Package Release Info


Update Info: Base Release
Available in Package Hub : 15 SP1





Change Logs

* Mon Feb 18 2019 Egbert Eich <>
- Expose PAPI_LIB from module also when no devel package is installed
* Thu Oct 04 2018 Michal Suchanek <>
- Fix pcp build (boo#1107637).
  libpapi-devel depends on papi which was dynamically linked in the past and
  pulled in libpapi. It is linked statically now so the libpapi dependency
  needs to be added.
* Wed Aug 22 2018
- Update to 5.6.0 (FATE#321456).
  Major changes:
  * Validation tests:  A substantial effort to add validation tests to PAPI to
    check and detect problems in the definition of PAPI preset events.
  * Event testing:  Thorough cleanup of code in the C and Fortran testing to
    add processor support, cleanup output and make the testing behavior
  * CUDA component:  Updated and rewritten to support CUPTI Metric API
    (combinations of basic events).  This component now supports NVLink
    information through the Metric API. Updated testing for the component.
  * NVML component: Updated to support power management limits and improved
    event names. Minor other bug fixes.
  * RAPL component: Added support for: Intel Atom models Goldmont / Gemini_Lake
    / Denverton, Skylake-X / Kabylake
  * PAPI preset events:  Many updates to the PAPI preset event mappings;
    Skylake X support, initial AMD fam17h, fix AMD fam16h, added more Power8
    events, initial Power9 events.
  Other changes:
  * Updating man and help pages for papi_avail and papi_native_avail.
  * Powercap component:  Added test for setting power caps via PAPI powercap
  * Infiniband component:  Bugfix for infiniband_umad component.
  * Uncore component:  Updated to support recent processors.
  * Lmsensors component updated to support correct runtime linking, better
    events name, and a number of bug fixes.
  * Updated and fixed timer support for multiple architectures.
  * All components: Cleanup and standardize testing behavior in the components.
  * Build system: Much needed cleanup of configure and make scripts.
  * Support for C++ was enhanced.
  * Enabling optional support for reading events using perfevent-rdpmc on
    recent Linux kernels can speed up PAPI_read() by a factor of 5.
  * Pthread testing limited to avoid excessive CPU consumption on highly
    parallel machines.
- Refresh papi-codecleanup.patch
- Remove papi-power9_event_list_presets.patch:
  obsoleted by update.
- Add papi-fix-ldflags.patch: fix libdl linking errors
* Fri Feb 09 2018
- Fix summary in module files (bnc#1080259)
* Sat Jan 06 2018
- Fix unexpanded rpm macro in environment module file for HPC (boo#1074897).
* Sun Nov 12 2017
- Add lua-lmod as dependency to HPC package.
* Mon Oct 30 2017
- Fix Requires for HPC build.
* Tue Oct 17 2017
- Add magic to limit the number of flavors built in the
  in the OBS ring.
* Thu Oct 12 2017
- Generate baselib.conf dynamically and only for the non-HPC
  build: this avoids issues with the source validator.
* Wed Oct 11 2017
- Work around openSUSE:Factory spec file validation test.
* Mon Sep 11 2017
- Converted to multibuild.
- Add HPC build using environment modules
* Mon Sep 11 2017
- Add papi-power9_event_list_presets.patch from IBM to add support
  for the POWER9 architecture (fate#322255).
* Thu Nov 24 2016
- Update to version 5.5.1:
  * Support for uncore performance monitoring events on Intel Xeon
    Phi Knights Landing (KNL).
  * Minor bugfixes.
* Thu Sep 15 2016
- Update to 5.5.0. Changes since 5.4.1 :
  New Platforms:
  * Added  Knights Landing (KNL) core events and preset events.
  * Added Intel Broadwell/Skylake/Knights Landing RAPL support
  * Updated PAPI preset event support for Intel Broadwell/Skylake
  New Implementations:
  * libmsr component: Using LLNL's libmsr library to access Intel
    RAPL (Running Average Power Limit) library adds power capping
    abilities to PAPI.
  * CUDA PC sampling: A new standalone CUDA sampling tool
    (papi_cuda_sampling) has been added to the CUDA component
    (components/cuda/sampling/) and can be used as a preloader to
    perform PC sampling on Nvidia GPUs which support the CUPTI
    sampling interface (e.g. Maxwell).
  * ARM Cortex A53 support: Event definitions added.
  New Component:
  * Powercap component: PAPI now supports the Linux Power Capping
    Framework which exposes power capping devices and power measurement
    to user space via a sysfs virtual file system interface.
  * Added Haswell-EP uncore support
  * Initial Broadwell, Skylake support
