summaryrefslogtreecommitdiffstats
path: root/drivers/base/base.h (follow)
Commit message (Collapse)AuthorAgeFilesLines
* driver core: module: make module_[add|remove]_driver take a const *Greg Kroah-Hartman2024-07-101-2/+2
| | | | | | | | | | The functions module_add_driver() and module_remove_driver() do not modify the struct device_driver structure directly, so they are safe to be marked as a constant pointer type. Cc: "Rafael J. Wysocki" <rafael@kernel.org> Link: https://lore.kernel.org/r/2024070850-entering-grandson-205e@gregkh Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* driver core: have match() callback in struct bus_type take a const *Greg Kroah-Hartman2024-07-031-2/+1
| | | | | | | | | | | | | | | | | | | | | | | In the match() callback, the struct device_driver * should not be changed, so change the function callback to be a const *. This is one step of many towards making the driver core safe to have struct device_driver in read-only memory. Because the match() callback is in all busses, all busses are modified to handle this properly. This does entail switching some container_of() calls to container_of_const() to properly handle the constant *. For some busses, like PCI and USB and HV, the const * is cast away in the match callback as those busses do want to modify those structures at this point in time (they have a local lock in the driver structure.) That will have to be changed in the future if they wish to have their struct device * in read-only-memory. Cc: Rafael J. Wysocki <rafael@kernel.org> Reviewed-by: Alex Elder <elder@kernel.org> Acked-by: Sumit Garg <sumit.garg@linaro.org> Link: https://lore.kernel.org/r/2024070136-wrongdoer-busily-01e8@gregkh Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* driver core: make [device_]driver_attach take a const *Greg Kroah-Hartman2024-06-201-2/+3
| | | | | | | | | | | | | | | | | | | | Change device_driver_attach() and driver_attach() to take a const * to struct device driver as neither of them modify the structure at all. Also, for some odd reason, drivers/dma/idxd/compat.c had a duplicate external reference to device_driver_attach(), so remove that to fix up the build, it should never have had that there in the first place. Cc: Rafael J. Wysocki <rafael@kernel.org> Cc: Fenghua Yu <fenghua.yu@intel.com> Cc: Dave Jiang <dave.jiang@intel.com> Cc: Vinod Koul <vkoul@kernel.org> Cc: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Cc: Petr Tesarik <petr.tesarik.ext@huawei.com> Cc: Alexander Lobakin <aleksander.lobakin@intel.com> Cc: dmaengine@vger.kernel.org Link: https://lore.kernel.org/r/2024061401-rasping-manger-c385@gregkh Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* driver core: mark async_driver as a const *Greg Kroah-Hartman2024-06-131-1/+1
| | | | | | | | | | | | | Within struct device_private, mark the async_driver * as const as it is never modified. This requires some internal-to-the-driver-core functions to also have their parameters marked as constant, and there is one place where we cast _back_ from the const pointer to a real one, as the driver core still wants to modify the structure in a number of remaining places. Cc: Rafael J. Wysocki <rafael@kernel.org> Link: https://lore.kernel.org/r/20240611130103.3262749-12-gregkh@linuxfoundation.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* driver core: make driver_detach() take a const *Greg Kroah-Hartman2024-06-131-1/+1
| | | | | | | | | | | driver_detach() does not modify the driver itself, so make the pointer constant. In doing so, the function driver_allows_async_probing() also needs to be changed so that the pointer type passes through to that function properly. Cc: Rafael J. Wysocki <rafael@kernel.org> Link: https://lore.kernel.org/r/20240611130103.3262749-11-gregkh@linuxfoundation.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* driver core: make device_release_driver_internal() take a const *Greg Kroah-Hartman2024-06-131-1/+1
| | | | | | | | | Change device_release_driver_internal() to take a const struct device_driver * as it is not modifying it at all. Cc: Rafael J. Wysocki <rafael@kernel.org> Link: https://lore.kernel.org/r/20240611130103.3262749-10-gregkh@linuxfoundation.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* driver core: driver: mark driver_add/remove_groups constantGreg Kroah-Hartman2024-06-131-2/+2
| | | | | | | | | driver_add_groups() and driver_remove_groups should take a constant pointer as the structure is not modified, so make the change. Cc: Rafael J. Wysocki <rafael@kernel.org> Link: https://lore.kernel.org/r/20240611130103.3262749-9-gregkh@linuxfoundation.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* module: don't ignore sysfs_create_link() failuresArnd Bergmann2024-04-111-3/+6
| | | | | | | | | | | | | | | | | | | | | | The sysfs_create_link() return code is marked as __must_check, but the module_add_driver() function tries hard to not care, by assigning the return code to a variable. When building with 'make W=1', gcc still warns because this variable is only assigned but not used: drivers/base/module.c: In function 'module_add_driver': drivers/base/module.c:36:6: warning: variable 'no_warn' set but not used [-Wunused-but-set-variable] Rework the code to properly unwind and return the error code to the caller. My reading of the original code was that it tries to not fail when the links already exist, so keep ignoring -EEXIST errors. Fixes: e17e0f51aeea ("Driver core: show drivers in /sys/module/") See-also: 4a7fb6363f2d ("add __must_check to device management code") Signed-off-by: Arnd Bergmann <arnd@arndb.de> Reviewed-by: Luis Chamberlain <mcgrof@kernel.org> Link: https://lore.kernel.org/r/20240408080616.3911573-1-arnd@kernel.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* block: make block_class constantRicardo B. Marliere2024-03-061-1/+1
