Joerg Schilling wrote:
> That is the reason for trying to use a generic name for a utility that
> only works with a limited subset of SCSI devices?

The goal of the project is to allow this to be used by things other than 
SCSI.

Quoting from the case material:

"This fasttrack addresses that request and in the process extends 
fwflash(1M) to also provide a generic, pluggable interface for other 
devices for which a firmware update mechanism might be required in the 
future."

The keywords here are "generic" and "other devices".

Then later on in the material:

"The existing InfiniBand functionality will be re-factored to work 
within the proposed plugin framework."

So not only does the case provide a framework for use by things other 
than SCSI it actually mentions an non SCSI consumer.

What is it about the above that isn't clear and makes you think that 
fwflash(1M) applies only to a subset of SCSI ?

I should also point out that the case material also says that 
/usr/sbin/fwflash already exits and was approved and introduced by 
PSARC/2005/126.

So if anything I believe this case is actually the one that addresses 
your concern that fwflash is a generic name and your issue would have 
been relevant to PSARC/2005/126.  2005/126 was originally proposed as 
ibflash(1M) because it was to be relevant only for InfiniBand, however 
the then PSARC committee asked the project to to prepare for having a 
generic interface for firmware flashing and rename it to fwflash(1M), 
*this* case (2008/151) is that future one that makes it generic.

-- 
Darren J Moffat

Reply via email to