ARM: multi_v7_defconfig: Replace CONFIG_USB_ISP1760_HCD by CONFIG_USB_ISP1760

Make USB support modular to reduce kernel size:

       text    data     bss     dec     hex   filename
    12979266 1433908  335640 14748814  e10c8e vmlinux.before
    12237024 1386396  333592 13957012  d4f794 vmlinux.after

USB_GSPCA, USB_EHCI_HCD_PLATFORM, USB_OHCI_HCD_PLATFORM, and PHY_SAMSUNG_USB2 are now enabled automatically by Kconfig logic, so they are dropped.

Tested on r8a7791/koelsch, which can't boot vmlinux.before anymore.

Signed-off-by Geert Uytterhoeven <[email protected]>

---
Against next-20150526

---
 arch/arm/configs/multi_v7_defconfig | 40  ++++++++++++++++-----------------
 1 file changed, 18 insertions(+), 22 deletions(-)

(quoted diff) +0 -0

b/arch/arm/configs/multi_v7_defconfig index
e3fc9799972367c3..5df66b3671946378 100644

arch/arm/configs/multi_v7_defconfig +18 -22

@@ -215,10 +215,10 @@ CONFIG_SMSC_PHY=y
 CONFIG_BROADCOM_PHY=y
 CONFIG_ICPLUS_PHY=y
 CONFIG_MICREL_PHY=y
-CONFIG_USB_PEGASUS=y
-CONFIG_USB_USBNET=y
-CONFIG_USB_NET_SMSC75XX=y
-CONFIG_USB_NET_SMSC95XX=y
+CONFIG_USB_PEGASUS=m
+CONFIG_USB_USBNET=m
+CONFIG_USB_NET_SMSC75XX=m
+CONFIG_USB_NET_SMSC95XX=m
 CONFIG_BRCMFMAC=m
 CONFIG_RT2X00=m
 CONFIG_RT2800USB=m

@@ -417,8 +417,7 @@ CONFIG_MEDIA_CAMERA_SUPPORT=y
 CONFIG_MEDIA_CONTROLLER=y
 CONFIG_VIDEO_V4L2_SUBDEV_API=y
 CONFIG_MEDIA_USB_SUPPORT=y
-CONFIG_USB_VIDEO_CLASS=y
-CONFIG_USB_GSPCA=y
+CONFIG_USB_VIDEO_CLASS=m
 CONFIG_V4L_PLATFORM_DRIVERS=y
 CONFIG_SOC_CAMERA=m
 CONFIG_SOC_CAMERA_PLATFORM=m

@@ -452,7 +451,7 @@ CONFIG_FRAMEBUFFER_CONSOLE_ROTATION=y
 CONFIG_SOUND=y
 CONFIG_SND=y
 CONFIG_SND_DYNAMIC_MINORS=y
-CONFIG_SND_USB_AUDIO=y
+CONFIG_SND_USB_AUDIO=m
 CONFIG_SND_SOC=y
 CONFIG_SND_ATMEL_SOC=m
 CONFIG_SND_ATMEL_SOC_WM8904=m

@@ -467,24 +466,22 @@ CONFIG_SND_SOC_TEGRA_ALC5632=y
 CONFIG_SND_SOC_TEGRA_MAX98090=y
 CONFIG_SND_SOC_AK4642=m
 CONFIG_SND_SOC_WM8978=m
-CONFIG_USB=y
-CONFIG_USB_XHCI_HCD=y
-CONFIG_USB_XHCI_MVEBU=y
-CONFIG_USB_EHCI_HCD=y
-CONFIG_USB_EHCI_EXYNOS=y
-CONFIG_USB_EHCI_TEGRA=y
-CONFIG_USB_EHCI_HCD_STI=y
-CONFIG_USB_EHCI_HCD_PLATFORM=y
-CONFIG_USB_ISP1760=y
-CONFIG_USB_OHCI_HCD=y
-CONFIG_USB_OHCI_HCD_STI=y
-CONFIG_USB_OHCI_HCD_PLATFORM=y
+CONFIG_USB=m
+CONFIG_USB_XHCI_HCD=m
+CONFIG_USB_XHCI_MVEBU=m
+CONFIG_USB_EHCI_HCD=m
+CONFIG_USB_EHCI_HCD_STI=m
+CONFIG_USB_EHCI_TEGRA=m
+CONFIG_USB_EHCI_EXYNOS=m
+CONFIG_USB_ISP1760=m
+CONFIG_USB_OHCI_HCD=m
+CONFIG_USB_OHCI_HCD_STI=m
 CONFIG_USB_OHCI_EXYNOS=m
 CONFIG_USB_R8A66597_HCD=m
 CONFIG_USB_RENESAS_USBHS=m
-CONFIG_USB_STORAGE=y
+CONFIG_USB_STORAGE=m
 CONFIG_USB_DWC3=y
-CONFIG_USB_CHIPIDEA=y
+CONFIG_USB_CHIPIDEA=m
 CONFIG_USB_CHIPIDEA_HOST=y
 CONFIG_AB8500_USB=y
 CONFIG_KEYSTONE_USB_PHY=y

