Selective branch prediction reversal by correlating with data values and control flow

Modern wide-issue superscalar processors tend to adopt deeper pipelines in order to attain high clock rates. This trend increases the number of on-the-fly instructions in processors and a mispredicted branch can result in substantial amounts of wasted work. In order to mitigate these wasted works, an accurate branch prediction is required for the high performance processors. In order to improve the prediction accuracy, we propose the bimode++ branch predictor. It is an enhanced version of the bimode branch predictor. Throughout execution from the start to the end of a program, some branch instructions have the same result at all times. These branches are defined as extremely biased branches. The bimode++ branch predictor is unique in predicting the output of an extremely biased branch with a simple hardware structure. In addition, the bimode++ branch predictor improves the accuracy using the refined indexing and a fusion function. Our experimental results with benchmarks from SpecFP, SpecINT, multi-media and server area show that the bimode++ branch predictor can reduce the misprediction rate by 13.2% to the bimode and by 32.5% to the gshare.

Branch misprediction is one of the important causes of performance degradation in superpipelined and superscalar processors. Most of the existing branch predictors, based on the exploiting of branch history, suffer from prediction accuracy decrease caused by frequent context switches. The goal of this research is to reduce misprediction rate (MPR) when the context switches are frequent, and not to increase the MPR when the context switches are relatively rare. We propose two independent, but closely related modifications of global adaptive prediction mechanisms: first, to flush only the branch history register (BHR) at context switch, instead of reinitialization of the whole predictor, and second, to use two separated BHRs, one for user and one for kernel branches, instead of one global history register. We have evaluated the ideas by measurements on real traces from IBS (Instruction Benchmark Set), and have shown that both modifications reduce MPR at negligible hardware cost.

Modern embedded processors use small and simple branch predictors to improve performance. Using complex and accurate branch predictors, while desirable, is not possible as such predictors impose high power and area overhead which is not affordable in an embedded processor. As a result, for some applications, misprediction rate can be high. Such mispredictions result in energy wasted down the mispredicted path. We introduce area-aware and low-complexity pipeline gating mechanisms to reduce energy lost to possible branch mispredictions in embedded processors. We show that by using a simple gating mechanism which comes with 33-bit area overhead, on average, we can reduce the number of executed instructions by 17% (max: 30%) while paying a negligible performance cost (average 1.1%).KeywordsArea OverheadEmbed ProcessorConfidence EstimationWaste ActivityBranch InstructionThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Branch prediction reversal has been proved to be an effective alternative approach to dropping misprediction rates by means of adding a Confidence Estimator to a correlating branch predictor. This paper presents a Branch Prediction Reversal Unit (BPRU) especially oriented to enhance correlating branch predictors, such as the gshare and the Alpha 21264 metapredictor. The novelty of this proposal lies on the inclusion of data values in the confidence estimation process. Confidence metrics show that the BPRU can correctly tag 43% of branch mispredictions as low confident predictions, whereas the SBI (a previously proposed estimator) just detects 26%. Using the BPRU to reverse the gshare branch predictions leads to misprediction reductions of 15% for the SPECint2000 (up to 27% for some applications). Furthermore, the BPRU+gshare predictor reduces the misprediction rate of the SBI+gshare by an average factor of 10%. Performance evaluation of the BPRU in a superscalar processor obtains speedups of up to 9%. Similar results are obtained when the BPRU is combined with the Alpha 21264 branch predictor.KeywordsConfidence EstimatorConfidence EstimationValue PredictorBranch PredictionHistory RegisterThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

As Out-of-Order (OOO) cores scale to very large instruction windows to extract higher instruction level parallelism (ILP), the cost of mis-speculation increases tremendously. Branch predictors sitting at the head of the pipeline are a significant contributor to the speculatively executed code. It is critical to limit the execution time down the wrong path for the sake of performance and energy efficiency. Architects continue to increase the branch predictor sizes and improve the prediction algorithms to mitigate the increasing cost of the mis-speculations. Unfortunately, there still exists branches whose outcomes are predominantly data dependent, and that significantly contribute to the overall mispredictions. This work is the first to characterize data dependent branches at a scale spanning 100+ workloads drawn from several application categories and establish a clear motivation to address them. We find that branches which have only one load instruction feeding them and only simple operations to compute the branch direction from the load value are responsible for a significant fraction of the overall mispredictions. We call such branches Direct Data Dependent (3D) branches. We develop a family of synergistic techniques to avoid mispredictions from such 3D-branches. We describe 3D-Branch Overrider, our novel branch overriding technique that operates in the front-end of the pipeline to minimize the branch misprediction penalties. On a modern Icelake-like OOO core equipped with the state-of-the-art branch predictor, we find that our technique provides a 12.7% reduction in branch mispredictions resulting in 3.1% Instructions Per Cycle (IPC) gain while needing just 5.7KB in additional storage. As future cores become wider and deeper, we show that the gains from 3D-Branch Overrider nicely increases, even if the size of the underlying branch predictor is aggressively scaled.

Modern processors usually have a deep pipeline, superscalar architecture to obtain higher performance. As pipelines are getting deeper, accurate branch prediction is critical to achieve high performance, since fetched instructions after a branch have to be flushed from inside the pipeline when the prediction is wrong. This paper studies the performance of several types of branch predictors, starting from local branch predictor and global branch predictor. Simulation results show that the global history predictor outperforms the local history predictor due to the characteristic that branches tend to be correlated. However, the global history predictor still suffers an aliasing problem that degrades performance. Four techniques are proposed to alleviate the aliasing problem. The performance is evaluated by using Simplescalar with SPEC CINT95 benchmark programs. The proposed predictors display better performance over the conventional predictors after careful configuration of each.

As the pipeline length increases, the accuracy in a branch prediction gets critical to overall performance. In designing a branch predictor, in addition to accuracy, microarchitects should consider power consumption, especially in embedded processors. In this paper, we propose a low power branch predictor, which is based on the gshare predictor, by accessing the BTB (Branch Target Buffer) only when the prediction from the PHT (Prediction History Table) is taken. To enable this, the PHT is accessed one cycle earlier to prevent the additional delay. As a side effect, two predictions from the PHT are obtained at one access to the PHT, which leads to more power reduction. The proposed branch predictor reduces the power consumption, not requiring any additional storage arrays, not incurring additional delay (except just one MUX delay) and never harming accuracy. The simulation results show that the proposed predictor reduces the power consumption by 43-52%.KeywordsPower ConsumptionPower ReductionTotal Power ConsumptionAdditional DelayEmbed ProcessorThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Recently, there have been significant advances in deep neural networks (DNNs) and they have shown distinctive performance in speech recognition, natural language processing, and image recognition. In this paper, we explore DNNs to push the limit for branch prediction. We treat branch prediction as a classification problem and employ both deep convolutional neural networks (CNNs), ranging from LeNet to ResNet-50, and deep belief network (DBN) for branch prediction. We compare the effectiveness of DNNs with the state-of-the-art branch predictors, including the perceptron, our prior work, Multi-poTAGE+SC, and MTAGE+SC branch predictors. The last two are the most recent winners of championship branch prediction (CBP) contests. Several interesting observations emerged from our study. First, for branch prediction, the DNNs outperform the perceptron model as high as 60-80%. Second, we analyze the impact of the depth of CNNs (i.e., number of convolutional layers and pooling layers) on the misprediction rates. The results confirm that deeper CNN structures can lead to lower misprediction rates. Third, the DBN could outperform our prior work, but not outperform the state-of-the-art TAGE-like branch predictor; the ResNet-50 could not only outperform our prior work, but also the Multi-poTAGE+SC and MTAGE+SC.

Many aspects of speculative multithreading have been under constant and crucial research in the recent times with the increased importance in exploiting parallelism in single thread applications. One of the important architectural optimizations that is very pertinent in this scenario is branch prediction. Branch Prediction assumes increased importance for multi-threading systems that execute threads speculatively, since wrong predictions can be much costlier here, in terms of threads, than a few instructions that occupy the pipeline in a uni-processor. Conventional branch prediction techniques have provided increasingly better prediction accuracies for uni-core processing. But the branch prediction itself takes on a whole new dimension when applied to multi-core architectures based on Speculative Multithreading. Dependence on global branch history has helped branch predictors to achieve high prediction accuracy in single thread applications. The discontinuity of global history created at the thread boundaries cripple the performance of branch predictors in a multi-threaded environment. Many studies in the past have tried to address the branch history problem to improve the prediction accuracy. Most of these have been found either to be architecture specific or complex in terms of the hardware needed to recreate or approximate the right history to be given to the threads when they start executing out of order. This hardware overhead increases as the number and size of threads increase thereby limiting the scalability of the algorithms proposed so far. The current thesis takes a different direction and proposes a simple and scalable solution to effectively reduce the misprediction rates in Speculative Multithreaded systems. This is accomplished by making use of a synergistic interaction between threads to boost the inherent biased nature of branches and using less complex hardware to reduce aliasing between branches in the threads. The study proposes a new scheme called the Global Broadcast Buffer scheme to effectively reduce branch mispredictions in Speculative Multithreaded architectures.

There is wide agreement that one of the most important impediments to the performance of current and future pipelined superscalar processors is the presence of conditional branches in the instruction stream. Speculative execution seems to be one solution of choice to the branch problem, but speculative work is discarded if a branch is mispredicted. Therefore, we need a very accurate branch predictor; 95% accuracy is not good enough. This paper proposes branch classification to help improve the accuracy of branch predictors. Branch classification allows an individual branch instruction to be associated with the branch predictor best suited to predict its direction. Using this approach, a hybrid branch predictor can be constructed such that each component branch predictor predicts those branches for which it is best suited. This paper suggests one classification scheme, analyzes several branch predictors, and proposes a hybrid branch predictor that achieves higher prediction accuracy than any branch predictor previously reported in the literature.

