Historique des commits

Auteur SHA1 Message Date
  Nathalie Furmento 6cacff4980 mpi/tests/user_defined_datatype.c: make sure tags are different for each iteration il y a 12 ans
  Nathalie Furmento 8ef6b950f3 tests/loader.c: send SIGQUIT on timeout to generate core file il y a 12 ans
  Andra Hugo 5c1a97e5c7 fix resize for lp2 (compute correctly the bw && the latency) il y a 12 ans
  Samuel Thibault d95e1c8398 explain helgrind about some races we are fine with il y a 12 ans
  Samuel Thibault 95e0b19b3e Add some missing mutex/cond/rwlock_destroy il y a 12 ans
  Samuel Thibault 4b2844fb36 Also automatically release opencl memory on shutdown il y a 12 ans
  Andra Hugo a603573008 fix valgrind issue il y a 12 ans
  Samuel Thibault 32e124f6b6 revert spurious change in r9000, keep the suppression for now il y a 12 ans
  Samuel Thibault c44a90ff25 after r8987, mem chunks can now not have a data pointer il y a 12 ans
  Samuel Thibault add756f1af Note about racyness. We can't really add a suppression for this case, it would hide potential other races il y a 12 ans
  Samuel Thibault 0cd466b90a revert r8991: the busy_mutex must be taken after the spinlock, so we can not take the spinlock after the mutex. Reading an int really is atomic on all our targets anyway. il y a 12 ans
  Samuel Thibault 5e18100b8c more suppressions il y a 12 ans
  Samuel Thibault 78e6c17e27 Fix setting memory limit above 2GiB il y a 12 ans
  Samuel Thibault d3065fdc1b Make valgrind yield to other threads when busy-looping, otherwise we may get stuck il y a 12 ans
  Samuel Thibault a1db18eda4 count data as freed only after it is actually freed il y a 12 ans
  Samuel Thibault 5b3d87d1c3 Permit to disable sequential consistency for a given task. il y a 12 ans
  Samuel Thibault 450dc26dec more suppression il y a 12 ans
  Samuel Thibault bc6f8a0868 add more valgrind suppression counterpart il y a 12 ans
  Samuel Thibault bb2c29edc8 Fix race on a system where reading an int wouldn't be atomic. il y a 12 ans
  Samuel Thibault 0fe579e3ec suppress more race warnings about termination, which is not racy il y a 12 ans
  Samuel Thibault e95a4579fd Avoid taking the global tag lock after having taken some tag spinlock. Fixes race in main/tag_wait_api il y a 12 ans
  Samuel Thibault c8ff9b7dc6 Get size before freeing interfaces il y a 12 ans
  Samuel Thibault f24ad3cbab rework deallocating handle memory chunks: we do not actually need to take the burden of going through the whole mc_list, since we already have the mc pointer. This also permits to properly mark per-worker chunks as deallocated il y a 12 ans
  Samuel Thibault dc4867f7e8 tasks[i] are not actually tasks, but arrays of tasks :) il y a 12 ans
  Samuel Thibault b1ae105330 With the locking fixes from r8919, _starpu_request_mem_chunk_removal does not need to access the child handle any more il y a 12 ans
  Samuel Thibault 3ff66999cd Add missing task cleaning il y a 12 ans
  Nathalie Furmento 5ea90822ca mpi/examples/stencil/stencil5.c: reduce values when quick-check is enabled il y a 12 ans
  Andra Hugo 2ac86bbbe1 fix lp2: take latency & bw into account il y a 12 ans
  Nathalie Furmento e3df31128b mpi/src/starpu_mpi_collective.c: only allocate memory if needed il y a 12 ans
  Nathalie Furmento 32cfa5ab67 mpi: New configure option --enable-mpi-progression-hook to enable the il y a 12 ans