From mboxrd@z Thu Jan 1 00:00:00 1970 From: Olof Johansson Subject: Re: [PATCH 2/4] USB: MUSB: Enable MUSB for OMAP4 in Kconfig Date: Mon, 1 Mar 2010 09:43:29 -0600 Message-ID: <20100301154329.GA4154@lixom.net> References: <1267079758-14480-1-git-send-email-x0082077@ti.com> <20100225184833.GG28173@atomide.com> <20100226004233.GA3241@lixom.net> <000001cab6bf$5df75720$808918ac@apr.dhcp.ti.com> <20100226185414.GA10835@lixom.net> <001b01cab901$dadec9a0$808918ac@apr.dhcp.ti.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: <001b01cab901$dadec9a0$808918ac-wD+IZp/g4/2mHdYHvhjUOg@public.gmane.org> Sender: linux-usb-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: Maulik Cc: 'Tony Lindgren' , linux-usb-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, linux-omap-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, 'Felipe Balbi' , 'Greg Kroah-Hartman' , 'David Brownell' List-Id: linux-omap@vger.kernel.org On Mon, Mar 01, 2010 at 11:11:29AM +0530, Maulik wrote: > > Olof, > > > > We have such an option "USB_MUSB_SOC" that takes care of all platforms > where > > MUSB is present. But the Inventra DMA case is typical. Not all > architectures > > that have MUSB use Inventra DMA for e.g Davinci uses CPPI DMA. > > >>Yeah, sorry. I didn't mean the INVENTRA_DMA config option as much as the > >>top > >>MUSB one. So you should be able to switch to that instead of adding a > >>check for > >>OMAP4 there then? > > Olof, > So you suggest adding something like below for OMAP4 and all others as well > right? > > comment "OMAP 44xx high speed USB support" > depends on USB_MUSB_HDRC && USB_MUSB_SOC Hmm. I must be missing something, but it looks odd to have both the USB_MUSB_HDRC and USB_MUSB_SOC high-level config options, especially since the depends are duplicated across them. In general, options like these tend to scale badly: config USB_MUSB_SOC boolean depends on USB_MUSB_HDRC default y if ARCH_DAVINCI default y if ARCH_OMAP2430 default y if ARCH_OMAP34XX default y if (BF54x && !BF544) default y if (BF52x && !BF522 && !BF523) I.e. every single platform needs to add a new line. It's better to have a config option that is selected from the platform code (arch/arm/mach-omap2/Kconfig for the specific SOC in this case), that for example does a : config ARCH_OMAP34XX bool "OMAP34xx Based System" depends on ARCH_OMAP3 select USB_ARCH_HAS_EHCI select USB_ARCH_HAS_MUSB and then just do the USB_MUSB_HDRC or USB_MUSB_SOC config option based on: config USB_MUSB_SOC boolena depends on USB_ARCH_HAS_MUSB default y That way a new platform doesn't need to change the common files. I also don't see the direct need to provide a specific comment saying "44xx high speed USB support" when you configure for 44xx. With the work Tony is doing you can build one kernel that boots across OMAP2/3/4, which comment should then be displayed? It's better to just do a generic comment such as "OMAP on-chip OTG USB support" The DMA config is also a bit confusing. I scratched my head over that one a few times, the fact that one is described with OMAP in the name, and the other one (INVENTRA) is the one actually used on newer OMAP platforms. -Olof -- To unsubscribe from this list: send the line "unsubscribe linux-usb" in the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org More majordomo info at http://vger.kernel.org/majordomo-info.html