hci_h4p from linux-n900 branch: fix with new bluetooth for 3.13-rc1

hci_h4p from linux-n900 branch: fix with new bluetooth for 3.13-rc1

[Pavel Machek]

commit d18adf907651766c97eedae993f14c6e042914a6
Author: Pavel <pavel@ucw.cz>
Date:   Wed Nov 27 13:43:40 2013 +0100

Fix compilation of hci_h4p after bluetooth core changes. Only compile tested.
    
Signed-off-by: Pavel Machek <pavel@ucw.cz>

--

When you update to 3.13-rc1, you'll likely need this one.

(And good news is that it boots with that kernel, and video seems to work.)

index 7ed9d4e..dac3875 100644
--- a/drivers/bluetooth/hci_h4p/core.c
+++ b/drivers/bluetooth/hci_h4p/core.c
@@ -405,7 +405,7 @@ static inline void hci_h4p_recv_frame(struct hci_h4p_info *info,
 		}
 		hci_h4p_parse_fw_event(info, skb);
 	} else {
-		hci_recv_frame(skb);
+		hci_recv_frame(info->hdev, skb);
 		NBT_DBG("Frame sent to upper layer\n");
 	}
 }
@@ -789,10 +789,9 @@ static int hci_h4p_hci_close(struct hci_dev *hdev)
 	return 0;
 }
 
-static int hci_h4p_hci_send_frame(struct sk_buff *skb)
+static int hci_h4p_hci_send_frame(struct hci_dev *hdev, struct sk_buff *skb)
 {
 	struct hci_h4p_info *info;
-	struct hci_dev *hdev = (struct hci_dev *)skb->dev;
 	int err = 0;
 	unsigned long flags;
 
@@ -841,12 +840,6 @@ static int hci_h4p_hci_send_frame(struct sk_buff *skb)
 	return 0;
 }
 
-static int hci_h4p_hci_ioctl(struct hci_dev *hdev, unsigned int cmd,
-			     unsigned long arg)
-{
-	return -ENOIOCTLCMD;
-}
-
 static int hci_h4p_register_hdev(struct hci_h4p_info *info)
 {
 	struct hci_dev *hdev;
@@ -867,7 +860,6 @@ static int hci_h4p_register_hdev(struct hci_h4p_info *info)
 	hdev->close = hci_h4p_hci_close;
 	hdev->flush = hci_h4p_hci_flush;
 	hdev->send = hci_h4p_hci_send_frame;
-	hdev->ioctl = hci_h4p_hci_ioctl;
 	set_bit(HCI_QUIRK_RESET_ON_CLOSE, &hdev->quirks);
 
 	SET_HCIDEV_DEV(hdev, info->dev);
diff --git a/drivers/bluetooth/hci_h4p/fw-csr.c b/drivers/bluetooth/hci_h4p/fw-csr.c
index af880d9..020fa52 100644
--- a/drivers/bluetooth/hci_h4p/fw-csr.c
+++ b/drivers/bluetooth/hci_h4p/fw-csr.c
@@ -31,7 +31,7 @@ void hci_h4p_bc4_parse_fw_event(struct hci_h4p_info *info, struct sk_buff *skb)
 {
 	/* Check if this is fw packet */
 	if (skb->data[0] != 0xff) {
-		hci_recv_frame(skb);
+		hci_recv_frame(info->hdev, skb);
 		return;
 	}
 

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

Re: hci_h4p from linux-n900 branch: fix with new bluetooth for 3.13-rc1

[Pali Rohár]

[-- Attachment #1: Type: Text/Plain, Size: 2507 bytes --]

Hi, I already have this patch prepared for 3.13.

On Wednesday 27 November 2013 13:51:50 Pavel Machek wrote:
> commit d18adf907651766c97eedae993f14c6e042914a6
> Author: Pavel <pavel@ucw.cz>
> Date:   Wed Nov 27 13:43:40 2013 +0100
> 
> Fix compilation of hci_h4p after bluetooth core changes. Only
> compile tested.
> 
> Signed-off-by: Pavel Machek <pavel@ucw.cz>
> 
> --
> 
> When you update to 3.13-rc1, you'll likely need this one.
> 
> (And good news is that it boots with that kernel, and video
> seems to work.)
> 
> index 7ed9d4e..dac3875 100644
> --- a/drivers/bluetooth/hci_h4p/core.c
> +++ b/drivers/bluetooth/hci_h4p/core.c
> @@ -405,7 +405,7 @@ static inline void
> hci_h4p_recv_frame(struct hci_h4p_info *info, }
>  		hci_h4p_parse_fw_event(info, skb);
>  	} else {
> -		hci_recv_frame(skb);
> +		hci_recv_frame(info->hdev, skb);
>  		NBT_DBG("Frame sent to upper layer\n");
>  	}
>  }
> @@ -789,10 +789,9 @@ static int hci_h4p_hci_close(struct
> hci_dev *hdev) return 0;
>  }
> 
> -static int hci_h4p_hci_send_frame(struct sk_buff *skb)
> +static int hci_h4p_hci_send_frame(struct hci_dev *hdev,
> struct sk_buff *skb) {
>  	struct hci_h4p_info *info;
> -	struct hci_dev *hdev = (struct hci_dev *)skb->dev;
>  	int err = 0;
>  	unsigned long flags;
> 
> @@ -841,12 +840,6 @@ static int hci_h4p_hci_send_frame(struct
> sk_buff *skb) return 0;
>  }
> 
> -static int hci_h4p_hci_ioctl(struct hci_dev *hdev, unsigned
> int cmd, -			     unsigned long arg)
> -{
> -	return -ENOIOCTLCMD;
> -}
> -
>  static int hci_h4p_register_hdev(struct hci_h4p_info *info)
>  {
>  	struct hci_dev *hdev;
> @@ -867,7 +860,6 @@ static int hci_h4p_register_hdev(struct
> hci_h4p_info *info) hdev->close = hci_h4p_hci_close;
>  	hdev->flush = hci_h4p_hci_flush;
>  	hdev->send = hci_h4p_hci_send_frame;
> -	hdev->ioctl = hci_h4p_hci_ioctl;
>  	set_bit(HCI_QUIRK_RESET_ON_CLOSE, &hdev->quirks);
> 
>  	SET_HCIDEV_DEV(hdev, info->dev);
> diff --git a/drivers/bluetooth/hci_h4p/fw-csr.c
> b/drivers/bluetooth/hci_h4p/fw-csr.c index af880d9..020fa52
> 100644
> --- a/drivers/bluetooth/hci_h4p/fw-csr.c
> +++ b/drivers/bluetooth/hci_h4p/fw-csr.c
> @@ -31,7 +31,7 @@ void hci_h4p_bc4_parse_fw_event(struct
> hci_h4p_info *info, struct sk_buff *skb) {
>  	/* Check if this is fw packet */
>  	if (skb->data[0] != 0xff) {
> -		hci_recv_frame(skb);
> +		hci_recv_frame(info->hdev, skb);
>  		return;
>  	}

