On Sun, Nov 24, 2013 at 7:33 AM, Eric Covener <cove...@gmail.com> wrote:
> On Sun, Nov 24, 2013 at 5:24 AM, Rainer Jung <rainer.j...@kippdata.de> > wrote: > > On 24.11.2013 01:03, Eric Covener wrote: > >>>> I'm curious what other scenarios will fail though. I can try Sun > Studio on > >>>> Solaris 10 x86_64 (32-bit and 64-bit builds) "soon". But Sun Studio > on > >>>> SPARC presumably uses different explicit code in APR and I don't have > access > >>>> to that. > >>> > >>> I will try to get that one up and running on 2.4.7. > >> > >> No startup error with 2.4.7 + event on sparc32 or sparc64 with sun > >> studio 12 (cc 5.9) > > > > For the sake of completeness: was apr build with the nonportable atomics > > switch enabled? Don't know whether that really matters, but want to make > > sure we understand the test case. > > Defaults before, retested with no difference under > --enable-nonportable-atomics > Solaris 10 on Intel/AMD, Sun Studio, --enable-nonportable-atomics, 32-bit and 64-bit, 2.4.x HEAD, apr 1.5.x HEAD, apr-util 1.5.x HEAD no problem; uses Solaris-specific atomic implementation, which uses the return-new-value version of decrement The portable, "mutex" version of atomics clearly returns the new value too, though I didn't test with it. -- Born in Roswell... married an alien... http://emptyhammock.com/
