Lost fixes?

Lost fixes?

[Russell King - ARM Linux]

These fixes are a collection from other people which resolve various
issues found with the new kautobuilder.  They're now a month old, and
as far as I can see, have not been merged into Linus' tree.

What's the status of them, and when are they going to be merged?

Thanks.

 arch/arm/mach-omap2/Makefile                 |    5 +----
 arch/arm/mach-omap2/clkt2xxx_virt_prcm_set.c |    2 +-
 arch/arm/mach-omap2/clock.c                  |    2 +-
 arch/arm/mach-omap2/clock.h                  |    2 +-
 arch/arm/mach-omap2/smartreflex.c            |    2 +-
 arch/arm/plat-omap/clock.c                   |    2 +-
 arch/arm/plat-omap/dma.c                     |    2 +-
 arch/arm/plat-omap/include/plat/clock.h      |    4 ++--
 drivers/cpufreq/Kconfig.arm                  |    6 ++++++
 drivers/cpufreq/Makefile                     |    2 +-
 drivers/mfd/da9052-spi.c                     |    4 ++--
 kernel/irq/chip.c                            |    3 +--
 12 files changed, 19 insertions(+), 17 deletions(-)

[PATCH 2/6] CPUFREQ: Fix exposure of ARM_EXYNOS4210_CPUFREQ

[Russell King]

exynos4210-cpufreq.c is not buildable on non-exynos builds, so it's
pointless allowing this option to be exposed.  Fix this by adding a
dependency on ARCH_EXYNOS.

drivers/cpufreq/exynos4210-cpufreq.c:20:29: error: mach/regs-clock.h: No such file or directory
drivers/cpufreq/exynos4210-cpufreq.c:21:26: error: mach/cpufreq.h: No such file or directory

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 drivers/cpufreq/Kconfig.arm |    1 +
 1 files changed, 1 insertions(+), 0 deletions(-)

diff --git a/drivers/cpufreq/Kconfig.arm b/drivers/cpufreq/Kconfig.arm
index e0664fe..c8bde43 100644
--- a/drivers/cpufreq/Kconfig.arm
+++ b/drivers/cpufreq/Kconfig.arm
@@ -34,6 +34,7 @@ config ARM_EXYNOS_CPUFREQ
 
 config ARM_EXYNOS4210_CPUFREQ
 	bool "Samsung EXYNOS4210"
+	depends on ARCH_EXYNOS
 	help
 	  This adds the CPUFreq driver for Samsung EXYNOS4210
 	  SoC (S5PV310 or S5PC210).
-- 
1.7.4.4

[PATCH 6/6] ARM: OMAP: fix cpufreq build

[Russell King]

OMAPs cpufreq requires the frequency table support, but nothing ensures
that this is selected.  This can result in configurations which fail to
build:

drivers/built-in.o:(.data+0x5238): undefined reference to `cpufreq_freq_attr_scaling_available_freqs'
drivers/cpufreq/omap-cpufreq.c:88: undefined reference to `cpufreq_frequency_table_target'
drivers/cpufreq/omap-cpufreq.c:60: undefined reference to `cpufreq_frequency_table_verify'
drivers/cpufreq/omap-cpufreq.c:186: undefined reference to `cpufreq_frequency_table_cpuinfo'
drivers/cpufreq/omap-cpufreq.c:190: undefined reference to `cpufreq_frequency_table_get_attr'

Fix this by introducing a new configuration variable and having that
select CPU_FREQ_TABLE.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
---
 arch/arm/mach-omap2/clkt2xxx_virt_prcm_set.c |    2 +-
 arch/arm/mach-omap2/clock.c                  |    2 +-
 arch/arm/mach-omap2/clock.h                  |    2 +-
 arch/arm/plat-omap/clock.c                   |    2 +-
 arch/arm/plat-omap/include/plat/clock.h      |    4 ++--
 drivers/cpufreq/Kconfig.arm                  |    5 +++++
 drivers/cpufreq/Makefile                     |    2 +-
 7 files changed, 12 insertions(+), 7 deletions(-)

diff --git a/arch/arm/mach-omap2/clkt2xxx_virt_prcm_set.c b/arch/arm/mach-omap2/clkt2xxx_virt_prcm_set.c
index 39f9d5a..4705b50 100644
--- a/arch/arm/mach-omap2/clkt2xxx_virt_prcm_set.c
+++ b/arch/arm/mach-omap2/clkt2xxx_virt_prcm_set.c
@@ -165,7 +165,7 @@ int omap2_select_table_rate(struct clk *clk, unsigned long rate)
 	return 0;
 }
 
