diff mbox

[1/1] ARM: dt: add header to define tegra20 clocks

Message ID 1360823899-17846-1-git-send-email-hdoyu@nvidia.com (mailing list archive)
State New, archived
Headers show

Commit Message

Hiroshi DOYU Feb. 14, 2013, 6:38 a.m. UTC
To replace magic number in "clocks = <&tegra_car 28>;"

Signed-off-by: Hiroshi Doyu <hdoyu@nvidia.com>
---
This patch depends on:

  [PATCH 0/9] ARM: tegra: use new dtc+cpp feature
  http://lists.infradead.org/pipermail/linux-arm-kernel/2013-February/149613.html

This patch is the experiment for Tegra20. The same replacement can be
done for Tegra{30,114}.

Usage:
	Modified arch/arm/boot/dts/tegra20.dtsip

Comments

Peter De Schrijver Feb. 14, 2013, 10:12 a.m. UTC | #1
On Thu, Feb 14, 2013 at 07:38:17AM +0100, Hiroshi Doyu wrote:
> To replace magic number in "clocks = <&tegra_car 28>;"
> 

But then we should probably also remove the enum tegra20_clk and replace it
with this header file. We should also consider removing the list of IDs from
the documentation and refer to the header file instead. Otherwise we will have
3 places which define the same IDs and we need to be sure they stay in sync...

Cheers,

Peter.
Hiroshi DOYU Feb. 14, 2013, 1:42 p.m. UTC | #2
Peter De Schrijver <pdeschrijver@nvidia.com> wrote @ Thu, 14 Feb 2013 11:12:20 +0100:

> On Thu, Feb 14, 2013 at 07:38:17AM +0100, Hiroshi Doyu wrote:
> > To replace magic number in "clocks = <&tegra_car 28>;"
> > 
> 
> But then we should probably also remove the enum tegra20_clk and replace it
> with this header file. We should also consider removing the list of IDs from
> the documentation and refer to the header file instead. Otherwise we will have
> 3 places which define the same IDs and we need to be sure they stay in sync...

Actually I generate "arch/arm/boot/dts/tegra{20,30}-car.h" by parsing
"enum tegra{20,30}_clk {..};" with a python script.

There could be the same kind of requirement from others that both
"dtsip" and "C-source" want to share the same header file. If we
consider that DT is the exact H/W description, C-source should be able
to include a DT header file.

For example, a C-header file for "enum tegra{20,30}_clk" should be
generated from "arch/arm/boot/dts/tegra{20,30}-car.h" at runtime, or
alternatively just "drivers/clk/tegra/clk-tegra{20,30}.c" can include
"arch/arm/boot/dts/tegra{20,30}-car.h"? The latter may be easier.
Peter De Schrijver Feb. 14, 2013, 2:58 p.m. UTC | #3
> There could be the same kind of requirement from others that both
> "dtsip" and "C-source" want to share the same header file. If we
> consider that DT is the exact H/W description, C-source should be able
> to include a DT header file.
> 
> For example, a C-header file for "enum tegra{20,30}_clk" should be
> generated from "arch/arm/boot/dts/tegra{20,30}-car.h" at runtime, or
> alternatively just "drivers/clk/tegra/clk-tegra{20,30}.c" can include
> "arch/arm/boot/dts/tegra{20,30}-car.h"? The latter may be easier.

Including arch/arm/boot/dts/tegra{20,30}-car.h seems to make much more sense
to me than generating a enum based on #defines... :)

Cheers,

Peter.
Stephen Warren Feb. 14, 2013, 5:54 p.m. UTC | #4
On 02/13/2013 11:38 PM, Hiroshi Doyu wrote:
> To replace magic number in "clocks = <&tegra_car 28>;"

I like the concept here; I was thinking about doing this today, but you
beat me to it:-) Feel free to create the Tegra30 header too, and modify
all the *.dts* files.

