[RFC PATCH 0/8] clk: qoriq: Move chip-specific knowledge into driver

STALE3940d

7 messages, 4 authors, 2015-10-22 · open the first message on its own page

[RFC PATCH 0/8] clk: qoriq: Move chip-specific knowledge into driver

From: Scott Wood <hidden>
Date: 2015-06-19 02:49:10

The existing device tree bindings are error-prone and inflexible. 
Correct the mistake by moving the knowledge into the driver, which
has more flexibility in describing the quirks of each chip.  This leaves
the device tree to its proper role of identifying a programming interface
rather than describing its individual registers.

For more detail, see the commit message of patch 4.

Scott Wood (8):
  ARM: dts: ls1021a: Fix clockgen node
  cpufreq: qoriq: Don't look at clock implementation details
  powerpc/fsl: Move fsl_guts.h out of arch/powerpc
  clk: qoriq: Move chip-specific knowledge into driver
  clk: qoriq: Redirect legacy clock nodes to new clocks
  cpufreq: qoriq: Remove frequency masking and minimum
  clk: qoriq: Expose OF clocks directly from the clockgen node
  powerpc/fsl: Use new clockgen binding

 .../devicetree/bindings/clock/qoriq-clock.txt      |   57 +-
 arch/arm/boot/dts/ls1021a.dtsi                     |    3 +
 arch/powerpc/boot/dts/fsl/b4420si-pre.dtsi         |    4 +-
 arch/powerpc/boot/dts/fsl/b4860si-pre.dtsi         |    8 +-
 arch/powerpc/boot/dts/fsl/b4si-post.dtsi           |   15 -
 arch/powerpc/boot/dts/fsl/p2041si-post.dtsi        |   18 -
 arch/powerpc/boot/dts/fsl/p2041si-pre.dtsi         |    8 +-
 arch/powerpc/boot/dts/fsl/p3041si-post.dtsi        |   18 -
 arch/powerpc/boot/dts/fsl/p3041si-pre.dtsi         |    8 +-
 arch/powerpc/boot/dts/fsl/p4080si-post.dtsi        |   70 --
 arch/powerpc/boot/dts/fsl/p4080si-pre.dtsi         |   16 +-
 arch/powerpc/boot/dts/fsl/p5020si-pre.dtsi         |    4 +-
 arch/powerpc/boot/dts/fsl/p5040si-post.dtsi        |   18 -
 arch/powerpc/boot/dts/fsl/p5040si-pre.dtsi         |    8 +-
 arch/powerpc/boot/dts/fsl/qoriq-clockgen1.dtsi     |   50 +-
 arch/powerpc/boot/dts/fsl/qoriq-clockgen2.dtsi     |   33 +-
 arch/powerpc/boot/dts/fsl/t1023si-post.dtsi        |   16 -
 arch/powerpc/boot/dts/fsl/t102xsi-pre.dtsi         |    4 +-
 arch/powerpc/boot/dts/fsl/t1040si-post.dtsi        |   44 -
 arch/powerpc/boot/dts/fsl/t104xsi-pre.dtsi         |    8 +-
 arch/powerpc/boot/dts/fsl/t2081si-post.dtsi        |   22 -
 arch/powerpc/boot/dts/fsl/t208xsi-pre.dtsi         |    8 +-
 arch/powerpc/boot/dts/fsl/t4240si-post.dtsi        |   61 -
 arch/powerpc/boot/dts/fsl/t4240si-pre.dtsi         |   24 +-
 arch/powerpc/platforms/85xx/mpc85xx_mds.c          |    2 +-
 arch/powerpc/platforms/85xx/mpc85xx_rdb.c          |    2 +-
 arch/powerpc/platforms/85xx/p1022_ds.c             |    2 +-
 arch/powerpc/platforms/85xx/p1022_rdk.c            |    2 +-
 arch/powerpc/platforms/85xx/smp.c                  |    2 +-
 arch/powerpc/platforms/85xx/twr_p102x.c            |    2 +-
 arch/powerpc/platforms/86xx/mpc8610_hpcd.c         |    2 +-
 drivers/clk/clk-qoriq.c                            | 1191 ++++++++++++++++----
 drivers/cpufreq/qoriq-cpufreq.c                    |  139 +--
 drivers/iommu/fsl_pamu.c                           |    2 +-
 .../asm/fsl_guts.h => include/linux/fsl/guts.h     |    6 +-
 sound/soc/fsl/mpc8610_hpcd.c                       |    2 +-
 sound/soc/fsl/p1022_ds.c                           |    2 +-
 sound/soc/fsl/p1022_rdk.c                          |    2 +-
 38 files changed, 1107 insertions(+), 776 deletions(-)
 rename arch/powerpc/include/asm/fsl_guts.h => include/linux/fsl/guts.h (98%)

