Re: Panics and Logging Messages
Re: Panics and Logging Messages
- Subject: Re: Panics and Logging Messages
- From: Carlos Calderon <email@hidden>
- Date: Tue, 23 Feb 2010 13:25:35 -0800
Hi Duane,
> Is that are reasonable expectation? That the functions that are
> printing in the middle of the kernel panic are likely not the cause of
> the kernel panic itself?
>
> My concern and reason for the question is what level of buffering may
> be involved to delay the printing of the message?
Yes, that's a reasonable expectation (different processors/cores logging). This is because FireWireKPrintf is synchronous in that it will put the log onto the wire before returning from kprintf(). Additionally, kprintf() is kernel lock synchronized, so as to order all log messages (and make them readable). So, there isn't be any sort of delay "buffering," simply that of transmission, but that should be ordered. Of course, until a processor is halted, it will diligently continue to do work, even when others are in the process of giving up.
// carlos
On Feb 23, 2010, at 11:57 AM, Duane Murphy wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA256
>
> I'm packing a couple of questions into this message.
>
> I'm trying to track down the cause of an elusive kernel panic. I have
> the target machine connected over firewire using fwkpfv. I've adjusted
> the debugging strings in my kext to print using kprintf.
>
> Question 1
>
> I'm trying to determine which functions may have triggered the kernel
> panic. When the kernel panic backtrace is printed to fwkpfv, there are
> messages interspersed with the kernel panic. My expectation is that
> while the back trace is being calculated, the other processor is still
> running. These other messages are from that processor while the kernel
> panic is being processed. Then the system is halted.
>
> Is that are reasonable expectation? That the functions that are
> printing in the middle of the kernel panic are likely not the cause of
> the kernel panic itself?
>
> My concern and reason for the question is what level of buffering may
> be involved to delay the printing of the message?
>
> Question 2
>
> I've noticed this message after of the kernel panic logged to fwkpfv:
>
> Attempting to commit panic log to NVRAM
> EFI: couldn't save panic info (err = 80000009)
>
> I've never noticed the second message before (EFI: etc). Can someone
> translate that into English?
>
> Thank you for your insight,
>
> ...Duane
>
>
>
>
> -----BEGIN PGP SIGNATURE-----
> Version: PGP Desktop 9.10.0 (Build 500)
> Charset: US-ASCII
>
> wsBVAwUBS4QzeErg9acQ4r2CAQhPGwf8DXCDttP0k7qXqt1emXH3bFApNqOYF9mR
> mzDLsO1eci+scjzwE9Fybuv5oUoJjobaZWSFcs+tYR1nO9eYqp/l0VUc/cyigDji
> Sk6eZeg5xMmUTWdhkcHvSuCZVAoxEWbJhhoLLnKoOddSDt1k8f8NlyyQNa7FrKXH
> Em27w6Qe22Tuvo5nIw8dW/zrNu7A7tfsTQ25kVhj0kHLXCKkpieJncbAOZ6tnnzI
> JwsOE57oHLjAd+nr9SVANRRhrVxrddLvwXOhycfqoujmIEBEz1MKms70MO92GfrK
> Jkdwa6ExQHGhdMK8KrI6o2kStq+x9RBtbRs7A9PHblB/8KQwR/ZEHw==
> =N1iq
> -----END PGP SIGNATURE-----
> _______________________________________________
> Do not post admin requests to the list. They will be ignored.
> Darwin-kernel mailing list (email@hidden)
> Help/Unsubscribe/Update your Subscription:
>
> This email sent to email@hidden
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Darwin-kernel mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden