cvs commit: src/sys/alpha/alpha support.s src/sys/i386/i386 swtch.s src/sys/kern kern_shutdown.c src/sys/sys systm.h

Greg 'groggy' Lehey grog at FreeBSD.org
Tue Jan 20 16:41:13 PST 2004


On Tuesday, 20 January 2004 at 16:00:19 -0500, John Baldwin wrote:
> On Tuesday 20 January 2004 03:39 pm, Poul-Henning Kamp wrote:
>> In message <200401201445.24897.jhb at FreeBSD.org>, John Baldwin writes:
>>> On Tuesday 20 January 2004 01:23 pm, Poul-Henning Kamp wrote:
>>>> In message <200401201234.45472.jhb at FreeBSD.org>, John Baldwin writes:
>>>>> On Monday 19 January 2004 04:27 pm, Poul-Henning Kamp wrote:
>>>>>>   Log:
>>>>>>   Add linenumber and source filename to panic(9) output.
>>
>> May I kindly remind you all that even if one person doesn't find
>> that it adds anything to his already trained skills as a debugger,
>> it does make life easier for other less gifted people in our project,
>> or God forbid: actual users who may try to report a problem.
>
> By adding even more clutter to a panic message (esp. the fatal page faults
> that happen more often than any other panic) you make it even harder for any
> debugger to glean out the useful information from all of the rest.

OK, I haven't seen this output, but if it's just line number and
source file name, I think it's a Good Idea.

>> And no, this does not solve the mid-east crisis, but it is still
>> an improvement, even for me:  If I save 1 minute because I do not
>> have to hunt for the panic in the first place, then that is one
>> minute more I can spend on the code.
>
> It is not an improvement in all cases and most of the people in this
> thread have opposed this.

I'm very much in favour.

> The only response you got on the mailing list to your post was a
> "please do not commit" from Bruce and you went ahead and committed
> anyway.  Do all of our opinions just not count when the Almighty
> Poul-Henning has a patch he wants to commit?

Certainly committing without consensus is wrong.  I've been out of
touch for the last week due to a conference, so I didn't see the
discussion, but that's a separate issue.  I'll concentrate on the
technical issue in this message.

>> So if this makes your eyes water, I suggest you comment it out in
>> your local source tree and pop in on the next meeting in your
>> local user-group for some much needed perspective.
>
> I have been very busy helping people with bug reports including
> closing PR's etc.  I'm not hiding in my white tower cursing users as
> you seem to imply, so you can lay off that lame argument.  In my
> real world experience, the panic message is enough to find where the
> problem is.

OK, fine for you.  And yes, I can grep too.  But finding bugs is hard
enough; any help can only be an improvement.

> Several of them already include far more relevant file and line
> numbers anyways.

These are obviously not the newbies on FreeBSD-questions who say "My
machine panicked.  What went wrong?".

On Tuesday, 20 January 2004 at 22:33:18 +0100, Poul-Henning Kamp wrote:
> In message <200401201600.19855.jhb at FreeBSD.org>, John Baldwin writes:
>
> I suggest you and everybody else calm down and let dust settle for
> a couple of days, maybe other people should have a chance to say
> their opinion.  If there is a clear concensus that this is bad
> (maybe somebody neutral should take a count ?), then we'll back it
> out.

Sounds reasonable to me.

Greg
--
See complete headers for address and phone numbers.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 187 bytes
Desc: not available
Url : http://lists.freebsd.org/pipermail/cvs-src/attachments/20040121/675835c7/attachment.bin


More information about the cvs-src mailing list