Store Bound
Metric DescriptionThis metric shows how often CPU was stalled on store operations. Consider False Sharing analysis as your next step.Parent topic: CPU MetricsSee AlsoReference for Performance...
View ArticleSplit Stores
Metric DescriptionThis metric represents a rate of split store accesses. Consider aligning your data to the 64-byte cache line granularity.Possible IssuesA significant proportion of cycles is spent...
View ArticleSplit Loads
Metric DescriptionThroughout the memory hierarchy, data moves at cache-line granularity (64 bytes per line). Although this is much larger than many common data types (such as integer, float, or...
View ArticleSpin Time
Metric DescriptionSpin time is Wait Time during which the CPU is busy. This often occurs when a synchronization API causes the CPU to poll while the software thread is waiting. Some Spin Time may be...
View ArticleSMC Machine Clear
Metric DescriptionCertain events require the entire pipeline to be cleared and restarted from just after the last retired instruction. This metric measures only self-modifying code (SMC) event. This...
View ArticleClik here to view.
Overhead and Spin Time
Overhead TimeOverhead time is the time the system takes to deliver a shared resource from a releasing owner to an acquiring owner. Ideally, the Overhead time should be close to zero because it means...
View ArticleOverhead Time
Metric DescriptionOverhead time is CPU time spent on the overhead of known synchronization and threading libraries, such as system synchronization APIs, Intel TBB, and OpenMP.Possible IssuesA...
View ArticleIntel® Fortran Vectorization Diagnostics
Intel® Fortran Compiler Vectorization Diagnostics We have a similar catalog of vectorization diagnostics for the Intel® C++ Compiler HERE The following diagnostic messages from the vectorization report...
View ArticleDiagnostic 15011: loop was not vectorized: statement cannot be vectorized...
Cause:The loop contains a statement that cannot be mapped to SIMD instructions by the vectorizer. In the example below, the loop contains an assignment to a derived data type which is not directly...
View ArticleClik here to view.
Inspector XE 2013 Update 9 线程检查时带来的整体性能提升
Intel(R) Inspector XE 2013 帮助用户进行线程检查，包含：线程间的共享数据冲突和死锁的检测。当选择”Locate Deadlocks and Data Races“选项时，工具额外开销加程序本身达到原来的40-160倍。在新版的Inspector...
View Article使用英特尔® Composer XE为至强融核™ 协处理器进行Offload 编译
Offload编译指的是在一个可运行的主机代码中加入编译指示或者某些新的关键字使代码段运行在基于英特尔集成众核架构(英特尔MIC架构)的协处理器上。其编程方式类似于使用OpenMP* 指示或英特尔 Cilk™ Plus关键字在串行代码中加入并行。...
View ArticleClik here to view.
Doctor Fortran in "It's a Modern Fortran World"
I recently received a copy of "Numerical Computing with Modern Fortran", by Richard Hanson and Tim Hopkins, and noted how many books on Fortran are being published recently with "Modern Fortran" in the...
View ArticleDiagnostic 15033: loop was not vectorized: modifying order of operation not...
Cause:A summation operation or reduction can be vectorized by breaking it up into a separate partial sum for each vector lane, and then adding together the partial sums at the end. Because this changes...
View ArticleDiagnostic 15015: loop was not vectorized: unsupported data type (Fortran)
Cause:The vectorizer does not support certain data types because there is no corresponding SIMD instruction. For example, the compiler supports REAL(16) arithmetic (quad precision) through a software...
View ArticleStatic linking with -mkl, -ipp or -tbb may give unresolved references
Reference Number : dpd200252274Version : Intel® C++ Compiler and Intel® Fortran Compiler versions 13.0.1, 13.1.any, 14.0.0 and 14.0.1Operating System : Linux*, OS X*Problem Description : When an...
View ArticleClik here to view.
Upgrade Your Fortran Compiler
Still using an old version of Intel® Fortran Compiler or Intel® MKL? You're probably leaving performance on the table.You have an opportunity to increase your applications performance by taking...
View ArticleClik here to view.
Viewing load imbalance in OpenMP* applications with Intel® VTune™ Amplifier XE
With Intel® VTune™ Amplifier XE 2013 Update 12 and earlier it was possible to profile OpenMP applications with parallel regions as described in the article Profiling OpenMP* applications with Intel®...
View ArticleInspector XE's pinbin relocation error
Problem:When the user ran Inspector XE's thread check (Update 7-9), like as -inspxe-cl -collect ti3 -- applicationThen displayed,/opt/intel/inspector_xe_2011/bin32/pinbin: relocation error:...
View ArticleClik here to view.
Debugging Intel® Xeon Phi™ Applications on Windows* Host
ContentsIntroductionDebug Solution for Intel® MICHow to get it?Debug Solution as IntegrationComponents RequiredConfigure & TestPrerequisite for DebuggingDebugging Applications with Offload...
View ArticleClik here to view.
Debugging Intel® Xeon Phi™ Applications on Linux* Host
ContentsIntroductionDebug Solution for Intel® MICHow to get it?Why use the provided GNU* GDB from Intel?Why is Intel providing a Command Line and Eclipse* IDE Integration?Depreciation...
View Article