On Friday,  5 April 2002 at 14:18:38 -0800, Doug White wrote:
> On Fri, 5 Apr 2002, Alessandro de Manzano wrote:
>
>> On Fri, Apr 05, 2002 at 12:00:05PM -0800, Alfred Perlstein wrote:
>>
>> Wow, thanks for the super-fast answer! :))
>>
>>
>>>> on my production servers' kernel so in the very rare case of crash I'll
>>>> got a crash dump ( I'ld use also options DDB_UNATTENDED) and could
>>>> immediately have a backtrace report.
>>>>
>>>> ..Am I crazy ? :-))
>>>
>>> I don't think you'll notice a difference for most stuff, this is how
>>
>> does the "-g" option (GCC option I guess) disable the "-O" optimizing
>> option ?
>> If "-g" simply attach the symbols and similar debug info to the
>> executable I guess the kernel should not be slower, but I don't know
>> GCC very well...
>
> Note that the kernel binary with debugging symbols is left in
> /sys/compile/MYKERNEL/kernel.debug while the actual kernel is stripped
> before installation into /kernel.
>
> If the debugging kernel was actually loaded it would be gigantic :)

No, since the transition to ELF, none of the debugging information
gets loaded into core.  Try it.

> This is all detailed in the Handbook section on kernel debugging,
> btw.

Hmm, that needs to be fixed, then.

Greg
--
See complete headers for address and phone numbers

To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-hackers" in the body of the message

Reply via email to