drm: move pinning/unpinning to buffer attach

This allows importing bo's to own device to work without requiring that the buffer is pinned in GART.

Change-Id: If20aa4e1a112118901baa4110ccb19d68b2a5ea5
Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Dave Airlie <airlied@redhat.com>

drm: add unpin function to prime helpers

Prevents buffers from being pinned forever.

Change-Id: Ifee56b3eb72bf4c5e349e0daf5ebf3156c096247
Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com>
Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Dave Airlie <airlied@redhat.com>

drm: GEM CMA: Add DRM PRIME support

Change-Id: Ib62164dfdbafad48053ecbc0b447aa3c9dc81482
Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Reviewed-by: Rob Clark <robdclark@gmail.com>

drm: GEM CMA: Split object mapping into GEM mapping and CMA mapping

The CMA-specific mapping code will be used to implement dma-buf mmap
support.

Change-Id: I8c60b83683b72bc9d5ba352b6c72e111bed86b83
Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Signed-off-by: Rob Clark <robdclark@gmail.com>

drm: GEM CMA: Split object creation into object alloc and DMA memory alloc

This allows creating a GEM CMA object without an associated DMA memory
buffer, and will be used to implement DRM PRIME support.

Change-Id: I9e9e25f7323829939dae993bb7b343e3dfdd8014
Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Reviewed-by: Rob Clark <robdclark@gmail.com>

drm/omap: Use drm_gem_mmap_obj() to implement dma-buf mmap

The dma-buf mmap code was copied from the GEM mmap implementation.
Replace it with the new drm_gem_mmap_obj() function.

Change-Id: I5dd5b285891debd85f458b66e0f3778d8dcec8d5
Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Reviewed-by: Rob Clark <robdclark@gmail.com>

drm/gem: Split drm_gem_mmap() into object search and object mapping

The drm_gem_mmap() function first finds the GEM object to be mapped
based on the fake mmap offset and then maps the object. Split the object
mapping code into a standalone drm_gem_mmap_obj() function that can be
used to implement dma-buf mmap() operations.

Change-Id: I0be7c06a06dad981ea5f2e7dddbf55e47301bf31
Signed-off-by: Laurent Pinchart <laurent.pinchart+renesas@ideasonboard.com>
Reviewed-by: Rob Clark <robdclark@gmail.com>

Revert "gpu/drm: Allow drm_prime_add_buf_handle to use outside drm_prime.c"

This reverts commit e3bc8acb9541c75789cc486aeff1bd51124676ff.

Revert "drm/prime: resolve race for drm_gem_prime_handle_to_fd()"

This reverts commit cf0ea65cd31feb199d78b15ab0fbcedde852019f.

Revert "drm/sprd: Avoid dma_buf creation on every prime_handle_fd()"

This reverts commit 9b842ab3914d407545fb2903adedb617e6dd4923.

Revert "drm/sprd: Add lookup for gem_handle/dma_buf"

This reverts commit 826db2d77c3fe927c9218cbddddfbf1ead8d2486.

Smack: type confusion in smak sendmsg() handler

Smack security handler for sendmsg() syscall
is vulnerable to type confusion issue what
can allow to privilege escalation into root
or cause denial of service.

A malicious attacker can create socket of one
type for example AF_UNIX and pass is into
sendmsg() function ensuring that this is
AF_INET socket.

Remedy
Do not trust user supplied data.
Proposed fix below.

Change-Id: I6d8b3a3eb9560c0e6d6bfef59e56d6ec659e2d3d
Signed-off-by: Roman Kubiak <r.kubiak@samsung.com>
Signed-off-by: Mateusz Fruba <m.fruba@samsung.com>
Acked-by: Casey Schaufler <casey@schaufler-ca.com>
[Backport from linux-next of v4.5-rc]
Signed-off-by: Seung-Woo Kim <sw0312.kim@samsung.com>

spec: copy modules instead of move

This patch copy modules to proper directory instead of move to
install two sets of modules.

Change-Id: I1d6c44cf8ebbfb44bf99c821cf3b1f6d7f2f5825
Signed-off-by: Seung-Woo Kim <sw0312.kim@samsung.com>

spec: remove unnecessary version from package name

There was unnecessary and wrong date and commit id in the package
name, so this patch removes it.