@@ -632,7 +629,6 @@ CONFIG_PHY_STIH41X_USB=y
 CONFIG_PHY_STIH407_USB=y
 CONFIG_PHY_SUN4I_USB=y
 CONFIG_PHY_SUN9I_USB=y
-CONFIG_PHY_SAMSUNG_USB2=m
 CONFIG_EXT4_FS=y
 CONFIG_AUTOFS4_FS=y
 CONFIG_MSDOS_FS=y

Quoted 1 line · On Tuesday 26 May 2015 12:58:38 Geert Uytterhoeven wrote:

Make USB support modular to reduce kernel size:

This means we lose support for NFS root on USB-based ethernet controllers. I'm not sure that's a very good idea.

P.S. Strange, USB_NET_DRIVERS is tristate, but setting it to "y" doesn't
mean all USB network drivers become builtin, as USB_NET_DRIVERS is never used in a "depends" construct...

Since commit 100832abf065bc18 ("usb: isp1760: Make HCD support optional"), CONFIG_USB_ISP1760_HCD is automatically selected when needed. Enabling that option in the defconfig is now a no-op, and no longer enables ISP1760 HCD support.

Re-enable the ISP1760 driver in the defconfig by enabling USB_ISP1760_HOST_ROLE instead.

Signed-off-by Geert Uytterhoeven <[email protected]>

Quoted 8 lines · Geert Uytterhoeven

Since commit 100832abf065bc18 ("usb: isp1760: Make HCD support optional"), CONFIG_USB_ISP1760_HCD is automatically selected when needed. Enabling that option in the defconfig is now a no-op, and no longer enables ISP1760 HCD support.

Re-enable the ISP1760 driver in the defconfig by enabling USB_ISP1760_HOST_ROLE instead.

Quoted 1 line

Signed-off-by Geert Uytterhoeven <[email protected]>

Thanks, I have queued this up for v4.3.

I would not object at all if the ARM SoC maintainers wish to take this earlier and I'd be happy to prepare a pull request to that effect once the change has sat in next. Unfortunately I anticipate that will be after the standard rc6 deadline.

Quoted 13 lines · On Thursday 28 May 2015 09:39:08 Simon Horman wrote:

Quoted 9 lines · Geert Uytterhoeven

Since commit 100832abf065bc18 ("usb: isp1760: Make HCD support optional"), CONFIG_USB_ISP1760_HCD is automatically selected when needed. Enabling that option in the defconfig is now a no-op, and no longer enables ISP1760 HCD support.

Re-enable the ISP1760 driver in the defconfig by enabling USB_ISP1760_HOST_ROLE instead.

Signed-off-by Geert Uytterhoeven <[email protected]>

Thanks, I have queued this up for v4.3.

Quoted 4 lines

I would not object at all if the ARM SoC maintainers wish to take this earlier and I'd be happy to prepare a pull request to that effect once the change has sat in next. Unfortunately I anticipate that will be after the standard rc6 deadline.

I would consider patch 1 as a bug-fix, and put that into 4.1 or 4.2, depending on timing. Patch 2 looks like 4.3 material.

Quoted 19 lines · Arnd Bergmann

Quoted 17 lines · On Thursday 28 May 2015 09:39:08 Simon Horman wrote:

Quoted 9 lines · Geert Uytterhoeven

Since commit 100832abf065bc18 ("usb: isp1760: Make HCD support optional"), CONFIG_USB_ISP1760_HCD is automatically selected when needed. Enabling that option in the defconfig is now a no-op, and no longer enables ISP1760 HCD support.

Re-enable the ISP1760 driver in the defconfig by enabling USB_ISP1760_HOST_ROLE instead.

Signed-off-by Geert Uytterhoeven <[email protected]>

Thanks, I have queued this up for v4.3.

I would not object at all if the ARM SoC maintainers wish to take this earlier and I'd be happy to prepare a pull request to that effect once the change has sat in next. Unfortunately I anticipate that will be after the standard rc6 deadline.

Quoted 2 lines

I would consider patch 1 as a bug-fix, and put that into 4.1 or 4.2, depending on timing. Patch 2 looks like 4.3 material.

Thanks Arnd, would you care to take pick it up directly for 4.1 (or 4.2)? If so I'll drop it from my tree.