Commit graph

505,617 commits

Author SHA1 Message Date
Mike Marciniszyn
3033771feb IB/ipath: Change get_user_pages() usage to always NULL vmas
The static helper routine, __ipath_get_user_pages(), accepts a vma
arg, but current use always passes NULL.

This has caused some confusion associated with the correct use of this
argument, but since the current use case doesn't require the
flexiblity, the best thing to do is to simplfy the code to always pass
NULL to get_user_pages().

Reviewed-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Mike Marciniszyn <mike.marciniszyn@intel.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
2014-09-19 10:03:11 -07:00
Brian Norris
ab95eac99c MAINTAINERS: add l2-mtd.git, 'next' tree for MTD
We've been semi-officially queueing patches here for a while, and it's
in linux-next, so let's advertise it in MAINTAINERS.

Signed-off-by: Brian Norris <computersforpeace@gmail.com>
Acked-by: Artem Bityutskiy <artem.bityutskiy@linux.intel.com>
2014-09-19 10:01:10 -07:00
Bjorn Helgaas
7a0b33d4a4 Revert "PCI: Don't scan random busses in pci_scan_bridge()"
This reverts commit fc1b253141 ("PCI: Don't scan random busses in
pci_scan_bridge()") because it breaks CardBus on some machines.

David tested a Dell Latitude D505 that worked like this prior to
fc1b253141:

    pci 0000:00:1e.0: PCI bridge to [bus 01]
    pci 0000:01:01.0: CardBus bridge to [bus 02-05]

Note that the 01:01.0 CardBus bridge has a bus number aperture of
[bus 02-05], but those buses are all outside the 00:1e.0 PCI bridge bus
number aperture, so accesses to buses 02-05 never reach CardBus.  This is
later patched up by yenta_fixup_parent_bridge(), which changes the
subordinate bus number of the 00:1e.0 PCI bridge:

    pci_bus 0000:01: Raising subordinate bus# of parent bus (#01) from #01 to #05

With fc1b253141, pci_scan_bridge() fails immediately when it notices that
we can't allocate a valid secondary bus number for the CardBus bridge, and
CardBus doesn't work at all:

    pci 0000:01:01.0: can't allocate child bus 01 from [bus 01]

I'd prefer to fix this by integrating the yenta_fixup_parent_bridge() logic
into pci_scan_bridge() so we fix the bus number apertures up front.  But
I don't think we can do that before v3.17, so I'm going to revert this to
avoid the problem while we're working on the long-term fix.

Link: https://bugzilla.kernel.org/show_bug.cgi?id=83441
Link: http://lkml.kernel.org/r/1409303414-5196-1-git-send-email-david.henningsson@canonical.com
Reported-by: David Henningsson <david.henningsson@canonical.com>
Tested-by: David Henningsson <david.henningsson@canonical.com>
Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
CC: stable@vger.kernel.org	# v3.15+
2014-09-19 10:56:06 -06:00
Shawn Bohrer
87773dd56d IB: ib_umem_release() should decrement mm->pinned_vm from ib_umem_get
In debugging an application that receives -ENOMEM from ib_reg_mr(), I
found that ib_umem_get() can fail because the pinned_vm count has
wrapped causing it to always be larger than the lock limit even with
RLIMIT_MEMLOCK set to RLIM_INFINITY.

The wrapping of pinned_vm occurs because the process that calls
ib_reg_mr() will have its mm->pinned_vm count incremented.  Later a
different process with a different mm_struct than the one that
allocated the ib_umem struct ends up releasing it which results in
decrementing the new processes mm->pinned_vm count past zero and
wrapping.

