linux-fbdev-devel — linux-fbdev mailing list for project developers

Re: [Linux-fbdev-devel] Framebuffer support for HTC Dream

[Pavel M. <pa...@uc...>]

> > diff --git a/drivers/video/Kconfig b/drivers/video/Kconfig
> > index fb19803..4a7873b 100644
> > --- a/drivers/video/Kconfig
> > +++ b/drivers/video/Kconfig
> > @@ -2119,6 +2119,14 @@ config FB_PRE_INIT_FB
> >  	  Select this option if display contents should be inherited as set by
> >  	  the bootloader.
> >  
> > +config FB_MSM
> > +	tristate
> > +	depends on FB && ARCH_MSM
> > +	select FB_CFB_FILLRECT
> > +	select FB_CFB_COPYAREA
> > +	select FB_CFB_IMAGEBLIT
> > +	default y
> 
> I'm seeing gpio calls in there.  Do we need a gpio dependency, or did
> ARCH_MSM give us gpio anyway?

AFAICT, ARCH_MSM needs GENERIC_GPIO, so we are ok.

Thanks,
									Pavel

-- 
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html

Re: [Linux-fbdev-devel] [PATCH 2/3] drivers/video/imxfb.c: Fix resource size off by 1 error

[Sascha H. <s....@pe...>]

Hi Julia,

On Sat, Jul 04, 2009 at 11:32:29PM +0200, Julia Lawall wrote:
> From: Julia Lawall <ju...@di...>
> 
> In the first case, the change is not semantics-preserving, but
> resource_size is used at the corresponding call to request_mem_region in
> the same function.
> 
> In the second case, the change is semantics-preserving.
> 
> The problem was found using the following semantic patch:
> (http://www.emn.fr/x-info/coccinelle/)
> 
> // <smpl>
> @@
> struct resource *res;
> @@
> 
> - (res->end - res->start) + 1
> + resource_size(res)
> 
> @@
> struct resource *res;
> @@
> 
> - res->end - res->start
> + BAD(resource_size(res))
> // </smpl>
> 
> Signed-off-by: Julia Lawall <ju...@di...>
> 
> ---
>  drivers/video/imxfb.c |    4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff -u -p a/drivers/video/imxfb.c b/drivers/video/imxfb.c
> --- a/drivers/video/imxfb.c 2009-04-13 16:04:25.000000000 +0200
> +++ b/drivers/video/imxfb.c 2009-07-04 21:37:48.000000000 +0200
> @@ -754,7 +754,7 @@ failed_map:
>  failed_getclock:
>  	iounmap(fbi->regs);
>  failed_ioremap:
> -	release_mem_region(res->start, res->end - res->start);
> +	release_mem_region(res->start, resource_size(res));
>  failed_req:
>  	kfree(info->pseudo_palette);
>  failed_init:
> @@ -785,7 +785,7 @@ static int __devexit imxfb_remove(struct
>  	framebuffer_release(info);
>  
>  	iounmap(fbi->regs);
> -	release_mem_region(res->start, res->end - res->start + 1);
> +	release_mem_region(res->start, resource_size(res));
>  	clk_disable(fbi->clk);
>  	clk_put(fbi->clk);

I already have a patch queued for this issue in imxfb.c

Sascha

-- 
Pengutronix e.K.                           |                             |
Industrial Linux Solutions                 | http://www.pengutronix.de/  |
Peiner Str. 6-8, 31137 Hildesheim, Germany | Phone: +49-5121-206917-0    |
Amtsgericht Hildesheim, HRA 2686           | Fax:   +49-5121-206917-5555 |

Re: [Linux-fbdev-devel] [PATCH v5] Frame Buffer driver for TI DA8xx/OMAP-L1xx

[<krz...@po...>]

Rajashekhara, Sudhakar napisał(a):
> On Wed, Jul 08, 2009 at 01:39:43, Krzysztof Helt wrote:
> > On Mon,  6 Jul 2009 01:29:11 -0400
> > "Rajashekhara, Sudhakar" >sud...@ti...> wrote:
> > 
> > > Adds LCD controller (LCDC) driver for TI's DA8xx/OMAP-L1xx
> architecture.
> > > LCDC specifications can be found at
> http://www.ti.com/litv/pdf/sprufm0a.
> > > 
> > > LCDC on DA8xx consists of two independent controllers, the Raster 
> > > Controller and the LCD Interface Display Driver (LIDD) controller. 
> > > LIDD further supports character and graphic displays.
> > > 
> > > This patch adds support for the graphic display (Sharp LQ035Q3DG01) 
> > > found on the DA830 based EVM. The EVM details can be found at:
> > > http://support.spectrumdigital.com/boards/dskda830/revc/.
> > > 
> > > Signed-off-by: Sudhakar Rajashekhara >sud...@ti...>
> > > Signed-off-by: Pavel Kiryukhin >pki...@ru...>
> > > Signed-off-by: Steve Chen >sc...@mv...>
> > > ---
> > >  This patch applies to Linus's Kernel tree.
> > > 
> > 
> > Acked-by: Krzysztof Helt >krz...@wp...>
> > 
> 
> Thanks. What I should do to get this patch to Linus's tree?
> 

Send it to Andrew Morton <ak...@li...>

Regards,
Krzysztof

----------------------------------------------------------------------
Twoja pensja x 3? Otworz Konto Direct!
Sprawdz >>  	 http://link.interia.pl/f2209

Re: [Linux-fbdev-devel] [PATCH v5] Frame Buffer driver for TI DA8xx/OMAP-L1xx

[Rajashekhara, S. <sud...@ti...>]

On Wed, Jul 08, 2009 at 01:39:43, Krzysztof Helt wrote:
> On Mon,  6 Jul 2009 01:29:11 -0400
> "Rajashekhara, Sudhakar" <sud...@ti...> wrote:
> 
> > Adds LCD controller (LCDC) driver for TI's DA8xx/OMAP-L1xx architecture.
> > LCDC specifications can be found at http://www.ti.com/litv/pdf/sprufm0a.
> > 
> > LCDC on DA8xx consists of two independent controllers, the Raster 
> > Controller and the LCD Interface Display Driver (LIDD) controller. 
> > LIDD further supports character and graphic displays.
> > 
> > This patch adds support for the graphic display (Sharp LQ035Q3DG01) 
> > found on the DA830 based EVM. The EVM details can be found at:
> > http://support.spectrumdigital.com/boards/dskda830/revc/.
> > 
> > Signed-off-by: Sudhakar Rajashekhara <sud...@ti...>
> > Signed-off-by: Pavel Kiryukhin <pki...@ru...>
> > Signed-off-by: Steve Chen <sc...@mv...>
> > ---
> >  This patch applies to Linus's Kernel tree.
> > 
> 
> Acked-by: Krzysztof Helt <krz...@wp...>
> 

Thanks. What I should do to get this patch to Linus's tree?

Regards,
Sudhakar

Re: [Linux-fbdev-devel] Framebuffer support for HTC Dream

[Pavel M. <pa...@uc...>]

Hi!

> >> > Will it have a maintainer?
> >>
> >> I did the original quick and dirty version for bringup.  Rebecca took
> >> over and (re)wrote the bulk of the driver, getting things stable for
> >> production ship of Dream and Sapphire, and Dima is currently adding
> >> support for later Qualcomm chipsets (QSD8x50, etc).  Looping them into
> >> this discussion.
> >
> > I can maintain it if noone else steps up. Of course, someone who
> > actually knows the hw/has the docs would be preferable.
> 
> As Brian mentioned, I'm the last one to touch it here, so I now own
> it. I've done a bunch refactoring to add support for the qsd8x50

Can you submit a patch adding yourself into MAINTAINERS file?

> devices. I'll post those up onto our experimental git tree (hopefully
> sometime today), and once this patch goes in, I'll make sure that we
> apply cleanly against it. Alternatively, if you like, I can
> incorporate your comments into the refactoring work I'm doing, and can
> push out the combined patch. In the future though, I'll try to be a
> lot more aggressive at sending these out for review earlier so you
> don't have to clean up our mess.

I believe it makes sense to submit patches relative to -mm tree now.


> Please feel free to CC Rebecca and myself on all the msm_fb related
> changes/discussions.
> 
> >> It would probably be worth referring to drivers like this as "for
> >> Qualcomm MSM/QSD SoCs" or the like, since they do work with a lot of
> >> hardware beyond just Dream.
> >
> > Well, the Dream is why I'm doing it, hence patch title. Of course,
> > Kconfig should say 'Qualcomm MSM/QSD SoCs', too.
> 
> I second the request that the patches for qualcomm msm7k/qsd8k generic
> support would have descriptions that refer to them as such, since the
> htc dream specific code is only in the board files in
> arch/arm/mach-msm. Note that the HTC Magic (aka the Ion, myTouch 3G,

Ok, I guess someone should do a picture of all the connections and
dictionary of all the acronyms :-).
								Pavel
-- 
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html

Re: [Linux-fbdev-devel] matroxfb: fix regression with uninitalized fb_info->mm_lock mutex (second head)

[<krz...@po...>]

Linus Torvalds napisał(a):
> 
> 
> On Tue, 7 Jul 2009, Krzysztof Helt wrote:
> > 
> > Remove redundant locking by the mm_lock mutex before a second head of
> matrox 
> > framebuffer is registered.
> 
> Why do you write misleading commentary like this.
> 

I have not intention to mislead someone. English is not my primary language and this can be a source of this unfortunate wording.

> > +/*
> > + * This function is called before the register_framebuffer so
> > + * no locking is needed.
> > + */
> 
> Or this?
> 
> It's not about "needed". The locking is not only not needed, it would be 
> BUGGY.
> 
> And it's not "redundant". That implies that it's done somewhere else. It's
> 
> more than "not needed" - it would be actively buggy to lock things there.
> 
> I really don't like how you're approaching this. You're ignoring the real
> 
> issues I ask you, you're writing misleading comments and commit messages,
> 
> and the end result is fragile code. I still don't understand why you 
> insist on initializing those things late, which is the primary problem 
> here.
> 

The mm_lock mutex is used only inside the fb_mmap() function and driver's specific code. The fb_mmap() 
cannot be called before a framebuffer is registered. That's why it is enough to initialize the mutex
just before registering the framebuffer (or inside the register_framebuffer()). A current approach 
allows for statically declared fb_info structures (you consider it wrong).

It is easy to find affected drivers: "grep -rl mm_lock drivers/video/*". I went too far with some
drivers and added mm_lock mutex to functions which are called only from a probing function. They do
not need any locking yet. This is similar as marking them as __init or __devinit which makes incorrect
calling them after driver initialization. Some of these functions are so short they can be unrolled
into the probing functions. Does it make removal of the unneeded mutexing correct then?

All other drivers just have, let's say, inefficient code - calling a function which is called just
a moment later or set values which are not used before set again later (i810fb, matroxfb, sisfb).
The mm_lock patch converted this inefficient code into broken code. 

I understand you choose different approach. In order to keep this inefficient code in place you 
decided to change some correctly working legacy drivers and disallow statically declared fb_info 
structures. I see I cannot convince you it is worse solution so I am not trying any more.

However, you can still count my patches as small code improvements.

My fault is that I haven't tested the patch on a hardware and driver with added mm_lock mutex and
broken by this (like sisfb or matroxfb). I have not known I broke them.