-- 
2.1.4

Re: [RFC PATCH 0/8] clk: qoriq: Move chip-specific knowledge into driver

From: Michael Turquette <hidden>
Date: 2015-08-11 18:25:35

Hi Scott,

Quoting Scott Wood (2015-06-18 19:49:10)
The existing device tree bindings are error-prone and inflexible. =
Correct the mistake by moving the knowledge into the driver, which
has more flexibility in describing the quirks of each chip.  This leaves
the device tree to its proper role of identifying a programming interface
rather than describing its individual registers.
Sorry for not responding to this one sooner. Fell through the cracks.

All of the changes to drives/clk/clk-qoriq.c look great to me. I assume
you need to keep all of these patches together and want to the take
through the freescale tree? If so feel free to add,

Acked-by: Michael Turquette <mturquette@baylibre.com>

Regards,
Mike
=
For more detail, see the commit message of patch 4.
=
Scott Wood (8):
  ARM: dts: ls1021a: Fix clockgen node
  cpufreq: qoriq: Don't look at clock implementation details
  powerpc/fsl: Move fsl_guts.h out of arch/powerpc
  clk: qoriq: Move chip-specific knowledge into driver
  clk: qoriq: Redirect legacy clock nodes to new clocks
  cpufreq: qoriq: Remove frequency masking and minimum
  clk: qoriq: Expose OF clocks directly from the clockgen node
  powerpc/fsl: Use new clockgen binding
=
 .../devicetree/bindings/clock/qoriq-clock.txt      |   57 +-
 arch/arm/boot/dts/ls1021a.dtsi                     |    3 +
 arch/powerpc/boot/dts/fsl/b4420si-pre.dtsi         |    4 +-
 arch/powerpc/boot/dts/fsl/b4860si-pre.dtsi         |    8 +-
 arch/powerpc/boot/dts/fsl/b4si-post.dtsi           |   15 -
 arch/powerpc/boot/dts/fsl/p2041si-post.dtsi        |   18 -
 arch/powerpc/boot/dts/fsl/p2041si-pre.dtsi         |    8 +-
 arch/powerpc/boot/dts/fsl/p3041si-post.dtsi        |   18 -
 arch/powerpc/boot/dts/fsl/p3041si-pre.dtsi         |    8 +-
 arch/powerpc/boot/dts/fsl/p4080si-post.dtsi        |   70 --
 arch/powerpc/boot/dts/fsl/p4080si-pre.dtsi         |   16 +-
 arch/powerpc/boot/dts/fsl/p5020si-pre.dtsi         |    4 +-
 arch/powerpc/boot/dts/fsl/p5040si-post.dtsi        |   18 -
 arch/powerpc/boot/dts/fsl/p5040si-pre.dtsi         |    8 +-
 arch/powerpc/boot/dts/fsl/qoriq-clockgen1.dtsi     |   50 +-
 arch/powerpc/boot/dts/fsl/qoriq-clockgen2.dtsi     |   33 +-
 arch/powerpc/boot/dts/fsl/t1023si-post.dtsi        |   16 -
 arch/powerpc/boot/dts/fsl/t102xsi-pre.dtsi         |    4 +-
 arch/powerpc/boot/dts/fsl/t1040si-post.dtsi        |   44 -
 arch/powerpc/boot/dts/fsl/t104xsi-pre.dtsi         |    8 +-
 arch/powerpc/boot/dts/fsl/t2081si-post.dtsi        |   22 -
 arch/powerpc/boot/dts/fsl/t208xsi-pre.dtsi         |    8 +-
 arch/powerpc/boot/dts/fsl/t4240si-post.dtsi        |   61 -
 arch/powerpc/boot/dts/fsl/t4240si-pre.dtsi         |   24 +-
 arch/powerpc/platforms/85xx/mpc85xx_mds.c          |    2 +-
 arch/powerpc/platforms/85xx/mpc85xx_rdb.c          |    2 +-
 arch/powerpc/platforms/85xx/p1022_ds.c             |    2 +-
 arch/powerpc/platforms/85xx/p1022_rdk.c            |    2 +-
 arch/powerpc/platforms/85xx/smp.c                  |    2 +-
 arch/powerpc/platforms/85xx/twr_p102x.c            |    2 +-
 arch/powerpc/platforms/86xx/mpc8610_hpcd.c         |    2 +-
 drivers/clk/clk-qoriq.c                            | 1191 ++++++++++++++=
++----
 drivers/cpufreq/qoriq-cpufreq.c                    |  139 +--
 drivers/iommu/fsl_pamu.c                           |    2 +-
 .../asm/fsl_guts.h =3D> include/linux/fsl/guts.h     |    6 +-
 sound/soc/fsl/mpc8610_hpcd.c                       |    2 +-
 sound/soc/fsl/p1022_ds.c                           |    2 +-
 sound/soc/fsl/p1022_rdk.c                          |    2 +-
 38 files changed, 1107 insertions(+), 776 deletions(-)
 rename arch/powerpc/include/asm/fsl_guts.h =3D> include/linux/fsl/guts.h=
 (98%)
=
-- =
2.1.4
=20

