Intel® Advisor Release Notes and New Features

This page provides the current Release Notes for Intel® Advisor. The notes are categorized by year, from newest to oldest, with individual releases listed within each year.

Click a version to expand it into a summary of new features and changes in that version since the last release, and access the download buttons for the detailed release notes, which include important information, such as pre-requisites, software compatibility, installation instructions, and known issues.

You can copy a link to a specific version's section by clicking the chain icon next to its name.

All files are in PDF format - Adobe Reader* (or compatible) required.
To get product updates, log in to the Intel® Software Development Products Registration Center.
For questions or technical support, visit Intel® Software Developer Support.

2018

Beta Update 1

Linux* Release NotesWindows* Release Notes

Find out more and join the beta here!

  • Hierarchical Roofline preview feature. Set the environment variable ADVIXE_EXPERIMENTAL=roofline_ex before launching Advisor (data re-collection may be required).
  • Experimental support for accessing Advisor data via Python API.
  • New recommendations:
    • Force scalar remainder for loops with low mask utilization on AVX512.
    • Extended “Gather recommendation” with “Constant (non-unit)” pattern.
    • Reporting names of serialized or scalar functions preventing vectorization.
  • Roofline enhancements:
    • Customization of roof values and chart borders.
    • Persistence for roofs visibility and custom values.
    • Improved scaling of histogram.
  • Joined Survey and Refinement CLI report.
  • MKL breakdown in Summary.
  • Divided read/write access in Memory Access Patterns report.
  • Added support for Microsoft Visual Studio* 2017.
Beta

Linux* Release NotesWindows* Release Notes

Find out more and join the beta here!

  • Use fast Intel® Advanced Vector Extensions 512 (Intel® AVX-512) instructions on Intel® Xeon® and Intel® Xeon® Phi™ processors and coprocessors
  • Intel® Advisor - Roofline finds high impact, but under optimized loops
  • Intel® Distribution for Python* - Faster Python* applications
  • Stay up-to-date with the latest standards and IDE:
    • C++2017 draft
    • Full Fortran* 2008Fortran 2015 draft
    • OpenMP* 5.0 draft, Microsoft Visual Studio* 2017
  • Accelerate MPI applications with Intel® Omni-Path Architecture Support

Flexibility for Your Needs

  • Application Performance Snapshot – Get quick answers: Does my hybrid code need optimization?
  • Intel® VTune™ Amplifier – Profile private clouds with Docker* and Mesos* containers, Java* daemons

Hear more about our Beta product features in this recorded webinar: Get Onboard with Intel® Parallel Studio XE 2018 Beta

Latest and Greatest Documentation Available Online

2017

Update 4

Linux* Release NotesWindows* Release Notes

Overview:

  • Bug fixes
Update 3

Linux* Release NotesWindows* Release Notes

Overview:

  • Hierarchical Roofline preview feature. Set the environment variable ADVIXE_EXPERIMENTAL=roofline_ex before launching Advisor (data re-collection may be required).
  • Experimental support for accessing Advisor data via Python API. See examples in the {install_dir}/pythonapi/examples. API is subject to change in the future releases.
  • New recommendations:
    • Force scalar remainder for loops with low mask utilization on AVX512.
    • Extended “Gather recommendation” with “Constant (non-unit)” pattern.
  • Roofline enhancements:
    • Customization of roof values and chart borders.
    • Persistence for roofs visibility and custom values (note that auto save delay is 20 seconds).
    • Improved scaling of histogram.
  • Joined Survey and Refinement CLI report.
  • MKL breakdown in Summary.
  • Divided read/write access in Memory Access Patterns report.
  • Added support for Microsoft Visual Studio* 2017.
Update 2

Linux* Release NotesWindows* Release Notes

Overview

  • Cache-Aware Roofline modeling
  • Improved Trip Counts and FLOPS
  • Filtering by Module
  • Dynamic Instruction Mixes
  • Re-finalization
  • Updated OS Support:
    • Support for Windows* Server 2016, Fedora* 25, and Ubuntu* 16.10
    • All license types now support cross-OS analysis

Details:

Cache-Aware Roofline

Cache-Aware Roofline, introduced as a preview in the previous update, is now an official feature of Intel® Advisor, and no longer requires setting an environment variable to access. This visual performance model provides insight into the source of performance limitations and reveals how much potential performance your application is currently leaving on the table, allowing you to spend your time efficiently in optimizing where it counts most.

