Commit Graph

137903 Commits

Author SHA1 Message Date
Douglas Kosovic
ade0815c16 V4L/DVB (10299): bttv: Add support for IVCE-8784 support for V4L2 bttv driver
It's a quad Bt878 PCI-e x1 capture board that's basically the same as the
IVC-200 (quad Bt878 PCI) capture board that's currently supported in
the V4L2 bttv driver.

Manufacturer's web page for IVCE-8784 with photo and info:
  http://www.iei.com.tw/en/product_IPC.asp?model=IVCE-8784

Signed-off-by: Douglas Kosovic <douglask@itee.uq.edu.au>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2009-03-30 12:42:26 -03:00
Mauro Carvalho Chehab
be9ed5117d V4L/DVB (10298): remove err macro from few usb devices
Patch removes err() macros from few usb devices.
It places pr_err in pvrusb2-v4l2.c, dev_err in dabusb and in usbvision
drivers. Beside placing dev_err, patch defines new s2255_dev_err macro
with S2255_DRIVER_NAME in s2255 module.

Signed-off-by: Alexey Klimov <klimov.linux@gmail.com>
Acked-by: Thierry Merle <thierry.merle@free.fr>
Acked-by: Mike Isely <isely@pobox.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2009-03-30 12:42:26 -03:00
Laurent Pinchart
f180152376 V4L/DVB (10296): uvcvideo: Fix memory leak in input device handling
The dynamically allocated input_dev->phys buffer isn't freed when
unregistering the device. As the input layer doesn't provide any release
callback, use a fixed-size buffer inside the uvc_device structure.

Signed-off-by: Laurent Pinchart <laurent.pinchart@skynet.be>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2009-03-30 12:42:26 -03:00
Laurent Pinchart
efdc8a9585 V4L/DVB (10295): uvcvideo: Retry URB buffers allocation when the system is low on memory.
URB buffers for video transfers are sized to UVC_MAX_PACKETS bulk/isochronous
packets by default. If the system is too low on memory try successively
smaller numbers of packets until allocation succeeds.

Tested-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: Laurent Pinchart <laurent.pinchart@skynet.be>
Reviewed-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2009-03-30 12:42:26 -03:00
Laurent Pinchart
f61d1d8a56 V4L/DVB (10294): uvcvideo: Add support for the Alcor Micro AU3820 chipset.
The Alcor Micro AU3820 chipset (found in the Future Boy PC USB webcam)
requires the MINMAX quirk. Add a corresponding entry in the device IDs list.

Signed-off-by: Laurent Pinchart <laurent.pinchart@skynet.be>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2009-03-30 12:42:26 -03:00
Laurent Pinchart
d0ebf3073f V4L/DVB (10293): uvcvideo: replace strn{cpy,cat} with strl{cpy,cat}.
strncpy is unsafe as it doesn't append a terminating NUL character when the
source string doesn't fit in the destination buffer. Replace it with strlcpy.
strncat is misused as its size argument refers to the source string, not the
destination buffer. Replace it with strlcat.

Signed-off-by: Laurent Pinchart <laurent.pinchart@skynet.be>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2009-03-30 12:42:26 -03:00
Hans Verkuil
07f7db4ce7 V4L/DVB (10291): em28xx: fix VIDIOC_G_CTRL when there is no msp34xx device.
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2009-03-30 12:42:26 -03:00
Antti Palosaari
71bf2e08ce V4L/DVB (10286): af9015: add new USB ID for KWorld DVB-T 395U
Add new USB ID 1b80:e39b for KWorld DVB-T 395U. This device revision does
have Quantek QT1010 silicon tuner.

Thanks-to: Ray Chen <swyear@gmail.com>
Signed-off-by: Antti Palosaari <crope@iki.fi>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2009-03-30 12:42:25 -03:00
Andy Walls
6ce9ee539c V4L/DVB (10284): cx18: Add initial entry for a Leadtek DVR3100 H hybrid card
Card report provided by Patryk on the ivtv-devel list:
http://ivtvdriver.org/pipermail/ivtv-devel/2009-January/005922.html

Reported-by: Pat <tatyrza@mm.pl>
Signed-off-by: Andy Walls <awalls@radix.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2009-03-30 12:42:25 -03:00
Andy Walls
be411df610 V4L/DVB (10283): cx18: Call request_module() with proper argument types.
request_module() needs to be called with a string literal for a
format string or with 1 or more variable arguments to avoid
compiler warnings and possible exploits, if someone could cause
us to get a format string with a '%' code in the format string
when we make the call.

