| Message ID | 20130924131206.d57d07f5faa5cb0773fc328c@ops.dti.ne.jp (mailing list archive) |
|---|---|
| State | Changes Requested |
| Headers | show |
Hi Yoshii-san, Thank you for the patch. On Tuesday 24 September 2013 13:12:06 takasi-y@ops.dti.ne.jp wrote: > Make sh clock framework core depend on HAVE_MACH_CLKDEV, and > set it > - y on sh for backward compatibility > - !CONFIG_COMMON_CLK on sh-mobile > This is a preparation for migration to common clock framework > from sh clock framework on sh-mobile. While I agree with this patch, I believe the use of the HAVE_MACH_CLKDEV configuration option to select whether to compile core.o in is a bit of an abuse. I would have created a new configuration option (SH_CLK_CORE in drivers/sh/Kconfig for instance) as 'def_bool y' that depends on SH || (ARM && !COMMON_CLK). However, as all ARCH_SHMOBILE platforms should be converted to the common clock framework, this is only temporary and could be revisited later, so I'm fine with keeping the patch as-is. > Signed-off-by: Takashi Yoshii <takashi.yoshii.zj@renesas.com> > --- > arch/arm/Kconfig | 2 +- > arch/sh/Kconfig | 1 + > drivers/sh/clk/Makefile | 3 +-- > 3 files changed, 3 insertions(+), 3 deletions(-) > > diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig > index 3f7714d..53044ca 100644 > --- a/arch/arm/Kconfig > +++ b/arch/arm/Kconfig > @@ -650,7 +650,7 @@ config ARCH_SHMOBILE > select HAVE_ARM_SCU if SMP > select HAVE_ARM_TWD if SMP > select HAVE_CLK > - select HAVE_MACH_CLKDEV > + select HAVE_MACH_CLKDEV if !COMMON_CLK > select HAVE_SMP > select MIGHT_HAVE_CACHE_L2X0 > select MULTI_IRQ_HANDLER > diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig > index 224f4bc..f57e47f 100644 > --- a/arch/sh/Kconfig > +++ b/arch/sh/Kconfig > @@ -41,6 +41,7 @@ config SUPERH > select MODULES_USE_ELF_RELA > select OLD_SIGSUSPEND > select OLD_SIGACTION > + select HAVE_MACH_CLKDEV > help > The SuperH is a RISC processor targeted for use in embedded systems > and consumer electronics; it was also used in the Sega Dreamcast > diff --git a/drivers/sh/clk/Makefile b/drivers/sh/clk/Makefile > index 5d15ebf..e73b031 100644 > --- a/drivers/sh/clk/Makefile > +++ b/drivers/sh/clk/Makefile > @@ -1,3 +1,2 @@ > -obj-y := core.o > - > +obj-$(CONFIG_HAVE_MACH_CLKDEV) += core.o > obj-$(CONFIG_SH_CLK_CPG) += cpg.o
On Tue, Sep 24, 2013 at 1:12 PM, <takasi-y@ops.dti.ne.jp> wrote: > Make sh clock framework core depend on HAVE_MACH_CLKDEV, and > set it > - y on sh for backward compatibility > - !CONFIG_COMMON_CLK on sh-mobile > This is a preparation for migration to common clock framework > from sh clock framework on sh-mobile. > > Signed-off-by: Takashi Yoshii <takashi.yoshii.zj@renesas.com> > --- > arch/arm/Kconfig | 2 +- > arch/sh/Kconfig | 1 + > drivers/sh/clk/Makefile | 3 +-- > 3 files changed, 3 insertions(+), 3 deletions(-) Hi Yoshii-san, Thanks for your patch. I'm sure there is a reason behind this, but I'm trying to understand why you need this modification. It looks to me like you're trying to enable COMMON_CLK on ARCH_SHMOBILE but in my mind it is enough to only enable COMMON_CLK in the case of ARCH_SHMOBILE_MULTI. During my test using ARCH_SHMOBILE_MULTI on KZM9D I omitted this patch and only used patch 1, 2, 4, 5, and a modified 6 from your series. This worked just fine, but I may be missing something. Let me know if you really want to keep this patch or not. If it's not needed for MULTIPLATFORM then I suggest that we just drop it. Cheers, / magnus -- To unsubscribe from this list: send the line "unsubscribe linux-sh" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Hi Laurent, > While I agree with this patch, I believe the use of the HAVE_MACH_CLKDEV > configuration option to select whether to compile core.o in is a bit of an > abuse. <snip> Well, yes, indeed. It does not mean having include/mach/clkdev.h, but means somewhat such like !COMMON_CLK. # I think there is similar usage in arch/mips, though they define but not use. # ... is an execution. > However, as all ARCH_SHMOBILE platforms should be converted to the common > clock framework, this is only temporary and could be revisited later, so I'm > fine with keeping the patch as-is. Sure. Thank you for your review. /yoshii -- To unsubscribe from this list: send the line "unsubscribe linux-sh" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig index 3f7714d..53044ca 100644 --- a/arch/arm/Kconfig +++ b/arch/arm/Kconfig @@ -650,7 +650,7 @@ config ARCH_SHMOBILE select HAVE_ARM_SCU if SMP select HAVE_ARM_TWD if SMP select HAVE_CLK - select HAVE_MACH_CLKDEV + select HAVE_MACH_CLKDEV if !COMMON_CLK select HAVE_SMP select MIGHT_HAVE_CACHE_L2X0 select MULTI_IRQ_HANDLER diff --git a/arch/sh/Kconfig b/arch/sh/Kconfig index 224f4bc..f57e47f 100644 --- a/arch/sh/Kconfig +++ b/arch/sh/Kconfig @@ -41,6 +41,7 @@ config SUPERH select MODULES_USE_ELF_RELA select OLD_SIGSUSPEND select OLD_SIGACTION + select HAVE_MACH_CLKDEV help The SuperH is a RISC processor targeted for use in embedded systems and consumer electronics; it was also used in the Sega Dreamcast diff --git a/drivers/sh/clk/Makefile b/drivers/sh/clk/Makefile index 5d15ebf..e73b031 100644 --- a/drivers/sh/clk/Makefile +++ b/drivers/sh/clk/Makefile @@ -1,3 +1,2 @@ -obj-y := core.o - +obj-$(CONFIG_HAVE_MACH_CLKDEV) += core.o obj-$(CONFIG_SH_CLK_CPG) += cpg.o
Make sh clock framework core depend on HAVE_MACH_CLKDEV, and set it - y on sh for backward compatibility - !CONFIG_COMMON_CLK on sh-mobile This is a preparation for migration to common clock framework from sh clock framework on sh-mobile. Signed-off-by: Takashi Yoshii <takashi.yoshii.zj@renesas.com> --- arch/arm/Kconfig | 2 +- arch/sh/Kconfig | 1 + drivers/sh/clk/Makefile | 3 +-- 3 files changed, 3 insertions(+), 3 deletions(-)