Regards,
Krzysztof

----------------------------------------------------------------------
Teraz AC/OC + ubezpieczenie kosztow leczenia za granica
http://link.interia.pl/f2230

Re: [Linux-fbdev-devel] [PATCH] au1200fb: fix oops due to uninitialized fb_info.lock mutex

[Manuel L. <man...@go...>]

Hi Krzysztof,

On Tue, Jul 7, 2009 at 5:44 PM, Krzysztof Helt<krz...@po...> wrote:
> On Tue, 07 Jul 2009 10:29:51 +0200
> Manuel Lauss <man...@go...> wrote:
>
>> commit 4148df9b0f38bdd362dd91d52076926c11cbe5a9 breaks au1200fb due to
>> fb_info.lock now being uninitialized at the time of the
>> register_frambuffer call:
>>
>> CPU 0 Unable to handle kernel paging request at virtual address 00000000, epc == 8038ccd4, ra == 802412d4
>> Call Trace:
>> [<8038ccd4>] __mutex_lock_slowpath+0x3c/0xe4
>> [<802412d4>] lock_fb_info+0x20/0x50
>> [<80242158>] register_framebuffer+0x264/0x2a4
>> [<8024f8c4>] au1200fb_drv_probe+0x308/0x444
>> [<80273ef8>] driver_probe_device+0xbc/0x184
>> [<8027402c>] __driver_attach+0x6c/0xa4
>> [<802736bc>] bus_for_each_dev+0x60/0xb0
>> [<80272e20>] bus_add_driver+0xc8/0x24c
>> [<802743f4>] driver_register+0xdc/0x190
>> [<8047d130>] au1200fb_init+0x39c/0x3cc
>> [<80100460>] _stext+0x60/0x1c8
>> [<8046f334>] kernel_init+0xc8/0x134
>> [<80106f64>] kernel_thread_helper+0x10/0x18
>>
>> This band-aid fixes this oops.
>>
>
> Please wait a little. Currently, I, Paul Mundt and Linus discuss how to fix all drivers broken by the commit 4148df9b0f38bdd362dd91d52076.

Alright, I'll wait. In the meantime I've refined the patch a bit and
converted the driver to use the framebuffer_alloc() call.

      Manuel Lauss

Re: [Linux-fbdev-devel] matroxfb: fix regression with uninitalized fb_info->mm_lock mutex (second head)

[Linus T. <tor...@li...>]

On Tue, 7 Jul 2009, Krzysztof Helt wrote:
> 
> Remove redundant locking by the mm_lock mutex before a second head of matrox 
> framebuffer is registered.

Why do you write misleading commentary like this.

> +/*
> + * This function is called before the register_framebuffer so
> + * no locking is needed.
> + */

Or this?

It's not about "needed". The locking is not only not needed, it would be 
BUGGY.

And it's not "redundant". That implies that it's done somewhere else. It's 
more than "not needed" - it would be actively buggy to lock things there.

I really don't like how you're approaching this. You're ignoring the real 
issues I ask you, you're writing misleading comments and commit messages, 
and the end result is fragile code. I still don't understand why you 
insist on initializing those things late, which is the primary problem 
here.

		Linus

[Linux-fbdev-devel] matroxfb: fix regression with uninitalized fb_info->mm_lock mutex (second head)

[Krzysztof H. <krz...@po...>]

From: Krzysztof Helt <krz...@wp...>

Remove redundant locking by the mm_lock mutex before a second head of matrox 
framebuffer is registered.

This fixes a problem with uninitialized the fb_info->mm_lock mutex introduced 
by the commit 537a1bf059f " fbdev: add mutex for fb_mmap locking"

Signed-off-by: Krzysztof Helt <krz...@wp...>

---

diff --git a/drivers/video/matrox/matroxfb_crtc2.c b/drivers/video/matrox/matroxfb_crtc2.c
index 909e10a..6caa369 100644
--- a/drivers/video/matrox/matroxfb_crtc2.c
+++ b/drivers/video/matrox/matroxfb_crtc2.c
@@ -289,16 +289,18 @@ static int matroxfb_dh_release(struct fb_info* info, int user) {
 #undef m2info
 }
 
+/*
+ * This function is called before the register_framebuffer so
+ * no locking is needed.
+ */
 static void matroxfb_dh_init_fix(struct matroxfb_dh_fb_info *m2info)
 {
 	struct fb_fix_screeninfo *fix = &m2info->fbcon.fix;
 
 	strcpy(fix->id, "MATROX DH");
 
-	mutex_lock(&m2info->fbcon.mm_lock);
 	fix->smem_start = m2info->video.base;
 	fix->smem_len = m2info->video.len_usable;
-	mutex_unlock(&m2info->fbcon.mm_lock);
 	fix->ypanstep = 1;
 	fix->ywrapstep = 0;
 	fix->xpanstep = 8;	/* TBD */

----------------------------------------------------------------------
Sprawdz promocje ubezpieczen komunikacyjnych w Ergo  Hestia
http://link.interia.pl/f222c

Re: [Linux-fbdev-devel] [PATCH v5] Frame Buffer driver for TI DA8xx/OMAP-L1xx

[Krzysztof H. <krz...@po...>]

On Mon,  6 Jul 2009 01:29:11 -0400
"Rajashekhara, Sudhakar" <sud...@ti...> wrote:

> Adds LCD controller (LCDC) driver for TI's DA8xx/OMAP-L1xx architecture.
> LCDC specifications can be found at http://www.ti.com/litv/pdf/sprufm0a.
> 
> LCDC on DA8xx consists of two independent controllers, the Raster Controller
> and the LCD Interface Display Driver (LIDD) controller. LIDD further supports
> character and graphic displays.
> 
> This patch adds support for the graphic display (Sharp LQ035Q3DG01) found on
> the DA830 based EVM. The EVM details can be found at:
> http://support.spectrumdigital.com/boards/dskda830/revc/.
> 
> Signed-off-by: Sudhakar Rajashekhara <sud...@ti...>
> Signed-off-by: Pavel Kiryukhin <pki...@ru...>
> Signed-off-by: Steve Chen <sc...@mv...>
> ---
>  This patch applies to Linus's Kernel tree.
> 

Acked-by: Krzysztof Helt <krz...@wp...>

----------------------------------------------------------------------
Zaufalo nam ponad 700 000 Polakow. Dolacz do AXA OFE.
http://link.interia.pl/f2257

Re: [Linux-fbdev-devel] [PATCH v5] Frame Buffer driver for TI DA8xx/OMAP-L1xx

[Krzysztof H. <krz...@po...>]

On Mon,  6 Jul 2009 01:29:11 -0400
"Rajashekhara, Sudhakar" <sud...@ti...> wrote:

> Adds LCD controller (LCDC) driver for TI's DA8xx/OMAP-L1xx architecture.
> LCDC specifications can be found at http://www.ti.com/litv/pdf/sprufm0a.
> 
> LCDC on DA8xx consists of two independent controllers, the Raster Controller
> and the LCD Interface Display Driver (LIDD) controller. LIDD further supports
> character and graphic displays.
> 
> This patch adds support for the graphic display (Sharp LQ035Q3DG01) found on
> the DA830 based EVM. The EVM details can be found at:
> http://support.spectrumdigital.com/boards/dskda830/revc/.
> 
> Signed-off-by: Sudhakar Rajashekhara <sud...@ti...>
> Signed-off-by: Pavel Kiryukhin <pki...@ru...>
> Signed-off-by: Steve Chen <sc...@mv...>
> ---
>  This patch applies to Linus's Kernel tree.
> 