More information about the Roofline feature can be found in the Intel® Advisor Roofline main article.

Roofline modeling was first proposed in a paper by Berkeley researchers in 2009. The model has been improved since then, particularly in a 2013 paper from the Technical University of Lisbon that introduced the Cache-Aware model, which is now available in Advisor as an automatically-generated chart when both Survey and Trip Counts with FLOPS have been collected. The Roofline chart and Survey report can either be viewed on their own using the toggle bar on the left to switch between them, or side by side by clicking the white ribbon with four gray dots (initially located by the toggle bar).

Improved Trip Counts and FLOPS

The general quality and coverage of the Trip Counts collection has been improved. Collecting Trip Counts data also collects a new Call Count metric for functions.

Trip Counts and FLOPS can now be collected independently of each other. In the GUI, there are checkboxes in the work flow. On the command line, FLOPS can be enabled by adding the -flops-and-masks option to a tripcounts collection, while Trip Counts can be disabled by adding -no-trip-counts.

Example Command Lines:

### Collect only FLOPS ###
advixe-cl -collect tripcounts -no-trip-counts -flops-and-masks -project-dir MyProject -- MyApplication
### Collect only Trip Counts ###
advixe-cl -collect tripcounts -project-dir MyProject -- MyApplication
### Collect both ###
advixe-cl -collect tripcounts -flops-and-masks -project-dir MyProject -- MyApplication
Filtering By Module

Application analysis scope can now be narrowed down before collection begins to cut out extraneous data and reduce overhead for Survey and Trip Counts analysis types, by restricting collection to specific modules. Modules can be filtered by excluding unwanted files or by restricting collection to a specific list of desired modules, using either the command line or the GUI project properties.

Example Command Lines:

advixe-cl -collect survey -module-filter-mode=exclude -module-filter=DoNotAnalyze.so -project-dir MyProject -- MyApplication
advixe-cl -collect survey -module-filter-mode=include -module-filter=AnalyzeMyApp.exe,AnalyzeThisToo.dll -project-dir MyProject -- AnalyzeMyApp.exe
Updated Operating System Support

Intel® Advisor now supports Windows* Server 2016, Fedora* 25, and Ubuntu* 16.10.

All license types now have support for cross-operating system analysis. This allows for the use of a single license for both Linux* and Windows* systems, making it easier to collect data on one operating system and view it on the other. Installation packages for additional operating systems can be downloaded from the Registration Center. More information about this feature is available here.

Dynamic Instruction Mixes

Dynamic Instruction Mixes are a new variation of the Instruction Mixes feature. While the previously-existing Static Instruction Mixes report how many instructions of a given type exist in the code, the Dynamic Instruction Mixes report how many instructions of a given type were actually executed. It should be noted that the Static Mix is counted per iteration while the Dynamic Mix is for the entire execution of the application.

Dynamic Instruction Mixes are currently available only on the command line by adding the –mix option to a survey or trip counts report command, and only on a result that has had trip counts collected with the -flops-and-masks flag.

Example Command Line:

advixe-cl -report survey -mix -project-dir MyProject
Re-Finalization

Survey and Trip Counts analysis results can now be re-finalized from the GUI, allowing for easy correction or updating of binary and source search directories. This may be necessary if source files are moved after compilation, or remotely collected results are opened on a viewing machine before setting the search paths appropriately. Correcting these situations is now as simple as changing the search directories and re-finalizing.

Update 1

Linux* Release NotesWindows* Release Notes

Overview:

  • Cache-Aware Roofline preview feature
  • Analysis workflow improvements
  • Recommendations display in Summary and Refinement Reports
  • New recommendations
  • Ability to stop refinement analysis collection if every site has executed at least once.

Details:

Cache-aware roofline modeling

The Intel Advisor now offers a great step forward in visual performance optimization with a new Roofline model analysis feature. This new feature provides insights beyond vectorization, such as memory usage and the quality of algorithm implementation.

To enable this preview feature, set the environment variable ADVIXE_EXPERIMENTAL=roofline before launching the Intel Advisor.

