Skip to content
Commits on Apr 11, 2012
  1. @UlfNorell

    erlang:now() doesn't make a good pulse seed

    it improves shrinking, but makes recheck not work
    UlfNorell committed Apr 11, 2012
  2. @UlfNorell

    added no_side_effect on erlang:now()

    also use erlang:now() instead of pulse:seed() to make
    quickcheck pick a fresh seed in every shrinking step
    UlfNorell committed Apr 11, 2012
  3. @UlfNorell
  4. @UlfNorell

    Better command data for open

    UlfNorell committed Apr 11, 2012
Commits on Apr 4, 2012
  1. @hanssv

    No need to explicitly start pulse_application_controller

    This is done automatically when calling (pulse_)application:start/1 in
    the latest pulse_otp.
    hanssv committed Apr 4, 2012
  2. @hanssv
Commits on Apr 3, 2012
  1. @hanssv
  2. @hanssv

    Updated pulse_send

    Now uses the correct format required by the improved PULSE. Also turned the
    function call into a Macro in order to get access to source location!
    hanssv committed Apr 3, 2012
Commits on Mar 30, 2012
  1. @UlfNorell
Commits on Mar 28, 2012
  1. @hanssv
  2. @hanssv
  3. @hanssv

    Added a note to rebar.config

    The tests can't be run without the pulsified versions of some OTP
    libraries (pulse_gen, pulse_gen_server, etc...) and a new version of
    PULSE.
    hanssv committed Mar 28, 2012
  4. @hanssv
  5. @hanssv

    Patching bitcask_nifs for PULSE usage

    Erlang messages are sent from the C-code (bitcask_nifs.c). This
    will not work when scheduled by PULSE. Since it is not possible
    
    Patching bitcask_nifs for PULSE usage
    
    Erlang messages are sent from the C-code (bitcask_nifs.c). This
    will not work when scheduled by PULSE. Since it is not possible
    to send messages to registered processes from C, quite a bit of
    machinery is needed to keep track of the current PULSE process
    so that messages can be re-routed properly.
    
    Everything is ifdef:ed, so when compiling normally, no code is
    added to bitcask.so.
    
    Usage:
    bitcask_nifs:set_pulse_pid(Pid) is used to store the current
    PULSE pid, after that messages sent from the C code will be
    routed properly.
    If no call is made to bitcask_nifs:set_pulse_pid is made, the
    behavior is unspecified!!
    hanssv committed Mar 28, 2012
  6. @hanssv

    Add pulse_instrument compile options

    More bitcask modules are instrumented by PULSE.
    All compile options are ifdef:ed with PULSE
    hanssv committed Mar 28, 2012
  7. @hanssv

    Changed rebar.config

    The new PULSE tests consider a lot more side-effects and uses
    pulse_gen_server, pulse_supervisor, etc.
    
    We also need to instrument the C-code, therefore there is a line
    in port_envs for setting -D PULSE.
    hanssv committed Mar 28, 2012
  8. @hanssv

    Removing old pulse-test

    The old eqc pulse property does not work with the
    new pulse/rebar.config.
    hanssv committed Mar 28, 2012
Commits on Mar 6, 2012
  1. @jaredmorrow

    Roll version 1.5.1

    jaredmorrow committed Mar 6, 2012
  2. @jtuple
Commits on Mar 4, 2012
  1. @jtuple

    Fix incorrect NIF error tuples

    jtuple committed Mar 3, 2012
Commits on Feb 19, 2012
  1. @jaredmorrow

    Roll version 1.5.0

    jaredmorrow committed Feb 19, 2012
Commits on Feb 16, 2012
  1. @Vagabond

    Merge pull request #38 from basho/adt-fix-eqc-test

    Fix incorrect timeout and remove flaky counterexample code
    Vagabond committed Feb 15, 2012
Commits on Feb 15, 2012
  1. @Vagabond

    Fix incorrect timeout and remove flaky counterexample code

    Also, cleanup an unused var.
    Vagabond committed Feb 15, 2012
Commits on Feb 9, 2012
  1. @jonmeredith
  2. @jonmeredith

    Removed unused bitcask_fileops:temp_file function.

    Was used by the old hintfile code but is no longer needed.
    jonmeredith committed Feb 9, 2012
  3. @jonmeredith
  4. @jonmeredith

    Restrict regex for finding bitcask files to starting with numbers.

    Thanks for the review Scott and Dan.
    jonmeredith committed Feb 8, 2012
  5. @jonmeredith

    Tighten bitcask data file regex (currently picks up NNNN.bitcask.data*)

    Make sure the regex is tight so that if people move bitcask files
    out of the way during recovery e.g. 123.bitcask.data.bak then they
    are not included when deciding on the next filename.
    jonmeredith committed Feb 8, 2012
Commits on Feb 8, 2012
  1. @jonmeredith
  2. @jonmeredith
Commits on Feb 6, 2012
  1. @jonmeredith

    Added require_hint_crc entry to bitcask.app.src and enabled for eqc t…

    …est.
    
    Without the setting EQC finds and edge case where it truncates
    the hint file on a hint record boundary.  required_hint_crc will
    be set to true by default in a future release.
    jonmeredith committed Feb 6, 2012
  2. @jonmeredith

    Removed original dead hintfile creation code.

    In the distant past casks were written, then reread and a hint file
    was written for them.  Now hints are written concurrently with the
    main cask file so all of this code is dead.
    jonmeredith committed Feb 6, 2012
  3. @jonmeredith

    Added env var to require strict hintfile checking.

    Otherwise it is possible to truncate at the end of a hintfile
    record and be treated as a hint file without CRCs.
    jonmeredith committed Feb 6, 2012
  4. @jonmeredith

    Updated truncated_datafile_test for hintfile changes.

    With the new 'recovery' option for keyfolding, the cask files
    are scanned after the hint file read discovers an object beyond
    the end of the cask file.
    jonmeredith committed Feb 5, 2012
  5. @jonmeredith

    Fixed truncated_merge_test.

    The truncated hint files are now detected so only the two
    data file corruptions are bad.
    jonmeredith committed Feb 5, 2012
Something went wrong with that request. Please try again.