-#ifdef CONFIG_CPU_FREQ
+#ifdef CONFIG_ARM_OMAP2PLUS_CPUFREQ
 /*
  * Walk PRCM rate table and fillout cpufreq freq_table
  * XXX This should be replaced by an OPP layer in the near future
diff --git a/arch/arm/mach-omap2/clock.c b/arch/arm/mach-omap2/clock.c
index f57ed5b..450feb0 100644
--- a/arch/arm/mach-omap2/clock.c
+++ b/arch/arm/mach-omap2/clock.c
@@ -536,7 +536,7 @@ struct clk_functions omap2_clk_functions = {
 	.clk_set_rate		= omap2_clk_set_rate,
 	.clk_set_parent		= omap2_clk_set_parent,
 	.clk_disable_unused	= omap2_clk_disable_unused,
-#ifdef CONFIG_CPU_FREQ
+#ifdef CONFIG_ARM_OMAP2PLUS_CPUFREQ
 	/* These will be removed when the OPP code is integrated */
 	.clk_init_cpufreq_table	= omap2_clk_init_cpufreq_table,
 	.clk_exit_cpufreq_table	= omap2_clk_exit_cpufreq_table,
diff --git a/arch/arm/mach-omap2/clock.h b/arch/arm/mach-omap2/clock.h
index b8c2a68..f331cfc 100644
--- a/arch/arm/mach-omap2/clock.h
+++ b/arch/arm/mach-omap2/clock.h
@@ -146,7 +146,7 @@ extern const struct clksel_rate gpt_sys_rates[];
 extern const struct clksel_rate gfx_l3_rates[];
 extern const struct clksel_rate dsp_ick_rates[];
 
-#if defined(CONFIG_ARCH_OMAP2) && defined(CONFIG_CPU_FREQ)
+#if defined(CONFIG_ARCH_OMAP2) && defined(CONFIG_ARM_OMAP2PLUS_CPUFREQ)
 extern void omap2_clk_init_cpufreq_table(struct cpufreq_frequency_table **table);
 extern void omap2_clk_exit_cpufreq_table(struct cpufreq_frequency_table **table);
 #else
diff --git a/arch/arm/plat-omap/clock.c b/arch/arm/plat-omap/clock.c
index 567e4b5..2e2efb6 100644
--- a/arch/arm/plat-omap/clock.c
+++ b/arch/arm/plat-omap/clock.c
@@ -399,7 +399,7 @@ struct clk dummy_ck = {
 	.ops	= &clkops_null,
 };
 
-#ifdef CONFIG_CPU_FREQ
+#ifdef CONFIG_ARM_OMAP2PLUS_CPUFREQ
 void clk_init_cpufreq_table(struct cpufreq_frequency_table **table)
 {
 	unsigned long flags;
diff --git a/arch/arm/plat-omap/include/plat/clock.h b/arch/arm/plat-omap/include/plat/clock.h
index 240a7b9..879ba27 100644
--- a/arch/arm/plat-omap/include/plat/clock.h
+++ b/arch/arm/plat-omap/include/plat/clock.h
@@ -283,7 +283,7 @@ struct clk_functions {
 	void		(*clk_allow_idle)(struct clk *clk);
 	void		(*clk_deny_idle)(struct clk *clk);
 	void		(*clk_disable_unused)(struct clk *clk);
-#ifdef CONFIG_CPU_FREQ
+#ifdef CONFIG_ARM_OMAP2PLUS_CPUFREQ
 	void		(*clk_init_cpufreq_table)(struct cpufreq_frequency_table **);
 	void		(*clk_exit_cpufreq_table)(struct cpufreq_frequency_table **);
 #endif
@@ -301,7 +301,7 @@ extern void recalculate_root_clocks(void);
 extern unsigned long followparent_recalc(struct clk *clk);
 extern void clk_enable_init_clocks(void);
 unsigned long omap_fixed_divisor_recalc(struct clk *clk);
-#ifdef CONFIG_CPU_FREQ
+#ifdef CONFIG_ARM_OMAP2PLUS_CPUFREQ
 extern void clk_init_cpufreq_table(struct cpufreq_frequency_table **table);
 extern void clk_exit_cpufreq_table(struct cpufreq_frequency_table **table);
 #endif
diff --git a/drivers/cpufreq/Kconfig.arm b/drivers/cpufreq/Kconfig.arm
index c8bde43..74e8b55 100644
--- a/drivers/cpufreq/Kconfig.arm
+++ b/drivers/cpufreq/Kconfig.arm
@@ -2,6 +2,11 @@
 # ARM CPU Frequency scaling drivers
 #
 
