Re: [PATCH 00/11] ARM: OMAP3: legacy clock data move under clk driver

2015-02-03 Thread Arnd Bergmann 
On Thursday 08 January 2015, Tony Lindgren wrote:

 Great, hopefully this will finally allow Mike to make the
 generic struct clk private to drivers/clk :)
 
 Seems to boot just fine based on a quick legacy booting test
 on n900.
 
 Mike, assuming no other issues, can you please apply these into a
 immutable branch against v3.19-rc1 that Paul and I can also merge
 in as needed?
 
 Please also feel free to add:
 
 Acked-by: Tony Lindgren t...@atomide.com

The series has arrived in linux-next and promptly triggered a few
randconfig build errors. Here is a patch that fixes it. Feel free
to replace it with a different patch if you don't like this version.

8
Subject: clk: omap: compile legacy omap3 clocks conditionally

The 'ARM: OMAP3: legacy clock data move under clk driver' patch series
causes build errors when CONFIG_OMAP3 is not set:

drivers/clk/ti/dpll.c: In function 'ti_clk_register_dpll':
drivers/clk/ti/dpll.c:199:31: error: 'omap3_dpll_ck_ops' undeclared (first use 
in this function)
  const struct clk_ops *ops = omap3_dpll_ck_ops;
   ^
drivers/clk/ti/dpll.c:199:31: note: each undeclared identifier is reported only 
once for each function it appears in
drivers/clk/ti/dpll.c:259:10: error: 'omap3_dpll_per_ck_ops' undeclared (first 
use in this function)
   ops = omap3_dpll_per_ck_ops;
  ^

drivers/built-in.o: In function `ti_clk_register_gate':
drivers/clk/ti/gate.c:179: undefined reference to 
`clkhwops_omap3430es2_dss_usbhost_wait'
drivers/clk/ti/gate.c:179: undefined reference to 
`clkhwops_am35xx_ipss_module_wait'
-in.o: In function `ti_clk_register_interface':
drivers/clk/ti/interface.c:100: undefined reference to 
`clkhwops_omap3430es2_iclk_hsotgusb_wait'
drivers/clk/ti/interface.c:100: undefined reference to 
`clkhwops_omap3430es2_iclk_dss_usbhost_wait'
drivers/clk/ti/interface.c:100: undefined reference to 
`clkhwops_omap3430es2_iclk_ssi_wait'
drivers/clk/ti/interface.c:100: undefined reference to 
`clkhwops_am35xx_ipss_wait'
drivers/built-in.o: In function `ti_clk_register_composite':
:(.text+0x3da768): undefined reference to `ti_clk_build_component_gate'

In order to fix that problem, this patch makes the omap3 legacy code
compiled only when both CONFIG_OMAP3 and CONFIG_ATAGS are set.

Signed-off-by: Arnd Bergmann a...@arndb.de

diff --git a/drivers/clk/ti/Makefile b/drivers/clk/ti/Makefile
index 14e6686a5eea..105ffd0f5e79 100644
--- a/drivers/clk/ti/Makefile
+++ b/drivers/clk/ti/Makefile
@@ -1,4 +1,3 @@
-ifneq ($(CONFIG_OF),)
 obj-y  += clk.o autoidle.o clockdomain.o
 clk-common = dpll.o composite.o divider.o gate.o \
  fixed-factor.o mux.o apll.o
@@ -6,10 +5,13 @@ obj-$(CONFIG_SOC_AM33XX)  += $(clk-common) 
clk-33xx.o
 obj-$(CONFIG_SOC_TI81XX)   += $(clk-common) fapll.o clk-816x.o
 obj-$(CONFIG_ARCH_OMAP2)   += $(clk-common) interface.o clk-2xxx.o
 obj-$(CONFIG_ARCH_OMAP3)   += $(clk-common) interface.o \
-  clk-3xxx.o clk-3xxx-legacy.o
+  clk-3xxx.o
 obj-$(CONFIG_ARCH_OMAP4)   += $(clk-common) clk-44xx.o
 obj-$(CONFIG_SOC_OMAP5)+= $(clk-common) clk-54xx.o
 obj-$(CONFIG_SOC_DRA7XX)   += $(clk-common) clk-7xx.o \
   clk-dra7-atl.o
 obj-$(CONFIG_SOC_AM43XX)   += $(clk-common) clk-43xx.o
+
+ifdef CONFIG_ATAGS
+obj-$(CONFIG_ARCH_OMAP3)+= clk-3xxx-legacy.o
 endif
diff --git a/drivers/clk/ti/clk.c b/drivers/clk/ti/clk.c
index 546dae405402..e22b95646e09 100644
--- a/drivers/clk/ti/clk.c
+++ b/drivers/clk/ti/clk.c
@@ -186,6 +186,7 @@ void ti_dt_clk_init_retry_clks(void)
}
 }
 
+#if defined(CONFIG_ARCH_OMAP3)  defined(CONFIG_ATAGS)
 void __init ti_clk_patch_legacy_clks(struct ti_clk **patch)
 {
while (*patch) {
@@ -308,3 +309,4 @@ int __init ti_clk_register_legacy_clks(struct ti_clk_alias 
*clks)
 
return 0;
 }
+#endif
diff --git a/drivers/clk/ti/composite.c b/drivers/clk/ti/composite.c
index 3a9665fce041..3654f61912eb 100644
--- a/drivers/clk/ti/composite.c
+++ b/drivers/clk/ti/composite.c
@@ -118,6 +118,7 @@ static inline struct clk_hw *_get_hw(struct 
clk_hw_omap_comp *clk, int idx)
 
 #define to_clk_hw_comp(_hw) container_of(_hw, struct clk_hw_omap_comp, hw)
 
