From mboxrd@z Thu Jan 1 00:00:00 1970 From: Russell King - ARM Linux Subject: Re: [RFT/PATCH] serial: omap: prevent resume if device is not suspended. Date: Tue, 25 Sep 2012 12:15:06 +0100 Message-ID: <20120925111506.GO31374@n2100.arm.linux.org.uk> References: <1347972050-3509-1-git-send-email-sourav.poddar@ti.com> <20120925083029.GG31374@n2100.arm.linux.org.uk> <20120925083118.GI9137@arwen.pp.htv.fi> <20120925091228.GI31374@n2100.arm.linux.org.uk> <20120925091112.GK9137@arwen.pp.htv.fi> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: <20120925091112.GK9137@arwen.pp.htv.fi> Sender: linux-kernel-owner@vger.kernel.org To: Felipe Balbi Cc: "Poddar, Sourav" , gregkh@linuxfoundation.org, khilman@ti.com, paul@pwsan.com, tony@atomide.com, linux-kernel@vger.kernel.org, santosh.shilimkar@ti.com, linux-serial@vger.kernel.org, linux-omap@vger.kernel.org, linux-arm-kernel@lists.infradead.org, alan@linux.intel.com List-Id: linux-serial@vger.kernel.org On Tue, Sep 25, 2012 at 12:11:14PM +0300, Felipe Balbi wrote: > that's most likely, of course. But should we cause a regression to > beagleboard XM because of that ? Also, if you look into chapter 9 of the > runtime_pm documentation, starting on line 822 you'll see documentation > suggests the use of mystruct->is_suspended flag. BTW, I'll point out a fatal flaw in your justification above. If you read the entire example, you'll see that the is_suspended flag is _not_ used to prevent resumes without suspends, but is used as a flag to control whether the driver processes requests or not. That's entirely functionally different from using a "is_suspended" flag in the way you mention above. Section 5 is quite clear about the requirements at initialization time for runtime PM, and nothing in section 9 contradicts that, and the is_suspended flag in that example has nothing to do with any of this.