Michael Neuling [Sat, 28 Jan 2012 19:35:24 +0000 (19:35 +0000)]
drivers/video: compile fixes for fsl-diu-fb.c
Fix a compiler errors introduced in:
commit ddd3d905436b572ebadc09dcf2d12ca5b37020a0
Author: Timur Tabi <timur@freescale.com>
drivers/video: fsl-diu-fb: merge all allocated data into one block
Signed-off-by: Michael Neuling <mikey@neuling.org> Acked-by: Timur Tabi <timur@freescale.com> Signed-off-by: Florian Tobias Schandinat <FlorianSchandinat@gmx.de>
Tomi Valkeinen [Tue, 17 Jan 2012 09:09:57 +0000 (11:09 +0200)]
OMAPDSS: HDMI: PHY burnout fix
A hardware bug in the OMAP4 HDMI PHY causes physical damage to the board
if the HDMI PHY is kept powered on when the cable is not connected.
This patch solves the problem by adding hot-plug-detection into the HDMI
IP driver. This is not a real HPD support in the sense that nobody else
than the IP driver gets to know about the HPD events, but is only meant
to fix the HW bug.
The strategy is simple: If the display device is turned off by the user,
the PHY power is set to OFF. When the display device is turned on by the
user, the PHY power is set either to LDOON or TXON, depending on whether
the HDMI cable is connected.
The reason to avoid PHY OFF when the display device is on, but the cable
is disconnected, is that when the PHY is turned OFF, the HDMI IP is not
"ticking" and thus the DISPC does not receive pixel clock from the HDMI
IP. This would, for example, prevent any VSYNCs from happening, and
would thus affect the users of omapdss. By using LDOON when the cable is
disconnected we'll avoid the HW bug, but keep the HDMI working as usual
from the user's point of view.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Tomi Valkeinen [Tue, 17 Jan 2012 08:59:00 +0000 (10:59 +0200)]
OMAPDSS: remove wrong HDMI HPD muxing
"hdmi_hpd" pin is muxed to INPUT and PULLUP, but the pin is not
currently used, and in the future when it is used, the pin is used as a
GPIO and is board specific, not an OMAP4 wide thing.
So remove the muxing for now.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com> Acked-by: Tony Lindgren <tony@atomide.com>
Tomi Valkeinen [Tue, 17 Jan 2012 08:49:38 +0000 (10:49 +0200)]
OMAP: 4430SDP/Panda: rename HPD GPIO to CT_CP_HPD
The GPIO 60 on 4430sdp and Panda is not HPD GPIO, as currently marked in
the board files, but CT_CP_HPD, which is used to enable/disable HPD
functionality.
This patch renames the GPIO.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com> Acked-by: Tony Lindgren <tony@atomide.com>
Tomi Valkeinen [Mon, 23 Jan 2012 11:23:08 +0000 (13:23 +0200)]
OMAPDSS: use sync versions of pm_runtime_put
omapdss doesn't work properly on system suspend. The problem seems to be
the fact that omapdss uses pm_runtime_put() functions when turning off
the hardware, and when system suspend is in process only sync versions
are allowed.
Using non-sync versions normally and sync versions when suspending would
need rather ugly hacks to convey the information of
suspending/not-suspending to different functions. Optimally the driver
wouldn't even need to care about this, and the PM layer would handle
syncing when suspend is in process.
This patch changes all omapdss's pm_runtime_put calls to
pm_runtime_put_sync. This fixes the suspend problem, and probably the
performance penalty of always using sync versions is negligible.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com> Acked-by: Kevin Hilman <khilman@ti.com>
Nicolas Ferre <nicolas.ferre@atmel.com> wrote:
"Unfortunately this is not true for all the SoC that embed the
atmel_lcdfb... So I may need to rework this patch but it is certainly
not applicable in the current form."
Mythri P K [Mon, 2 Jan 2012 08:32:38 +0000 (14:02 +0530)]
OMAPDSS: HDMI: Disable DDC internal pull up
Disables the internal pull resistor for SDA and SCL which are enabled by
default, as there are external pull up's in 4460 and 4430 ES2.3
SDP, Blaze and Panda Boards, It is done to avoid the EDID read failure.
Signed-off-by: Ricardo Salveti de Araujo <ricardo.salveti@linaro.org> Signed-off-by: Mythri P K <mythripk@ti.com> Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Daniel Mack [Tue, 20 Dec 2011 09:54:32 +0000 (10:54 +0100)]
OMAP: DSS2: Support for UMSH-8173MD TFT panel
This patch adds support for Microtip Technologies' UMSH-8173MD 800x480
RGB TFT panel. Tested with an OMAP3 board.
Signed-off-by: Daniel Mack <zonque@gmail.com> Cc: Tomi Valkeinen <tomi.valkeinen@ti.com> Cc: Florian Tobias Schandinat <FlorianSchandinat@gmx.de>
[tomi.valkeinen@ti.com: changed the panel name string] Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Ricardo Neri [Sun, 27 Nov 2011 22:18:01 +0000 (16:18 -0600)]
ASoC: OMAP: HDMI: Move HDMI codec trigger function to generic HDMI driver
The function hdmi_audio_trigger is a callback used by ASoC to stop/start
HDMI audio. Also, it does not perform IP-specific configuration directly.
Hence, it should be placed in the general portion of the HDMI driver,
along with the other ASoC callbacks.
Also, make this function static.