-- 
Pali Rohár
pali.rohar@gmail.com

[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

Re: hci_h4p from linux-n900 branch: fix with new bluetooth for 3.13-rc1

[Pavel Machek]

Hi!

> Hi, I already have this patch prepared for 3.13.

Cool :-). Do you plan to publish / is there 3.13 based tree I could
pull somewhere?

Thanks,	
								Pavel

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

Re: hci_h4p from linux-n900 branch: fix with new bluetooth for 3.13-rc1

[Pali Rohár]

[-- Attachment #1: Type: Text/Plain, Size: 456 bytes --]

On Wednesday 27 November 2013 20:13:20 Pavel Machek wrote:
> Hi!
> 
> > Hi, I already have this patch prepared for 3.13.
> 
> Cool :-). Do you plan to publish / is there 3.13 based tree I
> could pull somewhere?
> 
> Thanks,
> 								Pavel

Yes, I will rebase n900 patches on top of next 3.13 rc release 
which will have some patches already included. I have not created 
full n900 3.13 branch yet.

-- 
Pali Rohár
pali.rohar@gmail.com

[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

Re: hci_h4p from linux-n900 branch: fix with new bluetooth for 3.13-rc1

[Pali Rohár]

[-- Attachment #1: Type: Text/Plain, Size: 654 bytes --]

On Wednesday 27 November 2013 20:44:51 Pali Rohár wrote:
> On Wednesday 27 November 2013 20:13:20 Pavel Machek wrote:
> > Hi!
> > 
> > > Hi, I already have this patch prepared for 3.13.
> > 
> > Cool :-). Do you plan to publish / is there 3.13 based tree
> > I could pull somewhere?
> > 
> > Thanks,
> > 
> > 								Pavel
> 
> Yes, I will rebase n900 patches on top of next 3.13 rc release
> which will have some patches already included. I have not
> created full n900 3.13 branch yet.

New branch 3.13-rc2 is now on gitorious linux-n900 tree:
git://gitorious.org/linux-n900/linux-n900.git

-- 
Pali Rohár
pali.rohar@gmail.com

[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

Re: hci_h4p from linux-n900 branch: fix with new bluetooth for 3.13-rc1

[Pavel Machek]

Hi!

> > > > Hi, I already have this patch prepared for 3.13.
> > > 
> > > Cool :-). Do you plan to publish / is there 3.13 based tree
> > > I could pull somewhere?
> > > 
> > > Thanks,
> > > 
> > > 								Pavel
> > 
> > Yes, I will rebase n900 patches on top of next 3.13 rc release
> > which will have some patches already included. I have not
> > created full n900 3.13 branch yet.
> 
> New branch 3.13-rc2 is now on gitorious linux-n900 tree:
> git://gitorious.org/linux-n900/linux-n900.git

Branch seems to be named "v3.13-rc2-n900". Thanks!
									Pavel
-- 
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html

device-tree based boot on linux-n900 branch

[Pavel Machek]

On Sat 2013-11-30 18:29:38, Pavel Machek wrote:
> Hi!
> 
> > > > > Hi, I already have this patch prepared for 3.13.
> > > > 
> > > > Cool :-). Do you plan to publish / is there 3.13 based tree
> > > > I could pull somewhere?
> > > > 
> > > > Thanks,
> > > > 
> > > > 								Pavel
> > > 
> > > Yes, I will rebase n900 patches on top of next 3.13 rc release
> > > which will have some patches already included. I have not
> > > created full n900 3.13 branch yet.
> > 
> > New branch 3.13-rc2 is now on gitorious linux-n900 tree:
> > git://gitorious.org/linux-n900/linux-n900.git
> 
> Branch seems to be named "v3.13-rc2-n900". Thanks!

Is this one expected to boot in devicetree-based setup? If so, does
someone have .config?

