avx512 onednn 3.0 ryzen 9 7950x AMD Ryzen 9 7950X 16-Core testing with a ASUS ROG CROSSHAIR X670E HERO (0805 BIOS) and AMD Radeon RX 7900 XTX 24GB on Ubuntu 22.04 via the Phoronix Test Suite. a: Processor: AMD Ryzen 9 7950X 16-Core @ 5.88GHz (16 Cores / 32 Threads), Motherboard: ASUS ROG CROSSHAIR X670E HERO (0805 BIOS), Chipset: AMD Device 14d8, Memory: 32GB, Disk: Western Digital WD_BLACK SN850X 1000GB + 2000GB, Graphics: AMD Radeon RX 7900 XTX 24GB (3220/1249MHz), Audio: AMD Device ab30, Monitor: ASUS MG28U, Network: Intel I225-V + Intel Wi-Fi 6 AX210/AX211/AX411 OS: Ubuntu 22.04, Kernel: 5.15.0-56-generic (x86_64), Desktop: GNOME Shell 42.5, Display Server: X Server 1.21.1.3 + Wayland, OpenGL: 4.6 Mesa 22.3.0-devel (LLVM 15.0.3 DRM 3.49), OpenCL: OpenCL 2.1 AMD-APP (3513.0), Compiler: GCC 11.3.0, File-System: ext4, Screen Resolution: 3840x2160 b: Processor: AMD Ryzen 9 7950X 16-Core @ 5.88GHz (16 Cores / 32 Threads), Motherboard: ASUS ROG CROSSHAIR X670E HERO (0805 BIOS), Chipset: AMD Device 14d8, Memory: 32GB, Disk: Western Digital WD_BLACK SN850X 1000GB + 2000GB, Graphics: AMD Radeon RX 7900 XTX 24GB (3220/1249MHz), Audio: AMD Device ab30, Monitor: ASUS MG28U, Network: Intel I225-V + Intel Wi-Fi 6 AX210/AX211/AX411 OS: Ubuntu 22.04, Kernel: 5.15.0-56-generic (x86_64), Desktop: GNOME Shell 42.5, Display Server: X Server 1.21.1.3 + Wayland, OpenGL: 4.6 Mesa 22.3.0-devel (LLVM 15.0.3 DRM 3.49), OpenCL: OpenCL 2.1 AMD-APP (3513.0), Compiler: GCC 11.3.0, File-System: ext4, Screen Resolution: 3840x2160 cc: Processor: AMD Ryzen 9 7950X 16-Core @ 5.88GHz (16 Cores / 32 Threads), Motherboard: ASUS ROG CROSSHAIR X670E HERO (0805 BIOS), Chipset: AMD Device 14d8, Memory: 32GB, Disk: Western Digital WD_BLACK SN850X 1000GB + 2000GB, Graphics: AMD Radeon RX 7900 XTX 24GB (3220/1249MHz), Audio: AMD Device ab30, Monitor: ASUS MG28U, Network: Intel I225-V + Intel Wi-Fi 6 AX210/AX211/AX411 OS: Ubuntu 22.04, Kernel: 5.15.0-56-generic (x86_64), Desktop: GNOME Shell 42.5, Display Server: X Server 1.21.1.3 + Wayland, OpenGL: 4.6 Mesa 22.3.0-devel (LLVM 15.0.3 DRM 3.49), OpenCL: OpenCL 2.1 AMD-APP (3513.0), Compiler: GCC 11.3.0, File-System: ext4, Screen Resolution: 3840x2160 d: Processor: AMD Ryzen 9 7950X 16-Core @ 5.88GHz (16 Cores / 32 Threads), Motherboard: ASUS ROG CROSSHAIR X670E HERO (0805 BIOS), Chipset: AMD Device 14d8, Memory: 32GB, Disk: Western Digital WD_BLACK SN850X 1000GB + 2000GB, Graphics: AMD Radeon RX 7900 XTX 24GB (3220/1249MHz), Audio: AMD Device ab30, Monitor: ASUS MG28U, Network: Intel I225-V + Intel Wi-Fi 6 AX210/AX211/AX411 OS: Ubuntu 22.04, Kernel: 5.15.0-56-generic (x86_64), Desktop: GNOME Shell 42.5, Display Server: X Server 1.21.1.3 + Wayland, OpenGL: 4.6 Mesa 22.3.0-devel (LLVM 15.0.3 DRM 3.49), OpenCL: OpenCL 2.1 AMD-APP (3513.0), Compiler: GCC 11.3.0, File-System: ext4, Screen Resolution: 3840x2160 oneDNN 3.0 Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU ms < Lower Is Better a .. 1135.35 |================================================================= b .. 1135.16 |================================================================= cc . 1132.81 |================================================================= d .. 1134.70 |================================================================= oneDNN 3.0 Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better a .. 1136.91 |================================================================= b .. 1133.90 |================================================================= cc . 1133.54 |================================================================= d .. 1134.22 |================================================================= oneDNN 3.0 Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU ms < Lower Is Better a .. 1135.50 |================================================================= b .. 1129.74 |================================================================= cc . 1135.82 |================================================================= d .. 1134.55 |================================================================= oneDNN 3.0 Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU ms < Lower Is Better a .. 582.38 |================================================================== b .. 576.96 |================================================================= cc . 569.85 |================================================================ d .. 583.77 |================================================================== oneDNN 3.0 Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU ms < Lower Is Better a .. 581.00 |================================================================== b .. 583.07 |================================================================== cc . 582.25 |================================================================== d .. 581.67 |================================================================== oneDNN 3.0 Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better a .. 580.58 |================================================================== b .. 582.96 |================================================================== cc . 579.43 |================================================================== d .. 583.75 |================================================================== oneDNN 3.0 Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU ms < Lower Is Better a .. 4.36135 |=========================================================== b .. 3.76311 |=================================================== cc . 2.56838 |=================================== d .. 4.77297 |================================================================= oneDNN 3.0 Harness: IP Shapes 1D - Data Type: bf16bf16bf16 - Engine: CPU ms < Lower Is Better a .. 0.702572 |=========================================================== b .. 0.768543 |================================================================ cc . 0.631368 |===================================================== d .. 0.718084 |============================================================ oneDNN 3.0 Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better a .. 0.458029 |================================================================ b .. 0.357162 |================================================== cc . 