onednn 3.0 whiskey Intel Core i7-8565U testing with a Dell 0KTW76 (1.17.0 BIOS) and Intel UHD 620 WHL GT2 15GB on Ubuntu 22.04 via the Phoronix Test Suite. a: Processor: Intel Core i7-8565U @ 4.60GHz (4 Cores / 8 Threads), Motherboard: Dell 0KTW76 (1.17.0 BIOS), Chipset: Intel Cannon Point-LP, Memory: 16GB, Disk: SK hynix PC401 NVMe 256GB, Graphics: Intel UHD 620 WHL GT2 15GB (1150MHz), Audio: Realtek ALC3271, Network: Qualcomm Atheros QCA6174 802.11ac OS: Ubuntu 22.04, Kernel: 5.19.0-rc6-phx-retbleed (x86_64), Desktop: GNOME Shell 42.2, Display Server: X Server + Wayland, OpenGL: 4.6 Mesa 22.0.1, OpenCL: OpenCL 3.0, Vulkan: 1.3.204, Compiler: GCC 11.3.0, File-System: ext4, Screen Resolution: 1920x1080 b: Processor: Intel Core i7-8565U @ 4.60GHz (4 Cores / 8 Threads), Motherboard: Dell 0KTW76 (1.17.0 BIOS), Chipset: Intel Cannon Point-LP, Memory: 16GB, Disk: SK hynix PC401 NVMe 256GB, Graphics: Intel UHD 620 WHL GT2 15GB (1150MHz), Audio: Realtek ALC3271, Network: Qualcomm Atheros QCA6174 802.11ac OS: Ubuntu 22.04, Kernel: 5.19.0-rc6-phx-retbleed (x86_64), Desktop: GNOME Shell 42.2, Display Server: X Server + Wayland, OpenGL: 4.6 Mesa 22.0.1, OpenCL: OpenCL 3.0, Vulkan: 1.3.204, Compiler: GCC 11.3.0, File-System: ext4, Screen Resolution: 1920x1080 c: Processor: Intel Core i7-8565U @ 4.60GHz (4 Cores / 8 Threads), Motherboard: Dell 0KTW76 (1.17.0 BIOS), Chipset: Intel Cannon Point-LP, Memory: 16GB, Disk: SK hynix PC401 NVMe 256GB, Graphics: Intel UHD 620 WHL GT2 15GB (1150MHz), Audio: Realtek ALC3271, Network: Qualcomm Atheros QCA6174 802.11ac OS: Ubuntu 22.04, Kernel: 5.19.0-rc6-phx-retbleed (x86_64), Desktop: GNOME Shell 42.2, Display Server: X Server + Wayland, OpenGL: 4.6 Mesa 22.0.1, OpenCL: OpenCL 3.0, Vulkan: 1.3.204, Compiler: GCC 11.3.0, File-System: ext4, Screen Resolution: 1920x1080 oneDNN 3.0 Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better a . 5293.77 |========================================================= b . 6173.65 |================================================================== c . 5926.93 |=============================================================== oneDNN 3.0 Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU ms < Lower Is Better a . 5544.45 |=========================================================== b . 6204.84 |================================================================== c . 5574.51 |=========================================================== oneDNN 3.0 Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU ms < Lower Is Better a . 5748.76 |============================================================== b . 6094.99 |================================================================== c . 5500.81 |============================================================ oneDNN 3.0 Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU ms < Lower Is Better a . 12062.30 |================================================================= b . 11451.00 |============================================================== c . 9754.05 |===================================================== oneDNN 3.0 Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better a . 10826.9 |============================================================== b . 11511.0 |================================================================== c . 10646.9 |============================================================= oneDNN 3.0 Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU ms < Lower Is Better a . 11022.0 |=============================================================== b . 11620.3 |================================================================== c . 10876.5 |============================================================== oneDNN 3.0 Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU ms < Lower Is Better a . 15.06 |======================================================== b . 18.19 |==================================================================== c . 16.59 |============================================================== oneDNN 3.0 Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU ms < Lower Is Better a . 8.78070 |=========================================================== b . 9.88295 |================================================================== c . 8.27289 |======================================================= oneDNN 3.0 Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better a . 3.84594 |========================================================= b . 4.41605 |================================================================== c . 3.68125 |======================================================= oneDNN 3.0 Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better a . 6.47655 |================================================================== b . 6.34864 |================================================================= c . 6.27607 |================================================================ oneDNN 3.0 Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU ms < Lower Is Better a . 5.85417 |================================================================= b . 5.96239 |================================================================== c . 5.70501 |=============================================================== oneDNN 3.0 Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better a . 3.27762 |================================================================= b . 3.30977 |================================================================== c . 3.26082 |================================================================= oneDNN 3.0 Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU ms < Lower Is Better a . 17.64 |==================================================================== b . 17.70 |==================================================================== c . 15.46 |=========================================================== oneDNN 3.0 Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better a . 4.13532 |================================================================= b . 4.17287 |================================================================== c . 4.12731 |================================================================= oneDNN 3.0 Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better a . 21.21 |================================================================== b . 21.71 |==================================================================== c . 21.21 |================================================================== oneDNN 3.0 Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU ms < Lower Is Better a . 22.29 |================================================================= b . 23.32 |==================================================================== c . 22.34 |================================================================= oneDNN 3.0 Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU ms < Lower Is Better a . 16.53 |=================================================================== b . 16.77 |==================================================================== c . 16.56 |=================================================================== oneDNN 3.0 Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU ms < Lower Is Better a . 10.45330 |================================================================= b . 9.99794 |============================================================== c . 10.03140 |==============================================================