Re: [PATCH] spi/bcm63xx: fix multi transfer messages

[Florian Fainelli <florian-p3rKhJxN3npAfugRpC6u6w@public.gmane.org>]

On Thursday 15 November 2012 10:15:08 Mark Brown wrote:
> On Thu, Nov 15, 2012 at 12:33:31AM +0100, Jonas Gorski wrote:
> > On 15 November 2012 00:17, Mark Brown
> 
> > > This is only going to work around a relatively small subset of cases so
> > > it doesn't seem worth rushing in.  The normal fix for such issues is to
> > > control /CS as a GPIO.
> 
> > Well, one of the "small" subsets is SPI attached flash, those need
> > write then read or write then write with CS asserted for most
> > operations, too. And I'm not sure if controlling CS as a GPIO is an
> 
> This is the case for essentially all devices with registers too.  If it
> were always fixing the issue (eg, by allocating a buffer if the one
> supplied isn't suitable) that'd be fine but instead it's going to work
> some but not all of the time which seems non-ideal

Devices that we care about on bcm63xx, which are connected to these built-in
chip-selects are:
- SPI flashes
- SPI Ethernet switches (register based)
- SPI SLIC/SLAC (register based as well)

I assume any hardware design which needs specific CS treatment would wire it
to a spare GPIO to fully control its assertion.

Would you accept this patch provided that Jonas updates his commit log with
this description of devices we support?
--
Florian


------------------------------------------------------------------------------
Monitor your physical, virtual and cloud infrastructure from a single
web console. Get in-depth insight into apps, servers, databases, vmware,
SAP, cloud infrastructure, etc. Download 30-day Free Trial.
Pricing starts from $795 for 25 servers or applications!
http://p.sf.net/sfu/zoho_dev2dev_nov