[PATCH 0/2] ARM: OMAP4: clock data fixes for early 3.8-rc

linux-arm-kernel.lists.infradead.org archive mirror
 help / color / mirror / Atom feed

* [PATCH 0/2] ARM: OMAP4: clock data fixes for early 3.8-rc
@ 2012-12-14  7:32 Paul Walmsley
  2012-12-14  7:32 ` [PATCH 1/2] ARM: OMAP4: clock data: div_iva_hs_clk is a power-of-two divider Paul Walmsley
                   ` (2 more replies)
  0 siblings, 3 replies; 6+ messages in thread
From: Paul Walmsley @ 2012-12-14  7:32 UTC (permalink / raw)
  To: linux-arm-kernel

A few bugs in the OMAP4 common clock framework data caused boot-time
warnings on OMAP4460 PandaBoard-ES; these patches fix those.  They
apply after the arm-soc merge during pre-3.8-rc1.

Also needed are Jon Hunter's clock fixes series; the pull request will
combine his patches and this series.


- Paul

---

Paul Walmsley (2):
      ARM: OMAP4: clock data: div_iva_hs_clk is a power-of-two divider
      ARM: OMAP4: clock data: DPLLs are missing bypass clocks in their parent lists


 arch/arm/mach-omap2/cclock44xx_data.c |   37 +++++++++++++++++++++++++--------
 1 file changed, 28 insertions(+), 9 deletions(-)

^ permalink raw reply	[flat|nested] 6+ messages in thread

* [PATCH 1/2] ARM: OMAP4: clock data: div_iva_hs_clk is a power-of-two divider
  2012-12-14  7:32 [PATCH 0/2] ARM: OMAP4: clock data fixes for early 3.8-rc Paul Walmsley
@ 2012-12-14  7:32 ` Paul Walmsley
  2012-12-14 16:46   ` Mike Turquette
  2012-12-14  7:32 ` [PATCH 2/2] ARM: OMAP4: clock data: DPLLs are missing bypass clocks in their parent lists Paul Walmsley
  2012-12-14 18:34 ` [PATCH 0/2] ARM: OMAP4: clock data fixes for early 3.8-rc Tony Lindgren
  2 siblings, 1 reply; 6+ messages in thread
From: Paul Walmsley @ 2012-12-14  7:32 UTC (permalink / raw)
  To: linux-arm-kernel

The OMAP4 clock divider "div_iva_hs_clk" is listed in the clock data
as an OMAP HSDIVIDER, but it's actually a power-of-two divider.  This
causes a warning during boot on an OMAP4460 Pandaboard-ES with a
recent u-boot:

WARNING: at arch/arm/mach-omap2/clkt_clksel.c:143 omap2_clksel_recalc+0xf4/0x12c()
clock: div_iva_hs_clk: could not find fieldval 0 for parent dpll_core_m5x2_ck

Fix by converting the data for this clock to a power-of-two divider.

Signed-off-by: Paul Walmsley <paul@pwsan.com>
Cc: Mike Turquette <mturquette@linaro.org>
---
 arch/arm/mach-omap2/cclock44xx_data.c |    6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/arch/arm/mach-omap2/cclock44xx_data.c b/arch/arm/mach-omap2/cclock44xx_data.c
index aa56c3e..a3c54b2 100644
--- a/arch/arm/mach-omap2/cclock44xx_data.c
+++ b/arch/arm/mach-omap2/cclock44xx_data.c
@@ -286,9 +286,9 @@ DEFINE_CLK_DIVIDER(div_core_ck, "dpll_core_m5x2_ck", &dpll_core_m5x2_ck, 0x0,
 		   OMAP4430_CM_CLKSEL_CORE, OMAP4430_CLKSEL_CORE_SHIFT,
 		   OMAP4430_CLKSEL_CORE_WIDTH, 0x0, NULL);
 