I'm not entirely sure what circumstances cause a different process to
release the ib_umem than the one that allocated it but the kernel
stack trace of the freeing process from my situation looks like the
following:

    Call Trace:
     [<ffffffff814d64b1>] dump_stack+0x19/0x1b
     [<ffffffffa0b522a5>] ib_umem_release+0x1f5/0x200 [ib_core]
     [<ffffffffa0b90681>] mlx4_ib_destroy_qp+0x241/0x440 [mlx4_ib]
     [<ffffffffa0b4d93c>] ib_destroy_qp+0x12c/0x170 [ib_core]
     [<ffffffffa0cc7129>] ib_uverbs_close+0x259/0x4e0 [ib_uverbs]
     [<ffffffff81141cba>] __fput+0xba/0x240
     [<ffffffff81141e4e>] ____fput+0xe/0x10
     [<ffffffff81060894>] task_work_run+0xc4/0xe0
     [<ffffffff810029e5>] do_notify_resume+0x95/0xa0
     [<ffffffff814e3dd0>] int_signal+0x12/0x17

The following patch fixes the issue by storing the pid struct of the
process that calls ib_umem_get() so that ib_umem_release and/or
ib_umem_account() can properly decrement the pinned_vm count of the
correct mm_struct.

Signed-off-by: Shawn Bohrer <sbohrer@rgmadvisors.com>
Reviewed-by: Shachar Raindel <raindel@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
2014-09-19 09:55:42 -07:00
Devesh Sharma
f93439e476 RDMA/ocrdma: Do not skip setting deferred_arm
When ib_request_notify_cq() is called for the first time, ocrdma tries
to skip setting deffered_arm flag. This may lead CQ to an un-armed
state thus never generating a CQ event and leaving consumer hung.

This patch removes the part of code that skips setting deferred_arm.

Signed-off-by: Devesh Sharma <devesh.sharma@emulex.com>
Signed-off-by: Selvin Xavier <selvin.xavier@emulex.com>
Signed-off-by: Mitesh Ahuja <mitesh.ahuja@emulex.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
2014-09-19 09:53:35 -07:00
Devesh Sharma
d6a488f21c RDMA/ocrdma: Report correct value of max_fast_reg_page_list_len
Fix ocrdma_query_device() to report correct value of max_fast_reg_page_list_len.

Signed-off-by: Devesh Sharma <devesh.sharma@emulex.com>
Signed-off-by: Selvin Xavier <selvin.xavier@emulex.com>
Signed-off-by: Mitesh Ahuja <mitesh.ahuja@emulex.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
2014-09-19 09:52:33 -07:00
Frans Klaver
d48690383f eeepc-laptop: store_cpufv: return error if set_acpi fails
The result of set_acpi is left unchecked, but it may return errors. If
one occurs, send the error to the caller. There's no reason to lie about
it, if set_acpi fails.

Signed-off-by: Frans Klaver <fransklaver@gmail.com>
Signed-off-by: Darren Hart <dvhart@linux.intel.com>
2014-09-19 09:42:12 -07:00
Frans Klaver
a5c155b16f eeepc-laptop: check proper return values in get_cpufv
In get_cpufv the return value of get_acpi is stored in the cpufv struct.
Right before this value is checked for errors, it is and'ed with 0xff.
This means c->cur can never be less than zero. Besides that, the actual
error value is ignored.

c->num is also and'ed with 0xff, which means we can ignore values below
zero.

Check the result of get_acpi() right away. While at it, propagate the
error if we got one.

Signed-off-by: Frans Klaver <fransklaver@gmail.com>
Signed-off-by: Darren Hart <dvhart@linux.intel.com>
2014-09-19 09:42:11 -07:00
Frans Klaver
48d4a5b29c eeepc-laptop: make fan1_input really read-only
In the instantiation of the fan1_input device attribute, NULL is passed
as set function to store_sys_hwmon. The function pointer is never
checked before dereferencing it. This is fine if we can guarantee that
it will never be called with an invalid pointer, but we can't. If
someone from user space decides to change the permissions on this
attribute and write to it, kernel will crash.

Introduce EEEPC_CREATE_SENSOR_ATTR_RO() to instantiate a read-only
attribute, and declare fan1_input with it. This ensures store_sys_hwmon
is never called with NULL parameters. If someone tries to write the
attribute, the system will at least keep its sanity.

This also causes EEEPC_CREATE_SENSOR_ATTR() to be only used for R/W
attributes.This enables us to drop the _mode argument from the macro
and use DEVICE_ATTR_RW() internally while we're at it. Append _RW to the
name for readability.