Reported-by: Brandon Jenkins <bcjenkins@tvwhere.com>
Signed-off-by: Andy Walls <awalls@radix.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2009-03-30 12:42:25 -03:00
Andy Walls
888cdb0774 V4L/DVB (10281): cx18: Conversion to new V4L2 framework: use v4l2_device object
First step in conversion to the new V4L2 framework.  Added per cx18 device
instance of the v4l2_device and its registration.

Signed-off-by: Andy Walls <awalls@radix.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2009-03-30 12:42:25 -03:00
Andy Walls
3d05913d89 V4L/DVB (10280): cx18: Rename structure members: dev to pci_dev and v4l2dev to video_dev
Renamed structure member name to be more specific to type in anticipation
of updating to the v4l2_device/v4l2_subdev framework.  Too many objects named
"dev" and /v4l2_\{0,1\}dev/ would be to confusing.

Signed-off-by: Andy Walls <awalls@radix.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2009-03-30 12:42:25 -03:00
Andy Walls
e0f28b6a69 V4L/DVB (10279): cx18: Print driver version number when logging status
Make sure v4l2-ctl --log-status outputs the driver version.

Reported-by: David Dombroski <davedom@optonline.net>
Signed-off-by: Andy Walls <awalls@radix.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2009-03-30 12:42:25 -03:00
Andy Walls
350145a4f7 V4L/DVB (10278): cx18: Fix bad audio in first analog capture.
Normalize the APU state before the second firmware load so that audio
for the first analog capture is correct.

Many thanks to Conexant for supporting me in finding a solution for this
problem.

Signed-off-by: Andy Walls <awalls@radix.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2009-03-30 12:42:25 -03:00
Andy Walls
31230c5f6f V4L/DVB (10277): cx18, cx2341x: Fix bugs in cx18 AC3 control and comply with V4L2 spec
Fix bugs in the cx18 AC3 control implementation that would have affected
ivtv and other drivers via the cx2341x module.  Bring AC3 controls
behavior into comliance with V4L2 specification.  Thanks to Hans Verkuil
for reviewing the previous patch and pointing out the problems.

Reported-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Andy Walls <awalls@radix.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2009-03-30 12:42:25 -03:00
Andy Walls
0d82fe801d V4L/DVB (10276): cx18, cx2341x, ivtv: Add AC-3 audio encoding control to cx18
Initial addition of controls to set AC-3 audio encoding for the CX23418 - it
does not work yet due to firmware or cx18 driver issues.  This change affects
the common cx2341x and ivtv modules due to shared structures and
common functions.

Signed-off-by: Andy Walls <awalls@radix.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2009-03-30 12:42:24 -03:00
Andy Walls
5029999418 V4L/DVB (10275): cx18: Additional debug to display outgoing mailbox parameters
Added debug display of outgoing mailbox arguments.  Fixed a minor problem
with display of stale incoming mailbox contents, when user was not looking for
debug warnings.

Signed-off-by: Andy Walls <awalls@radix.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2009-03-30 12:42:24 -03:00
Andy Walls
903bfeac50 V4L/DVB (10274): cx18: Fix a PLL divisor update for the I2S master clock
A redundant PLL divisior update for the I2S master clock after AV core
firmware load was missed in earlier PLL parameter changes.  This one really
doesn't matter because it's redundant and gets overwritten, but the driver
should be self consistent in the values used.

Signed-off-by: Andy Walls <awalls@radix.net>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2009-03-30 12:42:24 -03:00
Hans Verkuil
f97d2074e3 V4L/DVB (10272): av7110: test type field in VIDIOC_G_SLICED_VBI_CAP
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2009-03-30 12:42:24 -03:00
Hans Verkuil
b960074fec V4L/DVB (10271): saa7146: convert to video_ioctl2.
The conversion to video_ioctl2 is the first phase to converting this driver
to the latest v4l2 framework.

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2009-03-30 12:42:24 -03:00
Igor M. Liplianin
c9b8b04b26 V4L/DVB (10269): Add support for DVBWorld DVBS2 PCI-e 2005.
DVBWorld DVBS2 PCI-e 2005 card contains cx23885 PCI-e bridge
and cx24116 demodulator.
http://www.linuxtv.org/wiki/index.php/DVBWorld_DVB-S2_2005_PCI-Express_Card
The card tested by me (Igor).

Signed-off-by: Igor M. Liplianin <liplianin@me.by>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2009-03-30 12:42:24 -03:00
Igor M. Liplianin
c7bdcd0f54 V4L/DVB (10268): Proper implement set_voltage in cx24116.
Now there is a card, which uses cx24116 to control LNB DC voltage.
It is DVBWorld DVBS2 PCI-e 2005.
The patch is nedded to add support for that card.
1. Rename CMD_SET_TONEPRE to CMD_LNBDCLEVEL.
2. Fill set_voltage with actually control voltage code.
3. Correct set_tone to not affect voltage.

Signed-off-by: Igor M. Liplianin <liplianin@me.by>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2009-03-30 12:42:24 -03:00
Igor M. Liplianin
579943f548 V4L/DVB (10267): Add support for TeVii S470 DVB-S2 PCI-e card.
TeVii S470 DVB-S2 PCI-e card contains cx23885 PCI-e bridge
and cx24116 demodulator.
http://www.linuxtv.org/wiki/index.php/TeVii_S470
The card tested by me (Igor).

Signed-off-by: Igor M. Liplianin <liplianin@me.by>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2009-03-30 12:42:23 -03:00
Igor M. Liplianin
96318d0cca V4L/DVB (10266): Add support for TurboSight TBS6920 DVB-S2 PCI-e card.
TurboSight TBS6920 DVB-S2 PCI-e card contains cx23885 PCI-e bridge
and cx24116 demodulator.
http://www.linuxtv.org/wiki/index.php/TBS_6920
The card tested by me (Igor).

Signed-off-by: Igor M. Liplianin <liplianin@me.by>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2009-03-30 12:42:23 -03:00
Antoine Jacquet
c76b638ca2 V4L/DVB (10263): zr364xx: add support for Aiptek DV T300
Tested-by: Hámorszky Balázs <balihb@freepop.hu>
Signed-off-by: Antoine Jacquet <royale@zerezo.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2009-03-30 12:42:23 -03:00
Mike Isely
94b5ff9cf3 V4L/DVB (10259): pvrusb2: Code module name directly in printk
The name of the pvrusb2 module is not likely to ever change, and there
are plenty of other places where the name is directly coded, so there
is little utility in using a macro to infer the module name here.  In
addition, using that macro complicates other uses of the driver
involving older kernels where this macro works differently.  Yes I
know for many places we don't have to worry about that.  But my
alternative is that I have to build special logic in the pvrusb2
standalone driver to special-case what is otherwise costmetic and that
is just plain nuts for something as trivial as this, especially since
this change does not at all have any compile time or run time impact
on the driver.  I'm just removing a nicety that didn't have a lot of
value here to begin with.

Signed-off-by: Mike Isely <isely@pobox.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2009-03-30 12:42:23 -03:00
Mike Isely
2cc3b6bff4 V4L/DVB (10258): pvrusb2: Issue VIDIOC_INT_INIT to v4l2 modules when they first attach
It appears that various v4l-dvb drivers are changing to require
explicit initialization before use.  This change to the pvrusb2 driver
implements an automatic issuance of VIDIOC_INT_INIT when a module is
bound to the driver, thus conforming to the new behavior.

Signed-off-by: Mike Isely <isely@pobox.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2009-03-30 12:42:23 -03:00
Robert Krakora
8ed06fd472 V4L/DVB (10255): em28xx: Clock (XCLK) Cleanup
Clock (XCLK) Cleanup

Signed-off-by: Robert Krakora <rob.krakora@messagenetsystems.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2009-03-30 12:42:23 -03:00
Hans Verkuil
f5360bdc55 V4L/DVB (10252): v4l2 doc: explain why v4l2_device_unregister_subdev() has to be called.
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2009-03-30 12:42:23 -03:00
Hans Verkuil
6ca187abb2 V4L/DVB (10251): cx25840: add comments explaining what the init() does.
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2009-03-30 12:42:23 -03:00
Hans Verkuil
c7d29e2f53 V4L/DVB (10249): v4l2-common: added v4l2_i2c_tuner_addrs()
Add v4l2_i2c_tuner_addrs() to obtain the various I2C tuner addresses.
This will be used in several drivers, so make this a common function
as we do not want to have these I2C addresses all over the place.

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2009-03-30 12:42:22 -03:00
Hans Verkuil
fac6986c47 V4L/DVB (10247): saa7134: convert to the new v4l2 framework.
Register v4l2_device and switch to v4l2_subdev to access the i2c modules.

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2009-03-30 12:42:22 -03:00
Hans Verkuil
5b73e98c83 V4L/DVB (10246): saa6752hs: convert to v4l2_subdev.
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2009-03-30 12:42:22 -03:00
Hans Verkuil
b634a93f78 V4L/DVB (10244): v4l2: replace a few snprintfs with strlcpy
Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2009-03-30 12:42:22 -03:00
Nam Phạm Thành
e32a7eccd7 V4L/DVB (10242): pwc: add support for webcam snapshot button
This patch adds support for Philips webcam snapshot button as an
event input device, for consistency with other webcam drivers.