+#if defined(CONFIG_ARCH_OMAP3)  defined(CONFIG_ATAGS)
 struct clk *ti_clk_register_composite(struct ti_clk *setup)
 {
struct ti_clk_composite *comp;
@@ -153,6 +154,7 @@ struct clk *ti_clk_register_composite(struct ti_clk *setup)
 
return clk;
 }
+#endif
 
 static void __init _register_composite(struct clk_hw *hw,
   struct device_node *node)
diff --git a/drivers/clk/ti/dpll.c b/drivers/clk/ti/dpll.c
index 47ebff772b13..81dc4698dc41 100644
--- a/drivers/clk/ti/dpll.c
+++

Re: [PATCH 00/11] ARM: OMAP3: legacy clock data move under clk driver

2015-02-03 Thread Mike Turquette 
On Tue, Feb 3, 2015 at 11:04 AM, Tony Lindgren t...@atomide.com wrote:
 * Arnd Bergmann a...@arndb.de [150203 09:03]:
 On Thursday 08 January 2015, Tony Lindgren wrote:

  Great, hopefully this will finally allow Mike to make the
  generic struct clk private to drivers/clk :)
 
  Seems to boot just fine based on a quick legacy booting test
  on n900.
 
  Mike, assuming no other issues, can you please apply these into a
  immutable branch against v3.19-rc1 that Paul and I can also merge
  in as needed?
 
  Please also feel free to add:
 
  Acked-by: Tony Lindgren t...@atomide.com

 The series has arrived in linux-next and promptly triggered a few
 randconfig build errors. Here is a patch that fixes it. Feel free
 to replace it with a different patch if you don't like this version.

 8
 Subject: clk: omap: compile legacy omap3 clocks conditionally

 The 'ARM: OMAP3: legacy clock data move under clk driver' patch series
 causes build errors when CONFIG_OMAP3 is not set:

 drivers/clk/ti/dpll.c: In function 'ti_clk_register_dpll':
 drivers/clk/ti/dpll.c:199:31: error: 'omap3_dpll_ck_ops' undeclared (first 
 use in this function)
   const struct clk_ops *ops = omap3_dpll_ck_ops;
^
 drivers/clk/ti/dpll.c:199:31: note: each undeclared identifier is reported 
 only once for each function it appears in
 drivers/clk/ti/dpll.c:259:10: error: 'omap3_dpll_per_ck_ops' undeclared 
 (first use in this function)
ops = omap3_dpll_per_ck_ops;
   ^

 drivers/built-in.o: In function `ti_clk_register_gate':
 drivers/clk/ti/gate.c:179: undefined reference to 
 `clkhwops_omap3430es2_dss_usbhost_wait'
 drivers/clk/ti/gate.c:179: undefined reference to 
 `clkhwops_am35xx_ipss_module_wait'
 -in.o: In function `ti_clk_register_interface':
 drivers/clk/ti/interface.c:100: undefined reference to 
 `clkhwops_omap3430es2_iclk_hsotgusb_wait'
 drivers/clk/ti/interface.c:100: undefined reference to 
 `clkhwops_omap3430es2_iclk_dss_usbhost_wait'
 drivers/clk/ti/interface.c:100: undefined reference to 
 `clkhwops_omap3430es2_iclk_ssi_wait'
 drivers/clk/ti/interface.c:100: undefined reference to 
 `clkhwops_am35xx_ipss_wait'
 drivers/built-in.o: In function `ti_clk_register_composite':
 :(.text+0x3da768): undefined reference to `ti_clk_build_component_gate'

 In order to fix that problem, this patch makes the omap3 legacy code
 compiled only when both CONFIG_OMAP3 and CONFIG_ATAGS are set.

 Looks OK to me:

 Acked-by: Tony Lindgren t...@atomide.com

Thanks for the fix Arnd and for the Ack Tony. I beautified the commitlog a bit.

Applied to clk-next.

Regards,
Mike


 Signed-off-by: Arnd Bergmann a...@arndb.de
 
 diff --git a/drivers/clk/ti/Makefile b/drivers/clk/ti/Makefile
 index 14e6686a5eea..105ffd0f5e79 100644
 --- a/drivers/clk/ti/Makefile
 +++ b/drivers/clk/ti/Makefile
 @@ -1,4 +1,3 @@
 -ifneq ($(CONFIG_OF),)
  obj-y+= clk.o autoidle.o 
 clockdomain.o
  clk-common   = dpll.o composite.o divider.o gate.o \
 fixed-factor.o mux.o apll.o
 @@ -6,10 +5,13 @@ obj-$(CONFIG_SOC_AM33XX)+= $(clk-common) 
 clk-33xx.o
  obj-$(CONFIG_SOC_TI81XX) += $(clk-common) fapll.o clk-816x.o
  obj-$(CONFIG_ARCH_OMAP2) += $(clk-common) interface.o clk-2xxx.o
  obj-$(CONFIG_ARCH_OMAP3) += $(clk-common) interface.o \
 -clk-3xxx.o clk-3xxx-legacy.o
 +clk-3xxx.o
  obj-$(CONFIG_ARCH_OMAP4) += $(clk-common) clk-44xx.o
  obj-$(CONFIG_SOC_OMAP5)  += $(clk-common) clk-54xx.o
  obj-$(CONFIG_SOC_DRA7XX) += $(clk-common) clk-7xx.o \
  clk-dra7-atl.o
  obj-$(CONFIG_SOC_AM43XX) += $(clk-common) clk-43xx.o
 +
 +ifdef CONFIG_ATAGS
 +obj-$(CONFIG_ARCH_OMAP3)+= clk-3xxx-legacy.o
  endif
 diff --git a/drivers/clk/ti/clk.c b/drivers/clk/ti/clk.c
 index 546dae405402..e22b95646e09 100644
 --- a/drivers/clk/ti/clk.c
 +++ b/drivers/clk/ti/clk.c
 @@ -186,6 +186,7 @@ void ti_dt_clk_init_retry_clks(void)
   }
  }