Change-Id: I14ed31d71dc7e1c0a95c843563dadf500f5958c2
Signed-off-by: Seung-Woo Kim <sw0312.kim@samsung.com>

Smack: File receive for sockets

The existing file receive hook checks for access on
the file inode even for UDS. This is not right, as
the inode is not used by Smack to make access checks
for sockets. This change checks for an appropriate
access relationship between the receiving (current)
process and the socket. If the process can't write
to the socket's send label or the socket's receive
label can't write to the process fail.

This will allow the legitimate cases, where the
socket sender and socket receiver can freely communicate.
Only strangly set socket labels should cause a problem.

Change-Id: Id37df53243264ac843f9c6693ba99aba9779f05e
Signed-off-by: Casey Schaufler <casey@schaufler-ca.com>
[backport to 3.10 from smack-next commit 79be093500791cc25cc31bcaec5a4db62e21497b]
Signed-off-by: Seung-Woo Kim <sw0312.kim@samsung.com>

smack: Add missing logging in bidirectional UDS connect check

During UDS connection check, both sides are checked for write access to
the other side. But only the first check is performed with audit support.
The second one didn't produce any audit logs. This simple patch fixes that.

Change-Id: I8f98910cec80687a18debfac818b50c9b0a9ddbd
Signed-off-by: Rafal Krypa <r.krypa@samsung.com>
[Backport from mainline to 3.10]
Signed-off-by: Seung-Woo Kim <sw0312.kim@samsung.com>

smack: Fix a bidirectional UDS connect check typo

The 54e70ec5eb090193b03e69d551fa6771a5a217c4 commit introduced a
bidirectional check that should have checked for mutual WRITE access
between two labels. Due to a typo subject's OUT label is checked with
object's OUT. Should be OUT to IN.

Change-Id: I020748ac01bb12ed421be4ee20ab8d922c3399fe
Signed-off-by: Zbigniew Jasinski <z.jasinski@samsung.com>
[Backport from mainline to 3.10]
Signed-off-by: Seung-Woo Kim <sw0312.kim@samsung.com>

Fix a bidirectional UDS connect check typo

The 54e70ec5eb090193b03e69d551fa6771a5a217c4 commit introduced a
bidirectional check that should have checked for mutual WRITE access
between two labels. Due to a typo the second check was incorrect.

Change-Id: I0ea34a529bc37e71c7342fe42b6d1c003837c5cd
Signed-off-by: Lukasz Pawelczyk <l.pawelczyk@samsung.com>
[Backport from mainline to 3.10]
Signed-off-by: Seung-Woo Kim <sw0312.kim@samsung.com>

Smack: bidirectional UDS connect check

Smack IPC policy requires that the sender have write access
to the receiver. UDS streams don't do per-packet checks. The
only check is done at connect time. The existing code checks
if the connecting process can write to the other, but not the
other way around. This change adds a check that the other end
can write to the connecting process.

Targeted for git://git.gitorious.org/smack-next/kernel.git

Change-Id: Ibecf37e5644f66885acc02ae0ee96f8009a0b5fa
Signed-off-by: Casey Schuafler <casey@schaufler-ca.com>
[Backport from mainline to 3.10]
Signed-off-by: Seung-Woo Kim <sw0312.kim@samsung.com>

drm/sprd: Add lookup for gem_handle/dma_buf

This patch add support for translation gem_handle to dma_buf
and vice-versa. This allow us to have single gem_obj/gem_handle
associated with a buffer.It will be used in prime_fd_to_handle
and prime_handle_to_fd conversion.

Change-Id: I7333a08939112fa8e926a122b765a162e4f1811f
Signed-off-by: Rohit kumar <rohit.kr@samsung.com>

drm/sprd: Avoid dma_buf creation on every prime_handle_fd()

This patch stores the dma_buf associated with the gem_object in
obj->export_dma_buf and reuse it in prime_handle_to_fd().

Change-Id: I679b3eed769d334ed39392e9c7c4011cc2652fbc
Signed-off-by: Rohit kumar <rohit.kr@samsung.com>

drm/prime: resolve race for drm_gem_prime_handle_to_fd()

This patch resolves race for drm_gem_prime_handle_to_fd().
There could be a race when drm_gem_prime_handle_to_fd() is
requested during __fput() is executed with same dmabuf.