Signed-off-by: Ricardo Neri <ricardo.neri@ti.com> Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Ricardo Neri [Sun, 27 Nov 2011 22:09:58 +0000 (16:09 -0600)]
OMAPDSS: HDMI: Create function to enable HDMI audio
In order to separate clearly IP-specific code from general DSS code,
a function for OMAP4 audio enable is created. This function is
included in the HDMI IP ops to align with the current implementation
of the DSS HDMI driver. This function is to be used by the ASoC
HDMI audio codec.
Signed-off-by: Ricardo Neri <ricardo.neri@ti.com> Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Ricardo Neri [Wed, 23 Nov 2011 02:50:50 +0000 (20:50 -0600)]
ASoC: OMAP: HDMI: Correct signature of ASoC functions
These functions require access to IP-secific data. However, it is not possible
to pass such data as a function argument as such functions have a specific
signature specified by ASoC. Instead, they will have access to the IP-specific
data by calling snd_soc_codec_get_drvdata. The codec driver data is set
at probe time.
Signed-off-by: Ricardo Neri <ricardo.neri@ti.com> Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Ricardo Neri [Wed, 23 Nov 2011 02:09:41 +0000 (20:09 -0600)]
ASoC: OMAP: HDMI: Introduce driver data for audio codec
Under the new DSS architecture for HDMI, there is a clear separation
between general DSS code and HDMI IP-specific data. Functions
that require access to the HDMI driver IP-specific data receive an
hdmi_ip_data structure. The ASoC codec require access to such
IP-specific data. Then, instead of accessing it directly, it will be
passed as codec driver data. This also helps to have a clear separation
between DSS and ASoC portions of the code.
Signed-off-by: Ricardo Neri <ricardo.neri@ti.com> Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Sam Ravnborg [Tue, 27 Dec 2011 21:55:49 +0000 (22:55 +0100)]
grvga: fix section mismatch warnings
Fix following section mismatch warnings:
WARNING: drivers/video/built-in.o(.devinit.text+0x110): Section mismatch in reference from the function grvga_probe() to the function .init.text:grvga_parse_custom()
The function __devinit grvga_probe() references
a function __init grvga_parse_custom().
If grvga_parse_custom is only used by grvga_probe then
annotate grvga_parse_custom with a matching annotation.
WARNING: drivers/video/built-in.o(.devinit.text+0x1f8): Section mismatch in reference from the function grvga_probe() to the variable .init.data:grvga_fix
The function __devinit grvga_probe() references
a variable __initdata grvga_fix.
If grvga_fix is only used by grvga_probe then
annotate grvga_fix with a matching annotation.
WARNING: drivers/video/built-in.o(.devinit.text+0x204): Section mismatch in reference from the function grvga_probe() to the variable .init.data:grvga_fix
The function __devinit grvga_probe() references
a variable __initdata grvga_fix.
If grvga_fix is only used by grvga_probe then
annotate grvga_fix with a matching annotation.
grvga_fix is used in a function annotated __devinit - so
match this using a __devinitdata annotation on grvga_fix.
grvga_parse_custom() is used in a function annotated
__devinit - so match this by annotating grvga_parse_custom()
with __devinit too.
Mark Brown [Tue, 27 Dec 2011 14:16:12 +0000 (14:16 +0000)]
video: s3c-fb: Don't keep device runtime active when open
Allow the controller to be runtime suspended when the screen is blanked
by not taking a runtime reference while the device is open. This allows
greater system wide power savings when used with a standard application
layer and ensures that the screen does not blank unless requested.
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com> Acked-by: Jingoo Han <jg1.han@samsung.com> Signed-off-by: Florian Tobias Schandinat <FlorianSchandinat@gmx.de>
Mark Brown [Tue, 27 Dec 2011 14:16:11 +0000 (14:16 +0000)]
video: s3c-fb: Hold runtime PM references when touching registers
Take a runtime PM reference whenever updating registers in preparation
for suspending the device when the framebuffer is blanked.
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com> Acked-by: Jingoo Han <jg1.han@samsung.com> Signed-off-by: Florian Tobias Schandinat <FlorianSchandinat@gmx.de>
Mark Brown [Tue, 27 Dec 2011 14:16:10 +0000 (14:16 +0000)]
video: s3c-fb: Take a runtime PM reference when unblanked
When the framebuffer is unblanked hold a runtime PM reference. This
prevents us powering down when userspace has left an image on the
framebuffer and prepares the way for being able to power down the hardware
when an application still has the device open.
Since we now hold a runtime PM reference whenever the display is unblanked
there is no need for the runtime power management to disable and enable
the display, and doing so would lead to runtime PM trying to recurse into
itself when called from the blanking code, so split the runtime PM into
separate functions which only deal with the clocks. The PM core will
runtime resume the device prior to system suspend.
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com> Acked-by: Jingoo Han <jg1.han@samsung.com> Signed-off-by: Florian Tobias Schandinat <FlorianSchandinat@gmx.de>
Mark Brown [Tue, 27 Dec 2011 14:16:09 +0000 (14:16 +0000)]
video: s3c-fb: Disable runtime PM in error paths from probe
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com> Acked-by: Jingoo Han <jg1.han@samsung.com> Signed-off-by: Florian Tobias Schandinat <FlorianSchandinat@gmx.de>
Mark Brown [Tue, 27 Dec 2011 14:16:08 +0000 (14:16 +0000)]
video: s3c-fb: Use s3c_fb_enable() to enable the framebuffer
The s3c-fb driver has a function called s3c_fb_enable() which turns on
and off the physical output. However it is only actually used in paths
which disable the screen, the enabling just writes to the register. Make
the code less confusing by ensuring that the enable also goes through
the same path.