Acked-by: Krzysztof Helt <krz...@wp...>

>  Since the previous version, fb_setcolreg function has been modified for
>  8 bit mode.
> 
>  drivers/video/Kconfig    |   11 +
>  drivers/video/Makefile   |    1 +
>  drivers/video/da8xx-fb.c |  900 ++++++++++++++++++++++++++++++++++++++++++++++
>  include/video/da8xx-fb.h |  106 ++++++
>  4 files changed, 1018 insertions(+), 0 deletions(-)
>  create mode 100644 drivers/video/da8xx-fb.c
>  create mode 100644 include/video/da8xx-fb.h
> 
> diff --git a/drivers/video/Kconfig b/drivers/video/Kconfig
> index 8afcf08..d048b7e 100644
> --- a/drivers/video/Kconfig
> +++ b/drivers/video/Kconfig
> @@ -2038,6 +2038,17 @@ config FB_SH7760
>  	  and 8, 15 or 16 bpp color; 90 degrees clockwise display rotation for
>  	  panels <= 320 pixel horizontal resolution.
>  
> +config FB_DA8XX
> +        tristate "DA8xx/OMAP-L1xx Framebuffer support"
> +        depends on FB && ARCH_DAVINCI_DA830
> +	select FB_CFB_FILLRECT
> +	select FB_CFB_COPYAREA
> +	select FB_CFB_IMAGEBLIT
> +	---help---
> +          This is the frame buffer device driver for the TI LCD controller
> +	  found on DA8xx/OMAP-L1xx SoCs.
> +          If unsure, say N.
> +
>  config FB_VIRTUAL
>  	tristate "Virtual Frame Buffer support (ONLY FOR TESTING!)"
>  	depends on FB
> diff --git a/drivers/video/Makefile b/drivers/video/Makefile
> index 01a819f..288d9b0 100644
> --- a/drivers/video/Makefile
> +++ b/drivers/video/Makefile
> @@ -136,6 +136,7 @@ obj-$(CONFIG_FB_OF)               += offb.o
>  obj-$(CONFIG_FB_BF54X_LQ043)	  += bf54x-lq043fb.o
>  obj-$(CONFIG_FB_BFIN_T350MCQB)	  += bfin-t350mcqb-fb.o
>  obj-$(CONFIG_FB_MX3)		  += mx3fb.o
> +obj-$(CONFIG_FB_DA8XX)		  += da8xx-fb.o
>  
>  # the test framebuffer is last
>  obj-$(CONFIG_FB_VIRTUAL)          += vfb.o
> diff --git a/drivers/video/da8xx-fb.c b/drivers/video/da8xx-fb.c
> new file mode 100644
> index 0000000..4b9d80b
> --- /dev/null
> +++ b/drivers/video/da8xx-fb.c
> @@ -0,0 +1,900 @@
> +/*
> + * Copyright (C) 2008-2009 MontaVista Software Inc.
> + * Copyright (C) 2008-2009 Texas Instruments Inc
> + *
> + * Based on the LCD driver for TI Avalanche processors written by
> + * Ajay Singh and Shalom Hai.
> + *
> + * This program is free software; you can redistribute it and/or modify
> + * it under the terms of the GNU General Public License as published by
> + * the Free Software Foundation; either version 2 of the License, or
> + * (at your option)any later version.
> + *
> + * This program is distributed in the hope that it will be useful,
> + * but WITHOUT ANY WARRANTY; without even the implied warranty of
> + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
> + * GNU General Public License for more details.
> + *
> + * You should have received a copy of the GNU General Public License
> + * along with this program; if not, write to the Free Software
> + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
> + */
> +#include <linux/module.h>
> +#include <linux/kernel.h>
> +#include <linux/fb.h>
> +#include <linux/dma-mapping.h>
> +#include <linux/device.h>
> +#include <linux/platform_device.h>
> +#include <linux/uaccess.h>
> +#include <linux/device.h>
> +#include <linux/interrupt.h>
> +#include <linux/clk.h>
> +#include <video/da8xx-fb.h>
> +
> +#define DRIVER_NAME "da8xx_lcdc"
> +
> +/* LCD Status Register */
> +#define LCD_END_OF_FRAME0		BIT(8)
> +#define LCD_FIFO_UNDERFLOW		BIT(5)
> +#define LCD_SYNC_LOST			BIT(2)
> +
> +/* LCD DMA Control Register */
> +#define LCD_DMA_BURST_SIZE(x)		((x) << 4)
> +#define LCD_DMA_BURST_1			0x0
> +#define LCD_DMA_BURST_2			0x1
> +#define LCD_DMA_BURST_4			0x2
> +#define LCD_DMA_BURST_8			0x3
> +#define LCD_DMA_BURST_16		0x4
> +#define LCD_END_OF_FRAME_INT_ENA	BIT(2)
> +#define LCD_DUAL_FRAME_BUFFER_ENABLE	BIT(0)
> +
> +/* LCD Control Register */
> +#define LCD_CLK_DIVISOR(x)		((x) << 8)
> +#define LCD_RASTER_MODE			0x01
> +
> +/* LCD Raster Control Register */
> +#define LCD_PALETTE_LOAD_MODE(x)	((x) << 20)
> +#define PALETTE_AND_DATA		0x00
> +#define PALETTE_ONLY			0x01
> +
> +#define LCD_MONO_8BIT_MODE		BIT(9)
> +#define LCD_RASTER_ORDER		BIT(8)
> +#define LCD_TFT_MODE			BIT(7)
> +#define LCD_UNDERFLOW_INT_ENA		BIT(6)
> +#define LCD_MONOCHROME_MODE		BIT(1)
> +#define LCD_RASTER_ENABLE		BIT(0)
> +#define LCD_TFT_ALT_ENABLE		BIT(23)
> +#define LCD_STN_565_ENABLE		BIT(24)
> +
> +/* LCD Raster Timing 2 Register */
> +#define LCD_AC_BIAS_TRANSITIONS_PER_INT(x)	((x) << 16)
> +#define LCD_AC_BIAS_FREQUENCY(x)		((x) << 8)
> +#define LCD_SYNC_CTRL				BIT(25)
> +#define LCD_SYNC_EDGE				BIT(24)
> +#define LCD_INVERT_PIXEL_CLOCK			BIT(22)
> +#define LCD_INVERT_LINE_CLOCK			BIT(21)
> +#define LCD_INVERT_FRAME_CLOCK			BIT(20)
> +
> +/* LCD Block */
> +#define  LCD_CTRL_REG				0x4
> +#define  LCD_STAT_REG				0x8
> +#define  LCD_RASTER_CTRL_REG			0x28
> +#define  LCD_RASTER_TIMING_0_REG		0x2C
> +#define  LCD_RASTER_TIMING_1_REG		0x30
> +#define  LCD_RASTER_TIMING_2_REG		0x34
> +#define  LCD_DMA_CTRL_REG			0x40
> +#define  LCD_DMA_FRM_BUF_BASE_ADDR_0_REG	0x44
> +#define  LCD_DMA_FRM_BUF_CEILING_ADDR_0_REG	0x48
> +
> +#define WSI_TIMEOUT	50
> +#define PALETTE_SIZE	256
> +#define LEFT_MARGIN	64
> +#define RIGHT_MARGIN	64
> +#define UPPER_MARGIN	32
> +#define LOWER_MARGIN	32
> +
> +static resource_size_t da8xx_fb_reg_base;
> +
> +static inline unsigned int lcdc_read(unsigned int addr)
> +{
> +	return (unsigned int)__raw_readl(da8xx_fb_reg_base + (addr));
> +}
> +
> +static inline void lcdc_write(unsigned int val, unsigned int addr)
> +{
> +	__raw_writel(val, da8xx_fb_reg_base + (addr));
> +}
> +
> +struct da8xx_fb_par {
> +	wait_queue_head_t da8xx_wq;
> +	resource_size_t p_palette_base;
> +	unsigned char *v_palette_base;
> +	struct clk *lcdc_clk;
> +	unsigned int irq;
> +	unsigned short pseudo_palette[16];
> +	unsigned int databuf_sz;
> +	unsigned int palette_sz;
> +};
> +
> +/* Variable Screen Information */
> +static struct fb_var_screeninfo da8xx_fb_var __devinitdata = {
> +	.xoffset = 0,
> +	.yoffset = 0,
> +	.transp = {0, 0, 0},
> +	.nonstd = 0,
> +	.activate = 0,
> +	.height = -1,
> +	.width = -1,
> +	.pixclock = 46666,	/* 46us - AUO display */
> +	.accel_flags = 0,
> +	.left_margin = LEFT_MARGIN,
> +	.right_margin = RIGHT_MARGIN,
> +	.upper_margin = UPPER_MARGIN,
> +	.lower_margin = LOWER_MARGIN,
> +	.sync = 0,
> +	.vmode = FB_VMODE_NONINTERLACED
> +};
> +
> +static struct fb_fix_screeninfo da8xx_fb_fix __devinitdata = {
> +	.id = "DA8xx FB Drv",
> +	.type = FB_TYPE_PACKED_PIXELS,
> +	.type_aux = 0,
> +	.visual = FB_VISUAL_PSEUDOCOLOR,
> +	.xpanstep = 1,
> +	.ypanstep = 1,
> +	.ywrapstep = 1,
> +	.accel = FB_ACCEL_NONE
> +};
> +
> +struct da8xx_panel {
> +	const char	name[25];	/* Full name <vendor>_<model> */
> +	unsigned short	width;
> +	unsigned short	height;
> +	int		hfp;		/* Horizontal front porch */
> +	int		hbp;		/* Horizontal back porch */
> +	int		hsw;		/* Horizontal Sync Pulse Width */
> +	int		vfp;		/* Vertical front porch */
> +	int		vbp;		/* Vertical back porch */
> +	int		vsw;		/* Vertical Sync Pulse Width */
> +	int		pxl_clk;	/* Pixel clock */
> +};
> +
> +static struct da8xx_panel known_lcd_panels[] = {
> +	/* Sharp LCD035Q3DG01 */
> +	[0] = {
> +		.name = "Sharp_LCD035Q3DG01",
> +		.width = 320,
> +		.height = 240,
> +		.hfp = 8,
> +		.hbp = 6,
> +		.hsw = 0,
> +		.vfp = 2,
> +		.vbp = 2,
> +		.vsw = 0,
> +		.pxl_clk = 0x10,
> +	},
> +	/* Sharp LK043T1DG01 */
> +	[1] = {
> +		.name = "Sharp_LK043T1DG01",
> +		.width = 480,
> +		.height = 272,
> +		.hfp = 2,
> +		.hbp = 2,
> +		.hsw = 41,
> +		.vfp = 2,
> +		.vbp = 2,
> +		.vsw = 10,
> +		.pxl_clk = 0x12,
> +	},
> +};
> +
> +/* Disable the Raster Engine of the LCD Controller */
> +static int lcd_disable_raster(struct da8xx_fb_par *par)
> +{
> +	int ret = 0;
> +	u32 reg;
> +
> +	reg = lcdc_read(LCD_RASTER_CTRL_REG);
> +	if (reg & LCD_RASTER_ENABLE) {
> +		lcdc_write(reg & ~LCD_RASTER_ENABLE, LCD_RASTER_CTRL_REG);
> +		ret = wait_event_interruptible_timeout(par->da8xx_wq,
> +						!lcdc_read(LCD_STAT_REG) &
> +						LCD_END_OF_FRAME0, WSI_TIMEOUT);
> +	}
> +
> +	if (ret < 0)
> +		return ret;
> +	if (ret == 0)
> +		return -ETIMEDOUT;
> +
> +	return 0;
> +}
> +
> +static void lcd_blit(int load_mode, struct da8xx_fb_par *par)
> +{
> +	u32 tmp = par->p_palette_base + par->databuf_sz - 4;
> +	u32 reg;
> +
> +	/* Update the databuf in the hw. */
> +	lcdc_write(par->p_palette_base, LCD_DMA_FRM_BUF_BASE_ADDR_0_REG);
> +	lcdc_write(tmp, LCD_DMA_FRM_BUF_CEILING_ADDR_0_REG);
> +
> +	/* Start the DMA. */
> +	reg = lcdc_read(LCD_RASTER_CTRL_REG);
> +	reg &= ~(3 << 20);
> +	if (load_mode == LOAD_DATA)
> +		reg |= LCD_PALETTE_LOAD_MODE(PALETTE_AND_DATA);
> +	else if (load_mode == LOAD_PALETTE)
> +		reg |= LCD_PALETTE_LOAD_MODE(PALETTE_ONLY);
> +
> +	lcdc_write(reg, LCD_RASTER_CTRL_REG);
> +}
> +
> +/* Configure the Burst Size of DMA */
> +static int lcd_cfg_dma(int burst_size)
> +{
> +	u32 reg;
> +
> +	reg = lcdc_read(LCD_DMA_CTRL_REG) & 0x00000001;
> +	switch (burst_size) {
> +	case 1:
> +		reg |= LCD_DMA_BURST_SIZE(LCD_DMA_BURST_1);
> +		break;
> +	case 2:
> +		reg |= LCD_DMA_BURST_SIZE(LCD_DMA_BURST_2);
> +		break;
> +	case 4:
> +		reg |= LCD_DMA_BURST_SIZE(LCD_DMA_BURST_4);
> +		break;
> +	case 8:
> +		reg |= LCD_DMA_BURST_SIZE(LCD_DMA_BURST_8);
> +		break;
> +	case 16:
> +		reg |= LCD_DMA_BURST_SIZE(LCD_DMA_BURST_16);
> +		break;
> +	default:
> +		return -EINVAL;
> +	}
> +	lcdc_write(reg | LCD_END_OF_FRAME_INT_ENA, LCD_DMA_CTRL_REG);
> +
> +	return 0;
> +}
> +
> +static void lcd_cfg_ac_bias(int period, int transitions_per_int)
> +{
> +	u32 reg;
> +
> +	/* Set the AC Bias Period and Number of Transisitons per Interrupt */
> +	reg = lcdc_read(LCD_RASTER_TIMING_2_REG) & 0xFFF00000;
> +	reg |= LCD_AC_BIAS_FREQUENCY(period) |
