Date: Sun, 04 Oct 1998 12:06:50 -0700 From: Manfred Antar <mantar@netcom.com> To: Mike Smith <mike@smith.net.au> Cc: Mike Smith <mike@smith.net.au>, Robert Nordier <rnordier@nordier.com>, osa@etrust.ru, current@FreeBSD.ORG Subject: Re: CRUSH after recompile kernel... Message-ID: <4.1.19981004120024.00a3dd10@192.168.0.1> In-Reply-To: <199810040115.SAA00490@dingo.cdrom.com> References: <Your message of "Sat, 03 Oct 1998 16:45:44 PDT." <4.1.19981003164142.00a422a0@192.168.0.1>
next in thread | previous in thread | raw e-mail | index | archive | help
At 06:15 PM 10/3/98 -0700, Mike Smith wrote: >> > >> Just out of curiosity I built an elf kernel without the debugger and >> tried the /boot/loader. >> It worked fine. >> so the problem had something to do with the debugger code. >> I'll test it more Sunday. > >Ok. I have a pretty good idea as to what might be going on now; >thanks for the input. If you want to do something really helpful, >print the values of bootinfo.bi_symtab and bootinfo.bi_esymtab in the >non-debugger kernel, and then load it with btxboot and the loader. > >Put a printf somewhere in machdep.c, eg. around about where it tests to >call userconfig(). > >My guess is that there's something wrong with the symbol table >information coming in, and it causes ddb to explode. > > I'm not sure how to do the above. But I have been following this thread and put esym = ssym = 0; in line 81 of /sys/i386/libi386/elf_freebsd.c and I can now boot a current elf kernel with the standard bootblocks with /boot/loader. Manfred ============================== || mantar@netcom.com || || pozo@infinex.com || || Ph. (415) 681-6235 || ============================== To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4.1.19981004120024.00a3dd10>