Analysis workflow improvements:

Analysis workflow improvements include

  • Intel® Math Kernel Library (Intel® MKL) support: Intel Advisor results now show Intel MKL function calls.
  • Improved FLOPs analysis performance.
  • Decreased Survey analysis overhead for the Intel® Xeon Phi™ processor.
  • New category for instruction mixed data: compute with memory operands.
  • Finalize button for “no-auto-finalize” results, when result is finalized on a separate machine. On collecting speciy the -no-auto-finalize option. Then you can copy your project to a different system, Open the project, specify new paths to your source and binary and then you will be ready to finalize.
  • MPI support in the command line dialog. We now have ability to generate the MPI command line. After clicking on the command-line button click on generate MPI command-line.
Recommendations:

Recommendations are now cached in result snapshots to speed up display. Additionally, recommendations now display in in the Summary and Refinement reports.

New recommendation: Vectorize call(s) to virtual method.

Memory analysis:

You can now track a refinement analysis progress, and can stop collection if every site executes at least once.

Initial Release

Linux* Release NotesWindows* Release Notes

Overview:

  • Workflow improvements
    • Batch Mode
    • Improved MPI workflow
  • Memory Access Patterns, Survey and Loop Analytics improvements

Details:

Workflow

Batch mode, which lets you to automate collecting multiple analysis types at once. You can collect Survey and Trip Counts in single run – Advisor will run the application twice, but automatically, without user actions. For Memory Access Patterns (MAP) and Dependencies analyses, there are pre-defined auto-selection criterias, e.g. check Dependencies only for loops with “Assumed dependencies” issue.

Improved MPI workflow allows you to create snapshots for MPI results, so you can collect data in CLI and transfer self-contained packed result to a workstation with GUI for analysis. We also fixed some GUI and CLI interoperability issues.

Memory Access Patterns

MAP analysis now detects Gather instruction usage, unveiling more complex access patterns. A SIMD loop with Gather instructions will work faster than scalar one, but slower, than SIMD loop without Gather operations. If a loop has “Gather stride” category, check new “Details” tab in Refinement report for information about strides and mask shape for the gather operation. One of possible solutions is to inform compiler about your data access patterns via OpenMP 4.x options – for cases, when gather instructions are not necessary actually.

MAP report in enriched with Memory Footprint metric – distance in address ranges touched by given instruction. The value represents maximal footprint across all loop instances.

Variable name is now reported for memory accesses, in addition to source line and assembly instruction. Therefore, you have more accuracy in determining the data structure of interest. Advisor can detect global, static, stack and heap-allocated variables.

We added new recommendation to use SDLT for loops with an “Ineffective memory access” issue.

Survey and Loop Analytics

Loop Analytics tab has got trip counts and extended instruction mix, so you can see compute vs memory instruction distribution, scalar vs vector, ISA details, etc.

We have improved usability of non-Executed code paths analysis, so that you can see ISA and traits in the virtual loops and sort and find AVX512 code paths more easily.

Loops with vector intrinsics are exposed as vectorized in the Survey grid now.

2016

Update 4

Linux* Release NotesWindows* Release Notes

Full support for all analysis types on the second generation Intel® Xeon™ Phi processor (code named Knights Landing)

FLOPS and mask utilization

Tech Preview feature! Accurate hardware independent FLOPS measurement tool. (AVX512 only) Mask aware . Unique capability to correlate FLOPS with performance data.

Workflow

Batch mode, which lets you to automate collecting multiple analysis types at once.  You can collect Survey and Trip Counts in single run – Advisor will run the application twice, but automatically, without user actions. For Memory Access Patterns (MAP) and Dependencies analyses, there are pre-defined auto-selection criterias, e.g. check Dependencies only for loops with “Assumed dependencies” issue.

Improved MPI workflow allows you to create snapshots for MPI results, so you can collect data in CLI and transfer self-contained packed result to a workstation with GUI for analysis. We also fixed some GUI and CLI interoperability issues.

Memory Access Patterns

MAP analysis now detects Gather instruction usage, unveiling more complex access patterns. A SIMD loop with Gather instructions will work faster than scalar one, but slower, than SIMD loop without Gather operations.  If a loop has “Gather stride” category, check new “Details” tab in Refinement report for information about strides and mask shape for the gather operation. One of possible solutions is to inform compiler about your data access patterns via OpenMP 4.x options – for cases, when gather instructions are not necessary actually.