To address other comments in this thread: Yes, I think that we will want
to modify the clock driver to include this header to avoid
duplication/errors (that will require adjusting Linux's include path),
and also remove the list of IDs from the binding document; it can just
refer the the new header by name and cause the header to *be* part of
the binding document.
Shawn Guo Feb. 19, 2013, 5:31 a.m. UTC | #5
On Thu, Feb 14, 2013 at 10:54:28AM -0700, Stephen Warren wrote:
> On 02/13/2013 11:38 PM, Hiroshi Doyu wrote:
> > To replace magic number in "clocks = <&tegra_car 28>;"
> 
> I like the concept here; I was thinking about doing this today, but you
> beat me to it:-) Feel free to create the Tegra30 header too, and modify
> all the *.dts* files.
> 
> To address other comments in this thread: Yes, I think that we will want
> to modify the clock driver to include this header to avoid
> duplication/errors (that will require adjusting Linux's include path),
> and also remove the list of IDs from the binding document; it can just
> refer the the new header by name and cause the header to *be* part of
> the binding document.

I like it, since doing so will help us hijack kernel to have dts stay
in the tree rather than going to a separate repository :)

Seriously, is maintaining dts in a separate repository still a plan?
If yes, we will have duplication problem someday anyway.

For reason of it, I vote for having dts stay in the kernel tree.

Shawn
Stephen Warren Feb. 19, 2013, 5:35 p.m. UTC | #6
On 02/18/2013 10:31 PM, Shawn Guo wrote:
> On Thu, Feb 14, 2013 at 10:54:28AM -0700, Stephen Warren wrote:
>> On 02/13/2013 11:38 PM, Hiroshi Doyu wrote:
>>> To replace magic number in "clocks = <&tegra_car 28>;"
>>
>> I like the concept here; I was thinking about doing this today, but you
>> beat me to it:-) Feel free to create the Tegra30 header too, and modify
>> all the *.dts* files.
>>
>> To address other comments in this thread: Yes, I think that we will want
>> to modify the clock driver to include this header to avoid
>> duplication/errors (that will require adjusting Linux's include path),
>> and also remove the list of IDs from the binding document; it can just
>> refer the the new header by name and cause the header to *be* part of
>> the binding document.
> 
> I like it, since doing so will help us hijack kernel to have dts stay
> in the tree rather than going to a separate repository :)

Well, at least the header files would need to stay in the kernel. The
.txt binding documentation wouldn't have to though.

I guess this is a good argument for not putting these header files into
Documentation/devicetree/bindings, and adding that to the include path,
but rather putting the headers somewhere else.

> Seriously, is maintaining dts in a separate repository still a plan?
> If yes, we will have duplication problem someday anyway.
> 
> For reason of it, I vote for having dts stay in the kernel tree.
> 
> Shawn
>
diff mbox

Patch

diff --git a/arch/arm/boot/dts/tegra20.dtsip b/arch/arm/boot/dts/tegra20.dtsip
index 7b05f53..6edd397 100644
--- a/arch/arm/boot/dts/tegra20.dtsip
+++ b/arch/arm/boot/dts/tegra20.dtsip
@@ -1,6 +1,7 @@ 
 #include "skeleton.dtsi"
 #include "tegra-gpio.h"
 #include "arm-gic.h"
+#include "tegra20-car.h"

 / {
	compatible = "nvidia,tegra20";
@@ -19,7 +20,7 @@ 
		reg = <0x50000000 0x00024000>;
		interrupts = <GIC_SPI 65 IRQ_TYPE_LEVEL_HIGH>, /* syncpt */
			     <GIC_SPI 67 IRQ_TYPE_LEVEL_HIGH>; /* general */
-               clocks = <&tegra_car 28>;
+               clocks = <&tegra_car CLK_HOST1X>;

Signed-off-by: Hiroshi Doyu <hdoyu@nvidia.com>
---
 arch/arm/boot/dts/tegra20-car.h |  126 +++++++++++++++++++++++++++++++++++++++
 1 file changed, 126 insertions(+)
 create mode 100644 arch/arm/boot/dts/tegra20-car.h

