Skip site navigation (1)Skip section navigation (2)
Date:      Sun, 17 Feb 2019 23:19:16 +0200
From:      Konstantin Belousov <kostikbel@gmail.com>
To:        "Herbert J. Skuhra" <herbert@gojira.at>
Cc:        src-committers@freebsd.org, svn-src-all@freebsd.org, svn-src-stable@freebsd.org, svn-src-stable-12@freebsd.org
Subject:   Re: svn commit: r344157 - stable/12/lib/libc/x86/sys
Message-ID:  <20190217211916.GD2420@kib.kiev.ua>
In-Reply-To: <87d0nqazuq.wl-herbert@gojira.at>
References:  <201902151133.x1FBXmHU060955@repo.freebsd.org> <87d0nqazuq.wl-herbert@gojira.at>

next in thread | previous in thread | raw e-mail | index | archive | help
On Sun, Feb 17, 2019 at 10:07:09PM +0100, Herbert J. Skuhra wrote:
> On Fri, 15 Feb 2019 12:33:48 +0100, Konstantin Belousov wrote:
> > 
> > Author: kib
> > Date: Fri Feb 15 11:33:48 2019
> > New Revision: 344157
> > URL: https://svnweb.freebsd.org/changeset/base/344157
> > 
> > Log:
> >   MFC r343855, r343859:
> >   Use ifunc to select the barrier instruction for RDTSC.
> > 
> > Modified:
> >   stable/12/lib/libc/x86/sys/__vdso_gettc.c
> > Directory Properties:
> >   stable/12/   (props changed)
> 
> After upgrading my Soekris Net 6501-70 (amd64) to this revision the
> system is unusable. Basically all programs produce a "Segmentation
> fault (core dumped)":
> 
> pid 856 (ls), jid 0, uid 1001: exited on signal 11 (core dumped)
> pid 857 (ls), jid 0, uid 1001: exited on signal 11 (core dumped)
> pid 858 (bsdtar), jid 0, uid 0: exited on signal 11 (core dumped)
> pid 859 (sh), jid 0, uid 0: exited on signal 11 (core dumped)
> pid 876 (dmesg), jid 0, uid 1001: exited on signal 11 (core dumped)
> pid 877 (su), jid 0, uid 0: exited on signal 11
> pid 880 (more), jid 0, uid 1001: exited on signal 11 (core dumped)
> pid 885 (sudo), jid 0, uid 0: exited on signal 11

What do you have in your make.conf and src.conf ?
Can you show me CPU identification lines from dmesg for a verbose boot ?

I suspect that you can take /lib/libc.so.7 from 12.0 RELEASE and restore
the system.  If it helps, please move your existing libc.so.7 into some
directory, reproduce the problem with 'LD_LIBRARY_PATH=<path to dir with
bad libc.so.7> /bin/ls' and load the core into gdb.  I want to see
the backtrace for start.



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20190217211916.GD2420>