Mythri P K [Thu, 8 Sep 2011 13:36:26 +0000 (19:06 +0530)]
OMAP4: DSS2: HDMI: Function pointer approach to call
HDMI IP fundamentally has replaceable core PHY and PLL blocks.
These blocks might vary across OMAP's but the end functionality such as to
enable or disable PLL, PHY, function to read EDID would remain the same.
Thus to make the current hdmi DSS driver compatible with different OMAP's having
different IP blocks( A combination of different core, PHY, PLL blocks), function
pointer approach is introduced.
With function pointer, relevant IP dependent functions are mapped to the generic
functions used by DSS during the initialization based on the OMAP compiled.
Thus making hdmi DSS driver IP agnostic.
Signed-off-by: Mythri P K <mythripk@ti.com> Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Mythri P K [Thu, 8 Sep 2011 13:36:23 +0000 (19:06 +0530)]
OMAP4: DSS2: HDMI: Split the current HDMI driver to move
Split the current HDMI driver to move the HDMI IP dependent ( PLL/PHY/Core
configuration code) to a new IP file (ti_hdmi_4xxx_ip.c.
This is to separate IP dependent OMAP agnostic code from OMAP specific DSS
dependent code.
Signed-off-by: Mythri P K <mythripk@ti.com> Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Mythri P K [Thu, 8 Sep 2011 13:36:21 +0000 (19:06 +0530)]
OMAP4: DSS2: HDMI: Move HDMI IP independent generic header
Some of the header file definitions that are there in the hdmi.h are generic
and can be used across OMAP's, Thus moving generic definition to new file.
Signed-off-by: Mythri P K <mythripk@ti.com> Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Mythri P K [Thu, 8 Sep 2011 13:36:20 +0000 (19:06 +0530)]
OMAP4: DSS2: HDMI: Use specific HDMI timings structure
As hdmi has few additional parameters such as vsync and hsync
polarity which is missing in DSS timing structure, define HDMI timings
structure for hdmi to use instead of OMAP DSS timing structure.
Signed-off-by: Mythri P K <mythripk@ti.com> Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Mythri P K [Thu, 8 Sep 2011 13:36:19 +0000 (19:06 +0530)]
OMAP4: DSS2: HDMI: Move pll and video configuration
As the pll and the video configuration info are part of the ip_data, pll
and video structures are moved to the ip_data structure. Also the pll and
video configuration functions are modified accordingly to take care of the
structure movement.
Signed-off-by: Mythri P K <mythripk@ti.com> Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Mythri P K [Thu, 8 Sep 2011 13:36:18 +0000 (19:06 +0530)]
OMAP4: DSS2: HDMI: HDMI clean up to pass base_address
As the base_address of the HDMI might differ across SoC's, offset of the HDMI
logical blocks(PHY, PLL and Core) and base address procured from the platform
data are passed dynamically to the functions that modify HDMI IP registers.
Signed-off-by: Mythri P K <mythripk@ti.com> Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Add initial support for DSI video mode panels:
- Add a new structure omap_dss_dsi_videomode_data in the member "panel" in
omap_dss_device struct. This allows panel driver to configure dsi video_mode
specific parameters.
- Configure basic DSI video mode timing parameters: HBP, HFP, HSA, VBP, VFP, VSA,
TL and VACT.
- Configure DSI protocol engine registers for video_mode support.
- Introduce functions dsi_video_mode_enable() and dsi_video_mode_disable() which
enable/disable video mode for a given virtual channel and a given pixel format
type.
Things left for later
- Add functions to check for errors in video mode timings provided by panel.
- Configure timing registers required for command mode interleaving.
Signed-off-by: Archit Taneja <archit@ti.com> Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
OMAP: DSS2: DSI: Send zero length packet in dsi_vc_send_null()
dsi_vc_send_null() currently sends a long packet with data type
MIPI_DSI_NULL_PACKET and packet length 4. Modify it to send a zero
length long packet. This leads to sending only the long packet header
and no payload packets and hence the transfer completes faster.
The function can be modified later if there is a need to send null
packets of a non-zero length.
Signed-off-by: Archit Taneja <archit@ti.com> Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Currently, DSI pixel info is only represented by the pixel size in bits using
the pixel_size parameter in omap_dss_device struct's ctrl member.
