Re: [cp-patches] FYI: automake 1.10 fixlet for plugin

2007-04-17 Thread Mark Wielaard
Hi Dalibor, On Tue, 2007-04-17 at 17:17 +0200, Dalibor Topic wrote: > Thank you for putting them on the release branch, I've got a couple more > to come > (include/Makefile.am cleanup, JAVAH handling on regen headers, > lib/Makefile.am portability, > autogen.sh simplification). And there is a pe

[cp-patches] FYI: Fix Solaris FIONREAD issues

2007-04-17 Thread Andrew John Hughes
This fixes an issue with importing FIONREAD on Solaris. We do things right in cpio.c but not in some other files. ChangeLog: 2007-04-17 Andrew John Hughes <[EMAIL PROTECTED]> * native/jni/java-net/local.c: Fix import of FIONREAD. * native/jni/java-nio/gnu_java_nio_VMCh

[cp-patches] FYI: improved javah checking

2007-04-17 Thread Dalibor Topic
Hi all, I've improved the configure check for javah a little bit. cheers, dalibor topic 2007-04-17 Dalibor Topic <[EMAIL PROTECTED]> * m4/acinclude.m4 (CLASSPATH_CHECK_JAVAH): Use AC_CHECK_PROGS and AC_MSG_ERROR. Check for gjavah, gcjh-wrapper-4.1 and gcjh-4.1. Error ou

[cp-patches] UnicastConnectionManager

2007-04-17 Thread Andrew Haley
This bug has taken me three solid days to find. :-( The symptom was that all the JBoss tests failed with "connection refused". A trace on the interface revealed that the listener, explicitly set by the user to "localhost", was listening on 127.0.0.1, but the client connection was attempting to c

[cp-patches] FYI: conditional javah check in configure

2007-04-17 Thread Dalibor Topic
Hi all, I've made the javah check in configure conditional on whether the JNI headers need to actually be regenerated, as that's the only place where we use javah. cheers, dalibor topic 2007-04-17 Dalibor Topic <[EMAIL PROTECTED]> * configure.ac (CLASSPATH_WITH_JAVAH): Only check i

Re: [cp-patches] FYI: automake 1.10 fixlet for plugin

2007-04-17 Thread Dalibor Topic
Mark Wielaard wrote: Hi Dalibor, On Fri, 2007-04-06 at 00:43 +0200, Dalibor Topic wrote: the attached patch adds AM_LDFLAGS to the plugin's LDFLAGS to compensate for automake 1.10 striping them away. Thanks for your various build system cleanup patches. It took some time to test the

[cp-patches] FYI: cleaned up JAVAH from lib makefile

2007-04-17 Thread Dalibor Topic
Hi all, the attached patch removes an unused variable from lib/Makefile.am cheers, dalibor topic 2007-04-17 Dalibor Topic <[EMAIL PROTECTED]> * lib/Makefile.am (JAVAH): Removed unused variable. Index: lib/Makefile.am ===

Re: [cp-patches] FYI: automake 1.10 fixlet for plugin

2007-04-17 Thread Mark Wielaard
Hi Dalibor, On Fri, 2007-04-06 at 00:43 +0200, Dalibor Topic wrote: > the attached patch adds AM_LDFLAGS to the plugin's LDFLAGS to > compensate for automake 1.10 striping them away. Thanks for your various build system cleanup patches. It took some time to test them all. But I added them (plu

Re: [cp-patches] Patch: FYI: minor fix in SSLSocketFactoryImpl

2007-04-17 Thread Mark Wielaard
On Mon, 2007-04-16 at 13:55 -0600, Tom Tromey wrote: > I'm checking this in to gcc svn trunk, RH 4.1 branch, and Classpath > cvs. After some additional testing (I installed cacert files on builder.classpath.org). Also added to the classpath-0.95 release branch now. > from Tom Tromey <[EMAIL PRO

[cp-patches] Always use underlyingSocket in SSLSocketImpl

2007-04-17 Thread Mark Wielaard
Hi, This adds a patch from Casey which was already applied and tested in libgcj. 2007-03-27 Casey Marshall <[EMAIL PROTECTED]> PR classpath/31302: * gnu/javax/net/ssl/provider/SSLSocketImpl.java (SSLSocketImpl): Always make a new socket. (bind, connect, getInetA