Signed-off-by: Frans Klaver <fransklaver@gmail.com>
Signed-off-by: Darren Hart <dvhart@linux.intel.com>
2014-09-19 09:42:11 -07:00
Frans Klaver
28ac85f71a eeepc-laptop: pull out SENSOR_STORE_FUNC and SENSOR_SHOW_FUNC macros
Pull out EEEPC_SENSOR_STORE_FUNC and EEEPC_SENSOR_SHOW_FUNC. These
macros define functions that call store_sys_hwmon() and show_sys_hwmon()
respectively. This helps prevent duplication later on.

Signed-off-by: Frans Klaver <fransklaver@gmail.com>
Signed-off-by: Darren Hart <dvhart@linux.intel.com>
2014-09-19 09:42:11 -07:00
Frans Klaver
6fe3a77f62 eeepc-laptop: tell sysfs that the disp attribute is write-only
The disp attribute is write-only, but sysfs doesn't know this. Currently
show_sys_acpi() is mimicking sysfs behavior, if the underlying acpi call
should fail. This was introduced in 6dff29b63a "eeepc-laptop:
disp attribute should be write-only". This is not ideal; behaving like
sysfs is better left to sysfs.

Introduce EEEPC_CREATE_DEVICE_ATTR_WO() to instantiate a write-only
attribute, and declare the disp attribute with it. Sysfs makes sure
userspace can only write to disp at all times. This removes the need for
mimicking the sysfs behavior in show_sys_acpi() and store_sys_acpi(),
but we'll stick with -EIO, as changing sysfs return values should not be
taken lightly.

This change also causes EEEPC_CREATE_DEVICE_ATTR() to be used only for
R/W attributes. This enables us to drop the _mode argument from the
macro and use DEVICE_ATTR_RW() internally while we're at it. Append _RW
to the name for readability.

Signed-off-by: Frans Klaver <fransklaver@gmail.com>
Signed-off-by: Darren Hart <dvhart@linux.intel.com>
2014-09-19 09:42:11 -07:00
Frans Klaver
9797132577 eeepc-laptop: pull out ACPI_STORE_FUNC and ACPI_SHOW_FUNC macros
Pull out macros EEEPC_ACPI_STORE_FUNC and EEEPC_ACPI_SHOW_FUNC. These
macros define functions that call store_sys_acpi() and show_sys_acpi()
respectively. This helps prevent duplication later on.

Signed-off-by: Frans Klaver <fransklaver@gmail.com>
Signed-off-by: Darren Hart <dvhart@linux.intel.com>
2014-09-19 09:42:11 -07:00
Frans Klaver
bb382dbaba eeepc-laptop: use DEVICE_ATTR* to instantiate device_attributes
Device attributes are instantiated manually, while we have DEVICE_ATTR*
macros available to do much of the work for us. Let's use them.

Signed-off-by: Frans Klaver <fransklaver@gmail.com>
Signed-off-by: Darren Hart <dvhart@linux.intel.com>
2014-09-19 09:42:11 -07:00
Frans Klaver
8c72fc8bd7 eeepc-laptop: change sysfs function names to API expectations
The eeepc-laptop driver follows the function naming convention
<action>_<attrname>(), while the sysfs macros are built around the
convention <attrname>_<action>(). Rename the sysfs functions to the
convention used by sysfs. This makes it easier to use the available API
later on.

Signed-off-by: Frans Klaver <fransklaver@gmail.com>
Signed-off-by: Darren Hart <dvhart@linux.intel.com>
2014-09-19 09:42:11 -07:00
Frans Klaver
39a3e17e0d eeepc-laptop: clean up coding style
Correct indentation and brace usage to comply with
Documentation/CodingStyle.

Signed-off-by: Frans Klaver <fransklaver@gmail.com>
Signed-off-by: Darren Hart <dvhart@linux.intel.com>
2014-09-19 09:42:11 -07:00
Masahiro Yamada
8125450cd8 mtd: denali: fix indents and other trivial things
- Fix indents
- Do not break a line unless it is longer than 80 columns
- Do not insert a whitespace before ';'
- Use whitespaces around operators
- Use braces for a "else" block where the "if" block uses ones.

Besides, eliminate all the warnings reported by checkpatch.pl:
- WARNING: quoted string split across lines
- WARNING: else is not generally useful after a break or return
- WARNING: Missing a blank line after declarations
- WARNING: Avoid line continuations in quoted strings

Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2014-09-19 09:39:27 -07:00
Masahiro Yamada
7d14ecd050 mtd: denali: remove unnecessary parentheses
We should use parentheses only when they are necessary
or they really improve the readability.

Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com>
Reviewed-by: Josh Triplett <josh@joshtriplett.org>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2014-09-19 09:39:26 -07:00
Brian Norris
ba5f2bc2af mtd: denali: remove another set-but-unused variable
The variable "irq_status" in denali_read_page_raw() is set, but not used.

Signed-off-by: Masahiro Yamada <yamada.m@jp.panasonic.com>
Signed-off-by: Brian Norris <computersforpeace@gmail.com>
2014-09-19 09:38:54 -07:00
Andrzej Hajda
72ed6ccd08 drm/exynos: switch to universal plane API
The patch replaces legacy functions
drm_plane_init() / drm_crtc_init() with
drm_universal_plane_init() and drm_crtc_init_with_planes().
It allows to replace fake primary plane with the real one.
Additionally the patch leaves cleanup of crtcs to core,
this way planes and crtcs are cleaned in correct order.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
2014-09-20 01:17:53 +09:00
Linus Torvalds
7a5e87867e Merge branch 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull x86 fixes from Ingo Molnar:
 "Misc fixes:

  EFI fixes, a build fix, a page table dumping (debug) fix and a clang
  build fix"

* 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  efi/arm64: Fix fdt-related memory reservation
  x86/mm: Apply the section attribute to the variable, not its type
  x86/efi: Fixup GOT in all boot code paths
  x86/efi: Only load initrd above 4g on second try
  x86-64, ptdump: Mark espfix area only if existent
  x86, irq: Fix build error caused by 9eabc99a63
2014-09-19 09:07:47 -07:00
Linus Torvalds
9f47112975 Merge branch 'upstream' of git://git.linux-mips.org/pub/scm/ralf/upstream-linus
Pull MIPS fixes from Ralf Baechle:
 "A fair number of build fixes for various configurations.

  Fixes to BPF, and the BCM47xx platform code, a preemption fix for the
  Loongson core, a syscall auditing fix, wire up the new getrandom and
  memfd_create.  Several patches for EVA"

* 'upstream' of git://git.linux-mips.org/pub/scm/ralf/upstream-linus: (30 commits)
  MIPS: SmartMIPS: Disable assembler warnings
  MIPS: Move CPU topology macros to topology.h
  MIPS: Wire up new syscalls getrandom and memfd_create.
  MIPS: Fix a warning for virt_to_page
  arch/mips/net/bpf_jit.c: fix failure check
  MIPS: COP2: CPP macro safety fixes.
  MIPS: Kconfig: Select SMP symbols for CMP
  MIPS: ZBOOT: add missing <linux/string.h> include
  MIPS: IP28: Fix/clean spaces.h
  MIPS: IP28: Select correct L1_CACHE_SHIFT
  MIPS: BCM63xx: delete double assignment
  MIPS: Spelling s/confugrations/configurations/
  MIPS: OCTEON: make get_system_type() thread-safe
  MIPS: CPS: Initialize EVA before bringing up VPEs from secondary cores
  MIPS: Malta: EVA: Rename 'eva_entry' to 'platform_eva_init'
  MIPS: EVA: Add new EVA header
  MIPS: scall64-o32: Fix indirect syscall detection
  MIPS: syscall: Fix AUDIT value for O32 processes on MIPS64
  MIPS: Loongson: Fix COP2 usage for preemptible kernel
  MIPS: NL: Fix nlm_xlp_defconfig build error
  ...
2014-09-19 09:06:39 -07:00
Inki Dae
832316c704 drm/exynos: use drm generic mmap interface
This patch removes DRM_EXYNOS_GEM_MMAP ictrl feature specific
to Exynos drm and instead uses drm generic mmap.

We had used the interface specific to Exynos drm to do mmap directly,
not to use demand paging which maps each page with physical memory
at page fault handler. We don't need the specific mmap interface
because the drm generic mmap which uses vm offset manager stuff can
also do mmap directly.

This patch makes a userspace region to be mapped with whole physical
memory region allocated by userspace request when mmap system call is
requested.