diff --git a/arch/arm/boot/dts/tegra20-car.h b/arch/arm/boot/dts/tegra20-car.h
new file mode 100644
index 0000000..6426320
--- /dev/null
+++ b/arch/arm/boot/dts/tegra20-car.h
@@ -0,0 +1,126 @@ 
+#define CLK_CPU		0
+/* UNUSED 1 */
+/* UNUSED 2 */
+#define CLK_AC97	3
+#define CLK_RTC		4
+#define CLK_TIMER	5
+#define CLK_UARTA	6
+/* UNUSED 7 */
+#define CLK_GPIO	8
+#define CLK_SDMMC2	9
+/* UNUSED 10 */
+#define CLK_I2S1	11
+#define CLK_I2C1	12
+#define CLK_NDFLASH	13
+#define CLK_SDMMC1	14
+#define CLK_SDMMC4	15
+#define CLK_TWC		16
+#define CLK_PWM		17
+#define CLK_I2S2	18
+#define CLK_EPP		19
+/* UNUSED 20 */
+#define CLK_GR2D	21
+#define CLK_USBD	22
+#define CLK_ISP		23
+#define CLK_GR3D	24
+#define CLK_IDE		25
+#define CLK_DISP2	26
+#define CLK_DISP1	27
+#define CLK_HOST1X	28
+#define CLK_VCP		29
+/* UNUSED 30 */
+#define CLK_CACHE2	31
+#define CLK_MEM		32
+#define CLK_AHBDMA	33
+#define CLK_APBDMA	34
+/* UNUSED 35 */
+#define CLK_KBC		36
+#define CLK_STAT_MON	37
+#define CLK_PMC		38
+#define CLK_FUSE	39
+#define CLK_KFUSE	40
+#define CLK_SBC1	41
+#define CLK_NOR		42
+#define CLK_SPI		43
+#define CLK_SBC2	44
+#define CLK_XIO		45
+#define CLK_SBC3	46
+#define CLK_DVC		47
+#define CLK_DSI		48
+/* UNUSED 49 */
+#define CLK_MIPI	50
+#define CLK_HDMI	51
+#define CLK_CSI		52
+#define CLK_TVDAC	53
+#define CLK_I2C2	54
+#define CLK_UARTC	55
+/* UNUSED 56 */
+#define CLK_EMC		57
+#define CLK_USB2	58
+#define CLK_USB3	59
+#define CLK_MPE		60
+#define CLK_VDE		61
+#define CLK_BSEA	62
+#define CLK_BSEV	63
+#define CLK_SPEEDO	64
+#define CLK_UARTD	65
+#define CLK_UARTE	66
+#define CLK_I2C3	67
+#define CLK_SBC4	68
+#define CLK_SDMMC3	69
+#define CLK_PEX		70
+#define CLK_OWR		71
+#define CLK_AFI		72
+#define CLK_CSITE	73
+#define CLK_PCIE_XCLK	74
+#define CLK_AVPUCQ	75
+#define CLK_LA		76
+/* UNUSED 77-83 */
+#define CLK_IRAMA	84
+#define CLK_IRAMB	85
+#define CLK_IRAMC	86
+#define CLK_IRAMD	87
+#define CLK_CRAM2	88
+#define CLK_AUDIO_2X	89
+#define CLK_CLK_D	90
+/* UNUSED 91 */
+#define CLK_CSUS	92
+#define CLK_CDEV1	93
+#define CLK_CDEV2	94
+/* UNUSED 95 */
+#define CLK_UARTB	96
+#define CLK_VFIR	97
+#define CLK_SPDIF_IN	98
+#define CLK_SPDIF_OUT	99
+#define CLK_VI		100
+#define CLK_VI_SENSOR	101
+#define CLK_TVO		102
+#define CLK_CVE		103
+#define CLK_OSC		104
+#define CLK_CLK_32K	105
+#define CLK_CLK_M	106
+#define CLK_SCLK	107
+#define CLK_CCLK	108
+#define CLK_HCLK	109
+#define CLK_PCLK	110
+#define CLK_BLINK	111
+#define CLK_PLL_A	112
+#define CLK_PLL_A_OUT0	113
+#define CLK_PLL_C	114
+#define CLK_PLL_C_OUT1	115
+#define CLK_PLL_D	116
+#define CLK_PLL_D_OUT0	117
+#define CLK_PLL_E	118
+#define CLK_PLL_M	119
+#define CLK_PLL_M_OUT1	120
+#define CLK_PLL_P	121
+#define CLK_PLL_P_OUT1	122
+#define CLK_PLL_P_OUT2	123
+#define CLK_PLL_P_OUT3	124
+#define CLK_PLL_P_OUT4	125
+#define CLK_PLL_U	126
+#define CLK_PLL_X	127
+#define CLK_AUDIO	128
+#define CLK_PLL_REF	129
+#define CLK_TWD		130
+#define CLK_MAX		131