25 May, 2010
40 commits
-
Since the drivers probe call was changed from .init.text to .devinit.text
in commit c2e13037e6794bd0d9de3f9ecabf5615f15c160b ("platform-drivers:
move probe to .devinit.text in drivers/video") the fb_fix_screeninfo and
fb_var_screeninfo structures must be changed from .init.data to
.devinit.data, too.Also the drivers remove routine should be moved from .exit.text to
.devexit.textSigned-off-by: Henrik Kretzschmar
Cc: Paul Mundt
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
FBIO_WAITFORVSYNC is currently implemented by matroxfb, atyfb, intelfb and
more. All of them keep redefining the same FBIO_WAITFORVSYNC macro over
and over again, so move it to linux/fb.h and clean up those duplicate
defines.Signed-off-by: Grazvydas Ignotas
Cc: Ville Syrjala
Cc: Grant Likely
Cc: Maik Broemme
Cc: Petr Vandrovec
Cc: Benjamin Herrenschmidt
Cc: Krzysztof Helt
Cc: "Hiremath, Vaibhav"
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
This work includes the following:
- Implement handler for FBIO_WAITFORVSYNC ioctl.
- Allocate the data and palette buffers separately. A consequence of
this is that the palette and data loading is now done in different
phases. And that the LCD must be disabled temporarily after the palette
is loaded but this will only happen once after init and each time the
palette is changed. I think this is OK.- Allocate two (ping and pong) framebuffers from memory.
- Add pan_display handler which toggles the LCDC DMA registers between
the ping and pong buffers.Signed-off-by: Martin Ambrose
Cc: Chaithrika U S
Cc: Sudhakar Rajashekhara
Cc: Krzysztof Helt
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Fix up the section in the w100fb driver, by moving:
* w100fb_remove() from .text to .devexit.text
* w100_get_xtal_table() from .text to .devinit.text
* w100fb_init() from .devinit.text to .init.text
Signed-off-by: Henrik Kretzschmar
Acked-by: Uwe Kleine-König
Cc: [if "platform-drivers: move probe to .devinit.text in drivers/video" was merged]
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Fix up the sections in the vga16fb driver, by moving:
* the variables vga16_defined and vga16fb
from .init.data to .devinit.data* vga16fb_setup() from .text to .init.text
* vga16fb_remove() from .text. to .devexit.text
This fixes the following warnings issued by modpost:
WARNING: drivers/video/built-in.o(.devinit.text+0x1a420): Section mismatch in re
ference from the function vga16fb_probe() to the (unknown reference) .init.data:
(unknown)
The function __devinit vga16fb_probe() references
a (unknown reference) __initdata (unknown).
If (unknown) is only used by vga16fb_probe then
annotate (unknown) with a matching annotation.WARNING: drivers/video/built-in.o(.devinit.text+0x1a437): Section mismatch in reference from the function vga16fb_probe() to the variable .init.data:vga16fb_defined
The function __devinit vga16fb_probe() references
a variable __initdata vga16fb_defined.
If vga16fb_defined is only used by vga16fb_probe then
annotate vga16fb_defined with a matching annotation.WARNING: drivers/video/built-in.o(.devinit.text+0x1a457): Section mismatch in reference from the function vga16fb_probe() to the variable .init.data:vga16fb_fix
The function __devinit vga16fb_probe() references
a variable __initdata vga16fb_fix.
If vga16fb_fix is only used by vga16fb_probe then
annotate vga16fb_fix with a matching annotation.Signed-off-by: Henrik Kretzschmar
Acked-by: Uwe Kleine-König
Cc: [if "platform-drivers: move probe to .devinit.text in drivers/video" was merged]
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Fix up the section in the vfb driver, by moving the variables vfb_default
and vfb_fix from .init.data to .devinit.dataThis fixes the following warnings issued by modpost:
WARNING: drivers/video/vfb.o(.devinit.text+0xf8): Section mismatch in reference from the function vfb_probe() to the variable .init.data:vfb_default
The function __devinit vfb_probe() references
a variable __initdata vfb_default.