+config ARM_OMAP2PLUS_CPUFREQ
+	bool
+	depends on ARCH_OMAP2PLUS
+	select CPU_FREQ_TABLE
+
 config ARM_S3C64XX_CPUFREQ
 	bool "Samsung S3C64XX"
 	depends on CPU_S3C6410
diff --git a/drivers/cpufreq/Makefile b/drivers/cpufreq/Makefile
index ac000fa..fda94c7 100644
--- a/drivers/cpufreq/Makefile
+++ b/drivers/cpufreq/Makefile
@@ -44,7 +44,7 @@ obj-$(CONFIG_ARM_S3C64XX_CPUFREQ)	+= s3c64xx-cpufreq.o
 obj-$(CONFIG_ARM_S5PV210_CPUFREQ)	+= s5pv210-cpufreq.o
 obj-$(CONFIG_ARM_EXYNOS_CPUFREQ)	+= exynos-cpufreq.o
 obj-$(CONFIG_ARM_EXYNOS4210_CPUFREQ)	+= exynos4210-cpufreq.o
-obj-$(CONFIG_ARCH_OMAP2PLUS)            += omap-cpufreq.o
+obj-$(CONFIG_ARM_OMAP2PLUS_CPUFREQ)     += omap-cpufreq.o
 
 ##################################################################################
 # PowerPC platform drivers
-- 
1.7.4.4

Re: Lost fixes?

[Samuel Ortiz]

On Thu, Mar 15, 2012 at 10:28:45AM +0000, Russell King - ARM Linux wrote:
> These fixes are a collection from other people which resolve various
> issues found with the new kautobuilder.  They're now a month old, and
> as far as I can see, have not been merged into Linus' tree.
> 
> What's the status of them, and when are they going to be merged?
For the drivers/mfd/da9052-spi.c part: A patch from Axel Lin has been living
in my for-next branch for a while now. It's a section mismatch warning fix, so
not something I consider worth sending to Linus after the merge window is
closed. So it's queued for the next one.

Cheers,
Samuel.

-- 
Intel Open Source Technology Centre
http://oss.intel.com/

Re: [PATCH 6/6] ARM: OMAP: fix cpufreq build

[Kevin Hilman]

Russell King <rmk+kernel@arm.linux.org.uk> writes:

> OMAPs cpufreq requires the frequency table support, but nothing ensures
> that this is selected.  This can result in configurations which fail to
> build:
>
> drivers/built-in.o:(.data+0x5238): undefined reference to `cpufreq_freq_attr_scaling_available_freqs'
> drivers/cpufreq/omap-cpufreq.c:88: undefined reference to `cpufreq_frequency_table_target'
> drivers/cpufreq/omap-cpufreq.c:60: undefined reference to `cpufreq_frequency_table_verify'
> drivers/cpufreq/omap-cpufreq.c:186: undefined reference to `cpufreq_frequency_table_cpuinfo'
> drivers/cpufreq/omap-cpufreq.c:190: undefined reference to `cpufreq_frequency_table_get_attr'
>
> Fix this by introducing a new configuration variable and having that
> select CPU_FREQ_TABLE.
>
> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
> ---
>  arch/arm/mach-omap2/clkt2xxx_virt_prcm_set.c |    2 +-
>  arch/arm/mach-omap2/clock.c                  |    2 +-
>  arch/arm/mach-omap2/clock.h                  |    2 +-
>  arch/arm/plat-omap/clock.c                   |    2 +-
>  arch/arm/plat-omap/include/plat/clock.h      |    4 ++--