Changelog v2:
- do not set VM_IO, VM_DONTEXPEND and VM_DONTDUMP. These flags were already
  set by drm_gem_mmap
- do not include <linux/anon_inodes.h>, which isn't needed anymore.

Signed-off-by: Inki Dae <inki.dae@samsung.com>
2014-09-20 01:00:13 +09:00
Inki Dae
d931589c01 drm/exynos: remove DRM_EXYNOS_GEM_MAP_OFFSET ioctl
This interface and relevant codes aren't used anymore.

Signed-off-by: Inki Dae <inki.dae@samsung.com>
2014-09-20 01:00:07 +09:00
Chris Wilson
1893a71b1e drm/i915: Inline feature detection into sanitize_enable_ppgtt
Rather than splitting and hiding away critical parts of
sanitize_enable_ppgtt() into single use macros in the headers, inline
them into the function for clarity.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2014-09-19 17:58:42 +02:00
Joonyoung Shim
4a3ffedda2 drm/exynos: factor out initial setting of each driver
From fimd driver and vidi driver, dev->irq_enabled and
dev->vblank_disable_allowed are set and also mixer needs them even if
missed. It's duplicated so set them when loads drm driver.

Signed-off-by: Joonyoung Shim <jy0922.shim@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
2014-09-20 00:56:39 +09:00
Brad Volkin
00caf0199f drm/i915: Log a message when rejecting LRM to OACONTROL
The other paths in the command parser that reject a batch all
log a message indicating the reason. We simply missed this one.

Signed-off-by: Brad Volkin <bradley.d.volkin@intel.com>
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
2014-09-19 17:56:20 +02:00
Andrzej Hajda
ad279310d7 drm/exynos/hdmi: unregister connector on removal
During component removal driver should unregister connector.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
2014-09-20 00:56:15 +09:00
Andrzej Hajda
7c61b1ecab drm/exynos/dp: unregister connector on removal
During component removal driver should unregister connector.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
2014-09-20 00:56:15 +09:00
Andrzej Hajda
90eac89726 drm/exynos/dpi: unregister connector and panel on removal
During component removal it should unregister connector
and optionally detach the panel.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
2014-09-20 00:56:15 +09:00
Andrzej Hajda
0ae460159f drm/exynos/dsi: unregister connector on removal
During component unbind connector should be unregistered.
Also DSI host should be unregistered after KMS cleanup.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
2014-09-20 00:56:15 +09:00
Andrzej Hajda
849b43187a drm/exynos/fb: free exynos framebuffer on error
In case drm_framebuffer_init fails exynos_fb should be freed
before returning an error.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
2014-09-20 00:56:14 +09:00
Andrzej Hajda
78ef31e204 drm/exynos/fbdev: fix fbdev gem object cleanup
exynos_gem_obj is used by exynos_drm_fbdev_destroy so it cannot be destroyed
before calling the latter. exynos_gem_obj will be destroyed anyway by
exynos_drm_fbdev_destroy->...->exynos_drm_fb_destroy.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
2014-09-20 00:56:14 +09:00
Andrzej Hajda
9f3dd7dbc5 drm/exynos: fix drm driver de-initialization order
Since components have their own cleanup routines calling
drm_mode_config_cleanup before component_unbind_all causes errors
due to double free of KMS objects. The patch fixes it by changing
de-initialization order. Now it is exactly opposite to init order.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
2014-09-20 00:56:14 +09:00
Andrzej Hajda
a36ed46687 drm/exynos/ipp: traverse ipp drivers list safely
On ipp subsystem removal list of ipp drivers is traversed
and their members are deleted. To do it properly safe version
of list_for_each* should be used.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
2014-09-20 00:56:13 +09:00
Inki Dae
53c5558d95 drm/exynos: update to use component match support
Update Exynos's DRM driver to use component match support rater than
add_components.

Changelog v2:
- release devices and drivers if failed.
- change compare_of to compare_dev.