> +		LCD_AC_BIAS_TRANSITIONS_PER_INT(transitions_per_int);
> +	lcdc_write(reg, LCD_RASTER_TIMING_2_REG);
> +}
> +
> +static void lcd_cfg_horizontal_sync(int back_porch, int pulse_width,
> +		int front_porch)
> +{
> +	u32 reg;
> +
> +	reg = lcdc_read(LCD_RASTER_TIMING_0_REG) & 0xf;
> +	reg |= ((back_porch & 0xff) << 24)
> +	    | ((front_porch & 0xff) << 16)
> +	    | ((pulse_width & 0x3f) << 10);
> +	lcdc_write(reg, LCD_RASTER_TIMING_0_REG);
> +}
> +
> +static void lcd_cfg_vertical_sync(int back_porch, int pulse_width,
> +		int front_porch)
> +{
> +	u32 reg;
> +
> +	reg = lcdc_read(LCD_RASTER_TIMING_1_REG) & 0x3ff;
> +	reg |= ((back_porch & 0xff) << 24)
> +	    | ((front_porch & 0xff) << 16)
> +	    | ((pulse_width & 0x3f) << 10);
> +	lcdc_write(reg, LCD_RASTER_TIMING_1_REG);
> +}
> +
> +static int lcd_cfg_display(const struct lcd_ctrl_config *cfg)
> +{
> +	u32 reg;
> +
> +	reg = lcdc_read(LCD_RASTER_CTRL_REG) & ~(LCD_TFT_MODE |
> +						LCD_MONO_8BIT_MODE |
> +						LCD_MONOCHROME_MODE);
> +
> +	switch (cfg->p_disp_panel->panel_shade) {
> +	case MONOCHROME:
> +		reg |= LCD_MONOCHROME_MODE;
> +		if (cfg->mono_8bit_mode)
> +			reg |= LCD_MONO_8BIT_MODE;
> +		break;
> +	case COLOR_ACTIVE:
> +		reg |= LCD_TFT_MODE;
> +		if (cfg->tft_alt_mode)
> +			reg |= LCD_TFT_ALT_ENABLE;
> +		break;
> +
> +	case COLOR_PASSIVE:
> +		if (cfg->stn_565_mode)
> +			reg |= LCD_STN_565_ENABLE;
> +		break;
> +
> +	default:
> +		return -EINVAL;
> +	}
> +
> +	/* enable additional interrupts here */
> +	reg |= LCD_UNDERFLOW_INT_ENA;
> +
> +	lcdc_write(reg, LCD_RASTER_CTRL_REG);
> +
> +	reg = lcdc_read(LCD_RASTER_TIMING_2_REG);
> +
> +	if (cfg->sync_ctrl)
> +		reg |= LCD_SYNC_CTRL;
> +	else
> +		reg &= ~LCD_SYNC_CTRL;
> +
> +	if (cfg->sync_edge)
> +		reg |= LCD_SYNC_EDGE;
> +	else
> +		reg &= ~LCD_SYNC_EDGE;
> +
> +	if (cfg->invert_pxl_clock)
> +		reg |= LCD_INVERT_PIXEL_CLOCK;
> +	else
> +		reg &= ~LCD_INVERT_PIXEL_CLOCK;
> +
> +	if (cfg->invert_line_clock)
> +		reg |= LCD_INVERT_LINE_CLOCK;
> +	else
> +		reg &= ~LCD_INVERT_LINE_CLOCK;
> +
> +	if (cfg->invert_frm_clock)
> +		reg |= LCD_INVERT_FRAME_CLOCK;
> +	else
> +		reg &= ~LCD_INVERT_FRAME_CLOCK;
> +
> +	lcdc_write(reg, LCD_RASTER_TIMING_2_REG);
> +
> +	return 0;
> +}
> +
> +static int lcd_cfg_frame_buffer(struct da8xx_fb_par *par, u32 width, u32 height,
> +		u32 bpp, u32 raster_order)
> +{
> +	u32 bpl, reg;
> +
> +	/* Disable Dual Frame Buffer. */
> +	reg = lcdc_read(LCD_DMA_CTRL_REG);
> +	lcdc_write(reg & ~LCD_DUAL_FRAME_BUFFER_ENABLE,
> +						LCD_DMA_CTRL_REG);
> +	/* Set the Panel Width */
> +	/* Pixels per line = (PPL + 1)*16 */
> +	/*0x3F in bits 4..9 gives max horisontal resolution = 1024 pixels*/
> +	width &= 0x3f0;
> +	reg = lcdc_read(LCD_RASTER_TIMING_0_REG);
> +	reg &= 0xfffffc00;
> +	reg |= ((width >> 4) - 1) << 4;
> +	lcdc_write(reg, LCD_RASTER_TIMING_0_REG);
> +
> +	/* Set the Panel Height */
> +	reg = lcdc_read(LCD_RASTER_TIMING_1_REG);
> +	reg = ((height - 1) & 0x3ff) | (reg & 0xfffffc00);
> +	lcdc_write(reg, LCD_RASTER_TIMING_1_REG);
> +
> +	/* Set the Raster Order of the Frame Buffer */
> +	reg = lcdc_read(LCD_RASTER_CTRL_REG) & ~(1 << 8);
> +	if (raster_order)
> +		reg |= LCD_RASTER_ORDER;
> +	lcdc_write(reg, LCD_RASTER_CTRL_REG);
> +
> +	switch (bpp) {
> +	case 1:
> +	case 2:
> +	case 4:
> +	case 16:
> +		par->palette_sz = 16 * 2;
> +		break;
> +
> +	case 8:
> +		par->palette_sz = 256 * 2;
> +		break;
> +
> +	default:
> +		return -EINVAL;
> +	}
> +
> +	bpl = width * bpp / 8;
> +	par->databuf_sz = height * bpl + par->palette_sz;
> +
> +	return 0;
> +}
> +
> +static int fb_setcolreg(unsigned regno, unsigned red, unsigned green,
> +			      unsigned blue, unsigned transp,
> +			      struct fb_info *info)
> +{
> +	struct da8xx_fb_par *par = info->par;
> +	unsigned short *palette = (unsigned short *)par->v_palette_base;
> +	u_short pal;
> +
> +	if (regno > 255)
> +		return 1;
> +
> +	if (info->fix.visual == FB_VISUAL_DIRECTCOLOR)
> +		return 1;
> +
> +	if (info->var.bits_per_pixel == 8) {
> +		red >>= 4;
> +		green >>= 8;
> +		blue >>= 12;
> +
> +		pal = (red & 0x0f00);
> +		pal |= (green & 0x00f0);
> +		pal |= (blue & 0x000f);
> +
> +		palette[regno] = pal;
> +
> +	} else if ((info->var.bits_per_pixel == 16) && regno < 16) {
> +		red >>= (16 - info->var.red.length);
> +		red <<= info->var.red.offset;
> +
> +		green >>= (16 - info->var.green.length);
> +		green <<= info->var.green.offset;
> +
> +		blue >>= (16 - info->var.blue.length);
> +		blue <<= info->var.blue.offset;
> +
> +		par->pseudo_palette[regno] = red | green | blue;
> +
> +		palette[0] = 0x4000;
> +	}
> +
> +	return 0;
> +}
> +
> +static int lcd_reset(struct da8xx_fb_par *par)
> +{
> +	int ret = 0;
> +
> +	/* Disable the Raster if previously Enabled */
> +	if (lcdc_read(LCD_RASTER_CTRL_REG) & LCD_RASTER_ENABLE)
> +		ret = lcd_disable_raster(par);
> +
> +	/* DMA has to be disabled */
> +	lcdc_write(0, LCD_DMA_CTRL_REG);
> +	lcdc_write(0, LCD_RASTER_CTRL_REG);
> +
> +	return ret;
> +}
> +
> +static int lcd_init(struct da8xx_fb_par *par, const struct lcd_ctrl_config *cfg,
> +		struct da8xx_panel *panel)
> +{
> +	u32 bpp, ret = 0;
> +
> +	ret = lcd_reset(par);
> +	if (ret != 0)
> +		return ret;
> +
> +	/* Configure the LCD clock divisor. */
> +	lcdc_write(LCD_CLK_DIVISOR(panel->pxl_clk) |
> +			(LCD_RASTER_MODE & 0x1), LCD_CTRL_REG);
> +
> +	/* Configure the DMA burst size. */
> +	ret = lcd_cfg_dma(cfg->dma_burst_sz);
> +	if (ret < 0)
> +		return ret;
> +
> +	/* Configure the AC bias properties. */
> +	lcd_cfg_ac_bias(cfg->ac_bias, cfg->ac_bias_intrpt);
> +
> +	/* Configure the vertical and horizontal sync properties. */
> +	lcd_cfg_vertical_sync(panel->vbp, panel->vsw, panel->vfp);
> +	lcd_cfg_horizontal_sync(panel->hbp, panel->hsw, panel->hfp);
> +
> +	/* Configure for disply */
> +	ret = lcd_cfg_display(cfg);
> +	if (ret < 0)
> +		return ret;
> +
> +	if (QVGA != cfg->p_disp_panel->panel_type)
> +		return -EINVAL;
> +
> +	if (cfg->bpp <= cfg->p_disp_panel->max_bpp &&
> +	    cfg->bpp >= cfg->p_disp_panel->min_bpp)
> +		bpp = cfg->bpp;
> +	else
> +		bpp = cfg->p_disp_panel->max_bpp;
> +	if (bpp == 12)
> +		bpp = 16;
> +	ret = lcd_cfg_frame_buffer(par, (unsigned int)panel->width,
> +				(unsigned int)panel->height, bpp,
> +				cfg->raster_order);
> +	if (ret < 0)
> +		return ret;
> +
> +	/* Configure FDD */
> +	lcdc_write((lcdc_read(LCD_RASTER_CTRL_REG) & 0xfff00fff) |
> +		       (cfg->fdd << 12), LCD_RASTER_CTRL_REG);
> +
> +	return 0;
> +}
> +
> +static irqreturn_t lcdc_irq_handler(int irq, void *arg)
> +{
> +	u32 stat = lcdc_read(LCD_STAT_REG);
> +	struct da8xx_fb_par *par = arg;
> +	u32 reg;
> +
> +	if ((stat & LCD_SYNC_LOST) && (stat & LCD_FIFO_UNDERFLOW)) {
> +		reg = lcdc_read(LCD_RASTER_CTRL_REG);
> +		lcdc_write(reg & ~LCD_RASTER_ENABLE, LCD_RASTER_CTRL_REG);
> +		lcdc_write(stat, LCD_STAT_REG);
> +		lcdc_write(reg | LCD_RASTER_ENABLE, LCD_RASTER_CTRL_REG);
> +	} else
> +		lcdc_write(stat, LCD_STAT_REG);
> +
> +	wake_up_interruptible(&par->da8xx_wq);
> +	return IRQ_HANDLED;
> +}
> +
> +static int fb_check_var(struct fb_var_screeninfo *var,
> +			struct fb_info *info)
> +{
> +	int err = 0;
> +
> +	switch (var->bits_per_pixel) {
> +	case 1:
> +	case 8:
> +		var->red.offset = 0;
> +		var->red.length = 8;
> +		var->green.offset = 0;
> +		var->green.length = 8;
> +		var->blue.offset = 0;
> +		var->blue.length = 8;
> +		var->transp.offset = 0;
> +		var->transp.length = 0;
> +		break;
> +	case 4:
> +		var->red.offset = 0;
> +		var->red.length = 4;
> +		var->green.offset = 0;
> +		var->green.length = 4;
> +		var->blue.offset = 0;
> +		var->blue.length = 4;
> +		var->transp.offset = 0;
> +		var->transp.length = 0;
> +		break;
> +	case 16:		/* RGB 565 */
> +		var->red.offset = 0;
> +		var->red.length = 5;
> +		var->green.offset = 5;
> +		var->green.length = 6;
> +		var->blue.offset = 11;
> +		var->blue.length = 5;
> +		var->transp.offset = 0;
> +		var->transp.length = 0;
> +		break;
> +	default:
> +		err = -EINVAL;
> +	}
> +
> +	var->red.msb_right = 0;
> +	var->green.msb_right = 0;
> +	var->blue.msb_right = 0;
> +	var->transp.msb_right = 0;
> +	return err;
> +}
> +
> +static int __devexit fb_remove(struct platform_device *dev)
> +{
> +	struct fb_info *info = dev_get_drvdata(&dev->dev);
> +	int ret = 0;
> +
> +	if (info) {
> +		struct da8xx_fb_par *par = info->par;
> +
> +		if (lcdc_read(LCD_RASTER_CTRL_REG) & LCD_RASTER_ENABLE)
> +			ret = lcd_disable_raster(par);
> +		lcdc_write(0, LCD_RASTER_CTRL_REG);
> +
> +		/* disable DMA  */
> +		lcdc_write(0, LCD_DMA_CTRL_REG);
> +
> +		unregister_framebuffer(info);
> +		fb_dealloc_cmap(&info->cmap);
> +		dma_free_coherent(NULL, par->databuf_sz + PAGE_SIZE,
> +					info->screen_base,
> +					info->fix.smem_start);
> +		free_irq(par->irq, NULL);
> +		clk_disable(par->lcdc_clk);
> +		clk_put(par->lcdc_clk);
> +		framebuffer_release(info);
> +
> +	}
> +	return ret;
> +}
> +
> +static int fb_ioctl(struct fb_info *info, unsigned int cmd,
> +			  unsigned long arg)
> +{
> +	struct lcd_sync_arg sync_arg;
> +
> +	switch (cmd) {
> +	case FBIOGET_CONTRAST:
> +	case FBIOPUT_CONTRAST:
> +	case FBIGET_BRIGHTNESS:
> +	case FBIPUT_BRIGHTNESS:
> +	case FBIGET_COLOR:
> +	case FBIPUT_COLOR:
> +		return -EINVAL;
> +	case FBIPUT_HSYNC:
> +		if (copy_from_user(&sync_arg, (char *)arg,
> +				sizeof(struct lcd_sync_arg)))
> +			return -EINVAL;
> +		lcd_cfg_horizontal_sync(sync_arg.back_porch,
> +					sync_arg.pulse_width,
> +					sync_arg.front_porch);
> +		break;
> +	case FBIPUT_VSYNC:
> +		if (copy_from_user(&sync_arg, (char *)arg,
> +				sizeof(struct lcd_sync_arg)))
> +			return -EINVAL;
> +		lcd_cfg_vertical_sync(sync_arg.back_porch,
> +					sync_arg.pulse_width,
> +					sync_arg.front_porch);
