Fix handling of the Cirrus Logic GD5434 chip.  Distinguish this chip from the
GD5430.  It allows detecting memory size for both models correctly.
Signed-off-by: NKrzysztof Helt <krzysztof.h1@wp.pl>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

A memory clock (MCLK) is set at various places.  Move the setting into one
place.

Set the MCLK only for Zorro cards as the x86 cards should be initialized by
BIOS.

Improve handling of the GD5434 (SD64).

Kill one annoying debug output "virtual offset: ...".
Signed-off-by: NKrzysztof Helt <krzysztof.h1@wp.pl>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Add an accelerator constant so almost all Cirrus are recognized as
accelerators by the fbset command.
Signed-off-by: NKrzysztof Helt <krzysztof.h1@wp.pl>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Cirrus' Alpine and Picasso4 chips uses DAC clock doubling to achieve full
range of pixclock frequencies.

[akpm@linux-foundation.org: fix spelling, use usual comment layout]
Signed-off-by: NKrzysztof Helt <krzysztof.h1@wp.pl>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

The 32bpp is supported only on the latest Cirrus Logic chips.  Use the
24bpp which is supported at least since Alpine chips (GD543x).

Change 32bpp mode setting to 24bpp mode.  Change acceleration as well.
Signed-off-by: NKrzysztof Helt <krzysztof.h1@wp.pl>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Various fixes to make Cirrus GD5446 chip work.

Another Cirrus chip works with the cirrusfb.  The gd5446 seems very
similar to Alpine chips.
Signed-off-by: NKrzysztof Helt <krzysztof.h1@wp.pl>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Balance iomap and iounmap and alloc and free calls in case of error druing
device register (probing).
Signed-off-by: NKrzysztof Helt <krzysztof.h1@wp.pl>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Add hardware color expansion (imageblit) function.  It roughly doubles
scrolling speed of my Alpine card (GD5430).
Signed-off-by: NKrzysztof Helt <krzysztof.h1@wp.pl>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

- Fix color expansion for 16bpp and 32bpp modes in the
  cirrusfb_RectFill().

- Make a function with a common blitter code (cirrusfb_set_blitter).

- Add fb_sync function to allow a higher layer synchronize with the
  blitter.

- Kill one redundant blitter reset.
Signed-off-by: NKrzysztof Helt <krzysztof.h1@wp.pl>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

- The Laguna GD5465 (AGP) has one register more than non-AGP chips.
  Recognize the AGP version and write a tile control register only on the
  AGP version.  Tested only on an AGP card.

- Bump up RAMDAC frequencies after X11 code.  This allow to drive a flat
  panel resolution 1680x1050 at 16bpp from the 4MB card.

- Fix screen start address overflow bits on Laguna cards (CRT1D
  register).

- Fix exit path in the cirrusfb_pci_register() in case of error.
Signed-off-by: NKrzysztof Helt <krzysztof.h1@wp.pl>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Break cirrusfb_decode_var() function into two parts:
cirrusfb_check_pixclock() which can be called from the
cirrusfb_check_var() aand merge rest into the cirrusfb_set_par_foo().
This allows rejecting modes with too high pixclock before before any
change to hardware state (and a console is messed up).

Also, fix RGB field's lengths for 8bpp modes to correct ones so X11 works
with fbdev driver with cirrusfb.

Kill some redundant function calls or register loads.
Signed-off-by: NKrzysztof Helt <krzysztof.h1@wp.pl>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Fix 8bpp mode by adding handling of the Laguna chipsets to various places
and stop trashing a HDR register which probably does not exist on the
Laguna.

Fix compilation warnings about uninitialized variables also.

Finally, all 8bpp, 16bpp and 32bpp modes work on the Laguna chipset.
Signed-off-by: NKrzysztof Helt <krzysztof.h1@wp.pl>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Various improvements to the code:
- kill a structure with only one
  field: multiplexing and use the
  field directly
- move the cirrusfb_ops structure
  down the file to kill forward
  declarations
- move cirrusfb_init() to kill
  forward declaration
- kill register loads done already
  in the init_vgachip()
- kill assigments done by higher
  layer in the cirrusfb_pan_display()
- do not overwrite line pitch bit in
  the CL_CRT1D register
- kill btype variables if they were
  used only once or twice
- add cpu_relax() in the busy waiting
  loop

The fix to the CL_CRT1D register handling makess the 1024x768 32bpp mode
work.  Previously, only lower resolution modes have worked with 32bpp.
Signed-off-by: NKrzysztof Helt <krzysztof.h1@poczta.fm>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Use the 5-6-5 RGB mode instead of the 5-5-5 mode at 16bpp depth.

It fixes colors in the 16bpp modes on Cirrus Laguna chips.
Signed-off-by: NKrzysztof Helt <krzysztof.h1@wp.pl>
Cc: Geert Uytterhoeven <geert.uytterhoeven@gmail.com>
Cc: Arthur Marsh <arthur.marsh@internode.on.net>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

A line length is calculated twice: first in the cirrusfb_decode_var() then
in the cirrusfb_set_par_foo().