For AVX512 MAP analysis also detects Gather/scatter instruction usage, these instructions allow more code to vectorize but you can obtain greater performance by avoiding these types of instructions.

MAP report in enriched with Memory Footprint metric – distance in address ranges touched by given instruction. The value represents maximal footprint across all loop instances.

Variable name is now reported for memory accesses, in addition to source line and assembly instruction. Therefore, you have more accuracy in determining the data structure of interest. Advisor can detect global, static, stack and heap-allocated variables.

We added new recommendation to use SDLT for loops with an “Ineffective memory access” issue.

Survey and Loop Analytics

Loop Analytics tab has got trip counts and extended instruction mix, so you can see compute vs memory instruction distribution, scalar vs vector, ISA details, etc.

We have improved usability of non-Executed code paths analysis, so that you can see ISA and traits in the virtual loops and sort and find AVX512 code paths more easily.

Loops with vector intrinsics are exposed as vectorized in the Survey grid now.

Update 3

Linux* Release NotesWindows* Release Notes

Support for Intel® Xeon Phi™ processor (codename: Knights Landing)

You can run the Survey analysis on the Intel® Xeon Phi™ processors (codename Knights Landing), see Intel AVX-512 ISA metrics and new Intel AVX-512-specific traits, e.g. Scatter, Compress/Expand, masking manipulations, etc.

Other analysis types are not available for Intel® Xeon Phi™ processors yet. We will extend the platform support in future updates.

Analysis of non-Executed code paths

Using the Intel Advisor XE, you can get your code ready for the next generation CPUs and Intel® Xeon Phi™ coprocessors even if you don’t have access to the hardware yet. Enable this functionality by generating code for multiple vector instruction sets (including Intel AVX-512) using the Intel Compiler
–ax option, and then analyze the resulting binary with Vectorization Advisor.

Read the detailed article about using the feature with some examples here.

New tab: “Loop Analytics”

“Loop Analytics” combines several loop metrics in single place, with easy-to-read visualization. The tab contains brand new metric “Instruction mix” - percentage of Memory, Compute and other instructions with distribution to vector/scalar instruction types. Other metrics are ISA, Traits and Efficiency. In future updates we plan to extend this tab with even more information, such as detailed instruction mix break-downs,  explanation and performance impact break-down for various sophisticated vectorization notions (multi-pumping, vectorized remainder) and more.

Filter results data by thread

You can now filter data by thread in the Survey report view, to narrow down performance metrics:

New recommendations

We have updated recommendation for Fortran "contiguous" attribute and added two new recommendations:

  1. “Indirect function calls” recommendation, that is useful for vectorizing C++ codes with virtual methods/functions.

  1. Recommendation for better utilization of FMA instructions:

Extended CLI snapshot

You can now pack your Advisor results with caching of sources and binaries from command line. So, you can make self-contained packed result snapshot on a remote server or a cluster node, which then can be easily copied to your workstation for investigation in GUI. This option is also handy for keeping historical results:

advixe-cl --snapshot --project-dir ./my_proj --pack --cache-sources --cache-binaries -- /tmp/my_proj_snapshot
Update 2

Linux* Release NotesWindows* Release Notes

Loops, Functions and SIMD-enabled Functions

We have extended the Survey grid with functions (in addition to loops). Therefore, you can explore the complete hotspots list and call tree, including loops, functions, and SIMD-enabled functions:

New Recommendations

New issue “Vector register spilling” supersedes previous “High register pressure” issue. The new issue is more accurate and the recommendations are more actionable. The number of actually used vector registers appears in a separate column. If many vector registers are in use, there can be not enough registers to store some data, and compiler will have to generate extra code for storing some other data in memory. This increases number of instructions used and incurs additional memory accesses, thus negatively affecting performance.

