From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Google-Smtp-Source: AB8JxZqVJ9tzRDS7Apm2ZWdqGAp+yNr6IpW1GFbOuq19LtTwO94l/RhS5CheJvyq5a9b5b9KYpSu ARC-Seal: i=1; a=rsa-sha256; t=1525534851; cv=none; d=google.com; s=arc-20160816; b=JsiXLlnD+m9ZjEbBBSP3Lm+DsxkP/Zly/QyuoIIIpBnDVJJrGZzyGVBD4pd5NcSDZm P7V7ERDhY314uKW53Dv5rPsD8M9DgqioT6iNwEQ17O9JfE5F7PwuoxZzRfzP7eJaR/9U H5iBHXufFWotG/s7gVdA2dAphBp1tg/DzftQ7FS4vQXiGPsw/IRQZBOWa2I8OqDgYFYh F9ous3cYcbc5ulI2EVuHzUKy0vja0lmXjYAMulHMRsbjfPYPGxK9JadGBes7FCZx1yeG o5JMJ1+dsrV0DoXy5qn8ukroy14imC7hOQjuH2+6ZfVZw/FUPa0oLm//bp1YeXAz53kC pl5A== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=arc-20160816; h=references:in-reply-to:message-id:date:subject:to:from :dkim-signature:arc-authentication-results; bh=jH41MutgCV2CuZW+iGOgIx2JxZezF4sEzd+KoagsfsA=; b=FlzlZjMVspeB4GD29JQ9LcbWeciWgkAGDEOz2H2qeXhi3AdZ8GSZqhMHfw5ChUU9Kr KjeJ0fdp5bqDLwqT2pe5bI3mJhJ9AQpQcsftDuUaH387NCBbhmo2Yex7GVPvsGfigTNN nQI7vexcUYT0GFBDSLwz4EkKOrMYJIJTkFdXqc5Jd77jJG+Q41dfHxMi4svyJ6QqJL1M AEe0WuErbqKXtjVy4TOX1X2VTCCYYdKCaApCcFMTfUMbjVgN/Y0weWdV4/9ztLV4Wrag E5DIW4NlcWgjp4AvQxBEh3luTbV0xWrBaeSCl1H+VAO8CEudzmpZG1ny830lRu43Ends xsBw== ARC-Authentication-Results: i=1; mx.google.com; dkim=pass header.i=@oracle.com header.s=corp-2017-10-26 header.b=JW7qFBeU; spf=pass (google.com: domain of pasha.tatashin@oracle.com designates 141.146.126.79 as permitted sender) smtp.mailfrom=pasha.tatashin@oracle.com; dmarc=pass (p=NONE sp=NONE dis=NONE) header.from=oracle.com Authentication-Results: mx.google.com; dkim=pass header.i=@oracle.com header.s=corp-2017-10-26 header.b=JW7qFBeU; spf=pass (google.com: domain of pasha.tatashin@oracle.com designates 141.146.126.79 as permitted sender) smtp.mailfrom=pasha.tatashin@oracle.com; dmarc=pass (p=NONE sp=NONE dis=NONE) header.from=oracle.com From: Pavel Tatashin To: pasha.tatashin@oracle.com, steven.sistare@oracle.com, daniel.m.jordan@oracle.com, linux-kernel@vger.kernel.org, jeffrey.t.kirsher@intel.com, intel-wired-lan@lists.osuosl.org, netdev@vger.kernel.org, gregkh@linuxfoundation.org, alexander.duyck@gmail.com, tobin@apporbit.com Subject: [PATCH v2 1/1] drivers core: multi-threading device shutdown Date: Sat, 5 May 2018 11:40:40 -0400 Message-Id: <20180505154040.28614-2-pasha.tatashin@oracle.com> X-Mailer: git-send-email 2.17.0 In-Reply-To: <20180505154040.28614-1-pasha.tatashin@oracle.com> References: <20180505154040.28614-1-pasha.tatashin@oracle.com> X-Proofpoint-Virus-Version: vendor=nai engine=5900 definitions=8884 signatures=668698 X-Proofpoint-Spam-Details: rule=notspam policy=default score=0 suspectscore=2 malwarescore=0 phishscore=0 bulkscore=0 spamscore=0 mlxscore=0 mlxlogscore=999 adultscore=0 classifier=spam adjust=0 reason=mlx scancount=1 engine=8.0.1-1711220000 definitions=main-1805050154 X-getmail-retrieved-from-mailbox: INBOX X-GMAIL-THRID: =?utf-8?q?1599639232732130480?= X-GMAIL-MSGID: =?utf-8?q?1599639232732130480?= X-Mailing-List: linux-kernel@vger.kernel.org List-ID: When system is rebooted, halted or kexeced device_shutdown() is called. This function shuts down every single device by calling either: dev->bus->shutdown(dev) dev->driver->shutdown(dev) Even on a machine with just a moderate amount of devices, device_shutdown() may take multiple seconds to complete. This is because many devices require a specific delays to perform this operation. Here is a sample analysis of time it takes to call device_shutdown() on a two socket Intel(R) Xeon(R) CPU E5-2630 v4 @ 2.20GHz machine. device_shutdown 2.95s ----------------------------- mlx4_shutdown 1.14s megasas_shutdown 