> +		break;
> +	default:
> +		return -EINVAL;
> +	}
> +	return 0;
> +}
> +
> +static struct fb_ops da8xx_fb_ops = {
> +	.owner = THIS_MODULE,
> +	.fb_check_var = fb_check_var,
> +	.fb_setcolreg = fb_setcolreg,
> +	.fb_ioctl = fb_ioctl,
> +	.fb_fillrect = cfb_fillrect,
> +	.fb_copyarea = cfb_copyarea,
> +	.fb_imageblit = cfb_imageblit,
> +};
> +
> +static int __init fb_probe(struct platform_device *device)
> +{
> +	struct da8xx_lcdc_platform_data *fb_pdata =
> +						device->dev.platform_data;
> +	struct lcd_ctrl_config *lcd_cfg;
> +	struct da8xx_panel *lcdc_info;
> +	struct fb_info *da8xx_fb_info;
> +	struct resource *lcdc_regs;
> +	struct clk *fb_clk = NULL;
> +	struct da8xx_fb_par *par;
> +	resource_size_t len;
> +	int ret, i;
> +
> +	if (fb_pdata == NULL) {
> +		dev_err(&device->dev, "Can not get platform data\n");
> +		return -ENOENT;
> +	}
> +
> +	lcdc_regs = platform_get_resource(device, IORESOURCE_MEM, 0);
> +	if (!lcdc_regs) {
> +		dev_err(&device->dev,
> +			"Can not get memory resource for LCD controller\n");
> +		return -ENOENT;
> +	}
> +
> +	len = lcdc_regs->end - lcdc_regs->start + 1;
> +
> +	lcdc_regs = request_mem_region(lcdc_regs->start, len, lcdc_regs->name);
> +	if (!lcdc_regs)
> +		return -EBUSY;
> +
> +	da8xx_fb_reg_base = (resource_size_t)ioremap(lcdc_regs->start, len);
> +
> +	fb_clk = clk_get(&device->dev, NULL);
> +	if (IS_ERR(fb_clk)) {
> +		dev_err(&device->dev, "Can not get device clock\n");
> +		ret = -ENODEV;
> +		goto err_request_mem;
> +	}
> +	ret = clk_enable(fb_clk);
> +	if (ret)
> +		goto err_clk_put;
> +
> +	for (i = 0, lcdc_info = known_lcd_panels;
> +		i < ARRAY_SIZE(known_lcd_panels);
> +		i++, lcdc_info++) {
> +		if (strcmp(fb_pdata->type, lcdc_info->name) == 0)
> +			break;
> +	}
> +
> +	if (i == ARRAY_SIZE(known_lcd_panels)) {
> +		dev_err(&device->dev, "GLCD: No valid panel found\n");
> +		ret = ENODEV;
> +		goto err_clk_disable;
> +	} else
> +		dev_info(&device->dev, "GLCD: Found %s panel\n",
> +					fb_pdata->type);
> +
> +	lcd_cfg = (struct lcd_ctrl_config *)fb_pdata->controller_data;
> +
> +	da8xx_fb_info = framebuffer_alloc(sizeof(struct da8xx_fb_par),
> +					&device->dev);
> +	if (!da8xx_fb_info) {
> +		dev_dbg(&device->dev, "Memory allocation failed for fb_info\n");
> +		ret = -ENOMEM;
> +		goto err_clk_disable;
> +	}
> +
> +	par = da8xx_fb_info->par;
> +
> +	if (lcd_init(par, lcd_cfg, lcdc_info) < 0) {
> +		dev_err(&device->dev, "lcd_init failed\n");
> +		ret = -EFAULT;
> +		goto err_release_fb;
> +	}
> +
> +	/* allocate frame buffer */
> +	da8xx_fb_info->screen_base = dma_alloc_coherent(NULL,
> +					par->databuf_sz + PAGE_SIZE,
> +					(resource_size_t *)
> +					&da8xx_fb_info->fix.smem_start,
> +					GFP_KERNEL | GFP_DMA);
> +
> +	if (!da8xx_fb_info->screen_base) {
> +		dev_err(&device->dev,
> +			"GLCD: kmalloc for frame buffer failed\n");
> +		ret = -EINVAL;
> +		goto err_release_fb;
> +	}
> +
> +	/* move palette base pointer by (PAGE_SIZE - palette_sz) bytes */
> +	par->v_palette_base = da8xx_fb_info->screen_base +
> +				(PAGE_SIZE - par->palette_sz);
> +	par->p_palette_base = da8xx_fb_info->fix.smem_start +
> +				(PAGE_SIZE - par->palette_sz);
> +
> +	/* the rest of the frame buffer is pixel data */
> +	da8xx_fb_info->screen_base = par->v_palette_base + par->palette_sz;
> +	da8xx_fb_fix.smem_start = par->p_palette_base + par->palette_sz;
> +	da8xx_fb_fix.smem_len = par->databuf_sz - par->palette_sz;
> +	da8xx_fb_fix.line_length = (lcdc_info->width * lcd_cfg->bpp) / 8;
> +
> +	par->lcdc_clk = fb_clk;
> +
> +	init_waitqueue_head(&par->da8xx_wq);
> +
> +	par->irq = platform_get_irq(device, 0);
> +	if (par->irq < 0) {
> +		ret = -ENOENT;
> +		goto err_release_fb_mem;
> +	}
> +
> +	ret = request_irq(par->irq, lcdc_irq_handler, 0, DRIVER_NAME, par);
> +	if (ret)
> +		goto err_release_fb_mem;
> +
> +	/* Initialize par */
> +	da8xx_fb_info->var.bits_per_pixel = lcd_cfg->bpp;
> +
> +	da8xx_fb_var.xres = lcdc_info->width;
> +	da8xx_fb_var.xres_virtual = lcdc_info->width;
> +
> +	da8xx_fb_var.yres = lcdc_info->height;
> +	da8xx_fb_var.yres_virtual = lcdc_info->height;
> +
> +	da8xx_fb_var.grayscale =
> +	    lcd_cfg->p_disp_panel->panel_shade == MONOCHROME ? 1 : 0;
> +	da8xx_fb_var.bits_per_pixel = lcd_cfg->bpp;
> +
> +	da8xx_fb_var.hsync_len = lcdc_info->hsw;
> +	da8xx_fb_var.vsync_len = lcdc_info->vsw;
> +
> +	/* Initialize fbinfo */
> +	da8xx_fb_info->flags = FBINFO_FLAG_DEFAULT;
> +	da8xx_fb_info->fix = da8xx_fb_fix;
> +	da8xx_fb_info->var = da8xx_fb_var;
> +	da8xx_fb_info->fbops = &da8xx_fb_ops;
> +	da8xx_fb_info->pseudo_palette = par->pseudo_palette;
> +
> +	ret = fb_alloc_cmap(&da8xx_fb_info->cmap, PALETTE_SIZE, 0);
> +	if (ret)
> +		goto err_free_irq;
> +
> +	/* First palette_sz byte of the frame buffer is the palette */
> +	da8xx_fb_info->cmap.len = par->palette_sz;
> +
> +	/* Flush the buffer to the screen. */
> +	lcd_blit(LOAD_DATA, par);
> +
> +	/* initialize var_screeninfo */
> +	da8xx_fb_var.activate = FB_ACTIVATE_FORCE;
> +	fb_set_var(da8xx_fb_info, &da8xx_fb_var);
> +
> +	dev_set_drvdata(&device->dev, da8xx_fb_info);
> +	/* Register the Frame Buffer  */
> +	if (register_framebuffer(da8xx_fb_info) < 0) {
> +		dev_err(&device->dev,
> +			"GLCD: Frame Buffer Registration Failed!\n");
> +		ret = -EINVAL;
> +		goto err_dealloc_cmap;
> +	}
> +
> +	/* enable raster engine */
> +	lcdc_write(lcdc_read(LCD_RASTER_CTRL_REG) |
> +			LCD_RASTER_ENABLE, LCD_RASTER_CTRL_REG);
> +
> +	return 0;
> +
> +err_dealloc_cmap:
> +	fb_dealloc_cmap(&da8xx_fb_info->cmap);
> +
> +err_free_irq:
> +	free_irq(par->irq, NULL);
> +
> +err_release_fb_mem:
> +	dma_free_coherent(NULL, par->databuf_sz + PAGE_SIZE,
> +				da8xx_fb_info->screen_base,
> +				da8xx_fb_info->fix.smem_start);
> +
> +err_release_fb:
> +	framebuffer_release(da8xx_fb_info);
> +
> +err_clk_disable:
> +	clk_disable(fb_clk);
> +
> +err_clk_put:
> +	clk_put(fb_clk);
> +
> +err_request_mem:
> +	release_mem_region(lcdc_regs->start, len);;
> +
> +	return ret;
> +}
> +
> +#ifdef CONFIG_PM
> +static int fb_suspend(struct platform_device *dev, pm_message_t state)
> +{
> +	 return -EBUSY;
> +}
> +static int fb_resume(struct platform_device *dev)
> +{
> +	 return -EBUSY;
> +}
> +#else
> +#define fb_suspend NULL
> +#define fb_resume NULL
> +#endif
> +
> +static struct platform_driver da8xx_fb_driver = {
> +	.probe = fb_probe,
> +	.remove = fb_remove,
> +	.suspend = fb_suspend,
> +	.resume = fb_resume,
> +	.driver = {
> +		   .name = DRIVER_NAME,
> +		   .owner = THIS_MODULE,
> +		   },
> +};
> +
> +static int __init da8xx_fb_init(void)
> +{
> +	return platform_driver_register(&da8xx_fb_driver);
> +}
> +
> +static void __exit da8xx_fb_cleanup(void)
> +{
> +	platform_driver_unregister(&da8xx_fb_driver);
> +}
> +
> +module_init(da8xx_fb_init);
> +module_exit(da8xx_fb_cleanup);
> +
> +MODULE_DESCRIPTION("Framebuffer driver for TI da8xx/omap-l1xx");
> +MODULE_AUTHOR("Texas Instruments");
> +MODULE_LICENSE("GPL");
> diff --git a/include/video/da8xx-fb.h b/include/video/da8xx-fb.h
> new file mode 100644
> index 0000000..5f77675
> --- /dev/null
> +++ b/include/video/da8xx-fb.h
> @@ -0,0 +1,106 @@
> +/*
> + * Header file for TI DA8XX LCD controller platform data.
> + *
> + * Copyright (C) 2008-2009 MontaVista Software Inc.
> + * Copyright (C) 2008-2009 Texas Instruments Inc
> + *
> + * This file is licensed under the terms of the GNU General Public License
> + * version 2. This program is licensed "as is" without any warranty of any
> + * kind, whether express or implied.
> + */
> +
> +#ifndef DA8XX_FB_H
> +#define DA8XX_FB_H
> +
> +enum panel_type {
> +	QVGA = 0
> +};
> +
> +enum panel_shade {
> +	MONOCHROME = 0,
> +	COLOR_ACTIVE,
> +	COLOR_PASSIVE,
> +};
> +
> +enum raster_load_mode {
> +	LOAD_DATA = 1,
> +	LOAD_PALETTE,
> +};
> +
> +struct display_panel {
> +	enum panel_type panel_type; /* QVGA */
> +	int max_bpp;
> +	int min_bpp;
> +	enum panel_shade panel_shade;
> +};
> +
> +struct da8xx_lcdc_platform_data {
> +	const char manu_name[10];
> +	void *controller_data;
> +	const char type[25];
> +};
> +
> +struct lcd_ctrl_config {
> +	const struct display_panel *p_disp_panel;
> +
> +	/* AC Bias Pin Frequency */
> +	int ac_bias;
> +
> +	/* AC Bias Pin Transitions per Interrupt */
> +	int ac_bias_intrpt;
> +
> +	/* DMA burst size */
> +	int dma_burst_sz;
> +
> +	/* Bits per pixel */
> +	int bpp;
> +
> +	/* FIFO DMA Request Delay */
> +	int fdd;
> +
> +	/* TFT Alternative Signal Mapping (Only for active) */
> +	unsigned char tft_alt_mode;
> +
> +	/* 12 Bit Per Pixel (5-6-5) Mode (Only for passive) */
> +	unsigned char stn_565_mode;
> +
> +	/* Mono 8-bit Mode: 1=D0-D7 or 0=D0-D3 */
> +	unsigned char mono_8bit_mode;
> +
> +	/* Invert pixel clock */
> +	unsigned char invert_pxl_clock;
> +
> +	/* Invert line clock */
> +	unsigned char invert_line_clock;
> +
> +	/* Invert frame clock  */
> +	unsigned char invert_frm_clock;
> +
> +	/* Horizontal and Vertical Sync Edge: 0=rising 1=falling */
> +	unsigned char sync_edge;
> +
> +	/* Horizontal and Vertical Sync: Control: 0=ignore */
> +	unsigned char sync_ctrl;
> +
> +	/* Raster Data Order Select: 1=Most-to-least 0=Least-to-most */
> +	unsigned char raster_order;
> +};
> +
> +struct lcd_sync_arg {
> +	int back_porch;
> +	int front_porch;
> +	int pulse_width;
> +};
> +
> +/* ioctls */
> +#define FBIOGET_CONTRAST	_IOR('F', 1, int)
> +#define FBIOPUT_CONTRAST	_IOW('F', 2, int)
> +#define FBIGET_BRIGHTNESS	_IOR('F', 3, int)
> +#define FBIPUT_BRIGHTNESS	_IOW('F', 3, int)
> +#define FBIGET_COLOR		_IOR('F', 5, int)
> +#define FBIPUT_COLOR		_IOW('F', 6, int)
> +#define FBIPUT_HSYNC		_IOW('F', 9, int)
> +#define FBIPUT_VSYNC		_IOW('F', 10, int)
> +
> +#endif  /* ifndef DA8XX_FB_H */
> +
> -- 
> 1.5.6
> 
> 