 +#if defined(CONFIG_ARCH_OMAP3)  defined(CONFIG_ATAGS)
  void __init ti_clk_patch_legacy_clks(struct ti_clk **patch)
  {
   while (*patch) {
 @@ -308,3 +309,4 @@ int __init ti_clk_register_legacy_clks(struct 
 ti_clk_alias *clks)

   return 0;
  }
 +#endif
 diff --git a/drivers/clk/ti/composite.c b/drivers/clk/ti/composite.c
 index 3a9665fce041..3654f61912eb 100644
 --- a/drivers/clk/ti/composite.c
 +++ b/drivers/clk/ti/composite.c
 @@ -118,6 +118,7 @@ static inline struct clk_hw *_get_hw(struct 
 clk_hw_omap_comp *clk, int idx)

  #define to_clk_hw_comp(_hw) container_of(_hw, struct clk_hw_omap_comp, hw)

 +#if defined(CONFIG_ARCH_OMAP3)  defined(CONFIG_ATAGS)
  struct clk *ti_clk_register_composite(struct ti_clk *setup)
  {
   struct ti_clk_composite *comp;
 @@

Re: [PATCH 00/11] ARM: OMAP3: legacy clock data move under clk driver

2015-02-03 Thread Tony Lindgren 
* Arnd Bergmann a...@arndb.de [150203 09:03]:
 On Thursday 08 January 2015, Tony Lindgren wrote:
 
  Great, hopefully this will finally allow Mike to make the
  generic struct clk private to drivers/clk :)
  
  Seems to boot just fine based on a quick legacy booting test
  on n900.
  
  Mike, assuming no other issues, can you please apply these into a
  immutable branch against v3.19-rc1 that Paul and I can also merge
  in as needed?
  
  Please also feel free to add:
  
  Acked-by: Tony Lindgren t...@atomide.com
 
 The series has arrived in linux-next and promptly triggered a few
 randconfig build errors. Here is a patch that fixes it. Feel free
 to replace it with a different patch if you don't like this version.
 
 8
 Subject: clk: omap: compile legacy omap3 clocks conditionally
 
 The 'ARM: OMAP3: legacy clock data move under clk driver' patch series
 causes build errors when CONFIG_OMAP3 is not set:
 
 drivers/clk/ti/dpll.c: In function 'ti_clk_register_dpll':
 drivers/clk/ti/dpll.c:199:31: error: 'omap3_dpll_ck_ops' undeclared (first 
 use in this function)
   const struct clk_ops *ops = omap3_dpll_ck_ops;
^
 drivers/clk/ti/dpll.c:199:31: note: each undeclared identifier is reported 
 only once for each function it appears in
 drivers/clk/ti/dpll.c:259:10: error: 'omap3_dpll_per_ck_ops' undeclared 
 (first use in this function)
ops = omap3_dpll_per_ck_ops;
   ^
 
 drivers/built-in.o: In function `ti_clk_register_gate':
 drivers/clk/ti/gate.c:179: undefined reference to 
 `clkhwops_omap3430es2_dss_usbhost_wait'
 drivers/clk/ti/gate.c:179: undefined reference to 
 `clkhwops_am35xx_ipss_module_wait'
 -in.o: In function `ti_clk_register_interface':
 drivers/clk/ti/interface.c:100: undefined reference to 
 `clkhwops_omap3430es2_iclk_hsotgusb_wait'
 drivers/clk/ti/interface.c:100: undefined reference to 
 `clkhwops_omap3430es2_iclk_dss_usbhost_wait'
 drivers/clk/ti/interface.c:100: undefined reference to 
 `clkhwops_omap3430es2_iclk_ssi_wait'
 drivers/clk/ti/interface.c:100: undefined reference to 
 `clkhwops_am35xx_ipss_wait'
 drivers/built-in.o: In function `ti_clk_register_composite':
 :(.text+0x3da768): undefined reference to `ti_clk_build_component_gate'
 
 In order to fix that problem, this patch makes the omap3 legacy code
 compiled only when both CONFIG_OMAP3 and CONFIG_ATAGS are set.

Looks OK to me:

Acked-by: Tony Lindgren t...@atomide.com
 
 Signed-off-by: Arnd Bergmann a...@arndb.de
 
 diff --git a/drivers/clk/ti/Makefile b/drivers/clk/ti/Makefile
 index 14e6686a5eea..105ffd0f5e79 100644
 --- a/drivers/clk/ti/Makefile
 +++ b/drivers/clk/ti/Makefile
 @@ -1,4 +1,3 @@
 -ifneq ($(CONFIG_OF),)
  obj-y+= clk.o autoidle.o 
 clockdomain.o
  clk-common   = dpll.o composite.o divider.o gate.o \
 fixed-factor.o mux.o apll.o
 @@ -6,10 +5,13 @@ obj-$(CONFIG_SOC_AM33XX)+= $(clk-common) 
 clk-33xx.o
  obj-$(CONFIG_SOC_TI81XX) += $(clk-common) fapll.o clk-816x.o
  obj-$(CONFIG_ARCH_OMAP2) += $(clk-common) interface.o clk-2xxx.o
  obj-$(CONFIG_ARCH_OMAP3) += $(clk-common) interface.o \
 -clk-3xxx.o clk-3xxx-legacy.o
 +clk-3xxx.o
  obj-$(CONFIG_ARCH_OMAP4) += $(clk-common) clk-44xx.o
  obj-$(CONFIG_SOC_OMAP5)  += $(clk-common) clk-54xx.o
  obj-$(CONFIG_SOC_DRA7XX) += $(clk-common) clk-7xx.o \
  clk-dra7-atl.o
  obj-$(CONFIG_SOC_AM43XX) += $(clk-common) clk-43xx.o
 +
 +ifdef CONFIG_ATAGS
 +obj-$(CONFIG_ARCH_OMAP3)+= clk-3xxx-legacy.o
  endif
 diff --git a/drivers/clk/ti/clk.c b/drivers/clk/ti/clk.c
 index 546dae405402..e22b95646e09 100644
 --- a/drivers/clk/ti/clk.c
 +++ b/drivers/clk/ti/clk.c
 @@ -186,6 +186,7 @@ void ti_dt_clk_init_retry_clks(void)
   }
  }
  