0.24s ixgbe_shutdown 0.37s x 4 (four ixgbe devices on this machine). the rest 0.09s In mlx4 we spent the most time, but that is because there is a 1 second sleep, which is defined by hardware specifications: mlx4_shutdown mlx4_unload_one mlx4_free_ownership msleep(1000) With megasas we spend a quarter of a second, but sometimes longer (up-to 0.5s) in this path: megasas_shutdown megasas_flush_cache megasas_issue_blocked_cmd wait_event_timeout Finally, with ixgbe_shutdown() it takes 0.37 for each device, but that time is spread all over the place, with bigger offenders: ixgbe_shutdown __ixgbe_shutdown ixgbe_close_suspend ixgbe_down ixgbe_init_hw_generic ixgbe_reset_hw_X540 msleep(100); 0.104483472 ixgbe_get_san_mac_addr_generic 0.048414851 ixgbe_get_wwn_prefix_generic 0.048409893 ixgbe_start_hw_X540 ixgbe_start_hw_generic ixgbe_clear_hw_cntrs_generic 0.048581502 ixgbe_setup_fc_generic 0.024225800 All the ixgbe_*generic functions end-up calling: ixgbe_read_eerd_X540() ixgbe_acquire_swfw_sync_X540 usleep_range(5000, 6000); ixgbe_release_swfw_sync_X540 usleep_range(5000, 6000); While these are short sleeps, they end-up calling them over 24 times! 24 * 0.0055s = 0.132s. Adding-up to 0.528s for four devices. Also we have four msleep(100). Totaling to: 0.928s While we should keep optimizing the individual device drivers, in some cases this is simply a hardware property that forces a specific delay, and we must wait. So, the solution for this problem is to shutdown devices in parallel. However, we must shutdown children before shutting down parents, so parent device must wait for its children to finish. With this patch, on the same machine devices_shutdown() takes 1.142s, and without mlx4 one second delay only 0.38s Signed-off-by: Pavel Tatashin --- drivers/base/core.c | 275 ++++++++++++++++++++++++++++++++++++-------- 1 file changed, 225 insertions(+), 50 deletions(-) diff --git a/drivers/base/core.c b/drivers/base/core.c index b610816eb887..110d7970deef 100644 --- a/drivers/base/core.c +++ b/drivers/base/core.c @@ -25,6 +25,7 @@ #include #include #include +#include #include "base.h" #include "power/power.h" @@ -2102,6 +2103,59 @@ const char *device_get_devnode(struct device *dev, return *tmp = s; } +/** + * device_children_count - device children count + * @parent: parent struct device. + * + * Returns number of children for this device or 0 if none. + */ +static int device_children_count(struct device *parent) +{ + struct klist_iter i; + int children = 0; + + if (!parent->p) + return 0; + + klist_iter_init(&parent->p->klist_children, &i); + while (next_device(&i)) + children++; + klist_iter_exit(&i); + + return children; +} + +/** + * device_get_child_by_index - Return child using the provided index. + * @parent: parent struct device. + * @index: Index of the child, where 0 is the first child in the children list, + * and so on. + * + * Returns child or NULL if child with this index is not present. + */ +static struct device * +device_get_child_by_index(struct device *parent, int index) +{ + struct klist_iter i; + struct device *dev = NULL, *d; + int child_index = 0; + + if (!parent->p || index < 0) + return NULL; + + klist_iter_init(&parent->p->klist_children, &i); + while ((d = next_device(&i))) { + if (child_index == index) { + dev = d; + break; + } + child_index++; + } + klist_iter_exit(&i); + + return dev; +} + /** * device_for_each_child - device child iterator. * @parent: parent struct device. @@ -2765,71 +2819,192 @@ int device_move(struct device *dev, struct device *new_parent, } EXPORT_SYMBOL_GPL(device_move); +/* + * device_shutdown_one - call ->shutdown() for the device passed as + * argument. + */ +static void device_shutdown_one(struct device *dev) +{ + /* Don't allow any more runtime suspends */ + pm_runtime_get_noresume(dev); + pm_runtime_barrier(dev); + + if (dev->class && dev->class->shutdown_pre) { + if (initcall_debug) + dev_info(dev, "shutdown_pre\n"); + dev->class->shutdown_pre(dev); + } + if (dev->bus && dev->bus->shutdown) { + if (initcall_debug) + dev_info(dev, "shutdown\n"); + dev->bus->shutdown(dev); + } else if (dev->driver && dev->driver->shutdown) { + if (initcall_debug) + dev_info(dev, "shutdown\n"); + dev->driver->shutdown(dev); + } + + /* decrement the reference counter */ + put_device(dev); +} + +/** + * Passed as an argument to device_shutdown_child_task(). + * child_next_index the next available child index. + * tasks_running number of tasks still running. Each tasks decrements it + * when job is finished and the last task signals that the + * job is complete. + * complete Used to signal job competition. + * parent Parent device. + */ +struct device_shutdown_task_data { + atomic_t child_next_index; + atomic_t tasks_running; + struct completion complete; + struct device *parent; +}; + +static int device_shutdown_child_task(void *data); + +/** + * These globals are used by tasks that are started for root devices. + * device_root_tasks_finished Number of root devices finished shutting down. + * device_root_tasks_started Total number of root devices tasks started. + * device_root_tasks_done The completion signal to the main thread. + */ +static atomic_t device_root_tasks_finished; +static atomic_t device_root_tasks_started; +struct completion device_root_tasks_done; + +/** + * Shutdown device tree with root started in dev. If dev has no children + * simply shutdown only this device. If dev has children recursively shutdown + * children first, and only then the parent. For performance reasons children + * are shutdown in parallel using kernel threads. dev is locked, so its children + * cannot be removed while this functions is running. + */ +static void device_shutdown_tree(struct device *dev) +{ + int children_count = device_children_count(dev); + + if (children_count) { + struct device_shutdown_task_data tdata; + int i; + + init_completion(&tdata.complete); + atomic_set(&tdata.child_next_index, 0); + atomic_set(&tdata.tasks_running, children_count); + tdata.parent = dev; + + for (i = 0; i < children_count; i++) { + kthread_run(device_shutdown_child_task, + &tdata, "device_shutdown.%s", + dev_name(dev)); + } + wait_for_completion(&tdata.complete); + } + device_shutdown_one(dev); +} + +/** + * Only devices with parent are going through this function. The parent is + * locked and waits for all of its children to finish shutting down before + * calling shutdown function for itself. + */ +static int device_shutdown_child_task(void *data) +{ + struct device_shutdown_task_data *tdata = data; + int cidx = atomic_inc_return(&tdata->child_next_index) - 1; + struct device *dev = device_get_child_by_index(tdata->parent, cidx); + + /* ref. counter is going to be decremented in device_shutdown_one() */ + get_device(dev); + device_lock(dev); + device_shutdown_tree(dev); + device_unlock(dev); + + /* If we are the last to exit, signal the completion */ + if (atomic_dec_return(&tdata->tasks_running) == 0) + complete(&tdata->complete); + return 0; +} + +/** + * On shutdown each root device (the one that does not have a parent) goes + * through this function. + */ +static int device_shutdown_root_task(void *data) +{ + struct