----------------------------------------------------------------------
Rozwiaz krzyzowke i  wygraj nagrody! 
Sprawdz >>  http://link.interia.pl/f2232 

[Linux-fbdev-devel] matroxfb: fix regression with uninitalized fb_info->mm_lock mutex

[Krzysztof H. <krz...@po...>]

From: Krzysztof Helt <krz...@wp...>

Remove redundant call to the matroxfb_update_fix() before matrox frambuffer
is registered.

This fixes a problem with uninitialized the fb_info->mm_lock mutex introduced 
by the commit 537a1bf059f " fbdev: add mutex for fb_mmap locking"

Signed-off-by: Krzysztof Helt <krz...@wp...>

---

This one I have been able to test. It fixed regression with the mm_lock mutex.

It may not fix regression for a second head. I will investigate this issue.

diff --git a/drivers/video/matrox/matroxfb_base.c b/drivers/video/matrox/matroxfb_base.c
index 76bc51b..721a87d 100644
--- a/drivers/video/matrox/matroxfb_base.c
+++ b/drivers/video/matrox/matroxfb_base.c
@@ -1876,7 +1876,6 @@ static int initMatrox2(WPMINFO struct board* b){
 	}
 	matroxfb_init_fix(PMINFO2);
 	ACCESS_FBINFO(fbcon.screen_base) = vaddr_va(ACCESS_FBINFO(video.vbase));
-	matroxfb_update_fix(PMINFO2);
 	/* Normalize values (namely yres_virtual) */
 	matroxfb_check_var(&vesafb_defined, &ACCESS_FBINFO(fbcon));
 	/* And put it into "current" var. Do NOT program hardware yet, or we'll not take over

----------------------------------------------------------------------
Promocja ubezpieczen komunikacyjnych Ergo Hestia. Sprawdz!
http://link.interia.pl/f222f

[Linux-fbdev-devel] atmelfb: fix regression with uninitalized fb_info->mm_lock mutex

[Krzysztof H. <krz...@po...>]

From: Krzysztof Helt <krz...@wp...>

Remove redundant locking of the fb_info->mm_lock mutex before the frambuffer 
is registered.

This fixes a problem with uninitialized the fb_info->mm_lock mutex introduced 
by the commit 537a1bf059f " fbdev: add mutex for fb_mmap locking"

Signed-off-by: Krzysztof Helt <krz...@wp...>
---

Nicolas, could you test this patch?

Kind regard,
Krzysztof

diff --git a/drivers/video/atmel_lcdfb.c b/drivers/video/atmel_lcdfb.c
index cb88394..da05f08 100644
--- a/drivers/video/atmel_lcdfb.c
+++ b/drivers/video/atmel_lcdfb.c
@@ -261,6 +261,9 @@ static inline void atmel_lcdfb_free_video_memory(struct atmel_lcdfb_info *sinfo)
 /**
  *	atmel_lcdfb_alloc_video_memory - Allocate framebuffer memory
  *	@sinfo: the frame buffer to allocate memory for
+ * 	
+ * 	This function is called only from the atmel_lcdfb_probe()
+ * 	so no locking by fb_info->mm_lock around smem_len setting is needed.
  */
 static int atmel_lcdfb_alloc_video_memory(struct atmel_lcdfb_info *sinfo)
 {
@@ -270,9 +273,7 @@ static int atmel_lcdfb_alloc_video_memory(struct atmel_lcdfb_info *sinfo)
 
 	smem_len = (var->xres_virtual * var->yres_virtual
 		    * ((var->bits_per_pixel + 7) / 8));
-	mutex_lock(&info->mm_lock);
 	info->fix.smem_len = max(smem_len, sinfo->smem_len);
-	mutex_unlock(&info->mm_lock);
 
 	info->screen_base = dma_alloc_writecombine(info->device, info->fix.smem_len,
 					(dma_addr_t *)&info->fix.smem_start, GFP_KERNEL);


----------------------------------------------------------------------
Daj odpoczac plecom - zmien auto na rower 
Sprawdz >>>http://link.interia.pl/f2255

[Linux-fbdev-devel] atmelfb: fix regression with uninitalized fb_info->mm_lock mutex

[Krzysztof H. <krz...@po...>]

From: Krzysztof Helt <krz...@wp...>

Remove redundant locking of the fb_info->mm_lock mutex before the frambuffer 
is registered.

This fixes a problem with uninitialized the fb_info->mm_lock mutex introduced 
by the commit 537a1bf059f " fbdev: add mutex for fb_mmap locking"

Signed-off-by: Krzysztof Helt <krz...@wp...>
---

diff --git a/drivers/video/atmel_lcdfb.c b/drivers/video/atmel_lcdfb.c
index cb88394..da05f08 100644
--- a/drivers/video/atmel_lcdfb.c
+++ b/drivers/video/atmel_lcdfb.c
@@ -261,6 +261,9 @@ static inline void atmel_lcdfb_free_video_memory(struct atmel_lcdfb_info *sinfo)
 /**
  *	atmel_lcdfb_alloc_video_memory - Allocate framebuffer memory
  *	@sinfo: the frame buffer to allocate memory for
+ * 	
+ * 	This function is called only from the atmel_lcdfb_probe()
+ * 	so no locking by fb_info->mm_lock around smem_len setting is needed.
  */
 static int atmel_lcdfb_alloc_video_memory(struct atmel_lcdfb_info *sinfo)
 {
@@ -270,9 +273,7 @@ static int atmel_lcdfb_alloc_video_memory(struct atmel_lcdfb_info *sinfo)
 
 	smem_len = (var->xres_virtual * var->yres_virtual
 		    * ((var->bits_per_pixel + 7) / 8));
-	mutex_lock(&info->mm_lock);
 	info->fix.smem_len = max(smem_len, sinfo->smem_len);
-	mutex_unlock(&info->mm_lock);
 
 	info->screen_base = dma_alloc_writecombine(info->device, info->fix.smem_len,
 					(dma_addr_t *)&info->fix.smem_start, GFP_KERNEL);


----------------------------------------------------------------------
O czym spiewa Ayo w piosence &#8222;Lonely&#8221;?
U nas teledysk razem z tekstem piosenki! 
>>> http://link.interia.pl/f2240

[Linux-fbdev-devel] fsl-diu-fb: fix regression with uninitalized fb_info->mm_lock mutex

[Krzysztof H. <krz...@po...>]

From: Krzysztof Helt <krz...@wp...>

Remove redundant call to the fsl_diu_set_par before the register_framebuffer().

This fixes a problem with uninitialized the fb_info->mm_lock mutex introduced 
by the commit 537a1bf059f " fbdev: add mutex for fb_mmap locking"

Signed-off-by: Krzysztof Helt <krz...@wp...>

---
If anybody could test this patch with the 2.6.30-rc2 kernel I am deeply grateful.

The fb_set_par() function is called after register_framebuffer() when fbcon
is taking over the console so a call just before the register_framebuffer()
is redundant.

Regards,
Krzysztof

diff --git a/drivers/video/fsl-diu-fb.c b/drivers/video/fsl-diu-fb.c
index 0bf2190..72d68b3 100644
--- a/drivers/video/fsl-diu-fb.c
+++ b/drivers/video/fsl-diu-fb.c
@@ -1223,12 +1223,6 @@ static int __devinit install_fb(struct fb_info *info)
 		return -EINVAL;
 	}
 
-	if (fsl_diu_set_par(info)) {
-		printk(KERN_ERR "fb_set_par failed");
-		fb_dealloc_cmap(&info->cmap);
-		return -EINVAL;
-	}
-
 	if (register_framebuffer(info) < 0) {
 		printk(KERN_ERR "register_framebuffer failed");
 		unmap_video_memory(info);

----------------------------------------------------------------------
Rozwiaz krzyzowke i  wygraj nagrody! 
Sprawdz >>  http://link.interia.pl/f2232 

[Linux-fbdev-devel] w100fb: fix regression with uninitalized fb_info->mm_lock mutex

[Krzysztof H. <krz...@po...>]

From: Krzysztof Helt <krz...@wp...>

Remove redundant call to the w100fb_set_par() before w100 frambuffer is registered.

This fixes a problem with uninitialized the fb_info->mm_lock mutex introduced 
by the commit 537a1bf059f " fbdev: add mutex for fb_mmap locking"

Signed-off-by: Krzysztof Helt <krz...@wp...>

---

diff --git a/drivers/video/w100fb.c b/drivers/video/w100fb.c
index 8a141c2..2376f68 100644
--- a/drivers/video/w100fb.c
+++ b/drivers/video/w100fb.c
@@ -748,8 +748,6 @@ int __init w100fb_probe(struct platform_device *pdev)
 		goto out;
 	}
 
-	w100fb_set_par(info);
-
 	if (register_framebuffer(info) < 0) {
 		err = -EINVAL;
 		goto out;

----------------------------------------------------------------------
Promocja ubezpieczen komunikacyjnych Ergo Hestia. Sprawdz!
http://link.interia.pl/f222f

[Linux-fbdev-devel] mx3fb: fix regression with uninitalized fb_info->mm_lock mutex

[Krzysztof H. <krz...@po...>]

From: Krzysztof Helt <krz...@wp...>

Remove redundant call to the mx3fb_set_par before the register_framebuffer().

This fixes a problem with uninitialized the fb_info->mm_lock mutex introduced 
by the commit 537a1bf059f " fbdev: add mutex for fb_mmap locking"

Signed-off-by: Krzysztof Helt <krz...@wp...>

---
Daniel,

Could you test this patch?

The mx3fb_set_par() is called from the register_framebuffer() anyway, so please check what 
to add to correctly unblank console for the first time. Maybe setting 

mx3_fbi->blank = FB_BLANK_UNBLANK;

before the call to register_framebuffer() is enough.

Regards,
Krzysztof

diff --git a/drivers/video/mx3fb.c b/drivers/video/mx3fb.c
index 567fb94..f8778cd 100644
--- a/drivers/video/mx3fb.c
+++ b/drivers/video/mx3fb.c
@@ -1365,11 +1365,6 @@ static int init_fb_chan(struct mx3fb_data *mx3fb, struct idmac_channel *ichan)
 	init_completion(&mx3fbi->flip_cmpl);
 	disable_irq(ichan->eof_irq);
 	dev_dbg(mx3fb->dev, "disabling irq %d\n", ichan->eof_irq);
-	ret = mx3fb_set_par(fbi);
-	if (ret < 0)
-		goto esetpar;
-
-	mx3fb_blank(FB_BLANK_UNBLANK, fbi);
 
 	dev_info(dev, "registered, using mode %s\n", fb_mode);
 

----------------------------------------------------------------------
Promocja ubezpieczen komunikacyjnych Ergo Hestia. Sprawdz!
http://link.interia.pl/f222f

[Linux-fbdev-devel] atafb: fix regression with uninitalized fb_info->mm_lock mutex

[Krzysztof H. <krz...@po...>]

From: Krzysztof Helt <krz...@wp...>

Remove redundant locking of the fb_info->mm_lock mutex before the frambuffer 
is registered.