Re: [RFC PATCH 0/8] clk: qoriq: Move chip-specific knowledge into driver

From: Scott Wood <hidden>
Date: 2015-08-15 06:43:37

On Tue, 2015-08-11 at 11:25 -0700, Michael Turquette wrote:
Hi Scott,

Quoting Scott Wood (2015-06-18 19:49:10)
quoted
The existing device tree bindings are error-prone and inflexible. 
Correct the mistake by moving the knowledge into the driver, which
has more flexibility in describing the quirks of each chip.  This leaves
the device tree to its proper role of identifying a programming interface
rather than describing its individual registers.
Sorry for not responding to this one sooner. Fell through the cracks.

All of the changes to drives/clk/clk-qoriq.c look great to me. I assume
you need to keep all of these patches together and want to the take
through the freescale tree? If so feel free to add,

Acked-by: Michael Turquette <mturquette@baylibre.com>
I just sent a non-RFC v2, with improved compatibility with old device trees 
(especially ls1021a).  It depends on the cpufreq patch though (at least, to 
avoid breaking qoriq-cpufreq until that patch is merged), so I'll also need 
an ack from Rafael for that if I'm taking it through my tree.

-Scott

Re: [RFC PATCH 0/8] clk: qoriq: Move chip-specific knowledge into driver

From: Scott Wood <hidden>
Date: 2015-10-02 00:23:54

On Tue, 2015-08-11 at 11:25 -0700, Michael Turquette wrote:
Hi Scott,

Quoting Scott Wood (2015-06-18 19:49:10)
quoted
The existing device tree bindings are error-prone and inflexible. 
Correct the mistake by moving the knowledge into the driver, which
has more flexibility in describing the quirks of each chip.  This leaves
the device tree to its proper role of identifying a programming interface
rather than describing its individual registers.
Sorry for not responding to this one sooner. Fell through the cracks.