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com> Acked-by: Jingoo Han <jg1.han@samsung.com> Signed-off-by: Florian Tobias Schandinat <FlorianSchandinat@gmx.de>
Mark Brown [Tue, 27 Dec 2011 14:16:07 +0000 (14:16 +0000)]
video: s3c-fb: Make runtime PM functional again
The change in "video: s3c-fb: modify runtime pm functions" (commit
35784b) renders the runtime power management for the device completely
ineffectual as while it leaves runtime power management notionally
enabled a runtime power reference is held for the entire time the device
is registered meaning it will never actually do anything.
A further issue is introduced as runtime power management is added
during the system suspend path which is not something which drivers are
supposed to do and would interact poorly if there were any operations
done in the runtime power management callbacks.
While this does make things simpler (the main motivation for the
original change) it will not only cause us to use more power in the
framebuffer controller but will also prevent us entering lower power
domain and SoC wide states as we can never power down the domain
containing the device. Since neither of these things is desirable
revert the change.
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com> Acked-by: Jingoo Han <jg1.han@samsung.com> Signed-off-by: Florian Tobias Schandinat <FlorianSchandinat@gmx.de>
The DIU driver requires some platform-specific functions to be defined,
but two them can be optional because most platforms implement them the
same way. Functions set_gamma_table() and get_pixel_format() are only
needed because of quirks in the Freescale MPC8610 HPCD reference board.
For other boards, a generic implementation works, so we shouldn't
require the platform code to define them.
Signed-off-by: Timur Tabi <timur@freescale.com> Signed-off-by: Florian Tobias Schandinat <FlorianSchandinat@gmx.de>
Timur Tabi [Mon, 19 Dec 2011 22:26:16 +0000 (16:26 -0600)]
drivers/video: fsl-diu-fb: remove broken reference count enabling the display
The functions enable_lcdc() and disable_lcdc() used the variable fb_enable
to keep a reference count of when the display is enabled, but the code is
broken. The display is always disabled when disable_lcdc(), and it is
always enabled when enable_lcdc() is called, regardless of the value of
fb_enable. Obviously, we don't need to keep a reference count, so just
remove the variable.
Signed-off-by: Timur Tabi <timur@freescale.com> Signed-off-by: Florian Tobias Schandinat <FlorianSchandinat@gmx.de>
Timur Tabi [Mon, 19 Dec 2011 22:26:14 +0000 (16:26 -0600)]
drivers/video: fsl-diu-fb: merge init_fbinfo() into install_fb()
Function init_fbinfo() is called only from install_fb(), and it's only a few
lines long. Plus, it ignores the return code from fb_alloc_cmap(). Merge
its contents into install_fb() and handle errors properly.
Signed-off-by: Timur Tabi <timur@freescale.com> Signed-off-by: Florian Tobias Schandinat <FlorianSchandinat@gmx.de>
Clock requirements for scaling in OMAP2, OMAP3 and OMAP4 are different. In
OMAP2 and OMAP3 the required clock rate is a function of pixel clock, vertical
downscale ratio and horizontal downscale ratio whereas in OMAP4 it is a
function of pixel clock and horizontal downscale ratio only. Selection of 3-tap
vs 5-tap coefficients depends on clock rate line buffer width in OMAP3 whereas
in OMAP4 it is independent of clock rate and line buffer width. In OMAP2 3-tap
for vertical and 5-tap for horizontal scaling is used. In OMAP4 5-tap is used
both for horizontal and vertical scaling for better performance. Also, the
number and width of line buffers differs in OMAP3 and OMAP4.
So, clock functions have been fined tuned for OMAP3 and support has been added
added for OMAP4. This code has been tested on OMAP2, OMAP3 and OMAP4, and
scaling issues due to clock errors have been resolved.
Signed-off-by: Chandrabhanu Mahapatra <cmahapatra@ti.com> Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
The FIR coefficients present in kernel are being updated to new coefficients
consisting of 24 coefficient tables, with 12 each for 3 tap and 5 tap scenario,
which are chosen on the basis of DISPC up/downsampling filters M value. M is
the inverse of low pass cut off frequency of the sampling filter. For vertical
scaling 3 tap or 5 tap tables are used based on the clock rate and width of
the line buffer whereas in OMAP2 3 tap is always used. For horizontal scaling
however 5 tap tables are always used.
New coefficients and the corresponding logic have been tested on OMAP2, OMAP3
and OMAP4. Horizontal and vertical scaling worked fine except for some 3 tap
vs 5 tap issue during vertical upscaling and clock failing issues which is
acknowledged in the next patch. Vertical upscaling was found to perform better
under 5 taps. The 24 coefficient tables have been moved to another file
dispc_coefs.c for proper maintainance.
This code is written based on code written by Lajos Molnar <lajos@ti.com> in
Android Kernel for scaling. Lajos Molnar <lajos@ti.com> had fine tuned the FIR
coefficient selection process and reduced outliness and blockiness around
images when upscaling more than 2 times.
