Re: [V3] tty: serial: qcom-geni-serial: Fix get_clk_div_rate() which otherwise could return a sub-optimal clock rate.

[Vijaya Krishna Nivarthi <quic_vnivarth@quicinc.com>]

On 7/8/2022 6:41 PM, Greg KH wrote:
> On Fri, Jul 08, 2022 at 12:47:37AM +0530, Vijaya Krishna Nivarthi wrote:
>> In the logic around call to clk_round_rate(), for some corner conditions,
>> get_clk_div_rate() could return an sub-optimal clock rate. Also, if an
>> exact clock rate was not found lowest clock was being returned.
>>
>> Search for suitable clock rate in 2 steps
>> a) exact match or within 2% tolerance
>> b) within 5% tolerance
>> This also takes care of corner conditions.
>>
>> Fixes: c2194bc999d4 ("tty: serial: qcom-geni-serial: Remove uart frequency table. Instead, find suitable frequency with call to clk_round_rate")
>> Signed-off-by: Vijaya Krishna Nivarthi <quic_vnivarth@quicinc.com>
>> ---
>> v3: simplified algorithm further, fixed robot compile warnings
>> v2: removed minor optimisations to make more readable
>> v1: intial patch contained slightly complicated logic
>> ---
>>   drivers/tty/serial/qcom_geni_serial.c | 88 +++++++++++++++++++++--------------
>>   1 file changed, 53 insertions(+), 35 deletions(-)
>>
>> diff --git a/drivers/tty/serial/qcom_geni_serial.c b/drivers/tty/serial/qcom_geni_serial.c
>> index 2e23b65..ac2df1c 100644
>> --- a/drivers/tty/serial/qcom_geni_serial.c
>> +++ b/drivers/tty/serial/qcom_geni_serial.c
>> @@ -943,52 +943,71 @@ static int qcom_geni_serial_startup(struct uart_port *uport)
>>   	return 0;
>>   }
>>   
>> -static unsigned long get_clk_div_rate(struct clk *clk, unsigned int baud,
>> -			unsigned int sampling_rate, unsigned int *clk_div)
>> +static unsigned long find_clk_rate_in_tol(struct clk *clk, unsigned int desired_clk,
>> +			unsigned int *clk_div, unsigned int percent_tol)
>>   {
>> -	unsigned long ser_clk;
>> -	unsigned long desired_clk;
>> -	unsigned long freq, prev;
>> +	unsigned long freq;
>>   	unsigned long div, maxdiv;
>> -	int64_t mult;
>> -
>> -	desired_clk = baud * sampling_rate;
>> -	if (!desired_clk) {
>> -		pr_err("%s: Invalid frequency\n", __func__);
>> -		return 0;
>> -	}
>> +	u64 mult;
>> +	unsigned long offset, abs_tol, achieved;
>>   
>> +	abs_tol = div_u64((u64)desired_clk * percent_tol, 100);
>>   	maxdiv = CLK_DIV_MSK >> CLK_DIV_SHFT;
>> -	prev = 0;
>> -
>> -	for (div = 1; div <= maxdiv; div++) {
>> -		mult = div * desired_clk;
>> -		if (mult > ULONG_MAX)
>> +	div = 1;
>> +	while (div <= maxdiv) {
>> +		mult = (u64)div * desired_clk;
>> +		if (mult != (unsigned long)mult)
>>   			break;
>>   
>> -		freq = clk_round_rate(clk, (unsigned long)mult);
>> -		if (!(freq % desired_clk)) {
>> -			ser_clk = freq;
>> -			break;
>> -		}
>> +		offset = div * abs_tol;
>> +		freq = clk_round_rate(clk, mult - offset);
>>   
>> -		if (!prev)
>> -			ser_clk = freq;
>> -		else if (prev == freq)
>> +		/* Can only get lower if we're done */
>> +		if (freq < mult - offset)
>>   			break;
>>   
>> -		prev = freq;
>> +		/*
>> +		 * Re-calculate div in case rounding skipped rates but we
>> +		 * ended up at a good one, then check for a match.
>> +		 */
>> +		div = DIV_ROUND_CLOSEST(freq, desired_clk);
>> +		achieved = DIV_ROUND_CLOSEST(freq, div);
>> +		if (achieved <= desired_clk + abs_tol &&
>> +			achieved >= desired_clk - abs_tol) {
>> +			*clk_div = div;
>> +			return freq;
>> +		}
>> +
>> +		div = DIV_ROUND_UP(freq, desired_clk);
>>   	}
>>   
>> -	if (!ser_clk) {
>> -		pr_err("%s: Can't find matching DFS entry for baud %d\n",
>> -								__func__, baud);
>> -		return ser_clk;
>> +	return 0;
>> +}
>> +
>> +static unsigned long get_clk_div_rate(struct clk *clk, unsigned int baud,
>> +			unsigned int sampling_rate, unsigned int *clk_div)
>> +{
>> +	unsigned long ser_clk;
>> +	unsigned long desired_clk;
>> +
>> +	desired_clk = baud * sampling_rate;
>> +	if (!desired_clk) {
>> +		pr_err("%s: Invalid frequency\n", __func__);
> Note, this is a driver, ALWAYS use dev_err() and friends instead.
>
> Also do not allow userspace to flood the kernel logs like this looks is
> possible, this should just be dev_dbg().
>
> And of course, never use __func__, it's not needed anymore for
> dev_dbg().

Ok.

>
>> +		return 0;
> Why if you have a error, are you returning 0?

Yes, and it has been so earlier too.

0 is an invalid clock rate and will be handled accordingly by caller.

>>   	}
>>   
>> -	*clk_div = ser_clk / desired_clk;
>> -	if (!(*clk_div))
>> -		*clk_div = 1;
>> +	/*
>> +	 * try to find a clock rate within 2% tolerance, then within
>> +	 */
>> +	ser_clk = find_clk_rate_in_tol(clk, desired_clk, clk_div, 2);
>> +	if (!ser_clk)
>> +		ser_clk = find_clk_rate_in_tol(clk, desired_clk, clk_div, 5);
>> +
>> +	if (!ser_clk)
>> +		pr_err("Couldn't find suitable clock rate for %lu\n", desired_clk);
> return an error?
>
> dev_err().

As mentioned, we didn't (and don't) return error from here but 0.

>
>> +	else
>> +		pr_debug("desired_clk-%lu, ser_clk-%lu, clk_div-%lu\n",
>> +			desired_clk, ser_clk, *clk_div);
> dev_dbg()?
Ok.
>
> Also, as the kernel test robot says, this does not build cleanly :(

change to dev_dbg should take care of these.

Will do.

Thank you.

Vijay/


>
> thanks,
>
> greg k-h