The arch/arm/* stuff is all code that has been removed, but the pull
request didn't make it in time for v3.4:

   http://www.spinics.net/lists/arm-kernel/msg164545.html

Initially, this was intended as a cleanup because it was just dead code
removal, but since we missed v3.4, maybe we should pull out 
"ARM: OMAP: clock: cleanup CPUfreq leftovers" and submit for 3.4-rc.

>  drivers/cpufreq/Kconfig.arm                  |    5 +++++
>  drivers/cpufreq/Makefile                     |    2 +-

This change was included in my CPUfreq pull request to Dave Jones.  He
has pulled it into his fixes branch[1], but I don't see it upstream.
I've just sent a mail to Dave asking about it.

Kevin

[1] git://git.kernel.org/pub/scm/linux/kernel/git/davej/cpufreq.git fixes

Re: [PATCH 6/6] ARM: OMAP: fix cpufreq build

[Paul Walmsley]

On Thu, 15 Mar 2012, Kevin Hilman wrote:

> Initially, this was intended as a cleanup because it was just dead code
> removal, but since we missed v3.4, maybe we should pull out 
> "ARM: OMAP: clock: cleanup CPUfreq leftovers" and submit for 3.4-rc.

If it doesn't go upstream for the 3.4 merge window, I'll submit it for 
3.4-rc.


- Paul

Re: [PATCH 6/6] ARM: OMAP: fix cpufreq build

[Russell King - ARM Linux]

Kevin,

I only just found this via the arm-soc tree, while trying to merge the
kautobuildv2 build tree.

You changed my patch and broke it:

config ARM_OMAP2PLUS_CPUFREQ
        bool "TI OMAP2+"
        default ARCH_OMAP2PLUS
        select CPU_FREQ_TABLE

So now, OMAP2 CPUFREQ is selectable on ANYTHING but it will only build
successfully on OMAP.

If you're going to change someone elses patch, do them the favour of
re-posting it back to them for comment *BEFORE* you commit it and send
it upstream, rather than making it look like they're a total tosser for
authoring a patch which fucks up other architecture builds.

Many thanks for that.