This is not sufficient information for DSI video mode usage, as two of the
supported formats(RGB666 loosely packed, and RGB888) have the same pixel
container size, but different data_type values for the video mode packet header.
Create enum "omap_dss_dsi_pixel_format" which describes the pixel data format
the panel is configured for. Create helper function dsi_get_pixel_size() which
returns the pixel size of the given pixel format.
Modify functions omapdss_default_get_recommended_bpp() and dss_use_replication()
to use dsi_get_pixel_size().
Signed-off-by: Archit Taneja <archit@ti.com> Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Archit Taneja [Mon, 22 Aug 2011 12:11:57 +0000 (17:41 +0530)]
OMAP: DSS2: Clean up stallmode and io pad mode selection
Split the function dispc_set_parallel_interface_mode() into 2 separate
functions called dispc_mgr_set_io_pad_mode() and dispc_mgr_enable_stallmode().
The current function tries to set 2 different modes(io pad mode and stall mode)
based on a parameter omap_parallel_interface_mode which loosely corresponds to
the panel interface type.
This isn't correct because a) these 2 modes are independent to some extent,
b) we are currently configuring gpout0/gpout1 for DSI panels which is
unnecessary, c) a DSI Video mode panel won't get configured correctly.
Splitting the functions allows the interface driver to set these modes
independently and hence allow more flexibility.
Signed-off-by: Archit Taneja <archit@ti.com> Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Archit Taneja [Tue, 30 Aug 2011 10:37:39 +0000 (16:07 +0530)]
OMAP: DSS2: DSI: Introduce generic read functions
Introduce read functions which use generic Processor-to-Peripheral
transaction types. These are needed by some devices which may not support
corresponding DCS commands.
Add function dsi_vc_generic_send_read_request() which can send
a short packet with 0, 1 or 2 bytes of request data and the corresponding
generic data type.
Rename function dsi_vc_dcs_read_rx_fifo() to dsi_vc_read_rx_fifo() and modify
it to take the enum "dss_dsi_content_type" as an argument to use either DCS
or GENERIC Peripheral-to-Processor transaction types while parsing data read
from the device.
Signed-off-by: Archit Taneja <archit@ti.com> Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Archit Taneja [Tue, 30 Aug 2011 10:18:23 +0000 (15:48 +0530)]
OMAP: DSS2: DSI: Split dsi_vc_dcs_read() into 2 functions
Split dsi_vc_dcs_read() into the functions:
- dsi_vc_dcs_send_read_request(): This is responsible for sending the short
packet command with the read request.
- dsi_vc_dcs_read_rx_fifo(): This parses the DSI RX fifo of the given virtual
channel, identifies the type of data received, and fills a buffer with the data
provided by the panel.
Signed-off-by: Archit Taneja <archit@ti.com> Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Archit Taneja [Mon, 22 Aug 2011 09:14:27 +0000 (14:44 +0530)]
OMAP: DSS2: DSI: Remove functions dsi_vc_dcs_read_1() and dsi_vc_dcs_read_2()
Remove functions dsi_vc_dcs_read_1() and dsi_vc_dcs_read_2(), these are used
when the panel is expected to return 1 and 2 bytes respecitvely. This was manily
used for debugging purposes. These functions should be implemented in the panel
driver if needed.
Signed-off-by: Archit Taneja <archit@ti.com> Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Intoduce enum "dss_dsi_content_type" to differentiate between DCS and generic
content types.
Introduce short and long packet write functions which use generic
Processor-to-Peripheral transaction types. These are needed by some devices
which may not support corresponding DCS commands. Create common write functions
which allow code reuse between DCS and generic write functions.
Signed-off-by: Archit Taneja <archit@ti.com> Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Archit Taneja [Mon, 22 Aug 2011 06:28:08 +0000 (11:58 +0530)]
OMAP: DSS2: DSI: Represent L4 and VP as sources of VC instead of modes
The enum type dsi_vc_mode is a bit misleading as L4 slave port and video port
are sources to VC rather than the mode of operation. Rename then enum type and
its members. Merge dsi_vc_config_vp() and dsi_vc_config_l4() into a single
function called dsi_vc_config_source() which takes dsi_vc_source enum as an
extra argument.