Thanks,
								Pavel

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

Re: device-tree based boot on linux-n900 branch

[Pali Rohár]

[-- Attachment #1: Type: Text/Plain, Size: 1205 bytes --]

On Sunday 01 December 2013 00:15:32 Pavel Machek wrote:
> On Sat 2013-11-30 18:29:38, Pavel Machek wrote:
> > Hi!
> > 
> > > > > > Hi, I already have this patch prepared for 3.13.
> > > > > 
> > > > > Cool :-). Do you plan to publish / is there 3.13 based
> > > > > tree I could pull somewhere?
> > > > > 
> > > > > Thanks,
> > > > > 
> > > > > 								Pavel
> > > > 
> > > > Yes, I will rebase n900 patches on top of next 3.13 rc
> > > > release which will have some patches already included.
> > > > I have not created full n900 3.13 branch yet.
> > > 
> > > New branch 3.13-rc2 is now on gitorious linux-n900 tree:
> > > git://gitorious.org/linux-n900/linux-n900.git
> > 
> > Branch seems to be named "v3.13-rc2-n900". Thanks!
> 
> Is this one expected to boot in devicetree-based setup? If so,
> does someone have .config?
> 
> Thanks,
> 								Pavel

No, this my tree has rx51_defconfig which is used for classic 
booting with board files.

Sebastian, can you create new rx51_defconfig or at least patch 
which enable DT booting? Also some steps how to compile DT kernel 
will be usefull (maybe on http://elinux.org/N900).

-- 
Pali Rohár
pali.rohar@gmail.com

[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

Re: device-tree based boot on linux-n900 branch

[Sebastian Reichel]

[-- Attachment #1: Type: text/plain, Size: 415 bytes --]

On Sun, Dec 01, 2013 at 10:50:18AM +0100, Pali Rohár wrote:
> No, this my tree has rx51_defconfig which is used for classic 
> booting with board files.
> 
> Sebastian, can you create new rx51_defconfig or at least patch 
> which enable DT booting? Also some steps how to compile DT kernel 
> will be usefull (maybe on http://elinux.org/N900).

http://elinux.org/N900#DT_Kernel_Compilation

-- Sebastian

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 836 bytes --]

Re: hci_h4p from linux-n900 branch: fix with new bluetooth for 3.13-rc1

[Sebastian Reichel]

[-- Attachment #1: Type: text/plain, Size: 165 bytes --]

Hi Pali,

While talking about the hci_h4p: What's the status of the mainline
inclusion? I saw, that you sent it to linux-bluetooth guys some time
ago.

-- Sebastian

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 836 bytes --]

Re: hci_h4p from linux-n900 branch: fix with new bluetooth for 3.13-rc1

[Pali Rohár]

[-- Attachment #1: Type: Text/Plain, Size: 347 bytes --]

On Sunday 01 December 2013 11:48:40 Sebastian Reichel wrote:
> Hi Pali,
> 
> While talking about the hci_h4p: What's the status of the
> mainline inclusion? I saw, that you sent it to
> linux-bluetooth guys some time ago.
> 
> -- Sebastian

Yes, but code needs to be changed... see that thread.

-- 
Pali Rohár
pali.rohar@gmail.com

[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

Re: hci_h4p from linux-n900 branch: fix with new bluetooth for 3.13-rc1

[Sebastian Reichel]

[-- Attachment #1: Type: text/plain, Size: 527 bytes --]

On Sun, Dec 01, 2013 at 12:16:42PM +0100, Pali Rohár wrote:
> On Sunday 01 December 2013 11:48:40 Sebastian Reichel wrote:
> > While talking about the hci_h4p: What's the status of the
> > mainline inclusion? I saw, that you sent it to
> > linux-bluetooth guys some time ago.
> 
> Yes, but code needs to be changed... see that thread.

I only saw the few small suggestions from [0]. That's
why I was wondering about the status. Have there been
more mails?

-- Sebastian

[0] https://lkml.org/lkml/2013/10/24/375

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 836 bytes --]

Compile failure on linux-n900 branch -- MEDIA_CAMERA_SUPPORT needed for n900 compilation

[Pavel Machek]

Hi!

I just hit this one:

  CC      arch/arm/mach-omap2/board-rx51-camera.o
In file included from arch/arm/mach-omap2/board-rx51-camera.c:34:0:
arch/arm/mach-omap2/../../../drivers/media/platform/omap3isp/isp.h: In
function 'isp_pad_buffer_type':
arch/arm/mach-omap2/../../../drivers/media/platform/omap3isp/isp.h:347:19:
error: 'const struct v4l2_subdev' has no member named 'entity'
arch/arm/mach-omap2/../../../drivers/media/platform/omap3isp/isp.h:350:12:
error: 'const struct v4l2_subdev' has no member named 'entity'
make[1]: *** [arch/arm/mach-omap2/board-rx51-camera.o] Error 1
make: *** [arch/arm/mach-omap2] Error 2
make: *** Waiting for unfinished jobs....
Command exited with non-zero status 2

Problem seems to be that isp.h expects CONFIG_MEDIA_CONTROLLER to be set,
but board-rx51-camera.c is always compiled. We could do

config MACH_NOKIA_RX51
        bool "Nokia N900 (RX-51) phone"
-        depends on ARCH_OMAP3
+        depends on ARCH_OMAP3 && MEDIA_CONTROLLER
        default y
	select OMAP_PACKAGE_CBB
 
...but that's cruel to the user. Would this be acceptable?

config MACH_NOKIA_RX51
        bool "Nokia N900 (RX-51) phone"
        depends on ARCH_OMAP3
        default y
	select OMAP_PACKAGE_CBB
+	select MEDIA_CONTROLLER
+       select MEDIA_CAMERA_SUPPORT

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

Re: Compile failure on linux-n900 branch -- MEDIA_CAMERA_SUPPORT needed for n900 compilation

[Pali Rohár]

[-- Attachment #1: Type: Text/Plain, Size: 1778 bytes --]