0.357673 |================================================== d .. 0.368789 |==================================================== oneDNN 3.0 Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better a .. 0.342068 |========================================================== b .. 0.329774 |======================================================== cc . 0.368081 |============================================================== d .. 0.376915 |================================================================ oneDNN 3.0 Harness: Deconvolution Batch shapes_1d - Data Type: bf16bf16bf16 - Engine: CPU ms < Lower Is Better a .. 3.34345 |================================================================ b .. 3.37554 |================================================================= cc . 3.36730 |================================================================= d .. 3.36723 |================================================================= oneDNN 3.0 Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better a .. 0.421798 |================================================================ b .. 0.420827 |================================================================ cc . 0.420617 |================================================================ d .. 0.420822 |================================================================ oneDNN 3.0 Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU ms < Lower Is Better a .. 1.71973 |================================================================ b .. 1.71247 |================================================================ cc . 1.72827 |================================================================ d .. 1.74523 |================================================================= oneDNN 3.0 Harness: Matrix Multiply Batch Shapes Transformer - Data Type: bf16bf16bf16 - Engine: CPU ms < Lower Is Better a .. 0.209725 |=============================================================== b .. 0.212361 |================================================================ cc . 0.211933 |================================================================ d .. 0.210342 |=============================================================== oneDNN 3.0 Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU ms < Lower Is Better a .. 0.436887 |================================================================ b .. 0.439701 |================================================================ cc . 0.438763 |================================================================ d .. 0.439201 |================================================================ oneDNN 3.0 Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better a .. 0.129773 |================================================================ b .. 0.130403 |================================================================ cc . 0.129929 |================================================================ d .. 0.129658 |================================================================ oneDNN 3.0 Harness: IP Shapes 3D - Data Type: bf16bf16bf16 - Engine: CPU ms < Lower Is Better a .. 1.54703 |================================================================ b .. 1.57145 |================================================================= cc . 1.57079 |================================================================= d .. 1.54588 |================================================================ oneDNN 3.0 Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU ms < Lower Is Better a .. 3.24945 |================================================================= b .. 3.23468 |================================================================ cc . 3.27078 |================================================================= d .. 3.23747 |================================================================ oneDNN 3.0 Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU ms < Lower Is Better a .. 5.60386 |================================================================= b .. 5.60673 |================================================================= cc . 5.61189 |================================================================= d .. 5.60781 |================================================================= oneDNN 3.0 Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better a .. 5.20379 |================================================================ b .. 5.19629 |================================================================ cc . 5.22347 |================================================================ d .. 5.27147 |================================================================= oneDNN 3.0 Harness: Convolution Batch Shapes Auto - Data Type: bf16bf16bf16 - Engine: CPU ms < Lower Is Better a .. 1.69724 |================================================================= b .. 1.69655 |================================================================= cc . 1.69867 |================================================================= d .. 1.69653 |================================================================= oneDNN 3.0 Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU ms < Lower Is Better a .. 2.36413 |============================================================ b .. 2.55866 |================================================================= cc . 2.35925 |============================================================ d .. 2.35522 |============================================================ oneDNN 3.0 Harness: Deconvolution Batch shapes_3d - Data Type: bf16bf16bf16 - Engine: CPU ms < Lower Is Better a .. 1.41319 |================================================================= b .. 1.40158 |================================================================ cc . 1.40127 |================================================================ d .. 1.39831 |================================================================ oneDNN 3.0 Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better a .. 0.587832 |================================================================ b .. 0.586989 |================================================================ cc . 0.584807 |================================================================ d .. 0.588675 |================================================================