If vfb_default is only used by vfb_probe then
annotate vfb_default with a matching annotation.WARNING: drivers/video/vfb.o(.devinit.text+0x114): Section mismatch in reference from the function vfb_probe() to the variable .init.data:vfb_fix
The function __devinit vfb_probe() references
a variable __initdata vfb_fix.
If vfb_fix is only used by vfb_probe then
annotate vfb_fix with a matching annotation.Signed-off-by: Henrik Kretzschmar
Acked-by: Uwe Kleine-König
Cc: [if "platform-drivers: move probe to .devinit.text in drivers/video" was merged]
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Fix up the sections in the hgafb driver, by
* moving hga_default_var and hga_fix from .init.data to .devinit.data
* moving hga_detect() from .init.text to .devinit.text
* moving hga_fb_remove() from .text to .devexit.text
This fixes the following warnings issued by modpost:
WARNING: drivers/video/hgafb.o(.devinit.text+0x18): Section mismatch in referenc
e from the function hgafb_probe() to the function .init.text:hga_card_detect()
The function __devinit hgafb_probe() references
a function __init hga_card_detect().
If hga_card_detect is only used by hgafb_probe then
annotate hga_card_detect with a matching annotation.WARNING: drivers/video/hgafb.o(.devinit.text+0xfe): Section mismatch in referenc
e from the function hgafb_probe() to the variable .init.data:hga_fix
The function __devinit hgafb_probe() references
a variable __initdata hga_fix.
If hga_fix is only used by hgafb_probe then
annotate hga_fix with a matching annotation.WARNING: drivers/video/hgafb.o(.devinit.text+0x105): Section mismatch in reference from the function hgafb_probe() to the variable .init.data:hga_default_var
The function __devinit hgafb_probe() references
a variable __initdata hga_default_var.
If hga_default_var is only used by hgafb_probe then
annotate hga_default_var with a matching annotation.Signed-off-by: Henrik Kretzschmar
Acked-by: Uwe Kleine-König
Cc: [if "platform-drivers: move probe to .devinit.text in drivers/video" was merged]
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Fix the sections in the arcfb driver, by moving:
* the variables arcfb_fix and arcfb_var from .init.data to .devinit.data
* arcfb_remove() from .text to .devexit.text
This fixes the following warnings issued by modpost:
WARNING: drivers/video/built-in.o(.devinit.text+0x543): Section mismatch in reference from the function arcfb_probe() to the variable .init.data:arcfb_var
The function __devinit arcfb_probe() references
a variable __initdata arcfb_var.
If arcfb_var is only used by arcfb_probe then
annotate arcfb_var with a matching annotation.WARNING: drivers/video/built-in.o(.devinit.text+0x558): Section mismatch in reference from the function arcfb_probe() to the variable .init.data:arcfb_fix
The function __devinit arcfb_probe() references
a variable __initdata arcfb_fix.
If arcfb_fix is only used by arcfb_probe then
annotate arcfb_fix with a matching annotation.Signed-off-by: Henrik Kretzschmar
Acked-by: Uwe Kleine-König
Cc: [if "platform-drivers: move probe to .devinit.text in drivers/video" was merged]
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
These are different size buffers (40 chars vs 16), we may as well be
cautious.Signed-off-by: Dan Carpenter
Cc: Thomas Winischhofer
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
This lets us support the new BF527-EZKIT V2.1 via platform resources
tweaks only.Signed-off-by: Michael Hennerich
Signed-off-by: Mike Frysinger
Cc: Bryan Wu
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Now that the WM831x core uses genirq for the IRQ controller there is no
need to use the WM831x-specific wrappers to request interrupts so convert
to use genirq directly.Also use more meaningful strings to make /proc/interrupts more readily
legible.Signed-off-by: Mark Brown
Cc: Alessandro Zummo
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
The bug is an oops when dev_get_drvdata() returned null in
cmos_update_irq_enable(). The call tree looks like this:
rtc_dev_ioctl()
=> rtc_update_irq_enable()
=> cmos_update_irq_enable()It's caused by a race condition in the module initialization. It is
rtc_device_register() which makes the ioctl operations live so I moved
the call to dev_set_drvdata() before the call to rtc_device_register().Addresses https://bugzilla.kernel.org/show_bug.cgi?id=15963
Reported-by: Randy Dunlap
Signed-off-by: Dan Carpenter
Tested-by: Randy Dunlap
Cc: Alessandro Zummo
Cc: Paul Gortmaker
Cc: Malte Schroder
Cc: Ralf Baechle
Cc: Herton Ronaldo Krzesinski
Cc:
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
The current ds1302 driver (or at least the one that lives in /drivers/rtc)
seems to be designed for memory mapped devices only. This make it quite
hard to add support for GPIO-based implementations (as this is the case
for the upcoming Arcom Vulcan).This patch moves the direct register access to inline functions with
explicit names. Still not as good as a proper platform driver, but at
least neater.Signed-off-by: Marc Zyngier
Cc: Paul Gortmaker
Cc: Alessandro Zummo
Cc: Paul Mundt
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Instead of individually creating and removing the sysfs device attribute
files, wrap them in an attribute_group and use
sysfs_{create/remove}_group.Signed-off-by: H Hartley Sweeten
Cc: Alessandro Zummo
Cc: Paul Gortmaker
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
On imx SoCs rtc clock parent is CKIL, but clock rate shall be determined
using rtc clock itself, that eliminates CKIL clock usage in the driver.Signed-off-by: Vladimir Zapolskiy
Cc: Alessandro Zummo
Cc: Daniel Mack
Cc: Andrew Morton
Cc: Paul Gortmaker
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Add support for the S3C64xx SoC to the generic S3C RTC driver.
Signed-off-by: Maurus Cuelenaere
Acked-by: Ben Dooks
Cc: Frans Pop
Cc: Paul Gortmaker
Cc: Alessandro Zummo
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Remove duplicated #include('s) in fs/smbfs/symlink.c
Signed-off-by: Huang Weiyi
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Fix build error when CONFIG_MAGIC_SYSRQ is not enabled:
drivers/xen/manage.c:223: error: implicit declaration of function 'handle_sysrq'
Signed-off-by: Randy Dunlap
Acked-by: Jeremy Fitzhardinge
Cc: Chris Wright
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
The MSR IA32_TEMPERATURE_TARGET contains the TjMax value in the newer
Intel processors.Signed-off-by: Huaxu Wan
Signed-off-by: Carsten Emde
Cc: Jean Delvare
Cc: Valdis Kletnieks
Cc: Henrique de Moraes Holschuh
Cc: Yong Wang
Cc: Rudolf Marek
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
The thermal sensors of Intel(R) CPUs can be detected by CPUID instruction,
indicated by CPUID.06H.EAX[0].Signed-off-by: Huaxu Wan
Signed-off-by: Carsten Emde
Reviewed-by: Valdis Kletnieks
Cc: Jean Delvare
Cc: Henrique de Moraes Holschuh
Cc: Yong Wang
Cc: Rudolf Marek
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: Paul Thomas
Cc: Jonathan Cameron
Cc: Jean Delvare
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Set valid adjustment window (0 - 2000ms).
Signed-off-by: Samu Onkalo
Acked-by: Eric Piel
Cc: Daniel Mack
Cc: Pavel Machek
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Content for the 8bit device threaded interrupt handlers. Depending on the
interrupt line and chip configuration, either click or wakeup / freefall
handler is called. In case of click, BTN_ event is sent via input device.
In case of wakeup or freefall, input device ABS_ events are updated
immediatelly.It is still possible to configure interrupt line 1 for fast freefall
detection and use the second line either for click or threshold based
interrupts. Or both lines can be used for click / threshold interrupts.Polled input device can be set to stopped state and still get coordinate
updates via input device using interrupt based method. Polled mode and
interrupt mode can also be used parallel.BTN_ events are remapped based on existing axis remapping information.
Signed-off-by: Samu Onkalo
Acked-by: Eric Piel
Cc: Daniel Mack
Cc: Pavel Machek
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Original lis3 driver didn't provide interrupt handler(s) for click or
threshold event handling. This patch adds threaded handlers for one or
two interrupt lines for 8 bit device. Actual content for interrupt
handling is provided in the separate patch.Signed-off-by: Samu Onkalo
Tested-by: Daniel Mack
Acked-by: Eric Piel
Cc: Pavel Machek
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
8 bit device has two wakeup / free fall units. It was not possible to
configure the second unit. This patch introduces configuration entry to
the platform data and also corresponding changes to the 8 bit setup
function.High pass filters were enabled by default. Patch introduces configuration
option for high pass filter cut off frequency and also possibility to
disable or enable the filter via platform data. Since the control is a
new one and default state was filter enabled, new option is used to
disable the filter. This way old platform data is still compatible with
the change.Signed-off-by: Samu Onkalo
Acked-by: Eric Piel
Tested-by: Daniel Mack
Cc: Pavel Machek
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Separate configuration function for 8 bit version of the chip. This way
generic part of the init function stays little bit more readable.Signed-off-by: Samu Onkalo
Acked-by: Eric Piel
Tested-by: Daniel Mack
Cc: Pavel Machek
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Definitions for click were missing.
Signed-off-by: Samu Onkalo
Acked-by: Eric Piel
Tested-by: Daniel Mack
Cc: Pavel Machek
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Since crc32.c contains a nifty test program that can be executed in user
space, make sure endian detection works reliably in user space too.Signed-off-by: Joakim Tjernlund
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Precompute more crc32 values(0xcc00, 0xcc0000 and 0xcc000000) into tables.
This increases the table size from 1KB to 4KB but the performance benfit
makes it worth it:28% faster on MPC8321, 266 MHz
2x faster on Core 2 Duo, 3.1GHz[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: Joakim Tjernlund
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
The C99 specification states in section 6.11.5:
The placement of a storage-class specifier other than at the beginning
of the declaration specifiers in a declaration is an obsolescent
feature.Signed-off-by: Tobias Klauser
Acked-by: Jean Delvare
Cc: Andy Whitcroft
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
I've seen various new Kconfigs with rather unhelpful one liner
descriptions. Add a Kconfig warning for a minimum length of the Kconfig
help section.Right now I arbitarily chose 4. The exact value can be debated.
[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: Andi Kleen
Cc: Andy Whitcroft
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Remove own implementation of hex_to_bin().
Signed-off-by: Andy Shevchenko
Cc: Len Brown
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Instead of using own implementation involve hex_to_bin() function.
Signed-off-by: Andy Shevchenko
Acked-by: John W. Linville
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Remove own implementation of hex_to_bin().
Signed-off-by: Andy Shevchenko
Cc: "Richard Russon (FlatCap)"
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Instead of using own implementation involve hex_to_bin() function.
Signed-off-by: Andy Shevchenko
Cc: Greg Kroah-Hartman
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Remove own implementation of hex_to_bin().
Signed-off-by: Andy Shevchenko
Cc: Eric W. Biederman
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Instead of using own implementation which potentialy has bugs involve
hex_to_bin() function. It requires to have hex_to_bin() implementation
introduced by starter patch in series.Signed-off-by: Andy Shevchenko
Cc: Duncan Sands
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Remove own implementation of hex_to_bin().
Signed-off-by: Andy Shevchenko
Acked-by: Tilman Schmidt
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
hex_to_bin() is a little method which converts hex digit to its actual
value. There are plenty of places where such functionality is needed.[akpm@linux-foundation.org: use tolower(), saving 3 bytes, test the more common case first - it's quicker]
[akpm@linux-foundation.org: relocate tolower to make it even faster! (Joe)]
Signed-off-by: Andy Shevchenko
Cc: Tilman Schmidt
Cc: Duncan Sands
Cc: Eric W. Biederman
Cc: Greg Kroah-Hartman
Cc: "Richard Russon (FlatCap)"
Cc: John W. Linville
Cc: Len Brown
Cc: Joe Perches
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds -
Reduce char linebuf[200] to the actual size required., which is 32 * 3 + 2
+ 32 + 1, ie: linebuf[131].Change examples to use bool true not int 1.
Align multiline argument indentation to open parenthesis.
Use temporary for ptr[j] so trigraph fits on single line.
Convert printk ptr from %*p, (int)(2 * sizeof(void *)) to %p as %p uses
the same calculation for size.Signed-off-by: Joe Perches
Cc: Randy Dunlap
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