-DEFINE_CLK_OMAP_HSDIVIDER(div_iva_hs_clk, "dpll_core_m5x2_ck",
-			  &dpll_core_m5x2_ck, 0x0, OMAP4430_CM_BYPCLK_DPLL_IVA,
-			  OMAP4430_CLKSEL_0_1_MASK);
+DEFINE_CLK_DIVIDER(div_iva_hs_clk, "dpll_core_m5x2_ck", &dpll_core_m5x2_ck,
+		   0x0, OMAP4430_CM_BYPCLK_DPLL_IVA, OMAP4430_CLKSEL_0_1_SHIFT,
+		   OMAP4430_CLKSEL_0_1_WIDTH, CLK_DIVIDER_POWER_OF_TWO, NULL);
 
 DEFINE_CLK_DIVIDER(div_mpu_hs_clk, "dpll_core_m5x2_ck", &dpll_core_m5x2_ck,
 		   0x0, OMAP4430_CM_BYPCLK_DPLL_MPU, OMAP4430_CLKSEL_0_1_SHIFT,

^ permalink raw reply related	[flat|nested] 6+ messages in thread

* [PATCH 1/2] ARM: OMAP4: clock data: div_iva_hs_clk is a power-of-two divider
  2012-12-14  7:32 ` [PATCH 1/2] ARM: OMAP4: clock data: div_iva_hs_clk is a power-of-two divider Paul Walmsley
@ 2012-12-14 16:46   ` Mike Turquette
  0 siblings, 0 replies; 6+ messages in thread
From: Mike Turquette @ 2012-12-14 16:46 UTC (permalink / raw)
  To: linux-arm-kernel

On Thu, Dec 13, 2012 at 11:32 PM, Paul Walmsley <paul@pwsan.com> wrote:
> The OMAP4 clock divider "div_iva_hs_clk" is listed in the clock data
> as an OMAP HSDIVIDER, but it's actually a power-of-two divider.  This
> causes a warning during boot on an OMAP4460 Pandaboard-ES with a
> recent u-boot:
>
> WARNING: at arch/arm/mach-omap2/clkt_clksel.c:143 omap2_clksel_recalc+0xf4/0x12c()
> clock: div_iva_hs_clk: could not find fieldval 0 for parent dpll_core_m5x2_ck
>
> Fix by converting the data for this clock to a power-of-two divider.
>
> Signed-off-by: Paul Walmsley <paul@pwsan.com>
> Cc: Mike Turquette <mturquette@linaro.org>
> ---
>  arch/arm/mach-omap2/cclock44xx_data.c |    6 +++---
>  1 file changed, 3 insertions(+), 3 deletions(-)
>
> diff --git a/arch/arm/mach-omap2/cclock44xx_data.c b/arch/arm/mach-omap2/cclock44xx_data.c
> index aa56c3e..a3c54b2 100644
> --- a/arch/arm/mach-omap2/cclock44xx_data.c
> +++ b/arch/arm/mach-omap2/cclock44xx_data.c
> @@ -286,9 +286,9 @@ DEFINE_CLK_DIVIDER(div_core_ck, "dpll_core_m5x2_ck", &dpll_core_m5x2_ck, 0x0,
>                    OMAP4430_CM_CLKSEL_CORE, OMAP4430_CLKSEL_CORE_SHIFT,
>                    OMAP4430_CLKSEL_CORE_WIDTH, 0x0, NULL);
>
> -DEFINE_CLK_OMAP_HSDIVIDER(div_iva_hs_clk, "dpll_core_m5x2_ck",
> -                         &dpll_core_m5x2_ck, 0x0, OMAP4430_CM_BYPCLK_DPLL_IVA,
> -                         OMAP4430_CLKSEL_0_1_MASK);
> +DEFINE_CLK_DIVIDER(div_iva_hs_clk, "dpll_core_m5x2_ck", &dpll_core_m5x2_ck,
> +                  0x0, OMAP4430_CM_BYPCLK_DPLL_IVA, OMAP4430_CLKSEL_0_1_SHIFT,
> +                  OMAP4430_CLKSEL_0_1_WIDTH, CLK_DIVIDER_POWER_OF_TWO, NULL);
>

Hi Paul,

Looks good to me.

