Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751180AbdFAWED (ORCPT ); Thu, 1 Jun 2017 18:04:03 -0400 Received: from bh-25.webhostbox.net ([208.91.199.152]:53947 "EHLO bh-25.webhostbox.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751078AbdFAWEC (ORCPT ); Thu, 1 Jun 2017 18:04:02 -0400 Date: Thu, 1 Jun 2017 15:03:58 -0700 From: Guenter Roeck To: Christopher Bostic Cc: wim@iguana.be, joel@jms.id.au, linux-kernel@vger.kernel.org, linux-watchdog@vger.kernel.org, andrew@aj.id.au Subject: Re: [PATCH] drivers/watchdog: Add sys files to modify watchdog behavior Message-ID: <20170601220358.GC8803@roeck-us.net> References: <20170601213521.58141-1-cbostic@linux.vnet.ibm.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20170601213521.58141-1-cbostic@linux.vnet.ibm.com> User-Agent: Mutt/1.5.24 (2015-08-30) X-Authenticated_sender: guenter@roeck-us.net X-OutGoing-Spam-Status: No, score=-1.0 X-AntiAbuse: This header was added to track abuse, please include it with any abuse report X-AntiAbuse: Primary Hostname - bh-25.webhostbox.net X-AntiAbuse: Original Domain - vger.kernel.org X-AntiAbuse: Originator/Caller UID/GID - [47 12] / [47 12] X-AntiAbuse: Sender Address Domain - roeck-us.net X-Get-Message-Sender-Via: bh-25.webhostbox.net: authenticated_id: guenter@roeck-us.net X-Authenticated-Sender: bh-25.webhostbox.net: guenter@roeck-us.net X-Source: X-Source-Args: X-Source-Dir: Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 6177 Lines: 176 On Thu, Jun 01, 2017 at 04:35:21PM -0500, Christopher Bostic wrote: > Add files to enable/disable Aspeed watchdog: > * External signal after timeout > * Reset system after timeout > The subject line is heavily misleading. > Signed-off-by: Christopher Bostic > --- > .../ABI/testing/sysfs-platform-aspeed-wdt | 40 +++++++++++ > drivers/watchdog/aspeed_wdt.c | 79 +++++++++++++++++++++- > 2 files changed, 118 insertions(+), 1 deletion(-) > create mode 100644 Documentation/ABI/testing/sysfs-platform-aspeed-wdt > > diff --git a/Documentation/ABI/testing/sysfs-platform-aspeed-wdt b/Documentation/ABI/testing/sysfs-platform-aspeed-wdt > new file mode 100644 > index 0000000..a582176 > --- /dev/null > +++ b/Documentation/ABI/testing/sysfs-platform-aspeed-wdt > @@ -0,0 +1,40 @@ > +What: /sys/bus/platform/devices/1e785000.wdt/watchdog/watchdog0/ > + reset_system > +Date: May 2017 > +KernelVersion: 4.12 > +Contact: Christopher Bostic > +Description: > + Value indicates if a watchdog time out results in a reset of > + system. 'off': no system reset on timeout. 'on' there will > + be a system reset on timeout. > + > + Default: off > + > + - Enable 'reset system' on watchdog timeout: > + echo on > /sys/bus/platform/devices/1e785000.wdt/watchdog/ > + watchdog0/reset_system; > + > + - Disable 'reset system' on watchdog timeout: > + echo off > /sys/bus/platform/devices/1e785000.wdt/watchdog/ > + watchdog0/reset_system; > + > +What: /sys/bus/platform/devices/1e785000.wdt/watchdog/watchdog0/ > + ext_signal > +Date: May 2017 > +KernelVersion: 4.12 > +Contact: Christopher Bostic > +Description: > + Value indicates if a watchdog time out results in an active > + high pulse sent out an output pin. 'off': No output pulse > + generated on watchdog time out. 