RSS Reader

Today Logitech announced a long-term partnership with Baidu Brain, beginning with the launch of its intuitive new Logitech Voice M380 Wireless Mouse with Speech Input in China. Designed especially for people who create large amounts of content, this innovative product lets you dictate with your voice, creating content two or three times faster than typing. The Logitech Voice M380 Wireless Mouse is powered exclusively by intelligent Baidu Speech technology* from Baidu Brain and features the comfort, performance and quality that users expect in a Logitech

Open Source LLM

ROCm AOMP 13.0 Released For OpenMP Offloading To Radeon GPUs

Phoronix | April 13, 2021, 6:41 a.m. | Original

AOMP 13.0 is out today as the first update to AMD's Radeon OpenMP offloading compiler since AOMP 11.0 last December.

AOMP is part of the Radeon Open eCosystem (ROCm) for offloading OpenMP work to AMD GPUs. AOMP is a downstream of the LLVM Clang compiler stack. AMD engineers have been working to upstream their OpenMP offloading patches albeit is a work-in-progress and they churn out new patches faster than what it takes to get reviewed and upstreamed. AOMP is not to be

Open Source LLM

HPVM 1.0 Released As LLVM-Based Compiler For CPUs / GPUs / FPGAs / Accelerators

Phoronix | April 9, 2021, 5:57 a.m. | Original

The latest open-source compiler infrastructure effort seeking to target a wide spectrum of devices from CPUs through GPUs, FPGAs, and accelerators is HPVM. The HPVM project today celebrated its 1.0 milestone.

Like most compiler projects these days, HPVM is based on the LLVM compiler stack. HPVM was also born at the University of Illinois where LLVM itself was first started. We covered the initial work on HPVM more than a year ago in University of Illinois Releases HPVM As Heterogeneous Parallel Systems Compiler .

HPVM uses a target-

Open Source LLM

Google Posts Initial Code For Lyra Speech Codec

Phoronix | April 6, 2021, 8:16 p.m. | Original

Back in February we covered Google's work on the Lyra voice/audio codec designed for fitting with very low bit-rate audio for speech compression in use-cases like WebRTC and video chatting even on the most limited Internet connections. Thanks to leveraging machine learning, Lyra can function at just 3kbps. The code to Lyra is now public.

Lyra when originally announced by Google back in February sounded quite promising with their effort to allow for video chats over 56kbps Internet connections when Lyra is paired with the

Open Source LLM

LLVM 12.0-rc4 Released For Squaring Up This Open-Source Compiler

Phoronix | April 1, 2021, 5:23 a.m. | Original

LLVM 12.0 was supposed to be out around the end of February but blocker bugs have resulted in additional release candidates as the developers work to button up this open-source compiler stack release.

LLVM release manager Tom Stellard overnight issued LLVM 12.0-rc4 as the latest unplanned release candidate for this compiler update. More fixes have landed and ideally this will be the last release candidate and what the official release should look like, assuming no more pressing issues turn up.

LLVM 12.0

Open Source LLM

NVIDIA's Open-Source DALI Reaches Version 1.0

Phoronix | March 24, 2021, 6:30 p.m. | Original

Announced nearly three years ago by NVIDIA as one of their open-source projects was the DALI library for GPU-accelerated data augmentation and image loading . The DALI library today reached the v1.0.0 milestone.

NVIDIA DALI is summed up as a data loading library with a focus on data loading and pre-processing for deep learning software. DALI provides various building blocks particularly around image, video, and audio processing. Of course, the GPU-accelerated library is optimized for NVIDIA's software/hardware

Open Source LLM

AMD Sends Out Latest Patches For HMM-Based SVM Memory Manager

Phoronix | March 23, 2021, 9:11 a.m. | Original

Published back in January was the initial work on a HMM-based SVM memory manager for the open-source Radeon compute stack. A second version of that work is now available as it continues working towards the mainline kernel.

This Shared Virtual Memory (SVM) memory manager implementation makes use of the Linux kernel's modern Heterogeneous Memory Management (HMM) code. As part of working on continuing to advance their ROCm compute stack and preparing for their big HPC/super-computer deployments using a combination of AMD EPYC CPUs

Open Source LLM

OpenBLAS 0.3.14 Released With Performance Improvements For AMD Ryzen, POWER10

Phoronix | March 18, 2021, midnight | Original

OpenBLAS 0.3.14 is out today as the newest version of this open-source BLAS (Basic Linear Algebra Subprograms) library that continues to work on maximizing the performance for x86_64 and other architectures.

OpenBLAS 0.3.14 on the x86_64 has an optimized BFloat16 GEMM kernel for Intel Cooper Lake processors, auto-detection is added for Rocket Lake and Tiger Lake, and AMD Ryzen processors are enjoying improved performance for SASUM / DASUM /

Open Source LLM

AMD AOCC 3.0 Released As Zen 3 Optimized LLVM Clang 12 Based Compiler

Phoronix | March 15, 2021, 11:46 a.m. | Original

With today's AMD EPYC 7003 "Milan" launch there is also the public availability of AOCC 3.0 as their LLVM/Clang downstream now carrying patches for optimized Zen 3 support.

AOCC 3.0 is AMD's downstream of LLVM/Clang focused on offering leading AMD CPU support. With AOCC 3.0 they have re-based against the in-development LLVM 12 trunk code while have enablined additional tuning/optimizations around the EPYC 7003 series. AOCC 3.

Open Source LLM

LLVM 12.0 Should Be Released Soon Following RC3 Release

Phoronix | March 11, 2021, 12:05 a.m. | Original

LLVM 12.0 was aiming for release at the start of March but it missed that goal post and a third release candidate was tagged on Wednesday. But, fortunately, it looks like the official release will be out soon.

Tom Stellard as the LLVM release manager announced LLVM 12.0-RC3 on Wednesday and indicated that it should be the last release candidate barring any new major issues from becoming known. This will be the first major LLVM release of 2021 and keeping up with their

(PR) Logitech and Baidu Brain Partner to Transform the Way We Work Using AI and Voice

ROCm AOMP 13.0 Released For OpenMP Offloading To Radeon GPUs

HPVM 1.0 Released As LLVM-Based Compiler For CPUs / GPUs / FPGAs / Accelerators

Google Posts Initial Code For Lyra Speech Codec

LLVM 12.0-rc4 Released For Squaring Up This Open-Source Compiler

NVIDIA's Open-Source DALI Reaches Version 1.0

AMD Sends Out Latest Patches For HMM-Based SVM Memory Manager

OpenBLAS 0.3.14 Released With Performance Improvements For AMD Ryzen, POWER10

AMD AOCC 3.0 Released As Zen 3 Optimized LLVM Clang 12 Based Compiler

LLVM 12.0 Should Be Released Soon Following RC3 Release

Menu