public inbox for u-boot@lists.denx.de
 help / color / mirror / Atom feed
From: Lukasz Majewski <lukma@denx.de>
To: u-boot@lists.denx.de
Subject: [U-Boot] [i.MX8MM+CCF 11/41] clk: fixed_rate: export clk_fixed_rate
Date: Wed, 8 May 2019 09:40:09 +0200	[thread overview]
Message-ID: <20190508094009.33082ad8@jawa> (raw)
In-Reply-To: <AM0PR04MB4481898838E3DD7B533A7F4188320@AM0PR04MB4481.eurprd04.prod.outlook.com>

On Wed, 8 May 2019 06:51:46 +0000
Peng Fan <peng.fan@nxp.com> wrote:

> > -----Original Message-----
> > From: Lukasz Majewski [mailto:lukma at denx.de]
> > Sent: 2019年5月8日 14:46
> > To: Peng Fan <peng.fan@nxp.com>
> > Cc: sbabic at denx.de; festevam at gmail.com; dl-uboot-imx
> > <uboot-imx@nxp.com>; sjg at chromium.org; jagan at amarulasolutions.com;
> > sr at denx.de; u-boot at lists.denx.de; trini at konsulko.com
> > Subject: Re: [i.MX8MM+CCF 11/41] clk: fixed_rate: export
> > clk_fixed_rate
> > 
> > On Tue, 7 May 2019 13:27:45 +0000
> > Peng Fan <peng.fan@nxp.com> wrote:
> >   
> > > > Subject: Re: [i.MX8MM+CCF 11/41] clk: fixed_rate: export
> > > > clk_fixed_rate
> > > >
> > > > Hi Peng,
> > > >  
> > > > > Export the structure for others to use.
> > > > >
> > > > > Signed-off-by: Peng Fan <peng.fan@nxp.com>
> > > > > ---
> > > > >  drivers/clk/clk_fixed_rate.c | 8 +-------
> > > > > include/linux/clk-provider.h | 7 +++++++
> > > > >  2 files changed, 8 insertions(+), 7 deletions(-)
> > > > >
> > > > > diff --git a/drivers/clk/clk_fixed_rate.c
> > > > > b/drivers/clk/clk_fixed_rate.c index 089f060a23..069e643fbc
> > > > > 100644 --- a/drivers/clk/clk_fixed_rate.c
> > > > > +++ b/drivers/clk/clk_fixed_rate.c
> > > > > @@ -6,13 +6,7 @@
> > > > >  #include <common.h>
> > > > >  #include <clk-uclass.h>
> > > > >  #include <dm.h>
> > > > > -
> > > > > -struct clk_fixed_rate {
> > > > > -	struct clk clk;
> > > > > -	unsigned long fixed_rate;
> > > > > -};
> > > > > -
> > > > > -#define to_clk_fixed_rate(dev)	((struct clk_fixed_rate
> > > > > *)dev_get_platdata(dev)) +#include <linux/clk-provider.h>
> > > > >
> > > > >  static ulong clk_fixed_rate_get_rate(struct clk *clk)  { diff
> > > > > --git a/include/linux/clk-provider.h
> > > > > b/include/linux/clk-provider.h index 3ed0db86d2..b2bed768b6
> > > > > 100644 --- a/include/linux/clk-provider.h
> > > > > +++ b/include/linux/clk-provider.h
> > > > > @@ -112,6 +112,13 @@ struct clk_fixed_factor {  #define
> > > > > to_clk_fixed_factor(_clk) container_of(_clk, struct
> > > > > clk_fixed_factor,\ clk)
> > > > >
> > > > > +struct clk_fixed_rate {
> > > > > +	struct clk clk;
> > > > > +	unsigned long fixed_rate;
> > > > > +};  
> > > >
> > > > I think that this struct shall stay where it was. Moreover, the
> > > > clk-provider.h is not the API to be used by other parts of the
> > > > clock API.
> > > >
> > > > The clk_fixed_rate shall be accessed via get_rate() only and in
> > > > IMX6Q it is available in early SPL (parsed from dts /clocks
> > > > property
> > > > - the 24MHz OSC)  
> > > > > +
> > > > > +#define to_clk_fixed_rate(dev)	((struct clk_fixed_rate
> > > > > *)dev_get_platdata(dev)) +
> > > > >  int clk_register(struct clk *clk, const char *drv_name,
> > > > >  		 ulong drv_data, const char *name,
> > > > >  		 const char *parent_name);  
> > > >
> > > > Please explain why iMX8MM needs such global export?  
> > >
> > > In clk-imx8mm.c, first configure ARM clk to osc24M fixed clk to
> > > change pll clock.
> > > +       /* Configure ARM to osc24M */
> > > +       clk_get_by_id(IMX8MM_CLK_A53_SRC, &clkp);
> > > +       uclass_get_device_by_name(UCLASS_CLK, "clock-osc-24m",  
> > &devp);  
> > > +       clkp1 = &to_clk_fixed_rate(devp)->clk;
> > > +       clk_set_parent(clkp, clkp1);  
> > 
> > This code looks a bit strange to me. Why imx8mm sets parent here?  
> 
> The A53 clk could not change on the fly. There is a mux here, one is
> PLL, one is OSC, And there are others. If we want to change the pll
> clock which is currently being used by A53, we need first switch the
> A53 clk to source from OSC, then change pll clock, then switch A53
> clk back to PLL.

