This patch fixes the problem I noticed, but there might be more of the kind... The patch only needs to be applied if the ipv6 was also applied. --- ucspi-tcp-0.88.orig/tcpserver.c 2003-04-06 11:01:41.000000000 +0200 +++ ucspi-tcp-0.88/tcpserver.c 2003-04-06 11:21:23.000000000 +0200 @@ -376,7 +376,7 @@ se = getservbyname(x,"tcp"); if (!se) strerr_die3x(111,FATAL,"unable to figure out port number for ",x); - uint16_unpack_big((char*)&se->s_port,&localport); + localport = ntohs (se->s_port); } if (!*argv) usage(); Reproducible: Always Steps to Reproduce: 1. 2. 3. Portage 2.0.47-r10 (default-ppc-1.4, gcc-3.2.2, glibc-2.3.2-r0) ================================================================= System uname: 2.4.20-ben5-int1 ppc GENTOO_MIRRORS="http://gentoo.oregonstate.edu/ http://www.ibiblio.org/pub/Linux/distributions/gentoo" CONFIG_PROTECT="/etc /var/qmail/control /usr/share/config /usr/kde/2/share/config /usr/kde/3/share/config" CONFIG_PROTECT_MASK="/etc/gconf /etc/env.d" PORTDIR="/usr/portage" DISTDIR="/usr/portage/distfiles" PKGDIR="/usr/portage/packages" PORTAGE_TMPDIR="/var/tmp" PORTDIR_OVERLAY="/home/portage" USE="dvd libwww mitshm esd mozilla ppc gdbm berkdb slang python crypt encode imap ipv6 maildir mbox ncurses oggvorbis oss pam pic readline ssl spell tcpd tetex xml2 zlib perl -cups -gif -gpm -gtk -imlib -arts -avi -java -jpeg -kde -gnome -motif -mpeg -mikmod -nls -opengl -png -qt -quicktime -sdl -svga -truetype -X -xv -xmms" COMPILER="gcc3" CHOST="powerpc-unknown-linux-gnu" CFLAGS="-O2 -pipe -fsigned-char" CXXFLAGS="-O2 -pipe -fsigned-char" ACCEPT_KEYWORDS="ppc ~ppc" MAKEOPTS="-j2" AUTOCLEAN="yes" SYNC="rsync://rsync.gentoo.org/gentoo-portage" FEATURES="ccache"
bouncing back to bug-wranglers as we don't have a ppc-dev running ipv6 to test this, maybe someone familiar with the ipv6 stuff or one of the other bigendian arches can take a look at this
guys help each other
I don't know how this ebuild works but I can test it on a hppa station which is bigendian. can you tell me how to reproduce this and explain the problem you've noticed ?
clearing up my bug tracking
I have been running .88-r5 on an ipv6 use-flag enabled machine. If you could pass me a test case I can give it a test.
Reassigning to ppc, keeping net-mail in CC
Marking as TEST-REQUEST then.
What's status on the test case?
Fixed in CVS, ucspi-tcp-0.88-r9