This fixes a problem with uninitialized the fb_info->mm_lock mutex introduced 
by the commit 537a1bf059f " fbdev: add mutex for fb_mmap locking"

Signed-off-by: Krzysztof Helt <krz...@wp...>

---

diff --git a/drivers/video/atafb.c b/drivers/video/atafb.c
index 497ff8a..8cd279b 100644
--- a/drivers/video/atafb.c
+++ b/drivers/video/atafb.c
@@ -2405,6 +2405,9 @@ static int do_fb_set_var(struct fb_var_screeninfo *var, int isactive)
 	return 0;
 }
 
+/* fbhw->encode_fix() must be called with fb_info->mm_lock held
+ * if it is called after the register_framebuffer() - not a case here
+ */
 static int atafb_get_fix(struct fb_fix_screeninfo *fix, struct fb_info *info)
 {
 	struct atafb_par par;
@@ -2414,9 +2417,7 @@ static int atafb_get_fix(struct fb_fix_screeninfo *fix, struct fb_info *info)
 	if (err)
 		return err;
 	memset(fix, 0, sizeof(struct fb_fix_screeninfo));
-	mutex_lock(&info->mm_lock);
 	err = fbhw->encode_fix(fix, &par);
-	mutex_unlock(&info->mm_lock);
 	return err;
 }
 

----------------------------------------------------------------------
Ubezpieczenie komunikacyjne wraz z ubezpieczeniem 
kosztow leczenia za granic±. http://link.interia.pl/f222b

[Linux-fbdev-devel] i810fb: fix regression with uninitalized fb_info->mm_lock mutex

[Krzysztof H. <krz...@po...>]

From: Krzysztof Helt <krz...@wp...>

Remove redundant call to the encode_fix() before i810 frambuffer is registered.

This fixes a problem with uninitialized the fb_info->mm_lock mutex introduced 
by the commit 537a1bf059f " fbdev: add mutex for fb_mmap locking"

Signed-off-by: Krzysztof Helt <krz...@wp...>

---

diff --git a/drivers/video/i810/i810_main.c b/drivers/video/i810/i810_main.c
index 7196067..5743ea2 100644
--- a/drivers/video/i810/i810_main.c
+++ b/drivers/video/i810/i810_main.c
@@ -2060,8 +2060,7 @@ static int __devinit i810fb_init_pci (struct pci_dev *dev,
 
 	fb_var_to_videomode(&mode, &info->var);
 	fb_add_videomode(&mode, &info->modelist);
-	encode_fix(&info->fix, info); 
-	 	    
+
 	i810fb_init_ringbuffer(info);
 	err = register_framebuffer(info);
 


----------------------------------------------------------------------
Najlepsze OC i AC tylko w Ergo Hestia
http://link.interia.pl/f222

Re: [Linux-fbdev-devel] [PATCH] au1200fb: fix oops due to uninitialized fb_info.lock mutex

[Krzysztof H. <krz...@po...>]

On Tue, 07 Jul 2009 10:29:51 +0200
Manuel Lauss <man...@go...> wrote:

> commit 4148df9b0f38bdd362dd91d52076926c11cbe5a9 breaks au1200fb due to
> fb_info.lock now being uninitialized at the time of the
> register_frambuffer call:
> 
> CPU 0 Unable to handle kernel paging request at virtual address 00000000, epc == 8038ccd4, ra == 802412d4
> Call Trace:
> [<8038ccd4>] __mutex_lock_slowpath+0x3c/0xe4
> [<802412d4>] lock_fb_info+0x20/0x50
> [<80242158>] register_framebuffer+0x264/0x2a4
> [<8024f8c4>] au1200fb_drv_probe+0x308/0x444
> [<80273ef8>] driver_probe_device+0xbc/0x184
> [<8027402c>] __driver_attach+0x6c/0xa4
> [<802736bc>] bus_for_each_dev+0x60/0xb0
> [<80272e20>] bus_add_driver+0xc8/0x24c
> [<802743f4>] driver_register+0xdc/0x190
> [<8047d130>] au1200fb_init+0x39c/0x3cc
> [<80100460>] _stext+0x60/0x1c8
> [<8046f334>] kernel_init+0xc8/0x134
> [<80106f64>] kernel_thread_helper+0x10/0x18
> 
> This band-aid fixes this oops.
> 

Please wait a little. Currently, I, Paul Mundt and Linus discuss how to fix all drivers broken by the commit 4148df9b0f38bdd362dd91d52076.

Kind regards,
Krzysztof

 	 
----------------------------------------------------------------------
Sprawdz, czy wziac klapki, czy parasolke
Pogoda na dzis, jutro i pojutrze >> http://link.interia.pl/f224d

[Linux-fbdev-devel] [PATCH 3/4] [ARM] s3c-fb: Add support for display panning

[Pawel O. <p.o...@sa...>]

Added support for panning the display in all directions.


Reviewed-by: Marek Szyprowski <m.s...@sa...>
Reviewed-by: Kyungmin Park <kyu...@sa...>
Signed-off-by: Pawel Osciak <p.o...@sa...>



diff --git a/drivers/video/s3c-fb.c b/drivers/video/s3c-fb.c
index ff0b1a3..77b77a2 100644
--- a/drivers/video/s3c-fb.c
+++ b/drivers/video/s3c-fb.c
@@ -294,6 +294,8 @@ static int s3c_fb_set_par(struct fb_info *info)
 	}
 
 	info->fix.line_length = (var->xres_virtual * var->bits_per_pixel) / 8;
+	info->fix.xpanstep = info->var.xres_virtual > info->var.xres ? 1 : 0;
+	info->fix.ypanstep = info->var.yres_virtual > info->var.yres ? 1 : 0;
 
 	/* disable the window whilst we update it */
 	writel(0, regs + WINCON(win_no));
@@ -640,6 +642,38 @@ static int s3c_fb_blank(int blank_mode, struct fb_info *info)
 	return 0;
 }
 
+/**
+ * s3c_fb_pan_display() - Pan the display.
+ *
+ * Note, that the offsets can be written to the device at any time, as their
+ * values are latched at each vsync automatically. This also means that only
+ * the last call to this function will have any effect on next vsync, but
+ * there is no need to sleep waiting for it to prevent tearing.
+ *
+ * @var: The screen information to verify.
+ * @info: The framebuffer device.
+ */
+static int s3c_fb_pan_display(struct fb_var_screeninfo *var,
+			      struct fb_info *info)
+{
+	struct s3c_fb_win *win	= info->par;
+	struct s3c_fb *sfb	= win->parent;
+	unsigned int start_byte_offset, end_byte_offset;
+
+	/* Offset in bytes to the start of the displayed area */
+	start_byte_offset = var->yoffset * info->fix.line_length
+			    + var->xoffset * (info->var.bits_per_pixel) / 8;
+	writel(info->fix.smem_start + start_byte_offset,
+	       sfb->regs + VIDW_BUF_START(win->index));
+
+	/* Offset in bytes to the end of the displayed area */
+	end_byte_offset = start_byte_offset + var->yres * info->fix.line_length;
+	writel(info->fix.smem_start + end_byte_offset,
+	       sfb->regs + VIDW_BUF_END(win->index));
+
+	return 0;
+}
+
 static struct fb_ops s3c_fb_ops = {
 	.owner		= THIS_MODULE,
 	.fb_check_var	= s3c_fb_check_var,
@@ -649,6 +683,7 @@ static struct fb_ops s3c_fb_ops = {
 	.fb_fillrect	= cfb_fillrect,
 	.fb_copyarea	= cfb_copyarea,
 	.fb_imageblit	= cfb_imageblit,
+	.fb_pan_display	= s3c_fb_pan_display,
 };
 
 /**

[Linux-fbdev-devel] [PATCH 1/4] [ARM] s3c-fb: Fix an off-by-one error at window release

[Pawel O. <p.o...@sa...>]

    
This bug caused the release function being called on the first
uninitialized window, instead of starting with the last successfully
initialized one.    

Reviewed-by: Marek Szyprowski <m.s...@sa...>
Reviewed-by: Kyungmin Park <kyu...@sa...>
Signed-off-by: Pawel Osciak <p.o...@sa...>



diff --git a/drivers/video/s3c-fb.c b/drivers/video/s3c-fb.c
index bb63c07..ff0b1a3 100644
--- a/drivers/video/s3c-fb.c
+++ b/drivers/video/s3c-fb.c
@@ -926,7 +926,7 @@ static int __devinit s3c_fb_probe(struct platform_device *pdev)
 		ret = s3c_fb_probe_win(sfb, win, &sfb->windows[win]);
 		if (ret < 0) {
 			dev_err(dev, "failed to create window %d\n", win);
-			for (; win >= 0; win--)
+			for ( win = win - 1; win >= 0; win--)
 				s3c_fb_release_win(sfb, sfb->windows[win]);
 			goto err_ioremap;
 		}

[Linux-fbdev-devel] [PATCH 0/4] [ARM] S3C-fb update

[Pawel O. <p.o...@sa...>]

This patch series adds support for VSYNC wait ioctl
(FBIO_WAITFORVSYNC) and display panning for the S3C framebuffer.

It also contains minor fixes for the driver.

The series contains:

[PATCH 1/4] [ARM] s3c-fb: Fix an off-by-one error at window release
[PATCH 2/4] [ARM] s3c-fb: Include fixes
[PATCH 3/4] [ARM] s3c-fb: Add support for display panning
[PATCH 4/4] [ARM] s3c-fb: Add wait for vsync support


Best regards
--
Pawel Osciak
Linux Platform Group
Samsung Poland R&D Center

[Linux-fbdev-devel] [PATCH 4/4] [ARM] s3c-fb: Add wait for vsync support

[Pawel O. <p.o...@sa...>]

Added VSYNC interrupt support.
Added FBIO_WAITFORVSYNC ioctl that allows sleeping on vsync
waitqueue.


