[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <a079022b-8434-a204-c9b4-696a0580fc8c@ozlabs.org>
Date: Thu, 11 May 2017 09:58:16 +0800
From: Jeremy Kerr <jk@...abs.org>
To: Christopher Bostic <cbostic@...ux.vnet.ibm.com>,
Joel Stanley <joel@....id.au>
Cc: Rob Herring <robh+dt@...nel.org>,
Mark Rutland <mark.rutland@....com>,
Russell King <linux@...linux.org.uk>, rostedt@...dmis.org,
mingo@...hat.com, Greg KH <gregkh@...uxfoundation.org>,
devicetree <devicetree@...r.kernel.org>,
linux-arm-kernel@...ts.infradead.org,
Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
Andrew Jeffery <andrew@...id.au>,
Alistair Popple <alistair@...ple.id.au>,
Benjamin Herrenschmidt <benh@...nel.crashing.org>,
"Edward A . James" <eajames@...ibm.com>
Subject: Re: [PATCH v6 19/23] drivers/fsi: Add GPIO based FSI master
Hi Chris,
> I don't think we'd want this per master. The lock is for the 'top'
> master issuing commands. Only the top master can initiate any
> transactions on the bus to any devices connected downstream. Downstream
> masters such as hub masters, etc... cannot initiate a command.
I think what Joel meant there was that we have it per *GPIO* master; if
there are two GPIO masters on a system, there's no need to provide
mutual exclusion to each (separate) set of GPIOs.
To implement this, we'd just move the lock into struct fsi_master_gpio.
Cheers,
Jeremy
Powered by blists - more mailing lists