<html><head><meta http-equiv="Content-Type" content="text/html charset=utf-8"></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space;" class=""><div class="" style="-webkit-text-stroke-width: initial; margin: 0px; font-size: 11px; -webkit-text-stroke-color: rgb(0, 0, 0);"><span style="-webkit-text-stroke-width: initial;" class="">Recently members of the JavaScript community at Intel and Mozilla have</span><span style="-webkit-text-stroke-width: initial;" class=""> </span><a href="http://www.2ality.com/2013/12/simd-js.html" class="" style="-webkit-text-stroke-width: initial;">suggested</a><span style="-webkit-text-stroke-width: initial;" class=""> </span><span style="-webkit-text-stroke-width: initial;" class="">adding SIMD types to the JavaScript language. In this email would like to share my thoughts about this proposal and to start a technical discussion about SIMD.js support in Webkit. I BCCed some of the authors of the proposal to allow them to participate in this discussion. </span></div><div class="" style="-webkit-text-stroke-width: initial; margin: 0px; font-size: 11px; -webkit-text-stroke-color: rgb(0, 0, 0);"><br class=""></div><div class="" style="-webkit-text-stroke-width: initial; margin: 0px; font-size: 11px; -webkit-text-stroke-color: rgb(0, 0, 0);"><span style="-webkit-text-stroke-width: initial;" class="">Modern processors feature </span><a href="http://en.wikipedia.org/wiki/SIMD" class="" style="-webkit-text-stroke-width: initial;">SIMD (Single Instruction Multiple Data)</a><span style="-webkit-text-stroke-width: initial;" class=""> instructions, which perform the same arithmetic operation on a vector of elements. SIMD instructions are used to accelerate compute intensive code, like image processing algorithms, because the same calculation is applied to every pixel in the image. A single SIMD instruction can process 4 or 8 pixels at the same time. Compilers try to make use of SIMD instructions in an optimization that is called vectorization. </span></div><div class="" style="-webkit-text-stroke-width: initial; margin: 0px; font-size: 11px; -webkit-text-stroke-color: rgb(0, 0, 0);"><div class=""><br class=""></div></div><div class="" style="-webkit-text-stroke-width: initial; margin: 0px; font-size: 11px; -webkit-text-stroke-color: rgb(0, 0, 0);">The <a href="http://wiki.ecmascript.org/doku.php?id=strawman:simd_number" class="">SIMD.js API</a> adds new types, such as float32x4, and operators that map to vector instructions on most processors. The idea behind the proposal is that manual use of vector instructions, just like intrinsics in C, will allow developers to accelerate common compute-intensive JavaScript applications. The idea of using SIMD instructions to accelerate JavaScript code is compelling because high performance applications in JavaScript are becoming very popular. </div><div class="" style="-webkit-text-stroke-width: initial; margin: 0px; font-size: 11px; -webkit-text-stroke-color: rgb(0, 0, 0); min-height: 13px;"><br class=""></div><div class="" style="-webkit-text-stroke-width: initial; margin: 0px; font-size: 11px; -webkit-text-stroke-color: rgb(0, 0, 0);">Before I became involved with JavaScript through my work on the <a href="https://www.webkit.org/blog/3362/introducing-the-webkit-ftl-jit/" class="">FTL project</a>, I developed the <a href="http://llvm.org/docs/Vectorizers.html" class="">LLVM vectorizer</a> and worked on a vectorizing compiler for a data-parallel programming language. Based on my experience with vectorization, I believe that the current proposal to include SIMD types in the JavaScript language is not the right approach to utilize SIMD instructions. In this email I argue that vector types should not be added to the JavaScript language.