Signed-off-by: Chandrabhanu Mahapatra <cmahapatra@ti.com> Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Rob Clark [Sat, 17 Dec 2011 19:28:52 +0000 (13:28 -0600)]
OMAPDSS: APPLY: fix NULL pointer deref when mgr is not set
extra_info_update_ongoing() goes through all overlays, but doesn't check
if the overlay is connected to a manager. This leads to a crash whenever
an overlay has been detached.
Add a check to skip the non-connected overlays.
Reported-by: Rob Clark <rob@ti.com> Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Archit Taneja [Mon, 12 Dec 2011 06:17:42 +0000 (11:47 +0530)]
OMAPDSS: Panel NEC: Set omap_dss_device states correctly
The display state parameter of omap_dss_device struct is not being set correctly
in the panel driver NEC panel driver panel-nec-nl8048hl11-01b.c. Set the correct
states in the panel's enable/disable/suspend/resume functions.
CC: Erik Gilling <konkers@android.com> Signed-off-by: Archit Taneja <archit@ti.com> Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Archit Taneja [Mon, 12 Dec 2011 06:17:41 +0000 (11:47 +0530)]
OMAPDSS: DSI: Fix HSDIV related PLL info in dsi_dump_clocks()
The clock names of DSI_PLL_HSDIV_DISPC and DSI_PLL_HSDIV_DSI was made
dynamic based on the current value of DISPC and DSI FCLK sources. This
doesn't need to be done since we are just interested in the clock names,
and not the current clock sources for DISPC and DSI FCLKs.
Use only the generic and omap specific names for the DSI PLL's HSDIV
clocks.
Signed-off-by: Archit Taneja <archit@ti.com> Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Tomi Valkeinen [Tue, 13 Dec 2011 11:18:52 +0000 (13:18 +0200)]
OMAPDSS: APPLY: move simple_check functions
The functions dss_ovl_simple_check() and dss_mgr_simple_check() are not
really part of the apply mechanism, and can be moved to overlay.c and
manager.c.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Rob Clark [Sun, 11 Dec 2011 20:02:27 +0000 (14:02 -0600)]
OMAPDSS: fix potential NULL pointer ref in OCP_ERR handling path
The dispc's error handler tries to disable all outputs when OCP_ERR
happens. However, the code doesn't check if there actually is a display
on each particular output, nor if there's a driver for the display. This
may lead to NULL pointer reference.
Signed-off-by: Rob Clark <rob@ti.com>
[tomi.valkeinen@ti.com: added patch description] Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Axel Lin [Fri, 9 Dec 2011 01:59:56 +0000 (09:59 +0800)]
video: omap: convert drivers/video/omap/* to use module_platform_driver()
This patch converts the drivers in drivers/video/omap/* to use the
module_platform_driver() macro which makes the code smaller and a bit
simpler.
Cc: Jonathan McDowell <noodles@earth.li> Cc: Cory Maccarrone <darkstar6262@gmail.com> Cc: Laurent Gonzalez <palmte.linux@free.fr> Cc: Marek Vasut <marek.vasut@gmail.com> Signed-off-by: Axel Lin <axel.lin@gmail.com> Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Tomi Valkeinen [Thu, 8 Dec 2011 08:32:37 +0000 (10:32 +0200)]
OMAPDSS: APPLY: move check functions
The functions dss_ovl_check, dss_mgr_check_zorder, dss_mgr_check in
apply.c are not really part of the apply mechanism, and can be moved to
overlay.c and manager.c.
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ti.com>
Laurent Pinchart [Tue, 20 Dec 2011 19:00:11 +0000 (20:00 +0100)]
fbdev: matroxfb: Fix compilation after fb_var_screeninfo change
Commit fb21c2f42879 ("fbdev: Add FOURCC-based format configuration API")
modified the layout of the fb_var_screeninfo structure. Remove zero
fields from the static initializers in the matroxfb driver accordingly.
Ondrej Zary [Mon, 19 Dec 2011 22:39:47 +0000 (23:39 +0100)]
s3fb: fix Virge/VX
Add memory size detection for Virge/VX and small delay in mode setting (same
as in X.org driver) to fix blank screen problem.
Also adjust DTPC position to fix garbled screen in some modes (tested that
this adjustment does not break other cards - at least Trio32, Trio64V+,
Trio64V2/DX, Virge, Virge/DX).
Tested on ELSA Winner 2000AVI/3D.
Signed-off-by: Ondrej Zary <linux@rainbow-software.org> Signed-off-by: Florian Tobias Schandinat <FlorianSchandinat@gmx.de>
Axel Lin [Fri, 9 Dec 2011 01:57:53 +0000 (09:57 +0800)]
video: convert mbxfb to use module_platform_driver()
This patch converts mbxfb to use the module_platform_driver() macro
which makes the code smaller and a bit simpler.
Cc: Mike Rapoport <mike@compulab.co.il> Signed-off-by: Axel Lin <axel.lin@gmail.com> Signed-off-by: Florian Tobias Schandinat <FlorianSchandinat@gmx.de>
Jingoo Han [Mon, 5 Dec 2011 02:42:46 +0000 (11:42 +0900)]
video: s3c-fb: modify runtime pm functions
Runtime suspend and runtime resume are modified in order to
reduce the complexity and improve the usability of runtime pm.
After probe function, s3c-fb driver is not suspended until
suspend or remove is called.
The scheme is changed as follows:
runtime_get is only called in probe and resume.
runtime_put is only called in remove and suspend.
open/close cannot call the runtime_get/put.