On Sunday 01 December 2013 20:21:20 Pavel Machek wrote:
> Hi!
> 
> I just hit this one:
> 
>   CC      arch/arm/mach-omap2/board-rx51-camera.o
> In file included from
> arch/arm/mach-omap2/board-rx51-camera.c:34:0:
> arch/arm/mach-omap2/../../../drivers/media/platform/omap3isp/
> isp.h: In function 'isp_pad_buffer_type':
> arch/arm/mach-omap2/../../../drivers/media/platform/omap3isp/i
> sp.h:347:19: error: 'const struct v4l2_subdev' has no member
> named 'entity'
> arch/arm/mach-omap2/../../../drivers/media/platform/omap3isp/
> isp.h:350:12: error: 'const struct v4l2_subdev' has no member
> named 'entity' make[1]: ***
> [arch/arm/mach-omap2/board-rx51-camera.o] Error 1 make: ***
> [arch/arm/mach-omap2] Error 2
> make: *** Waiting for unfinished jobs....
> Command exited with non-zero status 2
> 
> Problem seems to be that isp.h expects CONFIG_MEDIA_CONTROLLER
> to be set, but board-rx51-camera.c is always compiled. We
> could do
> 
> config MACH_NOKIA_RX51
>         bool "Nokia N900 (RX-51) phone"
> -        depends on ARCH_OMAP3
> +        depends on ARCH_OMAP3 && MEDIA_CONTROLLER
>         default y
> 	select OMAP_PACKAGE_CBB
> 
> ...but that's cruel to the user. Would this be acceptable?
> 
> config MACH_NOKIA_RX51
>         bool "Nokia N900 (RX-51) phone"
>         depends on ARCH_OMAP3
>         default y
> 	select OMAP_PACKAGE_CBB
> +	select MEDIA_CONTROLLER
> +       select MEDIA_CAMERA_SUPPORT
> 
> 									Pavel

board-rx51-camera.c is not part of mainline kernel and will never 
be (due to DT migration). So maybe it could be better to add 
#ifdef to board-rx51-camera.c. Keeping changes for Makefile/Kconfig 
files is harder because of future rebase...

-- 
Pali Rohár
pali.rohar@gmail.com

[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

Re: Compile failure on linux-n900 branch -- MEDIA_CAMERA_SUPPORT needed for n900 compilation

[Pavel Machek]

Hi!

> board-rx51-camera.c is not part of mainline kernel and will never 
> be (due to DT migration). So maybe it could be better to add 
> #ifdef to board-rx51-camera.c. Keeping changes for Makefile/Kconfig 
> files is harder because of future rebase...

Ok, that makes sense.

This breaks compilation, too, but I don't think I want to fix that.

  Example code (PVR_EXAMPLES) [N/m/y] (NEW) 

(Select Y to break compile).

Thanks,
								Pavel

---

Fix compilation in !CONFIG_MEDIA_CAMERA_SUPPORT case, and make it such
that it is unlikely to cause merge conflicts later.

Signed-off-by: Pavel Machek <pavel@ucw.cz>

diff --git a/arch/arm/mach-omap2/board-rx51-camera.c b/arch/arm/mach-omap2/board-rx51-camera.c
index 8b4b59a..bae2cc0 100644
--- a/arch/arm/mach-omap2/board-rx51-camera.c
+++ b/arch/arm/mach-omap2/board-rx51-camera.c
@@ -31,6 +31,7 @@
 
 #include <asm/gpio.h>
 
+#ifdef CONFIG_MEDIA_CAMERA_SUPPORT
 #include "../../../drivers/media/platform/omap3isp/isp.h"
 #include "../../../drivers/media/platform/omap3isp/ispreg.h"
 
@@ -375,9 +376,11 @@ static struct isp_platform_data rx51_isp_platform_data = {
 		{ .dev_id = "2-0010"},
 	.subdevs = rx51_camera_subdevs,
 };
+#endif
 
 void __init rx51_camera_init(void)
 {
+#ifdef CONFIG_MEDIA_CAMERA_SUPPORT
 	if (rx51_camera_hw_init()) {
 		printk(KERN_WARNING "%s: Unable to initialize camera\n",
 		       __func__);
@@ -387,4 +390,5 @@ void __init rx51_camera_init(void)
 	if (omap3_init_camera(&rx51_isp_platform_data) < 0)
 		printk(KERN_WARNING "%s: Unable to register camera platform "
 		       "device\n", __func__);
+#endif
 }




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

Re: Compile failure on linux-n900 branch -- MEDIA_CAMERA_SUPPORT needed for n900 compilation

[Pavel Machek]

Hi!

> > I just hit this one:
> > 
> >   CC      arch/arm/mach-omap2/board-rx51-camera.o
> > In file included from
> > arch/arm/mach-omap2/board-rx51-camera.c:34:0:
> > arch/arm/mach-omap2/../../../drivers/media/platform/omap3isp/
> > isp.h: In function 'isp_pad_buffer_type':
> > arch/arm/mach-omap2/../../../drivers/media/platform/omap3isp/i
> > sp.h:347:19: error: 'const struct v4l2_subdev' has no member
> > named 'entity'
> > arch/arm/mach-omap2/../../../drivers/media/platform/omap3isp/
> > isp.h:350:12: error: 'const struct v4l2_subdev' has no member
> > named 'entity' make[1]: ***
> > [arch/arm/mach-omap2/board-rx51-camera.o] Error 1 make: ***
> > [arch/arm/mach-omap2] Error 2
> > make: *** Waiting for unfinished jobs....
> > Command exited with non-zero status 2
> > 
> > Problem seems to be that isp.h expects CONFIG_MEDIA_CONTROLLER
> > to be set, but board-rx51-camera.c is always compiled. We
> > could do

Interesting; with Sebastian DT config
(http://elektranox.org/n900-kernel-config.txt), I get the same