On Thu, Mar 15, 2012 at 10:31:00AM +0000, Russell King wrote:
> OMAPs cpufreq requires the frequency table support, but nothing ensures
> that this is selected.  This can result in configurations which fail to
> build:
> 
> drivers/built-in.o:(.data+0x5238): undefined reference to `cpufreq_freq_attr_scaling_available_freqs'
> drivers/cpufreq/omap-cpufreq.c:88: undefined reference to `cpufreq_frequency_table_target'
> drivers/cpufreq/omap-cpufreq.c:60: undefined reference to `cpufreq_frequency_table_verify'
> drivers/cpufreq/omap-cpufreq.c:186: undefined reference to `cpufreq_frequency_table_cpuinfo'
> drivers/cpufreq/omap-cpufreq.c:190: undefined reference to `cpufreq_frequency_table_get_attr'
> 
> Fix this by introducing a new configuration variable and having that
> select CPU_FREQ_TABLE.
> 
> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
> ---
>  arch/arm/mach-omap2/clkt2xxx_virt_prcm_set.c |    2 +-
>  arch/arm/mach-omap2/clock.c                  |    2 +-
>  arch/arm/mach-omap2/clock.h                  |    2 +-
>  arch/arm/plat-omap/clock.c                   |    2 +-
>  arch/arm/plat-omap/include/plat/clock.h      |    4 ++--
>  drivers/cpufreq/Kconfig.arm                  |    5 +++++
>  drivers/cpufreq/Makefile                     |    2 +-
>  7 files changed, 12 insertions(+), 7 deletions(-)
> 
> diff --git a/arch/arm/mach-omap2/clkt2xxx_virt_prcm_set.c b/arch/arm/mach-omap2/clkt2xxx_virt_prcm_set.c
> index 39f9d5a..4705b50 100644
> --- a/arch/arm/mach-omap2/clkt2xxx_virt_prcm_set.c
> +++ b/arch/arm/mach-omap2/clkt2xxx_virt_prcm_set.c
> @@ -165,7 +165,7 @@ int omap2_select_table_rate(struct clk *clk, unsigned long rate)
>  	return 0;
>  }
>  
> -#ifdef CONFIG_CPU_FREQ
> +#ifdef CONFIG_ARM_OMAP2PLUS_CPUFREQ
>  /*
>   * Walk PRCM rate table and fillout cpufreq freq_table
>   * XXX This should be replaced by an OPP layer in the near future
> diff --git a/arch/arm/mach-omap2/clock.c b/arch/arm/mach-omap2/clock.c
> index f57ed5b..450feb0 100644
> --- a/arch/arm/mach-omap2/clock.c
> +++ b/arch/arm/mach-omap2/clock.c
> @@ -536,7 +536,7 @@ struct clk_functions omap2_clk_functions = {
>  	.clk_set_rate		= omap2_clk_set_rate,
>  	.clk_set_parent		= omap2_clk_set_parent,
>  	.clk_disable_unused	= omap2_clk_disable_unused,
> -#ifdef CONFIG_CPU_FREQ
> +#ifdef CONFIG_ARM_OMAP2PLUS_CPUFREQ
>  	/* These will be removed when the OPP code is integrated */
>  	.clk_init_cpufreq_table	= omap2_clk_init_cpufreq_table,
>  	.clk_exit_cpufreq_table	= omap2_clk_exit_cpufreq_table,
> diff --git a/arch/arm/mach-omap2/clock.h b/arch/arm/mach-omap2/clock.h
> index b8c2a68..f331cfc 100644
> --- a/arch/arm/mach-omap2/clock.h
> +++ b/arch/arm/mach-omap2/clock.h
> @@ -146,7 +146,7 @@ extern const struct clksel_rate gpt_sys_rates[];
>  extern const struct clksel_rate gfx_l3_rates[];
>  extern const struct clksel_rate dsp_ick_rates[];
>  
> -#if defined(CONFIG_ARCH_OMAP2) && defined(CONFIG_CPU_FREQ)
> +#if defined(CONFIG_ARCH_OMAP2) && defined(CONFIG_ARM_OMAP2PLUS_CPUFREQ)
>  extern void omap2_clk_init_cpufreq_table(struct cpufreq_frequency_table **table);
>  extern void omap2_clk_exit_cpufreq_table(struct cpufreq_frequency_table **table);
>  #else
> diff --git a/arch/arm/plat-omap/clock.c b/arch/arm/plat-omap/clock.c
> index 567e4b5..2e2efb6 100644
> --- a/arch/arm/plat-omap/clock.c
> +++ b/arch/arm/plat-omap/clock.c
> @@ -399,7 +399,7 @@ struct clk dummy_ck = {
>  	.ops	= &clkops_null,
>  };
>  
> -#ifdef CONFIG_CPU_FREQ
> +#ifdef CONFIG_ARM_OMAP2PLUS_CPUFREQ
>  void clk_init_cpufreq_table(struct cpufreq_frequency_table **table)
>  {
>  	unsigned long flags;
> diff --git a/arch/arm/plat-omap/include/plat/clock.h b/arch/arm/plat-omap/include/plat/clock.h
> index 240a7b9..879ba27 100644
> --- a/arch/arm/plat-omap/include/plat/clock.h
> +++ b/arch/arm/plat-omap/include/plat/clock.h
> @@ -283,7 +283,7 @@ struct clk_functions {
>  	void		(*clk_allow_idle)(struct clk *clk);
>  	void		(*clk_deny_idle)(struct clk *clk);
>  	void		(*clk_disable_unused)(struct clk *clk);
> -#ifdef CONFIG_CPU_FREQ
> +#ifdef CONFIG_ARM_OMAP2PLUS_CPUFREQ
>  	void		(*clk_init_cpufreq_table)(struct cpufreq_frequency_table **);
>  	void		(*clk_exit_cpufreq_table)(struct cpufreq_frequency_table **);
>  #endif
> @@ -301,7 +301,7 @@ extern void recalculate_root_clocks(void);
>  extern unsigned long followparent_recalc(struct clk *clk);
>  extern void clk_enable_init_clocks(void);
>  unsigned long omap_fixed_divisor_recalc(struct clk *clk);
> -#ifdef CONFIG_CPU_FREQ
> +#ifdef CONFIG_ARM_OMAP2PLUS_CPUFREQ
>  extern void clk_init_cpufreq_table(struct cpufreq_frequency_table **table);
>  extern void clk_exit_cpufreq_table(struct cpufreq_frequency_table **table);
>  #endif
> diff --git a/drivers/cpufreq/Kconfig.arm b/drivers/cpufreq/Kconfig.arm
> index c8bde43..74e8b55 100644
> --- a/drivers/cpufreq/Kconfig.arm
> +++ b/drivers/cpufreq/Kconfig.arm
> @@ -2,6 +2,11 @@
>  # ARM CPU Frequency scaling drivers
>  #
>  
> +config ARM_OMAP2PLUS_CPUFREQ
> +	bool
> +	depends on ARCH_OMAP2PLUS
> +	select CPU_FREQ_TABLE
> +
>  config ARM_S3C64XX_CPUFREQ
>  	bool "Samsung S3C64XX"
>  	depends on CPU_S3C6410
> diff --git a/drivers/cpufreq/Makefile b/drivers/cpufreq/Makefile
> index ac000fa..fda94c7 100644
> --- a/drivers/cpufreq/Makefile
> +++ b/drivers/cpufreq/Makefile
> @@ -44,7 +44,7 @@ obj-$(CONFIG_ARM_S3C64XX_CPUFREQ)	+= s3c64xx-cpufreq.o
>  obj-$(CONFIG_ARM_S5PV210_CPUFREQ)	+= s5pv210-cpufreq.o
>  obj-$(CONFIG_ARM_EXYNOS_CPUFREQ)	+= exynos-cpufreq.o
>  obj-$(CONFIG_ARM_EXYNOS4210_CPUFREQ)	+= exynos4210-cpufreq.o
> -obj-$(CONFIG_ARCH_OMAP2PLUS)            += omap-cpufreq.o
> +obj-$(CONFIG_ARM_OMAP2PLUS_CPUFREQ)     += omap-cpufreq.o
>  
>  ##################################################################################
>  # PowerPC platform drivers
> -- 
> 1.7.4.4
> 
> 
> _______________________________________________
> linux-arm-kernel mailing list
> linux-arm-kernel@lists.infradead.org
> http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

