kernel ddb help

2001-09-05 Thread Zhihui Zhang
I know gdb can source stepping the kernel. But without two machines, you can not do it. Now I have only one machine and the system panic: db trace bqrelse(cxxx, cxxx, cxxx, c, cxxx) at bqrelse+0x25 is there a way to use these addresses to figure out which line or lines of source are

RE: kernel ddb help

2001-09-05 Thread Alexander N. Kabaev
You can use gdb on the dump file or even on live kernel after reboot to figure out exactly what the problem was. Use gdb -k ./kernel.debug /dev/mem or gdb -k ./kernel.debug /var/crash/vmcore.num On 05-Sep-2001 Zhihui Zhang wrote: I know gdb can source stepping the kernel. But without two

Re: kernel ddb help

2001-09-05 Thread Julian Elischer
you can gdb -k mykernel /dev/mem and do list bqrelse+0x25 (I think) alternatively, in ddb you can do: x/iii bqrelse and work out what is wrong by reading the machine instructions WHen I have one machine I usually debug by running the new kernel within a VMWARE virtual

Re: kernel ddb help

2001-09-05 Thread Zhihui Zhang
Your snapshot is cool and I have found your old mail regarding VMWARE. One more question: Is X-windows needed for this stuff? Thanks, -Zhihui To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-hackers in the body of the message