</div><div class="" style="-webkit-text-stroke-width: initial; margin: 0px; font-size: 11px; -webkit-text-stroke-color: rgb(0, 0, 0);"><br class=""></div><div class="" style="-webkit-text-stroke-width: initial; margin: 0px; font-size: 11px; -webkit-text-stroke-color: rgb(0, 0, 0);">Vector instruction sets are sparse, asymmetrical, and vary in size and features from one generation to another. For example, some Intel processors feature <a href="https://software.intel.com/en-us/blogs/2013/avx-512-instructions" class="">512-bit wide vector instructions</a>. This means that they can process 16 floating point numbers with one instruction. However, today’s high-end ARM processors feature <a href="http://www.arm.com/products/processors/technologies/neon.php" class="">128-bit wide vector instructions</a> and can only process 4 floating point elements. ARM processors support byte-sized blend instructions but only recent Intel processors added support for byte-sized blends. ARM processors support variable shifts but only Intel processors with AVX2 support variable shifts. Different generations of Intel processors support different instruction sets with different features such as broadcasting from a local register, 16-bit and 64-bit arithmetic, and varied shuffles. Modern processors even feature predicated arithmetic and scatter/gather instructions that are very difficult to model using target independent high-level intrinsics. </div><div class="" style="-webkit-text-stroke-width: initial; margin: 0px; font-size: 11px; -webkit-text-stroke-color: rgb(0, 0, 0);"><span style="-webkit-text-stroke-width: initial;" class="">The designers of the high-level target independent API should decide if they want to support the union of all vector instruction sets, or the intersection. A subset of the vector instructions that represent the intersection of all popular instruction sets is not useable for writing non-trivial vector programs. And the superset of the vector instructions will cause huge performance regressions on platforms that do not support the used instructions.</span></div><div class="" style="-webkit-text-stroke-width: initial; margin: 0px; font-size: 11px; -webkit-text-stroke-color: rgb(0, 0, 0); min-height: 13px;"><br class=""></div><div class="" style="-webkit-text-stroke-width: initial; margin: 0px; font-size: 11px; -webkit-text-stroke-color: rgb(0, 0, 0);">Code that uses SIMD.js is not performance-portable. Modern vectorizing compilers feature complex cost models and heuristics for deciding when to vectorize, at which vector width, and how many loop iterations to interleave. The cost models takes into account the features of the vector instruction set, properties of the architecture such as the number of vector registers, and properties of the current processor generation. Making a poor selection decision on any of the vectorization parameters can result in a major performance regression. Executing vector intrinsics on processors that don’t support them is slower than executing multiple scalar instructions because the compiler can’t always generate efficient with the same semantics.</div><div class="" style="-webkit-text-stroke-width: initial; margin: 0px; font-size: 11px; -webkit-text-stroke-color: rgb(0, 0, 0);">I don’t believe that it is possible to write non-trivial vector code that will show performance gains on processors from different families. Executing vector code with insufficient hardware support will cause major performance regressions. <span class="" style="-webkit-text-stroke-width: initial;">One of the motivations for SIMD.js was to allow </span><a href="https://developer.mozilla.org/en-US/docs/Mozilla/Projects/Emscripten" class="" style="-webkit-text-stroke-width: initial;">Emscripten</a><span class="" style="-webkit-text-stroke-width: initial;"> to vectorize C code and to emit JavaScript SIMD intrinsics. One problem with this suggestion is that the Emscripten compiler should not be assuming that the target is an x86 machine and that a specific vector width and interleave width is the right answer. Targeting a specific processor will surely cause regressions on other processors. </span></div><div class="" style="-webkit-text-stroke-width: initial; margin: 0px; font-size: 11px; -webkit-text-stroke-color: rgb(0, 0, 0); min-height: 13px;"><br class=""></div><div class="" style="-webkit-text-stroke-width: initial; margin: 0px; font-size: 11px; -webkit-text-stroke-color: rgb(0, 0, 0);">SIMD.js does not make good use of modern vector instruction sets. Modern vector processors feature large vectors (up to 512-bit), predication of arithmetic and memory operations, scatter/gather memory operations, advance shuffles and broadcasts and other features that make vectorization efficient. The current SIMD.js proposal is limited to a small number of arithmetic operations on 128-bit vector data types.