Reviewed-by: Marek Szyprowski <m.s...@sa...>
Reviewed-by: Kyungmin Park <kyu...@sa...>
Signed-off-by: Pawel Osciak <p.o...@sa...>



diff --git a/drivers/video/s3c-fb.c b/drivers/video/s3c-fb.c
index 77b77a2..0c064c2 100644
--- a/drivers/video/s3c-fb.c
+++ b/drivers/video/s3c-fb.c
@@ -21,6 +21,8 @@
 #include <linux/clk.h>
 #include <linux/fb.h>
 #include <linux/io.h>
+#include <linux/uaccess.h>
+#include <linux/interrupt.h>
 
 #include <mach/map.h>
 #include <mach/regs-fb.h>
@@ -48,6 +50,13 @@
 	__raw_writel(v, r); } while(0)
 #endif /* FB_S3C_DEBUG_REGWRITE */
 
+/* Bit in irq_flags used to mark that interrupts are activated in device */
+#define S3C_FB_IRQ_ENABLED_BIT	0
+
+#ifndef FBIO_WAITFORVSYNC
+#define FBIO_WAITFORVSYNC _IOW('F', 0x20, __u32)
+#endif
+
 struct s3c_fb;
 
 /**
@@ -59,6 +68,7 @@ struct s3c_fb;
  * @pseudo_palette: For use in TRUECOLOUR modes for entries 0..15/
  * @index: The window number of this window.
  * @palette: The bitfields for changing r/g/b into a hardware palette entry.
+ * @vsync: Vsync handling helper.
  */
 struct s3c_fb_win {
 	struct s3c_fb_pd_win	*windata;
@@ -72,14 +82,27 @@ struct s3c_fb_win {
 };
 
 /**
+ * struct s3c_fb_vsync - vsync information for display.
+ * @vsync_queue: A queue for processes waiting for vsync.
+ * @vsync_count: Vsync interrupt count.
+ */
+struct s3c_fb_vsync {
+	wait_queue_head_t	vsync_queue;
+	unsigned long		vsync_count;
+};
+
+/**
  * struct s3c_fb - overall hardware state of the hardware
  * @dev: The device that we bound to, for printing, etc.
  * @regs_res: The resource we claimed for the IO registers.
  * @bus_clk: The clk (hclk) feeding our interface and possibly pixclk.
  * @regs: The mapped hardware registers.
  * @enabled: A bitmask of enabled hardware windows.
+ * @open_instances: How many times the driver has been opened.
  * @pdata: The platform configuration data passed with the device.
  * @windows: The hardware windows that have been claimed.
+ * @irq_no: IRQ number for the device (acquired from platform data).
+ * @vsync_info: VSYNC-related information (count, queues...)
  */
 struct s3c_fb {
 	struct device		*dev;
@@ -87,10 +110,16 @@ struct s3c_fb {
 	struct clk		*bus_clk;
 	void __iomem		*regs;
 
-	unsigned char		 enabled;
+	unsigned char		enabled;
+	atomic_t		open_instances;
 
 	struct s3c_fb_platdata	*pdata;
 	struct s3c_fb_win	*windows[S3C_FB_MAX_WIN];
+
+	unsigned int		irq_no;
+	unsigned long		irq_flags;
+
+	struct s3c_fb_vsync	vsync_info;
 };
 
 /**
@@ -643,6 +672,104 @@ static int s3c_fb_blank(int blank_mode, struct fb_info *info)
 }
 
 /**
+ * s3c_fb_enable_irq() - enable interrupts in hardware.
+ * @sfb: The base resources for the hardware.
+ */
+static int s3c_fb_enable_irq(struct s3c_fb *sfb)
+{
+	u32	     irq_ctrl_reg;
+	void __iomem *regs = sfb->regs;
+
+	if (!test_and_set_bit(S3C_FB_IRQ_ENABLED_BIT, &sfb->irq_flags)) {
+		/* IRQ was disabled, enable it */
+		irq_ctrl_reg = readl(regs + VIDINTCON0);
+
+		irq_ctrl_reg |= VIDINTCON0_INT_ENABLE;
+		irq_ctrl_reg |= VIDINTCON0_INT_FRAME;
+
+		irq_ctrl_reg &= ~VIDINTCON0_FRAMESEL0_MASK;
+		irq_ctrl_reg |= VIDINTCON0_FRAMESEL0_VSYNC;
+		irq_ctrl_reg &= ~VIDINTCON0_FRAMESEL1_MASK;
+		irq_ctrl_reg |= VIDINTCON0_FRAMESEL1_NONE;
+
+		writel(irq_ctrl_reg, regs + VIDINTCON0);
+	}
+
+	return 0;
+}
+
+/**
+ * s3c_fb_disable_irq() - disable interrupts in hardware.
+ * @sfb: The base resources for the hardware.
+ */
+static int s3c_fb_disable_irq(struct s3c_fb *sfb)
+{
+	u32	     irq_ctrl_reg;
+	void __iomem *regs = sfb->regs;
+
+	if (test_and_clear_bit(S3C_FB_IRQ_ENABLED_BIT, &sfb->irq_flags)) {
+		/* IRQ was enabled, disable it */
+		irq_ctrl_reg = readl(regs + VIDINTCON0);
+
+		irq_ctrl_reg &= ~VIDINTCON0_INT_FRAME;
+		irq_ctrl_reg &= ~VIDINTCON0_INT_ENABLE;
+
+		writel(irq_ctrl_reg, regs + VIDINTCON0);
+	}
+
+	return 0;
+}
+
+static irqreturn_t s3c_fb_interrupt(int irq, void *dev_id)
+{
+	u32	      irq_sts_reg;
+	struct s3c_fb *sfb = dev_id;
+	void __iomem  *regs = sfb->regs;
+
+	irq_sts_reg = readl(regs + VIDINTCON1);
+
+	if (irq_sts_reg & VIDINTCON1_INT_FRAME) {
+
+		/* VSYNC interrupt, accept it */
+		writel(VIDINTCON1_INT_FRAME, regs + VIDINTCON1);
+
+		sfb->vsync_info.vsync_count++;
+		wake_up_interruptible(&sfb->vsync_info.vsync_queue);
+	}
+
+	/* We support only waiting for VSYNC interrupt for now,
+	 * so it's safe to disable irqs here. */
+	s3c_fb_disable_irq(sfb);
+
+	return IRQ_HANDLED;
+}
+
+/**
+ * s3c_fb_wait_for_vsync() - sleep until next VSYNC interrupt or timeout.
+ * @sfb: The base resources for the hardware.
+ * @crtc: Head index.
+ */
+static int s3c_fb_wait_for_vsync(struct s3c_fb *sfb, u32 crtc)
+{
+	unsigned long count;
+	int ret;
+
+	if (crtc != 0)
+		return -ENODEV;
+
+	s3c_fb_enable_irq(sfb);
+	count = sfb->vsync_info.vsync_count;
+	ret = wait_event_interruptible_timeout(sfb->vsync_info.vsync_queue,
+				       count != sfb->vsync_info.vsync_count,
+				       HZ / 10);
+	if (ret == 0)
+		return -ETIMEDOUT;
+
+
+	return 0;
+}
+
+/**
  * s3c_fb_pan_display() - Pan the display.
  *
  * Note, that the offsets can be written to the device at any time, as their
@@ -674,8 +801,53 @@ static int s3c_fb_pan_display(struct fb_var_screeninfo *var,
 	return 0;
 }
 
+static int s3c_fb_ioctl(struct fb_info *info, unsigned int cmd,
+			unsigned long arg)
+{
+	struct s3c_fb_win *win = info->par;
+	struct s3c_fb *sfb = win->parent;
+
+	switch (cmd) {
+	case FBIO_WAITFORVSYNC:
+		{
+			u32 crtc;
+			if (get_user(crtc, (u32 __user *)arg))
+				return -EFAULT;
+
+			return s3c_fb_wait_for_vsync(sfb, crtc);
+		}
+	default:
+		return -ENOTTY;
+	}
+
+	return 0;
+}
+
+static int s3c_fb_open(struct fb_info *info, int user)
+{
+	struct s3c_fb_win *win = info->par;
+	struct s3c_fb *sfb = win->parent;
+
+	atomic_inc(&sfb->open_instances);
+
+	return 0;
+}
+
+static int s3c_fb_release(struct fb_info *info, int user)
+{
+	struct s3c_fb_win *win = info->par;
+	struct s3c_fb *sfb = win->parent;
+
+	if (atomic_dec_and_test(&sfb->open_instances))
+		s3c_fb_disable_irq(sfb);
+
+	return 0;
+}
+
 static struct fb_ops s3c_fb_ops = {
 	.owner		= THIS_MODULE,
+	.fb_open	= s3c_fb_open,
+	.fb_release	= s3c_fb_release,
 	.fb_check_var	= s3c_fb_check_var,
 	.fb_set_par	= s3c_fb_set_par,
 	.fb_blank	= s3c_fb_blank,
@@ -683,6 +855,7 @@ static struct fb_ops s3c_fb_ops = {
 	.fb_fillrect	= cfb_fillrect,
 	.fb_copyarea	= cfb_copyarea,
 	.fb_imageblit	= cfb_imageblit,
+	.fb_ioctl	= s3c_fb_ioctl,
 	.fb_pan_display	= s3c_fb_pan_display,
 };
 
@@ -908,6 +1081,7 @@ static int __devinit s3c_fb_probe(struct platform_device *pdev)
 
 	sfb->dev = dev;
 	sfb->pdata = pd;
+	init_waitqueue_head(&sfb->vsync_info.vsync_queue);
 
 	sfb->bus_clk = clk_get(dev, "lcd");
 	if (IS_ERR(sfb->bus_clk)) {
@@ -939,6 +1113,20 @@ static int __devinit s3c_fb_probe(struct platform_device *pdev)
 		goto err_req_region;
 	}
 
+	res = platform_get_resource(pdev, IORESOURCE_IRQ, 0);
+	if (!res) {
+		dev_err(dev, "failed to acquire irq resource\n");
+		ret = -ENOENT;
+		goto err_ioremap;
+	}
+	sfb->irq_no = res->start;
+	ret = request_irq(sfb->irq_no, s3c_fb_interrupt,
+			  0, "s3c_fb", sfb);
+	if (ret) {
+		dev_err(dev, "irq request failed\n");
+		goto err_ioremap;
+	}
+
 	dev_dbg(dev, "got resources (regs %p), probing windows\n", sfb->regs);
 
 	/* setup gpio and output polarity controls */
@@ -963,7 +1151,7 @@ static int __devinit s3c_fb_probe(struct platform_device *pdev)
 			dev_err(dev, "failed to create window %d\n", win);
 			for ( win = win - 1; win >= 0; win--)
 				s3c_fb_release_win(sfb, sfb->windows[win]);
-			goto err_ioremap;
+			goto err_irq;
 		}
 	}
 
@@ -971,6 +1159,9 @@ static int __devinit s3c_fb_probe(struct platform_device *pdev)
 
 	return 0;
 
+err_irq:
+	free_irq(sfb->irq_no, sfb);
+
 err_ioremap:
 	iounmap(sfb->regs);
 
@@ -1003,6 +1194,8 @@ static int __devexit s3c_fb_remove(struct platform_device *pdev)
 		if (sfb->windows[win])
 			s3c_fb_release_win(sfb, sfb->windows[win]);
 
+	free_irq(sfb->irq_no, sfb);
+
 	iounmap(sfb->regs);
 
 	clk_disable(sfb->bus_clk);

[Linux-fbdev-devel] [PATCH 2/4] [ARM] s3c-fb: Include fixes

[Pawel O. <p.o...@sa...>]

Corrected defines for FRAMESEL1 framebuffer register.


Reviewed-by: Marek Szyprowski <m.s...@sa...>
Reviewed-by: Kyungmin Park <kyu...@sa...>
Signed-off-by: Pawel Osciak <p.o...@sa...>



diff --git a/arch/arm/plat-s3c/include/plat/regs-fb.h b/arch/arm/plat-s3c/include/plat/regs-fb.h
index e9ee599..8048cae 100644
--- a/arch/arm/plat-s3c/include/plat/regs-fb.h
+++ b/arch/arm/plat-s3c/include/plat/regs-fb.h
@@ -291,11 +291,12 @@
 #define VIDINTCON0_FRAMESEL0_ACTIVE		(0x2 << 15)
 #define VIDINTCON0_FRAMESEL0_FRONTPORCH		(0x3 << 15)
 
-#define VIDINTCON0_FRAMESEL1			(1 << 14)
-#define VIDINTCON0_FRAMESEL1_NONE		(0x0 << 14)
-#define VIDINTCON0_FRAMESEL1_BACKPORCH		(0x1 << 14)
-#define VIDINTCON0_FRAMESEL1_VSYNC		(0x2 << 14)
-#define VIDINTCON0_FRAMESEL1_FRONTPORCH		(0x3 << 14)
+#define VIDINTCON0_FRAMESEL1_MASK		(0x3 << 13)
+#define VIDINTCON0_FRAMESEL1_SHIFT		(13)
+#define VIDINTCON0_FRAMESEL1_NONE		(0x0 << 13)
+#define VIDINTCON0_FRAMESEL1_BACKPORCH		(0x1 << 13)
+#define VIDINTCON0_FRAMESEL1_VSYNC		(0x2 << 13)
+#define VIDINTCON0_FRAMESEL1_FRONTPORCH		(0x3 << 13)
 
 #define VIDINTCON0_INT_FRAME			(1 << 12)
 #define VIDINTCON0_FIFIOSEL_MASK		(0x7f << 5)