Also, runtime_susepnd/resume are just called by runtime pm,
not doing suspend/resume routine any longer. This is because
open/close cannot call the runtime_get/put; the suspend/resume
routine in runtime_suspend/resume were previously used when
open and close were called.
The name of s3c-fb dev_pm_ops is changed from s3cfb_pm_ops to
s3c_fb_pm_ops in order to use more consistent naming.
Signed-off-by: Jingoo Han <jg1.han@samsung.com> Signed-off-by: Florian Tobias Schandinat <FlorianSchandinat@gmx.de>
Timur Tabi [Fri, 2 Dec 2011 23:03:27 +0000 (17:03 -0600)]
drivers/video: fsl-diu-fb: merge all allocated data into one block
The Freescale DIU driver allocates multiple blocks of memory, including
multiple DMA buffers. Merge all of these blocks into one data structure.
Specifically:
1) struct fsl_diu_data now contains everything that needs to be allocated,
except for the framebuffers themselves. DMA'able objects are aligned correctly
within the structure.
2) struct diu_addr is no longer needed, because we don't have to manage
multiple blocks of DMA memory.
3) Since there's no diu_addr any more, macro DMA_ADDR is used to calculate
the DMA address of any field in fsl_diu_data.
4) Functions allocate_buf() and free_buf() are no longer needed, because we
now assume that dma_alloc_coherent() will allocate a page-aligned block,
and everything is properly aligned with fsl_diu_data already, so we no longer
need to align any memory blocks ourselves.
5) The "dummy" area descriptor is now defined separately from the other
five ADs, so NUM_AOIS (previously called FSL_AOI_NUM) is now set to five
instead of six. Previously, all six were combined together to avoid a
separate call to allocate_buf() just for the dummy AD.
6) framebuffer_alloc() and framebuffer_release() are no longer used. The
framebuffer is initialized manually.
7) Error handling is simplified since there's only one memory buffer
allocated.
Signed-off-by: Timur Tabi <timur@freescale.com> Signed-off-by: Florian Tobias Schandinat <FlorianSchandinat@gmx.de>
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/sparc:
sparc32: Be less strict in matching %lo part of relocation.
sbus: convert drivers/sbus/char/* to use module_platform_driver()
bbc_i2c: Remove unneeded err variable
sparc: Use kmemdup rather than duplicating its implementation
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net:
ipv6: Check dest prefix length on original route not copied one in rt6_alloc_cow().
sch_gred: should not use GFP_KERNEL while holding a spinlock
ipip, sit: copy parms.name after register_netdevice
ipv6: Fix for adding multicast route for loopback device automatically.
ssb: fix init regression with SoCs
rtl8192{ce,cu,de,se}: avoid problems because of possible ERFOFF -> ERFSLEEP transition
mac80211: fix another race in aggregation start
fsl_pq_mdio: Clean up tbi address configuration
ppp: fix pptp double release_sock in pptp_bind()
net/fec: fix the use of pdev->id
ath9k: fix check for antenna diversity support
batman-adv: delete global entry in case of roaming
batman-adv: in case of roaming mark the client with TT_CLIENT_ROAM
Bluetooth: Correct version check in hci_setup
btusb: fix a memory leak in btusb_send_frame()
Bluetooth: bnep: Fix module reference
Bluetooth: cmtp: Fix module reference
Bluetooth: btmrvl: support Marvell Bluetooth device SD8797
Linus Torvalds [Fri, 16 Dec 2011 20:15:50 +0000 (12:15 -0800)]
Merge branches 'for-linus' and 'for-linus-3.2' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs:
Btrfs: unplug every once and a while
Btrfs: deal with NULL srv_rsv in the delalloc inode reservation code
Btrfs: only set cache_generation if we setup the block group
Btrfs: don't panic if orphan item already exists
Btrfs: fix leaked space in truncate
Btrfs: fix how we do delalloc reservations and how we free reservations on error
Btrfs: deal with enospc from dirtying inodes properly
Btrfs: fix num_workers_starting bug and other bugs in async thread
BTRFS: Establish i_ops before calling d_instantiate
Btrfs: add a cond_resched() into the worker loop
Btrfs: fix ctime update of on-disk inode
btrfs: keep orphans for subvolume deletion
Btrfs: fix inaccurate available space on raid0 profile
Btrfs: fix wrong disk space information of the files
Btrfs: fix wrong i_size when truncating a file to a larger size
Btrfs: fix btrfs_end_bio to deal with write errors to a single mirror
* 'for-linus-3.2' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs:
btrfs: lower the dirty balance poll interval
Linus Torvalds [Fri, 16 Dec 2011 19:27:56 +0000 (11:27 -0800)]
Merge branch 'drm-intel-fixes' of git://people.freedesktop.org/~keithp/linux
* 'drm-intel-fixes' of git://people.freedesktop.org/~keithp/linux:
drm/i915/dp: Dither down to 6bpc if it makes the mode fit
drm/i915: enable semaphores on per-device defaults
drm/i915: don't set unpin_work if vblank_get fails
drm/i915: By default, enable RC6 on IVB and SNB when reasonable
iommu: Export intel_iommu_enabled to signal when iommu is in use
drm/i915/sdvo: Include LVDS panels for the IS_DIGITAL check
drm/i915: prevent division by zero when asking for chipset power
drm/i915: add PCH info to i915_capabilities
drm/i915: set the right SDVO transcoder for CPT
drm/i915: no-lvds quirk for ASUS AT5NM10T-I
drm/i915: Treat pre-gen4 backlight duty cycle value consistently
drm/i915: Hook up Ivybridge eDP
drm/i915: add multi-threaded forcewake support
Linus Torvalds [Fri, 16 Dec 2011 19:14:42 +0000 (11:14 -0800)]
Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound:
ALSA: HDA: Use LPIB Position fix for Intel SCH Poulsbo
ALSA: hda: fix mute led polarity for HP laptops with buggy BIOS
ALSA: HDA: Set position fix to LPIB for an Atom/Poulsbo based device
ASoC: Fix hx4700 error handling to free gpios if snd_soc_register_card fails
ASoC: WM8958: correctly show firmware magic on mismatch
ASoC: mxs: Add appropriate MODULE_ALIAS()
ASoC: mxs: Add missing MODULE_LICENSE("GPL")
ASoC: Fix WM8996 24.576MHz clock operation
ASoC: Include linux/module.h for smdk2443_wm9710
ASoC: Fix a typo in jive_wm8750
ASoC: Fix build dependency for SND_SOC_JZ4740_CODEC
ASoC: Include linux/io.h for jz4740 codec
Linus Torvalds [Fri, 16 Dec 2011 19:13:49 +0000 (11:13 -0800)]
Merge branch 'gpio/merge' of git://git.secretlab.ca/git/linux-2.6
* 'gpio/merge' of git://git.secretlab.ca/git/linux-2.6:
gpio: Fix DA9052 GPIO build errors.