| | | | | | | | | | | | | | | Since commit 43a7206b0963 ("driver core: class: make class_register() take a const *"), the driver core allows for struct class to be in read-only memory, so move the block_class structure to be declared at build time placing it into read-only memory, instead of having to be dynamically allocated at boot time. Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Suggested-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Ricardo B. Marliere <ricardo@marliere.net> Reviewed-by: Christoph Hellwig <hch@lst.de> Link: https://lore.kernel.org/r/20240305-class_cleanup-block-v1-1-130bb27b9c72@marliere.net Signed-off-by: Jens Axboe <axboe@kernel.dk>
* driver core: class: make class_register() take a const *Greg Kroah-Hartman2023-04-031-1/+1
| | | | | | | | | | Now that the class code is cleaned up to not modify the class pointer registered with it, change class_register() to take a const * to allow the structure to be placed into read-only memory. Acked-by: Rafael J. Wysocki <rafael@kernel.org> Link: https://lore.kernel.org/r/2023040248-customary-release-4aec@gregkh Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* driver core: make sysfs_dev_char_kobj staticGreg Kroah-Hartman2023-03-311-3/+0
| | | | | | | | | | Nothing outside of drivers/base/core.c uses sysfs_dev_char_kobj, so make it static and document what it is used for so we remember it the next time we touch it 15 years from now. Acked-by: Rafael J. Wysocki <rafael@kernel.org> Link: https://lore.kernel.org/r/20230331093318.82288-7-gregkh@linuxfoundation.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* driver core: clean up the logic to determine which /sys/dev/ directory to useGreg Kroah-Hartman2023-03-311-0/+10
| | | | | | | | | | | | | | | | | | | | | | When a dev_t is set in a struct device, an symlink in /sys/dev/ is created for it either under /sys/dev/block/ or /sys/dev/char/ depending on the device type. The logic to determine this would trigger off of the class of the object, and the kobj_type set in that location. But it turns out that this deep nesting isn't needed at all, as it's either a choice of block or "everything else" which is a char device. So make the logic a lot more simple and obvious, and remove the incorrect comments in the code that tried to document something that was not happening at all (it is impossible to set class->dev_kobj to NULL as the class core prevented that from happening. This removes the only place that class->dev_kobj was being used, so after this, it can be removed entirely. Acked-by: Rafael J. Wysocki <rafael@kernel.org> Link: https://lore.kernel.org/r/20230331093318.82288-4-gregkh@linuxfoundation.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* driver core: core: move to use class_to_subsys()Greg Kroah-Hartman2023-03-311-0/+2
| | | | | | | | | | | | | There are a number of places in core.c that need access to the private subsystem structure of struct class, so move them to use class_to_subsys() instead of accessing it directly. This requires exporting class_to_subsys() out of class.c, but keeping it local to the driver core. Reviewed-by: Rafael J. Wysocki <rafael@kernel.org> Link: https://lore.kernel.org/r/20230331093318.82288-1-gregkh@linuxfoundation.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* driver core: move sysfs_dev_char_kobj out of class.hGreg Kroah-Hartman2023-03-271-0/+3
| | | | | | | | | | The structure sysfs_dev_char_kobj is local only to the driver core code, so move it out of the global class.h file and into the internal base.h file as no one else should be touching this symbol. Acked-by: Rafael J. Wysocki <rafael@kernel.org> Link: https://lore.kernel.org/r/20230327160319.513974-1-gregkh@linuxfoundation.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* driver core: bus: move documentation for lock_key to proper location.Greg Kroah-Hartman2023-03-251-0/+1
| | | | | | | | | | | | | In commit 37e98d9bedb5 ("driver core: bus: move lock_class_key into dynamic structure"), the lock_key variable moved out of struct bus_type and into struct subsys_private, yet the documentation for it did not move. Fix that up and place the documentation comment in the correct location. Cc: "Rafael J. Wysocki" <rafael@kernel.org> Fixes: 37e98d9bedb5 ("driver core: bus: move lock_class_key into dynamic structure") Link: https://lore.kernel.org/r/20230324090814.386654-1-gregkh@linuxfoundation.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* driver core: base.h: remove extern from function prototypesGreg Kroah-Hartman2023-03-241-46/+42
| | | | | | | | | | The kernel coding style does not require 'extern' in function prototypes in .h files, so remove them from drivers/base/base.h as they are not needed. Acked-by: Rafael J. Wysocki <rafael@kernel.org> Link: https://lore.kernel.org/r/20230324122711.2664537-4-gregkh@linuxfoundation.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* driver core: bus: constantify bus_register()Greg Kroah-Hartman2023-03-231-1/+1
