stable.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	stable@vger.kernel.org, Logan Gunthorpe <logang@deltatee.com>,
	Dan Williams <dan.j.williams@intel.com>,
	Hans Verkuil <hans.verkuil@cisco.com>,
	Alexandre Belloni <alexandre.belloni@free-electrons.com>,
	Ben Hutchings <ben.hutchings@codethink.co.uk>,
	Sasha Levin <sashal@kernel.org>
Subject: [PATCH 4.9 37/90] chardev: add helper function to register char devs with a struct device
Date: Mon, 18 May 2020 19:36:15 +0200	[thread overview]
Message-ID: <20200518173458.652973043@linuxfoundation.org> (raw)
In-Reply-To: <20200518173450.930655662@linuxfoundation.org>

From: Logan Gunthorpe <logang@deltatee.com>

commit 233ed09d7fdacf592ee91e6c97ce5f4364fbe7c0 upstream.

Credit for this patch goes is shared with Dan Williams [1]. I've
taken things one step further to make the helper function more
useful and clean up calling code.

There's a common pattern in the kernel whereby a struct cdev is placed
in a structure along side a struct device which manages the life-cycle
of both. In the naive approach, the reference counting is broken and
the struct device can free everything before the chardev code
is entirely released.

Many developers have solved this problem by linking the internal kobjs
in this fashion:

cdev.kobj.parent = &parent_dev.kobj;

The cdev code explicitly gets and puts a reference to it's kobj parent.
So this seems like it was intended to be used this way. Dmitrty Torokhov
first put this in place in 2012 with this commit:

2f0157f char_dev: pin parent kobject

and the first instance of the fix was then done in the input subsystem
in the following commit:

4a215aa Input: fix use-after-free introduced with dynamic minor changes

Subsequently over the years, however, this issue seems to have tripped
up multiple developers independently. For example, see these commits:

0d5b7da iio: Prevent race between IIO chardev opening and IIO device
(by Lars-Peter Clausen in 2013)

ba0ef85 tpm: Fix initialization of the cdev
(by Jason Gunthorpe in 2015)

5b28dde [media] media: fix use-after-free in cdev_put() when app exits
after driver unbind
(by Shauh Khan in 2016)

This technique is similarly done in at least 15 places within the kernel
and probably should have been done so in another, at least, 5 places.
The kobj line also looks very suspect in that one would not expect
drivers to have to mess with kobject internals in this way.
Even highly experienced kernel developers can be surprised by this
code, as seen in [2].

To help alleviate this situation, and hopefully prevent future
wasted effort on this problem, this patch introduces a helper function
to register a char device along with its parent struct device.
This creates a more regular API for tying a char device to its parent
without the developer having to set members in the underlying kobject.

This patch introduce cdev_device_add and cdev_device_del which
replaces a common pattern including setting the kobj parent, calling
cdev_add and then calling device_add. It also introduces cdev_set_parent
for the few cases that set the kobject parent without using device_add.

[1] https://lkml.org/lkml/2017/2/13/700
[2] https://lkml.org/lkml/2017/2/10/370

Signed-off-by: Logan Gunthorpe <logang@deltatee.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Reviewed-by: Hans Verkuil <hans.verkuil@cisco.com>
Reviewed-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Ben Hutchings <ben.hutchings@codethink.co.uk>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 fs/char_dev.c        | 86 ++++++++++++++++++++++++++++++++++++++++++++
 include/linux/cdev.h |  5 +++
 2 files changed, 91 insertions(+)

diff --git a/fs/char_dev.c b/fs/char_dev.c
index 23e0477edf7d6..1bbb966c0783f 100644
--- a/fs/char_dev.c
+++ b/fs/char_dev.c
@@ -477,6 +477,85 @@ int cdev_add(struct cdev *p, dev_t dev, unsigned count)
 	return 0;
 }
 
+/**
+ * cdev_set_parent() - set the parent kobject for a char device
+ * @p: the cdev structure
+ * @kobj: the kobject to take a reference to
+ *
+ * cdev_set_parent() sets a parent kobject which will be referenced
+ * appropriately so the parent is not freed before the cdev. This
+ * should be called before cdev_add.
+ */
+void cdev_set_parent(struct cdev *p, struct kobject *kobj)
+{
+	WARN_ON(!kobj->state_initialized);
+	p->kobj.parent = kobj;
+}
+
+/**
+ * cdev_device_add() - add a char device and it's corresponding
+ *	struct device, linkink
+ * @dev: the device structure
+ * @cdev: the cdev structure
+ *
+ * cdev_device_add() adds the char device represented by @cdev to the system,
+ * just as cdev_add does. It then adds @dev to the system using device_add
+ * The dev_t for the char device will be taken from the struct device which
+ * needs to be initialized first. This helper function correctly takes a
+ * reference to the parent device so the parent will not get released until
+ * all references to the cdev are released.
+ *
+ * This helper uses dev->devt for the device number. If it is not set
+ * it will not add the cdev and it will be equivalent to device_add.
+ *
+ * This function should be used whenever the struct cdev and the
+ * struct device are members of the same structure whose lifetime is
+ * managed by the struct device.
+ *
+ * NOTE: Callers must assume that userspace was able to open the cdev and
+ * can call cdev fops callbacks at any time, even if this function fails.
+ */
+int cdev_device_add(struct cdev *cdev, struct device *dev)
+{
+	int rc = 0;
+
+	if (dev->devt) {
+		cdev_set_parent(cdev, &dev->kobj);
+
+		rc = cdev_add(cdev, dev->devt, 1);
+		if (rc)
+			return rc;
+	}
+
+	rc = device_add(dev);
+	if (rc)
+		cdev_del(cdev);
+
+	return rc;
+}
+
+/**
+ * cdev_device_del() - inverse of cdev_device_add
+ * @dev: the device structure
+ * @cdev: the cdev structure
+ *
+ * cdev_device_del() is a helper function to call cdev_del and device_del.
+ * It should be used whenever cdev_device_add is used.
+ *
+ * If dev->devt is not set it will not remove the cdev and will be equivalent
+ * to device_del.
+ *
+ * NOTE: This guarantees that associated sysfs callbacks are not running
+ * or runnable, however any cdevs already open will remain and their fops
+ * will still be callable even after this function returns.
+ */
+void cdev_device_del(struct cdev *cdev, struct device *dev)
+{
+	device_del(dev);
+	if (dev->devt)
+		cdev_del(cdev);
+}
+
 static void cdev_unmap(dev_t dev, unsigned count)
 {
 	kobj_unmap(cdev_map, dev, count);
@@ -488,6 +567,10 @@ static void cdev_unmap(dev_t dev, unsigned count)
  *
  * cdev_del() removes @p from the system, possibly freeing the structure
  * itself.
+ *
+ * NOTE: This guarantees that cdev device will no longer be able to be
+ * opened, however any cdevs already open will remain and their fops will
+ * still be callable even after cdev_del returns.
  */
 void cdev_del(struct cdev *p)
 {
@@ -576,5 +659,8 @@ EXPORT_SYMBOL(cdev_init);
 EXPORT_SYMBOL(cdev_alloc);
 EXPORT_SYMBOL(cdev_del);
 EXPORT_SYMBOL(cdev_add);
+EXPORT_SYMBOL(cdev_set_parent);
+EXPORT_SYMBOL(cdev_device_add);
+EXPORT_SYMBOL(cdev_device_del);
 EXPORT_SYMBOL(__register_chrdev);
 EXPORT_SYMBOL(__unregister_chrdev);
diff --git a/include/linux/cdev.h b/include/linux/cdev.h
index f8763615a5f2d..408bc09ce497b 100644
--- a/include/linux/cdev.h
+++ b/include/linux/cdev.h
@@ -4,6 +4,7 @@
 #include <linux/kobject.h>
 #include <linux/kdev_t.h>
 #include <linux/list.h>
+#include <linux/device.h>
 
 struct file_operations;
 struct inode;
@@ -26,6 +27,10 @@ void cdev_put(struct cdev *p);
 
 int cdev_add(struct cdev *, dev_t, unsigned);
 
+void cdev_set_parent(struct cdev *p, struct kobject *kobj);
+int cdev_device_add(struct cdev *cdev, struct device *dev);
+void cdev_device_del(struct cdev *cdev, struct device *dev);
+
 void cdev_del(struct cdev *);
 
 void cd_forget(struct inode *);
-- 
2.20.1




  parent reply	other threads:[~2020-05-18 17:43 UTC|newest]

Thread overview: 95+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-05-18 17:35 [PATCH 4.9 00/90] 4.9.224-rc1 review Greg Kroah-Hartman
2020-05-18 17:35 ` [PATCH 4.9 01/90] USB: serial: qcserial: Add DW5816e support Greg Kroah-Hartman
2020-05-18 17:35 ` [PATCH 4.9 02/90] dp83640: reverse arguments to list_add_tail Greg Kroah-Hartman
2020-05-18 17:35 ` [PATCH 4.9 03/90] fq_codel: fix TCA_FQ_CODEL_DROP_BATCH_SIZE sanity checks Greg Kroah-Hartman
2020-05-18 17:35 ` [PATCH 4.9 04/90] net: macsec: preserve ingress frame ordering Greg Kroah-Hartman
2020-05-18 17:35 ` [PATCH 4.9 05/90] net/mlx4_core: Fix use of ENOSPC around mlx4_counter_alloc() Greg Kroah-Hartman
2020-05-18 17:35 ` [PATCH 4.9 06/90] net: usb: qmi_wwan: add support for DW5816e Greg Kroah-Hartman
2020-05-18 17:35 ` [PATCH 4.9 07/90] sch_choke: avoid potential panic in choke_reset() Greg Kroah-Hartman
2020-05-18 17:35 ` [PATCH 4.9 08/90] sch_sfq: validate silly quantum values Greg Kroah-Hartman
2020-05-18 17:35 ` [PATCH 4.9 09/90] bnxt_en: Fix VLAN acceleration handling in bnxt_fix_features() Greg Kroah-Hartman
2020-05-18 17:35 ` [PATCH 4.9 10/90] net/mlx5: Fix forced completion access non initialized command entry Greg Kroah-Hartman
2020-05-18 17:35 ` [PATCH 4.9 11/90] net/mlx5: Fix command entry leak in Internal Error State Greg Kroah-Hartman
2020-05-18 17:35 ` [PATCH 4.9 12/90] bnxt_en: Improve AER slot reset Greg Kroah-Hartman
2020-05-18 17:35 ` [PATCH 4.9 13/90] Revert "ACPI / video: Add force_native quirk for HP Pavilion dv6" Greg Kroah-Hartman
2020-05-18 17:35 ` [PATCH 4.9 14/90] binfmt_elf: move brk out of mmap when doing direct loader exec Greg Kroah-Hartman
2020-05-18 17:35 ` [PATCH 4.9 15/90] USB: uas: add quirk for LaCie 2Big Quadra Greg Kroah-Hartman
2020-05-18 17:35 ` [PATCH 4.9 16/90] USB: serial: garmin_gps: add sanity checking for data length Greg Kroah-Hartman
2020-05-18 17:35 ` [PATCH 4.9 17/90] tracing: Add a vmalloc_sync_mappings() for safe measure Greg Kroah-Hartman
2020-05-18 17:35 ` [PATCH 4.9 18/90] mm/page_alloc: fix watchdog soft lockups during set_zone_contiguous() Greg Kroah-Hartman
2020-05-18 17:35 ` [PATCH 4.9 19/90] batman-adv: fix batadv_nc_random_weight_tq Greg Kroah-Hartman
2020-05-18 17:35 ` [PATCH 4.9 20/90] batman-adv: Fix refcnt leak in batadv_show_throughput_override Greg Kroah-Hartman
2020-05-18 17:35 ` [PATCH 4.9 21/90] batman-adv: Fix refcnt leak in batadv_store_throughput_override Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 22/90] batman-adv: Fix refcnt leak in batadv_v_ogm_process Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 23/90] objtool: Fix stack offset tracking for indirect CFAs Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 24/90] scripts/decodecode: fix trapping instruction formatting Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 25/90] binfmt_elf: Do not move brk for INTERP-less ET_EXEC Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 26/90] ext4: add cond_resched() to ext4_protect_reserved_inode Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 27/90] net: ipv6: add net argument to ip6_dst_lookup_flow Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 28/90] net: ipv6_stub: use ip6_dst_lookup_flow instead of ip6_dst_lookup Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 29/90] blktrace: Fix potential deadlock between delete & sysfs ops Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 30/90] blktrace: fix unlocked access to init/start-stop/teardown Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 31/90] blktrace: fix trace mutex deadlock Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 32/90] blktrace: Protect q->blk_trace with RCU Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 33/90] blktrace: fix dereference after null check Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 34/90] ptp: do not explicitly set drvdata in ptp_clock_register() Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 35/90] ptp: use is_visible method to hide unused attributes Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 36/90] ptp: create "pins" together with the rest of attributes Greg Kroah-Hartman
2020-05-18 17:36 ` Greg Kroah-Hartman [this message]
2020-05-18 17:36 ` [PATCH 4.9 38/90] ptp: Fix pass zero to ERR_PTR() in ptp_clock_register Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 39/90] ptp: fix the race between the release of ptp_clock and cdev Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 40/90] ptp: free ptp device pin descriptors properly Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 41/90] shmem: fix possible deadlocks on shmlock_user_lock Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 42/90] net/sonic: Fix a resource leak in an error handling path in jazz_sonic_probe() Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 43/90] net: moxa: Fix a potential double free_irq() Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 44/90] drop_monitor: work around gcc-10 stringop-overflow warning Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 45/90] scsi: sg: add sg_remove_request in sg_write Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 46/90] spi: spi-dw: Add lock protect dw_spi rx/tx to prevent concurrent calls Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 47/90] cifs: Check for timeout on Negotiate stage Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 48/90] cifs: Fix a race condition with cifs_echo_request Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 49/90] dmaengine: pch_dma.c: Avoid data race between probe and irq handler Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 50/90] dmaengine: mmp_tdma: Reset channel error on release Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 51/90] ALSA: hda/hdmi: fix race in monitor detection during probe Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 52/90] drm/qxl: lost qxl_bo_kunmap_atomic_page in qxl_image_init_helper() Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 53/90] ipc/util.c: sysvipc_find_ipc() incorrectly updates position index Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 54/90] pinctrl: cherryview: Add missing spinlock usage in chv_gpio_irq_handler Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 55/90] i40iw: Fix error handling in i40iw_manage_arp_cache() Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 56/90] netfilter: conntrack: avoid gcc-10 zero-length-bounds warning Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 57/90] IB/mlx4: Test return value of calls to ib_get_cached_pkey Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 58/90] pnp: Use list_for_each_entry() instead of open coding Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 59/90] gcc-10 warnings: fix low-hanging fruit Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 60/90] kbuild: compute false-positive -Wmaybe-uninitialized cases in Kconfig Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 61/90] Stop the ad-hoc games with -Wno-maybe-initialized Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 62/90] net: phy: micrel: Use strlcpy() for ethtool::get_strings Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 63/90] gcc-10: avoid shadowing standard library free() in crypto Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 64/90] gcc-10: disable zero-length-bounds warning for now Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 65/90] gcc-10: disable array-bounds " Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 66/90] gcc-10: disable stringop-overflow " Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 67/90] gcc-10: disable restrict " Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 68/90] net: fix a potential recursive NETDEV_FEAT_CHANGE Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 69/90] netlabel: cope with NULL catmap Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 70/90] Revert "ipv6: add mtu lock check in __ip6_rt_update_pmtu" Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 71/90] net: ipv4: really enforce backoff for redirects Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 72/90] netprio_cgroup: Fix unlimited memory leak of v2 cgroups Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 73/90] ALSA: hda/realtek - Limit int mic boost for Thinkpad T530 Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 74/90] ALSA: rawmidi: Initialize allocated buffers Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 75/90] ALSA: rawmidi: Fix racy buffer resize under concurrent accesses Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 76/90] ALSA: usb-audio: Add control message quirk delay for Kingston HyperX headset Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 77/90] USB: gadget: fix illegal array access in binding with UDC Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 78/90] usb: xhci: Fix NULL pointer dereference when enqueuing trbs from urb sg list Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 79/90] ARM: dts: imx27-phytec-phycard-s-rdk: Fix the I2C1 pinctrl entries Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 80/90] x86: Fix early boot crash on gcc-10, third try Greg Kroah-Hartman
2020-05-18 17:36 ` [PATCH 4.9 81/90] exec: Move would_dump into flush_old_exec Greg Kroah-Hartman
2020-05-18 17:37 ` [PATCH 4.9 82/90] usb: gadget: net2272: Fix a memory leak in an error handling path in net2272_plat_probe() Greg Kroah-Hartman
2020-05-18 17:37 ` [PATCH 4.9 83/90] usb: gadget: audio: Fix a missing error return value in audio_bind() Greg Kroah-Hartman
2020-05-18 17:37 ` [PATCH 4.9 84/90] usb: gadget: legacy: fix error return code in gncm_bind() Greg Kroah-Hartman
2020-05-18 17:37 ` [PATCH 4.9 85/90] usb: gadget: legacy: fix error return code in cdc_bind() Greg Kroah-Hartman
2020-05-18 17:37 ` [PATCH 4.9 86/90] Revert "ALSA: hda/realtek: Fix pop noise on ALC225" Greg Kroah-Hartman
2020-05-18 17:37 ` [PATCH 4.9 87/90] ARM: dts: r8a73a4: Add missing CMT1 interrupts Greg Kroah-Hartman
2020-05-18 17:37 ` [PATCH 4.9 88/90] ARM: dts: r8a7740: Add missing extal2 to CPG node Greg Kroah-Hartman
2020-05-18 17:37 ` [PATCH 4.9 89/90] KVM: x86: Fix off-by-one error in kvm_vcpu_ioctl_x86_setup_mce Greg Kroah-Hartman
2020-05-18 17:37 ` [PATCH 4.9 90/90] Makefile: disallow data races on gcc-10 as well Greg Kroah-Hartman
2020-05-19  8:19 ` [PATCH 4.9 00/90] 4.9.224-rc1 review Naresh Kamboju
2020-05-19  8:49 ` Jon Hunter
2020-05-19 15:08 ` shuah
2020-05-19 16:27 ` Guenter Roeck

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20200518173458.652973043@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=alexandre.belloni@free-electrons.com \
    --cc=ben.hutchings@codethink.co.uk \
    --cc=dan.j.williams@intel.com \
    --cc=hans.verkuil@cisco.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=logang@deltatee.com \
    --cc=sashal@kernel.org \
    --cc=stable@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).