"usb: gadget: pch_udc: Detecting VBUS through GPIO with interrupt" breaks sparc

PATCH Superseded Paul Gortmaker 4 participants 9 messages Mar 2, 2012 – Mar 7, 2012

Subsystems (2)

USB Linux-next tree

lore

Share & tools

lore

Paul Gortmaker

author

2012-03-02 22:01 UTC #0 lore raw

This commit:

---------------
commit 637b78eb31e0b167ed913f1750bb645dfeda38f0

Author: Tomoya MORINAGA <[email protected]>
Date:   Fri Feb 3 16:14:18 2012 +0900

usb: gadget: pch_udc: Detecting VBUS through GPIO with interrupt

Problem:
pch_udc continues operation even if VBUS becomes Low. pch_udc performs D+ pulling up before VBUS becomes High. USB device should be controlled according to VBUS state.

Root cause:
The current pch_udc is not always monitoring VBUS.

Solution:
The change of VBUS is detected using an interrupt of GPIO. If VBUS became Low, pch_udc handles 'disconnect'. After VBUS became High, a pull improves D+, and pch_udc handles 'connect'.

[ [email protected] : make it actually compile ]

Signed-off-by Tomoya MORINAGA <[email protected]>

Signed-off-by Felipe Balbi <[email protected]>

---------------

breaks the sparc builds in next:

http://kisskb.ellerman.id.au/kisskb/buildresult/5771890/
http://kisskb.ellerman.id.au/kisskb/buildresult/5771905/

Thanks,
Paul.

Stephen Rothwell

2012-03-02 23:41 UTC #1 lore raw

[Just cc'ing Greg as this commit is in the usb tree]

Quoted 37 lines · On Fri, 2 Mar 2012 17:01:16 -0500 Paul Gortmaker…

This commit:
---------------
commit 637b78eb31e0b167ed913f1750bb645dfeda38f0
Author: Tomoya MORINAGA <[email protected]>
Date:   Fri Feb 3 16:14:18 2012 +0900
usb: gadget: pch_udc: Detecting VBUS through GPIO with interrupt
Problem:
pch_udc continues operation even if VBUS becomes Low. pch_udc performs D+ pulling up before VBUS becomes High. USB device should be controlled according to VBUS state.
Root cause:
The current pch_udc is not always monitoring VBUS.
Solution:
The change of VBUS is detected using an interrupt of GPIO. If VBUS became Low, pch_udc handles 'disconnect'. After VBUS became High, a pull improves D+, and pch_udc handles 'connect'.
[ [email protected] : make it actually compile ]
Signed-off-by Tomoya MORINAGA <[email protected]>
Signed-off-by Felipe Balbi <[email protected]>
---------------
breaks the sparc builds in next:
http://kisskb.ellerman.id.au/kisskb/buildresult/5771890/
http://kisskb.ellerman.id.au/kisskb/buildresult/5771905/
Thanks,
Paul.

Signature · 2 lines

Cheers,
Stephen Rothwell                    [email protected]

Greg KH

maintainer

2012-03-05 16:53 UTC #2 lore raw

Quoted 37 lines · Stephen Rothwell

[Just cc'ing Greg as this commit is in the usb tree]
Quoted 34 lines · On Fri, 2 Mar 2012 17:01:16 -0500 Paul Gortmaker…
This commit:
---------------
commit 637b78eb31e0b167ed913f1750bb645dfeda38f0
Author: Tomoya MORINAGA <[email protected]>
Date:   Fri Feb 3 16:14:18 2012 +0900
usb: gadget: pch_udc: Detecting VBUS through GPIO with interrupt
Problem:
pch_udc continues operation even if VBUS becomes Low. pch_udc performs D+ pulling up before VBUS becomes High. USB device should be controlled according to VBUS state.
Root cause:
The current pch_udc is not always monitoring VBUS.
Solution:
The change of VBUS is detected using an interrupt of GPIO. If VBUS became Low, pch_udc handles 'disconnect'. After VBUS became High, a pull improves D+, and pch_udc handles 'connect'.
[ [email protected] : make it actually compile ]
Signed-off-by Tomoya MORINAGA <[email protected]>
Signed-off-by Felipe Balbi <[email protected]>
---------------
breaks the sparc builds in next:
http://kisskb.ellerman.id.au/kisskb/buildresult/5771890/
http://kisskb.ellerman.id.au/kisskb/buildresult/5771905/

Ick, not good.

Tomoya, can you please send me a patch to fix this? Felipe is off-line for a few days and can't do this himself.

greg k-h

Tomoya MORINAGA

2012-03-06 00:07 UTC #3 lore raw

Hi Greg

Sorry, for inconvenience.
I'll re-submit ASAP.

thanks

Signature · 4 lines

ROHM Co., Ltd.
tomoya

2012年3月6日1:43 Greg KH <[email protected]>:

Quoted 46 lines · Stephen Rothwell

Quoted 37 lines
[Just cc'ing Greg as this commit is in the usb tree]
Quoted 34 lines · On Fri, 2 Mar 2012 17:01:16 -0500 Paul Gortmaker…
This commit:
---------------
commit 637b78eb31e0b167ed913f1750bb645dfeda38f0
Author: Tomoya MORINAGA <[email protected]>
Date:   Fri Feb 3 16:14:18 2012 +0900
usb: gadget: pch_udc: Detecting VBUS through GPIO with interrupt
Problem:
pch_udc continues operation even if VBUS becomes Low. pch_udc performs D+ pulling up before VBUS becomes High. USB device should be controlled according to VBUS state.
Root cause:
The current pch_udc is not always monitoring VBUS.
Solution:
The change of VBUS is detected using an interrupt of GPIO. If VBUS became Low, pch_udc handles 'disconnect'. After VBUS became High, a pull improves D+, and pch_udc handles 'connect'.
[ [email protected] : make it actually compile ]
Signed-off-by Tomoya MORINAGA <[email protected]>
Signed-off-by Felipe Balbi <[email protected]>
---------------
breaks the sparc builds in next:
http://kisskb.ellerman.id.au/kisskb/buildresult/5771890/
http://kisskb.ellerman.id.au/kisskb/buildresult/5771905/
Ick, not good.
Tomoya, can you please send me a patch to fix this? Felipe is off-line for a few days and can't do this himself.
greg k-h

Tomoya MORINAGA

2012-03-06 02:35 UTC #4 lore raw

PATCH 1/1 usb/gadget/pch_udc: Fix compile error

Though I've tested this patch,
http://marc.info/?l=linux-usb&m=132825305710285&w=2,
I've received the following reports.
http://kisskb.ellerman.id.au/kisskb/buildresult/5771890/
http://kisskb.ellerman.id.au/kisskb/buildresult/5771905/

So, I added header file for these symbols.
Using this patch, this compile error must be disappeared.

Signed-off-by Tomoya MORINAGA <[email protected]>

---
 drivers/usb/gadget/pch_udc.c |    1 +
 1 files changed, 1 insertions(+), 0 deletions(-)

drivers/usb/gadget/pch_udc.c +1 -0

index 350dbcd..6530706 100644
@@ -16,6 +16,7 @@
 #include <linux/usb/ch9.h>
 #include <linux/usb/gadget.h>
 #include <linux/gpio.h>
+#include <linux/irq.h>
 
 /* GPIO port for VBUS detecting */
 static int vbus_gpio_port = -1;		/* GPIO port number (-1:Not used) */

Signature · 1 line

1.7.7.6

Tomoya MORINAGA

2012-03-06 02:49 UTC #5 lore raw

PATCH RESEND 1/1 usb/gadget/pch_udc: Fix compile error

In reply to Greg KH

Greg's e-mail address was old. So, I resend it.

So, I added header file for these symbols.
Using this patch, this compile error must be disappeared.

Signed-off-by Tomoya MORINAGA <[email protected]>

---
 drivers/usb/gadget/pch_udc.c |    1 +
 1 files changed, 1 insertions(+), 0 deletions(-)

drivers/usb/gadget/pch_udc.c +1 -0

index 350dbcd..6530706 100644
@@ -16,6 +16,7 @@
 #include <linux/usb/ch9.h>
 #include <linux/usb/gadget.h>
 #include <linux/gpio.h>
+#include <linux/irq.h>
 
 /* GPIO port for VBUS detecting */
 static int vbus_gpio_port = -1;		/* GPIO port number (-1:Not used) */

Signature · 1 line

1.7.7.6

Tomoya MORINAGA

2012-03-06 06:39 UTC #6 lore raw

Re: [PATCH][linux-next][RESEND] usb/gadget/pch_udc: Fix compile error

Add information.

My develop environment is on x86.

According to following report,
http://kisskb.ellerman.id.au/kisskb/buildresult/5771890/
http://kisskb.ellerman.id.au/kisskb/buildresult/5771905/

This occurred on sparc64.

I suspect this issue ins GPIO implementation for sparc64. Does sparc64 support GPIO interrupt ?
(Even x86, before 2.6.39, GPIO interrupt is not implemented !)

Additionally, this driver is for companion chip of Intel Atom E6xx series. So, this driver is never used for except x86, e.g. sparc.

thanks

Signature · 2 lines

ROHM Co., Ltd.
tomoya

Greg Kroah-Hartman

maintainer

2012-03-06 15:01 UTC #7 lore raw

Re: [PATCH][linux-next][RESEND] usb/gadget/pch_udc: Fix compile error

Quoted 10 lines · Tomoya MORINAGA

Hi
Add information.
My develop environment is on x86.
According to following report,
http://kisskb.ellerman.id.au/kisskb/buildresult/5771890/
http://kisskb.ellerman.id.au/kisskb/buildresult/5771905/

Quoted 8 lines

This occurred on sparc64.
I suspect this issue ins GPIO implementation for sparc64. Does sparc64 support GPIO interrupt ?
(Even x86, before 2.6.39, GPIO interrupt is not implemented !)
Additionally, this driver is for companion chip of Intel Atom E6xx series. So, this driver is never used for except x86, e.g. sparc.

Then the Kconfig file should probably only enable it on x86 to prevent these kinds of things :)

thanks,

greg k-h

Tomoya MORINAGA

2012-03-07 00:44 UTC #8 lore raw

Re: [PATCH][linux-next][RESEND] usb/gadget/pch_udc: Fix compile error

Quoted 5 lines · Greg Kroah-Hartman

Quoted 2 lines
Additionally, this driver is for companion chip of Intel Atom E6xx series. So, this driver is never used for except x86, e.g. sparc.
Then the Kconfig file should probably only enable it on x86 to prevent these kinds of things :)

Yes, you are right.
I should have added "x86" to Kconfig.

thanks,

Signature · 2 lines

ROHM Co., Ltd.
tomoya