| | | | | | | | | bus_register() is now safe to take a constant * to bus_type, so make that change and mark the subsys_private bus_type * constant as well. Cc: "Rafael J. Wysocki" <rafael@kernel.org> Link: https://lore.kernel.org/r/20230313182918.1312597-24-gregkh@linuxfoundation.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* driver core: bus: move dev_root out of struct bus_typeGreg Kroah-Hartman2023-03-231-0/+2
| | | | | | | | | | | | | Now that all accesses of dev_root is through the bus_get_dev_root() call, move the pointer out of struct bus_type and into the private dynamic structure, subsys_private. With this change, there is no modifiable portions of struct bus_type so it can be marked as a constant structure and moved to read-only memory. Cc: "Rafael J. Wysocki" <rafael@kernel.org> Link: https://lore.kernel.org/r/20230313182918.1312597-22-gregkh@linuxfoundation.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* driver core: Make state_synced device attribute writeableSaravana Kannan2023-03-101-0/+8
| | | | | | | | | | | | | | | | | If the file is written to and sync_state() hasn't been called for the device yet, then call sync_state() for the device independent of the state of its consumers. This is useful for supplier devices that have one or more consumers that don't have a driver but the consumers are in a state that don't use the resources supplied by the supplier device. This gives finer grained control than using the fw_devlink.sync_state=timeout kernel commandline parameter. Signed-off-by: Saravana Kannan <saravanak@google.com> Link: https://lore.kernel.org/r/20230304005355.746421-3-saravanak@google.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* driver core: Add fw_devlink.sync_state command line paramSaravana Kannan2023-03-101-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | When all devices that could probe have finished probing (based on deferred_probe_timeout configuration or late_initcall() when !CONFIG_MODULES), this parameter controls what to do with devices that haven't yet received their sync_state() calls. fw_devlink.sync_state=strict is the default and the driver core will continue waiting on all consumers of a device to probe successfully before sync_state() is called for the device. This is the default behavior since calling sync_state() on a device when all its consumers haven't probed could make some systems unusable/unstable. When this option is selected, we also print the list of devices that haven't had sync_state() called on them by the time all devices the could probe have finished probing. fw_devlink.sync_state=timeout will cause the driver core to give up waiting on consumers and call sync_state() on any devices that haven't yet received their sync_state() calls. This option is provided for systems that won't become unusable/unstable as they might be able to save power (depends on state of hardware before kernel starts) if all devices get their sync_state(). Signed-off-by: Saravana Kannan <saravanak@google.com> Link: https://lore.kernel.org/r/20230304005355.746421-2-saravanak@google.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* Revert "devtmpfs: remove return value of devtmpfs_delete_node()"Greg Kroah-Hartman2023-02-141-2/+2
| | | | | | | | | | This reverts commit 9d3fe6aa6b9517408064c7c3134187e8ec77dbf7 as it is reported to cause boot regressions. Link: https://lore.kernel.org/r/Y+rSXg14z1Myd8Px@dev-arch.thelio-3990X Reported-by: Nathan Chancellor <nathan@kernel.org> Cc: Longlong Xia <xialonglong1@huawei.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* devtmpfs: remove return value of devtmpfs_delete_node()Longlong Xia2023-02-111-2/+2
| | | | | | | | | | The only caller of device_del() does not check the return value. And there's nothing we can do when cleaning things up on a remove path. Let's make it a void function. Signed-off-by: Longlong Xia <xialonglong1@huawei.com> Link: https://lore.kernel.org/r/20230210095444.4067307-4-xialonglong1@huawei.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* driver core: create bus_is_registered()Greg Kroah-Hartman2023-02-091-0/+1
| | | | | | | | | | | A local function to the driver core to determine if a bus really is registered with the kernel or not. To be used only by the driver core code, as part of the driver registration path as it's not really "safe" because the bus could be unregistered instantly after being called. Cc: "Rafael J. Wysocki" <rafael@kernel.org> Link: https://lore.kernel.org/r/20230208111330.439504-17-gregkh@linuxfoundation.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* driver core: add local subsys_get and subsys_put functionsGreg Kroah-Hartman2023-02-091-0/+13
| | | | | | | | | | | We need to control the reference count of the subsys private structure instead of directly manipulating the kset reference count of it, so wrap that logic up in a subsys_get() and subsys_put() function to make it more obvious as to what is happening. Cc: "Rafael J. Wysocki" <rafael@kernel.org> Link: https://lore.kernel.org/r/20230208111330.439504-2-gregkh@linuxfoundation.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* driver core: bus: move lock_class_key into dynamic structureGreg Kroah-Hartman2023-02-011-0/+2