Change-Id: I8ab68a7184ce974ded84c75a7805b9aa1d021ee2
Signed-off-by: YoungJun Cho <yj44.cho@samsung.com>
Signed-off-by: Rohit kumar <rohit.kr@samsung.com>

gpu/drm: Allow drm_prime_add_buf_handle to use outside drm_prime.c

Make drm_prime_add_buf_handle non-static to use it from other files.

Change-Id: I95e9442deca42adbe8a1293588f82391ac46556d
Signed-off-by: Rohit kumar <rohit.kr@samsung.com>

Smack: pipefs fix in smack_d_instantiate

This fix writes the task label when
smack_d_instantiate is called, before the
label of the superblock was written on the
pipe's inode.

Change-Id: I3d4fcf5b8e652d6f2abfe5ef0dfd96306f2c8219
Signed-off-by: Roman Kubiak <r.kubiak@samsung.com>
Acked-by: Casey Schaufler <casey@schaufler-ca.com>
Signed-off-by: Seung-Woo Kim <sw0312.kim@samsung.com>

smack: introduce a special case for tmpfs in smack_d_instantiate()

Files created with __shmem_file_stup() appear to have somewhat fake
dentries which make them look like root directories and not get
the label the current process or ("*") star meant for tmpfs files.

Change-Id: Id97004f2a5090cee9c16778109d849d433bd39c9
Signed-off-by: Łukasz Stelmach <l.stelmach@samsung.com>
Signed-off-by: Seung-Woo Kim <sw0312.kim@samsung.com>

ARM: dts: sc7730: add Performance Monitoring Unit(PMU) node

This patch is to support perf command with performance monitoring
unit (PMU) of SC7730S CPU on Z3 device. If you want to monitor
hardware information such as cpu cycles, insns per cycle, cache miss,
and branches, you must enable this patch.

For example,
sh-3.2# uname -a
Linux localhost 3.10.65-00016-g465f771-dirty #2-Tizen SMP PREEMPT Fri Dec 4 08:38:31 KST 2015 armv7l GNU/Linux

sh-3.2#
sh-3.2# perf list | grep "Hardware event"
  cpu-cycles OR cycles                               [Hardware event]
  instructions                                       [Hardware event]
  cache-references                                   [Hardware event]
  cache-misses                                       [Hardware event]
  branch-instructions OR branches                    [Hardware event]
  branch-misses                                      [Hardware event]
  bus-cycles                                         [Hardware event]
  stalled-cycles-frontend OR idle-cycles-frontend    [Hardware event]
  stalled-cycles-backend OR idle-cycles-backend      [Hardware event]
  ref-cycles                                         [Hardware event]
sh-3.2#
sh-3.2#
sh-3.2# perf stat ls

 Performance counter stats for 'ls':

          5.950928 task-clock                #    0.666 CPUs utilized
                 5 context-switches          #    0.840 K/sec
                 0 cpu-migrations            #    0.000 K/sec
               224 page-faults               #    0.038 M/sec
           5962814 cycles                    #    1.002 GHz
   <not supported> stalled-cycles-frontend
   <not supported> stalled-cycles-backend
           1629441 instructions              #    0.27  insns per cycle
            197736 branches                  #   33.228 M/sec
             55299 branch-misses             #   27.97% of all branches

       0.008941650 seconds time elapsed
sh-3.2#
sh-3.2# perf record -a
^C (Press Ctrl + C key)
[ perf record: Woken up 1 times to write data ]
[ perf record: Captured and wrote 0.947 MB perf.data (~41368 samples) ]

sh-3.2# cat /proc/sys/kernel/kptr_restrict
2
sh-3.2# echo 0 > /proc/sys/kernel/kptr_restrict
sh-3.2# cat /proc/sys/kernel/kptr_restrict
0