[   17.967651] omap_i2c 48070000.i2c: Too much work in one IRQ
[   17.968109] omap_i2c 48070000.i2c: Too much work in one IRQ
[   17.968566] omap_i2c 48070000.i2c: Too much work in one IRQ
[   17.969055] omap_i2c 48070000.i2c: Too much work in one IRQ
[   17.969512] omap_i2c 48070000.i2c: Too much work in one IRQ
[   17.969970] omap_i2c 48070000.i2c: Too much work in one IRQ
[   17.970428] omap_i2c 48070000.i2c: Too much work in one IRQ
[   17.970886] omap_i2c 48070000.i2c: Too much work in one IRQ
[   17.971313] omap_i2c 48070000.i2c: Too much work in one IRQ
[   17.967651] omap_i2c 48070000.i2c: Too much work in one IRQ
[   17.968109] omap_i2c 48070000.i2c: Too much work in one IRQ
[   17.968566] omap_i2c 48070000.i2c: Too much work in one IRQ
[   17.969055] omap_i2c 48070000.i2c: Too much work in one IRQ
[   17.969512] omap_i2c 48070000.i2c: Too much work in one IRQ
[   17.969970] omap_i2c 48070000.i2c: Too much work in one IRQ
[   17.970428] omap_i2c 48070000.i2c: Too much work in one IRQ
[   17.970886] omap_i2c 48070000.i2c: Too much work in one IRQ
[   17.971313] omap_i2c 48070000.i2c: Too much work in one IRQ

...infinite loop during bootup. So, Sebastian, it looks problem is in
linux-n900, too. Will switch to defconfig, that worked...

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

Re: Compile failure on linux-n900 branch -- MEDIA_CAMERA_SUPPORT needed for n900 compilation

[Pali Rohár]

[-- Attachment #1: Type: Text/Plain, Size: 1904 bytes --]

On Sunday 01 December 2013 21:01:13 Pavel Machek wrote:
> Hi!
> 
> > board-rx51-camera.c is not part of mainline kernel and will
> > never be (due to DT migration). So maybe it could be better
> > to add #ifdef to board-rx51-camera.c. Keeping changes for
> > Makefile/Kconfig files is harder because of future
> > rebase...
> 
> Ok, that makes sense.
> 
> This breaks compilation, too, but I don't think I want to fix
> that.
> 
>   Example code (PVR_EXAMPLES) [N/m/y] (NEW)
> 
> (Select Y to break compile).
> 
> Thanks,
> 								Pavel
> 
> ---
> 
> Fix compilation in !CONFIG_MEDIA_CAMERA_SUPPORT case, and make
> it such that it is unlikely to cause merge conflicts later.
> 
> Signed-off-by: Pavel Machek <pavel@ucw.cz>
> 
> diff --git a/arch/arm/mach-omap2/board-rx51-camera.c
> b/arch/arm/mach-omap2/board-rx51-camera.c index
> 8b4b59a..bae2cc0 100644
> --- a/arch/arm/mach-omap2/board-rx51-camera.c
> +++ b/arch/arm/mach-omap2/board-rx51-camera.c
> @@ -31,6 +31,7 @@
> 
>  #include <asm/gpio.h>
> 
> +#ifdef CONFIG_MEDIA_CAMERA_SUPPORT
>  #include "../../../drivers/media/platform/omap3isp/isp.h"
>  #include "../../../drivers/media/platform/omap3isp/ispreg.h"
> 
> @@ -375,9 +376,11 @@ static struct isp_platform_data
> rx51_isp_platform_data = { { .dev_id = "2-0010"},
>  	.subdevs = rx51_camera_subdevs,
>  };
> +#endif
> 
>  void __init rx51_camera_init(void)
>  {
> +#ifdef CONFIG_MEDIA_CAMERA_SUPPORT
>  	if (rx51_camera_hw_init()) {
>  		printk(KERN_WARNING "%s: Unable to initialize camera\n",
>  		       __func__);
> @@ -387,4 +390,5 @@ void __init rx51_camera_init(void)
>  	if (omap3_init_camera(&rx51_isp_platform_data) < 0)
>  		printk(KERN_WARNING "%s: Unable to register camera 
platform
> " "device\n", __func__);
> +#endif
>  }

Thanks, I commited patch to linux-n900 tree.

-- 
Pali Rohár
pali.rohar@gmail.com

[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

Re: Compile failure on linux-n900 branch -- MEDIA_CAMERA_SUPPORT needed for n900 compilation

[Pali Rohár]

[-- Attachment #1: Type: Text/Plain, Size: 1647 bytes --]

On Sunday 01 December 2013 21:25:00 Pavel Machek wrote:
> Hi!
> 
> Interesting; with Sebastian DT config
> (http://elektranox.org/n900-kernel-config.txt), I get the same
> 
> 
> 
> [   17.967651] omap_i2c 48070000.i2c: Too much work in one IRQ
> [   17.968109] omap_i2c 48070000.i2c: Too much work in one IRQ
> [   17.968566] omap_i2c 48070000.i2c: Too much work in one IRQ
> [   17.969055] omap_i2c 48070000.i2c: Too much work in one IRQ
> [   17.969512] omap_i2c 48070000.i2c: Too much work in one IRQ
> [   17.969970] omap_i2c 48070000.i2c: Too much work in one IRQ
> [   17.970428] omap_i2c 48070000.i2c: Too much work in one IRQ
> [   17.970886] omap_i2c 48070000.i2c: Too much work in one IRQ
> [   17.971313] omap_i2c 48070000.i2c: Too much work in one IRQ
> [   17.967651] omap_i2c 48070000.i2c: Too much work in one IRQ
> [   17.968109] omap_i2c 48070000.i2c: Too much work in one IRQ
> [   17.968566] omap_i2c 48070000.i2c: Too much work in one IRQ
> [   17.969055] omap_i2c 48070000.i2c: Too much work in one IRQ
> [   17.969512] omap_i2c 48070000.i2c: Too much work in one IRQ
> [   17.969970] omap_i2c 48070000.i2c: Too much work in one IRQ
> [   17.970428] omap_i2c 48070000.i2c: Too much work in one IRQ
> [   17.970886] omap_i2c 48070000.i2c: Too much work in one IRQ
> [   17.971313] omap_i2c 48070000.i2c: Too much work in one IRQ
> 
> ...infinite loop during bootup. So, Sebastian, it looks
> problem is in linux-n900, too. Will switch to defconfig, that
> worked...
> 
> 									Pavel

