• Open Menu Close Menu
  • Apple
  • Shopping Bag
  • Apple
  • Mac
  • iPad
  • iPhone
  • Watch
  • TV
  • Music
  • Support
  • Search apple.com
  • Shopping Bag

Lists

Open Menu Close Menu
  • Terms and Conditions
  • Lists hosted on this site
  • Email the Postmaster
  • Tips for posting to public mailing lists
Re: cdevsw close never called
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: cdevsw close never called


  • Subject: Re: cdevsw close never called
  • From: mm w <email@hidden>
  • Date: Fri, 6 Mar 2009 13:06:56 -0800

Ok Andrew, I misunderstood in the first place,

it's because someone is knocking the door when you want to remove it

kstart:
  create dev

kstop:
  remove dev failed because something is using it

kstart again failed device existing

is that your case?

On Fri, Mar 6, 2009 at 12:41 PM, Andrew Gallatin <email@hidden> wrote:
> mm w wrote:
>>
>> Hello Andrew,
>>
>> a simple question,
>>
>> A has opened /dev/file000
>> B for obscur reason wants to access  /dev/file000 (could create an alias)
>
> B is a process out of my control, looking for something to open, and
> is iterating over file0..N, attempting to open each one until
> he succeeds or runs out of device nodes to try.
> Heck, it is a potentially malicious program opening and closing each
> device node in a tight loop, for all I know.  I've got to deal with
> whatever happens in the wild.
>
>> tuple processes
>>
>> if (proc (A...B C...) hold /dev/file000 || /dev/file000 on job)
>>  return copy refA;
>> else
>>  ... open stuff
>>
>> I think it's at your design level that you have to deal with that
>> you have to create your own stack to know what's going on, but maybe
>> there is a prebuilt
>> stuff to handle it I don't know, but I almost sure not
>
> I'm not sure what you mean.  I don't control anything in userspace.
> I only control the driver side.   I'd love to have a factory device like
> Solaris, which sort of does what I think you're trying to describe.
> But I've got to work with what I'm given.
>
> The underlying statement remains:  It really looks like there might
> be a bug in Darwin's implementation of last close semantics for
> device nodes, which can lead to a driver's close never getting
> called.
>
> Drew
>
>



--
-mmw
 _______________________________________________
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

References: 
 >cdevsw close never called (From: Andrew Gallatin <email@hidden>)
 >Re: cdevsw close never called (From: mm w <email@hidden>)
 >Re: cdevsw close never called (From: Andrew Gallatin <email@hidden>)

  • Prev by Date: Re: cdevsw close never called
  • Next by Date: Re: cdevsw close never called
  • Previous by thread: Re: cdevsw close never called
  • Next by thread: Re: cdevsw close never called
  • Index(es):
    • Date
    • Thread