Use the first calculated value.  A nice side effect is that 32bpp mode
works now.
Signed-off-by: NKrzysztof Helt <krzysztof.h1@wp.pl>
Cc: Geert Uytterhoeven <geert.uytterhoeven@gmail.com>
Cc: Arthur Marsh <arthur.marsh@internode.on.net>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

The Laguna chipsets use special registers which are available through the
mmio area.  The cirrusfb driver does not use memory mapped registers for
the PCI cards.

Add the memory mapped area for Laguna chipsets and add basic usage of the
special Laguna registers after SVGALIB code.

This gives readable console at 16bpp on the GD-5465 (Laguna AGP).  The
8bpp and 32bpp depths are still broken.
Signed-off-by: NKrzysztof Helt <krzysztof.h1@wp.pl>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Add additional overflow register setting for Laguna chips.

Also, simplify some code in the cirrusfb_pan_display() and
cirrusfb_blank().
Signed-off-by: NKrzysztof Helt <krzysztof.h1@wp.pl>
Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Fix memory detection and clock setting for Cirrus Laguna chipsets
(GD5464/GD5465).  The changes are done after the Xorg code.

The driver still does not display anything on the GD5465 but it switches
resolutions correctly at least.
Signed-off-by: NKrzysztof Helt <krzysztof.h1@wp.pl>
Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Convert all printks to dev_dbg, dev_info or dev_err.  Kill some excessive
debug information and code in the process.

[akpm@linux-foundation.org: printk fixes]
[akpm@linux-foundation.org: cleanups]
Signed-off-by: NKrzysztof Helt <krzysztof.h1@poczta.fm>
Cc: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

GFP_KERNEL is legal here - we don't need to use gfp_any().

Cc: Evgeniy Polyakov <johnpol@2ka.mipt.ru>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Andi Kleen <andi@firstfloor.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Change the page_mkwrite prototype to take a struct vm_fault, and return
VM_FAULT_xxx flags.  There should be no functional change.

This makes it possible to return much more detailed error information to
the VM (and also can provide more information eg.  virtual_address to the
driver, which might be important in some special cases).

This is required for a subsequent fix.  And will also make it easier to
merge page_mkwrite() with fault() in future.
Signed-off-by: NNick Piggin <npiggin@suse.de>
Cc: Chris Mason <chris.mason@oracle.com>
Cc: Trond Myklebust <trond.myklebust@fys.uio.no>
Cc: Miklos Szeredi <miklos@szeredi.hu>
Cc: Steven Whitehouse <swhiteho@redhat.com>
Cc: Mark Fasheh <mfasheh@suse.com>
Cc: Joel Becker <joel.becker@oracle.com>
Cc: Artem Bityutskiy <dedekind@infradead.org>
Cc: Felix Blyakher <felixb@sgi.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Setting ->owner as done currently (pde->owner = THIS_MODULE) is racy
as correctly noted at bug #12454. Someone can lookup entry with NULL
->owner, thus not pinning enything, and release it later resulting
in module refcount underflow.

We can keep ->owner and supply it at registration time like ->proc_fops
and ->data.

But this leaves ->owner as easy-manipulative field (just one C assignment)
and somebody will forget to unpin previous/pin current module when
switching ->owner. ->proc_fops is declared as "const" which should give
some thoughts.

->read_proc/->write_proc were just fixed to not require ->owner for
protection.

rmmod'ed directories will be empty and return "." and ".." -- no harm.
And directories with tricky enough readdir and lookup shouldn't be modular.
We definitely don't want such modular code.

Removing ->owner will also make PDE smaller.

So, let's nuke it.

Kudos to Jeff Layton for reminding about this, let's say, oversight.

http://bugzilla.kernel.org/show_bug.cgi?id=12454Signed-off-by: NAlexey Dobriyan <adobriyan@gmail.com>

Use __pci_complete_power_transition() to finalize the transition into
D2 after programming the PMCSR of the device directly.
Signed-off-by: NRafael J. Wysocki <rjw@sisk.pl>
Acked-by: NJesse Barnes <jbarnes@virtuousgeek.org>

We read from the PCR reg to determine whether to use TFT mode or not.
This is not possible because it may not have been initialized with
the correct value yet. Select it using fbi->pcr instead.
Signed-off-by: NSascha Hauer <s.hauer@pengutronix.de>

v2: Added change from Martin Fuzzey: pixclock should be in pico seconds
    instead of MHz.
Signed-off-by: NSascha Hauer <s.hauer@pengutronix.de>

Signed-off-by: NRussell King <rmk+kernel@arm.linux.org.uk>

This patch fixes the bug reported in
	http://bugzilla.kernel.org/show_bug.cgi?id=11681.