Signed-off-by: Pham Thanh Nam <phamthanhnam.ptn@gmail.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2009-03-30 12:42:22 -03:00
Mike Isely
be4f4aecf8 V4L/DVB (10239): pvrusb2: Fix misleading comment caused by earlier commit
Previous v4l-dvb changeset id 4cc8ed11e2e0 changed the pvrusb2-hdw
internal API regarding i2c chip debug register access.  However that
change failed to also update the corresponding function comment
describing the API.  As driver maintained I never saw a request for an
ack on that change; there probably should have been one especially
since the manner in which this API operates was changed - its
interface is now entangled with a v4l specific struct and I would have
preferred to keep this API clear of moving-target v4l-isms such as
this one if at all possible which is why I had done it the way I did
before.  But whatever.  This commit at least fixes the comment issue.

Signed-off-by: Mike Isely <isely@pobox.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2009-03-30 12:42:22 -03:00
Mike Isely
730e92c9ac V4L/DVB (10238): pvrusb2: Change sysfs serial number handling
Use the new pvrusb2 internal API to grab the device identifier, rather
than generating it directly.  This unifies some code and make possible
use of that identifier in places other than sysfs.

Signed-off-by: Mike Isely <isely@pobox.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2009-03-30 12:42:22 -03:00
Mike Isely
13a887971b V4L/DVB (10237): pvrusb2: Generate a device-unique identifier
Implement a new internal function to create a string device
identifier.  This ID stays with the specific device, making it useful
to user space to identify specific devices.  We use the serial number
if available; otherwise we give up and just spit out a unit/instance ID.

Signed-off-by: Mike Isely <isely@pobox.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2009-03-30 12:42:21 -03:00
Mike Isely
d166b02ea6 V4L/DVB (10236): pvrusb2: Stop advertising VBI capability - it isn't there
Signed-off-by: Mike Isely <isely@pobox.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2009-03-30 12:42:21 -03:00
Hans Verkuil
952617f259 V4L/DVB (10231): v4l2-subdev: add v4l2_ext_controls support
The saa6752hs module needs this.

Signed-off-by: Hans Verkuil <hverkuil@xs4all.nl>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2009-03-30 12:42:20 -03:00
Mauro Carvalho Chehab
e164b58a84 V4L/DVB (10211): vivi: Implements 4 inputs on vivi
This patch adds the capability of selecting between 4 different inputs
on vivi driver. Input 0 is the normal color bar, while inputs 1-3 are
modified color bars.

This allows testing input selection on userspace applications and
serves as an implementation model for other drivers.

The current approach allows a maximum of 10 different inputs, since the
input name generator assumes that we need just one digit to present the
input. It shouldn't be hard to modify it to present a bigger name of
inputs, but, in fact, it doesn't make much sense of doing it for this
test driver.

Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
2009-03-30 12:42:20 -03:00
Linus Torvalds
0d34fb8e93 Merge branch 'bzip2-lzma-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/x86/linux-2.6-tip
* 'bzip2-lzma-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/x86/linux-2.6-tip:
  bzip2/lzma: don't ask for compression mode for the default initramfs
  bzip2/lzma: consistently capitalize LZMA in Kconfig
  bzip2/lzma: clarify the meaning of the CONFIG_RD_ options
  bzip2/lzma: move CONFIG_RD_* options under CONFIG_EMBEDDED
2009-03-29 13:05:41 -07:00
Ralf Baechle
503e576303 Fix build error in <linux/irq.h>
<linux/irq.h> relies on <linux/gfp.h> and <linux/topology.h> having been
included previous.  If not, the errors like below will result.

    CC      arch/mips/mti-malta/malta-int.o
  In file included from arch/mips/mti-malta/malta-int.c:25:
  include/linux/irq.h: In function ‘init_alloc_desc_masks’:
  include/linux/irq.h:444: error: implicit declaration of function ‘cpu_to_node’
  include/linux/irq.h:446: error: ‘GFP_ATOMIC’ undeclared (first use in this function)
  include/linux/irq.h:446: error: (Each undeclared identifier is reported only once
  include/linux/irq.h:446: error: for each function it appears in.)
  make[3]: *** [arch/mips/mti-malta/malta-int.o] Error 1
  make[2]: *** [arch/mips/mti-malta] Error 2
  make[1]: *** [sub-make] Error 2

Fixed by including the two missing headers.

Signed-off-by: Ralf Baechle <ralf@linux-mips.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-03-29 08:19:09 -07:00
Randy Dunlap
d5ac537e5f sched: fix errors in struct & function comments
Fix kernel-doc errors in sched.c:  the structs don't have
kernel-doc notation and the short function description needs to
be one line only.

