From mboxrd@z Thu Jan  1 00:00:00 1970
From: t-kristo@ti.com (Tero Kristo)
Date: Thu, 1 Aug 2013 18:34:19 +0300
Subject: [PATCHv4 09/33] CLK: omap: add omap4 clock init file
In-Reply-To: <51FA7341.7040303@ti.com>
References: <1374564028-11352-1-git-send-email-t-kristo@ti.com>
 <1374564028-11352-10-git-send-email-t-kristo@ti.com>
 <51F81516.3050302@ti.com> <51F924CA.6000803@ti.com> <51FA7341.7040303@ti.com>
Message-ID: <51FA7FFB.8000309@ti.com>
To: linux-arm-kernel@lists.infradead.org
List-Id: linux-arm-kernel.lists.infradead.org

On 08/01/2013 05:40 PM, Nishanth Menon wrote:
> On 07/31/2013 09:52 AM, Tero Kristo wrote:
>> On 07/30/2013 10:33 PM, Nishanth Menon wrote:
> [...]
>
>>>
>>>> +    DT_CLK(NULL,    "cpufreq_ck",    "dpll_mpu_ck"),
>>> please remove cpufreq.
>>
>> Hmm why?
>>
>> Because cpufreq is completely broken now and your current work on it? :)
>
> cpufreq fixing can be done separately and has abillity to pick up clock
> nodes by describing clocks=<&dpll_mpu_ck> in cpu0 node. absolutely no
> need to carry on this hack - See my comment on [1]
>
> yes, I know I will have to redo my cpufreq series, but thanks to your
> work, it is a 10 min job to enable them back online :) dont worry about
> the broken cpufreq, we will fix it properly and it is not even needed
> for basic boot to shell (example omap2plus_defconfig does not even
> enable CONFIG_CPU_FREQ).

Ok, I'll drop the cpufreq one out at least. :)

>
>
>>>
>>>> +    abe_dpll_ref = clk_get_sys(NULL, "abe_dpll_refclk_mux_ck");
>>>> +    sys_32k_ck = clk_get_sys(NULL, "sys_32k_ck");
>>>> +    rc = clk_set_parent(abe_dpll_ref, sys_32k_ck);
>>>> +    abe_dpll = clk_get_sys(NULL, "dpll_abe_ck");
>>>> +    if (!rc)
>>>> +        rc = clk_set_rate(abe_dpll, OMAP4_DPLL_ABE_DEFFREQ);
>>>> +    if (rc)
>>>> +        pr_err("%s: failed to configure ABE DPLL!\n", __func__);
>>>> +
>>>> +    /*
>>>> +     * Lock USB DPLL on OMAP4 devices so that the L3INIT power
>>>> +     * domain can transition to retention state when not in use.
>>>> +     */
>>>> +    usb_dpll = clk_get_sys(NULL, "dpll_usb_ck");
>>>> +    rc = clk_set_rate(usb_dpll, OMAP4_DPLL_USB_DEFFREQ);
>>>> +    if (rc)
>>>> +        pr_err("%s: failed to configure USB DPLL!\n", __func__);
>>>
>>> ^^^^ why cant we have a generic property that does this "automagically"
>>> from dts property to the node?
>>> a) be able to select a parent
>>> b) be able to set a frequency.
>>> c) how do we ensure that CLK_SET_RATE_PARENT are setup for the clock
>>> nodes defined by dt?
>>
>> This is probably for Mike to answer, currently such mechanism does not
>> exist.
>
> I think you already dealt with (c). in the same vein, we can do (a) and
> (b). Dont think anyone else has gone about this to the extent we are
> attempting to do till date, so as pioneers, we might as well suggest
> with a patch :D ?

Yea, for c I have a patch as that is clearly required. For a/b, we can 
currently workaround. Again, I don't want to implement everything in 
this set at once, as the result would be a huge behemoth (this is 
complicated enough as is imo.)

>
> [1] http://marc.info/?l=linux-omap&m=137536714605620&w=2