From: Roger Quadros Subject: Re: [PATCH 1/2] phy: ti-pipe3: Disable clocks on system suspend Date: Mon, 12 Jan 2015 11:21:51 +0200 Message-ID: <54B3922F.3010501@ti.com> References: <1418990720-3638-1-git-send-email-rogerq@ti.com> <1418990720-3638-2-git-send-email-rogerq@ti.com> <54AFDE4A.2000205@ti.com> Mime-Version: 1.0 Content-Type: text/plain; charset="windows-1252" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <54AFDE4A.2000205@ti.com> Sender: linux-kernel-owner@vger.kernel.org To: Kishon Vijay Abraham I , tony@atomide.com Cc: nm@ti.com, balbi@ti.com, george.cherian@ti.com, nsekhar@ti.com, linux-omap@vger.kernel.org, linux-kernel@vger.kernel.org List-Id: linux-omap@vger.kernel.org Kishon, On 09/01/15 15:57, Kishon Vijay Abraham I wrote: > Hi Roger, > > On Friday 19 December 2014 05:35 PM, Roger Quadros wrote: >> On system suspend, the runtime_suspend() driver hook doesn't get >> called and so the clocks are not disabled in the driver. >> This causes the L3INIT_960M_GFCLK and L3INIT_480M_GFCLK to remain >> active on the DRA7 platform while in system suspend. >> >> Add suspend/resume hooks to the driver. >> In case of pcie-phy, the runtime_suspend hook gets called after > > This contradicts with the first line of your commit message. Is pcie-phy driver > is an exception? Yes in the pcie-phy case it behaves differently. I'll rewrite the message. cheers, -roger > > Thanks > Kishon > >> the suspend hook so we introduce a flag phy->enabled to keep >> track if our clocks are enabled or not to prevent multiple >> enable/disables. >> >> Move enabling/disabling clock code into helper functions. >> >> Reported-by: Nishant Menon >> Signed-off-by: Roger Quadros