 +#if defined(CONFIG_ARCH_OMAP3)  defined(CONFIG_ATAGS)
  void __init ti_clk_patch_legacy_clks(struct ti_clk **patch)
  {
   while (*patch) {
 @@ -308,3 +309,4 @@ int __init ti_clk_register_legacy_clks(struct 
 ti_clk_alias *clks)
  
   return 0;
  }
 +#endif
 diff --git a/drivers/clk/ti/composite.c b/drivers/clk/ti/composite.c
 index 3a9665fce041..3654f61912eb 100644
 --- a/drivers/clk/ti/composite.c
 +++ b/drivers/clk/ti/composite.c
 @@ -118,6 +118,7 @@ static inline struct clk_hw *_get_hw(struct 
 clk_hw_omap_comp *clk, int idx)
  
  #define to_clk_hw_comp(_hw) container_of(_hw, struct clk_hw_omap_comp, hw)
  
 +#if defined(CONFIG_ARCH_OMAP3)  defined(CONFIG_ATAGS)
  struct clk *ti_clk_register_composite(struct ti_clk *setup)
  {
   struct ti_clk_composite *comp;
 @@ -153,6 +154,7 @@ struct clk *ti_clk_register_composite(struct ti_clk 
 *setup)
  
   return clk;
  }
 +#endif
  
  static void __init _register_composite(struct clk_hw *hw,

Re: [PATCH 00/11] ARM: OMAP3: legacy clock data move under clk driver

2015-01-30 Thread Tero Kristo 

On 01/30/2015 02:42 AM, Mike Turquette wrote:

Quoting Tero Kristo (2015-01-29 12:19:29)

On 01/08/2015 01:00 AM, Tony Lindgren wrote:

* Tero Kristo t-kri...@ti.com [141216 08:22]:

Hi,

These patches move the legacy clock data for omap3 under drivers/clk/ti.
After these patches are applied, it should be possible to get rid of
clk-private.h (long pending project for Mike.)

Testing done (on top of 3.18-rc1):

omap3-beagle: boot / suspend-resume (ret/off) / cpuidle (ret/off)
omap3-beagle-xm: boot upto fs mount (see note below)
sdp3430: boot
n900: boot

Note: beagle-xm failed with FS mount on the board I have access to, but
this happens with clean 3.18-rc1 and linux-next also at the moment.
The board has probably corrupted filesystem image but I am unable
to fix this atm (remote board.)

Test branch:
tree: https://github.com/t-kristo/linux-pm.git
branch: 3.18-rc1-omap3-clk-rework


Great, hopefully this will finally allow Mike to make the
generic struct clk private to drivers/clk :)

Seems to boot just fine based on a quick legacy booting test
on n900.

Mike, assuming no other issues, can you please apply these into a
immutable branch against v3.19-rc1 that Paul and I can also merge
in as needed?

Please also feel free to add:

Acked-by: Tony Lindgren t...@atomide.com



I just rebased these patches on top of 3.19-rc1, and noticed a problem
with dpll5 on beagle-xm (basically a divide-by-zero error + locking
issue during boot.) The extra diff at the end of this email fixes the
problems, I will also send the updated two patches as v2. Updated branch
available in my tree under name 3.19-rc1-omap3-clk-rework.


I've applied these 11 patches on top of v3.19-rc1, including the two V2
patches for #6 and #8 to the clk-omap-legacy branch here:

git://git.kernel.org/pub/scm/linux/kernel/git/clk/linux.git clk-omap-legacy

Let me know if I screwed anything up. I've merged this immutable branch
into clk-next.

In addition I've made some changes to remove clk-private.h permanently.
I'll post those later today. Again, please let me know if I screwed
anything up.


I can't see the clk-omap-legacy merged to clk-next so far, maybe you did 
not push it yet? The branch itself is identical copy of mine, so can't 
see any issues there so far.