Today freemangordon wrote (on IRC), that i2c is broken in 3.13.

-- 
Pali Rohár
pali.rohar@gmail.com

[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

linux-n900: minor coding style fixes

[Pavel Machek]

Replace RX51 with N900 (RX-51) in user-visible places.

Fix comment style.

Simplify obvious code.

Add a fixme where error is not checked.

Signed-off-by: Pavel Machek <pavel@ucw.cz>


index e7604e5..64f7aca 100644
--- a/drivers/hsi/clients/cmt_speech.c
+++ b/drivers/hsi/clients/cmt_speech.c
@@ -642,7 +642,7 @@ static void cs_hsi_peek_on_data_complete(struct hsi_msg *msg)
 		cs_hsi_data_read_error(hi, msg);
 }
 
-/**
+/*
  * Read/write transaction is ongoing. Returns false if in
  * SSI_CHANNEL_STATE_POLL state.
  */
@@ -652,7 +652,7 @@ static inline int cs_state_xfer_active(unsigned int state)
 		(state & SSI_CHANNEL_STATE_READING);
 }
 
-/**
+/*
  * No pending read/writes
  */
 static inline int cs_state_idle(unsigned int state)
@@ -841,7 +841,7 @@ static int check_buf_params(struct cs_hsi_iface *hi,
 	return r;
 }
 
-/**
+/*
  * Block until pending data transfers have completed.
  */
 static int cs_hsi_data_sync(struct cs_hsi_iface *hi)
@@ -864,7 +864,7 @@ static int cs_hsi_data_sync(struct cs_hsi_iface *hi)
 			r = -ERESTARTSYS;
 			goto out;
 		}
-		/**
+		/*
 		 * prepare_to_wait must be called with hi->lock held
 		 * so that callbacks can check for waitqueue_active()
 		 */
diff --git a/drivers/media/i2c/Kconfig b/drivers/media/i2c/Kconfig
index 4544798..31ead21 100644
--- a/drivers/media/i2c/Kconfig
+++ b/drivers/media/i2c/Kconfig
@@ -271,7 +271,7 @@ config VIDEO_AD5820
 	depends on I2C && VIDEO_V4L2 && MEDIA_CONTROLLER
 	---help---
 	  This is a driver for the AD5820 camera lens voice coil.
-	  It is used for example in Nokia RX51.
+	  It is used for example in Nokia N900 (RX-51).
 
 config VIDEO_SAA7110
 	tristate "Philips SAA7110 video decoder"
diff --git a/drivers/media/i2c/ad5820.c b/drivers/media/i2c/ad5820.c
index d2574b7..0f91673 100644
--- a/drivers/media/i2c/ad5820.c
+++ b/drivers/media/i2c/ad5820.c
@@ -1,5 +1,5 @@
 /*
- * drivers/media/video/ad5820.c
+ * drivers/media/i2c/ad5820.c
  *
  * AD5820 DAC driver for camera voice coil focus.
  *
@@ -101,9 +101,11 @@ static int ad5820_read(struct ad5820_device *coil)
 	return be16_to_cpu(data);
 }
 
-/* Calculate status word and write it to the device based on current
+/*
+ * Calculate status word and write it to the device based on current
  * values of V4L2 controls. It is assumed that the stored V4L2 control
- * values are properly limited and rounded. */
+ * values are properly limited and rounded.
+ */
 static int ad5820_update_hw(struct ad5820_device *coil)
 {
 	u16 status;
@@ -119,15 +121,15 @@ static int ad5820_update_hw(struct ad5820_device *coil)
 	return ad5820_write(coil, status);
 }
 
-/* --------------------------------------------------------------------------
+/*
  * Power handling
  */
-
 static int ad5820_power_off(struct ad5820_device *coil, int standby)
 {
 	int ret = 0;
 
-	/* Go to standby first as real power off my be denied by the hardware
+	/*
+	 * Go to standby first as real power off my be denied by the hardware
 	 * (single power line control for both coil and sensor).
 	 */
 	if (standby) {
@@ -172,10 +174,9 @@ fail:
 	return ret;
 }
 
-/* --------------------------------------------------------------------------
+/*
  * V4L2 controls
  */
-
 static int ad5820_set_ctrl(struct v4l2_ctrl *ctrl)
 {
 	struct ad5820_device *coil =
@@ -244,7 +245,8 @@ static int ad5820_init_controls(struct ad5820_device *coil)
 
 	v4l2_ctrl_handler_init(&coil->ctrls, ARRAY_SIZE(ad5820_ctrls) + 1);
 
