Re: _REENTRANT and _THREAD_SAFE

2017-02-27 Thread Bruno Haible
Joseph Myers wrote: > Now, the effect is to declare all POSIX.1 (199506) symbols rather than > just getlogin_r. They still have no effect unless you define feature test > macros for an older standard, or select an ISO C conformance mode without > defining feature test macros for a newer standar

Re: _REENTRANT and _THREAD_SAFE

2017-02-27 Thread Joseph Myers
On Mon, 27 Feb 2017, Bruno Haible wrote: > Does it mean that _REENTRANT and _THREAD_SAFE now prevent some symbols > from being declared? No. The effect of those macros used to be: declare getlogin_r. Because getlogin_r is part of POSIX.1 (199506) and the default POSIX version in the glibc hea

_REENTRANT and _THREAD_SAFE

2017-02-27 Thread Bruno Haible
Hi all, gnulib uses the _REENTRANT and _THREAD_SAFE macros in a couple of modules: m4/threadlib.m4 m4/readutmp.m4 Does anyone know in detail what this glibc change [1] means in detail? * The nonstandard feature selection macros _REENTRANT and _THREAD_SAFE are now treated as compatibility s

z/OS and symlink()

2017-02-27 Thread Daniel Richard G.
Currently, test-symlink and test-symlinkat are failing on z/OS: $ ./test-symlink .../test-symlink.h:64: assertion 'errno == ENOTDIR || errno == ENOENT' failed CEE5207E The signal SIGABRT was received. ABORT instruction $ ./test-symlinkat .../test-symlink.h:64: assertion