-Tero



Regards,
Mike



- Tero

-

diff --git a/drivers/clk/ti/clk-3xxx-legacy.c
b/drivers/clk/ti/clk-3xxx-legacy.c
index 81ad510..e0732a4 100644
--- a/drivers/clk/ti/clk-3xxx-legacy.c
+++ b/drivers/clk/ti/clk-3xxx-legacy.c
@@ -136,6 +136,7 @@ static struct ti_clk_dpll dpll3_ck_data = {
 .idlest_mask = 0x1,
 .auto_recal_bit = 0x3,
 .max_divider = 0x80,
+   .min_divider = 0x1,
 .recal_en_bit = 0x5,
 .max_multiplier = 0x7ff,
 .enable_mask = 0x7,
@@ -307,6 +308,7 @@ static struct ti_clk_dpll dpll4_ck_data = {
 .idlest_mask = 0x2,
 .auto_recal_bit = 0x13,
 .max_divider = 0x80,
+   .min_divider = 0x1,
 .recal_en_bit = 0x6,
 .max_multiplier = 0x7ff,
 .enable_mask = 0x7,
@@ -507,6 +509,7 @@ static struct ti_clk_dpll dpll5_ck_data = {
 .idlest_mask = 0x1,
 .auto_recal_bit = 0x3,
 .max_divider = 0x80,
+   .min_divider = 0x1,
 .recal_en_bit = 0x19,
 .max_multiplier = 0x7ff,
 .enable_mask = 0x7,
@@ -1271,6 +1274,7 @@ static struct ti_clk_dpll dpll1_ck_data = {
 .idlest_mask = 0x1,
 .auto_recal_bit = 0x3,
 .max_divider = 0x80,
+   .min_divider = 0x1,
 .recal_en_bit = 0x7,
 .max_multiplier = 0x7ff,
 .enable_mask = 0x7,
@@ -2154,6 +2158,7 @@ static struct ti_clk_dpll dpll2_ck_data = {
 .idlest_mask = 0x1,
 .auto_recal_bit = 0x3,
 .max_divider = 0x80,
+   .min_divider = 0x1,
 .recal_en_bit = 0x8,
 .max_multiplier = 0x7ff,
 .enable_mask = 0x7,
@@ -2513,6 +2518,7 @@ static struct ti_clk_dpll dpll4_ck_omap36xx_data = {
 .idlest_mask = 0x2,
 .auto_recal_bit = 0x13,
 .max_divider = 0x80,
+   .min_divider = 0x1,
 .recal_en_bit = 0x6,
 .max_multiplier = 0xfff,
 .enable_mask = 0x7,
diff --git a/drivers/clk/ti/clock.h b/drivers/clk/ti/clock.h
index 8d9c603..404158d 100644
--- a/drivers/clk/ti/clock.h
+++ b/drivers/clk/ti/clock.h
@@ -148,6 +148,7 @@ struct ti_clk_dpll {
 u32 sddiv_mask;
 u16 max_multiplier;
 u16 max_divider;
+   u8 min_divider;
 u8 auto_recal_bit;
 u8 recal_en_bit;
 u8 recal_st_bit;
diff --git a/drivers/clk/ti/dpll.c b/drivers/clk/ti/dpll.c
index 7d67639..47ebff7 100644
--- a/drivers/clk/ti/dpll.c
+++ b/drivers/clk/ti/dpll.c
@@ -243,6 +243,7 @@ struct clk *ti_clk_register_dpll(struct ti_clk *setup)
 dd-sddiv_mask = dpll-sddiv_mask;
 dd-dco_mask = dpll-dco_mask;
 dd-max_divider = dpll-max_divider;
+   dd-min_divider = dpll-min_divider;

Re: [PATCH 00/11] ARM: OMAP3: legacy clock data move under clk driver

2015-01-30 Thread Mike Turquette 
Quoting Tero Kristo (2015-01-30 07:20:36)
 On 01/30/2015 02:42 AM, Mike Turquette wrote:
  Quoting Tero Kristo (2015-01-29 12:19:29)
  On 01/08/2015 01:00 AM, Tony Lindgren wrote:
  * Tero Kristo t-kri...@ti.com [141216 08:22]:
  Hi,
 
  These patches move the legacy clock data for omap3 under drivers/clk/ti.
  After these patches are applied, it should be possible to get rid of
  clk-private.h (long pending project for Mike.)
 
  Testing done (on top of 3.18-rc1):
 
  omap3-beagle: boot / suspend-resume (ret/off) / cpuidle (ret/off)
  omap3-beagle-xm: boot upto fs mount (see note below)
  sdp3430: boot
  n900: boot
 
  Note: beagle-xm failed with FS mount on the board I have access to, but
  this happens with clean 3.18-rc1 and linux-next also at the 
  moment.
  The board has probably corrupted filesystem image but I am unable
  to fix this atm (remote board.)
 
  Test branch:
  tree: https://github.com/t-kristo/linux-pm.git
  branch: 3.18-rc1-omap3-clk-rework
 
  Great, hopefully this will finally allow Mike to make the
  generic struct clk private to drivers/clk :)
 
  Seems to boot just fine based on a quick legacy booting test
  on n900.
 
  Mike, assuming no other issues, can you please apply these into a
  immutable branch against v3.19-rc1 that Paul and I can also merge
  in as needed?
 
  Please also feel free to add:
 
  Acked-by: Tony Lindgren t...@atomide.com
 
 
  I just rebased these patches on top of 3.19-rc1, and noticed a problem
  with dpll5 on beagle-xm (basically a divide-by-zero error + locking
  issue during boot.) The extra diff at the end of this email fixes the
  problems, I will also send the updated two patches as v2. Updated branch
  available in my tree under name 3.19-rc1-omap3-clk-rework.
 
  I've applied these 11 patches on top of v3.19-rc1, including the two V2
  patches for #6 and #8 to the clk-omap-legacy branch here:
 
  git://git.kernel.org/pub/scm/linux/kernel/git/clk/linux.git clk-omap-legacy
 
  Let me know if I screwed anything up. I've merged this immutable branch
  into clk-next.
 
  In addition I've made some changes to remove clk-private.h permanently.
  I'll post those later today. Again, please let me know if I screwed
  anything up.
 
 I can't see the clk-omap-legacy merged to clk-next so far, maybe you did 
 not push it yet? The branch itself is identical copy of mine, so can't 
 see any issues there so far.

That is correct. There are a lot of unrelated changes in clk-next right
now so I didn't push this OMAP stuff last night. Turns out to be a good
thing since linux-next had some breakage today. Once that is resolved
I'll push this out, but clk-omap-legacy IS merged locally in my tree,
just not on the public git that linux-next pulls.

