[cgl_discussion] devfs & CGL

Steven Dake sdake at mvista.com
Thu Feb 6 16:45:52 PST 2003


In the current implementation for ga mapping (mapping a scsi device for 
a disk in a certain slot/chassis), we do use devfs.  I've come to my 
senses and I am going to change this in our CGE 3.1 product to make 
enumeration of those devices take place in userspace without the need 
for devfs.  Also, the work I have done on our automatic storage 
multipathing enumeration occurs in userspace, following this philosophy. 
 The work Mark Bellon is doing for device enumeration (the next-gen 
MontaVista HDI work) is based in userspace and does not depend on devfs 
at all.

devfs is an interesting idea, but controlling naming/permission policy 
in kernel has its share of problems.  The only downside to controlling 
permission/naming in userspace is it is currently done in a static form 
via MAKEDEV which is inapproriate for a variety of reasons.  

This is why a next-generation HDI solution is needed to solve the naming 
problem in userspace, not in the kernel (with some kernel help to tell 
userspace of insertions/extractions that the hardware may be aware of).

Thanks
-steve

Lynch, Rusty wrote:

>IIRC, the Monta Visa implementation of HDI utilized devfs, but that is the
>only dependency that I know of.
>
>I have been assuming devfs was dead for a long time now.
>
>    -rustyl
>
>-----Original Message-----
>From: Patrick Mochel [mailto:mochel at osdl.org] 
>Sent: Thursday, February 06, 2003 6:43 AM
>To: Mika Kukkonen
>Cc: cgl_discussion at osdl.org
>Subject: Re: [cgl_discussion] devfs & CGL
>
>
>On 6 Feb 2003, Mika Kukkonen wrote:
>
>  
>
>>I am starting to get worried about devfs in 2.6; see for example this:
>>http://marc.theaimsgroup.com/?l=linux-kernel&m=104456842930178&w=2
>>    
>>
>
>This should make you very happy. :)
>
>  
>
>>Our specs don't explicitly name devfs (I searched) anywhere, but how 
>>about POC; is there and how much of our features are dependent on devfs?
>>I thought some of the hotplugging and maybe scalability requirements
>>needed it?
>>    
>>
>
>They absolutely should NOT. There are no kernel developers that I know of 
>-- besides Richard -- that are using devfs. In fact, they staunchly refuse 
>to, especially those working on hotplug and device naming issues. 
>
>	-pat
>
>
>_______________________________________________
>cgl_discussion mailing list
>cgl_discussion at lists.osdl.org
>http://lists.osdl.org/mailman/listinfo/cgl_discussion
>
>
>
>  
>




More information about the cgl_discussion mailing list