aboutsummaryrefslogtreecommitdiffstats
path: root/include/linux/devm-helpers.h
diff options
context:
space:
mode:
authorLinus Torvalds <torvalds@linux-foundation.org>2021-04-26 11:05:36 -0700
committerLinus Torvalds <torvalds@linux-foundation.org>2021-04-26 11:05:36 -0700
commitc01c0716ccf5db2086d9693033472f37de96a699 (patch)
tree330fbb4c8a386073a25fb17ee29be15e21266ab3 /include/linux/devm-helpers.h
parentMerge tag 'char-misc-5.13-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc (diff)
parentdevm-helpers: Fix devm_delayed_work_autocancel() kerneldoc (diff)
downloadlinux-dev-c01c0716ccf5db2086d9693033472f37de96a699.tar.xz
linux-dev-c01c0716ccf5db2086d9693033472f37de96a699.zip
Merge tag 'driver-core-5.13-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core
Pull driver core updates from Greg KH: "Here is the "big" set of driver core changes for 5.13-rc1. Nothing major, just lots of little core changes and cleanups, notable things are: - finally set 'fw_devlink=on' by default. All reported issues with this have been shaken out over the past 9 months or so, but we will be paying attention to any fallout here in case we need to revert this as the default boot value (symptoms of problems are a simple lack of booting) - fixes found to be needed by fw_devlink=on value in some subsystems (like clock). - delayed work initialization cleanup - driver core cleanups and minor updates - software node cleanups and tweaks - devtmpfs cleanups - minor debugfs cleanups All of these have been in linux-next for a while with no reported issues" * tag 'driver-core-5.13-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/driver-core: (53 commits) devm-helpers: Fix devm_delayed_work_autocancel() kerneldoc PM / wakeup: use dev_set_name() directly software node: Allow node addition to already existing device kunit: software node: adhear to KUNIT formatting standard node: fix device cleanups in error handling code kobject_uevent: remove warning in init_uevent_argv() debugfs: Make debugfs_allow RO after init Revert "driver core: platform: Make platform_get_irq_optional() optional" media: ipu3-cio2: Switch to use SOFTWARE_NODE_REFERENCE() software node: Introduce SOFTWARE_NODE_REFERENCE() helper macro software node: Imply kobj_to_swnode() to be no-op software node: Deduplicate code in fwnode_create_software_node() software node: Introduce software_node_alloc()/software_node_free() software node: Free resources explicitly when swnode_register() fails debugfs: drop pointless nul-termination in debugfs_read_file_bool() driver core: add helper for deferred probe reason setting driver core: Improve fw_devlink & deferred_probe_timeout interaction of: property: fw_devlink: Add support for remote-endpoint driver core: platform: Make platform_get_irq_optional() optional driver core: Replace printf() specifier and drop unneeded casting ...
Diffstat (limited to 'include/linux/devm-helpers.h')
-rw-r--r--include/linux/devm-helpers.h54
1 files changed, 54 insertions, 0 deletions
diff --git a/include/linux/devm-helpers.h b/include/linux/devm-helpers.h
new file mode 100644
index 000000000000..f40f77717a24
--- /dev/null
+++ b/include/linux/devm-helpers.h
@@ -0,0 +1,54 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+#ifndef __LINUX_DEVM_HELPERS_H
+#define __LINUX_DEVM_HELPERS_H
+
+/*
+ * Functions which do automatically cancel operations or release resources upon
+ * driver detach.
+ *
+ * These should be helpful to avoid mixing the manual and devm-based resource
+ * management which can be source of annoying, rarely occurring,
+ * hard-to-reproduce bugs.
+ *
+ * Please take into account that devm based cancellation may be performed some
+ * time after the remove() is ran.
+ *
+ * Thus mixing devm and manual resource management can easily cause problems
+ * when unwinding operations with dependencies. IRQ scheduling a work in a queue
+ * is typical example where IRQs are often devm-managed and WQs are manually
+ * cleaned at remove(). If IRQs are not manually freed at remove() (and this is
+ * often the case when we use devm for IRQs) we have a period of time after
+ * remove() - and before devm managed IRQs are freed - where new IRQ may fire
+ * and schedule a work item which won't be cancelled because remove() was
+ * already ran.
+ */
+
+#include <linux/device.h>
+#include <linux/workqueue.h>
+
+static inline void devm_delayed_work_drop(void *res)
+{
+ cancel_delayed_work_sync(res);
+}
+
+/**
+ * devm_delayed_work_autocancel - Resource-managed delayed work allocation
+ * @dev: Device which lifetime work is bound to
+ * @w: Work item to be queued
+ * @worker: Worker function
+ *
+ * Initialize delayed work which is automatically cancelled when driver is
+ * detached. A few drivers need delayed work which must be cancelled before
+ * driver is detached to avoid accessing removed resources.
+ * devm_delayed_work_autocancel() can be used to omit the explicit
+ * cancelleation when driver is detached.
+ */
+static inline int devm_delayed_work_autocancel(struct device *dev,
+ struct delayed_work *w,
+ work_func_t worker)
+{
+ INIT_DELAYED_WORK(w, worker);
+ return devm_add_action(dev, devm_delayed_work_drop, w);
+}
+
+#endif