Regards,
Mike

>  DEFINE_CLK_DIVIDER(div_mpu_hs_clk, "dpll_core_m5x2_ck", &dpll_core_m5x2_ck,
>                    0x0, OMAP4430_CM_BYPCLK_DPLL_MPU, OMAP4430_CLKSEL_0_1_SHIFT,
>
>

^ permalink raw reply	[flat|nested] 6+ messages in thread

* [PATCH 2/2] ARM: OMAP4: clock data: DPLLs are missing bypass clocks in their parent lists
  2012-12-14  7:32 [PATCH 0/2] ARM: OMAP4: clock data fixes for early 3.8-rc Paul Walmsley
  2012-12-14  7:32 ` [PATCH 1/2] ARM: OMAP4: clock data: div_iva_hs_clk is a power-of-two divider Paul Walmsley
@ 2012-12-14  7:32 ` Paul Walmsley
  2012-12-14 16:51   ` Mike Turquette
  2012-12-14 18:34 ` [PATCH 0/2] ARM: OMAP4: clock data fixes for early 3.8-rc Tony Lindgren
  2 siblings, 1 reply; 6+ messages in thread
From: Paul Walmsley @ 2012-12-14  7:32 UTC (permalink / raw)
  To: linux-arm-kernel

Booting OMAP4460 Pandaboard ES with a recent u-boot results in this
warning:

WARNING: at arch/arm/mach-omap2/dpll3xxx.c:427 omap3_noncore_dpll_enable+0xf4/0x110()

The OMAP4 DPLL parent clock name arrays only listed the reference
clocks, not the bypass clocks.  Fix by adding the bypass clocks to the
DPLL parent lists.

Signed-off-by: Paul Walmsley <paul@pwsan.com>
Cc: Mike Turquette <mturquette@linaro.org>
---
 arch/arm/mach-omap2/cclock44xx_data.c |   31 +++++++++++++++++++++++++------
 1 file changed, 25 insertions(+), 6 deletions(-)

diff --git a/arch/arm/mach-omap2/cclock44xx_data.c b/arch/arm/mach-omap2/cclock44xx_data.c
index a3c54b2..4536bae 100644
--- a/arch/arm/mach-omap2/cclock44xx_data.c
+++ b/arch/arm/mach-omap2/cclock44xx_data.c
@@ -233,7 +233,7 @@ static struct dpll_data dpll_core_dd = {
 
 
 static const char *dpll_core_ck_parents[] = {
-	"sys_clkin_ck",
+	"sys_clkin_ck", "core_hsd_byp_clk_mux_ck"
 };
 
 static struct clk dpll_core_ck;
@@ -363,6 +363,10 @@ static struct dpll_data dpll_iva_dd = {
 	.min_divider	= 1,
 };
 
+static const char *dpll_iva_ck_parents[] = {
+	"sys_clkin_ck", "iva_hsd_byp_clk_mux_ck"
+};
+
 static struct clk dpll_iva_ck;
 
 static struct clk_hw_omap dpll_iva_ck_hw = {
@@ -373,7 +377,7 @@ static struct clk_hw_omap dpll_iva_ck_hw = {
 	.ops		= &clkhwops_omap3_dpll,
 };
 
-DEFINE_STRUCT_CLK(dpll_iva_ck, dpll_core_ck_parents, dpll_abe_ck_ops);
+DEFINE_STRUCT_CLK(dpll_iva_ck, dpll_iva_ck_parents, dpll_abe_ck_ops);
 
 static const char *dpll_iva_x2_ck_parents[] = {
 	"dpll_iva_ck",
@@ -416,6 +420,10 @@ static struct dpll_data dpll_mpu_dd = {
 	.min_divider	= 1,
 };
 