gpio: mpc8xxx: don't allow input-only pins to be output for MPC5121
gpio-ml-ioh: Add the irq_disable/irq_enable hooks for ml-ioh irq chip
gpio-ml-ioh: fix a bug in the interrupt handler
gpio: pl061: drop extra check for NULL platform_data
Linus Torvalds [Fri, 16 Dec 2011 18:07:50 +0000 (10:07 -0800)]
Merge branch 'fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc
* 'fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc:
ARM: OMAP: rx51: fix USB
ARM: OMAP: mcbsp: Fix possible memory corruption
arm/imx: fix power button on imx51 babbage board
ARM: imx: fix cpufreq build errors
ARM: mx5: add __initconst for fec pdata
MXC PWM: should active during DOZE/WAIT/DBG mode
ARM: EXYNOS: Fix build error without CONFIG_LOCAL_TIMERS
ARM: EXYNOS: Fix for stall in case of cpu hotplug or sleep
ARM: S5PV210: Set 1000ns as PWM backlight period on SMDKV210
ARM: SAMSUNG: remove duplicated header include
Wu Fengguang [Fri, 16 Dec 2011 17:32:57 +0000 (12:32 -0500)]
btrfs: lower the dirty balance poll interval
Tests show that the original large intervals can easily make the dirty
limit exceeded on 100 concurrent dd's. So adapt to as large as the
next check point selected by the dirty throttling algorithm.
Signed-off-by: Wu Fengguang <fengguang.wu@intel.com> Signed-off-by: Chris Mason <chris.mason@oracle.com>
v2: Rebased to current code, resulting in the fix applying to EDP panels as
well. Also changed from spatio-temporal to just spatial dithering on
pre-ironlake, to be conssitent (and less visual flicker)
Signed-off-by: Adam Jackson <ajax@redhat.com> Signed-off-by: Eric Anholt <eric@anholt.net> Tested-by: Eric Anholt <eric@anholt.net> Tested-by: Dirk Hohndel <hohndel@infradead.org> Signed-off-by: Keith Packard <keithp@keithp.com>
Eugeni Dodonov [Sat, 10 Dec 2011 01:16:37 +0000 (17:16 -0800)]
drm/i915: enable semaphores on per-device defaults
This adds a default setting for semaphores parameter, and enables
semaphores by default on IVB.
For now, as semaphores interaction with VTd causes random issues on
SNB, we do not enable them by default. But they can still be enabled
via the semaphores=1 kernel parameter.
v2: enables semaphores on SNB when IO remapping is disabled, with base
on Keith Packard patch.
CC: Daniel Vetter <daniel.vetter@ffwll.ch> CC: Ben Widawsky <ben@bwidawsk.net> CC: Keith Packard <keithp@keithp.com> CC: Jesse Barnes <jbarnes@virtuousgeek.org> CC: Chris Wilson <chris@chris-wilson.co.uk>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=42696
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=40564
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=41353
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=38862 Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch> Signed-off-by: Eugeni Dodonov <eugeni.dodonov@intel.com> Signed-off-by: Keith Packard <keithp@keithp.com>
Jesse Barnes [Mon, 29 Aug 2011 16:45:28 +0000 (09:45 -0700)]
drm/i915: don't set unpin_work if vblank_get fails
This fixes a race where we may try to finish a page flip and decrement
the refcount even if our vblank_get failed and we ended up with a
spurious flip pending interrupt.
Keith Packard [Thu, 17 Nov 2011 06:24:52 +0000 (22:24 -0800)]
drm/i915: By default, enable RC6 on IVB and SNB when reasonable
RC6 should always work on IVB, and should work on SNB whenever IO
remapping is disabled. RC6 never works on Ironlake. Make the default
value for the parameter follow these guidelines. Setting the value
to either 0 or 1 will force the specified behavior.
