Skip to contentSkip to conversation

3.20-rc0 breaks nfsroot over usb

PATCH Superseded Pavel Machek 5 participants 9 messages Feb 17, 2015 – Feb 27, 2015

Subsystems (2)
USB MUSB multipoint high speed dual-role controller
lore
Share & tools
lore

Pavel Machek

author
#0 lore raw

Hi!

It no longer works after I merged

commit 796e1c55717e9a6ff5c81b12289ffa1ffd919b6f
Merge: 8c334ce 45ee2db
Author: Linus Torvalds <[email protected]>
Date: Mon Feb 16 15:48:00 2015 -0800

Merge branch 'drm-next' of
git://people.freedesktop.org/~airlied/linux

Pull drm updates from Dave Airlie:
"This is the main drm pull, it has a shared branch with some alsa
crossover but everything should be acked by relevant people.

If it works for you or if you have an idea, let me know...
Pavel

Signature · 2 lines
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html
author
#1 lore raw

Hi!

Quoted 10 lines

It no longer works after I merged

commit 796e1c55717e9a6ff5c81b12289ffa1ffd919b6f
Merge: 8c334ce 45ee2db
Author: Linus Torvalds <[email protected]>
Date: Mon Feb 16 15:48:00 2015 -0800

Merge branch 'drm-next' of
git://people.freedesktop.org/~airlied/linux

Quoted 7 lines

Pull drm updates from Dave Airlie:
"This is the main drm pull, it has a shared branch with some alsa
crossover but everything should be acked by relevant people.

If it works for you or if you have an idea, let me know...

Ok, it seems that new config dependencies broke my kconfig. These will not be enabled if you just to make oldconfig; you want to make sure they are on.

OMAP2430 and onwards (USB_MUSB_OMAP2PLUS) [N/m/y] (NEW)
TWL4030 USB Transceiver Driver (TWL4030_USB) [N/m/y/?] (NEW)
Pavel

Signature · 2 lines
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html
#2 lore raw

Hi,

Quoted 6 lines · Pavel Machek

Ok, it seems that new config dependencies broke my kconfig. These will not be enabled if you just to make oldconfig; you want to make sure they are on.

OMAP2430 and onwards (USB_MUSB_OMAP2PLUS) [N/m/y] (NEW)
TWL4030 USB Transceiver Driver (TWL4030_USB) [N/m/y/?] (NEW)

Yes, also other users have reported "USB breakage" with 4.0-rc1 on N900 etc caused by this.

Commit fbba7db3990cb707ff91cd6507d53a0a730afe97 (usb: musb:
add omap-control dependency) changed USB_MUSB_OMAP2PLUS to depend on OMAP_CONTROL_PHY but that seems to be wrong at least for OMAP3. omap_control_phy.h provides dummy functions when it's not enabled. So there should be no dependency.

A.

#3 lore raw
Quoted 10 lines · Aaro Koskinen

Hi,

Quoted 6 lines · Pavel Machek

Ok, it seems that new config dependencies broke my kconfig. These will not be enabled if you just to make oldconfig; you want to make sure they are on.

OMAP2430 and onwards (USB_MUSB_OMAP2PLUS) [N/m/y] (NEW)
TWL4030 USB Transceiver Driver (TWL4030_USB) [N/m/y/?] (NEW)

Quoted 8 lines

Yes, also other users have reported "USB breakage" with 4.0-rc1 on N900 etc caused by this.

Commit fbba7db3990cb707ff91cd6507d53a0a730afe97 (usb: musb:
add omap-control dependency) changed USB_MUSB_OMAP2PLUS to depend on OMAP_CONTROL_PHY but that seems to be wrong at least for OMAP3. omap_control_phy.h provides dummy functions when it's not enabled. So there should be no dependency.

I guess the solution would be to make drivers/phy handle whatever omap_control_usb_set_mode is doing in a generic way.

Regards,

Tony

#4 lore raw
Quoted 11 lines · On Friday 27 February 2015 07:38:09 Tony Lindgren wrote:
Quoted 10 lines · Aaro Koskinen

Hi,

Quoted 6 lines · Pavel Machek

Ok, it seems that new config dependencies broke my kconfig. These will not be enabled if you just to make oldconfig; you want to make sure they are on.

OMAP2430 and onwards (USB_MUSB_OMAP2PLUS) [N/m/y] (NEW)
TWL4030 USB Transceiver Driver (TWL4030_USB) [N/m/y/?] (NEW)

Quoted 8 lines
Quoted 8 lines

Yes, also other users have reported "USB breakage" with 4.0-rc1 on N900 etc caused by this.

Commit fbba7db3990cb707ff91cd6507d53a0a730afe97 (usb: musb:
add omap-control dependency) changed USB_MUSB_OMAP2PLUS to depend on OMAP_CONTROL_PHY but that seems to be wrong at least for OMAP3. omap_control_phy.h provides dummy functions when it's not enabled. So there should be no dependency.

The problem I fixed was a link error when USB_MUSB_OMAP2PLUS is built-in but OMAP_CONTROL_PHY is a loadable module.