The “OpenMP function call(s) present” recommendation suggests to refactor code that uses routines like “omp_get_thread_num()” inside a parallel region, since they may constrain performance:

Intel Advisor can now detect scalar math function calls in GCC-compiled code. They can be replaced with SIMD analogues, e.g. from Intel Short Vector Math Library (SVML). Below is an example of a GCC-compiled code that doesn’t use vector math functions by default. Intel Advisor recommends methods for enabling them in the code and in the GCC compiler command line:

Persistent Loops Selection for Easier CLI/GUI Interaction

The list of selected loops is now stored in a separate file and is persistent in the Intel Advisor project. This makes possible such workflows as:

  • If you work on a single machine, collect results in CLI and browse them in GUI:
    • When browsing in GUI, select loops of interest for deeper analysis (Dependencies or Memory Access Patterns)
    • Run deeper analysis in CLI under the same project. The selection made in GUI will be already applied for the CLI collection. No need in using the “-mark-up-list” option.
  • Collect results on a Linux server with CLI, then transfer results to another workstation to browse in GUI:
    • Select loops of interest in GUI, then copy the project back to a server. If project is too big, you can transfer only this file: “e000/loop_hashes.def” (to save the selection, close the results before transferring the file).
    • On server, simply run deeper analysis (Dependencies or Memory Access Patterns) without specifying the “-mark-up-list” option. The selection will be automatically picked up from the project.
    • If you need to change the selection in CLI, use the “advixe-cl -mark-up-loops” command.
    • Loop selection is persistent, so you do not need to specify the markup list for the “-collect” command.

Decreased Finalization Time for Trip Counts Collection

You can now run the Trip Counts analysis independently from the Survey analysis, so you may run them in any order.

  • If you collect Survey data first, and run Trip Counts later on, the behavior doesn’t change.
  • If you collect Trip Counts first, without running Survey, the collection will finalize only after running the Survey analysis. This triggers only one finalization for the two analyses, saving time on results post-processing. You will be able to view both Trip Counts and Survey results when both collections complete.

Memory Footprint and Variable Names

The Memory Access Patterns analysis now has the Estimated Memory Footprint metric in the MAP Source View. So, you get an idea about the amount of memory touched during the loop execution:

Support of the variable names for object-oriented code in MAP/Dependencies analysis was extended. Now variable names are detected for global and stack-allocated objects on Windows and Linux operating systems:

Update 1

Linux* Release NotesWindows* Release Notes

Intel Advisor XE now supports Microsoft Windows* 10 and integrates into Microsoft Visual Studio* 2015.

We significantly improved performance and responsiveness of data loading and finalization in Survey, Trip Counts, MAP and Dependencies reports.

We have improved recommendations accuracy, and made them separate for C++ and Fortran codes. So that you’ll see more relevant code samples and recommendations.

You can now create Data Snapshot to transfer read-only result as a single file from a remote server to your workstation:

Linux OS system administrators and users will enjoy new capability to load/unload Advisor as an Environment Module when setting up product environment.

Summary pane

Advisor XE Vectorization Summary pane has gotten plenty of improvements. For example, now it has a list of top time consuming loops, so you can quickly focus on the hotspots and jump into details by clicking a loop hyperlink:

In “Refinement analysis data” section on Summary pane, we have added new column “Site Name”. This is a hyperlink to corresponding site (selected loop) on “Refinement” Tab.

Survey analysis

We have improved data representation in Survey report: parent-child relationship, filtering and highlighting. Also, we improved co-existence of Threading and Vectorization Advisor layouts.

Advisor XE now differentiates inlined functions in the Survey Top Down grid, so you know more about your application call tree:

Sometimes compiler can estimate loop trip count to be 1, and don’t generate actual loop in binary code. With Advisor XE 2016 Update 1 you can see those loops in Bottom Up and Top Down windows. Source location for completely unrolled loops and loops with trip count=1 is attributed to an outer loop source line:

We have improved outer loops vectorization analysis by introducing special Loop Type: “Inside Vectorized”. It helps to distinguish outer SIMD loops from their child loops, where vectorization is only done along the induction variable of parent loop.

