summaryrefslogtreecommitdiff
path: root/sccomp/qa
AgeCommit message (Collapse)AuthorFilesLines
2017-11-30Weaken SwarmSolverTest::testUnconstrained even further for nowStephan Bergmann1-1/+1
...after 1fa761af825641da5c87f80c2a17135f92418960 "Ridiculously large delta for SwarmSolverTest::testUnconstrained for now", to accommodate further Jenkins lo_ubsan failures with actual values of 3.67055466470122 (<https://ci.libreoffice.org/job/lo_ubsan/741/console>) and 3.88389164367578 (<https://ci.libreoffice.org/job/lo_ubsan/743/console>). Change-Id: Ibacb25ba82c2c279ef8dcd19c5ce7f6d5d8014d5 Reviewed-on: https://gerrit.libreoffice.org/45520 Reviewed-by: Stephan Bergmann <sbergman@redhat.com> Tested-by: Stephan Bergmann <sbergman@redhat.com>
2017-11-25Ridiculously large delta for SwarmSolverTest::testUnconstrained for nowStephan Bergmann1-1/+1
At least the Jenkins lo_ubsan build sporadically keeps failing CppunitTest_sccomp_swarmsolvertest due to rather widely off values here (3.00064356398118 in <https://ci.libreoffice.org/job/lo_ubsan/736/console> and 2.8639022684852 in <https://ci.libreoffice.org/job/lo_ubsan/738/console>), so suggestion by Tomaž Vajngerl was: "Let's adapt the delta for now. Generally anything close to 3 should be acceptable as the algorithm greatly depends on random values." Change-Id: Id8f8657e521381decb1c719a1cffdf959ae74ef4 Reviewed-on: https://gerrit.libreoffice.org/45231 Tested-by: Jenkins <ci@libreoffice.org> Reviewed-by: Tomaž Vajngerl <quikee@gmail.com>
2017-11-21clang-format fixesMiklos Vajna1-1/+0
Hopefully the last such commit, now that the in-gerrit verification is in place and also a warning is printed locally for those who push directly. Change-Id: Ib80cbca0b29dc7f71cbb182a38ad6839a659ff41 Reviewed-on: https://gerrit.libreoffice.org/45002 Reviewed-by: Miklos Vajna <vmiklos@collabora.co.uk> Tested-by: Jenkins <ci@libreoffice.org>
2017-11-21CppunitTest_swarm_solver_test also consistently fails for UBSan buildStephan Bergmann1-2/+2
...at least for my local one, with > /data/sbergman/lo-san/core/sccomp/qa/unit/SwarmSolverTest.cxx:387:(anonymous namespace)::SwarmSolverTest::testMultipleVariables > double equality assertion failed > - Expected: 0 > - Actual : 1 > - Delta : 1e-05 so disable that part unconditionally for now. Change-Id: Id8c091781baf4381696e2a8f0501df1f3f83724f Reviewed-on: https://gerrit.libreoffice.org/44977 Tested-by: Jenkins <ci@libreoffice.org> Reviewed-by: Tomaž Vajngerl <quikee@gmail.com>
2017-11-20WaE: -Werror=unused-functionCaolán McNamara1-4/+16
Change-Id: I773224bd332669c86648578a47612fa6d639c9c5 Reviewed-on: https://gerrit.libreoffice.org/44945 Tested-by: Jenkins <ci@libreoffice.org> Reviewed-by: Caolán McNamara <caolanm@redhat.com> Tested-by: Caolán McNamara <caolanm@redhat.com>
2017-11-18Swarm based (uses PSO or DE) experimental non-linear solverTomaž Vajngerl4-0/+399
This is a new, simple non-linear solver that uses a swarm (population) to do global optimization. It uses two algoritms - Particle Swarm Optimization (PSO) or Differential Evolution (DE) to find a (non-optimal) solution. It is experimental as not all functions are implemented and it needs a lot more testing so that it performs well. Change-Id: If55dad7eda17394851a9d178ad892de771eca7c9 Reviewed-on: https://gerrit.libreoffice.org/44382 Tested-by: Jenkins <ci@libreoffice.org> Reviewed-by: Tomaž Vajngerl <quikee@gmail.com>
2017-08-27sccomp_lpsolver -> sccomp_solverRene Engelhard1-0/+0
This test also tests the CoinMP solver, no _lpsolver is misleading Change-Id: If56de3aa84ec80b1888aadeb940e08e65083c9e9 Reviewed-on: https://gerrit.libreoffice.org/41607 Reviewed-by: Rene Engelhard <rene@debian.org> Tested-by: Rene Engelhard <rene@debian.org>
2017-04-06loplugin:redundantcast find cstyle double castsNoel Grandin1-1/+1
Change-Id: I5507be190dac781e5cdb545a60acf3d50056c9f8 Reviewed-on: https://gerrit.libreoffice.org/36187 Tested-by: Jenkins <ci@libreoffice.org> Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
2017-03-02change the solver test to explicitly test the available solversTomaž Vajngerl1-35/+23
This changes the lpsolver test to explicitly test the available linear solvers we have - either CoinMP, LpSolver or both. This prevents that a newly added solver will be automatically tested as it can have a different behaviour for the tested input values. Change-Id: I0c4d2f9c561d1e834ca51196b7b5ecf7d89ba550 Reviewed-on: https://gerrit.libreoffice.org/34813 Tested-by: Jenkins <ci@libreoffice.org> Reviewed-by: Tomaž Vajngerl <quikee@gmail.com>
2015-10-28com::sun::star->css in scaddins,sccomp,scriptingNoel Grandin1-1/+1
Change-Id: Id9167341940bac65f055e1e33ff0670f6fa0f6c4 Reviewed-on: https://gerrit.libreoffice.org/19629 Tested-by: Jenkins <ci@libreoffice.org> Reviewed-by: Noel Grandin <noelgrandin@gmail.com>
2015-10-12Replace "SAL_OVERRIDE" with "override" in LIBO_INTERNAL_ONLY codeStephan Bergmann1-2/+2
Change-Id: I2ea407acd763ef2d7dae2d3b8f32525523ac8274
2014-04-02sccomp: Add very simple unit test for lp solvers.Matúš Kukan1-0/+128
Change-Id: I359737ad57decf2b1988afbceab85645fbbf1f33