sh-3.2# perf report
Samples: 2K of event 'cycles', Event count (approx.): 331006766
13.53%          swapper  [kernel.kallsyms]           [k] sc_enter_idle
  8.59%          swapper  [kernel.kallsyms]           [k] tick_nohz_idle_exit
  7.67%    enlightenment  libc-2.20-2014.11.so        [.] 0x000bdf34
  4.92%          swapper  [kernel.kallsyms]           [k] _raw_spin_unlock_irq
  4.61%    enlightenment  [kernel.kallsyms]           [k] _raw_spin_unlock_irqr
  4.11%    enlightenment  [kernel.kallsyms]           [k] user_exit
  3.98%    enlightenment  [kernel.kallsyms]           [k] _raw_spin_unlock_irq
  2.94%          swapper  [kernel.kallsyms]           [k] tick_nohz_idle_enter
  1.83%  tizenstoreservi  [kernel.kallsyms]           [k] _raw_spin_unlock_irq
  1.61%    enlightenment  libecore.so.1.13.0          [.] 0x0000ba3c
  1.00%    enlightenment  libpthread-2.20-2014.11.so  [.] recvmsg
  0.89%    enlightenment  libeo.so.1.13.0             [.] _eo_call_resolve
  0.75%    enlightenment  [kernel.kallsyms]           [k] fget_light
  0.74%    enlightenment  [kernel.kallsyms]           [k] __gptimer_clocksource
  0.73%    enlightenment  libpthread-2.20-2014.11.so  [.] 0x00009db0
  0.72%  tizenstoreservi  libc-2.20-2014.11.so        [.] 0x000d9fa8
  0.70%    enlightenment  libeo.so.1.13.0             [.] eo_data_scope_get
  0.65%    enlightenment  [kernel.kallsyms]           [k] do_select
  0.64%    enlightenment  libecore_x.so.1.13.0        [.] 0x000322c6
  0.64%    enlightenment  libeo.so.1.13.0             [.] _eo_do_start
  0.59%  cloud-pdm-serve  [kernel.kallsyms]           [k] _raw_spin_unlock_irq
  0.55%    enlightenment  enlightenment               [.] 0x00051120

sh-3.2#

Change-Id: Icc16380e3dc7027d45b7a6a202c589375d68b255
Reported-by: Geunsik Lim <geunsik.lim@samsung.com>
Tested-by: Geunsik Lim <geunsik.lim@samsung.com>
Signed-off-by: Geunsik Lim <geunsik.lim@samsung.com>
Signed-off-by: Congfu Zhao <Congfu.Zhao@spreadtrum.com>
CC: Vishnu Pratap Singh <vishnu.ps@samsung.com>
CC: Manjunatha Achar <a.manjunatha@samsung.com>
CC: Rajkumar Chandrasekar <c.rajkumar@samsung.com>
CC: Pintu Kumar <pintu.k@samsung.com>
CC: MyungJoo Ham <myungjoo.ham@samsung.com>

serial: sprd: Fix wrong console output on tty port 1

Now, tty port 1 for serial console has a problem that serial console
output shows like ����������������������������������������.

The reason is by the function that resets the serial port in shutdown().
This patch reverts it to old codes instead of reset for tty port 1.

Change-Id: I83817394bda9226e2d35e9072dfde616d6d9a24f
Signed-off-by: Joonyoung Shim <jy0922.shim@samsung.com>

kernel: swap-modules: remove swap-modules codes

swap-modules will be loaded as a kernel module

Change-Id: I28d8d5c9fc7f2f8cb218f8b59bd322899db51232
Signen-off-by: Hyeongsik Min <hyeongsik.min@samsung.com>

ARM: tizen_tm1_defconfig: Enable missing features related with CGROUPS

CONFIG_CGROUP_DEVICE=y
CONFIG_CPUSETS=y
CONFIG_PROC_PID_CPUSET=y

Especially, the security manager needs CONFIG_CPUSETS and
CONFIG_PROC_PID_CPUSET.

Change-Id: Idd49d81fabc937eb259e0654502681e358d90e19
Signed-off-by: Joonyoung Shim <jy0922.shim@samsung.com>

ARM: tizen_tm1_defconfig: Enable namespaces

To fix that systemd-hostnamed is failed.