The above description looks like a "standard" procedure for bypassing
PLL when it is going to be locked.

The same is also performed on IMX6Q:

https://elixir.bootlin.com/linux/v5.1/source/drivers/clk/imx/clk-imx6q.c#L88

But I've not ported that part from the original Linux source code.

> 
> > 
> > In the imx6q I write "osc" as a part of the clock tree:
> > 
> >         clk_dm(IMX6QDL_CLK_PLL2,
> >                imx_clk_pllv3(IMX_PLLV3_GENERIC, "pll2_bus", "osc",
> >                              base + 0x30, 0x1));
> > 
> > And here the parent is set as "osc".
> > 
> > Moreover, the "osc" or in your case "clock-osc-24m" shall be
> > available in "clocks" DTS node and hence parsed / setup from there
> > (and be available in the clk uclass list).
> >   
> > > +
> > > +       /* Configure ARM PLL to 1.2GHz */
> > > +       clk_get_by_id(IMX8MM_ARM_PLL, &clkp1);
> > > +       clk_set_rate(clkp1, 1200000000UL);  
> > 
> > Shouldn't this be set in DTS ? As it is now - it seems like a
> > hardcoded value for early SPL clock setup. Am I right?  
> 
> You mean get freq from cpu opp and set cpu freq? I do not have good
> idea how to set in DTS.

I'm thinking about following description of the "osc" fixed clock:
https://elixir.bootlin.com/linux/v5.1/source/arch/arm/boot/dts/imx6qdl.dtsi#L64

and

http://git.denx.de/?p=u-boot/u-boot-imx.git;a=blob;f=arch/arm/dts/imx6qdl.dtsi;h=c0a94780087382a6e2805b7d6572f3e5e294e302;hb=HEAD#l53

The "fixed-clock" code has been adjusted to comply with the above DTS
(which is provided in pre-reloc SPL - even before console and ddr
setup).

(I had to use debug uart for debugging).

> 
> Thanks,
> Peng.
> 
> >   
> > > +       clk_get_by_id(IMX8MM_ARM_PLL_OUT, &clkp1);
> > > +       clk_enable(clkp1);
> > > +       clk_set_parent(clkp, clkp1);
> > > +
> > > +       /* Configure DIV to 1.2GHz */
> > > +       clk_get_by_id(IMX8MM_CLK_A53_DIV, &clkp1);
> > > +       clk_set_rate(clkp1, 1200000000UL);
> > >
> > > Regards,
> > > Peng.
> > >  
> > > >
> > > >
> > > > Best regards,
> > > >
> > > > Lukasz Majewski
> > > >
> > > > --
> > > >
> > > > DENX Software Engineering GmbH,      Managing Director: Wolfgang
> > > > Denk HRB 165235 Munich, Office: Kirchenstr.5, D-82194
> > > > Groebenzell, Germany Phone: (+49)-8142-66989-59 Fax:
> > > > (+49)-8142-66989-80 Email: lukma at denx.de  
> > 
> > 
> > 
> > 
> > Best regards,
> > 
> > Lukasz Majewski
> > 
> > --
> > 
> > DENX Software Engineering GmbH,      Managing Director: Wolfgang
> > Denk HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell,
> > Germany Phone: (+49)-8142-66989-59 Fax: (+49)-8142-66989-80 Email:
> > lukma at denx.de  




Best regards,

Lukasz Majewski

--

DENX Software Engineering GmbH,      Managing Director: Wolfgang Denk
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: (+49)-8142-66989-59 Fax: (+49)-8142-66989-80 Email: lukma at denx.de
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 488 bytes
Desc: OpenPGP digital signature
URL: <http://lists.denx.de/pipermail/u-boot/attachments/20190508/c8c9ec12/attachment.sig>

  reply	other threads:[~2019-05-08  7:40 UTC|newest]