-	/* V4L2_CID_FOCUS_ABSOLUTE
+	/*
+	 * V4L2_CID_FOCUS_ABSOLUTE
 	 *
 	 * Minimum current is 0 mA, maximum is 100 mA. Thus, 1 code is
 	 * equivalent to 100/1023 = 0.0978 mA. Nevertheless, we do not use [mA]
@@ -273,10 +275,9 @@ static int ad5820_init_controls(struct ad5820_device *coil)
 	return 0;
 }
 
-/* --------------------------------------------------------------------------
+/*
  * V4L2 subdev operations
  */
-
 static int
 ad5820_registered(struct v4l2_subdev *subdev)
 {
@@ -323,7 +324,8 @@ ad5820_set_power(struct v4l2_subdev *subdev, int on)
 
 	mutex_lock(&coil->power_lock);
 
-	/* If the power count is modified from 0 to != 0 or from != 0 to 0,
+	/*
+	 * If the power count is modified from 0 to != 0 or from != 0 to 0,
 	 * update the power state.
 	 */
 	if (coil->power_count == !on) {
@@ -365,7 +367,7 @@ static const struct v4l2_subdev_internal_ops ad5820_internal_ops = {
 	.close = ad5820_close,
 };
 
-/* --------------------------------------------------------------------------
+/*
  * I2C driver
  */
 #ifdef CONFIG_PM
diff --git a/drivers/media/i2c/smia/Kconfig b/drivers/media/i2c/smia/Kconfig
index ed089e4..13ca043 100644
--- a/drivers/media/i2c/smia/Kconfig
+++ b/drivers/media/i2c/smia/Kconfig
@@ -14,4 +14,4 @@ config VIDEO_ET8EK8
 	select VIDEO_SMIAREGS
 	---help---
 	  This is a driver for the Toshiba ET8EK8 5 MP camera sensor.
-	  It is used for example in Nokia RX51.
+	  It is used for example in Nokia N900 (RX-51).
diff --git a/drivers/media/i2c/smia/smiaregs.c b/drivers/media/i2c/smia/smiaregs.c
index f88ac38..50193da 100644
--- a/drivers/media/i2c/smia/smiaregs.c
+++ b/drivers/media/i2c/smia/smiaregs.c
@@ -1,5 +1,5 @@
 /*
- * drivers/media/video/smiaregs.c
+ * drivers/media/i2c/smia/smiaregs.c
  *
  * Copyright (C) 2008 Nokia Corporation
  *
@@ -31,11 +31,8 @@
 #include <media/smiaregs.h>
 
 /*
- *
  * Video control helpers
- *
  */
-
 int smia_ctrl_find(const struct v4l2_queryctrl *ctrls, size_t nctrls, int id)
 {
 	size_t i;
@@ -183,11 +180,8 @@ int smia_mode_g_ctrl(const __u32 *ctrls, size_t nctrls, struct v4l2_control *vc,
 EXPORT_SYMBOL_GPL(smia_mode_g_ctrl);
 
 /*
- *
  * Reglist helpers
- *
  */
-
 static int smia_reglist_cmp(const void *a, const void *b)
 {
 	const struct smia_reglist **list1 = (const struct smia_reglist **)a,
@@ -197,22 +191,22 @@ static int smia_reglist_cmp(const void *a, const void *b)
 	if ((*list1)->type == SMIA_REGLIST_MODE &&
 	    (*list2)->type != SMIA_REGLIST_MODE)
 		return -1;
-	else if ((*list1)->type != SMIA_REGLIST_MODE &&
-		 (*list2)->type == SMIA_REGLIST_MODE)
+	if ((*list1)->type != SMIA_REGLIST_MODE &&
+	    (*list2)->type == SMIA_REGLIST_MODE)
 		return 1;
 
 	/* Descending width. */
 	if ((*list1)->mode.window_width > (*list2)->mode.window_width)
 		return -1;
-	else if ((*list1)->mode.window_width < (*list2)->mode.window_width)
+	if ((*list1)->mode.window_width < (*list2)->mode.window_width)
 		return 1;
 
 	if ((*list1)->mode.window_height > (*list2)->mode.window_height)
 		return -1;
-	else if ((*list1)->mode.window_height < (*list2)->mode.window_height)
+	if ((*list1)->mode.window_height < (*list2)->mode.window_height)
 		return 1;
-	else
-		return 0;
+
+	return 0;
 }
 
 /*
diff --git a/drivers/media/radio/radio-bcm2048.c b/drivers/media/radio/radio-bcm2048.c
index d948cac..087e321 100644
--- a/drivers/media/radio/radio-bcm2048.c
+++ b/drivers/media/radio/radio-bcm2048.c
@@ -353,6 +353,9 @@ static struct region_info region_configs[] = {
 
 /*
  *	I2C Interface read / write
+ *
+ * Note: callers use | operation to combine errors from multiple
+ * calls. So this has to return just single error value.
  */
 static int bcm2048_send_command(struct bcm2048_device *bdev, unsigned int reg,
 					unsigned int value)
@@ -492,6 +495,8 @@ static int bcm2048_set_rds_no_lock(struct bcm2048_device *bdev, u8 rds_on)
 		memset(&bdev->rds_info, 0, sizeof(bdev->rds_info));
 	}
 
+	/* FIXME: if (err) return err ? */
+
 	err = bcm2048_send_command(bdev, BCM2048_I2C_FM_RDS_SYSTEM,
 					bdev->cache_fm_rds_system);
 
@@ -633,10 +638,7 @@ static int bcm2048_get_fm_frequency(struct bcm2048_device *bdev)
 	if (err)
 		return err;
 
-	err = compose_u16(msb, lsb);
-	err += BCM2048_FREQUENCY_BASE;
-
-	return err;
+	return compose_u16(msb, lsb) + BCM2048_FREQUENCY_BASE;
 }
 
 static int bcm2048_set_fm_af_frequency(struct bcm2048_device *bdev,
@@ -678,10 +680,7 @@ static int bcm2048_get_fm_af_frequency(struct bcm2048_device *bdev)
 	if (err)
 		return err;
 
