----- Original Message ----- > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > Hello, > > On 23.09.2011 17:52, Igor Galić wrote: > > in a long time to do so now. Please cast your votes or concerns > > for the remaining. > > First, 3.0.x as of right now builds fine on my Debian box. However I > noticed a non critical issue. Regressions checks don't compile on gcc > 4.6 (once again) in combination with -Werror which is used in default > builds: > > test_atomic.cc: In function ‘int main(int, const char**)’: > test_atomic.cc:214:7: error: cast to pointer from integer of > different > size [-Werror=int-to-pointer-cast] > cc1plus: all warnings being treated as errors > > This does not occur in trunk, as Igor consolidated string functions > there, which, "by accident" also fixed that compile time issue. > Probably > its not desirable to backport the full set of invasive changes to > 3.0.x > though.
Patch+Jira: http://svn.apache.org/viewvc?rev=1171365&view=rev https://issues.apache.org/jira/browse/TS-969 I added a STATUS entry. > Also I will probably patch my Debian build with my patch proposed in > TS-967 due to a compile time build flags transition in Debian which > passes several CFLAGS now, including but not limited to hardening > flags. > I am, however, perfectly fine if that fix is not (yet?) backported to > a > 3.0.x stable release. I haven't had any cycles to review that patch, although it looked very fine on first sight. > Finally I was informed by an user in #trafficserver that parent > caching > is broken in 3.0, but working in 3.1. I see several related patches Does that include 3.0.1 or > in > trunk which could have caused that that fix - but I am not exactly in > the position to judge here. Presuming there is a real issue with > parent > caching (I can't test right now) that should definitively be > backported. Can we track down what exactly is broken and what has fixed it? https://issues.apache.org/jira/browse/TS-906 http://svn.apache.org/viewvc?view=rev&revision=1154744 Which is proposed only seems to deal with auth. It also seems, and I only noticed that now, to break ABI compatibility. Or am I mislead here? > - -- > with kind regards, > Arno Töll > IRC: daemonkeeper on Freenode/OFTC > GnuPG Key-ID: 0x9D80F36D i -- Igor Galić Tel: +43 (0) 664 886 22 883 Mail: i.ga...@brainsware.org URL: http://brainsware.org/ GPG: 571B 8B8A FC97 266D BDA3 EF6F 43AD 80A4 5779 3257
