Permalink
Commits on Apr 6, 2012
  1. Add jnr-x86asm.jar to the build, so jnr-ffi can use it to generate na…

    …tive stubs
    Wayne Meissner committed Apr 6, 2012
  2. Rework FFI jit to wire up DynamicMethod subclasses directly into the …

    …module method cache
    Wayne Meissner committed Apr 6, 2012
  3. Simplify some of the ffi method wrappers

    Wayne Meissner committed Apr 5, 2012
  4. Jitted FFI invokers no longer need a fallback path, as they can handl…

    …e all arg types.
    Wayne Meissner committed Apr 5, 2012
  5. Allow setting of nested struct fields

    Wayne Meissner committed Apr 4, 2012
Commits on Apr 5, 2012
Commits on Apr 4, 2012
  1. Fix TCPServer.accept NPE under concurrent accepts.

    When ServerSocketChannel is woken without a ready connection, such
    as when doing multiple threads accepting on the same server, we
    may get back a null from ServerSocketChannel.accept. Assume that
    means another thread got our channel and go back into the select
    logic.
    headius committed Apr 4, 2012
  2. Mitigation for JRUBY-6576

    Because in JRuby, concurrent selects can trigger blocking mode
    exceptions, I've modified IO.select logic to raise a concurrency
    error when multiple threads try to select on the same sockets at
    the same time (and fail to do so cleanly). At present, there's
    no way to do this since we use these channels in both blocking and
    nonblocking mode; we can't synchronize the blocking mode changes
    with multiple threads all trying to do select operations.
    headius committed Apr 4, 2012
  3. IR: gruesome hack to make sure default listener is conditional for JR…

    …uby::IR.compiler_debug=true/false
    enebo committed Apr 4, 2012
  4. Update to jline2.

    headius committed Apr 4, 2012
  5. Remove dependency enum type and make it passes responsibility to rese…

    …t dependent passes manually
    enebo committed Apr 4, 2012
  6. Fix JRUBY-6570: autoload :Time, 'time' doesn't work

    Instead of always using Object when calling autoload from an
    instance method, we use self's real class (skipping singletons)
    as the autoload location. This isn't exactly how MRI does it
    (they use cbase, which I think is euivalent to 1.8's ruby_class),
    but it should be correct for the majority of cases.
    headius committed Apr 4, 2012
Commits on Apr 3, 2012
  1. Store the result of loading the kernel32 library, so it doesn't get r…

    …eloaded on every call to WindowsFFI.getKernel32().
    Wayne Meissner committed Apr 3, 2012
  2. Store the native callback in the FFI handle slot on the object, where…

    … possible.
    Wayne Meissner committed Apr 3, 2012
  3. Replace all the Updater subclasses/instances with specific methods to…

    … allocate each of the ID, Cext and FFI variable accessors. Reduces and simplifies the code.
    Wayne Meissner committed Apr 3, 2012
  4. Replace some calls to ThreadContext.getRuntime() in helper functions …

    …with ThreadContex.runtime
    Wayne Meissner committed Apr 3, 2012
Commits on Apr 2, 2012
  1. Test for JRUBY-5863.

    headius committed Apr 2, 2012
  2. Fix JRUBY-5863: Named captures cause crash when there is no match

    Compiler calls match2AndUpdateScope[19], which did not pass through
    the match result and instead fell back to logic that used the
    scope's match and assumed it was non-nill.
    headius committed Apr 2, 2012
Commits on Apr 1, 2012
  1. Switch FFI jitted methods to invokeI<x>, invokeL<x>, invokeN<x> helpe…

    …r methods
    Wayne Meissner committed Apr 1, 2012
  2. Rebase PointerParameterStrategy hierarchy on jnr.ffi.ObjectParameterS…

    …trategy
    Wayne Meissner committed Apr 1, 2012
  3. Upgrade jffi to 1.1.0 and jnr-ffi to 0.7.1

    Wayne Meissner committed Apr 1, 2012
Commits on Mar 31, 2012