-	err = compose_u16(msb, lsb);
-	err += BCM2048_FREQUENCY_BASE;
-
-	return err;
+	return compose_u16(msb, lsb) + BCM2048_FREQUENCY_BASE;
 }
 
 static int bcm2048_set_fm_deemphasis(struct bcm2048_device *bdev, int d)
diff --git a/drivers/staging/tidspbridge/core/tiomap3430.c b/drivers/staging/tidspbridge/core/tiomap3430.c
index a4b4300..66e7fed 100644
--- a/drivers/staging/tidspbridge/core/tiomap3430.c
+++ b/drivers/staging/tidspbridge/core/tiomap3430.c
@@ -527,7 +527,7 @@ static int bridge_brd_start(struct bridge_dev_context *dev_ctxt,
 		/* Set the DSP clock rate */
 		(void)dev_get_symbol(dev_context->dev_obj,
 				     "_BRIDGEINIT_DSP_FREQ", &ul_dsp_clk_addr);
-		/*Set Autoidle Mode for IVA2 PLL */
+		/* Set Autoidle Mode for IVA2 PLL */
 		(*pdata->dsp_cm_write)(1,
 				OMAP3430_IVA2_MOD, OMAP3430_CM_AUTOIDLE_PLL);
 
diff --git a/sound/soc/omap/Kconfig b/sound/soc/omap/Kconfig
index a8319e6..45ea88a 100644
--- a/sound/soc/omap/Kconfig
+++ b/sound/soc/omap/Kconfig
@@ -25,14 +25,14 @@ config SND_OMAP_SOC_N810
 	  Say Y if you want to add support for SoC audio on Nokia N810.
 
 config SND_OMAP_SOC_RX51
-	tristate "SoC Audio support for Nokia RX-51"
+	tristate "SoC Audio support for Nokia N900 (RX-51)"
 	depends on SND_OMAP_SOC && I2C && ARM && (MACH_NOKIA_RX51 || COMPILE_TEST)
 	select SND_OMAP_SOC_MCBSP
 	select SND_SOC_TLV320AIC3X
 	select SND_SOC_TPA6130A2
 	help
-	  Say Y if you want to add support for SoC audio on Nokia RX-51
-	  hardware. This is also known as Nokia N900 product.
+	  Say Y if you want to add support for SoC audio on Nokia N900
+	  (RX-51) hardware.
 
 config SND_OMAP_SOC_AMS_DELTA
 	tristate "SoC Audio support for Amstrad E3 (Delta) videophone"

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

[patch] fix n900 bt error handling

[Pavel Machek]

Hi!

> I think that we can drop "new" version, because it has several
> problems and "old" version is working. So now it not make sense
> to use hack or develop "new" version anymore.

Ok... but this seems still needed. Please apply,
							Pavel

commit 6c65774280a7095dce8b1e995d86b2e079ef5fdb
Author: Pavel <pavel@ucw.cz>
Date:   Thu Dec 5 12:54:31 2013 +0100

Cleanups should be performed even if firmware load fails.
   
Signed-off-by: Pavel Machek <pavel@ucw.cz>

diff --git a/drivers/bluetooth/hci_h4p/core.c b/drivers/bluetooth/hci_h4p/core.c
index ead25c4..e76e889 100644
--- a/drivers/bluetooth/hci_h4p/core.c
+++ b/drivers/bluetooth/hci_h4p/core.c
@@ -942,7 +942,7 @@ again:
 	err = hci_h4p_read_fw(info, &fw_queue);
 	if (err < 0) {
 		dev_err(info->dev, "Cannot read firmware\n");
-		return err;
+		goto err_clean;
 	}
 
 	err = hci_h4p_send_fw(info, &fw_queue);



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

Re: [patch] fix n900 bt error handling

[Pali Rohár]

[-- Attachment #1: Type: Text/Plain, Size: 1132 bytes --]

On Thursday 05 December 2013 12:56:08 Pavel Machek wrote:
> Hi!
> 
> > I think that we can drop "new" version, because it has
> > several problems and "old" version is working. So now it
> > not make sense to use hack or develop "new" version
> > anymore.
> 
> Ok... but this seems still needed. Please apply,
> 							Pavel
> 
> commit 6c65774280a7095dce8b1e995d86b2e079ef5fdb
> Author: Pavel <pavel@ucw.cz>
> Date:   Thu Dec 5 12:54:31 2013 +0100
> 
> Cleanups should be performed even if firmware load fails.
> 
> Signed-off-by: Pavel Machek <pavel@ucw.cz>
> 
> diff --git a/drivers/bluetooth/hci_h4p/core.c
> b/drivers/bluetooth/hci_h4p/core.c index ead25c4..e76e889
> 100644
> --- a/drivers/bluetooth/hci_h4p/core.c
> +++ b/drivers/bluetooth/hci_h4p/core.c
> @@ -942,7 +942,7 @@ again:
>  	err = hci_h4p_read_fw(info, &fw_queue);
>  	if (err < 0) {
>  		dev_err(info->dev, "Cannot read firmware\n");
> -		return err;
> +		goto err_clean;
>  	}
> 
>  	err = hci_h4p_send_fw(info, &fw_queue);

Done, pushed to 3.12-rc5-n900 and 3.13-rc2-n900 branches.

-- 
Pali Rohár
pali.rohar@gmail.com

[-- Attachment #2: This is a digitally signed message part. --]
[-- Type: application/pgp-signature, Size: 198 bytes --]