With new ability to sort by loop selection column you can clearly see which loops from your long list were selected and manage that accordingly before running Dependencies or Memory Access Patterns analysis. Just click on the “Selection” column title (with the black drop icon):

Refinement analysis

We extended filtering capability to Refinement report, so you can manage Dependencies and Memory Access Patterns reports more efficiently:

Improved variable resolution in Memory Access or Dependencies Analysis allows you to see nested data objects, like C++ class members and understand which particular object has been loaded/stored in unit-stride or irregular manner:

We have added new “stride” type: Uniform Stride, that is accessing to the same memory location within a loop iterations:

Initial Release

Linux* Release NotesWindows* Release Notes

Intel Advisor XE 2016 is a successor product to Intel® Advisor XE 2015. Intel Advisor XE 2016 provides a set of tools that help you decide where to add threading and now vectorization to your applications. The key features provided by this new major version since the Intel Advisor XE 2015 release include:

  • Vectorization assistance in addition to the threading assistance available in the previous product
  • Significant changes to the product workflow because of the addition of vectorization assistance

2015

Update 1

Linux* Release NotesWindows* Release Notes

  • Improved the custom install directory process
  • Incorporated new Intel® Cilk™ Plus overheads
Initial Release

Linux* Release NotesWindows* Release Notes

Intel Advisor XE 2015 is a successor product to Intel® Advisor 2013.

  • Significant new modeling capabilities in the Suitability Report window, including Intel® Xeon Phi™ processor analysis, loop iteration and duration modeling options, and improved scalability graph features and explanations.
  • Faster and more efficient annotation collection provides a more responsive GUI and less system load during annotation collection.
  • The ability to create a snapshot of the active result, save it, and view its data. Saved result snapshots are read-only, but can be compared visually with the current active result or other read-only result snapshots.
  • You can now pause (and later resume) data collection for uninteresting parts of your target program. Use either the Pause Collection and Resume Collection annotations or the pause/resume buttons in the command toolbar of the Survey Report and Suitability Report.
  • New command line options for the advixe command: no-auto-finalize, data-limit, interval, mrte-mode, no-follow-child, start-paused, resume-after, and target-duration-type.
  • Various usability and stability improvements.

2013

Update 5

Linux* Release NotesWindows* Release Notes

  • Support for Microsoft Windows* 8.1 and Visual Studio* 2013 (Windows* only)
  • A new Option lets you choose which Linux editor appears when you double-click a Source window line. The corresponding file and location appears in the specified Linux editor. You can still specify a Linux editor using an environment variable. (Linux* only)
  • The annotation assistant pane now appears with the Suitability Report and Correctness Report "No Data messages" to help you insert annotations (as well as in Survey windows).
  • The annotation wizard drop-down list now provides two new items that let you copy code snippets for both site and task annotations.
  • New command line options for the collect action: data-limit, interval, mrte-mode, no-follow-child, and target-duration-type.
  • Various usability and stability improvements, such as improved wording in the annotation assistant pane.
Update 4

Linux* Release NotesWindows* Release Notes

  • Increased values for the Target CPU Count in the Suitability Report window, including the scalability graph. To modify the current value, specify the Maximum CPU Count in the Options > General dialog box.
  • Ways to skip program initialization for Survey and Suitability analysis, by either starting the tools in a paused collection state or pausing collection for a specified time. To resume collection, either use the Resume Collection annotation or click the Resume button.
  • The side command toolbar is more accessible. For example, to start Survey or Suitability analysis in a paused collection state, use the << button to show it. After analysis completes, use the >> button to hide it, which increases the width available for the Report window.
  • New options that let you display or hide the warning message for missing debug information (Survey) and set the Suitability Report's Maximum CPU Count.
  • Several new messages, a visual indication in the result tab name when a tool is waiting for you to click a button, and other usability enhancements.
Update 3

Linux* Release NotesWindows* Release Notes

  • Improved assistance window
  • Snapshot copy procedure cancellation functionality
  • Improved suitability by excluding paused time
  • New educational sample for matrix multiplication
  • Several usability improvements
Update 2

Linux* Release NotesWindows* Release Notes

  • New Pause/Resume API and GUI functionality
  • New “make snapshot” feature for taking snapshots of results
  • Fortran annotations API improvements
  • Fix for external editor support on Linux*
  • Usability improvements in the Intel® Advisor XE Standalone GUI (both on Windows* and Linux*) as well as the Microsoft Visual Studio* integrated GUI (Windows* only)
Update 1

Linux* Release NotesWindows* Release Notes

  • A small toolbar for the Survey report
  • A Getting Started Tutorial for the Fortran nqueens sample
  • Various bug fixes and improvements
Initial Release

Linux* Release NotesWindows* Release Notes

This is a new product.

Para obtener información más completa sobre las optimizaciones del compilador, consulte nuestro Aviso de optimización.