| Message ID | 201202080953.44853.hartleys@visionengravers.com |
|---|---|
| State | New |
| Headers | show |
On Wed, Feb 08, 2012 at 09:53:44AM -0700, H Hartley Sweeten wrote: > Fix build breakage due to the following commits: > > Commit bd5f12a24766c1f299def0a78b008d4746f528f2 > ARM: 7042/3: mach-ep93xx: break out GPIO driver specifics > > Commit 257af9f9725aa8a863b306659208a031135d59e7 > ARM: 7041/1: gpio-ep93xx: hookup the to_irq callback in the driver > > The vision_ep9307 machine uses the ep93xx build-in gpios and needs to > include <mach/gpio-ep93xx.h> to pickup the defines. > > The gpio_to_irq() call is now a callback to the gpio-ep93xx.c driver > and cannot be used as a constant initializer for the .irq member of > struct i2c_board_info. > > Signed-off-by: Hartley Sweeten <hsweeten@visionengravers.com> > Acked-by: Ryan Mallon <rmallon@gmail.com> Acked-by: Mika Westerberg <mika.westerberg@iki.fi>
On Thursday, February 09, 2012 11:47 PM, Mika Westerberg wrote: > On Wed, Feb 08, 2012 at 09:53:44AM -0700, H Hartley Sweeten wrote: >> Fix build breakage due to the following commits: >> >> Commit bd5f12a24766c1f299def0a78b008d4746f528f2 >> ARM: 7042/3: mach-ep93xx: break out GPIO driver specifics >> >> Commit 257af9f9725aa8a863b306659208a031135d59e7 >> ARM: 7041/1: gpio-ep93xx: hookup the to_irq callback in the driver >> >> The vision_ep9307 machine uses the ep93xx build-in gpios and needs to >> include <mach/gpio-ep93xx.h> to pickup the defines. >> >> The gpio_to_irq() call is now a callback to the gpio-ep93xx.c driver >> and cannot be used as a constant initializer for the .irq member of >> struct i2c_board_info. >> >> Signed-off-by: Hartley Sweeten <hsweeten@visionengravers.com> >> Acked-by: Ryan Mallon <rmallon@gmail.com> > > Acked-by: Mika Westerberg <mika.westerberg@iki.fi> Russell, Can this go to your Patch system or should it go thru the arm-soc tree? The 3.2 kernel will not currently build with CONFIG_MACH_VISION_EP9307=y and needs this patch. Thanks, Hartley
On Friday 10 February 2012, H Hartley Sweeten wrote: > On Thursday, February 09, 2012 11:47 PM, Mika Westerberg wrote: > > On Wed, Feb 08, 2012 at 09:53:44AM -0700, H Hartley Sweeten wrote: > >> Fix build breakage due to the following commits: > >> > >> Commit bd5f12a24766c1f299def0a78b008d4746f528f2 > >> ARM: 7042/3: mach-ep93xx: break out GPIO driver specifics > >> > >> Commit 257af9f9725aa8a863b306659208a031135d59e7 > >> ARM: 7041/1: gpio-ep93xx: hookup the to_irq callback in the driver > >> > >> The vision_ep9307 machine uses the ep93xx build-in gpios and needs to > >> include <mach/gpio-ep93xx.h> to pickup the defines. > >> > >> The gpio_to_irq() call is now a callback to the gpio-ep93xx.c driver > >> and cannot be used as a constant initializer for the .irq member of > >> struct i2c_board_info. > >> > >> Signed-off-by: Hartley Sweeten <hsweeten@visionengravers.com> > >> Acked-by: Ryan Mallon <rmallon@gmail.com> > > > > Acked-by: Mika Westerberg <mika.westerberg@iki.fi> > > Russell, > > Can this go to your Patch system or should it go thru the arm-soc tree? > > The 3.2 kernel will not currently build with CONFIG_MACH_VISION_EP9307=y > and needs this patch. Hi Hartley, Ryan and Mika, It should indeed go through the arm-soc tree, I'm sorry for having been so unresponsive recently. Olof was taking care of the other pull requests and bug fixes but unfortunately was not on Cc on this one, so we both ended up missing it until now. I've applied it on the "fixes" branch of the arm-soc tree now, it will get sent to Linus on the next opportunity. I had a little trouble applying the patch from base64 format, but since you've had to wait so long already and I'm not completely sure which side the problem was on I took the time to recreate the diff from the base64 encoding and fix up the whitespace damage I got. Arnd
