RE: kev_vendor_code_find() call causes kernel panic...
RE: kev_vendor_code_find() call causes kernel panic...
- Subject: RE: kev_vendor_code_find() call causes kernel panic...
- From: "Alexander Klyuev" <email@hidden>
- Date: Fri, 13 Aug 2010 18:24:25 -0400
Hi Vincent,
Thanks, of course, I'm gonna do that.
Cheers,
- Alexander
-----Original Message-----
From: Vincent Lubet [mailto:email@hidden]
Sent: August 13, 2010 6:18 PM
To: Alexander Klyuev
Cc: email@hidden
Subject: Re: kev_vendor_code_find() call causes kernel panic...
Alexander,
To report a such bug you should open a problem report at
https://bugreport.apple.com/
Thanks,
Vincent
On Aug 13, 2010, at 2:38 PM, Alexander Klyuev wrote:
> Hi All,
>
> I already wrote about this issue: kev_vendor_code_find() call causes
> kernel panic when network stack is not entirely up (net_str_id_lock
> object is still uninitialized (NULL).
>
> Here is simplified call sequence:
> --------------------------------------
> kev_vendor_code_find(...)
> net_str_id_find_internal(...)
> lck_mtx_lock(net_str_id_lock) <= where net_str_id_lock value is
NULL.
> ... kernel panic
> -------------------------------------
>
> However IMHO such a function MUST NEVER cause kernel crash whenever it
> is called because it's possible to check whether a pointer argument
> ("net_str_id_lock" in this case) is NULL before call to
> lck_mtx_lock()
> and return an error code not causing kernel panic. So that is
> definitely a bug.
>
> Am I missing something?
>
> Thanks for further responses.
_______________________________________________
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