泰晓资讯·11月 / 第三期 / 2015

LWN: 内核峰会 2015 成功在首尔举办

泰晓科技对这次峰会有详细介绍，见：2015 内核峰会简报

Khronos: 发布 OpenCL 2.1 与 SPIR-V 1.0 规范

SPIR-V 1.0 is an intermediate language fully defined by Khronos with unique, native support for graphics shaders and computational kernels.

The OpenCL C++ kernel language released in the OpenCL 2.1 provisional specification is being finalized and will be released imminently, also using SPIR-V for run-time execution.

AMD 耕耘多年的异构计算体系，渐渐成形。

对了，坐等 Vulkan 规范发布。

Phoronix: AMD 计划发布 HCC（Heterogeneous Compute Compiler)

AMD has been open-sourcing several components of their Linux HSA (Heterogeneous System Architecture) stack for the past several months including the AMDKFD kernel driver and HSAKMT run-time.

…, they’re reportedly planning to publish a Heterogeneous Compute Compiler (HCC) that utilizes LLVM.

AMD 在 LLVM 开发者会议 2015 上透露了该计划。

从面向应用开发者的 OpenCL，到中间的实现支撑库 HSAKMT，再到底层驱动 AMDKFD，AMD 平台异构计算的软件栈正在形成。

独有无偶，Imagination 也已公布异构计算的初步路线图，详见此(PDF)

Phoronix: Google 的开源 CUDA 编译器 GPUCC，碉堡了

GPUCC is their name for an open-source GPGPU compiler built atop LLVM.

They call it “the first fully-functional, open-source high performance CUDA compiler” that is up to 51% faster on internal end-to-end benchmarks, on par with open-source benchmarks, compile time is 8% faster on average and 2.4x faster for pathological compilations compared to NVIDIA’s official CUDA compiler (NVCC).

CUDA 是来自 NVIDIA、出道早于 OpenCL 的异构计算框架，有较广泛应用。

Google 在 LLVM 开发者会议 2015 介绍了 GPUCC（PDF）

顺便说一句，LLVM 05 的其他幻灯在此公布，视频发布在你懂的某个不存在的网站。

Intel GFX: 发布 XenGT，一个图形仲裁透传（Mediated Graphics Passthrough）方案

A virtual GPU instance is maintained for each VM, with part of performance critical resources directly assigned.

The capability of running native graphics driver inside a VM, without hypervisor intervention in performance critical paths, achieves a good balance among performance, feature, and sharing capability.

Xen is currently supported on Intel Processor Graphics (a.k.a. XenGT); and the core logic can be easily ported to other hypervisors.

LWN：无 UI 也是一种 UI

Aubé’s premise is that “invisible” applications—those that use text-messaging or voice-recognition rather than on-screen interfaces—are the future of UI design.

Langridge, however, contends that “until very recently, and honestly pretty much still, a computer can’t understand the nuance of language. So ‘use language to control computers’ meant ‘learn the computer’s language’, not ‘the computer learns yours’.”

咳咳，大哥你把 “UI” 混同 “GUI” 也太不专业了吧(☆_☆)

回到话题，GUI 常以一组又一组图形控件的颜面，来给用户做选择题。所需呈现的事物越复杂，则控件数量及关联越多，由此干扰交互，至而搞晕用户。这在手机为代表的、屏幕尺寸有限的移动设备上更易显现。

移动场景下使用设备，需干净利落地人机交互。这便要借助对语境的识别，即“机器智能”，而不是徒徒增加屏幕尺寸。

我们在 移动终端 UI 设计新思路：从图形界面回归“命令行”? 、大开脑洞，做一个比表屏大，比 phone 便携的移动设备 也表达了相似的想法。