On Monday 13 February 2012, Arnd Bergmann wrote: > It should indeed go through the arm-soc tree, I'm sorry for having > been so unresponsive recently. Olof was taking care of the other > pull requests and bug fixes but unfortunately was not on Cc on this > one, so we both ended up missing it until now. > > I've applied it on the "fixes" branch of the arm-soc tree now, it > will get sent to Linus on the next opportunity. > > I had a little trouble applying the patch from base64 format, but > since you've had to wait so long already and I'm not completely sure > which side the problem was on I took the time to recreate the diff > from the base64 encoding and fix up the whitespace damage I got. > One more comment: if you want to ensure a patch or pull request for arm-soc reaches us, please send it to arm@kernel.org. This should let us make sure that at least one of us reacts in a way that you deserve instead of letting it fall under the table. Arnd
On Sunday, February 12, 2012 11:58 PM, Arnd Bergmann wrote: > On Friday 10 February 2012, H Hartley Sweeten wrote: >> On Thursday, February 09, 2012 11:47 PM, Mika Westerberg wrote: >>> On Wed, Feb 08, 2012 at 09:53:44AM -0700, H Hartley Sweeten wrote: >>>> Fix build breakage due to the following commits: >>>> >>>> Commit bd5f12a24766c1f299def0a78b008d4746f528f2 >>>> ARM: 7042/3: mach-ep93xx: break out GPIO driver specifics >>>> >>>> Commit 257af9f9725aa8a863b306659208a031135d59e7 >>>> ARM: 7041/1: gpio-ep93xx: hookup the to_irq callback in the driver >>>> >>>> The vision_ep9307 machine uses the ep93xx build-in gpios and needs to >>>> include <mach/gpio-ep93xx.h> to pickup the defines. >>>> >>>> The gpio_to_irq() call is now a callback to the gpio-ep93xx.c driver >>>> and cannot be used as a constant initializer for the .irq member of >>>> struct i2c_board_info. >>>> >>>> Signed-off-by: Hartley Sweeten <hsweeten@visionengravers.com> >>>> Acked-by: Ryan Mallon <rmallon@gmail.com> >>> >>> Acked-by: Mika Westerberg <mika.westerberg@iki.fi> >> >> Russell, >> >> Can this go to your Patch system or should it go thru the arm-soc tree? >> >> The 3.2 kernel will not currently build with CONFIG_MACH_VISION_EP9307=y >> and needs this patch. > > Hi Hartley, Ryan and Mika, > > It should indeed go through the arm-soc tree, I'm sorry for having > been so unresponsive recently. Olof was taking care of the other > pull requests and bug fixes but unfortunately was not on Cc on this > one, so we both ended up missing it until now. > > I've applied it on the "fixes" branch of the arm-soc tree now, it > will get sent to Linus on the next opportunity. Thanks! >I had a little trouble applying the patch from base64 format, but >since you've had to wait so long already and I'm not completely sure >which side the problem was on I took the time to recreate the diff >from the base64 encoding and fix up the whitespace damage I got. Hmm... I'm not sure what caused the base64 encoding. I recently updated my host machine but I don't think anything I did would have caused the mail to be encoded differently. I'll look into it and try to find the problem. Thanks, Hartley
diff --git a/arch/arm/mach-ep93xx/vision_ep9307.c b/arch/arm/mach-ep93xx/vision_ep9307.c index 03dd401..d5fb44f 100644 --- a/arch/arm/mach-ep93xx/vision_ep9307.c +++ b/arch/arm/mach-ep93xx/vision_ep9307.c @@ -32,6 +32,7 @@ #include <mach/hardware.h> #include <mach/fb.h> #include <mach/ep93xx_spi.h> +#include <mach/gpio-ep93xx.h> #include <asm/mach-types.h> #include <asm/mach/map.h> @@ -153,7 +154,6 @@ static struct i2c_board_info vision_i2c_info[] __initdata = { }, { I2C_BOARD_INFO("pca9539", 0x74), .platform_data = &pca953x_74_gpio_data, - .irq = gpio_to_irq(EP93XX_GPIO_LINE_F(7)), }, { I2C_BOARD_INFO("pca9539", 0x75), .platform_data = &pca953x_75_gpio_data, @@ -348,6 +348,8 @@ static void __init vision_init_machine(void) "pca9539:74")) pr_warn("cannot request interrupt gpio for pca9539:74\n"); + vision_i2c_info[1].irq = gpio_to_irq(EP93XX_GPIO_LINE_F(7)); + ep93xx_register_i2c(&vision_i2c_gpio_data, vision_i2c_info, ARRAY_SIZE(vision_i2c_info)); ep93xx_register_spi(&vision_spi_master, vision_spi_board_info,