Signed-off-by: Archit Taneja <archit@ti.com> Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Archit Taneja [Thu, 25 Aug 2011 12:55:03 +0000 (18:25 +0530)]
OMAP: DSS2: Use MIPI DSI enums from include/video/mipi_display.h
MIPI DSI Transaction types and DCS commands are currently defined as
macros in dsi.c and panel-taal.c, remove these and replace them with
enum members defined in include/video/mipi_display.h.
Signed-off-by: Archit Taneja <archit@ti.com>
[tomi.valkeinen@ti.com: reformatted the commit message] Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Tomi Valkeinen [Tue, 16 Aug 2011 10:45:15 +0000 (13:45 +0300)]
OMAP: DSS2: DISPC: rename manager related funcs
Rename dispc's manager related functions as follows:
- Remove prepending underscores, which were originally used to inform
that the clocks needs to be enabled. This meaning is no longer valid.
- Prepend the functions with dispc_mgr_*
- Remove "channel" from the name, e.g. dispc_enable_channel ->
dispc_mgr_enable
The idea is to group manager related functions so that it can be deduced
from the function name that it writes to manager spesific registers.
All dispc_mgr_* functions have enum omap_channel as the first parameter.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Tomi Valkeinen [Tue, 16 Aug 2011 10:25:00 +0000 (13:25 +0300)]
OMAP: DSS2: DISPC: rename overlay related funcs
Rename dispc's overlay related functions as follows:
- Remove prepending underscores, which were originally used to inform
that the clocks needs to be enabled. This meaning is no longer valid.
- Prepend the functions with dispc_ovl_*
- Remove "plane" from the name, e.g. dispc_set_plane_ba0 ->
dispc_ovl_set_ba0
The idea is to group overlay related functions so that it can be deduced
from the function name that it writes to overlay spesific registers.
All dispc_ovl_* functions have enum omap_plane as the first parameter.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Tomi Valkeinen [Mon, 15 Aug 2011 12:18:20 +0000 (15:18 +0300)]
OMAP: DSS2: Add GLOBAL_ALPHA & PRE_MULT_ALPHA to ovl caps
Add OMAP_DSS_OVL_CAP_GLOBAL_ALPHA and OMAP_DSS_OVL_CAP_PRE_MULT_ALPHA to
overlay capabilities. Use these instead of FEAT_GLOBAL_ALPHA,
FEAT_GLOBAL_ALPHA_VID1 and FEAT_PRE_MULT_ALPHA in code.
Remove FEAT_GLOBAL_ALPHA_VID1 and FEAT_PRE_MULT_ALPHA which are no
longer used. FEAT_GLOBAL_ALPHA is still used to decide if the HW has
global alpha register.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com> Acked-by: Archit Taneja <archit@ti.com>
Tomi Valkeinen [Mon, 15 Aug 2011 08:22:21 +0000 (11:22 +0300)]
OMAP: DSS2: Remove support for non-DISPC overlays
Remove support for non-DISPC overlays and overlay managers.
The support to possibly have non-DISPC overlays and managers was made to
make it possible to use CPU and/or sDMA to update RFBI or DSI command
mode displays. It is ok to remove the support, because:
- No one has used the feature.
- Display update without DISPC is very slow, so it is debatable if the
update would even be usable.
- Removal cleans up code.
- If such a feature is needed later, it is better implemented outside
omapdss driver.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com> Acked-by: Archit Taneja <archit@ti.com>
Tomi Valkeinen [Tue, 16 Aug 2011 09:56:19 +0000 (12:56 +0300)]
OMAP: DSS2: Handle manager change in apply
Currently when changing the manager of an overlay, set_manager() directly
calls dispc to set the overlay's destination.
Change this to be more in line with other overlay configurations, and
this will also remove the need to have dispc clocks enabled when calling
set_manager().
A new field is added to overlay struct, "manager_changed". This is
similar to "display_changed" field in manager struct, and is used to
inform apply that the manager has changed and thus write to the
registers is needed.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Tomi Valkeinen [Thu, 4 Aug 2011 11:37:29 +0000 (14:37 +0300)]
OMAP: OMAPFB: make omapfb start even when a display is missing a driver
Currently omapfb wants that all the display devices have a driver,
otherwise omapfb refuses to start. There's no real requirement to act
like that, and this patch will make omapfb give a warning and skip that
device.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Tomi Valkeinen [Wed, 15 Jun 2011 12:22:47 +0000 (15:22 +0300)]
OMAP: DSS2: Implement dsi_mux_pads for OMAP4
Implement dsi_mux_pads for OMAP4. On enable the function enables the DSI
pins and disables pull down. On disable the function disables the pins
and enables pull down.
It is unclear from the TRM whether the pull down is active if the pins
are disabled, so this implementation may leave the pins floating when
the DSI device is disabled.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Tomi Valkeinen [Wed, 15 Jun 2011 12:21:12 +0000 (15:21 +0300)]
OMAP: DSS2: DSI: Improve dsi_mux_pads parameters
dsi_mux_pads() needs to know about the DSI HW module and the DSI lanes
used. Split the function into two, enable and disable, which take
necessary arguments, and add empty implementations for both.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Tomi Valkeinen [Wed, 3 Aug 2011 11:13:52 +0000 (14:13 +0300)]
OMAP4: TWL: Add common omapdss supplies
OMAP DSS normally gets power from VCXIO on OMAP4. Add configuration for
this into twl-common.c
Mark VCXIO as always_on, as VCXIO is used by multiple components,
including the MPU, and turning it off when DSS doesn't need it would
lead the device to halt.
Acked-by: Tony Lindgren <tony@atomide.com> Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Tomi Valkeinen [Wed, 3 Aug 2011 11:00:57 +0000 (14:00 +0300)]
OMAP: DSS2: Change DSI device naming
Currently, there are 2 differently named platform devices generated for
the 2 DSS DSI modules. In order to use the same driver, the dsi devices
should be 2 instances of the same platform device.
Change the platform device names from "omapdss_dsi1" and "omapdss_dsi2"
to omapdss_dsi", and set the device indices to 0 and 1.
Signed-off-by: Archit Taneja <archit@ti.com> Acked-by: Tony Lindgren <tony@atomide.com> Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Tomi Valkeinen [Tue, 2 Aug 2011 08:54:39 +0000 (11:54 +0300)]
OMAP: DSS2: remove unneeded fck enable/disables
Now that the HWMOD fmwk handles the fcks of DSS modules properly, the
DSS driver no longer needs to explicitely enable/disable the fck.
This patch removes the enables/disables of fck from dispc, dsi and dss.
The clk_get(fck) is still needed there, as the modules need to know the
frequency of the clock.
For hdmi and venc this patch also removes the clk_get(fck), as they
don't need the clock at all.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Archit Taneja [Fri, 5 Aug 2011 13:36:04 +0000 (19:06 +0530)]
OMAP: DSS2: Use a macro to declare size of the fifo_size array in dispc.c
The array size of fifo_size array in the global dispc struct is currently
hardcoded to 3. Replace this with the MAX_DSS_OVERLAYS macro in dss_features.h,
use dss_features function to get the number of overlays instead of the
ARRAY_SIZE macro in dispc_read_plane_fifo_sizes().
Signed-off-by: Archit Taneja <archit@ti.com> Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Archit Taneja [Fri, 5 Aug 2011 13:36:00 +0000 (19:06 +0530)]
OMAP: DSS2: DISPC: Prepare dispc_dump_regs() for shortening
Prepare dispc_dump_regs() to iterate over manager and overlay id's. Doing this
requires modifications of the macro "DUMPREG" which currently needs us to specify
the manager/overlay name to get the correct result. For example, in order to
print the register DISPC_TIMING_H(OMAP_DSS_CHANNEL_LCD), we can't iterate over
a varaible i and get the desired result through DUMPREG(DISPC_TIMING_H(i)).
Split the registers into 3 sections, the first with no arguments(common
registers), the second with one argument(manager/overlay id), and the third with
two arguments(overlay id and coefficient index), redefine DUMPREG macros for
each of these.
Signed-off-by: Archit Taneja <archit@ti.com> Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Daniel Morsing [Wed, 3 Aug 2011 20:10:51 +0000 (22:10 +0200)]
OMAP: DSS2: Don't allow moving managers away from enabled displays
If a manager is moved while attached to an enabled display, the DSS
system will be left in an inconsistent state. This will eventually cause
a kernel oops when the enabled display is disabled.
Fix this by not allowing the user to move a manager away from an enabled
display.
Signed-off-by: Daniel Morsing <daniel.morsing@gmail.com> Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Tomi Valkeinen [Thu, 23 Jun 2011 13:38:21 +0000 (16:38 +0300)]
OMAP: DSS2: check for manager when enabling display
None of the DSS interface drivers check if an overlay manager is
connected to the display when the display is being enabled. This leads
to null pointer crash if the display has no manager.
This patch checks for the manager and returns an error if it is null.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Tomi Valkeinen [Thu, 23 Jun 2011 13:36:36 +0000 (16:36 +0300)]
OMAP: DSS2: PicoDLP: fix error handling in power_on
Fix two problems in PicoDLP driver's error handling on
picodlp_panel_power_on:
- If omapdss_dpi_display_enable() failed, the its error value was not
returned
- If picodlp_i2c_init() failed, dssdev->state was erroneously set to
OMAP_DSS_DISPLAY_ACTIVE
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
DLP used in OMAP4 is dpp2600 (DLP Pico Projector) The DLP requires
commands to be sent over i2c for configurations. To know more about
dpp2600 commands please visit:
https://focus.ti.com/myti/docs/extranet.tsp?sectionId=403
The picodlp module consists of a dss driver and an i2c_client.
To know more please visit:
http://www.omappedia.org/wiki/PicoDLP_projector_guide
Based on original design from Mythri P K <mythripk@ti.com>
Signed-off-by: Mayuresh Janorkar <mayur@ti.com> Signed-off-by: Mythri P K <mythripk@ti.com>
[tomi.valkeinen@ti.com: squashed commits] Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Merge branch 'drm-fixes' of git://people.freedesktop.org/~airlied/linux
* 'drm-fixes' of git://people.freedesktop.org/~airlied/linux:
drm: Remove duplicate "return" statement
drm/nv04/crtc: Bail out if FB is not bound to crtc
drm/nouveau: fix nv04_sgdma_bind on non-"4kB pages" archs
drm/nouveau: properly handle allocation failure in nouveau_sgdma_populate
drm/nouveau: fix oops on pre-semaphore hardware
drm/nv50/crtc: Bail out if FB is not bound to crtc
drm/radeon/kms: fix DP detect and EDID fetch for DP bridges
Merge branch 'fixes' of git://git.linaro.org/people/arnd/arm-soc
* 'fixes' of git://git.linaro.org/people/arnd/arm-soc:
ARM: CSR: add missing sentinels to of_device_id tables
ARM: cns3xxx: Fix newly introduced warnings in the PCIe code
ARM: cns3xxx: Fix compile error caused by hardware.h removed
ARM: davinci: fix cache flush build error
ARM: davinci: correct MDSTAT_STATE_MASK
ARM: davinci: da850 EVM: read mac address from SPI flash
OMAP: omap_device: fix !CONFIG_SUSPEND case in _noirq handlers
OMAP2430: hwmod: musb: add missing terminator to omap2430_usbhsotg_addrs[]
OMAP3: clock: indicate that gpt12_fck and wdt1_fck are in the WKUP clockdomain
OMAP4: clock: fix compile warning
OMAP4: clock: re-enable previous clockdomain enable/disable sequence
OMAP: clockdomain: Wait for powerdomain to be ON when using clockdomain force wakeup
OMAP: powerdomains: Make all powerdomain target states as ON at init
The LTTng 2.0 kernel tracer (stand-alone module package, available at
http://lttng.org) uses the 0xF6 ioctl range for tracer control and
transport operations.