All of the changes to drives/clk/clk-qoriq.c look great to me. I assume
you need to keep all of these patches together and want to the take
through the freescale tree? If so feel free to add,

Acked-by: Michael Turquette <mturquette@baylibre.com>
Is the ack still valid for the v3 patchset?

-Scott

Re: [RFC PATCH 0/8] clk: qoriq: Move chip-specific knowledge into driver

From: Scott Wood <hidden>
Date: 2015-10-02 00:26:29

[Resending to updated e-mail address]

On Tue, 2015-08-11 at 11:25 -0700, Michael Turquette wrote:
Hi Scott,

Quoting Scott Wood (2015-06-18 19:49:10)
quoted
The existing device tree bindings are error-prone and inflexible. 
Correct the mistake by moving the knowledge into the driver, which
has more flexibility in describing the quirks of each chip.  This leaves
the device tree to its proper role of identifying a programming interface
rather than describing its individual registers.
Sorry for not responding to this one sooner. Fell through the cracks.

All of the changes to drives/clk/clk-qoriq.c look great to me. I assume
you need to keep all of these patches together and want to the take
through the freescale tree? If so feel free to add,

Acked-by: Michael Turquette <mturquette@baylibre.com>
Is the ack still valid for the v3 patchset?

-Scott

Re: [RFC PATCH 0/8] clk: qoriq: Move chip-specific knowledge into driver

From: Scott Wood <hidden>
Date: 2015-10-09 23:57:44

On Thu, 2015-10-01 at 19:26 -0500, Scott Wood wrote:
[Resending to updated e-mail address]

On Tue, 2015-08-11 at 11:25 -0700, Michael Turquette wrote:
quoted
Hi Scott,

Quoting Scott Wood (2015-06-18 19:49:10)
quoted
The existing device tree bindings are error-prone and inflexible. 
Correct the mistake by moving the knowledge into the driver, which
has more flexibility in describing the quirks of each chip.  This leaves
the device tree to its proper role of identifying a programming 
interface
rather than describing its individual registers.
Sorry for not responding to this one sooner. Fell through the cracks.

All of the changes to drives/clk/clk-qoriq.c look great to me. I assume
you need to keep all of these patches together and want to the take
through the freescale tree? If so feel free to add,

Acked-by: Michael Turquette <mturquette@baylibre.com>
Is the ack still valid for the v3 patchset?
ping

-Scott

Re: [RFC PATCH 0/8] clk: qoriq: Move chip-specific knowledge into driver

From: Michael Turquette <mturquette@baylibre.com>
Date: 2015-10-22 10:11:16

Quoting Scott Wood (2015-10-09 16:57:44)
On Thu, 2015-10-01 at 19:26 -0500, Scott Wood wrote:
quoted
[Resending to updated e-mail address]
=
quoted
On Tue, 2015-08-11 at 11:25 -0700, Michael Turquette wrote:
quoted
Hi Scott,
=
quoted
quoted
Quoting Scott Wood (2015-06-18 19:49:10)
quoted
The existing device tree bindings are error-prone and inflexible. =
quoted
quoted
quoted
Correct the mistake by moving the knowledge into the driver, which
has more flexibility in describing the quirks of each chip.  This l=
eaves
quoted
quoted
quoted
the device tree to its proper role of identifying a programming =
quoted
quoted
quoted
interface
rather than describing its individual registers.
=
quoted
quoted
Sorry for not responding to this one sooner. Fell through the cracks.
=
quoted
quoted
All of the changes to drives/clk/clk-qoriq.c look great to me. I assu=
me
quoted
quoted
you need to keep all of these patches together and want to the take
through the freescale tree? If so feel free to add,
=
quoted
quoted
Acked-by: Michael Turquette <mturquette@baylibre.com>
=
quoted
Is the ack still valid for the v3 patchset?
Yes. Looks Stephen beat me to it though.

Regards,
Mike
=
ping
=
-Scott
=20
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help