</div><div class="" style="-webkit-text-stroke-width: initial; margin: 0px; font-size: 11px; -webkit-text-stroke-color: rgb(0, 0, 0); min-height: 13px;"><br class=""></div><div class="" style="-webkit-text-stroke-width: initial; margin: 0px; font-size: 11px; -webkit-text-stroke-color: rgb(0, 0, 0);">So far, I’ve explained why I believe SIMD.js will not be performance-portable and why it will not utilize modern instruction sets, but I have not made a suggestion on how to use vector instructions to accelerate JavaScript programs. Vectorization, like instruction scheduling and register allocation, is a code-generation problem. In order to solve these problems, it is necessary for the compiler to have intimate knowledge of the architecture. Forcing the compiler to use a specific instruction or a specific data-type is the wrong answer. We can learn a lesson from the design of compilers for data-parallel languages. GPU programs (shaders and compute languages, such as OpenCL and GLSL) are written using vector instructions because the domain of the problem requires vectors (colors and coordinates). One of the first thing that data-parallel compilers do is to break vector instructions into scalars (this process is called scalarization). After getting rid of the vectors that resulted from the problem domain, the compiler may begin to analyze the program, calculate profitability, and make use of the available instruction set. </div><div class="" style="-webkit-text-stroke-width: initial; margin: 0px; font-size: 11px; -webkit-text-stroke-color: rgb(0, 0, 0); min-height: 13px;"><br class=""></div><div class="" style="-webkit-text-stroke-width: initial; margin: 0px; font-size: 11px; -webkit-text-stroke-color: rgb(0, 0, 0);">I believe that it is the responsibility of JIT compilers to use vector instructions. In the implementation of the Webkit’s FTL JIT compiler, we took one step in the direction of using vector instructions. LLVM already vectorizes some code sequences during instruction selection, and we started investigating the use of LLVM’s Loop and SLP vectorizers. We found that despite nice performance gains on a number of workloads, we experienced some performance regressions on Intel’s Sandybridge processors, which is currently a very popular desktop processor. JavaScript code contains many branches (due to dynamic speculation). Unfortunately, branches on Sandybridge execute on Port5, which is also where many vector instructions are executed. So, pressure on Port5 prevented performance gains. The LLVM vectorizer currently does not model execution port pressure and we had to disable vectorization in FTL. In the future, we intend to enable more vectorization features in FTL.</div><div class="" style="-webkit-text-stroke-width: initial; margin: 0px; font-size: 11px; -webkit-text-stroke-color: rgb(0, 0, 0); min-height: 13px;"><br class=""></div><div class="" style="-webkit-text-stroke-width: initial; margin: 0px; font-size: 11px; -webkit-text-stroke-color: rgb(0, 0, 0);">To summarize, SIMD.js will not provide a portable performance solution because vector instruction sets are sparse and vary between architectures and generations. Emscripten should not generate vector instructions because it can’t model the target machine. SIMD.js will not make use of modern SIMD features such as predication or scatter/gather. Vectorization is a compiler code generation problem that should be solved by JIT compilers, and not by the language itself. JIT compilers should continue to evolve and to start vectorizing code like modern compilers.</div><div class="" style="-webkit-text-stroke-width: initial; margin: 0px; font-size: 11px; -webkit-text-stroke-color: rgb(0, 0, 0);"><br class=""></div><div class="" style="-webkit-text-stroke-width: initial; margin: 0px; font-size: 11px; -webkit-text-stroke-color: rgb(0, 0, 0);">Thanks,</div><div class="" style="-webkit-text-stroke-width: initial; margin: 0px; font-size: 11px; -webkit-text-stroke-color: rgb(0, 0, 0);">Nadav</div><div class="" style="-webkit-text-stroke-width: initial; margin: 0px; font-size: 11px; -webkit-text-stroke-color: rgb(0, 0, 0);"><br class=""></div></body></html>