RE: [PATCH 7/7] CCISS: run through Lindent

[Cameron, Steve]

Sometimes there are new devices that need device specific
handling (e.g. max number of commands, no. of SG elements, etc
are bigger or smaller) and an old driver on a new board
might do bad things.  Currently, you can run an old driver
on a new board by explictly telling it to make the attempt
via a command line option, iirc, but having it do it always 
and without user prodding it to do so is less safe.

-- steve


-----Original Message-----
From: Helgaas, Bjorn
Sent: Thu 6/15/2006 10:42 AM
To: Miller, Mike (OS Dev)
Cc: ISS StorageDev; linux-kernel@xxxxxxxxxxxxxxx; Andrew Morton
Subject: Re: [PATCH 7/7] CCISS: run through Lindent
 
On Thursday 15 June 2006 08:49, Miller, Mike (OS Dev) wrote
> > -----Original Message-----
> > From: Helgaas, Bjorn 
> > Sent: Wednesday, June 14, 2006 6:14 PM
> > To: Miller, Mike (OS Dev)
> > Cc: ISS StorageDev; linux-kernel@xxxxxxxxxxxxxxx; Andrew Morton
> > Subject: [PATCH 7/7] CCISS: run through Lindent
> > 
> > cciss is full of inconsistent style ("for (" vs. "for(", lines that
> > end with whitespace, lines beginning with a mix of spaces & 
> > tabs, etc).
> > 
> > This patch changes only whitespace.
> > 
> > Signed-off-by: Bjorn Helgaas <bjorn.helgaas@xxxxxx>
> 
> I agree that we had lots of whitespace garbage and inconsistent styles
> in the driver but I'm not sure I like all the indentation being removed
> from the product table. It makes it a bit harder to read, IMO.

If you are OK with the patch other than the product table, how about
if you apply the patch as-is, and I post a follow-on patch to fix the
indentation?

I'm contemplating more than just a white-space change, and it'd
probably be better to keep the "indent" patch to be white-space
changes only.

The cciss_pci_device_id[] table lists a bunch of subvendor and
subdevice IDs.  Is there any reason to check those sub-IDs
explicitly?  In other words, we currently list:

        {PCI_VENDOR_ID_COMPAQ, PCI_DEVICE_ID_COMPAQ_CISS,
         0x0E11, 0x4070, 0, 0, 0},

Could we replace that with:

        {PCI_VENDOR_ID_COMPAQ, PCI_DEVICE_ID_COMPAQ_CISS,
         PCI_ANY_ID, PCI_ANY_ID, 0, 0, 0},

This would potentially make the driver claim additional devices.
But do COMPAQ/CISS devices with sub-IDs other than the listed ones
really exist anyway?

Bjorn

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/