Regards,
Mike

 
 -Tero
 
 
  Regards,
  Mike
 
 
  - Tero
 
  -
 
  diff --git a/drivers/clk/ti/clk-3xxx-legacy.c
  b/drivers/clk/ti/clk-3xxx-legacy.c
  index 81ad510..e0732a4 100644
  --- a/drivers/clk/ti/clk-3xxx-legacy.c
  +++ b/drivers/clk/ti/clk-3xxx-legacy.c
  @@ -136,6 +136,7 @@ static struct ti_clk_dpll dpll3_ck_data = {
   .idlest_mask = 0x1,
   .auto_recal_bit = 0x3,
   .max_divider = 0x80,
  +   .min_divider = 0x1,
   .recal_en_bit = 0x5,
   .max_multiplier = 0x7ff,
   .enable_mask = 0x7,
  @@ -307,6 +308,7 @@ static struct ti_clk_dpll dpll4_ck_data = {
   .idlest_mask = 0x2,
   .auto_recal_bit = 0x13,
   .max_divider = 0x80,
  +   .min_divider = 0x1,
   .recal_en_bit = 0x6,
   .max_multiplier = 0x7ff,
   .enable_mask = 0x7,
  @@ -507,6 +509,7 @@ static struct ti_clk_dpll dpll5_ck_data = {
   .idlest_mask = 0x1,
   .auto_recal_bit = 0x3,
   .max_divider = 0x80,
  +   .min_divider = 0x1,
   .recal_en_bit = 0x19,
   .max_multiplier = 0x7ff,
   .enable_mask = 0x7,
  @@ -1271,6 +1274,7 @@ static struct ti_clk_dpll dpll1_ck_data = {
   .idlest_mask = 0x1,
   .auto_recal_bit = 0x3,
   .max_divider = 0x80,
  +   .min_divider = 0x1,
   .recal_en_bit = 0x7,
   .max_multiplier = 0x7ff,
   .enable_mask = 0x7,
  @@ -2154,6 +2158,7 @@ static struct ti_clk_dpll dpll2_ck_data = {
   .idlest_mask = 0x1,
   .auto_recal_bit = 0x3,
   .max_divider = 0x80,
  +   .min_divider = 0x1,
   .recal_en_bit = 0x8,
   .max_multiplier = 0x7ff,
   .enable_mask = 0x7,
  @@ -2513,6 +2518,7 @@ static struct ti_clk_dpll dpll4_ck_omap36xx_data = {
   .idlest_mask = 0x2,
   .auto_recal_bit = 0x13,
   .max_divider = 0x80,
  +   .min_divider = 0x1,
   .recal_en_bit = 0x6,
   .max_multiplier = 0xfff,
   .enable_mask = 0x7,
  diff --git a/drivers/clk/ti/clock.h b/drivers/clk/ti/clock.h
  index 8d9c603..404158d 100644
  --- a/drivers/clk/ti/clock.h
  +++ b/drivers/clk/ti/clock.h
  @@ -148,6

Re: [PATCH 00/11] ARM: OMAP3: legacy clock data move under clk driver

2015-01-29 Thread Tero Kristo 

On 01/08/2015 01:00 AM, Tony Lindgren wrote:

* Tero Kristo t-kri...@ti.com [141216 08:22]:

Hi,

These patches move the legacy clock data for omap3 under drivers/clk/ti.
After these patches are applied, it should be possible to get rid of
clk-private.h (long pending project for Mike.)

Testing done (on top of 3.18-rc1):

omap3-beagle: boot / suspend-resume (ret/off) / cpuidle (ret/off)
omap3-beagle-xm: boot upto fs mount (see note below)
sdp3430: boot
n900: boot

Note: beagle-xm failed with FS mount on the board I have access to, but
   this happens with clean 3.18-rc1 and linux-next also at the moment.
   The board has probably corrupted filesystem image but I am unable
   to fix this atm (remote board.)

Test branch:
tree: https://github.com/t-kristo/linux-pm.git
branch: 3.18-rc1-omap3-clk-rework


Great, hopefully this will finally allow Mike to make the
generic struct clk private to drivers/clk :)

Seems to boot just fine based on a quick legacy booting test
on n900.

Mike, assuming no other issues, can you please apply these into a
immutable branch against v3.19-rc1 that Paul and I can also merge
in as needed?

Please also feel free to add:

Acked-by: Tony Lindgren t...@atomide.com



I just rebased these patches on top of 3.19-rc1, and noticed a problem 
with dpll5 on beagle-xm (basically a divide-by-zero error + locking 
issue during boot.) The extra diff at the end of this email fixes the 
problems, I will also send the updated two patches as v2. Updated branch 
available in my tree under name 3.19-rc1-omap3-clk-rework.


- Tero

-

diff --git a/drivers/clk/ti/clk-3xxx-legacy.c 
b/drivers/clk/ti/clk-3xxx-legacy.c

index 81ad510..e0732a4 100644
--- a/drivers/clk/ti/clk-3xxx-legacy.c
+++ b/drivers/clk/ti/clk-3xxx-legacy.c
@@ -136,6 +136,7 @@ static struct ti_clk_dpll dpll3_ck_data = {
.idlest_mask = 0x1,
.auto_recal_bit = 0x3,
.max_divider = 0x80,
+   .min_divider = 0x1,
.recal_en_bit = 0x5,
.max_multiplier = 0x7ff,
.enable_mask = 0x7,
@@ -307,6 +308,7 @@ static struct ti_clk_dpll dpll4_ck_data = {
.idlest_mask = 0x2,
.auto_recal_bit = 0x13,
.max_divider = 0x80,
+   .min_divider = 0x1,
.recal_en_bit = 0x6,
.max_multiplier = 0x7ff,
.enable_mask = 0x7,
@@ -507,6 +509,7 @@ static struct ti_clk_dpll dpll5_ck_data = {
.idlest_mask = 0x1,
.auto_recal_bit = 0x3,
.max_divider = 0x80,
+   .min_divider = 0x1,
.recal_en_bit = 0x19,
.max_multiplier = 0x7ff,
.enable_mask = 0x7,
@@ -1271,6 +1274,7 @@ static struct ti_clk_dpll dpll1_ck_data = {
.idlest_mask = 0x1,
.auto_recal_bit = 0x3,
.max_divider = 0x80,
+   .min_divider = 0x1,
.recal_en_bit = 0x7,
.max_multiplier = 0x7ff,
.enable_mask = 0x7,
@@ -2154,6 +2158,7 @@ static struct ti_clk_dpll dpll2_ck_data = {
.idlest_mask = 0x1,
.auto_recal_bit = 0x3,
.max_divider = 0x80,
+   .min_divider = 0x1,
.recal_en_bit = 0x8,
.max_multiplier = 0x7ff,
.enable_mask = 0x7,
@@ -2513,6 +2518,7 @@ static struct ti_clk_dpll dpll4_ck_omap36xx_data = {
.idlest_mask = 0x2,
.auto_recal_bit = 0x13,
.max_divider = 0x80,
+   .min_divider = 0x1,
.recal_en_bit = 0x6,
.max_multiplier = 0xfff,
.enable_mask = 0x7,
diff --git a/drivers/clk/ti/clock.h b/drivers/clk/ti/clock.h
index 8d9c603..404158d 100644
--- a/drivers/clk/ti/clock.h
+++ b/drivers/clk/ti/clock.h
@@ -148,6 +148,7 @@ struct ti_clk_dpll {
u32 sddiv_mask;
u16 max_multiplier;
u16 max_divider;
+   u8 min_divider;
u8 auto_recal_bit;
u8 recal_en_bit;
u8 recal_st_bit;
diff --git a/drivers/clk/ti/dpll.c b/drivers/clk/ti/dpll.c
index 7d67639..47ebff7 100644
--- a/drivers/clk/ti/dpll.c
+++ b/drivers/clk/ti/dpll.c
@@ -243,6 +243,7 @@ struct clk *ti_clk_register_dpll(struct ti_clk *setup)
dd-sddiv_mask = dpll-sddiv_mask;
dd-dco_mask = dpll-dco_mask;
dd-max_divider = dpll-max_divider;
+   dd-min_divider = dpll-min_divider;
dd-max_multiplier = dpll-max_multiplier;
dd-auto_recal_bit = dpll-auto_recal_bit;
dd-recal_en_bit = dpll-recal_en_bit;

--
To unsubscribe from this list: send the line unsubscribe linux-omap in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 00/11] ARM: OMAP3: legacy clock data move under clk driver

2015-01-29 Thread Mike Turquette 
Quoting Tero Kristo (2015-01-29 12:19:29)
 On 01/08/2015 01:00 AM, Tony Lindgren wrote:
  * Tero Kristo t-kri...@ti.com [141216 08:22]:
  Hi,
 
  These patches move the legacy clock data for omap3 under drivers/clk/ti.
  After these patches are applied, it should be possible to get rid of
  clk-private.h (long pending project for Mike.)
 
  Testing done (on top of 3.18-rc1):
 
  omap3-beagle: boot / suspend-resume (ret/off) / cpuidle (ret/off)
  omap3-beagle-xm: boot upto fs mount (see note below)
  sdp3430: boot
  n900: boot
 
  Note: beagle-xm failed with FS mount on the board I have access to, but
 this happens with clean 3.18-rc1 and linux-next also at the moment.
 The board has probably corrupted filesystem image but I am unable
 to fix this atm (remote board.)
 
  Test branch:
  tree: https://github.com/t-kristo/linux-pm.git
  branch: 3.18-rc1-omap3-clk-rework
 
  Great, hopefully this will finally allow Mike to make the
  generic struct clk private to drivers/clk :)
 
  Seems to boot just fine based on a quick legacy booting test
  on n900.
 
  Mike, assuming no other issues, can you please apply these into a
  immutable branch against v3.19-rc1 that Paul and I can also merge
  in as needed?
 
  Please also feel free to add:
 
  Acked-by: Tony Lindgren t...@atomide.com
 
 
 I just rebased these patches on top of 3.19-rc1, and noticed a problem 
 with dpll5 on beagle-xm (basically a divide-by-zero error + locking 
 issue during boot.) The extra diff at the end of this email fixes the 
 problems, I will also send the updated two patches as v2. Updated branch 
 available in my tree under name 3.19-rc1-omap3-clk-rework.

I've applied these 11 patches on top of v3.19-rc1, including the two V2
patches for #6 and #8 to the clk-omap-legacy branch here:

git://git.kernel.org/pub/scm/linux/kernel/git/clk/linux.git clk-omap-legacy

Let me know if I screwed anything up. I've merged this immutable branch
into clk-next.

In addition I've made some changes to remove clk-private.h permanently.
I'll post those later today. Again, please let me know if I screwed
anything up.

