From mboxrd@z Thu Jan 1 00:00:00 1970 From: Tony Lindgren Subject: Re: [PATCH] wlcore: sdio: Fixup power on/off sequence Date: Wed, 16 Jan 2019 07:40:29 -0800 Message-ID: <20190116154029.GO5544@atomide.com> References: <20190115150433.12558-1-ulf.hansson@linaro.org> <20190115185509.GJ5544@atomide.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: Sender: linux-kernel-owner@vger.kernel.org To: Ulf Hansson Cc: Kalle Valo , Eyal Reizer , linux-wireless@vger.kernel.org, Ricardo Salveti , Kishon Vijay Abraham I , Anders Roxell , John Stultz , Jan Kiszka , Linux Kernel Mailing List , linux-omap List-Id: linux-omap@vger.kernel.org * Ulf Hansson [190116 11:44]: > On Tue, 15 Jan 2019 at 19:55, Tony Lindgren wrote: > > # while [ 1 ]; do ifconfig wlan0 down; usleep 20000; \ > > ifconfig wlan0 up; done > > > > Otherwise I get the following on warning pandaboard-es: > > > > WARNING: CPU: 0 PID: 1770 at drivers/bus/omap_l3_noc.c:147 > > l3_interrupt_handler+0x2f8/0x388 > > 44000000.ocp:L3 Custom Error: MASTER MPU TARGET L4PER2 (Read): > > Data Access in User mode during Functional access > > > > Meaning that we access registers while not clocked > > somewhere. I think that warning is different from the > > earlier errors though :) And we could add a msleep(50) > > to the end to mostly avoid it until we have a better > > fix available? > > Well, that would paper over the problem, let's see if we can avoid it. > > I realized that I have not invoked mmc_hw_reset() with the sdio host > claimed. This could be the reason and is fixed in a v2, please try and > see if that solves the "new" problem. If id doesn't, you could also > convert the pm_runtime_put() to a pm_runtime_put_sync() in > wl12xx_sdio_power_off() and see if that makes a difference. Yes you're right v2 works for me no problem. I'll go ack it there. Tony