better Clang 11 . 205224.87 |========================================================= Clang 12 . 205740.39 |========================================================= Clang 13 . 205977.68 |========================================================= oneDNN 2.1.2 Harness: IP Shapes 1D - Data Type: bf16bf16bf16 - Engine: CPU ms < Lower Is Better Clang 11 . 2.78642 |=========================================================== Clang 12 . 2.79658 |=========================================================== Clang 13 . 2.78910 |=========================================================== dav1d 0.9.1 Video Input: Chimera 1080p 10-bit FPS > Higher Is Better Clang 11 . 842.95 |============================================================ Clang 12 . 844.78 |============================================================ Clang 13 . 843.96 |============================================================ NCNN 20210720 Target: CPU - Model: regnety_400m ms < Lower Is Better Clang 11 . 36.26 |============================================================ Clang 12 . 37.11 |============================================================= Clang 13 . 23.98 |======================================= NCNN 20210720 Target: CPU - Model: squeezenet_ssd ms < Lower Is Better Clang 11 . 19.72 |============================================================= Clang 12 . 19.26 |============================================================ Clang 13 . 19.21 |=========================================================== NCNN 20210720 Target: CPU - Model: resnet50 ms < Lower Is Better Clang 11 . 22.36 |============================================================= Clang 12 . 21.48 |=========================================================== Clang 13 . 21.71 |=========================================================== NCNN 20210720 Target: CPU - Model: alexnet ms < Lower Is Better Clang 11 . 9.97 |======================================================= Clang 12 . 11.04 |============================================================= Clang 13 . 9.64 |===================================================== NCNN 20210720 Target: CPU - Model: googlenet ms < Lower Is Better Clang 11 . 21.59 |============================================================= Clang 12 . 21.24 |============================================================ Clang 13 . 17.87 |================================================== NCNN 20210720 Target: CPU - Model: blazeface ms < Lower Is Better Clang 11 . 6.07 |====================================================== Clang 12 . 6.94 |============================================================== Clang 13 . 4.86 |=========================================== NCNN 20210720 Target: CPU - Model: efficientnet-b0 ms < Lower Is Better Clang 11 . 15.32 |============================================================= Clang 12 . 13.75 |======================================================= Clang 13 . 12.66 |================================================== NCNN 20210720 Target: CPU - Model: mnasnet ms < Lower Is Better Clang 11 . 10.90 |============================================================ Clang 12 . 11.01 |============================================================= Clang 13 . 8.79 |================================================= NCNN 20210720 Target: CPU - Model: shufflenet-v2 ms < Lower Is Better Clang 11 . 11.46 |============================================================ Clang 12 . 11.62 |============================================================= Clang 13 . 7.84 |========================================= PostgreSQL pgbench 13.0 Scaling Factor: 100 - Clients: 250 - Mode: Read Only - Average Latency ms < Lower Is Better Clang 11 . 0.260 |============================================================= Clang 12 . 0.255 |============================================================ Clang 13 . 0.257 |============================================================ PostgreSQL pgbench 13.0 Scaling Factor: 100 - Clients: 250 - Mode: Read Only TPS > Higher Is Better Clang 11 . 967214 |=========================================================== Clang 12 . 981525 |============================================================ Clang 13 . 973432 |============================================================ GraphicsMagick 1.3.33 Operation: Resizing Iterations Per Minute > Higher Is Better Clang 11 . 473 |============================================================= Clang 12 . 452 |=========================================================== Clang 13 . 485 |===============================================================