  Error(kernel/sched.c:3197): cannot understand prototype: 'struct sd_lb_stats '
  Error(kernel/sched.c:3228): cannot understand prototype: 'struct sg_lb_stats '
  Error(kernel/sched.c:3375): duplicate section name 'Description'

Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-03-29 08:12:39 -07:00
Randy Dunlap
ee665ecca6 maple: fix Error in kernel-doc notation
Fix kernel-doc error in maple (it's not kernel-doc):

  Error(drivers/sh/maple/maple.c:782): cannot understand prototype: 'struct bus_type maple_bus_type = '

Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
cc: Paul Mundt <lethal@linux-sh.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-03-29 08:12:39 -07:00
Linus Torvalds
c31f403de6 Merge branch 'futexes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip
* 'futexes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
  futex: remove the pointer math from double_unlock_hb, fix
  futex: remove the pointer math from double_unlock_hb
  futex: clean up fault logic
  futex: unlock before returning -EFAULT
  futex: use current->time_slack_ns for rt tasks too
  futex: add double_unlock_hb()
  futex: additional (get|put)_futex_key() fixes
  futex: update futex commentary
2009-03-28 17:32:14 -07:00
Linus Torvalds
7541bba880 Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jmorris/security-testing-2.6
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jmorris/security-testing-2.6:
  smack: Add a new '-CIPSO' option to the network address label configuration
  netlabel: Cleanup the Smack/NetLabel code to fix incoming TCP connections
  lsm: Remove the socket_post_accept() hook
  selinux: Remove the "compat_net" compatibility code
  netlabel: Label incoming TCP connections correctly in SELinux
  lsm: Relocate the IPv4 security_inet_conn_request() hooks
  TOMOYO: Fix a typo.
  smack: convert smack to standard linux lists
2009-03-28 17:30:42 -07:00
David Howells
795e2fe0a3 Annotate struct fs_struct's usage count restriction
Annotate struct fs_struct's usage count to indicate the restrictions upon it.
It may not be incremented, except by clone(CLONE_FS), as this affects the
check in check_unsafe_exec() in fs/exec.c.

Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Cc: stable@kernel.org
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-03-28 17:30:00 -07:00
Hugh Dickins
7c2c7d9930 fix setuid sometimes wouldn't
check_unsafe_exec() also notes whether the fs_struct is being
shared by more threads than will get killed by the exec, and if so
sets LSM_UNSAFE_SHARE to make bprm_set_creds() careful about euid.
But /proc/<pid>/cwd and /proc/<pid>/root lookups make transient
use of get_fs_struct(), which also raises that sharing count.

This might occasionally cause a setuid program not to change euid,
in the same way as happened with files->count (check_unsafe_exec
also looks at sighand->count, but /proc doesn't raise that one).

We'd prefer exec not to unshare fs_struct: so fix this in procfs,
replacing get_fs_struct() by get_fs_path(), which does path_get
while still holding task_lock, instead of raising fs->count.

Signed-off-by: Hugh Dickins <hugh@veritas.com>
Cc: stable@kernel.org
___

 fs/proc/base.c |   50 +++++++++++++++--------------------------------
 1 file changed, 16 insertions(+), 34 deletions(-)
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-03-28 17:30:00 -07:00
Hugh Dickins
e426b64c41 fix setuid sometimes doesn't
Joe Malicki reports that setuid sometimes doesn't: very rarely,
a setuid root program does not get root euid; and, by the way,
they have a health check running lsof every few minutes.

Right, check_unsafe_exec() notes whether the files_struct is being
shared by more threads than will get killed by the exec, and if so
sets LSM_UNSAFE_SHARE to make bprm_set_creds() careful about euid.
But /proc/<pid>/fd and /proc/<pid>/fdinfo lookups make transient
use of get_files_struct(), which also raises that sharing count.

There's a rather simple fix for this: exec's check on files->count
has been redundant ever since 2.6.1 made it unshare_files() (except
while compat_do_execve() omitted to do so) - just remove that check.

[Note to -stable: this patch will not apply before 2.6.29: earlier
releases should just remove the files->count line from unsafe_exec().]

Reported-by: Joe Malicki <jmalicki@metacarta.com>
Narrowed-down-by: Michael Itz <mitz@metacarta.com>
Tested-by: Joe Malicki <jmalicki@metacarta.com>
Signed-off-by: Hugh Dickins <hugh@veritas.com>
Cc: stable@kernel.org
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-03-28 17:30:00 -07:00