| | | | | | | | | | | | | | | | | | | Move the lock_class_key structure out of struct bus_type and into the dynamic structure we create already for all bus_types registered with the kernel. This saves on static space and removes one more writable field in struct bus_type. In the future, the same field can be moved out of the struct class logic because it shares this same private structure. Most everyone will never notice this change, as lockdep is not enabled in real systems so no memory or logic changes are happening for them. Cc: "Rafael J. Wysocki" <rafael@kernel.org> Acked-by: Rafael J. Wysocki <rafael@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Link: https://lore.kernel.org/r/20230201083349.4038660-1-gregkh@linuxfoundation.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* driver core: device_get_devnode() should take a const *Greg Kroah-Hartman2023-01-271-0/+2
| | | | | | | | | | | | | | | | | | device_get_devnode() should take a constant * to struct device as it does not modify it in any way, so modify the function definition to do this and move it out of device.h as it does not need to be exposed to the whole kernel tree. Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Jason Gunthorpe <jgg@ziepe.ca> Cc: Ira Weiny <ira.weiny@intel.com> Cc: Dan Williams <dan.j.williams@intel.com> Cc: Won Chung <wonchung@google.com> Acked-by: Rafael J. Wysocki <rafael@kernel.org> Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Acked-by: Hans de Goede <hdegoede@redhat.com> Link: https://lore.kernel.org/r/20230111113018.459199-8-gregkh@linuxfoundation.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* driver core: bus: move bus notifier logic into bus.cGreg Kroah-Hartman2023-01-181-0/+1
| | | | | | | | | | | The logic to touch the bus notifier was open-coded in numberous places in the driver core. Clean that up by creating a local bus_notify() function and have everyone call this function instead, making the reading of the caller code simpler and easier to maintain over time. Reviewed-by: Rafael J. Wysocki <rafael@kernel.org> Link: https://lore.kernel.org/r/20230111092331.3946745-2-gregkh@linuxfoundation.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* driver core: change to_subsys_private() to use container_of_const()Greg Kroah-Hartman2023-01-111-1/+1
| | | | | | | | | | | | | The macro to_subsys_private() needs to switch to using container_of_const() as it turned out to being incorrectly casting a const pointer to a non-const one. Make this change and fix up the one offending user to be correctly handling a const pointer properly. Cc: "Rafael J. Wysocki" <rafael@kernel.org> Reviewed-by: Rafael J. Wysocki <rafael@kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Link: https://lore.kernel.org/r/20230111093327.3955063-1-gregkh@linuxfoundation.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* driver core: mark driver_allows_async_probing staticChristoph Hellwig2022-11-101-1/+0
| | | | | | | | | driver_allows_async_probing is only used in drivers/base/dd.c, so mark it static and remove the declaration in drivers/base/base.h. Signed-off-by: Christoph Hellwig <hch@lst.de> Link: https://lore.kernel.org/r/20221030092255.872280-1-hch@lst.de Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* driver core: remove make_class_name declarationGaosheng Cui2022-09-091-2/+0
| | | | | | | | | | make_class_name has been removed since commit 39aba963d937 ("driver core: remove CONFIG_SYSFS_DEPRECATED_V2 but keep it for block devices"), so remove it. Signed-off-by: Gaosheng Cui <cuigaosheng1@huawei.com> Link: https://lore.kernel.org/r/20220909063337.1146151-1-cuigaosheng1@huawei.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* driver core: Add wait_for_init_devices_probe helper functionSaravana Kannan2022-06-101-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Some devices might need to be probed and bound successfully before the kernel boot sequence can finish and move on to init/userspace. For example, a network interface might need to be bound to be able to mount a NFS rootfs. With fw_devlink=on by default, some of these devices might be blocked from probing because they are waiting on a optional supplier that doesn't have a driver. While fw_devlink will eventually identify such devices and unblock the probing automatically, it might be too late by the time it unblocks the probing of devices. For example, the IP4 autoconfig might timeout before fw_devlink unblocks probing of the network interface. This function is available to temporarily try and probe all devices that have a driver even if some of their suppliers haven't been added or don't have drivers. The drivers can then decide which of the suppliers are optional vs mandatory and probe the device if possible. By the time this function returns, all such "best effort" probes are guaranteed to be completed. If a device successfully probes in this mode, we delete all fw_devlink discovered dependencies of that device where the supplier hasn't yet probed successfully because they have to be optional dependencies. This also means that some devices that aren't needed for init and could have waited for their optional supplier to probe (when the supplier's module is loaded later on) would end up probing prematurely with limited functionality. So call this function only when boot would fail without it. Tested-by: Geert Uytterhoeven <geert+renesas@glider.be> Signed-off-by: Saravana Kannan <saravanak@google.com> Link: https://lore.kernel.org/r/20220601070707.3946847-5-saravanak@google.