Merge branch 'for-linus' of git://github.com/chrismason/linux
* 'for-linus' of git://github.com/chrismason/linux:
Btrfs: add dummy extent if dst offset excceeds file end in
Btrfs: calc file extent num_bytes correctly in file clone
btrfs: xattr: fix attribute removal
Btrfs: fix wrong nbytes information of the inode
Btrfs: fix the file extent gap when doing direct IO
Btrfs: fix unclosed transaction handle in btrfs_cont_expand
Btrfs: fix misuse of trans block rsv
Btrfs: reset to appropriate block rsv after orphan operations
Btrfs: skip locking if searching the commit root in csum lookup
btrfs: fix warning in iput for bad-inode
Btrfs: fix an oops when deleting snapshots
If we write some data into the data hole of the file(no preallocation for this
hole), Btrfs will allocate some disk space, and update nbytes of the inode, but
the other element--disk_i_size needn't be updated. At this condition, we must
update inode metadata though disk_i_size is not changed(btrfs_ordered_update_i_size()
return 1).
Btrfs: fix the file extent gap when doing direct IO
When we write some data to the place that is beyond the end of the file
in direct I/O mode, a data hole will be created. And Btrfs should insert
a file extent item that point to this hole into the fs tree. But unfortunately
Btrfs forgets doing it.
The following is a simple way to reproduce it:
# mkfs.btrfs /dev/sdc2
# mount /dev/sdc2 /test4
# touch /test4/a
# dd if=/dev/zero of=/test4/a seek=8 count=1 bs=4K oflag=direct conv=nocreat,notrunc
# umount /test4
# btrfsck /dev/sdc2
root 5 inode 257 errors 100
Liu Bo [Sun, 11 Sep 2011 14:52:24 +0000 (10:52 -0400)]
Btrfs: fix misuse of trans block rsv
At the beginning of create_pending_snapshot, trans->block_rsv is set
to pending->block_rsv and is used for snapshot things, however, when
it is done, we do not recover it as will.
Signed-off-by: Liu Bo <liubo2009@cn.fujitsu.com> Signed-off-by: Chris Mason <chris.mason@oracle.com>
Liu Bo [Sun, 11 Sep 2011 14:52:24 +0000 (10:52 -0400)]
Btrfs: reset to appropriate block rsv after orphan operations
While truncating free space cache, we forget to change trans->block_rsv
back to the original one, but leave it with the orphan_block_rsv, and
then with option inode_cache enable, it leads to countless warnings of
btrfs_alloc_free_block and btrfs_orphan_commit_root:
WARNING: at fs/btrfs/extent-tree.c:5711 btrfs_alloc_free_block+0x180/0x350 [btrfs]()
...
WARNING: at fs/btrfs/inode.c:2193 btrfs_orphan_commit_root+0xb0/0xc0 [btrfs]()
Signed-off-by: Liu Bo <liubo2009@cn.fujitsu.com> Signed-off-by: Chris Mason <chris.mason@oracle.com>
Josef Bacik [Sun, 11 Sep 2011 14:52:24 +0000 (10:52 -0400)]
Btrfs: skip locking if searching the commit root in csum lookup
It's not enough to just search the commit root, since we could be cow'ing the
very block we need to search through, which would mean that its locked and we'll
still deadlock. So use path->skip_locking as well. Thanks,
Signed-off-by: Josef Bacik <josef@redhat.com> Signed-off-by: Chris Mason <chris.mason@oracle.com>
Signed-off-by: Sergei Trofimovich <slyfox@gentoo.org> CC: Konstantin Khlebnikov <khlebnikov@openvz.org> Tested-by: David Sterba <dsterba@suse.cz> CC: Josef Bacik <josef@redhat.com> CC: Chris Mason <chris.mason@oracle.com> Signed-off-by: Chris Mason <chris.mason@oracle.com>
When it comes to btrfs_lookup_dentry, we may set a snapshot's inode->i_ino
to BTRFS_EMPTY_SUBVOL_DIR_OBJECTID instead of BTRFS_FIRST_FREE_OBJECTID,
while the snapshot's location.objectid remains unchanged.
However, btrfs_ino() does not take this into account, and returns a wrong ino,
and causes the oops.
Signed-off-by: Liu Bo <liubo2009@cn.fujitsu.com> Signed-off-by: Chris Mason <chris.mason@oracle.com>
Florian Mickler [Wed, 10 Aug 2011 10:05:20 +0000 (07:05 -0300)]
[media] vp7045: fix buffer setup
dvb_usb_device_init calls the frontend_attach method of this driver which
uses vp7045_usb_ob. In order to have a buffer ready in vp7045_usb_op, it has to
be allocated before that happens.
