svnlite segfaults a lot

Tim Kientzle tim at kientzle.com
Thu Jul 3 01:00:11 UTC 2014


On Jul 2, 2014, at 1:20 AM, Andrew Turner <andrew at fubar.geek.nz> wrote:

> On Tue, 1 Jul 2014 10:30:46 +0100 (BST)
> Anton Shterenlikht <mexas at bris.ac.uk> wrote:
> 
>> # uname -a
>> FreeBSD raspberry-pi 11.0-CURRENT FreeBSD 11.0-CURRENT #0 r267801:
>> Tue Jun 24 11:03:28 UTC 2014
>> root at grind.freebsd.org:/usr/obj/arm.armv6/usr/src/sys/RPI-B  arm
>> 
>> I'm trying to pull the ports tree to rasp. pi model B.
>> svnlite up (or co) segfaults after pulling 1-5 MB.
>> Is this expected?
>> 
>> I'm powering via a bench power supply, so I can
>> monitor the current. It is only about 400mA,
>> occasionally raising to maybe 450mA.
>> So my earlier suspicion, that an inadequate
>> power was to blame, was wrong.
>> 
>> Anybody else seeing svnlite segfaults?
> 
> Yes, however I never tracked it down. I found svn from ports to work as
> expected.

It wasn’t this way a month or two ago when I used it last.

But I tried just now and it’s pretty bad.  Trying to check out
http://svn.freebsd.org/base/head segfaults after a thousand or so
items (not always at the same point).   Of course, gdb doesn’t
seem to be able to make any sense of the core file:

$ gdb /usr/bin/svnlite
...
(gdb) core svnlite.core
… loading many symbols ...
(gdb) bt
#0  0x0008dbac in ?? ()
(gdb)

…

Trying to “run update” from inside GDB immediately hits
SIGILL in _rtld_get_stack_prot()

Tim




More information about the freebsd-arm mailing list