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* driver core: Extend deferred probe timeout on driver registrationSaravana Kannan2022-05-191-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The deferred probe timer that's used for this currently starts at late_initcall and runs for driver_deferred_probe_timeout seconds. The assumption being that all available drivers would be loaded and registered before the timer expires. This means, the driver_deferred_probe_timeout has to be pretty large for it to cover the worst case. But if we set the default value for it to cover the worst case, it would significantly slow down the average case. For this reason, the default value is set to 0. Also, with CONFIG_MODULES=y and the current default values of driver_deferred_probe_timeout=0 and fw_devlink=on, devices with missing drivers will cause their consumer devices to always defer their probes. This is because device links created by fw_devlink defer the probe even before the consumer driver's probe() is called. Instead of a fixed timeout, if we extend an unexpired deferred probe timer on every successful driver registration, with the expectation more modules would be loaded in the near future, then the default value of driver_deferred_probe_timeout only needs to be as long as the worst case time difference between two consecutive module loads. So let's implement that and set the default value to 10 seconds when CONFIG_MODULES=y. Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net> Cc: Rob Herring <robh@kernel.org> Cc: Linus Walleij <linus.walleij@linaro.org> Cc: Will Deacon <will@kernel.org> Cc: Ulf Hansson <ulf.hansson@linaro.org> Cc: Kevin Hilman <khilman@kernel.org> Cc: Thierry Reding <treding@nvidia.com> Cc: Mark Brown <broonie@kernel.org> Cc: Pavel Machek <pavel@ucw.cz> Cc: Geert Uytterhoeven <geert@linux-m68k.org> Cc: Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com> Cc: Paul Kocialkowski <paul.kocialkowski@bootlin.com> Cc: linux-gpio@vger.kernel.org Cc: linux-pm@vger.kernel.org Cc: iommu@lists.linux-foundation.org Reviewed-by: Mark Brown <broonie@kernel.org> Acked-by: Rob Herring <robh@kernel.org> Signed-off-by: Saravana Kannan <saravanak@google.com> Link: https://lore.kernel.org/r/20220429220933.1350374-1-saravanak@google.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* software nodes: Split software_node_notify()Rafael J. Wysocki2021-07-161-0/+3
| | | | | | | | | | | | | | | | Split software_node_notify_remove) out of software_node_notify() and make device_platform_notify() call the latter on device addition and the former on device removal. While at it, put the headers of the above functions into base.h, because they don't need to be present in a global header file. No intentional functional impact. Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Reviewed-by: Heikki Krogerus <heikki.krogerus@linux.intel.com> Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
* driver core: Export device_driver_attach()Jason Gunthorpe2021-06-211-1/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This is intended as a replacement API for device_bind_driver(). It has at least the following benefits: - Internal locking. Few of the users of device_bind_driver() follow the locking rules - Calls device driver probe() internally. Notably this means that devm support for probe works correctly as probe() error will call devres_release_all() - struct device_driver -> dev_groups is supported - Simplified calling convention, no need to manually call probe(). The general usage is for situations that already know what driver to bind and need to ensure the bind is synchronized with other logic. Call device_driver_attach() after device_add(). If probe() returns a failure then this will be preserved up through to the error return of device_driver_attach(). Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Cornelia Huck <cohuck@redhat.com> Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Link: https://lore.kernel.org/r/20210617142218.1877096-6-hch@lst.de Signed-off-by: Alex Williamson <alex.williamson@redhat.com>
* driver core: Improve fw_devlink & deferred_probe_timeout interactionSaravana Kannan2021-04-051-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | deferred_probe_timeout kernel commandline parameter allows probing of consumer devices if the supplier devices don't have any drivers. fw_devlink=on will indefintely block probe() calls on a device if all its suppliers haven't probed successfully. This completely skips calls to driver_deferred_probe_check_state() since that's only called when a .probe() function calls framework APIs. So fw_devlink=on breaks deferred_probe_timeout. deferred_probe_timeout in its current state also ignores a lot of information that's now available to the kernel. It assumes all suppliers that haven't probed when the timer expires (or when initcalls are done on a static kernel) will never probe and fails any calls to acquire resources from these unprobed suppliers. However, this assumption by deferred_probe_timeout isn't true under many conditions. For example: - If the consumer happens to be before the supplier in the deferred probe list. - If the supplier itself is waiting on its supplier to probe. This patch fixes both these issues by relaxing device links between devices only if the supplier doesn't have any driver that could match with (NOT bound to) the supplier device. This way, we only fail attempts to acquire resources from suppliers that truly don't have any driver vs suppliers that just happen to not have probed yet. Signed-off-by: Saravana Kannan <saravanak@google.com> Link: https://lore.kernel.org/r/20210402040342.2944858-3-saravanak@google.