Hi,
On Thu, Dec 16, 2010 at 06:05:02PM +0800, Ming Lei wrote:
Seems the option(USB_MUSB_OMAP2PLUS) shouldn't declared as bool
since it may depend on one 'm' option(USB_MUSB_HDRC), so omap2430.o
is not linked into vmlinux.
Then the problem will return to the topic discussed in musb-cleanup
thread this days:
- driving multiple musb hw controllers using one same set of musb
binary drivers
I think it makes sense, for example mach-omap2 may support tusb, am35x
and omap2430, so it is valuable for os-distribution(such as Ubuntu/angstrom/...)
that one same omap2 image can support all the musb controllers for different
omap2 boards.
If we agree on the feature above makes sense, building glue layer
driver as module is to be supported for memory footprint reason.
that's actually what that huge patchset I wrote is starting to do. Only
not possible due to the exported symbols.
Felipe and all, if you don't have objections, I will go ahead to work on
musb-cleanup v1. Or any suggestions?
Seems fine to me. So you would do something like below ?
diff --git a/drivers/usb/musb/Kconfig b/drivers/usb/musb/Kconfig
index 4cbb7e4..cf8d85c 100644
--- a/drivers/usb/musb/Kconfig
+++ b/drivers/usb/musb/Kconfig
@@ -38,32 +38,39 @@ choice
depends on USB_MUSB_HDRC
config USB_MUSB_DAVINCI
- bool "DaVinci"
+ tristate "DaVinci"
depends on ARCH_DAVINCI_DMx
+ depends on USB_MUSB_HDRC
config USB_MUSB_DA8XX
- bool "DA8xx/OMAP-L1x"
+ tristate "DA8xx/OMAP-L1x"
depends on ARCH_DAVINCI_DA8XX
+ depends on USB_MUSB_HDRC
config USB_MUSB_TUSB6010
- bool "TUSB6010"
+ tristate "TUSB6010"
depends on ARCH_OMAP
+ depends on USB_MUSB_HDRC
config USB_MUSB_OMAP2PLUS
- bool "OMAP2430 and onwards"
+ tristate "OMAP2430 and onwards"
depends on ARCH_OMAP2PLUS
+ depends on USB_MUSB_HDRC
config USB_MUSB_AM35X
- bool "AM35x"
+ tristate "AM35x"
depends on ARCH_OMAP
+ depends on USB_MUSB_HDRC
config USB_MUSB_BLACKFIN
- bool "Blackfin"
+ tristate "Blackfin"
depends on (BF54x && !BF544) || (BF52x && ! BF522 && !BF523)
+ depends on USB_MUSB_HDRC
config USB_MUSB_UX500
- bool "U8500 and U5500"
+ tristate "U8500 and U5500"
depends on (ARCH_U8500 && AB8500_USB) || (ARCH_U5500)
+ depends on USB_MUSB_HDRC
endchoice
but that series can't go now, at least not entirely. It's too much new
stuff for -rc I think. Anyway, send the patches and let's work on them
:-)
--
balbi
--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