+static const char *dpll_mpu_ck_parents[] = {
+	"sys_clkin_ck", "div_mpu_hs_clk"
+};
+
 static struct clk dpll_mpu_ck;
 
 static struct clk_hw_omap dpll_mpu_ck_hw = {
@@ -426,7 +434,7 @@ static struct clk_hw_omap dpll_mpu_ck_hw = {
 	.ops		= &clkhwops_omap3_dpll,
 };
 
-DEFINE_STRUCT_CLK(dpll_mpu_ck, dpll_core_ck_parents, dpll_abe_ck_ops);
+DEFINE_STRUCT_CLK(dpll_mpu_ck, dpll_mpu_ck_parents, dpll_abe_ck_ops);
 
 DEFINE_CLK_FIXED_FACTOR(mpu_periphclk, "dpll_mpu_ck", &dpll_mpu_ck, 0x0, 1, 2);
 
@@ -464,6 +472,9 @@ static struct dpll_data dpll_per_dd = {
 	.min_divider	= 1,
 };
 
+static const char *dpll_per_ck_parents[] = {
+	"sys_clkin_ck", "per_hsd_byp_clk_mux_ck"
+};
 
 static struct clk dpll_per_ck;
 
@@ -475,7 +486,7 @@ static struct clk_hw_omap dpll_per_ck_hw = {
 	.ops		= &clkhwops_omap3_dpll,
 };
 
-DEFINE_STRUCT_CLK(dpll_per_ck, dpll_core_ck_parents, dpll_abe_ck_ops);
+DEFINE_STRUCT_CLK(dpll_per_ck, dpll_per_ck_parents, dpll_abe_ck_ops);
 
 DEFINE_CLK_DIVIDER(dpll_per_m2_ck, "dpll_per_ck", &dpll_per_ck, 0x0,
 		   OMAP4430_CM_DIV_M2_DPLL_PER, OMAP4430_DPLL_CLKOUT_DIV_SHIFT,
@@ -559,6 +570,10 @@ static struct dpll_data dpll_usb_dd = {
 	.min_divider	= 1,
 };
 
+static const char *dpll_usb_ck_parents[] = {
+	"sys_clkin_ck", "usb_hs_clk_div_ck"
+};
+
 static struct clk dpll_usb_ck;
 
 static struct clk_hw_omap dpll_usb_ck_hw = {
@@ -569,7 +584,7 @@ static struct clk_hw_omap dpll_usb_ck_hw = {
 	.ops		= &clkhwops_omap3_dpll,
 };
 
-DEFINE_STRUCT_CLK(dpll_usb_ck, dpll_core_ck_parents, dpll_abe_ck_ops);
+DEFINE_STRUCT_CLK(dpll_usb_ck, dpll_usb_ck_parents, dpll_abe_ck_ops);
 
 static const char *dpll_usb_clkdcoldo_ck_parents[] = {
 	"dpll_usb_ck",
@@ -696,9 +711,13 @@ DEFINE_CLK_DIVIDER(syc_clk_div_ck, "sys_clkin_ck", &sys_clkin_ck, 0x0,
 		   OMAP4430_CM_ABE_DSS_SYS_CLKSEL, OMAP4430_CLKSEL_0_0_SHIFT,
 		   OMAP4430_CLKSEL_0_0_WIDTH, 0x0, NULL);
 
+static const char *dbgclk_mux_ck_parents[] = {
+	"sys_clkin_ck"
+};
+
 static struct clk dbgclk_mux_ck;
 DEFINE_STRUCT_CLK_HW_OMAP(dbgclk_mux_ck, NULL);
-DEFINE_STRUCT_CLK(dbgclk_mux_ck, dpll_core_ck_parents,
+DEFINE_STRUCT_CLK(dbgclk_mux_ck, dbgclk_mux_ck_parents,
 		  dpll_usb_clkdcoldo_ck_ops);
 