Signed-off-by: Inki Dae <inki.dae@samsung.com>
Tested-by: Andrzej Hajda <a.hajda@samsung.com>
2014-09-20 00:56:13 +09:00
Andrzej Hajda
18383cb92f drm/exynos/ipp: add file checks for ioctls
Process should not have access to ipp nodes created by another
process. The patch adds necessary checks.
It also simplifies lookup for command node.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Reviewed-by: Joonyoung Shim <jy0922.shim@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
2014-09-20 00:56:13 +09:00
Andrzej Hajda
d9b9734c5d drm/exynos/ipp: remove file argument from node related functions
Since file pointer is preserved in c_node passing it
as argument in node functions is redundant.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Reviewed-by: Joonyoung Shim <jy0922.shim@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
2014-09-20 00:56:13 +09:00
Andrzej Hajda
20ed715ebb drm/exynos/fimc: fix source buffer registers
FIMC in default mode of operation uses only one input buffer,
but the driver used also second buffer, as a result only the
first frame was processed correctly. The patch fixes it.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Reviewed-by: Joonyoung Shim <jy0922.shim@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
2014-09-20 00:56:13 +09:00
Andrzej Hajda
56442d8340 drm/exynos/fimc: simplify buffer queuing
The patch removes redundant checks, redundant HW reads
and simplifies code.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Reviewed-by: Joonyoung Shim <jy0922.shim@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
2014-09-20 00:56:12 +09:00
Andrzej Hajda
c7b3014bf9 drm/exynos/fimc: do not enable fimc twice
The patch removes redundant H/W activation.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Reviewed-by: Joonyoung Shim <jy0922.shim@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
2014-09-20 00:56:12 +09:00
Andrzej Hajda
7794a775f4 drm/exynos/fimc: avoid clearing overflow bits
Overflow bits shall be cleared by H/W.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Reviewed-by: Joonyoung Shim <jy0922.shim@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
2014-09-20 00:56:12 +09:00
Andrzej Hajda
c0592c8549 drm/exynos/ipp: remove events during command cleaning
Events were removed only during stop command, as a result
there were memory leaks if program prematurely exited.
This patch fixes it.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Reviewed-by: Joonyoung Shim <jy0922.shim@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
2014-09-20 00:56:12 +09:00
Andrzej Hajda
8aa99dd376 drm/exynos/ipp: stop hardware before freeing memory
Memory shouldn't be freed when hardware is still running.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Reviewed-by: Joonyoung Shim <jy0922.shim@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
2014-09-20 00:56:12 +09:00
Andrzej Hajda
05afb1ac53 drm/exynos/ipp: replace work_struct casting with better constructs
Type casting should be avoided if possible. In case of
work_struct it can be simply replaced by reference to member field.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Reviewed-by: Joonyoung Shim <jy0922.shim@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
2014-09-20 00:56:12 +09:00
Andrzej Hajda
22e816f87c drm/exynos/ipp: clean memory nodes on command node cleaning
The nodes should be removed before removing command node.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Reviewed-by: Joonyoung Shim <jy0922.shim@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
2014-09-20 00:56:11 +09:00
Andrzej Hajda
c4a856a733 drm/exynos/ipp: move nodes cleaning to separate function
The patch introduces ipp_clean_mem_nodes function which replaces
redundant code. Additionally memory node function definitions
are moved up to increase its visibility.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Reviewed-by: Joonyoung Shim <jy0922.shim@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
2014-09-20 00:56:11 +09:00
Andrzej Hajda
6602ffb849 drm/exynos/ipp: free partially allocated resources on error
In case of allocation errors some already allocated buffers
were not freed. The patch fixes it.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Reviewed-by: Joonyoung Shim <jy0922.shim@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
2014-09-20 00:56:11 +09:00
Andrzej Hajda
985c293d09 drm/exynos/ipp: remove unused field in command node
Since command node have file pointer dev field became useless.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Reviewed-by: Joonyoung Shim <jy0922.shim@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
2014-09-20 00:56:10 +09:00
Andrzej Hajda
21a825ee1f drm/exynos/ipp: remove only related commands on file close
On file close driver should remove only command nodes created
via this file.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Reviewed-by: Joonyoung Shim <jy0922.shim@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
2014-09-20 00:56:10 +09:00
Andrzej Hajda
945a0aad96 drm/exynos/ipp: move file reference from memory to command node
Command node should contain file reference to distinguish commands
created by different processes.

Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Reviewed-by: Joonyoung Shim <jy0922.shim@samsung.com>
Signed-off-by: Inki Dae <inki.dae@samsung.com>
2014-09-20 00:56:10 +09:00