Re: [PATCH 6/6] ARM: OMAP: fix cpufreq build

[Kevin Hilman]

Russell King - ARM Linux <linux@arm.linux.org.uk> writes:

> Kevin,
>
> I only just found this via the arm-soc tree, while trying to merge the
> kautobuildv2 build tree.
>
> You changed my patch and broke it:
>
> config ARM_OMAP2PLUS_CPUFREQ
>         bool "TI OMAP2+"
>         default ARCH_OMAP2PLUS
>         select CPU_FREQ_TABLE
>
> So now, OMAP2 CPUFREQ is selectable on ANYTHING but it will only build
> successfully on OMAP.

Yes, I screwed it up.

I wanted it to be selectable, but majorly screwed it up and it is
certainly not randconfig friendly.

This was also pointed out by Grazvydas soon after merging[1], and I was
expecting a patch from him to remedy that, but I will take care of it
and submit a patch for v3.4-rc right away.
 
> If you're going to change someone elses patch, do them the favour of
> re-posting it back to them for comment *BEFORE* you commit it and send
> it upstream, 

I did.  On Feb 16th, I posted my modified version with you and
linux-arm-kernel Cc'd:

   http://lists.infradead.org/pipermail/linux-arm-kernel/2012-February/085053.html

and then I didn't send the pull request for this for another 3 weeks.

> rather than making it look like they're a total tosser for
> authoring a patch which fucks up other architecture builds.

/me has cache miss on tosser in American jargon cache
/me looks up in urban dictionary.

> Many thanks for that.

Many apologies.

Kevin

[1] http://marc.info/?l=linux-omap&m=133174677506294&w=2