Quoted 2 lines

I guess the solution would be to make drivers/phy handle whatever omap_control_usb_set_mode is doing in a generic way.

I'm very sorry about the breakage in a patch that is marked for backports.

Please accept my apology and the workaround patch in the follow-up mail.

Arnd

#5 lore raw

PATCH 1/1 usb: musb: fix Kconfig regression

In reply to Tony Lindgren

A recent bug fix I did that was marked for stable backports introduced a slightly wrong dependency on CONFIG_OMAP_CONTROL_PHY.

I was missing the fact that the PHY driver already stubs out the omap_control_usb_set_mode, and we only need to add a dependency to prevent the musb-omap2430 driver from being built-in when the phy driver is a loadable module, but we should not prevent it from being built altogether when the phy driver is disabled.

This changes the dependency to allow OMAP_CONTROL_PHY to be disabled, but to disallow USB_MUSB_OMAP2PLUS from being built-in if OMAP_CONTROL_PHY is a loadable module.

Signed-off-by Arnd Bergmann <[email protected]>
Fixes ca784be36cc725 ("usb: start using the control module driver")
Cc <[email protected]> # v3.9+
Cc Felipe Balbi <[email protected]>
drivers/usb/musb/Kconfig +2 -1
index 14e1628483d9..39db8b603627 100644
@@ -79,7 +79,8 @@ config USB_MUSB_TUSB6010
 
 config USB_MUSB_OMAP2PLUS
 	tristate "OMAP2430 and onwards"
-	depends on ARCH_OMAP2PLUS && USB && OMAP_CONTROL_PHY
+	depends on ARCH_OMAP2PLUS && USB
+	depends on OMAP_CONTROL_PHY || !OMAP_CONTROL_PHY
 	select GENERIC_PHY
 
 config USB_MUSB_AM35X
#6 lore raw

Re: [PATCH] usb: musb: fix Kconfig regression

Quoted 9 lines · Arnd Bergmann

A recent bug fix I did that was marked for stable backports introduced a slightly wrong dependency on CONFIG_OMAP_CONTROL_PHY.

I was missing the fact that the PHY driver already stubs out the omap_control_usb_set_mode, and we only need to add a dependency to prevent the musb-omap2430 driver from being built-in when the phy driver is a loadable module, but we should not prevent it from being built altogether when the phy driver is disabled.

Quoted 8 lines

This changes the dependency to allow OMAP_CONTROL_PHY to be disabled, but to disallow USB_MUSB_OMAP2PLUS from being built-in if OMAP_CONTROL_PHY is a loadable module.

Signed-off-by Arnd Bergmann <[email protected]>
Fixes ca784be36cc725 ("usb: start using the control module driver")
Cc <[email protected]> # v3.9+
Cc Felipe Balbi <[email protected]>

please resend with linux-usb in Cc and myself in Cc too ;-)

Signature · 8 lines
balbi
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: Digital signature
URL: <http://lists.infradead.org/pipermail/linux-arm-kernel/attachments/20150227/f2421702/attachment.sig>
author Acked-by on v1 1/1
#7 lore raw

Re: [PATCH] usb: musb: fix Kconfig regression

In reply to Arnd Bergmann
Quoted 9 lines · Arnd Bergmann

A recent bug fix I did that was marked for stable backports introduced a slightly wrong dependency on CONFIG_OMAP_CONTROL_PHY.

I was missing the fact that the PHY driver already stubs out the omap_control_usb_set_mode, and we only need to add a dependency to prevent the musb-omap2430 driver from being built-in when the phy driver is a loadable module, but we should not prevent it from being built altogether when the phy driver is disabled.

Quoted 8 lines

This changes the dependency to allow OMAP_CONTROL_PHY to be disabled, but to disallow USB_MUSB_OMAP2PLUS from being built-in if OMAP_CONTROL_PHY is a loadable module.

Signed-off-by Arnd Bergmann <[email protected]>
Fixes ca784be36cc725 ("usb: start using the control module driver")
Cc <[email protected]> # v3.9+
Cc Felipe Balbi <[email protected]>
Acked-by Pavel Machek <[email protected]>
Signature · 2 lines
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html
author
#8 lore raw
In reply to Aaro Koskinen
Quoted 7 lines · Aaro Koskinen

Hi,

Quoted 4 lines · Pavel Machek

Ok, it seems that new config dependencies broke my kconfig. These will not be enabled if you just to make oldconfig; you want to make sure they are on.

Quoted 8 lines
Quoted 2 lines

OMAP2430 and onwards (USB_MUSB_OMAP2PLUS) [N/m/y] (NEW)
TWL4030 USB Transceiver Driver (TWL4030_USB) [N/m/y/?] (NEW)

Yes, also other users have reported "USB breakage" with 4.0-rc1 on N900 etc caused by this.

Commit fbba7db3990cb707ff91cd6507d53a0a730afe97 (usb: musb:
add omap-control dependency) changed USB_MUSB_OMAP2PLUS to depend

Too bad noone told me. I had a lot of fun trying to bisect config problem. I have yet to try the nfsroot, but I guess it should be ok now.

Thanks,

Pavel

Signature · 2 lines
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html