Signed-off-by: Keith Packard <keithp@keithp.com> Reviewed-by: Kenneth Graunke <kenneth@whitecape.org> Reviewed-by: Eugeni Dodonov <eugeni.dodonov@intel.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=38567 Cc: Ted Phelps <phelps@gnusto.com> Cc: Peter <pab1612@gmail.com> Cc: Lukas Hejtmanek <xhejtman@fi.muni.cz> Cc: Andrew Lutomirski <luto@mit.edu>
Eugeni Dodonov [Wed, 23 Nov 2011 18:42:14 +0000 (16:42 -0200)]
iommu: Export intel_iommu_enabled to signal when iommu is in use
In i915 driver, we do not enable either rc6 or semaphores on SNB when dmar
is enabled. The new 'intel_iommu_enabled' variable signals when the
iommu code is in operation.
Cc: Ted Phelps <phelps@gnusto.com> Cc: Peter <pab1612@gmail.com> Cc: Lukas Hejtmanek <xhejtman@fi.muni.cz> Cc: Andrew Lutomirski <luto@mit.edu> CC: Daniel Vetter <daniel.vetter@ffwll.ch> Cc: Eugeni Dodonov <eugeni.dodonov@intel.com> Signed-off-by: Keith Packard <keithp@keithp.com>
Chris Wilson [Mon, 20 Jun 2011 13:45:50 +0000 (14:45 +0100)]
drm/i915/sdvo: Include LVDS panels for the IS_DIGITAL check
We were checking whether the supplied edid matched the connector it was
read from. We do this in case a DDC read returns an EDID for another
device on a multifunction or otherwise interesting card. However, we
failed to include LVDS as a digital device and so rejecting an otherwise
valid EDID.
Fixes the detection of the secondary SDVO LVDS panel on the Libretto
W105.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by: Adam Jackson <ajax@redhat.com>
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=39216 Signed-off-by: Keith Packard <keithp@keithp.com>
Eugeni Dodonov [Thu, 10 Nov 2011 15:55:15 +0000 (13:55 -0200)]
drm/i915: prevent division by zero when asking for chipset power
This prevents an in-kernel division by zero which happens when we are
asking for i915_chipset_val too quickly, or within a race condition
between the power monitoring thread and userspace accesses via debugfs.
The issue can be reproduced easily via the following command:
while ``; do cat /sys/kernel/debug/dri/0/i915_emon_status; done
This is particularly dangerous because it can be triggered by
a non-privileged user by just reading the debugfs entry.
This issue was also found independently by Konstantin Belousov
<kostikbel@gmail.com>, who proposed a similar patch.
Reported-by: Konstantin Belousov <kostikbel@gmail.com> Acked-by: Jesse Barnes <jbarnes@virtuousgeek.org> Acked-by: Keith Packard <keithp@keithp.com> Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> Cc: <stable@vger.kernel.org> Signed-off-by: Eugeni Dodonov <eugeni.dodonov@intel.com> Signed-off-by: Keith Packard <keithp@keithp.com>
Paulo Zanoni [Fri, 14 Oct 2011 21:17:41 +0000 (18:17 -0300)]
drm/i915: add PCH info to i915_capabilities
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk> Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com> Signed-off-by: Keith Packard <keithp@keithp.com>
Paulo Zanoni [Fri, 14 Oct 2011 21:16:22 +0000 (18:16 -0300)]
drm/i915: set the right SDVO transcoder for CPT
v2: add a CPT-specific macro, make code cleaner
v3: fix commit message
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=41272 Cc: stable@kernel.org Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com> Reviewed-by: Chris Wilson <chris@chris-wilson.co.uk> Signed-off-by: Keith Packard <keithp@keithp.com>
Linus Torvalds [Thu, 15 Dec 2011 22:15:01 +0000 (14:15 -0800)]
Merge branch 'staging-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging
* 'staging-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/staging:
staging: r8712u: Add new USB ID
staging: tidspbridge: request dmtimer clocks on init
staging: tidspbridge: include module.h by default
Uwe Kleine-König [Thu, 15 Dec 2011 20:47:56 +0000 (21:47 +0100)]
ARM: unwinder: fix bisection to find origin in .idx section
The bisection implemented in unwind_find_origin() stopped to early. If
there is only a single entry left to check the original code just took
the end point as origin which might be wrong.
This was introduced in commit de66a979012d ("ARM: 7187/1: fix unwinding
for XIP kernels").
Tejun Heo [Thu, 15 Dec 2011 19:03:04 +0000 (20:03 +0100)]
block: don't kick empty queue in blk_drain_queue()
While probing, fd sets up queue, probes hardware and tears down the
queue if probing fails. In the process, blk_drain_queue() kicks the
queue which failed to finish initialization and fd is unhappy about
that.
Linus Torvalds [Thu, 15 Dec 2011 18:52:40 +0000 (10:52 -0800)]
Merge branch 'stable/for-linus-fixes-3.2' of git://git.kernel.org/pub/scm/linux/kernel/git/konrad/xen
* 'stable/for-linus-fixes-3.2' of git://git.kernel.org/pub/scm/linux/kernel/git/konrad/xen:
xen/swiotlb: Use page alignment for early buffer allocation.
xen: only limit memory map to maximum reservation for domain 0.
xen/swiotlb: Use page alignment for early buffer allocation.
