MAD24-320 Qualcomm and Upstream Kernel BoF

• Thursday, 16 May 13:45 - 14:10
• Room: Session 2 | Tenerife I

BoF aims at providing platform to discuss state of Qualcomm SoCs and Platforms support in Mainline Linux Kernel. Focusing mainly on. -> Where we are, Current Status of Qualcomm drivers. -> Ongoing efforts on new Drivers/Frameworks. -> Pending problems to solve and plan of attack. Along with open discussion around multiple ongoing efforts around Qualcomm Linux Eco system. This will be co-presented by some of Qualcomm Landing Team Engineers.

Qualcomm Demo Booth

Qualcomm Landing Team want to show case the work they are up to with multiple demo items ranging from SystemReady , GPU to Tensorflow AI. 1. “Qualcomm compute platforms” Summary: This demo will showcase features like external display support, GPU rendering on Adreno, video decoding offloaded to Video Accelerator IP IRIS, TensorFlow MobileNet Model running on Hexagon Compute DSP, audio over Display Port running on Audio DSP and MIPI CSI camera using libcamera SoftISP, with browsing over Wi-fi, and also Bluetooth connectivity. All while monitoring system performance. Both laptops will be plugged in all the time. The Linux distro running on each of these devices will be either Debian (sid) or Ubuntu official image, both running with custom kernel for more support. The demo will emphasise on Hardware capabilities and end-to-end support for the Qualcomm based platform by Linux distros (both in kernel and userspace). 2. System Ready and Qualcomm IoT platforms (caleb) Summary: demo U-Boot booting into Linux with EFI. op6 can boot from USB with a powered hub, otherwise booting from UFS. Show GNOME firmware app with DMI/SMBIOS info? Use pmOS and/or Fedora images for SR/IR 3. TensorFlow and AI demo? https://aihub.qualcomm.com/ Summary: This demo show cases running TensorFlow Lite on RBX platforms and Qualcomm compute platforms like X13s/Hamoa. Demo will show image classification and Image identification including boxing using Moblie Net and MobileNet SSD models. 4 OpenCL and Qcom GPU. Summary: This Demo with showcase running OpenCL on Qualcomm GPU, which can be used by multiple usecase like AI. SoftISP, GPT-2.