On 04/10/12(Thu) 22:44, matt wrote: > I assume I or my hardware is doing something stupid and obvious. > I've been trying to successfully build OpenBSD for the first time on > a 2002 G4 (Mirror Drive Door) dual 1ghz. The RAM is new, and > slightly faster than the bus speed demands (pc3200 instead of > pc2700). Processer is a PowerPC 7455B. > > I am following the guide for building OpenBSD from source to the > letter (Although a misread is not out of the question!). First I use > cvs to checkout the sources for -CURRENT (Including Xenocara and > Ports, but I've never gotten that far). The kernel builds fine, and > I install it and reboot. On an earlier attempt, I verified it was > installing properly by changing WS_KERNEL_BG. > > I am starting from the latest snapshot and sources, and have tried > multiple cvs revisions and gone through two latest snapshots. > > Filebin link for the 7.3mb script log of my build attempt is > here:http://filebin.ca/I1aLqmsKata/buildlog > > I did use -j4, but the problem is the same without it. The result: > > --- > > install -c -o root -g bin -m 444 /usr/src/bin/md5/sha256.1 > /usr/share/man/man1/sha256.1 > *** Signal 6 in target maninstall > > Stop in /usr/src/bin/md5: > Received signal 6 (line 71 of <bsd.man.mk>, > target maninstall: @l=/usr/share/man/man1/cksum.1; > t=/usr/share/man/man1/sum.1; echo $t -\> $l; rm -f $t; ln $l $t;) > *** Error code 2 in target realinstall > > Stop in /usr/src/bin: > Exit status 2 (line 48 of <bsd.subdir.mk>, target realinstall) > *** Error code 2 in target realinstall > > Stop in /usr/src: > Exit status 2 (line 48 of <bsd.subdir.mk>, target realinstall) > *** Error code 2 in target build > > Stop in /usr/src: > Exit status 2 (line 85 of Makefile, target build) > > --- > > All executables installed before that give "Abort trap" signal 6 > results. The system can barely be used, and once logout occurs, the > system is hosed. No login is possible, nothing can be executed, > single user mode does not work. > > It looks like I am somehow getting broken executables and when ln is > called after its installation, it aborts. I assume that line is the > first time an installed executable is called, but I'm not sure why > my executables are aborting. > > It's still logged in if anyone has any ideas other than rm -rf /, > which aborts :)
I guess this is related to the fact that you are using macppc snapshots, they are PIE but not all the share/mk bits are committed until someone fix PIE for socppc. Until then, stick to the snaps ;) M.
