From mboxrd@z Thu Jan 1 00:00:00 1970 From: Olof Johansson Date: Mon, 16 Dec 2013 05:39:50 +0000 Subject: Re: [PATCH v4 01/03] clocksource: Add Kconfig entries for CMT, MTU2, TMU and STI Message-Id: <20131216053950.GA5822@quad.lixom.net> List-Id: References: <20131211235616.12661.19702.sendpatchset@w520> <20131211235626.12661.8865.sendpatchset@w520> <20131212010503.GI21531@quad.lixom.net> In-Reply-To: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: Magnus Damm Cc: linux-kernel , Kevin Hilman , Arnd Bergmann , SH-Linux , Daniel Lezcano , "Simon Horman [Horms]" , John Stultz , Thomas Gleixner On Fri, Dec 13, 2013 at 07:12:58PM +0900, Magnus Damm wrote: > Hi Olof, > > On Thu, Dec 12, 2013 at 10:05 AM, Olof Johansson wrote: > > Hi, > > > > A couple of small comments below. > > Thanks for your feedback! > > > On Thu, Dec 12, 2013 at 08:56:26AM +0900, Magnus Damm wrote: > >> From: Magnus Damm > >> > >> Add Kconfig entries for CMT, MTU2, TMU and STI to > >> drivers/clocksource/Kconfig. This will allow us to > >> get rid of duplicated entires in architecture code > >> such as arch/sh and arch/arm/mach-shmobile. > >> > >> Signed-off-by: Magnus Damm > >> --- > >> > >> drivers/clocksource/Kconfig | 44 +++++++++++++++++++++++++++++++++++++++++++ > >> 1 file changed, 44 insertions(+) > >> > >> --- 0001/drivers/clocksource/Kconfig > >> +++ work/drivers/clocksource/Kconfig 2013-12-12 08:41:55.000000000 +0900 > >> @@ -134,3 +134,47 @@ config VF_PIT_TIMER > >> bool > >> help > >> Support for Period Interrupt Timer on Freescale Vybrid Family SoCs. > >> + > >> +config SYS_SUPPORTS_CMT > >> + bool > >> + > >> +config SYS_SUPPORTS_TMU > >> + bool > >> + > >> +config SYS_SUPPORTS_MTU2 > >> + bool > >> + > >> +config SYS_SUPPORTS_STI > >> + bool > > > > Maybe a prefix to avoid namespace collissions here? > > Sure, that is fine with me. I based the ones above on already existing > ones used by SH but I don't mind reworking those. > > How about SYS_SUPPORTS_CLKSRC_xxx? Oh, I was hoping for something shmobile-specific. Other platforms might have some other clock source that they name STI or MTU, for example. > >> +config SH_TIMER_CMT > >> + bool "Renesas CMT timer driver" if COMPILE_TEST > >> + default SYS_SUPPORTS_CMT > > > > It might be useful to have an explicit depends on ARCH_SH || ARCH_SHMOBILE > > || COMPILE_TEST on these, just to make it easier for someone reading > > the code later on. > > I have any strong feelings one way or the other about that myself, but > I got the impression that John Stultz preferred to not allow manual > selection at all, and my compromise here is to allow it only when > COMPILE_TEST is selected. > > Would you be ok to keep it as-is for now? If needed then we can add an > incremental patch later to enable more flexible selection. Yeah, if there's no manual selection at all we might be fine. -Olof