Regards,
Mike

 
 - Tero
 
 -
 
 diff --git a/drivers/clk/ti/clk-3xxx-legacy.c 
 b/drivers/clk/ti/clk-3xxx-legacy.c
 index 81ad510..e0732a4 100644
 --- a/drivers/clk/ti/clk-3xxx-legacy.c
 +++ b/drivers/clk/ti/clk-3xxx-legacy.c
 @@ -136,6 +136,7 @@ static struct ti_clk_dpll dpll3_ck_data = {
 .idlest_mask = 0x1,
 .auto_recal_bit = 0x3,
 .max_divider = 0x80,
 +   .min_divider = 0x1,
 .recal_en_bit = 0x5,
 .max_multiplier = 0x7ff,
 .enable_mask = 0x7,
 @@ -307,6 +308,7 @@ static struct ti_clk_dpll dpll4_ck_data = {
 .idlest_mask = 0x2,
 .auto_recal_bit = 0x13,
 .max_divider = 0x80,
 +   .min_divider = 0x1,
 .recal_en_bit = 0x6,
 .max_multiplier = 0x7ff,
 .enable_mask = 0x7,
 @@ -507,6 +509,7 @@ static struct ti_clk_dpll dpll5_ck_data = {
 .idlest_mask = 0x1,
 .auto_recal_bit = 0x3,
 .max_divider = 0x80,
 +   .min_divider = 0x1,
 .recal_en_bit = 0x19,
 .max_multiplier = 0x7ff,
 .enable_mask = 0x7,
 @@ -1271,6 +1274,7 @@ static struct ti_clk_dpll dpll1_ck_data = {
 .idlest_mask = 0x1,
 .auto_recal_bit = 0x3,
 .max_divider = 0x80,
 +   .min_divider = 0x1,
 .recal_en_bit = 0x7,
 .max_multiplier = 0x7ff,
 .enable_mask = 0x7,
 @@ -2154,6 +2158,7 @@ static struct ti_clk_dpll dpll2_ck_data = {
 .idlest_mask = 0x1,
 .auto_recal_bit = 0x3,
 .max_divider = 0x80,
 +   .min_divider = 0x1,
 .recal_en_bit = 0x8,
 .max_multiplier = 0x7ff,
 .enable_mask = 0x7,
 @@ -2513,6 +2518,7 @@ static struct ti_clk_dpll dpll4_ck_omap36xx_data = {
 .idlest_mask = 0x2,
 .auto_recal_bit = 0x13,
 .max_divider = 0x80,
 +   .min_divider = 0x1,
 .recal_en_bit = 0x6,
 .max_multiplier = 0xfff,
 .enable_mask = 0x7,
 diff --git a/drivers/clk/ti/clock.h b/drivers/clk/ti/clock.h
 index 8d9c603..404158d 100644
 --- a/drivers/clk/ti/clock.h
 +++ b/drivers/clk/ti/clock.h
 @@ -148,6 +148,7 @@ struct ti_clk_dpll {
 u32 sddiv_mask;
 u16 max_multiplier;
 u16 max_divider;
 +   u8 min_divider;
 u8 auto_recal_bit;
 u8 recal_en_bit;
 u8 recal_st_bit;
 diff --git a/drivers/clk/ti/dpll.c b/drivers/clk/ti/dpll.c
 index 7d67639..47ebff7 100644
 --- a/drivers/clk/ti/dpll.c
 +++ b/drivers/clk/ti/dpll.c
 @@ -243,6 +243,7 @@ struct clk *ti_clk_register_dpll(struct ti_clk *setup)
 dd-sddiv_mask = dpll-sddiv_mask;
 dd-dco_mask = dpll-dco_mask;
 dd-max_divider = dpll-max_divider;
 +   dd-min_divider = dpll-min_divider;
 dd-max_multiplier = dpll-max_multiplier;
 dd-auto_recal_bit = dpll-auto_recal_bit;
 dd-recal_en_bit = dpll-recal_en_bit;
 
--
To

Re: [PATCH 00/11] ARM: OMAP3: legacy clock data move under clk driver

2015-01-07 Thread Tony Lindgren 
* Tero Kristo t-kri...@ti.com [141216 08:22]:
 Hi,
 
 These patches move the legacy clock data for omap3 under drivers/clk/ti.
 After these patches are applied, it should be possible to get rid of
 clk-private.h (long pending project for Mike.)
 
 Testing done (on top of 3.18-rc1):
 
 omap3-beagle: boot / suspend-resume (ret/off) / cpuidle (ret/off)
 omap3-beagle-xm: boot upto fs mount (see note below)
 sdp3430: boot
 n900: boot
 
 Note: beagle-xm failed with FS mount on the board I have access to, but
   this happens with clean 3.18-rc1 and linux-next also at the moment.
   The board has probably corrupted filesystem image but I am unable
   to fix this atm (remote board.)
 
 Test branch:
 tree: https://github.com/t-kristo/linux-pm.git
 branch: 3.18-rc1-omap3-clk-rework

Great, hopefully this will finally allow Mike to make the
generic struct clk private to drivers/clk :)

Seems to boot just fine based on a quick legacy booting test
on n900.

Mike, assuming no other issues, can you please apply these into a
immutable branch against v3.19-rc1 that Paul and I can also merge
in as needed?

Please also feel free to add:

Acked-by: Tony Lindgren t...@atomide.com
--
To unsubscribe from this list: send the line unsubscribe linux-omap in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH 00/11] ARM: OMAP3: legacy clock data move under clk driver

2014-12-16 Thread Tero Kristo 
Hi,

These patches move the legacy clock data for omap3 under drivers/clk/ti.
After these patches are applied, it should be possible to get rid of
clk-private.h (long pending project for Mike.)

Testing done (on top of 3.18-rc1):

omap3-beagle: boot / suspend-resume (ret/off) / cpuidle (ret/off)
omap3-beagle-xm: boot upto fs mount (see note below)
sdp3430: boot
n900: boot

Note: beagle-xm failed with FS mount on the board I have access to, but
  this happens with clean 3.18-rc1 and linux-next also at the moment.
  The board has probably corrupted filesystem image but I am unable
  to fix this atm (remote board.)

Test branch:
tree: https://github.com/t-kristo/linux-pm.git
branch: 3.18-rc1-omap3-clk-rework

-Tero

--
To unsubscribe from this list: send the line unsubscribe linux-omap in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html