From mboxrd@z Thu Jan 1 00:00:00 1970 Message-Id: <200005120507.HAA02656@denx.local.net> To: "Shuangjun Zhu" cc: "Lucinda Schafer" , linuxppc-embedded@lists.linuxppc.org Subject: Re: Help about the libc-2.1.1 segmentation fault From: Wolfgang Denk Mime-version: 1.0 Content-type: text/plain; charset=ISO-8859-1 In-reply-to: Your message of "Fri, 12 May 2000 09:05:32 +0800." <001e01bfbbae$2d0e03a0$69f001dc@sjzhu.sps.mot.com> Date: Fri, 12 May 2000 07:07:52 +0200 Sender: owner-linuxppc-embedded@lists.linuxppc.org List-Id: In message <001e01bfbbae$2d0e03a0$69f001dc@sjzhu.sps.mot.com> you wrote: > > Firstly, thank you for your kind help. > In fact, my root file system is libc-1.99.so. It assures me to begin to > work. > I build another file system to test the libc-2.1.1.so. It likes > ---- lib/libc-2.1.1.so > | > /mnt ---------- bin -------- hello.dynamic, which built with libc-2.1.1.so > library > | |--- hello.static, which built with static library > | > ------ sbin/ Combining libc5 (1.99) and libc6 (2.1.x) withing one system is not easily done. The only way I got this working is _strict_ separation of both environments; separate directories is not enough. Try the following: install a STATICALLY linked shell (for instance, sash) as /mnt/bin/sh; then run "chroot /mnt /bin/sh", so that you have a staicallly linked shell running in a chrooted environment where ONLY libc-2.1.x files (binaries, libraries, loader) are visible. This way I've been testing my 2.1.x (x=1, 2 and 3) libs and binaries for a while. Works fine for me, hope it helps. Wolfgang Denk -- Software Engineering: Embedded and Realtime Systems, Embedded Linux Phone: (+49)-8142-4596-87 Fax: (+49)-8142-4596-88 Email: wd@denx.de C++ was an interesting and valuable experiment, but we've learned its lessons and it's time to move on. - Peter Curran in ** Sent via the linuxppc-embedded mail list. See http://lists.linuxppc.org/