  * Added a general CUDA example (components/cuda/test) that uses
    LD_PRELOAD to attach to a running CUcontext.
  * Added "-check" flag to papi_avail and papi_native_avail to
    test counter availability/validity.
  * Add support for multiple flavors of POWER8 processors.
  * Force all processors to check event schedulability by checking that
    PAPI can successfully read the counters.
  * Support for Intel Broadwell-EP, Skylake, Goldmont, Haswell-EP
    inherited from libpfm4.
  * Shared memory object (.so) naming is made more limited so that minor
    updates do not break ABI compatibility.
  Bug Fixes:
  * Clean output from papi_avail tool when there are no user defined events.
  * Support PAPI_GRN_SYS granularity for perf component.
  * Bug fix for infiniband_umad component.
  * Bug fix for vmware component.
  * Bug fix for NVML component.
  * Fixed RAPL component so it reports unsupported inside a guest VM.
  * Cleanup ARM CPU detection.
  * Bug fix for PAPI_overflow issue for multiple eventsets.
  * Increased PERF_EVENT_MAX_MPX_COUNTERS to 192 from 128.
  * Fixed memory leak in papi_preset.c.
  * Free allocated memory in the stealtime component.
  * Improve testlib error messages if a component fails to initialize.
  * Fix _papi_hwi_postfix_calc parsing and robustness.
  * Clean build rules for CUDA sampling subcomponent.
  * Correct IBM Power7 and Power8 computation of PAPI_L1_DCA.
  * Eliminate the sole use of ftests_skip subroutine.
  * Correct the event string names for tenth.c.
  * Have Fortran test support code report errors more clearly.
  * Cleanup output from libmsr component.
  * PAPI internal functions were marked as static to avoid exposing them
  * Multiple component were fixed to make internal functions static
    where possible, to avoid exposing the functions as externally
    accessible entry points.
  * CUDA component configuration bug fixed.
* Wed Mar 04 2015
- Replace ./configure with macro %configure
* Wed Mar 04 2015
- fix typo in excludearch statement
* Wed Mar 04 2015
- Cleanup spec file with spec-cleaner
- Update to 5.4.1
  * Updated support for Intel Haswell and Haswell-EP
  * Added ARM Cortex A7
  * Added ARM 1176 CPU (original Raspberry Pi)
  * Enhance PAPI preset events to allow user defined events.
  * User defined events are set up via a user event definition file.
  * New test demonstrating attaching an eventset to a single CPU
    rather than a thread.
  * Use the term "event qualifiers" instead of "event masks" to
    clarify understanding.
  * Added pkg-config support to PAPI.
  * Fixed lustre segfault bug in lustre component.
  * Fixed compilation in the absence of a Fortran compiler.
  * Fixed bug in krental_pthreads ctest to join threads properly
    on exit.
  * Fixed bug in perf_events where event masks were not getting
    cleared properly.
  * Fixed memory leak bug in perf_events.
* Tue Nov 18 2014
- Update to 5.4.0:
  New Platforms:
  - ------------
  * EMON power component for IBM Blue Gene/Q
  * Support for the Applied Micro X-Gene processor
  * Support for IBM POWER non-virtualized platform
  * RAPL support for Intel Haswell models (60,69,71)
  - -----------
  * Added list of supported PMU names (core/uncore components)
  * Support for extended event masks (core/uncore components)
  * Extension of the RAPL energy measurements on Intel via msr-safe
  * Updated IBM POWER7, POWER8 presets
  * 'papi_native_avail --validate' supports events that require
    multiple masks to be valid
  Bug Fixes:
  - ---------
  * HW counter and event count added/fixed for BGPM components
  * Reduce cost of using PAPI_name_to_code
  * Non-null terminated strings fixed
  * Growing list of native events in core/uncore components fixed
  * Cleaned up Intel IvyBridge presets
  * Addressed Coverity reported issues
  For specific and detailed information on changes made in this
  release, grep ChangeLogP540.txt in the package documentation for
  keywords of interest or go directly to the PAPI git repo.
- Remove papi-PPC64-sys_mem_info-array-size-is-wrong.patch:
  obsoleted by update.
* Mon Oct 20 2014
- Add papi-PPC64-sys_mem_info-array-size-is-wrong.patch from IBM to
  fix access beyond array bounds on POWER8 (bnc#901134).
* Fri Oct 03 2014
- Update to version 5.3.2
  - NVML component updates;
  - Addressed appio memory leaks;
  - Support for Haswell-EP added to RAPL component;
  - The perf_event_uncore component event enumeration works now;
  - All components now have appropriate domain and granularities.