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* driver core: Update device link status properly for device_bind_driver()Saravana Kannan2021-03-231-0/+1
| | | | | | | | | | | | | | Device link status was not getting updated correctly when device_bind_driver() is called on a device. This causes a warning[1]. Fix this by updating device links that can be updated and dropping device links that can't be updated to a sensible state. [1] - https://lore.kernel.org/lkml/56f7d032-ba5a-a8c7-23de-2969d98c527e@nvidia.com/ Tested-by: Jon Hunter <jonathanh@nvidia.com> Signed-off-by: Saravana Kannan <saravanak@google.com> Link: https://lore.kernel.org/r/20210302211133.2244281-3-saravanak@google.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* driver core: auxiliary bus: Fix calling stage for auxiliary bus initDave Jiang2021-02-111-0/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When the auxiliary device code is built into the kernel, it can be executed before the auxiliary bus is registered. This causes bus->p to be not allocated and triggers a NULL pointer dereference when the auxiliary bus device gets added with bus_add_device(). Call the auxiliary_bus_init() under driver_init() so the bus is initialized before devices. Below is the kernel splat for the bug: [ 1.948215] BUG: kernel NULL pointer dereference, address: 0000000000000060 [ 1.950670] #PF: supervisor read access in kernel mode [ 1.950670] #PF: error_code(0x0000) - not-present page [ 1.950670] PGD 0 [ 1.950670] Oops: 0000 1 SMP NOPTI [ 1.950670] CPU: 0 PID: 1 Comm: swapper/0 Not tainted 5.10.0-intel-nextsvmtest+ #2205 [ 1.950670] Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.14.0-0-g155821a1990b-prebuilt.qemu.org 04/01/2014 [ 1.950670] RIP: 0010:bus_add_device+0x64/0x140 [ 1.950670] Code: 00 49 8b 75 20 48 89 df e8 59 a1 ff ff 41 89 c4 85 c0 75 7b 48 8b 53 50 48 85 d2 75 03 48 8b 13 49 8b 85 a0 00 00 00 48 89 de <48> 8 78 60 48 83 c7 18 e8 ef d9 a9 ff 41 89 c4 85 c0 75 45 48 8b [ 1.950670] RSP: 0000:ff46032ac001baf8 EFLAGS: 00010246 [ 1.950670] RAX: 0000000000000000 RBX: ff4597f7414aa680 RCX: 0000000000000000 [ 1.950670] RDX: ff4597f74142bbc0 RSI: ff4597f7414aa680 RDI: ff4597f7414aa680 [ 1.950670] RBP: ff46032ac001bb10 R08: 0000000000000044 R09: 0000000000000228 [ 1.950670] R10: ff4597f741141b30 R11: ff4597f740182a90 R12: 0000000000000000 [ 1.950670] R13: ffffffffa5e936c0 R14: 0000000000000000 R15: 0000000000000000 [ 1.950670] FS: 0000000000000000(0000) GS:ff4597f7bba00000(0000) knlGS:0000000000000000 [ 1.950670] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 1.950670] CR2: 0000000000000060 CR3: 000000002140c001 CR4: 0000000000f71ef0 [ 1.950670] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 1.950670] DR3: 0000000000000000 DR6: 00000000fffe07f0 DR7: 0000000000000400 [ 1.950670] PKRU: 55555554 [ 1.950670] Call Trace: [ 1.950670] device_add+0x3ee/0x850 [ 1.950670] __auxiliary_device_add+0x47/0x60 [ 1.950670] idxd_pci_probe+0xf77/0x1180 [ 1.950670] local_pci_probe+0x4a/0x90 [ 1.950670] pci_device_probe+0xff/0x1b0 [ 1.950670] really_probe+0x1cf/0x440 [ 1.950670] ? rdinit_setup+0x31/0x31 [ 1.950670] driver_probe_device+0xe8/0x150 [ 1.950670] device_driver_attach+0x58/0x60 [ 1.950670] __driver_attach+0x8f/0x150 [ 1.950670] ? device_driver_attach+0x60/0x60 [ 1.950670] ? device_driver_attach+0x60/0x60 [ 1.950670] bus_for_each_dev+0x79/0xc0 [ 1.950670] ? kmem_cache_alloc_trace+0x323/0x430 [ 1.950670] driver_attach+0x1e/0x20 [ 1.950670] bus_add_driver+0x154/0x1f0 [ 1.950670] driver_register+0x70/0xc0 [ 1.950670] __pci_register_driver+0x54/0x60 [ 1.950670] idxd_init_module+0xe2/0xfc [ 1.950670] ? idma64_platform_driver_init+0x19/0x19 [ 1.950670] do_one_initcall+0x4a/0x1e0 [ 1.950670] kernel_init_freeable+0x1fc/0x25c [ 1.950670] ? rest_init+0xba/0xba [ 1.950670] kernel_init+0xe/0x116 [ 1.950670] ret_from_fork+0x1f/0x30 [ 1.950670] Modules linked in: [ 1.950670] CR2: 0000000000000060 [ 1.950670] --[ end trace cd7d1b226d3ca901 ]-- Fixes: 7de3697e9cbd ("Add auxiliary bus support") Reported-by: Jacob Pan <jacob.jun.pan@intel.com> Reviewed-by: Dan Williams <dan.j.williams@intel.com> Acked-by: Dave Ertman <david.m.ertman@intel.com> Signed-off-by: Dave Jiang <dave.jiang@intel.com> Link: https://lore.kernel.org/r/20210210201611.1611074-1-dave.jiang@intel.com Cc: stable <stable@vger.kernel.org> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* driver core: make driver_probe_device() staticJulian Wiedmann2020-12-091-1/+0
| | | | | | | | It's only used inside drivers/base/dd.c Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com> Link: https://lore.kernel.org/r/20201123111938.18968-1-jwi@linux.ibm.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* Revert "driver core: fw_devlink: Add support for batching fwnode parsing"Saravana Kannan2020-12-091-1/+0