Luckily we can use the whole private data as the buffer as it gets separately
allocated on the heap via kzalloc in dvb_usb_device_init and is thus apt for
use via usb_control_msg.
This fixes a
BUG: unable to handle kernel paging request at 0000000000001e78
reported by Tino Keitel and diagnosed by Dan Carpenter.
Cc: stable@kernel.org # For v3.0 and upper Tested-by: Tino Keitel <tino.keitel@tikei.de> Signed-off-by: Florian Mickler <florian@mickler.org> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
Jarod Wilson [Mon, 8 Aug 2011 20:20:40 +0000 (17:20 -0300)]
[media] nuvoton-cir: simplify raw IR sample handling
The nuvoton-cir driver was storing up consecutive pulse-pulse and
space-space samples internally, for no good reason, since
ir_raw_event_store_with_filter() already merges back to back like
samples types for us. This should also fix a regression introduced late
in 3.0 that related to a timeout change, which actually becomes correct
when coupled with this change. Tested with RC6 and RC5 on my own
nuvoton-cir hardware atop vanilla 3.0.0, after verifying quirky
behavior in 3.0 due to the timeout change.
Reported-by: Stephan Raue <sraue@openelec.tv> CC: Stephan Raue <sraue@openelec.tv> CC: stable@vger.kernel.org Signed-off-by: Jarod Wilson <jarod@redhat.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
Jesper Juhl [Mon, 1 Aug 2011 21:39:17 +0000 (18:39 -0300)]
[media] [Resend] viacam: Don't explode if pci_find_bus() returns NULL
In the unlikely case that pci_find_bus() should return NULL
viacam_serial_is_enabled() is going to dereference a NULL pointer and
blow up. Better safe than sorry, so be defensive and check the
pointer.
Dan Carpenter [Sat, 23 Jul 2011 18:53:03 +0000 (15:53 -0300)]
[media] pwc: precedence bug in pwc_init_controls()
'!' has higher precedence than '&' so we need parenthesis here.
Signed-off-by: Dan Carpenter <error27@gmail.com> Signed-off-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
Jonas Bonn [Mon, 5 Sep 2011 11:47:10 +0000 (13:47 +0200)]
Add missing DMA ops
For the initial architecture submission, not all of the DMA ops were
implemented. This patch adds the *map_page and *map_sg variants of the
DMA mapping ops.
This patch is currently of interest mainly to some drivers that haven't
been submitted upstream yet.
Merge branch 'fixes' of http://ftp.arm.linux.org.uk/pub/linux/arm/kernel/git-cur/linux-2.6-arm
* 'fixes' of http://ftp.arm.linux.org.uk/pub/linux/arm/kernel/git-cur/linux-2.6-arm:
ARM: 7088/1: entry: fix wrong parameter name used in do_thumb_abort
ARM: 7080/1: l2x0: make sure I&D are not locked down on init
ARM: 7081/1: mach-integrator: fix the clocksource
NET: am79c961: fix race in link status code
ARM: 7067/1: mm: keep significant bits in pfn_valid
ARM: 7088/1: entry: fix wrong parameter name used in do_thumb_abort
Commit be020f8618ca, "ARM: entry: abort-macro: specify registers to be
used for macros", while replacing register numbers with macro parameter
names, mismatched the name used for r1. For me, this resulted in user
space built for EABI with -march=armv4t -mtune=arm920t -mthumb-interwork
-mthumb broken on my OMAP1510 based Amstrad Delta (old ABI and no thumb
still worked for me though).
Fix this by using correct parameter name fsr instead of mismatched psr,
used by callers for another purpose.
Signed-off-by: Janusz Krzysztofik <jkrzyszt@tis.icnet.pl> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Since backlight_types[] isn't modified, let's declare it const. That
was probably the intention of the author of commit bb7ca747f8d6
("backlight: add backlight type"), via which the "const char const *"
construct was introduced. The duplicate const was detected by sparse.
Signed-off-by: Bart Van Assche <bvanassche@acm.org> Cc: Matthew Garrett <mjg@redhat.com> Cc: Richard Purdie <rpurdie@rpsys.net> Cc: Florian Tobias Schandinat <FlorianSchandinat@gmx.de> Cc: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Merge branch 'for-linus' of git://neil.brown.name/md
* 'for-linus' of git://neil.brown.name/md:
md: Fix handling for devices from 2TB to 4TB in 0.90 metadata.
md/raid1,10: Remove use-after-free bug in make_request.
md/raid10: unify handling of write completion.