 /* Leaf clocks controlled by modules */

^ permalink raw reply related	[flat|nested] 6+ messages in thread

* [PATCH 2/2] ARM: OMAP4: clock data: DPLLs are missing bypass clocks in their parent lists
  2012-12-14  7:32 ` [PATCH 2/2] ARM: OMAP4: clock data: DPLLs are missing bypass clocks in their parent lists Paul Walmsley
@ 2012-12-14 16:51   ` Mike Turquette
  0 siblings, 0 replies; 6+ messages in thread
From: Mike Turquette @ 2012-12-14 16:51 UTC (permalink / raw)
  To: linux-arm-kernel

On Thu, Dec 13, 2012 at 11:32 PM, Paul Walmsley <paul@pwsan.com> wrote:
> Booting OMAP4460 Pandaboard ES with a recent u-boot results in this
> warning:
>
> WARNING: at arch/arm/mach-omap2/dpll3xxx.c:427 omap3_noncore_dpll_enable+0xf4/0x110()
>
> The OMAP4 DPLL parent clock name arrays only listed the reference
> clocks, not the bypass clocks.  Fix by adding the bypass clocks to the
> DPLL parent lists.
>
> Signed-off-by: Paul Walmsley <paul@pwsan.com>
> Cc: Mike Turquette <mturquette@linaro.org>

That is really strange.  I wonder if something went wrong with the
autogenerated data or perhaps later on when the file was being edited
by hand...

Patch looks good to me.

Regards,
Mike

^ permalink raw reply	[flat|nested] 6+ messages in thread

* [PATCH 0/2] ARM: OMAP4: clock data fixes for early 3.8-rc
  2012-12-14  7:32 [PATCH 0/2] ARM: OMAP4: clock data fixes for early 3.8-rc Paul Walmsley
  2012-12-14  7:32 ` [PATCH 1/2] ARM: OMAP4: clock data: div_iva_hs_clk is a power-of-two divider Paul Walmsley
  2012-12-14  7:32 ` [PATCH 2/2] ARM: OMAP4: clock data: DPLLs are missing bypass clocks in their parent lists Paul Walmsley
@ 2012-12-14 18:34 ` Tony Lindgren
  2 siblings, 0 replies; 6+ messages in thread
From: Tony Lindgren @ 2012-12-14 18:34 UTC (permalink / raw)
  To: linux-arm-kernel

* Paul Walmsley <paul@pwsan.com> [121213 23:37]:
> A few bugs in the OMAP4 common clock framework data caused boot-time
> warnings on OMAP4460 PandaBoard-ES; these patches fix those.  They
> apply after the arm-soc merge during pre-3.8-rc1.
> 
> Also needed are Jon Hunter's clock fixes series; the pull request will
> combine his patches and this series.

OK good to hear, I was wondering about that patch. No need for me to
reply to that thread then :)

Regards,

Tony

^ permalink raw reply	[flat|nested] 6+ messages in thread

end of thread, other threads:[~2012-12-14 18:34 UTC | newest]

Thread overview: 6+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2012-12-14  7:32 [PATCH 0/2] ARM: OMAP4: clock data fixes for early 3.8-rc Paul Walmsley
2012-12-14  7:32 ` [PATCH 1/2] ARM: OMAP4: clock data: div_iva_hs_clk is a power-of-two divider Paul Walmsley
2012-12-14 16:46   ` Mike Turquette
2012-12-14  7:32 ` [PATCH 2/2] ARM: OMAP4: clock data: DPLLs are missing bypass clocks in their parent lists Paul Walmsley
2012-12-14 16:51   ` Mike Turquette
2012-12-14 18:34 ` [PATCH 0/2] ARM: OMAP4: clock data fixes for early 3.8-rc Tony Lindgren

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).