| | | | | | | | | | | | | | | | This reverts commit 716a7a25969003d82ab738179c3f1068a120ed11. The fw_devlink_pause/resume() APIs added by the commit being reverted were a first cut attempt at optimizing boot time. But these APIs don't fully solve the problem and are very fragile (can only be used for the top level devices being added). This series replaces them with a much better optimization that works for all device additions and also has the benefit of reducing the complexity of the firmware (DT, EFI) specific code and abstracting out common code to driver core. Signed-off-by: Saravana Kannan <saravanak@google.com> Link: https://lore.kernel.org/r/20201121020232.908850-7-saravanak@google.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* Revert "driver core: Don't do deferred probe in parallel with kernel_init ↵Saravana Kannan2020-12-091-0/+1
| | | | | | | | | | | | | | | | thread" This reverts commit cec72f3efc6272420c2c2c699607f03d09b93e41. Commit cec72f3efc62 ("driver core: Don't do deferred probe in parallel with kernel_init thread") was fixing a commit 716a7a259690 ("driver core: fw_devlink: Add support for batching fwnode parsing"). Since the commit being fixed itself is going to be reverted, the fix can also be reverted. Signed-off-by: Saravana Kannan <saravanak@google.com> Link: https://lore.kernel.org/r/20201121020232.908850-4-saravanak@google.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* driver core: add deferring probe reason to devices_deferred propertyAndrzej Hajda2020-07-301-0/+3
| | | | | | | | | | | | | | | | | | /sys/kernel/debug/devices_deferred property contains list of deferred devices. This list does not contain reason why the driver deferred probe, the patch improves it. The natural place to set the reason is dev_err_probe function introduced recently, ie. if dev_err_probe will be called with -EPROBE_DEFER instead of printk the message will be attached to a deferred device and printed when user reads devices_deferred property. Signed-off-by: Andrzej Hajda <a.hajda@samsung.com> Reviewed-by: Mark Brown <broonie@kernel.org> Reviewed-by: Javier Martinez Canillas <javierm@redhat.com> Reviewed-by: Andy Shevchenko <andy.shevchenko@gmail.com> Reviewed-by: Rafael J. Wysocki <rafael@kernel.org> Link: https://lore.kernel.org/r/20200713144324.23654-3-a.hajda@samsung.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* driver core: Don't do deferred probe in parallel with kernel_init threadSaravana Kannan2020-07-101-1/+0
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The current deferred probe implementation can mess up suspend/resume ordering if deferred probe thread is kicked off in parallel with the main initcall thread (kernel_init thread) [1]. For example: Say device-B is a consumer of device-A. Initcall thread Deferred probe thread =============== ===================== 1. device-A is added. 2. device-B is added. 3. dpm_list is now [device-A, device-B]. 4. driver-A defers probe of device-A. 5. device-A is moved to end of dpm_list 6. dpm_list is now [device-B, device-A] 7. driver-B is registereed and probes device-B. 8. dpm_list stays as [device-B, device-A]. The reverse order of dpm_list is used for suspend. So in this case device-A would incorrectly get suspended before device-B. Commit 716a7a259690 ("driver core: fw_devlink: Add support for batching fwnode parsing") kicked off the deferred probe thread early during boot to run in parallel with the initcall thread and caused suspend/resume regressions. This patch removes the parallel run of the deferred probe thread to avoid the suspend/resume regressions. [1] - https://lore.kernel.org/lkml/CAGETcx8W96KAw-d_siTX4qHB_-7ddk0miYRDQeHE6E0_8qx-6Q@mail.gmail.com/ Fixes: 716a7a259690 ("driver core: fw_devlink: Add support for batching fwnode parsing") Signed-off-by: Saravana Kannan <saravanak@google.com> Tested-by: Geert Uytterhoeven <geert+renesas@glider.be> Link: https://lore.kernel.org/r/20200701194259.3337652-2-saravanak@google.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* driver core: fw_devlink: Add support for batching fwnode parsingSaravana Kannan2020-05-151-0/+1
| | | | | | | | | | | | | | The amount of time spent parsing fwnodes of devices can become really high if the devices are added in an non-ideal order. Worst case can be O(N^2) when N devices are added. But this can be optimized to O(N) by adding all the devices and then parsing all their fwnodes in one batch. This commit adds fw_devlink_pause() and fw_devlink_resume() to allow doing this. Signed-off-by: Saravana Kannan <saravanak@google.com> Link: https://lore.kernel.org/r/20200515053500.215929-4-saravanak@google.com Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* device.h: move devtmpfs prototypes out of the fileGreg Kroah-Hartman2019-12-161-0/+8
| | | | | | | | | | | | The devtmpfs functions do not need to be in device.h as only the driver core uses them, so move them to the private .h file for the driver core. Cc: "Rafael J. Wysocki" <rafael@kernel.org> Cc: Suzuki K Poulose <suzuki.poulose@arm.com> Cc: Saravana Kannan <saravanak@google.com> Cc: Heikki Krogerus <heikki.krogerus@linux.intel.com> Link: https://lore.kernel.org/r/20191209193303.1694546-3-gregkh@linuxfoundation.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* drivers/base: base.h: add proper copyright and header infoGreg Kroah-Hartman2019-12-121-0/+11
| | | | | | | | | base.h didn't have any copyright information in it, so update it with the correct information. Cc: "Rafael J. Wysocki" <rafael@kernel.org> Link: https://lore.kernel.org/r/20191209193303.1694546-2-gregkh@linuxfoundation.org Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* driver/core: Convert to use built-in RCU list checkingJoel Fernandes (Google)2019-08-131-0/+1
| | | | | | | | | This commit applies the consolidated hlist_for_each_entry_rcu() support for lockdep conditions. Acked-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Signed-off-by: Joel Fernandes (Google) <joel@joelfernandes.org> Signed-off-by: Paul E. McKenney <paulmck@linux.ibm.com>
* driver core: Probe devices asynchronously instead of the driverAlexander Duyck2019-01-311-0/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Probe devices asynchronously instead of the driver. This results in us seeing the same behavior if the device is registered before the driver or after. This way we can avoid serializing the initialization should the driver not be loaded until after the devices have already been added. The motivation behind this is that if we have a set of devices that take a significant amount of time to load we can greatly reduce the time to load by processing them in parallel instead of one at a time. In addition, each device can exist on a different node so placing a single thread on one CPU to initialize all of the devices for a given driver can result in poor performance on a system with multiple nodes. This approach can reduce the time needed to scan SCSI LUNs significantly. The only way to realize that speedup is by enabling more concurrency which is what is achieved with this patch. To achieve this it was necessary to add a new member "async_driver" to the device_private structure to store the driver pointer while we wait on the deferred probe call. Reviewed-by: Bart Van Assche <bvanassche@acm.org> Reviewed-by: Dan Williams <dan.j.williams@intel.com> Signed-off-by: Alexander Duyck <alexander.h.duyck@linux.intel.com> Reviewed-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* device core: Consolidate locking and unlocking of parent and deviceAlexander Duyck2019-01-311-0/+2
| | | | | | | | | | | | | | | | | | | | | Try to consolidate all of the locking and unlocking of both the parent and device when attaching or removing a driver from a given device. To do that I first consolidated the lock pattern into two functions __device_driver_lock and __device_driver_unlock. After doing that I then created functions specific to attaching and detaching the driver while acquiring these locks. By doing this I was able to reduce the number of spots where we touch need_parent_lock from 12 down to 4. This patch should produce no functional changes, it is meant to be a code clean-up/consolidation only. Reviewed-by: Luis Chamberlain <mcgrof@kernel.org> Reviewed-by: Bart Van Assche <bvanassche@acm.org> Reviewed-by: Dan Williams <dan.j.williams@intel.com> Reviewed-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Alexander Duyck <alexander.h.duyck@linux.intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* driver core: Establish order of operations for device_add and device_del via ↵Alexander Duyck2019-01-311-0/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | bitflag Add an additional bit flag to the device_private struct named "dead". This additional flag provides a guarantee that when a device_del is executed on a given interface an async worker will not attempt to attach the driver following the earlier device_del call. Previously this guarantee was not present and could result in the device_del call attempting to remove a driver from an interface only to have the async worker attempt to probe the driver later when it finally completes the asynchronous probe call. One additional change added was that I pulled the check for dev->driver out of the __device_attach_driver call and instead placed it in the __device_attach_async_helper call. This was motivated by the fact that the only other caller of this, __device_attach, had already taken the device_lock() and checked for dev->driver. Instead of testing for this twice in this path it makes more sense to just consolidate the dev->dead and dev->driver checks together into one set of checks. Reviewed-by: Dan Williams <dan.j.williams@intel.com> Reviewed-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Alexander Duyck <alexander.h.duyck@linux.intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* driver core: move device->knode_class to device_privateWei Yang2019-01-181-0/+4
| | | | | | | | | | | | | | | As the description of struct device_private says, it stores data which is private to driver core. And it already has similar fields like: knode_parent, knode_driver, knode_driver and knode_bus. This look it is more proper to put knode_class together with those fields to make it private to driver core. This patch move device->knode_class to device_private to make it comply with code convention. Signed-off-by: Wei Yang <richardw.yang@linux.intel.com> Reviewed-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>