Change-Id: I95efe13133bec30a69d12699d627cd0554e35d4f
Signed-off-by: Joonyoung Shim <jy0922.shim@samsung.com>

arm: config: tm1: enable netfilter multiport configuration

Enable the multiport.
Previous config is wrong. Re-commit this patch.

Change-Id: Ie2b38a4c2aaa922dfcd1fd3de47fc781bda7c5b3
Signed-off-by: Jaehoon Chung <jh80.chung@samsung.com>

Revert "arm: config: tm1: enable netfilter multiport configuration"

This reverts commit 908f8b80a3eadca0a9515f71a3c1ab739c74e0a9.

Change-Id: I1563acc745813075494c427b496cf0fda1af8c17

BATTERY: Add missing get_property() of some property

The property of power should be implemented both get_property() and
set_property() but some property has only set_property() and it causes
some problem on tizen_3.0 platform like belows.

[   15.804351]  [3:systemd-journal:  188] power_supply sec-charger: driver failed to report `voltage_max' property: -22
[   27.924804]  [3:systemd-journal:  188] power_supply sec-fuelgauge: driver failed to report `status' property: -22
[   28.080047]  [3:systemd-journal:  179] power_supply sec-fuelgauge: driver failed to report `charge_full' property: -22
[   20.704162]  [0:systemd-journal:  180] power_supply sec-fuelgauge: driver failed to report `present' property: -22
[   19.580261]  [1:systemd-journal:  181] power_supply sec-fuelgauge: driver failed to report `energy_now' property: -22
[   20.989288]  [2:systemd-journal:  167] power_supply sec-fuelgauge: driver failed to report `temp' property: -22
[   20.832183]  [2:systemd-journal:  177] power_supply sec-fuelgauge: driver failed to report `temp_ambient' property: -22

This is workaround so driver should implement get_property() exactly
later.

Change-Id: I9d9650b748bdb73ae3149f0c4709908d0c0bb962
Signed-off-by: Min Kang <min1023.kang@samsung.com>
Signed-off-by: Joonyoung Shim <jy0922.shim@samsung.com>

net: wirelss: sc2331: Apply the tethering APIs patch

Applied tethering APIs patch received from spreadtrum.
(This is related with MAC control.)

Change-Id: I992d810fb3e2813dd77ba3e0684c73e58a2d2a75
Signed-off-by: Jaehoon Chung <jh80.chung@samsung.com>

arm: config: tm1: enable netfilter multiport configuration

Enable CONFIG_NETFILTER_XT_MATCH_MULTIPORT.

Change-Id: I0a504c030549880f88e27c887216bfa4dde14010
Signed-off-by: Jaehoon Chung <jh80.chung@samsung.com>

Update from product codes

Until commit 5cfebdc36f5b88225c4f8c032f7cf511ebba1cc5

Change-Id: I0f838df8640acefc0cdecd628e7c7cd10360d78f
Signed-off-by: Joonyoung Shim <jy0922.shim@samsung.com>

spec: fix badness problem by rpmlint

Errors:

[  978s] RPMLINT report:
[  978s] ===============
[  978s] qemu: Unsupported syscall: 311
[ 1870s] (none): E: badness 6147800 exceeds threshold 2000, aborting.

Badness cases:

E: spurious-executable-perm (Badness: 50)
E: non-standard-group (Badness: 200)
E: license-placeholder-not-allowed (Badness: 200)
E: invalid-license (Badness: 200)
E: group-placeholder-not-allowed (Badness: 200)
E: files-duplicated-waste (Badness: 100)
E: dir-or-file-in-tmp (Badness: 200)

Changes:

- Fix licenses and group tags
- s/var\/tmp/boot/
- add command in %install of spec
find %{buildroot}/boot/kernel/ -name "*.h" -exec chmod 644 {} \;

Change-Id: Idf55ca574cf7d15453a23ebb72cf717778ed75d6
Signed-off-by: Joonyoung Shim <jy0922.shim@samsung.com>

spec: do not check target_name

Now public Tizen 3.0 doesn't support target name
so the relevant script should be removed to build
kernel.

Change-Id: I21ea402c2274e76b2fd197bca1505413a4616b0a
Signed-off-by: Inki Dae <inki.dae@samsung.com>

tizen 2.4 release

Initial empty repository