'on': output pulse > + generated on watchdog time out. > + > + Default: off > + > + - Enable 'external signal' on watchdog timeout: > + echo on > /sys/bus/platform/devices/1e785000.wdt/watchdog/ > + watchdog0/ext_signal; > + > + - Disable 'external signal' on watchdog timeout: > + echo off > /sys/bus/platform/devices/1e785000.wdt/watchdog/ > + watchdog0/ext_signal; I'll want to see some discussion on those, both for the values written (on/off vs. 0/1) and the action to be taken. Both seem to be system parameters rather than something to set from user space. Also, there is no explanation what else the watchdog should do on timeout if not to reset the system. If it doesn't do anything, why run it in the first place ? Also, specifying absolute path names seems wrong. There is no real guarantee that the device is always watchdog0, that the path always includes "1e785000.wdt", or that the path name will always start with /sys/bus/platform/devices. Thanks, Guenter > diff --git a/drivers/watchdog/aspeed_wdt.c b/drivers/watchdog/aspeed_wdt.c > index 1c65258..66b27e8 100644 > --- a/drivers/watchdog/aspeed_wdt.c > +++ b/drivers/watchdog/aspeed_wdt.c > @@ -136,6 +136,83 @@ static int aspeed_wdt_restart(struct watchdog_device *wdd, > .identity = KBUILD_MODNAME, > }; > > +static ssize_t aspeed_wdt_cntl_get_state(struct device *dev, char *buf, > + uint32_t mask) > +{ > + struct platform_device *pdev = to_platform_device(dev); > + struct aspeed_wdt *wdt = platform_get_drvdata(pdev); > + int ret; > + > + if (readl(wdt->base + WDT_CTRL) & mask) > + ret = sprintf(buf, "on\n"); > + else > + ret = sprintf(buf, "off\n"); > + > + return ret; > +} > + > +static ssize_t aspeed_wdt_cntl_set_state(struct device *dev, const char *buf, > + size_t count, uint32_t mask) > +{ > + struct platform_device *pdev = to_platform_device(dev); > + struct aspeed_wdt *wdt = platform_get_drvdata(pdev); > + > + if (!strncmp(buf, "on", strlen("on"))) { > + wdt->ctrl |= mask; > + writel(wdt->ctrl, wdt->base + WDT_CTRL); > + } else if (!strncmp(buf, "off", strlen("off"))) { > + wdt->ctrl &= ~mask; > + writel(wdt->ctrl, wdt->base + WDT_CTRL); > + } else { > + dev_warn(dev, "Unknown reset system mode command: [%s]\n", buf); > + return -EINVAL; > + } > + > + return count; > +} > + > +static ssize_t aspeed_wdt_reset_system_store(struct device *dev, > + struct device_attribute *attr, > + const char *buf, size_t count) > +{ > + return aspeed_wdt_cntl_set_state(dev, buf, count, > + WDT_CTRL_RESET_SYSTEM); > +} > + > +static ssize_t aspeed_wdt_reset_system_show(struct device *dev, > + struct device_attribute *attr, > + char *buf) > +{ > + return aspeed_wdt_cntl_get_state(dev, buf, WDT_CTRL_RESET_SYSTEM); > +} > + > +static DEVICE_ATTR(reset_system, 0644, > + aspeed_wdt_reset_system_show, aspeed_wdt_reset_system_store); > + > +static ssize_t aspeed_wdt_ext_signal_store(struct device *dev, > + struct device_attribute *attr, > + const char *buf, size_t count) > +{ > + return aspeed_wdt_cntl_set_state(dev, buf, count, WDT_CTRL_WDT_EXT); > +} > + > +static ssize_t aspeed_wdt_ext_signal_show(struct device *dev, > + struct device_attribute *attr, > + char *buf) > +{ > + return aspeed_wdt_cntl_get_state(dev, buf, WDT_CTRL_WDT_EXT); > +} > + > +static DEVICE_ATTR(ext_signal, 0644, > + aspeed_wdt_ext_signal_show, aspeed_wdt_ext_signal_store); > + > +static struct attribute *aspeed_wdt_attrs[] = { > + &dev_attr_reset_system.attr, > + &dev_attr_ext_signal.attr, > + NULL > +}; > +ATTRIBUTE_GROUPS(aspeed_wdt); > + > static int aspeed_wdt_probe(struct platform_device *pdev) > { > struct aspeed_wdt *wdt; > @@ -160,7 +237,7 @@ static int aspeed_wdt_probe(struct platform_device *pdev) > wdt->wdd.ops = &aspeed_wdt_ops; > wdt->wdd.max_hw_heartbeat_ms = WDT_MAX_TIMEOUT_MS; > wdt->wdd.parent = &pdev->dev; > - > + wdt->wdd.groups = aspeed_wdt_groups; > wdt->wdd.timeout = WDT_DEFAULT_TIMEOUT; > watchdog_init_timeout(&wdt->wdd, 0, &pdev->dev); > > -- > 1.8.2.2 >