Avoid dereferencing a 'request_queue' after last close.
md: Fix handling for devices from 2TB to 4TB in 0.90 metadata.
0.90 metadata uses an unsigned 32bit number to count the number of
kilobytes used from each device.
This should allow up to 4TB per device.
However we multiply this by 2 (to get sectors) before casting to a
larger type, so sizes above 2TB get truncated.
Also we allow rdev->sectors to be larger than 4TB, so it is possible
for the array to be resized larger than the metadata can handle.
So make sure rdev->sectors never exceeds 4TB when 0.90 metadata is in
used.
Also the sanity check at the end of super_90_load should include level
1 as it used ->size too. (RAID0 and Linear don't use ->size at all).
A write can complete at two different places:
1/ when the last member-device write completes, through
raid10_end_write_request
2/ in make_request() when we remove the initial bias from ->remaining.
These two should do exactly the same thing and the comment says they
do, but they don't.
So factor the correct code out into a function and call it in both
places. This makes the code much more similar to RAID1.
The difference is only significant if there is an error, and they
usually take a while, so it is unlikely that there will be an error
already when make_request is completing, so this is unlikely to cause
real problems.
Avoid dereferencing a 'request_queue' after last close.
On the last close of an 'md' device which as been stopped, the device
is destroyed and in particular the request_queue is freed. The free
is done in a separate thread so it might happen a short time later.
__blkdev_put calls bdev_inode_switch_bdi *after* ->release has been
called.
Since commit f758eeabeb96f878c860e8f110f94ec8820822a9
bdev_inode_switch_bdi will dereference the 'old' bdi, which lives
inside a request_queue, to get a spin lock. This causes the last
close on an md device to sometime take a spin_lock which lives in
freed memory - which results in an oops.
So move the called to bdev_inode_switch_bdi before the call to
->release.
Jon Mason [Thu, 8 Sep 2011 21:41:18 +0000 (16:41 -0500)]
PCI: Remove MRRS modification from MPS setting code
Modifying the Maximum Read Request Size to 0 (value of 128Bytes) has
massive negative ramifications on some devices. Without knowing which
devices have this issue, do not modify from the default value when
walking the PCI-E bus in pcie_bus_safe mode. Also, make pcie_bus_safe
the default procedure.
Tested-by: Sven Schnelle <svens@stackframe.org> Tested-by: Simon Kirby <sim@hostway.ca> Tested-by: Stephen M. Cameron <scameron@beardog.cce.hp.com> Reported-and-tested-by: Eric Dumazet <eric.dumazet@gmail.com> Reported-and-tested-by: Niels Ole Salscheider <niels_ole@salscheider-online.de>
References: https://bugzilla.kernel.org/show_bug.cgi?id=42162 Signed-off-by: Jon Mason <mason@myri.com> Acked-by: Jesse Barnes <jbarnes@virtuousgeek.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Fix pointer dereference before call to pcie_bus_configure_settings
Commit b03e7495a862 ("PCI: Set PCI-E Max Payload Size on fabric")
introduced a potential NULL pointer dereference in calls to
pcie_bus_configure_settings due to attempts to access pci_bus self
variables when the self pointer is NULL.
To correct this, verify that the self pointer in pci_bus is non-NULL
before dereferencing it.
Merge branch 'for-linus' of git://dev.laptop.org/users/cjb/mmc
* 'for-linus' of git://dev.laptop.org/users/cjb/mmc:
mmc: sdhci-s3c: Fix mmc card I/O problem
mmc: sd: UHS-I bus speed should be set last in UHS initialization
mmc: sdhi: initialise mmc_data->flags before use
mmc: core: use non-reentrant workqueue for clock gating
mmc: core: prevent aggressive clock gating racing with ios updates
mmc: rename mmc_host_clk_{ungate|gate} to mmc_host_clk_{hold|release}
mmc: sdhci-esdhc-imx: add missing inclusion of linux/module.h
projects / karo-tx-linux.git / log