Re: [PATCH v6] clk: bcm2835: Add support for programming the audio domain clocks.

From: Eric Anholt
Date: Thu Oct 08 2015 - 21:32:01 EST


Stephen Boyd <sboyd@xxxxxxxxxxxxxx> writes:

> Please drop the full-stop from your subject lines.
>
> On 10/08, Eric Anholt wrote:
>> This adds support for enabling, disabling, and setting the rate of the
>> audio domain clocks. It will be necessary for setting the pixel clock
>> for HDMI in the VC4 driver and let us write a cpufreq driver. It will
>> also improve compatibility with user changes to the firmware's
>> config.txt, since our previous fixed clocks are unaware of it.
>>
>> The firmware also has support for configuring the clocks through the
>> mailbox channel, but the pixel clock setup by the firmware doesn't
>> work, and it's Raspberry Pi specific anyway. The only conflicts we
>> should have with the firmware would be if we made firmware calls that
>> result in clock management (like opening firmware V3D or ISP access,
>> which we don't support in upstream), or on hardware over-thermal or
>> under-voltage (when the firmware would rewrite PLLB to take the ARM
>> out of overclock). If that happens, our cached .recalc_rate() results
>> would be incorrect, but that's no worse than our current state where
>> we used fixed clocks.
>>
>> The existing fixed clocks in the code are left in place to provide
>> backwards compatibility with old device tree files.
>>
>> Signed-off-by: Eric Anholt <eric@xxxxxxxxxx>
>> Tested-by: Martin Sperl <kernel@xxxxxxxxxxxxxxxx>
>> ---
>
> There's a variable length array in here, causing sparse to
> complain:
>
> drivers/clk/bcm/clk-bcm2835.c:1408:41:
> warning: Variable length array is used.
>
> This one looks easy enough to fix with another allocation.

We know the bounds of the allocation, which is up to 10 entries
populated in the 4-bit bitfield. I've switched to using 1 <<
CM_SRC_BITS to shut up sparse.

> But then there's some weird casting warning coming from sparse
> that I honestly don't understand:
>
> drivers/clk/bcm/clk-bcm2835.c:370:36: warning: cast truncates bits
>from constant value (3fffffffff8000 becomes ffff8000)
>drivers/clk/bcm/clk-bcm2835.c:372:19: warning: cast truncates bits from
>constant value (3ffffffff80 becomes ffffff80)
>drivers/clk/bcm/clk-bcm2835.c:378:37: warning: cast truncates bits from
>constant value (fffffffff80000 becomes fff80000)
>drivers/clk/bcm/clk-bcm2835.c:380:42: warning: cast truncates bits from
>constant value (1fffffffff becomes ffffffff)
>
> I guess I'll just ignore that for now. A couple nitpicks are
> left, but nothing major.

That's from GENMASK, which returns an unsigned long, unfortunately.
>
>> + /* Wait for the PLL to lock. */
>> + start = ktime_get();
>> + while (!(cprman_read(cprman, CM_LOCK) & data->lock_mask)) {
>> + ktime_t delta = ktime_sub(ktime_get(), start);
>> +
>> + if (ktime_to_ms(delta) > LOCK_TIMEOUT_MS) {
>
> Didn't notice this one before. Why not add the LOCK_TIMEOUT_MS to
> start, and then call ktime_get() in the loop and use
> ktime_after() to figure out if we're beyond the timeout?

I've switched to that, but I was just modeling off of samsung/clk-pll.c
(one of the few drivers that doesn't infinite loop).

I've fixed the other devm_clk_register() and the extra parens as well.

Attachment: signature.asc
Description: PGP signature