lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-Id: <1430752248-15401-1-git-send-email-johan@kernel.org>
Date:	Mon,  4 May 2015 17:10:25 +0200
From:	Johan Hovold <johan@...nel.org>
To:	Linus Walleij <linus.walleij@...aro.org>
Cc:	Alexandre Courbot <gnurou@...il.com>, linux-gpio@...r.kernel.org,
	linux-kernel@...r.kernel.org, Johan Hovold <johan@...nel.org>,
	Jonathan Corbet <corbet@....net>,
	Harry Wei <harryxiyou@...il.com>,
	Arnd Bergmann <arnd@...db.de>, linux-doc@...r.kernel.org,
	linux-kernel@...kernel.org, linux-arch@...r.kernel.org
Subject: [PATCH v2 00/23] gpio: sysfs: fixes and clean ups

These patches fix a number of issues with the gpio sysfs interface,
including

 - fix memory leaks and crashes on device hotplug
 - straighten out the convoluted locking
 - reduce sysfs-interface latencies through more fine-grained locking
 - more clearly separate the sysfs-interface implementation from gpiolib
   core

The first patch is marked for stable and could go into 4.1. [ May
already have been applied but not pushed by Linus, but included in v2
for completeness.  ]

Unfortunately we can't just kill the gpio sysfs interface, but these
patches will make it more manageable and should allow us to implement a
new user-space interface while maintaining the old one (for a while at
least) without losing our sanity.

Note that there is still a race between chip remove and gpiod_request (and
therefore sysfs export), which needs to be fixed separately (for instance as
part of a generic solution to chip hotplugging).

Johan


Changes since v1:
 - Keep explicit lock-as-irq call in sysfs and add comment that it
   should be removed once the broken drivers have been fixed (patch
   3/23).
 - Add comment that the class-device field of struct gpiochip is used by
   the sysfs interface (patch 6/23).
 - Add "sysfs"-infix to gpiochip sysfs registration functions as suggested
   by Alexandre (patch 7/23).


Johan Hovold (23):
  gpio: sysfs: fix memory leaks and device hotplug
  gpio: clean up gpiochip_remove
  gpio: sysfs: fix redundant lock-as-irq handling
  gpio: sysfs: preparatory clean ups
  gpio: sysfs: reduce gpiochip-export locking scope
  gpio: sysfs: clean up chip class-device handling
  gpio: sysfs: rename gpiochip registration functions
  gpio: remove gpiod_sysfs_set_active_low
  gpio: sysfs: use DEVICE_ATTR macros
  gpio: sysfs: release irq after class-device deregistration
  gpio: sysfs: remove redundant export tests
  gpio: sysfs: add gpiod class-device data
  gpio: sysfs: remove redundant gpio-descriptor parameters
  gpio: sysfs: clean up interrupt-interface implementation
  gpio: sysfs: only call irq helper if needed
  gpio: sysfs: split irq allocation and deallocation
  gpio: sysfs: clean up edge_store
  gpio: sysfs: clean up gpiod_export_link locking
  gpio: sysfs: use per-gpio locking
  gpio: sysfs: fix race between gpiod export and unexport
  gpio: sysfs: rename active-low helper
  gpio: sysfs: remove FLAG_SYSFS_DIR
  gpio: sysfs: move irq trigger flags to class-device data

 Documentation/gpio/gpio-legacy.txt |   9 -
 Documentation/gpio/sysfs.txt       |   8 -
 Documentation/zh_CN/gpio.txt       |   8 -
 drivers/gpio/gpiolib-sysfs.c       | 578 ++++++++++++++++++-------------------
 drivers/gpio/gpiolib.c             |  18 +-
 drivers/gpio/gpiolib.h             |  16 +-
 include/asm-generic/gpio.h         |   5 -
 include/linux/gpio.h               |   7 -
 include/linux/gpio/consumer.h      |   6 -
 include/linux/gpio/driver.h        |   4 +-
 10 files changed, 290 insertions(+), 369 deletions(-)

-- 
2.0.5

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