This fixes an odd bug found on a Dell PowerEdge 1850/0RC130
(BIOS A05 01/09/2006) where all of the modules doing pci_set_dma_mask
would fail with:
ata_piix 0000:00:1f.1: enabling device (0005 -> 0007)
ata_piix 0000:00:1f.1: can't derive routing for PCI INT A
ata_piix 0000:00:1f.1: BMDMA: failed to set dma mask, falling back to PIO
The issue was the Xen-SWIOTLB was allocated such as that the end of
buffer was stradling a page (and also above 4GB). The fix was
spotted by Kalev Leonid which was to piggyback on git commit e79f86b2ef9c0a8c47225217c1018b7d3d90101c "swiotlb: Use page alignment
for early buffer allocation" which:
We could call free_bootmem_late() if swiotlb is not used, and
it will shrink to page alignment.
So alloc them with page alignment at first, to avoid lose two pages
Ian Campbell [Wed, 14 Dec 2011 12:16:08 +0000 (12:16 +0000)]
xen: only limit memory map to maximum reservation for domain 0.
d312ae878b6a "xen: use maximum reservation to limit amount of usable RAM"
clamped the total amount of RAM to the current maximum reservation. This is
correct for dom0 but is not correct for guest domains. In order to boot a guest
"pre-ballooned" (e.g. with memory=1G but maxmem=2G) in order to allow for
future memory expansion the guest must derive max_pfn from the e820 provided by
the toolstack and not the current maximum reservation (which can reflect only
the current maximum, not the guest lifetime max). The existing algorithm
already behaves this correctly if we do not artificially limit the maximum
number of pages for the guest case.
With this change "xl mem-set <domain> 512M" will successfully increase the
guest RAM (by reducing the balloon).
There is no change for dom0.
Reported-and-Tested-by: George Shuklin <george.shuklin@gmail.com> Signed-off-by: Ian Campbell <ian.campbell@citrix.com> Cc: stable@kernel.org Reviewed-by: David Vrabel <david.vrabel@citrix.com> Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Josef Bacik [Tue, 13 Dec 2011 21:04:54 +0000 (16:04 -0500)]
Btrfs: only set cache_generation if we setup the block group
A user reported a problem booting into a new kernel with the old format inodes.
He was panicing in cow_file_range while writing out the inode cache. This is
because if the block group is not cached we'll just skip writing out the cache,
however if it gets dirtied again in the same transaction and it finished caching
we'd go ahead and write it out, but since we set cache_generation to the transid
we think we've already truncated it and will just carry on, running into
cow_file_range and blowing up. We need to make sure we only set
cache_generation if we've done the truncate. The user tested this patch and
verified that the panic no longer occured. Thanks,
Reported-and-Tested-by: Klaus Bitto <klaus.bitto@gmail.com> Signed-off-by: Josef Bacik <josef@redhat.com>
Josef Bacik [Tue, 13 Dec 2011 17:55:58 +0000 (12:55 -0500)]
Btrfs: don't panic if orphan item already exists
I've been hitting this BUG_ON() in btrfs_orphan_add when running xfstest 269 in
a loop. This is because we will add an orphan item, do the truncate, the
truncate will fail for whatever reason (*cough*ENOSPC*cough*) and then we're
left with an orphan item still in the fs. Then we come back later to do another
truncate and it blows up because we already have an orphan item. This is ok so
just fix the BUG_ON() to only BUG() if ret is not EEXIST. Thanks,
Josef Bacik [Fri, 9 Dec 2011 18:26:22 +0000 (13:26 -0500)]
Btrfs: fix leaked space in truncate
We were occasionaly leaking space when running xfstest 269. This is because if
we failed to start the transaction in the truncate loop we'd just goto out, but
we need to break so that the inode is removed from the orphan list and the space
is properly freed. Thanks,
Josef Bacik [Fri, 9 Dec 2011 16:18:51 +0000 (11:18 -0500)]
Btrfs: fix how we do delalloc reservations and how we free reservations on error
Running xfstests 269 with some tracing my scripts kept spitting out errors about
releasing bytes that we didn't actually have reserved. This took me down a huge
rabbit hole and it turns out the way we deal with reserved_extents is wrong,
we need to only be setting it if the reservation succeeds, otherwise the free()
method will come in and unreserve space that isn't actually reserved yet, which
can lead to other warnings and such. The math was all working out right in the
end, but it caused all sorts of other issues in addition to making my scripts
yell and scream and generally make it impossible for me to track down the
original issue I was looking for. The other problem is with our error handling
in the reservation code. There are two cases that we need to deal with
1) We raced with free. In this case free won't free anything because csum_bytes
is modified before we dro the lock in our reservation path, so free rightly
doesn't release any space because the reservation code may be depending on that
reservation. However if we fail, we need the reservation side to do the free at
that point since that space is no longer in use. So as it stands the code was
doing this fine and it worked out, except in case #2
2) We don't race with free. Nobody comes in and changes anything, and our
reservation fails. In this case we didn't reserve anything anyway and we just
need to clean up csum_bytes but not free anything. So we keep track of
csum_bytes before we drop the lock and if it hasn't changed we know we can just
decrement csum_bytes and carry on.
Because of the case where we can race with free()'s since we have to drop our
spin_lock to do the reservation, I'm going to serialize all reservations with
the i_mutex. We already get this for free in the heavy use paths, truncate and
file write all hold the i_mutex, just needed to add it to page_mkwrite and
various ioctl/balance things. With this patch my space leak scripts no longer
scream bloody murder. Thanks,
projects / linux-beck.git / log