Author: Armin Rigo <ar...@tunes.org> Branch: extradoc Changeset: r5685:430b20e00289 Date: 2016-08-27 16:35 +0200 http://bitbucket.org/pypy/extradoc/changeset/430b20e00289/
Log: More things to do later diff --git a/planning/py3.5/2016-august-progress.rst b/planning/py3.5/2016-august-progress.rst --- a/planning/py3.5/2016-august-progress.rst +++ b/planning/py3.5/2016-august-progress.rst @@ -38,6 +38,11 @@ * compare ``dir(posix)`` on py3.5 and cpython 3.5. +* review all unwrap_spec() that are meant to pass booleans (now + with '=int'). Argument clinic turns these to PyObject_IsTrue(), i.e. + accepting any object whatsoever(?!), which is supposedly a feature + (see http://bugs.python.org/issue14705). + Milestone 1 (Aug-Sep-Oct 2016) ------------------------------ diff --git a/planning/py3.5/cpython-crashers.rst b/planning/py3.5/cpython-crashers.rst --- a/planning/py3.5/cpython-crashers.rst +++ b/planning/py3.5/cpython-crashers.rst @@ -23,8 +23,17 @@ threads concurrently. It will make two stat objects and leak one of them. -* (not a crasher) on modern Linux: if the first call in the process to + +Other bugs +---------- + +* on modern Linux: if the first call in the process to socketpair() ends in a EINVAL, then cpython will (possibly wrongly) assume it was caused by SOCK_CLOEXEC and not use SOCK_CLOEXEC at all in the future +* fcntl.ioctl(x, y, buf, mutate_flag): mutate_flag is there for the case + of buf being a read-write buffer, which is then mutated in-place. + But if we call with a read-only buffer, mutate_flag is ignored (instead + of rejecting a True value)---ioctl(x, y, "foo", True) will not actually + mutate the string "foo", but the True is completely ignored. _______________________________________________ pypy-commit mailing list pypy-commit@python.org https://mail.python.org/mailman/listinfo/pypy-commit