device *dev = (struct device *)data; + int root_devices; + + device_lock(dev); + device_shutdown_tree(dev); + device_unlock(dev); + + /* If we are the last to exit, signal the completion */ + root_devices = atomic_inc_return(&device_root_tasks_finished); + if (root_devices == atomic_read(&device_root_tasks_started)) + complete(&device_root_tasks_done); + return 0; +} + /** * device_shutdown - call ->shutdown() on each device to shutdown. */ void device_shutdown(void) { - struct device *dev, *parent; + int root_devices = 0; + struct device *dev; - spin_lock(&devices_kset->list_lock); - /* - * Walk the devices list backward, shutting down each in turn. - * Beware that device unplug events may also start pulling - * devices offline, even as the system is shutting down. + atomic_set(&device_root_tasks_finished, 0); + atomic_set(&device_root_tasks_started, 0); + init_completion(&device_root_tasks_done); + + /* Shutdown the root devices in parallel. The children are going to be + * shutdown first in device_shutdown_tree(). */ + spin_lock(&devices_kset->list_lock); while (!list_empty(&devices_kset->list)) { - dev = list_entry(devices_kset->list.prev, struct device, - kobj.entry); + dev = list_entry(devices_kset->list.next, struct device, + kobj.entry); - /* - * hold reference count of device's parent to - * prevent it from being freed because parent's - * lock is to be held - */ - parent = get_device(dev->parent); - get_device(dev); - /* - * Make sure the device is off the kset list, in the - * event that dev->*->shutdown() doesn't remove it. + /* Make sure the device is off the kset list, in the event that + * dev->*->shutdown() doesn't remove it. */ list_del_init(&dev->kobj.entry); - spin_unlock(&devices_kset->list_lock); - - /* hold lock to avoid race with probe/release */ - if (parent) - device_lock(parent); - device_lock(dev); - - /* Don't allow any more runtime suspends */ - pm_runtime_get_noresume(dev); - pm_runtime_barrier(dev); - if (dev->class && dev->class->shutdown_pre) { - if (initcall_debug) - dev_info(dev, "shutdown_pre\n"); - dev->class->shutdown_pre(dev); + /* Here we start tasks for root devices only */ + if (!dev->parent) { + /* Prevents devices from being freed. The counter is + * going to be decremented in device_shutdown_one() once + * this root device is shutdown. + */ + get_device(dev); + + /* We unlock list for performance reasons, + * dev->*->shutdown(), may try to take this lock to + * remove us from kset list. To avoid unlocking this + * list we could replace spin lock in: + * dev->kobj.kset->list_lock with a dummy one once + * device is locked in device_shutdown_root_task() and + * in device_shutdown_child_task(). + */ + spin_unlock(&devices_kset->list_lock); + + root_devices++; + kthread_run(device_shutdown_root_task, + dev, "device_root_shutdown.%s", + dev_name(dev)); + spin_lock(&devices_kset->list_lock); } - if (dev->bus && dev->bus->shutdown) { - if (initcall_debug) - dev_info(dev, "shutdown\n"); - dev->bus->shutdown(dev); - } else if (dev->driver && dev->driver->shutdown) { - if (initcall_debug) - dev_info(dev, "shutdown\n"); - dev->driver->shutdown(dev); - } - - device_unlock(dev); - if (parent) - device_unlock(parent); - - put_device(dev); - put_device(parent); - - spin_lock(&devices_kset->list_lock); } spin_unlock(&devices_kset->list_lock); + + /* Set number of root tasks started, and waits for completion */ + atomic_set(&device_root_tasks_started, root_devices); + if (root_devices != atomic_read(&device_root_tasks_finished)) + wait_for_completion(&device_root_tasks_done); } /* -- 2.17.0