Thread overview: 84+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-04-30 10:17 [U-Boot] [i.MX8MM+CCF 00/41] i.MX8MM + CCF Peng Fan
2019-04-30 10:17 ` [U-Boot] [i.MX8MM+CCF 01/41] clk: correct get clk_x pointer Peng Fan
2019-05-06 21:36   ` Lukasz Majewski
2019-04-30 10:17 ` [U-Boot] [i.MX8MM+CCF 02/41] clk: fixed-factor: fix get clk_fixed_factor Peng Fan
2019-05-06 21:41   ` Lukasz Majewski
2019-04-30 10:17 ` [U-Boot] [i.MX8MM+CCF 03/41] clk: introduce clk_dev_binded Peng Fan
2019-05-06 21:57   ` Lukasz Majewski
2019-05-07 13:22     ` Peng Fan
2019-05-08  7:30       ` Lukasz Majewski
2019-05-08  7:41         ` Peng Fan
2019-05-08 22:24           ` Lukasz Majewski
2019-04-30 10:17 ` [U-Boot] [i.MX8MM+CCF 04/41] clk: use clk_dev_binded Peng Fan
2019-05-06 21:59   ` Lukasz Majewski
2019-05-08  7:27   ` Lukasz Majewski
2019-04-30 10:17 ` [U-Boot] [i.MX8MM+CCF 05/41] clk-provider: sync more clk flags from Linux Kernel Peng Fan
2019-05-06 22:01   ` Lukasz Majewski
2019-04-30 10:17 ` [U-Boot] [i.MX8MM+CCF 06/41] cmd: clk: print err value when clk_get_rate failed Peng Fan
2019-05-06 22:03   ` Lukasz Majewski
2019-04-30 10:17 ` [U-Boot] [i.MX8MM+CCF 07/41] clk: mux: add set parent support Peng Fan
2019-05-06 22:04   ` Lukasz Majewski
2019-05-07 13:23     ` Peng Fan
2019-04-30 10:17 ` [U-Boot] [i.MX8MM+CCF 08/41] clk: export mux/divider ops Peng Fan
2019-05-06 22:06   ` Lukasz Majewski
2019-05-07 13:25     ` Peng Fan
2019-05-08  6:31       ` Lukasz Majewski
2019-05-08  6:35         ` Peng Fan
2019-04-30 10:18 ` [U-Boot] [i.MX8MM+CCF 09/41] clk: add clk-gate support Peng Fan
2019-04-30 10:18 ` [U-Boot] [i.MX8MM+CCF 10/41] divider set rate supporrt Peng Fan
2019-05-06 22:08   ` Lukasz Majewski
2019-04-30 10:18 ` [U-Boot] [i.MX8MM+CCF 11/41] clk: fixed_rate: export clk_fixed_rate Peng Fan
2019-05-06 22:15   ` Lukasz Majewski
2019-05-07 13:27     ` Peng Fan
2019-05-08  6:46       ` Lukasz Majewski
2019-05-08  6:51         ` Peng Fan
2019-05-08  7:40           ` Lukasz Majewski [this message]
2019-05-08  7:45             ` Peng Fan
2019-05-08 22:27               ` Lukasz Majewski
2019-05-09  1:13                 ` Peng Fan
2019-05-16  8:55                   ` Lukasz Majewski
2019-05-16  9:58                     ` Peng Fan
2019-05-16 10:08                       ` Lukasz Majewski
2019-04-30 10:18 ` [U-Boot] [i.MX8MM+CCF 12/41] clk: fixed_rate: add pre reloc flag Peng Fan
2019-05-06 22:16   ` Lukasz Majewski
2019-05-07 13:29     ` Peng Fan
2019-05-07 14:00       ` Bin Meng
2019-05-08  7:00         ` Lukasz Majewski
2019-05-08  6:57       ` Lukasz Majewski
2019-05-08  7:21         ` Peng Fan
2019-04-30 10:18 ` [U-Boot] [i.MX8MM+CCF 13/41] clk: imx: import clk heplers Peng Fan
2019-05-06 22:17   ` Lukasz Majewski
2019-04-30 10:18 ` [U-Boot] [i.MX8MM+CCF 14/41] clk: imx: gate2 add set rate Peng Fan
2019-04-30 10:18 ` [U-Boot] [i.MX8MM+CCF 15/41] linux: compat: guard PAGE_SIZE Peng Fan