"Lots of device drivers register a 'struct device_driver' with
the '.bus' member set to '&platform_bus_type'. This is wrong,
since the platform_bus functions expect the 'struct device_driver'
to be wrapped up in a 'struct platform_driver' which provides
some additional callbacks (like suspend_late, resume_early).
The effect may be that platform_suspend_late() uses bogus data
outside the device_driver struct as a pointer pointer to the
device driver's suspend_late() function or other hard to
reproduce failures."(Lothar Wassmann)
Signed-off-by: NMing Lei <tom.leiming@gmail.com>
Acked-by: NHenrique de Moraes Holschuh <hmh@hmh.eng.br>
Acked-by: NDavid Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

Acked-by: NGreg Kroah-Hartman <gregkh@suse.de>
Signed-off-by: NKay Sievers <kay.sievers@vrfy.org>

This fixes a regression introduced when we switched to using the core
pci_set_power_state().  The chip seems to need the state to be written
over and over again until it sticks, so we do that.

Note that the code is a bit blunt, without timeout, etc...  but that's
pretty much because I put back in there the code exactly as it used to
be before the regression.  I still add a call to pci_set_power_state()
at the end so that ACPI gets called appropriately on x86.
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>
Tested-by: NRaymond Wooninck <tittiatcoke@gmail.com>
Acked-by: NRafael J. Wysocki <rjw@sisk.pl>
Cc: Jesse Barnes <jbarnes@virtuousgeek.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

pxa25x doesn't support overlay in its LCD controller, this patch adds
pxafb_overlay_supported() functions to check the initialization is
necessary.
Signed-off-by: NEric Miao <eric.miao@marvell.com>

With arch/sh/drivers/dma/ always being built, the Dreamcast DMA engines
are being unconditionally built in, regardless of whether the DMA API is
enabled or not. This is a regression from previous behaviour, but there
is not much advantage in building them all in unconditionally regardless.
Add a new config option to make it optional, and update the only user of
it to reflect that.
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

Impact: help prevent extinction of species

The Tasmanian Devil is a shy iconic Australian creature named for its
spine-chilling screech.  It is threatened with extinction due to a
scientifically interesting but horrific transmissible facial cancer.

This one is standing in for Tux for one release using the far less-known
Devil Facial Tux Disguise.

	Save The Tasmanian Devil http://tassiedevil.com.auSigned-off-by: NLinux.conf.au Hobart Team <contact@marchsouth.org>
Signed-off-by: NRusty Russell <rusty@rustcorp.com.au>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

This patch adds suspend/resume support to the LCDC
driver for SuperH Mobile - sh_mobile_lcdcfb.

We simply stop hardware on suspend and start it again
on resume. For RGB panels this is trivial, but for SYS
panels in deferred io mode this becomes a bit more
difficult - we need to wait for a frame end interrupt
to make sure the clocks are balanced before stopping
the actual hardware.
Signed-off-by: NMagnus Damm <damm@igel.co.jp>
Signed-off-by: NPaul Mundt <lethal@linux-sh.org>

This patch adds clkdev support for i.MX31. This is done in a
similar way done previously for i.MX27
Signed-off-by: NSascha Hauer <s.hauer@pengutronix.de>

The var->hsync_len, var->right_margin and var->left_margin fields
should contain the real values, not the hardware dependent values.
Signed-off-by: NSascha Hauer <s.hauer@pengutronix.de>

Signed-off-by: NSascha Hauer <s.hauer@pengutronix.de>

v2: As pointed out by Hans J. Koch we have to claim we can do 24bit to
    make software like X work. We are lucky on i.MX that 18bit support
    has the necessary gaps in the fields to do so.
Signed-off-by: NSascha Hauer <s.hauer@pengutronix.de>

Signed-off-by: NSascha Hauer <s.hauer@pengutronix.de>

This patch mimicks what Martin wrote on the mailing list:

* move arch/arm/mach-imx/include/mach/imxfb.h into
  arch/arm/mach-mxc/include/mach/imxfb.h
* changes Kconfig so that CONFIG_FB_IMX is selectable
* adds a platform device (copied from some pengutronix
  patches)
Signed-off-by: NHolger Schurig <hs4233@mail.mn-solutions.de>
Signed-off-by: NSascha Hauer <s.hauer@pengutronix.de>

radeonfb and aty128fb have a special hook called by the PowerMac platform
code very very early on resume from sleep to bring the screen back. This
is useful for debugging wakup problems, but unfortunately, this also became
a source of problems of its own.

The hook is called extremely early, with interrupts still off, and the code
path involved with that code nowadays rely on things like taking mutexes,
GFP_KERNEL allocations, etc...

In addition, the driver now relies on the PCI core to restore the standard
config space before calling resume which doesn't happen with this early
code path.

I'm keeping the code in but commented out along with a fixup call to
pci_restore_state(). The reason is that I still want to make it easy to
re-enable temporarily to track wake up problems, and it's possible that
I can revive it at some stage if we make sleeping things save to call
in early resume using a system state.

In the meantime, this should fix several reported regressions.
Signed-off-by: NBenjamin Herrenschmidt <benh@kernel.crashing.org>