When CPU enters the era of instruction level parallelism, the execution cycle of instructions has become shorter and shorter. Meanwhile, the processing speed of CPU has become faster and faster. However, only from the execution process of CPU instructions, there is a key factor that limits the speed of CPU processing instructions, namely, the branch instruction processing. The branch predictor is used to predict the execution result of the current branch instruction (taken/non-taken), which can make the program counter get the target instruction to continue to execute according to the prediction result before calculating the target jump address. If the prediction is correct, it will greatly improve the execution efficiency of CPU. This paper starts from the two basic categories of static branch prediction and dynamic branch prediction. Then analyse their prediction principle, prediction accuracy and component consumption used for prediction. For static branch predictor, its hardware consumption is less than that of dynamic branch predictor, and its accuracy is also lower than that of dynamic branch predictor. Moreover, their shortcomings are pointed out in this paper. At last, based on the original branch predictor, better branch predictors developed in recent years are introduced. In the future, considering the hardware consumption, I think the development of branch predictor will continue to optimize the branch predictor combining global history and local history, from the number of stored entries, the correlation between branch instructions and so on. On the premise of executing a large number of instructions, every little improvement in the efficiency of the predictor will bring huge benefits.

A value's degree of use---the number of dynamic uses of that value---provides the most essential information needed to optimize its communication. We present simulation results demonstrating the properties of degree of use of values, including their predictability: most static instructions generate values with few degrees of use and these exhibit temporal locality. We use these results to guide the design of a degree of use predictor. The development and detailed characterization of this predictor is the focus of this paper. Our predictor leverages future control flow information (e.g., branch predictions) to select among different possible degrees of use. We study the effects of several optimizations and variations in the predictor's algorithms to tune the predictor for maximum performance. The resulting predictor generates correct degree of use predictions for over 92% of all dynamic values and has a misprediction rate below 2.5%. Such a predictor has a wide range of potential applications in optimizing value communication.

A value's degree of use---the number of dynamic uses of that value---provides the most essential information needed to optimize its communication. We present simulation results demonstrating the properties of degree of use of values, including their predictability: most static instructions generate values with few degrees of use and these exhibit temporal locality. We use these results to guide the design of a degree of use predictor. The development and detailed characterization of this predictor is the focus of this paper. Our predictor leverages future control flow information (e.g., branch predictions) to select among different possible degrees of use. We study the effects of several optimizations and variations in the predictor's algorithms to tune the predictor for maximum performance. The resulting predictor generates correct degree of use predictions for over 92% of all dynamic values and has a misprediction rate below 2.5%. Such a predictor has a wide range of potential applications in optimizing value communication.

Unlike traditional superscalar processors, Simultaneous Multithreaded processors can explore both instruction level parallelism and thread level parallelism at the same time. With a same fetch width, SMT does not fetch instructions from a single thread as deeply as in traditional superscalar processors. Meanwhile, all the instructions from different threads share the same Function Units in SMT. All the characteristics make it possible to enhance the performance of SMT by reducing the branch mis-predictions. Based on the fact that about 15% of branch instructions directions can be definitely known at predicting cycle, a simple and effective bypass mechanism is proposed. This scheme doesn't depend on any existing branch predictors, and it can be used as an effective enhancement to any one of them. Execution-driven simulation results show that the branch miss prediction rates of our predictor decrease by more than 15% on average compared to a simple base line (g-share) predictor and improve the instruction throughput by about 2.5%.

Neural-inspired branch predictors achieve very low branch misprediction rates. However, previously proposed implementations have a variety of characteristics that make them challenging to implement in future high-performance processors. In particular, the the original Perceptron branch predictor suffers from a long access latency, and the faster path-based neural predictor (PBNP) requires deep pipelining and additional area to support checkpointing for misprediction recovery. The complexity of the PBNP predictor stems from the fact that the path history length, which determines the number of tables and pipeline stages, is equal to the history length, which is typically very long for high accuracy. We propose to decouple the path-history length from the outcomehistory length through a new technique called modulo-path history. By allowing a shorter path history, we can implement a PBNP with significantly fewer tables and pipeline stages while still exploiting a traditional long branch outcome history. The pipeline length reduction results in decreased power and implementation complexity. We also propose folded modulo-path history to allow the number of pipeline stages to differ from the path history length. We show that our modulo-path PBNP at 8KB can achieve prediction accuracy and overall performance within 0.8% (SPECint) of the original PBNP while simultaneously reducing predictor energy consumption by \sim 29% per access and predictor die area by \sim 35%. Our folded modulo-path history PBNP achieves performance within 1.3% of ideal, with a \sim 37% energy reduction and \sim 36% predictor area reduction.