2019-04-30 10:18 ` [U-Boot] [i.MX8MM+CCF 16/41] drivers: core: use strcmp when find device by name Peng Fan
2019-04-30 10:18 ` [U-Boot] [i.MX8MM+CCF 17/41] ddr: imx8m: fix ddr firmware location when enable SPL OF Peng Fan
2019-04-30 10:18 ` [U-Boot] [i.MX8MM+CCF 18/41] imx: add IMX8MQ kconfig entry Peng Fan
2019-04-30 10:18 ` [U-Boot] [i.MX8MM+CCF 19/41] imx: add IMX8MM " Peng Fan
2019-04-30 10:18 ` [U-Boot] [i.MX8MM+CCF 20/41] imx: imx8mm: add clock bindings header Peng Fan
2019-04-30 10:18 ` [U-Boot] [i.MX8MM+CCF 21/41] imx: add i.MX8MM cpu type Peng Fan
2019-04-30 10:18 ` [U-Boot] [i.MX8MM+CCF 22/41] imx: spl: add spl_board_boot_device for i.MX8MM Peng Fan
2019-04-30 10:18 ` [U-Boot] [i.MX8MM+CCF 23/41] imx8m: update imx-regs " Peng Fan
2019-04-30 10:18 ` [U-Boot] [i.MX8MM+CCF 24/41] imx: add get_cpu_rev support " Peng Fan
2019-04-30 10:18 ` [U-Boot] [i.MX8MM+CCF 25/41] imx8m: rename clock to clock_imx8mq Peng Fan
2019-04-30 10:19 ` [U-Boot] [i.MX8MM+CCF 26/41] imx8m: restructure clock.h Peng Fan
2019-04-30 10:19 ` [U-Boot] [i.MX8MM+CCF 27/41] imx8m: add clk support for i.MX8MM Peng Fan
2019-04-30 10:19 ` [U-Boot] [i.MX8MM+CCF 28/41] imx8m: soc: probe clk before relocation Peng Fan
2019-04-30 10:19 ` [U-Boot] [i.MX8MM+CCF 29/41] imx8m: add pin header for i.MX8MM Peng Fan
2019-04-30 10:19 ` [U-Boot] [i.MX8MM+CCF 30/41] imx: add i.MX8MM PE property Peng Fan
2019-04-30 10:19 ` [U-Boot] [i.MX8MM+CCF 31/41] imx8m: Fix MMU table issue for OPTEE memory Peng Fan
2019-04-30 10:19 ` [U-Boot] [i.MX8MM+CCF 32/41] imx8m: set BYPASS ID SWAP to avoid AXI bus errors Peng Fan
2019-04-30 10:19 ` [U-Boot] [i.MX8MM+CCF 33/41] imx8m: Configure trustzone region 0 for non-secure access Peng Fan
2019-04-30 10:19 ` [U-Boot] [i.MX8MM+CCF 34/41] imx8m: soc: enable SCTR clock before timer init Peng Fan
2019-04-30 10:19 ` [U-Boot] [i.MX8MM+CCF 35/41] serial: Kconfig: make MXC_UART usable for MX7 and IMX8M Peng Fan
2019-04-30 10:19 ` [U-Boot] [i.MX8MM+CCF 36/41] clk: imx: add Kconfig entry for i.MX8MM Peng Fan
2019-04-30 10:19 ` [U-Boot] [i.MX8MM+CCF 37/41] clk: imx: add pll14xx driver Peng Fan
2019-04-30 10:19 ` [U-Boot] [i.MX8MM+CCF 38/41] clk: add composite clk support Peng Fan
2019-04-30 10:19 ` [U-Boot] [i.MX8MM+CCF 39/41] clk: imx: add i.MX8MM " Peng Fan
2019-04-30 10:19 ` [U-Boot] [i.MX8MM+CCF 40/41] clk: imx: add i.MX8MM clk driver Peng Fan
2019-04-30 10:19 ` [U-Boot] [i.MX8MM+CCF 41/41] imx: add i.MX8MM EVK board support Peng Fan
2019-05-06  7:57 ` [U-Boot] [i.MX8MM+CCF 00/41] i.MX8MM + CCF Schrempf Frieder
2019-05-06  8:26   ` Schrempf Frieder
2019-05-06  9:08     ` Peng Fan
2019-05-06  9:55       ` Lukasz Majewski
2019-05-06 21:32 ` Lukasz Majewski
2019-05-07 13:16   ` Peng Fan

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20190508094009.33082ad8@jawa \
    --to=lukma@denx.de \
    --cc=u-boot@lists.denx.de \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox