Commit History

Autor SHA1 Mensaxe Data
  Samuel Thibault aeff50217e Accept going outside of benchmarked nl regression by 10%. Force calibration otherwise %!s(int64=14) %!d(string=hai) anos
  Samuel Thibault d963e59e8d starpu_perfmodel_display can now access to history using the public interface %!s(int64=14) %!d(string=hai) anos
  Samuel Thibault 2b4e61f351 Always make sure that starpu.h gets included first, since that one knows which order has to be used for the rest. Expose history entries to applications. %!s(int64=14) %!d(string=hai) anos
  Nathalie Furmento e482ecdf0a doc: use @var instead of @code when appropriate %!s(int64=14) %!d(string=hai) anos
  Cyril Roelandt 4c1d6fa28d Avoid mixing signed and unsigned (cf. r4584) %!s(int64=14) %!d(string=hai) anos
  Olivier Aumage fc07bb4782 - avoid mixing signed and unsigned %!s(int64=14) %!d(string=hai) anos
  Cyril Roelandt dc3ce715e4 Fix the SSE example. %!s(int64=14) %!d(string=hai) anos
  Nathalie Furmento 3d931891f4 configure.ac: when CUDA is enabled, enable ICC only if it is compatible with CUDA %!s(int64=14) %!d(string=hai) anos
  Samuel Thibault b0bbe73041 include source headers before system headers, to fix build when older starpu is already installed %!s(int64=14) %!d(string=hai) anos
  Samuel Thibault 6c4cb74536 Start plot at 1 %!s(int64=14) %!d(string=hai) anos
  Samuel Thibault bf9bc759c4 Document that non-linear regression reverts to history-based on first run %!s(int64=14) %!d(string=hai) anos
  Samuel Thibault 4b2f334cc5 fix plot when calibrations were not in the same order %!s(int64=14) %!d(string=hai) anos
  Samuel Thibault 089b5aca13 Make nl regression revert to history-based when it is not valid yet, but enough history is available. This permits nice auto-calibration %!s(int64=14) %!d(string=hai) anos
  Samuel Thibault e5c40b85fd drop spurious debugging print %!s(int64=14) %!d(string=hai) anos
  Samuel Thibault c02c566746 Fix plotting measures for non-linear regression %!s(int64=14) %!d(string=hai) anos
  Samuel Thibault b0b2e42d5f Only use non-linear regression if the requested size falls inside the covered measures %!s(int64=14) %!d(string=hai) anos
  Samuel Thibault ad7869523c Add missing piece of commit %!s(int64=14) %!d(string=hai) anos
  Samuel Thibault 0a06263c45 Only validate a regression model if we have at least 10% difference in input size. Do not validate the non-linear regression model just due to number of samples. %!s(int64=14) %!d(string=hai) anos
  Samuel Thibault 2d8c3bd434 mention which schedulers support choosing the implementation %!s(int64=14) %!d(string=hai) anos
  Samuel Thibault 7a0f65790f Also build vector_scal with icc when available %!s(int64=14) %!d(string=hai) anos
  Samuel Thibault f1cb2d818f replace tabs with spaces %!s(int64=14) %!d(string=hai) anos
  Ludovic Courtès 863adbed76 doc: Clarify the wording in "Partitioning Vector Data". %!s(int64=14) %!d(string=hai) anos
  Ludovic Courtès eb918faec0 Remove `doc/version.texi' from the repository. %!s(int64=14) %!d(string=hai) anos
  Ludovic Courtès bc7cb0d9cf test_interfaces: Remove malloc from the `vector' test; use a descriptive name. %!s(int64=14) %!d(string=hai) anos
  Ludovic Courtès 33a2088388 test_interfaces: Change `register_func_t' to return a `starpu_data_handle'. %!s(int64=14) %!d(string=hai) anos
  Ludovic Courtès 6d314b81aa Properly parenthesize `STARPU_MULTIPLE_*_IMPLEMENTATIONS' macros. %!s(int64=14) %!d(string=hai) anos
  Samuel Thibault de83330e6a We have to also dump empty models, for implementations which aren't calibrated yet %!s(int64=14) %!d(string=hai) anos
  Samuel Thibault c163358f96 Handle non-calibrated implementations %!s(int64=14) %!d(string=hai) anos
  Nathalie Furmento b78b99afa3 doc/chapters/perf-feedback.texi: typo %!s(int64=14) %!d(string=hai) anos
  Nathalie Furmento 235b99337f tools/dev: update w.r.t coccinelle version 1.0.0-rc7 %!s(int64=14) %!d(string=hai) anos